Textbook of Pleural Diseases 2008 PDF
Textbook of Pleural Diseases 2008 PDF
Textbook of Pleural Diseases 2008 PDF
Second edition
and
www.hoddereducation.com
Whilst the advice and information in this book are believed to be true
and accurate at the date of going to press, neither the author[s] nor the
publisher can accept any legal responsibility or liability for any errors or
omissions that may be made. In particular (but without limiting the
generality of the preceding disclaimer) every effort has been made to
check drug dosages; however it is still possible that errors have been
missed. Furthermore, dosage schedules are constantly being revised and
new side-effects recognized. For these reasons the reader is strongly
urged to consult the drug companies’ printed instructions before
administering any of the drugs recommended in this book.
2 3 4 5 6 7 8 9 10
What do you think about this book? Or any other Hodder Arnold
title? Please send your comments to www.hoddereducation.com
Contents
Contributors ix
Preface xiii
Acknowledgements xv
Glossary xvii
Reference annotations xxi
The color plate section appears between pages 298 and 299
EFFUSIONS
24 Effusions from cardiac diseases 315
Gary T Kinasewitz and Kellie R Jones
25 Effusions from malignancy 323
Francisco Rodriguez-Panadero
26 Effusions from infections: parapneumonic effusions and empyema 341
Najib M Rahman and Robert JO Davies
27 Effusions from infections: tuberculosis 367
Esteban Pérez-Rodriguez and Richard W Light
28 Effusions from infections: atypical infections 379
Lisete R Teixeira and Francisco S Vargas
29 Effusions from lymphatic disruptions 389
Gunnar Hillerdal
30 Effusions from vascular causes 397
José M Porcel and Richard W Light
31 Effusions in immunocompromised hosts 409
Bekele Afessa and John J Mullon
32 Effusions from connective tissue diseases 421
Demosthenes Bouros and Dimitris A Vassilakis
33 Effusions caused by drugs 431
Ioannis Kalomenidis
34 Effusions after surgery 445
Oner Dikensoy and Richard W Light
35 Hepatic hydrothorax 455
José Castellote and Xavier Xiol
36 Effusions caused by gastrointestinal disease 465
Charlie Strange
37 Effusions of obstetric or gynecological origin 473
Richard W Light
38 Benign fibrous tumor of the pleura 483
Marc de Perrot
39 Undiagnosed pleural effusions 491
Nick A Maskell
ASBESTOS-RELATED DISEASES
40 Asbestos-related pleural diseases 499
A William Musk and Nicholas H de Klerk
Contents vii
PNEUMOTHORAX
42 Spontaneous pneumothorax 515
Andrew C Miller
43 Non-spontaneous pneumothorax 533
Michael H Baumann
PEDIATRIC CONSIDERATIONS
44 Pediatric pleural diseases 545
Elizabeth A Perkett and Paul E Moore
INTERVENTION PROCEDURES
45 Drainage techniques 551
Henri Colt
46 Pleurodesis 569
Helen E Davies and YC Gary Lee
47 Medical thoracoscopy 583
Robert Loddenkemper
48 Surgery for pleural diseases 599
David A Waller and Antonio E Martin-Ucar
49 Gene therapy in pleural diseases 613
Steven M Albelda and Daniel H Sterman
CONCLUSION
50 Future directions 621
YC Gary Lee and Richard W Light
Index 629
This page intentionally left blank
Contributors
We have been most encouraged and delighted by the very pleural disease). Nonetheless, a definitive etiology remains
positive responses received since the publication of the elusive in many pleural effusions (see Undiagnosed pleural
first edition of the Textbook of Pleural Diseases in 2003. It effusions), and serves as a reminder of the inadequacy of
has been our aim to provide a comprehensive, yet easily our current knowledge of pleural pathologies.
readable, reference text that covers both basic and clinical Controversies have grown around how best to manage
science on pleural diseases. By doing so, we hope that this recurrent effusions and a separate chapter is introduced on
text will serve to stimulate much needed clinical and Pleurodesis. A fascinating new chapter on History of pleural
research interests in pleural diseases. diseases provides us an opportunity to reflect on how far
Knowledge concerning pleural diseases has grown in research on pleural diseases has progressed in the past cen-
stature and complexity in the past few years. Seven new turies.
chapters have been added to this edition and the remain- In this edition, we have kept the popular format of pro-
ing chapters substantially updated to reflect the expansion viding a bullet point summary in each chapter as well as
in knowledge in the field. Ten of the fifty chapters in this highlighting seminal papers in the reference lists. We are
edition are written by new expert authors providing fresh confident that this edition will find a useful place in all
ideas and views. Recent years have seen the incorporation hospitals and medical libraries.
of new technologies in clinical management of pleural dis-
eases, hence the new chapters on Pleural ultrasound and Richard W Light
Pleural manometry. Likewise, new translational research YC Gary Lee
techniques are likely to enhance our understanding of March 2008
pathophysiology of pleural diseases (see Proteomics in
Pleural disease affects over 3000 subjects per million pop- This book provides a reference text for both clinicians
ulation each year. While the pleural cavity is a confined and scientists whose clinical or research work involves the
space, pleural disease can originate from a broad spectrum pleura. The 50 chapters are divided into the Basic Science
of pathologies: from local diseases of the lung and pleura and the Clinical Science sections. Experts on the respective
to systemic conditions such as collagen vascular diseases or topics from ten countries and four continents provide
drug reactions. Although the clinical presentations of leading edge information in the context of their years of
pleural diseases are usually limited to pleuritic pain and experience in the field. In each chapter, the author(s) has
dyspnea, the determination of the underlying cause(s) is prepared a summary that underscores the ‘take-home’
frequently a clinical challenge despite the use of modern messages. Interested readers can also pursue further read-
diagnostic modalities. Because the pleura is involved in ings from the highlighted references at the end of each
such a wide range of conditions, contributions into pleural chapter.
disease research have come from diverse backgrounds. Yet The Basic Science section covers the key aspects of pre-
there are few comprehensive texts dedicated to pleural clinical topics involving the pleura. It contains state-of-
disease, and none that covers both the basic science and the-art scientific knowledge but is written with sufficient
clinical aspects of the pleura. clarity that it can easily be understood by clinicians with
xiv Preface to the first edition
limited basic science background. The Clinical Science It is our hope that this book will provide an easy refer-
section begins with chapters general to pleural diseases, ence for clinicians faced with specific management issues,
including the approach to patients, histology and radiol- yet allow interested readers to gather a comprehensive
ogy. The subsequent chapters discuss specific groups of picture of the topic including the immuno-pathological
pleural diseases in detail. Each of those chapters shares the validation behind the disease development. It is our hope
same outline format, covering etiology, presentation, diag- that this book will stimulate interest in pleural disease,
nosis, treatment and complications. Any published guide- both in its research and clinical practice.
lines are also included. In the final chapter, we endeavor to
make brave predictions of how we think the area of pleural YC Gary Lee
disease may develop over the next decade or two. Richard W Light
July 2002
Acknowledgements
We are grateful to the many readers who gave us extremely Philip Shaw, who worked tirelessly to ensure all aspects of
positive and constructive feedback on the first edition of the book met the highest standard we set.
this book. These encouragements have supported us Once again, our spouses (Audrey and Judi) have been
through the many long hours spent in preparing this new most patient with us throughout the many months of pro-
edition. duction while we spent most evenings and weekends
We must thank all the authors for their energy and time indulging in the pleasure of reading and thinking about
in writing their superb chapters, without monetary reward pleural diseases.
but merely out of their passion on the subjects. The
smooth production of this book would not have been pos- YC Gary Lee
sible without the help of the dedicated and experienced Richard W Light
team at Hodder Arnold, in particular Amy Mulick and March 2008
This page intentionally left blank
Glossary
The reference lists are annotated, where appropriate, to guide readers to key primary papers and major review articles as
follows:
We hope that this feature will render extensive lists of references more useful to the reader and help to encourage self-
directed learning among both trainees and practicing physicians.
This page intentionally left blank
1
Pleural disease: historic perspective
For centuries the pleura as an anatomical entity was either Another advance came from the School of Alexandria
ignored or confused with the thoracic wall.1 In ancient with Erasistratus (310–250BC), who distinguished between
Greek, pleuron meant flank or side and was used generi- illnesses which affected only the lung and those which
cally in other fields such as mythology, zoology, geometry affected the hymen hypezocota (undergirding membrane)
and botany. Pleuron was an important figure in Greek which covers the thorax.3
mythology (the brother of Calydon and son of Aetolus); in Pliny (23–79AD) said ‘Nature, in its foresight, has
zoology, pleuron is the lateral part of the thorax of an enclosed all the principal viscera in membranes (membranae
insect; in Plato’s Timaeus (c. 360BC) the side of a triangle propriae) as if inside a special sheath.’4
is called a pleuron (Timaeus, 53c). There was also the Soranus (98–138AD) and Galen (129–200AD) continued
city of Pleuron in Aetolia, cited by Homer in the Iliad Erasistratus’s approach, describing the ‘membrana succin-
(Book II, 639) and by Ovid in Metamorphoses (Book VII, gens’ (girding membrane): ‘Nature created another struc-
382), so-called perhaps because it was built on the side of ture of the same substance as the peritoneum and performing
a hill. for the organs of the pneuma as does the peritoneum for the
organs of assimilation. It is as fine as a spider’s web; it is
homogeneous and a true and real membrane.’ He attempted
THE PLEURA AS AN ANATOMICAL ENTITY to define their function: ‘These membranes are present on
the inside of the entire thoracic cavity. They are located in
There are no specific references to the pleura in the pre- order to provide protection for the lungs, preventing them
Greek period, even though the ancient Egyptians could from striking the bare bone during breathing.’5,6
well have known of its existence, given their extensive Aretaeus of Cappadocia (200AD) is more precise: ‘Under
knowledge of mummification and their practice of remov- the ribs, the spine, and the internal part of the thorax as far
ing the organs from a body to be placed in their own as the clavicles, there is stretched a thin strong membrane,
funerary (canopic) jar. Having been extracted through the adhering to the bones, which is named succingens’.7
side of the diaphragm, the lungs were placed in a jar dedi- Avicenna (980–1037AD), a follower of the works of
cated to the baboon-headed god Hapi. Aristotle, advanced the hypothesis in the book the
The first to introduce the concept of a membrane was Canon of Medicine that the pleura also covers the thoracic
Aristotle (384–322BC). In his History of Animals he wrote organs.8
‘In all sanguineous animals membranes are found. And However, it was only during the Renaissance when
membrane resembles a thin close-textured skin, but its autopsies were carried out that a more accurate definition
qualities are different, as it admits neither of cleavage nor of could be given. Mondino de Liuzzi (1275–1326), in his
extension. Membrane envelops each one of the bones anatomical treatise Anothomia (1316), describes the pari-
and each one of the viscera, both in the larger and the etal pleura. ‘The membranes are three: the mediastinal
smaller animals; though in the smaller animals the mem- pleura which divides the thoracic cavity medially in an ante-
branes are indiscernible from their extreme tenuity and rior–posterior direction. Its function is that if for some reason
minuteness.’2 an empyema, an accumulation of pus in the chest, forms, the
2 Pleural disease: Historic perspective
pus does not drain to the other part. Then there is the costal- 400BC and 1600AD, death was the result of ‘pleural infec-
vertebral pleura, which is a strong membrane with a nervous tion’ in 80 percent of cases, in 70 percent between 1600
structure; it is large and covers all the ribs and is thus in and 1800, in 50 percent until 1900, in 30 percent until
contact with all the organs of the thoracic cavity. Because of 1950, with a gradual reduction to the present-day 3–5
the sensitivity of the pleural membrane, an infection is fol- percent. Each step of this improvement is the result of a
lowed by a sharp pain in the side of the body. The third mem- significant therapeutic advance.
brane is the diaphragm.’9 The first to illustrate pleurisy was Hippocrates of Kos
Thus the pleura begins to take on its own identity, being (460–377BC), who frequently discussed it in his aphorisms
divided into its principal parts, each with its own charac- and treatise on diseases. The symptoms described were: a
teristics, and with accurate anatomical references. constant fever, sweat, a cough and pain. He said ‘When
Three centuries later, Caspar Bauhin, in his Theatrum pleuritis arises, a person suffers the following: he has pain in
Anatomicum (1592), was still more precise when he accu- his side, fever and shivering; he breathes rapidly and he has
rately described visceral pleura. With him, the ‘membrana orthopnea …’12
costas succigens’ definitely becomes ‘pleura.’10 Many of his aphorisms – short sentences which were for
However, it was only in the monumental work of many years the only source of information for practicing
Francois Xavier Bichat (1771–1802) that the function of doctors – deal with pleurisy: ‘Pleuritis that does not clear up
the pleura was considered, rather than being simply in fourteen days results in empyema.’ ‘Pneumonia coming on
regarded as a covering. In his ‘Traité des membranes en pleurisy is bad.’ ‘Pains and fevers occur rather at the forma-
général et de diverses membranes en particulier’ (treatise on tion of pus than when it is already formed.’ Others refer to
membranes in general and various membranes in particu- the prognosis: ‘Persons who become affected with empyema
lar), he wrote that ‘The serous membranes are characterized after pleurisy, if they get clear of it in forty days from the
by the lymphatic fluid which incessantly lubricates them and breaking of it, escape the disease; but if not, it passes into
that every serous membrane represents a sack without an phthisis.’ And therapy: ‘When empyema is treated by the
opening spread over the respective organs which it embraces. cautery or incision, if pure and white pus flow from the
Their first function is doubtless to form about the essential wound, the patients recover; but if mixed with blood, slimy
organs a boundary which separates them from those of their and fetid, they die.’13
vicinity. A second function is to facilitate the moving of the Hippocrates recommended treating empyema by
organs.’ Also, ‘the surface of the pleura has a smooth and draining through the mouth, or with purgatives, or induc-
pale aspect because of a serous fluid of which the membrane ing vomiting, but also attempted a more ‘modern’
is the source. This fluid is constantly produced and reab- approach, advising an oblique cut in the lowest intercostals
sorbed. It is so thin that the color of the underlying parts can space, perforation or removal of a piece of rib, thereby cre-
be distinguished.’11 ating the first open drainage system, and filling the incision
with linen gauze. He suggested washing the cavity with oil
and warm wine; an empyema can drain first from the
PLEURAL DISEASES lungs rather than from the thoracic wall; he describes
(digital) clubbing in chronic suppuration. He also identi-
It should be emphasized that the discussion of pleural fied the most appropriate location for the incision in the
infections in a historical context, and then making a sys- thoracic wall with ‘the sign of damp earth’, which was
tematic reconstruction, are extremely difficult owing to achieved by covering the patient’s thorax in mud and
the imprecise nature of the terminology used. For making the incision in the part which dried first. He was
example, the word ‘empyema’ was used not only to also responsible for ‘hippocratic succussion’, which
describe the disease but also for the operation performed involves shaking the patient by the shoulders in order to
to drain the empyema. The nature of the material removed identify the sound of liquid inside the thorax. This maneu-
in an intervention was not always specified, meaning that ver only had positive results in the presence of a certain
they could have been transudates, inflammatory exudates, level of air and liquid, and was therefore only possible in
or purulent effusions, since ‘paracentesis thoracis’ (tho- the final stages of the disease.14
racic paracentesis) was carried out in the presence of any From the ancient Roman period there is not only doc-
accumulation of liquid, almost always with the broad defi- umentary evidence regarding the history of pleural disease,
nition of ‘bad humor’. Also, an initially clear liquid often but also archaeological discoveries from excavations at
became purulent with the use of unsterile instruments. Pompei which have produced surgical instruments that
The first recognized and described pleural disease was were apparently used to drain purulent accumulations.
infectious pleurisy, both because the symptoms were often Aulus Cornelius Celsus (25BC to 50AD) identified the
clear, such as fever and characteristic thoracic pain, and characteristics of inflammation (rubor, tumor, calor,
because of its frequent evolution into empyema. dolor), applicable also to thoracic suppuration, and he
The frequency, morbidity and mortality for this disease describes pleurisy in his De Medicina (book IV): ‘The
were extremely high until the twentieth century. An stomach is girt about by the ribs, and in these also severe
approximate attempt at quantification is that between pains occur. And the commencement either is from a chill, or
Pleural diseases 3
from a blow, or from excessive running, or from disease. But ifests itself when the patient breathes strongly… the third
at times pain is all there is the matter, and this is recovered symptom is difficulty in breathing and frequency of breath-
from be it slowly or quickly; at times it goes on until it is dan- ing. The fourth symptom is a rapid and weak pulse. The fifth
gerous, and the acute disease arises which the Greeks call symptom is the cough.’8
pleurisy. To the aforesaid pain in the side is added fever and Until this time, descriptions of therapy tend to be brief
cough; and by means of the cough, phlegm is expectorated and then in the twelfth and thirteenth centuries there
when the disease is less serious, but blood when it is grave.’15 appears to have been a proactive approach towards surgery
He suggests treatment by bleeding or cupping glass, advis- by, among others, Henri de Mondeville (1260–1320).17
ing walks and potions of hyssop with dried figs, rubbing Ambrosie Paré (1510–1590) was a leading surgeon and
the shoulder blades, light food, the application to the chest specialist in battlefield surgery, and a pioneer in the
of a poultice of ground salt and abstinence from wine. medical field. (He was responsible for concocting an early
There is a curious but perhaps significant anecdote ‘antiseptic’ for treating wounds, made from egg-yolk, oil of
related by Pliny the Elder (23–79AD) in his Naturalis roses and turpentine.) He appears to have been particu-
Historia in which he tells the story of the consul Publius larly interested in empyema and had a series of instru-
Cornelius Rufus who, suffering from an empyema and ments made for the drainage of effusions, and suggested
convinced of his imminent end, launches himself boldly the advantage of enlarging the breach in the thorax to facil-
into battle and is struck by an enemy arrow in the thorax, itate the emptying of purulent substances and then
causing the draining of the effusion and consequently his washing the pleura with a disinfecting solution: ‘the pus
recovery.4 and matter must be evacuated little by little at several times,
Soranus (98–138AD) offers some elementary epidemi- and the capacity of the chest cleansed by a detergent injection
ological notions, confirming that pleurisy is more frequent of barley water and honey of rose.’ He pointed out the
in old age than in youth, in women than in men, and serious effect of the entry of air into the thorax (so that the
strikes mainly in winter months. vital spirits do not escape). He also described subcuta-
Galen (129–216AD) noted the importance of the pulse, neous emphysema resulting from a costal fracture (Figure
which is faster in cases of inflammatory diseases of the 1.1).18
thorax, and describes the occurrence of pleural effusion: ‘A The work of Fabrizio d’Acquapendente (1537–1619) is
sense of heaviness can persist inside the ribs. This means a also important. Having declared the empyema to be ‘a col-
quantity of pus or other humor has gathered in the thoracic lection of rotten material’, he proposed ‘manuariam oper-
cavity causing dyspnea.’6 ationem’ (manual operation) for removal, indicating that
There are numerous references to pleurisy in ancient the effusion should be drained gradually, the thoracotomy
Roman literature. For example Horace (65–8BC) called it breach left open, a perforated cannula introduced and the
‘laterum dolor’ (side pain) in his first book of the Satires. pleural cavity washed with wine and oil.19
Vitruvius (c. 80–25BC) in his De Architectura, Book I, said In 1528 the first book entirely dedicated to pleurisy
that it is a disease common in areas exposed to winds. appeared, entitled De incisione vene in pleuritide by Andrea
Aretaeus (c. 80AD), speaking of peripneumonia, writes: Turini. This was followed in 1536 by Liber de pleuritide ad
‘This is what we call Peripneumonia, being an inflammation Galeni e Hippocratis scopum by Benedetto Vittorio. Other
of the lungs, with acute fever, when they are attended with books of note were Tentamen medicum de pleuritide vera
heaviness of the chest, freedom from pain, provided the lungs by Leonardus Bardon di Montpellier (1777) and, in 1740,
alone are inflamed; for they are naturally insensible, being of the systematic treatment of pleurisy in the volume De pleu-
loose texture, like wool. But branches of the aspera arteria are ritide ejusque curatione (Figure 1.2) written by Daniel
spread through them, of a cartilaginous nature, and these, Wilhelm Triller, a doctor and philosopher in Wittemberg.
also, are insensible; muscles there are nowhere, and the nerves In 1664 the famous anatomist Frederik Ruysch graduated
are small, slender, and minister to motion. This is the cause at Leiden with a thesis entitled Disputatio medica inaugu-
of the insensibility to pain. But if any of the membranes, by ralis de pleuritide.
which it is connected with the chest, be inflamed, pain also is In his Armementarium Chirurgicum, Johann Scultetus
present; respiration bad, and hot; they wish to get up into an (1595–1645), student of Fabrizio d’Acquapendente, pub-
erect posture, as being the easiest of all postures for the respi- lished some fine illustrations showing the operation of
ration.’7 empyema, and underlined the active role the surgeon
In subsequent centuries, references to pleurisy are less should have regarding drainage in thoracic diseases.20
numerous, but in the Regimen Sanitatis Salernitanum In 1648, Riolano described thoracentesis of air ‘flatus
(1000AD) is written: ‘Pleuresia est vera cum spirandi gravi- cum violentia displosus’ (violent expulsion of air). More
tate, febreque continua, tussi, laterioque dolor’ (pleurisy is incisive is the chapter on empyema in the excellent book of
real if there is difficulty in breathing and constant fever, 1710 by Dionis, which has extensive and practical descrip-
cough and lateral pain).16 Avicenna (980–1037AD), in the tions and is backed up by showing a set of surgical instru-
third book of the Canon, described the clinical character- ments made especially for the operation.21
istics: ‘The symptoms of simple pleurisy are clear: constant In 1699, Giorgio Baglivi in De praxi medica followed
fever, violent pain under the ribs which sometimes only man- Galen’s line: ‘If you would discover pleurisy, place your chief
4 Pleural disease: Historic perspective
Figure 1.1 Ambroise Paré’s 1582 book. Figure 1.2 Triller’s 1740 De pleuritide ejusque curatione (On
pleurisy and its treatment).
care in observing the nature of the pulse. The hardness of the though not necessarily the result of diseases such as
pulse is almost an infallible sign of all pleurisies’.22 pleurisy or pneumonia.
In his Observationes Medicae (1676), Thomas The origin of pleural adhesions gave rise to some fanci-
Sydenham classified pleurisy as an epidemic disease. Willis ful theories, the most original of which was surely that of
(1621–1675) considered pleurisy and peripneumonia to be the French zoologist Duvernoy (1777–1855), who attrib-
separate diseases. Boerhaave (1668–1738) described the uted them to laughter: ‘le rire en effet n’appartient qu’aux
variation in degrees of pain experienced by the patient seuls humains, et seulement après qu’ils sont nés’ (only
caused by the act of breathing and pneumothorax follow- humans laugh, and only after they are born).
ing the rupture of the esophagus. In 1767, Hewson claimed In a series of 3000 autopsies in 1767, Joseph Lietaud
that removal of air from the thorax was also possible. described the presence of two pleural tumors, possibly
Giambattista Morgagni, in 1761 in De Sedibus et Causis mesothelioma: ‘Reperitur in dextro thoracis latere tumor
Morborum (Letters 20 and 21), discussed pleurisy at length ingens pleuram occupans’ (On the right side of the thorax
and remarked on the frequency of adhesions. Their pres- there was found a large tumor in the pleura).23
ence had already been noted in the sixteenth century, the In the nineteenth century, more emphasis was placed
period in which the practice of autopsy had become more on attempting early diagnosis. Until then interventions
widespread, and Isbrand di Diemerbroek, professor of often took place when the symptoms of disease had
Anatomy in Utrecht from 1651 to 1674, speculated that up become very obvious, i.e. at a very late stage.
to a third of the ‘normal’ population could be subject to Progress in this direction can be attributed to semeiotic
this ‘anomaly’. His contemporary, Nicolas Tulpius from respiration initiated by Leopold Auenbrugger and his
Amsterdam, famous for being immortalized in a painting ‘Inventum Novum’ of 1761 (Figure 1.3) in which he pre-
by Rembrandt, considered that a lung entirely free from sented chest percussion as a method of physical examina-
adhesions was a rarity in adults. This opinion was shared tion: ‘The thorax of a healthy person sounds when struck. If
by Morgagni, who, having observed their absence in a sonorous region of the chest appears, on percussion, entirely
fetuses, considered that they were not congenital, even destitute of the natural sound, disease exists in that region’).
Pleural diseases 5
However, perhaps even more significant was ‘De After Laennec diagnosis became more accurate, and the
L’Auscultation Médiate’ by R.T.H. Laennec in 1819 in identification of suspected pleurisy could be made earlier.
which he systematically described thoracic diseases. In 1828 Piorry introduced indirect percussion and in 1843
Laennec distinguished between acute and chronic Damoiseau, and later Ellis in 1874, identified the semi-
pleurisy, classified pleural disease, separates them from elliptical curve, which is highest in the axilla, to define the
peripneumonia and described the hemorrhagic form and limit of an effusion. In 1874 Garland described a paraver-
fibrothorax. With the introduction of the stethoscope, the tebral triangular area of relative resonance in the lower
presence of liquid in the thorax could be diagnosed at an back between the spine and Damoiseau-line found in the
earlier stage. He identified the major obstacle to the suc- same side as an effusion. In 1902 Grocco described a trian-
cessful removal of liquid as being the pressure of the lung gular area of dullness at the base of the chest near the
on the mediastinum and vertebral column and understood spinal column on the side opposite a pleural effusion
the danger of exposing the pleural cavity to the external (Grocco’s triangle) due to mediastinal shift caused by the
environment, preferring repeated thoracentesis to effusion.
trepanatation of the rib or an incision in the intercostal William Stokes (1804–1878), known for the epony-
space. He considered the fourth–fifth intercostals space the mous ‘periodic breathing’ (Cheyne–Stokes respiration)
ideal location for thoracentesis because of its centrality.24 and for atrioventricular block, described 20 cases of
In 1804 John Bell began to describe pneumothorax, pleurisy treated successfully. He thought the results of tho-
even though it had not yet been called that. He considered racentesis were often not encouraging because it took
the cause to be either thoracic traumas, putrefaction of the place too late; he described contraindications in the pres-
pleural cavity, or erosion of the lung.25 The term ‘pneu- ence of a fistula and considered that pleurisy was often
mothorax’ was coined by a student of Laennec, Jean Marc caused by tuberculosis. Perhaps he was also the first to
Gaspard Itard, who in 1803 described five cases of tuber- create a unidirectional valve. He proposed a higher posi-
culosis with air in the pleural cavity, even though he did tion for the drain to facilitate spontaneous outflow of the
not know of the existence of the spontaneous variant.26 liquid and prevent possible lesions to the diaphragm if
The latter form was described by his mentor: ‘Aeriform entry was too low.27
fluid can be exhaled in the pleura without visible alterations Henry Bowditch (1808–1892) was an internist and
in the membrane. I have twice seen an abundant simple physiologist from Boston who in 1852 published the
pneumothorax accompanied by such dryness of the pleura results of paracentesis of the thorax in 65 cases. The treat-
that the membrane resembled parchment’.24 ment was used in the initial phases of exudate, based on
clinical evidence of dyspnea and the presence of liquid,
which was examined under a microscope, and in his
opinion thoracentesis should not have been considered an
extreme intervention, but a simple remedy.28 This view
was also held by his contemporary, Armand Trousseau
(1801–1867), a French internist: ‘I do not believe that I
invented thoracentesis, in the same way that I did not invent
the instruments to perform it, but I do think I am among the
first to recommend its application in cases of excessive liquid
in the thorax.’ He thought thoracentesis simple and that it
could brilliantly resolve many clinical cases, even if often
performed late, and that the introduction of air did not
cause damage even if the pleura was inflamed.29
The physiology of pleura also became better under-
stood. In 1863 Recklinghausen published an important
study on the mesothelial stomata.30 Carl Ludwig first made
graphic recordings of intrapleural pressure in 1847 by use
of a water-filled balloon placed in the pleural space and in
1900 Aron reported the first measurement of intrapleural
pressure in a healthy human.31 West proposed the hypoth-
esis that the interface between the parietal wall and the
lungs is separated by the pleurae.31 Ernest Starling pro-
duced the equation which regulates hydrostatic and
oncotic forces in 1895.32 In 1870 Wagner gave the first
anatomical–pathological description of a primary pleural
tumor and in 1882 Ehrlich realized the possibility of iden-
Figure 1.3 Inventum novum (The new discovery) the first tifying carcinoma cells in pleural liquid, describing the
description of percussion by Leopold Auenbrugger. characteristic rosette-shaped aggregation.33
6 Pleural disease: Historic perspective
immersed in a bottle containing sodium and potassium direct passage was made between the pleural cavity and the
permanganate (as an antiseptic). Pleural fluid could be skin to allow open drainage of pus. The skin was folded
drained slowly through the cannula and the force of back around the edges. This method is still used in a
aspiration could be increased by gradually lowering the minority of cases which cannot be treated with less inva-
bottle. sive methods.46
Gotthard Bülau, still today regarded as the leading
pioneer of the water valve, described the application thus:
‘I have always believed that the principal advantage of CONCLUSIONS
siphon-drainage is that it lowers the pressure within the
pleural space, thereby bringing about re-expansion of the The famous French surgeon Guillaume Dupuytren
lung. The drain can function as a valve, allowing escape of (1777–1835) consulted five doctors when he felt ill with
pus and the air which has entered during the operation, while empyema. Having listened to their conflicting diagnoses,
preventing entry of air. With each forcible expiration against he said, ‘I prefer to die by the hand of God than with the
a closed glottis, the air in the opposite (healthy) lung is forced help of a surgeon.’
into the partially collapsed one. With the next inspiration the This chapter outlined the slow and difficult journey
valve closes, the lungs can expand, the expansion is main- traveled in the development of better management strate-
tained… ’.41 gies for patients with pleural diseases, especially in estab-
Another innovative concept was that introduced in lishing interventions which were both simple and effective.
1877 by the Finn J.A. Estlander, who was the first to Removing pathological material from an easily accessible
understand the beneficial effects on the lung of collapsing (pleural) cavity would appear a simple task in theory. In
the thoracic wall in the treatment of chronic empyema. He reality, it has taken centuries of research to arrive at our
designed the thoracoplasty which, with numerous current practices.47
variations, would be used until the 1950s for the treatment
of suppurative lung diseases.42 This procedure was further
developed by the Swiss surgeon Edouard de Cérenville
(1843–1915), who in 1885 widened its application to the
tuberculous cavities, abscesses and suppurative infarcts,
and proposed the total resection of one or more ribs.43 KEY POINTS
In 1893 the French surgeon Edmond Delorme per-
formed the first decortication on a young French soldier ● Parietal pleura was recognized as a separate
suffering from tuberculous empyema and removed the anatomical entity and was subdivided into its
visceral pleura to re-expand the lungs.44 principal parts (mediastinal, costal-vertebral and
In 1918 the world was ravaged by Spanish influenza, a diaphragmatic pleura) in the treatise Anothomia
terrible scourge which killed 20 million people worldwide. (1316) by Mondino de Liuzzi.
The disease was often complicated by development of a ● Visceral pleura was described in detail by Caspar
pleural empyema, which led to attempts to apply various Bauhin in his Theatrum Anatomicum (1592).
techniques for therapeutic purposes. Within a short time ● The first to illustrate pleurisy was Hippocrates of
the USA initiated a Commission for the study of Kos (460–377BC), who frequently discussed it in
empyema, and a brilliant young surgeon, Evarts Graham his aphorisms and treatise on diseases.
(1883–1957), was put in charge. He subsequently became ● A systematic treatment of pleurisy was demon-
famous for carrying out the first pneumonectomy for a strated in the 1740 volume De pleuritide ejusque
pulmonary tumor on April 5, 1933. He noted that curatione (On pleurisy and its treatment) written
premature use of open drainage, in particular in the by Daniel Wilhelm Triller, a doctor and philoso-
presence of streptococcus infection which produces pher in Wittemberg.
‘liquid’ pus, was detrimental because it might provoke a ● Thoracoscopy was first started in 1910 by the
mediastinal shift that could be fatal. However, in Swedish internist Hans Christian Jacobaeus.
empyemas in which the pus was more viscous and ● Gotthard Bülau, a German internist, the leading
adhesions formed more rapidly, open drainage was less pioneer of the water valve, introduced the closed
dangerous. He also noted that the first type of infection water-seal drainage for empyema in 1875.
was more common in soldiers fighting in World War I, ● Thoracoplasty in the treatment of chronic
whilst the second was more common in civilians. Soldiers empyema was designed in 1877 by J.A. Estlander.
were therefore always treated with closed drainage and a ● The first decortication with removal of the vis-
water seal, and within a few months the mortality rate ceral pleura to re-expand the lungs was per-
dropped from 30 to 4 percent.45 formed in 1893 by the French surgeon E.
One further surgical approach to be considered is that Delorme on a young French soldier suffering
of Leo Eloesser, who in 1935 developed ‘skin flap open from tuberculous empyema.
drainage’, for the treatment of tuberculous empyema. A
References 9
INTRODUCTION erated between these two surfaces, the inner surface of the
thoracic cage and the outer surface of the lung are covered
The precise structure and function of the pleural space is by a serous, elastic membrane with a smooth lubricating
not fully understood. There is marked inter-species varia- surface – the pleura.
tion in the ultrastructure of the pleura; in humans the The pleural cavity describes the slit-like fluid filled
pleural cavities are separated by the mediastinum, whereas space between these surfaces. It is thus an entirely sealed
other mammals (e.g. mice, American buffalo) lack com- cavity inserted between the chest wall and lung, main-
plete separation between the left and right pleurae, allow- tained at 10–20 mm across. This arrangement is crucial to
ing free movement of air and fluid. The adult elephant is the efficient function of the lung, in a manner analogous to
the only mammal that does not possess a pleural cavity – a the pericardium and pericardial space to the heart.
normal pleural space is present in utero, but in late gesta-
tion the parietal pleural is replaced with a dense sheet of
connective tissue.1 The pleural space is then obliterated EMBRYOLOGY
with loose connective tissue, permitting movement of the
lung against the chest wall. Theories have been advanced as All body cavities, including the pleural peritoneal and peri-
to why the elephant lacks a pleural space,2 but the varia- cardial cavities, are derived from the coelom, the primitive
tions seen between mammals in the structure of the pleural body cavity. The coelom derives from the primitive meso-
cavity is as yet to be explained. derm of the embryo and forms a cavity lined by serous
In humans, each hemithorax is constructed like a verti- membrane before all internal organs develop.
cal, cone-shaped bellow with the diaphragm as the moving In the human embryo, the primitive mesoderm on both
part at the caudal and widest end, and the trachea and sides of the notochord divides first into the medial seg-
nasal structures superiorly creating a narrow outlet/inlet mented and natural non-segmented plates. The left and
and providing protection and functional adaptation for right medial segmented plates later develop into the skull,
the lung. Within the thoracic cage, the lung must be able vertebra, ribs and the thick muscles of the dorsal (rear)
to both move and change volume with the respiratory parts of the body wall.
cycle. In contrast, the lateral non-segmented plates split into
The lung expands during inhalation and deflates during the internal splanchnopleure (the precursor of the internal
exhalation, creating movement between the lung surface organs) and the lateral or external somatopleure (the pre-
and the inner chest wall. In order to decrease friction gen- cursor of the anterior and lateral body wall), between
14 Anatomy of the pleura
which structures a slit-like cavity is formed. The paired left ● the cervical pleura, rising in to the neck and extending
and right splanchnopleure and somatopleure with their above the first rib;
associated cavities extend along the length of the embryo ● the mediastinal pleura, adherent to the mediastinal
in a cephalo-caudal direction, and out and over the surface structures.
of the yolk sacked ventrally or anteriorly. The fusion of the
left and right somatopleure and two cavities ventrally (i.e.
the fusion and closure of the ventral wall) creates a sealed THE PLEURAL CAVITY
primitive body cavity, an intra-embryonic coelom, in the
seventh a week of gestation.3 At this early stage the cavity The pleural cavity is a sealed but expandable space that is
is already completely covered by a layer of serous mem- formed between the visceral and parietal pleura.5 The right
brane with mesothelial cells on the surface.4 and left pleural cavities in humans are completely separate
With the shrinkage of the yolk sack, the coelom from one another, and from the mediastinum and pericar-
expands. The internal organs move around within the dial cavities. The dome, or cupola, of the pleural cavity
embryo, changing in size and shape during development. extends above the first rib for 2–3 cm along the medial
Organs protrude as they develop into the body cavities one-third of the clavicle behind the sternocleidomastoid
and, as they protrude, are enveloped by a layer of serous muscles. The pleural space may therefore be entered in
membrane that covers the inner surface of the body cavi- trauma to the lower neck, procedures such as central
ties. Meanwhile, the coelom divides into the pleural and venous catheter insertion or surgical dissection of the
pericardial cavities by the fusion of the transverse septum lymph nodes, resulting in pneumothorax.
arising from the ventral, as well as left and right, pleu- During development, several structures within the tho-
roperitoneal folds from the dorsal walls. When the two racic cavity acquire a double layer of parietal pleura. In the
pleuroperitoneal folds fuse, the two pleural cavities are lower mediastinum, the dorsal and ventral mediastinal
completely separated from each other and from the peri- parietal pleura are pulled into the thoracic cavity as the
cardial cavity.4 This arrangement allows flexibility for the lung develops, and form vertical structures from the hilum
organs to expand, retract, deform or displace each other as of the lung to the diaphragm, resulting in a pair of back-
they develop and grow in the limited space of the four to-back layers of parietal pleura. These may persist into
body cavities. adulthood and are called the pulmonary ligaments.
Pulmonary ligaments may divide the pleural space below
the hilum into anterior and posterior compartments,6 and
may contain large lymphatic vessels. During surgery to the
GROSS ANATOMY lung, incomplete ligation or damage to these lymphatic
vessels may result in post-operative pleural effusion.7 It has
To the naked eye, the normal pleural surfaces are smooth, been suggested that their presence may prevent torsion of
wet and semi-transparent. In humans, the left and right the lower lobes.
pleural cavities are entirely separated from one another (by Inferiorly, the pleura reflect at the lower boundary of
the mediastinum) and from the pericardial space. the thoracic cage, but often extend beyond the costal
Although the visceral and parietal pleura originate from margin in the right infrasternal region and at the costo-
the same serous membrane of the coelom, they appear vertebral angles bilaterally. A radiological study has
macroscopically different because of different underlying demonstrated that the lung lies at or below the level of the
structures and topography (see section ‘The regional dif- 12th rib anteriorly in 80 percent of patients, and in 18
ference’). percent the lung reaches the level of the body of the L1 ver-
The visceral pleura covers the entire surface of the lung, tebra posteriorly.8
including the interlobular fissures. The parietal pleura During deep inspiration, the lung fills the pleural cavity
covers the inner surface of the entire thoracic cage, includ- entirely. As the lung deflates during expiration, the most
ing the mediastinal surfaces and diaphragm. The visceral inferior and distal parts of the parietal pleura extend
and parietal pleura coalesce at the lung hilae, where the beyond the costal margins and may come into direct
major airways and pulmonary vessels penetrate. The area contact with one another to form the costophrenic recesses
of the entire pleural surface has been estimated to be 2000 – any increase in the amount of fluid in the pleural cavity
cm2 in an average adult male. accumulates here first. Attempts to aspirate small amounts
The visceral pleura adheres tightly to the lung surface of pleural fluid within this recess, or approaching the
throughout the thorax. The parietal pleura is further sub- upper abdominal structures (liver, adrenals or kidneys)
divided in to different anatomical regions as follows: posteriorly during medical or surgical procedures, may
result in damage to the lung and pleura, and in iatrogenic
● the costal pleura, lining the inner surface of the ribs and pneumothorax or hemothorax.8
intercostal muscles; The visceral pleura extends into the interlobular space,
● the diaphragmatic pleura, covering the convex surface and therefore each lobe of the lung may expand or collapse
of the diaphragm; individually. Abnormal divisions of lobes and segments
Mesothelial cells 15
are common5 and interlobar fissures may be incompletely 4 a loose connective tissue layer (contains nerves, blood
or completely separated by septa. These altered fissures vessels, lymphatics);
may appear radiologically as linear shadows, and occasion- 5 a deep fibroelastic layer (often fused to underlying
ally cause ‘vanishing tumours’ when pleural fluid is tissue).
trapped within them, due to conditions in which fluid may
fluctuate such as heart failure.9 The thickness of each layer is variable between species, and
The major fissure of the lung extends obliquely down- displays variation between regions in the pleural space in
wards, posteriorly to anteriorly, paralleling the sixth rib. both animal and humans5,11–15 (see section ‘The regional
An intercostal tube placed through the fifth, sixth or difference’).
seventh intercostal space may therefore enter the pleural The thickness and boundaries of the superficial connec-
cavity near the major fissure, and if the tube is directed tive and elastic fiber layers (i.e. the second and third layers)
centrally, may enter and become trapped within the are usually imprecise. The loose fourth connective tissue
fissure, resulting in ineffective drainage.10 layer contains adipose tissue, fibroblasts, mast cells and
other mononuclear cells, blood vessels, nerves and lym-
phatics, and often serves as the cleavage plane at pleurec-
MICROSCOPIC ANATOMY tomy. The fifth deep fibroelastic layer often adheres tightly
to, or is fused with, the underlying tissue (i.e. lung
Layers of the pleural membrane parenchyma, mediastinum, diaphragm, ribs or intercostal
muscles). The fibroelastic layer is highly variable, accord-
The visceral and parietal pleura in humans are approxi- ing to its position within the pleural cavity. The overall
mately 40 mm across. By light microscopy, the pleura is amount, proportion and integration of the fibroelastic
generally divided into five layers (see Figure 2.1), consist- meshwork are directly proportional to the extent of pleural
ing of a single cellular layer and four subcellular layers. excursion during the respiratory cycle.13
Proceeding from the pleural surface, the layers are as
follows:
MESOTHELIAL CELLS
1 a single layer of mesothelial cells;
2 a thin subendothelial connective tissue layer, including A single layer of mesothelial cells covers the surface of both
a basal lamina; the parietal and visceral pleura. Although there are mor-
3 a thin superficial and elastic layer (often merged with phological differences in mesothelial cells found in differ-
the second in layer); ent areas,16 no significant differences have been identified
1. Mesothelial cells
2. Basal lamina
3. Superficial elastic layer
Capillaries
Fat cell
Valves Nerve
Lymphatic valves
Figure 2.1 A schematic drawing of the pleura. On light microscopy, the pleura is divided into five layers. From the pleural surface, the
layers are: first, a single layer of mesothelial cells; second, a thin submesothelial connective tissue lair, including a basal lamina; third, a
thin superficial elastic layer; fourth, a loose connective tissue layer; and fifth, a deep fibroelastic layer. The thickness of each layer is
markedly varied between species and between regions within the same individual.
16 Anatomy of the pleura
out from the cell surface, slightly twisting and interwoven The venous drainage of the parietal pleura follows its arte-
with each other13 (Figures 2.2 and 2.3). rial supply, with the majority eventually draining into the
The density of the carpet of microvilli varies according azygos vein and subsequently into the superior vena cava.
to position within the thoracic cavity, ranging from only a Venous blood derived from the diaphragm drains either
few to more than 600 per 100 mm2, with an average of caudally into the inferior vena cava through the inferior
300 per 100 mm2.11,14,18,19 A higher density of microvilli (in phrenic veins or cranially into the superior vena cava
association with abundant hyaluronic acid) is found in the through the superior phrenic veins, which run parallel
most actively moving parts of the lung, i.e. the lower tho- with the internal mammary artery and thence into the
racic cavity,11,13,14 where the contraction and relation of the brachiochepalic trunk.
diaphragm and expansion and retraction of the lung and
classic caged is highest. A higher density of microvilli is
also found in the visceral rather than the parietal pleura at The blood supply of the visceral pleura
any given level.11,13,14 The lowest density of microvilli is
found on the inner surface of the ribs. Animal studies The arterial blood supply of the visceral pleura in humans
suggest that the microvilli carpet develops over the first is still debated. Animals with thick pleura (such as horses,
few months of life, and subsequently changes and decays pigs or sheep), derive their arterial blood supply for the
with increasing age.23 visceral pleura from the bronchial arteries. Animals with
The function of microvilli is not completely clear. The thin pleura (such as mice, rats and rabbits) tend to derive
cell surface area of the mesothelial cell layer is substantially visceral pleural blood supply from the pulmonary circula-
increased by the presence of microvilli, and it is presumed tion. Albertine et al.14 have demonstrated that in young
that this also increases the cell membrane dependent func- adult sheep the bronchial artery supplies the visceral
tions, such as that of a variety of receptors ligands and pleura completely and exclusively.
enzyme productions such as metalloproteinases. As men- Humans have thick visceral pleura, and therefore it is
tioned above, microvilli are associated with pinocytototic expected that the bronchial circulation should supply the
vessels, implying an important role in transcellular trans- human visceral pleura. However, this has not been clearly
port.24 Given the apparent relationship between microvilli demonstrated.26 There is agreement that the bronchial
density and degree of movement of the lung, it seems likely artery supplies most of the pleura facing the mediastinum,
that the microvilli’s role in enmeshing glycoproteins rich the pleura covering the interlobular surfaces, and a part of
in hyaluronic acid to lubricate the pleural surface and the diaphragmatic surface. The blood supply for the
lesson friction between the lung and thorax is a key part of remaining visceral pleura (i.e. the entire convex costal lung
their function.11,19 surface including the apex and the greater part of the
Hyaluronic acid is secreted by the mesothelial cell, and diaphragmatic surface) is less certain.27 It has been sug-
by mesenchymal cells in the submesothelial interstitial gested that these parts of the visceral pleura are supplied by
tissue. Hyaluronic acid may be demonstrated by alcian pulmonary arteries which arise beneath the pleura from
blue or colloidal iron stains using light microscopy (LM) the pulmonary circulation.5 Using techniques to delineate
and electron microscopy (EM).11 pulmonary and bronchial vessels, Milne and Pistolesi28
concluded that the visceral pleural circulation is derived
from and continuous with the pulmonary circulation.
THE BLOOD SUPPLY OF THE PLEURA Disagreement continues about the blood supply of visceral
pleura in humans (see section ‘Shunting and pathological
The blood supply of the parietal pleura changes with age’).
The greater part of the visceral pleura supplied by the
The parietal pleura is richly supplied with arterial blood, bronchial artery is drained through the pulmonary veins,
derived from multiple branches of adjacent systemic arter- except for a small area around the hilum where the pleural
ies according to region:5,15,25 veins drain into the bronchial veins.
which is deprived of the original bronchial blood supply. contact between the visceral and parietal pleura,34 suggest-
This phenomenon is most apparent in bullae of the lung. ing that the pleural space is a real rather than a potential
Within the aged lung, and in association with many space.
chronic lung and pleural diseases, the bronchial arteries Normal pleural fluid contains 1–2 g of protein per
proliferate around the airway and in the interlobular 100 mL, a figure similar to the concentration detected in
septum. Bronchial artery proliferation may be seen within the interstitial fluid of both animals and humans.29,35
the pleura in inflammation and fibrosis, and systemic However, the concentration of large molecular weight pro-
arteries may invade into the visceral pleura from the pari- teins (e.g. lactate dehydrogenase, molecular weight
etal site, especially when adhesions between the lung and 134 000) in pleural fluid is less than half that of serum,
the chest wall develop. implying some regulation of molecular passage into the
pleural cavity. There are 1400 to 4500 cells per microliter
of pleural fluid in animals or humans,29,30,35 largely made
INNERVATION up of macrophages with a few leukocytes and red blood
cells, again implying a restriction to cellular passage into
The costal parietal pleura and the peripheral part of the the pleural cavity. (For further description of normal
diaphragm are innervated by somatic intercostals pleural fluid composition, see Chapter 4, Normal physio-
nerves,15,25 thus pain felt in these areas is referred to the logical fluid and cellular contents by M Noppen.)
adjacent chest wall. The central portion of the diaphragm
is innervated by the phrenic nerve, resulting in referred
pain to the ipsilateral shoulder tip during central
diaphragm irritation. TRANSPORT ACROSS THE MESOTHELIAL CELL
The visceral pleura is extensively innervated by pul- AND PLEURA
monary branches of the vagus nerve and sympathetic
trunk. However, the visceral pleural contains no pain Pleural fluid is produced by the parietal pleura, originating
fibers in contrast to the parietal pleura, and this is of clin- from the systemic circulation, and production occurs
ical relevance. The presence of pleuritic chest pain there- mostly in the less dependent region of the pleural cavity
fore usually indicates involvement of the parietal pleural in where blood vessels are closest to the mesothelial surface.
the disease process. Reabsorption of pleural fluid occurs mainly through lym-
phatic drainage in the most dependant part of the pleural
cavity, and again occurs exclusively on the parietal pleural
CONTENTS OF THE PLEURAL SPACE side. Drainage occurs from the parietal thoracic, mediasti-
nal and diaphragmatic surfaces (see section ‘The pleuro-
The volume and characteristics of liquid in the pleural lymphatic communication’).36,37
space is determined by a combination of dynamic phe- Fluid filters out of parietal pleural capillaries and into
nomenon, involving the pulmonary and systemic circula- the pleural space (or vice versa) according to the net
tion, the lymphatic drainage, the mechanical movement of hydrostatic–oncotic pressure gradient.37 As fluid in the
the thoracic cage and the movement of the heart.6 pleural space actively alters transpleural forces in respira-
The volume of pleural fluid in health is small. tion, optimal volume and thickness are closely main-
Experiments in rabbits have demonstrated a total pleural tained.37,38
fluid content of 0.2 mL, and previous data has suggested Water and molecules less than 4 nm in size can freely
that the normal human pleural fluid volume is less than pass between mesothelial cells. Intrapleural injections of
1 mL.29 In a subsequent study, the volume of pleural fluid hypotonic and hypertonic fluids have been demonstrated
collected from a single pleural cavity was 0.98 mL in the to induce an increase in the number and size of pinocyto-
rabbit and 2.35 mL in the dog.30 Noppen et al.,31 using an totic and cytoplasmic vesicles within mesothelial cells.39
interesting pleural lavage technique, estimated the mean Intrapleural injection of larger particles, for example fer-
pleural fluid volume in healthy human subjects under- ritin (11 nm), carbon (20–50 nm) and polystyrene (up to
going medical thoracoscopy and found the mean pleural 1000 nm) result in the appearance of these substances
fluid volume in the right pleural cavity of healthy humans within cytoplasmic vesicles of mesothelial cells.40 Smaller
to be 8 mL. The amount of pleural fluid may be related to particles are subsequently identified within the mesenchy-
physical excursion of the chest, with Yamada29 demon- mal cells of the pleural wall.20 Taken together, these find-
strating an increase in the volume of pleural fluid in mili- ings suggest that transcytoplasmic transport is active
tary recruits after exercise. within mesothelial cells.
The small volume of pleural fluid forms a thin film Movement of fluid in between the pleural space and the
between the visceral and parietal pleura, 10 mm thick at alveolar or pulmonary interstitium is restricted by the
least.32,33 This prevents contact between the visceral and presence of tight junctions between visceral mesothelial
parietal pleura throughout their surfaces. In support of cells, as previously described (section ‘Mesothelial cells’).
this, experiments in sheep have demonstrated no direct However, during disease states, such as congestive cardiac
The pleuro-lymphatic communication 19
failure or adult respiratory distress syndrome, both the Lymphatic circulation of the parietal pleura
endothelial and mesothelial barriers are damaged, permit-
ting alveolar and pulmonary interstitial fluid movement The lymphatic drainage of the parietal pleura varies
into the pleural space. In this context, the pleural space is according to region. In the costal parietal pleura, lym-
considered as one of the main important exits for lung phatic plexuses are confined to intercostal spaces and are
edema fluid.38 absent or minimal over the inner surface of the ribs.27
Particles greater in size than 1000 nm are engulfed by Lymph collected in the costal pleura drains ventrally
mesothelial cells but are not transported across the basal towards nodes in the internal mammary nodes, or dorsally
lamina.20 Effective removal of large particles or cells towards the intercostal lymph nodes near the heads of the
through the pleura is therefore unlikely, unless the basal ribs.
lamina plus or minus deeper layers of the pleura have been The lymphatic vessels of the mediastinal pleura are seen
damaged. However, the intrapleural injection of labeled in areas with abundant fatty tissue, the lymph collected
red blood cells results in their appearance in the systemic draining to the tracheobronchial and mediastinal nodes.
circulation intact,41,42 and large molecular weight proteins In more caudal areas, the mediastinal lymphatics are often
are absorbed rapidly through the pleural lymphatics.42 associated with Kampmeier’s foci (see section ‘Kampmeier’s
Therefore, communication passages between the pleural foci’). Lymphatic vessels from the diaphragmatic pleura
cavity and the circulation system must exist, which are drain into parasternal, middle phrenic and posteriorly
larger and faster than the cytoplasmic route (see section mediastinal nodes.
‘The pleuro-lymphatic communication’). The lymphatic system of the parietal pleura is more
extensive, more complex and less restricted in direction
and passage of flow than the visceral lymphatic system,
LYMPHATICS mirroring the arterial and venous supply. The parietal
lymphatics are therefore thought to play a key role in the
The lymphatics within the lung are divided into two formation and removal of pleural fluid. In normal and
systems: the superficial or pleural plexus, localized in the pathological states, the effective removal of fluid, cells and
subpleural connective tissue layer of the visceral pleura, and cellular debris relies upon the presence of the pleuro-
the deep plexus located in the bronchovascular bundles. lymphatic communications (see next section).
The deep plexus includes peribronchial, peripulmonary
vascular and interlobular septum or connectivity tissue.
Communications between the two plexuses exist only at the
junction of the pleura and the interlobular septum.5,15 THE PLEURO-LYMPHATIC COMMUNICATION
Figure 2.4 Two isolated round openings of the preformed paravertebral fatty tissue in the dorsocaudal region. In this
stomata (p) are shown in the subcostal region of the parietal study, no lymphatic stomata were found on the pleural
pleura. Mouse, scanning electron microscopy, ¥1680. (With surface of the diaphragm.50
permission from Wang NS. Morphological data of pleura – normal Other studies have demonstrated stomata in the pari-
conditions. In Chretien J, Hirsch H (eds). Diseases of the pleura. etal and diaphragmatic pleura, and stomata seem to be
New York: Masson Publishing USA, Inc, 1983: 10–24.) abundant in the diaphragmatic peritoneum of both
humans and animals.45–47,49 No study to date has demon-
strated stomas in the visceral pleura.
The process of formation of the pleural stomata is as yet
unknown. Studies in rats have suggested that parietal
stoma appear in the first few days of life post delivery.51
The diaphragmatic peritoneal stoma in rats are formed as
a result of the breakdown of intercellular junctions in both
the endothelial and mesothelial cell layers,52 and it seems
likely that the process in the parietal pleura is similar.
pleural and lymphatic surfaces are disrupted (Figures 2.7 membrane cribriformis has been measured at 7–60 mm in
and 2.8). diameter.56
Similar to the stomal distribution, the membrana crib-
riformis is abundant on the peritoneal diaphragmatic
surface,55 and has not been documented in the visceral
Lacuna and lymphatic channels
pleura of either humans or animals. In the rabbit, the
Beneath the stoma and membrana cribriformis is a lacuna,
which is the terminal dilatation of a lymphatic channel.
Each lacuna is connected at one end to the pleural cavity
by a small number of stomas, and drains via a lymphatic
channel with checking-valves at the other end (Figure
2.6).20,47
Movements of the lung and thoracic cage during respi-
ration alter the rate of removal of particles, cells and fluid
from the pleural cavity.39,41,47 During inspiration, the chest
wall expands and the intercostal spaces widen, resulting in
the stomas and lacunae being pulled open. Fluid and par-
ticles enter the lacuna by a combination of the negative
pressure generated within lacunae and the expanding lung
pushing on the pleural contents. During expiration, the
diameter of the stomas decrease as the chest wall contracts,
resulting in compression of the lacunae and expulsion of
Figure 2.7 The covering mesothelial cells are mostly broken in
fluid, particles and cells in to the draining lymphatic chan-
the mediastinal pleura of this patient with massive pleural
nels. Retrograde flow of material within the lymphatic
effusion, exposing the lamina cribriformis. The stomas between
channel is prevented during the next inspiratory move-
the collagen bundles are quite variable in size, as is the thickness
ment by the presence of the lymphatic valves. Stomas and
of the collagen bundles. Rupture of the thin bundle (arrow) may
lacunae appear to function in a similar manner on the
change the adjacent stomas into much larger fenestrae. Scanning
peritoneal diaphragmatic surface.46,55
electron microscopy, ¥8250. (With permission from Wang NS.
Morphological data of pleura – normal conditions. In Chretien J,
Hirsch H (eds). Diseases of the pleura. New York: Masson Kampmeier’s foci
Publishing USA, Inc, 1983: 10–24.)
In 1928, Kampmeier described small milky spots in the
dorsal and caudal portion of the human mediastinum.57 At
light microscopy, these foci are made up of modified
cuboidal mesothelial cells with stomas and are associated
with an aggregate of lymphocytes, histiocytes, plasma cells
and other mononuclear cells, located around a central
lymphatic or vascular vessel. The mesothelial cells have
increased cytoplasmic mass and granules, suggesting cellu-
lar activity (see section ‘The resting and reactive mesothe-
lial cell’). Under scanning electron microscopy, the foci
appear as irregular, elevated mound-like structures.20
Similar foci have been identified in the thoracic cavity of
dogs and in the mesentery of many species.58,59 It is postu-
lated that these structures act as local host defence mecha-
nisms, similar to the tonsils in the oropharynx or Peyer’s
patches in the gut. Infectious organisms and noxious par-
ticles may bypass these foci, and appear via the draining
lymphatic channels in the parasternal lymph nodes.60
Figure 2.8 Mesothelial cells (m) with microvilli and lymphatic
endothelial cells are in the process of disruption, or formation of The crevices or fenestrae
stomas or fenestrae, on the lamina cribriformis. The stretched
remnant of a lymphatic endothelial cell appears just broken Openings much larger than stomas, around 10–50 mm, have
(arrow). Debris of foreign particles (d) are present over a been documented in the mediastinal pleura of aged mice
lymphatic endothelial cell, which is almost intact. Human parietal and are called crevices.61 Similar structures have been
pleura, ¥3105. observed in rabbits (personal observation). The precise
22 Anatomy of the pleura
mechanism of their formation is uncertain. However, rare. Acquired defects are thought to arise from thinning
crevices are only found in areas where stomas are normally and eventual separation of collagenous fibres within the
present, and it is therefore postulated that they may be the tendinous part of the diaphragm.
result of fusion of adjacent stomas as a result of breakdown Small diaphragmatic defects usually only become appar-
of the collagen bundle meshwork, perhaps owing to diges- ent clinically in the context of ascites associated with pleural
tive enzyme release in pleuritis (Figure 2.8). This process is effusion, for example hepatic hydrothorax,67–69 continuous
analogous to the development of pulmonary emphy- ambulatory peritoneal dialysis70 or Meig’s syndrome. Such
sema,62,63 in which digestive enzyme-mediated breakdown defects may be demonstrated at thoracoscopy69, by the
of elastic fibers results in fusion of the interalveolar pores of injection of air70 or labeled tracer into the peritoneal space
Kohn. with subsequent chest radiology, or at autopsy.68
In the absence of diaphragmatic defects, fluid may enter
the pleural space from a peritoneal cavity distended with
The pleuro-lymphatic communication – does it fluid (or rarely vice versa), through reversal of the normal
exist? pressure gradient within the lymphatics and valve insuffi-
ciency in the thoracic duct and its attributes. Peritonitis or
Although several animal and human studies have demon- subphrenic abscess is a recognized cause of pleural reac-
strated the presence of stomas connecting the pleural tion, whereas lower lobe pneumonia resulting in intra-
cavity with the lymphatics,45–47,49,50 some studies have abdominal abscess is rare.
found no evidence of stomas in human pleura.64 This may
be due to their paucity, the limited availability of normal
human tissue for study, or obscuration of the stomas by THE REGIONAL DIFFERENCE
fibrin and cellular debris.47,53,65
Stomas are found readily in the peritoneal cavity,46,66 There are substantial morphological differences in regions
with one study demonstrating around 250 stomas per of the pleura. These differences include the mesothelial cell
mm2 of diaphragmatic peritoneum.66 In contrast, the characteristics (e.g. size and shape, density of microvilli),
pleural diaphragmatic density of stomas is as little as the pleural substructure and the number of pleuro-
1 per mm2 in small mammals47 and sheep.65 The reason for lymphatic communications, including the Kapmeier’s
this difference is unclear, but may reflect different foci.11,13,57
demands on the pleural and abdominal cavities. The
pleural space is sealed from the external environment in Visceral pleura
health, and relatively little fluid and cellular content
requires removal. The peritoneal cavity is likely to require
In the apical portion of the hemithorax, the visceral pleura
a greater capacity to remove fluid and cellular content, for
is relatively thin with flattened mesothelial cells and sparse
example as the result of ovulation.
microvilli, reflecting paucity of movement in the statically
Stomas may therefore be relatively unimportant in
expanded upper lung. Beneath the mesothelial cell layer,
health for the pleural cavity, but be recruited in disease
the basal lamina and deeper three layers are often difficult
states in which pleural fluid and cellular material clearance
to distinguish, especially in the apex where the systemic
becomes important. Inflammation and chronic pleural
arterial supply is replaced by a pulmonary supply (see
effusion appears to easily disrupt the thin mesothelial and
section ‘The blood supply of the visceral pleura’). This
lymphatic lining cells covering the lamina cribriformis47
thin pleura is the site of bleb formation seen in some
(Figures 2.7 and 2.8), although no cause and effect rela-
patients with spontaneous pneumothorax and is often the
tionship has been established. Stomas may increase in both
site of bullae formation in chronic obstructive pulmonary
size and number with age.61
disease. Rupture of these structures results in
pneumothorax.
In more basal areas where the lung moves and stretches
THE PLEURO-PERITONEAL COMMUNICATION more, the visceral pleura is thicker with cuboidal mesothe-
AND DIAPHRAGMATIC DEFECTS lial cells showing increased microvilli.11,13,14 The amount of
collagen and elastic fibers increases within the deeper
The lymphatic plexuses of the diaphragmatic pleura and
layers toward the lower part of the lung.
diaphragmatic peritoneum are separate, with communi-
cations between the two being poorly formed or infre-
quent. This is presumably an adaptation to prevent Parietal pleura
inflammatory or infected fluid from entering the pleural
space and interfering with respiration. Pleura overlying the inner rib surfaces is thin with flat-
Small diaphragmatic defects probably exist more often tened mesothelial cells, sparse microvilli and thin subcellu-
than is clinically suspected. Severe congenital diaphrag- lar layers. The dense fifth layer of fibroelastic tissue fuses
matic defects or anomalies may be fatal in utero, but are with the perichondrium or periosteum of the rib.
Key points 23
In areas of parietal pleura overlying loose substructures, SUBCLINICAL ALTERATIONS AND REPAIR OF
for example the mediastinum, the costophrenic recesses THE PLEURA
and the subcostal margins, lining mesothelial cells are
cuboidal and prominent. The second and third layers are Pleural effusion or pleuritis may occur and regress sponta-
well defined, whereas the fourth layer is often merged with neously without the need for pleural intervention, for
a deeper and wider interstitial space, which contains a example in heart failure, pulmonary infarction and some
poorly formed or absent fifth layer. This loose fourth layer cases of parapneumonic effusion. The course and mecha-
often serves as the cleavage plane in pleurectomy. nism of spontaneous resolution in pleural inflammation is
The pleura over the diaphragm and intercostal muscles not clear.80 Experimental animal studies demonstrate that
is of moderate thickness, with characteristics somewhere mesothelial cells become reactive, proliferate and migrate
between pleura overlying ribs and pleura overlying looser in response to injury. These changes appear to facilitate the
structures. The underlying tissue covered by the pleura removal of fibrin and inflammatory debris and to allow
therefore influences the cellular and non-cellular compo- repair of the pleural surface, maintaining the integrity of
nents of the pleural layers. the pleural cavity while preserving drainage.75,81,82
Albertine et al.65 showed that unlike the varied thick- Whether the pleura recovers completely or progresses to
ness and appearance of the visceral pleura, the parietal the development of fibrosis appears to be related to the
pleura in sheep has a relatively uniform thickness over degree of damage to the basal lamina.83
chest wall, diaphragm and mediastinum. There was a sig- Human CT and MRI studies suggest that unsuspected
nificant difference in the distance from capillary to pleural pleural lesions are common, especially in smokers.84
surface between parietal (10–12 mm) and visceral Pleural changes have been observed in patients with pneu-
(18–56 mm, depending on region of the lung studied) monitis, lung cancer and myocardial ischaemia, without
pleura.65 Transportation of fluid and large molecules is clinical or basic radiological evidence of pleural disease.80
therefore anatomically easier through the parietal side. It is possible that minor damage and subsequent repair
occur in the pleura frequently without any clinical mani-
festation and, in this case, the reactive and reparative prop-
THE RESTING AND REACTIVE MESOTHELIAL erties of the mesothelial cell are important.
CELL
REFERENCES 25. Clemente CD. Anatomy of the human body, 30th edn. Philadelphia:
Lea & Febiger, 1985.
26. Davila RM, Crouch EC. Anatomic organisation and function of the
● = Key primary paper human pleura. Semin Respir Crit Care Med 1979; 82: 783–97.
◆ = Major review article 27. Bernaudin JF, Fleury JY. Anatomy of the blood and lymphatic
circulation of the pleural serosa. In: Chretien J, Bignon J, Hirsch A
1. West JB, Fu Z, Gaeth AP, Short RV. Fetal lung development in the (eds). The pleural in health and disease. New York: Marcel Dekker;
elephant reflects the adaptations required for snorkeling in adult 1985: 101–24.
life. Respir Physiol Neurobiol 2003; 138: 325–33. 28. Milne ENC, Pistolesi M. Reading the chest radiograph: a physiologic
2. West JB. Snorkel breathing in the elephant explains the unique approach. St Louis: Mosby; 1993.
anatomy of its pleura. Respir Physiol 2001; 126: 1–8. ●29. Yamada S. Uber die serose Flussigkeit in der Pleurahohle der
3. Patten BM, Carlson BW. Foundations of embryology, 3rd edn. New gesunde Menschen. Z Ges Exp Med 1933; 90: 342–8.
York: McGraw-Hill; 1974. ●30. Miserocchi G, Agostoni E. Contents of the pleural space. J Appl
●4. Hesseldahl H, Larsen JF. Ultrastructure of human yolk sac: Physiol 1971; 30: 208–13.
endoderm, mesenchyme, tubules and mesothelium. Am J Anat ●31. Noppen M, De Waele M, Li R, et al. Volume and cellular content of
1969; 126: 315–35. normal pleural fluid in humans examined by pleural lavage. Am J
◆5. von Hayek H. The parietal pleura and visceral pleura. New York: Respir Crit Care Med 2000; 162: 1023–6.
Hafner; 1960. 32. Butler JP, Huang J, Loring SH, et al. Model for a pump that drives
6. Pistolesi M, Miniati M, Giuntini C. Pleural liquid and solute circulation of pleural fluid. J Appl Physiol 1995; 78: 23–9.
exchange. Am Rev Respir Dis 1989; 140: 825–47. 33. Agostoni E, Miserocchi G, Bonanni MV. Thickness and pressure of
7. Rabinowitz JG, Cohen BA, Mendleson DS. Symposium on the pleural liquid in some mammals. Respir Physiol 1969; 6:
nonpulmonary aspects in chest radiology. The pulmonary ligament. 245–56.
Radiol Clin North Am 1984; 22: 659–72. ●34. Albertine KH, Wiener-Kronish JP, Bastacky J, Staub NC. No
8. Morrissey BM, Bisset RA. The right inferior lung margin: anatomy evidence for mesothelial cell contact across the costal pleural
and clinical implication. Br J Radiol 1993; 66: 503–5. space of sheep. J Appl Physiol 1991; 70: 123–34.
9. Satoh K, Sato A, Kobayashi T, et al. Septal structure of incomplete ●35. Sahn SA, Willcox ML, Good JT Jr, Potts DE, Filley GF.
interlobar fissures of the lung. Acad Radiol 1996; 3: 475–8. Characteristics of normal rabbit pleural fluid: physiologic and
10. Webb WR, LaBerge JM. Radiographic recognition of chest tube biochemical implications. Lung 1979; 156: 63–9.
malposition in the major fissure. Chest 1984; 85: 81–3. 36. Miserocchi G. Physiology and pathophysiology of pleural fluid
●11. Wang NS. The regional difference of pleural mesothelial cells in turnover. Eur Respir J 1997; 10: 219–25.
rabbits. Am Rev Respir Dis 1974; 110: 623–33. ◆37. Staub NC. New concepts about the pathophysiology of pulmonary
12. Michailova KN. The serous membranes in the cat. Electron edema. J Thorac Imaging 1988; 3: 8–14.
microscopic observations. Ann Anat 1996; 178: 413–24. 38. Miserocchi G, Venturoli D, Negrini D, Del Fabbro M. Model of
13. Mariassy AT, Wheeldon EB. The pleura: a combined light pleural fluid turnover. J Appl Physiol 1993; 75: 1798–806.
microscopic, scanning, and transmission electron microscopic 39. Shumko JZ, Feinberg RN, Shalvoy RM, DeFouw DO. Responses of
study in the sheep. I. Normal pleura. Exp Lung Res 1983; 4: rat pleural mesothelia to increased intrathoracic pressure. Exp
293–314. Lung Res 1993; 19: 283–97.
●14. Albertine KH, Wiener-Kronish JP, Roos PJ, Staub NC. Structure, ◆40. Wang NS. Morphological data of pleura – normal conditions. In:
blood supply, and lymphatic vessels of the sheep’s visceral pleura. Chretien J, Hirsch A (eds). The pleural in health and disease. New
Am J Anat 1982; 165: 277–94. York: Masson Publishing USA Inc; 1983: 10–24.
15. Nagaishi C. Functional anatomy and histology of the lung. Tokyo: ◆41. Courtice FC, Simmonds WJ. Physiological significance of lymph
Igaku Shoin; 1972. drainage of the serous cavities and lungs. Physiol Rev 1954; 34:
16. Michailova K, Wassilev W, Wedel T. Scanning and transmission 419–48.
electron microscopic study of visceral and parietal peritoneal 42. Wilson JL, Herrod CM, Searle GL, et al. The absorption of blood
regions in the rat. Ann Anat 1999; 181: 253–60. from the pleural space. Surgery 1960; 48: 766–74.
17. Odor DL. Observations of the rat mesothelium with the electron ●43. von Recklinghausen FV. Zur Fettresorption. Virchow Arch (Pathol
and phase microscopes. Am J Anat 1954; 95: 433–65. Anat) 1863; 26: 172–278.
●18. Legrand M, Pariente R, Andre J, Chretien J, Brouet G. 44. Dybkowsky. Ueber Aufsaugang und Absonderung der Pleurawand.
[Ultrastructure of the human parietal pleura]. Presse Med 1971; Ber d Kgl Sachs Gesellsch d Wissensch Math-physik Kl 1866; 18:
79: 2515–20. 191–218.
●19. Andrews PM, Porter KR. The ultrastructural morphology and 45. Wheeldon EB, Mariassy AT, McSporran KD. The pleura: a combined
possible functional significance of mesothelial microvilli. Anat Rec light microscopic and scanning and transmission electron
1973; 177: 409–26. microscopic study in the sheep. II. Response to injury. Exp Lung
◆20. Wang NS. Mesothlelial cells in situ. In: Chretien J, Bignon J, Hirsch Res 1983; 5: 125–40.
A (eds). The pleura in health and disease. New York: Marcel Dekker, 46. Tsilibary EC, Wissig SL. Absorption from the peritoneal cavity: SEM
1985: 23–42. study of the mesothelium covering the peritoneal surface of the
21. Inoue T, Osatake H. Three-dimensional demonstration of the muscular portion of the diaphragm. Am J Anat 1977; 149:
intracellular structures of mouse mesothelial cells by scanning 127–33.
electron microscopy. J Submicrosc Cytol Pathol 1989; 21: 215–27. ●47. Wang NS. The preformed stomas connecting the pleural cavity and
●22. Simionescu M, Simionescu N. Organization of cell junctions in the the lymphatics in the parietal pleura. Am Rev Respir Dis 1975;
peritoneal mesothelium. J Cell Biol 1977; 74: 98–110. 111: 12–20.
23. Nomura K, Kida K, Kudoh S. [A morphological study to elucidate 48. Miura T, Shimada T, Tanaka K, Chujo M, Uchida Y. Lymphatic
the differences in visceral pleura in young and old mice]. Nippon drainage of carbon particles injected into the pleural cavity of
Ika Daigaku Zasshi 1998; 65: 227–35. the monkey, as studied by video-assisted thoracoscopy and
24. Madison LD, Bergstrom-Porter B, Torres AR, Shelton E. Regulation electron microscopy. J Thorac Cardiovasc Surg 2000; 120:
of surface topography of mouse peritoneal cells. Formation of 437–47.
microvilli and vesiculated pits on omental mesothelial cells by 49. Li J. Ultrastructural study on the pleural stomata in human. Funct
serum and other proteins. J Cell Biol 1979; 82: 783–97. Dev Morphol 1993; 3: 277–80.
References 25
50. Shinohara H. Distribution of lymphatic stomata on the pleural Distribution of diaphragmatic lymphatic lacunae. J Appl Physiol
surface of the thoracic cavity and the surface topography of the 1992; 72: 1166–72.
pleural mesothelium in the golden hamster. Anat Rec 1997; 249: 67. Emerson PA, Davies JH. Hydrothorax complicating ascites. Lancet
16–23. 1955; 268: 487–8.
51. Ohtani O, Ohtani Y, Li RX. Phylogeny and ontogeny of the ◆68. Alberts WM, Salem AJ, Solomon DA, Boyce G. Hepatic
lymphatic stomata connecting the pleural and peritoneal cavities hydrothorax. Cause and management. Arch Intern Med 1991; 151:
with the lymphatic system – a review. Ital J Anat Embryol 2001; 2383–8.
106: 251–9. 69. Nakamura A, Kojima Y, Ohmi H, Yamada J, Yamada Y. Peritoneal-
52. Nakatani T, Tanaka S, Mizukami S, et al. Peritoneal lymphatic pleural communications in hepatic hydrothorax demonstrated by
stomata of the diaphragm in the mouse: process of their thoracoscopy. Chest 1996; 109: 579–81.
formation. Anat Rec 1997; 248: 121–8. 70. Nomoto Y, Suga T, Nakajima K, et al. Acute hydrothorax in
53. Miura T, Shimada T, Tanaka K, Chujo M, Uchida Y. Lymphatic continuous ambulatory peritoneal dialysis – a collaborative study
drainage of carbon particles injected into the pleural cavity of the of 161 centers. Am J Nephrol 1989; 9: 363–7.
monkey, as studied by video-assisted thoracoscopy and electron 71. Whitaker D, Papadimitriou JM, Walters MN. The mesothelium: a
microscopy. J Thorac Cardiovasc Surg 2000; 120: 437–47. histochemical study of resting mesothelial cells. J Pathol 1980;
●54. Kihara T. Das extravasculare Saftbahn System. Okazima Fol Anat 132: 273–84.
Jpn 1956; 28: 601–21. 72. Hott JW, Sparks JA, Godbey SW, Antony VB. Mesothelial cell
55. Oya M, Shimada T, Nakamura M, Uchida Y. Functional morphology response to pleural injury: thrombin-induced proliferation and
of the lymphatic system in the monkey diaphragm. Arch Histol chemotaxis of rat pleural mesothelial cells. Am J Respir Cell Mol
Cytol 1993; 56: 37–47. Biol 1992; 6: 421–5.
56. Li YY, Li JC. Ultrastructure and three-dimensional study of the 73. Whitaker D, Papadimitriou JM, Walters MN. The mesothelium: a
lymphatic stomata in the costal pleura of the rabbit. Microsc Res cytochemical study of ‘activated’ mesothelial cells. J Pathol 1982;
Tech 2003; 62: 240–6. 136: 169–79.
●57. Kampmeier OF. Concerning certain mesothelial thickenings and 74. Coene MC, Van Hove C, Claeys M, Herman AG. Arachidonic acid
vascular plexus of the mediastinal pleura associated with metabolism by cultured mesothelial cells. Different
histiocyte and fat cell production in the human newborn. Anat Rec transformations of exogenously added and endogenously. Biochim
1928; 39: 201–8. Biophys Acta 1982; 710: 437–45.
58. Cooray GH. Defensive mechanisms in the mediastinum with 75. Whitaker D, Papadimitriou JM, Walters M. The mesothelium: its
special reference to the mechanics of pleural absorption. J Pathol fibrinolytic properties. J Pathol 1982; 136: 291–9.
Bacteriol 1949; 61: 551–67. ◆76. Ryan GB, Grobety J, Majno G. Mesothelial injury and recovery. Am
59. Lang J, Liebich HG. Uber eigenartige Kapillarkonvolute der Pleural J Pathol 1973; 71: 93–112.
parietalis. III. Elektronenmikroskopische Untersuchungen. Z 77. Gabrielson EW, Gerwin BI, Harris CC, et al. Stimulation of DNA
Mikrosk-Anat Forsch 1976; 9: 1092. synthesis in cultured primary human mesothelial cells by specific
60. Burke HE, Wilson JA. A new method for establishing the diagnosis growth factors. FASEB J 1988; 2: 2717–21.
of pleural disease – parasternal lymph node biopsy. Am Rev Respir 78. Adamson IY, Bakowska J, Bowden DH. Mesothelial cell
Dis 1966; 93: 201–8. proliferation: a nonspecific response to lung injury associated with
◆61. Kanazawa K. Exchange through the pleura. Cells and particles. In: fibrosis. Am J Respir Cell Mol Biol 1994; 10: 253–8.
Chretien J, Bignon J, Hirsch A (eds). The pleura in health and 79. Adamson IY, Bakowska J, Bowden DH. Mesothelial cell
disease. New York: Marcel Dekker; 1985: 195–231. proliferation after instillation of long or short asbestos fibers into
62. Wang NS. Scanning electron microscopy of the lung. In: Lenfant C, mouse lung. Am J Pathol 1993; 142: 1209–16.
Schraunagel DE (eds). Electron micsroscopy of the lung: lung 80. Peng MJ, Wang NS, Vargas FS, Light RW. Subclinical surface
biology in health and disease. New York: Marcel Dekker; 1990: alterations of human pleura. A scanning electron microscopic
517–55. study. Chest 1994; 106: 351–3.
63. Boren HG. Alveolar fenestrae. Relationship to the pathology and 81. Whitaker D, Papadimitriou J. Mesothelial healing: morphological
pathogenesis of pulmonary emphysema. Am Rev Respir Dis 1962; and kinetic investigations. J Pathol 1985; 145: 159–75.
85: 328–44. 82. Watters WB, Buck RC. Scanning electron microscopy of
64. Gaudio E, Rendina EA, Pannarale L, Ricci C, Marinozzi G. Surface mesothelial regeneration in the rat. Lab Invest 1972; 26: 604–9.
morphology of the human pleura. A scanning electron microscopic 83. Davila RM, Crouch EC. Role of mesothelial and submesothelial
study. Chest 1988; 93: 149–53. stromal cells in matrix remodeling following pleural injury. Am J
●65. Albertine KH, Wiener-Kronish JP, Staub NC. The structure of the Pathol 1993; 142: 547–55.
parietal pleura and its relationship to pleural liquid dynamics in ◆84. Kohda E, Suzuki K, Tanaka M, et al. [Radiological approach to the
sheep. Anat Rec 1984; 208: 401–9. pleura and pleural cavity with CT and MRI]. Nihon Kyobu Shikkan
66. Negrini D, Del Fabbro M, Gonano C, Mukenge S, Miserocchi G. Gakkai Zasshi 1994; 32(Suppl): 148–54.
This page intentionally left blank
3
Mesothelial cells
Introduction 27 Conclusions 34
Mesothelial cell morphology 27 Key points 34
Regulation of mesothelial cell proliferation 30 References 35
Physiopathology of mesothelial cells 32
sarcomatoid and biphasic forms,12 which are also referred mesothelial cells. Coexpression of cytokeratins and
to as epithelial, sarcomatous, or mixed subtypes, vimentin is characteristic of MM cells. Other membrane
respectively. These different features reflect the plasticity proteins – mesothelin, ME1 and HBME1 – are also present
of mesothelial cells. In humans, epithelial subtype is on these cells.
predominant (approximately 55 percent) followed by
sarcomatous (approximately 22 percent) and mixed
(approximately 24 percent).13 The typical light microscopy Mesothelial cells in culture
appearance of epithelioid mesothelioma is ill-defined
tubular, papillary and loose solid nests of epithelial cells. Pleural mesothelial cells have been successfully explanted
The well-differentiated mesothelioma cells have similar and cultured. Early studies using imprint technique have
phenotypes to normal cells, exhibiting cubic, polygonal, or permitted identification of some of the proteins expressed
flattened morphologies.14,15 by mesothelial cells.20 Rodent mesothelial cells can prolif-
The ultrastructural features of mesothelioma cells are erate in culture and showed an ability to phagocytose
characterized in the epithelia form by the presence of asbestos fibres.21 In culture, mesothelial cells form a
brush-like elongated, slender and branching microvilli monolayer with cobblestone morphology. In vitro,
(Figure 3.2).16,17 The microvilli have a length to width ratio mesothelial cells maintain their morphological character-
of approximately 12 within the range of that of normal istics, i.e. presence of microvilli, perinuclear bundles of
cells.18 Microvilli are found on the luminal surface, in the intermediate filaments and desmosomes.22 Cell dedifferen-
intercellular space, and also on the abluminal surface tiation occurs following a large number of population
where, if abundant enough, they can disrupt the basal doublings. In rats, after approximately 40 passages
lamina.19 Mesothelial characteristics of abundant cytoplas- (approximately 100 doublings), phenotypical changes
mic perinuclear intermediate filaments, multiple junctions appear in the culture layer, as foci of cells with looser con-
including desmosomes, are found in epithelial malignant tacts and loss of contact inhibition. However, these
mesothelial cells. In the sarcomatous cells, these features changes are not associated with the acquisition of tumori-
are less typical or absent. Nevertheless, microvilli and genic potency, as assessed by inoculation in nude mice.23
desmosomes may be occasionally found by TEM analysis. More sophisticated culture medium, supplemented
Immunohistochemical analysis of MM have demon- with transferrin, insulin and growth factors, has been
strated the presence of antigens usually present in normal made to grow human pleural mesothelial cells from non-
cancerous pleural fluids.24 These systems of cultured
mesothelial cells have helped researchers to explore the
biological and physiological features of mesothelial cells,
especially their participation in inflammatory processes
and responses to asbestos fiber exposure. Long-term cul-
tures for periods up to 5–6 months can be obtained by
transfection of human normal pleural mesothelial cells
with a plasmid containing the coding sequence of the
simian virus (SV40) early region DNA.25 These cells were
not tumorigenic when injected subcutaneously in nude
mice, even after a delay of 1 year.25 One cell line (Met 5A)
generated in this study is frequently used as a surrogate of
normal pleural mesothelial cells.
Similarly, MM cells from tumors have been grown in
culture. All mesothelioma cell lines showed strong
immunoreactivity to anti-cytokeratin and -vimentin anti-
bodies, and did not express carcinoembryonic antigen,
factor VIII or Leu-M1 (Figure 3.3). Cultured cells of
epithelial and sarcomatous morphological phenotypes
both co-expressed cytokeratins and vimentin.26 Ultra-
structural features of mesothelioma cells in culture show a
variety of differentiation status. While some cell lines
exhibited well-differentiated microvilli, abundant perinu-
clear bundles of intermediate filaments and typical junc-
tions, less differentiated cell lines were also found. These
Figure 3.2 Transmission electron micrograph of human malignant differences may be related to the original location of the
mesothelial cells in culture. Mesothelial differentiation is mesothelial cells in the pleural space or to culture condi-
demonstrated particularly by the occurrence of typical microvilli tions. Interestingly, when injected subcutaneously in nude
(䊳) (¥3276). mice, some cell lines demonstrated higher differentiation
30 Mesothelial cells
remains that epithelial mesothelial cells have the capability and growth factors. EGF, basic FGF (bFGF) and PDGF
to transdifferentiate from an epithelial to mesenchymal have been shown to stimulate mesothelial cell proliferation
phenotype. This potential is illustrated in repair processes in vivo and in vitro.24,40–43 Cytokines such as TNF-a, trans-
and in tumorigenesis, as it is not rare to observe not only forming growth factor (TGF)-b and IL-1 stimulate
sarcomatous differentiation, but also osseous and cartilagi- mesothelial cells to produce growth factors.31 TGF-b exerts
nous differentiation in human MM.38 Murine models of inhibitory effect on normal pleural mesothelial (NPM) cell
MM also exhibit these differential features.39 proliferation (Figure 3.4).44
Normal mesothelial cells from different individual
donors have differential responses to growth factors, e.g.
Growth factors regulating cell growth EGF.24,42 Normal human mesothelial cells express low
levels of mRNA of PDGF-A chain and the mRNA for
Proliferation of pleural mesothelial cells has been found to PDGF-b was not detectable.40 Messenger RNA transcripts
be modulated by several growth factors, including fibrob- for insulin-like growth factor (IGF)-I and its receptor,
last growth factor (FGF), hepatocyte growth factor (HGF), IGF-I receptor, were also expressed.45 IGF-I appears to
keratinocyte growth factor (KGF), platelet-derived growth function as an autocrine growth stimulus in human
factor (PDGF) and vascular endothelial growth factor mesothelial cells. IL-6 is another auto-regulatory factor in
(VEGF). Productions of, and response to growth factors, human mesothelial cell proliferation. A complete
have been investigated in the context of inflammatory autocrine loop was demonstrated by the expression of
processes following pleural injury, especially to explain the IL-6 mRNA transcripts and IL-6 receptor subunits.46
mechanism of pleural fibrosis. During the inflammatory Platelet-derived growth factor-BB- or TGF-b1-stimulated
process, macrophages release several factors such as normal human mesothelial cells express hyaluronan syn-
hyaluronic acid, interleukin (IL)-1b, tumor necrosis factor thase and hyaluronan, an important constituent of the
(TNF)-a and interferon (IFN)-g, which in turn may stim- extracellular matrix.47,48 Hyaluronan does not appear to be
ulate mesothelial cells to produce cytokines, chemokines involved in proliferation of normal mesothelial cells since
TNF- ␣
IGF-I
TGF- HGF/SF IL-6 IG F-BP3
IGF-II
PAI-1
Fn PAI-2
Lam
Elast TF
Tn Normal cell
⫹ EGF
Coll
HA Met
IGF-R PDGF
ECM IL6-R IGF-I
EGF-R PDGF IGF-R II
␣-R
VEGF-R PDGF-R
HA
IGF-BP3
CD44
IL-6
GM-CSF
IL-8
G-CSF
TGF- M-CSF
Figure 3.4 Schematic representation of the factors produce by mesothelial cells and receptors (human, 䡲; rat, ▫; both species, 䡲) present
in normal and malignant mesothelial cells. See text for details. Thin arrows indicate modulation of the expression of one factor (➝) by
another one (●). Coll, collagen; ECM, extracellular matrix; EGF, epidermal growth factor; EGF-R, epidermal growth factor receptor; Elast,
elastin; Fn, fibronectin; G-, M- and GM-CSF, granulocyte-, macrophage- and granulocyte/macrophage colony stimulating factors,
respectively; HA, hyaluronan; HGF/SF, hepatocyte growth factor/scattering factor; IGF-I, insulin-like growth factor receptor; IL-6,
interleukin 6; IL-8, interleukin 8; Lam, laminine; PAI, plasminogen activator inhibitor; PDGF, platelet-derived growth factor; PDGF-R,
platelet-derived growth factor receptor; TF, tissue factor; TGF-b, transforming growth factor b; Tn, tenascin; TNF-a, tumor necrosis factor
a; VEGF-R, vascular endothelial growth factor receptor.
32 Mesothelial cells
they do not express the hyaluronan receptor CD44.49 Human MM cells produce mRNA transcripts for IGF-
Production of VEGF and its receptors is another feature of I, IGF-binding protein 3 and of IGF-I receptor; the expres-
human mesothelial cells.50,51 In contrast, no expression of sions of IGF-I and IGF-binding protein 3 at a protein level
scattering factor HGF/(SF) and the c-met receptor were have also been confirmed.45 IGF-I appears to be an impor-
detected in normal mesothelial cells.52 tant regulator of MM cell growth (Figure 3.4).
Normal mesothelial cells from the rat possess EGF The HGF/SF and c-Met pathway could be involved in
receptors but the effect of EGF appears to depend on an autocrine fashion in the control of MM cell prolifera-
culture conditions, as proliferation is inhibited in serum tion. Several studies have demonstrated the expression of
free medium.35 Like human cells, rat mesothelial cells HGF/SF and c-Met in MM cell lines, and in paraffin sec-
produce IGF-I and IGF-II and autocrine growth stimula- tions.52,62–67 Nevertheless, expression of HGF/SF could be
tion by IGF-II has been suggested.53 Other growth factors, limited to spindle-shaped cells.63 In addition to the cell
PDGF, HGF and KGF, have also been demonstrated to growth regulation, HGF/SF exerts stimulatory effect on the
stimulate the proliferation of rat mesothelial cells.54,55 motility, spreading and proliferation of MM cells and
Thus, normal pleural mesothelial cells have the capabil- stimulated the expression of matrix metallopro-
ity to produce different growth factors which may in turn teinases.52,63,65
induce environment-dependent biological responses from Vascular endothelial growth factor plays a role in the
the mesothelial cells. The mesothelial cells are also able to growth of mesothelioma. MM cells express both VEGF
produce and respond to various inflammatory factors. and VEGF receptors [fms-like tyrosine kinase (Flt-1) or
These functions confirm that mesothelial cells play a fetal liver kinase (Flk-1)], and their expression appears to
crucial role in the maintenance of the pleural integrity. be enhanced in MM cells when compared with benign
To date, the mechanisms whereby mesothelial cells cells.51 An autocrine role of VEGF has been suggested,
survive, or are committed to apoptosis, remain unknown. using neutralizing antibodies against VEGF or the VEGF
Recent data suggested a list of inhibitors of apoptosis receptors Flt-1 and Flk-1.50
expressed in a primary human mesothelial cell line.56 Gene Malignant mesothelioma cells are resistant to apopto-
and protein analyses revealed inhibitors of apoptosis sis, but the mechanism remains to be elucidated. The anti-
(IAPs), including IAP-1, IAP-2, XIAP, survivin and livin. apoptotic factor, Bcl-x, seems to play a role in this
Immunohistochemistry and Western blot analysis have resistance.68–70 In MM cells, survival could be promoted by
demonstrated the production of IAP-1, IAP-2 and XIAP.56 an activation of the AKT/PKB pathway.71
(VCAM)-1, E-cadherin and several types of selectins.29,73 mesothelial cells also ingest asbestos fibers.82,83 Fiber uptake
Release of growth factors TGF-b, b-FGF, PDGF and EGF is facilitated by the fiber coating with vitronectin, a serum
stimulate cell proliferation and mesothelium regeneration. component interacting with integrin receptors present in
TGF-b is a multifunctional cytokine that may act as a the mesothelial cell membrane.84 Several forms of integrins,
chemotactic factor for lung fibroblasts that play a role in alpha2, alpha3, alpha 5, beta1, beta3 and alpha vbeta3, have
the synthesis of collagen and the development of fibrotic been found to be highly expressed in cultures of primary
processes, and TGF-b2 and -b3 have been shown to upreg- pleural mesothelial cells.85 In culture, the interaction
ulate mesothelial cell collagen expression.74,75 TGF-b between mesothelial cells and asbestos fibers results in the
enhances pleural fluid formation in part by inducing the adaptation to oxidative stress, activation of signalling path-
production of VEGF; it stimulates mesothelial cells to syn- ways and the release of cytokine and growth factor mediators
thesize extracellular matrix components as well as to already mentioned. Reactive oxygen species (ROS) and reac-
produce matrix metalloproteinases and inhibitors of met- tive nitrogen species (RNS) are generated in cells exposed to
alloproteinases.44,75 Pleural mesothelial cells may be acti- asbestos.86 Activation of signaling pathways involved the
vated to release inflammatory cytokines in vitro and to activation of transcription factors, such as nuclear factor
induce neutrophil recruitment in vivo by the stimulation (NF)-kB and AP-1, that control expression of other genes,
of proteinase-activated receptor-2, a novel seven-trans- including those governing cell survival.87 In serum-free con-
membrane receptor with immunomodulatory roles.76 ditions, EGF receptor appears to be activated by asbestos
Transforming growth factor b also regulates the coagu- fibers, leading to signaling via the extracellular signal-regu-
lation process, suppressing fibrinolysis by reduction of lated kinase (ERK) mitogen activated protein (MAP) kinase
plasminogen activator production and stimulation of pathway.87 In addition to production of ROS and RNS,
mesothelial cells to secrete plasminogen activator cytokines and chemotactic factors are also expressed on
inhibitors. In addition, TGF-b can also induce angiogene- mesothelial cells exposed to asbestos.88–90
sis via synthesis of VEGF by mesothelial cells.51 VEGF is a Exposure of mesothelial cells to asbestos fibers results in
key mediator in pleural effusion formation.77 cell injury that may be responsible for the carcinogenicity
Platelet-derived growth factor is a growth factor for of the fibers. Fiber internalization can impair the progres-
fibroblasts and, as mentioned above, PDGF is mitogenic sion of mitosis, as demonstrated by the formation of
for mesothelial cells. Basic FGF has been shown to stimu- micronuclei, bi- and multi-nucleated mesothelial cells,
late growth of both fibroblasts and mesothelial cells. induction of anaphase/telophase abnormalities and alter-
Mesothelial cells play an important role in the regulation ation of cytokinesis.91,92 DNA damage, resulting from the
of fibrin deposition in the pleura. Normal pleural mesothe- production of ROS, has been demonstrated using methods
lial cells exhibit procoagulant activity due to their ability to allowing the detection of base oxidation and DNA break-
release tissue factor, which initiates the coagulation cascade age.92,93 Indirectly, DNA damage has been demonstrated
and the deposition of transitional fibrin. This process is by the occurrence of DNA repair processes and activation
facilitated by the production of plasminogen activator of cell cycle control checkpoints.92,94,95 Liu et al.96 have
inhibitors (PAI), PAI-1 and PAI-2.73,78 Mesothelial cells also reported that selective decrease in fiber uptake reduced the
secrete components of the fibrinolysis system. Fibrinolytic adverse effects of asbestos, including DNA strand break-
activity is mediated by tissue plasminogen activator (tPA) age. These results indicate that the phagocytosis function
and urokinase PA (uPA) that activates plasminogen into of mesothelial cells is important and may contribute to the
plasmin, which degrades fibrin. Mesothelial cells are the asbestos-induced fibrogenesis and cancer development in
main source of tPA in serosal cavities but also secretes lower the pleura. (The molecular genetics of mesothelioma are
levels of uPA.29 (See also Chapter 9, Pleural fibrosis, for also discussed in Chapter 11.)
details on fibrinolytic pathways in the pleura.) Talc is a mineral used to induce pleurodesis in patients
with malignant pleural effusions; its introduction in the
pleural cavity produces pleural symphysis. Mesothelial
Phagocytosis cells stimulated with talc were found to release higher
amount of b-FGF than untreated cells, in agreement with
Mesothelial cells have been shown to phagocytose mineral the fibrogenic process described above.97 Use of talc pleu-
particles such as asbestos fibers, talc and quartz, as well as rodesis remains controversial as it prevents further use of
glass microbeads.72 It has been demonstrated that asbestos intrapleural therapy, a strategy that may be of interest for
fibres can be translocated from the lung to the pleura; hence treatment of MM. (The clinical management of mesothe-
mesothelial cells can interact directly with fibers and lioma is detailed in Chapter 41, Malignant mesothelioma.)
respond to circulating mediators.79–81 (See also Chapter 10,
Pleural reaction to mineral dusts.) The response of mesothe- Somatic genetic changes in malignant
lial cells has been largely studied in the context of fibrogenic mesothelioma cells
and carcinogenic potency of the fibers. Internalization of
asbestos fibers by rat mesothelial cells results in a lysosomal Cytogenetic and loss of heterozygozity analysis of MM
degranulation in the phagocytic vacuole.21 Human cells have shown the occurrence of multiple abnormalities
34 Mesothelial cells
and frequent deletions (see also Chapter 11, Genetics of regulating inflammatory responses, survival and apoptosis
malignant mesothelioma). Numerical and structural chro- in normal mesothelial cells. Determination of the somatic
mosome changes are shown in chromosomes 1, 3, 4, 6, 9, genetic mutations in MM cells, and understanding the
13, 15 and 22.98,99 These cytogenetic abnormalities could pathophysiology of alterations of cell regulation pathways,
affect various initiation/progression steps of neoplastic may explain how mesothelial cells acquire a neoplastic
progression. Candidate tumor suppressor genes in MM phenotype, and thereby allow better prevention and treat-
include P16/CDKN2A, P14/ARF and P15/CDKN2B, ment of pleural diseases.
located at the INK4 locus on chromosome 9, a frequently
deleted region. In contrast, the TP53 tumour suppressor
gene is not frequently mutated.100–102
Malignant transformation of mesothelial cell is associ-
ated with abolishment of the G1 to S regulation of transi-
tion which may play a role in the genetic instability found KEY POINTS
in cancer cells. Nevertheless, MM cells appear to maintain
the ability to control DNA damage, as the p53 protein
remains wild type in most cases. ● The pleural mesothelium consists of a single-
NF2 is another candidate tumor suppressor gene whose layer epithelial sheet over a basal lamina, sup-
alterations have been found in approximately 50 percent ported by the submesothelial connective tissue.
of mesotheliomas.103–105 This gene product helps link pro- The apical surface of mesothelial cells shows het-
teins from the cytoskeleton to membrane proteins. It plays erogeneity from numerous randomly-oriented
a role in the stabilization of adherens junctions. So far, the long microvilli to relatively few microvilli. The
role of the NF2 protein in the physiology of mesothelial cytoplasm of mesothelial cells contains scattered
cells is unknown but it may hold specific pathophysio- microtubules, dispersed microfilaments and
logical functions. bundled intermediate filaments. Numerous plas-
The exact steps involved in the neoplastic transforma- malemmal vesicles suggest intense pinocytic and
tion of mesothelial cells are unknown. However, inactiva- intracellular trafficking activity of mesothelial
tion of the tumor suppressor genes P16/CDKN2A and cells.
P15/CDKN2B has been proposed as an early event, fol- ● The mesothelium acts as a barrier for exchange of
lowed by an inactivation of NF2, followed by other genes ions and small molecules. Mesothelial cells can
located on chromosomes 11, 6 and 3.99 Accordingly, in a secrete immunomodulatory molecules, procoag-
model of murine MM, we found that NF2 does not appear ulants and fibrinolytic mediators. Mesothelial
to play a role in the initial steps of neoplastic progres- cells can phagocytose foreign particles. These
sion.106 multiple functions make the mesothelial cell
Hypotheses on the mechanism of mesothelial oncogen- crucial in maintaining pleural integrity.
esis, especially in response to asbestos fibers, can be tested ● The cellular source involved in the regeneration
with experimental mesotheliomas in knockout mice. of normal or damaged mesothelium is contro-
(Experimental models for mesothelioma are further dis- versial. Mechanisms of mesothelium regenera-
cussed in Chapter 15.) Development of tumor models is a tion include centripetal migration of mesothelial
useful means to identify key genes related to specific onco- cells at the border of the damaged area, and/or
genic processes. MM models have been obtained by attachment of exfoliated mesothelial cells or pre-
intraperitoneal inoculation of asbestos fibers or ceramic cursor cells. It has been suggested that mesothe-
fibers in hemizygous NF2 +/- mice.39,106–108 Somatic genetic lial cells could be multipotent cells.
alterations were similar to those observed in human MM: ● Multiple cytogenetic alterations, loss of heterozy-
frequent deletions of the orthologous genes at the Ink4 gozity and frequent deletions are common fea-
locus, as well as a lower frequency of Trp53 mutations, tures in malignant mesothelioma cells. Exposure
were detected. Further studies are needed to define the of mesothelial cells to asbestos fibers results in
nature of other somatic genetic changes in murine fiber- cell injury that may be responsible for the car-
induced MM, and provide data to better define the charac- cinogenicity of the fibers.
teristics of this type of tumor. ● Development of animal models of mesothelioma
help identify key genes related to the underlying
oncogenic processes. Somatic genetic alterations
CONCLUSIONS in malignant mesotheliomas obtained by
intraperitoneal inoculation of asbestos or
A better knowledge of the biology of mesothelial cells is ceramic fibers in hemizygous NF2+/- mice were
needed to improve our ability to treat pleural diseases. similar to those observed in human malignant
Further research should focus on the multipotent proper- mesothelioma.
ties of mesothelial cells and better define the pathways
References 35
43. Mutsaers SE, McAnulty RJ, Laurent GJ, et al. Cytokine regulation 62. Harvey P, Warn A, Newman P, et al. Immunoreactivity for
of mesothelial cell proliferation in vitro and in vivo. Eur J Cell Biol hepatocyte growth factor/scatter factor and its receptor, met, in
1997; 72: 24–9. human lung carcinomas and malignant meesotheliomas. J Pathol
44. Zhong J, Gencay MM, Bubendorf L, et al. ERK1/2 and p38 MAP 1996; 180: 389–94.
kinase control MMP-2, MT1-MMP, and TIMP action and affect cell 63. Harvey P, Warn A, Dobbin S, et al. Expression of HGF/SF in
migration: a comparison between mesothelioma and mesothelial mesothelioma cell lines and its effects on cell motility,
cells. J Cell Physiol 2006; 207: 540–52. proliferation and morphology. Br J Cancer 1998; 77: 1052–9.
45. Lee TC, Zhang YH, Aston C, et al. Normal human mesothelial cells 64. Tolnay E, Kuhnen C, Wiethege T, et al. Hepatocyte growth
and mesothelioma cell lines express insulin-like growth factor-I factor/scatter factor and its receptor c-Met are overexpressed and
and associated molecules. Cancer Res 1993; 53: 2858–64. associated with an increased microvessel density in malignant
46. Fujino S, Yokoyama A, Kohno N, Hiwada K. Interleukin 6 is an pleural mesothelioma. J Cancer Res Clin Oncol 1998; 124: 291–6.
autocrine growth factor for normal human pleural mesothelial 65. Harvey P, Clark IM, Jaurand MC, et al. Hepatocyte growth
cells. Am J Respir Cell Mol Biol 1996; 14: 508–15. factor/scatter factor enhances the invasion of mesothelioma cell
47. Heldin P, Asplund T, Ytterberg D, et al. Characterization of the lines and the expression of matrix metalloproteinases. Br J Cancer
molecular mechanisms involved in the activation of hyaluronan 2000; 83: 1147–53.
synthetase by platelet-derived growth factor in human 66. Thirkettle I, Harvey P, Hasleton PS, et al. Immunoreactivity for
mesothelial cells. Biochem J 1992; 283: 165–70. cadherins, HGF/SF, met, and erbB-2 in pleural malignant
48. Jacobson A, Brinck J, Briskin MJ, et al. Expression of human mesotheliomas. Histopathol 2000; 36: 522–8.
hyaluronan synthases in response to external stimuli. Biochem J 67. Jagadeeswaran R, Ma PC, Seiwert TY, et al. Functional analysis of
2000; 348: 29–35. c-Met/hepatocyte growth factor pathway in malignant pleural
49. Asplund T, Heldin P. Hyaluronan receptors are expressed on human mesothelioma. Cancer Res 2006; 66: 352–61.
malignant mesothelioma cells but not on normal mesothelial cells. 68. Cao XX, Mohuiddin I, Chada S, et al. Adenoviral transfer of mda-7
Cancer Res 1994; 54: 4516–23. leads to BAX up-regulation and apoptosis in mesothelioma cells,
50. Strizzi L, Catalano A, Vianale G, et al. Vascular endothelial growth and is abrogated by over-expression of BCL-XL. Mol Med 2002; 8:
factor is an autocrine growth factor in human malignant 869–76.
mesothelioma. J Pathol 2001; 193: 468–75. 69. Hopkins-Donaldson S, Cathomas R, Simoes-Wust AP, et al.
●51. Lee YCG, Melkerneker D, Thompson PJ, et al. Transforming growth Induction of apoptosis and chemosensitization of mesothelioma
factor b induces vascular endothelial growth factor elaboration cells by Bcl-2 and Bcl-xL antisense treatment. Int J Cancer 2003;
from pleural mesothelial cells in vivo and in vitro. Am J Respir Crit 106: 160–6.
Care Med 2002; 165: 88–94. 70. Ozvaran MK, Cao XX, Miller SD, et al. Antisense oligonucleotides
52. Klominek J, Baskin B, Liu Z, Hauzenberger D. Hepatocyte growth directed at the bcl-xl gene product augment chemotherapy
factor/scatter factor stimulates chemotaxis and growth of response in mesothelioma. Mol Cancer Ther 2004; 3: 545–50.
malignant mesothelioma cells through c-met receptor. Int J 71. Altomare DA, You H, Xiao GH, et al. Human and mouse
Cancer 1998; 76: 240–9. mesotheliomas exhibit elevated AKT/PKB activity, which can be
53. Rutten AA, Bermudez E, Stewart W, et al. Expression of insulin-like targeted pharmacologically to inhibit tumor cell growth.
growth factor II in spontaneously immortalized rat mesothelial Oncogene 2005; 24: 6080–9.
and spontaneous mesothelioma cells : a potential autocrine role of ◆72. Jaurand MC, Pinchon MC, Bignon J. Mesothelial cells in vitro. In:
insulin-like growth factor II. Cancer Res 1995; 55: 3634–9. Chrétien J, Bignon J, Hirsch A (eds). The pleura in health and
54. Owens MW, Milligan SA. Growth factor modulation of rat pleural disease. New York: Marcel Dekker Inc. 1985: 43–67.
mesothelial cell mitogenesis and collagen synthesis – effects of ◆73. Jantz MA, Antony VB. Pleural fibrosis. Clin Chest Med 2006; 27:
epidermal growth factor and platelet-derived growth factor. 181–91.
Inflammation 1994; 18: 77–87. 74. Kuwahara M, Kuwahara M, Bijwaard KE, et al. Mesothelial cells
55. Adamson IYR, Bakowska J, Prieditis H. Proliferation of rat pleural produce a chemoattractant for lung fibroblasts – role of
mesothelial cells in response to hepatocyte and keratinocyte fibronectin. Am J Respir Cell Mol Biol 1991; 5: 256–64.
growth factors. Am J Respir Cell Mol Biol 2000; 23: 345–9. 75. Lee YC, Lane KB, Zoia O, et al. Transforming growth factor-beta
56. Gordon GJ, Mani M, Mukhopadhyay L, et al. Expression patterns of induces collagen synthesis without inducing IL-8 production in
inhibitor of apoptosis proteins in malignant pleural mesothelioma. mesothelial cells. Eur Respir J 2003; 22: 197–202.
J Pathol 2007; 211: 447–54. 76. Lee YC, Knight DA, Lane KB, et al. Activation of proteinase-
57. Versnel MA, Claessonwelsh L, Hammacher A, et al. Human activated receptor-2 in mesothelial cells induces pleural
malignant mesothelioma cell lines express PDGF beta-receptors inflammation. Am J Physiol Lung Cell Mol Physiol 2005; 288:
whereas cultured normal mesothelial cells express predominantly L734–40.
PDGF alpha-receptors. Oncogene 1991; 6: 2005–11. ◆77. Grove CS, Lee YC. Vascular endothelial growth factor: the key
58. Ramael M, Buysse C, Vandenbossche J, et al. Immunoreactivity for mediator in pleural effusion formation. Curr Opin Pulm Med 2002;
the beta-chain of the platelet-derived growth factor receptor in 8: 294–301.
malignant mesothelioma and non-neoplastic mesothelium. J 78. Idell S, Zwieb C, Kumar A, et al. Pathways of fibrin turnover of
Pathol 1992; 167: 1–4. human pleural mesothelial cells in vitro. Am J Respir Cell Mol Biol
◆59. Fredriksson L, Li H, Eriksson U. The PDGF family: four gene 1992; 7: 414–26.
products form five dimeric isoforms. Cytokine Growth Factor Rev ●79. Boutin C, Dumortier P, Rey F, et al. Black spots concentrate
2004; 15: 197–204. oncogenic asbestos fibers in the parietal pleura: thoracoscopic
60. Metheny-Barlow LJ, Flynn B, van Gijssel HE, et al. Paradoxical and mineralogic study. Am J Respir Crit Care Med 1996; 153:
effects of platelet-derived growth factor-A overexpression in 444–9.
malignant mesothelioma. Antiproliferative effects in vitro and 80. Gelzleichter TR, Bermudez E, Mangum JB, et al. Pulmonary and
tumorigenic stimulation in vivo. Am J Respir Cell Mol Biol 2001; pleural responses in Fischer 344 rats following short-term
24: 694–702. inhalation of a synthetic vitreous fiber. II. Pathobiologic responses.
61. Walker C, Bermudez E, Stewart W, et al. Characterization of Fundam Appl Toxicol 1996; 30: 39–46.
patelet-derived growth factor and platelet-derived growth factor 81. Mitchev K, Dumortier P, De Vuyst P. ‘Black Spots’ and hyaline
receptor expression in asbestos-induced rat mesothelioma. Cancer pleural plaques on the parietal pleura of 150 urban necropsy
Res 1992; 52: 301–6. cases. Am J Surg Pathol 2002; 26: 1198–206.
References 37
●82. Lechner JF, Tokiwa T, LaVeck M, et al. Asbestos-associated 95. Levresse V, Moritz S, Renier A, et al. Effect of simian virus large T
chromosomal changes in human mesothelial cells. Proc Natl Acad antigen expression on cell cycle control and apoptosis in rat
Sci, USA 1985; 82: 3884–8. pleural mesothelial cells exposed to DNA damaging agents.
83. Pelin K, Kivipensas P, Linnainmaa K. Effects of asbestos and man- Oncogene 1998; 16: 1041–53.
made vitreous fibers on cell division in cultured human 96. Liu W, Ernst JD, Broaddus VC. Phagocytosis of crocidolite asbestos
mesothelial cells in comparison to rodent cells. Environ Mol induces oxidative stress, DNA damage, and apoptosis in
Mutagen 1995; 25: 118–25. mesothelial cells. Am J Respir Cell Mol Biol 2000; 23: 371–8.
84. Wu J, Liu W, Koenig K, et al. Vitronectin adsorption to chrysotile 97. Antony VB, Nasreen N, Mohammed KA, et al. Talc pleurodesis:
asbestos increases fiber phagocytosis and toxicity for mesothelial basic fibroblast growth factor mediates pleural fibrosis. Chest
cells. Am J Physiol Lung Cell Mol Physiol 2000; 279: L916–23. 2004; 126: 1522–8.
85. Liaw YS, Yu CJ, Shun CT, et al. Expression of integrins in human 98. Murthy SS, Testa JR. Asbestos, chromosomal deletions, and tumor
cultured mesothelial cells: the roles in cell-to-extracellular matrix suppressor gene alterations in human malignant mesothelioma. J
adhesion and inhibition by RGD-containing peptide. Respir Med Cell Physiol 1999; 180: 150–7.
2001; 95: 2216. ◆99. Sandberg AA, Bridge JA. Updates on the cytogenetics and
◆86. Kamp DW, Graceffa P, Pryor WA, Weitzman SA. The role of free molecular genetics of bone and soft tissue tumors. Mesothelioma.
radicals in asbestos-induced diseases. Free Rad Biol Med 1992; 12: Cancer Genet Cytogenet 2001; 127: 93–110.
293–315. 100. Metcalf RA, Welsh JA, Bennett WP, et al. p53 and Kirsten-ras
87. Ramos-Nino ME, Haegens A, Shukla A, Mossman BT. Role of mutations in human mesothelioma cell lines. Cancer Res 1992; 52:
mitogen-activated protein kinases (MAPK) in cell injury and 2610–5.
proliferation by environmental particulates. Mol Cell Biochem 101. Mor O, Yaron P, Huszar M, et al. Absence of p53 mutations in
2002; 234–235: 111–8. malignant mesothelioma. Am J Respir Cell Mol Biol 1997; 16:
◆88. Antony VB, Owen CL, Hadley KJ. Pleural mesothelial cells 9–13.
stimulated by asbestos release chemotactic activity for neutrophils 102. Vivo C, Lecomte C, Levy F, et al. Cell cycle checkpoint status in
in vitro. Am Rev Respir Dis 1989; 139: 199–206. human malignant mesothelioma cell lines: response to gamma
89. Boylan AM, Ruegg C, Kim KJ, et al. Evidence of a role for radiation. Br J Cancer 2003; 88: 388–95.
mesothelial cell-derived interleukin-8 in the pathogenesis of 103. Arakawa H, Hayashi N, Nagase H, et al. Alternative splicing of the
asbestos-induced pleurisy in rabbits. J Clin Invest 1992; 89: NF2 gene and its mutation analysis of breast and colorectal
1257–67. cancers. Hum Mol Genet 1994; 3: 565–8.
90. Kuwahara M, Kuwahara M, Verma K, et al. Asbestos exposure ●104. Bianchi AB, Mitsunaga S, Cheng J, et al. High frequency of
stimulates pleural mesothelial cells to secrete the fibroblast inactivating mutations in the neurofibromatosis type 2 gene (NF2)
chemoattactant, fibronectin. Am J Respir Cell Mol Biol 1994; 10: in primary malignant mesothelioma. Proc Natl Acad Sci USA 1995;
167–76. 92: 10854–8.
●91. Yegles M, Saint-Etienne L, Renier A, et al. Induction of metaphase 105. Yaegashi S, Sachse R, Ohuchi N, et al. Low incidence of a
and anaphase/telophase abnormalities by asbestos fibers in rat nucleotide sequence alteration of the neurofibromatosis 2 gene in
pleural mesothelial cells in vitro. Am J Respir Cell Mol Biol 1993; human breast cancers. Jpn J Cancer Res 1995; 86: 929–33.
9: 186–91. ●106. Fleury-Feith J, Lecomte C, Renier A, et al. Hemizygosity of Nƒ2 is
92. Jaurand MC. Use of in-vitro genotoxicity and cell transformation associated with increased susceptibility to asbestos-induced
assays to evaluate potential carcinogenicity of fibres. In: Kane AB, peritoneal tumours. Oncogene 2003; 22: 3799–805.
Boffetta P, Sarracci R, Wilbourn JD (eds). Mechanisms in fiber 107. Altomare DA, Vaslet CA, Skele KL, et al. A mouse model
carcinogenesis. Lyon: IARC, 1996: 55–72. recapitulating molecular features of human mesothelioma. Cancer
●93. Jaurand MC, Broaddus C. Asbestos fibers and their interaction Res 2005; 65: 8090–5.
with mesothelial cells in vitro and in vivo. In: Robinson BWS, 108. Lecomte C, Andujar P, Renier A, et al. Similar tumor suppressor
Chahinian A (eds). Mesothelioma. London: Martin Dunitz, 2002: gene alteration profiles in asbestos-induced murine and human
273–94. mesothelioma. Cell Cycle 2005; 4: 1862–9.
94. Levresse V, Renier A, Fleury-Feith J, et al. Analysis of cell cycle
disruptions in cultures of rat pleural mesothelial cells exposed to
asbestos fibres. Am J Respir Cell Mol Biol 1997; 17: 660–71.
This page intentionally left blank
4
Normal physiological fluid and cellular contents
MARC NOPPEN
Study Species Mean volume Total white Macrophages Monocytes Mesothelial Lymphocytes
(ref) (right and left blood cell (%) (%) cells (%) (%)
pleural space) count
(mL/kg) (cells/mL)
Differential cell counts varied with the different methods percent), globulins (35 percent) and fibrinogen.3,9 Levels
of fixation (95 percent alcohol and Papinacolaou stain of large molecular weight proteins, such as lactate dehy-
versus 50 percent alcohol, 1 percent polyethylene glycol drogenase, in the pleural fluid are less than half of that
and Papinicolaou stain) between 38.6 and 70.1 percent found in serum.
monocytes, 10 and 10.6 percent lymphocytes and 5.5 and
16.6 percent macrophages. Sahn et al.9 aspirated costodi-
aphragmatic fluid in rabbits. Total volume of the free
pleural fluid in both pleural spaces was 0.45 ± 0.90 mL HUMAN STUDIES
(0.13 mL/kg). Total white blood cell count was 1503 ± 414
cells/mL, with 70.1 ± 3.6 percent monocytes, 10.6 ± 1.8 Reliable data on the volume and cellular content of pleural
percent lymphocytes, 8.9 ± 1.6 percent mesothelial cells fluid in normal humans are scarce because of the obvious
and 7.5 ± 1.5 percent macrophages. Novakov and Peshev10 difficulties in retrieving this small amount of fluid without
performed aspiration and lavage in rabbits. Volumes and ‘disturbing’ the pleural environment. The first study
total white blood cell counts were not reported; differential addressing this issue was that of Yamada,18 published in
cell counts included 9.25 percent macrophages, 66.5 1933, who punctured the ninth or tenth intercostals space
percent monocytes, 8 percent mesothelial cells and 9.75 on the dorsal axillary line in a group of healthy Japanese
percent lymphocytes after aspiration, and 5 percent soldiers. In approximately 30 percent of cases, some fluid
macrophages, 60.17 percent monocytes, 10 percent was aspirated after a period of rest, whereas in approxi-
mesothelial cells and 11.08 percent lymphocytes after mately 70 percent of cases some fluid was retrieved after
lavage. Other measurements of pleural fluid volume have exercise. Usually only a few drops of foam was aspirated
been made by Broaddus and Araya,11,Wang and Lai- but, in a few cases, up to 20 mL could be retrieved. Total
Fook12 and Agostoni and Zocchi13 in rabbits, Mellins et white blood cell count was 4500 cells/mL (range
al.14 in dogs, Wiener-Kronish et al.15 and Broaddus et al.16 1700–6200). Differential cell count showed 53.7 percent
in sheep and Miserocchi et al.17 in various animal models cells similar to monocytes, 10.2 percent lymphocytes, 3
(cats, dogs and pigs), as part of studies for purposes other percent mesothelial cells, 3.6 percent granulocytes and
than actual volume measurements. All measurements 29.5 percent ‘deteriorated cells of difficult classification’.
(except those in puppies) yielded total volumes between More recently, a pleural lavage technique was used to
0.04 and 0.28 mL/kg. retrieve the few milliliters of pleural fluid present in the
The solute composition of normal pleural fluid is pleural space of otherwise healthy participants undergoing
similar to that of interstitial fluid of other organs and con- thoracoscopic sympathectomy for the treatment of essen-
tains 1–2 g/100 mL, mainly consisting of albumin (50 tial hyperhidrosis.19 In analogy with bronchoalveolar
References 41
lavage (a technique enabling retrieval of small volumes of of a cell smear is shown in Figure 4.1. In a second study
epithelial lining fluid from the lung), 150 mL of pre- using a similar lavage technique, lymphocyte subtyping
warmed saline was injected in, and immediately aspirated showed a lower proportion of CD4+ T cells (30 percent
from, the right pleural space, after induction of a pneu- versus 45.8 percent) and a higher proportion of CD8+ T
mothorax in the setting of a thoracoscopic sympathec- cells (11.78 versus 9.6 percent) and regulatory
tomy performed for the treatment of essential (CD4+CD25+) T cells in pleural fluid in normal subjects
hyperhidrosis.20 With urea used as an endogenous marker compared with blood, which may suggest that previously
of dilution, measured mean right-sided pleural fluid described abnormalities in lymphocyte subsets in pleural
volume was 8.4 ± 4.3 mL. In a subgroup of subjects, right- effusions may not only be a result of the pleural disease
and left sided pleural fluid volumes were shown to be but may also be a characteristic of the pleural compart-
similar. Expressed per kg of body mass, total pleural fluid ment itself.21 Interestingly, a small but statistically signifi-
volume in non-smoking, healthy subjects is 0.26 mL/kg, cant increase in pleural fluid neutrophils was observed in
which corresponds well with values obtained in animal smoking subjects. In addition to revealing the volume and
studies. Total white blood cell count in the pleural fluid of cellular composition of normal pleural fluid (which may
normal non-smoking subjects was 1716 cells/mL. be helpful in understanding cellular events occurring in
Differential cell count yielded a predominance of disorders characterized by pleural effusions), this pleural
macrophages (median 75 percent, interquartile range 16 lavage technique allows the study of the pathophysiologi-
percent) and lymphocytes (median 23 percent, interquar- cal events in pleural disorders that typically are not associ-
tile range 18 percent). Mesothelial cells, neutrophils and ated with pleural effusions, such as pneumothorax22 and
eosinophils were only marginally present. A typical image asbestos-related pleurisy.23
KEY POINTS
● In normal animals and humans, the pleural space
contains a small volume of pleural fluid. In differ-
ent adult animal species this volume varies
between 0.04 and 0.60 mL/kg. In normal humans,
the pleural fluid volume is 0.26 mL/kg.
● This fluid has the solute characteristics of all
interstitial fluids, and contains a total of 1000 to
2500 white blood cells per mL. Macrophages/
monocytes and lymphocytes are the predomi-
nant cell types.
● Pleural lavage is a safe and simple technique
allowing the study of normal pleural fluid, and of
pleural disease which is not characterized by
pleural effusions
REFERENCES
● = Key primary paper
◆ = Major review article
5. Mills PC, Chen Y, Hills YC, Hills BA. Comparison of surfactant lipids and entry rates in pleural fluid and plasma in sheep. J Appl Physiol
between pleural and pulmonary lining fluids. Pulm Pharmacol Ther 1984; 56: 459–63.
2006; 19: 292–6. 16. Broaddus VC, Araya M, Carlton DP, Bland RD. Developmental
6. Noppen M. Normal volume and cellular contents of pleural fluid. changes in pleural liquid protein concentration in sheep. Am Rev
Curr Opin Pulm Med 2001; 7: 180–82. Respir Dis 1991; 143: 38–41.
●7. Miserocchi G, Agostoni E. Contents of pleural space. J Appl Physiol 17. Miserocchi G, Negrini D, Mortola J. Comparative features of
1971; 30: 208–13. Starling-lymphatic interaction at the pleural level in mammals. J
8. Stauffer JL, Potts DE, Sahn SA. Cellular contents of the normal Appl Physiol 1984; 56: 1151–6.
rabbit pleural space. Acta Cytol 1978; 22: 570–74. ●18. Yamada S. Uber die seröse Flüssigkeit in der Pleurahöhle der
9. Sahn SA, Willcox ML, Good JT , Potts DE, Filley DF. Characteristics gesunden Menschen. Z Ges Exp Med 1933; 90: 342–8.
of normal rabbit pleural fluid: physiologic and biochemical ●19. Noppen M, De Waele M, Li R, et al. Volume and cellular content of
implications. Lung 1979; 156: 63–9. normal pleural fluid in humans examined by pleural lavage. Am J
10. Novakov IP, Peshev ZP. Cell types in the normal pleural fluid from Respir Crit Care Med 2001; 7: 180–82.
rabbits. Trakia J Sci 2005; 3: 22–5. 20. Noppen M, Herregodts P, D’haese J, Vincken W, Dhaens J. A
11. Broaddus VC, Araya M. Liquid and protein dynamics using a new simplified thoracoscopic sympathicolysis technique for essential
minimally invasive pleural catheter in rabbits. J Appl Physiol 1992; hyperhidrosis: results in 100 consecutive patients. J Laparoendosc
72: 851–7. Surg 1996; 6: 151–9.
12. Wang PM, Lai-Fook SJ. Pleural tissue hyaluran produced by post- 21. Scherpereel A, Madsen P, Chahine B, et al. T cell subsets in pleural
mortem ventilation in rabbits. Lung 2000; 178: 1–12. fluid of healthy subjects. Am J Respir Crit Care Med 2007; 175:
13. Agostoni E, Zocchi L. Starling forces and lymphatic drainage in A452.
pleural liquid and protein exchanges. Respir Physiol 1991; 86: 22. De Smedt A, Vanderlinden E, Demanet C, et al. Characterisation of
271–81. pleural inflammation occurring after primary spontaneous
14. Mellins RB, Levine OR, Fishman AP. Effects of systemic and pneumothorax. Eur Respir J 2004; 23: 896–900.
pulmonary venous hypertension on pleural and pericardial fluid 23. Noppen M, Vanderlinden E, Demanet C, De Waele M. Pleural
accumulation. J Appl Physiol 1970; 29: 546–9. lavage in non-exudative benign asbestos-related pleural disease.
15. Wiener-Kronish JP, Albertine KH, Licko V, Staub NC. Protein egress Am J Respir Crit Care Med 2002; 165: A33.
5
Physiology: fluid and solute exchange in normal
physiological states
V COURTNEY BROADDUS
transudative liquid with a protein concentration between study, spontaneously hypertensive rats were found to have
2–3 g/dL, represents a definite increase in the protein con- lower total protein and albumin concentration ratios and
centration above the normal level. The low protein a higher pleural space thickness than in the control, nor-
concentration of normal liquid indicates a high degree of motensive rats.12 In the second study, sheep were studied
sieving of the protein molecules during fluid filtration from at different stages of development. As mammals grow
the microvasculature. The concentration, which represents from fetal to newborn to adult life, systemic arterial pres-
a protein ratio of 0.15–0.2, is in keeping with systemic inter- sure increases while pulmonary arterial pressure decreases.
stitial liquids of the body and is very different from the Thus, if pleural liquid arose from the systemic circulation,
higher protein ratio of pulmonary filtrate (0.6–0.7).10,11 pleural liquid protein concentrations would be expected to
Thus, the low protein concentration is strong evidence that decrease; if pleural liquid arose from the pulmonary circu-
normal pleural liquid arises from systemic vessels. lation, the opposite would be expected. In pleural liquid
Additional evidence for the systemic origin of pleural collected from sheep at different ages, pleural liquid
liquid derives from studies in animals in which systemic protein concentration ratios decreased progressively; in
pressure is found to vary. In animals with an increased sys- fetuses, the ratio was 0.50, in newborns 0.27 and in adults
temic arterial (and microvascular) pressure, filtration at 0.15, supporting a systemic origin of the liquid.13
the microvessels and the sieving of protein would be The likely systemic sources of liquid lie in the adjacent
expected to be higher and, if systemic vessels were the pleural membranes themselves (Figures 5.1 and 5.2). In
source of pleural liquid, the pleural protein concentration effect, the pleural space is sandwiched between two sys-
ratio (pleural/plasma) would be expected to be lower than temic circulations: the intercostal arterial circulation of the
in animals with a lower systemic pressure. In the first parietal pleura (Figure 5.1, see B) and the bronchial arterial
20 mm
Figure 5.1 Parietal pleura. The parietal pleura is lined by Figure 5.2 Visceral pleura. The visceral pleura (VP) lies between
mesothelial cells (M) adjacent to the pleural space (PS). The blood the pleural space (PS) and the alveoli of the lung parenchyma,
supply is via the intercostal arteries (B). The parietal pleura, but and is lined by mesothelial cells (M). The blood supply to the
not the visceral pleura, contains the lymphatics (L) that drain visceral pleura is via the bronchial arteries (A). (Reproduced from
pleural liquid via stomata that open into the pleural space. Reference 3 by courtesy of Marcel Dekker, Inc.)
Pleural fluid production 45
circulation of the visceral pleura (Figure 5.2, see A). It is Parietal pleura Visceral pleura
interesting that the parietal pleural circulation is constant
Pleural
among species with a morphology almost interchangeable space Pulmonary
from small mammals to humans.7 By contrast, the visceral capillaries
pleural circulation changes drastically depending on
whether the visceral pleura is ‘thick’ as in humans, sheep
and most large animals, or is ‘thin’ as in smaller mammals
like dogs, rabbits and mice.6 Thick visceral pleura has a sys- Lymphatic
temic bronchial blood supply while the thin visceral pleura stomata
has no systemic circulation itself but is nourished by the
underlying pulmonary circulation. For several reasons, the
systemic blood supply of the parietal pleura is thought to be
the major source of normal pleural liquid. First, despite the
great differences in visceral pleural anatomy and blood
supply, measured rates of pleural liquid production are
Alveoli
similar among different species, suggesting the constant
parietal pleura as the source.9,14 Second, the parietal pleural 15 μm 40 μm
microvessels are closer to the pleural space (10–15 mm)
than are those of the visceral pleura (20–50 mm).6,7 Finally, Intercostal Bronchial
the parietal pleural vessels are likely to have a higher microvessels microvessels
microvascular pressure due to their drainage into systemic
venules while the visceral bronchial vessels drain into lower Figure 5.3 Schema of pleural liquid entry and exit in the
resistance pulmonary venules.3 normal state. The microvascular filtrate of the arterial blood
Once the liquid filters across the systemic vessels, it can supply flows across the leaky mesothelial layer into the lower
then flow along a pressure gradient across the mesothelial pressure pleural space. From the pleural space pleural liquid exits
layer into the pleural space (Figure 5.3). The pressure gra- via the lymphatic stomata into the parietal pleural lymphatics.
dient exists from the high pressure pleural systemic
microvessels into its surrounding interstitial tissue and
from the interstitial tissues into the subatmospheric
pleural space.15 The mesothelial layer separating the inter-
stitial tissue from the pleural space is leaky, especially rela- serum in patients.19 Thus, passive forces alone may be suf-
tive to other barriers such as the epithelial barriers of the ficient to explain the small difference in ionic concentra-
alveoli or the kidney. Mesothelium has been shown, both tions between pleural liquid and serum and are more
in in vitro and in situ studies, to offer little resistance to the consistent with the leaky nature of the mesothelium.8,20
movement of liquid and protein.8,16 Thus, the liquid and Once it enters the pleural space, pleural liquid very
protein filtered from the pleural microvessels (as well as slowly flows toward the dependent regions of the pleural
liquid arising anywhere in the body) can flow across the space.21 Such flow has been proposed based on findings of
mesothelium along a pressure gradient into the pleural pleural pressure gradients. Pleural pressure has been
space. By virtue of its large size and surface area, its sub- exceedingly difficult to measure: even by placing a small
atmospheric pressure and its relative leaky borders, the catheter into the pleural space, the space is greatly widened
pleural space is clearly vulnerable to the accumulation of and pressures are measured consistent with a static
liquids. column of liquid. However, when pleural pressures are
The mesothelium has been proposed to be capable of measured with micropipettes that do not distort the space,
active transport. This possibility is particularly intriguing pleural pressure gradients are consistent with a gradual
when examining the distribution of ions in pleural liquid flow of liquid from the top to bottom.22,23
and serum because pleural concentrations of some ionic Based on noninvasive studies in different species, the
solutes such as bicarbonate have been noted to be slightly production of pleural liquid is normally slow. In earlier
different from those in serum.17,18 Although a higher studies, measured flows were probably overestimated due
pleural concentration has raised the possibility of an active to inflammation and distortion caused by invasive tech-
transport mechanism for bicarbonate into the pleural niques used to measure the fluid turnover. Based upon
space, the distribution of ions may also be explained by a radiolabel studies or minimally invasive techniques that
passive process, the Donnan equilibrium. In such an equi- leave the pleural space intact, pleural liquid production has
librium, differences in protein concentrations may pas- been measured at approximately 0.01 mL/(kg ¥ h) in
sively alter ionic balances between two electrolytic awake sheep9 and 0.02 in rabbits.14 The half-time of pleural
solutions separated by a semipermeable membrane. Such a liquid turnover in sheep and rabbits is 6–8 hours.14 At
mechanism has been proposed to explain similar differ- these rates, there would be an entry (and exit) of 15 mL per
ences in ion concentrations between body fluids and day in a 60 kg human.
46 Physiology: fluid and solute exchange in normal physiological states
tion beyond that point. Although over time this maximal Respiratory Medicine, 4th ed. Philadelphia: Saunders, 2000:
rate could increase, for example by increases in stomata 1913–60.
●5. Broaddus VC, Wiener-Kronish JP, Berthiaume Y, Staub NC.
number or size or increases in lymphatic contractility, Removal of pleural liquid and protein by lymphatics in awake
lymphatic capacity would set a definite limit in the ability sheep. J Appl Physiol 1988; 64: 384–90.
of the pleural space to handle liquid. 6. Albertine KH, Wiener-Kronish JP, Roos PJ, Staub NC. Structure,
Water channels have been investigated as a possible blood supply, and lymphatic vessels of the sheep’s visceral pleura.
means of clearance of pleural fluid.38,39 Aquaporins (AQP) Am J Anat 1982; 165: 277–94.
7. Albertine KH, Wiener-Kronish JP, Staub NC. The structure of the
present in mesothelial cells, particularly AQP1, have been parietal pleura and its relationship to pleural liquid dynamics in
shown to participate in the rapid osmotic shifts that sheep. Anat Rec 1984; 208: 401–9.
mediate movement of pure water. However, pleural clear- 8. Kim KJ, Critz AM, Crandall ED. Transport of water and solutes
ance of isoosmotic fluids, such as saline, is not affected by across sheep visceral pleura. Am Rev Resp Dis 1979; 120: 883–
the presence or absence of aquaporins. Therefore the main 92.
●9. Wiener-Kronish JP, Albertine KH, Licko V, Staub NC. Protein egress
route of clearance of fluid and protein is still most likely and entry rates in pleural fluid and plasma in sheep. J Appl Physiol
via the lymphatics. 1984; 56: 459–63.
10. Erdmann AJ, Vaughan TR, Brigham KL, Woolverton WC, Staub NC.
Effect of increased vascular pressure on lung fluid balance in
unanesthetized sheep. Circ Res 1975; 37: 271–84.
◆11. Wiener-Kronish JP, Broaddus VC. Interrelationship of pleural and
KEY POINTS
pulmonary interstitial liquid. Annu Rev Physiol 1993; 55: 209–26.
12. Lai-Fook SJ, Kaplowitz MR. Pleural protein concentration and
● Intrapleural pressure is lower than the interstitial liquid volume in spontaneously hypertensive rats. Microv Res
pressure of either of the pleural tissues. This pres- 1988; 35: 101–8.
sure difference constitutes a gradient for liquid 13. Broaddus VC, Araya M, Carlton DP, Bland RD. Developmental
movement into, but not out of, the pleural space. changes of pleural liquid protein concentration in sheep. Am Rev
Resp Dis 1991; 143: 38–41.
● The pleural membranes are leaky to liquid and 14. Broaddus VC, Araya M. Liquid and protein dynamics using a new,
protein providing little resistance to protein minimally invasive pleural catheter in rabbits. J Appl Physiol 1992;
movement. 72: 851–7.
● The entry of pleural liquid is normally slow and 15. Bhattacharya J, Gropper MA, Staub NC. Interstitial fluid pressure
is compatible with known interstitial flow rates, gradient measured by micropuncture in excised dog lung. J Appl
Physiol 1984; 56: 271–7.
approximately 0.5 mL hourly in a grown man. 16. Parameswaran S, Brown LV, Ibbott GS, Lai-Fook SJ. Hydraulic
● Most liquid exits the pleural space by bulk flow, conductivity, albumin reflection and diffusion coefficients of pig
not by diffusion. This is evident because the mediastinal pleura. Microvasc Res 1999; 58: 114–27.
protein concentration of pleural effusions does 17. D’Angelo E, Heisler N, Agostoni E. Acid–base balance of pleural
not change as the effusion is absorbed. liquid in dogs. Respir Physiol 1979; 37: 137–49.
18. Rolf LL, Travis DM. Pleural fluid-plasma bicarbonate gradients in
● The major exit of liquid and protein is via the oxygen-toxic and normal rats. Am J Physiol 1973; 224: 857–61.
parietal pleural stomata (2–6 mm diameter) and 19. Gilligan DR, Volk MC, Blumgart HL. Observations on the chemical
the pleural lymphatics. These lymphatics have a and physical relation between blood serum and body fluids. J Clin
large capacity for absorption, increasing up to 30 Invest 1934; 13: 365–81.
times the baseline exit rate, effusion and thereby 20. Payne DK, Kinasewitz GT, Gonzalez E. Comparative permeability of
canine visceral and parietal pleura. J Appl Physiol 1988; 65:
resisting formation. 2558–64.
21. Lai-Fook S. Pleural mechanics and fluid exchange. Physiol Rev
2004; 84: 385–410.
22. Lai-Fook SJ, Rodarte JR. Pleural pressure distribution and its
relationship to lung volume and interstitial pressure. J Appl Physiol
REFERENCES 1991; 70: 967–78.
23. Lai-Fook S, Price D, Staub N. Liquid thickness vs vertical pressure
● = Key primary paper gradient in a model of the pleural space. J Appl Physiol 1987; 62:
◆ = Major review article 1747–54.
24. Agostoni E, Zocchi L. Active Na+ transport and coupled liquid
1. Noppen M, De Waele M, Li R, et al. Volume and cellular content of outflow from hydrothoraces of various size. Respir Physiol 1993;
normal pleural fluid in humans examined by pleural lavage. Am J 92: 101–13.
Respir Crit Care Med 2000; 162: 1023–6. 25. Negrini D, Ballard ST, Benoit JN. Contribution of lymphatic
2. Albertine KH, Wiener-Kronish JP, Bastacky J, Staub NC. No myogenic activity and respiratory movements to pleural lymph
evidence for mesothelial cell contact across the costal pleural flow. J Appl Physiol 1994; 76: 2267–76.
space of sheep. J Appl Physiol 1991; 70: 123–4. 26. Agostoni E, Bodega F, Zocchi L. Albumin transcytosis from the
◆3. Staub NC, Wiener-Kronish JP, Albertine KH. Transport through the
pleural space. J Appl Physiol 2002; 93: 1806–12.
●27. Wang N-S. The preformed stomas connecting the pleural cavity
pleura: physiology of normal liquid and solute exchange in the
pleural space. In: Chretien J, Bignon J, Hirsch A (eds). The pleura in and the lymphatics in the parietal pleura. Am Rev Respir Dis 1975;
health and disease. New York: Marcel Dekker, Inc., 1985: 174–5. 111: 12–20.
◆4. Broaddus VC, Light RW. Pleural effusion. In: Mason RJ, Murray JF,
28. Shinohara H. Distribution of lymphatic stomata on the pleural
Broaddus VC, Nadel JA (eds). Murray Nadel’s Textbook of surface of the thoracic cavity and the surface topography of the
48 Physiology: fluid and solute exchange in normal physiological states
pleural mesothelium in the golden hamster. Anat Rec 1997; 249: monkey, as studied by video-assisted thoracoscopy and electron
16–23. microscopy. J Thorac Cardiovasc Surg 2000; 120: 437–47.
29. Li Y-Y, Li J-C. Ultrastructure and three-dimensional study of the 35. Shinohara H, Kominami R, Taniguchi Y, Yasutaka S. The
lymphatic stomata in the costal pleura of the rabbit. Micros Res distribution and morphology of lymphatic vessels on the
Techn 2003; 62: 240–46. peritoneal surface of the adult human diaphragm, as revealed by
30. Liu J, Wong HL, Moselhy J, et al. Targeting colloidal particulates to an ink-absorption method. Okajimas Folia Anat Jpn 2003; 79:
thoracic lymph nodes. Lung Cancer 2006; 51: 377–86. 175–83.
31. Wang Q, Ohtani O, Saitoh M, Ohtani Y. Distribution and 36. Grimaldi A, Moriondo A, Sciacca L, et al. Functional arrangement
ultrastructure of the stomata connecting the pleural cavity with of rat diaphragmatic initial lymphatic network. Am J Physiol Heart
lymphatics in the rat costal pleura. Acta Anat (Basel) 1997; 158: Circ Physiol 2006; 291: H876–85.
255–65. 37. Leckie WJH, Tothill P. Albumin turnover in pleural effusions. Clin
32. Gaudio E, Rendina EA, Pannarale L, Ricci C, Marinozzi G. Surface Sci 1965; 29: 339–52.
morphology of the human pleura. A scanning electron microscopic ●38. Song Y, Yang B, Matthay MA, Ma T, Verkman AS. Role of
study. Chest 1988; 93: 149–53. aquaporin water channels in pleural fluid dynamics. Am J Physiol
33. Li J. Ultrastructural study on the pleural stomata in human. Funct Cell Physiol 2000; 279: C1744–50.
Develop Morphol 1993; 3: 277–80. 39. Jiang J, Hu J, Bai CX. Role of aquaporin and sodium channel in
●34. Miura T, Shimada T, Tanaka K, Chujo M, Uchida Y. Lymphatic pleural water movement. Respir Physiol Neurobiol 2003; 139:
drainage of carbon particles injected into the pleural cavity of the 83–8.
6
Physiology: changes with pleural effusion and
pneumothorax
RICHARD W LIGHT
In this chapter the effects of pleural fluid or pleural air on pleural pressure becomes positive. When there is sufficient
pleural pressures, the lung, the diaphragm, the heart, pleural fluid such that the lung is separated from the chest
pulmonary gas exchange and exercise tolerance will be wall, there is a vertical gradient of 1 cm H2O per cm verti-
discussed. cal height because of the weight of the fluid.1 If there is a
hydrostatic column 40 cm high in a hemithorax, then the
pressure at the bottom of the column would be expected to
EFFECTS OF EFFUSION ON THE PLEURAL be approximately 40 cm H2O.
PRESSURE In one older study we measured the pleural pressure in
52 patients with significant pleural effusions (median
When pleural fluid is present, its volume must be compen- amount of fluid greater than 1000 mL). Overall, the mean
sated for by an increase in the size of the thoracic cavity, a pleural pressure was approximately 0, but there was a wide
decrease in the size of the lung or the heart, or a combina- range in the pleural pressures with a range of -21 cmH2O
tion of these changes. The thoracic cavity, the lungs and to +8 cmH2O (Figure 6.1).2 Pleural pressures of -5 cmH2O
the heart are all distensible objects, which means that the and lower were only seen with a trapped lung or with
volume of each is dependent upon the pressure inside malignancy. Villena et al.3 measured the pleural pressure
minus the pressure outside. The presence of pleural fluid in 61 patients with pleural effusions of varying etiology.
increases the pleural pressure. Since the distending pres- These workers found that the initial pleural pressure
sure of the thoracic wall is the atmospheric pressure minus ranged from -12 to +25 cmH2O and the mean pressure in
the pleural pressure, an increase in the pleural pressure will their patients was approximately +5 cmH2O. The reason
lead to an increase in the distending pressure of the tho- that the pleural pressures were not more positive in the
racic cavity and an increase in the volume of the thoracic above two studies is probably due to the insertion of the
cavity. In contrast, the distending pressure of the lungs is thoracentesis needle closer to the superior than the inferior
the alveolar pressure minus the pleural pressure. end of the hydrostatic column. The pleural pressure at
Therefore, an increase in the pleural pressure will lead to a times can be quite positive with a pleural effusion. Neff
decreased lung volume. The distending pressure of the and Buchanan4 reported that the initial pleural pressure
heart is the intracardiac pressure minus the pleural pres- was 76 cmH2O in a patient with a pleural effusion second-
sure, so an increase in the pleural pressure will lead to a ary to pneumothorax therapy for tuberculosis many years
decrease in the size of the heart. previously.
The pleural pressure is normally negative. However, When pleural fluid is removed with thoracentesis, the
when more than minimal pleural fluid accumulates, the fluid volume removed is compensated for by an increase in
50 Physiology: changes with pleural effusion and pneumothorax
110
0
Pleural pressure (cmH2O)
210
220
230
They reported that the lung did not re-expand (trapped forced expiratory volume in one second (FEV1) each
lung) in 11 of the 14 patients that had a pleural elastance increased approximately 400 mL. In other words, for every
greater than 19 cmH2O. Pleurodesis was attempted in the 1000 mL pleural fluid removed, the FVC and FEV1
other three patients with a high pleural elastance and it improve approximately 200 mL. In this study the TLC
failed in all three. In contrast, only three of 51 patients with increased almost twice as much as did the FVC or FEV1.
pleural elastance less than 19 cmH2O had a trapped lung, This is in contrast to the dog study outlined above where
and pleurodesis was successful in 42 of 43 patients (98 the TLC was affected less than the FRC. One possible
percent) who returned for evaluation.6 explanation for the varying results in humans and in dogs
is that in humans the lower lobe is frequently completely
atelectatic when a large pleural effusion is present. When
EFFECTS OF EFFUSION ON PULMONARY fluid is then removed, the lower lobe re-expands and the
FUNCTION residual volume in the reexpanded lobe increases the TLC
more than just the increase in the vital capacity of that
The effects of a pleural effusion on pulmonary function are lobe.
difficult to determine. Many diseases that cause pleural Although there is approximately a 20 mL increase in the
effusions, such as congestive heart failure, malignancy, FVC for every 100 mL of pleural fluid withdrawn, there is
pneumonia and pulmonary embolism, also affect the pul- much inter-individual variability (Figure 6.3). Changes in
monary parenchyma. Therefore, it is difficult to determine the FVC are related to pressure measurements during tho-
what part of the pulmonary dysfunction is caused by the racentesis.8 Patients with higher initial pleural pressures
pleural effusion and what part is caused by disease in the and patients with smaller changes in the pleural pressure as
underlying lung. fluid is removed are more likely to have larger increases in
There have been limited studies of the effects of a their vital capacity. Nevertheless, even by taking into con-
pleural effusion on the pulmonary function of animals. sideration the amount of fluid removed, the initial pleural
Krell and Rodarte7 studied the volume changes in the lung pressure and the pleural elastance, the multiple regression
and hemithorax after 200 to 1200 mL pleural fluid was coefficient with the FVC as the dependent variable never
added to the right hemithorax of dogs. They found that the exceeded 0.60. This indicates that less than 40 percent of
decrease in lung volume at functional residual capacity the variance in the change in the FVC was related to the
(FRC) was approximately one-third of the added saline amount of fluid removed and the measures of pleural pres-
volume, while the decrease in the lung volume at total lung sure. Possible explanations for the poor correlation are: (a)
capacity (TLC) was one-fifth of the added saline volume. the pulmonary function testing was not performed until
Consequently, the chest wall volume increased by two- 24 hours after the thoracentesis and the fluid might have
thirds the added saline volume at FRC and by four-fifths of reaccumulated to a variable degree during this time
the added saline volume at TLC.7 At a given esophageal period; (b) the pleural pressure changes recorded reflected
pressure (which is taken as a measure of pleural pressure), the elastance of the pleural space during the thoracentesis
the chest wall volume was higher and the lung volume was and if the lung had been atelectatic for a prolonged period,
lower when saline had been added to the hemithorax. it may take several hours or days for the lung to re-expand;
There was a larger decrease in the lower lobe volume than and (c) in some patients with large effusions, there is
in upper lobe volume.7 mediastinal shift to the opposite side, but this does not
There have been several studies evaluating the pul- always occur with malignancy because of fixation of the
monary function of patients with pleural effusions. We mediastinum by the malignant process.8
performed pulmonary function tests before and 24 hours Estenne and colleagues9 measured the changes in respi-
after thoracentesis in 26 patients from whom a mean of ratory mechanics in nine patients 2 hours after the
1740 mL pleural fluid was withdrawn (Table 6.1).8 We removal of a mean 1818 mL pleural fluid. They reported a
found that the mean forced vital capacity (FVC) and the mean increase of 300 mL in the vital capacity, which was
Table 6.1 Mean pulmonary functions baseline and 24 hours after therapeutic thoracentesis
1000
Improvement in FVC (ml)
500
similar to what we reported. They also found that the TLC therapeutic thoracentesis has relatively little effect on the
increased approximately twice as much as did the FVC.9 In arterial blood gas results.
addition, they studied the maximal pressures generated by When experimental bilateral pleural effusions were
the inspiratory muscles and found that the pressures were induced in pigs, there was a mild decrease in the partial
much greater post thoracentesis at a given lung volume. pressure of arterial O2 (PaO2) while the partial pressure of
The greater inspiratory pressures were attributed to a arterial CO2 (PaCO2) remained stable as the total amount
decrease in the thoracic cage volume.9 For example, the of pleural fluid was increased to 30 mL/kg (Figure 6.4).12
maximal inspiratory pressure (MIP) at TLC was However, after the amount of pleural fluid exceeded
-16 cmH2O before thoracentesis and increased to 30 mL/kg, the PaO2 dropped precipitously and the PaCO2
-25 cmH2O after thoracentesis, while the highest MIP started to increase. When the amount of pleural fluid
went from -41 cmH2O before thoracentesis to reached 80 mL/kg, the PaO2 (on 100 percent oxygen) was
-52 cmH2O post thoracentesis.9 I believe that relief of the less than 80 mmHg while the PaCO2 had increased from
downward displacement of the diaphragm by the pleural 34 to 51 mmHg. When the pleural fluid was removed in
fluid is probably the primary explanation for the improve- these normal pigs, the PaO2 and PaCO2 rapidly returned to
ment in the ability of the patient to generate more negative normal (Fig. 6.4).13
inspiratory pressures after thoracentesis.10 Things seem more complicated in humans. In an early
Some support for the last statement has been provided study of 16 patients with pleural effusions, Brandstetter
by Wang and Tseng.11 These researchers selected 21 and Cohen13 obtained blood gases before, and then 20
patients who had a pleural effusion and an inverted minutes, two hours and 24 hours after a thoracentesis in
diaphragm and measured pulmonary function before and which 150 to 1600 mL pleural fluid was removed. They
24 hours after the removal of a mean of 1610 mL pleural reported that the mean PaO2 at baseline was 70.4 mmHg,
fluid. They found that the mean FVC increased by 317 mL and this decreased significantly to 61.2 mmHg 20 minutes
while the mean FEV1 increased by 234 mL, which were following thoracentesis. In every patient there was a
changes similar to those observed by Estenne et al.9 and decrease in the PaO2 over this 20-minute period. The PaO2
our group.8 Interestingly, the patients in this study were remained significantly reduced at 2 hours (64.4 mmHg)
very dyspneic prior to thoracentesis and their dyspnea but had returned to baseline 24 hours later. In this study
improved markedly following the thoracentesis. They there were no significant changes in the pH or the
attributed the decreased dyspnea to the fact that the PaCO2.13
diaphragm was no longer inverted. The effects of a therapeutic thoracentesis on the blood
A therapeutic thoracentesis has essentially no effect on gases of patients on mechanical ventilation is not clear. In
the diffusion capacity of the lung (DLCO)8 or the specific one study14 a chest tube was placed in 19 patients with
airway conductance.8,9 acute respiratory distress syndrome (ARDS) who had
pleural effusions and refractory hypoxemia.14 After place-
ment of the chest tube, the compliance of their lungs
EFFECTS OF EFFUSION ON BLOOD GASES improved immediately, and 24 hours after insertion of the
chest tube, the PaO2/fraction of inspired air that is O2
Although patients with pleural effusions frequently have (FIO2) had improved to 245 ± 29 from 151 ± 13.14
abnormal arterial blood gas results, the performance of a However, in a second study, nine patients on mechanical
Effects of effusion on blood gases 53
100 52
90 MAP 50 PaCO2
80 48
70 46
60 44
(mmHg)
(torr)
50 42
40 40
30 38
20 36
10 34
0 32
0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70
Volpl (mL/kg) Volpl (mL/kg)
600 50
PaO2 Qs/Qt
45
500
40
400
35
(torr)
(%)
300 30
25
200
20
100
15
0 10
0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70
Volpl (mL/kg) Volpl (mL/kg)
Figure 6.4 Mean arterial pressure (MAP), arterial partial pressure of O2 and CO2 (PaO2, PaCO2) and intrapulmonary shunt (Qs/Qt) in pigs.
Solid circles indicate increasing intrapleural volume (Volpl) and open triangles indicate decreasing Volpl. Reprinted with permission from
Reference 12.
ventilation were subjected to a therapeutic thoracentesis investigators also showed that the main mechanism
(mean 1495 mL pleural fluid) and immediately after tho- underlying arterial hypoxemia in patients with pleural
racentesis there was no significant change in the lung com- effusion is an intrapulmonary shunt, which does not
pliance or the alveolar–arterial oxygen gradient (AaPO2).15 change significantly after thoracentesis.
The dissimilar results in the two studies may be due to the There have been four separate studies that examined
time the measurements were obtained or the poorer oxy- the effects of position on the oxygenation status of patients
genation status of the patients in the first study. with pleural effusions.17–20 In each of the four studies the
Agusti et al.16 attempted to determine the mechanisms oxygenation status was slightly better when the patients
for the hypoxemia associated with pleural effusions and were positioned with the side of the effusion superior
the effects of thoracentesis on the hypoxemia by studying (Table 6.2). The improvement was thought to be due to
the oxygenation status of nine patients before and imme- gravity distributing more blood to the lung that was not
diately after thoracentesis (mean 693 ± 424 mL) with the partially compressed by the pleural effusion. However, the
multiple inert gas technique. They reported that the mean differences in the mean levels of the oxygenation were not
difference in the AaPO2 was 29 mmHg before thoracente- statistically significant in any of the studies and are proba-
sis and remained at 29 mmHg post thoracentesis. These bly not clinically significant either.
Table 6.2 Effects of position on oxygenation status of patients with pleural effusion
Sonnenblick et al.17 8 66.7 ± 8.7 mmHg (PO2) 71.9 ± 9.3 mmHg (PO2)
Chang et al.19 21 66.0 ± 66.0 mmHg (PO2) 69.6 ± 14.6 mmHg (PO2)
Romero et al.20 33 78.0 ± 12.5 mmHg (PO2) 81.4 ± 8.5 mmHg (PO2)
Neagley et al.18 10 93.4 ± 2.1% (SaO2) 94.7 ± 2.1% (SaO2)
PO2, partial pressure of O2; SaO2, arterial O2 saturation.
54 Physiology: changes with pleural effusion and pneumothorax
EFFECTS OF EFFUSION ON EXERCISE indicated by the high resting pulse (Table 6.3) and the
TOLERANCE reduced oxygen pulse (O2 pulse), which is a reflection of
the stroke volume.
There has been limited research on the effects of a pleural The performance of a therapeutic thoracentesis (mean
effusion on the exercise tolerance of patients. We obtained 1612 mL) had relatively little influence on the exercise tol-
maximum exercise tests on 15 patients before and after erance of the 15 patients (Table 6.3). Although the mean
they underwent a therapeutic thoracentesis. The FEV1 and FVC both improved significantly (Table 6.3),
symptom-limited exercise tests were conducted on a there was no significant change in the maximum workload
bicycle ergometer with 15 watt increments every or the mean maximum oxygen consumption per minute
·
minute.21,22 The mean age of the patients was 64.7 and (VO2max). Overall, after thoracentesis, five patients had an
most of them had malignant pleural effusions. Pre-thora- improvement in their workload, five patients had a
centesis, the mean FEV1 and FVC were only 43 ± 17 decrease in the workload and five patients had no change
percent and 49 ± 17 percent of predicted, respectively. in the workload. The change in exercise capacity was not
Seven of the 15 had obstructive lung disease, as reflected by significantly correlated with the amount of fluid removed
an FEV1/FVC ratio of less than 0.70. or with the changes in pleural pressure. However, there
The exercise tolerance of these elderly patients was sig- was a significant correlation between changes in the
·
nificantly reduced prior to the thoracentesis. The mean VO2max and changes in the FEV1 (r = 0.576, p < 0.05),
maximum workload prior to thoracentesis was only 79 changes in the FVC (r = 0.610, p < 0.05) and changes in the
watts (43 percent of predicted) while the mean maximum maximum O2 pulse (r = 0.78, p < 0.05).
· In summary, based on this series, elderly patients with
oxygen consumption per minute (VO2) was only
907 mL/minute (37 percent of predicted) (Table 6.3). moderate to large pleural effusions have a marked reduc-
When the individual exercise tests were examined, the tion in their exercise capacity. The lung function, as
explanation for the reduced exercise tolerance was not reflected by the FEV1 and the FVC, and the cardiac func-
obvious. Eight of the patients appeared to be ventilatory tion, as reflected by the O2 pulse, are both reduced and
· contribute to the exercise limitation. However, the per-
limited (minute ventilation at maximum exercise [VE max]
greater than 80 percent of predicted maximum at exhaus- formance of a therapeutic thoracentesis does not result in
tion), and four of these also appeared to be cardiac limited a significant improvement in exercise tolerance of many
(maximum heart rate greater than 80 percent of predicted patients.
at exhaustion). There were two additional patients who
appeared to be only cardiac limited. At the maximum tol-
erated workload (Emax) the remaining five patients EFFECTS OF EFFUSION ON THE DIAPHRAGM
appeared to be neither ventilatory nor cardiac limited. In
general, the patients’ ventilation was inefficient, as shown When pleural fluid is present, the diaphragm on the side of
· · the effusion is profoundly affected by the weight of the
by their high ventilatory equivalents for oxygen (VE/VO2)
· · fluid on the diaphragm. The changes in the diaphragm
and carbon dioxide (VE/VCO2) (Table 6.3). In addition,
the patients’ cardiac function appeared to be impaired, as have been classified into three categories by Mulvey23
Table 6.3 Results of maximal exercise tests before and after a therapeutic thoracentesis in 15 patients from whom a mean of 1612 mL
pleural fluid was removed
FEV1, L (%pred) 1.56 ± 0.63 (43%) 1.74 ± 0.69 (47%) 0.18 ± 0.23 0.007
FVC, L (%pred) 2.32 ± 0.76 (49%) 2.63 ± 0.81 (56%) 0.31 ± 0.43 0.013
Max Work, watts (%pred) 77.7 ± 44.5 (43%) 79.0 ± 40.7 (44%) 1.3 ± 19.4 0.794
·
VO2 max, mL/min (%pred) 992 ± 431 (41%) 1038 ± 395 (43%) 46 ± 226 0.449
·
VE max, L/min (%pred) 45.1 ± 20.2 (79%) 48.2 ± 18.8 (77%) 3.1 ± 11.8 0.321
· ·
VE/VO2 max (%pred) 46.1 ± 9.9 (158%) 47.3 ± 12.0 (162%) 1.2 ± 5.2 0.394
· ·
VE/VCO2 max (%pred) 45.6 ± 7.4 (172%) 44.7 ± 8.1 (1.68%) -0.9 ± 4.7 0.454
HR rest, bpm 93.4 ± 16.6 93.6 ± 17.2 0.2 ± 12.9 0.953
HR max, bpm (%pred) 120.7 ± 15.6 (78%) 114.6 ± 17.3 (74%) -6.1 ± 10.6 0.049
O2 pulse rest, mL/beat 3.28 ± 0.72 3.38 ± 0.53 0.11 ± 0.67 0.547
O2 pulse SW, mL/beat 8.02 ± 2.94 8.59 ± 2.68 0.58 ± 1.14 0.083
O2 pulse max, mL/beat (%pred) 8.21 ± 2.99 (61%) 9.04 ± 3.00 (67%) 0.83 ± 1.57 0.070
· ·
FEV1, forced expiratory volume in 1 second; FVC, forced vital capacity; V O2, oxygen consumption per minute; V E max, minute ventilation at maximum
·
exercise; V CO2, CO2 respired per minute.
Effects of pneumothorax on pleural pressure 55
based on the findings on the plain film and fluoroscopy. In ray).26 It was unusual to have chamber collapse with iso-
the first, or least severe category, the hemidiaphragm is lated right pleural effusion. Hemodynamics were studied in
domed and functions normally. Patients in this category 22 mechanically ventilated patients with moderate to large
are usually asymptomatic even though the effusion may be pleural effusions before and after they underwent drainage
large. In the second category, the diaphragm is flattened of the effusions with a pigtail catheter by Ahmed et al.27
and does not move with respiration. Patients in this cate- They reported that the mean cardiac output increased from
gory frequently complain of dyspnea, which is likely to be 7.7 to 8.4 L/minute, but that this change was not statistically
relieved with a therapeutic thoracentesis. In the third cate- significant. However, the pulmonary capillary wedge pres-
gory, the diaphragm is inverted and there may be paradox- sure and the central venous pressure both decreased
ical movements on respiration. Patients in this category significantly after the pleural fluid was drained.
usually have severe dyspnea that is markedly relieved with There have been three reports of patients with compro-
a therapeutic thoracentesis. mised cardiac output attributed to large pleural effu-
The percentage of patients in each of the three cate- sions.28–30 Negus et al.28 reported a 60-year-old woman
gories has not been studied carefully. It is likely that inver- who presented with a large left sided pleural effusion with
sion of the diaphragm may be more common than is marked mediastinal shift to the right. Shortly after presen-
generally realized. Wang and Tseng11 were able to docu- tation, her blood pressure became unobtainable and her
ment diaphragmatic inversion in 21 patients over a 3-year carotid and femoral pulses were very weak. When a chest
period. Interestingly, when these patients underwent ther- tube was placed and 1125 mL of pleural fluid was with-
apeutic thoracentesis they experienced marked relief of drawn, the blood pressure rose to 140/86 and the pulses
their dyspnea. became bounding.28. Kisanuki et al.29 reported a 68-year-
old man who presented with a blood pressure of 90/60 and
a large left encapsulated pleural effusion. A two-dimen-
EFFECTS OF EFFUSION ON THE HEART sional echocardiogram revealed that the pleural effusion
compressed the lateral wall of the left ventricle. With M-
The presence of pleural fluid may also adversely influence mode echocardiogram, the left ventricular collapse was
cardiac function because the increase in the pleural pres- observed throughout diastole. After drainage of 500 mL of
sure can decrease the distending pressures of the heart pleural fluid, the blood pressure rose from 90/60 to 120/80
chambers. Vaska et al.24 studied seven spontaneously and the left ventricular collapse during diastole resolved.29
breathing dogs with a two-dimensional echocardiograph Kopterides et al.30 reported two patients who had hemody-
during infusions of saline into both pleural spaces. They namic compromise with large left-sided effusions in whom
reported that right ventricular diastolic collapse began left ventricular diastolic collapse was demonstrated by
when the mean pleural pressure increased by 5 mmHg. transthoracic echocardiograph. It is interesting that all
When the mean pleural pressure had increased by four of the patients in the above three reports had left sided
15 mmHg, the stroke volume had fallen by nearly 50 effusions. It is probable that the increased pleural pressure
percent and the cardiac output had fallen by 33 percent.24 resulting from the pleural fluid is responsible for the
In contrast, Nishida et al.12 reported that the infusion of decreased cardiac output.
20 mL/kg saline into each pleural space had no effect on
the cardiac output in anesthetized pigs. It should be noted
that Vaska et al.24 infused more than 50 mL/kg into the EFFECTS OF PNEUMOTHORAX ON PLEURAL
pleural spaces of their dogs. PRESSURE
It appears that the presence of a large effusion frequently
adversely affects cardiac function. In a study of 27 patients When a pneumothorax is present, the pleural pressure
who had more than a hemithorax occupied by pleural fluid, increases as it does with the presence of a pleural effusion.
Traylor et al.25 reported that eight subjects had elevated However, with a pneumothorax, the pressure is the same
jugular venous pressure, eight had pulsus paradoxus, six throughout the entire pleural space. In contrast, with a
had right ventricular diastolic collapse and 23 had flow pleural effusion there is a gradient in the pleural pressure
velocity paradoxus. Post thoracentesis or chest tube place- due to the hydrostatic column of fluid so that the pleural
ment, all these abnormalities resolved in all patients but one pressure in the dependent part of the hemithorax is much
in whom only 900 mL pleural fluid was withdrawn.25 greater than it is in the superior part of the hemithorax.
Sadaniantz et al.26 reviewed the echocardiograms of 116 Another way to look at this is that with a pneumothorax
patients with pleural effusion and observed cardiac the lung sinks to the bottom of the hemithorax because it
chamber collapse in 21 (18 percent). All had right atrial col- is heavier than air, while with a pleural effusion, the lung
lapse while one had concomitant right ventricular collapse, rises to the top of the hemithorax because it is lighter than
four had left atrial collapse and none had left ventricular fluid and is floating in the fluid. The net result is that with
collapse.26 Of the 21 patients with chamber collapse, 13 had a pneumothorax, the upper lobe is affected more than the
large, three had moderate, two had small and three had lower lobe while with a pleural effusion the lower lobe is
unknown amounts of left pleural effusion (no chest X- affected more than the upper lobes.
56 Physiology: changes with pleural effusion and pneumothorax
60
50
Pressure volume curve
lung
C
Thoracic cavity
40
volume (8% VC)
Vital capacity (%)
in five (4 percent).35 Of course, the abnormalities in the to less than 40 mmHg. The development of a tension
blood gases may have been due at least in part to the pneumothorax in man is also associated with impaired
underlying lung disease in this study.35 hemodynamics. Beards and Lipman37 recorded the hemo-
When the air is evacuated from patients with a pneu- dynamics of three patients who developed a tension pneu-
mothorax, the oxygenation levels return to normal more mothorax while on mechanical ventilation. The mean
slowly than they do in the experimental animals. Norris et cardiac outputs, which were 7.3, 4.8 and 3.6 L/minute/m2
al.34 reported that in patients with an initial anatomic at baseline, fell to 3.0, 3.1 and 1.4 L/minute/m2 with the
shunt above 20 percent, there was a reduction of at least 10 development of the tension pneumothorax. In a similar
percent in their shunt 30–90 minutes after tube thoracos- manner, the mean arterial pressures that were 97, 96 and
tomy was performed. Nevertheless, all patients had a shunt 68 mmHg fell to 33, 68 and 57 mmHg, respectively. The
above 5 percent at 90 minutes. Moreover, three additional probable mechanism for the decreased cardiac output is
patients with anatomic shunts between 10 and 20 percent decreased venous return due to the increased pleural pres-
had no change in their shunt when the air was removed.34 sures.
EFFECTS OF PNEUMOTHORAX ON
DIAPHRAGMATIC FUNCTION KEY POINTS
To my knowledge, there have been no studies evaluating ● When fluid or air is present in the pleural space,
the effects of a pneumothorax on diaphragmatic function. the lung must get smaller or the hemithorax
I would anticipate that the presence of a pneumothorax must get larger, or a combination of the two
would have less effect on the diaphragmatic function than must occur.
would a pleural effusion of comparable volume, since the ● When the patient is upright, the presence of fluid
pleural pressure would increase much more with the in the pleural space affects mostly the lower lobes
pleural fluid. The diaphragmatic inversion, which is seen while the presence of air in the pleural space
relatively frequently with a pleural effusion, is not seen affects the upper lobes more than the lower lobes.
with pneumothorax. With a tension pneumothorax, the ● When a thoracentesis is performed, the vital
diaphragm may be displaced inferiorly due to the capacity increases by an average of 20 mL for
increased pleural pressure but the functional significance each 100 mL fluid removed, but there is marked
of this displacement is not known. inter-individual variation.
● A thoracentesis has variable effects on blood
gases.
EFFECTS OF PNEUMOTHORAX ON EXERCISE ● The exercise tolerance of many patients with
TOLERANCE pleural effusion does not improve after a thera-
peutic thoracentesis.
There have been no studies on the effects of a pneumo- ● The presence of a pleural effusion can lead to
thorax on the exercise tolerance of either animals or man. diaphragmatic inversion which is associated with
However, it would be anticipated that the exercise toler- marked dyspnea.
ance would be markedly impaired since many patients are ● The presence of a pleural effusion can compress
dyspneic at rest. the heart and lead to a compromised cardiac
output.
4. Neff TA, Buchanana BD. Tension pleural effusion: a delayed thoracentesis improve the exercise capacity of patients with
complication of pneumothorax therapy in tuberculosis. Am Rev pleural effusion? Am Rev Respir Dis 1987; 135: A244.
Respir Dis 1973; 111: 543–8. 22. Shinto RA, Stansbury DW, Fischer CE, Light RW. The effect of
5. Feller-Kopman D, Walkey A, Berkowitz D, et al. The relationship of thoracentesis on central respiratory drive in patients with large
pleural pressure to symptom development during therapeutic pleural effusions. Am Rev Respir Dis 1988; 137: A112.
thoracentesis. Chest 2006; 129: 1556–60. ●23. Mulvey RB. The effect of pleural fluid on the diaphragm. Radiology
●6. Lan RS, Lo SK, Chuang ML, et al. Elastance of the pleural space: a 1965; 84: 1080–86.
predictor for the outcome of pleurodesis in patients with ●24. Vaska K, Wann LS, Sagar K, Klopfenstein HS. Pleural effusion as a
malignant pleural effusion. Ann Intern Med 1997; 126: 768–74. cause of right ventricular diastolic collapse. Circulation 1992; 86:
●7. Krell WS, Rodarte JR. Effects of acute pleural effusion on 609–17.
respiratory system mechanics in dogs. J Appl Physiol 1985; 59: 25. Traylor JJ, Chan K, Wong I, et al. Large pleural effusions producing
1458–63. signs of cardiac tamponade resolved by thoracentesis. Am J
8. Light RW, Stansbury DW, Brown SE. The relationship between Cardiol 2002; 89: 106–8.
pleural pressures and changes in pulmonary function following 26. Sadaniantz A, Anastacio R, Verma V, et al. The incidence of
therapeutic thoracentesis. Am Rev Respir Dis 1986; 133: 658–61. diastolic right atrial collapse in patients with pleural effusion in
9. Estenne M, Yernault J-C, De Troyer A. Mechanism of relief of the absence of pericardial effusion. Echocardiography 2003; 20:
dyspnea after thoracocentesis in patients with large pleural 211–15.
effusions. Am J Med 1983; 74: 813–9. 27. Ahmed SH, Ouzounian SP, Dirusso S, et al. Hemodynamic and
◆10. Light RW. Pleural diseases, 5th edn. Baltimore: Lippincott, pulmonary changes after drainage of significant pleural effusions
Williams and Wilkins, 2007. in critically ill, mechanically ventilated surgical patients. J Trauma
●11. Wang JS, Tseng CH. Changes in pulmonary mechanics and gas 2004; 57: 1184–8.
exchange after thoracentesis on patients with inversion of a 28. Negus RA, Chachkes JS, Wrenn K. Tension hydrothorax and shock
hemidiaphragm secondary to large pleural effusion. Chest 1995; in a patient with a malignant pleural effusion. Am J Emerg Med
107: 1610–14. 1990; 8: 205–7.
●12. Nishida O, Arellano R, Cheng DC, DeMajo W, Kavanagh BP. Gas 29. Kisanuki A, Shono H, Kiyonaga K, et al. Two-dimensional
exchange and hemodynamics in experimental pleural effusion. Crit echocardiographic demonstration of left ventricular diastolic
Care Med 1999; 27: 583–7. collapse due to compression by pleural effusion. Am Heart J 1991;
●13. Brandstetter RD, Cohen RP. Hypoxemia after thoracentesis: a 122: 1173–5.
predictable and treatable condition. J Am Med Assoc 1979; 242: 30. Kopterides P, Lignos M, Papanikolaou S, et al. Pleural effusion
1060–61. causing cardiac tamponade. Report of two cases and review of the
14. Talmor M, Hydo L, Gershenwald JG, Barie PS. Beneficial effects of literature. Heart Lung 2006; 35: 66–7.
chest tube drainage of pleural effusion in acute respiratory failure ●31. Rutherford RB, Hurt HH, Brickman RD, Tubb JM. The
refractory to positive end-expiratory pressure ventilation. Surgery pathophysiology of progressive, tension pneumothorax. J Trauma
1998; 123: 137–43. 1968; 8: 212–27.
15. Doelken P, Abreu R, Sahn SA, et al. Effect of thoracentesis on 32. Moran JF, Jones RH, Wolfe WG. Regional pulmonary function
respiratory mechanics and gas exchange in the patient receiving during experimental unilateral pneumothorax in the awake state. J
mechanical ventilation. Chest 2006; 130: 1354–61. Thorac Surg 1977; 74: 396–402.
16. Agusti AG, Cardus J, Roca J, et al. Ventilation–perfusion mismatch 33. Anthonisen NR. Regional function in spontaneous pneumothorax.
in patients with pleural effusion: effects of thoracentesis. Am J Am Rev Respir Dis 1977; 115: 873–6.
Respir Crit Care Med 1997; 156: 1205–9. 34. Norris RM, Jones JG, Bishop JM. Respiratory gas exchange in
17. Sonnenblick M, Melzer E, Rosin AJ. Body positional effect on gas patients with spontaneous pneumothorax. Thorax 1968; 23:
exchange in unilateral pleural effusion. Chest 1983; 83: 784–6. 427–33.
18. Neagley SR, Zwillich CW. The effect of positional changes on 35. Light RW, O’Hara VS, Moritz TE, et al. Intrapleural tetracycline for
oxygenation in patients with pleural effusions. Chest 1985; 88: the prevention of recurrent spontaneous pneumothorax. J Am Med
714–17. Assoc 1990; 264: 2224–30.
19. Chang SC, Shiao GM, Perng RP. Postural effect on gas exchange in 36. Carvalho P, Hilderbrandt J, Charan NB. Changes in bronchial and
patients with unilateral pleural effusions. Chest 1989; 96: 60–3. pulmonary arterial blood flow with progressive tension
20. Romero S, Martin C, Hernandez L, et al. Effect of body position on pneumothorax. J Appl Physiol 1996; 81: 1664–9.
gas exchange in patients with unilateral pleural effusion: 37. Beards SC, Lipman J. Decreased cardiac index as an indicator of
influence of effusion volume. Respir Med 1995; 89: 297–301. tension pneumothorax in the ventilated patient. Anaesthesia
21. Shinto RA, Stansbury DW, Brown SE, Light RW. Does therapeutic 1994; 49: 137–41.
7
Pleural inflammation and infection
DEFENSE MECHANISMS OF THE PLEURA mesothelial cells to the extracellular matrix and to other
cell types.34 Mesothelial cells express a2, a3, a5, b1, b3,
Consequences of microbial invasion of the and anb 3 integrins in high quantities. Expression of a1 is
pleural space noted in intermediate quantities, while expression of a6 is
low and seen in less than 30 percent of mesothelial cells.35
A massive attack by microbial factors in the pleural space The ability of mesothelial cells to adhere to its extracellular
has several deleterious consequences if not aggressively matrix is a function of some of these integrins.
countered. Microbial invasion of the pleural space results Importantly, mesothelial cells also express membrane
in increased volume of pleural fluid with increased protein type matrix metalloproteinase (MT-MMP)-1, 2, 3 and 9,
concentration and cellularity. Pleural infection also results and tissue inhibitor of metalloproteinase (TIMP)-I, -II and
in structural and functional changes of the mesothelium -III. In mesothelial cells, the differential expression of
and its basement membrane. Clearly, pleural infection is a MMPs and TIMPs is influenced by the pro-inflammatory
continuum from the early development of a small sterile cytokine IL-1b and the anti-inflammatory cytokine TGF-
parapneumonic effusion to the actual invasion of the b1.36,37 A balance of these factors is important in inhibiting
pleural space by microorganisms. Under favourable condi- adherence of other cells as well as in maintaining the
tions, infection of the pleural space resolves with steriliza- integrity of the pleural monolayer.
tion of the pleural fluid and a return to normal Multiple membrane-bound mucin-like molecules have
homeostasis with normal pleural fluid and normal pleural been demonstrated on the pleural surface. Pleural
structure. Under unfavorable circumstances, infection of mesothelial cells produce a significant amount of sialo-
the pleural space resolves with residual pockets of infected mucins and proteoglycans that coat the surface of the
proteinaceous fluid contained by thickened fibrotic pleura mesothelium.38 The membrane-bound mucins have a
and adhesions between opposing pleural surfaces. hydrophobic membrane-spanning domain and an extra-
Drainage of the pleural space prior to the development cellular domain containing numerous serine and threo-
of significant adhesions is a critical method of removing nine residues. The extracellular domain of the mucins is
large numbers of bacteria from the pleural space, as well as abundantly glycosylated by oligosaccharides O-linked to
proteolytic material that may cause further damage to the the serine and threonine residues.39–43 Thus, mucins are
mesothelium. The use of fibrinolytics is also based on the very large molecules that contain 60–80 percent carbohy-
same principles, namely lysis of fibrin pockets which drate by weight.44 Mucins serve as lubricants and protect
allows for removal of pus and proteases from the infected the underlying mesothelial cells. The extracellular domain
pleural space. Appropriate antibiotic therapy, initiated of the mucins is anionic, thus the free surface of the
early in the process of development of pleural infection, mesothelium has a negative charge. Mutual repulsion of
however, remains the major mechanism of achieving the negatively charged surfaces maintains the pleural
microbial killing and mitigating their effect on the pleura. cavity. The negative charge also repels bacteria.45
(For management of empyema, see also Chapter 26, Mesothelial cells also produce fibronectin. Fibronectin is a
Effusions from infections: parapneumonic effusions and large glycoprotein that resists adherence of microbes, such
empyema.) as P. aeruginosa and others.
Sialomucin complex (SMC) is a family of het-
erodimeric glycoproteins expressed by mesothelial cells on
their surface. These sialomucins contain abundant sialic
The pleural mesothelium as a physical barrier acid, accordingly they are strongly anionic and they resist
the adherence of bacteria and inflammatory cells.
The pleural mesothelium is an important host defense. As Members of the SMC family include CD34, ascites sialo-
a physical barrier the pleura may be seen to comprise mul- glycoprotein (ASGP)-1 and podocalyxin.46 CD34 is
tiple elements: the basement membrane, the mesothelial important for its ability to act as a ligand of L-selectin.
cell monolayer, the extracellular domain of the mesothe- The pleural membrane is also bathed in secretions con-
lium and the pleural fluid bathing the pleural surface. The taining microbicidal proteins, such as lysozyme. Lysozyme
physical properties of each element are critically important is an enzyme that degrades the peptidoglycan in the cell
to preventing the entry of organisms into the pleural space. wall of Gram-positive bacteria by hydrolyzing the b-(1,4)-
The pleural basement membrane comprises proteins, glycosidic linkage from N-acetyl muramic acid to N-acetyl
laminin, fibronectin, elastin and collagen types I, III and glucosamine.47 Normal pleural fluid also contains
IV,2 and the hyaluronan (a glycosaminoglycan). immunoglobulins, principally IgG and IgA. Pleural fluid
The pleural mesothelium is an intact monolayer of cells contains complement which is a group of proteins that
that adhere to each other and the basement membrane. interact with each other in a cascade when activated.48
The integrity of the pleural mesothelium is dependent on Complement activation can lead to microbial lysis and
the organization of the intermesothelial adherens junc- may also play a role in amplifying inflammation with
tions. Human mesothelial cells express a large number of cytokine production and increased phagocytosis of cells.49
integrins which play a critical role in the adherence of Although the formation of pleural fluid has been thought
62 Pleural inflammation and infection
to represent a detrimental effect of pleural infection, it not very well recognized, but is a critical response to the
might be considered as an appropriate response to infec- presence of microbes. As a consequence of phagocytosis, a
tion inasmuch as there is an exuberant flow of proteins number of responses are initiated by the mesothelial cell.
and cells into the pleural space that may kill or inactivate An important and immediate response is the release of
the infectious organism. toxic oxygen metabolites such as hydrogen peroxide,
superoxide and nitric oxide.58 The release of oxygen radi-
cals is also associated with the release of other antibacterial
molecules such as cathepsin-G and defensins, which kill
Immune responses of mesothelial cells microbes and some viruses by permeabilizing their mem-
branes.59 Phagocytosis of the microbe may be made more
The inflammatory stimuli to the mesothelium include efficient by the coating of particles with antibody. All three
microbes or microbial products, allergens, autoantigens, classes of IgG Fc receptors can mediate phagocytosis.60
alloantigens, tumor cells, etc. Interaction between the When a microbe is coated with antibody as well as comple-
inflammatory stimuli and the mesothelial cell are medi- ment, even encapsulated bacteria will be phagocytosed by
ated by membrane-spanning proteins expressed on the mesothelial cells and phagocytic cells.61 Antibodies pro-
mesothelial cell surface. In the presence of inflammatory duced by mesothelial cells also inhibit the ability of extra-
stimuli, the mesothelial cell itself may act as an effector cell cellular viruses to infect either the recruited phagocyte or
or it may signal other cell types to participate in a coordi- the mesothelial cell. Antibody responses, however, are less
nated reaction. Immune responses of the mesothelial cell effective against viruses such as the human immunodefi-
orchestrate a combined response comprising both innate ciency virus (HIV), which can enter the mesothelial cell
and acquired immunity. through multiple receptors, including C-C chemokine
The various classes of pathogens (Gram-negative bacte- receptor-5 (CCR5).
ria, Gram-positive bacteria, mycobacteria, fungi, parasites, Pleural mesothelial cells also recruit inflammatory cells
viruses, etc.) have characteristic molecular structures upon to the pleural space by secreting chemokines and regulate
their surfaces, such as lipopolysaccharide (LPS), bacteria their activity by secreting cytokines. Pleural mesothelial
lipoprotein (BLP), flagellin, peptidoglycan (PGN), LTA, cells can secrete several critically important chemokines of
lipoarabinomannan (LAM), viral glycoproteins, etc. both the C-X-C, C-C and C-X3-C families. Human pleural
Collectively, these microbial molecular structures are mesothelial cells activated by LPS secrete monocyte
called pathogen-associated molecular patterns (PAMPs). chemotactic protein-1 (MCP-1) and interleukin-8 (IL-
Pleural mesothelial cells express pattern recognition recep- 8).62 MCP-1 is a member of the C-C chemokine family and
tors (PRRs).50 These allow the mesothelial cell to respond is chemotactic for monocytes. IL-8 is a member of the C-
to the various classes of pathogen with specificity.51 X-C family and is chemotactic for neutrophils. Mesothelial
Among the PRRs on pleural mesothelial cells are b2 inte- cells also express fractalkine, the only known member of
grins (CD11, CD18) complement receptors, C-type lectins the C-X3-C chemokine family. In its soluble form,
and toll-like receptors (TLRs).52 Interaction of PAMPs fractalkine is chemotactic for T-cells, natural killer (NK)
with PRRs triggers a defensive reaction mobilizing the cells and monocytes; in its membrane bound form it pro-
both innate and adaptive immune responses, which could motes adhesion by those cell types.63 Mesothelial cells also
include the release of reactive oxygen intermediates (ROIs) coordinate with the acquired immune response by presen-
or reactive nitrogen intermediates (RNIs), or secretion of tation of antigen.64
antimicrobial peptides, inflammatory cytokines and Apoptosis, or programmed cell death, is thought to
chemokines.53 contribute to the homeostasis of the functional leukocyte
One of the innate immune responses of the pleural pool in the pleural space. Interestingly, pleural fluid of
mesothelial cell is the release of ROIs and RNIs. The nitric patients with complicated parapneumonic effusions
oxide radical (∑NO) is a diatomic molecule containing an contain significantly high levels of granulocyte/
unpaired electron that permits it to react with other mole- macrophage colony stimulating factor (GM-CSF), and
cules.54,55 The reaction of ∑NO with superoxide anion leads neutrophils in empyema pleural fluids demonstrate a
to formation of the peroxynitrite anion (ONOO-) and per- decrease in apoptosis when compared with neutrophils in
oxynitrous acid. The inducible form of NO synthase (iNOS) uncomplicated parapneumonic effusions.65 Neutrophils
is capable of producing micromolar quantities of ∑NO over exposed to mesothelial cells express the Bcl-xL gene, which
a prolonged period. Pleural mesothelial cells produce large is an anti-apoptotic gene.66 This anti-apoptotic gene is the
quantities of ∑NO in response to stimulation with cytokines, counterpart of Bak gene expression.67 Thus, mesothelial
LPS and particulates.56 Thus the inducible isoform of NO cells can control the initiation of the inflammatory
synthase contributes to the control of a variety of infections response to infection in the pleural space, and may also
in the pleural space. control the resolution of the inflammatory response by
Pleural mesothelial cells are actively phagocytic cells.57 regulating the changes in the level of the Bak gene expres-
This phenomenon of phagocytosis by mesothelial cells is sion.
Pleural inflammatory cells 63
Invasion by phagocytic cells linking innate and acquired immunity are multiple. The
TLRs, which function as pattern recognition receptors on
Over 100 years ago Elie Metchnikoff described the acute mesothelial cells, play an essential role in recognition of
inflammatory response as a reaction of phagocytes against microbial components. These TLRs allow the cells to
a harmful agent.68 This theory is classically demonstrated present antigen to naive T cells, which in turn regulate the
in the pleural space, since there is a rapid and site-directed development of Th1/Th2 cell development. The TLRs
movement of leukocytes into the pleura following inflam- induce the production of cytokines such as
mation. This transfer of leukocytes from the vascular com- IL-12 and IL-18 in antigen-presenting cells. These
partment into the pleural space involves a multistep cytokines are instructive cytokines and drive the naive T
paradigm of leukocyte recruitment involving margination, cell to differentiate into Th1 cells. The TLRs also recognize
capture of the free-flowing leukocytes in the vascular com- the presence of LPS, PGN and glycolipids. Activation of
partment via leukocyte rolling, activation and movement TLRs is involved in the recognition of M. tuberculosis as
to the surface of the pleura. When the leukocyte encoun- well as in the killing of these organisms.72 Th1 T cells can
ters the pleura, it initiates a similar process, but with the produce interferon-gamma (IFN-g) in the presence of
pleural mesothelial cell instead of the endothelial cell, and effector cytokines, while Th2 cells will produce IL-4, IL-5,
initiates movement under the direction of chemokines, IL-10 and IL-13. T cells can be functionally divided into
from the basilar surface of the mesothelium out onto the cells that provide help for other immune cell types such as
apical surface of the mesothelium and into the pleural B cells and cells that mediate cytotoxicity.73 Helper T cells
space. Malignant cells that invade the pleural space activate express the glycoprotein CD4 while cytotoxic T cells
similar responses.69 Mesothelial cell expression of adhe- express CD8. CD4+ and CD8+ T cells can be selectively
sion molecules, including the immunoglobulin superfam- called into the pleural space in response to infections, such
ily cell adhesion molecules (IGSF CAMs), such as the as tuberculosis. When mature T cells are exposed to an
intercellular adhesion molecules (ICAMs), selectins, such antigen, such as a mycobacterial antigen, e.g. the 65 kilo-
as L-, P- and E-selectin, and other adhesion molecules, dalton mycobacterial heat shock protein or lipoarabino-
such as CD44, come into play during the movement of mannan (LAM), their function and phenotype changes
cells into the pleural space. L-selectin is constitutively and remains persistently changed thenceforth. Enhanced
expressed on almost all leukocytes as well as mesothelial responses by these T cells can be observed for decades after
cells. Selectins are monomeric molecules that span the the initial exposure to the antigen. These CD4+ ab T cells
plasma membrane and contain complement-controlled are contained predominantly within a subset that com-
protein-like repeats.70 The calcium-dependent lectin prises approximately 40 percent of these cells in the adult
domain at the NH2-terminus defines their ability to bind circulation.
to specific ligands. Integrin expression by leukocytes Monocytes and macrophages are important sources of
allows for firm adhesion between the invading phagocytic a specific group of cytokines. A classic example of this is
and the mesothelial cell. In particular, b2 integrin IFN-g, which, in tuberculous effusions, is found in very
(CD11/CD18) on the surface of neutrophils binds to the high quantities in the pleural space, as are CD4+ T cells.74
ICAMs that are expressed by mesothelial cells. These IFN-g not only improves phagocytosis and killing of
ICAMs, namely ICAM-1 (CD54), ICAM-2 (CD102), mycobacteria, it may also mediate the level of expression
ICAM-3 (CD50) and VCAM-1 (CD106), are upregulated of chemokine receptors on the recruited mononuclear
on mesothelial cells during transfer of neutrophils, phagocyte. Thus, while in the vascular compartment, the
mononuclear cells and lymphocytes into the pleural mononuclear cell has a high level of CCR2 expression
space.71 when it enters the pleural space, the presence of IFN-g
downregulates the expression of CCR2, in effect capturing
it and localizing it to the pleural space. T lymphocytes are
Acquired immune responses in the pleural important sources of other cytokines, such as IL-2 and
space IL-9, and lymphotoxins.
stem cells in the bone marrow. Several low molecular cells are in the peripheral blood circulation, but is signifi-
weight proteins such as colony-stimulating factor are cantly reduced when the mononuclear cells reach the
responsible for the production, maturation and prolifera- pleural space. This mechanism serves to localize and
tion of these cells.75 Neutrophils are armed with an immobilize the monocytes once they reach the pleural
azurophil or primary granules as well as specific or second- cavity.82 Also, the C-X3-C chemokine fractalkine in its
ary granules. These granules contain peroxidases, phos- soluble form is chemotactic for monocytes and in its mem-
phatases and gelatinases that, when in contact with the brane-bound form promotes strong adhesion by mono-
microorganism, allow for microbial killing and diges- cytes.83
tion.76 Secondary granules are true lysosomes and contain
acid hydrolases as well as proteases and cationic proteins.77
Neutrophils enter the pleural space via diapedesis Eosinophils
across the pleural monolayer. The pattern of migration is
not unique since it closely follows the migratory patterns Eosinophils are identified as playing an important role in
and pathways in inflammation where recruited neu- the pathogenesis of idiopathic or allergic responses to
trophils traffic. pleural injury, parasitic diseases, the presence of air or
Specific granules contain the CD11b/CD18 receptor. blood in the pleural space and in hypersensitivity
CD11b/CD18 receptor mediates neutrophil chemotactic responses to certain drugs.84 The mechanisms and control
activity. Although only approximately 5 percent of the of the accumulation of these cells in the pleural space is
granulocyte pool is in the intravascular compartment, the unclear.
granulocytes can move from the vascular compartment
into the pleural space in as little as 2 hours. The classical
description of the phagocytic cell moving out of the vascu- Lymphocytes
lature involves rolling adhesion, firm adhesion and trans-
migration. Similar processes occur at the level of the The pleural space has a small population of lymphocytes,
pleural mesothelium, however, as phagocytes move from both B-lymphocytes and T-lymphocytes.85 The T-lympho-
the basilar surface of the pleura towards the apical surface cytes include gd-T cells and CD4-/CD8-ab-T cells. During
along paracellular channels and out into the pleural space. granulomatous inflammation, the number of lymphocytes
Mesothelial cells express intercellular adhesion molecules can increase dramatically reflecting either a Th-1, CD4 T-
(ICAM-1) integrins and interdigitate with the CD11/CD18 cell predominant response in diseases such as tuberculosis,
integrin on the surface of the neutrophil.78 or a CD8+ response in some diseases as in certain lym-
Although the intravascular half-life of neutrophils is phomas, etc. The production of antibodies by B cells and
6–8 hours, they may persist far longer in the pleural space. associated lymphoid tissue is important for resistance to
The lifespan of a neutrophil is short compared with other infectious processes; however, little is known about this
leukocytes and neutrophils are constitutively committed immune processing pathway in the pleural space.
to apoptosis. Interestingly, the mesothelial cell regulates
the process of neutrophil apoptosis by the release of
factors, such as GM-CSF, granulocyte colony-stimulating
factor (G-CSF) or IL-8.79 In particular, GM-CSF inhibits INITIATION OF INFLAMMATION
neutrophil apoptosis during pleural responses to infec-
tions.80 GM-CSF inhibits neutrophil apoptosis via modu- The mesothelial cell plays a critical role in the initiation of
lation of Bcl-xL. inflammatory responses in the pleural space because it is
the first cell to recognize the invasion of the pleural space.
Pleural inflammation is not only associated with an influx
Mononuclear cells of a large number of inflammatory cells, but also with a
transfer of proteins and a change in the permeability of the
Mesothelial cells recruit significant numbers of mononu- pleura. Pleural mesothelial cells release chemokines in a
clear cells to the pleural space during a variety of infec- polar fashion, with a higher concentration being released
tions. This influx may be strong and persistent in response on the apical surface, which leads to directed migration of
to infections such as tuberculosis.81 Mesothelial cells phagocytic cells into the pleural space. However, the
release several C-C chemokines, which recruit mononu- mechanisms whereby pleural integrity to proteins and
clear cells to the pleural space. Among the C-C fluid is breached with the development of an exudative
chemokines are regulated upon activation, normally T-cell high-protein-containing effusion are beginning to be elu-
expressed and secreted (RANTES) and MCP-1, -2 and -3. cidated. An infectious agent can initiate a cascade of events
The expression of C-C receptors, specifically CCR2 on which include release of nitric oxide and production of
mononuclear cells, is regulated in part by factors produced vascular permeability factor (VPF)/vascular endothelial
at local sites of inflammation such as the pleural space. growth factor (VEGF) through the accumulation of
Thus, CCR2 expression is high while the mononuclear HIF-1a. VEGF downregulates both cadherins and
Perpetuation of inflammation 65
catenins, which leads to increased pleural permeability and the most common connecting cell–cell link between
movement of cells, proteins and fluid across the interme- mesothelial cells. This cadherin family of proteins is a
sothelial gaps, leading to pleural effusion formation. major class of homophilic cell adhesion molecules that
Pleural mesothelial cell release of RNIs has been mediate calcium-dependent cell–cell interactions.99 These
demonstrated to lead to the accumulation of transcription transmembrane cadherin proteins function as a zipper
factors such as hypoxia-inducible transcription factor between cells allowing for a change in permeability to occur
(HIF)-1a as well as nuclear factor (NF)-kB. HIF-1 is a het- via signaling mechanisms that lead to contraction of the
erodimer. Both subunits are basic helix-loop-helix pro- intracellular actin cytoskeletal filaments and to gap forma-
teins containing a PAS domain containing proteins. The tion between mesothelial cells.100 When adherens junctions
first subunit is HIF-1a; the second subunit is aryl hydro- are stabilized under normal conditions, the majority of n-
carbon receptor nuclear translocator (ARNT), also known cadherin loses tyrosine phosphorylation and combines
as HIF-1b.86–90 HIF-1a accumulates under multiple with plakoglobin and actin; however, when cells have weak
pathophysiological conditions, including hypoxia, hence junctions, n-cadherin is heavily phosphorylated in tyrosine
its name. It has been documented, however, to accumulate and there is also decreased expression of b-catenin.101 Thus,
in pleural mesothelial cells that have been stimulated via n-cadherin and b-catenin are critical determinants of
release of ∑NO and other inflammatory mediators. The mesothelial paracellular permeability. This interaction is a
availability of HIF-1b is mainly determined by HIF-1a. dynamic one, since permeability is reversible. Monoclonal
HIF-1a dimerizes with HIF-1b, translocates to the antibodies directed against n-cadherin modulate pleural
nucleus, and binds to the target DNA sequence in the pro- permeability.
moter region of various genes. HIF-1a response correlates
with ∑NO formation, and administration of the iNOS
inhibitor carboxy-2-phenyl-4,4,5,5-tetramethylimidazide- PERPETUATION OF INFLAMMATION
1-oxyl-3-oxide (carboxy PTIO) to scavenge ∑NO sup-
presses HIF-1a accumulation. Among the various factors Following initiation of the inflammatory process, several
that cause an increase in accumulation of HIF-1a are mediators – cells and extracellular matrix components –
thrombin, platelet-derived growth factor and angiotensin- play a critical role in perpetuating the process of inflam-
2. mation. An example of an extracellular matrix component
A critically important regulatory cytokine in the pleural that plays a role in perpetuation of inflammation is
space is VPF/VEGF.91 This is upregulated in mesothelial hyaluronan. Hyaluronan is a high molecular weight non-
cells through an HIF-1 dependent pathway. VPF/VEGF is sulfated glycosaminoglycan produced by mesothelial cells,
a 35–43 kDa dimeric polypeptide expressed in several iso- under both normal and inflammatory conditions.
forms that result from alternative mRNA splicing of a Hyaluronan is present in the pericellular and extracellular
single gene.92 It was initially discovered because of its matrix of mesothelial cells. Hyaluronan is a linear polysac-
ability to increase vascular permeability. The molecule was charide composed of repeating alternation of N-acetyl glu-
first called vascular permeability factor. It is now recog- cosamine and glucuronide with each N-acetyl glucosamine
nized as a pivotal angiogenic factor mediating neovascu- joined to gluconuride by a b-1,4 glycosidic linkage and
larization under many conditions, but remains an each glucuronide joined to the next N-acetyl glucosamine
extremely potent inducer of permeability.93 Recent reports by a b-1,3 glycosidic linkage.102 Each molecule of hyaluro-
demonstrate that VPF/VEGF plays a central role in the for- nan may contain 10 000 or more disaccharide repeats and
mation of ascites and pleural effusions in animal models.94 have a molecular mass of several million Daltons. This
In patients with inflammatory pleural effusions such as high molecular weight hyaluron is a relatively inactive
empyema and tuberculosis, a significant amount of VEGF component but when it undergoes hydrolysis it produces
is found in the pleural fluids. Interestingly, recent investi- fragments of low molecular weight, which may mediate
gations suggest that VEGF is essential for sufficient multiple processes. Low molecular weight fragments of
formation of pleural fluid in animal models of lung hyaluronan induce mesothelial cells to express the
cancer-induced pleural effusions.95 chemokines MCP-1 and IL-8.103 Hyaluronan interacts
Bacteria such as S. aureus have been shown to cause gap with specific malignant cell receptors, such as CD44 and
formation between mesothelial cells.96 The major ubiqui- the receptor for hyaluronic acid mediated motility
tous type of mesothelial cell–cell junction are adherens (RHAMM).104 We have demonstrated that the standard
junctions. In the mesothelium the adherens junctions are isoform, CD44s, which does not contain variant exon
predominantly neural cadherin (n-cadherin) and b- product and is highly expressed in malignant breast cancer
catenins.97 Dysregulation of these proteins causes cells, is a specific mechanism whereby breast cancer cells
intercellular gap formation and increases permeability to adhere to pleural mesothelial cells.105 Addition of CD44s
cells, fluids and proteins. The mesothelium is a continuous antibody to the media blocks adherence of malignant
monolayer of cells linked by adhesive structures which are breast cancer MCF7 cells to mesothelial monolayers.106
involved in the control of pleural permeability to plasma Regulation of the degree of adherence can be regulated by
proteins, phagocytic cells, and allow for cell polarity.98 It is the presence or absence of the variant exon products. The
66 Pleural inflammation and infection
CD44–hyaluron complex is internalized by the malignant An interesting and important finding was that in patients
cells, whereupon acid hydrolases in lysosomal compart- there was a significant inverse correlation between the
ments break it down via hydrolysis to several small, acti- release of b-FGF into the pleural fluids and the tumor size,
vated low molecular weight fragments that then can as evaluated by an objective grading scale during thora-
participate in the migration process of a malignant cell coscopy. This implies that pleurodesis requires the pres-
through the pleural surface. Hyaluronan is acted upon by ence of normal mesothelial cells to release growth factors
lysosomal hyaluronidase allowing for the cleavage of for fibroblasts. Recent evidence suggests that talc also
N-acetyl hexosaminidic linkages107 with a series of saccha- causes pleural mesothelial cells to secrete endostatin,
rides being formed. The oligosaccharides are chemo- which inhibits angiogenesis, thus changing the milieu of
attractant for tumor cells. They are also growth factors for the pleural space from angiogenic to angiostatic.111
malignant cells and can mediate permeability of the Importantly, mesothelial cells also express MT-
monolayer. These fragments of hyaluronan, because of MMP-1, 2, 3 and 9, and TIMP-I, -II and -III. In mesothe-
their ability to perpetuate the process of inflammation by lial cells, the differential expression of MMPs and TIMPs is
attracting malignant cells and mononuclear cells, change influenced by the pro-inflammatory cytokine IL-1b and
the cell population in the pleural space during the process the anti-inflammatory cytokine TGF-b1.36,37 Under
of development of metastatic pleural effusions.108 normal conditions, the balance of these factors is impor-
tant in maintaining the integrity of the pleural monolayer;
in the context of recovery from infection the balance of
RESOLUTION AND REPAIR these factors determines the outcome of remodeling and
repair.
The process of resolution of pleural inflammation is just
beginning to be elucidated. It is recognized that the lifes-
pan of the inflammatory responses in the pleural space is INHIBITION OF PLEURAL INFLAMMATORY
in part regulated by the mesothelial cell through secretion RESPONSES
of antiinflammatory cytokines and regulation of apoptosis
of neutrophils, monocytes and lymphocytes. Under certain circumstances, such as malignancy and
It is important to note that pleural space inflammation acquired immune deficiency syndrome (AIDS), there is
may eventually result in either a normal pleural mesothe- an inhibition of normal pleural inflammatory responses
lial monolayer without the presence of remodeling and leading to deleterious end results for the host. The inter-
fibrosis or in the development of multiple adhesions, action of malignant and mesothelial cells, as well as their
fibrosis and loss of integrity of the pleural membranes. The extracellular matrix, is an example of one of the multiple
factors that direct the process of remodeling of the pleura pathways that allow malignant cells to elude control by
remain unclear. Resolution of pleural inflammation and the host and local regulatory cells, such as the mesothelial
pleural remodeling and fibrosis may be mediated in part cell. Tumor cells, for example, may themselves produce
by transforming growth factor beta (TGF-b). TGF-b is an large amounts of VEGF and growth factors that allow
anti-inflammatory cytokine. TGF-b2 induces mesothelial autocrine regulation of their ability to grow.
cells to synthesize collagen.109 Mesothelioma cells may themselves produce IL-8 which
Importantly, under certain circumstances, pleural scle- contains the Glu-Leu-Arg (ELR) motif making it an
rosis is a defined therapeutic goal. In patients with malig- autocrine growth factor and an angiogenic factor.112
nant effusions or in patients with pneumothorax, During angiogenesis, new vessels emerge from existing
inflammation is produced via the introduction of talc or endothelial vessels. This invasive process allows feeder
other agents into the pleural space. An exuberant fibrotic blood vessels to be generated near and within the tumor
response develops on both the surfaces of the pleura, con- tissue. Although mesothelial cells have recently been rec-
necting the two and obliterating differentiating margins. ognized as producing significant anti-angiogenic factors,
The study of this process has allowed some insight into the such as endostatin, their role in defense mechanisms
mechanisms of pleural fibrosis. Patients that had pleu- against invasion by malignant cells is still unclear. In dis-
rodesis attempted via talc insufflation had a rapid and eases such as AIDS, multiple systemic responses are weak-
marked increase in the amount of basic fibroblast growth ened which make the patient susceptible to opportunistic
factor in the pleural fluid.110 Pleural mesothelial cells stim- infections. Patients with AIDS have a higher incidence of
ulated by talc also release basic fibroblast growth factor (b- development of parapneumonic effusions, both uncom-
FGF) in vitro. This process is inhibited in vitro through the plicated and complicated, as well as a higher incidence of
use of cycloheximide, which prevents protein synthesis, or effusions secondary to diseases such as tuberculosis,96
by the use of colchicine, which prevents phagocytosis and where delayed hypersensitivity responses are key. Kaposi’s
adherence of talc. Thus, it appears that when studied in sarcoma is also known to metastasize from the lung into
vitro, mesothelial cells were required to phagocytose or the pleural space with great ease. It appears that the
adhere to talc particles prior to the release of b-FGF; mesothelial cell loses its regulatory capabilities when the
b-FGF was also actively synthesized by mesothelial cells. host is immunocompromised. In patients with tubercu-
References 67
lous pleural effusions, the mechanisms of development of the lymphatics in the parietal pleura. Am Rev Respir Dis 1975;
pleural effusions are now being described. It appears that 111: 12–20.
◆4. Sahn SA. State of the art. The pleura. Am Rev Respir Dis 1988;
not only is there loss of Th-1 cells and cytokines, but 138: 184–234.
other local immune responses regulated by the mesothe- 5. Willis T. The London practice of physics. In: Classics of medicine.
lial cell are abnormal. New York: Library Division of Gryphon, 1992: 113–20.
6. Anderson KV. Toll signaling pathways in the innate immune
response. Curr Opin Immunol 2000; 12: 13–19.
7. Hewlett EL. Toxins and other virulence factors. In: Mandell G,
Douglas R, Bennett J (eds). Principles and practice of infectious
KEY POINTS diseases, Vol. 1. New York, Edinburgh: Churchill Livingstone, 1995:
2–11.
● During pleural inflammation, mesothelial cells 8. Christensen GD, Simpson WA, Beachey EH. Adhesion of bacteria to
play an active role secreting cytokines and animal tissues: complex mechanisms. In: Savage DCFM (ed).
Bacterial adhesion. New York: Plenum Press, 1985: 279–306.
chemokines that attract and activate multiple 9. Ofek I. General concepts and principles of bacterial adherence in
inflammatory cell types. animals and man. In: Beachey EH. (ed.) Bacterial adherence.
● During pleural inflammation there is a flow of London: Chapman and Hall, 1980: 1–29.
inflammatory cells and protein into the pleural 10. Arp L. Bacterial infection of mucosal surfaces: an overview of
space. cellular and molecular mechanisms. In: Roth J (ed). Virulence
mechanisms of bacterial pathogens. Washington DC: American
● Pleural inflammation may resolve with restora- Society of Microbiology, 1988: 3–27.
tion of normal pleural structure and return to 11. Schlievert PM. Role of superantigens in human disease. J Infect Dis
normal function; alternatively, resolution of 1993; 167: 997–1002.
pleural inflammation may lead to pleural fibrosis 12. Birnbaumer LM, Yatani, A. Recent advances in the understanding
and gross distortion of pleural architecture. of multiple roles of G proteins in coupling of receptors to ionic
channels and other effectors. In: Moss J, Vaughan M (eds). ADP-
● Microbial virulence factors are aimed at over- ribosylating toxins and G proteins: insights into signal
coming specific host-defense mechanisms. transduction. Washington, DC: American Society for Microbiology,
● The outcome of pleural infection depends upon 1990: 225–66.
the virulence of the invading microbe and the 13. Middeldorp JM, Witholt B. K88-mediated binding of Escherichia
pleural defense mechanisms. coli outer membrane fragments to porcine intestinal epithelial cell
brush borders. Infect Immun 1981; 31: 42–51.
● Passive pleural defenses against pleural infection 14. Mollby R. Bacterial phospholipases. In: Jeljaszewicz JWT (ed).
include the microbicidal proteins within the Bacterial toxins and cell membranes. New York: Academic Press,
pleural fluid, the biophysical properties of the 1978: 367–424.
extracellular domain at the mesothelium surface, 15. Lee PK, Vercellotti GM, Deringer JR, Schlievert PM. Effects of
the structural integrity of the mesothelial cell staphylococcal toxic shock syndrome toxin 1 on aortic endothelial
cells. J Infect Dis 1991; 164: 711–19.
monolayer itself and the pleural basement mem- 16. Marrack P, Kappler J. The staphylococcal enterotoxins and their
brane. relatives. Science 1990; 248: 705–11.
● Microbial invasion of the pleural space provokes 17. Smith H, Huggins M. Further observations on the association of
both innate and adaptive immune responses; the the colicine V plasmic of Escherichia coli with pathogenicity and
pleural mesothelial cell participates in and coor- with survival in the alimentary tract. J Gen Microbiol 1976; 92:
335–50.
dinates both the innate and the adaptive immune 18. Mohammed KA, Nasreen N, Ward MJ, Antony VB. Induction of
responses. acute pleural inflammation by Staphylococcus aureus. I. CD4+ T
● The consequences of pleural infection may cells play a critical role in experimental empyema. J Infect Dis
include pleural effusion, empyema and pleural 2000; 181: 1693–9.
fibrosis. 19. Brubaker RR. Mechanisms of bacterial virulence. Annu Rev
Microbiol 1985; 39: 21–50.
20. Meyer TF. Pathogenic Neisseriae – model of bacterial virulence
and genetic flexibility. Immun Infekt 1989; 17: 113–23.
21. McCutchan JA, Katzenstein D, Norquist D, et al. Role of blocking
REFERENCES antibody in disseminated gonococcal infection. J Immunol 1978;
121: 1884–8.
22. Hirakata Y, Kaku M, Mizukane R, et al. Potential effects of
● = Key primary paper erythromycin on host defense systems and virulence of
◆ = Major review article Pseudomonas aeruginosa. Antimicrob Agents Chemother 1992; 36:
1922–7.
1. Wang NS. Mesothelial cells in situ. In: Chrétien J, Bignon J, Hirsch 23. Nasreen N, Mohammed KA, Dowling PA, et al. Talc induces
A (eds). The pleura in health and disease, Vol. 30. New York: apoptosis in human malignant mesothelioma cells in vitro. Am J
Marcel Dekker, 1985: 23–42. Respir Crit Care Med 2000; 161: 595–600.
2. Rennard S, Jaurand M-C, Bignon J, Ferrans V, Crystal R. Connective 24. Eidels L, Proia RL, Hart DA. Membrane receptors for bacterial
tissue matrix of the pleura. In: Chrétien J, Bignon J, Hirsch A (eds). toxins. Microbiol Rev 1983; 47: 596–620.
The pleura in health and disease, Vol. 30. New York: Marcel 25. Falkow S, Small P, Isberg R, Hayes SF, Corwin D. A molecular
Dekker, 1985: 69–85. strategy for the study of bacterial invasion. Rev Infect Dis 1987; 9
3. Wang NS. The preformed stomas connecting the pleural cavity and (Suppl 5): S450–5.
68 Pleural inflammation and infection
26. Finlay BB, Falkow S. Common themes in microbial pathogenicity. 47. Frank MM, Joiner K, Hammer C. The function of antibody and
Microbiol Rev 1989; 53: 210–30. complement in the lysis of bacteria. Rev Infect Dis 1987; 9 (Suppl
27. Jones GW, Isaacson RE. Proteinaceous bacterial adhesins and their 5): S537–45.
receptors. Crit Rev Microbiol 1983; 10: 229–60. 48. Fearon D, Austen K. The alternative pathway of complement: a
28. Beachey EH, Ofek I. Epithelial cell binding of group A streptococci system for host resistance to microbial infection. N Engl J Med
by lipoteichoic acid on fimbriae denuded of M protein. J Exp Med 1980; 303: 259–63.
1976; 143: 759–71. 49. Joiner KA. Complement evasion by bacteria and parasites. Annu
29. Ohtsuka A, Yamana S, Murakami T. Localization of membrane- Rev Microbiol 1988; 42: 201–30.
associated sialomucin on the free surface of mesothelial cells of 50. Medzhitov R, Janeway CA Jr. Innate immunity: impact on the
the pleura, pericardium, and peritoneum. Histochem Cell Biol adaptive immune response. Curr Opin Immunol 1997; 9: 4–9.
1997; 107: 441–7. 51. Gorbach S, Bartlett J, Blacklow N. Infectious diseases. In:
30. Heinzel F. Antibodies. In: Mandell G, Douglas R, Bennett J (eds). Blacklow N (ed). Host factors. Philadelphia: Saunders, 1992:
Principles and practice of infectious diseases, Vol. 1. New York, 37.
Edinburgh: Churchill Livingston, 1995: 36–57. 52. Zhang G, Ghosh S. Toll-like receptor-mediated NF-kappaB
31. Payne NR, Horwitz MA. Phagocytosis of Legionella pneumophila is activation: a phylogenetically conserved paradigm in innate
mediated by human monocyte complement receptors. J Exp Med immunity. J Clin Invest 2001; 107: 13–19.
1987; 166: 1377–89. 53. Toews GB. Pulmonary clearance of infectious agents. In: Fishman
32. Dowling JN, Saha AK, Glew RH. Virulence factors of the family AP, Elias JA (eds). Pulmonary diseases and disorders. New York:
Legionellaceae. Microbiol Rev 1992; 56: 32–60. McGraw-Hill 1997: 1891–904.
33. Hacker J, Ott M, Ludwig B, Rdest U. Intracellular survival and 54. Arriero MM, Rodriguez-Feo JA, Celdran A, et al. Expression of
expression of virulence determinants of Legionella pneumophila. endothelial nitric oxide synthase in human peritoneal tissue:
Infection 1991; 19 (Suppl 4): 198–201. regulation by Escherichia coli lipopolysaccharide. J Am Soc
34. Spurzem J, Rennard S, Romberger D. Interactions between Nephrol 2000; 11: 1848–56.
pulmonary epithelial cells and extracellular matrix. In: Ward P, 55. Chen JY, Chiu JH, Chen HL, et al. Human peritoneal mesothelial
Fantone J (eds). Adhesion molecules and the lung, Vol. 89. New cells produce nitric oxide: induction by cytokines. Perit Dial Int
York: Marcel Dekker, 1996: 127–46. 2000; 20: 772–7.
35. Liaw YS, Yu CJ, Shun CT, et al. Expression of integrins in human 56. Owens MW, Grisham MB. Nitric oxide synthesis by rat pleural
cultured mesothelial cells: the roles in cell-to-extracellular matric mesothelial cells: induction by cytokines and lipopolysaccharide.
adhesion and inhibition by RGD-containing peptide. Respir Med Am J Physiol 1993; 265: L110–16.
2001; 95: 221–6. 57. Antony VB. Pathogenesis of malignant pleural effusions and talc
36. Ma C, Tarnuzzer RW, Chegini N. Expression of matrix pleurodesis. Pneumologie 1999; 53: 493–8.
metalloproteinases and tissue inhibitor of matrix 58. Owens MW, Milligan SA, Grisham MB. Nitric oxide synthesis by rat
mettaloproteinases in mesothelial cells and their regulation by pleural mesothelial cells: induction by growth factors and
transforming growth factor-beta 1. Wound Repair Regen 1999; lipopolysaccharide. Exp Lung Res 1995; 21: 731–42.
7:477–85. 59. Lehrer RI, Barton A, Daher KA, et al. Interaction of human
37. Martin J, Yung S, Robson RL, Steadman R, Davies M. Production defensins with Escherichia coli. Mechanism of bactericidal activity.
and regulation of matrix metalloprotienases and their inhibitors by J Clin Invest 1989; 84: 553–61.
human peritoneal mesothelial cells. Perit Dial Int 2000; 20: 5. 60. Burton DR, Woof JM. Human antibody effector function. Adv
38. Price-Schiavi SA, Zhu X, Aquinin R, Carraway KL. Sialomucin Immunol 1992; 51: 1–84.
complex (rat Muc4) is regulated by transforming growth factor 61. Spitznagel J. Non-oxidative antimicrobial reactions of leukocytes.
beta in mammary gland by a novel post-translational mechanism. Contemp Top Immunobiol 1984; 14: 283.
J Biol Chem 2000; 275: 17800–7. 62. Antony VB, Hott JW, Kunkel SL, et al. Pleural mesothelial cell
39. Zannettino AC, Buhring HJ, Niutta S, et al. The sialomucin CD164 expression of C-C (monocyte chemotactic peptide) and C-X-C
(MGC-24v) is an adhesive glycoprotein expressed by human (interleukin 8) chemokines. Am J Respir Cell Mol Biol 1995; 12:
hematopoietic progenitors and bone marrow stromal cells that 581-8.
serves as a potent negative regulator of hematopoiesis. Blood 63. Bazan JF, Bacon KB, Hardiman G, et al. A new class of membrane
1998; 92: 2613–28. chemokine with a CX3C motif. Nature 1997; 385: 640–4.
40. Ohtsuka A, Yamana S, Murakami T. Localization of membrane- 64. Hausmann MJ, Rogachev B, Weiler M, Chaimovitz C, Douvdevani
associated sialomucin on the free surface of mesothelial cells of A. Accessory role of human peritoneal mesothelial cells in
the pleura, pericardium, and peritoneum. Histochem Cell Biol antigen presentation and T-cell growth. Kidney Int 2000; 57: 476-
1997; 107: 441–7. 86.
41. Sassetti C, van Zante A, Rosen SD. Identification of endoglycan, a ●65. Nasreen N, Mohammed KA, Sanders KL, et al. Differential
member of the CD34/podocalyxin family of sialomucins. J Biol expression of C-phos, C-June, and apoptosis in pleural mesothelial
Chem 2000; 275: 9001–10. cells exposed to Staph. aureus. Am J Respir Crit Care Med 2001;
42. Doyonnas R, Kershaw DB, Duhme C, et al. Anuria, omphalocele, 163: A772.
and perinatal lethality in mice lacking the CD34-related protein 66. Payne CM, Glasser L, Tischler ME, et al. Programmed cell death of
podocalyxin. J Exp Med 2001; 194: 13–27. the normal human neutrophil: an in vitro model of senescence.
43. Hilkens J, Ligtenberg MJ, Vos HL, Litvinov SV. Cell membrane- Microsc Res Technol 1994; 28: 327–44.
associated mucins and their adhesion-modulating property. Trends 67. Weinmann P, Gaehtgens P, Walzog B. Bcl-Xl- and Bax-alpha-
Biochem Sci 1992; 17: 359–63. mediated regulation of apoptosis of human neutrophils via
44. Hjelle JT, Golinska BT, Waters DC, et al. Lectin staining of caspase-3. Blood 1999; 93: 3106–15.
peritoneal mesothelial cells in vitro. Perit Dial Int 1991; 11: 68. Metchnikoff M. Lectures on the comparative pathology of
307–16. inflammation. London: Kegan Paul, Trench, Trubner & Co., 1893:
45. Leak LV. Distribution of cell surface charges on mesothelium and 1845–916.
lymphatic endothelium. Microvasc Res 1986; 31: 18–30. 69. Antony VB. Pathogenesis of malignant pleural effusions and talc
46. Andrews PM, Porter KR. The ultrastructural morphology and pleurodesis. Pneumologie 1999; 53: 493–8.
possible functional significance of mesothelial microvilli. Anat Rec 70. Barclay A, Brown M, Law S, et al. The leukocyte antigen facts
1973; 177: 409–26. book. San Diego: Academic Press, 1997.
References 69
71. Jonjic N, Peri G, Bernasconi S, et al. Expression of adhesion 92. Mazure NM, Chen EY, Laderoute KR, Giaccia AJ. Induction of
molecules and chemotactic cytokines in cultured human vascular endothelial growth factor by hypoxia is modulated by a
mesothelial cells. J Exp Med 1992; 176: 1165–74. phosphatidylinositol 3-kinase/Akt signaling pathway in Ha-ras-
72. Krieger M, Stern DM. Series introduction: multiligand receptors transformed cells through a hypoxia inducible factor-1
and human disease. J Clin Invest 2001; 108: 645–7. transcriptional element. Blood 1997; 90: 3322–31.
73. Locksley R, Wilson C. Cell-mediated immunity and its role in host 93. Becker PM, Alcasabas A, Yu AY, Semenza GL, Bunton TE. Oxygen-
defense. In: Mandell G, Douglas R, Bennett J (eds). Principles and independent upregulation of vascular endothelial growth factor
practice of infectious diseases, Vol. 1. New York: Churchill and vascular barrier dysfunction during ventilated pulmonary
Livingstone, 1995: 102–49. ischemia in isolated ferret lungs. Am J Respir Cell Mol Biol 2000;
74. Ellner JJ, Barnes PF, Wallis RS, Modlin RL. The immunology of 22: 272–9.
tuberculous pleurisy. Semin Respir Infect 1988; 3: 335–42. 94. Thickett DR, Armstrong L, Millar AB. Vascular endothelial growth
◆75. Malech H. Phagocytic cells: egress from marrow and diapedesis. factor (VEGF) in inflammatory and malignant pleural effusions.
In: Gallin J, Goldstein I, Snyderman R (eds). Inflammation: basic Thorax 1999; 54: 707–10.
principles and clinical correlates. New York: Raven Press, 1988; 95. D’Arcangelo D, Facchiano F, Barlucchi LM, et al. Acidosis inhibits
297–308. endothelial cell apoptosis and function and induces basic
76. Weiss J, Elsbach P, Olsson I, Odeberg H. Purification and fibroblast growth factor and vascular endothelial growth factor
characterization of a potent bactericidal and membrane active expression. Circ Res 2000; 86: 312–18.
protein from the granules of human polymorphonuclear 96. Mohammed KA, Nasreen N, Ward MJ, Antony VB. Induction of
leukocytes. J Biol Chem 1978; 253: 2664–72. acute pleural inflammation by Staphylococcus aureus. I. CD4+ T
77. Ganz T, Selsted ME, Szklarek D, et al. Defensins. Natural peptide cells play a critical role in experimental empyema. J Infect Dis
antibiotics of human neutrophils. J Clin Invest 1985; 76: 1427–35. 2000; 181: 1693–9.
78. Nasreen N, Hartman D, Mohammed K, Antony V. Talc induces 97. Zhang XY, Pettengell R, Nasiri N, et al. Characteristics and growth
pleural mesothelial cell expression of proinflammatory cytokines patterns of human peritoneal mesothelial cells: comparison
and intracellular adhesion molecule-1 (ICAM-1). Am J Respir Crit between advanced epithelial ovarian cancer and non-ovarian
Care Med 1998; 158: 971–8. cancer sources. J Soc Gynecol Invest 1999; 6: 333–40.
79. Cox G, Gauldie J, Jordana M. Bronchial epithelial cell-derived 98. Nasreen N, Mohammed KA, Hardwick J, et al. Polar production of
cytokines (G-CSF and GM-CSF) promote the survival of peripheral interleukin-8 by mesothelial cells promotes the transmesothelial
blood neutrophils in vitro. Am J Respir Cell Mol Biol 1992; 7: migration of neutrophils: role of intercellular adhesion molecule-
507–13. 1. J Infect Dis 2001; 183: 1638–45.
80. Coxon A, Tang T, Mayadas TN. Cytokine-activated endothelial cells 99. Corada M, Liao F, Lindgren M, et al. Monoclonal antibodies
delay neutrophil apoptosis in vitro and in vivo. A role for directed to different regions of vascular endothelial cadherin
granulocyte/macrophage colony-stimulating factor. J Exp Med extracellular domain affect adhesion and clustering of the protein
1999; 190: 923–34. and modulate endothelial permeability. Blood 2001; 97:
81. Antony VB, Sahn SA, Antony AC, Repine JE. Bacillus 1679–84.
Calmette–Guerin-stimulated neutrophils release chemotaxis for 100. Blankesteijn WM, van Gijn ME, Essers-Janssen YP, Daemen MJ,
monocytes in rabbit pleural spaces and in vitro. J Clin Invest 1985; Smits JF. Beta-catenin, an inducer of uncontrolled cell
76: 1514–21. proliferation and migration in malignancies, is localized in the
82. Mantovani A, Garlanda C. Novel pathways for negative regulation cytoplasm of vascular endothelium during neovascularization after
of inflammatory cytokines centered on receptor expression. Dev myocardial infarction. Am J Pathol 2000; 157: 877–83.
Biol Stand 1999; 97: 97–104. 101. Dejana E. Endothelial adherens junctions: implications in the
83. Imai T, Heishima K, Haskell C, et al. Identification and molecular control of vascular permeability and angiogenesis. J Clin Invest
characterization of fractaline receptor CX3CR1, which mediates 1996; 98: 1949–53.
both leukocyte migration and adhesion. Cell 1997; 91: 1-20 102. Bourguignon LY, Lokeshwar VB, Chen X, Kerrick WG. Hyaluronic
84. Klein A, Talvani A, Silva PM, et al. Stem cell factor-induced acid-induced lymphocyte signal transduction and HA receptor
leukotriene B4 production cooperates with eotaxin to mediate the (GP85/CD44)-cytoskeleton interaction. J Immunol 1993; 151:
recruitment of eosinophils during allergic pleurisy in mice. J 6634–44.
Immunol 2001; 167: 524–31. 103. Haslinger B, Mandl-Weber S, Sellmayer A, Sitter T. Hyaluronan
85. Light RW. Parapneumonic effusions and infections of the pleural fragments induce the synthesis of MCP-1 and IL-8 in cultured
space. In: Light RW (ed). Pleural diseases. Philadelphia: Lea and human peritoneal mesothelial cells Cell Tissue Res 2001; 305:79-
Febiger, 1990. 86.
86. Wang GL, Jiang BH, Rue EA, Semenza GL. Hypoxia-inducible factor 104. Ponta H. The CD44 protein family. Int J Biochem Cell Biol 1998;
1 is a basic-loop-helix-loop-PAS hetero dimmer regulated by 30: 299–305.
cellular O2 tension. Proc Natl Acad Sci USA 1995; 92: 5510–4. 105. Hott J, Godbey S, Antony V. The role of the mesothelial cell in
87. Marti HJ, Bernaudin M, Bellail A, et al. Hypoxia-induced vascular pleural metastasis: inhibition of breast carcinoma cell adherence
endothelial growth factor expression precedes neovascularization to pleural cells by hyaluronidase. FASEB J 1992; 6: 1919.
after cerebral ischemia. Am J Pathol 2000; 156: 965–76. 106. Hott J, Godbey S, Antony VB. The role of the mesothelial cell in
88. Richard DE, Berra E, Pouyssegur J. Nonhypoxic pathway mediates pleural metastasis: breast carcinoma cell adherence to pleural
the induction of hypoxia-inducible factor 1alpha in vascular mesothelial cells by a mechanism involving hyaluronate and CD44.
smooth muscle cells. J Biol Chem 2000; 275: 26765–71. Am Rev Respir Dis 1993; 147: A794.
89. Sandau KB, Fandrey J, Brune B. Accumulation of HIF-1alpha under 107. Johnson JP. Cell adhesion molecules of the immunoglobulin
the influence of nitric oxide. Blood 2001; 97: 1009–15. supergene family and their role in malignant transformation and
90. Semenza GL. HIF-1 and human disease: one highly involved factor. progression to metastatic disease. Cancer Metastasis Rev 1991;
Gene Dev 2000; 14(16): 1983–91. 10: 11–22.
●91. Yano S, Shinohara H, Herbst RS, et al. Production of experimental 108. Underhill C. CD44: the hyaluronan receptor. J Cell Sci 1992; 103
malignant pleural effusions is dependent on invasion of the pleura (Pt 2): 293–8.
and expression of vascular endothelial growth factor/vascular 109. Lee YCG, Lane, KB, Zoia O, et al. Transforming growth factor-b
permeability factor by human lung cancer cells. Am J Pathol 2000; induces collagen synthesis without inducing IL-8 production in
157: 1893–903. mesothelial cells. Eur Respir J 2003; 22:197–202.
70 Pleural inflammation and infection
110. Antony VB, Nasreen N, Mohammed KA, et al. Talc pleurodesis: 112. Galffy G, Mohammed K, Ward M, Dowling PA, Antony V.
basic fibroblast growth factor mediates pleural fibrosis. Chest Interleukin-8: an autocrine growth factor for malignant
2004; 126:1522-8 mesothelioma. Am Cancer Res 1999; 59: 367–71.
111. Nasreen N, Mohammed, KA, Brown S, et al. Talc mediates
angiostasis in malignant pleural effusions via endostatin
induction. Eur Respir J 2007; 29: 761–9.
8
Immunology
Introduction 71 Summary 89
Cytokine growth factors in pleural fibrosis 71 Key points 89
Cytokines in pleural fluid formation 79 References 89
Immunomodulatory cytokines 81
A host of cytokine growth factors have been demonstrated Overproduction of TGF-b is the principal abnormality in
in the pleura, and are likely to be of biological significance. most fibrotic diseases, and its direct administration
potently induces fibrosis.5 After tissue injury, a repair
process is initiated with increased expression of TGF-b by
Transforming growth factor beta parenchymal cells, infiltrating macrophages and lympho-
cytes.6 Latent TGF-b can be activated within 1 hour, fol-
Transforming growth factor beta (TGF-b) is currently lowed by a second wave of activation several days later.7
considered the most potent pro-fibrotic cytokine and its TGF-b induces fibrosis by increasing the expression of
72 Immunology
most extracellular matrix proteins, as well as by inhibiting sions19 have been achieved in animal models by neutraliz-
their degradation8 (Figure 8.1). ing TGF-b functions, or by increasing the levels of
Excessive formation of adhesions and resultant pleural inhibitory Smad proteins.20
fibrosis are commonly seen in pleural diseases (e.g. The use of pan-specific neutralizing anti-TGF-b anti-
empyema, fibrothorax following tuberculosis). Increased bodies have been shown to decrease the volume of pus and
levels of TGF-b have been observed in the pleural fluid the adhesions formed during Pasteurella multocida-
during these pathologies. During the progression of induced empyema in rabbits.21 Importantly, the micro-
empyema, pleural fluid TGF-b1 increases. This correlates scopic pleural thickness and fibroblast scores were also
with the pleural fibrosis that takes place, measured by decreased. Similarly, the volume of pleural effusion and
microscopic thickness and fibroblast score.9 TGF-b is a recruitment of inflammatory cells to the pleural cavity
potent chemoattractant for fibroblasts, which are impor- were decreased by neutralizing pan anti-TGF-b antibodies
tant in the synthesis and deposition of collagen, leading to during Mycobacterium tuberculosis antigen-specific pleural
the progression of pleural fibrosis. However, the mesothe- effusion formation.22
lium itself also participates in the matrix turnover and
deposition. Upon stimulation by TGF-b, mesothelial cells PLEURODESIS
can synthesize matrix proteins and collagen,10 matrix met-
alloproteinase (MMP)-1, MMP-9 and tissue inhibitor of Although the potent pro-fibrotic properties of TGF-b are
matrix metalloproteinases (TIMP)-2.11,12 TGF-b can con- implicated in the pathogenesis of excessive pleural fibrosis,
tribute to the profibrotic environment by also suppressing these properties can be utilized for therapeutic pleurodesis
fibrinolysis via reduction of tissue plasminogen activators (i.e. iatrogenic induction of pleural fibrosis to obliterate
and via increasing the mesothelial cell production of plas- the pleural space). Pleurodesis is generally performed by
minogen activator inhibitor (PAI)-1 and -2.13,14 Collagen, intrapleural administration of a chemical agent or by
PAI-1 and fibronectin production by stimulated mesothe- mechanical abrasion during surgery. These processes aim
lial cells can be downregulated by TGF-b1 short hairpin to provoke acute pleural inflammation,1 which, if intense
RNA.15 enough, will result in pleural fibrosis (Figure 8.2).
Strategies to antagonize TGF-b activity therefore hold Direct intrapleural administration of TGF-b2 and -b3
promise to prevent pleural fibrosis. Significant reductions induce excellent pleurodesis in different animal
in bleomycin-induced pulmonary fibrosis,16 glomeru- models23–27 (Figure 8.3). Histologically, intrapleural deliv-
losclerosis,17 wound scarring18 and intra-abdominal adhe-
Mechanism of Pleurodesis
Stimulating factors
(e.g. tissue injury; irradiation; TGF-; Injection of Pleurodesing Agent
angiotensin II; thromboxane; glucose and
glycosylation products, etc.)
Inhibited by Steroid
Activators
To convert latent TGF-b
into its active form TGF- Pleural Fibrosis / Pleurodesis
(e.g. plasmin, ROS,
thrombospondin)
Figure 8.2 The intrapleural administration of conventional
pleurodesing agents induces an acute pleural injury and
inflammation. This process is inhibited by corticosteroids. Pain
Matrix Matrix and fever associated with pleurodesis are presumed to be a result
synthesis degradation of the acute pleural inflammation. The inflammation may heal
with restoration of normal pleura (failed pleurodesis). However, if
the inflammation is sufficiently intense, it will progress to chronic
Pleural fibrosis inflammation and pleural fibrosis (successful pleurodesis).
Transforming growth factor (TGF)-b is most likely the mediator of
Figure 8.1 A simplified diagram of the role of transforming the fibrotic process.21 Reprinted from Lee YCG, Lane KB. Curr Opin
growth factor (TGF)-b in pleural fibrosis. ROS, reactive oxygen Pulm Med 2001; 7: 173–9, with permission from the publisher,
species. Lippincott, Williams & Wilkins.
Cytokine growth factors in pleural fibrosis 73
(a) (b)
Figure 8.3 The pleura (indicated by the arrows) following treatment with transforming growth factor (TGF)-b2 injection (a) is significantly
thickened with a large amount of collagen and fibrous tissue deposition compared with pleura from the control side (b). Reprinted from Lee
YCG, Lane KB, Parker RE, et al. Thorax 2000; 55: 1058–62 with permission from the publisher, the BMJ Publishing Group.
ery of TGF-b2 stimulates marked collagen deposition,24 cal advantage of producing pleurodesis with less pain or
resulting in dramatic pleural fibrosis and adhesions,26 fever than conventional agents (Figure 8.2).
more rapidly than talc (Figure 8.3). Injection of Although TGF-b appears promising as a pleurodesing
fibronectin, a downstream matrix protein induced by agent, its long-term effects and effectiveness in abnormal
TGF-b, fails to reproduce the fibrotic effects of TGF-b in (e.g. malignant) pleural surfaces requires evaluation in
the pleura.28 TGF-b upregulation of mesothelial cell colla- humans.1 Importantly, results to date challenge the tradi-
gen synthesis is at least as potent as talc and doxycycline.24 tional concept that creation of pleurodesis must be accom-
Interestingly, intrapleural introduction of TGF-b2 and -b3 panied with pleural injury and inflammation.1
in rabbit models induces an increase in pleural fluid
volume before pleurodesis ensues.27,29 IMMUNE MODULATION
Intrapleural administration of TGF-b in low doses
appears safe. No acute cardiopulmonary abnormalities or Transforming growth factor-b plays a critical role in the
histological extrapulmonary abnormalities at 14 days are immune regulation. Most commonly, it behaves as a
found in sheep after TGF-b pleurodesis.30 The systemic potent anti-inflammatory cytokine36 and suppresses the
levels of TGF-b1 and -b2 in sheep that received TGF-b2 production of tumor necrosis factor alpha (TNF-a), inter-
pleurodesis are no different from those given talc or leukin (IL)-137–39 and IL-8,40 thus deactivating neutrophil
bleomycin.30 In humans, regular intravenous therapy with and macrophage functions. TGF-b1 knockout mice
TGF-b2 was shown to be safe in a phase I clinical trial in succumb to overwhelming systemic inflammation medi-
multiple sclerosis patients.31 ated by lymphocytes41 with significantly elevated levels of
The concept of administration of TGF-b to initiate TNFa and IL-1. TGF-b inhibits the production of and the
fibrosis has the advantage of bypassing the pleural inflam- response to cytokines associated with both Th1 and Th2
matory process (Figure 8.2). Additionally, TGF-b has lymphocytes.42 In addition, TGF-b inhibits natural killer
potent anti-inflammatory properties. The pleural fluid cell functions and the generation of lymphokine-activated
induced after the intrapleural injection of TGF-b has sig- killer cells.36,43 TGF-b also blocks leukocyte adhesion and
nificantly lower inflammatory indices23,24 and IL-8 levels32 recruitment to inflammation sites.44. While IL-1 can stim-
when compared with those after talc or doxycycline ulate TGF-b in cultured mesothelial cell,.45,46 TGF-b in
administration. After intrapleural talc introduction, TGF- turn suppresses the mesothelial production of IL-832 and
b1 steadily increases over time.33 In cultured mesothelial nitric oxide,47 in keeping with its anti-inflammatory role.
cells, talc and doxycycline, but not TGF-b, stimulates sig- The therapeutic potential of the anti-inflammatory prop-
nificant IL-8 release.32 Systemic corticosteroids reduce the erties of TGF-b in inflammatory pleural conditions
pleural inflammation and hence the efficacy of talc and warrant investigation.
doxycycline pleurodesis in animal models.34,35 However,
TGF-b-induced pleurodesis remains effective in the pres- CARCINOGENESIS
ence of corticosteroids,25 supporting the notion that TGF-
b can induce pleurodesis without necessitating significant Transforming growth factor-b bears a biphasic ‘love-hate’
pleural inflammation. TGF-b therefore carries a theoreti- relationship with cancer. It is a potent inhibitor of cell pro-
74 Immunology
liferation. TGF-b acts as a tumor suppressor by also pro- be associated with variation in circulating TGF-b levels.54
moting cellular differentiation or apoptosis48 in normal Homozygosity for arginine at codon 25 of the leader
human cells, including mesothelial cells,49 and during the sequence of TGF-b1 correlates with higher TGF-b produc-
early stage of carcinogenesis. However, advanced tumors tion in vitro, and is associated with fibrotic lung pathology
are often resistant to such TGF-b-mediated growth arrest, before lung transplantation and with the development of
probably as a result of reduced TGF-b signaling secondary fibrosis in the graft.55 The effect of these polymorphisms
to mutations in genes encoding TGF-b signaling media- on pleural fibrosis awaits investigation.
tors.48 Once resistance develops, the more aggressive forms
of cancer cells can make use of TGF-b to enhance their
growth and metastasis in several ways. TGF-b signaling in Hepatocyte growth factor
the tumor microenvironment significantly affects carci-
noma initiation, progression and metastasis via epithelial Hepatocyte growth factor (HGF), also known as scatter
cell autonomous and interdependent stromal–epithelial factor and tumor cytotoxic factor, is a multifunctional
interactions (see review50). TGF-b can promote angiogen- polypeptide involved in embryonic development, tissue
esis essential for tumor growth, and increases extracellular repair and cancer growth. HGF forms part of a family of
matrix synthesis by cancer cells, allowing their binding to plasminogen-related growth factors along with HGF-
cell-adhesion molecules, thereby facilitating distant metas- like/macrophage stimulating protein, plasminogen and
tasis. The immunosuppressive functions of TGF-b may apolipoprotein – all of which are thought to have evolved
enable cancer cells to escape immune surveillance.48 from the same ancestral gene.56
The effect of TGF-b on pleural malignancies is largely
unknown. In mesothelioma cells, TGF-b increases the syn- STRUCTURE
thesis of matrix protein10 and urokinase-type plasminogen
activator. Antagonizing TGF-b activity can inhibit The gene encoding HGF maps on chromosome 7q21.1 in
mesothelioma cell proliferation and tumor growth in humans and is composed of 18 exons interrupted by 17
vivo.51 Small molecule inhibitor of ALK5 kinase (TBR-I) introns spanning approximately 70 kb. The protein is
can prevent tumor recurrence in a malignant mesothe- derived from a biologically inactive single chain propep-
lioma model which overexpresses TGF-b.52 tide (pro-HGF) of 728 amino acids, with a similar amino
Given the complex relationship between TGF-b and acid sequence as plasminogen.56,57 The propeptide is
cancer, the potential application of TGF-b or its antagonists cleaved extracellularly by serine proteases, such as plas-
as cancer therapy require careful clinical assessments.1 minogen activator, to give an active heterodimeric glyco-
protein consisting of an a and b chain held together by a
TGF-b ACTIVATION disulfide bond.58
Hepatocyte growth factor is produced not only by
Transforming growth factor-b is secreted in trimeric or hepatocytes but also by a wide variety of cell types includ-
dimeric latent form, according to cell type and environ- ing mesothelial cells, platelets, monocytes, fibroblasts and
ment. The dimer consists of the TGF-b itself, and the certain tumor cells including malignant mesothelioma.59
latency associated peptide (LAP), which confers inactivity Inflammatory cytokines and mediators produced by
on the molecule. In the trimeric form, this dimer, also tumor cells including IL-1b, IL-6, IL-8, TNF-a,
known as the small latent complex, is bound to extracellu- prostaglandin E2, basic fibroblast growth factor (b-FGF),
lar matrix by the latent TGF-b binding protein (LTBP). epidermal growth factor (EGF) and platelet-derived
The most crucial step of TGF-b bioavailability is post- growth factor (PDGF) can stimulate HGF production,60
translational activation, taking the form of the separation whereas TGF-b has been shown to downregulate HGF
of the TGF-b from the LAP, allowing it to become active. expression.61
Proteolytic cleavage and conformational change can bring The HGF receptor is the met protein, a product of the
this about, and there is a diverse range of activators includ- c-met proto-oncogene (p190 met) found principally on
ing matrix metalloproteinases (MMPs), integrins anb6 epithelial and endothelial cells, mesothelial cells and a
and anb8, thrombospondin (TSP)-1, and physiological variety of tumors, including mesothelioma.62 c-met con-
extremes such as low pH or mechanical stress. The post- sists of a transmembrane 145-kDa b-chain and an extracel-
translational regulation of TGF-b activity in the pleura has lular 50-kDa a-chain to form a dimeric 190 kDa protein
not been explored, but may represent a target point for with structural features of a tyrosine kinase receptor. HGF
therapeutic intervention. also binds to cell surface heparan sulfate proteoglycans
that serve as low affinity HGF receptors and modulate the
GENETICS interaction between HGF and c-met receptor. c-met
signals largely via the Ras signaling pathway. Data has
The TGF-b1 gene is located on human chromosome shown that activation of the Ras pathway leads to cell pro-
19q13. There are at least five polymorphisms in the TGF- liferation whereas the PI3 kinase pathway is needed to
b1 gene and several others in its promoter region53 and can induce mitogenesis. Both pathways are essential for
Cytokine growth factors in pleural fibrosis 75
ble resistance mechanisms to EGFR tyrosine kinase effusions (including mesotheliomas) compared with non-
inhibitors.129 malignant ones. Significantly, an inverse correlation is
The effect of farnesyltransferase inhibitors on mesothe- observed between b-FGF levels found in mesothelioma
lioma cell growth has also been examined in vitro. effusions and patient survival. High serum b-FGF levels are
Farnesylation of Ras (downstream molecule in EGFR sig- also associated with reduced survival.133 However, another
naling) is the obligatory first step in the switch from an study reported a negative correlation between b-FGF and
inactive to an active Ras-GTP bound form. Preventing Ras tumor size observed during thoracoscopy.135
farnesylation was proposed to inhibit Ras functions and Interleukin-1 induced mesothelial cell production of b-
hence block downstream effects resulting from EGFR acti- FGF in one study134 but not in another.106 In both studies,
vation. Growth of mesothelioma cell lines were not effected TNFa did not stimulate b-FGF release. IL-2, however, can
by a number of farnesyltransferase inhibitors, highlighting produce a marked suppression of b-FGF expression.106 In
the concept that the same signaling pathway can be regu- mesothelioma cells, b-FGF increases glycosaminoglycans
lated in different ways and these regulations can differ production but has no significant effect on heparan sulfate
between different cells of different origin.130 EGFR status in synthesis.138
malignant cells recovered from pleural effusions may have
prognostic implications. In one study, breast cancer
patients with negative EGFR status in the malignant cells in Platelet-derived growth factor
pleural fluids had the longest survival time, whereas
patients with positive EGFR but negative estrogen and The platelet-derived growth factor family of cytokines
progesterone receptor status had the worst prognosis.131 consists of dimers of two related polypeptide chains desig-
nated A, B, C and D that can exist as heterodimers (PDGF-
AB) or homodimers (PDGF-AA, PDGF-BB, PDGF-CC
Basic fibroblast growth factor and PDGF-DD).139–141 These factors exert their cellular
effects through structurally similar PDGF protein tyrosine
Basic fibroblast growth factor is present in pleural effu- kinase receptors (PDGFR-a and -b). The PDGFR-a binds
sions of various etiologies.132,133 Mesothelial cells synthe- to A-, B- and C- chains with high affinity, whereas
size and release considerable amounts of b-FGF.106,134 The PDGFR-b only binds the B- and D- chains. PDGF-C and
majority (80 percent) of b-FGF is localized intracellularly, PDGF-D require proteolytic activation before binding to
and the remainder is associated with extracellular matrix and activating PDGFR. A heterodimeric PDGF-a/b
components on the mesothelial cell surface.134 b-FGF is complex has also been identified which can be activated by
known to stimulate mesothelial cell proliferation in vitro PDGF-AB, PDGF-BB and PDGF-CC.139–142
and in vivo.105 Early evidence to date suggests a role of b- Platelet-derived growth factor ligand binding induces
FGF in pleural fibrosis and malignant pleural diseases. dimerization of both receptors and subsequently
autophosphorylation of the tyrosine kinase domain. This
PLEURAL FIBROSIS in turn phosphorylates numerous signaling molecules that
initiate intracellular signaling cascades, including the
In vitro talc-stimulated mesothelial cells increased their b- ras/MAPK and PI3K/Akt pathways, ultimately promoting
FGF transcription and secretion.135 Pleural fibroblasts pro- cell migration, proliferation, differentiation and survival,
liferate when incubated with conditioned medium from and tumor resistance to radiotherapy and chemother-
talc stimulated mesothelial cells – this response is dimin- apy.142
ished by neutralizing antibodies against b-FGF.135 A higher This factor is produced by a variety of cells including
pleural level of b-FGF was seen in patients who had a suc- mesothelial cells, macrophages, smooth muscle cells,
cessful talc-induced pleurodesis, compared with those in endothelial cells, fibroblasts and many types of tumors,
whom talc pleurodesis failed.135 and large stores of PDGF-AB are found in a granules of
platelets. PDGF exerts its potent mitogenic and chemotac-
MALIGNANT PLEURAL DISEASES tic effects in a variety of mesenchymal cells such as fibrob-
lasts, vascular smooth muscle cells, glomerular mesangial
Basic fibroblast growth factor is a potent angiogenic factor, cells and brain glial cells.142
and is implicated in cancer growth and metastasis. Protein In the pleura, PDGF is thought to have active roles in
expression of b-FGF in solid tumors is correlated with normal and pathologic conditions. Mesothelial cells
heparanase expression, their co-expression is thought to produce PDGF which has potent mitogenic effects on
allow heparanase mediated release of b-FGF from the extra- fibroblasts and mesothelial cells.102,105 PDGF is a potent
cellular matrix.136 Mesothelioma proliferation in vitro can mediator of asbestos-induced pleural fibroblast prolifera-
be inhibited using a-tocopheryl succinate (a-TOS) which tion.143 Fibroblast proliferation was blocked using neutral-
downregulates FGF-2 transcription, thereby interrupting izing antibodies against PDGF.143 PDGF is also a
the FGF autocrine growth loop.137 One study reported that chemoattractant for neutrophils and monocytes and stim-
the mean b-FGF level is significantly lower in malignant ulates the activation of macrophages.144,145 It upregulates
78 Immunology
fibronectin gene expression,146 hyaluronan synthesis,147 gemcitabine and pemetrexed164 and combinational
procollagen synthesis102,148 and increases collagenase activ- chemotherapy for mesothelioma using imatinib mesylate
ity,149 processes essential for connective tissue remodeling awaits evaluation. Several other tyrosine kinase inhibitors
and fibrosis. The combination of EGF and PDGF produce are now available which will block PDGFR but their use in
a synergistic effect in stimulating collagen production by pleural disease has not been reported.166
mesothelial cells.102 PDGF also induces the expression of
TGF-b further potentiating the fibrotic response.150
Insulin-like growth factor
MESOTHELIOMA
The insulin-like growth factors (IGF-I and IGF-II) are
Elevated levels of PDGF-AB are also seen in patients with single-chain polypeptides with structural homology to
pleural mesothelioma compared with high risk and proinsulin. They are produced by most cells in the body
normal controls, and positive PDGF-AB levels were asso- and are abundant in the circulation, usually bound to high
ciated with lower survival.151 Mesotheliomas produce sig- affinity binding proteins (IGFBP), of which six have been
nificant amounts of PDGF. The PDGF-A chain, and less identified.167 These binding proteins protect IGF from
frequently the PDGF-B chain, is highly overexpressed in proteolysis and modulate its interaction with the IGF
mesothelioma compared with normal mesothelial cells, receptor (IGFR). The liver is the major source of circulat-
suggesting a role for PDGF in the pathogenesis of ing IGFs, which are synthesized in response to growth
mesothelioma.152,153 This is supported by studies showing hormone, and controls the growth and differentiation of
that the SV40 T antigen-immortalized human mesothelial most tissues of the body.168
cells were only tumorigenic after ectopic PDGF-A over- The IGFs bind with high affinity to two cell surface
expression.154 receptors, IGF-RI (or type I receptor) and IGF-RII (or type
Normal mesothelial cells predominantly express one II receptor). The IGF-RI receptor has a high degree of
PDGF receptor, PDGFR-a, but have weak to undetectable homology to the insulin receptor and binds IGF-I with
expression of another receptor, PDGFR-b.155,156 higher affinity than either IGF-II or insulin.167,169 It exists
Stimulation with TGF-b1 decreases PDGFR-a mRNA at the cell surface as a heterotetramer consisting of two a
expression in normal mesothelial cells.157 and two b subunits joined by disulfide bonds. When the
On the other hand, most mesothelioma cells express ligand binds to the extracellular a subunit there is a con-
PDGFR-b with little PDGFR-a.155,158,159 It has been sug- formational change that induces autophosphorylation of
gested that mesothelioma cells in effusions may express the tyrosine residues within the intracellular segment of the b
receptor and therefore possess an autocrine stimulatory subunit. This leads to receptor tyrosine kinase activity and
loop.160 However, such an autocrine loop is likely to only the induction of various intracellular signaling cascades,
exist for PDGF-B. PDGF-B, acting in conjunction with including MAP kinase and phosphoinositide 3¢-kinase
integrin a3b1, is a chemoattractant for mesothelioma pathways, which leads to growth and metabolic
cells.161 Overexpression studies also showed that PDGF-A responses.169
is tumorigenic in vivo through paracrine, but not The IGF-RII is identical to the cation-independent
autocrine, mechanisms of cell proliferation.155 mannose 6-phosphate receptor. It exhibits higher affinity
Platelet-derived growth factor has also been recovered for IGF-II than IGF-I and does not bind insulin.168 The
in higher levels in pleural effusions from adenocarcinomas IGF-RII has no intrinsic tyrosine kinase activity, hence the
of the lung than those from small cell lung cancers or intracellular mechanisms by which the receptor can
benign effusions.162 mediate its biological effects, including lysosomal enzyme
trafficking, clearance and/or activation of a variety of
THERAPEUTIC APPROACHES TO INHIBIT PDGF growth factors and endocytosis-mediated degradation of
IGF-II, is unclear.169
Studies applying dipyridamole on mesothelial cells
demonstrated inhibition of PDGF-induced proliferation MESOTHELIOMA
through attenuated ERK activity, preservation of
p27(Kip1) and decreased pRB phosphorylation,163 This Normal human mesothelial cells, as well as mesothelioma
may represent a potential therapeutic strategy against cell lines, express IGF-I, IGF-II, some of the IGF binding
pleural fibrosis. proteins (IGFBP), IGF-RI and IGF-RII,124,170,171 and IGF-1
Imatinib mesylate selectively blocks tyrosine kinases, is mitogenic for mesothelial cells in vitro.170 SV40-induced
including PDGFR-b. Imatinib induces cytotoxicity and transformation of mesothelial cells requires the presence of
apoptosis selectively on PDGFR-b positive mesothelioma a functional IGFR, suggesting an important role for IGF in
cells, via blockade of receptor phosphorylation and inter- the development of mesothelioma.172 This is supported by
ference with the Akt pathway.164 However, phase II clinical studies showing decreased growth and tumorigenicity of
trials using imatinib mesylate alone against mesothelioma SV40-induced hamster mesotheliomas following treat-
showed no obvious benefits.165 Imatinib synergises with ment with antisense IGFR transcripts.173 More recently, a
Cytokines in pleural fluid formation 79
novel IGF-R1 inhibitor, NVP-AEW541, has shown con- The hypersecretion of pro-IGF-II is associated with
centration-dependent inhibitory effects on cultured suppression of growth hormone secretion. Application of
mesothelioma cells.174 Further evaluation of NVP- recombinant human growth hormone (rhGH), and in
AEW541 is required for its possible therapeutic potential in some studies with glucocorticoid treatment, reverses the
vivo. The IGFR adaptor proteins, insulin receptor substrate suppression and has been successfully used as an alterna-
(IRS)-1 and -2, also appear to play roles in the phenotype tive to surgery in a small number of patients to alleviate
of mesothelioma. Mesotheliomas that signal through hypoglycemia.186,187 The mechanism by which the rhGH
IRS-1 have increased cellular growth whereas those that controls hypoglycemia in non-islet cell tumors, including
signal through IRS-2 have increased motility.170 This may SFT, is unclear, but it is likely to involve multiple path-
affect tumor pathology as preferential signaling through ways, one of which is through reduction of the bioavail-
IRS-1 may lead to increased local spread of the tumor ability of IGF-II.187
whereas utilization of IRS-2 may promote metastases.175 The possible use of IGF-1 and IGFBP-2 as potential
Mesotheliomas express IGFBP 2, 4 and 5 but not 1, 3 or markers of malignant effusions has also been suggested.188
6. The absence of the beneficial IGFBP-3 together with the In 25 patients with malignant, infective or congestive heart
presence of the deleterious IGFBP-4 would allow for a failure effusions, IGF-1 and IGFBP-2 levels in effusions of
more aggressive phenotype.176 However, in another study, malignant solid tumors were significantly higher than in
IGFBP1-4 were overexpressed whereas IGFBP-5 was lymphoma, followed by infection and transudative effu-
underexpressed.177 In this case, IGFBP-5 may be an sion of congestive heart failure. In effusion of solid tumors,
inhibitor of IGF-1 activation and its decrease could lead to IGFBP-2 levels were higher than those in corresponding
over stimulation of the receptor and autocrine stimulation sera, which suggests local production of this binding
or growth.170 More studies need to confirm these observa- protein.188
tions.
Solitary fibrous tumor (SFT) is a mesenchymal tumor Vascular endothelial growth factor
which arises at a variety of sites including the pleura (see
Chapter 38). Approximately 10–15 percent of tumors Vascular hyperpermeability and plasma leakage is funda-
behave in a malignant fashion.178 mental to the development of most exudative pleural effu-
Many SFTs secrete IGF-II which is believed to be the sions. Compelling experimental evidence demonstrates
cause of intermittent hypoglycemia present in 4–5 percent that vascular endothelial growth factor (VEGF) is a crucial
of patients. In several case studies, high levels of IGF-II mediator in pleural fluid formation, and clinical trials are
were reported in the patient’s serum prior to surgery.179–181 underway using VEGF antagonists in the management of
Larger amounts of high molecular weight IGF-II were also malignant pleural effusions (see review189).
found in the tumor cystic fluid than in the serum.182 The This factor, initially known as vascular permeability
high molecular weight IGF-II (‘big’ IGF-II) is most likely factor, is a potent inducer of vascular permeability and the
an incompletely processed IGF-II precursor (pro-IGF-II). permeability of the mesothelial monolayer.190 VEGF pro-
In all cases, the hypoglycemia resolved upon removal of motes microvascular permeability by enhancing the activ-
the tumor183 and IGF-II levels in the serum returned to ity of vesicular vacuolar organelles, through which
normal.182 Others have suggested that altered IGF-1 activ- macromolecules extravazate. It can also increase active
ity may also play a role in hypoglycemia associated with trans-endothelial transport via pinocytotic vesicles. VEGF
these tumors, although varying results have been reported increases the capillary and venular leakage by opening the
for IGF-1.184 Chang and colleagues184 found that serum endothelial intercellular junctions and by inducing fenes-
IGF-I levels were normal in patients with SFT and IGF-1R trae development in endothelia. In vivo it is also a vasodila-
was highly expressed on tumor cells. However, Li and tor. Exogenous VEGF administration induces a reversible
coworkers185 failed to find IGF-IR expression on these increase in vascular permeability within minutes. VEGF
tumors. also has diverse effects on vascular endothelial cells and is
The downstream oncogenic pathways of IGF-II are not capable of inducing morphological changes, stimulating
clear but a recent study showed that insulin receptor (IR) cell proliferation and migration, altering their gene expres-
signaling pathways are constitutively activated in SFTs.185 sion and inhibiting apoptosis.
