ECMO-Extracorporeal+Life+Support+in+Adul (Estudio)

ECMOExtracorporeal Life
Support in Adults
Fabio Sangalli
Nicol Patroniti
Antonio Pesenti
Editors
123
ECMO-Extracorporeal Life Support in Adults
[email protected]
[email protected]
Fabio Sangalli Nicol Patroniti

Antonio Pesenti
Editors
ECMO-Extracorporeal Life
Support in Adults
[email protected]
Editors
Fabio Sangalli
Department of Anaesthesia
and Intensive Care Medicine
San Gerardo Hospital
Monza (MB)
Antonio Pesenti
Health Science Department
Universit Milano Bicocca Facolt
Medicina e Chirurgia
Monza (MB)
Italy
Italy
Nicol Patroniti
Health Sciences Department,
Urgency and Emergency Department
Milano-Bicocca University
San Gerardo Hospital
Monza (MB)
Italy
ISBN 978-88-470-5426-4
ISBN 978-88-470-5427-1 (eBook)
DOI 10.1007/978-88-470-5427-1
Springer Milan Heidelberg New York Dordrecht London
Library of Congress Control Number: 2014934677
Springer-Verlag Italia 2014
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed. Exempted from this legal reservation are brief excerpts in connection
with reviews or scholarly analysis or material supplied specifically for the purpose of being entered and
executed on a computer system, for exclusive use by the purchaser of the work. Duplication of this
publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher's
location, in its current version, and permission for use must always be obtained from Springer.
Permissions for use may be obtained through RightsLink at the Copyright Clearance Center. Violations
are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for
any errors or omissions that may be made. The publisher makes no warranty, express or implied, with
respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)
[email protected]
Foreword
The best way to temporarily support or substitute vital organs is based on the availability of reliable and effective tools able to vicariate the failing natural organ. This
opportunity was achieved long ago for the kidney and, later on, for the heart and
lung. The technological improvement that miniaturized the apparatus improved the
vascular access, increased the performance of the artificial support, and has allowed
to expand the use of circulatory and respiratory extracorporeal support to several
clinical situations and to different ICUs (cardiac, respiratory, general). The advent
of new fulminant diseases (H1N1 respiratory failure) and the improvement of outof-hospital care for cardiac arrest are two situations that recently have seen extracorporeal support as a possible life-saving application. In order to correctly use the new
technologies, a specific competency and skills should be developed and implemented: as it happens for the achievement of positive results in the ICU setting, the
entire team (perfusionists, nurses, and doctors) has to be trained and should have
specific knowledge of the new technologies. Moreover, in this time where the adequate allocation of resources appears to be very important, it is mandatory that the
indications for the use of expensive and long-lasting techniques should be accurately weighed and shared among professionals.
The aim of this book is to provide readers with the theory and practical issues
that experts in the field of extracorporeal circulatory and respiratory support believe
could help in understanding and improving the practice of this medical device.
Milan, Italy
Roberto Fumagalli
[email protected]
[email protected]
Preface
Extracorporeal membrane oxygenation (ECMO) is not a new technique. It has been

used in clinical practice for the last four decades, but the complexity of management
and the relevant complications limited its diffusion to few specialized centers.
In recent years, the development of new materials and the simplification of the
procedure led to a dramatic increase in the centers providing extracorporeal life support (ECLS) and in the number of ECMO runs, for both respiratory and circulatory
indications.
A growing number of publications on all aspects pertaining to ECLS are populating the medical literature.
Despite this expansion in the use of ECLS and in ECLS-related research, the
clinical management of ECMO remains mainly based on local protocols and procedures, and guidelines are lacking on many aspects of this practice. The ELSO
(Extracorporeal Life Support Organization) registry and website, together with their
so-called Red Book, represent the most authoritative resource, and many websites
provide protocols and management guidelines from different ECMO centers. Still,
such indications are mainly locally based or not regularly updated.
For this reason we tried to collate the most relevant aspects pertaining to ECLS,
following two different approaches. Some chapters present an in-depth analysis of
the current evidence and literature on the different indications, while other chapters
face technical aspects with a more practical approach. These latter chapters are
obviously influenced by the practice in the authors centers, but we tried to integrate
this with literature and different experiences whenever possible, particularly for the
aspects where centers attitudes diverge, such as left ventricle venting, cannulation
techniques, and management of the lung during respiratory support, to name some.
ECLS remains a fast-evolving technique and some aspects still need research
and optimization. Some of these are outlined in the conclusive chapter of the book,
but more are still to be faced. Ample bibliographic references are provided at the
end of every chapter for the interested reader to further explore specific aspects.
ECLS represents a relatively easy technique, but it is not simply a procedure to
be learned and performed. ECMO is an excellent tool for organ support, but it
requires sound physiologic and pathophysiologic knowledge and needs to be combined with top-level standard care.
vii
[email protected]
viii
Preface
We are aware that, as a first edition, the readers will find aspects of the book that
might be improved, and we will welcome any suggestion in this regard. We still
hope that the present work will be useful in disseminating ECLS knowledge and
stimulate further study and research.
Fabio Sangalli, Nicol Patroniti, Antonio Pesenti, Monza (MB), Italy
[email protected]
Contents
Part I
History and Technical Aspects
History of Extracorporeal Life Support . . . . . . . . . . . . . . . . . . . . . . .

Fabio Sangalli, Chiara Marzorati, and Nerlep K. Rana
Developing a New ECMO Program . . . . . . . . . . . . . . . . . . . . . . . . . .

Antonio F. Arcadipane and Giovanna Panarello
11
Basic Aspects of Physiology During ECMO Support . . . . . . . . . . . .

Vittorio Scaravilli, Alberto Zanella, Fabio Sangalli,
and Nicol Patroniti
19
Percutaneous Cannulation: Indication, Technique,

and Complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Maurizio Migliari, Roberto Marcolin, Leonello Avalli,
and Michela Bombino
37
Surgical Cannulation: Indication, Technique, and Complications .

Francesco Formica, Silvia Mariani, and Giovanni Paolini
49
Materials: Cannulas, Pumps, Oxygenators . . . . . . . . . . . . . . . . . . . .

Umberto Borrelli and Cristina Costa
65
Coagulation, Anticoagulation, and Inflammatory Response . . . . . .

Marco Ranucci
77
Part II
8
ECMO for Circulatory Support
Extracorporeal Life Support: Interactions

with Normal Circulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Michele G. Mondino, Filippo Milazzo, Roberto Paino,
and Roberto Fumagalli
ECMO for Ischemic Cardiogenic Shock. . . . . . . . . . . . . . . . . . . . . . .
Francesco Formica, Fabio Sangalli, and Antonio Pesenti
93
105
ix
[email protected]
Contents
10
ECMO for Refractory Cardiac Arrest . . . . . . . . . . . . . . . . . . . . . . . .

