2 Respiratory Tract
2 Respiratory Tract
2 Respiratory Tract
Respiratory Tract
Christopher A. French
Exfoliative cytology was first used to study cells of the of direct sampling methods via bronchoscopy and fine-
respiratory tract in 1845.1 The ability to diagnose pulmo- needle aspiration (FNA),3 resulting in an armamentar-
nary diseases cytologically was appreciated as early as ium of sampling techniques. Since then, the improved
1919,2 but it was not until the 1950s and 1960s that pul- sensitivity (and common application) of thoracic imag-
monary cytology came into its own as a diagnostic disci- ing has created an ever-increasing need for the cytologic
pline. Its emergence was bolstered by the introduction evaluation of pulmonary lesions.
65
66 Respiratory Tract
Anatomy and cellular components of the Figure 2.1 Normal ciliated bronchial cells (bronchial brush-
ing). These columnar cells have oval nuclei and finely stippled
respiratory tract:
chromatin. Numerous cilia project from the apical surface.
Upper respiratory tract:
• ciliated columnar cells
thin and cover the gas exchange portion of the alveolar
• squamous cells
surface. The type II pneumocyte is more conspicuous:
Lower respiratory tract: plump and cuboidal rather than flat. It secretes pulmo-
• trachea and bronchi nary surfactant, seen ultrastructurally as osmiophilic
• ciliated columnar cells lamellar bodies. After lung injury, these cells function
• goblet cells as reserve cells for the delicate type I pneumocyte. On
• basal or reserve cells cytologic preparations, type II pneumocytes are round
• neuroendocrine cells and have vacuolated cytoplasm; they can be difficult to
• terminal bronchioles distinguish from macrophages.
• nonciliated cuboidal or columnar cells (Clara cells) Alveolar macrophages vary in appearance depend-
• alveoli ing on the amount and type of phagocytosed cytoplas-
• type I and II pneumocytes mic material. In general, they have one or more round
• alveolar macrophages to oval nuclei and lacy or bubbly cytoplasm (Fig. 2.2).
Numerous alveolar macrophages must be identified for
The trachea and bronchi are lined by a pseudostrat- a sputum sample to be judged adequate. Under normal
ified epithelium. The predominant cell is the ciliated
columnar cell, which has a basally placed nucleus and
finely textured chromatin. The luminal surface has
a thick terminal bar with cilia (Fig. 2.1). Goblet cells,
present in a ratio of approximately one per six cili-
ated cells, also have basally located nuclei, but they
lack cilia and their cytoplasm is distended by mucus.
The goblet cells secrete mucus, and the ciliated cells
move the mucus and entrapped contaminants up
the airway. Adjacent to the basement membrane are
basal or reserve cells: small, undifferentiated cells that
are the presumed forerunners of the ciliated and gob-
let cells. Neuroendocrine cells, or Kulchitsky cells, are
also present in the respiratory epithelium, but they are
identified only with special stains or ultrastructural
examination; they are argyrophil-positive and possess
dense-core granules.
The terminal bronchioles are lined by nonciliated
cuboidal to columnar cells called Clara cells, usually not Figure 2.2 Pulmonary alveolar macrophages (sputum).
Pulmonary macrophages have abundant foamy cytoplasm that
recognized as such on cytologic preparations. The alveo- often contains black carbon particles, as seen here. These mac-
lar lining consists of type I and type II pneumocytes. Type rophages fill with hemosiderin pigment following pulmonary
I pneumocytes, which are more numerous, are paper hemorrhage. Hemosiderin is golden brown rather than black.
Sampling Techniques, Preparation Methods, Reporting Terminology, and Accuracy 67
circumstances, a few white blood cells, such as neutrophils A simple method of sputum preparation is known as
and lymphocytes, are also found within the alveolar com- the “pick and smear” technique, whereby fresh sputum is
partment. An increased number of inflammatory cells is examined for tissue fragments, blood, or both. Smears are
abnormal: Abundant neutrophils indicate an acute pneu- prepared from areas that contain these elements and are
monia, and numerous lymphocytes are usually associ- immediately fixed in 95% ethanol. A modification of this is
ated with chronic inflammation. the Saccomanno method, which calls for sputum to be col-
lected in 50% ethanol and 2% carbowax.8 The specimen is
then homogenized in a blender and concentrated by cen-
trifugation. Improved sensitivity has been demonstrated
Sampling Techniques, by the use of dithiothreitol (DTT) for homogenization.9
Preparation Methods, Smears are made from the concentrated cellular material.
Reporting Terminology, The Saccomanno method must be performed in a biologic
and Accuracy safety hood as a result of the risks of infection from aero-
solization. Sputum can also be processed using thinlayer
Familiarity with the variety of sampling and preparation methods or embedded in paraffin for cell block sections.10
methods is crucial for cytologic interpretation because The adequacy of a sputum sample is established by
cytomorphology is different depending on the sampling finding numerous pulmonary macrophages. They indicate
and preparation method. The accuracy of respiratory that a deep cough specimen of the lower respiratory tract
cytology also varies depending on the specimen type. has been obtained. Specimens consisting merely of squa-
As with other nongynecologic cytology specimens, mous cells, bacteria, and Candida organisms are unsat-
respiratory tract diagnoses are typically reported as “nega- isfactory because they represent only oral contents. Even
tive for malignant cells,” “positive for malignant cells,” or ciliated cells, which also line the sinonasal passages, do not
“non-diagnostic (unsatisfactory),” followed by a descrip- guarantee that a sample is from the lower respiratory tract.
tive diagnosis. Inconclusive diagnoses are commonly The sensitivity of sputum cytology for the diagnosis
reported as “atypical cells present” (usually connoting a of malignancy increases with the number of specimens
lower degree of suspicion) or “suspicious for malignancy examined, from 42% with a single specimen to 91% with
(connoting a high degree of suspicion),” depending on the five specimens.11 The specificity of sputum examination
degree of suspicion. Cancer is confirmed in 40% of “atyp- is high, ranging from 96% to 99%, and the positive and
ical” respiratory specimens and in almost 70% of those negative predictive values are 100% and 15%, respec-
diagnosed as “suspicious.”4 Atypical or suspicious cases tively.12 Negative sputum results do not guarantee the
remain inconclusive even after careful retrospective reex- absence of a malignancy, especially in a patient suspected
amination, which fails to reveal any morphologic features of having lung cancer. The sensitivity of sputum cytology
to reliably distinguish benign from malignant specimens.5 depends on the location of the malignant tumor: 46%
to 77% of central lung cancers but only 31% to 47% of
peripheral cancers are diagnosed by sputum cytology.13,14
Sputum Surprisingly, sensitivity is independent of tumor stage
Sputum consists of a mixture of cellular and noncellu- and histologic type. Accuracy in tumor classification is
lar elements that are cleared by the mucociliary appa- 75% to 80%15 and is tumor type dependent.16
ratus. It was once the most common respiratory tract
specimen because it is relatively easy to obtain and
causes little if any discomfort. Unfortunately, screen-
Bronchial Specimens
ing asymptomatic smokers with sputum cytology does A pivotal improvement in sampling cells from the lower
not decrease mortality from lung cancer. Today, sputum respiratory tract occurred with the development of the
cytology is generally reserved for symptomatic individ- flexible bronchoscope in the late 1960s. Today, any part
uals. Even here, with the advent of bronchoscopy and of the respiratory tract can be sampled with this device.
FNA, its use as the mainstay in respiratory cytology has Complications of bronchoscopy are rare (0.5% and
declined significantly. 0.8% for major and minor complications, respectively),17
Collecting multiple sputum samples over several days and include laryngospasm, bronchospasm, distur-
optimizes sensitivity. Early morning, deep cough speci- bances of cardiac conduction, seizures, hypoxia, and
mens are preferred.6 If the patient is not able to expecto- sepsis. The incidence of major complications is higher
rate adequately, expectoration can be induced by having for transbronchial biopsy (6.8%).17
the patient inhale nebulized water or saline. Sputum
induction increases the detection of lung cancer.7 When
Bronchial Aspirations and Washings
prompt preparation of sputum is not possible, the
patient can expectorate into a 70% ethanol solution, Bronchial secretions can be aspirated directly from the
which prefixes the specimen. lower respiratory tract through the bronchoscope, but
68 Respiratory Tract
an alternative (and more common) method is to “wash” Cryptococcus.26 The frequency and distribution of infec-
the mucosa by instilling 3 to 10 mL of saline and aspirate tions has changed since the widespread use of highly
the washings. The fluid is centrifuged and the concen- active antiretroviral therapy (HAART) to treat HIV.27
trate used to make smears, thinlayer preparations, or Among individuals who are HIV-seropositive with non-
cell blocks; the latter are particularly useful when special specific cytologic results, 27% prove to have pathogens,
stains are needed. usually bacterial or fungal, by either culture or biopsy,28
which emphasizes the importance of a multimodal
Bronchial Brushings approach to diagnosis in this setting.
BAL is also used for the diagnosis of malignancy, with
Fiberoptic bronchoscopy allows direct visualization
sensitivity that ranges from 35% to 70%.29,30 The sensitiv-
and sampling of the tracheobronchial tree. A brush is
ity of BAL for detecting malignancy is higher for multifo-
applied to the surface of an endobronchial lesion, and
cal or diffuse tumors like bronchioloalveolar carcinoma.31
the entrapped cells are either smeared onto a glass slide
False-positive results are occasionally encountered as a
or rinsed in a collection medium for thinlayer or cell
result of atypical type II pneumocytes in the setting of
block preparation. If smears are made, immediate fixa-
pneumonia, diffuse alveolar damage,32 bone marrow
tion (by immersion into 95% ethanol or spray fixation) of
transplantation,33 and chemotherapy.34
the smears is essential to preserve morphologic detail.6
The diagnostic accuracy of bronchial washing or
brushing cytology is comparable to that of bronchial Transbronchial Fine-Needle
biopsy.18 Brushings with cell block preparation some- Aspiration
times detect malignancy more reliably than bronchial
Transbronchial FNA is especially useful for the diagno-
biopsies.19 Accuracy improves when clinical history is
sis of primary pulmonary lesions that lie beneath the
provided with the specimen.20 The diagnostic yield also
bronchial surface and for staging lung cancer patients by
improves when several different sampling methods are
sampling mediastinal lymph nodes.35-37 In these settings,
used in concert.21,22
the need for additional surgical procedures is eliminated
in 20% of patients, and the cost is one third that of medi-
Bronchoalveolar Lavage
astinoscopy.38 The lesion is aspirated with a retractable
The choice between using bronchoalveolar lavage (BAL) (Wang) needle passed through a flexible catheter that is
and bronchial washing depends on the location of the sent down the bronchoscope.
airway one desires to sample. With BAL, the broncho- When used to sample mediastinal lymph nodes, at
scope is wedged into position as far as it will go, and least a moderate number of lymphocytes must be pres-
distal airways are flushed with several aliquots of sterile ent to ensure the adequacy of the specimen and avoid
saline. The first aliquot is more representative of the cel- a false-negative result.39 Complications from transbron-
lular material from larger airways, whereas subsequent chial aspiration are rare and include endobronchial
aliquots reflect the alveolar compartment. bleeding, which is usually controlled by suctioning.