IGF-II has been thought to mediate many of its biological
effects through IGF-IR, however this study did not find VEGF LEVELS IN PLEURAL EFFUSIONS
expression of IGF-IR in these tumors. IGF-II is known to
bind to an IR isoform (IR-A), with an affinity similar to Vascular endothelial growth factor is present in significant
that of insulin. It was shown that the IGF-II binding IR-A quantities in pleural and peritoneal effusions of varying
was the predominant isoform in SFTs suggesting that IGF- etiologies, and its level is consistently higher in exudative
II/IR signaling plays an oncogenic role in SFTs.185 than in transudative pleural effusions.190–193 Although
80 Immunology
VEGF levels are higher in malignant than in benign effu- into mice, lung lesions from adenocarcinoma (high VEGF
sions,194 the VEGF level is of limited use as a diagnostic output) express more VEGF and the tumors are more vas-
tool191,192 for malignant effusions, or to predict cancer cular than the squamous cell cancer (which produce lower
staging193 and histological cancer cell types.193 VEGF) lesions. VEGF expression correlates directly with
Different isoforms of VEGF have been shown to be the volume of effusion formed. Importantly, transfection
involved in the evolution of malignant pleural effusions of adenocarcinoma cells with antisense VEGF165 gene, sig-
with different characteristics by the implantation of lung nificantly reduces the vascular permeability and pleural
cancer cells transfected with these isoforms.195 VEGF-A effusions. Conversely, transfecting the squamous cancer
promotes tumor cells dissemination, capillary neogenesis cells with sense VEGF gene increases their VEGF produc-
and bloody effusions, while VEGF-D promotes both tion, and results in significantly more fluid accumulation.
pleural and lymph node tumor dissemination.195 Similar results apply to malignant ascites development.217
The majority of VEGF in the effusions is believed to Promising results are rapidly accumulating on the use
originate from local pleural or peritoneal production.196,197 of VEGF inhibition in preventing pleural and peritoneal
Quiescent mesothelial cells produce large amounts of fluid accumulation.189 VEGF activity can be antagonized
VEGF, which can be increased at mRNA and protein level with various techniques, including antibodies that block
by FGF-2.198 Mesothelial cells are likely to represent the VEGF binding to its receptors, inhibitors of the tyrosine
principal source of pleural fluid VEGF,29,190 but infiltrating kinase functions of the VEGF receptors, and antisense
inflammatory cells and malignant cells (in malignant pleu- nucleic acids to disrupt cellular VEGF production.218,219
ritis) also contribute to VEGF production.189 Many of these methods have been successfully used to
Most malignant cell types overexpress VEGF.199,200 This inhibit tumor growth in animals and are now in phase
includes lung201 and breast202 carcinomas and mesothe- II/III clinical trials.219 Antisense oligonucleotide (ODN),
lioma.203 Immunohistochemistry has confirmed the pres- inhibiting VEGF and VEGF-C expression, is capable of
ence of VEGF in malignant cells in human pleural specifically inhibiting mesothelioma cell growth. VEGF-R2
tissue,194 suggesting that they may be a source of the con- and VEGF-R3 blocking antibodies act synergistically to
sistently high VEGF in malignant pleural effusions. VEGF inhibit mesothelioma cell growth.204
is likely to be an autocrine growth factor in mesothe- Several VEGF receptors have been described, amongst
lioma.197 Functional coexpression of VEGF-C (a protein which two signaling tyrosine kinase receptors, Flt-1
closely related to VEGF), as well as its receptor VEGFR-3, (VEGFR-1) and KDR/Flk-1 (VEGFR-2), are most
has been demonstrated in malignant mesothelioma.204 studied.213,220 VEGF-R3 is the receptor for VEGF-C, a mol-
Empyema fluids contain high levels of VEGF,191,205 sig- ecule closely related to VEGF.204 VEGF receptors are
nificantly above those in uncomplicated parapneumonic expressed primarily on endothelial cells, but are also found
effusions.190 Staphylococcus aureus, a common causative in many diverse normal and malignant cell types.213
organism in empyema, can stimulate a dose- and time- Pleural tissues, in both healthy and diseased states, express
dependent VEGF release from mesothelial cells.190 significant levels of VEGF receptors.205 Upon ligand
Acidosis206 and hypoglycemia,207 common biochemical binding to the receptors, the intracellular signaling for
characteristics of empyema fluids, are also known to various VEGF functions are mediated via the MAPK, PKC
induce VEGF. VEGF is also elevated and correlated with and PI/Akt pathways,221 which are potential target points
other cytokines such as IL-1b and TNFa in tuberculous for inhibition of VEGF activities.222
pleural effusions.208 Phosphorylation blockade of the VEGF receptor (with
In post-coronary artery bypass grafting (CABG) pleural an oral inhibitor of tyrosine kinase phosphorylation of
effusions, VEGF levels correlate with inflammatory cells KDR/Flk-1 and Flt-1 receptors) inhibits the formation of
and lactate dehydrogenase (LDH) levels, as well as with malignant effusion in mice with lung adenocarcinomas by
protein levels which reflect the degree of vascular hyper- reducing vascular permeability.223 A neutralizing antibody
permeability.209 VEGF is likely to contribute to the gener- (A4.6.1) that blocks VEGF access to both its receptors
alized hyperpermeability and the resultant formation of completely inhibits ascites formation, tumor growth and
large effusions in ovarian hyperstimulation syndrome210 prolongs the survival in mice inoculated intraperitoneally
and in Meigs’ syndrome.211 with human ovarian cancer cells.224 Antibodies against the
Hypoxia and ischemia are amongst the most estab- KDR/Flk-1 receptors (DC101) and exogenous soluble
lished stimulators of VEGF212–215 and may contribute to human Flt-1 receptors (as an inhibitor of VEGF binding)
the pathogenesis of pleural effusion from pulmonary also effectively inhibit malignant ascites accumulation in
emboli.191 mice.225,226
A VEGF tyrosine kinase inhibitor (SU5416) is under
TARGETING VEGF TO CONTROL PLEURAL OR PERITONEAL clinical trial against recurrent pleural effusion and ascites
EFFUSIONS formation.227 ZD6474, an antagonist of VEGFR-2 tyrosine
kinase, can inhibit formation of malignant pleural effu-
Vascular endothelial growth factor is essential in malig- sions resulting from adenocarcinoma cells inoculated into
nant effusion formation.216 When cancer cells are injected nude mice.228 Further clinical studies are anticipated to
Immunomodulatory cytokines 81
contrast, TNF-a causes mesothelial cell cycle arrest in the lioma cells in vitro.268 TNF-a and IL-1b both enhance
G0/G1 phase.256 attachment of colon and pancreatic carcinoma cell adhe-
sion to mesothelial monolayers in vitro.269,270 TNF-a has
HUMAN PLEURAL EFFUSIONS also been shown to promote peritoneal metastasis in
vivo.271 Recently, the role of TNF-a in malignant effusion
Both TNF-a and IL-1b are present in pleural effusions of formation has been confirmed in a murine model of
various causes. Their levels are significantly higher in exu- malignant pleural effusion using lung cancer cell lines,
dates than transudates, and in loculated than in free- raising the possibility of TNF-a as a novel therapeutic
flowing pleural effusions.256 In tuberculous and empyema target in the management of pleural malignancies.272
effusions their levels correlate with each other.257 Their
levels in pleural fluids likely reflect the degree of local
inflammation. Very high IL-1 levels have been reported in THERAPEUTIC INHIBITION OF TNF-a
empyema fluids.257 However, the pleural fluid levels of Anti-TNF-a agents are now available and their anti-
TNF-a and IL-1 are not diagnostically useful as consider- inflammatory effect in the pleura has been confirmed in
able overlap exists among effusions of various etiologies. animal models. TNF-a-converting enzyme (TACE)
Mesothelial cells are likely the major source of pleural cleaves the precursor form of TNF, allowing the mature
fluid TNF-a and IL-1b. Pleural fluid TNF-a is significantly form to be secreted into the extracellular space. In the
higher than serum TNF-a in the malignant and in para- pleural space, a TACE inhibitor significantly reduces the
pneumonic effusions.258,259 Conflicting results exist for pleural fluid TNF-a accumulation in zymosan-induced
IL-1: its concentrations were higher in serum than in pleural inflammation.273 Similarly, GW3333, a dual
pleural fluids in one study,259 but the reverse was reported inhibitor of TACE and MMPs, effectively inhibits the
in another.257 increase in TNF-a and the associated influx of inflamma-
tory cells in zymosan-induced pleuritis.274
PLEURAL FIBROSIS
Steroids can also inhibit TNF-a and IL-1 production in
Amongst its pleiotropic nature, TNF-a is known to stimu- the pleura. FR167653, a cytokine synthesis inhibitor, also
late fibroblast proliferation and regulate collagen pro- suppresses TNF-a and IL-1b accumulation in the pleura in
duction.260 TNF-a induce epithelial–mesenchymal rat carrageenin-induced pleurisy. As a result, it inhibits
transformation of mesothelial cells, altering expression of plasma exudation and leukocyte infiltration and signifi-
cytokeratins 8 and 18, MMP-9 and collagen, but not that cantly lowered the prostanoid levels in the exudates.275
of vimentin.261 It is likely that TNF-a has a role in the These agents thus may have a role in treating pleural
fibrotic processes in the pleura. In vitro, TNF-a has been inflammation and may also prevent the development of
shown to stimulate and reduce mesothelial cell production subsequent fibrothorax. Anti-IL-1 strategies are available
of plasminogen activator inhibitor (PAI)-1 and tissue-type but little is known about their clinical effectiveness in the
plasminogen activator (tPA).262 Experimentally, anti- pleural setting.
TNF-a antibodies inhibit talc pleurodesis.263 In clinical
studies, pleural fibrosis from tuberculous pleuritis occurs
more commonly in patients with higher pleural fluid
IL-2
TNF-a and IL-1 levels.257
Polymorphisms of TNF-a have recently been associated Interleukin 2 is a pleiotropic cytokine with a variety of
with silicosis264 and asbestosis265 and genetic predisposi- effects on the immune system. Its effects are often environ-
tion may have a role in pleural fibrosis. ment-specific; hence conflicting accounts have been
reported.
PLEURAL MALIGNANCY The interaction of IL-2 with the IL-2 receptor induces
proliferation and differentiation of various T lymphocyte
The role of TNF-a in the biology of mesothelioma is com- subsets, and stimulates a cytokine cascade that includes
plicated. Recent evidence suggests a critical role of TNF-a various interleukins, interferons and TNF-a. IL-2 induces
in asbestos-induced malignant transformation of mesothelial cell expression of CCR2 and cell proliferation.
mesothelial cells. Although asbestos is usually cytotoxic to Intrapleural IL-2 administration induces nitric oxide
mesothelial cells, treatment with TNF-a significantly accumulation in the pleural fluid.276 Haptotactic
reduced asbestos cytotoxicity via activation of nuclear migration is upregulated when mesothelial cells are
factor (NF)-kB. This protective mechanism allows more cultured with IL-2 (see Chapter 7, Pleural inflammation
asbestos-damaged mesothelial cells to survive and undergo and infection).
malignant transformation.266 TNF-a and IL-1b are
involved in malignant transformation of human benign IL-2 LEVELS IN PLEURAL EFFUSIONS
mesothelial cells induced by erionite.267
Tumor necrosis factor-a has also been shown to Levels of IL-2 as well as of soluble IL-2 receptor (sIL-2R)
enhance SN38-mediated apoptosis in malignant mesothe- are higher in tuberculous effusions than in malignant ones,
Immunomodulatory cytokines 83
though the large degree of overlap precludes their use as a Interleukin-2 has been used in the treatment of
diagnostic tool.277–279 sIL-2R is a marker of T-lymphocyte mesothelioma via different routes of delivery, though the
activity: high levels of sIL-2R are present in other inflam- results were generally disappointing. Thirty-one patients
matory effusions, especially rheumatoid pleuritis.280 It has were given repeated intrapleural instillation of IL-2 for
been suggested that effusion levels of sIL-2R and IFN-g are 4 weeks, followed by regular subcutaneous IL-2 injections
useful post-treatment markers of pleural thickening.278 for up to 6 months. Only one patient had a complete, and
six a partial, tumor response. Interestingly, 90 percent of
THERAPEUTIC APPLICATION OF IL-2 IN PLEURAL patients had no further or minimal (asymptomatic)
MALIGNANCIES pleural fluid collection. Side effects included fever (19
percent) and cardiac failure (3 percent) and occurred
Recombinant IL-2s (e.g. aldesleukin, teceleukin) are non- mainly after intrapleural instillations.290 In another study
glycosylated, modified forms of the endogenous com- using intrapleural IL-2 for 5 days in 22 mesothelioma
pound. High-dose IL-2 results in objective regression of patients, 11 partial responses and one complete response
metastatic melanoma and renal cell carcinoma in were detected after 36 days.291
approximately 15 percent of patients,281 though response Intravenous IL-2 therapy, followed by regular subcuta-
rates are lower in other malignancies.282 The growth neous injections, was well tolerated by 29 previously
inhibitory effect of IL-2 on cancer cells varies with the untreated mesothelioma patients, but only two patients
proliferative status of these cells. In highly proliferating achieved a partial response.292 Combining IL-2 with epiru-
mesothelioma cells, a reduction of malignant cells in the bicin results in significant toxicity, without improving the
S-phase, with an accumulation in G0/G1 of the cell cycle response rate (5 percent).293 IL-2 has also been applied to
followed by apoptosis, is observed. In cells proliferating malignant pleural mesothelioma during a phase II study of
at a lower rate, IL-2 produces late cytotoxic effects with a four-modality treatments that included preoperative IL-
resultant apoptosis, without obvious influence on the cell 2, pleurectomy/decortication, intrapleural IL-2, adjuvant
cycle.276 radiotherapy, systemic chemotherapy and long-term sub-
Interleukin-2 has anti-proliferative effects on human cutaneous IL-2. The median survival was 26 months for
malignant mesothelioma cells and thus has been used for patients with stage II/III mesothelioma.294
the management of malignant pleural effusions283 and
mesothelioma. IL-2 also augmented the in vitro cytolytic VASCULAR LEAK SYNDROME
activity of mesothelial cells and cytolytic T-cells, isolated
from malignant pleural effusions, against autologous When high dose intravenous IL-2 is used to treat metasta-
tumors. Lymphocytes recovered from malignant pleural tic diseases, vascular leak syndrome (VLS) can occur which
effusions show depressed proliferation, IFN-g production often limits the dosage that can be administered. VLS is
and cytolytic activity, as compared with lymphocytes from characterized by an increase in vascular permeability
tuberculous effusions. Both IL-7 and IL-12, as well as accompanied by significant extravazation of fluids and
TCR-CD3 (T-cell receptor-CD3), in the presence of IL-2, proteins resulting in interstitial edema, anasarca and, at
can restore the immunosuppressed cytolytic activity of the times, cardiovascular or respiratory failure. Pleural, peri-
lymphocytes of malignant effusions against autologous cardial and peritoneal effusions occur as part of the gener-
tumors.284,285 alized vascular hyperpermeability.295 VLS is generally
In a murine model of mesothelioma, intratumoral believed to involve endothelial cell damage and is medi-
injection of IL-2 induced tumor regression in part via ated by nitric oxide. CD4+/CD25+ T-regulatory cells have
CD8+ T-cells. Complete regression was seen with small an important regulatory role, as their depletion in murine
tumors, but less so with bulky lesions.286,287 models demonstrates a significant increase in IL-2 induced
In humans, IL-2 as a palliative therapy for malignant VLS.296 Natural killer (NK) and polymorphonuclear cells
effusions has been reported in case series. Intracavitary are important in late events, including edema, of VLS.297
administration of low-dose IL-2, as an initial treatment, The potential role of IL-2 in vascular leakage and pleural
resulted in an objective clinical response in 72 percent for fluid formation in other conditions has not been explored.
a median duration of 5 months in 100 patients with malig-
nant (68 percent pleural) effusions. The response is better
in pleural than in peritoneal effusion control.288 Similar IL-6
results were achieved in another study of 21 patients with
malignant pleural effusions from non-small cell lung car- Interleukin-6 is a member of a family of closely related
cinomas. Complete response was achieved in 33 percent, pleiotropic cytokines which also include IL-11,
and partial response in 29 percent, with a median duration IL-27, IL-31, leukemia inhibitory factor (LIF), oncostatin
of 8 months (range 4–10 months).289 Treatment was well M (OSM), ciliary neurotrophic factor (CNF) and car-
tolerated in both studies. The effectiveness of IL-2 has yet diotrophin 1 (CT-1), cardiotrophin-like cytokine (CLC)
to be compared with conventional therapies for malignant and neuropoietin (NPN).298 Individual family members
pleural effusions. play important roles in the immune, nervous, cardiovas-
84 Immunology
cular and hemopoietic systems, as well as bone metabo- nitric oxide synthase and cylco-oxygenase-2, are signifi-
lism, inflammation, wound repair, the acute phase cantly reduced in IL-6 knockout mice compared with
response and development of the embryo.299 Their actions wild-type controls.309 Pretreatment of the wild-type mice
are mediated through specific cell surface receptors, con- with IL-6 neutralizing antibodies before carrageenan treat-
sisting of a unique a or b chain and the shared signal trans- ment results in the same responses as for the knockout
ducing subunit, glycoprotein b-subunit (gp)130 as part of animals.309
a multimeric (a, b) receptor complex.300,301 By itself, gp130 Interleukin-6 dictates the transition from acute to
does not provide high-affinity binding to any of the IL6- chronic inflammation by changing the nature of the leuko-
family cytokines, but rather converts low-affinity binding cyte infiltrate (from polymorphonuclear neutrophils to
to ‘specific’ a-subunit receptors into a high-affinity recep- monocytes/macrophages).310–312 IL-6 is also involved in
tor complex.302 the development of specific cellular and humoral immune
Soluble forms of the a-subunits lacking the transmem- responses, including B cell proliferation and differentia-
brane and cytoplasmic domains have been identified for tion, immunoglobulin secretion and T-cell activation, thus
IL-6, IL-11 and LIF, as well as gp130.302 The soluble IL-6 favoring chronic inflammatory responses.313
receptor (sIL-6R) is generated either by limited proteolysis
of the IL-6R303 or translation from alternatively spliced IL-6 LEVELS IN PLEURAL EFFUSIONS
mRNA.304 sIL-6R can only bind its ligand, IL-6, however
the complex of IL-6 and sIL-6R can bind to and activate Interleukin-6 is produced by various cell types including T
gp130 on cells which do not express the membrane-bound cells, B cells, monocytes, fibroblasts, keratinocytes,
IL-6R, thereby initiating signaling.305 endothelial cells, mesangial cells and some tumor cells.308
Gp130 has no intrinsic tyrosine kinase activity and thus In serosal tissues, IL-6 is also secreted by mesothelial and
requires recruitment and activation of specific kinases and malignant mesothelioma cells.252,314–318 Elevated IL-6 levels
docking proteins. Following binding of the cytokine to its have been demonstrated in the pleura and other serosal
a or b subunit and subsequent dimerization with gp130, cavities following injury or disease. High levels of pleural
cytoplasmic janus kinases (JAKs) are recruited and phos- fluid IL-6 is associated with pleural tuberculosis (TB).
phorylate gp130.306 The subsequent activation of intracel- Initial IL-6 levels prior to treatment correlates with the
lular signaling is dependent on specific phosphotyrosine number of febrile days and the percentage change of
residues on gp130, which act as docking sites for SH2 cytokines after 2 weeks of treatment helps to predict resid-
domain-containing intermediate signaling molecules. For ual pleural scarring.319 Elevated IL-6 levels together with
example, the Src homology protein tyrosine phosphatase 2 increases in other known inflammatory cytokines was
(SHP2) binds to membrane proximal tyrosine residues measured in the pleural lavage of patients suffering from
(Y759 in humans, Y757 in mice) and is necessary and suf- primary spontaneous pneumothorax showing that this
ficient for activation of the MAPK/ERK pathway. In con- condition is associated with a substantial pleural inflam-
trast, members of the STAT transcription factors dock to matory reaction.320 Elevated pleural IL-6 has also been
several membrane distal phosphotyrosine residues. Once shown in experimental animal models using zymosan and
phosphorylated, STAT proteins form a homodimer, carrageenan to induce pleuritis.321,322 Intrapleural instilla-
translocate to the nucleus and activate target genes.307 This tion of TNF-a in rats caused a sharp rise in plasma IL-6
system is negatively regulated by a number of inhibitory which stimulated an increase in T-kininigen. Similarly, in
molecules, most notably the family of suppressor of a clinical trial involving mesothelioma patients,
cytokine signaling (SOCS) proteins. intrapleural administration of TNF-a induced a rise in IL-
6 levels both in pleural fluid and serum.322 IL-6 levels were
ROLE OF IL-6 IN INFLAMMATION also markedly increased in the pleural fluid following
intrapleural tetracycline injection in patients with malig-
Interleukin-6 plays important pro-inflammatory roles. In nant pleural effusion.323 Surgery can also stimulate pro-
the acute inflammation phase, IL-6 stimulates hepatocytes duction of IL-6 in serosal cavities. In the peritoneum
to produce acute phase proteins, e.g. C-reactive protein, 1 hour after laparotomy, the levels of IL-6 were signifi-
fibrinogen, a1-antitrypsin and serum amyloid A, and cantly greater than immediately after the procedure.324
simultaneously suppresses albumin production. It also Mesothelioma patients who underwent pleurectomy or
induces the secretion of the iron regulatory protein hep- extrapleural pneumonectomy followed by intraoperative
cidin, hence excessive production of IL-6 leads to hypofer- photodynamic therapy also showed significant increases in
remia of inflammation.299,308 Overexpression of IL-6 also serum IL-6 levels.325
causes leukocytosis and fever and promotes the produc- In disease states, IL-6 levels are consistently higher in
tion of VEGF.308 When exposed to carrageenan-induced the pleural fluid than in the corresponding sera or plasma,
pleuritis, knockout mice exhibit reduced pleural exudation with no direct correlation between pleural effusions and
and polymorphonuclear cell migration. Downstream peripheral blood.326–328 In one study, the IL-6 concentra-
inflammatory changes, including lung myeloperoxidase tion in the pleural fluid of patients with adenocarcinoma
activity, lipid peroxidation and the expression of inducible was 60 to 1400 times higher than in the serum.327 These
Immunomodulatory cytokines 85
data suggest that IL-6 is locally produced, probably by associated with thrombocytosis.330 It has been hypothe-
mesothelial and inflammatory cells, rather than from dif- sized that large amounts of IL-6 from the pleural fluid of
fusion from the systemic circulation. patients with mesothelioma leak into the systemic circula-
Pleural fluid IL-6 is higher in exudative versus tran- tion and induce the systemic inflammatory reactions
sudative effusions, malignant versus benign effusions and related to mesothelioma,330 including thrombocytosis,
TB pleuritis versus malignant or other parapneumonic fever and cachexia.338
effusions.259,326,328,329 The concentration of IL-6 is also Production of IL-6 by mesotheliomas has also been
higher in effusions from mesotheliomas than those from measured as a way to predict tumor response to
adenocarcinomas.330 However, the sensitivity and speci- chemotherapy. Gemcitabine, and to a lesser extant irinote-
ficity of IL-6 in any of these conditions is inadequate for it can, but not most other agents, inhibit IL-6 secretion at
to be used diagnostically. low doses but stimulate a surge in IL-6 at higher doses.315
Pleural effusion levels of IL-6 clearly reflect disease These results suggest a palliative role for low doses of these
states but changes in the concentration of serum IL-6 is chemotherapeutic agents in non-responders by decreasing
not so obvious. Serum IL-6 is not significantly different in the secretion of IL-6.
patients with exudates compared with those with transu-
dates329 and is only slightly raised in patients with tubercu-
lous pleurisy compared with controls.331 However, OTHER PLEURAL DISEASES
patients with malignant pleurisy more frequently had ele- There is growing evidence for a role of the IL-6 family of
vated serum IL-6, fibrinogen, fibrinogen degradation cytokines in fibrosis of various tissues, in particular IL-6,
products (FDP) and C-reactive protein levels compared IL-11 and OSM.339 To date, there have been no published
with lung cancer patients without malignant pleurisy.332 studies examining the effect of these molecules on pleural
This suggests that the pleural IL-6 is absorbed systemically, fibrosis but it is likely that they will play a significant role
which induces increases in plasma fibrinogen and subse- in the fibrotic pathway in pleural and other serosal cavities.
quently FDP. In contrast to IL-6, sIL-6R concentrations
are much higher in serum than in corresponding pleural
fluids in patients with malignant or infective pleural effu- OTHER IL-6 FAMILY PROTEINS IN PLEURAL DISEASE
sions.327,333 sIL-6R levels are also low in pleural effusions
when the levels of IL-6 are high, which may reflect a down- Very few studies have examined any of the IL-6 family
regulation of sIL-6R expression in the presence of exces- cytokines, other than IL-6, in pleural or serosal biology.
sive amounts of IL-6.333 LIF is present in higher levels in infective and malignant
A role for IL-6 in pleural fluid formation has also been pleural effusions than in transudates. Pleural fluid levels of
suggested in several gynecological conditions. Increased LIF correlated with IL-8 levels in malignant effusions and
amounts of pleural fluid and elevated IL-6 levels have been with IL-4 in infective effusions.340 Mesothelioma cells have
observed in Meigs’ syndrome132 and in severe ovarian been shown to secrete LIF as well as IL-6341 and mesothe-
hyperstimulation syndrome.334 lial cells express the OSM receptor.327,331 However, it is
likely that other IL-6 family proteins play important roles
in maintaining pleural and serosal hemostasis and tissue
MESOTHELIOMA
repair, as well as in the pathogenesis of pleural disease.
Normal mesothelial cells produce IL-6 in response to
stimuli and express gp130.314 Therefore, it has been pre-
dicted that IL-6 acts as an autocrine factor for mesothelial IL-8
cells.314 However, mesothelial cells do not express the
IL-6R, therefore activation of mesothelial cells by IL-6 Interleukin-8 is one of the most studied chemokines, and
must occur when complexed to the sIL-6R, possibly shed its role in pleural inflammation is well established.
by neutrophils following initiation of inflammation.310 Mesothelial cells express IL-8 basally.342,343 Inflammatory
Mesothelioma cells produce and secrete high levels of stimuli,344 asbestos fibers345 and infective agents,346,347
IL-6317,335,336 but express only low levels of IL-6R mRNA.337 among others, are known to stimulate significant increases
It was recently shown that IL-6 together with sIL-6R stim- in mesothelial cell production of IL-8.
ulated mesothelioma cell growth and induced expression Interleukin-8 is a downstream cytokine from TNF-a
of vascular endothelial growth factor in vitro, via STAT3 and IL-1. In humans, significant elevation of pleural levels
signaling.337 This suggests an autocrine role for IL-6 in the of IL-8 is seen after intrapleural administration of TNF-a
development of mesothelioma. A high incidence of throm- in mesothelioma patients.348 In vitro, exposure of mesothe-
bocytosis (48 percent) and a significant correlation lial cells to TNF-a or IL-1b stimulates IL-8 release in a
between platelet count and serum IL-6 levels have been time- and dose-dependent fashion,248,252,342,343,345,349–352
reported in mesothelioma patients.330 This is consistent which is diminished by IFN-g.353 The stimulatory effects of
with the observation that in patients with tuberculous TNF-a and IL-1 can be synergistic.342 IL-8 secretion from
pleurisy, high levels of IL-6 in the pleural fluid are also pleural fibroblast is also induced by IL-1, TNF-a and
86 Immunology
LPS.354 Conversely, antibodies to TNF-a or IL-1 inhibit a-melanocyte stimulating hormone, has been shown in
IL-8 release in mesothelial cells.352 The neutrophil chemo- mesothelioma cells. Interruption of this circuit enhanced
tactic activity of supernatants from mesothelial cells stim- IL-8 expression on mesothelioma cells.365
ulated with either TNF-a or IL-1 is completely neutralized
with IL-8 antiserum.248 IL-1 receptor antagonist also PLEURODESIS
inhibits asbestos-induced IL-8 production from mesothe-
lial cells.351 Similarly, in animal models of LPS-induced As discussed in previous sections (see under TGF-b), the
pleuritis, IL-8 release is inhibited by anti-TNF-a antibod- conventional method of pleurodesis involves inducing
ies, but the production of TNF-a is not affected by anti-IL- acute pleural inflammation, which then progresses to
8 treatment.355 chronic inflammation and pleural fibrosis. As such, most
pleurodesing agents stimulate an acute pleural injury, and
CHEMOTAXIS hence production of IL-8, which holds an important role
in successful pleurodesis.
Interleukin-8 plays a crucial role in neutrophil influx and Intrapleural instillations of talc,366,367 tetracycline323,368
also participates in the recruitment of monocytes and lym- and OK-432369,370 have all been shown to stimulate a rapid
phocytes from the vascular compartment to the pleural rise in mesothelial cell production of IL-8, or pleural IL-8
space.356,357 In vitro, mesothelial cells produce IL-8 in a polar accumulation in humans, which is then followed by
fashion358 by releasing more IL-8 toward the apical surface, pleural neutrophilia. A good correlation between IL-8
providing a gradient to attract neutrophils from the basal levels and neutrophil chemotactic response has been
side of the mesothelium towards the pleural cavity.358 observed.367,369 Specific neutralization or removal of IL-8
Similarly, IL-8 derived from peritoneal mesothelial cells by antibody column significantly inhibited the neutrophil
contributes to the intraperitoneal recruitment of leukocytes chemotaxis366,369 and is likely to inhibit pleural fibrosis/
during peritoneal inflammation. Neutrophil migration was pleurodesis. However, the intrapleural administration of
significantly reduced in the presence of IL-8 antibody.344 IL-8 itself was ineffective in producing pleurodesis or facil-
Pleural fluid levels of IL-8 are consistently higher than itating the pleurodesis with talc (Vargas and Light, per-
its serum levels, supporting local production as the pre- sonal communications).
dominant source of IL-8 in the pleura.259,331,359 Consistent
with its role as a neutrophil chemotaxin, IL-8 concentra-
tions in empyema are higher than in effusions of other IL-10
causes,356,359–361 and the pleural fluid neutrophil count
correlates with pleural fluid IL-8 levels in most Interleukin-10 is an antiinflammatory cytokine with
studies.356,359,361 In vitro, conditioned media from potent immunosuppressive properties. In humans, the
Mycobacterium tuberculosis infected pleural macrophages main sources of IL-10 are lymphocytes and monocytes,
induce IL-8 secretion from mesothelial cells, which may but macrophages, mast cells and eosinophils also synthe-
explain the high IL-8 concentrations detected in tubercu- size IL-10.371,372 Circulating levels of IL-10 are elevated in
lous pleural effusions.362 patients with sepsis and have been associated with an
Antagonizing IL-8 activity can effectively reduce adverse clinical outcome. Experimental studies show that
inflammatory cell influx and pleural inflammation. Anti- exogenous IL-10, as an anti-inflammatory agent, can
IL-8 antibodies decrease chemotactic activity in empyema improve outcome in sepsis372 and inhibit airway inflam-
liquids,356,361 significantly inhibit neutrophil recruitment mation in asthma.371 IL-10 also carries strong immuno-
in endotoxin-induced pleurisy346 and diminish the neu- suppressive actions mediated through the downregulation
trophil chemotaxis into the pleural space following local of pro-inflammatory cytokines, the major histocompati-
instillation of crocidolite asbestos.345 bility complex (MHC) class II molecules and T-cell medi-
ated inflammatory responses, including delayed
ANGIOGENESIS AND TUMOR GROWTH hypersensitivity and Th2-driven allergic responses.372
It is likely IL-10 plays a role during the early phases of
Interleukin-8 has angiogenic properties and has been pleural inflammation, in mediating cell trafficking to the
shown to play a critical role in endotoxin-induced vascular pleura and vascular leak.373,374 Intrapleural IL-10 inhibits
permeability.355 IL-8 induced angiogenesis is likely to be the early phase of carrageenan-induced pleural inflamma-
important in tumor-related neovascularization. tion in a murine model.373 IL-10 null mice had enhanced
Interleukin-8 also possesses a direct growth-potentiat- rates of pleural exudation and polymorphonuclear cell
ing effect on certain tumors. In mesothelioma cell lines, trafficking; while mice injected with anti-IL-10 neutraliz-
IL-8 causes a dose-dependent increase in proliferating ing antibodies had increased leukocyte influx and vascular
activity, and can also function as an autocrine growth leakage in the early phase of carrageenan-induced pleural
factor in tumors. Concurrently, neutralization of IL-8 sig- inflammation.373
nificantly abrogates mesothelioma proliferation.363,364 Interleukin-10 is present in tuberculous and malignant
Similarly, an autocrine growth inhibitory circuit, involving pleural effusions. Its pleural fluid concentrations are signifi-
Immunomodulatory cytokines 87
cantly higher than the corresponding serum levels, suggest- mediated immunity, fail to develop granulomatous reac-
ing that local production is its principal source.375 Resident tions and are prone to develop TB.387 In TB patients, the
pleural macrophages may be a source of pleural IL-10, as percentage of T cells expressing IL-12 receptors is signifi-
macrophage ablation in a murine model of carrageenan cantly decreased, and IFN-g production by peripheral
pleurisy has demonstrated a reduced level of IL-10 as well monocytes is also reduced.388 In TB pleuritis, the mean
as other inflammatory cytokines.376 Alveolar macrophages IL-12 concentrations are tenfold higher in pleural fluid
isolated from older rats produce less IL-10 than those iso- than in serum.389 In vitro, pleural fluid cells have been
lated from young rats.377 These rats also exhibit significantly shown to proliferate and produce bioactive IL-12 when
reduced pleural fluid IL-10 levels during carrageenan- stimulated with M. tuberculosis.389 IL-12 in turn induces
induced pleurisy than younger rats. However, the effusion mononuclear cells in pleural effusions to increase their
IL-10 level does not correlate with lymphocyte subpopula- killing activity dose-dependently and enhances their pro-
tions in the pleural fluids or in blood378 and does not predict duction of IFN-g.390
survival in malignant effusions.375
Tumor cells can produce IL-10, which is known to sup-
press the production of anti-tumor cytokines (such as Pleural malignancy
TNF-a and IL-1b) by pleural macrophages.379 IL-10 is
reported to promote the growth of activated or neoplastic
Interleukin-12 is considered to have potent anti-tumor
B lymphocytes380 and its expression has been shown in
effects. Systemic administration of recombinant IL-12
high levels in cell lines of primary effusion lymphoma381
induces a significant and persistent anti-tumor immune
and in pyothorax-associated lymphoma.382
response, mediated by CD4 and CD8 T lymphocytes, and
In tuberculous effusions, IL-10 levels decrease over the
results in growth inhibition or even regression in a murine
course of the disease.383 IL-10 may be induced in the early
model of mesothelioma.391 IL-12 may also be useful as a
stages of chronic disease by Th1 type cytokine profiles, but
candidate for gene therapy in malignant pleural diseases.
its host-protective downregulatory effects on immune
Paracrine secretion of IL-12, generated by gene transfer,
response may allow the mycobacterium to persist.383 An
can induce immunity against mesothelioma locally and at
inverse correlation also exists between IL-10 and mononu-
a distant site, without causing significant systemic compli-
clear cell proliferation in tuberculous pleuritis384 and high
cations.392
IL-10 levels have been associated with pleural necrosis.
A Th2 to Th1 shift is observed when lymphocytes from
In murine models of Staphylococcus aureus empyema,
malignant effusions are treated with IL-12. The use of lym-
IFN-g levels rise in the controls, whereas the interleukin-
phocytes treated in this fashion as a novel adoptive
10 level increases significantly in the CD4 knockout mice.
immunotherapy for cancer has been suggested.393 Other
Therapeutic use of IL-10, especially in sepsis, has been
studies have reported that, upon IL-12/IL-2 co-stimulation,
explored in phase I and II trials.385 Further investigations on
monocytes isolated from malignant pleural effusions are
its potential use in pleural diseases would be worthwhile.
able to produce cytokines of both the Th1 and Th2 types.394
Interleukin-12 also serves to augment the anti-cancer
IL-12 effects of other cytokines. In vitro, IL-12, in the presence of
IL-2, can restore the cytolytic activity of the lymphocytes
in malignant pleural effusion against autologous tumor.284
Interleukin-12 is a heterodimeric cytokine that enhances
In vivo, co-administration of IL-15 with IL-12 activates the
cell-mediated and cytotoxic immune responses to intracel-
CD8+ T lymphocytes and NK cells, providing strong anti-
lular pathogens and tumors.210 IL-12 is produced prima-
tumor activity against malignant pleuritis in mice. The
rily by antigen presenting cells. It is considered crucial in
therapeutic effect is probably mainly due to enhanced
promoting Th1 responses and subsequent cell mediated
IFN-g production as the administration of anti-IFN-g
immunity. These functions are facilitated by the ability of
antibodies inhibits the beneficial effect of IL-15 and IL-
IL-12 to stimulate T lymphocyte and NK cell proliferation,
12.395 IL-12 has also been shown to promote pleural and
their cytotoxicity and IFN-g production. IL-12 receptors
blood monocyte cytotoxic activity against a small cell lung
comprise of two subunits IL-12R-b1 and -b2, which are
cancer cell line.394
expressed mainly on activated T and NK cells.386,387
Interleukin-12 has now been used in small clinical trials
against a variety of malignancies. Whether it has a therapeu-
TUBERCULOUS PLEURITIS tic value against pleural mesothelioma awaits investigation.
The role of IL-12 in TB pleuritis has been studied. IL-12
contributes to the anti-mycobacterial immune response by
enhancing production of IFN-g, facilitating the develop- Interferons
ment of Th1 cells and augmenting cytotoxicity of antigen-
specific T cells and NK cells. Knockout mice with There are two types of interferon: type I and type II. The
disrupted IL-12 or IL-12Rb1 genes have defective cell type I IFNs consist of seven classes, of which the
88 Immunology
IFN-a and IFN-b are the most studied. There is only one contact, contributing to regulation of the immune response
human IFN-b but there are multiple IFN-a species. Type toward the Th1 profile. Because CD56bright cells retain
II interferon consists of IFN-g only. In addition, four IFN- homing receptors for lymph nodes, they maintain their
like cytokines have been reported.396 Type I IFNs are capacity to migrate to them. Therefore, these findings may
secreted by virus infected cells while the type II IFN is explain why tuberculous pleurisy results in the clearance of
secreted mainly by T cells, NK cells and macrophages.396 M. tuberculosis and resolves without treatment.403
Toll-like receptors play an important role in the expression Polymorphisms within the IFN-g/IFN-g receptor (IFN-
of IFNs397 as well as monocyte-derived pro-inflammatory gR) complex have been identified which, depending on the
cytokines such as IL-12, IL-15 and IL-18, especially in polymorphism or combination of polymorphisms, can
combination.394,398,399 confer susceptibility or protection from disease.404–408
The interferons and IFN-like molecules signal through Recent studies on genetics and TB have also examined
the Jak-Stat pathway. The type I interferons all share the gene variants involved in iron acquisition. In one study
same receptor complex, whereas type II IFN-g binds to a examining the iron regulatory genes haptoglobin and
distinct receptor. The type I IFNs predominantly signal NRAMP1, common polymorphic variants showed func-
through Stat1 and Stat2 but IFN-a and IFN-b pathways tionally distinct biochemical phenotypes that would be
can involve Stat3, Stat4, Stat5 and IFN regulatory factors in predicted to influence the course of TB infection in
various cells under different conditions. IFN-g activates humans.409,410
Jak1, Jak2 and Stat1 that in turn induce genes containing Pleural fluid INF-g is important not only in the patho-
the g-activation sequence in the promoter. IFN-g can also genesis but also in the diagnosis of TB (see Chapter 27 on
activate Stat3 and Stat5. The IFNs can also activate other TB pleural effusions).
pathways including PI3K, Akt, NF-kB, MAPK and
others.396 PLEURAL INFLAMMATION
The type I interferons exhibit a wide array of biological
activities including antiviral, antiproliferative and cytotox- Interferon-g has anti-inflammatory properties and can
icity, on a wide variety of immune cells to increase the inhibit basal and IL-1b- and TNF-a-induced production
expression of tumor-associated antigens and other surface of IL-8. In addition, it can antagonize the polymorphonu-
molecules such as MHC class I antigens, activation of pro- clear cell influx induced by IL-1b.251 IFN-g also controls
apoptotic proteins, modulation of differentiation and polymorphonuclear cell infiltration and modulates IL-6
antiangiogenic activity.396 signaling through sIL-6R to promote their apoptosis and
clearance.411 Therefore, IFN-g plays a role in controlling
TUBERCULOUS PLEURITIS the phenotype of infiltrating leukocytes during inflamma-
tion through regulation of resident cell cytokine and
Tuberculosis pleural effusion is common (see Chapter 27) chemokine synthesis.
and the pleural fluid is exudative with a mononuclear cell
predominance. High levels of IFN-g are present in the MESOTHELIOMA
pleural fluid which is thought to be important in the
pathogenesis of TB.400 There is growing evidence that immunotherapy may be
Interferon-g stimulates mesothelial cells to secrete MIP- useful in the treatment of mesothelioma. Several studies
1a and MCP-1 which mediates recruitment of have shown that IFN-a and IFN-g inhibits the growth of
mononuclear cells to the pleural space.401 A positive corre- human mesothelioma cell lines which is further enhanced
lation has also been seen between IFN-g and the by the combination with other cytokines or chemothera-
proliferative response of mononuclear cells induced by M. peutic agents.412–416 A combination of IFN-a and IFN-g
tuberculosis.384 IFN-g-mediated pathways are extremely augments the response of mesothelioma cells to metho-
important in cell-mediated protective immunity against M. trexate by as much as 75 percent.413 Interestingly however,
tuberculosis antigen. The cell-mediated immune response the sensitivity to IFN-g alone was shown to be cell line
in TB originates predominantly from IFN-g-releasing CD4 dependent.413 The antiproliferative effect of IFN-g on
and CD8 effector T cells. This Th1 response helps to limit mesothelioma cell lines appears to be mediated through
mycobacterial replication and spread, however, it can also the JAK–STAT pathway.417 Failure of cells to respond to
lead to significant immunopathology. IFN-g also inhibits IFN-g was related to the limited transcriptional activity of
the proliferation of Th2 cells and thereby modulates Th2 STAT1, or a defect in JAK2 expression.417 Studies of the
mediated responses (see review402). NK cells extravazate to gene expression profiles of human mesothelioma cell lines
the site of TB infection. The CD56dimCD16+ subset has an in response to IFN-g treatment revealed many differences
increased susceptibility to undergo apoptosis induced by between IFN-g-resistant and -sensitive pathways, includ-
heat-stable/labile mediators present in tuberculous effu- ing features other than just the antiproliferative
sions, leading to an enrichment of CD56bright cells. These response.418
NK cells cannot lyse M. tuberculosis-infected target cells, Gene transfer of IFN-g or IFN-b into established
but are larger producers of IFN-g upon M. tuberculosis murine mesotheliomas has demonstrated significant
References 89
mesothelioma cell-line cells. Am J Respir Cell Mol Biol 1998; 19: growth factor (TGF)-b2, talc and bleomycin as pleurodesing agents
445–52. in sheep. Respirology 2002; 7: 209–16.
11. Rougier JP, Moullier P, Piedaguel R, Ronco PM. Hyperosmolality 31. Calabresi PA, Fields NS, Maloni HW, et al. Phase I trial of
suppresses but TGF beta 1 increases MMP9 in human peritoneal transforming growth factor beta 2 in chronic progressive MS.
mesothelial cells. Kidney Int 1997; 51: 337–47. Neurology 1998; 51: 289–92.
12. Ma C, Tarnuzzer RW, Chegini N. Expression of matrix 32. Lee YCG, Lane KB, Zoia O, et al. Transforming growth factor-beta
metalloproteinases and tissue inhibitors of matrix induces collagen synthesis without inducing IL-8 production in
metalloproteinases in mesothelial cells and their regulation by mesothelial cells. Eur Respir J 2003; 22: 197–202.
transforming growth factor-beta 1. Wound Repair Regen 1999; 7: 33. Marchi E, Vargas FS, Acencio MM, et al. Evidence that mesothelial
477–85. cells regulate the acute inflammatory response in talc pleurodesis.
13. Idell S, Zweib C, Kumar A, Koenig KB, Johnson AR. Pathways of Eur Respir J 2006; 28: 929–32.
fibrin turnover of human pleural mesothelial cells in vitro. Am J 34. Teixeira LR, Wu W, Cheng DS, Light RW. The effect of
Respir Cell Mol Biol 1992; 7: 414–26. corticosteroids on pleurodesis induced by doxycycline in rabbits.
14. Falk P, Ma C, Chegini N, Holmdahl L. Differential regulation of Chest 2002; 121: 216–9.
mesothelial cells fibrinolysis by transforming growth factor beta 1. 35. Teixeira LR, Vargas FS, Acencio MM, et al. Influence of
Scan J Clin Lab Invest 2000; 60: 439–47. antiinflammatory drugs (methylprednisolone and diclofenac
15. Liu F, Liu H, Peng Y, et al. Inhibition of transforming growth factor sodium) on experimental pleurodesis induced by silver nitrate or
beta (TGFbeta1) expression and extracellular matrix secretion in talc. Chest 2005; 128: 4041–5.
human peritoneal mesothelial cells by pcDU6 vector-mediated 36. Kelley J. Transforming growth factor-beta. In: Kelley J (ed.).
TGFbeta1 shRNA and by pcDNA3.1(-)-mediated antisense Cytokines of the lung. New York: Marcel Dekker, 1993: 101–37.
TGFbeta1 RNA. Adv Petit Dial 2005; 21: 41–52. 37. Link J, He B, Navikas V, et al. Transforming growth factor-beta 1
16. Wang Q, Wang Y, Hyde DM, et al. Reduction of bleomycin induced suppresses autoantigen-induced expression of pro-inflammatory
lung fibrosis by transforming growth factor soluble receptor in cytokines but not of interleukin-10 in multiple sclerosis and
hamsters. Thorax 1999; 54: 805–12. myasthenia gravis. J Neuroimmunol 1995; 58: 21–35.
17. Border WA, Okuda S, Languino LR, Sporn MB, Ruoslahti E. 38. Zissel G, Schlaak J, Schlaak M, Muller-Quernheim J. Regulation of
Suppression of experimental glomerulonephritis by antiserum cytokine release by alveolar macrophages treated with interleukin-
against transforming growth factor b1. Nature 1990; 346: 371–4. 4, interleukin-10, or transforming growth factor beta. Eur Cytokine
18. Shah M, Foreman DM, Ferguson MW. Control of scarring in adult Netw 1996; 7: 59–66.
wounds by neutralising antibody to transforming growth factor b. 39. Karres I, Kremer JP, Steckholzer U, Kenney JS, Ertel W.
Lancet 1992; 339: 213–4. Transforming growth factor-beta 1 inhibits synthesis of cytokines
19. Lucas PA, Warejcka DJ, Young HE, Lee BY. Formation of abdominal in endotoxin-stimulated human whole blood. Arch Surg 1996;
adhesions is inhibited by antibodies to transforming growth 131: 1310–6.
factor-beta 1. J Surg Res 1996; 65: 135–8. 40. Smith WB, Noack L, Khew-Goddall Y, et al. Transforming growth
20. Nakao A, Fujii M, Matsumura R, et al. Transient gene transfer and factor-beta 1 inhibits the production of IL-8 and the
expression of Smad 7 prevents bleomycin-induced lung fibrosis in transmigration of neutrophils through activated endothelium. J
mice. J Clin Invest 1999; 104: 5–11. Immunol 1996; 157: 360–8.
●21. Kunz CR, Jadus MR, Kukes GD, et al. Intrapleural injection of 41. Diebold RJ, Eis MJ, Yin M, et al. Early-onset multifocal inflammation
transforming growth factor-b antibody inhibits pleural fibrosis in in the transforming growth factor beta 1-null mouse is lymphocyte
empyema. Chest 2004; 126: 1636–44. mediated. Proc Natl Acad Sci U S A 1995; 92: 12215–9.
●22. Allen SS, Cassone L, Lasco TM, McMurray DN. Effect of 42. Letterio JJ, Roberts AB. Regulation of immune responses by TGF-b.
neutralizing transforming growth factor beta 1 on the immune Annu Rev Immunol 1998; 16: 137–61.
response against mycobacterium tuberculosis in guinea pigs. 43. Chen W, Wahl SM. Manipulation of TGF-b to control autoimmune
Infect Immun 2004; 72: 1358–63. and chronic inflammatory diseases. Microbes Infect 1999; 1:
●23. Light RW, Cheng DS, Lee YCG, et al. A single intrapleural injection 1367–80.
of transforming growth factor beta-2 produces an excellent 44. Gamble JR, Khew-Goodall Y, Vadas MA. Transforming growth
pleurodesis in rabbits. Am J Respir Crit Care Med 2000; 162: factor-b inhibits E-selectin expression on human endothelial cells.
98–104. J Immunol 1993; 150: 4494–503.
24. Lee YCG, Teixeira LR, Devin CJ, et al. Transforming growth factor- 45. Denk PO, Roth-Eichhorn S, Gressner AM, Knorr M. Effect of
b2 induces pleurodesis significantly faster than talc. Am J Respir cytokines on regulation of the production of transforming growth
Crit Care Med 2001; 163: 640–4. factor beta-1 in cultured human Tenon’s capsule fibroblasts. Eur J
25. Lee YCG, Devin CJ, Teixiera LR, et al. Transforming growth factor Ophthalmol 2000; 10: 110–5.
beta-2 induced pleurodesis is not inhibited by corticosteroids. 46. Offner FA, Feichtinger H, Stadlmann S, et al. Transforming growth
Thorax 2001; 56: 643–8. factor-beta synthesis by human peritoneal mesothelial cells.
26. Lee YCG, Lane KB, Parker RE, et al. Transforming growth factor Induction by interleukin-1. Am J Pathol 1996; 148: 1679–88.
beta-2 (TGFb2) produces effective pleurodesis in sheep with no 47. Owens MW, Milligan SA, Grisham MB. Inhibition of rat pleural
systemic complications. Thorax 2000; 55: 1058–62. mesothelial cell nitric oxide synthesis by transforming growth
27. Kalomenidis I, Guo Y, Lane KB, Hawthorne M, Light RW. factor-beta 1. Inflammation 1996; 20: 637–46.
Transforming growth factor-beta3 induces pleurodesis in rabbits 48. Blobe GC, Schiemann WP, Lodish HF. Role of transforming growth
and collagen production of human mesothelial cells. Chest 2005; factor (beta) in human disease. N Engl J Med 2000; 342: 1350–8.
127: 1335–40. 49. Ikubo A, Morisaki T, Katano M, et al. A possible role of TGF-beta in
28. Lee YCG, Malkerneker D, Devin CJ, et al. Comparing transforming the formation of malignant effusions. Clin Immunol
growth factor beta-2 and fibronectin as pleurodesing agents. Immunopathol 1995; 77: 27–32.
Respirology 2001; 6: 281–6. ◆50. Stover DG, Bierie B, Moses HL. A delicate balance: TGF-beta and
29. Lee YCG, Malkerneker D, Thompson PJ, Light RW, Lane KB. the tumor microenvironment. J Cell Biochem 2007; 101: 851–61.
Transforming growth factor-b induces vascular endothelial growth 51. Marzo AL, Fitzpatrick DR, Robinson BW, Scott B. Antisense
factor elaboration from pleural mesothelial cells in vivo and in oligonucleotides specific for transforming growth factor beta2
vitro. Am J Respir Crit Care Med 2002; 165: 88–94. inhibit the growth of malignant mesothelioma both in vitro and in
30. Lee YCG, Yasay JR, Johnson JE, et al. Comparing transforming vivo. Cancer Res 1997; 57: 3200–7.
References 91
●52. Suzuki E, Kim S, Cheung HK, et al. A novel small-molecule mesothelial cells in pleural lavage fluid after intratracheal
inhibitor of transforming growth factor beta type 1 receptor asbestos. Exp Lung Res 2001; 27: 605–16.
kinase (SM16) inhibits murine mesothelioma tumor growth in vivo 74. Dworkin LD, Gong R, Tolbert E, et al. Hepatocyte growth factor
and prevents tumor recurrence after surgical resection. Cancer Res ameliorates progression of interstitial fibrosis in rats with
2007; 67: 2351–9. established renal injury. Kidney Int 2004; 65: 409–19.
53. Cambien F, Ricard S, Troesch A, et al. Polymorphisms of the 75. Hattori N, Mizuno S, Yoshida Y, et al. The plasminogen activation
transforming growth factor-beta 1 gene in relation to myocardial system reduces fibrosis in the lung by a hepatocyte growth factor-
infarction and blood pressure. The Etude Cas-Temon de l’Infarctus dependent mechanism. Am J Pathol 2004; 164: 1091–8.
du Myocarde (ECTIM) study. Hypertension 1996; 28: 881–7. 76. Matsuda Y, Matsumoto K, Yamada A, et al. Preventive and
54. Awad MR, El-Gamel A, Hasleton P, et al. Genotype variation in the therapeutic effects in rats of hepatocyte growth factor infusion on
transforming growth factor-beta 1 gene. Transplantation 1998; liver fibrosis/cirrhosis. Hepatology 1997; 26: 81–9.
66: 1014–20. 77. Matsuo K, Maeda Y, Naiki Y, et al. Possible effects of hepatocyte
55. El-Gamel A, Awad MR, Hasleton PS, et al. Transforming growth growth factor for the prevention of peritoneal fibrosis. Nephron
factor-beta (TGF-beta1) genotype and lung allograft fibrosis. J Exp Nephrol 2005; 99: e87–94.
Heart Lung Transplant 1999; 18: 517–23. 78. Mizuno S, Kurosawa T, Matsumoto K, et al. Hepatocyte growth
56. Stella MC, Comoglio PM. HGF: a multifunctional growth factor factor prevents renal fibrosis and dysfunction in a mouse model of
controlling cell scattering. Int J Biochem Cell Biol 1999; 31: 1357–62. chronic renal disease. J Clin Invest 1998; 101: 1827–34.
57. Weidner KM, Arakaki N, Hartmann G, et al. Evidence for the 79. Mizuno S, Matsumoto K, Li MY, Nakamura T. HGF reduces
identity of human scatter factor and human hepatocyte growth advancing lung fibrosis in mice: a potential role for MMP-
factor. Proc Natl Acad Sci U S A 1991; 88: 7001–5. dependent myofibroblast apoptosis. FASEB J 2005; 19: 580–2.
58. Naldini L, Tamagnone L, Vigna E, et al. Extracellular proteolytic 80. Watanabe M, Ebina M, Orson FM, et al. Hepatocyte growth factor
cleavage by urokinase is required for activation of hepatocyte gene transfer to alveolar septa for effective suppression of lung
growth factor/scatter factor. Embo J 1992; 11: 4825–33. fibrosis. Mol Ther 2005; 12: 58–67.
59. Warn R, Harvey P, Warn A, et al. HGF/SF induces mesothelial cell 81. Yaekashiwa M, Nakayama S, Ohnuma K, et al. Simultaneous or
migration and proliferation by autocrine and paracrine pathways. delayed administration of hepatocyte growth factor equally
Exp Cell Res 2001; 267: 258–66. represses the fibrotic changes in murine lung injury induced by
60. Jiang W, Hiscox S, Matsumoto K, Nakamura T. Hepatocyte growth bleomycin. A morphologic study. Am J Respir Crit Care Med 1997;
factor/scatter factor, its molecular, cellular and clinical 156: 1937–44.
implications in cancer. Crit Rev Oncol Hematol 1999; 29: 209–48. 82. Gong R, Rifai A, Tolbert EM, et al. Hepatocyte growth factor
61. Matsumoto K, Tajima H, Okazaki H, Nakamura T. Negative ameliorates renal interstitial inflammation in rat remnant kidney
regulation of hepatocyte growth factor gene expression in human by modulating tubular expression of macrophage chemoattractant
lung fibroblasts and leukemic cells by transforming growth factor- protein-1 and RANTES. J Am Soc Nephrol 2004; 15: 2868–81.
beta 1 and glucocorticoids. J Biol Chem 1992; 267: 24917–20. 83. Liu Y, Rajur K, Tolbert E, Dworkin LD. Endogenous hepatocyte
●62. Bottaro DP, Rubin JS, Faletto DL, et al. Identification of the growth factor ameliorates chronic renal injury by activating
hepatocyte growth factor receptor as the c-met proto-oncogene matrix degradation pathways. Kidney Int 2000; 58: 2028–43.
product. Science 1991; 251: 802–4. 84. Gong R, Rifai A, Tolbert EM, Centracchio JN, Dworkin LD.
63. Funakoshi H, Nakamura T. Hepatocyte growth factor: from Hepatocyte growth factor modulates matrix metalloproteinases
diagnosis to clinical applications. Clin Chim Acta 2003; 327: and plasminogen activator/plasmin proteolytic pathways in
1–23. progressive renal interstitial fibrosis. J Am Soc Nephrol 2003; 14:
64. Bladt F, Riethmacher D, Isenmann S, Aguzzi A, Birchmeier C. 3047–60.
Essential role for the c-met receptor in the migration of myogenic 85. Harvey P, Warn A, Dobbin S, et al. Expression of HGF/SF in
precursor cells into the limb bud. Nature 1995; 376: 768–71. mesothelioma cell lines and its effects on cell motility,
65. Maina F, Hilton MC, Andres R, et al. Multiple roles for hepatocyte proliferation and morphology. Br J Cancer 1998; 77: 1052–9.
growth factor in sympathetic neuron development. Neuron 1998; 86. Harvey P, Warn A, Newman P, et al. Immunoreactivity for
20: 835–46. hepatocyte growth factor/scatter factor and its receptor, met, in
66. Schmidt C, Bladt F, Goedecke S, et al. Scatter factor/hepatocyte human lung carcinomas and malignant mesotheliomas. J Pathol
growth factor is essential for liver development. Nature 1995; 1996; 180: 389–94.
373: 699–702. 87. Lengyel E, Prechtel D, Resau JH, et al. C-Met overexpression in
67. Uehara Y, Minowa O, Mori C, et al. Placental defect and embryonic node-positive breast cancer identifies patients with poor clinical
lethality in mice lacking hepatocyte growth factor/scatter factor. outcome independent of Her2/neu. Int J Cancer 2005; 113:
Nature 1995; 373: 702–5. 678–82.
68. Ohmichi H, Koshimizu U, Matsumoto K, Nakamura T. Hepatocyte 88. Tolnay E, Kuhnen C, Wiethege T, et al. Hepatocyte growth
growth factor (HGF) acts as a mesenchyme-derived morphogenic factor/scatter factor and its receptor c-Met are overexpressed
factor during fetal lung development. Development 1998; 125: and associated with an increased microvessel density in
1315–24. malignant pleural mesothelioma. J Cancer Res Clin Oncol 1998;
69. Padela S, Cabacungan J, Shek S, et al. Hepatocyte growth factor is 124: 291–6.
required for alveologenesis in the neonatal rat. Am J Respir Crit 89. Takayama H, LaRochelle WJ, Sharp R, et al. Diverse tumorigenesis
Care Med 2005; 172: 907–14. associated with aberrant development in mice overexpressing
70. Ito W, Kanehiro A, Matsumoto K, et al. Hepatocyte growth factor hepatocyte growth factor/scatter factor. Proc Natl Acad Sci U S A
attenuates airway hyperresponsiveness, inflammation, and 1997; 94: 701–6.
remodeling. Am J Respir Cell Mol Biol 2005; 32: 268–80. 90. Cacciotti P, Libener R, Betta P, et al. SV40 replication in human
71. Okunishi K, Dohi M, Nakagome K, et al. A novel role of hepatocyte mesothelial cells induces HGF/Met receptor activation: a model for
growth factor as an immune regulator through suppressing viral-related carcinogenesis of human malignant mesothelioma.
dendritic cell function. J Immunol 2005; 175: 4745–53. Proc Natl Acad Sci U S A 2001; 98: 12032–7.
72. Rampino T, Cancarini G, Gregorini M, et al. Hepatocyte growth 91. Altomare DA, You H, Xiao GH, et al. Human and mouse
factor/scatter factor released during peritonitis is active on mesotheliomas exhibit elevated AKT/PKB activity, which can be
mesothelial cells. Am J Pathol 2001; 159: 1275–85. targeted pharmacologically to inhibit tumor cell growth.
73. Adamson IY, Bakowska J. KGF and HGF are growth factors for Oncogene 2005; 24: 6080–9.
92 Immunology
92. Harvey P, Clark IM, Jaurand MC, Warn RM, Edwards DR. 113. Zanella CL, Timblin CR, Cummins A, et al. Asbestos-induced
Hepatocyte growth factor/scatter factor enhances the invasion of phosphorylation of epidermal growth factor receptor is linked to
mesothelioma cell lines and the expression of matrix c-fos and apoptosis. Am J Physiol 1999; 277(4 Pt 1): L684–L93.
metalloproteinases. Br J Cancer 2000; 83: 1147–53. 114. Lahat N, Froom P, Kristal-Boneh E, et al. Increased serum
93. Uchiyama A, Morisaki T, Beppu K, et al. Hepatocyte growth factor concentrations of growth factor receptors and Neu in workers
and invasion-stimulatory activity are induced in pleural fluid by previously exposed to asbestos. Occup Environ Med 1999; 56:
surgery in lung cancer patients. Br J Cancer 1999; 81: 721–6. 114–7.
94. Yashiro M, Chung YS, Inoue T, et al. Hepatocyte growth factor 115. Stanton MF, Layard M, Tegeris A, et al. Relation of particle
(HGF) produced by peritoneal fibroblasts may affect mesothelial dimension to carcinogenicity in amphibole asbestoses and other
cell morphology and promote peritoneal dissemination. Int J fibrous minerals. J Natl Cancer Inst 1981; 67: 965–75.
Cancer 1996; 67: 289–93. 116. Heintz NH, Janssen YM, Mossman BT. Persistent induction of c-fos
95. Saimura M, Nagai E, Mizumoto K, et al. Intraperitoneal injection of and c-jun expression by asbestos. Proc Natl Acad Sci U S A 1993;
adenovirus-mediated NK4 gene suppresses peritoneal 90: 3299–303.
dissemination of pancreatic cancer cell line AsPC-1 in nude mice. 117. Janssen YM, Heintz NH, Mossman BT. Induction of c-fos and c-jun
Cancer Gene Ther 2002; 9: 799–806. proto-oncogene expression by asbestos is ameliorated by N-
96. Tanaka T, Shimura H, Sasaki T, et al. Gallbladder cancer treatment acetyl-L-cysteine in mesothelial cells. Cancer Res 1995; 55:
using adenovirus expressing the HGF/NK4 gene in a peritoneal 2085–9.
implantation model. Cancer Gene Ther 2004; 11: 431–40. 118. Zanella CL, Posada J, Tritton TR, Mossman BT. Asbestos causes
97. Xian CJ. Roles of epidermal growth factor family in the regulation stimulation of the extracellular signal-related kinase 1 mitogen-
of postnatal somatic growth. Endocr Rev 2007; 28: 284–96. activated protein kinase cascade after phosphorylation of the
98. Carpenter G. The EGF receptor: a nexus for trafficking and epidermal growth factor receptor. Cancer Res 1996; 56: 5334–8.
signaling. Bioessays 2000; 22: 697–707. 119. Dazzi H, Hasleton PS, Thatcher N, et al. Malignant pleural
99. Palmer U, Liu Z, Broome U, Klominek J. Epidermal growth factor mesothelioma and epidermal growth factor receptor (EGF-R).
receptor ligands are chemoattractants for normal human Relationship of EGF-R with histology and survival using fixed
mesothelial cells. Eur Respir J 1999; 14: 405–11. paraffin embedded tissue and the F4, monoclonal antibody. Br J
100. Gabrielson EW, Gerwin BI, Harris CC, et al. Stimulation of DNA Cancer 1990; 61: 924–6.
synthesis in cultured primary human mesothelial cells by specific 120. Destro A, Ceresoli GL, Falleni M, et al. EGFR overexpression in
growth factors. FASEB J 1988; 2: 2717–21. malignant pleural mesothelioma. An immunohistochemical and
101. Goodglick LA, Vaslet CA, Messier NJ, Kane AB. Growth factor molecular study with clinico-pathological correlations. Lung
responses and protooncogene expression of murine mesothelial Cancer 2006; 51: 207–15.
cell lines derived from asbestos-induced mesotheliomas. Toxicol 121. Edwards JG, Swinson DE, Jones JL, Waller DA, O’Byrne KJ. EGFR
Pathol 1997; 25: 565–73. expression: associations with outcome and clinicopathological
102. Owens MW, Milligan SA. Growth factor modulation of rat pleural variables in malignant pleural mesothelioma. Lung Cancer 2006;
mesothelial cell mitogenesis and collagen synthesis. Effects of 54: 399–407.
epidermal growth factor and platelet-derived factor. Inflammation 122. Govindan R, Kratzke RA, Herndon JE, 2nd, et al. Gefitinib in
1994; 18: 77–87. patients with malignant mesothelioma: a phase II study by the
103. Pache JC, Janssen YM, Walsh ES, et al. Increased epidermal growth Cancer and Leukemia Group B. Clin Cancer Res 2005; 11:
factor-receptor protein in a human mesothelial cell line in 2300–4.
response to long asbestos fibers. Am J Pathol 1998; 152: 333–40. 123. Ramael M, Segers K, Buysse C, Van den Bossche J, Van Marck E.
104. van Rossen ME, Hofland LJ, van den Tol MP, et al. Effect of Immunohistochemical distribution patterns of epidermal growth
inflammatory cytokines and growth factors on tumour cell factor receptor in malignant mesothelioma and non-neoplastic
adhesion to the peritoneum. J Pathol 2001; 193: 530–7. mesothelium. Virchows Arch A Pathol Anat Histopathol 1991; 419:
105. Mutsaers SE, McAnulty RJ, Laurent GJ, et al. Cytokine regulation 171–5.
of mesothelial cell proliferation in vitro and in vivo. Eur J Cell Biol 124. Liu Z, Klominek J. Regulation of matrix metalloprotease activity in
1997; 72: 24–9. malignant mesothelioma cell lines by growth factors. Thorax 2003;
106. Jayne DG, Perry SL, Morrison E, Farmery SM, Guillou PJ. Activated 58: 198–203.
mesothelial cells produce heparin-binding growth factors: 125. Cole GW Jr, Alleva AM, Reddy RM, et al. The selective epidermal
implications for tumour metastases. Br J Cancer 2000; 82: 1233–8. growth factor receptor tyrosine kinase inhibitor PD153035
107. Faull RJ, Stanley JM, Fraser S, Power DA, Leavesley DI. HB-EGF is suppresses expression of prometastasis phenotypes in malignant
produced in the peritoneal cavity and enhances mesothelial cell pleural mesothelioma cells in vitro. J Thorac Cardiovasc Surg 2005;
adhesion and migration. Kidney Int 2001; 59: 614–24. 129: 1010–7.
108. Muller J, Yoshida T. Interaction of murine peritoneal leukocytes 126. Janne PA, Taffaro ML, Salgia R, Johnson BE. Inhibition of
and mesothelial cells: in vitro model system to survey cellular epidermal growth factor receptor signaling in malignant pleural
events on serosal membranes during inflammation. Clin Immunol mesothelioma. Cancer Res 2002; 62: 5242–7.
Immunopathol 1995; 75: 231–8. 127. She Y, Lee F, Chen J, et al. The epidermal growth factor receptor
109. Owens MW, Milligan SA, Grisham MB. Nitric oxide synthesis by rat tyrosine kinase inhibitor ZD1839 selectively potentiates radiation
pleural mesothelial cells: Induction by growth factors and response of human tumors in nude mice, with a marked
lipopolysaccharide. Exp Lung Res 1995; 21: 731–42. improvement in therapeutic index. Clin Cancer Res 2003; 9(10 Pt
110. Ohmura E, Tsushima T, Kamiya Y, et al. Epidermal growth factor 1): 3773–8.
and transforming growth factor alpha induce ascitic fluid in mice. 128. Cortese JF, Gowda AL, Wali A, et al. Common EGFR mutations
Cancer Res 1990; 15: 4915–7. conferring sensitivity to gefitinib in lung adenocarcinoma are not
111. Liaw YS, Yang PC, Yu CJ, et al. PKC activation is required by EGF- prevalent in human malignant mesothelioma. Int J Cancer 2006;
stimulated Na(+)-H+ exchanger in human pleural mesothelial 118: 521–2.
cells. Am J Physiol 1998; 274(5 Pt 1): L665–L72. 129. Garland LL, Rankin C, Gandara DR, et al. Phase II study of erlotinib
112. Faux SP, Houghton CE, Hubbard A, Patrick G. Increased expression in patients with malignant pleural mesothelioma: a Southwest
of epidermal growth factor receptor in rat pleural mesothelial cells Oncology Group Study. J Clin Oncol 2007; 25: 2406–13.
correlates with carcinogenicity of mineral fibres. Carcinogenesis 130. Cesario A, Catassi A, Festi L, et al. Farnesyltransferase inhibitors
2000; 21: 2275–80. and human malignant pleural mesothelioma: a first-step
References 93
comparative translational study. Clin Cancer Res 2005; 11: 151. Filiberti R, Marroni P, Neri M, et al. Serum PDGF-AB in pleural
2026–37. mesothelioma. Tumour Biol 2005; 26: 221–6.
131. Athanassiadou P, Athanassiades P, Kyrkou K, et al. Expression of 152. Gerwin BI, Lechner JF, Reddel RR, et al. Comparison of production
vimentin and epidermal growth factor receptor in effusions from of transforming growth factor-beta and platelet-derived growth
patients with breast cancer; correlation with oestrogen and factor by normal human mesothelial cells and mesothelioma cell
progesterone receptor status. Cytopathology 1993; 4: 91–8. lines. Cancer Res 1987; 47: 6180–4.
132. Abramov Y, Anteby SO, Fasouliotis SJ, Barak V. Markedly elevated 153. Versnel MA, Hagemeijer A, Bouts MJ, van der Kwast TH,
levels of vascular endothelial growth factor, fibroblast growth Hoogsteden HC. Expression of c-sis (PDGF B-chain) and PDGF A-
factor, and interleukin 6 in Meigs syndrome. Am J Obstet Gynecol chain genes in ten human malignant mesothelioma cell lines
2001; 184: 354–5. derived from primary and metastatic tumors. Oncogene 1988; 2:
133. Strizzi L, Vianale G, Catalano A, et al. Basic fibroblast growth 601–5.
factor in mesothelioma pleural effusions: correlation with patient 154. Van der Meeren A, Seddon MB, Betsholtz CA, Lechner JF, Gerwin
survival and angiogenesis. Int J Oncol 2001; 18: 1093–8. BI. Tumorigenic conversion of human mesothelial cells as a
134. Cronauer MV, Stadlmann S, Klocker H, et al. Basic fibroblast consequence of platelet-derived growth factor-A chain
growth factor synthesis by human peritoneal mesothelial cells: overexpression. Am J Respir Cell Mol Biol 1993; 8: 214–21.
induction by interleukin-1. Am J Pathol 1999; 155: 1977–84. 155. Metheny-Barlow LJ, Flynn B, van Gijssel HE, Marrogi A, Gerwin BI.
135. Antony VB, Nasreen N, Mohammed KA, et al. Talc pleurodesis: Paradoxical effects of platelet-derived growth factor-A
basic fibroblast growth factor mediates pleural fibrosis. Chest overexpression in malignant mesothelioma. Antiproliferative
2004; 126: 1522–8. effects in vitro and tumorigenic stimulation in vivo. Am J Respir
136. Davidson B, Vintman L, Zcharia E, et al. Heparanase and basic Cell Mol Biol 2001; 24: 694–702.
fibroblast growth factor are co-expressed in malignant 156. Ramael M, Buysse C, van den Bossche J, Segers K, van Marck E.
mesothelioma. Clin Exp Metastasis 2004; 21: 469–76. Immunoreactivity for the beta chain of the platelet-derived
137. Stapelberg M, Gellert N, Swettenham E, et al. Alpha-tocopheryl growth factor receptor in malignant mesothelioma and non-
succinate inhibits malignant mesothelioma by disrupting the neoplastic mesothelium. J Pathol 1992; 167: 1–4.
fibroblast growth factor autocrine loop: mechanism and the role 157. Langerak AW, De Laat PA, Van Der Linden-Van Beurden CA, et al.
of oxidative stress. J Biol Chem 2005; 280: 25369–76. Expression of platelet-derived growth factor (PDGF) and PDGF
138. Tzanakakis GN, Hjerpe A, Karamanos NK. Proteoglycan synthesis receptors in human malignant mesothelioma in vitro and in vivo.
induced by transforming and basic fibroblast growth factors in J Pathol 1996; 178: 151–60.
human malignant mesothelioma is mediated through specific 158. Langerak AW, van der Linden-van Beurden CA, Versnel MA.
receptors and the tyrosine kinase intracellular pathway. Biochimie Regulation of differential expression of platelet-derived growth
1997; 79: 323–32. factor alpha- and beta-receptor mRNA in normal and malignant
139. Bergsten E, Uutela M, Li X, et al. PDGF-D is a specific, protease- human mesothelial cell lines. Biochim Biophys Acta 1996; 1305:
activated ligand for the PDGF beta-receptor. Nat Cell Biol 2001; 3: 63–70.
512–6. 159. Versnel MA, Claesson-Welsh L, Hammacher A, et al. Human
140. Fredriksson L, Li H, Eriksson U. The PDGF family: four gene malignant mesothelioma cell lines express PDGF beta-receptors
products form five dimeric isoforms. Cytokine Growth Factor Rev whereas cultured normal mesothelial cells express predominantly
2004; 15: 197–204. PDGF alpha-receptors. Oncogene 1991; 6: 2005–11.
141. Li X, Ponten A, Aase K, et al. PDGF-C is a new protease-activated 160. Ascoli V, Scalzo CC, Facciolo F, Nardi F. Platelet-derived growth
ligand for the PDGF alpha-receptor. Nat Cell Biol 2000; 2: 302–9. factor receptor immunoreactivity in mesothelioma and
142. Li M, Jendrossek V, Belka C. The role of PDGF in radiation nonneoplastic mesothelial cells in serous effusions. Acta Cytol
oncology. Radiat Oncol 2007; 2: 5. 1995; 39: 613–22.
143. Adamson IYR, Prieditis H, Young L. Lung mesothelial cell and 161. Klominek J, Baskin B, Hauzenberger D. Platelet-derived growth
fibroblast responses to pleural and alveolar macrophage factor (PDGF) BB acts as a chemoattractant for human malignant
supernatants and to lavage fluids from crocidolite-exposed rats. mesothelioma cells via PDGF receptor beta-integrin alpha3beta1
Am J Respir Cell Mol Biol 1997; 16: 650–6. interaction. Clin Exp Metastasis 1998; 16: 529–39.
144. Deuel TF, Senior RM, Huang JS, Griffin GL. Chemotaxis of 162. Safi A, Sadmi M, Martinet N, et al. Presence of elevated levels of
monocytes and neutrophils to platelet-derived growth factor. J platelet-derived growth factor (PDGF) in lung adenocarcinoma
Clin Invest 1982; 69: 1046–9. pleural effusions. Chest 1992; 102: 204–7.
145. Tzeng DY, Deuel TF, Huang JS, Baehner RL. Platelet-derived growth 163. Hung KY, Chen CT, Yen CJ, et al. Dipyridamole inhibits PDGF-
factor promotes human peripheral monocyte activation. Blood stimulated human peritoneal mesothelial cell proliferation. Kidney
1985; 66: 179–83. Int 2001; 60: 872–81.
146. Blatti SP, Foster DN, Ranganathan G, Moses HL, Getz MJ. 164. Bertino P, Porta C, Barbone D, et al. Preliminary data suggestive of
Induction of fibronectin gene transcription and mRNA is a primary a novel translational approach to mesothelioma therapy: Imatinib
response to growth-factor stimulation of AKR-2B cells. Proc Natl mesylate with gemcitabine or pemetrexed. Thorax 2007; 62:
Acad Sci U S A 1988; 85: 1119–23. 690–5..
147. Heldin P, Asplund T, Ytterberg D, Thelin S, Laurent TC. 165. Mathy A, Baas P, Dalesio O, van Zandwijk N. Limited efficacy of
Characterization of the molecular mechanism involved in the imatinib mesylate in malignant mesothelioma: a phase II trial.
activation of hyaluronan synthetase by platelet-derived growth Lung Cancer 2005; 50: 83–6.
factor in human mesothelial cells. Biochem J 1992; 283: 165–70. 166. Board R, Jayson GC. Platelet-derived growth factor receptor
148. Butt RP, Laurent GJ, Bishop JE. Collagen production and (PDGFR): a target for anticancer therapeutics. Drug Resist Updat
replication by cardiac fibroblasts is enhanced in response to 2005; 8:75–83.
diverse classes of growth factors. Eur J Cell Biol 1995; 68: 330–5. 167. Duan C. Specifying the cellular responses to IGF signals: roles of
149. Bauer EA, Cooper TW, Huang JS, Altman J, Deuel TF. Stimulation of IGF-binding proteins. J Endocrinol 2002; 175: 41–54.
in vitro human skin collagenase expression by platelet-derived 168. Laviola L, Natalicchio A, Giorgino F. The IGF-I signaling pathway.
growth factor. Proc Natl Acad Sci U S A 1985; 82: 4132–6. Curr Pharm Des 2007; 13: 663–9.
150. Pierce GF, Mustoe TA, Lingelbach J, et al. Platelet-derived growth 169. Hawkes C, Kar S. The insulin-like growth factor-II/mannose-6-
factor and transforming growth factor-beta enhance tissue repair phosphate receptor: structure, distribution and function in the
activities by unique mechanisms. J Cell Biol 1989; 109: 429–40. central nervous system. Brain Res Brain Res Rev 2004; 44: 117–40.
94 Immunology
170. Hoang CD, Zhang X, Scott PD, et al. Selective activation of insulin 192. Cheng DS, Lee YC, Rogers JT, et al. Vascular endothelial growth
receptor substrate-1 and -2 in pleural mesothelioma cells: factor level correlates with transforming growth factor beta
association with distinct malignant phenotypes. Cancer Res 2004; isoform levels in pleural effusions. Chest 2000; 118: 1747–53.
64: 7479–85. 193. Yanagawa H, Takeuchi E, Suzuki Y, et al. Vascular endothelial
171. Lee TC, Zhang Y, Aston C, et al. Normal human mesothelial cells growth factor in malignant pleural effusion associated with lung
and mesothelioma cell lines express insulin-like growth factor I cancer. Cancer Immunol Immunother 1999; 48: 396–400.
and associated molecules. Cancer Res 1993; 53: 2858–64. 194. Ishimoto O, Saijo Y, Narumi K, et al. High level of vascular
172. Porcu P, Ferber A, Pietrzkowski Z, et al. The growth-stimulatory endothelial growth factor in hemorrhagic pleural effusion of
effect of simian virus 40 T antigen requires the interaction of cancer. Oncology 2002; 63: 70–5.
insulinlike growth factor 1 with its receptor. Mol Cell Biol 1992; 195. Ishii H, Yazawa T, Sato H, et al. Enhancement of pleural
12: 5069–77. dissemination and lymph node metastasis of intrathoracic lung
173. Pass HI, Mew DJ, Carbone M, et al. The effect of an antisense cancer cells by vascular endothelial growth factors (VEGFs). Lung
expression plasmid to the IGF-1 receptor on hamster Cancer 2004; 45: 325–37.
mesothelioma proliferation. Dev Biol Stand 1998; 94: 321–8. 196. Kraft A, Weindel K, Ochs A, et al. Vascular endothelial growth
174. Whitson BA, Jacobson BA, Frizelle S, et al. Effects of insulin-like factor in the sera and effusions of patients with malignant and
growth factor-1 receptor inhibition in mesothelioma. Ann Thorac nonmalignant disease. Cancer 1999; 85: 178–87.
Surg 2006; 82: 996–1001. 197. Strizzi L, Catalano A, Vianale G, et al. Vascular endothelial growth
175. Whitson BA, Kratzke RA. Molecular pathways in malignant pleural factor is an autocrine growth factor in human malignant
mesothelioma. Cancer Lett 2006; 239: 183–9. mesothelioma. J Pathol 2001; 193: 468–75.
176. Hodzic D, Delacroix L, Willemsen P, et al. Characterization of the 198. Sako A, Kitayama J, Yamaguchi H, et al. Vascular endothelial
IGF system and analysis of the possible molecular mechanisms growth factor synthesis by human omental mesothelial cells is
leading to IGF-II overexpression in a mesothelioma. Horm Metab augmented by fibroblast growth factor-2: possible role of
Res 1997; 29: 549–55. mesothelial cell on the development of peritoneal metastasis. J
177. Hoang CD, D’Cunha J, Kratzke MG, et al. Gene expression profiling Surg Res 2003; 115: 113–20.
identifies matriptase overexpression in malignant mesothelioma. 199. Brown LF, Detmar M, Claffey K, Nagy JA. Vascular permeability
Chest 2004; 125: 1843–52. factor/vascular endothelial growth factors: A multifunctional
178. Briselli M, Mark EJ, Dickersin GR. Solitary fibrous tumours of the angiogenic cytokine. EXS 1997; 79: 233–69.
pleura: eight new cases and review of 360 cases in the literature. 200. Ferrara N, Keyt B. Vascular endothelial growth factor: Basic
Cancer 1981; 47: 2678–89. biology and clinical implications. EXS 1997; 79: 209–32.
179. Cole FH Jr, Ellis RA, Goodman RC, Weber BC, Courington DP. 201. Takahama M, Tsutsumi M, Tsujiuchi T, et al. Enhanced expression
Benign fibrous pleural tumor with elevation of insulin-like growth of Tie2, its ligand angiopoietin-1, vascular endothelial growth
factor and hypoglycemia. South Med J 1990; 83: 690–4. factor, and CD31 in human non-small cell lung carcinomas. Clin
180. Fukasawa Y, Takada A, Tateno M, et al. Solitary fibrous tumor of Cancer Res 1999; 5: 2506–10.
the pleura causing recurrent hypoglycemia by secretion of insulin- 202. Yoshiji H, Gomez DE, Shibuya M, Thorgeirsson UP. Expression of
like growth factor II. Pathol Int 1998; 48: 47–52. vascular endothelial growth factor, its receptor, and other
181. Masson EA, MacFarlane IA, Graham D, Foy P. Spontaneous angiogenic factors in human breast cancer. Cancer Res 1996; 56:
hypoglycaemia due to a pleural fibroma: role of insulin like growth 2013–6.
factors. Thorax 1991; 46: 930–1. 203. Cacciotti P, Strizzi L, Vianale G, et al. The presence of simian-virus
182. Kishi K, Homma S, Tanimura S, Matsushita H, Nakata K. 40 sequences in mesothelioma and mesothelial cells is associated
Hypoglycemia induced by secretion of high molecular weight with high levels of vascular endothelial growth factor. Am J Respir
insulin-like growth factor-II from a malignant solitary fibrous Cell Mol Biol 2002; 26: 189–93.
tumor of the pleura. Intern Med 2001; 40: 341–4. 204. Masood R, Kundra A, Zhu S, et al. Malignant mesothelioma growth
183. Rena O, Filosso PL, Papalia E, et al. Solitary fibrous tumour of the inhibition by agents that target the VEGF and VEGF-C autocrine
pleura: surgical treatment. Eur J Cardiothorac Surg 2001; 19: 185–9. loops. Int J Cancer 2003; 104: 603–10.
184. Chang ED, Lee EH, Won YS, et al. Malignant solitary fibrous tumor 205. Thickett DR, Armstrong L, Millar AB. Vascular endothelial growth
of the pleura causing recurrent hypoglycemia; factor (VEGF) in inflammatory and malignant pleural effusions.
immunohistochemical stain of insulin-like growth factor i receptor Thorax 1999; 54: 707–10.
in three cases. J Korean Med Sci 2001; 16: 220–4. 206. D’Arcangelo D, Facchiano F, Barlucchi LM, et al. Acidosis inhibits
185. Li Y, Chang Q, Rubin BP, et al. Insulin receptor activation in endothelial cell apoptosis and function and induces basic
solitary fibrous tumours. J Pathol 2007; 211: 550–4. fibroblast growth factor and vascular endothelial growth factor
186. Bourcigaux N, Arnault-Ouary G, Christol R, et al. Treatment of expression. Circ Res 2000; 86: 312–8.
hypoglycemia using combined glucocorticoid and recombinant 207. Clauss M, Schaper W. Vascular endothelial growth factor: a jack
human growth hormone in a patient with a metastatic non-islet of all trades or a nonspecific stress gene? Circ Res 2000; 86:
cell tumor hypoglycemia. Clin Ther 2005; 27: 246–51. 251–2.
187. Drake WM, Miraki F, Siddiqi A, et al. Dose-related effects of growth 208. Momi H, Matsuyama W, Inoue K, et al. Vascular endothelial
hormone on IGF-I and IGF-binding protein-3 levels in non-islet cell growth factor and proinflammatory cytokines in pleural effusions.
tumour hypoglycaemia. Eur J Endocrinol 1998; 139: 532–6. Respir Med 2002; 96: 817–22.
188. Olchovsky D, Shimon I, Goldberg I, et al. Elevated insulin-like 209. Kalomenidis I, Stathopoulos GT, Barnette R, et al. Vascular
growth factor-1 and insulin-like growth factor binding protein-2 endothelial growth factor levels in post-CABG pleural effusions
in malignant pleural effusion. Acta Oncol 2002; 41: 182–7. are associated with pleural inflammation and permeability. Respir
189. Grove CS, Lee YCG. Vascular endothelial growth factor: the key Med 2007; 101: 223–9.
mediator in pleural effusion formation. Curr Opin Pulm Med 2002; 210. Light RW. Pleural diseases, 4th edn. Baltimore: Lippincott,
8: 294–301. Williams & Wilkins, 2001.
190. Mohammed KA, Nasreen N, Hardwick J, et al. Bacterial induction 211. Ishiko O, Yoshida H, Sumi T, Hirai K, Ogita S. Vascular endothelial
of pleural mesothelial monolayer barrier dysfunction. Am J Physiol growth factor levels in pleural and peritoneal fluid in Meigs’
Lung Cell Mol Physiol 2001; 281: L119–L25. syndrome. Eur J Obstet Gynecol Reprod Biol 2001; 98: 129–30.
191. Cheng D, Rodriguez RM, Perkett EA, et al. Vascular endothelial 212. Ferrara N. Molecular and biological properties of vascular
growth factor in pleural fluid. Chest 1999; 116: 760–5. endothelial growth factor. J Mol Med 1999; 77: 527–43.
References 95
213. Neufeld G, Cohen T, Gengrinovitch S, Poltorak Z. Vascular 233. Punglia RS, Lu M, Hsu J, et al. Regulation of vascular endothelial
endothelial growth factor and its receptors. FASEB J 1999; 13: growth factor expression by insulin-like growth factor I. Diabetes
9–22. 1997; 46: 1619–26.
214. Kranz A, Rau C, Kochs M, Waltenberger J. Elevation of vascular 234. Schams D, Kosmann M, Berisha B, Amselgruber WM, Miyamoto A.
endothelial growth factor-A serum levels following acute Stimulatory and synergistic effects of luteinising hormone and
myocardial infarction. Evidence for its origin and functional insulin like growth factor 1 on the secretion of vascular
significance. J Mol Cell Cardiol 2000; 32: 65–72. endothelial growth factor and progesterone of cultured bovine
215. Lee SH, Wolf PL, Escudero R, et al. Early expression of angiogenesis granulosa cells. Exp Clin Endocrinol Diabetes 2001; 109: 155–62.
factors in acute myocardial ischemia and infarction. N Engl J Med 235. Ryuto M, Ono M, Izumi H, et al. Induction of vascular endothelial
2000; 342: 626–33. growth factor by tumor necrosis factor alpha in human glioma
216. Yano S, Shinohara H, Herbst RS, et al. Production of experimental cells. Possible roles of SP-1. J Biol Chem 1996; 271: 28220–8.
malignant pleural effusions is dependent on invasion of the pleura 236. Nauck M, Roth M, Tamm M, et al. Induction of vascular
and expression of vascular endothelial growth factor/vascular endothelial growth factor by platelet-activating factor and
permeability factor by human lung cancer cells. Am J Pathol 2000; platelet-derived growth factor is downregulated by
157: 1893–903. corticosteroids. Am J Respir Cell Mol Biol 1997; 16: 398–406.
217. Yeo K-T, Wang HH, Nagy JA, et al. Vascular permeability factor 237. Saadeh PB, Mehrara BJ, Steinbrech DS, et al. Mechanisms of
(vascular endothelial growth factor) in guinea pig and human fibroblast growth factor-2 modulation of vascular endothelial
tumor and inflammatory effusions. Cancer Res 1993; 53: 2912–8. growth factor expression by osteoblastic cells. Endocrinology
218. Margolin K. Inhibition of vascular endothelial growth factor in the 2000; 141: 2075–83.
treatment of solid tumors. Curr Oncol Rep 2002; 4: 20–8. 238. Mandl-Weber S, Cohen CD, Haslinger B, Kretzler M, Sitter T.
219. Schlaeppi JM, Wood JM. Targeting vascular endothelial growth Vascular endothelial growth factor production and regulation in
factor (VEGF) for anti-tumor therapy, by anti-VEGF neutralizing human peritoneal mesothelial cells. Kidney Int 2002; 61: 570–8.
monoclonal antibodies or by VEGF receptor tyrosine-kinase 239. Gruden G, Thomas S, Burt D, et al. Interaction of angiotensin II and
inhibitors. Cancer Metastasis Rev 1999; 18: 473–81. mechanical stretch on vascular endothelial growth factor
220. Robinson CJ, Stringer SE. The splice variants of vascular production by human mesangial cells. J Am Soc Nephrol 1999; 10:
endothelial growth factor (VEGF) and their receptors. J Cell Sci 730–7.
2001; 114: 853–65. 240. Sauter M, Cohen CD, Wörnle M, et al. ACE inhibitor and AT1-
221. Zachary I, Gliki G. Signaling transduction mechanisms mediating receptor blocker attenuate the production of VEGF in mesothelial
biological actions of the vascular endothelial growth factor family. cells. Perit Dial Int 2007; 27: 162–72.
Cardiovasc Res 2001; 49: 568–81. 241. Cascinu S, Del Ferro E, Ligi M, et al. Inhibition of vascular
222. Giles FJ. The vascular endothelial growth factor (VEGF) signaling endothelial growth factor by octreotide in colorectal cancer
pathway: A therapeutic target in patients with hematologic patients. Cancer Invest 2001; 19: 8–12.
malignancies. Oncologists 2001; 6 (Suppl 5): 32–9. 242. Fischer S, Renz D, Schaper W, Karliczek GF. In vitro effects of
●223. Yano S, Herbst RS, Shinohara H, et al. Treatment for malignant dexamethasone on hypoxia-induced hyperpermeability and
pleural effusion of human lung adenocarcinoma by inhibition of expression of vascular endothelial growth factor. Eur J Pharmacol
vascular endothelial growth factor receptor tyrosine kinase 2001; 411: 231–43.
phosphorylation. Clin Cancer Res 2000; 6: 957–65. 243. Kirschenbaum A, Liu X-H, Yao S, Levine AC. The role of
224. Mesiano S, Ferrara N, Jaffe RB. Role of vascular endothelial cyclooxygenase-2 in prostate cancer. Urology 2001; 58 (Suppl.
growth factor in ovarian cancer: inhibition of ascites formation by 2A): 127–31.
immunoneutralization. Am J Pathol 1998; 153: 1249–56. 244. Renner W, Kotschan S, Hoffmann C, Obermayer-Pietsch B, Pilger E.
225. Yoshiji H, Kuriyama S, Hicklin DJ, et al. The vascular endothelial A common 936 C/T mutation in the gene for vascular endothelial
growth factor receptor KDR/Flk-1 is a major regulator of growth factor is associated with vascular endothelial growth
malignant ascites formation in the mouse hepatocellular factor plasma levels. J Vasc Res 2000; 37: 443–8.
carcinoma model. Hepatology 2001; 33: 841–7. 245. Hoyer RJ, Leung N, Witzig TE, Lacy MQ. Treatment of diuretic
226. Stoelcker B, Echtenacher B, Weich HA, et al. VEGF/Flk-1 refractory pleural effusions with bevacizumab in four patients
interaction, a requirement for malignant ascites recurrence. J with primary systemic amyloidosis. Am J Hematol 2007; 82:
Interferon Cytokine Res 2000; 20: 511–7. 409–13.
227. Verheul HMW, Hoekman K, Jorna AS, Smit EF, Pinedo HM. 246. Pichelmayer O, Zielinski C, Raderer M. Response of a nonmalignant
Targeting vascular endothelial growth factor blockade: ascites and pleural effusion to bevacizumab. N Engl J Med 2005 353: 740–1.
pleural effusion formation. Oncologist 2000; 5: 45–50. 247. Moldawer LL. Biology of proinflammatory cytokines and their
228. Matsumori Y, Yano S, Goto H, et al. ZD6474, an inhibitor of antagonists. Crit Care Med 1994; 22: S3–S7.
vascular endothelial growth factor receptor tyrosine kinase, 248. Goodman RB, Wood RG, Martin TR, Hanson-Painton O, Knasewitz
inhibits growth of experimental lung metastasis and production of GT. Cytokine-stimulated human mesothelial cells produce
malignant pleural effusions in a non-small cell lung cancer model. chemotactic activity for neutrophils including NAP-1/IL-8. J
Oncol Res 2006; 16: 15–26. Immunol 1992; 148: 457–65.
●229. Motomura Y, Kanbayashi H, Khan WI, et al. The gene transfer of 249. Zeillemaker AM, Mul FP, Hoynck van Papendrecht AA, et al.
soluble VEGF type I receptor (Flt-1) attenuates peritoneal fibrosis Neutrophil adherence to and migration across monolayers of
formation in mice but not soluble TGF-beta type II receptor gene human peritoneal mesothelial cells. The role of mesothelium in the
transfer. AJP – Gastrointestinal Liver Physiol 2005; 288: G143–G50. influx of neutrophils during peritonitis. J Lab Clin Med 1996; 127:
230. Boussat S, Eddahibi S, Coste A, et al. Expression and regulation of 279–86.
vascular endothelial growth factor in human pulmonary epithelial 250. Alkhamesi NA, Ziprin P, Pfistermuller K, Peck DH, Darzi AW. ICAM-
cells. Am J Physiol Lung Cell Mol Physiol 2000; 279: L371–8. 1 mediated peritoneal carcinomatosis, a target for therapeutic
231. Cohen T, Nahari D, Cerem LW, Neufeld G, Levi BZ. Interleukin 6 intervention. Clin Exp Metastasis 2005; 22: 449–59.
induces the expression of vascular endothelial growth factor. J 251. Robson RL, McLoughlin RM, Witowski J, et al. Differential
Biol Chem 1996; 271: 736–41. regulation of chemokine production in human peritoneal
232. Ravindranath N, Wion D, Brachet P, Djakiew D. Epidermal growth mesothelial cells: IFN-gamma controls neutrophil migration across
factor modulates the expression of vascular endothelial growth the mesothelium in vitro and in vivo. J Immunol 2001; 167:
factor in the human prostate. J Androl 2001; 22: 432–43. 1028–38.
96 Immunology
252. Zhang XY, Guckian M, Nasiri N, et al. Normal and SV40 271. Mochizuki Y, Nakanishi H, Kodera Y, et al. TNF-a promotes
transfected human peritoneal mesothelial cells produce IL-6 and progression of peritoneal metastasis as demonstrated using a
IL-8: implication for gynaecological disease. Clin Exp Immunol green fluorescence protein (GFP)-tagged human gastric cancer cell
2002; 129: 288–96. line. Clin Exp Metastasis 2004; 21: 39–47.
253. Fang CC, Yen CJ, Chen YM, et al. Diltiazem suppresses collagen ●272. Stathopoulos GT, Kollintza A, Moschos C, et al. Tumor necrosis
synthesis and IL-1beta-induced TGF-beta1 production on human factor-alpha promotes malignant pleural effusion. Cancer Res
peritoneal mesothelial cells. Nephrol Dial Transplant 2006; 21: 2007; 67: 9825–34.
1340–7. 273. Rabinowitz MH, Andrews RC, Becherer JD, et al. Design of
254. Braundmeier AG, Nowak RA. Cytokines regulate matrix selective and soluble inhibitors of tumor necrosis factor-alpha
metalloproteinases in human uterine endometrial fibroblast cells converting enzyme (TACE). J Med Chem 2001; 44: 4252–67.
through a mechanism that does not involve increases in 274. Conway JG, Andrews RC, Beaudet B, et al. Inhibition of tumor
extracellular matrix metalloproteinase inducer. Am J Reprod necrosis factor-alpha (TNF-alpha) production and arthritis in the
Immunol 2006; 56: 201–14. rat by GW3333, a dual inhibitor of TNF-alpha-converting enzyme
255. Lee YCG, Knight DA, Lane KB, et al. Activation of Proteinase and matrix metalloproteinases. J Pharmacol Exp Ther 2001; 298:
Activated Receptor-2 in mesothelial cells induces pleural 900–8.
inflammation. Am J Physiol Lung Cell Mol Physiol 2005; 288: 275. Hatanaka K, Kawamura M, Murai N, et al. FR167653, a cytokine
L734–L40. synthesis inhibitor, exhibits anti-inflammatory effects early in rat
256. Stadlmann S, Pollheimer J, Renner K, et al. Response of human carrageenin-induced pleurisy but no effect later. J Pharmacol Exp
peritoneal mesothelial cells to inflammatory injury is regulated by Ther 2001; 299: 519–27.
interleukin-1beta and tumor necrosis factor-alpha. Wound Repair 276. Porta C, Danova M, Orengo AM, et al. Interleukin-2 induces cell
Regen 2006; 14: 187–94. cycle perturbations leading to cell growth inhibition and death in
257. Hua CC, Chang LC, Chen YC, Chang SC. Proinflammatory cytokines malignant mesothelioma cells in vitro. J Cell Physiol 2000; 185:
and fibrinolytic enzymes in tuberculous and malignant pleural 126–34.
effusions. Chest 1999; 116: 1292–6. 277. Chang SC, Hsu YT, Chen YC, et al. Usefulness of soluble interleukin
258. Odeh M, Sabo E, Srugo I, Oliven A. Tumour necrosis factor alpha in 2 receptor in differentiating tuberculous and carcinomatous
the diagnostic assessment of pleural effusion. QJM 2000; 93: pleural effusions. Arch Intern Med 1994; 154: 1097–101).
819–24. 278. Kim YK, Lee SY, Kwon SS, et al. Gamma-interferon and soluble
259. Alexandrakis MG, Coulocheri SA, Bouros D, et al. Evaluation of interleukin 2 receptor in tuberculous pleural effusion. Lung 2001;
inflammatory cytokines in malignant and benign pleural effusions. 179: 175–84.
Oncol Rep 2000; 7: 1327–32. 279. Porcel JM, Gazquez I, Vives M, et al. Diagnosis of tuberculous
260. Rochester CL, Elias JA. Cytokines and cytokine networking in the pleuritis by the measurement of soluble interleukin 2 receptor in
pathogenesis of interstitial and fibrotic lung disorders. Semin pleural fluid. Int J Tuberc Lung Dis 2000; 4: 975–9.
Respir Med 1993; 14: 389–416. 280. Pettersson T, Soderblom T, Nyberg P, et al. Pleural fluid soluble
261. Zhu Z, Yao J, Wang F, Xu Q. TNF-alpha and the phenotypic interleukin 2 receptor in rheumatoid arthritis and systemic lupus
transformation of human peritoneal meosthelioma cell. Chin Med erythematous. J Rheumatol 1994; 21: 1820–4.
J (Engl) 2002; 115: 513–7. 281. Schwartzentruber DJ. Guidelines for the safe administration of
262. Haslinger B, Kleemann R, Toet KH, Kooistra T. Simvastatin high-dose interleukin-2. J Immunother 2001; 24: 287–93.
suppresses tissue factor expression and increases fibrinolytic 282. Whittington R, Faulds D. Interleukin-2. A review of its
activity in tumor necrosis factor-alpha-activated human pharmacological properties and therapeutic use in patients with
peritoneal mesothelial cells. Kidney Int 2003; 63: 2065–74. cancer. Drugs 1993; 46: 446–514.
263. Cheng DS, Rogers JT, Wheeler A, et al. The effects of intrapleural 283. Yanagawa H, Sone S, Munekata M, et al. IL-6 in malignant pleural
polyclonal anti-tumor necrosis factor alpha (TNF alpha) Fab effusions and its augmentation by intrapleural instillation of IL-2.
fragments on pleurodesis in rabbits. Lung 2000; 178: 19–29. Clin Exp Immunol 1992; 88: 207–12.
264. Corbett EL, Mozzato-Chamay N, Butterworth AE, et al. 284. Chen YM, Tsai CM, Whang-Peng J, Perng RP. Interleukin-7 and
Polymorphisms in the tumor necrosis factor-alpha gene promoter interleukin-12 have different effects in rescue of depressed
may predispose to severe silicosis in black South African miners. cellular immunity: comparison of malignant and tuberculous
Am J Respir Crit Care Med 2002; 165: 690–3. pleural effusions. J Interferon Cytokine Res 2001; 21: 249–56.
265. Musk AW, Wong KCC, De Klerk NH, Cookson WO, Moffatt MF. 285. Chen YM, Ting CC, Peng JW, et al. Restoration of cytotoxic T
Candidate gene polymorphism and pulmonary fibrosis in subjects lymphocyte function in malignant pleural effusion: interleukin-15
with asbestos exposure. Am J Respir Crit Care Med 2001; 163: A44. vs. interleukin-2. J Interferon Cytokine Res 2000; 20: 31–9.
●266. Yang H, Bocchetta M, Kroczynska B, et al. TNF-alpha inhibits 286. van Bruggen I, Nelson DJ, Currie AJ, Jackaman C, Robinson BW.
asbestos-induced cytotoxicity via a NF-kappaB-dependent Intratumoral poly-N-acetyl glucosamine-based polymer matrix
pathway, a possible mechanism for asbestos-induced oncogenesis. provokes a prolonged local inflammatory response that, when
Proc Natl Acad Sci 2006; 103: 10397–402. combined with IL-2, induces regression of malignant mesothelioma
267. Wang Y, Faux SP, Hallden G, et al. Interleukin-1beta and tumour in a murine model. J Immunother 2005; 28: 359–67.
necrosis factor-alpha promote the transformation of human 287. Jackaman C, Bundell CS, Kinnear BF, et al. IL-2 Intratumoral
immortalised mesothelial cells by erionite. Int J Oncol 2004; 25: immunotherapy enhances CD8+ T cells that mediate destruction
173–8. of tumor cells and tumor-associated vasculature: A novel
268. Russo P, Catassi A, Malacarne D, et al. Tumor necrosis factor mechanism for IL-2. J Immunol 2003; 171: 5051–63.
enhances SN38-mediated apoptosis in mesothelioma cells. Cancer 288. Lissoni P, Mandala M, Curigliano G, et al. Progress report on the
2005; 103: 1503–18. palliative therapy of 100 patients with neoplastic effusions by
269. van Grevenstein WM, Hofland L, Jeekel J, van Eijck CH. The intracavitary low-dose interleukin-2. Oncology 2001; 60: 308–12.
expression of adhesion molecules and the influence of inflammatory 289. Masotti A, Fumagalli L, Morandini GC. Intrapleural administration
cytokines on the adhesion of human pancreatic carcinoma cells to of recombinant interleukin-2 in non-small cell lung cancer with
mesothelial monolayers. Pancreas 2006; 32: 396–402. neoplastic pleural effusion. Monaldi Arch Chest Dis 1997; 52:
270. van Grevenstein W, Hofland L, van Rossen M, et al. Inflammatory 225–8.
cytokines stimulate the adhesion of colon carcinoma cells to 290. Castagneto B, Zai S, Mutti L, et al. Palliative and therapeutic
mesothelial monolayers. Dig Dis Sci 2007; 52: 2775–83. activity of IL-2 immunotherapy in unresectable malignant pleural
References 97
mesothelioma with pleural effusion: Results of a phase II study on 312. McLoughlin RM, Jenkins BJ, Grail D, et al. IL-6 trans-signaling via
31 consecutive patients. Lung Cancer 2001; 31: 303–10. STAT3 directs T cell infiltration in acute inflammation. Proc Natl
291. Astoul P, Picat-Joossen D, Viallat JR, Boutin C. Intrapleural Acad Sci USA 2005; 102: 9589–94.
administration of interleukin-2 for the treatment of patients with 313. Gabay C. Interleukin-6 and chronic inflammation. Arthritis Res
malignant pleural mesothelioma: a Phase II study. Cancer 1998; Ther 2006; 8 (Suppl 2): S3.
83: 2099–104. 314. Fujino S, Yokoyama A, Kohno N, Hiwada K. Interleukin 6 is an
292. Mulatero C, Surentheran T, Breuer J, Rudd RM. Simian virus 40 autocrine growth factor for normal human pleural mesothelial
and human pleural mesothelioma. Thorax 1999; 54: 60–1. cells. Am J Respir Cell Mol Biol 1996; 14: 508–15.
293. Bretti S, Berruti A, Dogliotti L, et al. Combined epirubicin and 315. McLaren BR, Robinson BW, Lake RA. New chemotherapeutics in
interleukin-2 regimen in the treatment of malignant malignant mesothelioma: effects on cell growth and IL-6
mesothelioma: a multicenter phase II study of the Italian Group on production. Cancer Chemother Pharmacol 2000; 45: 502–8.
Rare Tumors. Tumori 1998; 84: 558–61. 316. Menet E, Corbi P, Ancey C, et al. Interleukine-6 (IL-6) synthesis
294. Lucchi M, Chella A, Melfi F, et al. A phase II study of intrapleural and gp130 expression by human pericardium. Eur Cytokine Netw
immuno-chemotherapy, pleurectomy/decortication, radiotherapy, 2001; 12: 639–46.
systemic chemotherapy and long-term sub-cutaneous IL-2 in 317. Schmitter D, Lauber B, Fagg B, Stahel RA. Hematopoietic growth
stage II-III malignant pleural mesothelioma. Eur J Cardiothorac factors secreted by seven human pleural mesothelioma cell lines:
Surg 2007; 31: 529–33. interleukin-6 production as a common feature. Int J Cancer 1992;
295. Baluna R, Vitetta ES. Vascular leak syndrome: a side effect 51: 296–301.
of immunotherapy. Immunopharmacology 1997; 37: 318. Yao V, Platell C, Hall JC. Peritoneal mesothelial cells produce
117–32. inflammatory related cytokines. ANZ J Surg 2004; 74: 997–1002.
296. Melencio L, McKallip RJ, Guan H, et al. Role of CD4+CD25+ T 319. Wong CF, Yew WW, Leung SK, et al. Assay of pleural fluid
regulatory cells in IL-2-induced vascular leak. Int Immunol 2006; interleukin-6, tumour necrosis factor-alpha and interferon-gamma
18: 1461–71. in the diagnosis and outcome correlation of tuberculous effusion.
297. Assier E, Jullien V, Lefort J, et al. NK cells and polymorphonuclear Respir Med 2003; 97: 1289–95.
neutrophils are both critical for IL-2-induced pulmonary vascular 320. De Smedt A, Vanderlinden E, Demanet C, et al. Characterisation of
leak syndrome. J Immunol 2004; 172: 7661–8. pleural inflammation occurring after primary spontaneous
298. Rose-John S, Scheller J, Elson G, Jones SA. Interleukin-6 biology is pneumothorax. Eur Respir J 2004; 23: 896–900.
coordinated by membrane-bound and soluble receptors: role in 321. Torres SR, Frode TS, Nardi GM, et al. Anti-inflammatory effects of
inflammation and cancer. J Leukoc Biol 2006; 80: 227–36. peripheral benzodiazepine receptor ligands in two mouse models
299. Kishimoto T. The biology of interleukin-6. Blood 1989; 74: of inflammation. Eur J Pharmacol 2000; 408: 199–211.
1–10. 322. Utsunomiya I, Ito M, Oh-ishi S. Generation of inflammatory
300. Kishimoto T, Akira S, Narazaki M, Taga T. Interleukin-6 family of cytokines in zymosan-induced pleurisy in rats: TNF induces IL-6
cytokines and gp130. Blood 1995; 86: 1243–54. and cytokine-induced neutrophil chemoattractant (CINC) in vivo.
301. Scheller J, Rose-John S. Interleukin-6 and its receptor: from bench Cytokine 1998; 10: 956–63.
to bedside. Med Microbiol Immunol 2006; 195: 173–83. 323. Lin CC, Liu CC, Lin CY. Changes in cell population and tumor
302. Heinrich PC, Behrmann I, Haan S, et al. Principles of interleukin necrosis factor, interleukin-6, and interleukin-8 in malignant
(IL)-6-type cytokine signalling and its regulation. Biochem J 2003; pleural effusions after treatment with intrapleural tetracycline.
374 (Pt 1): 1–20. Am Rev Respir Dis 1993; 147: 1503–6.
303. Mullberg J, Schooltink H, Stoyan T, Heinrich PC, Rose-John S. 324. Yahara N, Abe T, Morita K, Tangoku A, Oka M. Comparison of
Protein kinase C activity is rate limiting for shedding of the interleukin-6, interleukin-8, and granulocyte colony-stimulating
interleukin-6 receptor. Biochem Biophys Res Commun 1992; 189: factor production by the peritoneum in laparoscopic and open
794–800. surgery. Surg Endosc 2002; 16: 1615–9.
304. Horiuchi S, Koyanagi Y, Zhou Y, et al. Soluble interleukin-6 325. Yom SS, Busch TM, Friedberg JS, et al. Elevated serum cytokine
receptors released from T cell or granulocyte/macrophage cell lines levels in mesothelioma patients who have undergone pleurectomy
and human peripheral blood mononuclear cells are generated or extrapleural pneumonectomy and adjuvant intraoperative
through an alternative splicing mechanism. Eur J Immunol 1994; photodynamic therapy. Photochem Photobiol 2003; 78: 75–81.
24: 1945–8. 326. Alexandrakis MG, Kyriakou DS, Bouros D, et al. Interleukin-6 and
305. Mackiewicz A, Schooltink H, Heinrich PC, Rose-John S. Complex of its relationships to acute phase proteins in serous effusion
soluble human IL-6-receptor/IL-6 up-regulates expression of differentiation. Oncol Rep 2001; 8: 415–20.
acute-phase proteins. J Immunol 1992; 149: 2021–7. 327. Marie C, Losser MR, Fitting C, et al. Cytokines and soluble cytokine
306. Rodig SJ, Meraz MA, White JM, et al. Disruption of the Jak1 gene receptors in pleural effusions from septic and nonseptic patients.
demonstrates obligatory and nonredundant roles of the Jaks in Am J Respir Crit Care Med 1997; 156: 1515–22.
cytokine-induced biologic responses. Cell 1998; 93: 373–83. 328. Xirouchaki N, Tzanakis N, Bouros D, et al. Diagnostic value of
307. Zhong Z, Wen Z, Darnell JE Jr. Stat3: a STAT family member interleukin-1alpha, interleukin-6, and tumor necrosis factor in
activated by tyrosine phosphorylation in response to epidermal pleural effusions. Chest 2002; 121: 815–20.
growth factor and interleukin-6. Science 1994; 264: 95–8. 329. Alexandrakis MG, Coulocheri SA, Bouros D, Eliopoulos GD.
◆308. Nishimoto N, Kishimoto T. Interleukin 6: from bench to bedside. Evaluation of ferritin, interleukin-6, interleukin-8 and tumor
Nat Clin Pract Rheumatol 2006; 2: 619–26. necrosis factor alpha in the differentiation of exudates and
309. Cuzzocrea S, Sautebin L, De Sarro G, et al. Role of IL-6 in the transudates in pleural effusions. Anticancer Res 1999; 19: 3607–12.
pleurisy and lung injury caused by carrageenan. J Immunol 1999; 330. Nakano T, Chahinian AP, Shinjo M, et al. Interleukin 6 and its
163: 5094–104. relationship to clinical parameters in patients with malignant
310. Hurst SM, Wilkinson TS, McLoughlin RM, et al. IL-6 and its soluble pleural mesothelioma. Br J Cancer 1998; 77: 907–12.
receptor orchestrate a temporal switch in the pattern of leukocyte 331. Hoheisel G, Izbicki G, Roth M, et al. Compartmentalization of pro-
recruitment seen during acute inflammation. Immunity 2001; 14: inflammatory cytokines in tuberculous pleurisy. Respir Med 1998;
705–14. 92: 14–7.
311. McLoughlin RM, Hurst SM, Nowell MA, et al. Differential 332. Yamaguchi T, Kimura H, Yokota S, et al. Effect of IL-6 elevation in
regulation of neutrophil-activating chemokines by IL-6 and its malignant pleural effusion on hyperfibrinogenemia in lung cancer
soluble receptor isoforms. J Immunol 2004; 172: 5676–83. patients. Jpn J Clin Oncol 2000; 30: 53–8.
98 Immunology
333. Hoheisel G, Izbicki G, Roth M, et al. Proinflammatory cytokine 352. Betjes MG, Tuk CW, Struijk DG, et al. Interleukin-8 production by
levels in patients with lung cancer and carcinomatous pleurisy. human peritoneal mesothelial cells in response to tumor necrosis
Respiration 1998; 65: 183–6. factor-alpha, interleukin-1, and medium conditioned by
334. Loret de Mola JR, Arredondo-Soberon F, Randle CP, Tureck RT, macrophages cocultured with Staphylococcus epidermidis. J Infect
Friedlander MA. Markedly elevated cytokines in pleural effusion Dis 1993; 168: 1202–10.
during the ovarian hyperstimulation syndrome: transudate or 353. Man L, Lewis E, Einbinder T, et al. Major involvement of CD40 in
ascites? Fertil Steril 1997; 67: 780–2. the regulation of chemokine secretion from human peritoneal
335. Higashihara M, Sunaga S, Tange T, Oohashi H, Kurokawa K. mesothelial cells. Kidney Int 2003; 64: 2064–71.
Increased secretion of interleukin-6 in malignant mesothelioma 354. Loghmani F, Mohammed K, Nasreen N, et al. Inflammatory
cells from a patient with marked thrombocytosis. Cancer 1992; cytokines mediate C–C (monocyte chemotactic protein 1) and
70: 2105–8. C–X–C (interleukin 8) chemokine expression in human pleural
336. Kimura N, Ogasawara T, Asonuma S, et al. Granulocyte-colony fibroblasts. Inflammation 2002; 26: 73–82.
stimulating factor- and interleukin 6-producing diffuse deciduoid 355. Fukumoto T, Matsukawa A, Yoshimura T, et al. IL-8 is an essential
peritoneal mesothelioma. Mod Pathol 2005; 18: 446–50. mediator of the increased delayed-phase vascular permeability in
337. Adachi Y, Aoki C, Yoshio-Hoshino N, et al. Interleukin-6 induces LPS-induced rabbit pleurisy. J Leukoc Biol 1998; 63: 584–90.
both cell growth and VEGF production in malignant 356. Antony VB, Godbey SW, Kunkel SL, et al. Recruitment of
mesotheliomas. Int J Cancer 2006; 119: 1303–11. inflammatory cells to the pleural space. Chemotactic cytokines, IL-
338. Monti G, Jaurand MC, Monnet I, et al. Intrapleural production of 8, and monocyte chemotactic peptide-1 in human pleural fluids. J
interleukin 6 during mesothelioma and its modulation by gamma- Immunol 1993; 151: 7216–23.
interferon treatment. Cancer Res 1994; 54: 4419–23. 357. Pace E, Gjomarkaj M, Melis M, et al. Interleukin-8 induces
339. Knight DA, Ernst M, Anderson GP, Moodley YP, Mutsaers SE. The lymphocyte chemotaxis into the pleural space. Role of pleural
role of gp130/IL-6 cytokines in the development of pulmonary macrophages. Am J Respir Crit Care Med 1999; 159: 1592–9.
fibrosis: critical determinants of disease susceptibility and 358. Nasreen N, Mohammed KA, Hardwick J, et al. Polar production of
progression? Pharmacol Ther 2003; 99: 327–38. interleukin-8 by mesothelial cells promotes the transmesothelial
340. Heymann D, L’Her E, Nguyen JM, et al. Leukaemia inhibitory factor migration of neutrophils: role of intercellular adhesion molecule-
(LIF) production in pleural effusions: comparison with production 1. J Infect Dis 2001; 183: 1638–45.
of IL-4, IL-8, IL-10 and macrophage-colony stimulating factor (M- 359. Ceyhan BB, Ozgun S, Celikel T, Yalcin M, Koc M. IL-8 in pleural
CSF). Cytokine 1996; 8: 410–6. effusion. Respir Med 1996; 90: 215–21.
341. Meysman M, Schoors DF, Noppen M, Vincken W, Dewilde P. 360. Segura RM, Alegre J, Varela E, et al. Interleukin-8 and markers of
Tuberculous pleural effusion following coronary artery bypass neutrophil degranulation in pleural effusions. Am J Respir Crit
graft. Acta Clin Belg 1995; 50: 305–9. Care Med 1998; 157: 1565–72.
342. Topley N, Brown Z, Jorres A, et al. Human peritoneal mesothelial 361. Broaddus VC, Hebert CA, Vitangcol RV, et al. Interleukin-8 is a
cells synthesize interleukin-8. Synergistic induction by interleukin- major neutrophil chemotactic factor in pleural liquid of patients
1 beta and tumor necrosis factor-alpha. Am J Pathol 1993; 142: with empyema. Am Rev Respir Dis 1992; 146: 825–30.
1876–86. 362. Park JS, Kim YS, Jee YK, Myong NH, Lee KY. Interleukin-8
343. Arici A, Tazuke SI, Attar E, Kliman HJ, Olive DL. Interleukin-8 production in tuberculous pleurisy: Role of mesothelial cells
concentration in peritoneal fluid of patients with endometriosis stimulated by cytokine network involving tumour necrosis factor-
and modulation of interleukin-8 expression in human mesothelial a; and interleukin-1b. Scand J Immunol 2003; 57: 463–9.
cells. Mol Hum Reprod 1996; 2: 40–5. 363. Galffy G, Mohammed KA, Nasreen N, Ward MJ, Antony VB.
344. Li FK, Davenport A, Robson RL, et al. Leukocyte migration across Inhibition of interleukin-8 reduces human malignant pleural
human peritoneal mesothelial cells is dependent on directed mesothelioma propagation in nude mouse model. Oncol Res 1999;
chemokine secretion and ICAM-1 expression. Kidney Int 1998; 54: 11: 187–94.
2170–83. 364. Galffy G, Mohammed KA, Dowling PA, et al. Interleukin 8: an
345. Boylan AM, Ruegg C, Kim KJ, et al. Evidence of a role for autocrine growth factor for malignant mesothelioma. Cancer Res
mesothelial cell-derived interleukin-8 in the pathogenesis of 1999; 59: 367–71.
asbestos-induced pleurisy in rabbits. J Clin Invest 1992; 89: 365. Catania A, Colombo G, Carlin A, et al. Autocrine inhibitory
1257–67. influences of a-melanocyte-stimulating hormone in malignant
346. Broaddus VC, Boylan AM, Hoeffel JM, et al. Neutralization of IL-8 pleural mesothelioma. J Leukoc Biol 2004; 75: 253–9.
inhibits neutrophil influx in a rabbit model of endotoxin-induced 366. Nasreen N, Hartman DL, Mohammed KA, Antony VB. Talc-induced
pleurisy. J Immunol 1994; 152: 2960–7. expression of C-C and C-X-C chemokines and intercellular
347. Visser CE, Steenbergen JJ, Betjes MG, et al. Interleukin-8 adhesion molecule-1 in mesothelial cells. Am J Respir Crit Care
production by human mesothelial cells after direct stimulation Med 1998; 158: 971–8.
with staphylococci. Infect Immun 1995; 63: 4206–9. 367. van den Heuvel MM, Smit HJ, Barbierato SB, et al. Talc-induced
348. Stam TC, Swaak AJ, Kruit WH, Stoter G, Eggermont AM. inflammation in the pleural cavity. Eur Respir J 1998; 12: 1419–23.
Intrapleural administration of tumour necrosis factor-alpha 368. Miller EJ, Kajikawa O, Pueblitz S, et al. Chemokine involvement in
(TNFalpha) in patients with mesothelioma: cytokine patterns and tetracycline-induced pleuritis. Eur Respir J 1999; 14: 1387–93.
acute-phase protein response. Eur J Clin Invest 2000; 30: 336–43. 369. Tsuchiya I, Kasahara T, Yamashita K, et al. Induction of
349. Witowski J, Thiel A, Dechend R, et al. Synthesis of C-X-C and C-C inflammatory cytokines in the pleural effusion of cancer patients
chemokines by human peritoneal fibroblasts: induction by after the administration of an immunomodulator, OK-432: role of
macrophage-derived cytokines. Am J Pathol 2001; 158: 1441–50. IL-8 for neutrophil infiltration. Cytokine 1993; 5: 595–603.
350. Antony VB, Hott JW, Kunkel SL, et al. Pleural mesothelial cell 370. Kataoka M, Morishita R, Hiramatsu J, et al. OK-432 induces
expression of C-C (monocyte chemotactic peptide) and C-X-C production of neutrophil chemotactic factors in malignant pleural
(interleukin 8) chemokines. Am J Respir Cell Mol Biol 1995; 12: effusion. Intern Med 1995; 34: 352–6.
581–8. 371. Bellinghausen I, Knop J, Saloga J. The role of interleukin 10 in the
351. Griffith DE, Miller EJ, Gray LD, Idell S, Johnson AR. Interleukin-1- regulation of allergic immune responses. Int Arch Allergy Immunol
mediated release of interleukin-8 by asbestos-stimulated human 2001; 126: 97–101.
pleural mesothelial cells. Am J Respir Cell Mol Biol 1994; 10: 372. Oberholzer A, Oberholzer C, Moldawer LL. Interleukin-10: a
245–52. complex role in the pathogenesis of sepsis syndromes and its
References 99
potential as an anti-inflammatory drug. Crit Care Med 2002; 30 (1 392. Caminschi I, Venetsanakos E, Leong CC, et al. Cytokine gene therapy
Suppl): S58–63. of mesothelioma. Immune and antitumor effects of transfected
373. Frode TS, Souza GE, Calixto JB. The effects of IL-6 and IL-10 and interleukin-12. Am J Respir Cell Mol Biol 1999; 21: 347–56.
their specific antibodies in the acute inflammatory responses 393. Chen YM, Yang WK, Ting CC, et al. Cross regulation by IL-10 and
induced by carrageenan in the mouse model of pleurisy. Cytokine IL-2/IL-12 of the helper T cells and the cytolytic activity of
2002; 17: 149–56. lymphocytes from malignant effusions of lung cancer patients.
374. Fine JS, Rojas-Triana A, Jackson JV, et al. Impairment of leukocyte Chest 1997; 112: 960–6.
trafficking in a murine pleuritis model by IL-4 and IL-10. 394. Takeuchi E, Yanagawa H, Suzuki Y, et al. IL-12-induced production
Inflammation 2003; 27: 161–74. of IL-10 and interferon-gamma by mononuclear cells in lung
375. Chen YM, Yang WK, Whang-Peng J, Tsai CM, Perng RP. An analysis cancer-associated malignant pleural effusions. Lung Cancer 2002;
of cytokine status in the serum and effusions of patients with 35: 171–7.
tuberculous and lung cancer. Lung Cancer 2001; 31: 25–30. 395. Kimura K, Nishimura H, Matsuzaki T, et al. Synergistic effect of
●376. Cailhier JF, Sawatzky DA, Kipari T, et al. Resident pleural interleukin-15 and interleukin-12 on antitumor activity in a
macrophages are key orchestrators of neutrophil recruitment in murine malignant pleurisy model. Cancer Immunol Immunother
pleural inflammation. Am J Respir Crit Care Med 2006; 173: 2000; 49: 71–7.
540–7. ◆396. Pestka S, Krause CD, Walter MR. Interferons, interferon-like
377. Corsini E, Di Paola R, Viviani B, et al. Increased carrageenan- cytokines, and their receptors. Immunol Rev 2004; 202: 8–32.
induced acute lung inflammation in old rats. Immunology 2005; 397. Ozato K, Tsujimura H, Tamura T. Toll-like receptor signaling and
115: 253–61. regulation of cytokine gene expression in the immune system.
378. Chen YM, Yang WK, Whang-Peng J, Kuo BI, Perng RP. Elevation of Biotechniques 2002; Suppl: 66–8, 70, 2 passim.
interleukin-10 levels in malignant pleural effusion. Chest 1996; 398. Baccala R, Kono DH, Theofilopoulos AN. Interferons as pathogenic
110: 433–6. effectors in autoimmunity. Immunol Rev 2005; 204: 9–26.
379. Yanagawa H, Takeuchi E, Suzuki Y, et al. Presence and potent 399. Okamoto M, Kawabe T, Iwasaki Y, et al. Evaluation of interferon-
immunosuppressive role of interleukin-10 in malignant pleural gamma, interferon-gamma-inducing cytokines, and interferon-
effusion due to lung cancer. Cancer Lett 1999; 136: 27–32. gamma-inducible chemokines in tuberculous pleural effusions. J
380. Kanno H, Naka N, Yasunaga Y, et al. Production of the Lab Clin Med 2005; 145: 88–93.
immunosuppressive cytokine interleukin-10 by Epstein–Barr-virus- 400. Gopi A, Madhavan SM, Sharma SK, Sahn SA. Diagnosis and
expressing pyothorax-associated lymphoma: possible role in the treatment of tuberculous pleural effusion in 2006. Chest 2007;
development of overt lymphoma in immunocompetent hosts. Am J 131: 880–9.
Pathol 1997; 150: 349–57. 401. Mohammed KA, Nasreen N, Ward MJ, et al. Mycobacterium-
381. Drexler HG, Uphoff CC, Gaidano G, Carbone A. Lymphoma cell mediated chemokine expression in pleural mesothelial cells: role
lines: in vitro models for the study of HHV-8+ primary effusion of C-C chemokines in tuberculous pleurisy. J Infect Dis 1998; 178:
lymphomas (body cavity-based lymphomas). Leukemia 1998; 12: 1450–6.
1507–17. ◆402. Yew WW, Leung CC. Update in tuberculosis 2006. Am J Respir Crit
382. Kanno H, Naka N, Yasunaga Y, Aozasa K. Role of an Care Med 2007; 175: 541–6.
immunosuppressive cytokine, interleukin-10, in the development 403. Schierloh P, Yokobori N, Aleman M, et al. Increased susceptibility
of pyothorax-associated lymphoma. Leukemia 1997; 11 (Suppl 3): to apoptosis of CD56dimCD16+ NK cells induces the enrichment
525–6. of IFN-gamma-producing CD56bright cells in tuberculous pleurisy.
383. Barbosa T, Arruda S, Chalhoub M, et al. Correlation between J Immunol 2005; 175: 6852–60.
interleukin-10 and in situ necrosis and fibrosis suggests a role for 404. Bulat-Kardum L, Etokebe GE, Knezevic J, et al. Interferon-gamma
interleukin-10 in the resolution of the granulomatous response receptor-1 gene promoter polymorphisms (G-611A; T-56C) and
of tuberculous pleurisy patients. Microbes Infect 2006; 8: susceptibility to tuberculosis. Scand J Immunol 2006; 63:
889–97. 142–50.
384. Arruda S, Chalhoub M, Cardoso S, Barral-Netto M. Cell-mediated 405. Cooke GS, Campbell SJ, Sillah J, et al. Polymorphism within the
immune responses and cytotoxicity to mycobacterial antigens in interferon-gamma/receptor complex is associated with pulmonary
patients with tuberculous pleurisy in Brazil. Acta Trop 1998; 71: tuberculosis. Am J Respir Crit Care Med 2006; 174: 339–43.
1–15. 406. Etokebe GE, Bulat-Kardum L, Johansen MS, et al. Interferon-
385. Huber TS, Gaines GC, Welborn MBR, et al. Anticytokine therapies gamma gene (T874A and G2109A) polymorphisms are associated
for acute inflammation and the systemic inflammatory response with microscopy-positive tuberculosis. Scand J Immunol 2006; 63:
syndrome: IL-10 and ischemia/reperfusion injury as a new 136–41.
paradigm. Shock 2000; 13: 425–34. 407. Ferrara G, Losi M, D’Amico R, et al. Use in routine clinical practice
386. Park AY, Scott P. IL-12: Keeping cell-mediated immunity alive. of two commercial blood tests for diagnosis of infection with
Scand J Immunol 2001; 53: 529–32. Mycobacterium tuberculosis: a prospective study. Lancet 2006;
387. Dorman SE, Holland SM. Interferon-g and interleukin-12 pathway 367: 1328–34.
defects and human disease. Cytokine Growth Factor Rev 2000; 11: 408. Henao MI, Montes C, Paris SC, Garcia LF. Cytokine gene
321–33. polymorphisms in Colombian patients with different clinical
388. Zhang M, Gong J, Presky DH, Xue W, Barnes PF. Expression of the presentations of tuberculosis. Tuberculosis (Edinb) 2006; 86: 11–9.
IL-12 receptor beta 1 and beta 2 subunits in human tuberculosis. J 409. Krithika R, Marathe U, Saxena P, et al. A genetic locus required for
Immunol 1999; 162: 2441–7. iron acquisition in Mycobacterium tuberculosis. Proc Natl Acad Sci
389. Zhang M, Gately MK, Wang E, et al. Interleukin 12 at the site of U S A 2006; 103: 2069–74.
disease in tuberculosis. J Clin Invest 1994; 93: 1733–9. 410. McDermid JM, Prentice AM. Iron and infection: effects of host iron
390. Hiramatsu K, Yanagawa H, Haku T, Sone S. Generation of killer status and the iron-regulatory genes haptoglobin and NRAMP1
activity by interleukin-12 of mononuclear cells in malignant (SLC11A1) on host-pathogen interactions in tuberculosis and HIV.
pleural effusions due to lung cancer. Cancer Immunol Immunother Clin Sci (Lond) 2006; 110: 503–24.
1998; 46: 1–6. 411. McLoughlin RM, Witowski J, Robson RL, et al. Interplay between
391. Caminschi E, Venetsanakos E, Leong CC, et al. Interleukin-12 IFN-gamma and IL-6 signaling governs neutrophil trafficking and
induces an effective antitumor response in malignant apoptosis during acute inflammation. J Clin Invest 2003; 112:
mesothelioma. Am J Respir Cell Mol Biol 1998; 19: 738–46. 598–607.
100 Immunology
412. Gattacceca F, Pilatte Y, Billard C, et al. Ad-IFN gamma induces interferon alpha-2b in the treatment of diffuse malignant pleural
antiproliferative and antitumoral responses in malignant mesothelioma. Am J Clin Oncol 1994; 17: 80–2.
mesothelioma. Clin Cancer Res 2002; 8: 3298–304. 423. Bard M, Ruffie P. Malignant mesothelioma. Medical oncology:
413. Hand A, Pelin K, Mattson K, Linnainmaa K. Interferon (IFN)-alpha standards, new trends, trials – the French experience. Lung Cancer
and IFN-gamma in combination with methotrexate: in vitro 2004; 45 (Suppl 1): S129–31.
sensitivity studies in four human mesothelioma cell lines. 424. Metintas M, Ozdemir N, Ucgun I, et al. Cisplatin, mitomycin, and
Anticancer Drugs 1995; 6: 77–82. interferon-alpha2a combination chemoimmunotherapy in the
414. Hand AM, Husgafvel-Pursiainen K, Pelin K, et al. Interferon-alpha treatment of diffuse malignant pleural mesothelioma. Chest 1999;
and -gamma in combination with chemotherapeutic drugs: in 116: 391–8.
vitro sensitivity studies in four human mesothelioma cell lines. 425. Monnet I, Breau JL, Moro D, et al. Intrapleural infusion of
Anticancer Drugs 1992; 3: 687–94. activated macrophages and gamma-interferon in malignant
415. Hand AM, Husgafvel-Pursiainen K, Tammilehto L, Mattson K, pleural mesothelioma: a phase II study. Chest 2002; 121: 1921–7.
Linnainmaa K. Malignant mesothelioma: the antiproliferative 426. Parra HS, Tixi L, Latteri F, et al. Combined regimen of cisplatin,
effect of cytokine combinations on three human mesothelioma doxorubicin, and alpha-2b interferon in the treatment of advanced
cell lines. Cancer Lett 1991; 58: 205–10. malignant pleural mesothelioma: a Phase II multicenter trial of the
416. Phan-Bich L, Buard A, Petit JF, et al. Differential responsiveness of Italian Group on Rare Tumors (GITR) and the Italian Lung Cancer
human and rat mesothelioma cell lines to recombinant interferon- Task Force (FONICAP). Cancer 2001; 92: 650–6.
gamma. Am J Respir Cell Mol Biol 1997; 16: 178–86. 427. Pass HI, Temeck BK, Kranda K, et al. Phase III randomized trial of
417. Buard A, Vivo C, Monnet I, et al. Human malignant mesothelioma surgery with or without intraoperative photodynamic therapy and
cell growth: activation of janus kinase 2 and signal transducer and postoperative immunochemotherapy for malignant pleural
activator of transcription 1alpha for inhibition by interferon- mesothelioma. Ann Surg Oncol 1997; 4: 628–33.
gamma. Cancer Res 1998; 58: 840–7. 428. Soulie P, Ruffie P, Trandafir L, et al. Combined systemic
418. Kettunen E, Vivo C, Gattacceca F, Knuutila S, Jaurand MC. Gene chemoimmunotherapy in advanced diffuse malignant
expression profiles in human mesothelioma cell lines in response mesothelioma. Report of a phase I-II study of weekly
to interferon-gamma treatment. Cancer Genet Cytogenet 2004; cisplatin/interferon alfa-2a. J Clin Oncol 1996; 14: 878–85.
152: 42–51. 429. Tansan S, Emri S, Selcuk T, et al. Treatment of malignant pleural
419. DeLong P, Tanaka T, Kruklitis R, et al. Use of cyclooxygenase-2 mesothelioma with cisplatin, mitomycin C and alpha interferon.
inhibition to enhance the efficacy of immunotherapy. Cancer Res Oncology 1994; 51: 348–51.
2003; 63: 7845–52. 430. Upham JW, Musk AW, van Hazel G, Byrne M, Robinson BW.
420. Kruklitis RJ, Singhal S, Delong P, et al. Immuno-gene therapy with Interferon alpha and doxorubicin in malignant mesothelioma: a
interferon-beta before surgical debulking delays recurrence and phase II study. Aust N Z J Med 1993; 23: 683–7.
improves survival in a murine model of malignant mesothelioma. J 431. Boutin C, Nussbaum E, Monnet I, et al. Intrapleural treatment with
Thorac Cardiovasc Surg 2004; 127: 123–30. recombinant gamma-interferon in early stage malignant pleural
421. Altinbas M, Er O, Ozkan M, et al. Ifosfamide, mesna, and mesothelioma. Cancer 1994; 74: 2460–7.
interferon-alpha2A combination chemoimmunotherapy in 432. Lissoni P, Barni S, Tancini G, et al. Intracavitary therapy of
malignant mesothelioma: results of a single center in central neoplastic effusions with cytokines: comparison among interferon
anatolia. Med Oncol 2004; 21: 359–66. alpha, beta and interleukin-2. Support Care Cancer 1995; 3:
422. Ardizzoni A, Pennucci MC, Castagneto B, et al. Recombinant 78–80.
9
Pleural fibrosis
Overview: The pathogenesis of pleural fibrosis and 101 Newly defined post-transcriptional pathways by which 104
disordered fibrin turnover uPA, uPAR and PAI-1 are regulated
Linkage between disordered fibrin turnover and fibrosis 102 Interventional fibrinolysins and anticoagulants to prevent 106
Coagulation and pleural injury 103 pleural injury
Fibrinolysis and pleural injury 103 Key points 108
References 108
OVERVIEW: THE PATHOGENESIS OF PLEURAL pursued over the next several decades using a variety of
FIBROSIS AND DISORDERED FIBRIN fibrinolysins. Recently, a consensus statement endorsed
TURNOVER intrapleural intervention with fibrinolytic agents as an
appropriate interventional option for patients with locu-
The pathogenesis of pleural fibrosis resembles that of lated parapneumonic effusions at risk for poor outcome.4
fibrosis in other organ systems. The basic derangements More recent clinical investigation has challenged this
recapitulate those associated with the development of approach and the use of fibrinolytic therapy is undergoing
fibrotic repair during wound healing.1 In all these situa- reassessment, as described later in this chapter.
tions, the inflammatory response is potentiated by activa- It is clear that many proinflammatory pathways are
tion of multiple proinflammatory pathways including activated during the continuum of evolving pleural injury
cellular, humoral, immunological, cytokine and other and fibrosis. Nonetheless, morphological findings strongly
mediator networks. Complex interactions between com- suggest that disordered fibrin turnover plays a central role
ponents of these proinflammatory pathways contribute to in the pathogenesis of pleural fibrosis.5 While intrapleural
the remodeling process and scarring. The role of selected fibrin is not observed in the absence of pleural disease,
pathways implicated in pleural inflammation and repair extravascular fibrin deposition that occurs at the visceral
are considered in other chapters of this volume (see and parietal pleural surfaces is a hallmark of early pleural
Chapter 7, Pleural inflammation and infection and injury.2,6 Adhesions between the visceral and parietal
Chapter 8, Immunology). pleural surfaces that form within 1 day after pleural injury
A large body of clinical, preclinical and basic investiga- are fibrinous, as confirmed by immunohistochemical
tion lends strong support to the hypothesis that disordered analyses.6 Collagen within the fibrils is detectable within
fibrin turnover and specific pathways of coagulation and 3 days of pleural injury induced by intrapleural admini-
fibrinolysis are integral to pleural remodeling and fibrosis.2 stration of tetracycline in rabbits.7 These observations
It has long been postulated that fibrin strands within confirm that fibrin deposition characterizes pleural injury,
organizing pleural exudates are responsible for loculation, suggesting that intrapleural fibrin formation and/or
and that clearance of intrapleural fibrin by intrapleural clearance could contribute to the development of pleural
administration of fibrinolysins is of therapeutic value. For fibrosis.
example, in the 1940s, Tillett and Sherry3 used prepara- Interestingly, the prominence of extravascular fibrin
tions of streptokinase or streptodornase to break up deposition in pleural injury and repair parallels the situa-
pleural loculations attributable to parapneumonic effu- tion in fibrosing lung injury.2 In acute respiratory distress
sions or hemothoraces. Their approach was in part predi- syndrome (ARDS), fibrin deposition is likewise character-
cated on surgical observations that intrapleural adhesions istic in the alveolar compartment.2,8,9 Alveolar fibrin depo-
appeared to be fibrinous and might therefore be cleared by sition commonly accompanies diffuse alveolar damage,
fibrinolytic agents. The early success of this approach was which is the histologic constellation that commonly occurs
102 Pleural fibrosis
in ARDS.10,11 Prominent fibrin deposition is similarly plasminogen activator inhibitors (PAI) as well as plas-
observed in a variety of acute lung injuries, indicating that minogen activators (PA), including both tissue-type PA
formation of a fibrinous neomatrix is a consistent feature (tPA) and urokinase-type PA (uPA).2
of acute alveolitis.12–15 Alveolar fibrin deposition also char- The major PA implicated in clearance of extravascular
acterizes interstitial lung diseases, including idiopathic fibrin in the lung is uPA (Figure 9.1).8 The relative expres-
pulmonary fibrosis with active alveolitis,16,17 further sup- sion of uPA versus that of PAIs and antiplasmins is a key
porting a link between extravascular fibrin deposition and determinant of local fibrinolytic capacity. With ongoing
subsequent fibrosis. remodeling of transitional fibrin, collagen deposition
eventually leads to progressive scarring and fibrotic repair.
The desmoplastic response associated with solid neo-
plasms is predicated on a similar sequence of events,1,19 as
LINKAGE BETWEEN DISORDERED FIBRIN is accelerated pulmonary fibrosis following ARDS2,8,20 and
TURNOVER AND FIBROSIS in various interstitial lung diseases.18,21,22
There is good evidence to support the hypothesis that
Fibrosis after tissue injury evolves through remodeling and extravascular fibrin in pleural (or other) diseases is patho-
organization of transitional fibrin.2 This paradigm is not physiologic rather than incidental.1,2,8,23–26 Perturbations
unique to injury in the pleural compartment. It is now well of either coagulation or fibrinolytic pathways can influ-
established that a wide range of inflammatory and neo- ence inflammation and tissue repair in several different
plastic diseases is associated with disordered fibrin ways. For example, components of coagulation and fibri-
turnover and fibrin deposition.18 Formation and subse- nolytic pathways are interactive with other proinflamma-
quent remodeling of the transitional fibrin neomatrix tory pathways, and interactions with the complement and
follows a common progression in virtually all forms of kinin systems can amplify the local inflammatory
tissue injury. Acute inflammation initially promotes response.2 The expression of tissue factor can also be stim-
increased microvascular permeability, facilitating the ulated by a variety of cytokines now known to be elabo-
passage of plasma coagulation substrates into the injured rated in pleural diseases.6,27 Thrombin can influence
tissue or inflamed body compartment. Next, coagulation cytokine expression and increase vascular permeability via
at sites of tissue injury is initiated, primarily by tissue cellular signaling.28,29 In addition, fibrin and its derivatives
factor (TF) associated with activated coagulation factor can independently influence the inflammatory response.
VII (Figure 9.1). The TF-VIIa, or extrinsic pathway For example, fibrin and its products can disrupt endothe-
complex, is primarily responsible for amplification of lial cell organization,30 suppress lymphocyte prolifera-
downstream coagulation following tissue injury and initi- tion,31 and promote directed migration of macrophages
ates formation of a fibrin neomatrix.1 Remodeling of the and fibroblasts.32 In addition, proteolytic fragments of
transitional fibrin next occurs via the elaboration and fibrin(ogen) can effect increased vascular permeabil-
release of proteases from inflammatory cells, including ity.30,33,34 Plasmin, which is generated from plasminogen
macrophages, and fibroblasts that invade the neomatrix. through the action of PAs, including uPA, can also activate
While persistent fibrin deposition occurs in ongoing transforming growth factor (TGF)-b and thereby promote
pleural and parenchymal lung injury, continued formation fibrotic repair.26 The induction of PAI-1 by TGF-b is likely
and resorption of extravascular fibrin is facilitated by to be involved in this process. These observations strongly
cytokines and other mediators in the inflammatory support the concept that disordered fibrin turnover is
microenvironment. These mediators can increase local central to the pathogenesis of pleural inflammation and
expression of tissue factor and induce expression of the repair.
major determinant of proteolysis and inflammatory cell been shown to be involved. These mechanisms operate at
traffic, as occurs in the injured pleural space.2,8 Cytokines the post-transcriptional level of mRNA stability. At this
expressed in the course of pleural injury (see also level, the stability of the respective mRNAs are contingent
Chapter 8, Immunology), including tumor necrosis factor on the interactions between mRNA sequences that contain
(TNF)-a and TGF-b can upregulate uPAR at the surface of regulatory information and newly recognized mRNA
cell types expressed in pleural injury.2 Exposure of binding proteins that bind one or more of the coding or
mesothelial cells to particulates can also influence uPAR untranslated mRNA regions (Figure 9.2).
expression. Along these lines, chrysotile or crocidolite
asbestos induce uPAR expression at the surface of rabbit Post-transcriptional regulation of
pleural mesothelial or MeT5A mesothelial cells.73 These uPA, uPAR and PAI-1 expression
observations indicate that pleural disease caused by
asbestos exposure involves alterations of the mesothelial
uPA–uPAR system. It has also been shown that expression
of uPA and uPAR by endothelial or epithelial cells is
increased by asbestos exposure.74–76 In all, these studies
demonstrate that the uPA–uPAR system plays a role in the mRNA binding
protein
pathogenesis of a range of pleural disorders.
Cell
Recent studies suggest that PAI-1, in particular, plays an cytoplasm
important role in the pathogenesis of pulmonary fibro-
sis.77,78 Overexpression of PAI-1 promotes alveolar and
small airway fibrosis in lung injury induced by bleomycin.79 uPA, uPAR or PAI-1 mRNA
Evidence developed by our laboratory and other investiga-
tors likewise strongly implicates PAI-1 as being central to
the pathogenesis of pleural injury and repair.2 Fibrinolytic
Phosphorylation
activity is virtually undetectable in pleural effusions of
patients with exudative pleuritis and in tetracycline-
induced pleural injury where much of the uPA expressed in Stabilized or destabilized mRNA
pleural fluids is bound to PAI-1.35 Interaction between uPA
and PA-inhibitors irreversibly blocks expression of uPA
activity. In addition, in pleural fluids in patients with exuda-
Altered protein expression
tive pleuritis and in tetracycline-induced pleural injury,
PAI-1 is also disproportionately (up to 1000-fold) increased
compared with plasma levels, strongly suggesting that it is Figure 9.2 Post-transcriptional regulation of uPA, uPAR and
a major inhibitor of fibrinolytic activity in exudative pleural PAI-1 expression.
fluids.35 Both PAI-1 and PAI-2 are products of mesothelial
cells, lung fibroblasts and several other cell types,39,40,80 so
that these inhibitors can be locally elaborated in the setting Post-transcriptional regulation of PAI-1
of pleural injury. PAI-1, but not PAI-2, is usually present in
plasma, so that increased microvascular permeability may Regulation of PAI-1 expression has been shown to occur
also influence levels of this inhibitor in pleural fluids. PAI- at multiple levels. In HepG2 cells, for example, expression
1 is also relatively overexpressed in asbestos-mediated of the PAI-1 gene is regulated at both the transcriptional
pleural injury.75,81,82 Levels of PAI-1 in pleural fluids of and post-transcriptional levels.83 In HepG2 cells, PAI-1
patients with congestive heart failure approximate those mRNA exists in two forms; a labile 3.2 kb (half-life
found in plasma, whereas the PAI-1 concentrations in 0.85 hours) and a relatively more stable 2.2 kb (half-life
exudative fluids are, as noted above, markedly increased. 2.5 hours) species. Post-transcriptional control is involved
in the regulation of both species. Steady-state levels of
PAI-1 mRNA increase in these cells in response to treat-
NEWLY DEFINED POST-TRANSCRIPTIONAL ment with insulin and insulin-like growth factor-I (IGF-
PATHWAYS BY WHICH uPA, uPAR AND PAI-1 1).84 IGF-I stabilizes both PAI-1 mRNA species while
ARE REGULATED insulin stabilizes the 3.2 kb PAI-1 mRNA form. mRNA-
binding multiprotein complexes with molecular masses
Since expression of PAI-1, uPA and uPAR are all linked to ranging from 38 to 76 kDa interact with cAMP responsive
disordered fibrin clearance, the mechanisms by which sequences to regulate PAI-1 mRNA.85 These observations
these proteins are regulated in the pleural compartment indicate that a post-transcriptional component con-
are germane to pleural repair. Using cultured pleural tributes to expression of PAI-1 in selected cell types and
mesothelial and mesothelioma cells, novel mechanisms that PAI-1 mRNA-binding protein interactions can influ-
involving ‘switch-off’ rather than gene activation have ence PAI-1 mRNA stability.
Newly defined post-transcriptional pathways by which uPA, uPAR and PAI-1 are regulated 105
We previously found that human tracheal epithelial shown to regulate uPAR expression in pleural mesothelial
cells express PAI-180 and recently determined that selected and mesothelioma cells. uPAR mRNA stability is
lung carcinoma cells express more PAI-1 antigen com- increased in these cells by treatment with cytokines and
pared with normal lung epithelial cells. Post-transcrip- translational inhibitors.94 The ability of translational
tional regulation accounts for, at least in part, the inhibitors to augment the stability of uPAR mRNA in
increased PAI-1 expression in lung cancer cells, attested to mesothelial cells suggests that a protein is involved in the
by increased PAI-1 mRNA stability after transcriptional regulatory process. In fact, a 50-kDa moiety; phospho-
blockade. Cytokines, including TNF-a and TGF-b, glycerate kinase (PGK) selectively interacts with a coding
increase the stability of PAI-1 mRNA, an effect that is also region sequence of the uPAR mRNA.95 The binding
regulated at the post-transcriptional level.86 uPA also sequence for PGK on uPAR mRNA has been mapped and
induces PAI-1 expression via post-transcriptional stabi- encompasses a 51 nucleotide sequence within the coding
lization of PAI-1 mRNA.87 Apart from its role in lung and region.94 It also appears that a 40 kDa protein, hnRNPC,96
pleural inflammation, the regulation of PAI-1 appears to and the 30 kDa protein HuR97 likewise regulate uPAR
be of clinical importance in neoplasia, as previous reports mRNA stability. The PGK–uPAR mRNA coding region
suggest that increased PAI-1 expression relates to poor interaction destabilizes uPAR mRNA whereas the
outcome in lung cancer patients.88,89 hnRNPC–3¢UTR interaction decreases uPAR mRNA
While the regulatory mechanism is incompletely degradation. However, the role of Hur on uPAR mRNA
understood at the present time, the interaction between a turnover is not well established but most likely involves
sequence within the untranslated region of PAI-1 mRNA stabilization of uPAR mRNA.97 Based on chimeric gene
and a 60 kDa PAI-1 mRNA binding protein appears to reg- analyses, it appears that both the PGK binding uPAR
ulate the stability of PAI-1 mRNA. In small airway epithe- coding region and the hnRNPC-binding 3¢UTR determi-
lial cells this interaction is readily detectable and correlates nants contain information for uPAR mRNA degradation.
with a relatively short PAI-1 mRNA half-life. In cytokine- The relative contribution of these determinants to the reg-
treated cells or malignant lung carcinoma derived cells the ulation of uPAR mRNA stability was further confirmed by
binding interaction is attenuated or undetectable. This creation of deletion mutant mRNAs lacking either coding
inverse relationship between PAI-1 mRNA stability and or 3¢UTR or both determinants. It was found that both
the PAI-1 mRNA binding protein-PAI-1 mRNA interac- the coding and 3¢UTR determinants contain information
tion suggests that the interaction is destabilizing. The for uPAR mRNA destabilization and deletion of both the
precise binding sequence within PAI-1 mRNA that inter- determinants additively stabilizes uPAR mRNA (Shetty,
acts with the protein remains to be elucidated and the reg- unpublished results). Interaction of PGK with the 51 nt
ulatory consequences of the binding interaction are yet to coding region determinant enhances the degradation of
be confirmed. uPAR mRNA and proinflammatory agents inhibit PGK
We determined that post-transcriptional regulation of binding to uPAR mRNA by tyrosine phosphorylation of
PAI-1 similarly occurs in MeT5A human pleural PGK.98 Alternatively, the binding of hnRNPC to the
mesothelial cells and in primary cultures of rabbit pleural 110 nt 3¢UTR binding sequence promotes uPAR mRNA
mesothelial cells (Shetty and Idell, unpublished results). stability.96
In these cells, PAI-1 mRNA stability is increased by Unstimulated MeT5A or primary human pleural
cytokine stimulation. In MeT5A cells, the 60 kDa PAI-1 mesothelial or mesothelioma cells demonstrate interaction
mRNA binding protein-PAI-1 mRNA interaction appears between the 5¢ coding region and PGK.94 TGF-b and
to be involved in the regulation of PAI-1 gene expression. TNF-a, cytokines known to participate in the pathogene-
These observations indicate that regulation of PAI-1 sis of pleural injury, block or attenuate the binding inter-
expression by pleural mesothelial cells involves a post- action and concurrently increase stability of uPAR mRNA.
transcriptional component. PAI-1 mRNA is also regulated These same changes in the binding interaction occur in
at the post-transcriptional level by 8-bromo-cAMP and rabbit pleural mesothelial cells,67 malignant mesothelioma
cAMP-mediated destabilization of PAI-1 mRNA and cells and lung carcinoma cells.69 uPA similarly induces
involves association of the 3¢-most 134 nt PAI-1 mRNA uPAR expression through post-transcriptional stabiliza-
sequence with a 50 kDa PAI-1-mRNA binding protein.90 tion of uPAR mRNA in lung epithelial cells,99 but this
Whether this regulatory pathway contributes to regula- pathway has yet to be demonstrated in pleural mesothelial
tion of PAI-1 in pleural mesothelial cells remains to be cells.
determined.
matrix remodeling by these cells. The mechanisms by loculation, malignant effusions with loculation and hemo-
which uPA is regulated by these cells remain to be eluci- thorax with pleural organization.111–121 The clinical end-
dated but recent evidence in other cell types raises the pos- points of efficacy in these studies included improvement in
sibility that this molecule may, like uPAR or PAI-1, be radiographic evidence of pleural injury, pleural drainage
subject to post-transcriptional control. In normal renal and the need for further surgical procedures. The
epithelial cells, uPA mRNA has been shown to be regu- intrapleural administration of fibrinolysins appears to be
lated at this level in a number of studies.100–103 It appears well tolerated, generally does not cause clinically impor-
that the 3¢ untranslated region of uPA mRNA contains tant systemic fibrinolysins and reduces the need for
multiple instability determinants that regulate the half-life surgery.110–114,122,123 Other reports have shown that the
of uPA mRNA.104 A 40 kDa AU-binding protein has been increased volume of pleural drainage associated with use
identified in breast cancer cells that may be involved in the of thrombolytic therapy does not uniformly alter the mor-
regulatory mechanism.105 Epithelial cells from the lung bidity or mortality of underlying empyema.115 In 2000, an
also regulate uPA gene expression at the post-transcrip- evidence-based guideline on the management of para-
tional level.106 Interaction of a 30 kDa uPA mRNA- pneumonic effusions from the American College of Chest
binding protein with a 66 nucleotide sequence of the 3¢ Physicians (ACCP) supported the optional use of fibri-
untranslated region of uPA mRNA appears to regulate nolytic interventions if the effusion was large, loculated or
uPA mRNA expression in Beas2B cells. By chimeric gene was associated with thickened pleura, positive culture, pus,
analyses, in which this binding sequence is inserted in b- or pH was < 7.2. Fibrinolytic therapy was associated with
globin mRNA, the binding interaction is destabilizing and lower mortality and a decreased need for second interven-
shortens the decay of uPA mRNA. Interestingly, the tion in patients with empyema and extensive disease and at
binding protein is found in the cytoplasm of Beas2B cells risk for poor outcome.4 A Cochrane review assessed the
but not in lung carcinoma cells. In carcinoma-derived evidence for efficacy of fibrinolytic therapy from available
cells, which exhibit a longer uPA mRNA half-life and randomized controlled trials in 2004. The review con-
express more uPA, the binding protein is found in the cluded that intrapleural fibrinolytic interventions is an
nucleus. These observations suggest that the adjunctive therapy for patients with empyema or compli-
nuclear–cytoplasmic distribution of the protein is a key cated parapneumonic effusions as it confers significant
determinant of the post-transcriptional regulatory mecha- benefits in terms of hospital stay, duration of fever, radi-
nism. Proinflammatory cytokines including TNF-a ographic improvement and the need for surgical treat-
enhance uPA expression at the post-transcriptional level ment. However, the routine use of fibrinolysins was not
by altering the interaction of uPA mRNA with a 30 kDa recommended because the numbers of patients in these
uPA mRNA-binding protein that is yet to be character- randomized trials were small and study designs hetero-
ized.107 The role of this pathway in fibrin clearance in the geneous.124,125
pleural compartment remains to be determined. More recently, a randomized controlled trial in 53 sub-
jects showed improved outcomes and decreased need for
surgery when fibrinolysins with streptokinase was used in
INTERVENTIONAL FIBRINOLYSINS AND adjunction to drainage.126 However, a large randomized
ANTICOAGULANTS TO PREVENT PLEURAL controlled clinical trial (MIST1) including 454 subjects
INJURY with pleural infection treated with intrapleural streptoki-
nase or placebo127 recently challenged the prevailing favor-
The introduction of fibrinolysins to treat pleural locula- able view of the efficacy of intrapleural fibrinolytic therapy.
tion occurred over 50 years ago.3 These agents continue to In this trial, streptokinase did not alter mortality, the need
be used in clinical practice today. Intrapleural fibrinolysins for surgery at 3 and 12 months, radiological appearance, or
upregulate local fibrinolytic activity and thereby reverse length of stay. A meta-analysis of all trials using fibri-
the fibrinolytic defect found in exudative pleuritis.2,35,61 nolysins for pleural infection arrived at results similar to
Interestingly, increased pleural fluid fibrinolytic activity is MIST1, in large part because the MIST1 trial contributed
associated with failure of pleurodesis, further suggesting 75 percent of the subjects in the analysis.128 These new
that pathways of fibrin clearance are related to the results have sparked discussions to explain the differences
outcome of pleural fibrosis.108 From a pathophysiology between these results and those of multiple trials support-
perspective, the ability of fibrinolysins to clear pleural loc- ing the use of intrapleural fibrinolysins.129–133 While the
ulation underscores the importance of intrapleural fibrin basis for the disparity is unclear, differences in outcomes
in the pathogenesis of pleural organization and may relate to enrolment of subjects with early effusions
scarring.109,110 versus those with relatively protracted illness and well-
Streptokinase and urokinase have frequently been used established pleural reaction where fibrinolysins may be less
as the interventional agents in recently reported litera- likely to be efficacious. The advanced age of subjects and
ture.111–115 Several reports indicate the efficacy of this commitment to intrapleural fibrinolysis because of limited
approach in a variety of circumstances, including pleural availability of surgery may have contributed to masking of
empyemas, complicated parapneumonic effusions with potential benefits of fibrinolysins in patients with pleural
Interventional fibrinolysins and anticoagulants to prevent pleural injury 107
loculations (see also Chapter 24, Effusions from cardiac tPA.2 A comparable paucity of pleural fluid fibrinolytic
diseases). activity is also seen in experimental pleural injury induced
A recent report comparing video-assisted thoracoscopy by tetracycline in rabbits.27 Interestingly, intrapleural
with intrapleural streptokinase in post-traumatic retained administration of either the anticoagulant heparin or the
hemothorax showed thoracoscopy to be associated with low molecular weight uPA, the form of uPA that was until
shorter hospitalization and a decreased need for decortica- recently most commonly used for clinical applications,
tion.134 A randomized controlled trial comparing the two attenuated pleurodesis in rabbits with tetracycline-
strategies for empyema also showed higher efficacy, induced pleural fibrosis.27 Presumably, the effects relate to
shorter hospitalization and lower costs for thoracoscopy the blockade of intrapleural fibrin, although it is possible
compared with catheter directed fibrinolytic therapy.135 that the attenuation of tissue injury could be attributable
These data support the use of early thoracoscopy in com- to altered cytokine networking or cross-talk, or cellular
plicated parapneumonic effusions, reserving fibrinolytics signaling. These studies suggest that the balance of plas-
for poor surgical candidates or circumstances where thora- minogen activators versus their inhibitors in organizing
coscopy is not available. Further studies are needed to pleuritis can therefore be altered to prevent fibrotic seque-
clarify if there are specific clinical circumstances in which lae of pleural injury.
fibrinolytic therapy with currently available or new agents Due to increased PAI-1 appearing to play a major role
can yield superior outcomes versus tube thoracostomy or in the downregulation of pleural fluid fibrinolytic activity,
thoracoscopy. a fibrinolytic agent that resists inhibition by PAI-1 would
The results of studies in adults may not be applicable to be desirable for intrapleural administration. Single-chain
the pediatric population with pleural infection. In children uPA exhibits this property when bound to uPAR. scuPA is
with complicated parapneumonic effusion, fibrinolysis has the form of urokinase secreted by cells and it can be
been achieved with streptokinase, urokinase, or alteplase. cleaved by proteases such as plasmin to the more active
Several reports show fibrinolytic therapy to be efficacious, tcuPA. Both forms of urokinase bind uPAR with high
safe and a less invasive option in children.136–140 Few ran- affinity.59,60 tcuPA is susceptible to inhibition by PAI-1 but
domized controlled trials of intrapleural fibrinolytic agents scuPA activity is increased when bound to uPAR and it
have been carried out in children. In one randomized con- is relatively resistant to inhibition by either PAI-1 or
trolled trial in children, urokinase reduced hospital stay.141 PAI-2.60,62,145 Thus, binding of scuPA to uPAR protects it
Another randomized controlled trial showed benefit in from conversion to tcuPA, potentiates its enzymatic activ-
terms of decreased pleural thickening at 1 month that was ity and retards its inhibition by PAI-1. Interestingly, this
attributable to fibrinolytic treatment in the subset of chil- agent has been used as an intervention for stroke, where
dren with multiloculated empyema.142 A prospective ran- plasma fibrinogen concentrations were preserved and
domized trial comparing primary video-assisted bleeding complications were reduced compared with other
thoracoscopic surgery (VATS) with tube thoracostomy for fibrinolysins including tPA.146 scuPA also binds rabbit
treatment of parapneumonic effusions showed decreased pleural mesothelial cells and lung fibroblasts with kinetics
length of hospital stay, number of chest tube days, narcotic comparable to that exhibited with binding to human cells
use, number of radiographic procedures and interven- (Mazar and Idell, unpublished data). In tetracycline-
tional procedures when children underwent VATS. No induced pleurodesis in the rabbit, intrapleural administra-
patient in the VATS surgery group required fibrinolytic tion of scuPA was found to be as effective as scuPA bound
therapy.143 Conversely, a recent randomized trial including to its recombinant receptor in terms of prevention of
60 patients demonstrated comparable length of hospital intrapleural adhesion formation.147 More recently, single
stay, failure rate, or radiologic outcome after 6 months of dose scuPA was found to provide excellent protection
follow-up in children with empyema who were treated against pleural adhesion formation in the same model,
with either video-assisted thoracoscopy or intrapleural whether given before, during, or after adhesion forma-
urokinase. Treatment costs were lower in the urokinase tion.148
treatment group.144 The authors concluded that urokinase It is possible that interventions for malignant mesothe-
represents the preferred option given its advantage of lioma may similarly be predicated on changing the local
economy and equivalent efficacy. Larger randomized con- fibrinolytic milieu. While the role of fibrinolysins and their
trolled studies may better delineate the role of fibrinolytic inhibitors in the pathogenesis of malignant mesothelioma
therapy versus surgical treatment for pleural infections in remains to be clarified, there is good reason to speculate
children. that they may play a role in the remodeling of the stroma
Ongoing studies are being carried out to evaluate the of this neoplasm. Fibrous and epithelioid subtypes of this
ability of new fibrinolysins to clear intrapleural fibrinous malignancy express uPA, uPAR and PAI-1.149 Extra-
loculations. At this stage, these agents are being tested in vascular fibrin is commonly found at the leading edge of
preclinical models. In many forms of pleural injury, tumor stroma invading the normal pleura, suggesting that
increased pleural fluid procoagulant activity is accompa- locally disordered fibrin turnover is central to the spread of
nied by a profound decrement in fibrinolytic activity that this neoplasm.149 Whether these observations can be
occurs despite local elaboration of uPA as well as that of exploited for therapeutic gain remains to be determined,
108 Pleural fibrosis
13. Idell S, James KK, Gillies C, et al. Abnormalities of pathways of fibrin turnover in the human pleural space. Am Rev Respir Dis
fibrin turnover in lung lavage of rats with oleic acid and 1991; 144: 187–94.
bleomycin-induced lung injury support alveolar fibrin deposition. 36. Chapman HA, Stahl M, Allen CL, et al. Regulation of the
Am J Pathol 1989; 135: 387–99. procoagulant activity within the bronchoalveolar compartment of
14. Idell S, Peterson BT, Gonzalez KK, et al. Local abnormalities of normal human lung. Am Rev Respir Dis 1988; 137: 1417–25.
coagulation and fibrinolysis and alveolar fibrin deposition in sheep 37. Drake TA, Morrissey JH, Edgington TS. Selective cellular expression
with oleic acid-induced lung injury. Am Rev Respir Dis 1988; 138: of tissue factor in human tissues. Implication for disorders of
1282–94. hemostasis and thrombosis. Am J Pathol 1989; 134: 1087–97.
15. Idell S, Gonzalez KK, MacArthur CK, et al. Bronchoalveolar lavage 38. McGee MP, Rothberger H. Tissue factor in bronchalveolar lavage
procoagulant activity in bleomycin-induced lung injury in fluids. Evidence for an alveolar macrophage source. Am Rev Respir
marmosets. Characterization and relationship to fibrin deposition Dis 1985; 131: 331–6.
and fibrosis. Am Rev Respir Dis 1987; 136: 124–33. 39. Idell S, Zwieb C, Kumar A, et al. Pathways of fibrin turnover of
16. Jackson LK. Idiopathic pulmonary fibrosis. Clin Chest Med 1982; 3: human pleural mesothelial cells in vitro. Am J Respir Cell Mol Biol
579–92. 1992; 7: 414–26.
17. Kuhn C III, Boldt J, King TE Jr, et al. An immunohistochemical 40. Idell S, Zwieb C, Boggaram J, et al. Mechanisms of fibrin formation
study of architectural remodeling and connective tissue and lysis by human lung fibroblasts: influence of TGF-beta and
synthesis in pulmonary fibrosis. Am Rev Respir Dis 1989; 140: TNF-alpha. Am J Physiol 1992; 263: L487–94.
1693–703. 41. Wyshock EG, Idell S, Colman RW. The contribution of factor V to
18. Brown LF, Dvorak AM, Dvorak HF. Leaky vessels, fibrin deposition, the coagulant property of pleural fluid. J Lab Clin Med 1992; 120:
and fibrosis: a sequence of events common to solid tumors and 726–34.
many other types of disease. Am Rev Respir Dis 1989; 140: 42. Griffith DE, Johnson AR, Kumar A, et al. Growth factors for human
1104–7. pleural mesothelial cells in soluble products from formed clots.
◆19. Dvorak HF, Senger DR, Dvorak AM. Fibrin as a component of the Thromb Res 1994; 74: 207–18.
tumor stroma: origins and biological significance. Cancer 43. Jonjic N, Peri G, Bernasconi S, et al. Expression of adhesion
Metastasis Rev 1983; 2: 41–73. molecules and chemotactic cytokines in cultured human
20. Marshall R, Bellingan G, Laurent G. The acute respiratory distress mesothelial cells. J Exp Med 1992; 176: 1165–74.
syndrome: fibrosis in the fast lane. Thorax 1998; 53: 815–17. 44. Griffith DE, Miller EJ, Gray LD, et al. Interleukin-1-mediated
21. Chapman HA Jr, Bertozzi P, Reilly JJ Jr. Role of enzymes mediating release of interleukin-8 by asbestos-stimulated human pleural
thrombosis and thrombolysis in lung disease. Chest 1988; 93: mesothelial cells. Am J Respir Cell Mol Biol 1994; 10: 245–52.
1256–63. 45. Gerwin BI, Lechner JF, Reddel RR, et al. Comparison of production
◆22. Dabbagh K, Chambers RC, Laurent GJ. From clot to collagen: of transforming growth factor-beta and platelet-derived growth
coagulation peptides in interstitial lung disease. Eur Respir J 1998; factor by normal human mesothelial cells and mesothelioma cell
11: 1002–5 lines. Cancer Res 1987; 47: 6180–4.
◆23. Dano K, Andreasen PA, Grondahl-Hansen J, et al. Plasminogen 46. Gerwin BI. Cytokine signaling in mesothelial cells: Receptor
activators, tissue degradation, and cancer. Adv Cancer Res 1985; expression closes the autocrine loop. Am J Respir Cell Mol Biol
44: 139–266. 1996; 14: 505–7.
◆24. Chapman HA. Plasminogen activators, integrins, and the 47. Fujino S, Yokoyama A, Kohno N, Hiwada K. Interleukin 6 is an
coordinated regulation of cell adhesion and migration. Curr Opin autocrine growth factor for normal human pleural mesothelial
Cell Biol 1997; 9: 714–24. cells. Am J Respir Cell Mol Biol 1996; 14: 508–515.
25. Abraham E. Coagulation abnormalities in acute lung injury and 48. Fitzpatrick DR, Peroni DJ, Bielefeldt-Ohmann H. The role of growth
sepsis. Am J Respir Cell Mol Biol 2000; 22: 401–4. factors and cytokines in the tumorigenesis and immunobiology of
◆26. Vassalli JD, Sappino AP, Belin D. The plasminogen malignant mesothelioma. Am J Respir Cell Mol Biol 1995; 12:
activator/plasmin system. J Clin Invest 1991; 88: 1067–72. 455–60.
27. Strange C, Baumann MH, Sahn SA, Idell S. Effects of intrapleural 49. Antony VB, Hott JW, Kunkel SL, et al. Pleural mesothelial cell
heparin or urokinase on the extent of tetracycline-induced pleural expression of C-C (monocyte chemotactic peptide) and C-X-C
disease. Am J Respir Crit Care Med 1995; 151: 508–15. (Interleukin 8) chemokines. Am J Respir Cell Mol Biol 1995; 12:
28. Stevens T, Garcia JG, Shasby DM, et al. Mechanisms regulating 581–8.
endothelial cell barrier function. Am J Physiol Lung Cell Mol 50. Shetty S, Kumar A, Pueblitz S, et al. Fibrinogen promotes adhesion
Physiol 2000; 279: L419–22. of monocytic to human mesothelioma cells. Thromb Haemost
◆29. Narayanan S. Multifunctional roles of thrombin. Ann Clin Lab Sci 1996; 75: 782–90.
1999; 29: 275–80. 51. Bajaj MS, Pendurthi U, Koenig K, et al. Tissue factor pathway
30. Dang CV, Bell WR, Kaiser D, Wong A. Disorganization of cultured inhibitor expression by human pleural mesothelial and
vascular endothelial cell monolayers by fibrinogen fragment D. mesothelioma cells. Eur Respir J 2000; 15: 1069–78.
Science 1985; 227: 1487–90. 52. Sabharwal AK, Bajaj SP, Ameri A, et al. Tissue factor pathway
31. Edgington TS, Curtiss LK, Plow EF. A linkage between the inhibitor and von Willebrand factor antigen levels in adult
hemostatic and immune systems embodied in the fibrinolytic respiratory distress syndrome and in a primate model of sepsis. Am
release of lymphocyte suppressive peptides. J Immunol 1985; 134: J Respir Crit Care Med 1995; 151: 758–67.
471–7. 53. Bertozzi P, Astedt B, Zenzius L, et al. Depressed bronchoalveolar
32. Ciano PS, Colvin RB, Dvorak AM, et al. Macrophage migration in urokinase activity in patients with adult respiratory distress
fibrin gel matrices. Lab Invest 1986; 54: 62–70. syndrome. N Engl J Med 1990; 322: 890–7.
33. Rowland FN, Donovan MJ, Picciano PT, et al. Fibrin-mediated 54. Idell S, James KK, Levin EG, et al. Local abnormalities in
vascular injury. Identification of fibrin peptides that mediate coagulation and fibrinolytic pathways predispose to alveolar fibrin
endothelial cell retraction. Am J Pathol 1984; 117: 418–28. deposition in the adult respiratory distress syndrome. J Clin Invest
34. Colvin RB, Gardner PI, Roblin RO, et al. Cell surface fibrinogen- 1989; 84: 695–705.
fibrin receptors on cultured human fibroblasts. Association with 55. Idell S, Koenig KB, Fair DS, et al. Serial abnormalities of fibrin
fibronectin (cold insoluble globulin, LETS protein) and loss in SV40 turnover in evolving adult respiratory distress syndrome. Am J
transformed cells. Lab Invest 1979; 41: 464–73. Physiol 1991; 261: L240–8.
35. Idell S, Girard W, Koenig KB, et al. Abnormalities of pathways of 56. Nakstad B, Lyberg T, Skjonsberg OH, Boye NP. Local activation of
110 Pleural fibrosis
the coagulation and fibrinolysis systems in lung disease. Thromb procoagulant and fibrinolytic gene expression during bleomycin-
Res 1990; 57: 827–38. induced lung injury in the mouse. J Clin Invest 1995; 96: 1621–30.
57. Saldeen T. The microembolism syndrome, a review. In: The 79. Eitzman DT, McCoy RD, Zheng X, et al. Bleomycin-induced
microembolism syndrome. Stockholm: Almqvist and Wiskell, 1979: pulmonary fibrosis in transgenic mice that either lack or
7–44. overexpress the murine plasminogen activator inhibitor-1 gene. J
58. Welty-Wolf KE, Carraway MS, Miller DL, et al. Coagulation Clin Invest 1996; 97: 232–7.
blockade prevents sepsis-induced respiratory and renal failure in 80. Idell S, Kumar A, Zwieb C, et al. Effects of TGF-beta and TNF-alpha
baboons. Am J Resp Crit Care Med 2001; 164: 1988–96. on procoagulant and fibrinolytic pathways of human tracheal
59. Mazar AP, Henkin J, Goldfarb RH. The urokinase plasminogen epithelial cells. Am J Physiol 1994; 267: L693–703.
activator system in cancer: implications for tumor angiogenesis 81. Li XY, Brown GM, Lamb D, Donaldson K. Increased production of
and metastasis. Angiogenesis 1999; 3: 15–32. plasminogen activator inhibitor in vitro by pleural leukocytes from
60. Higazi AA, Bdeir K, Hiss E, et al. Lysis of plasma clots by urokinase- rats intratracheally instilled with crocidolite asbestos. Environ Res
soluble urokinase receptor complexes. Blood 1998; 92: 2075–83. 1991; 55: 135–44.
61. Agrenius V, Chmielewska J, Widstrom O, Blomback M. Pleural 82. Donaldson K, Brown GM, Bolton RE, Davis JM. Fibrinolysis by rat
fibrinolytic activity is decreased in inflammation as demonstrated mesothelial cells in vitro: the effect of mineral dusts at non-toxic
in quinacrine pleurodesis treatment of malignant pleural effusion. doses. Br J Exp Pathol 1988; 69: 487–94.
Am Rev Respir Dis 1989; 140: 1381–5. 83. Bosma PJ, Kooistra T. Different induction of two plasminogen
62. Ellis V, Behrendt N, Dano K. Plasminogen activation by receptor- activator inhibitor 1 mRNA species by phorbol ester in human
bound urokinase: A kinetic study with both cell-associated and hepatoma cells. J Biol Chem 1991; 266: 17845–9.
isolated receptor. J Biol Chem 1991; 266: 12752–8. 84. Fattal PG, Schneider DJ, Sobel BE, Billadello JJ. Post-
63. Alfano D, Iaccarino I, Stoppelli MP. Urokinase signaling through its transcriptional regulation of expression of plasminogen activator
receptor protects against anoikis by increasing BCL-xL expression inhibitor type 1 mRNA by insulin and insulin-like growth factor 1.
levels. J Biol Chem 2006; 281: 17758–67. J Biol Chem 1992; 267: 12412–15.
64. Shetty S, Kumar A, Johnson A, et al. Urokinase receptor in human 85. Tillmann-Bogush M, Heaton JH, Gelehrter TD. Cyclic nucleotide
malignant mesothelioma cells: role in tumor cell mitogenesis and regulation of PAI-1 mRNAStability. Identification of cytosolic
proteolysis. Am J Physiol 1995; 268: L972–82. proteins that interact with an a-rich sequence. J Biol Chem 1999;
65. Shetty S, Kumar A, Johnson AR, Idell S. Regulation of mesothelial 274: 1172–9.
cell mitogenesis by antisense oligonucleotides for the urokinase 86. Shetty S, Idell S. Post-transcriptional regulation of plasminogen
receptor. Antisense Res Dev 1995; 5: 307–14. activator inhibitor-1 in human lung carcinoma cells in vitro. Am J
66. Sitrin RG, Todd RF III, Albrecht E, Gyetko MR. The urokinase Physiol Lung Cell Mol Physiol 2000; 278: L148–56.
receptor (CD87) facilitates CD11b/CD18-mediated adhesion of 87. Shetty S, Bdeir K, Cines DB, Idell S. Induction of plasminogen
human monocytes. J Clin Invest 1996; 97: 1942–51. activator inhibitor-1 by urokinase in lung epithelial cells. J Biol
67. Shetty S, Idell S. A urokinase receptor mRNA binding protein from Chem 2003; 278: 18124–31.
rabbit lung fibroblasts and mesothelial cells. Am J Physiol 1998; 88. Pedersen H, Brunner N, Francis D, et al. Prognostic impact of
274: L871–82. urokinase, urokinase receptor, and type 1 plasminogen activator
68. Shetty S, Kumar A, Johnson AR, et al. Differential expression of the inhibitor in squamous and large cell lung cancer tissue. Cancer Res
urokinase receptor in fibroblasts from normal and fibrotic human 1994; 54: 4671–5.
lungs. Am J Respir Cell Mol Biol 1996; 15: 78–87. 89. Pedersen H, Grondahl-Hansen J, Francis D, et al. Urokinase and
69. Shetty S, Idell S. Post-transcriptional regulation of urokinase plasminogen activator inhibitor type 1 in pulmonary
receptor gene expression in human lung carcinoma and adenocarcinoma. Cancer Res 1994; 54: 120–3.
mesothelioma cells in vitro. Mol Cell Biochem 1999; 199: 90. Heaton JH, Dlakic WM, Dlakic M, Gelehrter TD. Identification and
189–200. cDNA cloning of a novel RNA-binding protein that interacts with
70. Gyetko MR, Sitrin RG, Fuller JA, et al. Function of the urokinase the cyclic nucleotide-responsive sequence in the Type-1
receptor (CD87) in neutrophil chemotaxis. J Leukoc Biol 1995; 58: plasminogen activator inhibitor mRNA. J Biol Chem 2001; 276:
533–8. 3341–7.
71. Gyetko MR, Chen GH, McDonald RA, et al. Urokinase is required ◆91. Saksela O, Rifkin DB. Cell-associated plasminogen activation:
for the pulmonary inflammatory response to Cryptococcus regulation and physiological functions. Annu Rev Cell Biol 1988;
neoformans. A murine transgenic model. J Clin Invest 1996; 97: 4: 93–126.
1818–26. 92. Duggan C, Maguire T, McDermott E, et al. Urokinase plasminogen
72. Sprengers ED, Kluft C. Plasminogen activator inhibitors. Blood activator and urokinase plasminogen activator receptor in breast
1987; 69: 381–7. cancer. Int J Cancer 1995; 61: 597–600.
73. Perkins RC, Broaddus VC, Shetty S, et al. Asbestos upregulates 93. Blasi F, Vassalli JD, Dano K. Urokinase-type plasminogen activator:
expression of the urokinase-type plasminogen activator receptor proenzyme, receptor, and inhibitors. J Cell Biol 1987; 104: 801–4.
on mesothelial cells. Am J Respir Cell Mol Biol 1999; 21: 637–46. 94. Shetty S, Kumar A, Idell S. Post-transcriptional regulation of
74. Barchowsky A, Roussel RR, Krieser RJ, et al. Expression and activity urokinase receptor mRNA: identification of a novel urokinase
of urokinase and its receptor in endothelial and pulmonary receptor mRNA binding protein in human mesothelioma cells. Mol
epithelial cells exposed to asbestos. Toxicol Appl Pharmacol 1998; Cell Biol 1997; 17: 1075–83.
152: 388–96. 95. Shetty S, Muniyappa H, Halady PK, Idell S. Regulation of urokinase
75. Lenz SP, Green FH, Murphy PG, et al. Expression of plasminogen receptor expression by phosphoglycerate kinase. Am J Respir Cell
activator and plasminogen activator inhibitor by rat mesothelioma Mol Biol 2004; 31: 100–6.
induced by asbestos. Cancer Lett 1993; 68: 119–27. 96. Shetty S. Regulation of urokinase receptor mRNA stability by
76. Bajaj MS, Pendurthi U, Koenig K, et al. Tissue factor pathway hnRNP C in lung epithelial cells. Mol Cell Biochem 2005; 272:
inhibitor expression by human pleural mesothelial and 107–18.
mesothelioma cells. Eur Respir J 2000; 15: 1069–78. 97. Tran H, Maurer F, Nagamine Y. Stabilization of urokinase and
77. Barazzone C, Belin D, Piguet PF, et al. Plasminogen activator urokinase receptor mRNAs by HuR is linked to its cytoplasmic
inhibitor-1 in acute hyperoxic mouse lung injury. J Clin Invest accumulation induced by activated mitogen-activated protein
1996; 98: 2666–73. kinase-activated protein kinase 2. Mol Cell Biol 2003; 23:
78. Olman MA, Mackman N, Gladson CL, et al. Changes in 7177–88.
References 111
98. Shetty S, Idell S. Urokinase receptor mRNA stability involves intrapleural streptokinase in the treatment of multiloculated
tyrosine phosphorylation in lung epithelial cells. Am J Respir Cell pleural effusions and pleural empyemas. Thorac Cardiovasc Surg
Mol Biol 2004; 30: 69–75. 1996; 44: 252–6.
99. Shetty S, Idell S. Urokinase induces expression of its own receptor 118. Davies CW, Traill ZC, Gleeson FV, Davies RJ. Intrapleural
in Beas2B lung epithelial cells. J Biol Chem 2001; 276: streptokinase in the management of malignant multiloculated
24549–56. pleural effusions. Chest 1999; 115: 729–33.
100. Altus MS, Pearson D, Horiuchi A, Nagamine Y. Inhibition of protein 119. Kemper P, Kohler D. Current value of intrapleural fibrinolysis in the
synthesis in LLC-PK1 cells increases calcitonin-induced treatment of exudative fibrinous pleural effusions in pleural
plasminogen-activator gene transcription and mRNA stability. empyema and hemothorax. Pneumologie 1999; 53: 373–84.
Biochem J 1987; 242: 387–2. 120. Jerjes-Sanchez C, Ramirez-Rivera A, Elizalde JJ, et al. Intrapleural
101. Altus MS, Nagamine Y. Protein synthesis inhibition stabilizes fibrinolysis with streptokinase as an adjunctive treatment in
urokinase-type plasminogen activator mRNA. Studies in vivo and hemothorax and empyema: a multicenter trial. Chest 1996; 109:
in cell-free decay reactions. J Biol Chem 1991; 266: 21190–6. 1514–19.
102. Ziegler A, Knesel J, Fabbro D, Nagamine Y. Protein kinase C down- 121. de Gregorio MA, Ruiz C, Alfonso ER, et al. Drainage of loculated
regulation enhances cAMP-mediated induction of urokinase-type and/or multiloculated pleural effusions using a small caliber
plasminogen activator mRNA in LLC-PK1 cells. J Biol Chem 1991; catheter and urokinase (pleuro-fibrinolysis). Arch Bronconeumol
266: 21067–74. 1996; 32: 510–5.
103. Ziegler A, Hagmann J, Kiefer B, Nagamine Y. Ca2+ potentiates 122. Davies CW, Lok S, Davies RJ. The systemic fibrinolytic activity of
cAMP-dependent expression of urokinase-type plasminogen intrapleural streptokinase. Am J Respir Crit Care Med 1998; 157:
activator gene through a calmodulin- and protein kinase C- 328–330.
independent mechanism. J Biol Chem 1990; 265: 21194–201. 123. Berglin E, Ekroth R, Teger-Nilsson AC, William-Olsson G.
104. Nanbu R, Menoud PA, Nagamine Y. Multiple instability-regulating Intrapleural instillation of streptokinase. Effects on systemic
sites in the 3’ untranslated region of the urokinase-type fibrinolysis. Thorac Cardiovasc Surg 1981; 29: 124–6.
plasminogen activator mRNA. Mol Cell Biol 1994; 14: 4920–8. ◆124. Cameron R, Davies HR. Intra-pleural fibrinolytic therapy versus
105. Nanbu R, Montero L, D’Orazio D, Nagamine Y. Enhanced stability conservative management in the treatment of parapneumonic
of urokinase-type plasminogen activator mRNA in metastatic effusions and empyema. Cochrane Database Syst Rev 2004; 2:
breast cancer MDA-MB-231 cells and LLC-PK 1 cells down- CD002312.
regulated for protein kinase C-correlation with cytoplasmic ◆125. Cameron R. Intrapleural fibrinolytic therapy vs. conservative
heterogeneous nuclear ribonucleoprotein C. Eur J Biochem 1997; management in the treatment of parapneumonic effusions and
247: 169–74. empyema. Cochrane Database Syst Rev 2000; 3: CD002312.
106. Shetty S, Idell S. Post-transcriptional regulation of urokinase 126. Diacon AH, Theron J, Schuurmans MM, et al. Intrapleural
mRNA. Identification of a novel urokinase mRNA-binding protein streptokinase for empyema and complicated parapneumonic
in human lung epithelial cells in vitro. J Biol Chem 2000; 275: effusions. Am J Respir Crit Care Med 2004; 170: 49–53.
13771–9. ●127. Maskell NA, Davies CW, Nunn AJ, et al. UK Controlled trial of
107. Shetty S. Cytoplasmic-nuclear shuttling of the urokinase mRNA intrapleural streptokinase for pleural infection. N Engl J Med
binding protein regulates message stability. Mol Cell Biochem 2005; 352: 865–74.
2002; 237: 55–67. 128. Tokuda Y, Matsushima D, Stein GH, Miyagi S. Intrapleural
108. Rodriguez-Panadero F, Segado A, Martin Juan J, et al. Failure of fibrinolytic agents for empyema and complicated parapneumonic
talc pleurodesis is associated with increased pleural fibrinolysis. effusions: a meta-analysis. Chest 2006; 129: 783–90.
Am J Respir Crit Care Med 1995; 151: 785–90. 129. Idell S. Update on the use of fibrinolysins in pleural disease. Clin
109. Bouros D, Schiza S, Siafakas N. Utility of fibrinolytic agents for Pulm Med 2005; 12: 184–90.
draining intrapleural infections. Semin Respir Infect 1999; 14: 130. Lee YC. Ongoing search for effective intrapleural therapy for
39–47. empyema: is streptokinase the answer? Am J Respir Crit Care Med
110. Sahn SA. Use of fibrinolytic agents in the management of 2004; 170:1–2.
complicated parapneumonic effusions and empyemas. Thorax 131. Antoniou KM, Pataka A, Bourous D, Siafakas NM. Pathogenetic
1998; 53 (Suppl 2): S65–S72. pathways and novel pharmacotherapeutic targets in idiopathic
111. Bouros D, Schiza S, Patsourakis G, et al. Intrapleural streptokinase pulmonary fibrosis. Pulm Pharmacol Ther 2007; 20: 453–61.
versus urokinase in the treatment of complicated parapneumonic 132. Heffner JE. Multicenter trials of treatment for empyema – after all
effusions: a prospective, double-blind study. Am J Respir Crit Care these years. N Engl J Med 2005; 352: 926–8.
Med 1997; 155: 291–5. 133. Bouros D, Antoniou KM, Light RW. Intrapleural streptokinase for
112. Davies RJ, Traill ZC, Gleeson FV. Randomised controlled trial of pleural infection. BMJ 2006; 332: 133–4.
intrapleural streptokinase in community acquired pleural infection. 134. Oguzkaya F, Akcali Y, Bilgin M. Videothoracoscopy versus
Thorax 1997; 52: 416–21. intrapleural streptokinase for management of post traumatic
113. Tuncozgur B, Ustunsoy H, Sivrikoz MC, et al. Intrapleural urokinase retained haemothorax: a retrospective study of 65 cases. Injury
in the management of parapneumonic empyema: a randomised 2005; 36: 526–9.
controlled trial. Int J Clin Pract 2001; 55: 658–60. 135. Wait MA, Sharma S, Hohn J, Dal Nogare A. A randomized trial of
114. Bouros D, Schiza S, Tzanakis N, et al. Intrapleural urokinase versus empyema therapy. Chest 1997; 111: 1548–51.
normal saline in the treatment of complicated parapneumonic 136. Cochran JB, Tecklenburg FW, Turner RB. Intrapleural instillation of
effusions and empyema. A randomized, double-blind study. Am J fibrinolytic agents for treatment of pleural empyema. Pediatr Crit
Respir Crit Care Med 1999; 159: 37–42. Care Med 2003; 4: 39–43.
115. Chin NK, Lim TK. Controlled trial of intrapleural streptokinase in 137. Wells RG, Havens PL. Intrapleural fibrinolysis for parapneumonic
the treatment of pleural empyema and complicated effusion and empyema in children. Radiology 2003; 228: 370–8.
parapneumonic effusions. Chest 1997; 111: 275–9. 138. Yao CT, Wu JM, Liu CC, et al. Treatment of complicated
116. Basile A, Boullosa-Seoane E, Dominguez Viguera L, et al. parapneumonic pleural effusion with intrapleural streptokinase in
Intrapleural fibrinolysis in the management of empyemas and children. Chest 2004; 125: 566–71.
haemothoraces. Our experience. Radiol Med (Torino) 2003; 105: 139. Barnes NP, Hull J, Thomson AH. Medical management of
12–16. parapneumonic pleural disease. Pediatr Pulmonol 2005; 39:
117. Laisaar T, Puttsepp E, Laisaar V. Early administration of 127–34.
112 Pleural fibrosis
140. Chen JP, Lue KH, Liu SC, et al. Intrapleural urokinase treatment in receptor-bound urokinase mediates tissue invasion of the human
children with complicated parapneumonic effusion. Acta Paediatr lung carcinoma cell lines A549 and Calu-1. Cancer Res 1992; 52:
Taiwan 2006; 47: 61–6. 3043–7.
141. Thomson AH, Hull J, Kumar MR, et al. Randomised trial of 151. Bianchi E, Cohen RL, Thor AT, et al. The urokinase receptor is
intrapleural urokinase in the treatment of childhood empyema. expressed in invasive breast cancer but not in normal breast
Thorax 2002; 57: 343–7. tissue. Cancer Res 1994; 54: 861–6.
142. Singh M, Mathew JL, Chandra S, et al. Randomized controlled trial 152. Achbarou A, Kaiser S, Tremblay G, et al. Urokinase overproduction
of intrapleural streptokinase in empyema thoracis in children. Acta results in increased skeletal metastasis by prostate cancer cells in
Paediatr 2004; 93: 1443–5. vivo. Cancer Res 1994; 54: 2372–7.
●143. Kurt BA, Winterhalter KM, Connors RH, et al. Therapy of 153. Crowley CW, Cohen RL, Lucas BK, et al. Prevention of metastasis
parapneumonic effusions in children: video-assisted thoracoscopic by inhibition of the urokinase receptor. Proc Natl Acad Sci U S A
surgery versus conventional thoracostomy drainage. Pediatrics 1993; 90: 5021–5.
2006; 118: e547–53. 154. Min HY, Doyle LV, Vitt CR, et al. Urokinase receptor antagonists
●144. Sonnappa S, Cohen G, Owens CM, et al. Comparison of urokinase inhibit angiogenesis and primary tumor growth in syngeneic mice.
and video-assisted thoracoscopic surgery for treatment of Cancer Res 1996; 56: 2428–33.
childhood empyema. Am J Respir Crit Care Med 2006; 174: 221–7. 155. Shetty S, Idell S. A urokinase receptor mRNA binding protein-
145. Higazi AA, Mazar A, Wang J, et al. Single-chain urokinase-type mRNA interaction regulates receptor expression and function in
plasminogen activator bound to its receptor is relatively resistant human pleural mesothelioma cells. Arch Biochem Biophys 1998;
to plasminogen activator inhibitor type 1. Blood 1996; 87: 356: 265–79.
3545–9. 156. Vervloet MG, Thijs LG, Hack CE. Derangements of coagulation and
146. Vermeer F, Bosl I, Meyer J, et al. Saruplase Is a safe and effective fibrinolysis in critically ill patients with sepsis and septic shock.
thrombolytic agent; observations in 1,698 patients: results of the Semin Thromb Hemost 1998; 24: 33–44.
PASS Study. J Thromb Thrombolysis 1999; 8: 143–50. ◆157. ten Cate H. Pathophysiology of disseminated intravascular
147. Idell S, Mazar A, Cines D, et al. Single-chain urokinase alone or coagulation in sepsis. Crit Care Med 2000; 28: S9–11.
complexed to its receptor in tetracycline-induced pleuritis in 158. Esmon CT. Introduction: Are natural anticoagulants candidates for
rabbits. Am J Respir Crit Care Med 2002; 166: 920–6. modulating the inflammatory response to endotoxin? Blood 2000;
148. Idell S, Allen T, Chen S, et al. Intrapleural activation, processing, 95: 1113–16.
efficacy and duration of protection of single-chain urokinase in 159. de Fouw NJ, de Jong YF, Haverkate F, Bertina RM. Activated
evolving tetracycline-induced pleural injury in rabbits. Am J protein C increases fibrin clot lysis by neutralization of
Physiol Lung Cell Mol Physiol 2007; 292: L25–32. plasminogen activator inhibitor – no evidence for a cofactor role
149. Strange C, Baumann MH, Sahn SA, Idell S. Effects of intrapleural of protein S. Thromb Haemost 1988; 60: 328–33.
heparin or urokinase on the extent of tetracycline-induced pleural 160. Bajzar L, Nesheim ME, Tracy PB. The profibrinolytic effect of
disease. Am J Respir Crit Care Med 1995; 151: 508–15. activated protein C in clots formed from plasma is TAFI-
150. Bruckner A, Filderman AE, Kirchheimer JC, et al. Endogenous dependent. Blood 1996; 88: 2093–100.
10
Pleural reaction to mineral dusts
Introduction 113 Local and systemic reactions associated with talc 116
Pleural involvement by silica and non-fibrous silicates 113 pleurodesis
Dynamics of fibers and particles in lung and pleura 114 Key points 120
Mechanisms of mesothelial damage 114 References 120
minerals has not been shown to cause pleural malignan- material, dust would be carried towards the broncho-
cies. mediastinal lymphatic trunk, and backwards to the inter-
Fibrous silicates are more harmful for the pleura. costal vessels, leading to lymphangitis and reactive fibrosis.
Sepiolite exposure was associated with calcified pleural The finding of coal dust in intercostal vessels in cases of
plaque in Bulgaria,14 but there was no pleural involvement coal pneumoconiosis supports this theory.24
in 218 sepiolite workers studied in Spain.15 Erionite, of the
zeolite family, was related to fibrosis, plaques and
mesothelioma in Turkey16 and wollastonite produced MECHANISMS OF MESOTHELIAL DAMAGE
bilateral pleural thickening in 13 out of 46 (28 percent)
workers exposed for a mean of 22 years.17 The highest Asbestos
potential for injury to the pleura corresponds to asbestos.
Up to 82 percent of 2907 asbestos insulators with an expo- Asbestos consists of fibrous hydrated silicates which
sure time equal or greater than 40 years had pleural fibro- include serpentines and amphiboles (see Chapter 40,
sis18 (see also Chapter 40, Asbestos-related pleural Asbestos-related pleural diseases). The most common ser-
diseases). pentine is chrysotile, while amphiboles include crocidolite,
amosite, anthophilite, actinolite and tremolite. Chemi-
cally, the various asbestos fibers differ. Crocidolite con-
DYNAMICS OF FIBERS AND PARTICLES IN tains a considerable amount of iron, while chrysotile
LUNG AND PLEURA contains mostly magnesium. Although both amphiboles
and chrysotile are harmful in vitro for the mesothelial cell,
Inhaled mineral dust progresses through the airways crocidolite is the one mostly found in pleural samples of
depending on the physical characteristics of the particles patients with mesothelioma.25 This is thought to be due to
and the characteristics of the inhaled dust. Only particles the superior biopersistence of crocidolite in the lung com-
with a mean diameter of less than 5 mm reach the alveoli, pared with chrysotile (biopersistence being the physical
while larger particles are lodged at bronchial bifurcations. durability and chemical stability of fibers in tissue over a
Particles and fibers deposited in the respiratory tract are particular period of time). This is because chrysotile is
cleared by host defense mechanisms. The mucociliary soluble in organic fluids, more fragmentable and therefore
‘escalator’, comprising of ciliated epithelial cells and their eliminated more quickly. It is accepted that fibers measur-
covering of mucous fluid, helps remove the particles from ing less than 0.25 mm in diameter and more than 8 mm
the airways, where they are expectorated or swallowed. long are more carcinogenic.26
Uncleared particles are phagocytosed by alveolar Asbestos appears to damage the pleura by direct and
macrophages and epithelial cells and eliminated by ciliary indirect mechanisms. Inhalation or intratracheal adminis-
movement or via the lymphatic system, if the loaded tration of asbestos in rats induces epithelial cell prolifera-
macrophages penetrate the lung interstitium. Accumu- tion27 and phagocytosis of asbestos fibers in the first 24
lation of mineral deposits in the lung parenchyma appears hours.28 Mesothelial cells start to proliferate from 24 hours
to occur when the amount of inhaled dust exceeds the up to 14 days,27 while fibroblast proliferation persists for
capacity of the clearance system.19 6 weeks.29 In addition, inflammatory cells increase in
Part of the inhaled dust is deposited in the subpleural pleural lavage fluid at 1 week.29 Supernatants of alveolar
lung.20 Increased silicon deposition has been described in and pleural macrophages taken at 1 week stimulate prolif-
the visceral pleura of individuals with silicosis and in non- eration of mesothelial cells and fibroblasts, but by 6 weeks
silicotic individuals exposed to silica dust compared with only fibroblasts are stimulated. Fibroblast growth was
the reference population.21 In asbestos-exposed workers reduced by antibodies against platelet-derived growth
chrysotile is always found in the pleura; in contrast, factor (PDGF), whereas mesothelial proliferation was
amphiboles, the most pathogenic fibers, are rarely blocked by an antibody to keratinocyte growth factor
detected,22 see below. (KGF)29 (see also Chapter 8, Immunology). No correlation
The mechanism by which inhaled inorganic particles or was found between cell proliferation and the number of
fibers reach the pleura is not well understood. It has been fibers in the pleura,27 and asbestos fibers were not found in
proposed that asbestos fibers migrate towards the periph- pleural macrophages.29 Taken together, these data suggest
eral lung and visceral pleura, and the parietal pleura is sub- that mediators from asbestos-phagocytozing cells induce
sequently eroded during respiratory movements.23 This pleural accumulation of inflammatory cells and thereby
theory is not supported by the fact that crocidolite fibers, mesothelial cell proliferation. However, the studies men-
which are larger and more pathogenic than chrysotile tioned do not support the idea that asbestos directly stim-
fibers, and thus more able to irritate the parietal pleura, are ulates the mesothelium.
rarely found in pleural samples of exposed individuals.22 Since a percentage of asbestos fibers reach the pleura
Particles or fibers could also reach the parietal pleura from after being inhaled, fiber–mesothelium interaction merits
the lung retrogadely by the lymphatic system. Once hiliar special mention. In vitro studies have shown that mesothe-
and mediastinal lymph nodes are overloaded by mineral lial cells can phagocytose asbestos fibers, which are then
Mechanisms of mesothelial damage 115
incorporated into phagosomes.30 Vitronectin, and perhaps lial cells.47 Moreover, increased DNA synthesis and apop-
other serum adhesive proteins, covers asbestos fibers and tosis are in dynamic balance in asbestos-exposed mesothe-
permit their binding to vitronectin receptors of the lial cells.37 Apoptosis is the key mechanism for eliminating
mesothelial cell: integrins avb3 and avb5. The covering by mesothelial cells damaged by asbestos, ensuring the
vitronectin increases asbestos fiber phagocytosis, and removal of cells carrying genetic abnormalities with poten-
could therefore increase the tissue damage via an oxidative tial harmful biological effects.
effect.31 In vivo, phagocytosis of asbestos fibers probably Attention has recently been paid to the role of ROS in
occurs at the apical domain of mesothelial cells, which is cell toxicity caused by asbestos. Iron present in the surface
the region of the cell membrane facing the pleural space. of asbestos fibers and other silicates drive local generation
In addition to internalizing fibers, mesothelial cells of ROS via redox reactions.48 Oxidation also occurs after
exposed to asbestos would recruit inflammatory cells by the respiratory burst of phagocytosis of large asbestos
secreting interleukin (IL)-8 and monocyte chemoattrac- fibers.44 The participation of ROS in asbestos mesothelial
tant protein-1 (MCP-1).32–34 This process is potentiated in damage has been suggested by recent in vitro studies.
the presence of IL-1 and MCP-1. It has been shown that Culture medium from asbestos-exposed mesothelial cells
asbestos exposure increases the adhesion of pleural leuko- contains dose-dependent increases of oxidant-induced
cytes to rat pleural mesothelial cells, and upregulates the base modifications such as 8-oxo-2¢-deoxyguanosine and
expression of vascular cell adhesion molecule-1 (VCAM- 8-hydroxydeoxyguanosine, thereby indicating DNA
1) on rat pleural mesothelial cells.35 This effect, which is damage.49,50 However, overexpression of the antioxidant
dose-dependent and potentiated by IL-1, supports the idea enzymes heme oxygenase and manganese-containing
that asbestos induces an early pleural cellular response superoxide dismutase (Mn-SOD) in asbestos-exposed
leading to additional oxidative and inflammatory human mesothelial cells suggests a reaction against oxida-
damages. Finally, mesothelial cells can also recruit fibro- tive damage.51
blasts by secreting fibronectin and promote collagen syn- Recent in vitro data have contributed to the elucida-
thesis.36 The pathogenesis of benign asbestos pleural tion of the signal transduction cascade, through which
diseases is probably the result of a pleural inflammation asbestos fibers regulate gene expression in mesothelial
involving the direct and indirect mechanisms mentioned. cells. Phosphorylation of the epidermal growth factor
Pleural inflammation can provoke collagen deposition, as receptor (EGF-R) on mesothelial cells by asbestos, espe-
occurs in plaques, fibrosis and rounded atelectasis of the cially the longest fibers,52 appears to be the initial step,
pleura. Alternatively, asbestos-induced pleural inflamma- since it is known to induce cell proliferation.53
tion can lead to an increase in the capillary permeability in Phosphorylation of the mitogen-activated protein (MAP)
the pleura, resulting in the development of an effusion. kinases and extracellular signal-regulated kinases (ERK) 1
The reason(s) why any patient develops certain type(s) of and 2 would then occur, and these activated kinases
asbestos-induced pleural disease but not the others is would translocate to the nucleus to induce c-fos transac-
unknown. tivation.54 In a recent study, crocidolite was shown to be
Other in vitro effects of asbestos on mesothelial cells able to induce activation of the p38 arm of the MAP
include cell proliferation,37 cytotoxicity,38 genetic changes pathway in rat mesothelial cells.55
such as aneuploidy and diverse chromosomal alterations,39 After activation by MAPK signaling cascades, the prolif-
mitotic damage consisting of anaphase and telophase erative effect of asbestos on mesothelial cells can be
abnormalities,40 and DNA breakage (see also Chapter 11, mediated by upregulation of the early response proto-
Genetics of malignant mesothelioma).41,42 However, oncogenes c-fos and c-jun.56 The c-fos and c-jun proteins
asbestos-related genetic effects are not necessarily irre- form the AP-1 transcription factor, which binds to AP-1
versible or transmitted to daughter cells. Data from other DNA sites and regulates transition of the G1 to S phase of
studies reflect a regulatory reaction of mesothelial cells the cell cycle.57
after previous asbestos genetic damage. The finding that It is probable that biopersistence of asbestos, mainly
asbestos-exposed rat pleural mesothelial cells are arrested amphiboles, in lung and pleura cause a low-grade but sus-
in G0/G1, G1/S and G2/M stages43,44 suggests that the cell tained inflammation and cell proliferation via c-fos and
cycle of damaged cells is interrupted to permit repairs. c-jun and protein kinase C (PKC) pathways. Regulatory
Conversely, expression of the DNA-repairing enzyme mechanisms such as DNA repair and apoptosis would be
apurynic/apyrimidinic (AP) endonuclease in mesothelial bypassed, favoring uncontrolled cell growth in certain
cells by crocidolite44 further supports the idea that the cases and after a long latency period.
asbestos-damaged DNA may be repaired.45 Alternative mechanisms of DNA modification have
Apoptosis, or programmed cell death, is thought to be been proposed. Mechanical interference of asbestos fibers
a physiological mechanism for eliminating damaged cells, with chromosomes, including severing, piercing or disrup-
and is therefore an important cell-regulating phenome- tion of the mitotic spindle, are known to occur.30,40
non. Asbestos-damaged mesothelial cells undergo apopto- Another possibility is transfection of exogenous DNA into
sis,46,47 and this effect is mediated by reactive oxygen mesothelial cells. Transfection of plasmid DNA into
species (ROS) and by internalization of fibers by mesothe- mesothelial cells by asbestos fibers has been carried out in
116 Pleural reaction to mineral dusts
vitro,58 but the role of this mechanism in asbestos pleural contact with particles translocated to the pleura21 or indi-
disease remains unknown. rectly through pulmonary mediators and influx of inflam-
matory cells. Oxidant mechanisms are probably involved
in direct mesothelial damage by silica and non-fibrous sil-
Man-made vitreous fibers icates.72
To our knowledge, non-fibrous particles are not toxic
Man-made or synthetic mineral fibers include slag, rock or to mesothelial cells73 and do not produce genetic changes.
glass wools, continuous filament glass and refractory Mesothelioma has not been associated with silica or non-
ceramic fibers. These fibers are mainly used for thermal asbestos silicate human inhalation, and this tumor has not
and acoustic insulation. In vitro studies show that expo- developed in animals after intrapleural administration of
sure to man-made vitreous fibers (MMVF) induces cyto- quartz.74
toxicity,38 cell division,59 cell proliferation60 and oxidative
DNA damage61 in mesothelial cells. In vivo, intrapleural
introduction of MMVF with a high length–diameter ratio
causes malignant mesothelioma, but these studies cannot LOCAL AND SYSTEMIC REACTIONS
predict human carcinogenicity, owing to differences in ASSOCIATED WITH TALC PLEURODESIS
dose, fiber length and their pleural deposition.62 In animal
studies (using the inhalatory route), only refractory Background
ceramic fibers are able to induce mesothelioma.63
The occurrence of pleural plaques in workers exposed Talc is currently the most common agent used for pleu-
to refractory ceramic fibers, with greater risk associated rodesis. Talc pleurodesis can be performed either by insuf-
with latency time, duration, and intensity of exposure, has flating talc powder by thoracoscopy or by introducing a
recently been described either in cross-sectional64 and lon- slurry through a chest tube. Despite its high efficacy, con-
gitudinal studies.65 However, the industrial use of refrac- cerns are now focused on the possible development of
tory ceramic fibers is relatively recent and further studies acute respiratory distress syndrome (ARDS) after
with longer latency period are required to ascertain the intrapleural talc administration.
potential to cause benign pleural disease of these fibers. Although talc is used worldwide for pleurodesis, the
Regarding mesothelioma in humans, there are no data cellular and molecular basis of talc-induced pleural sym-
supporting the hypothesis that it can occur as a conse- physis has not been fully established. Local and systemic
quence of the exposure to MMVF.66 In conclusion, we inflammatory reactions, extrapleural talc dissemination,
believe that the International Association for Research on and their relationship with talc particle characteristics and
Cancer’s statement that MMVF are a possible human car- the dose used, are of special interest (see also Chapter 46,
cinogen, with classification in group 2B,67 remains valid. Pleurodesis).
This category is used to classify agents with limited or
inadequate evidence of carcinogenicity in humans, and
sufficient or insufficient evidence of carcinogenicity in Talc particle characteristics
animals. The lower pathogenicity of MMVF compared
with asbestos could be due to their lower lung biopersis- Talc is a pulverized, natural, sheet-like, hydrated magne-
tence.68 sium silicate with the approximate chemical formula of
Mg3(Si2O5)2(OH)2, although calcium, manganese, alu-
minum and iron are always present in variable amounts
Silica and non-asbestos silicates depending on the geographical origin of the talc. Non-talc
minerals associated with commercial talc vary from
The most important clinical effect of inhaled silica and deposit to deposit and may include calcite, magnesite,
non-asbestos silicates in the pleura is pleural fibrosis. dolomite, chlorite, serpentine, quartz and others.75
However, information on the effects of silica and non- During the treatment process, talc ore is first crushed
asbestos silicates on the mesothelium is scarce. and ground to a fineness which liberates it from other
Intratracheal silica instillation induces mesothelial prolif- associated non-talc minerals. After washing, the talc is
eration.69 In cultured mesothelial cells, silica exposure passed through mesh to eliminate the larger-sized talc par-
reduces protein and collagen synthesis70 and fibrinolytic ticles. The final talc may be 200-, 325- or 400-mesh. With
activity,71 and supernatant of silica-treated mesothelial a 400-mesh, 90–95 percent of the particles are smaller than
cells increase collagen synthesis in mesothelial cells and 37 mm, while with a 200-mesh 95–99 percent of the parti-
fibroblasts.70 cles are under 74 mm. In a study by Ferrer et al.,75 particle
It is likely, therefore, that mesothelial cells can drive size of talcs used for pleurodesis from different countries
pleural inflammation in individuals exposed to silica and was measured. The mean diameter ranged between 10.8
non-asbestos silicates. As mentioned previously for and 33.6 mm, and USA talcs had the lowest diameters
asbestos, the mesothelial reaction can be caused by direct (between 10.8 and 20.1 mm). The pattern of contaminant
Local and systemic reactions associated with talc pleurodesis 117
fibrinolytic cascades, such as thrombin-antithrombin III diameters (1–6 mm) uniformly distributed throughout the
complex and plasminogen activator inhibitor-1.98 nerve section. Nerves were always observed in association
Stabilization of this fibrin network99 and increased levels of with blood vessels, mainly arterioles, and followed a
basic fibroblast growth factor (bFGF) result in the recruit- sinuous course along the longitudinal axis of the adhe-
ment and proliferation of fibroblasts.100,101 Fibroblast pro- sion. It has been suggested that this association is a conse-
liferation is negatively regulated by transforming growth quence of the control role played by the angiogenic
factor (TGF)-b1, although, in contrast, this growth factor process in nerve growth during adhesion formation.104
promotes the synthesis of numerous matrix molecules Nerve fibers appeared to originate from the parietal
such as collagen, proteoglycans, fibronectin, throm- pleura, which is mainly innervated by the internal inter-
bospondin and osteonectin, and can simultaneously costal nerves (costal pleura and peripheral part of the
inhibit the expression of matrix metalloproteases. TGF-b diaphragmatic pleura) and the phrenic nerves (central
can be expressed by several cell effectors involved in the portion of the diaphragmatic pleura and mediastinal
pleurodesis, including the mesothelial cells, macrophages, pleura). Nerve fibers have also been found in murine105
endothelial cells and lymphocytes. The importance of this and human104,106–108 peritoneal adhesions. Although the
growth factor resides in its role in the regulation of the effect of innervation on adhesion function is unknown, an
fibrotic process. Light et al.102 and Lee et al.103 showed that interesting clinical question is whether talc-induced
intrapleural injections of TGF-b2 induced excellent pleu- pleural adhesions are potentially capable of conducting
rodesis in rabbits and in sheep, without producing an pain sensation.
inflammatory pleural effusion. Further, in vitro studies
revealed that TGF-b2 stimulated collagen synthesis in SYSTEMIC REACTIONS
primary cultures of rabbit pleural mesothelial cells.79 The
maturation of fibrotic tissue of the newly formed adhesion There is currently little information on the systemic effects
is associated with mesothelial re-epithelialization of their of talc pleurodesis. Available data indicate that a systemic
surface.96 (See Chapter 8, Immunology, for a detailed dis- inflammatory response can develop early after talc pleu-
cussion of cytokines in pleural diseases.) rodesis. In humans, the intrapleural administration of talc
Although adhesions are considered to play a critical has been associated with increases in both the percentage
role in the establishment of pleural symphysis, little is of leukocytes,109–111 the erythrocyte sedimentation rate110
known about their histopathogenesis. In a recent morph- and the levels of C-reactive protein (CRP)110–112 in periph-
ological and ultrastructural study following talc pleu- eral blood. Further, leukocytosis (34 ¥ 109/L) has been
rodesis, all rabbit pleural adhesions examined were reported in a patient suffering an acute respiratory failure
mesothelial-covered fibrovascular bands containing well- following talc pleurodesis.109 It is interesting to note that a
developed blood and lymphatic vessels establishing a recent trial involving 48 patients has shown that pleurode-
structural continuity between both pleural layers, more sis with mixed talc (50 percent of the particles <15 mm)
resembling newly formed pleural tissue than a simple produces more systemic inflammation and more hypox-
scar.96 Strikingly, nerves were present in adhesions from emia than pleurodesis with graded talc (50 percent of the
20 percent of the rabbits. They consisted of a single fasci- particles >25 mm).112 Nine of 22 (41 percent) patients
cle containing 5 to 20 thin myelinated axons of various receiving mixed talc developed fever, whereas fever was
Local and systemic reactions associated with talc pleurodesis 119
almost absent (1 of 24, 4 percent) in those receiving graded beyond the scope of this chapter. It has to be taken into
talc. In addition, the rise in plasma CRP was significantly account that detection of the total number of talc particles
greater after mixed-talc pleurodesis than it was after in a sample requires the use of electron microscopy,
graded-talc pleurodesis. because particles smaller than 1 mm can be missed by light
Data from animal models concur with these findings. microscopy. Particle analysis, in order to detect the char-
The intrapleural administration of talc in the rabbit model acteristic elemental composition of talc, is achieved by
has been associated with a transient increase in the count energy-dispersive X-ray analysis.
of leukocytes,80,81 the percentage80,81 and count95 of neu- After therapeutic pleurodesis with talc, some patients
trophils, the count of monocytes,95 the count of platelets,95 showed dissemination of particles in lung
the levels of IL-8,80,81 the levels of VEGF80,81 and the activ- parenchyma87,123 or in air spaces after bronchoalveolar
ity of angiotensin-converting enzyme91 in peripheral lavage (BAL).121,123,124 In one patient who died of ARDS
blood. Further, hyperplasia of the white periarteriolar sub- after talc pleurodesis, necropsy revealed talc crystals in
stance of the spleen was observed 24 h following almost every organ examined including lung, brain, liver,
200-mg/kg talc dose administration in rabbits.93 It has kidney, heart and skeletal muscle.123 However, no talc par-
been reported that this systemic inflammatory response is ticles have been found either in BAL samples128,129 or lung
a dose-related phenomenon.95 tissues130 from patients with no respiratory complications
after talc administration.
Several attempts to assess safety following intrapleural
Extrapleural talc deposition talc administration have been made in animal models.
Werebe et al.131 detected talc particles by light microscopy
Acute respiratory distress syndrome is a potentially lethal in every organ studied (chest wall, lungs, heart, brain,
respiratory failure that can develop very soon after talc spleen and kidney) in 100 percent of rats undergoing pleu-
administration either as a slurry113–118 or by insuffla- rodesis with two doses of talc (33 or 67 mg/kg). However,
tion115,118–124 in up to 9 percent of patients.115 In a recent Fraticelli et al.132 only found few talc particles in the liver
international survey of 859 pulmonologists, 58 percent (6 percent of total animals), spleen (3 percent) and brain
reported cases of respiratory failure after talc pleuro- (3 percent) after intrapleural administration of a 40-mg/kg
desis.125 Although the pathogenesis of talc-induced ARDS talc dose containing large particles in rats. No talc particles
is uncertain, it has been suggested that extrapleural talc were found in the blood, kidneys and contralateral lung.
dissemination results in cytokine release and increased With respect to the rabbit model, only some rabbits
permeability-type pulmonary edema.126,127 It has been undergoing experimental pleurodesis with a dose of
postulated that talc particles of small sizes are more likely 70 mg/kg showed talc dissemination to mediastinal lymph
to be absorbed via the parietal lymphatics into the systemic nodes (17 percent of total animals), kidney (17 percent)
circulation following intrapleural administration of talc. and spleen (40 percent), although no particles were
The detection of talc particles in a sample can be carried detected in lung parenchyma.88 In addition, it has been
out without destroying the sample or by removing the shown that in rabbits the probability of dissemination of
organic fraction by incineration or enzymatic degradation. talc particles to thoracic (lung, mediastinum and peri-
Methods to observe talc particles include light and electron cardium) and extrathoracic organs (liver, spleen and
microscopy, polarized light microscopy and X-ray diffrac- kidney) was related to dose95 and particle diameter93
tion, among others, but a description of each method is (Figure 10.2). Greater extrapleural talc dissemination was
(a) (b)
Figure 10.2 Extrapleural dissemination of talc to pulmonary parenchyma (a, ¥40) and mediastinum (b, ¥200). Talc particles are seen as
bright structures when observed by means of 45° polarized light microscopy.
120 Pleural reaction to mineral dusts
observed with the highest doses and smallest talc particle REFERENCES
size.
Taking into account this evidence, it remains possible ● = Key primary paper
that lung dissemination of talc particles leads to ARDS. ◆ = Major review article
However, the precise underlying causes and mechanisms
of post-talc ARDS are still unknown. 1. Weill H, Jones RN, Parkes WR. Silicosis and related diseases. In:
Parkes WR (ed.). Occupational lung diseases. London: Butterworth-
Heinemann, 1994: 285–339.
◆2. Craighead JE and the Silicosis and Silicate Disease Committee.
Diseases associated with exposure to silica and nonfibrous silicate
minerals. Arch Pathol Lab Med 1988; 112: 673–720.
3. Zeren EA, Colby TV, Roggli VL. Silica-induced pleural disease. An
KEY POINTS unusual case mimicking malignant mesothelioma. Chest 1997;
112: 1436–8.
4. Green FHY, Vallyathan V. Coal worker’s pneumoconiosis and
● Inhaled inorganic dusts can cause pleural disease.
pneumoconiosis due to other carbonaceous dusts. In: Churg A,
Asbestos and other fibrous silicates such as erion- Green FHY (eds). Pathology of occupational lung disease.
ite and wollastonite may induce benign and Baltimore: Williams & Wilkins, 1998:129–207.
malignant pleural diseases. Patients exposed to 5. Orriols R, Muñoz X, Sunyer J, et al. Radiologically recognized
non-fibrous minerals, such as silica, talc and pleural changes in nonpneumoconiotic silica-exposed coal miners.
Scand J Work Environ Health 2005; 31: 115–21.
mica, may develop pleural thickening and calcifi-
6. Hale LW, Gough J, King EJ, Nagelschmidt G. Pneumoconiosis of
cation. kaolin workers. Br J Ind Med 1956; 13: 251–9.
● Among various types of asbestos fibers, amphi- 7. Lapenas D, Gale P, Kennedy T, Rawlings W. Dietrich P. Kaolin
boles are more dangerous than chrysotile, pneumoconiosis. Radiologic, pathologic, and mineralogic findings.
because of their greater biopersistence in pul- Am Rev Respir Dis 1984; 130: 282–8.
8. Gibbs AR, Wagner JC. Diseases due to silica. In: Churg A, Green
monary tissue.
FHY (eds). Pathology of occupational lung disease. Baltimore:
● Upon ingestion of asbestos, phagocytes in the Williams & Wilkins, 1998: 209–33.
respiratory tract release mediators that induce 9. Vallyathan NV, Craighead JE. Pulmonary pathology in workers
mesothelial cell proliferation. Asbestos fibers are exposed to nonasbestiform talc. Hum Pathol 1981; 12: 28–35.
also endocytosed by mesothelial cells, triggering 10. Gamble JF, Fellner W, Dimeo MJ. An epidemiologic study of a
group of talc workers. Am Rev Respir Dis 1979; 119: 741–53.
the release of proinflammatory factors such as
11. Ross Smith A. Pleural calcification resulting from exposure to
IL-8 and MCP-1. certain dusts. Am J Roentgnol 1952; 67: 375–82.
● Direct or indirect asbestos-related mesothelial 12. Kleinfeld M. Pleural calcification as a sign of silicatosis. Am J Med
inflammation is probably involved in benign 1966; 251. 215–24.
asbestos pleural diseases. Asbestos can also 13. Skulberg KR, Gylseth B, Skaug V, Hanoa R. Mica pneumoconiosis. A
literature review. Scand J Work Environ Health 1985; 11: 65–74.
induce genetic changes. Sustained inflammation
14. Burilkov T, Michailova L. Sepiolite content of the soil in regions
and cell proliferation via c-fos, c-jun and protein with endemic pleural calcifications. Int Arch Arbeitsmed 1972; 29:
kinase C pathways would explain the carcino- 95–101.
genic effect of asbestos. 15. McConnochie K, Bevan C, Newcombe RG, et al. A study of spanish
● Talc is currently the most common agent used sepiolite workers. Thorax 1993; 48: 370–374.
16. Baris YI, Simonato L, Saracci R, Skidmore JW, Artvinli M.
for pleurodesis. Talc particle size and contami-
Malignant mesothelioma and radiological chest abnormalities in
nants vary markedly among talc preparations two villages in central Turkey. Lancet 1981; i: 984–7.
used in different countries. 17. Huuskonen MS, Tossavainen A, Koskinen H, et al. Wollastonite
● In the animal model, talc-induced pleural exposure and lung fibrosis. Environ Res 1983; 30: 291–304.
inflammation is driven by mesothelial cells and 18. Lilis R, Miller A, Godbold J, Chan E, Selikoff IJ. Radiographic
abnormalities in asbestos insulators: effects of duration from
macrophages, which release chemotactic factors
onset of exposure and smoking. Relationships of dyspnea with
for neutrophils and monocytes, e.g. IL-8 and parenchymal and pleural fibrosis. Am J Ind Med 1991; 20: 1–15.
MCP-1. Pleural inflammation results in the pro- 19. Rom WN, Crystal RG. Consequences of chronic inorganic dust
duction of a fibrinous exudate bridging both exposure. In: Crystal RG, West JB (eds). The lung. New York: Raven
pleural surfaces. This is followed by fibroblast Press, 1991: 1885–97.
20. Dodson RF, Ford JO. Early response to the visceral pleura following
infiltration and collagen deposition, which
asbestos exposure. J Toxicol Environ Health 1985; 15: 673–86.
results in adhesion formation and pleural sym- 21. Ferrer J, Orriols R, Tura JM, et al. Energy-dispersive X-ray analysis
physis. and scanning electron microscopy of pleura. Study of reference,
● After talc pleurodesis, talc particles can dissemi- non-pneumoconiotic and silicosis populations. Am J Respir Crit
nate to extrapleural organs including the lung. It Care Med 1994; 149: 888–92.
22. Sebastien P, Janson X, Gaudichet A. Asbestos retention in human
is possible that acute respiratory distress syn-
respiratory tissues. Comparative measurements in lung
drome induced after talc pleurodesis is caused by parenchyma and in parietal pleura. In: Wagner JC. (ed.). Biological
talc deposition in the lung. effects of mineral fibers, vol 1. Lyon: IARC Scientific publications
No 30, 1980: 237–46.
References 121
23. Heard BE, Williams R. The pathology of asbestosis with reference 44. Liu W, Ernst JD, Broaddus VC. Phagocytosis of crocidolite asbestos
to lung function. Thorax 1961; 16: 264–81. induces oxidative stress, DNA damage, and apoptosis in
●24. Taskinen E, Ahlman K, Wiikeri M. A current hypothesis of the mesothelial cells. Am J Respir Cell Mol Biol 2000; 23: 371–8.
lymphatic transport of inspired dust to the parietal pleura. Chest 45. Fung H, Kow YW, Van Houten B, et al. Asbestos increases
1973; 64: 193–6. mammalian AP-endonuclease gene expression, protein levels, and
25. Browne K. Asbestos-related disorders. In: Parkes WR (ed.). enzyme activity in mesothelial cells. Cancer Res 1998; 58:
Occupational lung diseases. London: Butterworth-Heineman, 189–94.
1994: 411–504. ●46. Broaddus VC, Yang L, Scavo LM, Ernst JD, Boylan AM. Asbestos
●26. Stanton MF, Layard M, Tegeris A, et al. Relation of particle induces apoptosis of human and rabbit pleural mesothelial cells
dimension to carcinogenicity in amphibole asbestoses and other via reactive oxygen species. J Clin Invest 1996; 98: 2050–9.
fibrous minerals. J Natl Cancer Inst. 1981; 67: 965–75. 47. BeruBe KA, Quinlan TR, Fung H, et al. Apoptosis is observed in
27. Sekhon H, Wright J, Churg A. Effects of cigarette smoke and mesothelial cells after exposure to crocidolite asbestos. Am J
asbestos on airway, vascular and mesothelial cell proliferation. Int Respir Cell Mol Biol. 1996;15:141–7.
J Exp Pathol 1995; 76: 411–8. 48. Weitzman SA, Graceffa P. Asbestos catalyzes hydroxyl and
28. Brodi AR, Hill LH, Adkins B, O’Connor RW. Chrysotile asbestos superoxide radical generation from hydrogen peroxide. Arch
inhalation in rats: deposition pattern and reaction of alveolar Biochem Biophys 1984; 228: 373–6.
epithelium and pulmonary macrophages. Am Rev Respir Dis 1981; 49. Chen Q, Marsh J, Ames B, Mossman BT. Detection of 8-oxo-2¢-
123: 670–9. deoxyguanosine, a marker of oxidative DNA damage, in culture
29. Adamson IY. Early mesothelial cell proliferation after asbestos medium from human mesothelial cells exposed to crocidolite
exposure: in vivo and in vitro studies. Environ Health Perspect asbestos. Carcinogenesis 1996; 17: 2525–7.
1997; 105 (Suppl 5): 1205–8. 50. Fung H, Kow YW, Van Houten B, Mossman BT. Patterns of 8-
30. Wang NS, Jaurand MC, Magne L, et al. The interactions between hydroxydeoxyguanosine formation in DNA and indications of
asbestos fibers and metaphase chromosomes of rat pleural oxidative stress in rat and human pleural mesothelial cells after
mesothelial cells in culture. A scanning electron microscopy study. exposure to crocidolite asbestos. Carcinogenesis 1997; 18:
Am J Pathol 1987; 126: 343–9. 825–32.
31. Boylan AM, Sanan DA, Sheppard D, Broaddus VC. Vitronectin 51. Janssen YM, Marsh JP, Absher MP, et al. Oxidant stress responses
enhaces internalization of crocidolite asbestos by rabbit pleural in human pleural mesothelial cells exposed to asbestos. Am J
mesothelial cells via the integrin alpha v beta 5. J Clin Invest Respir Crit Care Med. 1994; 149: 795–802.
1995; 96: 1987–2001. 52. Pache JC, Janssen YM, Walsh ES, et al. Increased epidermal growth
32. Tanaka S, Choe N, Iwagaki A, Hemenway DR, Kagan E. Asbestos factor-receptor protein in a human mesothelial cell line in
exposure induces MCP-1 secretion by pleural mesothelial cells. response to long asbestos fibers. Am J Pathol 1998;152: 333–40.
Exp Lung Res 2000; 26: 241–55. 53. Faux SP, Houghton CE, Hubbard A, Patrick G. Increased expression
33. Antony VB, Owen CL, Hadley KJ. Pleural mesothelial cells of epidermal growth factor receptor in rat pleural mesothelial cells
stimulated by asbestos release chemotactic activity for neutrophils correlates with carcinogenicity of mineral fibers. Carcinogenesis
in vitro. Am Rev Respir Dis 1898; 139: 199–206. 2000; 21: 2275–80.
34. Griffith DE, Miller EJ, Gray LD, Idell S, Johnson AR. Interleukin-1 54. Zanella CL, Posada J, Tritton TR, Mossman BT. Asbestos causes
mediated release of interleukin-8 by asbestos-stimulated human stimulation of the extracellular signal-regulated kinase Y
pleural mesothelial cells. Am J Respir Cell Mol Biol 1994; 10: mitogen-activated protein kinase after phosphorylation of the
245–52. epidermal growth factor receptor. Cancer Res 1996; 65: 5334–8.
35. Choe N, Zhang J, Iwagaki A, et al. Asbestos exposure upregulates 55. Swain WA, O’Byrne J, Faux SP. Activation of p38 MAP Kinase by
the adhesion of pleural leukocytes to pleural mesothelial cells via asbestos in rat mesothelial cells is mediated by oxidative stress.
VCAM-1. Am J Physiol 1999; 277: 292–300. Am J Physiol 2004; 286: 859–65.
36. Kuwahara M, Kuwahara M, Bijwaard KE, et al. Mesothelial cells ●56. Heintz NH., Janssen YM., Mossman BT. Persistent induction of c-
produce a chemoattractant for lung fibroblasts: role of fos and c-jun expression by asbestos. Proc Natl Acad Sci U S A
fibronectine. Am J Respir Cell Mol Biol 1991; 5: 256–64. 1993; 90: 3299–303.
37. Goldberg JL, Zanella CL, Janssen YM, et al. Novel cell imaging 57. Angel P, Karin M. The role of Jun, Fos and the AP-1 complex in
techniques show induction of apoptosis and proliferation in cell-proliferation and transformation. Biochem Biophys Acta 1991;
mesothelial cells by asbestos. Am J Respir Cell Mol Biol. 1997; 1072:129–57
17:265–71. 58. Gan L, Savransky EF, Fasy TM, Johnson EM. Transfection of human
38. Wang QE, Han CH, Wu WD, et al. Biological effects of man-made mesothelial cells mediated by different asbestos fiber types.
fibers (I) – Reactive oxygen species production and calcium Environ Res 1993; 62: 28–42.
homeostasis in alveolar macrophages. Ind Health 1999; 37: 62–7. 59. Pelin K, Kivipensas P, Linnainmaa K. Effects of asbestos and man-
●39. Lechner JF, Tokiwa T, LaVeck M, et al. Asbestos-associated made vitrous fibers on cell division in cultured human mesothelial
chromosomal changes in human mesothelial cells. Proc Natl Acad cells in comparison to rodent cells. Environ Mol Mutagen 1995;
Sci U S A. 1985; 82:3884–8. 25: 118–25.
40. Yegles M, Saint-Etienne L, Renier A, Janson X, Jaurand MC. 60. Rutten AA, Bermudez E, Mangum JB, et al. Mesothelial cell
Induction of metaphase/telophase abnormalities by asbestos fibers proliferation induced by intrapleural instillation of man-made
in rat pleural mesothelial cells in vitro. Am J Respir Cell Mol Biol fibers in rats and hamsters. Fundam Appl Toxicol 1994; 23:
1993; 9: 186–91. 107–16.
41. Levresse V, Renier A, Levy F, Broaddus VC, Jaurand M. DNA 61. Cavallo D, Campopiano A, Cardinally G, et al. Cytotoxic and
breakage in asbestos-treated normal and transformed (TSV40) rat oxidative effects induced by man-made vitreius fibers (MMVFs) in
pleural mesothelial cells. Mutagenesis 2000; 15: 239–44. human mesothelial cell line. Toxicology 2004; 201: 219–29.
42. Ollikainen T, Linnainmaa K, Kinnula VL. DNA single strand breaks 62. Gibbs AR, Wagner JC, Churg A. Diseases due to synthetic fibers. In:
induced by asbestos fibers in human pleural mesothelial cells in Churg A, Green FHY (eds). Pathology of occupational lung disease.
vitro. Environ Mol Mutagen 1999; 33: 153–60. Baltimore: Williams & Wilkins, 1998: 393–402.
43. Levresse V, Renier A, Fleury-Feith J, et al. Analysis of cell cycle 63. Mast RW, Hesterberg TW, Glass LR, et al. Chronic inhalation and
disruptions in cultures of rat pleural mesothelial cells exposed to biopersistence of refractory ceramic fiber in rats and hamsters.
asbestos fibers. Am J Respir Cell Mol Biol. 1997; 17: 660–71. Environ Health Perspect. 1994; 102(Suppl 5): 207–9.
122 Pleural reaction to mineral dusts
64. Lockey J, Lemasters G, Rice C, et al. Refractory ceramic fiber (interleukin 8) chemokines. Am J Respir Cell Mol Biol 1995; 12:
exposure and pleural plaques. Am J Respir Crit Care Med 1996; 581–8.
154:1405–10. 86. Mohammed KA, Nasreen N, Ward MJ, Antony VB. Macrophage
65. Lockey JE, Lemasters GK, Levin L, et al. A longitudinal study of inflammatory protein-1 C-C chemokine in parapneumonic pleural
chest radiographic changes of workers in the refractory ceramic effusions. J Lab Clin Med 1998; 132: 202–9.
fiber industry. Chest 2002; 121: 2044–51. 87. Krismann M, Pieper K, Müller K-M. Pleurale reaktionsmuster nach
◆66. De Vuyst P, Dumortier P, Swaen GMH, Pairon JC, Brochard P. talcum-pleurodese. Pathologe 1998; 19: 214–20.
Respiratory health effects man-made vitreous fibres. Eur Respir J 88. Kennedy L, Harley RA, Sahn SA, Strange C. Talc slurry pleurodesis.
1995; 8: 2149–73. Pleural fluid and histologic analysis. Chest 1995; 107: 1707–12.
◆67. IARC. Man-made mineral fibers and radon. In: IARC monographs 89. Light RW, Wang NS, Sassoon CSH, Gruer SE, Vargas FS. Talc slurry
in the evaluation of the carcinogenic risk of chemicals to humans, is an effective pleural sclerosant in rabbits. Chest 1995; 107:
vol. 43. Lyon: International Agency for Reasearch on Cancer, 1988. 1702–6.
68. Hesterberg TW, Miiler WC, Musselman RP, et al. Biopersistence of 90. Xie C, Teixeira LR, McGovern JP, Light RW. Systemic
man-made vitreous fibers and crocidolite asbestos in the rat lung corticosteroids decrease the effectiveness of talc pleurodesis. Am J
following inhalation. Fundam Appl Toxicol 1996; 29: 267–79. Respir Crit Care Med 1998; 157: 1441–4.
69. Adamson IYR, Bakowska J, Bowden DH. Mesothelial cell 91. Mitchem RE, Herndon BL, Fiorella RM, et al. Pleurodesis by
proliferation: a non-specific response to lung injury associated autologous blood, doxycycline, and talc in a rabbit model. Ann
with fibrosis. Am J Respir Cell Mol Biol 1994; 10: 253–8. Thorac Surg 1999; 67: 917–21.
70. Aalto M, Kulonen E, Penttinen R, Renvall S. Collagen synthesis in 92. Lee YC, Teixeira LR, Devin CJ, et al. Transforming growth factor-
cultured mesothelial cells. Response to silica. Acta Chir Scand beta2 induces pleurodesis significantly faster than talc. Am J
1981; 147: 1–6. Respir Crit Care Med 2001; 163: 640–4.
71. Donaldson K, Brown GM, Bolton RE, Davis JM. Fibrinolysis by rat 93. Ferrer J, Montes JF, Villarino MA, Light RW, García-Valero J.
mesothelial cells in vitro: the effect of mineral dusts at non-toxic Influence of particle size on extrapleural talc dissemination after
doses. Br J Exp Pathol 1988; 69: 487–94. talc slurry pleurodesis. Chest 2002; 122: 1018–27.
72. Ghio AJ, Kennedy TP, Schapira RM, Crumbliss AL, Hoidal JR. 94. Vargas FS, Teixeira LR, Antonangelo L, et al. Experimental
Hypothesis: is lung disease after silicate inhalation caused by pleurodesis in rabbits induced by silver nitrate or talc. 1-year
oxidant generation? Lancet 1990; 336: 967–9. follow-up. Chest 2001; 119: 1516–20.
73. Jaurand MC, Bastie-Sigeac Y, Renier A, Bignon J. Comparative 95. Montes JF, Ferrer J, Villarino MA, et al. Influence of talc dose on
toxicities of different forms of asbestos on rat pleural mesothelial extrapleural talc dissemination after talc pleurodesis. Am J Respir
cells. Environ Health Perspect 1983; 51: 153–8. Crit Care Med 2003; 168: 348–55.
74. Jaurand MC, Fleury J, Monchaux G, Nebut M, Bignon J. Pleural 96. Montes JF, García-Valero J, Ferrer J. Evidence of innervation in
carcinogenic potency of mineral fibers (asbestos, attapulgite) and talc-induced pleural adhesions. Chest 2006; 130: 702–9.
their cytotoxicity on culture cells. J Natl Cancer Inst 1987; 79: 97. Guo YB, Kalomenidis I, Hawthorne M, et al. Pleurodesis is inhibited
797–804. by anti-vascular endothelial growth factor antibody. Chest 2005;
75. Ferrer J, Villarino MA, Tura JM, Traveria A, Light RW. Talc 128: 1790–7.
preparations used for pleurodesis vary markedly from one 98. Rodríguez-Panadero F, Segado A, Martín Juan J, et al. Failure of
preparation to another. Chest 2001; 119: 1901–5. talc pleurodesis is associated with increased pleural fibrinolysis.
76. van den Heuvel MM, Smit HJM, Barbierato SB, et al. Talc-induced Am J Respir Crit Care Med 1995; 151: 785–90.
inflammation in the pleural cavity. Eur Respir J 1998; 12: ◆99. Kroegel C, Antony VB. Immunobiology of pleural inflammation:
1419–23. potential implications for pathogenesis, diagnosis and therapy. Eur
●77. Antony VB, Godbey SW, Kunkel SL, et al. Recruitment of Respir J 1997; 10: 2411–8.
inflammatory cells to the pleural space. Chemotactic cytokines, IL- 100. Antony VB, Kamal MA, Godbey S, Loddenkemper RW. Talc induced
8, and monocyte chemotactic peptide-1 in human pleural fluids. pleurodesis: role of basic fibroblast growth factor (bFGF). Eur
J Immunol 1993; 151: 7216–23. Respir J 1997; 10: 403S.
78. Miller EJ, Kajikawa O, Pueblitz S, et al. Chemokine involvement in 101. Antony VB, Nasreen N, Mohammed KA, et al. Talc pleurodesis.
tetracycline-induced pleuritis. Eur Respir J 1999; 14: 1387–93. Basic fibroblast growth factor mediates pleural fibrosis. Chest
79. Lee YCG, Lane KB, Zoia O, et al. Transforming growth factor-b 2004; 126: 1522–8.
induces collagen synthesis without inducing IL-8 production in 102. Light RW, Cheng D-S, Lee YC, et al. A single intrapleural injection
mesothelial cells. Eur Respir J 2003; 22: 197–202. of transforming growth factor-beta2 produces an excellent
80. Marchi E, Vargas FS, Acencio MMP, et al. Talc and silver nitrate pleurodesis in rabbits. Am J Respir Crit Care Med 2000; 162:
induce systemic inflammatory effects during the acute phase of 98–104.
experimental pleurodesis in rabbits. Chest 2004; 125: 2268–77. 103. Lee YC, Lane KB, Parker RE, et al. Transforming growth factor beta2
●81. Marchi E, Vargas FS, Teixeira LR, et al. Intrapleural low-dose silver (TGF-b2) produces effective pleurodesis in sheep with no systemic
nitrate elicits more pleural inflammation and less systemic complications. Thorax 2000; 55: 1058–62.
inflammation than low-dose talc. Chest 2005; 128: 1798–804. 104. Sulaiman H, Gabella G, Davis C, et al. Presence and distribution of
82. Marchi E, Vargas FS, Acencio MM, et al. Evidence that mesothelial sensory nerve fibers in human peritoneal adhesions. Ann Surg
cells regulate the acute inflammatory response in talc pleurodesis. 2001; 234: 256–61.
Eur Respir J 2006; 28: 929–32. 105. Sulaiman H, Gabella G, Davis C, et al. Growth of nerve fibres into
●83. Nasreen N, Hartman DL, Mohammed KA, Antony VB. Talc-induced murine peritoneal adhesions. J Pathol 2000; 192: 396–403.
expression of C-C and C-X-C chemokines and intercellular 106. Kligman I, Drachenberg C, Papadimitriou J, Katz E.
adhesion molecule-1 in mesothelial cells. Am J Respir Crit Care Immunohistochemical demonstration of nerve fibers in pelvic
Med 1998; 158: 971–8. adhesions. Obstet Gynecol 1993; 82: 566–8.
84. Green TR, Fisher J, Stone M, Wroblewski BM, Ingham E. 107. Tulandi T, Chen MF, Al-Took S, Watkin K. A study of nerve fibers
Polyethylene particles of a ‘critical size’ are necessary for the and histopathology of postsurgical, postinfectious, and
induction of cytokines by macrophages in vitro. Biomaterials endometriosis-related adhesions. Obstet Gynecol 1998; 92: 766–8.
1998; 19: 2297–302. 108. Herrick SE, Mutsaers SE, Ozua P, et al. Human peritoneal
85. Antony VB, Hott JW, Kunkel SL, et al. Pleural mesothelial cell adhesions are highly cellular, innervated, and vascularized. J
expression of C-C (monocyte chemotactic peptide) and C-X-C Pathol 2000; 192: 67–72.
References 123
109. Lineau C, Le Coz A, Quinquenel ML, et al. Insuffisance respiratoire pneumonitis with bilateral pleural effusion after talc pleurodesis.
aiguë après talcage pleural d’un pneumothorax. Rev Pneumol Clin Chest 1984; 86: 795–7.
1993; 49: 153–55. 122. Petrou M, Kaplan D, Goldstraw P. Management of recurrent
110. Ukale V, Agrenius V, Widström O, Hassan A, Hillerdal G. malignant pleural effusions. The complementary role of talc
Inflammatory parameters after pleurodesis in recurrent malignant pleurodesis and pleuroperitoneal shunting. Cancer 1995; 75:
pleural effusions and their predictive value. Respir Med 2004; 98: 801–5.
1166–72. 123. Milanez JR, Werebe EC, Vargas FS, Jatene FB, Light RW.
111. Froudarakis ME, Klimathianaki M, Pougounias M. Systemic Respiratory failure due to insufflated talc. Lancet 1997; 349:
inflammatory reaction after thoracoscopic talc poudrage. Chest 251–2.
2006; 129: 356–61. 124. Milanez JR, Vargas FS, Werebe EC, et al. Thoracoscopy talc
●112. Maskell NA, Lee YCG, Gleeson FV, et al. Randomized trials poudrage. A 15-year experience. Chest 2001; 119: 801–6.
describing lung inflammation after pleurodesis with talc of varying ●125. Lee YCG, Baumann MH, Maskell NA, et al. Pleurodesis practice for
particle size. Am J Respir Crit Care Med 2004; 170: 377–82. malignant pleural effusions in five English-speaking countries.
113. Rinaldo JE, Owens GR, Rogers RM. Adult respiratory distress Survey of pulmonologists. Chest 2003; 124: 2229–38.
syndrome following intrapleural instillation of talc. J Thorac 126. Light RW, Vargas FS. Pleural sclerosis for the treatment of
Cardiovasc Surg 1983; 85: 523–6. pneumothorax and pleural effusion. Lung 1997; 175: 213–23.
114. Kennedy L, Rusch VW, Strange C, Ginsberg RJ, Sahn SA. ◆127. Light RW. Diseases of the pleura: the use of talc for pleurodesis.
Pleurodesis using talc slurry. Chest 1994; 106: 342–6. Curr Opin Pulm Med 2000; 6: 255–8.
115. Rehse DH, Aye RW, Florence MG. Respiratory failure following talc 128. Boutin C, Mathlouthi A, Espinasse P, et al. Perméabilité alvéolaire
pleurodesis. Am J Surg 1999; 177: 437–40. au cours des épanchements et aprés symphyse pleurale au talc.
116. Brant A, Eaton T. Serious complications with talc slurry Rev Fr Mal Respir 1988; 5: R125.
pleurodesis. Respirology 2001; 6: 181–5. 129. Viallat JR, Rey F, Astoul P, Boutin C. Thoracoscopic talc poudrage
117. Bondoc AYP, Bach P, Vander Els N. Talc pneumonitis: incidence, pleurodesis for malignant effusions: a review of 360 cases. Chest
clinical features and outcome. Chest 1999; 116: 358S–9S. 1996; 110: 1387–93.
118. Dresler CM, Olak J, Herndon JE, et al. Phase III intergroup study of 130. Rodríguez-Panadero F. Current trends in pleurodesis. Curr Opin
talc poudrage vs talc slurry sclerosis for malignant pleural Pulm Med 1997; 3: 319–25.
effusion. Chest 2005; 127: 909–15. 131. Werebe EC, Pazetti R, Milanez JR, et al. Systemic distribution of
119. Nandy P. Recurrent spontaneous pneumothorax; an effective talc after intrapleural administration in rats. Chest 1999; 115:
method of talc poudrage. Chest 1980; 77: 493–5. 190–3.
120. Todd TR, Delarue NC, Ilves R, Pearson FG, Cooper JD. Talc poudrage 132. Fraticelli A, Robaglia-Schlupp A, Riera H, et al. Distribution of
for malignant pleural effusion. Chest 1980; 78: 542–3. calibrated talc after intrapleural administration. An experimental
121. Bouchama A, Chastre J, Gaudichet A, Soler P, Gibert C. Acute study in rats. Chest 2002; 122: 1737–41.
This page intentionally left blank
11
Genetic alterations in mesothelioma pathogenesis
Mechanisms of asbestos carcinogenicity 125 Erionite and possible genetic predisposition to malignant 131
Recurrent cytogenetic abnormalities 126 mesothelioma
Deletion mapping of recurrent chromosomal losses 126 Potential therapeutic targets in malignant mesothelioma 131
Involvement of tumor suppressor genes 128 Conclusions 132
An Nf2 mouse model recapitulating molecular features of 130 Key points 132
human malignant mesothelioma Acknowledgments 133
AKT/Akt kinase activation in malignant mesothelioma 131 References 133
Malignant mesothelioma (MM) is a tumor of adult life multistep tumorigenic process. We also discuss potential
that mostly affects individuals older than 50 years of age therapeutic approaches that target molecular alterations
and occurs more commonly in men than in women.1 found in this disease.
Approximately 3000 patients are diagnosed with MM in
the USA each year. MMs arise from serosal surfaces and
are approximately four times more common in the pleural MECHANISMS OF ASBESTOS
than in the peritoneal cavity. MM manifests as a diffuse or CARCINOGENICITY
localized growth, with the former accounting for approxi-
mately 75 percent of all cases.1 The increasing frequency of Whether asbestos fibers act directly on mesothelial cells or
this disease over the last 40 years is a reflection of exposure indirectly via induction of reactive oxygen species (ROS)
to asbestos fibers in industrialized countries, particularly and growth factors5,6 is not fully understood. In vitro
in connection with the mining and shipyard industries.2 A studies have demonstrated that asbestos can physically
history of asbestos exposure is associated with approxi- interact with the mitotic spindle apparatus, which could
mately 80 percent of the cases, and a lag phase of 20–40 result in aneuploidy and other forms of chromosome
years between exposure and tumor development is usual. damage.7 In vivo, however, iron-rich crocidolite asbestos
In pleural MM, the first symptoms are typically dyspnea fibers may lead to the release of ROS when hydrogen per-
and chest pain, and radiological examination often reveals oxide and superoxide react to form hydroxyl radicals.
a pleural effusion. Cytological discrimination between Asbestos fibers induce expression and enzymatic activa-
MM, benign inflammatory or reactive effusions, and tion of the mammalian DNA repair enzyme, apurinic/
metastatic carcinoma is not always clear.1 On the other apyrimidinic endonuclease, suggesting that ROS generated
hand, recurrent cytogenetic and molecular genetic abnor- by asbestos may produce DNA damage such as G to T
malities have been identified in MM, which can aid in dis- transversions.8 Recent evidence indicates that G to T trans-
criminating neoplastic disease from benign mesothelioma. versions, often induced by the premutagenic DNA adduct
Besides asbestos, some evidence has implicated Simian 8-hydroxydeoxyguanosine, are among the most prevalent
virus 40 (SV40) in the etiology of some MMs.3,4 (See also mutations detected following crocidolite treatment of
Chapter 41, Malignant mesothelioma.) omenta from lacI transgenic rats.9 Moreover, the inflam-
In this chapter, we provide an overview of both cytoge- matory response to asbestos leads to the generation of
netic and molecular genetic alterations in MM, mecha- various cytokines responsible for the local and systemic
nisms by which asbestos and other carcinogenic mineral immunosuppressive activity of asbestos.10 In addition,
fibers contribute to MM pathogenesis, and briefly discuss asbestos can induce autophosphorylation of the epidermal
potential signaling pathways by which alterations of growth factor (EGF) receptor, which results in increased
certain tumor suppressor genes (TSGs) play a role in the expression of c-fos and c-jun, which encode transcription
126 Genetic alterations in mesothelioma pathogenesis
factors that activate genes critical in the initiation of DNA documented in 26 cases (67 percent). These recurrent
synthesis.5 Induction of these transcription activators may losses of 1p, 3p, 6q, 9p, 17 and 22 frequently occurred in
enhance cellular proliferation and, thus, render cells more combination in a given tumor. The complexity of the cyto-
susceptible to subsequent mutations in TSGs. The genetic alterations observed suggest the emergence of
enhanced expression of proto-oncogenes and inactivation tumor progression-associated changes. However, because
of TSGs may cooperate in a multistep process leading to cytogenetic data are lacking for early mesothelial lesions, it
mesothelial cell oncogenesis. is not possible to distinguish between alterations associ-
In vivo, macrophages are known to phagocytose ated with tumor initiation and those associated with pro-
asbestos and, in response, release tumor necrosis factor gression of the disease. In any case, the accumulated loss of
alpha (TNF-a) and other cytokines that play a role in car- DNA sequences from chromosomes 1p, 3p, 6q, 9p, 17 and
cinogenesis through poorly understood mechanisms. In 22 appears to play a fundamental role in the pathogenesis
vitro, asbestos does not induce transformation of primary of MM.
human mesothelial cells, but instead causes extensive cell We also employed comparative genomic hybridization
death. Yang et al.11 recently reported that asbestos induces (CGH) analysis to detect recurrent genomic imbalances
the secretion of TNF-a and the expression of TNF-a in MM. CGH is a DNA-based molecular cytogenetic
receptor in primary human mesothelial cells.11 TNF-a was technique that permits the identification of chromosome
found to activate nuclear factor (NF)-kB and that NF-kB gains and losses in the entire tumor genome in a single
activation led to resistance to the cytotoxic effects of experiment. We performed metaphase-based CGH analy-
asbestos. TNF-a/NF-kB signaling increased cell survival ses on 24 MM cell lines derived from American
following asbestos exposure, permitting increased suscep- patients;15 each of these cell lines exhibited multiple
tibility of mesothelial cells to malignant transformation. (6–25) genomic imbalances. Losses involving 22q were
Cytogenetics supported this hypothesis, showing only rare, the most consistent change, being detected in 14 of 24 (58
aberrant metaphases in mesothelial cells exposed to percent) cell lines. The analyses also confirmed earlier
asbestos and an increased mitotic rate with fewer irregular conventional cytogenetic findings, with losses of 1p, 3p,
metaphases in mesothelial cells exposed to both TNF-a 6q and 9p each being detected in approximately 30–40
and asbestos. These data provide a mechanistic rationale percent of cell lines. Moreover, the CGH analysis uncov-
for the paradoxical inability of asbestos to transform ered other recurrent chromosome losses not detected by
mesothelial cells in vitro and suggest that TNF-a plays a conventional cytogenetic studies. Especially noteworthy,
significant role in asbestos pathogenesis in humans. 13 of 24 MMs (54 percent) showed losses of part or all of
15q, with the SRO being 15q11.1–21. Additionally, losses
of 14q24.2-qter and 13q12–14 were each observed in 42
RECURRENT CYTOGENETIC ABNORMALITIES percent of the cell lines. The most common overrepre-
sented chromosomal arm was 5p (54 percent of cases),
Conventional cytogenetic analysis of MMs, using chromo- suggesting the involvement of a putative oncogene(s) in
some banding techniques, has revealed that most MMs this region.
have complex karyotypes (reviewed in refs. 12 and 13). All Similar recurrent genomic imbalances were identified
of the 39 MMs we karyotyped (reviewed in ref. 14; A. Elahi in MM specimens from Finland.16 However, three promi-
and S.C. Jhanwar, unpublished data) displayed extensive nent imbalances reported in the series from the USA, i.e.
aneuploidy and structural rearrangements of various chro- losses of 15q11.1–21.1, 8p21-pter and 3p21, were each
mosomes, particularly the short (p) arms of chromosomes observed in only one of 42 Finnish cases. Discrepancies
1, 3 and 9 and the long (q) arm of chromosome 6. Loss of between the data from Finland and the USA may reflect
a copy of chromosome 22 is the single most consistent dissimilarities in the type of asbestos exposure or genetic
numerical change seen in MMs. In some series, losses or differences in the study populations.
rearrangements of chromosomes 4, 14 and 17 and gain of
chromosome 7 have also been commonly observed. In the
series by Taguchi et al.14 and an additional 19 cases ana- DELETION MAPPING OF RECURRENT
lyzed by A. Elahi and S.C. Jhanwar (unpublished data), CHROMOSOMAL LOSSES
deletions and unbalanced rearrangements accounted for
overlapping losses from the chromosome region 1p21–22 The pattern of recurrent genomic losses observed in MMs
in 32 of 39 (82 percent) cases. MMs also had interstitial is consistent with the possibility of a recessive mechanism
deletions or other rearrangements that resulted in losses of oncogenesis. The common sites of chromosomal loss
from 3p21. Twenty cases (51 percent) showed losses from are thought to represent the locations of putative TSGs
6q, with the shortest region of overlap (SRO) being that contribute to the development and/or progression of
6q15–21. Losses involving 9p were observed in 31 (79 MM. As an initial approach to isolate these putative TSGs,
percent) cases, with the SRO being 9p21–22. Loss or rela- the commonly deleted regions defined by cytogenetic
tive deficiency of chromosome 17 was detected in 11 of 39 studies of MM were mapped at the molecular genetic level
(28 percent) cases. Loss of a copy of chromosome 22 was by loss of heterozygosity (LOH) analysis using numerous
Deletion mapping of recurrent chromosomal losses 127
polymorphic DNA markers. Results of these investigations segment located between the interferon gene cluster and
have been reviewed in detail elsewhere12,13 and are briefly the marker D9S171 in 9p21.24 The CDKN2A/ARF locus,
summarized below. which encodes the alternative TSG products p16INK4a and
p14ARF, is located within this region. The cellular function
of p16INK4a and p14ARF and their potential role in MM
Chromosome 1p22 pathogenesis are discussed below.
allelic losses from both arms, indicating that concurrent INVOLVEMENT OF TUMOR SUPPRESSOR
loss from both arms of chromosome 17 may be common in GENES
MM. The relative deficiencies of either 17p or whole chro-
mosome 17 were not detected by CGH analysis; the reason p16INK4a
for this appears to be that abnormalities of chromosome 17
arise by either uniparental disomy or trisomy, as deter- One product of the CDKN2A/ARF locus, p16INK4a, is
mined by cytogenetic and restriction fragment length capable of binding to and inhibiting the cyclin-dependent
polymorphism (RFLP) analyses performed on the same kinase CDK4. Shortly after being cloned, the p16INK4a gene
specimens (SC Jhanwar and A Elahi, unpublished data). was identified as the 9p21 TSG, and homozygous deletions
Recurrent cytogenetic and molecular genetic abnor- of p16INK4a were detected at high frequencies in cell lines
malities observed in MM are summarized in Table 11.1. A derived from various types of cancer.26 To assess the possi-
depiction of frequently deleted chromosomal regions and ble involvement of p16INK4a in MM, we performed deletion
the location of TSGs known or potentially involved in MM mapping studies of 40 MM cell lines;27 34 (85 percent) of
is presented in Figure 11.1. the lines had homozygous deletions of one or more
Table 11.1 Summary of allelic losses and tumor suppressor genes associated with multistep tumorigenesis
in human malignant mesothelioma
1p22 72 –
3p21.3 63 –
6q14-25 60 –
9p21 83b p16INK4a, p14ARF, p15INK4b
13q13.2–14.2 67 –
15q15 48 –
17p13 40 TP53
17q21–25 70 –
22q12 72 NF2
aPercentages shown reflect incidence of allelic loss observed in mesothelioma cell lines examined by the authors. Note:
each of these common sites of allelic loss was confirmed in a subset of cases for which corresponding tumor tissue was
available.
bAt the p16INK4a/p14ARF locus, 85 percent of cell lines exhibited homozygous losses, nearly all of which affected both
p16INK4a and p14ARF; 78 percent of these cell lines also showed homozygous loss of p15INK4b.
p21.3
p22
MTAP
p16 p21 p53 p13
p14ARF
1 3 6 9 13 14 15 17 22
Figure 11.1 Idiograms of chromosomes frequently altered in malignant mesothelioma (MM), indicative of multistep tumorigenesis in
this disease. Brackets demarcate minimally deleted regions (SROs) in each chromosome. Locations of TSGs (p14ARF, p16INK4a, TP53 and NF2)
known to be either mutated or homozygously deleted in MM are shown; other candidate genes (MTAP, NM23) are also indicated. (See also
Color Plate 4.)
Involvement of tumour suppressor genes 129
p16INK4a exons and another had a point mutation in by immunohistochemistry in 8 of 20 (40 percent) cell lines
p16INK4a. Downregulation of p16INK4a was observed in four examined.
of the remaining cell lines. Homozygous deletions of While the TP53 gene is associated with 17p loss, the
p16INK4a were identified in 5 of 23 (22 percent) MM tumor putative TSG involved in 17q losses in MM is not known.
specimens. The higher frequency of p16INK4a alterations in The human nm23 gene (NME1), located at 17q21.3, repre-
MM cell lines than in tumor samples may be associated sents one candidate gene in this region. Reduced expres-
with a selective growth advantage provided by p16INK4a loss sion of nm23 is associated with a high potential for
during cell culturing. On the other hand, MM samples metastasis in some tumor types. In MM, we have detected
often contain a significant amount of contaminating LOH at the NME1 locus in 8 of 20 (40 percent) cell lines,
normal stroma, which can mask the existence of a four of which also showed downregulated expression com-
homozygous deletion in the malignant cell population. pared with control mesothelial cells (SC Jhanwar, unpub-
Downregulation of p16INK4a in MM cells may result from lished data).
5¢CpG island hypermethylation, as has been observed in
other kinds of cancer.28 At the protein level, abnormal
expression of p16INK4a has been reported in 12 of 12 MM NF2
specimens and 15 of 15 MM-derived cell lines examined
by immunohistochemistry.29 As noted earlier, numerical loss of a copy of chromosome
22 is a frequent occurrence in MM. Although germline
mutations of the neurofibromatosis type 2 TSG, NF2, pre-
p14ARF dispose affected individuals to tumors of neuroectodermal
origin, somatic mutations of NF2 have occasionally been
In most cases, homozygous deletion of the CDKN2A/ARF identified in seemingly unrelated malignancies.36 There-
locus also leads to inactivation of another putative TSG, fore, two groups independently embarked on mutational
p14ARF, because p16INK4a and p14ARF share exons two and studies of NF2 in MM. We identified nucleotide mutations
three, although their reading frames differ. p14ARF is essen- in 8 of 15 (53 percent) MM cell lines.37 The mutations,
tial for the activation of p53 in response to the action of which included deletions and insertions and one nonsense
certain oncogene products such as Ras or Myc.30 The mutation, predicted truncated forms of the NF2 protein,
p16INK4a product, on the other hand, induces cell cycle known as merlin or schwannomin. Similarly, Sekido and
arrest at the G1 phase by inhibiting the phosphorylation of colleagues38 detected somatic mutations in one MM spec-
the retinoblastoma protein, pRb. Therefore, homozygous imen and in 7 of 17 (41 percent) MM cell lines. In our
loss of p14ARF and p16INK4a would collectively affect both study, the mutations observed in cDNAs from MM cell
p53- and pRb-dependent growth regulatory pathways, lines were confirmed in genomic DNA from six matched
respectively. CDKN2B (p15INK4b), another gene located primary tumor specimens.37 The two cDNA alterations
near the CDKN2A/ARF locus, is also frequently deleted in that could not be confirmed by genomic analysis were
human MMs,31 although to date a critical role for p15INK4b both splicing related: i.e., deletion of exon 10 in one cell
in MM has not been clearly defined. line, and a 43-bp insertion between exons 13 and 14 in the
The methylthioadenosine phosphorylase gene (MTAP) other.
is also frequently co-deleted with the CDKN2A/ARF locus In a follow-up investigation, mutations in the NF2
in MM. For example, in one study of 95 MM cases, 70 coding region were detected in 12 of 23 (52 percent) addi-
tumors showed homozygous deletions of p16INK4a, 64 (91 tional MM cell lines.39 Western blot analyses revealed loss
percent) of which exhibited co-deletion of MTAP.32 of merlin expression in each of the 12 cell lines having
Whether loss of MTAP expression plays a fundamental alterations of the NF2 gene. In addition, two cell lines with
role in MM pathogenesis is unknown at this time, but this NF2 mutations reported in an earlier study were also
enzyme may represent an interesting therapeutic target examined, both of which lacked NF2 expression. LOH
(see below). analyses were performed on the entire 25 MM cell lines
using two polymorphic DNA markers residing at or near
the NF2 locus in chromosome 22q12. Eighteen of the 25
TP53 cell lines showed losses at one or both of these loci. All
cases exhibiting mutation and aberrant expression of NF2
While loss of p14ARF is frequently observed in MM, muta- displayed LOH, consistent with bi-allelic inactivation of
tions of the p53 gene (TP53) are less commonly reported NF2 in MM.
in this malignancy.33–35 TP53 is located at chromosome Merlin exhibits significant homology to the ezrin–
17p13, and loss or relative deficiency of the short arm of radixin–moesin (ERM) family of proteins known to play a
chromosome 17 is a recurrent change in MM. Immuno- role in cell surface dynamics by linking the cytoskeleton to
histochemical staining of p53 protein in MM cell lines was components of the cell membrane. However, the mecha-
performed by one of us (SCJ). Mutant p53 protein was nisms by which merlin exerts its tumor suppressor activity
detected in a significant percentage (>50 percent) of cells are incompletely understood. One mechanism involves
130 Genetic alterations in mesothelioma pathogenesis
the inhibition of cellular proliferation by repressing cyclin merlin attenuated FAK phosphorylation at the critical
D1 expression.40 Thus, we found that adenovirus-medi- phosphorylation site Tyr397 and disrupted the interaction
ated expression of merlin in NF2-deficient MM cells led to of FAK with its binding partners Src and p85, the regula-
cell cycle arrest at G1 phase, concomitant with decreased tory subunit of phosphatidylinositol-3-kinase (PI3K). In
expression of cyclin D1, inhibition of CDK4 activity, and addition, NF2-null MM cells stably overexpressing FAK
dephosphorylation of pRB. The effect of merlin on cell showed increased invasiveness, which decreased signifi-
cycle progression was partially overridden by ectopic cantly when merlin expression was restored. Altogether,
expression of cyclin D1. RNA interference experiments these findings suggest that merlin inactivation is a critical
with NF2-positive cells showed that silencing of the step in MM pathogenesis and is related, at least in part,
endogenous NF2 gene results in up-regulation of cyclin D1 with upregulation of FAK activity.
and S-phase entry. Recently, merlin has also been shown to suppress cellu-
Because Rho GTPase-mediated signaling phosphory- lar proliferation by inhibiting the activation of the small
lates ERM proteins, we and others tested whether merlin is G-protein Ras.47 Merlin was found to counteract the ERM-
also regulated by members of the Rho family of GTPases. dependent activation of Ras, which correlated with the for-
These investigations showed that merlin is phosphorylated mation of a complex comprising ERM proteins, Grb2,
in response to expression of activated Rac and activated SOS, Ras and filamentous actin. Thus, part of the tumor
Cdc42.41,42 Furthermore, we demonstrated that merlin suppressor function of merlin appears to be its interfer-
phosphorylation is mediated by p21-activated kinase ence with Ras- and Rac-dependent oncogenic signaling.