Leonello Avalli, Margherita Scanziani, Elena Maggioni,
and Fabio Sangalli
117
11
ECMO for Postcardiotomic Shock . . . . . . . . . . . . . . . . . . . . . . . . . . .

Massimo Baiocchi, Fabio Caramelli, and Guido Frascaroli
127
12
ECMO in Myocarditis and Rare Cardiomyopathies . . . . . . . . . . . . .

Barbara Cortinovis, Monica Scanziani,
and Simona Celotti
137
13
ECMO for High-Risk Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Fabio Ramponi, Paul Forrest, John F. Fraser, Korana Musicki,
and Michael P. Vallely
151
14
ECMO for Severe Accidental Hypothermia . . . . . . . . . . . . . . . . . . . .

Peter Mair and Elfriede Ruttmann
163
15
ECMO in Drug Intoxication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Piergiorgio Bruno, Piero Farina, and Massimo Massetti
171
16
Newer Indications for ECMO: Pulmonary Embolism,

Pulmonary Hypertension, Septic Shock and Trauma . . . . . . . . . . . .
Michela Bombino, Sara Redaelli, and Antonio Pesenti
179
17
Left Ventricular Rest and Unloading During VA ECMO . . . . . . . . .

Gianluca Greco, Barbara Cortinovis, and Leonello Avalli
193
18
Weaning from Extracorporeal Circulatory Support . . . . . . . . . . . . .

Anna Coppo, Lucia Galbiati, and Gianluigi Redaelli
207
19
Treatment Options for End-Stage Cardiac Failure . . . . . . . . . . . . . .

Gurmeet Singh
217
Part III
ECMO for Respiratory Support
20
Ventilatory Management of ARDS Before and During ECMO . . . .

Giacomo Bellani, Giacomo Grasselli, and Antonio Pesenti
239
21
Respiratory Monitoring of the ECMO Patient . . . . . . . . . . . . . . . . .

Alberto Zanella, Francesco Mojoli, Luigi Castagna,
and Nicol Patroniti
249
22
Structure of an ECMO Network for Respiratory Support . . . . . . . .

Maria Grazia Calabr, Federico Pappalardo,
and Alberto Zangrillo
265
23
ECMO and Thoracic Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Alia Noorani and Alain Vuylsteke
273
24
ECMO in the Awake/Extubated Patient . . . . . . . . . . . . . . . . . . . . . . .

Giorgio A. Iotti, Francesco Mojoli, and Mirko Belliato
281
[email protected]
Contents
xi
25
ECMO as a Bridge to Lung Transplant . . . . . . . . . . . . . . . . . . . . . . .

Stefania Crotti and Alfredo Lissoni
293
26
Low-Flow ECMO and CO2 Removal . . . . . . . . . . . . . . . . . . . . . . . . .

Vito Fanelli, Andrea Costamagna, Pierpaolo P. Terragni,
and V. Marco Ranieri
303
27
Weaning from VV ECMO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Giacomo Grasselli, Paolo Mangili, Simone Sosio,
and Nicol Patroniti
317
Part IV
ECMO for Organ Procurement
28
Heart-Beating and Non-Heart-Beating Donors . . . . . . . . . . . . . . . . .

Marinella Zanierato, Francesco Mojoli, and Antonio Braschi
327
29
Lung Reconditioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Franco Valenza, Jacopo Fumagalli, Valentina Salice,
and Luciano Gattinoni
337
Part V
Monitoring the ECMO Patient
30
Patient Care During ECMO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Michela Bombino, Sara Redaelli, and Nicol Patroniti
345
31
Echocardiography in Venoarterial and Venovenous ECMO . . . . . .

Nicola Bianco, Leonello Avalli, and Fabio Sangalli
361
32
Haemodynamic Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fabio Guarracino and Rubia Baldassarri
375
33
Respiratory Monitoring During VA ECMO. . . . . . . . . . . . . . . . . . . .

Daniela Pasero, Pietro Persico, Tommaso Tenaglia,
and Vito Marco Ranieri
383
34
Neurological Monitoring During ECMO . . . . . . . . . . . . . . . . . . . . . .

Paolo Zanatta, Enrico Bosco, Alessandro Forti,
Elvio Polesel, and Carlo Sorbara
389
35
Monitoring the ECMO Patient: The Extracorporeal Circuit . . . . . .

Stefano Isgr, Francesco Mojoli, and Leonello Avalli
401
Part VI
36
Complications of ECMO
Complications of Extracorporeal Support

and Their Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Antonio Rubino, Richard Haddon, Fabrizio Corti,
and Fabio Sangalli
[email protected]
415
xii
37
Contents
Troubleshooting Common and Less Common Problems . . . . . . . . .

Lisen Hockings and Alain Vuylsteke
Part VII
425
Transport of the ECMO Patient
38
Air Transport: Fixed-Wing and Helicopter . . . . . . . . . . . . . . . . . . . .

Antonio F. Arcadipane and Gennaro Martucci
445
39
Ground Transport: Ambulance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Stefano Isgr, Roberto Rona, and Nicol Patroniti
455
Part VIII
40
Conclusion
Newer Indications and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . .

Marco Giani, Alberto Zanella, Fabio Sangalli, and Antonio Pesenti
463
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
473
[email protected]
Part I
History and Technical Aspects
[email protected]
History of Extracorporeal Life Support

Fabio Sangalli, Chiara Marzorati, and Nerlep K. Rana
ECMO (extracorporeal membrane oxygenation), also called ECLS (extracorporeal

life support), in its actual application is an evolution of the heartlung machines
used in cardiac surgery. Depending on its configuration venovenous or venoarterial it is used to support respiratory function, circulation, or both. This treatment
provides a bridge, either to healing of the natural organs or to long-term devices or
transplantation. In fact, although ECMO has the capability to support cardiorespiratory function temporarily, it is not a cure for the underlying disease. As Warren
Zapol, one of the fathers of respiratory ECMO, pinpointed in an editorial in the New
England Journal of Medicine in 1972, the goal of ECLS is to buy time while
sustaining an adequate tissue perfusion [1].
Despite the fact that the origins of ECLS stem from cardiac surgery and the
heartlung machine, its main applications at least until recent years and most
of the related research were carried out in the setting of severe respiratory
failure.
Artificial oxygenation is a theme that has always fascinated scientists since the
beginning of modern medicine.
The first attempt to artificially oxygenate blood in an extracorporeal circulation
was achieved in 1869 by Ludwig and Schmidt, by shaking together defibrinated
blood with air in a balloon [2]. Next step was reached 10 years later, by artificially
perfusing for the first time an isolated kidney using the first simple bubble oxygenator. In the same year, Frey and Gruber described the first two-dimensional,
F. Sangalli (*) C. Marzorati