BAL is particularly useful for the diagnosis of oppor- Contraindications are coagulopathy, respiratory failure,
tunistic infections in patients who are immunocompro- and uncontrollable coughing.40
mised. The specimen can be examined cytologically and Transbronchial FNA augments the diagnostic accu-
a portion also submitted for microbiologic studies. The racy of bronchial washings, brushings, and endoscopic
distinction between oral contamination and a real bac- biopsies for the detection of primary pulmonary neo-
terial infection can be difficult, but an abundance of nor- plasms.41,42
mal squamous cells usually indicates contamination by
oral flora, whereas neutrophils imply a real infection.23 Accuracy of transbronchial fine-needle
In patients who are immunocompromised, the diag- aspiration:
nostic yield for infectious pathogens is 39%, the sen-
• sensitivity (FNA alone): 56%
sitivity 82%, and the specificity 53%.24 In patients with
• sensitivity (combined with bronchial brush, wash,
acquired immune deficiency syndrome (AIDS), BAL has
and biopsy): 72%
a sensitivity for documenting infection that is compara-
• specificity: 74%
ble to that for transbronchial biopsy (86%); when used
• positive predictive value: 100%
in combination with biopsy, the sensitivity increases to
• negative predictive value: 53% to 70%
98%.25 Historically, the most common pulmonary patho-
gens that were detected by BAL in individuals who were
human immunodeficiency virus (HIV) seropositive Transbronchial FNA is accurate in distinguishing small
were Pneumocystis carinii (78%) and bacteria (19%); the cell from non–small cell lung cancer.43 For mediastinal
remaining infections were the result of Mycobacterium staging of bronchogenic carcinoma, the negative pre-
tuberculosis, atypical mycobacteria, Histoplasma, and dictive value of transbronchial FNA increases from
Sampling Techniques, Preparation Methods, Reporting Terminology, and Accuracy 69
36% to 78% when negative specimens without sufficient passes through aerated lung and decreases if the path
lymphocytes are regarded as unsatisfactory for evalu- does not traverse aerated lung.53 Transient hemoptysis
ation.39 The most common cause of false-negatives is occurs in 5% to 10% of patients. Other complications
sampling error.42 The accuracy of mediastinal staging are rare, and include hemopericardium, hemothorax, air
improves with the use of ultrasound guidance.35,36 embolism, tumor seeding, and death.53,54
Percutaneous FNA is usually performed by radiolo-
Transesophageal Fine-Needle gists using computed tomography (CT) or ultrasound
guidance. The needle gauge ranges from 18 to 22, and
Aspiration many different types of needle devices are available.
Mediastinal lymph node sampling can be performed not Although many radiologists prefer 22-gauge Chiba nee-
only bronchoscopically, but also endoscopically by pass- dles, these require repuncture if more than one pass
ing the needle through the esophagus.44-46 The addition is needed. Another choice is the coaxial needle, with a
of ultrasound guidance improves the accuracy of medi- large-bore outer needle serving as the guide for a small-
astinal lymph node sampling.47 Like bronchoscopic FNA, bore inner needle. Once the outer needle is positioned,
endoscopic FNA realizes significant cost savings46 and more than one aspiration can be performed using the
reduces the number of unnecessary thoracotomies.45 inner needle.
Endoscopic transesophageal FNA has a diagnostic It can be helpful in some cases if a cytotechnologist
accuracy (70% to 80%)48,49 that approaches that of ultra- or cytopathologist attends the FNA procedure to assist
sound-guided transbronchial FNA (84% to 96%).37,44,45,47 with specimen handling and assess its adequacy on-site.
When used in combination with transbronchial FNA, After direct smears are prepared, the needle is rinsed
the diagnostic yield for mediastinal staging is greatly in a balanced electrolyte solution, Saccomanno’s fixa-
improved, with an accuracy that approaches 100%.50 As tive, 50% ethanol, or commercial preservative solution.
with transbronchial FNA, a moderate number of lym- The cellular suspension can be processed for microbio-
phocytes must be present to ensure the adequacy of the logic analysis, cytospins, thinlayer preparations, filters,
specimen and avoid a false-negative result. cell blocks, cytogenetic analysis, flow cytometric analy-
sis, or electron microscopy. Formalin-fixed cell blocks
Percutaneous Fine-Needle are indispensable for histochemical and immunohisto-
chemical stains. A decision regarding the need, if any, for
Aspiration special studies can be made by the cytotechnologist or
The ease, rapidity of diagnosis, and minimal morbid- cytopathologist after examination of the direct smears.
ity of percutaneous FNA make it an attractive alterna- Percutaneous FNA is a reliable and accurate way to
tive to surgical biopsy in the evaluation of the patient diagnose many pulmonary neoplasms.
with a pulmonary mass. Thus, FNA is of greatest benefit
to patients for whom it spares a more invasive surgical Accuracy of percutaneous fine-needle
procedure. Surgical intervention, in fact, can be avoided aspiration:
in up to 50% of patients with clinically suspected lung
• sensitivity: 89%
cancer.51 There are some contraindications, however.
• specificity: 96%
• positive predictive value: 98%
Relative contraindications to percutaneous
• negative predictive value: 70%
fine-needle aspiration:
• false-positive rate: 0.85%
• chronic obstructive pulmonary disease (COPD) • false-negative rate: 8%
• emphysema
• uncontrollable coughing
In a study of more than 13,000 FNA specimens from
• uncooperative patient
436 institutions, the diagnostic sensitivity was 89% for
• bleeding diathesis (i.e., anticoagulant therapy)
the procedure itself and 99% for the pathologist’s inter-
• severe pulmonary hypertension
pretation.55 This difference indicates that most false-
• arteriovenous malformation
negative results are the result of sampling error. About
• cardiac disease
15% of false-positive diagnoses and 5% of false-negative
• suspected echinococcal cyst52
diagnoses have a significant, permanent, or grave influ-
ence on patient outcome.55 The reliability of a negative
The most common complication of percutaneous FNA result is a matter of controversy, given that negative
FNA is a pneumothorax. A radiographically detectable predictive values range from 34% to 88%.56-59 For this rea-
pneumothorax occurs in 21% to 34% of patients;53 only son, most investigators recommend a repeat aspiration
10%, however, require intercostal drainage tubes.53 The or tissue biopsy when a specific benign diagnosis that
risk of a pneumothorax increases with the number of accounts for the lesion cannot be made with certainty.
70 Respiratory Tract
Figure 2.3 Mesothelial cells (fine-needle aspiration [FNA]). Benign mesothelial cells are occasionally seen in transthoracic fine-
needle aspirates. They are arranged in flat, cohesive sheets. The cells have round or oval nuclei, small nucleoli, and a moderate
amount of cytoplasm, and space between cells—windows—can be appreciated.
Benign Cellular Change 71
Figure 2.4 Reactive bronchial cells (bronchial brushing). Bronchial Reserve Cell Hyperplasia
Reactive bronchial cells may show marked nuclear size variation.
Note that cilia—evidence of their benign nature—are retained. As the surface epithelium of the respiratory tract is shed
during lung injury, reserve cells proliferate and are seen
in bronchial washings and brushings (Fig. 2.7).
A B
Figure 2.13 Alternaria (bronchoalveolar lavage [BAL]). This pigmented fungus is rarely pathogenic. It can contaminate virtually
any cytologic specimen, including cervicovaginal smears, cerebrospinal fluid, and urine. A, The slender septate stalks (conidiophores)
are occasionally branched (not shown). B, The conidia are snowshoe-shaped and have both transverse and longitudinal septations.
Legionella pneumonia is caused by the aerobic gram- be an abundance of acid-fast organisms but few well-
negative bacteria Legionella sp., of which the most formed granulomas. If Romanowsky-type stains are used
common is L. pneumophila. The organisms are seen in such cases, the abundant acid-fast organisms can be
well with silver stains like the Steiner, Warthin-Starry, or identified as negative images.
Dieterle stains. A specific identification of L. pneumoph-
ila can be made in BAL samples using immunohisto-
chemical or immunofluorescent methods.
Pulmonary Fungal Infections
Pulmonary fungal infections are readily diagnosed by
cytology, particularly in transthoracic FNAs, and should
Tuberculosis be suspected whenever there is granulomatous inflam-
Infection by M. tuberculosis commonly results in granu- mation or necrosis. Cell block material can be used for
lomatous inflammation (Fig. 2.15). Cytologic specimens silver or periodic acid-Schiff (PAS) stains. Many fungi
contain aggregates of epithelioid histiocytes, lympho- have a characteristic microscopic appearance that
cytes, and Langhans giant cells. Necrosis may or may not enables a rapid, specific diagnosis (Table 2.2).
be evident. Granulomas by themselves, however, are a
nonspecific finding and can be seen in other conditions
Cryptococcosis
like fungal infections and sarcoidosis. A definitive dia
gnosis of tuberculosis rests on identifying the organ- Cryptococcus neoformans is an inhabitant of the soil and is
isms with the help of a special stain (Ziehl-Neelsen) or found in bird droppings. It can act as both a primary and
by microbiologic culture. Cell block preparations are an opportunistic pathogen and may involve numerous
particularly useful for special stains, but rarely are more body sites. Pulmonary invasion by this fungus is heralded
than one or just a few organisms identified. (By contrast, clinically by a productive cough, fever, and weight loss.
infection by M. avium intracellulare, as seen in patients
who are immunocompromised patients, often yields
Histoplasmosis
innumerable acid fast organisms.) A sensitive (93%)
and specific (99%) assay called the Mycobacterium Histoplasma capsulatum, another soil inhabitant, is con-
Tuberculosis Direct Test (MTD) is also available for the tracted by the inhalation of spores and more commonly
detection of M. tuberculosis and can be applied to respi- affects patients who are immunocompromised. The dis-
ratory specimens like sputum.74 The assay amplifies ease can mimic tuberculosis clinically, in that peripheral
M. tuberculosis ribosomal ribonucleic acid (RNA) by the nodular lesions and mediastinal lymphadenopathy are
polymerase chain reaction. relatively common. The organism, often present within
In countries where M. tuberculosis is prevalent, the the cytoplasm of macrophages, is so small that it may be
yield of acid-fast bacteria among all clinically suspi- overlooked if silver stains are not used.
cious lung masses can be quite high.75,76 In patients who
are immunocompromised with tuberculosis, there may
Blastomycosis
Blastomyces dermatitidis inhabits wooded terrain.
Although the lung is the primary target of infection,
there may be distant spread to other organs, such as
skin, bone, and the urinary tract.
Coccidioidomycosis
Coccidioides immitis infection is quite common in
endemic areas of the Southwest and Western United
States, giving rise to positive skin tests in more than
80% of individuals in these areas. It produces a respira-
tory infection that usually resolves spontaneously, but
persists as a pulmonary mass in about 2% of patients.
Multiorgan dissemination is more common in the
patient who is immunocompromised. Because BAL
Figure 2.15 Granuloma (fine-needle aspiration [FNA], or bronchial washings detect less than 50% of culture-
M. tuberculosis infection). The nodular aggregate of epithe- positive cases,77 cytologic diagnosis is best documented
lioid histiocytes, which defines the granuloma, has a syncytial
appearance because individual cell borders are indistinct. Note by transthoracic FNA. The organisms appear as mature
the curved and elongated, boomerang shape of some of the (sporulating) or immature spherules, often with a frac-
histiocytic nuclei. Interspersed lymphocytes can also be seen. tured (broken ping-pong ball) appearance, and free
Infections 77
endospores (see Table 2.2). These are present on a back- erroneous diagnosis of SQC in cytologic material (see
ground of granular eosinophilic debris with little inflam- Fig. 2.25).80 There is a high sensitivity (50% to 90%) for
mation. Hyphae are seen in 5% of cases. FNA cytology the detection of the organism in cell block preparations
is diagnostically far superior to the culture of aspirated of sputum.81
materials.78
Sporotrichosis
Paracoccidioidomycosis
Pulmonary infection caused by Sporothrix schenckii
Paracoccidioidomycosis, also known as South American is uncommon and occurs mainly in patients who are
blastomycosis, is caused by the dimorphic fungus immunocompromised, including diabetics and alcohol-
Paracoccidioides brasiliensis (Fig. 2.16) and is the most ics. The clinical symptoms are nonspecific. Though often
common systemic mycosis in Latin America.79 It fre- self-limited, these infections can become chronic, with
quently involves the lungs and mucocutaneous areas mass lesions or cavitary nodule formation. The yeasts
of healthy individuals, and clinically simulates tubercu- resemble Cryptococcus, Histoplasma, and Candida, and
losis. There may be a hyperplasia of the bronchial epi- therefore culture or fluorescent antibody staining is
thelium overlying granulomas, which may lead to the necessary for definitive diagnosis.82
78 Respiratory Tract
A B
C D
Figure 2.17 Pneumocystis carinii (bronchoalveolar lavage [BAL]). A, With the Papanicolaou stain, the pneumocystis organisms
are not seen, but foamy proteinaceous spheres characteristic of this infection are identified. B, The cysts, which have a cup-shaped
configuration and a central dark zone, are seen with the methenamine silver stain. C, The Giemsa stain outlines the cysts as nega-
tive images, and stains the intracystic bodies or trophozoites. Each cyst, as seen here, contains eight intracystic bodies. D, The direct
immunofluorescence test is highly sensitive, revealing green fluorescent-stained organisms and their extracellular products.
Non-Neoplastic,
Noninfectious Pulmonary
Diseases
Sarcoidosis
This common disease of unknown etiology is character-
Figure 2.18 Strongyloides (sputum). These large round- ized by noncaseating granulomas in many organs, most
worms are distinguished by their short buccal cavities (arrow). commonly the lung.