(Pak), a common downstream target of both Rac and Merlin has also been shown to interact with Ral guanine
Cdc42.42 Various kinase assays demonstrated that Pak can nucleotide dissociation stimulator (RalGDS), a down-
directly phosphorylate merlin at serine 518, a site that stream molecule of Ras, and to inhibit its activity.48
affects merlin activity and localization. In other experi- Functional studies revealed that merlin inhibits RalGDS-
ments, we found that Pak1-stimulated cyclin D1 promoter induced RalA activation, colony formation and cell migra-
activity was repressed by co-transfection of NF2, and Pak tion in mammalian cells.
activity was inhibited by expression of merlin.40
Interestingly, a S518A mutant form of merlin, which is
refractory to phosphorylation by Pak, was more efficient
than the wild-type protein in inhibiting cell cycle progres- AN NF2 MOUSE MODEL RECAPITULATING
sion and in repressing cyclin D1 promoter activity. MOLECULAR FEATURES OF HUMAN
Collectively, our data indicate that merlin exerts its MALIGNANT MESOTHELIOMA
antiproliferative effect, at least in part, via repression of
Pak-induced cyclin D1 expression. To better understand the significance of NF2 inactivation
Other data indicate that merlin and Pak participate in a in MM and identify tumor suppressor gene alterations that
feedback loop, because merlin has been shown to inhibit cooperate with NF2 loss of function in MM pathogenesis,
activation of Pak1.43,44 Thus, loss of merlin expression we treated Nf2(+/-) knockout mice with asbestos to
results in the inappropriate activation of Pak1, whereas induce MMs.31 Asbestos-exposed Nf2(+/-) mice exhibited
overexpression of merlin in cells with high basal activity of markedly accelerated MM tumor formation compared
Pak1 inhibited Pak1 activation. Merlin’s inhibitory func- with asbestos-treated wild-type littermates. Loss of the
tion is mediated by impeding Pak1 recruitment to focal wild-type Nf2 allele, leading to bi-allelic inactivation, was
adhesions.43 observed in all nine asbestos-induced MMs from Nf2(+/-)
Importantly, Pak has been shown to regulate motility in mice and in 50 percent of MMs from asbestos-exposed
mammalian cells,45 which raised the intriguing possibility wild-type mice. For a detailed comparison with the murine
that merlin loss of function may contribute to invasiveness model, DNA analyses were also carried out on a series of
and/or metastasis in MM. We recently showed that re- human MM samples. Remarkably, similar to human MM,
expression of merlin in NF2-null MM cells inhibits inva- tumors from Nf2(+/-) mice showed frequent homologous
siveness and negatively regulates focal adhesion kinase deletions of the Cdkn2a/Arf locus and adjacent Cdkn2b
(FAK).46 Re-expression of merlin markedly inhibited cell tumor suppressor gene, as well as reciprocal inactivation of
motility, spreading and invasiveness, properties connected the p53 gene, Tp53, in a subset of tumors that retained the
with the malignant phenotype of MM cells. To test directly Arf locus. As in the human disease counterpart, MMs from
whether merlin inactivation promotes invasion in a non- Nf2(+/-) mice also showed frequent activation of Akt
malignant system, we used small interfering RNA to kinase, which plays a central role in tumorigenesis and
silence Nf2 in mouse embryonic fibroblasts (MEFs) and therapeutic resistance. Thus, this murine model of envi-
found that downregulation of merlin resulted in enhanced ronmental carcinogenesis faithfully recapitulated many of
cell spreading and invasion. To delineate signaling events the molecular features of human MM and has significant
connected with this phenotype, we investigated the effect implications for the further characterization of MM
of merlin expression on FAK, a key component of cellular pathogenesis and preclinical testing of novel therapeutic
pathways affecting migration and invasion. Expression of drugs.
Potential therapeutic targets in malignant mesothelioma 131
AKT/Akt KINASE ACTIVATION IN MALIGNANT susceptibility gene could lead to the development of thera-
MESOTHELIOMA peutic approaches for members of these families.
Moreover, it is possible that the same gene may be a target
The PI3K/AKT pathway has been implicated in tumor in sporadic MMs associated with exposure to asbestos and,
aggressiveness, in part by mediating cell survival and thus, the eventual isolation of this gene might enhance our
reducing sensitivity to chemotherapy. Using antibodies understanding of molecular mechanisms involved in the
recognizing the phosphorylated (active) form of AKT, we pathogenesis of MM generally.
observed elevated phospho-AKT staining in 17 of 26 (65
percent) human MM specimens.49 In addition, AKT phos-
phorylation was repeatedly observed in MMs arising in POTENTIAL THERAPEUTIC TARGETS IN
asbestos-treated mice and in xenografts of human MM MALIGNANT MESOTHELIOMA
cells. Consistent with reports implicating hepatocyte
growth factor (HGF)/Met receptor signaling in MM, all 14 During the last three decades, much has been learned
human and murine MM cell lines had HGF-inducible about the genetic alterations associated with various
AKT activity. One of nine human MM cell lines had ele- cancers. This has led to the realization that the same
vated AKT activity under serum-starvation conditions, genetic changes involved in the development of a particu-
which was connected with a homozygous deletion of lar cancer may also serve as markers of disease susceptibil-
PTEN. Treatment of this cell line with the mTOR inhibitor ity, initiation and progression. In addition, some of these
rapamycin resulted in growth arrest in G1 phase. alterations have also provided a genetic basis for targeted
Treatment of MM cells with the PI3K inhibitor LY294002 therapy. It is only in recent years that scientists have begun
in combination with cisplatin had greater efficacy in to utilize this information with reasonable success to
inhibiting cell proliferation and inducing apoptosis than develop clinical trials to treat some of the cancers refrac-
either agent alone. Taken together, these findings indicate tory to standard therapies (see also Chapter 41, Malignant
that both human and murine MMs frequently express ele- mesothelioma).
vated AKT activity, which may be targeted pharmacologi- Considering the fact that the initial clinical presentation
cally to enhance chemotherapeutic efficacy. of MM is localized only to the pleural or peritoneal cavity1
and that there is a lack of any effective treatment for this
lethal disease,54 in theory, gene replacement therapy in
ERIONITE AND POSSIBLE GENETIC MM is an attractive option. In order to achieve success,
PREDISPOSITION TO MALIGNANT however, it will be necessary to overcome various impedi-
MESOTHELIOMA ments to gene therapy.55 If obstacles such as toxicity and
delivery can be effectively addressed, molecularly-based
In the small villages of Karain, Tuzkoy and ‘Old’ Sarihidir therapies could become practical.
in Cappadocia, a region in Central Anatolya, Turkey, char- As an initial approach to developing and testing such
acterized by volcanic tuffs and natural caves, nearly 50 therapy, investigators have utilized MM cell lines that have
percent of deaths are caused by MM. This extremely high sustained complete loss of a TSG to determine if the trans-
incidence of MM is associated with exposure to erionite, a fer of a normal copy of that gene inhibits growth and
form of zeolite that causes MM in rodents.50,51 However, restores other biological features of normal mesothelial
houses in the nearby village of Karlik were built with the cells. For example, it has been shown that adenovirus-
same type of stones, yet no MMs were observed. mediated gene transfer of p16INK4a in MM cells results in
Moreover, in Karain and Tuzkoy, cancer occurred mostly cell-cycle arrest and cell death, as well as tumor suppres-
in certain houses where entire families had succumbed to sion and regression.54 Similarly, adenovirus-mediated
MM. The amount of erionite in stone samples from these replacement of p14ARF in MM cell lines restored p53 func-
houses was found to be comparable to that of other houses tion, inducing cell cycle arrest in G1 phase and cell death.56
in MM or non-MM villages, suggesting the possible We have utilized retroviral vectors and adenovirus to re-
involvement of a genetic susceptibility factor.51,52 Pedigree express the NF2 gene in MM cell lines, and various exper-
analysis suggested that MM was genetically transmitted, iments outlined above have shown that restoring
possibly as an autosomal dominant disease.52 Approxi- expression of merlin results in an altered cell phenotype
mately 50 percent of the descendents of affected parents and tumor suppression.41
developed MM, whereas MM was absent in other families. The chemotherapeutic drug L-alanosine, a strong
When members of unaffected families married into inhibitor of de novo AMP synthesis, has been used in
affected families, MM appeared in their descendents.52 The various clinical trials to treat patients with leukemia and
X-ray diffraction pattern and crystal structure of erionite various solid tumors. It has been suggested that this drug
did not appear to differ in households with high or no inci- is particularly effective in selectively killing tumor cells
dence of MM.53 Thus, it appears that in Karain, Tuzkoy deficient in methythioadenosine phosphorylase (MTAP),
and ‘Old’ Sarihidir, MM is genetically transmitted and that an important enzyme for the salvage of adenine and
erionite is a cofactor. The isolation of the putative MM methionine.57 Furthermore, it has been demonstrated that
132 Genetic alterations in mesothelioma pathogenesis
MTAP-deficient cells are highly sensitive to L-alanosine which p53-deficient animals were shown to be more sus-
treatment (reviewed in reference 57). It is, therefore, not ceptible to asbestos-induced MMs than wild-type mice.59
unreasonable to propose that MMs with homozygous In summary, there is now a large body of experimental
deletion of 9p21 encompassing MTAP may be candidates and epidemiological data in support of the assertion that
for such a therapeutic choice along with conventional asbestos, or at least amphibole asbestos, causes MM. These
therapies employed to treat MMs. data also suggest that exposure to asbestos is usually not
As discussed earlier, NF2 has been shown to directly sufficient for MM development and that other factors,
inhibit Pak1, which is thought to be essential for Ras trans- such as genetic predisposition, may render some individu-
formation. Furthermore, certain Pak1 inhibitors are als more susceptible to asbestos carcinogenicity. Cyto-
known to selectively inhibit the growth of NF2-null cancer genetic and molecular genetic studies indicate that MM
cells, but not NF2-positive cells.44 These results suggest results from the accumulation of numerous somatic
that Pak1-specific blocking drugs could potentially be genetic events, mainly deletions, suggesting a multistep
useful for the treatment of NF2-deficient MM. Moreover, cascade involving the inactivation of multiple TSGs. To
the frequent activation of AKT signaling observed in date, several TSGs have been shown to be frequently
human MMs suggests that this pathway could represent an altered in MMs, and their disruption would be expected to
attractive therapeutic target for the treatment of this highly have profound consequences on the growth and behavior
aggressive form of cancer. Importantly, numerous groups of a mesothelial cell. The identification of all of the critical
are attempting to identify specific chemical inhibitors of somatic genetic alterations in MM and understanding how
proteins implicated in MM tumorigenesis, such as Pak and each of them contributes to the pathogenesis of this malig-
AKT, by high throughput screening of unbiased chemical nancy may ultimately lead to the design of more effective
libraries. The identification of such lead compounds is therapeutic strategies.
likely to provide biologically meaningful approaches for
novel targeted therapeutic strategies. Moreover, the fact
that TNF-a inhibits asbestos-induced cytotoxicity via a
NF-kB-dependent survival pathway (see also Chapter 8, KEY POINTS
Immunology) suggests that TNF-a/NF-kB may also serve
as targets for chemoprevention in high-risk groups, such ● Cytogenetic analysis has revealed that most
as asbestos workers or Cappadocian villagers exposed to human mesotheliomas have complex karyotypes,
erionite. with the accumulation of extensive aneuploidy
and structural rearrangements of multiple chro-
mosomes.
CONCLUSIONS ● Despite the great genomic disarray seen in
mesotheliomas, a number of recurrent genomic
Multiple genetic changes are involved in the development imbalances are found, including deletions of the
of most cancers. Asbestos is known to cause genetic short (p) arms of chromosomes 1, 3 and 9 and
damage. In a normal cell, mitosis is controlled through a the long (q) arms of chromosomes 6, 13 and 15.
delicate balance of phosphorylation and dephosphoryla- Loss of a copy of chromosome 22 is the most
tion events. Autophosphorylation of the EGF receptor common numerical change. Losses of chromo-
results in phosphorylation of MEK1 kinase, which in turn somes 4 and 17 and gains of chromosome 7 and
phosphorylates ERK (MAP kinases). The activation of 5p have also been reported in some series.
these kinases causes activation of other intermediaries, and ● Deletion mapping studies have identified a set of
this in turn stimulates c-fos and c-jun and other members commonly deleted sites in several different chro-
of the AP-1 family and induces cell division.5 mosomes. These sites are thought to represent
Downregulation of AP-1 activity is achieved through the the locations of putative tumor suppressor genes
phosphatase PP2A, which dephosphorylates MAP kinases. that contribute to the development and/or pro-
Asbestos, working through this mechanism, may induce gression of mesothelioma.
tumor formation. Asbestos can induce both DNA damage ● Frequent homozygous deletions of chromosome
and autophosphorylation of the EGF receptor, which band 9p21 target a set of tumor suppressor genes
eventually leads to AP-1 expression and cell division.58 As located there, including the CDKN2A/ARF locus,
a result of inactivation of p53, mutations caused by which encodes p16INK4a and p14ARF. Inactivation
asbestos would not undergo repair at the G1/S checkpoint of p14ARF and p16INK4a would collectively disrupt
mediated by p53 through p21. While most DNA alter- both p53- and pRb-dependent growth regulatory
ations will either be of no significance or lead to cell death, pathways, respectively. The tumor suppressor
a few cells could potentially develop perturbations of key genes CDKN2B, encoding p15INK4b, is often co-
cell cycle regulatory genes,12 become immortalized deleted with the CDKN2A/ARF locus in
and tumorigenic. Evidence in support of this notion mesotheliomas.
comes from experiments with p53 knockout mice, in
References 133
factor-receptor protein in a human mesothelial cell line in potentially involved in genesis of many tumor types. Science 1994;
response to long asbestos fibers. Am J Pathol 1998; 152: 333–40. 264: 436–40.
7. Ault JG, Cole RW, Jensen CG, et al. Behavior of crocidolite ●27. Cheng JQ, Jhanwar SC, Klein WM, et al. p16 alterations and
asbestos during mitosis in living vertebrate lung epithelial cells. deletion mapping of 9p21-p22 in malignant mesothelioma. Cancer
Cancer Res 1995; 55: 792–8. Res 1994; 54: 5547–51.
134 Genetic alterations in mesothelioma pathogenesis
28. Merlo A, Herman JG, Mao L, et al. 5¢ CpG island methylation is Nf2 tumor suppressor gene, is an inhibitor of the p21-activated
associated with transcriptional silencing of the tumour suppressor kinase, Pak1. Mol Cell 2003; 12: 841–9.
p16/CDKN2/MTS1 in human cancers. Nat Med 1995; 1: 686–92. 44. Hirokawa Y, Tikoo A, Huynh J, et al. A clue to the therapy of
●29. Kratzke RA, Otterson GA, Lincoln CE, et al. Immunohistochemical neurofibromatosis type 2: NF2/merlin is a PAK1 inhibitor. Cancer J
analysis of the p16INK4 cyclin-dependent kinase inhibitor in 2004; 10: 20–6.
malignant mesothelioma. J Natl Cancer Inst 1995; 87: 1870–5. 45. Sells MA, Boyd JT, Chernoff J. p21-activated kinase 1 (Pak1)
30. Palmero I, Pantoja C, Serrano M. p19ARF links the tumour regulates cell motility in mammalian fibroblasts. J Cell Biol 1999;
suppressor p53 to Ras. Nature 1998; 395: 125–6. 145: 837–49.
●31. Altomare DA, Vaslet CA, Skele KL, et al. A mouse model 46. Poulikakos PI, Xiao GH, Gallagher R, et al. Re-expression of the
recapitulating molecular features of human mesothelioma. Cancer tumor suppressor NF2/merlin inhibits invasiveness in
Res 2005; 65: 8090–5. mesothelioma cells and negatively regulates FAK. Oncogene 2006;
32. Illei PB, Rusch VW, Zakowski MF, Ladanyi M. Homozygous deletion 25: 5960–8.
of CDKN2A and codeletion of the methylthioadenosine ●47. Morrison H, Sperka T, Manent J, et al. Merlin/neurofibromatosis
phosphorylase gene in the majority of pleural mesotheliomas. Clin type 2 suppresses growth by inhibiting the activation of Ras and
Cancer Res 2003; 9: 2108–13. Rac. Cancer Res 2007; 67: 520–7.
33. Cote RJ, Jhanwar SC, Novick S, Pellicer A. Genetic alterations of 48. Ryu CH, Kim SW, Lee KH, et al. The merlin tumor suppressor
the p53 gene are a feature of malignant mesothelioma. Cancer interacts with Ral guanine nucleotide dissociation stimulator and
Res 1991; 51: 5410–6. inhibits its activity. Oncogene 2005; 24: 5355–64.
34. Metcalf RA, Welsh JA, Bennett WP, et al. p53 and Kirstein-ras 49. Altomare DA, Wang HQ, Skele KL, et al. AKT and mTOR
mutations in human mesothelioma cell lines. Cancer Res 1992; 52: phosphorylation is frequently detected in ovarian cancer and can
2610–5. be targeted to disrupt ovarian tumor cell growth. Oncogene 2004;
35. Mor O, Yaron P, Huszar M, et al. Absence of p53 mutation in 23: 5853–7.
malignant mesothelioma. Am J Respir Cell Mol Biol. 1997; 16: 50. Baris B, Demir AU, Shehu V, et al. Environmental fibrous zeolite
9–13. (erionite) exposure and malignant tumors other than
●36. Bianchi AB, Hara T, Ramesh V, et al. Mutations in transcript mesothelioma. J Environ Pathol Toxicol Oncol 1996; 15: 183–9.
isoforms of the neurofibromatosis 2 gene in multiple human ◆51. Carbone M, Kratzke RA, Testa JR. The pathogenesis of
tumour types. Nat Genet 1994; 6: 185–92. mesothelioma. Semin Oncol 2002; 29: 2–17.
●37. Bianchi AB, Mitsunaga S-I, Cheng JQ, et al. High frequency of ●52. Roushdy-Hammady I, Siegel J, Emri S, et al. Genetic-susceptibility
inactivating mutations in the neurofibromatosis type 2 gene (NF2) factor and malignant mesothelioma in the Cappadocian region of
in primary malignant mesotheliomas. Proc Natl Acad Sci U S A Turkey. Lancet 2001; 357: 444–5.
1995; 92: 10854–8. 53. Dogan AU, Baris YI, Dogan M, et al. Genetic predisposition to fiber
38. Sekido Y, Pass HI, Bader S, et al. Neurofibromatosis type 2 (NF2) carcinogenesis causes a mesothelioma epidemic in Turkey. Cancer
gene is somatically mutated in mesothelioma but not in lung Res 2006; 66: 5063–8.
cancer. Cancer Res 1995; 55: 1227–31. ●54. Frizelle SP, Grim J, Zhou J, et al. Re-expression of p16INK4a in
39. Cheng JQ, Lee WC, Klein MA, et al. Frequent mutations of NF2 and mesothelioma cells results in cell cycle arrest, cell death, tumor
allelic loss from chromosome band 22q12 in malignant suppression and tumor regression. Oncogene 1998; 16: 3087–95.
mesothelioma: evidence for a two-hit mechanism of NF2 55. Blaese RM. Gene therapy for cancer. Sci Am 1997; 276: 111–15.
inactivation. Genes Chromosomes Cancer 1999; 24: 238–42. 56. Yang CT, You L, Yeh CC, et al. Adenovirus-mediated p14(ARF) gene
●40. Xiao GH, Gallagher R, Shetler J, et al. The NF2 tumor suppressor transfer in human mesothelioma cells. J Natl Cancer Inst 2000;
gene product, merlin, inhibits cell proliferation and cell cycle 92: 636–41.
progression by repressing Rac-induced cyclin D1 expression. Mol 57. Batova A, Diccianni MB, Omura-Minamisawa M, et al. Use of
Cell Biol 2005; 25: 2384–94. alanosine as a methylthioadenosine phosphorylase-selective
●41. Shaw RJ, Paez JG, Curto M, et al. The Nf2 tumor suppressor, therapy for T-cell acute lymphoblastic leukemia in vitro. Cancer
merlin, functions in Rac-dependent signaling. Dev Cell 2001; 1: Res 1999; 59: 1492–7.
63–72. ◆58. Robledo R, Mossman B. Cellular and molecular mechanisms of
●42. Xiao GH, Beeser A, Chernoff J, Testa JR. p21-activated kinase asbestos-induced fibrosis. J Cell Physiol 1999; 180: 158–66.
links Rac/Cdc42 signaling to merlin. J Biol Chem 2002; 277: 59. Marsella JM, Liu BL, Vaslet CA, Kane AB. Susceptibility of p53-
883–6. deficient mice to induction of mesothelioma by crocidolite
●43. Kissil JL, Wilker EW, Johnson KC, et al. Merlin, the product of the asbestos fibers. Environ Health Perspect 1997; 105: 1069–72.
12
Proteomics in pleural disease
Proteomics and its complexity 135 Current limitations of clinical proteomics 146
Biomarkers 136 Future perspective 147
‘Traditional’ biomarkers for pleural diseases 136 Key points 147
Novel biomarkers for pleural diseases 138 References 147
Current techniques for proteomic analysis to discover 138
novel biomarkers
PROTEOMICS AND ITS COMPLEXITY quantification of proteins but also involves the compre-
hensive study of their structure, localization, modification,
Each cell produces thousands of proteins, each with a spe- interactions, activities and function of all proteins in body
cific function. Some proteins are expressed at very low fluids, tissues or cell types under given conditions.1
levels, a few copies per cell, while others such as house- Proteomics is relatively more challenging than genomics
keeping gene products are extremely abundant. They may because a single gene can give rise to multiple protein
be expressed during short periods during the life of an products through alternative splicing, proteolysis of pro-
individual, for example during embryonic development, teins and post-translational modifications (20 000–
while others may be continually expressed but with very 25 000 different genes versus approximately 1 000 000
short half lives. The collection of proteins in a cell is known different proteins).2,3 Currently there are more than 300
as the proteome, and, unlike the genome which is constant different types of post-translational modifications known,
irrespective of cell type, it differs from cell to cell and is and new ones are regularly discovered.4,5 They are often
constantly changing through its biochemical interactions transient and occur in vivo only in a small fraction of pro-
with the genome and the environment (Figure 12.1). It teins (<1 percent) and include glycosylation, phosphoryla-
changes from moment to moment in response to tens of tion, acetylation, nitration, ubiquitation and disulfide
thousands of intra- and extra-cellular environmental bond formation.4–7 Post-translational modifications regu-
signals, such as other proteins, pH, hypoxia and drug late protein function, determining their activity state, cel-
administration, and changes continuously during multi- lular location and dynamic interactions with other
genic processes such as ageing, stress or disease. proteins or nucleic acids.6
Proteomics consists not only of the identification and Exploring the structure and activity of proteins is
increasingly used to address biomedical questions that
may help to elucidate the molecular basis of health and
degradomics serum peptidome glycomics disease and, for example, to address fundamental ques-
lipidome clinical proteomics tions in the progression of a disease from a normal to a
Genome: The entire collection of genes in the complete DNA sequence of an organism pathophysiologic state (clinical proteomics). We will
Transcriptome: The complete set of mRNAs that are transcribed from the genome
Proteome: The expressed set of proteins that are encoded by the genome discuss different aspects of proteomics, including new
Genomics: Investigations and techniques for identifying the genome
Proteomics: Investigations and techniques for identifying the proteome proteomic platforms as well as their limitations. It is of no
metabolome
toponomics functional proteomics surprise that none of these proteomic approaches cur-
phosphoproteomics rently come close to detecting all proteins present in
subproteomics
complex biological systems. When the obstacles can be
Figure 12.1 The ‘omes’ and ‘omics’ era. Because of the success overcome, it will enable abundant harvesting of diagnostic
of genomics and proteomics, the suffix ‘-ome’ and ‘omics’ has biomarkers leading to a new era of personalized clinical
now widely migrated to a host of other contexts. medicine.
136 Proteomics in pleural disease
Cap
Transfer
film
Cells
of interest
Microscope
slide
(b)
Infrared laser
pulse
Transfer of (c)
selected cells
(a)
(a)
Figure 12.2 Laser capture microdissection allows researchers to compare normal with
diseased cells by isolating distinct subpopulations from (stained) tissue sections under
direct microscopic visualization. (a) After locating the cellular population of interest (top
picture), a cap with film backing is placed over the target area (middle picture). Pulsing
the infrared laser locally expands the thermosensitive polymer film to reach down and
adhere the target cell(s) beneath the laser pulse. Lifting the cap from the tissue section (d)
removes the target cells now attached to the cap (lower picture). The cap holding the
captured cells is then transferred to a tube, where an extraction buffer is used to
remove the cells for further analysis. An example shows the laser outline of the cells to
be collected (b), the remaining cells after laser capture (c), and the cells collected from
the outlined area (d). (See also Color Plates 5–7.)
cancer, lymphoma, mesothelioma, gastric or esophageal (CEA), neuron-specific enolase (NSE), CA125, squamous
cancer and ovarian carcinoma. The diagnosis of a malig- cell carcinoma antigen, CA19-9, tissue polypeptide antigen
nant pleural effusion is established by demonstrating (TPA), a-fetoprotein, CYFRA 21-1 or osteopontin in the
malignant cells in the pleural fluid or in the pleura itself. effusions to discriminate malignant from benign pleural
Numerous papers have recommended various diagnostic exudates.25 Although the presence of these tumor markers
tests, such as cytological and chromosomal analysis of is highly suggestive when levels are high, it is not very
pleural cells, measurement of pH, glucose, amylase, or helpful if values are only modestly increased.26,27
measurement of proteins as carcinoembryonic antigen Therefore, diagnosis of a disease based on a pleural
138 Proteomics in pleural disease
Protein digestion
Trypsin
NOVEL BIOMARKERS FOR PLEURAL DISEASES Lys-C
Asp-N
Patients with malignancy, pulmonary embolus and infective Glu-C
effusions may have an elevated level of fibrin degradation Peptide separation
products in their pleural fluid.28 Also, elevated b2-
High-performance liquid chromatography (HPLC)
microglobulin levels have been associated with tuberculosis, Ion exchange
leukemia, lymphoma and some autoimmune disorders.29
Distinctive levels of pleural fluid adenosine deaminase Sample ionization
isoenzyme 2 (ADA2), specific anti-tuberculous antibodies,
lysozyme and interferon-g are present in the pleural effu- Electrospray ionization
sion of patients with tuberculous pleuritis compared with Matrix-assisted laser desorption–ionization (MALDI)
patients with effusions of other etiologies.25 ADA1 isoen-
zyme is elevated in parapneumonic effusions.30,31 Mass spectrometry
Quadrupole (Q)
Another example of a promising biomarker with diag- Time of flight (TOF)
nostic potential in pleural fluid is soluble mesothelin- Quadrupole ion traps (IT)
related protein (SMRP).32,33 Pleural and serum levels of Fourier-transform ion cyclotron resonance (FT-ICR)
SMRP are significantly higher in epithelioid mesothelioma Data analysis
patients than in those with benign pleuritis and pleural Peptide search
metastases.33 Serum SMRP levels are tumor-size related Sequest
Mascot
and decrease upon surgical cytoreduction, suggestive of a
role of SMRP in disease monitoring,34–37 see also Chapter Figure 12.3 An outline of a general strategy to perform
15, Experimental models: mesothelioma. Efforts are proteomics. The identification of proteins of interest relies mainly
underway to search for biomarkers like SMRP that can be on the separation of a complex protein mixture by
analyzed relatively noninvasively and economically in electrophoresis, mass measurement of peptides generated after
effusions or serum, for other pleural diseases. spot proteolysis by mass spectrometry and search in databases.
There are a number of options available to profile proteins High-resolution one- and two-dimensional gel elec-
and identify potential biomarkers. These rely mainly on the trophoresis (1D and 2D GE, respectively) has traditionally
separation of a complex mixture of proteins by elec- been the gold-standard discovery-based tool for pro-
trophoresis, mass measurement of peptides generated after teomics38,39 and was used for analyzing the proteome of
spot proteolysis by mass spectrometry (MS) and searches in body fluids, for instance for plasma,40 urine,41 cerebrospinal
databases (Figure 12.3), although various other approaches fluid42,43 and pleural fluid.44–47 Unfortunately, pleural
are now used to study differentially expressed proteins. fluids, like most of the body fluids, contain an enormous
Before starting a proteomics study, the advantages and dis- amount of different proteins and salt ions. Albumin and
advantages of various methods must be assessed in order to immunoglobulin fragments are the major protein compo-
choose the best suitable approach. The choice of an appro- nents, representing 50–70 percent and 10–20 percent of the
priate methodology will depend on the goals of the specific total pleural proteins, respectively (mg/mL range).
study, amount and number of samples, availability of Together with transferrin, fibrinogen, haptoglobin, antit-
resources and other factors. Although technical advances rypsin, complement components and a few other proteins,
have been significant in previous decades, we are still con- the top 20 proteins are responsible for approximately 99
fronted with the challenges of the evaluation and validation percent of the protein mass. Pleural effusions contain a
of the proteomic technologies, sample preparation and tremendous array of very-low-abundance molecules such
fractionation, understanding the massive volume of data, as signaling and regulatory proteins. For example, compar-
translating the information to fit clinical contents and atively small quantities of cytokines are detected by cytokine
incorporating it into clinical studies. We will now summa- arrays and ELISA, such as transforming growth factor-beta
rize some of the available technologies and discuss the (TGF-b), interleukin (IL) (IL-1, IL-6, IL-8, and IL-10) or
limitations that we still face. vascular endothelial growth factor (VEGF), but these are in
Current techniques for proteomic analysis to discover novel biomarkers 139
the ng/mL to pg/mL range, a difference of nine orders of on the basis of their molecular weights (Figure 12.4a).
magnitude or more. Therefore, removal of abundantly After electrophoresis, protein spots in a gel can be visual-
expressed proteins is a key element of proteome research to ized using a variety of radioactive, chemical stains or flu-
allow the visualization of co-migrating proteins on a 1D and orescent markers (for example, Coomassie Brilliant Blue
2D gel and to allow a higher sample load for improved visu- or silver). Depending on the type of staining, 200–3000
alization of low-abundance proteins.48–52 A convenient proteins per gel can be visualized (Figure 12.4b). A large
approach to remove high-abundance proteins from body number of spots in a 2D gel result from post-translational
fluids is affinity chromatography with resins carrying highly or proteolytic modifications of proteins: a protein may,
efficient and specific ligands for these proteins. Removal of therefore, be present in several locations in the gel.
these proteins using commercially available kits (e.g. Despite specialized software packages that allow compar-
ProteoPrep Blue Albumin Depletion Kit [Sigma-Aldrich, St isons of multiple gels, the matching of images of different
Louis, MO, USA], or Multiple Affinity Removal System gels can be difficult because they can be distorted by less-
[Agilent Technologies, Santa Clara, CA, USA]) improves defined, less well separated spots, shrinking or swelling of
the resolution and increases the protein spot count in non-backed gels and concentration differences between
depleted samples.52 However, removal may confound the the gels. There are potentially thousands of intact or
subsequent proteomic analysis because peptides and low cleaved proteins in the human pleural effusion proteome.
molecular weight proteins of interest may be bound to these Finding a single disease-related protein is like searching
large carrier proteins.53 for a needle in a haystack, requiring the separation and
Samples are normally purified by removal of salts, identification of these entities individually. This approach
lipids and other interfering substances and concentrated has recently been published for a composite pleural effu-
before applying onto the first dimension (e.g. by using the sion sample from seven lung adenocarcinoma patients.
2D clean-up kit [GE Healthcare, Fairfield, CT, USA]). In This study revealed at least 472 silver-stained protein
2D GE, protein mixtures are first separated by isoelectric spots to be present in a 2D gel map, half of which could be
focusing; on the application of a current, the charged pro- identified by liquid chromatography-tandem MS.46
teins migrate in a gel strip that contains an immobilized Although the results of these studies provide information
pH gradient (IPG) until they reach the pH at which their for a basic understanding of the protein composition of
overall charge is neutral (isoelectric point or pI) in the pleural effusions, the value for clinical medicine is limited.
first dimension.39 This gel strip is then applied onto a rec- Many of the proteins present in pleural fluid are likely to
tangular sodium dodecyl sulfate containing polyacry- have originated from serum.46,54–59 Of interest are those
lamide gel (SDS-PAGE), and the pI focused proteins proteins that have not previously been reported in the lit-
migrate by electric current into the gel and are separated erature to be present in serum. These proteins could then
Isoelectric point
3 10
200 pH
Molecular mass
kDa
Protein spot
10
(a)
originate from diseased cells and may contain disease-spe- teins in the effusion may be caused by specific proteolysis
cific information and with it represent potential candi- or by specific absorption from the circulation by tumor
dates for useful biomarkers concentrated in or only cells. A protein spot with the approximate molecular mass
measurable in pleural effusions. To discover these pro- of 30 kDa and pI of 5.5, significantly expressed in the serum
teins, the development of quantitative proteomics such as but not in the effusion, was selected for identification and
differential gel electrophoresis and isotope-coded affinity further analyses. Proteins were analyzed by matrix-assisted
tags has widened the applicability to detect proteins of laser desorption/ionization–time-of-flight (MALDI-TOF)
interest. MS analysis using peptide mass fingerprinting and 19
matched peptides with a total coverage of 75 percent are the
basis of the identification of the decreased spot in effusions
Differential gel electrophoresis as apolipoprotein A1 (accession number CAA00975), a
protein with a molecular mass of 28.061 Da and pI of 5.27.
A powerful quantitative technique currently applied is dif- This spot migrated differently from the major apolipopro-
ferential gel electrophoresis (DIGE).60–64 This technology is tein A1 spots and represents a small fraction of the total
commercially available (GE Healthcare). It has the poten- serum apolipoprotein A1 but is absent in pleural effusion.
tial to overcome many of the limitations of 2D This directly illustrates the advantage of 2D gel-based
electrophoretic studies by allowing the direct comparison approaches in visualizing changes in the molecular weight
in proteomic profile of different samples at a particular and pI of a protein. The different pI and slightly different
time, under a particular set of conditions.65 DIGE encom- molecular mass reflects biologically significant processing
passes a simple strategy involving three molecular weight- and pI-altering post-translational modifications such as
and charge-matched cyanine dyes (CyDyes: Cy2, Cy3 and phosphorylation, sulfation or (de-)acetylation. Thus, com-
Cy5 [GE Healthcare]) possessing unique absorption and parison of the protein spots from serum and effusion by
emission spectra.66,67 The fluorescent dyes bind to the ter- 2D-DIGE provides a very striking quantitative picture of
minal amino group of lysine side chains in proteins with no proteins absorbed or shed into body fluids. This truncation
change in protein charge and add only 0.5 kDa to the mass product has not been reported previously and it is not
of the protein, thereby minimizing dye-induced shifting known whether the fragmentation was due to in vivo bio-
during electrophoresis. Due to a minimal labeling (only logical processing or protease activity. Reductions in the
2–5 percent of the total number of lysine residues are serum levels of apolipoprotein A1 have also been correlated
labeled), binding of the dye to the protein appears to have with hepatitis B virus-induced diseases.73–75 An isoform of
no effect on MS analyses. Two different samples are labeled apolipoprotein A1 was detected by 2D GE in serum
with Cy3 and Cy5 and a third sample, labeled with Cy2, is obtained from individuals with high risk for the develop-
introduced as an internal control for each gel. The internal ment of, or those diagnosed with, hepatocellular
control is often a pooled sample comprising equal amounts carcinoma.76 Apolipoprotein A1 is a potential marker of
of each of the samples within the study. This allows normal- the aggression in colonic adenocarcinoma77 and is upregu-
ization and both inter- and intra-gel matching of proteins lated in primary carcinoma tissue of the vagina.78 However,
and is imperative for accurate protein quantification. Once a downregulation of apolipoprotein A1 in serum is
labeled, samples are mixed and isoelectrically focused on an described in early stage ovarian cancer.79,80
IPG strip and co-electrophoresed on the same 2D gel. The
spectrally distinct dyes allow co-separation of different
CyDye-labeled samples and ensure that all samples will be Isotope-coded affinity tagging
subjected to exactly the same 2D GE running conditions.
This limits the experimental variation and thus ensures Owing to the limitations of these gel-based technologies
accuracy within gel matching. Each dye is then scanned (Table 12.1), more versatile mass spectrometry-based
using different emission filters and images are analyzed approaches in conjunction with gel-free protein separa-
with DeCyder Differential In-gel Analysis software (GE tions have been developed in recent years. Isotope-coded
Healthcare). This software allows protein alignment and affinity tags (ICAT) has been the most widely practiced
quantification between scanned images. Spots may be MS-based, non-gel, quantitative approach for biomarker
directly picked through an automated system. The differ- discovery in the last few years but new developments, such
ential 2D DIGE has been used in the proteomic expression as tags for relative and absolute quantification (iTRAQ),81
analysis of several cancer cell systems.68–72 To discover the metabolic labeling82 and label-free liquid chromatography
proteins of interest, we used a strategy of comparative MS,83,84 are emerging. ICAT utilizes stable isotope labeling
analysis of serum proteome and pleural effusion proteome of cysteine-containing proteins to compare the relative
from the same mesothelioma patient using the DIGE tech- abundance between two comparative reduced protein
nology (Figure 12.5). Overexpression of proteins by mixtures.85 The affinity tags have different masses, but are
mesothelioma cells can result in their shedding in the structurally and chemically identical and covalently bind
pleural effusion and will lead to enhanced intensity spots to all cysteines within a protein. When the light tag (e.g.
compared with serum of the same patient. Absence of pro- linker possessing nine carbon-12 atoms) or the heavy tag
Current techniques for proteomic analysis to discover novel biomarkers 141
Patient
First dimension:
isoelectric focusing
Second dimension: pl
relative molecular MW SDS-PAGE
masses Figure 12.5 Flowchart of differential gel electrophoresis (DIGE)
analysis of serum and malignant pleural fluid derived from a patient
suffering from malignant mesothelioma. Samples to be compared
Imagin on typhoon are labeled with either Cy3 or Cy5, whereas the Cy2 is employed to
label a pooled sample comprising equal amounts of serum and
effusion within the study. The labeled samples are combined and
then run on a single two-dimensional gel. Proteins are detected
using a dual laser-scanning device equipped with different
excitation/emission filters in order to generate three separate
Cy3 Cy2 Cy5
images. The images are matched by a computer-assisted overlay
Analysis with DeCyder software method, signals are normalized using the corresponding Cy2 spot
intensities, and spots of interest are excised and analyzed by mass
Spot picking protiens of interest
spectrometry. Differentially expressed proteins in serum and pleural
Enzymatic digestion effusions of the same patient can be useful to discover proteins
Peptide mass fingerprinting by that may be the result of the cancer. SDS-PAGE, sodium dodecyl
mass spectrometry sulfate–polyacrylamide gel electrophoresis.
Table 12.1 Advantages and disadvantages of selected proteomic technologies for protein profiling
2D GE Good separation for the larger-molecular mass region Low throughput (one sample per gel)
of the proteome (between 10 and 150 kDa) Labour intensive
Thousands of proteins can be resolved; for each Large sample input (>100 mg)
protein the isoelectric point, MW and the relative Small proteins (<10 kDa) or very large proteins (>150 kDa),
quantity can be determined extremely acidic or basic proteins, hydrophobic or otherwise
Detection of post-translationally modified proteins insoluble proteins (e.g. membrane proteins) are poorly
Proteins can be stored within (dried) gels for months resolved on gel
and analyzed at a later date Low abundance proteins are not detected because of
sensitivity limits
Gel-to-gel variations confound the analysis process
No direct (online) protein identification
DIGE Direct comparison of up to three samples on one 2D gel Low throughput (three samples per gel)
Four orders of magnitude dynamic range and good Labeling dependent on lysine content
correlation between spot density and protein content Mass shift of c. 500 Da, impractical for subsequent MS
Internal standard allows for quantitative comparison of (post-staining required especially for low-molecular-
multiple gels weight proteins)
Gel spots only visible under fluorescent light, equipment
required for visualization and spot excision
ICAT Measures the relative abundance of heavy and light Low throughput (two samples per run)
peptides simultaneously Large sample input
Relative quantification and direct identification within Extensive sample fractionation before MS
a single analysis Involves radioisotope handling by user
Identification possible for only the differently expressed Targets only cysteine residues (5% of human proteins lack
proteins cysteine)
Does not require metabolic labeling
MALDI High throughput (up to 1536 samples per plate) Need for (offline) sample fractionation of complex samples
Expensive equipment and processing costs
Dependent on ionization efficiency
SELDI High throughput (up to 96 samples per bioprocessor) Spectral patterns of masses rather than actual protein
Direct application of sample (‘fast on-chip sample identifications are produced
cleanup’) Only useful for separating small-molecular-mass
Small amount of starting material proteins (<15 kDa)
Unbiased searches
Protein array Relatively high throughput Limited availability of antibodies with high specificity and
Semi-quantitative measurement of hundreds of proteins affinity for the proteins of interest
in parallel Possible cross-reactivity, giving false-positives
Clinically applicable Different affinities of antibodies on one chip
Untargeted proteins will remain undetected
Difficult to preserve proteins in their biologically active shape
and form
Abbreviations: 2D GE, two-dimensional gel electrophoresis; DIGE, differential gel electrophoresis; ICAT, isotope-coded affinity tag; MALDI, matrix-assisted
laser desorption/ionization; MS, mass spectrometry; SELDI, surface-enhanced laser desorption/ionization.
arginine residues into a mixture of peptides. The resulting pulses, the peptide/matrix crystals become detached (i.e.
peptides have, on average, the right size to be usefully desorption), and gaseous ions are liberated (i.e. ioniza-
detected by MS and are first mixed with UV-absorbing tion). The charged molecules are accelerated through a
organic acid, also known as matrix solution (e.g. a-cyano- strong electric field within a high vacuum, and a recording
4-hydroxy-trans-cinnamic acid) which causes the peptide is taken of how long the peptides take to travel a specified
to form crystals. When irradiated with brief UV laser distance and strike a detector. The longer the time of flight,
Current techniques for proteomic analysis to discover novel biomarkers 143
Sample 1 Sample 2
Enzymatic digestion
Abundance
ratio (light/heavy) Figure 12.6 Isotope-coded affinity tags (ICAT) utilizes chemical
Intensity
the more massive the particle is relative to its charge. correlate the experimental data with data derived from
Ionized peptides are thus separated on the basis of their peptide sequences in protein databases (e.g. Peptide
mass-to-charge (m/z) ratio. Detected peptides/ions are Sequence Tags, Sequest or Mascot).
displayed as a unique series of peaks that are referred to as Using the combination of 2D GE followed by MALDI-
the peptide mass fingerprint (PMF). This mass mapping TOF MS showed, for example, an increase in cyclin D2,
spectrum of peptide peaks is then compared with the XEDAR, p53-binding protein Mdm2, LIM and cysteine-rich
virtual peptide masses predicted from theoretical digestion domain protein 1, and HSP70-interacting protein and
of protein sequences currently contained within databases HSPC163 present or increased in lung squamous carcinoma
(e.g. UniProt, Swiss-Prot, NCBI) and the protein can be compared with normal bronchial tissue.89 It also allowed the
identified, considering the specificity of the protease identification of several individual proteins and specific
employed. For further identification, the actual amino acid protein isoforms that were increased in lung adenocarcino-
sequence (primary structure) information for the peptide mas90–93 and proteins (e.g. phosphoglycerate kinase 1) that
of interest can be obtained by tandem mass spectrometry can predict the survival of patients with early-stage lung
(commonly referred to as MS/MS or MS2), in which the cancer.92 This technology was also used to characterize the
particular peptide ion is isolated and fragmented by colli- proteins present in exosomes isolated from malignant
sion with an inert gas (such as nitrogen molecules, argon pleural effusions from different cancerous origin to gain
or helium atoms), and complex algorithms are used to information on their potential biological function(s).94
144 Proteomics in pleural disease
Surface-enhanced laser desorption/ patterns from a single mass-spectral training set, each
ionization–time-of-flight pattern combination re-adjusting as the models improve
in the adaptive mode. Unlike tandem MS, SELDI alone
Surface-enhanced laser desorption/ionization–time-of- cannot be used to identify individual proteins in a sample.
flight (SELDI-TOF, Vermillion [formerly Ciphergen Other limitations of SELDI are the overall lack of sensitiv-
Biosystems], Fremont, CA, USA) MS has been intensely ity and reproducibility, and the potential bias towards
controversial as a tool for quantitative analysis of protein certain proteins. It is, however, ideal for high-throughput
mixtures, mainly because of its methodological shortcom- protein profiling of a large numbers of samples, and is
ings and bioinformatics artifacts.95 This technique used for biomarker discovery in cancer.97–103 We studied
employs protein biochips, spotted with a protein capture pleural effusions from patients with confirmed mesothe-
bait such as a chemical affinity resin (i.e. hydrophobic, lioma (n = 54) and from patients with effusions due to
hydrophilic, metal affinity, cationic or anionic surfaces), other causes (n = 54, cancerous and non-cancerous) using
receptors and ligands, antibodies, DNA oligonucleotides SELDI-TOF. All samples were collected, processed and
or enzymes96 to enrich for the protein or peptide of inter- stored in the same way. Samples were fractionated using
est. Microliter quantities of crude protein extracts directly anion exchange chromatography and then bound to dif-
from their ‘native’ environments are applied to the ferent types of ProteinChip® array surfaces. Peak intensity
ProteinChip®, allowing proteins with physical or chemical data were subjected to classification algorithms in order to
affinities to the capture molecules to bind to the surface, identify potential classifier peaks that could be used to dis-
and then washed to remove impurities or loosely-bound criminate between mesothelioma and non-mesothelioma
proteins (Figure 12.7). Analytes are laser desorbed and samples. One such protein peak at m/z 6614 was character-
ionized directly from the chip for mass spectral analysis. A ized as apolipoprotein (Apo) CI and was decreased in
SELDI experiment produces a unique sample mass spec- pleural effusions due to mesothelioma. These molecules
tral profile (m/z) ranging from small peptides of <1000 Da represent highly abundant proteins produced by the liver,
up to proteins of >300 kDa and can distinguish differences whose concentration may be decreased due to cancer
in protein expression levels between samples, which cachexia or malnutrition, rather non-specific effects for
require highly ordered data mining operations for analysis. many cancer types. In this setting, however, the sensitivity
Artificial intelligence-based systems are uniquely suited for and specificity of this potential biomarker was 76 and 69
this because it learns, adapts and gains experience over percent, respectively. The area under the receiver operat-
time through constant retraining. It is possible to generate ing characteristic curve (AUC) for Apo CI was 0.755. We
not just one, but multiple combinations of proteomic were unable to identify SMRP, probably because of the low
sensitivity of the SELDI-TOF approach. External and thor-
ough validation studies are now underway to further
Crude sample Wash to
is placed directly reduce
investigate a possible role and function of the decrease of
on proteinChip non-specific Add matrix Apo CI in the oncogenesis of mesothelioma and to put the
array binding ions
SELDI technology into perspective.
(Figure 12.8b), protein or peptide samples in solution are tion on the basis of hydrophobicity. Ionized and frag-
passed through a fine needle to which high electrical mented protein samples are introduced into a mass ana-
potential (several kV) is applied, which results in a fine lyzer, which separates and detect the ions according to
spray of highly positively charged sample-containing their m/z ratios. Besides the already described time-of-
droplets. Samples are delivered to the mass analyzer after flight device (TOF as in MALDI-TOF and SELDI-TOF),
the breakup and evaporation of the solvent, which other basic types of mass analyzers are the quadrupole
decreases the size and increases the charge density of the (Q), quadrupole ‘ion trap’ (IT), and Fourier transform
droplets. Electrospray ionization is performed from a ion cyclotron (FT-ICR-MS or FTMS), each with its own
liquid sample and liquid-based chromatographic separa- strengths and weaknesses. Each of these instruments
tion (LC) systems, such as low-flow rate LC can be generates a mass spectrum, hence the term mass spec-
coupled to the needle to allow for protein fractionation of trometry.
the samples before mass analysis. It separates the samples A Q mass filter consists of four parallel rods through
by eluting from reverse-phase columns using increasing which direct current and radio frequency alternate electric
concentrations of organic solvents, which allow separa- fields are applied to sort the introduced ions. For each
combination of voltages and frequencies, only ions with a
specific m/z ratio pass undeflected through the Q mass
filter. Ions of the desired mass are fragmented in a gas col-
(a) Mass analyser lision cell, the masses of these daughter fragments then
acceleration by electric potentials
being measured by a TOF mass analyzer. Precise stepping
of the settings allows the quadrupole to be used as a mass
analyzer to scan for ions over a large m/z range. In IT mass
Matrix ions spectrometry, the ions of a selected mass are first caught
(trapped) in a dynamic electric field and are then sequen-
PU Analyte ions tially – according to their m/z value – ejected onto the
LS
ED detector with the help of another electric field. Trapped
LA
SE ions can also be fragmented by increasing the energy of the
R trap, with the mass detector measuring the masses of these
Analyte molecules
fragments for peptide identification. In contrast to IT,
FTMS keep the ions confined in the high magnetic field of
Matrix molecules a super-conducting magnet. The ions circle with frequen-
cies that are inversely proportional to their m/z value. This
circling induces an alternating current in the metal plates
that make up the trap. This time-varying current consti-
(b) M
tutes a frequency spectrum of the ion motion and is
Droplets releasing a converted by the mathematical operation Fourier transfor-
analyte ions
s mation – which explains the name – into a mass spectrum
s with high resolution and mass accuracy.
a The number of possible instrumental configurations is
n
SPRAY NEEDLE a
large because different analyzers can be coupled with
various sources, such as the popular quadrupole-time of
l
flight hybrid (Q-TOF) mass spectrometer. When more
Analyte mixture y
s accuracy and sensitivity than MALDI or SELDI is needed,
e for instance for the analysis of post-translational modifica-
Acceleration by electric potentials r tions of proteins, high pressure LC is coupled to an ESI
source and an FTMS instrument with electron capture dis-
Figure 12.8 ‘Soft’ ionization techniques as matrix-assisted laser sociation can be used. The LC-ESI-triple quadrupole with
desorption/ionization (MALDI) (a) and electrospray ionization (b) multiple ionization sources is the instrument of choice for
are used in biological mass spectrometry to transfer highly polar, looking at biomarkers or drug efficacy. However, as there
non-volatile peptides/proteins into the gas phase without is no single instrument that can do it all, state of the art
destroying them. (a) Sample and matrix are mixed and allowed to proteomics facilities need to be equipped with several dif-
dry forming crystals and in a mass spectrometer ions are formed ferent types of MS instruments.
by irradiating the crystals with a brief laser pulse. (b) The analyte Highly complex protein mixtures (e.g. unfractionated
solution is injected via a thin metallized needle on which a cellular lysates) that cannot be efficiently resolved on gel
potential is applied. A continuous spray forms at the tip of the can be reduced, alkylated, digested and fractionated
needle. Under the effect of the potential difference, the spray through a strong cationic exchange column, and further
process forms very small droplets which progressively desolvate, separated on a reverse-phase column in a technique known
liberating multiply charged ions. as ‘multidimensional protein identification technology’
146 Proteomics in pleural disease
(MudPIT).105 This approach, whereby 2D liquid chro- differences in (tumor) type, location, size, histology, grade
matography (cation exchange followed by reverse-phase and stage of the disease, and heterogeneity, including dif-
chromatography) is coupled to tandem MS, was applied to ferences in genetic background, age, sex, ethnic origin,
yeast proteomic analysis and a total of 1484 proteins were smoking, alcohol consumption, medication, exercise, diet
detected and identified, including the identification of 131 and many others. Besides this, it is crucial that protein
proteins that are membrane localized – a highly problem- samples are handled and stored consistently throughout
atic class of proteins due to their intrinsic hydrophobic the study. Many factors, such as time of collection, con-
nature. This technique reduces the bias against the detec- tainers used, preservatives and other additives, transport
tion of basic and hydrophobic proteins and is very sensitive to the laboratory, storage and prefractionation, affect the
and reproducible. quality of the samples and the stability of the proteins of
interest and must be considered at the initial collection
stage.
Protein array system Most discovery platforms require large cellular input
samples that are orders of magnitude greater than those
Protein (micro-)array platforms containing spotted anti- procured during a clinical biopsy.69,105,117–120 Pleural effu-
gens or antibodies have been developed to rapidly screen sions, on the other hand, are relatively easy to obtain in
changes in protein expression or modification on a global large amounts but there is no single test for exudates that
scale without much of the specialist and expensive equip- is totally reliable or accurate for the diagnosis of its specific
ment as required by MS-based proteomics.106,107 Protein cause as the same results can also be found in other condi-
microarrays have been prepared by printing the capture tions. However, a single biomarker–single disease correla-
proteins on filters or coated microscope slides using puri- tion will be very unlikely. Biomarker discovery is therefore
fied or recombinant proteins, crude mixtures or antibod- moving away from the idealized single disease-specific bio-
ies. Although protein microarrays have been anticipated marker to a panel of many markers to transcend the het-
for several years108 and are predicted to be the ‘most likely erogeneity to reach a higher level of specificity and
path to low-cost, routine measurement of large numbers sensitivity. Thus, bioinformatics tools are required that
of plasma markers required for an impact on medical can extract diagnostic patterns of proteins rather than a
practice’,109 the applications of antibody microarrays were single biomarker from complex protein mixtures full of
limited by a variety of fundamental problems. Among biological noise and individual variability, reliably recog-
them, cross-reactivity significantly limits the multiplexing nize them in different samples where the patterns may be
ability in parallel sandwich immunoassays, and the limita- slightly changed and eventually cross-correlate them with
tions to the normalization, sensitivity, robustness and the other parameters related to the patients’ health and treat-
standardization between experiments.110 New formats of ment status. Particularly interesting are attempts to use
antibody microarrays are now applied with high through- SELDI to identify and combine several potential classifier
put and parallel detection for low-abundance proteins in peaks for a disease profile, however, recent controversy
body fluids.111–115 Proteins arrayed on a flat solid phase concerning this technology shows the difficulties facing
were used to simultaneously detect multiple cytokine proteomic research.95 SELDI, like the majority of MS-
expression levels from mesothelioma patient’s pleural based methods, primarily identifies smaller abundant pep-
effusions.116 Angiogenin, VEGF, TGF-b, epithelial tides with a higher sensitivity and is therefore biased
neutrophil-activating protein-78 (ENA-78) and several towards a mass range between 1–10 kDa. The other
other proteins involved in immune suppression, angio- problem is the huge dynamic range of protein concentra-
genesis and plasma extravasation were detected in pleural tions in tissues and body fluids, which exceeds more than
effusions obtained from mesothelioma patients. Proteins 10 orders of magnitude. Components of high molar abun-
that cause leukocyte infiltration and activation, such as dance generally dominate the spectrum and tend to sup-
hepatocyte growth factor (HGF), macrophage inflamma- press detection of lower abundance proteins. It is clear that
tory protein (MIP)-1d, MIP-3a, neutrophil-activating significant technological advances will be needed to over-
peptide-2 (NAP-2), and pulmonary and activation- come the huge dynamic range and sensitivity requirements
regulated chemokine (PARC) were detected.116 The pres- for probing deeper into the proteome. There is no protein
ence of these cytokines can elucidate the role of the equivalent of polymerase chain reaction (PCR) for the
immune system, tumor micro-environment and the amplification of these low-abundance proteins that are
potential immunosuppression in mesothelioma. more likely to serve as specific biomarkers. Extended
analysis to these lower abundance components will there-
fore require extensive specimen pre-fractionation before
CURRENT LIMITATIONS OF CLINICAL analysis, highly specific reagents and sophisticated instru-
PROTEOMICS mentation. The different proteomic platforms are expen-
sive and require specialized facilities with skilled staff.
Clinical proteomics faces several big challenges. One is the Currently, there is no single platform that can provide a
biological variability of patients’ samples, stemming from complete proteomic coverage to identify the entire effu-
References 147
patterns in serum to identify ovarian cancer. Lancet 2002; 359: 35. Hassan R, Remaley AT, Sampson ML, et al. Detection and
572–7. quantitation of serum mesothelin, a tumor marker for patients
11. Stoeckli M, Chaurand P, Hallahan DE, Caprioli RM. Imaging mass with mesothelioma and ovarian cancer. Clin Cancer Res 2006; 12:
spectrometry: a new technology for the analysis of protein 447–53.
expression in mammalian tissues. Nat Med 2001; 7: 493–6. ●36. Robinson BW, Creaney J, Lake R, et al. Soluble mesothelin-related
12. Emmert-Buck MR, Bonner RF, Smith PD, et al. Laser capture protein – a blood test for mesothelioma. Lung Cancer 2005;
microdissection. Science 1996; 274: 998–1001. 49(Suppl 1): S109–11.
13. Emmert-Buck MR, Gillespie JW, Paweletz CP, et al. An approach to 37. Creaney J, Robinson BW. Detection of malignant mesothelioma in
proteomic analysis of human tumors. Mol Carcinog 2000; 27: asbestos-exposed individuals: the potential role of soluble
158–65. mesothelin-related protein. Hematol Oncol Clin North Am 2005;
14. Craven RA, Totty N, Harnden P, Selby PJ, Banks RE. Laser capture 19: 1025–40, v.
microdissection and two-dimensional polyacrylamide gel 38. Gorg A, Weiss W, Dunn MJ. Current two-dimensional
electrophoresis: evaluation of tissue preparation and sample electrophoresis technology for proteomics. Proteomics 2004; 4:
limitations. Am J Pathol 2002; 160: 815–22. 3665–85.
15. Ornstein DK, Gillespie JW, Paweletz CP, et al. Proteomic analysis of 39. Hanash SM. Biomedical applications of two-dimensional
laser capture microdissected human prostate cancer and in vitro electrophoresis using immobilized pH gradients: current status.
prostate cell lines. Electrophoresis 2000; 21: 2235–42. Electrophoresis 2000; 21: 1202–9.
16. Wulfkuhle JD, McLean KC, Paweletz CP, et al. New approaches to 40. Anderson L, Anderson NG. High resolution two-dimensional
proteomic analysis of breast cancer. Proteomics 2001; 1: electrophoresis of human plasma proteins. Proc Natl Acad Sci U S
1205–15. A 1977; 74: 5421–5.
17. Jones MB, Krutzsch H, Shu H, et al. Proteomic analysis and 41. Anderson NG, Anderson NL, Tollaksen SL, et al. Analytical
identification of new biomarkers and therapeutic targets for techniques for cell fractions. XXV. Concentration and two-
invasive ovarian cancer. Proteomics 2002; 2: 76–84. dimensional electrophoretic analysis of human urinary proteins.
18. Knezevic V, Leethanakul C, Bichsel VE, et al. Proteomic profiling of Anal Biochem 1979; 95: 48–61.
the cancer microenvironment by antibody arrays. Proteomics 42. Wiederkehr F, Ogilvie A, Vonderschmitt DJ. Cerebrospinal fluid
2001; 1: 1271–8. proteins studied by two-dimensional gel electrophoresis and
19. Nakagawa T, Huang SK, Martinez SR, et al. Proteomic profiling of immunoblotting technique. J Neurochem 1987; 49: 363–72.
primary breast cancer predicts axillary lymph node metastasis. 43. Wildenauer DB, Korschenhausen D, Hoechtlen W, et al. Analysis of
Cancer Res 2006; 66: 11825–30. cerebrospinal fluid from patients with psychiatric and neurological
20. Mehta AI, Ross S, Lowenthal MS, et al. Biomarker amplification by disorders by two-dimensional electrophoresis: identification of
serum carrier protein binding. Dis Markers 2003; 19: 1–10. disease-associated polypeptides as fibrin fragments.
21. Petricoin EF, Belluco C, Araujo RP, Liotta LA. The blood peptidome: Electrophoresis 1991; 12: 487–92.
a higher dimension of information content for cancer biomarker 44. Felgenhauer K, Hagedorn D. Two-dimensional separation of human
discovery. Nat Rev Cancer 2006; 6: 961–7. body fluid proteins. Clin Chim Acta 1980; 100: 121–32.
22. Liotta LA, Petricoin EF. Serum peptidome for cancer detection: 45. Giometti CS, Tollaksen SL, Chubb C, Williams C, Huberman E.
spinning biologic trash into diagnostic gold. J Clin Invest 2006; Analysis of proteins from human breast epithelial cells using two-
116: 26–30. dimensional gel electrophoresis. Electrophoresis 1995; 16:
23. Light RW, Macgregor MI, Luchsinger PC, Ball WC, Jr. Pleural 1215–24.
effusions: the diagnostic separation of transudates and exudates. 46. Tyan YC, Wu HY, Su WC, Chen PW, Liao PC. Proteomic analysis of
Ann Intern Med 1972; 77: 507–13. human pleural effusion. Proteomics 2005; 5: 1062–74.
24. Bartter T, Santarelli R, Akers SM, Pratter MR. The evaluation of 47. Hsieh WY, Chen MW, Ho HT, You TM, Lu YT. Identification of
pleural effusion. Chest 1994; 106: 1209–14. differentially expressed proteins in human malignant pleural
◆25. Burgess LJ. Biochemical analysis of pleural, peritoneal and effusions. Eur Respir J 2006; 28: 1178–85.
pericardial effusions. Clin Chim Acta 2004; 343: 61–84. 48. Chromy BA, Gonzales AD, Perkins J, et al. Proteomic analysis of
26. Marel M, Stastny B, Melinova L, Svandova E, Light RW. Diagnosis human serum by two- dimensional differential gel electrophoresis
of pleural effusions. Experience with clinical studies, 1986 to after depletion of high-abundant proteins. J Proteome Res 2004;
1990. Chest 1995; 107: 1598–603. 3: 1120–7.
27. San Jose ME, Alvarez D, Valdes L, et al. Utility of tumour markers 49. Ahmed N, Barker G, Oliva K, et al. An approach to remove albumin
in the diagnosis of neoplastic pleural effusion. Clin Chim Acta for the proteomic analysis of low abundance biomarkers in human
1997; 265: 193–205. serum. Proteomics 2003; 3: 1980–7.
28. Raja OG, Casson IF. Fibrinogen degradation products in pleural 50. Fountoulakis M, Juranville JF, Jiang L, et al. Depletion of the high-
effusions. Br J Dis Chest 1980; 74: 164–8. abundance plasma proteins. Amino Acids 2004; 27: 249–59.
29. Riska H, Pettersson T, Froseth B, Klockars M. Beta 2 microglobulin 51. Fu Q, Bovenkamp DE, Van Eyk JE. A rapid, economical, and
in pleural effusions. Acta Med Scand 1982; 211: 45–50. reproducible method for human serum delipidation and albumin
30. Ungerer JP, Oosthuizen HM, Retief JH, Bissbort SH. Significance of and IgG removal for proteomic analysis. Methods Mol Biol 2007;
adenosine deaminase activity and its isoenzymes in tuberculous 357: 365–71.
effusions. Chest 1994; 106: 33–7. 52. Darde VM, Barderas MG, Vivanco F. Depletion of high-abundance
31. Carstens ME, Burgess LJ, Maritz FJ, Taljaard JJ. Isoenzymes of proteins in plasma by immunoaffinity subtraction for two-
adenosine deaminase in pleural effusions: a diagnostic tool? Int J dimensional difference gel electrophoresis analysis. Meth Mol Biol
Tuberc Lung Dis 1998; 2: 831–5. 2007; 357: 351–64.
32. Beyer HL, Geschwindt RD, Glover CL, et al. MESOMARKTM: A 53. Granger J, Siddiqui J, Copeland S, Remick D. Albumin depletion of
potential test for malignant pleural mesothelioma. Clin Chem human plasma also removes low abundance proteins including the
2007. cytokines. Proteomics 2005; 5: 4713–8.
●33. Scherpereel A, Grigoriu B, Conti M, et al. Soluble mesothelin- 54. Pieper R, Gatlin CL, Makusky AJ, et al. The human serum
related peptides in the diagnosis of malignant pleural proteome: display of nearly 3700 chromatographically separated
mesothelioma. Am J Respir Crit Care Med 2006; 173: 1155–60. protein spots on two-dimensional electrophoresis gels and
34. Robinson BW, Creaney J, Lake R, et al. Mesothelin-family proteins identification of 325 distinct proteins. Proteomics 2003; 3:
and diagnosis of mesothelioma. Lancet 2003; 362: 1612–6. 1345–64.
References 149
55. Sloane AJ, Duff JL, Wilson NL, et al. High throughput peptide mass 76. Steel LF, Shumpert D, Trotter M, et al. A strategy for the
fingerprinting and protein macroarray analysis using chemical comparative analysis of serum proteomes for the discovery of
printing strategies. Mol Cell Proteomics 2002; 1: 490–9. biomarkers for hepatocellular carcinoma. Proteomics 2003; 3:
56. Wu SL, Amato H, Biringer R, et al. Targeted proteomics of low- 601–9.
level proteins in human plasma by LC/MSn: using human growth 77. Tachibana M, Ohkura Y, Kobayashi Y, et al. Expression of
hormone as a model system. J Proteome Res 2002; 1: 459–65. apolipoprotein A1 in colonic adenocarcinoma. Anticancer Res
57. Choudhary G, Wu SL, Shieh P, Hancock WS. Multiple enzymatic 2003; 23: 4161–7.
digestion for enhanced sequence coverage of proteins in complex 78. Hellman K, Alaiya AA, Schedvins K, et al. Protein expression
proteomic mixtures using capillary LC with ion trap MS/MS. J patterns in primary carcinoma of the vagina. Br J Cancer 2004; 91:
Proteome Res 2003; 2: 59–67. 319–26.
58. Sanchez JC, Appel RD, Golaz O, et al. Inside SWISS-2DPAGE 79. Zhang Z, Bast RC Jr, Yu Y, et al. Three biomarkers identified from
database. Electrophoresis 1995; 16: 1131–51. serum proteomic analysis for the detection of early stage ovarian
59. Adkins JN, Varnum SM, Auberry KJ, et al. Toward a human blood cancer. Cancer Res 2004; 64: 5882–90.
serum proteome: analysis by multidimensional separation coupled 80. Kuesel AC, Kroft T, Prefontaine M, Smith IC. Lipoprotein(a) and
with mass spectrometry. Mol Cell Proteomics 2002; 1: 947–55. CA125 levels in the plasma of patients with benign and malignant
60. Tonge R, Shaw J, Middleton B, et al. Validation and development ovarian disease. Int J Cancer 1992; 52: 341–6.
of fluorescence two-dimensional differential gel electrophoresis 81. Ross PL, Huang YN, Marchese JN, et al. Multiplexed protein
proteomics technology. Proteomics 2001; 1: 377–96. quantitation in Saccharomyces cerevisiae using amine-reactive
61. Von Eggeling F, Gawriljuk A, Fiedler W, et al. Fluorescent dual isobaric tagging reagents. Mol Cell Proteomics 2004; 3: 1154–69.
colour 2D-protein gel electrophoresis for rapid detection of 82. Ong SE, Blagoev B, Kratchmarova I, et al. Stable isotope labeling
differences in protein pattern with standard image analysis by amino acids in cell culture, SILAC, as a simple and accurate
software. Int J Mol Med 2001; 8: 373–7. approach to expression proteomics. Mol Cell Proteomics 2002; 1:
62. Yan JX, Devenish AT, Wait R, et al. Fluorescence two-dimensional 376–86.
difference gel electrophoresis and mass spectrometry based 83. Higgs RE, Knierman MD, Gelfanova V, Butler JP, Hale JE.
proteomic analysis of Escherichia coli. Proteomics 2002; 2: Comprehensive label-free method for the relative quantification of
1682–98. proteins from biological samples. J Proteome Res 2005; 4:
63. Nordvarg H, Flensburg J, Ronn O, et al. A proteomics approach to 1442–50.
the study of absorption, distribution, metabolism, excretion, and 84. Wiener MC, Sachs JR, Deyanova EG, Yates NA. Differential mass
toxicity. J Biomol Tech 2004; 15: 265–75. spectrometry: a label-free LC-MS method for finding significant
64. Alban A, David SO, Bjorkesten L, et al. A novel experimental design differences in complex peptide and protein mixtures. Anal Chem
for comparative two-dimensional gel analysis: two-dimensional 2004; 76: 6085–96.
difference gel electrophoresis incorporating a pooled internal 85. Gygi SP, Rist B, Gerber SA, et al. Quantitative analysis of complex
standard. Proteomics 2003; 3: 36–44. protein mixtures using isotope-coded affinity tags. Nat Biotechnol
65. Unlu M, Morgan ME, Minden JS. Difference gel electrophoresis: a 1999; 17: 994–9.
single gel method for detecting changes in protein extracts. 86. Karas M, Hillenkamp F. Laser desorption ionization of proteins
Electrophoresis 1997; 18: 2071–7. with molecular masses exceeding 10,000 daltons. Anal Chem
66. Patton WF. Detection technologies in proteome analysis. J 1988; 60: 2299–301.
Chromatogr B Analyt Technol Biomed Life Sci 2002; 771: 3–31. 87. Hillenkamp F, Karas M. Mass spectrometry of peptides and
67. Lilley KS, Friedman DB. All about DIGE: quantification technology proteins by matrix-assisted ultraviolet laser desorption/ionization.
for differential-display 2D-gel proteomics. Expert Rev Proteomics Methods Enzymol 1990; 193: 280–95.
2004; 1: 401–9. 88. Cho A, Normile D. Nobel Prize in Chemistry. Mastering
68. Morita A, Miyagi E, Yasumitsu H, et al. Proteomic search for macromolecules. Science 2002; 298: 527–8.
potential diagnostic markers and therapeutic targets for ovarian 89. Li C, Chen Z, Xiao Z, et al. Comparative proteomics analysis of
clear cell adenocarcinoma. Proteomics 2006; 6: 5880–90. human lung squamous carcinoma. Biochem Biophys Res Commun
69. Zhou G, Li H, DeCamp D, et al. 2D differential in-gel 2003; 309: 253–60.
electrophoresis for the identification of esophageal scans cell 90. Chen G, Wang H, Gharib TG, et al. Overexpression of oncoprotein
cancer-specific protein markers. Mol Cell Proteomics 2002; 1: 18 correlates with poor differentiation in lung adenocarcinomas.
117–24. Mol Cell Proteomics 2003; 2: 107–16.
70. Alfonso P, Nunez A, Madoz-Gurpide J, et al. Proteomic expression 91. Chen G, Gharib TG, Thomas DG, et al. Proteomic analysis of eIF-5A
analysis of colorectal cancer by two-dimensional differential gel in lung adenocarcinomas. Proteomics 2003; 3: 496–504.
electrophoresis. Proteomics 2005; 5: 2602–11. 92. Chen G, Gharib TG, Wang H, et al. Protein profiles associated with
71. Lee IN, Chen CH, Sheu JC, et al. Identification of human survival in lung adenocarcinoma. Proc Natl Acad Sci U S A 2003;
hepatocellular carcinoma-related biomarkers by two-dimensional 100: 13537–42.
difference gel electrophoresis and mass spectrometry. J Proteome 93. Chen G, Gharib TG, Huang CC, et al. Proteomic analysis of lung
Res 2005; 4: 2062–9. adenocarcinoma: identification of a highly expressed set of
72. Yu KH, Rustgi AK, Blair IA. Characterization of proteins in human proteins in tumors. Clin Cancer Res 2002; 8: 2298–305.
pancreatic cancer serum using differential gel electrophoresis and 94. Bard MP, Hegmans JP, Hemmes A, et al. Proteomic analysis of
tandem mass spectrometry. J Proteome Res 2005; 4: 1742–51. exosomes isolated from human malignant pleural effusions. Am J
73. Nayak SS, Kamath SS, Kundaje GN, Aroor AR. Diagnostic Respir Cell Mol Biol 2004; 31: 114–21.
significance of estimation of serum apolipoprotein A along with 95. Diamandis EP. Serum proteomic profiling by matrix-assisted laser
alpha-fetoprotein in alcoholic cirrhosis and hepatocellular desorption-ionization time-of-flight mass spectrometry for cancer
carcinoma patients. Clin Chim Acta 1988; 173: 157–64. diagnosis: next steps. Cancer Res 2006; 66: 5540–1.
74. Matsuura T, Koga S, Ibayashi H. Increased proportion of 96. Merchant M, Weinberger SR. Recent advancements in surface-
proapolipoprotein A-I in HDL from patients with liver cirrhosis and enhanced laser desorption/ionization-time of flight-mass
hepatitis. Gastroenterol Jpn 1988; 23: 394–400. spectrometry. Electrophoresis 2000; 21: 1164–77.
75. Fujii S, Koga S, Shono T, Yamamoto K, Ibayashi H. Serum 97. Oh JH, Nandi A, Gurnani P, et al. Proteomic biomarker
apoprotein A-I and A-II levels in liver diseases and cholestasis. Clin identification for diagnosis of early relapse in ovarian cancer. J
Chim Acta 1981; 115: 321–31. Bioinform Comput Biol 2006; 4: 1159–79.
150 Proteomics in pleural disease
98. Ciordia S, de Los Rios V, Albar JP. Contributions of advanced 111. Nitadori J, Ishii G, Tsuta K, et al. Immunohistochemical differential
proteomics technologies to cancer diagnosis. Clin Transl Oncol diagnosis between large cell neuroendocrine carcinoma and small
2006; 8: 566–80. cell carcinoma by tissue microarray analysis with a large antibody
99. Oh JH, Gao J, Nandi A, et al. Diagnosis of early relapse in ovarian panel. Am J Clin Pathol 2006; 125: 682–92.
cancer using serum proteomic profiling. Genome Inform 2005; 16: 112. Song S, Li B, Wang L, et al. A cancer protein microarray platform
195–204. using antibody fragments and its clinical applications. Mol Biosyst
100. Ho DW, Yang ZF, Wong BY, et al. Surface-enhanced laser 2007; 3: 151–8.
desorption/ionization time-of-flight mass spectrometry serum 113. Srivastava M, Eidelman O, Jozwik C, et al. Serum proteomic
protein profiling to identify nasopharyngeal carcinoma. Cancer signature for cystic fibrosis using an antibody microarray platform.
2006; 107: 99–107. Mol Genet Metab 2006; 87: 303–10.
101. Gourin CG, Xia ZS, Han Y, et al. Serum protein profile analysis in 114. Orchekowski R, Hamelinck D, Li L, et al. Antibody microarray
patients with head and neck squamous cell carcinoma. Arch profiling reveals individual and combined serum proteins
Otolaryngol Head Neck Surg 2006; 132: 390–7. associated with pancreatic cancer. Cancer Res 2005; 65:
102. Engwegen JY, Gast MC, Schellens JH, Beijnen JH. Clinical 11193–202.
proteomics: searching for better tumour markers with SELDI-TOF 115. Gao WM, Kuick R, Orchekowski RP, et al. Distinctive serum protein
mass spectrometry. Trends Pharmacol Sci 2006; 27: 251–9. profiles involving abundant proteins in lung cancer patients based
103. Rubin RB, Merchant M. A rapid protein profiling system that upon antibody microarray analysis. BMC Cancer 2005; 5: 110.
speeds study of cancer and other diseases. Am Clin Lab 2000; 19: 116. Hegmans JP, Hemmes A, Hammad H, et al. Mesothelioma
28–9. environment comprises cytokines and T-regulatory cells that
104. Aebersold R, Mann M. Mass spectrometry-based proteomics. suppress immune responses. Eur Respir J 2006; 27: 1086–95.
Nature 2003; 422: 198–207. 117. Shen Y, Smith RD, Unger KK, Kumar D, Lubda D. Ultrahigh-
105. Washburn MP, Wolters D, Yates JR 3rd. Large-scale analysis of the throughput proteomics using fast RPLC separations with ESI-
yeast proteome by multidimensional protein identification MS/MS. Anal Chem 2005; 77: 6692–701.
technology. Nat Biotechnol 2001; 19: 242–7. 118. Shen Y, Smith RD. Advanced nanoscale separations and mass
106. Kusnezow W, Hoheisel JD. Antibody microarrays: promises and spectrometry for sensitive high-throughput proteomics. Expert Rev
problems. Biotechniques 2002; Suppl: 14–23. Proteomics 2005; 2: 431–47.
107. James P. Chips for proteomics: a new tool or just hype? 119. Li J, LeRiche T, Tremblay TL, et al. Application of microfluidic
Biotechniques 2002; Suppl: 4–10, 12–13. devices to proteomics research: identification of trace-level
108. Ekins RP, Chu FW. Multianalyte microspot immunoassay – protein digests and affinity capture of target peptides. Mol Cell
microanalytical “compact disk” of the future. Clin Chem 1991; 37: Proteomics 2002; 1: 157–68.
1955–67. 120. MacBeath G. Proteomics comes to the surface. Nat Biotechnol
109. Anderson NL, Anderson NG. The human plasma proteome: history, 2001; 19: 828–9.
character, and diagnostic prospects. Mol Cell Proteomics 2002; 1: ◆121. Mischak H, Apweiler R, Banks RE, et al. Clinical proteomics: A need
845–67. to define the field and to begin to set adequate standards.
110. Kusnezow W, Syagailo YV, Goychuk I, Hoheisel JD, Wild DG. Proteomics Clin Appl 2007; 1: 148–156.
Antibody microarrays: the crucial impact of mass transport on
assay kinetics and sensitivity. Expert Rev Mol Diagn 2006; 6:
111–24.
13
Pleural pharmacokinetics
Plasma concentration
nent to discussion of drug delivery to the pleural space.
Since intravenous drug administration is most often the
approach used when treating pleural disease, the intrica-
cies of the movement of drug from the site of administra-
tion to the systemic circulation, known as absorption,
will not be addressed. For the purposes of this discussion, AUC
it is understood that intravenous administration consti-
tutes immediate placement of the agent into the systemic
circulation. Time
Vd = Dose/Cpmax
Central
compartment
The Vd is a property that is inherent to each drug.
However, it may be affected by disease states that alter total A0
Drug
body volume and protein stores that change distribution A1
characteristics of the drug. The Vd may range from one to
several liters, depending on the degree of tissue penetra-
tion and binding of an individual agent. For example, it
may be concluded that the distribution of a drug with a Vd k
similar to the amount of total body extracellular fluid
(15–18 L in a 70 kg person) is limited mainly to the
extravascular space. A drug with a Vd that exceeds the total
amount of body water is thought to have a large affinity for Figure 13.3 A one-compartment model. A0, amount of drug
the tissues and other extravascular spaces. administered; A1, amount of drug in body; k, elimination rate
It follows, therefore, that a drug with a large Vd at dis- constant.
tribution equilibrium would be expected to have a small
plasma concentration and the reverse is also true. Factors
associated with decreased plasma drug concentrations, distribution equilibrium is reached in those spaces are
such as decreased plasma protein binding, would conse- similar.
quently result in a larger Vd. Generally speaking, a drug Two-compartment and multi-compartment models
with a large Vd is thought to have a high degree of distri- are useful when characterizing drugs that equilibrate at
bution to extravascular tissues, spaces and fluids and a different rates to various regions of the body. The periph-
smaller plasma concentration. It must be kept in mind, eral compartment or second compartment, as denoted in
however, that the Vd does not provide insight into the a two-compartment model, includes less well-perfused
exact distribution sites, it simply estimates the affinity of organs, tissues and spaces that are not in rapid equilibrium
the drug for extravascular areas of the body. with the plasma, such as muscle and fat. Membrane pene-
tration, protein binding and other physiochemical proper-
ties slow distribution to sites in the second compartment.
Compartmental models A three-compartment model represents distribution from
the central compartment to two distinct peripheral com-
The human body is undoubtedly a very complex system partments following administration.
composed of a multitude of different tissues, spaces and In a two-compartment model, once the drug is distrib-
fluids. Compartment models are used in the area of phar- uted throughout the first compartment, the plasma con-
macokinetics to simplify the complexity of drug distribu- centration will decline slightly, reflecting a slow and
tion and elimination in order to allow the application of reversible distribution to the second compartment (Figure
mathematical principles to depict and quantify the time- 13.4). The rate constants, k12 and k21, quantify the
course of drug concentrations in the body. Spaces, tissues reversible distribution of the drug from the central com-
and fluids with similar types of drug distribution charac- partment to and from the peripheral compartment. As
teristics are grouped into the same compartment.
The one-compartment model is the most straightfor-
ward and holds the underlying assumption that the body
behaves as one large container (Figure 13.3). The central
Central Peripheral
compartment or first compartment is thought to be con- compartment compartment
gruent with the systemic circulation and consists of the A0 k12
Drug A1 A2
most highly perfused tissues and organs such as the heart,
kidneys and lungs. Since most drugs are metabolized and k21
eliminated via the liver and kidneys, elimination is
accepted to occur strictly from the first compartment.
Drugs that display one-compartment model behavior are
k10
assumed to be distributed and equilibrate very rapidly
throughout the body following administration. From
another perspective, areas in the body that may be consid- Figure 13.4 A two-compartment model. A0, amount of drug
ered as parts of the central compartment contain drug administered; A1, amount of drug in body; k10, elimination rate
concentrations that equilibrate at a rate that is propor- constant from the body; k12, rate constant of transfer from the
tionate to drug concentrations in the plasma. This is not central compartment to the peripheral compartment; k21, rate
to say that drug concentrations within spaces of a single constant of transfer from the peripheral compartment to the
compartment are identical, simply that the rate at which central compartment.
154 Pleural pharmacokinetics
stated previously, most drugs must be transferred to the elimination and the pharmacokinetic parameters do not
central compartment to be eliminated.1–4 change with the size of the dose given.
Examination of the plasma concentration–time profile Although the amount eliminated changes with time in
of drugs that follow multi-compartment pharmacokinetics a first-order process, the fraction of drug eliminated over a
reveals a curve with varying degrees of slope that corre- specific interval remains constant. This fraction of drug
spond to the various rates at which the drug is distributed eliminated over time is referred to as the elimination rate
and eliminated from each compartment (Figure 13.5). The constant (k). A plot of the drug plasma concentration–
number of compartments required to describe the behav- time profile for a first-order drug reveals a curve and a plot
ior of a drug is determined by the number of exponential of the natural log of the plasma concentration–time profile
terms that may be applied to the plasma concentration– will produce a straight line (Figure 13.6). As a result, avail-
time data. Three or more compartments may be needed to able plasma drug concentration data may be used to calcu-
accurately depict the data generated by some agents in the late pharmacokinetic parameters and predict plasma drug
plasma. concentrations and elimination over time.1–4
100
100
Log plasma concentration
Plasma concentration
distribution elimination 80
60
(mg/mL)
10 elimination distribution 40
20
0
1 0 1 2 3 4 5 6
0 2 4 6 8 10 12 Time (h)
Time (h)
100
Log plasma concentration
10
Elimination
allows for the calculation of k when two plasma concentra- PHARMACOKINETICS SPECIFIC TO THE
tion timepoints are known for a drug following first-order PLEURAL FLUID
elimination:
Introduction
k = ln Cp2 - ln Cp1/t2 - t1
The pharmacokinetics of drugs in the pleural space is not
where D ln Cp = the natural log of plasma concentration, a well-studied or established topic. Although there are
Dt = the corresponding values for time, and k = the elimi- several published studies examining antimicrobial concen-
nation rate constant. trations in pleural fluid and the intrapleural administra-
The half-life (t1/2) is the time it takes for the plasma tion of antineoplastic agents, many of these studies are not
concentration to be reduced by 50 percent (Figure 13.7). well controlled and consist of a small number of patients
Practically speaking, it takes five half-lives for the drug to with a wide variety of malignancies and pleural effusion
be 97 percent removed from the body. The plasma t1/2 may states. In order to best serve the purpose of this chapter,
be calculated by the formula: the basic principles of drug permeation through mem-
branes will be considered along with theoretical explo-
t1/2 = 0.693/k ration of drug delivery to the pleura combined with the
published data to form educated conclusions.
Clearance (Cl) refers to the inherent ability of the elim- Exudative effusions associated with infection and
inating organs of the body to remove drug and metabolites malignancy comprise the realm of pleural diseases in
from the blood per unit time. It is a function of capacity of which the target site of drug therapy is clearly the pleura/
organs such as the kidney and liver to process drugs and pleural space. Penetration of adequate drug concentra-
the degree of blood flow that presents these organs with tions directly to the pleura/pleural fluid is paramount in
drugs to eliminate. Clearance does not represent the the clinical picture of parapneumonic effusions, empyema,
amount of drug removed but the volume of blood that is malignant effusions and malignant mesothelioma. Hence,
cleared of drug over a certain interval of time. The amount our discussion will focus on penetration of antimicrobial
of drug cleared depends on the concentration of drug in and antineoplastic agents to the pleura/pleural fluid and
the plasma as well as the rate of blood flow to eliminating the intrapleural administration of these agents.3–7
organs. The extraction ratio is the fraction of drug pre-
sented to an eliminating organ that is cleared after one pass
through that organ. Many factors may affect clearance, Drug membrane penetration and protein
such as the extraction ratio, the body surface area of the binding: general considerations
patient, plasma protein binding, renal and hepatic func-
tion and the cardiac output of the patient, which controls The ability of a drug to ultimately reach the pleura/pleural
blood flow to the eliminating organs. When the rate of fluid from the plasma is influenced by several key factors.
administration is equal to the clearance, such as may be the These factors include, but are not limited to: (1) plasma
case in continuous infusion, the concentration in the drug concentration; (2) plasma and tissue protein binding;
plasma may remain constant or achieve steady-state (Cpss). (3) pleural membrane penetration; (4) blood perfusion of
The AUC, which accounts for total drug exposure, may be the pleura; and (5) effects of pleural disease on vascular
calculated if the clearance is known.1–4 permeability, membrane permeability and protein
binding.
AUC = Dose/Cl Diffusion down a concentration gradient is the mecha-
nism by which most drugs move out of the plasma, into
Cpmax the interstitial fluid and across membranes to the target
100 site of action. The size of the concentration gradient that
Log plasma concentration
t1/2
will power diffusion is dependent on the initial concentra-
tion achieved in the plasma. Since distribution and elimi-
10 nation commence simultaneously and immediately
following administration, it is imperative that a drug be
AUC dosed properly to achieve acceptable initial plasma con-
centrations. For example, concern should exist about
1
0 1 2 3 4 5 6 7
using an antibiotic of which initial plasma concentrations
Time (h)
barely exceed the minimum inhibitory concentration
(MIC) of a targeted pathogen in the pleural space. Because
Figure 13.7 First-order elimination of a one-compartment there may be hindrances of distribution to the site of
drug. AUC, area under the curve; Cpmax, maximum plasma action and the continuous impact of drug elimination, the
concentration; t1/2, half-life (the time it takes for the plasma concentration that reaches a site of interest may be much
concentration to be reduced by 50 percent). lower than that in the plasma.
156 Pleural pharmacokinetics
PROTEIN BINDING
Plasma Tissue
Protein binding usually refers to serum or plasma protein
binding. Drugs may bind to tissue proteins but this has not Db Db
been studied and this section is therefore only concerned
with serum or plasma protein binding. Serum drug ≠ ≠
protein binding is generally not an issue with which the cli-
nician needs to be concerned. Once an agent reaches the Df Df
market, recommended dosage regimens are based on clin-
ical efficacy with the agent’s capacity for protein binding
already taken into consideration. However, the implica- Figure 13.8 Plasma protein binding. Db, drug bound to protein;
tions of protein binding become very relevant when focus- Df, free drug.
ing on a specific site of penetration. Biological proteins
such as albumin are large and bulky molecules with low
lipid solubility. Their size hinders passage through capil-
laries and low lipophilicity slows transfer across cell mem-
branes. A drug that is protein bound is similarly detained have a very high degree of protein binding in order for
from transfer to tissues. It is widely accepted that the fluctuating drug concentrations occurring over time to
unbound or free drug is the active moiety available for have a direct impact on the fraction bound in the plasma.
equilibration between intravascular and extravascular It has been determined that a drug must exhibit at least
sites. Further, it is believed that a drug must dissociate 85–90 percent plasma protein binding until a change in
from protein in order to exert any type of pharmacological plasma concentration leads to a substantial impact in the
activity. Protein binding has the propensity to affect three free drug available for distribution.5,9,10 Peterson and
main pharmacokinetic/pharmacodynamic concerns: (1) Gerding10 found that plasma protein binding in an anti-
the propensity of the drug to distribute to body tissues, biotic must approach 90 percent in order to have any sig-
spaces and fluids; (2) the amount of drug that is available nificant effect on free drug concentration. Once a drug is
to be eliminated; and (3) interaction of the drug with the this highly protein bound, a smaller fraction may be
site of action.8 available to move into the tissue with lower plasma con-
The plasma contains three proteins that have an affin- centrations. In other words, a drug must exhibit greater
ity for binding drugs including albumin (normal concen- than 85 percent protein binding in order for protein
tration 35–45 g/L), a-1-acid glycoprotein (normal binding to affect the drug pharmacokinetics to any signif-
concentration 0.4–1 g/L) and lipoproteins (variable con- icant degree.
centrations). Assaying techniques available for most However, the degree of plasma protein binding does
agents generate total plasma drug concentration measure- not efficiently predict the tissue distribution or half-life of
ments as opposed to a free drug concentration. Thus, a an agent. In fact, it has been found that for drugs that have
high peak plasma concentration of a highly protein- a relatively short t1/2(<2–3 h), a high degree of protein
bound compound may be deceiving in that it does not binding (>90 percent) may contribute to persistence in an
accurately reflect the free drug that is readily available for extravascular space and therefore prolong subsistence
tissue distribution and activity. An inverse relationship periods and enhance tissue distribution.9
exists between plasma protein binding and the Vd. Most The implications of a high degree of protein binding
body fluids such as lymph, interstitial fluid and the may not always be negative when considering protein con-
normal pleural fluid contain much lower protein concen- tained in tissue spaces. Proteins do have the ability to
trations compared with the plasma and will therefore diffuse through tissues, albeit slowly. Although protein
reflect lower total drug concentrations for highly protein binding may hinder drug delivery, this also holds true for
bound drugs. clearance. Theoretically, if a pleural effusion contains a
Protein binding of a drug is rapidly reversible (Figure large amount of protein, a drug to which protein has a
13.8). The degree to which an agent is bound depends on high affinity may become ‘trapped’ in the effusion, leading
the affinity of the protein for the drug and the concentra- to prolonged drug exposure and delayed pleural clearance.
tions of the binding protein compared with the drug.
Drug binding to plasma proteins may range from 0 to 99 MEMBRANE PENETRATION
percent. A fair amount of controversy exists regarding the
true impact of plasma protein binding on drug distribu- Drug molecules must move through a series of spaces and
tion and limitations in the utility of drugs that are highly membranes in order to reach a target site of action and
protein bound. In most cases, the fraction of drug that is conversely, move back into the systemic circulation in
free is consistent and does not vary with differing drug order to be cleared from the body. The three modes by
concentrations because the number of protein binding which drug molecules accomplish this include: (1) diffu-
sites far exceeds the drug concentrations. A drug must sion through the capillary pores; (2) penetration through
Pharmacokinetics specific to the pleural fluid 157
capillary membranes; and (3) bulk movement through Drug penetration to normal pleura cavity:
capillary pores.7 In order to move back into the systemic theoretical discussion
circulation to be cleared from the body, a drug molecule
may do so by diffusion through the capillary pores, pene- From a theoretical perspective, it seems logical that drug
tration of capillary membrane and by bulk flow or diffu- delivery and elimination to and from the normal pleural
sion through the lymphatics. The route of drug molecule space may follow the same route as the pleural fluid. The
transfer is determined by the number and size of available pleural space is the cavity that lies between the visceral
pores and by physicochemical properties of the agent. and parietal pleura and, in healthy humans, contains
These physicochemical properties include the molecular approximately 0.26 mL/kg of pleural fluid (approximately
size and weight, the degree of lipophilicity and the degree 18 mL in a 70 kg man) that is relatively low in protein
of ionization.1,3,5–7 content (approximately 10 mg/mL).12 The source of
Drug molecules with a molecular weight of <1000 will pleural fluid is the pulmonary and extrapleural parietal
generally move through capillary and lymphatic pores interstitium, with a larger contribution from the latter.
with ease. Large, bulky molecules such as albumin must The fluid flows down a small pressure gradient at a rate
pass through capillary membranes by endocytosis or presumed to be approximately 0.02 mL/kg.hour and is
pinocytosis and have difficulty crossing lymphatic drained mainly through the parietal lymphatics back into
endothelium. Equilibrium between the plasma and lymph the systemic circulation.13–16
compartment of albumin was found to be realized only The pleural space may be thought of as an enlarged
after 24 hours.11 Agents that are bound to albumin are tissue compartment distinct from the lung parenchyma.
similarly impeded. Small, lipophilic, unionized molecules Drug delivery to this space may be influenced by five other
tend to pass easily through cell membranes, but the larger entities that include the parietal systemic capillaries, the
and more polar the molecule, the greater difficulty of parietal interstitium, the pleural space, the lung intersti-
membrane passage.1,3,5–7 tium and the visceral systemic capillaries. Because humans
Ionization of a drug molecule complicates passage have a thick visceral pleura as opposed to the thin visceral
through tissue membranes. Most drugs exist in solution as pleura found in other animals, the blood supply to both
either weak acids or weak bases and equilibrium exists the visceral and parietal pleura is provided by the systemic
between ionized and unionized fractions. Accumulation of circulation. Therefore, the proposed route of transfer of
an agent on the side of a membrane with a pH that favors drug molecules to the pleural space is directly from the sys-
greater ionization may prevent transfer. This is generally temic circulation through capillaries and through the vis-
not the case as the pH in tissues throughout the body is ceral and parietal interstitium. Although pleural fluid is
close to the physiological pH of 7.4. However, the pH asso- thought to enter the pleural space mainly from the parietal
ciated with an infected area tends to be lower and alkaline pleura, drug may potentially enter from both the visceral
antibiotics may become trapped in infected areas.1,3 and parietal systemic capillaries. The distance of the capil-
As stated previously, the penetration of a drug molecule laries from the visceral pleural membrane is approximately
to a specific site of action depends largely on the molecu- 56 mm compared with the shorter distance of 10–12 mm in
lar size and weight. Most antibiotics are relatively small the parietal pleura. Considering this with the pressure gra-
with a molecular weight of less than 1000 with the excep- dient and relative thickness of the visceral pleura, the most
tion of vancomycin (MW 1485.75). The antineoplastic direct route giving higher drug delivery is most likely
agents that would generally be used in the arena of malig- through the parietal pleura.
nant effusions also have molecular weights of less than Adequate penetration of drug molecules into the
1000. Some agents, however, are larger macromolecules. pleural space requires passage across the first barrier of the
Antibiotics that are poorly lipid-soluble include the peni- capillary endothelium, through the interstitium, then
cillins, cephalosporins, aminoglycosides, vancomycin and across the second barrier of the mesothelium, which lines
clindamycin. Agents such as metronidazole, rifampin, the pleural cavity. Drug concentrations in the capillaries
trimethoprim and the quinolones are lipid-soluble and are consistent with the plasma and the interstitial fluid is in
may have the propensity to cross cell membranes with rapid equilibrium with the plasma. Potential sources of
more ease. diminishment of drug penetration into the pleural space
Although free drug is believed to be the active moiety include the endothelial cell and mesothelial cell membrane
and protein binding slows the penetration of drugs barriers and the interstitium.13–16
through membranes, it must be kept in mind that protein The capillary endothelium is known to be relatively
binding is rapidly reversible. Although transfer is slow, permeable. Water is found to egress from capillaries
protein molecules such as albumin do have the ability to through interendothelial junctions and albumin must pass
penetrate to the extravascular space and potentially have through by either endocytosis or pinocytosis. Unbound,
the capacity to slow diffusion of drug out of a target drug lipid-soluble drug may pass directly through the cell mem-
site due to binding within that site. Table 13.1 lists the brane and water-soluble drug will easily pass through the
protein binding characteristics and molecular weights of inter-endothelial junctions. Protein-bound drug will
agents of interest. follow the same, slower route as albumin.
158 Pleural pharmacokinetics
Aminoglycosides
Gentamicin NA 0–30
Amikacin 781.75 4–11
Tobramycin 1425.45 0–30
Penicillins
Penicillin G potassium 372.48 65
Ampicillin/sulbactam 371.39/255.22 28/38
Piperacillin/tazobactam 539.6/322.3 26–33/26
Carbapenems
Meropenem 437.52 2
Imipenem/cilastin 317.37/380.43 20/40
Cephalosporins
Cefazolin 476.48 74–86
Cefoxitin 449.44 65–79
Cefotaxime 477.4 31–50
Cefuroxime 446.37 33–50
Ceftriaxone 661.59 85–90
Ceftazidime 636.6 17
Cefepime 571.5 16–19
Macrolides
Erythromycin 733.94 75–90
Azithromycin 785 7–50 (concentration dependent)
Clarithromycin 747.96 42–50
Quinolones
Ciprofloxacin 331.4 16–43
Ofloxacin 361.4 20
Levofloxacin 370.38 50
Moxifloxacin 437.9 50
Gatifloxacin 402.42 20
Miscellaneous
Vancomycin 1485.75 10–50
Linezolid 337.35 31
Tigecycline 585.65 71–89
Clindamycin 504.96 94
Metronidazole 171.16 <20
Rifampin 822.95 80
Isoniazid 137.14 10–15
Antineoplastics
Cyclophosphamide 279.1 60a
Cisplatin 300.1 Platinum 90b
Carboplatin 371.25 0 (platinum 30 irreversible)c
Doxorubicin 579.99 74–76
Etoposide 588.58 97
Mitoxantrone 517.41 78
Pemetrexed 597.49 73–81
NA, not applicable.
aRefers to the protein binding of cyclophosphamide metabolites including but not limited to 4-hydroxycyclophoshphamide, aldophosphamide,
The next obstacle is the interstitium, which is basically may wax and wane and is not predictably consistent. In the
a meshwork of collagen and fibers combined with glyco- case of a parapneumonic effusion, the chemistry of the
polysaccharides and glucosaminoglycans. The intersti- pleural fluid itself would be expected to change, consisting
tium, to all intents and purposes, is similar to a gel through of higher protein levels, a higher population of cells and, in
which drug filtration is determined by molecular size, the case of empyema, a lower pH and lower glucose level.
weight and charge. Interestingly, protein-bound drug will These changes would influence drug pharmacokinet-
pass through the interstitium at a faster rate than unbound ics.13,16,18,19
drug. The final barrier to pleural drug delivery is the In the case of malignancy, obstructive tumors may
mesothelium. In contrast to earlier reports of the mesothe- block lymphatic drainage of the pleural space. Tumor
lium as a very ‘leaky membrane’, Negrini and colleagues17 characteristics that determine the degree and type of
determined that the mesothelium in animals is very similar blockage cannot be expected to be uniform among
to the endothelium of ‘continuous capillaries’. The various patients. Different malignancy in the pleural space will
sized pores found in the mesothelium have been deter- therefore affect the pharmacokinetics of each drug differ-
mined to be larger and more penetrable to albumin than ently for each patient.13,19
the capillary endothelium. The limit of passage through The pharmacokinetics of agents in the pleural space will
the mesothelium appears to be the number of pores avail- be subject to variance based on the drug penetration char-
able, hence the proposed reason given by the investigators acteristics as well as the underlying condition of the
for the relatively low protein content of the pleural fluid. If patient. Penetration and elimination of drugs to and from
similar properties correspond to human mesothelium, the pleura would be expected to be markedly different in
drug passage through this barrier would be similar to that healthy versus diseased tissue. Inflammation may be
as through the capillary endothelium.17 expected to increase drug penetration and lymphatic
From a pharmacokinetic viewpoint, it seems that the drainage saturation or obstruction would be expected to
pleural space may be considered to be a part of the periph- impede drug clearance. This would result in higher Cmax
eral compartment in a two-compartment model. levels and prolonged elimination rate constants (k) and
Although the visceral and parietal pleura are well perfused, half-life (t1/2) observed in the diseased pleural space com-
concentrations of drug in the pleural space may lag behind pared with the healthy pleural space.
those of the serum because of the necessity of crossing two
membrane barriers and permeating the interstitium.
Stomata are believed to exist in the human parietal Intrapleural administration: theoretical
pleura that lead to lymphatic lacuna, which provide unidi- discussion
rectional drainage of the pleural space. The maximum
capacity of drainage that may be provided for by the pari- If one considers the intrapleural administration of drug,
etal lymph is thought to be around 1 mL/kg.hour. The one must envision administration directly to the periph-
route of drug elimination from the pleural space may also eral compartment of a multi-compartment model (Figure
be presumed to be through the parietal lymphatic system, 13.4). The tendency would be for the drug to equilibrate
consistent with that of the pleural fluid.13,16 with other tissues and spaces, including the central com-
partment. The movement of the agent would follow the
basic principles of pharmacokinetics in that the drug
Penetration from the plasma to the diseased would seek to diffuse down a concentration gradient from
pleura areas of higher concentration to equilibrate with other
spaces of the body that are at lower concentrations.
The preceding proposed pathway of drug delivery to the Distribution would be reversible and occur at the rate con-
pleural space seems logical. The problem with its concep- stants k12 and k21.
tion is that the entire topography of the pleural cavity and Elimination from the body would most likely take place
surrounding tissue changes in the event of an exudative strictly from the central compartment. Distribution equi-
pleural effusion – the type of effusion most often experi- librium might not be reached since elimination would
enced with infection or malignancy. By definition, the per- consistently be taking place from the central compart-
meability characteristics and clearance mechanisms of the ment, decreasing drug levels in the systemic circulation. A
pleural cavity are completely altered in this clinical setting. drug would continue to move with this type of behavior
Fluid accumulation in the face of an exudative pleural until it was fully eliminated from the body.
effusion is thought to be due to increased permeability Since the pleural fluid contains much less protein than
and either saturated or blocked lymphatic drainage. the plasma, there is less protein binding that would
Inflammation that accompanies either malignancy or inhibit transport of drug out of the pleural space. Access
infection may potentially increase the vascular permeabil- to the systemic circulation would be available by two
ity and the permeability of the mesothelial cells. Therefore, routes. The first route would be through the mesothelium
inflammation may probably lead to greater drug levels lining of the pleura into the interstitium, followed by
achieved in the pleural space. This inflammation, however, transfer across the capillary endothelium into the systemic
160 Pleural pharmacokinetics
circulation via either the parietal or visceral pleura. The maximum concentrations attainable. It is clear that anti-
second route would be through the lymphatics into the biotic pharmacokinetics in the pleural space is an area that
systemic circulation. The route, once again, would be requires further study.
expected to be the same as that of the pleural fluid. The Despite the limitations of the available data, several
rate-limiting factor that would slow drug transfer from conclusions may be made regarding general principles of
the pleural space to the systemic circulation is the number antibiotic pharmacokinetics in the pleural fluid relative to
of stomata present on the parietal pleura and the patency that of the plasma.
of the downstream lymphatic drainage. Fluid is thought Selected studies will be reviewed to illustrate some key
to accumulate in the pleural space because of increased points. Table 13.2 provides a summary of drug penetration
fluid formation, as well as obstruction or saturation of the and peak pleural fluid to serum ratios.20,23–34
lymphatic drainage of the pleural space. The route to the
systemic circulation is through the lymphatics and the LACK OF CORRELATION BETWEEN PLEURAL FLUID AND
main route is through the parietal pleura. It would appear ASCITIC FLUID PHARMACOKINETICS
that intrapleural administration of an agent would result
in extremely limited clearance from the pleural space Because there are similarities between the pleural and peri-
because of the limited clearance of pleural fluid. toneal cavity with regard to the mesothelium and lym-
Therefore, drug would be expected to diffuse to a great phatic system, an assumption is made by some clinicians
extent into the visceral and parietal interstitium owing to that antibiotic characteristics of penetration and elimina-
the high concentration present in the pleura. The levels in tion from ascitic fluid may be extrapolated to those of the
the pleura would be expected to consistently exceed those pleural fluid. Although this may be true to a certain extent
of the plasma in patients with normal hepatic and renal when considering the permeation of the mesothelium, dis-
function. criminating conclusions must be made.
Lechi and colleagues23 examined the pharmacokinetics
of cefuroxime, 1 g, administered intravenously to five
Antimicrobial agents patients with ascites and six patients with pleural effusion
(three transudates, three exudates). Simultaneous serum
Since the late 1970s, several studies have examined antimi- and ascitic fluid levels were followed over 12 hours and
crobial concentrations obtained in the pleural fluid. serum and pleural fluid levels were followed at 3-hour
Although several studies have been conducted and many intervals for each patient over 6 hours. In other words,
reference books and drug package inserts list the degree of continuous data did not exist for all patients in the pleural
pleural penetration of various agents; the data is severely fluid group.
lacking and unsound and reference data should be consid- The average maximum pleural fluid concentration
ered with caution. (Cmax, approximately 8 mg/mL) was reached more quickly
The rationale behind this conclusion is that the many in most patients (approximately 1.5–2 hours) compared
studies conducted examined penetration in relatively with that of the ascitic fluid (4–5 hours; approximately
small, heterogeneous patient groups. The study patients 45 mg/mL). The antibiotic penetrated to the ascitic fluid
consisted of those that had transudative effusions together more readily with a mean Cmax serum–ascitic fluid ratio of
with those that had exudative effusions. Many studies 26 percent compared with the mean Cmax serum–pleural
included patients that were not infected or combined fluid ratio of 8–10 percent (n = 5). The pleural fluid con-
those that were infected with those that had malig- centration never exceeded that of the serum concentra-
nancy.20–22 tion, displaying a maximum average pleural fluid–serum
Most studies involved the administration of a single ratio of approximately 50 percent at the end of 6 hours
dose of drug as opposed to multiple doses. This is not an when serum concentrations were at a minimum.
ideal condition under which to observe drug pharmacoki- Conversely, the cefuroxime ascitic fluid concentration
netics since the patients’ plasma drug levels would not be greatly exceeded that of the serum as serum levels declined
considered to be at a ‘steady-state’ distribution equilib- and at the end of 12 hours, the ascitic fluid–serum ratio
rium. Consequently, the pleural fluid levels would not was approximately 1.80. The mean elimination t1/2 from
reflect the distribution equilibrium with the plasma and the pleural fluid was approximately 5.1 hours, two times
may underestimate drug concentrations. These conditions longer than that of the serum (n = 4).
do not represent those observed in a clinical situation with There are many limitations to this study in that contin-
multiple dosing and possible drug accumulation. uous data was not supplied for the pleural patients and
Another limitation to the available literature is that numerical concentration data was not provided for the
many studies did not follow pleural fluid or plasma drug ascitic patients. However, two observations may be made.
concentration–time profiles. In many instances, non- First, it is clear that the pharmacokinetic drug profile in the
simultaneous plasma and pleural fluid levels were taken ascitic fluid contrasted with that of the pleural fluid. The
for each patient. Therefore, it is not known whether the time to Cmax (Tmax) and degree of penetration were quite
drug levels found in the pleural fluid actually reflect the different. Second, it is evident that elimination from the
Table 13.2 Antimicrobial agents in the pleural fluid
Agent Pleural disease Levels Dose Route Cmaxa AUCa t1/2 (h) t1/2 (h) Reference
(n) PF/Serum PF/Serum PF Serum
pleural space lagged behind and was twice as long of that Another study was conducted in 16 patients with either
in the serum. infected or sterile pleural effusions. The patients were
It must be pointed out that these patients received one divided into four separate groups and received either
dose of cefuroxime and were therefore not at steady-state single or multiple doses of ciprofloxacin. The study groups
distribution equilibrium. The degree of penetration of were divided as follows. Group 1: seven patients with
cefuroxime into the pleural fluid and ascitic fluid may not sterile effusions received a single 200 mg dose of intra-
be accurately depicted; however, the contrast in pharma- venous ciprofloxacin. Group 2: two patients with sterile
cokinetic profiles of the agent in the ascitic compared with effusions received a single 750 mg oral dose of
the pleural fluid seems accurate. ciprofloxacin. Group 3: three patients with sterile effusions
received six doses of 750 mg oral ciprofloxacin over 3 days.
SELECTED CLINICAL STUDIES Group 4: three patients with empyema received 20 doses of
oral ciprofloxacin over 10 days. Simultaneous serum and
The penetration of a variety of antibiotics, including peni- pleural fluid samples were taken from all patients and were
cillins, cephalosporins, clindamycin and the aminoglyco- obtained after three doses in those that received multiple
sides was evaluated in 16 patients with pleural empyema, doses.
uncomplicated parapneumonic effusion or carcinomatous This study makes interesting comparisons in that it
effusions, respectively.24 The mean pleural fluid to serum examines single versus multiple dosing, which may illus-
concentration ratios is displayed in Table 13.2. The pene- trate any potential increase in intrapleural drug levels once
tration for all antibiotics in the case of multiple doses in an a steady state is achieved. It also compares the penetration
infected effusion was very high with a pleural fluid to of uninfected with infected tissue to show any evidence of
serum ratio of 0.80 in all cases. Although a small number increased permeability in the face of infection. The results
of patients were studied, the penetration of penicillin, from this study are illustrated in Figure 13.9 and Table
cephalothin and oxacillin appeared to be higher with mul- 13.2.
tiple versus single dosing. Penetration of penicillin in the The ciprofloxacin peak pleural fluid concentration to
patients with carcinomatous effusions was found to be serum ratio was the highest and was reached most quickly
lower than that of infected patients. The limitation of the in patients who were infected and received multiple
study is that it did not follow simultaneous pleural fluid doses. Drug levels in both the serum and pleural fluid
and plasma drug concentrations over time. However, it were higher with multiple versus single dosing. The Tmax
illustrates that the penetration of antibiotics into infected of the pleural fluid lagged behind that of the serum by
pleural effusions was excellent and that penetration into 3–4 hours in Groups 1–3, which consisted of uninfected
the carcinomatous effusions was less favorable. Further, it patients. In the infected group (Group 4), the Tmax in the
hints that greater penetration or accumulation may occur pleural fluid was shorter than that of the plasma (1 hour
with multiple dosing once steady-state distribution equi- versus 2 hours, respectively), leading to the conclusion
librium is reached. that drug penetration may be faster and greater in
Group 1: Concentration-time profile following a single Group 2: Concentration-time profile following a single
ciprofloxacin 200 mg IV dose (n7) ciprofloxacin 750 mg PO dose (n3)
3 4
Drug concentration
Drug concentration
3
2
(mg/L)
(mg/L)
2
1
1
0 0
0 1 2 3 4 5 0 1 2 3 4 5 6
Time (h) Time (h)
Average plasma Average pleural fluid Average plasma Average pleural fluid
concentration concentration concentration concentration
Drug concentration
4
(mg/L)
(mg/L)
2
2
0 0
0 1 2 3 4 5 0 1 2 3
Time (h) Time (h)
Average plasma Average pleural fluid Average plasma Average pleural fluid
concentration concentration concentration concentration
Figure 13.9 Penetration of ciprofloxacin into sterile (Groups 1–3) and empyemic (Group 4) human pleural fluid.33
Pharmacokinetics specific to the pleural fluid 163
infected patients. Finally, in all patient groups, elimina- the serum at 0.5 and 3 hours and in the pleural fluid at
tion of drug from the pleural fluid was much slower than 1 and 4 hours post antibiotic dose. Although the authors
that of the plasma. state that many patients were at steady-state distribution
Vancomycin was found to readily penetrate to the equilibrium upon sampling with receipt of greater than six
pleural fluid of 16 uninfected patients who underwent tho- doses, the number of doses received appears to have
racotomy for the treatment of bronchial cancer. The ranged from one to greater than six and is not reported for
patients received vancomycin as either a 15 mg/kg inter- individual patients.
mittent infusion twice daily (n = 8) or a 500 mg loading The average concentrations of ampicillin were as
dose followed by a 30 mg/kg continuous infusion presum- follows [mean (range)]; serum at 0.5 hours 44.6 mg/mL
ably over 24 hours (n = 8).25 The vancomycin AUC0–12 (17.2–81); pleural fluid at 1 hour 28.4 mg/mL (14.6–41.3);
pleural fluid–serum ratio exceeded 0.80 in all patients, serum at 3 hours (n = 7), 14.8 mg/mL (2–33.4); pleural
with an average ratio value of 0.88 ± 0.07 and 0.86 ± 0.14 fluid at 4 hours 16.2 mg/mL (25–25.4). The average con-
in the intermittent and continuous infusion groups, centrations of penicillin were as follows [mean (range)];
respectively. The average vancomycin Cmax was observed at serum at 0.5 hours 23.8 mg/mL (10.3–34.8); pleural fluid at
1 hour (48.3 ± 14.9 mg/mL) in the serum and at 2 hours 1 hour 10.8 mg/mL (9.1–13.1); serum at 3 hours 3.2 mg/mL
(19 ± 4.8 mg/mL) in the pleural fluid, following receipt of (1–3.3); pleural fluid at 4 hours 7.7 mg/mL (3.6–11.8).
intermittent infusion. This indicates a 1-hour lag time for Although an average pleural fluid to serum ratio may not
vancomycin penetration from the blood to pleural fluid in be determined because of the disparity of sampling times,
these patients. The pleural fluid concentrations ranged the data indicate substantial penetration of both b-lactam
from 11.8 ± 2.7 mg/mL at 0 hours to 13.7 ± 3.5 mg/mL at 12 agents into the pleural fluid of these infected pediatric
hours, following the start of dose administration in the patients.
continuous infusion patient group. An average van-
comycin concentration of approximately 14 and 12 mg/mL INTRAPLEURAL ADMINISTRATION OF ANTIBIOTIC
was maintained in the serum and pleural fluid over 12
hours, respectively, in those patients who received a bolus The use of aminoglycosides in parapneumonic effusion or
followed by continuous infusion. The average vancomycin empyema is not recommended as first line therapy. This is
concentration at 12 hours in the intermittent infusion because the aminoglycoside antibacterial activity may
group was 6.6 ± 2.3 mg/mL. It should be noted that blood become compromised in an acidic environment with low
samples were not obtained beyond the 12-hour timepoint oxygen content, as may exist in an infectious locus.
or at steady-state distribution equilibrium. Thus, it would Thys and colleagues27 conducted a study comparing
be of interest to know the true lag time for pleural fluid to intrapleural with systemic administration of amikacin,
reach its Cmax in relation to serum had more samples been which clarifies several interesting points. A total of 20 post-
obtained once a steady state was achieved. thoracotomy patients were given a single dose of amikacin
While the average vancomycin AUC0–12 was similar either intravenously or intrapleurally. The pleural fluid of
between groups, higher concentrations were achieved these patients was not infected, but contained an increased
more quickly in patients who received the loading dose mean protein level of 3.4 ± 0.2 g/dL. Pleural fluid and
followed by continuous infusion. As may be expected, serum drug concentrations were simultaneously collected
higher concentrations were longer sustained in the pleural and compared.
fluid with administration via continuous infusion. Despite The serum Cmax at 0.5 hours post-dose was 31.2 ±
its large molecular size and poor lipid solubility, van- 2.3 mg/mL for systemic administration and 14.1 ±
comycin readily reached the pleural fluid in these patients 24.7 mg/mL for intrapleural administration. It is evident
with administration by both intermittent and continuous that systemic distribution did occur without delay in
infusion. A possible explanation of the large degree of intrapleural administration, but was half that compared
pleural fluid penetration may be lung inflammation that with systemic administration. The peak pleural fluid to
may be associated with malignancy. serum concentration ratio was 43 percent and occurred
Pharmacokinetic studies in the pediatric population are within 0.5–2 hours of systemic administration. A signifi-
infrequent, and even more so when exploring pleural fluid cant difference was found in the elimination rate constant,
penetration. Giachetto and colleagues26 reported the k (serum, 0.28 ± 0.02/hour; pleural fluid, 0.19 ± 0.02/hour;
pleural fluid and serum concentrations of hospitalized p < 0.025) and the t1/2 (serum, 2.62 ± 0.22 hours; pleural
children with community-acquired pneumonia and fluid, 4.3 ± 0.8 hours; p < 0.0010) of systemically adminis-
empyema who received either ampicillin (ages 10 months tered drug. Amikacin was cleared from the pleural fluid at
to 5 years) or penicillin (ages 5–12 years). Blood and a rate half that for serum.
pleural fluid samples were collected from the patients after The mean peak pleural fluid concentration following
receipt of ampicillin (n = 9) 400 mg/kg/day as an intra- intrapleural administration was extremely large at 3374 ±
venous bolus in six divided doses or penicillin (n = 4) 983 mg/mL. The elimination from the pleural fluid in these
200 000 IU/kg.day intravenous infusion over 20 minutes in patients was assisted by suction, so values are misleading;
six divided doses. Drug concentrations were determined in however, the t1/2 was approximately 2.88 hours in serum
164 Pleural pharmacokinetics
(comparable to the other group) and 1.58 ± 0.23 hours in model (Table 13.3) provides us with pertinent information
pleural fluid. that is difficult to gather in human research trials.35–40
It has become apparent that the clearance of drug from Although there are undoubtedly differences between the
the pleural fluid lags behind that of the plasma. This may rabbit and human pleural structures and physiology, these
be beneficial for agents that display ‘time-dependent’ studies provide an insight into the relationship between
versus ‘concentration-dependent’ activity, but toxicity pharmacokinetics of the serum and pleural fluid. These
may become an issue with agents that have a small thera- data also provide an insight into the degree of penetration
peutic window. Since most studies examine the adminis- of the mesothelium by antimicrobial agents in the face of
tration of one dose, the implications of drug accumulation infection.
are not known. Intrapleural administration resulted in an The assumption has been that an empyema is difficult
exceedingly high peak concentration of amikacin. This to penetrate because of fibrin encasement and viscous pus
suggests that a lower dose may be used in intrapleural content. These data indicate that drug penetration into
administration since toxicity is a concern. Systemic distri- empyema is actually quite good. Penetration in animals
bution occurred from the pleural space almost immedi- was compared not only by average Cmax pleural fluid to
ately, and the elimination rate was unchanged compared serum ratios, but also by AUC or total drug exposure. The
with the group that received systemic administration. AUC ratios are higher since the elimination from the
Since these patients had suction-assisted elimination, the pleural space lags behind that of the serum, prolonging
effect of the high pleural concentration compared with drug exposure. The higher the pleural fluid to serum AUC
serum concentration is unclear over time. Perhaps the ratio, the more equal the drug exposure between the serum
serum concentrations would have continued to increase and pleural space. Depending on the pharmacodynamic
over time as a result of distribution to the systemic circu- characteristics of the drug, this may or may not be more
lation from the pleural compartment. important than peak drug concentrations in relation to
killing of a pathogen.
ANIMAL DATA
CONCLUSIONS REGARDING ANTIBIOTICS
The main reason why data is lacking regarding the pene-
tration of antimicrobial agents in the setting of infected In general, it seems that most antibiotics do attain thera-
effusion is the difficulty in performing studies with multi- peutic concentrations in the pleural space (Table 13.2) to
ple doses of medication in acutely ill patients over an varying degrees; however, these concentrations lag behind
extended period of time. The pleural empyema rabbit those of the serum. Delayed peak concentration is unlikely
to be an issue with multiple dosing. The quinolones, peni- The reversible movement of drug from the pleural cavity
cillins and aminoglycosides seem to penetrate the pleural to the systemic circulation may be quantified by the distri-
fluid especially well, although the aminoglycosides cannot bution rate constants between compartments, k21 and k12
be recommended. Antibiotic elimination from the pleural (Figure 13.4). This movement would depend on the con-
space lags behind that of the serum. This may be beneficial centration gradient established between the pleural cavity
to treat infection but accumulation may occur with multi- and the rest of the body as well as the physiochemical
ple dosing and caution is advised when administering properties of the administered agent (molecular weight,
toxic agents. The lag in clearance of the pleural fluid indi- protein binding and isoelectric properties). Characteristics
cates that AUC pleural fluid to serum ratios will exceed specific to the individual patient would also certainly influ-
peak pleural fluid to serum ratios for many agents. ence the pharmacokinetics to a large degree. It is difficult,
Therefore, while the pleural fluid peak is lower, total drug therefore, to make generalizations among chemotherapy
exposure may be closer in range. Although unnecessary in patients. The pharmacokinetics may potentially be influ-
the majority of cases, intrapleural administration of antibi- enced by the local effects of malignant tumor, changes in
otics offers the advantage of achieving higher peak pleural physiology of the pleura due to surgery/treatment and the
fluid levels and may be considered in special cases. Further, underlying renal and hepatic function of the patient.
the effects of the extraordinary peaks and the possibility Elimination of drugs from the pleural fluid and the total
and degree of drug accumulation are not known. body will also be influenced by biomedical interventions,
such as chest tube drainage of the pleural cavity that is
often employed with anti-cancer intrapleural chemother-
Antineoplastic agents apy. Inflammation, as in the case of infection, would be
expected to increase drug permeability of the mesothe-
lium.
Since the pharmacological intent of anti-cancer agents is to
Elimination from the pleural cavity to the systemic cir-
destroy cells, and these agents effectively do so without dis-
culation, potentially supported mainly from the lymphat-
crimination between healthy and malignant cells, antineo-
ics of the parietal pleura, may be blocked by malignancy
plastic agents are undoubtedly the most toxic drugs
and might also become saturated. As a result, drug move-
administered to humans. The greatest challenge in the
ment out of the pleural cavity may be a very slow process
drug therapy of malignancy lies in the creation of a balance
depending on the physiochemical drug properties and the
between efficacy and safety. Dosage administration design
patient. Since total body elimination of drug takes place
presents an opportunity to artfully manipulate the balance
strictly from the central compartment for most drugs,
between efficacy and safety through not only dosage
clearance from the body would be dependent upon trans-
adjustment, but also by route of administration. Intra-
fer from the pleura to the systemic circulation. Although
pleural chemotherapy for treatment of pleural malignan-
intrapleural administration would diminish systemic drug
cies, especially mesothelioma, has attracted considerable
exposure, concerns still exist over local and systemic
attention. The final focus of this chapter is the pharmaco-
toxicity.
kinetics associated with the intrapleural administration of
Several well-conducted studies have examined the
anti-cancer chemotherapy agents.
pharmacological effect as well as the pharmacokinetics of
intrapleurally administered agents in the treatment of
INTRAPLEURAL ADMINISTRATION OF ANTI-CANCER malignancy. Although not all inclusive, examination of the
CHEMOTHERAPY AGENTS results of these pharmacokinetic studies (Table 13.4) leads
to several important conclusions.41–49
The rationale for the intrapleural administration of toxic
anti-cancer chemotherapy is to maximize drug exposure MAXIMIZATION OF INTRAPLEURAL DRUG EXPOSURE
and the subsequent killing of malignant cells while mini-
mizing systemic exposure of healthy cells to the toxic The area under the drug concentration–time curve reveals
agents. The main concern with intrapleural administration total drug exposure. Table 13.4 shows that the direct
is the movement of drug into the systemic circulation. intrapleural administration of antineoplastic agents maxi-
The theoretical movement of intrapleurally adminis- mizes drug exposure of malignant cells located in the
tered drug from the pleural space to the systemic circula- pleura or pleural effusion, resulting in very high AUCs and
tion is proposed to follow the basic pharmacokinetic peak (Cmax) concentrations in the pleural fluid. These con-
principles of a two-compartment model except that drug centrations are undoubtedly higher than those achievable
administration would take place directly to the peripheral with systemic drug administration. Further, while the
compartment (Figure 13.4). These are the same principles pleural fluid AUC and Cmax parameters are large, the
set out in the previous section to describe the pharmacoki- plasma or serum levels are lower, indicating a much
netics of an intrapleurally administered antibiotic. greater drug exposure in the pleura compared with the
High initial pleural drug concentrations in the pleural plasma. However, with such high concentrations, concern
fluid would drive equilibration with the rest of the body. exists about local toxicity in the pleura.
166 Pleural pharmacokinetics
Agent N Dose Route Concentrationa Cmax Serum AUCa AUCa Cl (L/h)b Cl (L/h)b Reference
(mg/m2) PF/Serum (mg/L) PF/Serum Type PF Serum
CDDP
total 10 80 IP 59.6 1.8 10.9 24 h 116.9c NA 47
free 10 80 IP 8.38 0.4 104 24 h 10.5c NA 47
CDDP
total 3 90 IP NA 1 47.5 4h 5.68 122 44
free 3 90 IP NA 2.3 NA NA NA NA 44
CDDP 7 90 IV NA 2.3 NA NA NA 132 44
CDDP
total 11 100 IP NA NA 28.6 4h 28 ± 14.3 NA 45
free 11 100 IP NA NA 47.6 4h NA NA 45
CDDPtotal 14 153–203 IP 129.25 1.5 NA NA NA 8.7 ± 6.5 48
CDDP
total 10 200 IP NA 2.1 27 ± 4.6 48 h NA NA 49
free 10 200 IP NA 0.8 23 ± 9.2 48 h NA NA 49
CDDP
total 7 200 IP NA 2.0 25 ± 6.2 48 h NA NA 49
free 7 200 IP NA 0.8 37 ± 7.7 48 h NA NA 49
MTC 11 8 IP NA NA 195.7 • 2.3 ± 1.0 5.5 ± 4 45
SYSTEMIC DRUG DISTRIBUTION AND ELIMINATION the average Cmax plasma concentration of those that
received intrapleural cisplatin (2.25 ± 0.58 and 0.96 ±
As mentioned previously, in order for the drug to be elim- 0.35 mg/L, respectively; p < 0.01). The average maximum
inated from the body, it must first be transferred from the level of free platinum concentration resulting from intra-
pleural cavity to the systemic circulation through which it venous administration (1.45 ± 0.55 mg/L) was more than
may reach the organs of metabolism and elimination, three times greater than the free peak level achieved with
mainly the liver and kidneys. Bogliolo and colleagues44 intrapleural administration (0.42 ± 0.18 mg/L).
administered cisplatin 90 mg/m2 intrapleurally to four The administration of 80 mg/m2 in 10 patients with
patients and intravenously to seven patients. The drug malignant pleural effusions in another study resulted in an
elimination t1/2 from the pleural fluid, representing either average Cmax pleural fluid concentration of total platinum
transfer to the systemic circulation or tissue binding, was of 104.4 mg/L compared with 1.75 mg/L in the plasma
2.08 ± 0.65 hours for intrapleurally administered drug. which was reached 8 hours following intrapleural admin-
The mean platinum AUC of the plasma after 4 hours was istration. The average total drug AUC of the pleural fluid
0.27 ± 0.03 mg/minute.mL compared with 12.83 ± was 1032.99 mg/hour.mL compared with 94.34
4.06 mg/minute.mL in the pleural fluid. The average total mg/hour.mL in the plasma. The free platinum Cmax was 88
platinum Cmax drug concentration of those patients that mg/L in the pleural fluid and 0.36 mg/L in the plasma. The
received intravenous cisplatin was two times greater than average Cmax of free platinum in the plasma occurred
References 167
12. Noppen M. Normal volume and cellular contents of pleural fluid. 32. Umut S, Demir T, Akkan G, et al. Penetration of ciprofloxacin into
Curr Opin Pulm Med 2001; 7: 180–2. pleural fluid. J Chemother 1993; 5: 110–2.
13. Light RW. Anatomy of the pleura; physiology of the pleural space. ●33. Joseph J, Vaughan LM, Basran GS. Penetration of intravenous and
In: Pleural diseases, 3rd edn. Baltimore: Williams & Wilkins, 1995; oral ciprofloxacin into sterile and empyemic human pleural fluid.
1–17. Ann Pharmacother 1994; 28: 313–5.
14. Lee KF, Olak J. Anatomy and physiology of the pleural space. Chest ●34. Kutty K, Neicheril JC. Treatment of pleural blastomycosis:
Surg Clin N Am 1994; 4: 391–403. penetration of amphotericin B into the pleural fluid. J Infect Dis
15. Wang NS. Anatomy and physiology of the pleural space. Clin Chest 1987; 156: 689–90.
Med 1985; 6: 3–16. 35. Shohet I, Yellin A, Meyerovitch J, Rubinstein E. Pharmacokinetics
◆16. Miserocchi G. Physiology and pathophysiology of pleural fluid and therapeutic efficacy of gentamicin in an experimental pleural
turnover. Eur Respir J 1997; 10: 219–25. empyema rabbit model. Antimicrob Agents Chemother 1987; 31:
17. Negrini D, Venturoli D, Townsley MI, Reed RK. Permeability of 982–5.
parietal pleura to liquid and proteins. J Appl Physiol 1994; 76: 36. Teixeira LR, Sasse SA, Villarino MA, et al. Antibiotic levels in
627–33. empyemic pleural fluid. Chest 2000; 117: 1734–9.
18. Hamm H, Light RW. Parapneumonic effusion and empyema. Eur 37. Strahilevitz J, Lev A, Levi I, et al. Experimental pneumococcal
Respir J 1997; 10: 1150–6. pleural empyema model: the effect of moxifloxacin. J Antimicrob
19. Sahn SA. Pleural diseases related to metastatic malignancies. Eur Chemother 2003; 51: 665–9.
Respir J 1997; 10: 1907–13. 38. Liapakis IE, Kottakis I, Tzatzarakis MN, et al. Penetration of newer
20. Lode H, Dzwillo G. Investigations of the diffusion of antibiotics quinolones in the empyema fluid. Eur Respir J 2004; 24: 466–70.
into the pleural space. Current Chemotherapy. Proceedings of the 39. Liapakis IE, Light RW, Pitiakoudis MS, et al. Penetration of
10th International Congress on Chemotherapy. Washington DC: clarithromycin in experimental pleural empyema model fluid.
American Society for Microbiology, 1978: 386–8. Respiration 2005; 72: 296–300.
21. Cole DR, Pung J. Penetration of cefazolin into pleural fluid. 40. Niwa T, Nakamura A, Kato T, et al. Pharmacokinetic study of
Antimicrob Agents Chemother 1977; 11: 1003–5. pleural fluid penetration of carbapenem antibiotic agents in
22. Daschner FD, Gier E, Lentzen H, et al. Penetration into the pleural chemical pleurisy. Respir Med 2006; 100: 324–31.
fluid after bacampicillin and amoxycillin. J Antimicrob Chemother ●41. Howell SB, Chu BB, Wung WE, et al. Long-duration intracavitary
1981; 7: 585–8. infusion of methotrexate with systemic leucovorin protection in
23. Lechi A, Arosio E, Xerri L, et al. The kinetics of cefuroxime in ascitic patients with malignant effusions. J Clin Invest 1981; 67:
and pleural fluid. Int J Clin Pharmacol Ther Toxicol 1982; 20: 1161–70.
493–6. 42. Jones JM, Olman EA, Egorin MJ, Aisner J. A case report and
●24. Taryle DA, Good JT Jr, Morgan EJ 3rd, et al. Antibiotic description of the pharmacokinetic behavior of intrapleurally
concentrations in human parapneumonic effusions. J Antimicrob instilled etoposide. Cancer Chemother Pharmacol 1985; 14:
Chemother 1981; 7: 171–7. 172–4.
●25. Byl B, Jacobs F, Wallemacq P, et al. Vancomycin penetration of 43. Holoye PY, Jeffries DG, Dhingra HM, et al. Intrapleural etoposide
uninfected pleural fluid exudates after continuous or intermittent for malignant effusion. Cancer Chemother Pharmacol 1990; 26:
infusion. Antimicrob Agents Chemother 2003; 47: 2015–7. 147–50.
26. Giachetto G, Pirez MC, Nanni L, et al. Ampicillin and penicillin ●44. Bogliolo GV, Lerza R, Bottino GB, et al. Regional pharmacokinetic
concentration in serum and pleural fluid of hospitalized children selectivity of intrapleural cisplatin. Eur J Cancer 1991; 27:
with community-acquired pneumonia. Pediatr Infect Dis J 2004; 839–42.
23: 625–9. 45. Rusch VW, Niedzwiecki D, Tao Y, et al. Intrapleural cisplatin and
27. Thys JP, Serruys-Schoutens E, Rocmans P, et al. Amikacin mitomycin for malignant mesothelioma following pleurectomy:
concentrations in uninfected postthoracotomy pleural fluid and in pharmacokinetic studies. J Clin Oncol 1992; 10: 1001–6.
serum after intravenous and intrapleural injection. Chest 1984; 46. Lerza R, Vannozzi MO, Tolino G, et al. Carboplatin and cisplatin
85: 502–5. pharmacokinetics after intrapleural combination treatment in
28. Walstad RA, Hellum KB, Blika S, et al. Pharmacokinetics and tissue patients with malignant pleural effusion. Ann Oncol 1997; 8:
penetration of ceftazidime: studies on lymph, aqueous humour, 385–91.
skin blister, cerebrospinal and pleural fluid. J Antimicrob ●47. Tohda Y, Iwanaga T, Takada M, et al. Intrapleural administration of
Chemother 1983; 12 (Suppl A): 275–82. cisplatin and etoposide to treat malignant pleural effusions in
29. Yamada H, Iwanaga T, Nakanishi H, et al. Penetration and patients with non-small cell lung cancer. Chemotherapy 1999; 45:
clearance of cefoperazone and moxalactam in pleural fluid. 197–204.
Antimicrob Agents Chemother 1985; 27: 93–5. 48. Monjanel-Mouterde S, Frenay C, Catalin J, et al. Pharmacokinetics
30. Yew WW, Lee J, Chan CY, et al. Ofloxacin penetration into of intrapleural cisplatin for the treatment of malignant pleural
tuberculous pleural effusion. Antimicrob Agents Chemother 1991; effusions. Oncol Rep 2000; 7:171–5.
35: 2159–60. 49. Shigemura N, Akashi A, Nakagiri T, et al. Pleural perfusion thermo-
●31. Morgenroth A, Pfeuffer HP, Seelmann R, Schweisfurth H. Pleural chemotherapy under VATS: a new less invasive modality for
penetration of ciprofloxacin in patients with empyema thoracis. advance lung cancer with pleural spread. Ann Thorac Surg 2004;
Chest 1991; 100: 406–9. 77:1016–22.
14
Experimental models: pleural diseases other than
mesothelioma
ANIMAL MODELS FOR PLEURAL DISEASES While most animal models cannot exactly emulate human
disease, a good animal model has sufficient similarity to
Why we need animal models provide useful insights.
A large number of animal models have been employed
The pleura is involved in many pulmonary and systemic in the investigation of pleural diseases. The design of
disorders. As a result, pleural effusions are common clini- meaningful in vivo experiments demands a good under-
cal presentations. In vivo studies have played an invalu- standing of the advantages and limitations of the available
able part in enhancing our understanding of the etiology models. This chapter outlines the species and methods
of various pleural diseases. While in vitro studies can used in pleural disease studies, as well as the pros and cons
provide information on isolated cell types, pleural of specific models for different pleural diseases.
pathologies are inevitably a result of complicated interac-
tions between residential mesothelial cells and infiltrating GENERAL RULES
(e.g. inflammatory, malignant) cells. The pleura is also
under close influence of products from the systemic circu- There is no substitute for careful planning: the objectives
lation (e.g. cytokines) that cross the vascular and and experimental endpoints should be clearly defined in
mesothelial barriers. These interactions can only be ade- advance. Investigators must attempt to minimize the
quately studied in vivo. number of animals sacrificed and the pain or distress to
Animal studies are also important in evaluating the effi- each animal.
cacy and safety of novel therapeutic modalities, informa- Approval from local animal care committees must be
tion difficult to obtain from humans and in vitro obtained before experiments are performed. National and
experiments. For example, animal studies have assessed international guidelines on standards of animal use must
the efficacy and adverse effects of various pleurodesing and be strictly adhered to. The Guide for the Care and Use of
anti-cancer agents. Animal studies can also provide useful Laboratory Animals is one of the most commonly used ref-
data on the pharmacokinetics of drug delivery into the erence guides.2 It is mandatory in many countries that
pleural space, such as antibiotic penetration in empyema, researchers must attend training courses prior to using
and on physiological responses in health and disease. animals for experiments. The laws and regulations govern-
The ideal animal model should accurately represent the ing laboratory animal research can be found elsewhere.3
human disease under investigation; be readily available, Expert veterinarian advice is invaluable to ensure minimal
affordable, and easy to handle; yield reproducible results animal discomfort, optimal surgical approach and efficient
and provide adequate biological samples for analysis.1 specimen collection.
170 Experimental models: pleural diseases other than mesothelioma
Which animal model to use Inbred animals share a great degree of genetic identity,
reducing experimental variability. Moreover, differences
CHOICE OF ANIMAL SPECIES between inbred mouse strains can be exploited to discover
mechanisms of disease.
‘… it is proper to choose certain animals which offer The use of novel molecular biology technologies in the
favorable anatomical arrangements or special suscepti- mouse has greatly enhanced research into the pathogene-
bility to certain influences. … the proper choice of sis of pleural diseases. The pleura has been used for over-
animals is so important that the solution of a physio- expression of gene products using various vectors.9,10
logical or pathological problem often depends solely on Several constitutive and conditional gene knockout and
the appropriate choice of the animal for the experiment knockin mice have been used in pleural disease investiga-
so as to make the result clear and searching.’ tions,11–13 and the development of systems that facilitate
conditional gene overexpression or silencing in the adult
Claude Bernard (1813–1878) – for many the founder of mouse, such as the tetracycline on–off models, is expected
experimental medicine – so stated in 1865.4 to greatly enhance research in this area.14,15 Recently,
Many animal species are used in pleural disease investi- methods to silence gene expression in vivo using RNA
gations. Rabbits, mice, rats and sheep are the most interference were introduced, which are anticipated to
common. Several factors govern the choice of species for further boost pleural disease investigations.16,17
experimentation, including size of animals, their anatomy,
costs, availability of reagents suitable for the species, possi- ACCESS TO PLEURAL SPACE
bility of genetic manipulation, etc.
Smaller animals (e.g. mice) cost less and are easier to The next question is how to deliver the experimental agent
handle. Small size, however, makes intrapleural injections to the pleural space. Direct intrapleural injection with a
difficult and provides a limited amount of biological mate- fine needle is the least invasive; a small injection volume is
rial for examination. Larger animals are costly, but provide adequate. Using Tc-99 labeled fluid, it has been shown that
larger sample quantities. an injection volume as low as 0.5 mL is enough to allow the
Investigators must realize two important anatomical injectate to be distributed throughout the whole pleural
characteristics of the species they use. First, many animals surface of rabbits. Rotation of the animals is unnecessary
(e.g. mice, dogs), have incomplete mediastina and the two for distribution of the injectate throughout the pleural
pleural cavities communicate freely, prohibiting the use of space.18
the contralateral pleura as a control.5 Second, larger Alternatively, small plastic ‘chest tubes’ can be inserted
animals (e.g. sheep) have a thick visceral pleura resembling into the pleural space (see Appendix 14.1). This has the
that of humans, whereas smaller animals (e.g. rabbits, advantage of allowing repeat intrapleural administration
mice) have a thin visceral membrane. This difference bears of reagents, pleural fluid sampling or lavage of the pleural
implications on fluid and particulate transport across the cavity. This provides longitudinal data on the biological
pleura. changes within the pleural space, helping reduce the
The choice of species may also be dictated by the avail- number of animals required in time-course studies. Chest
ability of reagents to process the experimental samples. tubes can induce mild inflammation,19 which in our expe-
Commercial enzyme-linked immunosorbent assays rience is insignificant. In studies of pleurodesis, the inser-
(ELISA) are commonly used for measurement of tion of chest tubes more closely resembles the procedure
cytokines, but are usually available only against humans performed in clinical practice.
and mice. Thoracotomy20 and thoracoscopy21 have also been used
The choice of species for experimentation is often influ- to deliver material into the pleural space, but are more
enced by the knowledge of the genomic sequence of the invasive. Systemic delivery of substances can be achieved
species and by the availability of research tools to manipu- by intravenous injections. For repeated blood sampling,
late gene expression in animals. The mouse (Mus muscu- central venous access can be established.
lus) has emerged as the most popular choice and
genetically engineered mice are increasingly used in EXPERIMENTAL END-POINTS
pleural disease investigations, providing valuable insights
into the molecular pathogenesis of pleural diseases. Several parameters are commonly used as endpoints in
Murine models have several advantages:6–8 (i) The mouse animal pleural studies. In lethal models, survival is the
genome shares sufficient homology with that of humans; definitive endpoint. Pleural tissues can be collected at
(ii) a wide array of genetically engineered mice are avail- autopsy for macroscopic and/or microscopic examination
able; (iii) the large litter size of Mus musculus makes breed- and semi-quantitative assessment for inflammatory or
ing timelier and easier; (iv) high-throughput genotyping malignant changes. For pleural fibrosis studies, the pleura
methods (e.g. using genomic DNA from tail fragments) can be macroscopically graded for adhesions (see Table
have been developed for the mouse; (v) many inbred 14.1), which correlates well with histological measure-
strains of Mus musculus have been isolated over the years. ments of collagen deposition and pleural thickening.22 The
Animal models for pleural diseases 171
Table 14.1 Pleurodesis grading scheme22–24 The rabbit model is most commonly used in pleurode-
sis studies,22,26 though mice,27 sheep28 (see Appendix 14.2),
Pleurodesis grading scale of 1 to 8:
rats,29 dogs21 and pigs30 have been employed. The results of
common pleurodesing agents applied to different species
1 No adhesions between the visceral and parietal pleura appear similar; hence the choice of species depends mainly
2 Rare adhesions between the visceral and parietal pleura on experience and cost.
with no symphysis With New Zealand white rabbits, the pleurodesing agent
3 A few scattered adhesions between the visceral and is injected either directly or via a chest tube, the method of
parietal pleura with no symphysis delivery having little effect on the outcome.31 Likewise, the
4 Many adhesions between the visceral and parietal pleura pH of the pleurodesing agent does not affect its effective-
with no symphysis ness.32,33 Conventional agents, e.g. talc, doxycycline and
5 Many adhesions between the visceral and parietal pleura bleomycin, induce acute pleural inflammation and
with symphysis involving <5% of the hemithorax denudement of mesothelial cells.23,34,35 Inflammation may
6 Many adhesions between the visceral and parietal pleura resolve (failed pleurodesis) or, if sufficiently intense, will
with symphysis involving 5–25% of the hemithorax persist and lead to collagen deposition, fibrosis and symph-
7 Many adhesions between the visceral and parietal pleura ysis.36 Pleurodesis is usually evident 14 to 28 days after
with symphysis involving 25–50% of the hemithorax administration of the pleurodesing agent. However, with
8 Many adhesions between the visceral and parietal pleura pro-fibrotic transforming growth factor beta (TGF-b), sig-
with symphysis involving >50% of the hemithorax nificant adhesions can be seen as soon as after 24 hours.22
Adhesions are defined as fibrous connections between the visceral and
Pleural fibrosis can be graded macroscopically at
parietal pleura. autopsy (Table 14.1). In addition, pleural thickening and
Symphysis is present if the visceral and parietal pleura are difficult to collagen deposition can be measured microscopically
separate as a result of adhesions. using multiple samples from different lung regions to
avoid sampling bias, and the contralateral pleural space as
the control.22,37 In our experience, chest tube insertion and
volume of pleural fluid produced, as well as its cellular and saline or albumin injections do not result in significant
protein contents, may serve as surrogate markers for adhesions. Hemothoraces, however, either from trauma or
inflammation or tumor progression. Pleural vascular per- the experimental agent, can induce adhesions. Recently, a
meability can be determined by measurement of the pleurodesis grading system based on transthoracic ultra-
pleural fluid levels of an albumin-binding substance after sound findings, specifically the disappearance of the
intravenous injection (e.g. Evans’ blue, H3-labeled normal pleural gliding sign, was developed in rabbits and
albumin). Blood samples can be obtained to compare with was validated against pleurodesis grading at autopsy23,24
the intrapleural levels of mediators, to evaluate systemic (Table 14.2). Whether this system is applicable to other
responses and to study the pharmacokinetics of drugs species, including humans, remains to be tested.
delivered via the pleural space. One concern of the published pleurodesis studies is that
Advanced imaging techniques (e.g. magnetic resonance they were performed in animals with normal pleura,
imaging [MRI]) are increasingly applied to animal whereas human pleurodesis is usually applied to patients
research to allow longitudinal monitoring of pathophysio- with abnormal pleurae (especially malignant pleural
logic processes. Recently, transthoracic ultrasonography metastases). However, pleurodesis results from animal
has been shown to be useful in detecting and grading pleu- models are similar to those from clinical investigations.
rodesis in rabbits,23,24 without sacrificing the animals. It is For example, talc was effective in producing pleurodesis,
anticipated that more imaging modalities will be adopted
for pleural research in coming years.
Table 14.2 Ultrasound pleurodesis grading scheme23,24
Specific pleural disease models Grade Gliding sign
and was significantly more so than bleomycin in both of human adenocarcinoma (e.g. A549) cells produces
rabbit38 and sheep pleurodesis studies39 – the same as in numerous lung lesions, pleural metastases and effusions,
randomized clinical trials.40 in contrast with human squamous or large cell carcinoma
There are no available models for pleurodesis in malig- cell lines, where malignant effusions are uncommon.47,51
nant pleural effusions. Most models for malignant effu- Transfection of antisense VEGF gene reduced VEGF
sions are created in mice (see below), which are too small expression, tumor vascularity, pleural metastases and effu-
for adequate assessment of pleurodesis. On the other hand, sions induced by adenocarcinoma cells.47
attempts have been made to mimic the setting of pleuro- Tumors can also be injected directly into the pleural
desis for pneumothorax, where chemical pleurodesis space, which is best performed by a left lateral sub-
remained effective despite the presence of air in the pleural diaphragmatic approach aimed cephalically. Extra-fine
space or active air leaks.38,41 needles (28G) should be used. Up to 1.0 mL of fluid can be
injected into an adult mouse, but small volumes (e.g.
MODELS FOR MALIGNANT PLEURAL EFFUSION 50–100 mL) usually suffice. Accurate intrathoracic delivery
is confirmed by transient chest expansion and dyspnea.
Malignant pleural effusions, most commonly resulting Alternatively, a small skin incision (~5 mm) can be made,
from adenocarcinomas of the lung and breast, affect about preferably on the left hemithorax, and tumor cells can be
200 000 patients each year in the USA alone, causing sig- injected into the pleural space under direct vision. This
nificant morbidity and mortality.42–45 The pathogenesis of method requires minimal surgery and has the advantages
malignant effusions and their best management are still of on-site confirmation of orthotopic tumor cell delivery
unclear. Animal studies can shed light on the mechanisms to the pleura and is reliably reproducible.50
of effusion formation, as in recent studies on vascular Intrapleural injections of tumor cells (usually 105–106
endothelial growth factor (VEGF),46–48 and on the proin- per mice) result in local implantations in the chest wall,
flammatory axes of interleukin (IL)-6/Stat349 and of tumor mediastinum, lungs and diaphragm by 1–2 weeks (Figure
necrosis factor (TNF)-a/nuclear factor (NF)-kB.50 14.1a). Pleural effusions (usually bloody exudates) develop
Most animal studies on malignant pleural effusions approximately 2–3 weeks after inoculation.50,51 At later
have been performed using mice. Athymic nude mice are stages, ascites may accumulate. When immunocompetent
commonly used as they allow the development of pleural mice are used with syngeneic tumor cells (e.g. C57BL/6
metastases by xenogenic (e.g. human) tumor cells. Various mice – Lewis lung adenocarcinoma cells), the effusions are
human cancer cell lines have been successfully introduced rich in inflammatory cells (Figure 14.1b).50 Most mice
into the pleural space of immunodeficient mice, which eventually develop respiratory distress and weight loss, and
give rise to malignant effusions.47,48 Among the different die from local effects of the effusion, cachexia, and distant
tumor cell types used, adenocarcinomas produce the metastases.
highest rate of effusions, similar to clinical presentations Weight loss and survival are commonly used endpoints.
where adenocarcinomas tend to produce pleural metas- The volume of pleural effusion, tumor load (number, size
tases.51 The shortcoming of these models is the need for and weight), and the presence of distant metastases are
immunodeficient (e.g. severe combined immunodeficient other parameters useful for assessing therapeutic response.
[SCID] or recombination activation gene-RAG-2 null) Pleural fluid and tissue collected allow the study of patho-
mice, which have impaired immunogenic responses to logical mediators. Effusion-associated vascular permeabil-
malignancies. ity can be easily determined by intravenous injection of an
Stathopoulos et al.50 recently described a murine model albumin-binding dye (see ‘Models for the study of pleural
of adenocarcinoma-induced malignant effusion in vascular permeability’ and Figure 14.1c).50
immunocompetent mice, which will allow better elucida- Using these models, the role of important mediators
tion of the mechanism of malignant effusion formation and biological pathways (e.g. VEGF/VEGFR, IL-6/Stat3
and its treatment. and TNF-a/NF-kB) in effusion formation have been
Malignant cells can be implanted into the pleural space uncovered.47–50 Other studies have assessed novel anti-
directly (by surgery or intrapleural injection) or indirectly cancer therapies, such as IL-12 and IL-15,53 and inhibitors
from metastases from tumors deposited in the lungs. of topoisomerase II51 and of VEGF receptor tyrosine
Orthotopic implantation of freshly isolated human adeno- kinase.48 The intrapleural injection model has also been
carcinoma tissues into nude mice can produce a high successfully applied to transgenic mice, such as nitric oxide
takeup rate of the cancer52 but requires thoracotomy and synthase knockouts.11
tying of tumor tissues into the visceral and parietal
pleurae. This method is more invasive and does not MODELS FOR PLEURAL INFLAMMATION
provide additional information compared with intra-
pleural or intravenous injection of tumor cells. In the latter The pleural cavity is regarded by some as ‘the ideal site for
cases, commercially available cancer cell lines grown in cell the induction of inflammatory reactions’.54 In the clinical
culture conditions are titrated before injection to stan- setting, inflammation is often assessed histologically,
dardize the tumor load per mouse. Intravenous injection which is subject to sampling error and can only provide
Animal models for pleural diseases 173
(a) (c)
Figure 14.1 (a) Histological section through a Lewis lung
adenocarcinoma (LLC)-induced pleural tumor in a C57BL/6 mouse,
stained with hematoxylin and eosin (¥ 40). Note that the tumor
bridges the parietal and visceral pleurae. cw, chest wall; l, lung; r,
rib; pt, pleural tumor; pc, pleural cavity. (b) Cytocentrifugal
preparation of malignant pleural effusion cells from the LLC-
C57BL/6 model, stained with May-Grünwald-Giemsa (¥ 400).
Alien adenocarcinoma cells are mixed with host inflammatory
cells. cc, cancer cell; m, mononuclear cell; l, lymphocyte; n,
neutrophil polymorphonuclear cell. (c) Vascular hyperpermeability
induced in the skin of a C57BL/6 mouse by malignant pleural
fluids from the LLC-C57BL/6 model, compared with PBS (negative
control). After intradermal injection of 50 mL pleural fluid or PBS,
(b) the mouse received 0.8 mg Evans’ blue in 200 mL normal saline.
The mouse was killed and the skin inverted and photographed
after 30 minutes. Circles represent areas of dye extravasation, and
numbers relative surface area in respect to PBS (control). Evans’
blue avidly binds to albumin; hence its extravasation indicates
vascular hyperpermeability. (See also Color Plates 8–10.)
qualitative rather than quantitative data. Also, if inflam- inflammatory cell type recruited to the pleural cavity, the
matory mediators are to be investigated, they have to be various animal models have been coined models of neu-
extracted from histology tissues. In vitro studies of inflam- trophilic, mononuclear or eosinophilic pleural inflamma-
matory cells fail to provide knowledge on the complex tion. For example, intrapleural lipopolysaccharide (LPS)
interactions between various cells and mediators. results in mainly neutrophilic, but also, to a lesser extent,
Therefore, investigators often employ animal models of mononuclear, and eosinophilic pleural inflammation.56–58
pleural inflammation that allow the study of cells and The most widely used pleural inflammation model is
fluids accumulated during inflammation.55 the carrageenan pleurisy model.55 Carrageenan can be
Pleural models of inflammation offer several advan- administered intrapleurally by a needle injection, causing
tages. The pleural cavity provides a confined compartment dose-dependent inflammation.55 The detailed onset, pro-
lined by mesothelial cells in close contact with the systemic gression and resolution of carrageenan pleurisy have been
circulation where inflammatory cells and mediators extensively studied and validated.59 In brief, pleural exuda-
collect. These can be monitored in a dynamic fashion by tion begins within an hour of injection, and is character-
assaying the pleural fluid. The histological changes of ized by neutrophil followed by monocytic influx, followed
inflammation can be assessed in pleural tissues and effects by vascular hyperpermeability and pleural exudation.55
of pro- or anti-inflammatory agents can be investigated Pleural fluids and serum can be serially collected for analy-
following their intrapleural administration. sis, and pleural tissues can be obtained at necropsy.
Various methods have been used to induce pleural Numerous studies have utilized this model successfully in
inflammation (Table 14.3), and each yields an inflamma- mice, rats and rabbits.60 For example, mice that do not
tory reaction characterized by a distinct profile of cell and express IL-612 or nitric oxide synthase61 exhibit signifi-
mediator accumulation. Depending on the predominant cantly reduced pleural inflammation, confirming the
174 Experimental models: pleural diseases other than mesothelioma
Table 14.3 A large variety of agents have been used to induce pleural inflammation. This
table outlines the common and some of the uncommon agents used. Interested readers can
refer to the references for individual agents for details
Most common
Carrageenan Bliven et al.55
Vinegar et al.59
Common
Endotoxin or lipopolysaccharide Broaddus et al.65
Fukumoto et al.66
Reverse Passive Arthur Reaction Yamamoto et al.54
Berkenkopf et al.167
Zymosan Utsunomiya et al.168
Uncommon
Azoles Hanada et al.169
Bradykinin Saleh et al.170
Calcium pyrophosphate crystals Perianin et al.171
Ionophore A23187 Wang et al.172
Kaolin Kawamura et al.173
Phorbol myristate acetate Oh-ishi et al.174
Platelet activating factor Tarayre et al.175
Substance P Frode-Saleh et al.176
essential roles of these compounds in the inflammatory responses. The pleural responses have been extrapolated to
process. explain the pathophysiology of other allergic diseases, such
Using the carrageenan model in conditional as asthma and atopic dermatitis.
macrophage-deficient mice, the important role of resident In these models, mice or rats are sensitized with ovalbu-
pleural macrophages in neutrophil recruitment to the min adsorbed to Al(OH)3 gel injected subcutaneously 14
inflamed pleura was elucidated.62 This probably occurs via days prior to the experiment. This sensitization process
production of 15-deoxy-prostaglandin (PG)J2 that elicits peripheral and pleural eosinophilia. The experiment
induces transcription factor Nrf2, to facilitate recruitment is then performed with an intrapleural injection of ovalbu-
of neutrophils to the pleura, to switch off neutrophilic and min (10 mg per pleural cavity), which induces mast cell
switch on monocytic inflammation.63 degranulation, immunoglobulin E (IgE) accumulation
Several authors have injected LPS into the pleural cavity and eosinophil chemotaxis.67,68 The role of mediators in
of rats and mice to generate inflammation. LPS induces eosinophilic recruitment and the efficacy of anti-allergy
influx of neutrophils within 4 hours, followed by mono- therapies can then be evaluated.67–70 Pleural lavage can be
cytes, lymphocytes and eosinophils.56–58 These studies have performed to quantify eosinophil influx and allergic medi-
yielded interesting results on the differential role of the ators, such as leukotrienes and platelet-activating factor.
various selectins (L-, P- and E-) and of IL-8 in inflamma- Using ovalbumin allergic pleurisy in nuclear factor of acti-
tory neutrophil and eosinophil recruitment to the vated T cells (NFAT)1-gene-defficient mice, the role of
pleura.64–66 NFAT1 in the suppression of T-helper type 2 immune-
The reverse passive Arthus reaction has also been responses was elucidated.71 In addition to ovalbumin,
studied, particularly in rats.54 Intravenous bovine serum direct intrapleural delivery of biological mediators has
albumin (BSA), followed by an intrapleural injection of been used to provoke eosinophilic inflammation.
purified anti-BSA 20 minutes later, induces pleural Intrapleural injection of chemokine ligand (CCL) 22,
inflammation characterized by fluid extravasation (peak at macrophage inflammatory protein (MIP)-1a, RANTES
6 hours), neutrophil (peak at 6 hours) and mononuclear (regulated upon activation, normally T-cell expressed
(peak at 12–24 hours) influx. and secreted) and eotaxin resulted in dose- and time-
dependent recruitment of eosinophils.72,73
MODELS FOR EOSINOPHILIC (ALLERGIC) PLEURITIS Another mouse model of eosinophilic pleural inflamma-
tion exploits the pleural eosinophilia observed in patients
While allergic pleuritis is uncommon in humans, the with pneumothoraces. In this model, large numbers of
pleural cavities of rats and mice have been used as surro- eosinophils, among other inflammatory cells, are retrieved
gate systems for the study of the mechanisms of allergic from the pleural lavage of mice after transthoracic injection
Animal models for pleural diseases 175
of air.74 Using IL-5- and IL-13-gene-deficient mice, inves- mouse strain, C57BL/6.13 Although intrapleural inocula-
tigators revealed that pleural recruitment of eosinophils is tion of the microorganism did not result in frank
dependent on IL-5 but not IL-13. empyema, inflammatory cell influx and local cytokine and
As described above, intrapleural injection of LPS in rats chemokine production was observed. This model is inter-
and mice results in recruitment of eosinophils, along with esting for two reasons. First, it can be applied to genetically
other cell types.56–58,64,65 Eosinophil levels rise significantly engineered animals. In the aforementioned study, CD4
after 24 hours and their recruitment appears to be medi- knockout mice showed reduced inflammatory response
ated by T lymphocytes.58 LPS-induced pleural eosinophilia and retarded pathogen clearance compared with wild-type
is mediated by P-selectin and IL-8 and is inhibited by cor- mice.13 Second, S. aureus is also pathogenic to humans,
ticosteroids.56,64,65 The clinical value of this model remains making these data more readily applicable to humans.
unclear. Other methods to introduce empyema in guinea pigs and
sheep have been published, but did not gain popularity.81–83
MODELS FOR PLEURAL INFECTION BY COMMON PATHOGENS There is one important limitation of all these models. In
(EMPYEMA) humans, empyema occurs usually as a complication of
pneumonia, while isolated pleural infection is uncommon.
Thoracic empyema remains a common disease with signif- The animal models used in the literature all involve direct
icant morbidity and mortality.75,76 Animal experiments on introduction of microbes into the pleural cavity and devel-
empyema are most commonly performed using New opment of isolated pleural infection without pneumonia.
Zealand white rabbits, though mice and sheep have been Hence, the results of these experiments may not be directly
used. extrapolated to humans.
Developing an adequate empyema model is difficult. Animal models of empyema have facilitated the study
Direct introduction of bacteria (such as Streptococcus or of the pathogenesis of the disease. Studies on rabbits and
Peptostreptococcus species) into the pleural space usually mice have revealed the important role of CD4+ lympho-
results in their complete clearance.77 To successfully initi- cytes in empyema-associated inflammation and bacterial
ate pleural infection, prior injury to the pleura may be clearance13 and of TGF-b1 in empyema-associated pleural
required. However, under such situations, overwhelming fibrosis.84 Other studies have provided valuable informa-
sepsis and death can occur if too large a pathogen load is tion on the optimal treatment of empyema. Animal studies
administered. have lent support to repeated early thoracentesis,85 early
Sahn et al.78 used an intrapleural injection of turpentine chest tube insertion86 and intrapleural (single or com-
in rabbits, resulting in inflammation and an exudative bined) fibrinolytics for empyema.79,87 Finally, the penetra-
neutrophilic pleural effusion by 72 hours.79 At that point, tion of antibiotics into the pleural space has been studied
bacteria (e.g. Streptococcus pneumonia,78 Escherichia coli, using rabbit models of empyema.88–90
Peptostreptococcus anaerobius, Bacillus fragilis79 or their
combinations) were introduced by thoracentesis into the MODELS FOR PLEURAL INFECTION BY MYCOBACTERIA
effusion to create an empyema. Turpentine, however, may (TUBERCULOUS PLEURITIS)
impose artifacts, and the authors reported that a high per-
centage of animals did not develop an exudative effusion Tuberculosis (TB) pleuritis continues to be a common
to allow bacterial inoculation.79 clinical challenge in the new millennium. Our understand-
Alternatively, Sasse et al.77 showed that intrapleural ing of its pathogenesis and best management strategies
injection of a potent rabbit pathogen, Pasteurella multo- remain limited. Animal models were used to study TB
cida, in brain–heart infusion agar could produce empyema pleuritis as early as 1917 when Patterson investigated the
without prior administration of turpentine. In this model, disease using guinea pigs.91 Rabbits and mice have also
rabbits required daily intramuscular penicillin injections, been used subsequently.
to prevent death from sepsis. One drawback of this model In the guinea pig model modified by Widstrom et
is that P. multocida is a rabbit pathogen but rarely infects al.92,93 outbred guinea pigs are first vaccinated with intra-
humans.80 Hence, this model is not suitable for certain dermal 0.1–0.4 mg Bacille Calmette–Guerin (BCG). A
investigations, such as studies of antimicrobial treatment higher dose of either BCG or heat killed Mycobacterium
of empyema. tuberculosis is injected intrapleurally 10–15 weeks later
These models are well validated and exhibit pleural using a blunt needle connected to a manometer93 or a 20G
changes similar to that of human empyemas. In the needle directed subdiaphragmatically into the pleural
models of Sahn et al.78 and Sasse et al.77 the induced pleural space.94 With prior vaccination, most guinea pigs remain
fluids showed significant increases in leukocytes and clinically well after intrapleural mycobacteria injection,
inflammatory indices, as well as markedly reduced pH and despite the development of TB pleuritis.92 Vaccination is
glucose. Pleural adhesions and macroscopic suppurative successful in >90 percent of cases and can be confirmed by
changes were evident at necropsy. purified protein derivative (PPD) testing 3 weeks later.
Another interesting model of pleural infection by Shorter time gaps between vaccination and intrapleural
Staphylococcus aureus was developed using a common infection produced inconsistent results and higher
176 Experimental models: pleural diseases other than mesothelioma
incidence of hemothoraces.92 Antony et al.95,96 applied pleural space. Using guinea pigs, Allen et al.99,100 have
similar strategies on New Zealand white rabbits to induce charted the time-course of pleural fluid accumulation,
a TB pleuritis, rabbits being relatively resistant to M. tuber- leukocyte influx and mediator expression after intrapleural
culosis, but susceptible to BCG. They also induced neu- injection of heat-inactivated M. tuberculosis, and reported
tropenia in the rabbits by pretreatment with continual increase of TGF-b1, even during the resolution
nitrogen-mustard, a modification that can allow the study phase of the pleuritis.
of TB pleuritis in immunocompromised hosts.96
Several important points concerning these models MODELS FOR BENIGN ASBESTOS-INDUCED PLEURAL
deserve mention. It is believed that the dose of organisms, DISEASES
but not their virulence, is important for short-term TB
pleuritis models. In the guinea pig model, both heat-inac- Asbestos is a recognized cause for various pleural diseases,
tivated M. tuberculosis94 and BCG92 induce TB pleuritis. such as circumscribed (plaque) and diffuse pleural thick-
Unilateral injections of mycobacteria in guinea pigs often ening, benign asbestos pleural effusion (BAPE), rounded
result in bilateral pleural reactions.92 Whether this is due to atelectasis and malignant mesothelioma.101–103 Mesothelial
incomplete mediastinal separation of the pleural cavities cells appear sensitive to the toxic effects of asbestos fibers,
or represents bacterial dissemination is unknown. either by direct injury or via indirect effects from other
Nonetheless, the contralateral pleural cavity should not be asbestos-exposed cells104 (see Chapter 10, Pleural reaction
used as a control. to mineral dusts). Extensive research has been conducted
Tuberculosis pleuritis induced using these animal in the pathogenesis and treatment for malignant meso-
models closely represents the disease in humans. Exudative thelioma (See Chapter 15, Experimental models:
pleural effusions develop following intrapleural BCG Mesothelioma), but relatively little work has been invested
injections, characterized by early neutrophilic, intermedi- in the study of benign asbestos-induced pleural diseases,
ate monocytic and late lymphocytic influx.92,93,95 Similar to despite their much higher prevalence. In addition to the
humans, lymphocytes in guinea pig TB pleuritis are mainly pleura, asbestos damages the lung parenchyma causing
CD2+ T-lymphocytes.94 However, while TB pleural effu- fibrosis (asbestosis) and increasing lung cancer risk.
sions in humans are sometimes characterized by low pH Experimental models for the study of these conditions are
and glucose, this is not the case in animal models, presum- outside the scope of this chapter.
ably due to lower levels of infection and inflammation.
Histologically, the pleura of infected animals shows caseat- Route of delivery
ing granulomata, multi-nucleated giant cells and late To study the effect of asbestos on the animal pleura, fibers
pleural fibrosis.92,94,95 Regional lymph nodes can also be can be introduced either via the respiratory tract or by
affected. If live mycobacteria are used, they can be recov- direct intrapleural injection.104,105 The former resembles
ered from cultures of the pleural fluid or lymph nodes.92 human exposure to asbestos and fibers can be delivered by
Smaller animals such as mice (either Balb/c or C57BL/6 intratracheal instillation,106–109, by direct delivery to a lobar
strains) have been used to investigate the early inflamma- bronchus110 or by inhalation of aerosolized fibers in a
tion accompanying TB pleuritis.97,98 Without prior vacci- closed chamber.111–114 The resulting pleural and pul-
nation, intrathoracic injection of BCG can effectively monary inflammatory and fibroproliferative reactions
induce pleuritis while inactivated Mycobacterium leprae have a dose–response relationship with the amount of
(isolated from livers of armadillo) cannot.97 While the fibers delivered.111,112
pleural fluid cellular composition in mouse TB pleurisy Both the inhalation and the intrapleural injection
appeared similar to that in humans, the histological methods have drawbacks. Administered by inhalation,
changes with this model have not been described. fiber deposition in the peripheral lung, and hence their
Investigators should be aware of certain limitations of toxic effects to the pleura, varies. In addition, pleural
existing models. In these models, isolated TB pleuritis is changes in animals may take months to years to develop,
induced by transthoracic mycobacterial inoculation, while similar to humans.115,116 By contrast, direct intrapleural
human TB pleuritis usually develops as a result of pleural injection can ensure immediate delivery of a known
spread from adjacent lung parenchymal infection. Also, amount of fibers into the pleural cavity and accurate iden-
only BCG or heat-killed M. tuberculosis, rather than live M. tification of time zero of pleural injury. It also facilitates
tuberculosis (the most common pathogen in humans), are the study of isolated fiber effects on the pleura without the
used in these models. Thus, it is difficult to determine if influence by other lung parenchymal changes. However,
results of therapeutic interventions using these models can this method of delivery differs significantly from how
be directly applied to humans.95 fibers reach the pleura in humans. Wagner et al.117 com-
Despite their limitations, the guinea pig and rabbit pared intrapleural and inhalation delivery of chrysotile to
models of tuberculous pleuritis have provided valuable rats. Mesotheliomas developed more frequently with
insights into the pathogenesis of the disease. Using rabbits, intrapleural injections, whereas malignant lung tumors
Antony et al.96 have shown the importance of neutrophils were much more common than mesotheliomas if fibers
in recruiting mononuclear cells into the TB-infected were delivered by inhalation.
Animal models for pleural diseases 177
fluid was reinjected into another rabbit, a polymorpho- leukocytosis, rising levels of protein, very high amylase and
nuclear neutrophil (PMN) response was also elicited.132 low glucose, and positive bacterial cultures, consistent with
Asbestos-related lung and pleural diseases are likely to the classical pleural fluid findings in patients with ruptured
continue to increase. While legislation has been imple- esophagus.136 Interestingly, when the animals were ren-
mented in most developed countries to minimize occupa- dered neutropenic by pretreatment with nitrogen mustard,
tional and environmental asbestos exposure since the early no reduction in pleural fluid pH was observed, suggesting
1970s, the long lag time between exposure and clinical that the reduced pH results from neutrophil metabolism
presentation means there is still a large population at risk rather than anaerobic bacterial infection.137
of developing disease. Also, chrysotile now constitutes 99
percent of current global asbestos production and sales MODELS FOR PLEURAL EFFUSIONS FROM FLUID
remain strong in developing nations despite the recent OVERLOADS
conclusion of the International Program on Chemical
Safety of the World Health Organization that ‘exposure to Human transudative pleural effusions, commonly caused
chrysotile poses increased risks for asbestosis, lung cancer by congestive cardiac failure, renal failure or hepatic cir-
and mesothelioma in a dose-dependent manner’.101 rhosis, are not easily recapitulated in animal models. A
Hence, it is anticipated that animal studies will remain sheep model of intravenous oleic acid infusion has been
important in the ongoing effort to understand the patho- used to induce pulmonary edema and bilateral pleural
genesis of asbestos pleural damage and to design new treat- effusions.138 Oleic acid resulted in reduced cardiac con-
ment strategies. tractility, raised pulmonary arterial pressures, alveolar
edema and pleural transudation. The biochemical compo-
MODELS FOR CHYLOTHORAX sition of the pleural fluid was similar to that of the alveolar
fluid. While this model is useful, it is not entirely similar to
Chylothorax, the accumulation of chyle in the pleural pleural effusion from cardiac failure. First, oleic acid infu-
space, results from impaired lymphatic drainage due to sion causes alveolar damage. Second, the pleural fluid to
various causes, such as surgery or trauma to the thoracic plasma protein ratio was 0.6–0.7, classifying the pleural
duct and malignancy (see Chapter 29, Effusions from lym- fluid as an exudate.
phatic disruptions). Surgical ligation or interruption of the A murine model of renal failure and fluid overload has
thoracic duct has been performed in a canine model. been described by Song et al.139 Renal failure is generated
Mongrel dogs were fed milk fat prior to surgery to increase by bilateral renal vessel ligation and fluid overload by
lymphatic drainage and allow easy identification and tran- intraperitoneal delivery of isotonic saline of 40 percent
section of the thoracic duct, leaving chest tubes in place for body weight. After 3 hours, the mice develop bilateral
drainage. Using this model, octreotide has been shown to pleural effusions (approximately 100 mL).139 This model
enhance closure of the fistula and reduce chyle leak.133 was developed to study the role of aquaporin water chan-
Congenital defects of the lymphatic duct and lym- nels in pleural fluid homeostasis, but can be used to study
phangiectasia are uncommon causes of chylothoraces. other aspects of pleural effusion secondary to renal failure.
Recently, mice homozygous for a null mutation of the
gene encoding the a9 subunit of the a9b1 integrin were MODELS FOR PLEURAL PHYSIOLOGY STUDIES
bred to examine the roles of the a9 integrins.
Unexpectedly, these mice had genetic defects in their lym- Numerous animal models have been employed for the
phatic system resulting in bilateral chylothoraces by 6–12 investigation of pleural physiology in health and disease.
days after birth, and died eventually of respiratory The main areas of interest are mechanisms of pleural fluid
failure.134 The role of the a9b1 integrin in lymphatic devel- formation and absorption and pleural pressure changes in
opment appears to be mediated via binding to lymphan- the presence of effusion or pneumothorax.
giogenic VEGF-C and -D.135 This model may be useful for To study the transfer of lung water into the pleural
further investigation of the mechanism of chylothorax space through the visceral pleura, Broaddus et al.140 sub-
development. jected anesthetized ventilated sheep to volume overload.
The chest was open and a bag was wrapped around the
MODELS FOR PLEURAL EFFUSIONS FROM ESOPHAGEAL exposed lung to collect the fluid leak from lung
RUPTURE parenchyma into the pleural space. Their experiments
confirmed that the pleural space provided an important
Esophageal perforation is an uncommon cause of pleural route of clearance of pulmonary edema. Similar methods
effusion. Only one model has been developed using inser- have been used in mongrel dogs to study the factors that
tion and overinflation of a 16F Foley catheter into the alter the permeability of the visceral pleura.141,142
esophagus of anesthetized New Zealand white rabbits.136,137 Different animal models have been used to study the
Pleural effusions developed 2 hours later and were bilateral removal of pleural fluid and proteins from the pleural
in half of the rabbits. Serial thoracenteses revealed exuda- space. Many studies have been performed in sheep143 and
tive effusions with progressively increasing acidity and rabbits,144 and occasionally dogs.145 Labeled particles, such
In vitro studies of mesothelial cells and in situ studies of mesothelial monolayers 179
as I125-albumin146 or florescent isothiocyanate-labeled orescence or scintillation), and reflects the rate of albumin
dextran,147 can be used as a tracer to follow the efflux and leakage into the pleural space.
reabsorption of protein in the pleural space. Another experimental approach, the Miles vascular
Readers should beware of potential drawbacks when permeability assay, determines the effects of mediators
extrapolating the results of these animal studies to human contained in pleural fluid on vascular permeability in the
physiology. First, fluid exchange mechanisms in animals mouse or rat skin (Figure 14.1c),48,49,155,156 Cell-free pleural
may not always parallel those of humans, as humans (and fluid supernatants are injected intradermally and Evans’
sheep) have thick visceral pleurae in contrast to the thin blue is administered intravenously. After a predetermined
visceral pleural membranes of rabbits: thus, the results of time interval, extravasation of the dye into the mouse
rabbit fluid absorption studies may not represent human dermis is determined by measuring Evans’ blue levels in
conditions. Second, in humans, pleural fluid often accu- tissue extracts or by morphometry. Vascular permeability
mulates in pathologic conditions with abnormal (e.g. induced by the fluid under examination can be compared
inflammed or malignant) pleura, and the pathophysiology with saline (negative control) or VEGF solutions (positive
of fluid formation and regression may well differ from control), and the contribution of individual mediators to
studies performed in animals with normal pleurae. Third, overall vascular permeability can be assessed after their
respiratory patterns may affect pleural fluid absorption, neutralization in the pleural fluid prior to injection into
making sedated animals not ideal representatives of ‘real the skin.
life’ situations. In rabbits, the rate of pleural absorption of Using these methods, the contribution of VEGF,47,48,156
particles can be influenced by their molecular weight.148 IL-649 and TNF-a (Stathopoulos et al., unpublished obser-
Whether this applies to other animals remains to be tested. vations) to the induction of vascular hyperpermeability in
Animal studies have also been used to assess cardiores- malignant pleural effusions has been established.
piratory impairment induced by pleural effusion or pneu-
mothorax. Fluid or air can be introduced intrapleurally OTHER MODELS
and physiological changes (e.g. electrocardiogram, hemo-
dynamic changes, arterial oxygenation, lung function) Various other models have been published over the years.
measured.149–152 Large pleural effusions affect the dynamic For example, the turpentine model (see Models for pleural
elastance and resistance of the respiratory system, and infection) has been used to mimic effusions from rheuma-
produce hypoxemia in a dose-dependent fashion. It is toid arthritis.157
noteworthy that in animal experiments effusions are pro- Intrapleural gene therapy for replacement therapy or
duced acutely, whereas most human effusions (e.g. malig- for mesothelioma has been studied in animal models. In
nant) accumulate over time. Chronic measurement of the our experience, mesothelial cells can be easily transfected
effects of artificially induced effusion is difficult, but has in vitro and in vivo. Plasmids delivered intrapleurally can
been reported by Murphy et al.153 who surgically inserted transfect the mesothelial cells and the protein product can
catheters in the serosal layer of the rat esophagus, allowing be recovered from pleural fluids and from plasma.158
measurement of pleural pressures for up to 14 weeks. Rabbits have also been used to study the pharmacoki-
netics of intrapleurally administered drugs.159,160 This may
be relevant to intrapleural chemotherapy, which has been
MODELS FOR THE STUDY OF PLEURAL VASCULAR
PERMEABILITY
increasingly used in clinical trials for control of malignant
mesothelioma.
Leakage of protein-rich exudate from blood capillaries in
the lung interstitium, and beneath the mesothelium into
the pleural space, has been implicated in the pathogenesis IN VITRO STUDIES OF MESOTHELIAL CELLS
of various pleural effusions and has been demonstrated in AND IN SITU STUDIES OF MESOTHELIAL
human disease and animal models.47–50,154–156 Albumin is MONOLAYERS
the most abundant protein contained in exudative pleural
effusions and is, as a result, most commonly used to deter- While in vitro studies have limitations, the study of cul-
mine vascular permeability. Studies of pleural vascular tured mesothelial cells in isolation can provide informa-
permeability are feasible in any species, but mice have been tion supplemental to animal studies. Mesothelial cell lines
used preferentially. are commercially available but are transformed by viral
Pleural vascular permeability can be assessed using two infection or transformation. Mesothelial cells from the
mutually complementary methods. Permeability can be pleura or the peritoneum can be harvested for primary
determined using an albumin tracer (e.g. Evans’ blue, culture. While their biological behaviors are likely to be
fluoro-isothiocyanate or radioisotope-labeled albumin) similar, this has seldom been compared or confirmed.
introduced intravenously into animals bearing pleural However, when in vitro studies of mesothelial cells are
effusion or inflammation.47–50 Shortly thereafter (e.g. 5–30 undertaken to address the pathophysiology of pleural dis-
minutes), the levels of the tracer in the pleural fluid or eases, it would be ideal to use pleural and not peritoneal
lavage can be determined (e.g. measuring absorbance, flu- mesothelial cells.
180 Experimental models: pleural diseases other than mesothelioma
Normal human pleural mesothelial cells are difficult to No animal model is ideal. Investigators should under-
obtain, and have most commonly been isolated from stand the advantages and limitations of the use of different
pleural effusions caused by heart failure. Culture methods animal species and models. Only through doing so would
of human and animal mesothelial cells are similar.104,161 they be able to choose or design the most suitable in vivo
However, mesothelial cells from human effusions, even model or in vitro experiment that provides the best chance
from transudative ones, have likely been exposed to medi- of answering the scientific question(s) raised. Animal
ators and may not truly represent ‘normal’ mesothelial studies should be planned, conducted and supervised with
cells. For that reason, studies often employ primary culture a similar degree of scrutiny as that applied to clinical trials.
of pleural mesothelial cells from animals, e.g. rabbits or The ultimate aim should always be to provide better care
mice. Principles of harvesting the cells and points of to patients with pleural diseases.
caution are summarized in Appendix 14.3. In our experi- Advances in other areas of biomedical sciences, espe-
ence, primary rabbit mesothelial cells grow rapidly and cially animal imaging techniques, should allow the design
maintain their biological activities up to seven or eight of more sophisticated animal models that will improve our
passages. In contrast, murine mesothelial cells divide very understanding and clinical management of pleural dis-
slowly (the initial growth rate increases with density of eases.
seeded cells), and rarely survive a third passage.
To overcome the limitations of human mesothelial cell
culture outlined above, Kim et al. devised a new in vitro
system for culture of mesothelioma tissues in the form of KEY POINTS
spheroids, based on a technique previously developed for
culture of intact bronchial mucosa.162,163 Using this ● Animal models have been invaluable in the study
method, mesothelioma tissue retained many of its in vivo of the pathogenesis of pleural diseases. In vivo
characteristics. This model closely emulates intrapleural studies are essential in evaluating new therapeu-
mesothelioma growth and is expected to greatly facilitate tic options for pleural diseases, in assessing the
studies on mesothelioma apoptotic resistance to novel pharmacokinetics of drug delivery to the pleura
therapies. Such techniques can be adopted for experiments and in examining physiological changes in the
on non-malignant pleural tissue cultured ex vivo. pleural space in health and disease.
Other studies have been performed on isolated ● It is critical for researchers to understand the
mesothelial barriers in situ, most commonly obtained characteristics and limitations of experimental
from sheep.164–166 Ussing chambers, special devices that models available in order to use the most appro-
function as voltage clamps, have been used to measure the priate method that can best answer the scientific
transpleural resistance in pleura stripped from animals, question asked.
before and after an intervention. The resistance measured ● Researchers must adhere to standard guidelines
supposedly reflects the permeability of the pleural barrier. for animal care and gain approval from local
This model has its limitations: pleural fluid formation is ethics committees. Every effort should be made
governed largely by vascular (rather than mesothelial) per- to minimize animal discomfort and the number
meability, and the findings of these studies of normal of animals required.
pleura may not be applicable to pleural structures in ● In vitro studies allow the study of mesothelial and
disease states. other cells that may engage in the pathogenesis of
pleural diseases in isolation. In vitro experiments
can help explain pathologies observed in animal
CONCLUSIONS or human studies. Conversely, novel information
derived from cell culture experiments can be
tested in vivo using appropriate animal models.
‘If we look carefully enough we will eventually find an ● Animal models exist for common types of pleural
animal model for every disease.’ diseases (e.g. pleural effusion due to malignancy,
infection and inflammation), as well as for
Leader (1969) at the Federation of American Societies for pleural pathologies that are uncommonly
Experimental Biology. encountered in clinical practice (e.g. chylotho-
Animal experimentation represents one of the funda- rax, esophageal rupture).
mental approaches in the long arduous path towards the ● Advancement in biomedical technology, e.g.
understanding of the pathogenesis of various pleural dis- novel methods for gene transfer and further
eases. In vivo studies are essential in the evaluation of novel development of genetically engineered mice, will
therapeutic approaches in pleural diseases, in the study of allow the design of increasingly sophisticated
pharmacokinetics of drug delivery in the pleural space and models to provide further significant insights
in the investigation of pleural physiological changes in into pleural diseases.
both normal and disease states.
References 181
●22.
REFERENCES Lee YCG, Teixeira LR, Devin CJ, et al. Transforming growth factor-
b2 induces pleurodesis significantly faster than talc. Am J Respir
Crit Care Med 2001; 163: 640–4.
● = Key primary paper 23. Dikensoy O, Zhu Z, Donnelly E, et al. Combination therapy with
◆ = Major review article intrapleural doxycycline and talc in reduced doses is effective in
producing pleurodesis in rabbits. Chest 2005; 128: 3735–42.
◆1. Migaki G, Capen CC. Animal models in biomedical research. In: Fox ●24. Zhu Z, Donnelly E, Dikensoy O, et al. Efficacy of ultrasound in the
JG, Cohen BJ, Loew FM (eds). Laboratory animal medicine. New diagnosis of pleurodesis in rabbits. Chest 2005; 128: 934–9.
York: Academic Press Inc., 1984: 667–98. ◆25. Lee YCG, Light RW. Management of malignant pleural effusions.
◆2. Institute for Laboratory Animal Research, Commission on Life Respirology 2004; 9: 148–56.
Sciences, National Research Council. Guide for the care and use of ●26. Light RW, Cheng DS, Lee YCG, et al. A single intrapleural injection
laboratory animals. Washington DC: National Academy Press, of transforming growth factor beta-2 produces an excellent
1996. pleurodesis in rabbits. Am J Respir Crit Care Med 2000; 162:
◆3. McPherson CW. Laws, regulations, and policies affecting the use 98–104.
of laboratory animals. In: Fox JG, Cohen BJ, Loew FM (eds). 27. Marchi E, Vargas FS, Acencio MMP, et al. Pleurodesis: a novel
Laboratory animal medicine. New York: Academic Press Inc., 1984: experimental model. Respirology 2007; 12: 500–4.
19–30. ●28. Lee YCG, Lane KB, Parker RE, et al. Transforming growth factor
4. Cohen BJ, Loew FM. Laboratory animal medicine: Historical beta-2 (TGFb2) produces effective pleurodesis in sheep with no
perspectives. In: Fox JG, Cohen BJ, Loew FM (eds). Laboratory systemic complications. Thorax 2000; 55: 1058–62.
animal medicine. New York: Academic Press Inc., 1984: 1–18. 29. Werebe EC, Pazetti R, Milanez de Campos JR, et al. Systemic
5. Dechman G, Mishima M, Bates JH. Assessment of acute pleural distribution of talc after intrapleural administration in rats. Chest
effusion in dogs by computed tomography. J Appl Physiol 1994; 1999; 115: 190–3.
76: 1993–8. 30. Cohen RG, Shely WW, Thompson SE, et al. Talc pleurodesis: Talc
◆6. Silver LM. Mouse genetics. New York/Oxford: Oxford University slurry versus thoracoscopic talc insufflation in porcine model. Ann
Press, 1995. Thorac Surg 1996; 62: 1000–2.
◆7. Bockamp E, Maringer M, Spangenberg C, et al. Of mice and 31. Wu W, Teixeira LR, Light RW. Doxycycline pleurodesis in rabbits:
models: improved animal models for biomedical research. Physiol comparison of results with and without chest tube. Chest 1998;
Genomics 2002; 11: 115–32. 114: 563–8.
◆8. Wade CM, Daly MJ. Genetic variation in laboratory mice. Nat 32. Sahn SA, Good JT, Jr. The effect of common sclerosing agents on
Genet 2005; 37: 1175–80. the rabbit pleural space. Am Rev Respir Dis 1981; 124: 65–7.
9. van der Most RG, Robinson BWS, Nelson DJ. Gene therapy for 33. Hurewitz AN, Lidonicci K, Wu CL, Reim D, Zucker S. Histologic
malignant mesothelioma: beyond the infant years. Cancer Gene changes of doxycycline pleurodesis in rabbits. Effect of
Ther 2006; 13: 897–904. concentration and pH. Chest 1994; 106: 1241–5.
10. Mae M, Crystal RG. Gene transfer to the pleural mesothelium as a 34. Kennedy L, Harley RA, Sahn SA, Strange C. Talc slurry pleurodesis.
strategy to deliver proteins to the lung parenchyma. Hum Gene Pleural fluid and histologic analysis. Chest 1995; 107: 1707–12.
Ther 2002; 13: 1471–82. 35. Bilaceroglu S, Guo Y, Hawthorne ML, et al. Oral forms of
11. Wang B, Xiong Q, Shi Q, et al. Genetic disruption of host nitric tetracycline and doxycycline are effective in producing
oxide synthase II gene impairs melanoma-induced angiogenesis pleurodesis. Chest 2005; 128: 3750–6.
and suppresses pleural effusion. Int J Cancer 2001; 91: 607–11. ◆36. Mutsaers SE, Kalomenidis I, Wilson NA, Lee YCG. Growth factors in
12. Cuzzocrea S, Sautebin L, De Sarro G, et al. Role of IL-6 in the pleural fibrosis. Curr Opin Pulm Med 2006; 12: 251–8.
pleurisy and lung injury caused by carrageenan. J Immunol 1999; 37. Lee YCG, Devin CJ, Teixiera LR, et al. Transforming growth factor
163: 5094–104. beta-2 induced pleurodesis is not inhibited by corticosteroids.
13. Mohammed KA, Nasreen N, Ward MJ, Antony VB. Induction of Thorax 2001; 56: 643–8.
acute pleural inflammation by Staphylococcus aureus. I. CD4+ T 38. Xie C, Teixeira LR, McGovern JP, Light RW. Effect of pneumothorax
cells play a critical role in experimental empyema. J Infect Dis on pleurodesis induced with talc in rabbits. Chest 1998; 114:
2000; 181: 1693–9. 1143–6.
●14. Gossen M, Bujard H. Tight control of gene expression in 39. Lee YCG, Yasay JR, Johnson JE, et al. Comparing transforming
mammalian cells by tetracycline-responsive promoters. Proc Natl growth factor (TGF)-b2, talc and bleomycin as pleurodesing agents
Acad Sci U S A 1992; 89: 5547–51. in sheep. Respirology 2002; 7: 209–16.
◆15. Lewandoski M. Conditional control of gene expression in the 40. Diacon AH, Wyser C, Bollinger CT, et al. Prospective randomized
mouse. Nature Rev Genet 2001; 2: 743–55. comparison of thoracoscopic talc poudrage under local anaethesia
●16. Elbashir SM, Harborth J, Lendeckel W, et al. Duplexes of 21- versus bleomycin instillation for pleurodesis in malignant pleural
nucleotide RNAs mediate RNA interference in cultured effusions. Am J Respir Crit Care Med 2000; 162: 1445–9.
mammalian cells. Nature 2001; 411: 494–8. 41. Macoviak JA, Stephenson LW, Ochs R, Edmunds LHJ. Tetracycline
◆17. Engels BM, Hutvagner G. Principles and effects of microRNA- pleurodesis during active pulmonary-pleural air leak for prevention
mediated post-transcriptional gene regulation. Oncogene 2006; of recurrent pneumothorax. Chest 1982; 81: 78–81.
25: 6163–9. 42. Antony VB, Loddenkemper R, Astoul P, et al. Management of
18. Vargas FS, Teixeira LR, Coelho IJ, et al. Distribution of pleural malignant pleural effusions. Eur Respir J 2001;18: 402–19.
injectate. Effect of volume of injectate and animal rotation. Chest 43. Antunes G, Neville E, Duffy J, Ali N. Pleural Diseases Group,
1994; 106: 1246–9. Standards of Care Committee, British Thoracic Society. BTS
19. Carvalho P, Kirk W, Butler J, Charan NB. Effects of tube guidelines for the management of malignant pleural effusions.
thoracostomy on pleural fluid characteristics in sheep. J Appl Thorax 2003; 58: ii29–38.
Physiol 1993; 74: 2782–7. 44. Sugiura S, Ando Y, Minami H, et al. Prognostic value of pleural
20. Bresticker MA, Oba J, LoCicero Jr, Greene R. Optimal pleurodesis: a effusion in patients with non-small cell lung cancer. Clin Cancer
comparison study. Ann Thorac Surg 1993; 55: 364–6. Res 1997; 3: 47–50.
21. Colt HG, Russack V, Chiu Y, et al. A comparison of thoracoscopic 45. Naito T, Satoh H, Ishikawa H, et al. Pleural effusion as a significant
talc insufflation, slurry, and mechanical abrasion pleurodesis. prognostic factor in non-small cell lung cancer. Anticancer Res
Chest 1997; 111: 442–8. 1997; 17: 4743–6.
182 Experimental models: pleural diseases other than mesothelioma
◆46. Lee YCG, Malkerneker D, Thompson PJ, Light RW, Lane KB. 65. Broaddus VC, Boylan AM, Hoeffel JM, et al. Neutralization of IL-8
Transforming growth factor-b induces vascular endothelial growth inhibits neutrophil influx in a rabbit model of endotoxin-induced
factor elaboration from pleural mesothelial cells in vivo and in pleurisy. J Immunol 1994; 152: 2960–7.
vitro. Am J Respir Crit Care Med 2002; 165: 88–94. 66. Fukumoto T, Matsukawa A, Yoshimura T, et al. IL-8 is an essential
●47. Yano S, Shinohara H, Herbst RS, et al. Production of experimental mediator of the increased delayed-phase vascular permeability in
malignant pleural effusions is dependent on invasion of the pleura LPS-induced rabbit pleurisy. J Leukoc Biol 1998; 63: 584–90.
and expression of vascular endothelial growth factor/vascular 67. Klein A, Talvani A, Cara DC, et al. Stem cell factor plays a major
permeability factor by human lung cancer cells. Am J Pathol 2000; role in the recruitment of eosinophils in allergic pleurisy in mice
157: 1893–1903. via the production of leukotriene B4. J Immunol 2000; 164:
48. Yano S, Herbst RS, Shinohara H, et al. Treatment for malignant 4271–6.
pleural effusion of human lung adenocarcinoma by inhibition of 68. Pasquale CP, e Silva PM, Lima MC, et al. Suppression by cetirizine
vascular endothelial growth factor receptor tyrosine kinase of pleurisy triggered by antigen in actively sensitized rats. Eur J
phosphorylation. Clin Cancer Res 2000; 6: 957–65. Pharmacol 1992; 223: 9–14.
49. Yeh HH, Lai WW, Chen HH, Liu HS, Su WC. Autocrine IL-6-induced 69. Martins MA, Castro-Faria-Neto HC, Bozza PT, et al. Role of PAF in
Stat3 activation contributes to the pathogenesis of lung the allergic pleurisy caused by ovalbumin in actively sensitized
adenocarcinoma and malignant pleural effusion. Oncogene 2006; rats. J Leukoc Biol 1993; 53: 104–11.
25: 4300–9. 70. Martins MA, Pasquale CP, e Silva PM, Cordeiro RS, Vargaftig BB.
●50. Stathopoulos GT, Zhu Z, Everhart MB, et al. Nuclear factor-kappaB Eosinophil accumulation in the rat pleural cavity after mast cell
affects tumor progression in a mouse model of malignant pleural stimulation with compound 48/80 involves protein synthesis and
effusion. Am J Respir Cell Mol Biol 2006; 34: 142–50. is selectively suppressed by dexamethasone. Int Arch Allergy Appl
51. Kraus-Berthier L, Jan M, Guilbaud N, et al. Histology and Immunol 1990; 92: 416–24.
sensitivity to anticancer drugs of two human non-small cell lung ●71. Viola JPB, Kiani A, Bozza PT, Rao A. Regulation of allergic
carcinomas implanted in the pleural cavity of nude mice. Clin inflammation and eosinophil recruitment in mice lacking the
Cancer Res 2000; 6: 297–304. transcription factor NFAT1: Role of interleukin-4 (IL-4) and IL-5.
52. Astoul P, Colt HG, Wang X, Hoffman RM. Metastatic human pleura Blood 1998; 91: 2223–30.
ovarian cancer model constructed by orthotopic implantation of 72. Pinho V, Oliveira SH, Souza DG, et al. The role of CCL22 (MDC) for
fresh histologically intact patient carcinoma in nude mice. the recruitment of eosinophils during allergic pleurisy in mice. J
Anticancer Res 1993; 13: 1999–2002. Leukoc Biol 2003;73: 356–62.
53. Kimura K, Nishimura H, Matsuzaki T, et al. Synergistic effect of 73. Klein A, Talvani A, Silva PMR, et al. Stem cell factor-induced
interleukin-15 and interleukin-12 on antitumor activity in a leukotriene B4 production cooperates with eotaxin to mediate the
murine malignant pleurisy model. Cancer Immunol Immunother recruitment of eosinophils during allergic pleurisy in mice1. J
2000; 49: 71–7. Immunol 2001; 167: 524–31.
54. Yamamoto S, Dunn CJ, Deporter DA, et al. A model for the 74. Kalomenidis I, Guo Y, Peebles RS, et al. Pneumothorax-associated
quantitative study of Arthus (immunologic) hypersensitivity in pleural eosinophilia in mice is interleukin-5 but not interleukin-13
rats. Agents Actions 1975; 5: 374–7. dependent. Chest 2005; 128: 2978–83.
55. Bliven ML, Otterness IG. Carrageenan pleurisy. Meth Enzymol 75. Light RW, Girard WM, Jenkinson SG. The incidence and
1988; 162: 334–9. significance of parapneumonic effusions. Am J Med 1980; 69:
56. Bozza PT, Castro-Faria-Neto HC, Martins MA, et al. 507–12.
Pharmacological modulation of lipopolysaccharide-induced pleural 76. Strange C, Sahn SA. Management of paraneumonic pleural
eosinophil in the rat: a role for a newly generated protein. Eur J effusions and empyema. Infect Dis Clin North Am 1991; 5: 539–59.
Pharmacol 1993; 248: 41–7. ●77. Sasse SA, Causing LA, Mulligan ME, Light RW. Serial pleural fluid
57. Bozza PT, Castro-Faria-Neto HC, Penido C, et al. Requirement for analysis in a new experimental model of empyema. Chest 1996;
lymphocytes and resident macrophages in LPS-induced pleural 109: 1043–48.
eosinophil accumulation. J Leukoc Biol 1994; 56: 151–8. 78. Sahn SA, Reller LB, Taryle DA, Antony VB, Good JT Jr. The
58. Penido C, Castro-Faria-Neto HC, Larangeira AP, et al. The role of contribution of leukocytes and bacteria to the low pH of empyema
gammadelta T lymphocytes in lipopolysaccharide-induced fluid. Am Rev Respir Dis 1983; 128: 811–15.
eosinophil accumulation into the mouse pleural cavity. J Immunol 79. Strange C, Allen ML, Harley R, Lazarchick J, Sahn SA. Intrapleural
1997; 159: 853–60. streptokinase in experimental empyema. Am Rev Respir Dis 1993;
◆59. Vinegar R, Truax JF, Selph JL, Voelker FA. Pathway of onset, 147: 962–6.
development and decay of carrageenan pleurisy in the rat. Fed 80. Boyce JM. Pasteurella multocida. In: Mandell GL (ed.). Principles
Proc 1982; 41: 2588–95. and practice of infectious diseases. New York: Churchill
60. Strange C, Tomlinson JR, Wilson C, et al. The histology of Livingstone, 1990: 1746–8.
experimental pleural injury with tetracycline, empyema and 81. Mavroudis C, Gamzel BL, Katzmark S, et al. Effect of hemothorax
carrageenan. Exp Mol Pathol 1989; 51: 205–19. on experimental empyema in the guinea pig. J Thorac Cardiovasc
●61. Cuzzocrea S, Mazzon E, Calabro G, et al. Inducible nitric oxide Surg 1985; 89: 42–49.
synthase-knockout mice exhibit resistance to pleurisy and lung 82. Mavroudis C, Gamzel BL, Cox SK, et al. Experimental aerobic-
injury caused by carrageenan. Am J Respir Crit Care Med 2000; anaerobic thoracic empyema in the guinea pig. Ann Thorac Surg
162: 1859–66. 1987; 43: 298–302.
●62. Cailhier JF, Sawatzky DA, Kipari T, et al. Resident pleural 83. Pfeffer A, Rogers KM, O’Keeffe L, Osborn PJ. Acute phase protein
macrophages are key orchestrators of neutrophil recruitment in response, food intake, live weight change and lesions following
pleural inflammation. Am J Respir Crit Care Med 2006; 173: intrathoracic injection of yeast in sheep. Res Vet Sci 1993; 55:
540–47. 360–6.
●63. Itoh K, Mochizuki M, Ishii Y, et al. Transcription factor Nrf2 84. Sasse SA, Jadus MR, Kukes GD. Pleural fluid transforming growth
regulates inflammation by mediating the effect of 15-deoxy- factor-beta1 correlates with pleural fibrosis in experimental
D12,14-prostaglandin J2. Mol Cell Biol 2004; 24: 36–45. empyema. Am J Respir Crit Care Med 2003; 168: 700–5.
64. Henriques GMO, Miotla JM, Cordeiro RSB, et al. Selectins mediate 85. Sasse S, Nguyen T, Texeira LR, Light RW. The utility of daily
eosinophil recruitment in vivo: a comparison with their role in therapeutic thoracentecis for the treatment of early empyema.
neutrophil influx. Blood 1996; 187: 5297–304. Chest 1999; 116: 1703–8.
References 183
86. Sasse S, Nguyen TK, Mulligan M, et al. The effects of early chest 107. Dodson RF, Atkinson MAL, Levin JL. Asbestos fiber length as
tube placement on empyema resolution. Chest 1997; 111: 1679–83. related to potential pathogenicity: a critical review. Am J Ind Med
87. Zhu Z, Hawthorne ML, Guo Y, et al. Tissue plasminogen activator 2003; 44: 291–7.
combined with human recombinant deoxyribonuclease is effective 108. Dodson RF, Ford JO. Early response of the visceral pleura following
therapy for empyema in a rabbit model. Chest 2006; 129: asbestos exposure: an ultrastructural study. J Toxicol Environ
1577–83. Health 1985; 15: 673–86.
88. Strahilevitz J, Lev A, Levi I, Fridman E, Rubinstein E. Experimental 109. Viallat JR, Raybuad F, Passarel M, Boutin C. Pleural migration of
pneumococcal pleural empyema model: the effect of moxifloxacin. chrysotile fibers after intratracheal injection in rats. Arch Environ
J Antimicrob Chemother 2003; 51: 665–9. Health 1986; 41: 282–6.
89. Teixeira LR, Sasse SA, Villarino MA, et al. Antibiotic levels in 110. Fasske E. Pathogenesis of pulmonary fibrosis induced by chrysotile
empyemic pleural fluid. Chest 2000; 117: 1734–9. asbestos. Longitudinal light and electron microscopic studies on
90. Liapakis IE, Light RW, Pitiakoudis MS, et al. Penetration of the rat model. Virchows Arch A Pathol Anat Histopathol 1986;
clarithromycin in experimental pleural empyema model fluid. 408: 329–46.
Respiration 2005; 72: 296–300. 111. Coin PG, Osornio-Vargas AR, Roggli VL, Brody AR. Pulmonary
91. Patterson RC. The pleural reaction to innoculation with tubercle fibrogenesis after three consecutive inhalation exposures to
bacilli in vaccinated and normal guinea pigs. Am Rev Tuberc 1917; chrysotile asbestos. Am J Respir Crit Care Med 1996; 154:
1: 353–71. 1511–19.
92. Widstrom O, Nillsson BS. Pleurisy induced by intrapleural BCG in 112. McConnell EE, Axten C, Hesterberg TW, et al. Studies on the
immunized guinea pigs. Eur Respir J 1982; 63: 425–34. inhalation toxicology of two fiberglasses and amosite asbestos in
93. Widstrom O, Egberg N, Chmielewska J, Blomback M. Fibrinolytic the Syrian golden hamster. Part II. Results of chronic exposure.
and coagulation mechanisms in stages of inflammation: a study of Inhal Toxicol 1999; 11: 785–835.
BCG-induced pleural exudate in guinea pig. Thromb Res 1983; 29: ●113. Tanaka S, Choe N, Hemenway DR, et al. Asbestos inhalation
511–19. induces reactive nitrogen species and nitrotyrosine formation in
●94. Phalen SW, McMurray DN. T-lymphocyte response in a guinea pig the lungs and pleura of the rat. J Clin Invest 1998; 102: 445–54.
model of tuberculous pleuritis. Infect Immun 1993; 61: 142–5. 114. Bozelka BE, Sestini P, Gaumer HR, et al. A murine model of
95. Antony VB, Repine JE, Harada RN, Good JT Jr, Sahn SA. asbestosis. Am J Pathol 1983; 112: 326–37.
Inflammatory responses in experimental tuberculosis pleurisy. Acta 115. Fasske E. Experimental lung tumors following specific
Cytol 1983; 27: 355–61. intrabronchial application of chrysotile asbestos. Longitudinal
●96. Antony VB, Sahn SA, Antony AC, Repine JE. Bacillus Calmette- light and electron microscopic investigations in rats. Respiration
Guerin-stimulated neutrophils release chemotaxins for monocytes 1988; 53: 111–27.
in rabbit pleural spaces and in vitro. J Clin Invest 1985; 76: 116. Humphrey EW, Ewing SL, Wrigley JV, et al. The production of
1514–21. malignant tumors of the lung and pleura in dogs from
97. Moura AC, Leonardo PS, Henriques MG, Cordeiro RS. Opposite intratracheal asbestos instillation and cigarette smoking. Cancer
effects of M. leprae or M. bovis BCG delipidation on cellular 1981; 47: 1994–9.
accumulation into mouse pleural cavity. Distinct accomplishment 117. Wagner JC, Berry G, Skidmore JW, Pooley FD. The comparative
of mycobacterial lipids in vivo. Inflamm Res 1999; 48: 308–13. effects of three chrysotiles by injection and inhalation in rats.
98. Menezes-de-Lima-Junior O, Werneck-Barroso E, Cordeiro RS, IARC Sci Publ 1980; 30: 362–72.
Henriques MG. Effects of inhibitors of inflammatory mediators and ◆118. Maxim LD, McConnell EE. Interspecies comparisons of the toxicity
cytokines on eosinophil and neutrophil accumulation induced by of asbestos and synthetic vitreous fibers: A weight-of-the-
Mycobacterium bovis bacillus Calmette–Guerin in mouse pleurisy. evidence approach. Regul Toxicol Pharmacol 2001; 33: 319–42.
J Leukoc Biol 1997; 62: 778–85. 119. Brass DM, Hoyle GW, Poovey HG, Liu J-Y, Brody AR. Reduced
●99. Allen SS, McMurray DN. Coordinate cytokine gene expression in tumor necrosis factor-a and transforming growth factor-b1
vivo following induction of tuberculous pleurisy in guinea pigs. expression in the lungs of inbred mice that fail to develop
Infect Immun 2003; 71: 4271–7. fibroproliferative lesions consequent to asbestos exposure. Am J
●100. Allen SS, Cassone L, Lasco TM, McMurray DN. Effect of Pathol 1999; 154: 853–62.
neutralizing transforming growth factor b1 on the immune 120. National Research Council. In: Council NR (ed.). Review of the US
response against Mycobacterium tuberculosis in guinea pigs. Navy’s exposure standard for manufactured vitreous fibers.
Infect Immun 2004; 72: 1358–63. Washington DC: National Academies Press, 2000: 1–63.
◆101. Lee YCG, De Klerk NH, Henderson DW, Musk AW. Malignant 121. Warheit D, Hartsky MA. Influence of gender, species and strain
mesothelioma. In: Hendrick DJ, Burge PS, Beckett WS, Churg A differences in pulmonary toxicological assessments of inhaled
(eds). Occupational disorders of the lung. Recognition, particles and/or fibers. In: Mohr U, Dungworth DL, Mauderly JL,
management and prevention. London, UK: W B Saunders & Co., Oberdorster G (eds). Toxic and carcinogenic effects of solid
2002: 359–379. particles in the respiratory tract. Washington DC: ILSI Press, 1994.
102. Chapman SJ, Cookson WO, Musk AW, Lee YC. Benign asbestos 122. Adamson IYR, Bowden DH. Response of mouse lung to crocidolite
pleural diseases. Curr Opin Pulm Med 2003; 9: 266–71. asbestos. 1. Minimal fibrotic reaction to short fibres. J Pathol
103. Stathopoulos GT, Karamessini M, Sotiriadi AE, Pastromas VG. 1987; 152: 99–107.
Rounded atelectasis of the lung. Respir Med 2005; 99: 615–23. 123. Adamson IYR, Bowden DH. Response of mouse lung to crocidolite
●104. Marchi E, Liu W, Broaddus VC. Mesothelial cell apoptosis is asbestos. 2. Pulmonary fibrosis after long fibres. J Pathol 1987;
confirmed in vivo by morphological change in cytokeratin 152: 109–17.
distribution. Am J Physiol Lung Cell Mol Physiol 2000; 278: 124. Hesterberg TW, Axten C, McConnell EE, et al. Studies on the
L528–35. inhalation toxicology of two fiberglasses and amosite asbestos in
●105. Boylan AM, Ruegg C, Kim KJ, et al. Evidence of a role for the syrian golden hamster. Part I. Results of a subchronic study
mesothelial cell-derived interleukin-8 in the pathogenesis of and dose selection for a chronic study. Inhal Toxicol 1999; 11:
asbestos-induced pleurisy in rabbits. J Clin Invest 1992; 89: 747–84.
1257–67. 125. Voss B, Kerenyi T, Muller KM, Wilhelm M. Scanning electron
106. Adamson IYR, Bakowska J, Bowden DH. Mesothelial cell microscopical investigations of broncho-alveolar casts after
proliferation after instillation of long or short asbestos fibers into intratracheal asbestos fibre instillation. Int J Hyg Environ Health
mouse lung. Am J Pathol 1993; 142: 1209–16. 2000; 203: 127–34.
184 Experimental models: pleural diseases other than mesothelioma
126. Choe N, Tanaka S, Xia W, et al. Pleural macrophage recruitment regional variation of pleural liquid thickness in rabbits. Lung 1997;
and activation in asbestos-induced pleural injury. Environ Health 175: 165–73.
Perspect 1997; 105 (Suppl 5): 1257–60. 148. Stashenko GJ, Robichaux A, Lee YCG, et al. Pleural fluid exchange
127. Oberdoerster G, Ferin J, Marcello NL, Meinhold SH. Effect of in rabbits. Respirol, in press.
intrabronchially instilled amosite on lavagable lung and pleural 149. Nishida O, Arellano R, Cheng DC, DeMajo W, Kavanagh BP. Gas
cells. Environ Health Perspect 1983; 51: 41–8. exchange and hemodynamics in experimental pleural effusion. Crit
●128. Zanella CL, Posada J, Tritton TR, Mossman BT. Asbestos causes Care Med 1999; 27: 583–7.
stimulation of the extracellular signal-related kinase 1 mitogen- 150. Sousa AS, Moll RJ, Pontes CF, Saldiva PH, Zin WA. Mechanical and
activated protein kinase cascade after phosphorylation of the morphometrical changes in progressive bilateral pneumothorax
epidermal growth factor receptor. Cancer Res 1996; 56: and pleural effusion in normal rats. Eur Respir J 1995; 8: 99–104.
5334–8. 151. Dechman G, Sato J, Bates JH. Effect of pleural effusion on
129. Robledo RF, Buder-Hoffman SA, Cummins AB, et al. Increased respiratory mechanics, and the influence of deep inflation, in dogs.
phosphorylated extracellular signal-regulated kinase Eur Respir J 1993; 6: 219–24.
immunoreactivity associated with proliferative and morphologic 152. Rush JE, Hamlin RL. Effects of graded pleural effusion on QRS in
lung alterations after chrysotile asbestos inhalation in mice. Am J the dog. Am J Vet Res 1985; 46: 1887–91.
Pathol 2000; 156: 1307–16. 153. Murphy DJ, Renninger JP, Gossett KA. A novel method for chronic
130. Adamson IYR, Prieditis H, Young L. Lung mesothelial cell and measurement of pleural pressure in conscious rats. J Pharmacol
fibroblast responses to pleural and alveolar macrophage Toxicol Methods 1998; 39: 137–41.
supernatants and to lavage fluids from crocidolite-exposed rats. 154. Wilkinson PD, Keegan J, Davies SW, Bailey J, Rudd RM. Changes in
Am J Respir Cell Mol Biol 1997; 16: 650–6. pulmonary microvascular permeability accompanying re-
131. Liu J-Y, Brass DM, Hoyle GW, Brody AR. TNF-a receptor knockout expansion oedema: evidence from dual isotope scintigraphy.
mice are protected from the fibroproliferative effects of inhaled Thorax 1990; 45: 456–9.
asbestos fibers. Am J Pathol 1998; 153: 1839–47. ●155. Clauss M, Sunderkoetter C, Sveinbjoernsson B, et al. A permissive
●132. Shore BL, Daughaday CC, Spilberg I. Benign asbestos pleurisy in role for tumor necrosis factor in vascular endothelial growth
the rabbit. A model for the study of pathogenesis. Am Rev Respir factor–induced vascular permeability. Blood 2001; 97: 1321–9.
Dis 1983; 128: 481–5. 156. Zebrowski BK, Yano S, Liu W, et al. Vascular endothelial growth
133. Markham KM, Glover JL, Welsh RJ, Lucas RJ, Bendick PJ. factor levels and induction of permeability in malignant pleural
Octreotide in the treatment of thoracic duct injuries. Am Surg effusions. Clin Cancer Res 1999; 5: 3364–8.
2000; 66: 1165–7. 157. Faurschou P, Grunnet N, Winding O, Dirksen A, Faarup P.
●134. Huang XZ, Wu JF, Ferrando R, et al. Fatal bilateral chylothorax in Rheumatoid arthritis cells and biological changes in turpentine-
mice lacking the integrin a9b1. Mol Cell Biol 2000; 20: 5208–15. induced pleuritis in rabbits. APMIS 1989; 97: 413–18.
135. Vlahakis NE, Young BA, Atakilit A, Sheppard D. The 158. Devin CJ, Lee YCG, Light RW, Lane KB. Pleural space as a site of
lymphangiogenic vascular endothelial growth factors VEGF-C and ectopic gene delivery: Transfection of pleural mesothelial cells
-D are ligands for the integrin a9b1. J Biol Chem 2005; 280: with systemic distribution of gene product. Chest 2003; 123: 202–
4544–52. 8.
136. Maulitz R, Good JT, Kaplan RL, Reller LB, Sahn SA. The 159. Katano K, Tsujitani S, Saito H, et al. Hypotonic intrapleural
pleuropulmonary consequences of esophageal rupture: An cisplatin chemotherapy as treatment for pleural carcinomatosis in
experimental model. Am Rev Respir Dis 1979; 120: 363–7. an experimental model. Anticancer Res 1997; 17: 4547–52.
137. Good JTJ, Antony VB, Reller B, Maulitz R, Sahn SA. The 160. Rosenfeldt FL, Glover JR, Marossy D. Systemic absorption of
pathogenesis of the low pleural fluid pH in esophageal rupture. noxythiolin from the pleural cavity in man and in the rabbit.
Am Rev Respir Dis 1983; 127: 702–4. Thorax 1981; 36: 278–81.
●138. Wiener-Kronish JP, Broaddus VC, Albertine KH, et al. Relationship ●161. Antony VB, Owen CL, Hadley KJ. Pleural mesothelial cells
of pleural effusions to increased permeability pulmonary edema in stimulated by asbestos release chemotactic activity for neutrophils
anesthetized sheep. J Clin Invest 1988; 82: 1422–9. in vitro. Am Rev Respir Dis 1989; 139: 199–206.
●139. Song Y, Yang B, Matthay MA, Ma T, Verkman AS. Role of 162. Fjellbirkeland L, Bjerkvig R, Steinsvag SK, Laerum OD. Nonadhesive
aquaporin water channels in pleural fluid dynamics. Am J Physiol stationary organ culture of human bronchial mucosa. Am J Respir
Cell Physiol 2000; 279: C1744–50. Cell Mol Biol 1996; 15: 197–206.
●140. Broaddus VC, Wiener-Kronish JP, Staub NC. Clearance of lung ●163. Kim K-U, Wilson SM, Abayasiriwardana KS, et al. A novel in vitro
edema into the pleural space of volume-loaded anesthetized model of human mesothelioma for studying tumor biology and
sheep. J Appl Physiol 1990; 68: 2623–30. apoptotic resistance. Am J Respir Cell Mol Biol 2005; 33: 541–8.
141. Kinasewitz GT, Groome LJ, Marshall RP, Leslie WK, Diana JN. Effect 164. Zarogiannis S, Hatzoglou C, Stefanidis I, et al. Adrenergic influence
of hypoxia on permeability of pulmonary endothelium of canine on the permeability of sheep diaphragmatic parietal pleura.
visceral pleura. J Appl Physiol 1986; 61: 554–60. Respiration 2007; 74: 118–20.
142. Ashino Y, Tanita T, Ono S, et al. Roles of the visceral pleura in the 165. Zarogiannis S, Hatzoglou C, Stefanidis I, et al. Effect of adrenaline
production of pleural effusion in permeability pulmonary edema. on transmesothelial resistance of isolated sheep pleura. Respir
Tohoku J Exp Med 1997; 182: 283–96. Physiol Neurobiol 2006; 150: 165–72.
143. Broaddus VC, Wiener-Kronish JP, Berthiauma Y, Staub NC. 166. Vogiatzidis K, Hatzoglou C, Zarogiannis S, et al. mu-Opioid
Removal of pleural liquid and protein by lymphatics in awake influence on transmesothelial resistance of isolated sheep pleura
sheep. J Appl Physiol 1988; 64: 384–90. and parietal pericardium. Eur J Pharmacol 2006; 530: 276–80.
144. Agostoni E, Zocchi L. Solute-coupled liquid absorption from the 167. Berkenkopf JW, Weichman BM. Comparison of several new 5-
pleural space. Respir Physiol 1990; 81: 19–27. lipoxygenase inhibitors in a rat Arthus pleurisy model. Eur J
145. Nakamura T, Tanaka Y, Fukabori T, et al. The role of lymphatics in Pharmacol 1991; 193: 29–34.
removing pleural liquid in discrete hydrothorax. Eur Respir J 1988; 168. Utsunomiya I, Ito M, Oh-ishi S. Generation of inflammatory
1: 826–31. cytokines in zymosan-induced pleurisy in rats: TNF induces IL-6
146. Broaddus VC, Araya M. Liquid and protein dynamics using a new and cytokine-induced neutrophil chemoattractant (CINC) in vivo.
minimally invasive pleural catheter in rabbits. J Appl Physiol 1992; Cytokine 1998; 10: 956–63.
72: 851–7. 169. Hanada S, Oga S, Hirano MT. Some characteristics of econazole-
147. Wang PM, Lai-Fook SJ. Effect of mechanical ventilation on induced pleurisy in rats. Gen Pharmacol 1985; 16: 637–40.
Appendix 2. Sheep pleurodesis model 185
170. Saleh TS, Calixto JB, Medeiros YS. Pro-inflammatory effects 4 A three-way stopcock is attached to the end of the chest
induced by bradykinin in a murine model of pleurisy. Eur J tube through which any aspirated air is immediately
Pharmacol 1997; 331: 43–52.
171. Perianin A, Roch-Arveiller M, Giroud JP, Hakim J. In vivo
evacuated from the pleural space.
interaction of nonsteroidal anti-inflammatory drugs on the 5 Reagents can be administered intrapleurally via the
locomotion of neutrophils elicited by acute non-specific chest tube, followed by the instillation of 1.0 mL of 0.9
inflammations in the rat – effect of indomethacin, ibuprofen and percent sodium chloride solution or sterile phosphate-
flurbiprofen. Biochem Pharmacol 1984; 33: 2239–43. buffered saline (PBS) to clear the dead space.
172. Wang JP, Ho TF, Lin CN, Teng CM. Effect of norathyriol, isolated
from Tripterospermum lanceolatum, on A23187-induced pleurisy
6 The chest tubes can be aspirated for pleural fluid.
and analgesia in mice. Naunyn Schmiedebergs Arch Pharmacol Alternatively, pleural lavage can be performed by the
1994; 350: 90–5. administration of 5–10 mL of sterile PBS via the chest
173. Kawamura K, Oh-ishi S. Rat pleurisy induced by kaolin or croton tube. The rabbit is then rotated and the PBS aspirated
oil: time course of fluid accumulation and white cell migration. Int back from the chest tube.
J Tissue React 1985; 7: 381–6.
174. Oh-ishi S, Hayashi I, Hayashi M, Yamaki K, Utsunomiya I.
7 The chest tube should be removed under light sedation
Pharmacological demonstration of inflammatory mediators using as soon as it is no longer required in order to minimize
experimental inflammatory models: rat pleurisy induced by risks of infection and discomfort.
carrageenan and phorbol myristate acetate. Dermatologica 1989; 8 At time of sacrifice, the rabbits are sedated and killed
179 (Suppl 1): 68–71. with carbon dioxide. The thorax is removed en bloc. The
175. Tarayre JP, Delhon A, Bruniquel F, et al. Exudative, cellular and
humoral reactions to platelet-activating factor (PAF-acether) in
lungs are expanded by the injection of 50 mL of 10
the pleural cavity of rats. Eur J Pharmacol 1986; 124: 317–23. percent neutral-buffered formalin into the exposed
176. Frode-Saleh TS, Calixto JB, Medeiros YS. Analysis of the trachea via a plastic catheter. The trachea is then ligated
inflammatory response induced by substance P in the mouse and the entire thorax submerged in 10 percent neutral-
pleurisy model. Peptides 1999; 20: 259–65. buffered formalin solution for at least 48 hours.
APPENDIX 14.1: RABBIT CHEST TUBE APPENDIX 14.2: SHEEP PLEURODESIS MODEL28
INSERTION22
1 Yearling sheep of mixed breeds can be anesthetized with
1 New Zealand white rabbits (>1.5 kg) are anesthetized an intravenous injection of 2.5 percent sodium
with an intramuscular injection of ketamine hydro- thiopental at 20 mg/kg.
chloride (35 mg/kg) and xylazine hydrochloride 2 The chest should be shaven and the skin sterilized with
(5 mg/kg). The chest is shaven and the skin sterilized 2 percent chlorhexidine and then with 10 percent
with 10 percent povidone iodine.* povidone iodine.
2 The rabbit is placed in the lateral decubitus position and 3 Using a laryngoscope, an endotracheal tube (8.5 mm
a small (<3 cm) skin incision is made midway between internal diameter) is inserted with an attached plastic
the tip of the scapula and the sternum approximately ‘bite block’ and secured with tape. Anesthesia is
2 cm above the costal margin. Chest tubes are made maintained with a gaseous mixture of room air, oxygen,
from intravenous solution set tubes with three extra and 1.5–2.5 percent halothane at a ventilation rate of 10
openings near the distal end of the tube to enhance breath cycles/minute with a volume of 15 mL/kg per
drainage. breath cycle. The sheep is placed on its side on a surgical
3 The chest tube is inserted by blunt dissection into the table and the feet secured to the table.
right pleural cavity and secured at the muscle layers 4 A 5 cm incision is made in the lateral chest wall at the
with purse-string sutures (3.0 ethilon). The proximal 7th intercostal space. By blunt dissection, an 18G
end of the chest tube is then tunneled underneath the French Foley ballooned-catheter with 30 mL balloon
skin and drawn out through the skin posteriorly and volume is inserted into the pleural space under aseptic
superiorly between the two scapulae. The exterior end conditions, tunneled underneath the skin and brought
of the chest tube is sealed with a one-way valve with cap to the surface just lateral to the vertebrae. The tube is
via an adapter and sutured to the skin using a 2.0 silk secured at the skin with purse-string sutures.
suture. 5 The sheep is then ventilated with a positive end
expiratory pressure of 15 cmH2O. A three-way stopcock
* An alternative method of pleural catheter placement has been is attached to the end of the Foley catheter through
described by Broaddus et al.65 Rabbits were anesthetized with which all air is evacuated from the pleural space
halothane and ventilated via tracheotomy. The upper abdomen was
immediately after the chest tube insertion.
opened and the diaphragm was punctured through which a fine
catheter was passed into the pleural space, and secured with sutures. 6 Intrapleural injection of agents can be made via the
This method is more invasive but is effective and can avoid any chest tube. The buffer or vehicle can be injected to the
potential lung trauma from blunt dissection into the pleura. contralateral side and serve as the control.
186 Experimental models: pleural diseases other than mesothelioma
7 The chest tube is aspirated (with the Foley catheter 3 Trypsin-EDTA (ethylenediamine tetraacetic acid) (0.25
balloon inflated) regularly for any pleural fluid pro- percent) is injected into the pleural cavity and left in situ
duced. To minimize discomfort and risk of infection, for 10 minutes during which the animal should be
we recommend that the chest tubes be removed as soon rotated. The solution, with the mesothelial cells, is then
as no further intrapleural injections or pleural sample aspirated and put into fetal calf serum (FCS) or culture
collections is needed. media (e.g. DMEM [Dulbecco’s modified Eagles
For pleural fibrosis/pleurodesis studies, the sheep are medium] with 10 percent FCS) on ice. The serum
killed 14 days after the chest tube insertion with an contains tryptase that will terminate the action of, and
intravenous injection of sodium phenobarbital any potential damage from, the injected trypsin. An
solution. injectate volume of 1 mL is used in mice, and 10 mL (for
each side) in rabbits.
4 The FCS is centrifuged at 1000 rpm for 5 minutes. The
cell pellet is washed and resuspended in DMEM with 1
APPENDIX 14.3: METHODS FOR HARVESTING percent (v/v) L-glutamine, 1 percent (v/v) penicillin–
RABBIT AND MICE MESOTHELIAL CELLS22,161 streptomycin and 10 percent (v/v) FCS.
5 The cells can be plated in standard cell culture flasks,
1 Pleural mesothelial cells are obtained from mice of and incubated at 37°C with 95 percent of air and 5
adult size or from New Zealand white rabbits percent of CO2. Initial cell population will consist of a
(commonly 2 kg). large number of erythrocytes and leukocytes as well as
2 After the animals are killed, the abdomen is opened to the mesothelial cells.
expose the diaphragm. Hank’s Balanced Salt Solution 6 The media should be changed after overnight
(HBSS) is injected into the pleural cavity from beneath incubation. Mesothelial cells should adhere to the
the hemi-diaphragms under direct vision and then culture flasks, and contaminating cells can usually be
aspirated out after 2 minutes. This is to remove surface removed with the media.
proteolytic enzymes to allow greater efficacy of trypsin 7 The cells can be stained for mesothelial markers to
(see below). In mice, there is no mediastinal separation confirm the epithelial origin of the cells. Fibroblasts that
and a single injection of 1mL of HBSS is sufficient to may have adhered to the culture flasks are cytokeratin
rinse the pleural surface in the left and right chest.* In negative. Most investigators are able to achieve a 95
rabbits, 10 mL is injected into each pleural cavity. percent purity of mesothelial cells using this technique.
Reasons for establishing models of malignant mesothelioma 187 Experimental models designed to characterize the role of 193
Human malignant mesothelioma as experimental models 187 the immune system in malignant mesothelioma
Animal malignant mesothelioma models 189 Key points 195
SV40 models 190 References 195
Uses of experimental models of malignant mesothelioma 191
Malignant mesothelioma (MM) is often diagnosed late in Generating cell lines from human MM primary
disease progression and, at this stage, is associated with a tumors
rapid decline in health and death within a short period of
time; the average period of survival being 9 months post- Examination of samples of excised tumors reveals useful
diagnosis. Additionally, in the absence of reliable specific information, however this approach is limited to a few
tumor markers that could be utilized for early diagnosis, technologies such as histopathology, although microarray
MM will remain undetectable until characteristic clinical analysis is now providing additional important data.
symptoms manifest, which then require rigorous Therefore, to gain further insights, cell lines that have
histopathological confirmation of disease. As a result, the been generated from human tumors and prepared as
tumor cell biology, as well as the specific and non-specific single-cell suspensions, cloned and their phenotypes,
immune responses to MM, are difficult to monitor in soluble factor secretion and responses to numerous
humans. Hence, experimental models of MM remain the agents including chemotherapeutics and gene transfer
only option available to facilitate a deeper understanding vectors, have been studied in detail. Much of this work
of (i) how this cancer progresses in vivo, (ii) how the has provided the basis of clinical trials and is the founda-
immune system is responding, and (iii) how we can alter tion for animal experimental models that are discussed in
tumor growth with varying single and combination thera- detail below.
pies. This chapter will explore the experimental
approaches that have been, or are being, undertaken to
address features of MM development. These approaches Identifying the cellular effects of asbestos
are divided into models that utilize human-derived MM injury and modeling the transformation
tumor cells, and environmentally (asbestos or glass fibers process
and/or Simian virus 40 [SV40]) induced animal MM
tumor systems. Examples of the experimental models used In 1960, Wagner et al.1 reported an association between
and their outcomes are discussed. asbestos and both pleural and peritoneal malignant
188 Experimental models: mesothelioma
mesothelioma in a case series from the North Western The role of Simian virus in the development
Cape Province of South Africa where blue asbestos (cro- of MM
cidolite) was mined. Since then, many reports from all
parts of the world have confirmed the relationship The transformation process may not be as simple as a
between exposure to asbestos and the development of direct relationship between asbestos fiber exposure and
MM. Once epidemiological studies had identified a clear mesothelial cells. Many humans were accidentally exposed
association between asbestos fibers and the subsequent to this small double-stranded DNA monkey virus via con-
development of MM in humans, a number of studies taminated polio vaccines (produced in monkey cells) in
were undertaken to examine the effects that exposure to the 1950s and 1960s. SV40 DNA sequences have been
asbestos fibers has on cell types located within the site found in a defined group of human cancers, of which MM
‘injured’ by these fibrous particles; i.e. serosal tissue con- has been reported to have the highest frequency of detec-
sisting of a surface mesothelial layer and subsurface spin- tion;12–14 although the latter is currently the subject of con-
dled connective tissue cells. Mesothelial cells were shown troversy.
to be inherently susceptible to asbestos fibers; they are The biology of SV40 has been well studied. Following
actively phagocytic in culture and ingest asbestos fibers.2 infection of a host cell, SV40 expresses two proteins, small
Mesothelial cells are up to 10 times more susceptible t (transforming) antigen (tag) and large T antigen (TAg),
than bronchial epithelial cells to the direct cytotoxic which interact with host cell proteins to cause cell prolifer-
effects of asbestos fibers.3 However, in vitro responses ation and DNA replication and hence the production of
were different depending on fiber types (chrysotile versus more viral particles. Tag is the replicase of SV40 and its
amphiboles) as well as the chemical state of these expression can lead to cellular transformation, principally
fibers.4,5 through inhibition of cellular p53 and retinoblastoma
The presence of asbestos fibers leads to the formation of (Rb) family proteins.15 SV40 infects cells from different
reactive oxygen metabolites which are directly toxic species: it causes lytic infection in the cells of its natural
causing DNA point mutations, as well as strand and chro- primate host where it does not induce tumors. In contrast,
mosomal breaks in mesothelial cells.6 These abnormalities it is non-lytic and highly oncogenic in rodents.16 Crucially,
usually result in cellular apoptosis. However, particular infection is semi-permissive in human cell lines and, in
mutations in combination with direct mitotic damage7 some reports, up to 60 percent of human mesotheliomas
and increased cell division may result in cell survival, contain SV40 DNA.14,17 According to microdissection
despite their profound genetic abnormalities, thereby experiments, SV40 is present in the malignant cells and
increasing the risk of neoplastic transformation. Studies sometimes in reactive mesothelial cells but not in normal
exposing fibroblasts to chrysotile fibers demonstrated dis- adjacent tissues; neither is it found in lung cancers.18,19 It
tinctive morphological changes within the first 12 hours of should be noted that although many different laboratories
exposure. Continued serial passage was associated with have reported the presence of SV40 in human MM and
further changes, including increasing cell size and loss of other cancers, there is still uncertainty over the prevalence
control of directional growth, indicating the beginning of of SV40 in mesothelioma. The most common method
transformation.4 used to detect SV40 sequences has been by polymerase
Understanding the molecular changes of mesothelial chain reaction (PCR). The potential risk of false positives
cells as they differentiate into MM is a topic of ongoing from contamination by the SV40 DNA contained in com-
research. These include studies demonstrating that monly used laboratory plasmids led to a lack of confidence
asbestos-transformed cells expressed transforming growth in the data. However, SV40 has been detected using other
factor (TGF)-a transcripts, whilst spontaneously trans- technical approaches and strains of SV40 never used in
formed cells do not. In addition, TGF-a inhibited only laboratories have been rescued from human biopsies.
the growth of spontaneously transformed mesothelial Nonetheless, this issue remains controversial.
cells, indicating that TGF-a acts as an autocrine growth Fibroblasts are transformed by SV40 at a low rate,
factor for asbestos-transformed rat mesothelial cells.8 whereas mesothelial cells are uniformly infected, but not
Similarly, we have demonstrated clear inhibition of MM lysed, leading to a high rate of transformation and immor-
cell growth using antisense oligonucleotides which block talization.20 Crocidolite asbestos increases the rate of
TGF-b and platelet-derived growth factor (PDGF)-A transformation, though the increment is not huge, perhaps
chain, indicating key roles for these factors in MM prolif- suggesting that asbestos and SV40 may be co-carcinogens.
eration.9,10 Differences in mesothelioma etiology may be It is feasible that asbestos has a more profound effect in
reflected in differences in the molecular alterations vivo because it is immunosuppressive, or because it causes
present in these tumors.8 Others have recently shown that the production of mutagenic free radicals from activated
proteoglycan (PG) expression is closely associated with macrophages. Taken together, these results suggest that
the morphology and biological behavior of tumor cells, mesothelial cells might be unusually susceptible to SV40
and that MM has a different PG profile from epithelial infection and transformation. Why this is so and how the
tumors.11 infected cells interact with asbestos is not fully understood.
Animal malignant mesothelioma models 189
Hahn and colleagues21 have shown that epithelial cells can biological characteristics of the original tumor. Examples
be transformed and become tumorigenic upon combined include studies such as those undertaken by Chahinian et
transfection with TAg, H-ras and the catalytic subunit of al.22 and Reale et al.23 who transplanted human pleural
telomerase. Given the fact that SV40 TAg inhibits both Rb MM from patients into nude mice. Tumors grew as a solid
and p53 proteins, they suggested that a minimum of four neoplastic mass. Light and electron microscopy, as well as
distinct signaling pathways need to be affected to trans- immunocytochemistry, demonstrated a similarity of the
form normal human cells. It would not be surprising then transplanted solid and fluid malignancies with the human
if SV40-infected mesothelial cells still need a strong prolif- primary MM. Both groups concluded that their models
eration-driving signal such as H-ras for tumorigenicity; mimicked the clinical behavior of human MM.
mutations of ras are not a feature of mesotheliomas, but
the pathway is functional.
It should be noted that epidemiological studies have Uncontrolled environmentally induced animal
been unable to link increased incidence of mesothelioma models
or other cancers with the contaminated vaccines; nonethe-
less, concern has been raised that vaccinated people world- Animals share man’s domicile environment, yet do not
wide may have been inadvertently exposed to an indulge in activities (e.g. smoking and working environ-
oncogenic virus. A thorough assessment of all the relevant ments) which confound interpretation of epidemiological
epidemiological data was investigated by the Institute of studies. Glickman et al.24 used pet dogs with spontaneous
Medicine; they concluded that the current evidence was (histologically confirmed) MM to identify environmental
insufficient and inadequate to accept or reject a direct exposures that might increase their owner’s risk of
causal relationship between SV40-containing vaccines and asbestos-related disease. An asbestos-related occupation,
cancer. or hobby, of a household member and use of flea repel-
To develop a useful model and explore the role of SV40 lents on the dog were significantly associated with MM. In
one approach was to transfect normal human mesothelial addition, there was an increased risk of MM within an
cells with a plasmid containing SV40 early region DNA, urban residence. Lung tissue from dogs with MM had
and select cells for their longevity; i.e. passaged continu- higher levels of chrysotile asbestos fibers than lung tissue
ously for more than 2 years.12 These cells expressed SV40 from control dogs. The authors argued that these findings
large T antigen, exhibited features of mesothelial cells, suggest that well-designed epidemiological studies of
including sensitivity to the cytotoxic effects of asbestos spontaneous tumors in pet animals may provide insight
fibers, but failed to develop into tumors when injected into the role of environmental factors in human cancers
subcutaneously or intraperitoneally into nude mice, sug- and serve as a valuable sentinel model to identify environ-
gesting that other co-factors (such as asbestos fibers) were mental health hazards for humans
a prerequisite for malignancy to develop. Thus, these cells
may represent an ideal model to assess the role of SV40 as
an MM co-factor. More recently, SV40 transgenic animal Controlled environmentally induced tumors
models have been developed and are discussed below.
A number of studies have examined the effects exposure to
asbestos fibers has in animal models. As early as 1969, rats
ANIMAL MALIGNANT MESOTHELIOMA were inoculated with asbestos fibers and the incidence of
MODELS MM determined.25 Detailed studies of pulmonary deposi-
tion, biodurability, biopersistence and carcinogenicity
Faithful animal models (experimentally reproducing the after asbestos fiber inhalation in rats, mice and hamsters
human disease) remain necessary to study the natural were conducted.26–29 Animals were examined for the pres-
history of MM, and to test standard or novel treatments. ence of benign and malignant lung tumors and MM. The
MM tumors usually occurred at intervals between 12 and
31 months after asbestos exposure. Whilst these models
Using human tumors in animals confirmed a direct association between inhalation of
asbestos fibers and MM, some concern has been expressed
Experimental models were often constructed by inoculat- that rodent inhalation models may not be sensitive enough
ing human tumors into immunologically deficient mice, to predict the cancer risk posed by varying fiber types for
such as the athymic bald mouse (nude mouse) or severe humans.
combined immunologically deficient (SCID) mice to Hesterberg et al.30 exposed hamsters and rats to a range
avoid immunological rejection. These murine models of man-made fibers and confirmed that exposure to croci-
often employ subcutaneous injection of cloned human dolite or chrysotile asbestos induced pulmonary fibrosis,
MM cells, or subcutaneous implantation of tumor frag- lung tumors and MM in rats. Interestingly, however, they
ments, as well as pleural implantation of intact human also demonstrated that exposure to refractory ceramic
tumors; the latter attempting to more closely conserve the fibers (RCF) resulted in significant increases in lung
190 Experimental models: mesothelioma
tumors and MM. In contrast, inhalation of fiber-glass inhalation model.31 Murine MM tumor cell lines have
(MMVF [man-made vitreous fiber] 10 or 11), slag wool been established after intraperitoneal inoculation of
(MMVF 22) and rock wool (stone wool: MMVF 21) was asbestos fibers into the major strains of mice, including
not associated with MM. These studies support the argu- CBA, BALB/c and C57BL/6, and are diagnostically similar
ment that chemical composition and the surface physico- to human MM tumors.31,34–36 Similar to human mesothe-
chemical properties of the fibers may play an important lial cells, growth of the murine MM cell lines can be stim-
role in MM development.4,5,31 ulated by epidermal growth factor.36 Upon subcutaneous
Kucharczuk et al.32 established a pleural-based model of injection of these murine MM cell lines into mice, solid
MM in immune-competent Fischer rats. This was tumors form and grow rapidly. There is minimal lympho-
achieved by placing a syngeneic MM cell line (II-45) into cytic infiltration and the most prominent infiltrating
the pleural cavity via a modified left anterior lateral thoro- leukocyte is the macrophage that makes up 50 percent of
cotomy. Pleural MM that closely resembles the human the tumor mass. These murine models allow evaluation of
disease was seen with animals dying within 1 month. different aspects of immune responsiveness and biological
However, this is an invasive and labor-intensive model and diversity (due to genetic or strain differences) and are the
is restricted in its experimental applicability. An alternative basis for experimental models designed to answer specific
approach involved exposing rat pleural mesothelial cells to questions (discussed below) (Table 15.1).
chrysotile fibers in vitro, and then prior to their transfor-
mation into MM, transplanting them into nude mice.33
Interestingly, tumors arose even from untreated rat SV40 MODELS
mesothelial cells, but the delay between cell injection and
tumor formation was 22 weeks, whereas only 2 weeks were The role of SV40 in the development of mesotheliomas has
needed with asbestos-treated cells. been explored in animal models in a variety of ways and
Similar studies have been conducted in rodents using SV40 alone induces mesotheliomas in hamsters. However,
the intraperitoneal route to inject asbestos and other we recently used the mesothelin promoter to construct
fibers. The intraperitoneal model may be more sensitive four mouse lines that express SV40 TAg in mesothelial
for testing the carcinogenicity of inorganic fibers than the cells at different levels.37 All of these mice show a relatively
Human MM Murine MM
Asbestos induced + +
Variable latency + +
Effusion + +
Histology:
Epithelial + +
Sarcomatous + +
Mixed + +
Ultrastructure:
Microvilli + +
Glycogen granules + +
Tight junctions + +
Growth in culture ± (approx. 12%) + (86%)
Variable morphology in culture + +
MHC surface expression:
Class I + ±
Class II - -
Tumorigenicity ± (nude mice) + (syngeneic mice)
Tumor suppressor expression:
p53 + +
Soluble factor secretion:
TGF-b + +
VEGF + +
PDGF-A chain + +
IL-6 + +
MHC, major histocompatibility complex; TGF-b, transforming growth factor beta; VEGF, vascular
endothelial growth factor; PDGF, platelet-derived growth factor; IL-6, interleukin 6.
Uses of experimental models of malignant mesothelioma 191
low level of spontaneous tumor development. When upregulated.53 However, no consistent association of these
exposed to asbestos fibers, high-copy mice rapidly devel- particular oncogenes to human MM has been demon-
oped faster growing and more invasive mesotheliomas strated.54,55
than those developing in wild-type or single-copy mice.
These data support the concept of co-carcinogenicity
between SV40 and asbestos. This model is ideal for in vivo To assess the role of SV40 in the evolution of
mesothelioma studies as it provides spontaneous tumors mesothelioma
that occur in response to asbestos and does not employ
transplantable tumor cell lines. An important link between mesothelioma and SV40 was
made in 1993 when injection of wild-type SV40 into the
pleural space of hamsters resulted in development of peri-
USES OF EXPERIMENTAL MODELS OF cardial or pleural mesothelioma in 100 percent of cases.16
MALIGNANT MESOTHELIOMA Interestingly, the presence of antisense IGF-1 receptor
DNA inhibited the tumorigenicity of SV40 in these ham-
Use of MM cell lines to identify genetic sters56 suggesting that SV40 uses the IGF signaling pathway
alterations arising from asbestos fiber injury: to induce MM.
oncogenes and tumor suppressor genes Simian virus 40 is highly immunogenic and has a
number of mapped immunodominant epitopes which are
Neoplastic transformation leads to genomic damage which recognized by cytotoxic T lymphocytes in mice.57 Thus,
should be detected and repaired by normal cellular Imperiale et al.58 speculated that the unique properties of
processes. Tumor suppressor (TS) genes are involved in the SV40 virus could be exploited to treat patients with
these processes and arrest the cell cycle to allow repair of SV40-positive MM. A modified SV40 T antigen, from
genetic abnormalities; if these are not repaired cells enter which the transforming domains were removed but
programmed cell death (apoptosis). Mutation or loss of a immunogenicity preserved, was cloned into a vaccinia
TS gene encourages unrestricted growth and proliferation virus vector and shown, in an animal tumor model, to be
of an altered cell. The best-described TS gene is p53 which effective against SV40 Tag expressing pre-existing tumors
is mutated in the majority of human cancers.38 Mutations and subsequent tumor challenge. Survival duration was
in p53 have been identified in 75 percent of murine MM increased when pre-existing tumors were treated with the
cell lines.39. However, p53 may overexpressed in most SV40 vaccinia virus construct in combination with
human MM cell lines and in human MM primary immunotherapy. Plans to investigate the potential of this
tumors.40–43 Similarly, the retinoblastoma gene was shown construct in a human trial are in progress.
to be expressed normally in a number of human MM cell
lines.44 However, use of monoclonal antibodies directed
against two different epitopes of the retinoblastoma gene To assess the efficiency of varying single or
product, and immunohistochemistry of primary tumors, multi-modality therapeutic approaches
revealed differing reaction patterns in neoplastic and non-
neoplastic mesothelial cells,44 suggesting that the expressed Much of the early work (often employing in vitro models)
retinoblastoma gene product may be abnormal in MM. examining a single therapy (such as chemotherapy, dis-
The Wilm’s tumor gene (WT-1) is another interesting cussed below) indicated doses and sensitivities for use in
TS gene which is often mutated in Wilm’s tumor and is clinical trials however, the results of these trials have gen-
expressed in mesothelium during embryogenesis. erally been disappointing. It is becoming increasingly clear
Expression of this gene has been detected in both human in human studies, as well as in animal models, that single-
MM cell lines and in primary tumors.45,46 One of the modality therapy is unlikely to result in long-term sur-
actions of the WT-1 protein is to control the transcription vival. Hence, relatively new techniques, particularly those
of genes such as PDGF-A,47 insulin-like growth factor that offer an adjuvant setting, are being considered for use
(IGF)-II48 and TGF-b.49 These agents have been described in combination with other treatments. Gene therapies are
as potential autocrine growth factors in MM and the dele- under active investigation employing either established
tion of WT-1 could lead to their de-regulated production. animal models or constructing new ones, and these are
However, it has recently been shown that there is no described in more detail below.
inverse correlation in expression of WT-1, IGF-II or
PDGF-A in MM.50
Specific oncogenes contribute directly to the progres- To assess the efficiency of chemotherapeutic
sion of malignancy and the v-src gene has been shown to drugs
cause MM in chickens,51 whilst transfection of the EJ-ras
gene into mesothelial cells causes neoplastic transforma- Standard chemotherapy drugs result in response rates that
tion.52 When rat pleural mesothelial cells are exposed to are usually less than 20 percent and with little evidence of
asbestos the levels of both c-fos and c-jun mRNA become any meaningful increased response using combination
192 Experimental models: mesothelioma
chemotherapy.59 A large number of new chemotherapy vivors in both murine models with the parental tumor cell
agents have been developed in the last decade. Many of line demonstrated the generation of specific, long-lasting
these have shown activity against a wide range of solid systemic immunity mediated by cytolytic CD8+ T cells and
tumors and some are now being tried in MM. It would be natural killer (NK) cells.
desirable to trial each of these drugs alone and in combi- Immunotoxins, constructed by linking ricin to murine
nation in MM; unfortunately, this is difficult given the monoclonal antibodies reactive with the human transfer-
small numbers of patients who are eligible for such trials in rin receptor, were shown to be potent in vitro cytotoxins
any one center. One extensively used alternative to assess against human MM cell lines. Their in vivo potential was
the clinical potential of single and combination chemo- evaluated in a nude mouse model of human MM.67 The
therapeutic drugs is to examine their in vitro effects on survival of tumor-bearing mice was significantly extended
MM cell lines.60,61 Similarly, much of the work involving but no long-term cure was reported.
human tumors transplanted into immunocompromised
mice was established primarily to examine the chemosen- Identifying molecular targets for
sitivity of these tumors in vivo.22,62 Based on the results, pharmacological therapies
several clinical trials have been conducted. For example,
after a combination of mitomycin C and cisplatin was Methionine aminopeptidase-2 (MetAP2) is the molecular
described as the most effective regimen for xenografted target of the angiogenesis inhibitor fumagillin which can
human MM, a clinical trial using the same chemothera- also inhibit cancer cell proliferation. MM cells express
peutic agents reported that four of 12 patients showed higher MetAP2 mRNA levels than primary normal
objective responses (one complete and three partial).63 mesothelial cells and fumagillin induced apoptosis that
However, the benefits of chemotherapy should not be was restricted to the malignant cells, suggesting that
exclusively assessed by measuring response rates as deter- MetAP2 inhibition may represent a potential target for
mined by tumor shrinkage. Where a cancer is not curable, therapeutic intervention in human mesothelioma.68
the aim of treatment can be prolongation of life or pallia-
tion of symptoms. Mesothelioma patients often exhibit
severe systemic features and it is thought that one of the Non-immunological gene therapy
major mediators is the cytokine interleukin (IL)-6. IL-6 is
a pro-inflammatory cytokine and its blockade with anti- The nature and accessibility of MM tumors in humans
IL-6 antibodies has been shown to reduce cachexia and means that they are suitable candidates for direct cytokine
improve clinical status in mice with mesothelioma. Using and gene-transfer therapeutic approaches. Gene therapy
in vitro analysis, we have shown that irinotecan and gem- represents the introduction of a gene of interest into a site
citabine are not only more likely to be active against where it should be most effective. The methods of gene
mesothelioma than other chemotherapy agents, but may delivery vary greatly in terms of complexity, and are often
also produce a palliative effect in non-responders to these determined by the ultimate experimental or therapeutic
agents by decreasing IL-6 secretion.64 aims. For example, the aim might be to understand the
role of a gene as a therapeutic agent, or it might be to
determine whether or not this gene is significant in the
To assess the efficiency of new forms natural progression of MM. The latter approach was often
of therapy employed to understand which immunologically active
molecules were important to induce or maintain an effec-
Bacterial products such as diphtheria toxin and mycobac- tive immune response and involved transfecting cDNA
terial heat shock protein 65 (HSP65)65,66 have been coding for these molecules into MM cell lines and deter-
assessed for their in vivo tumoricidal effects in animal mining their in vivo tumorigenicity (see below). The
models. A single intraperitoneal or IV dose of diphtheria cytokine gene therapy approaches are described elsewhere.
toxin consistently and rapidly cured athymic mice of Gene therapy using adenovirus to deliver Herpes
advanced stage experimental human MM. The complete simplex virus (HSV) thymidine kinase (Ad.HSVtk) was
and direct tumoricidal effect without an associated evaluated in immunocompetent mice with established
immune response implies that toxin readily reached, abdominal MM tumor.69 Mice were treated with multiple
entered and preferentially killed human cancer cells. intraperitoneal injections of Ad.HSVtk followed by daily
Similarly, long-term survival was seen in immunocompe- administration of the pro-drug ganciclovir, and showed
tent mice bearing intraperitoneal progressing syngeneic significantly improved survival versus singly injected
MM cells (AC29 or AB12) that had been treated with animals and control animals. However, the response was
cationic lipid complexed with plasmid DNA (pDNA) con- significantly improved in immunosuppressed mice, there-
taining hsp65.66 Interestingly, survivors were also observed fore immunosuppression may be a useful adjunct.70
in groups treated with lipid complexed with any pDNA, Similarly, a modified, non-neurovirulent HSV effectively
although lipid alone or DNA alone provided no demon- treated a localized intraperitoneal malignancy (human
strable survival advantage. Re-challenging long-term sur- MM cells) in SCID mice.71
Experimental models designed to characterize the role of the immune system in malignant mesothelioma 193
Pleural models of MM demonstrated a widespread dis- immune surveillance. We have shown that numerous
tribution of infectious virus particles throughout the immune cell types infiltrate MM tumors including T cells,
thorax after intrapleural treatment with adenovirus B cells, NK cells and large numbers of macrophages (our
vectors that resulted in tumor growth inhibition.32,72 The unpublished data).75,76 There is contradictory evidence
use of these models led to a number of clinical trials. regarding their local regulation although downregulation
of some lymphocyte surface markers, known to be
involved in T cell activation, has been demonstrated in
tumor infiltrating lymphocytes (TIL).77 Similarly, expres-
EXPERIMENTAL MODELS DESIGNED TO sion of MHC class II antigen and integrins was weak on
CHARACTERIZE THE ROLE OF THE IMMUNE tumor infiltrating macrophages (TIM), suggesting altered
SYSTEM IN MALIGNANT MESOTHELIOMA functional activity. Significant amounts of TGF-b, IL-6,
IL-1 and tumor necrosis factor (TNF) were produced
Many experimental models were established to under- during the course of MM tumor development which may
stand the role of the (innate and adaptive) immune system contribute both to derangement of anti-tumor effector
as the MM tumors naturally progress, and to identify mechanisms and to the clinical and pathological manifes-
which, if any, therapies that act as immune adjuvants tations of this disease.78,79
could be successfully translated into a clinical trial. More Although MM tumor cells have abundant MHC class I
recently, exquisitely sensitive models using T cell receptor molecules, they do not express the co-stimulatory mole-
(TCR) transgenic mice have been constructed and the cules necessary to ensure that CD8+ T cells will become
power of these experimental systems is described below. effector cells. Hence, MM tumor cells are not able to induce
a class I-restricted response by directly presenting their
own antigens to CD8+ T cells. Furthermore, MM cell lines
Use of transfection/transgenic models to express little or no MHC class II molecules and are there-
understand tumor-specific immune responses fore are unable to directly activate CD4+ helper T cells.
to MM Thus, the only possible mechanism through which MM-
specific immune responses can be generated is via a third
Dissecting out the critical components involved in a spe- party antigen presenting cell (APC) that has collected MM
cific immune response directed against MM has been tumor antigens after trafficking through the tumor on its
hampered by the current lack of known MM tumor anti- way to draining lymph nodes – a process referred to as
gens. Whilst identifying MM-specific tumor antigens rep- cross-presentation. Use of the transfection/transgenic
resents work-in-progress, one experimental approach that models showed that MM tumor growth in normal inbred
has been employed to overcome this obstacle is the con- mice is associated with constitutive tumor antigen cross-
struction of defined model systems using mouse MM presentation.79 Interestingly, in the odd instance when a
tumor cell lines transfected with nominal antigens such as mouse did not develop a solid tumor, in vivo tumor antigen
influenza hemagglutinin (HA)73 and ovalbumin (OVA),74 presentation was detectable for up 6 months after tumor
to which TCR transgenic mice are available. Importantly, cell inoculation, implying that either the immune system
expression of either HA or OVA as neo tumor antigens kept tumor growth in check or that the tumor antigen was
does facilitate tumor rejection in syngeneic, immunologi- somehow trapped within the draining lymph nodes.
cally intact animals. In both systems the class I and class II Adoptive transfer of both tumor-specific CD4+ and
peptide reactivities are well defined and, anti-HA as well as CD8+ T cells together offered complete protection from
anti-OVA, TCR transgenic mice are available with class I tumor growth,81 and induced regression in MM tumor-
and II specificities. The advantage of using these TCR bearing mice. Tumor-specific CD4+ T cells were required
transgenic mice is that they provide a virtually monoclonal for the prolonged survival or ‘maintenance’ of functional
source of cells of known specificity where the respective tumor-specific CD8+ T cells, as well as their emigration
roles of tumor-specific CD8+ and CD4+ T cells can be into the tumor mass itself. It was only in the presence of
evaluated. Thus, it is possible to investigate when and CD4+ T cells that MM tumors exhibited upregulation of
where MM-associated tumor antigens are presented in major histocompatibility complex (MHC) class II and
vivo during tumor progression, how the frequency of intercellular adhesion molecule (ICAM) expression which
tumor-specific T cells influences tumor immunity, and the may be necessary to generate an effective anti-tumor
role that specific CD4+ T cells play in modulating the immune response.
immune response. More recently, potent regulatory T cells (Treg) that can
be identified by their co-expression of CD4, CD25 and the
transcription factor FoxP3 have been demonstrated in
The pathobiology and immunobiology of MM murine and human MM tumors.82,83 Treg cells play a
major role in the maintenance of self-tolerance, the
Murine models for MM have been used to gain insight control of autoimmune and, more deleteriously, in anti-
into the mechanism(s) whereby MM might escape tumor immune responses. Depletion of Treg cells signifi-
194 Experimental models: mesothelioma
cantly reduced tumor growth underlining their regulatory released systemically, can only be realistically evaluated
capacity. These data suggest that combining therapies with using in vivo models.
Treg depletion may be clinically beneficial.
and permanent tumor regression. This response was asso- re-emerge. However, when debulking surgery was per-
ciated with tumor infiltrating CD8+ T cells and reduced formed in conjunction with vaccination using syngeneic
tumor-associated vascularity. Tumor progression MM tumor cells transfected with genes encoding for B7-1
inevitably occurred in mice with large tumors when treat- or high levels of GM-CSF, there was a statistically signifi-
ment was commenced. It was evident that there was only a cant delay of tumor growth.96 Debulking surgery appeared
small time-frame in which IL-2 treatment could be effec- to play a pivotal role in promoting an anti-tumor immune
tive. response induced by immunological gene therapy.
More recently, we have shown that partial and com-
plete debulking surgery was highly effective in eradicating
Use of agonist anti-CD40 antibody therapy in MM tumors when combined with chemotherapy and a
mesothelioma CD40-based immunotherapy.102 However, only those
mice with partially debulked tumors plus the combination
CD40 is an Mr 40 000 type I glycoprotein and a member of regimen generated long-term, tumor-specific, protective
the tumor necrosis factor receptor superfamily, which was memory. We postulate that chemotherapy induced apop-
initially identified on bladder carcinoma cells and later on tosis of residual tumor cells following incomplete resection
normal and malignant B cells. It is expressed on a range of is absolutely required for the induction of long-term
immune and non-immune cells, the most important immunological memory. These data are the basis of a
immune cell being dendritic cells (DCs). Its ligand, CD40L current clinical trial.
(CD154), is expressed on CD4 T cells shortly after TCR In conclusion, experimental models of MM continue to
triggering. CD40–CD154 interactions have an important offer meaningful information for direct clinical transla-
role in cytotoxic lymphocyte (CTL) priming and their tion.
interaction is central to the decision whether CTLs become
primed or tolerized. Thus, use of an activating (agonist)
anti-CD40 antibody can replace or augment CD4 help in KEY POINTS
priming DCs to activate CD8 T cells. Systemic treatment
with an activating anti-CD40 antibody, FGK45, in our ● Mesothelioma models are generally similar to
murine MM models causes transient tumor regression human mesothelioma – this is unusual for mouse
over the treatment period, followed by rapid tumor out- tumor models.
growth.100 In contrast, intra-tumoral anti-CD40 antibody ● Sarcomoatoid rather than epithelial mesothe-
can completely cure small but not large MM tumors (our lioma are more commonly seen in mice, in con-
unpublished data). These data suggest that CD40 trigger- trast to human mesothelioma.
ing can only be effective when combined with other thera- ● New transgenic models have added powerful
peutic modalities. tools to the existing repertoire of mesothelioma
models of transplantable or asbestos-induced in
situ models.
Chemotherapy and anti-MM immunity ● These models are proving useful for the analysis
of biological events in mesothelioma.
Cytotoxic chemotherapy is generally considered immuno- ● These models are also useful for preclinical
suppressive, with neutropenia and lymphopenia being screening of potential therapies.
common adverse side effects. However, we have shown
that whilst the cytidine analogue gemcitabine abolishes
humoral responses, it in fact augments antigen-specific
cellular anti-tumor immunity.101 This augmentation REFERENCES
occurs in the context of increased antigen cross-presenta-
tion, T lymphocyte expansion and tumor infiltration. ● = Key primary paper
Combining gemcitabine with a CD40-triggering ◆ = Major review article
immunotherapy resulted in synergistic, leading to long-
term, protective cure when applied to small tumors. Thus, ◆1. Wagner J, Sleggs C, Marchand P. Diffuse pleural mesothelioma
and asbestos exposure in the North Western Cape Province. Br J
chemotherapy has the capacity to augment cellular antitu-
Ind Med 1960; 17: 260–71.
mor immunity. 2. Jaurand M, Kaplan H, Thiollet J, et al. Phagocytosis of chrysotile
fibers by pleural mesothelial cells in culture. Am J Pathol 1979;
94: 529–38.
Combination immune therapy with debulking 3. Lechner J, Tokiwa T, LaVeck M, et al. Asbestos-associated
surgery in murine MM chromosomal changes in human mesothelial cells. Proc Natl Acad
Sci U S A 1985; 82: 3884–8.
4. Joseph L, Stephens R, Ottolenghi A, et al. Morphological
We have previously shown that debulking surgery is rarely transformation in vitro of normal human fibroblasts by chrysotile.
permanently effective as resected MM tumors eventually Environ Health Perspect 1983; 51: 17–22.
196 Experimental models: mesothelioma
5. Bignon J, Jaurand M. Biological in vitro and in vivo responses of 28. Miller B, Jones A, Searl A, et al. Influence of characteristics of
chrysotile versus amphiboles. Environ Health Perspect 1983; 51: inhaled fibers on development of tumours in the rat lung. Ann
73–80. Occup Hyg 1999; 43: 167–79.
6. Walker C, Everitt J, Barrett J. Possible cellular and molecular 29. Muhle H, Pott F. Asbestos as reference material for fiber-induced
mechanisms for asbestos carcinogenicity. Am J Ind Med 1992; 21: cancer. Int Arch Occup Environ Health 2000; 73 (Suppl): S53–9.
253–73. ●30. Hesterberg T, Miller W, Thevenaz P, Anderson R. Chronic inhalation
7. Yegles M, Saint-Etienne L, Renier A, et al. Induction of metaphase studies of man-made vitreous fibers: characterization of fibers in
and anaphase/telophase abnormalities by asbestos fibers in rat the exposure aerosol and lungs. Ann Occup Hyg 1995; 39: 637–53.
pleural mesothelial cells in vitro. Am J Respir Cell Mol Biol 1993; 31. Pott F. Detection of mineral fiber carcinogenicity with the
9: 186–91. intraperitoneal test: recent results and their validity. Ann Occup
8. Walker C, Everitt J, Ferriola P, et al. Autocrine growth stimulation Hyg 1995; 39: 771–9.
by transforming growth factor alpha in asbestos-transformed rat ◆32. Kucharczuk J, Elshami A, Zhang H, et al. Pleural-based
mesothelial cells. Cancer Res 1995; 55: 530–6. mesothelioma in immune competent rats: a model to study
9. Garlepp M, Christmas T, Manning L, et al. The role of platelet- adenoviral gene transfer. Ann Thorac Surg 1995; 60: 593–7.
derived growth factor in the growth of human malignant 33. Fleury-Feith J, Nebut M, Saint-Etienne L, et al. Occurrence and
mesothelioma. Eur Respir Rev 1993; 3: 189–91. morphology of tumors induced in nude mice transplanted with
10. Garlepp M, Christmas T, Mutsaers S, et al. Platelet-derived growth chrysotile-transformed rat pleural mesothelial cells. Biol Cell
factor as an autocrine factor in murine malignant mesothelioma. 1989; 65: 45–50.