Department of Anaesthesia and Intensive Care Medicine, San Gerardo Hospital,
University of Milano-Bicocca, Via Pergolesi 33, Monza 20900, Italy
e-mail: [email protected]; [email protected]
N.K. Rana
Anaesthesiology and Critical Care Department, Citt della Salute e della Scienza,
Ospedale S. Giovanni Battista-Molinette, Corso Bramante, 88, Turin 10126, Italy
e-mail: [email protected]
F. Sangalli et al. (eds.), ECMO-Extracorporeal Life Support in Adults,
DOI 10.1007/978-88-470-5427-1_1, Springer-Verlag Italia 2014
[email protected]
F. Sangalli et al.
direct-contact extracorporeal oxygenator that exposed a thin film of blood to air

in an inclined cylinder that was rotated at a frequency of 30/min by an electric
motor [37].
Several bubble- and surface-type oxygenators were developed in the first two
decades of the twentieth century. The main problems that hampered the development of the technique were thrombosis and hemolysis [8]. The turning point was the
discovery of heparin by Jay Maclean, in 1916. This led to overcome most of the
problems due to the contact of blood with air and the resulting prothrombotic activation [3, 4, 7].
The first whole-body extracorporeal perfusion was realized on a dog in 1929 by
Brukhonenko and Tchetchuline in Russia [3, 7, 911].
Between 1930 and 1953 three important oxygenators were developed; they
paved the way to apply the technique to men:
The film oxygenator developed by Gibbon between 1937 and 1953 consisted of
a stationary screen oxygenator [1215] made up of a series of six to eight wire
mesh screens arranged vertically and in parallel in a plastic container down
which the blood flowed, forming a stable film that was exposed to a flow of oxygen [5, 11]. Kirklin et al. [1619], at the Mayo Clinic in Rochester, Minnesota,
further developed the Gibbon-type stationary screen oxygenator into the MayoGibbon pump-oxygenator apparatus.
The rotating disc oxygenator was described in 1948 by Bjork. It was further
modified for clinical use by several scientists and improved with the development of materials.
The bubble oxygenator was described in 1952 by Clarke, Gollan, and Gupta. They
reported that although small bubbles with their large surface area to volume ratio
favored oxygen uptake, they were less buoyant. This means that smaller bubbles
are less likely to rise spontaneously to the surface and are more likely to remain in
suspension air embolism is therefore more likely. An optimum balance has therefore to be obtained. This optimum is believed to exist if the bubbles are between
2 mm and 7 mm in diameter. Alternatively, a mixture of small and big bubbles may
be used. This oxygenator was subsequently modified and improved until the
DeWall oxygenator, a sequential bubble oxygenator, i.e., its components (bubbler, defoamer, reservoir, and pump), are arranged linearly in series [7, 20].
The first successful extracorporeal cardiopulmonary bypass was performed in
1953, by the surgeon John Gibbon. In 1954, Gibbon described how the heartlung
machine could be used, in case of emergency, to support respiratory and circulatory
activities. This theoretical intuition clashed with the practical impossibility of
extending the duration of extracorporeal circulation over 6 h. This was mainly due
to the cellular damage caused by the direct exposure of blood to gas. Interposing a
gas exchange membrane between the blood and the gas flow solved most of this
problem, and with this technological innovation the machine became more effective, allowing to perform ECMO for longer periods.
The first successful use of prolonged life support with a heartlung machine was
conducted by J. Donald Hill in 1971. The patient was 24 years old affected by posttraumatic ARDS, who was supported with ECMO during the acute phase of his
[email protected]
Fig. 1.1 The first successful ECMO patient
pathology, for 3 days. The patient was eventually weaned from ECLS and survived
(Fig. 1.1) [21].
This success was of fundamental importance for the subsequent development
and spread of ECMO. In the same period, ICUs were developing and hemodialysis
was introduced for the treatment of acute renal failure. ARDS remained a fundamental issue for critically ill patients, and the ECMO success was a hope for a
definitive solution to this problem: thanks to that treatment physicians could allow
the functional recovery of the damaged lung. The interest linked to ECLS treatment
was especially about its effectiveness as a respiratory support. This led to the creation of the name ECMO (extracorporeal membrane oxygenation), which emphasized the aspect of artificial oxygenation.
In 1975 Bartlett successfully treated with ECMO the first newborn, a baby
called Esperanza. The success of this case led to a great enthusiasm, and in the following years a lot of other patients, both pediatric and adults, were effectively
treated with ECMO [22]. In 1974 the Lung Division of the National Heart and
Lung Institute started a large multicenter trial to test ECMO versus conventional
therapies in acute respiratory failure. The results were disappointing, with just
10 % survival in both groups and no significant difference between ECMO and
conventional therapy [23].
The results of the NIH trial led to a diminished attention to ECMO, but a few
centers continued improving the technique (Fig. 1.2).
In 1978 Kolobow and Gattinoni introduced a modified extracorporeal gas
exchange technique, called extracorporeal carbon dioxide removal (ECCO2R). The
[email protected]
F. Sangalli et al.
Fig. 1.2 VV ECMO in Monza, early 1990s
rationale of this technique was to reduce CO2 to decrease ventilation to the minimum
necessary to recruit alveoli. The new ECMO was performed at low extracorporeal
blood flows (2030 % of cardiac output), so that a venovenous bypass technique
instead of a venoarterial one sufficed, which turned out to be less detrimental to
blood cells, coagulation, and internal organs. Using LFPPVECCO2-R, Gattinoni
et al. reported survival rates of up to 49 %. In the following years several centers
corroborated the promising survival rates of around 50 % and higher [14, 2427].
The need for a coordination between ECMO centers led to the foundation in
1989 in New Orleans of ELSO (Extracorporeal Life Support Organization), a free
community of clinicians and researchers, with the aim to collect data from the
ECMO centers on a unique database and to standardize the procedures.
The evolution of venovenous and venoarterial ECMO diverged over time, with
VV ECMO consolidating its primary role in respiratory support and VA ECMO
assuming an increasing role in the advanced management of circulatory failure.
1.1
VV ECMO
After a period of disgrace, mainly due to the relevant complications and to the
appearance on the scene of new promising and apparently less invasive strategies, namely, inhaled nitric oxide and prone positioning, VV ECMO was subject of
[email protected]
a renewed interest after the publication of CESAR Trial [15]. This is a multicenter
study comparing conventional therapies to VV ECMO support in ARDS. Results
showed a higher survival and less disability at 6 months in the ECMO group.
Moreover, although not the primary outcome, an actual difference in survival of
around 25 % was observed for patients considered for ECMO treatment at 28 days,
the primary outcome of most ARDS literature.
Even if what this trial actually demonstrated was the importance of centralization of severe ARDS patients to a specialized center, this gave a great thrust to
research, and in the following years the final explosion of the application of this
extracorporeal support was due to the use of ECMO as a rescue therapy in Australia
and New Zealand during the H1N1 influenza pandemic, proving its power in
hypoxemic emergencies [28]. The results obtained during this pandemic, more
than any randomized trial, led to the worldwide acceptance of the use of membrane
lungs.
This led to the creation of ARDS Network, a clinical network initiated by the
National Heart, Lung, and Blood Institute, National Institutes of Health, developed