80 Respiratory Tract
Figure 2.20 Pulmonary alveolar proteinosis (bronchoalveolar lavage [BAL]). Hematoxylin-eosin (H & E) stained cell block sections
show numerous acellular eosinophilic aggregates.
A B
Figure 2.21 Organizing pneumonia (fine-needle aspiration [FNA]). Some pneumonias respond slowly to antibiotic treatment;
such lesions may mimic a neoplasm radiographically. A, Smears typically show an admixture of lymphocytes and foamy macrophages.
B, Cell block preparations show fragments of granulation tissue composed of loosely arranged fibroblasts admixed with chronic
inflammatory cells.
age of 40. IMT is usually a peripheral, discrete, solitary eosinophilic cytoplasm and are surrounded by a thick-
nodule. These tumors behave unpredictably: most patients ened basement membrane.
have an excellent prognosis, but some tumors are locally
aggressive.107 The neoplastic nature of this lesion was Cytomorphology of endobronchial
confirmed by the cloning of translocations involving the granular cell tumor:
anaplastic lymphoma kinase (ALK) gene (chromosome
• small clusters of macrophage-like cells
2p23) and the two tropomyosin genes, TPM3 and TPM4.108
• abundant granular cytoplasm
Anaplastic lymphoma kinase fusion partner variants
• small, uniform, round to oval nuclei
include the clathrin heavy chain gene109 and ATIC gene.110
B
Figure 2.22 Pulmonary hamartoma (fine-needle aspiration [FNA]). A, Smears from a pulmonary hamartoma show fragments of
myxoid material, chondroid material, or both. Chondrocytes in lacunae, which are green with Papanicolaou’s stain, but unstained on
hematoxylin-eosin (H & E) sections, are seen here. B, Epithelial cells and adipocytes (latter not shown) are also often present.
with SQC.119 Dysplastic changes have been demonstrated sure to tobacco smoking, and, as a result, lung cancer
to regress in experimental systems.120 is more common in developed countries. It is the most
common fatal malignancy in both men and women in
the United States, where it accounts for 31% of all cancer
Lung Cancer deaths in men and 26% in women.121
The causes of lung cancer are many. In men, 85%
Carcinoma of the lung is the most common cancer in the of lung cancers are attributed to cigarette smoking.122
world today. Its geographic distribution parallels expo- Tobacco smoke contains numerous carcinogens, tumor
84 Respiratory Tract
initiators, and radioactive compounds, but the biologic more important as novel therapies are developed for
mechanism by which tobacco substances initiate lung specific lung cancer subtypes, like adenocarcinomas
cancer is not known. Genetic alterations play an impor- with EGFR mutations.
tant role. Chromosome 3p deletion (seen commonly Lung cancers often show histologic heterogeneity.
in neuroendocrine tumors)118 and p53, K-ras, and p16 Almost 50% of lung cancers contain more than one his-
mutations are common in lung carcinomas.123,124 p53 tologic type.122 Thus, careful examination of all cytologic
mutation, the most frequent mutation in human cancers, material for more than one histologic component is
likely occurs in lung cancer via a G-C to T-A transversion essential for correct classification.
induced by cigarette smoke.125 Exposure to asbestos has
a synergistic effect with smoking: Asbestos workers who
smoke increase their risk of developing pulmonary can-
Squamous Cell Carcinoma
cer by at least 50-fold. The development of lung cancer SQC of the lung accounts for approximately one third of
has been linked to exposure to other substances, includ- all primary pulmonary malignancies. Two thirds of these
ing radon, crystalline silica, nickel compounds, and tumors occur in a central location, and the remainder
organic compounds like benzene and vinyl chloride. arise in smaller bronchi. Postobstructive pneumonia
Universal screening for lung cancer with chest radio- and cavitation of the tumor are common. Of all the his-
graphs and sputum cytology is not cost effective. The tologic subtypes, SQC is the one most commonly asso-
most comprehensive trial in the United States was ciated with hemoptysis. Apical SQCs in the superior
undertaken by three medical centers under the aus- sulcus are called Pancoast tumors, characterized by pos-
pices of the National Cancer Institute. Approximately terior rib destruction and neural invasion, resulting in
30,000 men were screened, and the overall mortality of severe pain and Horner syndrome (enophthalmos, pto-
the screened group was no lower than that of the con- sis, miosis, and ipsilateral decreased sweating). Several
trol group.126 High-risk individuals might benefit from histologic variants (papillary, clear cell, small cell, basa-
periodic screening, however, like workers in the asbes- loid) have been described.122
tos or uranium industries, persons over the age of 65
with a 20-year history of smoking, and those whose Cytomorphology of well-differentiated
radiographs show a persistent abnormality.127 The use squamous cell carcinoma:
of low-dose spiral computed tomography with spu-
tum cytology has shown mixed results;128-132 early can- • abundant dyshesive cells
cer detection comes at the cost of unnecessary benign • polymorphic cell shapes: polygonal, rounded,
nodule detection and removal, and exposure to poten- elongated (fiber-like), tadpole shaped
tially harmful, repeated radiation exposures.128,133 Large, • dense cytoplasmic orangeophilia (Papanicolaou
prospective, multi-institutional trials with high-risk stain)
patients are needed to prove the efficacy of low-dose • pyknotic nuclei
spiral computed tomography with sputum screen- • frequent anucleate cells
ing.134,135 Detection of epigenetic or genetic aberrations
in frequently mutated lung cancer genes (p53, K-ras, The cytologic features vary depending on the degree
p16, HVAL2, FHIT, SFTPC) in sputum samples shows of squamous differentiation. In a well-differentiated
promise but is not yet widely applied.124,136,137 SQC, the malignant cells are dyshesive, come in a variety
Patients with lung cancer often present with shortness of shapes (polygonal, spindle, tadpole, Fig. 2.23A and B),
of breath, cough, chest pain, hoarseness, hemoptysis, or and have abundant smooth and dense cytoplasm that is
pneumonia. Some tumors, particularly adenocarcino- filled with keratin. The cytoplasm stains green, yellow,
mas, are detected incidentally in individuals who are or orange with the Papanicolaou stain and robin’s egg
asymptomatic. blue with Romanowsky stains. Nuclei are usually small,
Four histologic types account for 95% of all pulmo- hyperchromatic, and smudgy, and nucleoli are often
nary tumors: SQC, adenocarcinoma, large cell carci- inconspicuous.
noma (LCC), and small cell carcinoma. Because it is
usually metastatic at the time of detection, small cell Cytomorphology of moderately and
carcinoma is treated with systemic chemotherapy. The poorly differentiated squamous cell
treatment for the non–small cell carcinomas (SQC, carcinomas:
adenocarcinoma, and LCC) is essentially identical: • large, cohesive clusters of elongated cells
lobectomy or pneumonectomy for localized tumors. • rare to absent keratinization
The subclassification of the non–small cell carcinomas • large nuclei
is thus less important than the distinction between • coarse chromatin texture (“Idaho potato”)
non–small cell and small cell carcinoma. Finer distinc- • ± prominent nucleoli
tions (histologic and even molecular) are becoming
Lung Cancer 85
A B
Figure 2.23 Squamous cell carcinoma ([SQC]; fine-needle aspiration [FNA]). A, Well-differentiated SQCs are composed of cells
with dense, orangeophilic cytoplasm and hyperchromatic, often pyknotic nuclei, some with angular contours. B, Bizarre, elongated,
spindle-shaped cells are common. Often there is abundant granular debris, seen here, and inflammation.
Figure 2.25 Atypical squamous metaplasia (fine-needle aspiration [FNA]). Cavitary fungal infections, such as this one by
Aspergillus (inset), are among the causes of reactive squamous atypia, a mimic of squamous cell carcinoma (SQC).
Cytomorphology of adenocarcinoma:
• honeycomb-like sheets, three-dimensional
clusters, acini, papillae Figure 2.27 Adenocarcinoma (bronchial washing). These
• eccentrically placed, round or irregular nuclei malignant cells have large, round nuclei with distinct nucleoli,
open chromatin, and lacy, clear cytoplasm, and form a honey-
• finely textured chromatin comb-like arrangement.
• large nucleoli
• mucin vacuoles
• translucent, foamy cytoplasm are prominent in many adenocarcinomas. Psammoma
bodies can be seen in the papillary and bronchioloal-
veolar variants.152,153
In cytologic preparations, adenocarcinoma cells can Cytologic preparations from a mucinous BAC show
be arranged in a variety of architectural patterns: three- uniform cells with pale, optically clear nuclei and incon-
dimensional clusters, flat sheets, acini, and papil- spicuous nucleoli. Grooves and nuclear pseudoinclu-
lae (Figs. 2.26 and 2.27). The clustered cells tend to sions are often present (Fig. 2.28). In some cases, the
be loosely cohesive, such that isolated cells and small cells of a mucinous BAC are dispersed and resemble
groups are also present. Cytoplasm is usually abundant, macrophages. A nonmucinous BAC is indistinguishable
with well-defined borders. It can be thin and translu- from an ordinary adenocarcinoma on cytologic prepa-
cent or foamy. Some tumors have cells with a promi- rations (Fig. 2.29). Neither the mucinous nor the non-
nent large mucin vacuole. Nuclei are often eccentrically mucinous subtype of BAC can be diagnosed reliably on
placed and usually have round, smooth contours, but cytologic specimens, because the diagnosis is based on
in some tumors the nuclear membranes can be highly architecture: the absence of stromal, vascular, or pleural
irregular. Chromatin is finely textured in most adeno- invasion.122 A mucinous BAC can be suspected, however,
carcinomas, but some poorly differentiated adenocar- when characteristic cytology is correlated with imaging
cinomas have coarsely textured chromatin. Nucleoli findings.122,154
88 Respiratory Tract
Figure 2.28 Bronchioloalveolar carcinoma, mucinous type (fine-needle aspiration [FNA]). The crowded sheets of these tumor
cells, with their pale nuclei, small nucleoli, and occasional nuclear grooves, and intranuclear pseudoinclusions (arrows), are reminis-
cent of papillary carcinoma of the thyroid. Abundant extracellular mucin (not shown) is often present.
Lung Cancer 89
Figure 2.30 Large cell carcinoma ([LCC]; bronchial wash- Various types of lung injury (e.g., irradiation, infarc-
ing). The cells of this tumor are loosely arranged in aggregates. tion) can result in highly atypical bronchial cells or pneu-
Nuclei are markedly enlarged with focal chromatin clearing and
mocytes that mimic LCC and other tumors. The cells of
multiple irregular nucleoli.
a florid type II pneumocyte hyperplasia can be large and
wildly pleomorphic, but most patients with this phe-
nomenon are acutely ill with diffuse alveolar damage.
roendocrine differentiation is required, with clear-cut In patients who are acutely ill or those with other injury
reactivity for at least one of the neuroendocrine mark- to the lung, a conservative approach to diagnosis is rec-
ers (synaptophysin, chromogranin, and CD56). Like ommended. When there is no question that the lesion
LCNEC, basaloid carcinoma shows nuclear palisading is malignant, adenocarcinoma and SQC are unlikely if
around the margins of cell groups. The rare lymphoep- there is no keratinization or mucinous or glandular dif-
ithelioma-like carcinoma contains interspersed lym- ferentiation, but because the entire tumor has not been
phoid cells. Clear cell carcinoma is comprised of large sampled, the possibility of an undersampled adenocar-
polygonal cells with abundant clear cytoplasm. The cells cinoma or SQC can never be entirely excluded. For this
of large cell carcinoma with rhabdoid features have reason, cytologic specimens with features of LCC are
large cytoplasmic globules. often resulted as “Non-small cell carcinoma.” One of the
A B
Figure 2.31 Large cell neuroendocrine carcinoma (LCNEC). A, Fine-needle aspiration (FNA). The key cytologic features include:
rosettes, prominent nucleoli, and carcinoid tumor-like nuclei with extreme atypia and enlargement. Mitoses (not seen) are fre-
quent. This constellation of findings is not always present, making distinction from non–small cell carcinoma sometimes impos-
sible. B, Lung, autopsy specimen, same patient as A. Note the organoid growth pattern, rosettes, and brisk mitotic activity in the
histologic material.