Eur Resp Rev 1993; 3: 192–4. ◆34. Christmas T, Manning L, Davis MR, et al. HLA antigen expression
11. Dobra K, Andang M, Syrokou A, et al. Differentiation of and malignant mesothelioma. Am J Respir Cell Mol Biol 1991; 5:
mesothelioma cells is influenced by the expression of 213–20.
proteoglycans. Exp Cell Res 2000; 258: 12–22. 35. Davis M, Manning L, Whitaker D, et al. Establishment of a murine
12. Ke Y, Reddel R, Gerwin B, et al. Establishment of a human in vitro model of malignant mesothelioma. Int J Cancer 1992; 52: 881–6.
mesothelial cell model system for investigating mechanisms of 36. Goodglick L, Vaslet C, Messier N, Kane A. Growth factor responses
asbestos-induced mesothelioma. Am J Pathol 1989; 134: 979–91. and protooncogene expression of murine mesothelial cell lines
●13. Procopio A, Strizzi L, Vianale G, et al. Simian virus-40 sequences derived from asbestos-induced mesotheliomas. Toxicol Pathol
are a negative prognostic cofactor in patients with malignant 1997; 25: 565–73.
pleural mesothelioma. Genes Chromosomes Cancer 2000; 29: 37. Robinson C, van Bruggen I, Segal A, et al. A novel SV40 TAg
173–9. transgenic model of asbestos-induced mesothelioma: malignant
14. McLaren B, Haenel T, Stevenson, et al. RA. SV40 like sequences in transformation is dose dependent. Cancer Res 2006; 66:
cell lines and tumour biopsies from Australian malignant 10786–94.
mesotheliomas. Aust NZ J Med 2000; 30: 450–6. 38. Hollstein M, Sidransky D, Vogelstein B, Harris C. p53 mutations in
15. Bryan T, Reddel R. SV40-induced immortalization of human cells. human cancers. Science 1991; 253: 49–53.
Crit Rev Oncog 1994; 5: 331–57. 39. Cora E, Kane A. Alterations in a tumour suppressor gene, p53, in
16. Cicala C, Pompetti F, Carbone M. SV40 induces mesotheliomas in mouse mesotheliomas induced by crocidolite asbestos. Eur Resp
hamsters. Am J Pathol 1993; 142: 1524–33. Rev 1993; 3: 148–150.
17. Butel J, Lednicky J. Cell and molecular biology of simian virus 40: 40. Metcalf R, Welsh J, Bennett W, et al. p53 and Kirsten-ras
Implications for human infections and disease. J Natl Cancer Inst mutations in human mesothelioma cell lines. Cancer Res 1992; 52:
1999; 91: 119–134. 2610–15.
18. Carbone M. Simian virus 40 and human tumours: It is time to 41. Mor O, Yaron P, Huszar M, et al. Absence of p53 mutations in
study mechanisms. J Cell Biochem 1999; 76: 189–93. malignant mesotheliomas. Am J Respir Cell Mol Biol 1997; 16:
19. Shivapurkar N, Wiethege T, Wistuba I, et al. Presence of simian 9–13.
virus 40 sequences in malignant mesotheliomas and mesothelial 42. Creaney J, McLaren B, Stevenson S, et al. p53 autoantibodies in
cell proliferations. J Cell Biochem 1999; 76: 181–8. patients with malignant mesothelioma: stability through disease
20. Bocchetta M, Di Resta I, Powers A, et al. Human mesothelial cells progression. Br J Cancer 2001; 84: 52–6.
are unusually susceptible to simian virus 40-mediated 43. Van der Meeren A, Seddon M, Betsholtz C, et al. Tumorigenic
transformation and asbestos cocarcinogenicity. Proc Nat Am Soc conversion of human mesothelial cells as a consequence of
2000; 97: 10214–19. platelet-derived growth factor-A chain overexpression. Am J
21. Hahn W, Counter C, Lundberg A, et al. Creation of human tumour Respir Cell Mol Biol 1993; 8: 214–21.
cells with defined genetic elements. Nature 1999; 400: 464–8. 44. Ramael M, Segers K, Van Marck E. Differential
22. Chahinian A, Beranek J, Suzuki Y, et al. Transplantation of human immunohistochemical staining of the retinoblastoma protein with
malignant mesothelioma into nude mice. Cancer Res 1980; 40: the antibodies C15 and IF8 in malignant mesothelioma. Pathol Res
181–5. Pract 1994; 190: 138–41.
23. Reale F, Griffin T, Compton J, et al. Characterization of a human 45. Walker C, Rutten F, Yuan X, et al. Wilms’ tumor suppressor gene
malignant mesothelioma cell line (H-MESO-1): a biphasic solid expression in rat and human mesothelioma. Cancer Res 1994; 54:
and ascitic tumor model. Cancer Res 1987; 47: 3199–205. 3101–6.
24. Glickman L, Domanski L, Maguire T, et al. Mesothelioma in pet 46. Amin K, Litzky L, Smythe W, et al. Wilms’ tumor 1 susceptibility
dogs associated with exposure of their owners to asbestos. Environ (WT1) gene products are selectively expressed in malignant
Res 1983; 32: 305–13. mesothelioma. Am J Pathol 1995; 146: 344–56.
◆25. Berry G, Wagner J. The application of a mathematical model 47. Wang Z, Madden S, Deuel T, Rauscher F. The Wilms’ tumor gene
describing the times of occurrence of mesotheliomas in rats product, WT1, represses transcription of the platelet-derived
following inoculation with asbestos. Br J Cancer 1969; 23: 582–6. growth factor A-chain gene. J Biol Chem 1992; 267: 21999–2002.
26. Fasske E. Experimental lung tumors following specific 48. Drummond I, Madden S, Rohwer-Nutter P, et al. Repression of the
intrabronchial application of chrysotile asbestos. Longitudinal insulin-like growth factor II gene by the Wilms tumor suppressor
light and electron microscopic investigations in rats. Respiration WT1. Science 1992; 257: 674–8.
1988; 53: 111–27. 49. Dey B, Sukhatme V, Roberts A, et al. Repression of the
27. Smith W, Hubert D, Holiat S, et al. An experimental model for transforming growth factor-beta 1 gene by the Wilms’ tumor
treatment of mesothelioma. Cancer 1981; 47: 658–63. suppressor WT1 gene product. Mol Endocrinol 1994; 8: 595–602.
References 197
50. Langerak A, Williamson K, Miyagawa K, et al. Expression of the experimental malignant mesothelioma using adenovirus vectors
Wilms’ tumor gene WT1 in human malignant mesothelioma cell encoding the HSVtk gene. Gene Ther 1997; 4: 280–7.
lines and relationship to platelet-derived growth factor A and 73. Marzo A, Kinnear B, Lake R, et al. Tumor-specific CD4+ T cells have
insulin-like growth factor 2 expression. Genes Chromosomes a major ‘post-licensing’ role in CTL mediated anti-tumor immunity.
Cancer 1995; 12: 87–96. J Immunol 2000; 165: 6047–55.
51. England J, Panella M, Ewert D, Halpern S. Induction of a diffuse ●74. Jackaman C, Bundell C, Kinnear B, et al. IL-2 intratumoral
mesothelioma in chickens by intraperitoneal inoculation of v-src immunotherapy enhances CD8+ T cells that mediate destruction of
DNA. Virology 1991; 182: 423–9. tumor cells and tumor-associated vasculature: a novel mechanism
52. Reddel R, Malan-Shibley L, Gerwin B, et al. Tumorigenicity of for IL-2. J Immunol 2003; 171: 5051–63.
human mesothelial cell line transfected with EJ-ras oncogene. J 75. Garlepp M, Leong C, Biological and immunological aspects of
Natl Cancer Inst 1989; 81: 945–8. malignant mesothelioma. Eur Respir J 1995; 8: 643–50.
53. Heintz N, Janssen Y, Mossman B. Persistent induction of c-fos and 76. Bielefeldt-Ohmann H, Fitzpatrick D, Marzo A, et al. Patho- and
c-jun expression by asbestos. Proc Natl Acad Sci U S A 1993; 90: immunobiology of malignant mesothelioma: characterisation of
3299–303. tumour infiltrating leucocytes and cytokine production in a murine
54. Fung H, Quinlan T, Janssen Y, et al. Inhibition of protein kinase C model. Cancer Immunol Immunother 1994; 39: 347–59.
prevents asbestos-induced c-fos and c- jun proto-oncogene 77. Fitzpatrick D, Bielefeldt-Ohmann H, Himbeck R, et al. Transforming
expression in mesothelial cells. Cancer Res 1997; 57: 3101–5. growth factor-beta: antisense RNA-mediated inhibition affects
55. Suzuki Y, Weston A. Immunocytochemical analysis of oncoproteins anchorage-independent growth, tumorigenicity and tumor-
and growth factors in human malignant mesothelioma. Oncol Rep infiltrating T-cells in malignant mesothelioma. Growth Factors
1995; 12: 897–902. 1994; 11: 29–44.
56. Pass H, Mew, D. Carbone, M. Inhibition of hamster mesothelioma 78. Fitzpatrick D, Peroni D, Bielefeldt-Ohmann H. The role of growth
tumorigenesis by an antisense expression plasmid to the insulin- factors and cytokines in the tumorigenesis and immunobiology of
like growth factor-1 receptor. Cancer Res 1996; 50: 4044–8. malignant mesothelioma. Am J Respir Cell Mol Biol 1995; 12:
57. Tevethia S. Recognition of simian virus 40 T antigen by cytotoxic T 455–60.
lymphocytes. Mol Biol Med 1990; 7: 83–96. 79. Bielefeldt-Ohmann H, Marzo A, Himbeck R, et al. Interleukin-6
58. Imperiale M, Pass H, Sanda M. Prospects for an SV40 vaccine. involvement in mesothelioma pathobiology: inhibition by
Semin Cancer Biol 2001; 11: 81–5. interferon alpha immunotherapy. Cancer Immunol Immunother
59. Ong S, Vogelzang N. Chemotherapy in malignant pleural 1995; 40: 241–50.
mesothelioma. A review. J Clin Oncol 1996; 14: 1007–17. 80. Marzo A, Lake R, Lo D, et al. Tumor antigens are constitutively
60. Goodman G, Yen Y, Cox T, Crowley J. Effect of verapamil on in presented in the draining lymph nodes. J Immunol 1999; 162:
vitro cytotoxicity of adriamycin and vinblastine in human tumor 5838–45.
cells. Cancer Res 1987; 47: 2295–304. 81. Marzo A, Lake R, Robinson B, Scott B. T-cell receptor transgenic
61. Bowman R, Manning L, Davis M. Chemosensitivity and cytokine analysis of tumor-specific CD8 and CD4 responses in the
sensitivity of malignant mesothelioma. Cancer Chemother eradication of solid tumors. Cancer Res 1999; 59: 1071–9.
Pharmacol 1991; 28: 420–6. 82. Meloni F, Morosini M, Solari N, et al. Foxp3 expressing CD4+
62. Arnold W, Tanneberger S, Nowak C, Naundorf H. Recent results CD25+ and CD8+CD28- T regulatory cells in the peripheral blood
with clinically used antineoplastic drugs and drug combinations in of patients with lung cancer and pleural mesothelioma. Hum
vivo. Oncology 1980; 37 (Suppl 1): 34–41. Immunol 2006; 67: 1–12
63. Chahinian A, Norton L, Holland J, et al. Experimental and clinical 83. Needham D, Lee J, Beilharz M. Intra-tumoural regulatory T cells: a
activity of mitomycin C and cis- diamminedichloroplatinum in potential new target in cancer immunotherapy. Biochem Biophys
malignant mesothelioma. Cancer Res 1984; 44: 1688–92. Res Commun 2006; 343: 684–91
64. McLaren BR, Robinson BWS, Lake RA. New chemotherapeutics in 84. Manning L, Bowman R, Darby S, Robinson B. Lysis of human
malignant mesothelioma: Effects on cell growth and IL6 malignant mesothelioma cells by natural killer (NK) and
production. Cancer Chemother Pharmacol 2000; 45: 502–8. lymphokine-activated killer (LAK) cells. Am Rev Respir Dis 1989;
65. Raso V, McGrath J. Cure of experimental human malignant 139: 1369–1374.
mesothelioma in athymic mice by diphtheria toxin. J Natl Cancer 85. Mavaddat N, Robinson B, Rose A, et al. An analysis of the
Inst 1989; 81: 622–7. relationship between gamma delta T cell receptor V gene usage
66. Lanuti M, Rudginsky S, Force S, et al. Cationic lipid:bacterial DNA and non-major histocompatibility complex-restricted cytotoxicity.
complexes elicit adaptive cellular immunity in murine Immunol Cell Biol 1993; 71: 27–37.
intraperitoneal tumor models. Cancer Res 2000; 60: 2955–63. 86. Versnel M, Hagemeijer A, Bouts M, et al. Expression of c-sis (PDGF
67. Griffin T, Richardson C, Houston L, et al. Antitumor activity of B-chain) and PDGF A-chain genes in ten human malignant
intraperitoneal immunotoxins in a nude mouse model of human mesothelioma cell lines derived from primary and metastatic
malignant mesothelioma. Cancer Res 1987; 47: 4266–70. tumors. Oncogene 1988; 2: 601–5.
68. Catalano A, Romano M, Robuffo I, et al. Methionine 87. Gerwin B, Lechner J, Reddel R, et al. Comparison of production of
aminopeptidase-2 regulates human mesothelioma cell survival : transforming growth factor-beta and platelet-derived growth
role of bcl-2 expression and telomerase activity. Am J Pathol factor by normal human mesothelial cells and mesothelioma cell
2001; 159: 721–31. lines. Cancer Res 1987; 47: 6180–4.
69. Lambright E, Force S, Lanuti M, et al. Efficacy of repeated 88. Demetri G, Zenzie B, Rheinwald J, Griffin J. Expression of colony-
adenoviral suicide gene therapy in a localized murine tumor stimulating factor genes by normal human mesothelial cells and
model. Ann Thorac Surg 2000; 70: 1865–70. human malignant mesothelioma cells lines in vitro. Blood 1989;
70. Elshami A, Kucharczuk J, Sterman D, et al. The role of 74: 940–6.
immunosuppression in the efficacy of cancer gene therapy using 89. Lee TC, Zhang Y, Aston C, et al. Normal human mesothelial cells
adenovirus transfer of the herpes simplex thymidine kinase gene. and mesothelioma cell lines express insulin-like growth factor I
Ann Surg 1995; 222: 298–307. and associated molecules. Cancer Res 1993; 53: 2858–64.
●71. Kucharczuk J, Randazzo B, Chang M, et al. Use of a ‘replication- 90. Lauber B, Leuthold M, Schmitter D, et al. An autocrine mitogenic
restricted’ herpes virus to treat experimental human malignant activity produced by a pleural human mesothelioma cell line. Int J
mesothelioma. Cancer Res 1997; 57: 466–71. Cancer 1992; 50: 943–50.
72. Esandi M, van Someren G, Vincent A, et al. Gene therapy of 91. Schmitter D, Lauber B, Fagg B, Stahel RA. Hematopoietic growth
198 Experimental models: mesothelioma
factors secreted by seven human pleural mesothelioma cell lines: of mesothelioma. Immune and antitumor effects of transfected
interleukin-6 production as a common feature. Int J Cancer 1992; interleukin-12. Am J Respir Cell Mol Biol 1999; 21: 347–56.
51: 296–301. 98. Caminschi I, Venetsanakos E, Leong C, et al. Interleukin-12 induces
92. Monti G, Jaurand M, Monnet I, et al. Intrapleural production of an effective antitumor response in malignant mesothelioma. Am J
interleukin 6 during mesothelioma and its modulation by gamma- Respir Cell Mol Biol 1998; 19: 738–46.
interferon treatment. Cancer Res 1994; 54: 4419–23. 99. Bielefeldt-Ohmann H, Fitzpatrick D, Marzo A, et al. Potential for
93. Leong C, Marley J, Loh S, et al. Transfection of the gene for B7-1 interferon-alpha-based therapy in mesothelioma: assessment in a
but not B7-2 can induce immunity to murine malignant murine model. J Interferon Cytokine Res 1995; 15: 213–23
mesothelioma. Int J Cancer 1997; 71: 476–82. 100. Stumbles P, Himbeck R, Frelinger J, et al. Cutting edge: tumor-
94. Leong C, Robinson B, Garlepp M. Generation of an antitumour specific CTL are constitutively cross-armed in draining lymph
immune response to a murine mesothelioma cell line by the nodes and transiently disseminate to mediate tumor regression
transfection of allogeneic MHC genes. Int J Cancer 1994; 59: following systemic CD40 activation. J Immunol 2004; 173:
212–16. 5923–8.
95. Mukherjee S, Nelson D, Loh S, et al. The immune anti-tumor ◆101. Nowak, A, Robinson B, Lake R. Synergy between chemotherapy
effects of GM-CSF and B7-1 gene transfection are enhanced by and immunotherapy in the treatment of established murine solid
surgical debulking of tumor. Cancer Gene Ther 2001; 8: 580–8 tumors. Cancer Res 2003; 63: 4490–6.
96. Leong C, Marley J, Loh S, et al. The induction of immune responses ●102. Broomfield S, Currie A, van der Most R. Partial, but not complete,
to murine malignant mesothelioma by IL-2 gene transfer. Immunol tumor-debulking surgery promotes protective antitumor memory
Cell Biol 1997; 75: 356–9. when combined with chemotherapy and adjuvant immunotherapy.
97. Caminschi I, Venetsanakos E, Leong C, et al. Cytokine gene therapy Cancer Res 2005; 65: 7580–4.
PART 2
CLINICAL SCIENCE
EFFUSIONS
24 Effusions from cardiac diseases 315
Gary T Kinasewitz and Kellie R Jones
25 Effusions from malignancy 323
Francisco Rodriguez-Panadero
26 Effusions from infections: parapneumonic effusions and empyema 341
Najib M Rahman and Robert JO Davies
27 Effusions from infections: tuberculosis 367
Esteban Pérez-Rodriguez and Richard W Light
28 Effusions from infections: atypical infections 379
Lisete R Teixeira and Francisco S Vargas
29 Effusions from lymphatic disruptions 389
Gunnar Hillerdal
30 Effusions from vascular causes 397
José M Porcel and Richard W Light
31 Effusions in immunocompromised hosts 409
Bekele Afessa and John J Mullon
32 Effusions from connective tissue diseases 421
Demosthenes Bouros and Dimitris A Vassilakis
33 Effusions caused by drugs 431
Ioannis Kalomenidis
34 Effusions after surgery 445
Oner Dikensoy and Richard W Light
35 Hepatic hydrothorax 455
José Castellote and Xavier Xiol
36 Effusions caused by gastrointestinal disease 465
Charlie Strange
37 Effusions of obstetric or gynecological origin 473
Richard W Light
38 Benign fibrous tumor of the pleura 483
Marc de Perrot
39 Undiagnosed pleural effusions 491
Nick A Maskell
ASBESTOS-RELATED DISEASES
40 Asbestos-related pleural diseases 499
A William Musk and Nicholas H de Klerk
41 Malignant mesothelioma 507
Bruce WS Robinson
PNEUMOTHORAX
42 Spontaneous pneumothorax 515
Andrew C Miller
43 Non-spontaneous pneumothorax 533
Michael H Baumann
PEDIATRIC CONSIDERATIONS
44 Pediatric pleural diseases 545
Elizabeth A Perkett and Paul E Moore
INTERVENTION PROCEDURES
45 Drainage techniques 551
Henri Colt
46 Pleurodesis 569
Helen Davies and YC Gary Lee
47 Medical thoracoscopy 583
Robert Loddenkemper
48 Surgery for pleural diseases 599
David A Waller and Antonio E Martin-Ucar
49 Gene therapy in pleural diseases 613
Steven M Albelda and Daniel H Sterman
CONCLUSION
50 Future directions 621
YC Gary Lee and Richard W Light
16
Approach to patients with pleural diseases
STEVEN A SAHN
Patients with pleural effusions may present with symp- Table 16.1 Patients presenting with a pleural effusion who are
toms, such as pleuritic chest pain, dyspnea or cough, or the commonly asymptomatic
effusion may be suspected on physical examination or
Commonly asymptomatic patients presenting with a pleural
observed on chest radiograph. The diagnosis of a pleural
effusion
effusion signifies that the physiologic balance between
normal pleural fluid formation and removal has been dis-
turbed, resulting in pleural fluid accumulation. Since a ● Benign asbestos pleural effusion (BAPE)
pleural effusion can be a manifestation of disease in virtu- ● Hypoalbuminemia
ally any organ in the body, its presence provides the clini- ● Nephrotic syndrome
cian with the opportunity to support or confirm their ● Peritoneal dialysis
clinical diagnosis. Awareness that not only disease in the ● Rheumatoid pleurisy
thorax can cause a pleural effusion, but abnormalities of ● Trapped lung
organs juxtaposed to the diaphragm, such as the liver or ● Urinothorax
spleen, can lead to earlier diagnosis.1 In addition, systemic ● Yellow nail syndrome
diseases, such as systemic lupus erythematosus and
rheumatoid arthritis, and diseases of the lymphatic system,
such as yellow nail syndrome, may also cause pleural effu-
sions. Therefore, the evaluation of a patient with a pleural
effusion starts with, and requires, a thorough history and
physical examination in conjunction with pertinent labo-
Table 16.2 Patients presenting with a pleural effusion who are
ratory tests to allow the clinician to formulate a pre-
typically symptomatic
thoracentesis diagnosis. Pleural fluid analysis can provide
a confident diagnosis when the likelihood of a clinical Typically symptomatic patients presenting with a pleural
diagnosis is high. effusion
● Bacterial pneumonia
HISTORY ● Carcinomatous pleural effusion
● Congestive heart failure
Patients presenting with a pleural effusion may be asymp-
● Lupus pleuritis
tomatic, such as with benign asbestos pleural effusions
● Malignant mesothelioma
(BAPE)2 or rheumatoid pleurisy3 (Table 16.1) or have
● Postcardiac injury syndrome (PCIS)
symptoms as with lupus pleuritis4 or bacterial pneumonia
● Pulmonary embolism
(Table 16.2). Patients with small pleural effusions and
● Tuberculous pleural effusion
without underlying cardiopulmonary disease may be
● Viral pleurisy
asymptomatic at presentation and the effusion discovered
202 Approach to patients with pleural diseases
on a routine chest radiograph. Dyspnea and chest pain are prior to pleural fluid analysis. Orthopnea, paroxysmal
the two most common presenting symptoms of a pleural nocturnal dyspnea, peripheral edema and decreased exer-
effusion. The patient may complain of dyspnea with a cise tolerance suggest that the patient’s effusions are the
massive (occupying the entire hemithorax) or large (occu- result of congestive heart failure (CHF). A history of loss of
pying >50 percent of the hemithorax) pleural effusion and consciousness in an alcoholic, 14 days prior, who presents
normal lungs, a moderate (one-third to <50 percent) effu- with fever and fatigue suggests that the patient has an
sion with underlying lung disease, or a small (less than anaerobic empyema. The onset of dyspnea in a patient
one-third of the hemithorax) effusion in patients with with a recent leg fracture that required a cast suggests the
severe underlying lung disease. A large or massive effusion effusion was caused by pulmonary thromboembolic
causes contralateral mediastinal shift, depression of the disease. A unilateral pleural effusion in a man who has
ipsilateral hemidiaphragm, outward movement of the ipsi- worked in a shipyard for 20 years should raise the suspi-
lateral chest wall, and lung compression, in the absence of cion of BAPE. The postcardiac injury syndrome (PCIS)
an endobronchial lesion causing atelectasis or a fixed should be suspected in a patient who underwent cardiac
mediastinum. It is postulated that dyspnea perceived by surgery 3 weeks previously and presents with fever,
these patients results from decreased compliance of the dyspnea, and left pleuritic chest pain.8 Esophageal rupture
chest wall and lung modulated by the input of neurogenic should be considered in a patient who had recent
receptors from these structures.5 While lung compression esophageal dilatation, presents with a history of severe
is caused by large to massive effusions, a small to moderate retching and upper abdominal or lower chest pain, or has
effusion tends to cause lung displacement and generally sustained severe chest trauma.9 Procainamide use for the
has minimal or no effect on pulmonary function.6 The past 12 months or a known diagnosis of systemic lupus
major cause of dyspnea in patients with a small to moder- erythematosus should suggest lupus pleuritis.10 A history
ate pleural effusion may be related to chest pain with of Stage II or III sarcoidosis, rheumatoid disease or chronic
splinting and atelectasis or a primary parenchymal hemodialysis (uremic pleural effusion, tuberculous
process, such as pneumonia. pleurisy) should suggest diagnostic possibilities. Although
Pleuritic chest pain is associated with pleural inflamma- the number of drugs associated with pleural disease is sig-
tion and is typically accompanied by a pleural effusion.7 nificantly less than those that are presumed to cause
Pleuritic chest pain varies with the intensity of the pleural parenchymal lung disease, drugs should always be consid-
inflammation. Patients have described pleuritic chest pain ered as a possible cause of pleural effusion or fibrosis.
as ‘stabbing’ or ‘shooting’ in quality and as having a ‘stitch Some of the drugs that have been associated with a pleural
in the side’. Pleuritic chest pain is typically exacerbated by effusion in more than a single report include bromocrip-
a deep inspiration, coughing or sneezing. Manual pressure tine, dantrolene, nitrofurantoin, mitomycin, practolol,
over the chest wall, which results in splinting, will mini- procarbazine, methotrexate, mesalamine and iso-
mize the pain; however, a splinting maneuver cannot dif- tretinoin11 (see Chapter 33, Effusions caused by drugs).
ferentiate pleural inflammation from other causes of chest
pain, such as a rib fracture.
Pleuritic chest pain may be focused over the precise PHYSICAL EXAMINATION
location of the inflammation or it may be referred. With
costal pleural inflammation, the pain tends to be localized Pleural fluid interferes with sound transmission from the
directly over the site of pleural involvement and is often lung to the stethoscope because it separates the lung from
associated with tenderness on pressure and cutaneous the chest wall. The physical signs of a pleural effusion
hypersensitivity; abdominal pain is absent. When the depend upon the volume of pleural fluid and the degree of
lateral, anterior and portions of the posterior diaphragm lung compression. The status of the underlying lung and
are inflamed, pain is perceived diffusely over the lower the patency of the bronchial tree will modulate the physi-
thorax, back and abdomen, associated with cutaneous cal findings.
hyperesthesia, and exacerbated by pressure over the site When only 250–300 cm3 of pleural fluid is present,
with muscle rigidity. In contrast, inflammation of the detection by physical examination will be problematic.12
central portion of the diaphragmatic pleura does not result At a pleural fluid volume of approximately 500 cm3, the
in local pain, as pain is referred to the ipsilateral posterior typical physical findings are: (1) dullness to percussion; (2)
neck, shoulder and trapezius muscle; this referred pain is decreased fremitus; and (3) normal vesicular breath
associated with tenderness, hyperesthesia, hyperalgesia and sounds of decreased intensity compared with the contra-
muscle spasm. Central diaphragmatic pleural inflamma- lateral side.12 At a pleural fluid volume exceeding
tion causes referred pain because the sensory fibers of the 1000 cm3, there usually is: (1) absence of inspiratory
phrenic nerve enter the spinal cord at the C4 level, which is retraction and mild bulging of the intercostal spaces; (2)
the usual entry point of sensation from the shoulder.7 decreased expansion of the ipsilateral chest wall; (3) dull-
Because the primary symptoms of a pleural effusion, ness to percussion up to the level of the scapula and axilla;
chest pain and dyspnea, are nonspecific, a more detailed (4) decreased or absent fremitus posteriorly and laterally;
history is critical in narrowing the differential diagnosis (5) bronchovesicular breath sounds, which may be of
Diagnostic tests 203
decreased intensity at the upper level of the effusion; and exudative effusions with a normal heart size are most com-
(6) egophony (E to A change) at the upper level of the effu- monly malignant but can also be seen with lupus pleuritis
sion. With more marked lung compression, auscultation and rheumatoid pleurisy.37
may elicit bronchial breath sounds.12 When the effusion When a chest radiograph shows a pleural effusion(s)
fills the entire hemithorax (massive), physical examination with interstitial infiltrates, the differential diagnosis
will show: (1) bulging of the intracostal spaces; (2) includes congestive heart failure, rheumatoid disease,15
minimal to no expansion of the ipsilateral chest wall; (3) a asbestos pleuropulmonary disease,2 lymphangitic carcino-
dull or flat percussion note over the entire hemithorax; (4) matosis,38 lymphangioleiomyomatosis (LAM),39 viral and
absent breath sounds over the majority of the chest with mycoplasma pneumonia,40 sarcoidosis41 and Pneumocystis
possible bronchovesicular bronchial breath sounds at the jiroveci pneumonia.42
apex; (5) egophony at the upper level of the pleural effu- Pleural effusions associated with multiple nodules
sion; and (6) a palpable liver or spleen due to significant suggest cancer (most common), Wegener’s granulomato-
diaphragmatic depression.12 sis,43 rheumatoid disease,2 septic pulmonary emboli,44 sar-
coidosis41 and tularemia.45
DIAGNOSTIC TESTS
Pleural fluid analysis
Radiology
Virtually all patients with a newly discovered pleural
The chest radiograph can provide further diagnostic effusion should undergo thoracentesis to assist in diagno-
insight. Specific diseases should be considered if the only sis and management. Exceptions would be a secure clinical
abnormal radiographic finding is a pleural effusion or if the diagnosis, such as typical congestive heart failure or a very
effusion is associated with other abnormalities. For small pleural effusion in a patient with presumed viral
example, when the only abnormality on the chest radio- pleurisy. Observation is warranted in the above examples;
graph is a pleural effusion, infectious causes such as a however, if the clinical situation worsens or is atypical, a
tuberculous pleural effusion,13 viral pleurisy14 or a small thoracentesis should be performed without delay. For
bacterial pneumonia are possibilities. An isolated pleural example, if the patient with CHF has pleuritic chest pain,
effusion is more commonly observed with lupus pleuritis4 fever, a unilateral pleural effusion, a normal cardiac silhou-
and rheumatoid pleurisy15 than with another thoracic ette or an oxygen tension out of proportion to the clinical
manifestation of these diseases. Metastatic cancer, non- situation, a thoracentesis should be performed promptly.
Hodgkin lymphoma, and leukemia can also present as a In a prospective study of 129 patients with pleural effu-
solitary pleural effusion. Other diseases where a pleural sions, thoracenteses provided a definitive diagnosis (i.e.
effusion is typically the only radiographic abnormality malignancy) in only 18 percent of patients and a presump-
include BAPE,2 pulmonary embolism,16 drug-induced tive diagnosis (i.e. CHF) in 55 percent of patients.46 In the
pleural disease,11 yellow nail syndrome,17 hypothy- remaining 27 percent of patients, the pleural fluid findings
roidism,18 uremic pleuritis,19 chylothorax20 and were not helpful diagnostically because the values were
constrictive pericarditis.21 When a massive effusion is compatible with two or more clinical possibilities;
present that causes contralateral mediastinal shift, the most however, in a number of these patients, the findings were
likely diagnosis is carcinoma, usually a non-lung primary.22 useful in excluding possible diagnoses, such as empyema.
With a large or massive pleural effusion without contralat- Therefore, the clinician must assess the history and physi-
eral shift, lung cancer23 and malignant pleural cal examination, radiological evaluation and ancillary
mesothelioma24 are most likely. blood tests in establishing a pre-thoracentesis diagnosis so
Solitary pleural effusions may also be associated with that the pleural fluid findings can provide a confident,
disease below the diaphragm.1 Transudates from hepatic clinical diagnosis if the results are not definitive. With a
hydrothorax,25 nephrotic syndrome,26 urinothorax27 and more complete knowledge of pleural fluid analysis, the
peritoneal dialysis28 can cause this radiographic pattern. cases of undiagnosed pleural effusions should continue to
Exudates from acute29 and chronic pancreatitis,30 Meigs decrease.
syndrome,31 chylous ascites32 and subphrenic,33 hepatic34 A definitive diagnosis can only be established by pleural
and splenic abscesses35 or splenic hematomas36 can also be fluid analysis in a limited number of diseases that include
causative. empyema, malignancy, chylothorax, rheumatoid pleurisy,
Bilateral pleural effusions are most commonly transu- and others47 (Table 16.3).
dates, as seen with congestive heart failure, nephrotic syn- If the pleural effusion is clearly a transudate (see
drome, hypoalbuminemia, peritoneal dialysis and Chapter 17, Pleural fluid analysis), with low protein and
constrictive pericarditis. The cardiac silhouette is virtually lactate dehydrogenase (LDH) values, the diagnosis is
always enlarged in congestive heart failure but may be of limited and usually easily discernible from the patient’s
normal size with nephrotic syndrome, other causes of clinical presentation. Most transudates are due to CHF
hypoalbuminemia and constrictive pericarditis.37 Bilateral with the next most common (but much less frequent)
204 Approach to patients with pleural diseases
Table 16.3 Diagnoses that can be established definitively by pleural fluid analysis
cause being hepatic hydrothorax. An exudative pleural pleural fluid/serum ratio <0.5,54–56 or pleural fluid LDH
effusion, in contrast, has a much larger differential diagno- <0.6754,55 or <0.8256 of the upper limits of normal of the
sis of over 50 causes (see Chapter 17). Transudative effu- serum LDH), further testing is unnecessary. If the fluid is
sions are caused by imbalances in hydrostatic and oncotic exudative, additional tests can be performed on the pleural
pressures with normal pleurae, while exudative effusions fluid saved by the laboratory. Using this algorithm, there
are the result of inflammatory processes, malignancy and can be significant savings for the patient. If the preclinical
lymphatic abnormalities. diagnosis suggests an exudative effusion, most commonly
The differential diagnosis of the exudate may be further from pneumonia, malignancy or pulmonary embolism,
narrowed by complete pleural fluid analysis. If the pleural total cell count with differential and pH or glucose should
fluid pH is <7.30 with a normal blood pH, the exudative be determined. If pancreatic disease or malignancy is in
effusion essentially can be limited to six diagnoses48,49 (see the differential, amylase concentration should be deter-
Chapter 17). These include a complicated parapneumonic mined. If there is clinical suspicion of rheumatoid pleurisy
effusion or empyema (most common), malignancy, or lupus pleuritis, rheumatoid factor and a search for
esophageal rupture, rheumatoid pleurisy, lupus pleuritis lupus erythematosus (LE) cells, respectively, should be
and tuberculous pleural effusion. Rare causes of a low pH assessed. If chylothorax or a cholesteroal effusion is a con-
exudative pleural effusion are herniated bowel with infarc- sideration or the fluid is milky or turbid, the fluid should
tion50 and a large hemothorax with an abnormal pleura.48 be centrifuged; if the supernatant remains milky or cloudy,
The only singular cause of a transudate with a low pH is cholesterol and triglycerides should be measured. Since it
urinothorax, which is associated with ipsilateral obstruc- is rare for a malignant effusion to be transudative, cytology
tive uropathy.51 The same differential associated with a low does not need to be ordered routinely unless, for example,
pleural fluid pH is found with a low pleural fluid glucose the patient is elderly and has concomitant CHF. Cytology
(<60 mg/dL or a pleural fluid/serum ratio of <0.5)52 (see should be ordered in most patients with unilateral and
Chapter 17). those with bilateral exudative effusions and a normal heart
The presence of ≥80 percent lymphocytes or ≥10 size, if they are over 40 years of age, have a history of
percent eosinophils of the total pleural fluid nucleated cell cancer or the diagnosis is uncertain. Gram, fungal and
count is found with a limited number of diseases47 (see acid-fast bacilli (AFB) stains and corresponding cultures
Chapter 17); a tuberculous pleural effusion is the most should be obtained in all exudates where infection is likely
common diagnosis associated with marked lymphocytosis based on the clinical presentation.
and hemothorax is most prevalent with eosinophilia.
It has been suggested that if the clinical likelihood of a
transudative effusion is high, only total protein and LDH TIME TO RESOLUTION
should initially be measured and fluid for other
chemistries and cell counts saved by the laboratory.53 If the Knowledge of the time of resolution, either spontaneous
protein and LDH are low (total protein <3.0 gm/dL or or with therapy, of pleural effusions can be helpful diag-
Undiagnosed effusions 205
nostically57 (Table 16.4). For example, an effusion from a treatment should be given. With a negative repeat skin test,
pulmonary embolism rarely persists for more than 1 the pleural fluid should be studied with flow cytometry to
month,16 BAPE may not resolve for 2–6 months,58–60 evaluate for lymphoma.
tuberculous pleural effusions resolve over 1–4 months61 If the patient has a pleural malignancy, a repeat thora-
and effusions from yellow nail syndrome62 and trapped centesis a few weeks later will increase the yield of a posi-
lung are persistent.63 tive cytology, as the disease has become more advanced.
Pleural fluid carcinoembryonic antigen (CEA) has a high
(85 percent) diagnostic accuracy in diagnosing a carcino-
UNDIAGNOSED EFFUSIONS matous pleural effusion.65 If the patient or physician wants
to establish the diagnosis sooner, an experienced thoraco-
In the patient with an undiagnosed exudative pleural effu- scopist will be able to diagnose malignancy in 95 percent of
sion following the initial pleural fluid analysis including cases.
cytology and cultures, the options include observation, The pleural fluid from pulmonary embolism is present
repeat thoracentesis with or without percutaneous pleural on admission to the hospital, is small in volume (less than
biopsy and medical or video-assisted thoracic surgery one-third of the hemithorax), peaks at 72 hours and
(VATS). The most likely causes of an undiagnosed effusion resolves within 7–10 days when radiographic infarction is
are early-stage malignancy (cancer and lymphoma not present.16 When radiographic infarction (consolida-
(usually Hodgkin’s lymphoma) and mesothelioma), less tion) is visualized, the effusion will also peak by 72 hours
commonly tuberculosis, and benign persistent effusions but will not resolve completely for 2–3 weeks.16
from yellow nail syndrome, trapped lung, constrictive If an undiagnosed effusion is persistent and relatively
pericarditis or drug-induced pleural effusions. unchanged for several months or years, malignancy is less
Whether or not the patient with a tuberculous pleural likely and trapped lung, yellow nail syndrome, other lym-
effusion is treated with anti-tuberculosis medications, the phatic abnormalities or a cholesterol effusion should be
effusion will resolve in 4–16 weeks, unless lung entrap- considered. A trapped lung is a unilateral effusion of vari-
ment develops; however, the untreated patient has a 65 able size that, following thoracentesis, will return to its
percent chance of developing active pulmonary or extra- pre-procedure volume within a few days. The initial
pulmonary tuberculosis within the next 5 years.61 Culture pleural liquid pressure is typically negative and the pleural
of pleural fluid and tissue from pleural biopsy and pleural space elastance is increased (>15 cmH2O/L of fluid
tissue histology should establish the diagnosis of tubercu- removed).63 Pleural fluid analysis with trapped lung typi-
losis in 75–80 percent of cases.64 If the aforementioned cally reveals a serous lymphocyte-predominant transudate;
studies are negative and the pleural fluid is a lymphocytic however, the protein concentration may be >3.0 g/dL if
(usually >80 percent) exudate and the patient is PPD the initial pleural injury is not completely resolved. The
(purified protein derivative) positive, the patient should be pleural fluid from yellow nail syndrome is a lymphocyte-
treated for tuberculosis. If the patient has a negative PPD predominant (≥80 percent) exudate and may be associated
skin test, it should be repeated in 6–8 weeks; if positive, with slow-growing yellow nails and lymphedema or may
the history, physical examination, chest radi- (eds). Drug-induced and iatrogenic lung disease. London: Hodder
ograph and laboratory tests. Arnold; 2009, in press.
12. Hopkins HU. Principles and methods of physical diagnosis, 3rd ed.
● On chest radiograph, finding a solitary pleural Philadelphia: W.B. Saunders, 1965.
effusion, massive effusion, bilateral effusions or ◆13. Gopi A, Madhavan SM, Sharma SK, Sahn SA. Diagnosis and treatment
associated interstitial disease or nodules narrows of tuberculous pleural effusion in 2006. Chest 2007; 131: 874–9.
the pre-thoracentesis diagnosis. 14. Alptekin F. An epidemic of pleurisy with effusion in Bitlis, Turkey: a
● Finding a transudative effusion limits the differ- study of 559 cases. US Arm Forces Med J 1958; 9: 1–11.
●15. Walker WC, Wright V. Pulmonary lesions in rheumatoid arthritis.
ential diagnosis, which is easily discernible from Medicine 1968; 47: 501–19.
the clinical presentation. ●16. Bynum LJ, Wilson JE III. Radiographic features of pleural effusions
in pulmonary embolism. Am Rev Respir Dis 1978; 117: 829–34.
References 207
17. Hiller E, Rosenow E, Olsen A. Pulmonary manifestations of the 43. Maguire R, Fauci AS, Doppman JL, et al. Unusual radiographic
yellow nail syndrome. Chest 1972; 61: 452–8. features of Wegener’s granulomatosis. AJR Am J Roentgenol 1978;
18. Gottehrer A, Roa J, Stanford GG, et al. Hypothyroidism and pleural 130: 233–8.
effusions. Chest 1990; 98: 1130–2. 44. Gumbs RV, McCauley DI. Hilar and mediastinal adenopathy in
●19. Berger HW, Rammohan G, Neff MS, Buhain WJ. Uremic pleural septic pulmonary embolic disease. Radiology 1982; 142: 313–15.
effusion. A study in 14 patients on chronic dialysis. Ann Intern 45. Rubin SA. Radiographic spectrum of pleuropulmonary tularemia.
Med 1975; 82: 362–4. AJR Am J Roentgenol 1978; 131: 277–81.
20. Reeder MM, Felson B. Gamuts in radiology. Cincinnati: Audiovisual ●46. Collins TR, Sahn SA. Thoracentesis: complications, patient
Radiology, 1975. experience and diagnostic value. Chest 1987; 91: 817–22.
●21. Tomasilli G, Gamsu G, Stulbarg MS. Constrictive pericarditis 47. Sahn SA. Diagnostic value of pleural fluid analysis. Sem Respir Crit
presenting as pleural effusion of unknown origin. Arch Intern Med Care Med 1995; 16: 269–78.
1989; 149: 201–3. ●48. Good JT, Taryle DA, Maulitz RM, et al. The diagnostic value of
22. Maher GG, Berger HW. Massive pleural effusions: malignant and pleural fluid pH. Chest 1980; 78: 55–59.
non-malignant causes in 46 patients. Am Rev Respir Dis 1972; ◆49. Sahn SA. Pleural fluid pH in the normal state and in diseases
105: 458–60. affecting the pleural space. In: Chritein J, Bignon J, Hirsch A (eds).
23. Lieberson M. Diagnostic significance of the mediastinal profile in The pleura in health and disease. New York: Marcel Dekker, 1985:
massive unilateral pleural effusions. Am Rev Respir Dis 1963; 88: 253–66.
176–80. 50. Sahn SA, Collins DD. Pleural fluid acidosis and diagphragmatic
24. Heller RM, Janower ML, Weber AL. The radiological manifestations hernia. Ann Intern Med 1982; 96: 380–1.
of malignant pleural mesothelioma. Am J Roentgenol 1971; 108: 51. Miller KS, Wooten S, Sahn SA. Urinothorax: a cause of a low pH
53–9. transudate. Am J Med 1988; 85: 448–9.
●25. Lieberman FL, Hidemura R, Peters RL, Reynolds TB. Pathogenesis in ◆52. Sahn SA. Pathogenesis and clinical features of diseases associated
treatment of hydrothorax complicating cirrhosis with ascites. Ann with a low pleural fluid glucose. In: Chretein J, Bignon J, Hirsch A
Intern Med 1966; 64: 341–51. (eds). The pleura in health and disease. New York: Marcel Dekker,
26. Cavina C, Vichi G. Radiological aspects of pleural effusions in 1985: 267–85.
medical nephropathy in children. Ann Radiol Diagn 1958; 31: 53. Peterman TA, Speicher CE. Evaluating pleural effusions. A two-
163–202. stage laboratory approach. J Am Med Assoc 1984; 252: 1051–3.
●27. Stark DD, Shanes JG, Baron RL, Roach DD. Biochemical features of ●54. Light RW, MacGregor I, Luchsinger PC, Ball WC Jr. Pleural effusion:
urinothorax. Arch Intern Med 1982; 42: 1505–11. the diagnostic separation of transudates and exudates. Ann Intern
28. Bargman JM. Complications of peritoneal dialysis related to Med 1972; 77: 507–13.
increased intra-abdominal pressure. Kidney Intl 1993; 43: S75–80. ●55. Heffner JE, Brown LK, Barbieri C. Diagnostic value of tests that
29. Kaye MD. Pleuropulmonary complications of pancreatitis. Thorax discriminate between exudative and transudative pleural
1968; 23: 297–306. effusions. Chest 1997; 111: 970–9.
30. Anderson WJ, Skinner DB, Zuidema GD, et al. A chronic pancreatic 56. Joseph J, Badrinath P, Basran G, Sahn SA. Is the pleural fluid
pleural effusion. Surg Gynecol Obstet 1973; 137: 827–30. transudate or exudate? A revisit of the diagnostic criteria. Thorax
31. Majzlin G, Stevens FL. Meigs’ syndrome: case report and review of 2001; 56: 867–70.
literature. J Int Coll Surg 1964; 42: 623–30. ●57. Cohen M, Sahn SA. Resolution of pleural effusions. Chest 2001;
32. Press OW, Press NO, Kaufman SD. Evaluation and management of 119: 1547–62.
chylous ascites. Ann Intern Med 1982; 96: 358–64. 58. Hillerdal G, Ozesmi M. Benign asbestos pleural effusion: 73
33. De Cosse JJ, Poulin TL, Fox PS, Condon RE. Subphrenic abscess. exudates in 60 patients. Eur J Respir Dis 1987; 71: 113–21.
Surg Gynecol Obstet 1974; 138: 841–6. 59. Mattson SB. Monosymptomatic exudative pleurisy in persons
34. Rubin RH, Swartz MN, Malt R. Hepatic abscess: changes in exposed to asbestos dust. Scand J Respir Dis 1997; 56: 263–72.
clinical, biologic, and therapeutic aspects. Am J Med 1974; 57: 60. Robinson BWS, Musk AW. Benign asbestos pleural effusion:
601–10. diagnosis and course. Thorax 1981; 36: 896–900.
35. Chun CH, Raff MF, Contreras L, et al. Splenic abscess. Medicine ●61. Roper WH, Waring JJ. Primary serofibrinous pleural effusion in
1980; 59: 50–64. military personnel. Am Rev Tuberc 1955; 71: 616–35.
36. Koehler PR, Jones R. Association of left-sided pleural effusions in ●62. Nordkild P, Kromann-Andersen H, Struve-Christsen E. Yellow nail
splenic hematomas. AJR Am J Roentgenol 1980; 135: 851–3. syndrome – the triad of yellow nails, lymph edema, and pleural
●37. Rabin CB, Blackman NS. Bilateral plerual effusions. Its significance effusions. Acta Med Scand 1986; 219: 221–27.
in association with a heart of normal size. J Mt Sinai Hosp 1957; 63. Huggins JT, Sahn SA, Heidecker J, et al. Characteristics of trapped
24: 45–53. lung: Pleural fluid analysis, manometry, and air-contrast CT. Chest
38. Janower ML, Blennerhassett JB. Lymphangetic spread of 2007; 131: 206–13.
metastatic cancer to the lung: a radiologic–pathologic 64. Ferrer J. Pleural tuberculosis. Eur Respir J 1997; 10: 942–7.
classification. Radiology 1971; 101: 267–73. 65. Shitrit D, Zingerman B, Shitrit A, et al. Diagnostic value of CYFRA-
●39. Taylor JR, Ryu J, Colby TV, Raffin TA. Lymphangioleiomyomatosis. 21, CEA, CA 19-9, CA 15-3 and CA 125 assays in pleural effusions:
Clinical course in 32 patients. N Engl J Med 1990; 323: 1254–60. Analysis of 116 cases and review of the literature. Oncologist
40. Fine NL, Smith LR, Sheedy PF. Frequency of pleural effusions in 2007; 10: 501–7.
mycoplasma and viral pneumonias. N Engl J Med 1970; 283: ◆66. Loddenkemper R. Thoracoscopy – State of the art. Eur Respir J
790–3. 1998; 11: 213–21.
41. Huggins JT, Doelken P, Sahn SA, et al. Pleural effusions in a series 67. Kataria YP, Khurshid I. Adenosine deaminase in the diagnosis of
of 181 outpatients with sarcoidosis. Chest 2006; 129: 1599–604. tuberculous pleural effusion. Chest 2001; 120: 334–6.
42. Horowitz ML, Schiff M, Samuels J, et al. Pneumocystis carinii
pleural effusion. Pathogenesis and pleural fluid analysis. Am Rev
Respir Dis 1993; 148: 232–4.
This page intentionally left blank
17
Pleural fluid analysis
Because pleural effusions can be caused by diseases in the of appropriate watchful waiting is in a patient with a viral
chest, organ dysfunction or infections below the syndrome with pleuritic chest pain and a small pleural
diaphragm, drugs1 and systemic disease, it is not surprising effusion that can only be demonstrated clearly on a lateral
that there is an estimated 1.3 million new cases of pleural decubitus radiograph or ultrasonography. Observation is
effusions annually in the USA.2 Approximately two-thirds also warranted in any clinically stable patient with a very
of these effusions result from congestive heart failure small pleural effusion; if the patient’s clinical condition
(CHF), pneumonia, malignancy and pulmonary progresses, the effusion usually will increase in volume and
embolism. thoracentesis can be performed safely.
Clinicians may first suspect a pleural effusion on phys-
ical examination or become aware of its presence on a
standard chest radiograph. If a standard radiograph is APPEARANCE
only suggestive of a pleural effusion, a lateral decubitus
film or pleural ultrasound may be confirmatory. Once the A diagnosis can be established at the bedside by visual
presence of a pleural effusion is established, in most examination of the pleural fluid as it is aspirated from the
instances a diagnostic thoracentesis should be performed pleural space. The color, character and odor of the fluid
to assess the characteristics and cause of the pleural fluid. may be either diagnostic or helpful in diagnosis (Table
If the fluid is free flowing and at least of moderate size, the 17.1) A clear, straw-colored fluid suggests a transudate but
physical examination can safely guide thoracentesis. If also may be seen in pauci-cellular exudates. A serosan-
uncertainty exists, ultrasound-guided thoracentesis guinous appearance typically signifies a pleural fluid
should be carried out, which will increase the safety of the hematocrit <1 percent and is diagnostically unhelpful.
procedure.3 However, a grossly bloody effusion narrows the differen-
Not all patients with pleural effusions require a diag- tial diagnosis to malignancy, benign asbestos pleural effu-
nostic thoracentesis. For example, if the patient has the sion (BAPE), post-cardiac injury syndrome (PCIS),
classic presentation of CHF, appropriate treatment can be pulmonary infarction and trauma. A hemothorax, which is
instituted and the patient’s course closely observed. most commonly due to trauma,4 may also occur with inva-
However, if the patient presents with atypical features for sive procedures,5 metastatic disease to the pleura, antico-
CHF, such as fever and pleuritic chest pain, or unusual agulation in the setting of a pulmonary infarction6 and
radiographic findings, such as a unilateral effusion, bilat- catamenial hemothorax.7 To diagnose a hemothorax, the
eral effusions of disparate size, left effusion greater than pleural fluid hematocrit should be compared with the
right effusion, or a normal heart size, diagnostic thoracen- blood hematocrit; if the pleural fluid hematocrit is at least
tesis should be performed without delay. Another example 50 percent of the peripheral blood hematocrit, a
210 Pleural fluid analysis
Suggested diagnosis
Color of fluid
Pale yellow (straw) Transudate, pauci-cellular exudate
Red (bloody)
Hematocrit <5% Malignancy, BAPE, PCIS, pulmonary infarction
Hematocrit PF/S ≥0.5 Trauma
White (milky) Chylothorax or cholesterol effusion
Brown Long-standing bloody effusion; rupture of amebic
liver abscess into pleural space
Black Spores of Aspergillus niger
Yellow-green Rheumatoid pleurisy
Color of enteral tube
Feeding or central
Venous line infusate Feeding tube has entered pleural space; extravascular
catheter migration into mediastinum/pleural space
Character of fluid
Pus Empyema
Viscous Mesothelioma
Debris Rheumatoid pleurisy
Turbid Inflammatory exudate or lipid effusion
Anchovy paste Amebic liver abscess rupture
Odor of fluid
Putrid Anaerobic empyema
Ammonia Urinothorax
BAPE, benign asbestos pleural effusion; PCIS, post-cardiac injury syndrome; PF, pleural fluid; S, serum.
hemothorax is defined and chest tube drainage should be described as a thick, viscous, yellow-white, opaque fluid. If
performed without delay. the pus has a putrid odor, anaerobic organisms are usually
Withdrawing a white or milky fluid from the pleural present. When the pleural fluid appears to contain debris,
space diagnoses either a chylothorax or a cholesterol effu- rheumatoid pleurisy with exfoliation of portions of
sion; occasionally an empyema may simulate this appear- rheumatoid nodules from the visceral pleural surface into
ance. Centrifugation of the fluid will separate a lipid the pleural space is likely causative.13 Anchovy-paste
effusion from an empyema; in the lipid effusion, the appearing pleural fluid is virtually diagnostic of an amebic
supernatant will remain white while with empyema sus- liver abscess that has ruptured into the pleural space.
pended cells settle and the supernatant clears. When pleural fluid smells of ammonia, the diagnosis of
A pleural effusion that appears brown usually repre- urinothorax, which is caused by obstructive uropathy, is
sents a long-standing bloody effusion, while brownish established.14
fluid that is viscous in nature may represent an amebic
liver abscess that has ruptured into the pleural space;8 the
fluid of amebiasis represents a combination of blood, small
pieces of liver parenchyma and cytolysed liver tissue. Black EXUDATIVE VERSUS TRANSUDATIVE
pleural fluid suggests Aspergillus niger infection,9 while a
yellow-green tint has been observed in some patients with After gross inspection of pleural fluid obtained by thora-
rheumatoid pleural effusion.10 If an enteral feeding tube centesis, classification of a pleural effusion as a transudate
has entered the pleural space, the pleural fluid color will or exudate by chemical testing represents the initial step in
mirror that of the feeding solution.11 Likewise, if a central determining the cause of an effusion.15 A broad array of
venous catheter has migrated out of the vasculature and underlying conditions cause exudative effusions (Table
into the mediastinum, the pleural fluid will be similar in 17.2), while a limited number of disorders cause transu-
color and composition to the infusate (white if lipid is dates (Table 17.3). It is important to categorize effusions
infused).12 accurately because exudates and transudates present dif-
The character of the fluid will also suggest a diagnosis. ferent diagnostic and therapeutic implications. The pres-
If pus is aspirated, an empyema is established. Pus is best ence of an exudative effusion, for instance, may warrant
Exudative versus transudative 211
Causes
Diagnosis Comment
Congestive heart failure Acute diuresis can increase pleural fluid protein and LDH concentrations
Cirrhosis Uncommon without clinical ascites
Nephrotic syndrome Typically small and bilateral; unilateral, larger effusion may be due to pulmonary embolism
Peritoneal dialysis Large right effusion may develop within 48 hours of initiating dialysis
Hypoalbuminemia Edema fluid rarely isolated to pleural space; small bilateral effusions
Urinothorax Unilateral effusion caused by ipsilateral obstructive uropathy
Atelectasis Small effusion caused by increased intrapleural negative pressure; common in ICU patients
Constrictive pericarditis Bilateral effusions with normal heart size
Trapped lung Unilateral effusion from imbalance in hydrostatic pressures from a remote inflammatory process
Superior vena caval obstruction Due to acute systemic venous hypertension or acute obstruction of lymphatics
Duropleural fistula Cerebrospinal fluid in pleural space; β2-transferrin diagnostic
ICU, intensive care unit; LDH, lactate dehydrogenase.
pleural biopsy or pleural space drainage because of the Pleural fluid tests that discriminate exudates from tran-
high incidence of underlying malignant and infectious sudates measure the concentration of large molecular
conditions. In contrast, most patients with transudates weight constituents in pleural fluid (Table 17.3). Because
have clinically apparent systemic disorders as the cause of exudative effusions typically result from increased perme-
their effusions, which frequently resolve with treatment of ability and passage between mesothelial cells, these large
these conditions, such as with diuretic therapy for heart molecules can enter the pleural space or undergo release
failure. from intrapleural inflammatory or neoplastic cells. In con-
212 Pleural fluid analysis
trast, transudative effusions result from hydrostatic mech- centration > two-thirds the laboratory’s upper limit of
anisms and have lower concentrations of large molecular normal for serum. Aggregate data from a meta-analysis of
weight constituents because intact pleural membranes act multiple studies demonstrates that Light’s criteria has a
as a diffusion barrier to these compounds. sensitivity of 98 percent and specificity of 74 percent in
Various pleural fluid testing strategies reported in the identifying exudative effusions.19
literature recommend either a single pleural fluid test Other tests have been reported to have good discrimi-
result or multiple tests combined in diagnostic rules to nating properties for identifying exudative versus tran-
establish the presence of an exudate. The multi-test strate- sudative effusions (Table 17.4). These tests include pleural
gies combine test results in ‘or’ rules, which means that fluid cholesterol,20–25 albumin,26 protein27,28 and adenosine
any one of multiple test results can diagnose an exudative deaminase 29 used as single tests or combined with serum
effusion if the single result exceeds a certain limit. As with results to calculate a ratio or gradient. Although multiple
any diagnostic test, combining multiple tests in an ‘or- reports have compared these single tests with the Light’s
rule,’ increases sensitivity with the inevitable consequence criteria multi-test strategy to determine their relative diag-
of decreasing specificity.16,17 Single test results, therefore, nostic performance, the inevitable effects of test combina-
should be used when the goal is to increase specificity and tions to increase sensitivity and decrease specificity have
thereby ‘rule in’ an exudative effusion. Multiple tests com- been largely overlooked. As would be expected, most of
bined with in ‘or-rules’ increase sensitivity and help clini- these studies with Light’s criteria report a higher sensitiv-
cians to ‘rule out’ an exudative effusion if the test result is ity and lower specificity compared with each of the indi-
negative for an exudate. A positive result of a multiple test vidual test strategies.
strategy, however, has a decreased positive predictive value A meta-analysis was used to examine the comparative
for identifying exudative effusions. In most circumstances, diagnostic accuracies of the available individual tests and
clinicians would favor an initial multi-test screening strat- multi-test strategies commonly used in clinical practice for
egy with a high sensitivity for evaluating patients with new discriminating between exudates and transudates.19 This
onset pleural effusions because of the importance of not analysis reported that all of the individual pleural fluid
missing an exudative effusion, which carries important tests, except for pleural fluid bilirubin, had similar diag-
prognostic considerations. nostic operating characteristics (Table 17.5). Other inves-
Light’s criteria comprise a three-test strategy that tigators have recently confirmed the similar performance
screens patients for exudates by measuring pleural fluid of individual tests commonly used in clinical practice.28,30
and concurrent serum concentrations of protein and This result would be expected because the evaluated
lactate dehydrogenase (LDH). Test results are combined pleural tests share a common approach for detecting exu-
into three criteria: (1) pleural fluid to serum protein ratio; dates, which is the measurement of large molecular weight
(2) pleural fluid to serum LDH ratio; and (3) pleural fluid pleural fluid constituents.
LDH concentration relative to the laboratory’s upper limit Because all of the individual pleural fluid tests, except
of normal for serum LDH.18 An exudative effusion is for bilirubin, had similar operating characteristics in the
defined by the presence of any one of the following crite- meta-analysis by Heffner and colleagues,19 their combina-
ria: a pleural fluid to serum protein ratio >0.5, pleural fluid tion in various multi-test combination ‘or-rules’ had
to serum LDH ratio >0.6 or a pleural fluid LDH con- similar overall diagnostic accuracy compared with Light’s
Table 17.4 Cut-off points derived from a meta-analysis of pooled data and cut-off points commonly recommended by individual studies
Pleural fluid test Meta-analysis ROC cutoff point5 Previously reported cutoff points
Pleural fluid test Sensitivity, % (95% CI) Specificity, % (95% CI) AUC (95% CI)
criteria. As expected, sensitivity increased and specificity respectively, compared with 97.3 and 80.3 percent with the
decreased compared with the individual test performance two-test ‘Abbreviated Light’s criteria’ that contains pleural
as these tests are combined into two-test and three-test fluid to serum protein and pleural fluid to serum LDH.19
strategies. The abbreviated Light’s criteria rule has been confirmed to
Advantages exist for using multi-test strategies other have similar diagnostic accuracy compared with the three-
than Light’s criteria, which requires concurrent serum criteria Light’s criteria in independent data sets.31,32
samples for LDH and protein assay and drives up the cost In evaluating patients for exudative pleural effusions,
and discomfort of pleural fluid analysis. The combination clinicians should recognize that considerable uncertainty
diagnostic rule of pleural fluid LDH (> two-thirds the exists in categorization when pleural fluid tests return
upper limits of normal for serum LDH) and pleural fluid results near cut-off points. For example, pleural fluid from
cholesterol (>45 mg/dL) and the three-test combination of patients with CHF treated with diuretic therapy has an
pleural fluid protein (>2.9 mg/dL), pleural fluid LDH increased incidence of misclassification as exudate when
(> two-thirds the upper limits of normal for serum LDH), evaluated by the three-test Light’s criteria.33 These misclas-
and pleural fluid cholesterol (>45 mg/dL) have a sensitiv- sifications represent false-positives for exudative effusions,
ity and specificity equal to the three-test Light’s criteria.19 and most such patients have borderline pleural fluid test
When combining tests into multi-test combinations, a results. Light’s criteria has an overall diagnostic accuracy
basic principle directs that each test used is independently greater than 90 percent, but its accuracy decreases below
of the other and does not measure the same patient attrib- 70–80 percent when any one of its three criteria returns a
ute. Tests that depend on a common patient attribute borderline test result (Figure 17.1).34
would be expected to correlate highly with each other and Evaluation of an additional single pleural fluid test in
degrade the performance of the multi-test strategy. Note certain settings may improve diagnostic accuracy when an
that Light’s criteria include the pleural fluid to serum LDH initial screening multi-test strategy, such as Light’s criteria,
ratio and the absolute value of pleural fluid LDH relative returns a borderline result. In the case of patients with
to laboratory normal values; both of these criteria include CHF undergoing diuresis, one study reported the serum to
the same factor of pleural fluid LDH. Because these two pleural fluid albumin gradient >1.2 g/dL correctly identi-
LDH criteria were found in the metanalysis by Heffner fied 96 percent of transudates among patients incorrectly
and coworkers19 to be highly correlated (Pearson product- classified as having exudates by Light’s criteria.31 The
moment correlation 0.84), one or the other can be omitted overall diagnostic accuracy of the albumin gradient for
from Light’s criteria without significantly altering diagnos- patients with and without heart failure in this study was
tic performance. The sensitivity and specificity of the only 75 percent. Another study demonstrated that neither
three-component Light’s criteria is 97.9 and 74.3 percent, the serum to pleural fluid albumin gradient nor the serum
214 Pleural fluid analysis
40
malignancy,18 parapneumonic effusions18 and Pneumo-
cystis carinii pneumonia should be considered.49 A pleural
35 fluid LDH concentration greater than three times the
upper limits of normal for serum LDH is typically seen
30 only in a complicated parapneumonic effusion or
empyema,50,51 rheumatoid pleurisy52 or pleural parago-
nimiasis53; it is less often observed with malignancy and
Likelihood ratio
25
rarely with a tuberculous pleural effusion.18
20
15 NUCLEATED CELLS
Disease Comment
Tuberculous effusion most common cause of lymphocyte predominant exudate; usually 90–95% lymphocytes
Chylothorax 400–6800 lymphocytes/μL; lymphoma most common cause
Lymphoma Often 100% of nucleated cells are lymphocytes; diagnostic yield on cytology or pleural biopsy
higher with non-Hodgkin lymphoma
Yellow nail syndrome A cause of a persistent effusion
Rheumatoid pleurisy (chronic) Usually associated with lung entrapment
Sarcoidosis Usually >90% lymphocytes; prevalence <2%
Acute lung rejection New or increased effusion 2–6 weeks after transplant
CABG surgery Occurs >2 months following surgery; not associated with unexpandable lung
Uremic pleuritis Renal failure present for 1–2 years; most effusion resolve with continued dialysis in 4–6 weeks
CABG, coronary artery bypass graft.
the pretest diagnosis is a transudate. Transudative effu- following hemothorax eosinophils do not appear in the
sions are mononuclear cell predominant, a combination of pleural space for 7–14 days.77 Furthermore, PFE is associ-
lymphocytes, macrophages and mesothelial cells. ated with peripheral blood eosinophilia following trauma
When the lymphocyte population is 80 percent or that does not clear until the pleural fluid completely
greater of the total nucleated cells, the differential resolves.78,79 Approximately 30 percent of individuals with
diagnosis of the exudative effusion is narrowed to the BAPE have been noted to have PFE with eosinophil per-
following diagnoses (Table 17.6).61 The most common of centages sometimes reaching 50 percent.80,81 These
these diagnoses is a tuberculous effusion.62 Lymphoma,62 asbestos related effusions are often hemorrhagic. Other
yellow nail syndrome,63 chronic rheumatoid pleurisy,56,64 causes of PFE, such as pulmonary embolism with infarc-
uremic pleuritis,65 sarcoidosis,66,67 chylothorax,68 acute tion and carcinoma, are also associated with pleural space
lung rejection69 and post-coronary artery by-pass surgery hemorrhage. Fungal disease, particularly coccidioidomy-
(CABG surgery)70 are also causes of >80 percent of cosis and histoplasmosis, drug-induced pleurisy, Churg–
lymphocytes in pleural fluid. All of the aforementioned Strauss syndrome, lymphoma, particularly Hodgkin
diagnoses can occur with a lymphocyte population of <80
percent; however, a lymphocyte population of <50 percent
is rarely found. In contrast to lymphoma, only about 60
percent of patients with metastatic carcinoma to the pleura Table 17.7 Diseases associated with pleural fluid eosinophiliaa
have a majority of lymphocytes (usually 50–75 percent of Disease Comment
total nucleated cells).62 An undiagnosed, lymphocyte-
predominant, exudative effusion is the most appropriate
indication for percutaneous pleural biopsy and the most Pneumothorax Effusion in 10–20%; tissue
sensitive diagnostic procedure in tuberculous pleural eosinophilia and PFE occurs early
effusion with the exception of thoracoscopy.71,72 and is a common finding
Lymphoma, carcinoma, sarcoidosis and occasionally Hemothorax PFE occurs 1–2 weeks following a
rheumatoid pleurisy may be diagnosed by percutaneous hemothorax
pleural biopsy. Benign asbestos 30% incidence of PFE; up to 50%
Pleural fluid eosinophilia (PFE) is defined as a pleural eosinophils/total nucleated cells
fluid eosinophil count ≥10 percent of the total nucleated Pulmonary embolism Associated with infarction and
cell count. It appears that eosinophils, produced in the pleural space hemorrhage
bone marrow, are attracted to the pleural space by the Parasitic disease Paragonimiasis, hydatid disease,
chemotactic factor, IL-5.73 Causes of PFE are shown in amebiasis, ascariasis
Table 17.7. Pneumothorax and hemothorax are the most Fungal disease Histoplasmosis, coccidioidomycosis
common causes of PFE. It had been thought that the Drug-induced Dantrolene, bromocriptine,
finding of PFE virtually excluded the diagnosis of malig- nitrofurantoin, valproic acid
nancy. However, recent studies have shown that the preva- Lymphoma Hodgkin disease
lence of malignancy is similar in both eosinophilic and Carcinoma 5–8% with PFE
non-eosinophilic pleural effusions.74,75 Eosinophilic pleu- Churg–Strauss Syndrome High pleural fluid eosinophil counts;
ritis is a common early finding in patients who require associated with blood eosinophilia
thoracotomy or thoracoscopy for spontaneous pneumoth- Tuberculous pleurisy Rare
orax. In contrast to the rapid movement of eosinophils Sarcoid pleurisy Rare
into pleural tissue and pleural fluid in pneumothorax,76 aPFE, pleural fluid eosinophils/total nucleated pleural fluid cells ≥10%.
pH 217
lymphoma, and parasitic diseases are also associated with fluid, lymphoma or chronic lymphatic leukemia should be
PFE.61,74,75 A tuberculous effusion has rarely been reported suspected.85
as a cause of PFE; it is unclear from the literature whether
a previous thoracentesis resulting in air or blood entering
the pleural space was the cause of the PFE. pH
Pleural fluid macrophages, which originate from blood
monocytes, are not diagnostic.77 Mesothelial cells are exfo- Pathogenesis
liated into normal pleural fluid in small numbers.77
Although common in transudative effusions and some A limited number of pleural effusions are associated with a
exudates, mesothelial cells are rarely found in a tubercu- low pleural fluid pH (<7.30) (Table 17.8).86, 87 Pleural fluid
lous effusion, most likely due to the extensive pleural pH should only be measured with a blood gas analyzer.
involvement by the hypersensitivity reaction to tuberculin Pleural fluid pH decreases as a result of high metabolic
protein, which inhibits mesothelial shedding.77,82 A activity of intrapleural cellular constituents or by an
paucity of mesothelial cells is the typical finding in other abnormal pleural membrane that blocks the efflux of
inflammatory processes, such as empyema, chemical pleu- protons and organic acids from the pleural space into the
rodesis, rheumatoid pleuritis and chronic malignant effu- circulation.87 Urinothorax is a rare cause of a low pleural
sions.77 fluid pH in patients with acidic urine and is the only tran-
A large number of plasma cells in pleural fluid suggest sudate associated with pleural fluid acidosis.88 The
pleural involvement with multiple myeloma,77 while a extravasated urine from the capsule of the kidney moves
small number of plasma cells is non-diagnostic and has retroperitoneally across the diaphragm into the ipsilateral
been observed in several non-malignant diseases.77 A few pleural space.
basophils, occasionally found in pleural fluid, are of no The presence of a low pleural fluid pH provides diag-
clinical significance. However, when basophils represent nostic and prognostic information in various clinical set-
>10 percent of the nucleated cells, leukemic involvement tings. A low pH may also confound pleural fluid analysis
of the pleura is likely.77 by suggesting an alternative diagnosis, such as pleural
The separation of T and B lymphocytes in lymphocyte space infection, when it occurs in the setting of a less
predominant effusions has not been of diagnostic value. common condition associated with a low pH effusion,
The T lymphocyte represents about 70 percent of the such as rheumatoid pleurisy.
pleural fluid lymphocyte population.83,84 However, if a In the only study of acid–base characteristics of normal
high percentage of B lymphocytes are found in pleural human pleural fluid, the pH was measured at 7.64, 0.23 pH
Table 17.8 Diagnoses associated with pleural fluid acidosis (pH <7.30)
Parapneumonic effusion
Uncomplicated 0–5 7.45–7.20 Nonpurulent, nonloculated fluid with negative
bacteriology; resolves with antibiotics only
Complicated or empyema ~100 7.29–5.00 Requires pleural space drainage for resolution
Esophageal rupture ~100 6.80–5.00 pH 6.00 and high salivary amylase
Rheumatoid pleurisy (chronic) ~100 7.15–6.80 Associated with glucose < 30 mg/dL and LDH >1000 IU/L
Malignant effusion 30–40 7.50–6.90 pH <7.30: worse survival, increased yield on cytology and
pleural biopsy, poorer response to pleurodesis
Lupus pleuritis 15–20 7.40–6.85 Associated with low glucose; diagnosis by LE cells in PF
Tuberculous effusion 10–20 7.40–6.95 Associated with low glucose; when pH low, usually
between 7.29 and 7.10
Hemothorax <10 7.50–7.17 Occurs when PF hematocrit approaches blood hematocrit
and associated pleural injury
Pancreatic effusion Rare 7.50–7.28 Occurs with very high amylase and severe pleural injury
Pulmonary infarction Rare 7.52–7.29 Occurs with grossly bloody PF and extensive pleural injury
Diaphragmatic hernia with Single report 7.15 Acid products of dead bowel overwhelm efflux capacity
bowel infarction of pleural space
LDH, lactate dehydrogenase; LE, lupus erythematosus; PF, pleural fluid.
218 Pleural fluid analysis
units greater than the simultaneously measured blood pH.89 show impaired glucose transfer.95 Following 2 minutes of
We and others have measured an alkaline pH (>7.60) in hyperventilation, patients with low pH malignant effusions
normal animals.90,91 As the partial pressure of CO2 (PCO2) did not lower their pleural fluid PCO2, while those with
in pleural fluid was equivalent to arterial PCO2, a bicarbon- normal pH malignant effusions had a significant fall in
ate gradient between pleural fluid and blood explains the pleural fluid PCO2.95 We also demonstrated a block in
alkaline pH of normal pleural fluid.89,90 Human transuda- oxygen transfer from blood to pleural fluid in low pH
tive effusions have a pleural fluid pH ranging from 7.45 to malignant effusions. Following 100 percent oxygen admin-
7.55. The vast majority of exudative effusions have pH istration for 20 minutes, those with low pH malignant
values that range from 7.45 to 7.30. However, only a small effusions had no change in the partial pressure of O2 (PO2)
number of exudates are associated with pleural fluid acido- in pleural fluid, while those with normal pH malignant effu-
sis (pH <7.30) and are the result of a substantial sions had a significant increase in pleural fluid PO2.95
accumulation of hydrogen ions in the pleural space.86,87 Therefore, in patients with low pH malignant effusions,
The pleural fluid pH is determined by hydrogen ion the pleural fluid glucose is low because, in addition to the
production by pleural fluid and pleural tissue, efflux of metabolism of glucose from cells in pleural fluid and
hydrogen ions from the pleural space and the buffering pleural tissue, there is a block of transfer of glucose from
capacity of pleural fluid.87,92 Since the pleural fluid protein blood to the pleural space. Furthermore, the CO2 and
concentration of inflammatory pleural effusions lactate from glycolysis accumulate in the pleural space
approaches that of plasma, a poor buffering capacity prob- because of a relative efflux block, which is due to tumor
ably plays a minor role in determination of the pleural and associated fibrosis on the pleural surface. Similarly, the
fluid pH.92 Furthermore, since a low pleural fluid pH is oxygen tension of a malignant pleural effusion is typically
found with diagnoses as diverse as empyema, malignancy, low as oxygen is utilized by malignant cells in pleural fluid
lupus pleuritis and rheumatoid pleurisy, the pathogenesis and pleural tissue, and there is a block of oxygen transfer
of the low pleural fluid pH in each of these diseases prob- from blood to pleural fluid.
ably differs. We have shown that, after incubating human
pleural fluid anaerobically, the rate of change of pH in
most effusions is similar whether or not the pH is <7.30 or Parapneumonic effusions
>7.30.92 Thus, the acid generating capacity of pleural fluid
is not the sole determinant of pleural fluid pH. However, Some experts categorize parapneumonic effusions as
some low pH fluids generate a substantial amount of acid uncomplicated or complicated. Uncomplicated parapneu-
(empyema), while other low pH effusions have minimal monic effusions resolve with antibiotic therapy alone.
metabolic activity (chronic rheumatoid pleurisy).92 Complicated effusions require pleural fluid drainage to
Malignant pleural effusions have an acid generation capac- accelerate patient recovery and avoid progression to an
ity intermediate between empyema and rheumatoid empyema or lung entrapment. Because these categoriza-
pleural fluid.92 Therefore, in a low pH rheumatoid effu- tions require knowledge of a patient’s subsequent clinical
sion, where the fluid has minimal acid generating capacity, course, various strategies have been investigated to esti-
impaired hydrogen ion efflux must play a major role in mate the likely course of a parapneumonic effusion upon
determining the low pH of the effusion. In malignancy, initial presentation. Pleural fluid pH has been found to be
there appears to be a contribution from both acid genera- lower in patients who follow a complicated rather than an
tion of the fluid and pleural tissue and impaired hydrogen uncomplicated course.50,51,86,96–99
ion efflux from the pleural space. A meta-analysis examined the prognostic performance
In an experimental empyema model, we have shown of pleural fluid pH in patients with parapneumonic effu-
that pleural fluid pH decreases within hours following the sions.100 This study combined patient-level data from
injection of bacteria into pleural fluid.93 The decrease in seven primary investigations that measured pleural fluid
pH is associated with an increase in pleural fluid PCO2 and pH in samples from patients with non-purulent parapneu-
lactate concentration and a concomitant fall in pleural monic effusions.50,51,86,96–99 Receiver operating characteris-
fluid glucose. We subsequently demonstrated, in in vivo tics were used to determine the diagnostic accuracy of
and in vitro experiments, that both leukocyte/phagocytosis pleural fluid pH, which was reported as the area under the
and bacterial metabolism contributes to the low pH of receiver operating characteristic curve (AUC).101 The met-
empyema fluid.94 analysis found that pleural fluid pH was lower in pleural
Patients with low pleural fluid pH malignant effusions fluid samples from patients who followed a complicated
have a lower pleural fluid glucose concentration and oxygen rather than an uncomplicated course and that the AUC for
tension and higher PCO2 and lactate concentration than pH was 0.89.100 This value for AUC indicated that pleural
patients with normal pH malignant effusions.95 In glucose fluid pH had good discriminative properties for differenti-
transfer experiments, we demonstrated that patients with ating between complicated and uncomplicated parapneu-
low pH malignant effusions had a block in both transfer of monic effusions.
glucose from blood to pleural fluid and from pleural fluid Published primary studies recommend various cut-
to blood; those with normal pH malignant effusions did not points below which a complicated effusion can be
Glucose 219
defined;46,51,86,96–99 these range between pH 7.10 and pH fidence interval [CI], 31.1 to 46.8) for patients with pleural
7.30. The metanalysis of pleural fluid pH recommended a fluid pH values ≤7.28 compared with 62 percent (95
Bayesian approach for using pleural fluid pH by consider- percent CI, 55.7 to 67.4) patients with pleural fluid pH
ing the likelihood that the patient had a complicated values >7.28 (n = 268) (p < 0.0001).112 Unfortunately,
parapneumonic effusion and the risk to the patient of mis- patients with the lowest pH values still had a 65 percent
classifying the effusion.100 A high-risk patient might be probability of a successful pleurodesis. Similarly, pH cor-
identified by a clinical presentation that suggests a compli- responded with survival but patients with the lowest pH
cated course and a poor ability to tolerate misclassification values still had nearly a 45 percent likelihood of survival
of an effusion as uncomplicated. Such a patient might have for greater than 3 months, thereby, warranting efforts for
a large effusion, radiographic evidence of thickened pleural pleurodesis to manage symptoms. At present, pleural fluid
membranes, pleural loculations and a Gram-negative pH may have an adjunctive value in determining potential
pneumonia, all of which are associated with a complicated for benefit from pleurodesis, but it requires further exam-
course. The patient might also be elderly with poor host ination in clinical practice.
responses to infection, which would decrease the patient’s
ability to tolerate misclassification of the effusion as
uncomplicated and a delay in pleural fluid drainage. A GLUCOSE
pleural fluid pH <7.30 would warrant pleural fluid drainage
in this high-risk patient. Conversely, a young, relatively In the normal physiological state, blood and fluid glucose
well-appearing patient with pneumococcal pneumonia concentration should be equivalent because glucose is of
(good tolerance of a misclassification of the parapneumonic low molecular weight and should be transported from
effusion) who has a small, free-flowing effusion (low pretest blood to pleural fluid by simple diffusion across the
suspicion for a complicated course) would require drainage endothelial and mesothelial membranes. It is unlikely that
of the pleural space only if the pH were below 7.10. a carrier transport system for glucose exists between blood
The Bayesian approach to using pleural fluid pH and pleural fluid, as is postulated between blood and cere-
described above recognizes that no single pH value can brospinal fluid,113 because all transudative effusions and
accurately dichotomize patients into complicated and most exudative effusions have a pleural glucose that
uncomplicated categories. Pleural fluid pH, therefore, mirrors blood glucose. A small group of exudative pleural
serves as adjunctive information that should be combined effusions, the same as those with a low pleural fluid pH,
with the clinician’s general impression as to the utility of may present with a low pleural fluid glucose concentra-
pleural fluid drainage. This adjunctive role of pleural fluid tion.114 A low pleural fluid glucose concentration has been
pH was recommended by the recent American College of defined as <60 mg/dL or a pleural fluid to serum glucose
Chest Physicians consensus statement on the management ratio of <0.5 (Table 17.9). Post-pneumonic empyema and
of empyema.102 the anaerobic empyema associated with esophageal
rupture have low pleural fluid glucose concentrations that
tend to increase towards normal with treatment. In exper-
Malignant pleural effusions imental empyema, the glucose falls rapidly following pen-
etration of bacteria into the pleural space.93 Glucose
Pleural fluid pH has been examined for its clinical utility in concentration correlates inversely with pleural fluid lactate
selecting patients with malignant effusions to undergo and pCO2, both in vivo and in vitro, suggesting that the
pleurodesis.103–109 Low pleural fluid pH values have been accumulation of these glucose end-products is responsible
observed in patients who fail pleurodesis, and pleural fluid for the decrease of pleural fluid pH in empyema.93 Glucose
pH correlates with survival among patients with malignant utilization is increased by neutrophil phagocytosis and
effusions, with shorter survivals being associated with bacterial metabolism and exceeds the replacement of
lower pleural pH values.103–109 It has been suggested that a glucose from blood to pleural fluid.94 While increased
low pH may identify patients who should be spared pleu- glucose utilization is the primary mechanism for low
rodesis because of a low likelihood of benefit.110 glucose in empyema, an abnormal pleural membrane pro-
Two recent meta-analyses,111,112 however, collected duced by the rheumatoid state is the cause of low glucose
primary patient-level data from seven investigators in 11 concentration in rheumatoid pleurisy.115,116 A pleural fluid
data sources and found pleural fluid pH alone had insuffi- glucose concentration of zero is found almost exclusively
cient prognostic accuracy to be useful in clinical practice. in patients with empyema and rheumatoid pleurisy.114
The studies found a highly statistically significant direct Like rheumatoid effusions, the major mechanism of the
correlation of pH with both survival and failure of pleu- low glucose concentration in malignancy is an abnormal
rodesis, but pH by itself was a weak explanatory variable pleura, which is due to tumor infiltration that decreases
for these outcomes. The sensitivity and specificity of a glucose movement into the pleural space. In some cases,
pleural pH <7.28 for identifying pleurodesis failure were free cancer cells contribute to increased glucose utilization
56 and 78 percent, respectively.111 Kaplan–Meier estimates and an increase in the end-products of glucose metabo-
of the 3-month survival were 39 percent (95 percent con- lism, CO2 and lactate acid, which accumulate due to
220 Pleural fluid analysis
Table 17.9 Diagnoses associated with low pleural fluid glucose concentration (glucose <60 mg/dL or PF/S <0.5)
Rheumatoid pleurisy 85–90% 0–118 Glucose <30 mg/dL in 75%; PF triad of glucose <30 mg/dL,
pH 7.00 and LDH >1000 IU/L; may precede articular
manifestations by 3 years
Empyema 80–90% 0–145 Low glucose not as sensitive marker as low pH but
correlation of glucose and pH is strong
Esophageal rupture 40–50% 15–120 Characteristic pH of 6.00 and high pleural fluid amylase
(salivary)
Malignant effusion 30–40% 15–167 Low glucose concentration with chronic effusion in far-
advanced pleural malignancy
Lupus pleuritis 20–30% 32–160 Transient; associated with severe pleural inflammation; LE
cells diagnostic
Tuberculous effusion 20–30% 10–140 No correlation between low glucose and clinical course or
pleural bacteriology
PF/S, pleural fluid/serum; LDH, lactate dehydrogenase; LE, lupus erythematosus.
impaired efflux and cause pleural fluid acidosis.117 We concentration returns to a concentration similar to serum
have previously demonstrated that patients with carcino- glucose.
matous pleurisy have pleural fluid acid generation inter- In an experimental model of esophageal rupture, we
mediate between empyema fluids and rheumatoid demonstrated that pleural fluid glucose concentration
effusions.92 began to decrease within 2 hours following rupture and
The literature in the pre-chemotherapy era suggested decreased below 60 mg/dL by 12 hours.121 The lowest
that a low pleural fluid glucose was a common finding in a pleural fluid glucose concentrations were associated with
tuberculous pleural effusion. It is probable that many of the highest pleural fluid leukocyte counts. Furthermore,
these patients had tuberculous empyema or a markedly the pleural fluid glucose concentration correlated directly
abnormal pleural membrane responsible for this finding. with pleural fluid pH and inversely with pleural fluid PCO2
However, in the post-chemotherapy era, a low pleural and lactate. It appears that the low pleural fluid glucose
fluid glucose is found in only approximately 20 percent of concentration in esophageal rupture results from the same
patients with a tuberculous pleural effusion.46,118 In three mechanisms as in post-pneumonic empyema – increased
patients who had glucose transport studies, two showed no glucose metabolism by constituents in pleural fluid, leuko-
impairment of glucose movement from blood to pleural cytes and bacteria. It is doubtful that there is a marked
fluid and one, with a markedly thick pleura and encapsu- decrease in glucose transfer from blood to pleural fluid fol-
lated effusion, showed no change in pleural fluid glucose lowing esophageal rupture, as at least initially only the
concentration.119 It is likely that the cause of low pleural mediastinal pleura is affected. The lag in the decrease in
fluid glucose concentration in tuberculous effusion is pleural fluid glucose and pH following esophageal rupture
increased glycolysis by metabolically active cells in the probably reflects the time required for the pleural space to
pleural fluid or pleural membrane. In an acute tuberculous become contaminated with microorganisms and for
pleural effusion, impaired glucose transfer from blood to leukocytes to move into the pleural space. A high pleural
pleural fluid appears unlikely. fluid amylase concentration is the earliest marker of
The low glucose in lupus pleuritis tends to be transient. esophageal rupture and represents a direct communica-
All three patients in one study with lupus pleuritis and a tion between the oral cavity and the pleural space.121
low pleural fluid glucose ratio had low pleural fluid nucle-
ated cell counts, suggesting that pleural fluid cellular
metabolism is not a major contributor to the low glucose AMYLASE
concentration.120 Therefore, it is likely that active glycoly-
sis by an inflamed pleura, in addition to impaired glucose An increased pleural fluid amylase, defined as either a
transport across this abnormal membrane, play important value greater than the upper limits of normal of the serum
roles in the low glucose concentration in lupus pleuritis. or a pleural fluid amylase ratio >1.0, is found in pancreatic
When patients are treated with corticosteroids and the disease,118,122–124 esophageal rupture110,111,121,125,126 and
pleural fluid inflammation resolves, pleural fluid glucose malignancy.111,112,123,127,128 Rarely, an amylase-rich pleural
Immunological studies 221
effusion has been reported in pneumonia, ruptured pretest suspicion of pancreatic disease, esophageal rupture
ectopic pregnancy, hydronephrosis and cirrhosis.123 Both or malignancy.133
acute and chronic pancreatitis can cause an amylase-rich
pleural effusion. In acute pancreatitis, the effusion appears
to result from direct contact of the pancreatic enzymes TRIGLYCERIDES AND CHOLESTEROL
with the diaphragmatic pleura and movement of pancre-
atic fluid into the pleural space through diaphragmatic When the supernatant of a milky or whitish pleural effu-
defects.129,130 Pleural fluid amylase increases in concentra- sion remains opaque after centrifugation, excluding a large
tion because it is not cleared rapidly by the pleural lym- number of leukocytes, the differentiation between a
phatics, while amylase is cleared quickly from the blood by chylous (chylothorax) and a cholesterol effusion (chyli-
the kidney resulting in an increased pleural fluid to serum form effusion or pseudochylothorax) needs to be estab-
amylase ratio. In early pancreatitis, pleural fluid amylase lished. Therefore, the first test that should be ordered is a
may be normal but increases to a high level a few days fol- pleural fluid triglyceride concentration. A diagnosis of
lowing the onset of the pleural effusion.118 In chronic pan- chylothorax can be made presumptively if the triglyceride
creatitis, fluid moves via a fistulous track from a concentration exceeds 110 mg/dL;134 if the triglyceride
pseudocyst, either directly into the pleural space or into concentration is <50 mg/dL, it is highly unlikely that the
the mediastinum with eventual rupture of the mediastinal patient has a chylothorax. When the triglyceride concen-
parietal pleura.124 In chronic pancreatitis, pleural fluid tration is between 50 and 110 mg/dL, the presence of chy-
amylase is always elevated and may reach extremely high lomicrons should be determined; if chylomicrons are
levels of over 100 000 IU/L.127 Serum amylase may be ele- present, the diagnosis of a chylothorax is established.134
vated in chronic pancreatitis due to back-diffusion from Importantly, in many cases of chylothorax, the pleural
the pleural space or it may be normal.131 effusion will not appear milky. Blood can mask the milky
An increased pleural fluid amylase concentration appearance and, at times, the fluid is serous or turbid if the
occurs in 10 to 14 percent of patients with a malignant patient has not eaten recently.130,131,134,135 Lymphocytes are
pleural effusion.118,123,128,132 By isoenzyme analysis, the the primary cells in chyle with counts ranging from 400 to
amylase is virtually all salivary type.118,123,128 Adeno- 6800/μL;68 the percentage of lymphocytes usually exceeds
carcinoma of the lung is the most common malignancy 80 percent.68
associated with a salivary amylase-rich pleural effu- In the patient with a milky effusion and a low triglyc-
sion,123,128 followed by adenocarcinoma of the ovary.128 eride concentration, pleural fluid cholesterol should be
Lymphoma, leukemia and other types of lung cancer have measured and the sediment evaluated microscopically. On
also been reported with a salivary isoamylase-rich pleural observation, the pleural fluid may appear to have a satin-
effusion.123,128 When a salivary isoamylase-rich pleural like sheen, which is distinctive from any other pleural effu-
effusion is found in a patient who clinically does not have sion. Under the microscope, cholesterol crystals can be
esophageal rupture (status post-endoscopy or Boerhaave recognized as large, polyhydric crystals. When pleural fluid
syndrome), there is a high likelihood that the effusion is cholesterol exceeds 200 mg/dL, it most likely represents a
malignant and, most commonly, from adenocarcinoma of cholesterol pleural effusion.136 However, it should be rec-
the lung. While it has been documented that tumor tissue ognized that high cholesterol levels may also be found in
can produce a salivary-like isoamylase,127 this has rarely some chylous effusions.134 In addition, some cholesterol
been shown with mesothelioma. The amylase in pleural effusions have triglyceride levels of >250 mg/dL.137
esophageal rupture has its origin in the salivary glands in If there is clinical uncertainty about the diagnosis, the
the mouth, and the enzyme enters the pleural space presence of chylomicrons should be determined.
through a rent in the esophagus and mediastinum. If the Patients with chylothorax tend to have an acute or sub-
diagnosis is not established early, an anaerobic empyema acute onset of dyspnea with exertion and may have con-
will ensue. In an experimental model, we have shown that tralateral mediastinal shift on chest radiograph; patients
amylase is increased within 2 hours following the with a cholesterol pleural effusion have long-standing
esophageal tear.121 pleural effusions associated with lung entrapment and an
In contrast to patients with acute pancreatitis who typ- absence of mediastinal shift.137
ically have symptoms of abdominal pain, fever and chest
pain, patients with chronic pancreatic pleural effusions
present predominantly with dyspnea and without evidence IMMUNOLOGICAL STUDIES
of acute pancreatitis. The pleural fluid amylase level is
highest in chronic pancreatitis, moderately elevated in Approximately half of the patients with systemic lupus
acute pancreatitis, and lowest in amylase-rich effusions erythematosus will develop pleuritic chest pain or a pleural
secondary to malignancy.123 effusion during the course of their disease and, in 5 percent
The routine measurement of pleural fluid amylase does of patients, pleuritis may be the presenting manifesta-
not appear to be clinically indicated or cost effective. tion.138 Approximately 5 percent of patients, mostly men
Pleural fluid amylase should be measured only if there is a with active rheumatoid disease, will develop pleurisy.139,140
222 Pleural fluid analysis
In patients with these two diseases, immunology studies the number of specimens submitted (yields tend to
are appropriate. We previously reported that pleural fluid increase with additional specimens due to exfoliation of
anti-nuclear antibodies (ANA) titer ≥1:160 in conjunction fresher cells),151 the stage of pleural involvement (the more
with a pleural fluid to serum ANA ratio of ≥1 was sugges- advanced the stage, the higher the diagnostic yield), and,
tive of lupus pleuritis.120 We further showed that pleural lastly, the interest and expertise of the cytopathologist. At
fluid ANA titers were negative in a large number of times, the pleural effusion is unrelated to the malignant
patients with pleural effusion of other etiologies.120 In a process, such as CHF or non-obstructive pneumonia.
small number of patients with systemic lupus erythema-
tosus (SLE) and an effusion of another cause, the pleural
fluid to serum ANA ratio was <1.120 Others have also FLOW CYTOMETRY
found elevated titers of pleural fluid ANA in patients who
do not have lupus pleuritis, but when ANA titers exceed While flow cytometry should not be routinely used to dif-
1:320, lupus pleuritis was likely.141 However, while the ferentiate benign and malignant pleural effusions, it can be
ANA titer and pleural fluid to serum ratios are suggestive helpful in the diagnosis of lymphoma involving the pleura.
of lupus pleuritis, only finding LE cells in pleural fluid is Flow cytometry can specifically define lymphocyte surface
diagnostic. Although most clinical laboratories do not markers.152 Therefore, it can define clonality of a popula-
perform LE tests because there is better immunological tion of lymphocytes to determine whether the cells are
testing for SLE, a simple Wright stain of pleural fluid sedi- from T- or B-cell lineage. Therefore, flow cytometry is
ment that has remained at room temperature for several most helpful in patients with a lymphocyte predominant
hours will increase the likelihood of finding LE cells and pleural effusion when lymphoma is in the differential diag-
confirm the diagnosis of lupus pleuritis. nosis.
When rheumatoid pleurisy is considered, a rheumatoid
factor should be measured in pleural fluid. If the pleural
fluid rheumatoid factor titer is ≥1:320 and equal to or ADENOSINE DEAMINASE
greater than the serum rheumatoid factor, it is likely that
the effusion is caused by rheumatoid arthritis.142 Lower Adenosine deaminase (ADA) is an enzyme found in most
titers of pleural fluid rheumatoid factor are non- cells and is important in the degradation of purines. It is
diagnostic and have been documented in patients with required for lymphoid cell differentiation and is also
parapneumonic effusions and malignancy.143 Pleural fluid involved in monocyte–macrophage maturation. The assay
complement levels, whether CH50, C3 or C4, are typically for ADA is easily and rapidly performed, is of relatively low
are low in patients with lupus pleuritis and rheumatoid cost and has been used in areas of high tuberculosis preva-
pleurisy.142,144–146 However, levels may be low in other lence to aid in the diagnosis of tuberculosis.153,154 Pleural
exudates. CH50 levels below 10 U/mL144 or C4 levels below fluid/serum ADA ratios of >1 have been observed, not
10 × 10−5 U/g protein142 are rarely seen with exudates other only in tuberculous effusions, but in rheumatoid disease
than lupus pleuritis or rheumatoid pleurisy.142 and empyema, while other exudates have similar ADA
Nevertheless, a rheumatoid effusion has a diagnostic cyto- levels in pleural fluid and serum.151,155 Furthermore,
logical picture of a background of degenerative cells and patients with tuberculous effusions, rheumatoid pleurisy,
large, elongated tadpole-shaped cells representing exfolia- empyema and malignancy have higher ADA activity in
tion from visceral pleural rheumatoid nodules.13 pleural fluid than other exudates and CHF.156 The diag-
nostic cut-off points for ADA have been reported from 40
to 60 U/L.157–161 Selecting a cut-off point of 40 U/L will
CYTOLOGICAL EXAMINATION increase the sensitivity of ADA but decrease its specificity,
while choosing a cut-off point of 60 U/L will increase
Pleural fluid cytology has a widely variable diagnostic specificity but decrease sensitivity.
yield, ranging from 40 to 90 percent of patients with Some authors have suggested that determination of
pleural malignancy.143,144,147,148 There are several reasons ADA isoenzymes will enhance its diagnostic utility. In a
for this variability. One important consideration is that the tuberculous effusion, ADA-2 is the predominant isoform
effusion may be paramalignant, which is defined as a accounting for over 80 percent of ADA activity, whereas
pleural effusion associated with a known malignancy but ADA-1 accounts for approximately 70 percent of the activ-
the pleura is not involved with tumor.149 These paramalig- ity of the total ADA activity in empyema.158 ADA-2 prob-
nant effusions result from local effects of the tumor, such ably reflects monocyte/macrophage origin, while ADA-1
as obstructive atelectasis or pneumonia, systemic effects originates from lymphocyte or neutrophil turnover.
such as pulmonary embolism, and complications of However, measurement of ADA-2 is substantially more
therapy, such as radiation pleuritis or drug-induced expensive than measurement of total ADA and, currently,
pleurisy. Other explanations for the variable cytological is not available clinically in the USA.153 The decline in the
diagnosis include the type of tumor (high positivity with number of cases of tuberculous pleural effusions in the
adenocarcinoma150 and low in Hodgkin lymphoma),151 USA may have contributed to the relative unavailability of
References 223
●14. Stark DD, Shanes JG, Baron RL, Roach DD. Biochemical features of 37. Light RW. The undiagnosed pleural effusion. Clin Chest Med 2006;
urinothorax. Arch Intern Med 1982; 42: 1505–11. 27: 309–19.
15. Heffner JE. Discriminating between transudates and exudates. Clin 38. Kolditz M, Halank M, Schiemanck CS, Schmeisser A, Hoffken G.
Chest Med 2006; 27: 241–52. High diagnostic accuracy of NT-proBNP for cardiac origin of
16. Heffner JE, Feinstein D, Barbieri C. Methodologic standards for pleural effusions. Eur Respir J 2006; 28: 144–50.
diagnostic test research in pulmonary medicine. Chest 1998; 114: 39. Gegenhuber A, Mueller T, Dieplinger B, et al. Plasma B-type
877–85. natriuretic peptide in patients with pleural effusions: preliminary
17. Heffner JE. Evaluating diagnostic tests in the pleural space. observations. Chest 2005; 12: 1003–9.
Differentiating transudates from exudates as a model. Clin Chest 40. Sackett DL, Straus SE, Richardson WS, Rosenberg W, Haynes RB.
Med 1998; 19: 277–93. Evidence-based medicine. How to practice and teach EBM, 2nd
●18. Light RW, MacGregor I, Luchsinger PC, Ball WC. Pleural effusion: edn. Edinburgh: Churchill Livingstone, 2000.
the diagnostic separation of transudates and exudates. Ann Intern ●41. Heffner JE, Sahn SA, Brown LK. Multilevel likelihood ratios for
Med 1972; 77: 507–13. identifying exudative pleural effusions. Chest 2002; 121:
19. Heffner JE, Brown LK, Barbieri C. Diagnostic value of tests that 1916–20.
discriminate between exudative and transudative pleural 42. Leuallen EC, Carr DT. Pleural effusion, a statistical study of 436
effusions. Chest 1997; 111: 970–9. patients. N Engl J Med 1955; 252: 79–83.
20. Costa M, Quiroga T, Cruz E. Measurement of pleural fluid 43. Light RW. Pleural diseases. Baltimore: Williams & Wilkins, 1995:
cholesterol and lactate dehydrogenase. A simple and accurate set 36–74.
of indicators for separating exudates from transudates. Chest 44. Meisel S, Shamiss A, Thaler M, et al. Pleural fluid to serum
1995; 108: 1260–3. bilirubin concentration for the separation of transudates and
21. Hamm H, Brohan U, Bohmer R, Missmahl H-P. Cholesterol in exudates. Chest 1990; 98: 141–4.
pleural effusions. A diagnostic aid. Chest 1987; 92: 296–302. 45. Burgess L, Maritz FJ, Taljaard JJF. Comparative analysis of the
22. Romero S, Candela A, Martín C, et al. Evaluation of different biochemical parameters used to distinguish between pleural
criteria for the separation of pleural transudates from exudates. transudates and exudates. Chest 1995; 107: 1604–9.
Chest 1993; 104: 399–04. ◆46. Sahn SA. State of the art. The pleura. Am Rev Respir Dis 1988;
23. Suay VG, Moragón EM, Viedma EC, et al. Pleural cholesterol in 138: 184–234.
differentiating transudates and exudates. A prospective study of 47. Winterbauer RH, Riggins RCK, Griesman FA, Bauermeister DE.
232 cases. Respiration 1995; 62: 57–63. Pleuropulmonary manifestions of Waldenstrom’s
24. Valdés L, Pose A, Suàrez J, et al. Cholesterol: a useful parameter macroglobulinemia. Chest 1974; 66: 368–75.
for distinguishing between pleural exudates and transudates. 48. Rodriguez JN, Pereira A, Martinez JC, et al. Pleural effusion in
Chest 1991; 99: 1097–102. multiple myeloma. Chest 1994; 105: 622–4.
25. Chibante A, Neves D, Miranda S, Dias R. [Cholesterol as a 49. Horwitz ML, Schiff M, Samuels J, et al. Pneumocystis carinii
diferential parameter to separate pleural transudates from pleural effusion. Pathogenesis and pleural fluid analysis. Am Rev
exsudates.]. Rev Port Pneumol 2006; 12 (6 Suppl 1): 25. Respir Dis 1993; 148: 232–4.
26. Roth BJ, O’Meara TF, Cragun WH. The serum-effusion albumin ●50. Light RW, Girard WM, Jenkinson SG, George RB. Parapneumonic
gradient in the evaluation of pleural effusions. Chest 1990; 98: effusions. Am J Med 1980; 69: 507–11.
546–9. ●51. Potts DE, Levin DC, Sahn SA. Pleural fluid pH in parapneumonic
27. Romero-Candeira S, Fernandez C, Martin C, Sanchez-Paya J, effusions. Chest 1976; 70: 328–31.
Hernandez L. Influence of diuretics on the concentration of 52. Pettersson T, Klockars M, Helmstrom PE. Chemical and
proteins and other components of pleural transudates in patients immunological features of pleural effusions: comparison between
with heart failure. Am J Med 2001; 110: 681–6. rheumatoid arthritis and other diseases. Thorax 1982; 37: 354–61.
28. Romero-Candeira S, Hernandez L, Romero-Brufao S, et al. Is it 53. Johnson JR, Falk A, Iber C, Davies S. Paragonimiasis in the United
meaningful to use biochemical parameters to discriminate States. A report of 9 cases in Hmong immigrants. Chest 1982; 82:
between transudative and exudative pleural effusions? Chest 168–71.
2002; 122: 1524–9. ●54. Light RW, Erozan YS, Ball WC Jr. Cells in pleural fluid: their value
29. Atalay F, Ernam D, Hasanoglu HC, Karalezli A, Kaplan O. Pleural and differential diagnosis. Arch Intern Med 1973; 132: 854–60.
adenosine deaminase in the separation of transudative and 55. Berger HW, Mejiia E. Tuberculous pleurisy. Chest 1973; 63:
exudative pleural effusions. Clin Biochem 2005; 38: 1066–70. 88–92.
30. Porcel JM, Pena JM, Vicente de Vera C, et al. Bayesian analysis 56. Pettersson T, Riska H. Diagnostic value of total and differential
using continuous likelihood ratios for identifying pleural exudates. leukocyte counts in pleural effusions. Acta Med Scand 1981; 210:
Respir Med 2006; 100: 1960–5. 129–35.
31. Gonlugur U, Gonlugur TE. The distinction between transudates and 57. Antony VB, Repine JE, Sahn SA. Experimental models of
exudates. J Biomed Sci 2005; 12: 985–90. inflammation in the pleural space. In: Chretein J, Bignon J, Hirsch
32. Joseph J, Badrinath P, Basran GS, Sahn SA. Is the pleural fluid A (eds). The pleural in health and disease. New York: Marcel
transudate or exudate? A revisit of the diagnostic criteria. Thorax Dekker, 1985: 387–400.
2001; 56: 867–70. 58. Antony VB, Godbey SW, Kunkel SL, et al. Recruitment of
33. Chakko SC, Caldwell SH, Sforza PP. Treatment of congestive heart inflammatory cells to the pleural space. Chemotactic cytokines, IL-
failure. Its effect on pleural fluid chemistry. Chest 1989; 95: 8, and monocyte chemotactic peptide-1 in human pleural fluids. J
798–802. Immunol 1993; 151: 7216–23.
34. Heffner JE, Highland K, Brown LK. A meta-analysis derivation of 59. Broaddus VA, Hebert CA, Vitangcol RV, et al. Interleukin-8 is a
continuous likelihood ratios for diagnosing pleural fluid exudates. major neutrophil chemotactic factor in pleural liquid of patients
Am J Respir Crit Care Med 2003; 167: 1591–9. with empyema. Am Rev Respir Dis 1992; 146: 825–30.
35. Tomcsanyi J, Nagy E, Somloi M, et al. NT-brain natriuretic peptide 60. Antony VB, Sahn SA, Antony AC, Repine JE. Bacillus
levels in pleural fluid distinguish between pleural transudates and Calmette–Guerin-stimulated neutrophils release chemotaxins for
exudates. Eur J Heart Fail 2004; 6: 753–6. monocytes in rabbit pleural spaces and in vitro. J Clin Invest 1985;
36. Porcel JM. The use of probrain natriuretic peptide in pleural fluid 76: 1514–21.
for the diagnosis of pleural effusions resulting from heart failure. ●61. Sahn SA. Diagnostic value of pleural fluid analysis. Semin Respir
Curr Opin Pulm Med 2005; 11: 329–33. Crit Care Med 1995; 16: 269–78.
References 225
●62. Yam LT. Diagnostic significance of lymphocytes in pleural effusion. 88. Miller KS, Wooten S, Sahn SA. Urinothorax: a cause of a low pH
Ann Intern Med 1967; 66: 972–82. transudate. Am J Med 1988; 85: 448–9.
●63. Nordkild P, Kromann-Andersen H, Struve-Christsen E. Yellow nail 89. Yamada S. Uber die serose Flussigkeit in der Plerachohle der
syndrome – the triad of yellow nails, lymph edema, and pleural gesunden menschen. Z Ges Exp Med 1993; 90: 343–8.
effusions. Acta Med Scand 1986; 219: 221–7. 90. Sahn SA, Willcox ML, Good JT Jr, et al. Characteristics of normal
64. Sahn SA, Kaplan RL, Maulitz RM, Good JT Jr. Rheumatoid pleurisy. rabbit pleural fluid: physiologic and biochemical implications.
Observation on the development of low pleural fluid pH and Lung 1979; 156: 63–9.
glucose levels. Arch Intern Med 1980; 140: 1237–8. 91. Rolf LL, Travis DM. Pleural fluid-plasma bicarbonate gradients in
●65. Berger HW, Rammohan G, Neff MS, Buhain WJ. Uremic pleural oxygen-toxic and normal rats. Am J Physiol 1973; 224: 857–61.
effusion. A study in 14 patients on chronic dialysis. Ann Intern 92. Taryle DA, Good JT Jr, Sahn SA. Acid generation by pleural fluid:
Med 1975; 82: 362–4. possible role in the determination of pleural fluid pH. J Lab Clin
66. Nicholls AJ, Friend JAR, Legge JS. Sarcoid pleural effusions: three Med 1979; 93: 1041–6.
cases and review of the literature. Thorax 1980; 35: 277–81. 93. Sahn SA, Taryle DA, Good JT Jr. Experimental empyema: time
67. Huggins JT, Doelken P, Sahn SA, et al. Pleural effusions in a series course and pathogenesis of pleural fluid acidosis and low pleural
of 181 outpatients with sarcoidosis. Chest 2006; 129: 1599–1604. fluid glucose. Am Rev Respir Dis 1979; 120: 355–61.
68. Miller JJ. Anatomy of the thoracic duct and chylothorax. In: ●94. Sahn SA, Reller LB, Taryle DA, et al. The contribution of leukocytes
Shields T, Locicero J, Ponn R (eds). General thoracic surgery, 6th and bacteria to the low pH of empyema fluid. Am Rev Respir Dis
edn. Philadelphia: Lippincott, Williams and Wilkins; 2005: 879–88. 1983; 128: 811–5.
●69. Judson MA, Handy JR, Sahn SA. Pleural effusion from acute lung ●95. Good JT Jr, Taryle DA, Sahn SA. The pathogenesis of low glucose,
rejection. Chest 1997; 111: 1128–30. low pH malignant effusions. Am Rev Respir Dis 1985; 131:
●70. Lee Y, Vaz M, Ely K, et al. Symtomatic persistent post-coronary 737–41.
artery bypass graft pleural effusions requiring operative treatment. 96. Poe RH, Marin MG, Israel RH, Kallay MC. Utility of pleural fluid
Chest 2001; 119: 795–800. analysis in predicting tube thoracostomy/decortication in
71. Scharer L, McClement JH. Isolation of tubercle bacilli from needle parapneumonic effusions. Chest 1991; 100: 963–7.
biopsy specimens of parietal pleura. Am Rev Respir Dis 1968; 97: 97. Potts DE, Taryle DA, Sahn SA. The glucose-pH relationship in
466–8. parapneumonic effusions. Arch Intern Med 1978; 138: 1378–80.
72. Levine H, Metzger W, Lacera D, Kay L. Diagnosis of tuberculous 98. Limthongkul S, Charoenlap P, Nuchprayoon C, Songkhla YN.
pleurisy by culture or pleural biopsy specimen. Arch Intern Med Diagnostic and prognostic significance of pleural fluid pH and
1970; 126: 269–71. pCO2 in the exudative phase of parapneumonic effusions. J Med
73. Nakamura Y, Ozaki T, Kamei T, et al. Factors that stimulate the Assoc Thailand 1983; 66: 762–8.
proliferation and survival of eosinophils in eosinophilic pleural 99. Light R, MacGregor M, Ball WC Jr, Luchsinger PC. Diagnostic
effusion: relationship to granulocytes/macrophage colony- significance of pleural fluid pH and PCO2. Chest 1973; 64: 591–6.
stimulating factor, interleukin-5 and interleukin-3. Am Rev Cell ●100. Heffner JE, Brown LK, Barbieri C. Pleural fluid chemical analysis in
Mol Biol 1993; 8: 605–11. parapneumonic effusions. A meta-analysis. Am J Respir Crit Care
●74. Rubins JB, Rubins HB. Etiology and prognostic signifcance of Med 1995; 151: 1700–8.
eosinophilic pleural effusions. A prospective study. Chest 1996; 101. Hanley JA, McNeil BJ. The meaning and use of the area under the
110: 1271–4. receiving operating characteristic (ROC) curve. Radiology 1982;
●75. Martinez-Garcia MA, Cases-Viedma E, Cordero-Rodriguez PJ, et al. 143: 29–36.
Diagnostic utility of eosinophils in the pleural fluid. Eur Respir J ◆102. Colice GL, Curtis A, Deslauriers J, et al. Medical and surgical
2000; 15: 166–9. treatment of parapneumonic effusions : an evidence-based
76. Askin FB, McCann BG, Kuhn C. Reactive eosinophilic pleuritis. Arch guideline. Chest 2000; 118: 1158–71.
Path Lab Med 1997; 101: 187–91. ●103. Sahn SA, Good JT. Pleural fluid pH in malignant effusions.
77. Spriggs AI, Boddington MM. The cytology of effusions, 2nd edn. Diagnostic, prognostic, and therapeutic implications. Ann Intern
New York: Grune & Stratton, 1968: 5–17. Med 1988; 108: 345–9.
78. Maltais F, Laberge F, Cormier Y. Blood hypereosinophilia in the 104. Sanchez-Armengol A, Rodriguez-Panadero F. Survival and talc
course of posttraumatic pleural effusion. Chest 1990; 98: 348–51. pleurodesis in metastatic pleural carcinoma revisited. Report of
79. Heidecker J, Sahn SA. The four faces of a parapneumonic effusion. 125 cases. Chest 1993; 104: 1317–19.
Respirology 2007; 12: 610–13. 105. Martinez-Moragon E, Aparicio J, Sanchis J, et al. Malignant pleural
80. Mattson SB. Monosymptomatic exudative pleurisy in persons effusion: prognostic factors for survival and response to chemical
exposed to asbestos dust. Scand J Respir Dis 1997; 56: 263–72. pleurodesis in a series of 120 cases. Respiration 1998; 65: 108–13.
●81. Hillerdal G, Ozesmi M. Benign asbestos pleural effusion: 73 106. Lan RS, Lo SK, Chuang ML, et al. Elastance of the pleural space: a
exudates in 60 patients. Eur J Respir Dis 1987; 71: 113–21. predictor for the outcome of pleurodesis in patients with
82. Hurwitz S, Leiman G, Shapiro C. Mesothelial cells in pleural fluid: malignant pleural effusion. Ann Intern Med 1997; 126: 768–74.
TB or not TB? S Afr Med J 1980; 57: 937–9. 107. Rodriguez-Panadero F, Segado A, Martin Juan J, et al. Failure of
83. Potrykus AM, Steinmann G, Stein E, Mertelsmann R. T- and B-cell talc pleurodesis is associated with increased pleural fibrinolysis.
responses in patients with malignant pleural effusions. Br J Cancer Am J Respir Crit Care Med 1995; 151: 785–90.
1981; 43: 471–7. 108. Rodriguez-Panadero F, Lopez Mejias J. Low glucose and pH levels
84. Pettersson T, Klockars M, Hellstrom PE, et al. T and B lymphocytes in malignant pleural effusions. Diagnostic significance and
in pleural effusions. Chest 1978; 73: 49–51. prognostic value in respect to pleurodesis. Am Rev Respir Dis
85. Domagala W, Emeson EE, Kos LG. T and B lymphocyte enumeration 1989; 139: 663–7.
in the diagnosis of lymphocyte-rich pleural fluids. Acta Cytol 109. Salomaa ER, Pulkki K, Helenius H. Pleurodesis with doxycycline or
1981; 25: 108–10. Corynebacterium parvum in malignant pleural effusion. Acta
●86. Good JT, Taryle DA, Maulitz RM, et al. The diagnostic value of Oncol 1995; 34: 117–21.
pleural fluid pH. Chest 1980; 78: 55–9. 110. Rodriguez-Panadero F, Antony VB. Pleurodesis: state of the art.
◆87. Sahn SA. Pleural fluid pH in the normal state and in diseases Eur Respir J 1997; 10: 1648–54.
affecting the pleural space. In: Chretien J, Bignon J, Hirsch A (eds). ●111. Heffner JE, Nietert PJ, Barbieri C. Pleural fluid pH as a predictor of
The pleura in health and disease. New York: Marcel Dekker, 1985: pleurodesis failure: analysis of primary data. Chest 2000; 117:
253–66. 87–95.
226 Pleural fluid analysis
●112. Heffner JE, Nietert PJ, Barbieri C. Pleural fluid pH as a predictor of ●137. Coe JE, Aikawa JK. Cholesterol pleural effusion. Arch Intern Med
survival for patients with malignant pleural effusions. Chest 2000; 1961; 108: 763–74.
117: 79–86. 138. Harvey AM, Shulman LE, Tumulty PA, et al. Systemic lupus
113. Fishman RA. Carrier transport and concentration of glucose in erythematosis: review of the literature and clinical analysis of 138
cerebrospinal fluid in meningeal diseases. Ann Intern Med 1965; cases. Medicine 1954; 33: 291–437.
63: 153–5. 139. Walker WC, Wright V. Rheumatoid pleuritis. Ann Rheum Dis 1967;
◆114. Sahn SA. Pathogenesis and clinical features of diseases associated 26: 467–74.
with a low pleural fluid glucose. In: Chretien J, Bignon J, Hirsch A 140. Horler AR, Thompson M. The pleural and pulmonary complications
(eds). The pleura in health and disease. New York: Marcel Dekker, of rheumatoid arthritis. Ann Intern Med 1959; 50: 1179–1203.
1985: 267–85. 141. Khare V, Baethge B, Lang S, et al. Antinuclear antibodies in pleural
115. Dodson WH, Hollingsworth JW. Pleural effusion in rheumatoid fluid. Chest 1994; 106: 866–71.
arthritis. Impaired transport of glucose. N Engl J Med 1966; 275: 142. Halla JT, Schrohenloher RE, Volanakis JE. Immune complexes and
1337–42. other laboratory features of pleural effusions. Ann Intern Med
●116. Carr DT, McGuckin WF. Pleural fluid glucose. Serial observations of 1980; 92: 748–52.
its concentration following oral administration of glucose in the 143. Levine H, Szanto M, Grieble HG, et al. Rheumatoid factor in non-
patient with rheumatoid pleural effusions and malignant rheumatoid pleural effusions. Ann Intern Med 1968; 69: 487–92.
effusions. Am Rev Respir Dis 1968; 97: 302–5. 144. Hunder GG, McDufie FC, Heppern GG. Pleural fluid complement in
117. Clarkson B. Relationship between cell type, glucose concentration, systemic lupus erythematosus in rheumatoid arthritis. Ann Intern
and response to treatment in neoplastic effusions. Cancer 1964; Med 1972; 76: 357–62.
17: 914–28. 145. Glovsky MM, Louie JS, Pitts WH Jr, Alenty A. Reduction of pleural
●118. Light RW, Ball WC Jr. Glucose and amylase in pleural effusions. J fluid complement activity in patients with systemic lupus
Am Med Assoc 1973; 225: 257–60. erythematosus and rheumatoid arthritis. Clin Immunol
119. Russakoff AH, le Maistre CA, Dewlett HJ. An evaluation of the Immunopathol 1976; 6: 31–41).
pleural fluid glucose determination. Am Rev Respir Dis 1962; 85: 146. Andrews BS, Arora NS, Shadforth MF, et al. The role of immune
220–3. complexes in the pathogenesis of pleural effusions. Am Rev Respir
●120. Good JT Jr, King TE, Antony VB, Sahn SA. Lupus pleuritis: clinical Dis 1981; 125: 115–20.
features and pleural fluid characteristics with special reference to 147. Grunze H. The comparative diagnostic accuracy, efficiency, and
pleural fluid antinuclear antibody titers. Chest 1983; 84: 714–8. specificity of cytologic techniques used in the diagnosis of
121. Maulitz RM, Good JT Jr, Kaplan RL, et al. The pleuropulmonary malignant neoplasm and serous effusions of the pleuroperitoneal
consequences of esophageal rupture in an experimental model. cavities. Acta Cytol 1964; 8: 150–64.
Am Rev Respir Dis 1979; 120: 363–7. 148. Jarvi OH, Kunnas RJ, Laitio MT, Tyrkko JES. The accuracy and
122. Kaye MD. Pleuropulmonary complications of pancreatitis. Thorax significance of cytologic cancer diagnosis of pleural effusions.
1968; 23: 297–306. Acta Cytol 1972; 16: 152–7.
123. Joseph J, Viney S, Beck P, et al. A prospective study of amylase- 149. Sahn SA. Malignant pleural effusions. Sem Respir Med 1987; 9:
rich pleural effusion with special reference to amylase isoenzyme 43–53.
analysis. Chest 1992; 102: 1455–9. 150. Naylor B, Schmidt RW. The case for exfoliative cytology of serous
124. Rockey DC, Cello JP. Pancreaticopleural fistula: a report of 7 cases effusions. Lancet 1964; i: 711–2.
and review of the literature. Medicine 1990; 69: 332–4. 151. Melamed MR. The cytologic presentation of malignant lymphomas
125. Abbott OA, Mansour KA, Logan WD Jr, et al. Atraumatic so-called in related diseases and effusions. Cancer 1963; 16: 413–31.
‘spontaneous’ rupture of the esophagus. A review of 47 personal 152. Moriarty AT, Wiersema L, Snyder W, et al. Immunophenotyping of
cases with comments on a new method of surgical therapy. J cytologic specimens by flow cytometry. Diagn Cytopathol 1993; 9:
Thorac Cardiovasc Surg 1970; 59: 67–83. 252–8.
126. Sherr HP, Light RW, Merson MH, et al. Origin of pleural fluid 153. Kataria YP, Khurshid I. Adenosine deaminase in the diagnosis of
amylase in esophageal rupture. Ann Intern Med 1972; 76: 785–6. tuberculous pleural effusion. Chest 2001; 120: 334–6.
127. Ende N. Studies of amylase activity in pleural effusions and 154. Roth BJ. Searching for tuberculosis in the pleural space. Chest
ascites. Cancer 1960; 13: 283–7. 1999; 116: 3–5.
◆128. Kramer MR, Sepero RJ, Pitchenik AE. High amylase in neoplasm- 155. Pettersson T, Kaarina O, Weber TH. Adenosine deaminase in the
related pleural effusions. Ann Intern Med 1989; 110: 567–9. diagnosis of pleural effusions. Acta Med Scand 1984; 215:
129. Perry TT. Role of lymphatic vessels in the transmission of lipase in 299–304.
deceminated pancreatic lymphatic necrosis. Arch Pathol 1947; 43: 156. Lee YCG, Rogers JT, Rodriguez RM, et al. Adenosine deaminase
456–65. levels in nontuberculous lymphocytic pleural effusions. Chest
130. Dumont AE, Doubilet H, Mulholind JH. Lymphatic pathways of 2001; 120: 356–61.
pancreatic secretion in man. Ann Surg 1960; 153: 403–9. 157. Valdes L, San Jose E, Alvares D, et al. Diagnosis of tuberculous
131. Lueng KC. Pancreatic pleural effusion with normal serum amylase pleurisy using the biologic parameters adenosine deaminase,
levels. J R Soc Med 1985; 78: 698. lysozyme, and interferon-gamma. Chest 1993; 103: 458–65.
132. Buckler H, Honeybourne D. Raised pleural fluid amylase level as an 158. Ungerer JPJ, Oosthuizen HM, Retief JH, Bissbort SH. Significance
aid in the diagnosis of adenocarcinoma of the lung. Br J Clin Pract of adenosine deaminase and its isoenzymes in tuberculous
1984; 38: 359–61, 371. effusions. Chest 1994; 106: 33–7.
133. Rodriguez BP, Rogers JT, Ayo DS, et al. Routine measurement of 159. Burgess LJ, Maritz FJ, Roux IL, Taljaard JJ. Use of adenosine
pleural fluid amylase is not indicated. Arch Intern Med 2001; 161: deaminase as a diagnostic tool for tuberculous pleurisy. Thorax
228–32. 1995; 50: 672–4.
●134. Staats BA, Ellefson RD, Budahn LL, et al. The lipoprotein profile of ●160. Valdes L, Alvarez D, San Jose E, et al. Tuberculous pleurisy: a study
chylous and nonchylous pleural effusions. Mayo Clin Proc 1980; of 254 patients. Arch Intern Med 1998; 158: 2017–21.
55: 700–4. 161. Valdes L, Alvares D, San Jose E, et al. Value of adenosine
135. Teba L, Dedhia AT, Bowen R, Alexander JC. Chylothorax review. Crit deaminase in the diagnosis of tuberculous pleural effusions in
Care Med 1985; 13: 49–52. young patients in a region of high prevalence of tuberculosis.
136. Hamm H, Pfalzer B, Fabel H. Lipoprotein analysis in a chyliform Thorax 1995; 50: 600–3.
pelural effusion: implications for pathogenesis and diagnosis.
Respiration 1991; 58: 294–300.
18
Pleural manometry
prior to the insertion of the device.10 Though there contin- Depending on the cause of the effusion, as fluid builds
ues to be great debate between the two dominant theories up in the pleural space, Ppl typically rises. As fluid is
of normal pleural pressure physiology,10,11 these concep- removed, one expects the lung to expand, the chest wall
tual differences may only be of practical importance at the contract and the Ppl reach its steady state at FRC. Pleural
termination of a thoracentesis, when there is only a physio- pressure, however, can be negative, as in the case of
logical amount (5–8 mL) of pleural fluid present. In the trapped lung, or start out positive and drop rapidly, as is
presence of even a small effusion (several hundred milli- the case with lung entrapment.
liters), one can measure Ppl with a variety of techniques as
the viscous resistance to flow becomes negligible. The
pressure measured, therefore, is an accurate representation DEFINITIONS: LUNG ENTRAPMENT VERSUS
of the hydrostatic pressure in the effusion at the level of the TRAPPED LUNG
catheter/transducer.
There has been some suggestion that when measuring Though confusing terms, ‘lung entrapment’ and ‘trapped
Ppl, the height of the manometer relative to the effusion is lung’ describe different pathophysiology and are thought
insignificant.12 The reason for this was assumed to be that to represent a spectrum of pleural inflammation and
as Pascal’s law states, in an enclosed system such as the repair. Lung entrapment describes an inability of the lung
chest, pressure is transmitted equally in all directions and to re-expand due to visceral pleural thickening, endo-
will exert the same force equally on the lung, chest wall and bronchial obstruction or diseases that lead to an increase
manometer, regardless of where the needle is inserted. in the elastic recoil of the lung such as interstitial lung
This is in contrast to an open system, where the only force disease and lymphangitic carcinomatosis. Patients often
is that of gravity.12 present with dyspnea related to the effusion as well as with
More realistically, in a hydrothorax, a hydrostatic pres- signs and symptoms related to the underlying disease.
sure gradient of 1 cmH2O/cm height is present, and so the Chest discomfort or other signs of pleural inflammation
pressure read by the manometer represents the pressure at may also be present. The effusion associated with lung
a specific level, and not the pressure thoughout the entrapment is typically exudative, representing an active
hydrothorax. With the removal of pleural fluid and a inflammation. With normal healing of the underlying
reduction in the height of the fluid column above the process, the effusion may completely resolve.
catheter, the influence of the hydrostatic pressure gradient Trapped lung, on the other hand, is the sequelae of
becomes less. The measured Ppl therefore does depend on prior pleural inflammation resulting in visceral pleural
where in the effusion the catheter is placed. Placing the scarring.13,14 This creates negative pressure in the pleural
catheter at the most dependent part of the effusion has the space and results in an ‘effusion ex-vacuo’. Since there is no
potential benefits of (1) maximizing the amount of fluid active pleural inflammation, patients typically present with
that is able to be removed, and (2) minimizing the creation a chronic, and asymptomatic, effusion that is identified on
of deformation forces from the contact of the catheter with routine physical examination or chest X-ray. As the pleural
the lung. With this approach, the pressure measured at the fluid formation results from an excess of negative pleural
level of the catheter will reflect the pressure in the pleural pressure, it is rare to see contralateral mediastinal shift on
space, and hence the pressure exerted on the lung and a chest radiograph, even in the presence of a moderate to
chest wall, at that level (Figure 18.1). large effusion. Likewise, as the effusion is due to an
imbalance of hydrostatic forces, it is always transudative in measure mean pleural pressure during thoracenteses in 52
nature. Since the large majority of these patients are patients with the goal of determining the clinical utility of
asymptomatic, therapy aimed at the pleural effusion is pleural manometry, and to evaluate the safety of large-
usually not required. Should the patient have exertional volume thoracentesis. Pleural fluid was removed until
dyspnea resulting from a restrictive ventilatory defect due either the mean Ppl fell to less than −20 cmH2O, no more
to the effusion, decortication will likely be required to fluid could be obtained or patients developed symptoms
expand the underlying lung. described as more than minimal in severity. Though the
As mentioned above, lung entrapment and trapped initial Ppl varied widely (−21 cmH2O to +8 cmH2O) an
lung are part of a continuum of the natural healing process initial pressure of less than −5 cmH2O was only seen in
of the underlying disease and, as such, one may occasion- patients with malignant effusions or trapped lung. They
ally obtain pleural fluid results that fall in the exudative also measured pleural elastance (change in pressure
range, even in the setting of trapped lung physiology, divided by change in volume) and described three distinct
depending on when the thoracentesis is performed in the pleural elastance curves: (1) removal of a large amount of
healing process.15 fluid with minimal change in pressure (as can be seen in
patients with hepatic hydrothorax); (2) a relatively normal
initial curve followed by a sharp drop in pressure; and (3) a
TECHNIQUE negative initial pressure with a rapid drop in pressure
(Figure 18.2). They felt that the first curve (open circles)
Pleural pressure can be measured via several techniques.
represented a normal pleural elastance, the second curve
These include using a U-shaped water manometer,16 an
(closed circles) that of lung entrapment, and the third
‘overdamped’ water manometer17 or sophisticated elec-
curve (x marks) was consistent with trapped lung. A
tronic transducer systems that allow sampling several
pleural elastance >25 cmH2O/L was seen in patients with
times a second as well as the ability to store data for further
malignancy or trapped lung.
analysis. A benefit of the U-shaped manometer is the fact
Light’s group later investigated the relationship
that it is relatively inexpensive. A disadvantage, however, is
between changes in Ppl during thoracentesis and improve-
the fact that it may be difficult to accurately record values
ment in lung function.18 They found that although the
due to the inspiratory and expiratory pleural pressure
improvement in mean forced vital capacity (FVC) follow-
swings. Electronic transducer systems can be configured to
ing thoracentesis was related to the volume of fluid
standard monitors in the intensive care unit (ICU),
removed (approximately 21 mL for each 100 mL
however, one needs to calibrate an ‘offset’ as these moni-
removed), the correlation coefficient was only 0.49.
tors are not calibrated to measure negative pressure.
Interestingly, the improvement in FVC was most closely
Additionally, ICU hemodynamic transducers report data
related to Ppl after the removal of 800 mL of fluid (r =
in mmHg, instead of the standard cmH2O typically used
−0.57, p < 0.01). Improvement in FVC was also signifi-
for Ppl measurements. This problem is easily resolved by
cantly (and negatively) related to initial Ppl, as well as the
adding the conversion factor of 1 mmHg = 1.36 cmH2O. A
change in Ppl after removal of 400 and 800 mL of pleural
distinct advantage of using an electronic transducer system
fluid. The negative correlation indicates that the larger
is the ability to examine the Ppl curves after the data has
pressure changes were associated with smaller improve-
been collected. This allows one to measure end inspiratory
and end expiratory, as well as mean Ppl. Though it is
unclear at this time which of these pressures is clinically
10
most important, it is likely that future studies will clarify
this issue. Most authors currently report mean or end- 0
Pleural pressure (cmH2O)
MANOMETRY IN THE CLINICAL SETTING Figure 18.2 Pleural elastance curves (from Light RW, Jenkinson
SG, Minh VD, George RB. Observations on pleural fluid pressures
In 1980, Light and colleagues16 used a water-filled U- as fluid is withdrawn during thoracentesis. Am Rev Respir Dis
shaped manometer connected to an Abram’s needle to 1980; 121: 799–804.
230 Pleural manometry
ments in FVC, consistent with the physiology of unex- define the relationships of Ppl to underlying diagnosis as
pandable lung. That is to say, the lack of increase in lung well as complications of therapeutic thoracentesis.
volume correlates with more negative pleural pressure. Although there was no significance to a negative initial Ppl,
Huggins and colleagues15 have recently described their all patients with an initial Ppl of less than −4 cmH2O and
use of an ‘air-contrast’ computed tomography (CT) scan an elastance of the pleural space (Eps) >33 cmH2O/L had
to visualize visceral pleural thickening and help define the trapped lung. There were no cases of re-expansion pul-
cause of unexpandable lung in a group of 247 consecutive monary edema, despite a mean removal of 1.45 L of
patients undergoing pleural manometry during thoracen- pleural fluid. We routinely use pleural manometry during
tesis. They identified 11 patients with a clinical diagnosis of therapeutic thoracentesis, and terminate the tap if patients
trapped lung. All of these patients developed a mean Ppl of develop chest discomfort, end-expiratory Ppl is less than
less than −25 cmH2O and had prior pleural fluid analysis −20 cmH2O, or when there is no more fluid. In our recent
that was not suggestive of malignancy or pleural inflam- series of over 185 large volume (>1 L) thoracenteses (mean
mation. At the termination of the therapeutic thoracen- 1.67 L, range 1000–6550 mL), only one patient developed
tesis, they instilled atmospheric air, a ‘diagnostic clinically significant RPE.22 There was no relationship to
pneumothorax’ with the goal of raising the Ppl to a more the volume of pleural fluid removed, Ppl, Eps or symp-
physiological mean of −5 cmH2O. A subsequent CT scan toms during the thoracentesis, suggesting that RPE is a
confirmed visceral pleural thickening in all 11 patients. As rare event, and that large effusions can be drained com-
expected, all of these patients had a high pleural elastance pletely, provided that patients do not develop chest dis-
(Eps >19 cmH2O/L). The authors favor using the air-con- comfort (see below) or Ppl is less than −20 cmH2O.
trast CT as part of the diagnostic approach to patients with
trapped lung as a way of minimizing additional pleural
interventions that will be of little value, and in fact have PLEURODESIS
incorporated the ‘diagnostic pneumothorax’ as part of
their routine protocol should patients develop significantly For pleurodesis to be successful, the pleural surfaces need
low Ppl and pleural fluid is still present. to appose each other. If the lung is entrapped and does
not re-expand during thoracentesis, the odds of successful
pleurodesis are reduced. This fact is likely the single
RE-EXPANSION PULMONARY EDEMA largest confounder in the multiple studies comparing
pleurodesis agents, as documentation of lung re-expan-
During a thoracentesis, one would ideally like to remove as sion was used as a criteria prior to randomization in only
much fluid as is safely possible. The goals of completely two studies.23,24 Lan and colleagues4 used pleural manom-
draining the pleural space include maximizing sympto- etry to predict lung re-expansion and found that a pleural
matic relief, sparing the patient multiple procedures, space elastance of ≥19 cmH2O after the removal of 500
increasing the yield of other diagnostic tests, such as a mL of pleural fluid predicted pleurodesis failure. This is
post-thoracentesis CT scan, and documenting lung re- clinically important, as patients with lung entrapment can
expansion prior to attempts at pleurodesis. In Light’s orig- still achieve pleural palliation with a significant reduction
inal study, they showed that if thoracenteses were in their dyspnea with the placement of a chronic
terminated when the Ppl dropped to less than −20 cmH2O, indwelling (PleurX™) catheter,25 and one should not
re-expansion pulmonary edema was avoided despite attempt pleurodesis prior to documenting full lung
removing large quantities of fluid.16 They conclude that, expansion by either manometry or imaging after complete
‘as the operator cannot easily estimate pleural pressure… removal of pleural fluid.
therapeutic thoracentesis should be limited to 1000 mL
unless pleural pressures are monitored’. A pressure of −20
cmH2O was arbitrarily chosen based on prior animal SYMPTOMS
studies19,20 that showed a minimal risk of re-expansion
pulmonary edema if Ppl was kept above The development of symptoms such as cough and chest
−20 mmHg (approximately −27 cmH2O), but a significant pain/discomfort is quite common during therapeutic tho-
risk was present with Ppl of −40 mmHg (approximately racentesis. As most physicians do not currently perform
−54 cmH2O). The above quote has led to the majority of pleural manometry during thoracentesis, we investigated
clinicians terminating thoracentesis after removing the relationship of Ppl to patient symptoms as pleural fluid
1000–1500 mL without regard to the amount of remaining is removed.3 We measured end expiratory Ppl during ther-
pleural fluid, the potential benefit of removing that fluid or apeutic thoracentesis in 169 consecutive patients undergo-
consideration of pleural pressure. ing therapeutic thoracentesis. Symptoms developed in 17
Pleural manometry allows for the safe drainage of large percent of patients (cough in 6 percent and chest discom-
volumes of fluid as well as avoiding the pressure-related fort in 11 percent). We distinguished between the sensa-
consequences of thoracentesis such as RPE. In a study of tions of a sharp, ipsilateral pain that is typically felt over
61 patients, Villena and colleagues21 used manometry to the ipsilateral shoulder/scapula and another, more vague
References 231
distinct entities of unexpandable lung. pleural fluid pressures as fluid is withdrawn during thoracentesis.
Am Rev Respir Dis 1980; 121: 799–804.
● The majority of patients with trapped lung do
●17. Doelken P, Huggins JT, Pastis NJ, Sahn SA. Pleural manometry:
not require specific therapy aimed at the pleural technique and clinical implications. Chest 2004; 126: 1764–9.
space. ●18. Light RW, Stansbury DW, Brown SE. The relationship between
● Pleural manometry should be used to best select pleural pressures and changes in pulmonary function after
patients with malignant pleural effusions for pal- therapeutic thoracentesis. Am Rev Respir Dis 1986; 133: 658–
61.
liation by pleurodesis or PleurX™ catheter place-
19. Pavlin J, Cheney FW, Jr. Unilateral pulmonary edema in rabbits
ment. after reexpansion of collapsed lung. J Appl Physiol 1979; 46:
● If formal manometry is not performed during 31–5.
thoracentesis, the development of chest discom- 20. Miller WC, Toon R, Palat H, Lacroix J. Experimental pulmonary
fort should raise the suspicion of unexpandable edema following re-expansion of pneumothorax. Am Rev Respir
Dis 1973; 108: 654–6.
lung.
21. Villena V, Lopez-Encuentra A, Pozo F, et al. Measurement of
● Introduction of air into the pleural space, a diag- pleural pressure during therapeutic thoracentesis. Am J Respir Crit
nostic pneumothorax, can be used to help visual- Care Med 2000; 162: 1534–8.
ize visceral pleural thickening in patients with ●22. Feller-Kopman D, Berkowitz D, Boiselle P, Ernst A. Large volume
unexpandable lung. thoracentesis and the risk of re-expansion pulmonary edema. Ann
Thorac Surg 2007; 84: 1656–62.
232 Pleural manometry
23. Diacon AH, Wyser C, Bolliger CT, et al. Prospective randomized talc poudrage vs talc slurry sclerosis for malignant pleural
comparison of thoracoscopic talc poudrage under local anesthesia effusion. Chest 2005; 127: 909–15.
versus bleomycin instillation for pleurodesis in malignant pleural 25. Tremblay A, Michaud G. Single-center experience with 250
effusions. Am J Respir Crit Care Med 2000; 162: 1445–9. tunnelled pleural catheter insertions for malignant pleural
24. Dresler CM, Olak J, Herndon JE, et al. Phase III intergroup study of effusion. Chest 2006; 129: 362–8.
19
Radiology: diagnostic
FERGUS V GLEESON
High-resolution computed tomography is of value in greater positron emission from such areas may enable the
determining whether possible pleural thickening/plaques detection and differentiation of malignant from benign
detected on a CXR are genuine, and should be performed pleural disease. One study of 98 patients showed 18FDG
with thin sections of less than 2 mm and a high spatial res- activity in 61 of 63 patients with pleural malignancy com-
olution reconstruction algorithm.9 CT has also been pared to an absence of activity in 31 of 35 patients with
shown to be of value in the quantification of pleural benign disease. 18FDG activity was intense in 51 of the
disease and to correlate with functional measures in patients with malignant pleural thickening and moderate
patients with impaired lung function secondary to pleural in 10.18
thickening.10
pleura and the endothoracic fascia thickens adjacent to the A massive effusion leads to complete or near-complete
lateral ribs, producing fat pads several millimeters thick, opacification of the hemithorax with mediastinal shift to
especially over the fourth to eighth ribs which can extend the contralateral side.47 A centrally located mediastinum in
into the intercostal spaces or fissures.38,39 the presence of a large effusion suggests that either the
The transverse thoracic and subcostal muscles may be mediastinum is fixed or there is associated pulmonary col-
visible internal to the end of a rib or costal cartilage, the lapse.48
former at the level of the heart adjacent to the lower Inversion of the diaphragm may occur, especially with
sternum or xiphoid process and the latter posteriorly at the large effusions (see Chapter 6, Physiology: changes with
same level. They should be smooth, uniform in thickness pleural effusion and pneumothorax). This occurs most
and bilaterally symmetrical.37 commonly on the left owing to the protective effect of
In a paravertebral location, the pleura and fascia the liver on the right (Figure 19.4).49,50 An inverted
combine to produce a thinner line than that seen laterally. diaphragm, which has reverted to a normal position, is one
The innermost muscle is absent. Occasionally, the paraver- explanation for an apparent failure of a pleural effusion to
tebral line is thicker than expected, which may reflect seg- decrease on a CXR post thoracentesis.
ments of the intercostal vessels.37,40
PLEURAL FLUID
Chest radiography
SUPINE RADIOGRAPHS
SUBPULMONARY EFFUSION
Ultrasound
Computed tomography
EMPYEMA
Chest radiography
Magnetic resonance imaging
An erect PA and lateral CXR should be performed in all
Pleural fluid returns low signal intensity on T1-weighted patients with suspected parapneumonic effusion or
images and a relatively high signal intensity on T2- empyema. If clinico-radiological doubt remains, then
weighted images (Figure 19.13). Magnetic resonance may either a lateral decubitus film or US may be required. In
be superior to CT in the differentiation of transudates and general, a pleural effusion is demonstrated often in associ-
exudates.78 Using a triple-echo pulse sequence and nor- ation with consolidation. The effusion is commonly uni-
malized MRI intensities, complex exudates have greater lateral, but if bilateral the infected side is usually the
signal intensity than simple exudates, which in turn are larger.80 If the effusion is uncomplicated, it behaves like
brighter than transudates.40,78 Magnetic resonance also other non-infected sterile effusions with a normal menis-
allows differentiation of pleural effusion from parenchy- cus sign and changes position and appearance when a
mal disease and from pleural tumor. Subacute and chronic decubitus film is performed, unless there has been prior
hemorrhage into the pleural space can also be recognized scarring of the pleural cavity.40
Empyema 241
If the effusion presents at a later stage or evolves from treatment outcome.74 As mentioned earlier, US may be
presentation, it may be loculated. This loculated effusion used to help determine the need for drainage and the use
will then have the characteristic appearance of a lentiform, of tube drainage or video assisted thoracoscopy.64,65
pleural-based opacity. If the collection is fissural it will
have the previously described appearance of a ‘pseudotu-
mor’. Suspicion that the patient may have a pulmonary Computed tomography
abscess rather than an empyema may arise either from the
clinical history or CXR.44 Although the differentiation may Critically ill patients and those who for other reasons are
be difficult, a pulmonary abscess tends to be smaller and unsuitable for assessment by US may be imaged by con-
rounder, with similar dimensions on both the PA and trast-enhanced CT. As with other exudative effusions,
lateral CXR;81 it forms acute angles with the chest wall and pleural enhancement is seen.75,76 Parietal pleural
has thicker, irregular walls if cavitation is present.82,83 CT enhancement and thickening is readily visualized,
may be used to confirm the diagnosis. although visceral pleural enhancement, often adjacent to
It was previously thought that empyema size on a CXR consolidated or atelectatic lung, may not be appreci-
correlated with the need for surgery.84 More recently, the ated.71,76,81 There is also an increase in thickening and
ability to predict outcome and requirement for surgery attenuation of the adjacent extraparietal pleural fat
assessed on the size of the effusion on CXR has been ques- (Figure 19.11).76 There is a trend for CT-detected pleural
tioned, perhaps owing to more aggressive medical man- thickening to increase with the stage of the effusion but
agement.74 it is unfortunately not possible to predict outcome in this
regard.74 It has also been shown that marked pleural
thickening or ‘pleural peel’ may completely resolve on
Ultrasound long-term follow-up.85
If a lung abscess is suspected from the CXR, then CT
If further imaging is required, US should be performed. may be used to confirm the diagnosis (Figure 19.14).43 As
The majority of parapneumonic effusions and empyemas on the CXR, pulmonary abscesses tend to be rounder than
requiring US will be septated and may be hyperechoic, empyemas and often make an acute angle with the chest
although anechoic effusions may be frank pus on aspira- wall compared with the obtuse angle commonly seen in an
tion.58,59 Whereas septations are readily visualized on US, empyema on CT.81 In addition, the ‘passive’ atelectasis
pleural thickening is poorly visualized and if confirma- seen adjacent to the empyema, as a result of reduction in
tion of this is required, contrast-enhanced CT should be hemithorax space, is not seen in patients with a pulmonary
performed.75,76 Unfortunately, there is no apparent corre- abscess as these replace lung rather than displace it.
lation between US appearance and the stage of evolution Pulmonary abscesses also tend to have significantly thicker
of the effusion, neither can it be used as a predictor of walls than empyemas.
(a) (b)
Figure 19.14 Characteristic pulmonary abscess on computed tomography forming an acute angle with the right chest wall. (a) The
abscess on mediastinal windows. (b) The abscess on lung windows. Note the absence of significant ‘passive’ atelectasis.
242 Radiology: diagnostic
In patients with a tuberculous empyema, the pleural As with parapneumonic effusions and empyema, malig-
appearance may be significantly different from those nant pleural effusions may be anechoic, echoic or complex
described previously. In this instance the typical findings septated.58 The detection of a pleural or diaphragmatic
are those of a moderate to large loculated pleural effusion nodule appears specific for malignancy93,94 and if US is
with pleural calcification and often enlargement of the used as an aid to thoracentesis, the assessment of the
overlying ribs.86 diaphragmatic pleural surface is of value since the major-
Computed tomography scanning shows a thick calcific ity of patients with malignant effusions have associated
pleural rind with rib thickening surrounding a loculated diaphragmatic pleural thickening that is detectable on US
pleural effusion.87 Complications associated with a (Figure 19.15).95 Pleural thickening outside the diaphrag-
tuberculous empyema include a bronchopleural fistula matic surface may be difficult to detect on US.
and empyema necessitans.88,89 The latter is seen on CT as a
thick-walled, well-encapsulated calcific pleural mass. The
differential diagnoses of this appearance include bacterial Computed tomography
empyema, lung abscess, blastomycosis and
actinomycosis.86 In patients with suspected malignant pleural effusion and
negative cytology on aspiration, contrast-enhanced CT is
the next most commonly performed investigation. As
with other causes of exudative effusions, pleural enhance-
MALIGNANT PLEURAL EFFUSIONS ment is commonly seen in patients with malignant effu-
sions.75,92 This is commonly nodular or irregular, or has a
Not all patients with pleural metastases have an associ- pleural thickness of >1 cm (Figure 19.16).95 Using these
ated pleural effusion. Imaging of pleural thickening criteria for assessment, contrast-enhanced CT has been
without an effusion is discussed later. The majority of shown to have a sensitivity of >80 per cent and to be
malignant pleural effusions (see Chapter 25, Effusions highly specific in the evaluation of patients with suspected
from malignancy) are secondary to lung or breast carci- malignant effusions.77,95 In addition, confirmation that
noma and these patients are likely to have an appropriate there is no apparent malignant cause for the effusion
clinical history and radiology.90,91 Most patients present- appears reliable with CT.77 Further information such as
ing with pleural metastases have large volume effusions, bone or liver metastases may also be apparent on scan-
with up to 10 percent of patients having massive effusions ning.77,90 The distribution of malignant pleural thickening
causing complete hemithorax opacification;92 the major- detected with CT appears to correlate with that seen at
ity of massive effusions are, in turn, secondary to malig- thoracoscopy. The majority of patients have thickening,
nancy.47 It should also be remembered that pleural which is either solely or maximally posteriorly and
effusions in patients with known malignancy may be basally. A secondary role for CT is the ability to either
non-malignant in etiology (i.e. associated with pul- directly or indirectly guide biopsy of a detected abnormal-
monary thromboembolic disease). In most instances ity (Figure 19.17).96
these effusions tend to be small and are often asympto-
matic.
If the primary malignancy is bronchogenic then the
effusion is likely to be ipsilateral, otherwise there is no ipsi-
lateral predilection and the effusions may on occasion be
bilateral.
Chest radiography
Figure 19.17 Computed tomography-guided pleural biopsy. The Positron emission CT is now being evaluated in patients
patient is in the prone position for biopsy. with suspected malignant pleural effusions.97–99 Un-
fortunately, it is likely to be a poor discriminator between
infective and malignant causes,100 and potentially in
patients with low-grade or relatively indolent malignancy
Magnetic resonance imaging such as fibrous mesothelioma. There may also be difficulty
in detecting small-volume disease such as ‘studding’ of the
The role of MRI in the investigation of suspected pleural pleural surface by tumor, commonly seen at thoracoscopy.
effusion is, at present, not clearly defined since there have
only been a few non-randomized studies. Because of its
very high contrast resolution, the ability to detect small CHYLOTHORAX AND PSEUDOCHYLOTHORAX
pleural nodules against the background of an effusion is
likely to be superior to contrast enhanced CT (Figure Chest radiography
19.18). However, this superior contrast resolution will be
offset by inferior spatial resolution and increased respira- Unless associated with lymphangioleiomyomatosis, chy-
tory and cardiac motion artifact. In addition to routinely lothoraces (see Chapter 29, Effusions from lymphatic dis-
using cardiac and respiratory gating, T2-weighted and ruptions) cannot be distinguished from other effusions on
post-contrast T1-weighted sagittal and axial sequences a standard radiograph. They can be unilateral or bilateral
should be performed. and can range in size from small to massive. A chylothorax
244 Radiology: diagnostic
Computed tomography
This technique uses 99mTc-sulphur microcolloid, anti- A PA and lateral CXR should always be performed as the
mony sulfide colloid, stannous phytate, rhenium sulfur initial investigation in detecting asbestos related disease
colloid, human serum albumin or dextran injected subcu- (see Chapter 40, Asbestos-related pleural diseases). The
taneously between the toes and scanning 2–6 hours later. criteria set by the International Labour Office105 suggest
Its use in investigating chylothorax has been limited to a that the sensitivity of a CXR in the detection of plaques
few case reports. It is able to demonstrate abnormal lym- ranges from 30 to 80 percent and the specificity ranges
phatic drainage in the cases of chylothorax.102 It may be from 60 to 80 percent. Their detection is dependent on the
possible to distinguish between lymphatic drainage and number, size, shape, position, degree of calcification and
transdiaphragmatic movement of chylous peritoneal fluid technical quality of the CXR (Figure 19.20).106
through the pleuroperitoneal canal by radioisotope migra- Pleural plaques almost always involve the parietal
tion speed. Overall, lymphoscintigraphy may be useful in pleura alone, although rarely they may involve the visceral
selecting those patients that would be suitable for surgery interlobular fissures, and may simulate a pulmonary
and to assess the effects of surgery.102 nodule.107 Oblique views are helpful in the diagnosis of the
plaques, although CT and HRCT are now more commonly
performed.108,109 En face imaging may cause problems
because of the difficulty in distinguishing between plaques
HEMOTHORAX and the normal muscle and fat companion shadows of the
chest wall.24 Extrapleural fat is suggested by a bilateral and
Hemothorax usually results from trauma, although several symmetric distribution along the midlateral chest wall.
other conditions may be causative.103 On a standard radi- Plaques are unilateral on a CXR in approximately 25
ograph the presence of rib fractures and a known clinical percent of cases, more frequently on the left, and if bilat-
history of trauma may help to distinguish a hemothorax eral are asymmetrical.110
from a pleural effusion. A CT scan may also show areas of As they calcify they produce a white line on the tangen-
active bleeding as foci of high attenuation (Figure tial views that parallels the chest wall, diaphragm or
19.19).104 Loculation is frequently seen and fibrin bodies cardiac border. If these calcifications are seen en face, they
may also form. The end stage of a hemothorax may result produce irregular linear or stippled calcifications with an
in massive pleural thickening. uneven density.
Focal pleural disease 245
The advent of multislice CT scanning has brought into Localized fibrous tumor of the pleura
question the use of HRCT in the detection of suspected
pleural plaques on the CXR or in differentiating plaques CHEST RADIOGRAPHY
from extrapleural fat shadows or other causes of
increased radio-opacity.111 Patients should be scanned Localized fibrous tumors are smooth rounded or oval
prone, enabling comment on possible pulmonary homogeneous masses on CXR (Figure 19.22). They have a
parenchymal asbestos-related disease that may also be sharply delineated contour that forms an obtuse angle with
present.112 Both HRCT and MSCT are far more sensitive the pleural surface in 74–94 percent of cases.24,91 They are
than the CXR in differentiating subpleural fat from seen more frequently in the lower half of the chest on
pleural plaques113,114 either side. The location may be in a fissure (30 percent),
Pleural plaques on MSCT and HRCT appear as circum- along the mediastinal pleura (18 percent), the thoracic
scribed areas of pleural thickening separated from the pleura (46 percent) or adjacent to the hemidiaphragm
underlying rib and extrapleural soft tissues by a thin layer (6 percent).91 Those on pedicles have a tendency to change
of fat (Figure 19.21). Any soft tissue internal to a rib position with respiration or posture.117
(remembering the comment describing the innermost
intercostal muscle) or paravertebral region is regarded as COMPUTED TOMOGRAPHY
abnormal.37 Both techniques are also able to detect small
foci of calcification within the plaques.115 Pleural plaques On unenhanced CT scans, localized fibrous tumors are
may be classified according to their CT appearance:116 homogeneous and sharply marginated, calcification is
rarely seen and, in larger tumors, the adjacent lung
1 minimal pleural plaques – less than 1 mm thick, parenchyma may be displaced with atelectasis and bowing
1–3 cm long, and few in number; of the bronchi and pulmonary vessels around the mass and
2 moderate pleural plaques – 1–3 mm thick, 2–5 cm long a smooth tapering margin at the junction of the mass with
and multiple; the pleura.117
3 severe pleural plaques – thicker than 3 mm, clearly Enhancement is variable after intravenous contrast; it is
indenting adjacent lung, up to 8 cm in craniocaudal always equal to or greater than that of other soft tissues,
dimension and extensive in width. and is most commonly homogeneous but may be hetero-
246 Radiology: diagnostic
Lymphoma
CHEST RADIOGRAPHY
(a)
CHEST RADIOGRAPHY
CHEST RADIOGRAPHY
Malignant mesothelioma
CHEST RADIOGRAPHY
pleural thickening in 90 percent, extending into the inter- MAGNETIC RESONANCE IMAGING
lobular fissures in 90 percent, pleural effusion in 70
percent, loss of volume of the affected hemithorax in 40 Along with CT, MRI allows clear visualization of the cir-
percent, pleural calcification in 20 percent and invasion of cumferential pleural thickening, fissural thickening,
the chest wall in up to 20 percent of patients.145 pleural effusion and diaphragmatic invasion. The full
The CT features suggestive of mesothelioma are similar extent of the tumor may be readily assessed on coronal and
to other causes of pleural malignancy described earlier. sagittal images. Mesothelioma returns an intermediate
However, the presence of calcification within the pleural signal on T1-weighted images and an increased signal rel-
thickening suggests a benign process, with 10 percent of ative to the chest wall musculature on T2-weighted
patients with mesothelioma having evidence of pleural cal- images;13 it may also show avid pleural enhancement on
cification compared with up to 50 per cent of patients with T1-weighted sequences post gadolinium, and this may be
benign pleural thickening.95,146 of value in differentiating benign from malignant
Although volume loss is often seen in mesothelioma it disease.13,141 Additional fissural and diaphragmatic inva-
is a relatively non-specific finding, since it is seen in both sion detected by MRI appears reliable in differentiating
benign and malignant disease.142,145 mesothelioma from other causes of malignant pleural
A benign pleural effusion (see Chapter 40, Asbestos- thickening.141 Although MRI offers no apparent advan-
related pleural diseases) may also occur in up to 3 percent tages over contrast-enhanced MSCT in staging
of the asbestos exposed population. The effusion is usually disease,91,152 when performed as a dynamic contrast
unilateral and other manifestations of asbestos-related enhanced examination it may be of value in predicting
disease may be present. Differentiation of a benign response to chemotherapy, although PET-CT may also be
asbestos effusion and asbestos-related mesothelioma can of value in this regard (see below).15
be difficult, but benign asbestos-related disease tends to be
a symmetrical process and, on CT, pleural involvement is
POSITRON EMISSION TOMOGRAPHY COMBINED WITH CT
bilateral in most patients.95 Also, as previously described,
malignant pleural disease tends to involve the entire Although PET-CT is not routinely used in the investiga-
pleural surface whereas benign disease does not involve the tion of mesothelioma, it has been shown to be of value in
mediastinal pleura.95 providing prognostic information and assessing response
Differentiation of mesothelioma from metastatic to chemotherapy in patients with mesothelioma (Figure
disease is limited radiologically. The presence of hilar 19.31).153 Tumours with a low standardized uptake value
adenopathy is more common in metastatic disease.95,147 are more likely to be epithelioid and to have a better prog-
True hilar involvement with no mediastinal involvement is nosis. A reduction in metabolic activity post chemother-
rare in metastatic pleural disease except for bronchogenic apy correlates with an increased time to progression and
carcinoma, lymphoma and renal cell carcinoma. increased survival. The pattern and measurement of meta-
Unfortunately, the tendency for mesothelioma to involve bolic activity may also correlate with the surgical stage. It
the inferior hemithorax appears non-specific.95 Rib may also be more accurate than the other imaging modal-
destruction and chest wall invasion are also good indica- ities in detecting distant metastatic disease.154 In a study of
tors of malignancy. Although rarely seen, infections such 29 patients, PET-CT detected additional metastatic disease
as actinomycosis, tuberculosis and nocardiosis may cause not identified on clinical examination and CT scanning in
rib destruction but often at a single site rather than the 7 of 29 patients.154
multiple sites often seen in malignancy.
Computed tomography not only has an important role
in diagnosis and staging but also in follow up.148 The
advent of new effective chemotherapeutic agents means Staging of mesothelioma
that the assessment of tumor response is now an important
criteria for both continuing chemotherapy and the evalua- Using the recent international TNM staging system pro-
tion of new therapies. Tumor response is assessed using posed by the International Mesothelioma Interest Group,
contrast enhanced CT and may be evaluated using either the CT and MR criteria for RESECTABILITY are as
the World Health Organisation (WHO)149 assessment cri- follows:
teria or now more commonly the Response Evaluation
Criteria in Solid Tumors (RECIST) guidelines.150,151 1 Preserved extrapleural fat planes.
Difficulties in accurate measurement of pleural thickening 2 Normal CT attenuation values and MR signal intensity
on CT, and in accurately measuring the same area of thick- characteristic of structures adjacent to the tumor.
ening on sequential scans, make the measurements less 3 Absence of extrapleural soft tissue masses.
reliable than in other organs. Interobserver variability is a 4 Smooth inferior diaphragmatic surface.
further confounding factor and the use of automated or
semi-automated software may produce more consistent The imaging criteria for UNRESECTABILITY are as
methods for measurement in the future.148 follows:
252 Radiology: diagnostic
(a) (b)
(c) (d)
Figure 19.31 (a) Positron emission tomography (PET) scan and (b) PET-computed tomography (CT) scan demonstrating high grade FDG
activity (arrowed) in sarcomatoid mesothelioma, with circumferential pleural thickening. (c) PET scan and (d) integrated PET-CT scan
demonstrating epithelioid mesothelioma (arrowed), with low grade fluorodeoxyglucose (FDG) activity, associated with pleural effusion.
(For (b) and (d), see also Color Plates 11 and 12.)
1 Tumor encasement of the diaphragm. inspiratory radiographs are recommended for the initial
2 Invasion of the extrapleural fat or soft tissues. examination. If, however, the pneumothorax is not defi-
3 Involvement of the ribs or bone destruction. nitely seen on the inspiratory film, then an expiratory film
4 Invasion of mediastinal structures. may help to confirm it,159 as the constant volume of the
pneumothorax is accentuated by an overall reduction in
Computed tomography and MR cannot easily distin- the size of the hemithorax on expiration.
guish between stages T1a, T1b and T2 as they cannot easily The radiographic diagnosis of a pneumothorax is based
distinguish the parietal and visceral pleura or detect on the identification of a visceral pleural line separated
diaphragmatic muscle invasion. They can, however, enable from the parietal pleura by a radiolucent airspace, demon-
distinction of stage 3 from stage 4.91,155,156 strated by the absence of pulmonary vessels beyond the
visceral line. Free air moves preferentially to the less
dependent pleural spaces, and in the upright patient tends
AIR to accumulate in an apicolateral location.24 The underlying
etiology of primary spontaneous pneumothorax is demon-
Pneumothorax strated in the minority (15 percent) of patients.160 The eti-
ology of secondary spontaneous pneumothorax is more
CHEST RADIOGRAPHY frequently demonstrated.
Curvilinear opacities projected over the chest from
The majority of pneumothoraces are identified on a stan- gowns, skinfolds or hair may mimic the visceral pleural
dard PA CXR taken on inspiration.157,158 Routinely, only surface. Following such a line outside the bony thorax is
Air 253
(a)
(a)
(b)
(b)
Figure 19.32 Left upper lobe bulla on chest radiography
(arrow), (a) readily differentiated from an apical left Figure 19.34 Large right hydropneumothorax identified on
pneumothorax with abnormal lung (arrow) by its concave chest radiography by the straight line of the air–fluid level
surface (b). (arrow).
254 Radiology: diagnostic
Bronchopleural fistula
ULTRASOUND
REFERENCES
● = Key primary paper
◆ = Major review article
patients with pleural disease has inevitably led to an pleural disease. Crit Rev Diagn Imaging 1997; 38: 1–58.
increase in their clinical use. The decreased cost of portable 10. Copley SJ, Well AU, Rubens MB, et al. Functional consequences of
pleural disease evaluated with chest radiography and CT.
US machines has led to an explosion in its use and a move Radiology 2001; 220: 237–43.
away from its provision by radiologists. The advent of ◆11. Knisely BL, Broderick LS, Kuhlman JE. MR imaging of the pleura
MSCT has also enabled CT scans to be performed on and chest wall. MRI Clin North Am 2000; 8: 125–41.
patients previously thought unfit to scan. The advent of ◆12. White CS. Magnetic resonance imaging of the chest. Respir Care
PET-CT has also opened the door to the use of PET scan- 2001; 46: 922–31.
◆13. Schmutz GR, Fisch-Ponsot C, Regent D, Sylvestre J. Computed
ning in patients with pleural malignancy. Increasingly tomography and magnetic resonance imaging of pleural masses.
these tests are being used not only in diagnosis and staging, Crit Rev Diagn Imaging 1993; 34: 309–83.
but also to guide therapeutic decisions and are being inves- ●14. Falaschi F, Battolla L, Mascalchi M, et al. Usefulness of MR signal
tigated as surrogate markers of function and morphology. intensity in distinguishing benign from malignant pleural disease.
It is perhaps their role as surrogate biomarkers of disease Am J Roentgenol 1996; 166: 963–8.
15. Giesel FL, Bischoff H, von Tengg-Kobligk H, et al. Dynamic
that is the next and most exciting step in their use. contrast-enhanced MRI of malignant pleural mesothelioma: A
feasibility study of noninvasive assessment, therapeutic follow-up,
and possible predictor of improved outcome. Chest 2006; 129:
1570–76.
KEY POINTS 16. Von Schulthess GK, Steinert HC, Hany TF. Integrated PET/CT:
Current applications and future directions. Radiology 2006; 238:
405–22.
● The majority of pleural effusions are reliably ◆17. Goldsmith SJ, Kostakoglu L. Nuclear medicine imaging of lung
diagnosed on CXR. cancer. Radiol Clin North Am 2000; 38: 511–24.
● Ultrasound may be used to confirm the presence 18. Duysinx B, Nguyen D, Louis R, et al. Evaluation of pleural disease
of a pleural effusion, assess features specific for with 18-fluorodeoxyglucose positron emission tomography
imaging. Chest 2004; 125: 489–493.
exudative effusions such as a septations and 19. Speckman JM, Gamsu G, Webb WR. Alterations in CT mediastinal
pleural nodularity, and to aid drain placement. anatomy produced by an azygos lobe. Am J Roentgenol 1981; 137:
● Multislice CT is accurate in the diagnosis of 47–50.
benign and malignant pleural thickening. 20. Godwin JD, Tarver RD. Accessory fissures of the lung. Am J
● Image-guided biopsy is of value in the diagnosis Roentgenol 1985; 144: 39–47.
21. Raasch BN, Carsky EW, Lane EJ, O’Callaghan JP, Heitzman ER.
of malignant pleural disease, including malignant Pleural effusion: explanation of some typical appearances. Am J
mesothelioma. Roentgenol 1982; 139: 899–904.
● Multislice CT is significantly more sensitive than 22. Proto AV, Ball JB. The superolateral major fissures. Am J
CXR in the detection of pneumothorax, and is of Roentgenol 1983; 140: 431–7.
particular use in trauma and intensive care 23. Austin JHM. The left minor fissure. Radiology 1986; 161: 433–6.
24. Fraser RG, Muller NL, Colman N, Pare PDL. Diagnosis of diseases of
patients. the chest, 4th edn. Philadelphia: WB Saunders Co, 1999.
● MSCT, DCEMRI and PET-CT may be used as 25. Armstrong P, Wilson AG, Dee P, et al. Imaging of diseases of the
correlates of function and morphology and act as chest, 3rd edn. London: Mosby, 2000.
surrogate biomarkers of disease processes. 26. Cimmono CV. The oesophageal–pleural stripe: an update.
Radiology 1981; 140: 607–13.
256 Radiology: diagnostic
27. Goodman TR, Traill ZC, Phillips AJ, Berger J, Gleeson FV. Ultrasound 55. Gryminski J, Krakowka P, Lypacewicz G. The diagnosis of pleural
detection of pneumothorax. Clin Radiol 1999; 54: 736–9. effusion by ultrasonic and radiologic techniques. Chest 1976; 70:
28. Sistrom CL, Reiheld CT, Gay SB, Wallace KK. Detection and 33–7.
estimation of the volume of pneumothorax using real-time 56. Lipscomb DJ, Flower CDR, Hadfield JW. Ultrasound of the pleura:
sonography: efficacy determined by receiver operating an assessment of its clinical value. Clin Radiol 1981; 32: 289–90.
characteristic analysis. Am J Roentgenol 1996; 166: 317–21. 57. Hirsch JH, Rogers JV, Mack LA. Real-time sonography of pleural
29. Lichtenstein DA, Menu Y. A bedside ultrasound sign ruling out opacities. Am J Roentgenol 1981; 136: 297–301.
pneumothorax in the critically ill. Lung sliding. Chest 1995; 108: 58. Yang P-C, Luh K-T, Chang D-B, Wu C-J, Kuo S-H. Value of
1345–8. sonography in determining the nature of pleural effusion: analysis
●30. Wernecke K, Galanski M, Peters PE, et al. Pneumothorax: of 320 cases. Am J Roentgenol 1992; 159: 29–33.
evaluation by ultrasound-preliminary results. Thorac Imaging 59. Lomas DJ, Padley SG, Flower CDR. The sonographic appearances of
1987; 2: 76–8. pleural fluid. Br J Radiol 1993; 66: 619–24.
31. Targhetta R, Bourgeois JM, Chavagneux R, et al. Ultrasonic signs 60. Chian CF, Su WL, Soh LH, et al. Echogenic swirling pattern as a
of pneumothorax: preliminary work. J Clin Ultrasound 1993; 21: predictor of malignant pleural effusions in patients with
245–50. malignancies. Chest 2004; 126: 129–34.
32. Ueki J, De Bruin PF, Pride NB. In vivo assessment of diaphragm 61. Marks WM, Filly RA, Callen PW. Real-time evaluation of pleural
contraction by ultrasound in normal subjects. Thorax 1995; 50: lesions: new observations regarding the probability of obtaining
1157–61. free fluid. Radiology 1982; 142: 163–4.
33. McCool FD, Benditt JO, Conomos P, et al. Variability of diaphragm 62. Rosenberg ER. Ultrasound in the assessment of pleural densities.
structure among healthy individuals. Am J Respir Crit Care Med Chest 1983; 84: 283–5.
1997; 155: 1323–8. 63. Tu CY, Hsu WH, Hsia TC, et al. Pleural effusions in febrile medical
34. Proto AV, Ball JB. Computed tomography of the major and minor ICU patients: Chest ultrasound study. Chest 2004; 126: 1274–80.
fissures. Am J Roentgenol 1983; 140: 439–48. 64. Brutsche MH, Tassi GF, Gyorik S, et al. Treatment of
35. Marks BW, Kuhns LR. Identification of the pleural fissures with sonographically stratified multiloculated thoracic empyema by
computed tomography. Radiology 1982; 143: 139–41. medical thoracoscopy. Chest 2005; 128: 3303–9.
36. Webb RW, Muller NL, Naidich DP. High resolution CT of the lung, 65. Wu RG, Yang PC, Kuo SH, Luh KT. ‘Fluid color’ sign: a useful
2nd edn. Philadelphia: Lippincott, Williams and Williams, 1996. indicator for discrimination between pleural thickening and
37. Im J-G, Webb WR, Rosen A, Gamsu G. Costal pleura: appearances pleural effusion. J Ultrasound Med 1995; 14: 767–9.
at high-resolution CT. Radiology 1989; 171: 125–31. 66. Wu RG, Yuan A, Liaw YS, et al. Image comparison of real-time
38. Vix VA. Extrapleural costal fat. Radiology 1974; 112: 563–5. gray-scale ultrasound and color doppler ultrasound for use in
39. Sargent EN, Boswell WD, Ralls PW, Markovitz A. Subpleural fat diagnosis of minimal pleural effusion. Am J Respir Crit Care Med
pads in patients exposed to asbestos: distinction from non- 1994; 150: 510–14.
calcified pleural plaques. Radiology 1984; 152: 273–7. 67. Mayo PH, Dolken P. Pleural ultrasonography. Clin Chest Med 2006;
◆40. Mueller NL. Imaging of the pleura. Radiology 1993; 186: 297–309. 27: 215–27.
41. Collins JD, Burwell D, Furmanski S, Lorber P, Steckel RJ. Minimal 68. Eibenberger KL, Dock WI, Ammann ME, et al. Quantification of
detectable pleural effusions. Radiology 1972; 105: 51–3. pleural effusions: sonography versus radiography. Radiology 1994;
42. Blackmore CC, Black WC, Dallas RV, Crow HC. Pleural fluid volume 191: 681–4.
estimation: a chest radiograph prediction rule. Acad Radiol 1996; 69. Roch A, Bojan M, Michelet P, et al. Usefulness of ultrasonography
3: 103–9. in predicting pleural effusions >500 mL in patients receiving
43. Davis S, Gardner Q. The shape of a pleural effusion. Br Med J mechanical ventilation. Chest 2005; 127: 224–32.
1963; 26: 436–7. 70. Zhu Z, Donnelly E, Dikensoy O, et al. Efficacy of ultrasound in the
◆44. Henschke CI, Davis SD, Romano PM, Yankelevitz DF. Pleural diagnosis of pleurodesis in rabbits. Chest 2005; 128: 934–39.
effusions: pathogenesis, radiologic evaluation, and therapy. J 71. Bressler EL, Francis IR, Glazer GM, Gross BH. Bolus contrast
Thorac Imag 1989; 4: 49–60. medium enhancement for distinguishing pleural from parenchymal
45. Oestreich AE, Haley C. Pleural effusion: the thorn sign. Chest lung disease: CT features. J Comput Assist Tomogr 1987; 11:
1981; 79: 365–6. 436–40.
◆46. Fleischner FG. Atypical arrangement of free pleural effusion. 72. Teplick JG, Teplick SK, Goodman L, Haskin ME. The interface sign: a
Radiol Clin North Am 1963; 1: 347–62. computed tomographic sign for distinguishing pleural and intra-
47. Maher GG, Berger HW. Massive pleural effusion: malignant and abdominal fluid. Radiology 1982; 144: 359–62.
non-malignant causes in 46 patients. Am Rev Respir Dis 1972; 73. Dwyer A. The displaced crus: a sign for distinguishing between
105: 458–60. pleural fluid and ascites on computed tomography. J Comput
48. Liberson M. Diagnostic significance of the mediastinal profile in Assist Tomogr 1978; 2: 598–9.
massive unilateral pleural effusions. Am Rev Respir Dis 1963; 88: 74. Kearney SE, Davies CWH, Davies RJO, Gleeson FV. Computed
176–80. tomography and ultrasound in parapneumonic effusions and
49. Mulvey RB. The effect of pleural fluid on the diaphragm. Am J empyema. Clin Radiol 2000; 55: 542–7.
Roentgenol 1965; 84: 1080–5. 75. Aquino SL, Webb WR, Gushiken BJ. Pleural exudates and
50. Yousef MMA. Case of the fall season. Semin Roentgenol 1980; 15: transudates: diagnosis with contrast-enhanced CT. Radiology
269–71. 1994; 192: 803–8.
51. Ruskin JA, Gurney JW, Thorsen MK, Goodman LR. Detection of 76. Waite RJ, Carbonneau RJ, Balikian JP, et al. Parietal pleural
pleural effusions on supine chest radiographs. Am J Roentgenol changes in empyema: appearances at CT. Radiology 1990; 175:
1987; 148: 681–3. 145–50.
52. Petersen JA. Recognition of infrapulmonary pleural effusion. 77. Traill ZC, Davies RJO, Gleeson FV. Thoracic computed tomography
Radiology 1960; 74: 34–41. in patients with suspected malignant pleural effusions. Clin Radiol
53. Schwartz MI, Marmorstein BL. A new radiologic sign of 2001; 56: 193–6.
subpulmonic effusion. Chest 1975; 67: 176–8. 78. Davis SD, Henschke CI, Yankelevitz DF, Cahill PT, Yi Y. MR imaging
54. Eibenberger KL, Dock WI, Ammann ME, et al. Quantification of of pleural effusions. J Comput Assist Tomogr 1990; 14: 192–8.
pleural effusions: sonography versus radiography. Radiology 1994; ◆79. McLoud TC. CT and MR in pleural disease. Clin Chest Med 1998;
191: 681–4. 19: 261–76.
References 257
80. Hanna JW, Reed JC, Choplin RH. Pleural infections: a ◆106. Peacock C, Copley SJ, Hansell DM. Asbestos-related benign pleural
clinical–radiologic review. J Thorac Imaging 1991; 6: 68–79. disease. Clin Radiol 2000; 55: 422–32.
81. Stark DD, Federle MP, Goodman PC, Podrasky AE, Webb WR. 107. Rockoff SD, Kagan E, Schwartz A, et al. Visceral pleural thickening
Differentiating lung abscess and empyema: radiography and in asbestos exposure: the occurrence and implications of
computed tomography. Am J Roentgenol 1983; 141: 163–7. thickened interlobar fissures. J Thorac Imaging 1987; 2: 58–66.
82. Friedman PJ, Hellekant AG. Radiologic recognition of ◆108. McLoud TC. Conventional radiography in the diagnosis of
bronchopleural fistula. Radiology 1977; 124: 289–95. asbestos-related disease. Radiol Clin North Am 1992; 30:
83. Hsu JT, Bennett GM, Wolff E. Radiologic assessment of 1177–89.
bronchopleural fistula with empyema. Radiology 1972; 103: 41–5. 109. Sargent EN, Gordonson J, Jacobson G, Birnbaum W, Shaub M.
84. Ferguson AD, Prescott RJ, Selkon JB, Watson D, Swinburn CR. Bilateral pleural thickening: a manifestation of asbestos dust
Empyema subcommittee of the Research Committee of the British exposure. Am J Roentgenol 1978; 131: 579–85.
Thoracic Society. The clinical course and management of thoracic 110. Proto AV. Conventional chest radiographs: anatomic
empyema. QJM 1996; 89: 285–9. understanding of newer observations. Radiology 1992; 183:
85. Neff CC, van Sonnenberg E, Lawson DW, Patton AS. CT follow-up 593–603.
of empyemas: pleural peels resolve after percutaneous catheter 111. Remy-Jardin M, Sobaszek A, Duhamel A, et al. Asbestos-related
drainage. Radiology 1990; 176: 195–7. pleuropulmonary diseases: Evaluation with low-dose four-detector
86. Sahn SA, Iseman MD. Tuberculous empyema. Semin Respir Infect row spiral CT. Radiology 2004; 233: 182–90.
1999; 14: 82–7. ◆112. Staples CA. Computed tomography in the evaluation of benign
87. Hulnick DH, Naidich DP, McCauley DI. Pleural tuberculosis asbestos-related disorders. Radiol Clin North Am 1992; 30:
evaluated by computed tomography. Radiology 1983; 149: 1191–207.
759–65. 113. Friedman AC, Fiel SB, Fisher MS, et al. Asbestos-related pleural
88. Bhatt GM, Austin HM. CT demonstration of empyema necessitatis. disease and asbestosis: a comparison of CT and chest radiography.
J Comp Assist Tomogr 1985; 9: 1108–9. Am J Roentgenol 1988; 150: 269–75.
89. Glicklich M, Mendelson DS, Gendal ES, et al. Tuberculous empyema 114. Aberle DR, Gamsu G, Roy CS. High resolution CT of benign
necessitatis. Computed tomography findings. Clin Imaging 1990; asbestos-related disorders: clinical and radiographic correlation.
14: 23–5. Am J Roentgenol 1988; 151: 883–91.
90. American Thoracic Society. Management of malignant pleural ◆115. Gefter WB, Epstein DM, Miller WT. Radiographic evaluation of
effusions. Am J Respir Crit Care Med 2000; 162: 1987–2001. asbestos-related chest disorders. Crit Rev Diagn Imaging 1984; 21:
◆91. Bonomo L, Feragalli B, Sacco R, Merlino B, Storto ML. Malignant 133–81.
pleural disease. Eur J Radiol 2000; 34: 98–118. ◆116. Lynch DA, Gamsu G, Aberle DR. Conventional and high resolution
◆92. Sahn SA. Pleural diseases related to metastatic malignancies. Eur computed tomography in the diagnosis of asbestos-related
Respir J 1997; 10: 1907–13. diseases. Radiographics 1989; 9: 523–51.
93. Goerg C, Schwerk WB, Goerg K, Walters E. Pleural effusion: an 117. Dedrick CG, McLoud TC, Shepard JO, Shipley RT. Computed
‘acoustic window’ for sonography of pleural metastases. J Clin tomography of localized pleural mesothelioma. Am J Roentgenol
Ultrasound 1991; 19: 93–7. 1985; 144: 275–80.
◆94. Goerg C, Restrepo I, Schwerk WB. Sonography of malignant 118. Mendelson DS, Meary E, Buy JN, Pigeau I, Kirschner PA. Localized
pleural effusion. Eur Radiol 1997; 7: 1195–8. fibrous pleural mesothelioma: CT findings. Clin Imaging 1991; 15:
●95. Leung AN, Muller NL, Miller RR. CT in differential diagnosis of 105–8.
diffuse pleural disease. Am J Roentgenol 1990; 154: 487–92. 119. Lee KS, Im JG, Choe KO, Kim CJ, Lee BH. CT findings in benign
96. Adams R, Gleeson FV. Percutaneous image guided cutting needle fibrous mesothelioma of the pleura: pathologic correlation in nine
biopsy of the pleura in the presence of a suspected malignant patients. Am J Roentgenol 1992; 158: 983–6.
effusion. Radiology 2001; 219: 510–14. 120. Saifuddin A, Da Costa P, Chalmers AG, Carey BM, Robertson RJH.
97. Bury TH, Paulus P, Dowlati A, et al. Evaluation of pleural diseases Primary malignant localized fibrous tumors of the pleura: clinical,
with FDG-PET imaging – preliminary report. Thorax 1997; 52: radiological, and pathological features. Clin Radiol 1992; 45:
187–9. 13–17.
98. Schaffler GJ, Wolf G, Schoellnast H, et al. Non-small cell lung ◆121. Dynes MC, White EM, Fry WA, Ghahremani GG. Imaging
cancer: Evaluation of pleural abnormalities on CT scans with 18F manifestations of pleural tumors. Radiographics 1992; 12:
FDG PET. Radiology 2004; 231: 858–65. 1191–201.
99. Gupta NC, Rogers JS, Graeber GM, et al. Clinical role of F-18 122. Epler GR, McLoud TC, Munn CS, Colby TV. Pleural lipoma: diagnosis
fluorodeoxyglucose positron emission tomography imaging in by computed tomography. Chest 1986; 90: 265–8.
patients with lung cancer and suspected malignant pleural 123. Filly R, Blank N, Castellino RA. Radiographic distribution of
effusion. Chest 2002; 122: 1918–24. intrathoracic disease in previously untreated patients with
100. Erasmus JJ, McAdams HP, Rossi SE, et al. FDG PET of pleural Hodgkin’s disease and non-Hodgkin’s lymphoma. Radiology 1976;
effusions in patients with non-small cell lung cancer. Am J 120: 277–81.
Roentgenol 2000; 175: 245–9. 124. Burgener FH, Hamlin DJ. Intrathoracic histiocytic lymphoma. Am J
101. Hillerdal G. Chylothorax and pseudochylothorax. Eur Respir J Roentgenol 1981; 136: 499–504.
1997; 10: 1157–62. 125. Balikian JP, Herman PG. Non-Hodgkin lymphoma of the lungs.
102. Pui MH, Yueh T-C. Lymphoscintigraphy in chyluria, Radiology 1979; 132: 569–76.
chyloperitoneum and chylothorax. J Nucl Med 1998; 39: 1292–6. 126. Castellino RA, Blank N, Hoppe RT, et al. Hodgkin disease:
103. Groskin SA. Selected topics in chest trauma. Radiology 1992; 183: contributions of chest CT in initial staging evaluation. Radiology
605–17. 1986; 160: 603–5.
104. Wolverson MK, Crepps LF, Sundaram M, et al. Hyperdensity of 127. Shuman LS, Libshitz HI. Solid pleural manifestations of lymphoma.
recurrent haemorrhage at body computed tomography: incidence Am J Roentgenol 1984; 142: 269–73.
and morphologic variation. Radiology 1983; 148: 779–84. 128. Okada F, Ando Y, Kondo Y, et al. Thoracic CT Findings of Adult T-
105. International Labour Office. Guidelines for the use of the ILO Cell Leukaemia or Lymphoma. Am J Roentgenol 2006; 182: 761–7.
international classification of radiographs of pneumoconiosis. 129. Carlsen SE, Bergin CJ, Hoppe RT. MR imaging to detect chest wall
Revised edition. Occupational Health and Safety Series, No. 22 and pleural involvement in patients with lymphoma: effect on
(Rev. 80). Geneva: International Labour Office,1980. radiation therapy planning. Am J Roentgenol 1993; 160: 1191–5.
258 Radiology: diagnostic
130. Negendank WG, Alkatib AM, Karanes C, Smith MR. Lymphomas: 153. Gerbaudo VH, Britz-Cunningam S, Sugarbaker DJ, Treves ST.
MR imaging contrast characteristics with clinico-pathologic Metabolic significance of the pattern, intensity and kinetics of
correlations. Radiology 1990; 177: 209–16. 18F-FDG uptake in malignant pleural mesothelioma. Thorax 2003;
131. McLoud TC, Woods BO, Carrington CB, Epler GR, Gaensler EA. 58: 1077–82.
Diffuse pleural thickening in an asbestos-exposed population: 154. Erasmus JJ, Truong MT, Smythe WR, et al. Integrated computed
prevalence and causes. Am J Roentgenol 1985; 144: 9–18. tomography-positron emission tomography in patients with
132. Copley SJ, Lee YCG, Hansell DM, et al. Asbestos-induced and potentially respectable malignant pleural mesothelioma: Staging
smoking-related disease: Apportioning pulmonary function deficit implications. J Thorac Cardiovasc Surg 2005; 129: 1364–70.
by using thin-section CT. Radiology 2007; 242: 258–66. 155. Heelan RT, Rusch VW, Begg CB, et al. Staging of malignant pleural
133. Murray JG, Patz EF Jr, Eramus JJ, Gilkeson RC. CT appearance of mesothelioma: comparison of CT and MR imaging. Am J
the pleural space after talc pleurodesis. Am J Roentgenol 1997; Roentgenol 1999; 172: 1039–46.
169: 89–91. 156. Patz EF, Rusch VW, Heelan R. The proposed new international TNM
134. Murray JG, Eramus JJ, Bahtiarian EA, Goodman PC. Talc staging system for malignant pleural mesothelioma: application to
pleurodesis simulating pleural metastases on 18F-fluoro- imaging. Am J Roentgenol 1996; 166: 323–7.
deoxygenase positron emission tomography. Am J Roentgenol 157. Bradley M, Williams C, Walshaw MJ. The value of routine
1997; 168: 359–60. expiratory films in the diagnosis of pneumothorax. Arch Emerg
135. Kwek BH, Aquino SL, Fischman AJ. Fluorodeoxyglucose positron Med 1991; 8: 115–16.
emission tomography and CT after talc pleurodesis. Chest 2004; 158. Bungay HK, Berger J, Traill ZC, Gleeson FV. Pneumothorax post
125: 2356–60. CT–guided lung biopsy: a comparison between detection on chest
136. Renner RR, Pernice NJ. The apical cap. Semin Roentgenol 1977; radiographs and CT. Br J Radiology 1999; 72: 1160–3.
12: 299–302. 159. Seow A, Kazerooni EA, Cascade PN, et al. Comparison of upright
137. Im JG, Webb WR, Han MC, et al. Apical opacity associated with inspiratory and expiratory chest radiographs for detecting
pulmonary tuberculosis: high resolution CT findings. Radiology pneumothoraces. Am J Roentgenol 1996; 166: 313–16.
1991; 178: 727–31. 160. Killen DA, Gobbel WG. Spontaneous pneumothorax. Boston: Little,
138. McLoud TC, Isler RJ, Novelline RA, et al. The apical cap, review. Am Brown & Company, 1968.
J Roentgenol 1981; 137: 299–306. 161. Fisher JK. Skinfold versus pneumothorax. Am J Roentgenol 1978;
139. Miller BH, Rosado-de-Christenson ML, Mason AC, et al. Malignant 130: 791–2.
pleural mesothelioma: radiologic–pathologic correlation. 162. Lane EJ, Proto AV, Phillips JW. Machbands and density perception.
Radiographics 1996; 16: 613–44. Radiology 1976; 121: 9–17.
140. Scott EM, Marshall TJ, Flower CD, Stewart S. Diffuse pleural 163. Greene R, McLoud TC, Stark P. Pneumothorax. Semin Roentgenol
thickening: percutaneous CT-guided cutting needle biopsy. 1977; 12: 313–25.
Radiology 1995; 194: 867–70. 164. Glazer HS, Anderson DJ, Wilson BS, et al. Pneumothorax:
141. Knuuttila A, Kivisaari L, Kivisaari A, et al. Evaluation of pleural appearance on lateral chest radiographs. Radiology 1989; 173:
disease using MR and CT with special reference to malignant 701–11.
pleural mesothelioma. Acta Radiol 2001; 42: 502–7. 165. Tocino IM, Miller MH, Frederick PR, Bahr AL, Thomas F. CT
142. Ng CS, Munden RF, Libshitz HI. Malignant pleural mesothelioma: detection of occult pneumothorax in head trauma. Am J
the spectrum of manifestations on CT in 70 cases. Clin Radiol Roentgenol 1984; 143: 987–90.
1999; 54: 415–21. 166. Wall SD, Federle MP, Jeffrey RB, Brett CM. CT diagnosis of
143. Boylan AM. Mesothelioma: new concepts in diagnosis and unsuspected pneumothorax after blunt abdominal trauma. Am J
management. Curr Opin Pulm Med 2000; 6: 157–63. Roentgenol 1983; 141: 919–21.
◆144. Astoul P. Pleural mesothelioma. Curr Opin Pulm Med 1999; 5: 167. Tocino IM, Miller MH, Fairfax WR. Distribution of pneumothorax in
259–68. the supine and semirecumbent critically ill adult. Am J Roentgenol
145. Kawashima A, Libshitz HI. Malignant pleural mesothelioma: CT 1985; 144: 901–5.
manifestations in 50 cases. Am J Roentgenol 1990; 155: 965–9. 168. Rhea JT, vanSonnenberg E, McLoud TC. Basilar pneumothorax in
146. Hierholzer J, Luo L, Bittner RC, et al. MRI and CT in the differential the supine adult. Radiology 1979; 133: 593–5.
diagnosis of pleural disease. Chest 2000; 118: 604–9. 169. Ziter FMH, Westcott JL. Supine subpulmonary pneumothorax. Am J
147. Adams VI, Unni KK, Muhm JR, et al. Diffuse malignant Roentgenol 1981; 137: 699–701.
mesothelioma of pleura: diagnosis and survival in 92 cases. Cancer 170. Moskowitz PS, Griscom NT. The medial pneumothorax. Radiology
1986; 58: 1540–51. 1976; 120: 143–7.
148. Armato III SG, Ogarek JL, Starkey A, et al. Variability in 171. Gordon R. The deep sulcus sign. Radiology 1980; 13: 25–7.
mesothelioma tumor response classification. Am J Roentgenol ◆172. Gallardo X, Castaner E, Mata JM. Benign pleural diseases. Eur J
2006;186: 1000–6. Radiol 2000; 34: 87–97.
149. Miller AB, Hogestraeten B, Staquet M, Winkler A. Reporting results 173. Slater A, Goodwin M, Anderson K, Gleeson FV. COPD can mimic
of cancer treatment. Cancer 1981; 47: 207–14. the appearances of pneumothorax on ultrasound. Chest 2006;
150. James K, Eisenhauer EA, Christian M, et al. Measuring response in 129: 545–50.
solid tumors: Unidimensional versus bidimensional measurement. J 174. vanSonnenberg E, Casola G, Ho M, et al. Difficult thoracic lesions:
Natl Cancer Inst 1999; 91: 523–8. CT-guided biopsy experience in 150 cases. Radiology 1988; 167:
151. Therasse P, Arbuck SG, Eisenhauer EA, et al. New guidelines to 457–61.
evaluate the response to treatment in solid tumors. J Natl Cancer 175. Westcott JL, Valpe JP. Peripheral bronchopleural fistula: CT
Inst 2000; 92: 205–16. evaluation in 20 patients with pneumonia, empyema, or post
152. Patz EF, Shaffer K, Piwnica-Worms DR, et al. Malignant pleural operative air leak. Radiology 1995; 196: 175–91.
mesothelioma: value of CT and MR imaging in predicting 176. Zelefski MN, Freeman LM, Stern H. A simple approach to the
resectability. Am J Roentgenol 1992; 159: 961–6. diagnosis of bronchopleural fistula. Radiology 1977; 124: 843–4.
20
Radiology: interventional
identified. Only brief periods of radiation exposure are fluid sent for analysis. The yield of cytology for malignant
needed for imaging localization and guidance for tube pleural effusions also improves with repeated pleural fluid
placement. cytologic specimens and when combined with a percuta-
Additional factors influencing the choice of imaging neous pleural biopsy.9
modality used to assist pleural interventional procedures
include the type, size, position and accessibility of the
pleural abnormality. In most practices, ultrasound is
usually used to assist the placement of a drainage catheter TUBE THORACOSTOMY
into uncomplicated effusions. CT is preferred for small
loculated or medially placed fluid collections or for biopsy Catheter thoracostomy is usually used to manage patients
of pleural masses. Fluoroscopy is used for pneumothorax with malignant pleural effusions, parapneumonic effu-
and chest tube placement. sions or empyemas, and pneumothoraces.
(a)
(b)
until fluid drainage ceases. Similar to the approach prevent it kinking and guide it to the correct location under
described above for pigtail catheter insertion, a guidewire is fluoroscopic guidance. Pleural fluid is then drained period-
placed into the pleural fluid, a 1-cm incision is made ically into 600 mL vacuum bottles by placing the firm
approximately 5–8 cm posterior to the entrance site of the dilator on the preconnected tube into the valve at the distal
guidewire and a subcutaneous chest wall tunnel is created. end of the Pleurx catheter.
The catheter is pulled through this tunnel to the entrance In patients with long-term indwelling catheters, the
skin incision of the guidewire. The wire tract is then dilated degree of symptomatic improvement in dyspnea is compa-
with a Teflon ‘peel away’ sheath. These tubes are soft and rable to pleurodesis.66,67,70–73 Additionally, in patients with
may kink during placement. The advantage of using trapped lung syndrome, most have symptomatic benefit
imaging guidance for their placement is to gain control with a decrease in dyspnea and improved exercise toler-
over the tube by inserting a guidewire or a dilator through ance.68,70 Late recurrence of effusions is uncommon (<15
one of the side holes of the indwelling catheter to stiffen it, percent) and approximately 50 percent of patients will
Tube thoracostomy 263
have spontaneous pleurodesis, usually within a month.70 Traditionally, complicated parapneumonic effusions
Tube malfunction is uncommon and the catheters usually and empyemas have been treated by large-bore catheter
remain in place until the patient’s demise (mean duration thoracostomy82–87 followed, if needed, by rib resection
56–115 days).66,68,72 However, because of their long-term and open drainage,88–91 video-assisted thoracoscopic
use, catheter infection with localized skin breakdown as surgery92,93 or decortication.75,94,95 However, studies have
well as tumor seeding along the tract has been reported as shown that these collections may be adequately treated by
a complication.68,70 imaging-assisted placement of small-bore catheters
(8–16 F) rather than using the standard 22–34 F catheters.
In fact, the insertion of small-bore catheters under
Parapneumonic effusion and empyema imaging-assisted guidance (CT or sonography), is strongly
recommended: the procedure is safe, whereas conven-
Patients with parapneumonic effusions have a high risk of tional trocar placement can be hazardous,96 the mean
morbidity and mortality and approximately 15–40 percent duration of approximately 6 days for image-guided thora-
require drainage and prolonged hospitalization.74,75 There costomy is shorter than for traditional catheter thoracos-
are sequential stages in the evolution of a parapneumonic tomy,32,97–99 and the success rate is higher than traditional
effusion and development of an empyema and the clinical catheter thoracostomy (81 versus 47 percent, respec-
determination as to which pleural effusions require tively)100,101 (Figure 20.2). In our experience, small-bore
drainage to avoid progression to an empyema can be diffi- catheter placement under image guidance is highly effec-
cult.76–80 To assist in this decision, the American College of tive for the treatment of parapneumonic effusions. These
Chest Physicians (ACCP) has published recommendations patients are always treated with intravenous antibiotics
on the clinical management of patients with parapneu- and typically show clinical improvement within 24–48
monic effusions based on the categorization of the effusion hours. If patients do not respond to therapy within several
according to size, chemistry and bacteriology.76 days then management changes including alternative
Thoracentesis and the categorization of parapneu- antibiotics, additional catheters, and sometimes surgical
monic effusions is recommended in all patients except if drainage may be required.
the effusion is minimal (<1 cm on decubitus radiograph)
and free-flowing (Category 1). Category 2 (small to mod- TECHNIQUE
erate [>1 cm less than 50 percent of hemithorax], free-
flowing effusions with pleural pH >7.2, glucose >60 The catheter insertion site is determined by the location of
mg/dL, lactate dehydrogenase [LDH] <3× the upper limit the effusion. If there are multiple loculations, more than
of serum and Gram stain and culture negative) usually do one drainage catheter may be required. CT is preferred
not require drainage as most patients respond to antibiotic over ultrasound to more accurately assess the location and
therapy. Category 3 (large [>50 percent of hemithorax], extent of loculated fluid. Additionally, CT better demon-
free-flowing effusions, loculated effusions or effusions strates the presence of a thick pleural rind that can pre-
with thickened parietal pleura and/or pleural pH <7.2, clude chest catheter placement as an initial treatment
and/or pleural glucose <60 mg/dL, and/or Gram stain and option and other anatomic abnormalities within the
culture positive) and Category 4 (frank pus in the pleural thorax. Following insertion, the catheter is connected via a
space), require drainage. three-way stopcock to an underwater drainage system
In a retrospective study, the value of the ACCP recom- such as a Pleur-evac (DeKnatel Division, Pfizer Hospital
mendations as well as the British Thoracic Society guide- Products Group, Fall River, MA, USA) with continuous
lines to predict which patients with non-purulent wall suction at 20–30 cmH2O. To prevent catheter occlu-
parapneumonic effusions warranted chest tube drainage sion, the catheter is flushed with 10 mL of normal saline
were evaluated.80 A pleural fluid pH <7.20, a pleural every eight hours.
glucose <60 mg/dL or a large pleural effusion size pre- Patients with a persistent complex pleural collection or
dicted with a likelihood of >5 ratio that the effusion would a decrease in fluid drainage but radiographically evident
not resolve unless drainage was instituted. Overall, the collection may benefit from the installation of a fibrinolytic
ACCP guidelines had a sensitivity of 97 percent and speci- agent into the pleural space through the drainage catheter.
ficity of 56 percent in distinguishing which non-purulent Specifically, intrapleural fibrinolytics, if used early in the
parapneumonic effusion warranted drainage. In con- fibrinopurulent stage of a parapneumonic effusion, may
tradistinction, there is no correlation between the CT and dissolve fibrinous adhesions, improve pleural fluid
sonographic appearance of parapneumonic effusions and drainage, prevent pleural fluid loculation and decrease the
the severity of infection as determined by established rate of surgical intervention and length of hospitalization.
microbiological and biochemical indicators.81 Further- The rational for this management, which is included in the
more, neither imaging modality can predict those patients guidelines of the ACCP, is based on small studies although
who will subsequently require surgical intervention after clinical evidence from high-quality trials is not avail-
failed management by chest tube drainage and intrapleural able.76,102–105 The consensus in favor of the use of
fibrinolytics.81 intrapleural fibrinolytics in the management of complex
264 Radiology: interventional
(a)
(b)
(c)
Figure 20.2 Forty-two-year-old woman presents with cough
and fever. (a) Frontal chest radiograph demonstrates a poorly
defined opacity in the left base. (b) Lateral radiograph suggests
this posterior opacity is in the pleural space. (c) Axial computed (d)
tomography image confirms a posterior fluid collection.
(d) Follow-up chest radiograph 2 days after the pig-tail chest
tube shows almost complete resolution of the infected collection.
parapneumonic pleural effusions is, to a large extent, based saline.104 In contradistinction, a recent randomized trial of
on increased drainage volume/radiologic improvement 454 patients by Maskell et al.107 showed that the use of
although the effect on duration of hospitalization, need for intrapleural fibrinolytics did not affect length of hospital-
surgery and long-term clinical outcome is controver- ization, rate of surgery or mortality.
sial.104–107 In a recent small randomized study by Diacon et Although the overall available evidence suggests that
al.,104 9 percent of patients with empyemas who received fibrinolytics are useful in patients with parapneumonic
intrapleural streptokinase required surgical drainage com- effusions, the choice of fibrinolytic agent remains contro-
pared with 45 percent of patients who received intrapleural versial. Different agents available for instillation into the
Tube thoracostomy 265
pleural space include streptokinase, urokinase, alteplase confirm complete lung re-expansion. The catheter is
and tissue plasminogen activator (tPA). Streptokinase is removed when the output is less than 50 cm3 a day, the
more commonly used because it is much cheaper than the patient is afebrile, leukocytosis has resolved and the patient
other agents but it may be antigenic and can cause fever in is ready to be placed on oral antibiotics.
up to 28 percent of patients. Urokinase is non-antigenic,
does not usually cause fever and may be slightly more
effective than streptokinase. Although temporarily with- Pneumothorax
drawn from the USA because of the associated risks of viral
disease transmission, urokinase is now available for clini- Imaging can be used to place a chest catheter in patients
cal use. Alteplase and tPA are less commonly used, have with a symptomatic pneumothorax, or in high-risk
been proven to be effective but are more expensive than patients with a small, but expanding pneumothorax. It is
streptokinase. In addition, some preliminary results very useful in patients with lung or pleural disease and a
suggest that tPA may be associated with an increased risk loculated pneumothorax that requires a chest tube.
of intrapleural bleeding in patients simultaneously receiv- Although pneumothoraces are not uncommon after
ing therapeutic levels of systemic anticoagulation transthoracic needle aspiration biopsies of the lung (up to
When streptokinase is chosen as the fibrinolytic agent, 61 percent),108–118 or transbronchial biopsy, chest catheter
it is common to begin by instilling 250 000 U of streptoki- placement is typically only performed when patients are
nase mixed in 100 cm3 of normal saline through the tube symptomatic or when the pneumothorax is increasing in
directly into the pleural space. The catheter is then closed size96,112,119,120 (Figure 20.3). Success rates of small-bore
for 2 hours to allow the agent to distribute throughout the catheters (7–10 F) is high (87–97 percent) with most pneu-
pleural space and to dissolve the fibrin membranes. It is mothoraces resolving within 24–72 hours following
then reopened to continuous suction. Instillation is catheter insertion.112,121,122 The treatment of a pneumo-
repeated on a daily basis as needed, sometimes up to five thorax with a small-bore catheter has a similar success rate
consecutive days if the results improve drainage. Patients to that of large-bore chest tube drainage. Lack of response
are evaluated daily to assure proper catheter function and is usually due to technical factors (catheter malposition
to record catheter output. Daily chest radiographs are not and occlusion) or a very large air leak.112,121,123,124
required when drainage continues without difficulty and
the patient’s fever and clinical symptoms continue to TECHNIQUE
improve. When drainage is minimal or ceases and the
patient’s temperature has returned to normal, a chest radi- Imaging-assisted guidance is chosen according to avail-
ograph is performed to exclude loculated fluid, and to ability and the clinical scenario: if the pneumothorax
(a) (b)
Figure 20.3 Fifty-seven-year-old man with a right upper lobe nodule presents for percutaneous biopsy. (a) Chest film immediately
following the biopsy shows the poorly defined right nodule (curved arrow) and a right pneumothorax (straight arrow). (b) Follow-up chest
film after the pig-tail catheter shows almost complete resolution of the pneumothorax.
266 Radiology: interventional
occurs during a biopsy, a chest catheter is inserted using require chemical pleurodesis using sclerosing agents such
the imaging modality used to perform the biopsy, while as doxycycline, talc or bleomycin, or placement of a large-
those that occur later are usually performed using fluoro- bore tube may be necessary.127–130
scopic guidance. CT guidance is used for small or loculated
pneumothoraces in ventilated patient as it helps to prevent
intraparenchymal catheter placement. The entry site PLEURAL BIOPSY
depends on the location and size of the pneumothorax.
The catheter is usually inserted in the sixth intercostal Infections, inflammation and neoplasms of the pleura can
space in the region of the mid-axillary line and the tip manifest as diffuse or focal pleural abnormalities. When
positioned in the upper, anterior aspect of the hemitho- there is an accompanying effusion, thoracentesis may be
rax.125 However, there is a tendency for the tip of small- diagnostic. However, a pleural biopsy is often required to
bore catheters to be displaced posteriorly with this establish a diagnosis as the yield from thoracentesis in
approach. Small-bore catheters can also be placed through some diseases, including tuberculosis or malignancy, is less
the third or fourth anterior intercostal space. This place- than optimal.131 In some cases, cytological evaluation of
ment results in more constant positioning of the catheter specimens obtained from transthoracic needle aspiration
tip in the apex of the pleural space.112,126 may be ambiguous, such as seen with benign disease, or it
The choice of drainage catheter depends on physician may be difficult to distinguish reactive mesothelial cell
preference and whether there is fluid within the pleural hyperplasia from metastatic adenocarcinoma and
space. An 8–10 F straight catheter can be used for rapid mesothelioma.132–135 Thus, depending on the clinical sce-
placement in a single-step procedure but is usually not rec- nario, a core-needle biopsy that can obtain larger tissue
ommended if there is an accompanying pleural effusion. samples may be required.133 These larger core specimens
The Seldinger technique can be used to place a guide-wire can be obtained using Cope or Abrams biopsy needles
into the pleural space prior to placement of an 8–10 F without imaging-assisted guidance, although imaging-
pigtail catheter. Although this prolongs the procedure by a assisted biopsy has a higher diagnostic yield. For instance,
few minutes, using a pigtail catheter is advantageous as it by using ultrasound guidance, the confirmation of tuber-
effectively drains any accompanying pleural fluid and culous pleural involvement increases from 20 to 86
rarely dislodges (a common problem of the straight percent when compared with procedures performed
catheter).124 without imaging-assisted guidance. Similarly, for patients
The catheter/trocar system is angled in a cranial direc- with either primary or metastatic pleural disease, the use of
tion adjacent to the superior aspect of the rib to avoid ultrasound or CT increases the diagnostic rate.136,137 In
injury to the intercostal neurovascular bundle and rare cases, additional tissue is needed for special pathology
advanced into the pleural space. A gush of air occurs when stains and analysis, and video-assisted thoracoscopic
the catheters tip transverses the parietal pleura. The trocar surgery (VATS) may be necessary.138 In one comparative
is then held in position and the catheter advanced to the study, the diagnostic sensitivity of VATS for mesothelioma
lung apex. Once the tip is in a satisfactory position the was 94 percent, compared with 86 percent for image
trocar is removed. Straight catheters are secured to the skin guided core needle biopsy, but the procedure requires that
with sutures, while pigtail catheters are secured to an adhe- the visceral and parietal pleurae not be adherent.139 A
sive disc. The catheter is then connected to a one-way further disadvantage is that chest wall seeding occurs in 22
Heimlich valve. Patients with a large air leak and non- percent of surgical biopsies as compared with only 4
expansion of the lung may have the catheter connected to percent in image guided core needle biopsy.139 This
an underwater drainage system such as a Pleur-evac with seeding along the needle tract can be eliminated by treat-
continuous suction (approximately 20 cmH2O). ing the region of biopsy with radiotherapy.140
Upright posterior–anterior and lateral chest radi-
ographs are obtained after the procedure to determine the
size of any residual pneumothorax and to confirm the Technique
position of the drainage catheter. A repeat radiograph is
performed after 18–24 hours of drainage. If the pneu- Large pleural masses can usually be biopsied under fluoro-
mothorax has resolved, the catheter is clamped for approx- scopic guidance, but ultrasound and CT are usually used.
imately 4 hours after which a repeat chest film is obtained. The needle insertion site should be selected to avoid tra-
If the lung remains completely expanded the catheter is versing lung or vessels. In this regard, it is best to place the
removed. If a pneumothorax persists and there is no pres- needle parallel to a small pleural abnormality (rather than
sure variation in the water-seal chamber of the Pleur-evac perpendicular), to enable adequate needle manipulation at
with breathing, the catheter is either malpositioned or the time of biopsy without violation of the visceral pleura.
occluded.112 However, there may also be a persistent tear A coaxial biopsy needle may be used so that multiple spec-
in the visceral pleura. In these cases drainage should con- imens can be obtained if necessary. The outer sheath is
tinue for several days, which may be sufficient time to introduced to the level of the parietal pleural. Multiple
allow the air leak to seal. If the leak persists, patients may specimens are then obtained with an automated cutting
References 267
needle device such as the Temno II Scalpel Tip Biopsy 2. Light RW. Pleural diseses, 3rd ed. Baltimore: Williams and Wilkins,
Device (Allegiance Healthcare Corporation, McGraw 1995.
3. Raptopoulos V, Davis LM, Lee G, et al. Factors affecting the
Park, IL, USA) or Achieve Biopsy System (Bauer Medical, development of pneumothorax associated with thoracentesis. AJR
Inc., Clearwater, FL, USA) as the diagnostic yield is Am J Roentgenol 1991; 156: 917–20.
increased with two or more specimens.141 Using a biopsy 4. Seneff MG, Corwin RW, Gold LH, Irwin RS. Complications
needle with a diameter of 18 gauge or larger can also associated with thoracocentesis. Chest 1986; 90: 97–100.
increase diagnostic yield.142,143 If no acute complication 5. Grogan DR, Irwin RS, Channick R, et al. Complications associated
with thoracentesis. A prospective, randomized study comparing
occurs, the patient is observed for two hours. If the patient three different methods. Arch Intern Med 1990; 150: 873–7.
remains asymptomatic and there are no acute abnormali- 6. Collins TR, Sahn SA. Thoracocentesis. Clinical value, complications,
ties on the follow-up radiograph, the patient is discharged. technical problems, and patient experience. Chest 1987; 91:
817–22.
7. Jones PW, Moyers JP, Rogers JT, et al. Ultrasound-guided
thoracentesis: is it a safer method? Chest 2003; 123: 418–23.
CONCLUSIONS AND FUTURE DIRECTION 8. Kohan JM, Poe RH, Israel RH, et al. Value of chest ultrasonography
versus decubitus roentgenography for thoracentesis. Am Rev
Improvements in percutaneous interventional techniques Respir Dis 1986; 133: 1124–6.
now allow patients to undergo image-guided diagnostic 9. Ong KC, Indumathi V, Poh WT, Ong YY. The diagnostic yield of
and therapeutic pleural procedures with less morbidity pleural fluid cytology in malignant pleural effusions. Singapore
Med J 2000; 41: 19–23.
and mortality than surgical intervention. The use of ultra- 10. Anderson CB, Philpott GW, Ferguson TB. The treatment of
sound, CT or fluoroscopy permits optimal needle or malignant pleural effusions. Cancer 1974; 33: 916–22.
catheter placement and safe and effective percutaneous 11. Bruneau R, Rubin P. The management of pleural effusions and
evacuation of pleural air and fluid collections. Image- chylothorax in lymphoma. Radiology 1965; 85: 1085–92.
guided percutaneous intervention should be considered 12. Fracchia AA, Knapper WH, Carey JT, Farrow JH. Intrapleural
chemotherapy for effusion from metastatic breast carcinoma.
the initial procedure performed to treat malignant effu- Cancer 1970; 26: 626–9.
sions, parapneumonic effusions and pneumothoraces. At 13. Johnston WW. The malignant pleural effusion. A review of
present, novel chemotherapeutic agents such as vascular cytopathologic diagnoses of 584 specimens from 472 consecutive
endothelial growth factor (VEGF) tyrosine kinase patients. Cancer 1985; 56: 905–9.
inhibitors are being tested for instillation through pleural 14. Sahn SA. Pleural effusion in lung cancer. Clin Chest Med 1993; 14:
189–200.
catheters. VEGF is over-expressed in several tumors, 15. Chernow B, Sahn SA. Carcinomatous involvement of the pleura: an
including non-small cell lung cancer, and treatment with a analysis of 96 patients. Am J Med 1977; 63: 695–702.
VEGF inhibitor has been shown to control malignant 16. Tattersall M. Pleural effusions. Curr Opin Oncol 1992; 4: 642–6.
pleural effusions caused by non-small-cell lung cancer.144 17. Hausheer FH, Yarbro JW. Diagnosis and treatment of malignant
We anticipate that in the future new diagnostic capabili- pleural effusion. Semin Oncol 1985; 12: 54–75.
18. Seaton KG, Patz EF Jr, Goodman PC. Palliative treatment of
ties, refinements in chest tubes and novel biologic scleros- malignant pleural effusions: value of small-bore catheter
ing agents should result in even further improvements in thoracostomy and doxycycline sclerotherapy. AJR Am J Roentgenol
patient care. 1995; 164: 589–91.
19. Andrews CO, Gora ML. Pleural effusions: pathophysiology and
management. Ann Pharmacother 1994; 28: 894–903.
20. Kennedy L, Sahn SA. Talc pleurodesis for the treatment of
pneumothorax and pleural effusion. Chest 1994; 106:
KEY POINTS 1215–22.
21. Clementsen P, Evald T, Grode G, et al. Treatment of malignant
● Imaging provides an accurate assessment of the pleural effusion: pleurodesis using a small percutaneous catheter.
pleural space. A prospective randomized study. Respir Med 1998; 92: 593–6.
22. Parulekar W, Di Primio G, Matzinger F, Dennie C, Bociek G. Use of
● Imaging is extremely useful in directing interven- small-bore vs large-bore chest tubes for treatment of malignant
tional procedures for pleural abnormalities. pleural effusions. Chest 2001; 120: 19–25.
● Image-guided drainage procedures are effective 23. Goff BA, Mueller PR, Muntz HG, Rice LW. Small chest-tube
in treating patients with malignant effusions, drainage followed by bleomycin sclerosis for malignant pleural
parapneumonic effusions, empyemas and pneu- effusions. Obstet Gynecol 1993; 81: 993–6.
24. Morrison MC, Mueller PR, Lee MJ, et al. Sclerotherapy of
mothoraces. malignant pleural effusion through sonographically placed small-
bore catheters. AJR Am J Roentgenol 1992; 158: 41–3.
25. O’Moore PV, Mueller PR, Simeone JF, et al. Sonographic guidance
in diagnostic and therapeutic interventions in the pleural space.
AJR Am J Roentgenol 1987; 149: 1–5.
REFERENCES 26. Marom EM, Patz EF Jr, Erasmus JJ, et al. Malignant pleural
effusions: treatment with small-bore-catheter thoracostomy and
talc pleurodesis. Radiology 1999; 210: 277–81.
1. Mathisen DJ. A surgeon’s view of interventional radiology in 27. Parker LA, Charnock GC, Delany DJ. Small bore catheter drainage
general thoracic surgery patients. Semin Intervent Radiol 1991; 8: and sclerotherapy for malignant pleural effusions. Cancer 1989;
85–7. 64: 1218–21.
268 Radiology: interventional
28. Barbetakis N, Antoniadis T, Tsilikas C. Results of chemical 50. McLeod DT, Calverley PM, Millar JW, Horne NW. Further
pleurodesis with mitoxantrone in malignant pleural effusion from experience of Corynebacterium parvum in malignant pleural
breast cancer. World J Surg Oncol 2004; 2: 16. effusion. Thorax 1985; 40: 515–18.
29. Light RW, Jenkinson SG, Minh VD, George RB. Observations on 51. Millar JW, Hunter AM, Horne NW. Intrapleural immunotherapy
pleural fluid pressures as fluid is withdrawn during thoracentesis. with Corynebacterium parvum in recurrent malignant pleural
Am Rev Respir Dis 1980; 121: 799–804. effusions. Thorax 1980; 35: 856–8.
30. Chang YC, Patz EF Jr, Goodman PC. Pneumothorax after small- 52. Moffett MJ, Ruckdeschel JC. Bleomycin and tetracycline in
bore catheter placement for malignant pleural effusions. AJR Am J malignant pleural effusions: a review. Semin Oncol 1992; 19(2
Roentgenol 1996; 166: 1049–51. Suppl 5): 59–62; discussion 62–3.
31. Bouros D, Schiza S, Patsourakis G, et al. Intrapleural streptokinase 53. Ostrowski MJ, Halsall GM. Intracavitary bleomycin in the
versus urokinase in the treatment of complicated parapneumonic management of malignant effusions: a multicenter study. Cancer
effusions: a prospective, double-blind study. Am J Respir Crit Care Treat Rep 1982; 66: 1903–7.
Med 1997; 155: 291–5. 54. Ostrowski MJ. An assessment of the long-term results of
32. Moulton JS, Benkert RE, Weisiger KH, Chambers JA. Treatment of controlling the reaccumulation of malignant effusions using
complicated pleural fluid collections with image-guided drainage intracavity bleomycin. Cancer 1986; 57: 721–7.
and intracavitary urokinase. Chest 1995; 108: 1252–9. 55. Paladine W, Cunningham TJ, Sponzo R, et al. Intracavitary
33. Kennedy L, Harley RA, Sahn SA, Strange C. Talc slurry bleomycin in the management of malignant effusions. Cancer
pleurodesis. Pleural fluid and histologic analysis. Chest 1995; 107: 1976; 38: 1903–8.
1707–12. 56. Pearson FG, MacGregor DC. Talc poudrage for malignant pleural
34. Bethune N. Pleural poudrage: a new technique for deliberate effusion. J Thorac Cardiovasc Surg 1966; 51: 732–8.
production of pleural adhesions as preliminary to lobectomy. J 57. Robinson LA, Fleming WH, Galbraith TA. Intrapleural doxycycline
Thorac Surg 1935; 4: 251–61. control of malignant pleural effusions. Ann Thorac Surg 1993; 55:
35. Card RY, Cole DR, Henscke UK. Summary of ten years of the use of 1115–21; discussion 21–2.
radioactive colloids in intracavitary therapy. J Nucl Med 1960; 1: 58. Ruckdeschel JC, Moores D, Lee JY, et al. Intrapleural therapy for
195–8. malignant pleural effusions. A randomized comparison of
36. Casali A, Gionfra T, Rinaldi M, et al. Treatment of malignant bleomycin and tetracycline. Chest 1991; 100: 1528–35.
pleural effusions with intracavitary Corynebacterium parvum. 59. Rusch VW, Figlin R, Godwin D, Piantadosi S. Intrapleural cisplatin
Cancer 1988; 62: 806–11. and cytarabine in the management of malignant pleural effusions:
37. Colt HG, Russack V, Chiu Y, et al. A comparison of thoracoscopic a Lung Cancer Study Group trial. J Clin Oncol 1991; 9: 313–19.
talc insufflation, slurry, and mechanical abrasion pleurodesis. 60. Sorensen PG, Svendsen TL, Enk B. Treatment of malignant pleural
Chest 1997; 111: 442–8. effusion with drainage, with and without instillation of talc. Eur J
38. Cunningham TJ, Olson KB, Horton J, et al. A clinical trial of Respir Dis 1984; 65: 131–5.
intravenous and intracavitary bleomycin. Cancer 1972; 29: 61. Trotter JM, Stuart JFB, McBeth F, et al. The management of
1413–19. malignant effusions with bleomycin. Br J Cancer 1979; 40: 310.
39. Davis M, Williford S, Muss HB, et al. A phase I–II study of 62. Webb HE, Oaten SW, Pike CP. Treatment of malignant ascitic and
recombinant intrapleural alpha interferon in malignant pleural pleural effusion with Corynebacterium parvum. Br Med J 1978; 1:
effusions. Am J Clin Oncol 1992; 15: 328–30. 338–40.
40. Fentiman IS, Rubens RD, Hayward JL. A comparison of 63. Yasumoto K, Ogura T. Intrapleural application of recombinant
intracavitary talc and tetracycline for the control of pleural interleukin-2 in patients with malignant pleurisy due to lung
effusions secondary to breast cancer. Eur J Cancer Clin Oncol cancer. A multi-institutional cooperative study. Biotherapy 1991;
1986; 22: 1079–81. 3: 345–9.
41. Fingar BL. Sclerosing agents used to control malignant pleural 64. Zimmer PW, Hill M, Casey K, Harvey E, Low DE. Prospective
effusions. Hosp Pharm 1992; 27: 622–8. randomized trial of talc slurry vs bleomycin in pleurodesis for
42. Gupta N, Opfell RW, Podova J, et al. Intrapleural bleomycin vs symptomatic malignant pleural effusions. Chest 1997; 112:
tetracycline for control of malignant pleural effusion: a 430–4.
randomized study. Proc Am Assoc Cancer Res 1980; 21: 366. 65. Patz EF Jr, McAdams HP, Goodman PC, Blackwell S, Crawford J.
43. Holoye PY, Jeffries DG, Dhingra HM, et al. Intrapleural etoposide Ambulatory sclerotherapy for malignant pleural effusions.
for malignant effusion. Cancer Chemother Pharmacol 1990; 26: Radiology 1996; 199: 133–5.
147–50. 66. van den Toorn LM, Schaap E, Surmont VF, et al. Management of
44. Ike O, Shimizu Y, Hitomi S, Wada R, Ikada Y. Treatment of recurrent malignant pleural effusions with a chronic indwelling
malignant pleural effusions with doxorubicin hydrochloride- pleural catheter. Lung Cancer 2005; 50: 123–7.
containing poly(L-lactic acid) microspheres. Chest 1991; 99: 67. Pollak JS, Burdge CM, Rosenblatt M, et al. Treatment of malignant
911–15. pleural effusions with tunneled long-term drainage catheters. J
45. Kitamura S, Sugiyama Y, Izumi T, Hayashi R, Kosaka K. Intrapleural Vasc Interv Radiol 2001; 12: 201–8.
doxycycline for control of malignant pleural effusion. Curr Ther 68. Pien GW, Gant MJ, Washam CL, Sterman DH. Use of an
Res 1981; 30: 515–21. implantable pleural catheter for trapped lung syndrome in patients
46. Liu X. Effectiveness of treatment with transfer of autologous or with malignant pleural effusion. Chest 2001; 119: 1641–6.
allogenic LAK cells combined with rIL-2 in 121 patients with 69. Putnam JB Jr, Walsh GL, Swisher SG, et al. Outpatient
malignant pleural effusion. Chinese Med J 1993; 15: 205–8. management of malignant pleural effusion by a chronic indwelling
47. Maiche AG, Virkkunen P, Kontkanen T, Moykkynen K, Porkka K. pleural catheter. Ann Thorac Surg 2000; 69: 369–75.
Bleomycin and mitoxantrone in the treatment of malignant 70. Putnam JB Jr, Light RW, Rodriguez RM, et al. A randomized
pleural effusions. A comparative study. Am J Clin Oncol 1993; 16: comparison of indwelling pleural catheter and doxycycline
50–3. pleurodesis in the management of malignant pleural effusions.
48. Mansson T. Treatment of malignant pleural effusion with Cancer 1999; 86: 1992–9.
doxycycline. Scand J Infect Dis Suppl 1988; 53: 29–34. 71. Musani AI, Haas AR, Seijo L, Wilby M, Sterman DH. Outpatient
49. Markman M, Cleary S, King ME, Howell SB. Cisplatin and management of malignant pleural effusions with small-bore,
cytarabine administered intrapleurally as treatment of malignant tunneled pleural catheters. Respiration 2004; 71: 559–66.
pleural effusions. Med Pediatr Oncol 1985; 13: 191–3. 72. Tremblay A, Michaud G. Single-center experience with 250
References 269
tunnelled pleural catheter insertions for malignant pleural lung abscess, and pneumothorax. AJR Am J Roentgenol 1995; 164:
effusion. Chest 2006; 129: 362–8. 581–8.
73. Ohm C, Park D, Vogen M, et al. Use of an indwelling pleural 97. Bouros D, Schiza S, Panagou P, Drositis J, Siafakas N. Role of
catheter compared with thorascopic talc pleurodesis in the streptokinase in the treatment of acute loculated parapneumonic
management of malignant pleural effusions. Am Surg 2003; 69: pleural effusions and empyema. Thorax 1994; 49: 852–5.
198–202. 98. Lee KS, Im JG, Kim YH, et al. Treatment of thoracic multiloculated
74. Davies CW, Kearney SE, Gleeson FV, Davies RJ. Predictors of empyemas with intracavitary urokinase: a prospective study.
outcome and long-term survival in patients with pleural infection. Radiology 1991; 179: 771–5.
Am J Respir Crit Care Med 1999; 160: 1682–7. 99. Pollak JS, Passik CS. Intrapleural urokinase in the treatment of
75. Ferguson AD, Prescott RJ, Selkon JB, Watson D, Swinburn CR. The loculated pleural effusions. Chest 1994; 105: 868–73.
clinical course and management of thoracic empyema. QJM 1996; 100. Merriam MA, Cronan JJ, Dorfman GS, Lambiase RE, Haas RA.
89: 285–9. Radiographically guided percutaneous catheter drainage of pleural
76. Colice GL, Curtis A, Deslauriers J, et al. Medical and surgical fluid collections. AJR Am J Roentgenol 1988; 151: 1113–6.
treatment of parapneumonic effusions: an evidence–based 101. Ulmer JL, Choplin RH, Reed JC. Image-guided catheter drainage of
guideline. Chest 2000; 118: 1158–71. the infected pleural space. J Thorac Imaging 1991; 6: 65–73.
77. Barnes NP, Hull J, Thomson AH. Medical management of 102. Misthos P, Sepsas E, Konstantinou M, et al. Early use of
parapneumonic pleural disease. Pediatr Pulmonol 2005; 39: intrapleural fibrinolytics in the management of postpneumonic
127–34. empyema. A prospective study. Eur J Cardiothorac Surg 2005; 28:
78. Davies CW, Gleeson FV, Davies RJ. BTS guidelines for the 599–603.
management of pleural infection. Thorax 2003; 58 (Suppl 2): 103. Bouros D, Schiza S, Patsourakis G, et al. Intrapleural streptokinase
ii18–28. versus urokinase in the treatment of complicated parapneumonic
79. Light RW. Parapneumonic effusions and empyema. Proc Am effusions: a prospective, double-blind study. Am J Respir Crit Care
Thorac Soc 2006; 3: 75–80. Med 1997; 155: 291–5.
80. Manuel Porcel J, Vives M, Esquerda A, Ruiz A. Usefulness of the 104. Diacon AH, Theron J, Schuurmans MM, Van de Wal BW, Bolliger
British Thoracic Society and the American College of Chest CT. Intrapleural streptokinase for empyema and complicated
Physicians guidelines in predicting pleural drainage of non- parapneumonic effusions. Am J Respir Crit Care Med 2004; 170:
purulent parapneumonic effusions. Respir Med 2006; 100: 933–7. 49–53.
81. Kearney SE, Davies CW, Davies RJ, Gleeson FV. Computed 105. Davies RJ, Traill ZC, Gleeson FV. Randomised controlled trial of
tomography and ultrasound in parapneumonic effusions and intrapleural streptokinase in community acquired pleural infection.
empyema. Clin Radiol 2000; 55: 542–7. Thorax 1997; 52: 416–21.
82. Vianna NJ. Nontuberculous bacterial empyema in patients with 106. Ozol D, Oktem S, Erdinc E. Complicated parapneumonic effusion
and without underlying diseases. J Am Med Assoc 1971; 215: and empyema thoracis: microbiologic and therapeutic aspects.
69–75. Respir Med 2006; 100: 286–91.
83. Smith JA, Mullerworth MH, Westlake GW, Tatoulis J. Empyema 107. Maskell NA, Davies CW, Nunn AJ, et al. U.K. Controlled trial of
thoracis: 14-year experience in a teaching center. Ann Thorac Surg intrapleural streptokinase for pleural infection. N Engl J Med
1991; 51: 39–42. 2005; 352: 865–74.
84. Ali I, Unruh H. Management of empyema thoracis. Ann Thorac 108. Anderson CL, Crespo JC, Lie TH. Risk of pneumothorax not
Surg 1990; 50: 355–9. increased by obstructive lung disease in percutaneous needle
85. LeBlanc KA, Tucker WY. Empyema of the thorax. Surg Gynecol biopsy. Chest 1994; 105: 1705–8.
Obstet 1984; 158: 66–70. 109. Berger R. Iatrogenous pneumothorax. Chest 1994; 105: 980–2.
86. Pothula V, Krellenstein DJ. Early aggressive surgical management 110. Berger R. Pleurodesis for spontaneous pneumothorax. Will the
of parapneumonic empyemas. Chest 1994; 105: 832–6. procedure of choice please stand up? Chest 1994; 106: 992–4.
87. Wehr CJ, Adkins RB Jr. Empyema thoracis: a ten-year experience. 111. Berger R, Smith D. Efficacy of the lateral decubitus position in
South Med J 1986; 79: 171–6. preventing pneumothorax after needle biopsy of the lung. South
88. Refaely Y, Weissberg D. Gangrene of the lung: treatment in two Med J 1988; 81: 1140–3.
stages. Ann Thorac Surg 1997; 64: 970–3; discussion 3–4. 112. Klein JS, Schultz S. Interventional chest radiology. Curr Probl
89. Serletti JM, Feins RH, Carras AJ, et al. Obliteration of empyema Diagn Radiol 1992; 21: 219–77.
tract with deepithelialized unipedicle transverse rectus abdominis 113. Westcott JL. Percutaneous transthoracic needle biopsy. Radiology
myocutaneous flap. J Thorac Cardiovasc Surg 1996; 112: 631–6. 1988; 169: 593–601.
90. Weissberg D, Refaely Y. Pleural empyema: 24-year experience. Ann 114. vanSonnenberg E, Casola G, Ho M, et al. Difficult thoracic lesions:
Thorac Surg 1996; 62: 1026–9. CT-guided biopsy experience in 150 cases. Radiology 1988; 167:
91. Garcia-Yuste M, Ramos G, Duque JL, et al. Open-window 457–61.
thoracostomy and thoracomyoplasty to manage chronic pleural 115. Fish GD, Stanley JH, Miller KS, Schabel SI, Sutherland SE.
empyema. Ann Thorac Surg 1998; 65: 818–22. Postbiopsy pneumothorax: estimating the risk by chest
92. Striffeler H, Gugger M, Im Hof V, et al. Video-assisted radiography and pulmonary function tests. AJR Am J Roentgenol
thoracoscopic surgery for fibrinopurulent pleural empyema in 67 1988; 150: 71–4.
patients. Ann Thorac Surg 1998; 65: 319–23. 116. Harter LP, Moss AA, Goldberg HI, Gross BH. CT-guided fine-needle
93. Cassina PC, Hauser M, Hillejan L, Greschuchna D, Stamatis G. aspirations for diagnosis of benign and malignant disease. AJR Am
Video-assisted thoracoscopy in the treatment of pleural empyema: J Roentgenol 1983; 140: 363–7.
stage-based management and outcome. J Thorac Cardiovasc Surg 117. Miller KS, Fish GB, Stanley JH, Schabel SI. Prediction of
1999; 117: 234–8. pneumothorax rate in percutaneous needle aspiration of the lung.
94. Renner H, Gabor S, Pinter H, et al. Is aggressive surgery in pleural Chest 1988; 93: 742–5.
empyema justified? Eur J Cardiothorac Surg 1998; 14: 117–22. 118. Poe RH, Kallay MC, Wicks CM, Odoroff CL. Predicting risk of
95. Thourani VH, Brady KM, Mansour KA, Miller JI Jr, Lee RB. pneumothorax in needle biopsy of the lung. Chest 1984; 85:
Evaluation of treatment modalities for thoracic empyema: a cost- 232–5.
effectiveness analysis. Ann Thorac Surg 1998; 66: 1121–7. 119. Lindskog GE, Halasz NA. Spontaneous pneumothorax: a
96. Klein JS, Schultz S, Heffner JE. Interventional radiology of the consideration of pathogenesis and management with review of
chest: image-guided percutaneous drainage of pleural effusions, seventy-two hospitalized cases. Arch Surg 1957; 75: 693–8.
270 Radiology: interventional
120. Rhea JT, DeLuca SA, Greene RE. Determining the size of 133. Miller BH, Rosado-de-Christenson ML, Mason AC, et al. From the
pneumothorax in the upright patient. Radiology 1982; 144: archives of the AFIP. Malignant pleural mesothelioma: radiologic–
733–6. pathologic correlation. Radiographics 1996; 16: 613–44.
121. Conces DJ Jr, Tarver RD, Gray WC, Pearcy EA. Treatment of 134. McCaughey WT, Al-Jabi M. Differentiation of serosal hyperplasia
pneumothoraces utilizing small caliber chest tubes. Chest 1988; and neoplasia in biopsies. Pathol Annu 1986; 21: 271–93.
94: 55–7. 135. Chan JK, Loo KT, Yau BK, Lam SY. Nodular histiocytic/mesothelial
122. Casola G, vanSonnenberg E, Keightley A, et al. Pneumothorax: hyperplasia: a lesion potentially mistaken for a neoplasm in
radiologic treatment with small catheters. Radiology 1988; 166: transbronchial biopsy. Am J Surg Pathol 1997; 21: 658–63.
89–91. 136. Heilo A, Stenwig AE, Solheim OP. Malignant pleural mesothelioma:
123. Reinhold C, Illescas FF, Atri M, Bret PM. Treatment of pleural US-guided histologic core-needle biopsy. Radiology 1999; 211:
effusions and pneumothorax with catheters placed percutaneously 657–9.
under imaging guidance. AJR Am J Roentgenol 1989; 152: 137. Metintas M, Ozdemir N, Isiksoy S, et al. CT-guided pleural needle
1189–91. biopsy in the diagnosis of malignant mesothelioma. J Comput
124. Peters J, Kubitschek KR. Clinical evaluation of a percutaneous Assist Tomogr 1995; 19: 370–4.
pneumothorax catheter. Chest 1984; 86: 714–17. 138. Boutin C, Rey F. Thoracoscopy in pleural malignant mesothelioma:
125. Erasmus JJ, Goodman PC, Patz EF Jr. Management of malignant a prospective study of 188 consecutive patients. Part 1: Diagnosis.
pleural effusions and pneumothorax. Radiol Clin North Am 2000; Cancer 1993; 72: 389–93.
38: 375–83. 139. Agarwal PP, Seely JM, Matzinger FR, et al. Pleural mesothelioma:
126. Tarver RD, Conces DJ Jr. Chest intervention: drainage and biopsies. sensitivity and incidence of needle track seeding after image-
In: Castaneda-Zuniga WR, Tadavarthy SM (eds). Interventional guided biopsy versus surgical biopsy. Radiology 2006; 241:
radiology, 2nd ed. Baltimore: Williams & Wilkins; 1997: 1357–81. 589–94.
127. Alfageme I, Moreno L, Huertas C, et al. Spontaneous 140. West SD, Foord T, Davies RJ. Needle-track metastases and
pneumothorax. Long-term results with tetracycline pleurodesis. prophylactic radiotherapy for mesothelioma. Respir Med 2006;
Chest 1994; 106: 347–50. 100: 1037–40.
128. Janzing HM, Derom A, Derom E, et al. Intrapleural quinacrine 141. Scott EM, Marshall TJ, Flower CD, Stewart S. Diffuse pleural
instillation for recurrent pneumothorax or persistent air leak. Ann thickening: percutaneous CT-guided cutting needle biopsy.
Thorac Surg 1993; 55: 368–71. Radiology 1995; 194: 867–70.
129. Milanez JR, Vargas FS, Filomeno LT, et al. Intrapleural talc for the 142. Andriole JG, Haaga JR, Adams RB, Nunez C. Biopsy needle
prevention of recurrent pneumothorax. Chest 1994; 106: characteristics assessed in the laboratory. Radiology 1983; 148:
1162–5. 659–62.
130. Perlmutt LM, Braun SD, Newman GE, et al. Transthoracic needle 143. Haaga JR, LiPuma JP, Bryan PJ, Balsara VJ, Cohen AM. Clinical
aspiration: use of a small chest tube to treat pneumothorax. AJR comparison of small-and large-caliber cutting needles for biopsy.
Am J Roentgenol 1987; 148: 849–51. Radiology 1983; 146: 665–7.
131. Jay SJ. Diagnostic procedures for pleural disease. Clin Chest Med 144. Matsumori Y, Yano S, Goto H, et al. ZD6474, an inhibitor of
1985; 6: 33–48. vascular endothelial growth factor receptor tyrosine kinase,
132. Diacon AH, Theron J, Schubert P, et al. Ultrasound-assisted inhibits growth of experimental lung metastasis and production of
transthoracic biopsy: fine-needle aspiration or cutting-needle malignant pleural effusions in a non-small cell lung cancer model.
biopsy? Eur Respir J 2007; 29: 357–62. Oncol Res 2006; 16: 15–26.
21
Radiology: pleural ultrasound
INTRODUCTION Table 21.1 The indications for pleural and chest wall
ultrasounda
Although the introduction of diagnostic sonography of the Indications
abdomen dates back to the late 1940s, ultrasonography of
the thorax lagged behind by many decades. The inability of 1 To detect a pleural effusion and guide
ultrasound (US) to penetrate aerated tissue has diverted thoracentesis and drainage, especially in small or
chest physicians from recognizing its excellent ability to loculated effusions
visualize the chest wall, pleura and pathology of lung abut- 2 To differentiate a subpulmonary effusion from a
ting the pleura. The major advantages of thoracic US subphrenic fluid accumulation and diaphragm
include its dynamic properties, low cost, lack of radiation, paralysis in radiographically elevated
mobility and short examination time.1–6 It is also well hemidiaphragms
suited for use in intensive care units, where suboptimal 3 To localize pleural tumors, pleural thickening and
conditions for radiography make the diagnosis of clinically lung tumors that abut the pleural surface and to
significant thoracic abnormalities difficult.3 Furthermore, guide needle aspiration and/or biopsies thereof
US of the chest is increasingly being used to guide inter- 4 To assist the evaluation of patients with pleuritic
ventional procedures, such as thoracentesis, biopsies of the chest pain
chest wall, pleura or abutting lung and the placement of 5 To clarify the nature of unknown pleural densities
intercostal drains. The indications for pleural and chest 6 To recognize pneumothorax, especially for
wall US are summarized in Table 21.1. The main aim of emergency situations, or when radiographic
this chapter is to demystify ultrasonography for the chest equipment is unavailable
physician by reviewing the basic principles and techniques 7 To localize pleural fluid prior to a thoracoscopy
from the perspective of the non-radiologist. 8 To evaluate empyema and a parapneumonic pleural
effusion in order to detect loculations and septae,
and to guide aspiration with or without tube drainage
a
Adapted with permission from: Tsai TH, Yang PC. Ultrasound in the
AN APPROACH TO THORACIC diagnosis and management of pleural disease. Curr Opin Pulm Med 2003;
ULTRASONOGRAPHY 9: 282–90.
Technical principles
sounds to generate an image is paramount in order to
‘Ultrasounds’ are acoustic waves with a frequency above comprehend its uses and limitations.
human hearing. Most US scanners operate in the fre- The US unit produces sound by means of a piezoelec-
quency range of 2–15 megahertz (MHz). A very basic tric transducer encased in a handheld probe that is
understanding of how the US scanner employs these ultra- attached to the processing unit with an electric cable. This
272 Radiology: pleural ultrasound
sound is focused in the transducer and is efficiently trans- lower penetration. A lower frequency probe (e.g. 3.5 MHz)
mitted into the thorax. Ultrasound waves are propagated with curvilinear shape for covering a large area is therefore
in liquid media (e.g. pleural effusions) or in tissues with a suitable for initial screening of superficial and deeper
high water content (e.g. muscle, liver, consolidated lung or structures, while the high frequency probe (e.g. 8 MHz)
tumors), but are reflected off interfaces between dissimilar with a linear shape is used for refined assessment of an
densities (e.g. gas or bone). If the US encounters gas (e.g. abnormal chest wall or pleural area.
normal aerated lungs or a pneumothorax) or solids (e.g. The three most important controls on a standard key-
ribs), the density difference is so great that most of the board are ‘depth’, ‘gain’ and ‘freeze’. The depth function is
acoustic energy is reflected. This phenomenon is known as a digital zoom that defines what portion of the scanned
acoustical impedance and explains why structures deeper image is displayed on the monitor at what magnification.
than the visceral pleura are invisible by means of ultra- The scale is displayed on a vertical axis. Obese subjects or
sonography in the non-diseased state. patients with a large effusion or intrathoracic tumors may
The reflected part of the sound waves is detected as an require a depth setting of up to 12 cm. High frequency
echo. The time delay between emitted US and the received scanning is performed at a maximum depth of around
echo is used to calculate the depth of the structure causing 3–4 cm. The gain is, in essence, a measure for the amplifi-
the reflection. Furthermore, the greater the difference cation of the echoes and determines the brightness of the
between acoustic impedances, the larger the echo will be. image. The freeze function allows for the capturing of still
The transducer captures sound waves returning to the images, and to perform measurements with the appropri-
probe and converts these echoes into electrical pulses, ate keys and the trackball. Only experienced users should
which are ultimately processed and transformed into a change advanced parameters, such as the frequency of a
digital image. The intensity and distribution of the pixels particular probe.
appearing on the screen is determined by three character-
istics of the echo, namely (1) its direction, (2) its intensity
and (3) the time elapsed from emission to capture. Images Patient positioning
may contain ‘hyperechoic’ or white areas caused by high-
amplitude echoes and ‘hypoechoic’ or dark areas from The optimal patient position for scanning is a paramount
low-amplitude echoes. but under-appreciated aspect. Sonographic access to the
chest can be achieved via the abdomen, intercostal spaces
or the upper thoracic aperture (supraclavicular fossa). It is
The ultrasound unit important to review a patient’s chest radiograph and com-
puted tomography (CT) scan prior to performing a chest
Adequate pleural and chest wall ultrasonography can be per- US examination. This will not only identify the area of
formed by means of the most basic, entry-level, interest, but will also guide the positioning of the patient.
two-dimensional black-and-white US equipment. Doppler The posterior chest is best scanned with the patient in the
and color flow echo are not required for routine pleural exam- sitting position using a bedside table as an armrest (Figure
ination. For documentation of still images, a basic thermal 21.1a,b), whereas the lateral and anterior chest wall can be
printer is sufficient. Most modern scanners also allow for the examined with the patient in either the lateral decubitus or
transferral of images to data storage devices and can capture even supine position (Figure 21.1c). Maximum visualiza-
dynamic information in video format, which is preferable. tion of the lung and pleura is achieved by examining along
The US scanner is adorned by a confusing array of con- the intercostal spaces. Raising the arm above the patient’s
trols and options. It is imperative that the occasional sono- head increases the intercostal space distance and facilitates
grapher familiarizes themself with the scanner and its most scanning in erect or recumbent positions. A patient can
important function keys. It is helpful to differentiate fold the arms across the chest in order to displace the
between settings to optimize the machine for thoracic US scapulae when surveying the upper posterior thorax.
in general, and controls for fine tuning the scanning for the Superior sulcus pathology can be visualized apically with
individual case. For occasional thoracic scanning, basic the patient in the supine or sitting position.
settings programmed for abdominal sonography will
suffice. However, if a machine is to be used mainly for tho-
racic US, we recommend calling upon an expert to assist Scanning
with the basic setting up of the machine, which includes
contrast and brightness of the monitor as well as default Once the patient is adequately positioned and the area of
settings of depth and gain when using different probes. interest is identified, liberal application of gel is the final
Thoracic US is best performed with two transducers. A step before scanning. It is advisable to hold the probe like
3.7 MHz (range: 2–5 MHz) curvilinear probe is compul- a pen for writing on paper, and not like chalk for writing
sory, and an 8 MHz (range: 5–10 MHz) linear probe is a on a blackboard (Figure 21.1b). Experienced sonographers
very helpful addition. As a rule, higher frequency gives keep their eyes on the screen while their hand moves the
better resolution closer to the probe, but at the cost of probe across the area of interest and provides the posi-
Diagnostic thoracic ultrasonography 273
(a) (b)
tional information. The probe is moved slowly, preferably with posterior acoustic shadowing (Figure 21.2b). When
along intercostal spaces, which are oblique and not hori- the ribs are scanned along the longitudinal, the anterior
zontal. Frequent pauses are needed for observing the spon- cortex appears as a continuous echogenic line.
taneous movement of structures with respiration. Unclear The visceral and parietal pleura can normally not be
findings can be compared with the contralateral side. differentiated by means of a low-frequency probe, which
instead displays one highly echogenic line representing
the pleura and pleuropulmonary surface. With a high-
DIAGNOSTIC THORACIC ULTRASONOGRAPHY resolution linear probe (e.g. 8 MHz), the visceral and pari-
etal portions of the pleura can be seen as two distinct
Normal chest wall and pleura echogenic lines, with the latter seemingly thinner in
appearance. The two layers can be seen to slide over each
The initial surveillance of a normal chest with the low fre- other with respiratory motion. The respiratory movement
quency probe will yield a series of echogenic layers of of the lung relative to the chest wall is visible with both
muscles and fascia planes (Figure 21.2a). The ribs appear probes and is called the ‘lung sliding’ sign. Its presence on
as curvilinear structures on transverse scans, associated real-time US is strong evidence against a pneumothorax.7
274 Radiology: pleural ultrasound
(a) (b)
The ‘curtain-sign’ describes the variable obscuring of often associated with strands of echogenic material and
underlying structures by air containing tissue. In normal septations which show more or less mobility with respira-
subjects, the curtain-sign is seen in the costophrenic angle. tion and the cardiac cycle. The presence of septae has
The upper abdominal organs are easily visible on expira- several implications. Chen et al.13 demonstrated that
tion, but during inspiration the normal air-filled lung is patients with septated effusions needed longer chest tube
moved downwards in front of the probe and temporarily drainage, longer hospital care and were more likely to
obscures the sonographic window. require fibrinolytic therapy or surgery compared with
The parenchyma of normal aerated lungs is invisible by those with unseptated effusions. Tu et al.14 recently con-
means of US. The large change in acoustic impedance at firmed some of these findings in medical intensive care
the pleura–lung interface causes horizontal artifacts that unit patients. Empyema may cause a strongly echogenic
are seen as a series of echogenic parallel lines equidistant effusion that may be mistaken for a solid pleural lesion.
from one another below the pleura. These bright but form-
less lines are known as reverberation artifacts and dimin- ESTIMATING THE VOLUME OF A PLEURAL EFFUSION BY
ish in intensity with increasing distance from the pleura ULTRASOUND
(Figure 21.2a). Vertical ‘comet-tail’ artifacts (Figure
21.2c), caused by fluid-filled subpleural interlobular Several studies have shown reasonable correlation between
septae, can also be seen originating at the pleura–lung the volume of an effusion estimated with planimetric
interface. The normal diaphragm is best seen through the measurements and its square dimensions.15–17 Such geo-
lower intercostal spaces or via the liver or spleen. It is seen metric calculations are hampered by the uneven distribu-
as an echogenic 1 mm thick line which contracts with tion of fluid in the presence of pleuropulmonary
inspiration. adhesions. We suggest the following practical way to clas-
sify the volume of an effusion: minimal, if the echo-free
space is confined to the costophrenic angle; small, if the
Pleural effusions space is greater than the costophrenic angle but still within
the range of the area covered with a 3.5 MHz curvilinear
SONOGRAPHIC DIAGNOSIS probe; moderate, if the space is greater than a one-probe
range but within a two-probe range; and large, if the space
The value of ultrasonography for detection and quantifica- is bigger than a two-probe range.
tion of pleural effusions is uncontested. Ultrasound is par-
ticularly helpful in determining the nature of localized or DIFFERENTIATION OF EFFUSION FROM PLEURAL
diffuse pleural opacities, and is more sensitive than decu- THICKENING
bitus expiratory films in identifying minimal or loculated
effusions.8 Sonographically, a pleural effusion appears as To distinguish small effusions from anechoic pleural thick-
an anechoic, homogeneous space between parietal and vis- ening can be challenging. Both may appear as anechoic on
ceral pleura (Figure 21.3). This space may change in shape US. Nearly 20 percent of echo-free pleural lesions will not
with respiration, and the atelectatic lung inside a large yield free fluid, whereas a significant percentage of
effusion may appear as a tongue-like structure within the complex-appearing lesions will do so. Mobility is a good
effusion. In inflammatory effusions, adhesions between sign for effusion. Marks et al.18 found that if a lesion
the two pleural surfaces may result in the absence of lung changed shape with respiratory excursion and if it con-
motion above the effusion. If an abnormal elevation of a tained movable strands or echo densities, the lesion was an
hemidiaphragm is noted on the chest radiograph, subpul- effusion. If a color Doppler is available, the fluid color sign
monary effusion can be differentiated from a subphrenic is the most sensitive and specific ultrasonographic evidence
fluid collection or diaphragm paralysis.9 of a small effusion. The sign refers to the presence of a color
signal within the fluid collection that is believed to arise
DETERMINING THE NATURE OF A PLEURAL EFFUSION from transmitted motion during respiratory or cardiac
cycles. This sign has a sensitivity of 89.2 percent and speci-
The sonographic appearance of a pleural effusion depends ficity of 100 percent in detecting minimal fluid collections.19
on its nature, cause and chronicity. Four appearances are
recognized based on the internal echogenicity: anechoic;
complex but non-septated; complex and septated; and Pleural thickening
homogenously echogenic. Transudates are invariably
anechoic, unseptated and free flowing, whereas complex, Pleural thickening is defined as focal lesions arising from
septated or echogenic effusions are usually exudates.10,11 the visceral or parietal pleura that is greater than 3 mm in
Malignant effusions are often anechoic. Nodular pleural width with or without an irregular margin (Figure 21.4). It
thickening is apparent in the minority of malignant effu- appears as broadening of the pleura and does not exhibit a
sions, and echogenic swirling patterns have recently been fluid color sign or display movement relative to the chest
linked to these effusions.12 Inflammatory effusions are wall. Pleural thickening most often appears hypoechoic,
276 Radiology: pleural ultrasound
(a) (b)
(c) (d)
(e) (f)
Diagnostic thoracic ultrasonography 277
Figure 21.3 (opposite) (a) Pleural effusion is presented as an echo-free space between the visceral and parietal pleura. Compressive
atelectasis of the lung may be seen in a large effusion. The effusion can be subclassified as anechoic (b), complex non-septated (c),
complex septated (d), or homogenously echogenic (e). Note the movable echogenic spots within the complex non-septated effusion, and
the floating strands and septa within the complex septated effusion (arrowheads). (f) Pleural effusion associated with pleural nodules or
nodular thickenings is characteristic of malignant effusion. PE, pleural effusion; D, diaphragm; RLL, right lower lobe; L, lung; T, pleural
tumor. Reproduced with permission from: Tsai TH, Yang PC. Ultrasound in the diagnosis and management of pleural disease. Curr Opin
Pulm Med 2003; 9: 282–90.
278 Radiology: pleural ultrasound
(b)
Pleural tumors
(a) (b)
Figure 21.8 (a) An ultrasound (US) image showing a lung tumor with posterior echo enhancement. Note that both of the visceral and
parietal pleural lines are intact. (b) This US shows tumor extension beyond the pleura. The visceral pleural line is interrupted, and the
respiratory movement of the tumor is disturbed in real-time US. Invasion of the pleural cavity by the tumor is evident. L, lung; T, tumor; Pv,
visceral pleural; Pp, parietal pleura. Reproduced with permission from: Diacon AH, Theron J, Bolliger CT. Transthoracic ultrasound for the
pulmonologist. Curr Opin Pulm Med 2005; 11: 307–12.
Apart from solid tumors, numerous pathological processes Ultrasound can serve for the acute bedside assessment of
can replace the air within lung tissue and thereby become patients presenting with possible pulmonary embolism.32
detectable with US, provided that the pleura is abutted. Pulmonary infarction is recognized as a peripheral wedge-
280 Radiology: pleural ultrasound
Thoracentesis
ciated effusion decreases the risk of visceral pleural lacera- whether a histological specimen obtained via cutting
tions, which is particularly relevant in cases with minimal needle biopsy is needed or not.43 In the absence of ROSE,
pleural effusion. cutting biopsies should be performed in all cases where
Conventionally, closed pleural biopsies (e.g. with the cytology is non-contributory and in cases where a diagno-
Abrams needle) could only safely be performed in the sis other than lung cancer is suspected. Conveniently,
presence of a sizeable pleural effusion or pneumothorax US is a good tool for exclusion of a pneumothorax post-
determined clinically or on chest radiography. However, aspiration or biopsy. If the lesion remains visible on US
US can demonstrate small fluid collections and facilitate and is unchanged in location, shape and size, it implies
the use of devices that were not primarily designed for that no free air is present between the sampled lesion and
fluid aspiration (e.g. Tru-cut needles). Numerous studies the visceral pleura, and that a clinically relevant pneu-
have concluded that US-guided Tru-cut needle biopsies mothorax is unlikely. US-guided biopsies are safe proce-
have higher sensitivity and specificity in the diagnosis of dures, with an overall complication rate of only 1–2
pleural malignancy and tuberculosis than unaided Abrams percent.42,43
needle biopsies.38–41
13. Chen KY, Liaw YS, Wang HC, et al. Sonographic septation: a useful
KEY POINTS prognostic indicator of acute thoracic empyema. J Ultrasound Med
2000; 19: 837–43.
14. Tu CY, Hsu WH, Hsia TC, et al. Pleural effusions in febrile medical
● Thoracic ultrasonography can be performed by ICU patients: chest ultrasound study. Chest 2004; 126: 1274–80.
means of the most basic ultrasound equipment. 15. Eibenberger KL, Dock WI, Ammann ME, et al. Quantification of
● In healthy individuals, ultrasound can visualize pleural effusions: sonography versus radiography. Radiology 1994;
the chest wall, the diaphragm and the pleura, but 191: 681–4.
16. Lorenz J, Borner N, Nikolaus HP. Sonographic volumetry of pleural
not the lung parenchyma.
effusions. Ultraschall Med 1988; 9: 212–15.
● The main domain of thoracic ultrasound is the 17. Balik M, Plasil P, Waldauf P, et al. Ultrasound estimation of
investigation of chest wall abnormalities, pleural volume of pleural fluid in mechanically ventilated patients.
thickening and pleural tumors, and the qualita- Intensive Care Med 2006; 32: 318–21.
tive and quantitative description of pleural effu- 18. Marks WM, Filly RA, Callen PW. Real-time evaluation of pleural
lesions: new observations regarding the probability of obtaining
sions.
free fluid. Radiology 1982; 142: 163–4.
● Ultrasound can visualize lung tumors and other 19. Wu RG, Yang PC, Kuo SH, Luh KT. ‘Fluid color’ sign: a useful
parenchymal pulmonary processes provided that indicator for discrimination between pleural thickening and
they abut the pleura. pleural effusion. J Ultrasound Med 1995; 14: 767–9.
●20. Chan SS. Emergency bedside ultrasound to detect pneumothorax.
● Ultrasound is the ideal tool to assist with thora-
Acad Emerg Med 2003; 10: 91–4.
centesis and drainage of effusions.
21. Chan SS. The comet tail artifact in the diagnosis of pneumothorax.
● Ultrasound-assisted fine needle aspiration and J Ultrasound Med 2002; 21: 1060.
cutting needle biopsy of lesions arising from the 22. Simon BC, Paolinetti L. Two cases where bedside ultrasound was
chest wall, pleura and lung are safe and have a able to distinguish pulmonary bleb from pneumothorax. J Emerg
high yield in the hands of chest physicians. Med 2005; 29: 201–5.
23. Slater A, Goodwin M, Anderson KE, Gleeson FV. COPD can mimic
● New applications of ultrasound include the diag-
the appearance of pneumothorax on thoracic ultrasound. Chest
nosis of pneumothorax and pulmonary 2006; 129: 545–50.
embolism. ●24. Herth FJ, Eberhardt R, Ernst A, et al. Diagnosis of pneumothorax by
means of transthoracic ultrasound: a prospective trial. Eur Respir J
2004; 24: S491.
25. Targhetta R, Bourgeois JM, Chavagneux R, et al. Ultrasonographic
REFERENCES approach to diagnosing hydropneumothorax. Chest 1992; 101:
931–4.
● = Key primary paper 26. Gorg C, Restrepo I, Schwerk WB: Sonography of malignant pleural
◆ = Major review article effusion. Eur Radiol 1997; 7: 1195–8.
●27. Sugama Y, Tamaki S, Kitamura S, et al. Ultrasonographic
◆1. Tsai TH, Yang PC. Ultrasound in the diagnosis and management of evaluation of pleural and chest wall invasion of lung cancer. Chest
pleural disease. Curr Opin Pulm Med 2003; 9: 282–90. 1988; 93: 275–9.
◆2. Diacon AH, Theron J, Bolliger CT. Transthoracic ultrasound for the 28. Suzuki N, Saitoh T, Kitamura S: Tumor invasion of the chest wall in
pulmonologist. Curr Opin Pulm Med 2005; 11: 307–12. lung cancer: diagnosis with US. Radiology 1993; 187: 39–42.
◆3. Yu CJ, Yang PC, Chang DB, Luh KT. Diagnostic and therapeutic use 29. Bruneton JN, Caramella E, Hery M et al. Axillary lymph node
of chest sonography: value in critically ill patients. Am J metastases in breast cancer: preoperative detection with US.
Roentgenol 1992; 159: 695–701. Radiology 1986; 158: 325–6.
◆4. Koh DM, Burke S, Davies N, et al. Transthoracic US of the chest: 30. Gottesman E, McCool MD. Ultrasound evaluation of the paralyzed
clinical uses and applications. Radiographics 2002; 22: E1. diaphragm. Am J Resp Crit Care Med 1997; 155: 1570–4.
●31. Yang PC, Chang DB, Yu CJ, et al. Ultrasound guided percutaneous
Available: http://radiographics.
rsnajnls.org/cgi/content/full/22/1/e1 cutting biopsy for the diagnosis of pulmonary consolidation of
◆5. Mayo PH, Doelken P. Pleural ultrasonography. Clin Chest Med unknown aetiology. Thorax 1992; 47: 457–60.
●32. Reissig A, Kroegel C. Transthoracic ultrasound of lung and pleura
2006; 27: 215–17.
◆6. Evans AL, Gleeson FV. Radiology in pleural disease: state of the art. in the diagnosis of pulmonary embolism: a novel non-invasive
Respirology 2004; 9: 300–12. bedside approach. Respiration 2003; 70: 441–52.
●7. Lichtenstein DA, Menu Y. A bedside ultrasound sign ruling out 33. Lichtenstein D, Meziere G. A lung ultrasound signs allowing
pneumothorax in the critically ill. Chest 1995; 108: 1345–8. bedside distinction between pulmonary edema and COPD: the
8. Kocijancic I, Kocijancic K, Cufer T. Imaging of pleural fluid in comet-tail artifact. Intensive Care Med 1998; 24: 1331–4.
healthy individuals. Clin Radiol 2004; 59: 826–9. 34. Diacon AH, Brutsche MH, Soler M. Accuracy of pleural puncture
9. Ko JC, Yang PC, Chang DB, et al. Ultrasonographic evaluation of sites: a prospective comparison of clinical examination with
peridiaphragmatic lesions: a prospective study. J Med Ultrasound ultrasound. Chest 2003; 123: 436–41.
●35. Yang PC, Kuo SH, Luh KT: Ultrasonography and ultrasound-guided
1994; 2: 84–92.
10. Yang PC, Luh KT, Chang DB, et al. Value of sonography in needle biopsy of chest diseases: indications, techniques,
determining the nature of pleural effusion: analysis of 320 cases. diagnostic yields and complications. J Med Ultrasound 1993; 1:
AJR Am J Roentgenol 1992; 159: 29–33. 53–63.
11. Hirsch JH, Rogers JV, Mack LA. Real-time sonography of pleural 36. Chen KY, Liaw YS, Wang HC, et al. Sonographic septation: a useful
opacities. Am J Roentgenol 1981; 136: 297–301. prognostic indicator of acute thoracic empyema. J Ultrasound Med
12. Chian CF, Su WL, Soh LH, et al. Echogenic swirling pattern as a 2000; 19: 837–43.
predictor of malignant pleural effusions in patients with 37. Tatersall DJ, Traill ZC, Gleeson FV. Chest drains: does size matter?
malignancies. Chest 2004; 126: 129–34. Clin Radiol 2000; 55: 415–21.
References 283
●38. Chang DB, Yang PC, Luh KT, Kuo SH, Yu CJ. Ultrasound guided ultrasound assisted transthoracic biopsy performed by
pleural biopsy with Tru-cut needle. Chest 1991; 100: 1328–33. pulmonologists. Respiration 2004; 71: 519–22.
39. Heilo A, Stenwig AE, Solheim OP. Malignant pleural mesothelioma: 43. Diacon AH, Theron J, Schubert P, et al. Ultrasound-assisted
US-guided histologic core needle biopsy. Radiology 1999; 211: transthoracic biopsy: fine-needle aspiration or cutting-needle
657–9. biopsy? Eur Respir J 2007; 29: 357–362.
40. McLeod DT, Ternouth I, Nkanza N. Comparison of the Tru-cut 44. Maskell NA, Gleeson FV, Davies RJ. Standard pleural biopsy versus
biopsy needle with the Abrams punch for pleural biopsy. Thorax CT-guided cutting-needle biopsy for diagnosis of malignant
1989; 44: 794–6. disease in pleural effusions: a randomised controlled trial. Lancet
41. Theron J, Diacon AH, Williams Z, Walzl G, Bolliger CT. Abrams 2003; 361: 1326–30.
versus TruCut needle in tuberculous pleuritis: a pilot study. Eur 45. Yang PC, Luh KT, Lee YC, et al. Lung abscesses: US examination
Respir J 2004; 24: 73s. and US-guided transthoracic aspiration. Radiology 1991; 180:
●42. Diacon AH, Schuurmans MM, Theron J, et al. Safety and yield of 171–75.