in order to carry out multicenter trials of ARDS treatment.
Similar experiences, with excellent results both from a clinical and an organizational point of view, were realized in Italy [29] as well as in many other countries [30].
1.2
VA ECMO
Although VA ECMO was originally applied for respiratory support, its main application is nowadays as a circulatory support. In this setting, VA ECMO was employed
almost exclusively as a support for postcardiotomic cardiogenic shock until recent
years.
In the past few decades, VA ECMO gained a place out of the operating theater to
become an advanced treatment for cardiogenic shock. As you will read in the following chapters, it is nowadays widely employed as a circulatory support for cardiogenic shock of any etiology. Its ease of application, which makes it possible to
institute the extracorporeal support virtually anywhere, and the relatively low costs
made it an appealing alternative to other mechanical circulatory support systems,
especially in the emergency setting.
Another emergency application where ECMO gained a pivotal role as a unique
option is that of refractory cardiac arrest. In selected populations, ECMO demonstrated an advantage in survival and neurological outcome in patients with an
expected mortality approaching 100 % [18].
The development of miniaturized systems and more biocompatible circuits
made it possible to bring ECMO everywhere in the hospital, to retrieve patients
from hospitals without ECMO facilities (Fig. 1.3) or even out of the hospital [19,
31]. This was simply unimaginable just two decades ago, as Fig. 1.3 demonstrates
clearly.
[email protected]
F. Sangalli et al.
Fig. 1.3 Retrieval of an acute cardiogenic shock patient from a peripheral hospital
[email protected]
1.3
Conclusion
The technological evolution and new directions expand every day the potential of
ECLS.
The relatively short history of ECMO is dotted with great discoveries and forward leaps and hampered with disillusions, but for sure most of this story has yet
to be written!
References
1. Zapol WM, Kitz RJ (1972) Buying time with artificial lungs. N Engl J Med 286:657658
2. Ludwig C, Schmidt A (1868) Das Verhalten der Gase, Welche mit dem Blut durch die reizbaren Sugethiermuskelz strmen. Leipzig Berichte 20:1272
3. Rendell-Baker L (1963) History of thoracic anaesthesia. In: Mushin WW (ed) Thoracic anaesthesia. Blackwell Scientific Publications, Oxford, pp 598661
4. Wylie WD, Churchill-Davidson HC (1972) A practice of anaesthesia, 3rd edn. Lloyd-Luke,
London, pp 691715
5. Hewitt RL, Creech O Jr (1966) History of the pump oxygenator. Arch Surg 93:680696
6. von Frey M, Gruber M (1885) Studies on metabolism of isolated organs. A respirationapparatus for isolated organs. Untersuchungenuber den stoffwechsel Isolierter organe. Ein
respirations-apparat fur isolierte organe [in German]. Virchows Archiv Physiol 9:519532
7. Lim MW (2006) The history of extracorporeal oxygenators. Anaesthesia 61:984995
8. Kirklin JW, Theye RA, Patrick RT (1958) The stationary vertical screen oxygenator. In: Allen
JG (ed) Extracorporeal circulation. Thesis. Charles C Thomas, Springfield, pp 5766
9. Lee LH, Krumhaar D, Fonkolsrud EW, Schjeide OA, Maloney JV (1961) Denaturation of
plasma proteins as a cause of morbidity and death after intracardiac operations. Surgery
50:2937
10. Probert WR, Melrose DG (1960) An early Russian heart-lung machine. Br Med J
1:10471048
11. Brukhonenko S (1929) Circulation artificielle du sang dans lorganisme entire dun chin avec
Coeur exclu. J Physiol Pathol Gen 27:251272
12. Brukhonenko S, Tchetchuline S (1929) Experiences avec la tete isolee du chien. J Physiol
Pathol Gen 27:3179
13. Miller BJ, Gibbon JH, Fineburg C (1953) An improved mechanical heart and lung apparatus;
its use during open cardiotomy in experimental animals. Med Clin North Am 1:16031624
14. Gattinoni L, Pesenti A, Mascheroni D, Marcolin R, Fumagalli R, Rossi F, Iapichino G,
Romagnoli G, Uziel L, Agostoni A (1986) Low-frequency positive pressure ventilation with
extracorporeal CO2 removal in severe acute respiratory failure. JAMA 256:881886
15. Peek GJ, Mugford M, Tiruvoipati R, Wilson A, Allen E, Thalanany MM, Hibbert CL, Truesdale
A, Clemens F, Cooper N, Firmin RK, Elbourne D (2009) Efficacy and economic assessment of
conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult
respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet 374:13511363
16. Gibbon JH Jr (1954) Application of a mechanical heart and lung apparatus to cardiac surgery.
Minn Med 37:171185
17. Jones RE, Donald DE, Swan JC, Harshbarger HG, Kirklin JW, Wood EH (1955) Apparatus of
the Gibbon type for mechanical bypass of the heart and lungs; preliminary report. Proc Staff
Meet Mayo Clin 30:105113
18. Avalli L, Maggioni E, Formica F, Redaelli G, Migliari M, Scanziani M, Celotti S, Coppo A,
Caruso R, Ristagno G, Fumagalli R (2012) Favourable survival of in-hospital compared to
out-of-hospital refractory cardiac arrest patients treated with extracorporeal membrane oxygenation: an Italian tertiary care centre experience. Resuscitation 83:579583
[email protected]
10
F. Sangalli et al.
19. Arlt M, Philipp A, Voekel S, Camboni D, Rupprecht L, Graf BM, Schmid C, Hilker M (2011)
Hand-held minimized extracorporeal membrane oxygenation: a new bridge to recovery in
patients with out-of-centre cardiogenic shock. Eur J Cardiothorac Surg 40:689694
20. Iwahashi H, Yuri K, Nos K (2004) Development of the oxygenator: past, present and future.
J Artif Organs 7:111120
21. Hill JD, OBrien TG, Murray JJ, Dontigny L, Bramson ML, Osborn JJ, Gerbode F (1972)
Extracorporeal oxygenation for acute post-traumatic respiratory failure (shock-lung syndrome): use of the Bramson Membrane Lung. N Engl J Med 286:629634
22. Bartlett RH, Gazzaniga AB, Jefferies R, Huxtable RF, Haiduc NJ, Fong SW (1976)
Extracorporeal membrane oxygenation (ECMO) cardiopulmonary support in infancy. Trans
Am Soc Artif Intern Organs 22:8088
23. Lewandowski K, Metz J, Deutschmann C, Preiss H, Kuhlen R, Artigas A, Falke KJ (1995)
Incidence, severity, and mortality of acute respiratory failure in Berlin, Germany. Am J Respir
Crit Care Med 151:11211125
24. Kolobow T, Gattinoni L, Tomlinson T, White D, Pierce J, Iapichino G (1977) The carbon
dioxide membrane lung (CDML): a new concept. Trans Am Soc Artif Intern Organs
23:1721
25. Kolobow T, Gattinoni L, Tomlinson TA, Pierce JE (1977) Control of breathing using an extracorporeal membrane lung. Anesthesiology 46:138141
26. Gattinoni L, Pesenti A (2005) The concept of baby lung. Intensive Care Med 31:776784
27. Gattinoni L, Agostoni A, Pesenti A, Pelizzola A, Rossi GP, Langer M, Vesconi S, Uziel L, Fox
U, Longoni F, Kolobow T, Damia G (1980) Treatment of acute respiratory failure with lowfrequency positive-pressure ventilation and extracorporeal removal of CO2. Lancet
2:292294
28. The Australia and New Zealand Extracorporeal Membrane Oxygenation (ANZ ECMO)
Influenza investigators (2009) Extracorporeal membrane oxygenation for 2009 influenza
A(H1N1) acute respiratory distress syndrome. JAMA 302:18881895
29. Patroniti N, Zangrillo A, Pappalardo F, Peris A, Cianchi G, Braschi A, Iotti GA, Arcadipane
A, Panarello G, Ranieri VM, Terragni P, Antonelli M, Gattinoni L, Oleari F, Pesenti A (2011)
The Italian ECMO network experience during the 2009 influenza A(H1N1) pandemic: preparation for severe respiratory emergency outbreaks. Intensive Care Med 37:14471457
30. Zangrillo A, Biondi-Zoccai G, Landoni G, Frati G, Patroniti N, Pesenti A, Pappalardo F (2013)
Extracorporeal membrane oxygenation (ECMO) in patients with H1N1 influenza infection: a
systematic review and meta-analysis including 8 studies and 266 patients receiving ECMO.
Crit Care 17:R30
31. Lebreton G, Pozzi M, Luyt CE, Chastre J, Carli P, Pavie A, Leprince P, Vivien B (2012) Outof-hospital extra-corporeal life support implantation during refractory cardiac arrest in a halfmarathon runner. Resuscitation 82:12391242
[email protected]
Developing a New ECMO Program