Lung Cancer 91
A B
Figure 2.32 Epithelioid angiosarcoma (EAS) of the lung. A, The large polygonal tumor cells resemble those of a large cell
carcinoma (LCC). B, Cell block sections show sheetlike growth.
sarcomatoid carcinomas should be considered if there is no differentiated spindle cell elements. Cytologic prep-
spindle or giant cell differentiation. arations show large malignant spindle cells with hyper-
EAS occurs as a primary tumor in the lung and other chromatic nuclei.
sites. EAS is a particularly good mimic of LCC because Giant cell carcinoma is composed of enormous,
it is a high-grade tumor, with large nuclei, prominent often multinucleated cells with no foci of adenocarci-
nucleoli, and brisk mitotic activity (Fig. 2.32). The diag- noma, SQC, or LCC. Cytologic preparations show dis-
nosis of EAS is established by demonstrating vascular persed, isolated, strikingly large, often multinucleated
differentiation with immunohistochemistry for CD31 or cells with round nuclei and prominent nucleoli (Fig.
CD34. About 30% are reactive for keratins, however, and 2.33). Neutrophils are often prominent. It is important to
thus a wider antibody panel is essential. recognize giant cell carcinoma because it is associated
Diffuse large B-cell lymphoma, anaplastic large cell with an aggressive clinical course.158
lymphoma, metastatic carcinoma, and melanoma can The diagnosis of a pleomorphic, spindle cell, or
usually be distinguished from LCC by an immunohisto- giant cell carcinoma can be suspected on a cytologic
chemical panel that includes CD20, CD3, ALK1, S-100, specimen, but precise classification depends on exten-
HMB45, keratins CK7 and CK20, and carcinoma-specific sive sampling and thus is best deferred to histologic
markers selected based on clinical correlation. examination.
The remaining two tumors are true biphasic neo-
plasms. Pulmonary blastoma, composed of primitive
Sarcomatoid Carcinoma glandular and stromal elements, is a malignant neo-
Sarcomatoid carcinoma is a family of poorly differenti- plasm named for its resemblance to fetal lung. It is
ated non–small cell carcinomas that show sarcomatoid slightly more common in males, and the mean age is in
or giant cell differentiation. The group includes pleo- the fourth decade.159 Histologically there is a spindle cell
morphic carcinoma, spindle cell carcinoma, giant cell component that can show myxoid, chondroid, osteoid,
carcinoma, carcinosarcoma, and pulmonary blastoma. or rhabdomyoblastic differentiation and an epithelial
These tumors are important to recognize because they component consisting of tubules of cuboidal to colum-
have a worse prognosis than the conventional non-small nar cells with frequent mitoses and subnuclear and
cell carcinomas. supranuclear vacuoles. The vacuoles contain glycogen
Pleomorphic carcinoma is defined histologically as a and impart a “piano key”–like, endometrioid appear-
poorly differentiated adenocarcinoma, SQC, or LCC, at ance to the epithelioid component. Solid nests of squa-
least 10% of which is a spindle or giant cell malignancy. moid cells (squamoid morulas) are sometimes present.
The spindle cell component shows no differentiated Tumors comprised of just the epithelial component of
sarcomatous elements. Cytologic preparations show, in the pulmonary blastoma are called fetal adenocarci-
addition to adenocarcinoma, SQC, or LCC, a population nomas.138 Pulmonary blastoma most likely arises from
of pleomorphic spindle or giant cells. Pleomorphic car- a totipotent precursor because identical p53 mutations
cinoma is often a large peripheral tumor with a tendency are present in both the epithelial and sarcomatous
to invade the chest wall. components within the same tumor.160 Immunohis
Spindle cell carcinoma is a non–small cell carcinoma tochemical demonstration of keratins in the epithelial
consisting only of spindle-shaped malignant cells with component and muscle-specific actin in the spindle
92 Respiratory Tract
Typical Carcinoid
At one end of the spectrum of neuroendocrine tumors
is the typical carcinoid (TC), which accounts for 2%
to 3% of all pulmonary tumors. Typical carcinoids are
uniformly distributed throughout the lungs. Centrally
located tumors are often submucosal, with a promi-
nent endobronchial component. When submucosal,
the typical carcinoid is usually covered by intact respira-
tory epithelium, and as a result sputum cytology is often
negative. TCs have a low metastatic rate: 10% to 15% of
patients have regional lymph node involvement,122,162
and 5% to 10% eventually metastasize to distant sites,122
but the prognosis is excellent with surgery, and 5-year
Figure 2.33 Giant cell carcinoma (fine-needle aspiration survival is 90% to 98%.122
[FNA]). The huge multinucleated cells with striking nuclear Histologic examination reveals uniform polygonal
atypia of this tumor are obviously malignant. cells with a variable growth pattern that may include
nests, ribbons, papillae, and rosette-like arrangements.
Occasionally the cells and nuclei are elongated (spindle
cells helps to confirm the diagnosis.161 Pitfalls in the cell carcinoid).163
cytologic diagnosis of blastoma occur when one of the
components is poorly represented, as when the stromal Cytomorphology of typical
component is misinterpreted as a small cell carcinoma. carcinoid:
Carcinosarcoma differs clinically from blastoma in
the mean age of patients at presentation (65 rather than • loosely cohesive groups and single cells
40 years). Morphologically, the spindle cell component • rosette-like structures
includes differentiated elements like malignant carti- • round, plasmacytoid, or elongated cells
lage, bone, or skeletal muscle. • uniform nuclei with “salt and pepper” chromatin
• ample granular cytoplasm
• branching capillaries
• mitoses uncommon
Pulmonary Neuroendocrine • no necrosis
Neoplasms
This group of neoplasms encompasses a morphologic
and biologic spectrum, divided by the World Health Cytologic preparations usually show a uniform pop-
Organization into four distinct diagnostic categories: ulation of isolated cells and loose clusters (Fig. 2.34).
typical carcinoid, atypical carcinoid, LCNEC, and small Large vascularized tissue fragments are sometimes
cell carcinoma.122 By electron microscopy, all contain present. TCs are vascular tumors, and sometimes a
cytoplasmic dense-core, membrane-bound granules in mesh of branching capillaries is encountered. In cell
variable quantities. Neuroendocrine tumors are immu- block sections, solid nests, trabeculae (ribbons), papil-
noreactive for one or more of the neuroendocrine mark- lae, and rosettes can be appreciated. The tumor cells
ers, chromogranin A (most specific), synaptophysin, and are round, oval (plasmacytoid), or elongated (spindle
Leu-7 (CD57). Adrenocorticotropic hormone (ACTH), cell carcinoid) and have moderate or abundant granu-
insulin, calcitonin, gastrin, and vasoactive intestinal lar cytoplasm. Nuclei are round or oval, with smooth
peptide are less consistently present. contours, finely speckled (salt and pepper) chromatin,
The critical feature that separates typical and atypi- and inconspicuous nucleoli. In some cases there may
cal carcinoids from LCNEC and small cell carcinoma be nuclear atypia and pleomorphism, but this has no
is the mitotic rate (Table 2.3). Even if a tumor has mor- clinical significance.
phologic features of small cell carcinoma, if it lacks fre-
quent mitoses or necrosis, it should not be diagnosed as Differential diagnosis of typical
small cell carcinoma. LCNEC was previously discussed carcinoid:
in greater detail. • benign bronchial epithelial cells
Typical and atypical carcinoids are usually positive • adenocarcinoma
for keratins, but up to 20% can be negative.122 Conflicting • small cell carcinoma/atypical carcinoid
results have been obtained with thyroid transcription • lymphoma
factor 1 (TTF-1): carcinoids are either negative, or they • mesenchymal tumors
can be positive in one third of cases.122
Lung Cancer 93
Image
*Cytologic features extrapolated from Travis WD, Linnoila RI, Tsokos MG, et al.: Neuroendocrine tumors of the lung with proposed criteria for large-cell
neuroendocrine carcinoma: An ultrastructural, immunohistochemical, and flow cytometric study of 35 cases. Am J Surg Pathol 1991;15:529–553, and
authors’ unpublished observations.
Atypical Carcinoid
In the hazy area between the typical carcinoid and small
cell carcinoma lies the atypical carcinoid (AT). It is more
aggressive than the typical carcinoid: the 5-year survival
rate is 61% to 73%.122
Cytomorphology of atypical
carcinoid:
• cells and architecture similar to typical carcinoid
• focal necrosis
• mitoses (moderate number)
• prominent nucleoli
Figure 2.34 Typical carcinoid (fine-needle aspiration [FNA]).
The tumor cells have a moderate amount of coarsely granular
cytoplasm and a “salt-and-pepper” chromatin pattern. Rosettes The cytomorphologic features of the neuroendocrine
are seen in some carcinoid tumors.
tumors are listed in Table 2.3. ATs share many of the mor-
phologic features of TCs (Fig. 2.35). They differ, however,
Because of their bland, monomorphous appearance, in several slight ways: the architectural arrangements may
the cells of the typical carcinoid may be mistaken for be looser; there is greater (but not brisk) mitotic activity;
benign bronchial cells. In contrast to bronchial cells, TC there may be focal necrosis.164 Although helpful, these fea-
cells are less cohesive and lack cilia. Because rosettes are tures do not always permit precise classification in cyto-
seen in some TCs, they are sometimes mistaken for an logic and bronchial biopsy samples. Precise classification
adenocarcinoma. In most adenocarcinomas, however, depends on thorough histologic sampling of a resected
there is greater variation in cell size and more group- specimen. In view of this, surgical excision should be con-
ing of cells into spheres and sheets than with carcinoid sidered for all localized neuroendocrine tumors.
tumors. Carcinoid tumors are distinguished from atyp-
ical carcinoids and small cell carcinomas by the lower Small Cell Carcinoma
mitotic rate and absence of necrosis (see Table 2.3).
There may be a resemblance to lymphoid cells, but lym- Small cell carcinoma (SMC) accounts for 20% to 25%
phoid cells have less cytoplasm and do not form clusters of all primary lung carcinomas, and 90% are centrally
or rosettes. located. The majority of patients are male smokers
94 Respiratory Tract
Figure 2.35 Atypical carcinoid (fine-needle aspiration Figure 2.36 Small cell carcinoma ([SMC]; bronchial brushing).
[FNA]). All the features of typical carcinoid can be seen, but The cells of SMC are tightly packed and show nuclear molding.
greater pleomorphism, slight nuclear enlargement, an increased They are fragile and occasionally degenerated. Well-preserved
number of mitoses, and focal necrosis are important distinguish- cells have a powdery chromatin texture.
ing elements.
(80%), and their prognosis is poor (5-year survival intermediate in size. Cytoplasm is scant and incon-
<10%).165 A morphologic variant is the “combined spicuous, and adjacent nuclei show frequent mold-
small cell carcinoma” (e.g., small cell/SQC, small ing (Fig. 2.36). The cytoplasm can contain occasional
cell/adenocarcinoma). paranuclear blue bodies, solitary round, light to dark
Histologically, SMC consists of small, round to fusi- blue, homogeneous spheres that indent the nucleus.166
form cells with scant cytoplasm. The tumor often under- Nuclei are round, oval, or stretched out (carrot-
mines the bronchial mucosa, with extensive invasion shaped). Chromatin is finely textured, and nucleoli
of lung parenchyma and lymphatics. Necrosis, a high are absent or uncommon. The cytologic appearance
mitotic rate, and nuclear encrustation of vascular walls varies somewhat with the sampling method. In spu-
are characteristic features. In some cases the cells are tum, streams of small hyperchromatic cells form loose
larger, with more cytoplasm and vesicular nuclei. aggregates. In bronchial brushings and FNA speci-
mens, the cells are better preserved.
rather than molded together; and they are one-half the include cough, dyspnea, and hemoptysis. Treatment is
size of SMC cells. surgical, but the long-term prognosis is poor, with 80% of
The distinction between an SMC and a TC is usually patients dead of disease within 20 years.169
straightforward, given that TC lack nuclear molding, These tumors are most often diagnosed by means of
necrosis, and mitotic activity. In addition, the chroma- bronchial brushings or transbronchial needle aspira-
tin texture of TC is more coarsely granular than that of tion. Cytologic preparations show cylinders or spheres
SMC. Some TC, like the spindle cell variant, however, of small, deceptively innocuous-appearing epithe-
bear a superficial resemblance to SMC. Therefore, care lial cells that surround basal lamina material (see Figs
must be taken not to make a diagnosis of SMC with- 10.20 to 22).