This page intentionally left blank
22
Pathology: histology
pleura included in a biopsy may provide clues to the cause pathognomonic and may be seen in other types of granu-
of pleuritis; for example, pneumonia observed in sub- lomas. These should not be mistaken for foreign material.
pleural lung tissue. Wegener’s granulomatosis (see Chapter 32, Effusions
from connective tissue diseases) may involve the pleura
and causes a less circumscribed type of granulomatous
Eosinophilic pleuritis inflammation rather than well-formed granulomas. The
classic histopathology picture is one of palisading histio-
In eosinophilic pleuritis there is an increased number of cytes mixed with multinucleated giant cells, other inflam-
eosinophils, sometimes strikingly so, in the inflammatory matory cells and fibroblasts surrounding a central area of
infiltrate which otherwise may show features of fibrinous necrosis. The necrosis begins as micro-abscesses with neu-
or fibrous pleuritis, mesothelial hyperplasia, etc. Since trophils and leukocytoclastic necrosis that coalesce and
eosinophilic pleuritis is often associated with some distinc- enlarge into so-called ‘geographic’ areas of basophilic
tive clinical situations, there may be accompanying necrosis. While these features and those of vasculitis can be
histopathological findings in any underlying lung tissue seen in the pleura, Wegener’s granulomatosis is usually
that may be sampled. A classic etiology of eosinophilic diagnosed on a biopsy of lung tissue.
pleuritis is spontaneous pneumothorax, which occurs gen- Foreign body granulomatous reactions may occur in
erally in young adults. In these cases, the eosinophilic pleu- the pleura. Most of these are the result of deliberate pleu-
ritis may be associated with focal blebs, bullae or focal rodesis or iatrogenic instillation of foreign material into
honeycombing of the underlying lung tissue. Pneumo- the pleural cavity to cause its obliteration for the purpose
thorax or hemothorax from a wide variety of causes can of treating recurrent pleural effusions.
potentially result in eosinophilic pleuritis and lymphangi-
oleiomyomatosis may be seen in the underlying lung tissue
in women of childbearing age. Eosinophilic pleuritis may Localized pleural fibrosis
also be associated with certain drug reactions and infec-
tions which may also produce characteristic findings in the Apical caps which also involve the subpleural tissue are
underlying lung tissue. often associated with emphysema. Apical caps consist of
elastotic scars with thick fragments of elastic tissue with a
grayish tint on hematoxylin and eosin (H&E) staining
Granulomatous pleuritis reminiscent of solar elastosis in the skin. An ischemic
origin has been suggested for apical caps and these lesions
Granulomatous pleuritis is characterized by the presence do have a resemblance to scars from pulmonary infarcts.2
of granulomatous inflammation, which may consist of Pleural plaques show a histopathological pattern of
well-formed granulomas or less circumscribed areas of dense, acellular collagen arranged in a distinctive basket-
granulomatous inflammation depending on the etiology. weave pattern. Additional findings may include calcifica-
Granulomatous pleuritis is often accompanied by features tions and ossification. Pleural plaques result from
of chronic pleuritis (i.e. chronic inflammation and organized pleuritis and thus can result from many causes,
fibrosis). including occupational levels of exposure to asbestos, chest
Infections, such as TB or fungal infections, that involve surgery and trauma, and infections involving the pleura.
the pleura classically produce well-formed granulomas Examination of the pleural plaques themselves for asbestos
with or without central necrosis, but may occasionally bodies is virtually always non-productive, even when the
form less circumscribed areas of granulomatous inflam- plaques are caused by asbestos exposure. The possible
mation. It is important to remember that special stains for association of a pleural plaque with asbestos exposure is
organisms, for example the acid-fast stain for tuberculosis based on (1) observation of increased asbestos bodies in
or the Grocott’s methenamine silver (GMS) stain for lung tissue sections and/or (2) the finding of elevated
fungus, are very helpful when positive, but a negative asbestos body or fiber concentrations on lung tissue diges-
special stain can occur even when infection is present. tion studies.
Therefore, cultures are always recommended even when
special stains are negative.
Sarcoidosis is characterized by well-formed, compact NEOPLASTIC LESIONS OF THE PLEURA
granulomas in a lymphangitic distribution that includes
the pleura. Significant necrosis is not a usual feature of sar- Benign or low-grade malignant mesothelial
coidosis granulomas, although small punctate areas of neoplasms
necrosis may be present in some granulomas. Sarcoidosis
granulomas may be surrounded by chronic inflammatory Well-differentiated papillary mesotheliomas are character-
infiltrates or by fibrosis. Endogenous materials, for ized by broad papillae lined by bland cuboidal mesothelial
example, Schaumann bodies or calcium oxalate crystals, cells (Figure 22.1). These are most common in women and
are often seen in the histiocytes of sarcoidosis, but are not are more common in the peritoneum than in the pleura.
Neoplastic lesions of the pleura 287
Histopathology patterns
Benign mesenchymal neoplasms of the pleura 1 Negativity for an antibody does not, by itself, confirm
mesothelioma. (i) Many of the immunostains typically
Solitary fibrous tumor (see Chapter 38) is a well-circum- used to rule out carcinoma apply primarily to lung
scribed mass arising in the pleura and sometimes in the cancers, particularly adenocarcinomas. One must
subpleural lung. Solitary fibrous tumor has several remember that carcinomas from primary sites other
histopathological patterns: a cellular pattern with spindle than the lung and malignancies other than carcinoma
cells to oval cells, a hemangiopericytoma pattern with
staghorn blood vessels and a relatively acellular collagen
pattern with slit-like spaces. Most solitary fibrous tumors
are immunopositive for CD34 and this is a useful marker
for confirming their diagnosis. Unlike mesotheliomas,
solitary fibrous tumors are immunonegative for keratin.
Other benign mesenchymal neoplasms of the pleura
include lipoma and Schwannoma.6,7
Figure 22.3 Sarcomatous malignant mesothelioma showing Figure 22.4 Immunostain for calretinin showing intranuclear
malignant spindle cells resembling a fibrosarcoma or malignant immunopositivity in an epithelial malignant mesothelioma. (See
fibrous histiocytoma. (See also Color Plate 15.) also Color Plate 16.)
(for example, melanoma) may metastasize to the and continuous around the cell border, but indicates
pleura. Metastatic renal cell carcinomas, for example, carcinoma when staining is thin, discontinuous or cyto-
are typically negative for all of the traditional lung car- plasmic. Specific criteria for interpretation also apply to
cinoma markers such as carcinoembryonic antigen other antibodies.
(CEA). Therefore, ruling out only adenocarcinoma of
the lung by immunostains does not, by itself, confirm a For the reasons given above, a panel of immunostains is
diagnosis of mesothelioma. (ii) Even a well-differenti- recommended to distinguish diffuse malignant mesothe-
ated adenocarcinoma of the lung may be immuno- lioma from other cancers (Figure 22.4). Characteristic
negative for one or more markers typically associated immunostain results for malignant mesothelioma versus
with lung cancer. selected malignancies of other types are given in Table
2 Negativity for an antibody does not, by itself, confirm or 22.2.4,5,8–19
exclude a diagnosis. It is not unusual for poorly differen- The diagnosis of diffuse malignant mesothelioma, like
tiated mesotheliomas, as well as poorly differentiated the diagnosis of any other malignancy, should be based on
malignancies of other types, to fail to express markers the routine histopathology and immunostaining pattern of
seen in better-differentiated tumors of the same type. a tumor tissue sample when the tumor has a gross distri-
Poorly differentiated, pleomorphic and sarcomatous bution potentially consistent with the diagnosis. The mere
lung carcinomas are often negative for CEA and other fact that an individual is reported to have a history of ele-
lung cancer markers and only immunopositive for vated asbestos exposure is not a basis for making the diag-
vimentin and keratin. Poorly differentiated and sarco- nosis of mesothelioma. Most individuals with a history of
matous mesotheliomas are often negative for calretinin elevated asbestos exposure who have cancer will have
and other mesothelioma markers and only immuno- cancers other than mesothelioma. Similarly, a diagnosis of
positive for vimentin and keratin. Therefore, immuno- mesothelioma should not be excluded simply because
stains might not distinguish between poorly there is no history of asbestos exposure.
differentiated or sarcomatous mesotheliomas and Distinguishing malignant mesothelioma from benign
poorly differentiated or sarcomatous carcinomas, or reactive mesothelial hyperplasias and fibrous pleuritis on
other poorly differentiated malignancies. biopsy can be a very difficult challenge, particularly on a
3 Staining with an antibody may require interpretation small biopsy (Figure 22.5). Benign proliferations may
according to very specific criteria before it can be consid- show features that mimic those seen in malignancy and
ered positive. For example, the reliability of calretinin malignant mesothelioma can be very bland or biopsies
depends on the antibody used and only nuclear may sample very early malignant lesions. Features that
immunostaining is meaningful. Cytoplasmic calretinin favor benign versus malignant mesothelial proliferations
immunostaining may be seen in many types of carci- are listed in Table 22.3. Ultimately, the presence of true
noma and even in mesotheliomas and should not be tissue invasion is the most reliable feature to distinguish a
considered support for a diagnosis of mesothelioma in malignant mesothelioma but, even then, the pathologist
those cases. The antihuman mesothelial cell antibody should be aware that entrapment of benign mesothelial
HBME-1 indicates mesothelioma when staining is thick cells within fibrous pleuritis may mimic invasion. It should
Other rare primary malignant neoplasms of the pleura 289
Table 22.2 Characteristic results for selected immunostains for malignant mesothelioma versus other malignancies
Suggestive features
◆21. Churg A, Colby TV, Cagle PT, et al. The separation of benign and 23. Parkash V, Gerald WL, Parma A, Miettinen M, Rosai J.
malignant mesothelial proliferations. Am J Surg Pathol 2000; 24: Desmoplastic small round cell tumor of the pleura. Am J Surg
1183–200. Pathol 1995; 19: 659–65.
22. Galateau-Sallé F, Cagle PT. Non-malignant versus malignant 24. Zhang PJ, Livolsi VA, Brooks JJ. Malignant epithelioid vascular
proliferations on pleural biopsy. In: Cagle PT (ed.). Diagnostic tumors of the pleura: report of a series and literature review. Hum
pulmonary pathology. New York: Marcel Dekker, 2000; 555–67. Pathol 2000; 31: 29–34.
This page intentionally left blank
23
Pathology: cytology
MARK R WICK
General clinicopathology considerations in the diagnosis 293 Adjunctive studies of pleural diseases 302
of pleural neoplasms Key points 310
Cytopathological ‘partitions’ in the diagnosis of pleural 296 References 310
neoplasms
Interpreting pleural fluid specimens is often a challenging orders, many mature lymphocytes are often observed and
task even for the experienced cytopathologists. Reactive may mimic a well-differentiated lymphoproliferative dis-
but benign epithelial cells may closely resemble malignan- order.6 Flow cytometry or immunohistology are com-
cies in effusion cytomorphology; conversely, carcinoma monly required. It is distinctly unusual to observe
cells in pleural effusions can be deceptively bland. granulomatous histiocytic arrays in pleural cytology speci-
Mesenchymal and hematopoietic proliferations likewise mens or in pleural biopsies.
may require extensive additional studies with adjunctive Metastatic carcinomas, melanomas and lymphomas in
techniques. This chapter aims to describe current the pleura often shed freely into effusions, but there still
approaches for the resolution of interpretative problems in remain a sizable proportion of such cases (30–40 percent)
cytopathology of the pleura. Conditions such as hemotho- for which cytological examination is non-diagnostic.
rax, chylothorax and pneumothorax relating to structural Reasons for that phenomenon are twofold: some pleural
disorders of the chest are not considered herein, because malignancies demonstrate a high degree of intercellular
their diagnosis are relatively straightforward. cohesion, with a limited tendency to seed the pleural fluid,
and, second, a fibrinofibrous ‘cap’ often covers malignan-
cies in the pleural soft tissue. This effectively walls off
GENERAL CLINICOPATHOLOGY underlying tumor cells from the interserosal space.
CONSIDERATIONS IN THE DIAGNOSIS OF Moreover, benign mesothelial cells on the surface of the
PLEURAL NEOPLASMS cap may shed into the pleural fluid and further obscure the
true nature of the underlying pathological process. The
Shortness of breath, chest pain and a flu-like illness are overall efficacy of cytopathological evaluation of pleural
common presentations of patients with inflammatory or fluid for the detection of mesothelioma approximates only
malignant pleural diseases.1,2 Thoracentesis is typically the 30 percent in the author’s experience. The diagnostic yield
first diagnostic procedure employed in evaluation of fluid is highest with epithelial mesothelioma and lowest in sar-
in the pleural cavities. In benign conditions, the cytologi- comatoid tumors.
cal presentation is usually that of moderate cellularity, When processing pleural fluid samples for microscopy,
showing only scant mixed inflammatory cells and benign it is important that a good cell-block preparation (forma-
mesothelial elements, or, alternatively, a predominance of lin-fixed and paraffin-embedded) be obtained in order to
specific leukocytes. The first situation may be seen in asso- evaluate cells by immunohistochemistry. This provision
ciation with cardiac, renal or hepatic insufficiency; some circumvents the many technical problems that can be
cases of viral pleuritis; and pneumoconioses affecting the encountered in attempting such studies with conventional
pleura (e.g. asbestos pleuritis).3 The second scenario cytological specimens (i.e. cytospins and filter prepara-
accompanies empyema or parapneumonic pleural effu- tions).7 A centrifuged pellet of cells from the effusion may
sions, in which cases the fluid is neutrophil-rich, as well as be suspended in nutrient solutions such as Michel’s
connective tissue diseases, tuberculosis, fungal pleuritis medium or RPMI medium, and submitted for flow cytom-
and most viral infections.4,5 In the last four of those dis- etry and cytogenetic analysis. For those adjunctive tech-
294 Pathology: cytology
group demonstrates more nuclear irregularity, apoptotic present as a malignant pleural tumor of uncertain type,
cellular dropout and mitotic activity, and also may show because of their relative commonality in adults.44 The his-
‘lymphoglandular bodies’ in cytological preparations.25,26 tological appearance is that of a variably-organoid prolif-
Rhabdomyosarcoma of the ‘solid-alveolar’ variety eration of extensively apoptotic small neoplastic cells with
demonstrates a somewhat greater degree of nuclear pleo- brisk mitotic activity, often demonstrating prominent
morphism than that seen in other SRCTs, occasionally crush artifact. Cytologically, these neoplasms are com-
with interspersed large cells possessing relatively generous posed of small cells with high nuclear-to-cytoplasmic
amounts of eosinophilic cytoplasm39,40 (Figure 23.9). ratios, nuclear ‘smearing,’ crush artifact, nuclear
Nuclei are not dissimilar in appearance to those of ‘molding,’ powdery chromatin, inconspicuous nucleoli,
ES/PNET, and LRMS also shares potential PAS-positivity scant cytoplasm and a tendency for loose cohesion and cel-
with neuroectodermal tumors. If foci suggesting an ‘alveo- lular dispersion45 (Figures 23.10 and 23.11). Staining with
lar’ (dyshesive) growth pattern are observed in a PAS- Romanowsky methods may reveal fine metachromatic
reactive SRCT, the diagnosis of rhabdomyosarcoma
should be favored.
A large percentage of neuroblastoma (NB) cases present
with metastases, with or without elevated urinary levels of
catecholamines or their metabolites.41,42 The primary
tumors may reside in the adrenal medulla or the remain-
der of the sympathetic nervous system. Metastatic NB may
be extremely difficult to distinguish from ES/PNET or
other SRCT, particularly in small biopsies or cytological
specimens. In particular, there is a rare glycogen-rich form
of NB which bears a striking morphological resemblance
to neuroectodermal neoplasms. Neuroblastoma is com-
posed of primitive round to angulated cells with scant
cytoplasm. Well-formed true rosettes and pseudorosettes,
neuropil formation and dystrophic calcification may aid in
the diagnosis of this lesion. A careful search is also worth-
while for primitive or mature ganglionic elements, repre-
sented by nucleolated cells with eccentric, relatively
abundant eosinophilic cytoplasm.43
Small-cell neuroendocrine carcinomas (SCNCs) are, Figure 23.10 Metastatic small-cell neuroendocrine carcinoma
perhaps, the most important of the SRCTs that may (‘oat-cell carcinoma’) of the lung in pleural fluid. The tumor cells
show nuclear molding and high nuclear-to-cytoplasmic ratios.
(See also Color Plate 27.)
Figure 23.9 Metastatic rhabdomyosarcoma in pleural fluid, Figure 23.11 Metastatic small-cell neuroendocrine carcinoma
demonstrating cellular dyshesion and slight nuclear in a pleural effusion specimen. Nuclear molding and dispersion of
pleomorphism in small tumor cells. (See also Color Plate 26.) chromatin is evident. (See also Color Plate 28.)
298 Pathology: cytology
cytoplasmic granules. Reactivity with argyrophilic histo- variety of terms, including oncocyte, oxyphil, Hurthle-cell,
chemical techniques such as the Sevier–Munger, Grimelius Ashkenazy-cell and others, although one classically regards
or Churukian–Schenk procedures is helpful in recognizing oncocytoid change to reflect to an abundance of mito-
that small-cell carcinomas have neuroendocrine features. chondria or other cytoplasmic organelles such as lyso-
Once metastatic SCNCs are identified as such, there are somes, neuroendocrine granules, cytofilaments and
few other nuances of morphology or biochemistry that can smooth endoplasmic reticulum.47
be used with certainty to predict their anatomic sources. In Oncocytic neoplasms arise most often in the salivary
particular, cellular peptide and amine products are broadly glands, thyroid, kidneys and parathyroid glands but can
shared among this group of neoplasms, regardless of their originate from virtually any organ, see Table 23.1.
topographic origins. Hence, anatomic patterns of metasta- Histologically, variably-organoid arrays of large cells
sis and the relative frequency of SCNC in various organ demonstrating round nuclei, inconstant mitotic activity,
systems must be used as the principal data in determining dispersed chromatin and indistinct nucleoli are present.
the likely source. The lung is by far the most common Cytologically, these tumors manifest as flat sheets, loose
origin for metastatic small-cell carcinomas in the pleura. groups, cords and singly-dispersed polygonal cells, usually
Small-cell mesotheliomas also exist, albeit rarely.46 with a strikingly monotonous appearance. The polyhedral
These tumors are rather nondescript cytologically and his- cells may be intermingled with spindled forms. The pres-
tologically, and they most closely simulate the appearances ence of neurosecretory granules, represented by fine red
of small-cell carcinomas. Adjunctive pathologic studies are cytoplasmic granules, is highly suggestive of neuro-
mandatory. endocrine differentiation and may be detected in
Romanowsky stains. Amyloidaceous matrical material
may be present.
Malignant large-polygonal-cell tumors Metastatic hepatocellular carcinomas (HCCs) often
exhibit granular cytoplasmic eosinophilia due to lipofuscin
Large-polygonal-cell tumors can be subdivided into neo- granules and abundant endoplasmic reticulum. Up to 15
plasms with amphophilic cytoplasm, cytoplasmic eosino- percent of HCCs show globular intracytoplasmic, PAS-
philia and clear-cell features. negative oxyphilic inclusions, which, when present, are
useful diagnostic clues. In fine-needle aspiration biopsies,
the tumor cells may be seen singly, in sheets or in compact
Malignant oncocytic tumors
cords. Intranuclear cytoplasmic invaginations, intracellu-
Cells showing intense cytoplasmic acidophilia lar cytoplasmic globules or bile pigment (highlighted with
(eosinophilia) (Figure 23.12) have been referred to using a the Fouchet stain) may be seen.48,49
Tumors
with other secondary tumors, especially those of renal, Metastatic classical seminoma/germinoma may be mis-
ovarian or adrenal origins, or with clear-cell mesothe- interpreted when it presents as a metastatic lesion. A
lioma.57 Clear-cell change in all of these tumors results diffuse arrangement of clear tumor cells is typical of this
from accumulations of intracytoplasmic fat or glycogen, or neoplasm, sometimes with irregular separation or com-
both. Hence, immunohistochemical or cytogenetic evalu- partmentalization by fibrous stroma and diffuse perme-
ations, rather than histochemical examination, are the ation by small mature lymphocytes.54 Other common
most helpful. cytomorphological features of seminoma/germinoma
Cytologically, smears of each of the lesions in this group include well-defined cell borders, evenly spaced nuclei and
are generally hypercellular. They consist of loosely cohe- single prominent central nucleoli.60 When present, multi-
sive groups and individually scattered malignant cells with nucleated synctiotrophoblastic cells and epithelioid granu-
anisonucleosis, nuclear hyperchromasia, irregular promi- lomas are also helpful interpretatively. Abundant
nent nucleoli and abundant finely vacuolated to clear cyto- intracytoplasmic glycogen is evident on PAS staining.
plasm.
Clear-cell thyroid carcinomas of both the papillary and
follicular types have been recognized.58 In extrathyroidal Malignant spindle-cell and pleomorphic
sites such as the pleura, these cancers share many micro- tumors
scopic features with other clear cell malignancies and
immunohistology or electron microscopy must be Metastatic carcinomas from various organ sites may occa-
employed to resolve such uncertainty. Cytological findings sionally have overwhelming spindle-cell components
in clear-cell papillary thyroid carcinoma (CCPTC) include resembling pleural sarcomas. A relatively common setting
the presence of nuclear pseudoinclusions and grooves, is that of a lesion composed of spindle cells for which a
similar to conventional papillary tumors. Unfortunately, primary mesothelial origin cannot be excluded on radi-
these features can also be observed in metastatic renal ographic and conventional pathology assessments. A pro-
tumors. In addition, multinucleated tumor cells may be portion of these cases actually represent metastatic
seen in CCPTC, and these are particularly helpful because sarcomatoid renal cell carcinomas, and additional detailed
they are not a part of the cytological spectrum of other imaging may reveal a mass in the kidney.61,62 In such
clear-cell carcinomas. Follicular formations containing instances, the primary tumors may be entirely sarcomatoid
colloid-like material are likewise supportive of clear-cell or represent a preponderance of clear-cells with only focal
follicular thyroid carcinoma or CCPTC. spindle cell differentiation. Morphogically similar primary
sarcoma-like tumors may also be encountered in the lungs,
urinary tract, female genital tract, alimentary tract, pan-
Clear-cell tumors lacking consistent architectural creas, thyroid, upper airway mucosa and in pleural sarco-
patterns matoid mesothelioma.
Other carcinomas that potentially assume a clear-cell Regardless of their specific lineages, the cytological
image and lack a consistent growth pattern are represented image of sarcomatoid malignancies is much more consis-
by metastatic adenocarcinomas and ‘hydropic’ squamous tent with that of a true sarcoma than with an epithelial
carcinomas of various anatomic sources. Other than neoplasm. Cellular dyshesion, anisocytosis, nuclear pleo-
anatomic patterns of spread, there are no reliable architec- morphism, unremarkable cytoplasmic details and nonde-
tural or cytological features that can be used to predict the script stroma are their usual features63 (Figure 23.14).
topographic sources of these lesions. Fortunately, the Adjunctive studies are mandatory for diagnosis.
immunophenotypes of clear-cell prostatic, pulmonary, True spindle-cell sarcomas only exceptionally manifest
renal, salivary and cutaneous tumors demonstrate many as pleural metastasis in the absence of a known primary
points of dissimilarity, making such profiles valuable in tumor; similarly, primary pleural sarcoma is extremely
differential diagnosis.53 rare.64
Metastatic melanomas composed of clear ‘balloon-’ or
signet-ring cells are capable of imitating a multiplicity of
other malignant clear-cell neoplasms, including carcino- Clinicopathological features of malignant
mas, sarcomas, lymphomas and germ-cell tumors.59 The mesothelioma
clear cells in melanomas result from cystic dilatation of
premelanosomes or the cytoplasmic accumulation of Pleural mesothelioma (Figure 23.15) is a rare tumor that
lipid or acid mucin. Nuclei in these cases may be may be challenging to recognize accurately. However, a
deceptively bland. In cytological preparations, the cells in firm diagnosis is crucial in the ever-growing number of
clear-cell melanomas can be misinterpreted as those of legal compensation claims.65,66
an adenocarcinoma or liposarcoma. Histochemistry Traditionally, three broad histopathological patterns of
and immunostaining or ultrastructural studies are neces- mesothelioma have been considered: epithelial (including
sary in amelanotic tumors to define their melanocytic oncocytoid/deciduoid, clear-cell and small-cell subtypes),
nature. sarcomatoid (including desmoplastic and ‘lymphohistio-
Cytopathological ‘partitions’ in the diagnosis of pleural neoplasms 301
Figure 23.15 Autopsy photograph showing the typical Figure 23.16 Pleural fluid specimen showing epithelioid
configuration of end-stage pleural mesothelioma, surrounding the malignant mesothelioma. The tumor cells exhibit increased
lungs and invading adjacent soft tissue with obliteration of the nuclear-to-cytoplasmic ratios and nucleoli, and form a gland-like
pleural space. (See also Color Plate 32.) array. Intercellular windows are evident. (See also Color Plate 33.)
302 Pathology: cytology
tumors are useful in this context, and they are best used as
batteries. These include CA15.3, CA19-9, B72.3/CA72-4
and anti-carcinoembryonic antigen (CEA).73–75 Another
polypeptide which is related to keratin-19 (CYFRA-21-1)
was found to have no value in this setting.76 Most studies
have shown that an elevated level of CEA (normal
£5 ng/mL) is the most sensitive single indicator of pleural
malignancies, and is also relatively reliable as an individual
marker of metastatic carcinoma (as opposed to other neo-
plasms). B72.3/CA72-4 has a similar value.
In the opinion of the author, at least two of the afore-
mentioned markers should be abnormal before a confident
biochemical diagnosis of a malignant effusion can be made.
None of the glycoproteins listed in this section is typically
Figure 23.17 Sarcomatoid mesothelioma in pleural fluid. The
elevated in malignant mesothelioma, and false positives
tumor cells are irregular in size and shape and show a tendency
from benign pleural diseases can occur. Cytological exam-
to dyshesion, simulating the appearance of sarcoma. (See also
ination is necessary in all cases with abnormal pleural fluid
Color Plate 34.)
tumor marker levels, as well as situations in which the clin-
ical suspicion for malignancy remains high.
‘Lymphohistiocytoid’ mesothelioma is another sarcoma-
toid variant, but it bears more resemblance to ‘lympho-
epithelioma-like’ carcinomas than to sarcomas.72 All forms Algorithmic immunohistochemistry of pleural
of sarcomatoid mesothelioma shed poorly into pleural malignant neoplasms of uncertain or
fluid, and manifest scarce atypical spindle cells therein in indeterminate nature
only a minority of cases.
Biphasic mesothelioma manifests a combination of the The overwhelming number of antibodies for diagnostic
epithelial and sarcomatoid patterns. immunohistology has made an algorithmic approach nec-
Histochemical studies still play an important role in essary, especially as no single reagent can provide a defini-
separating mesotheliomas from other neoplasms. The tive answer in all cases. Several caveats must be heeded
most useful histochemical procedures in this setting are before immunohistochemical algorithms can be safely and
the PAS stain with diastase digestion (D-PAS), the muci- effectively applied, as outlined in earlier publications.28,77
carmine method and the colloidal iron procedure with The statistical data used to construct the algorithms in
hyaluronidase digestion. The presence of neutral mucin in this chapter were gathered over a period of many years
the neoplastic cells on D-PAS or mucicarmine stains is a using specimens (including cell blocks) fixed routinely in
diagnostic feature of adenocarcinoma, though up to 5 10 percent neutral-buffered formalin; primary antibody
percent of mesotheliomas may show focal labeling for incubations at 4°C for 16–18 hours; and the Elite®
mucicarmine or D-PAS. avidin–biotin–peroxidase complex method of immuno-
detection (Vector Laboratories, Burlingame, CA, USA).
The antibodies that are most pertinent to differential diag-
ADJUNCTIVE STUDIES OF PLEURAL DISEASES nosis of tumors in the pleura are discussed below.
Thus, mixtures of monoclonal antibodies may be prepared visos to the last statement, however. First, true reactivity
to cover the widest range of kilodalton sizes.28,47,53,77–81 for EMA must be regarded as crisp labeling of the plas-
One approach to the subtyping of carcinomas involves malemma, with or without cytoplasmic staining as well;
the immunohistological ‘dissection’ of the keratin classes cytoplasmic labeling alone is a spurious finding that
which they manufacture. In fact, it has been found that should be considered a negative result. Second, hepatocel-
monospecific antibodies to keratin 7 (a simple glandular lular carcinomas, adrenocortical carcinomas and most
keratin) (Figure 23.18) and keratin 20 (a 46 kDa protein malignant germ cell neoplasms are characteristically EMA
that is relatively restricted to enteric-type tissues, urinary negative, even though they are epithelial in nature. Also,
bladder and selected neuroepithelial cells) provide helpful selected malignant lymphomas may exhibit an EMA-like
information in this regard. In the topic under discussion determinant. The use of other supplementary antibodies
here, keratin 20 is utilized as a ‘secondary’ reactant in the obviates the potential confusion that may be caused by
identification of MTUOTs, after application of the these expressions of EMA-like moieties.
broadly-reactive ‘screening’ keratin reagent. It has been contended that double-layered (‘tram-
Vimentin is a 55 kDa protein which is expressed widely track’) EMA-labeling of the surfaces of malignant epithe-
by many classes of neoplasia, but it may be used as a poten- lioid cells is strongly suggestive of mesothelioma and that
tially differential indicator of certain epithelial tumor mor- EMA positivity is restricted to malignant (and not benign)
photypes. For example, the majority of renal cell mesothelial cells. The author’s experience has led to a good
carcinomas, thyroid carcinomas and adrenocortical carci- deal of skepticism regarding these assertions.
nomas are vimentin positive.
Desmin is a 57 kDa moiety that is restricted in its distri- MOC-31
bution to cells and neoplasms that show myogenous dif-
ferentiation. Thus, it is most helpful in the recognition of MOC-31 is a 41 kDa membrane-based glycoprotein
rhabdomyosarcomas and leiomyosarcomas. However, widely-distributed in epithelial cells and tumors from
desmin reactivity is also present in a substantial propor- many tissue sites, but not the mesothelium. This offers
tion of mesotheliomas, probably as a reflection of ‘diver- diagnostic distinction between serosal adenocarcinoma
gent’ differentiation in those basically epithelial tumors. and mesothelioma (which typically lacks MOC-31).
However, antibodies to MOC-31 fail to label several carci-
nomas, including hepatocellular carcinoma, germ-cell
EPITHELIAL MEMBRANE ANTIGEN
malignancies and renal cell carcinoma.
Epithelial membrane antigen (EMA) is found on the
surface of most epithelial cells and is actually a family of TUMOR-ASSOCIATED GLYCOPROTEIN (TAG)-72/CA72-4
milk fat globule glycoproteins ranging in size from 40 to
The monoclonal antibodies known as B72.3 and CC49
425 kDa. Monoclonal antibodies to this discriminant are
label a plasmalemmal glycoprotein designated as tumor-
useful in determining the epithelial or mesenchymal
associated glycoprotein-72 or CA72-4. It was isolated from
nature of an undifferentiated tumor. There are some pro-
a human breast carcinoma cell line and appears to be vir-
tually ‘pan-carcinomatous’ in its distribution. Like MOC-
31, TAG-72 is characteristically absent in mesotheliomas,
adrenocortical carcinomas, hepatocellular carcinomas,
renal cell carcinoma, nasopharyngeal carcinoma, thyroid
carcinomas and malignant germ cell tumors.
CA-125
Figure 23.18 Immunoreactivity for cytokeratin-7 in metastatic CA-125 was characterized as a glycoproteinaceous mem-
pulmonary adenocarcinoma involving pleural fluid. (See also Color brane constituent of ovarian carcinoma cells. A closely-
Plate 35.) similar or identical moiety was expressed by mesothelial
304 Pathology: cytology
CA19-9
In addition to being a serological indicator for colon THYROGLOBULIN AND THYROID TRANSCRIPTION FACTOR-1
cancer, CEA is also expressed by many other epithelial
tumors. Monoclonal antibodies to this 180 kDa family of Thyroglobulin is a 660 kDa moiety that is restricted in its
markers represent prototypic epitope-specific probes that expression to the follicular thyroid epithelium and related
may be used in the selective recognition of small portions neoplasms. The greatest utility of this marker is in the
of a large antigen. Adenocarcinomas of the lung, breast and determination of origin for metastatic papillary carcino-
gastrointestinal tract have been uniformly reactive with a mas in the cervical lymph nodes or the lung, or for second-
particular commercial monoclonal anti-CEA (HO62) used ary deposits of more nondescript adenocarcinomas in
in the author’s laboratory (Figure 23.19), and, therefore, bone. A relatively newly-described determinant, thyroid
metastatic glandular neoplasms expressing this substance transcription factor-1 (TTF1), is a 38 kDa nuclear protein
are likely to have arisen in one of these anatomic sites. present in the majority of primary pulmonary carcinomas
Mesotheliomas are consistently non-reactive for CEA. (Figure 23.20) but not mesotheliomas, making it valuable
in that specific differential diagnosis.
CD15 (ANTI-LEU-M1)
GROSS CYSTIC DISEASE FLUID PROTEIN-15
CD15 is a hematopoietic cell-surface protein that is seen in
myelomonocytic elements and in the Reed–Sternberg cells Gross cystic disease fluid protein-15 (GCDFP-15) is a
of Hodgkin’s disease and some T-cell lymphomas. CD15 is soluble product found in the fluid contents of fibrocystic
present in adenocarcinomas from a wide spectrum of breasts and is strongly expressed by cells with apocrine
Adjunctive studies of pleural diseases 305
BG8 + 4 + 88
MoAb 44-3A6 + 100 + 8
Factor VIII ± Rare − N.S.
Surfactant apoprotein − − + 62
Anti-Lewis antigen + 11 + 76
Tn antigen − − + 62
E-cadherin + 10 + 77
TTF-1 + 68 + 100
MoAb SM3 + 52 + 100
Secretory component
(SC) + 0–62 + 60 Figure 23.22 This cell block preparation of pleural fluid shows
Pregnancy specific diffuse labeling of tumor cells for CD45 in a case of large-cell
protein + 0+6 + 34–59 non-Hodgkin’s lymphoma involving the pleural space. (See also
CA 19–9 − − + 39 Color Plate 39.)
OV632 + 85–91 + 20–63
NSE + 96 N.S. N.S.
CD57 + 70 N.S. N.S.
Mab 45 + N.S. + N.S. another highly selective determinant for lymphoid prolif-
HEA-125 − − + 75 erations.
Anti BRG − − + 83
ICAM-1 + 100 N.S. N.S.
IMMUNOHISTOLOGICAL EVALUATION OF SARCOMAS
VCAM + 87 N.S. N.S.
Parathyroid hormone + 84 + 11 Most sarcomas differ from metastatic sarcomatoid carci-
CD44H + 91 + 45 nomas and from sarcomatoid mesotheliomas in that they
IOB 3 — — + 100 are keratin negative. The principal exception is synovial
P63 — 0 ± 10b sarcoma,86 which is immunoreactive for epithelial
aTable modified from Reference 76. markers, including calretinin, and closely resembles
bMetastatic squamous cell carcinomas are virtually 100% reactive for mesothelioma morphologically and immunophenotypi-
nuclear p63 protein cally. However, synovial sarcoma was labeled by BER-Ep4
N.S., Not studied or not specified; MoAb, monoclonal antibody; NSE, in 90 percent of cases, compared with 13 percent for
neuron-specific enolase; ICAM, intercellular adhesion molecule; VCAM,
mesothelioma.87 Conversely, mesothelial tumors were
vascular cell adhesion molecule; BRG, retinoblastoma-gene-related
protein (other abbreviations are monoclonal antibody designations and immunoreactive for Wilms’ tumor protein-1 (WT1) and
have no expanded names). CD141 in the majority of cases, whereas synovial sarcoma
was typically negative. Ultimately, cytogenetic studies are
the most helpful in separating mesothelioma and synovial
sarcoma; the latter manifests a consistent t(X;18) chromo-
CD45 (LEUKOCYTE COMMON ANTIGEN) AND OTHER somal translocation that is not observed in mesothelial
HEMATOPOIETIC MARKERS lesions.88
Other immunohistochemical markers can be used to
CD45 is a cell membrane protein that is nearly universally determine the lineage of a sarcoma, once it has been
present in leukocytic proliferations, including leukemias defined as such. These include: indicators of a generic
and non-Hodgkin’s lymphomas (Figure 23.22), but it is myogenous nature, such as desmin; striated muscle
virtually never observed in other tumors. Thus, it is differentiation (e.g. myoglobin, myogenin, myo-D1 and
extremely valuable in the recognition of hematopoietic ‘fast’ myosin); proteins seen in smooth muscle cells that
neoplasms in the pleura and elsewhere. There are two include calponin and caldesmon; markers of peripheral
caveats to the latter statement; for unexplained reasons, nerve sheath differentiation such as S100, CD56, CD57
classic Reed–Sternberg cells and Reed variant cells of and nerve growth factor receptor; proteins seen in
Hodgkin’s lymphomas do not label for CD45, and approx- osteogenic sarcomas, including osteopontin and
imately 30 percent of large-cell anaplastic (Ki-1) non- osteonectin; and endothelial polypeptides such as CD31,
Hodgkin’s lymphomas also are non-reactive for this CD34, von Willebrand factor and receptor for Ulex
marker. Both of those tumors consistently express CD30, europaeus I lectin.
Adjunctive studies of pleural diseases 307
DES/MSA
DES/MSA -- NSE/SYN/CGA
+ Malignant
+ lymphoma + Malignant
+ melanoma
or leukemia
CK
CK -- LCA
LCA -- EMA
EMA -- S100/HMB-45/MART-1 -- VIM
VIM
Rhabdomyosarcoma --
Peripheral
neuroectodermal DES/MSA/MYOD-1/MYOGN CD57/SYN/MB2
l
tumor (PNET)
) --
-- Technically
PNET Figure 23.23 Algorithm for
PNET inadequate
-- NSE/SYN/CD99 specimen immunohistochemical diagnosis of
PNET
ET
PN small-cell malignancies.
+ Malignant melanoma
-- S100/HMB-45/MART-1/Tyrosinase
--
CD45 VIMENTIN
Malignant lymphoma
+
-- or leukemia
+
-- Calretinin Mesothelioma --
Calretinin
+ Technically
Technically
inferior specimen Figure 23.24 Algorithm for
+ Adenocarcinoma
BER-EP4/CD15/CEA/S100 immunohistochemical diagnosis of generic
large-cell malignancies.
308 Pathology: cytology
Malignant schwannoma or
+ + Angiosarcoma
neuroid malignant melanoma
--
-- HMB-45/MART-1 --
Vimentin
Vimentin
Malignant melanoma +
Technically
+ -- inadequate
specimen
Desmin/muscle-specific actin
Pleomorphic sarcoma, NOS
Leiomyosarcoma
-- +
Calretinin and
Probable
CD141
-- sarcomatoid squamous
Keratin and epithelial cell carcinoma Figure 23.25 Algorithm for
membrane antigen + immunohistochemical diagnosis of
Mesothelioma
pleomorphic malignancies.
PSA
Prostatic carcinoma Either or GCDFP alone-- Breast, salivary gland
TGB GCDFP
both CEA alone-- GI tract, lung, Mullerian tract, bladder
CEA-M S100 alone-- Mullerian tract, breast, salivary gland
PLAP alone-- Mullerian tract, GI tract, lung,
PSA S100 kidney, breast
Thyroid carcinoma PLAP CA125 alone-- Mullerian tract, biliary tract,
TGB pancreas, lung, breast, GI tract, liver
CA-125 CA19-9 alone-- GI tract, pancreas, Mullerian tract,
CA19-9 bladder, lung
CK20 alone-- GI tract, Mullerian tract, pancreas, lung
PSA/TGB Both CK20 ER alone-- Breast, Mullerian tract, bladder, liver,
() B72.3/MOC-31
ER stomach
GCDFPCEA-- Breast, salivary gland
GCDFPS100-- Breast, salivary gland
GCDFPPLAP-- Breast
GCDFPCA125-- Breast, salivary gland
Both -- Mesothelioma GCDFPCA19-9-- Salivary gland, breast
GCDFPCK20-- Not described in literature
GCDFPER-- Breast
CEAS100-- Breast, salivary gland, Mullerian tract,
CALRET/CD141 stomach, biliary tract
Keratin mixture NPC CEAPLAP-- GI tract, biliary tract, lung, Mullerian tract
CEACA125-- Mullerian tract, biliary tract,
-- pancreas, lung, GI tract, breast
CEACA19-9-- GI tract, pancreas, bladder, Mullerian
HCC -- tract, lung
VIM -- CEACK20-- GI or biliary tracts, pancreas,
-- endocervix, bladder, lung
PLAP
CEAER-- Breast, stomach, Mullerian tract
S100PLAP-- Mullerian tract, breast, stomach, kidney
Seminoma (PLAP; VIM) -- EMA S100CA125-- Mullerian tract, breast, biliary tract
or adrenocortical S100CA19-9-- Mullerian tract, biliary tree, stomach,
carcinoma Or kidney
EMA S100CK20-- Mullerian tract, biliary tree, stomach
sarcoma (VIM; PLAP–) Embryonal CA S100ER-- Breast, Mullerian tract
PLAPCA125-- Mullerian, biliary, or GI tracts
PLAPCA19-9-- Biliary, Mullerian, or GI tracts; lung
RCC or NPC -- PLAPCK20-- Biliary, GI, or Mullerian tracts; pancreas
PLAPER-- Breast, Mullerian tract
ALL NEGATIVE-- Renal cell carcinoma or
Germ cell tumor (rare examples) or
-- PLAP Nasopharyngeal carcinoma
ACC Embryonal CA
Figure 23.26 Algorithm for immunohistochemical diagnosis of epithelial malignancies of unknown origin.
Adjunctive studies of pleural diseases 309
PLAP
Keratin
Germ cell carcinoma
-- Malignant
S100
S100
melanoma
Adrenocortical
-- carcinoma or sarcoma
Vimentin Figure 23.27 Algorithm for
-- Technically unsatisfactory specimen
immunohistochemical diagnosis of oncocytic
malignant neoplasms.
Melanoma
---- S100/HMB-45/MART-1/TYR --
PLAP
Seminoma
Seminoma
--
CD45 VIMENTIN
Clear-cell Non-Hodgkin’s
lymphoma
Lymphoma
--
-- Adrenocortical l
carcinoma or
Clear-cell carcinoma sarcoma
(EMA/CalRet–) or
clear-cell mesothelioma
KERATIN (EMA/CalRet)
or clear-cell HCC or
non-seminomatous germ
-- cell tumor (NSGCT)
--
(both EMA–;
NSGCT also PLAP) Technically
inferior specimen
Figure 23.28 Algorithm for
BER-EP4/CD15/CEA/S100 Clear-cell immunohistochemical diagnosis of clear-
carcinoma, NOS
cell malignant neoplasms.
selected immunodeterminants; i.e. they appear at more results they produce are coupled with others that are
than one place in the algorithms. This is intentional, so linked to highly lineage-restrictive reagents. To state this
that the natural biological variability of certain neoplasms tenet another way, algorithmic ‘layering’ is a way of
is adequately represented. Another critical principle of obtaining specific information from relatively non-specific
algorithmic interpretation is that one moves from data markers.
with a relatively high predictive value to those with lesser Antibody reagents in these algorithms were chosen,
predictive values. If carried out properly, however, this based on their reliability in the author’s 15 years of prac-
‘layering’ paradigm contextually enhances the specificity tice. However, it should not be taken that this approach is
of antibodies that label a variety of neoplasms, because the the only workable one for the assessment of MTUOTs.
310 Pathology: cytology
Flow cytometry (FCM) is valuable in the assessment of ● Special processing is recommended if flow
lymphohematological infiltrates of the pleural space. The cytometry or cytogenetic studies of pleural effu-
clonality of constituent cells can be determined quite easily sions are desired. The specimen should be cen-
by such means if >5 mL of properly-prepared fresh speci- trifuged and a cellular pellet should be suspended
men is submitted, and a B-lymphocytic, T-lymphocytic or in nutrient medium for such evaluations.
myeloid lineage can be assigned based on cell-surface ● Pleural biopsies are much better acquired with
marker characteristics.25 If necessary, lymphoid gene video-assisted thoracoscopic technique than by
rearrangement analyses can be performed. ‘closed’ needle sampling.
Flow cytometry has a limited role in the assessment of ● Benign-reactive and malignant pleural lesions
pleural diseases. Cytometric evaluations has inadequate can be remarkably similar cytologically, regard-
accuracy in distinguishing benign from malignant effu- less of their cellular lineages. Hence, this area is a
sions,94–97 or adenocarcinoma from mesothelioma.98,99 difficult one in cytopathological diagnosis.
FCM studies with rather small cohorts suggested that ● Malignant tumors can be generically divided on
DNA-aneuploidy and an S (proliferative) phase of the cell cytological grounds into small round-cell, large
cycle (SPF) number <5 or 6 may correlate with shortened polygonal-cell and spindle-cell-pleomorphic
survival in mesothelioma.100–103 In that vein, it should be lesions. This system facilitates differential diag-
noted that a measurement equivalent to that of the SPF nosis.
can be obtained immunohistochemically with the anti- ● Immunohistochemical algorithms can be linked
body known as MIB-1/Ki-67.104 with cytological categories to yield logical
schemes for diagnostic interpretation. Flow
cytometry is best applied to pleural effusions in
Cytogenetic evaluation of pleural diseases the evaluation of clonality (malignancy) of
hematolymphoid proliferations. It has much less
Cytogenetic analyses, by classical karyotypic methods, flu- utility in the differential diagnosis and prognosti-
orescence in-situ hybridization, comparative genomic cation of ‘solid’ tumors.
hybridization, or other molecular methods, have con- ● Cytogenetic evaluation is a helpful adjunct in the
tributed to our understanding of the subcellular alter- characterization of hematolymphoid and mes-
ations in various human malignancies. Many enchymal malignancies in pleural fluid.
well-characterized chromosomal markers of lymphoid
neoplasms are recognized, e.g. t(8;14) translocation of
Burkitt’s lymphoma, t(14;18) translocation of follicular
lymphomas, t(2;5) translocation seen in a proportion of REFERENCES
anaplastic large-cell lymphomas, 11q23 deletion seen in
small-lymphocytic lymphoma/chronic lymphocytic ● = Key primary paper
leukemia, translocations affecting 3q27 in some diffuse ◆ = Major review article
large-cell non-Hodgkin’s lymphomas and t(15;17)
translocation in acute promyelocytic leukemia.105–111 Such 1. Antman KH. Clinical presentation and natural history of benign
cytogenetic tests may aid the diagnosis of lymphoma in the and malignant mesothelioma. Semin Oncol 1981; 8: 313–20.
2. Belani CP, Ziskind AA, Dhawan M, Lemmon CC. Management of
pleural fluid. malignant pleural and pericardial effusions. In: Aisner J, Arriagada
Similarly, cytogenetic studies have potential application R, Green MR, Martini N, Perry MC (eds). Comprehensive textbook
to ‘solid’ tumor pathology in the pleura. For example, of thoracic oncology. Baltimore: Williams & Wilkins, 1996;
Ewing’s sarcoma/PNET demonstrates a reproducible 880–905.
t(11;22) translocation that distinguishes it from other 3. Gaensler EA, Kaplan AI. Asbestos pleural effusion. Ann Intern Med
1971; 74: 178–91.
small-cell neoplasms.29,30 4. Luse SA, Reagan JW. A histologic study of effusions. I. Effusions
The cytogenetics of mesothelioma are detailed sepa- not associated with malignant tumors. Cancer 1954; 7: 1155–66.
rately in this text (see Chapter 11). 5. Reda MG, Raigelman W. Pleural effusion in systemic lupus
erythematosus. Acta Cytol 1980; 24: 553–56.
●6. Das DK, Gupta SK, Ayyagari S, et al. Pleural effusions in non-
Hodgkin’s lymphomas: a cytomorphologic, cytochemical, and
immunologic study. Acta Cytol 1987; 31: 119–24.
7. Leong ASY, Suthipinatwong C, Vinyuvat S. Immunostaining of
cytologic preparations: a review of technical problems. Appl
Immunohistochem Mol Morphol 1999; 7: 214–20.
8. Galan GG, Tarrazona-Hervas V, Morcillo-Aixela A, et al. The
indications for and results of videothoracoscopic surgery:
References 311
reflections on 152 procedures. Arch Bronchopneumonol 1999; 35: small cell tumor of the thoracopulmonary region (Askin tumor).
477–482. Acta Cytol 1994; 38: 702–6.
9. Sakuraba M, Masuda K, Hebisawa A, Sagara Y, Komatsu H. 34. Ordonez NG. Desmoplastic small round cell tumor. I. A
Diagnostic value of thoracoscopic pleural biopsy for pleurisy under histopathologic study of 39 cases with emphasis on unusual
local anesthesia. ANZ J Surg 2006; 76: 722–4. histological patterns. Am J Surg Pathol 1998; 22: 1303–13.
10. Matsubara O, Mark EJ, Ritter JH: Pseudoneoplastic lesions of the 35. Wolf AN, Ladanyi M, Paull G, Balugrund JE, Westra WH. The
lungs, pleural surfaces, and mediastinum. In: Wick MR, Humphrey expanding clinical spectrum of desmoplastic small round-cell
PA, Ritter JH (eds). Pathology of pseudoneoplastic lesions. tumor: a report of two cases with molecular confirmation. Hum
Philadelphia: Lippincott-Raven, 1997; 97–129. Pathol 1999; 30: 430–435.
11. Kutty CPK, Remeniuk E, Varkey B. Malignant-appearing cells in 36. Parkash V, Gerald WL, Parma, Miettinen M, Rosai J. Desmoplastic
pleural effusion due to pancreatitis. Acta Cytol 1981; 25: small round cell tumor of the pleura. Am J Surg Pathol 1995; 19:
412–16. 659–65.
●12. Ashton PR, Hollingsworth AS, Johnston WW: The cytopathology of 37. Bian Y, Jordan AG, Rupp M, et al. Effusion cytology of
metastatic breast cancer. Acta Cytol 1975; 19: 1–6. desmoplastic small round cell tumor of the pleura. Acta Cytol
13. Danner DE, Gmelich JT. A comparative study of tumor cells from 1993; 37: 77–82.
metastatic carcinoma of the breast in effusions. Acta Cytol 1975; 38. Argani P, Askin FB, Colombani P, Perlman EJ. Occult pulmonary
19: 509–18. synovial sarcoma confirmed by molecular techniques. Pediatr Dev
14. Mallonee MM, Lin F, Hassanein K. A morphologic analysis of the Pathol 2000; 3: 87–90.
cells of ductal carcinoma of the breast and of adenocarcinoma of ◆39. Hajdu SI, Hajdu EO. Exfoliative cytology of malignant
the ovary in pleural and abdominal effusions. Acta Cytol 1987; 31: lymphoreticular, soft tissue, and bone neoplasms. Pathol Annu
441–7. 1976; 11: 317–34.
15. Spriggs AI, Jerrome DW. Intracellular mucus inclusions: a feature 40. Nishikawa A, Tanaka T, Kanai N, et al. Exfoliative cytopathology of
of malignant cells in effusions in the serous cavities, particularly alveolar rhabdomyosarcoma. Acta Pathol Jpn 1987; 37:
due to carcinoma of the breast. J Clin Pathol 1975; 28: 929–36. 1003–1007.
16. Yamada T, Itou V, Watanabe Y, et al. Cytologic diagnosis of 41. Wong JW, Pitlik D, Abdul-Karim FW. Cytology of pleural,
malignant melanoma. Acta Cytol 1972; 16: 70–6. peritoneal, and pericardial fluids in children. A 40-year summary.
17. Hajdu SI, Savino A. Cytologic diagnosis of malignant melanoma. Acta Cytol 1997; 41: 467–73.
Acta Cytol 1973; 11: 320–7. 42. Helson L, Krochmal P, Hajdu SI. Diagnostic value of cytologic
●18. Longatto-Filho A, deCarvalho LV, Santos GD, et al. Cytologic specimens obtained from children with cancer. Ann Clin Lab Sci
diagnosis of melanoma in serous effusions: a morphologic and 1975; 5: 294–7.
immunocytochemical study. Acta Cytol 1995; 39: 481–4. 43. Farr GH, Hajdu SI. Exfoliative cytology of metastatic
19. Denton KJ, Stretch J, Athanasou N. Osteoclast-like giant cells in neuroblastoma. Acta Cytol 1972; 16: 203–6.
malignant melanoma. Histopathology 1992; 20: 179–181. 44. Sahn SA. Pleural effusion in lung cancer. Clin Chest Med 1993; 14:
20. Suster S, Moran CA. Unusual manifestations of metastatic tumors 189–200.
to the lungs. Semin Diagn Pathol 1995; 12: 193–206. 45. Banner BF, Warren WH, Gould VE. Cytomorphology and marker
21. Nakhleh RE, Wick MR, Rocamora A, Swanson PE, Dehner LP. expression of malignant neuroendocrine cells in pleural effusions.
Morphologic diversity in malignant melanomas. Am J Clin Pathol Acta Cytol 1986; 30: 99–104.
1990; 93: 731–40. ●46. Mayall FG, Gibbs AR. The histology and immunohistochemistry of
22. Roberts GH, Campbell GH. Exfoliative cytology of diffuse small-cell mesothelioma. Histopathology 1992; 20: 47–51.
mesothelioma. J Clin Pathol 1972; 25: 557–82. ●47. Nappi O, Ferrara G, Wick MR. Neoplasms composed of eosinophilic
●23. Ansari MQ, Dawson DB, Nador R, et al. Primary body cavity-based polygonal cells: an overview with consideration of different
AIDS-related lymphomas. Am J Clin Pathol 1996; 105: 221–9. cytomorphologic patterns. Semin Diagn Pathol 1999; 16: 82–90.
24. Melamed MR. The cytological presentation of malignant 48. Rosendale BE, Dusenberry D. Cytology of hepatocellular carcinoma
lymphomas and related diseases in effusions. Cancer 1963; 16: in serous fluids: a report of three cases. Diagn Cytopathol 1996;
413–31. 15: 127–31.
●25. O’Hara MF, Cousar JB, Glick AD, et al. Multiparametric approach to 49. Falconieri G, Zanconati F, Colautti I, et al. Effusion cytology of
the diagnosis of hematopoietic-lymphoid neoplasms in body hepatocellular carcinoma. Acta Cytol 1995; 39: 893–7.
fluids. Diagn Cytopathol 1985; 1: 33–8. 50. Hajdu SI, Koss LG. Cytologic diagnosis of metastatic myosarcomas.
●26. Spriggs AE, Vanhegan RI. Cytological diagnosis of lymphoma in Acta Cytol 1969; 13: 545–51.
serous effusions. J Clin Pathol 1981; 34: 1311–25. 51. Wick MR, Ritter JH, Dehner LP. Malignant rhabdoid tumors: a
◆27. Cerilli LA, Wick MR. Metastatic malignancies of unknown origin: a clinicopathologic review and conceptual discussion. Semin Diagn
histological and cytological approach to diagnosis. Pathol Case Pathol 1995; 12: 233–48.
Rev 2001; 6: 137–45. 52. Puttagunta L, Vriend RA, Nguyen GK. Deciduoid epithelioid
28. Wick MR, Cerilli LA. Applications of immunohistochemistry in the mesothelioma of the pleura with focal rhabdoid change. Am J
diagnosis of undifferentiated tumors. In: Lloyd RV (ed.). Surg Pathol 2000; 24: 1440–3.
Morphology methods. Totowa: Humana Press, 2001; 323–60. ◆53. Nappi O, Mills SE, Swanson PE, Wick MR. Clear cell tumors of
29. Askin FB, Perlman EJ. Neuroblastoma and peripheral unknown nature and origin: a systematic approach to diagnosis.
neuroectodermal tumors. Am J Clin Pathol 1998; 109 (Suppl): Semin Diagn Pathol 1997; 14: 164–74.
S23–30. 54. Humphrey PA. Clear cell neoplasms of the urinary tract and male
●30. Winters JL, Geil JD, O’Connor WN. Immunohistology, cytogenetics, reproductive system. Semin Diagn Pathol 1997; 14: 240–52.
and molecular studies of small round cell tumors of childhood: a 55. Ritter JH, Mills SE, Gaffey MJ, Nappi O, Wick MR. Clear cell tumors
review. Ann Clin Lab Sci 1995; 25: 66–78. of the alimentary tract and abdominal cavity. Semin Diagn Pathol
31. DeAlava E, Pardo J. Ewing tumor: tumor biology and clinical 1997; 14: 213–19.
applications. Int J Surg Pathol 2001; 9: 7–17. 56. Matias-Guiu X, Lerma E, Prat J. Clear cell tumors of the female
32. Sahu K, Pai RR, Khadilkar UN. Fine needle aspiration cytology of genital tract. Semin Diagn Pathol 1997; 14: 233–9.
the Ewing’s sarcoma family of tumors. Acta Cytol 2000; 44: 57. Murakata LA, Ishak KG, Nzeako UC. Clear cell carcinoma of the
332–36. liver: a comparative immunohistochemical study with renal clear-
33. Kumar PV. Fine needle aspiration cytologic findings in malignant cell carcinoma. Mod Pathol 2000; 13: 874–81.
312 Pathology: cytology
58. Ropp BG, Solomides C, Palazzo J, Bibbo M. Follicular carcinoma of ●81. Saad RS, Lindner JL, Lin X, Liu YL, Silverman JE. The diagnostic
the thyroid with extensive clear-cell differentiation: a potential utility of D2-40 for malignant mesothelioma versus pulmonary
diagnostic pitfall. Diagn Cytopathol 2000; 23: 222–3. carcinoma with pleural involvement. Diagn Cytopathol 2006; 34:
59. Perniciaro C. Dermatopathologic variants of malignant melanoma. 801–6.
Mayo Clin Proc 1997; 72: 273–79. 82. Kahn HJ, Bailey D, Marks A. Monoclonal antibody D2-40, a
●60. Hajdu SI, Nolan MA. Exfoliative cytology of malignant germ cell new marker of lymphatic endothelium, reacts with Kaposi’s
tumors. Acta Cytol 1975; 19: 255–60. sarcoma and a subset of angiosarcomas. Mod Pathol 2002;
61. Azuma T, Nishimatsu H, Nakagawa T, et al. Metastatic renal cell 15: 434–40.
carcinoma mimicking pleural mesothelioma. Scan J Urol Nephrol 83. Lau SK, Weiss LM, Chu PG. D2-40 immunohistochemistry in the
1999; 33: 140–1. differential diagnosis of seminoma and embryonal carcinoma: a
62. Kutty K, Varkey B. Incidence and distribution of intrathoracic comparative immunohistochemical study with KIT (CD117) and
metastases from renal cell carcinoma. Arch Intern Med 1984; 144: CD30. Mod Pathol 2007; 20: 320–5.
273–6. 84. Browning L, Parker A, Bailey D. D2-40 is a sensitive and specific
63. Silverman JF, Dabbs DJ, Finley JL, Geisinger KR. Fine needle marker in differentiating primary adrenal cortical tumors from
aspiration biopsy of pleomorphic (giant cell) carcinoma of the both metastatic clear cell renal cell carcinoma and
pancreas: cytologic, immunocytochemical, and ultrastructural pheochromocytoma. J Clin Pathol 2008; 61: 293–6.
findings. Am J Clin Pathol 1988; 89: 714–20. ●85. Wick MR, Moran CA, Mills SE, Suster S. Immunohistochemical
64. Bonomo L, Feragalli B, Sacco R, Merlino B, Storto ML. Malignant differential diagnosis of pleural effusions, with emphasis on
pleural disease. Eur J Radiol 2000; 34: 98–118. malignant mesothelioma. Curr Opin Pulm Med 2001; 7: 187–92.
●65. Newhouse M. Epidemiology of asbestos-related tumors. Semin 86. Gaertner E, Zeren EH, Fleming MV, Colby TV, Travis WD. Biphasic
Oncol 1981; 8: 250–7. synovial sarcomas arising in the pleural cavity: a clinicopathologic
●66. Peterson JT, Greenberg SD, Buffler P. Non-asbestos-related study of five cases. Am J Surg Pathol 1996; 20: 36–45.
malignant mesothelioma: a review. Cancer 1984; 54: 951–60. 87. Miettinen M, Limon J, Niezabitowski A, Lasota J. Calretinin and
67. Attanoos RL, Gibbs AR. Pathology of malignant mesothelioma. other mesothelioma markers in synovial sarcoma: analysis of
Histopathology 1997; 30: 403–18. antigenic similarities and differences with malignant
68. Wick MR, Mills SE. Mesothelial proliferations: an increasing mesothelioma. Am J Surg Pathol 2001; 25: 610–17.
morphological spectrum. Am J Clin Pathol 2000; 113: 619– 22. 88. Argani P, Zakowski MF, Klimstra DS, Rosai J, Ladanyi M. Detection
69. Carter D, Otis CN. Three types of spindle cell tumors of the pleura: of the SYT-SSX chimeric RNA of synovial sarcoma in paraffin-
fibroma, sarcoma, and sarcomatoid mesothelioma. Am J Surg embedded tissue and its application in problematic cases. Mod
Pathol 1988; 12: 747–53. Pathol 1998; 11: 65–71.
70. Cagle PT, Truong LD, Roggli VL, Greenberg SD: 89. Cury PM, Butcher DN, Corrin D, Nicholson AG. The use of
Immunohistochemical differentiation of sarcomatoid histological and immunohistochemical markers to distinguish
mesotheliomas from other spindle cell neoplasms. Am J Clin pleural malignant mesothelioma and in-situ mesothelioma from
Pathol 1989; 92: 566–71. reactive mesothelial hyperplasia and reactive pleural fibrosis. J
◆71. Rdzanek M, Fresco R, Pass HI, Carbone M. Spindle cell tumors of Pathol 1999; 189: 251–7.
the pleura: differential diagnosis. Semin Diagn Pathol 2006; 23: 90. Mayall FG, Jacobson G, Wilkins R. Mutations of p53 gene and
44–55. SV40 sequences in asbestos-associated and non-asbestos-
72. Khalidi HS, Medeiros LJ, Battifora H. Lymphohistiocytoid associated mesotheliomas. J Clin Pathol 1999; 52: 291–3.
mesothelioma: an often-misdiagnosed variant of sarcomatoid ●91. Stoetzer OJ, Munker R, Darsow M, Wilmanns W. p53-
malignant mesothelioma. Am J Clin Pathol 2000; 113: 649–54. immunoreactive cells in benign and malignant effusions:
●73. Villena V, Lopez-Encuentra A, Echave-Sustaeta J, et al. Diagnostic diagnostic value using a panel of monoclonal antibodies and
value of CA72-4, carcinoembryonic antigen, CA15-3, and CA19-9 comparison with CEA staining. Oncol Rep 1999; 6: 455–8.
assay in pleural fluid: a study of 207 patients. Cancer 1996; 78: ●92. Mangano WE, Cagle PT, Churg A, Vollmer RT, Roggli VL. The
736–40. diagnosis of desmoplastic malignant mesothelioma and its
●74. Mezger J, Stotzer O, Schilli G, Bauer S, Wilmanns W. Identification distinction from fibrous pleurisy: a histologic and
of carcinoma cells in ascitic and pleural fluid: comparison of four immunohistochemical analysis of 31 cases including p53
panepithelial antigens and carcinoembryonic antigen. Acta Cytol immunostaining. Am J Clin Pathol 1998; 110: 191–9.
1992; 36: 75–81. ◆93. Churg A, Colby TV, Cagle PT, et al. The separation of benign and
75. Shimokata K, Totani Y, Nakanishi K, et al. Diagnostic value of malignant mesothelial proliferations. Am J Surg Pathol 2000; 24:
cancer antigen 15-3 (CA15-3) detected by monoclonal antibodies 1183–1200.
(115D8 and DF3) in exudative pleural effusions. Eur Respir J 1988; ●94. Ceyhan BB, Demiralp E, Celikel T. Analysis of pleural effusions
1: 341–4. using flow cytometry. Respiration 1996; 63: 17–24.
●76. Romero S, Fernandez C, Arriero JM, et al. CEA, CA15-3, and CYFRA ●95. Joseph MG, Banerjee D, Harris P, Gibson S, McFadden RG.
21-1 in serum and pleural fluid of patients with pleural effusions. Multiparameter flow cytometric DNA analysis of effusions: a
Eur Respir J 1996; 9: 17–23. prospective study of 36 cases compared with routine cytology and
●77. DeYoung BR, Wick MR. Immunohistologic analysis of metastatic immunohistochemistry. Mod Pathol 1995; 8: 686–93.
carcinomas of unknown origin: an algorithmic approach. Semin 96. Pinto MM. DNA analysis of malignant effusions: comparison with
Diagn Pathol 2000; 17: 184–93. cytologic diagnosis and carcinoembryonic antigen content. Anal
◆78. Ordonez NG. What are the current best immunohistochemical Quant Cytol Histol 1992; 14: 222–6.
markers for the diagnosis of epithelioid mesothelioma? A review 97. Nance KV, Silverman JF. The utility of ancillary techniques in
and update. Hum Pathol 2007; 38: 1–16. effusion cytology. Diagn Cytopathol 1992; 8: 185–9.
79. Kachali C, Eltoum I, Horton D, Chhieng DC. Use of mesothelin as a 98. Esteban JM, Sheibani K. DNA ploidy analysis of pleural
marker for mesothelial cells in cytologic specimens. Semin Diagn mesotheliomas: its usefulness for their distinction from lung
Pathol 2006; 23: 20–4. adenocarcinomas. Mod Pathol 1992; 5: 626–30.
80. Li Q, Bavikatty N, Michael CW. The role of immunohistochemistry 99. El-Naggar AK, Ordonez NG, Garnsey L, Batsakis JG. Epithelioid
in distinguishing squamous cell carcinoma from mesothelioma and pleural mesotheliomas and pulmonary adenocarcinomas: a
adenocarcinoma in pleural effusions. Semin Diagn Pathol 2006; comparative DNA flow-cytometric study. Hum Pathol 1991; 22:
23: 15–19. 972–8.
References 313
●100. Emri S, Akbulut H, Zorlu F, et al. Prognostic significance of flow molecular features of B-cell lymphoma. Hematology 2000; 1:
cytometric DNA analysis in patients with malignant pleural 180–204.
mesothelioma. Lung Cancer 2001; 33: 109–14. 106. Kuppers R, Dalla-Favera R. Mechanisms of chromosomal
101. Isobe H, Sridhar KS, Doria R, et al. Prognostic significance of DNA translocation in B-cell lymphomas. Oncogene 2001; 20:
aneuploidy in diffuse malignant mesothelioma. Cytometry 1995; 5580–94.
19: 86–91. 107. Karnolsky IN. Cytogenetic abnormalities in chronic lymphocytic
102. Pyrhonen S, Laasonen A, Tammilehto L, et al. Diploid leukemia. Folia Med 2000; 42: 5–10.
predominance and prognostic significance of S-phase cells in ●108. Hegde U, Wilson WH. Gene expression profiling of lymphomas.
malignant mesothelioma. Eur J Cancer 1991; 27: 197–200. Curr Oncol Rep 2001; 3: 243–249.
103. Dazzi H, Thatcher N, Hasleton PS, Chatterjee AK, Lawson RA. DNA 109. Appelbaum FR, Rowe JM, Radich J, Dick JE. Acute myeloid
analysis by flow cytometry in malignant pleural mesothelioma: leukemia. Hematology 2001; 2: 62–86.
relationship to histology and survival. J Pathol 1990; 162: 51–5. 110. Lin RJ, Sternsdorf T, Tini M, Evans RM. Transcriptional regulation in
104. Comin CE, Anichini C, Boddi V, Novelli L, Dini S. MIB-1 acute promyelocytic leukemia. Oncogene 2001; 20: 7204–15.
proliferation index correlates with survival in pleural malignant 111. Padua RA, McGlynn A, McGlynn H. Molecular, cytogenetic, and
mesotheliomas. Histopathology 2000; 36: 26–31. genetic abnormalities in MDS and secondary AML. Cancer Treat
105. Macintyre E, Willerford D, Morris SW. Non-Hodgkin’s lymphoma: Res 2001; 108: 111–57.
This page intentionally left blank
24
Effusions from cardiac diseases
fluid formed correlated with the amount of pulmonary ≥20 mmHg. In contrast, no significant difference in right
edema. Pleural effusions did not develop unless pul- atrial pressure was noted between those with and without
monary edema was present. Broaddus et al.14 produced pleural effusions. Subsequently, these investigators studied
pulmonary edema in sheep by infusing them with large a group of 18 patients with right atrial hypertension due to
volumes of saline and found that as much as 25 percent of pulmonary vascular disease and normal pulmonary capil-
the infused volume escaped into the pleural cavity. The lary wedge pressures.22 Pleural effusions were not found in
resolution of alveolar edema requires the reabsorption of any of these patients with isolated right heart failure. Thus,
the alveolar fluid into the interstitium of the lung for sub- while right ventricular failure may facilitate the formation
sequent removal by the pulmonary lymphatics.15 This of pleural effusions, left ventricular failure appears to be
reabsorption of alveolar fluid into the interstitium of the essential for their development.
lung raises lung interstitial pressure and promotes the leak The pathogenesis of pleural effusions in patients with
of fluid from the subpleural lung through the visceral pericardial disease is less clear. In some patients, it is likely
mesothelium into the pleural space. The mesothelium per that the pericardium and pleura are affected by the same
se is a very porous membrane that offers little resistance to process. For example, the incidence of pleural effusions is
the flow of fluid.16 high in patients with tuberculous pericarditis, and infec-
Acute elevations in systemic venous pressure produced tion of the mesothelium lining of both the pericardial and
by volume loading and vena caval obstruction can also pleural cavities is usually present.9 Simultaneous involve-
induce pleural effusions in experimental animals.17,18 The ment of the pericardium and pleura probably accounts for
effusions which develop are a consequence of increased fil- the finding of malignant effusions in over one-third of
tration from the pleural capillaries and decreased reab- patients with malignant pericardial disease.8 Malignant
sorption due to the combined effects of increased pleural invasion of the mediastinal lymph nodes, through which
capillary hydrostatic pressure and lymphatic outflow the pleural and pericardial lymphatics drain, contributes
obstruction of the pleural lymphatics which ultimately to the simultaneous development of pericardial and
empty into the superior vena cava. Lowering of the plasma pleural effusions. However, the frequent occurrence of
oncotic pressure due to saline infusion undoubtedly is yet unilateral left-sided pleural effusions in those with idio-
another factor contributing to the formation of pleural pathic and acute inflammatory pericardial disease is more
fluid in these animals. However, hypoalbuminemia in the difficult to explain.7 The left sided effusion may represent
absence of venous hypertension is unlikely to cause pleural a sympathetic response to contiguous inflammation or be
effusions in humans.19 When the obstruction to lymphatic due to an impairment in lymphatic drainage. The former
outflow is chronic, the lymphatic pumping mechanism mechanism probably accounts for the simultaneous pres-
can compensate for the venous obstruction.20 Once lym- ence of both pericardial and pleural effusions in patients
phatic flow is restored, both the edema fluid in the lung with acute pancreatitis.23
and the associated pleural effusions resolve.21
The clinical correlate of these experimental studies is
that pleural effusions are common in patients with left CLINICAL PRESENTATION
heart failure but rare in those with right heart failure.
Weiner-Kronish and colleagues4 prospectively studied 37 The clinical presentation is generally typical of patients
patients admitted to a coronary care unit with congestive with congestive heart failure. Increasing dyspnea, orthop-
heart failure and found that 19 (51 percent) had pleural nea and edema are the most common presenting symp-
effusions. The pulmonary capillary wedge pressure was sig- toms. Patients with heart failure rarely complain of
nificantly greater and alveolar edema was more common in pleuritic pain. Frequently, the effusions may be asympto-
the patients with pleural effusions (Table 24.1). Sixteen of matic radiographic findings in a patient with known con-
19 patients with pleural effusions (but only 3 of 18 without gestive heart failure. Physical examination commonly
effusions) had a pulmonary capillary wedge pressure reveals evidence of biventricular failure with distended
neck veins, rales, an S3 gallop and peripheral edema. in 7 percent of patients with heart failure, a difference that
Dullness to percussion at the lung base, decreased fremitus probably is not clinically significant.5,6 A superimposed
and egophony indicate the presence of a pleural effusion. process such as thromboembolism should be suspected
In contrast to the painless dyspnea of the patient with whenever a unilateral effusion is found in a patient with
heart failure, patients with pericarditis typically complain heart failure. The presence of a large left-sided effusion in
of chest pain that may be pleuritic in nature as well as a patient with cardiomegaly should raise the suspicion of
dyspnea.24 A pericardial friction rub is frequently audible pericarditis.3,7
and the electrocardiogram (ECG) may display the typical When pleural effusions develop in a patient with heart
diffuse ST-segment elevations. Low voltage on the ECG failure, the fluid is usually free flowing and will layer on a
and distant heart sounds in a patient with cardiomegaly decubitus radiograph. However, unusual loculations may
are highly suggestive but uncommon clinical findings.24 occur, particularly in the presence of previous pleural
Hypotension, tachycardia and pulsus paradoxus may disease. The presence of a pseudotumor which might be
signal impending tamponade. When the pericarditis is one mistaken for a lung mass may be suspected by its charac-
manifestation of a systemic disease such as lupus erythe- teristic convex radiographic appearance within a fissure
matosus or rheumatoid arthritis, other stigmata of the and confirmed if the ‘mass’ disappears with diuretic
underlying disorder may be readily appreciated. therapy (Figure 24.1).29
RADIOLOGY
patients. Nonetheless, these more specific indicators still prompt a search for an etiology other than congestive
would have misclassified 20 percent of the effusions as exu- heart failure.
dates. It is clear that when thoracentesis is delayed for Exudates are much more likely when patients with peri-
several days while the patient is being treated for congestive cardial disease develop an effusion. In general, the pleural
heart failure, the results of pleural fluid chemistries must be and pericardial fluid will have the same characteristics.
interpreted with caution. The presence of a plasma to Indeed, examination of the pleural fluid can often establish
pleural fluid albumin gradient greater than 1.2 g/dL indi- the nature of the underlying problem. The finding of
cates that the effusion was formed as a transudate malignant cells or positive acid-fast smears of the pleural
irrespective of whether or not it meets traditional protein fluid will indicate the probable cause of the pericardial
and LDH criteria at the time of thoracentesis.42 disease.8,9 High anti-nuclear antibody (ANA) titers may be
B-type natriuretic peptide (BNP) is secreted by the found in both the pericardial and the pleural fluid from
cardiac ventricles in response to acute distention. Porcel patients with lupus erythematosis.48 Pericardial effusions
and colleagues43 found that BNP levels greater than due to heart failure do occur; Kataoka et al.5 found pericar-
1500 pg/mL in the pleural fluid distinguished 44 patients dial effusions in 20 percent of patients with decompen-
with effusions due to heart failure from 73 controls with sated congestive heart failure, whereas 87 percent of these
hepatic hydrothorax or exudates of varying etiology. patients had pleural effusions. None of the pericardial
Gegenhuber et al.44 found that plasma BNP levels were ele- effusions were large. Both the pericardial and pleural fluid
vated in patients with effusions due to heart failure and have the characteristics of a transudate.
that the elevated plasma levels persisted at 24 hours despite
removing most of the effusion by thoracentesis. Since
pleural fluid originates as an ultrafiltrate of plasma, it is TREATMENT
not surprising that the plasma and pleural fluid BNP levels
are highly correlated.45 An elevated BNP level may help The pleural membranes per se are intact in the patient with
identify the cardiac origin of an effusion in a patient with congestive heart failure so that restoration of a normal
heart failure, even if the effusion meets one or more of balance of Starling forces should permit the reabsorption
Light’s criteria for an exudate after diuresis. of the pleural fluid. Diuretics, afterload reduction and dig-
Because congestive heart failure is such a common con- italis are the mainstays of therapy. Howard and Dunn
dition, it is not uncommon to encounter a patient who has demonstrated that aggressive diuretic therapy with contin-
an exudative effusion caused by a coexistent condition uous IV furosemide will reduce the length (and associated
such as pneumonia. Generally, the etiology of the effusion cost) of hospital stay without any increase in morbidity.49
is obvious in such cases. Gotsman et al.46 reported on 47 If treatment of the underlying heart failure is successful,
patients with exudative effusions in the setting of conges- the effusion will resolve.
tive heart failure. Infection, cardiac surgery and malig- Since they often have the same etiology, the pleural effu-
nancy accounted for most of the identified etiologies. sions in patients with pericardial disease will respond to
However, 20 of the 47 patients had no other obvious cause treatment of the underlying process. Anti-inflammatory
for their exudative effusions. Patients with no obvious therapy for serositis and anti-tuberculous therapy for the
cause were more likely to have received acute diuretic patient with tuberculosis will effectively treat both the
therapy compared with those with transudates. pleural effusion and the pericardial disease. However, if
We recently reviewed our experience with 770 patients there is clinical evidence of tamponade, immediate pericar-
seen at our medical center with congestive heart failure dial drainage is necessary and can be lifesaving. Malignant
and radiographic evidence of pleural effusions.47 The pericardial disease usually signals far advanced disease and
majority had typical clinical presentation and were treated often requires definitive local therapy for its control.
medically without thoracentesis. Only 175 had a thoracen- Pericardial sclerosis with doxycycline or bleomycin may
tesis and in this select group, 89 patients had exudates. prevent the reaccumulation of fluid in about three-quarters
Exudates were significantly more common in those who of treated patients.50–52 Alternatively, a pericardial window
had a remote history of coronary artery bypass grafting can be created to allow the fluid to drain into the pleural or
(CABG). A non-cardiac cause was ultimately identified in peritoneal cavities or the preperitoneal subcutaneous
77 of these patients including 11 who had undergone space.53–56 Pericardiectomy may be required for the patient
CABG one or more years prior to the development of the with symptomatic constrictive pericardial disease.11
effusion. The effusions in four of the remaining patients When a patient with large pleural effusion remains dys-
could be explained by red blood cell (RBC) contamination pneic despite intensive medical therapy, therapeutic thora-
during a traumatic tap, leaving only eight patients of the centesis is indicated. Often the removal of 500–1000 mL of
original 770 patients with unexplained exudates. All of fluid may produce immediate and dramatic symptomatic
these patients had received diuretic therapy prior to thora- relief, even before there is an increase in vital capacity or
centesis. Thus, unexplained exudates are rare in patients the partial pressure of arterial O2 (PaO2). The beneficial
with congestive heart failure. The presence of an exudate effects of therapeutic thoracentesis include a reduction in
in a patient who has not had previous chest surgery should the resting volume of the chest wall which enables the
320 Effusions from cardiac diseases
of left-sided effusions with pericardial disease, remains to pleural effusions to pulmonary hemodynamics in patients with
be elucidated. BNP is a promising marker indicating a congestive heart failure. Am Rev Respir Dis 1985; 132: 1253–6.
5. Kataoka H. Pericardial and pleural effusions in decompensated
cardiac contribution to a pleural effusion. Cardiac disease chronic heart failure. Am Heart J 2000; 139: 918–23.
is so common in our society that physicians can expect to 6. Race GA, Scheifly CH, Edwards JE. Hydrothorax in congestive heart
encounter patients with an effusion caused by cardiac failure. Am J Med 1957; 22: 83–9.
failure and a concomitant second problem contributing to ●7. Weiss JM, Spodick DH. Association of left pleural effusion with
the effusion. While additional testing such as cytological pericardial disease. N Engl J Med 1983; 308: 696–7.
8. DeCamp MM, Mentzer SJ, Swanson SJ, et al. Malignant effusive
examination may identify the second problem, one still disease of the pleura and pericardium. Chest 1997; 112: 291S–5.
must rely on clinical intuition and the recognition of an 9. Casas E, Blanco JR, Ibarra V, et al. Incidence of pericardial
atypical presentation to prompt additional testing. A infusion in pulmonary tuberculosis. Int J Tuberc Lung Dis 2000; 4:
marker for the ‘complicated’ cardiac effusion is needed. 1173–5.
10. Hara KS, Ballard DJ, Ilstrup DM, et al. Rheumatoid pericarditis:
Clinical features and survival. Medicine 1990; 69: 81–91.
11. Kelly CA, Bourke JP, Malcolm A, et al. Chronic pericardial disease
in patients with rheumatoid arthritis: a longitudinal study. Q J
KEY POINTS Med 1990; 75: 461–70.
12. Kinasewitz GT, Groome LJ, Marshall RP, et al. Role of pulmonary
● Congestive heart failure is the most common lymphatics and interstitium in visceral pleural fluid exchange. J
Appl Physiol 1984; 56: 355–63.
cause of transudative pleural effusions.
13. Allen S, Gabel J, Drake R. Left atrial hypertension causes pleural
● Pleural effusions caused by heart failure are bilat- effusion formation in unanesthetized sheep. Am J Physiol 1989;
eral in over 80 percent of patients. 257: H690–2.
● A large left-sided effusion in a patient with car- 14. Broaddus VC, Wiener-Kronish JP, Staub NC. Clearance of lung
diomegaly should raise suspicion of pericardial edema into the pleural space of volume-loaded anesthetized
sheep. J Appl Physiol 1990; 68: 2623–30.
disease.
15. Staub NC. Pulmonary edema. Physiol Rev 1974; 54: 678–809.
● Right heart failure facilitates the development of 16. Payne DK, Kinasewitz GT, Gonzalez E. Comparative permeability of
pleural effusions, but left heart failure is essential. canine visceral and parietal pleura. J Appl Physiol 1988; 65:
2558–64.
References 321
17. Mellins RB, Levine OR, Fishman AP. Effect of systemic and diuretics on the concentration of proteins and other components
pulmonary venous hypertension on pleural and pericardial fluid of pleural transudates in patients with heart failure. Am J Med
accumulation. J Appl Physiol 1970; 29: 564–9. 2001; 110: 681–6.
18. Pang LM, Mellins RB, Rodriguez-Martinez F. Effect of acute 42. Broaddus VC. Diuresis and transudative effusions – Changing the
lymphatic obstruction on fluid accumulation in the chest in dogs. rules of the game. Am J Med 2001; 110: 732–5.
J Appl Physiol 1975; 39: 985–9. ●43. Porcel JM, Vives M, Cao G, et al. Measurement of pro-brain
19. Eid A, Keddissi JI, Kinasewitz GT. Hypoalbuminemia as a cause of natriuretic peptide in pleural fluid for the diagnosis of pleural
pleural effusions. Chest 1999; 115: 1066–9. effusions due to heart failure. Am J Med 2004; 116: 417–20.
20. Szabó G, Magyar Z. Effect of increased systemic venous pressure 44. Gegenhuber A, Mueller T, Dieplinger B, et al. Plasma B-type
on lymph pressure and flow. Am J Physiol 1967; 212: 1469–74. natriuretic peptide in patients with pleural effusions: preliminary
21. Allen SJ, Drake RE, Laine GA, et al. Effect of thoracic duct drainage observations. Chest 2005; 128: 1003–9.
on hydrostatic pulmonary edema and pleural effusion in sheep. J 45. Kolditz M, Halank M, Schiemanck CS, et al. High diagnostic
Appl Physiol 1991; 71: 314–16. accuracy of NT-proBNP for cardiac origin of pleural effusions. Eur
22. Wiener-Kronish JP, Goldstein R, Matthay RA, et al. Lack of Respir J 2006; 28: 144–50.
association of pleural effusion with chronic pulmonary arterial and 46. Gotsman I, Fridlender Z, Meirovitz A, et al. The evaluation of
right atrial hypertension. Chest 1987; 92: 967–70. pleural effusions in patients with heart failure. Am J Med 2001;
23. Maringhini A, Ciambra M, Patti R, et al. Ascites, pleural, and 111: 375–8.
pericardial effusions in acute pancreatitis: A prospective study of 47. Eid A, Keddissi JI, Samaha M, et al. Exudative effusions in
incidence, natural history, and prognostic role. Dig Dis Sci 1996; congestive heart failure. Chest 2002; 122: 1518–23.
41: 848–52. 48. Wang D-Y, Yang P-C, Yu W-L, et al. Serial antinuclear antibodies
24. Agner RC, Gallis HA. Pericarditis: Differential diagnostic titre in pleural and pericardial fluid. Eur Respir J 2000; 15:
considerations. Arch Intern Med 1979; 139: 407–12. 1106–10.
25. Kataoka H, Takada S. The role of thoracic ultrasonography for 49. PA Howard, MI Dunn. Aggressive diuresis for severe heart failure
evaluation of patients with decompensated chronic heart failure. J in the elderly. Chest 2001; 119: 807–10.
Am Coll Cardiol 2000; 35: 1638–46. 50. Maher EA, Shepherd FA, Todd TJR. Pericardial sclerosis as the
26. Peterman TA, Brothers SK. Pleural effusions in congestive heart primary management of malignant pericardial effusion and
failure and in pericardial disease. N Engl J Med 1983; 309: 313. cardiac tamponade. J Thorac Cardiovasc Surg 1996; 112: 637–43.
27. Woodring JH. Distribution of pleural effusion in congestive heart 51. Vaitkus PT, Hermann HC, LeWinter MM. Treatment of malignant
failure: what is atypical? South Med J 2005; 98: 518–23. pericardial effusion. J Am Med Assoc 1994; 272: 59–64.
28. Rabin CB, Blackman NS. Bilateral pleural effusion: its significance 52. Yano T, Yokoyama H, Inoue T, et al. A simple technique to manage
in association with a heart of normal size. J Mt Sinai Hosp 1957; malignant pericardial effusion with a local instillation of
24: 45–63. bleomycin in non-small cell carcinoma of the lung. Oncology
29. Van Gelderen WFC. Vanishing pleural fluid collections in cardiac 1994; 51: 507–9.
failure simulating lung tumours. Australas Radiol 1994; 38: 53. Hazelrigg SR, Mack MJ, Landreneau RJ, et al. Thorascopic
93–6. pericardiectomy for effusive pericardial disease. Ann Thorac Surg
30. Weingardt JP, Guico RR, Nemcek AA Jr, et al. Ultrasound findings 1993; 56: 792–5.
following failed clinically directed thoracenteses. J Clin Ultrasound ◆54. Piehler JM, Pluth JR, Schaff HV, et al. Surgical management of
1994; 22: 419–26. effusive pericardial disease: Influence of extent of pericardial
31. Wu RG, Yuan A, Liaw YS, et al. Image comparison of real-time resection on clinical course. J Thorac Cardiovasc Surg 1985; 90:
gray-scale ultrasound and color doppler ultrasound for use in 506–16.
diagnosis of minimal pleural effusion. Am J Respir Crit Care Med 55. Olson JE, Ryan MB, Blumenstock DA. Eleven years experience with
1994; 150: 510–14. pericardial–peritoneal window in the management of malignant
32. Matalon TA, Neiman HL, Mintzer RA. Noncardiac chest and benign pericardial effusions. Ann Surg Oncol 1995; 2: 165–9.
sonography. Chest 1983; 83: 675–8. 56. Van Trigt P, Douglas J, Smith PK, et al. A prospective trial of
∗33. Horowitz MS, Schultz CS, Stinson EB. Sensitivity and specificity of subxiphoid pericardiotomy in the diagnosis and treatment of large
echocardiographic diagnosis of pericardial effusion. Circulation pericardial effusion. Ann Surg 1993; 218: 777–82.
1974; 50: 239–47. 57. Estenne M, Yernault JC, De Troyer A. Mechanism of relief of
∗34. Singh S, Wann SL, Schuchard GH, et al. Right ventricular and right dyspnea after thoracocentesis in patients with large pleural
atrial collapse in patients with cardiac tamponade – a combined effusions. Am J Med 1983; 74: 813–19.
echocardiographic and hemodynamic study. Circulation 1984; 70: 58. Sudduth CD, Sahn SA. Pleurodesis for nonmalignant pleural
966. effusions. Chest 1992; 102: 1855–60.
●35. Hall WJ, Mayewski RJ. Diagnostic thoracentesis and pleural biopsy 59. Glazer M, Berkman N, Lafair JS, et al. Successful talc slurry
in pleural effusions. Ann Intern Med 1985; 103: 799–802. pleurodesis in patients with nonmalignant pleural effusion:
36. Sokolowski JW, Burgher LW, Jones FL Jr, et al. Guidelines for Report of 16 cases and review of the literature. Chest 2000; 117:
thoracentesis and needle biopsy of the pleura. Am Rev Respir Dis 1404–9.
1989; 140: 257–8. 60. Davidoff D, Naparstek Y, Eliakim M. The use of pleurodesis for
37. Light RW, MacGregor IM, Luchsinger PC, et al. Pleural effusions: intractable pleural effusion due to congestive heart failure.
The diagnostic separation of transudates and exudates. Ann Intern Postgrad Med J 1983; 59: 330–1.
Med 1972; 77: 507–13. 61. Dorsey JS, Cogordan JA. Pleuroperitoneal shunt for intractable
38. Light RW, Erozan YS, Ball WC Jr. Cells in pleural fluid. Arch Intern pleural effusion. Can J Surg 1984; 27: 598–9.
Med 1973; 132: 854–60. 62. Ponn RB, Blancaflor J, D’Agostino RS, et al. Pleuroperitoneal
39. Shinto RA, Light RW. Effects of diuresis on the characteristics of shunting for intractable pleural effusions. Ann Thorac Surg 1991;
pleural fluid in patients with congestive heart failure. Am J Med 51: 605–9.
1990; 88: 230–4. 63. Cimochowski GE, Joyner LR. Pleural peritoneal shunting for benign
40. Chakko SC, Caldwell SH, Sforza PP. Treatment of congestive heart and malignant pleural effusions. Surg Annu 1989; 21: 49–71.
failure: Its effect on pleural fluid chemistry. Chest 1989; 95: 64. Little AG, Kadowaki MH, Ferguson MK, et al. Pleuro-peritoneal
798–802. shunting: Alternative therapy for pleural effusions. Ann Surg 1988;
41. Romero-Candeira S, Fernandez C, Martin C, et al. Influence of 208: 443–50.
This page intentionally left blank
25
Effusions from malignancy
FRANCISCO RODRIGUEZ-PANADERO
INTRODUCTION PATHOGENESIS
Malignant pleural effusion is one of the most common The pleura can be invaded directly from neighboring
problems faced by clinicians in their everyday practice. structures (lung, chest wall – including breast in some
However, it is less commonly recognized that a pleural cases – diaphragm and mediastinum), but most of the
malignancy can occur without the presence of effusion. pleural malignancies arise, according to two necropsy
This is not unusual in cases of mesothelioma, where the studies, from tumor emboli to the visceral pleura, with sec-
effusion can be very small or even absent from the begin- ondary seeding to the parietal pleura (Figures 25.1 and
ning of the disease. However, metastatic neoplasms can 25.2).1,4 The effusion can develop as a direct consequence
also be present without any effusion in the pleural space. of neoplastic pleural involvement, but it can also occur in
In one autopsy series from our group we found metastatic cases with lymphatic blockade at the mediastinal level
pleural involvement in 29 percent of 191 cases with a (Figure 25.3). This last mechanism was invoked by Meyer
malignant tumor elsewhere, but pleural effusion was in 1966,4 and confirmed by us in another necropsy study
present in little more than half of these cases.1 Therefore, many years later.5 However, the exact mechanisms
the incidence of malignant pleural effusion in this particu- involved in the development of effusion in the presence of
lar autopsy series was 15 percent. This fact would be par- tumor implants in the pleural space are not fully under-
ticularly relevant when ‘dry’ diffuse pleural involvement is stood. It is likely that increased pleural permeability caused
suspected in cases of lung cancer that might otherwise be directly by the tumor and the accompanying inflammation
subjected to resection surgery. Therefore, exploratory are responsible for the development of the effusion.6 In
thoracoscopy is often advocated in those circumstances, some circumstances, a malignant tumor can coexist with
just before proceeding to thoracotomy. an effusion without direct pleural involvement, and those
In a prospective study including 1000 consecutive effusions were defined by Sahn as ‘paramalignant’.7 We
patients who were submitted to thoracentesis, Villena and found this phenomenon in up to 17 percent of patients
coworkers2 found that the etiology of the effusion was having a malignant tumor elsewhere. Pulmonary
malignant in 364 patients (36 percent), and it is widely rec- embolism, pneumonia and lymphatic mediastinal block-
ognized that the incidence of malignant pleural effusion ade were found to be the most frequent causes in a
has increased in recent years. Approximately one-third of necropsy study from our group.8
the malignant pleural effusions are related to lung cancer,
and metastatic carcinoma of the breast is the second most
frequent cause. While mesothelioma incidence began to CLINICAL PRESENTATION
decline a few years ago in some countries that banned
asbestos use early, it is still rising in many countries, and is Most of the malignant effusions present with dyspnea on
not expected to decline before year 2020.3 exertion, which is progressive as the effusion is becoming
324 Effusions from malignancy
RADIOGRAPHIC APPEARANCE OF MALIGNANT indicated. In this setting, the following four conditions
EFFUSIONS should be considered, according to Sahn:11
(a)
differential white cell count typically shows a predomi- ing patients for pleurodesis, and this lack of predictive
nance of either lymphocytes or other mononuclear cells. power was probably associated to the influence of
The presence of neutrophils or eosinophils is much less mesothelioma cases in their study.
common but does not exclude a malignant effusion. In conclusion, pH can be helpful in clinical practice
Adenosine deaminase (ADA) determination is routinely when used in conjunction with the patient’s performance
recommended in countries with medium to high preva- status, primary tumor type and response to therapeutic
lence of tuberculosis. It can yield false positive results in thoracentesis.
some cases of mesothelioma and lymphoma. Pleural fluid cytology is the simplest definitive method
Almost all malignant effusions are exudates, but a few available for obtaining a diagnosis of malignant pleural
can be transudates (approximately 1 percent in our thora- effusion. The diagnostic yield is dependent on such factors
coscopy series). Malignant transudative effusions can as extent of disease and the nature of the primary malig-
occur in lung cancer with obstruction of the mainstem nancy. Thus, in our thoracoscopy series including 545
bronchus, but also in lymphoma and other malignancies.17 patients with malignant pleural effusion, the yield of cytol-
Blockade of lymphatic drainage may be invoked as one of ogy ranged from 80 percent in metastatic breast cancer to
the possible pathogenetic mechanisms,18 but an underly- 70 percent in lung cancer, 56 percent in mesothelioma, 40
ing cardiac failure or other conditions that are usually percent in lymphoma and 20 percent in sarcoma (unpub-
associated with transudative effusions should also be taken lished data) (Figure 25.7). The cytological diagnosis of
into account.19 malignancy can therefore be very difficult in some cases of
Approximately one-third of malignant effusions will mesothelioma and lymphoma, and the diagnosis with
demonstrate a pleural fluid pH of less than 7.30 at presen- these tumors frequently requires a more invasive proce-
tation (118 out of 359 with pH measured at the time of dure, such as thoracoscopy.
diagnostic thoracoscopy in our series); this low pH corre- In our experience, low pH/glucose malignant effusions
lates with glucose values of less than 60 mg/dL. The cause have a significantly higher yield of cytology than those with
of these low glucose, low pH malignant effusions appears normal pH levels (Figure 25.8).
to be an increased tumor burden,20 resulting in decreased Most of the current guidelines recommend the addition
glucose transfer into the pleural space and decreased efflux of a biopsy procedure when a first cytology is negative in
of acidic byproducts of glucose metabolism.21 For that effusions of unknown origin.26 Percutaneous needle pleural
reason, malignant effusions with these properties have biopsy is frequently advised in those circumstances.2,27
been shown to have a higher diagnostic yield on cytology However, closed pleural biopsy adds little to the cytological
and poorer response to pleurodesis (Figure 25.6), and they diagnosis in most cases and this is related to the scarce and
are associated with a decreased overall survival.22,23 When irregular distribution of the tumor lesions in the pleural
considering pleural pH as a predicting factor for survival, cavity when cytology is negative.28 With recent advances of
one has to take into account that patients with mesothe- image techniques some authors prefer computed tomo-
lioma tend to survive longer than those with metastatic graphy (CT)-guided needle biopsy, which could replace
pleural carcinoma;24 they also have a tendency to show a
lower pH because of the marked pleural thickening caused
by mesothelioma. 100
Heffner and coworkers25 found in a metanalysis study
that although pH was by itself an independent predictor of
80
survival, it had insufficient predictive accuracy for select-
Cytology
60
40
20
0
t
ow ng
in
ey
ot lon
al a
s
s
he
rie
ar
Ly om
om
om
ac
ea
ig
dn
Lu
Ov
Co
Ot
om
se
or
Br
li
ph
rc
Ki
he
Sa
n
St
ob
es
Gl
kn
M
Un
P 0.0002
Small effusion
(1/3 of hemithorax)
Cytology Cytology
Cytology Cytology
Chemotherapy? Thoracoscopy
Thoracentesis ultrasound
Successful Unsuccessful
(liquid obtained)
Figure 25.12 Management of apparently large effusions with the mediastinum is not displaced or ipsilaterally shifted.
Management of pleural effusion in special conditions 329
known, chemotherapy could be tried in selected patients, chest X-ray films. The remaining 116 effusions were
but a pleurodesis procedure would be needed in most. If detected on CT or ultrasound examination, or were found
the first cytology is negative, thoracoscopy and talc pleu- only at thoracotomy. Although cytology was positive in 40
rodesis would be strongly recommended as the second percent of the effusions that were visible on chest radi-
step following the initial thoracentesis. ographs, pleural metastases were actually found in up to 75
Approximately two-thirds of patients with malignant percent of cases.
effusions require pleurodesis sooner or later.37 In the pres- If the mediastinum is midline or shows an ipsilateral
ence of a large and recurrent effusion the choice is clearly shift, obstruction of the mainstem bronchus should be
pleurodesis, preferably using talc (see below). This should suspected (see Figure 25.5), and bronchoscopy performed.
be carried out as early as possible in order to prevent the In the remaining cases, we strongly recommend perform-
development of a trapped lung, which could prevent lung ing exploratory thoracoscopy before attempting tumor
re-expansion and make a successful pleurodesis problem- resection, in order to detect unsuspected pleural metas-
atic. tases.40–42
When there appears to be a large effusion and the medi- When the effusion is found at thoracotomy only, one
astinum is midline or ipsilaterally shifted, a contrast CT could think about the possibility of a paramalignant
scan would be recommended as the first approach. The CT pleural effusion (associated with obstructive pneumonitis,
scan can be helpful in choosing bronchoscopy, thora- atelectasis or lymphatic blockade), and resection of the
coscopy or transthoracic needle biopsy (TTNB) to obtain tumor has to be considered. However, the prognosis is
the diagnosis. Exploratory thoracotomy should be the poorer in those patients than in those without pleural effu-
choice if all those procedures fail, but according to my sion.43
experience it is seldom needed for diagnosis in effusions Pleural cytology can be eventually positive without
suspicious of being malignant. macroscopically visible lesions on the pleural surface, and
Diagnostic and therapeutic bronchoscopy is mandatory this appears to be associated with a better prognosis.44 In
if a central bronchial obstruction is suspected (see Figure some cases with tight adhesions between the lung and
25.5). Management of trapped lung can be much more chest wall, tumor resection could be achieved despite pos-
complex: it can be detected on contrast CT scans, and itive cytology.41,45
pleural pressure measurement during thoracentesis might The finding of a positive cytology in pleural lavage per-
be useful. In patients with mediastinum centered or with formed at thoracotomy has been associated with a worse
ipsilateral shift, the likelihood of a precipitous drop in prognosis in cases submitted to resection.46–48
pleural pressure is increased, and either pleural pressure The success rate of pleurodesis in pleural effusion sec-
should be monitored during thoracentesis or only a small ondary to lung cancer appears to be lower than in other
volume of fluid removed. malignancies (65 percent complete success in our thora-
When the effusion is relatively small (less than one- coscopy series, compared with an average 76 percent in the
third of the hemithorax), the question is whether pleu- remaining cases of metastatic carcinomas). Although the
rodesis is necessary. The answer is obvious if the effusion presence of trapped lung might play a role, we found no
remains stable and well tolerated. However, in an unde- clear differences in this respect in our thoracoscopy series.
fined proportion of cases the effusion would enlarge and, Another likely explanation would be that complete lung
as the disease advances, be accompanied by a trapped lung re-expansion is hard to achieve in the presence of lung
at which time successful pleurodesis will become unlikely. cancer, compared with other metastatic malignancies.
(b)
(a)
(c) (d)
Figure 25.13 Malignant mesothelioma in a 75-year-old man with a past history of asbestos exposure. No pleural fluid obtained. (a)
Chest radiograph at presentation (with chest pain). (b and c) Appearance on computed tomography scan. (d) Chest radiograph 17 months
later. Only palliative treatment for chest pain was applied.
Cytological diagnosis is rather difficult with mesothe- but there are a few details that should be taken into
lioma because it is sometimes hard to establish a clear account:
distinction between mesothelioma and metastatic adeno-
carcinoma. Also, differentiating between reactive and ● The aspect of the pleural fluid can be chylous more fre-
malignant mesothelial cells can be difficult. Cytology was quently than in other malignancies because the thoracic
positive for malignancy in 56 percent in our cases of duct can be disrupted by lymphoma involvement:
mesothelioma, but the cytological diagnoses was correct among 51 cases of lymphoma in our thoracoscopy
only in two-thirds of cases. series, we found a chyliform appearance in 15.6 percent
of the cases, while it was present in only 2 percent in our
series overall. A right-sided chylous pleural effusion in
Pleural effusion in lymphoma lymphoma is mostly associated with involvement of the
thoracic duct at the lower part of the paravertebral zone
Pleural involvement by lymphoma can show no significant in the hemithorax (see also Chapter 29, Effusions from
differences with other types of effusions at presentation, lymphatic disruptions).
Considerations on pleurodesis for malignant pleural effusions 331
● Cytological diagnosis can be difficult in lymphoma, median pleural fluid D-dimer levels were higher in the
because a lymphocytic pleural effusion is frequently group requiring pleurodesis.
found and the lymphocytes show a normal appearance
in many cases. Cytology was positive in only 40 percent CONTRAINDICATIONS TO PLEURODESIS IN MPE
of our cases, compared with 63.5 percent in the total
series. Flow cytometry analysis can be helpful in sus- If thoracoscopic pleurodesis is considered, patients with
pected cases of lymphoma.35,36 severe chronic obstructive pulmonary disease (COPD)
● When there is a chylothorax coexisting with lym- and consequent respiratory insufficiency, with hypoxemia
phoma, pleurodesis can be very effective if oral feeding (partial pressure of O2 [PO2] <50 mmHg) and hypercap-
is stopped for a few days before, during and after the nia, will not tolerate induction of a pneumothorax without
pleurodesis attempt (we had a 72 percent successful talc further deterioration of the gas exchange, and are therefore
pleurodesis in our lymphoma series, including those not suitable candidates for thoracoscopy. Patients with
with chyliform effusion). unstable cardiovascular status should not undergo thora-
coscopy. Moreover, thoracoscopic pleurodesis should not
be attempted in patients with a significant contralateral
involvement of the lung, since they are likely to develop an
CONSIDERATIONS ON PLEURODESIS FOR acute respiratory insufficiency (Figure 25.14). The pres-
MALIGNANT PLEURAL EFFUSIONS ence of a bilateral pleural effusion might also be problem-
atic, unless all the effusion is removed from the
Pleurodesis is addressed separately in Chapter 46. Here, I contralateral side before attempting pleurodesis on one
emphasize the key points regarding its role in the manage- side. Simultaneous pleurodesis should not be attempted
ment of malignant pleural effusions (MPE). since the likelihood of developing complications is high.
When talc poudrage is not feasible, other alternatives
should be considered, including talc slurry or doxycycline
Indications and contraindications for application through the chest tube after removing the
pleurodesis pleural fluid.
(a)
(b)
Moreover, iodide might provoke severe adverse effects PERSISTENT AIR LEAK
when instilled into the pleural space.55
Air leak can occur during lung re-expansion,56 especially in
patients with necrotic tumor nodules in the visceral pleura.
Side effects and complications of pleurodesis In our experience, this especially occurs in patients that
have been submitted to previous chemotherapy, even if no
Pain and transient fever, due to release of pro-inflamma- biopsies of the visceral pleural were taken (Figure 25.15).
tory mediators, are common side effects associated with
pleurodesis performed with practically any sclerosant. ACUTE RESPIRATORY DISTRESS OR PNEUMONITIS
However, there are other worrying complications that have
been reported with the procedure, as discussed below. Acute respiratory distress or pneumonitis has been
described in some cases of talc pleurodesis.50,57–59 The
RE-EXPANSION EDEMA precise pathophysiological mechanism responsible for this
severe complication is still unclear, but it appears that a
In order to prevent re-expansion lung edema, careful and high dose of talc used might have played a significant role
graded suction should be applied. We usually leave the in some cases. Also, the size of talc particles used for pleu-
drain connected to water-seal without suction for at least rodesis appears to be critical.60 In a study on experimental
3 hours following the pleurodesis procedure, and then talc slurry pleurodesis, Kennedy and coworkers61 found
apply increasing suction gradually. prominent perivascular infiltrates with mononuclear
Considerations on pleurodesis for malignant pleural effusions 333
(a)
(b)
(c) (d)
Figure 25.15 Metastatic cancer of the colon. Chemotherapy was given, but the effusion could not be controlled. (a) Chest radiograph
before thoracentesis. (b) Iatrogenic pneumothorax occurred after therapeutic thoracentesis. (c) Necrotic tumor nodules and diffuse
lymphangitis were observed on the visceral pleura at thoracoscopy. One of them (top of the figure, dark umbilicated lesion) was
spontaneously ruptured, with no biopsy performed. (d) Air leak and persistence of pneumothorax was observed after unsuccessful talc
pleurodesis.
inflammation in the underlying lung, and they speculated ent, since no talc was found beyond the pleura in any
that some mediators might spread through the pulmonary case. It seems that the size of particles may play an
circulation. important role in the whole process,64 and a recent
There is some concern about the systemic absorption European multicenter study on safety of talc poudrage
of the sclerosing agents, and this is suspected to be the using large-size particle talc (with 25.6 μm median diam-
rule for almost all of the soluble agents that are instilled eter) found no cases of acute respiratory distress in a
into the pleural space. In contrast, talc is thought to series including 558 patients.65
persist in the pleura for a long time, thus accounting, at
least in part, for its better results in pleurodesis. POSSIBLE ACTIVATION OF THE SYSTEMIC COAGULATION
However, there are some disturbing reports on the
finding of talc particles in distant organs after talc pleu- One special aspect to be considered is the possible
rodesis, both in animals62 and humans.63 Our experience activation of the systemic coagulation following
with four autopsies in patients that had undergone tho- pleurodesis. Agrenius and coworkers66,67 reported an
racoscopic talc poudrage is, however, completely differ- increase in coagulation and inhibition of fibrinolytic
334 Effusions from malignancy
activity in the pleural space after instillation of quinacrine Indwelling pleural catheter
as a sclerosing agent. Since it is assumed that a fibrin mesh
formation is a necessary step for the fibrotic process, these In patients with an expected short survival (poor perform-
findings make sense in the context of the mechanisms ance status and usually presenting with very low pleural
which lead to pleural symphysis. We also demonstrated pH), placement of an indwelling pleural catheter con-
similar effects after talc pleurodesis in our patients,10 and nected to a vacuum bottle or a disposable bag can be an
were subsequently concerned about the possible systemic acceptable choice. Tremblay and Michaud72 have reported
implications of the pleural coagulation/fibrinolysis very good results in a series of 250 cases treated with a
imbalance that is involved in the pleurodesis process itself. tunneled pleural catheter for MPEs.
Prompted by this concern and by our finding of two cases
of massive pulmonary embolism after talc pleurodesis, we
performed a preliminary study on simultaneous Repeated thoracentesis
pleural/plasma determination of markers for coagulation
and fibrinolysis. We found that an activation of the In cases with very poor general condition, repeated thora-
systemic coagulation is frequently observed after talc centeses may be the only choice available. However, this
poudrage68 and that this side effect can be partially option should be kept as a last choice since discomfort,
controlled with prophylactic heparin.69 The relevance of risks of infection and protein depletion can significantly
this finding in clinical practice is still unclear, but some adversely affect the already poor quality of life of those
early deaths (less than 30 days) following pleurodesis patients.
procedures (up to 43 percent in the series of Seaton and
coworkers70) may in part be related to an undetected
pulmonary embolism, and not to advanced neoplastic
disease, as is commonly believed. PROSPECTS FOR CLINICAL STUDIES
14. Macha HN, Reichle G, von Zwehl D, et al. The role of ultrasound carbohydrate antigen 15-3, and cytokeratin 19 fragments) in
assisted thoracoscopy in the diagnosis of pleural disease. Clinical pleural fluid for the differential diagnosis of benign and malignant
experience in 687 cases. Eur J Cardiothorac Surg 1993; 7: 19–22. effusions. Chest 2004; 126: 1757–63.
15. Kaplan LM, Epstein SK, Schwartz SL, Cao QL, Pandian NG. Clinical, 35. Ceyhan BB, Demiralp E, Celikel T. Analysis of pleural effusions
echocardiographic, and hemodynamic evidence of cardiac using flow cytometry. Respiration 1996; 63: 17–24.
tamponade caused by large pleural effusions. Am J Respir Crit 36. Kayser K, Blum S, Beyer M, et al. Routine DNA cytometry of benign
Care Med 1995; 151: 904–8. and malignant pleural effusions by means of the remote
∗16. Antony VB, Loddenkemper R, Astoul P, et al. Management of quantitation server Euroquant: a prospective study. J Clin Pathol
malignant pleural effusions. Eur Respir J 2001; 18: 402–19. 2000; 53: 760–4.
17. Moltyaner Y, Miletin MS, Grossman RF. Transudative pleural ◆37. Rodríguez-Panadero F, Anthony VB. Pleurodesis: State of the art.
effusions. False reassurance against malignancy. Chest 2000; 118: Eur Respir J 1997; 10: 1648–54.
885. 38. Naito T, Satoh H, Ishikawa H, et al. Pleural effusion as a significant
18. Fernandez C, Martin C, Aranda I, Romero S. Malignant transient prognostic factor in non-small cell lung cancer. Anticancer Res
pleural transudate: a sign of early lymphatic tumoral obstruction. 1997; 17: 4743–6.
Respiration 2000; 67: 333–6. 39. Martín Díaz E, Arnau Obrer A, Martorell Cebollada M, Cantó
19. Assi Z, Caruso JL, Herndon J, Patz Jr EF. Cytologically proved Armengod A. La toracocentesis en la evaluación del cáncer de
malignant pleural effusions. Distribution of transudates and pulmón con derrame pleural. [Thoracocentesis for the assessment
exudates. Chest 1998; 113: 1302–4. of lung cancer with pleural effusion]. Arch Bronconeumol 2002;
20. Rodriguez-Panadero F, Lopez-Mejias J. Low glucose and pH levels 38: 479–84.
in malignant effusions; Diagnostic significance and prognostic 40. Rodríguez-Panadero F. Lung cancer and ipsilateral pleural
value in respect to pleurodesis. Am Rev Respir Dis 1989; 139: effusion. Ann Oncol 1995; 6 (Suppl 3): S25–7.
663–7. 41. Cantó A, Arnau A, Guijarro R, Centeno A, Marorell M. Actitud
21. Good JT, Taryle DA, Sahn SA. The pathogenesis of low glucose, quirúrgica en el carcinoma broncopulmonar que se acompaña de
low pH malignant effusions. Am Rev Respir Dis 1985; 131: un derrame pleural homolateral. Arch Bronconeumol 1992; 28:
737–41. 332–6.
22. Sahn SA., Good JT Jr. Pleural fluid pH in malignant effusions: 42. Roberts JR, Blum MG, Arildsen R, et al. Prospective comparison of
Diagnostic, Prognostic, and therapeutic implications. Ann Int Med radiologic, thoracoscopic, and pathologic staging in patients with
1988; 108: 345–9. early non-small cell lung cancer. Ann Thorac Surg 1999; 68:
23. Sanchez-Armengol A, Rodriguez-Panadero F. Survival and talc 1154–8.
pleurodesis in metastatic pleural carcinoma, revisited. Report on 43. Sawabata N, Matsumura A, Motohiro A, et al. Malignant minor
125 cases. Chest 1993; 104: 1482–5. pleural effusion detected on thoracotomy for patients with non-
24. Rodríguez-Panadero F, Del Rey Pérez JJ. Survival of malignant small cell lung cancer: Is tumor resection beneficial for prognosis?
pleural mesotheliomas as compared to metastatic carcinomas. Eur Ann Thorac Surg 2002; 73: 412–15.
Respir Rev 1993; 3: 208–10. 44. Cantó A, Arnau A, Galbis J, et al. [The so-called malignant pleural
25. Heffner JE, Nietert PJ, Barbieri C. Pleural fluid pH as a predictor of effusion: A new review of direct data obtained with diagnostic
survival for patients with malignant pleural effusions. Chest 2000; pleuroscopy]. Arch Bronconeumol 1996; 32: 453–8.
117: 79–86. 45. Usuda K, Saito Y, Endo C, et al. Cytologic assessment of
26. Antunes G, Neville E, Duffy J, Ali N. BTS guidelines for the peroperative pleural effusion and prognosis in lung cancer
management of malignant pleural effusions. Thorax 2003; 58 patients who underwent resection. Tohoku J Exp Med 1992; 167:
(Suppl 2): ii29–38. 219–30.
27. Escudero Bueno C, García Clemente M, Cuesta Castro B, et al. 46. Kondo H, Asamura H, Suemasu K, et al. Prognostic significance of
Cytologic and bacteriologic analysis of fluid and pleural biopsy pleural lavage cytology immediately after thoracotomy in patients
specimens with Cope’s needle. Study of 414 patients. Arch Intern with lung cancer. J Thorac Cardiovasc Surg 1993; 106: 1092–7.
Med 1990; 150: 1190–4. 47. Higashiyama M, Doi O, Kodama K, et al. Pleural lavage cytology
●28. Canto A, Rivas J, Saumench J, Morera R, Moya J. Points to consider immediately after thoracotomy and before closure of the thoracic
when choosing a biopsy method in cases of pleurisy of unknown cavity for lung cancer without pleural effusion and dissemination:
origin. Chest 1983; 84: 176–9. clinicopathologic and prognostic analysis. Ann Surg Oncol 1997;
29. Maskell NA, Gleeson FV, Davies RJ. Standard pleural biopsy versus 4: 409–15.
CT-guided cutting-needle biopsy for diagnosis of malignant 48. Dresler CM, Fratelli C, Babb J. Prognostic value of positive pleural
disease in pleural effusions: a randomised controlled trial. Lancet lavage in patients with lung cancer resection. Ann Thorac Surg
2003; 361: 1326–30. 1999; 67: 1435–9.
◆30. Loddenkemper R. Thoracoscopy: State of the art. Eur Respir J 49. Romero Romero B, Diaz-Cañaveral L, Laserna E, et al. The need for
1998; 11: 213–21. chemical pleurodesis in patients with malignant pleural effusion
31. Boutin C, Viallat JR, Cargnino P, Farisse P. Thoracoscopy in is associated to decline of pH and high levels of D-dimer in serial
malignant pleural effusions. Am Rev Respir Dis 1981; 124: pleural fluid samples. Am J Respir Crit Care Med 2001; 163:
588–92. A903.
32. Loddenkemper R, Grosser H, Gabler A, et al. Prospective evaluation 50. Dresler CM, Olak J, Herndon JE, et al. Phase III intergroup study of
of biopsy methods in the diagnosis of malignant pleural effusions: talc poudrage vs talc slurry sclerosis for malignant pleural
intrapatient comparison between pleural fluid cytology, blind effusion. Chest 2005; 127: 909–15.
needle biopsy and thoracoscopy. Am Rev Respir Dis 1983; 127 51. Yim AP, Chan AT, Lee TW, Wan IY, Ho JK. Thoracoscopic talc
(Suppl 4): 114. insufflation versus talc slurry for symptomatic malignant pleural
33. Villena V, Lopez Encuentra A, De Pablo A, et al. Diagnóstico effusion. Ann Thorac Surg 1996; 62: 1655–8.
ambulatorio de los pacientes que precisan biopsia pleural. Estudio 52. Mañes N, Rodriguez-Panadero F, Bravo JL, Hernandez H, Alix A.
de 100 casos consecutivos. [Ambulatory diagnosis of the patients Talc pleurodesis: Prospective and randomized study. Clinical
requiring a pleural biopsy. Study of 100 consecutive cases]. Arch follow-up. Chest 2000; 118 (Suppl): 131S.
Bronconeumol 1997; 33: 395–8. ◆53. Rodriguez-Panadero F, Jannsen JP, Astoul P. Thoracoscopy: general
34. Porcel JM, Vives M, Esquerda A, et al. Use of a panel of tumor overview and place in the diagnosis and management of pleural
markers (carcinoembryonic antigen, cancer antigen 125, effusion. Eur Respir J 2006; 28: 409–21.
References 337
54. Xie C, McGovern JP, Wu W, Wang NS, Light RW. Comparisons of ●65. Jannsen JP, Collier G, Astoul P, et al. Safety of talc poudrage in
pleurodesis induced by talc with or without thymol iodide in malignant pleural effusion. Lancet 2007; 369: 1535–9.
rabbits. Chest 1998; 113: 795–9. ●66. Agrenius V, Chmielewska J, Widström O., Blombäck M. Increased
55. Olivares-Torres CA, Laniado-Laborin R, Chaves-Garcia C, et al. coagulation activity of the pleura after tube drainage and
Iodopovidone pleurodesis for recurrent pleural effusions. Chest quinacrine instillation in malignant pleural effusion. Eur Respir J
2002; 122: 581–3. 1991; 4: 1135–9.
56. Chang YC, Patz EF Jr, Goodman PC. Pneumothorax after small- ●67. Agrenius V, Chmielewska J, Widström O, Blombäck M. Pleural
bore catheter placement for malignant pleural effusions. AJR Am J fibrinolytic activity is decreased in inflammation as demonstrated
Roentgenol 1996; 166: 1049–51. in quinacrine pleurodesis treatment of malignant pleural effusion.
57. Rinaldo JE, Owens GR, Rogers RM. Adult respiratory distress Am Rev Resp Dis 1989; 140: 1381–5.
syndrome following intrapleural instillation of talc. J Thorac 68. Rodriguez-Panadero F, Segado A, Torres I, et al. Thoracoscopy and
Cardiovasc Surg 1983; 85: 523–6. talc poudrage induce an activation of the systemic coagulation
58. Bouchama A, Chastre J, Gaudichet A, Soler P, Gibert C. Acute system. Am J Respir Crit Care Med 1995; 151: A357.
pneumonitis with bilateral pleural effusion after talc pleurodesis. 69. Rodriguez Panadero F, Segado A, Martin Juan J, et al. Activation of
Chest 1984; 86: 795–7. systemic coagulation in talc poudrage can be (partially) controlled
59. Rehse DH, Aye RW, Florence MG. Respiratory failure following talc with prophylactic heparin. Am J Respir Crit Care Med 1996; 153:
pleurodesis. Am J Surg 1999; 177: 437–40. A458.
●60. Maskell NA, Lee YC, Gleeson FV, et al. Randomized trials describing 70. Seaton KG, Patz EF Jr, Goodman PC. Palliative treatment of
lung inflammation after pleurodesis with talc of varying particle malignant pleural effusions: Value of small-bore catheters,
size. Am J Respir Crit Care Med 2004; 170: 377–82. thoracostomy and doxycycline sclerotherapy. Am J Roentgenol
61. Kennedy L, Harley RA, Sahn SA, Strange C. Talc slurry pleurodesis: 1995; 164: 589–91.
Pleural fluid and histologic analysis. Chest 1995; 107: 1707–12. 71. Petrou M, Kaplan D, Goldstraw P. The management of recurrent
62. Campos Werebe E, Pazetti R, Milanez de Campos JR, et al. malignant pleural effusions: The complementary role of talc
Systemic distribution of talc after intrapleural administration in pleurodesis and pleuro-peritoneal shunting. Cancer 1995; 75:
rats. Chest 1999; 115: 190–3. 801–5.
63. Milanez de Campos JR, Campos Werebe E, Vargas FS, et al. 72. Tremblay A, Michaud G. Single-center experience with 250
Respiratory failure due to insuflated talc. Lancet 1997; 349: tunnelled pleural catheter insertions for malignant pleural
251–2. effusion. Chest 2006; 129: 362–8.
●64. Ferrer J, Villarino MA, Tura JM, Traveria J, Light RW. Talc ●73. Mohanty SK, Dey P. Serous effusions: Diagnosis of malignancy
preparations used for pleurodesis vary markedly from one beyond cytomorphology. An analytic review. Postgrad Med J 2003;
preparation to another. Chest 2001; 119: 1901–5. 79: 569–74.
This page intentionally left blank
This page intentionally left blank
This page intentionally left blank
26
Effusions from infections: parapneumonic effusion
and empyema
The clinical importance of infection in the pleural space 341 Radiology 351
Historical perspective 342 Antibiotics 353
The epidemiology of pleural infection 342 Chest catheter drainage 354
The pathophysiology of pleural infection 343 Intrapleural fibrinolytics 354
Clinico-pathological stage correlations in pleural infection 345 Other fibrinolytics studies 357
Bacteriology 345 Future directions 361
The diagnosis and clinical assessment of pleural infection 348 Key points 361
Differential diagnosis 349 References 362
Predictors of clinical outcome in pleural infection 350
optimal methods for diagnosis and treatment where The implementation of these changes reduced the mortal-
possible. ity to 4.3 percent35 and they remain the core aims of phys-
ical and supportive therapy to this day.
The next major advance was the discovery of penicillin
that further decreased empyema mortality in the 1940s
HISTORICAL PERSPECTIVE and was also associated in a change in the microbiology of
the disease to produce the modern bacterial spectrum
The first recorded description of empyema is attributed described below.
to the Egyptian physician Imhotep in around 3000BC,25 Intrapleural fibrinolytic therapy was tried by Tillett and
in which he described ‘An abscess with prominent head Sherry36 in the 1940s but did not become standard practice
from the breast’, perhaps suggestive of empyema compli- because of frequent antigenic side effects from this impure
cated by chest wall invasion (empyema necessitans). product.
However, the first reliable description of pleural Surgical techniques improved beyond open drainage in
empyema and its treatment is attributed to Hippocrates the late nineteenth century, when thoracoplasty was
in about 500BC, who treated cases of pleural infection described by Estlander (1897) and Schede (1890).37,38 This
with open thoracic drainage.26 This remained the primary eliminated the empyema cavity but required major surgery
treatment for this disorder until the First World War, and was disfiguring. Decortication of the pleura was first
though Hewitt and Bulau27,28 described methods for described by Fowler and Beck at the end of the nineteenth
closed pleural drainage using a chest tube and an under- century39,40 and most recently video-assisted thoraco-
water seal decades before (1876–1891). Napoleon’s scopic surgery (VATS) has been introduced to allow
surgeon, Guillaume Dupuytren (1777–1835), after whom pleural debridement without formal thoracotomy in some
digital contractures associated with liver disease are patients.41
named, succumbed to empyema after announcing that
we would ‘rather die at the hands of God than of sur-
geons’.29,30 The famous Italian tenor Enrico Caruso THE EPIDEMIOLOGY OF PLEURAL INFECTION
(1873–1921) is likely to have died from empyema. One of
the most graphic historical accounts of pleural empyema Pleural infection affects patients of all ages but is more
was written by the Oxford physician, Sir William Osler, common in the elderly21 and in childhood.42 Men are
(after whom our Unit is named) who described his own, affected twice as often as women.43,44 Its incidence is
ultimately terminal, illness in 1919, after refusing thoracic higher in those with diabetes, alcoholism and substance
surgery for his condition.31 abuse, rheumatoid arthritis and coincidental chronic lung
The influenza pandemic of 1919, during the First disease.4,20,43,45 Diabetes is present in 10 to 23 percent of
World War, led to clusters of empyema cases among the patients, five times the background prevalence of this
fit young army recruits living in army camps. The treat- disease4,19,20,43 and excess alcohol intake is present in 6 to
ment of choice was open surgical drainage resulting in a 10 percent of patients.19,20,43,46 Poor dentition and risk
mortality as high as 70 percent in some outbreaks,32,33 factors for aspiration (seizures, gastro-esophageal reflux,
probably attributable to respiratory failure induced by mental retardation, alcohol use and sedative drug use) are
the large open pneumothorax produced by the thoraco- believed to be associated with an increased prevalence of
tomy.32 Since Streptococcus haemolyticus infection was anaerobic infection.19
responsible for many of these cases, it is interesting to The markedly different bacteriological patterns
consider that the streptokinase produced by this organ- observed in pneumonia and empyema (see below) sug-
ism may have prevented fibrinous adhesions forming in gests that they are microbiologically distinct diseases.
the pleural cavity, exacerbating the problem by facilitat- Other causes include surgery, trauma and iatrogenic
ing complete lung collapse.34,35 The US Army Empyema insults,20,43,47–49 including thoracentesis for pneumothorax
Commission was formed to address this problem and or pleural effusion (Table 26.1). Up to one-third of cases
this proved to be a landmark event in the care of this occur without any as yet identified risk factors.19 Where
disorder.34 The Commission noted that dogs with the pleural infection is acquired in hospital, the prognosis
empyema died more often if treated with early open is worse, and recent evidence suggests that the microbio-
drainage rather than delayed intervention, and it advo- logical pattern of infection influences mortality independ-
cated the closed tube drainage techniques described by ent of this (see below).50 The duration of recovery is also
Hewitt and Bulau.27,28 Their summary recommendations prolonged in these patients with the median hospital stay
were: being 58 days for hospital-acquired empyema in one rep-
resentative series,51 compared with 12–13 days for the
1. Adequate pus drainage with a closed chest tube. entire group in randomised trials.20 Hospital-acquired
2. Avoidance of early open drainage. infections also exhibit a microbiology that is strikingly
3. Obliteration of the pleural space. different from that of community-acquired empyema
4. Proper nutritional support. (Figure 26.2).43,50 The clinical picture of hospital-acquired
The pathophysiology of pleural infection 343
Table 26.1 Frequency and cause of 701 patients with pleural THE PATHOPHYSIOLOGY OF PLEURAL
infection19,42,46–48 INFECTION
Causes of pleural space infection Frequency (%)
This can be divided into three stages as discussed below.
Parapneumonic effusion 70
Post-bacterial pneumonia
Development of the initial pleural effusion:
Hospital-acquired pneumonia
‘the exudative phase’
Primary empyema 4
Postoperative 12 Animal work suggests that a pleural effusion may be a nec-
Traumatic 3 essary substrate for sustained pleural infection. In animal
Blunt trauma models of empyema, direct inoculation of bacteria into the
Penetrating trauma pleural cavity in the absence of pleural effusion results
Iatrogenic 4 either in overwhelming sepsis or spontaneous recovery –
e.g. post chest tube insertion and not sustained bacterial growth as characterizes clinical
Abdominal infection 2 empyema.52,53 However, experiments in rabbits suggest
e.g. subphrenic abscess that empyema (in rabbits) may be induced by direct inoc-
Miscellaneous 5 ulation of bacteria without an initial pleural effusion if the
Esophageal perforation bacteria are in a thicker broth.54,55 The development of the
Bacteremia initial effusion is due to increased permeability of the
Rupture of lung abscess into pleural space pleural membranes, in response to inflammation in the
intravenous drug abuse (contaminated needles) underlying lung parenchyma, which is thought to result in
transfer of interstitial fluid across the visceral pleura.
Indirect clinical evidence suggests that pleural inflamma-
tion alone is insufficient to result in significant pleural
empyema is sufficiently different from that of community- fluid, as many patients complain of pleuritic chest pain
acquired disease that it should probably be considered a during a pneumonic illness, and yet only a minority
different entity from both the epidemiological and thera- develop radiologically detectable pleural fluid.
peutic standpoints. Given the correct circumstances, fluid moves into the
pleural space due to locally increased capillary vascular
permeability and the activation of immune processes such
as neutrophil migration. Pro-inflammatory cytokines
Other streptococci 7%
including interleukin (IL)-6, IL-8 and tumor necrosis
Strep milleri Strep pneumoniae 13%
factor alpha (TNF-a) produce changes in the anatomical
group 32% shape of pleural mesothelial cells creating intercellular
‘gaps’ which further enhance permeability and additional
Staphylococci 11% fluid accumulates.56–58 The accumulating pleural fluid has
a normal glucose level (>40 mg/dL) and pH (>7.20), with
no detectable bacteria, and hence no microbiological or
Other 8% biochemical evidence of bacterial invasion. The effusion
Proteus 3% Anaerobes 16% will usually resolve spontaneously with antibiotic therapy
Enterobacteriacea Haemophilus for the underlying pneumonia.
7% influenza 3%
causes.59,60 This leads to fibrin deposition over the visceral and transforming growth factor beta (TGF-β).54,56,59,65 This
and parietal pleura, with the division of the pleural space forms an inelastic peel on both pleural surfaces with dense
by fibrinous septae, producing fluid loculation and pleural fibrous septations across the pleural cavity. As this solid
adhesions. While effusions of any cause may become locu- fibrous peel replaces the soft fibrin, lung re-expansion is
lated, the depression of the fibrinolytic system (elevated prevented, impairing lung function. Interesting recent evi-
PAI level, depressed tissue plasminogen activator [tPA]) dence points to a potential therapeutic target in the
has only been observed in pleural infection, and not in mediators thought to drive this process; the administration
effusions secondary to malignancy or transudates.59,60 This of anti-TGF-β antibodies during pleural infection results in
division of the pleural cavity provides the skeleton for the significantly less pleural thickening, in a well established
later invasion of fibrous tissue and the ‘natural’ process of animal model of empyema.55 The establishment of the
scarring and healing. However, it also creates multiple sep- phase of scar tissue formation is another important clinical
arate pockets of infected pleural fluid, and hence impairs landmark. It marks the point where pus drainage with a
the drainage of pus and is a major factor in reducing chest chest tube, even if supplemented by fibrinolytics that may
tube efficacy. Bacterial metabolism and neutrophil phago- lyse fibrin but not collagenous fibrous tissue, is likely to fail.
cytic activity induced by bacterial cell wall-derived frag- Interestingly there is marked inter-individual variation and
ments and proteases lead to increased lactic acid sometimes a prolonged delay in the rapidity with which this
production61 and thus a fall in pleural fluid pH and glucose happens. Approximately 50 percent of patients do not have
– the biochemical hallmarks of early transition to the collagenous fibrous pleural scarring even 3 weeks after the
infected state.62 As the numbers of neutrophils in the onset of pleural infection,66 implying that tube drainage
pleural space rise, lactate dehydrogenase is released pro- and thoracoscopic surgery may still be effective in patients
ducing the high levels typically seen in infected pleural with a long history at presentation. The subsequent clinical
fluids,24,63 and ultimately the pleural fluid becomes frankly course after the organizing stage is entered is also variable,
purulent, secondary to bacterial and inflammatory cell with some patients undergoing spontaneous resolution of
death and lysis. pleural thickening at 12 weeks67 while others develop
chronic sepsis and lung function deficits.65
In summary, pleural infection is a progressive process
Natural healing: ‘the organizing stage’ in which a self-resolving parapneumonic pleural effusion
can progress to a complicated, multi-septated fibrotic col-
Finally, there is the proliferation of fibroblasts and the lection that is only amenable to surgery. This evolution
evolution of pleural scarring, with animal model data sug- does not occur in a linear fashion but is summarized in
gesting this process is driven by mediators such as Table 26.2 using the classification of Light and col-
platelet-derived growth factor-like growth factor (PDGF)64 leagues.68
Parapneumonic effusion
Pleural pH
7.20 7.20
complicated simple
parapneumonic parapneumonic
effusion effusion
Despite this variation, there emerges a consistent micro- percent of entire group, = 62 percent of microbiology-
biological pattern which is distinct between community- positive samples) were due to a single aerobic organism,
and hospital-acquired empyema, and is furthermore 9 percent due to a single anaerobic organism and 13
distinct from the microbiology of pneumonia. percent due to polymicrobial infection. A distinct micro-
Detailed microbiological data were recently reported50 biological pattern was seen in community- and hospital-
from the largest randomized trial in pleural infection acquired disease.50
(the MIST1 study20). This represents the largest single Combining all previous data over the past 10
cohort of well-defined microbiology in pleural infection. years4,50,70–116 (Table 26.4), a microbiological diagnosis was
In this study, a microbiological diagnosis was achieved reached in 1253 out of 2501 cases (50.1 percent) with
using standard methods in 58 percent (mirroring previ- aerobic organisms the most frequent organisms identified
ous data), with a further 16 percent achieving diagnosis from infected pleural fluid. These are most commonly
using nucleic acid amplification techniques (bacterial Gram-positive organisms from streptococcal species fol-
DNA polymerase chain reaction).50 The use of DNA lowed by Staphylococcus aureus. In the modern era, around
amplification reduced the number of cases remaining 25 percent of culture positive cases are caused by
microbiologically undiagnosed from 42 percent with Streptococcus pneumoniae whereas it accounted for 60
standard techniques to 26 percent.50 While blood cul- percent of cases prior to the advent of antibiotics.50,117
tures were positive in only 12 percent of cases, they were However, many currently culture negative cases may be
often the only positive microbiological test in these due to this bacterium in patients who may have been
patients.50 The majority of culture positive cases (35 administered antibiotics prior to sampling.
Bacteriology 347
Table 26.4 A summary of the bacteriology of all positive Peptostreptococcus, Fusobacterium, Bacteroides and
cultures from clinical studies between 1996 and 2006 published Prevotella species. The positive diagnostic rate from
in the English language where details of bacteriology were pleural fluid culture may be significantly increased if
presented4,50,70–73,75–78,80–86,89,91,92,94–96,100,103–107,112,114–116 aerobic and anaerobic ‘blood culture’ bottles are inocu-
lated with the fresh pleural fluid sample and are sent to
Organism Number % of total
complement the standard ‘sterile vial’ sample.119,120
Microbiology from the MIST1 trial cohort showed very
Aerobes – Gram positive 948 63 substantial differences between community-acquired
Streptococci 680 45 empyema and hospital-acquired infection, and probably
Streptococcus milleri group 169 11.2 represents the best base for empirical antibiotic choices.
Streptococcus pneumoniae 346 22.9 Within this cohort, community-acquired disease (whether
Streptococcus pyogenes 36 2.4 arising following an evident community-acquired pneu-
Other streptococcus species 117 7.8 monia or as a ‘primary empyema’ without obvious under-
Staphylococci 248 16 lying parenchymal lung infection) was caused by
Staphylococcus aureus 197 13.1 streptococcal disease in over 50 percent of microbiology
Methicillin resistant S. aureus 40 2.7 positive cases (largely Streptococcus pneumoniae and
Staphylococcus epidermidis 11 0.7 Streptococcus intermedius50). Staphylococci, anaerobes and
Enterococcus spp. 20 1 Gram-negative organisms accounted for the majority of
Aerobes – Gram negatives 228 15 the remainder of cases. This translates to around 50
Escherichia coli 52 3.4 percent of microbiologically positive cases associated with
Klebsiella 64 4.2 penicillin-resistant organisms,50 and 30 percent associated
Other coliforms 37 2.4 with anaerobic infection (often as co-infection with
Proteus 15 1 aerobes), which has clear implications for empirical antibi-
Enterobacter spp. 18 1.2 otic choices (see later).
Pseudomonas aeruginosa 42 2.8 In contrast, hospital-acquired empyema, which occurs
Anaerobes 206 14 after hospital acquired pneumonia, surgery or iatrogenic
Fusobacterium 51 3.4 causes, was dominated by Staphylococcus, Gram-negative
Bacteroides 51 3.4 organisms (e.g. Enterobactericae, coliforms), Enterococcus
Peptostreptococcus 50 3.3 species and a high proportion of methicillin-resistant
Mixed anaerobes/unclassified 21 1.4 Staphylococcus aureus (MRSA = 28 percent) (Figure
Prevotella 31 2.1 26.2).50 This observation fits with previous surgical series
Clostridium 2 0.1 where 14 percent of the culture positive cases were attrib-
Mycobacterium tuberculosis 22 1.4 utable to MRSA.47 Streptococcal disease was almost absent
Actinomyces spp. 7 0.5 as a cause of hospital-acquired empyema, resulting in bac-
Other 96 6 teria resistant to standard antibiotic therapies for pneumo-
Total 1508 100 nia in the majority of cases of hospital-acquired disease.50
The so called ‘atypical’ organisms (e.g. Mycoplasma,
Legionella) were not detected as a cause of pleural infec-
tion.50
Gram-negative aerobic organisms are also common Previous data from lung abscess treatment has suggested
causes of pleural infection. The most frequent of these are that ‘mixed’ infections are associated with a poorer prog-
Haemophilus influenzae, Escherichia coli, Pseudomonas spp. nosis,121,122 although this does not appear to be true for
and Klebsiella pneumoniae (Table 26.4). pleural infection.50 Mortality is higher in hospital- than in
Anaerobic pleural infection is more likely to have an community-acquired disease.50 Streptococcal disease has a
insidious clinical onset, with less fever and greater weight 1-year mortality of 17 percent.50 The survival with culture-
loss. The delayed clinical diagnosis of anaerobic negative disease intriguingly mirrors that of streptococcal
empyema is therefore frequent, and a high clinical index disease,50 perhaps lending weight to the theory that culture-
of suspicion should be maintained if life-threatening and negative disease is frequently a result of antibiotic-treated
entirely treatable disease is not to be overlooked. streptococcal disease. In contrast to this, mortality in
Anaerobic infection is commonly associated with poor patients with Gram-negative disease, staphylococcal and
dental hygiene and aspiration pneumonia and often has a mixed aerobic disease is much higher at 45 percent at 1
mixed bacterial flora. The frequency of anaerobic organ- year.50 This difference in prognosis seen between bacterial
isms isolated from published series of pleural infection subgroups may be a potential area for targeting more
ranges from 14 to 32 percent of positive cul- aggressive therapy in the future (Figure 26.4).
tures,4,50,70–78,80–82,84–86,89,91,92,94–96,101,103–107,112,114–118 with Gram-negative empyema is more frequent in patients
an overall combined rate of 14 percent (Table 26.4). The with underlying diseases, especially those with diabetes or
anaerobic pathogens most commonly identified are alcoholism.4 Some bacterial pneumonias more commonly
348 Effusions from infections: parapneumonic effusion and empyema
1.0
0.9
0.8 pH = 7.2
0.7
0.5
0.4
0.3
0.2
0.1
0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
(a) False positive fraction
Doppler of Small rim of Figure 26.6 Receiver operating characteristic (ROC) curve for
blood vessels pleural fluid the diagnostic accuracy of pleural fluid pH for pleural infection.130
within lung
RADIOLOGY
Table 26.5 Differentiating between a lung abscess and empyema on contrast enhanced thoracic computed tomography150,152,153
(a) (b)
Figure 26.11 Computed tomography scan (a) and magnetic resonance imaging (MRI) scan (b), T2 weighted images of septated pleural
effusion. The septations are clearly demonstrated on the MRI image. Image courtesy of Dr Fergus Gleeson, Oxford.
Antibiotics 353
ening (cold on PET scanning159,160), but is unable to reli- empyema is substantial in the hospital-acquired group.50
ably separate infection and malignancy in the context of Although the majority of hospital-acquired infections are
pleural effusion. treated intravenously for the duration of treatment, oral
regimens exist which cover the majority of likely
pathogens. Exact regimens will be influenced by local prac-
ANTIBIOTICS tice and bacterial resistance patterns, but suitable illustra-
tive regimes are shown in Table 26.6. Oral regimens are
All patients with parapneumonic effusions or empyema not recommended for the treatment of hospital-acquired
should receive antibiotics from the time of diagnosis. infection, but may be required for the long-term treatment
Where possible, the antibiotics prescribed should be of patients with hospital-acquired disease who are unfit for
guided by bacterial culture results. However, these results invasive procedures.
are usually not immediately available and approximately Penicillins, cephalosporins, clindamycin, carbapenems
40 percent of cases will be persistently culture-nega- and metronidazole all show good and similar penetration
tive.4,50,70–116 The patient’s clinical setting and the underly- of the pleural space.161–163 However, aminoglycosides
ing cause of the empyema should therefore dictate the appear to penetrate poorly into the pleural space and may
initial choice of empirical antibiotic therapy, which will be be inactive in purulent, acidic pleural fluid. They should
required for the duration of therapy in 40 percent of therefore be avoided in treating pleural infection.161,162 In
cases. all cases, positive pleural fluid or blood cultures should be
used to narrow antibiotic therapy tailored to the organ-
ism(s) found. However, mixed infections occur in up to 13
Community-acquired pleural infection percent of cases, especially relevant to anaerobic bacteria,
and these organisms may be difficult to isolate.
In order to provide cover for penicillin-resistant organ- Consideration should therefore be given to continuing
isms and anaerobes described above, a combination of a concurrent anaerobic antibiotics, even when a positive
second generation cephalosporin or aminopenicillin plus a microbiological diagnosis has been made.
β-lactamase inhibitor plus anaerobic cover will provide The duration of antibiotic treatment depends on the
adequate empirical cover for the duration of therapy. bacteriology, the efficacy of pleural drainage and the speed
Given the low prevalence of Legionella and Mycoplasma in of resolution of the patient’s symptoms. There are no
pleural infection,50 in contrast to community-acquired studies directly addressing the length of treatment
pneumonia, addition of a macrolide is unnecessary. required in pleural infection, although at least 2 weeks
therapy is not uncommon (following paradigms for the
treatment of pulmonary abscess). It is our practise to give
Hospital-acquired pleural infection at least the first week of antibiotic therapy intravenously,
with a subsequent change to oral regimens dictated by the
A combination of a broad-spectrum antibiotic (such as a patient’s response. A combination of an oral aminopeni-
carbapenems or the anti-pseudomonal penicillins) and an cillin with a beta lactamase inhibitor (or clindamycin and
agent active against MRSA is required for the hospital- ciprofloxacin in penicillin-allergic patients) is often ade-
acquired subgroup, to provide cover for multiresistant quate as oral sustained therapy. Decisions on the length of
organisms and anaerobes.50 Postoperative, trauma and treatment can be guided by repeated measurements of
iatrogenic related empyema will also require anti- serum markers of the acute-phase reaction, such as the
staphylococcal coverage as the incidence of MRSA C-reactive protein. The monitoring of fever lysis is also
Hospital-acquireda Clindamycin 300 mg qds + ciprofloxacin 500 mg bd Meropenem 500 mg tds + vancomycin 1 g bd
(+ rifampicin 300 mg bd if MRSA is suspected)
Community-acquired Co-Amoxiclav 625 mg tds + metronidazole 400 mg tds Cefuroxime 1.5 g tds (or co-amoxiclav 1.2 g tds) +
OR metronidazole 500 mg tds (meropenem 500 mg
clindamycin 300 mg qds + ciprofloxacin 500 mg bd in penicillin-allergic patients)
aOral
regimens are not recommended for hospital-acquired infection, but may be required for long-term treatment (see text).
MRSA, methicillin-resistant Staphylococcus aureus.
354 Effusions from infections: parapneumonic effusion and empyema
valuable, though elderly patients and those with indolent, small studies have directly compared small- and large-bore
often anaerobic empyema, frequently fail to mount a fever catheters for the drainage of intra-abdominal pus, both of
and here indices such as the C-reactive protein seem par- which reported no advantage of drains larger than 8
ticularly helpful. F.175,176 Assessing clinical studies over the last 18 years of
treatment of pleural infection in which different catheter
sizes were used, no difference between outcomes with dif-
ferent catheter size are reported.75,77,78,86,171,177 The use of
CHEST CATHETER DRAINAGE regular saline flushes and suction may prevent small
catheter blockage and aid the drainage of pus.
The optimal size of catheter for the drainage of an infected
pleural space remains a cause of vigorous debate. Previous
management guidelines have suggested that larger bore INTRAPLEURAL FIBRINOLYTICS
catheters are required to allow drainage of high viscosity
empyema fluid, citing evidence that inadequate tube bore Background
is a predictor of need for surgery.21,23,164,165 However, flex-
ible smaller bore catheters are less traumatic to insert and The use of intrapleural fibrinolytic agents to chemically
more comfortable for the patient after insertion, extrapo- disrupt the fibrinous pleural septations of empyema
lating data from catheter treatment of malignant pleural (Figure 26.12) has been used to aid the drainage of infected
effusion.166 Numerous observational series, including pleural fluids for over 50 years. It was first described in
hundreds of patients, show that good outcomes can be 1949 by Tillet and Sherry,36 who used partially purified
achieved with smaller catheters (<12 F).127,128,167–173 streptococcal fibrinolysin that contained both streptoki-
Given the extent and importance of the ‘chest tube size’ nase and strepdornase (a DNAse) to drain infected post-
debate, consideration of the background to this argument operative haemothoraces. This was associated with
is appropriate. There are three reasons why infected immunological side effects and the use of this agent failed
pleural fluid may resist drainage through a catheter: to become routine practice. Subsequently, the availability
drainage may fail if the fluid is of high viscosity and of highly purified streptokinase and non-antigenic uroki-
directly blocks the tube; it may fail because the balance of nase was, in part, responsible for renewed interest in this
forces drawing it down the tube is inadequate; and it may therapy.
fail if the fluid is partitioned by fibrinous septae. Although Streptokinase is a proteolytic enzyme derived from a
a large-bore tube may improve the first problem, smaller bacterial protein of group C beta-haemolytic streptococci.
bore catheters can be kept clear by flushing and a large- It forms a complex with plasminogen that then converts
bore tube is unlikely to decrease the other difficulties.
During the chest tube drainage of non-septated pleural
contents, provided that the sucker does not block, the neg-
ative tube tip pressure is transmitted through the pleural
fluid to the lung surface. Therefore, whether fluid flow
occurs at all is related to the balance between the negative
suction pressure and the compliance of the underlying
lung and not to catheter bore. In contrast, the rate at which
fluid drains is clearly dependent on tube calibre, as well as
fluid viscosity. Therefore, provided that the tube does not
block, the rapidity of chest tube drainage might be
improved by increasing the drain size, but the likelihood of
eventual successful drainage is unchanged. In the presence
of a multi-septated effusion the advantages of a large drain
seem even smaller. Here, as some fluid drainage occurs,
the fibrinous septae distort, distributing the drainage
forces across the many locules as well as the lung surface.
It is the pressure gradients developed across the walls of
these locules that predict whether they will rupture and
drain, not the rapidity of flow down the drainage catheter.
Here again, provided that the catheter is patent, its bore
would seem to be irrelevant.
There is some indirect evidence in support of the above
arguments. In vitro studies suggest that increased tube Figure 26.12 The macroscopic appearance of pleural fibrinous
bore is associated with increased fluid flow, but the septation at thoracoscopy in an infected malignant pleural
increase in flow is small with catheters above 7 F.174 Two effusion.
Intrapleural fibrinolytics 355
additional circulating plasminogen to plasmin. Plasmin the streptokinase arm compared with placebo.20 Assessing
lyses fresh fibrin clots and digests prothrombin and fib- whether it does induce a systemic antibody response is
rinogen.178 Because it is derived from a bacterial source it complicated by the fact that many pleural infections are
is antigenic,179 unlike urokinase. themselves streptococcal and hence are likely to induce an
Streptokinase is usually administered as a solution of antibody response as a direct consequence of the bacterial
250 000 IU in 30 mL of sterile saline via the chest tube and infection. Currently, cautious best practice would seem to
the tube clamped for 2–4 hours before returning to normal indicate that subjects who have had intrapleural streptoki-
drainage. This has been given daily71,77,83,85,95 or repeated nase should be managed as if they had received intra-
twice daily for several days.20 Urokinase can be used in a venous streptokinase. These patients should be given a
similar fashion with 100 000 IU as the standard daily streptokinase exposure card and an alternative fibrinolytic
dose.78 agent used if needed.
These agents theoretically improve pleural pus drainage
through the disruption of fibrinous septae and the dissolu-
tion of the visceral fibrinous pleural rind. Lysis of this rind Randomized trials using fibrinolytics
may then aid lung expansion and so help to obliterate the
infected pleural space. Uncontrolled observational case To date, 14 randomized trials have been published using
series2,10,16,17,51,89,180,181 report good outcomes for patients intrapleural fibrinolytics: 10 in adults20,71,76–78,85,86,101,111,177
treated with these drugs. and four in children92,107,108,116 (Table 26.7). Eight of these
studies are ‘efficacy’ studies assessing fibrinolytic versus
placebo 20,78,85,86,107,111,116,177 (two pediatric107,116), one
The safety of intrapleural fibrinolytic agents assesses chest drainage plus fibrinolytic versus chest
drainage only101 (i.e. no placebo) and one is a comparison
Potential side effects of fibrinolytic therapy include hem- of two different fibrinolytic agents.77. The remaining three
orrhage, pleuritic pain after drug administration and studies assess the early use of VATs versus a drainage plus
fever.77,89,181 There have been a few case reports of both fibrinolytic strategy76,92,108 (two pediatric92,108).
systemic and local hemorrhage after intrapleural The randomized trials of fibrinolytic therapy until 2004
fibrinolytic administration,182–184 but the majority of the all assessed clinical surrogates for treatment success,
published studies report no bleeding complica- usually using volume of fluid drained, radiographic
tions.2,3,10,16,17,83,86,92,101,107,108,111,116,181 Within the MIST1 outcome or length of hospital stay as outcome measures
cohort, there was a serious adverse event rate of 3 percent (Table 26.7). A selected few assessed the need for surgical
in the placebo arm and 7 percent of the streptokinase arm, drainage as a primary outcome measure, but the criteria
largely due to immunological reactions to the streptoki- for surgical drainage were not clear and may simply be a
nase and with no excess of bleeding complications.20 further surrogate for poor radiographic outcomes. The
Studies of whether fibrinolytic drugs given into the majority of these studies found significant benefit for the
pleural space induce systemic activation of fibrinolytic use of intrapleural fibrinolytic therapy. These studies were
mechanisms are also reassuring. Systemic fibrinolytic acti- all small, the largest number of cases in a single study being
vation is best quantified from the thrombin time, fibrino- 60, and the total number of patients in these studies
gen levels and the presence of fibrinogen degradation amounting to 272 (100 of these being children). A meta-
products. Two studies have assessed whether these change analysis conducted in 2004 assessing the publishes studies
after the administration of intrapleural streptokinase. at the time concluded that while there seemed to be evi-
They studied a total of 26 patients with thrombin time, fib- dence of benefit from intrapleural thrombolytic therapy
rinogen levels and fibrinogen degradation products being (decreased hospital stay, improved radiographic appear-
quantified before and after the administration of up to 1.5 ance, lower surgical rate), there was insufficient evidence
million units of intrapleural streptokinase given in to recommend its use routinely.188
repeated doses. Neither of these studies revealed any The first randomised trial to demonstrate improvement
detectable changes in the coagulation indices when com- in clinically meaningful outcomes with intrapleural fibri-
pared with baseline.185,186 nolytic, such as clinical success rate and need for surgery,
A further important safety issue relates to the systemic was conducted by Diacon et al.86 in 2004. A total of 44
antigenic effects of intrapleural streptokinase and hence patients were randomized to intrapleural streptokinase or
the necessity for the use of a different fibrinolytic drug for intrapleural saline in addition to pleural rinses through the
later clinical indications such as myocardial infarction. chest drain, and decisions about further management,
Anti-streptokinase immunoglobulin (Ig)G detectable in including surgery, were made in a blinded and consistent
blood after prior exposure to streptokinase or streptococ- manner. After 3 days of intrapleural therapy, there was no
cal infection is associated with reduced fibrinolytic efficacy difference between the active and placebo groups, whereas
and hence reduced efficacy in achieving coronary reperfu- after 7 days, there was a significantly increased clinical
sion.187 Within the MIST1 study, at 3 months there was a success rate and a significantly lower surgical rate in the
highly significant rise in anti-streptococcal antibodies in streptokinase group.86 However, this study was not
Table 26.7 Randomized controlled trials of fibrinolytic therapy in adult empyema
Bouros et al.77 SK versus UK 50 250 000 IU SK Randomized Similar fluid drainage both groups
25 SK versus 25 UK 100 000 IU UK Double blind Two patients in each group required surgery
repeated daily if
required Transient fever 7/25 SK, versus 0/25 UK
Davies et al.85 SK versus saline 24 250 000 IU SK daily Randomized Significantly greater volume of pleural fluid drainage in
for 3 days SK group (p < 0.001) and radiographic improvement (p < 0.05)
12 SK versus 12 saline Double blind Higher surgical referral rate in control group (0/13 versus 3/12,
non-significant)
Bouros et al.78 UK versus saline 31 100 000 IU SK daily Randomized Faster mean time to defervescence and shorter hospital stay in
for 3 days UK group (p < 0.01)
15 UK versus 16 saline Double blind Significantly greater radiographic improvement in UK group
Tuncozgur et al.177 UK versus saline 49 100 000 IU UK for Randomized Shorter time to defervesence (p < 0.01), shorter hospital stay
5 days (p < 0.001) in UK group, volume drained greater with UK (p <
0.001). Lower decortication rate (60% versus 29%, p < 0.001)
24 UK versus 25 saline Blinding unclear
Decision to surgery included radiological assessment
High rate of failure in saline group
Talib et al.111 SK versus saline 24 250 000 IU SK for Randomized Increased drainage, shorter duration of drainage in SK group.
6 days Significant difference in radiographic improvement (p < 0.001)
12 SK versus 12 saline Non-blinded
Diacon et al.86 SK versus saline 44 250 000 IU SK daily Randomized One mortality in each group. Longer duration of SK treatment
and continued as (4.5 versus 3 days). No difference in outcome after 3 days, but
needed after 7 days, higher clinical success rate (p < 0.01) and fewer
surgical referrals (p < 0.02)
22 SK versus 22 saline Placebo controlled
High rate of withdrawal
Wide inclusion criteria
Surgical referral based on radiology
Maskell et al.20 SK versus saline 430 250 000 IU SK bd for Randomized No difference in mortality, need for surgery or combined endpoint.
3 days No difference in hospital stay, radiographic outcome or lung
function at 3 months. No difference between groups in subgroups
(purulent, short clinical history)
208 SK versus 222 saline Double blind Surgical referral based on clinical opinion (not radiographic alone)
Misthos et al.101 Intercostal drain 127 250 000 IU SK od for Randomized Higher treatment success in SK group (67% versus 87%, p < 0.05)
versus intercostal 3 days (tube clamped
drain + SK for 4 hours)
70 intercostal drain Non-controlled, Shorter hospital stay and lower mortality in SK group (p < 0.001)
versus 57 intercostal non-blinded
drain + SK
SK, streptokinase; UK, urokinase.
Other fibrinolytics studies 357
powered to address surgical outcomes and there was a high tion unblinded, which made the differences in the surgical
rate of subject withdrawal (Table 26.7). and hospital stay outcomes questionable.
Against this background, the MIST1 trial20 reported its A meta-analysis performed in 2006189 and including the
findings in 2005. The study recruited 454 patients across five methodologically sound trials (575 patients in total)
52 UK hospitals, and was powered to address whether concluded that the routine use of fibrinolytic therapy for
streptokinase altered combined death and surgical rate. all patients was not supported by the current evidence,
Patients were recruited on the basis of established criteria although it suggested that some patients may benefit from
for pleural infection (appropriate clinical scenario and one treatment given the trial heterogeneity.189
or more of: pleural fluid pH < 7.2, purulent pleural fluid, Does the MIST1 trial therefore end 50 years of clinical
microbiologically positive fluid) and randomized to investigation into intrapleural thrombolytics? There is
intrapleural streptokinase (250000 IU bd for 3 days) or direct evidence that fibrinous septations are divided by
intrapleural saline (bd for 3 days) in addition to standard intrapleural streptokinase in malignant pleural effusion.190
local care (chest drain, antibiotics).20 Intrapleural agents There are reasons to believe that fibrinolytic agents other
were given to 430 patients and primary outcome data was than streptokinase may be more efficacious in breaking
available in 99 percent. No difference was found in the down adhesions and loculations intrapleurally. The mode
combined rate of death and surgery at 3 months (primary of action of streptokinase relies upon a minimum level of
outcome measure) in the placebo and streptokinase (SK) plasminogen in order to produce fibrinolytic plasmin, and
groups, the outcome occurring in around 30 percent of there is evidence that the intrapleural concentration of
patients (Figure 26.13). There was no difference between plasminogen is low. Agents such as direct plasminogen
SK and placebo groups in death or surgery analyzed sepa- activators are not limited by the endogenous plasminogen
rately, length of hospital stay, radiographic change or lung level and therefore may be more successful in dividing sep-
function at 3 months. Subgroup analyses demonstrated no tations within an infected pleural space. There are several
difference between streptokinase and placebo regardless of case reports and series in both adults191,192 and chil-
characteristic studied (pleural fluid purulence, compli- dren193–198 reporting success using intrapleural tPA, with
cated parapneumonic effusion, short clinical history, an apparently good safety profile. Clearly, further random-
initial amount of radiographic effusion). There was a small ized studies are needed to elucidate the role of this agent in
excess of adverse effects in the SK group, mainly associated pleural infection, and a UK based randomized trial using
with allergic or immunological reactions, but no excess of tPA in the treatment of pleural infection is currently under
systemic thrombolysis or bleeding.20 way.
A study in 127 patients was published later in 2005,
assessing intrapleural streptokinase (n = 47) versus normal
chest drainage (n = 70), with no placebo used.101 Once OTHER FIBRINOLYTICS STUDIES
again, treatment success, length of hospital stay and surgi-
cal referral were lower in the streptokinase group, and this Whether one fibrinolytic agent is better than another was
study found a significant mortality benefit in the SK group addressed by Bouros et al.77 who compared streptokinase
(4.2 percent versus 1.7 percent, p < 0.001). However, the and urokinase. This study was effectively a pilot of only 50
study was not placebo controlled, and therefore by defini- patients and allows a power calculation of the size of study
that would be needed to establish whether there is a differ-
ence between the two drugs. In this study both agents
failed in 8 percent of subjects. If it is assumed that the
failure rates do differ and are 8 percent versus 12 percent,
a randomized study of 3000 subjects would be needed to
1.00
assess whether there is a difference at the 5 percent level
Proportion surviving
Placebo
0.50 Streptokinase
Intrapleural DNAse
0.25
bination streptokinase and streptodornase (streptodornase without surgical interventions and surgical options are asso-
is streptococcal DNAse) or saline. Liquefaction of the exu- ciated with significant morbidity. It is likely that patients
dates was only achieved with the combination with late fibropurulent or organizing empyema will require
streptokinase and streptodornase.200 The second study a surgical procedure to establish complete drainage and ade-
involved nine specimens of pus collected from patients quate lung re-expansion, and a proportion of these patients
with either pleural empyema or abscess elsewhere.201 These may require additional surgery.203
specimens were also incubated with saline, combination Management guidelines suggest that surgical referral
streptokinase and streptodornase, human recombinant should be made after 7 days of failed medical therapy.6
DNAse or combination human recombinant DNAse and There are currently a number of surgical options available
streptokinase. Each of the DNAse-containing incubations to achieve these goals and include;
again effectively reduced the pus viscosity by over 95
percent with little change seen in the other groups.201 1. VATS;
2. mini-thoracotomy;
3. open thoracotomy with decortications;
Nutrition 4. rib resection with open drainage.
Adequate nutrition was identified as a key aim of therapy, Formal thoracotomy with decortication is a major pro-
and an important determinant of outcome in pleural cedure, and the techniques involved have changed little
infection during the First World War. Unfortunately, it is over the past 50 years.204 This procedure involves removal
still sometimes overlooked, while complex modern thera- of all fibrous tissue and blood clot from the visceral pleura
pies with adjunctive intrapleural agents and/or early mini- and evacuation of all the pus and fibrous tissue from the
mally invasive surgery receive much attention. Empyema pleural cavity. This eliminates sepsis and allows the lung to
patients characteristically suffer the protracted catabolic re-expand. The outcome from thoracotomy and decorti-
consequences of chronic infection, including immunode- cation is excellent, with reported success rates of 95
ficiency and slow recovery. In one series of 80 patients with percent118 in these highly selected patients, although its
infected pleural fluid, a low blood albumin level was the morbidity is substantial. Thoracotomy in general is associ-
most important determinant of a fatal outcome.202 It is ated with significant morbidity;99 one study reported that
therefore essential to provide adequate nutritional support over 80 percent will have post-thoracotomy pain at 3
from the time of diagnosis. This may require nasogastric months, decreasing only slightly to 61 percent at 1 year,
or occasionally intravenous feeding. and between 3 and 5 percent of patients experience severe
pain.146 Around two-thirds of patients require treatment
for pain up to 6 months after thoracotomy.145 A longer
Bronchoscopy follow-up study reported 55 percent of patients experienc-
ing chronic pain over 1 year after surgery, with 45 percent
A small number of empyema patients have a proximal experiencing pain at 2 years, 38 percent at 3 years and 30
obstructing lesion as the cause of their empyema. In a UK percent over 4 years after surgery.144 While the reported
series of 119 cases, 40 percent underwent bronchoscopy pain intensity was low, 23 percent of patients felt it inter-
and bronchial disease was identified in only five (<4 fered with their lives, and 9 percent required daily analge-
percent of the total sample).19 Bronchoscopy is therefore sia or intervention (anaesthetic block or referral to pain
not routinely appropriate in these patients.6 However, clinic).144
failure of the empyema to resolve or clinical or radi- The increasingly popular mini-thoracotomy is similar
ographic features of bronchial obstruction should indicate to a formal thoracotomy and decortication but involves a
either thoracic computed tomography or bronchoscopy to much smaller incision. This is possible as the operation is
identify any bronchial pathology. assisted by the use of a camera that allows the surgeon to
work with open instruments through the small incision.
The introduction of VATS in the late 1980s offered a
Surgery for pleural infection less traumatic alternative to thoracotomy and its use in the
treatment of pleural infection has become wide-
Medical therapy will fail in up to 30 percent of patients with spread.66,205–210 Its exact niche is still under evaluation, but
empyema,20,118 and these patients are referred for surgical recent studies suggest that VATS is effective and is a less
options. There is currently no definitive data that helps invasive strategy than thoracotomy and decortication in
define the point at which a patient with empyema should certain patient cohorts. Published retrospective series on
proceed to surgical intervention. Previous observational VATS management of empyema are generally favorable,
studies have suggested that the presence of purulent fluid or with reduced hospital inpatient time, postoperative com-
loculations increase the chances of surgical intervention,21 plications and length of operating time.18,76,99,113,207–214
whereas other studies have not found these parameters to be Conversion rates to formal thoracotomy are highly vari-
predictive.20,152 Many such patients will recover fully able in previous studies, ranging from 8 to 59
Other fibrinolytics studies 359
percent.18,99,113,211–215 Most VATS failures are with VATS with primary tube drainage in 18 patients, with
empyema in the organizing stage of the disease, and non- thrombolytic therapy permitted in both limbs after initial
blinded studies have cited certain bacteriological patterns intervention. They found a decreased length of hospital
and delayed referral (more than around 25 days), as pre- stay, decreased tube dwell time and decreased analgesic use
dictive of VATS failure.18,214 in the VATS group.92 However, in a larger study, Sonnappa
There are now two randomized prospective trials com- et al.108 randomized 60 children to VATS or chest tube +
paring VATS treatment to intercostal drainage in urokinase. No difference in terms of hospital stay, failure
association with fibrinolytics in adults71,76 (Table 26.8). Wait rate or radiological outcome at 6 months was seen between
et al.71 randomized 20 patients to chest tube drainage with the groups, with VATS being substantially more expensive
intrapleural streptokinase or VATS. Twenty patients were ($11 000 versus $9000).108 Therefore, further investigation
randomized, nine to chest tube with fibrinolytics and 11 to is required to fully elucidate the role of VATS in paediatric
VATS surgery. Five of the nine controls failed to resolve with empyema, although the largest randomized study to date
medical therapy compared with only one of the patients ran- has shown no benefit (and a cost harm) for VATs.
domized to VATS.71 However, the positive result in this trial Medical thoracoscopy (thoracoscopy performed under
is actually attributable to the very high control failure rate of local anesthetic and sedation usually by a physician) is very
53 percent71 (compared with approximately 20 percent in commonly used in the investigation and management of
most other studies), and the primary outcome (clinical patients with malignant pleural disease, and there is now
response at 3 days post randomization) was assessed in an increased interest in its use for infected pleural effusion. A
unblinded manner, which renders the result liable to sub- retrospective series of 127 patients was recently published,79
stantial observer bias. A Cochrane review conducted in 2005 in which medical thoracoscopy was used as primary therapy
evaluating surgical versus non-surgical treatment of for cases of multiloculated pleural infection identified at
empyema71 was only able to include this study (as the only thoracic ultrasound. This included cases of failed chest tube
randomized study at the time) and concluded that the study drainage. ‘Successful’ treatment was achieved using medical
was both underpowered and methodologically flawed, sug- thoracoscopy as primary therapy in 91 percent of cases, with
gesting that further evidence was required.71 a 6 percent thoracotomy conversion rate.79 Further studies
There has since been a larger randomized prospective are needed to elucidate the role of medical thoracoscopy in
trial comparing VATS to tube drainage in 70 adult sub- the treatment of pleural infection.
jects.76 A significantly lower proportion of patients under- Unfortunately, there are no radiological techniques that
going VATS as primary therapy required subsequent reliably predict which group of empyema patients requir-
thoracotomy and decortication (17 percent versus 37 ing surgical intervention may be amenable to VATS,
percent, p < 0.05),76 and VATS was associated with a although authors have suggested that delayed referral and
shorter hospital stay. However, the decision to proceed to certain bacteriological patterns are associated with worse
thoracotomy was not made blind to initial treatment allo- outcomes from VATs.
cation, and details of the tube only group (size of tube, use Open drainage by rib resection has the advantage that it
of flushes, suction and thrombolytics) are not described. can be performed under local or general anesthesia. It is
A VATS procedure is well tolerated with a low compli- not as major a procedure as decortication and therefore
cation rate in studies assessing complications for all indi- often more appropriate for patients at high operative
cations including pleural infection.216 The postsurgical risk.204 In this procedure, segments of one to three ribs
morbidity rate is very much lower than for thoracotomy in overlying the lower part of the empyema are resected and
a retrospective case series217 (96 percent of patients a large-bore tube inserted. This can be connected to a
symptom free at 2 years), although there are no direct colostomy bag to enable the patient to leave hospital. One
comparative studies. Therefore, the early indications are disadvantage of this technique is the length of time needed
that VATs may be a valuable tool in the management of to achieve complete healing (median time 6 months).204 A
empyema, but larger and more robust studies are needed modified version of the rib resection ‘open window thora-
to demonstrate this. costomy’ can also be used. This was devised by Eloesser221
There are a number of retrospective, non-randomized and involves the formation of a skin-lined track connect-
studies suggesting benefit from early VATS in pediatric ing the empyema cavity.
empyema, in terms of length of hospital stay, tube dwell
time and analgesic use.74,110,198,218,219 A study aggregating
the previous retrospective evidence on VATS in children Treatment options in patients with persisting
suggested an overall lower in hospital mortality rate from sepsis who are not fit enough for a general
primary operative management compared with primary anesthesia
tube drainage,220 but the pooled papers reviewed were
highly likely to suffer from selection bias being non- The frail patient who is a poor risk for general anesthesia
randomized. There are two randomized studies assessing or who declines surgical intervention despite failure of
the use of VATS in pediatric empyema,92,108 with conflict- their empyema to resolve with standard drainage and
ing results (Table 26.8). Kurt et al.92 compared primary antibiotics, presents a particular clinical challenge. There
Table 26.8 Randomized trials of surgery for pleural infection
Wait et al.71 SK versus VATS 20 250 000 IU SK daily for Randomized Reduced hospital stay in VATS group
3 days
9 SK versus 11 VATS Assessor unblinded 5/9 required VATS in SK group
High rate of medical All surgery in SK group managed with VATS
therapy failure
Bilgin et al.76 VATS versus intercostal 70 n/a Randomized Lower need for decortication in VATS group (17% versus 37%,
drain 35 VATS versus 35 p < 0.05). Shorter hospital stay with VATS (p < 0.05). One
intercostal drain Assessor unblinded death in drain group only.
Sonnappa et al.108 (Pediatric) 60 children 40 000u or 10 000 IU Randomized No difference in length of hospital stay, failure rate or
VATS versus drain + UK (30 drain + UK) versus UK bd for 3 days radiological outcome at 6 months. Lower cost in UK group
(24 VATS, 4 VATS + (according to child age) (p < 0.001)
mini-thoracotomy, Assessor unblinded
1 repeat VATS)
Kurt et al.92 (Paediatric) 18 children Reteplase 0.02 IU/kg Randomized Shorter length of stay, less analgesic use, fewer radiographs
VATS versus intercostal 10 VATS versus 8 up to 4 times/day Assessor unblinded and fewer subsequent intervenions required in VATS group
drain (fibrinolytic intercostal drain for up to 5 days None of VATS group required fibrinolytic
permissible after in
both groups)
SK, streptokinase; VATS, video-assisted thoracoscopic surgery.
Key points 361
49. Yeh TJ, Hall DP, Ellison RG. Empyema thoracis: A review of 110 infectious pleural effusions by use of reagent strips. Clin Infect Dis
cases. Am Rev Respir Dis 1963; 88: 785–90. 2000; 31: 914–19.
●50. Maskell NA, Batt S, Hedley EL, et al. The bacteriology of pleural 74. Bailey KA, Bass J, Rubin S, Barrowman N. Empyema management:
infection by genetic and standard methods and its mortality twelve years’ experience since the introduction of video-assisted
significance. Am J Respir Crit Care Med 2006; 174: 817–23. thoracoscopic surgery. J Laparoendosc Adv Surg Tech A 2005; 15:
◆51. Alfageme I, Munoz F, Pena N, Umbria S. Empyema of the thorax in 338–41.
adults. Etiology, microbiologic findings, and management. Chest ◆75. Barnes NP, Hull J, Thomson AH. Medical management of
1993; 103: 839–43. parapneumonic pleural disease. Pediatr Pulmonol 2005; 39:
●52. Sahn SA, Taryle DA, Good JT Jr. Experimental empyema. Time 127–34.
course and pathogenesis of pleural fluid acidosis and low pleural ●76. Bilgin M, Akcali Y, Oguzkaya F. Benefits of early aggressive
fluid glucose. Am Rev Respir Dis 1979; 120: 355–61. management of empyema thoracis. ANZ J Surg 2006; 76: 120–2.
53. Mavroudis C, Ganzel BL, Katzmark S, Polk HC Jr. Effect of ●77. Bouros D, Schiza S, Patsourakis G, et al. Intrapleural streptokinase
hemothorax on experimental empyema thoracis in the guinea pig. versus urokinase in the treatment of complicated parapneumonic
J Thorac Cardiovasc Surg 1985; 89: 42–9. effusions: a prospective, double-blind study. Am J Respir Crit Care
●54. Sasse SA, Jadus MR, Kukes GD. Pleural fluid transforming growth Med 1997; 155: 291–5.
factor-beta1 correlates with pleural fibrosis in experimental ●78. Bouros D, Schiza S, Tzanakis N, et al. Intrapleural urokinase versus
empyema. Am J Respir Crit Care Med 2003; 168: 700–5. normal saline in the treatment of complicated parapneumonic
●55. Kunz CR, Jadus MR, Kukes GD, et al. Intrapleural injection of effusions and empyema. A randomized, double-blind study. Am J
transforming growth factor-beta antibody inhibits pleural fibrosis Respir Crit Care Med 1999; 159: 37–42.
in empyema. Chest 2004; 126: 1636–44. 79. Brutsche MH, Tassi GF, Gyorik S, et al. Treatment of
◆56. Kroegel C, Antony VB. Immunobiology of pleural inflammation: sonographically stratified multiloculated thoracic empyema by
potential implications for pathogenesis, diagnosis and therapy. Eur medical thoracoscopy. Chest 2005; 128: 3303–9.
Respir J 1997; 10: 2411–18. 80. Buckingham SC, King MD, Miller ML. Incidence and etiologies of
●57. Broaddus VC, Boylan AM, Hoeffel JM, et al. Neutralization of IL-8 complicated parapneumonic effusions in children, 1996 to 2001.
inhibits neutrophil influx in a rabbit model of endotoxin-induced Pediatr Infect Dis J 2003; 22: 499–504.
pleurisy. J Immunol 1994; 152: 2960–7. 81. Castellote J, Lopez C, Gornals J, Domingo A, Xiol X. Use of reagent
●58. Broaddus VC, Hebert CA, Vitangcol RV, et al. Interleukin-8 is a strips for the rapid diagnosis of spontaneous bacterial empyema. J
major neutrophil chemotactic factor in pleural liquid of patients Clin Gastroenterol 2005; 39: 278–81.
with empyema. Am Rev Respir Dis 1992; 146: 825–30. 82. Chen LE, Langer JC, Dillon PA, et al. Management of late-stage
59. Idell S, Girard W, Koenig KB, McLarty J, Fair DS. Abnormalities of parapneumonic empyema. J Pediatr Surg 2002; 37: 371–4.
pathways of fibrin turnover in the human pleural space. Am Rev ●83. Chin NK, Lim TK. Controlled trial of intrapleural streptokinase in
Respir Dis 1991; 144: 187–94. the treatment of pleural empyema and complicated
60. Aleman C, Alegre J, Monasterio J, et al. Association between parapneumonic effusions. Chest 1997; 111: 275–9.
inflammatory mediators and the fibrinolysis system in infectious 84. Chu MW, Dewar LR, Burgess JJ, Busse EG. Empyema thoracis: lack
pleural effusions. Clin Sci (Lond) 2003; 105: 601–7. of awareness results in a prolonged clinical course. Can J Surg
●61. Sahn SA, Reller LB, Taryle DA, Antony VB, Good JT Jr. The 2001; 44: 284–8.
contribution of leukocytes and bacteria to the low pH of empyema ●85. Davies RJ, Traill ZC, Gleeson FV. Randomised controlled trial of
fluid. Am Rev Respir Dis 1983; 128: 811–15. intrapleural streptokinase in community acquired pleural infection.
●62. Light RW, MacGregor MI, Ball WC Jr, Luchsinger PC. Diagnostic Thorax 1997; 52: 416–21.
significance of pleural fluid pH and PCO2. Chest 1973; 64: ●86. Diacon AH, Theron J, Schuurmans MM, Van de Wal BW, Bolliger
591–6. CT. Intrapleural streptokinase for empyema and complicated
◆63. Miserocchi G. Physiology and pathophysiology of pleural fluid parapneumonic effusions. Am J Respir Crit Care Med 2004; 170:
turnover. Eur Respir J 1997; 10: 219–25. 49–53.
64. Mutsaers SE, Kalomenidis I, Wilson NA, Lee YC. Growth factors in 87. Dzielicki J, Korlacki W. The role of thoracoscopy in the treatment
pleural fibrosis. Curr Opin Pulm Med 2006; 12: 251–8. of pleural empyema in children. Surg Endosc 2006; 20: 1402–5.
◆65. Hamm H, Light RW. Parapneumonic effusion and empyema. Eur 88. Freitas M, Castelo A, Petty G, Gomes CE, Carvalho E. Viridans
Respir J 1997; 10: 1150–6. streptococci causing community acquired pneumonia. Arch Dis
66. Landreneau RJ, Keenan RJ, Hazelrigg SR, Mack MJ, Naunheim KS. Child 2006; 91: 779–80.
Thoracoscopy for empyema and hemothorax. Chest 1996; 109: 89. Jerjes-Sanchez C, Ramirez-Rivera A, Elizalde JJ, et al. Intrapleural
18–24. fibrinolysis with streptokinase as an adjunctive treatment in
67. Neff CC, vanSonnenberg E, Lawson DW, Patton AS. CT follow-up hemothorax and empyema: a multicenter trial. Chest 1996; 109:
of empyemas: pleural peels resolve after percutaneous catheter 1514–19.
drainage. Radiology 1990; 176: 195–7. 90. Kalfa N, Allal H, Lopez M, et al. Thoracoscopy in pediatric pleural
68. Light RW. Pleural diseases, 3rd edn. Baltimore: Williams and empyema: a prospective study of prognostic factors. J Pediatr Surg
Wilkins, 1995. 2006; 41: 1732–7.
69. Rahman NM, Batt S, Gleeson FV, et al. The relationship between 91. Kuboi S, Nomura H. Clinical background of cases showing a
lung parenchymal CT appearances and pleural bacteriology in positive culture of pleural effusion at Shin-Kokura Hospital over a
pleural infection [abstract]. Am J Respir Crit Care Med 2006; 173: period of 5 years. J Infect Chemother 2006; 12: 264–8.
A11. ●92. Kurt BA, Winterhalter KM, Connors RH, Betz BW, Winters JW.
70. Ahmed RA, Marrie TJ, Huang JQ. Thoracic empyema in patients Therapy of parapneumonic effusions in children: video-assisted
with community-acquired pneumonia. Am J Med 2006; 119: thoracoscopic surgery versus conventional thoracostomy drainage.
877–83. Pediatrics 2006; 118: e547–53.
●71. Wait MA, Sharma S, Hohn J, Dal Nogare A. A randomized trial of 93. Lahti E, Mertsola J, Kontiokari T, et al. Pneumolysin polymerase
empyema therapy. Chest 1997; 111: 1548–51. chain reaction for diagnosis of pneumococcal pneumonia and
72. Alfaro C, Fergie J, Purcell K. Emergence of community-acquired empyema in children. Eur J Clin Microbiol Infect Dis 2006; 25:
methicillin-resistant Staphylococcus aureus in complicated 783–9.
parapneumonic effusions. Pediatr Infect Dis J 2005; 24: 274–6. 94. Le Monnier A, Carbonnelle E, Zahar JR, et al. Microbiological
73. Azoulay E, Fartoukh M, Galliot R, et al. Rapid diagnosis of diagnosis of empyema in children: comparative evaluations by
364 Effusions from infections: parapneumonic effusion and empyema
culture, polymerase chain reaction, and pneumococcal antigen 115. Ozol D, Oktem S, Erdinc E. Complicated parapneumonic effusion
detection in pleural fluids. Clin Infect Dis 2006; 42: 1135–40. and empyema thoracis: microbiologic and therapeutic aspects.
95. Lim TK, Chin NK. Empirical treatment with fibrinolysis and early Respir Med 2006; 100: 286–91.
surgery reduces the duration of hospitalization in pleural sepsis. ●116. Thomson AH, Hull J, Kumar MR, Wallis C, Balfour LI. Randomised
Eur Respir J 1999; 13: 514–18. trial of intrapleural urokinase in the treatment of childhood
96. Lindstrom ST, Kolbe J. Community acquired parapneumonic empyema. Thorax 2002; 57: 343–7.
thoracic empyema: predictors of outcome. Respirology 1999; 4: 117. Estrera AS, Platt MR, Mills LJ, Shaw RR. Primary lung abscess. J
173–9. Thorac Cardiovasc Surg 1980; 79: 275–82.
97. Manuel PJ, Vives M, Esquerda A, Ruiz A. Usefulness of the British 118. LeMense GP, Strange C, Sahn SA. Empyema thoracis. Therapeutic
Thoracic Society and the American College of Chest Physicians management and outcome. Chest 1995; 107: 1532–7.
guidelines in predicting pleural drainage of non-purulent 119. Runyon BA, Antillon MR, Akriviadis EA, McHutchison JG. Bedside
parapneumonic effusions. Respir Med 2006; 100: 933–7. inoculation of blood culture bottles with ascitic fluid is superior to
98. Margenthaler JA, Weber TR, Keller MS. Predictors of surgical delayed inoculation in the detection of spontaneous bacterial
outcome for complicated pneumonia in children: impact of peritonitis. J Clin Microbiol 1990; 28: 2811–12.
bacterial virulence. World J Surg 2004; 28: 87–91. 120. Ferrer A, Osset J, Alegre J, et al. Prospective clinical and
99. Melloni G, Carretta A, Ciriaco P, et al. Decortication for chronic microbiological study of pleural effusions. Eur J Clin Microbiol
parapneumonic empyema: results of a prospective study. World J Infect Dis 1999; 18: 237–41.
Surg 2004; 28: 488–93. 121. Andersen MN, McDonald KE. Prognostic factors and results of
100. Menezes-Martins LF, Menezes-Martins JJ, Michaelsen VS, et al. treatment in pyogenic pulmonary abscess. J Thorac Cardiovasc
Diagnosis of parapneumonic pleural effusion by polymerase chain Surg 1960; 39: 573– 8.
reaction in children. J Pediatr Surg 2005; 40: 1106–10. 122. Hagan JL, Hardy JD. Lung abscess revisited. A survey of 184 cases.
●101. Misthos P, Sepsas E, Konstantinou M, et al. Early use of Ann Surg 1983; 197: 755–62.
intrapleural fibrinolytics in the management of postpneumonic 123. Torres A, Serra-Batlles J, Ferrer A, et al. Severe community-
empyema. A prospective study. Eur J Cardiothorac Surg 2005; 28: acquired pneumonia. Epidemiology and prognostic factors. Am Rev
599–603. Respir Dis 1991; 144: 312–18.
102. Mitri RK, Brown SD, Zurakowski D, et al. Outcomes of primary 124. Ko SC, Chen KY, Hsueh PR, Luh KT, Yang PC. Fungal empyema
image-guided drainage of parapneumonic effusions in children. thoracis: an emerging clinical entity. Chest 2000; 117: 1672–8.
Pediatrics 2002; 110: e37. 125. Eibenberger KL, Dock WI, Ammann ME, et al. Quantification of
103. Ozel SK, Kazez A, Kilic M, et al. Conservative treatment of pleural effusions: sonography versus radiography. Radiology 1994;
postpneumonic thoracic empyema in children. Surg Today 2004; 191: 681–4.
34: 1002–5. 126. Jones PW, Moyers JP, Rogers JT, et al. Ultrasound-guided
104. Paganini H, Guinazu JR, Hernandez C, et al. Comparative analysis thoracentesis: is it a safer method? Chest 2003; 123: 418–23.
of outcome and clinical features in children with pleural 127. O’Moore PV, Mueller PR, Simeone JF, et al. Sonographic guidance
empyema caused by penicillin-nonsusceptible and penicillin- in diagnostic and therapeutic interventions in the pleural space.
susceptible Streptococcus pneumoniae. Int J Infect Dis 2001; 5: AJR Am J Roentgenol 1987; 149: 1–5.
86–8. 128. Stavas J, vanSonnenberg E, Casola G, Wittich GR. Percutaneous
105. Saglani S, Harris KA, Wallis C, Hartley JC. Empyema: the use of drainage of infected and noninfected thoracic fluid collections. J
broad range 16S rDNA PCR for pathogen detection. Arch Dis Child Thorac Imaging 1987; 2: 80–7.
2005; 90: 70–3. 129. Potts DE, Taryle DA, Sahn SA. The glucose-pH relationship in
106. Satish B, Bunker M, Seddon P. Management of thoracic empyema parapneumonic effusions. Arch Intern Med 1978; 138: 1378–80.
in childhood: does the pleural thickening matter? Arch Dis Child ●130. Heffner JE, Brown LK, Barbieri C, DeLeo JM. Pleural fluid chemical
2003; 88: 918–21. analysis in parapneumonic effusions. A meta-analysis. Am J Respir
●107. Singh M, Mathew JL, Chandra S, Katariya S, Kumar L. Randomized Crit Care Med 1995; 151: 1700–8.
controlled trial of intrapleural streptokinase in empyema thoracis 131. Lesho EP, Roth BJ. Is pH paper an acceptable, low-cost alternative
in children. Acta Paediatr 2004; 93: 1443–5. to the blood gas analyzer for determining pleural fluid pH? Chest
●108. Sonnappa S, Cohen G, Owens CM, et al. Comparison of urokinase 1997; 112: 1291– 2.
and video-assisted thoracoscopic surgery for treatment of 132. Maskell NA, Gleeson FV, Darby M, Davies RJ. Diagnostically
childhood empyema. Am J Respir Crit Care Med 2006; 174: significant variations in pleural fluid pH in loculated
221–7. parapneumonic effusions. Chest 2004; 126: 2022–4.
109. Soriano T, Alegre J, Aleman C, et al. Factors influencing length of 133. Joseph J, Sahn SA. Connective tissue diseases and the pleura.
hospital stay in patients with bacterial pleural effusion. Chest 1993; 104: 262–70.
Respiration 2005; 72: 587–93. 134. Romano M, Catalano A, Nutini M, et al. 5-lipoxygenase regulates
110. Subramaniam R, Joseph VT, Tan GM, Goh A, Chay OM. Experience malignant mesothelial cell survival: involvement of vascular
with video-assisted thoracoscopic surgery in the management of endothelial growth factor. FASEB J 2001; 15: 2326–36.
complicated pneumonia in children. J Pediatr Surg 2001; 36: 135. Strizzi L, Vianale G, Catalano A, Muraro R, Mutti L, Procopio A.
316–19. Basic fibroblast growth factor in mesothelioma pleural effusions:
●111. Talib SH, Verma GR, Arshad M, Tayade BO, Rafeeque A. Utility of correlation with patient survival and angiogenesis. Int J Oncol
intrapleural streptokinase in management of chronic empyemas. J 2001; 18: 1093– 8.
Assoc Physicians India 2003; 51: 464–8. 136. Edwards JG, Cox G, Andi A, et al. Angiogenesis is an independent
112. Tsai TH, Jerng JS, Chen KY, Yu CJ, Yang PC. Community-acquired prognostic factor in malignant mesothelioma. Br J Cancer 2001;
thoracic empyema in older people. J Am Geriatr Soc 2005; 53: 85: 863–8.
1203–9. 137. Aelony Y, Yao JF, King RR. Prognostic value of pleural fluid pH in
113. Wurnig PN, Wittmer V, Pridun NS, Hollaus PH. Video-assisted malignant epithelial mesothelioma after talc poudrage.
thoracic surgery for pleural empyema. Ann Thorac Surg 2006; 81: Respiration 2006; 73: 334–9.
309–13. 138. Heffner JE, Heffner JN, Brown LK. Multilevel and continuous
114. Yao CT, Wu JM, Liu CC, et al. Treatment of complicated pleural fluid pH likelihood ratios for evaluating malignant pleural
parapneumonic pleural effusion with intrapleural streptokinase in effusions. Chest 2003; 123: 1887–94.
children. Chest 2004; 125: 566–71. 139. Heffner JE, Nietert PJ, Barbieri C. Pleural fluid pH as a predictor of
References 365
survival for patients with malignant pleural effusions. Chest 2000; 166. Patz EF Jr, Goodman PC, Erasmus JJ. Percutaneous drainage of
117: 79–86. pleural collections. J Thorac Imaging 1998; 13: 83–92.
140. Martinez-Moragon E, Aparicio J, Sanchis J, et al. Malignant pleural 167. Merriam MA, Cronan JJ, Dorfman GS, Lambiase RE, Haas RA.
effusion: prognostic factors for survival and response to chemical Radiographically guided percutaneous catheter drainage of pleural
pleurodesis in a series of 120 cases. Respiration 1998; 65: 108–13. fluid collections. AJR Am J Roentgenol 1988; 151: 1113–16.
◆141. Light RW, Erozan YS, Ball WC Jr. Cells in pleural fluid. Their value 168. Silverman SG, Mueller PR, Saini S, et al. Thoracic empyema:
in differential diagnosis. Arch Intern Med 1973; 132: 854–60. management with image-guided catheter drainage. Radiology
◆142. Light RW. Pleural effusion due to pulmonary emboli. Curr Opin 1988; 169: 5–9.
Pulm Med 2001; 7: 198–201. 169. Ulmer JL, Choplin RH, Reed JC. Image-guided catheter drainage of
143. Light RW. Pleural effusions. Med Clin North Am 1977; 61: 1339–52. the infected pleural space. J Thorac Imaging 1991; 6: 65–73.
144. Dajczman E, Gordon A, Kreisman H, Wolkove N. Long-term 170. vanSonnenberg E, Nakamoto SK, Mueller PR, et al. CT- and
postthoracotomy pain. Chest 1991; 99: 270–4. ultrasound-guided catheter drainage of empyemas after chest-
145. Kalso E, Perttunen K, Kaasinen S. Pain after thoracic surgery. Acta tube failure. Radiology 1984; 151: 349–53.
Anaesthesiol Scand 1992; 36: 96–100. 171. Crouch JD, Keagy BA, Delany DJ. ‘Pigtail’ catheter drainage in
146. Perttunen K, Tasmuth T, Kalso E. Chronic pain after thoracic thoracic surgery. Am Rev Respir Dis 1987; 136: 174–5.
surgery: a follow-up study. Acta Anaesthesiol Scand 1999; 43: 172. Hunnam GR, Flower CD. Radiologically guided percutaneous
563–7. catheter drainage of empyemas. Clin Radiol 1988; 39: 121–6.
147. Cham CW, Haq SM, Rahamim J. Empyema thoracis: a problem 173. Klein JS, Schultz S, Heffner JE. Interventional radiology of the
with late referral? Thorax 1993; 48: 925–7. chest: image-guided percutaneous drainage of pleural effusions,
148. Ashbaugh DG. Empyema thoracis. Factors influencing morbidity lung abscess, and pneumothorax. AJR Am J Roentgenol 1995; 164:
and mortality. Chest 1991; 99: 1162–5. 581–8.
149. Huang HC, Chang HY, Chen CW, Lee CH, Hsiue TR. Predicting 174. Park JK, Kraus FC, Haaga JR. Fluid flow during percutaneous
factors for outcome of tube thoracostomy in complicated drainage procedures: an in vitro study of the effects of fluid
parapneumonic effusion for empyema. Chest 1999; 115: 751–6. viscosity, catheter size, and adjunctive urokinase. AJR Am J
150. Armstrong P. Imaging of diseases of the chest, 3rd edn. London: Roentgenol 1993; 160: 165–9.
Mosby; 2001. 175. Gobien RP, Stanley JH, Schabel SI, et al. The effect of drainage
151. Wu RG, Yuan A, Liaw YS, et al. Image comparison of real-time tube size on adequacy of percutaneous abscess drainage.
gray-scale ultrasound and color Doppler ultrasound for use in Cardiovasc Intervent Radiol 1985; 8: 100–2.
diagnosis of minimal pleural effusion. Am J Respir Crit Care Med 176. Rothlin MA, Schob O, Klotz H, Candinas D, Largiader F.
1994; 150: 510–14. Percutaneous drainage of abdominal abscesses: are large-bore
◆152. Kearney SE, Davies CW, Davies RJ, Gleeson FV. Computed catheters necessary? Eur J Surg 1998; 164: 419–24.
tomography and ultrasound in parapneumonic effusions and ●177. Tuncozgur B, Ustunsoy H, Sivrikoz MC, et al. Intrapleural urokinase
empyema. Clin Radiol 2000; 55: 542–7. in the management of parapneumonic empyema: a randomised
153. Stark DD, Federle MP, Goodman PC, Podrasky AE, Webb WR. controlled trial. Int J Clin Pract 2001; 55: 658–60.
Differentiating lung abscess and empyema: radiography and 178. Barletta JF. Streptokinase and urokinase for the treatment of
computed tomography. AJR Am J Roentgenol 1983; 141: 163–7. pleural effusions and empyemas. Ann Pharmacother 1999; 33:
154. Brenner DJ. Radiation risks potentially associated with low-dose 495–8.
CT screening of adult smokers for lung cancer. Radiology 2004; 179. Bruserud O, Elsayed S, Pawelec G. At least five antigenic epitopes
231: 440–5. on the streptokinase molecule are recognized by human CD4+ TCR
155. Beir V. National Reserach Council. Committee on the Biological alpha beta+ T cells. Mol Immunol 1992; 29: 1097–104.
Effects of Ionising Radiations. Health effects of exposure to low 180. Bouros D, Schiza S, Panagou P, Drositis J, Siafakas N. Role of
levels of ionizing radiation. Washington, DC: National Academy streptokinase in the treatment of acute loculated parapneumonic
Press, 1990. pleural effusions and empyema. Thorax 1994; 49: 852–5.
156. Davis SD, Henschke CI, Yankelevitz DF, Cahill PT, Yi Y. MR imaging 181. Aye RW, Froese DP, Hill LD. Use of purified streptokinase in
of pleural effusions. J Comput Assist Tomogr 1990; 14: 192–8. empyema and hemothorax. Am J Surg 1991; 161: 560–2.
◆157. Evans AL, Gleeson FV. Radiology in pleural disease: state of the art. 182. Srivastava P, Godden DJ, Kerr KM, Legge JS. Fatal haemorrhage
Respirology 2004; 9: 300–12. from aortic dissection following instillation of intrapleural
158. Waite RJ, Carbonneau RJ, Balikian JP, et al. Parietal pleural streptokinase. Scott Med J 2000; 45: 86–7.
changes in empyema: appearances at CT. Radiology 1990; 175: 183. Temes RT, Follis F, Kessler RM, Pett SB Jr, Wernly JA. Intrapleural
145–50. fibrinolytics in management of empyema thoracis. Chest 1996;
159. Duysinx B, Nguyen D, Louis R, et al. Evaluation of pleural disease 110: 102–6.
with 18-fluorodeoxyglucose positron emission tomography 184. Godley PJ, Bell RC. Major hemorrhage following administration of
imaging. Chest 2004; 125: 489–93. intrapleural streptokinase. Chest 1984; 86: 486–7.
160. Duysinx BC, Larock MP, Nguyen D, et al. 18F-FDG PET imaging in 185. Davies CW, Lok S, Davies RJ. The systemic fibrinolytic activity of
assessing exudative pleural effusions. Nucl Med Commun 2006; intrapleural streptokinase. Am J Respir Crit Care Med 1998; 157:
27: 971–6. 328–30.
161. Taryle DA, Good JT Jr, Morgan EJ III, Reller LB, Sahn SA. Antibiotic 186. Berglin E, Ekroth R, Teger-Nilsson AC, William-Olsson G.
concentrations in human parapneumonic effusions. J Antimicrob Intrapleural instillation of streptokinase. Effects on systemic
Chemother 1981; 7: 171–7. fibrinolysis. Thorac Cardiovasc Surg 1981; 29: 124–6.
162. Hughes CE, Van Scoy RE. Antibiotic therapy of pleural empyema. 187. Juhlin P, Bostrom PA, Torp A, Bredberg A. Streptokinase
Semin Respir Infect 1991; 6: 94–102. antibodies inhibit reperfusion during thrombolytic therapy with
163. Niwa T, Nakamura A, Kato T, et al. Pharmacokinetic study of streptokinase in acute myocardial infarction. J Intern Med 1999;
pleural fluid penetration of carbapenem antibiotic agents in 245: 483–8.
chemical pleurisy. Respir Med 2006; 100: 324–31. ●188. Cameron R, Davies HR. Intrapleural fibrinolytic therapy versus
164. Light RW. Management of parapneumonic effusions. Arch Intern conservative management in the treatment of parapneumonic
Med 1981; 141: 1339–41. effusions and empyema. Cochrane Database Syst Rev 2004; 2:
165. Miller KS, Sahn SA. Chest tubes. Indications, technique, CD002312.
management and complications. Chest 1987; 91: 258–64. ●189. Tokuda Y, Matsushima D, Stein GH, Miyagi S. Intrapleural
366 Effusions from infections: parapneumonic effusion and empyema
fibrinolytic agents for empyema and complicated parapneumonic 211. Solaini L, Prusciano F, Bagioni P. Video-assisted thoracic surgery
effusions: a meta-analysis. Chest 2006; 129: 783–90. in the treatment of pleural empyema. Surg Endosc 2007; 21:
190. Maskell NA, Gleeson FV. Images in clinical medicine. Effect of 280–4.
intrapleural streptokinase on a loculated malignant pleural 212. Petrakis IE, Kogerakis NE, Drositis IE, et al. Video-assisted
effusion. N Engl J Med 2003; 348: e4. thoracoscopic surgery for thoracic empyema: primarily, or after
191. Walker CA, Shirk MB, Tschampel MM, Visconti JA. Intrapleural fibrinolytic therapy failure? Am J Surg 2004; 187: 471–4.
alteplase in a patient with complicated pleural effusion. Ann 213. Luh SP, Chou MC, Wang LS, Chen JY, Tsai TP. Video-assisted
Pharmacother 2003; 37: 376–9. thoracoscopic surgery in the treatment of complicated
192. Skeete DA, Rutherford EJ, Schlidt SA, et al. Intrapleural tissue parapneumonic effusions or empyemas: outcome of 234 patients.
plasminogen activator for complicated pleural effusions. J Trauma Chest 2005; 127: 1427–32.
2004; 57: 1178–83. 214. Lardinois D, Gock M, Pezzetta E, et al. Delayed referral and Gram-
193. Feola GP, Shaw LC, Coburn L. Management of complicated negative organisms increase the conversion thoracotomy rate in
parapneumonic effusions in children. Tech Vasc Interv Radiol patients undergoing video-assisted thoracoscopic surgery for
2003; 6: 197–204. empyema. Ann Thorac Surg 2005; 79: 1851– 6.
194. Bishop NB, Pon S, Ushay HM, Greenwald BM. Alteplase in the 215. Hope WW, Bolton WD, Stephenson JE. The utility and timing of
treatment of complicated parapneumonic effusion: a case report. surgical intervention for parapneumonic empyema in the era of
Pediatrics 2003; 111: E188–90. video-assisted thoracoscopy. Am Surg 2005; 71: 512–14.
195. Wells RG, Havens PL. Intrapleural fibrinolysis for parapneumonic 216. Katlic MR. Video-assisted thoracic surgery utilizing local
effusion and empyema in children. Radiology 2003; 228: 370–8. anesthesia and sedation. Eur J Cardiothorac Surg 2006; 30:
196. Weinstein M, Restrepo R, Chait PG, et al. Effectiveness and safety 529–32.
of tissue plasminogen activator in the management of complicated 217. Stammberger U, Steinacher C, Hillinger S, et al. Early and long-
parapneumonic effusions. Pediatrics 2004; 113: e182–5. term complaints following video-assisted thoracoscopic surgery:
197. Ray TL, Berkenbosch JW, Russo P, Tobias JD. Tissue plasminogen evaluation in 173 patients. Eur J Cardiothorac Surg 2000; 18:
activator as an adjuvant therapy for pleural empyema in pediatric 7–11.
patients. J Intensive Care Med 2004; 19: 44–50. 218. Goldschlager T, Frawley G, Crameri J, et al. Comparison of
198. Gates RL, Hogan M, Weinstein S, Arca MJ. Drainage, fibrinolytics, thoracoscopic drainage with open thoracotomy for treatment of
or surgery: a comparison of treatment options in pediatric paediatric parapneumonic empyema. Pediatr Surg Int 2005; 21:
empyema. J Pediatr Surg 2004; 39: 1638–42. 599–603.
199. Jaffe A, Bush A. Cystic fibrosis: review of the decade. Monaldi Arch 219. Suchar AM, Zureikat AH, Glynn L, et al. Ready for the frontline: is
Chest Dis 2001; 56: 240–7. early thoracoscopic decortication the new standard of care for
●200. Light RW, Nguyen T, Mulligan ME, Sasse SA. The in vitro efficacy advanced pneumonia with empyema? Am Surg 2006; 72: 688–92.
of varidase versus streptokinase or urokinase for liquefying thick 220. Avansino JR, Goldman B, Sawin RS, Flum DR. Primary operative
purulent exudative material from loculated empyema. Lung 2000; versus nonoperative therapy for pediatric empyema: a meta-
178: 13–18. analysis. Pediatrics 2005; 115: 1652–9.
●201. Simpson G, Roomes D, Heron M. Effects of streptokinase and 221. Eloesser L. Of an operation for tuberculous empyema. Ann Thorac
deoxyribonuclease on viscosity of human surgical and empyema Surg 1969; 8: 355–7.
pus. Chest 2000; 117: 1728–33. 222. Tremblay A, Michaud G. Single-center experience with 250
202. Nwiloh J, Freeman H, McCord C. Malnutrition: an important tunnelled pleural catheter insertions for malignant pleural
determinant of fatal outcome in surgically treated pulmonary effusion. Chest 2006; 129: 362–8.
suppurative disease. J Natl Med Assoc 1989; 81: 525–9. 223. Furman AC, Jacobs J, Sepkowitz KA. Lung abscess in patients with
203. Martella AT, Santos GH. Decortication for chronic postpneumonic AIDS. Clin Infect Dis 1996; 22: 81–5.
empyema. J Am Coll Surg 1995; 180: 573–6. 224. Afessa B. Pleural effusion and pneumothorax in hospitalized
204. Katariya K, Thurer RJ. Surgical management of empyema. Clin patients with HIV infection: the Pulmonary Complications, ICU
Chest Med 1998; 19: 395–406. support, and Prognostic Factors of Hospitalized Patients with HIV
205. Lackner RP, Hughes R, Anderson LA, Sammut PH, Thompson AB. (PIP) Study. Chest 2000; 117: 1031–7.
Video-assisted evacuation of empyema is the preferred procedure 225. Minkin R, Shapiro JM. Corynebacterium afermentans lung abscess
for management of pleural space infections. Am J Surg 2000; 179: and empyema in a patient with human immunodeficiency virus
27–30. infection. South Med J 2004; 97: 395–7.
206. Waller DA, Rengarajan A. Thoracoscopic decortication: a role for 226. Shapiro JM, Romney BM, Weiden MD, White CS, O’Toole KM.
video-assisted surgery in chronic postpneumonic pleural Rhodococcus equi endobronchial mass with lung abscess in a
empyema. Ann Thorac Surg 2001; 71: 1813–16. patient with AIDS. Thorax 1992; 47: 62–3.
207. Angelillo Mackinlay TA, Lyons GA, Chimondeguy DJ, et al. VATS 227. Marron A, Roson B, Mascaro J, Carratala J. Listeria monocytogenes
debridement versus thoracotomy in the treatment of loculated empyema in an HIV infected patient. Thorax 1997; 52: 745–6.
postpneumonia empyema. Ann Thorac Surg 1996; 61: 1626–30. 228. Mulanovich VE, Dismukes WE, Markowitz N. Cryptococcal
208. Cunniffe MG, Maguire D, McAnena OJ, Johnston S, Gilmartin JJ. empyema: case report and review. Clin Infect Dis 1995; 20:
Video-assisted thoracoscopic surgery in the management of 1396–8.
loculated empyema. Surg Endosc 2000; 14: 175–8. 229. Wolday D, Seyoum B. Pleural empyema due to Salmonella
209. Powell LL, Allen R, Brenner M, Aryan HE, Chen JC. Improved paratyphi in a patient with AIDS. Trop Med Int Health 1997; 2:
patient outcome after surgical treatment for loculated empyema. 1140–2.
Am J Surg 2000; 179: 1–6. 230. Khwaja S, Rosenbaum DH, Paul MC, et al. Surgical treatment of
210. Podbielski FJ, Maniar HS, Rodriguez HE, Hernan MJ, Vigneswaran thoracic empyema in HIV-infected patients: severity and
WT. Surgical strategy of complex empyema thoracis. JSLS 2000; 4: treatment modality is associated with CD4 count status. Chest
287–90. 2005; 128: 246–9.
27
Effusions from infections: tuberculosis
of a TB patient being HIV positive, the presence of pleural release interleukin (IL)-1 to IL-6, tumor necrosis factor
effusion did not predict HIV seropositivity. Pleural effu- alpha (TNF-a) as well as alpha and beta chemokines.
sion was present in 58 of 185 (31 percent) of the HIV- Alpha chemokines (IL-8, neutrophil activating protein
positive and 38 of 119 (32 percent) of the negative subjects. [NAP-2]) are chemotactic for neutrophils, lymphocytes or
Similar results were obtained in Zimbabwe,19 where no both, while beta chemokines (macrophage inflammatory
association was found between the presence of pleural protein [MIP-1] and monocyte chemoattractant protein
effusion and HIV status in a study of 422 patients (202 [MCP-1]) are chemotactic for monocytes and
HIV-positive and 220 negative). macrophages.33–35
This disparity in the results could be explained by the Mohammed et al.33–34 reported that pleural fluid from
immunological status of the HIV positive subjects. Jones et patients with tuberculosis pleuritis contained significantly
al.20 observed that pleural TB in HIV-positive patients was (p < 0.001) more biologically active MIP-1 and MCP-1
more frequent in patients with high CD4 counts. than did fluids from patients with congestive heart failure.
Tuberculous pleural effusion is one of the most fre- Unfortunately, they did not measure these cytokines in
quent causes of exudates in large series of pleural effusions other types of exudative pleural fluids. Antigenic MIP-1
in immunocompetent patients. If we exclude patients with and MCP-1 were detected by immunocytochemistry in
underlying pulmonary disease, TB is the most common pleural biopsy sections from patients with tuberculous
cause of pleural exudates in many areas of the world.21,22 pleurisy. In vitro, pleural mesothelial cell stimulated with
Bacillus Calmette-Guérin (BCG) produced MIP-1 and
MCP-1. Reverse transcription-polymerase chain reaction
ETIOLOGY AND PATHOGENESIS studies confirmed that BCG and interferon (IFN)-g
induced MIP-1 and MCP-1 expression in mesothelial cells,
Pleural TB reflects a compartmentalized activation of demonstrating that Th1 and Th2 cytokines may regulate
immunity mediated by cells in the pleural space. the C-C chemokines expression in mesothelial cells and
Mycobacterial proteins access this space through rupture thus play a biologically important role in the recruitment
of a subpleural focus 6–12 weeks after a primary infec- of mononuclear cells to the pleural space. Also, large con-
tion,23 and rarely from a vertebral focus by direct exten- centrations of IL-8 have been found by Pace et al.36 in
sion.24 With fewer new infections, pleural disease now cancer and tuberculosis pleural effusions. The concentra-
more commonly represents reactivation than primary tion of IL-8 correlated best with lymphocyte recruitment
infection and is present in approximately 7 percent of and not with neutrophil recruitment. This chemotactic
cases of active pulmonary TB. In a study from Maryland, activity was partly blocked by anti-IL-8 antibody and was
USA, organisms isolated from patients with pleural TB stimulated by the induction of messenger RNA for IL-8
were significantly less likely to be genotypically clustered and IL8 protein production in pleural macrophages. These
than patients presenting with pulmonary TB, suggesting findings demonstrate that pleural macrophages and
that pleural TB is more frequently reactivation TB than is mesothelial cells produce IL-8 in tuberculous pleurisy and
pulmonary TB.25 In a study in Edinburgh, UK, from this is an important factor for the recruitment of lympho-
1980–91, Moudgil et al.26 attributed 64 percent of the cytes into pleural space.37
new cases of pleural TB to reactivation TB. This explains why the pleural fluid from the patient
In an animal model with pleural effusion induced by with TB is predominantly lymphocytic. In fact, values
the intrapleural injection of purified protein derivative higher than 85 percent are reported to be very suggestive of
(PPD) in sensitized animals, neutrophils were the pre- this diagnosis,38–40 and a predominance of neutrophils are
dominant cells in the initial phase, and after the third day only found in the early phase of TB.41
lymphocytes predominated.27 (See also Chapter 14, In HIV-positive patients, it seems reasonable to
Experimental models: pleural disease other than mesothe- hypothesize that pleural TB would be manifested with a
lioma.) The pleural fluid in TB is predominantly lympho- less intense immunological response. The findings of
cytic. Many of the lymphocytes are T-helper type 1 (Th1) Kitinya et al.42 support this. In their series of 57 patients
cells,28 and the CD4/CD8 ratio observed was 4.3 in pleural with pleural TB, 69 percent of the 36 HIV-positive patients
fluid, while it was 1.6 in peripheral blood.29–31 had mycobacteria in the pleural fluid, whereas only 38
This cellular component reflects the immunological percent of the 21 HIV-negative patients had mycobacteria
reaction of tuberculous pleurisy in which mesothelial cells, in the pleural fluid. Also, the presence of granulomas on
neutrophils, lymphocytes, monocytes and their cytokines biopsy was less frequent in the HIV-positives (61 percent)
are involved. The most probable hypothesis for the patho- than in the HIV-negatives (90 percent).
genesis of pleural TB suggests that the endothelial and Some authors43,44 report that HIV-positive patients
mesothelial cells are incipient protagonists of the respond to tuberculous infection with local recruitment of
process.32 T lymphocytes, despite their decrease in the serum. Jones
These cells, exposed to inflammatory stimuli derived et al.20 found that patients with HIV and pleural TB with
from mycobacteria (glycolipids and lipoproteins) medi- CD4 counts >200/mL were more likely to develop a pleural
ated chemokine expression in pleural mesothelial cells, effusion (8 of 29, 28 percent) than those with CD4 counts
Diagnosis 369
<200/mL (6 of 58, 10 percent). This suggests that the obtained in the first thoracentesis.54 A finding of >5
immune state must be relatively well preserved in order for percent of mesothelial cells is exceptional in pleural TB
pleural TB to develop. and serves to exclude the diagnosis.55
Although the immune mechanisms linked to the activa-
tion of the Th1 cells are important in the development of
tuberculous pleural effusion, pleuritis after the primary RADIOLOGY
infection does not confer protection from subsequent TB,
as 65 percent of untreated patients develop active pul- Approximately 30 percent of patients with pleural TB
monary disease45,46 after the tuberculous pleuritis resolves. present with ipsilateral radiological disease in the lung
parenchyma as assessed with the chest radiograph. This
percentage may be as high as 86 percent with chest com-
SYMPTOMS AND LABORATORY FINDINGS puted tomography (CT).56
The pleural effusion is generally unilateral and small to
Pleural tuberculosis affects men more frequently than moderate in size. Massive effusions are observed in 12–29
women, with an approximate ratio of 3:1 and affects percent of cases, and tuberculosis is one of the three most
mainly young adults with a mean age around 35 years.47 frequent causes of massive pleural exudates: malignancy
However, more recent studies report an older mean age and empyema being the other two.57,58 Joseph et al.59
with 10–15 percent of the patients being over the age of reported that massive effusions were more frequent in
70 years. HIV-positive patients. Less than 10 percent of tuberculous
The clinical presentation is usually acute or subacute effusions are bilateral and bilateral effusions are generally
and the interval from the beginning of symptoms to the in HIV-positive patients.
diagnosis is less than 1 month in most patients. The most Identifying whether pleural TB is a primary infection or
common symptoms are: cough 71–94 percent; fever a reinfection can be quite difficult.60 Some authors have
71–100 percent; chest pain 78–82 percent; and dyspnea reported that patients with pleural tuberculosis were sig-
12–15 percent.48,49 Rarely are patients asymptomatic, but nificantly less likely to be genotypically clustered than
this is more common among the aged. Physical examina- patients presenting with pulmonary tuberculosis.25
tion reveals a decrease in the breath sounds over the Normally, radiological criteria are used to differentiate
affected area and dullness to percussion. A pleural friction primary and reinfection tuberculosis. Primary TB mani-
rub is seldom present. The Mantoux skin test is negative in fests as lymphadenopathy or lower lobe infiltrates. In cases
30 percent of the cases with the standard five-unit test, reported as pleural TB reactivation, the following charac-
more than half are negative with one unit test. However, teristics are described: advanced age, tobacco habit,
the skin test usually becomes become positive with repeti- alcohol abuse, significant weight loss, presence of
tion after 2–6 weeks.48,49 Mycobacterium in sputum and pleural effusions of smaller
In patients with HIV, the clinical presentation of TB is volume. However, 8–20 percent of patients with pul-
different.50 In Tanzania, Richter et al.51 found that HIV- monary TB have atypical radiological appearances,60 and
positive patients with pleural TB present with a more pro- diagnosis often requires the use of immunological crite-
longed interval between the beginning of symptoms and ria.61
diagnosis than do the HIV-negative patients. The HIV-
positive patients are also more symptomatic, with fever,
dyspnea, night sweats and asthenia, while the physical DIAGNOSIS
examination often reveals hepatomegaly, splenomegaly
and lymphadenopathy. The Mantoux skin test is more fre- The clinical likelihood of the diagnosis of pleural TB varies
quently negative in HIV-positive (47 percent) than in depending on the prevalence of TB and HIV co-infection.
HIV-negative patients (12 percent). In a significant percentage of cases, pleural TB is found in
The rate of identification of mycobacteria in sputum adults and elderly people. Elderly individuals tend to have
from patients with isolated pleural TB is low (4–7 percent) additional comorbidities, making the diagnosis more diffi-
but is much higher (28–50 percent) if there are associated cult. We have reported a diagnosis sensitivity of 58 percent
pulmonary infiltrates.49,52 Routine laboratory studies are using age, clinical data and thorax X-ray.62
nonspecific. A frequent finding is a high sedimentation Pleural fluid with TB pleuritis is usually an exudate
rate with normal leukocyte count.49 containing predominantly lymphocytes, with few
Thoracentesis shows a clear yellow or a serosan- eosinophils (<10 percent), or mesothelial cells (<5
guineous liquid. The pleural fluid is an exudate and usually percent).63
contains predominantly lymphocytes – a finding very sug- A definitive diagnosis of tuberculous pleurisy requires
gestive of TB pleuritis if the patient is young.53 The pleural demonstration of M. tuberculosis in sputum or pleural
fluid may contain predominantly neutrophils in the very specimens or the presence of caseating granulomas in the
early phase.41 Eosinophilia >10 percent is exceptional and pleura. Sputum culture is positive in 30–50 percent of
practically excludes the diagnosis, as long as this finding is patients with pleural and pulmonary tuberculosis, but in
370 Effusions from infections: tuberculosis
only 4 percent in patients with isolated pleural effusion.64 The low sensitivity of finding mycobacteria in the
However, this increased to 55 percent with sputum induc- pleural fluid explains why blind pleural biopsy has been
tion.65 The sensitivity of pleural fluid culture for the diag- the conventional diagnostic procedure for tuberculous
nosis of tuberculous pleuritis is 23–86 percent and that of pleurisy for the last 40 years. Pleural biopsy is an invasive
needle pleural biopsy is 39–71 percent. technique, is difficult to perform on infants, takes a long
The finding of necrotizing caseating granulomas in time for culture results and requires experience.
biopsy specimens is most likely to provide a diagnosis (in Alternative methods have been developed that attempt to
51–88 percent of cases) (Table 27.1).1,21,26,50,66,67,68,69 The diagnose TB pleuritis from tests on the pleural fluid.
use of all the above-mentioned tests provides a diagnosis New alternatives for diagnosis have been aimed at the
in 82–98 percent. Increasing numbers of specimens actual detection of mycobacteria (BACTEC [see below]
obtained by blind pleural biopsy in pleural TB does not and polymerase chain reaction [PCR]) and at identifying
significantly increase the diagnostic performance. This is biomarkers based on understanding of the immunological
probably because tuberculous pleuritis is a diffuse process mechanisms of TB pleuritis (adenosine deaminase [ADA],
affecting the entire pleural surface, unlike malignant ADA isoenzymes, IFN-g, lysozyme, tuberculostearic acid,
disease in which involvement of the parietal pleural is SC5b-9, IL-2, IL-6, IL-12).
patchy or non-existent.70 The radiometric mycobacterial culture system
Diagnostic performance of the various tests differs (BACTEC; Becton Dickinson, Franklin Lakes, NJ, USA)
depending on HIV status. Anergy is more frequent in accelerates diagnosis by 2–3 weeks compared with conven-
HIV-positive patients (47–59 percent) than HIV-negative tional culture. Bedside inoculation of pleural fluid pro-
patients (12–24 percent). The likelihood of isolation of vides more positive results (11 of 24) than laboratory
mycobacteria from sputum is higher in HIV-positive inoculation (4 of 24).71,72
patients (53 versus 23 percent).42,52 Direct smear of pleural For the detection of M. tuberculosis deoxyribonucleic
fluid (15 versus 8 percent) and culture of pleural biopsy acid (DNA) in sputum samples, PCR (Amplicor kit; Roche
tissue (91 versus 78 percent) are more likely to yield Molecular Systems, Branchburg, NJ, USA) has a false posi-
mycobacteria in HIV-positive patients. Findings of granu- tive rate of less than 1 percent.73,74 In pleural fluid, various
lomas are less frequent (77 versus 88 percent) in HIV pos- investigators have reported disparate results with PCR
itive patients. Kitinya et al.42 described three (sensitivity 61–94 percent and specificity 78–100
morphological types of TB related to HIV status – reactive, percent);75–77 hence, its clinical use is not recommended.
hyporeactive and non-reactive – according to whether well Querol et al.76 found that the sensitivity of PCR was 100
formed, poorly formed or no granulomas were present on percent in pleural tuberculosis with positive cultures and
biopsy. In 36 HIV-positive patients, 14 were non-reactive 60 percent in cases with negative cultures, in both cases
or hyporeactive, while this occurred in only 2 out of 21 superior to both pleural fluid acid-fast bacilli (AFB) smear
HIV-negative patients. In the 14 non-reac- (14 percent) and culture (52 percent). However, other
tive/hyporeactive HIV-positive patients, the culture was authors have reported much poorer results.
positive in 11, compared with 14 out of 22 of the reactive The most promising new test in the diagnosis of pleural
patients. These data contrast with those of other authors tuberculosis is ADA. In 1973, Piras et al.78 first reported
and can probably be explained by the different CD4 status that the ADA levels were elevated in the cerebrospinal fluid
in the series studied, as reported by Jones et al.20 of patients with tuberculous meningitis, and in 1978 the
same authors reported that the pleural fluid ADA levels lymphocytes in pleural fluid. In a series of 303 cases of
were elevated in tuberculous pleuritis. which 143 were TB, they found a sensitivity of 91 percent
There is much evidence supporting the usefulness of an and a specificity of 81 percent using an ADA cut-off value
increased level of pleural fluid ADA in establishing the of 50 IU. When the patients were also required to have a
diagnosis of tuberculous pleurisy (Table 127.2).79–88 lymphocyte/neutrophil ratio equal or greater than 0.75,
Nevertheless, high levels of ADA can also be found in other the sensitivity fell to 88 percent while the specificity
diseases such as empyemas, malignant lymphomas and increased to 95 percent.
collagen–vascular diseases. Its use in the diagnosis of Pérez-Rodriguez et al.,85 in their initial series of 304
pleural TB shows a sensitivity of 77–100 percent and a consecutive patients with pleural effusion undergoing tho-
specificity of 81–97 percent. racentesis, including 48 with pleural TB, found the sensi-
Bañales et al.82 performed a metanalysis of 2251 cases in tivity and specificity of using ADA test (with a cut-off level
which ADA was carried out by the Blake–Berman of 40 IU) to be 87.5 and 96.8 percent, respectively. Six were
method.87 They reported a sensitivity of 99 percent and a false negatives (1.9 percent) and eight were false positives
specificity of 89 percent. There were 116 false positives (5.7 (2.6 percent), including four empyemas, three lymphomas
percent) and these included 20 carcinomas, 18 lym- and a melanoma.
phomas, 52 empyemas and 6 mesotheliomas. The cut-off In view of the above, one can conclude that the ADA
level varied between 40 and 71 IU/L, which identified the level is a good parameter for identifying pleural TB.
need for individual cut-off levels of reference in respective Adding the criteria of pleural fluid lymphocyte/neutrophil
centers. ratio above 0.75 and an age limit of <35 years both
In a study of 405 cases of pleural effusion including 91 improve diagnostic specificity.
due to TB, Valdes et al.80 found a sensitivity of 100 percent In HIV-positive patients, the immunological response
and specificity of 95 percent using a cut-off value equal to is diminished (as observed by Kitinya et al.42) and one
or greater than 47 IU/L. The number of false positives was might expect the ADA to be less useful diagnostically.
16 (3.5 percent) (malignant 10, parapneumonic 1 and However, Riantawan et al.44 found that the diagnostic
empyema 5). In a series of 129 patients under the age of 35 utility of ADA was as good in HIV patients (n = 37) as in
of whom 39 had TB, these same authors found a sensitiv- HIV-negative (n = 52) ones. The average levels of ADA
ity of 100 percent and specificity of 87.5 percent using a were 110 and 114 IU, respectively, and when a cut-off
cut-off value of 47 IU. All the false positives were empye- value of 60 IU was applied, the sensitivity in both groups
mas. The study suggested that biopsy may be unnecessary was 95 percent. The percentage of lymphocytes in serum
in regions where the prevalence of TB is high in individu- was less in the HIV positive subjects than in the HIV-
als under the age of 35 with ADA higher than 47 IU, given negative subjects, while in pleural fluid the lymphocyte
that false-positive empyemas are easily identifiable. percentage was similar in both groups.
Burgess et al.79 analyzed the diagnostic performance of Some authors believe that ADA is a better negative pre-
individual ADA and of that associated with the value of dictive parameter than the combination of symptoms plus
Table 27.2 Utility of adenosine deaminase (ADA) in the diagnosis of tuberculous pleurisy
lymphocytes in pleural fluid plus PPD. In a series of 19 Pleural fluid levels of lysozyme are also frequently
pleural effusions of unidentified etiologies with positive higher in pleural TB than in pleural fluids of other etiolo-
PPD and ADA less than 47 IU, none developed tuberculo- gies. It is an enzyme present in the epithelioid cells of gran-
sis during a 69-month follow-up.89 ulomas, activated macrophages and certain tumor cells. Its
ADA in pleural fluid represents the sum of two isoen- diagnostic effectiveness in pleural TB has been described
zymes ADA1 + ADA2.34 ADA1 is ubiquitous and is neces- with varying sensitivities (66–100 percent) and a low speci-
sary for the breakdown of the substrate adenosine to ficity (66 percent).80,95 False positives have been reported
2′deoxyadenosine. This enzyme is important because a low in parapneumonic effusions, empyemas, malignancies,
level of 2′deoxyadenosine is essential for the proper func- post-thoracic surgery, collagen–vascular disease, heart
tioning of immune cells. In contrast, ADA2 is not ubiqui- failure and, on rare occasions, pleural sarcoidosis. Its
tous, and coexists with ADA1 only in monocytes and routine use is not recommended.
macrophages. Other less relevant parameters studied are: tubercu-
Gakis90 demonstrated that the level of ADA in mono- lostearic acid, monoclonal antibodies, SC5b-9 and other
cytes and macrophages is always low, except on certain cytokines (IL-2, -3, -4, -5, -6, -10 and -12).
occasions in which ADA2 is dramatically increased due to Tuberculostearic acid is a structural component of
infections by intracellular organisms. From this it can be mycobacteria and actinomycetes. Measured in pleural
deduced that ADA1 is produced by lymphocytes and fluid, its sensitivity for diagnosing tuberculous pleuritis is
monocytes, while ADA2 is only produced by monocytes. reported as 68–90 percent and its specificity 52 percent.96,97
This explains why some studies find a correlation Anti-mycobacterial antibodies have seldom been
between ADA and monocytes but not lymphocytes in studied in pleural TB.98 The only one to have been devel-
pleural TB, and why ADA in false-positive cases is prima- oped to date is anti-P32 (antigen specific to mycobacteria).
rily ADA1.91 P32 antigen is restricted to Mycobacterium tuberculosis,
For a better discriminating capability in the diagnosis of bovis, kansasi and avium. Pleural fluid and the correspon-
pleural TB, some authors have used the levels of ADA2 and ding sera were obtained from five patients with pleural TB
found a high sensitivity and specificity of 100 and 91 and 14 patients from pleural effusions of other origins. The
percent, respectively.92 Another measure that has been pleural fluids and sera from patients with pleural TB con-
used is the ADA1/ADA total. With this measure a cut-off tained a significantly higher proportion of antiP32 IgG and
level of 0.42 (values <0.42 representing TB) yielded an IgA antibodies than samples obtained from non-tubercu-
improved sensitivity of 100 percent and specificity of lous pleural effusions.
97–98.6 percent (Table 27.3). A monoclonal antibody against soluble phase-terminal
The levels of ADA in the pleural fluid decline with time, complement complex (SC5b-9) has been analyzed in
at ambient temperatures. However, the addition of stabi- pleural fluid to differentiate tuberculous effusions from
lizing agents (glycerol and ethylene glycol each at 5 percent malignant effusions and effusions of other causes. In one
concentration) to pleural fluid specimens allows the trans- study, 26 patients with TB pleural effusions had a signifi-
port of the specimens to distant laboratories at ambient cantly higher SC5b-9 level than did the 38 cases of pleural
temperatures without a decline in the ADA levels.93 effusions of other etiologies. With a cut-off level of
Pleural fluid levels of IFN-g are increased in pleural TB 2.0 mg/L, the specificity and sensitivity of SC5b-9 for
compared with other exudative effusions. When a cut-off pleural TB was 74 and 100 percent, respectively.99
value of 140 pg/mL or 3.9 U/mL is used, a sensitivity of In tuberculous pleuritis, high pleural fluid levels of
74–100 percent and a specificity of 91–100 percent have IL-2, IL-6 and IL-12 and TNF-a have been described (see
been reported.80,94 False positives have been described in also Chapter 8, Immunology). The levels of IL-2 and IL-6
parapneumonic effusions, lymphoproliferative disorders have been found up to 15 times higher in pleural fluid than
and other malignancies. The cost and speed of results have in serum.100 IL-12 has also been reported to be elevated in
limited its routine use. pleural TB and its low levels have been related to poor
Table 27.3 Adenosine deaminase (ADA) and ADA1/ADApleural activity in the diagnosis of tuberculous pleuritis
prognoses.101 However, these cytokines have never been Treatment of patients with tuberculosis should gener-
shown to have significant diagnostic utility, because their ally be monitored by official public health centers to
levels overlap with those in exudates of other causes. ensure: administration of correct therapy, which helps
In summary, non-invasive diagnosis using a ADA cut- prevent the emergence of resistant mycobacteria; coordi-
off level of 40 IU with lymphocyte/neutrophil ratio above nation of contact tracing; monitoring of the pattern of
0.75 can be used to diagnose pleural tuberculosis in resistances to drugs in the community; adequate education
patients <35 years old from countries with a high preva- of patients; and identification of possible outbreaks.105
lence of tuberculosis and a low level of mycobacterial Therapy of pleural TB, as for pulmonary TB, is based on
resistance102,103 (Figure 27.1, diagnostic algorithm). In the use of therapeutic regimens with combination of drugs
other situations, closed pleural biopsy is required, though directed at reducing the population of mycobacteria
more invasive procedures such as thoracoscopy are usually without creating resistance and at sterilizing the lesions
not necessary.104 during the prolonged treatment phase. Duration of treat-
ment, number of drugs, use of steroids, therapeutic
drainage and intrapleural urokinase are aspects of possible
controversy.
TREATMENT AND MANAGEMENT One of the most widely recommended therapeutic reg-
imens is a 6-month treatment course. An initial phase with
In most cases, pleural TB without treatment resolves spon- isoniazid (5–10 mg/kg.day up to 300 mg/day), rifampin
taneously in 2–4 months. However, if no treatment is (600 mg/day) and pyrazinamide (1.5–2 g/day) adminis-
received, 65 percent (92 out of 114) of patients end up tered daily for 2 months is followed by isoniazid and
developing pulmonary tuberculosis in the following 5 rifampin daily during the following 4 months.106 If the
years.45 This is the rationale of treatment for tuberculous patient lives in a region with more than 4 percent primary
pleuritis. resistance to isoniazid, comes from an area where high
Pleural effusion
Tuberculosis suspected
(symptoms, thoracic X-ray
and other analyses)
Tuberculosis suspected by
Tuberculosis clinically
biochemical parameters
suspected
from pleural fluid
Negative Mycobacterium
ADA 35 IU
in pleural fluid or tissue
Lymph/neutrophil 0.75
and no presence of
age 35 years
necrotizing granulomas
Mycobacterium
in pleural fluid or tissue
Tuberculosis or
necrotizing granulomas
levels of resistant microorganisms are reported, has than 10 mm. The relationship between residual pleural
received anti-TB drugs previously or has been exposed to thickening and different variables (low concentration of
patients with known multi-resistant TB, ethambutol (15 glucose in fluid and low pH level, high lysozyme level and
mg/kg.day) should be added until the results of the sensi- TNF-a level, presence of mycobacteria in pleural fluid or
tivity studies become known.107 The scarce bacillary pres- biopsy, high effusion level, duration of symptoms, low
ence in pleural fluid and biopsy suggests that a less intense hemoglobin level and loss of weight)121,123 have been weak.
regimen may be adequate. The use of only two drugs, iso- Residual pleural thickening has been reported to be less in
niazid and rifampin, daily for 6 months has been shown to HIV-positive than in HIV-negative patients.124 However,
be effective in immunocompetent patients as well as HIV- functional sequelae related to the pleural thickening are
positive patients, without any evidence of relapse infrequent.125
(0 percent) during a follow-up for more than 3 years.108,109 Tuberculous empyema (pus in pleural fluid) is an
There is complete resolution after 6 months in 75 percent unusual complication, normally related to the presence of
of cases and in practically all cases 14 months after the ini- bronchopleural fistulae.126 The response to medical treat-
tiation of treatment.110 During the initial phase of treat- ment is limited by loculated pleural effusion and pleural
ment, some patients may show an increase of pleural thickening127 and in these cases drainage and decortication
effusion, and this does not necessarily indicate a therapeu- are should be considered.128,129 However, good results have
tic failure.111 been reported with repeated thoracentesis and medical
Generally, the same treatment regime is recommended treatment for 24 months.130
for HIV-positive patients. However, treatment should be Isolated cases of empyema necessitatis and costal
prolonged if the clinical or bacteriological response is slow involvement by proximity due to M. tuberculosis have been
or less than optimal.112,113 described.131,132 On rare occasions, fistulas and abscesses of
Systemic corticosteroids have been proposed in the the thoracic wall in the path of the pleural biopsy have
treatment of pleural tuberculosis because pleural tubercu- been reported.133 In our experience this happened once in
losis is associated with inflammation and fibrosis, and 106 biopsy cases.
steroids can reduce the degree of pleural inflammation.31 A
recent review by Matchaba et al.114 found that only 3 out of
10 studies met randomization, double blind, placebo-con- FUTURE DIRECTIONS OF DEVELOPMENT
trolled criteria.7,115,116 They analyzed the effects of corti-
coids in terms of symptoms, residual pleural fluid and 1. Reinforce mechanisms for prevention: education, vac-
residual pachypleuritis and concluded that there is insuffi- cinations, chemoprophylaxis and adequate treatments,
cient evidence to define whether steroids are useful in available in any geographical area of the world.
pleural TB. 2. Facilitate diagnosis:
Reasonable management of pleural TB could include (i) Obviate the necessity for pleural biopsy for diagno-
performing a therapeutic thoracentesis at the same time as sis through the search for parameters in the pleural
the diagnostic thoracentesis is carried out, especially in fluid (ADA, ADA isoenzymes, IFN-g, PCR). This
large-volume effusions, with the intention of draining all would allow for easier diagnoses in regions with
the fluid possible and initiating the antituberculous treat- limited health care resources. Improving the diag-
ment. Routine drainage with a pigtail catheter does not nostic rate may result in provision of treatment to
reduce the level of residual pleural thickening.117 Some populations that are currently not diagnosed,
authors have used intrapleural urokinase in loculated TB including children who are may not be able to
pleural effusions,118 but there are no randomized studies cooperate with pleural biopsy, and those who live
evaluating this treatment and it is not recommended. in areas where there are no trained personel to
perform pleural biopsy.
(ii) Improve, standardize and make reproducible the
COMPLICATIONS diagnostic methods that are offering better per-
formances in various geographic areas with differ-
Residual pleural thickening, empyema and thoracic wall ent prevalences of the disease.
infection are the most frequently described complications. 3. Develop therapeutic aspects:
Others, such as non-Hodgkin’s lymphomas, sarcomas and (i) Given that pleural tuberculosis has a low mycobac-
other malignancies can present themselves after chronic terial population, are three anti-TB drugs needed?
inflammatory processes119,120 but these complications are Can the treatment be shortened?
very unusual. (ii) In the adjuvant treatment, are corticoids useful
Residual pleural thickening is the most frequent com- (when, how, why, for what)? Is pleural drainage
plication. Its incidence varies according to the time of eval- necessary? Could intrapleural fibrinolytic drugs be
uation and the degree of thickness. At the conclusion of useful if there is an increase in procoagulant activ-
treatment, 43–50 percent of patients have pleural thicken- ity? Could treatments with IFN-g or IL-12 be
ing of more than 2 mm121,122 and 20 percent have more useful?
References 375
15. Liam CK, Lim KH, Wong CM. Tuberculosis pleurisy as a space: role of intercellular adhesion molecule-1 in tuberculous
manifestation of primary and reactivation disease in a region with pleurisy. J Infect Dis 1999; 180: 1616–23.
a high prevalence of tuberculosis. Int J Tuberc Lung Dis 1999; 3: ●36. Pace E, Gjomarkaj M, Melis M, et al. Interleukin-8 induces
816–22. lymphocyte chemotaxis into the pleural space: role of pleural
16. Kramer F, Modilevsky T, Waliany AR, Leedom JM, Barnes PF. macrophages. Am J Respir Crit Care Med 1999; 159: 1592–9.
Delayed diagnosis of tuberculosis in patients with human 37. Park JS, Kim YS, Jee YK, et al. Interleukin-8 production in
immunodeficiency virus infection. Am J Med 1990; 89: 451–6. tuberculosis pleurisy: role of mesothelial cells stimulated by
17. Davies PDO. The association between HIV and tuberculosis in the cytokine network involving tumor necrosis factor-alpha and
developing world. In: Davies PDO (ed.). Clinical tuberculosis. interleukine-1beta. Scand J Immunol 2003; 57: 463–9.
London: Arnold, 1995: 253. 38. Light RW. Tuberculous pleural effusions. In: Light RW (ed.). Pleural
◆18. Mlika-Cabanne N, Brauner M, Mugusi F, et al. Radiographic diseases, 3rd edn. Baltimore: Williams and Wilkins, 1995: 154–66.
abnormalities in tuberculosis and risk of coexisting human 39. Sahn SA. State of the art: The pleura. Am Rev Respir Med 1988;
deficiency virus infection. Results from Dar-es-Salaam, Tanzania, 138: 184–234.
and scoring system. Am Rev Respir Crit Care Med 1995; 152: 40. Perez-Rodriguez E, Jimenez Castro D, Gaudo Navarro J. Valoracion
786–93. diagnostica del derrame pleural. Rev Clin Esp 2000; 200: 74–7.
◆19. Pozniak AL, MacLeod GA, Ndlovu D, et al. Clinical and radiographic 41. Antony VB, Sahn SA, Antony AC, Repine JE. Bacillus
features of tuberculosis associated with human immunodeficiency Calmette–Guerin-stimulated neutrophils release chemotaxins for
virus in Zimbabwe. Am J Respir Crit Care Med 1995; 152: monocytes in rabbit pleural spaces and in vitro. J Clin Invest 1985;
1558–61. 76: 1514–21.
20. Jones BE, Young SM, Antoniskis D, et al. Relationship of the ◆42. Kitinya JN, Richter C, Perenboom R, et al. Influence of HIV status
manifestations of tuberculosis to CD4 cell counts in patients with on pathological changes in tuberculous pleuritis. Tuberc Lung Dis
human immunodeficiency virus infection. Am J Respir Crit Care 1994; 75: 195–8.
Med 1993; 148: 1292–7. 43. Law KF, Jagirdar J, Weiden MD, et al. Tuberculosis in HIV-positive
◆21. Valdes L, Alvarez D, San José E, et al. Value of adenosine patients: cellular response and immune activation in the lung. Am
deaminase in the diagnosis of tuberculous pleural effusions in J Respir Crit Care Med 1996; 153: 1377–84.
young patients in a region of high prevalence of tuberculosis. ◆44. Riantawan P, Chaowalit P, Wongsangiem M, Rojanaraweewong P.
Thorax 1995; 50: 600–603. Diagnostic value of pleural fluid adenosine deaminase in
22. Pérez-Rodriguez E, Ortiz de Saracho J, Sanchez JJ, et al. Is tuberculous pleuritis with reference to HIV coinfection and a
adenosine deaminase in pleural effusion an expression of Bayesian analysis. Chest 1999; 116: 97–103.
lymphocytic or monocytic activity? Eur Respir J 1995; 8: 309s. 45. Sibley JC. A study of 200 cases of tuberculous pleurisy with
23. Stead W, Eichenholz A, Stauss HK. Operative and pathologic effusion. Am Rev Tuberc 1950; 62: 314–23.
findings in twenty-four patients with syndrome of idiopathic 46. Roper W, Waring JJ. Primary serofbrinous pleural effusion in
pleurisy with effusion, presumably tuberculous. Am Rev Tuberc military personnel. Am Rev Tuberc 1955; 71: 616–34.
1955; 71: 473–502. ◆47. Aktogu S, Yorgancioglu A, Cirak K, Kose T, Dereli SM. Clinical
24. Stevenson F. The natural history of pleural effusion and spectrum of pulmonary and pleural tuberculosis: a report of 5480
orthopaedic tuberculosis. J Bone Joint Surg Br 1955; 37: 80–91. cases. Eur Respir J 1996; 9: 2031–5.
25. Torgersen J, Dorman SE, Baruch N, Hooper N, Cronin W. Molecular 48. Berger HW, Mejia E. Tuberculous pleurisy. Chest 1973; 63: 88–92.
epidemiology of pleural and other extrapulmonary tuberculosis: a 49. Chan C, Arnold M, Chan CY, Mak T, et al. Clinical and pathological
Maryland state review. Clin Infect Dis 2006; 42: 1375–82. features of tuberculous pleural effusion and its long-term
26. Moudgil H, Sridhar G, Leitch AG. Reactivation disease: the consequences. Respiration 1991; 58: 171–5.
commonest form of tuberculosis pleural effusion in 50. Barnes PF, Bloch AB, Davidson PT, Snider DE. Tuberculosis in
Edinburgh,1980–1991. Respir Med 1994; 88: 301–4. patients with human immunodeficiency virus infection. N Engl J
27. Widstrom O, Nilsson BS. Pleurisy induced by intrapleural BCG in Med 1991; 324: 1644–50.
immunized guinea pigs. Eur J Respir Dis 1982; 63: 425–34. ◆51. Richter C, Perenboom R, Mtoni I, et al. Clinical features of HIV-
28. Okamoto M, Hasegawa Y, Hara T, et al. T-helper type 1/T-helper seropositive and HIV-seronegative patients with tuberculous
type 2 balance in malignant pleural effusions compared to pleural effusion in Dar es Salaam. Tanzania. Chest 1994; 106:
tuberculosis pleural effusions. Chest 2005; 128: 4030-5. 1471–5.
29. Rossi GA, Balbi B, Manca F. Tuberculous pleural effusions: ◆52. Arriero JM, Romero S, Hernandez L, et al. Tuberculous pleurisy
evidence for a selective presence of PPD-specific T-lymphocytes at with or without radiographic evidence of pulmonary disease. Is
site of inflammation in the early phase of the infection. Am Rev there any difference? Int J Tuberc Lung Dis 1998; 2: 513–17.
Respir Dis 1987; 136: 575–9. 53. Light RW, Erozan YS, Ball WC. Cells in pleural fluid. Their value in
30. Fujiwara H, Okuda Y, Fukakawa T, Tsuyuguchi I. In vitro tuberculin differential diagnosis. Arch Intern Med 1973; 132: 854–60.
reactivity of lymphocytes from patients with tuberculous pleurisy. ◆54. Diaz Nuevo G, Jimenez Castro D, Perez-Rodriguez E, Prieto Yaya E,
Infect Immun 1982; 35: 402–9. Sueiro Bendito A. Pleural eosinophilia: its diagnostic and
31. Baganha FM, Pego A, Lima AA, Gaspar EV, Cordeiro RA. Serum and prognostic significance. Rev Clin Esp 1999; 199: 573–5.
pleural adenosine deaminase: correlation with lymphocytic 55. Spriggs AI, Boddington MM. Absence of mesothelial cells from
populations. Chest 1990; 97: 605–10. tuberculous pleural effusions. Thorax 1960; 15: 169–71.
◆32. Pérez-Rodriguez E, Jimenez Castro D. The use of adenosine 56. Kim HJ, Lee HJ, Kwon SY, et al. The prevalence of pulmonary
deaminase isoenzymes in the diagnosis of tuberculous pleuritis. parenchymal tuberculosis in patients with tuberculous
Curr Opin Pulm Med 2000; 6: 259–66. pleuritis.Chest 2006; 129: 1253–8.
33. Mohamed KA, Nasreen N, Ward MJ, et al. Mycobacterium 57. Porcel JM, Vives M. Etiology and pleural fluid characteristics of
mediated chemokine expression in pleural mesothelial: role of C–C large and massive effusions. Chest 2003; 124: 978–83.
chemokines in tuberculosis. J Infect Dis 1998; 178: 1450–8. 58. Gil Montoro D, Perez-Rodriguez E, de Francisco G, et al. Derrames
●34 Mohamed KA, Nasreen N, Ward MJ, et al. Helper T cell type 1 and pleurales masivos: incidencia y significado en 1098 casos. Rev
2 cytokines regulate C–C chemokine expression in mouse pleural Patol Respir 2001; 4: 67–70.
mesothelial cells. Am J Respir Crit Care Med 1999; 159: 1653–9. 59. Joseph J, Strange C, Sahn SA. Pleural effusions in hospitalized
●35. Nasreen N, Mohammed KA, Ward MJ, Antony VB. Mycobacterium- patients with AIDS. Ann Intern Med 1993; 118: 856–9.
induced transmesothelial migration of monocytes into pleural 60. Navio P, Jimenez D, Pérez-Rodriguez E, et al. Atypical locations of
References 377
pulmonary tuberculosis and the influence of the roentgenographic fluids: a useful diagnostic tool in tuberculosis. Clin Biochem 1989;
patterns and sample type in its diagnosis: Respiration 1997; 64: 22: 141–8.
296–9. 84. Ocaña I, Martinez Vazquez JM, Segura RM, Fernandez de Sevilla T,
61. Farman DP, Speir WA Jr. Initial roentgenographic manifestations Capdevila JA. Adenosine deaminase in pleural fluid: test for
of bacteriologically proven Mycobacterium tuberculosis. Typical or diagnosis of tuberculous pleural effusion. Chest 1983; 84: 51–3.
atypical? Chest 1986; 89: 75–7. 85. Perez-Rodriguez E, Ortiz de Caracho J, Sanchez JJ, et al. Pleural
62. Jimenez Castro D, Díaz Nuevo G, Izquierdo Alonso JL, Pérez- adenosine deaminase (ADAp) and its iso-enzymes in tuberculous
Rodriguez E. Valor diagnóstico de presunción en los derrames effusions. Eur Respir J 1995; 8: 554s.
pleurales. Rev Patol Respir 2001; 4: 5–8. 86. De Oliveira HG, Rossatto ER, Prolla JC. Pleural fluid adenosine
◆63. Ferrer J. Pleural tuberculosis. Eur Respir J 1997; 10: 942–7. deaminase and lymphocyte proportion: clinical usefulness in the
◆64. Morehead RS. Tuberculosis of the pleura. South Med J 1998; 91: diagnosis of tuberculosis. Cytopathology 1994; 5: 27–32.
630–6. ●87. Blake J. Berman P. The use of adenosine deaminase assays in the
65. Conde MB, Loivos AC, Rezende VM, et al. Yield of sputum diagnosis of tuberculosis. S Afr Med J 1982; 62: 19–21.
induction in the diagnosis of pleural tuberculosis. Am J Respir Crit 88. Maritz FJ, Malan C, Le Roux I. Adenosine deaminase estimations in
Care Med 2003; 167: 723–5. the differentiation of pleural effusions. S Afr Med J 1982; 62:
66. Antoniskis D, Amin K, Barnes P. Pleuritis as a manifestation of 556–8.
reactivation tuberculosis. Am J Med 1990; 89: 447–50. 89. Ferrer JS, Muñoz XG, Orriols RM, Light RW, Morell FB. Evolution of
67. Maertens G, Bateman ED. Tuberculous pleural effusions: Increased idiopathic pleural effusion: a long-term follow-up study. Chest
culture yield with beside inoculation of pleural fluid and poor 1996; 109: 1508–13.
diagnostic value of adenosine deaminase. Thorax 1991; 46: 96–9. ●90. Gakis C. Adenosine deaminase (ADA) isoenzymes ADA1 and ADA2:
68. Kirsch CM, Kroe DM, Azzi RL, et al. The optimal number of pleural diagnostic and biological role. Eur Respir J 1996; 9: 632–3.
biopsy specimens for a diagnosis of tuberculous pleurisy. Chest ◆91. Perez-Rodriguez E, Perez-Walton IJ, Sanchez Hernandez JJ, et al.
1997; 112: 702–6. ADA1/ADAp ratio in pleural tuberculosis: an excellent diagnostic
◆69. Valdes L, Alvarez D, San José E, et al. Tuberculous pleurisy: a study parameter in pleural fluid. Respir Med 1999; 93: 816–21.
of 254 patients. Arch Intern Med 1998; 158: 2017–21. 92. Inase N, Tominaga S, Yasui M, et al. Adenosine deaminase 2 in the
70. Jimenez D, Perez-Rodriguez E, Diaz G, Fogue L, Light RW. diagnosis of tuberculous pleuritis. Kekkaku 2005; 80: 731–4.
Determining the optimal number of specimens to obtain with 93. Miller KD, Barnette R, Light RW. Stability of adenosine deaminase
needle biopsy of the pleura. Respir Med 2002; 96: 14–17. during transport. Chest 2004; 126: 1933–7.
71. Maartens G, Bateman ED. Tuberculous pleural effusions: Increased 94. Villena V, Lopez-Encuentra A, Echave-Sustaeta J, et al. Interferon-
culture yield with beside inoculation of pleural fluid and poor gamma in 388 immunocompromised and immunocompetent
diagnostic value of adenosine deaminase. Thorax 1991; 46: 96–9. patients for diagnosing pleural tuberculosis. Eur Respir J 1996; 9:
72. Schluger NW, Rom WN. Current approaches to the diagnosis of 2635–9.
active pulmonary tuberculosis. Am J Respir Crit Care Med 1994; 95. Moriwaki Y, Kohjiro N, Itoh M, et al. Discrimination of tuberculous
149: 264–7. from carcinomatous pleural effusion by biochemical markers:
73. Schirm J, Oostendorp LA, Mulder JG. Comparison of Amplicor, in- adenosine deaminase, lysozyme, fibronectin and
house PCR, and conventional culture for detection of carcinoembryobnic antigen. Jpn J Med 1989; 28: 478–84.
Mycobacterium tuberculosis in clinical samples. J Clin Microbiol 96. Yew WW, Chan CY, Kwan S, Cheung SW, French GL. Diagnosis of
1995; 33: 3221–4. tuberculosis pleural effusions by the detection of tubercolostearic
74. Bennedsen J, Thomsen VO, Pfyffer GE, et al. Utility of PCR in acid in pleural aspirates. Chest 1991; 100: 1261–3.
diagnosing pulmonary tuberculosis. J Clin Microbiol 1996; 34: 97. Muranishi H, Nakashima M, Hirano H, et al. Simultaneous
1407–11. measurements of adenosine deaminase activity and
75. Nagesh BS, Sehgal S, Jindal SK, Arora SK. Evaluation of polymerase tuberculostearic acid in pleural effusions for the diagnosis of
chain reaction for detection of Mycobacterium tuberculosis in tuberculous pleuritis. Intern Med 1992; 31: 752–5.
pleural fluid. Chest 2001; 119: 1737–41. 98. Van Vooren JP, Farber CM, De Bruyn J, Yernault JC.
◆76. Querol JM, Minguez J, Garcia Sanchez E, et al. Rapid diagnosis of Antimycobacterial antibodies in pleural effusion. Chest 1990; 97:
pleural tuberculosis by polymerase chain reaction. Am J Respir Crit 88–90.
Care Med 1995; 152: 1977–81. 99. Hara N, Abe M, Inuzuka S, Kawarada Y, Shigematsu N. Pleural
77. Villena V, Rebollo MJ, Aguado JM, et al. Polymerase chain reaction SC5b-9 in differential diagnosis of tuberculous, malignant and
for the diagnosis of pleural tuberculosis in immunocompromised other effusions. Chest 1992; 102: 1060–64.
and immunocompetent patients. Clin Infect Dis 1998; 26: 212–14. 100. Barnes PF, Lu S, Abrams JS, et al. Cytokine production at the site
●78. Piras MA, Gakis G, Budroni M, Andreoni G. Adenosine deaminase of disease in human tuberculosis. Infect Immun 1993; 61: 3482–9.
activity in pleural effusions: an aid to differential diagnosis. Br 101. Zhang M, Gately MK, Wang E, et al. Interleukin 12 at the site of
Med J 1978; 2: 1751–2. disease in tuberculosis. J Clin Invest 1994; 93: 1733–9.
◆79. Burgess LJ, Maritz FJ, Le Roux I, Talijaard JJ. Combined use of 102. Laniado-Laborin R. Adenosine deaminase in the diagnosis of
pleural adenosine deaminase with lymphocyte/neutrophil ratio. tuberculosis pleural effusion: is it really an ideal test? A word of
Increased specificity for the diagnosis of tuberculous pleuritis. caution. Chest 2005; 127: 417–18.
Chest 1995; 109: 414–19. 103. Diacon AH, Van de Wal BW, Wyser C, et al. Diagnostic tools in
◆80. Valdes L, San José E, Alvarez D, et al. Diagnosis of tuberculous tuberculosis pleurisy: a direct comparative study. Eur Respir J
pleurisy using the biologic parameters adenosine deaminase, 2003; 22: 589–91.
lysozyme and interferon gamma. Chest 1993; 103: 458–65. 104. Sakuraba M, Masuda K, Hebisawa A, Sagara Y, Komatsu H.
◆81. Valdes L, San José E, Alvarez D, Valle JM. Adenosine deaminase Thoracoscopic pleural biopsy for tuberculosis pleurisy under local
(ADA) isoenzyme analysis in pleural effusion: diagnostic role, anesthesia. Ann Thorac Cardiovasc Surg 2006; 12: 245–8
relevance to the origin of increased ADA in tuberculous pleurisy. 105. Small PM, Fujiwara PI. Medical progress: Management of
Eur Respir J 1996; 9: 747–75. tuberculosis in the United States. New Engl J Med 2001; 345:
82. Bañales JL, Pineda PR, Fitzgerald JM, et al. Adenosine deaminase 189–200.
in the diagnosis of tuberculosis pleural effusions. A report of 218 106. Bass JB, Farer LS, Hopewell PC, et al. Treatment of tuberculosis
patients and review of the literature. Chest 1991; 99: 355–7. and tuberculosis infection in adults and children. Am J Respir Crit
83. Segura RM, Pascual C, Ocaña I, et al. Adenosine deaminase in body Care Med 1994; 149: 1359–74.
378 Effusions from infections: tuberculosis
107. Joint Tuberculosis Committee of the British Thoracic Society. report of an unusual case. J Thorac Cardiovasc Surg 1994: 108:
Chemotherapy and management of tuberculosis in the United 395–6.
Kingdom: recommendations 1998. Thorax 1998; 53: 536–48. ◆121. de Pablo A, Villena V, Echave Sustaeta J, Encuentra A. Are pleural
◆108. Dutt AK, Moers D, Stead WW. Tuberculous pleural effusion: 6- fluid parameters related to the development of residual pleural
month therapy with isoniazid and rifampin. Am Rev Respir Dis thickening in tuberculosis? Chest 1997; 112: 1293–7.
1992; 145: 1429–32. 122. Barbas CS, Cukier A, de Varvalho CR, Barbas-Filho JV, Light RW.
◆109. Cañete C, Galarza I, Granados A, et al. Tuberculous pleural The relationship between pleural fluid findings and the
effusion: experience with six months of treatment with isoniazide development of pleural thickening in patients with pleural
and rifampicin. Thorax 1994; 49: 1160–61. tuberculosis. Chest 1991; 100: 1264–7.
110. Ormerod LP, McCarthy OR, Rudd RM, Horsfield N. Short-course 123. Flores J, Pérez-Rodriguez E, Ferrando C, et al. Parámetros no
chemotherapy for tuberculous pleural effusion and culture- terapéuticos que definen el pronóstico de la tuberculosis pleural.
negative pulmonary tuberculosis. Tuberc Lung Dis 1995; 76: 25–7. Arch Bronconeumol 1993; 29: 104–105.
111. Al-Majed SA. Study of paradoxical response to chemotherapy in 124. Lawn SD, Evans AJ, Sedgwick PM, Acheampong JW. Pulmonary
tuberculous pleural effusion. Respir Med 1996; 90: 211–14. tuberculosis: radiological features in west Africans coinfected with
112. Prevention and treatment of tuberculosis among patients infected HIV. Br J Radiol 1999; 72: 339–44.
with human immunodeficiency virus: principles of therapy and 125. Candela A, Andujar J, Hernandez L, et al. Functional sequelae of
revised recommendations. MMWR Morb Mortal Wkly Rep 1998; tuberculous pleurisy in patients correctly treated. Chest 2003;
47: 1–58. 123: 1996–2000.
113. Zumla A, Malon P, Henderson J, Grange JM. Impact of HIV 126. Jenssen AO. Chronic calcified pleural empyema. Scand J Respir Dis
infection on tuberculosis. Postgrad Med J 2000; 76: 259–68. 1969; 50: 19–27.
◆114. Matchaba PT, Volmink J. Steroids for treating tuberculous pleurisy. 127. Elliot AM, Berning SE, Iseman M, Peloquin CA. Failure of drug
Cochrane Database Syst Rev 2000; 2: CD001876. penetration and acquisition of drug resistance in chronic
115. Lee CH, Wang WJ, Lan RS, et al. Corticosteroids in the treatment tuberculous empyema. Tuberc Lung Dis 1995; 76: 463–7.
of tuberculous pleurisy: a double-blind, placebo-controlled, 128. Ali S, Siddiqui A, McLaughlin J. Open drainage of massive
randomized study. Chest 1988; 94: 1256–9. tuberculous empyema with progressive reexpansion of the lung:
116. Galarza I, Cañete C, Granados A, Estopa R, Manresa F. Randomised an old concept revisited. Ann Thorac Surg 1996; 62: 218–24.
trial of corticosteroids in the treatment of tuberculous pleurisy. ◆129. Mushegera CK, Mbuyi Muamba JM, Kabemba MJ. Indications and
Thorax 1995; 50: 1305–307. results of pleuropulmonary decortications in the university
117. Lai YF, Chao TY, Wang YH, Lin AS. Pigtail drainage in the hospital of Kinshasa. Acta Chir Belg 1996; 96: 217–22.
treatment of tuberculous pleural effusions: a randomized study. 130. Neihart RE, Hof DG. Successful nonsurgical treatment of
Thorax 2003; 58: 149–51. tuberculous empyema in an irreducible pleural space. Chest 1985;
118. Cases Viedma E, Lorenzo Dus MJ, Gonzalez-Molina A, Sanchis 88: 792–4.
Aldas JL. A study of loculated tuberculous pleural effusions treated 131. Peterson MW, Austin JH, Yip CK, McManus RP, Jaretzki A. CT
with intrapleural urokinase. Respir Med 2005; 100: 2037–42. findings in transdiaphragmatic empyema necessitatis due to
119. Fujiwara T, Kasahara H, Tanohata K, Nagase M. Fast spin-echo MR tuberculosis. J Comput Assist Tomogr 1987; 11: 704–6.
imaging of non-Hodgkin lymphoma arising from chronic 132. Ip M, Chen NK, So SY, Chiu SW, Lam WK. Unusual rib destruction
tuberculous empyema. J Thorac Imaging 1995; 10: 82–4. in pleuropulmonary tuberculosis. Chest 1989; 95: 242–4.
120. Takanami I, Imamura T, Morota N, Kodaira S. Malignant fibrous 133. Guest JL Jr, Anderson JN, Simmons EM Jr. Dumbbell
histiocytoma of the chest wall developing after granulomatous abscess of the chest wall following needle biopsy
pleuropneumonectomy performed for tuberculous pyothorax: of the pleura. South Med J 1976; 69: 1513–15.
28
Effusions from infections: atypical infections
percent).7,17,20 The pleural fluid is an exudate with a high The clinical picture is similar to that observed in tuber-
level of protein with a mix of lymphocytes and neu- culosis or fungal infections. The chest radiograph shows an
trophils. A cytological examination showing the character- infiltrative consolidation with multiple cavities associated
istic yeast forms is considered diagnostic. However, the to pleural thickening or effusion. Nowadays destruction of
usual method for diagnosis is culture, which takes 2–6 ribs or vertebrae is uncommon and sinus tracts of the chest
weeks. The accuracy of the pleural biopsy is around 59 wall are rarely seen.24,25
percent.1,20 The pleural fluid is exudative with a predominance of
The South American blastomycosis, caused by lymphocytes or an empyema with a predominance of
Paracoccidioides brasiliensis rarely causes pleural effusion. polymorphonuclear leukocytes. Sulfur granules in the
Anecdotal cases have been reported, the pleural effusions pleural fluid are strongly suggestive of actinomycosis
may be unilateral or bilateral with a small amount of fluid.21 (Figure 28.5). The granules are yellow-white on inspection
The treatment includes antifungal drugs and trimetho- and appear as irregular masses of branched, Gram-positive
prim–sulfamethoxazole.20,21 filaments.1,25 The fluid is an important source for isolation
of the organisms; it should be Gram-stained and cultured
anaerobically. Video-assisted thoracic surgery (VATS)
Sporotrichosis seems to be the most effective diagnostic method.24
The local management of the pleural effusion includes
Sporotrichosis is caused by the fungus Sporothrix schenckii, tube thoracostomy if the fluid is an empyema; decortica-
which is found in decaying vegetation, sphagnum moss tion is sometimes necessary. The specific treatment con-
and soil. The usual mode of infection is by cutaneous inoc- sists of prolonged administration of high doses of
ulation of the organism. Infection can be related to antibiotics.1,24
zoonotic spread from infected cats or scratches from
digging animals.22
Extracutaneous manifestations are unusual. Pulmonary Nocardiosis
sporotrichosis has rarely been reported; when present, it
simulates tuberculosis.22 Morissey and Caso23 described a Nocardiosis is usually caused by Nocardia asteroides, a
case of a pleural exudate with fungus stains positive for Gram-positive aerobic bacteria. Other pathogenic human
S. schenkii. Examination of the pleura revealed non- species include Nocardia brasiliensis and Nocardia caviae.9
caseating granulomas. Cultures of fluid and pleural tissue Nocardia species are not human commensals; their
were negative for mycobacteria. presence in the upper respiratory tract should be consid-
Treatment includes local measures (hyperthermia), ered as evidence of infection. They produce an acute, sub-
solution of potassium iodide, azoles, polyenes and ally- acute or chronic pneumonia.9,26
lamines.22 The signs and symptoms of nocardiosis include
anorexia, weight loss, dyspnea, cough with purulent
sputum and occasionally hemoptysis. The presence of
BACTERIA pleuritic pain suggests pleural involvement. Patients with
pleural disease (25–50 percent of cases) usually have
Actinomyces or Nocardia parenchymal infiltrates and cavitation.9,27
Actinomycosis
The pleural fluid ranges from a serous exudate to frank exposition. Pleural effusion and pulmonary infiltrate are
pus. When Nocardia are not identified in pleural effusion, present in 10–36 percent of patients. The pleural and lung
bronchoscopy provides material for identification.28 The lesions are probably due to vasculitis with increased per-
organism is aerobic and easily cultured on blood agar. The meability of the capillaries. The characteristics of the
presence of N. asteroides in the sputum does not mean the pleural fluid have not been well-defined. The treatment of
presence of pulmonary disease; 45 percent of patients with choice is doxycycline.1,30
positive culture did not have radiographic abnormalities.