Antonio F. Arcadipane and Giovanna Panarello
Since the first successful use of an artificial heart/lung apparatus by John Gibbon the
extracorporeal circulation technique has been optimized, and its applicability
expanded to multiple clinical settings, in recent decades, extracorporeal membrane
oxygenation has become the first line of mechanical circulatory support for cases of
severe cardiopulmonary failure not responsive to conventional therapy.
The diffusion of this complex technology in clinical practice and the good results
in terms of morbidity and mortality explain the desire that many hospitals feel to
exploit this treatment, though awareness of the technical skills and clinical competencies required for proper management of such an invasive and high-risk treatment
has limited its application. Only highly specialized centers equipped with specific
infrastructural characteristics, knowledge, experience, and organizational models
are suited to make use of extracorporeal circulation.
Since the H1N1 pandemic influenza in 2009, and following the publication of the
CESAR [1] trials results and the Anzic [2] study, the medical community has felt
the need to increase the availability and the number of centers specialized in extracorporeal circulation.
This chapter is intended for health-care givers already expert in intensive care
and willing to set up an ECMO program.
The Extracorporeal Life Support Organization (ELSO [3], an international organization founded in 1989) has published a list of recommendations and requirements that a center should satisfy in order to be recognized as suitable for managing
extracorporeal support. These guidelines are reviewed and updated every 3 years to
keep pace with the continuous improvement in technique and scientific understanding. All of the 240 international centers adhering to ELSO are required to meet the
standards of ELSOs prerequisites.
A.F. Arcadipane, MD (*) G. Panarello, MD

Department of Anesthesia and Critical Care, ISMETT (Mediterranean Institute for
Transplantation and Advanced Specialized Therapies), Via Tricomi 5, Palermo 90127, Italy
[email protected]
11
12
A.F. Arcadipane and G. Panarello
Though this treatment is burdened by intrinsic risks, the morbidity and mortality rates can be contained in centers with specific management protocols, careful selection of candidates for extracorporeal circulation support, and
latest-generation technology. A learning curve is unavoidable, but certain technical, clinical, and scientific standards should be met before implementing an
ECMO program.
The support of already experienced centers with recognized competence in the
field is an essential aid, and their assistance during the learning phase, when sharing
decision-making and programs may ensure better results, is crucial.
Specific steps can be identified in the set up of an ECLS program, and all the
passages must be fully analyzed to obtain the best results before the program starts.
2.1
Organization
Ideally, an ECMO center should be located in a tertiary hospital where all ventilation modes and/or rescue therapies can be guaranteed. There should also be availability of rapid consultation by a wide range of specialists, which is often necessary
for critical patients. According to the ELSO guidelines, the regionalization of a
referring system, with predefined centers covering precise geographic areas, is
advisable. Regionalization can have several advantages: from an organizational
standpoint, it facilitates coordination of activity within a geographic area; from a
clinical standpoint, it allows concentration of patient volume in specialized centers
in order to guarantee at least six cases a year, the minimum recognized as sufficient
for maintaining clinical expertise and better outcomes [4, 8, 9]. The relation among
outcome, regionalization, and high-volume programs is even stronger for lowvolume procedures performed in high-risk patients (such as ECMO candidates),
and though the relation between outcome, ECMO, and population volume has never
been formally addressed, we can deduce from studies done in highly specialized
adult and pediatric ICU cases that centralization is an effective tool for optimizing
results and costs.
2.2
Planning
Defining the scope of the program, and the role of the center, in the local health-care
system is the first step in clarifying not only the duties but principally the limits of
this highly complex clinical activity.
An exhaustive plan starts with an assessment of needs, which means verifying
the requests from the medical community and defining which tools (human and
instrumental) the new center should rely on to satisfy the request.
Assessment of needs consists of:
1. Identification of the manageable patient population
2. Identification of the personnel necessary to run the project
3. Evaluation of required equipment
4. Identification of financial support
[email protected]
2 Developing a New ECMO Program
2.3
13
Manageable Patient Population
The demand for the new center should be measured considering currently unmet
needs and the potentially increasing request for treatment as soon as the project
starts. A further consideration is proximity of potential referring hospitals. This is
essential for better defining the volume of patients the referral center might be asked
to respond to. Patients already managed by the referral center can also be the beneficiaries of the new program, though it is possible that an entirely new population of
patients should be included.
At the outset of the program, a center may be not ready to manage all subtypes of
extracorporeal support and all classes of patients. Age, disease requiring ECMO support,
and already consolidated expertise should drive the starting choices. A lack of neonatal/
pediatric expertise should not preclude the development of an adult ECMO service, and
a cardiac surgery center, for example, could start by offering only cardiac ECMO support
for respiratory cases. A wise starting point might be to begin with a select group of
patients suffering from a specific disease with more predictable outcome, and only later
expand the program to include patients affected with more complex clinical conditions,
and with a higher risk of complications and less predictable outcomes [5, 6].
2.4
Identification of Personnel
In setting up a new ECMO program, a steering group must be identified. The components of the steering group are both medical and administrative personnel with
responsibility of:
Identifying the programs purpose
Setting up the program
Identifying achievable results and defining performance indicators, ideally compared with benchmarks of similar centers
Implementing the program
Defining a business plan for predicting expenses and potential revenues, not only
monetary (QUALY adjusted) [11].
2.5
Staff
For an ECMO program to be developed, a dedicated team, led by a coordinator,