out abundant mitotic activity and necrosis. The distinc- Mucoepidermoid carcinomas of the trachea and
tion between TC and AC (see Fig. 2.35), and between bronchi are uncommon, representing only 0.2% of lung
AC and SMC, is more difficult; distinguishing features tumors. They are defined by the presence of squamous,
are summarized in Table 2.3. In general, ACs have more glandular, and intermediate cells, and by their origin
numerous mitoses than TCs, and SMCs have signifi- within cartilage-containing airways. The diagnosis “ade-
cantly more mitoses than either TCs or ACs, but a final nosquamous carcinoma” should be restricted to tumors
determination may not be possible without thorough arising at the lung periphery beyond a recognizable
histologic sampling. bronchus. Mucoepidermoid carcinomas can develop in
SMC is sometimes difficult to distinguish from the persons of all ages, including children. They are endo-
non–small cell carcinomas, particularly the small cell bronchial tumors like those of adenoid cystic carcinoma
variant of SQC and those adenocarcinomas composed and other salivary gland tumors, and the presenting
of relatively small cells. The most helpful features in symptoms are similar. Cytologically, the tumors are
making this distinction are the powdery chromatin tex- diagnosed by finding squamous cells, intermediate cells,
ture, inconspicuous nucleoli, prominent nuclear mold- and mucinous cells (see Figs 10.17, 10.18). High-grade
ing, and scant cytoplasm of SMC. Care should be taken tumors show marked nuclear atypia. Mucoepidermoid
not to misinterpret the solid paranuclear blue bod- carcinoma is another one of a growing number of neo-
ies as mucin vacuoles (which are clear), otherwise an plasms for which a recurrent and tumor-specific genetic
SMC might be misidentified as an adenocarcinoma.166 aberration is present: the t(11;19), forming the MECT1-
Immunohistochemistry for p63 can be quite helpful MAML2 fusion transcript.170
in distinguishing SQC, which is immunoreactive, from Other bronchial gland tumors include pleomorphic
SMC, which is negative.167,168 Even after careful examina- adenoma, acinic cell carcinoma, and oncocytoma.
tion and immunohistochemical study, not all tumors are
easily classified. In such cases, the possibility of a mixed
tumor, a common occurrence in the lungs, should be
Clear Cell Tumor (“Sugar Tumor”)
considered. The clear cell tumor is extremely rare and can occur in
Clinical history, particularly patient age, provides persons of all ages, most of whom are asymptomatic. It
clues to the diagnosis of the other small round cell is usually a peripheral mass and ranges from 1 to 7 cm
tumors. The sarcomatous component of pulmonary in greatest diameter. Clear cell tumors most likely arise
blastomas resembles a SMC; identifying the epithelial from perivascular epithelioid cells.122,171 Histologically
areas is crucial in making this distinction. and cytologically, the tumor consists of bland polygo-
nal and spindle-shaped cells with a central oval or elon-
gated nucleus and clear, glycogen-rich cytoplasm.172
Uncommon Pulmonary The differential diagnosis includes the clear cell vari-
Tumors ant of adenocarcinoma or SQC of the lung, granular cell
tumor, and metastatic tumors with clear-cell features
Adenoid Cystic Carcinoma and (renal cell carcinoma, adrenal cortical carcinoma, some
melanomas, and others).172,173 Immunohistochemistry
Other Bronchial Gland Tumors is helpful for most of these distinctions except with
The submucosal glands of the trachea and bronchi give melanoma: most clear cell tumors are positive for
rise to neoplasms that resemble those occurring in the HMB-45 and Melan-A and negative for cytokeratins
salivary glands. The most common of these is the adenoid and CEA.171,174
cystic carcinoma. This malignant tumor accounts for
20% to 35% of all cancers in the trachea but also occurs in
the major bronchi. The tumors grow as polypoid masses,
Sarcomas
undermining intact overlying respiratory epithelium. Primary malignant mesenchymal tumors of the lung
They often infiltrate the cartilage and soft tissues of the are rare, occurring in adults of either sex. They can be
major airways. Symptoms are related to obstruction, and large masses or small endobronchial lesions. One of the
96 Respiratory Tract
Granulomatous inflammatory processes need to be Metastases are diagnosed in sputum in up to 70% of cases.192
distinguished from Hodgkin lymphoma. Histiocytes About 40% of metastatic tumors in exfoliative specimens
can mimic Reed-Sternberg cells, and granulomas can are squamous carcinomas from an adjacent site in the
occur in Hodgkin lymphoma. Reed-Sternberg cells are aerodigestive tract. The remainder of the cases includes
distinguished from histiocytes because RS cells have adenocarcinomas (34%)—most commonly of the breast, kid-
large, inclusion-like nucleoli and highly irregular nuclei. ney, and colon—and hematopoietic malignancies (8%).193
Melanoma and LCC can be excluded with immunohisto- Because confirming metastatic disease is a major indi-
chemical studies for keratins, S-100, and HMB45. cation for transthoracic FNA, metastatic tumors account
About 10% of diffuse pulmonary infiltrates in patients for between 14% and 26% of transthoracic FNA speci-
with leukemia result from leukemic involvement of the mens.194 In a study of more than 1000 cases, malignant
lung. At autopsy, infiltrates are seen in 25% to 65% of melanoma accounted for 27%, urinary and male geni-
cases, most commonly acute leukemia. Leukemic involve- tal tract neoplasms for 17%, breast carcinoma for 15%,
ment of the lung may also present as a mass lesion. The female genital tract neoplasms for 13%, and gastrointes-
cells are dispersed as isolated cells, as with lympho- tinal tract neoplasms for 10%.193 Another study reported
mas, and cellular monomorphism is characteristic. that the most common metastases (in decreasing fre-
BAL is useful for demonstrating leukemic involvement, quency) are those from breast cancer, colon cancer, renal
especially when biopsy is contraindicated because of coag- cell carcinoma, bladder cancer, and melanoma.195
ulopathy.188 The diagnosis can be confirmed using immu- Primary sites for some metastatic carcinomas can be
nohistochemistry on cell block, cytospin preparations, or inferred based on their characteristic cytologic features.
by flow cytometry (i.e., blasts are commonly CD34+). The cells of the usual type of colon carcinoma have a dis-
Other hematopoietic malignancies that can involve tinctive tall, columnar “picket fence” appearance, with
the lung include myeloma189 and mycosis fungoides.190 hyperchromatic nuclei and “dirty” necrosis. Malignant
melanoma often demonstrates a single-cell arrangement,
with cytoplasmic pigment, intranuclear cytoplasmic
Metastatic Cancers To invaginations, and classic “mirror-image” binucleation.
The Lung Breast carcinoma often shows a single-file arrangement of
cells, some with mucin-containing intracellular lumens
The lungs receive the entire cardiac output, so it is not surpris- (Fig. 2.38); renal cell carcinoma cells are large cells with
ing that they are a common site for metastasis; 30% to 50% abundant clear cytoplasm and poorly defined cell mem-
of extrapulmonary cancers involve the lungs at autopsy.191 branes; and metastatic papillary thyroid carcinomas have
A
Figure 2.38 Metastatic ductal carcinoma of the breast. This patient had a previous history of ductal breast carcinoma. A, Fine-needle
aspiration (FNA) of the lung. The smear shows clusters of tumor cells with prominent nucleoli and mucin-filled intracellular lumens.
Illustration continued on following page
98 Respiratory Tract
B
Figure 2.38 (Continued) Metastatic ductal carcinoma of the breast. B, Breast lumpectomy. Comparison with the previously
resected tumor shows close similarity, supporting the diagnosis of metastasis to the lung from the breast primary.
nuclear grooves, nuclear pallor, and intranuclear pseu- cytopathologic diagnoses during the years 1970 to 1974. Acta
doinclusions, a morphologic appearance mimicked by Cytol 1982;26:759-766.
5. Wax TD, Coogan AC, Johnston WW: Clinical significance
the mucinous type of bronchioloalveolar carcinoma.
of an inconclusive cytopathologic diagnosis: A five-year
The distinction of primary lung carcinoma from a experience at Duke University Medical Center. II. Analysis and
metastasis is often facilitated by immunohistochem- follow-up of specimens from the respiratory tract. Acta Cytol
istry. Most lung carcinomas are reactive for CK7 and 1997;41(4):1053-1057.
nonreactive for CK20 (as contrasted with metastatic 6. Guidelines of the Papanicolaou Society of Cytopathology
for the examination of cytologic specimens obtained
colon cancer, for example, which is CK7- and CK20+).
from the respiratory tract. The Papanicolaou Society of
Additionally, lung carcinomas can be distinguished Cytopathology Task Force on Standards of Practice. Mod
from non–lung carcinomas, except those of thy- Pathol 1999;12(4):427-436.
roid origin, with antibodies to thyroid transcription 7. Khajotia RR, Mohn A, Pokieser L, et al.: Induced sputum
factor 1.196 Interestingly, whereas thyroid transcription and cytological diagnosis of lung cancer. Lancet
1991;338:976-977.
factor 1 was once thought to be simply a marker of lung
8. Saccomanno G, Saunders RP, Ellis H, et al.: Concentration
origin, there is now evidence that it is involved in the of carcinoma or atypical cells in sputum. Acta Cytol
pathogenesis of lung cancer: 12% of lung carcinomas 1963;7:305-310.
have amplification of thyroid transcription factor 1.197 9. Tang CS, Tang CM, Lau YY, Kung IT: Sensitivity of sputum
Determining the primary site of the metastasis (see Fig. cytology after homogenization with dithiothreitol in lung
cancer detection. Two years of experience. Acta Cytol
2.38) is greatly aided by comparing the patient’s cytologic
1995;39(6):1137-1140.
specimen with any previous pathologic material. It is par- 10. Böcking A, Biesterfeld S, Chatelain R, et al: Diagnosis of
ticularly important to obtain sufficient cytologic material bronchial carcinoma on sections of paraffin-embedded
for cell block preparations so that tissue architecture and sputum: Sensitivity and specificity of an alternative to routine
immunohistochemical markers can be assessed. cytology. Acta Cytol 1992;36:37-47.
11. Erozan YS, Frost JK: Cytopathologic diagnosis of cancer in
pulmonary material: A critical histopathologic correlation.
Acta Cytol 1970;14:560-565.
References 12. Fraire AE, Underwood RD, McLarty JW, Greenberg SD:
1. Johnston WW, Frable WJ: Diagnostic Respiratory Conventional respiratory cytology versus fine needle
Cytopathology. New York, Masson Publishing, 1979. aspiration cytology in the diagnosis of lung cancer. Acta Cytol
2. Linder J: Lung cancer cytology. Something old, something 1991;35(4):385-388.
new. Am J Clin Pathol 2000;114:169-171. 13. Rosenthal DL, (ed): Cytopathology of Pulmonary Disease.
3. Dahlgren S, Nordenstrom B: Transthoracic Needle Biopsy. Basel, S. Karger, 1988.
Chicago: Year Book Medical Publishers, 1966. 14. Sing A, Freudenberg N, Kortsik C, et al.: Comparison of the
4. Johnston WW: Ten years of respiratory cytopathology at Duke sensitivity of sputum and brush cytology in the diagnosis of
University Medical Center. III. The significance of inconclusive lung carcinomas. Acta Cytol 1997;41(2):399-408.
References 99
15. Suprun H, Pedio G, Ruttner JR: The diagnostic reliability of aspiration for sampling mediastinal lymph nodes. Thorax
cytologic typing in primary lung cancer with a review of the 2006;61(9):795-798.
literature. Acta Cytol 1980;24:494-500. 37. Yasufuku K, Chiyo M, Koh E, et al.: Endobronchial ultrasound
16. Liang XM: Accuracy of cytologic diagnosis and cytotyping of guided transbronchial needle aspiration for staging of lung
sputum in primary lung cancer: Analysis of 161 cases. J Surg cancer. Lung Cancer 2005;50(3):347-354.