If nocardiosis is suspected clinically, transbronchial
biopsy, percutaneous lung aspiration, needle biopsy or Ehrlichiosis
open lung biopsy must be performed.27
Sulfonamides or trimethoprim–sulfamethoxazole are There are two human ehrlichial diseases: a monocytic
the drugs of choice. The antibiotics are used in conjunc- disease caused by Ehrlichia chaffeensis and a granulocytic
tion with appropriate surgical management of the pleural disease caused by Anaplasma phagocytophilum. The disease
effusion.9,26 is not uncommon.The symptoms begin 7 days after a tick
bite and are characterized by high fever, headache, myal-
gias, nausea, vomiting and anorexia. Monoclonal antibod-
Chlamydiae and Rickettsiae ies have been used to detect the organism.31
Pulmonary infiltrates and a pleural effusion develop in
These organisms are grouped together because of their approximately 50 percent of patients. The infiltrates prob-
small size, the association with eukaryotic cells and the ably represent non-cardiogenic pulmonary edema with a
similarity of identification techniques. pathogenesis similar to the hantavirus syndrome. The
The Chlamydiae are among the more common treatment of choice is tetracycline or doxycycline.1,31
pathogens. Although recently the Chlamydiae have been
related to multiple diseases, including respiratory infec-
tions, an extensive search in Medline did not reveal any Mycoplasma pneumoniae
report of pleural effusion.
The order Rickettsiales includes a very diverse group of The organism is a small bacterium derived from ancestral
microorganisms. Recent characterizations at the molecular anaerobic bacteria (clostridia). Bronchopneumonia
level demonstrate a relationship between the genera involving one or more lobes develops in 3–10 percent of
Coxiella and Rickettsia, excluded from this classification is infected patients.32 Pleural effusion, generally small, occurs
the genus Ehrlichlia.7 in 5–20 percent of patients; empyema is a rare complica-
tion.1,33 The pathogenesis of pleural effusion is unclear, but
Q FEVER it is speculated that the presence of M. pneumoniae DNA
may elicit stronger immunological reaction causing lung
Q fever is caused by the rickettsial agent Coxiella burnetti. damage, with interleukin (IL)-8 and IL-18 playing a role in
It is acquired by the inhalation of contaminated dust par- the reaction.1,33 The diagnosis is established by increasing
ticles or by drinking infected and unpasteurized milk.7 specific antibody titers or by isolating M. pneumoniae from
Half of patients do not exhibit any respiratory symptoms the pleural fluid. The detection of M. pneumoniae DNA in
and one-third have pleuritic chest pain.29 throat swab specimens through the application of poly-
Pleural effusion is usually small and found in 10–35 merase chain reaction (PCR) has been found to be a sensi-
percent of patients. The fluid is a mononuclear exudate, tive and specific diagnostic technique and the positive PCR
although on occasion there has been an eosinophilic effu- results using pleural fluid samples have been found to be
sion. The related increase of adenosine deaminase (ADA) strongly associated with the abnormalities observed on the
activity is probably due to the lymphocytic activation. The radiograms.34
diagnosis is established by demonstrating a positive C. bur- Macrolides are the drugs of choice. No specific treat-
netti complement fixation reaction. The treatment of ment is necessary for pleural effusion; however, a diagnos-
choice is tetracycline or its derivatives (minocycline and tic thoracentesis should exclude a complicated
doxycicline).1,29 parapneumonic effusion.1,33
tion, the trophozoites develop, colonize the intestine, dazole or albendazole should be considered as a supple-
migrate to the liver and give rise to liver abscess.1 A solitary ment.1,36,37
rupture into the pleural cavity producing a pleural effusion
is much less common than directly into the lungs. The
pleural fluid is described as ‘chocolate sauce’ or ‘anchovy Paragonimiasis
paste’.1,35
The patients complain of shortness of breath, cough Paragonimiasis is caused by numerous species of lung
and pleuritic chest pain with an abrupt worsening when a flukes, (genus Paragonimus). The pulmonary form is the
transdiaphragmatic rupture occurs. The chest radiograph most common presentation of Paragonimus westermani
shows a small to massive pleural effusion, occasionally and Paragonimus miyazakii. It is a food-borne parasitic
with a contralateral mediastinum shift, and elevation of disease common in southeast Asia.38
the hemidiaphragm and atelectasis at the base.1,35 Humans acquire paragonimiasis by eating under-
The diagnosis is confirmed by the pleural fluid appear- cooked crustaceans contaminated with metacercariae of
ance and by serological test for amebiasis. Amoebas are the worm. After ingestion, the metacercariae hatch in the
seen in the exudative fluid in less than 10 percent of duodenum and the larval fluke bore through the intestinal
patients.1,35 wall, enter the peritoneal cavity and travel through the
The treatment of choice is metronidazole associated diaphragm to reach the visceral pleura and lung. In the
with a therapeutic thoracocentesis or a tube thoracostomy. lung, the adult remains for years producing eggs which are
One-third of patients have a bacterial infection of the expectorated or swallowed and excreted in the stools. Once
pleural space, and should be treated with antibiotics.1 in water, the eggs turn into ciliated miracidia that infect
freshwater snails. Another larval form develops and is lib-
erated as cercariae that penetrate crayfish and crabs. Adult
Echinococcosis–hydatidosis lung flukes can live in the human host for as long as 20
years, highlighting the need for clinicians to be aware of
Human echinococcosis is caused by three species of the disease in areas where the disease is not endemic.1
Echinococcus: E. granulosus (producing cystic hydatic in 90 In pulmonary paragonimiasis, pleural disease occurs in
percent of cases), E. multilocularis and E. vogeli.7 When approximately 60 percent of patients. The pleural involve-
feces containing the parasite’s eggs are ingested by ment is represented in 60 percent by pleural effusion (80
humans, larvae emerge in the duodenum, enter the blood percent unilateral), in 30 percent by hydropneumothorax
and lodge in the liver or in the lung. Echinococcus has a (half are bilateral) and in 10 percent by pleural thickening.
worldwide distribution; however, it is seen more often in Pleural paragonimiasis should be suspected in oriental
most sheep- and cattle-raising areas.1 patients or in patients who have traveled to the Orient.1,39
The organs most commonly affected are the liver and The diagnosis is made by detecting eggs in sputum,
the lungs. Pleural involvement is rare, it may be secondary stool, bronchoscopic lavage or biopsy specimens, or by the
to hematogenous dissemination of the larvae, however, it presence of a positive anti-Paragonimus antibody test. The
usually follows the rupture of a pulmonary or hepatic cyst pleural fluid is an eosinophilic exudate with low pH
releasing its contents into the pleural cavity. Since the cyst (<7.10), low glucose (<10 mg/dL) and high LDH
is not fertile after rupture in about 90 percent of cases, (>1000 IU/L). The pleural fluid white-cell count is usually
pleural hydatidosis is rare, occurring in less than 10 less than 2000/mm3 with marked eosinophilia. The blood
percent of the episodes. When the cyst ruptures into the and sputum commonly show an elevated number of
pleural space, the patient becomes acutely ill, with chest eosinophils.1,39,40
pain, dyspnea and occasionally goes into shock. In about It has been demonstrated that the levels of immuno-
50 percent of patients, the rupture occurs simultaneously globulin (Ig)E and P. westermani-specific IgE and IgG
into the pleural space and into the tracheobronchial tree.36 immunoglobulins are elevated and higher in the pleural
In geographic areas where echinococcosis occurs fre- fluid than in the serum, suggesting that these antibodies
quently, the diagnosis is usually apparent from the clinical are produced in the pleural space. Recently, it was demon-
picture, radiological findings and results of serological and strated that interleukin (IL)-5 concentrations in pleural
skin tests. In non-endemic areas, where physicians are effusions are higher in paragonimiasis than in transudates,
unfamiliar with echinococcosis, the diagnosis is suspected empyema, lung cancer or tuberculosis and that these levels
only after the cytological examination of the pleural fluid are correlated with the percentage of eosinophils in
with demonstration of echinococcal scoleces. Eosinophils peripheal blood and pleural fluid.1,39,40
are frequently present in the pleural fluid, unless it The drug of choice to treat is praziquantel or bithionol.
becomes secondarily infected.1,37 The biochemical charac- Pulmonary disease is rarely fatal. Thoracotomy with
teristics of the pleural fluid have not been reported. decortication may be necessary when pleural surfaces are
In patients with a rupture of a hydatid cyst, a thoraco- abnormally thickened and penetration of the drugs into
tomy to remove the parasite is recommended. Medical the pleural space insufficient to eradicate the infec-
treatment with benzimidazole compounds such as meben- tion.1,39,40
Viruses 385
Dengue fever
Herpes virus
●24.
REFERENCES Kobashi Y, Yoshida K, Miyashita N, et al. Thoracic actinomycosis
with mainly pleural involvement. J Infect Chemother 2004; 10:
172–7.
● = Key primary paper ◆25. Fife TD, Finegold SM, Grennan T. Pericardial actinomycosis: case
◆ = Major review article report and review. Rev Infect Dis 1991; 13: 120–26.
26. Yoshida K, Bandoh S, Fujita J, et al. Pyothorax caused by Nocardia
1. Light RW. Pleural diseases, 5th edn. Baltimore: Lippincott Williams otitidiscaviarum in a patient with rheumatoid vasculitis. Intern
& Wilkins, 2007. Med 2004; 43: 615–19.
●2. Chen KY, Ko SC, Hsueh PR, et al. Pulmonary fungal infection. 27. Uttamchandani RB, Daikos GL, Reyes RR, et al. Nocardiosis in 30
Emphasis on microbiological spectra, patient outcome, and patients with advanced human immunodeficiency virus infection:
prognostic factors. Chest 2001; 120: 177–84. clinical features and outcome. Clin Infect Dis 1994; 18: 348–53.
3. Schacherer D, Mayer S, Borisch I, et al. Esophagobronchial and 28. Toukap NA, Hainaut P, Moreau M, et al. Nocardiosis: a rare cause
esophagomediastinal fistula, pleural and pericardial effusion due of pleuropulmonary disease in the immunocompromised host. Acta
to severe pseudodiverticulosis of the esophagus. Z Gastroenterol Clin Belg 1996; 51:161–5.
2006; 44: 491–5. 29. Okikamoto N, Asaoka N, Osaki K, et al. Clinical features of Q fever
4. Mora Duarte J, Betts R, Rotstein C, et al. Comparison of pneumonia. Respirology 2004; 9: 278–82.
caspofungin and amphotericin B for invasive candidiasis. N Engl J 30. Byrd RP Jr, Vasquez J, Roy TM. Respiratory manifestations of tick-
Med 2002; 347: 2020–29. borne diseases in the southeastern United States. South Med J
5. Wex P, Utta E, Drozdz W. Surgical treatment of pulmonary and 1997; 90: 1–4.
pleuro-pulmonary Aspergillus disease. Thorac Cardiovasc Surg 31. Scully RE, Mark Ej, Neely WF. Human granulocytic ehrlichiosis.
1993; 41: 64–70. Case records of the Massachusetts General Hospital. N Engl J Med
6. Soto-Hurtado EJ, Marín-Gámez E, Segura-Dominguez N, et al. 2001; 345: 1627–34.
Pleural aspergillosis with bronchopleurocutaneous fistula and ◆32. Waites KB, Talkington DF. Mycoplasma pneumonie and its role as
costal bone destruction: a case report. Lung 2005; 183: 417–23. a human pathogen. Clin Microbiol Rev 2004; 17: 697–728.
7. Murray PR. Manual of clinical microbiology, 6th edn. Washington, 33. Shuvy M, Rav-Acha M, Izhar U, et al. Massive empyema caused by
DC: ASM Press, 1995. Mycoplasma pneumoniae in an adult: a case report. BMC Infect
8. Regnard JF, Icard P, Nicolosi M, et al. Aspergilloma: A series of 89 Dis 2006; 6: 18.
surgical cases. Ann Thorac Surg 2000; 69: 898–903. 34. Narita M, Matsuzono Y, Itakura O, et al. Analysis of mycoplasmal
◆9. George RB, Penn RL, Kinasewitz GT. Mycobacterial, fungal, pleural effusion by the polymerase chain reaction. Arch Dis Child
actinomycotic, and nocardial infections of the pleura. Clin Chest 1998; 78: 67–9.
Med 1985; 6: 63–75. 35. Kumar R, Dasan B, Choudhury S, et al. Hemoptysis demonstrated
10. Kawabata T, Matsuyama W, Higashimoto I, et al. Pleural on Tc99m-sulfur colloid scanning. A rare complication of amoebic
cryptococcosis with idiopathic CD4 positive T-lymphocytopenia. liver abscess. Clin Imaging 2002; 26: 296–8.
Intern Med 2004; 43: 977–81. ◆36. Skerrett SJ, Plorde JJ. Parasitic infections of the pleura space.
◆11. Young EJ, Hirsh DD, Fainstein V, et al. Pleural effusions due to Semin Respir Med 1992; 13: 242–58.
Cryptococcus neoformans: a review of the literature and report of 37. Pfefferkorn U, Viehl CT, Barras JP. Rupture hydatid cyst in the right
two cases with Cryptococcal antigen determinations. Am Rev thorax: differential diagnosis to pleural empyema. Thorac
Respir Dis 1980; 121: 743–7. Cardiovasc Surg 2005; 53: 250–51.
◆12. Batra P. Pulmonary coccidioidomycosis. J Thorac Imaging 1992; 7: 38. Singh TN, Kananbala S, Devi KS. Pleuropulmonary paragonimiasis
29–38. mimicking pulmonary tuberculosis – a report of three cases. Indian
13. Osaki T, Morishita H, Maeda H, et al. Pulmonary J Med Microbiol 2005; 23: 131–4.
coccidioidomycosis that formed a fungus ball with 8-years ●39. Mukae H, Taniguchi H, Matsumoto N, et al. Clinicoradiologic
duration. Intern Med 2005; 44: 141–4. features of pleuropulmonary paragonimus westermani on Kyusyu
14. Chiller TM, Galgiani JN, Stevens DA. Coccidioidomycosis. Infect Dis Island, Japan. Chest 2001; 120: 514–20.
Clin North Am 2003; 17: 41–57. 40. Robertson KB, Janssen WJ, Saint S, et al. Clinical problem-solving
15. Mortara L, Bayer AS. Fever, cough, pleuritic chest pain and pleural – the missing piece. N Engl J Med 2006; 355: 1913–18.
fluid eosinophilia in a 30-year-old-man. Chest 1994; 105: 41. Lewis KL, Doherty DE, Ribes J, et al. Empyema caused by
918–9. trichomonas. Chest 2003; 123: 291–2.
16. Kapotsis GE, Daniil Z, Malagan K, et al. A young male with chest 42. Klion AD, Eisenstein EM, Smirniotopoulos TT, et al. Pulmonary
pain, cough and fever. Eur Respir J 2004; 24: 506–9. involvement in Loiasis. Am Rev Respir Dis 1992; 145: 961–3.
17. Fraser RS, Muller NL, Colman N, Pare PD. Histoplasmosis. In: ●43. Hong JY, Lee HJ, Piedra PA, et al. Lower respiratory tract infections
Diagnosis of diseases of the chest, 4th edn. Philadelphia: due to adenovirus in hospitalized Korean children: epidemiology,
W Saunders, 1999: 876–90. clinical features, and prognosis. Clin Infect Dis 2001; 32: 1423–9.
18. Kilburn CD, McKinsey S. Recurrent massive pleural effusion due to 44. Bustamante EA, Levy H, Simpson SQ. Pleural fluid characteristics
pleural, pericardial and epicardial fibrosis in histoplasmosis. Chest in Hantavirus pulmonary syndrome. Chest 1997; 112: 1133–6.
1991; 100: 1715–17. ●45. Fabbri M, Maslow MJ. Hantavirus pulmonary syndrome in the
19. Carter AB, Hunninghake GW. Massive pleural effusion in diffuse United States. Curr Infect Dis Rep 2001; 3: 258–65.
granulomatous disease. Chest 1997; 112: 284–8. 46. Delfs-Jegge S, Dalquen P, Hurwitz N. Cytomegalovirus-infected
20. Alvarez GG, Burns BF, Desjardins M, et al. Blastomycosis in a cells in a pleural effusion from an acquired immunodeficiency
young African man presenting with a pleural effusion. Can Respir J syndrome patient. Acta Cytol 1994; 38: 70–72.
2006; 13: 441–4. 47. Beers MF, Sohn M, Swartz M. Recurrent pneumothorax in AIDS
21. Pellegrino A, De Capriles CH, Magaldi S, et al. Case report: Severe patients with pneumocystis pneumonia: A clinicopathological
juvenile type paracoccidioiidomycosis with hepatitis C. Am J Trop report of three cases and review of the literature. Chest 1990; 98:
Med 2003; 68: 301–3. 266–70.
●22. Kauffman CA, Hajjeh R, Chapman R. Practice guidelines for the ●48. Nador RG, Cesarman E, Chadburn A, et al. Primary effusion
management of patients with sporotrichosis. Clin Infect Dis 2000; lymphoma: a distinct clinicopathologic entity associated with the
30: 684–7. Kaposi’s sarcoma-associated herpes virus. Blood 1996; 88:
23. Morissey R, Caso R. Pleural sporotrichosis. Chest 1983; 84: 507. 645–56.
388 Effusions from infections: atypical infections
49. Selimoglu MA, Ziraatci O, Tan H, et al. A rare complication of 51. Chen J, Konstantinopoulos PA, Satyal S, et al. Just another simple
hepatitis A: pleural effusion. J Emerg Med 2005; 28: 229–30. case of infectious mononucleosis? Lancet 2003; 361: 1182.
50. Cloney DL, Kugler JÁ, Donowitz LG, et al. Infectious mononucleosis 52. Avirutnan P, Malasit P, Seliger B. Dengue virus infection of human
with pleural effusion. South Med J 1988; 81: 1441–2. endothelial cells leads to chemokine production, complement
activation and apoptosis. J Immunol 1998; 161: 6338–46.
29
Effusions from lymphatic disruptions
GUNNAR HILLERDAL
Chylus or chyle is lymph coming mainly from the gas- Chylothorax is a rare disease, which has become more
trointestinal tract. Except for medium-chained triglyc- common with increasing activities of thoracic surgeons.
erides, most types of fat will be resorbed into the The incidence after chest surgery, mainly coronary artery
lymphatics of the intestines and delivered to the general surgery, is around 0.5 percent1–3 and is by far the most
circulation via the thoracic duct. This explains the peculiar common cause. The other causes of chylothorax, i.e. after
composition of chylus with high levels of triglycerides, lymphomas, traffic accidents, etc., are of a sporadic occur-
cholesterol and chylomicrons. The content varies with rence.
whether the person is fasting or has just enjoyed a meal Pseudochylothorax is an even rarer occurrence. The
rich in calories, i.e. fat. Eating will considerably increase most common causes, large fibrotic pleural peels after
the flow of chylus and its content of fat components. tuberculosis pleurisy or pneumothorax treatment, are now
Chylothorax is the occurrence of chylus in the pleural very rare because there are few such patients alive.
space and results from leakage of the thoracic duct due to
damage of the wall, either by trauma or blockage of the
duct with subsequent rupture. The diagnosis is made by Etiology of chylothorax
analysis of the pleural fluid.
Pseudochylothorax or cholesterol pleurisy is a pleural Chylothorax is caused by leakage of chyle from the tho-
fluid with a very high content of cholesterol and occurs racic duct. This leakage is secondary to trauma, weakening
when a fluid has been present for a long time in the pleural of the wall, or blockage of the duct (Table 29.1). The
space surrounded by a fibrotic pleura with poor vascular- thoracic duct is fairly weak and even an intense sudden
ization. There are no chylomicrons in the fluid and usually stretching of it can cause rupture, for example a forceful
the level of triglycerides is also low. The cholesterol has cough or emesis. It has also been described after the strains
been formed in the encapsulated fluid and has no connec- of childbirth both in mother and child, more usually in the
tion with lymphatic vessels or chylus. infant.19 It must be remembered that when a fairly trivial
Both conditions have a common characteristic: the damage such as a cough causes a chylothorax, some
pleural fluid is usually thick, opalescent, whitish or the underlying cause must always be suspected and appropri-
colour of café-au-lait or chocolate milk, due to the very ate investigations started. Similarly, when chylothorax
high fat content. Apart from that, and the high level of develops in a child, a malformation must be the first sus-
cholesterol, the two conditions have nothing in common. picion.
390 Effusions from lymphatic disruptions
Table 29.1 Causes of chylothorax usually an extremely rare complication of most of the dif-
Causes of chylothorax
ferent diseases listed in Table 29.1. For example, case
reports in the world literature of chylothorax as a compli-
cation of sarcoidosis, goiter or tuberculosis are easily
A. Traumatic: counted on one hand.
Non-iatrogenic: Congenital chylothorax is caused by malformation of
Any accident with damage or stretching of the chest wall or the thoracic duct more often than by the trauma of birth.20
thoracic spine
One differential diagnosis could be parenteral nutrition
Forceful cough or emesis
Childbirth
via a subclavian vein with penetration of the vessel and
Iatrogenic: leakage into the pleura of the parenteral fluid. Analysis of
Surgery of chest the pleural content and comparison with the intravenous
Head and neck surgery solution can give the diagnosis.21
Radiation (often late sequale)
Sclerotherapy of the esophagus4
B. Diseases: Anatomy of the thoracic duct and
Malignant: pathogenesis of chylothorax
Lymphomas5
Other malignancies6 The lymph vessels from the lower parts of the body
Benign tumors:
combine with those from the peritoneal cavity and come
Retrosternal goiter7
Diseases affecting the lymph vessels:
together behind the aorta below the diaphragm and form
Sarcoidosis8,9 the thoracic duct. Usually, there is a widening of the duct
Lymphangioleiomyomatosis where it starts, which has been termed the cisterna chyli.
Hemangiomatosis (Gorham’s syndrome)10 The duct passes through the diaphragm behind the aorta
Congenital and continues upwards on the right side of the vertebral
Filiariasis11 column between the azygos vein and the aorta. At the
Tuberculosis12 height of the third or fourth vertebra it turns to the left,
Amyloidosis crosses the midline and continues upwards behind the
Yellow nail syndrome esophagus, medially of and behind the subclavian artery. It
Diseases affecting venous pressure: then turns laterally again, often after making a small loop
Thrombosis of the superior vena cava or other central veins
up into the neck region, and finally empties into the left
Heart failure with increased venous pressure13–15
Transdiaphragmal movement of chylous ascites15–17
vena subclavia between the jugular and the vertebral veins.
C. Congenital18 This anatomy explains why chylothoraces most often
D. Idiopathic18 occur on the right side. The largest part of the duct is
within the right hemithorax and this is also where it is
most easily damaged due to stretching. When the leakage
from the duct occurs where it passes over the mid-line, a
Traumatic rupture of the duct can be an early or late bilateral chylothorax can occur.22,23 If the duct is damaged
complication of traffic accidents but the vast majority of at the level of the aorta, the chylothorax tends to appear on
cases are now seen after various types of surgery involving the left side.7,24,25 What has been described here is,
heart, lung, or the head and neck region. It can occur even however, only the most common anatomy. There can be
after thoracoscopic surgery. The incidence of chylothorax very large variations. Two or more branches of the duct
after chest surgery of various kinds is, as mentioned, can be seen. The most common variation of this is one
around 0.5 percent.1–3 The most common cause in child- main duct and a number of small collaterals. Where the
hood is a postoperative complication to surgery of cardiac duct crosses the midline can also vary and it can even
malformations.18 empty into the right subclavian artery.
Of the non-traumatic causes, the most common is a When the duct starts to leak, a collection of chyle below
malignant lymphoma. Whenever a chylothorax of the pleura is formed. This is called a ‘chyloma’ and usually
unknown etiology occurs, the first suspicion should be of is very short-lived because the pleura soon ruptures and
an underlying malignant lymphoma. The lymphoma is the chylus empties into the pleural space. Thus, chylomas
most often of the non-Hodgkin’s type. Malignant cells are are only rarely seen clinically, for instance as a swelling of
not seen in the pleural fluid and, if the chylothorax is the the supraclavicular fossa.13 The rupture of a ‘chyloma’ can
first symptom of the disease, the diagnosis can cause great be a very dramatic clinical event with acute chest pain
difficulties.5 Other malignancies can cause blockage of causing dyspnoea and tachycardia, suggesting myocardial
the ductus by metastatic spread, but this is a fairly rare infarction or pulmonary embolism. Such episodes seem to
occurrence.6 occur particularly after traumatic rupture of the thoracic
Apart from those diseases which directly affect the duct. Depending on where the rupture occurs, some very
lymph vessels and where chylothorax is common, it is rare entities can be found: chylomediastinum, where the
Clinical presentation 391
chylus collects in the mediastinum without breaking diagnosis of a turbid or milky fluid is an empyema; cen-
through the pleura26 and chylopericardium, where it trifugation of the fluid will in the case of an empyema
empties into the pericardial sac.27–29 Very rarely, when an show a clear supernatant. The clinical picture is also differ-
abnormal communication between the ductus and the ent in most cases, the patient with empyema being obvi-
bronchi is formed or when there is a broncopleural fistula ously ill, with high fever, etc. The cell picture in empyemas
in chylothorax, chyloptysis can occur.30 is also different with almost 100 percent granulocytes. In
The thoracic duct contains many valves forcing the rare cases, a chylothorax can become secondarily infected.
chylus in one direction only. Movement of the thorax
propels the chyle forwards. There are numerous small con-
nections with veins, making it possible to ligate the ductus Investigations
without any resulting problems.
Apart from its protein and fat content, the chyle also Once the diagnosis of chylothorax has been established,
contains a large number of lymphocytes and is bacteriosta- investigations as to the cause should be carried out. If there
tic. The normal daily flow in an adult is around 2 L.31,32 is no history of trauma or surgery, or if the trauma seems
The flow of chylus increases substantially with intake of to be fairly trivial (such as a violent cough), a malignant
food and drink, and decreases to a small trickle with star- lymphoma should be excluded. A computed tomography
vation. With repeated thoracenteses or continuous (CT) scan of the thorax and upper abdomen should be
drainage of a chylothorax, large amounts of fat, proteins performed to visualize any enlarged lymph nodes or other
and lymphocytes are lost, quickly resulting in negative signs of tumor, and to enable scrutiny of the lungs.
effects on the patient’s nutritional and immunological Lymphography can be made to show where the leakage or
status. blockage is situated.34 This can be of importance for the
clinical decisions of where to take biopsies.
If no cause has been found, thoracic surgery might be
CLINICAL PRESENTATION necessary. Biopsies of any suspect area should be taken,
and before the operation it should also be decided whether
The clinical presentation of chylothorax is as for any other ligation of the ductus should be performed at the same
pleural effusion, the main symptom being dyspnea. The setting. Biopsy of the lung, especially any suspicious part
chest X-ray shows the effusion, which should lead to a seen at CT, should be performed, since this is the best way
diagnostic thoracentesis. The typical whitish fluid should to diagnose a lymphangioleiomyomatosis. It should be
alert the clinician to the possibility of the diagnosis. remembered, however, that even thoracic surgery might
not give a definitive diagnosis. A malignant lymphoma can
be very difficult to diagnose. In all unclear cases, treatment
Diagnosis of chylothorax of the chylothorax should be performed and the patient
then followed up.
The diagnosis is by analysis of the fats within the pleural
fluid. Triglyceride levels greater than 1.2 mmol/L are
highly suggestive of a chylous effusion. In equivocal cases Treatment
with lower triglycerides, a lipid electrophoresis can clarify
the diagnosis.33 Very low triglycerides virtually exclude the A large effusion causes respiratory distress, which can be
diagnosis of chylothorax, but it should be remembered relieved by thoracentesis. However, the chylus usually
that after prolonged fasting the levels can be quite low. In soon reaccumulates, and repeated thoracenteses can be
these cases, one can measure the pleural levels of triglyc- necessary. Apart from the diminished quality of life, the
erides after a meal containing a lipid composition, when patient will lose protein and other nutrients, so some alter-
the levels should increase, thereby proving that the pleural native therapy will become necessary in these cases. A
effusion is in fact a chylothorax. number of different treatments have been tried (Table
Cholesterol values should also be measured simultane- 29.2).
ously. High triglyceride levels can occur in pseudo- Treatment of the underlying disease is important and
chylothorax,34 but in this disease the cholesterol level is will in some cases cause the chylothorax to disappear.
always very high (more than 200 mg/dL) and at Examples are corticosteroids in sarcoidosis8 or treatment
microscopy cholesterol crystals can be seen, which are of heart failure. In many instances, for example malignant
thought to be diagnostic. lymphomas, specific treatment will have a good effect on
Cells are also increased, usually more than 1000/mL. the underlying disease,5 but the damage to the ductus can
There is a dominance of monocytes and, in particular, of remain and consequently so can the chylothorax; there-
lymphocytes. fore, further measures will be necessary.
The gross appearance of the fluid can be misleading, A low-fat diet, with medium-chain triglycerides, which
and in any pleural effusion of undetermined cause, a to the large extent are absorbed directly into the blood, will
lipoprotein analysis can be helpful.34 Another differential cause the chyle to decrease in amount, but there will still be
392 Effusions from lymphatic disruptions
Table 29.2 Treatment of chylothorax vention, a minimally invasive procedure, is now recom-
Treatment of chylothorax
mended by many, especially if the patient’s nutritional
status is already poor.2
Chemical pleurodesis for pleural effusions has been
A. Treatment of the underlying disease (lymphoma, other used with varying success for decades. Many different
malignancy) drugs have been used, the choice of which depended less
B. Conservative measures: on scientific investigations than on local customs. All of
Repeated thoracocenteses these different drugs have also been used for chylothorax.
Continuous drainage Examples are tetracycline, bleomycin43 and talc.44 Talc is
Pharmacologic: Octreotide now the most popular substance and good results have
Diminishing food intake been reported. The general impression is that it is more
Low-fat diet, medium-chain triglyceride diet difficult to achieve a chemical pleurodesis in chylothorax
Total parenteral nutrition than in malignant pleurisy, probably due to the normal
C. Medical pleurodesis (instillation of talc or other sclerosing pleura and perhaps a neutralizing effect of the chyle.
agent) An alternative to pleurodesis where for some reasons this
D. Surgical measures: is not feasible, is the pleuro-peritoneal shunt, which in prin-
Pleuro-peritoneal pump ciple is a one-way subcutaneous connection between the
Fibrin glue to close the leak in the duct35 pleura and the peritoneum with a pump that can be acti-
Ligation of the thoracic duct vated by light pressure. It requires daily pumping and thus
By thoracoscopy a cooperative patient. It has been used in some cases of chy-
By thoracotomy lothorax11,37 With time, the rupture in the duct often heals,
Pleurectomy making it possible to remove the shunt, and it is therefore
recommended especially for infants and children.45
When conservative measurements and pleurodesis have
failed, ligation of the thoracic duct is an alternative. It is
successful in up to 90 percent of patients.46 However, as
a flow.36 This can be tried for up to 2 weeks. The next step mentioned, there can be large variations in the anatomy of
is total parenteral nutrition, which will diminish the the duct and there can also be smaller tributaries which
chylous flow through the ductus even further. In around can leak, and therefore sometimes a full thoracotomy is
50 percent of patients with traumatic chylothorax, sponta- necessary.47 If the patient has drunk full cream before the
neous healing will occur with conservative treatment3,37 operation, the leakage is often visible at surgery,48 but
and occurs even more in children.18,20 Thus, a trial of such whether this procedure is worthwhile is controversial. The
treatment is usually recommended.1,38,39 The longer the ductus is ligated shortly above the diaphragm, and where
time of conservative therapy, the more patients will be the leakage occurs is thus unimportant if it occurs in the
cured spontaneously. Conservative measures are often main duct. The surgical procedure can also be performed
effective, especially in children.20 For example, in 39 child- by thoracoscopy with good results.49,50 Due to the rich
ren with mainly postoperative chylothorax, the effusion network of collaterals there are never any problems with
had disappeared after 14 days in 15 and after 45 days in lymph stasis afterwards.
30.18 Thus, especially in children, it seems advisable to wait
a longer time, but this will necessitate long hospitalization
and a risk of complications such as empyema. SOME DISEASES WITH SPECIAL ASPECTS OF
In recent years, a new pharmacological treatment has CHYLOTHORAX
been of great value: subcutaneously injected octreotide.
This is a peptide which regulates the release of growth Lymphomas
hormone and thyrotropin, and also has effects on the
gastro-intestinal tract, where it inhibits glandular secre- Most commonly, a non-Hodgkin’s lymphoma causes the
tion, neurotransmission, smooth-muscle contraction and chylothorax. In most instances the lymphoma is diagnosed
absorption of nutrients. Adverse effects are nausea, before the complication of chylothorax occurs, but it can
abdominal cramps, diarrhea, malabsorption of fat and also be the first manifestation of the disease. As already
flatulence. Long-term treatment can cause an increase in mentioned, even if the malignant disease can be treated, it
the number of cholesterol gallstones. Because of the inhi- is often necessary to treat the chylothorax separately.5, 38, 43
bition of absorption of fats and other nutrients, octreotide Chemical pleurodesis is one option and the best alternative
has been shown to be useful in chylothorax from many dif- seems to be talc,40 ligation of the thoracic duct is another,
ferent causes.40–42 and sometimes these two measures can be combined.
If conservative measures fail, further therapy becomes The etiology behind the chylothorax is probably inva-
necessary. One can try with medical pleurodesis, but in the sion of the thoracic duct wall by lymphoma cells, causing
end surgery might become necessary. Thoracoscopic inter- it to become more brittle. Lymphoma cells are, however,
Etiology and pathogenesis of pseudochylothorax 393
Lymphangioleiomyomatosis
If the thickening has increased considerably in size, or if suction is applied. This is immediately felt by the patient
no earlier films can be recovered and there is a large pleural and further fluid removal is painful and meaningless,
thickening and the patient has some symptoms, thoracen- unless air is allowed to enter to equalize the pressures. This
tesis should be performed. This is important not only to can be helpful in delineating the cavity later (Figure 29.2).
relieve dyspnoea but also to prevent complications which
can occur in untreated cholesterol pleurisy. A CT scan is
very helpful to illustrate the anatomy. Great care should be YELLOW NAIL SYNDROME
taken before a pleural peel is declared to be ‘inactive’.
Apart from respiratory insufficiency, the pseudochy- In 1966, Emerson described a triad of yellow nails, lym-
lothorax fluid can cause a number of complications. phedema, and pleural effusion58 (Figure 29.3). The etiol-
Infections can occur, the most common of which is a reac- ogy is thought to be a dysfunction of the lymphatics. In the
tivation of tuberculosis. Cultures for tuberculosis (TB) nails, this causes disfiguration, thickening and a yellow
should always be made on the fluid at puncture. Even if the color. In the pleural space, there is a large daily flow of
cultures are negative, a patient with some complications fluid, and almost all of this fluid is resorbed via the lym-
from a pseudochylothorax due to old TB who has never phatic vessels in the parietal pleura. It is thought that in
had modern tuberculosis treatment (which applies to the yellow nail syndrome these vessels are also affected, which
majority of these patients) should be treated with a course causes collection of fluid in the pleura.
of chemotherapy. The syndrome is defined as at least two of the three car-
Non-specific infection also occurs and can be difficult dinal findings. Often, the patient also has a long-standing
to treat. Another complicating infection can be a fungal cough and/or repeated infections which turn out to be due
one, and particularly the Aspergillus species can cause very to small bronchiectasies. Yellow nails without pleural effu-
troublesome infections which are hard to treat.57 Any sion is more common than the full-blown syndrome but
infection can also lead to bronchopleural or pleurocuta- still rare. Pleural effusions can develop after many years of
neous fistulas. yellow nails.
Two points should be stressed regarding thoracocente- The amount of fluid is highly variable and in many
sis in these patients: first, the peel is often of considerable cases small, while in others there can be considerable
thickness and in addition calcified, which can make punc- amounts which need to be treated. The fluid is an exudate
ture difficult. Second, the walls around the fluid are stiff and not chylus since the mechanism is different from what
and thus cannot adapt easily when the fluid is withdrawn, occurs in chylothorax. In rare cases, however, the fluid is a
and therefore a negative pressure can develop if forceful frank chylothorax.
(a) (b)
Figure 29.2 Pseudochylothorax, before and after thoracocentesis with letting in of air.
References 395
19. Cammarata SK, Brush RE, Hyzy RC. Chylothorax after childbirth. ◆40. Lamberts SWJ, van der Lely AJ, de Herder WW, Hofland LJ.
Chest 1991; 99: 1539–40. Octreotide. New Engl J Med 1996; 334: 246–54.
20. van Straaten HLM, Gerards LJ, Krediet TG. Chylothorax in the ●41. Kalomenidis I. Octerotide and chylothorax. Curr Opin Pulm Med
neonatal period. Eur J Pediatr 1993; 52: 2–5. 2006; 12: 264–7.
21. Wolthuis A, Lamdewé RBM, Theunissen PHMH, Westerhuis LWJJM. 42. Mares DC, Mather PN. Medical thoracoscopic talc pleurodesis for
Chylothorax or leakage of total parenteral nutrition? Eur Respir J chylothorax due to lymphoma. Chest 1998; 114: 731–5.
1998; 12: 1233–5. 43. Fogli L, Gorini P, Belcastro S. Conservative management of
22. Restoy EG, Cueto FB, Arenas EE, Duch AA. Spontaneous bilateral traumatic chylothorax: a case report. Intensive Care Med 1993;
chylothorax: uniform features of a rare condition. Eur Respir J 19; 176–7.
1988; 1: 872–3. 44. Norum J, Aasebo U. Intrapleural bleomycin in the treatment of
23. Flaherty S, Ellison R. Bilateral chylothorax following thymectomy: chylothorax. J Chemother 1994; 6: 427–30.
resolution following unilateral drainage. Mil Med 1994; 159: 627–8. 45. Graham DD, McGahren ED, Tribble CG, Daniel TM, Rodgers BM.
24. Janssen JP, Joosten HJM, Postmus PE. Thoracoscopic treatment of Use of video-assisted thoracic surgery in the treatment of
postoperative chylothorax after coronary bypass surgery. Thorax chylothorax. Ann Thorac Surg 1994; 57: 1507–12.
1994; 49: 1273. 46. Murphy MC, Newman BM, Rodgers BM. Pleuroperitoneal shunts in
25. Cheng WC, Chang CN, Lin TK. Chylothorax after endoscopic the management of persistent chylothorax. Ann Thorac Surg 1989;
sympathectomy; case report. Neurosurgery 1994; 35: 330–2. 48: 195–200.
26. Riquet M, Darse-Derippe J, Saab M, et al. Chylome médiastin après 47. Noel AA, Gloviczki P, Bender CE, et al. Treatment of symptomatic
médiastinoscopie. A propos d’une observation. Rev Mal Respir primary chylous disorders. J Vasc Surg 2001; 34: 785–91.
1993; 10: 473–6. 48. Zoetmulder F, Rutgers E, Baas P. Thoracoscopic ligation of a
27. de Winter RJ, Bresser P, Remer JWP, Kromhout JG, Reekers J. thoracic duct leakage. Chest 1994; 106: 1233–4.
Idiopathic chylopericardium with bilateral reflux of chyle. Am 49. Kent RB, Pinson TW. Thoracoscopic ligation of the thoracic duct.
Heart J 1994; 127: 936–9. Surg Endosc 1993; 7: 52–3.
28. Rose DM, Colvin SB, Danilowicz D, Isom OW. Cardiac tamponade ●50. Taylor JR, Ryu J, Colby TV, Raffin TA. Lymphangioleiomyomatosis:
secondary to chylopericardium following cardiac surgery: case report clinical course in 32 patients. N Engl J Med 1990; 323: 1254–60.
and review of the literature. Ann Thorac Surg 1982; 34: 333–6. 51. Johnson S. Lymphangioleiomyomatosis. Eur Respir J 2006; 27:
29. Dunn RP. Primary chylopericardium: a review of the literature and 1056–65.
an illustrated case. Am Heart J 1975; 89: 369–77. 52. Almoosa KF, McCormack FX, Sahn SA. Pleural disease in
30. Chyloptysis in adults. Presentation, recognition, and differential lymphangioleiomyomatosis. Clin Chest Med 2006; 27: 355–68.
diagnosis. Chest 2004; 125: 336–40. ●53. Hillerdal G. Chyliform (cholesterol) pleural effusion. Chest 1985;
31. MacFarlane JR, Holman CW. Chylothorax. Am Rev Respir Dis 1972; 88: 426–8.
105: 287–91. 54. Ferguson GC. Cholesterol pleural effusion in rheumatoid lung
◆32. Valentine VG, Raffn TA. The management of chylothorax. Chest disease. Thorax 1966; 21: 577–82.
1992; 102: 586–91. 55. Debieuvre D, Gury JP, Ory P, Jobard JM. Association
●33. Staats BA, Ellefson RD, Budahn LL, et al. The lipoprotein profile of pseudochylothorax et tuberculose pleurale. Rev Pneumol Clin
chylous and nonchylous pleural effusions. Mayo Clin Proc 1980; 1994; 50: 175–7.
55: 700–4. 56. Garcia-Pachon E, Fernandez LC, Lopez-Azorin F, Padilla-Navas I.
34. Hamm H, Pfalzer B, Fabel H. Lipoprotein analysis in a chyliform Pseudo-chylothorax in pleural effusion due to coronary artery
pleural effusion: implications for pathogenesis and diagnosis. bypass surgery. Eur Respir J 1999; 13: 1487–8.
Respiration 1991; 58: 294–300. 57. Hillerdal G. Pulmonary aspergillus infection invading the pleura.
35. Ngan H, Fok M, Wong J. The role of lymphography in chylothorax Thorax 1981; 36: 745–51.
following thoracic surgery. Br J Radiol 1988; 61: 1032–6. ●58. Emerson PA. Yellow nails, lymphedema, and pleural effusions.
36. Akaogi E, Mitsui K, Sohara Y, et al. Treatment of postoperative Thorax 1966; 21: 247–53.
chylothorax with intrapleural fibrin glue. Ann Thorac Surg 1989; 59. Iqbal M, Rossoff LJ, Marzouk KA, Steinberg HN. Yellow nail
48: 116–18. syndrome. Resolution of yellow nails after successful treatment of
37. Jensen GL, Mascioloi EA, Meyer LP, et al. Dietary modification of breast cancer. Chest 2000; 117: 1516–18.
chyle composition in chylothorax. Gastroenetrology 1989; 87: 761–5. 60. Brofman JD, Hall JB, Scott W, Little AG. Yellow nails, lymphedema
◆38. Paes ML, Powell H. Chylothorax: an update. Br J Hosp Med 1994; and pleural effusion. Chest 1990; 97: 743–5.
51: 482–90. 61. Makrilakis K, Pavlatos S, Giannikoupolos G, Toubanakis C,
39. Marts BC, Naunheim KS, Fiore AC, Pennington DG. Conservative Katsilambros N. Successful Octreotide treatment of chylous
versus surgical management of chylothorax. Am J Surg 1992; 164: pleural effusion and lymphedema in the Yellow Nail Syndrome.
532–5. Ann Int Med 2004; 141: 246–7.
30
Effusions from vascular causes
PLEURAL EFFUSIONS CAUSED BY One wonders how many of these 25 patients with pleural
PULMONARY EMBOLI effusions of undetermined etiology had pulmonary
emboli, since these authors did not do any routine evalua-
Incidence and epidemiology tion for pulmonary embolism. Gunnels6 reported on a
series of 27 patients with exudative pleural effusion who
Pulmonary embolism is a common disease, with an inci- did not have any diagnosis after the initial work-up, which
dence of approximately 200 per 100 000 person-years.1 It is included a needle biopsy of the pleura but no evaluation
the fourth leading cause of pleural effusion in the USA fol- for pulmonary embolus. She reported that during the
lowing congestive heart failure, parapneumonic effusions follow-up period, two of the 19 patients without malignant
and malignant pleural effusions;2 it has been estimated disease died and both had pulmonary emboli at autopsy.6
that 180 000 cases of pleural effusion from pulmonary Pulmonary emboli might have been discovered in more
emboli occur each year. This estimate is based on the fact patients if the diagnosis was considered.
that at least 600 000 persons have a pulmonary embolic
event each year3 and at least 30 percent of those with pul-
monary emboli have a pleural effusion.2 In an epidemio- Pathogenesis
logical study from the Czech Republic in the early 1990s,
pulmonary embolism was the fourth leading cause of Although the precise pathogenesis of the pleural effusions
pleural effusion.4 However, in most series of patients who associated with pulmonary emboli is not known, the fact
undergo thoracentesis, pulmonary embolism accounts for that the associated pleural fluid can either be an exudate or
less than 5 percent of pleural effusions. Two reasons may (rarely) a transudate suggests that there may be two sepa-
explain this apparent paradox. First, the majority of rate mechanisms.
pleural effusions associated with pulmonary embolism The mechanism for the exudative pleural effusion is
cause only blunting of the costrophrenic angles, which probably related to increased permeability of the pul-
preclude a safe thoracentesis. Second, most patients with monary capillaries to fluid and protein. As a result of the
moderate or high clinical probability of pulmonary increased permeability, excess fluid enters the interstitial
embolism are immediately anticoagulated while awaiting spaces of the lung and some of this interstitial fluid trav-
confirmatory tests, and this increases the risk of a diagnos- erses the visceral pleural. In the experimental situation,
tic thoracentesis which is not necessary once the diagnosis more than 20 percent of interstitial fluid exits the lung by
of pulmonary embolus is established. passing through the visceral pleura.7 Leckie and Tothill8
Pulmonary embolus is probably responsible for a sig- have shown that patients with exudative pleural effusion
nificant percentage of undiagnosed pleural effusions. For secondary to pulmonary emboli have a large amount of
example, Storey and coworkers5 reported the etiology of a protein entering and leaving the pleural space.
series of 133 patients with pleural effusions and noted that The factor(s) responsible for the increased permeability
only three were caused by pulmonary emboli, but causes of of the pulmonary capillaries is not definitely known; it is
the pleural effusions were not determined in 25 patients. probable that the increased permeability results from the
398 Effusions from vascular causes
release of cytokines or inflammatory mediators from the and (3) syncope with circulatory collapse. In the Registro
platelet-rich thrombi. Ischemia of the capillaries in the vis- Informatizado de la Enfermedad TromboEmbólica
ceral pleura probably does not play a significant role since (RIETE) registry, of 3391 patients with pulmonary
these capillaries are supplied by the bronchial circulation,9 embolism who had no chronic lung disease or heart
and probably do not develop ischemia with pulmonary failure, 1709 patients presented with pleuritic chest pain or
emboli. However, ischemia of the capillaries distal to the hemoptysis and 27 percent had pleural effusion, 1083 pre-
embolus may play a contributory role in increasing the vas- sented with isolated dyspnea and 12 percent had pleural
cular permeability. One cytokine that may be at least partly effusion while 599 presented with circulatory collapse and
responsible for the increased vascular permeability is vas- 16 percent had pleural effusion.19
cular endothelial growth factor (VEGF). This is one of the Multiple symptoms can occur with pulmonary
most potent agents known for increasing vascular perme- embolism. In the RIETE study, the proportion of 4444
ability. Platelets contain large quantities of VEGF.10 The patients with the following symptoms was: dyspnea, 83
pleural fluid levels of VEGF were very high in one patient percent; chest pain, 52 percent; leg pain or swelling, 37
with a pleural effusion secondary to pulmonary emboli.11 percent; cough, 20 percent; syncope, 14.5 percent; fever
The mechanism for the rare transudative pleural effu- >38°C, 12 percent; and hemoptysis, 7 percent.20
sions associated with pulmonary embolism is probably Physical findings also vary in patients with pulmonary
related to increased pressures in either the pulmonary or emboli. When two reports are combined,16,17 the propor-
systemic circulation. Pulmonary embolism is associated tion of 357 patients with the following physical findings
with both increased pulmonary arterial pressures and was: tachypnea ≥20/minute, 68 percent; rales, 31 percent;
increased systemic vascular pressures from the associated tachycardia >100/minute, 26 percent; accentuated pul-
right heart failure. Increased pressures in either of these monic component of second heart sound, 27 percent; and
locations can lead to increased pleural fluid formation, wheezes, pleural friction rub and Homans’ sign, less than 5
although it is more likely that increased pressure in the sys- percent each.
temic capillaries12 rather than increased pulmonary artery
pressure13 is directly related to the accumulation of pleural ARTERIAL BLOOD GASES AND ELECTROCARDIOGRAPHY
fluid.
Approximately one of every four patients with pulmonary
embolism has a normal pulse oximetry or partial pressure
Clinical presentation of alveolar O2 (PaO2).21 Electrocardiographic abnormali-
ties historically considered to be suggestive of pulmonary
RISK FACTORS embolism, such as the S1Q3T3 pattern or the right bundle
There are many factors that predispose patients to pul- branch block, are actually present in approximately 15
monary embolism. The most common are major surgery, percent of patients.20
previous venous thromboembolism, trauma, immobility
(e.g. infection, acute rheumatism, falls without fractures, CHEST RADIOGRAPHS
prolonged air travel), malignancy and its treatment, estro-
gen therapy, hereditable and acquired thrombophilia (e.g. Pleural effusions occur in 20–50 percent of patients with
activated protein C resistance, prothrombin G20210A pulmonary embolism. Worsley and associates22 reviewed
mutation, antiphospholipid antibody syndrome), preg- the chest radiographs of 383 patients in the Prospective
nancy and the postpartum state, and hospitalization for Investigation of Pulmonary Embolism Diagnosis
medical conditions including heart failure, stroke and (PIOPED) study with angiographically proven pulmonary
chronic lung disease. Less common risk factors are emboli and reported that 36 percent had a pleural effusion.
advanced age, obesity, central venous catheters, acute Two large multicenter studies comprising 231923 and
infections, or certain medical disorders such as myelopro- 403320 patients with pulmonary embolism reported that
liferative diseases, nephrotic syndrome, inflammatory the incidence of pleural effusion was 23 and 21 percent,
bowel disease or paroxysmal nocturnal hemoglobin- respectively, when assessed by chest radiograph.
uria.14,15 Importantly, approximately 20 percent of Recently, Porcel and colleagues24 reviewed the radiolog-
patients with pulmonary embolism have no predisposing ical characteristics of pleural effusions in 230 patients with
factors.16,17 Pulmonary embolism is thought to be a disease pulmonary embolism. Pleural effusions were observed in
of old people, yet 12 percent of 400 patients with pul- 32 and 47 percent of patients by chest radiograph and com-
monary embolism in one series were under 40 years old.18 puted tomography (CT), respectively. Several concepts
arose from this investigation. First, pleural effusions sec-
ondary to pulmonary embolism are usually unilateral and
SYMPTOMS AND SIGNS
small. Among the 73 patients with a pleural effusion visible
Patients with pulmonary emboli present with three differ- on the chest radiograph, 34 (47 percent) had unilateral left-
ent symptom complexes: (1) pleuritic chest pain or sided effusions, 28 (38 percent) had unilateral right-sided
hemoptysis (pulmonary infarction); (2) isolated dyspnea; effusions and 11 (15 percent) had bilateral effusions. In 66
Pleural effusions caused by pulmonary emboli 399
(90 percent) of the 73 patients, effusions occupied a third Classically, it has been taught that the pleural effusion
or less of the hemithorax, while in the remaining 7 (10 secondary to pulmonary embolism may be a transudate or
percent) patients the effusions were large or massive.24 an exudate, but two recent series cast doubt on this dogma.
Second, pulmonary embolism should be considered in the In these two studies, pleural effusions from 60 and 26
differential diagnosis of a loculated pleural effusion, in patients with pulmonary embolism, respectively, all fell
addition to complicated parapneumonic effusion, tubercu- into the exudative category when Light’s criteria were
losis, hemothorax and long-standing malignant effusion. applied.24,26 In fact, the potential transudative nature of the
CT demonstrated fluid loculations in 21 percent of patients effusions associated with pulmonary embolism is reported
with pleural effusion for whom the diagnosis of pulmonary in a single publication that had significant methodological
embolism was deferred for a mean of 12 days after symp- limitations.27
toms developed.24 Third, contrary to some previous The pleural fluid is yellow and not blood-tinged or
findings,25 the size of the effusions tended to be smaller if bloody in approximately 40 percent of patients.24,26 The
pulmonary infarction was present. Peripheral wedge- pleural fluid white blood cell (WBC) count ranges from
shaped opacities, likely to represent pulmonary infarction, under 100 to more than 50 000 cells/mm3.24,26,27 The dif-
were observed in 50 (38 percent) of 133 patients with pul- ferential WBC may reveal predominantly neutrophils,
monary embolism on CT scan. The frequency of pleural lymphocytes, mononuclear cells, mesothelial cells or
effusion was not significantly different in patients with or eosinophils.
without pulmonary infarction (50 percent versus 46
percent), but effusion median volumes were smaller in the
former (200 mL) than in the latter (600 mL).24 There are Investigations
two possible explanations for this finding: (1) large
amounts of pleural fluid with the resulting passive atelecta- CLINICAL PROBABILITY OF PULMONARY EMBOLISM
sis may conceal the identification of some parenchymal
opacities; and (2) ischemia may induce mesothelial cells to An important component for the strategy for the optimal
secrete various coagulation cascade proteins, such as the diagnosis of pulmonary embolism using noninvasive tests
tissue factor. The resulting procoagulant activity and fibrin is the classification of the clinical probability of pulmonary
deposition in the pleural space may lead to a spontaneous embolism as high, intermediate or low according to well-
pleurodesis, thereby precluding pleural fluid accumula- validated prediction rules (e.g. modified Wells and Geneva
tion. Finally, an absolute correlation between the sidedness scores).28,29 Table 30.1 presents two approaches to estimate
of the pleural effusion and that of the pulmonary embolism the pretest clinical probability of pulmonary embolism.
is lacking. Of the 93 patients with pleural effusion detected Although widely used, the Wells score28 contains a highly
by chest radiograph or CT, the pulmonary embolism was subjective variable, the physician’s opinion as to whether
unilateral in 61 and the pleural effusion was on the ipsilat- pulmonary embolism is the most likely diagnosis. On the
eral side in 38 and on the contralateral side in 7, whereas it other hand, the revised Geneva score29 is standardized and
was bilateral in 16. The pulmonary embolism was bilateral relies exclusively on clinical variables, but it should be
in 31 patients and the pleural effusion was also bilateral in tested in a formal outcome study. Prospective studies
6, but unilateral in 25 patients.24 The presence of an effu- using these algorithms have shown that the prevalence of
sion on the opposite side of an intraluminal filling defect pulmonary embolism with low, intermediate and high
may simply reflect the poor sensitivity of conventional clinical probability scores is approximately 10, 30 and 70
computed tomographic pulmonary angiography (CTA) percent, respectively.30
for emboli in subsegmental pulmonary arteries. The diagnosis of pulmonary embolism should be con-
sidered in every patient with an undiagnosed pleural effu-
PLEURAL FLUID FINDINGS sion. If the patient has a predisposing factor for pulmonary
emboli and has pleuritic chest pain or dyspnea that is out
The pleural fluid findings associated with pulmonary of proportion to the size of the effusion, the patient should
embolism vary widely and do not help in establishing a be started on heparin treatment and a test for pulmonary
diagnosis. Nonetheless, in patients with more than a embolism should be performed before a thoracentesis is
minimal pleural effusion, thoracentesis should be attempted.31
attempted, at least to exclude other potential etiologies. In
our experience, even in fully-anticoagulated patients D-DIMER
pending confirmatory tests for pulmonary embolism, tho-
racentesis can be safe if performed with a 25-gauge needle, The best screening test for pulmonary embolus is measure-
preferably under ultrasound guidance. If the diagnosis of ment of the D-dimers in the peripheral blood. D-dimers
pulmonary embolism is already established, a thoracente- are degradation products that result from the breakdown
sis is indicated if the patient is febrile to rule out a pleural of cross-linked fibrin by plasmin. Fibrin is cross-linked by
infection or if the effusion is increasing in size to rule out factor XIII and is the primary component of thrombus
a hemothorax. material. Increased levels of D-dimer are found in condi-
400 Effusions from vascular causes
Prediction rules
Wells criteria
Clinical signs and symptoms of deep venous thrombosis 3
Alternative diagnosis less likely than pulmonary embolism 3
Heart rate >100 beats/minute 1.5
Surgery or bedridden for ≥3 days in the previous 4 weeks 1.5
Previous deep venous thrombosis or pulmonary embolism 1.5
Hemoptysis 1
Active cancer (treatment within 6 months, or palliative) 1
Clinical probability
Low <2
Intermediate 2–6
High >6
Pulmonary embolism likely >4
Pulmonary embolism unlikely ≤4
tions that result in the activation of the fibrinolytic system. The utility of using a D-dimer test as a screening test
Thus, D-dimer tests lack specificity for thromboembolism, has been demonstrated in a recent prospective cohort
although it appears that normal levels should be useful in study that included 3306 patients with suspected pul-
excluding it.32 Elevated levels of D-dimer are also found in monary embolism.34 Patients were categorized as having
patients with recent surgery, infection, malignancy and pulmonary embolism ‘likely’ (n = 1100), or ‘unlikely’
liver disease. More than 50 percent of hospitalized patients (n = 2206) using the dichotomized version of the Wells
have elevated levels of D-dimer. Nevertheless, if the levels clinical decision rule. If the D-dimer test was negative
are normal and an appropriate test is used, the diagnosis of (VIDAS D-dimer assay or the Tinaquant assay) in patients
pulmonary embolus can be excluded. However, if the D- with an unlikely clinical probability score, pulmonary
dimer test is positive, an additional test is necessary to defi- embolism was considered excluded and no additional tests
nitely diagnose pulmonary embolism.32 were performed. When the D-dimer test was performed
It is important for physicians to understand the charac- on 2206 patients with pulmonary embolism unlikely, it
teristics and limitations of the test that is performed at was normal in 1057 (48 percent). These patients did not
their hospital. Rapid enzyme-linked immunosorbent assay undergo further evaluation or receive treatment and only
(ELISA) tests (e.g. VIDAS, Instant IA, Nycocard), and four (0.4 percent) developed a pulmonary embolus during
latex immunoturbidimetric assays (e.g. Tinaquant, the subsequent 3 months.34
HemosIL D-dimer HS, Liatest, Plus) have a high sensitiv-
ity in screening for venous thromboembolism (approxi- PERFUSION LUNG SCANS
mately 95 percent), while the red cell (e.g. SimpliRED) or
the first-generation latex agglutination tests are, in general, For the past several decades, the perfusion lung scan has
much less sensitive.33 been the primary test by which the possibility of pul-
Pleural effusions caused by pulmonary emboli 401
monary embolus has been evaluated. If the lung scan is radiation exposure (e.g. pregnancy, women of reproduc-
negative, a pulmonary embolus is virtually ruled out. If the tive age).
perfusion lung scan is a high-probability lung scan, 87 In general, the sensitivity of single-detector CTA in
percent of the patients will have a pulmonary embolism diagnosing pulmonary embolism has ranged from 75 to
and when a high-probability lung scan is coupled with a 100 percent, with a specificity that averages 95 percent.39,40
high clinical probability of embolism, the positive pre- In the past several years, CT technology has evolved from
dicted value increases to 96 percent.2 Nevertheless, the single-detector CT to multidetector CT and from 4-slice to
utility of the lung scan is limited by the fact that many 64-slice scanners, which has significantly improved resolu-
patients with a high suspicion of pulmonary embolism will tion and thus sensitivity for the detection of subsegmental
have an intermediate- or low-probability lung scan and pulmonary embolism.
therefore will require additional diagnostic tests. For If CTA scans are used for the diagnosis of pulmonary
example, in the PIOPED study, 931 patients underwent emboli, it is important that the pulmonary arteries be
both scintigraphy and pulmonary angiography. Of the 116 examined on the video monitor and not just on the hard
patients with high-probability scans and definitive copies of the scans. False-positive and false-negative results
angiograms, 102 (88 percent) had pulmonary embolism. are common when the length of the pulmonary artery is
Of the 322 patients with intermediate-probability scans, not scrutinized on the monitor. In a meta-analysis that
105 (33 percent) had pulmonary embolism, while of the included 4657 patients with suspected pulmonary
493 patients with low-probability scans, 59 (12 percent) embolism who did not receive anticoagulation after nega-
had pulmonary embolism.35 In the PIOPED study, 164 tive results on CTA, the 3-month risks for a subsequent
patients from 266 (62 percent) with positive arteriograms venous thromboembolic event and fatal pulmonary
did not have high probability lung scans. embolism were 1.4 and 0.51 percent, respectively.41 This,
If a pleural effusion is present, the perfusion lung scan and other studies,42 suggest that CTA is at least as accurate
is more difficult to interpret.36 A large effusion severely as invasive pulmonary angiography to rule out pulmonary
restricts the ability of the lung to expand and causes a shift embolism.
of perfusion to the contralateral lung. The presence of the Stein and colleagues43 investigated the use of multide-
pleural fluid itself makes interpretation of the lung scans tector CTA alone and combined with imaging of the pelvic
more difficult. An apparent perfusion defect can be and thigh veins (CTA/CTV) in 824 patients with suspected
caused by the presence of fluid anywhere in the chest. If pulmonary embolism. They reported that the sensitivity of
perfusion and ventilation scans are obtained with the CTA/CTV (90 percent) in the diagnosis of pulmonary
patient in different positions, then the lung scan may be embolism exceeds that of CTA alone (83 percent), with
misinterpreted as a ventilation/perfusion mismatch.2 If a similar specificity (95 percent). However, an accompany-
patient has more than a small pleural effusion, it is prefer- ing editorial questions if the small increase in the negative
able to perform a therapeutic thoracentesis before the predictive value from 95 to 97 percent is enough to justify
lung scan is obtained.2 the additional irradiation by CTV.44 To reduce radiation,
the PIOPED II investigators recommend CTV of only the
PULMONARY ARTERIOGRAMS femoral and popliteal veins.45
We believe that CTA is the best way to evaluate the pos-
The pulmonary arteriogram remains the gold standard for sibility of pulmonary emboli in patients with a pleural
diagnosing pulmonary emboli. However, pulmonary arte- effusion. Patients with a pleural effusion are likely to have
riography still has some limitations. First, the arteriogra- an embolus in the central, lobar, segmental or subsegmen-
phy must be performed in a special facility to which the tal pulmonary arteries and these are the areas in which
patient must be transported. Second, there are limitations CTA can detect an embolus. The additional advantage of
in the interpretation of pulmonary arteriograms. The obtaining a CTA in a patient with a pleural effusion who is
interpretation of the pulmonary angiogram is influenced being evaluated for pulmonary embolism is that the CTA
heavily by three factors: (1) the location of the throm- can also demonstrate pulmonary infiltrates or masses,
boembolic obstruction; (2) the quality of the images; and pleural nodules or thickening, or mediastinal abnormali-
(3) the experience of the interpreters.2 Angiography with ties, which may provide clues to the etiology of the pleural
digital subtraction is less time-consuming and has a diag- effusion if a pulmonary embolus is not present.
nostic accuracy similar to pulmonary arteriograms and
greater than film-screen angiography.37 DUPLEX ULTRASONOGRAPHY OF LEG VEINS
phy with venous compression. This test has a sensitivity Treatment and management
that exceeds 90 percent and a specificity that exceeds 95
percent.35,40 If deep venous thrombosis is demonstrated In the treatment of patients with pulmonary embolus, the
with ultrasonography, the diagnosis of pulmonary presence of a pleural effusion does not alter the therapy.
embolus is likely. The patient needs to be anticoagulated The primary treatment is anticoagulation (normally
and usually no further diagnostic tests directed toward the heparin followed by vitamin-K antagonists). The presence
pleural effusions are indicated. Unfortunately, deep of bloody pleural fluid should not serve as a contraindica-
venous thrombosis is not demonstrated in about 40 tion to anticoagulation, since the vast majority of such
percent of patients with pulmonary embolism,20,46 and in patients will not have significant bleeding into their pleural
these patients additional studies are necessary to delineate space if they are anticoagulated.
the etiology of the pleural effusion. Among patients with Either unfractionated heparin or low molecular weight
negative multidetector CTA scans, leg ultrasonography heparin (LMWH) is appropriate for the initial treatment
yielded an incremental rate of detection of venous throm- of pulmonary embolism.51 The advantages of LMWH
bosis of only 0.9 percent in one study.47 Thus, it seems that compared with conventional heparin include the follow-
patients undergoing multidetector CT scanning do not ing: (1) it is not necessary to monitor coagulation param-
require ultrasonography as a complementary diagnostic eters; (2) outpatient treatment is possible; (3)
test. heparin-induced thrombocytopenia is less frequent; and
(4) heparin-associated osteopenia is less frequent. The
duration of anticoagulation with vitamin-K antagonists
THORACIC ULTRASONOGRAPHY
varies according to the risk of recurrent pulmonary
Recently, a prospective multicenter study reported the embolism. Patients with a first episode of pulmonary
accuracy of thoracic ultrasound in the diagnosis of 352 embolism and a transient risk factor can be treated for
patients suspected of having pulmonary embolism.48 3–6 months. An extended-duration therapy (>12 months
Pulmonary embolism was considered definite when two or or indefinite) should be considered for idiopathic or recur-
more typical triangular or rounded pleural-based lesions rent pulmonary embolism.51
were demonstrated, and probable when one typical lesion Therapeutic thoracentesis is usually not necessary in the
with pleural effusion was present. Pulmonary embolism treatment of the pleural effusion secondary to pulmonary
was diagnosed in 194 patients, mostly by CTA. In 49 embolus because the effusions tend to be small.
percent of these patients, investigators found small pleural
effusions by thoracic ultrasound. Ultrasonography had a
sensitivity of 74 percent, and a specificity of 95 percent for
diagnosing peripheral pulmonary embolism.48 Complications
For unstable patients with suspected massive pul-
monary embolism, echocardiography searching for right
The two primary complications related to the pleural effu-
ventricular enlargement or poor ventricular function, in
sion in patients with pulmonary embolus are hemothorax
combination with leg ultrasonography, are recommended
and pleural infection. If a pleural effusion enlarges in size
as rapidly obtainable bedside tests.45
or if a contralateral pleural effusion develops after the ini-
tiation of anticoagulant therapy in a patient with a pleural
SUMMARY effusion caused by pulmonary embolism, the patient prob-
ably has one of these two complications or recurrent pul-
The following procedure is recommended when patients monary emboli.
with pleural effusion are evaluated for pulmonary emboli. Patients who develop hemothorax as a complication of
If the clinical probability of pulmonary embolus is not therapy for pulmonary embolus usually develop an
high, then a D-dimer test (one with a high sensitivity for increase in the size of their pleural effusion 4–7 days after
pulmonary emboli) should be obtained. If this is negative, anticoagulant therapy is initiated.52,53 The coagulation
it is unlikely that the patient has a pulmonary embolus and studies in patients who develop this complication are
no further evaluation for pulmonary emboli is usually usually within an acceptable therapeutic range. The patho-
indicated.49,50 If the clinical probability of pulmonary genesis of the hemothoraces in these instances is not clear.
embolus is high or if the D-dimer test is positive, addi- Although they have been attributed to rupture of the pul-
tional tests are indicated. One can proceed with the com- monary infarction,54 the evidence for this is somewhat
bination of CTA and CT venography of the femoral and lacking. The treatment for spontaneous hemothorax com-
popliteal veins.45 If no venous thrombosis or pulmonary plicating anticoagulant therapy is transfusion, as needed,
embolism is demonstrated, a pulmonary digital subtrac- plus the immediate discontinuation of the anticoagulant
tion angiography or serial leg ultrasonography should be therapy and insertion of chest tubes.2 If brisk bleeding per-
obtained if the clinical probability of pulmonary embolus sists, a thoracotomy with resection of the infarcted lobe
is high (Figure 30.1). may be necessary.54
Pleural effusions caused by other pulmonary vascular diseases 403
Yes No
Thoracentesis Clinical assessment of PE probability
(Wells or Geneva scores)
Transudate Exudate
Normal CTA/CTV
Segmental or more
Nondiagnostic findings Negative
proximal PE
(subsegmental PE)
(and/or proximal DVT)
Low or moderate High PE
PE probability probability
Figure 30.1 Suggested workup for patient with pleural effusion and suspected pulmonary embolus. PE, pulmonary embolism; CTA,
computed tomographic pulmonary angiography; CTV, computed tomographic venography; ELISA, enzyme-linked immunosorbent assay;
DVT, deep venous thrombosis.
*If measured by latex agglutination a normal D-dimer is only sufficient to exclude PE if the clinical probability of PE is low
†If multidetector CTA is used, evaluation for deep venous thrombosis may not be necessary
PLEURAL EFFUSIONS CAUSED BY OTHER Ten effusions were unilateral left-sided, 17 were unilateral
PULMONARY VASCULAR DISEASES right-sided and the remaining 17 were bilateral. The major-
ity of pleural effusions were small; 61 percent occupied less
There are other pulmonary vascular diseases that cause than 25 percent of the affected hemithorax, and only three
pleural effusions in addition to non-septic pulmonary patients with malignancy had large effusions. Pleural fluid
emboli. These include the superior vena cava (SVC) syn- was analyzed in 22 patients; none had transudates and four
drome, thoracic aortic dissection, septic pulmonary met the standard criteria of chylothorax.55 The reason for
emboli, sickle cell anemia and pulmonary veno-occlusive the exudative nature of these effusions is elusive, while the
disease. chylous effusions probably result from the increased hydro-
static pressure in the SVC and thoracic duct and its
subsequent non-traumatic rupture. Pleural fluid cytology
Superior vena cava syndrome was positive in 9 of the 17 effusions sampled from patients
with malignancy, which suggests that at least in some cases
The SVC syndrome results from the obstruction of SVC, the effusions were secondary to pleural metastases rather
with severe reduction in venous return from the head, neck, than to the SVC occlusion.
and upper extremities. A recent study focuses on the pleural
findings in patients with SVC syndrome caused by malig-
nant (e.g. lung cancer) or benign (e.g. intravascular device) Thoracic aortic dissection
conditions.55 Pleural effusions were demonstrated in 44 of
the 67 patients (66 percent) with radiological studies, either Thoracic aortic dissection is a catastrophic life-threatening
chest radiographs or CT scans, regardless of the etiology. condition that typically presents with severe, sharp, tearing
404 Effusions from vascular causes
chest pain. In one series, CT scan detected pleural effusion lactate dehydrogenase level is usually relatively high and
in 42 of 48 (87.5 percent) patients with thoracic aortic dis- may be elevated to 10 times or more the upper limit for
section.56 Pleural effusions appeared at a mean of 4.5 days serum.
(range 1–15 days) after the onset of dissection, but they Lemierre’s syndrome is a special case of septic emboli
were seen on the first day of hospitalization in 18 patients. in which an oropharyngeal infection is followed by inter-
Effusions were bilateral in 31 patients (74 percent) and left nal jugular vein septic thrombophlebitis and metastatic
sided in the remainder 11 patients (26 percent). emboli, most frequently to the lungs and joints.62
Thoracentesis was performed in six patients because of Patients frequently present with symptoms and signs of
dyspnea secondary to a massive pleural effusion, and the septic emboli to the lungs including dyspnea, pleuritic
pleural fluid was bloody in half of these cases. Most of the chest pain, fever and hemoptysis. The syndrome is most
effusions resolved before discharge from the hospital. commonly associated with the anaerobic Gram-negative
Others authors have reported effusions to be less common. rod Fusobacterium necrophorum. The diagnosis is estab-
For example, Sato et al,57 found pleural effusions in 16 of lished with evidence of metastatic infection and internal
66 (24 percent) patients with thoracic aortic dissection, jugular vein thrombophlebitis. The diagnostic procedure
mostly (13 patients) of the Standford type B (i.e. no of choice is a contrast-enhanced CT of the neck and
involvement of the ascending aorta). However, it was chest, which typically reveals distended veins with
unclear what reference standard, either chest radiograph enhancing walls, low-attenuation intraluminal filling
or CT, was used to identify the effusions. In rare cases, defects and localized soft tissue edema.63 The treatment
pleural effusion or hemothorax, rather than chest or back should include an extended course of a β-lactamase-
pain, is the first sign of thoracic aortic dissection.58 The resistant antibiotic and surgical drainage of any purulent
presence of pleural effusion is neither an indication nor a fluid collections. The role of anticoagulant therapy
contraindication for surgery. remains controversial.
Pleural effusions occur with Lemierre’s syndrome as
with any septic emboli. In one review of 109 patients
reported in the literature during a 20-year period, 43
Septic pulmonary emboli percent had pleural effusion.64 Empyema and pneumotho-
rax have also been reported.65 If the fluid is infected, tube
Septic emboli occur when fragments of the thrombus thoracostomy or thoracoscopy with the breakdown of
contain organisms, usually bacteria but occasionally fungi adhesions may be necessary.66
or parasites. Most septic emboli originate from the heart in
association with endocarditis of the tricuspid valve (e.g.
intravenous drug users), a ventricular septal defect, from
an infected venous catheter (including pacemaker leads) Sickle cell anemia
or septic thrombophlebitis. The organism most commonly
associated with septic emboli is Staphylococcus aureus. There is a high incidence of pleural effusions in patients
Patients with septic emboli present with a picture of the with sickle cell anemia who develop acute chest syn-
systemic inflammatory response syndrome. The chest drome. This syndrome is characterized by fever, cough,
radiograph usually reveals multiple, ill-defined, round or pleuritic chest pain, dyspnea, leukocytosis and new lung
wedge-shaped opacities in the periphery of the lung. The opacities on radiographs. The acute chest syndrome is the
opacities may be uniform or may vary widely in size, leading cause of death in sickle cell disease. This syn-
reflecting recurrent showers of emboli. Cavitation is fre- drome occurs in up to 50 percent of patients who have
quent and many occur rapidly; the cavities are usually sickle cell disease and is second only to pain as a cause for
thin-walled and many have no fluid level.59 hospitalization.67 The pathogenesis of the syndrome is
The reported incidences of pleural effusion with right- probably multifactorial. Possible etiological factors
sided endocarditis have varied from 25 to 69 percent. In include infection (e.g. Chlamydia pneumoniae,
one study, unilateral or bilateral effusions were present by Mycoplasma pneumoniae and respiratory viruses), in situ
chest radiograph in 7 of 13 patients (54 percent) with microvascular thrombosis and fat embolism from
septic pulmonary embolism, and by CT in two additional infarcted bone marrow.
effusions (sensitivity 69 percent).60 Two patients required When patients are seen with acute chest syndrome, the
drainage of the pleural space by tube thoracostomy or chest radiograph frequently shows bilateral patchy areas of
thoracoscopy. However, a review article concluded that consolidation. The incidence of pleural effusion with the
the incidence of pleural effusion with right-sided acute chest syndrome is 35–55 percent.68,69 In a report of
endocarditis is approximately 25 percent.61 The pleural 107 episodes of the acute chest syndrome in 77 adults, uni-
fluid is an exudate, but the cultures are negative in the lateral pleural effusions were present in 35 percent while
majority of cases. The differential count can reveal bilateral pleural effusions were present in an additional 14
neutrophils, lymphocytes or monomesothelial cells. One percent.68 In another study, a pleural effusion was present
interesting characteristic of the pleural fluid is that the in 55 percent of 537 patients with acute chest syndrome.69
Hemothorax 405
37. Smith TP, Ryan JM, Brodwater BK. Acute pulmonary 58. Little S, Johnson J, Moon BY, Mehta S. Painless left hemorrhagic
thromboembolism: comparison of the diagnostic capabilities of pleural effusion. An unusual presentation of dissecting ascending
conventional film-screen and digital angiography. Chest 2002; aortic aneurysm. Chest 1999; 116: 1478–80.
122: 968–72. 59. Han D, Lee KS, Franquet T, et al. Thrombotic and nonthrombotic
38. Weiss CR, Scatarige JC, Diette GB, et al. CT pulmonary pulmonary arterial embolism: spectrum of imaging findings.
angiography is the first-line imaging test for acute pulmonary Radiographics 2003; 23: 1521–39.
embolism: a survey of US clinicians. Acad Radiol 2006; 13: ●60. Cook RJ, Ashton RW, Aughenbaugh GL, Ryu JH. Septic pulmonary
434–46. embolism. Presenting features and clinical course of 14 patients.
39. Hogg K, Brown G, Dunning J, et al. Diagnosis of pulmonary Chest 2005; 128: 162–6.
embolism with CT pulmonary angiography: a systematic review. 61. Sexauer WP, Quezado Z, Lippmann ML, Goldberg SK. Pleural
Emerg Med J 2006; 23: 172–8. effusions in right-sided endocarditis: characteristics and
◆40. Segal JB, Eng J, Tamariz LJ, Bass EB. Review of the evidence on pathophysiology. South Med J 1992; 85: 1176–80.
diagnosis of deep venous thrombosis and pulmonary embolism. 62. Riordan T, Wilson M. Lemierre’s syndrome: more than a historical
Ann Fam Med 2007; 5: 63–73. curiosa. Postgrad Med J 2004; 80: 328–34.
41. Moores LK, Jackson WL Jr, Shorr AF, Jackson JL. Meta-analysis: ◆63. Armstrong AW, Spooner K, Sanders JW. Lemierre’s Syndrome. Curr
outcomes in patients with suspected pulmonary embolism Infect Dis Rep 2000; 2: 168–73.
managed with computed tomographic pulmonary angiography. 64. Chirinos JA, Lichtstein DM, Garcia J, Tamariz LJ. The evolution of
Ann Intern Med 2004; 141: 866–74. Lemierre syndrome: report of 2 cases and review of the literature.
●42. Quiroz R, Kucher N, Zou KH, et al. Clinical validity of a negative Medicine (Baltimore) 2002; 81: 458–65.
computed tomography scan in patients with suspected pulmonary 65. Gowan RT, Mehran RJ, Cardinal P, Jones G. Thoracic complications
embolism: a systematic review. J Am Med Assoc 2005; 293: 2012–7. of Lemierre syndrome. Can Respir J 2000; 7: 481–5.
●43. Stein PD, Fowler SE, Goodman LR, et al. Multidetector computed 66. Alifano M, Venissac N, Guillot F, Mouroux J. Lemierre’s syndrome
tomography for acute pulmonary embolism. N Engl J Med 2006; with bilateral empyema thoracis. Ann Thorac Surg 2000; 69:
354: 2317–27. 930–1.
44. Perrier A, Bounameaux H. Accuracy or outcome in suspected ◆67. Siddiqui AK, Ahmed S. Pulmonary manifestations of sickle cell
pulmonary embolism. N Engl J Med 2006; 354: 2383–5. disease. Postgrad Med J 2003; 79: 384-90.
◆45. Stein PD, Woodard PK, Weg JG, et al. Diagnostic pathways in 68. Maitre B, Habibi A, Raided-Thoraval F, et al. Acute chest syndrome
acute pulmonary embolism: recommendations of the PIOPED II in adults with sickle cell disease. Chest 2000; 117: 1386–92.
investigators. Am J Med 2006; 119: 1048–55. ●69. Vichinsky EP, Neumayr LD, Earles AN, et al. Causes and outcomes
●46. Girard P, Sanchez O, Leroyer C, et al. Deep venous thrombosis in of the acute chest syndrome in sickle cell disease. National Acute
patients with acute pulmonary embolism. Prevalence, risk factors, Chest Syndrome Study Group. N Engl J Med 2000; 342: 1855–65.
and clinical significance. Chest 2005; 128: 1593–1600. 70. Mandel J, Mark EJ, Hales CA. Pulmonary veno-occlusive disease.
47. Perrier A, Roy PM, Sanchez O, et al. Multidetector-row computed Am J Respir Crit Care Med 2000; 162: 1964–73.
tomography in suspected pulmonary embolism. N Engl J Med 71. Swensen SJ, Tashjian JH, Myers JL, et al. Pulmonary venoocclusive
2005; 352: 1760–8. disease: CT findings in eight patients. AJR Am J Roentgenol 1996;
●48. Mathis G, Blank W, Reissig A, et al. Thoracic ultrasound for 167: 937–40.
diagnosing pulmonary embolism: a prospective multicenter study 72. Resten A, Maitre S, Humbert M, et al. Pulmonary hypertension: CT
of 352 patients. Chest 2005; 128: 1531–8. of the chest in pulmonary venooclusive disease. AJR Am J
●49. Kearon C, Ginsberg JS, Douketis J, et al. An evaluation of D-dimer Roentgenol 2004; 183: 65–70.
in the diagnosis of pulmonary embolism: a randomized trial. Ann ●73. Hsu NY, Shih CS, Hsu CP, Chen PR. Spontaneous
Intern Med 2006; 144: 812–21. hemopneumothorax revisited: clinical approach and systemic
◆50. Qaseem A, Snow V, Barry P, et al. Current diagnosis of venous review of the literature. Ann Thorac Surg 2005; 80: 1859–63.
thromboembolism in primary care: a clinical practice guideline 74. Stafford RE, Linn J, Washington L. Incidence and management of
from the American College of Physicians and the American College occult hemothoraces. Am J Surg 2006; 192: 722–6.
of Family Physicians. Ann Intern Med 2007; 146: 454–8. ◆75. Porcel JM, Light RW. Thoracentesis.
◆51. Snow V, Qaseem A, Barry P, et al. Management of venous http://pier.acponline.org/physicians/procedures/physpro899/physpr
thromboembolism: a clinical practice guideline from the American o899. [Data accessed: 2006 Nov 6]. In: PIER [online database].
College of Physicians and the American College of Family Philadelphia: American College of Physicians; 2007.
Physicians. Ann Intern Med 2007; 146: 204–10. 76. Sanabria A, Valdivieso E, Gomez G, Echeverry G. Prophylactic
52. Rostand RA, Feldman RL, Block ER. Massive hemothorax antibiotics in chest trauma: a meta-analysis of high-quality
complicating heparin anticoagulation for pulmonary embolus. studies. World J Surg 2006; 30: 1843–7.
South Med J 1977; 70:1128–30. 77. Ambrogi MC, Lucchi M, Dini P, Mussi A, Angeletti CA.
53. Mrug M, Mishra PV, Lusane HC, Cunningham JM, Alpert MA. Videothoracoscopy for evaluation and treatment of hemothorax. J
Hemothorax and retroperitoneal hematoma after anticoagulation Cardiovasc Surg (Torino) 2002; 43: 109–12.
with enoxaparin. South Med J 2002; 95: 936–8. 78. Oguzkaya F, Akcali Y, Bilgin M. Videothoracoscopy versus
54. Wick MR, Ritter JH, Schuller D. Ruptured pulmonary infarction: a intrapleural streptokinase for management of post traumatic
rare, fatal complication of thromboembolic disease. Mayo Clin retained haemothorax: a retrospective study of 65 cases. Injury
Proc 2000; 75: 639–42. 2005; 36: 526–9.
●55. Rice TW, Rodriguez RM, Barnette R, Light RW. Prevalence and 79. Navsaria PH, Vogel RJ, Nicol AJ. Thoracoscopic evacuation of
characteristics of pleural effusions in superior vena cava retained posttraumatic hemothorax. Ann Thorac Surg 2004; 78:
syndrome. Respirology 2006; 11: 299–305. 282–6.
●56. Hata N, Tanaka K, Imaizumi T, et al. Clinical significance of pleural 80. Inci I, Ozcelik C, Ulku R, Tuna A, Eren N. Intrapleural fibrinolytic
effusion in acute aortic dissection. Chest 2002; 121: 825–30. treatment of traumatic clotted hemothorax. Chest 1998; 114:
57. Sato F, Kitamura T, Kongo M, et al. Newly diagnosed acute aortic 160–5.
dissection. Characteristics, treatment modifications, and ◆81. Ng CSh, Yim AP. Spontaneous hemopneumothorax. Curr Opin
outcomes. Int Heart J 2005; 46: 1083–98. Pulm Med 2006; 12: 273–7.
31
Effusions in immunocompromised hosts
Humoral Phagocytic
X-linked agammaglobulinemia Neutropenia
Autosomal recessive agammaglobulinemia Congenital
Common variable immunodeficiency Cyclic
Hyperimmunoglobulin M syndromes Chronic granulomatous disease
Immunoglobulin A deficiency Leukocyte adhesion deficiency
Immunoglobulin G subclass deficiency Hyperimmunoglobulin E syndrome
Combined humoral and cellular deficiency Myeloperoxidase deficiency
Severe combined immune deficiency (SCID) Chediak–Higashi syndrome
Inherited complement deficiencies
410 Effusions in immunocompromised hosts
Underlying conditions
Table 31.3 Causes of pleural effusion in patients with human immunodeficiency virus (HIV) infection
commonly caused by Streptococcus pneumoniae, merase chain reaction (PCR), lysozyme, interferon gamma
Staphylococcus aureus, Haemophilus influenzae and (IFN-g) and adenosine deaminase (ADA) measurements
Pseudomonas aeruginosa.2 However, parapneumonic effu- in pleural fluid for the diagnosis of tuberculous pleuritis
sions and empyema can be caused by other pathogens such have not been clearly defined because of their inability to
as Rhodococcus equi, Legionella pneumophila, Salmonella provide culture and drug sensitivity information. In one
spp. and Bordetella bronchiseptica. study from Thailand, pleural fluid ADA level at a cutoff
In the absence of pulmonary nodules, unilateral effu- value of 60 U/L had sensitivity, specificity and positive and
sion with focal air space consolidation suggests parapneu- negative predictive values of 95 percent or higher for the
monic effusion.6 In addition to antibiotics, the diagnosis of pleural tuberculosis, independent of patients’
management of complicated pleural effusions and empye- HIV status.12 The diagnostic use of ADA in areas with low
mas requires serial thoracentesis, urgent drainage and incidence of tuberculosis may lead to high false-positive
close clinical surveillance to discern the need for decortica- rates, although the finding of a low ADA level has high
tion or rib resection, and open drainage.2 Video-assisted negative predictive value and may be useful in excluding
thoracoscopic surgery (VATS) has been shown to be effi- tuberculous pleurisy.
cacious and safe for the treatment of empyema in HIV- Most HIV infected patients with tuberculous pleurisy
infected patients.7 should be started on four-drug therapy, consisting of iso-
niazid, rifampin, pyrazinamide and ethambutol. The
TUBERCULOSIS majority of HIV infected patients with tuberculous pleuri-
tis respond favorably to treatment. In one study from
Approximately two billion people are infected with South Carolina, the mortality of HIV infected patients
Mycobacterium tuberculosis.8 Although tuberculosis is a with pleural tuberculosis was 9 percent after an average
worldwide pandemic, over 75 percent of it occurs in Africa follow-up period of 31 months.13 In contrast, higher mor-
and Asia. Of the two million annual deaths due to tuber- tality rates have been reported from Africa.2 There is insuf-
culosis, 250 000 of them are associated with HIV infection. ficient evidence for the efficacy of corticosteroids in the
Pleural effusion is seen in 8–21 percent of HIV infected treatment of tuberculosis pleurisy.14
patients with pulmonary tuberculosis.2,5 In certain regions
of the world, tuberculosis is the most common cause of
pleural effusion.5 Among HIV infected patients with PNEUMOCYSTIS JIROVECII
tuberculosis, the frequency of pleural effusion is higher in Pleural effusions have been reported in 6–15 percent of
patients with CD4+ lymphocyte counts above 200 HIV infected patients with Pneumocystis jirovecii pneumo-
cells/mL.9 nia (PCP).2 Approximately 4 percent of pleural effusions
Chest radiograph findings of unilateral effusion with in HIV infected patients are associated with PCP (Table
miliary nodules and/or mediastinal lymphadenopathy 31.3). However, confirmed Pneumocystis jirovecii involve-
suggest tuberculosis.6 The diagnostic approach to tubercu- ment of the pleura is limited to few case reports.
lous pleuritis in patients with HIV is similar to that in Pneumocystis pleural disease appears to be an anatomic
other patients.10 The tuberculin skin test is positive in extension of smoldering subpleural PCP and it can be
12–63 percent of HIV infected patients with tuberculous diagnosed by staining pleural fluid for the organism.2
pleurisy, with positivity higher in those with higher CD4 Treatment requires antibiotics for the underlying PCP
counts.9 Sputum should be obtained for acid-fast bacilli and tube thoracostomy if a bronchopleural fistula is
(AFB) smear and mycobacterial culture, even in the present.
absence of parenchymal involvement. The pleural fluid
AFB smear is positive in 6–15 percent.2 Positive pleural
fluid AFB smear is more likely when the blood CD4+ lym- MISCELLANEOUS INFECTIOUS CAUSES
phocyte count is <200 cells/mL. Pleural biopsy AFB has
Pleural effusion has been reported in 5 percent of HIV
been reported to have positive rates of 44 and 69 percent.2
infected patients with pulmonary cryptococcosis and may
Pleural fluid and biopsy cultures are each positive in 30–50
precede meningitis.15 There are case reports of pleural
percent of pleural tuberculosis. In the appropriate clinical
effusions associated with various infections including
setting, the diagnosis of tuberculous pleurisy can be estab-
histoplasmosis, candidiasis, amebiasis, microfilariasis and
lished by demonstrating granulomas in pleural biopsy.
Mycobacterium avium complex.
Despite the depressed T-lymphocyte function, granulomas
have been reported in 44–88 percent of HIV infected
patients.2
Although tuberculous pleural effusions are usually Malignant causes
characterized by lymphocytosis and paucity of mesothelial
cells and eosniophils in non-HIV infected patients, cases of Ninety percent of the malignancies reported in acquired
tuberculous pleurisy with numerous mesothelial cells have immune deficiency syndrome (AIDS) are either KS or non-
been reported in patients with HIV.11 The roles of poly- Hodgkin’s lymphoma (NHL).2 Most of the AIDS-defining
412 Effusions in immunocompromised hosts
malignancies are associated with secondary viral infections pleural fluid and closed needle pleural biopsy are 75 and
by Human herpes virus (HHV)-8, Epstein–Barr virus (EBV) 100 percent, respectively. Despite chemotherapy, progno-
or Human papilloma virus.16 Since the wide-spread use of sis is poor.
HAART, the incidence rates of KS and NHL have declined.
However, KS remains the most common AIDS-associated
malignancy in the developed world and is one of the most PRIMARY EFFUSION LYMPHOMA
common cancers in developing nations. Compared with
Primary effusion lymphoma (PEL), or body-cavity-based
empyema, KS and lymphoma are more likely to be associ-
lymphoma, is a form of NHL characterized by pleural,
ated with bilateral pleural effusions.6 Current treatment
pericardial or peritoneal lymphomatous effusions in the
options for HHV-8 associated disease are ineffective,
absence of a solid tumor mass.20 It is associated with HHV-
unavailable or toxic.17
8 and accounts for 1–3 percent of all AIDS-related lym-
phomas.2 Although PEL affects mainly homosexuals with
KAPOSI SARCOMA advanced AIDS, it has also been reported in injection drug
users. The median CD4+ lymphocyte counts range
Kaposi sarcoma affects mainly homosexual and bisexual
between 34 and 84 cells/mL at the time of PEL diagnosis.
men. Pulmonary KS occurs in 47–75 percent of HIV
The PEL cells are immunophenotypically indetermi-
infected patients with cutaneous KS and may present
nate, lacking B- and T-cell-associated antigens, but
without mucocutaneous involvement.2 Pleural effusions
expressing markers associated with the late stages of B-cell
occur in 15–89 percent of pulmonary KS.2 Most KS-
differentiation. Infection of the tumor clone with HHV-8
associated pleural effusions are visibly bloody or blood-
constitutes the genetic hallmark of PEL.2 Although EBV
tinged.2 Although uncommon, chylous effusion associated
infects PEL cells, it is absent in some, in contrast to HHV-
with KS has been reported.18
8, which is present in all.
A chest radiograph finding of bilateral effusions with
Radiographic studies in PEL show pleural effusion
intrapulmonary nodules, and/or hilar lymphadenopathy
without detectable mass, parenchymal opacities or lym-
suggests KS.6 However, it is difficult to make a definitive
phadenopathy.2 In addition to pleural effusion, plain chest
diagnosis of pleural KS. Since there are no characteristic
radiographs and computed tomography show pleural
KS cells and most KS lesions are found in the visceral
thickening, pericardial thickening and pericardial and
pleura, pleural fluid and closed needle biopsies are usually
peritoneal effusions. Pleural fluid analysis shows elevated
non-diagnostic. Thoracoscopic surgical finding of the
lactate dehydrogenase and protein. PCR of the fluid shows
characteristic KS lesions in the pleura establishes the diag-
HHV-8. Although lymphoma spread outside of the body
nosis. In most cases, the presumptive diagnosis of pleural
cavity is unusual in PEL, minor involvement of the lymph
KS can be made in the appropriate clinical setting by visu-
nodes and bone marrow have been reported.2
alization of the characteristic endobronchial lesions bron-
Although some patients with PEL have been treated
choscopically and excluding other causes.
with chemotherapy, the impact of treatment on outcome
Although radiation therapy, chemotherapy and bio-
is either unknown or poor. Tube thoracostomy drainage of
logical modifiers have been used to treat KS, the major
the pleural effusion can be used for palliation. The
goal of treatment of pleural KS is palliation. HAART is
reported median survival of patients with PEL is less than
associated with regression of KS and improved survival.
7 months.2,21
Thoracentesis and tube thoracostomy are used to drain
large, symptomatic, pleural effusions associated with KS.
However, the effusions recur and sclerotherapy is usually OTHER MALIGNANCIES
ineffective. Thoracoscopic talc pleurodesis, pleuroperi-
toneal shunt or indwelling pleural catheter are other alter- Bronchogenic carcinoma is an uncommon cause of pleural
natives. The median survival of patients with pulmonary effusion in HIV infected patients. There are reports of
KS is 2–10 months, and this is shorter with pleural involve- malignant mesothelioma in HIV infected patients without
ment.2 history of asbestos exposure.22
Multicentric Castleman’s disease is a lymphoprolifera-
NON-HODGKIN’S LYMPHOMA tive disorder of the plasma cell type, which has been
reported in patients with HIV infection. Tissues involved
The incidence of NHL has declined and it occurs mainly in by the disease frequently contain HHV-8.20 Reports of
patients who are refusing or failing HAART. Among a HIV infected patients with Castleman’s disease show
cohort of 304 439 adults with AIDS in the USA, 14 lym- pleural effusion in 24 percent. Other manifestations
phomas, 4 of which involved the pleura, were identified.19 include dyspnea, cough, malaise, fever, hepato-
Pulmonary involvement, with pleural effusion in 68 splenomegaly and lymphadenopathy. Patients with
percent, is reported in 1–14 percent of AIDS-related Castleman’s disease respond to treatment with steroids
NHL.2 Pleural effusions are absent in AIDS related and chemotherapy. However, relapse of the disease and
primary pulmonary lymphoma. The diagnostic yield of deterioration to lymphoma are frequent.
Pleural diseases in transplant recipients 413
other solid organ transplant. There is paucity of data about are case reports of pleural involvement by cryptococcal,
pulmonary and pleural complications in intestinal trans- histoplasma, Pneumocystis, Mucor, Rhizopus and
plant recipients. Pleural effusion develops in 10–39 Chaetomium fungal infections.
percent of HSCT recipients.30
OTHER INFECTIONS
MYCOBACTERIA
OTHER HEMATOLOGICAL MALIGNANCIES
In endemic areas and in patients at high risk for M. tuber-
culosis infection, tuberculous pleural effusion should be Transplant recipients are at risk for the development of
included in the differential diagnoses. The measurement of NHL with pleural effusion. There are reports of pleural
ADA activity in pleural fluid provides a good tool for the effusions caused by Burkitt-like lymphoma, HHV-8 asso-
diagnosis of tuberculous pleural effusion in transplant ciated KS and PEL in solid organ transplant recipients.48,49
recipients.38 Relapse of the underlying hematological malignancy may
involve the pleura in HSCT recipients.50
FUNGI
NON-HEMATOLOGICAL MALIGNANCIES
Candidiasis and Aspergillosis infections are common in
transplant recipients, especially those with prolonged neu- Transplant recipients with smoking and asbestos exposure
tropenia. However, pleural involvement is uncommon. history are at high risk for bronchogenic carcinoma and
Candida albicans has been associated with empyema malignant mesothelioma. Several reports have shown an
requiring decortication and chest tube drainage in a increased risk of secondary malignancy following HSCT,
cardiac transplant recipient.39 Hemothorax may be seen in some of them associated with malignant pleural effu-
transplant recipients with Aspergillus pneumonia.40 There sion.51,52
Pleural diseases in hematological malignancies 415
myelodysplastic syndromes are rarely accompanied by cause pleural effusion in patients with leukemia.76–78
pleural involvement. Extramedullary hematopoiesis may cause pleural effusion
Pleural involvement by the primary lymphoma is the in chronic myelocytic leukemia.
most common cause of effusion in patients with lym- Pleural effusion occurs in approximately 6 percent of
phoma. Other causes include drug toxicity, underlying patients with multiple myeloma and may be caused by
infections, radiation therapy, secondary malignancy or, nephrotic syndrome, pulmonary embolism and congestive
rarely, autoimmune diseases.30 The size of the effusions heart failure. Myelomatous pleural involvement is rare,
may vary from very small to massive. Depending on the occurring in <1 percent of cases and mostly in IgA type
size and the cause of the effusion, patients may be asymp- multiple myeloma.
tomatic or present with life-threatening respiratory failure
or hemodynamic compromise. Although the first diagnos-
tic evaluation is usually thoracentesis, it may not give a
specific diagnosis because of the sparse malignant cells in PLEURAL EFFUSION IN PRIMARY IMMUNE
the pleural fluid. Closed or thoracoscopic pleural biopsy is DEFICIENCIES
required for diagnosis if pleural fluid cytology fails to
determine the cause of the effusion. Although pleural effu- Approximately 50 000 new cases of primary immune defi-
sions in lymphomas are usually exudates, they may also be ciencies are diagnosed annually in the USA.79 The diag-
transudates caused by venous compression or as reaction noses are delayed until adolescence or early adulthood in
to lymphomatous involvement of lung parenchyma. approximately 40 percent. Humoral deficiencies are the
Obstruction of the thoracic duct by lymphoma or radia- most common.80 Infectious complications are frequent
tion induced fibrosis may lead to chylothorax.30 In most manifestations of primary immune deficiencies (Table
patients with lymphoma, the pleural fluid responds to 31.5). Although pulmonary infections are common, the
treatment of the primary disease, whereas resistant or incidence of parapneumonic effusions and empyema has
relapsing cases may necessitate pleurodesis. not been well described.81
Pyothorax-associated lymphoma, mostly reported from
Japan, is a non-Hodgkin’s lymphoma that develops in the
pleural cavity after a long-standing history of pyothorax,
usually in patients who have undergone therapeutic pneu- PLEURAL DISEASES IN OTHER
mothorax for the treatment of pulmonary tuberculosis.75 IMMUNOCOMPROMISED PATIENTS
It develops in approximately 2 percent of patients with
chronic pyothorax, 19–64 years after the therapeutic pneu- The spleen acts as a mechanical filter for particulate anti-
mothorax. In the majority of these patients, the lymphoma gens and microorganisms in the circulation and plays roles
cells are large atypical B cells, and express latent gene prod- in both the non-specific and specific immune responses.82
ucts of EBV. Most of the patients are elderly with multiple In addition to congenital asplenia and splenectomy, func-
co-morbidities. Treatment includes chemotherapy, radia- tional hyposplenism is associated with sickle cell disease,
tion therapy and surgery, alone or in combination.75 The celiac disease, sarcoidosis, systemic lupus erythematosus,
5-year survival rate is approximately 35 percent. ulcerative colitis and amyloidosis. Polysaccharide capsular
Although rare during life, pleural infiltration with organisms, including S. pneumoniae, H. influenzae and
malignant cells is a common finding at autopsy of patients Neisseria meningitidis, are the most common pathogens
with acute leukemia.30 Pleural effusions are uncommon in causing sepsis in patients with hyposplenism. Although
chronic lymphocytic and myelocytic leukemias. The effu- patients with hyposplenism are at high risk for compli-
sions may be hemorrhagic if they are due to leukemic cated parapneumonic effusions and empyema, there is
pleural infiltration. Bacterial, viral and fungal infections paucity of data in the published literature.
14. Elliott AM, Luzze H, Quigley MA, et al. A randomized, double-blind, 36. Thacker WL, Benson RF, Schifman RB, et al. Legionella tucsonensis
placebo-controlled trial of the use of prednisolone as an adjunct sp. nov. isolated from a renal transplant recipient. J Clin Microbiol
to treatment in HIV-1-associated pleural tuberculosis. J Infect Dis 1989; 27: 1831–4.
2004; 190: 869–78. 37. Natrajan S, Hadeli O, Quan SF. Infected spontaneous chylothorax.
15. Batungwanayo J, Taelman H, Bogaerts J, et al. Pulmonary Diagn Microbiol Infect Dis 1998; 30: 31–2.
cryptococcosis associated with HIV-1 infection in Rwanda: a 38. Chung JH, Kim YS, Kim SI, et al. The diagnostic value of the
retrospective study of 37 cases. AIDS 1994; 8: 1271–6. adenosine deaminase activity in the pleural fluid of renal
16. Bernstein WB, Little RF, Wilson WH, et al. Acquired transplant patients with tuberculous pleural effusion. Yonsei Med
immunodeficiency syndrome-related malignancies in the era of J 2004; 45: 661–4.
highly active antiretroviral therapy. Int J Hematol 2006; 84: 39. Canver CC, Patel AK, Kosolcharoen P, et al. Fungal purulent
3–11. constrictive pericarditis in a heart transplant patient. Ann Thorac
17. Casper C, Wald A. The use of antiviral drugs in the prevention and Surg 1998; 65: 1792–4.
treatment of Kaposi sarcoma, multicentric Castleman disease and ●40. Lenner R, Padilla ML, Teirstein AS, et al. Pulmonary complications
primary effusion lymphoma. Curr Top Microbiol Immunol 2007; in cardiac transplant recipients. Chest 2001; 120: 508–13.
312: 289–307. 41. Collet G, Marty P, Le Fichoux Y, et al. Pleural effusion as the first
18. Maradona JA, Carton JA, Asensi V, et al. AIDS-related Kaposi’s manifestation of pulmonary toxoplasmosis in a bone marrow
sarcoma with chylothorax and pericardial involvement transplant recipient. Acta Cytol 2004; 48: 114–16.
satisfactorily treated with liposomal doxorubicin. AIDS 2002; 16: 42. Schulman LL, Reison DS, Austin JH, et al. Cytomegalovirus
806. pneumonitis after cardiac transplantation. Arch Intern Med 1991;
19. Mbulaiteye SM, Biggar RJ, Goedert JJ, et al. Pleural and peritoneal 151: 1118–24.
lymphoma among people with AIDS in the United States. J Acquir 43. Abe K, Suzuki K, Kamata N, et al. [High-resolution CT findings in
Immune Defic Syndr 2002; 29: 418–21. cytomegalovirus pneumonitis after bone marrow transplantation].
20. Hengge UR, Ruzicka T, Tyring SK, et al. Update on Kaposi’s Nippon Igaku Hoshasen Gakkai Zasshi 1998; 58: 7–11.
sarcoma and other HHV8 associated diseases. Part 2: 44. Harris NL, Ferry JA, Swerdlow SH. Posttransplant
pathogenesis, Castleman’s disease, and pleural effusion lymphoma. lymphoproliferative disorders: summary of Society for
Lancet Infect Dis 2002; 2: 344–52. Hematopathology Workshop. Semin Diagn Pathol 1997; 14: 8–14.
21. Boulanger E, Gerard L, Gabarre J, et al. Prognostic factors and ●45. Halkos ME, Miller JI, Mann KP, et al. Thoracic presentations of
outcome of human herpesvirus 8-associated primary effusion posttransplant lymphoproliferative disorders. Chest 2004; 126:
lymphoma in patients with AIDS. J Clin Oncol 2005; 23: 4372–80. 2013–20.
22. Behling CA, Wolf PL, Haghighi P. AIDS and malignant 46. Ohori NP, Whisnant RE, Nalesnik MA, et al. Primary pleural
mesothelioma – is there a connection? Chest 1993; 103: 1268–9. effusion posttransplant lymphoproliferative disorder: Distinction
23. Lipman M, Breen R. Immune reconstitution inflammatory from secondary involvement and effusion lymphoma. Diagn
syndrome in HIV. Curr Opin Infect Dis 2006; 19: 20–5. Cytopathol 2001; 25: 50–3.
24. Buckingham SJ, Haddow LJ, Shaw PJ, et al. Immune reconstitution 47. Lamba M, Jabi M, Padmore R, et al. Isolated pleural PTLD after
inflammatory syndrome in HIV-infected patients with cardiac transplantation. Cardiovasc Pathol 2002; 11: 346–50.
mycobacterial infections starting highly active anti-retroviral 48. Wolford JF, Krause JR. Posttransplant mediastinal Burkitt-like
therapy. Clin Radiol 2004; 59: 505–13. lymphoma. Diagnosis by cytologic and flow cytometric analysis of
25. Schroeder SA, Beneck D, Dozor AJ. Spontaneous pneumothorax in pleural fluid. Acta Cytol 1990; 34: 261–4.
children with AIDS. Chest 1995; 108: 1173–6. 49. Jones D, Ballestas ME, Kaye KM, et al. Primary-effusion lymphoma
26. Transplants in the US by state. The Organ Procurement and and Kaposi’s sarcoma in a cardiac-transplant recipient. N Engl J
Transplantation Network. Available from: Med 1998; 339: 444–9.
http://www.optn.org/latestData/rptData.asp 50. Park J, Park SY, Cho HI, et al. Isolated extramedullary relapse in
27. Report on state of the art in blood and marrow transplantation. the pleural fluid of a patient with acute myeloid leukemia
Center for International Blood & Marrow Transplant Research. following allogeneic BMT. Bone Marrow Transplant 2002; 30:
CIBMTR Newsletter. Available from: 57–9.
http://www.optn.org/latestData/rptData.asp 51. Germing U, Kobbe G, Sohngen D, et al. Early occurrence of an
◆28. Judson MA, Sahn SA. The pleural space and organ transplantation. adenocarcinoma after allogeneic bone marrow transplantation in
Am J Respir Crit Care Med 1996; 153: 1153–65. a patient with AML. Oncol Rep 1999; 6: 855–7.
29. Candan S, Pirat A, Varol G, et al. Respiratory problems in renal 52. Motherby H, Ross B, Kube M, et al. Pleural carcinosis confirmed by
transplant recipients admitted to intensive care during long-term adjuvant cytological methods: a case report. Diagn Cytopathol
follow-up. Transplant Proc 2006; 38: 1354–6. 1998; 19: 370–4.
◆30. Alexandrakis MG, Passam FH, Kyriakou DS, et al. Pleural effusions 53. Judson MA, Handy JR, Sahn SA. Pleural effusions following lung
in hematologic malignancies. Chest 2004; 125: 1546–55. transplantation. Time course, characteristics, and clinical
●31. Ferrer J, Roldan J, Roman A, et al. Acute and chronic pleural implications. Chest 1996; 109: 1190–4.
complications in lung transplantation. J Heart Lung Transplant 54. Shitrit D, Izbicki G, Fink G, et al. Late postoperative pleural
2003; 22: 1217–25. effusion following lung transplantation: characteristics and
32. Herridge MS, de Hoyos AL, Chaparro C, et al. Pleural complications clinical implications. Eur J Cardiothorac Surg 2003; 23: 494–6.
in lung transplant recipients. J Thorac Cardiovasc Surg 1995; 110: ●55. Judson MA, Sahn SA, Hahn AB. Origin of pleural cells after lung
22–6. transplantation: from donor or recipient? Chest 1997; 112: 426–9.
33. Backhus LM, Sievers EM, Schenkel FA, et al. Pleural space 56. Liguori C, Schulman LL, Weslow RG, et al. Late pulmonary venous
problems after living lobar transplantation. J Heart Lung complications after lung transplantation. J Am Soc Echocardiogr
Transplant 2005; 24: 2086–90. 1997; 10: 763–7.
34. Ma YK, Yan LN, Li B, et al. Diagnosis and treatment of bacterial 57. Khan SU, Salloum J, O’Donovan PB, et al. Acute pulmonary edema
pneumonia in liver transplantation recipients: report of 33 cases. after lung transplantation: the pulmonary reimplantation
Chin Med J (Engl ) 2005; 118: 1879–85. response. Chest 1999; 116: 187–94.
35. Janssens W, Van Raemdonck D, Dupont L, et al. Listeria pleuritis 1 58. Lee JT, Durzinsky DS, Wilson WR, et al. Pericardial tamponade and
week after lung transplantation. J Heart Lung Transplant 2006; 25: massive pleural effusion complicating orthotopic heart
734–7. transplantation. J Cardiovasc Surg (Torino) 1999; 40: 377–9.
References 419
●59. Hong SK, Hwang S, Lee SG, et al. Pulmonary complications 70. Ueda T, Manabe A, Kikuchi A, et al. Massive pericardial and pleural
following adult liver transplantation. Transplant Proc 2006; 38: effusion with anasarca following allogeneic bone marrow
2979–81. transplantation. Int J Hematol 2000; 71: 394–7.
60. Spizarny DL, Gross BH, McLoud T. Enlarging pleural effusion after 71. Oeda E, Shinohara K, Kamei S, et al. Capillary leak syndrome likely
liver transplantation. J Thorac Imaging 1993; 8: 85–7. the result of granulocyte colony-stimulating factor after high-
61. Zajko AB, Claus D, Clapuyt P, et al. Obstruction to hepatic venous dose chemotherapy. Intern Med 1994; 33: 115–19.
drainage after liver transplantation: treatment with balloon 72. Goldsby R, Pulsipher M, Adams R, et al. Unexpected pleural
angioplasty. Radiology 1989; 170: 763–5. effusions in 3 pediatric patients treated with STI-571. J Pediatr
62. Montalbano M, Neff GW, Yamashiki N, et al. A retrospective Hematol Oncol 2002; 24: 694–5.
review of liver transplant patients treated with sirolimus from a 73. Duran FG, Piqueras B, Romero M, et al. Pulmonary complications
single center: an analysis of sirolimus-related complications. following orthotopic liver transplant. Transpl Int 1998; 11(Suppl
Transplantation 2004; 78: 264–8. 1): S255–9.
63. Carcillo J Jr, Salcedo JR. Urinothorax as a manifestation of 74. Afessa B, Litzow MR, Tefferi A. Bronchiolitis obliterans and other
nondilated obstructive uropathy following renal transplantation. late onset non-infectious pulmonary complications in
Am J Kidney Dis 1985; 5: 211–13. hematopoietic stem cell transplantation. Bone Marrow Transplant
64. DeCamp MM, Tilney NL. Late development of intractable 2001; 28: 425–34.
lymphocele after renal transplantation. Transplant Proc 1988; 20: 75. Narimatsu H, Ota Y, Kami M, et al. Clinicopathological features of
105–9. pyothorax-associated lymphoma; a retrospective survey involving
65. Schmidt RJ, Bender FH, Chang WW, et al. Sarcoidosis after renal 98 patients. Ann Oncol 2007; 18: 122–8.
transplantation. Transplantation 1999; 68: 1420–3. 76. Desselle BC, Bozeman PM, Patrick CC. Diagnostic utility of
66. Ozkaynak MF, Weinberg K, Kohn D, et al. Hepatic veno-occlusive thoracentesis for neutropenic children with cancer. Clin Infect Dis
disease post-bone marrow transplantation in children 1995; 21: 887–90.
conditioned with busulfan and cyclophosphamide: incidence, risk 77. Mori M, Imamura Y, Maegawa H, et al. Cytology of pleural
factors, and clinical outcome. Bone Marrow Transplant 1991; 7: effusion associated with disseminated infection caused by
467–74. varicella-zoster virus in an immunocompromised patient. A case
67. van den Bosch MA, van Hoe L. MR imaging findings in two report. Acta Cytol 2003; 47: 480–4.
patients with hepatic veno-occlusive disease following bone 78. Trudo FJ, Gopez EV, Gupta PK, et al. Pleural effusion due to herpes
marrow transplantation. Eur Radiol 2000; 10: 1290–3. simplex type II infection in an immunocompromised host. Am J
68. Seber A, Khan SP, Kersey JH. Unexplained effusions: association Respir Crit Care Med 1997; 155: 371–3.
with allogeneic bone marrow transplantation and acute or chronic 79. Verbsky JW, Grossman WJ. Cellular and genetic basis of primary
graft-versus-host disease. Bone Marrow Transplant 1996; 17: immune deficiencies. Pediatr Clin North Am 2006; 53: 649–84.
207–11. 80. Ballow M. Primary immunodeficiency disorders: antibody
69. Toren A, Nagler A. Massive pericardial effusion complicating the deficiency. J Allergy Clin Immunol 2002; 109: 581–91.
course of chronic graft-versus-host disease (cGVHD) in a child 81. Hollingsworth CL. Thoracic disorders in the immunocompromised
with acute lymphoblastic leukemia following allogeneic bone child. Radiol Clin North Am 2005; 43: 435–47.
marrow transplantation. Bone Marrow Transplant 1997; 20: 82. Sumaraju V, Smith LG, Smith SM. Infectious complications in
805–7. asplenic hosts. Infect Dis Clin North Am 2001; 15: 551–65.
This page intentionally left blank
32
Effusions from connective tissue diseases
tis is the first manifestation of SLE in only 5–10 percent,24 Fluid lupus cells
but is an early feature in 25–30 percent, usually preceded
by arthralgia, and is sometimes associated with pneumoni- Lupus erythematosus (LE) cells are found occasionally in
tis and pericarditis.15 The presence of pleural disease is serous effusions in SLE5,48–54 and may appear at the onset
usually accompanied with multisystem involvement.25 The or later in the disease course.55 Some have regarded the
prevalence of pleural disease at presentation increases in presence of LE cells in serous effusions as virtually diag-
the elderly,26 but is also frequent in children.27–29 Pleuritis nostic of SLE,48–51 an approach which is no longer favored
is also a frequent feature of drug-induced lupus syndromes since they are subject to observer variation,50 are not easy
(not covered here, see Chapter 33, Effusions caused by to detect and have also been reported in effusions from
drugs). Finally, pleural effusions may result from compli- rheumatoid joints,56 malignant pleural effusions57 and in
cations of SLE. pleural fluid, without clinical evidence of SLE.55
Additionally, LE cells are not found in the pleural fluid of
all patients with lupus pleuritis58 and are usually associated
Radiographic imaging with the presence of serum LE cells; thus, the added diag-
nostic value of pleural fluid LE cells is questionable.58
On chest radiography, pleural effusions are generally
small, but occasionally massive.30–36 They are bilateral in
approximately 50 percent, with no predilection for the Pleural fluid antinuclear antibodies
right or left side4 and may, in serial chest radiographs,
often change sides.14 Pleural fluid antinuclear antibodies (ANA) titers >1:60
and pleural fluid to serum ANA ratios >1 are suggestive
but not diagnostic of SLE pleuritis; high pleural fluid titers
Diagnosis (up to 1:640) are seen occasionally in patients with non-
SLE exudative effusions.59,60 Sometimes, demonstrable
In patients with known SLE, the diagnosis is often obvious, pleural fluid ANA may be absent.61 Patients with SLE pleu-
but in patients with pleuritis associated with non-specific ritis sometimes have higher pleural fluid ANA, ssDNA,
arthritis, the major differential diagnosis is rheumatoid dsDNA, smooth muscle and ribonucleoprotein titers than
pleural disease (distinction between the two is covered in in the serum.59,62 Pleural fluid ANA titer levels may be
the section on rheumatoid arthritis). SLE pleuritis should useful in distinguishing between lupus pleuritis and other
be considered in any patient with an exudative pleural causes of pleural disease in SLE patients, in which pleural
effusion of unknown etiology.15 Measurement of lactic fluid ANA titers tend to be low or absent.58 The authors in
acid has been proposed a rapid tool to distinguish between a recent study regarding the diagnostic value of ANA in
bacterial pleural inflammation and other causes of exuda- SLE pleural effusion reached the same conclusion.63
tive effusions.37 However, a new pleural effusion in
patients with SLE remains a diagnostic challenge.38
Treatment
Pleural fluid analysis Due to the small number of reported patients, the best type
of intervention remains uncertain.64 Pleuritic pain in SLE
The pleural fluid is usually exudative, yellow or serosan- may respond to non-steroidal anti-inflammatory agents,
guineous.5,15 Hemothorax has also been reported.36,39 and almost always responds to corticosteroid therapy.
Neutrophils or monocytes predominate, but lymphocytes Sometimes, high doses may be needed for large effusions
are common in chronic effusions.3,5 Pleural fluid and severe pleuritis.65 Immunosuppressants (azathioprine
eosinophilia is generally considered to rule out underlying and hydroxychloroquine) may be efficacious32,66,67 and
SLE.40 The pleural fluid pH is usually higher than 7.20, the monthly cyclosporine courses have been used with a good
glucose concentrations are slightly decreased,5,41 but outcome.67 Recurrent pleural effusions usually respond to
usually higher than 60 mg/dL, and lactate dehydrogenase pleurodesis.32,68,69
(LDH) levels are less than the upper normal limits of
serum LDH.3
Reductions in pleural fluid complement levels have Prognosis
been observed,1–3,15,42–44 possibly reflecting complement
conversion by immune complexes.3,5,42–44 Immune Small effusions do not require any treatment, usually
complex deposition may engender pleural effusions by resolve spontaneously and have no known prognostic sig-
increasing capillary permeability.45 Elevated CA-125 levels nificance.15 Conversely, pleuritic pain appears to be an
have been reported in the pleural fluid of a number of con- adverse prognostic marker,15,70,71 with a mean survival of
nective tissue diseases, including SLE.46,47 less than 4 years in affected cases.15
Rheumatoid arthritis 423
Most of the patients are asymptomatic, the effusions are The fluid is exudative and non-odorous and may be
small and resolve spontaneously.77 The pleural effusion cloudy, greenish yellow or opalescent.85 Glucose levels
may present in the absent of arthritis.78 In large effusions, exceed 30 mg/100 mL in 20–30 percent, but it has been
breathlessness may result from pulmonary compression argued that normal glucose concentrations may indicate
due to the size of the effusion. Occasional complaints are causes other than RA.80 Other biochemical findings
fever, cough and pleuritic pain.79 The presence of pleural include pH >7.20, LDH levels more than twice the upper
disease has not been linked to more severe systemic limit of the normal serum value, low complement and
disease, but is associated with a higher prevalence of immune complex levels and rheumatoid factor titers
cardiac and ocular lesions.73 In a study of 1968 subjects, 81 (>1:320) that exceed serum titers.79 Whole complement
percent of patients who were found to have rheumatoid activity and C3, C4 levels are less in RA pleural fluid than
pleural effusions were male, and the average age of onset in non-rheumatoid effusions.44 The complement cascade
was 51 years (range 35–69 years).73 In over half of the is activated through both classic and the alternative path-
cases, rheumatoid effusions are associated with subcuta- ways in rheumatic pleurisy; in one study, determinations
neous nodules73,70,80 and usually follow the onset of joint of SC5b-9 and C4d/C4 content in pleural fluid most accu-
manifestations. Only very rarely (6 percent) do effusions rately distinguished between rheumatic, tuberculous and
precede arthritis and, in 11 percent, pleural and systemic malignant effusions.86
disease present concurrently. As in empyema and tuberculous effusions, the activity
of adenosine deaminase in rheumatoid effusions is higher
in pleural fluid than in serum, indicating local synthesis of
Radiographic imaging ADA by cells within the pleural cavity in RA.87
diagnostic value88,91 since they may also be seen in tuber- are pleural rheumatoid nodules demonstrated which are
culous effusions and empyema.83 diagnostic.96,97 At thoracoscopy the parietal pleura has a
granular appearance and the histopathological changes in
tissue gained from thoracoscopic biopsies are often diag-
Glucose nostic.85
◆6. Alarcon-Segovia D, Alarcon DG. Pleuro-pulmonary manifestations 30. Bouros D, Panagou P, Papandreou L, Kottakis I, Tegos C. Massive
of systemic lupus erythematosus. Dis Chest 1961; 39: 7–17. bilateral pleural effusion as the only first presentation of systemic
●7. Man BL, Mok CC. Serositis related to systemic lupus lupus erythematosus. Respiration 1992; 59: 173–5.
erythematosus: prevalence and outcome. Lupus 2005; 14: 822–6. 31. Elborn JS, Conn P, Roberts SD. Refractory massive pleural effusion
8. Ropes MW. Systemic lupus erythematosus. Cambridge: Harvard in systemic lupus erythematosus treated by pleurectomy. Ann
University Press; 1976. Rheum Dis 1987; 46: 77–80.
◆9. Miller LR, Greenberg SD, McLarty JW. Lupus lung. Chest 1985; 88: 32. Kaine JL. Refractory massive pleural effusion in systemic lupus
265–9. erythematosus treated with talc poudrage. Ann Rheum Dis 1985;
●10. Haupt HM, Moore GW, Hutchins GM. The lung in systemic lupus 44: 61–4.
erythematosus. Analysis of the pathologic changes in 120 33. Bulgrin JG, Dubois EL, Jacobson G. Chest roentgenographic
patients. Am J Med 1981; 71: 791–8. changes in systemic lupus erythematosus. Radiology 1960; 74:
11. Gross M, Esterly J R, Earle R H. Pulmonary alterations in systemic 42.
lupus erythematosus. Am Rev Respir Dis 1972; 105: 572–7. 34. Taylor TL, Ostrum H. The roentgen evaluation of systemic lupus
●12. Miller MH, Urowitz MB, Gladman DD, Killinger DW. Systemic erythematosus. Am J Roentgenol 1959; 82: 95.
lupus erythematosus in males. Medicine (Baltimore) 1983; 62: 35. Mathlouthi A, Ben M’rad S, Merai S, et al. Massive pleural effusion
327–34. in systemic lupus erythematosus: thoracoscopic and
13. Mayor AM, Vila LM. Gender differences in a cohort of Puerto immunohistological findings. Monaldi Arch Chest Dis 1998; 53:
Ricans with systemic lupus erythematosus. Cell Mol Biol (Noisy-le- 34–6.
grand) 2003; 49: 1339–44. 36. Mulkey D, Hudson L. Massive spontaneous unilateral hemothorax
◆14. Harvey AM, Shulman LE, Tumulty PA, et al. Systemic lupus in systemic lupus erythematosus. Am J Med 1974; 56: 570.
erythematosus: a review of the literature and clinical analysis of 37. Brook I. Measurement of lactic acid in pleural fluid. Respiration
138 cases. Medicine (Baltimore) 1954; 33: 291–437. 1980; 40: 344–8.
15. Winslow WA, Ploss LN, Loitman B. Pleuritis in systemic lupus 38. Wang DY. Diagnosis and management of lupus pleuritis. Curr Opin
erythematosus: it’s importance as an early manifestation in Pulm Med 2002; 8: 312–16.
diagnosis. Ann Intern Med 1958; 49: 70–88. 39. Passero FC, Myers AR. Hemopneumothorax in systemic lupus
●16. Segasothy M, Phillips PA. Systemic lupus erythematosus in erythematosus. J Rheumatol 1980; 7: 183–6.
Aborigines and Caucasians in central Australia: a comparative 40. Lakhotia M, Mehta SR, Mathur D, Baid CS, Varma AR. Diagnostic
study. Lupus 2001; 10: 439–44. significance of pleural fluid eosinophilia during initial
17. Camilleri F, Mallia C. Male SLE patients in Malta. Adv Exp Med thoracocentesis. Indian J Chest Dis Allied Sci 1989; 31: 259–64.
Biol 1999; 455: 173–9. 41. Carr DT, Lillington GA, Mayne JG. Pleural-fluid glucose in systemic
●18. Molina JF, Molina J, Garcia C, et al. Ethnic differences in the lupus erythematosus. Mayo Clin Proc 1970; 45: 409–12.
clinical expression of systemic lupus erythematosus: a 42. Hunder GG, McDuffie FC, Huston KA, Elveback LR, Hepper NG.
comparative study between African-Americans and Latin Pleural fluid complement, complement conversion, and immune
Americans. Lupus 1997; 6: 63–7. complexes in immunologic and nonimmunologic diseases. J Lab
19. Chang CC, Shih TY, Chu SJ, et al. Lupus in Chinese male: a Clin Med 1977; 90: 971–80.
retrospective study of 61 patients. Chung Hua I Hsueh Tsa Chih 43. Glovsky MM, Louie JS, Pitts WH Jr, Alenty A. Reduction of pleural
(Taipei) 1995; 55: 143–50. fluid complement activity in patients with systemic lupus
20. Costallat LT, Coimbra AM. Systemic lupus erythematosus in 18 erythematosus and rheumatoid arthritis. Clin Immunol
Brazilian males: clinical and laboratory analysis. Clin Rheumatol Immunopathol 1976; 6: 31–41.
1993; 12: 522–5. 44. Hunder GG, McDuffie FC, Hepper NG. Pleural fluid complement in
21. Al Rawi Z, Al Shaarbaf H, Al Raheem E, Khalifa SJ. Clinical features systemic lupus erythematosus and rheumatoid arthritis. Ann Intern
of early cases of systemic lupus erythematosus in Iraqui patients. Med 1972; 76: 357–63.
Br J Rheumatol 1983; 22: 165–71. 45. Andrews BS, Arora NS, Shadforth MF, Goldberg SK, Davis JS. The
22. Bell R, Lawrence DS. Chronic pleurisy in systemic lupus role of immune complexes in the pathogenesis of pleural
erythematosus treated with pleurectomy. Br J Dis Chest 1979; 73: effusions. Am Rev Respir Dis 1981; 124: 115–20.
314–16. ●46. Kimura K, Ezoe K, Yokozeki H, Katayama I, Nishioka K. Elevated
23. Wang DY, Chang DB, Kuo SH, et al. Systemic lupus erythematosus serum CA125 in progressive systemic sclerosis with pleural
presenting as pleural effusion: report of a case. J Formos Med effusion. J Dermatol 1995; 22: 28–31.
Assoc 1995; 94: 746–9. ●47. Yucel AE, Calguneri M, Ruacan S. False positive pleural biopsy and
24. Mitra B, Sengupta P, Saha K, Sarkar N, Pal J. Systemic lupus high CA125 levels in serum and pleural effusion in systemic lupus
erythematosus presenting with recurrent pleural effusion without erythematosus. Clin Rheumatol 1996; 15: 295–7.
any systemic manifestation. J Assoc Physicians India 2005; 53: 48. Carel RS, Shapiro MS, Cordoba O, Taragan R, Gutman A. LE cells in
1073–6. pleural fluid. Arthritis Rheum 1979; 22: 936–7.
25. Paran D, Fireman E, Elkayam O. Pulmonary disease in systemic 49. Keshgegian AA. Lupus erythematosus cells in pleural fluid. Am J
lupus erythematosus and the antiphospholipid syndrome. Clin Pathol 1978; 69: 570–1.
Autoimmun Rev 2004; 3: 70–5. 50. Naylor B. Cytological aspects of pleural, peritoneal and pericardial
26. Baker SB, Rovira JR, Campion EW, Mills JA. Late onset systemic fluids from patients with systemic lupus erythematosus.
lupus erythematosus. Am J Med 1979; 66: 727–32. Cytopathology 1992; 3: 1–8.
27. Chantarojanasiri T, Sittirath A, Preutthipan A, Tapaneya-Olarn W, 51. Yoshiyuki OR, Shioya S, Handa K, Shimizu K. Lupus erythematosus
Suwanjutha S. Pulmonary involvement in childhood systemic cells in pleural fluid cytologic diagnosis in two patients. Acta Cytol
lupus erythematosus. J Med Assoc Thai 1999; 82(Suppl 1): 1977; 21: 215–17.
S144–8. 52. Reda MG, Baigelman W. Pleural effusion in systemic lupus
28. Gedalia A, Molina JF, Molina J, et al. Childhood-onset systemic erythematosus. Acta Cytol 1980; 24: 553–7.
lupus erythematosus: a comparative study of African Americans 53. Makashir R, Jayaram G. Lupus erythematosus cells in pleural fluid.
and Latin Americans. J Natl Med Assoc 1999; 91: 497–501. Diagn Cytopathol 1988; 4: 273–4.
●29. Nadorra RL, Landing BH. Pulmonary lesions in childhood onset 54. Sethi S, Pooley RJ, Yu GH. Lupus erythematosus (LE) cells in pleural
systemic lupus erythematosus: analysis of 26 cases, and summary fluid: initial diagnosis of systemic lupus erythematosus by
of literature. Pediatr Pathol 1987; 7: 1–18. cytologic examination. Cytopathology 1996; 7: 292–4.
428 Effusions from connective tissue diseases
55. Chao TY, Huang SH, Chu CC. Lupus erythematosus cells in pleural ◆77. Balbir-Gurman A, Yigla M, Nahir AM, Braun-Moscovici Y.
effusions: diagnostic of systemic lupus erythematosus? Acta Cytol Rheumatoid pleural effusion. Semin Arthritis Rheum 2006; 35:
1997; 41: 1231–3. 368–78.
56. Hunder GG, Pierre RV. In vivo LE cell phenomenon. Arthritis Rheum 78. Allan JS, Donahue DM, Garrity JM. Rheumatoid pleural effusion in
1970; 13: 570–1. the absence of arthritic disease. Ann Thorac Surg 2005; 80:
57. Greis M, Atay Z. Zytomorphologische Begleitreaction bei malignen 1519–21.
Pleuraerguessen. Pneumonologie (suppl) 1990; 44: 262–4. 79. Halla JT, Schronhenloher RE, Volanakis JE. Immune complexes and
●58. Wang DY, Yang PC, Yu WL, et al. Comparison of different other laboratory features of pleural effusions. Ann Intern Med
diagnostic methods for lupus pleuritis and pericarditis: a 1980; 92: 748–52.
prospective three-year study. J Formos Med Assoc 2000; 99: 80. Lillington GA, Carr DT, Mayne GJ. Rheumatoid pleurisy with pleural
375–80. effusion. Arch Intern Med 1971; 128: 764–8.
●59. Khare V, Baethge B, Lang S, Wolf RE, Campbell GD Jr. Antinuclear 81. Brennan SR, Daly JJ. Large pleural effusions in rheumatoid
antibodies in pleural fluid. Chest 1994; 106: 866–71. arthritis. Br J Dis Chest 1979; 73: 133–40.
●60. Win T, Groves AM, Phillips GD. Antinuclear antibody positive 82. Pritikin JD, Jensen WA, Yenokida GG, Kirsch CM, Fainstat M.
pleural effusion in a patient with tuberculosis. Respirology 2003; Respiratory failure due to a massive rheumatoid pleural effusion. J
8: 396–7. Rheumatol 1990; 17: 673–5.
●61. Ferreiro JE, Reiter WM, Saldana MJ. Systemic lupus erythematosus 83. Joseph J, Sahn SA. Connective tissue diseases and the pleura.
presenting as chronic serositis with no demonstrable antinuclear Chest 1993; 104: 262–70.
antibodies. Am J Med 1984; 76: 1100–5. ●84. Ferrer JS, Munoz XG, Orriols RM, Light RW, Morell FB. Evolution of
●62. Riska H, Fyhrquist F, Selander RK, Hellstrom PE. Systemic lupus idiopathic pleural effusion: a prospective, long-term follow-up
erythematosus and DNA antibodies in pleural effusions. Scand J study. Chest 1996; 109: 1508–13.
Rheumatol 1978; 7: 159–60. 85. Faurschou P, Francis D, Faarup P. Thoracoscopic, histological, and
63. Porcel JM, Ordi-Ros J, Esquerda A, et al. Antinuclear antibody clinical findings in nine case of rheumatoid pleural effusion.
testing in pleural fluid for the diagnosis of lupus pleuritis. Lupus Thorax 1985; 40: 371–5.
2007; 16: 25–7. 86. Salomaa ER, Viander M, Saaresranta T, Terho EO. Complement
64. Breuer GS, Deeb M, Fisher D, Nesher G. Therapeutic options for components and their activation products in pleural fluid. Chest
refractory massive pleural effusion in systemic lupus 1998; 114: 723–30.
erythematosus: a case study and review of the literature. Semin ●87. Pettersson T, Ojala K, Weber TH. Adenosine deaminase in the
Arthritis Rheum 2005; 34: 744–9. diagnosis of pleural effusions. Acta Med Scand 1984; 215:
65. Brasington RD, Furst DE. Pulmonary disease in systemic lupus 299–304.
erythematosus. Clin Exp Rheumatol 1985; 3: 269–76. 88. Naylor B. The pathognomonic cytologic picture of rheumatoid
66. Ben Chetrit E, Putterman C, Naparstek Y. Lupus refractory pleural pleuritis. Acta Cytol 1990; 34: 465–73.
effusion: transient response to intravenous immunoglobulins. J 89. Aru A, Engel U, Francis D. Characteristic and specific histological
Rheumatol 1991; 18: 1635–7. findings in rheumatoid pleurisy. Acta Pathol Microbiol Immunol
67. Sherer Y, Langevitz P, Levy Y, Fabrizzi F, Shoenfeld Y. Treatment of Scand [A] 1986; 94: 57–62.
chronic bilateral pleural effusions with intravenous ◆90. Sahn SA. Immunologic diseases of the pleura. Clin Chest Med
immunoglobulin and cyclosporin. Lupus 1999; 8: 324–7. 1985; 6: 103–12.
●68. McKnight KM, Adair NE, Agudelo CA. Successful use of 91. Faurschou P. Decreased glucose in RA-cell-positive pleural
tetracycline pleurodesis to treat massive pleural effusion effusion: correlation of pleural glucose, lactic dehydrogenase and
secondary to systemic lupus erythematosus. Arthritis Rheum 1991; protein concentration to the presence of RA-cells. Eur J Respir Dis
34: 1483–4. 1984; 65: 272–7.
●69. Gilleece MH, Evans CC, Bucknall RC. Steroid resistant pleural 92. Carr DT, McGuckin WF. Pleural fluid glycose. Am Rev Resp Dis
effusion in systemic lupus erythematosus treated with tetracycline 1968; 97: 302–5.
pleurodesis. Ann Rheum Dis 1988; 47: 1031–2. 93. Ferguson GC. Cholesterol pleural effusion in rheumatoid lung
70. Cook RJ, Gladman DD, Pericak D, Urowitz MB. Prediction of short disease. Thorax 1966; 21: 577–82.
term mortality in systemic lupus erythematosus with time 94. Jones FL, Blodget RC. Empyema in rheumatoid pleuropulmonary
dependent measures of disease activity. J Rheumatol 2000; 27: disease. Ann Intern Med 1971; 74: 665–71.
1892–5. 95. Dieppe PA. Empyema in rheumatoid arthritis. Ann Rheum Dis
●71. Cervera R, Khamashta MA, Font J, et al. Morbidity and mortality in 1975; 34: 181–5.
systemic lupus erythematosus during a 5-year period. A 96. Anonymous. Pleural effusion in rheumatoid arthritis (Editorials).
multicenter prospective study of 1,000 patients. European Lancet 1972; i: 480–1.
Working Party on Systemic Lupus Erythematosus. Medicine 97. Faurschou P. Rheumatoid pleuritis and thoracoscopy. Scand J
(Baltimore) 1999; 78: 167–75. Respir Dis 1974; 55: 277–83.
72. Fuller HM. On rheumatism, rheumatic gout and sciatica: Their ◆98. Ward R. Pleural effusion and rheumatoid disease. Lancet 1961; ii:
pathology, symptoms and treatment. New York: S. S. & W. Wood, 1336.
1854. 99. Russell ML, Gladman DD, Mintz S. Rheumatoid pleural effusion:
73. Walker WC, Wright V. Pulmonary lesions and rheumatoid arthritis. lack of response to intrapleural corticosteroid. J Rheumatol 1986;
Medicine (Baltimore) 1968; 47: 501–20. 13: 412–15.
●74. Hyland RH, Gordon DA, Broder I, et al. A systematic controlled 100. Emerson RA. Pleural effusion complicating rheumatoid arthritis. Br
study of pulmonary abnormalities in rheumatoid arthritis. J Med J 1956; 1: 428.
Rheumatol 1983; 10: 395–405. 101. Walker WC, Wright W. Rheumatoid pleuritis. Ann Rheum Dis
75. Dodson WH, Hollingsworth JW. Pleural effusion in rheumatoid 1967; 26: 467.
arthritis. Impaired transport of glucose. N Engl J Med 1966; 275: 102. Brunk JR, Drash EC, Swineford O. Rheumatoid pleuritis successfully
1337–42. treated with decortication. Am J Med 1966; 251: 545.
●76. Jurik AG, Davidsen D, Graudal H. Prevalence of pulmonary 103. Yarbrough JW, Sealy WC, Miller JA. Thoracic surgical problems
involvement in rheumatoid arthritis and its relationship to some associated with rheumatoid arthritis. J Thorac Cardiovasc Surg
characteristics of the patients. A radiological and clinical study. 1975; 69: 347–54.
Scand J Rheumatol 1982; 11: 217–24. 104. Yigla M, Simsolo C, Goralnik L, Balabir-German A, Nahir AM. The
References 429
problem of empyematous pleural effusion in rheumatoid arthritis: 126. Dudley-Hart F, Bogdanovich A, Nichol WD. The thorax in
report of two cases and review of the literature. Clin Rheumatol ankylosing spondilitis. Ann Rheum Dis 1950; 9: 116–31.
2002; 21: 180–3. 127. Kinnear WJ, Shneerson JM. Acute pleural effusions in inactive
105. Chapman PT, O’ Donnell JL, Moller PW. Rheumatoid pleural ankylosing spondylitis. Thorax 1985; 40: 150–1.
effusion: response to intrapleural corticosteroid. Rheumatology 128. Haslock I. Ankylosing spondylitis. Baillieres Clin Rheumatol 1993;
1992; 19: 478–80. 7: 99–115.
106. Light RW, Jenkinson SG, Minh Vea. Observations on pleural ◆129. Zorab PA. The lungs in ankylosing spondylitis. Q J Med 1962; 31:
pressures as fluid is withdrawn during thoracentesis. Am Rev Resp 267–80.
Dis 1980; 121: 799–804. 130. Crompton GK, Cameron SJ, Langlands AO. Pulmonary fibrosis,
107. Funauchi M, Ikoma S, Yu H, et al. A case of progressive systemic pulmonary tuberculosis and ankylosing spondylitis. Br J Dis Chest
sclerosis complicated by massive pleural effusion with elevated 1974; 68: 51–6.
CA125. Lupus 2000; 9: 382–5. ●131. Spencer DG, Park WM, Dick HM, Papazoglou SN, Buchanan WW.
108. Hiramatsu K, Takeda N, Okumura S, Takuno H, Yasuda K. Radiological manifestations in 200 patients with ankylosing
[Progressive systemic sclerosis associated with massive pleural and spondylitis: correlation with clinical features and HLA B27. J
pericardial effusion in a 90-year-old woman]. Nippon Ronen Reumatol 1979; 6: 305–15.
Igakkai Zasshi 1996; 33: 535–9. ◆132. Rosenow E, Strimlan CV, Muhm JR, Ferguson RH. Pleuropulmonary
109. Lee YH, Ji JD, Shim JJ, Kang KH, Song GG. Exudative pleural manifestations of ankylosing spondylitis. Mayo Clin Proc 1977; 52:
effusion and pleural leukocytoclastic vasculitis in limited 641–9.
scleroderma. J Rheumatol 1998; 25: 1006–8. ◆133. Prakash UB. Respiratory complications in mixed connective tissue
110. Thompson AE, Pope JE. A study of the frequency of pericardial and disease. Clin Chest Med 1998; 19: 733–46, ix.
pleural effusions in scleroderma. Br J Rheumatol 1998; 37: 134. Battista G, Zompatori M, Poletti V, Canini R. Thoracic
1320–3. manifestations of the less common collagen diseases. A pictorial
111. Takeda N, Teramoto S, Ihn H, et al. [A case of very late onset essay. Radiol Med (Torino) 2003; 106: 445–1.
overlap syndrome of systemic sclerosis and systemic lupus ◆135. Bull TM, Fagan KA, Badesch DB. Pulmonary vascular
erythematosus]. Nippon Ronen Igakkai Zasshi 2000; 37: 74–9. manifestations of mixed connective tissue disease. Rheum Dis Clin
112. Lakhanpal S, Lie JT, Conn DL, Martin WJ. Pulmonary disease in North Am 2005; 31: 451–64, vi.
polymyositis/dermatomyositis: a clinicopathological analysis of 65 ◆136. Aiello M, Chetta A, Marangio E, Zompatori M, Olivieri D. Pleural
autopsy cases. Ann Rheum Dis 1987; 46: 23–9. involvement in systemic disorders. Curr Drug Targets Inflamm
113. Ozawa Y, Kurosaka D, Hashimoto N. [An autopsy case of Allergy 2004; 3: 441–7.
dermatomyositis with rapidly progressive interstitial pneumonia]. ●137. Sullivan WD, Hurst DJ, Harmon CE, et al. A prospective evaluation
Nihon Rinsho Meneki Gakkai Kaishi 1995; 18: 552–8. emphasizing pulmonary involvement in patients with mixed
114. Schwarz MI, Matthay RA, Sahn SA, et al. Interstitial lung disease connective tissue disease. Medicine (Baltimore) 1984; 63:
in polymyositis and dermatomyositis: analysis of six cases and 92–107.
review of the literature. Medicine (Baltimore) 1976; 55: 89–104. 138. Beier JM, Nielsen HL, Nielsen D. Pleuritis-pericarditis – an unusual
115. Miyata M, Fukaya E, Takagi T, et al. Two patients with polymyositis initial manifestation of mixed connective tissue disease. Eur Heart
or dermatomyositis complicated with massive pleural effusion. J 1992; 13: 859–61.
Intern Med 1998; 37: 1058–63. 139. Richard P, Sabouret P, Vayre F, Desrame J, Ollivier JP.
●116. Kawamata K, Haraoka H, Hirohata S, et al. Pleurisy in primary [Pleuropericarditis complicated of tamponade disclosing mixed
Sjogren’s syndrome: T cell receptor beta-chain variable region connective tissue disease. Remission with non-steroidal anti-
gene bias and local autoantibody production in the pleural inflammatory agents. Apropos of a case]. Ann Cardiol Angeiol
effusion. Clin Exp Rheumatol 1997; 15: 193–6. (Paris) 1996; 45: 513–15.
●117. Bloch KJ, Buchanan WW, Wohl MJ, Bunim JJ. Sjogren’s syndrome. 140. Hoogsteden HC, van Dongen JJ, van der Kwast TH, Hooijkaas H,
A clinical, pathological, and serological study of sixty-two cases. Hilvering C. Bilateral exudative pleuritis, an unusual pulmonary
Medicine (Baltimore) 1992; 71: 386–401. onset of mixed connective tissue disease. Respiration 1985; 48:
●118. Constantopoulos SH, Papadimitriou CS, Moutsopoulos HM. 164–7.
Respiratory manifestations in primary Sjogren’s syndrome. A 141. Ilan Y, Ben Yehuda A, Okon E, Breuer R. Mixed connective tissue
clinical, functional, and histologic study. Chest 1985; 88: 226–9. disease presenting as a left sided pleural effusion. Ann Rheum Dis
●119. Papathanasiou MP, Constantopoulos SH, Tsampoulas C, Drosos AA, 1992; 51: 1157–8.
Moutsopoulos HM. Reappraisal of respiratory abnormalities in 142. Swygert LA, Maes EF, Sewell LE, et al. Eosinophilia–myalgia
primary and secondary Sjogren’s syndrome. A controlled study. syndrome. Results of national surveillance. J Am Med Assoc 1990;
Chest 1986; 90: 370–4. 264: 1698–703.
◆120. Strimlan CV, Rosenow EC III, Divertie MB, Harrison EG Jr. ●143. Williamson MR, Eidson M, Rosenberg RD, Williamson SL.
Pulmonary manifestations of Sjogren’s syndrome. Chest 1976; 70: Eosinophilia–myalgia syndrome: findings on chest radiographs in
354–61. 18 patients. Radiology 1991; 180: 849–52.
121. Kashiwabara K, Kishi K, Narushima K, et al. [Primary Sjogren’s 144. Killen JW, Swift GL, White RJ. Eosinophilic fasciitis with
syndrome accompanied by pleural effusion]. Nihon Kyobu Shikkan pulmonary and pleural involvement. Postgrad Med J 2000; 76:
Gakkai Zasshi 1995; 33: 1325–9. 36–7.
122. Tanaka A, Tohda Y, Fukuoka M, Nakajima S. [A case of Sjogren’s 145. Cullen MH, Stansfeld AG, Oliver RT, Lister TA, Malpas JS. Angio-
syndrome with pleural effusion]. Nihon Kokyuki Gakkai Zasshi immunoblastic lymphadenopathy: report of ten cases and review
2000; 38: 628–31. of the literature. Q J Med 1979; 48: 151–77.
◆123. Ogihara T, Nakatani A, Ito H, et al. Sjogren’s syndrome with pleural 146. Sugiyama H, Kotajima F, Kamimura M, et al. [Pulmonary
effusion. Intern Med 1995; 34: 811–14. involvement in immunoblastic lymphadenopathy: case reports and
124. Alvarez-Sala R, Sanchez-Toril F, Garcia-Martinez J, Zaera A, Masa review of literature published in Japan]. Nihon Kyobu Shikkan
JF. Primary Sjogren syndrome and pleural effusion. Chest 1989; Gakkai Zasshi 1995; 33: 1276–82.
96: 1440–1. ●147. Guillevin L, Cohen P, Gayraud M, et al. Churg–Strauss syndrome.
125. Tanaka H, Itoh E, Shibusa T, et al. Pleural effusion in ankylosing Clinical study and long-term follow-up of 96 patients. Medicine
spondylitis; successful treatment with intra-pleural steroid (Baltimore) 1999; 78: 26–37.
administration. Respir Med 1995; 89: 509–11. ◆148. Lanham JG, Elkon KB, Pusey CD, Hughes GR. Systemic vasculitis
430 Effusions from connective tissue diseases
with asthma and eosinophilia: a clinical approach to the 160. Romero S, Vela P, Padilla I, et al. Pleural effusion as manifestation
Churg–Strauss syndrome. Medicine (Baltimore) 1984; 63: 65–81. of temporal arteritis. Thorax 1992; 47: 398–9.
●149. Choi YH, Im JG, Han BK, et al. Thoracic manifestation of 161. Turiaf J, Valere PE, Gubler MC. [Recurrent pleurisy during temporal
Churg–Strauss syndrome: radiologic and clinical findings. Chest arteritis]. Poumon Coeur 1967; 23: 633–52.
2000; 117: 117–24. 162. Garcia-Alfranca F, Solans R, Simeon C, et al. Pleural effusion as a
150. Erzurum SC, Underwood GA, Hamilos DL, Waldron JA. Pleural form of presentation of temporal arteritis. Br J Rheumatol 1998;
effusion in Churg–Strauss syndrome. Chest 1989; 95: 1357–9. 37: 802–3.
●151. Hirasaki S, Kamei T, Iwasaki Y, et al. Churg–Strauss Syndrome with 163. Ramos A, Laguna P, Cuervas V. Pleural effusion in giant cell
pleural involvement. Intern Med 2000; 39: 976–8. arteritis. Ann Intern Med 1992; 116: 957.
●152. Gonzalez L, Van Ordstrand HS. Wegener’s granulomatosis. Review 164. Gur H, Ehrenfeld M, Izsak E. Pleural effusion as a presenting
of 11 cases. Radiology 1973; 107: 295–300. manifestation of giant cell arteritis. Clin Rheumatol 1996; 15:
153. Bambery P, Katariya S, Sakhuja V, et al. Wegener’s granulomatosis 200–3.
in north India. Radiologic manifestations in eleven patients. Acta 165. Umezawa T, Saji T, Matsuo N, Odagiri K. Chest x-ray findings in
Radiol 1988; 29: 11–13. the acute phase of Kawasaki disease. Pediatr Radiol 1989; 20:
●154. Cordier JF, Valeyre D, Guillevin L, Loire R, Brechot JM. Pulmonary 48–51.
Wegener’s granulomatosis. A clinical and imaging study of 77 ●166. Tunaci A, Berkmen YM, Gokmen E. Thoracic involvement in
cases. Chest 1990; 97: 906–12. Behcet’s disease: pathologic, clinical, and imaging features. AJR
●155. Fauci AS, Wolff SM. Wegener’s granulomatosis: studies in Am J Roentgenol 1995; 164: 51–6.
eighteen patients and a review of the literature. Medicine 167. Walsh FW, Solomon DA, Espinoza LR, Adams GD, Whitelocke HE.
(Baltimore) 1973; 73: 315–24. Human adjuvant disease. A new cause of chylous effusions. Arch
156. Lohrmann C, Uhl M, Schaefer O, et al. Serial high-resolution Intern Med 1989; 149: 1194–6.
computed tomography imaging in patients with Wegener 168. Pasteur M, Laroche C, Keogan M. Pleuropericardial effusion in a 50
granulomatosis: differentiation between active inflammatory and year old woman. Pleuropericardial effusion caused by adult inset
chronic fibrotic lesions. Acta Radiol 2005; 46: 484–91. Still’s disease. Postgrad Med J 2001; 77: 346, 355–7.
157. Blundell AG, Roe S. Wegener’s granulomatosis presenting as a 169. Nishio J, Koike R, Iizuka H, et al. [A refractory case of adult-onset
pleural effusion. Br Med J 2003; 327: 95–6. Still’s disease]. Nihon Rinsho Meneki Gakkai Kaishi 1997; 20:
158. Weir IH, Muller NL, Chiles C, et al. Wegener’s granulomatosis: 191–8.
findings from computed tomography of the chest in 10 patients. 170. Cheema GS, Quismorio FP Jr. Pulmonary involvement in adult-
Can Assoc Radiol J 1992; 43: 31–4. onset Still’s disease. Curr Opin Pulm Med 1999; 5: 305–9.
159. Bosch X, Ramirez J. [Bilateral lung images and respiratory
insufficiency in an 86-year-old man with polyarteritis nodosa].
Med Clin (Barc) 1999; 113:189–97.
33
Effusions caused by drugs
IOANNIS KALOMENIDIS
A variety of commonly used medications may cause be obtained. It should be remembered that a patient may not
pleural disorders, most commonly, pleural effusion. reveal the use of a drug unless asked specifically.1 Treatment
Pleural fibrosis and, rarely, hemothorax can be associated of drug-induced pleural disease mainly consists of discon-
with the use of certain agents. The pathogenesis of drug- tinuation of the offending drug, a measure usually adequate
induced pleural disease is poorly understood in the major- to lead to partial or complete remission. Corticosteroids are
ity of the cases. However, oxidant-induced mesothelial sometimes administered, more commonly in patients with
and/or endothelial cell injury, acute hypersensitivity-type severe clinical manifestations, based on the belief that the
reaction, direct dose-related toxic effect or chemically underlying mechanism may include pleural inflammation.
induced pleural inflammation, are among the possible However, with the exception of certain conditions, the
mechanisms.1,2 benefit of corticosteroids is uncertain.
Pleural disease can be the dominant or even the sole In this chapter, I describe the clinical, radiographical
manifestation of an adverse drug reaction. Otherwise, it and serological abnormalities that characterize pleural
may appear in combination with drug-associated lung disease resulting from drug reactions. When available,
disease or even as a component of generalized syndromes. information concerning the pathogenesis and the pathol-
The most common symptom on patient presentation is ogy, the clinical course and the response to the treatment
dyspnea on exertion, which may be due to either underly- of the diseases will be also provided. A drug is thought to
ing parenchymal disease or the accumulation of excessive be the cause of a pleural abnormality when the disease
amounts of pleural fluid, or even the progressive pleural occurs after the administration of the drug, other usual or
fibrosis. Cough, thoracic pain and constitutional symp- possible etiologies have been convincingly ruled out, and
toms, such as fever, malaise and weight loss, are sometimes the disease resolves when the drug is discontinued. An
present. even stronger proof is the recrudescence of the disorder
The clinical importance of prompt recognition of an when the drug is readministered. As will be noted, the clin-
adverse drug reaction as the cause of pleural disease lies on ical data presented by the authors of some case reports to
the fact that in most cases, pleural disease resolves upon the support a casual relationship between the use of a drug and
withdrawal of the drug. For this reason, and although drug the subsequent pleural abnormalities are rather weak to
reactions account for a small percentage of pleural abnor- definitely establish such a relationship. However, these
malities, the clinician must consider the possibility of cases will be presented to facilitate possible related future
drug-induced pleural disease in all patients with pleural effu- clinical observations, which may lead to outright con-
sions or fibrosis and a detailed history of drug intake must clusions.
432 Effusions caused by drugs
sion recurred, but again resolved after the cessation of the Fluid retention is a frequent adverse effect of rhIL-11 treat-
therapy. ment and can manifest as peripheral and pulmonary
edema as well as pleural effusions.21 The frequency of
pleural effusions in one study of patients with breast
Imidapril cancer patients was 42 percent.22 Although all of these
patients had distant metastases and thoracentesis was not
Yoshida and associates reported on a patient with hyper- performed in any, the fact that all the pleural effusions
tension treated with imidapril for 6 years who developed responded well to diuretic treatment suggests that the
low-grade fever and a left-sided pleural effusion17 with most probable explanation of the effusions is fluid over-
pleural fluid and peripheral eosinophilia (38 and 28 load.
percent, respectively) that did no respond to antibiotics.
Interferons
CYTOKINES AND IMMUNOMODULATING
AGENTS Interferons are used in the treatment of various malignan-
cies and chronic hepatitis. Interferons can cause several
Interleukin 2 significant adverse effects including depression, bone
marrow suppression, interstitial pneumonitis, retinopa-
Recombinant human interleukin 2 (rhIL-2), which has thy, hearing loss, vitiligo and autoimmune diseases, as well
been used in the treatment of melanoma or renal cell car- as milder disorders including flu-like symptoms,
cinoma, is associated with a variety of reversible adverse headache, irritability and alopecia. Pleural effusion, poss-
effects, including fever, chills, lethargy, diarrhea, anemia, ibly caused by recombinant human interferon-a, has been
thrombocytopenia, eosinophilia, confusion and diffuse reported in a patient with chronic hepatitis C.23 The
erythroderma.18 Respiratory abnormalities include patient developed an asymptomatic, moderate, right-sided
parenchymal infiltrates and pleural effusion. Two exudative, lymphocytic pleural effusion 14 days after
studies,18,19 including a total of 108 patients, indicated that beginning the treatment. Although a slight increase of
approximately half of those treated with rhIL-2 may serum antinuclear antibody (ANA) titer was observed, the
develop pleural effusion. Diffuse infiltrates resembling patient’s symptoms and signs did not satisfy the criteria of
pulmonary edema (41–80 percent) and focal infiltrates (22 any autoimmune disorders, including systemic lupus ery-
percent) may also be present. Pleural effusion may be thematosus (SLE), a disease that may either be exacerbated
found in the absence of parenchymal abnormalities. The or induced by the drug. Tuberculus pleuritis, which is
characteristics of the fluid were not reported. Patients encountered with increased frequency in patients on inter-
receiving bolus rather than constant intravenous therapy feron treatment, was ruled out. The pleural effusion grad-
were more likely to develop respiratory complications.18 ually resolved upon the discontinuation of the interferon.
Severe respiratory failure was reported in 35 percent of the
patients and some of them required intubation and
mechanical ventilation. As reported in one of these studies, Granulocyte-colony stimulating growth factor
the pleural effusions tended to improve, but they persisted
for 4 weeks following therapy cessation in 17 percent of Busmanis et al.24 reported a case of a 43-year-old woman
patients.18 Pleural effusions resolved later than the with breast carcinoma who developed a pleural effusion
parenchymal infiltrates. after a cycle of chemotherapy supported by granulocyte-
The respiratory adverse effects of the rhIL-2 are prob- colony stimulating growth factor (G-CSF). Although
ably a manifestation of the generalized capillary leak syn- infection and malignancy were ruled out, the possible role
drome that may occur after the administration of this of the chemotherapeutic regimen in the development of
cytokine. Berthianume et al.20 proposed that pulmonary the pleural effusion was not discussed. Subsequently,
edema related to the rhIL-2 administration may be the Nakamura et al.25 reported a case of a 57-year-old man
combined result of increased permeability and hydrostatic with rheumatoid arthritis and polycystic kidneys who
pressure. Thus, it is possible that pleural fluid originates received G-CSF for the treatment of methotrexate-
from the leaky capillaries in the lung. induced pancytopenia. The patient was in a chronic
hemodialysis program. He developed bilateral exudative
pleural effusion 9 days after the end of a 10-day course of
Interleukin 11 G-CSF. The effusion persisted for more than 20 days and
resolved after the administration of prednisolone. In my
Recombinant human interleukin 11 (rhIL-11) is used to opinion, other possible etiologies of the pleural effusion,
prevent severe thrombocytopenia and to reduce the need such as rheumatoid pleuritis, uremia or hemodialysis,
for platelet transfusions following myelosuppressive were not effectively ruled out. Thus, I do not think that
chemotherapy in patients with non-myeloid malignancies. this evidence linking the administration of G-CSF with the
434 Effusions caused by drugs
development of pleural effusion is convincing. However, stimulate fibroblast growth.1 A rather rare mechanism
deAvezo et al.26 reported a life-threatening capillary leak could be fluid retention, as reported in a patient who was
syndrome after G-CSF was used for mobilization and col- taking pergolide and developed bilateral pleural effusion
lection of peripheral blood progenitor cells for allogeneic and peripheral edema, both of which responded to diuretic
transplantation in a 37-year-old donor. During therapy.41 Furthermore, some authors proposed that prior
leukophoresis she suddenly developed hypotention, exposure to asbestos represents a specific risk factor for the
hypoxemia, ascites, pericardial and pleural effusions, development of ergot-related pleural disease.36,42
edema, neurologic changes and hepatocellular injury. G- Respiratory symptoms usually arise after a treatment
SCF was withdrawn and she was treated with the intra- period of more than 6 months, though latent periods of 3
venous administration of fluid and methylprednizolone weeks to 30 years have been reported.37 Dyspnea develops
until she fully recovered. insidiously and may be accompanied by dry cough, pleu-
ritic or non-pleuritic chest pain, malaise and recurrent or
chronic moderate to low-grade fever.1,37 The radiographic
Immunoglobulin abnormalities consist of unilateral or bilateral pleural
thickening and/or effusion. Rounded atelectasis and occa-
Bolanos-Meade et al.27 reported a patient with recurrent sional linear pleural calcification have been also
pleural effusions attributed to the use of intravenous observed.37 Pleural thickening is often massive and irregu-
immunoglobulin (IVIG) for the treatment of idiopathic lar, resembling malignant mesothelioma. The pleural fluid
thrombocytopenic purpura. She was receiving a 2-day is an exudate with predominantly lymphocytes and some-
course of the IVIG every four to six weeks. On the second times eosinophils.32,37,43 Bloody effusion has been reported
day of the seventh or the eighth course, bilateral exdutative in a patient taking methysergide.30 The erythrocyte sedi-
lymphocytic pleural effusions developed, but resolved mentation rate (ESR) and C-reactive protein (CRP) levels
completely 2 weeks later. The effusions could not be attrib- are frequently elevated14,36,37 and normochromic anemia35
uted to infectious, malignant or autoimmune disease. The as well as weakly positive serum ANA or rheumatoid factor
same thing happened with the next course of the IVIG. (RF) have been reported.37 Pleural biopsy reveals non-
However, when a different preparation of IVIG was specific fibrosis of the parietal pleura with few inflamma-
administered, there was no recurrence of the pleural effu- tory cells.37
sions. Thus, I would suggest that pleural effusion could A few patients on treatment with bromocryptine, ergo-
represent a reaction to a component of the first prepara- tamine, cabergoline or dihydroergocristine develop
tion and not of the IVIG itself. parenchymal disease, usually but not always associated
with pleural disease.14,37 Chest X-ray revealed interstitial
infiltrates in all patients and apical patchy fibrosis in one
ERGOT DERIVATIVES patient. Interestingly, in patients with pleural fibrosis with
or without obvious parenchymal disease, bronchoalveolar
Ergot derivatives have been used in the treatment and pro- lavage (BAL) discloses polymorphonuclear or lymphocytic
phylaxis of migraine and cluster headaches (methysergide, alveolitis.
ergotamine) and Parkinson’s disease (bromocriptine, The vast majority of patients will experience gradual
dihydroergotamine, nicergoline, pergolide, dopergine, improvement of the symptoms upon the withdrawal of the
lisuride, dihydroergocristine, dihydroergocryptine, caber- drug. Classically, the pleural abnormalities will gradually
golide). The long-term administration of any of these clear over a period of months to 4 years often leaving
drugs can lead to a distinctive pattern of pleuropulmonary residual pleural thickening.14,37,44 Parenchymal infiltrates
abnormalities, consisting of pleural fibrosis with or tend to resolve more rapidly and completely.37
without effusion, occasionally associated with parenchy- Corticosteroids are often used, but their necessity or their
mal fibrosis. First described by Graham et al.28 in 1966 in a effectiveness are not established. Importantly, Ling et al.35
patient receiving methysergide, the potential of ergot reported a patient with pleuropulmonary fibrosis due to
derivatives to produce these fibrosing abnormalities was cabergoline who had persistent if not progressive disease,
subsequently confirmed and thoroughly character- despite the discontinuation of the drug and the adminis-
ized.1,14,29–39 A variety of other fibrosing disorders of the tration of long-term corticosteroid treatment. The patient
retroperitoneum, mediastinum, pericardium and heart died a few months after the diagnosis from respiratory
valves, either singly or, rarely, in combination with pleural insufficiency and, although the death was not attributed to
disease has been also reported.14,35,37,39,40 the pleural thickening by the authors, this could be the
The pathogenesis of the fibrosis which is induced by the only fatal case of pleuropulmonary fibrosis related to an
ergot derivatives remains unclear. It has been suggested that ergot derivative.
fibrinogenesis may be related to the serotoninergic activity Should the drug definitely be discontinued after it is
of the drugs, as similar fibrosing disorders may be seen in recognized as the cause of pleuropulmonary fibrosis?
association with serotonin-secreting carcinoid tumors.37 Miller14 reported that in seven of eight patients with pleu-
Serotonin can cause either pleural vasoconstriction14 or ropulmonary disease caused by bromocriptine who con-
Chemotherapeutic agents 435
tinued the drug after the diagnosis was made had no clin- treated with ATRA, this syndrome was the reported reason
ical or radiographic evidence of progressive disease putting of death in nine patients.48,49 Early administration of high-
into question the necessity of drug withdrawal. However, dose corticosteroid treatment usually leads to prompt
Kok-Jensen et al.,29 who followed-up 12 patients with symptomatic improvement, while full radiographic recov-
pleural fibrosis caused by methysergide, reported that ery may require several days and residual pleural thicken-
those who continued to take methysergide for the longest ing may remain in an occasional patient.49,50
period after the onset had the worst disease after 6 months.
Therefore, given the possibility that the disease is dose-
related and related to the length of the time the patient has CHEMOTHERAPEUTIC AGENTS
been on the drug, as well as the excellent outcome after the
drug withdrawal, I would agree with the recommendation The most common respiratory adverse effect caused by
that patients who are taking ergot alkaloids on a long-term cytotoxic agents is interstitial pneumonitis that may
basis should have a radiograph annually; if there is evi- present with fever, cough and progressive dyspnea, evolve
dence of pleural or parenchymal disease, strong consider- into ARDS and be fatal.1,8 Interstitial lung fibrosis, pleural
ation should be given to stopping the ergot alkaloids and effusion and pleural fibrosis may occur. Drug withdrawal
using an alternative drug for the treatment of Parkinson’s and corticosteroids usually lead to resolution of the symp-
disease.45 toms, although the radiographic clearance may be only
partial. Pneumothorax has been rarely reported in patients
with lung metastases treated with combination chemo-
RETINOIDS therapy regimens.14,51 The extent to which the drugs con-
tribute in this complication is not defined. It is possible
Isotretinoin that the administration of the chemotherapeutic agent
leads to the necrosis of a pleural metastases and a commu-
Isoretinoin, a retinoid compound, has been used in the nication between the alveolus and the pleural space.
treatment of cystic acne and may be also useful as chemo-
preventive agent for various malignancies.8 There are two
case reports indicating that the drug can cause eosinophilic Methotrexate
pleural effusion 1–6 months after the initiation of treat-
ment.46,47 Parenchymal disease or blood eosinophilia was The best known respiratory complication of methotrexate
absent. The effusion resolved 1–3 months after drug dis- treatment is life-threatening pneumonitis, which occurs in
continuation. Fever and cough were reported in one of patients who are taking long-term therapy for rheumatoid
these patients, while the other experienced only dyspnea arthritis or other autoimmune diseases.8,52 Lung histology
on exertion. The manufacturer of isotretinoin has on file reveals non-specific inflammatory findings. Pleural disease
three other cases of pleural effusion occurring in patients has not been reported in association with interstitial pneu-
who took isotretinoin for acne.47 monia. Two groups reported pleural disease associated
with methotrexate in patients with malignancies who
received high dose courses of methotrexate. Walden and
All- trans -retinoic acid co-workers53 reported that 14 of 317 patients (4.5 percent)
with trophoblastic tumors who had been treated with
All-trans-retinoic acid (ATRA) is used in the treatment of methotrexate developed pleuritic chest pain 2–5 days after
acute promyelocytic leukemia.48 Up to one-quarter of the the injection. Pleural effusions were found in four. Pleural
patients taking the drug develop a life-threatening reaction disease was also reported in 18 of 210 patients (8.5
that resembles capillary-leak syndrome which is attributed percent) who were treated for osteogenic sarcoma with
to IL-2.49,50 Symptoms may develop 1–22 days after the weekly courses of methotrexate.54 Most of the patients pre-
initiation of the treatment and include fever, fluid reten- sented with severe pleuritic chest pain of sudden onset
tion, dyspnea, multiple organ failure and hemorrhagic and after the third or fourth course of the treatment. Chest X-
thrombotic manifestations. Leukocytosis is usually ray revealed thickening of the interlobar pleura. There is
present. Chest X-ray reveals a combination any of the fol- no report of pleural effusion. The pain resolved after 3–5
lowing: signs of fluid overload, ground glass opacities, days but the pleural thickening persisted, sometimes even
nodules, consolidation and pericardial and pleural effu- for years.
sions. In one series, 11 of 15 patients with the syndrome
developed either right-sided or bilateral pleural effusion.
The characteristics of the pleural fluid were not reported. Procarbazine
Extreme respiratory manifestations of the syndrome,
such as diffuse alveolar hemorrhage or acute respiratory Procarbazine has been used in the treatment of Hodgkin’s
distress syndrome (ARDS), may be fatal. In fact, among disease and lymphomas. Two case reports have described a
148 patients with acute promyelocytic leukemia who were hypersensitivity-like reaction related to procarbazine
436 Effusions caused by drugs
Docetaxel Tizanidine
It has been reported that patients may develop exudative Moufarrege and associates66 reported on a patient who
pleural effusions of apparently indeterminate etiology developed a right-sided pleural effusion that was inciden-
after an average time of approximately 5 months from the tally found in a spinal computed tomography (CT), 3
first docetaxel administration.2 The effusion gradually months after the initiation of tizanidine was prescribed for
resolved over a period of several weeks after the end of the persistent post-traumatic muscle and bone pain.66 The
therapy. fluid was an exhudate with 10 percent eosinophils. Other
Antidiabetic agents 437
common etiologies were ruled out and the effusion disap- year-old woman who presented with cough, pleuritic chest
peared over the first month after the discontinuation of the pain, pink sputum, dyspnea and fever 2 months after
drug. mesalamine was first administered. She had bilateral
pleural effusions and a pulmonary infiltrate. The effusion
was said to be a ‘transudate’ but the characteristics of the
VITAMINS fluid were not reported. The disease did not respond to
empirical treatment and the symptoms finally resolved
A case of life-threatening eosinophilic pleural and pericar- only after mesalamine discontinuation. Subsequently,
dial effusion has been reported in a 76-year-old woman Trisolini et al.71 reported a patient with an eosinophilic
taking vitamins B6 and H for alopecia.67 The patient pre- pleural effusion associated with mesalamine treatment.
sented with chest pain and dyspnea, attributed to cardiac There is also another case report of a patient who devel-
temponade and pleural effusion. Peripheral blood oped a pleuropericarditis 7 months after starting treat-
eosinophilia was detected. The pleural effusion relapsed ment with the drug.72 The serological findings were
despite pericardiotomy and serial thoracenteses. thought to be suggestive of a lupus-like reaction and the
Infectious, allergic, autoimmune and malignant diseases syndrome remitted after the discontinuation of the drug.
were appropriately ruled out. Histological examination of
both the pleura and the pericardium revealed eosinophilic
infiltrate. Symptoms and eosinophilia resolved 1 week CLOZAPINE
after the withdrawal of the vitamins and the patient was
symptom-free at a 2-year follow-up visit. Clozapine, a commonly used antipsychotic agent, has been
associated with the development of pleural effusions.73–77
The pleural effusions develop within 7 days to 2 months of
ANTICOAGULANTS initial administration of the drug and are usually, but not
always,74 bilateral. Pericardial effusion73,77 and even con-
Eosinophilic pleural effusion has been reported in a strictive pericarditis77 may accompany the pleural
patient who presented with a dry cough and low-grade effusions. Pleuritic chest pain, fever, headache, skin rash or
fever, 9 months after the initiation of warfarin treatment.68 peripheral eosinophilia have also been observed in some of
A right-sided exudative pleural effusion with 57 percent these patients. In one patient, the pleural effusions were part
eosinophils, as well as blood eosinophilia, was found. No of a more generalized syndrome consisting of hepatitis,
other etiologies for the syndrome were determined after hyperglycemia and glomerular injury.75 Pleural fluid analy-
extensive evaluation. The blood eosinophilia declined after sis, reported in only one case, revealed a neutrophil
the discontinuation of the drug. When warfarin was predominant exudate.73 Symptoms and radiographic
administered again, a left-sided pleural effusion appeared abnormalities resolve within several days after the drug is
and the peripheral eosinophilia increased. Both the discontinued. When two of the above patients underwent
peripheral eosinophilia and the pleural effusions gradually rechallenge, the symptoms and the effusions recurred.73,74
resolved, after the final withdrawal of the drug.
Hemothorax can rarely happen in patients receiving
anticoagulants or intrapleural fibrinolytics.14,69 Most com- ANTIDIABETIC AGENTS
monly (9 of 12 reported cases), hemothorax occur in
patients being treated for pulmonary infarction. In all of Gliclazide
those cases, the anticoagulant used was heparin and the
complication occurred within the first week of the A 52-year-old diabetic patient was reported, who devel-
therapy.14 In another patient who was treated with oped a mild pneumonitis accompanied by an exudative
heparin, the bleeding source was pneumonia. The remain- eosinophilic (pleural fluid eosinophils were 80 percent)
ing two cases were taking warfarin and the hemothorax ipsilateral pleural effusion, accompanied by peripheral
occurred contralaterally to the pulmonary embolus. blood eosinophilia 2 weeks after beginning therapy with
Importantly, in the majority of the reported cases, the pro- gliclazide, an oral hypoglycemic agent. The peripheral
thrombin time (PT) and partial thromboplastin time blood eosinophil count was 20 percent. The effusion and
(PTT) were within ‘safe’ ranges. the blood eosinophilia resolved over the first month after
the withdrawal of gliclazide.78
MESALAMINE
Troglitazone
Mesalamine, an anti-inflammatory drug used in the treat-
ment of inflammatory bowel disease, has been associated A 47-year old man presented with cough and dyspnea
with the development of pulmonary infiltrates with or 1 week after the first dose of troglitazone.79 Small bilateral
without pleural effusion.70 There is a case report of a 72- pleural effusions were evident 2 weeks later. The symp-
438 Effusions caused by drugs
toms resolved within 24 hours after the discontinuation of will develop DIL.1,8 Pleural disease affects 30 percent of
the drug and no pleural fluid was found in a chest X-ray them, while parenchymal infiltrates are the sole respiratory
obtained 6 days later. manifestation in 5 percent. Arthralgias are the rule (85
percent) and fever is not uncommon (>40 percent), while
renal disease may be more common compared with DIL
caused by other agents.
DRUG-INDUCED LUPUS PLEURITIS
With 15 000–20 000 new patients every year in the USA, Isoniazid
drug-induced lupus (DIL) pleuritis is the most common
cause of drug-associated pleural effusion.8 While more Isoniazid (INH) may induce ANA in one-quarter of the
than 80 drugs have been associated with the syndrome, patients who take the drug, but INH-associated DIL is very
definite evidence exists for a minority of them including rare.82–84 However, an increase in the size of a tuberculous
procainamide, hydralazine, isoniazide, methyldopa, chlor- effusion or the development of a new pleural effusion 3–12
promazine, d-penicillamine and quinidine.2,80 DIL differs weeks after the initiation of anti-tuberculous treatment for
from the idiopathic SLE in that: (i) it is reversible after pulmonary or pleural tuberculosis could be due to a poorly
drug withdrawal; (ii) it has a milder clinical course, with defined ‘paradoxical response’.84–90 It is speculated that
predominant arthralgias, myalgias, rash and serositis and this phenomenon represents an excessive response of the
infrequent renal and central nervous system involvement; cell-mediated immunity to an increased antigen load
(iii) it shows no sex predominance. resulting from the mycobacterial lysis. Gupta et al.85
Pleural effusions, typically bilateral, are the most reported that pleural fluid was smear-positive in two, and
common radiographic abnormality. Pericardial effusions culture positive in four of the 29 of patients who developed
and, rarely, patchy infiltrates of acute onset may be seen. this ‘paradoxical response’.85 Pleural biopsy was negative
The patients with DIL pleuritis are usually symptomatic for mycobacteria or tuberculous histology in 15 of the 20
with fever and pleuritic chest pain.1 Arthralgias and myal- biopsies carried out. The majority of these patients (83
gias are very common.8 The pleural fluid is exudate, with percent), as well as all of the patients reported by
varying nucleated cell counts. The pleural fluid:serum others,86–90 recovered without any modification in
ANA ratio may be ≥1. The presence of LE cells in the chemotherapy.84 Hiroaka et al.84 reported two patients
pleural fluid, although unusual, strongly suggests the diag- who developed pleural effusion during anti-tuberculous
nosis. The erythrocyte sedimentation rate and serum ANA treatment for pulmonary tuberculosis.84 In both, the fluid
titers are elevated. Among ANA, the anti-double-stranded was a lymphocytic exudate with normal glucose, high ANA
DNA antibodies are very seldom detected while the anti- titer, low CH50 and low ADA, thus resembling lupus effu-
histone antibodies are suggestive, though not specific, for sion. The first patient had progressive rash and the INH
DIL.2,14 Drug withdrawal typically leads to rapid patient was replaced with pyrazinamide. The INH-containing
improvement. However, non-steroid anti-inflammatory regimen was continued in the second patient. Pleural effu-
drugs or a short course of low-dose prednisolone may be sion disappeared in both. These two cases raise a question
beneficial when a patient is markedly symptomatic. A drug about the relationship of what is thought to be a ‘paradox-
that is known to cause DIL pleuritis should never been ical response’ to anti-tuberculous treatment and INH-
given to a patient with SLE is because there is significant induced lupus pleuritis. Speaking practically, when should
risk of aggravation of the symptoms.1 the clinician continue or withdraw INH, in the face of
newly formed pleural effusion? The presence of pericardi-
tis or generalized symptoms, such as rash and arthralgias,
Procainamide as well as positive serum ANA, makes the DIL pleuritis
more likely. Furthermore, bilateral effusions are more
Approximately one-third of the patients taking the drug likely due to DIL, while all the 43 pleural effusions, previ-
will develop DIL pleuritis within the first year of the ously reported as the result of ‘paradoxical response’ were
therapy, although the disorder can appear anytime unilateral.84–90 High pleural fluid ADA or the presence of
between the first month and the twelfth year of beginning granuloma in the pleural biopsy suggest a tuberculous
treatment.8 Pleural involvement occurs in more than half, origin. Pleural fluid ANA titers are not particularly useful,
while patchy parenchymal infiltrates may be seen in 40 because 25 percent of the patients that take INH have ele-
percent of patients.81 vated ANA8 and pleural fluid ANA usually reflects the
serum titers.45 However, a negative serum ANA virtually
rules out DIL. If DIL-associated symptoms are present, the
Hydralazine INH should be changed to another effective drug. If DIL
pleuritis is not clinically probable, the regimen should be
Although half of the patients taking hydralazine have pos- continued and if the effusion is symptomatic, a therapeu-
itive ANA, only 2–20 percent, most commonly women, tic aspiration should be performed. If the effusions tend to
Other drugs 439
relapse, a low dose, short-term course of corticosteroids of BAL did not reveal any infective agent. Although one
usually hastens its resolution.84 can notice that the hemoptysis suggests the presence of an
underlying lung disease and that the results of the BAL and
the lung scan do not completely rule out pulmonary
OTHER DRUGS embolism or pneumonia, the authors reported that acy-
clovir discontinuation was followed by fever remission.
Simvastatin The pleural effusion and the pulmonary infiltrates resolved
within 10 days of the drug withdrawal.
Simvastatin is an HMG-CoA reductase blocker used in the
treatment of hypercholesterolemia. There is one case
report of a patient who developed cough, progressive Metronidazole
dyspnea and fatigue 6 months after the initiation of the
therapy. Chest X-ray disclosed bilateral interstitial infil- There is one report97 of a 42-year-old atopic woman who
trates and a moderate sized right-sided pleural effusion. developed fever, skin rash and bilateral pneumonitis and
Elevated liver function tests and serum IgE were detected. pleural effusions within a day of starting of oral metron-
BAL revealed eosinophilia of 34 percent. The pleural fluid idazole. The drug was discontinued and she recovered
was described as darkish brown but analysis was not per- completely over the next few days. Six months later, she
formed. Thoracoscopic biopsies revealed interstitial lung developed a similar syndrome, without pneumonitis, with
fibrosis and no pleural abnormalities. The discontinuation the first dose of metronidazole. Drug withdrawal again led
of simvastatin and prednisone administration led to clini- to rapid improvement of the symptoms and complete
cal improvement, normalization of the liver tests and fall recovery.
of the BAL eosinophil count.91 Roncato-Saberan et al.92
reported on a patient who presented with large bilateral
pleural effusions and weight loss while being on simvas- Valproic acid
tatin treatment for 13 years. He had peripheral and pleural
fluid eosinophilia. The pleural effusions resolved and the Valproic acid, an antiepileptic drug, may cause pleural
patient regained weight within a month of the discontinu- effusion. There is a report of a patient who developed
ation of simvastatin with no relapse during a 2-year long exudative pleural effusion 9 months after the first admin-
follow-up period. In addition to the above, simvastatin has istration of the drug.98 The pleural fluid and the blood
been reported to cause drug-induced lupus syndrome.93,94 eosinophil percentages were 62 and 26 percent, respec-
tively. The fluid gradually disappeared within several days
after the withdrawal of valproic and did not recur during a
L-Tryptophan 6-month follow-up period. However, the fact that the
pleural fluid appeared a week after the onset of ipsilateral
l-Tryptophan has been used as sleep-promoting drug. pneumonia makes it possible that it was really a post-
This drug is considered as the causal factor of vasculitis, pneumonia pleural effusion. Subsequently, eosinophilic
which is clinically expressed as eosinophilia–myalgia syn- pleural effusion and peripheral eosinophilia that resolved
drome.8 The patients present with blood eosinophilia, after valproic withdrawal was reported in another patient.
myalgias and fatigue, progressive dyspnea and cough. The More importantly, re-challenge with valproic acid pro-
chest radiographs demonstrate bilateral intersti- duced recurrent symptoms.99 Another patient who was
tial/alveolar infiltrates and pleural effusions.95 However, taking valproic acid100 developed a febrile lymphocyte pre-
pleural fluid characteristics have not been reported. Drug dominant exudative pleural effusion. Microbiology
withdrawal and corticosteroid administration were fol- studies, chest CT angiography and abdominal CT were
lowed by clinical and radiographic resolution. normal and thoracoscopic biopsy revealed non-specific
pleural inflammation. The effusion disappeared when val-
proic acid was substituted by another anticonvulsant. The
Acyclovir authors did not state how long the patient had been in val-
proic acid treatment before the onset of pleuritis. More
Acyclovir is commonly used in the treatment of herpes recently, Savvas and associates101 attributed an episode of
virus disease. There is only one report of pleuropulmonary bilateral febrile eosinophilic pleural effusions that devel-
disease attributed to this drug.96 A 71-year-old man who oped 20 days after cardiac surgery to the valproic acid that
was being treated with acyclovir for ophthalmic herpes was administered because of seizures in the early postoper-
zoster developed fever and bilateral pulmonary infiltrates, ative period. However, other etiologies, including a post-
3 days after the initiation of the treatment. The next day he traumatic pleural effusion and Dressler syndrome, were
presented with a left-sided pleural effusion and hemopty- not considered in this case. Interestingly, the disease per-
sis. A ventilation-perfusion lung scan was ‘not indicative’ sisted after the discontinuation of the valproic acid and
of pulmonary embolism and microbiological examination resolved only with systemic corticosteroids.
440 Effusions caused by drugs
Dapsone causes a hypersensitivity reaction known as As new drugs will be continuously released in the market
sulfone syndrome, consisting of fever, malaise, acute hepa- and some of the currently used ones will be administered
titis, exfoliative dermatitis and hemolytic anemia.102 to more patients, additional clinical observations concern-
Symptoms typically appear within 2–6 weeks after the insti- ing drug-induced pleural disease should be expected.
tution of dapsone therapy. There is one report where a Future reports should incorporate a thorough diagnostic
patient had a large unilateral exudative pleural effusion.102 work-up that convincingly rules out all the alternative eti-
ologies, before the disease is attributed to the suspected
drug. Future studies will also need to elucidate the patho-
D-Penicillamine genesis of the drug-induced pleural abnormalities. Given
the small number of the patients with these abnormalities,
There is only one case report of a woman who developed animal studies will be particularly useful in addressing this
an unexplained massive left-sided pleural effusion two issue. Finally, even though drug withdrawal is a convenient
years after d-penicillamine treatment was initiated for treatment option for the majority of the patients, it may
Wilson’s disease.103 Video-assisted thoracoscopic surgery not be sufficient in the settings of a life-threatening drug-
(VATS) revealed white plaques over the parietal and vis- induced syndrome. Moreover, drug discontinuation is not
ceral pleura, and the lung was encased within a thickened desirable for some medications that are uniquely active
cortex. Talc pleurodesis was unsuccessful. Eventually, she against serious diseases. For these reasons, even given the
underwent decortication and histology revealed changes of relative scarcity of patients for the conduction of random-
chronic non-specific inflammation. The drug was contin- ized-controlled studies, the role of the anti-inflammatory
ued without further sequelae. treatment, i.e. corticosteroids, should be further examined
and new treatment modalities may need to be developed.
Propylthiouracil
4. Rosenow EC III. Drug-induced bronchopulmonary pleural disease. J 27. Bolanos-Meade J, Keng YK, Cobos E. Recurrent lymphocytic pleural
Allergy Clin Immunol 1987; 80: 780–7. effusion after intravenous immunoglobulin. Am J Hematol 1999;
5. Cooper JA, White DA, Matthay RA. Drug-induced pulmonary 60: 248–9.
disease. Am Rev Respir Dis 1986; 133: 488–505. ●28. Graham JR, Suby HI, LeCompte PR, Sadowski NL. Fibrotic disorders
●6. Holmberg L, Boman G. Pulmonary reactions to nitrofurantoin: 447 associated with methysergide therapy for headache. N Engl J Med
cases reported to the Swedish adverse drug reaction committee, 1966; 274: 350–68.
1966–1976. Eur J Respir Dis 1981; 62: 180–9. 29. Kok-Jensen A, Lindeneg O. Pleurisy and fibrosis of the pleura
7. Geller M, Flaherty DK, Dickie HA, Reed CE. Lymphopenia in acute during methysergide treatment of hemicrania. Scand J Respir Dis
nitrofurantoin pleuropulmonary reactions. J Allergy Clin Immunol 1970; 51: 218–222.
1977; 59: 445–8. 30. Hindle W, Posner E, Sweetnam MT, Tan RS. Pleural effusion and
◆8. Morelock SY, Sahn SA. Drugs and the pleura. Chest 1999; 116: fibrosis during treatment with methysergide. Br Med J 1970; 1:
212–21. 605–6.
9. Martin WJ II, Rosenow EC III. Amiodarone pulmonary toxicity. ●31. Rinne UK. Pleuropulmonary changes during long-term
Chest 1988; 93: 1067–74. bromocriptine treatment for Parkinson’s disease. Lancet 1981; i:
10. Pollak PT. Clinical organ toxicity of antiarrhythmic compounds: 44.
ocular and pulmonary manifestations. Am J Cardiol 1999 84: 32. McElvaney NG, Wilcox PG, Churg A, Fleetham JA. Pleuropulmonary
37R–45R. disease during bromocriptine treatment of Parkinson’s disease.
11. Gonzalez-Rothi RJ, Hannan SE, Hood I, Franzini DA. Amiodarone Arch Intern Med 1988; 148: 2231–6.
pulmonary toxicity presenting as bilateral exudative pleural 33. Bhatt MH, Keenan SP, Fleetham JA, Calne DB. Pleuropulmonary
effusions. Chest 1987; 92: 179–82. disease associated with dopamine agonist therapy. Ann Neurol
12. Stein B, Zaatari GS, Pine JR. Amiodarone pulmonary toxicity. 1991; 30: 613–16.
Clinical, cytologic and ultrastructural findings. Acta Cytol 1987; 34. Frans E, Dom R, Demedts M. Pleuropulmonary changes during
31: 357–61. treatment of Parkinson’s disease with a long-acting ergot
13. Akoun GM, Cadranel JL, Blanchette G, Milleron BJ, Mayaud CM. derivative, cabergoline. Eur Respir J 1992; 5: 263–5.
Pleural T-lymphocyte subsets in amiodarone-associated 35. Ling LH, Ahlskog JE, Munger TM, Limper AH, Oh JK. Constrictive
pleuropneumonitis. Chest 1989; 95: 596–7. pericarditis and pleuropulmonary disease linked to ergot dopamine
◆14. Miller WT. Pleural and mediastinal disorders related to drug use. agonist therapy (cabergoline) for Parkinson’s disease. Mayo Clin
Semin Roentgenol 1995; 30: 35–48. Proc 1999; 74: 371–5.
15. Page RL. Progressive pleural thickening during oxprenolol therapy. 36. De Vuyst P, Pfitzenmeyer P, Camus P. Asbestos, ergot drugs and
Br J Dis Chest 1979; 73: 195–9. the pleura. Eur Respir J 1997; 10: 2695–8.
16. Webb DB, Whale RJ. Pleuropericardial effusion associated with ◆37. Pfitzenmeyer P, Foucher P, Dennewald G, et al. Pleuropulmonary
minoxidil administration. Postgrad Med J 1982; 58: 319–20. changes induced by ergoline drugs. Eur Respir J 1996; 9:
17. Yoshida H, Hasegawa R, Hayashi H, Irie Y. Imidapril-induced 1013–19.
eosinophilic pleurisy. Case report and review of the literature. 38. Oechsner M, Groenke L, Mueller D. Pleural fibrosis associated with
Respiration 2005; 72: 423–6. dihydroergocryptine treatment. Acta Neurol Scand 2000; 101:
18. Vogelzang PJ, Bloom SM, Mier JW, Atkins MB: Chest 283–5.
roentgenographic abnormalities in IL-2 recipients. Incidence and 39. Graham JR. Cardiac and pulmonary fibrosis during methysergide
correlation with clinical parameters. Chest 1992; 101: 746–52. therapy for headache. Am J Med Sci 1967; 254: 1–12.
●19. Saxon RR, Klein JR, Bar MH, et al. Pathogenesis of pulmonary 40. Shaunak S, Wilkins A, Pilling JB, Dick DJ. Pericardial,
edema during interleukin-2 therapy: correlation of chest retroperitoneal, and pleural fibrosis induced by pergolide. J Neurol
radiographic and clinical findings in 54 patients. AJR Am J Neurosurg Psychiatry 1999; 66: 79–81.
Roentgenol 1991; 156: 281–5. 41. Varsano S, Gershman M, Hamaoui E. Pergolide-induced dyspnea,
20. Berthianume Y, Boiteau P, Fick G, et al. Pulmonary edema during bilateral pleural effusion and peripheral edema. Respiration 2000;
IL-2 therapy: Combined effect of increased permeability and 67: 580–2.
hydrostatic pressure. Am J Respir Crit Care Med 1995; 152: 42. Knoop C, Mairesse M, Lenclud C, Gevenois PA, De Vuyst P. Pleural
329–35. effusion during bromocriptine exposure in two patients with pre-
21. Smith JW. Tolerability and side-effect profile of rhIL-11. Oncology existing asbestos pleural plaques: a relationship? Eur Respir J
(Williston Park) 2000; 14: 41s–7s. 1997; 10: 2898–901.
22. Takeda A, Ikegame K, Kimura Y, et al. Pleural effusion during 43. Kinnunen E, Viljanen A. Pleuropulmonary involvement during
interferon treatment for chronic hepatitis C. bromocriptine treatment. Chest 1988; 94: 1034–6.
Hepatogastroenterology 2000; 47: 1431–5. 44. Mohan K, Owen S, McCallum S, Magennis R. A patient with
23. Isaacs C, Robert NJ, Bailey FA, et al. Randomized placebo- tremors and breathlessness. Prim Care Respir J 2005; 14: 47–50.
controlled study of recombinant human interleukin-11 to prevent ◆45. Light RW. Pleural diseases, 4th edn. Philadelphia: Lippincott
chemotherapy-induced thrombocytopenia in patients with breast Williams & Wilkins, 2001: 265–70.
cancer receiving dose-intensive cyclophosphamide and 46. Bunker CB, Sheron N, Maurice PD, et al. Isotretinoin and
doxorubicin. J Clin Oncol 1997; 15: 3368–77. eosinophilic pleural effusion. Lancet 1989; 1: 435–6.
24. Busmanis IA, Beaty AE, Basser RL. Isolated pleural effusion with 47. Milleron BJ, Valcke J, Akoun GM, Mayaud CM. Isotretinoin-related
hematopoietic cells of mixed lineage in a patient receiving eosinophilic pleural effusion. Chest 1996; 110: 1128.
granulocyte-colony-stimulating factor after high-dose 48. Warrell RP Jr, Maslak P, Eardley A, et al. Treatment of acute
chemotherapy. Diagn Cytopathol 1998; 18: 204–7. promyelocytic leukemia with all-trans retinoic acid: an update of
25. Nakamura M, Sakemi T, Fujisaki T, et al. Sudden death or the New York experience. Leukemia 1994; 8: 929–33.
refractory pleural effusion following treatment with granulocyte ●49. Jung JI, Choi JE, Hahn ST, et al. Radiologic features of all-trans-
colony-stimulating factor in two hemodialysis patients. Nephron retinoic acid syndrome. AJR Am J Roentgenol 2002; 178: 475–80.
1999; 83: 178–9. 50. Frankel SR, Eardley A, Lauwers G, Weiss M, Warrell RP Jr. The
26. de Azevedo AM, Goldberg Tabak D. Life-threatening capillary leak ‘retinoic acid syndrome’ in acute promyelocytic leukemia. Ann
syndrome after G-CSF mobilization and collection of peripheral Intern Med 1992; 117: 292–6.
blood progenitor cells for allogeneic transplantation. Bone Marrow 51. Bini A, Zompatori M, Ansaloni L, et al. Bilateral recurrent
Transplant 2001; 28: 311–2. pneumothorax complicating chemotherapy for pulmonary
442 Effusions caused by drugs
metastatic breast ductal carcinoma: report of a case. Surg Today proteinuria occurring early in clozapine treatment. Int Clin
2000; 30: 469–72. Psychopharmacol 1998; 13: 95–8.
●52. Imokawa S, Colby TV, Leslie KO, Helmers RA. Methotrexate 76. Stanislav SW, Gonzalez-Blanco M. Papular rash and bilateral
pneumonitis: review of the literature and histopathological pleural effusion associated with clozapine. Ann Pharmacother
findings in nine patients. Eur Respir J 2000; 15: 373–81. 1999; 33: 1008–9.
53. Walden PAM, Mitchell-Heggs PF, Coppin C, et al. Pleurisy and ●77. Catalano G, Catalano MC, Frankel Wetter RL. Clozapine induced
methotrexate treatment. Br Med J 1977; 2: 867. polyserositis. Clin Neuropharmacol 1997; 20: 352–6.
54. Urban C, Nirenberg A, Caparros B, et al. Chemical pleuritis as the 78. Tzanakis N, Bouros D, Siafakas N. Eosinophilic pleural effusion due
cause of acute chest pain following high-dose methotrexate to gliclazide. Respir Med 2000; 94: 94.
treatment. Cancer 1983; 51: 34–7. 79. Koshida H, Shibata K, Kametani T. Pleuropulmonary disease in a
55. Gunstream SR, Seidenfield JJ, Sobonya RE, McMahon LJ. man with diabetes who was treated with troglitazone. N Engl J
Mitomycin-associated lung disease. Cancer Treat Rep 1983; 67: Med 1998; 339: 1400–1.
301–4. ◆80. Yung RL, Richardson BC. Drug-induced lupus. Rheum Dis Clin
56. Schaap N, Raymakers R, Schattenberg A, et al. Massive pleural North Am 1994; 20: 61–85.
effusion attributed to high-dose cyclophosphamide during 81. Cush JJ, Goldings EA. Drug-induced lupus; clinical spectrum and
conditioning for MBT. Bone Marrow Transplant 1996; 18: 247–8. pathogenesis. Am J Med Sci 1985; 290: 36–45.
57. Jones SE, Moore M, Blank N, Castellino RA. Hypersensitivity to 82. Salazar-Paramo M, Rubin RL, Garcia-De La Torre I. Systemic lupus
procarbazine (Matulane) manifested by fever and pleuropulmonary erythematosus induced by isoniazid. Ann Rheum Dis 1992; 51:
reaction. Cancer 1972; 29: 498–500. 1085–7.
58. Ecker MD, Jay B, Keohane MF. Procarbazine lung. AJR Am J 83. Guleria R, Behera D, Jindal SK. Systemic lupus erythematosus
Roentgenol 1978; 131: 527–8. during isoniazid therapy. Indian J Chest Dis Allied Sci 1990; 32:
59. Goldsby R, Pulsipher M, Adams R, et al. Unexpected pleural 55–8.
effusions in 3 pediatric patients treated with STI-571. J Pediatr ●84. Hiraoka K, Nagata N, Kawajiri T, et al. Paradoxical pleural response
Hematol Oncol 2002; 24: 694–5. to antituberculous chemotherapy and isoniazid-induced lupus.
60. Ishii Y, Shoji N, Kimura Y, Ohyashiki K. Prominent pleural effusion Respiration 1998; 65:152–5.
possibly due to imatinib mesylate in adult Philadelphia 85. Gupta RC, Dixit R, Purohit SD, Saxena A. Development of pleural
chromosome-positive acute lymphoblastic leukemia. Intern Med effusion in patients during anti-tuberculous chemotherapy:
2006; 45: 339–40. analysis of twenty-nine cases with review of literature. Indian J
●61. Breccia M, D’Elia GM, D’Andrea M, Latagliata R, Alimena G. Chest Dis Allied Sci 2000; 42: 161–6.
Pleural-pericardic effusion as uncommon complication in CML 86. Al-Ali MA, Almasri NM. Development of contralateral pleural
patients treated with Imatinib. Eur J Haematol 2005; 74: 89–90. effusion during chemotherapy for tuberculous pleurisy. Saudi Med
●62. Petusevsky ML, Faling J, Rocklin RE, et al. Pleuropericardial J 2000; 21: 574–6.
reaction to treatment with dantrolene. J Am Med Assoc 1979; 87. Puri MM, Arora VK. Contralateral pleural effusion during
242: 2772–4. chemotherapy for tuberculous pleural effusion. Med J Malaysia
63. Mahoney JM, Bachtel MD. Pleural effusion associated with chronic 2000; 55: 382–4.
dantrolene administration. Ann Pharmacother 1994; 28: 587–9. 88. Al-Majed SA.Study of paradoxical response to chemotherapy in
64. Felz MW, Haviland-Foley DJ. Eosinophilic pleural effusion due to tuberculous pleural effusion. Respir Med 1996; 90: 211–14.
dantrolene: resolution with steroid therapy. South Med J 2001; 89. Vilaseca J, Lopez-Vivancos J, Arnau J, Guardia J. Contralateral
94: 502–4. pleural effusion during chemotherapy for tuberculous pleurisy.
65. Le-Quang B, Camels P, Valayer-Caleat E, Fayolle-Minon I, Tubercle 1984; 65: 209–10.
Gautheron V. Dantrolene and pleural effusion: case report and 90. Matthay RA, Neff TA, Iseman MD. Tuberculous pleural effusions
review of literature. Spinal Cord 2004; 42: 317–20. developing during chemotherapy for pulmonary tuberculosis. Am
66. Moufarrege G, Frank E, Carstens DD. Eosinophilic exudative pleural Rev Respir Dis 1974; 109: 469–72.
effusion after initiation of tizanidine treatment: a case report. 91. De Groot RE, Willems LN, Dijkman JH. Interstitial lung disease with
Pain Med 2003; 4: 85–90. pleural effusion caused by simvastin. J Intern Med 1996; 239:
67. Debourdeau PM, Djezzar S, Estival JL, et al. Life-threatening 361–3.
eosinophilic pleuropericardial effusion related to vitamins B5 and 92. Roncato-Saberan M, Hustache-Mathieu L, Hoen B. Eosinophilic
H. Ann Pharmacother 2001; 35: 424–6. pleural effusion caused by simvastatin after 13 years of exposure.
68. Kuwahara T, Hamada M, Inoue Y, Aono S, Hiwada K. Warfarin- Eur J Intern Med 2006; 17: 450.
induced eosinophilic pleurisy. Intern Med 1995; 34: 794–6. 93. Khosla R, Butman A N, Hammer D F. Simvastatin-induced lupus
69. Bouros D, Schiza S, Siafakas N. Utility of fibrinolytic agents for erythematosus. South Med J 1998; 91: 873–4
draining intrapleural infections. Semin Respir Infect 1999; 14: 94. Bannwarth B, Miremont G, Papapietro P-M. Lupuslike syndrome
39–47. associated with simvastatin. Arch Intern Med 1992; 152: 1093–4.
70. Sesin GP, Mucciardi N, Almeida S. Mesalamine-associated pleural 95. Shore ET. L-tryptophane induced cough and pleural effusions
effusion with pulmonary infiltration. Am J Health Syst Pharm associated with the eosinophelia–myalgia syndrome. Chest 1990;
1998; 55: 2304–5. 98: 1540.
71. Trisolini R, Dore R, Biagi F, et al. Eosinophilic pleural effusion due 96. Pusateri DW, Muder RR. Fever, pulmonary infiltrates and pleural
to mesalamine. Report of a rare occurrence. Sarcoidosis Vasc effusion following acyclivir therapy for Herpes zoster opthalmicus.
Diffuse Lung Dis 2000; 17: 288–91. Chest 1990; 98: 754–6.
72. Pent MT, Anapathy S, Holdswotrth CD, et al. Mesalamine-induced 97. Kristenson M, Fryden A. Pneumonitis caused by metronidazole. J
lupus-like syndrome. Br Med J 1992; 305:159. Am Med Assoc 1988; 260: 184.
73. Daly JM, Godber RJ, Braman SS. Polyserositis associated with 98. Kaufman J, O’Shaughnessy IM. Eosinophilic pleural effusion
clozapine treatment. Am J Psychiatry 1992; 149: 1274–5. associated with valproic acid administration. South Med J 1995;
74. Chatterjee A, Saffermaan A. Cellulitis, eosinophilia and unilateral 88: 881–2.
pleural effusion associated with clozapine treatment. J Clin 99. Kavetz JD, Federman DG. Valproic acid-induced eosinophilic
Psychopharmacol 1997; 17: 323–33. pleural effusion. South Med J 2003; 96: 803–6.
75. Thompson J, Chengappa KN, Good CB, et al. Hepatitis, 100. Andre S, Drowart A, De Bels D. Lymphocytic pleural effusion
hyperglycemia, pleural effusion, eosinophilia, hematuria and associated with valproic acid. Eur J Intern Med 2005; 16: 535.
References 443
101. Savvas SP, Dimopoulou E, Koskinas J. Valproic acid-associated 104. Middleton KL, Santella R, Couser JI. Eosinophilic pleuritis due to
eosinophilic pleural effusion treated with corticosteroids. Eur J propylthiouracil. Chest 1993; 103: 955–6.
Intern Med 2006; 17: 71. 105. Gonther J, Lode H, Raffenberg M, et al. Development of pleural
102. Corp CC, Ghishan FK. The sulfone syndrome complicated by and pericardial effusions during itraconazole therapy of
pancreatitis and pleural effusion in an adolescent receiving pulmonary aspergillosis. Eur J Clin Microbiol Inf Dis 1993; 12:
dapsone for treatment of acne vulgaris. J Pediatr Gastroenterol 723–4.
Nutr 1998; 26: 103–5. 106. Bando T, Fujimura M, Noda Y, et al. Minocycline-induced
103. Karkos C, Moore A, Manche A, Thorpe JA. Pleural effusion pneumonitis with bilateral hilar lymphadenopathy and pleural
associated with D -penicillamine therapy: a case report. J Clin effusion. Intern Med 1994; 33: 177–9.
Pharm Therpeut 1996; 21: 15–17.
This page intentionally left blank
34
Effusions after surgery
were <4 mm on the decubitus film in 50 patients whereas be categorized into two groups based on time course:
it was between 4 and 10 mm in 26 patients, and >10 mm ‘early’ and ‘late’.14 Early effusions are those that occur
in 21 patients. However, a post-operative pleural effusion within 30 days of CABG surgery, and late effusions are
should be considered in cases that have an unexpected those that occur after 30 days. Light et al.16 graded these
clinical deterioration following surgery. Although not effusions between 0 and 5 based on the volume of fluid
common, large pleural effusions or infected pleural fluid within the pleural space (Table 34.2).
can cause significant clinical symptoms and require inter-
ventions such as tube thoracostomy. In a case of acute
dyspnea and a minimal pleural effusion, pulmonary Table 34.2 Quantification of size of pleural effusions after
embolus should also be considered. post-coronary artery bypass graft surgery
Hurlbut et al.20 obtained chest radiographs 8 weeks post- CABG surgery. The incidence of pleural effusion in the SV
operatively on 76 patients who had received IMA grafts, only group was 23/54 (43 percent) while the incidence in
and reported that five (9.1 percent) of 55 patients who the IMA group was 28/68 (41 percent). The difference
underwent pleurotomy and three (14.5 percent) of 21 between the two groups was not statistically significant.
patients who had an IMA without pleurotomy had a Landymore and Howell19 investigated the role of pleuro-
pleural effusion. In a recent retrospective study of 410 tomy in 106 patients, including 34 patients that underwent
patients who had undergone CABG surgery, the incidence IMA grafting with pleurotomy and 33 without pleuro-
of patients diagnosed with symptomatic newly developed tomy. They included 39 patients undergoing valve replace-
large pleural effusions from 30 days to 6 months post- ment or revascularization with SV graft as controls. The
CABG was 3.1 percent.24 patients were followed by chest X-rays that were per-
formed prior to discharge and at 3 months post-opera-
ETIOLOGY
tively. The incidence of pleural effusion prior to discharge
was 91 percent (31 out of 34) in the patients that had IMA
Several potential causative factors have been suggested for grafting with pleurotomy, 58 percent (19 out of 33) in the
early and late effusions, but their relative roles and impor- patients that had IMA grafting without pleurotomy and 31
tance have not been completely defined. The early effu- percent (12 out of 39) in the control patients. They con-
sions are usually bloody, with a median hematocrit above cluded that pleurotomy has a role in the formation of
5 percent.25 The most likely underlying mechanism for pleural effusions after surgery and that the IMA should be
these bloody effusions is trauma from the surgery and exposed without pleurotomy.
post-operative bleeding into the pleural space.25 Opening of the pericardium during surgery does not
Topical cardiac cooling with ice is another potential appear to be primarily responsible for the effusion because
pathogenetic factor for early effusions.2,26 In one retro- patients who have valve replacement have a lower preva-
spective study, pleural effusions occurred in 50 percent of lence of effusion than do patients who undergo CABG
patients who underwent topical cardiac cooling with ice surgery.7
but in only 14 and 18 percent of patients who received Patients who undergo CABG surgery off-pump
topical cold saline or no topical cardioplegia, respec- (without cardiopulmonary bypass) (OP-CABG) have been
tively.27 Similar results were reported in a different study of reported to have a lower prevalence of large pleural effu-
191 patients.26 The reason why iced slush is associated with sion 4 weeks post-operatively than do those who undergo
an increased prevalence of pleural effusions is not known, CABG surgery with cardiopulmonary bypass.32 In the
but it has been speculated cold injury to the phrenic nerve series by Mohamed et al.,32 the prevalence of effusions at
may cause atelectasis which leads to development of 30 days post-operatively that occupied more than 25
pleural effusions.28 percent of the hemithorax was only 3 percent in the OP-
In contrast to the early effusions, late effusions tend to CABG patients but was 10 percent in the CABG patients
be serous in appearance with a predominance of lympho- undergoing cardiopulmonary bypass. It is not clear
cytes in the pleural fluid.1 Kim et al.29 suggested that these whether the decreased prevalence of larger pleural effusion
late effusions might be a manifestation of the PCIS. with OP-CABG surgery is caused by less trauma, omission
However, the two main clinical manifestations of PCIS are of the iced slush hypothermia or omission of cardiopul-
chest pain and fever,30 and both of these symptoms are monary bypass, which is known to be associated with the
uncommon in patients with pleural effusions more than systemic release of inflammatory mediators.33
30 days after CABG surgery.1
Lymphocytic predominance with evidence of lympho- CLINICAL MANIFESTATIONS
cytic pleuritis found in the late effusions after CABG
surgery suggests an immunological mechanism.28 When Small, usually left-sided, pleural effusions that typically
immunohistochemical analysis is performed on the lym- encompass a radiographic area smaller than two inter-
phocytes in the pleural specimens, there is a mixed popu- costal spaces are common in the early post-operative
lation of lymphocytes with a predominance of B cells.1 course following CABG procedures, and such effusions are
However, an autoimmune etiology has not yet been usually asymptomatic.2,15 However, occasionally large
demonstrated. pleural effusions causing significant symptoms may
It is not clear whether the type of the graft or pleuro- occur.16 The primary symptom of patients with large
tomy contributes to the development of late pleural effu- pleural effusion after CABG is dyspnea. In the series of
sions after CABG surgery.14 Pleurotomy is frequently Light et al.,23 22 (75.9 percent) of 29 patients with an effu-
performed in patients undergoing placement of an IMA sion occupying more than 25 percent of the hemithorax
coronary artery graft, and late pleural effusions have been complained of dyspnea, but only three (10.3 percent)
suggested to be more common after IMA–CABG opera- complained of chest pain, and only one (3.4 percent) com-
tions compared with those who had saphenous vein (SV) plained of fever.
grafts alone.19–21,31 In the study by Peng and colleagues,15 The early pleural effusions after CABG surgery are
51 out of 122 patients developed pleural effusion after almost always small and unilateral on the left side.
Effusions after cardiac surgery 449
sion is hemorrhagic in 70 percent of cases and frankly of these four effusions were exudative, suggesting that con-
bloody in 30 percent. The pleural fluid pH and glucose are gestive heart failure was not the primary underlying cause.
normal in most cases and the predominant cells in the However, it is likely that at least some of the effusions were
effusion are polymorphonuclear leukocytes during the due to cardiac dysfunction. In their series, none of the
acute phase and mononuclear cells later in the course.40 patients had any evidence for PCIS, such as pericarditis or
pneumonitis. Other possible explanations suggested for
MANAGEMENT these pleural effusions were interruption of the lymphatics
that normally drain the pleural space, leakage of fluid from
The majority of patients with PCIS respond well to the ini- the mediastinum, damage from topical hypothermia or a
tiation of anti-inflammatory agents such as aspirin or hypersensitivity reaction to a drug.47
indomethacin.1 Corticosteroids may be used in cases unre-
sponsive to anti-inflammatory drugs.45 Pleural effusion CLINICAL MANIFESTATIONS
due to PCIS rarely requires invasive interventions such as
therapeutic thoracentesis or any further procedure. The data on the clinical manifestations of pleural effusion
after post-OHT surgery is limited to one study.47 In this
study, Misra et al. reported that pleural effusions were
Pleural effusion after heart transplantation more common after OHT than pre-transplantation,
despite the presence of severe cardiac dysfunction preop-
INCIDENCE eratively. Sixty-one of the 72 patients (85 percent) had an
effusion demonstrated at some point in the 12 months fol-
Nearly 2300 heart transplants are performed each year in lowing transplantation. Ten of the 72 (14 percent) had a
the USA.16 While the prevalence of pleural effusions after unilateral effusion on the left while only two (3 percent)
CABG surgery and other cardiac surgeries has been well had a unilateral effusion on the right. Forty-nine patients
established, only two studies to date have described the (68 percent) developed an effusion on each side at least
prevalence of pleural effusions in the post-orthotopic heart once in the first 12 months after transplantation, and all of
transplant (OHT) population.18,46 Lenner and associates18 these patients also had concurrent bilateral pleural effu-
retrospectively evaluated the frequency of all pulmonary sions noted at some point during that same time period.47
complications occurring after 159 orthotopic heart trans- The majority of the post-operative effusions were
plantation surgery. They reported 81 pulmonary compli- small.47 However, eight patients (13 percent) had 10 effu-
cations in 47 heart recipients. Pleural effusions were sions occupying from 25 up to 50 percent of the hemitho-
accepted as complication of the surgery if seen on a chest rax, and seven of those effusions were left-sided. Two
radiograph or if an intervention, such as diagnostic thora- different patients had an effusion occupying 50 percent or
centesis, diuresis, antibiotic therapy or tube thoracostomy, more of the hemithorax. Effusions were largest at median
was required. They reported effusions in only 10 (6.7 post-operative day 6.5 (5.0–7.0) on the left and 6.0 (5.0–
percent) subjects. 9.0) on the right, expressed as mean day and 25th to 75th
The first study evaluating specifically the prevalence of percentiles. The effusions tended to resolve with time.47
pleural effusions post-orthotopic heart transplantation
was done in the authors’ institution.47 In this study, 81 CHARACTERISTICS OF THE EFFUSIONS
patients who underwent OHT were screened by reviewing
chest radiographs and chest computed tomography (CT) Four patients with fluid studies were available for review in
scans, and 72 patients were included in the study. Sixty- the study by Misra et al.47 The fluid samples were obtained
one patients (85 percent) developed an effusion at some at a variety of times during the post-operative period, from
time during the first 365 post-operative days.47 post-operative day 12 to post-operative day 128. Two of
the four effusions were exudates with LDH levels of
ETIOLOGY 385 IU/L and 1623 IU/L (upper normal limit for LDH is
220 IU/L). The protein level was measured in only one of
The reported prevalence of post-OHT effusions is similar these exudates and it was 3.8 g/dL.47
to that seen in the post-CABG patient population.47 Misra
et al.47 hypothesized that the effusions post-OHT result MANAGEMENT
from the actual surgical procedure of entering the chest
cavity and disrupting the pleural space. While it is also It appears from the very limited data that effusions after
possible that the occurrence of pleural effusions in the post-OHT tend to resolve with time and no intervention is
post-OHT population could be related to induction or necessary for the management of these effusions. In the
maintenance immunosuppressive regimens, it seems less study by Misra et al.47 only eight patients (13 percent) had
likely given the similarity in occurrence to the post-CABG 10 effusions occupying from 25 up to 50 percent of the
population. hemithorax, and two different patients had an effusion
In the study by Misra et al.,47 biochemical analysis for occupying 50 percent or more of the hemithorax.
the pleural fluid was available for only four patients. Two Effusions were largest at median post-operative day 6.5
Pleural effusion after liver transplantation 451
(5.0–7.0) on the left and 6.0 (5.0–9.0) on the right, istics in nine single lung transplant recipients. Ipsilateral
expressed as mean day and 25th to 75th percentile.47 pleural fluid occurred immediately following the surgery
and continued for up to 9 days. The effusions were exuda-
tive in all patients. The fluid was bloody in appearance
PLEURAL EFFUSION AFTER LUNG with a high number of red blood cells (331 627 ± 122 583
TRANSPLANTATION mm3, mean ± SEM). The white blood cell count in the
fluid was 9803 ± 3470/mm3 and decreased steadily after
Incidence the first post-operative day. The initial LDH and protein
values were approximately 3000 IU/L and 3.3 ± 0.2 g/dL,
Pleural effusion occurs in almost all lung transplant recip- respectively. These values decreased by day nine.
ients in the early post-operative period (within 9 days).6 Shtrit et al.46 reported the characteristics of pleural
Chiles et al.6 reported a 100 percent incidence in 10 fluids after single lung transplantation in seven patients.
patients after heart–lung transplantation. Judson et al.48 All the effusions were of medium size, with a median fluid
reported that nine patients underwent unilateral lung volume of 700 mL (range 100–1300). Pleural fluids were
transplantation. All of these patients developed a pleural exudative with lymphocyte predominance in all patients.
effusion. These effusions are usually bloody, exudative, In one patient, eosinophils were noted and in two patients,
neutrophil predominant and usually small to moderate in a bloody fluid was observed. LDH levels ranged between
size. They tend to resolve spontaneously within 9 days of 322 and 560 IU/L (normal upper limit for LDH =
transplantation.5 There are only two reports on late (more 220 IU/L), and protein levels ranged between 4.9 and
than 14 days) pleural effusions following lung transplanta- 6.3 g/dL.
tion.46,49
Herridge et al.49 reviewed the pleural complications of
unilateral and bilateral lung transplantation. They Management
reported that none of the 53 single-lung transplant recipi-
ents had pleural effusions while 14 of 91 (15 percent) Because these effusions are usually either self-limited or
double lung transplant recipients did so. The effusions easily treated systemically, it is seldom necessary to either
consisted of empyemas in seven patients, parapneumonic examine or drain small to moderate pleural fluid collec-
effusions that resolved spontaneously in four, hemothorax tions in the first 10 days after transplantation.48 When new
in two and chylothorax in one. Shitrit et al.46 investigated or persistent pleural effusions are observed beyond the
the characteristics of late pleural effusions occurring 14–45 first few weeks after transplantation, the following etiolo-
days in patients who underwent lung transplantation. gies should be considered: empyema or parapneumonic
Seven out of 35 patients (20 percent) had late pleural effu- effusion, acute rejection, organizing pleural hematoma,
sion. The median time for the appearance of pleural effu- lymphoproliferative disorder and cardiac or renal failure.51
sion was 23 days (range 14–34 days). It has been our A diagnostic thoracentesis should be performed to help
observation that patients who experience rejection fre- identify the etiology of the effusion.
quently had pleural effusions.
◆17. Light RW. Pleural effusions following cardiac injury and coronary 38. Snow M, Agatston, AS, Kramer, HC, Samet, P The postcardiotomy
artery bypass graft surgery. Semin Respir Crit Care Med 2001; 22: syndrome following transvenous pacemaker insertion. Pacing Clin
657–64. Electrophysiol 1987; 10: 934.
●18. Lenner R, Padilla ML, Teirstein AS, et al. Pulmonary complications 39. Dressler W. The post-myocardial-infarction syndrome: a report on
in cardiac transplant recipients. Chest 2001; 120: 508–13. forty-four cases. AMA Arch Intern Med 1959; 103: 28–42.
19. Landymore RW, Howell F. Pulmonary complications following 40. Stelzner TJ, King TE Jr, Antony VB, et al. The pleuropulmonary
myocardial revascularization with the internal mammary artery manifestations of the postcardiac injury syndrome. Chest 1983;
graft. Eur J Cardiothorac Surg 1990; 4: 156–61; discussion 161–2. 84: 383–7.
20. Hurlbut D, Myers ML, Lefcoe M, et al. Pleuropulmonary morbidity: 41. Engle MA, Zabriskie JB, Senterfit LB, et al. Postpericardiotomy
internal thoracic artery versus saphenous vein graft. Ann Thorac syndrome. A new look at an old condition. Mod Concepts
Surg 1990; 50: 959–64. Cardiovasc Dis 1975; 44: 59–64.
21. Vargas F, Cukier A, Hueb W, et al. Relationship between pleural 42. Spodick DH. Decreased recognition of the post-myocardial
effusion and pericardial involvement after myocardial infarction (Dressler) syndrome in the postinfarct setting: does it
revascularization. Chest 1994; 105: 1748–52. masquerade as ‘idiopathic pericarditis’ following silent infarcts?
●22. Daganou M, Dimopoulou I, Michalopoulos N, et al. Respiratory Chest 2004; 126: 1410–11.
complications after coronary artery bypass surgery with unilateral 43. Spodick D. The pericardium: a comprehensive textbook. New York:
or bilateral internal mammary artery grafting. Chest 1998; 113: Dekker, 1997.
1285–9. 44. Welin L, Vedin A, Wilhelmsson C. Characteristics, prevalence, and
●23. Light R, Rogers J, Moyers J, et al. Prevalence and clinical course of prognosis of postmyocardial infarction syndrome. Br Heart J 1983;
pleural effusions at 30 days after coronary artery and cardiac 50: 140–5.
surgery. Am J Respir Crit Care Med 2002; 166: 1567–71. 45. Gregoratos G. Pericardial involvement in acute myocardial
●24. Peng M, Hou C, Li J, et al. Prevalence of symptomatic large pleural infarction. Cardiol Clin 1990; 8: 601–8.
effusions first diagnosed more than 30 days after coronary artery 46. Shitrit D, Izbicki G, Fink G, et al. Late postoperative pleural
bypass graft surgery. Respirology 2007; 12: 122–6. effusion following lung transplantation: characteristics and
●25. Sadikot R, Rogers J, Cheng D, et al. Pleural fluid characteristics of clinical implications. Eur J Cardiothorac Surg 2003; 23: 494–6.
patients with symptomatic pleural effusion after coronary artery ●47. Misra H, Dikensoy O, Rodriguez RM, et al. The prevalence of
bypass graft surgery. Arch Intern Med 2000; 160: 2665–8. pleural effusions post-orthotopic heart transplantation.
26. Nikas D, Ramadan F, Elefteriades J. Topical hypothermia: Respirology 2007; 12: 887–90.
ineffective and deleterious as adjunct to cardioplegia for ●48. Judson MA, Handy JR, Sahn SA. Pleural effusions following lung
myocardial protection. Ann Thorac Surg 1998; 65: 28–31. transplantation. Time course, characteristics, and clinical
27. Allen B, Buckberg G, Rosenkranz E, et al. Topical cardiac implications. Chest 1996; 109: 1190–4.
hypothermia in patients with coronary disease. An unnecessary 49. Herridge MS, de Hoyos AL, Chaparro C, et al. Pleural complications
adjunct to cardioplegic protection and cause of pulmonary in lung transplant recipients. J Thorac Cardiovasc Surg 1995; 110:
morbidity. J Thorac Cardiovasc Surg 1992; 104: 626–31. 22–6.
◆28. Light R. Pleural effusions after coronary artery bypass graft 50. Rodriguez RM MJ, Rogers JT, et al. Prevalence and clinical course
surgery. Curr Opin Pulm Med 2002; 8: 308–11. of pleural effusion at 30 days post coronary artery bypass surgery.
29. Kim Y, Mohsenifar Z, Koerner S. Lymphocytic pleural effusion in Chest 1999, 116: 282S.
postpericardiotomy syndrome. Am Heart J 1988; 115: 1077–79. 51. Marom EM, Palmer SM, Erasmus JJ, et al. Pleural effusions in lung
30. Mott A, Fraser CJ, Kusnoor A, et al. The effect of short-term transplant recipients: image-guided small-bore catheter drainage.
prophylactic methylprednisolone on the incidence and severity of Radiology 2003; 228: 241–5.
postpericardiotomy syndrome in children undergoing cardiac ●52. Afessa B, Gay PC, Plevak DJ, et al. Pulmonary complications of
surgery with cardiopulmonary bypass. J Am Coll Cardiol 2001; 37: orthotopic liver transplantation. Mayo Clin Proc 1993; 68: 427–34.
1700–6. 53. Spizarny DL, Gross BH, McLoud T. Enlarging pleural effusion after
31. Jain U, Rao T, Kumar P, et al. Radiographic pulmonary liver transplantation. J Thorac Imaging 1993; 8: 85–7.
abnormalities after different types of cardiac surgery. J 54. Olutola PS, Hutton L, Wall WJ. Pleural effusion following liver
Cardiothorac Vasc Anesth 1991; 5: 592–5. transplantation. Radiology 1985; 157: 594.
32. Mohamed KH JT, Rodriguez RM, Light RW. Pleural effusions post 55. Costello P, Williams CR, Jenkins RW, et al. The incidence and
coronary artery bypass grafting performed with or without a implications of chest radiographic abnormalities following
bypass pump [Abstract]. Am J Respir Crit Care Med 2001; 163: orthotopic liver transplantation. Can Assoc Radiol J 1987; 38:
A901. 90–5.
33. Ganapathy S, Murkin J, Dobkowski W, et al. Stress and 56. Jiang Y, Lv LZ, Cai QC, et al. Liver transplant for 70 patients with
inflammatory response after beating heart surgery versus end-stage liver diseases. Hepatobiliary Pancreat Dis Int 2007; 6:
conventional bypass surgery: the role of thoracic epidural 24–8.
anesthesia. Heart Surg Forum 2001; 4: 323–7. 57. Golfieri R, Giampalma E, Sama C, et al. Pulmonary complications
34. Cohen M, Sahn S. Resolution of pleural effusions. Chest 2001; following orthotopic liver transplant: radiologic patterns and
119: 1547–62. epidemiologic considerations in 100 cases. Rays 1994; 19:
35. Paull D, Delahanty T, Weber F, et al. Thoracoscopic talc pleurodesis 319–38.
for recurrent, symptomatic pleural effusion following cardiac 58. Duran FG, Piqueras B, Romero M, et al. Pulmonary complications
operations. Surg Laparosc Endosc Percutan Tech 2003; 13: following orthotopic liver transplant. Transpl Int 1998; 11(Suppl
339–44. 1): S255–9.
36. Jerjes-Sanchez C, Ibarra-Perez C, Ramirez-Rivera A, et al. Dressler- 59. Pirat A, Ozgur S, Torgay A, et al. Risk factors for postoperative
like syndrome after pulmonary embolism and infarction. Chest respiratory complications in adult liver transplant recipients.
1987; 92: 115–17. Transplant Proc 2004; 36: 218–20.
37. Tabatznik B, Isaacs JP. Postpericardiotomy syndrome following 60. Bilik R, Yellen M, Superina RA. Surgical complications in children
traumatic hemopericardium. Am J Cardiol 1961; 7: 83–96. after liver transplantation. J Pediatr Surg 1992; 27: 1371–5.
This page intentionally left blank
35
Hepatic hydrothorax
ETIOLOGY
Incidence
It is now broadly accepted that hepatic hydrothorax is sec-
Hepatic hydrothorax accounts for 2–3 percent of all ondary to transfer of peritoneal fluid directly via defects in
pleural effusions.6–10 The incidence increases to 9.9 percent the diaphragm: this mechanism was suggested by the
when only massive pleural effusions are considered.10 In observation of pneumothorax after injection of air into the
patients with cirrhosis the incidence of hepatic hydrotho- peritoneal cavity, and confirmed by higher radioactivity in
rax is low but depends on the severity of the cirrhosis and pleural fluid than in lymph or plasma after injection of
the presence of ascites. In a recent report of 862 Chinese albumin labeled with iodine-131 into ascitic fluid. Many
cirrhotic patients,11 132 (15 percent) had pleural effusion other studies have proved that the injection of air, dyes or
although in most of them the effusion was only detectable radiolabeled material into the peritoneal cavity of patients
by ultrasonography, and thoracentesis could only be per- with hydrothorax is associated with the rapid movement
formed in 56 (6.5 percent). The four larger series of cir- of these materials into the pleural space.1,22
rhotic patients with ascites,12–15 including a total of 1155 Many otherwise normal people probably have tiny con-
cases, show a very similar incidence of hepatic hydrotho- genital holes in the diaphragm. In patients with ascites, the
456 Hepatic hydrothorax
increasing abdominal pressure and the diaphragmatic The diagnosis of hepatic cirrhosis is mainly histological
thinning secondary to malnutrition of cirrhotic patients but, in general practice, the diagnosis is based on clinical,
enlarges the defects. Blebs of herniated peritoneum pro- analytical and ultrasonographic findings. The presence of
trude through these defects and, if the blebs burst, a com- ascites favors the diagnosis but its absence does not rule
munication between peritoneal and pleural spaces is out hepatic hydrothorax.
formed. There is a pressure gradient between peritoneal To exclude a primary cardiac, pulmonary or pleural
and pleural spaces that favors the unidirectional passage of disease, a chest radiograph and thoracentesis with pleural
ascitic fluid into the chest. A valvular mechanism may con- fluid analysis should be performed.
tribute to this unidirectional flow. The congenital A chest radiograph should always be obtained to
diaphragmatic holes are frequently seen in the tendinous confirm the effusion, rule out cardiomegaly and pul-
portion of the right diaphragm and less frequently in the monary or pleural pathology. In patients with massive
left diaphragm, which is thicker and more muscular than pleural effusion the radiograph should be repeated when
the right. This fact, plus the piston effect of the liver, is why the effusion has decreased considerably (after diuresis or
most hepatic hydrothoraces are right-sided. The diaphrag- therapeutic thoracentesis) to evaluate pulmonary or
matic defects can also be secondary to previous trauma or pleural pathology that was masked by the effusion,18
surgery. Huang et al.23 have classified the diaphragmatic although this also can be evaluated by a thoracic scan.
defects associated with the development of hepatic Thoracentesis and pleural fluid study should be per-
hydrothorax on thoracoscopy into four morphological formed despite the appearance of a normal chest radi-
types: type I, no obvious defect; type II, blebs lying on the ograph, because diagnostic thoracentesis with a fine needle
diaphragm; type III, broken defects (fenestrations) in the (21 G) has very few complications and pleural fluid analy-
diaphragm and type IV, multiple gaps in the diaphragm. sis is the best option to rule out pleural pathology and to
In patients without ascites the mechanism of formation detect the main complication of hepatic hydrothorax that
is the same. In these cases all the ascitic fluid formed is SBEM (the bacterial infection of a pre-existing
rapidly crosses the diaphragm and becomes pleural fluid. hydrothorax). In 139 consecutive diagnostic thoracentesis
This has been confirmed by scintigraphic studies.24–26 performed with a fine needle, the most severe complica-
tion was pneumothorax that appeared in two cases (1.3
percent).28 Despite a severe coagulation disorder in some
patients, the incidence of bleeding complications was small
CLINICAL PRESENTATION (4 percent) and without clinical impact.28 This observation
confirms a previous report of the absence of increased
Hepatic hydrothorax should be suspected when a cirrhotic
bleeding after paracentesis and thoracentesis in patients
patient, especially with ascites, develops a pleural effusion.
with coagulation abnormalities.29 There is a wide experi-
It should also be suspected in patients with an isolated
ence of paracentesis in cirrhotic patients with coagulation
right pleural effusion with transudative characteristics,
abnormalities,30 and thoracentesis, if performed correctly
especially if it is massive. Hepatic hydrothorax can be
avoiding the intercostal vessels, is a procedure very similar
asymptomatic, or present with symptoms that can range
to paracentesis.
from dyspnea on exertion to overt respiratory failure,
In a study of 60 consecutive cirrhotic patients admitted
depending on various factors such as the volume of the
to the hospital with pleural effusion,18 only 42 (70 percent)
effusion, the amount of ascites present, the speed of the
were considered to have uncomplicated hepatic hydrotho-
accumulation of pleural fluid and the presence of associ-
rax based on pleural fluid analysis. Of the other 18 (30
ated pulmonary disease, which is not infrequent in alco-
percent), nine had spontaneous bacterial empyema, two
holic patients. Life-threatening dyspnea secondary to an
had pleural tuberculosis, two had adenocarcinoma, two
acute hepatic hydrothorax has been reported, probably
had parapneumonic empyema and three had undiagnosed
secondary to a sudden increase in intra-abdominal pres-
exudates. In those cases in which the diagnosis was other
sure, such as straining or coughing that caused rupture of
than hydrothorax, the chest radiograph was only able to
the pleuroperitoneal bleb.27
diagnose the patients with parapneumonic empyema.
When the pleural effusion was right-sided, 37/46 (80
percent) were uncomplicated hydrothorax. Conversely,
DIAGNOSIS when it was left-sided only, 5/14 (35 percent) were
hydrothorax and none of the five patients with left pleural
The diagnosis of hepatic hydrothorax is based on: effusion and absence of ascites had uncomplicated hepatic
hydrothorax. Furthermore, according to this study, diag-
1. Presence of hepatic cirrhosis with portal hypertension; nosis other than hydrothorax often cannot be suspected by
2. Exclusion of a primary cardiac, pulmonary or pleural ascitic fluid analysis only.18 This concurs with previous
disease; reports20,31 that the presence of a pleural effusion in a cir-
3. Eventual confirmation of the passage of ascites to the rhotic patient should not automatically lead to the diagno-
pleural space. sis of hepatic hydrothorax, especially if the effusion is
Treatment 457
left-sided. Pleural fluid analysis is mandatory and may add pleural fluid amylase (in alcoholic patients) and adenosine
valuable information to ascitic fluid analysis for patients deaminase (ADA), especially in countries where tubercu-
with ascites and pleural effusion. This approach has been losis is prevalent, should be considered when the fluid is an
recently emphasized by Light32 who concluded in a recent exudate or when the effusion is left-sided. In this last case,
review of pleural effusion that ‘a thoracentesis should be thoracic computed tomography (CT) can help to rule out
performed in patients with a pleural effusion of unknown pulmonary or pleural pathology.
case unless the effusion is small (less than 10 mm on ultra- The best way to confirm the communication between
sonography) or the patient has congestive heart failure and pleural and peritoneal space is with a scintigraphic study,
bilateral pleural effusions’. although there has been a report of cases of hepatic
Pleural fluid of hepatic hydrothorax has low protein and hydrothorax in which the communication between peri-
lactate dehydrogenase (LDH) concentration, very similar toneal and pleural space was detected by ultrasonography
to those of ascitic fluid, and thus it has transudative charac- or magnetic resonance.38 These studies are not necessary
teristics,33 although pleural fluid total proteins are greater for the diagnosis of hepatic hydrothorax in clinical prac-
than in ascites.34 Light’s criteria remain the best means of tice. It can be used for confirming a hydrothorax when
separating transudates from exudates.35 However, it should pleural fluid has exudative characteristics, or if is planned
be pointed out that a small number of patients with tran- to close the communications between peritoneal and
sudative pleural effusions are misclassified as having pleural spaces with video-assisted thoracoscopic surgery
exudates when applying Light’s criteria, mostly because (VATS).
total pleural fluid protein increases with diuresis.36 In these
cases, when hydrothorax is suspected and exudative crite-
ria are met, a serum albumin gradient should be calculated TREATMENT
(serum albumin minus pleural fluid albumin), and if it is
above 12 g/L the effusion probably is a transudate,35,37 Hepatic hydrothorax is secondary to the passage of ascites
although studies including a higher number of cirrhotic through a diaphragmatic defect, and ascites is secondary to
patients should be carried out to confirm the utility of portal hypertension and salt retention by the kidney.
albumin gradient in the diagnosis of hepatic hydrothorax. Treatment can be directed to improve salt retention
In conclusion, the most advisable approach is to (diuretics), to reduce portal hypertension (transjugular
perform a chest radiograh, paracentesis if the patient has intrahepatic portosystemic shunt, TIPS), to close the
ascites and thoracentesis in all patients with suspected diaphragmatic defects (through VATS, with concomitant
hepatic hydrothorax. Pleural fluid analysis should include talc pleurodesis) or to solve hepatic cirrhosis by liver trans-
protein and LDH determinations, polymorphonuclear plantation. Therapeutic options and their relationship
count and culture in a blood culture bottle, in order to with the pathophysiological cascade of hepatic hydrotho-
confirm that the fluid is transudate (applying Light crite- rax are shown in Table 35.1.
ria), to rule out SBEM and to rule out other causes of The first line of treatment is the same as for ascites:
pleural effusion.3,32 A smear to exclude malignancy, sodium restriction and diuretic therapy, spironolactone,
Table 35.1 Therapeutic options depending on the physiopathological steps of hepatic hydrothorax
Therapeutic options
and if there is no response to sequential increments or pneumothorax, but also to evaluate pulmonary or pleural
complications such as hyperkalemia, the addition of pathology that was masked by the effusion. Re-expansion
furosemide. In case of intolerance to spironolactone due to pulmonary edema, a very rare complication of thoracente-
gynecomastia, triamterene or amiloride may be consid- sis, is associated with previous pulmonary disease and
ered. evacuation of a high pleural fluid volume in a short period
Those that do not respond to medical therapy are con- of time.43 In a series of 76 therapeutic thoracenteses in cir-
sidered to have refractory hydrothorax (RH), that can be rhotic patients we did not have any post-evacuation pul-
defined, using the same criteria for refractory ascites intro- monary edema. It has been suggested that intravascular
duced in a Consensus Conference in 1996,39 as a pleural expansion with albumin may prevent this complication.41
effusion that cannot be mobilized (with diuretics, paracen- In any case, evacuation of more than 2000 mL of pleural
tesis or thoracentesis) or the early recurrence of which fluid and use of vacuum is not advisable. Thus, for patients
cannot be satisfactorily prevented by medical therapy. with refractory hydrothorax whose hepatic function is not
Similar definitions have been given recently: pleural effu- expected to improve spontaneously, as occurs after
sion requiring repeated thoracenteses despite treatment variceal bleeding, bacterial infection or recent alcohol
with the highest tolerable doses of spironolactone and intake, repeated thoracenteses are not the treatment of
furosemide17 or recurrent pleural effusion that, despite salt choice and alternatives should be investigated.
restriction and diuretic therapy, requires repeated thera- Chest tube insertion should be avoided because it does
peutic thoracenteses to control symptoms.40 The incidence not solve the problem and produces massive fluid loss that
of refractory hydrothorax is not clearly established. In a can lead to the death of the patient.44 In a retrospective
retrospective study by our group,14 405 cirrhotic patients analysis, mortality of cirrhotic patients undergoing a chest
admitted over a 5 year period with ascites were studied. tube placement mostly for hepatic hydrothorax and spon-
Seven out of 27 patients with hepatic hydrothorax (26 taneous bacterial empyema was 40 percent in patient with
percent) were considered refractory to medical treatment. Child–Turcotte–Pugh (CTP) C and 16 percent in CTP B.45
In a prospective study of 60 cirrhotic patients with hepatic If a chest tube is placed, appropriate replacement of fluid
hydrothorax,34 13 (22 percent) were considered refractory. losses and albumin of 6–8 g/L of pleural fluid removed,
Thus, we can estimate that the prevalence of refractory should be administered. It has been anecdotally reported
hepatic hydrothorax is 25 percent. that the concomitant administration of vasoconstrictor
Management of refractory hydrothorax is a clinical chal- drugs (as usually used in hepatorenal syndrome) can
lenge because most cirrhotic patients with RH have severely decrease the output from the chest tube (Table 35.1).46,47
impaired liver function, frequently with associated renal Pleural sclerosis with tetracycline or talc has been used
insufficiency, and aggressive therapy in these very fragile in the past. The irritant is administered by a chest tube or
patient results in high morbidity and mortality rates. With thoracoscopy. Positive results were achieved in less than 50
the exception of liver transplantation, no single therapy has percent of patients, although figures are better in the few
been shown to be completely satisfactory because of associ- cases without ascites.48 Failure is caused by the continuous
ated morbidity or low efficacy. Controlled trials are lacking passage of ascitic fluid from abdominal cavity resulting in
and therapeutic decisions must therefore be taken on the the irritant dilution. The results may be improved if irri-
basis of anecdotal reports or small series of patients. tant is administered by VATS.49,50 To decrease this flood by
Thoracentesis is the most effective method for the rapid decreasing the gradient pressure between pleural and peri-
relief of dyspnea associated with hepatic hydrothorax. toneal cavity, continuous positive airway pressure (CPAP)
Owing to the fact that paracentesis has almost no side has been used in association with chemical pleurodesis51 or
effects41 and can improve dyspnea by decreasing intra- alone.52
abdominal pressure, in patients with ascites and hydrotho- Peritoneovenous shunt was used in the past for refrac-
rax it is advisable to drain ascites first and only perform tory ascites and also for hepatic hydrothorax.19 Problems
thoracentesis if there is no ascites or paracentesis does not associated with the shunt, such as obstruction or infection
ameliorate the dyspnea. Thoracentesis relieves patient and lack of efficacy, limit its use. Anecdotal use of other
symptoms easily, with few complications, except pneumo- shunts, such as the pleurovenous shunt, has been
thorax.18 If performed correctly the risk of bleeding is reported.53
small,5,42 even without administering fresh frozen plasma. Closure of transdiaphragmatic fenestrations can be per-
Protein depletion can be overcome by albumin adminis- formed by VATS with concomitant talc pleurodesis.54–56
tration. However, the risk of pneumothorax is not negligi- Mouroux et al.54 found demonstrable diaphragmatic
ble. In 19 patients that were treated with 76 therapeutic defects in six of eight patients (75 percent). The main
thoracentesis, five patients had seven pneumothoraces (9 duration of post-operative chest tube drainage was 7.6
percent of the procedures, 26 percent of the patients). days for the patients with demonstrable fenestrations and
Although there was no mortality associated with pneu- 16.5 days in the two patients without demonstrable pleural
mothorax, a chest tube had to be inserted because of fenestrations. The pleural effusion did not recur in the six
dyspnea in four patients. A chest radiograph after thera- patients with fenestrations closed, but the other two
peutic thoracentesis is advisable not only for detection of patients had recurrent effusion and died 1 month and
Treatment 459
2 months after the procedure, respectively. In the series Transjugular intrahepatic portosystemic shunt is an
reported by de Campos et al.55 which included 18 patients, interventional radiology technique that creates an anasto-
the main duration of the post-operative chest tube mosis between portal and hepatic veins that behaves as a
drainage was 13 days (range 4–38 days). Two patients side-to-side portocaval shunt. It decompresses the hepatic
developed empyema and six developed hyponatremia and and splanchnic vascular bed, causing the portal pressure
hypoalbuminemia through chest tube drainage. Only in to fall. Introduced for the treatment of variceal bleeding,
five patients could a suture be performed; three had a good it also works for ascites as portal hypertension is a prereq-
response (60 percent), one died of post-operative pneu- uisite for ascites formation. It is useful for refractory
monia and liver failure and the other had empyema and ascites and for refractory hydrothorax. The main advan-
drained fluid for 1 month. Ten of the 21 procedures, (47.6 tages of TIPS are the lack of post-operative complications
percent) had a good response. Seven out of the 18 patients related to laparotomy, it is effective in patients without
died 12–40 days after the procedure. Successful thoraco- ascites,58 hepatic function may improve in long-term sur-
scopic pleura or Mersilene mesh onlay reinforcement of vivors and it does not preclude a future liver trans-
diaphragm in 10 cirrhotic patients with hepatic hydrotho- plantation.
rax has been reported recently.57 Although it can be successfully performed in patients
In conclusion, VATS with concomitant talc pleurodesis with high post-operative risk, one of its main drawbacks is
for refractory hydrothorax is successful in 40–75 percent impairment of liver function due to the reduction of effec-
of patients, but may result in prolonged hospitalization tive portal perfusion to the liver. This impairment can be
and pleural space intubation with severe secondary effects dramatic for patients with poor liver function in whom it
(hyponatremia and empyema), and considerable mortal- is contraindicated. Other problems are that it may induce
ity. Patients with demonstrable diaphragmatic defects or worsen hepatic encephalopathy. In addition, in most
treated with closure have better results than those without cases, stenosis or occlusion of the stents occurs in the
demonstrable fenestration. In patients with refractory medium or long term, although these complications are
ascites and refractory hydrothorax, closure of fenestration decreased with the new covered stents.
does not solve the refractory ascites. Its main indication To date there have been seven reports of refractory
are those patients with refractory hydrothorax without hepatic hydrothorax treated by TIPS (Table 35.2). The first
ascites or with low volume of ascites, or those with rela- three series16,59,60 included 41 patients and reported a high
tively good hepatic function in which TIPS is contraindi- mortality. However, the last reports achieved a consider-
cated because of advanced age, although more experience ably better survival. Chalasani et al.40 reported a survival of
is needed to confirm its role in the management of refrac- 80 percent at 6 months and 56 percent at 1 year. Factors
tory hydrothorax. influencing survival were bilirubin concentration, variceal
Table 35.2 Results of transjugular intrahepatic portosystemic shunt (TIPS) in refractory hydrothorax
hemorrhage requiring emergent TIPS, encephalopathy In the cases prospectively studied in our center,4,5,18 53
unrelated to bleeding previous to TIPS and an alanine percent were associated with spontaneous bacterial peri-
aminotransferase (ALT) level of greater than 100 IU/L. In tonitis, 30 percent had no ascites and 17 percent had non-
2001, Siegerstetter et al.17 reported a 1-year survival of 64 infected ascites. In the Chinese series, SBEM were
percent. The probability of being free of relapse was only associated with SBP in 56 percent of cases.11 Other cases of
35 percent, although in most cases revision of TIPS solved SBEM without infected ascites have been reported,66–68
the problem. Hydrothorax response and survival showed a suggesting that SBEM is not necessarily secondary to spon-
significant inverse correlation with age over 60 years. taneous bacterial peritonitis. Probably, pathogens arrive in
Spencer et al.61 reported a complete response of 63 percent the pleural space through a bacteremia, in most cases orig-
but with an early mortality of 29 percent and Nuñez et al.62 inating in the gut, the same pathogenesis reported in spon-
reported 60 percent encephalopathy. taneous bacterial peritonitis,65 although spreading of an
In conclusion, in selected patients with refractory infected ascitic fluid through diaphragmatic defects cannot
hepatic hydrothorax and a conserved hepatic function be excluded in some cases. As SBEM can be present
(especially those younger than 60 years old, with low levels without a simultaneous SBP, both paracentesis and thora-
of bilirubin and absence of encephalopathy), TIPS can be centesis should be performed when an infection is sus-
an effective treatment. It has been suggested that a ‘model pected in a cirrhotic patient with ascites and hydrothorax.
of end-stage liver disease’ (MELD) <17 could be a good The fact that a thoracentesis is a procedure with low mor-
predictor of 3-month survival after TIPS.63 It can be used bidity supports this approach.
as a bridge to hepatic transplantation or as definite therapy Risk factors for developing SBEM have been defined.
in those in whom hepatic transplantation is contraindi- Sese et al.34 found that low pleural protein and C3 levels
cated. and higher CTP score were associated with SBEM. Chen et
Hepatic transplantation is the best treatment for al.11 identified pleural fluid protein <12 g/L and a simulta-
decompensated hepatic cirrhosis and therefore for neous SBP as independent risk factors for the development
patients with hydrothorax. Most series reported few cases of SBEM.
but the results are the same as those of patients with Clinical manifestations of SBEM are shown in Table
ascites.16,40 It is indicated in refractory hydrothorax but 35.3. It is noteworthy that thorax-related symptoms are
also in patients with hydrothorax and poor hepatic func- scanty and thus, a high index of suspicion and a prompt
tion (CTP B or C) or after an episode of SBEM. In a retro- diagnostic thoracentesis should be performed to diagnose
spective study of our group we compared 28 patients with SBEM when non-specific signs or symptoms are present
hepatic hydrothorax with 56 cirrhotic patients trans- such as renal insufficiency or hepatic encephalopathy.
planted without hydrothorax.64 There were no differences Criteria for diagnosis include: positive pleural fluid culture
in peri- or post-operative complications, including trans- and a pleural fluid polymorphonuclear count greater than
fusions, days of mechanical ventilation, intensive care unit 250 cells/mL, and exclusion of parapneumonic infection;
(ICU) admission days or post-operative mortality. There evidence of pleural fluid before infection or transudate
were no differences in either long-term survival or graft characteristics during infection, and chest radiograph or
survival. In conclusion, liver transplantation is the best CT scan without evidence of pneumonia.4,5 Culture-nega-
therapy for patients with hepatic hydrothorax; there are tive SBEM is diagnosed when patients have a compatible
no further complications related to pleural effusion and clinical course, negative pleural fluid culture and a poly-
long-term survival is similar to that in patients without morphonuclear count greater than 500 cells/mL. Culture of
hydrothorax. pleural fluid should be carried out by inoculating 10 mL of
pleural fluid into a blood culture bottle at the bedside
because a causal bacteria is identified in 75 percent using
this method while this rate is only 33 percent when con- Because hydrothorax is secondary to ascites, the new
ventional microbiological techniques are used.5,11 treatments for ascites can be applied to hepatic hydro-
During the infectious episode, pleural fluid characteris- thorax. The introduction of selective vasoconstrictors to
tics may change, although protein and glucose levels stay revert the splanchnic vasodilatation responsible of ascites
stable, LDH increases and in some cases the pleural fluid in cirrhotic patients and the use of aquaretics may be
can be defined as an exudate by Light’s criteria due to LDH future alternatives.72 Video-assisted thoracoscopy with
increment.5 It has been suggested that the serum–pleura concomitant talc pleurodesis is useful therapeutically in
albumin gradient >12 g/L accurately differentiate between the patients in which a communication between peritoneal
hepatic hydrothorax or SBEM and other exudative pleural and pleural cavity is detected. Identification of these
effusions.37 patients before intervention can considerably improve the
In the four series reported including 65 cases,4,5,11,66 34 results of this technique.
were culture negative and 31 (48 percent) were culture
positive, although etiological bacteria could be identified
in 39 cases (60 percent): Escherichia coli in 21 cases,
Streptococcus species in six, Klebsiella pneumoniae in five,
Enterococcus species in four, Clostridium perfringens in one,
Pseudomonas stutzeri in one and Gemella morbillorum in
one. Due to the delay in culture results, pleural polymorph- KEY POINTS
onuclear neutrophil (PMN) count is the cornerstone of
the diagnosis. Recently, the use of reagent strips to detect ● The presence of a pleural effusion in a cirrhotic
leukocyte esterase in urine has been shown to be a very patient should not automatically lead to the diag-
sensitive and specific method for a rapid diagnosis of SBP nosis of hepatic hydrothorax, especially if the
in cirrhotic patients with ascites69 and has also been vali- effusion is left-sided. Because diagnostic thora-
dated in pleural fluid to diagnose SBEM.70 centesis is a safe procedure in patients with cir-
Mortality in the four series was 29 percent (19 from 65). rhosis, pleural fluid analysis is mandatory and
Intravascular expansion with albumin improves survival may add valuable information to ascitic fluid
in cirrhotic patients with SBP,71 although this approach analysis for patients with ascites and pleural
has not been validated in patients with SBEM, it should effusion.
improve prognosis because an impairment of systemic cir- ● Patients with hydrothorax and poor hepatic
culation after SBEM should be expected. function or those with refractory hydrothorax
Treatment with a third-generation cephalosporin, cefo- should be considered for liver transplantation.
taxime or ceftriaxone, should be initiated without waiting ● Transjugular intrahepatic portosystemic shunt is
for the pleural fluid culture when pleural fluid polymor- an effective treatment for refractory hydrothorax
phonuclear count is over 250 cells/mL. Aminoglycosides of selected patients with relatively conserved
are contraindicated in cirrhotic patients because of signifi- hepatic function, especially those with low levels
cant side effects, especially renal insufficiency. None of the of bilirubin and absence of encephalopathy. It
patients reported was treated with a chest tube, because can be used as a bridge to hepatic transplantation
pleural fluid did not meet the biochemical criteria for its or as a definitive therapy in those in whom liver
insertion. Since most patients were cured of the infection transplantation is contraindicated.
without a chest tube and its insertion can be harmful in ● Video-assisted thoracoscopy with concomitant
patients with hepatic hydrothorax,44,45 a chest tube should talc pleurodesis can be an alternative, especially
not be used for treating SBEM. Owing to frequent recidi- when no refractory ascites is associated.
vism of the infection, prophylaxis with norfloxacin is rec- ● When a cirrhotic patient with pleural effusion is
ommended to all survivors of an episode of SBEM. admitted to hospital or presents with fever, chills,
Long-term survival of patients with SBEM is poor. In a ret- encephalopathy or abdominal or chest pain,
rospective series from our hospital, median survival of cir- spontaneous bacterial empyema should be sus-
rhotic patients with SBEM was 13 months; thus, a SBEM pected. Thoracentesis with pleural fluid poly-
episode should be considered an indication for liver trans- morphonuclear count and culture, inoculating
plantation.5 10 mL of pleural fluid to a blood culture bottle at
bedside, should be performed in order to rule
our SBEM, even when spontaneous bacterial
peritonitis has been excluded. Antibiotic therapy
FUTURE DIRECTIONS with third-generation cephalosporins should be
initiated without awaiting pleural fluid culture
The utility of albumin gradient in the diagnosis of hepatic when the pleural fluid PMN count is over 250
hydrothorax has not been studied and should be investi- cells/mL.
gated before recommending its application.
462 Hepatic hydrothorax
REFERENCES 23. Huang PM, Chang YL, Yang CY, Lee YC. The morphology of
diaphragmatic defects in hepatic hydrothorax. J Thorac Cardiovasc
Surg 2005; 130: 141–5.
● = Key primary paper 24. Rubinstein D, Mclnnes IE, Dudley FJ. Hepatic hydrothorax in the
◆ = Major review article absence of clinical ascites: diagnosis and management.
∗ = Paper that represents the first formal publication of a Gastroenterology 1985; 88: 188–91
management guideline 25. Mentes BB, Kayhan B, Gorgul A, Unal S. Hepatic hydrothorax in
the absence of ascites. Report of two cases and review of the
mechanism. Dig Dis Sci 1997; 42: 781–8.
●1. Alberts WM, Salem AJ, Solomon DA, Boyce G. Hepatic 26. Kataki S, Yoshinaga YT, Takayama OHT, et al. Hepatic hydrothorax
hydrothorax. Cause and management. Arch Intern Med 1991; 151: in the absence of ascites. Liver 1998; 18: 216–20.
2383–8. 27. Castellote J, Gornals J, López C, Xiol X. Acute tension hydrothorax:
◆2. Lazaridis KN, Frank JW, Krowka MJ, Kamath PS. Hepatic a life threatening complication of cirrhosis. J Clin Gastroenterol
hydrothorax: pathogenesis, diagnosis and management. Am J Med 2002; 34: 588–9.
1999; 107: 262–7 28. Castellote J, Xiol X, Cortés-Beut R, et al. Complicaciones de la
3. Xiol X, Guardiola J. Hepatic hydrothorax. Curr Opin Pulm Med toracentesis en pacientes cirróticos con derrame pleural. Rev Esp
1998; 4: 239–42. Enferm Dig 2001; 93: 566–70.
●4. Xiol X, Castellote J, Baliellas C, et al. Spontaneous bacterial 29. McVay PA, Toy PTC. Lack of increased bleeding after paracentesis
empyema in cirrhotic patients: analysis of eleven cases. and thoracentesis in patients with mild coagulation abnormalities.
Hepatology 1990; 11: 365–70. Transfusion 1991; 31: 164–71.
5. Xiol X, Castellvi JM, Guardiola J, et al. Spontaneous bacterial 30. Runyon BA. Paracentesis of ascitic fluid. A safe procedure. Arch
empyema in cirrhotic patients: a prospective study. Hepatology Intern Med 1986; 146: 2259–61.
1996; 23: 719–23. 31. Mirouze D, Juttner HU, Reynolds TB. Left pleural effusion in
6. Tinney WS, Olsen AM. The significance of fluid in pleural space: a patients with chronic liver disease and ascites. Dig Dis Sci 1981;
study of 274 cases. Proc Mayo Clin 1945; 20: 81–7. 26: 984–8.
7. Leuallen EC, Carr DT. Pleural effusion: a statistical study of 436 32. Light R. Pleural effusion. N Engl J Med 2002; 346: 1971–7.
patients. N Engl J Med 1955; 255: 79–83. 33. Sahn S. The pleura. Am Rev Respir Dis 1988; 138: 184–234
8. Light RW, MacGregor I, Luchsinger PC, Ball WC. Pleural effusions: ●34. Sese E, Xiol X, Castellote J, Rodríguez-Fariñas E, Tremosa G. Low
the diagnostic separation of transudates and exudates. Ann Intern complement levels and opsonic activity in hepatic hydrothorax. Its
Med 1972; 77: 507–13. relationship with spontaneous bacterial empyema. J Clin
9. Romero S, Candela A, Martin C, et al. Evaluation of different Gastroenterol 2003; 36: 75–7.
criteria for the separation of pleural transudates from exudates. 35. Light RW. Diagnostic principles in pleural disease. Eur Respir J
Chest 1993; 104: 399–404. 1997; 10: 476–81.
10. Jimenez D, Díaz G, Gil D, et al. Etiology and prognostic 36. Romero S, Fernandez C, Martin C, et al. Influence of diuretics on
significance of massive pleural effusions. Respir Med 2005; 99: the concentration of proteins and other components of pleural
1183–7. transudates in patients with heart failure. Am J Med 2001; 110:
11. Chen TA, Lo GH, Lai KH. Risk factors for spontaneous bacterial 681–6.
empyema in cirrhotic patients with hydrothorax. J Chin Med Assoc 37. Tremosa G, Xiol X, Rodríguez E, et al. Valor del gradiente de
2003; 66: 579–85. albúmina en el diagnóstico diferencial de los derrames pleurales
12. Jhonston RF, Loo RV. Hepatic hydrothorax. Ann Intern Med 1964; asociados a cirrosis hepática (abst). Gastreonterologia y
61: 385–401. Hepatologia 2002; 25: 113–14.
13. Lieberman FL, Peters RL. Cirrhotic hydrothorax. Arch Intern Med 38. Zenda T, Miyamoto S, Murata S, Mabuchi H. Detection of
1970; 125: 114–17. diaphragmatic defect as the cause of severe hepatic hydrothorax
14. Esteve M, Xiol X, Fernadez F, et al. Treatment and outcome of with magnetic resonance. AJR Am J Roentgenol 1998; 93:
hydrothorax in liver cirrhosis. J Clin Nutr Gastroenterol 1986; 1: 2288–9.
139–44. 39. Arroyo V, Gines P, Gerbes AL, et al. Definition and diagnostic
15. Giacobbe A, Facciorusso D, Barbano F, Andriulli A, Frusciante V. criteria of refractory ascites and hepatorenal syndrome in
Hepatic hydrothorax. Diagnosis and management. Clin Nucl Med cirrhosis. Hepatology 1996; 23: 164–73.
1996; 21: 56–60. 40. Chalasani N, Clark WS, Martin LG, et al. Determinants of mortality
16. Jeffries MA, Kazanjian S, Wilson M, Punch J, Fontana RJ. in patients with advanced cirrhosis after transjugular intrahepatic
Transjugular intrahepatic portosystemic shunts and liver portosystemic shunting. Gastroenterology 2000; 118: 138–44.
transplantation in patients with refractory hepatic hydrothorax. 41. Runyon BA. Paracentesis of ascitic fluid. A safe procedure. Arch
Liver Transpl Surg 1998; 4: 416–23. Intern Med 1996; 124: 816–20.
17. Siegerstetter V, Deibert P, Ochs A, et al. Treatment of refractory 42. McVay PA, Toy PTC. Lack of increased bleeding after paracentesis
hepatic hydrothorax with transjugular portosystemic shunt: long and thoracentesis in patients with mild coagulation annormalities.
term results in 40 patients. Eur J Gastroenterol Hepatol 2001: 13: Transfusion 1991; 31: 164–71.
529–34. 43. Trachiotis GD, Vricella LA, Aaron BL, Hix WR. As originally
18. Xiol X, Castellote J, Cortes-Beut R, et al. Usefulness and published in 1988: Reexpansion pulmonary edema. Updated 1997.
complications of thoracentesis in cirrhotic patients. Am J Med Ann Thorac Surg 1997; 4: 1206–7.
2001; 111: 67–9. 44. Runyon BA, Greenblatt M, Ming HC. Hepatic hydrothorax is a
19. LeVeen HH, Piccone VA, Hutto RB. Management of ascites with relative contraindication to chest tube insertion. Am J
hydrothorax. Am J Surg 1984; 148: 210–13. Gastroenterol 1986; 7: 566–7.
20. Ackerman Z, Reynolds TB. Evaluation of pleural fluid in patients 45. Liu LU, Haddadin HA, Bodian CA, et al. Outcome analysis of
with cirrhosis. J Clin Gastroenterol 1997; 25: 619–22. cirrhotic patients undergoing chest tube placement. Chest 2004;
21. Kinasewitz GT. Transudative effusions. Eur Respir J 1997; 10: 126: 142–8.
714–18. 46. Dumortier J, Lepetre J, Scalone O, et al. Succesful treatment of
22. Strauss RM, Boyer TD. Hepatic hydrothorax. Sem Liv Dis 1997; 17: hepatic hydrothorax with octreotide. Eur J Gastroenterol Hepatol
227–32. 2000; 12: 817–20.
References 463
47. Ibrisim D, Cakaloglu Y, Akyuz F, et al. Treatment of hepatic symptomatic cirrhotic hydrothorax. Am J Gastroenterol 1994; 89:
hydrothorax with terlipressin in a cirrhotic patient. Scand J 1520–2.
Gastroenterol 2006; 41: 862–5. 60. Gordon FD, Anastopoulos HT, Crenshaw W, et al. The successful
48. Cantó A, Gonzalez A, Moya J, et al. La pleurodesis con talco en los treatment of symptomatic refractory hepatic hydrothorax with
derrames pleurales masivos y recidivantes de las cirrosis hepáticas. transjugular intrahepatic portosystemic shunt. Hepatology 1997;
Arch Bronconeumol 1989; 25: 256–8. 25: 1366–9.
49. Ferrante D, Arguedas M, Cerfolio RJ, et al. Video-assisted 61. Spencer EB, Cohen DT, Darcy MD. Safety and efficacy of
thoracoscopy surgery with talc pleurodesis in the management of transjugular intrahepatic portosystemic shunt creation for the
symptomatic hepatic hydrothorax. Am J Gastroenterol 2002; 97: treatment of hepatic hydrothorax. J Vasc Interv Radiol 2002; 13:
3172–5. 385–90.
50. Cerfolio RJ, Bryant AS. Efficacy of video-assisted thoracocscopic 62. Nuñez O, Garcia A, Rincón D, et al. Derivación portosistémica
surgery with talc pleurodesis for porous diaphragm syndrome in percutanea intrahepática como tratamiento del hidrotórax
patients with refractory hepatic hydrothorax. Ann Thorac Surg hepático refractario. Gastroenterol Hepatol 2002; 25: 143–7.