must be available daily for 24-h coverage. Supportive personnel is important: consultants and rehabilitation specialists are of extreme value in meeting the needs of
ECMO patients, both during and after the Extracorporeal circulatory support.
2.5.1
Coordinator
At least one ECMO coordinator (the ideal number of leaders will depend on the
volume of ECMO service activity) should be designated. Part of this responsibility
[email protected]
14
is the identification of skilled personnel, the selection of equipment instrumentation,

the organization of periodic and repeated in-services with technical and scientific
updates, verification of expertise and competency, and organization of daily operational activity.
An additional duty of the coordinator is to oversee the drafting of protocols
addressing the following topics:
Indications and contraindications
Clinical management of patients during ECMO, including weaning and
decannulation
Maintenance of equipment and updates in technologies
Interruption of ECMO support
Follow-up of patients after decannulation
2.5.2
Team
An ECMO team should be staffed by intensive care physicians and intensive care
nurses with working knowledge of management of ECMO patients. Some centers
will be able to include cardiothoracic surgeons and perfusionists in the ECMO
team, though this is not mandatory.
The multidisciplinary composition of the team, with the constant presence of a
cardiothoracic surgeon, is certainly an added value, though the absence of such a condition will not preclude development of an ECMO service if rapid consultation by a
cardiovascular surgical service is ensured so that vascular-hemorrhagic complications
can be immediately addressed. The final composition of the team must be based on
the scope of the ECMO service. If the plan is to offer extracorporeal cardiopulmonary
resuscitation (ECPR) and/or VAD support as a bridge to transplantation along with
the ECMO support, cardiac surgery expertise must be included in the ECMO staff [7].
Experience in percutaneous, large-bore vascular access placement and extracorporeal circulation management is a required competency in the core group of an
ECMO team. Similarly, technical skills in managing emergency troubleshooting
(clinical and/or instrumental) are also requisites.
Physicians selected to be components of an ECMO team must have vast critical
care experience and the working knowledge required for proper clinical management
of patients suffering from end-stage organ disease. This means robust clinical and
scientific training and knowledge of respiratory and cardiac failure in order to guarantee the most appropriate patient management, particularly in the period immediately preceding extracorporeal support placement, when candidates are more fragile
and major disabilities and irreversible organ damage can occur. ELSO has not produced specific recommendations to define the expertise level required for ECMO
specialists, thus leaving each center the autonomy to define competence. Each component of the team can have different roles and discrete autonomy in activity according to a recognized competency. An inclination for teamwork, a multidisciplinary
approach, and, principally, the ability to transfer know-how are also essential.
[email protected]
15
The final composition of an ECMO team is not defined by any clear-cut indications, but is generally the result of organizational and financial consideration specific to each center. As already mentioned, a perfusionist may not always be
involved. In Europe and Australia, critical care nurses with additional training in
extracorporeal circulation play a central role. However, the ideal condition for rapid
implementation of an ECMO service and optimal resource management may be
found in post cardiac surgery intensive care, where medical and paramedical staff
already possess the necessary knowledge to ensure a successful program of extracorporeal support.
Bedside care is not based on a fixed model and depends mainly on staff organization and the volume of patients, so that in some circumstances, it may become
necessary to have a dual-provider model to ensure full-time exclusive supervision of
both the patient and the ECMO circuits.
The staff selected for in-house management of patients on extracorporeal circulatory support may not always have the necessary technical skills and/or resources
for interhospital transfer, so it may be necessary to rely on the collaboration of other
centers.
2.5.3
Supportive Personnel
Consultants who are expert in a wide range of specialties outside the intensive care
unit may become necessary while managing critical patients on extracorporeal circulatory support. Being located in a tertiary intensive care unit can facilitate the
rapid assistance of necessary consultants and ensure the support of services essential not only during, but even after, an ECMO run.
According to the expectations of a tertiary level intensive care unit, daily aroundthe-clock availability of the following services must be guaranteed:
Clinical laboratory
Blood bank
Radiology department ensuring necessary instrumentation for bedside radiologic
and fluoroscopic exams
Operating theater equipped for cardiothoracic surgery (indispensable for managing patients requiring cardiorespiratory support with VA ECMO)
Special consideration must be given to rehabilitation services: physical and
respiratory therapies. In some lung transplant centers, the current tendency is to
have ECMO-supported patients awake and extubated, making rehabilitative
measures extremely important. Surviving patients could have clinically
significant respiratory and musculoskeletal disabilities requiring long-term
rehabilitation.
Finally, a nutrition support service needs to be involved because of the patients
decline in nutritional status, body composition changes, and sarcopenia, a relevant
problem in critically ill patients, and even more significant in patients with respiratory failure [12].
[email protected]
16
2.6
Evaluation of Necessary Equipment
Because of rapid advances in ECMO technology, selection of equipment should be