Oncol 1989;40:107-111. 38. Harrow EM, Oldenburg FA, Smith AM: Transbronchial needle
17. Pue CA, Pacht ER: Complications of fiberoptic bronchoscopy aspiration to clinical practice. Thorax 1985;40:756-759.
at a university hospital. Chest 1995;107:430-432. 39. Baker JJ, Solanki PH, Schenk DA, et al.: Transbronchial
18. Popp W, Rauscher H, Ritschka L, et al.: Diagnostic sensitivity fine needle aspiration of the mediastinum: Importance of
of different techniques in the diagnosis of lung tumors with lymphocytes as an indicator of specimen adequacy. Acta
the flexible fiberoptic bronchoscope: Comparison of brush Cytol 1990;3:517-523.
biopsy, imprint cytology of forceps biopsy, and histology of 40. Rosenthal DL, Wallace JM: Fine needle aspiration of
forceps biopsy. Cancer 1991;67:72-75. pulmonary lesions via fiberoptic bronchoscopy. Acta Cytol
19. Kawan E, Ulrich W, Redtenbacher S, et al.: Kawan bronchial 1984;28:203-210.
brush/cell block technique. Facilitation of the routine diagnosis 41. Bhat N, Bhagat P, Pearlman ES, et al.: Transbronchial needle
of bronchial neoplasms. Acta Cytol 1998;42:1409-1413. aspiration biopsy in the diagnosis of pulmonary neoplasms.
20. Raab SS, Oweity T, Hughes JH, et al.: Effect of clinical history on Diagn Cytopathol 1990;6:14-17.
diagnostic accuracy in the cytologic interpretation of bronchial 42. Zakowski MF, Gatscha RM, Zaman MB: Negative predictive
brush specimens. Am J Clin Pathol 2000;114(1):78-83. value of pulmonary fine needle aspiration cytology. Acta Cytol
21. Johnston WW: Ten years of respiratory cytopathology at Duke 1992;6:283-286.
University Medical Center. 1: The cytopathologic diagnosis 43. Wagner ED, Ramzy I, Greenberg SD, Gonzalez JM:
of lung cancer during the years 1970 to 1974, noting the Transbronchial fine-needle aspiration: Reliability and
significance of specimen number and type. Acta Cytol limitations. Am J Clin Pathol 1989;92:36-41.
1981;25:103-107. 44. Rintoul RC, Skwarski KM, Murchison JT, et al.: Endobronchial
22. Mak VH, Johnston ID, Hetzel MR, Grubb C: Value of washings and endoscopic ultrasound-guided real-time fine-needle aspi
and brushings at fiberoptic bronchoscopy in the diagnosis of ration for mediastinal staging. Eur Respir J 2005;25(3):416-421.
lung cancer. Thorax 1990;45:373-376. 45. Singh P, Camazine B, Jadhav Y, et al.: Endoscopic ultrasound as a
23. Kahn FW, Jones JM: Diagnosing bacterial respiratory infection first test for diagnosis and staging of lung cancer: A prospective
by bronchoalveolar lavage. J Infect Dis 1987;155(5):862-869. study. Am J Respir Crit Care Med 2007;175(4):345-354.
24. Pisani RJ, Wright AJ: Clinical utility of bronchoalveolar 46. Eloubeidi MA, Tamhane A, Chen VK, Cerfolio RJ: Endoscopic
lavage in immunocompromised hosts. Mayo Clin Proc ultrasound-guided fine-needle aspiration in patients with non-
1992;67:221-227. small cell lung cancer and prior negative mediastinoscopy.
25. Broaddus C, Dake MD, Stulbarg MS, et al.: Bronchoalveolar Ann Thorac Surg 2005;80(4):1231123-9.
lavage and transbronchial biopsy for the diagnosis of 47. Herth F, Becker HD, Ernst A: Conventional vs endobronchial
pulmonary infections in the acquired immunodeficiency ultrasound-guided transbronchial needle aspiration: a
syndrome. Ann Intern Med 1985;102:747-752. randomized trial. Chest 2004;125(1):322-325.
26. Baughman RP, Dohn MN, Frame PT: The continuing utility of 48. Emery SC, Savides TJ, Behling CA: Utility of immediate
bronchoalveolar lavage to diagnose opportunistic infection in evaluation of endoscopic ultrasound-guided transesophageal
AIDS patients. Am J Med 1994;97(6):515-522. fine needle aspiration of mediastinal lymph nodes. Acta Cytol
27. Atzori C, Angeli E, Mainini A, et al.: In vitro activity of 2004;48(5):630-634.
human immunodeficiency virus protease inhibitors against 49. Fernandez-Esparrach G, Gines A, Belda J, et al.: Trans
Pneumocystis carinii. J Infect Dis 2000;181(5):1629-1634. esophageal ultrasound-guided fine needle aspiration
28. Piller CF, Clark DP: Pathologic outcome in HIV-seropositive improves mediastinal staging in patients with non-small
individuals with nonspecific bronchoalveolar lavage cytology. cell lung cancer and normal mediastinum on computed
Acta Cytol 1998;42:913-917. tomography. Lung Cancer 2006;54(1):35-40.
29. Pirozynski M: Bronchoalveolar lavage in the diagnosis of 50. Vilmann P, Krasnik M, Larsen SS, et al.: Transesophageal
peripheral, primary lung cancer. Chest 1992;102(2):372-374. endoscopic ultrasound-guided fine-needle aspiration (EUS-
30. Rennard SI: Bronchoalveolar lavage in the diagnosis of cancer. FNA) and endobronchial ultrasound-guided transbronchial
Lung 1990;168(Suppl):1035-1040. needle aspiration (EBUS-TBNA) biopsy: A combined
31. Semenzato G, Poletti V: Bronchoalveolar lavage in lung cancer. approach in the evaluation of mediastinal lesions. Endoscopy
Respiration 1992;9(Suppl 1):44-46. 2005;37(9):833-839.
32. Stanley MW, Henry-Stanley MJ, Gajl-Peczalska KJ, Bitterman 51. Westcott JL, Rao N, Colley DP: Transthoracic needle biopsy of
PB: Hyperplasia of type II pneumocytes in acute lung injury: small pulmonary nodules. Radiology 1997;202:97-103.
Cytologic findings of sequential bronchoalveolar lavage. Am J 52. Agarwal PK, Husain N, Singh BN: Cytologic findings in
Clin Pathol 1992;97:669-677. aspirated hydatid fluid. Acta Cytol 1989;33:652-654.
33. Gulbahce HE, Baker KS, Kumar P, et al.: Atypical cells in 53. Austin JHM, Cohen MB: Value of having a cytopathologist
bronchoalveolar lavage specimens from bone marrow present during percutaneous fine-needle aspiration biopsy
transplant recipients. A potential pitfall. Am J Clin Pathol of lung: Report of 55 cancer patients and metaanalysis of the
2003;120(1):101-106. literature. Am J Roent 1993;160:175-177.
34. Policarpio-Nicolas ML,Wick MR: False-positive interpretations 54. Sacchini V, Galimberti V, Marchini S, Luini A: Percutaneous
in respiratory cytopathology: Exemplary cases and literature transthoracic needle aspiration biopsy: A case report of
review. Diagn Cytopathol 2008;36(1):13-19. implantation metastasis. Eur J Surg Oncol 1989;15:179-183.
35. Herth FJ, Becker HD, Ernst A: Ultrasound-guided 55. Zarbo RJ, Fenoglio-Preiser CM: Interinstitutional database for
transbronchial needle aspiration: an experience in 242 comparison of performance in lung fine-needle aspiration
patients. Chest 2003;123(2):604-607. cytology: A College of American Pathologists Q-probe study
36. Herth FJ, Eberhardt R, Vilmann P, et al.: Real-time of 5264 cases with histologic correlation. Arch Pathol Lab Med
endobronchial ultrasound guided transbronchial needle 1992;116:463-470.
100 Respiratory Tract
56. Yazdi HM, MacDonald LL, Hickey NM: Thoracic fine needle clinical specimens by amplification of ribosomal RNA and its
aspiration biopsy versus fine needle cutting biopsy: A clinical application]. Kekkaku 1996;71(10):573-585.
comparative study of 40 patients. Acta Cytol 1988;32:635-640. 75. Qadri SM, Ali MA: Sensitivity of fine-needle aspiration
57. Arslan S, Yilmaz A, Bayramgurler B, et al.: CT-guided biopsy in the detection of mycobacterial infections. Diagn
transthoracic fine needle aspiration of pulmonary lesions: Cytopathol 1991;7:142-146.
accuracy and complications in 294 patients. Med Sci Monit 76. Das DK, Bhambhani S, Pant JN, et al.: Superficial and deep-
2002;8(7):CR493-497. seated tuberculosis lesions: Fine-needle aspiration cytology
58. Di Donna A, Bazzocchi M, Dolcet F, Springolo E: [CT-guided diagnosis of 547 cases. Diagn Cytopathol 1992;8:211-215.
transthoracic needle aspiration of solitary lung lesions. 77. DiTomasso JP, Ampel NM, Sobonya RE, Bloom JW:
Personal experience in 118 cases]. Radiol Med (Torino) Bronchoscopic diagnosis of pulmonary coccidioidomycosis.
1995;89(3):287-294. Comparison of cytology, culture, and transbronchial biopsy.
59. Hayes MM, Zhang DY, Brown W: Transthoracic fine-needle Diagn Microbiol Infect Dis 1994;18:83-87.
aspiration biopsy cytology of pulmonary neoplasms. Diagn 78. Raab SS, Silverman JF, Zimmerman KG: Fine-needle
Cytopathol 1994;10(4):315-319. aspiration biopsy of pulmonary coccidioidomycosis:
60. Arakawa H, NakajimaY, KuriharaY, et al.: CT-guided transthoracic Spectrum of cytologic findings in 73 patients. Am J Clin Pathol
needle biopsy: A comparison between automated biopsy gun 1993;99:582-587.
and fine needle aspiration. Clin Radiol 1996;51(7):503-506. 79. Iwama de Mattos MC, de Oliveira ML: Respiratory
61. Diacon AH, Theron J, Schubert P, et al.: Ultrasound-assisted cytopathology: Paracoccidioidomycosis associated either with
transthoracic biopsy: Fine-needle aspiration or cutting- tuberculosis or bronchogenic carcinoma. Diagn Cytopathol
needle biopsy? Eur Respir J 2007;29(2):357-362. 1992;8:198-199.
62. Greif J, Marmur S, Schwarz Y, et al.: Percutaneous core cutting 80. Iwama de Mattos MC, de Oliveira ML: Pseudoepitheliomatous
needle biopsy compared with fine-needle aspiration in the proliferation, a pitfall in sputum cytology. Diagn Cytopathol
diagnosis of peripheral lung malignant lesions: Results in 156 1991;7:656-657.
patients. Cancer 1998;84(3):144-147. 81. Iwama de Mattos MC, Mendes RP, Marcondes-
63. Cristallini EG, Ascani S, Farabi R, et al.: Fine needle aspiration Machado J, et al.: Sputum cytology in the diagnosis of
biopsy in the diagnosis of intrathoracic masses. Acta Cytol pulmonary paracoccidioidomycosis. Mycopathologica
1992;36:416-422. 1991;114:187-191.
64. Castella X, Valles J, Cabezuelo MA, et al.: Fat embolism 82. Farley ML, Fagan MF, Mabry LC, Wallace RJ, Jr: Presentation
syndrome and pulmonary microvascular cytology. Chest of Sporothrix schenkii in pulmonary cytology specimens. Acta
1992;101:1710-1711. Cytol 1991;35:389-395.
65. Abati A, Landucci D, Danner RL, Solomon D: Diagnosis of 83. Ness MJ, Rennard SI, Vaughn WP, et al.: Detection of
pulmonary microvascular metastases by cytologic evaluation Candida antigen in bronchoalveolar lavage fluid. Acta Cytol
of pulmonary artery catheter-derived blood specimens. Hum 1988;32:347-352.
Pathol 1994;25:257-262. 84. Edman JC, Kovacs JA, Masur H, et al.: Ribosomal RNA
66. Masson RG, Ruggieri J: Pulmonary microvascular cytology. A sequence shows Pneumocystis carinii to be a member of the
new diagnostic application of the pulmonary artery catheter. fungi. Nature 1988;334:519-522.