2006; 82: 457–9. 63. Salerno F, Merli M, Cazzaniga M, et al. MELD score is better than
51. Drohuin F, Fischer D, Law Koune D, et al. Traitement des Child–Pugh score in predicting 3-month survival of patients
hydrothorax des cirrhoses par pleurodèse chimique associeé à une undergoing transjugular intrahepatic portosystemic shunt. J
ventilation avec pression positive continue. Résultats Hepatol 2002; 36: 494–500.
préliminaires. Gastroenterol Clin Biol 1991; 15: 271–2. ●64. Xiol X, Tremosa G, Castellote J, et al. Liver transplantation in
52. Takahashi K, Chin K, Sumi K, et al. Resistant hepatic hydrothorax: patients with hepatic hydrothorax. Transpl Int 2005; 18: 672–5.
a succesful case with treatmeny by nCPAP. Respir Med 2005; 99: ◆65. Strauss RM, Boyer TD. Diagnosis and management of cirrhotic
262–4. ascites. In: Zakim D, Boyer TD (eds). Hepatology: a textbook of liver
53. Hadsaitong D, Suttithawil W. Pleurovenous shunt in treating disease. Philadelphia: WB Saunders Co, 1996: 764–88.
refractory nonmalignant hepatic hydrothorax: a case report. Respir 66. Chesta J, Ponichik J, Brahm J, et al. Spontaneous bacterial pleuritis
Med 2005; 99: 1603–5. in patients with liver cirrhosis. Rev Med Chile 1991; 119: 295–8.
∗54. Mouroux J, Perrin C, Venissac N, Blaive B, Richelme H. 67. Malnick SDH, Somin M, Zimchoni SO, Sthoeger ZM. Spontaneous
Management of pleural effusion of cirrhotic origin. Chest 1996; bacterial empyema in a patient with hepatitis C virus cirrhosis and
109: 1093–6. sterile ascitic fluid. Clin Infect Dis 1996; 23: 834–5.
55. De Campos JRM, Filho LO, Werebe EC, et al. Thoracoscopy and talc 68. Abba A, Laajam MA, Zargar SA. Spontaneous neutrocytic hepatic
poudrage in the management of hepatic hydrothorax. Chest 2000; hydrothorax without ascites. Respir Med 1996; 90: 631–4.
118: 13–17. 69. Castellote J, Lopez C, Gornals J, et al. Rapid diagnosis of
56. Kirsch CM. Cirrhotic hydrothorax and the law of unintended spontaneous bacterial peritonitis by use of reagent strips.
consequences. Chest 2000; 118: 2–4. Hepatology 2003; 37: 893–6.
57. Huang PM, Kuo SW, Lee JM. Thoracoscopic diaphragmatic repair 70. Castellote J, Lopez C, Gornals J, et al. Use of reagent strips for the
for refractory hepatic hydrothorax: application of pleural flap and diagnosis of spontaneous bacterial empyema. J Clin Gastroenterol
mesh onlay reinforcement. Thorac Cardiov Surg 2006; 54: 47–50. 2005; 39: 278–81.
58. Andrade RJ, Martin-Palanca A, Fraile JM, et al. Transjugular 71. Sort P, Navasa M, Arroyo V, et al. Effect of intravenous albumin on
intrahepatic portosystemic shunt for the management of hepatic renal imparment and mortality in patients with cirrhosis and
hydrothorax in absence of ascites. J Clin Gastroenterol 1996; 22: spontaneous bacterial peritonitis. N Eng J Med 1999; 341: 403–9.
305–7 72. Costello-Boerrigter LC, Boerrigter G, Burnett JC Jr. Revisiting salt
∗59. Strauss RM, Martin LG, Kaufman SL, Boyer TD. Transjugular and water retention: new diuretics, aquaretics and natriuretics.
intrahepatic portal systemic shunt for the management of Med Clin North Am 2003; 87: 475–91.
This page intentionally left blank
36
Effusions caused by gastrointestinal disease
CHARLIE STRANGE
the pancreatitis abates. They are invariably gone in 2 obtained. Although serum amylase is elevated in this con-
months or less.2 These effusions are exudates with a neu- dition, it may be only mildly so, since the majority of
trophil predominance and can have significant numbers of ductal drainage is released into the pleural space. Pleural
red blood cells. Pleural fluid amylase can be normal but as fluid from pseudocysts or pancreatic fistulae is always
disease progresses usually becomes greater than two times exudative with a neutrophil-rich fluid. Bloody effusions
the blood amylase.3,4 Left-sided effusions are more can be associated with pleural fluid eosinophilia.12
common than bilateral or unilateral right effusions. Fistulae between the pancreatic body and the pleural
The presence of a pleural effusion has recently transi- space release large amounts of amylase-rich pancreatic
tioned from being a marker of disease to becoming a key fluid into the pleural space. The effusions are often large,
indicator of the severity of pancreatitis. Effusions are more filling one-half of the hemithorax. Right-sided pancreato-
commonly seen with pancreatic necrosis (Balthazar’s score pleural effusions have been described with fistulae that tra-
D and E).5 By including the presence or absence of a verse the mediastinum where the esophagus penetrates the
pleural effusion, a newly proposed EPIC scoring system diaphragm before entry into the pleural space through the
(ExtraPancreatic Inflammation on CT score)6 has superior mediastinal pleura. Large bilateral effusions also have been
receiver operating characteristic scores to predict morbid- seen.13
ity and mortality than other scoring systems that grade the It should be remembered that other conditions cause
severity of pancreatic injury alone. Although the effusions pleural fluid to have high amylase levels. Benign condi-
in acute pancreatitis are usually small, their importance is tions include esophageal rupture, tuberculosis and cirrho-
large since systemic inflammation, diaphragmatic inflam- sis.14 Rarely, pleural fluid amylase may be elevated in
mation, associated pulmonary atelectasis or early acute pneumonia.15 Malignant effusions metastatic from many
respiratory distress syndrome (ARDS) are the proposed organs have been described that produce salivary amylase.
mechanisms for pleural fluid formation. In the setting of amylase-rich peritoneal fluid, the possibil-
The second form of pancreatic pleural effusion occurs ity of ovarian carcinoma or ectopic pregnancy should not
with disruption of the pancreatic duct. Most commonly, be forgotten.16
this abnormality presents with chronic pancreatitis with Successful therapy of pancreatic fistulae with or without
ductal stenosis or stones but can also occur after pancreatic pseudocyst has most readily been obtained by 2–4 weeks of
surgery, intraductal tumors7 or trauma. Anterior fluid col- pancreatic rest, hyperalimentation and serial thoracentesis
lections may rupture through the omentum and cause that is successful approximately 50 percent of the time.17,18
pancreatic ascites. Posterior rupture places fluid in the No specific therapy is necessary for the pleural space.
retroperitoneal space where it can reach the pleural space. Endoscopic management by stenting of the pancreatic duct
Importantly, these fistulae continue to advance until pan- is possible in some cases.19 Successful treatment series using
creatic output is slowed, pancreatic ductal drainage is re- nasopancreatic drains placed by ERCP have been reported.
established, or the area of abnormality is surgically This modality allows for serial imaging by fistulograms
resected. Pancreatograms performed during endoscopic through the nasal side of the tube.20,21 Ligation of the
retrograde cholangiopancreatography (ERCP) can fistula,22 radiation of the fistula23 and subtotal pancreatec-
demonstrate these fistulae and on occasion the contrast tomy have been used in some case series.
material can reflux into the pleural space. As would be Pancreatic abscess is a rare complication of acute pan-
expected, the amylase concentrations in the pleural fluid creatitis that occurs in approximately 4 percent of cases.
are extremely high with this condition, often reaching Secondary infection of a pseudocyst can produce a simi-
values greater than 100 000 IU/mL. larly severe condition. Delays in abscess drainage may be
Occasionally, a pancreatic pseudocyst migrates toward fatal. Pleural effusions occurred in 39 percent of a series of
the diaphragmatic surface without rupture. Although 63 patients with pancreatic abscess and contributed to sig-
intact pseudocysts can migrate into the mediastinum, or nificant morbidity and mortality in these cases.24
rarely into the pleural space and present as mass lesions,
more commonly the cyst leaks or ruptures into the sub-
diaphagmatic, mediastinal or pleural space. Rare ruptures SPLENIC DISEASES
into the airways have been described.8,9 Because the pleural
space is under negative pressure, migration of pseudocyst Splenic infarctions, splenic hematomas and splenic
fluid into the pleural space is often the path of least resist- abscesses often produce pleural effusions. These effusions
ance. Because the pseudocyst fluid is extremely caustic and are always left-sided and often accompanied by pleuritic
impaired pancreatic drainage creates a continuous supply pain. The effusions are usually small, neutrophil-rich exu-
of pancreatic fluid, conditions are ripe for fistula forma- dates that require no specific therapy. Yet, recognition of
tion. the underlying diagnosis does make a difference. Rare pre-
A pleural effusion can be the initial manifestation of a sentations with tumors of the spleen, epidermoid cysts,25
pancreatopleural fistula.10 Interestingly, many patients splenic vein thrombosis26 and primary rupture of the
have few if any abdominal symptoms.11 For that reason, spleen with amyloidosis have been associated with pleural
the diagnosis can be missed if a pleural fluid amylase is not effusions.
Esophageal rupture 467
Splenic abscess is a rare condition that occurs in a may rupture the esophagus as may occur following a
variety of situations. Clostridium perfringens abscesses have Heimlich maneuver40 or automobile accidents.
been described in patients with sickle cell anemia and The gastroesophageal junction lies adjacent to the
intravenous drug use.27 Other causes have included mediastinal pleura. Esophageal rupture from any cause
trauma, infection in contiguous areas and endocarditis. may allow esophageal fluids access to the mediastinum.
Rare splenic abscesses with tuberculosis28 have been Since the esophageal mucosa is not sterile, the introduc-
described. tion of bacteria into the mediastinum and pleural space is
Splenic hematomas usually are formed from trauma. associated with mediastinitis and empyema, respectively.
Yet, pleural effusions associated with a subphrenic Although some case series have suggested that mediastini-
hematoma may be distant from the time of trauma.29 tis without pleural rupture carries a better prognosis, the
Some patients may not volunteer the trauma history since difference in time to recognition and initiation of appro-
minor trauma can cause splenic hematomas. Rarely, priate therapy is likely the most important factor in this
splenic subcapsular hematomas from pancreatitis, associ- observation. Most of the morbidity of esophageal perfora-
ated with pleural effusions have been described.30 Pleural tion occurs as a consequence of infectious mediastinitis,
fluid may or may not be hemorrhagic in this condition and usually with a combination of aerobic and anaerobic
usually resolves spontaneously over 2 weeks while the pathogens.41 When pleural effusions develop, they usually
hematoma can persist for a longer time.31 need drainage since most qualify as empyemas.
Splenic infarctions result from vascular obstruction. Pleural fluid characteristics reflect the origin of the fluid.
The splenic arteries are single end arteries without signifi- Pleural fluid amylase is elevated and, when fractionated, is
cant collateral circulation. This anatomical arrangement of salivary origin. A rare case of normal pleural amylase in a
makes the spleen particularly susceptible to sickle cell patient with Sjögren’s syndrome has been described.42
occlusion that may occur in unusual settings such as in Pleural fluid pH is low even if gastric pH is pharmacologi-
sickle cell heterozygotes at altitude.32 Therapeutic splenic cally raised. This suggests that the low pH is from products
infarction by embolization has been used to control the of bacterial metabolism and polymorphonuclear neutrophil
hypersplenism of cirrhosis. When 80 percent of the spleen (PMN) sequestration in the pleural space.43 Effusions can be
is infarcted, pleural effusions are common.33 Splenic large and either unilateral or bilateral; isolated right effu-
infarction associated pleural effusions are usually small. sions are uncommon. When the etiology of pleural fluid is
Other causes of splenomegaly may also be associated unclear, cytology for foodstuff can be helpful.44
with pleural effusions. Castleman’s disease, sarcoidosis Diagnosis should be made as quickly as possible. Any
and lymphomas would fit in this category. history of retching or esophageal procedures should
prompt evaluation for esophageal rupture if any pul-
monary symptoms are present. Corticosteroids may alter
ESOPHAGEAL RUPTURE the clinical presentation by preventing fever and chest
pain.45 An upright chest radiograph should evaluate
Esophageal rupture is usually associated with trauma, whether pleural fluid is present and whether air is in the
often from endoscopy or surgery. Spontaneous esophageal mediastinum. Plain radiographs often give hints about the
rupture is a rare disease usually seen after protracted vom- site of esophageal perforation. Left hydropneumothorax is
iting but may be seen as a complication of esophageal car- commonly seen with perforation of the lower esophagus.
cinoma, perforation of Barrett’s ulcer, intensive radiation Right-sided effusions usually occur after perforation of the
therapy, esophageal Crohn’s disease,34 herpetic esophagi- mid-esophagus.46 Plain films have been recorded as
tis35 or tuberculous esophagitis.36 Delayed recognition of normal in 12 percent of cases in one series.46
this disease is common, accounting for significant morbid- Computerized tomography is more sensitive and specific
ity and mortality. for mediastinal fluid collections and mediastinal air.47
Iatrogenic rupture following dilation procedures for The diagnostic test of choice is an esophagram. Small
esophageal stricture is the most common cause of esophageal perforations can only be detected by contrast-
rupture.37 Other endoscopic procedures in which pleural enhanced swallowing studies. In larger perforations, oper-
effusions have been described include the removal of ative planning can be aided by site-specific information.
foreign bodies and sclerosis of esophageal varices.38 The An initial esophagram should be performed with a water-
hallmark symptom of chest pain that is persistent for a few soluble contrast agent since some esophageal leaks are
hours after the procedure should be followed by a contrast quite large. If negative, a barium swallow should follow the
esophagram to establish the diagnosis, the discontinuation water-soluble contrast study. The higher radiodensity of
of oral feeding and the administration of antibiotics. the barium allows for more sensitive detection of
Traumatic or post-surgical esophageal perforation esophageal perforations. In one series of suspected
occurs from a variety of esophageal insults. Surgeries for esophageal perforation, 15 percent of perforations were
esophageal carcinoma are prone to leakage at the primary missed by a water-soluble contrast agent.48
anastamotic site; occasionally the esophagus is injured Therapy of this disorder remains controversial. For per-
during other thoracic operations.39 Rarely, blunt trauma forations that are small and recognized at the time of
468 Effusions caused by gastrointestinal disease
may be the presenting manifestation of disease.68 Cases of Effusions also accompany some rare infections of the
amebic empyema may result from rupture of liver abscess liver. Hepatic hemorrhage occurs from dengue fever72
into the pleural space. where effusions are more common in severe disease.73,74
Pyogenic liver abscesses can occur with many organ- Pleural effusions have also been noted in Hantavirus infec-
isms although Escherichia coli is the most frequent tions whose severity is correlated with gallbladder wall
pathogen. Antibiotics alone are associated with mortality thickness.75 Effusions have also been seen with chronic
as high as 95 percent. Percutaneous drainage of abscesses active Epstein–Barr virus infections that cause
under CT imaging allows for culture and improves sur- hepatomegaly and gallbladder wall thickening.76 Non-
vival.69 infectious causes of inflammatory hepatic effusions
Pleural fluid analysis usually reveals a neutrophil pre- include those following transhepatic catheter ablations of
dominant exudate without infectious organisms present. hepatomas.77
Effusions are typically small but may become large if the
abscess ruptures into the pleural space creating an
empyema. GENERALIZED APPROACH TO THERAPY
Biliopleural fistula PMN predominant exudate 1000–15 000 rarely PF = serum PF/serum bilirubin >1 while fistula
pus is open; effusion becomes non-
specific exudate when fistula closes
Esophageal rupture Exudative, usually with >10 000 occasionally Elevated Morbidity secondary to empyema.
pus or positive PF cultures pus Pneumothorax seen in up to 25%.
Surgical correction of esophageal
leak often needed
Hepatitis Mononuclear predominant <1000 PF = serum Rare, spontaneously resolves.
exudate Consider unusual causes of hepatitis
if effusion is significant in size
Pancreatitis PMN predominant exudate 1000–50 000 >2¥ serum Occasionally hemorrhagic. No
specific therapy needed. Always
resolves within 2 months
Pancreatic fistula PMN predominant exudate <10 000 Often >100 000 ERCP and abdominal CT indicated to
determine site of fistula.
Nasopancreatic catheter may allow
serial fistulograms
Splenic infarction PMN predominant exudate <10 000 PF = serum Usually associated with pleuritic
chest pain. Occasionally
hemorrhagic. Spontaneously resolves
Upper abdominal and PMN predominant exudate 1000–50 000 PF = serum Anchovy paste pleural effusion
hepatic abscesses occasionally pus has been described with amebic
abscesses. Contrast-enhanced CT
essential for site-specific
localization of pathology
Abbreviations: CT, computed tomography; ERCP, endoscopic retrograde cholangiopancreatography; IU, international units; PF, pleural fluid; PMN,
polymorphonuclear neutrophil; WBC, white blood cell count.
470 Effusions caused by gastrointestinal disease
23. Greenwald RA, Deluca RF, Raskin JB. Pancreatic–pleural fistula: esophagus: correlation of site and cause with plain film findings.
demonstration by endoscopic retrograde AJR Am J Roentgenol 1985; 145: 537–40.
cholangiopancreatography (ERCP) and successful treatment with 47. White CS, Templeton PA, Attar S. Esophageal perforation: CT
radiation therapy. Dig Dis Sci 1979; 24: 240–4. findings. AJR Am J Roentgenol 1993; 160: 767–70.
24. Miller TA, Lindenauer SM, Frey CF, Stanley JC. Pancreatic abscess. 48. Tanomkiat W, Galassi W. Barium sulfate as contrast medium for
Arch Surg 1974; 108: 545–50. evaluation of postoperative anastomotic leaks. Acta Radiol 2000;
25. Caillet B, Hakim F, Perol M, Bayle JY, Guerin JC. [A rare cause of 41: 482–5.
pleurisy from subdiaphragmatic origin: epidermoid cyst of the ◆49. Kotsis L, Kostic S, Zubovits K. Multimodality treatment of
spleen]. Rev Pneumol Clin 1990; 46: 114–5. esophageal disruptions. Chest 1997; 112: 1304–9.
26. Warren MS, Gibbons RB. Left-sided pleural effusion secondary to 50. Tilanus HW, Bossuyt P, Schattenkerk ME, Obertop H. Treatment of
splenic vein thrombosis. A previously unrecognized relationship. oesophageal perforation: a multivariate analysis. Br J Surg 1991;
Chest 1991; 100: 574–5. 78: 582–5.
27. Gangahar DM, Delany HM. Intrasplenic abscess: two case reports 51. Salo JA, Isolauri JO, Heikkila LJ, et al. Management of delayed
and review of the literature. Am Surg 1981; 47: 488–91. esophageal perforation with mediastinal sepsis. Esophagectomy or
28. Aubry P, Reynaud JP, Nbonyingingo C, Ndabaneze E, Mucikere E. primary repair? J Thorac Cardiovasc Surg 1993; 106: 1088–91.
[Ultrasonographic data of the solid organs of the abdomen in 52. Larsson S, Pettersson G, Lepore V. Esophagocutaneous drainage to
stage IV human immunodeficiency virus infection. A prospective treat late and complicated esophageal perforation. Eur J
study of 101 cases in central Africa]. Ann Gastroenterol Hepatol Cardiothorac Surg 1991; 5: 579–82.
(Paris) 1994; 30: 43–52. ◆53. Bufkin BL, Miller JI Jr, Mansour KA. Esophageal perforation:
29. Koehler PR, Jones R. Association of left-sided pleural effusions and emphasis on management. Ann Thorac Surg 1996; 61: 1447–51;
splenic hematomas. AJR Am J Roentgenol 1980; 135: 851–3. discussion 51–2.
30. Vitaux J, Theodore C, Molas G, et al. [Subcapsular haematoma of 54. Serna DL, Vovan TT, Roum JH, Brenner M, Chen JC. Successful
the spleen as complication of chronic pancreatitis. Contribution of nonoperative management of delayed spontaneous esophageal
echotomography to the diagnosis in two cases (author’s transl)]. perforation in patients with human immunodeficiency virus. Crit
Nouv Presse Med 1981; 10: 495–8. Care Med 2000; 28: 2634–7.
31. Lupien C, Sauerbrei EE. Healing in the traumatized spleen: 55. Dougenis D, Petsas T, Bouboulis N, et al. Management of non
sonographic investigation. Radiology 1984; 151: 181–5. resectable malignant esophageal stricture and fistula. Eur J
32. Cox RE. Splenic infarct in a white man with sickle cell trait. Ann Cardiothorac Surg 1997; 11: 38–45.
Emerg Med 1982; 11: 668–9. 56. Howell HS, Prinz RA, Pickleman JR. Anaerobic mediastinitis. Surg
33. Hirai K, Kawazoe Y, Yamashita K, et al. Transcatheter partial Gynecol Obstet 1976; 143: 353–9.
splenic arterial embolization in patients with hypersplenism: a 57. Saugier B, Emonot A, Plauchu M, Galy P. [Effusions rich in amylase
clinical evaluation as supporting therapy for hepatocellular without pancreatitis. 14 cases]. Nouv Presse Med 1976; 5:
carcinoma and liver cirrhosis. Hepatogastroenterology 1986; 33: 2777–80.
105–8. 58. Asch MJ, Coran AG, Johnston PW. Gastric perforation secondary to
34. Cosme A, Bujanda L, Arriola JA, Ojeda E. Esophageal Crohn’s blunt trauma in children. J Trauma 1975; 15: 187–9.
disease with esophagopleural fistula. Endoscopy 1998; 30: S109. 59. Baker A, Volberg F, Sumner T, Moran R. Childhood Menetrier’s
35. Cronstedt JL, Bouchama A, Hainau B, et al. Spontaneous disease: four new cases and discussion of the literature.
esophageal perforation in herpes simplex esophagitis. Am J Gastrointest Radiol 1986; 11: 131–4.
Gastroenterol 1992; 87: 124–7. 60. Domej W, Kullnig P, Petritsch W, et al. Colobronchial fistula: a rare
36. Ramo OJ, Salo JA, Isolauri J, Luostarinen M, Mattila SP. complication of Crohn’s colitis. Am Rev Respir Dis 1990; 142:
Tuberculous fistula of the esophagus. Ann Thorac Surg 1996; 62: 1225–7.
1030–2. 61. Al-Emadi M, Helmy I, Nada MA, Al-Jaber H. Laparoscopic repair of
37. Avanoglu A, Ergun O, Mutaf O. Management of instrumental Bochdalek hernia in an adult. Surg Laparosc Endosc Percutan Tech
perforations of the esophagus occurring during treatment of 1999; 9: 423–5.
corrosive strictures. J Pediatr Surg 1998; 33: 1393–5. 62. McAllister JD, D’Altorio RA, Snyder A. CT findings after
38. Bacon BR, Camara DS, Duffy MC. Severe ulceration and delayed uncomplicated percutaneous laparoscopic cholecystectomy. J
perforation of the esophagus after endoscopic variceal Comput Assist Tomogr 1991; 15: 770–2.
sclerotherapy. Gastrointest Endosc 1987; 33: 311–5. 63. Cunningham LW, Grobman M, Paz HL, Hanlon CA, Promisloff RA.
39. Massard G, Ducrocq X, Hentz JG, et al. Esophagopleural fistula: an Cholecystopleural fistula with cholelithiasis presenting as a right
early and long-term complication after pneumonectomy. Ann pleural effusion. Chest 1990; 97: 751–2.
Thorac Surg 1994; 58: 1437–40; discussion 41. 64. Delco F, Domenighetti G, Kauzlaric D, Donati D, Mombelli G.
40. Haynes DE, Haynes BE, Yong YV. Esophageal rupture complicating Spontaneous biliothorax (thoracobilia) following cholecystopleural
Heimlich maneuver. Am J Emerg Med 1984; 2: 507–9. fistula presenting as an acute respiratory insufficiency. Successful
●41. Maulitz RM, Good JT Jr, Kaplan RL, Reller LB, Sahn SA. The removal of gallstones from the pleural space. Chest 1994; 106:
pleuropulmonary consequences of esophageal rupture: an 961–3.
experimental model. Am Rev Respir Dis 1979; 120: 363–7. 65. Chan SY, Osborne AW, Purkiss SF. Cholelithoptysis: an unusual
42. Rudin JS, Ellrodt AG, Phillips EH. Low pleural fluid amylase complication following laparoscopic cholecystectomy. Dig Surg
associated with spontaneous rupture of the esophagus. Arch 1998; 15: 707–8.
Intern Med 1983; 143: 1034–5. ●66. Strange C, Allen ML, Freedland PN, Cunningham J, Sahn SA.
●43. Good JT Jr, Antony VB, Reller LB, Maulitz RM, Sahn SA. The Biliopleural fistula as a complication of percutaneous biliary
pathogenesis of the low pleural fluid pH in esophageal rupture. drainage: experimental evidence for pleural inflammation. Am Rev
Am Rev Respir Dis 1983; 127: 702–4. Respir Dis 1988; 137: 959–61.
44. Drury M, Anderson W, Heffner JE. Diagnostic value of pleural fluid 67. DeCosse JJ, Poulin TL, Fox PS, Condon RE. Subphrenic abscess.
cytology in occult Boerhaave’s syndrome. Chest 1992; 102: Surg Gynecol Obstet 1974; 138: 841–6.
976–8. 68. Dietrick DB. Experience with liver abscess. Am J Surg 1984; 147:
45. Klygis LM, Jutabha R, McCrohan MB, Vanagunas AD. Esophageal 288–91.
perforations masked by steroids. Abdom Imaging 1993; 18: 10–2. ●69. Perera MR, Kirk A, Noone P. Presentation, diagnosis and
46. Han SY, McElvein RB, Aldrete JS, Tishler JM. Perforation of the management of liver abscess. Lancet 1980; ii: 626–32.
472 Effusions caused by gastrointestinal disease
●70. Katsilabros L, Triandafillow G, Kontoyiannis P, Katsilabros N. 76. Moritani T, Aihara T, Oguma E, et al. Spectrum of Epstein–Barr
Pleural effusion and hepatitis. Gastroenterology 1972; 63: virus infection in Japanese children: a pictorial essay. Clin Imaging
718–22. 2001; 25: 1–8.
71. Flacks LM, Lees D. A case of hepatitis B with pleural effusion. Aust 77. Furui S, Otomo K, Itai Y, Iio M. Hepatocellular carcinoma treated
N Z J Med 1977; 7: 636–7. by transcatheter arterial embolization: progress evaluated by
72. Pramuljo HS, Harun SR. Ultrasound findings in dengue computed tomography. Radiology 1984; 150: 773–8.
haemorrhagic fever. Pediatr Radiol 1991; 21: 100–2. ●78. Branca P, Rodriguez RM, Rogers JT, et al. Routine measurement of
73. Thulkar S, Sharma S, Srivastava DN, et al. Sonographic findings in pleural fluid amylase is not indicated. Arch Intern Med 2001; 161:
grade III dengue hemorrhagic fever in adults. J Clin Ultrasound 228–32.
2000; 28: 34–7. ●79. Teplick JG, Teplick SK, Goodman L, Haskin ME. The interface sign: a
74. Setiawan MW, Samsi TK, Wulur H, Sugianto D, Pool TN. Dengue computed tomographic sign for distinguishing pleural and intra-
haemorrhagic fever: ultrasound as an aid to predict the severity of abdominal fluid. Radiology 1982; 144: 359–62.
the disease. Pediatr Radiol 1998; 28: 1–4. 80. Griffin DJ, Gross BH, McCracken S, Glazer GM. Observations on CT
75. Kim YO, Chun KA, Choi JY, et al. Sonographic evaluation of differentiation of pleural and peritoneal fluid. J Comput Assist
gallbladder-wall thickening in hemorrhagic fever with renal Tomogr 1984; 8: 24–8.
syndrome: prediction of disease severity. J Clin Ultrasound 2001; ◆81. Light RW. Exudative pleural effusions secondary to
29: 286–9. gastrointestinal diseases. Clin Chest Med 1985; 6: 103–11.
37
Effusions of obstetric or gynecological origin
RICHARD W LIGHT
Pleural effusions can occur with a variety of gynecological and coworkers2 reviewed the medical charts and all avail-
and obstetrical conditions. Patients who are pregnant can able imaging studies of 771 women who had undergone
have pleural effusions from common causes such as pneu- induction of superovulation with gonadotropins during
monia, viral infections or pulmonary emboli. However, the period October 1990 to July 1992 at University
there are a few types of pleural effusions that are specific to Hospital in Western Ontario and reported that the inci-
obstetric and gynecological conditions, including the dence of severe OHSS was 3 percent. In another report, the
ovarian hyperstimulation syndrome, fetal pleural effusions medical records of all OHSS patients hospitalized between
and the pleural effusions that occur in the post-partum January 1987 and December 1996 in 16 of the 19 tertiary
period. medical centers in Israel were reviewed.3 The authors were
able to find a total of 2902 patients (3305 hospitalizations)
with OHSS of whom 196 (6.7 percent) had severe and 13
(0.4 percent) had critical OHSS. In this series, 27 of the 209
OVARIAN HYPERSTIMULATION SYNDROME patients (12.9 percent) underwent thoracentesis,3 while
130 patients were dyspneic and received chest radiographs.
Definition Thirty eight of the 130 patients (29 percent) had a pleural
effusion.
The ovarian hyperstimulation syndrome (OHSS) is a
serious complication of ovulation induction with human
chorionic gonadotropin (hCG) and occasionally Etiology and pathogenesis
clomiphene. This syndrome is characterized by ovarian
enlargement and fluid shifts resulting in intravascular The exact etiology of the pleural effusion seen with severe
volume depletion.1 Severe OHSS is a life-threatening con- OHSS is not known. There are two primary components
dition characterized by clinical and sonographic evidence to OHSS: (a) enlargement of the ovaries accompanied by
of massive ascites or hydrothorax, as well as by breathing the formation of follicular, luteal and hemorrhagic ovarian
difficulties, increased blood viscosity, renal/hepatic dys- cysts and edema of the stroma, and (b) an acute shift of
function or anasarca. Critical OHSS includes, in addition, fluid out of the intravascular space.2 In patients undergo-
overt renal failure, thromboembolic phenomena, tense ing ovulation induction, several risk factors have been
ascites or the acute respiratory distress syndrome (ARDS).1 reported to be associated with OHSS. These include young
age (<35 years), asthenic habitus, patients with polycystic
ovaries, pregnancy resulting from stimulation, hCG sup-
Incidence and epidemiology plementation, high serum estradiol (>2500 pg/mL) and
multiple follicles.4
Severe OHSS occurs in approximately 3 percent of patients At one time, it was thought that OHSS resulted from
undergoing superovulation with gonadotropins.2 Levin high local concentrations of estrogen in the ovaries causing
474 Effusions of obstetric or gynecological origin
than 15 ovarian follicles with a high proportion of small of the women miscarried.24 In this series the karyotype was
and intermediate-sized follicles.18 However, there is no abnormal in 9 of 11 instances (82 percent) in which it was
universal agreement as to what level of estradiol should performed.
serve as an indication for cessation of the injections.
Treatment and management In view of the above, the following approach to fetuses
with a pleural effusion is recommended. When a fetus with
The best way to manage patients with fetal pleural effu- hydrothorax develops acute fetal distress, a thoracentesis
sions is controversial.20 If the pleural effusions are not should be performed. At the time of thoracentesis, tests
treated, some will resolve spontaneously. Spontaneous such as karyotype, fetal blood count, maternal serology
regression of the effusion occurred in 22 percent of the 89 and meticulous echography are performed in an attempt
untreated patients in the series of Aubard et al.31 Effusions to ascertain if there are other congenital abnormalities. If
which resolve spontaneously are more likely to be unilat- there are abnormalities, then a decision must be made
eral, the diagnosis made early in the second trimester, and concerning the termination of the pregnancy. If there are
are not associated with polyhydramnios or hydrops.31 no abnormalities and hydrops is present, a shunt is placed.
However, in others the effusion will increase in size and One report suggested that the presence of a contralateral
the fetus will develop generalized hydrops. If the effusion mediastinal shift and/or diaphragmatic inversion indi-
persists, the underlying lung will not develop and upon cated that the intrathoracic pressure was high and should
delivery the baby may die from pulmonary hypoplasia.20 be indications for placement of the shunt.28 Otherwise, if
A literature review of 204 cases of fetal hydrothoraces the effusion is well tolerated, the fetus is treated conserva-
showed a mortality of 39 percent in 89 fetuses who tively with a repeat ultrasound scan in 2 weeks. If upon
received no in utero therapy. The mortality was higher if repeat ultrasound scan, the effusion is shown to be enlarg-
the fetus had hydrops (76 percent) or if the effusion was ing and the fetus is less than 32 weeks gestation, a shunt is
bilateral (47 percent).31 placed. If the fetus is more than 32 weeks gestation a tho-
Therapeutic maneuvers had been attempted in the racentesis is performed. If the effusion is not enlarging,
majority of patients. In Aubard’s review 29 fetuses were repeat examinations are performed at 2-week intervals.
treated with thoracentesis and the mortality was 45 Any fetus that has a significant effusion just prior to birth
percent.31 The principal drawback of thoracentesis is the is subjected to a thoracentesis.31
rapid reaccumulation of the effusion. Twenty-five of the 29 The prognosis of fetuses that receive pleuroamniotic
patients (86 percent) had either one or two thoracenteses. shunts and who survive appears to be good. Thompson et
There is one report where the administration of medium al.37 studied 17 infants who had undergone pleuroamni-
chain triglycerides to the mother was thought to slow the otic shunting for a fetal pleural effusion at a median age of
reaccumulation of pleural fluid.32 However, physiologically 12 months. They reported that respiratory symptoms and
this is difficult to explain since the triglycerides ingested by function were no different in these 17 infants than those in
the mother never get into the thoracic duct of the fetus. a control group.37
Pleuroamniotic shunting, as described by Rodeck and
coworkers,33 is frequently used to treat fetal pleural effu-
sions. With this technique a metal trocar with cannula is Complications
introduced through the maternal abdominal wall and
through the fetal thorax as close as possible to the midax- The placement of pleuroamniotic shunts is not without
illary line at the level of the base of the scapula. A double complication. Catheter migration occurred 10 times in 80
pigtail catheter is introduced through the trocar lumen. A fetuses in one review, but this did not contraindicate the
short introducer rod is used to position the distal catheter placement of a new shunt. Other complications include
loop inside the fetal thorax. The trocar is then used to posi- obstruction of the shunt and migration of the catheter into
tion the proximal catheter loop in the amniotic cavity. the maternal peritoneal cavity or the fetal thoracic cavity.38
Positioned in this manner, the catheter creates a perma-
nent communication between the pleural space and the
amniotic cavity. Future directions of development
In a review of 80 fetuses treated with pleuroamniotic
shunts, the overall mortality was 26 percent.31 In 43 fetuses There are two areas for future development in the manage-
where this information was specified, a pleuroamniotic ment of fetal pleural effusions. First, it would be important
shunt was placed bilaterally in 24 and unilaterally in 19 to develop indicators as to which fetuses with hydrothorax
cases. The shunts become displaced intrathoracically in a will progress and which ones will regress. Second, there
significant percentage of fetuses, but since there appears to needs to be improvements in the design of the catheters so
be no long-term pulmonary complications, such shunts that they are less likely to migrate and less likely to become
need not be removed.34 occluded.
Some have advocated creation of a pleurodesis in the
fetus by the intrapleural injection of the immunostimulant
OK-43235 or maternal blood.36 However, since the long- PLEURAL EFFUSIONS DURING PREGNANCY
term side effects from a pleurodesis performed in a fetus
remain to be determined, pleurodesis is not recom- There has never been a systematic study on the diseases
mended. causing pleural effusion in women that are pregnant.
Post-partum pleural effusion (delayed) 477
Incidence and epidemiology ery was 46 percent and the effusions were bilateral in 75
percent.43 In a prospective study by the same researchers
Pregnant women frequently have small pleural effusions the prevalence of pleural effusion was 67 percent in a
demonstrable by ultrasonography. Kocijancic et al.39 group of 30 patients and the effusions were bilateral in 55
studied 47 women at a mean gestation of 24.4 weeks and percent.43 Decubitus radiographs confirmed the effusion
reported that 28 (59.5 percent) had free pleural fluid that in seven of the ten women with effusions who had a decu-
was bilateral in 18 (38.3 percent) and unilateral in 10 (21.2 bitus radiograph. In contrast, Udeshi et al.44 reported that
percent). The mean thickness of the pleural fluid was 2.9 ± the prevalence of pleural effusion was only 2 percent in a
1.1 mm and the patients were asymptomatic. The preva- series of 50 patients who had ultrasonography of the chest.
lence of demonstrable pleural fluid was 25 percent in 106 The one patient in this latter series who had a pleural effu-
normal individuals studied by the same investigators. sion had severe pre-eclampsia with apparent pulmonary
edema. An intermediate prevalence was reported by Wallis
and coworkers45 who prospectively studied 34 patients
Etiology and pathogenesis with moderate to severe pre-eclampsia with ultrasound
and reported that six (17.6 percent) of the patients had a
When a woman develops a symptomatic pleural effusion pleural effusion. The explanation for the marked discrep-
early in pregnancy, the possibility of the ovarian hyper- ancy in the prevalence of pleural effusions in the different
stimulation syndrome should be considered. This entity is series is not clear.
discussed early in the chapter. The distribution of the diag-
noses responsible for pleural effusions in pregnant women
is probably similar to that in non-pregnant women of the Etiology and pathogenesis
same age.40 Pulmonary embolism is probably the leading
cause of pleural effusion in the pregnant woman40 as the The etiology of the post-partum pleural effusion is
incidence of pleural embolism is higher in pregnant than unknown. Hughson and coworkers43 hypothesized that
in non-pregnant women.41 The second leading cause is the following two factors were contributory. First, the
probably pneumonia.40 In women with the HELLP syn- colloid osmotic pressure is decreased during pregnancy
drome (hemolysis, elevated liver enzymes and low platelet and this decreased colloid osmotic pressure should lead to
count) or with severe pre-eclampsia, the prevalence of increased pleural fluid formation. Second, the Valsalva
pleural effusion is about 3 percent.42 If a pregnant woman maneuvers typical of the second stage of labor could
has a transudative pleural effusion, peripartum cardiomy- impair lymphatic drainage of the pleural space by elevating
opathy should be considered. systemic venous pressure. In the study of Wallis and asso-
ciates of patients with moderate to severe pre-eclampsia,
the patients with pleural effusion did not have a greater
Clinical presentation and treatment degree of hypertension or proteinuria.45
sidered in patients with pleural effusions in the weeks to period for early signs or symptoms of pleuropulmonary or
months after delivery is pulmonary embolism. The risk of cardiac disease or thrombotic episodes. When such post-
pulmonary embolism is 15 times higher in the post- partum complications are recognized, assays should be
partum period than during pregnancy.41 performed for antiphospholipid antibodies and, if they are
positive, consideration should be given to immunosup-
Etiology and pathogenesis pressive therapy after pulmonary embolus and infection
have been carefully ruled out.46
The pathogenesis of the syndrome appears to be related to
the presence of antiphospholipid antibodies. There are two Complications
different antiphospholipid antibodies: the lupus anticoagu-
lant and the anticardiolipin antibody. The lupus
The primary complications are myocarditis with conges-
anticoagulant was first discovered in patients who had sys-
tive heart failure and venous thrombosis.
temic lupus erythematosus, but is now known to be found
in other disease states, as well as in totally asymptomatic
patients. It is a monoclonal antibody that reacts with platelet Future directions of development
factor 3 to cause incomplete formation of the prothrombin
activator complex. The lupus anticoagulant has been asso-
It is unclear as to what is the true incidence of this syn-
ciated with the inhibition of prostacyclin production by
drome. In the future, prospective studies will delineate the
endothelial cells. Other theories of its activity include inter-
prevalence of this syndrome and its optimal treatment.
ference with the anticoagulation system and inhibition of
fibrinolysis. The anticardiolipin antibody also reacts with
negatively charged phospholipids. Its mode of action is
believed to be similar to that of the lupus anticoagulant.47
MEIGS’ SYNDROME
Clinical presentation When this syndrome was originally described by Meigs in
1937, the syndrome consisted of the presence of ascites
Patients with this syndrome typically present with fever, and pleural effusions in patients with benign solid ovarian
pleural effusions and pulmonary infiltrates within a few tumors.48 Moreover, the syndrome required that when the
weeks of delivery.46 The patients also frequently present ovarian tumor was removed, the ascites and the pleural
with chest pain. Some of the patients also have evidence of effusion both had to resolve.48 Subsequent to the initial
myocarditis and/or of thrombosis of various vessels.46 The description, it became apparent that a similar syndrome
patients have either lupus anticoagulant or anticardiolipin could occur with benign cystic ovarian tumors, benign
antibodies or both, but do not have antinuclear antibodies; tumors of the uterus (fibromyomata), low-grade ovarian
neither do they fulfill the criteria for the diagnosis of sys- malignant tumors without evidence of metastases49 and
temic lupus erythematosus.46 All of the reported patients endometriomas.50 Although Meigs still prefers to reserve
reported had false-positive tests for syphilis.46,47 It should his name for only those cases in which the primary neo-
be noted that the most well known complication of the plasm is a benign solid ovarian tumor, I classify any patient
antiphospholipid syndrome is spontaneous abortion, most with a pelvic neoplasm associated with ascites and pleural
commonly in the second trimester. effusion in whom surgical removal of the tumor results in
permanent disappearance of the ascites and pleural effu-
Investigations sion as having had Meigs’ syndrome.6
Treatment and management The reason that patients with Meigs’ syndrome develop
ascites appears to be the secretion of large amounts of fluid
Pregnant patients known to have antiphospholipid anti- by the primary tumor. When tumors have been resected
bodies should be carefully monitored in the post-partum and placed in dry containers, the secretion of large
Endometriosis and pleural effusions 479
amounts of fluid continues.49 Larger tumors are more Investigations and treatment
likely to be associated with free peritoneal fluid. Samanth
and Black52 reported that free peritoneal fluid was only Before a female with a pelvic mass, ascites and a pleural
found with tumors with a diameter greater than 11 cm. effusion is labeled as having an inoperable malignancy,
It appears that the pathogenesis of the pleural effusion definite evidence of malignancy in a body fluid or a biopsy
in patients with Meigs’ syndrome is similar to that in specimen should be obtained. The initial evaluation
patients with cirrhosis, ascites and a pleural effusion. It is should include cytology on the ascitic and pleural fluid. If
thought that fluid passes from the peritoneal cavity into these cytologies are negative and there is no other evidence
the pleural cavity through small holes in the diaphragm.53 of malignancy elsewhere, exploratory laparotomy should
Supporting evidence for this theory includes the following: be considered. If there is no evidence of malignancy, the
(a) the ascitic and the pleural fluid have similar character- primary neoplasm should be removed. The diagnosis is
istics, (b) the pleural fluid rapidly recurs following thora- confirmed when the ascites and the pleural fluid resolve
centesis, and (c) some patients have ovarian tumors and post-operatively and do not recur. The disappearance of
ascites without pleural effusion. It should be noted, these fluids post-operatively usually occurs within 2
however, that some authors believe that the pleural fluid weeks.61
arises from the transdiaphragmatic transfer of ascitic fluid
by the lymphatic vessels.33,35,38
It appears that the fluid secreted by the tumors into the Future directions of development
peritoneal cavity has high levels of factors which may
increase the permeability of vessels in the peritoneal and In the future, additional studies should reveal more pre-
pleural cavities and lead to more fluid formation in these cisely the role of various vasoactive substances and
body cavities. The serum levels of IL-1, IL-6, IL-8, TNF-a cytokines in the pathogenesis of Meigs’ syndrome.
and VEGF are all elevated in patients with Meigs’ syn-
drome compared with controls. The levels of these factors
are higher in the ascitic and pleural fluid than in the ENDOMETRIOSIS AND PLEURAL EFFUSIONS
serum.54–56
The tumors most commonly responsible for Meigs’ On occasion, severe endometriosis is complicated by
syndrome (my definition) are ovarian fibroma, followed massive ascites63 and a pleural effusion is present in
by ovarian cysts, thecomas, granulosal cell tumors and approximately 30 percent of these patients.63
leiomyomas of the uterus.49
The pleural fluid has not been systematically studied, The patients usually present with dyspnea or chest pain.
but it is usually described as bloody or chocolate-colored.64 The right hemithorax is universally involved.65
Cytology of the fluid has shown numerous hemosiderin-
laden macrophages with an elevated CA-125.63 The pleural
effusions have been unilateral right-sided in the majority Investigations
of cases.64
Thoracentesis reveals a bloody pleural fluid. This finding,
in conjunction with the history showing the relationship
Investigation to menstruation and pelvic endometriosis, are usually suf-
ficient to establish the diagnosis.
This diagnosis should be considered in any menstruating
female who develops bloody or chocolate-colored pleural
fluid. The diagnosis is usually definitively established at the Treatment and management
time of laparotomy.
The treatment of choice appears to be total hysterectomy
and bilateral salpingo-oophorectomy because hormonal
Treatment and management suppressive therapy appears to provide only temporary
improvement. If the hemothorax is large, the patient
The treatment of the massive ascites, pleural effusion and should be treated with tube thoracostomy.
endometriosis is not easy. Hormonal therapy (progesta-
tional agents, danazol, or luprolide acetate [Lupron]) fails
in more than 50 percent of cases. Most commonly, these Future directions of development
patients are treated with total abdominal hysterectomy
and bilateral salpingo-oophorectomy, but the presence of Since this disease is so uncommon, it is difficult to know
the pelvic endometriosis makes this surgery difficult. the future directions. A better therapy would be one that
does not involve total hysterectomy and bilateral salpingo-
oophorectomy in these young patients.
Future directions of development
KEY POINTS
CATAMENIAL HEMOTHORAX
● Pleural effusions occur as a complication of the
Incidence and epidemiology ovarian hyperstimulation syndrome which
occurs in approximately 3 percent of patients
A review of the literature on catamenial hemothoraces in who have undergone induction of superovula-
1993 revealed a total of 16 cases.65 tion with gonadotropins. The pleural effusions
are secondary to a generalized capillary leak syn-
drome plus movement of ascitic fluid from the
Etiology and pathogenesis peritoneal cavity to the pleural cavity.
● Fetal pleural effusions which persist cause pul-
The majority of patients with catamenial hemothoraces monary hypoplasia at birth. Fetuses with persist-
have associated pelvic and abdominal endometriosis. The ent or recurrent pleural effusions are best
right hemithorax is almost always involved and some of managed with intra-uterine shunts.
the patients have diaphragmatic fenestrations. It has been ● A sizeable fraction of patients have very small
hypothesized that the endometrial tissue enters the thorax pleural effusions in the immediate post-partum
through the diaphragmatic fenestrations. When the period.
patient menstruates the endometrial tissue is shed into the ● The presence of a pelvic mass, ascites and a
pleural space and a hemothorax ensues. pleural effusion does not necessarily indicate dis-
seminated malignancy – Meigs’ syndrome
should be considered.
Clinical presentation ● Pleural effusions at times complicate endo-
metriosis with ascites or endometriosis of the
Most patients who present with catamenial hemothorax pleura.
are nulliparous and the average age has been 32 years.65
References 481
●20.
REFERENCES Hagay Z, Reece A, Roberts A, Hobbins JC. Isolated fetal pleural
effusion: a prenatal management dilemma. Obstet Gynecol 1993;
81: 147–52.
● = Key primary paper 21. Rocha G, Fernandes P, Rocha P, et al. Pleural effusions in the
◆ = Major review article neonate. Acta Paediatr 2006; 95: 791–8.
22. Eddleman KA, Levine AB, Chitkara U, Berkowitz RL. Reliability of
1. Delvigne A, Rozenberg S. Review of clinical course and treatment pleural fluid lymphocyte counts in the antenatal diagnosis of
of ovarian hyperstimulation syndrome (OHSS). Hum Reprod congenital chylothorax. Obstet Gynecol 1991; 78: 530–2.
Update 2003; 9: 77–96. 23. Chang YL, Lien R, Wang CJ, et al. Congenital chylothorax in three
2. Levin MF, Kaplan BR, Hutton LC. Thoracic manifestations of ovarian siblings. Am J Obstet Gynecol 2005; 192: 2065–6.
hyperstimulation syndrome. Can Assoc Radiol J 1995; 46: 23–6. 24. Hashimoto K, Shimizu T, Fukuda M, et al. Pregnancy outcome of
●3. Abramov Y, Elchalal U, Schenker JG. Pulmonary manifestations of embryonic/fetal pleural effusion in the first trimester. J Ultrasound
severe ovarian hyperstimulation syndrome: a multicenter study. Med 2003; 22: 501–5.
Fertil Steril 1999: 71: 645–51. 25. Benacerraf BR, Frigoletto FD Jr, Wilson M. Successful midtrimester
4. Elchalal U, Schenker JG. The pathophysiology of ovarian thoracentesis with analysis of the lymphocyte population in the
hyperstimulation syndrome – views and ideas. Hum Reprod 1997; pleural effusion. Am J Obstet Gynecol 1986; 155: 398–9.
12: 1129–37. 26. Chernick V, Reed MH. Pneumothorax and chylothorax in the
5. Yarali H, Fleige-Zahradka BG, Yuen BH, McComb PF. The ascites in neonatal period. J Pediatr 1970; 76: 624–32.
ovarian hyperstimulation syndrome does not originate from the 27. Tsukimori K, Nakanami N, Fukushima K, et al. Pleural fluid/serum
ovary. Fertil Steril 1993; 59: 657–61. immunoglobulin ratio is a diagnostic marker for congenital
6. Light RW. Pleural diseases, 5th edn. Baltimore: Lippincott, chylothorax in utero. J Perinat Med 2006; 34: 313–17.
Williams and Wilkins, 2007. 28. Yamamoto M, Insunza A, Carrillo J, et al. Intrathoracic pressure in
7. Wang TH, Horng SG, Chang CL, et al. Human chorionic congenital chylothorax. Keystone for the rationale of
gonadotropin-induced ovarian hyperstimulation syndrome is thoracoamniotic shunting? Fetal Diagn Ther 2007; 22: 169–171.
associated with up-regulation of vascular endothelial growth 29. Murayama K, Jimbo T, Matsumoto Y, Mitsuishi C, Nishida H. Fetal
factor. J Clin Endocrinol Metab 2002; 87: 3300–8. pulmonary hypoplasia with hydrothorax. Am J Obstet Gynecol
●8. Goldsman MP, Pedram A, Dominguez CE, et al. Increased capillary 1987; 157: 119–20.
permeability induced by human follicular fluid: a hypothesis for an 30. Waller K, Chaithongwongwatthana S, Yamasmit W, et al.
ovarian origin of the hyperstimulation syndrome. Fertil Steril Chromosomal abnormalities among 246 fetuses with pleural
1995; 63: 268–72. effusions detected on prenatal ultrasound examination: Factors
9. McClure N, Healy DL, Rogers PAW, et al. Vascular endothelial associated with an increased risk of aneuploidy. Genet Med 2005;
growth factor as capillary permeability agent in ovarian 7: 417–21.
hyperstimulation syndrome. Lancet 1994; 344: 235–6. ◆31. Aubard Y, Derouineau I, Aubard V, Chalifour V, Preux PM. Primary
10. Krasnow JS, Berga SL, Gusick DS, et al. Vascular permeability fetal hydrothorax: A literature review and proposed antenatal
factor and vascular endothelial growth factor in ovarian clinical strategy. Fetal Diagn Ther 1998; 13: 325–33.
hyperstimulation syndrome: a preliminary report. Fertil Steril 32. Bartha JL, Comino-Delgado R. Fetal chylothorax response to
1996; 65: 552–5. maternal dietary treatment. Obstet Gynecol 2001; 97 (5 Suppl 1):
●11. Chen CD, Wu MY, Chen HF, et al. Prognostic importance of serial 820–3.
cytokine changes in ascites and pleural effusion in women with 33. Rodeck CH, Fisk NM, Fraser DI, Nicolini U. Long-term in utero
severe ovarian hyperstimulation syndrome. Fertil Steril 1999; 72: drainage of fetal hydrothorax. N Engl J Med 1988; 319: 1135–8.
286–92. 34. Sepulveda W, Galindo A, Sosa A, et al. Intrathoracic dislodgement
12. Loret de Mola JR, Farredondo-Soberon F, Randle CP, et al. of pleuro-amniotic shunt. Three case reports with long-term
Markedly elevated cytokines in pleural effusion during the ovarian follow-up. Fetal Diagn Ther 2005; 20: 102–5.
hyperstimulation syndrome: transudate or ascites? Fertil Steril 35. Chen M, Chen CP, Shih JC, et al. Antenatal treatment of
1997; 67: 780–2. chylothorax and cystic hygroma with OK-432 in nonimmune
13. Gregory WT, Patton PE. Isolated pleural effusion in severe ovarian hydrops fetalis. Fetal Diagn Ther 2005; 20: 309–15.
hyperstimulation: A case report. Am J Obstet Gynecol 1999; 180: 36. Parra J, Amenedo M, Muniz-Diaz E, et al. A new successful therapy
1468–71. for fetal chylothorax by intrapleural injection of maternal blood.
14. Rabinerson D, Shalev J, Royburt M, Ben-Rafael Z, Dekel A. Severe Ultrasound Obstet Gynecol 2003; 22: 290–4.
unilateral hydrothorax as the only manifestation of the ovarian 37. Thompson P, Greenough A, Nicolaides KH. Respiratory function in
hyperstimulation syndrome. Gynecol Obstet Invest 2000; 49: infancy following pleuro-amniotic shunting. Fetal Diagn Ther
140–2. 1993; 8: 79–83.
15. Yuen BH, McComb P, Sy L, et al. Plasma prolactin, human ◆38. Wittman BK, Martin KA, Wilson RD, Peacock D. Complications of
chorionic gonadotropin, estradiol, testosterone, and progesterone long-term drainage of fetal pleural effusion: case report and
in the ovarian hyperstimulation syndrome. Am J Obstet Gynecol review of the literature. Am J Perinatol 1997; 14: 443–7.
1979; 133: 316–20. 39. Kocijancic I, Pusenjak S, Kocijancic K, et al. Sonographic detection
16. Abramov Y, Elchalal U, Schenker JG. Febrile morbidity in severe of physiologic pleural fluid in normal pregnant women. J Clin
and critical ovarian hyperstimulation syndrome: a multicentre Ultrasound 2005; 33: 63–6.
study. Hum Reprod 1998; 13: 3128–31. 40. Light RW. Pleural diseases in pregnancy. Int Pleural News 2006; 4:
17. Abramov Y, Fatum M, Abrahamov D, Schenker JG. 6–7.
Hydroxyethylstarch versus human albumin for the treatment of 41. Heit JA, Kobbervig CE, James AH, et al. Trends in the indicence of
severe ovarian hyperstimulation syndrome: a preliminary report. venous thromboembolism during pregnancy or postpartum: a 30-
Fertil Steril 2001; 75: 1228–30. year population-based study. Ann Intern Med 2005; 143: 749–50.
18. Rizk B, Aboulghar M. Modern management of ovarian 42. Haddad B, Barton JR, Livingston JC, et al. HELLP (hemolysis,
hyperstimulation syndrome. Hum Reprod 1991; 6: 1082–7. elevated liver enzymes, and low platelet count) syndrome versus
19. Semba S, Moriya T, Youssef EM, Sasano H. An autopsy case of severe preeclampsia: onset at £ 28 weeks’ gestation. Am J Obstet
ovarian hyperstimulation syndrome with massive pulmonary Gynecol 2000; 183: 1475–9.
edema and pleural effusion. Pathol Int 2000; 50: 549–52. ●43. Hughson WG, Friedman PJ, Feigin DS, Resnick R, Moser KM.
482 Effusions of obstetric or gynecological origin
Postpartum pleural effusion: a common radiologic finding. Ann levels of vascular endothelial growth factor, fibroblast growth
Intern Med 1982; 97: 856–8. factor, and interleukin 6 in Meigs’ syndrome. Am J Obstet Gynecol
●44. Udeshi UL, McHugo JM, Selwyn CJ. Postpartum pleural effusion. Br 2001; 184: 354–5.
J Obstet Gynaecol 1988; 95: 894–7. 56. Abramov Y, Anteby SO, Fasouliotis SJ, et al. The role of
45. Wallis MG, McHugo JM, Carruthers DA, Selwyn Crawford J. The inflammatory cytokines in Meigs’ syndrome. Obstet Gynecol 2002;
prevalence of pleural effusions in pre-eclampsia: an ultrasound 99 (5 Suppl 1): 917–9.
study. Br J Obstet Gynaecol 1989; 96: 431–3. 57. Patsner B. Meigs’ syndrome and “false positive” preoperative
●46. Kochenour NK, Branch DW, Rote NS, et al. A new postpartum serum CA-125 levels: analysis of ten cases. Eur J Gynaecol Oncol
syndrome associated with antiphospholipid antibodies. Obstet 2000; 21: 362–3
Gynecol 1987; 69: 460–8. 58. Loizzi V, Cormio G, Resta L, et al. Pseudo-Meigs syndrome and
47. Ayres MA, Sulak PJ. Pregnancy complicated by antiphospholipid elevated CA125 associated with struma ovarii. Gynecol Oncol
antibodies. South Med J 1991; 84: 266–9. 2005; 97: 282–4.
●48. Meigs JV, Cass JW. Fibroma of the ovary with ascites and 59. O’Flanagan SJ, Tighe BF, Egan TJ, Delaney PV. Meigs’ syndrome and
hydrothorax. Am J Obstet Gynecol 1937; 33: 249–67. pseudo-Meigs’ syndrome. J R Soc Med 1987; 80: 252–3.
49. Meigs JV. Pelvic tumors other than fibromas of the ovary with 60. Hurlow RA, Greening WP, Krantz E. Ascites and hydrothorax in
ascites and hydrothorax. Obstet Gynecol 1954; 3: 471–86. association with stroma ovarii. Br J Surg 1976; 63: 110–2.
50. Yu J, Grimes DA. Ascites and pleural effusions associated with 61. Jimerson SD. Pseudo-Meigs’ syndrome: an unusual case with
endometriosis. Obstet Gynecol 1991; 78: 533–4. analysis of the effusions. Obstet Gynecol 1973; 42: 535–7.
51. Majzlin G, Stevens FL. Meigs’ syndrome: case report and review of 62. Neustadt JE, Levy RC. Hemorrhagic pleural effusion in Meigs’
literature. J Int Coll Surg 1964; 42: 625–30. syndrome. J Am Med Assoc 1968, 204: 179–80.
52. Samanth KK, Black WC III. Benign ovarian stromal tumors 63. Muneyyirci-Delale O, Neil G, Serur E, et al. Endometriosis with
associated with free peritoneal fluid. Am J Obstet Gynecol 1970; massive ascites. Gynecol Oncol 1998; 69: 42–6.
107: 538–45. ◆64. Bhojawala J, Heller DS, Cracchiolo B, Sama J. Endometriosis
53. Kirschner PA. Porous diaphragm syndromes. Chest Surg Clin N Am presenting as bloody pleural effusion and ascites – report of a
1998; 8: 449–72. case and review of the literature. Arch Gynecol Obstet 2000; 264:
54. Ishiko O, Yoshida H, Sumi T, Hirai K, Ogita S. Vascular endothelial 39–41.
growth factor levels in pleural and peritoneal fluid in Meigs’ 65. Shepard MK, Mancini MC, Campbell GD, George RB. Right-sided
syndrome. Eur J Obstet Gynecol Reprod Biol 2001; 98: 129–30. hemothorax and recurrent abdominal pain in a 34-year-old
●55. Abramov Y, Anteby SO, Fasouliotis SJ, Barak V. Markedly elevated woman. Chest 1993; 103: 1239–40.
38
Benign fibrous tumor of the pleura
MARC DE PERROT
HISTORICAL BACKGROUND Table 38.1 Terms given to solitary fibrous tumors of the pleura
(SFTP) over the years
The first report of a primary localized pleural tumor is
attributed to Wagner in 1870.5 However, it was only in Given terms
1931 that Klemperer and Rabin6 published the first accu-
rate pathological description, and classified mesothelioma
Localized mesothelioma
as either ‘localized’ or ‘diffuse’. Since tissue culture7 and
Localized fibrous mesothelioma
ultrastructural analysis8,9 demonstrated the presence of
Localized fibrous tumors
epithelial-like cells in localized mesotheliomas, in 1942
Solitary fibrous mesothelioma
Stout and Murray7 claimed that localized mesothelioma
Fibrous mesothelioma
had a mesothelial origin. However, other investigators
Pleural fibroma
have shown that the mesothelial layer covering the tumor
Submesothelial fibroma
was intact and they postulated that the epithelial cells seen
Subserosal fibroma
could have been trapped within growing fibrous mes-
484 Benign fibrous tumor of the pleura
de Rijn and Rouse17 have described this distinctive group RADIOLOGICAL FEATURES
of cells as ‘dendritic interstitial cells’ and have raised the
possibility that solitary fibrous tumors seen in the lung and Benign and malignant SFTP usually appear as a well-
in other sites may originate from such cells. defined, homogeneous and rounded mass on the initial
chest radiograph.1,19–21 Rarely, a pleural effusion is associ-
ated with malignant SFTP.1,20,21 While small tumors
CLINICAL FEATURES arising from the parietal pleura classically form obtuse
angles with the chest wall, large or pedunculated SFTP may
Solitary fibrous tumors of the pleura have been described form acute angles and be confused with intrapulmonary
in all age groups from 5 to 87 years, but they peak in the masses.27,28 Pedunculated SFTP have occasionally been
sixth and seventh decade of life with an even distribution reported to be moving on successive chest radiographs
between men and women.1,3,18 The majority of benign because of their pedunculated attachment in the fissure of
tumors are small and pedunculated, whereas most malig- the lung.29,30
nant SFTP are large and symptomatic.1,4,19,20 Symptoms Computed tomography (CT) scan usually demon-
usually include cough, chest pain, and dyspnea (Table strates a well delineated, homogeneous and occasionally
38.2). More rarely, hemoptysis and obstructive pneumoni- lobulated mass of soft tissue attenuation.31 Although no
tis are observed as a result of airway obstruction.1,4 specific CT features have been described for the diagnosis
of SFTP, the tumors typically appear in contact with the
pleural surface and show displacement or invasion of the
surrounding structures.20 Heterogeneity may be observed
Table 38.2 Approximate incidence of symptoms with benign and malignant variants of SFTP because of
myxoid degeneration, hemorrhage or necrosis (Figure
Approximate incidence %
38.1). Tumors arising in an interlobar fissure may be more
difficult to differentiate from an intraparenchymal mass,
Asymptomatic 60 because the lesion appears to be surrounded by lung
Chest pain 18 parenchyma.32 Calcifications may be observed in a few
Dyspnea 15 tumors regardless of their benign or malignant histological
Cough 12 features.1,4,20
Weight loss 2 Magnetic resonance imaging (MRI) has limited use in
HPO 2 the assessment of pleural disease.33 However, the morphol-
Hypoglycemiaa 1 ogy and relationship of large SFTP to adjacent mediastinal
Fever 1 and major vascular structures may be better appreciated
a
Doege–Potter syndrome.
with MRI than with CT.34 MRI is helpful in differentiating
HPO, hypertrophic pulmonary osteoarthropathy. the tumor from other structures and in confirming
intrathoracic localization when the tumor abuts the
diaphragm.4,31,35 Fibrous tissue as seen in both benign and The presence of occasional large bizarre cells or focal high
malignant SFTP has low signal intensity on T1-weighted cellularity in the absence of cellular atypia or mitosis is
images.34,36,37 On T2-weighted images, however, mature usually not sufficient to categorize the tumor as malig-
fibrous tissue containing few cells and abundant collagen nant.1,43,46
stroma has a low intensity, whereas malignant fibrosis The use of immunohistochemistry has been an
invariably appears with high signal intensity because of extremely useful tool to differentiate SFTP from mesothe-
increased vascularity, edema and cellularity.34,36,37 liomas and sarcomas over the last few years. Indeed, SFTP
Unfortunately, benign SFTP can often also have areas of by definition is vimentin positive and keratin negative. In
high signal intensity on T2-weighted images because of addition, CD34 is positive in most benign and malignant
intratumoral necrosis or myxoid degeneration and thus SFTP, whereas it remains negative for most other pul-
may not be differentiable from malignant SFTP.34,38,39 monary tumors. Occasionally, malignant SFTP may be
Positron-emission tomography (PET) scan has CD34 negative.13,43 The expression of bcl-2 can be a useful
minimal utility in the evaluation of these tumors. Few marker to confirm the diagnosis of SFTP if CD34 is nega-
cases have been reported in the literature and the tumor tive.47,48
showed no or minimal uptake on PET.40,41 Recently, some authors have demonstrated that CD99
and factor XIIIa could be expressed by solitary fibrous
tumors located in the pleura and in other locations.48–51
HISTOPATHOLOGY However, in contrast to CD34 and bcl-2, CD99 and factor
XIIIa are not strongly expressed by SFTP and can also fre-
Benign and malignant SFTP are widely distributed in the quently be positive with other tumors such as synovial sar-
chest. While most of the benign SFTP are small peduncu- comas or neural tumors and, thus, are less specific for the
lated tumors, the malignant variants are often larger than diagnosis of SFTP.52
10 cm and grow beneath the parietal pleura of the chest
wall, diaphragm or mediastinum.1,4,20,42
Macroscopically, benign and malignant tumors appear MANAGEMENT
as firm, smoothly lobulated masses. Most of them are
encapsulated by a thin, translucent membrane, containing Complete en bloc surgical resection is the mainstay of
a reticulated vascular network. Firm adhesion without therapy for all benign and malignant SFTP. A distance of
signs of invasion may be present between visceral and pari- 1–2 cm from the tumor is usually recommended to have
etal pleura at the surface of the tumor. The cut surface adequate margins. While pedunculated tumors can be
appears gray-white to tan with a whorled pattern, and may safely resected with a wedge resection of the lung, large
disclose areas of hemorrhage and necrosis.1,43 sessile tumors can be difficult to resect because of extensive
Hemorrhagic and necrotic areas may be present in benign adhesions and may occasionally require a lobectomy or a
tumors, but they usually predominate in the malignant pneumonectomy in order to achieve complete resec-
form, most likely because of their larger size.1 tion.53–55 Frozen sections can be helpful to demonstrate
Microscopically, SFTP are characterized by a prolifera- that the resection margins are free of tumors, but are not
tion of uniform elongated spindle cells intimately inter- routinely required.
twining with various amounts of connective tissue. Zones Tumors adherent to the parietal pleura require an
of hypercellularity may alternate with hypocellular or extra-pleural dissection.22,53 However, concomitant chest
fibrous areas within the same tumor. Typically, fibroblasts wall resection can be necessary if the tumor densely
and connective tissue are arranged in a so-called ‘pattern- adheres to or invades the chest wall.19 In 3 percent or less
less pattern’ or ‘storiform pattern’, characterized by a hap- of cases the tumor can be ‘inverted’ and grow inside the
hazard distribution of spindle cells and collagen fibrils.44,45 lung parenchyma (Figure 38.2).56,57 These tumors often
Occasionally, an increased amount of blood vessels within require a lobectomy or a sleeve lobectomy.43,53,58,59 In our
the tumor cause a hemangiopericytoma-like pattern.1,44 experience in Toronto, a sleeve lobectomy was required in
More rarely, other patterns such as herringbone and one case for a solitary fibrous tumor growing from the left
neural palisading are also observed inside the tumor.1 lower lobe bronchus inside the left main bronchus.
Histological signs of malignancy include: Interestingly, the tumor was growing without invading the
main bronchus and could be simply pulled back from the
1. high mitotic counts defined as more than four mitoses main bronchus.52
per 10 high-power field (HPF); Thoracoscopic approaches can be safely used to remove
2. mild to marked pleomorphism based on nuclear size, small pedunculated tumors located on the visceral
irregularity and nuclear prominence; pleura.53,60 Some authors have also recommended the
3. bundles of high cellularity with crowding and overlap- assistance of a video-camera to obtain a more precise view
ping of nuclei; of the resection margins in some large broad-based tumors
4. presence of necrotic and/or hemorrhagic zones; of the parietal pleura.53 Extreme caution should be used to
5. stromal and/or vascular invasion. avoid contact between the tumor and the thoracoscopic
486 Benign fibrous tumor of the pleura
PROGNOSIS
Morphological and histological parameters are important
predictors of outcome.64 In 1981, Briselli and coworkers3
presented eight new cases and reviewed 360 cases from the
literature. Twelve percent of the tumors followed a malig-
nant course and led to death. The authors observed that
the growth pattern of the tumor was more important for
prognosis than the histological characteristics. More
recently, England and coworkers1 reported 223 cases from
the files of the Armed Forced Institute of Pathology, of
which 82 were described as histologically malignant. While
* none of the patients with a histologically benign disease
died, 55 percent of those with a malignant form died
because of recurrences and/or metastases. The authors
observed that among the malignant variants, complete
resection was the single most important predictor of
outcome.
In order to stratify the risk of recurrence, a staging
system was developed after reviewing all publications with
Figure 38.2 Solitary fibrous tumor growing inside the lung adequate follow-up for patients with a diagnosis of SFTP
parenchyma and located at the bifurcation between the proven by histology and immunohistochemistry.52 Among
segmental bronchi of the right upper lobe (*). A lobectomy was a total of 185 reported cases, 35 (19 percent) presented
required to remove this tumor. Histological evaluation with recurrence and 16 of them died (Table 38.3). Of the
demonstrated a benign solitary fibrous tumor of the pleura. The 35 patients who presented with recurrent tumor, 27 had a
risk of recurrence of benign intraparenchymal tumors is estimated primary sessile tumor with histological sign of malignancy,
to be approximately 8 percent. whereas five presented with a primary sessile histologically
benign tumor and two with a primary pedunculated histo-
logically malignant tumor. One patient with a peduncu-
lated histologically benign tumor presented a recurrence
sites, since contact metastasis and local recurrence at the (or a new primary tumor) and died 10 years after the initial
port sites have been reported.53,54 resection.43
The risk of recurrence was therefore observed to be the
highest in patients with malignant sessile tumors (esti-
ADJUVANT THERAPY mated to be approximately 63 percent) and this type of
SFTP were classified as stage III. Patients with malignant
The role of adjuvant therapy in SFTP has not been system- pedunculated tumor were found to have a 14 percent risk
atically explored because of the limited number of of recurrence and were classified as stage II. Patients with
patients.61 However, evidence suggests that radiotherapy benign sessile tumor had an 8 percent risk of recurrence
and chemotherapy could be beneficial in some patients. and were classified as stage I. Patients presenting with
Suter and colleagues19 have reported one patient who is benign pedunculated tumor were found to have a risk of
alive with no evidence of disease more than 20 years after recurrence of less than 2 percent and were considered as
subtotal resection of the tumor followed by radiotherapy, benign, therefore classified as stage 0 (Table 38.4).
and Veronesi and colleagues62 have observed significant Some authors observed that tumors larger than 10 cm
reduction of an inoperable recurrent SFTP with ifosfamide had a worse prognosis than smaller tumors.65,66 However,
and adriamycine. Although neoadjuvant therapy could be the worse prognosis is related to the higher frequency of
helpful in large malignant tumors, its use is limited by the malignant histological characteristics present in these large
difficulty to obtain a precise preoperative diagnosis even tumors and to the higher rate of incomplete resection in
with an open biopsy.62 Currently, we would recommend this group of patients. In our review of the literature,52 we
the administration of adjuvant therapy after resection of observed that all patients with a histologically benign
malignant sessile tumors, in particular if they are recur- tumor larger than 10 cm that was completely resected had
rent. a good clinical outcome, whereas 16 out of the 28 tumors
Additional therapies such as brachytherapy and photo- larger than 10 cm with malignant characteristics were
dynamic therapy have been developed for malignant associated with recurrence. Hence, although the majority
mesothelioma and could be applied for other pleural of malignant tumors are larger than 10 cm, histological
tumors. However, their use in SFTP has rarely been characteristics and resectability rather than size are the
reported and their utility remains unproven.61,63 principal indicators of clinical outcome.
Future directions of development 487
Table 38.3 Summary of recent publications on solitary fibrous tumors of pleura (SFTP)a
FOLLOW-UP CONCLUSIONS
Recurrence after surgical resection is most often located in Solitary fibrous tumor of the pleura is a mesenchymal
the same hemithorax and has been reported up to 17 years tumor that has been increasingly recognized over the past
after resection.21 Intrathoracic recurrence may be fatal few years. The tumor was initially described in the pleura
because of mediastinal compression and inferior vena cava but has been reported in many other sites lately. Although
obstruction.3,67 Metastases, if present, are usually blood- the majority of these tumors have a benign course, the
borne and located by order of frequency, in the liver, malignant form still remains enigmatic. Their behavior is
central nervous system, spleen, peritoneum, adrenal gland, often unpredictable and does not always correlate with his-
gastrointestinal tract, kidney and bone.1 tological findings. In addition, benign tumors may remain
The risk of recurrence is high after resection of malig- silent for several years before degenerating into a malig-
nant sessile SFTP. However, most of the recurrences are nant form. Complete en bloc surgical resection is the main-
initially located inside the pleural cavity and distant metas- stay of therapy for all benign and malignant SFTP.
tasis seems to be a late event in the evolution of the disease. Histological and morphological characteristics are the
Recurrences are often associated with a progression in the principal indicators of clinical outcome.
degree of malignancy.19,43 The majority of recurrence after
resection of malignant sessile tumors occurs within the
first 24 months after the initial resection. Hence, half-
yearly radiological control with chest X-ray or CT scan
during the initial 2 years after the resection and yearly FUTURE DIRECTIONS OF DEVELOPMENT
thereafter seems warranted and may help to reduce the
mortality from malignant SFTP. In case of recurrence, Better understanding of the molecular mechanisms
aggressive surgical resection remains the treatment of leading to the malignant form of solitary fibrous tumors
choice and may lead to long-term survival.20 Adjuvant will be important in the future. This may help to develop
therapy should be considered if the tumor appears histo- novel targeted therapy for patient at risk of recurrence or
logically malignant. with advanced disease.
488 Benign fibrous tumor of the pleura
34. Ferretti GR, Chiles C, Cox JE, Choplin RH, Coulomb M. Localized 51. Alawi F, Stratton D, Freedman PD. Solitary fibrous tumor of the
benign fibrous tumors of the pleura: MR appearance. J Comput oral soft tissues. A clinicopathologic and immunohistochemical
Assist Tomogr 1997; 21: 115–20. study of 16 cases. Am J Surg Pathol 2001; 25: 900–10.
35. Norton SA, Clark SC, Sheehan AL, Ibrahim NBN, Jeyasingham K. ◆52. de Perrot M, Fischer S, Brundler MA, Sekine Y, Keshavjee S.
Solitary fibrous tumour of the diaphragm. J Cardiovasc Surg 1997; Solitary fibrous tumors of the pleura. Ann Thorac Surg 2002; 74:
38: 685–6. 285–93.
36. Arrive L, Hricak H, Tavares NJ, Miller TR. Malignant versus 53. Cardillo G, Facciolo F, Cavazzana AO, et al. Localized (solitary)
nonmalignant retroperitoneal fibrosis: differentiation with MR fibrous tumors of the pleura: an analysis of 55 patients. Ann
imaging. Radiology 1989; 172: 139–43. Thorac Surg 2000; 70: 1808–12.
37. Falaschi F, Battolla L, Mascalchi M, et al. Usefullness of MR signal: 54. Nomori H, Horio H, Fuyuno G, Morinaga S. Contacting metastasis
intensity in distinguishing benign from malignant pleural disease. of a fibrous tumor of the pleura. Eur J Cardiothorac Surg 1997; 12:
AJR Am J Roentgenol 1966; 166: 963–8. 928–30.
38. Verluis PJ, Lamers RJ. Localized pleural fibroma: radiological 55. Sung SH, Chang JW, Kim J, et al. Solitary fibrous tumors of the
features. Eur J Radiol 1994; 18: 124–5. pleura: surgical outcome and clinical course. Ann Thorac Surg
39. Harris GN, Rozenstein A, Schiff MJ. Benign fibrous mesothelioma 2005; 79: 303–7.
of the pleura: MR imaging findings. AJR Am J Roentgenol 1995; 56. Yousem SA, Flynn SD. Intrapulmonary localized fibrous tumor.
165: 1143–4. Intraparenchymal so-called localized fibrous mesothelioma. Am J
40. Alexander M, Yang S, Yung R, Brasic JR, Pannu H. Diagnosis of Clin Pathol 1988; 89: 365–9.
benign solitary fibrous tumors by positron emission tomography. 57. Patsios D, Hwang DM, Chung TB. Intraparenchymal solitary fibrous
South Med J 2004; 97: 1264–7. tumor of the lung: an uncommon cause of a pulmonary nodule. J
41. Kramer H, Pieterman RM, Slebos DJ, et al. PET for the evaluation Thorac Imaging 2006; 21: 50–3.
of pleural thickening observed on CT. J Nucl Med 2004; 45: 995–8. 58. Khalifa M, Montgomery EA, Azumi N, et al. Solitary fibrous
42. Witkin GB, Rosai J. Solitary fibrous tumor of the mediastinum. A tumors: a series of lesions, some in unusual sites. South Med J
report of 14 cases. Am J Surg Pathol 1989; 13: 547–57. 1997; 90: 793–9.
43. Yokoi T, Tsuzuki T, Yatabe Y, et al. Solitary fibrous tumour: 59. Aufiero TX, McGary SA, Campbell DB, Phillips PP. Intrapulmonary
significance of p53 and CD34 immunoreactivity in its malignant benign fibrous tumor of the pleura. J Thorac Cardiovasc Surg
transformation. Histopathology 1998; 32: 423–32. 1995; 110: 549–51.
44. Moran CA, Suster S, Koss MN. The spectrum of histologic growth 60. Sanguinetti CM, Marchesani F, Ranaldi R, Pela R, Cecarini L.
patterns in benign and malignant fibrous tumours of the pleura. Localized fibrous pleural tumour of the interlobular pleura. Eur
Semin Diagn Pathol 1992; 9: 169–80. Respir J 1996; 9: 1094–6.
45. Stout AP. Tumors of the pleura. Harlem Hosp Bull 1971; 5: 54–7. ∗61. de Perrot M. Fibrous tumors of the pleura. Curr Treat Options
46. Hanau CA, Miettinen M. Solitary fibrous tumor: histological and Oncol 2000; 1: 293–8.
immunohistochemical spectrum of benign and malignant variants 62. Veronesi G, Spaggiari L, Mazzarol G, et al. Huge malignant
presenting at different sites. Hum Pathol 1995; 26: 440–9. localized fibrous tumor of the pleura. J Cardiovasc Surg 2000; 41:
47. Chilosi M, Facchetti F, Dei Tos AP, et al. bcl-2 expression in pleural 781–4.
and extrapleural solitary fibrous tumours. J Pathol 1997; 181: 63. Martini N, McCormack PM, Bains MS, et al. Pleural mesothelioma.
362–7. Ann Thorac Surg 1987; 43: 113–20.
48. Suster S, Fisher C, Moran CA. Expression of bcl-2 oncoprotein in 64. Chan JKC. Solitary fibrous tumour – everywhere, and a diagnosis
benign and malignant spindle cell tumors of soft tissue, skin, in vogue. Histopathology 1997; 31: 568–76.
serosal surfaces, and gastrointestinal tract. Am J Surg Pathol 65. Brozzetti S, D’Andrea N, Limiti MR, et al. Clinical behavior of
1998; 22: 863–72. solitary fibrous tumors of the pleura. An immunohistochemical
49. Vallat-Decouvelaere AV, Dry SM, Fletcher CDM. Atypical and study. Anticancer Res 2000; 20: 4701–6.
malignant solitary fibrous tumors in extrathoracic locations. 66. Chang YL, Lee YC, Wu CT. Thoracic solitary fibrous tumor: clinical
Evidence of their comparability to intra-thoracic tumors. Am J and pathological diversity. Lung Cancer 1999; 23: 53–60.
Surg Pathol 1998; 22: 1501–11. 67. Dalton WT, Zolliker AS, McCaughey WTE, Jacques J, Kannerstaein
50. Mentzel T, Bainbridge TC, Katenkamp D. Solitary fibrous tumour: M. Localized primary tumors of the pleura. An analysis of 40 cases.
clinicopathological, immunohistochemical, and ultrastructural Cancer 1979; 44: 1465–75.
analysis of 12 cases arising in soft tissues, nasal cavity and
nasopharynx, urinary bladder and prostate. Virchows Arch 1997;
430: 445–53.
This page intentionally left blank
39
Undiagnosed pleural effusions
NICK A MASKELL
Table 39.3 Some of the common conditions causing an sistent undiagnosed effusions because they are both life-
exudative pleural effusion threatening and treatable. The effusions are usually unilat-
Common conditions
eral and occupy less than one-third of the hemithorax.3 In
80 percent of cases they are exudates, 20 percent transu-
dates and the pleural fluid is often blood stained. The
Parapneumonic effusion Pulmonary infarction pleural fluid is often lymphocytic, but there are no charac-
Malignancy Connective tissue diseases teristic features which enable pulmonary embolus to be
Tuberculosis Fungal infections completely ruled out. Common presenting features
Drugs Benign asbestos effusion include pleuritic chest pain with associated shortness of
Post-cardiac injury syndrome (PCIS) Yellow nail syndrome breath.5 If suspected, a D-dimer measurement is a useful
Rheumatoid arthritis Pancreatitis starting point, followed by a computer tomography pul-
Trapped lung (when inflammatory monary angiogram (CTPA) if the D-dimer test is positive.6
process not remote) (Further discussion can be found in Chapter 30, Effusions
from vascular causes.)
Is it an exudate or transudate
(use Light’s criteria)
Exudate Transudate
Could pulmonary embolus be a possibility. If so, consider Could pulmonary embolus be a possibility. If so, consider
a d-dimer and if positive a CTPA scan to exclude a d-dimer and if positive a CTPA scan to exclude
(80% are exudates) (20% are transudates)
(a) (c)
Tuberculous pleurisy
True chylous effusions result from disruption of the tho- of triglycerides. Chylothorax must be distinguished from
racic duct or its tributaries. This leads to the presence of pseudochylothorax or ‘cholesterol pleurisy’ which is due
chyle in the pleural space. Approximately 50 percent are to the accumulation of cholesterol crystals in a long-
due to malignancy (particularly lymphoma), 25 percent standing pleural effusion. In these cases the pleura is
trauma (especially during surgery) and the rest are miscel- usually markedly thickened and fibrotic.24
laneous causes such as TB, sarcoid, amyloidosis, aneurysm In the past, the most common causes of a pseudochy-
of the thoracic aorta, lymphangioleiomyomatosis and lous effusion were tuberculosis and artificial pneumotho-
tuberous sclerosis.23 The pleural fluid contains high levels rax. Chronic rheumatoid pleurisy is now the usual cause.
496 Undiagnosed pleural effusions
These two types of effusion can be discriminated by lipid mild weight loss. The pleural fluid aspirated may also not
analysis of the fluid. A true chylothorax will usually have a reveal any clues as it might still be straw colored even it is
pleural fluid triglyceride level of >1.24 mmol/L has been present for some time. The diagnostic problems
(110 mg/dL), and can usually be excluded if the triglyc- do not stop there as up to 40 percent of pleural fluid cul-
eride is <0.56 mmol/L (50 mg/dL). Asking the laboratory tures are negative. Sometimes the diagnosis is only made at
to look for the presence of chylomicrons is useful to the time of video-assisted thoracoscopic surgery (VATS)
confirm the diagnosis of chylothorax. In a pseudochy- where biopsies show evidence of a chronic empyema
lothorax the cholesterol level is greater than 5.18 mmol/L rather than the malignancy. The key message here is to
(200 mg/dL), chylomicrons are not found and cholesterol keep an open mind and always ask the question ‘could this
crystals are often seen at microscopy.25 Effusions from dis- be due to infection?’
ruption of the lymphatics are discussed in Chapter 29.
Cardiac failure
Pleural malignancy
Cardiac failure is the most common cause of a transuda-
Many undiagnosed effusions will eventually turn out to be tive effusion (Chapter 24). Usually, this diagnosis is made
a malignant process. As cytology alone has a sensitivity of clinically and the chest X-ray reveals bilateral effusions.
only 60 percent and often the CT thorax does not reveal an However, they can occur unilaterally, and when they do
area of focal pleural thickening to undertake a percuta- they are twice as common on the right side.29
neous biopsy, the clinician will often find himself in the Although these effusions are usually transudates, they
position where malignancy is suspected but the patient is occasional present as exudates, particularly in patients
too frail for thoracoscopy. In these situations, watchful already on diuretics. The predominant cell count can be
waiting in the outpatient setting is often appropriate. lymphocytic and the fluid often also contains mesothelial
Thoracoscopy remains the diagnostic tool of choice for cells. Plasma brain natriuretic peptide (BNP) level can be
undiagnosed effusions where malignancy is suspected. very useful in diagnosing these less obvious cases.
Harris et al.26 described 182 consecutive patients who Measuring pleural fluid BNP level does not convey any
underwent thoracoscopy over a 5-year period and showed additional benefit to the plasma level alone.30,31
it to have a diagnostic sensitivity of 95 percent for malig-
nancy. Malignancy was shown by thoracoscopy in
66 percent of patients who had previously had a non-
Hepatic hydrothorax
diagnostic closed pleural biopsy and in 69 percent of
patients who had had two negative pleural cytological
specimens.26 A similar sensitivity for malignant disease was Some patients present with a hepatic hydrothorax
described by Page et al.27 in 121 patients with undiagnosed without ascites and, therefore, needs to be thought about
effusion. In addition, it has a very low complication rate. in persistently undiagnosed pleural effusions. In one
In patients with a prior asbestos history, mesothelioma series 27 of the 28 cases were right-sided. The explanation
should always be considered. In this setting the yield from of the pleural effusion in the absence of overt ascites is
cytology alone can be as low as 25 percent and thora- that patients have defects in their diaphragm and the
coscopy improves this to 95 percent in some series.27 In fluid flows into the pleural space because of negative
addition to obtaining a tissue diagnosis, several liters of pleural pressure. Hepatic hydrothorax is secondary to the
fluid can be removed during the procedure and the oppor- passage of ascites through diaphragmatic defects and the
tunity is also provided for talc pleurodesis. Thoracoscopy ascites is secondary to portal hypertension and salt reten-
may therefore be therapeutic as well as diagnostic28 (see tion. First-line treatment is sodium restriction and spiro-
also Chapters 47 and 48). lactone.
The vast majority of these cases present with classical This causes bilateral transudative effusions in 25 percent of
symptoms of infection and cause no diagnostic difficulty. patients. Checking the patients’ renal function, serum
The patient will often have a fever, high inflammatory protein and urine for protein is important if this condition
markers and a chest radiograph showing consolidation is suspected. In addition patients with nephrotic syndrome
and an effusion. The management and investigation of are at increased risk of pulmonary embolus so a low
these have been covered in Chapter 26. threshold for CTPA is needed.32
It is, however, important to state that occasionally the Table 39.4 summarizes some useful additional tests to
presentation will be much less clear cut. There may be no order in selected cases of persistent undiagnosed pleural
fever or sweats, just a history of generalized lethargy and effusions.
References 497
WHAT TO DO IF TESTS REMAIN PERSISTENTLY patients are frail and watchful waiting is usually appropri-
NEGATIVE – A SUGGESTED MANAGEMENT ate.
PLAN
REFERENCES
SUMMARY
● = Key primary paper
After the standard initial workup for undiagnosed pleural ◆ = Major review article
effusions, the cause is undetermined in around 15 percent
of cases. When faced with this problem it is sensible to first 1. Turton CW Troublesome pleural fluid. Br J Dis Chest 1987; 81:
revisit the history (including drugs) and clinical examina- 217–23.
tion. The next step is to re-evaluate whether it is a transu- 2. Light RW, MacGregor MI, Luchsinger PC, Ball WCJ. Pleural
date or exudate by Light’s criteria. This will help channel effusions: the diagnostic separation of transudates and exudates.
Ann Intern Med 1972; 77: 507–13.
further investigations. ◆3. Light RW Pleural effusion due to pulmonary emboli. Curr Opin
All treatable causes of pleural effusions need to be con- Pulm Med 2001; 7: 198–201.
sidered and ruled out. These include pulmonary embolus ●4. Marel M, Arusiova M, Stasny B, et al. Incidence of pleural effusion
where a CTPA scan needs to be performed if suspected. in a well defined region: epidemiologic study in Central Bohemia.
Serum BNP levels are another useful investigation Chest 1993: 104: 1466–99.
5. Fedullo PF. Pulmonary thromboembolism. In: Murray JF, Nadel JA
where congestive heart failure is thought possible. Special (eds). Textbook of respiratory medicine. Philadelphia: WB
tests to exclude TB and lymphoma may also be necessary. Saunders, 2000: 1503–31.
Unfortunately, many of the remaining effusions will 6. Goodman PC. Spiral CT for pulmonary embolism. Semin Resp Crit
eventually turn out to be due to malignancy. Often these Care Med 2000; 21: 508–10.
498 Undiagnosed pleural effusions
7. Light RW, Rogers JT, Moyers JP, et al. Prevalance and clinical 21. Leechawengwong M, Berger HW, Sukumaran M. Diagnostic
course of pleural effusions at 30 days post coronary artery bypass significance of antinuclear antibodies in pleural effusion. Mt Sinai
surgery. Am J Respir Crit Care Med 2002: 166: 1563–6. J Med 1979: 46: 137–9
8. Meigs JV, Cass JW. Fibroma of the ovary with ascities and ●22. Joseph J, Sahn SA. Connective tissue diseases and the pleura.
hydrothorax. Am J Obstet Gynecol 1937; 33: 249–67. Chest 1993; 104: 262–70.
◆9. Sahn SA. The pleura – state of the art. Am Rev Respir Dis 1988; 23. Turton CW. Troublesome pleural fluid. Br J Dis Chest 1987; 81:
138: 184–23. 217–24.
10. Majzlin G, Stevens FL. Meigs’ syndrome: case report and review of 24. Hillerdal G. Chylothorax and pseudochylothorax. Eur Respir J
literature. J Int Coll Surg 1964: 42: 625–30. 1997; 10: 1157–62
11. Meigs JV. Fibroma of the ovary with ascities and hydrothorax – 25. Romero S, Martin C, Hernandez L, et al. Chylothorax in cirrhosis of
Meigs’ syndrome. Am J Obstet Gynecol 1954: 67: 962–87. the liver: analysis of its frequency and clinical characteristics.
12. Robinson BWS, Musk AW. Benign asbestos pleural effusion: Chest 1998; 114: 154–9.
diagnosis and course. Thorax 1981; 36: 896–90. 26. Harris RJ, Kavuru MS, Rice TW, Kirby TJ. The diagnostic and
●13. Joseph J, Strange C, Shan SA. Pleural effusions in hospitalised therapeutic utility of thoracoscopy. A review. Chest 1995; 108:
patients with AIDS. Ann Intern Med 1983; 118: 856–859. 828–41.
14. Moore PJ, Thomas PA. The trapped lung with chronic pleural space, ◆27. Page RD, Jeffrey RR, Donnelly RJ. Thoracoscopy: a review of 121
a cause of recurring pleural effusion. Military Med 1967; 132: consecutive surgical procedures. Ann Thorac Surg 1989; 48:
998–1002. 66–8.
15. Farber JE, Lincoln NS. The unexpandable lung. I. Statement of the 28. Loddenkemper R. Thoracoscopy – state of the art. Eur Respir J
problem. Am Rev Tuberc 1939: 40: 704–9. 1998; 11: 213–21.
16. Berger HW, Mejia E. Tuberculous pleurisy. Chest 1973; 63: 88–92. 29. Race GA, Scheifley CH, Edwards JE. Hydrothorax in congestive
17. Idell S. Evaluation of perplexing pleural effusions. Ann Intern Med heart failure. Am J Med Sci 1947; 214: 243–7.
1994; 110: 567–9. 30. Gegenhuber A, Muller T, Dieplinger B, et al. Plasma B-type
18. Ocana IM, Martinez Vazquez JM, Seguna R, et al. Adenosine natriuretic peptide in patients with pleural effusions. Chest 2005;
deaminase in pleural fluids. Chest 1983; 84: 51–3. 128: 1003–9.
19. Idell S. Granulomatous disease of the pleura. Semin Respir Med ●31. Muller T, Haltmayer,M. Natriurectic peptide measurements as part
1994; 6: 31–9. of the diagnostic work up in pleural effusions. Eur Resp J 2006;
20. Good JT Jr, King TE, Antony VB, et al. Lupus pleuritis. Clinical 28: 7–9.
features and pleural fluid characteristics with special reference to ◆32. Light RW The undiagnosed pleural effusion. Clin Chest Med 2006;
pleural fluid antinuclear antibodies. Chest 1983; 84: 714–18. 27: 309–19.
40
Asbestos-related pleural diseases
● circumscribed pleural plaques (CPP); thickening in which the lung tissues are drawn in to the
● diffuse pleural thickening (DPT); pleural or subpleural fibrosis.
● benign asbestos pleural effusion (BAPE); Transpulmonary bands are coarse bands of fibrosis
● rolled atelectasis (RA); which also occur in the lung parenchyma in the presence
● transpulmonary bands (TPB). of diffuse pleural thickening.
become visible on plain chest x-rays within the first 10 It is widely accepted that circumscribed plaques form as
years following initial exposure to asbestos but do not sub- a direct result of the presence of asbestos fibers inciting an
sequently progress in size/extent. However, they may inflammatory response in the parietal pleura, possibly as a
become more easily seen on radiographs due to increasing result of fibers protruding from the visceral pleural surface
calcification.16,17 Initially the calcification is punctate but causing mechanical irritation during the movement of the
becomes denser with time. lung during respiration,23 even though pathological data
Talc and kaolin exposure may also cause pleural has not shown that there are fibers projecting from the vis-
plaques although with talc this may be a result of asbestos ceral pleura or even that fibers are long enough to be
contamination. anchored in the visceral pleura to cause this process to
occur. It has also been suggested that fibers reach the
Benign asbestos pleural effusion pleural space by a transpleural route and enter the parietal
pleura through (hypothetical) stomata in the pleural
surface or by retrograde drainage from and intercostal
Benign asbestos pleural effusions tend to occur earlier fol-
lymphatics.23
lowing asbestos exposure than mesothelioma, at least in
Many cases of diffuse pleural thickening appear to
crocidolite exposed subjects.16,18 Their average latency is
follow benign asbestos pleural effusion16 and it is thought
also shorter than is the development of pleural plaques or
that this is the most likely inflammatory pathway for all
asbestosis. The incidence of benign asbestos pleural effu-
cases.23
sions16,19 also increases with the dose of asbestos exposure.
Pleural fibrosis (hyaline or diffuse) appears to resemble
Cookson et al.16 showed that the occurrence of benign
fibrosis in other tissues with excessive deposition of matrix
asbestos pleural effusion in the Wittenoom cohort was
components and destruction of normal architecture.24
uncommon beyond 25 years after first exposure to amphi-
Complex interactions have been demonstrated between
bole asbestos (the median time following first exposure
resident and imported inflammatory cells, mediators,
was 16 years). This contrasts with malignant mesothe-
coagulant and fibrinolytic pathways etc (see also Chapter
lioma, the incidence of which rises exponentially with
10, Pleural reactions to mineral dusts).
increasing time after initial exposure to asbestos (mainly if
Rolled atelectasis has been postulated as most likely to
not exclusively the amphibole varieties).20
occur as a result of a localized low-grade inflammatory
reaction in the pleura involving both pleural surfaces with
Diffuse pleural thickening progressive thickening and contraction of the fibrosis
resulting in retraction and contraction of the underlying
Diffuse pleural thickening appears more closely related to lung tissue. The indrawn lung takes its airway and vessels
amphibole than chrysotile exposure.8 It is more common with it producing a characteristic (almost diagnostic)
in those more heavily exposed and increases in incidence ‘comet tail’ appearance radiographically.25
with increasing time since first exposure.17,21 There is no
consistent relationship between smoking status and diffuse
pleural thickening.
Rolled atelectasis
AETIOLOGY/PATHOGENESIS
The main specific management issue with DPT is its diag- The banning of the use of amphibole asbestos worldwide
nostic differentiation from primary or secondary malig- would significantly reduce the future occurrence of
nancy of the pleura, especially desmoplastic malignant asbestos-related pleural diseases if the ban were observed.
mesothelioma. It is important to be precise as manage- The ongoing use of chrysotile, however, means that such
ment is determined by the cause of the pleural thicken- cases will still present in future. The study of molecular
ing. Mesothelioma is increasingly attracting treatment biology has provided some hints to the etiology of
with chemotherapy (despite an abysmal response rate) asbestos-related pleural diseases. Recently, early evidence
and even surgery (despite the absence of any scientifically from genetic studies suggest that genetic predisposition to
valid evidence that it improves quality of life or survival). the effects of asbestos exposure may exist and, in future,
Also, secondary tumors (especially breast and prostate may serve as a way to identify individuals at risk.
cancers) in the pleura may require specific treatment
measures.
Decortication of the lung to improve lung volumes and KEY POINTS
ventilatory capacity has not been a useful procedure.
Unlike pleural thickening in tuberculosis or other infective ● The main significance of circumscribed pleural
causes of pleural thickening following empyema, a plane of plaques is that they are indicators of past asbestos
dissection/cleavage does not exist, making hemostasis exposure. They cause minor impairment of lung
difficult and the response of lung volume unsatisfactory. function and are statistically associated with
chest pain which has features of angina.
● Diffuse pleural thickening involves the visceral
Benign asbestos pleural effusion and parietal pleura and may thereby cause signif-
icant impairment of lung function and give rise
The main management issues with BAPE relate to estab- to exertional breathlessness.
lishing its diagnosis by exclusion of other disease ● Benign asbestos pleural effusion is an exudative
processes, the control of dyspnea by pleural aspiration or effusion which follows asbestos exposure but
drainage, and very occasionally the need to perform some which has no specific diagnostic features and is
form of pleurodesis to prevent further recurrence. The therefore a diagnosis of exclusion.
pleurodesis may be performed surgically at open thoraco- ● Benign asbestos pleural effusion may be recur-
tomy (pleurectomy, pleural abrasion or talc – or other rent and may be followed by diffuse pleural
agent – insufflation), at pleuroscopy (now usually video- thickening.
assisted), also with talc insufflation or simply with talc
slurry following tube drainage.
There are no specific management issues relating to these ● = Key primary paper
entities. The main issue with RA is the exclusion of a ◆ = Major review article
peripheral lung cancer. While the radiological features are
often sufficiently characteristic to permit confidence in the 1. Pooley F. Asbestos mineralogy. In: Autman K, Aisner J (eds).
Asbestos-related malignancy. Orlando: Grune and Stratton 1987:
diagnosis, fine needle aspiration biopsy may be indicated if 3–30.
there is doubt and follow-up radiographic (usually CT) 2. World Health Organization. Asbestos and other mineral fibers.
examination is usually suggested for reassurance of the Geneva: WHO 1986.
patient and the physician. 3. Davis J. Mineral fiber carcinogenesis: experimental data relating to
References 505
the importance of fiber type, size, deposition, dissolution and 26. Hillerdal G. The pathogenesis of pleural plaques and pulmonary
migration. Lyon: IARC Scientific Publications, 1989: 33–45. asbestosis: possibilities and impossibilities. Am Rev Respir Dis
4. Oberdorster G. Determinants of the pathogenicity of man-made 1980; 61: 129–38.
vitreous fibers (MMVF). Int Arch Occup Environ Health 2000; 73 ●27. Mukherjee S, de Klerk N, Palmer LJ, et al. Chest pain in asbestos-
(Suppl): S60–8. exposed individuals with benign pleural and parenchymal disease.
5. Rockhoff SD, Chu J, Rubin LJ. Special report: asbestos induced Am J Respir Crit Care Med 2000; 162: 1807–11.
pleural plaques; a disease process associated with ventilatory ●28. Kilburn KH, Warshaw R. Pulmonary functional impairment
impairment and respiratory symptoms. Clin Pulm Med 2002; 9: associated with pleural asbestos disease. Circumscribed and
113–24. diffuse thickening. Chest 1990; 98: 965–72.
6. Roach HD, Davies GJ, Attanoos R, et al. Asbestos: when the dust 29. Hilt B, Lien JT, Lund-Larsen PG. Lung function and respiratory
settles an imaging review of asbestos-related disease. symptoms in subjects with asbestos-related disorders: a cross-
Radiographics 2002; 22: S167–84. sectional study. Am J Ind Med 1987; 11: 517–28.
●7. Albelda SM, Epstein DM, Gefter WB, Miller WT. Pleural thickening: 30. Peacock C, Copley SJ, Hansell DM. Asbestos-related benign pleural
its significance and relationship to asbestos dust exposure. Am disease. Clin Radiol 2000; 55: 422–32.
Rev Respir Dis 1982; 126: 621–4. 31. Ameille J, Brochard P, Brechot JM, et al. Pleural thickening: a
●8. Hillerdal G, Musk AW. Pleural lesions in crocidolite workers from comparison of oblique chest radiographs and high-resolution
Western Australia. Br J Ind Med 1990; 47: 782–3. computed tomography in subjects exposed to low levels of
●9. Hillerdal G, Zitting A, van Assendelft AH, Kuusela T. Rarity of asbestos pollution. Int Arch Occup Environ Health 1993; 64:
mineral fiber pleurisy among persons exposed to Finnish 545–8.
anthophyllite and with low risk of mesothelioma. Thorax 1984; 39: 32. International Labour Office. Guidelines for the use of ILO
608–11. international classification of radiographs of pneumoconiosis.
10. Evans AL, Gleeson FV. Radiology in pleural disease: state of the art. Revised 1980. Geneva: ILO, 1980.
Respirology 2004; 9: 300–12. ●33. Lee YC, Runnion CK, Pang SC, de Klerk NH, Musk AW. Increased
11. Hillerdal G. Pleural changes and exposure to fibrous minerals. body mass index is related to apparent circumscribed pleural
Scand J Work Environ Health 1984; 10: 473–9. thickening on plain chest radiographs. Am J Ind Med 2001; 39:
12. Craighead JE, Abraham JL, Churg A, et al. The pathology of 112–16.
asbestos-associated diseases of the lungs and pleural cavities: 34. Webb WR, Muller NL, Naidich DP. Standardized terms for high-
diagnostic criteria and proposed grading schema. Report of the resolution computed tomography of the lung: a proposed glossary.
Pneumoconiosis Committee of the College of American J Thorac Imaging 1993; 8: 167–75.
Pathologists and the National Institute for Occupational Safety 35. Boraschi P, Neri S, Braccini G, et al. Magnetic resonance
and Health. Arch Pathol Lab Med 1982; 106: 544–96. appearance of asbestos-related benign and malignant pleural
13. Hillerdal G, Lindgren A. Pleural plaques: correlation of autopsy diseases. Scand J Work Environ Health 1999; 25: 18–23.
findings to radiographic findings and occupational history. Eur J 36. Kramer H, Pieterman RM, Slebos DJ, et al. PET for the evaluation
Respir Dis 1980; 61: 315–19. of pleural thickening observed on CT. J Nucl Med 2004; 45:
14. Karjalainen A, Karhunen PJ, Lalu K, et al. Pleural plaques and 995–8.
exposure to mineral fibers in a male urban necropsy population. 37. Duysinx B, Nguyen D, Louis R, et al. Evaluation of pleural disease
Occup Environ Med 1994; 51: 456–60. with 18–fluorodeoxyglucose positron emission tomography
15. Zitting AJ, Karjalainen A, Impivaara O, et al. Radiographic small imaging. Chest 2004; 125: 489–93.
lung opacities and pleural abnormalities in relation to smoking, 38. Kwek BH, Aquino SL, Fischman AJ. Fluorodeoxyglucose positron
urbanization status, and occupational asbestos exposure in emission tomography and CT after talc pleurodesis. Chest 2004;
Finland. J Occup Environ Med 1996; 38: 602–9. 125: 2356–60.
●16. Cookson WO, De Klerk NH, Musk AW, et al. Benign and malignant 39. Van Cleemput J, De Raeve H, Verschakelen JA, et al. Surface of
pleural effusions in former Wittenoom crocidolite millers and localized pleural plaques quantitated by computed tomography
miners. Aust N Z J Med 1985; 15: 731–7. scanning: no relation with cumulative asbestos exposure and no
●17. de Klerk NH, Cookson WO, Musk AW, Armstrong BK, Glancy JJ. effect on lung function. Am J Respir Crit Care Med 2001; 163:
Natural history of pleural thickening after exposure to crocidolite. 705–10.
Br J Ind Med 1989; 46: 461–7. 40. Broderick A, Fuortes LJ, Merchant JA, Galvin JR, Schwartz DA.
●18. Robinson BW, Musk AW. Benign asbestos pleural effusion: Pleural determinants of restrictive lung function and respiratory
diagnosis and course. Thorax 1981; 36: 896–900. symptoms in an asbestos-exposed population. Chest 1992; 101:
●19. Epler GR, McLoud TC, Gaensler EA. Prevalence and incidence of 684–91.
benign asbestos pleural effusion in a working population. J Am 41. Cotes JE, King B. Relationship of lung function to radiographic
Med Assoc 1982; 247: 617–22. reading (ILO) in patients with asbestos related lung disease. Thorax
◆20. De Klerk NH, Armstrong BK. The epidemiology of asbestos and 1988; 43: 777–83.
mesothelioma. In: Henderson DW, Shilkin KB, Langlois SL, 42. Cookson WO, Musk AW, Glancy JJ. Pleural thickening and gas
Whitaker D (eds). Malignant mesothelioma. New York: transfer in asbestosis. Thorax 1983; 38: 657–61.
Hemisphere, 1992: 223–50. 43. Cotes JE, Steel J. Work-related lung disorders. London: Blackwell,
◆21. Rudd R. Benign pleural disease. In: Hendrick DJ, Burge PS, Beckett 1987.
WS, Churg A (eds). Occupational disorder of the lung recognition, 44. Sandrini A, Johnson AR, Thomas PS, Yates DH. Fractional exhaled
management and prevention. London: WB Saunders and Co, 2002: nitric oxide concentration is increased in asbestosis and pleural
343–57. plaques. Respirology 2006; 11: 325–9.
◆22. Rizzo P, Bocchetta M, Powers A, et al. SV40 and the pathogenesis ◆45. American Thoracic Society. Diagnosis and initial management of
of mesothelioma. Semin Cancer Biol 2001; 11: 63–71. nonmalignant diseases related to asbestos. Am J Respir Crit Care
23. Huggins JT, Sahn SA. Causes and management of pleural fibrosis. Med 2004; 170: 691–715.
Respirology 2004; 9: 441–7. ●46. Reid A, de Klerk N, Ambrosini G, et al. The additional risk of
24. Mutsaers SE, Prele CM, Brody AR, Idell S. Pathogenesis of pleural malignant mesothelioma in former workers and residents of
fibrosis. Respirology 2004; 9: 428–40. Wittenoom with benign pleural disease or asbestosis. Occup
◆25. Cugell DW, Kamp DW. Asbestos and the pleura: a review. Chest Environ Med 2005; 62: 665–9.
2004; 125: 1103–17.
This page intentionally left blank
41
Malignant mesothelioma
BRUCE WS ROBINSON
Metastatic deposits of mesothelioma are common at tinguish mesothelioma from adenocarcinoma and is occa-
post mortem but are not usually detected clinically.22 Hilar, sionally necessary to distinguish desmoplas-
mediastinal, internal mammary and supraclavicular lymph tic/sarcomatoid mesothelioma from fibrous pleuritis.
node spread is common. Metastases to major organs and
occasionally miliary spread have been reported.23
Imaging
HISTOPATHOLOGY
Figure 41.3 Positron emission tomography (PET) in DNA microarray and mesothelioma diagnosis
mesothelioma. Mesothelioma patient showing multiple
metastatic deposits (black spots) with only small amounts of There have been several studies of microarray findings in
tumor in the involved lung (right side). (Image courtesy of Dr A mesothelioma,40,41 with one study comparing expression
van der Schaff). levels of a small number of genes reportedly able to accu-
1.50
Non-exposed
Asb-exposed
Infection
Inflammatory
Transudates
Malignancy
Plaques
Lung cancer
Breast cancer
Other cancer
Asbestosis
Sarcoidosis
SLE
RA Lung
Other ILD
TB
Gross
thickening
IPF
rately discriminate between mesothelioma and lung tion. They decrease tumor burden, improve symptoms
cancer. The method depends upon quantitative poly- and prolong survival. Pemetrexed (or raltitrexed) plus cis-
merase chain reaction (PCR) to measure gene expression platin, a combination of a multi-targeted anti-folate and a
ratios of paired markers.42 In another study, a panel of 15 platinum compound, improved overall survival by nearly
genes was able to distinguish these two tumors, partly 3 months with an objective response rate of 41 percent.57
because the authors subtracted genes from reactive Gemcitabine plus cisplatin has objective response rates of
mesothelial cells in their analysis.43 48 and 33 percent in a total of 74 patients, with sympto-
matic improvement and quality of life benefits.54
Although PDGF and epidermal growth factor (EGF)
Staging and prognosis signaling paths may always be involved in mesothelioma,
imatinib mesylate (Gleevec) and gefitinib (Iressa), which
Staging has implications for surgical management if the block each of these pathways, have demonstrated no con-
latter is being considered. The International Mesothelioma vincing evidence of response.55
Interest Group (IMIG) staging system is the most widely
used.44 Although accurate preoperative assessment
requires CT, PET and often thoracoscopy and medi- Radiotherapy
astinoscopy, final staging only occurs at surgery.44
Mesothelioma remains a universally fatal disease of
Radiotherapy results have been disappointing except for
increasing incidence worldwide. Median survival from the
local post-surgical radiotherapy to prevent seeding of
time of presentation is 9–12 months.45–47 Prognostication
tumor cells.56 The diffuse nature of the tumor makes it dif-
for mesothelioma is currently dependent on relatively
ficult to undertake radical radiotherapy without causing
crude criteria, such as tumor extension and differentiation,
pneumonitis. Intensity modulated radiotherapy (IMRT) is
rather than on genetic information.48 Poor prognosis is
being used in some centers with benefit.57 Local radio-
associated with a poor performance status, a high white
therapy, e.g. radioactive colloids and other forms of
blood cell (WBC) count, a probable/possible histological
brachytherapy, have been disappointing.
diagnosis of mesothelioma, male gender and having the
sarcomatoid histological subtype.49 A better prognosis is
essentially the opposite and the absence of pain in the pre-
vious 6 months.50 Immunotherapy
A small microarray study in mesothelioma identified
four genes which were predictive of outcome.30 Mesothelioma appears to be sensitive to destruction by
immunotherapies though none have entered routine clin-
ical use.58 Anti-mesothelioma immune responses can be
MANAGEMENT detected in many patients.59 Recombinant interferon alpha
given systemically produced response rates of around
More effective chemotherapies and new approaches to 10–15 percent, and although that approach is complicated
surgery have occurred in recent years. by lethargy, weight loss and fevers, some patients exhibit
long-lasting responses.60
Surgery
Gene therapy
Diagnostic surgery includes video thoracoscopy, open
pleural biopsy and/or mediastinoscopy or laparoscopy to
Gene therapy using either ‘suicide gene’ therapy or
establish a diagnosis of mesothelioma.51 Palliative surgery
‘immunogene’ therapy have been used clinically in
includes partial pleurectomy with pleurodesis, thora-
mesothelioma. These pioneering approaches are reviewed
coscopy with pleurodesis and pleuroperitoneal shunting.52
elsewhere (see Chapter 49, Gene therapy in pleural
‘Potentially curative’ surgery, which is only used in a
disease).
limited number of centres, involves extrapleural pneu-
monectomy usually followed by some form of adjuvant
therapy. The difficulty of the surgery and peri-operative
mortality rate combined to make the role of radical Other therapies
surgery controversial.53–56
Photodynamic therapy, involving the generation of toxic
oxygen radicals in mesothelioma cavities, is laborious and
Chemotherapy time-consuming and is an effective form of cytoreduc-
tion.61 Clinical trials using other novel agents (e.g. the anti-
Cures are rare with chemotherapy regimes but recently angiogenic agents SU5416, bevacizumab, thalidomide and
several regimes have been shown to be of value for pallia- PTK/ZK 787) have been or currently are being tried.62,63
512 Malignant mesothelioma
Inhibitors of histone deacetylase superoylanilide and 3. Greenberg AK, Lee TC, Rom WN. The North American experience
hydroxamic acid (SAHA), a proteosome inhibitor (PS- with malignant mesothelioma. In: Robinson BWS, Chahinian AP
(eds). Mesothelioma. London: Martin Dunitz, 2002; 1–27.
341), another histone deacetyalase inhibitor (PXD101), ●4. Wagner JC, Sleggs CA, Marchand P. Diffuse pleural mesothelioma.
and another vascular endothelial growth factor (VEGF) Br J Ind Med 1960; 17: 260–71.
antagonist (AZD2171) are also being studied.64,65 ●5. Baris YL, Simonato L, Artvinli M, et al. Epidemiological and
Immunotoxin trials using monoclonal antibodies labeled environmental evidence of the health effects of exposure to
with toxins are also underway.66 erionite fibers. Int J Cancer 1987; 39: 10–17.
6. Comin CE, de Klerk NH, Henderson DW. Malignant mesothelioma.
Ultrastruct Pathol 1997; 21: 315–20
7. de Klerk NH, Musk AW. Epidemiology of mesothelioma. In:
PALLIATION Robinson BWS, Chahinian AP (eds). Mesothelioma. London: Martin
Dunitz, 2002: 339–50.
Recurrent pleural effusions are best controlled by pleu- 8. de Klerk NH, Musk AW, Williams VM, et al. Comparison of
measures of exposure to asbestos in former crocidolite workers
rodesis, e.g. talc instillation or occasionally surgery. from Wittenoom Gorge, W. Australia. Am J Ind Med 1996; 30:
Invasion of the chest wall can cause localized somatic pain, 579–87.
intercostal nerve invasion or vertebral invasion can cause 9. Sebastien P, Janson X, Gaudichet A, Hirsch A, Bignon J. Asbestos
neuropathic pain and lung invasion can cause diffuse vis- retention in human respiratory tissues: comparative measurements
ceral pain.67 There is no limit of dose for opioids to control in lung parenchyma and in parietal pleura. IARC Sci Publ 1980; 30:
237–46.
this pain, e.g. use liquid morphine plus sustained release 10. Ault, JG, Cole RW, Jensen CG, et al. Behavior of crocidolite
morphine in doses that do not cause unnecessary side asbestos during mitosis in living vertebrate lung epithelial cells.
effects. Somatic pain responds to non-steroidal anti- Cancer Res 1995; 55: 792–8.
inflammatory drugs and neuropathic pain requires an 11. Shatos, MA, Doherty JM, Marsh JP, Mossman BT. Prevention of
anti-convulsant such as carbamazepine or sodium val- asbestos-induced cell death in rat lung fibroblasts and alveolar
macrophages by scavengers of active oxygen species. Environ Res
proate.68 Some patients require procedural pain relief, e.g. 1987; 44: 103–16.
intrathecal analgesia or nerve block.67 Weight loss and 12. Zanella CL, Posada J, Tritton TR, Mossman, BT. Asbestos causes
anorexia respond to dietary advice and to drug therapy, stimulation of the extracellular signal-regulated kinase 1
e.g. dexamethasone. Dyspnea due to pleural effusion, or mitogen-activated protein kinase cascade after phosphorylation of
more commonly tumor spread, is also common. the epidermal growth factor receptor. Cancer Res 1996; 56:
5334–8.
Psychosocial care is also important in palliation given 13. Marzo AL, Fitzpatrick DR, Robinson BWS, Scott B. Antisense
the fear, anger and suffering associated with this disease. oligonucleotides specific for transforming growth factor beta 2
inhibit the growth of malignant mesothelioma both in vitro and in
vivo. Cancer Res 1997; 57: 3200–7.
14. Gazdar AF, Carbone M. Molecular pathogenesis of mesothelioma
KEY POINTS and its relationship to Simian Virus 40. Clin Lung Cancer 2003; 5:
177–81.
15. Shivapurkar N, Wiethege T, Wistuba II, et al. Presence of simian
● Asbestos is almost always the causative agent. virus 40 sequences in malignant mesotheliomas and mesothelial
● A role for SV40 remains uncertain. cell proliferations. J Cell Biochem 1999; 76: 181–8.
● Most patients present with pleural effusion plus 16. Murthy SS, Testa JR. Asbestos, chromosomal deletions, and tumor
chest wall pain. suppressor gene alterations in human malignant mesothelioma. J
Cell Physiol 1999; 180: 150–7.
● New biomarkers such as SMRP are useful for
17. Pylkkanen L, Sainio M, Ollikainen T, et al. Concurrent LOH at
diagnosis. multiple loci in human malignant mesothelioma with
● Pemetrexed, or gemcitabine, plus cisplatin are preferential loss of NF2 gene region. Oncol Rep 2002; 9:
useful chemotherapy regimes. 955–9.
● Positron-emission tomography is essential for 18. Scott B, Mukherjee S, Lake R, Robinson BWS. Malignant
mesothelioma. In: Hanson H (ed.). Textbook of lung cancer.
preoperative staging if resection is planned.
London: Martin Dunitz, 2000; 273–93.
19. Berry G, de Klerk NH, Reid A, et al. Malignant pleural and
peritoneal mesotheliomas in former miners and millers of
crocidolite at Wittenoom, Western Australia. Occup Environ Med
2004; 61: 14.
20. Hirano H, Maeda T, Tsuji M, et al. Malignant mesothelioma of the
REFERENCES pericardium: case reports and immunohistochemical studies
including Ki-67 expression. Pathol Int 2002; 52: 669–76.
21. Abe K, Kato N, Miki K, et al. Malignant mesothelioma of testicular
● = Key primary paper tunica vaginalis. Int J Urol 2002; 9: 602–3.
◆ = Major review article 22. Lumb PD, Suvarna SK. Metastasis in pleural mesothelioma.
Immunohistochemical markers for disseminated disease.
◆1. Robinson BWS, Lake RL. Advances in malignant mesothelioma. Histopathology 2004; 44: 345–52.
New Eng J Med 2005; 353: 1591–603. 23. Musk AW. More cases of miliary mesothelioma. Chest 1995; 108:
2. Cancerstats Incidence. UK. Mesothelioma. 587.
www.cancerresearchuk.org/statistics 24. Segal A, Whitaker D, Henderson D, Shilkin K. Pathology of
References 513
mesothelioma. In: Robinson BWS, Chahinian AP (eds). 45. Leigh J, Robinson BWS. The history of mesothelioma in Australia
Mesothelioma. London: Martin Dunitz, 2002: 143–184. 1045–2001. In: Robinson BWS, Chahinian AP (eds). Mesothelioma.
◆25. Whitaker D, Shilkin KB. Diagnosis of pleural malignant London: Martin Dunitz, 2002: 55–110.
mesothelioma in life – a practical approach. J Pathol 1984; 143: 46. de Klerk NH, Olsen N, Threlfall T, et al. Mesothelioma survival in
147–75. Western Australia. Perth: 1st Perth Mesothelioma Centre
26. Sterrett GF, Whitaker D, Shilkin KB, Walters MN. Fine needle Symposium 2004.
aspiration cytology of malignant mesothelioma. Acta Cytol 1987; 47. Armstrong BK, Musk AW, Baker JE, et al. Epidemiology of
31: 185–93. mesothelioma in Western Australia. Med J Aust 1984; 141:
●27. Wolanski KD, Whitaker D, Shilkin KB, Henderson DW. The use of 86–8.
epithelial membrane antigen and silver-stained nucleolar 48. di Muzio M, Spoletini L, Strizzi L, et al. Prognostic significance of
organizer regions testing in the differential diagnosis of presence and reduplication of basal lamina in malignant pleural
mesothelioma from benign reactive mesothelioses. Cancer 1998; mesothelioma. Hum Pathol 2000; 31: 1341–5.
82: 583–90. 49. Curran D, Sahmoud T, Therasse P, et al. Prognostic factors in
28. Nguyen GK, Akin MR, Villanueva RR, Slatnik J. Cytopathology of patients with pleural mesothelioma: the European Organization
malignant mesothelioma of the pleura in fine-needle aspiration for Research and Treatment of Cancer experience. J Clin Oncol
biopsy. Diagn Cytopathol 1999; 21: 253–9. 1998; 16: 145–52.
29. Segal A, Whitaker D, Henderson D, Shilkin K. Pathology and 50. Boutin C, Schlesser M, Frenay C, Astoul P. Malignant pleural
mesothelioma. In: Robinson BWS, Chahinian AP (eds). mesothelioma. Eur Respir J 1998; 12: 972–81.
Mesothelioma. London: Martin Dunitz, 2002: 143–84. 51. Waller DA.. Malignant mesothelioma – British surgical strategies.
30. Whitaker D, Henderson DW, Shilkin KB. The concept of Lung Cancer 2004; 45: S81–4.
mesothelioma in situ: implications for diagnosis and histogenesis. 52. Sugarbaker DJ, Jaklitsch MT, Bueno R, et al. Prevention, early
Semin Diagn Pathol 1992; 9: 151–61. detection, and management of complications after 328
31. Rabinowitz JG. Imaging in mesothelioma. In: Robinson BWS, consecutive extrapleural pneumonectomies. J Thorac Cardiovasc
Chahinian AP (eds). Mesothelioma. London: Martin Dunitz, 2002: Surg 2004; 128: 138–46.
201–8. 53. Stewart DJ, Martin-Ucar A, Pilling JE, et al. The effect of extent of
◆32. Wang ZJ, Reddy GP, Gotway MB, et al. Malignant pleural local resection on patterns of disease progression in malignant
mesothelioma: evaluation with CT, MR imaging, and PET. pleural mesothelioma. Ann Thorac Surg 2004; 78: 245–52.
Radiographics 2004; 24: 105–19. 54. Rusch VW. Indications for pneumonectomy. Extrapleural
33. Kebapci M, Vardareli E, Adapinar B, Acikalin M. CT findings and pneumonectomy. Chest Surg Clin N Am 1999; 9: 327–38.
serum CA125 levels in malignant peritoneal mesothelioma: report 55. Pass HI, Kranda K, Temeck BK, Feuerstein I, Steinberg SM.
of 11 new cases and review of the literature. Eur Radiol 2003; 13: Surgically debulked malignant pleural mesothelioma: results and
2620–6. prognostic factors. Ann Surg Oncol 1997; 4: 215–22.
◆34. Marom EM, Erasmus JJ, Pass HI, Patz EF Jr. The role of imaging in 56. Treasure T, Sedrakyan A. Pleural mesothelioma: little evidence, still
malignant pleural mesothelioma. Semin Oncol 2002; 29: 26–35. time to do trials. Lancet 2004; 364: 1183–5.
35. Weber MA, Bock M, Plathow C, et al. Asbestos-related pleural 57. Kindler HL. Malignant pleural mesothelioma. Curr Treat Options
disease: value of dedicated magnetic resonance imaging Oncol 2000; 1: 313–26.
techniques. Invest Radiol 2004; 39: 554–64. ●58. Vogelzang NJ, Rusthoven JJ, Symanowski J, et al. Phase III study of
36. Flores RM, Akhurst T, Gonen M, Larson SM, Rusch VW. Positron pemetrexed in combination with cisplatin versus cisplatin alone in
emission tomography defines metastatic disease but not patients with malignant pleural mesothelioma. J Clin Oncol 2003;
locoregional disease in patients with malignant pleural 21: 2636–44.
mesothelioma. J Thorac Cardiovasc Surg 2003; 126: 11–16. ●59. Nowak AK, Byrne MJ, Williamson R, et al. A multicentre phase II
●37. Robinson BWS, Creaney J, Lake RA, et al. Mesothelin-family study of cisplatin and gemcitabine in malignant mesothelioma. Br
proteins and diagnosis of mesothelioma. Lancet 2003; 362: J Cancer 2002; 87: 491–6.
1612–16. 60. Nowak A, Lake RA, Kindler HL, Robinson BWS. New approaches for
●38. Creaney J, Yeoman D, Naumoff L, et al. Soluble mesothelin in mesothelioma: biologics, vaccines, gene therapy, and other novel
effusions – a useful tool for the diagnosis of malignant agents. Semin Oncol 2002; 29: 82–96.
mesothelioma. Thorax 2007; 62: 569–76. 61. Baldini EH. External beam radiation therapy for the treatment of
39. Pass HI, Lott D, Lonardo F, et al. Asbestos exposure, pleural pleural mesothelioma. Thorac Surg Clin 2004; 14: 543.
mesothelioma, and serum osteopontin levels. N Engl J Med 2005; 62. Ahamad A, Stevens CW, Smythe WR, et al. Intensity-modulated
353: 1564–73. radiation therapy: a novel approach to the management of
40. Ramos-Nino ME, Scapoli L, Martinelli M, Land S, Mossman BT. malignant pleural mesothelioma. Int J Radiat Oncol Biol Phys
Microarray analysis and RNA silencing link fra-1 to cd44 and 2003; 55: 768–75.
c-met expression in mesothelioma. Cancer Res 2003; 63: 63. Mukherjee S, Robinson BWS. Immunotherapy of malignant
3539–45. mesothelioma. In: Robinson BWS, Chahinian AP (eds).
41. Singhal S, Wiewrodt R, Malden LD, et al. Gene expression profiling Mesothelioma. London: Martin Dunitz, 2002: 325–38.
of malignant mesothelioma. Clin Cancer Res 2003; 9: 3080–97. 64. Kindler HL. Moving beyond chemotherapy: novel cytostatic agents
●42. Gordon GJ, Jensen RV, Hsiao LL, et al. Translation of microarray for malignant mesothelioma. Lung Cancer 2004; 45: S125–7.
data into clinically relevant cancer diagnostic tests using gene 65. Kindler HL. Malignant pleural mesothelioma. Curr Treat Options
expression ratios in lung cancer and mesothelioma. Cancer Res Oncol 2000; 1: 313–26.
2002; 62: 4963–7. 66. Hassan R, Bera T, Pastan I. Mesothelin: a new target for
●43. Holloway AJ, Diyagama DS, Opeskin K, et al. A molecular immunotherapy. Clin Cancer Res 2004; 10: 3937–42.
diagnostic test for distinguishing lung adenocarcinoma from 67. Lee, YC, Thompson RI, Dean A, Robinson BWS. Clinical and
malignant mesothelioma using cells collected from pleural palliative care aspects of malignant mesothelioma. In: Robinson
effusions. Clin Cancer Res 2006; 12: 5129–35. BWS, Chahinian AP (eds). Mesothelioma. London: Martin Dunitz,
44. Rusch VW. A proposed new international TNM staging system for 2002: 111–26.
malignant pleural mesothelioma. From the International 68. Finnerup NB, Gottrup H, Jensen TS. Anticonvulsants in central
Mesothelioma Interest Group. Chest 1995; 108: 1122–8. pain. Expert Opin Pharmacother 2002; 3: 1411–20.
This page intentionally left blank
42
Spontaneous pneumothorax
ANDREW C MILLER
DEFINITIONS Incidence
‘Spontaneous’ pneumothorax by definition occurs unex- The incidence of pneumothorax in those presenting as an
pectedly. It is traditional to subdivide it into ‘secondary’ emergency to hospital has been calculated to be 8–12 per
and ‘primary’ depending upon whether or not the patient 100 000 per year.4–7 This was confirmed in a much large
has significant underlying lung disease (Table 42.1). study for a 5-year period,8 with an additional 17 per
Semantically, this is not an accurate distinction because 100 000 annually presenting first to their family doctor –
the great majority of primary cases do have identifiable an overall incidence of 40.7 (men) and 15.6 (women) per
underlying pathology (see section Primary pneumo- 100 000 per year. The latter report is of particular impor-
thorax). tance because most published series underestimate the
516 Spontaneous pneumothorax
Primary Secondary
true incidence by being skewed towards those either accurate than the male preponderance of 80–95 percent
referred to chest clinics from family doctors, or who self- in other studies.4,10–14 Catamenial cases account for about
refer to hospital but may not be admitted, or who are 3 percent of those in young women, in whom smoking
transferred to specialist thoracic units. and body shape are less relevant than in men.15 There are
few reports of pneumothorax in pregnancy,16 half of
these women having other attacks outside pregnancy;
Gender half of the episodes occurred during or shortly after
labor. Another cause exclusive to females is lymphangi-
The above male:female ratio of 2.5:1, similar to that in oleiomyomatosis.17–20
large series from tertiary centers,9 is likely to be more
Pathogenesis 517
Again, pulmonary metastatic disease only causes pneumo- Acquired immune deficiency syndrome
thorax when the visceral pleura is involved, as has been
documented in a wide variety of tumors arising from the Spontaneous pneumothorax is a common – and ominous
breast, bowel, uterus as well as sarcomas and germ-call – complication of AIDS, several specialist units reporting
tumors; it may be induced by tissue death due to more than 25 such cases.131–137 Major associations are
Clinical features 519
Pain
Dyspnea
CLINICAL FEATURES
Allowing for the severity of chest pain, the degree of
Occasionally, spontaneous pneumothorax, even bilateral, breathlessness in primary pneumothorax is variable; it is
may be detected in an asymptomatic patient;108,109,146 in a frequently minimal,22,26–28 and otherwise is often only mild
survey of 994 Air Force personnel 7 percent were only or moderate even in complete collapse, so that many are
detected on routine chest radiography.58 However, most not tachypneic.115 On the other hand, dyspnea (worse than
patients present because of chest pain and/or dyspnea. usual) is a prominent symptom in secondary pneumo-
Pain is almost always the main presenting symptom in thorax, where even a small degree of collapse can be life-
primary pneumothorax;22,25–28,115 in secondary cases, threatening; it is often assumed that such patients are
which includes most elderly patients with pneumothorax, suffering an exacerbation of their known lung disease, and
the dominant symptom is dyspnea.111,118,147,148 In primary the pneumothorax is often hard to detect clinically, so mis-
cases symptoms usually come on suddenly and subside diagnosis here is common.111,115,159
520 Spontaneous pneumothorax
Other the error rate is halved when paired expiratory films are
available.158 The variable quality of emergency room radio-
Of the classical clinical signs, the most common are graphs can make a small pneumothorax easy to overlook,
reduced ipsilateral chest expansion21 and breath sounds,25 but a closer scrutiny of the film may be prompted by iden-
hyper-resonance being found mainly in moderate or com- tifying a basal pleural reaction, a change found in two-
plete collapse. Chest examination is unhelpful in some thirds of patients with primary pneumothorax, often with
primary and most secondary118 cases. An occasional aus- an air–fluid level.38,157 With careful inspection small blebs
cultatory finding with shallow left-sided collapse is a click- or larger bullae may be apparent.80,157 The latter can be
ing, crunching or crackling sound at the cardiac apex in mistaken for pneumothorax by the unwary, with poten-
time with the pulse,160,161 distinct from those described in tially adverse consequences if drainage is attempted; a sug-
mediastinal emphysema.162 Some with ‘noisy pneumo- gestive sign is air surrounding both walls of the bulla,170
thorax’ are themselves aware of these sounds.160 Although and CT is useful where uncertainty remains.171 Lateral
not widely known, commonly patients with resolving radiographs occasionally demonstrate a small pneumo-
pneumothorax (whether right- or left-sided) describe or thorax not detected on a standard film, but have a high
admit to bubbling sounds in that lung, particularly on false-negative rate172 and are not recommended.
lying down and varying with posture, disappearing when In secondary pneumothorax, chest radiography usually
the lung fully re-inflates. Some even have electrocardio- also reflects the underlying pathology (Figures 42.1 and
graphic changes leading to a mistaken diagnosis of peri- 42.2); exceptions include early cases of lymphangioleiomy-
carditis, myocardial infarction or pulmonary embolism. omatosis and pleural-based malignancy.
Although subcutaneous emphysema and pneumo-
mediastinum may occur in traumatic pneumothorax and
with barotrauma, they are rare in spontaneous pneumo-
Size of pneumothorax
thorax at presentation although they commonly compli-
cate drainage procedures;13 in experimental or artificial Although attempts have been made to calculate the per-
pneumothorax air cannot be forced into the mediastinum. centage of lung collapse from standard radiographs,173–177
Where they are associated, the pneumothorax is small and only the simplest (the ‘Light index’176) correlates well with
complicates the pneumomediastinum rather than the the volume of air that can be aspirated.178 The latter is pre-
other way round.69,162,163 The pathogenesis of mediastinal dicted best by CT, probably because the lung collapses
emphysema differs in that rupture occurs in alveoli con- asymmetrically; in any case the authors179 also concluded
tiguous with connective tissue in the pulmonary vascula- ‘perhaps more emphasis should be placed on the clinical
ture and then spreads towards the mediastinum,69 which status of the patient when making a decision of treatment
requires much less pressure to rupture than visceral strategy’. This latter premise is reflected in the 1993 British
pleura. Spontaneous pneumomediastinum is a benign Thoracic Society (BTS) guidelines, which instead used
condition which can nearly always be diagnosed on clini- three defined categories (‘small’, ‘moderate’ and ‘large’).
cal grounds and good-quality chest radiography; manage- However, in line with others, 157,180 the BTS now prefers
ment is conservative164 and a good clinician will not two groupings only, using a plain film rim of air of 2 cm
request CT scanning165 or tests to exclude esophageal per- (equivalent to 50 percent collapse) as cut-off.166 Some
foration. studies refer to patients with ‘complete’ or ‘total’ pneu-
mothorax even though semantically these terms should be
reserved for those with an airless lung. In any case, initial
management depends not only upon the size of pneu-
CHEST RADIOGRAPHY mothorax but also its cause and degree of associated
breathlessness.
Technical considerations
Since expiratory films exaggerate the radiographic changes Tension
of pneumothorax, standard practice was to request paired
films in suspected cases, a practice that is now discour- The BTS also redefined tension as ‘any pneumothorax
aged,166 and in reality only inspiratory films are routinely with cardio-respiratory collapse’, because of the wide-
requested for patients presenting with chest pain or spread misunderstanding that the presence of mediastinal
dyspnea. Apparent justification for inspiratory-only films shift plus depression and flattening of the diaphragm on
comes from two studies in which the only errors were in radiography represents a medical emergency. Such
two (out of 138) patients with minimal collapse,167,168 but changes have been recorded in 19 percent of all cases of
these films were reported by experienced radiologists, pneumothorax;181 they are very common in large pneu-
whereas in most emergency rooms decisions are based on mothorax,38,53,157 and many such patients are not unduly
those viewed by junior doctors. The latter are less accurate distressed (Figure 42.3).181,182 In contrast with true tension
in recognizing pneumothorax than their seniors,169 and (discussed in Chapter 43, Non-spontaneous pneumotho-
Medical management 521
(b)
(a)
Figure 42.3 This is not a tension pneumothorax. A 51-year-old man referred with a week of mild dyspnea. Radiograph (a) showed a
large left pneumothorax with displacement of heart and hemidiaphragm, fully resolving after simple aspiration in clinic (b). He was not
admitted and has had no recurrence.
rax), fluoroscopic screening shows that these shifts disap- place of outpatient observation, initial drainage method,
pear during deep inspiration.183 size of intercostal tube, value of suction, clamping chest
Cardiovascular collapse is rare in primary pneumo- drains and indications for surgery.187–189 The American
thorax;180,184 yet even recently it has been stated that College of Chest Physicians developed a consensus docu-
patients with radiographic tension ‘are at risk of sudden ment on these issues from recognized international
deterioration and possibly cardiac arrest… potentially life- authorities,190 but on several of these issues there was a
threatening situation’. However, only one of the four wide range of views, and its effect on standardizing
patients described was unstable,185 and this was due to American practice is unclear. In that document simple
hemopneumothorax, as in one of two previously reported aspiration, although recommended by some of its contrib-
who were severely ill.21 Kjærgaard did describe eight such utors,6,191 was nevertheless strongly discouraged, a view
patients, but these were collected from a large area over 20 that is no longer tenable.
years and only three were sufficiently ill to require urgent
drainage.3 There seems no reason to dissent from the view
of Nissen180 that untreated pneumothorax in otherwise Historical
healthy people is ‘very seldom fatal’. Conversely, a patient
with secondary pneumothorax who rapidly deteriorates In the early part of the last century most cases, even with
even when radiological collapse is small should be decom- major lung collapse, were treated with bed rest because it
pressed by immediate cannulation. was assumed that they had underlying tuberculosis. As
spontaneous pneumothorax was increasingly recognized,
some physicians familiar with inducing artificial pneu-
MEDICAL MANAGEMENT mothorax in tuberculosis (often in outpatients who con-
tinued at work) used the same apparatus for aspirating
Only one national guideline has been introduced, by the pneumothorax,49,192 which others avoided because of fear
BTS in 1993, which 10 years later was updated, modified that the metal needles available might puncture the
and expanded.6,166 The BTS approach is now included in lung.3,86,115,193 Increasing thoracic surgical experience with
all UK textbooks of general, respiratory and emergency rubber tube drains (originally used mainly in post-
medicine, and has been introduced elsewhere.7,186 operative cases) led to their adoption by other specialists
Otherwise, there is widespread variation in practice for pneumothorax, and references to metal needles were
with little agreement on fundamental issues such as the withdrawn in later editions of medical textbooks.192,194 The
522 Spontaneous pneumothorax
Observation
years in Britain with no recorded fatalities. Where death
It is widely accepted that a primary pneumothorax of less has been recorded in primary pneumothorax, either it
than 15 percent does not require intervention, but the BTS occurred during inpatient observation,214 was a complica-
documents agree with earlier clinical practice that observa- tion of treatment,215 diagnosis had been overlooked104 or
tion is appropriate in greater degrees of collapse provided was due to concomitant bleeding.3 Overnight observation
that the patient is only mildly breathless. In the UK this is indicated if the patient lives remote from the hospital; all
can be carried out as an outpatient procedure because agree that this is mandatory in secondary pneumothorax,
much of the population live relatively near their district although many of these can be discharged if stable.9
general hospital. The great majority of patients with minor
lung collapse have a closed pneumothorax. It has been
known since the nineteenth century that intrapleural air is Drainage
rapidly reabsorbed; early animal experiments showed this
to be accelerated by breathing an oxygen-rich If the pneumothorax is closed then any method of drain-
mixture.205,206 This has been confirmed by more sophisti- ing the air will be successful. Although it has long been
cated recent research in rabbits207,208 as well as in human assumed, and taught, that the presence of a large-bore
studies.209,210 An early study calculated that air reabsorbs at intercostal drain for a few days promotes pleural adher-
a rate of approximately 1.25 percent of the lung volume ence, thus allegedly reducing the risk of recurrence,86 a
per day,211 but a more accurate figure is 2.2 percent (95 randomized controlled trial showed that in this respect it
percent confidence interval 1.4–3.0 percent).212 The sug- was no better than simple aspiration.204 This British study
gestion that patients with primary pneumothorax may not has since been confirmed by two larger trials from
need hospitalization has been criticized as being poten- Belgium216 and Kuwait,217 as well as in a French compari-
tially dangerous,213 but this has been widely practiced for son of indwelling catheter with large-bore drains.218
Medical management 523
Compiling available information from all studies on but then after a few minutes further air can again be aspi-
primary pneumothorax, the immediate success rate for rated there is a small air leak;232 and (3) if after coughing,
simple aspiration is 73 percent (440/606 – 16 no more air can be aspirated after several minutes, it is a
studies217,219,220) and for indwelling catheter 80 percent closed pneumothorax and after confirmatory chest radi-
(441/560 – 12 studies53,219,221,222). The advantage of the ograph the patient can be discharged or the cannula can be
latter is that it allows time for a continuing small air leak to secured in place for a few hours.233 The accepted approach
heal, which is why its success rates approach the 83 if aspiration is unsuccessful is insertion of an indwelling
percent (938/1134 – 11 studies217,219,221,222) for large-bore catheter; whether this is always necessary is discussed later
drains. The pros and cons of these three procedures are (see section Persistent air leak). Unlike catheters, plastic
outlined in Table 42.3. The high success rate of simple cannulae (which are very cheap, readily available and
aspiration suggests that the initial pleural tear has already extremely easy for inexperienced doctors to use) are easily
sealed by the time of diagnosis, and such rapid healing is dislodged from the pleural cavity if left in situ for the effect
even seen in traumatic pneumothorax where most chest of aspiration to be assessed. Also, the cannula must be at
drains can be removed within a day.223 It is often assumed least 5 cm long to be sure of penetrating the pleura234 and
that complete lung collapse suggests an open pneumotho- is perhaps preferably inserted using an axillary rather than
rax, but in one study where three-quarters of patients had an anterior approach.235
complete collapse, simple aspiration was highly effec- There is less information on simple aspiration in sec-
tive.216 With the exception of hemopneumothorax, there ondary pneumothorax, where it is successful in 47 percent
is now little justification for using large-bore tubes for (32/68 – 4 studies219); the latest of these was the largest, in
initial drainage of either primary or secondary pneumoth- which an indwelling drain was only required in 19 of 35.236
orax; one French group switched over entirely to As well as being effective in many, in those who fail the size
indwelling catheters in 1977.224 Harvey, co-author of the of the air leak can be inferred, as above. Although standard
BTS guidelines, has calculated that adoption of simple practice has been to use a large-bore drain (at least 16 FG),
aspiration throughout Europe and the USA would avoid usually an indwelling catheter is effective,222 although less
75 000 intercostal drains per annum in patients with advisable in those who clearly have a large air leak and who
primary pneumothorax. are likely to require prolonged drainage. BTS guidelines no
There are three reasons why simple aspiration should longer advocate large-bore bore tube drains, and catheter
be considered as the first-line intervention in such drainage instead is recommended.166
patients. First, it requires minimal training, an important
consideration when many are initially seen by non-
specialist junior doctors, who often have never inserted a Management of chest drains
chest tube,225 have inadequate technique,226 use an inap-
propriate site227,228 and may cause major complica- Since experience has shown that small-bore catheters may
tions.229,230 Second, hospital admission is often become displaced, blocked, kinked or are poorly posi-
unnecessary; even in patients with complete collapse it is tioned,7,200,224,237 it is advisable to choose one 12–14 FG size
successful in over 60 percent,203,216,231 and yet many are and relatively stiff. Those introduced by a Seldinger tech-
reluctant to attempt aspiration in such patients.25,189 nique are marketed by several manufacturers and are
Third, it can usually predict both the existence and size of becoming increasingly popular; they have proved success-
an air-leak: (1) if air continues to be easily aspirated a large ful in various types of pleural drainage with few complica-
air leak is present; (2) should the lung appear to re-inflate tions.238
Advantages Disadvantages
The traditional method of attaching the drain to an this is that deferring such a decision by clamping is unnec-
underwater seal allows bubbling of air to be easily recog- essary, a practice which most British pulmonologists aban-
nized. However, using a flutter valve instead has the major doned following advice from thoracic surgeons and the
advantage of allowing the option of outpatient manage- 1993 BTS guidelines. If air leakage has ceased, clamping
ment199,239 and there is some evidence that it is more effec- confers no advantage, but if not a dangerous situation may
tive;240 however, care is needed to ensure it is connected develop unrecognized. Such a consensus has not been
the correct way.241 Newer devices have been developed to reached elsewhere.190 It makes no difference whether the
simplify outpatient drainage;242–245 although effective, drain is removed in expiration or inspiration.255
these are expensive and are unlikely to replace catheter In the absence of a persistent air leak, failure of the lung
plus flutter valve. to re-expand is nearly always due to the drain having
The widespread practice of applying suction to a chest becoming blocked, displaced (Figure 42.6), disconnected
drain derives from thoracic surgical practice of draining or clamped; occasionally it results from lung fibrosis or
viscous secretions in post-resection, empyema and trauma bronchial obstruction.256
cases, but its logic in drainage of air is dubious and early
textbooks had polarized views.194,246 Some imagine that air
leak is more likely to stop if the lung can be brought into OPERATIVE MANAGEMENT
apposition against the chest wall, but this is no more con-
vincing than the argument that a collapsed lung is more Urgent surgical intervention is required in those with
likely to heal. The clinical impression that applying suction spontaneous hemo-pneumothorax who are hemodynami-
increases the risk of rare247 but potentially fatal re-expan- cally compromised, although milder cases can be managed
sion pulmonary edema, particularly if the lung has been conservatively.153,257 Non-urgent cases may be treated
collapsed for several days (Figure 42.5),157,248,249 has been either by medical thoracoscopists (Chapter 47) or thoracic
elegantly confirmed in a monkey model.250 Three groups surgeons (Chapter 48), who strongly disagree about which
who compared standard suction with none found no procedure is best. Different practices both between and
advantage either in sealing air leaks or reducing hospital within countries are largely governed by who admits the
stay;237,251,252 they disagreed about whether, once the lung patient: (1) thoracic surgeon, or (2) non-interventional
re-expands with no further air leak, the drain can be pulmonologists with established surgical links, or (3) pul-
removed immediately, and many practitioners still prefer monologist experienced at medical thoracoscopy.
delayed removal.188 Randomized studies following pul- Whichever is preferred, patients can be reassured that this
monary resection253 and video-assisted surgery254 show will have no long-term effect on lung function or physical
that applying suction to chest drains delays tube removal fitness.258,259 However, there is a significant incidence of
and lengthens hospital stay. An important implication of chronic troublesome post-operative pain and paraesthe-
sia,103,260,261 sometimes not mentioned in ongoing debates
about indications for, timing of and technique of operative
intervention.
●10. Stradling P, Poole G. Conservative management of spontaneous 36. Bitar ZI, Ahmed S, Amin AE, Jamal K, Ridha M. Prevalence of mitral
pneumothorax. Thorax 1966; 21: 145–9. valve prolapse in primary spontaneous pneumothorax. Prim Care
11. Jansveld CAF, Dijkman JH. Primary spontaneous pneumothorax Respir J 2006; 15: 342–5.
and smoking. Br Med J 1975; 4: 559–60. 37. Lopes C, Manique A, Sotto-Mayor R, et al. Ehlers-Danlos syndrome
12. Hart GJ, Stokes TC, Couch AH. Spontaneous pneumothorax in – a rare cause of spontaneous pneumothorax [Portuguese]. Rev
Norfolk. Br J Dis Chest 1983; 77: 164–70. Port Pneumol 2006; 12: 471–80.
13. Teixidor Sureda J, Estrada Salo G, Sole Montserrat J, et al. 38. Hallgrimsson JG. Spontaneous pneumothorax in Iceland with
Spontaneous pneumothorax: a review of 2,507 cases [Spanish]. special reference to idiopathic type: a clinical and epidemiological
Arch Bronconeumol 1994; 30: 131–5. investigation. Scand J Thorac Cardiovasc Surg 1978; Suppl 21:
14. Ji Woong Son, Jae Yong Park, Kwan Young Kim, et al. Clinical 1–85.
analysis of spontaneous pneumothorax. Tubercul Respir Dis 1999; 39. Lenler-Petersen P, Grunnet N, Jespersen TW, Jaeger P. Familial
47: 374–82. spontaneous pneumothorax. Eur Respir J 1990; 3: 342–5.
15. Nakamura H, Konishiike J, Sugamura A, Takeno Y. Epidemiology of 40. Nickoladze GD. Surgical management of familial spontaneous
spontaneous pneumothorax in women. Chest 1986; 89: 378–82. pneumothorax. Respir Med 1990; 84: 107–9.
16. Van Winter JT, Nichols IIIFC, Pairolero PC, Ney JA, Ogburn PL Jr. 41. Abolnik IZ, Lossos IS, Zlotogora J, Brauer R. On the inheritance of
Management of spontaneous pneumothorax during pregnancy: primary spontaneous pneumothorax. Am J Med Genet 1991; 40:
case report and review of the literature. Mayo Clin Proc 1996; 71: 155–8.
249–52. 42. Yellin A, Shiner RJ, Lieberman Y. Familial multiple bilateral
17. Cohen MM, Pollock-Barziv S, Johnson S. The emerging clinical pneumothorax associated with Marfan syndrome. Chest 1991;
picture of lymphangioleiomyomatosis. Thorax 2005; 60: 875–9. 100: 577–8.
18. Almoosa KF, Ryu JH, Mendez J, et al. Management of 43. Cheng YJ, Chou SH, Kao EL. Familial spontaneous pneumothorax:
pneumothorax in lymphangioleiomyomatosis: effects on report of seven cases in two families [Chinese]. Kao J Med Sci
recurrence and lung transplantation complications. Chest 2006; 1992; 8: 390–4.
129: 1274–81. 44. Morrison PJ, Lowry RC, Nevin NC. Familial primary spontaneous
19. Ryu JH, Moss J, Beck GJ, et al. The NHLBI pneumothorax consistent with true autosomal dominant
lymphangioleiomyomatosis registry: characteristics of 230 inheritance. Thorax 1998; 53: 151–2.
patients at enrollment. Am J Respir Crit Care Med 2006; 173: 45. Chiu HT, Garcia CK. Familial spontaneous pneumothorax. Curr
105–11. Opin Pulm Med 2006; 12: 268–72.
20. Taveira-DaSilva AM, Steagall WK, Moss J. 46. Toro JR, Pautler SE, Stewart L, et al. Lung cysts, spontaneous
Lymphangioleiomyomatosis. Cancer Control 2006; 13: 276–85. pneumothrorax and genetic associations in 89 families with
21. Myers JA. Simple spontaneous pneumothorax. Dis Chest 1954; 26: Birt–Hogg–Dube syndrome. Am J Respir Crit Care Med 2007; 22:
420–41. 22.
22. Seremetis MG. The management of spontaneous pneumothorax. 47. Elfeldt RJ, Schroder D, Schroeder P, Schaube H, Nissen R. Relation
Chest 1970; 57: 65–8. between smoking and idiopathic spontaneous pneumothorax: a
23. Watt AG. Spontaneous pneumothorax: a review of 210 case–control study. Theor Surg 1993; 8: 131–5.
consecutive admissions to Royal Perth Hospital. Med J Austr 1978; 48. Lindskog GE, Halasz NA. Spontaneous pneumothorax. Arch Surg
1: 186–8. 1957; 75: 693–8.
24. Nakamura H, Izuchi R, Hagiwara T, et al. Physical constitution and 49. Lenox-Smith I. Spontaneous pneumothorax: a study of 94 cases.
smoking habits of patients with idiopathic spontaneous Br J Dis Chest 1962; 56: 1–10.
pneumothorax. Jpn J Med 1983; 22: 2–8. 50. Klassen KP, Meckstroth CV. Treatment of spontaneous
25. Abolnik IZ, Lossos IS, Gillis D, Breuer R. Primary spontaneous pneumothorax: prompt expansion with controlled thoracotomy
pneumothorax in men. Am J Med Sci 1993; 305: 297–303. tube suction. J Am Med Assoc 1962; 182: 111–5.
26. Getz SB, Beasley WE. Spontaneous pneumothorax. Am J Surg 51. Voge VM, Anthracite R. Spontaneous pneumothorax in the USAF
1983; 145: 823–7. aircrew population: a retrospective study. Aviat Space Environ
27. Escribano Montaner A, Herrejon Silvestre A, Simo Mompo M, Med 1986; 57: 939–49.
Blanquer Olivas J. Spontaneous idiopathic pneumothorax in 52. Bense L, Wiman LG, Hedenstierna G. Onset of symptoms in
adolescents [Spanish]. Ana Espan Ped 1995; 42: 261–4. spontaneous pneumothorax: correlations to physical activity. Eur J
28. Wilcox DT, Glick PL, Karamanoukian HL, Allen JE, Azizkhan RG. Respir Dis 1987; 71: 181–6.
Spontaneous pneumothorax: a single-institution, 12-year 53. Marquette CH, Marx A, Leroy S, et al. Simplified stepwise
experience in patients under 16 years of age. J Pediatr Surg 1995; management of primary spontaneous pneumothorax: a pilot study.
30: 1452–4. Eur Respir J 2006; 27: 470–6.
29. Wong KS, Liu HP, Yeow KM. Spontaneous pneumothorax in 54. Bense L. Spontaneous pneumothorax related to falls in
children. Acta Paediatr Taiwan 2000; 41: 263–5. atmospheric pressure. Eur J Respir Dis 1984; 65: 544–6.
30. Vassallo R, Ryu JH, Schroeder DR, Decker PA, Limper AH. Clinical 55. Ozenne G, Poignie P, Lemercier JP, Nouvet G, Grancher G.
outcomes of pulmonary Langerhans’-cell histiocytosis in adults. N Meteorological conditions and spontaneous pneumothorax: a
Engl J Med 2002; 346: 484–90. retrospective study of 165 cases in the Rouen region. Rev Pneumol
31. Melton LJ 3rd, Hepper NG, Offord KP. Influence of height on the Clin 1984; 40: 27–33.
risk of spontaneous pneumothorax. Mayo Clin Proc 1981; 56: 56. Smit HJM, Deville WL, Schramel FMNH, et al. Atmospheric
678–82. pressure changes and outdoor temperature changes in relation to
32. Kawakami Y, Irie T, Kamishima K. Stature, lung height, and spontaneous pneumothorax. Chest 1999; 116: 676–81.
spontaneous pneumothorax. Respiration 1982; 43: 35–40. 57. Sok M, Mikulecky M, Erzen J. Onset of spontaneous pneumothorax
33. Fujino S, Inoue S, Tezuka N, et al. Physical development of and the synodic lunar cycle. Med Hypotheses 2001; 57: 638–41.
surgically treated patients with primary spontaneous 58. Cran IR, Rumball CA. Survey of spontaneous pneumothorax in the
pneumothorax. Chest 1999; 116: 899–902. Royal Air Force. Thorax 1967; 22: 462–5.
●34. West JB. Distribution of mechanical stress in the lung: a possible 59. Morales Suarez-Varela M, Martinez-Selva MI, Llopis-Gonzalez A,
factor in localisation of pulmonary disease. Lancet 1971; i: 839–41. Martinez-Jimeno JL, Plaza-Valia P. Spontaneous pneumothorax
35. Margaliot SZ, Barzilay J, Bar-David M, et al. Spontaneous related with climatic characteristics in the Valencia area. Eur J
pneumothorax and mitral valve prolapse. Chest 1986; 89: 93–4. Epidem 2000; 16: 193–8.
528 Spontaneous pneumothorax
60. Ayed AK, Bazerbashi S, Ben-Nakhi M, et al. Risk factors of pneumothorax, and its treatment, on the smoking behaviour of
spontaneous pneumothorax in Kuwait. Med Princ Pract 2006; 15: young adult smokers. Respir Med 1998; 92: 1132–6.
338–42. 84. Van Belle AF, Lamers RJ, ten Velde GP, Wouters EF. Diagnostic
61. Noppen M, Verbanck S, Harvey J, et al. Music: a new cause of yield of computed tomography and densitometric measurements
primary spontaneous pneumothorax. Thorax 2004; 59: 722–4. of the lung in thoracoscopically-defined idiopathic spontaneous
62. Blanco-Perez J, Bordon J, Pineiro-Amigo L, et al. Pneumothorax in pneumothorax. Respir Med 2001; 95: 292–6.
active pulmonary tuberculosis: resurgence of an old complication? 85. Baranofsky ID, Warden HG, Kaufman JL, Whately J, Hanner JM.
Respir Med 1998; 92: 1269–73. Bilateral therapy for unilateral spontaneous pneumothorax. J
63. Javaid A, Amjad M, Khan W, et al. Pneumothorax: aetiology, Thorac Cardiovasc Surg 1957; 34: 1767–9.
complications and outcome. J Coll Phys Surg Pakistan 1998; 8: 86. Mills M, Baisch BF. Spontaneous pneumothorax: a series of 400
14–16. cases. Ann Thorac Surg 1965; 1: 286–97.
64. Hussain SF, Aziz A, Fatima H. Pneumothorax: a review of 146 adult 87. Radomsky J, Becker HP, Hartel W. Pleural porosity in idiopathic
cases admitted at a university teaching hospital in Pakistan. J spontaneous pneumothorax [German]. Pneumologie 1989; 43:
Pakistan Med Assoc 1999; 49: 243–6. 250–3.
65. Gupta D, Mishra S, Faruqi S, Aggarwal AN. Aetiology and clinical 88. Janssen JP, Schramel FM, Sutedja TG, Cuesta MA, Postmus PE.
profile of spontaneous pneumothorax in adults. Indian J Chest Dis Videothoracoscopic appearance of first and recurrent
Allied Sci 2006; 48: 261–4. pneumothorax. Chest 1995; 108: 330–4.
66. Mezghani S, Abdelghani A, Njima H, et al. Tuberculous 89. Ayed AK, Chandrasekaran C, Sukumar M. Video-assisted
pneumothorax: retrospective study of 23 cases in Tunisia [French]. thoracoscopic surgery for primary spontaneous pneumothorax:
Rev Pneumol Clin 2006; 62: 13–18. clinicopathological correlation. Eur J Cardiothorac Surg 2006; 29:
67. Darwish B. Clinical and radiological manifestations of 206 patients 221–5.
with pulmonary hydatidosis over a ten-year period. Prim Care 90. Lichter I, Gwynne JF. Spontaneous pneumothorax in young
Respir J 2006; 15: 246–51. subjects: a clinical and pathological study. Thorax 1971; 26:
68. West S. The Bradshawe lecture: on pneumothorax. Br Med J 1887; 409–17.
ii: 393–400. 91. Tomashefski JF, Dahms B, Bruce M. Pleura in pneumothorax:
●69. Macklin CC. Pneumothorax with massive collapse from comparison of patients with cystic fibrosis and idiopathic
experimental over-inflation of the lung substance. Can Med J spontaneous pneumothorax. Arch Pathol Lab Med 1985; 109:
1937; 36: 414–20. 910–16.
70. Calder IM. Autopsy and experimental observations on factors 92. Becker HP, Weidringer JW, Willy C, Hartel W, Blumel G. Light and
leading to barotrauma in man. Undersea Biomed Res 1985; 12: scanning electron microscopy studies of the pleura: a contribution
165–81. to the pathogenesis of spontaneous pneumothorax in young
71. Marnejon T, Sarac S, Cropp AJ. Spontaneous pneumothorax in patients [German]. Langenbecks Arch Chir 1991; 376: 295–301.
weightlifters. J Sports Med Phys Fitness 1995; 35: 124–6. 93. Haraguchi S, Fukuda Y. Histogenesis of abnormal elastic fibers in
72. Reeder SR. Subcutaneous emphysema, pneumomediastinum, and blebs and bullae of patients with spontaneous pneumothorax:
pneumothorax in labor and delivery. Am J Obstet Gynecol 1986; ultrastructural and immunohistochemical studies. Acta Pathol Jpn
154: 487–9. 1993; 43: 709–22.
73. Nieboer B, Aboosy N, Verschoor L, Huisman A. 94. Luna E, Tomashefski JF Jr, Brown D, Clarke RE, Kleinerman J.
Pneumomediastinum as a cause of acute chest pain postpartum. J Reactive eosinophilic pulmonary vascular infiltration in patients
Matern Fetal Neonatal Med 2006; 19: 243–5. with spontaneous pneumothorax. Am J Surg Pathol 1994; 18:
74. Heppner HJ, Sieber C, Schmitt K. ‘Usual’ cannabis abuse producing 195–9.
unusual consequences [German]. Deutsch Med Wochenschr 2007; 95. Cyr PV, Vincic L, Kay JM. Pulmonary vasculopathy in idiopathic
132: 560–2. spontaneous pneumothorax in young subjects. Arch Pathol Lab
75. Reddymasu S, Borhan-Manesh F, Jordan PA. Spontaneous Med 2000; 124: 717–20.
pneumomediastinum due to achalasia: a case report. South Med J 96. Smit HJ, Wienk MA, Schreurs AJ, Schramel FM, Postmus PE. Do
2006; 99: 892–3. bullae indicate a predisposition to recurrent pneumothorax? Br J
76. Manco JC, Terra-Filho J, Silva GA. Pneumomediastinum, Radiol 2000; 73: 356–9.
pneumothorax and subcutaneous emphysema following the 97. Noppen M. CT scanning and bilateral surgery for unilateral primary
measurement of maximal expiratory pressure in a normal subject. pneumothorax? Chest 2001; 119: 1293–4.
Chest 1990; 98: 1530–2. 98. Smit HJM, Golding RP, Schramel FMNH, et al. Lung density
77. Dermskian G, Lamb LE. Spontaneous pneumothorax in apparently measurements in spontaneous pneumothorax demonstrate air
healthy flying personnel. Ann Intern Med 1959; 51: 39–51. trapping. Chest 2004; 125: 2083–90.
78. Robb DJ. Cases from the aerospace medicine residents’ teaching 99. Ayed AK, Chandrasekaran C, Sukumar M. Video-assisted
file: Case H57 – complete spontaneous pneumothorax in-flight in thoracoscopic surgery for primary spontaneous pneumothorax:
an F-16 pilot during a high-G maneuver. Aviat Space Environ Med clinicopathological correlation. Eur J Cardiothorac Surg 2006; 29:
1994; 65: 170–2. 221–5.
79. Pedersen MN, Jensen BN. Pneumothorax after ‘reversed bungee 100. Lipscomb VJ, Hardie RJ, Dubielzig RR. Spontaneous pneumothorax
jump’ [Danish]. Ugeskr Laeger 1999; 161: 5547–8. caused by pulmonary blebs and bullae in 12 dogs. J Am Anim Hosp
80. Jordan KG, Kwong JS, Flint J, Muller NL. Surgically treated Assoc 2003; 39: 435–45.
pneumothorax: radiologic and pathologic findings. Chest 1997; 101. De Troyer A, Yernault JC, Rodenstein D, Englert M, de Coster A.
111: 280–5. Pulmonary function in patients with primary spontaneous
81. Lesur O, Delorme N, Fromaget JM, Bernadac P, Polu JM. Computed pneumothorax. Bull Eur Physiopathol Respir 1978; 14: 31–9.
tomography in the etiologic assessment of idiopathic spontaneous 102. Gobbel WG. Spontaneous pneumothorax. J Thorac Cardiovasc Surg
pneumothorax. Chest 1990; 98: 341–7. 1963; 46: 331–45.
82. Mitlehner W, Friedrich M, Dissmann W. Value of computer 103. Sihoe AD, Au SS, Cheung ML, et al. Incidence of chest wall
tomography in the detection of bullae and blebs in patients with paresthesia after video-assisted thoracic surgery for primary
primary spontaneous pneumothorax. Respiration 1992; 59: spontaneous pneumothorax. Eur J Cardiothorac Surg 2004; 25:
221–7. 1054–8.
83. Smit HJ, Chatrou M, Postmus PE. The impact of spontaneous 104. Simonsen K. Sudden death due to bilateral spontaneous
References 529
pneumothorax caused by rupture of congenital lung cysts. Z 129. Akelsson IG, Eklund A, Skold CM, Tornling G. Bilateral spontaneous
Rechtsmed 1990; 103: 379–83. pneumothorax and sarcoidosis. Sarcoidosis 1990; 7: 136–8.
105. Hay E, Sternfeld M, Rashid A, Kunichevsky S, Eliraz A. 130. Fraticelli A, Picat-Joossen D, Astoul P. Pneumothorax and multiple
Simultaneous bilateral spontaneous pneumothorax: case report. lung masses: unusual manifestations of sarcoidosis [French]. Rev
Am J Emerg Med 1992; 10: 50–2. Mal Respir 2000; 17: 885–7.
106. Kendall IG, Harborne DJ, Premswarup I. Beware spontaneous 131. Byrnes TA, Brevig JK, Chin Bor Y. Pneumothorax in patients with
bilateral pneumothorax. Arch Emerg Med 1992; 9: 51–3. acquired immunodeficiency syndrome. J Thorac Cardiovasc Surg
107. Hatta T, Mastuda S, Kuris S, et al. A case of simultaneous bilateral 1989; 98: 546–50.
spontaneous pneumothorax. Jpn J Thorac Surg 1993; 46: 287–9. 132. Pellerin M, Lafontaine E, Toma E, Bard C. Surgical treatment of
108. Kadokura M, Yamamoto S, Nonaka M, et al. Asymptomatic pneumothorax in patients with AIDS. Ann Chir 1993; 47: 844–7.
bilateral spontaneous pneumothorax [Japanese]. Kyobu Geka 133. Metersky ML, Colt HG, Olson LK, Shanks TG. AIDS-related
1994; 47: 215–18. spontaneous pneumothorax: risk factors and treatment. Chest
109. Ohara K, Yamazaki T, Sakaguchi K, Nakayama M, Kobayashi A. A 1995; 108: 946–51.
case of simultaneous bilateral spontaneous pneumothorax 134. Pastores SM, Garay SM, Naidich D, Rom WN. Review:
[Japanese]. Jpn J Thorac Surg 1994; 47: 1110–11. pneumothorax in patients with AIDS-related Pneumocystis carinii
110. Sayar A, Turna A, Metin M, et al. Simultaneous bilateral pneumonia. Am J Med Sci 1996; 312: 229–34.
spontaneous pneumothorax report of 12 cases and review of the 135. Spivak H, Keller S. Spontaneous pneumothorax in the AIDS
literature. Acta Chir Belg 2004; 104: 572–6. population. Am Surg 1996; 62: 753–6.
111. Dines DE, Clagett OT, Payne WS. Spontaneous pneumothorax in 136. Trachiotis GD, Vricella LA, Alyono D, Aaron BL, Hix WR.
emphysema. Mayo Clin Proc 1970; 45: 481–7. Management of AIDS-related pneumothorax. Ann Thorac Surg
112. Lai RS, Perng RP, Chang SC. Primary lung cancer complicated with 1996; 62: 1608–13.
pneumothorax. Jpn J Clin Oncol 1992; 22: 194–7. 137. Tumbarello M, Tacconelli E, Pirronti T, Cauda R, Ortona L.
113. Sakai S, Matsuoka H, Tsuwano S, et al. Lung cancer associated Pneumothorax in HIV-infected patients: role of Pneumocystis
with bulla found during an operation of pneumothorax; report of a carinii pneumonia and pulmonary tuberculosis. Eur Respir J 1997;
case [Japanese]. Kyobu Geka 2006; 59: 1209–12. 10: 1332–5.
114. Houacine S, Belhamri A, Damache MB, Ihaddaden A, Drif M. 138. Kuhlman JE, Knowles MC, Fishman EK, Sigelman SS. Premature
Recurrent pneumothorax during asthma attacks. Rev Pneumol Clin bullous damage in AIDS: CT diagnosis. Radiology 1989; 173: 23–6.
1986; 42: 204–6. ●139. Maurer ER, Schaal JA, Mendez FL. Chronic recurrent spontaneous
115. Hyde L. Benign spontaneous pneumothorax. Ann Intern Med 1962; pneumothorax due to endometriosis of the diaphragm. J Am Med
56: 746–51. Assoc 1958; 168: 2013–14.
116. Burke GJ. Pneumothorax complicating acute asthma. S Afr Med J 140. Suzuki S, Yasuda K, Matsumura Y, Kondo T. Left-side catamenial
1979; 55: 508–10. pneumothorax with endometrial tissue on the visceral pleura. Jpn
117. Pickup CM, Nee PA, Randall PE. Radiographic features in 1016 J Thorac Cardiovasc Surg 2006; 54: 225–7.
adults admitted to hospital with acute asthma. J Accid Emerg Med 141. Stern H, Toole AL, Merino M. Catamenial pneumothorax. Chest
1994; 11: 234–7. 1980; 78: 480–2.
118. Limthongkul S, Wongthim S, Udompanich V, Charoenlap P, 142. Guerin JC, Champel F, Martinat Y, Boniface E. Thoracoscopic study
Nuchprayoon C. Status asthmaticus: an analysis of 560 episodes of 6 cases of catamenial pneumothorax [French]. Rev Mal Respir
and comparison between mechanical and non-mechanical 1987; 4: 167–71.
ventilation groups. J Med Assoc Thai 1990; 73: 495–501. ◆143. Joseph J, Sahn SA. Thoracic endometriosis syndrome: new
119. Srinivas S, Varadhachary G. Spontaneous pneumothorax in observations from an analysis of 110 cases. Am J Med 1996; 100:
malignancy: a case report and review of the literature. Ann Oncol 164–70.
2000; 11: 887–9. 144. Bagan P, le Pimpec Barthes F, Assouad J, Souilamas R, Riquet M.
120. Ouellette D, Inculet R. Unsuspected metastatic choriocarcinoma Catamenial pneumothorax: retrospective study of surgical
presenting as unilateral spontaneous pneumothorax. Ann Thorac treatment. Ann Thorac Surg 2003; 75: 378–81.
Surg 1992; 53: 144–5. 145. Morcos M, Alifano M, Gompel A, Regnard JF. Life-threatening
121. Stein ME, Haim N, Drumea K, Ben-Itzhak O, Kuten A. Spontaneous endometriosis-related hemopneumothorax. Ann Thorac Surg 2006;
pneumothorax complicating chemotherapy for metastatic 82: 726–9.
seminoma: a case report and a review of the literature. Cancer 146. Kadokura M, Nonaka M, Yamamoto S, et al. Five cases of
1995; 75: 2710–3. asymptomatic spontaneous pneumothorax. Ann Thorac Cardiovasc
122. Furrer M, Althaus U, Ris HB. Spontaneous pneumothorax from Surg 1999; 5: 187–90.
radiographically occult metastatic sarcoma. Eur J Cardiothorac 147. Liston R, McLoughlin R, Clinch D. Acute pneumothorax: a
Surg 1997; 11: 1171–3. comparison of elderly with younger patients. Age Ageing 1994;
123. Sakurai H, Hada M, Miyashita Y, et al. Simultaneous bilateral 23: 393–5.
spontaneous pneumothorax secondary to metastatic 148. Loscertales J, Garcia Diaz F, Jimenez Merchan R, et al. Treatment
angiosarcoma of the scalp: report of a case. Surg Today 2006; 36: of spontaneous pneumothorax in patients over 70 years of age
919–22. [Spanish]. Arch Bronconeumol 1994; 30: 344–7.
●124. Heimlich HJ. Valve drainage of the pleural cavity. Dis Chest 1968; 149. Hagen RH, Reed W, Solheim K. Spontaneous pneumothorax. Scand
53: 282–7. J Thorac Cardiovasc Surg 1987; 21: 183–5.
125. Kolber C, Souquet PJ, Geriniere L, et al. Spontaneous ●150. Miller AC. Hypothesis: chest pain in primary spontaneous
pneumothorax revealing malignant mesothelioma. Rev Pneumol pneumothorax. Int J Clin Pract 2007; 61: 290–2.
Clin 1996; 52: 42–4. 151. Forlanini C. Early results of artificial pneumothorax in pulmonary
126. Alkhuja S, Miller A, Mastellone AJ, Markowitz S. Malignant pleural phthisis [Italian]. Gazz med Torina 1894; 45: 381–4, 401–3.
mesothelioma presenting as spontaneous pneumothorax: a case 152. Wu YC, Lu MS, Yeh CH, et al. Justifying video-assisted thoracic
series and review. Am J Ind Med 2000; 38: 219–23. surgery for spontaneous hemopneumothorax. Chest 2002; 122:
127. Weissberg D, Refaely Y. Pneumothorax: experience with 1,199 1844–7.
patients. Chest 2000; 117: 1279–85. ◆153. Hsu NY, Shih CS, Hsu CP, Chen PR. Spontaneous
128. Froudarakis ME, Bouros D, Voloudaki A, et al. Pneumothorax as a hemopneumothorax revisited: clinical approach and systemic
first manifestation of sarcoidosis. Chest 1997; 112: 278–80. review of the literature. Ann Thorac Surg 2005; 80: 1859–63.
530 Spontaneous pneumothorax
154. Askin FB, McCann BG. Reactive eosinophilic pleuritis. Arch Pathol 178. Noppen M, Alexander P, Driesen P, Slabbynck H, Verstraete A.
Lab Med 1977; 101: 187–91. Quantification of the size of primary spontaneous pneumothorax:
155. De Smedt A, Vanderlinden E, Demanet C, et al. Characterisation of accuracy of the Light index. Respiration 2001; 68: 396–9.
pleural inflammation occurring after primary spontaneous 179. Engdahl O, Toft T, Boe J. Chest radiograph – a poor method for
pneumothorax. Eur Respir J 2004; 23: 896–900. determining the size of a pneumothorax. Chest 1993; 103: 26–9.
156. Noppen M. Normal volume and cellular contents of pleural fluid. 180. Nissen H. Spontaneous pneumothorax: a clinical and histological
Paediatr Respir Rev 2004; 5: (Suppl A):S201–3. study of aetiological factors, pathogenesis and management
●157. Vincent M, Celard P, Pinet F, et al. Radiology of spontaneous [Danish]. Copenhagen (Thesis); 1969.
pneumothorax in young patients: a study of 200 cases [French]. 181. Clark S, Ragg M, Stella J. Is mediastinal shift on chest X-ray of
Sem Hop 1984; 60: 759–65. pneumothorax always an emergency? Emerg Med 2003; 15:
158. Aitchison F, Bleetman A, Munro P, McCarter D, Reid AW. Detection 429–33.
of pneumothorax by accident and emergency officers and 182. Kupferschmid JP, Carr T, Fonger JD, Aldea GS, Vosburgh E. Chronic
radiologists on single chest films. Arch Emerg Med 1993; 10: tension pneumothorax mimicking tension bullae: use of video-
343–6. assisted thoracoscopy for diagnosis. Chest 1993; 104: 1913–14.
159. Bulynin VI, Red’kin AN, Solod NV. Prehospital diagnostic errors and 183. Teplick SK, Clark RE. Various faces of tension pneumothorax.
therapeutic policy in spontaneous pneumothorax [Russian]. Klin Postgrad Med 1974; 56: 87–92.
Med 1997; 75: 41–3. ◆184. Gilbert TB, McGrath BJ. Tension pneumothorax: aetiology,
●160. Scadding JG, Wood P. Systolic clicks due to left-sided diagnosis, pathophysiology, and management. J Intensive Care
pneumothorax. Lancet 1939; ii: 1208–11. Med 1994; 9: 139–50.
161. Semple T, Lancaster WM. Noisy pneumothorax: observations based 185. Holloway VJ, Harris JK. Spontaneous pneumothorax: is it under
on 24 cases. Br Med J 1961; 2: 1342–6. tension? J Accid Emerg Med 2000; 17: 222–3.
162. Hamman L. Spontaneous mediastinal emphysema. Bull John 186. Chan SS. Current opinions and practices in the treatment of
Hopkins Hosp 1939; 64: 1–21. spontaneous pneumothorax. J Accid Emerg Med 2000; 17: 165–9.
163. Aisner M, Franco J. Mediastinal emphysema. New Engl J Med 187. Janssen JP, Cuesta MA, Postmus PE. Treatment of spontaneous
1949; 241: 818–25. pneumothorax: survey among Dutch pneumonologists [Dutch].
164. Freixinet J, Garcia F, Rodriguez PM, et al. Spontaneous Ned Tijdschr Geneeskd 1994; 138: 661–4.
pneumomediastinum: long-term follow-up. Respir Med 2005; 99: 188. Baumann MH, Strange C. The clinician’s perspective on
1160–3. pneumothorax management. Chest 1997; 112: 822–8.
165. Koullias GJ, Korkolis DP, Wang XJ, Hammond GL. Current 189. Sutherland M, Burdon J, Hart D. Primary spontaneous
assessment and management of spontaneous pneumothorax: treatment practices in Australia. Respirology 2000;
pneumomediastinum: experience in 24 adult patients. Eur J 5: 277–80.
Cardiothorac Surg 2004; 25: 852–5. ∗190. Baumann MH, Strange C, Heffner JE, et al. Management of
166. Henry M, Arnold T, Harvey J. BTS guidelines for the management spontaneous pneumothorax: an American College of Chest
of spontaneous pneumothorax. Thorax 2003; 58 (Suppl 2): Physicians Delphi Consensus Statement. Chest 2001; 119:
39S–52S. 590–602.
167. Bradley M, Williams C, Walshaw MJ. The value of routine 191. Light RW. Disorders of the pleura. In: Fauci AS, Braunwald E,
expiratory films in the diagnosis of pneumothorax. Arch Emerg Isselbacher KJ, et al. (eds). Harrison’s principles of internal
Med 1991; 8: 115–16. medicine, 14th edn. New York: McGraw Hill, 1998: 1472–5.
168. Schramel FM, Wagenaar M, Sutedja TG, Golding RP, Postmus PE. 192. Crofton J, Douglas A. Respiratory diseases, 1st edn. Oxford:
Diagnosis of pneumothorax not improved by additional roentgen Blackwell Scientific Publications, 1969.
pictures of the thorax in the expiration phase [Dutch]. Ned 193. Coope R. Diseases of the chest. Edinburgh: Livingstone, 1944.
Tijdschr Geneeskd 1995; 139: 131–3. 194. Crofton J, Douglas A. Respiratory diseases. 3rd ed. Oxford:
169. Eisen LA, Berger JS, Hegde A, Schneider RF. Competency in chest Blackwell Scientific Publications; 1981.
radiography: a comparison of medical students, residents, and 195. Rubin EH, Rubin M. Thoracic disease. Philadelphia: Saunders, 1961.
fellows. J Gen Intern Med 2006; 21: 460–5. 196. Cumming G, Semple SJ. Disorders of the respiratory system.
170. Waitches GM, Stern EJ, Dubinsky TJ. Usefulness of the double-wall Oxford: Blackwell Scientific Publications, 1973.
sign in detecting pneumothorax in patients with giant bullous 197. Jones JS. A place for aspiration in the treatment of spontaneous
emphysema. Am J Roentgenol 2000; 174: 1765–8. pneumothorax. Thorax 1985; 40: 66–7.
171. Bourgouin P, Cousineau G, Lemire P, Hebert G. Computed 198. Green RA. Pneumothorax. In: Baum GL (ed.). Textbook of
tomography used to exclude pneumothorax in bullous lung pulmonary diseases. London: Churchill, 1965: 694–5.
disease. J Can Assoc Radiol 1985; 36: 341–2. 199. Mercier C, Page A, Verdant A, et al. Outpatient management of
172. Glazer HS, Anderson DJ, Wilson BS, Molina PL, Sagel SS. intercostal tube drainage in spontaneous pneumothorax. Ann
Pneumothorax: appearance on lateral chest radiographs. Thorac Surg 1976; 22: 163–5.
Radiology 1989; 173: 707–11. 200. Conces DJ, Tarver RD, Gray WC, Pearcy EA. Treatment of
173. Axel L. A simple way to estimate the size of a pneumothorax. pneumothoraces utilizing small caliber chest tubes. Chest 1988;
Invest Radiol 1981; 16: 165–6. 94: 55–7.
174. Rhea J, de Luca S, Greene R. Determining the size of 201. Raja OG, Lalor AJ. Simple aspiration of spontaneous
pneumothorax in the upright patient. Radiology 1985; 144: pneumothorax. Br J Dis Chest 1981; 75: 207–8.
733–6. 202. Flint K, Hillawi AH, Johnson AM. Conservative management of
175. Collins CD, Lopez A, Mathie A, et al. Quantification of spontaneous pneumothorax. Lancet 1984; i: 687–8.
pneumothorax size on chest radiographs using interpleural 203. Archer GJ, Hamilton AA, Upadhyay R, Finlay M, Grace PM. Results
distances: regression analysis based on volume measurements of simple aspiration of pneumothoraces. Br J Dis Chest 1985; 79:
from helical CT. Am J Roentgenol 1995; 165: 1127–30. 177–82.
176. Light RW. Pneumothorax. In: Light RW (ed). Pleural diseases, 3rd 204. Harvey JE, Prescott RJ. Simple aspiration versus intercostal tube
edn. Baltimore: Williams & Wilkins, 1995: 242–77. drainage for spontaneous pneumothorax in patients with normal
177. Choi BG, Park SH, Yun EH, Chae KO, Shinn KS. Pneumothorax size: lungs. Br Med J 1994; 309: 1338–9.
correlation of supine anteroposterior with erect posteroanterior 205. Skoda J. Pneumothorax [German]. Allgem Wiener Med Zeitung
chest radiographs. Radiology 1998; 209: 567–9. 1857: 113.
References 531
206. Rodet N. Experimental research in pneumothorax [French]. 231. Ng AW, Chan KW, Lee SK. Simple aspiration of pneumothorax.
Congres Franc Med 1895: 894. Singapore Med J 1994; 35: 50–2.
207. Hill RC, DeCarlo DP, Jr., Hill JF, et al. Resolution of experimental 232. Mark JBD, p344 in Gobbel WG. Spontaneous pneumothorax. J
pneumothorax in rabbits by oxygen therapy. Ann Thorac Surg Thorac Cardiovasc Surg 1963; 46: 331–45.
1995; 59: 825–7; discussion 7–8. 233. Light RW. Management of spontaneous pneumothorax. Am Rev
208. England GJ, Hill RC, Timberlake GA, et al. Resolution of Respir Dis 1993; 148: 245–8.
experimental pneumothorax in rabbits by graded oxygen therapy. J 234. Britten S, Palmer SH, Snow TM. Needle thoracocentesis in tension
Trauma 1998; 45: 333–4. pneumothorax: insufficient cannula length and potential failure.
209. Northfield TC. Oxygen therapy for spontaenous pneumothorax. Br Injury 1996; 27: 321–2.
Med J 1971; 4: 86–8. 235. Rawlins R, Brown KM, Carr CS, Cameron CR. Life threatening
210. Chadha TS, Cohn MA. Noninvasive treatment of pneumothorax haemorrhage after anterior needle aspiration of pneumothoraces.
with oxygen inhalation. Respiration 1983; 44: 147–52. A role for lateral needle aspiration in emergency decompression of
●211. Kircher LT, Swartzel RL. Spontaneous pneumothorax and its spontaneous pneumothorax. Emerg Med J 2003; 20: 383–4.
treatment. J Am Med Assoc 1954; 153: 24–9. 236. Kiely DG, Ansari S, Davey WA, et al. Bedside tracer gas technique
212. Kelly AM, Loy J, Tsang AY, Graham CA. Estimating the rate of re- accurately predicts outcome in aspiration of spontaneous
expansion of spontaneous pneumothorax by a formula derived pneumothorax. Thorax 2001; 56: 617–21.
from computed tomography volumetry studies. Emerg Med J 2006; 237. So SY, Yu DY. Catheter drainage of spontaneous pneumothorax:
23: 780–2. suction or no suction, early or late removal? Thorax 1982; 37:
213. Baumann MH, Strange C. Treatment of spontaneous 46–8.
pneumothorax: a more aggressive approach? Chest 1997; 112: 238. Horsley A, Jones L, White J, Henry M. Efficacy and complications
789–804. of small-bore, wire-guided chest drains. Chest 2006; 130:
214. O’Rourke JP, Yee ES. Civilian spontaneous pneumothorax: 1857–63.
treatment options and long-term results. Chest 1989; 96: 1302–6. 239. Choi SH, Lee SW, Hong YS, et al. Can spontaneous pneumothorax
215. Peatfield RC, Edwards PR, Johnson NM. Two unexpected deaths patients be treated by ambulatory care management? Eur J
from pneumothorax. Lancet 1979; i: 356–8. Cardiothorac Surg 2007; 8: 8.
216. Noppen M, Alexander P, Driesen P, Slabbynck H, Verstraeten A. 240. Waller DA, Edwards JG, Rajesh PB. A physiological comparison of
Manual aspiration versus chest tube drainage in first episodes of flutter valve drainage bags and underwater seal systems for
primary spontaneous pneumothorax: a multicenter, prospective, postoperative air leaks. Thorax 1999; 54: 442–3.
randomized pilot study. Am J Respir Crit Care Med 2002; 165: 241. Mainini SE, Johnson FE. Tension pneumothorax complicating
1240–4. small-caliber chest tube insertion. Chest 1990; 97: 759–60.
217. Ayed AK, Chandrasekaran C, Sukumar M. Aspiration versus tube 242. Samelson SL, Goldberg EM, Ferguson MK. The thoracic vent:
drainage in primary spontaneous pneumothorax: a randomised clinical experience with a new device for treating simple
study. Eur Respir J 2006; 27: 477–82. pneumothorax. Chest 1991; 100: 880–2.
218. Andrivet P, Djedaini K, Teboul JL, Brochard L, Dreyfuss D. 243. Martin T, Fontana G, Olak J, Ferguson M. Use of pleural catheter
Spontaneous pneumothorax: comparison of thoracic drainage vs for the management of simple pneumothorax. Chest 1996; 110:
immediate or delayed needle aspiration. Chest 1995; 108: 335–9. 1169–72.
219. Miller AC. Spontaneous pneumothorax. In: Light RW, Lee YCG. 244. Roggla M, Wagner A, Brunner C, Roggla G. The management of
(eds). Textbook of pleural diseases, 1st edn. London: Arnold, 2003: pneumothorax with the thoracic vent versus conventional
445–63. intercostal tube drainage. Wien Klin Wochenschr 1996; 108:
220. Camuset J, Laganier J, Brugiere O, et al. Needle aspiration as first- 330–3.
line management of primary spontaneous pneumothorax. Presse 245. Dernevik L, Roberts D. Mini-drainage in pneumothorax is
Med 2006; 35: 765–8. expensive, but still beneficial [Swedish]. Lakartidningen 2000; 97:
221. Vedam H, Barnes DJ. Comparison of large- and small-bore 3726–8.
intercostal catheters in the management of spontaneous 246. Holman CW, Muschenheim C. Bronchopulmonary diseases and
pneumothorax. Intern Med J 2003; 33: 495–9. related disorders. Baltimore: Harper & Row, 1972.
222. Tsai WK, Chen W, Lee JC, et al. Pigtail catheters vs large-bore 247. Rozenman J, Yellin A, Simansky DA, Shiner RJ. Re-expansion
chest tubes for management of secondary spontaneous pulmonary oedema following spontaneous pneumothorax. Respir
pneumothoraces in adults. Am J Emerg Med 2006; 24: 795–800. Med 1996; 90: 235–8.
223. Knottenbelt JD, van der Spuy JW. Traumatic pneumothorax: a 248. Mahfood S, Hix WR, Aaron BL. Re-expansion pulmonary edema.
scheme for rapid patient turnover. Injury 1990; 21: 77–80. Ann Thorac Surg 1988; 45: 340–4.
224. Burgaud JC, Offenstadt G, Bencharif G, Hericord P, Amstutz P. 249. Matsuura Y, Nomimura T, Nurikami H. Clinical analysis of re-
Spontaneous pneumothorax: drainage by catheter or drain? Rev expansion pulmonary edema. Chest 1991; 100: 1562–6.
Pneumol Clin 1985; 41: 317–19. 250. Miller WC, Toon R, Palat H, Lacroix J. Experimental pulmonary
225. Lee P, Chee CB, Wang YT. Questionnaire survey on management of edema following re-expansion of pneumothorax. Am Rev Respir
spontaneous pneumothorax. Singapore Med J 2000; 41: 538–41. Dis 1973; 108: 664–6.
226. Courtney PA, McKane WR. Audit of the management of 251. Sharma TN, Agnihotri SP, Jain NK, Madan A, Deopura G.
spontaneous pneumothorax. Ulster Med J 1998; 67: 41–3. Intercostal tube thoracostomy in pneumothorax: factors
227. Baldt MM, Bankier AA, Germann PS, et al. Complications after influencing re-expansion of lung. Indian J Chest Dis Allied Sci
emergency tube thoracostomy: assessment with CT. Radiology 1988; 30: 32–5.
1995; 195: 539–43. 252. Reed MF, Lyons JM, Luchette FA, Neu JA, Howington JA.
228. Griffiths JR, Roberts N. Do junior doctors know where to insert Preliminary report of a prospective, randomized trial of
chest drains safely? Postgrad Med J 2005; 81: 456–8. underwater seal for spontaneous and iatrogenic pneumothorax. J
229. Burge TS. Complications of prophylactic intercostal tube drainage Am Coll Surg 2007; 204: 84–90.
– including tension pneumothorax. J R Army Med Corps 1992; 253. Marshall M, Deeb M, Bleier J, et al. Suction vs water seal after
138: 138–9. pulmonary resection. Chest 2002; 121: 831–5.
230. Landay M, Oliver Q, Estrera A, et al. Lung penetration by 254. Ayed AK. Suction versus water seal after thoracoscopy for primary
thoracostomy tubes: imaging findings on CT. J Thorac Imaging spontaneous pneumothorax: prospective randomized study. Ann
2006; 21: 197–204. Thorac Surg 2003; 75: 1593–6.
532 Spontaneous pneumothorax
255. Bell RL, Ovadia P, Abdullah F, Spector S, Rabinovici R. Chest tube 269. Chee CB, Abisheganaden J, Yeo JK, et al. Persistent air-leak in
removal: end-inspiration or end-expiration? J Trauma Injury Infect spontaneous pneumothorax: clinical course and outcome. Respir
Crit Care 2001; 50: 674–7. Med 1998; 92: 757–61.
256. Tanaka F, Ezaki H, Isobe J, et al. Four cases of spontaneous 270. Curtis HJ, Bourke SJ, Dark JH, Corris PA. Lung transplantation
pneumothorax with no reexpansion of collapsed lung despite outcome in cystic fibrosis patients with previous pneumothorax. J
chest tube drainage [Japanese]. Jpn J Thorac Surg 1992; 45: Heart Lung Transplant 2005; 24: 865–9.
515–18. 271. Noyes BE, Orenstein DM. Treatment of pneumothorax in cystic
257. Kakaris S, Athanassiadi K, Vassilikos K, Skottis I. Spontaneous fibrosis in the era of lung transplantation. Chest 1992; 101:
hemopneumothorax: a rare but life-threatening entity. Eur J 1187–8.
Cardiothorac Surg 2004; 25: 856–8. 272. Edenborough FP, Hussain I, Stableforth DE. Use of a Heimlich
258. Fackeldey V, Franke A, Schachtrupp A, Becker HP, Schwab R. flutter valve for pneumothorax in cystic fibrosis. Thorax 1994; 49:
Physical fitness after apical resection for the treatment of primary 1178–9.
spontaneous pneumothorax. Mil Med 2005; 9: 760–3. 273. Spector ML, Stern RC. Pneumothorax in cystic fibrosis: a 26-year
259. Cardillo G, Carleo F, Carbone L, et al. Long-term lung function experience. Ann Thorac Surg 1989; 47: 204–7.
following videothoracoscopic talc poudrage for primary 274. Flume PA, Strange C, Ye X, et al. Pneumothorax in cystic fibrosis.
spontaneous recurrent pneumothorax. Eur J Cardiothorac Surg Chest 2005; 128: 720–8.
2007; 28: 28. 275. Vricella LA, Trachiotis GD. Heimlich valve in the management of
260. Passlick B, Born C, Sienel W, Thetter O. Incidence of chronic pain pneumothorax in patients with advanced AIDS. Chest 2001; 120:
after minimal-invasive surgery for spontaneous pneumothorax. 15–18.
Eur J Cardiothor Surg 2001; 19: 355–8. ●276. Ong YE, Sheth A, Simmonds NJ, et al. Radiotherapy: a novel
261. Ben-Nun A, Soudack M, Best LA. Video-assisted thoracoscopic treatment for pneumothorax. Eur Respir J 2006; 27: 427–9.
surgery for recurrent spontaneous pneumothorax: the long-term 277. Tripp HF, Thomas LP, Obney JA. Current therapy of catamenial
benefit. World J Surg 2006; 30: 285–90. pneumothorax. Heart Surg Forum 1998; 1: 146–9.
262. Schramel FM, Sutedja TG, Braber JC, van Mourik JC, Postmus PE. 278. Marshall MB, Ahmed Z, Kucharczuk JC, Kaiser LR, Shrager JB.
Cost-effectiveness of video-assisted thoracoscopic surgery versus Catamenial pneumothorax: optimal hormonal and surgical
conservative treatment for first time or recurrent spontaneous management. Eur J Cardiothorac Surg 2005; 27: 662–6.
pneumothorax. Eur Respir J 1996; 9: 1821–5. 279. Alfageme I, Moreno L, Huertas C, et al. Spontaneous
263. Abdala OA, Levy RR, Bibiloni RH, et al. Advantages of video pneumothorax: long-term results with tetracycline pleurodesis.
assisted thoracic surgery in the treatment of spontaneous Chest 1994; 106: 347–50.
pneumothorax. Medicina 2001; 61: 157–60. ◆280. Schramel FM, Postmus PE, Vanderschueren RG. Current aspects of
264. Torresini G, Vaccarili M, Divisi D, Crisci R. Is video-assisted thoracic spontaneous pneumothorax. Eur Respir J 1997; 10: 1372–9.
surgery justified at first spontaneous pneumothorax? Eur J 281. Lippert HL, Lund O, Blegvad S, Larsen HV. Independent risk factors
Cardiothorac Surg 2001; 20: 42–5. for cumulative recurrence rate after first spontaneous
265. Sawada S, Watanabe Y, Moriyama S. Video-assisted thoracoscopic pneumothorax. Eur Respir J 1991; 4: 324–31.
surgery for primary spontaneous pneumothorax: evaluation of 282. Videm V, Pillgram-Larsen J, Ellingsen O, Andersen G, Ovrum E.
indications and long-term outcome compared with conservative Spontaneous pneumothorax in chronic obstructive pulmonary
treatment and open thoracotomy. Chest 2005; 127: 2226–30. disease: complications, treatment and recurrences. Eur J Respir Dis
266. Iablonskii PK, Atiukov MA, Pishchik VG, El’kina EA. Diagnostic and 1987; 71: 365–71.
treatment strategy for patients with the first episode of primary 283. Sadikot RT, Greene T, Meadows K, Arnold AG. Recurrence of
spontaneous pneumothorax [Russian]. Vestn Khir Im I I Grek 2005; primary spontaneous pneumothorax. Thorax 1997; 52: 805–9.
164: 11–14. 284. Sihoe ADL, Yim APC, Lee TW, et al. Can CT scanning be used to
267. Herrejon A, Inchaurraga I, Vivas C, Custardoy J, Marin J. Initial select patients with unilateral primary spontaneous pneumothorax
pleural pressure measurement in spontaneous pneumothorax. for bilateral surgery? Chest 2000; 118: 380–3.
Lung 2000; 178: 309–16. 285. Haid MM, Paladini P, Maccherini M, et al. Air transport and the
268. Mathur R, Cullen J, Kinnear WJ, Johnston ID. Time course of fate of pneumothorax in pleural adhesions. Thorax 1992; 47:
resolution of persistent air leak in spontaneous pneumothorax. 833–4.
Respir Med 1995; 89: 129–32.
43
Non-spontaneous pneumothorax
MICHAEL H BAUMANN
INTRODUCTION Classification
Pneumothorax references continue to subdivide pneumo- Traumatic pneumothoraces may be classified by mecha-
thorax into spontaneous and traumatic.1–3 The preceding nism of injury into penetrating and non-penetrating
chapter (Chapter 42) dealt with spontaneous pneumo- (blunt).2 The object penetrating the chest disrupts the vis-
thoraces. Spontaneous pneumothoraces occur without ceral pleural surface prompting pleural air entry or allows
preceding trauma or obvious underlying precipitating air entry through the breached chest wall. The etiology of
cause and are classified as primary (no clear underlying pneumothorax arising from blunt chest trauma is less
lung disease) and secondary (an underlying causal lung immediately apparent in the absence of a rib fracture lac-
disease present). This chapter will focus upon traumatic erating the visceral pleura. Blunt chest trauma may
and tension pneumothoraces. Traumatic pneumothoraces prompt abrupt alveolar pressure elevations causing alveo-
occur as a result of direct or indirect trauma and can be lar rupture. Air may then dissect into the lung interstitium
subdivided into non-iatrogenic and iatrogenic. Non- and then to the visceral pleural or the mediastinal pleural
iatrogenic pneumothoraces occur as a result of direct or surface. Rupture of either the visceral or mediastinal
indirect trauma, generally to the chest, unrelated to any pleura may then cause a pneumothorax.
medical procedure. Traumatic pneumothoraces resulting
from medical interventions are termed iatrogenic pneu-
mothoraces. Tension pneumothorax may occur with any Clinical presentation
of the aforementioned categories of pneumothorax.
Up to 6 percent of traumatic pneumothoraces present with
tension.8 Alternatively, a substantial number of traumatic
NON-IATROGENIC TRAUMATIC pneumothoraces may be occult, not seen on an initial chest
PNEUMOTHORAX radiograph but found by additional imaging. The propor-
tion of trauma-related pneumothoraces that are occult
Incidence and epidemiology compared with those seen on chest radiography ranges
from 29 to 72 percent,9,10 emphasizing the need for a high
Trauma related deaths exceed 50 000 per year in the index of suspicion and incorporating early routine com-
USA. Chest trauma is causal in 25 percent of these deaths puted tomography (CT) in all chest and multi-trauma
and plays a significant role in an additional 50 percent. patients.7 Subcutaneous emphysema, pulmonary contu-
Pneumothorax ranks second to rib fractures as the most sion, and rib fracture(s) are independent predictors of
common manifestation of chest injury and may be noted an occult pneumothorax (Table 43.1).10 CT imaging
in 40–50 percent of patients with chest trauma.4–6 detects other abnormalities in addition to pneumothorax,
Traumatic pneumothorax may also present after blunt including lung contusion, diaphragmatic rupture and
trauma to the abdomen.4,6 At least 20 percent of trau- hemothorax leading to an alteration in patient care in up to
matic pneumothoraces have an accompanying hemo- 41 percent of cases.7 Pneumothorax detection is critical
thorax.7 given concerns for progression, particularly if the patient is
534 Non-spontaneous pneumothorax
Predictor Sensitivity (%) Specificity (%) PPV (%) NPV (%) Odds ratio p -value
mechanically ventilated or anesthesia becomes neces- failure with a conservative approach appears significant.
sary.5,6,9,10 Consideration and evaluation for disruption of Bridges and colleagues6 noted that of 25 patients with
major airways should be considered in the traumatic pneu- occult traumatic pneumothoraces treated conservatively
mothorax patient, often with a persistent air leak, where (none intubated), five (20 percent) required chest tube
their condition deteriorates out of proportion to the find- placement due to respiratory distress, hemodynamic insta-
ings on the chest radiograph or physical examination.11,12 bility or pneumothorax progression. Further, in this retro-
spective study, all occult pneumothorax patients requiring
intubation (an additional 10 patients) were, by preference,
Treatment and management treated with a chest tube.6 Enderson and colleagues’5
prospective study of 40 trauma patients sustaining occult
Generally, traumatic pneumothoraces have been treated pneumothorax randomized to chest tube placement
with placement of a chest tube.5,13 This approach remains (n = 19) or observation (n = 21) noted that 8 of the 21 (38
the subject of debate and scrutiny in the setting of both percent) patients conservatively managed had pneumo-
non-occult and occult traumatic pneumothoraces.5,8–10,13,14 thorax progression with mechanical ventilation with three
of the eight (38 percent) developing a tension pneumo-
NON-OCCULT TRAUMATIC PNEUMOTHORACES thorax. The authors conclude that patients with traumatic
occult pneumothoraces requiring positive-pressure venti-
Conservative management, i.e. observation without chest lation should have a chest tube placed.
tube placement, has been reported in selected patients sus- By contrast, a retrospective study of occult pneumo-
taining non-occult traumatic pneumothoraces with a thorax promotes the safety of withholding immediate
failure rate of 7 to 9 percent.8,13 Johnson13 retrospectively chest tube placement in 13 of 26 patients treated conserv-
reported 29 patients with minimal, small or moderate atively.14 Six of these conservatively managed patients
sized traumatic pneumothoraces managed without received intubation and positive-pressure ventilation with
drainage; two patients (6.9 percent) required chest tube one (17 percent) requiring chest tube placement. An addi-
drainage due to asymptomatic radiographic pneumo- tional patient initially receiving immediate chest tube
thorax enlargement. Johnson emphasized that drainage is placement removed his chest tube during mechanical ven-
mandatory when positive pressure ventilation is required tilation and required emergent replacement due to recur-
or the patient sustains respiratory compromise from lung rent pneumothorax. Despite a nearly 20 percent
collapse.13 Knottenbelt and van der Spuy8 prospectively complication rate in pneumothorax patients without chest
limited chest tube placement in 804 traumatic pneumo- tubes on mechanical ventilation, the authors conclude that
thorax patients (n = 645, penetrating knife wounds) to the their data do not support the use of prophylactic chest
following: those with a significant pneumothorax by chest tube placement before mechanical ventilation in patients
radiograph (>1.5 cm lung collapse, arbitrarily selected); with occult traumatic pneumothoraces.14
smaller pneumothoraces if bilateral; patients requiring Brasel and colleagues15 also provide evidence regarding
mechanical ventilation; and if the patient’s respiratory conservative management of CT diagnosed occult pneu-
reserve was limited for any reason. Forty-one percent mothoraces even during mechanical ventilation. The
(n = 329) of patients met criteria for conservative manage- authors note no difference in complication rates in 39
ment; 29 of these 329 patients (8.8 percent) treated conser- patients with 44 pneumothoraces prospectively random-
vatively required chest tube placement for an enlarging ized to chest tube placement (n = 18) versus observation
pneumothorax.8 Only four of the 804 patients suffered a (n = 21). Nine patients in each group were mechanically
gunshot wound and all patients were observed in hospital. ventilated. Interestingly, the four patients in the chest tube
group having pneumothorax progression (three on
OCCULT TRAUMATIC PNEUMOTHORACES mechanical ventilation) were not on pleural suction and
had suction reinstituted as therapy for pneumothorax pro-
Occult pneumothoraces may also be successfully treated gression. Lack of suction may account for the reported
conservatively, arguably, even in the setting of positive- pneumothorax progression. Three of the observation
pressure mechanical ventilation. However, the risk of group had pneumothorax progression, including two on
Non-iatrogenic traumatic pneumothorax 535
mechanical ventilation. If the three patients in the chest ment. A prospective utilization of CT classification of 44
tube group on mechanical ventilation had been main- traumatic pneumothoraces (36 patients), by size from
tained on suction and had no pneumothorax progression, minuscule to limited anterior to anterolateral (smallest to
the authors’ reported lack of difference in complication largest), touts the safety of close inpatient observation in
rates between the two groups would disappear. Hence, this the first two groups.4 Eighty-nine percent, or 24 of 27
study does not provide definitive evidence for superiority patients in the minuscule and anterior pneumothorax size
of conservative management of occult traumatic pneumo- groups, did not require tube thoracostomy. Chest tubes
thoraces, particularly in the presence of mechanical venti- were placed in all anterolateral pneumothoraces based on
lation. the preference of the institution’s surgeons and chest tubes
Ball and colleagues10 provide the largest prospective were preferentially placed in all anterior pneumothoraces
study of occult traumatic pneumothoraces revealing a pref- receiving mechanical ventilation.4 A similar but retrospec-
erence for chest tube placement in occult pneumothorax tive study incorporating CT size classification of occult
patients that are mechanically ventilated. Paired chest radi- traumatic pneumothoraces suggests that small occult
ographs and CT scans were available in 338 of 761 trauma pneumothoraces may be managed by observation even in
patients (44 percent) (98.5 percent blunt trauma). One- the presence of positive-pressure ventilation. Larger pneu-
hundred and three pneumothoraces were present in 89 mothoraces and those associated with two rib fractures
patients and 57 (55 percent) of these pneumothoraces were may require early chest tube placement.16
occult. Independent risk factors for occult pneumothorax Assimilating this information, chest tube placement in
were identified (see above, Table 43.1). In this single insti- patients with traumatic pneumothoraces seems a reason-
tution study, no specific protocol was in place prompting able approach in many but not all patients (Figure 43.1).
chest tube placement. Twenty-three of 49 patients (47 Such tubes may serve to treat not only the pneumothorax
percent) with an occult pneumothorax received a chest but also the accompanying hemothorax found in 20
tube (28 French in 33 percent and 32 French in 57 percent, percent of cases.7 After initial chest radiographic and CT
p = 0.02). Occult pneumothoraces were found to be treated assessment of the chest, if the patient has a large pneumo-
differently if the patients required mechanical ventilation. thorax (arbitrarily, greater than 1.5 cm collapse of the lung
In the occult pneumothorax non-ventilated patient sub- from the parietal pleura seen by chest radiograph), hemo-
group, 10 of 32 (31 percent) received a chest tube compared dynamic or respiratory instability, or the presence of an
with 13 of 17 (76 percent) of the subgroup being mechani- accompanying hemothorax, a large-bore chest tube
cally ventilated (p = 0.03). No differences in complications (28–32 F) should be strongly considered. If only a chest
rate were noted between the ventilated and non-ventilated radiograph is available, the presence of subcutaneous
chest tube groups. Of four of 17 (24 percent) ventilated emphysema, pulmonary contusion or rib fracture(s)
patients and of 22 of 32 (69 percent) nonventilated patients should prompt consideration of an occult pneumothorax
who were observed without chest tube insertion, two (8 (Table 43.1). A large-bore tube is suggested to effectively
percent) later required tube placement due to pneumo- drain any potential blood or large air leak. If the pneumo-
thorax progression (one patient each in the ventilated and thorax is small (less than 1.5 cm collapse by chest radi-
nonventilated group). ograph) or occult by CT (minuscule or anterior) and
Classification of the size of CT classification of size may the patient is hemodynamically and respiratory stable,
be useful in determining the need for chest tube place- very close in-hospital observation with a series of chest
Society notes that this rule seems arbitrary without ciated with significant morbidity, occasional mortality and
accounting for type of pneumothorax, underlying disease, increased hospital stay and cost.29,32 Unfortunately, despite
therapeutic interventions or demographic factors but ulti- increased national emphasis and interest in patient
mately suggests that patients may travel safely 6 weeks after safety,33 recent measures to reduce iatrogenic complica-
a definitive surgical recurrence prevention procedure or tions have not reduced the occurrence of IP. Resident
chest radiographic resolution of unspecified types of pneu- work-hour limits have not reduced the incidence of IP.34
mothoraces.24 Also, the Veteran’s Administration system-wide imple-
mentation of patient safety indicators, including those for
IP, actually resulted in increased rates of IP.35 This may not
IATROGENIC PNEUMOTHORAX reflect worse care but increased reporting of IP.35
potentially important in detecting delayed pneumo- placed or simple aspiration ranges from 546 to 53 percent.43
thoraces. Prospectively, each of 64 patients underwent a The incidence of post-biopsy pneumothorax does not
CT reimaging with two additional views immediately after appear to differ whether using CT or fluoroscopy guid-
needle biopsy, and an expiratory chest radiograph was ance.47 Finally, placing the patient in the lateral decubitus
obtained 4 hours post procedure.39 Only 2 of 64 patients (3 position with biopsy side down does not reduce the inci-
percent) developed a delayed pneumothorax detected by dence of transthoracic needle biopsy related IP.48–50
the 4-hour post-procedure chest radiograph. Only one of The value of preoperative lung function predicting IP
these two patients required chest tube drainage (1.5 occurrence is debatable.39,51–54 Three studies,39,42,53 includ-
percent). The authors conclude that post-procedure chest ing one likely suffering beta error,39 have found no corre-
radiographs should be omitted and instead state that lation between lung function findings compatible with
patients should be instructed on signs and symptoms of obstructive lung disease and CT-directed needle biopsy IP
pneumothorax and seek medical attention if any of these rates. An early study by Poe and colleagues54 of fluoro-
occur.39 By contrast, Choi and colleagues40 report that 15 scopically directed needle biopsy of lung lesions found a
of 458 patients (3.3 percent) undergoing transthoracic correlation between increased total lung capacity (a poten-
needle biopsy under fluoroscopic (n = 280), CT (n=21) or tial measure of obstruction) and pneumothorax occur-
ultrasound (n = 157) guidance developed a delayed pneu- rence. The same data published in two venues noted a
mothorax (occurring ≥3 hours post procedure, range correlation of obstructive lung function changes and CT-
5–120 hours) not detected on an immediate post-proce- guided needle biopsy related pneumothorax.51,52
dure chest radiograph. Three of these 15 patients (20 Obstructive lung disease predicted a 46 percent pneumo-
percent) required tube drainage.40 The authors considered thorax event rate compared with a 19 percent rate in
the incidence of delayed pneumothorax and subsequent patients with normal lung function;51 the FEV1 (forced
need for tube drainage clinically important.40 expiratory volume in one second) was the most predictive
single variable.52 Notably the later two studies obtained
pulmonary function in only 35 percent of reported
Risk factors for iatrogenic pneumothorax patients,51,52 raising the issue of study bias. However, a
more recent study supported the thought that a lower
Transthoracic needle aspiration and intensive care setting FEV1 is predictive of iatrogenic pneumothorax occurrence
related IP have associated variables that may predict pneu- in univariant but not multiple regression analysis.43
mothorax risk. Multiple studies have assessed potential The incidence of mechanical ventilation related pneu-
risk factors associated with the development of a pneumo- mothorax has declined,29,31 likely secondary to newer
thorax after transthoracic needle biopsy (Table 43.2).40–46 modes of mechanical ventilation29 and lung protective
Greater lesion depth is consistently associated with a ventilation strategies in acute respiratory distress syn-
higher likelihood of an associated pneumothorax.41–46 drome (ARDS).55,56 However, pneumothorax remains one
Smaller lesions size also appears to be a risk factor for of the most common iatrogenic complications in the
pneumothorax.40,41,44,46 Less consistently, the presence of intensive care unit (ICU).57 IP in the ICU occurs primarily
or severity of emphysema is reported as a risk factor for as a complication of volutrauma from mechanical ventila-
pneumothorax.40,41,45 Paradoxically, the number of needle tion or after a procedure.57 A large prospective observa-
passes41,44–46 and needle size41,45,46 are not associated with tional study of 3430 patients in the ICU for more than 24
pneumothorax development. The percentage of patients hours recently identified associated risks for the develop-
with a pneumothorax subsequently having a drainage tube ment of IP in the ICU (Table 43.3).57 IP was seen in 94
Table 43.2 Potential risk factors associated with the development of a pneumothorax after transthoracic needle biopsy
Author Study type n (biopsies Iatrogenic Lesion Small Emphysema Number Needle
or patients) pneumothorax depth lesion of needle size
( n , %) size passes
Table 43.3 Statistically significant risk factors (multivariate analysis) associated with the occurrence of an iatrogenic pneumothorax in
the intensive care unit setting57
patients (2.7 percent) within the first 30 days. The most recommended.2 Given that the chest radiograph may be a
common causes of IP, in descending order, were volu- poor guide to predicting pneumothorax size59 and the for-
trauma (n = 42), and invasive procedures including central mulae to calculate percent pneumothorax size can be cum-
venous catheter insertion (n = 28), thoracentesis (n = 21), bersome, a simpler approach may be useful. A
and other procedures in three. Mortality risk was 2.6 times measurement from the visceral to parietal pleural surface
greater in patients with IP versus those without (95 percent as used in traumatic pneumothorax8 may be easier.
confidence interval, 1.3–4.9). The simplified acute physio- Suggested size cut points outlined in the American College
logic score (SAPS II) on the first ICU day was greater in of Chest Physicians (ACCP) statement on management of
patients suffering an IP (p = 0.001). spontaneous pneumothorax58 may be used: <3 cm and
≥3 cm lung collapse, for <15 percent and >15 percent
pneumothorax size, respectively.
Treatment and management Observation and simple aspiration may be successful in
managing selected patients with an IP.44 However, given
Current management of patients suffering iatrogenic the variety of commercially available small-bore
pneumothoraces is variable, with the development of catheters60 that can be attached to a Heimlich valve or
treatment protocols complicated by inadequate monitor- similar one-way valve device, incorporating a versatile
ing and reporting of these iatrogenic events.32 As opposed small-bore catheter instead of a simple aspiration is rec-
to spontaneous pneumothorax,1,58 pneumothorax recur- ommended where drainage is deemed necessary. Other
rence considerations are not an issue with patients suffer- commercially packaged alternatives containing a catheter
ing an iatrogenic pneumothorax. Instead, treatment and one-way valve system in one unit are also available
focuses upon the least invasive intervention appropriate to and noted to be successful in iatrogenic pneumothoraces
the patient’s underlying lung health and associated clinical (TRU-CLOSE‰ Thoracic Vent; Davis and Geck, Wayne,
circumstances (Figure 43.2). NJ, USA).61 Caution is urged when incorporating a
A recent text recommends observation and oxygen sup- Heimlich valve, or similar one-way valve device, particu-
plementation for patients not mechanically ventilated with larly for patients discharged to home. Clear instructions
minimal symptoms and a limited (<15 percent) pneumo- regarding device orientation to ensure airflow out of the
thorax. If a patient has more than minimal symptoms or a chest and device maintenance are requisite to prevent
larger pneumothorax (>15 percent), simple aspiration is complications, including tension pneumothorax.62,63
Clinically stable Symptomatic or Small-bore chest tube Figure 43.2 Management algorithm for iatrogenic
COPD** COPD pneumothorax. *See text for details; **COPD, chronic
obstructive pulmonary disease; +, positive; -, negative.
Any patient on mechanical ventilation requires a large-
Observe Small-bore chest tube bore chest tube.
540 Non-spontaneous pneumothorax
The presence of obstructive lung disease predicts the completed treatment in 1–7 days.31 Patients with chronic
need for chest tube placement in patients undergoing a obstructive pulmonary disease (COPD) required signifi-
needle biopsy of the lung39,41,51,53 and for longer duration cantly longer treatment times than those without COPD.
of treatment post pneumothorax compared with patients Thoracotomy had to be performed in one of 535 (0.2
without a history of obstructed lung disease.31 Cox and percent) patients because of a persistent air leak.
colleagues41 reported a significantly greater need for chest Schoenenberger and colleagues70 retrospective review
tube placement in patients with underlying computed of 47 patients provides more direct information regarding
tomographic evidence of emphysema (27 percent with evi- the occurrence of persistent air leaks and the role and
dence versus 9 percent without, p < 0.01) sustaining a timing of surgical intervention in patients suffering an
needle lung biopsy related iatrogenic pneumothorax.41 iatrogenic pneumothorax. Patients with at least a 20
Hence, initial placement of a small-bore catheter and for- percent iatrogenic pneumothorax and placement of a 20 to
going observation is preferred in such patients. 24 F chest tube were analyzed. Percent air leak resolution
Safe outpatient management utilizing a small-bore was 100 percent at 72 hours after chest tube placement for
catheter attached to a Heimlich valve has been reported in patients without underlying lung disease and 71 percent
central venous catheter related iatrogenic pneumothoraces for those with an underlying lung disease. Air leak resolu-
patients.37 This retrospective study notes only 60 percent tion reached a plateau at 72 hours for those with underly-
success (resolution of pneumothorax with no subsequent ing lung disease; no statistically significant increase in air
air leak or pneumothorax enlargement) with observation leak resolution occurred after 72 hours.70 Hence, use of an
and a 40 percent failure rate requiring placement of a chest invasive intervention including VATS or thoracotomy
catheter. Alternatively, 85 percent success accompanies the should be strongly considered for patients suffering an
initial placement of a small-bore catheter (8.5 F) attached iatrogenic pneumothorax with an air leak persisting at 72
to a Heimlich valve with a mean catheter residence time of hours after chest tube placement.
only 1.6 days. Similar reports of successful and safe man-
agement of iatrogenic pneumothoraces with small-bore
catheters, including in outpatients, placed initially or after TENSION PNEUMOTHORAX
short periods of unsuccessful observation have been pub-
lished.64,65 Incidence, epidemiology and pathophysiology
All mechanically ventilated patients developing an iatro-
genic pneumothorax from any cause should receive a chest Both spontaneous and traumatic (non-iatrogenic and
tube. Iatrogenic pneumothoraces resulting directly from iatrogenic) pneumothoraces may present with or develop
positive-pressure mechanical ventilation may be accompa- into tension pneumothorax. Tension pneumothorax is
nied by bronchopleural fistula and the development of present when the intrapleural pressure exceeds atmos-
tension pneumothorax.66,67 Air leaks in this setting may be pheric pressure throughout expiration and frequently
quite large and variably contribute to alteration in physio- during inspiration.2 Many patients suffering a tension
logical gas exchange including increased carbon dioxide pneumothorax have positive pressure applied to their
excretion and decreased utilization of inspired oxygen due airways during mechanical ventilation71 or during resusci-
to escape of carbon dioxide and oxygen through the tation.72 Seventy-one of 74 patients developing pneumo-
fistula.66 Mechanically ventilated patients with an iatro- thorax during mechanical ventilation had a tension
genic pneumothorax require prompt chest tube placement, component in Steier and colleagues review.26 Presentation
often large bore,66–69 and foregoing consideration of obser- is particularly abrupt in patients receiving mechanical ven-
vation. In addition to the placement of an appropriate size tilation and may carry a particularly high mortality rate.26
chest tube, adjusting the ventilator mode to limit the Intra- and post-cardiopulmonary resuscitation (CPR)
amount of positive pressure delivered to the patient may patients should receive special scrutiny for the possibility
assist in minimizing the air leak and healing the fistula.66,67 of tension of pneumothorax, particularly if positive-pres-
sure ventilation is administered. Unsuspected or untreated
tension pneumothoraces were found in 12 of 3500 autop-
Surgical management and persistent air leaks sies. Ten of these 12 patients were receiving some form of
endotracheal positive pressure. Eight of these 10 patients
The incidence of iatrogenic pneumothorax related air were mechanically ventilated; two were receiving intermit-
leaks appears related to the underlying etiology of the tent positive pressure. Seven of these 10 patients had also
pneumothorax and any associated lung disease. The received CPR.72 Successful diagnosis and treatment of
Veterans Administration cooperative study does not intra- and post-resuscitation tension pneumothorax could
clearly delineate the treatment modalities incorporated in mean the difference between successful and unsuccessful
patients with an iatrogenic pneumothorax. However, CPR. Tension may also develop in spontaneously breath-
using duration of treatment as a rough surrogate for pre- ing patients but inherent is the presumed presence of a
sumed accompanying air leak, 288 of 385 patients (75 one-way valve process permitting air to enter the pleural
percent) in the Veterans Administration cooperative study space during inspiration but not exit during expiration.
Future directions 541
The dramatic clinical presentation of tension marked cycle ventilation. Worsening in both dynamic and static
by cardiopulmonary compromise is likely multifactorial compliance can also be found in pulmonary edema,
in etiology. The decrease in the partial pressure of arterial ARDS, pneumonia and atelectasis.82 The rapidity of dete-
oxygen (PaO2) and increase in alveolar–arterial oxygen rioration in lung compliance may be a clue to tension
difference that may be seen with any pneumothorax pneumothorax over the other possibilities. If the patient is
results from alterations in ventilation–perfusion relation- on pressure-cycled ventilation, decreasing delivered tidal
ships, anatomic shunt and dead space.73,74 Anatomic volumes is noted.
shunt may in fact increase secondary to worsening venti- The chest radiograph itself may be a misleading tool to
lation–perfusion relationships after pleural air drainage diagnose tension. The classic radiographic findings of
with improvement delayed up to 90 minutes emphasizing tension may indeed not represent tension.2 Classic radi-
the need for supplemental oxygen in pneumothorax ographic findings include a shift of the mediastinum to the
patients.73 contralateral side, ipsilateral enlargement of the hemitho-
Decreased PaO2 and impaired cardiac output, alone or rax and depressed hemidiaphragm. Such findings may
in combination, play a role in the clinical decompensation occur due to the natural recoil properties of the chest wall
noted in tension pneumothorax. Earlier studies performed and the lung when air enters the pleural space without
in goats and monkeys with induced tension pneumothorax intrapleural pressures exceeding the surrounding atmos-
note no fall in cardiac output in goats and a minimal drop pheric pressure throughout the expiratory cycle.2 Further,
in monkeys. The animals’ clinical distress was attributed to radiographic findings of pneumothorax may be altered in
the precipitous fall in PaO2.75 Subsequent reports in sheep the supine patient, as is often the case for critically ill
found a similar lack of cardiac output compromise with patients on mechanical ventilation. This may be further
induced tension pneumothorax.76 More recent animal compounded by extensive pulmonary parenchymal
investigations in sheep note a drop in cardiac output disease and accompanying alterations in pulmonary com-
accompanied by a fall in PaO2. The abnormalities in gas pliance not allowing ready collapse of the lung upon air
exchange persisted for up to 60–90 minutes after recovery.77 entry into the pleural space. A chest radiographic change
Similarly, Barton78 found a drop in cardiac output and PaO2 found in supine patients wherein pleural air tracks anteri-
in ventilated pigs with hypoxemia developing early and pre- orly and inferiorly creating a deep lateral costophrenic
ceding the development of hypotension. In four sulcus is the ‘deep sulcus sign’.83
documented cases of tension pneumothorax in three
patients on pressure control ventilation, significant Treatment
decreases in cardiac index were observed.79 Similarly,
Connolly80 reported a drop in cardiac index in a mechani-
Urgent treatment of tension pneumothorax is key. As
cally ventilated patient with tension pneumothorax.
noted, mortality may be considerably increased awaiting
chest radiographic confirmation. Treatment provides con-
Clinical presentation and diagnosis firmation of the diagnosis. Prompt drainage by placement
of an intravenous needle (14–16 F) in the pleural space will
usually elicit an audible rush of air due to air rushing out
Patients with tension pneumothorax typically present with
of the pressurized pleural space. Alternatively, the needle
severe distress, displaying cyanosis, diaphoresis, tachycar-
may be fitted with a partially water (sterile) filled syringe
dia, hypotension and labored respirations.2,26 Unusually,
and introduced into the pleural space until air is aspirated.
patients may present quiescently even during mechanical
Marked bubbling after the syringe plunger is removed
ventilation.80,81 The diagnosis should and can be made
confirms pleural air under pressure. Needle placement
clinically in the majority of cases based on the usual dra-
through the second anterior intercostal space has been
matic presentation. Treatment should generally not be
advocated,2 however, placement of the needle through the
delayed for radiographic confirmation, particularly in the
lateral fifth intercostals space may be safer.84 Prompt sub-
mechanically ventilated patient. Such a delay can increase
sequent placement of a chest tube is required regardless of
mortality more than fourfold. Twelve of 74 (16 percent)
whether an audible air rush or water syringe bubbling is
mechanically ventilated patients with pneumothorax died
noted, given the potential for having induced a pneumo-
in Steier and colleagues’ review.26 Seventy-one of these 74
thorax in an already compromised patient by placing a
patients had a tension component. Treatment was delayed
needle into the chest. Placement of a larger-bore chest tube
from 30 minutes to 8 hours in 29 patients awaiting chest
in the setting of a tension pneumothorax in a mechanically
radiographic confirmation. Nine of these 29 (31 percent)
ventilated patient is preferred.
patients died of pneumothorax compared with a 7 percent
mortality rate in the 45 patients with a clinical diagnosis
and immediate treatment.26 A clue to the development of FUTURE DIRECTIONS
a pneumothorax including tension pneumothorax is wors-
ening dynamic and static lung compliance (increasing Additional prospective studies of all aspects of the man-
peak and plateau pressures, respectively) while on volume agement of both spontaneous and non-pneumothorax are
542 Non-spontaneous pneumothorax
●8. Knottenbelt JD, van der Spuy JW. Traumatic pneumothorax: a ●34. Poulose BK, Ray WA, Arbogast PG, et al. Resident work hour limits
scheme for rapid patient turnover. Br J Accident Surg 1990; 21: and patient safety. Ann Surg 2005; 241: 847–56.
77–80. ●35. Rosen AK, Zhao S, Rivard P, et al. Tracking rates of patient safety
◆9. Ball C, Hameed S, Evans D, et al. Occult pneumothorax in the indictors over time. Lessons from the Veterans Administration.
mechanically ventilated trauma patient. Can J Surg 2003; 46: Med Care 2006; 44: 850–61.
373–9. 36. Thomas C, Butler C. Delayed pneumothorax and hydrothorax with
●10. Ball C, Kirkpatrick A, Laupland K, et al. Incidence, risk factors, and central venous migration. Anaesthesia 1999; 54: 987–90.
outcomes for occult pneumothoraces in victims of major trauma. J 37. Laronga C, Meric F, Truong MT, et al. A treatment algorithm for
Trauma 2005; 59: 917–25. pneumothoraces complicating central venous catheter insertion.
11. Guest JL, Anderson JN. Major airway injury in closed chest trauma. Am J Surg 2000; 180: 523–7.
Chest 1977; 72: 63–6, 38. Plaus WJ. Delayed pneumothorax after subclavian vein
12. Carrillo EH, Schmacht DC, Gable DR, et al. Thoracoscopy in the catheterization. J Parenter Enter Nutr 1990; 14: 414–15.
management of postraumatic persistent pneumothorax. J Am Coll 39. Byrd RP, Fields-Ossorio C, Roy TM. Delayed chest radiographs
Surg 1998; 186: 636–40. and the diagnosis of pneumothorax following CT-guided fine
13. Johnson G. Traumatic pneumothorax: is a chest drain always needle aspiration of pulmonary lesions. Respir Med 1999; 93:
necessary? J Accid Emerg Med 1996; 13: 173–4. 379–81.
14. Collins JC, Levine G, Waxman K. Occult traumatic pneumothorax: ●40. Choi C-M, Um S-W, Yoo C-G, et al. Incidence and risk factors of
immediate tube thoracostomy versus expectant management. Am delayed pneumothorax after transthoracic needle biopsy of the
Surg 1992; 58: 743–6. lung. Chest 2004; 126: 1616–21.
15. Brasel KJ, Stafford RE, Weigelt JA, et al. Treatment of occult ●41. Cox JE, Chiles C, McManus CM, et al. Transthoracic needle
pneumothoraces from blunt trauma. J Trauma 1999; 46: 987–91. aspiration biopsy: variables that affect risk of pneumothorax.
16. Garramone RR, Jacobs LM, Sahdev P. An objective method to Radiology 1999; 212: 165–8.
measure and manage occult pneumothorax. Surg Gynecol Obstet ●42. Laurent F, Michel P, Latrabe V, et al. Pneumothoraces and chest
1991; 173: 257–61. tube placement after CT-guided transthoracic lung biopsy using
17. Etoch SW, Bar-Natan MF, Miller FB, et al. Tube thoracostomy. coaxial technique: incidence and risk factors. AJR Am J
Factors related to complications. Arch Surg 1995; 130: 521–6. Roentgenol 1999; 172: 1049–53.
18. Helling TS, Gyles NR, Eisenstein CL, et al. Complications following ●43. Saji H, Nakamura H, Tsuchida T, et al. The incidence and the risk of
blunt and penetrating injuries in 216 victims of chest trauma pneumothorax and chest tube placement after percutaneous CT-
requiring tube thoracostomy. J Trauma 1989; 29: 1367–70. guided lung biopsy. The angle of the needle trajectory is a novel
19. Divisi D, Battaglia C, De Berardis B, et al. Video assisted predictor. Chest 2002; 121: 1521–6.
thoracoscopy in thoracic injury: early or delayed indication? Acta ●44. Yamagami T, Nakamura T, Iida S, et al. Management of
Biomed Ateneo Parmense 2004; 75: 158–63. pneumothorax after percutaneous CT-guided lung biopsy. Chest
20. Golden P. Follow-up chest radiographs after traumatic 2002; 121: 1159–64.
pneumothorax or hemothorax in the outpatient setting: a ●45. Topal U, Ediz B. Transthoracic needle biopsy: factors affecting risk
retrospective review. Int J Trauma Nurs 1999; 5: 88–94. of pneumothorax. Eur J Radiol 2003; 48: 263–7.
21. Misthos P, Kakaris S, Sepsas E, et al. A prospective analysis of ●46. Yeow K-M, Su I-H, Pan K-T, et al. Risk factors of pneumothorax
occult pneumothorax, delayed pneumothorax and delayed and bleeding. Multivariant analysis of 660 CT-guided coaxial
hemothorax after minor blunt thoracic trauma. Eur J Cardiothorac cutting needle lung biopsies. Chest 2004; 126: 748–54.
Surg 2004; 25: 859–64. 47. Bungay H, Berger J, Traill Z, et al. Pneumothorax post CT-guided
●22. Cheatham ML, Safcsak K. Air travel following traumatic lung biopsy: a comparison between detection on chest
pneumothorax: when is it safe? Am Surg 1999; 65: 1160–4. radiographs and CT. Br J Radiol 1999; 72: 1160–3.
∗23. Committee AMAATM. Medical guidelines for air travel. Aviat Space 48. Tanisaro K. Patient positioning after fine needle lung biopsy-effect
Environ Med 1996; 67 (Supplement): B1–16. on pneumothorax rate. Acta Radiologica 2003; 44: 52–5.
24. British Thoracic Society Standards of Care Committee. Managing 49. Masterson AV, Haslam P, Logan PM, et al. Patient positioning after
passengers with respiratory disease planning air travel: British lung biopsy: influence on the incidence of pneumothorax. Can
Thoracic Society recommendations. Thorax 2002; 57: 289–304. Assoc Radiol J 2003; 54: 31–4.
●25. Steier M, Ching N, Bonfils-Roberts E, et al. Iatrogenic causes of 50. Berger R, Smith D. Efficacy of the lateral decubitus position in
pneumothorax: increasing incidence with advances in medical preventing pneumothorax after needle biopsy of the lung. South
care. N Y State J Med 1973; 173: 1296–8. Med J 1988; 81: 1140–3.
26. Steier M, Ching N, Bonfils Roberts E, et al. Pneumothorax 51. Fish G, Stanley J, Miller KS, et al. Postbiopsy pneumothorax:
complicating continuous ventilatory support. J Thoracic estimating the risk by chest radiography and pulmonary function
Caridiovasc Surg 1974; 67: 17–23. tests. AJR Am J Roentgenol 1988; 150: 71–4.
27. Janmeja A, Tandon S, Gupta K, et al. Uncommon iatrogenic 52. Miller K, Fish G, Stanley J, et al. Prediction of pneumothorax rate
pneumothorax. J Assoc Physicians India 1999; 47: 560–1. in percutaneous needle aspiration of the lung. Chest 1988; 93:
28. Weissberg D, Refaely Y. Pneumothorax. Experience with 1,199 742–5.
patients. Chest 2000; 117: 1279–85. 53. Anderson CLV, Crespo JC, Lie Th. Risk of pneumothorax not
●29. Despars JA, Sassoon CSH, Light RW. Significance of iatrogenic increased by obstructive lung disease in percutaneous needle
pneumothoraces. Chest 1994; 105: 1147–50. biopsy. Chest 1994; 105: 1705–8.
30. Sznajder JI, Zveibil FR, Bitterman H, et al. Central vein 54. Poe RH, Kallay MC, Wicks CM, et al. Predicting risk of
catheterization. Failure and complication rates by three pneumothorax in needle biopsy of the lung. Chest 1984; 85:
percutaneous approaches. Arch Intern Med 1986; 146: 259–61. 232–5.
●31. Sassoon CSH, Light RW, O’Hara VS, et al. Iatrogenic 55. Amato M, Barbas C, Medeiros D, et al. Effect of protective-
pneumothorax: etiology and morbidity. Respiration 1992; 59: ventilation strategy on mortality in the acute respiratory distress
215–20. syndrome. N Engl J Med 1998; 338: 347–54.
32. Berger R. Iatrogenic pneumothorax. Chest 1994; 105: 980–2. 56. Network ARDS. Ventilation with lower tidal volumes as
33. Committee on data standards for patient safety. Patient safety. compared with traditional tidal volumes for acute lung injury and
Achieving a new standard for care. Washington, DC: National the acute respiratory distress syndrome. N Engl J Med 2000; 342:
Academies Press, 2004. 1301–8.
544 Non-spontaneous pneumothorax
●57. de Lassence A, Timsit J-F, Tafflet M, et al. Pneumothorax in the conventional chest tube therapy for iatrogenic pneumothorax.
intensive care unit. Incidence, risk factors, and outcome. Chest 1993; 104: 1770–2.
Anesthesiology 2006; 104: 5–13. 71. Barton E, Rhee P, Hutton K. The pathophysiology of tension
∗58. Baumann MH, Strange C, Heffner JE, et al. Management of pneumothorax in ventilated swine. J Emerg Med 1997; 15:
spontaneous pneumothorax. An American College of Chest 147–53.
Physicians Delphi consensus statement. Chest 2001; 119: 72. Ludwig J, Kienzle GD. Pneumothorax in a large autopsy population.
590–602. Am J Clin Pathol 1978; 70: 24–6.
59. Engdahl O, Toft T, Boe J. Chest radiograph – a poor method for 73. Norris RM, Jones JG, Bishop JM. Respiratory gas exchange in
determining size of a pneumothorax. Chest 1993; 103: 26–9. patients with spontaneous pneumothorax. Thorax 1968; 23:
●60. Baumann MH, Patel PB, Roney CW, et al. Comparison of function 427–33.
of commercially available pleural drainage units and catheters. 74. Moran JF, Jones RH, Wolfe WG. Regional pulmonary function
Chest 2003; 123: 1878–86. during experimental unilateral pneumothorax in the awake state. J
61. Roggla M, Wagner A, Brunner C, et al. The management of Thorac Cardiovasc Surg 1977; 74: 396–402.
pneumothorax with the thoracic vent versus conventional 75. Rutherford RB, Hurt HH, Brickman RD, et al. The pathophysiology
intercostal tube drainage. Wien Klin Wochenschr 1996; 108: of progressive, tension pneumothorax. J Trauma 1968; 8: 212–27.
330–3. 76. Gustman P, Yerger L, Wanner A. Immediate cardiovascular effects
62. Mainini SE, Johnson FE. Tension pneumothorax complicating of tension pneumothorax. Am Rev Respir Dis 1983; 127: 171–4.
small-caliber chest tube insertion. Chest 1990; 97: 759–60. 77. Carvalho P, Hilderbrandt J, Charan NB. Changes in bronchial and
63. Crocker HL, Ruffin RE. Patient-induced complications of a pulmonary arterial blood flow with progressive tension
Heimlich flutter valve. Chest 1998; 113: 838–9. pneumothorax. J Appl Physiol 1996; 81: 1664–9.
64. Delius RE, Obeid F, Horst M, et al. Catheter aspiration for simple ◆78. Barton ED. Tension pneumothorax. Curr Opin Pulm Med 1999; 5:
pneumothorax. Arch Surg 1989; 124: 833–6. 269–74.
65. Brown KT, Brody LA, Getrajdman GI, et al. Outpatient treatment of 79. Beards S, Lipman J. Decreased cardiac index as an indicator of
iatrogenic pneumothorax after needle biopsy. Radiology 1997; tension pneumothorax in the ventilated patient. Anaesthesia
205: 249–52. 1994; 49: 137–41.
66. Baumann MH, Sahn SA. Medical management and therapy of 80. Connolly J. Hemodynamic measurements during tension
bronchopleural fistulas in the mechanically ventilated patient. pneumothorax. Crit Care Med 1993; 21: 294–6.
Chest 1990; 97: 721–8. 81. Baumann MH, Sahn SA. Tension pneumothorax: diagnostic and
◆67. Lois M, Noppen M. Bronchopleural fistuals. An overview of the therapeutic pitfalls. Crit Care Med 1993; 21: 177–9.
problem with special focus on endoscopic management. Chest 82. Pilbeam SP. Basic patient assessment and methods to improve
2005; 128: 3955–65. ventilation. Mechanical ventilation. Physiological and clinical
68. Baumann MH, Strange C. Treatment of spontaneous applications. St Louis: Mosby, 1998; 244–61.
pneumothorax. A more aggressive approach? Chest 1997; 112: ●83. Gordon R. The deep sulcus sign. Radiology 1980; 136: 25–7.
789–804. 84. Rawlins R, Brown KM, Carr CS, et al. Life threatening haemorrhage
69. Baumann MH, Strange C. The clinician’s perspective on after anterior needle aspiration of pneumothoraces. A role for
pneumothorax management. Chest 1997; 112: 822–8. lateral needle aspiration in emergency decompression of
●70. Schoenenberger R, Haefeli W, Weiss P, et al. Evaluation of spontaneous pneumothorax. Emerg Med J 2003; 20: 383–4.
44
Pediatric pleural diseases
may be indicated, but in many cases a fetal thoracentesis is atrics, thoracentesis is frequently performed and is defi-
needed to distinguish primary from secondary effusions. nitely needed for diagnostic studies of acute infectious
The underlying condition is the most important factor in effusion, malignancies and chylothorax. Pediatric parap-
determining overall outcome. neumonic effusions are sometimes detected when the
Pleural effusions detected in utero may persist into the patient is in the recovery phase of an infection and the
newborn period. Significant accumulations will cause res- effusion is not causing any symptoms. In such patients,
piratory distress at birth, both by limiting lung expansion diagnostic thoracentesis is often not necessary, but close
and by causing underlying pulmonary hypoplasia. At the follow-up is indicated to determine that there has been
time of delivery, personnel should be present to provide resolution of the effusion.
resuscitation and drainage to relieve respiratory distress.
Additional therapies may be indicated if an underlying dis-
order, such as congenital heart disease, was also detected in
utero. CHYLOTHORAX
invasive approach for ligation of the thoracic duct, but in has been frustrating and conventional therapy with chest
the very small infant the technology may be limited, par- tube drainage, dietary interventions or pleurodesis have
ticularly if there has been prior thoracic surgery. variable success.24 Aggressive attention to fluid manage-
ment, oxygen therapy and other medical therapies has
been reported to decrease the duration of the effusions.25
Pulmonary lymphangiectasis Octreotide has been reported to have benefit for the chy-
lothorax following congenital heart surgery,26 however, no
Pulmonary lymphangiectasis is associated with abnormal prospective randomized controlled trials have been carried
pulmonary lymphatics in the pleura and interstitium of out.
the lung.20 The overall incidence is unknown and scattered
cases continue to be reported, but pulmonary lymphang-
iectasis is very rare. Esther and Barker21 proposed catego- Malignancies
rizing pulmonary lymphangiectasis as primary or
secondary (which is related to pulmonary venous or lym-
Although a small percentage of pediatric pleural effusions
phatic obstruction). Diagnosis may be suspected by the
are associated with malignancies, it is a very important cat-
presence of increased interstitial markings that are appar-
egory. Malignancy must be considered in a child who pres-
ent on chest radiograph secondary to the dilated lymphat-
ents with an effusion, without symptoms or history
ics throughout the pleural and connective tissue. Open
suggesting infection or trauma. Lymphomas are common
lung biopsy is usually required for definitive diagnosis.
childhood cancers and frequently present with a mediasti-
Patients with pulmonary lymphangiectasis who present
nal mass and pleural effusion.27,28 One report found 71
with respiratory distress in the newborn period had previ-
percent of patients with lymphoblastic lymphoma had a
ously been thought to have a very poor prognosis, but with
pleural effusion at presentation while only 11 percent of
aggressive supportive care pleural effusions seem to
patients with Hodgkin’s disease presented with effusions.27
subside, and infants are surviving beyond the neonatal
A large effusion may result in respiratory compromise,
period.
necessitating drainage, but pleural fluid analysis can also
provide a definitive diagnosis in most cases.29 Pleural effu-
sions have been seen with other intrathoracic malignancies
including leukemia, neuroblastoma, Wilm’s tumor, sarco-
PEDIATRIC EFFUSIONS mas and mesothelioma (extremely rare in pediatrics).30,31
Pleurodesis is used in adult oncology patients but there
Pleural effusions are uncommon findings on pediatric
is little information about its usefulness in pediatric
radiographs. Beyond the newborn period, a newly detected
patients. In a report of seven end-stage pediatric oncology
pleural effusion is most likely secondary to infection – a
patients with intractable pleural effusions, pleurodesis
parapneumonic effusion. The true incidence of pleural
with doxycycline provided significant relief of respiratory
effusions in the pediatric population is not known because
symptoms.32
most studies are retrospective and based on hospitalized
children. Alkrinawi and Chernick22,23 reviewed charts of
127 pediatric patients (<18 years of age) admitted with
pleural effusions from 1987 to 1995. Fifty percent of the Infections
patients had pneumonia with a parapneumonic effusion.
Other etiologies included congenital heart disease (17 Tuberculous pleural effusions occur in children, but less
percent – mostly post-operative, 2 percent congestive frequently than in adults. A review of 175 children <18
heart failure), malignancy (10 percent), renal (9 percent) years of age with primary pulmonary tuberculosis (TB)
and trauma (7 percent). As in other studies, pleural fluid revealed that 22 percent had an effusion identified on the
was more than twice as likely to occur in boys. chest radiograph.33 The average age of the children with
effusions was significantly greater – 13.5 years versus 7.0
years – than the age of patients with parenchymal TB. The
Congenital heart disease effusion was unilateral and was the only radiographic
finding in 41 percent. TB should be considered in cases of
In pediatrics, congestive heart failure is not common, pleural effusions where other etiologies are not clearly
although it can occur with congenital heart disease. One of identified. TB workup would include a thorough history
the challenging pediatric pleural effusions is that associ- for contacts and skin testing. In older children workup
ated with single ventricle congenital heart disease and would be similar to adults, including assessment of
post-operatively with the Fontan procedure. The precise sputum and pleural fluid. In young children, induced
etiology of the post-Fontan pleural effusion is not clear, sputum or gastric aspirate may be needed.
and is likely multifactorial – including fluids, pressures, In pediatric patients infections with Mycoplasma pneu-
arrhythmias and inflammatory cytokines. Management moniae and viruses are very common. Pleural effusions
548 Pediatric pleural diseases
may occur with M. pneumoniae pneumonia, but the only a few reports available, their role in pediatrics is not
effusions are generally small and do not cause respiratory yet clear.35
symptoms.34 The incidence of pleural effusions in associa- Management of parapneumonic effusions and empye-
tion with viral pneumonia is not known, since most mas in children is controversial. Sonographic evaluation
patients with viral respiratory disease are seen as out- is carried out to define whether loculations are present
patients and chest radiographs are not routinely obtained. and provide guidance for the thoracentesis. Intravenous
There are anecdotal reports of the discovery of large para- antibiotics should be started, with coverage for the
pneumonic effusions on chest radiographs taken after res- common organisms S. aureus and S. pneumoniae, with
olution of an acute respiratory illness. It is presumed that consideration given to the pattern of antibiotic resistance
these effusions are the result of a viral pneumonia, but in the community. Therapy can be adjusted once an
since the patients are asymptomatic, no specific interven- organism has been identified from either pleural fluid or
tion is performed. Viral infections can be associated with blood culture. After adequate antibiotic coverage, there is
large effusions which result in respiratory distress and tho- significant debate about additional therapies of pleural
racentesis may be needed both for diagnostic and thera- infections, in particular about surgical intervention with
peutic reasons. chest tube drainage, fibrinolytic therapy and VATS. All
Bacterial pneumonia is the most common cause of these therapies have been reported to be successful.
pleural effusions in children and there are some reports Most clinicians agree that if the fluid accumulation is
which suggest that the incidence of empyema in children is large enough to cause significant respiratory symptoms,
increasing.35 Historically, three specific organisms have then the fluid should be drained with a chest tube;
been associated with pleural effusions: Staphylococcus many would suggest that chest tubes are needed for all
aureus, Haemophilus influenzae and Streptococcus pneumo- empyemas.
niae. Over the past decade, universal vaccination against Studies of intrapleural fibrinolytic therapy in adults do
H. influenzae and S. pneumoniae has reduced the impor- not support its use for the treatment of pleural infec-
tance of these pathogens, although in most reports S. tion.40,41 In contrast, a randomized trial of intrapleural
pneumoniae is still the most common pathogen. A recent urokinase in children was associated with shorter hospital
retrospective review of empyema suggested that stay.4 A prospective study comparing urokinase and VATS
Staphylococcus aureus, particularly methicillin-resistant found no difference in outcomes, and recommended
S. aureus, was becoming more prevalent.36 However, in urokinase as the treatment option.42 A retrospective report
that report, causative organisms were detected in only 43 of primary VATS versus non-operative therapy43 and a
percent of the cases, perhaps related to previous antibiotic small prospective study of VATS44 suggested that primary
therapy. Anaerobic organisms are rarely isolated in chil- VATS therapy was superior. There is a paucity of data on
dren, but infrequent use of anaerobic culture media may simple medical therapy. In a report of 14 patients treated
have reduced the detection of anaerobes. with intravenous antibiotics and simple chest tube drain,
The emergence of antibiotic resistant organisms, all patients completely recovered without aggressive surgi-
including methicillin-resistant S. aureus and S. pneumo- cal intervention.45 Routine chest tubes for all empyemas in
niae, may be associated with more complicated effusions children was challenged in study of 65 patients.46 In the
and empyema. However, overall, patients still respond well conservatively managed group, the patients received a
if treated with appropriate antibiotics.37,38 The increase in chest tube only if there was significant mediastinal shift,
antibiotic resistant infections secondary to S. aureus and respiratory distress or uncontrolled infection. These
S. pneumoniae underscores the importance of a diagnostic patients did just as well as the patients who had chest tubes
tap in children who present with a pleural effusion. placed for large pleural effusions and/or fibrinopurulent
Identification of an organism with accompanying drug effusion.
susceptibilities allows the clinician to tailor antibiotic All agree that appropriate antibiotic therapy is the
therapy and may prevent the need for more invasive pro- mainstay of therapy. While the controversy continues
cedures. Although the prior use of antibiotics may limit regarding other aspects of care, the good news is that
the ability to recover an organism from culture, even its children do well with all the currently proposed
presence on Gram stain may help guide therapy. The use therapies.35 There are several factors that likely contribute
of polymerase chain reaction (PCR) and pneumococcal to the conflicting results. Many studies are small and ret-
antigen detection in pleural fluid specimens can be helpful rospective. There is no consistent definition of the
for more rapid diagnosis of S. pneumoniae when culture patients in the studies. Many patients are referred after
results are negative.39 failure of primary antibiotic therapy. Not all studies uti-
As additional PCR-based detection studies become lized similar diagnostic studies, including analyses of
available they will improve the ability to rapidly detect pleural fluid. Studies are not consistent in the outcome
other organisms. Analyses of protein, lactate dehydroge- measures, should it be length of stay or pulmonary out-
nase (LDH), glucose and pH play important roles in the comes. In pediatrics, one of the goals of therapy is not to
management of pleural effusions in adult. These studies just maintain lung function, but also not to interfere with
are not always routinely performed in pediatrics, and with future lung growth.
References 549
There is little information about the long term conse- ●1. Montgomery MSD. Air and liquid in the pleural space. In: Chernick
quences of empyema in children. A small studied looked at VBT, Wilmott RW, Bush A (eds). Kendig’s disorders of the
respiratory tract in children, 7th edn. Phildelphia: Saunders, 2006:
the results of spirometry and exercise testing in children 368–87.
following recovery from empyema.47 No differences were 2. Bush A. Update in pediatric lung disease 2006. Am J Respir Crit
found in children who had chest tube drainage and those Care Med 2007; 175: 532–40.
who only received antibiotics. Another report measured 3. Maskell NA, Davies CW, Nunn AJ, et al. U.K. Controlled trial of
lung function 3–24 months after discharge and found no intrapleural streptokinase for pleural infection. N Engl J Med; 352:
865–74.
evidence of restrictive lung disease in children with anti- ●4. Thomson AH, Hull J, Kumar MR, Wallis C, Balfour Lynn IM.
biotics and simple chest tube drainage.45 Yet another study Randomised trial of intrapleural urokinase in the treatment of
found a restrictive defect in patients at 3 months after childhood empyema. Thorax 2002; 57: 343–7.
treatment for empyema but most had normalized in ∗5. Balfour-Lynn IM, Abrahamson E, Cohen G, et al. BTS guidelines for
1 year.48 the management of pleural infection in children. Thorax; 60 (Suppl
1): i1–21.
6. Longaker MT, Laberge JM, Dansereau J, et al. Primary fetal
hydrothorax: natural history and management. J Pediatr Surg
FUTURE THERAPIES 1989; 24: 573–6.
7. Bianchi D. Hydrothorax. In: Bianchi DW, Crombleholm TM, D’Alton
Improved diagnostic testing is needed to better character- ME (eds). Fetology: Diagnosis and management of the fetal
patient. New York: McGraw-Hill, 2000; 313–21.
ize pediatric pleural effusions. A better characterization of 8. Waller K, Chaithongwongwatthana S, Yamasmit W, Donnenfeld
patients, and the nature of their effusions, will result in AE. Chromosomal abnormalities among 246 fetuses with pleural
improved understanding of the natural history of the effu- effusions detected on prenatal ultrasound examination: factors
sions and optimal management. associated with an increased risk of aneuploidy. Genet Med 2005;
7: 417–21.
9. Bellini C, Boccardo F, Campisi C, Bonioli E. Congenital pulmonary
lymphangiectasia. Orphanet J Rare Dis 2006; 1: 43.
SUMMARY 10. Jacquemont S, Barbarot S, Boceno M, Stalder JF, David A. Familial
congenital pulmonary lymphangectasia, non-immune hydrops
The incidence of pleural disease and effusions is uncom- fetalis, facial and lower limb lymphedema: confirmation of
mon in children when compared with that in adults and, Njolstad’s report. Am J Med Genet 2000; 93: 264–8.
11. Klam S, Bigras JL, Hudon L. Predicting outcome in primary fetal
in general, pediatric outcomes are better than that in hydrothorax. Fetal Diagn Ther 2005; 20: 366–70.
adults which may reflect less severe co-morbidities. The 12. Smith RP, Illanes S, Denbow ML, Soothill PW. Outcome of fetal
underlying etiologies in the fetus and young children are pleural effusions treated by thoracoamniotic shunting. Ultrasound
frequently related to other congenital conditions. In the Obstet Gynecol 2005; 26: 63–6.
neonate, chylothorax is the most common pleural effu- 13. Kalfa N, Allal H, Lopez M, et al. Thoracoscopy in pediatric pleural
empyema: a prospective study of prognostic factors. J Pediatr Surg
sion. As children progress from infancy to childhood, 2006; 41: 1732–7.
parapneumonic effusions become more frequent. 14. Kearney SE, Davies CWH, Davies RJO, Gleeson FV. Computed
Pediatric-specific diagnostic and therapeutic approaches tomography and ultrasound in parapneumonic effusions and
need to be developed. empyema. Clin Radiol 2000; 55: 542–7.
15. Heller RM, Hernanz-Schulman M. Applications of new imaging
modalities to the evaluation of common pediatric conditions. J
Pediatr 1999; 135: 632–9.
KEY POINTS 16. Rocha G, Fernandes P, Rocha P, et al. Pleural effusions in the
neonate. Acta Paediatr 2006; 95: 791–8.
17. Beghetti M, La Scala G, Belli D, et al. Etiology and management of
● Pleural effusions in the fetus are frequently pediatric chylothorax. J Pediatr 2000; 136: 653–8.
related to underlying congenital anomalies. 18. Geismar SL, Tilelli JA, Campbell JB, Chiaro JJ. Chylothorax as a
● Chylothorax is the most common pleural effu- manifestation of child abuse. Pediatr Emerg Care 1997; 13: 386–9.
sion in newborns. 19. Siu SL, Lam DS. Spontaneous neonatal chylothorax treated with
octreotide. J Paediatr Child Health 2006; 42: 65–7.
● Parapneumonic effusions are the most common 20. Noonan JA, Walters LR, Reeves JT. Congenital pulmonary
pleural effusions in children. lymphangiectasis. Am J Dis Child 1970; 120: 314–19.
● Management of parapneuomic effusions in chil- 21. Esther CR Jr, Barker PM. Pulmonary lymphangiectasia: diagnosis
dren is controversial. and clinical course. Pediatr Pulmonol 2004; 38: 308–13.
22. Alkrinawi S, Chernick V. Pleural fluid in hospitalized pediatric
patients. Clin Pediatr (Phil) 1996; 35: 5–9.
23. Alkrinawi S, Chernick V. Pleural infection in children. Semin Respir
REFERENCES Infect 1996; 11: 148–54.
24. Kiziltepe U, Eyileten ZB, Uysalel A, Akalin H. Prolonged pleural
effusion following Fontan operation: effective pleurodesis with
● = Key primary paper talc slurry. Int J Cardiol 2002; 85: 297–9.
◆ = Major review article 25. Cava JR, Bevandic SM, Steltzer MM, Tweddell JS. A medical
∗ = Paper that represents the first formal publication of a strategy to reduce persistent chest tube drainage after the fontan
management guideline operation. Am J Cardiol 2005; 96: 130–3.
550 Pediatric pleural diseases
26. Chan SY, Lau W, Wong WH, et al. Chylothorax in children after outcome of children with pneumonia attributable to penicillin-
congenital heart surgery. Ann Thorac Surg 2006; 82: 1650–6. susceptible and penicillin-nonsusceptible Streptococcus
27. Chaignaud BE, Bonsack TA, Kozakewich HP, Shamberger RC. pneumoniae. Pediatrics 1998; 102: 1369–75.
Pleural effusions in lymphoblastic lymphoma: a diagnostic 39. Le Monnier A, Carbonnelle E, Zahar JR, et al. Microbiological
alternative. J Pediatr Surg 1998; 33: 1355–7. diagnosis of empyema in children: comparative evaluations by
28. Pietsch JB, Whitlock JA, Ford C, Kinney MC. Management of culture, polymerase chain reaction, and pneumococcal antigen
pleural effusions in children with malignant lymphoma. J Pediatr detection in pleural fluids. Clin Infect Dis 2006; 42: 1135–40.
Surg 1999; 34: 635–8. 40. Maskell N, Nunn A, Davies RJ. Intrapleural streptokinase for
29. Wong JW, Pitlik D, Abdul-Karim FW. Cytology of pleural, pleural infection. Br Med J 2006; 332: 552.
peritoneal and pericardial fluids in children. A 40-year summary. 41. Cameron R, Davies HR. Intra-pleural fibrinolytic therapy versus
Acta Cytol 1997; 41: 467–73. conservative management in the treatment of parapneumonic
30. Easa D, Balaraman V, Ash K, Thompson B, Boychuk R. Congenital effusions and empyema. Cochrane Database Syst Rev 2004; 2:
chylothorax and mediastinal neuroblastoma. J Pediatr Surg 1991; CD002312.
26: 96–8. 42. Sonnappa S, Cohen G, Owens CM, et al. Comparison of urokinase
31. Goyal M, Swanson KF, Konez O, Patel D, Vyas PK. Malignant and video-assisted thoracoscopic surgery for treatment of
pleural mesothelioma in a 13-year-old girl. Pediatr Radiol 2000; childhood empyema. Am J Respir Crit Care Med 2006; 174: 221–7.
30: 776–8. 43. Avansino JR, Goldman B, Sawin RS, Flum DR. Primary operative
32. Hoffer FA, Hancock ML, Hinds PS, et al. Pleurodesis for effusions in versus nonoperative therapy for pediatric empyema: a meta-
pediatric oncology patients at end of life. Pediatr Radiol 2007; 37: analysis. Pediatrics 2005; 115: 1652–9.
269–73. 44. Kurt BA, Winterhalter KM, Connors RH, Betz BW, Winters JW.
33. Merino JM, Carpintero I, Alvarez T, et al. Tuberculous pleural Therapy of parapneumonic effusions in children: video-assisted
effusion in children. Chest 1999; 115: 26–30. thoracoscopic surgery versus conventional thoracostomy drainage.
34. Ward MA. Lower respiratory tract infections in adolescents. Pediatrics 2006; 118: e547–53.
Adolesc Med 2000; 11: 251–62. 45. Satish B, Bunker M, Seddon P. Management of thoracic empyema
●35. Jaffe A, Balfour-Lynn IM. Management of empyema in children. in childhood: does the pleural thickening matter? Arch Dis Child
Pediatr Pulmonol 2005; 40: 148–56. 2003; 88: 918–21.
36. Schultz KD, Fan LL, Pinsky J, et al. The changing face of pleural 46. Epaud R, Aubertin G, Larroquet M, et al. Conservative use of chest-
empyemas in children: epidemiology and management. Pediatrics tube insertion in children with pleural effusion. Pediatr Surg Int
2004; 113: 1735–40. 2006; 22: 357–62.
37. Hardie WD, Roberts NE, Reising SF, Christie CD. Complicated 47. Redding GJ, Walund L, Walund D, et al. Lung function in children
parapneumonic effusions in children caused by penicillin- following empyema. Am J Dis Child 1990; 144: 1337–42.
nonsusceptible Streptococcus pneumoniae. Pediatrics 1998; 101: 48. Kohn GL, Walston C, Feldstein J, et al. Persistent abnormal lung
388–92. function after childhood empyema. Am J Respir Med 2002; 1:
38. Tan TQ, Mason EO Jr, Barson WJ, et al. Clinical characteristics and 441–5.
45
Drainage techniques
HENRI COLT
Thoracentesis instrumentation
General techniques
space above the initial insertion site. In general, it is almost index finger while the proximal end of the needle is braced
always safe to use a small fine needle of 22 or 25 gauge. against the palm of the hand. The needle is then advanced
through the subcutaneous tissue above the rib below and
up to the parietal pleura.16 The sharp trochar is replaced by
Examples of specific techniques the blunt trochar just before traversing the parietal pleural.
The advantages of this needle are less risk of traumatic
BEDSIDE THORACENTESIS IN CRITICALLY ILL PATIENTS USING puncture of the lung, simple use of accessory suction
‘THE ROLLED BED SHEET TECHNIQUE’ systems or manometers and the relatively large bore of the
needle, which facilitates removal of large amounts of fluid
The rolled bed sheet technique allows for the placement of as well as fibrin, blood or pus. Because the needle is
the seated patient into an optimum position for thoracen- reusable, there is an initial non-recurring cost for the
tesis or for ultrasound guided thoracentesis.15 I have found instrument, but the cost of each subsequent use is nil.
this technique to be particularly helpful because the risks Disadvantages include possible air entry at the time of the
of patient positioning, which in the intensive care unit may exchange between sharp and blunt pleural trochars and
be associated with accidental extubation, loss of central the potential laceration of an intercostal vessel because the
venous access or deterioration of hemodynamic status, are sharp trochar itself has a cutting, pyramidal tip design.
diminished. A bedsheet is rolled lengthwise and then
placed gently behind the patient’s back. The head of the THORACENTESIS USING THE ARGYLE–TURKEL SAFETY
bed is raised, as the bedsheet is easily slid down behind the THORACENTESIS SET
patient’s back to the level of the scapula before lowering
the head of the bed again. While each end of the bedsheet The system consists of a blunt multisided spring-loaded
is held by an assistant and pulled forward toward the foot inner cannula coaxilly housed within a 16 gauge conven-
of the bed, the operator has room to work behind the tional sharp beveled hollow needle. While the needle is
patient, who is kept in the sitting position throughout the advanced through the skin and intercostal tissues, the
procedure (Figure 45.6). blunt cannula is forced into the shaft of the needle. When
the tip of the needle encounters low resistance, such as
THE REUSABLE BOUTIN PLEURAL PUNCTURE NEEDLE pleural effusion or loss of tissue resistance, the spring-
loaded cannula automatically extends beyond the bevel,
The Boutin pleural needle manufactured by the Richard protecting underlying tissue from further penetration.
Wolf Company (Vernon Hills, IL, USA) is a stainless steel Although the procedure requires some feel, a great advan-
reusable needle designed for performing thoracentesis. It is tage for some operators is a green indicator on the needle
a large-bore 2.8 mm diameter three-part instrument with housing which identifies the position of the blunt cannula.
an 80 mm working length. It is comprised of an outer If resistance is being felt such that the sharp tip of the
cannula with a open–shut tab at one end, a blunt trochar needle is exposed, the indicator is red. When resistance is
and a sharp inner trochar (Figure 45.7). After making a lost, the indicator turns green. Therefore, once the needle
small stab incision with a #11 scalpel, the Boutin needle is is in pleural space, the indicator turns green. A one-way
assembled with the sharp trochar inside the outer cannula. valve prevents any air entry after the needle is removed
The shaft of the needle is held between the thumb and (Figure 45.4).