based not only on clinical considerations but also on the awareness that there may
be investment in equipment that will soon be outdated. A criterion to be followed is
the aim toward uniformity of materials, since this will facilitate staff training,
enhance familiarity with equipment, and reduce the risk of error. Specialized technical support is often required for such sophisticated technology. A perfusion service
and/or biomedical engineering department can be responsible for material revision
and maintenance, but agreement with the seller company may be needed for maintenance and replacement of equipment.
The actual case load is the first criterion for determining the minimum necessary
storage, though if a sudden rise in activity can be anticipated, additional supplies
must be guaranteed.
Essential available endowment consists of backup components of the ECMO
system and circuits and instrumentation (including a light source) to support bedside surgical procedures, such as surgical revision of cannulas or management of
hemorrhagic complications. If a central ECMO has been placed, instrumentation for
immediate surgical reopening of the chest must be ready and personnel trained in
management of such complications available.
2.7
Identification of Financial Support and Cost-Benefit Ratio
ECLS requires highly sophisticated technology, is labor intensive and resource dependent, and requires highly specialized personnel, all of which determine the entity of
financial support to be planned in the early phase of program development. A clear
business plan should define the necessary starting budget, the magnitude of which is
based on the anticipated costs of equipment, supplies, infrastructure, and personnel.
Less manifest might be the expectation of income and benefits deriving from an ECMO
program. Revenue from ECLS will depend on the regional/national health-care system
of reimbursement. The cost-benefit ratio for such high technology applied to often fatal
diseases can be calculated by weighing the treatment expense against the survival rate
and the number of quality-of-life years (QALY) after the treatment. In the neonatal
setting, positive survival outcomes for ECLS patients have been clearly verified [13,
14]. The CESAR trial had similarly positive results in adult patients, with a 20 % better
survival when ECMO, in place of conventional therapy, was used for ARDS [15].
2.8
Training
Similarly to what is advisable for the implementation of any new project, medical,
nursing, and paramedic personnel must be trained from both the theoretical and
practical points of view, and the acquired skills must be tested and verified repeatedly while running the program.
[email protected]
17
ELSO has dedicated a significant part of its scientific activity to the publication of
valuable recommendations for setting up a comprehensive educational plan. It has
published Guidelines for Training and Continuing Education of ECMO Specialists
[4], the ELSO Red Book [10], and the ELSO ECMO Specialist Training Manual. The
result has been the drafting by ELSO of specific educational requirements expected
from ECMO specialists, though some deviation from ELSO guidelines can be
expected because of regional-institutional organization and regulations. Moreover, it
is quite common for each center to adopt local training programs consisting of didactic courses and hands-on training. According to the ELSO guidelines, an ECMO
training course should last at least 1 week and should include 2436 h dedicated to
didactics and 816 h to hands-on training in order to review ECLS equipment components and functional checks, basic and emergency procedures, and patient safety.
Didactic courses must cover the following topics:
Indications, contraindications, and evaluation of the risk/benefit ratio
Pathophysiology of diseases requiring ECMO support
Selection of the most appropriate ECLS support (VA, VV, VA-V)
Physiology of extracorporeal membrane function, pathophysiology of oxygen
delivery and consumption, and physiology of venoarterial and venovenous ECMO
Knowledge of ECMO equipment, cannulas, circuits, and materials necessary for
extracorporeal support
Daily management of patients and circuits
Identification and management of clinical and mechanical emergencies
Weaning from ECMO, decannulation
Coagulation
Post-ECMO complications and post-ECMO outcomes
Each ECMO specialist is expected to attend training courses, be updated periodically, and review protocols and results. The level and depth of an educational program
will be determined by the existing competencies in each center. The ECMO coordinator is responsible for the training of the ECMO team, as well as for the verification of
level of competency, and compliance with standards, international and/or internal.
Sharing results and experiences with other established ECMO centers is an
essential component of the training/updating process and offers a wider and more
critical review of each activity. This will help ensure rectification of management
defects and optimization of treatment modalities.
2.9
Architectural and Infrastructural Features
Hospital infrastructures must be able to accommodate both the patient and the entire
kit of devices required for clinical management over the ECMO run. The ideal
available space for a single patient should be about 2226 m2 according to the structure of the intensive care unit, whether it be open space or single room. The number
of sockets and sources of medical gas and air supply available in each patient location must be adequate to ensure proper functioning of all devices required to support
vitals.
[email protected]
18
Intrahospital transport routes are further factors to be considered while assessing

fitness of the ECMO center. Patients might need to be moved from one unit to
another, and the access route should never be impeded by architectural barriers.
Adjunctive equipment might become necessary over the ECMO run and include
those for renal replacement therapies, plasmapheresis, nitric oxide supply, and
intra-aortic balloon pump counterpulsation. Space and infrastructure should never
limit a proper clinical approach to the patient.
Space scheme and proximity must be considered. Rapid access to storage areas
where equipment and devices can be placed is essential.
With the encouraging results obtained over the last 10 years, and the increasingly
simplified use of latest-generation miniaturized devices, extracirculatory support is
bound to become more widespread. In any event, it is essential that the human and
material requirements reported in this brief chapter are met in order to help guarantee the success of a newly implemented ECMO program.
References
1. Peek G et al (2009) Efficacy and economic assessment of conventional ventilatory support
versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a
multicentre randomised controlled trial. Lancet 374(17):13511363
2. Davis et al (2009) Australia and New Zeland Extracorporeal membrane oxygenation
(ANZ ECMO) influenza investigators. JAMA 302(17):18881895
3. ELSO Guidelines for ECMO Centers, version 1.7 Feb 2010, pp 17
4. ELSO guidelines for the training and continuous education of ECMO specialists. Version 1.5
(2012) Available at: http://www.elsonet.org/index.php/resources/guidelines.html
5. MaClaren G et al (2007) Extracorporeal membrane oxygenation and sepsis. Crit Care Resusc
9:7680
6. Skinner SC et al (2012) Improved survival in venovenous vs venoarterial extracorporeal membrane oxygenation for pediatric non cardiac septic patients. A study of the Extracorporeal Life
Support Organization registry. J Pediatr Surg 47:6367
7. Sung K et al (2006) Improved survival after cardiac arrest using emergent autopriming percutaneous cardiopulmonary support. Ann Thorac Surg 82:651656
8. Halm EA et al (2002) Is volume related to outcome in health care? A systematic review and
methodologic critique of the literature. Ann Intern Med 137(6):511520
9. Kahn JM (2007) Volume, outcome and the organization of intensive care. Crit Care 11(3):129
10. Ogino MT et al (2012) ECMO Administrative and Training Issues, and Sustaining Quality. In
Annich G (ed) ECMO: extracorporeal cardiopulmonary support in critical care, 4th edn.
ELSO, Ann Arbor. pp. 479497
11. Extracorporeal Life Support Organization (2010) Extracorporeal: ECMO specialist training
manual. ELSO, Ann Arbor
12. Sheean PM et al (2013) The prevalence of sarcopenia in patients with respiratory failure classified as normally nourished using computed tomography and subjective global assessment.
JPEN J Parent Enterl Nutr 20(10):17
13. Roberts TE (1998) Economic evaluation and randomised controlled trial of extracorporeal
membrane oxygenation: UK collaborative trial. The Extracorporeal Membrane Oxygenation
Economics Working Group. BMJ 317:911916
14. Petrou S et al (2004) Cost effectiveness of neonatal extracorporeal membrane oxygenation
based on four years results from the UK Collaborative ECMO trial. Arch Dis Child Fetal
Neonatal Ed 89:F263F268
15. Peek GJ et al (2009) Efficacy and economic assessment of conventional ventilatory support
multicentre randomised controlled trial. Lancet 374:13521363
[email protected]
Basic Aspects of Physiology