Chest 1985;88:908-914. 85. Ypma-Wong MF, Fonzi WA, Sypherd PS: Fungus-specific
67. Dhooge I, Joos G, Van Cauwenberge P: Value of rigid translation elongation factor 3 gene present in Pneumocystis
bronchoscopy and cytodiagnosis of bronchial washings carinii. Infect Immun 1992;60:4140-4145.
in detecting bronchial carcinoma in the presence of a 86. Hartmann B, Koss M, Hui A, et al.: Pneumocystis carinii
carcinoma of the upper aerodigestive tract. J Otolaryngol pneumonia in the acquired immunodeficiency syndrome
1996;25:219-222. (AIDS): Diagnosis with bronchial brushings, biopsy, and
68. Roggli VL, Piantadosi CA, Bell DY: Asbestos bodies in bronchoalveolar lavage. Chest 1985;87:603-607.
bronchoalveolar lavage fluid: A study of 20 asbestos-exposed 87. Maiese SC, Naryshkin S: Diagnosis of Pneumocystis carinii
individuals and comparison to patients with other chronic by cytologic examination of Papanicolaou-stained sputum
interstitial lung diseases. Acta Cytol 1986;30:470-476. specimens. Diagn Cytopathol 1991;7:111-112.
69. Strickler JG, Manivel JC, Copenhaver CM, KubicVL: Comparison 88. Naryshkin S, Daniels J, Freno E, Cunningham L: Cytology
of in situ hybridization and immunohistochemistry for of treated and minimal Pneumocystis carinii pneumonia
detection of cytomegalovirus and herpes simplex virus. Hum and a pitfall of the Grocott methenamine silver stain. Diagn
Pathol 1990;21(4):443-448. Cytopathol 1991;7:41-47.
70. Weiss RL, Snow GW, Schumann GB, Hammond ME: Diagnosis 89. Kovacs JA, Ng VL, Masur H, et al.: Diagnosis of Pneumocystis
of cytomegalovirus pneumonitis on bronchoalveolar lavage carinii pneumonia: improved detection in sputum with use of
fluid: Comparison of cytology, immunofluorescence, and monoclonal antibodies. N Engl J Med 1988;318:589-593.
in situ hybridization with viral isolation. Diagn Cytopathol 90. Midgley J, Parsons PA, Shanson DC, et al.: Monoclonal
1991;7:243-247. immunofluorescence compared with silver stain for
71. Weinberg A, Hodges TN, Li S, et al.: Comparison of PCR, investigating Pneumocystis carinii pneumonia. J Clin Pathol
antigenemia assay, and rapid blood culture for detection 1991;44:75-76.
and prevention of cytomegalovirus disease after lung 91. Hawkins AG, Hsiu JG, Smith RM3rd, et al.: Pulmonary
transplantation. J Clin Microbiol 2000;38(2):768-772. dirofilariasis diagnosed by fine needle aspiration biopsy: A
72. Harboldt SL, Dugan JM, Tronic BS: Cytologic diagnosis of case report. Acta Cytol 1985;29:19-22.
measles pneumonia in a bronchoalveolar lavage specimen. A 92. Ingram EA, Helikson MA: Echinococcosis (hydatid disease) in
case report. Acta Cytol 1994;38:403-406. Missouri: Diagnosis by fine-needle aspiration of a lung cyst.
73. Pierce CH, Knox AW: Ciliocytophthoria in sputum from Diagn Cytopathol 1991;7:527-531.
patients with adenovirus infections. Proc Soc Exp Biol Med 93. Aisner SC, Gupta PK, Frost JK: Sputum cytology in pulmonary
1960;104:492-495. sarcoidosis. Acta Cytol 1977;21:394-398.
74. Osumi M, Toyoda T, Kawashiro T, Aoyagi T: [Evaluation of a 94. Travis WD, Colby TV, Koss MN, et al.: Atlas of Nontumor
new detection method of Mycobacterium tuberculosis in Pathology: Non-neoplastic Disorders of the Lower Respiratory
references 101
Tract. Washington, DC, American Registry of Pathology and 115. Nadal A, Campo E, Pinto J, et al: p53 expression in normal,
the Armed Forces Institute of Pathology, 2002. dysplastic, and neoplastic laryngeal epithelium. Absence of a
95. Kitamura T, Tanaka N, Watanabe J, et al: Idiopathic correlation with prognostic factors. J Pathol 1995;175:181-188.
pulmonary alveolar proteinosis as an autoimmune disease 116. Rusch V, Klimstra D, Linkov I, Dmitrovsky E: Aberrant
with neutralizing antibody against granulocyte/macrophage expression of p53 or the epidermal growth factor receptor
colony-stimulating factor. J Exp Med 1999;190(6):875-880. is frequent in early bronchial neoplasia and coexpression
96. Uchida K, Beck DC, Yamamoto T, et al GM-CSF autoantibodies precedes squamous cell carcinoma development. Cancer Res
and neutrophil dysfunction in pulmonary alveolar proteinosis. 1995;55:1365-1372.
N Engl J Med 2007;356(6):567-579. 117. Vine MF, Schoenbach VJ, Hulka BS, et al: Atypical metaplasia
97. Beskow CO, Drachenberg CB, Bourquin PM, et al: Diffuse and incidence of bronchogenic carcinoma. Am J Epidemiol
alveolar damage. Morphologic features in bronchoalveolar 1990;131:781-793.
lavage fluid. Acta Cytol 2000;44:640-646. 118. Onuki N, Wistuba, II, Travis WD, et al: Genetic changes
98. Hyakudo T, Yoshii C, Nikaido Y, et al: A case of bronchiolitis in the spectrum of neuroendocrine lung tumors. Cancer
obliterans organizing pneumonia: Diagnostic utility of 1999;85:600-607.
bronchoalveolar lavage and CT scan. Nihon Kyobu Shikkan 119. Lam S, Becker HD: Future diagnostic procedures. Chest Surg
Gakkai Zasshi 1994;32:768-773. Clin N Am 1996;6:363-380.
99. Majori M, Poletti V, Curti A, et al: Bronchoalveolar lavage 120. Hammond WG, Teplitz RL, Benfield JR: Variable regression of
in bronchiolitis obliterans organizing pneumonia primed experimental bronchial preneoplasia during carcinogenesis. J
by radiation therapy to the breast. J Allergy Clin Immunol Thorac Cardiovasc Surg 1991;101:800-806.
2000;105:239-244. 121. Jemal A, Siegel R, Ward E, et al: Cancer statistics, 2007. CA
100. Poletti V, Cazzato S, Minicuci N, et al: The diagnostic Cancer J Clin 2007;57(1):43-66.
value of bronchoalveolar lavage and transbronchial lung 122. Travis WD, Brambilla E, Muller-Hermelink HK, Harris CC
biopsy in cryptogenic organizing pneumonia. Eur Respir J (eds): World Health Organization Classification of Tumours.
1996;9:2513-2516. Pathology and Genetics of Tumours of the Lung, Pleura,
101. Tiroke AH, Bewig B, Haverich A: Bronchoalveolar lavage Thymus and Heart. Lyon, IARC Press, 2004.
in lung transplantation. State of the art. Clin Transplant 123. Ahrendt SA, Chow JT, Xu LH, et al.: Molecular detection of
1999;13:131-157. tumor cells in bronchoalveolar lavage fluid from patients with
102. Walts AE, Marchevsky AM, Morgan M: Pulmonary cytology early stage lung cancer. J Natl Cancer Inst 1999;91:332-339.
in lung transplant recipients: Recent trends in laboratory 124. Somers VA, Pietersen AM, Theunissen PH, Thunnissen FB:
utilization. Diagn Cytopathol 1991;7:353-358. Detection of K-ras point mutations in sputum from patients
103. Minoletti F, Sozzi G, Calderone C, et al: Variant translocation with adenocarcinoma of the lung by point-EXACCT. J Clin
t (6;10)(p21;q22) in pulmonary chondroid hamartoma. Genes Oncol 1998;16:3061-3068.
Chromosomes Cancer 1996;15:246-248. 125. Hussain SP, Harris CC: p53 mutation spectrum and load: The
104. Xiao S, Lux ML, Reeves R, et al: HMGI(Y) activation generation of hypotheses linking the exposure of endogenous
by chromosome 6p21 rearrangements in multilineage or exogenous carcinogens to human cancer. Mutat Res
mesenchymal cells from pulmonary hamartoma. Am J Pathol 1999;428:23-32.
1997;150:901-910. 126. Fontana RS, Sanderson DR, Woolner LB, et al.: Lung cancer
105. Dunbar F, Leiman G: The aspiration cytology of pulmonary screening: The Mayo program. J Occup Med 1986;28:746-750.
hamartomas. Diagn Cytopathol 1989;5:174-180. 127. McGee JMC: Screening for lung cancer. Semin Surg Oncol
106. Hughes JH, Young NA, Wilbur DC, et al: Fine-needle aspiration 1989;5:179-185.
of pulmonary hamartoma: A common source of false- 128. Swensen SJ, Jett JR, Sloan JA, et al: Screening for lung cancer
positive diagnoses in the College of American Pathologists with low-dose spiral computed tomography. Am J Respir Crit
Interlaboratory Comparison Program in Nongynecologic Care Med 2002;165(4):508-513.
Cytology. Arch Pathol Lab Med 2005;129(1):19-22. 129. Diederich S, Thomas M, Semik M, et al.: Screening for early
107. Meis-Kindblom JM, Kjellstrom C, Kindblom LG: Inflammatory lung cancer with low-dose spiral computed tomography:
fibrosarcoma: Update, reappraisal, and perspective on its Results of annual follow-up examinations in asymptomatic
place in the spectrum of inflammatory myofibroblastic smokers. Eur Radiol 2004;14(4):691-702.
tumors. Semin Diagn Pathol 1998;15:133-143. 130. Patz EF Jr, Swensen SJ, Herndon JE 2nd. Estimate of lung
108. Lawrence B, Perez-Atayde A, Hibbard MK, et al: TPM3-ALK cancer mortality from low-dose spiral computed tomography
and TPM4-ALK oncogenes in inflammatory myofibroblastic screening trials: implications for current mass screening
tumors. Am J Pathol 2000;157:377-384. recommendations. J Clin Oncol 2004;22(11):2202-2206.
109. Bridge JA, Kanamori M, Ma Z, et al: Fusion of the ALK gene 131. Sone S, Li F, Yang ZG, et al.: Results of three-year mass
to the clathrin heavy chain gene, CLTC, in inflammatory screening programme for lung cancer using mobile low-
myofibroblastic tumor. Am J Pathol 2001;159(2):411-415. dose spiral computed tomography scanner. Br J Cancer
110. Debiec-Rychter M, Marynen P, Hagemeijer A, Pauwels P: ALK- 2001;84(1):25-32.
ATIC fusion in urinary bladder inflammatory myofibroblastic 132. Sone S, Nakayama T, Honda T, et al.: Long-term follow-
tumor. Genes Chromosomes Cancer 2003;38(2):187-190. up study of a population-based 1996–1998 mass screening
111. Usuda K, Saito Y, Imai T, et al: Inflammatory pseudotumor of programme for lung cancer using mobile low-dose spiral
the lung diagnosed as granulomatous lesion by preoperative computed tomography. Lung Cancer 2007;58(3):329-341.
brushing cytology: A case report. Acta Cytol 1990;34:685-689. 133. Jett JR: Limitations of screening for lung cancer with low-dose
112. Füzesi L, Höer PW, Schmidt W: Exfoliative cytology of spiral computed tomography. Clin Cancer Res 2005;11(13 Pt 2):
multiple endobronchial granular cell tumor. Acta Cytol 4988s-4992s.
1989;33:516-518. 134. Manser R, Dalton A, Carter R, et al.: Cost-effectiveness
113. Mermolja M, Rott T: Cytology of endobronchial granular cell analysis of screening for lung cancer with low dose spiral
tumor. Diagn Cytopathol 1991;7:524-526. CT (computed tomography) in the Australian setting. Lung
114. Coissard CJ, Besson G, Polette MC, et al: Prevalence of human Cancer 2005;48(2):171-185.
papillomaviruses in lung carcinomas: a study of 218 cases. 135. Pastorino U: Early detection of lung cancer. Respiration
Mod Pathol 2005;18(12):1606-1609. 2006;73(1):5-13.