Figure 45.6 Bedside ultrasound-guided thoracentesis using the Figure 45.7 Boutin pleural puncture needle (Richard Wolf Corp,
‘rolled bedsheet’ technique. Bloomington, IN, USA).
556 Drainage techniques
THORACENTESIS USING THE BARD MILROSE WANG AND Vasovagal reactions are characterized by decreased
GARG THORACENTESIS NEEDLES AND KITS stroke volume, fallen cardiac output, bradycardia and
hypotension. Anxiety, pain, sight of blood and apprehen-
Each of these needles is 17 gauges, with a catheter sheath sion are promoting factors. When the reaction is noted,
diameter of 2.3 mm and length of 8 cm (Figure 45.3). the procedure should be terminated and the patient
Needles are available individually or in a procedural kit should be placed in the reverse Trendelenburg position
intended for single use. A small side port catheter is while vital signs are assessed. Recuperation usually occurs
attached to the Wang needle sheath to facilitate fluid within 15 minutes.
removal into Vacutainer bottles. The sheath septum allows Hemothorax is an extremely infrequent complication,
for multiple insertions of the needle without passage of air. but can result from laceration of an intercostal vein or
Rather than a small-attached tubing, this Garg needle artery. Hemothorax is probably more frequent after closed
comes with a stopcock feature, which prevents passage of needle pleural biopsy than after thoracentesis but, regard-
air and also permits multiple insertions of the needle into less of etiology, may require thoracoscopic exploration and
the sheath. even open thoracotomy for additional diagnosis and man-
agement. A hemothorax should be suspected when a
follow-up chest radiograph reveals an immediate reaccu-
mulation of pleural fluid. Hemothorax should also be sus-
Complications from thoracentesis pected if bright red blood suddenly appears within the
syringe during fluid removal.
Complications from thoracentesis are the same as those In patients with fever, new pleural effusions or symp-
from closed needle pleural biopsy or even chest tube toms suggestive of lung or pleural infection, and who have
drainage. The most common complication is a pneumo- had a recent history of thoracentesis, bacterial contamina-
thorax.17–22 It appears, however, that the incidence of tion of the pleural space should be suspected. In these
pneumothorax is reduced if experienced individuals cases, I suggest repeating the thoracentesis in order to
perform the procedure. Iatrogenic pneumothorax is also identify a new etiology for the pleural effusion. Other
decreased when pleural ultrasonography is used for guid- potential complications from thoracentesis include liver or
ance. In general, there are three reasons why a pneumo- spleen laceration. Soft tissue infection is also a possibility.
thorax may occur following thoracentesis. The first, and Potential seeding of the needle track with tumor cells has
potentially most consequential, is that of an accidental lac- been rarely reported in patients undergoing percutaneous
eration of the lung parenchyma. This is easily recognized lung biopsy, but is a feared complication of pleural proce-
at the time of thoracentesis by air bubbles coming up into dures in patients with malignant mesothelioma.
the pleural fluid. A second reason for air in the pleural Comments are warranted regarding the risk for hepati-
space is that air was accidentally entered the pleural space tis or Human immunodeficiency virus (HIV) infection due
through the catheter and needle at the time of thoracente- to operator injury. Needles should never be recapped
sis. This rarely results in a large pneumothorax and air can during a procedure. Neither should needles should be
be readily removed by attaching the catheter to a suction placed onto gauze pads or back into the procedure tray
device or syringe. If left in the hemithorax, air will sponta- because one might inadvertently pick up the gauze pad and
neously be absorbed at a rate of approximately 1 percent thus stick oneself. When discarding sharps, they should be
per day. A third reason is the presence of a trapped lung.23 grasped using a hemostat, rather than one’s hands.
This might occur in patients with chronic pleural effu- In view of increasing concerns regarding patient safety,
sions, empyema or pleural thickening, or malignancy and a few comments are warranted pertaining to the use of
malignant mesothelioma. A thick pleural peel is present pleural ultrasonography for thoracentesis. This technology
over the parietal and visceral pleura. The lung is not fully is frequently used in training programs, even for many
expanded, so after fluid is removed, the chest radiograph routine thoracenteses, and helps decrease the frequency of
reveals a pneumothorax. thoracentesis-related complications, improve physical
Other complications from thoracentesis include vaso- diagnostic skills and examine the pleural space for multi-
vagal reactions, cough, chest pain and hemothorax. Cough loculations that might impact management decisions, par-
most frequently complicates a thoracentesis when large ticularly for malignant effusions being considered for
amounts of pleural fluid are removed. If a patient begins to pleurodesis, and parapneumonic effusions being consid-
cough excessively, the procedure should be stopped. Chest ered for drainage. However, ultrasound units are not
pain is infrequent, but might be related to anesthetic tech- always readily accessible from radiology departments and
nique, needle insertion or evacuation of pleural fluid. the cost of a portable ultrasound machine is currently
Sometimes, chest discomfort can be decreased by decreas- more than $10 000. Certainly, if pleural ultrasonography is
ing the rapidity with which pleural fluid is withdrawn, or not needed routinely, patients can at the least be risk-
by removing pleural fluid using manual syringe suction stratified so that high-risk blind thoracenteses are avoided,
techniques rather than a high negative-pressure and ultrasound can be desirable when effusions are noted
Vacutainer bottle. to be small (on chest radiograph or computed tomography
Closed needle pleural biopsy 557
scan), multiloculated or in patients with poor lung func- inserting the biopsy needle, abundant local anesthesia
tion or significant bullous disease. should be administered to the skin, intercostal tissues and
parietal pleura. A wider and larger field of local anesthesia
is warranted than for simple thoracentesis because biopsies
might be obtained from several quadrants surrounding the
CLOSED NEEDLE PLEURAL BIOPSY initial needle insertion site, all through the initial needle
tract. I often administer a narcotic intramuscularly or
Definition, indications and contraindications intravenously as well.31–34
Insertion techniques
pleural biopsy needle samples the liver, spleen or kidney.
One should therefore be particularly careful to never INSERTING SMALL-BORE CHEST TUBES
perform closed needle pleural biopsy too low.
A comment is warranted pertaining to the apparent Small-bore chest tubes (less than 20 French) are usually
decline in the number of closed-needle pleural biopsies placed to evacuate small or loculated pleural effusions and
being performed in Europe, Asia and the USA. to evacuate air in case of iatrogenic pneumothorax.51–53
Thoracoscopy and flex-rigid pleuroscopy37,38 are excellent This might also be necessary when pneumothorax occurs
alternatives to blind sampling, and provide the added after a transbronchial lung biopsy. Small-caliber tubes are
benefit of removing all pleural fluid, evaluating the visceral inserted using catheter systems such as the catheter
and parietal pleural surfaces, breaking down adhesions to over/through needle technique and the catheter over guide
enhance lung expansion, and obtain forceps biopsies. In wire (Seldinger) technique (Figure 45.12). When using the
addition, increasingly, measurements of adenosine deami- catheter over/through needle technique, procedures are
nase assist in making a diagnosis of tuberculous pleurisy, begun just as if performing a thoracentesis. With the
and increased use of computed tomography helps with dif- needle attached to a syringe, the needle is pointed towards
ferential diagnosis of malignant and nonmalignant effu- the intrapleural location of interest. A catheter sheath is
sions. Only time will tell whether closed needle biopsy usually inserted over or through the needle while holding
becomes an antiquated technique.39 an introducer. In case of pneumothorax, these catheters
are directed towards the apex, whereas in case of pleural
effusions, catheters are directed inferiorly and posteriorly
CHEST TUBES into the costal diaphragmatic recess. The catheter is con-
nected to either a bag or negative pressure suction system
Definition, indications and contraindications and is tied to the skin using non-absorbable sutures.
When using the Seldinger technique, an 18-gauge
Chest tube drainage consists of a percutaneous insertion of needle is usually advanced into the pleural space attached
a small- or large-bore tube, usually made of silicone or to a syringe. The syringe is disconnected while pointing the
polyurethane, into the pleural cavity.40 This procedure is needle in the desired direction and a guide wire is passed
warranted in certain patients with pleural and pulmonary through the needle into the pleural space. After removal, a
560 Drainage techniques
(a)
(a) (b)
(b) (c)
(c)
(d)
(d) (e)
(e)
Figure 45.12 Catheter over guidewire (Seldinger technique) of a small-bore chest tube insertion. (a) The pleural space is accessed. (b)
The guidewire is threaded through the needle. (c) The dilator is threaded over the guidewire, then removed. (d) The chest tube is threaded
over the guidewire. (e) The guidewire is removed, leaving the chest tube in position. The chest tube will be secured using a simple stitch
tied down onto the chest tube using a surgeon’s knot. (From Colt HG, Manual of pleural procedures, Lippincott, Williams and Wilkins, 1999,
Figure 12.5, with permission.)
Figure 45.16 Transparent dressings allow rapid inspection of Figure 45.17 Home drainage apparatus (disposable suction
dressing sites. bottle) by Denver Biomedical, Boulder CO, USA, for use with
indwelling pleural catheters (hospital or home aspiration and
pleurodesis).
and another bottle connected serially to it, serving as a COMMERCIALLY AVAILABLE DRAINAGE SYSTEMS
water seal chamber. Fluid accumulates in this first bottle,
and air passes through the first bottle into a shorter cannula These systems consist of disposable molded plastic units
and into a second bottle, which is the water seal chamber. with three chambers duplicating the classic three-bottle
A suction pump such as an Emerson pump (which is a reg- system (Figure 45.20). These initially used water columns
ulated negative pressure device) can be attached to the in order to regulate the amount of negative pressure
bottles. The amount of negative pressure applied during exerted upon the pleural cavity. More recently they use dry
aspiration can be most readily controlled. However, a third suction units in which a valve replaces the water column.
suction control bottle can be added to the two-bottle The advantage of these ‘dry’ units is their silence, and rel-
system. In this case, a vent (cannula) in the suction control ative absence of spillage of fluid if tipped over. In addition,
bottle is connected to a vent in the water seal bottle. When pleural fluid can be removed from the system for analysis,
suction is applied, air enters the bottle through the rigid and patient ambulation is facilitated (Figure 45.21).
cannula if the pressure in the bottle is more negative than It is noteworthy that in both dry and water column
the depth of the cannula below the fluid. The three-bottle systems, the water seal chamber always contains approxi-
system was used for many years, but has the inconveniences mately 2 cm of colored fluid. This is the chamber reflect-
of spillage, tubing, possible disconnection, fragility and ing the pleural pressures within the pleural cavity. Using
decreased mobility because of multiple bottles on the floor these systems, negative suction can be applied at levels
at a patient’s bedside (Figure 45.19). ranging from 0 to -40 cmH2O of suction. Intermittent or
constant bubbling within the water seal chamber is indica-
tive of an air leak. This leak could be occurring from any-
To patient where within the system, from the lung and pleural space
To suction
to the suction tubing to connections to cracks within the
Air plastic containers themselves. Patency of the chest tube is
verified by observing inspiratory and expiratory fluctua-
tion of the fluid in the water seal chamber when the patient
is taken off suction. If there is no fluctuation of this fluid
column with the patient off suction, the tube is said to be
20 cm H2O 20 cm H2O 18 cm H2O
occluded. This means that there is no communication with
the pleural space and that pleural pressures are not being
communicated to the chest tube drainage device. Either
(1) the chest tube is kinked or obstructed by clot or fibri-
Suction control Water-seal Collection nous debris inside the pleural cavity; (2) the chest tube
bottle bottle bottle catheter or rubber tubing going to the chest drainage
Figure 45.19 Three-bottle chest tube system. (From Light RW, device is kinked or crushed, perhaps by the wheel of a bed
Pleural diseases, Lippincott, Williams and Wilkins, 1995, Figure or a chair; (3) the lung is completely expanded, obliterat-
24.7, with permission.) ing the drainage holes of the chest tube inside the chest; or
(4) the chest tube or chest tube tubing outside the pleural
Suction cavity is obstructed by fibrin or blood.
To patient
KEY POINTS
The one-way Heimlich valve
● Drainage techniques of the pleural cavity are
The Heimlich valve is used most frequently in patients based on the premise of safe access, sterile tech-
with pneumothorax who are going to be managed as out- nique, satisfactory re-expansion of underlying
patients. The chest tube is attached to a plastic one-way lung and avoidance of pleural or pulmonary
flutter valve using a five-in-one connector. As inspiration injury, and adequate evacuation of fluid or air.
occurs, the thin flexible rubber tubing within the plastic ● A variety of needles, tubes and catheters are used.
container of the Heimlich valve collapses. This is because Each requires specific expertise and manual dex-
pressure outside the tubing is greater than pressure inside terity.
the tubing. During expiration, pleural pressure becomes ● Closed needle biopsy, although still performed
positive, so the flexible rubber tubing inside the container for diagnosis of cancer or tuberculosis, is being
is kept open, allowing air to escape from the pleural cavity, increasingly replaced by thoracoscopy in many
through the container and into the atmosphere. Because centers.
some patients with pneumothorax may also have small ● There is no clear cut demonstrated advantage for
amounts of fluid draining from their chest, it is reasonable small- or large-bore chest tubes. Therefore, oper-
to attach the one-way Heimlich valve to a bag collection ators should be aware of the advantages and dis-
system that can then be taped to the patient’s chest or advantages of each.
abdomen. One of the upper corners of the collection bag ● New catheters and indwelling tubes that allow
should be cut so that any air accompanying the fluid can permanent evacuation of fluid, particularly in
be easily evacuated (Figure 45.22). patients suffering from recurrent effusion and
trapped lung, warrant greater attention and
increased use.
FUTURE DIRECTIONS
2. O’Moore PV, Mueller PR, Simone JF. Sonographic guidance in 27. Emad A, Rezaian GR. Diagnostic value of closed percutaneous
diagnostic and therapeutic interventions in the pleural space. AJR pleural biopsy vs. pleuroscopy in suspected malignant pleural
Am J Roentgenol 1987; 149: 1–5. effusion or tuberculous pleurisy in a region with a high incidence
3. Petersen S, Freitag M, Albert W, Tempel S, Ludwig K. Ultrasound- of tuberculosis: a comparative, age-dependent study. Respir Med
guided thoracentesis in surgical intensive care patients [letter] 1998; 92: 488–92.
[see comments]. Intensive Care Med 1999; 25: 1029. 28. Nance KV, Shermer RW, Askin FB. Diagnostic efficacy of pleural
4. Sheth S, Hamper UM, Stanley DB, Wheeler JH, Smith PA. US biopsy as compared with that of pleural fluid examination.
guidance for thoracic biopsy: a valuable alternative to CT. Modern Pathol 1991; 4: 320–4.
Radiology 1999; 210: 721–6. 29. Roggli VL. Role of closed-needle biopsy in the diagnosis of
5. Yang PC. Ultrasound-guided transthoracic biopsy of peripheral malignant mesothelioma of the pleura [letter; comment]. Chest
lung, pleural, and chest-wall lesions. J Thorac Imag 1997; 12: 1994; 105: 321–2.
272–84. 30. Suri JC, Goel A, Gupta DK, Bhatia A. Role of serial pleural biopsies
6. Gervais DA, Petersein A, Lee MJ, et al. US-guided thoracentesis: in the diagnosis of pleural effusions. Indian J Chest Dis Allied Sci
requirement for postprocedure chest radiography in patients who 1991; 33: 63–7.
receive mechanical ventilation versus patients who breathe 31. Kirsch CM, Kroe DM, Jensen WA et al. A modified Abrams needle
spontaneously. Radiology 1997; 204: 503–6. biopsy technique. Chest 1995; 108: 982–6.
7. Godwin JE, Sahn SA. Thoracentesis: a safe procedure in 32. O’Connor S, Yung T. A comparison of Abrams and Raja pleural
mechanically ventilated patients [see comments]. Ann Intern Med biopsy needles. Austr N Z J Med 1992; 22: 237–9.
1990; 113: 800–2. 33. Screaton NJ, Flower CD. Percutaneous needle biopsy of the pleura.
8. Hsu WH, Chiang CD, Hsu JY, et al. Value of ultrasonically guided Radiol Clin North Am 2000; 38: 293–301.
needle biopsy of pleural masses: an under-utilized technique. J 34. Ogirala RG, Agarwal V, Vizioli, Pinsker KL, Aldrich TK. Comparison
Clin Ultrasound 1997; 25: 119–25. of the Raja and the Abrams pleural biopsy needles in patients with
9. Keske U. Ultrasound-aided thoracentesis in intensive care patients pleural effusion. Am Rev Respir Dis 1993; 147: 1291–4.
[editorial; comment]. Intensive Care Med 1999; 25: 896–7. 35. Eng P, Colt HG. Fracture of Cope’s needle in closed pleural biopsy
10. Lichtenstein D, Hulot JS, Rabiller A, Tostivint I, Mezière G. [letter]. Chest 1994; 106: 977–8.
Feasibility and safety of ultrasound-aided thoracentesis in 36. Bhandari S, Sellars L. Splenic hematoma complicating pleural
mechanically ventilated patients. Intensive Care Med 1999; 25: biopsy [letter]. Postgrad Med J 1995; 71: 319.
955–8. 37. Lee P, Colt HG. Rigid and semirigid pleuroscopy: the future is
●11. McCartney JP, Adams JWD, Hazard PB. Safety of thoracentesis in bright. Respirology 2005; 10: 418–25.
mechanically ventilated patients. Chest 1993; 103: 1920–1. 38. Lee P, Hsu A, Lo C, Colt HG. Prospective evaluation of flex-rigid
◆12. Moulton JS. Image-guided drainage techniques. Semin Respir pleuroscopy for indeterminate pleural effusion: accuracy, safety
Infect 1999; 14: 59–72. and outcome. Respirology 2007; 12: 881–6.
◆13. Strange C. Pleural complications in the intensive care unit. Clin 39. Silviestri GA, Strange C. Rest in peace: the decline on training and
Chest Med 1999; 20: 317–27. use of the closed pleural biopsy. J Bronchol 2005; 12: 131–32.
14. Schachner A, Aisenberg R, Levy MJ. Retrograde embolization of a 40. Tomlinson MA, Treasure T. Insertion of a chest drain: how to do it.
detached polyethylene catheter. Chest 1981; 79: 600–1. Br J Hosp Med 1997; 58: 248–52.
15. Mahon RT, Colt HG. Bedside thoracentesis in critically ill patients: 41. Rashid MA, Wikström T, Ortenwall P. A simple technique for
the ‘rolled bedsheet’ technique. J Bronchol 2000; 7: 340–2. anchoring chest tubes. Eur Respir J 1998; 12: 958–9.
16. Clark SJ, Vanselow CHG. Use of the reusable Boutin pleural needle 42. Smith L, Baker F, McDougall C, Stead L. Removal of chest drains.
for thoracentesis. Bronchol, 1999; 6: 207–10. Nursing Times 1999; 95: 1–2.
17. Alemán C, Alegre J, Armadans L, et al. The value of chest 40. Palesty JA, McKelvey AA, Dudrick SJ. The efficacy of X-rays after
roentgenography in the diagnosis of pneumothorax after chest tube removal. Am J Surg 2000; 179: 13–6.
thoracentesis. Am J Med 1999; 107: 340–3. 43. Pacanowski JP, Waack ML, Daley BJ, et al. Is routine
18. Bartter T, Mayo PD, Pratter MR, et al. Lower risk and higher yield roentgenography needed after closed tube thoracostomy removal?
for thoracentesis when performed by experienced operators. Chest J Trauma 2000; 48: 684–8.
1993; 103: 1873–6. ◆44. Carroll P. What’s new in chest-tube management. RN 1991; 54:
19. Collins TR, Sahn SA. Thoracocentesis: clinical value, complications, 34–8, 40.
technical problems, and patient experience. Chest 1987; 91: 817–22. ◆45. Carroll P. Exploring chest drain options. RN 2000; 63: 50–4; quiz
●20. Colt HG, Brewer N, Barbur E. Evaluation of patient-related and 56.
procedure-related factors contributing to pneumothorax following 46. Collop NA, Kim S, Sahn SA. Analysis of tube thoracostomy
thoracentesis. Chest 1999; 116: 134–8. performed by pulmonologists at a teaching hospital. Chest 1997;
21. Grogan DR, Irwin RS, Channick R et al. Complications associated 112: 709–13.
with thoracentesis. A prospective, randomized study comparing 47. Gordon P, Norton JM. Managing chest tubes: what is based on
three different methods. Arch Intern Med 1990; 150: 873–7. research, and what is not? Dimens Crit Care Nurs 1995; 14: 14–16.
●22. Capizzi SA, Prakash UB. Chest roentgenography after outpatient 48. Gross SB. Current challenges, concepts, and controversies in
thoracentesis. Mayo Clin Proc 1998; 73: 948–50. chest tube management. AACN Clin Iss Crit Care Nurs 1993; 4:
23. Villena V, Lopez-Encuentra A, Pozo F, De-Pablo A, Martin- 260–75.
Escribano P. Measurement of pleural pressure during therapeutic 49. Martino K, Merrit S, Boyakye, et al. Prospective randomized trial of
thoracentesis. Am J Respir Crit Care Med 2000; 162: 1534–8. thoracostomy removal algorithms. J Trauma 1999; 46: 369–71;
24. Walsh LJ, Macfarlane JT, Manhire AR, Sheppard M, Jones JS. Audit discussion 372–3.
of pleural biopsies: an argument for a pleural biopsy service. 50. Altman E, Ben-Nun A, Curtis W Jr, Best LA. Modified Seldinger
Respir Med 1994; 88: 503–5. technique for the insertion of standard chest tubes. Am J Surg
25. Kirsch CM, Kroe DM, Azzi RL, et al. The optimal number of pleural 2001; 181: 354–5.
biopsy specimens for a diagnosis of tuberculous pleurisy. Chest 51. Tattersall DJ, Traill ZC, Gleeson FV. Chest drains: does size matter?
1997; 112: 702–6. Clin Radiol 2000; 55: 415–21.
26. Chen NH, Hsieh IC, Tsao TC. Comparison of the clinical diagnostic 52. Gammie JS, Banks MC, Fuhrman, et al. The pigtail catheter for
value between pleural needle biopsy and analysis of pleural pleural drainage: a less invasive alternative to tube thoracostomy.
effusion. Chang-Keng i Hsueh Tsa Chih 1997; 20: 11–16. J Soc Laparoendosc Surg 1999; 3: 57–61.
References 567
53. Grodzin CJ, Balk RA. Indwelling small pleural catheter needle 63. Cloutier R, Gignac MA. Pneumothorax following tube
thoracentesis in the management of large pleural effusion. Chest thoracostomy and water seal drainage. Can J Surg 2001; 44:
1997; 111: 981–8. 387.
54. Bell RL, Ovadia P, Abdullah F, Spector S, Rabinovici R. Chest tube ●64. Pien GW, Gant MJ, Washam Cl, Sterman DH. Use of an
removal: end-inspiration or end-expiration? J Trauma, 2001; 50: implantable pleural catheter for trapped lung syndrome in patients
674–7. with malignant pleural effusion. Chest 2001; 119: 1641–6.
55. Bailey RC. Complications of tube thoracostomy in trauma. J Accid 65. Pollak JS, Burdge CM, Rosenblatt M, et al. Treatment of malignant
Emerg Med 2000; 17: 111–14. pleural effusion with tunneled long-term drainage catheters.
56. Baldt MM, Bankier A, Germann PS, et al. Complications after J Vasc Interv Radiol 2001; 12: 201–8.
emergency tube thoracostomy: assessment with CT. Radiology 66. Putnam JB Jr, Light RW, Rodriguez RM, et al. A randomized
1995; 195: 539–43. comparison of indwelling pleural catheter and doxycycline
57. Chan L, Reilly KM, Henderson C, Kahn F, Salluzzo RF. Complication pleurodesis in the management of malignant pleural effusion. Am
rates of tube thoracostomy. Am J Emerg Med 1997; 15: 368–70. Cancer Soc 1999; 86: 1992–9.
58. Desai AV, Phipps PR, Barnes DJ. Shock and ipsilateral pulmonary 67. Putnam JB Jr, Walsh GL, Swisher SG, et al. Outpatient
oedema after tube thoracostomy for spontaneous pneumothorax. J management of malignant pleural effusion by chronic indwelling
Accid Emerg Med, 1999; 16: 454–5. pleural catheter. Ann Thorac Surg 2000; 69: 369–75.
59. Foresti V, Villa A, Casati O, Parisio E, De Filippi G. Abdominal 68. Vricella LA, Trachiotis GD. Heimlich valve in the management of
placement of tube thoracostomy due to lack of recognition of pneumothorax in patients with advanced AIDS. Chest 2001; 120:
paralysis of hemidiaphragm. Chest 1992; 102: 292–3. 15–18.
60. Etoch SW, Bar-Natan MF, Miller FB, Richardson JD. Tube 69. Tremblay A, Michaud G. Single center experience with 250
thoracostomy. Factors related to complications. Arch Surg 1995; tunneled pleural catheter insertions for malignant pleural
130: 521–5; discussion 525–6. effusions. Chest 2006; 129: 362–8.
61. Shapira OM, Aldea GS, Kupferschmid J, Shemin RJ. Delayed 70. Van den Toorn LM, Schaap E, Surmont VFM, et al. Management of
perforation of the esophagus by a closed thoracostomy tube. recurrent malignant pleural effusions with a chronic indwelling
Chest 1993; 104: 1897–8. pleural catheter. Lung Cancer 2005; 50: 123–7.
62. Zieren J, Enzweiler C, Müller JM. Tube thoracostomy complicates 71. Lee P, Colt HG. Flex-rigid pleuroscopy step-by-step. Singapore:
unrecognized diaphragmatic rupture. Thoracic Cardiovasc Surg CMP Medica, 2005.
1999; 47: 199–202.
This page intentionally left blank
46
Pleurodesis
Pleurodesis is the iatrogenic induction of symphysis of the Symptomatic fluid reaccumulation following drainage
visceral and parietal pleura. The aim of pleurodesis is to occurs in 70–90 percent of patients5 and can significantly
obliterate the pleural space and prevent the accumulation reduce quality of life. Pleurodesis may provide effective
of fluid or air. control of fluid reaccumulation (see also Chapter 25).
The main indications for pleurodesis are recurrent, symp- Pleurodesis in pneumothorax is indicated primarily to
tomatic malignant pleural effusion or pneumothorax. In prevent recurrence (see also Chapter 42).
selected cases pleurodesis is performed for benign, recur- For patients with primary spontaneous pneumo-
rent pleural effusion. thorax, recurrence rates increase with successive events6
In the future there is likely to be a rise in the number of and the risk of contralateral pneumothorax is also
potential candidates for pleurodesis reflecting the higher.7 Pleurodesis is indicated in patients with a second
increased incidence of malignant effusion in an ageing ipsilateral pneumothorax, first contralateral pneumo-
population and cases of secondary pneumothorax related thorax, synchronous pneumothoraces, a continuous air
to underlying chronic obstructive pulmonary disease. leak or a spontaneous hemopneumothorax as well as in
those with job restrictions, e.g. divers and airline staff.8
No study has directly compared surgical and bedside
pleurodesis in the setting of pneumothoraces. Based on
Malignant pleural effusion results from observational studies, surgical pleurodesis is
generally preferred over bedside pleurodesis via a chest
Malignant pleural effusions are common and affect 660 tube, the latter usually being reserved for patients unwill-
patients per million population per year.1 In 20 percent of ing or unsuitable for surgery.
patients they represent the initial presentation of malig- In those with secondary pneumothorax who have failed
nancy. Up to 50 percent of patients with breast cancer, a conservative therapy, pleurodesis is advocated as 45–67
quarter of those with bronchogenic carcinoma and over 95 percent of patients with a secondary pneumothorax will
percent of patients with mesothelioma will develop a suffer recurrence9–11 which are often symptomatic and
pleural effusion during their disease course.2–4 may be life threatening.
570 Pleurodesis
PRACTICAL ASPECTS OF PLEURODESIS All pleurodesis agents have differing degrees of efficacy
and side effect profiles, and suggested newer alternatives to
Types of pleurodesis date have not been rigorously compared with conventional
agents.
Current practice is largely anecdotal. Success rates
MECHANICAL PLEURODESIS
quoted are often over-inflated by selection bias and the
Mechanical pleurodesis may be accomplished by pleural rates are much lower in ‘real life’ clinical practice. Over 30
abrasion or parietal pleurectomy. Video-assisted thoraco- randomized controlled trials comparing different agents
scopic surgery (VATS) is now the procedure of choice have been performed but patient numbers are small and
amongst surgeons and is associated with a shorter hospital few results reach statistical significance.34 Pleurodesis
stay and recovery time than open thoracotomy and success rates of different sclerosing agents vary (Table
transaxillary mini-thoracotomy. 46.1).
Pleural abrasion is technically easier to perform but
pleurectomy had a slightly lower recurrence rate in one TALC
large series; 0.4 percent (n = 752) versus 2.3 percent
(n = 301).30 Total parietal pleurectomy is associated with Talc is the most commonly used agent worldwide. It is a
higher post-operative morbidity than apical pleurectomy heterogeneous group of minerals rather than a pure com-
alone or pleural abrasion. Hemothorax occurs in approxi- pound and has the chemical formula Mg3(Si2O5)2(OH)2. It
mately 4 percent of cases.31 is mined from different deposits throughout the world,
each with their own unique composition. Modern talc
supplies are asbestos-free although mineral dusts such as
CHEMICAL PLEURODESIS
calcite, quartz, dolomite and others may be present.
The instillation of pleurodesing agents is usually via a chest Sterilization is achieved using dry heat, ethylene oxide or g
tube, except for talc which may also be administered tho- irradiation, although potential contamination with media-
racoscopically. Pleurodesis via tube thoracostomy has tors such as endotoxin or other materials may occur.
comparable success rates to surgical pleurodesis, is poten- During the preparatory process commercial talc is
tially cheaper and does not involve general anesthetic risks sifted through mesh of varying pore size; smaller particles
(which can be significant in elderly patients with malig- are commonly selected for use in the USA (11–20 mm)
nant effusions or secondary spontaneous pneumo- whereas in France larger talc particles are chosen (mean
thoraces). 34 mm). This difference may explain why talc-related acute
respiratory distress syndrome (ARDS) is more frequently
reported in the USA (see below).
TECHNICAL ASPECTS OF PLEURODESIS Talc can be administered as a powder (poudrage) insuf-
flated at thoracoscopy or thoracotomy (Figure 46.1) or as
Choice of pleurodesis agent ‘talc slurry’ (talc powder mixed with normal saline and
agitated to form a suspension) via a chest tube.35–37
Over 30 pleurodesis agents have been proposed since 1935 Dosages previously used have ranged from 1 to 14 g,
yet none is ideal. In a survey of over 800 pulmonologists, however, less than 5 g is advised as adverse effects seem to
physicians were only ‘somewhat satisfied’ with available be more common with higher talc doses. Success rates for
preparations.32,33 pleurodesis are similar for both thoracoscopic talc
Table 46.1 Success rates and relative efficacy of commonly used pleurodesis agents22,38,51,56,57,60,61,70,129
CYCLINES
Table 46.2 Prospective randomized trials comparing talc with other agents
tries.51 However, oral forms of doxycycline dissolved in (500 mg) compared with 96 percent with talc (5 g)
normal saline demonstrated similar efficacy in rabbits and recently.59 However, case reports of patients experiencing
may offer a less expensive, safe alternative to the parenteral temporary confusion, hallucinations and convulsions fol-
forms in humans (not yet tested).52 lowing intrapleural administration rendered it unpopular
in large parts of Europe and the USA and serious central
CYTOTOXIC AGENTS nervous system toxicity has been described in rabbits.60
QUINACRINE
Combinations of different sclerosing agents, including
uniting surgical and chemical pleurodesis methods, to
Quinacrine, an anti-malarial agent, has been used rou- exploit a potential synergistic action have been evaluated
tinely in Scandinavia for decades.59 Ukale et al.59 reported in few published studies.54,69–72 These have failed to
pleurodesis success rates of 89 percent with quinacrine show conclusive benefit over single-agent pleurodesis.
574 Pleurodesis
Randomized trials are required to determine optimal com- control of the source of the air leak as well as mechanical
binations and compare with current practice. pleurodesis which carries a recurrence rate of <5 percent.
Talc poudrage via VATS demonstrates comparable efficacy
and is favored by some surgeons.
TIMING OF PLEURODESIS Traditionally, chemical pleurodesis via the chest drain
is advocated only if the patient is unsuitable for surgery.
Malignant pleural effusion However, in certain centers, patients with spontaneous
pneumothorax for whom chest drainage is required
No consensus exists regarding the timing of pleurodesis. undergo first-line medical thoracoscopy. This allows
Some pulmonologists advise pleurodesis as soon as a pneumothorax evacuation, assessment and coagulation of
malignant pleural effusion is diagnosed whereas others accessible blebs or bullae, as well as pleurodesis by talc
delay until symptomatic recurrence. The reported success poudrage without the risk of general anesthesia.
rate of pleurodesis between these groups is similar;33 In patients with secondary pneumothorax and a per-
however more advanced pleural malignancy reduces the sistent air leak, early referral (day two to four) is recom-
likelihood of success.73 Delay also increases the risk of mended. Video-assisted thoracoscopy is superior to
eventual development of a ‘trapped lung’ which will pre- chemical pleurodesis via chest tube with recurrence rates
clude pleurodesis. of <5 percent and approximately 20 percent respectively.
Timing of pleurodesis in chemotherapy-sensitive However, if the patient is not fit for general anesthesia, e.g.
tumor is controversial. In ISPP, two-thirds favored waiting because of poor lung functions, chemical pleurodesis via a
to see if chemotherapy is effective in controlling effusion chest drain may be appropriate.
accumulation, but one-third would pleurodese.
Oncologists often favor the former option.74 No data exist
Congestive cardiac failure
on how often these chemotherapy-sensitive tumors recur,
and no definitive strategy is known. Risks of not pleu-
rodesing include a potential site of infection during neu- In selected patients with congestive cardiac failure and
tropenic sepsis and third spacing of chemotherapeutic pleural effusion(s) refractory to conventional therapy,
drugs with resultant toxicity.75 The Food and Drug pleurodesis may be considered.78,79 Pleural fluid accumu-
Administration (FDA) body recommends that pleural lates as an escape route for interstitial fluid in congestive
effusions secondary to malignant pleural mesothelioma be cardiac failure and pleurodesis carries a theoretical risk of
drained prior to chemotherapy with Pemetrexed. worsening contralateral fluid accumulation or alveolar
After the chest tube insertion, instillation of the pleu- edema,80,81 but this has not been demonstrated clinically.
rodesis agent is traditionally delayed until the daily pleural
fluid drainage rate is <150 mL. This practice is supported Hepatic hydrothorax
by the majority of pulmonologists,33 but increases the
period of drainage, prolongs hospital stay and has not been Often, hepatic hydrothoraces form, in part, by the trans-
shown to be more effective. The amount of fluid drained, diaphragmatic migration of ascitic fluid. Caution should
when the X-ray shows no residual fluid, likely represents be exercised against routine chest drain placement in such
the amount of fluid formation and may not be further patients as massive fluid loss may occur associated with
reduced by prolonged drainage. There is an equal chance secondary loss of electrolytes and protein. If medical treat-
of success if pleurodesis is performed when radiographic ment fails and repeated therapeutic thoracenteses for
lung re-expansion is achieved and this practice is recom- symptomatic relief are required, pleurodesis may be
mended.76 attempted. The concomitant application of continuous
positive airway pressure ventilation to decrease the perito-
neopleural pressure gradient (and reduce passage of ascitic
Pneumothorax fluid into the pleural cavity via diaphragmatic stomata)
showed promising initial results but these have not been
Debate surrounding the optimal timing and type of pleu- validated in larger studies.82 Video-assisted thoracoscopy
rodesis in patients with primary spontaneous pneumo- with closure of demonstrable diaphragmatic defects and
thorax continues. Historical series suggest that surgical simultaneous talc pleurodesis offers a potential therapeu-
intervention with pleurectomy or mechanical abrasion is tic strategy although identification of patients with such
superior, however, little evidence-based justification exists defects is not yet possible.83,84
for this approach over chemical pleurodesis in primary
pneumothoraces.
Discussion and referral for surgical intervention is gen- Chylothorax
erally recommended at day five to seven in patients with a
persistent air leak despite tube drainage, although this cut- After conservative measures have failed, and/or ligation of
off time is arbitrary.11,77,78 VATS allows identification and the thoracic duct is not suitable, pleurodesis should be
Clinical considerations at pleurodesis 575
considered.85–87 It was previously believed that successful cavity.96,97 Cytological examination is frequently positive.
chemical pleurodesis for chylothorax was difficult to A direct correlation has been observed between low pleural
achieve: this has been attributed to normal underlying fluid pH (<7.30) and reduced survival rates;98 however,
pleura or chemical composition of chyle. In recent years, data only weakly supports its use (pH <7.20) as a discrim-
however, large series have shown that pleurodesis can inative tool to determine likelihood of pleurodesis
work for recurrent chylothoraces but ongoing medical success.94 Therefore, although low pleural pH may indicate
management to lessen chyle production should continue patients in whom the likelihood of pleurodesis success
following all attempts.73 may be lower, it should not be used in isolation and this
group should not be denied pleurodesis on this basis alone.
PATIENT SELECTION
TRANSPLANT CONSIDERATIONS
Performance status reflects the predicted survival of
patients with malignant pleural effusion and should be Many patients who eventually require lung transplantation
assessed in all patients prior to pleurodesis.88–90 If the encounter complications which raise the question of the
expected survival is short (arbitrary suggestion, less than appropriateness of pleurodesis, e.g. recurrent pneumo-
3 months), less invasive procedures are appropriate (e.g. thoraces in chronic obstructive pulmonary disease
therapeutic thoracentesis). Patients with a good perform- (COPD), cystic fibrosis or lymphangioleiomyomatosis
ance status are more likely to derive benefit from pleu- (LAM). Previously, pleurodesis was seen as a contraindica-
rodesis.89,90 tion to transplantation and avoidance of pleurodesis with
Pleurodesis is only recommended in malignant pleural postponement of definitive preventative measures was rec-
effusion if symptomatic improvement from thoracentesis ommended in any potential future transplant candidate.
is exhibited.91,92 Dyspnea is often multi-factorial and up to However, recent evidence from patients with recurrent
50 percent of patients do not have demonstrable sympto- pneumothoraces and underlying lymphangioleio-
matic relief following fluid evacuation.93 In this group, myomatosis suggests pleural adhesion/fibrosis does not
alternative causes for dyspnea should be addressed, e.g. preclude successful transplantation, although it does
obstructive airways disease, lymphangitis, pulmonary increase the rate of perioperative bleeding and technical
embolism or ‘trapped lung’. The presence of ‘trapped difficulty.10 Liaison with a specialist transplantation unit is
lung’, when lung expansion is restricted either by visceral recommended prior to pleurodesis in any potential future
pleura tumor encasement or by endobronchial obstruc- lung transplant candidate.
tion, prohibits effective pleurodesis as the pleural surfaces
do not oppose. Pleurodesis may also precipitate fibrosis of
the visceral pleura exacerbating restriction of the underly- CLINICAL CONSIDERATIONS AT PLEURODESIS
ing lung.
In patients with metastatic malignant disease involving Size of chest tube for pleurodesis
serosal surfaces, pleural, pericardial and ascitic fluid collec-
tions may concomitantly occur. Clinical management of No adequately powered randomized trials define an
such patients is often difficult. Statistically, patients with optimal chest drain size to maximize pleurodesis efficacy.
extensive disease have a very short prognosis (e.g. weeks) Traditionally, large-bore chest tubes have been employed;
particularly if other treatment modalities (e.g. chemother- however, pleurodesis via small-bore tubes (<16 F) is asso-
apy) are not a viable option. Repeated therapeutic aspira- ciated with comparable success rates and a large number of
tion may be more appropriate. Insertion of an ambulatory non-randomized studies suggest they are better tolerated
indwelling pleural catheter may also provide relief in such than larger bore drains.94,99–106 Interestingly, despite this
cases. Ascitic fluid should be drained prior to attempted evidence, small-bore drains are still under-utilized with
pleurodesis to reduce the transdiaphragmatic migration of larger (28–32 F) tubes preferred in the USA and Canada,
fluid, which may reduce likelihood of success. and 20–24 F favored in UK and Australasia.33 Ambulatory
The identification of clinical or biochemical parameters small-bore catheters enable pleurodesis to be performed
to guide selection of patients for pleurodesis has been the on an outpatient basis obviating the need for hospital
focus of much previous research. Several measures have admission and associated costs.107,108 This has been suc-
been suggested including patients’ performance status, size cessfully applied with most pleurodesing agents, including
of effusion on chest X-ray and pleural fluid LDH (>2¥ talc slurry.
upper normal limit for serum), glucose (<60 mg/dL) or
pH (<7.20).90,94,95 Low pleural fluid glucose (and pH)
appears to arise secondary to tumor infiltration, which Patient rotation
decreases glucose transport into the pleural space, as well
as via increased glucose utilization by cancer cells and Whether rotation of the patient following administration
reduced efflux of acidic byproducts from the pleural of the pleurodesing agent aids its distribution and
576 Pleurodesis
improves success remains debatable. The success rate of Narcotic analgesia and/or conscious sedation should be
pleurodesis and distribution of talc throughout the pleural used as pre-medication provided that no contraindications
cavity appear independent of rotation in small exist. There are no clinical studies evaluating efficacy of
studies.109,110 Concerns, however, have been expressed for different analgesic agents following pleurodesis.
talc slurry as it is not soluble and particles gravitate
towards the bases of the pleural cavity. As rotation is a safe
procedure, there is little to contradict its use. Effect of non-steroidal anti-inflammatory
drugs and prednisolone
Clamping time Modulation of the inflammatory cascade, theorized to play
a salient role in effective pleurodesis, may limit pleurode-
Whether prolonged contact of a pleurodesing agent with sis efficacy. Experimental animal studies have shown that
the pleural surface is required to induce inflammation and concomitant administration of corticosteroids or non-
successful pleural union has not been assessed, and no steroidal anti-inflammatory drugs (NSAIDs) reduces for-
studies address the optimal chest tube clamping time. mation of talc- and doxycycline-induced pleural
Significant variations in practice occur and most pulmo- adhesions.18,113,114 Conflicting evidence exists; in another
nologists clamp the tube for 1–4 hours.33 animal study using short-term NSAIDs no effect was seen,
and the systemic use of these agents did not influence the
outcome of silver nitrate- or TGF-b-induced pleurode-
Suction sis.28,114 Until adequate human studies have been per-
formed, it is recommended that wherever possible these
In order to achieve opposition of the pleural membranes, agents should be avoided prior to, and in the immediate
suction is applied in some centers following release of tube period following, pleurodesis. A randomized trial is under-
clamping. No universal guideline exists but some advocate way to assess pleurodesis efficacy with either NSAIDs or
suction until tube drainage is less than 100 mL daily. If opiate analgesia used at the time of pleurodesis.
used, careful graded suction should be initiated to prevent
re-expansion pulmonary edema.
Effect of heparin
Chest tube removal
Activation of the coagulation cascade has been shown to
play an important role in organ fibrosis and stimulation of
The optimal duration for drainage following pleurodesis is
systemic and pleural coagulation pathways have been
unknown, though most chest drains can be taken out at 48
demonstrated following talc poudrage. The use of prophy-
hours. Some pulmonologists use daily fluid drainage to
lactic heparin (or its derivatives) to prevent thromboem-
guide removal whereas others remove the drain at a spe-
bolic disease is common in clinical practice, however, it
cific time after pleurodesis regardless of fluid volume.33 A
has been postulated that their use may reduce pleurodesis
recent study (n = 41) showed no difference in pleurodesis
efficacy via an inhibitory action on pleural coagula-
success rates in patients randomized to drain removal at 24
tion.115,116 Further study to address the significance of this
hours compared with those whose drain was removed at
effect is needed.
72 hours. However, length of hospital stay was signifi-
cantly reduced in the former group (4 days versus 8 days;
p < 0.01).111
SIDE EFFECTS/COMPLICATIONS OF
PLEURODESIS
Best analgesia
Common side effects of all sclerosing agents include chest
Pleurodesis can be painful, presumably from the intense pain and fever, presumably from the intense pleural
pleural inflammation provoked. Instillation of intrapleural inflammation provoked.
analgesia is recommended before administration of the
pleurodesing agent99 but evidence supporting this practice
is scant and questions remain on whether the effectiveness Talc and its safety
of the anesthetic is maintained following dilution when the
pleurodesing agent is subsequently introduced. Only one Whilst talc is effective, concerns regarding potential lethal
small non-randomized (n = 20) study has compared side effects persist.
250 mg versus 200 mg of intrapleural lidocaine and the Talc-induced ARDS has been widely reported,115,117–124
result was difficult to interpret.112 Further randomized, and its incidence ranges from 0–9 percent in different
blinded trials are needed. studies. The mechanism remains poorly understood.
Alternatives – what to do if pleurodesis fails 577
Acute respiratory distress syndrome has been seen after Repeat pleurodesis
both talc slurry instillation and talc poudrage,122 and with
low (2 g) and high (10 g) talc doses.118 It is hypothesized Repeated talc pleurodesis may be appropriate in patients
that acute lung inflammation secondary to the systemic with good performance status, although anecdotally it is
distribution of talc particles occurs and a resultant pneu- associated with lower success rates. One comparison study
monitis precipitates respiratory failure. To date, many between 500 mg quinacrine (n = 54) and 5 g talc slurry
ARDS cases have been reported from the USA where, (n = 56) reported the need for repeat administration in 31
interestingly, commercial preparations used have predom- versus 7 percent of patients respectively following initial
inantly smaller (<20 mm) size talc particles.125 Randomized treatment. Repeat talc instillation was successful in 50
controlled trials support the theory that talc preparations percent of patients (2/4), increasing the overall success rate
consisting of mainly small particle sizes are absorbed in by only 3.6 percent.51
higher concentrations systemically in animals, possibly A different agent may be used but quality evidence to
through parietal pleural pores, and induce more lung and support this practice is lacking.
systemic inflammation.74 Further studies demonstrate
deposition of talc in extra-pleural tissues following pleu-
rodesis and in rabbit models there is a dose-dependent Ambulatory indwelling pleural catheter
effect.115,120 However, in all the animal and human studies,
few (if any) actually developed ARDS despite the demon- The insertion of a long term indwelling tunneled pleural
stration of extrapleural talc deposition and raised systemic catheter allows outpatient management of refractory
inflammation. malignant pleural effusion (Figure 46.2). It is also indi-
In addition to small-size talc particles, impurities and cated in patients with a symptomatic effusion and under-
other contaminants of talc preparations may also be con- lying ‘trapped lung’. The catheters can be inserted as a day
tributing causes of ARDS. A large prospective study of over case obviating hospital admission and their presence
550 patients in continental Europe failed to identify any induces a complete or partial pleurodesis in up to 58
cases of ARDS after pleurodesis using graded talc of mainly percent of patients.104,108 The patient (or carer) can drain
large particle size.126 the effusion as guided by symptoms avoiding hospitaliza-
At present, the evidence is not sufficient to discourage tion, and the catheters are generally well tolerated.
the use of talc, but small particle size talc preparations Recognized complications include local tumor invasion at
should be avoided. Special care should be exercised in the insertion site, pleural infection and catheter displace-
patients with pre-existing hypoxia or advanced lung ment.
disease who need pleurodesis. An alternative agent (e.g.
tetracycline) should be considered in such circumstances.
Pleuroperitoneal shunt
ALTERNATIVES – WHAT TO DO IF
PLEURODESIS FAILS Repeated thoracentesis
No good data exist to guide the management of patients Serial therapeutic thoracenteses to relieve breathlessness
who have failed pleurodesis, but the following options can should be considered in patients with a poor performance
be considered. status, a short predicted life expectancy (e.g. less than
578 Pleurodesis
Non-invasive palliation
FUTURE DIRECTIONS
4. A deeper understanding of the process by which pleural 5. Anderson CB, Philpott GW, Ferguson TB. The treatment of
fluid initially develops is required and would allow malignant pleural effusions. Cancer 1974; 33: 916–22.
6. Gobbel WGJ, Rhea WGJ, Nelson IA. Spontaneous pneumothorax. J
primary prevention of malignant pleural effusion Thorac Cardiovasc Surg 1963; 46: 331–45.
rather than secondarily addressing the fluid once it has 7. Sadikot RT, Greene T, Meadows K, Arnold AG. Recurrence of
accumulated. primary spontaneous pneumothorax. Thorax 1997; 52: 805–9.
8. Henry M, Arnold T, Harvey J. BTS guidelines for the management of
The potential role of intrapleural therapy with chemother- spontaneous pneumothorax. Thorax 2003; 58 (Suppl 2): ii39– 52.
9. Lippert HL, Lund O, Blegvad S, Larsen HV. Independent risk factors
apeutic agents, immune modulators and gene therapy for cumulative recurrence rate after first spontaneous
offers a novel approach. Gene therapy is the subject of pneumothorax. Eur Respir J 1991; 4: 324–31.
exciting ongoing research and although still in its infancy, 10. Almoosa KF, Ryu JH, Mendez J, et al. Management of
may provide a future treatment of pleural malignancies pneumothorax in lymphangioleiomyomatosis: effects on
(see Chapter 49). recurrence and lung transplantation complications. Chest 2006;
129: 1274–81.
11. Light RW, O’Hara VS, Moritz TE, et al. Intrapleural tetracycline for
the prevention of recurrent spontaneous pneumothorax. Results of
a Department of Veterans Affairs cooperative study. J Am Med
KEY POINTS Assoc 1990; 264: 2224–30.
12. Kennedy L, Harley RA, Sahn SA, Strange C. Talc slurry pleurodesis.
Pleural fluid and histologic analysis. Chest 1995; 107: 1707–12.
● Pleurodesis is an effective therapeutic option for 13. Montes JF, Garcia-Valero J, Ferrer J. Evidence of innervation in
recurrent pleural effusions and pneumothoraces. talc-induced pleural adhesions. Chest 2006; 130: 702–9.
It should be considered in patients with recurrent 14. Guo YB, Kalomenidis I, Hawthorne M, et al. Pleurodesis is inhibited
effusions who have symptomatic relief of by anti-vascular endothelial growth factor antibody. Chest 2005;
dyspnea upon fluid removal. 128: 1790–7.
15. Tzouvelekis A, Anevlavis S, Bouros D. Angiogenesis in interstitial
● Various surgical and chemical pleurodesing lung diseases: a pathogenetic hallmark or a bystander? Respir Res
methods/agents exist, each with its advantages 2006; 7: 82.
and side effect profiles. 16. Genofre EH, Vargas FS, Antonangelo L, et al. Ultrastructural acute
● Talc is the most commonly used and effective features of active remodeling after chemical pleurodesis induced
agent, though it carries potentially lethal compli- by silver nitrate or talc. Lung 2005; 183: 197–207.
17. Marchi E, Vargas FS, Acencio MM, et al. Evidence that mesothelial
cations. Talc poudrage and slurry are of similar cells regulate the acute inflammatory response in talc pleurodesis.
efficacy in randomized studies. Eur Respir J 2006; 28: 929–32.
● The best protocol of administration of chemical 18. Xie C, Teixeira LR, McGovern JP, Light RW. Systemic
pleurodesis (e.g. tube size, analgesia, use of corticosteroids decrease the effectiveness of talc pleurodesis. Am J
suction etc.) remains poorly defined. Respir Crit Care Med 1998; 157: 1441–4.
19. Agrenius V, Gustafsson LE, Widstrom O. Tumour necrosis factor-
● Indwelling pleural catheters should be consid- alpha and nitric oxide, determined as nitrite, in malignant pleural
ered for ambulatory drainage in patients who are effusion. Respir Med 1994; 88: 743–8.
unsuitable for, or have failed, pleurodesis. 20. van den Heuvel MM, Smit HJ, Barbierato SB, et al. Talc-induced
● Success of pleurodesis should be defined by best inflammation in the pleural cavity. Eur Respir J 1998; 12: 1419–23.
symptomatic relief to the patients rather than by 21. Nasreen N, Hartman DL, Mohammed KA, Antony VB. Talc-induced
expression of C-C and C-X-C chemokines and intercellular
radiographic measures. adhesion molecule-1 in mesothelial cells. Am J Respir Crit Care
Med 1998; 158: 971–8.
22. Walker-Renard PB, Vaughan LM, Sahn SA. Chemical pleurodesis for
malignant pleural effusions. Ann Intern Med 1994; 120: 56–64.
23. Lee YCG, Teixeira LR, Devin CJ, et al. Transforming growth factor-
beta2 induces pleurodesis significantly faster than talc. Am J
REFERENCES Respir Crit Care Med 2001; 163: 640–4.
24. Light RW, Cheng DS, Lee YC, et al. A single intrapleural injection
of transforming growth factor-beta(2) produces an excellent
1. Marel M, Zrustova M, Stasny B, Light RW. The incidence of pleural pleurodesis in rabbits. Am J Respir Crit Care Med 2000; 162:
effusion in a well-defined region. Epidemiologic study in central 98–104.
Bohemia. Chest 1993; 104: 1486–9. 25. Lee YC, Lane KB, Parker RE, et al. Transforming growth factor
2. Lee YCG, De Clerk NH, Henderson DW, Musk AW. Malignant beta(2) (TGF beta(2)) produces effective pleurodesis in sheep with
mesothelioma. In Hendrick DJ, Burge PS, Beckett WS, Churg A no systemic complications. Thorax 2000; 55: 1058–62.
(eds). Occupational disorders of the lung. Recognition, 26. Lee YC, Yasay JR, Johnson JE, et al. Comparing transforming
management and prevention. London: WB Saunders & Co., 2002: growth factor-beta2, talc and bleomycin as pleurodesing agents in
359–79. sheep. Respirology 2002; 7: 209–16.
3. Lee YCG, Dean A, Thompson RI, Robinson BWS. Clinical and 27. Lee YCG, Lane KB, Zoia O, et al. Transforming growth factor-beta
palliative care aspects of malignant mesothelioma. In Robinson induces collagen synthesis without inducing IL-8 production in
BWS, Chahinian P (eds). Mesothelioma. London: Martin Dunitz, mesothelial cells. Eur Respir J 2003; 22: 197–202.
2002: 111–26. 28. Lee YC, Devin CJ, Teixeira LR, et al. Transforming growth factor
4. Sahn SA. Malignancy metastatic to the pleura. Clin Chest Med beta2 induced pleurodesis is not inhibited by corticosteroids.
1998; 19: 351–61. Thorax 2001; 56: 643–8.
580 Pleurodesis
29. Mutsaers SE, Kalomenidis I, Wilson NA, Lee YC. Growth factors in symptomatic malignant pleural effusions. Chest 1997; 112:
pleural fibrosis. Curr Opin Pulm Med 2006; 12: 251–8. 430–4.
30. Weeden D, Smith GH. Surgical experience in the management of 51. Dikensoy O, Light RW. Alternative widely available, inexpensive
spontaneous pneumothorax, 1972–82. Thorax 1983; 38: 737–43. agents for pleurodesis. Curr Opin Pulm Med 2005; 11: 340–4.
31. Thomas P, Le Mee F, Le Hors H, et al. [Results of surgical treatment 52. Bilaceroglu S, Guo Y, Hawthorne ML, et al. Oral forms of
of persistent or recurrent pneumothorax]. Ann Chir 1993; 47: tetracycline and doxycycline are effective in producing
136–40. pleurodesis. Chest 2005; 128: 3750–6.
32. Bethune N. Pleural poudrage: a new technique for the deliberate 53. Marchi E, Vargas FS, Teixeira LR, et al. Comparison of nitrogen
production of pleural adhesions as a preliminary to lobectomy. J mustard, cytarabine and dacarbazine as pleural sclerosing agents
Thorac Surg 1935; 4: 251–61. in rabbits. Eur Respir J 1997; 10: 598–602.
33. Lee YC, Baumann MH, Maskell NA, et al. Pleurodesis practice for 54. Kishi K, Homma S, Sakamoto S, et al. Efficacious pleurodesis with
malignant pleural effusions in five English-speaking countries: OK-432 and doxorubicin against malignant pleural effusions. Eur
survey of pulmonologists. Chest 2003; 124: 2229–38. Respir J 2004; 24: 263–6.
34. Tan C, Sedrakyan A, Browne J, Swift S, Treasure T. The evidence on 55. Brock RC. The use of silver nitrate in the production of aseptic
the effectiveness of management for malignant pleural effusion: a obliterative pleuritis. Guys Hosp Rep 1942; 91: 99–103.
systematic review. Eur J Cardiothorac Surg 2006; 29: 829–38. 56. Brock RC. Recurrent and chronic spontaneous pneumothorax.
35. Antony VB, Loddenkemper R, Astoul P, et al. Management of Thorax 1948; 3: 88–91.
malignant pleural effusions. Eur Respir J 2001; 18: 402–19. 57. Paschoalini MS, Vargas FS, Marchi E, et al. Prospective randomized
36. Kennedy L, Sahn SA. Talc pleurodesis for the treatment of trial of silver nitrate vs talc slurry in pleurodesis for symptomatic
pneumothorax and pleural effusion. Chest 1994; 106: 1215–22. malignant pleural effusions. Chest 2005; 128: 684–9.
37. Yim AP, Chan AT, Lee TW, Wan IY, Ho JK. Thoracoscopic talc 58. Olivares-Torres CA, Laniado-Laborin R, Chavez-Garcia C, et al.
insufflation versus talc slurry for symptomatic malignant pleural Iodopovidone pleurodesis for recurrent pleural effusions. Chest
effusion. Ann Thorac Surg 1996; 62: 1655–8. 2002; 122: 581–3.
38. Dresler CM, Olak J, Herndon JE, et al. Phase III intergroup study of 59. Ukale V, Agrenius V, Hillerdal G, Mohlkert D, Widstrom O.
talc poudrage vs talc slurry sclerosis for malignant pleural Pleurodesis in recurrent pleural effusions: a randomized
effusion. Chest 2005; 127: 909–15. comparison of a classical and a currently popular drug. Lung
39. Webb WR, Ozmen V, Moulder PV, Shabahang B, Breaux J. Iodized Cancer 2004; 43: 323–8.
talc pleurodesis for the treatment of pleural effusions. J Thorac 60. Bjorkman S, Elisson LO, Gabrielsson J. Pharmacokinetics of
Cardiovasc Surg 1992; 103: 881–5. quinacrine after intrapleural instillation in rabbits and man. J
40. Adler RH, Sayek I. Treatment of malignant pleural effusion: a Pharm Pharmacol 1989; 41: 160–3.
method using tube thoracostomy and talc. Ann Thorac Surg 1976; 61. Foresti V. Corynebacterium parvum for malignant pleural
22: 8–15. effusions. Thorax 1995; 50: 104.
41. Diacon AH, Wyser C, Bolliger CT, et al. Prospective randomized 62. Bartal AH, Gazitt Y, Zidan G, Vermeulen B, Robinson E. Clinical and
comparison of thoracoscopic talc poudrage under local anesthesia flow cytometry characteristics of malignant pleural effusions in
versus bleomycin instillation for pleurodesis in malignant pleural patients after intracavitary administration of methylprednisolone
effusions. Am J Respir Crit Care Med 2000; 162: 1445–9. acetate. Cancer 1991; 67: 3136–40.
42. Fentiman IS, Rubens RD, Hayward JL. Control of pleural effusions 63. North SA, Au HJ, Halls SB, Tkachuk L, Mackey JR. A randomized,
in patients with breast cancer. A randomized trial. Cancer 1983; phase III, double-blind, placebo-controlled trial of intrapleural
52: 737–9. instillation of methylprednisolone acetate in the management of
43. Fentiman IS, Rubens RD, Hayward JL. A comparison of malignant pleural effusion. Chest 2003; 123: 822–7.
intracavitary talc and tetracycline for the control of pleural 64. Ren S, Terman DS, Bohach G, et al. Intrapleural staphylococcal
effusions secondary to breast cancer. Eur J Cancer Clin Oncol superantigen induces resolution of malignant pleural effusions
1986; 22: 1079–81. and a survival benefit in non-small cell lung cancer. Chest 2004;
44. Haddad FJ, Younes RN, Gross JL, Deheinzelin D. Pleurodesis in 126: 1529–39.
patients with malignant pleural effusions: talc slurry or 65. Dumire R, Crabbe MM, Mappin FG, Fontenelle LJ. Autologous
bleomycin? Results of a prospective randomized trial. World J Surg “blood patch” pleurodesis for persistent pulmonary air leak. Chest
2004; 28: 749–53. 1992; 101: 64–6.
45. Hamed H, Fentiman IS, Chaudary MA, Rubens RD. Comparison of 66. Williams P, Laing R. Tension pneumothorax complicating
intracavitary bleomycin and talc for control of pleural effusions autologous “blood patch” pleurodesis. Thorax 2005; 60: 1066–7.
secondary to carcinoma of the breast. Br J Surg 1989; 76: 67. Kinoshita T, Miyoshi S, Katoh M, et al. Intrapleural administration
1266–7. of a large amount of diluted fibrin glue for intractable
46. Hartman DL, Gaither JM, Kesler KA, et al. Comparison of pneumothorax. Chest 2000; 117: 790–5.
insufflated talc under thoracoscopic guidance with standard 68. Guerin JC, Van Derschueren RG. [Treatment of recurrent
tetracycline and bleomycin pleurodesis for control of malignant pneumothorax applying fibrin adhesive under endoscopy]. Rev Mal
pleural effusions. J Thorac Cardiovasc Surg 1993; 105: 743–7. Respir 1989; 6: 443–5.
47. Lynch T, Kalish L, Mentzer SJ, et al. Optimal therapy of malignant 69. Chen JS, Hsu HH, Kuo SW, et al. Effects of additional minocycline
pleural effusions: report of a randomised trial of bleomycin, pleurodesis after thoracoscopic procedures for primary
tetracycline, and talc and a meta-analysis. Int J Oncol 1996; 8: spontaneous pneumothorax. Chest 2004; 125: 50–5.
183–90. 70. Dikensoy O, Zhu Z, Donnelly E, et al. Combination therapy with
48. Noppen M, Degreve J, Mignolet M, Vincken W. A prospective, intrapleural doxycycline and talc in reduced doses is effective in
randomised study comparing the efficacy of talc slurry and producing pleurodesis in rabbits. Chest 2005; 128: 3735–42.
bleomycin in the treatment of malignant pleural effusions. Acta 71. Rusch VW, Figlin R, Godwin D, Piantadosi S. Intrapleural cisplatin
Clin Belg 1997; 52: 258–62. and cytarabine in the management of malignant pleural
49. Ong KC, Indumathi V, Raghuram J, Ong YY. A comparative study of effusions: a Lung Cancer Study Group trial. J Clin Oncol 1991; 9:
pleurodesis using talc slurry and bleomycin in the management of 313–9.
malignant pleural effusions. Respirology 2000; 5: 99–103. 72. Tanaka A, Sato T. [Adhesion therapy for malignant pleural effusion
50. Zimmer PW, Hill M, Casey K, Harvey E, Low DE. Prospective (intrapleural administration of OK-432 with minocycline)]. Nihon
randomized trial of talc slurry vs bleomycin in pleurodesis for Kokyuki Gakkai Zasshi 1999; 37: 531–7.
References 581
73. Rodriguez-Panadero F, Antony VB. Pleurodesis: state of the art. and response to treatment in neoplastic effusions. Cancer 1964;
Eur Respir J 1997; 10: 1648–54. 17: 914–28.
74. Maskell NA, Lee YC, Gleeson FV, et al. Randomized trials describing 97. Good JT Jr, Taryle DA, Sahn SA. The pathogenesis of low glucose,
lung inflammation after pleurodesis with talc of varying particle low pH malignant effusions. Am Rev Respir Dis 1985; 131:
size. Am J Respir Crit Care Med 2004; 170: 377–82. 737–41.
75. Li J,.Gwilt P. The effect of malignant effusions on methotrexate 98. Sahn SA, Good JT Jr. Pleural fluid pH in malignant effusions.
disposition. Cancer Chemother Pharmacol 2002; 50: 373–82. Diagnostic, prognostic, and therapeutic implications. Ann Intern
76. Villanueva AG, Gray AW Jr, Shahian DM, Williamson WA, Beamis Med 1988; 108: 345–9.
JF Jr. Efficacy of short term versus long term tube thoracostomy 99. Antunes G, Neville E, Duffy J, Ali N. BTS guidelines for the
drainage before tetracycline pleurodesis in the treatment of management of malignant pleural effusions. Thorax 2003; 58
malignant pleural effusions. Thorax 1994; 49: 23–5. (Suppl 2): ii29–38.
77. O’Rourke JP, Yee ES. Civilian spontaneous pneumothorax. 100. Bloom AI, Wilson MW, Kerlan RK Jr, Gordon RL, LaBerge JM. Talc
Treatment options and long-term results. Chest 1989; 96: 1302–6. pleurodesis through small-bore percutaneous tubes. Cardiovasc
78. Sudduth CD, Sahn SA. Pleurodesis for nonmalignant pleural Intervent Radiol 1999; 22: 433–6.
effusions. Recommendations. Chest 1992; 102: 1855–60. 101. Marom EM, Patz EF Jr, Erasmus JJ, et al. Malignant pleural
79. Glazer M, Berkman N, Lafair JS, Kramer MR. Successful talc slurry effusions: treatment with small-bore-catheter thoracostomy and
pleurodesis in patients with nonmalignant pleural effusion. Chest talc pleurodesis. Radiology 1999; 210: 277–81.
2000; 117: 1404–9. 102. Morrison MC, Mueller PR, Lee MJ, et al. Sclerotherapy of
80. Davidoff D, Naparstek Y, Eliakim M. The use of pleurodesis for malignant pleural effusion through sonographically placed small-
intractable pleural effusion due to congestive heart failure. bore catheters. AJR Am J Roentgenol 1992; 158: 41–3.
Postgrad Med J 1983; 59: 330–1. 103. Musani AI, Haas AR, Seijo L, Wilby M, Sterman DH. Outpatient
81. Glazer M, Berkman N, Lafair JS, Kramer MR. Successful talc slurry management of malignant pleural effusions with small-bore,
pleurodesis in patients with nonmalignant pleural effusion. Chest tunneled pleural catheters. Respiration 2004; 71: 559–66.
2000; 117: 1404–9. 104. Saffran L, Ost DE, Fein AM, Schiff MJ. Outpatient pleurodesis of
82. Boiteau R, Tenaillon A, Law Koune JD, et al. Treatment for cirrhotic malignant pleural effusions using a small-bore pigtail catheter.
hydrothorax with CPAP on mask and tetracycline pleural sclerosis Chest 2000; 118: 417–21.
(abstract). Am Rev Respir Dis 1990; A770: 141. 105. Spiegler PA, Hurewitz AN, Groth ML. Rapid pleurodesis for
83. De Campos, Filho LO, Werebe EC, et al. Thoracoscopy and talc malignant pleural effusions. Chest 2003; 123: 1895–8.
poudrage in the management of hepatic hydrothorax. Chest 2000; 106. Walsh FW, Alberts WM, Solomon DA, Goldman AL. Malignant
13–7. pleural effusions: pleurodesis using a small-bore percutaneous
84. Mouroux J, Perrin C, Venissac N, Blaive B, Richelme H. catheter. South Med J 1989; 82: 963–5, 972.
Management of pleural effusion of cirrhotic origin. Chest 1996; 107. Putnam JB Jr, Light RW, Rodriguez RM, et al. A randomized
109: 1093–6. comparison of indwelling pleural catheter and doxycycline
85. Fogli L, Gorini P, Belcastro S. Conservative management of pleurodesis in the management of malignant pleural effusions.
traumatic chylothorax: a case report. Intensive Care Med 1993; Cancer 1999; 86: 1992–9.
19: 176–7. 108. Putnam JB Jr, Walsh GL, Swisher SG, et al. Outpatient
86. Mares DC, Mathur PN. Medical thoracoscopic talc pleurodesis for management of malignant pleural effusion by a chronic indwelling
chylothorax due to lymphoma: a case series. Chest 1998; 114: pleural catheter. Ann Thorac Surg 2000; 69: 369–75.
731–5. 109. Dryzer SR, Allen ML, Strange C, Sahn SA. A comparison of rotation
87. Norum J, Aasebo U. Intrapleural bleomycin in the treatment of and nonrotation in tetracycline pleurodesis. Chest 1993; 104:
chylothorax. J Chemother 1994; 6: 427–30. 1763–6.
88. Bernard A, de Dompsure RB, Hagry O, Favre JP. Early and late 110. Mager HJ, Maesen B, Verzijlbergen F, Schramel F. Distribution of
mortality after pleurodesis for malignant pleural effusion. Ann talc suspension during treatment of malignant pleural effusion
Thorac Surg 2002; 74: 213–17. with talc pleurodesis. Lung Cancer 2002; 36: 77–81.
89. Burrows CM, Mathews WC, Colt HG. Predicting survival in patients 111. Goodman A, Davies CW. Efficacy of short-term versus long-term
with recurrent symptomatic malignant pleural effusions: an chest tube drainage following talc slurry pleurodesis in patients
assessment of the prognostic values of physiologic, morphologic, with malignant pleural effusions: a randomised trial. Lung Cancer
and quality of life measures of extent of disease. Chest 2000; 117: 2006; 54: 51–5.
73–8. 112. Sherman S, Ravikrishnan KP, Patel AS, Seidman JC. Optimum
90. Martinez-Moragon E, Aparicio J, Sanchis J, et al. Malignant anesthesia with intrapleural lidocaine during chemical pleurodesis
pleural effusion: prognostic factors for survival and response to with tetracycline. Chest 1988; 93: 533–6.
chemical pleurodesis in a series of 120 cases. Respiration 1998; 113. Teixeira LR, Wu W, Chang DS, Light RW. The effect of
65: 108–13. corticosteroids on pleurodesis induced by doxycycline in rabbits.
91. Lee YCG, Rodrigues RM, Lane KB, Light RW. Pleurodesis for Chest 2002; 121: 216–9.
recurrent pleural effusions in the new millennium. Recent Adv Res 114. Teixeira LR, Vargas FS, Acencio MM, et al. Influence of
Updates 2001; 2: 81–9. antiinflammatory drugs (methylprednisolone and diclofenac
92. Light RW. Pleural diseases. Baltimore: Lippincott, Williams & sodium) on experimental pleurodesis induced by silver nitrate or
Wilkins, 2001. talc. Chest 2005; 128: 4041–5.
93. Shinto RA, Stansbury DW, Fischer CE, Light RW. Does therapeutic 115. Rodriguez-Panadero F, Segado A, Torres I, et al. Thoracoscopy and
thoracentesis improve the exercise capacity of patients with talc poudrage induce an activation of the systemic coagulation
pleural effusion? Am Rev Respir Dis 1988; A244: 135. system. Am J Respir Crit Care Med 1995; 151: 785–90.
94. Heffner JE, Nietert PJ, Barbieri C. Pleural fluid pH as a predictor of 116. Rodriguez-Panadero F, Segado A, Martin J, et al. Activation of
pleurodesis failure: analysis of primary data. Chest 2000; 117: systemic coagulation in talc poudrage can be (partially) controlled
87–95. with prophylactic heparin. Am J Respir Crit Care Med 1996; 152:
95. Loddenkemper R. Are prognostic factors helpful in determining the A458.
indication for pleurodesis in malignant pleural effusions? 117. Bouchama A, Chastre J, Gaudichet A, Soler P, Gibert C. Acute
Respiration 1998; 65: 106–7. pneumonitis with bilateral pleural effusion after talc pleurodesis.
96. Clarkson B. Relationship between cell type, glucose concentration, Chest 1984; 86: 795–7.
582 Pleurodesis
118. Campos JR, Werebe EC, Vargas FS, Jatene FB, Light RW. 125. Ferrer J, Villarino MA, Tura JM, Traveria A, Light RW. Talc
Respiratory failure due to insufflated talc. Lancet 1997; 349: preparations used for pleurodesis vary markedly from one
251–2. preparation to another. Chest 2001; 119: 1901–5.
119. Marel M, Skacel Z, Bednar M, Julak J, Light RW. Corynebacterium 126. Janssen JP, Collier G, Astoul P, et al. Safety of pleurodesis with talc
parvum, bleomycin and talc in the treatment of malignant pleural poudrage in malignant pleural effusion: a prospective cohort
effusions. J BUON 1998; 1: 165–70. study. Lancet 2007; 369: 1535–9.
120. Migueres J, Jover A. [Indications for intrapleural talc under 127. Lange P, Mortensen J, Groth S. Lung function 22–35 years after
pleuroscopic control in malignant recurrent pleural effusions. treatment of idiopathic spontaneous pneumothorax with talc
Based on 26 cases (author’s transl)]. Poumon Coeur 1981; 37: poudrage or simple drainage. Thorax 1988; 43: 559–61.
295–7. 128. Gyorik S, Erni S, Studler U, et al. Long term follow up of
121. Nandi P. Recurrent spontaneous pneumothorax; an effective thoracoscopic talc pleurodesis for primary spontaneous
method of talc poudrage. Chest 1980; 77: 493–5. pneumothorax. Eur Respir J 2006.
122. Rehse DH, Aye RW, Florence MG. Respiratory failure following talc 129. Vargas FS, Teixeira LR, Vaz MA, et al. Silver nitrate is superior to
pleurodesis. Am J Surg 1999; 177: 437–40. talc slurry in producing pleurodesis in rabbits. Chest 2000; 118:
123. Rinaldo JE, Owens GR, Rogers RM. Adult respiratory distress 808–13.
syndrome following intrapleural instillation of talc. J Thorac 130. Sorensen PG, Svendsen TL, Enk B. Treatment of malignant pleural
Cardiovasc Surg 1983; 85: 523–6. effusion with drainage, with and without instillation of talc. Eur J
124. Todd TR, Delarue NC, Ilves R, Pearson FG, Cooper JD. Talc poudrage Respir Dis 1984; 65: 131–5.
of malignant pleural effusion. Chest 1980; 78: 542–3.
47
Medical thoracoscopy
ROBERT LODDENKEMPER
Richard Cruise in Ireland most probably was the first who with all technical procedures, there is a learning curve
in 1866 performed a binocular thoracoscopy.13,14 After the before full competence is achieved.32 Appropriate training
report by Jacobaeus, thoracoscopy was used for diagnostic is therefore mandatory. The technique is actually very
purposes by some other pulmonary specialists, mainly in similar to chest-tube insertion by means of a trocar, the
Scandinavia, Germany, Italy and other European coun- difference being that the pleural cavity can be visualized in
tries.15 Jacobaeus published his vast diagnostic experiences addition and biopsies can be taken from all areas of the
in 1925, describing in detail his studies of the etiology and pleural cavity including the chest wall, diaphragm, medi-
staging of tuberculous pleurisy, of malignant effusion, astinum and lung. If indicated, talc poudrage can be per-
rheumatoid effusion, empyema, parapneumonic effusion, formed prior to chest-tube insertion. In our experience,
as well as idiopathic pneumothorax.16 medical thoracoscopy is easier to learn than flexible bron-
However, Jacobaeus himself initiated the therapeutic choscopy, if sufficient expertise in thoracentesis and chest-
application of thoracoscopy for lysis of pleural adhesions tube placement has already been gained.5
by means of thoracocautery to facilitate pneumothorax Jacobaeus1 in his pioneering paper defined three main
treatment of tuberculosis (Jacobaeus’ operation).2 During prerequisites for the examination of serous cavities. First,
the ensuing 40 years, his technique of using one entry for the possibility to introduce a trocar (or puncture needle)
the thoracoscope and another for the electrocautery was into the respective cavity without lacerating the inner
applied worldwide for this specific therapeutic purpose,12 organs and without causing too much pain. Second, the
until antibiotic therapy of tuberculosis was introduced and introduction of a transparent medium into the cavity – he
proved much more successful. However, during these used filtered air for this purpose. Third, an endoscope of
times thoracoscopy was also used by several authors, again such small dimensions that it can be introduced through
mainly in Europe, as a therapeutic tool in non-tuberculous the trocar. He felt that these three main prerequisites were
diseases, especially in the treatment of patients with idio- fulfilled even better by thoracoscopy than by laparoscopy.
pathic spontaneous pneumothorax.17 Later, in 1963, the An absolute prerequisite for thoracoscopy is the pres-
first report on talc poudrage during thoracoscopy for pleu- ence of an adequate pleural space, which should be at least
rodesis in chronic, mainly malignant pleural effusions was 6–10 cm in diameter. If not already present, a pneumo-
published.18 This technique is now frequently applied thorax is induced, immediately before thoracoscopy under
because of its high success rate.19 Between 1950 and 1960, fluoroscopic/sonographic control or the day before under
a generation of chest physicians already familiar with the radiographic control. If extensive pleuropulmonary adhe-
therapeutic application of thoracoscopy began to use the sions exist, ‘extended’ thoracoscopy without creating a
technique on a much broader basis in pleuropulmonary pneumothorax can be carried out, but this requires special
biopsy diagnosis, even for localized and diffuse lung dis- skills and should not be undertaken without special train-
eases and, in the following years, the technique became ing.33 The use of mini-thoracoscopy may be a less invasive
popular mainly in continental Europe,20,21 and later in alternative in patients with loculated effusions.34
other parts of the world.22–24 The main advantage of medical thoracoscopy com-
The excellent results of laparoscopic surgery and the pared with VATS is that the examination can be per-
tremendous advances in endoscopic technology stimulated formed under local anesthesia or conscious sedation
many thoracic surgeons almost simultaneously in Europe (neuroleptic analgesia) after adequate premedication, and
and the USA to develop minimally invasive techniques thus without the support of an anesthetist. Furthermore,
which were termed ‘therapeutic’ or ‘surgical thoracoscopy’ medical thoracoscopy is also less expensive because it may
as well as ‘video-controlled’ or ‘video-thoracoscopic be safely performed with non-disposable instruments and
surgery’ or ‘video-assisted thoracic surgery’ (VATS).4,25,26 in an appropriate endoscopy room.
Particularly in the USA, where thoracoscopy was per- There are two different techniques of diagnostic and
formed only by few pulmonary physicians, a heated debate therapeutic thoracoscopy, as performed by the pneumolo-
was started whether thoracoscopy should be the domain of gist. The one method, very similar to the technique first
the pulmonologists or of the thoracic surgeon,26,27 while in described by Jacobaeus for diagnostic purposes, uses a
most parts of Europe this was not a controversial issue as single entry with a rigid 9 mm thoracoscope (Storz
many pneumologists had been performing thoracoscopy for Company/Olympus) with a working channel for accessory
several years before the introduction of VATS.28,29 Today, instruments and an optical biopsy forceps under local
American thoracic surgeons agree that medical thora- anesthesia3,20,35 (Figure 47.1).
coscopy (pleuroscopy) is in the domain of pulmonologists, The single entry technique has now been modified by
but correctly claim that adequate training is needed.30,31 the introduction of an autoclavable semi-rigid pleuroscope
(Olympus) which has the advantage that handling is very
simple, similar to a flexible bronchovideoscope (Figure
TECHNIQUES 47.2).24,36,37
The other technique, as used by Jacobaeus for lysis of
Medical thoracoscopy is an invasive technique, which adhesions, uses two entries, one with a 7 mm trocar for the
should be used only when other simple methods fail. As rigid examination telescope and the other with a 5 mm
Techniques 585
(a)
(b)
(c)
(d)
Figure 47.4 Thoracoscopic view into the left pleural cavity with
normal upper and lower lobes of the lung. The chest wall pleura is
covered by a whitish layer of small tumor nodules (diffuse
malignant pleural mesothelioma). (See also Color Plate 49.)
book Practical Thoracoscopy by Boutin, Viallat and thoracic wall over all puncture sites. None of the 20
Aelony.21 mesothelioma patients who received radiotherapy devel-
oped malignant outgrowths from the chest wall, whereas
of the 20 who were not irradiated, nine (45 percent,
CONTRAINDICATIONS AND PREVENTION OF p < 0.05) developed nodules. However, tumor growth may
COMPLICATIONS occur even after diagnostic thoracentesis with small
needles, or after closed needle biopsy. The usefulness of
Medical thoracoscopy is safe if the contraindications are prophylactic radiotherapy in mesothelioma was confirmed
observed, and if certain standard criteria are fulfilled.20,21,41 in a report from the UK.46
An obliterated pleural space is an absolute contraindica-
tion. Relative contraindications include bleeding disor-
ders, hypoxemia and an unstable cardiovascular status, as INDICATIONS
well as persistent uncontrollable cough.
Cardiorespiratory functions should be monitored Medical thoracoscopy is primarily a diagnostic proce-
during the procedure by electrocardiogram (ECG), meas- dure.3,5 The indications for its use include evaluation of
urement of blood pressure and continuous oximetry. exudative effusions of unknown cause, staging of malig-
Complications such as benign cardiac arrhythmias, low- nant mesothelioma or lung cancer, and treatment of
grade hypertension or hypoxemia can be almost com- malignant or other recurrent effusions with talc pleurode-
pletely prevented by nasal administration of oxygen.38,42 sis. The procedure is not only the gold standard in the
Since significant hypoventilation has been observed under diagnosis of pleural effusions, but it is also a remarkable
sedation-assisted local anaesthesia, the additional meas- tool for research as a gold standard in the study of pleural
urement of cutaneous carbon dioxide tension (PcCO2) is effusions. It can also be useful in the management of early
recommended.43 empyema and of pneumothorax. If placement of a chest
The most serious, but fortunately least frequent, com- tube is indicated anyhow, and if the facilities are available,
plication is severe hemorrhage due to blood vessel injury medical thoracoscopy should be performed at the time of
during the procedure. This, and pulmonary perforations, chest-tube insertion by the trocar technique. For those
can be avoided by using safe points of entry and a cautious familiar with the technique, other, mainly diagnostic indi-
biopsy technique. In case of persistent bleeding, electroco- cations are biopsies from the diaphragm, the lung, the
agulation may be necessary. In our experience of almost mediastinum and the pericardium, but today there is a
7000 thoracoscopies, surgical intervention was never nec- definite trend towards its use almost exclusively in pleural
essary to stop bleeding caused by medical thoracoscopy. effusions and pneumothorax.
The most serious complication of pneumothorax induc- The changed indications for medical thoracoscopy are
tion is air or gas embolism, which occurs very rarely (<0.1 reflected by the development at our own institution,
percent), provided that the necessary precautionary meas- Lungenklinik Heckeshorn in Berlin/Germany.15 Pleural
ures are taken.14 effusions are by far the most frequent application of
The reported mortality rates (<0.01 percent) are very medical thoracoscopy (over 90 percent). A decline in the
low.38,44 Even several liters of fluid can be completely other indications is explained by several factors. Medical
removed during thoracoscopy with little risk of pulmonary thoracoscopy for diagnostic purposes in localized lung and
re-expansion edema because immediate equilibration of chest wall lesions has been almost abandoned since
pressures is provided by direct entrance of air through the imaging techniques such as computed tomography (CT)
cannula into the pleural space. If the re-expansion poten- or magnetic resonance imaging (MRI) very often deliver
tial of the lung appears to be diminished, at most only low- the diagnosis or allow differentiation between malignant
pressure suction should be applied through the and benign disease. Furthermore, VATS or surgical thora-
intrapleural drainage tube, which is always placed at the coscopy has the advantage that the lesion can be removed
end of thoracoscopy. Following lung biopsy, a bron- simultaneously. Our indications for lung biopsies in
chopleural fistula may result. This may require longer than diffuse lung disease have decreased, too. This is due to the
the usual suction periods of 3–5 days, particularly in cases improved diagnostic results of bronchoscopy using trans-
with stiff lungs. After talc poudrage, fever may occur. Local bronchial lung biopsies and bronchoalveolar lavage, as
site infection is, however, uncommon, and empyema has well as to the development of high-resolution CT (HRCT),
been reported only very rarely,44 often as sequelae of long- which has considerably improved the diagnostic evalua-
term drainage with bronchopleural fistula.20 tion of diffuse lung diseases. At our institution, for many
In cases of mesothelioma, radiotherapy should be years, the department of thoracic surgery has been respon-
carried out after medical thoracoscopy to prevent the late sible for almost all patients with spontaneous pneumo-
complications of tumor growth at the sites of entry:45 thorax. However, the thoracic surgeons today routinely
10–12 days after thoracoscopy, once the incision has apply medical thoracoscopy at the time of chest-tube
healed, Boutin and co-workers in a randomized trial insertion through the cannula, under local anesthesia, in
applied electron therapy between 12.5 and 15 MeV to the their cases with spontaneous pneumothorax.47 They also
588 Medical thoracoscopy
use this approach frequently for cases of empyema, where percent of the whole group. In 15 percent of the patients,
the medical thoracoscopic technique is applied at the time the diagnosis of non-specific pleuritis after thoracoscopy
of chest-tube placement.48 appeared to be false-negative. The authors calculated a
The absolute number of pleural effusions at our institu- positive predictive value of 100 percent and a negative pre-
tion has risen markedly because more and more cases of dictive value of 92 percent. In another retrospective study
pleural effusion are referred for diagnostic and therapeutic of 75 patients diagnosed with non-specific pleuritis after
purposes, more than two-thirds are malignant effusions. thoracoscopy, follow-up revealed a malignant origin in 8.3
The number of tuberculous pleurisies fell considerably due percent (three cases with lung cancer, two cases with
to the decrease in incidence, and pleural effusions of other mesothelioma).55 The authors conclude that a histological
origin now comprise about 25 percent. Here, medical tho- diagnosis of non-specific pleuritis does not necessarily
racoscopy was performed mainly for differential diagnosis, imply a clinical diagnosis of idiopathic pleuritis. They rec-
in particular to exclude malignancy or tuberculosis (TB), ommend a close follow-up, particularly in patients with a
in a few cases also for talc pleurodesis in cases of recurrent history of asbestos exposure. Autofluorescence videotho-
pleural effusion of non-malignant etiology. If one takes the racoscopy may help in the future to avoid some of the
estimate of Light for the annual incidences of 750 000 non- false-negative results.56
cardiac and non-parapneumonic pleural effusions in the In comparison, even in a study on thoracotomy in
USA, one would, by that rationale, expect about 6000 patients with pleural effusion of undetermined etiology
pleural effusions annually in our catchment area, which after a pathological diagnosis of a benign pleural process,
comprises mainly the western part of Berlin with approxi- 25 percent of patients were still diagnosed with a malig-
mately two million inhabitants. Of these, approximately nancy within 6 months.57
2000 would be of malignant origin, and less than 5–10 Because of its high diagnostic accuracy, we perform
percent of these would finally undergo medical thora- diagnostic medical thoracoscopy in almost all cases of
coscopy.5 exudates in which the etiology remains undetermined
after pleural fluid analysis. Due to the low additional
diagnostic yield of closed blind needle biopsy, we recom-
PLEURAL EFFUSIONS mend medical thoracoscopy after inconclusive diagnostic
work-up of the pleural fluid if the facilities for medical
Conservative estimates suggest that approximately 25 thoracoscopy are available.19 The procedure allows fast
percent of the cases seen in a general pulmonologist’s prac- and more definite biopsy diagnosis, including a high yield
tice demonstrate pleural effusions.49,50 In approximately in tuberculosis cultures, and determination of hormone
25 percent of these cases, a specific diagnosis cannot be receptors in some malignancies. Furthermore, staging in
made, even after thoracentesis and closed pleural biopsy.51 lung cancer and diffuse mesothelioma is possible. The
In a series by Boutin et al.32 of 1000 consecutive patients exclusion of malignancy or TB is provided with high
with pleural effusion, 215 cases remained undiagnosed probability. Furthermore, medical thoracoscopy allows
even after repeated pleural fluid analysis and performance complete fluid removal and evaluation of the re-expan-
of pleural biopsies. This is in agreement with the results of sion potential of the lung. Fibrinous loculations in tuber-
several other authors, who report that without thora- culosis and empyema are easily removed, thus creating a
coscopy, at least 20–25 percent of pleural effusions remain single pleural cavity, which allows more efficient treat-
undiagnosed, although this certainly depends strongly on ment.5
patient populations.
Several studies have tried to determine the diagnostic
accuracy of medical thoracoscopy in the setting of undiag- MALIGNANT PLEURAL EFFUSIONS
nosed pleural effusion, but the results vary widely, with a
range of 69 to 90 percent. Closer evaluation of the study Malignant pleural effusions are today the leading diagnos-
designs reveals that the duration of follow-up was occa- tic and therapeutic indication for medical thora-
sionally short and frequently not mentioned at all. One coscopy.27,28,58,59 In a prospective intrapatient comparison,
well-designed study reported by Menzies52 with follow-up the diagnostic yield of non-surgical biopsy methods in
periods between 1 and 2 years found a sensitivity of 91 malignant pleural effusions was studied simultaneously in
percent, specificity of 100 percent, accuracy of 96 percent, 208 patients, including 58 diffuse malignant mesothe-
and negative predictive value of 93 percent. Boutin et al.32 liomas, 29 cancers of the lung, 116 metastatic pleural effu-
reported a false-negative rate of 15 percent within 1 year of sions with 28 breast cancers, 30 cancers of various other
follow-up, while Enk and Viskum53 reported a diagnostic organs, 58 of undetermined origin and 5 malignant lym-
accuracy of 69 percent with a 5-year follow-up period. A phomas.60 The diagnostic yield was 62 percent by pleural
retrospective study of 709 patients with pleural effusion fluid cytology, 44 percent by closed pleural biopsy and 95
showed that 29 percent remained inconclusive after thora- percent by medical thoracoscopy (Table 47.1). The sensi-
coscopy.54 After long-term follow-up, a malignancy was tivity of medical thoracoscopy was higher than that of
found in 4.3 percent and a true benign disease in 24.7 cytology and closed pleural biopsy combined (95 versus 74
Malignant pleural effusions 589
Table 47.1 Sensitivity (%) of non-surgical biopsy methods in scopies in 45 patients with lung cancer and pleural effu-
malignant pleural effusions. Prospective simultaneous comparison sion. In 37, they found pleural invasion; three patients had
(n = 206) mediastinal disease, the remaining five had no evident
metastatic disease, and, therefore, no contraindication for
Method Sensitivity (%)
tumor resection. Canto et al.63 found no thoracoscopic
evidence of pleural involvement in 8 of 44 patients, six
Fluid cytology (FC) 62 proceeded to resection with no pleural involvement
Closed needle (CN) 44 found.63
FC + CN 74 In diffuse malignant mesothelioma, medical thora-
Medical thoracoscopy (MT) 95 coscopy can provide an earlier diagnosis and better histo-
MT + FC 96 logic classification than closed pleural biopsy because of
MT + FC + CN 97 larger and more representative biopsies, and more accu-
Taken from Loddenkemper et al.60 rate staging.64,65 This may have important therapeutic
implications, since much better responses to local
immunotherapy or local chemotherapy have been
observed in the early stages (I and II).66–68 The technique is
percent, p > 0.001). The combined non-surgical methods also helpful in the diagnosis of benign asbestos-related
were diagnostic in 97 percent of malignant pleural effu- pleural effusion (BAPE) by excluding mesothelioma or
sions. In six of the 208 cases (2.8 percent), an underlying malignancies. Fibrohyaline or calcified, thick and pearly
neoplasm was suspected at thoracoscopy, but confirmed white pleural plaques may be found, indicating possible
only by thoracotomy or autopsy. Similar results have been asbestos exposure.68 Thoracoscopic pulmonary biopsies
reported by other investigators.32,52,61 and even biopsies from special lesions on the parietal
The reasons for false-negative results of medical thora- pleura may demonstrate high concentrations of asbestos
coscopy include insufficient and non-representative biop- fibers and thereby provide further support for a diagnosis
sies that depend largely on the experience of the of asbestos-induced disease.69
thoracoscopist and the presence of adhesions preventing A further advantage of medical thoracoscopy in
access to the neoplastic tissue.28,32 An additional reason metastatic pleural disease is that biopsies of the visceral
could be that early lesions are not detected by normal, but and diaphragmatic pleura are possible under direct obser-
only by autofluorescence light.56 vation. In addition, because of the larger size of thoraco-
The diagnostic sensitivity of medical thoracoscopy is scopic biopsies, these may provide easier identification of
similar for all types of malignant effusion (Table 47.2). The primary tumor, including hormone receptor determina-
overall yield in 287 cases was 62 percent for cytology and tion in breast cancer,70,71 and improved morphological
95 percent for medical thoracoscopy; the yield for cytology classification in lymphomas.72 In addition, the extent of
and in particular medical thoracoscopy did not vary intrapleural tumor spread can be semiquantified using a
greatly between lung carcinomas (n = 67) with 67 versus scoring system which has been shown to correlate quite
96 percent, extrathoracic primaries (n = 154) with 62 closely with survival,73 and with the success of talc pleu-
versus 95.5 percent, and diffuse malignant mesotheliomas rodesis.74
(n = 66) with 58 versus 92 percent.5,19 The main advantage in using thoracoscopy to diagnose
Medical thoracoscopy may be useful in staging lung malignancy is that talc poudrage can be performed during
cancer, diffuse malignant mesothelioma (Figure 47.4) and medical thoracoscopy (Figures 47.7 and 47.8), which is
metastatic cancer (Figures 47.7 and 47.8). In lung cancer probably the best conservative option today for pleurode-
patients, thoracoscopy can help to determine whether the sis, with success rates of more than 80 percent.28,75 These
effusion is malignant or paramalignant.28 As a result, it high success rates may be due to the more even distribu-
may be possible to avoid exploratory thoracotomy for tion of talc powder to all parts of the pleura. It has also
tumor staging. Weissberg et al.62 performed thoraco- proven very efficient in the treatment of lymphomatous
Table 47.2 Sensitivity (%) of medical thoracoscopy and pleural fluid cytology in different malignant etiologies. Prospective simultaneous
comparison at Lungenklinik Heckeshorn, Berlin, Germany
chylothorax, even after all other treatment options have procedures or are related to talc per se has not been deter-
failed.76 mined.19 Acute respiratory distress syndrome (ARDS),
acute pneumonitis and respiratory failure have also been
reported to occur after both talc poudrage and
TALC POUDRAGE slurry.84,86,87 The development of respiratory failure most
likely is related to dose and, in particular, particle size.88–90
Talc poudrage is the most widely reported method of talc It is remarkable that several large series from Europe and
instillation into the pleural space.19 It can be easily per- from Israel did not observe ARDS after thoracoscopic talc
formed at medical thoracoscopy (Figure 47.3) under local insufflations.28,75,91–93 This was also found in a large multi-
anesthesia with some additional pain medication, if neces- center, open-label, prospective cohort study of 558
sary. patients with malignant pleural effusions who underwent
As a prerequisite for successful pleurodesis, all pleural medical thoracoscopy and talc poudrage with 4 g of cali-
fluid must be removed before spraying talc. Removal can brated French graded large-particle talc. No patient devel-
be easily accomplished during thoracoscopy, as air is oped acute respiratory distress syndrome, proving that the
entering the pleural cavity, thus creating the desired equi- use of large-particle talc for pleurodesis is safe.77,94
librium in pressures. Complete collapse of the lung is In vitro studies suggest that talc induces apoptosis in
desirable, because it permits wide and uniform distribu- human malignant mesothelioma cells and might therefore
tion of the talc.28,75 have an anti-tumorous effect.95 One explanation may be
The optimal dose of talc for poudrage is not known, but that talc mediates angiostasis in malignant pleural effusion
usually about 4–6 g (8–12 mL) are recommended for via endostatin induction.96
malignant effusions.39,77 Thoracoscopic inspection of the
pleural cavity should be done during talc insufflation to
ensure that talc is uniformly distributed over pleural sur- TUBERCULOUS PLEURAL EFFUSIONS
faces (Figures 47.7 and 47.8).
After talc poudrage, an 8–11 mm (24–30 French) chest Although the diagnostic yield of pleural fluid TB culture
tube should always be inserted. Suction should be applied and of closed needle biopsy combined is quite high, there
carefully and progressively to avoid creation of air leaks may be indications for medical thoracoscopy in otherwise
which can be caused by necrotic tissue in the visceral uncertain pleural effusions.5 The diagnostic accuracy of
pleura. The chest tube can be removed when the daily thoracoscopy is almost 100 percent because the patholo-
amount of fluid production is lower than 100 mL. gist is provided with multiple, selected biopsies and
Pleurodesis success rates with talc poudrage are usually because the cultural proof of tubercle bacilli growth is
above 80 percent, but there are only few prospective more frequent.
studies comparing talc with other pleurodesis agents.78–83 In a prospective intrapatient comparison, the immedi-
The advantage of talc poudrage compared with slurry is ate diagnosis in 100 TB cases was established histologically
probably the more even distribution over the whole by medical thoracoscopy in 94 percent, compared with
pleural surface, which can be achieved under visual only 38 percent with needle biopsy.97 This may be of clin-
control.75 In our own experience, we observe fewer locula- ical importance, because anti-tuberculous chemotherapy
tions than with talc slurry. In a large randomized trial of can be started without delay. The combined yield of histol-
talc poudrage versus talc slurry with 501 patients, an ogy and bacteriological culture was 99 percent for medical
advantage of poudrage was seen only in those patients thoracoscopy and 51 percent for needle biopsy, and 61
(whether they lived 30 days or not) who had more than 90 percent when culture results from effusions were added
percent lung re-expansion (78 versus 71 percent efficacy of (Table 47.3). The percentage of positive TB cultures was
sclerosis with prevention of recurrence, p = 0.045).84 If the twice as high from thoracoscopic biopsies, including cul-
two most common etiologies (breast and lung cancer) for tures from fibrinous membranes (78 percent), as the per-
malignant pleural effusion were examined, there was also a centage from pleural fluid and needle biopsies combined
statistical difference favoring poudrage (p = 0.022) in (39 percent), allowing bacteriological confirmation of the
patients living 30 days with more than 90 percent re- diagnosis and, furthermore, susceptibility tests. In 5 of the
expansion. The authors concluded that prevention of 78 positive cases (6.4 percent), resistance to one or multi-
recurrence of malignant pleural effusions from breast or ple anti-tuberculous drugs was found, which influenced
lung cancers can be achieved with a success rate of 82 therapy and prognosis. The chance for positive TB cultures
percent with poudrage versus 67 percent with slurry. was much higher (78 percent) in cases with fibrin produc-
Talc is inexpensive and highly effective. Its most tion. This type, with a diffusely thickened pleura, multiple
common short-term adverse effects include fever, due to adhesions, and sometimes formation of encapsulated
systemic inflammatory reaction, 85 and pain. Cardio- membranes with fluid loculations, was present in 75
vascular complications such as arrhythmias, cardiac arrest, percent of cases (Figure 47.9). By comparison, the typical
chest pain, myocardial infarction or hypotension have picture of sago-like early pleuritis with miliary tuberculous
been noted; whether these complications result from the granulomas and without fibrin layers was seen in only 25
Tuberculous pleural effusions 591
Table 47.3 Sensitivity (%) of non-surgical biopsy methods in tuberculous pleural effusions (histological
and bacteriological results). Prospective simultaneous comparison (n = 100)
Effusion (E) – 28 28
Closed needle (N) 38 25 51
E+N 38 39 61
Medical thoracoscopy (MT) 94 76 99
MT + N 95 76 99
MT + E 94 78 100
Figure 47.9 Tuberculous pleural effusion with fibrinous Figure 47.10 Tuberculous pleural effusion with typical sago-
adhesions between the lung and the chest wall. (See also Color like nodules on the inflamed pleura. (See also Color Plate 54.)
Plate 53.)
percent (Figure 47.10). Here, positive TB cultures were ment of TB pleurisy, the authors found that the initial
obtained from all materials in only 50 percent, giving a complete drainage of the effusion, performed during and
highly significant difference (p > 0.0005). Interestingly, after thoracoscopy, was associated with greater sympto-
this study also showed that the chance of positive TB cul- matic improvement than any subsequent therapy.99 No
tures from pleural effusion alone was statistically much studies are known which compare the influence of thora-
better in cases with a low pleural glucose level (<50 mg per coscopy with its early diagnosis and complete drainage and
dL), indicating an increased metabolism by TB bacilli subsequent early drug treatment to a group with drug
and/or higher degree of inflammation (59 percent positive treatment alone. In our institution, Lungenklinik
versus 25 percent with glucose levels above 50 mg per dL, Heckeshorn, no single case has been observed during the
p > 0.005). last decade which needed decortication because of devel-
In another prospective study, medical thoracoscopy in opment of fibrothorax, perhaps a result of routine medical
40 cases from South Africa had a diagnostic yield of 98 thoracoscopy with evacuation of all pleural fluid and
percent in comparison with an 80 percent diagnostic yield opening of intrapleural loculations during medical thora-
with Abram’s needle biopsies.98 This led to the conclusion coscopy.5 In a recent study from the same institution in
that in areas with a high prevalence of TB, Abram’s closed South Africa, 51 patients with undiagnosed exudative
needle biopsy (in this case three biopsies were obtained pleural effusion were recruited for a prospective, direct
and each examined histologically and microbiologically) comparison between bronchial wash, pleural fluid micro-
can contribute significantly to a diagnosis. However, in a biology, and biochemistry (adenosine deaminase [ADA]
further study on the effect of corticosteroids in the treat- and cell count), closed needle biopsy and medical thora-
592 Medical thoracoscopy
coscopy.100 The final diagnosis was TB in 41 patients (82 nosis of benign asbestos-related pleural effusions, which,
percent). Sensitivity of histology, culture and combined by definition, are a diagnosis of exclusion.68
histology/culture was 66, 48 and 79 percent, respectively, In other pleural effusions, when the origin is unknown,
for closed needle biopsy and 100, 76 and 100 percent the main diagnostic value of medical thoracoscopy lies in
respectively, for thoracoscopy. Since the combination of its ability to exclude, with high probability, malignant and
ADA, lymphocyte/neutrophil ratio ≥0.75 plus closed tuberculous disease.5,59 By means of thoracoscopy, the
needle biopsy reached 93 percent sensitivity and 100 proportion of so-called idiopathic pleural effusions usually
percent specificity, the authors concluded that this high falls markedly below 10 percent, whereas studies which
diagnostic accuracy is sufficient in areas with a high inci- have not used thoracoscopy report failure to obtain a diag-
dence of tuberculosis. However, if this test is negative nosis in over 20 percent.51 However, this certainly also
despite a high clinical suspicion of tuberculous pleurisy, if depends on the selection of patients and on the definition
antibiotic resistance is of concern or if other possible diag- of ‘idiopathic’.54,55
noses are considered, they recommend medical thora- It is occasionally impossible to perform thoracoscopy in
coscopy as the method of choice. the presence of effusion because of dense pleuropul-
It is our policy not to treat patients with anti-tubercu- monary adhesions. In these cases, multiple closed needle
lous drugs merely on the suspicion of tuberculous pleurisy. biopsies should be performed, or more invasive surgical
At least in countries with a low prevalence of TB, where procedures should be considered.4
even other laboratory tests may not be very distinctive, In some selected cases of recurrent pleural effusion of
medical thoracoscopy should be performed when needle non-malignant etiology including chylothorax, pleurode-
biopsies show negative histological results, in order to sis may be induced by applying talc poudrage during
prove or exclude TB. In addition, the high yield in positive medical thoracoscopy.75,76,105
TB cultures from thoracoscopic biopsies gives rise to the
possibility of obtaining susceptibility tests which, in partic-
ular in patients born in countries with high rates of multi-
drug resistance (MDR) or extensive drug resistance EMPYEMA
(XDR),101 may have a considerable impact on treatment
and prognosis. In a study from Japan using the semi-rigid Medical thoracoscopy can also be used in the management
thoracoscope under local anaesthesia, 30 out of 32 patients of early empyema.48,106–108 In cases with multiple locula-
(94 percent) were successfully diagnosed by histology of tions, it is possible to open these spaces to remove the fib-
the pleural biopsy, whereas bacteriological examination of rinopurulent membranes (Figure 47.11) by forceps and to
the pleural fluid was positive only in 11 cases (eight create one single cavity, which can be drained and irrigated
culture-positive, three PCR-positive).102 Twenty-one of much more successfully. This was demonstrated in a retro-
the cases (66 percent) were diagnosed by histology alone. spective study of 127 patients of whom 94 percent were
The low yield of bacteriological confirmation (compared treated successfully for multiloculated empyema. Only 6
with the results of the other studies mentioned97,98,100) is percent of the patients required a surgical approach.108
explained by the fact that biopsy specimens were not taken This treatment should be carried out early in the course of
for bacteriological examination. The finding of only five parapneumonic effusion/empyema, before the adhesions
positive thoracoscopic TB cultures was surprisingly low
compared with the results of the other studies.97,98,100
become too fibrous and adherent. Thus, if the indication adhesions; stage III with small bullae and blebs (<2 cm in
for placement of a chest tube is present and if the facilities diameter); and stage IV with numerous large bullae
are available, medical thoracoscopy should be performed (>2 cm in diameter).113 In 1047 cases where medical tho-
at the time of chest-tube insertion. For patients with com- racoscopy was used by three different teams,114–116 patho-
plicated parapneumonic effusions, morbidity is lower in logical lesions were detected in approximately 70 percent
those who are treated with thoracoscopy or VATS than in of the cases with only slightly differing percentages for
those who received tube thoracostomy.109 The consensus is stages II to IV. Blebs and bullae were present in 45–62
to perform thoracoscopy earlier in patients with compli- percent. A false classification of stage I was shown at
cated parapneumonic effusions in whom tube thoracos- surgery in 8–28 percent of cases. Although the detection
tomy and fibrinolytic therapy have failed.110 Overall, rates of blebs and bullae are higher (76–100 percent) in a
medical thoracoscopy is a procedure similar to chest-tube series with VATS or thoracotomy,111 it is unlikely that
placement, but enables the creation of one single pleural larger bullae and blebs or fistulae would not be detected
cavity, allowing much better local treatment.48 However, during medical thoracoscopy. However, fluorescein-
prospective studies on the use of medical thoracoscopy in enhanced autofluorescence thoracoscopy detected more
the treatment of early empyema have not yet been carried abnormalities compared with white light thoracoscopy.117
out. Medical thoracoscopy offers the possibility to combine
chest drainage with coagulation of blebs and bullae as well
as pleurodesis by talc poudrage.39,109 Talc poudrage
achieves the best conservative treatment results with recur-
SPONTANEOUS PNEUMOTHORAX rence rates below 10 percent.39,118 Subsequent surgical
intervention was necessary in 4–10 percent of the cases in
In spontaneous pneumothorax, medical thoracoscopy can the three above-mentioned series.114–116 In stage IV with
be applied easily for diagnostic and therapeutic purposes, numerous large bullae, VATS or thoracotomy is usually
if the skills and facilities for this technique are avail- indicated. These patients should be transferred directly to
able.20,28,39,111,112 In particular, if a chest tube is introduced the surgical department after chest-tube insertion. Talc
by the trocar technique, it is easy to use an optique for poudrage and/or coagulation of bullae are performed only
visual inspection of the lung and pleural cavity, before in cases where surgery is contraindicated (e.g. because of
insertion of the chest tube through this cannula. On respiratory insufficiency, secondary to severe bronchitis or
inspection during medical thoracoscopy, the underlying other advanced pulmonary diseases).119 Talc poudrage
lesions can be directly assessed (Figure 47.12) according to under thoracoscopy is safe, even in patients with advanced
the classification of Vanderschueren: stage I with an endo- chronic obstructive pulmonary disease (COPD),120 con-
scopically normal lung; stage II with pleuropulmonary firming previous studies.112
In our view, medical thoracoscopy is routinely justified
in all patients with spontaneous pneumothorax where
tube drainage is indicated, since several advantages are
offered: precise assessment of underlying lesions under
direct visual control, choice of best (conservative or surgi-
cal) treatment measures,47 direct treatment by coagulation
of blebs and bullae, and by severing of adhesions, if neces-
sary, followed by talc poudrage, as well as selection of the
best location for chest-tube placement.28 In the case of
recurrent or persistent pneumothorax, simple talc
poudrage under medical thoracoscopy has been shown to
be safe, cost-effective and no more painful than conserva-
tive treatment using a chest tube.40
For talc poudrage in pneumothorax patients, a mere
2–4 mL of talc are sufficient for effective pleurodesis.39 No
serious short-term complications have been seen if graded
large-particle talc is used.90,94,112 No long-term sequelae
were observed 22–35 years after talc poudrage of pneumo-
thorax: total lung capacity (TLC) averaged 89 percent pre-
dicted in 46 patients, whereas TLC was 97 percent
predicted in 29 patients treated with tube thoracostomy
alone.121 None of the poudrage group developed mesothe-
Figure 47.12 Thoracoscopic view of an emphysematous bulla in lioma over the 22- to 35-year follow-up. Although talc
the apex of the right upper lobe (stage IV). (See also Color Plate poudrage may result in minimally reduced TLC, as well as
56.) pleural thickening on chest radiography, these changes
594 Medical thoracoscopy
◆3. Mathur PN, Boutin C, Loddenkemper R. ‘Medical’ thoracoscopy: ◆28. Loddenkemper R, Boutin C. Thoracoscopy: present diagnostic and
Technique and indications in pulmonary medicine. J Bronchol therapeutic indications. Eur Respir J 1993; 6: 1544–55.
1994; 1: 228–39. ◆29. Maiwand MO. Video-assisted thoracoscopic surgery: Current
◆4. Loddenkemper R, McKenna RJ. Pleuroscopy, thoracoscopy and applications, advantages, and limitations (review). J Bronchol
other invasive procedures. In: Mason, RJ, Broaddus, VC, Murray, JF, 1997; 4: 321–8.
Nadel, JA. (eds). Textbook of respiratory medicine. Philadelphia: 30. Lewis JW Jr. Thoracoscopy: a surgeon’s or a pulmonologist’s
Elsevier Saunders, 2005: 651–70. domain; pro pulmonologist. J Bronchol 1994; 1: 152–4.
◆5. Loddenkemper R. Thoracoscopy – state of the art. Eur Respir J 31. Soffer A. Who should perform thoracoscopy? Chest 1992; 102:
1998; 11: 213–21. 1553–5.
6. Dijkman JH, Martinez Gonzales del Rio J, Loddenkemper R, Prowse ●32. Boutin C, Viallat JR, Cargnino C, Farisse P. Thoracoscopy in
K, Siafakas N. Report of the working party of the ‘UEMS malignant pleural effusions. Am Rev Respir Dis 1981; 124: 588–92.
monospecialty section on pneumology’ on training requirements ●33. Janssen J, Boutin C. Extended thoracoscopy: a method to be used
and facilities in Europe. Eur Respir J 1994: 1019–22. in case of pleural adhesions. Eur Respir J 1992; 5: 763–6.
7. Tape TG, Blank LL, Wigton RS. Procedural skills of practicing ●34. Tassi G, Marchetti G. Minithoracoscopy. A less invasive approach
pulmonologists: a national survey of 1,000 members of the to thoracoscopy. Chest 2003; 124: 1975–7.
American College of Physicians. Am J Respir Crit Care Med 1995; ◆35. Mathur PN. How I do it. ‘Medical’ thoracoscopy. J Bronchol 1994;
151: 282–7. 2: 144–51.
8. Pastis NJ, Nietert PJ, Silvestri GA. Variation in training for ●36. Ernst A, Hersh CP, Herth F, et al. A novel instrument for the
interventional pulmonary procedures among US pulmonary/critical evaluation of the pleural space: an experience in 34 patients.
care fellowships – a survey of fellowship directors. Chest 2005; Chest 2002; 122: 1530–4.
127: 1614–8. 37. Munavvar M, Khan MAI, Edwards J, Waqaruddin Z, Mills J. The
9. Lee P, Lan RS, Colt HG. Survey of pulmonologists’ perspectives on autoclavable semirigid thoracoscope: the way forward in pleural
thoracoscopy. J Bronchol 2003; 10: 99–106. disease? Eur Respir J 2007; 29: 571–4.
10. Burrows NJ, Ali NJ, Cox GM. The use and development of medical 38. Loddenkemper R. Thoracoscopy under local anesthesia. Is it safe?
thoracoscopy in the United Kingdom over the past 5 years. Respir (Editorial) J Bronchol 2000; 7: 207–9.
Med 2006; 7: 1234–8. ◆39. Boutin C, Astoul P, Rey F, Mathur PN. Thoracoscopy in the
◆11. Seijo LM, Sterman DH. Interventional pulmonology. N Engl J Med diagnosis and treatment of spontaneous pneumothorax. Clin Chest
2001; 344: 740–9. Med 1995; 16: 497–503.
12. Piquet A, Giraud A. La pleuroscopie et la section des adhérences 40. Tschopp J-M, Boutin C, Astoul P, et al. Talcage by medical
intrapleural au cours du pneumothorax thérapeutique. La presse thoracoscopy for primary spontaneous pneumothorax is more
médicale 1923; 23. cost-effective than drainage: a randomised study. Eur Respir J
13. Hoksch B, Birken-Bertsch H, Müller JM. Thoracoscopy before 2002; 20: 1003–9.
Jacobaeus. Ann Thorac Surg 2002; 74: 1288–90. ◆41. Colt HG. Thoracoscopy: a prospective study of safety and outcome.
◆14. Moisiuc FV, Colt HG. Thoracoscopy: origins revisited. Respiration Chest 1995; 108: 324–9.
2007; 74: 455–55. 42. Faurschou P, Madsen F, Viskum K. Thoracoscopy: influence of the
◆15. Loddenkemper R. Medical thoracoscopy – historical perspective. procedure on some respiratory and cardiac values. Thorax 1983;
In: Beamis JF, Mathur P, Mehta A (eds.). Interventional pulmonary 38: 341–3.
medicine. New York: Marcel Dekker, Inc, 2004: 411–29. 43. Chhajed PN, Kaegi B, Rajasekaran R, Tamm MD. Detection of
◆16. Jacobaeus HC. Die Thorakoskopie und ihre praktische Bedeutung. hypoventilation during thoracoscopy. Combined cutaneous carbon
Ergebn ges Med 1925; 7: 112–66. dioxide tension and oximetry monitoring with a new digital
●17. Sattler A. Zur Behandlung des Spontanpneumothorax mit sensor. Chest 2005; 127: 585–8.
besonderer Berücksichtigung der Thorakoskopie. Beitr Klin Tuberk ◆44. Viskum K, Enk B. Complications of thoracoscopy. Poumon Cœur
1937; 89: 395–408. 1981; 37: 25–8.
●18. Roche G, Delanoe Y, Moayer N. Talcage de la plèvre sous ●45. Boutin C, Rey F, Viallat JR. Prevention of malignant seeding after
pleuroscopie. Résultats, indications, technique (A propos de 14 invasive diagnostic procedures in patients with pleural
observations). J Fr Med Chir Thorac 1963; 21: 177–95. mesothelioma. A randomized trial of local radiotherapy. Chest
∗19. Antony VB, Loddenkemper R, Astoul P, et al. Management of 1995; 108: 754–8.
malignant pleural effusions (ERS/ATS Statement). Am J Respir Crit 46. Low EM, Khoury GG, Matthews AW, Neville E. Prevention of
Care Med 2000; 162: 1987–2001. Eur Respir J 2001; 18: 402–19. tumour seeding following thoracoscopy in mesothelioma by
◆20. Brandt HJ, Loddenkemper R, Mai J. Atlas of diagnostic prophylactic radiotherapy. Clin Oncol 1995; 7: 317–8.
thoracoscopy. Indications – Technique. New York: Thieme Inc, 47. Kaiser D. Indikationen zur chirurgischen Therapie beim
1985. Spontanpneumothorax. Chir Praxis 2000; 57: 239–48.
◆21. Boutin C, Viallat JR, Aelony Y. Practical thoracoscopy. Berlin: 48. Kaiser D. Indikationen zur Thorakoskopie beim Pleuraempyem.
Springer, 1991. Pneumologie 1989; 43: 76–9.
◆22. Beamis JF, Mathur P, Mehta A. (eds.). Interventional pulmonary 49. Collins TR, Sahn SA. Thoracentesis: Clinical value, complications,
medicine. New York: Marcel Dekker, Inc, 2004. technical problems, and patient experience. Chest 1987; 91:
◆23. Buchanan DR, Neville E. (eds.). Thoracoscopy for physicians. A 817–22.
practical guide. London: Arnold Hodder, 2004. ◆50. Light RW. Diagnostic principles in pleural disease. Eur Respir J
◆24. Lee P, Colt HG. Flex-rigid pleuroscopy. Step-by-step. Singapore: 1997; 10: 476–81.
CMPMedica Asia Pte Ltd, 2005. ◆51. Storey DD, Dines DE, Coles DT. Pleural effusion. A diagnostic
25. LoCicero J. Minimally invasive thoracic surgery, video-assisted dilemma. J Am Med Assoc 1976; 236: 2183–6.
thoracic surgery and thoracoscopy (Editorial). Chest 1992; 102: 52. Menzies R. Charbonneau M. Thoracoscopy for the diagnosis of
330–1. pleural disease. Ann Intern Med 1991; 114: 271–6.
26. Faber LP. Thoracoscopy: A surgeon’s or a pulmonologist’s domain. 53. Enk B, Viskum K. Diagnostic thoracoscopy. Eur J Respir Dis 1981;
Pro surgeon. J Bronchol 1994; 1: 155–9. 114: 271–6.
◆27. Harris RJ, Kavuru MS, Rice TW, Kirby TI. The diagnostic and 54. Janssen JP, Ramlala S, Mravunac M. The long-term follow up of
therapeutic utility of thoracoscopy. A review. Chest 1995; 108: exudative pleural effusion after nondiagnostic thoracoscopy. J
828–41. Bronchol 2004; 11: 169–74.
596 Medical thoracoscopy
55. Venekamp LN, Velkeniers B, Noppen M. Does ‘idiopathic pleuritis’ talc poudrage in malignant pleural effusion: a prospective cohort
exist? Natural history of non-specific pleuritis diagnosed after study. Lancet 2007; 369: 1535–9.
thoracoscopy. Respiration 2005; 72: 74–8. 78. Boutin C, Rey F, Viallat JR. Etude randomisée de l’efficacité du
●56. Chrysanthidis MG, Janssen JP. Autofluorescence talcage thoracoscopique et de l’instillation de tétracycline dans le
videothoracoscopy in exudative pleural effusions: preliminary traitement des pleurésies cancéreuses récidivantes. Rev Mal Resp
results. Eur Respir J 2005; 26: 989–92. 1985; 2: 374–8.
●57. Ryan CJ, Rodgers RF, Uni UK, Hepper NG. The outcome of patients 79. Fentiman IS, Rubens RD, Hayward JL. A comparison of
with pleural effusion of indeterminate cause at thoracotomy. intracavitary talc and tetracycline for the control of pleural
Mayo Clin Proc 1981; 56: 145–9. effusions secondary to breast cancer. Eur J Cancer Clin Oncol
●58. Lee P, Colt HG. Using diagnostic thoracoscopy to optimal effect. J 1986; 22: 1079–81.
Respir Dis 2003; 24: 503–9. 80. Hartmann DL, Gaither JM, Kesler KA, et al. Comparison of
●59. Rodriguez-Panadero F, Janssen JP, Astoul P. Thoracoscopy: general insufflated talc under thoracoscopic guidance with standard
overview and place in the diagnosis and management of pleural tetracycline and bleomycin pleurodesis for control of malignant
effusion. Eur Respir J 2006; 28: 409–21. pleural effusions. J Thorac Cardiovasc Surg 1993; 105: 743–8.
●60. Loddenkemper R, Grosser H, Gabler A, et al. Prospective evaluation 81. Yin AC, Chan AT, Lee TW, Wan IY, Ho JK. Thoracoscopic talc
of biopsy methods in the diagnosis of malignant pleural effusions. insufflation versus talc slurry for symptomatic malignant pleural
Intrapatient comparison between pleural fluid cytology, blind effusion. Ann Thorac Surg 1996; 62: 1655–8.
needle biopsy and thoracoscopy. Am Rev Respir Dis 1983; 127 82. Diacon AH, Wyser C, Bolliger CT, et al. Prospective randomized
(Suppl. 4): 114. comparison of thoracoscopic talc poudrage under local anesthesia
61. Oldenburg FA Jr, Newhouse MT. Thoracoscopy. A safe, accurate versus bleomycin instillation for pleurodesis in malignant pleural
diagnostic procedure using the rigid thoracoscope and local effusions. Am J Respir Crit Care Med 2000; 162: 1445–9.
anesthesia. Chest 1979; 75: 45–50. ◆83. Tan C, Sedrakyan A, Browne J, Swift S, Treasure T. The evidence on
●62. Weissberg D, Kaufmann M, Schwecher I. Pleuroscopy in clinical the effectiveness of management for malignant pleural effusion: a
evaluation and staging of lung cancer. Poumon Coeur 1981; 37: systematic review. Eur J Cardiothorac Surg 2006; 29: 829–38.
241–3. ●84. Dresler CM, Olak J, Herndon JE, et al. Phase III intergroup study of
63. Canto A, Ferrer G, Romagosa V, Moya J. Bernat R. Lung cancer and talc poudrage vs talc slurry sclerosis for malignant pleural
pleural effusion. Clinical significance and study of pleural effusion. Chest 2005; 127: 909–15.
metastatic locations. Chest 1985; 87: 649–52. 85. Froudarakis ME, Klimathianaki M, Pougounias M. Systemic
●64. Boutin C, Rey F. Thoracoscopy in pleural malignant mesothelioma: inflammatory reaction after thoracoscopic talc poudrage. Chest
a prospective study of 188 consecutive patients. Part 1: Diagnosis. 2006; 129: 356–61.
Cancer 1993; 72: 389–93. ●86. Campos JR, Werebe EC, Vargas FS, Jatene FB, Light RW.
●65. Boutin C, Rey F, Gouvernet J, et al. Thoracoscopy in pleural Respiratory failure due to insufflated talc. Lancet 1997; 349:
malignant mesothelioma. Part 2: Prognosis and staging. Cancer 251–2.
1993; 72: 394–404. 87. Rehse DH, Aye RW, Florence MG. Respiratory failure following talc
●66. Boutin C, Nussbaum E, Monnet I, et al. Intrapleural treatment with pleurodesis. Am J Surg 1999; 177: 437–440.
gamma-interferon in early stage malignant mesothelioma. Cancer ●88. Ferrer J, Villarino MA, Tura JM, Traveria A, Light RW. Talc
1994; 74: 2460–7. preparations used for pleurodesis vary markedly from one
67. Goey SH, Eggemont AMM, Punt CJA, et al. Intrapleural preparation to another. Chest 2001; 119: 1901–5.
administration of interleukin 2 in pleural mesothelioma: a phase 89. Ferrer J, Montes JF, Villarino MA, Light RW, Garcia-Valero J.
I–II study. Br J Cancer 1995; 72: 1283–8. Influence of particle size on extrapleural talc dissemination after
68. Boutin C, Schlesser M, Frenay C, Astoul PH. Malignant pleural talc slurry pleurodesis. Chest 2002; 122: 1018–27.
mesothelioma. Eur Respir J 1998; 12: 972–81. ●90. Maskell NA, Lee YC, Gleeson FV, et al. Randomized trials describing
●69. Boutin C, Dumortier P, Rey F, Viallat JR, DeVuyst P. Black spots lung inflammation after pleurodesis with talc of varying particle
concentrate oncogenic asbestosis fibers in the parietal pleura: size. Am J Resp Crit Care Med 2004; 170: 377–82.
thoracoscopic and mineralogic study. Am J Respir Crit Care Med 91. Weissberg D, BenZeev I. Talc pleurodesis: experience with 360
1996; 153: 111–9. patients. J Thorac Cardiovasc Surg 1993; 106: 689–95.
●70. Levine MN, Young JE, Ryan ED, Newhouse MT. Pleural effusion in 92. Viallat JR, Rey F. Astoul Pl, Boutin C. Thoracoscopic talc poudrage
breast cancer: thoracoscopy for hormone receptor determination. pleurodesis for malignant effusions: a review of 360 cases. Chest
Cancer 1986; 57: 324–7. 1996; 110: 1387–93.
71. Schwarz C, Lübbert Hl, Rahn W, et al. Medical thoracoscopy: ◆93. Cardillo G, Facciolo F, Carbone L, et al. Long term follow up of
hormone receptor content in pleural metastases due to breast video assisted talc pleurodesis in malignant recurrent pleurals
cancer. Eur Respir J 2004; 24: 728–30. effusions. Eur J Cardiothorac Surg 2002; 21: 302–6.
●72. Celikoglu F, Teirstein AS, Krellenstein DJ, Strauchen JA. Pleural 94. Noppen M. Who’s (still) afraid of talc? Eur Respir J 2007; 29:
effusion in non-Hodgkins lymphoma. Chest 1992; 101: 619–21.
1357–60. 95. Nasreen N, Mohammed KA, Dowling PA, et al. Talc induces
●73. Sanchez-Armengol A, Rodriguez-Panadero F. Survival and talc apoptosis in human malignant mesothelioma cells in vitro. Am J
pleurodesis in metastatic pleural carcinoma, revisited. Report of Respir Crit Care Med 2000; 161: 595–600.
125 cases. Chest 1993; 104: 1482–5. ●96. Najmunnisa N, Mohammed KA, Brown S, et al. Talc mediates
●74. Antony VB, Nasreen N, Mohammed KA, et al. Talc pleurodesis. angiostasis in malignant pleural effusions via endostatin
Basic fibroblast growth factor mediates pleural fibrosis. Chest induction. Eur Respir J 2007; 29: 761–9.
2004; 126: 1522–8. ◆97. Loddenkemper R, Grosser H, Mai J, et al. Diagnostik des
◆75. Rodriguez-Panadero F, Antony VB. Pleurodesis. State of the art. tuberkulösen Pleuraergusses: Prospektiver Vergleich
Eur Respir J 1997; 10: 1648–54. laborchemischer, bakteriologischer, zytologischer und
●76. Mares CC, Mathur PN. Thoracoscopic talc pleurodesis for histologischer Untersuchungsergebnisse. Prax Klin Pneumol 1983;
lymphoma induced chylothorax, a case series of twenty two 37: 1153–6.
treated hemithoraces in eighteen patients. Am J Respir Crit Care 98. Walzl G, Wyser D, Smedema J, Corbett C, van de Wal B. Comparing
Med 1997; 155: A481. the diagnostic yield of Abrams needle pleural biopsy and
●77. Janssen JP, Collier G, Astoul P, et al. Safety of talc pleurodesis with thoracoscopy. Am J Respir Crit Care Med 1996; 153: A460.
References 597
99. Wyser C, Walzl G. Smedema JP, et al. Corticosteroids in the spontaneous pneumothorax: state of the art. Eur Respir J 2006;
treatment of tuberculous pleurisy. A double-blind, placebo- 28: 637–50.
controlled, randomized study. Chest 1996; 110: 333–8. ●113. Vanderschueren RG. Le talcage pleural dans le pneumothorax
◆100. Diacon AH, Van de Wal BW, Wyser C, et al. Diagnostic tools in spontané. Poumon Cœur 1981; 37: 273–6.
tuberculous pleurisy: a direct comparative study. Eur Respir J 114. vd Brekel JA, Duurkens VAM, Vanderschueren RGJRA.
2003; 22: 589–91. Pneumothorax: results of thoracoscopy and pleurodesis with talc
101. Aziz MA, Wright A, Laszlo A, et al. Epidemiology of poudrage and thoracotomy. Chest 1993; 103: 345–7.
antituberculosis drug resistance (the Global Project on Anti- 115. El Khawand C, Marchandise FX, Maynel A, et al. Pneumothorax
tuberculosis Drug Resistance Surveillance): an updated analysis. spontané. Résultats du talcage pleural sous thoracoscopie. Rev
Lancet 2006; 368: 2142–54. Med Resp 1995; 12: 275–81.
102. Sakuraba M, Masuda K, Hebisawa A, Sagara Y, Komatsu H. 116. Hausmann M, Keller R. Thorakoskopische Pleurodese beim
Thoracoscopic pleural biopsy for tuberculous pleurisy under local Spontanpneumothorax. Schweiz Med Wochenschr 1994; 124:
anesthesia. Ann Thorac Cardiovasc Surg 2006; 12: 245–8. 97–104.
∗103. Maskell NA, Butland RJ, Pleural Diseases Group, Standards of Care ●117. Noppen M, Dekeukeleire T, Hanon S, et al. Fluorescein-enhanced
Committee, British Thoracic Society. BTS guidelines for the autofluorescence thoracoscopy in patients with primary
investigation of a unilateral pleural effusion in adults. Thorax spontaneous pneumothorax and normal subjects. Am J Respir Crit
2003; 58 (Suppl 2): ii8–17. Care Med 2006; 174: 26–30.
∗104. Ernst A, Silvestri GA, Johnstone D, American College of Chest 118. Almind M, Lange P, Viskum K. Spontaneous pneumothorax:
Physicians. Interventional pulmonary procedures. Guidelines from comparison of simple drainage, talc pleurodesis and tetracycline
the American College of Chest Physicians. Chest 2003; 123: pleurodesis. Thorax 1989; 44: 627–30.
1693–1717. 119. Tschopp J-M, Brutsche M, Frey JG. Treatment of complicated
●105. Sudduth C, Sahn SA. Pleurodesis for non-malignant pleural spontaneous pneumothorax by simple talc pleurodesis under
effusions. Recommendations. Chest 1992; 102: 1855–60. thoracoscopy and local anaesthesia. Thorax 1997; 52: 329–32.
●106. Weissberg D. Pleuroscopy in empyema: is it ever necessary? 120. Lee P, Yap WS, Pek WY, et al. An audit of medical thoracoscopy
Poumon Coeur 1981; 37: 269–72. and talc poudrage for pneumothorax prevention in advanced
◆107. Weissberg D. Handbook of practical pleuroscopy. Mount Kisko: COPD. Chest 2004; 125: 1315–20.
Futura, 1991. ●121. Lange P, Mortensen J, Groth S. Lung function 22–25 years after
●108. Brutsche MH, Tassi GF, Györik S, et al. Treatment of treatment of idiopathic spontaneous pneumothorax with talc
sonographically stratified multiloculated thoracic empyema by poudrage or simple drainage. Thorax 1988; 43: 559–61.
medical thoracoscopy. Chest 2005; 128: 3303–9. 122. Györik S, Erni S, Studler Ul, et al. Long-term follow-up of
109. Lee P, Colt HG. Thoracoscopy: an update on therapeutic thoracoscopic talc pleurodesis for primary spontaneous
applications. J Respir Dis 2003; 24: 530–6. pneumothorax. Eur Respir J 2007; 29: 757–60.
∗110. Colice GL, Curtis A, Deslauriers J, et al. Medical and surgical ◆123. Judson MA, Sahn SA. The pleural space and organ transplantation.
treatment of parapneumonic effusions. an evidence-based Am J Respir Crit Care Med 1996; 153: 1153–65.
guideline. Chest 2000; 118: 1158–71. (Erratum in: Chest 2001; 124. Tassi GF, Davies RJ, Noppen M. Advanced techniques in medical
119: 319.) thoracoscopy. Eur Respir J 2006; 28: 1051–9.
111. Schramel FMNH, Postmus PE, Vanderschueren RGJRA. Current ◆125. Noppen M. Medical thoracoscopy. Techniques for thoracic
aspects of spontaneous pneumothorax. Eur Respir J 1997; 10: sympathectomy. In: Beamis JF, Mathur P, Mehta A (eds).
1372–9. Interventional pulmonary medicine. New York: Marcel Dekker, Inc:
◆112. Tschopp JM, Rami-Porta R, Noppen M, Astoul P. Management of 2004: 483–502.
This page intentionally left blank
48
Surgery for pleural diseases
Surgical method
Is bleb resection necessary? the reduced release of inflammatory and vasoactive medi-
ators after VATS. This may explain why VATS is better tol-
Failure to identify or ablate a macroscopic bleb at VATS erated.21
has been found to be an independent predictor of recur- In one of the few prospective randomized comparisons
rence.13 Recurrence rates of 23 percent were reported com- of the use of VATS versus open surgery (limited postero-
pared with 1.8 percent when a bleb was ablated. Routine lateral thoracotomy) for spontaneous pneumothorax we
blind resection of the apex of the upper lobe together with found that stapled bullectomy and apical parietal pleurec-
a procedure to obtain pleural symphysis has therefore been tomy could be performed equally reliably for PSP.9
advocated. An alternative method of dealing with a bleb is Operating time was no longer for VATS but the post-
VATS ligation using a pre-tied Roeder slip knot intro- operative analgesic requirement and hospital stay were
duced by an external applicator. Unfortunately, this tech- reduced. VATS resulted in less early post-operative respi-
nique was associated with a treatment failure rate of nearly ratory dysfunction with a drop in FEV1 (forced expiratory
20 percent and is therefore not recommended.14 volume in 1 second) of 29 percent and in FVC (forced vital
capacity) of 19 percent compared with falls of 43 and 39
percent, respectively, after lateral thoracotomy.
Which type of pleural procedure? There has been no similar prospective randomized
comparison of VATS and limited axillary thoracotomy
Whilst there has been a report of successful pneumothorax (LAT). Whilst Dumont et al.22 showed shorter stay, less
prevention by apical bullectomy without pleurodesis15 it is pain and similar recurrence than LAT in a retrospective
accepted practice that a procedure to promote pleural review of only 79 patients, Miller et al.23 found no benefits
adhesions is the norm.1 Mechanical pleurodesis can be from VATS. Similar findings have been described by other
accomplished by parietal pleurectomy or pleural abrasion. authors24,25 who showed perioperative benefits from VATS
In PSP when an apical source of air leak is the norm, an but higher recurrence related to resection of fewer bullae
apical parietal pleurectomy is preferred. This theoretically than limited axillary thoracotomy. These findings may be
allows the possibility of future thoracotomy for other indi- explained by unfamiliarity with the newer technique of
cations. Pleural abrasion16 is technically easier to perform VATS. Indeed, the early comparisons of VATS with tradi-
then pleurectomy by VATS and preserves the extrapleural tional methods may not have appreciated the technical dif-
plane and is therefore favored by some surgeons. In a ficulties in learning a new technique. It has been shown
review of these two alternatives there was a slightly lower that VATS results improve with surgical experience with
recurrence rate after pleurectomy than abrasion: 0.4 fewer complications26 and a decrease in operating time
percent (n = 752) versus 2.3 percent (n = 301).17 and hospital stay;10 in the largest reported series of VATS,
In secondary SP, parietal pleurectomy may be avoided the fact that the recurrence rates are slightly higher than
in favor of talc insufflation which may shorten the dura- the accepted rates for open surgery (Table 48.1) may be
tion of anaesthesia, be less painful and for which there are explained by this. In a systematic review of four trials of
less potential long-term problems in this age group. VATS versus thoracotomy in over 200 patients, Sedryakan
Talc is the chemical sclerosant of choice with a lower et al.27 found that VATS resulted in a reduced need for
recurrence rate than tetracycline in a comparative trial of 96 analgesia and length of stay.
patients: 8 percent versus 13 percent.4 In larger reviews the
recurrence rate for talc pleurodesis alone was 10–15 percent
and therefore mechanical methods should be preferred. Medical thoracoscopy
The choice of pleural procedure in combination with bleb
excision is less clear. Cardillo et al.18 reported that the addi- The use of video-assisted thoracoscopy in a spontaneously
tion of talc poudrage rather than subtotal pleurectomy to ventilating patient without double-lumen intubation has
stapled excision of the bleb by VATS carried a significantly become to be known as ‘medical thoracoscopy’. There are
lower recurrence rate. Their long-term experience of over several proponents of this technique in the treatment of
800 cases reported no adult repiratory distress and an spontaneous pneumothorax with simple talc insufflation
overall recurrence rate of only 1.73 percent.19 being the usual method employed. Indeed, there are pro-
Fibrin glue has also been used by insufflation to obtain ponents of its use in first-time PSP in preference to tube
pleurodesis but has been associated with high recurrence drainage alone.28 Whilst early recurrence is reported to be
of up to 25 percent.20 low, recurrence is higher than open surgery when macro-
scopic bullae are present.29 This results from a failure to
treat the source of the air leak attributable to the absence
Which surgical approach: VATS versus of single-lung ventilation which precludes complete
thoracotomy? intrathoracic inspection. In the event of recurrence after
initial medical thoracoscopic talc insufflations, subsequent
Video-assisted thoracoscopic surgery has been shown to be VATS is precluded by patchy adhesions, thoracotomy is
a less traumatic approach than thoracotomy as shown by required and the benefits of VATS are lost. We do not
602 Surgery for pleural diseases
Author Study Number Deaths Conversion Persistent Drain Hospital Median Recurrence
to air leak time stay follow-up
thoractomy (days) (days)
Special considerations
CYSTIC FIBROSIS
CATAMENIAL PNEUMOTHORAX
used to exclude and treat an anatomical cause either by
This condition defines a cyclical pneumothorax which stapling of an apical bleb or a diaphragmatic fenestration
occurs in young females in relation to the menstrual cycle. (Figure 48.3) before hormonal treatment is commenced.
Several causative theories have been forwarded including An additional procedure to achieve pleurodesis at the level
alveolar rupture secondary to the effects of raised of the diaphragm is also advisable.36
prostaglandin F2 levels and ectopic foci of endometriosis
in the lung.33 The finding of diaphragmatic fenestrations
SPONTANEOUS HEMOPNEUMOTHORAX
supports a theory of the passage of free air from the peri-
toneal cavity due to absence of the normal cervical mucous This is defined as the accumulation of over 400 mL of
plug during ovulation34 and mandates surgical explo- blood in the pleural cavity associated with spontaneous
ration. In a review of 154 operated cases in the published pneumothorax. It occurs in up to 5 percent of cases37 and
literature: endometriosis was found in 52 percent and is associated with hypovolemic shock in up to one in three
diaphragmatic lesions in 39 percent.35 Hormonal treat- cases. It is caused by the tearing of an arterial branch from
ment (luteinizing hormone-releasing hormone analogue) the parietal pleural surface to the lung as the lung collapses
alone, intended to treat pleural endometriosis as a cause, is following pneumothorax. Treatment may be successful by
not adequate since it will be appropriate in only 20 percent VATS and includes identification and either clipping or
of cases where this condition coexists.33 VATS should be cautery of the feeding vessel.
Pleural empyema 603
There are two main principles to surgical intervention in VATS VERSUS FIBRINOLYSIS
pleural empyema: debridement of the infected cavity and
decortication of the underlying lung. The former contributes Initial reports of the use of VATS in the management of
to the control of sepsis and the latter allows for re-expansion the fibrinopurulent phase of empyema reported an imme-
of the lung and obliteration of the empyema cavity.
SURGICAL APPROACH
Table 48.2 Results of video-assisted thoracoscopic surgery (VATS) for pleural empyema
be carried out by open thoracotomy. Landreneau et al.51 ment of a TB empyema.58 VATS decortication of the
found that 17 percent of his series required open decorti- trapped lung in TB is also possible.
cation and in these patients 93 percent had an empyema
for more than 3 weeks. Certainly, thoracotomy with decor- MALIGNANT EMPYEMA
tication has an excellent success rate of around 95 percent
in treating empyema at any stage with low associated mor- A malignant pleural effusion may become secondarily
bidity and mortality.52 Nevertheless, with technical refine- infected by percutaneous intervention or intrinsic pul-
ment, VATS can successfully treat even the most chronic monary infection predisposed by collapse of the
empyema. We have reported successful VATS decortica- parenchyma (Figure 48.6). Repeated pleural aspirations
tion in patients with duration of symptoms of nearly 2 render the patient particularly susceptible to secondary
months.53 With increasing experience in the use of VATS infection. Because of the frailty of these patients, fenestra-
to treat all cases of chronic empyema primarily, we are tion may be all that is possible. However, the benefits of
currently successful in around 60 percent of cases.54 VATS are especially pertinent in this group of patients.
RIB RESECTION/THORACOSTOMY
MALIGNANT PLEURAL EFFUSION
Rib resection to drain the empyema cavity is indicated for
patients at high-risk from more major surgery with thora- Indications for surgery
cotomy or who could not tolerate single-lung ventilation.
It may be performed with the patient spontaneously venti- PLEURAL BIOPSY
lating with the resection of a short segment of rib over the
most dependent part of the empyema cavity to allow Diagnostic pleural biopsy by VATS has a high sensitivity of
drainage of and limited breakdown of loculations in the over 90 percent59 as it allows full inspection of the pleural
infected space. A large bore tube is left in situ.55 In patients surfaces under magnification and illumination. This pro-
in whom the empyema cavity is thought unlikely to close cedure can be carried out in a spontaneously ventilating
or to be sterilized, a permanent drainage site can be patient if the pleural disease is diffuse. If the patient is to be
formed by resecting two to three ribs and by suturing the considered for more radical surgery, then the diagnostic
exposed parietal pleura to the skin of the raised cutaneous pleural biopsy should be performed via as few incisions as
flaps. The cavity can then be easily irrigated and dressed as possible (ideally in the line of a future thoracotomy inci-
an outpatient. This procedure is described as the Eloesser sion) to minimize the risk of tumor implantation.
flap drainage procedure.56 Thoracoscopy is superior to closed pleural biopsy in the
diagnosis of malignant pleural effusion.60
TUBERCULOUS EMPYEMA
PLEURODESIS
The original description of thoracoscopy was for adhesio-
lysis for tuberculosis (TB).57 VATS has been shown to be Control of pleural fluid production can be achieved by
useful in the diagnosis of TB effusion and in the debride- inducing pleurodesis with talc insufflation administered
by video-assisted or direct thoracoscopy and can be effec-
tively administered in a spontaneously ventilating patient
under local anesthesia.61 Thoracoscopic talc poudrage is
superior to talc slurry administered via tube thoracos-
tomy.62 In mesothelioma, cytoreductive parietal pleurec-
tomy may have a longer lasting control of symptoms than
chemical pleurodesis alone.63 Cytoreductive pleurectomy
may also have a survival benefit over pleural biopsy
alone.64 (Pleurodesis is also discussed in further detail in
Chapter 46.)
TRAPPED LUNG
fort, avoids prolonged hospital stay and relieves symptoms nodal size,70 therefore cervical mediastinoscopy is manda-
during the terminal phase of the disease. tory. Future preoperative selection may include biological
Surgical treatment is directed towards either evacuation markers of angiogenesis, i.e tumor microvessel counts or
of fluid from the thoracic cavity by the insertion of a pleu- matrix metalloproteinase expression.71
roperitoneal shunt or by attempting to re-expand the lung The operation of extrapleural pneumonectomy (EPP)
by visceral decortication. Whilst evacuation of the pleural involves the en-bloc excision of the entire pleural surface
space will improve dyspnoea, the patient will remain com- with underlying lung and adjacent pericardium and
promised by the collapsed lung. hemidiaphragm. Mediastinal lymph node clearance is also
Pleuroperitoneal shunts are usually inserted using advocated. The pericardium and diaphragm are replaced
VATS although mini-thoracotomy may be required if with prosthetic patches.The operation is usually per-
repeated prior aspiration has resulted in pleural adhesions. formed via a lateral thoracotomy but we have found post-
Operative complications are generally low but post- operative benefits from a median sternotomy approach.72
operative complications, such as infection, shunt blockage The initial reported operative mortality of 25 percent
and tumor seeding at the port site, may occur in up to 15 for extrapleural pneumonectomy73 was initially thought to
percent of patients. However, effective palliation can be be preclusive but recent series reflecting improvements in
achieved in 95 percent of cases with no evidence of tumor perioperative care report operative mortality of as low as
seeding into peritoneum.65 3.8 percent.68 Unfortunately, despite wider application,
We have found tumor decortication with visceral radical treatment has had little impact on overall survival
pleurectomy to be beneficial with symptomatic relief of from mesothelioma. The best 5-year survival figures are
dyspnea for over 3 months with acceptable operative mor- reported for those with the prognostic factors above and
tality and morbidity.66 Thoracotomy, however, should be these represent a minority of those referred for surgery.
reserved for those patients with epithelioid mesothelioma With multi-modality treatment including radical surgery,
and before significant weight loss since post-operative radical hemithorax irradiation and chemotherapy median
recovery may significantly erode the remaining symptom- survival ranges from 33 to 51 months.68,74 Despite these
free survival. For this reason we have extended the role of aggressive attempts at local disease control, including
VATS decortication to mesothelioma with survival benefit, intracavitary chemotherapy,75 survival from mesothelioma
especially in the elderly.67 surgery is thwarted by distant disease progression,76,77
Our current management protocol for malignant commonly in the peritoneal cavity or the contralateral
pleural effusion comprises early assessment by VATS and pleural cavity. Several authors therefore advocate radical
incorporates all the above treatment options (Figure 48.7). pleurectomy/decortication, with preservation of the lung,
in preference to EPP78,79 as it is associated with lower oper-
RADICAL SURGERY ative mortality and potentially better functional status. We
utilize this form of radical debulking in those with positive
In selected cases, malignant pleural effusion secondary to mediastinoscopy whose survival is limited.80
malignant mesothelioma may be treated by extrapleural
pneumonectomy. This operation is only advocated by a
small number of surgical centers and is reserved for a MISCELLANEOUS
selected number of patients. The indicators of favorable
post-operative prognosis include epithelioid histology, clear Clotted hemothorax
surgical resection margins and the absence of extrapleural
lymph node involvement.68 Thus, careful preoperative A retained hemothorax may be expected in around 2
staging is imperative. Malignant lymph node involvement percent of patients with thoracic injuries.81 CT is needed to
cannot be accurately predicted from nodal uptake on predict the need for evacuation of the clotted blood as
positron-emission tomography (PET) scanning69 or from chest radiography alone is insensitive. VATS is effective in
Cytology/closed biopsy
VATS
SURGICAL INDICATIONS
SURGICAL RESULTS
Cirrhotic hydrothorax
Pleural biopsy
The most likely mechanism for this condition, which
develops in approximately 5 percent of patients with cir- Malignant seeding of tumor cells in thoracoscopy port
rhosis, is the passive collection of pleural fluid due to the sites may occur in up to 40 percent of patients after biopsy
movement of ascites along a pressure gradient through of malignant pleural disease. A prospective randomized
small diaphragmatic fenestrations.87 Repeated thoracocen- trial by Boutin and colleagues94 demonstrated a preventive
tesis or tube drainage are contraindicated due to the result- benefit from prophylactic external beam radiotherapy to
ant hypovolemia, protein loss and infection. Therapy these sites.
608 Surgery for pleural diseases
●
Talc pleurodesis Day case surgery by VATS with in situ Heimlich valve
drainage system.
● Incorporating the principles of lung volume reduction
The fear of mesothelioma development as a long-term
sequelae of talc poudrage is unfounded in the era of surgery in patients presenting with secondary sponta-
asbestos-free talc.95 Long-term restrictive respiratory neous pneumothorax.
impairment is also not a major concern.96 A potentially
more serious risk from talc pleurodesis in the surgical
setting is the development of acute respiratory distress
Pleural empyema
after insufflations.97 However, recent large series19 have
● Earlier surgical intervention by VATS debridement
not seen this complication, particularly where large-
particle talc is used.98 rather than extended tube drainage or intrapleural fib-
rinolysis.
● Wider application of VATS to more chronic post-
Decortication pneumonic and tuberculous cases.
Tumor decortication must ensure full lung re-expansion Malignant pleural effusion
as these patients are particularly at risk from infection of
any residual space with the formation of an empyema. In ● Increased application of medical thoracoscopy with
our own series we have found this complication in 4
pleurodesis as the initial diagnostic procedure.
percent of patients.66 Other specific complications include ● Prospective randomized controlled trials of radical
an iatrogenic parenchymal lung injury with resultant per-
surgery in mesothelioma.
sistent air leak, post-operative hemothorax and iatrogenic ● Improved management of the trapped lung by VATS
injuries to the azygos vein, thoracic duct or esophagus (the
decortication.
intraoperative placement of an esophageal bougie aids ● Prospective randomized controlled trials of VATS
identification of the organ and reduces the risk of inadver-
debulking surgery.
tent injury).
Hemothorax
Extrapleural pneumonectomy
The earlier use of VATS in the assessment of major tho-
Major morbidity is experienced by over 40 percent of racic trauma may prevent unnecessary thoracotomy from
patients undergoing this extensive operation,99 most com- compounding the risks of the injury itself.
monly supraventricular cardiac arrythmias. Specific tech-
nical complications include detachment of prosthetic
patches, bronchopleural fistulae, esophageal rupture and
post-operative hemorrhage. We have also observed rapid
filling of the pneumonectomy space with serous fluid
leading to mediastinal shift and hemodynamic compro- KEY POINTS
mise.100 There is debate whether the use of induction
chemotherapy increases the risk of post-operative compli- ● Early surgical intervention by VATS is indicated
cations.100,101 in both spontaneous pneumothorax and pleural
empyema to maximize the benefits of minimally
invasive surgery.
● Percutaneous fibrinolytic treatment should not
FUTURE SURGICAL DIRECTIONS delay surgical intervention in pleural empyema.
● Multiple preoperative pleural interventions are
The following are areas of surgical practice which are unnecessary and may prejudice the outcome of
under evaluation at present and which in the future may surgery by inducing pleural sepsis and adhesions.
be incorporated into routine management protocols for ● VATS has many advantages over thoracotomy in
pleural disease. the management of pleural disease but is subject
to a learning curve.
● The role of VATS debulking in mesothelioma
Spontaneous pneumothorax remains to be determined.
● The benefit of extrapleural pneumonectomy in
● Surgical intervention by VATS after the first episode of mesothelioma remains to be proven.
primary spontaneous pneumothorax.
References 609
REFERENCES 21. Gebhard FT, Becker HP, Gerngross H, Bruckner UB. Reduced
inflammatory response in minimal invasive surgery of
pneumothorax. Arch Surg 1996; 131: 1079–82.
● = Key primary paper 22. Dumont P, Diemont F, Massard G, et al. Does a thoracoscopic
◆ = Major review article approach for surgical treatment of spontaneous pneumothorax
represent progress? Eur J Cardiothorac Surg 1997; 11: 27–31.
◆1. Baumann MH, Strange C, Heffner JE, et al. Management of 23. Miller JD, Simone C, Kahnamoui K, et al. Comparison of
spontaneous pneumothorax: an American College of Chest videothoracoscopy and axillary thoracotomy for the treatment of
Physicians Delphi consensus statement. Chest 2001; 119: spontaneous pneumothorax. Am Surg 2000; 66: 1014–15.
590–602. 24. Horio H, Nomori H, Fuyuno G, Kobayashi R, Suemasu K. Limited
2. Waller DA, McConnell SA, Rajesh PB. Delayed referral reduces the axillary thoracotomy vs video-assisted thoracoscopic surgery for
success of video-assisted thoracoscopic surgery for spontaneous spontaneous pneumothorax. Surg Endosc 1998; 12: 1155–8.
pneumothorax. Respir Med 1998; 92: 246–9. 25. Kim KH, Kim HK, Han JY, et al. Transaxillary minithoracotomy
3. Sihoe AD, Yim AP, Lee TW, et al. Can CT scanning be used to select versus video-assisted thoracic surgery for spontaneous
patients with unilateral primary spontaneous pneumothorax for pneumothorax. Ann Thorac Surg 1996; 61: 1510–12.
bilateral surgery? Chest 2000; 118: 380–3. 26. Jimenez-Merchan R, Garcia-Diaz F, Arenas-Linares C, et al.
4. Almind M, Lange P, Viskum K. Spontaneous pneumothorax: Comparative retrospective study of surgical treatment of
comparison of simple drainage,talc pleurodesis, tetracycline spontaneous pneumothorax. Thoracotomy vs thoracoscopy. Surg
pleurodesis. Thorax 1989; 44: 627–30. Endosc 1997; 11: 919–22.
5. Torresini G, Vaccarili M, Divisi D, Crisci R. Is video-assisted thoracic ◆27. Sedryakan A, van der Meulen J, Lewsey J, Treasure T. Video
surgery justified at first spontaneous pneumothorax? Eur J assisted thoracic surgery for treatment of pneumothorax and lung
Cardiothorac Surg 2001; 20: 42–5. resections: systematic review of randomized clinical trials. Br Med
6. Abdala OA, Levy RR, Bibiloni RH, et al. Advantages of video- J 2004; 329: 1008–11.
assisted thoracic surgery in the treatment of spontaneous 28. Tschopp JM, Boutin C, Astoul P, et al. Talcage by medical
pneumothorax. Medicina (B Aires) 2001; 61: 157–60. thoracoscopy for primary spontaneous pneumothorax is more
7. Deslauriers J, Beaulieu M, Despres JP, et al. Transaxillary cost-effective than drainage: a randomized study. Eur Respir J
pleurectomy for treatment of spontaneous pneumothorax. Ann 2002; 20: 1003–9.
Thorac Surg 1980; 30: 569–74. 29. Tschopp JM, Brutsche M, Frey JG. Treatment of complicated
●8. Waller DA, Forty J, Morritt GN. Video-assisted thoracoscopic spontaneous pneumothorax by simple talc pleurodesis under
surgery versus thoracotomy for spontaneous pneumothorax. Ann thoracoscopy and local anaesthesia. Thorax 1997; 52: 329–32.
Thorac Surg 1994; 58: 372–6. 30. Waller DA. Video-assisted thoracoscopic surgery (VATS) in the
9. Waller DA, Forty J, Soni AK, Conacher ID, Morritt GN. management of spontaneous pneumothorax. Thorax 1997; 52:
Videothoracoscopic operation for secondary spontaneous 307–8.
pneumothorax. Ann Thorac Surg 1994; 57: 1612–15. 31. Spector ML, Stern RC. Pneumothorax in cystic fibrosis: a 26 year
10. Waller DA. Video-assisted thoracoscopic surgery for spontaneous experience. Ann Thorac Surg 1989; 47: 204–7.
pneumothorax – a 7-year learning experience. Ann R Coll Surg 32. Noyes BE, Orenstein DM. Treatment of pneumothorax in cystic
Engl 1999; 81: 387–92. fobrosis in the era of lung transplantation. Chest 1992; 101:
11. McManus KG, Spence GM, McGuigan JA. Outpatient chest tubes. 1187–8.
Ann Thorac Surg 1998; 66: 299–300. ◆33. Joseph J, Sahn SA. Thoracic endometriosis syndrome: new
12. Gelb AF, McKenna RJ Jr, Brenner M, Epstein JD, Zamel N. Lung observations from an analysis of 110 cases. Am J Med 1996; 100:
function 5 years after lung volume reduction surgery for 164–70.
emphysema. Am J Respir Crit Care Med 2001; 163: 1562–6. 34. Fonseca P. Catamenial pneumothorax: a multifactorial etiology. J
◆13. Naunheim KS, Mack MJ, Hazelrigg SR, et al. Safety and efficacy of Thorac Cardiovasc Surg 1998; 116: 872–3.
video-assisted thoracic surgical techniques for the treatment of 35. Korom S, Canyurt H, Missbach A, et al. Catamenial pneumothorax
spontaneous pneumothorax. J Thorac Cardiovasc Surg 1995; 109: evisited. J Thorac Cardiovasc Surg 2004; 128: 502–8.
1198–203. 36. Marshall MB, Ahmed Z, Kucharczuk JC, Kaiser LR, Shager JB.
14. Inderbitzi RG, Leiser A, Furrer M, Althaus U. Three years’ Catamenial pneumothorax: optimal hormonal and surgical
experience in video-assisted thoracic surgery (VATS) for management. Eur J Cardiothorac Surg 2005; 27: 662–6.
spontaneous pneumothorax. J Thorac Cardiovasc Surg 1994; 107: 37. Hsu NY, Shih CS, Hsu CP, Chen PR.Spontaneous
1410–15. hemopneumothorax revisited. Ann Thorac Surg 2005; 80: 1859.
15. Ferguson LJ, Imrie CW, Hutchison J. Excision of bullae without ◆38. Colice GL, Curtis A, Deslauriers J, et al. Medical and surgical
pleurectomy in patients with spontaneous pneumothorax. Br J treatment of parapneumonic effusions: an evidence-based
Surg 1981; 68: 214–16. guideline. Chest 2000; 118: 1158–71.
16. Melvin WS, Krasna MJ, McLaughlin JS. Thoracoscopic 39. Ridley PD, Braimbridge MV. Thoracoscopic debridement and
management of spontaneous pneumothorax. Chest 1992; 102: pleural irrigation in the management of empyema thoracis. Ann
1877–9. Thorac Surg 1991; 51: 461–4.
17. Weeden D, Smith GH. Surgical experience in the management of 40. Nakamura H, Taniguchi Y, Makihara K, Ohgi S. Ultrasonic pleural
spontaneous pneumothorax, 1972–82. Thorax 1983; 38: 737–43. debridement of empyema. Ann Thorac Cardiovasc Surg 2001; 7:
18. Cardillo G, Facciolo F, Giunti R, et al. Videothoracoscopic 62–4.
treatment of primary spontaneous pneumothorax: a 6-year 41. Angelillo Mackinlay TA, Lyons GA, Chimondeguy DJ, et al. VATS
experience. Ann Thorac Surg 2000; 69: 357–61. debridement versus thoracotomy in the treatment of loculated
19. Cardillo G, Carleo F, Giunti R, et al. Video thoracoscopic talc postpneumonia empyema. Ann Thorac Surg 1996; 61: 1626–30.
poudrage in primary spontaneous pneumothorax: a single- 42. Lawrence DR, Ohri SK, Moxon RE, Townsend ER, Fountain SW.
institution experience in 861 cases. J Thorac Cardiovasc Surg Thoracoscopic debridement of empyema thoracis. Ann Thorac Surg
2006; 131: 322–8. 1997; 64: 1448–50.
20. Guerin JC, van Derschueren RG. Treatment of recurrent 43. Striffeler H, Gugger M, Im Hof V, et al. Video-assisted
pneumothorax applying fibrin adhesive under endoscopy. Rev Mal thoracoscopic surgery for fibrinopurulent pleural empyema in 67
Respir 1989; 6: 443–5. patients. Ann Thorac Surg 1998; 65: 319–23.
610 Surgery for pleural diseases
●44. Wait MA, Sharma S, Hohn J, Dal Nogare A. A randomized trial of 66. Martin-Ucar AE, Edwards JG,Rengarajan A, Muller S, Waller DA.
empyema therapy. Chest 1997; 111: 1548–51. Palliative surgical debulking in malignant mesothelioma.
45. Waller DA, Rengarajan A, Nicholson FH, Rajesh PB. Delayed Predictors of survival and symptom control. Eur J Cardiothorac
referral reduces the success of video-assisted thoracoscopic Surg 2001; 20: 1117–21.
debridement for post-pneumonic empyema. Respir Med 2001; 95: 67. Nakas A, Martin-Ucar AE, Barlow A, et al. The extent of surgery for
836–40. malignant mesothelioma in patients over the age of 65: a
46. Lardinois D, Gock M, Pezzetta E, et al. Delayed referral and Gram therapeutic dilemma. Thorax 2006; 61: 45S
negative organisms increase the conversion to thoracotomy rate in ●68. Sugarbaker DJ, Flores RM, Jaklitsch MT, et al. Resection margins,
patients undergoing video-assisted thoracoscopic surgery for extrapleural nodal status, cell type determine postoperative long-
empyema. Ann Thorac Surg 2005; 79: 1851–6. term survival in trimodality therapy of malignant pleural
47. Lim TK, Chin NK. Empirical treatment with fibrinolysis and early mesothelioma: results in 183 patients. J Thorac Cardiovasc Surg
surgery reduces the duration of hospitalization in pleural sepsis. 1999; 117: 54–63.
Eur Respir J 1999; 13: 514–18. 69. Flores RM, Akhurst T, Gonen M, et al. Positron emission
●48. Maskell NA, Davies CWH, Nunn AJ, et al. UK controlled trial of tomography predicts survival in malignant pleural mesothelioma. J
intrapleural streptokinase for pleural infection. N Engl J Med Thorac Cardiovasc Surg 2006; 132: 763–8.
2005; 352: 865–74. 70. Pilling JE, Stewart DJ, Martin-Ucar AE, et al. The case for routine
49. Cohen G, Hjortal V, Ricci M. Primary thoracoscopic treatment of cervical mediastinoscopy prior to radical surgery fro malignant
empyema in children. J Thorac Cardiovasc Surg 2003; 125: pleural mesothelioma. Eur J Cardiothorac Surg 2004; 25:
79–83. 497–501.
50. Chin NK, Lim TK. Controlled trial of intrapleural streptokinase in ◆71. O’Byrne KJ, Edwards JG, Waller DA. Clinico-pathological and
the treatment of pleural empyema and complicated biological prognostic factors in pleural malignant mesothelioma.
parapneumonic effusions. Chest 1997; 111: 275–9. Lung Cancer 2004; 45: S45–8.
51. Landreneau RJ, Keenan RJ, Hazelrigg SR, Mack MJ, Naunheim KS. 72. Edwards JG, Martin-Ucar AE, Stewart DJ, Waller DA. Right
Thoracoscopy for empyema and hemothorax. Chest 1996; 109: extrapleural pneumonectomy for malignant mesothelioma via
18–24. median sternotomy or thoracotomy? Short and long-term results.
52. Hoover EL, Hsu HK, Ross MJ, et al. Reappraisal of empyema Eur J Cardiothorac Surg 2007; 31: 759–64.
thoracis. Surgical intervention when the duration of illness is ●73. Butchart EG, Ashcroft T, Barnsley WC, Holden MP.
unknown. Chest 1986; 90: 511–15. Pleuropneumonectomy in the management of diffuse malignant
53. Waller DA, Rengarajan A. Thoracoscopic decortication: a role for mesothelioma of the pleura. Experience with 29 patients. Thorax
video-assisted surgery in chronic postpneumonic pleural 1976; 31: 15–24.
empyema. Ann Thorac Surg 2001; 71: 1813–16. 74. Rusch VW, Rosenzweig K, Venkatraman E, et al.A phase II trial of
54. Jutley RS, Coneybeare A, Rengarajan A,Waller DA. Video assisted surgical resection and adjuvant high-dose hemithoracic radiation
thoracoscopic surgery for chronic post-pneumonic empyema: a for malignant pleural mesothelioma. J Thorac Cardiovasc Surg
missed opportunity? Thorax 2005; 60: 31S 2001; 122: 788–95.
55. Samson PC. Empyema thoracis, essentials of present day 75. Richards WG, Zellos L, Bueno R, et al. Phase I to II study of
management. Ann Thorac Surg 1971; 11: 210. pleurectomy/decortication and intraoperative intracavitary
56. Hurvitz RJ, Thucker BL. The Eloesser flap: past and present. J hyperthermic cisplatin lavage for mesothelioma. J Clin Oncol
Thorac Cardiovasc Surg 1986; 42: 958. 2006; 24: 1561–7.
57. Jacobaeus H. The cauterization of adhesions in artificial 76. Baldini EH, Recht A, Strauss GM, et al. Patterns of failure after
pneumothrax treatment of tuberculosis under thoracoscopic trimodality therapy for malignant pleural mesothelioma. Ann
control. Proc R Soc Med 1922–3; 16: 45–60. Thorac Surg 1997; 63: 334–8.
58. Yim AP, Izzat MB, Lee TW. Thoracoscopic surgery for pulmonary 77. Stewart DJ, Martin-Ucar AE, Pilling JE, et al. The effect of extent
tuberculosis. World J Surg 1999; 23: 1114–17. of local resection on patterns of disease progression in malignant
59. Waller DA, Hasan A, Forty J, Morritt GN. Videothoracoscopy in the pleural mesothelioma. Ann Thorac Surg 2004; 78: 245–52.
diagnosis of intrathoracic pathology: early experience. Ann R Coll 78. Rusch VW, Saltz L, Venkatraman E, et al. A phase II trial of
Surg Engl 1994; 76: 123–6. pleurectomy/decortication followed by intrapleural and systemic
60. Emad A, Rezaian GR. Diagnostic value of closed percutaneous chemotherapy for malignant pleural mesothelioma. J Clin Oncol
pleural biopsy vs pleuroscopy in suspected malignant pleural 1994; 12: 1156–63.
effusion or tuberculous pleurisy in a region with a high incidence 79. Pass HI, Kranda K, Temeck BK, Feuerstein I, Steinberg SM.
of tuberculosis: a comparative, age-dependent study. Respir Med Surgically debulked malignant pleural mesothelioma: results and
1998; 92: 488–92. prognostic factors. Ann Surg Oncol 1997; 4: 215–22.
61. Danby CA, Adebonojo SA, Moritz DM. Video-assisted talc 80. Martin-Ucar AE, Nakas A, Edwards JG, Waller DA. Case-control
pleurodesis for malignant pleural effusions utilizing local study between extrapleural pneumonectomy and radical
anesthesia and IV sedation. Chest 1998; 113: 739–42. pleurectomy/decortication for pathological N2 malignant pleural
62. Stefani A, Natali P, Casali C, Morandi U. Talc poudrage versus talc mesothelioma. Eur J Cardiothorac Surg 2007; 31: 765–70.
slurry in the treatment of malignant pleural effusion. A 81. Velmahos GC, Demetriades D. Early thoracoscopy for the
prospective comparative study. Eur J Cardiothorac Surg 2006; 30: evacuation of undrained haemothorax. Eur J Surg 1999; 165:
827–32. 924–9.
63. Waller DA, Morritt GN, Forty J. Video-assisted thoracoscopic ●82. Meyer DM, Jessen ME, Wait MA, Estrera AS. Early evacuation of
pleurectomy in the management of malignant pleural effusion. traumatic retained hemothoraces using thoracoscopy: a
Chest 1995; 107: 1454–6. prospective, randomized trial. Ann Thorac Surg 1997; 64:
64. Halstead JC, Lim E, Venkateswaran RM, et al. Improved survival 1396–400.
with VATS pleurectomy–decortication in advanced malignant 83. Milsom JW, Kron IL, Rheuban KS, Rodgers BM. Chylothorax: an
mesothelioma. Eur J Surg Oncol 2005; 31: 314–20. assessment of current surgical management. J Thorac Cardiovasc
◆65. Genc O, Petrou M, Ladas G, Goldstraw P. The long-term Surg 1985; 89: 221–7.
morbidity of pleuroperitoneal shunts in the management of 84. Patterson GA, Todd TR, Delarue NC, et al. Supradiaphragmatic
recurrent malignant effusions. Eur J Cardiothorac Surg 2000; 18: ligation of the thoracic duct in intractable chylous fistula. Ann
143–6. Thorac Surg 1981; 32: 44–9.
References 611
85. Graham DD, McGahren ED, Tribble CG, Daniel TM, Rodgers BM. 100. Stewart DJ, Martin-Ucar AE, Edwards JG, West K, Waller DA.
Use of video-assisted thoracic surgery in the treatment of Extrapleural pneumonectomy for malignant pleural mesothelioma;
chylothorax. Ann Thorac Surg 1994; 57: 1507–11. the risks of induction chemotherapy, right-sided procedures and
86. Mares DC, Mathur PN. Medical thoracoscopic talc pleurodesis for prolonged operations. Eur J Cardiothorac Surg 2005; 27: 373–8.
chylothorax due to lymphoma: a case series. Chest 1998; 114: 101. Weder W, Kestenholz P, Taverna C, et al. Neoadjuvant
731–5. chemotherapy followed by extrapleural pneumonectomy in
87. Strauss RM, Boyer TD. Hepatic hydrothorax. Semin Liver Dis 1997; malignant pleural mesothelioma. J Clin Oncol 2004; 22: 3451–7.
17: 227–32. 102. Liu HP, Yim AP, Izzat MB, et al. Thoracoscopic surgery for
88. Scholz DG, Nagorney DM, Lindor KD. Poor outcome from spontaneous pneumothorax. World J Surg 1999; 23: 1133–6.
peritoneovenous shunts for refractory ascites. Am J Gastroenterol 103. Hürtgen M, Linder A, Friedel G, et al. Video-assisted thoracoscopic
1989; 84: 540–3. pleurodesis. A survey conducted by the German Society for
89. Milanez de Campos JR, Filho LO, de Campos Werebe E, Sette H Jr, Thoracic Surgery. Thorac Cardiovasc Surg 1996; 44: 199–203.
Fernandez A, Filomeno LT, Jatene FB. Thoracoscopy and talc 104. Ohno K, Miyoshi S, Minami M, et al. Ipsilateral recurrence
poudrage in the management of hepatic hydrothorax. Chest 2000; frequency after video-assisted thoracoscopic surgery for primary
118: 13–17. spontaneous pneumothorax. Jpn J Thorac Cardiovasc Surg 2000;
90. Lazaridis KN, Frank JW, Krowka MJ, Kamath PS. Hepatic 48: 757–60.
hydrothorax: pathogenesis, diagnosis, management. Am J Med 105. Yim APC. Video-assisted thoracoscopic management of primary
1999; 107: 262–7. spontaneous pneumothorax. Ann Acad Med Singapore 1996; 25:
91. Thomas P, Le Mee F, Le Hors H, et al. Results of surgical treatment 668–72.
of persistent or recurrent pneumothorax. Ann Chir 1993; 47: 106. Bertrand PC, Regnard JF, Spaggiari L, et al. Immediate and long-
136–40. term results after surgical treatment of primary spontaneous
92. Weeden D, Smith GH. Surgical experience in the management of pneumothorax by VATS. Ann Thorac Surg 1996; 61: 1641–5.
spontaneous pneumothorax, 1972–82. Thorax 1983; 38: 737–43. 107. Freixinet J, Canalis E, Rivas JJ, et al. Surgical treatment of primary
93. Passlick B, Born C, Sienel W, Thetter O. Incidence of chronic pain spontaneous pneumothorax with video-assisted thoracic surgery.
after minimal-invasive surgery for spontaneous pneumothorax. Eur Respir J 1997; 10: 409–11.
Eur J Cardiothorac Surg 2001; 19: 355–8; discussion 8–9. 108. Chan P, Clarke P, Daniel FJ, et al. Efficacy study of video-assisted
94. Boutin C, Rey F, Vaillat J-R. Prevention of malignant seeding after thoracoscopic surgery pleurodesis for spontaneous pneumothorax.
invasive diagnostic procedures in patients with pleural Ann Thorac Surg 2000; 70: 253–7.
mesothelioma. Chest 1995; 108: 754–8. 109. Hatz RA, Kaps MF, Meimarakis G, et al. Long-term results after
95. Chappel A, Johnson A, Charles J, et al. A survey of the long-term video-assisted thoracoscopic surgery for first-time and recurrent
effects of talc and kaolin pleurodesis. Research Committee of the spontaneous pneumothorax. Ann Thorac Surg 2000; 70: 253–7.
British Thoracic Association and the Medical Research Council 110. Solaini L, Prusciano F, Bagioni P. Video-assisted thoracic surgery
Pneumoconiosis Unit. Br J Dis Chest 1979; 73: 285–8. in the treatment of pleural empyema. Surg Endosc 2007;
96. Lange P, Mortensen J, Groth S. Lung function 22–35 years after 21: 280–4.
treatment of idiopathic spontaneous pneumothorax with talc 111. Lackner RP, Hughes R, Anderson LA, et al. Video-assisted
poudrage or simple drainage. Thorax 1988; 43: 559–61. evacuation of empyema is the preferred procedure for
97. Brant A, Eaton T. Serious complications with talc slurry management of pleural space infections. Am J Surg 2000; 179:
pleurodesis. Respirology 2001; 6: 181–5. 27–30.
98. Janssen JP, Collier G, Astoul P, et al. Safety of pleurodesis with talc 112. Cassina PC, Hauser M, Hillejan L, et al. Video-assisted
poudrage in malignant pleural effusion: a prospective cohort thoracoscopy in the treatment of pleural empyema: stage-based
study. Lancet 2007; 369: 1535–9. management and outcome. J Thorac Cardiovasc Surg 1999; 117:
●99. Sugarbaker DJ, Jaklitsch MT, Bueno R, et al. Prevention, early 234–8.
detection and management of complications after 328 113. Cunniffe MG, Maguire D, McAnena OJ, et al. Video-assisted
consecutive extrapleural pneumonectomies. J Thorac Cardiovasc thoracoscopic surgery in the management of loculated empyema.
Surg 2004; 128: 138–46. Surg Endosc 2000; 14: 175–8.
This page intentionally left blank
49
Gene therapy in pleural diseases
The first requirement for successful gene therapy is effi- Pleural injection of plasmid/liposome or adenoviral
cient gene delivery. A variety of viral and non-viral gene vectors can produce biologically active proteins, but for
transfer vectors have been developed.2 As summarized in only for brief periods of time (i.e. days).8,9 In contrast, it
Table 49.1, each of these vectors has certain advantages has been recently demonstrated that intrapleural adminis-
with regard to DNA carrying capacity, types of cells tar- tration of AAV vectors can lead to high-level and persist-
614 Gene therapy in pleural diseases
ent systemic expression of transgenes.1,10,11 De et al.10 com- malignant pleural diseases are those related to lung or
pared the efficiency of an AAV5-based vector expressing pleural infection and those secondary to underlying sys-
a1-antitrypsin (a1AT) with that of an AAV2 vector temic diseases (i.e. congestive heart failure), neither of
expressing the same transgene by intrapleural and intra- which is particularly amenable to local therapy. However,
muscular routes in mice and found that the AAV5 vector one could image a situation, such as refractory air leaks,
achieved lung (bronchoalveolar lavage) and serum levels where it could be beneficial to introduce a gene (i.e.
that were 10-fold greater than the AA2 vector. At 40 weeks platelet-derived growth factor) that might accelerate
post-instillation, a1AT levels were a remarkable healing of bronchopleural fistulae.
900 mg/mL – 1.6-fold higher than the accepted therapeutic
serum level of 570 mg/mL. In a follow-up study,11 the same
group performed a comprehensive screen of 25 AAV Malignant pleural diseases
serotype vectors and showed the highest efficiencies were
seen with AAV5, AAV8, AAV9 and AAV rhesus monkey- In the near future, the most likely use for pleural gene
10 vectors. Application of this approach in non-human therapy will be in the treatment of malignant diseases
primate models is ongoing, but has not yet been published. including malignant mesothelioma (MM) and metastatic
Issues of importance would be how long the transfected pleural disease.12 A number of clinical trials have already
mesothelial cells would remain alive and secretory, as well been performed (see below). Pleural tumors have several
as development of potential immune responses against the characteristics that make them attractive targets for gene
transgene or AAV proteins. These studies raise the exciting therapy, including: (1) the absence of standard, effective
possibility of using AAV gene delivery to the pleural space therapy; (2) the accessibility of the pleural space for biopsy,
for genetic pulmonary diseases such as a1AT or other sys- vector delivery and analysis of treatment effects; and (3)
temic diseases, such as hemophilia, lysosomal storage dis- the availability of therapeutic strategies that require only
orders or diabetes.1 transient gene expression. MM is an especially attractive
target since local extension of disease, rather than distant
metastases, is responsible for much of its morbidity and
GENE THERAPY FOR THE TREATMENT OF mortality.
PLEURAL DISEASES
It is currently difficult to envision a large number of clini- A large number of gene therapy strategies for pleural
cal scenarios where pleural gene therapy would be both malignancy have been explored using cell culture and
useful and cost-effective. The two large classes of non- animal models (Table 49.2).
Gene therapy for the treatment of pleural diseases 615
Principal Examples
Suicide gene therapy Herpes simplex thymidine kinase gene plus ganciclovir
Cytosine deaminase gene plus 5-flurocytosine
Induction of apoptosis p53, p16, p14ARF, Bak, Anti-sense SV40-T antigen
Anti-angiogenesis Soluble form of the VEGF receptor (Flt-1), anti-angiogenic pigment
epithelium-derived factor
Immunogene therapy
Cytokine therapy Interleukin-2, interleukin-12, Type 1 and Type 2 interferons, GM-CSF
Non-specific induction of innate and acquired immunity Liposome/DNA complexes, mycobacterial heat shock protein gene
ligand (HSP-65), anti-CD40
Vaccination SV40 T-antigen
Tumor-selective replicating virus Herpes virus, Vaccinia virus, Adenovirus
GM-CSF, granulocyte macrophage colony-stimulating factor; SV40, Simian virus 40; VEGF, vascular endothelial growth factor.
Delivery of the wild-type p53 gene has been the most fre- An attractive approach in cancer gene therapeutics is ‘sui-
quent method of experimental gene therapy of solid cide’ gene therapy where a neoplasm is transduced with a
tumors, since mutations in the p53 tumor suppressor gene cDNA encoding for an enzyme rendering tumor cells sen-
account for the majority of genetic abnormalities in solid sitive to a benign agent by converting the ‘prodrug’ to a
tumors. Even though most MMs contain wild-type p53, toxic metabolite. Enzymes used most commonly are the
over-expression of p53 using an adenoviral vector has herpes simplex virus-1 thymidine kinase (HSVtk) gene
inhibited cell growth.13 Other molecular approaches used which makes cells sensitive to the nucleoside analog ganci-
in mesotheliomas have been re-expression of p16INK4a,14 or clovir (GCV) or the yeast enzyme cytosine deaminase
the p14ARF protein complex.15 An alternate method of which converts 5-fluorocytosine to the toxic 5-fluorouracil.
inhibiting MM cells is the introduction of ‘downstream’ Therapeutic efficacy in the former is enhanced by a
promoters of apoptosis such as the pro-apoptotic Bcl-2 ‘bystander’ effect that involves passage of toxic GCV
family member Bak.16 The potential role for SV40 as a metabolites from transduced to non-transduced cells via
causative factor in MM oncogenesis and proliferation is gap junctions or apoptotic vesicles and induction of anti-
the rationale for experiments by Schrump and colleagues tumor immune responses capable of killing tumor cells not
showing that anti-sense oligonucleotides designed to abro- expressing the transgene.
gate Simian virus 40 (SV40) Tag expression induce apop- The transfer of HSVtk DNA to target pleural or peri-
tosis and enhance sensitivity to chemotherapeutic agents toneal tumor cells has been accomplished using a variety
in SV40-positive MM cells in vitro.17 of delivery systems including carrier cells,20,21 lipo-
Given that in vivo gene transfer is quite inefficient, a somes,22,23 plasmid DNA–polyethylenimine complexes22
major limitation of all of these approaches is that inducing and AAV2 vectors.24 The most effective vector, however,
cell death in only the transduced cells (without bystander has been adenovirus. Initial experiments demonstrated
effects), will have only limited therapeutic efficacy. that replication-deficient adenoviral HSVtk vectors
(Ad.HSVtk) efficiently transduced mesothelioma cells
ANTI-ANGIOGENESIS both in tissue culture and in animal models and facilitated
HSVtk-mediated killing of human MM cells in the pres-
Inhibiting tumor blood vessel growth as a treatment for ence of low concentrations of GCV.25 Subsequently,
cancer has been an area of active interest, resulting in a Ad.HSVtk/GCV gene therapy was used successfully to treat
number of approved therapies, such as an anti-vascular established intraperitoneal human MM tumors and lung
endothelial cell growth factor (VEGF) antibody. cancers in immunodeficient mice26 and in rat models of
Adenoviral vectors have been used to deliver a soluble pleural MM.5,27
form of the VEGF receptor (Flt-1) or the anti-angiogenic
pigment epithelium-derived factor to the pleural space of IMMUNO-GENE THERAPY
mice with lung cancers or MMs with some success.18,19 The
short duration of expression of Ad (adenoviral) vectors Innate and adaptive anti-tumor immune responses can be
will likely limit this strategy in the clinic, but development elicited by delivering non-specific immunostimulatory
of AAV vectors expressing anti-angiogenic molecules genes. Delivery of the immunogenic heat shock protein
(including antibodies) may be an effective primary treat- gene-65 via cationic liposomes showed efficacy in a syn-
ment or chemotherapy adjuvant in the future. geneic murine model of MM. This effect, however,
616 Gene therapy in pleural diseases
appeared to be related to non-specific effects of CD8+ T lymphocytes in pleural fluid but no clinical
lipid–pDNA complexes likely due to the unmethylated responses were noted.
CpG motifs of the prokaryotic DNA in the vector plasmids Sterman et al.46–49 conducted a series of Phase 1 clinical
that were sufficient to activate ‘danger signals’ and initiate trials of a replication-incompetent adenoviral vector
innate and adaptive anti-tumor immune responses.28,29 encoding HSVtk (Ad.HSVtk) delivered intrapleurally (fol-
Cytokines are known to have both direct anti-prolifer- lowed by ganciclovir) into more than 30 patients with
ative effects upon mesothelioma cells, as well as activating pleural MM. Dose-limiting toxicity was not reached, side
intrapleural and intratumoral immune effector cells in effects were minimal and gene transfer was confirmed in a
vivo. Several published Phase I and Phase II clinical trials dose-related fashion with clearly detectable gene transfer
have documented MM tumor responses to intrapleural (indicated by immunostaining) at tumor surfaces and up
infusion of interleukin-2 (IL-2), interferon beta (IFN-b), to 30–50 cell layers deep (Figure 49.1). Anti-tumor anti-
and interferon gamma (IFN-g). The rationale for the use bodies and strong anti-adenoviral immune responses,
of gene therapy is that expression of cytokine genes by including high titers of neutralizing antibody and prolifer-
tumor cells generates high levels of intratumoral cytokines ative T-cell responses, were generated, but with no obvious
in a paracrine fashion, inducing powerful local cytokine adverse clinical effects. Interestingly, a number of clinical
effects without significant systemic toxicity. responses were seen at the higher dose levels, including
Animal studies showing good anti-MM efficacy have two patients that remain alive (one without disease) more
been published using a variety of pro-inflammatory than 8 years after vector instillation.49
cytokines including IL-2,30 IL-12,31 granulocyte-mono-
cyte-colony stimulating factor (GM-CSF),32,33 IFN-g34,35,
IL-24 (mda-7)36 and CD40-ligand.37
Our group has explored the use of genes encoding type
I (a, b) interferons. A single intraperitoneal (i.p.) injection
of a recombinant adenovirus engineered to express the
murine b-interferon gene (Ad.muIFN-b) eradicated syn-
geneic murine MM in >90 percent of animals tested.38
Intraperitoneal Ad.muIFN-b gene therapy also resulted in
a significant reduction of subcutaneous tumors at a distant
site and was mediated by CD8+ T lymphocytes.39
Clinical trials
(b)
A cell-transfer trial was conducted using an irradiated Figure 49.1 Gene transfer into mesothelioma. Forty-eight hours
ovarian carcinoma cell line retrovirally transfected with after intrapleural gene transfer, tumor biopsies were taken and
HSVtk (PA1-STK cells). Cells were instilled via an stained for the transgene Herpes simplex virus (HSV)-thymidine
indwelling pleural catheter followed by systemic adminis- kinase. Immunohistochemical staining (red color) showed clear
tration of GCV.21,45 Minimal side effects were seen; there expression in the surface layers of tumor cells in both nuclear (a)
were some post-treatment increases in the percentage of and cytoplasmic (b) locations. (See also Color Plates 57 and 58.)
Key points 617
Zarogoulidis et al.50 treated six lung cancer patients only 2 weeks, inhibiting transgene production after the
with malignant pleural effusions intrapleurally with an second dose. Future plans are to administer two doses of
adenovirus encoding the suicide gene cytosine deaminase Ad.IFNb more closely together in time and in combina-
followed by the prodrug flucytosine for 14 days. The treat- tion with chemotherapy for patients with MM and
ment was well tolerated and inhibited pleural fluid reaccu- metastatic pleural effusions from non-small cell lung
mulation in two patients, but no clear clinical responses cancer (NSCLC). Based on other preclinical studies,55 a
were noted. neoadjuvant surgery trial is also being planned where
Vaccinia virus (VV) is a double-stranded DNA virus vector would be given to patients with mesothelioma fol-
that relies primarily on its own proteins for DNA replica- lowed by a maximal debulking procedure and adjuvant
tion and mRNA synthesis, and so has minimal interactions chemo-radiotherapy.
with host proteins, facilitating production of daughter
viral particles immediately after cell entry. Owing to its
role in the eradication of smallpox, it has been used exten- CONCLUSIONS
sively in humans.51 Mukherjee et al.44 injected a replica-
tion-restricted recombinant VV expressing the human Pleural gene therapy could potentially be used to treat sys-
IL-2 gene intra-tumorally into patients with MM. temic diseases by providing a large and easily accessible
Toxicities were minimal and there was no clinical or sero- cellular target for gene transduction and protein secretion
logical evidence of spread of virus to patient contacts. No or for the treatment of a number of pleural diseases.
significant tumor regression was seen in any of the patients Studies using novel AAV serotypes are especially promis-
and only modest intra-tumoral T-cell infiltration was ing in this regard. The other area where pleural gene
detected. VV IL-2 mRNA was detected by reverse tran- therapy is likely to become useful is in the treatment of
scriptase-polymerase chain reaction (PCR) in serial tumor pleural diseases, especially pleural malignancies. A number
biopsies for up to 6 days after injection, but declined to low of clinical trials in cancer patients have been completed
levels by day 8. that show the safety and feasibility of this approach.
The use of a cellular vector to deliver the IL-2 gene in Gene therapy has not yet been proven as a useful thera-
patients with pleural MM has been reported in abstract peutic tool for the treatment of pleural diseases. However,
form.52 Fourteen patients received intratumoral injections the field is less than 15 years old and great initial progress
(four courses) of xenogenic fibroblasts (Vero cells) has been made. The authors have no doubt that gene
expressing IL-2. The treatment was well tolerated and cir- therapy will be an important treatment strategy in the near
culating levels of IL-2 were detected in seven patients. One future.
patient showed temporary tumor shrinkage and one had
disease stabilization for 4 months, but this approach is not
being pursued further.
Based on the preclinical data described above, our
group at the University of Pennsylvania has recently com-
pleted a Phase 1 dose escalation study evaluating the safety KEY POINTS
and feasibility of single-dose intrapleural IFN-b gene
transfer using an adenoviral vector (Ad.IFNb) in seven ● Gene therapy (the treatment of disease based
patients with MM and three patients with metastatic upon the transfer of genetic material) involving
pleural effusions (MPE).53,54 Intrapleural Ad.INFb was the pleural space offers a number of potential
generally well tolerated. Gene transfer was documented in advantages and could be used to produce
seven of the ten patients by demonstration of IFN-b secreted gene products or in the treatment of
message or protein in pleural fluid. Anti-tumor immune pleural diseases.
responses were elicited in seven of the ten patients, includ- ● AAV vectors have been shown to lead to high-
ing humoral responses to known tumor antigens (SV40 level and persistent expression of transgene and
virus large T-antigen and mesothelin) and unknown are being investigated for the treatment of a1-
tumor antigens (seven patients). Four of ten patients antitrypsin deficiency.
showed meaningful clinical responses defined as disease ● The most likely use for pleural gene therapy will
stability and/or regression on fluorodeoxyglucose be in the treatment of malignant diseases includ-
positron-emission tomograpy (FDG-PET) and computed ing MM and metastatic pleural disease.
tomography (CT) scans at day 60 post-vector infusion.18 ● A large number of gene therapy strategies for
Given these results, a second Phase 1 trial, using two pleural malignancy have been explored using cell
doses of Ad.IFNb separated by 2 weeks was conducted in culture and animal models, including induction
10 patients. Vector was generally well tolerated and of apoptosis, anti-angiogenesis, suicide gene
induced similar anti-tumor humoral immune responses, therapy, immuno-gene therapy and replicating,
along with two clinical responses. However, very high tumor-selective viruses.
titers of neutralizing anti-Ad antibodies were induced after
618 Gene therapy in pleural diseases
by surgical debulking of tumor. Cancer Gene Ther 2001; 8: 45. Harrison LH Jr, Schwarzenberger PO, Byrne PS, et al. Gene-
580–8. modified PA1-STK cells home to tumor sites in patients with
33. Triozzi PL, Aldrich W, Allen KO, et al. Antitumor activity of the malignant pleural mesothelioma. Ann Thorac Surg 2000; 70:
intratumoral injection of fowlpox vectors expressing a triad of 407–11.
costimulatory molecules and granulocyte/macrophage colony ●46. Sterman DH, Treat J, Litzky LA, et al. Adenovirus-mediated herpes
stimulating factor in mesothelioma. Int J Cancer 2005; 113: simplex virus thymidine kinase/ganciclovir gene therapy in
406–14. patients with localized malignancy: results of a phase I clinical
34. Gattacceca F, Pilatte Y, Billard C, et al. Ad-IFN gamma induces trial in malignant mesothelioma. Hum Gene Ther 1998; 9:
antiproliferative and antitumoral responses in malignant 1083–92.
mesothelioma. Clin Cancer Res 2002; 8: 3298–304. ●47. Molnar-Kimber KL, Sterman DH, Chang M, et al. Impact of
35. Cordier Kellerman L, Valeyrie L, Fernandez N, et al. Regression of preexisting and induced humoral and cellular immune responses in
AK7 malignant mesothelioma established in immunocompetent an adenovirus-based gene therapy phase I clinical trial for
mice following intratumoral gene transfer of interferon gamma. localized mesothelioma. Hum Gene Ther 1998; 9: 2121–33.
Cancer Gene Ther 2003; 10: 481–90. 48. Sterman DH, Molnar-Kimber K, Iyengar T, et al. A pilot study of
36. Cao XX, Mohuiddin I, Chada S, et al. Adenoviral transfer of mda-7 systemic corticosteroid administration in conjunction with
leads to BAX up-regulation and apoptosis in mesothelioma cells, intrapleural adenoviral vector administration in patients with
and is abrogated by over-expression of BCL-XL. Mol Med 2002; 8: malignant pleural mesothelioma. Cancer Gene Ther 2000; 7:
869–76. 1511–8.
37. Friedlander PL, Delaune CL, Abadie JM, et al. Efficacy of CD40 ●49. Sterman DH, Recio A, Vachani A, et al. Long-term follow-up of
ligand gene therapy in malignant mesothelioma. Am J Respir Cell patients with malignant pleural mesothelioma receiving high-dose
Mol Biol 2003; 29: 321–30. adenovirus herpes simplex thymidine kinase/ganciclovir suicide
●38. Odaka M, Sterman DH, Wiewrodt R, et al. Eradication of gene therapy. Clin Cancer Res 2005; 11: 7444–53.
intraperitoneal and distant tumor by adenovirus-mediated 50. Zarogoulidis K, Kontakiotis T, Papagiannis A, Xafenias A, Kortsaris
interferon-beta gene therapy is attributable to induction of A. Management of resistant lung cancer malignant pleural
systemic immunity. Cancer Res 2001; 61: 6201–12. effusion by intrapleural gene therapy. J Clin Oncol 2004; 22: 3168.
39. Odaka M, Wiewrodt R, DeLong P, et al. Analysis of the ◆51. Shen Y, Nemunaitis J. Fighting cancer with vaccinia virus:
immunologic response generated by Ad.IFN-beta during successful teaching new tricks to an old dog. Mol Ther 2005; 11: 180–95.
intraperitoneal tumor gene therapy. Mol Ther 2002; 6: 210–8. 52. Pitako J, Squiban P, Acres B, Digel W. A randomized phase II single
40. Zhu ZB, Makhija SK, Lu B, et al. Incorporating the survivin center study of gene transfer-based non-specific immunotherapy
promoter in an infectivity enhanced CRAd-analysis of oncolysis of malignant mesothelioma (MM) by intratumoral injections of an
and anti-tumor effects in vitro and in vivo. Int J Oncol 2005; 27: interleukin-2 producing vero cells. Proc Am Soc Clin Oncol 2003;
237–46. 22(abstr 920).
●41. Kucharczuk JC, Randazzo B, Chang MY, et al. Use of a ‘replication- ●53. Sterman DH, Gillespie CT, Carroll RG, et al. Interferon beta
restricted’ herpes virus to treat experimental human malignant adenoviral gene therapy in a patient with ovarian cancer. Nat Clin
mesothelioma. Cancer Res 1997; 57: 466–71. Pract Oncol 2006; 3: 633–9.
42. Adusumilli PS, Stiles BM, Chan MK, et al. Imaging and therapy of 54. Sterman DH, Recio A, Carroll RG, et al. A Phase I clinical trial of
malignant pleural mesothelioma using replication-competent single-dose intrapleural interferon-beta gene transfer for
herpes simplex viruses. J Gene Med 2006; 8: 603–15. malignant mesothelioma and metastatic pleural effusion: high
43. Stiles BM, Adusumilli PS, Bhargava A, et al. Minimally invasive rate of antitumor immune responses. Clin Cancer Res 2007; 13(15
localization of oncolytic herpes simplex viral therapy of metastatic Pt1): 4456–66.
pleural cancer. Cancer Gene Ther 2006; 13: 53–64. 55. Kruklitis RJ, Singhal S, DeLong P, et al. Immunogene therapy with
●44. Mukherjee S, Haenel T, Himbeck R, et al. Replication-restricted interferon-b before surgical debulking delays recurrence and
vaccinia as a cytokine gene therapy vector in cancer: persistent improves survival in a murine model of malignant mesothelioma. J
transgene expression despite antibody generation. Cancer Gene Thorac Cardiovasc Surg 2004; 127: 123–30.
Ther 2000; 7: 663–70.
This page intentionally left blank
50
Future directions
Introduction 621 Our predictions for the evolution of pleural disease 623
Pleural disease is common and important 621 during the first part of the twenty-first century
Recent advances and controversies 622 Key points 626
Current state of affairs 623 References 626
To bring forward the field of pleural disease 623
pleural involvement is not only common with internal (tPA), recombinant human deoxyribonuclease (rhDNase),
medical conditions, but often crosses the boundaries of both tPA and rhDNase, or placebo in the management of
other specialities, such as obstetrics and gynecology, and complicated parapneumonic effusions.
frequently complicates surgical procedures. Thus, it is A randomized controlled study comparing talc slurry
important that physicians of any specialty possess a good and talc insufflation for the treatment of malignant pleural
understanding of basic pleural anatomy, physiology and effusions with more than 400 patients provided valuable
knowledge of common pleural pathologies. Unfortunately, data showing no differences between the two methods of
clinicians often view pleural pathology as a ‘side issue’ of talc delivery in the primary endpoints.4 A randomized trial
the principal diagnosis. Delays in the recognition and showed that small particle size talc induced pulmonary
management of pleural involvement of various diseases are and systemic inflammation compared with talc prepara-
common occurrences in both primary and tertiary centers. tions with larger median particle size.5 The hypothesis that
Pleural disease, as shown in the preceding chapters, is small talc particles are responsible for the inflammation is
an exciting and dynamic field. New disease entities are further supported by a large prospective observation study
continually being recognized. For example, pleural effu- from 14 European and South African centers demonstrat-
sions are a recognized complication of organ transplanta- ing no cases of acute respiratory distress syndrome
tions and coronary artery bypass grafting (CABG), which (ARDS) following pleurodesis using calibrated talc prepa-
are likely to be of greater significance as more of these rations.6 Another study is underway to compare the value
operations are performed each year worldwide. The of tunneled indwelling catheter (without pleurodesis)
Human immunodeficiency virus (HIV) epidemic has seen a versus conventional talc pleurodesis as first-line treatment
dramatic increase in effusions due to Karposi’s sarcoma. for malignant pleural effusions. This trial will use breath-
The increasing application of in vitro fertilization has led to lessness score and quality of life measures as its key end-
a new complication – the ovarian hyperstimulation syn- points.
drome – which often presents with large pleuro-peritoneal Malignant pleural mesothelioma continues to be an
effusions. active area of research and controversies. Two large multi-
The epidemiology of pleural diseases is dependent upon center trials with 456 and 250 patients were the first ran-
socioeconomical and geographic factors. The causative domized studies to demonstrate treatment (with
agents for pleural space infection and their sensitivities pemetrexed7 and raltitrexed,8 respectively) can prolong
vary from region to region, and with time. In developing survival in mesothelioma. The role of Simian virus 40
countries, TB pleuritis is significantly more common than (SV40) in mesothelioma development continues to be
in the developed countries. The incidence of asbestos- debated; new evidence casts further doubts on its role.9
related pleural diseases, especially malignant mesothe- Prophylactic radiotherapy for pleural puncture sites in
lioma, varies dramatically by geographic regions and mesothelioma patients, once well accepted, is now being
reflects the level of asbestos mining or processing several challenged after two randomized studies showed no signif-
decades before. It is for all these reasons that pleural icant benefits.10,11
disease remains a challenging field and for which a contin- The growth in clinical research in pleural disease is not
uous update of knowledge is necessary. paralleled by a significant increase in basic research of
pleural pathologies. For example, little research has been
undertaken in the pharmacokinetics of drug delivery to the
RECENT ADVANCES AND CONTROVERSIES pleural space. Likewise, despite the fact that genetic
research is a flourishing area in modern medicine, there
Clinicians continue to strive for better management of has been practically no research work performed on
pleural diseases with new strategies being developed con- genetic aspects of pleural diseases, other than the genetics
stantly. Controversies continue to arise as new treatment of mesothelioma development.
options are developed. It is heartening to note that several So why is pleural disease research being neglected? To
large multi-center clinical trials have been completed since start with, pleural disease is not commonly recognized as a
the previous edition of this book, all of which impact on sub-specialty of its own right. In most cases, pleural disease
clinical practice worldwide. The Multi-center Intrapleural is managed by specialists who are involved with a narrow
Sepsis Trial (MIST)2 randomized 454 patients with com- spectrum of the pleural involvement in their own specialty
plicated parapneumonic effusions to intrapleural streptok- practice, who often see pleural disease as a ‘side issue’. For
inase or saline and showed no benefit from the example, oncologists are more likely to focus their research
streptokinase. This was corroborated by a smaller study in management of the primary tumors than in malignant
from South Africa3 which also showed no difference pleuritis; surgeons often regard post-operative pleural
between the two groups at day 3 after treatment. involvement as a nuisance rather than recognizing it as an
Controversies exist, however, as there was a small benefit important morbidity; infectious disease specialists caring
at day 7 in the latter trial favoring streptokinase. A large for patients with pneumonia are more interested in select-
multicenter trial (MIST2) is presently under way in the UK ing the proper antibiotics than draining the parapneu-
which randomizes patients to tissue plasminogen activator monic effusion.
Our predictions for the evolution of pleural disease during the first part of the twenty-first century 623
Lack of commercial interest and drug company spon- TO BRING FORWARD THE FIELD OF PLEURAL
sorship is a significant disadvantage in pleural research, DISEASE
when compared with more ‘marketable’ conditions such
as asthma or pneumonia. Many physicians also view To improve the clinical practice of pleural disease, first and
pleural disease (especially malignant effusions) as a termi- foremost we need to increase the awareness among physi-
nal event not worthy of efforts to improve its practice. The cians of the significance of pleural diseases, and the conse-
lack of funding and interest result in a lack of good quences of overlooking pleural abnormalities. This should
research, making it difficult to attract funding and recruit take place in different levels simultaneously. Pul-
talented young investigators. The net result is a vicious monologists and thoracic surgeons should be encouraged
cycle in which pleural disease research continues to be to develop sub-specialty interests in the investigation and
ignored. management of pleural diseases. Respiratory trainees
The pleural disease research community lacks a central should have mandatory experience and be competent in
organization to liase and put together investigators of basic pleural procedures (e.g. thoracentesis, chest tube
similar interests. This partly explains the relative lack of insertion and pleural biopsies). Junior staff, medical stu-
large multi-center collaboration efforts. There exist no dents and paramedical staff should be adequately educated
pleural disease interest groups within the major thoracic to handle pleural emergencies, such as tension pneumo-
societies (e.g. the American Thoracic Society, European thoraces, as well as to manage common pleural conditions
Respiratory Society or Thoracic Society of Australia and such as the appropriate approach to a patient with a
New Zealand), as there are for airways diseases, pulmonary pleural effusion or the day-to-day management of chest
vascular disease or lung fibrosis. There are no large patient tubes. Development of clinical guidelines for common
support groups, such as those for asthma or lymphangio- pleural presentations may also help staff unfamiliar with
leiomyomatosis. No specialized forum exists for these diseases.
researchers in pleural disease to exchange ideas. Unlike for In the long run, it is important that pleural disease be
asthma or lung cancer, there are no scientific publications recognized as a specialty. Physicians should be encouraged
with a focus on pleural disease research. The International to refer patients with difficult pleural problems to pulmo-
Pleural Newsletter, a free publication available at nologists with a special interest in pleural diseases. In the
www.musc.edu/pleuralnews, is an attempt to provide bigger picture, developing a global league of interested
useful knowledge on pleural diseases to the wider medical physicians is the best way to raise the profile of pleural
community, with the hope of promoting interests and disease within the professional communities. Such an
research in pleural disease. organization will facilitate the performance of multi-
center trials often necessary to answer crucial clinical ques-
tions. Having regular and dedicated forums for discussion
CURRENT STATE OF AFFAIRS of current issues in pleural diseases, such as regular confer-
ences and publications, is most worthwhile.
As a consequence of the relative lack of evidence-based Improving funding and resources for pleural disease
medicine, clinical practices in pleural disease are com- research is essential to improve its standards. Programs to
monly based on results of small trials, on ‘traditional’ prac- raise public awareness will help attract research funding.
tice or on anecdotal experience of individual physicians. Clinicians should work together with scientists, as the need
Hence, there are major variations in practice among prac- to apply the latest technology in pleural research cannot be
titioners of different countries, or sometimes even differ- over-emphasized.
ent units within the same city.
Very often, adoption of advances in pleural disease
management (e.g. thoracoscopy or using adenosine deam- OUR PREDICTIONS FOR THE EVOLUTION OF
inase [ADA] to diagnose tuberculosis) is relatively slow by PLEURAL DISEASE DURING THE FIRST PART
clinicians. The lack of interest and proper recognition of OF THE TWENTY-FIRST CENTURY
the importance of pleural disease are not without practical
consequences. Around the world daily, there are patients Our first prediction is that at least 40 percent of what is
whose pleural effusions are not investigated early. Delay in written in this book will be proven wrong or at least out-
drainage or management of parapneumonic effusions dated over the next 20 years. The difficult aspect is to
happens regularly throughout the world, despite the predict which 40 percent is wrong. We readily admit that
repeated call that ‘the sun should never set on a parapneu- we are not clairvoyants and do not necessarily know any
monic effusion’. Some clinicians are not comfortable in more than do other people who are interested in pleural
dealing with pleural diseases while others think it is a ‘pure disease. Nevertheless, we will offer the following predic-
nuisance’. Hence, many clinicians tend to either ignore tions, albeit with some trepidation. Although authors of
important clinical presentations or have a low threshold the clinical chapters have stated their own views about the
for referring patients to other specialists, including sur- future of their specific area of pleural disease, our predic-
geons. tions and their predictions differ on many occasions.
624 Future directions
Indeed, the two editors do not necessarily agree in our pre- ual susceptibility to various pleural diseases.
dictions. Pharmacogenetics may also help explain the differ-
ences in inter-individual responses to treatments of
pleural diseases.
Overall predictions ii. Powerful screening tools (e.g. microarray technol-
ogy and proteomics) will be applied to hasten
The incidence of pleural diseases will continue to rise. This advances of our knowledge in the pathophysiology
is because the general population will live longer and the of pleural diseases. These tools may help diagnose
elderly population is more likely to develop diseases asso- the cause of pleural effusions, predict therapeutic
ciated with pleural effusions such as congestive heart responses and prognosis of patients with various
failure, malignancy and pneumonia. As medical practice conditions.
evolves, there are more interventions that lead to pleural 4. Gene therapy is only in its infantile stage and numerous
disease (e.g. transthoracic needle aspiration and radiofre- hurdles lie ahead, but continual advances will be made
quency ablation causing pneumothorax; CABG surgery towards large-scale clinical application. Gene therapy
and fertility induction producing pleural effusions). can potentially be applied: using the pleural space as a
Historically, high-profile pleural researchers are pre- gene receptacle for systemic gene therapy; and using
dominantly based in the USA, from where a large propor- local gene therapy in the pleural space to treat pleural
tion of high-impact clinical studies and cutting edge basic disorders such as mesothelioma or pleural metastases,
research on pleural diseases were published. With rising or to produce pleurodesis.
interests in pleural disease worldwide and globalization of 5. The future direction for the management of pleural
technology, YCGL predicts that pleural research will be effusions should lie in targeting pleural fluid formation,
more evenly distributed around the world. European and rather than to manage the fluid after it is formed with
Australian centers have already taken a lead in pleural procedures such as pleurodesis. We predict that signifi-
infection and asbestos-related pleural diseases. It would be cant advances will be gained in other disciplines (such
encouraging to see the number of centers with pleural as research in pulmonary edema) that target the funda-
interests continue to grow around the world. mental control of fluid exudation.
6. There is an urgent need to elucidate the pharmacoki-
netics of the pleura. How best to deliver pharmacologi-
Basic science predictions cal agents to the pleura in adequate concentrations, and
how to enhance removal of pleural substances are
1. There will be further increases in the recognition of the important yet unstudied questions. We are not entirely
importance of resident pleural mesothelial cells in optimistic, however, that such questions would attract
pleural diseases (e.g. inflammation, fibrosis, fluid accu- much interest (for the reasons outlined earlier in this
mulation). The differing roles of the mesothelial cells chapter).
and pleural macrophages in the pleura will be defined. 7. At present, basic research in mesothelioma accounts for
This contrasts with the focus of previous research, which a huge proportion of laboratory research in all pleural
tended to concentrate on cell types that are recruited diseases. With the expected peak and the subsequent
into the pleural cavity (e.g. neutrophils and fibroblasts). decline in the incidence of mesothelioma, the propor-
2. Cytokines and their manipulation will provide the most tion of resources in pleural diseases that goes into
immediate areas of advances in understanding the mesothelioma research will eventually decrease.
pathophysiology of pleural disease in the next 10–20
years. This area will provide the basis for novel treat-
ment options. Advances in the knowledge of intracellu- Clinical practice predictions
lar signaling mechanisms will provide the focus for the
next wave of basic science research in pleural disease, Diagnosis of pleural disease:
though clinical benefits from this knowledge remain to
be seen. 1. The separation of pleural effusions into transudates and
3. Genetics in pleural diseases: exudates will be forgotten. This division really is artifi-
i. Genetics is a broad new field of active research in all cial. Instead, patients will be given a specific diagnosis,
areas of medicine, yet few steps have been taken to possibly by new diagnostic methods that allow the
apply the advances in genetics in pleural research. determination of the pleural fluid characteristics in a
Why do some people develop pleural fibrosis and non-invasive fashion.12 An example of this is using the
others do not following the same exposure to irradi- NT-pro BNP levels to establish the diagnosis of conges-
ation, asbestos or drugs? Why do certain patients tive heart failure.
develop effusions but others with the same condi- 2. Non-radiologists, especially chest physicians, critical
tion do not? We predict that genetic predisposition care physicians, emergency room physicians and tho-
will explain at least part of the differences in individ- racic surgeons will perform ultrasound more frequently
Our predictions for the evolution of pleural disease during the first part of the twenty-first century 625
themselves for the diagnosis and management of tion are concerns on the long-term use of indwelling
pleural disease. This prediction has already been proven catheters.
true since the last edition of this book. 2. YCGL advocates that we should scrutinize our current
3. Advances in laboratory diagnostic techniques (e.g. concepts of management of malignant pleural effu-
polymerase chain reaction, PCR) will allow identifica- sions. The aim should be to improve quality of life, in
tion of most infective organisms in parapneumonic part via preventing dyspnea from fluid accumulation.
effusions, pleural tuberculosis and pleural fungal dis- YCGL predicts that more clinical studies will adopt
eases. quality of life (instead of radiographic improvement) as
4. Many of the pleural effusions which are presently principle endpoints. RWL believes that quality of life is
labeled as idiopathic will be proven to be due to viral very important but feels that studies on this are difficult
infections by more sophisticated diagnostic technology. because distinguishing symptom progression due to
5. Better imaging techniques will provide better delin- recurrence of the effusion or progression of the tumor
eation of the invasion of the underlying tissue in the is often difficult. Talc will gradually be replaced by
pleura by malignancy. Such techniques will be used as other compounds. When talc is used, it will only be
part of the routine workup of possible pleural malig- with large particle size talc. Talc is an inhomogeneous
nancies. substance that is basically dirt and is associated with sig-
6. Identification of the cellular origin of metastatic pleural nificant pulmonary problems. RWL maintains that in
malignancies will become more important as more the third millennium we should be able to produce a
malignancies become amenable to chemotherapy. pleurodesis with more sophisticated agents. YCGL
believes that research efforts should focus on stopping
Treatment of parapneumonic effusion: fluid accumulation rather than secondary prevention
such as pleurodesis.
1. RWL predicts that the combination of fibrinolytics and
3. RWL predicts that low doses of silver nitrate (20 mL 0.5
human recombinant DNAase, but not fibrinolytics by
percent) or iodopovidone (Betadine) (100 mL 2.0
themselves, will be cost effective in the treatment of
percent ) will become the most common agents for
parapneumonic effusions.
pleurodesis in developing countries because they are
2. RWL predicts that monoclonal antibodies or receptor
effective, inexpensive, widely available and are associ-
agonists for the cytokines responsible for the fibrosis
ated with tolerable side effects. In developed countries,
and pleural loculations will be used therapeutically to
pleurodesis will be created by using more sophisticated
prevent pleural fluid loculation and fibrous covering of
agents (such as TGF-b) that can produce a pleurodesis
the visceral pleura. This treatment will follow the eluci-
without inducing pleural inflammation. YCGL would
dation of the cytokines responsible for the pleural fluid
like to reserve judgment until more clinical data is
loculations and pleural fibrosis that make treatment of
available.
this disease difficult. A leading candidate as an agent
4. Therapies targeting the actual pathophysiology of
that is responsible for the loculations and fibrosis is
pleural fluid formation, such as inhibitors of vascular
transforming growth factor beta (TGF-b). YCGL is less
endothelial growth factor, will be used to decrease
optimistic along this vein and predicts that application
pleural fluid production in malignant pleural effusions.
of cytokine manipulation for therapeutic use would
turn out to be very complicated due to the multifunc-
Management of pneumothorax:
tional nature of most of the cytokines.
1. RWL predicts that more patients with primary sponta-
Treatment of malignant pleural effusion:
neous pneumothorax will be initially managed with
1. RWL predicts that the increase in the use of outpatient aspiration and not be hospitalized. YCGL notes that
therapy with indwelling cathers for malignant pleural RWL is out of date and that this was already happening
effusion seen since the previous edition of this book will in many countries before the millennium. If this aspira-
continue. This will be done with tunneled indwelling tion is successful, the patients will be sent home with a
pleural catheters. When the amount of drainage with an device which will measure the amount of air leak.
indwelling catheter shows no strong tendency to Patients with primary spontaneous pneumothorax who
decrease, attempts to create a pleurodesis will be made are not initially managed successfully with aspiration or
by injecting pleurodesing agents through the indwelling who have a recurrent pneumothorax will be subjected
catheter. YCGL agrees with such an approach but pre- to early thoracotomy with stapling of blebs and pleural
dicts that general use of indwelling catheter treatment abrasion.
may prove difficult. Cost is likely to be a prohibitory 2. Iatrogenic pneumothoraces will be managed more fre-
factor in the near future. Such a catheter system quently with observation or aspiration and less fre-
requires either the patient or a dedicated caregiver to quently with tube thoracostomy.
manage the drainage in an aseptic fashion, which is not 3. YCGL predicts that the underlying genetic or anatomi-
always easy. Catheter blockage and bacterial coloniza- cal features that predispose to recurrence of pneumo-
626 Future directions
9. Lopez-Rios F, Illei PB, Rusch V, Ladanyi M. Evidence against a role 11. O’Rourke N, Garcia JC, Paul J, et al. A randomised controlled trial
for SV40 infection in human mesotheliomas and high risk of false- of intervention site radiotherapy in malignant pleural
positive PCR results owing to presence of SV40 sequences in mesothelioma. Radiother Oncol 2007; 84: 18–22.
common laboratory plasmids. Lancet 2004; 364: 1157–66. 12. Lee YCG, Davies RJO, Light RW. Diagnosing pleural effusion:
10. Bydder S, Phillips M, Joseph DJ, et al. A randomised trial of single- moving beyond transudate-exudate separation. Chest 2007; 131:
dose radiotherapy to prevent procedure tract metastasis by 942–3.
malignant mesothelioma. Br J Cancer 2004; 91: 9–10.
This page intentionally left blank
Index
carbapenems 158, 353 cell-block preparation 293 empyema 240–1, 351, 351
carcinoembryonic antigen (CEA) 205, cephalosporins esophageal rupture 467
302, 304, 304 hospital-acquired pleural infection 353 hantavirus pulmonary syndrome 385
carcinoma immunostaining, malignant protein binding 158 heart failure patients 315
mesothelioma vs. 289 spontaneous bacterial empyema 461 hemothorax 244, 406
carcinoma, metastatic 293 cervical pleura hepatic hydrothorax 456
cardiac chamber collapse 55 blood supply 17 iatrogenic pneumothorax 537–8
cardiac cooling, post-CABG pleural gross anatomy 14 immune reconstitution inflammatory
effusions 448 chemical pleurodesis syndrome 413
cardiac disease-related effusions 315–21 chylothorax 392 interlobar fissure fluid collection 236,
clinical presentation 316–17, 496 cirrhotic hydrothorax 607 237
coexistent conditions 319 chemokine release, mesothelial cells 32 Kaposi sarcoma 412
delayed thoracentesis 318–19 chemotherapy lamellar pleural effusions 237
epidemiology 315 interstitial pneumonitis induction lateral 233
exudative 319 435–6 liposarcoma 246
future directions 320 malignant mesothelioma 191–2, 195, localized fibrous tumors 245, 246
incidence 315 511 loculated effusions 236–7
lung interstitium 315–16 malignant pleural effusions 327 lung abscess 241
misclassification 318–19 pleural effusion induction 435–6 lymphoma 246–7
noncardiac causes 319 solitary fibrous tumor of the pleura malignant mesothelioma 250, 509
pathogenesis 315–16 486 malignant pleural effusions 242, 261,
pleural fluid characteristics 318–19 see also individual drugs 325, 325
pleural peritoneal shunting 320 chest drainage Meigs’ syndrome 493
pulmonary edema in 316 future directions 565 meniscus sign 236
radiology 317, 317–18 medical thoracoscopy 586 mesothelioma 249, 249
transudate conversion to exudate pleural infection 354 ‘middle lobe step’ 236
318–19 techniques 551–67 nitrofurantoin-induced effusions 432
treatment 319–20 tension pneumothorax 541 normal anatomy 234
cardiac function see also individual techniques pediatric pleural effusion 546
pleural effusion effects 55, 179 chest drainage devices 563–5 pleural fluid 236–8
pneumothorax effects 57, 179 air leak indications 564 pleural infection 351, 351
cardiac output bottle systems 563–5, 564 pleural lipoma 246, 247
pleural effusion 55 commercially available 564, 564–5 pleural plaques 244, 245
pneumothorax 57 contraindications 563 pneumothorax see pneumothorax,
tension pneumothorax 541 definitions 563 spontaneous
cardiac surgery, post-operative effusions dry suction units 564 posterior-anterior (PA) 233
447–51 historical perspectives 2, 7–8 pseudochylothorax 243–4
cardiopulmonary resuscitation (CPR), indications 563 pulmonary embolism 398
tension pneumothorax 540 monitoring 565 pulmonary veno-occlusive disease 405
cardiovascular agents, effusion see also chest tubes rheumatoid arthritis 495
induction 432 chest expansion, spontaneous septic pulmonary emboli 404
carrageenan pleurisy model 173–4 pneumothorax 520 solitary fibrous tumor of the pleura
caspofungin 379 chest pain 484
Castleman’s disease, multicentric 412 malignant pleural effusions 324 subpulmonary effusion 237–8
catamenial hemothorax 480 thoracentesis-induced 556 supine 233
catamenial pneumothorax 515 chest percussion 4 tension pneumothorax 253–4, 254,
gender differences 516 chest radiography (CXR) 233 541
management 525 acute chest syndrome 404 tuberculosis, HIV patients 411
pathogenesis 519 All-trans-retinoic acid-induced Wegener’s granulomatosis 426
surgery 602, 602 pleural disease 435 chest syndrome, acute 404–5
catheters, indwelling see indwelling amebiasis 384 chest trauma, hemothorax 405
pleural catheters apical pleural thickening 248–9 chest tubes 559–63
CC49 monoclonal antibody 303 asbestos-related pleural diseases 244, contraindications 559
CD15 304 502–3 definition 559
CD34 483, 485 benign asbestos pleural effusion 503 dressing 561, 562
CD44 65–6 bronchopleural fistula 254 hemothorax 406
CD45 306, 306 chylothorax 243–4 hepatic hydrothorax 458
CD99 485 circumscribed pleural plaques 502 iatrogenic pneumothorax 540
CD141 (thrombomodulin) 305, 305 diaphragmatic inversion 236, 236 indications 559
Cdkn2b tumor suppressor gene 130 diffuse pleural thickening 247, 500 insertion
cefotaxime 461 benign, non-asbestos related 248, animal model 170, 185
ceftriaxone 461 248 complications 561–2
cefuroxime 160 malignant 249, 249 techniques 559–61
632 Index
pleural plaques 245 cryptococcosis 380, 380, 411 desmoplastic small round cell tumor
pneumothorax 254, 254 CTNNB1 gene 127 (DSRCT) 290, 296
pulmonary embolism 399 cultures, pleural infection 348 diabetes, pleural infection 342
pulmonary veno-occlusive disease cyclines 572–3 diaphragm
405 cyclophosphamide 436 inflammation 465
rolled atelectasis 501 cystic fibrosis, spontaneous innervation 18
solitary fibrous tumor of the pleura pneumothorax 518, 602 pleural effusion effects 54–5
484 cystic/multicystic mesothelioma 287 pneumothorax effects 57
spontaneous pneumothorax 517, 518 cytogenetic analysis 293–4, 310 ultrasound 234
transudative vs. exudative effusion cytokeratins 302–3 venous drainage 17
239, 240 cytokine(s) diaphragm peritonectomy, post-
tuberculous empyema 242 future directions 624 operative effusions 446
Wegener’s granulomatosis 426 gene therapy 616 diaphragm resection, post-operative
computed tomography (CT)-guided immunomodulatory 81–2 effusions 446
pleural biopsy 266 pleural effusion induction 433–4 diaphragmatic defects
malignant pleural effusions 242, 243, pleural fluid formation 79–81 acquired 22
326–7 cytokine growth factors hepatic hydrothorax 455, 456
congenital abnormalities, fetal pleural malignant mesothelioma 194 pleuro-peritoneal communication 22
effusions 476, 545 pleural fibrosis 71–9 diaphragmatic fenestrations, catamenial
congenital heart disease 547 cytokine receptor agonists 625 hemothorax 480
congestive heart failure (CHF) cytology 293–313 diaphragmatic paralysis 279
effusions adjunctive studies 302–10 diaphragmatic pleura
clinical presentation 316–17 benign conditions 293 blood supply 17
epidemiology 315 ‘classical’ forms 294 gross anatomy 14
incidence 315 cytopathological ‘partitions’ 296–302 lymphatic circulation 19
noncardiac causes 319 malignant large-polygonal-cell referred pain 202
pathogenesis 316 tumors 298–300 diaphragmatic rupture 468
radiographic findings 315, 316 malignant pleomorphic tumors 300 diet, chylothorax 391–2
transudate 318 malignant pleural effusions 326, 326, differential gel electrophoresis (DIGE)
treatment 319–20 327 140, 141, 142
hemodynamic findings 316 malignant spindle-cell tumors 300, diffuse pleural thickening (DPT) 500
history 202 301 clinical features 502
connective tissue diseases (CTD), pleural pleural fluid analysis 204, 222 definition 500
effusions 421–30 preparation 293–4 epidemiology 501
future directions 426 rheumatoid arthritis 423–4 etiology/pathogenesis 501
see also individual disorders superior vena cava syndrome 403 management 504
contrast enhanced ultrasound (CEUS) Cytomegalovirus (CMV) infection pulmonary function tests 503
233 385–6, 386 digital clubbing 484
cope needle 557, 558 cytosine deaminase 617 dihydroergocristine 434
coronary artery bypass graft (CABG) cytotoxic agents, pleurodesis 573 diphtheria toxin 192
surgery cytotoxic T cells 63 dipyridamole 78
postoperative effusions see post- distribution, drugs 152–3
CABG pleural effusions D-dimer test, pulmonary embolism distribution equilibrium 152
pseudochylothorax 393 399–400 diuretic therapy
corticosteroids ‘D’ shaped subpleural opacity 351, 351 cardiac disease-induced effusions 319
rheumatoid arthritis 424 dantrolene sodium 436 hepatic hydrothorax 457–8
systemic lupus erythematosus 422 dapsone 440 DNA microarray, malignant
tuberculosis 374 debridement, empyema 603, 603 mesothelioma 510–11
Corynebacterium parvum 573 decortication DNAse, intrapleural 357–8
costal pleura 14 complications 608 docetaxel 436
blood supply 17 diffuse pleural thickening 504 Doege–Potter syndrome (refractory
innervation 18 empyema 604–5 hypoglycemia) 484
lymphatic circulation 19 rheumatoid arthritis 424 dogs, pleural fluid studies 39
costophrenic recesses 14 with visceral pleurectomy 606 Donnan equilibrium 45
cough DeCyder Differential In-gel Analysis double lung transplant 451
malignant pleural effusions 324 software 140 doxycyline 572–3
thoracentesis-induced 556 ‘deep sulcus sign’ 541 drainage see chest drainage
Coxiella burnetti 383 deep vein thrombosis 401–2 drainage bag 563
crevices (fenestrae) 21, 21–2 dengue fever 386, 469 drug-induced effusions 431–43
cribriform lamina (membrana Denver Pleurx catheter 562, 562–3 future directions 440
cribriformis) 20–1, 21 Denver shunt 320, 563, 563 symptoms 431
crocidolite 114, 177, 499, 500–1 dermatomyositis 425 see also individual drugs
Crohn’s disease 468 desmin 303 drug-induced lupus (DIL) pleuritis 438–9
634 Index
factor XIIIa 485 bronchopleural 254, 255, 587 gastric lymphoma 468
farnesyltransferase inhibitors 77 gallbladder 468 gastrointestinal disease-associated,
fenestrae (crevices) 21–2, 22 pancreatic 469 pleural effusions 465–72
fetal pleural effusions 475–6, 545–6 pancreatopleural 466 pleural fluid characteristics 469
chylous effusion 545 flow cytometry (FCM) therapeutic approaches 469–70
clinical presentation 475 malignant pleural effusions 327 see also individual diseases
complications 476 pleural effusions 310 gefitinib 76
epidemiology 475 pleural fluid analysis 222 gemcitabine 85
etiology 475 preparation 293–4 gene therapy 613–19
future development directions 476 fluconazole 381 animal models 174
incidence 475, 545 18-fluorodeoxyglucose (18FDG) 234 future directions 624
management 476 fluoroscopic guidance 259–60 malignant mesothelioma 194–5
newborn period 545–6 flutter valves 524 malignant pleural disease 614
pathogenesis 475 focal adhesion kinase 130 anti-angiogenesis 615
primary 545 Fontan procedure, post-operative apoptosis induction 615
secondary 545–6 effusions 547 clinical trials 616, 616–17
spontaneous regression 476 forced expiratory volume in one second immuno-gene therapy 615–16
treatment 476 (FEV1), pleural effusion 51 preclinical studies 614–16, 615
fever, pleurodesis 118–19, 570 exercise tolerance 54 replicating, tumor-selective vectors
FGK45, malignant mesothelioma 195 post-thoracentesis 52 616
fibrin forced vital capacity (FVC) suicide gene therapy 615
alveolar deposition, ARDS 101–2 pleural effusion 51 metastatic pleural disease 614
disordered turnover 101–2 exercise tolerance 54 non-malignant diseases 614
inflammatory response 102 inter-individual variability 51, 52 systemic disease treatment 613–14
fibrin degradation products 138 post-thoracentesis 229 vectors 613, 614
fibrin glue Fourier transform ion cyclotron (FT- gene transfer, soluble Flt-1 81
pleurodesis 573 ICR-MS; FTMS) 145 genetics, future directions 624
spontaneous pneumothorax 601 FR16753 82 gliclazide 437
fibrinogen 103 fractalkine 62 glucose, pleural fluid analysis 219–20
fibrinolysins fraction of inspired O2 air that is O2 low 219, 220
malignant mesothelioma 107–8 (FIO2) 52 malignant pleural effusions 218,
parapneumonic effusion, children free (unbound) drug 156 219–20, 326, 326
107 fumagillin 192 pleural infection 344
pleural fibrosis 106–8 functional residual capacity (FRC) pleurodesis 575
preclinical models 107 animals 51 rheumatoid arthritis 423, 424
thoracoscopy vs. 107 pleural pressure 227 tuberculous effusions 220, 220
fibrinolysis fungal infection 379–82 Gram-negative organisms 60, 347, 347
empyema 106 empyema 348, 379 Gram-positive organisms 346, 347
VATS vs. 603–4 pseudochylothorax 394 granulocyte-colony stimulating growth
pathways 102, 102 transplant recipients 414 factor (G-CSF) 433–4
pleural infection 61 see also individual infections granulomas 286, 370
pleural injury and 103–4 furosemide 319 granulomatous pleuritis, histology 286
fibrinolytics, intrapleural see intrapleural fusiform cells 2989 Grocco’s triangle 5
fibrinolytics Fusobacterium necrophorum 404 gross cystic disease fluid protein-15
fibrinopeptide A 103 future directions 621–7 (GCDFP-15) 304–5
fibrinous pleuritis 285–6 diagnosis 624–5 group-A streptococci 60
fibroblasts predictions 623–6 GW3333 82
growth, asbestos inhalation 114 clinical practice 624–6 gynecological effusions 473–82
pleurodesis 570 incidence 624
fibronectin 61 scientific 624 Haemophilus influenzae 548
fibrous pleuritis, histology 285–6 recent advances 622–3 ‘hairy plaques’ 500
malignant mesothelioma vs. 288–9, as specialty 623 half-life (t1/2), drugs 154–5, 155
289 training/education 623 Hantavirus infection 385, 469
fibrous silicates 114 hantavirus pulmonary syndrome 385
fibrous tumors gallbladder disease-associated pleural HBME-1 immunostaining 288, 305
benign 483–9 effusions 468 heart, distending pressure 49
localized, radiology 245–6, 246 gallbladder empyema 468 heart failure
First Multicentre Intrapleural Sepsis gallbladder fistulae 468 effusions 317, 317, 317, 496
Trial (MIST1) 346–7, 355, 357, gd T cells 63 pseudotumor 317, 317
357, 604 Garg thoracentesis needle 553, 556 heart-lung transplant, post-operative
fissures, HRCT 235, 235 gastric carcinoma 299, 468 effusions 451
fistula(e) gastric disease-associated pleural heart transplant recipients, pleural
biliopleural 469 effusions 468 effusion 413, 415, 450–1
636 Index
heat shock protein-65 192, 615–16 intercostal stripe 235, 235 incidence 537
Heimlich valve 539, 565, 565, 578 normal anatomy 235, 235 management 539, 539–40, 625
HELLP syndrome 477 pleural plaques 245, 245 observation 539
hematological malignancies 414, 415–16 hippocratic succussion 2 persistent air leak 540
hematoma 240 histology 285–91 risk factors 538, 538–9
hematopoietic stem cell transplant future directions 290 surgical management 540
recipients 415 non-neoplastic pleural lesions 285–6 symptoms 537
hemithorax 13 pleural neoplastic lesions 286–9 thoracentesis-induced 556
pneumothorax 56, 56 rare primary malignancies 289–90 transthoracic needle biopsy 537–8,
hemopneumothorax, spontaneous 405, histoplasmosis 381, 381 538
602 historical perspectives, pleural disease treatment 539–40
hemorrhage 1–9 ICU hemodynamic transducers 229
hepatic 469 HIV infection/AIDS IFN-g/IFN-g receptor (IFN-gR) complex
medical thoracoscopy 587 empyema 361 polymorphism 88
hemothorax 405–6 herpes simplex virus infection 386 imatinib mesylate
catamenial 480 malignancies 411–12 platelet-derived growth factor
complications 406 pleural disease 410–13 inhibition 78
diagnosis 405–6, 406 pleural effusions 410, 410–12, 494 pleural effusion induction 436
drug-induced 437 pleural inflammatory response imidapril 433
empyema 406 inhibition 66 immune deficiencies
etiology 405, 405–6 risk of, thoracentesis 556 primary 409, 416
lung transplant recipients 415 spontaneous pneumothorax 518–19 complications 416, 416
pleural fluid hematocrit 209–10 tuberculous effusions 367–8, 367–9, secondary 410
pleural thickening 406 411 immune reconstitution inflammatory
pleurectomy complication 607 clinical features 369 syndrome (IRIS) 413
pulmonary embolism 402 treatment 374, 411 immuno-gene therapy 615–16
radiology 244, 244, 406 tuberculous pleuritis 370, 371 immunocompromised host, effusions
residual clots 406, 606–7 Hodgkin’s disease 247, 247, 547 409–19
retained 606–7 horizontal fissure, imaging 234, 235 future directions 417
surgery, future directions 608 hormonal treatment, catamenial immunoglobulin(s) 61, 434
thoracentesis-induced 556 pneumothorax 602 immunoglobulin E (IgE),
treatment 406 Horner’s syndrome 607 paragonimiasis 384
heparanase 77 HRCT see high-resolution computed immunoglobulin G (IgG),
heparin 108, 437, 576 tomography (HRCT) paragonimiasis 384
hepatic abscess 468–9, 469 Human herpes virus (HHV)-8 412 immunohistochemistry, solitary fibrous
hepatic hemorrhage 469 Human immunodeficiency virus (HIV) tumor of the pleura 485
hepatic hydrothorax 455–63, 496 see HIV infection/AIDS immunological studies, pleural fluid
chest radiography 456 Human papilloma virus 412 analysis 221–2
clinical presentation 456 hyaluronan (hyaluronic acid) 17, 32, immunology 71–100
complications 460–1 65–6 immunomodulating agents 433–4
diagnosis 456–7 hydralazine 438 immunosuppressants, systemic lupus
differential diagnosis 457 hydropneumothorax 277 erythematosus 422
etiology 455–6 hydrostatic theory 227 immunotherapy, malignant
future directions 461 hydrothorax mesothelioma 511
incidence 455 hepatic see hepatic hydrothorax experimental models 194–5
pleural fluid analysis 456–7 pressure measurement 228, 228 immunotoxins 192
treatment 457, 457–60 refractory 458–9 indwelling pleural catheters
hepatitis 386, 469, 556 hydroxyethylstarch 474 ambulatory
hepatitis B virus 469 hypertrophic pulmonary insertion equipment 578
hepatocellular carcinomas (HCCs) 298, osteoarthropathy (HPO, pleurodesis alternative 577, 578
299–300 Pierre–Marie–Bamberg syndrome) spontaneous pneumothorax 600
hepatocyte growth factor (HGF) 74–5 484 malignant pleural effusions 334
mesothelioma 75, 131 hyphae, aspergillosis 379, 380 future directions 625
hepatocyte growth factor (HGF)/ hypoproteinemia 468, 474 long-term 261–3, 262
Met receptor signaling 131 hyposplenism 416 pleural infection 361
hepatocyte growth factor receptor 74 hypoxia-inducible transcription factor spontaneous pneumothorax drainage
HepG2 cells 104 (HIF)-1a 65 522–3, 523
herpes simplex virus infection 386 infection see pleural infection
herpes simplex virus thymidine kinase iatrogenic pneumothorax (IP) 537–40 infectious mononucleosis 386
192, 615 clinical presentation 537–8 inflammation see pleural inflammation
high-resolution computed tomography diagnosis 537–8 influenza hemagglutinin (HA) 193
(HRCT) 233–4 epidemiology 537 influenza pandemic (1919) 342
asbestos-related pleural diseases 503 etiology 537 inhibitors of apoptosis (IAPs) 32
Index 637
octreotide 392, 546 parapneumonic effusions 341–66 pleural fluid pharmacokinetics 162,
OK432 573 antibiotics 353–4 163
oligosaccharides 66 categorization 263 peptide mass fingerprint (PMF) 143
oncocytic tumors, malignant 298 children 548 percutaneous needle pleural biopsy 326
cytology 298, 298–9 classification 344 perfusion lung scans, pulmonary
diffuse cytoplasmic eosinophilia 299 complicated 218–19, 345, 345 embolism 400–1
fibrillary cytoplasmic eosinophilia diagnosis 496 pergolide 434
298–9 future directions 625 pericardial disease-associated effusions
globular cytoplasmic eosinophilia HIV patients 411 clinical presentation 317
299 medical thoracoscopy 592–3 epidemiology 315
immunohistological diagnostic pleural fluid pH 218–19 incidence 315
algorithm 309 presentation 496 pathogenesis 316
oncogenes, malignant mesothelioma rheumatoid arthritis vs. 349–50 pleural fluid characteristics 319
191 simple 345, 345 radiology 318
one-compartment model, surgery 603 pericardial drainage 319
pharmacokinetics 153, 153 transplant recipients 414 pericardial sclerosis 319
open window thoracostomy 359 tube thoracostomy 263–5 pericardial window 319
orthotopic heart transplant (OHT)- ultrasound 241 pericardiectomy 319
associated, pleural effusions 450–1 uncomplicated 218 peritoneovenous shunt 458
osteopontin 510 parasitic infections 383–6 PET see positron emission tomography
ovalbumin (OVA) 193 paravertebral line 236 (PET)
ovarian hyperstimulation syndrome parietal pleura 44 phagocytosis 33, 62
(OHSS) 473–5 blood supply 17, 44, 44–5 pharmacokinetics 151–68
clinical presentation 474 gross anatomy 14 clearance 155
management/treatment 474–5 innervation 18 compartment models 153, 153–4
lymphatic circulation 19 definition 151, 151
p14ARF, malignant mesothelioma 129, pleural fluid production 18 diffusion 155–7
131 regional differences 22–3 distribution 152–3
P16/CDKN2A tumor suppressor gene 34 stomata 46 elimination see elimination, drug
p16INK4a, malignant mesothelioma thickness 23 future research directions 624
128–9, 131 ultrasound 273 half-life 154–5, 155
p21-activated kinase (Pak) 130 venous drainage 17 malignancy 159
p21-activated kinase (Pak) inhibitors parietal pleurectomy plasma drug concentration and time
132 failed pleurodesis 577 152, 152
p53 gene pleurodesis 571 pleural fluid specific 155–67
gene therapy 615 spontaneous pneumothorax 600, principles 151–5
malignant mesothelioma 191 600, 601 protein binding 155–7, 156, 158
palliation/palliative care partial pressure of arterial CO2 (PaCO2) reversibility 156
failed pleurodesis 578 pleural effusion effects 52, 53 see also individual drugs
malignant mesothelioma 512 pneumothorax effects 56–7 phosphoglycerate kinase (PGK) 105
malignant pleural effusions 83 partial pressure of arterial O2 (PaO2) photodynamic therapy 511
palliative procedure, malignant pleural pleural effusion effects 52, 53 PI3 kinase pathway 74–5
effusions 260 pneumothorax effects 56 PI3K inhibitors 131
pancreatic abscess 466 tension pneumothorax 541 Pierre–Marie–Bamberg syndrome
pancreatic disease-associated pleural PAS stain with diastase digestion (D- (hypertrophic pulmonary
effusions 465–6 PAS) 302 osteoarthropathy, HPO) 484
pancreatic duct disruption, pleural Pasteurella multocida 175 PIOPED study 401
effusions 466 pathogen-associated molecular patterns placental alkaline phosphatase (PLAP)
pancreatic fistula 469 (PAMPs) 62 304
pancreatic necrosis 466 pathology plasma cells 217
pancreatic pseudocysts 466 cytology see cytology plasma proteins 156
pancreatitis, acute histology see histology plasmin 102
pleural effusions 465–6, 469 patient, approaches to 201–7 plasminogen activator inhibitor 1
pleural fluid amylase 221 pattern recognition receptors (PPRs) 62 (PAI-1)
pancreatitis, chronic pediatric pleural diseases 545–50 overexpression 104
pleural effusions 466, 494 future therapies 549 post-transcriptional regulation 104,
pleural fluid amylase 221 malignancy 547 104–5
pancreatopleural fistula 466 pleural effusion diagnosis 546 uPA inhibition 103
papillary mesotheliomas 286–7, 287 see also individual diseases plasminogen activator inhibitor 2
paracentesis 456, 458 D-penicillamine 440 (PAI-2) 103
paragonimiasis 384 penicillin(s) plasminogen activator inhibitors (PAI)
paramalignant pleural effusions 222, historical perspectives 342 33, 102
323, 329 hospital-acquired pleural infection 353 plasminogen activators (PA) 102
642 Index
platelet-derived growth factor (PDGF) diagnostic tests 203–4, 206, 546 viral hepatitis 386
77–8 diaphragmatic effects 54–5 viral infections 385–6
asbestos-induced pleural fibroblast differential diagnosis 202 pleural elastance, pleural manometry
proliferation 77 drug-induced see drug-induced and 229, 229
inhibition, therapeutic approaches 78 effusions pleural fibrosis 101–12
malignant mesothelioma 32, 78 endocarditis 404 animal models 171–2
mesothelial cell proliferation 31, 31 exercise tolerance 54, 54 anticoagulants 108
platelet-derived growth factor receptors exudative see exudative pleural asbestos-related pleural diseases 501
(PDGFRs) 77, 78 effusion basic fibroblast growth factor 77
pleura fetal see fetal pleural effusions cytokine growth factors 71–9
anatomy 13–25 flow cytometry 310 disordered fibrin turnover link 102,
historical perspectives 1–2 fluid overload induced 178 102
inter-species variation 13 fungal infections 379–82 ergot derivative-induced 434
blood supply 17–18 history 201, 201–2 fibrinolysins 106–8
defense mechanisms 61–3 undiagnosed effusion 491 interleukin-1 82
acquired immune responses 63 hypoxemia 53 localized, histology 286
leukocyte recruitment 63 idiopathic, medical thoracoscopy 592 non-asbestos silicates 116
phagocytic cell invasion 63 infection 341–66 pathogenesis 101–2
embryology 13–14 inflammatory 275, 469 silica 116
extent 14 interleukin-1 82 transforming growth factor beta
gross anatomy 14 interleukin 2 82–3 71–2, 72
inflammatory cells 63–4 interleukin-6 84–5 tumor necrosis factor-a 82
innervation 18 lamellar, imaging 237 pleural fluid
layers 15, 15 Lemierre’s syndrome 404 absorption 46–7
lymphatics 19 malignant see malignant pleural animal models 178–9
microscopic anatomy 15 effusions analysis see pleural fluid analysis
mineral dust fibers/particles dynamics malignant mesothelioma 329–30, bulk flow 46
114 330, 508 cellular content 18, 39–42
physiology massive 203 CT 239–40
animal models 178–9 MRI 240 formation
fluid exchange, normal paramalignant 222, 323, 329 animal models 178
physiological states 43–8 pediatric 547–9 cytokines 79–81
historical perspectives 5 physical examination 202–3 future research directions 624
solute exchange, normal physiological changes 49–55 function 39
physiological states 43–8 pleural pressure effects 49–51, 50 lymphatic clearance 46
reflections 14 position effects, oxygenation status microbiology-positive 345
regional differences 22–3 53, 53 normal physiology 39–42
repair 23 pulmonary embolism 397–402 animal studies 39–40, 40
subclinical alterations 23 pulmonary function and 51, 51–2 human studies 40–1
transport across 18–19 animal models 179 pleural infection 345
pleural abrasion 571, 601 pulmonary veno-occlusive disease 405 production 18, 43–5
pleural adhesions salivary isoamylase-rich 221 pressure gradient 45, 45
chest tubes insertion complications septic pulmonary emboli 404 rates 45
562 superior vena cava syndrome 403 protein content 18, 43–4, 138
histopathogenesis 118 thoracic aortic dissection 404 radiology 236–40
historical perspectives 4 time to resolution 204–5, 205 reabsorption 18
nerves in 118 transforming growth factor beta 71 smokers 41
pleural infection 344 transplant recipients 413, 413–15 solute composition 40
pleural cavity 13 transudative see transudative pleural systemic origins 44–5
anatomy 14–15 effusions thickness 43
dome/cupola 14 tuberculosis see tuberculous effusions ultrasound 238, 238–9
embryology 13–14 tumor necrosis factor-a 82 volume 18, 27
respiratory movement 14 ultrasound see ultrasound animals 39–40
pleural effusions undiagnosed 205–6, 491–8 humans 41
acute chest syndrome 404–5 follow-up 497 increases, infection 61
approaches to patient 201–7 malignancy 496 pleural effusion 202–3
atypical infections 379–88 persistently negative results 497 white blood cell counts 39–40, 41
bacterial infections 382–3 tests 497 pleural fluid analysis 203–4, 209–26
biochemical tumor markers 302 transudate vs. exudate 491, 491 abdominal surgery, post-operative
blood gases 52–3 treatable disease exclusion 492, effusions 447
cardiac effects 55, 179 492–6 adenosine deaminase 222–3
coagulation proteins 103 vascular endothelial growth factor amylase 220–1
CT 239, 239–40 79–80 appearance 209–10, 210
Index 643
bedside examination 209–10 pleural infection pleural injury, fibrinolysis and 103–4
character 210, 210 anaerobic 347, 347 pleural manometry 227–32
cholesterol 221 animal models 175 air-contrast CT scan 230
color 209–10, 210 antibiotics 353–4 clinical setting 229
complement levels 222 atypical 379–88 physiology 228, 228
cytology 204, 222 bacterial see bacterial infection pleural elastance and 229, 229
definitive diagnoses 203, 204 bacteriology 343, 345–8, 347 pleurodesis 230
exudative vs. transudative 210–15, causes 342–3, 343 therapeutic thoracentesis and 230
212, 213 clinical assessment 348–9 pleural neoplasms/tumors
abbreviated Light’s criteria rule clinical outcome predictors 350–1 benign mesenchymal neoplasms 287
213 clinico-pathological stage correlations diagnosis 293–6
large molecular weight constituents 345, 345 diagnostic algorithms 307–9
211–12 community-acquired disease 347, clear-cell malignant neoplasms 309
Light’s criteria 212–15, 214 353 epithelial malignancies, unknown
likelihood ratio 214, 215 comorbid disease 341 origin 308
misclassifications 213–14 diagnosis 348–9 large-cell malignancies 307
‘pretest’ clinical suspicion 214 diagnostic algorithm 346 oncocytic neoplasms 309
serum to pleural fluid albumin differential diagnosis 349–50 pleomorphic malignancies 308
gradient 213 epidemiology 342–3 predictive value 309
test strategies 212, 213 exudative phase 343 small-cell malignancies 307
flow cytometry 222 fibropurulent phase 343–4 immunohistochemistry
future directions 206, 223 fungal see fungal infection algorithmic 302
glucose see glucose, pleural fluid future directions 361 markers 302–3
analysis granuloma formation 286 ultrasound 278, 278
heart transplant recipients 450 historical perspective 342 see also individual types
hepatic hydrothorax 456–7 hospital-acquired 342–3 pleural nodules 238, 238
immunological studies 221–2 antibiotics 353, 353–4 pleural peels infectious 344
lactate dehydrogenase see lactate bacteriology 343, 347 pleural plaques
dehydrogenase (LDH) incidence 342 asbestos exposure 286
lung transplant recipients 451 inflammation 59–70 circumscribed see circumscribed
lymphocyte count 204 loculation 344 pleural plaques
malignant pleural effusions 325–6 malignancy vs. 350 histology 286
mesothelial cells 217 mixed infection 347 radiology 244–5, 245
nucleated cells 215–17 mortality 341, 341, 347, 348 pleural pressure
paragonimiasis 384 natural healing/organizing stage 344 changes, peri-thoracentesis 229
pH see pleural fluid pH parasitic 383–6 measurement 227–8
pleural effusion 203–4 pathophysiology 343–4 normal physiology 227
pleural infection 348–9, 349 pediatric pleural effusions 547–9 pleural effusion effects 49–51, 50
post-CABG pleural effusions 449 pleural fluid analysis 348–9, 349 pneumothorax effects 55
primary effusion lymphoma 412 pleural scarring 344 pleural rub, asbestos-related pleural
protein 204, 215 pulmonary embolism 402 diseases 502
pulmonary embolism 399 radiology 351–3 pleural scarring, pleural infection 344
rheumatoid factor 222 rheumatoid arthritis 424 pleural sclerosis 66, 458–9
spontaneous bacterial empyema surgery 358–9, 360 pleural space
460–1 treatment, non-surgical patients acquired immune responses 63
superior vena cava syndrome 403 359–61 contents 18
transudate 203–4 viral 385–6, 548 elastance 50
trapped lung 205 see also specific infections pleurodesis success and 50–1
triglycerides 221 pleural inflammation 59–70 microbial invasion 61
tuberculosis pleuritis 369 animal models 172–4 pleural symphysis
yellow nail syndrome 205–6 histologic assessment 172–3 adhesions 118
pleural fluid eosinophilia (PFE) 216, asbestos-induced 115 talc pleurodesis 117–18, 334
216, 422 drug penetration 159 pleural thickening
pleural fluid pH 204, 217–19 epidermal growth factor 76 ankylosing spondylitis 425
esophageal rupture 467 induction techniques 173–4, 174 definition 275
low 217, 217 infection 59–70 diffuse, apical pleural cap 248–9
malignant pleural effusions 218, 219, inhibition 66–7 diffuse benign
326, 326, 327 initiation 64–5 asbestos related 247–8
normal 217–18 interferons 88 non-asbestos related 248, 248, 249
parapneumonic effusions 218–19 mesothelial cell proliferation 23 diffuse malignant
pathogenesis 217–18 perpetuation 65–6 benign vs. 249
pleural infection 344, 348–9, 349 repair 66 radiology 249–50
pleurodesis indications 219, 575 resolution 66 ergot derivative-induced 434
644 Index
Q fever 383 regulatory T cells (Treg), malignant SC5b-9 monoclonal antibody 372
Q (quadrupole) mass filter 145 mesothelioma 193–4 scarring, pleural infection 344
quinacrine 573 renal cell carcinoma, metastatic 299 scatter factor see hepatocyte growth
quinolones 158 renal failure 178 factor (HGF)
repeat pleurodesis 577 schwannomin 129–30
rabbits malignant pleural effusions 334 scintigraphy, hepatic hydrothorax 457
mesothelial cell harvesting 186 replication-deficient adenoviral HSVtk sclerosant instillation, tube
pleural fluid studies 39 vectors 615 thoracostomy 261
tuberculous pleurisy/pleuritis models research, neglect in 622 sclerosing agents 333
176 residual air, traumatic pneumothorax screening tools, future directions 624
radiology 536 Seldinger technique 266, 559–60, 560
air 252–4 retinoblastoma gene 191 L-selectin 63
diagnostic 233–58 retinoids 435 seminoma/germinoma, metastatic 300
techniques 233–4 retroviral gene therapy vectors 614 sepiolite exposure 114
diffuse pleural disease 247–52 reverse passive Arthus reaction 174 septated pleural effusions, ultrasound
focal pleural disease 244–7 revised Geneva score 399, 400 275
future directions 255 rhabdoid 299 septic pulmonary emboli 404
hemothorax 244 rhabdomyosarcoma 297, 297 serum (blood) peptidome 136
interventional see interventional rheumatoid arthritis (RA) sheep pleurodesis model 185–6
radiology cytology 423–4 sialomucin complex (SMC) 61
malignant mesothelioma 250–1 parapneumonic effusion vs. 349–50 sickle cell anemia 404–5
normal anatomy 234–6 pleural effusions 423–4, 495 ‘the sign of damp earth’ 2
pleural effusion 203 pleural fluid analysis 210, 423, 495, silica
pleural fluid 236–40 495 mesothelial damage mechanisms 116
pleural infection 351–3 pleural infection 424 pleural fibrosis 116
tuberculous effusions 369 SLE vs. 423 pleural involvement 113–14
see also individual techniques rheumatoid arthritis cell (RA cells; silicosis 113
radiotherapy ragocytes) 423–4 silver nitrate pleurodesis 573, 625
malignant mesothelioma 511, 587 rheumatoid disease, pleural fluid Simian virus 40 (SV40)
solitary fibrous tumor of the pleura analysis 221–2 biology 188
486 rheumatoid factor 222 malignant mesothelioma 75, 188–9,
Ral guanine nucleotide dissociation rib(s) 508
stimulator (RalGDS) 130 bony metastases 279 models 190–1
rapamycin 131 ultrasound 273, 274, 279 sequence detection 188
Ras signaling pathway 74–5 rib fracture 279 T antigen 191
Raynold’s phenomenon 495 rib resection simple aspiration, spontaneous
re-expansion pulmonary edema 230, empyema 605 pneumothorax 522, 522, 523, 523
332, 458, 524 pleural infection 359 simvastatin 439
reactive nitrogen species (RNS) 33, 62 ricin 192 Sin Nombre virus 385
reactive oxygen species (ROS) Rickettsiae infection 383 single-chain uPA (scuPA, prouPA) 103,
asbestos-related mesothelial damage rifampin 373–4 107
33, 115, 125–6 Rocky Mountain spotted fever 383 Sjögren’s syndrome 425
malignant mesothelioma 508 rolled atelectasis (RA) 500 sleeve lobectomy 485
mesothelial cell immune response 62 ‘comet tail’ appearance 501 small-cell mesotheliomas 298
receptor for hyaluronic acid mediated epidemiology 501 small-cell neuroendocrine carcinomas
motility (RHAMM) 65–6 etiology/pathogenesis 501 (SCNCs, oat-cell carcinoma) 297,
recombinant human growth hormone management 504 297, 298
(rhGH) 79 ‘rolled bed sheet technique,’ small intestinal disease-associated
recombinant human interleukin 2 (rhIL- thoracentesis 555, 555 pleural effusions 468
2) 83, 194–5, 433 small molecule inhibitor of ALK5kinase
recombinant human interleukin 11 S100 protein 305 (TBR-1) 74
(rhIL-11) 433 salivary isoamylase-rich pleural effusion small round-cell tumors 296–8
recombinant interferon alpha 194, 433, 221 smoking, spontaneous pneumothorax
511 saphenous vein (SV) grafts 448 517, 518
referred pain, diaphragm 18 sarcoidosis sodium channels 46
refractory ceramic fibers (RCF) 116, granulomas 286 sodium restriction, hepatic hydrothorax
189–90 spontaneous pneumothorax 518, 519 457–8
refractory hydrothorax (RH) 458–9 sarcoma solitary fibrous tumor of the pleura
refractory hypoglycemia (Doege–Potter cytology 296 (SFTP) 483–9
syndrome) 484 immunohistology 289, 306 adjuvant therapy 486
Registro Informatizado de la malignant mesothelioma vs. 289 clinical features 484
Enfermedad TromboEmbólica metastatic 296 follow-up 487
(RIETE) registry 398 synovial 296–7, 306 future developments 487
Index 647
thoracentesis – contd sago-like nodules 590–1, 591 liver function impairment 459
thoracic aortic dissection 404 VATS vs. 584 model of end-stage liver disease 460
tuberculosis 369, 374 thoracotomy refractory hepatic hydrothorax 459,
ultrasound-guided 280 with decortication, pleural infection 459–60
thoracic aortic dissection 403–4 358 transplant recipients, pleural disease
thoracic duct 390–1 empyema 605 413–15
chylothorax 389, 390–1, 607 persistent air leak, traumatic causes 413, 413, 414
ligation 392, 607 pneumothorax 536 pneumothorax 415
traumatic rupture 390 VATS vs. 600, 601, 604–5 transpulmonary bands (TPB) 500, 504
valves 391 thrombin 102 transthoracic fine needle aspirations
variations 390 a-thrombin 103 (TTFNA) 281
thoracocautery (Jacobaeus operation) 6, thrombomodulin (CD141) 305, 305 rapid on-site evaluation (ROSE) 281
583, 584 thyroglobulin 304 transudative pleural effusions
thoracoplasty 8, 342 thyroid transcription factor-1 (TTF1) causes 211, 491
thoracoscopic sympathectomy 594 304, 305 exudative pleural effusion vs. 238,
thoracoscopy, medical 583–97 tissue factor (TF) 102, 102, 103 491, 491
benign asbestos-related pleural tissue factor pathway inhibitors (TFPI) nucleated cells 215
effusion 589 103 pleural fluid analysis 203–4, 218
biopsy 585 tissue inhibitor of metalloproteinase trapped lung 228–9
cardiorespiratory function (TIMP)-1 61 diagnosis 50, 494
monitoring 587 tissue inhibitor of metalloproteinases malignant pleural effusions 605–6
chest tube introduction 586 (TIMPs) 66 management 329
complication prevention 587 tissue plasminogen activator (tPA) 33, pleural fluid analysis 205
contraindications 587 265 pleurodesis 575
empyema 592, 592–3 tissue plasminogen activator inhibitor 1 thoracentesis 556
fibrinolysins vs. 107 343–4 traumatic pneumothorax
fluoroscopy in 585 tissue plasminogen activator inhibitor 2 air travel 536–7
future directions/predictions 594, 626 343–4 conservative management 534
historical perspectives 6–7, 7, 583–4 tizanidine 436–7 delayed 536
indications 583, 587–8 toll like receptors (TLRs) 63 follow-up issues 536
instrumentation 584, 585 total lung capacity (TLC), pleural effusion iatrogenic see iatrogenic
malignant mesothelioma 589 animals 51 pneumothorax (IP)
malignant pleural effusions 327, humans 51, 52 management algorithm 535
588–90 total parenteral nutrition, chylothorax non-iatrogenic 533–7
sensitivity 588–9, 589 392, 546 classification 533
staging 586, 589 Toxoplasma gondii 414 clinical presentation 533–4
metastatic pleural disease 589 TP53 129 epidemiology 533
mortality rates 587 transforming growth factor alpha incidence 533
parapneumonic effusions 592–3 (TGF-a) 188 management 534–6
pleural effusion 587, 588 transforming growth factor beta treatment 534–6
pleural fibrosis 107 (TGF-b) 71–4 non-occult 534
pleural infection 359 activation 74 non-penetrating 533
pleural malignancy 496 angiogenesis promotion 74 observation 535–6
pleural space requirements 584 antagonists 72 occult
pleurodesis and 585, 585 anti-inflammatory properties 73 clinical presentation 533–4
solitary fibrous tumor of the pleura carcinogenesis 73–4 non-ventilated patients 535
485–6 forms 74 risk predictors 534
spontaneous pneumothorax 525, genetics 74 treatment/management 534–6
587, 593–4, 601–2 immune modulation 73 penetrating 533
talc poudrage and 593–4 inflammation 33, 66 persistent air leaks 536
surgical vs. 583 parapneumonic effusions 625 residual air 536
talc pleurodesis in 586, 586 pleural effusion 71 surgical management 536
malignant pleural effusions 586, pleural fibrosis 71–2, 72 tremolite 499
589 pleurodesis 72, 72–3, 73, 570 trichomoniasis 385
techniques 584–7 transforming growth factor beta-1 triglycerides
endoscopy 585 (TGF-b1) 31, 118, 194 chylothorax 391
single-entry 584, 585–6, 586 transgenic models, malignant pleural fluid analysis 221
two-entry 584–5, 586–7 mesothelioma 193 troglitazone 437–8
training 583 transinfection models, malignant trypsin 141–2
tuberculous pleural effusions 590–2 mesothelioma 193 L-tryptophan 439
Abram’s needle biopsy vs. 591 transjugular intrahepatic portosystemic tube thoracostomy 260–6
fibrous adhesions 590, 591 shunt (TIPS) empyema 263–5, 264
other techniques vs. 590, 591 cirrhotic hydrothorax 607 intrapleural fibrinolytics 263–5
Index 649
ultrasound-guided interventions – contd vascular caused effusions 397–408, 406 pleural infection 358–9, 360, 548
pleural infection 348, 349 vascular endothelial growth factor spontaneous pneumothorax 525,
principles 280 (VEGF) 79–81 600, 600, 602
thoracentesis 280, 552, 556–7 catheter instillation 267 thoracic duct ligation 607, 607
transthoracic fine needle aspirations empyema fluids 80 thoracoscopy vs. 584
281 inflammation 33, 64–5 thoracotomy vs. 600, 601, 604–5
unfractionated heparin 402 inhibitors 80 vimentin 288, 303
uPA-PAi-urokinase receptor (uPAR) malignant mesothelioma growth viral hepatitis 386
system 103, 104 regulation 32 viral infection 385–6, 548
upper abdominal abscess-associated malignant pleural effusion 80 visceral pleura 14, 44
pleural effusions 468–9, 469 ovarian hyperstimulation syndrome bleb formation 22
urinothorax 474 blood supply 17, 44, 44–5
kidney transplant recipients 415 pleural effusion 79–80 innervation 18
pleural fluid analysis 204, 210, 217 pleural fluid formation 79–81 lymphatic circulation 19
urokinase pulmonary embolism 398 mesothelial cells 22
empyema 265 targeting 80–1 radiology 234, 235, 273
parapneumonic effusions 265 vascular endothelial growth factor regional differences 22
pleural fibrosis therapy 106 (VEGF) antagonists 80–1 vitamin(s) 437
pleural infection 354–5 vascular endothelial growth factor vitamin B6 437
urokinase receptor (uPAR) (VEGF) receptors 80 vitamin H 437
expression 103–4 vascular leak syndrome (VLS) 83 vitamin-K antagonists 402
expression blockage, malignant vascular permeability factor see vascular vitronectin 33, 115
mesothelioma therapy 108 endothelial growth factor (VEGF) volume of distribution (Vd) 152–3
phosphoglycerate kinase binding 105 vasovagal reactions, thoracentesis-
post-transcriptional regulation 104, induced 556 Wang thoracentesis needle 553, 556
105 venous thrombosis 280 warfarin 437
urokinase-type plasminogen activators ventricular diastolic collapse 55 water channels 47
(uPA) 33, 102, 102 Vero cells 617 water manometers 229
cytokine regulation 103 video-assisted thoracoscopic surgery water valve 8
inhibition 103 (VATS) Wegener’s granulomatosis 286, 426
malignant mesothelioma 108 catamenial pneumothorax 602 Wells score, pulmonary embolism 399,
post-transcriptional regulation 104, chylothorax 546–7, 607, 607 400
105–6 clotted hemothorax 606–7 white blood cell (WBC) count,
US Army Empyema Commission 342 empyema 603, 603, 604 pulmonary embolism 399
Ussing chambers 180 fibrinolysis vs. 603–4 Wilm’s tumor gene (WT-1) 191
guided pleural biopsy vs. 266 Wilm’s tumor gene product (WT1
v-src gene 191 hemothorax 406 protein) 305
Vaccinia virus, gene therapy 614, 617 hepatic hydrothorax 458–9 wollastonite 114, 177, 499
valproic acid 439 intercostal drainage vs. 359
Valsalva maneuvers 477 malignant empyema 605, 605
vancomycin 163 parapneumonic effusion 107 yellow nail syndrome 205–6, 394–5,
Vanderschueren classification, pediatric pleural effusion 546 395
spontaneous pneumothorax 593 persistent air leaks, traumatic
‘vanishing tumors’ 15 pneumothorax 536 ZD6464 80