102 Respiratory Tract
136. Li R, Todd NW, Qiu Q, et al.: Genetic deletions in sputum as Diagnosis by fine needle aspiration biopsy with histologic and
diagnostic markers for early detection of stage I non-small cell ultrastructural confirmation. Diagn Cytopathol 1985;1:205-215.
lung cancer. Clin Cancer Res 2007;13(2 Pt 1):482-487. 154. Ohori NP, Santa Maria EL: Cytopathologic diagnosis of
137. Kersting M, Friedl C, Kraus A, et al.: Differential frequencies of bronchioloalveolar carcinoma: does it correlate with the
p16(INK4a) promoter hypermethylation, p53 mutation, and 1999 World Health Organization definition? Am J Clin Pathol
K-ras mutation in exfoliative material mark the development 2004;122(1):44-50.
of lung cancer in symptomatic chronic smokers. J Clin Oncol 155. Mhoyan A, Weidner N, Shabaik A: Epithelioid hemangio
2000;18(18):3221-3229. endothelioma of the lung diagnosed by transesophageal
138. Proctor L, Folpe AL, Esper A, et al.: Well-differentiated fetal endoscopic ultrasound-guided fine needle aspiration: A case
adenocarcinoma of the lung: Cytomorphologic features on fine- report. Acta Cytol 2004;48(4):555-559.
needle aspiration with emphasis on use of beta-catenin as a 156. Travis WD, Linnoila RI, Tsokos MG, et al: Neuroendocrine
useful diagnostic marker. Diagn Cytopathol 2007;35(1):39-42. tumors of the lung with proposed criteria for large-cell
139. Paez JG, Janne PA, Lee JC, et al.: EGFR mutations in lung neuroendocrine carcinoma: An ultrastructural, immuno
cancer: correlation with clinical response to gefitinib therapy. histochemical, and flow cytometric study of 35 cases. Am J
Science 2004;304(5676):1497-1500. Surg Pathol 1991;15:529-553.
140. Marchetti A, Martella C, Felicioni L, et al.: EGFR mutations 157. Hammond ME, Sause WT: Large cell neuroendocrine
in non-small-cell lung cancer: Analysis of a large series of tumors of the lung: Clinical significance and histopathologic
cases and development of a rapid and sensitive method definition. Cancer 1985;56:1624-1629.
for diagnostic screening with potential implications on 158. Iwasaki Y: Large cell carcinoma. Nippon Rinsho
pharmacologic treatment. J Clin Oncol 2005;23(4):857-865. 2000;58:1127-1131.
141. Haneda H, Sasaki H, Lindeman N, et al: A correlation between 159. Koss MN, Hochholzer L, O’Leary T: Pulmonary blastomas.
EGFR gene mutation status and bronchioloalveolar carcinoma Cancer 1991;67:2368-2381.
features in Japanese patients with adenocarcinoma. Jpn J Clin 160. Holst VA, Finkelstein S, Colby TV, et al.: p53 and K-ras
Oncol 2006;36(2):69-75. mutational genotyping in pulmonary carcinosarcoma, spindle
142. Chen YR, Fu YN, Lin CH, et al: Distinctive activation patterns cell carcinoma, and pulmonary blastoma: Implications for
in constitutively active and gefitinib-sensitive EGFR mutants. histogenesis. Am J Surg Pathol 1997;21:801-811.
Oncogene 2006;25(8):1205-1215. 161. Cosgrove MM, Chandrasoma PT, Martin SE: Diagnosis of
143. Okabe T, Okamoto I, Tamura K, et al: Differential constitutive pulmonary blastoma by fine-needle aspiration biopsy:
activation of the epidermal growth factor receptor in non- cytologic and immunocytochemical findings. Diagn Cytopathol
small cell lung cancer cells bearing EGFR gene mutation and 1991;7:83-87.
amplification. Cancer Res 2007;67(5):2046-2053. 162. Warren WM, Faber LP, Gould VE: Neuroendocrine neoplasms
144. Greulich H, Chen TH, Feng W, et al: Oncogenic transformation of the lung: a clinicopathologic update. J Thorac Cardiovasc
by inhibitor-sensitive and -resistant EGFR mutants. PLoS Med Surg 1989;98:321-332.
2005;2(11):e313. 163. Fekete PS, Cohen C, DeRose PB: Pulmonary spindle cell carcinoid:
145. Lynch TJ, Bell DW, Sordella R, et al: Activating mutations in the Needle aspiration biopsy, histologic and immunohistochemical
epidermal growth factor receptor underlying responsiveness findings. Acta Cytol 1990;34:50–56.
of non-small-cell lung cancer to gefitinib. N Engl J Med 164. Jordan AG, Predmore L, Sullivan MM, Memoli VA: The
2004;350(21):2129-2139. cytodiagnosis of well-differentiated neuroendocrine carcinoma:
146. Inoue A, Suzuki T, Fukuhara T, et al: Prospective phase II A distinct clinicopathologic entity. Acta Cytol 1987;31:464–470.
study of gefitinib for chemotherapy-naive patients with 165. Travis WD, Rush W, Flieder DB, et al.: Survival analysis of
advanced non-small-cell lung cancer with epidermal 200 pulmonary neuroendocrine tumors with clarification of
growth factor receptor gene mutations. J Clin Oncol criteria for atypical carcinoid and its separation from typical
2006;24(21):3340-3346. carcinoid. Am J Surg Pathol 1998;22(8):934–944.
147. Jackman DM, Yeap BY, Lindeman NI, et al: Phase II clinical trial 166. Renshaw AA, Voytek TM, Haja J, Wilbur DC: Distinguishing
of chemotherapy-naive patients > or = 70 years of age treated small cell carcinoma from non-small cell carcinoma of the
with erlotinib for advanced non-small-cell lung cancer. J Clin lung: correlating cytologic features and performance in the
Oncol 2007;25(7):760-766. College of American Pathologists Non-Gynecologic Cytology
148. Engelman JA, Zejnullahu K, Mitsudomi T, et al: MET ampli Program. Arch Pathol Lab Med 2005;129(5):619–623.
fication leads to gefitinib resistance in lung cancer by acti 167. Kalhor N, Zander DS, Liu J: TTF-1 and p63 for distinguishing
vating ERBB3 signaling. Science 2007;316(5827):1039-1043. pulmonary small-cell carcinoma from poorly differentiated
149. Riely GJ, Kris MG, Zhao B, et al: Prospective assessment of squamous cell carcinoma in previously pap-stained cytologic
discontinuation and reinitiation of erlotinib or gefitinib in material. Mod Pathol 2006;19(8):1117–1123.
patients with acquired resistance to erlotinib or gefitinib 168. Wu M, Wang B, Gil J, et al.: p63 and TTF-1 immunostaining.
followed by the addition of everolimus. Clin Cancer Res A useful marker panel for distinguishing small cell carcinoma
2007;13(17):5150-5155. of lung from poorly differentiated squamous cell carcinoma
150. Smouse J, Cibas E, Janne P, et al: Cytology specimens compare of lung. Am J Clin Pathol 2003;119(5):696–702.
favorably with surgical specimens for EGFR mutation detection 169. Dalton ML, Gatling RR: Peripheral adenoid cystic carcinoma
in patients with NSCLC. Mod Pathol 2007;20 (Suppl 2):85A. of the lung. South Med J 1990;83:577–579.
151. Horiike A, Kimura H, Nishio K, et al: Detection of epidermal 170. Tonon G, Modi S, Wu L, et al: t(11;19)(q21;p13) translocation
growth factor receptor mutation in transbronchial in mucoepidermoid carcinoma creates a novel fusion
needle aspirates of non-small cell lung cancer. Chest product that disrupts a Notch signaling pathway. Nat Genet
2007;131(6):1628-1634. 2003;33(2):208–213.
152. Chen KT: Psammoma bodies in fine-needle aspiration 171. Lantuejoul S, Isaac S, Pinel N, et al.: Clear cell tumor of the lung:
cytology of papillary adenocarcinoma of the lung. Diagn An immunohistochemical and ultrastructural study supporting
Cytopathol 1990;6:271-274. a pericytic differentiation. Mod Pathol 1997;10:1001–1008.
153. Silverman JF, Finley JL, Park HK, et al: Psammoma bodies and 172. Policarpio-Nicolas ML, Covell J, Bregman S, Atkins K: Fine
optically clear nuclei in bronchioloalveolar cell carcinoma: needle aspiration cytology of clear cell “sugar” tumor
references 103
(PEComa) of the lung: Report of a case. Diagn Cytopathol pulmonary mucosa associated lymphoid tissue lymphomas.
2008;36(2):89–93. Mod Pathol 1998;11:525–532.
173. Husain M, Nguyen GK: Cytopathology of granular-cell tumor 186. Sprague RI, deBlois GG: Small lymphocytic pulmonary
of the lung. Diagn Cytopathol 2000;23(4):294–295. lymphoma: diagnosis by transthoracic fine needle aspiration.
174. Gal AA, Koss MN, Hochholzer L, Chejfec G: An Chest 1989;96:929–930.
immunohistochemical study of benign clear cell (“sugar”) 187. Reale FR, Variakojis D, Compton J, Bibbo M: Cytodiagnosis
tumor of the lung. Arch Pathol Lab Med 1991;115:1034–1038. of Hodgkin’s disease in sputum specimens. Acta Cytol
175. Fleming WH, Jove DF: Primary leiomyosarcoma of the lung 1983;27:258–261.
with positive sputum cytology. Acta Cytol 1975;19:14–20. 188. Rossi GA, Balbi B, Risso M, et al: Acute myelomonocytic
176. Moran CA, Suster S: Tumors of the lung and pleura. In leukemia: Demonstration of pulmonary involvement by
Fletcher C (ed): Diagnostic Histopathology of Tumors, 3rd ed. bronchoalveolar lavage. Chest 1985;87:259–260.
New York, Churchill Livingstone, 2007, pp. 181–214. 189. Riazmontazer N, Bedayat G: Cytology of plasma cell myeloma
177. Stanfield BL, Powers CN, Desch CE, et al: Fine-needle aspiration in bronchial washing. Acta Cytol 1989;33:519–522.
cytology of an unusual primary lung tumor, chondrosarcoma: 190. Rosen SE, Vonderheid EC, Koprowska I: Mycosis fungoides
Case report. Diagn Cytopathol 1991;7:423–426. with pulmonary involvement: cytopathologic findings. Acta
178. Nguyen GK: Exfoliative cytology of angiosarcoma of the Cytol 1984;28:51–57.
pulmonary artery. Acta Cytol 1985;29:624–627. 191. Rosenblatt MB, Lisa JR, Trinidad S: Pitfalls in the clinical and
179. Crosby JH, Hoeg K, Hager B: Transthoracic fine needle histologic diagnosis of bronchogenic carcinoma. Dis Chest
aspiration of primary and metastatic sarcomas. Diagn 1966;49:396–404.
Cytopathol 1985;1:221–227. 192. Kern WH, Schweizer CW: Sputum cytology of metastatic
180. Kim K, Naylor B, Han IH: Fine needle aspiration carcinoma of the lung. Acta Cytol 1976;20:514–520.
cytology of sarcomas metastatic to the lung. Acta Cytol 193. Johnston WW: Cytologic correlations. In Dail K, Hammer
1986;30(6):688–694. S(eds): Pulmonary Pathology. New York, Springer-Verlag, 1987,
181. Flieder DB, Yousem SA: Pulmonary lymphomas and lymphoid pp. 1058.
hyperplasias. In Knowles D(ed) Neoplastic Hematopathology. 194. Johnston WW: Percutaneous fine needle aspiration biopsy of
Baltimore, Lippincott, Williams & Wilkins, 2001, pp. 1263–1301. the lung: a study of 1,015 patients. Acta Cytol 1984;28:218–224.
182. Habermann TM, Ryu JH, Inwards DJ, Kurtin PJ: Primary 195. Zaman MB, Hajdu SI, Melamed MR, Watson RC: Transthoracic
pulmonary lymphoma. Semin Oncol 1999;26:307–315. aspiration cytology of pulmonary lesions. Semin Diagn Pathol
183. Pisani RJ, DeRemee RA: Clinical implications of the 1986;3:176–187.
histopathologic diagnosis of pulmonary lymphomatoid 196. Holzinger A, Dingle S, Bejarano PA, et al: Monoclonal
granulomatosis. Mayo Clin Proc 1990;65:151–163. antibody to thyroid transcription factor-1: Production,
184. Myers JL, Kurtin PJ, Katzenstein AL, et al: Lymphomatoid characterization, and usefulness in tumor diagnosis.
granulomatosis. Evidence of immunophenotypic diversity Hybridoma 1996;15:49–53.
and relationship to Epstein-Barr virus infection. Am J Surg 197. Weir BA, Woo MS, Getz Get al.: Characterizing the cancer
Pathol 1995;19:1300–1312. genome in lung adenocarcinoma. Nature 2007;
185. Zaer FS, Braylan RC, Zander DS, et al: Multiparametric flow 450(7171):893–898.
cytometry in the diagnosis and characterization of low-grade