During ECMO Support
Vittorio Scaravilli, Alberto Zanella, Fabio Sangalli,
and Nicol Patroniti
3.1
Introduction
Every life form, except for Archaebacteria [1], relies on cellular respiration to create vital energy; hence, it consumes oxygen and produces carbon dioxide. Since
gases move always downstream partial pressure gradient, life can only exist if oxygen diffusion is continuously guaranteed from the outer ambient to the mitochondria of each cell and the reverse for carbon dioxide [2]. This process is also called
the oxygen cascade. In unicellular organisms, gas homeostasis is achieved by
simple transmembrane gas diffusion. Contrarily, multicellular organisms had to
develop complex cardiorespiratory systems to absorb, transport, deliver, and eliminate vital gases.
V. Scaravilli (*)
Dipartimento di Scienze della Salute, University of Milan-Bicocca,
San Gerardo Hospital, Via Donizetti 106, Monza 20900, Italy
A. Zanella
Dipartimento di Scienze della Salute, University of Milan-Bicocca,
San Gerardo Hospital, Via Donizetti 106, Monza, Milan 20900, Italy
Department of Experimental Medicine, University of Milano-Bicocca,
San Gerardo Hospital, Via Donizetti 106, Monza 20900, Italy
F. Sangalli
N. Patroniti, MD
Health Sciences Department, Urgency and Emergency Department,
University of Milano-Bicocca, San Gerardo Hospital,
Via Pergolesi 33, Monza (MB) 20900, Italy
[email protected]
19
20
V. Scaravilli et al.
The mainstay of modern intensive care medicine is to guarantee the function of

the cardiorespiratory system. Extracorporeal membrane oxygenation (ECMO) represents nowadays the only therapy capable of completely replacing these vital
functions.
ECMO consists of a life support technique based on the patients venous blood
diversion towards an artificial gas exchanger, which provides blood oxygenation
and decarboxylation. Subsequently, arterialized blood is returned to the patient.
Blood stream can be redirected to either the arterial or the venous circulation,
through a central or peripheral vessel cannulation. In the venoarterial setting, the
membrane lung (ML) is in parallel to the natural lung (NL) and the extracorporeal
blood pump provides circulatory support. In the venovenous setting, ML is in series
to the NL and the blood pump does not support systemic circulation.
Thus, a venoarterial ECMO (VA ECMO) replaces both heart and lung function and
can be applied for cardiac and lung failure [3], while venovenous ECMO (VV ECMO)
substitutes only native lung function and is used for respiratory failure [4].
Management of ECMO support is the last frontier of applied physiology. A solid
knowledge of hemodynamic and respiratory physiology is mandatory to take care of
patients undergoing ECMO. In this chapter, a review of physiology essentials during ECMO support will be presented. A particular interest will be directed on oxygenation, decarboxylation, and hemodynamics during ECMO support, with the
necessary distinction between VA and VV ECMO.
3.2
The Artificial Lung
During ECMO support, oxygen delivery and carbon dioxide removal are determined by a close interaction between the artificial lung performances, the natural
lung (NL) function, and the cardiac output (CO) of the patient. The gas transfer in
the artificial lung is an essential step of this process.
A comprehensive description of mass gas transfer and membrane lung engineering goes beyond the scope of this chapter. We address the interested reader to other
excellent publications on this specific topic [5]. We will introduce here the essentials for the management of patients undergoing ECMO.
Modern blood oxygenators are membrane gas exchangers and therefore are commonly called membrane lung (ML). They are comprised of microporous hollow
fiber membranes, made of hydrophobic polymers (e.g. polymethylpentene). The
sweep gas flows through the lumen of these fibers while blood flows on their outside. Differently from bubble oxygenators, membrane lungs avoid direct contact
between blood and gases. Asymmetrical composite hollow fibers and heparin coated
surfaces have been recently introduced to limit common problems encountered
using first-generation membrane lungs, such as plasma leakage and coagulation
activation; see Chap. 6 for more details.
Despite these technical progresses, developing a device capable of substituting
the lung function is still a great technical challenge. This can be easily understood
through a comparison between native and artificial lungs. In the native lungs, gases
[email protected]
3 Basic Aspects of Physiology During ECMO Support
VV LF - ECCO2R
Membrane lung gas transfer (mL/min)
300
21
250
200
VCO2ML
VO ML
2
150
VA ECMO
100
VV ECMO
50
VV ECCO R
2
0
0
0.5
2
3
ECMO blood flow (L/min)
Fig. 3.1 Oxygen delivery (VO2ML) and carbon dioxide removal (VCO2ML) as a function of
ECMO blood flow (BF). Operative range of BF of main CO2 removal techniques is represented.
VV-LF-ECCO2R venovenous low-flow extracorporeal CO2 removal, VV-ECCO2R venovenous
extracorporeal CO2 removal, VV ECMO venovenous ECMO, VA ECMO venoarterial ECMO
move across the alveolar-capillary membrane, which is about 150m2 wide and
13m thick. This huge exchange area is compacted in a total volume of only 5L,
leading to a surface/blood volume ratio of about 300cm1. Under stress, human
respiratory system is able to guarantee oxygen delivery (VO2NL) and carbon dioxide removal (VCO2NL) up to 3,000mL /min [6].
In comparison, modern MLs are much less efficient. They have an exchange
surface lower than 4m2 wide and a surface/blood volume ratio of 30cm1. In the
artificial lungs, the interface between blood and gases is 1030m thick. For these
reasons, a ML provides a gas transfer just barely adequate for the metabolic requirements of a resting man. Indeed, even in the best conditions, oxygen delivery and
carbon dioxide removal of 250200mL/min can be obtained through an artificial
lung.
Gas transfer capabilities through ML are consequent to their intrinsic performances, which are directly proportional to the membrane surface area and dependent on hollow fiber characteristics, such as thickness and material. While these
determinants cannot be altered at the bedside, the clinician can act on the blood flow
(BF) and sweep gas flow (GF) to modify VO2ML and VCO2ML (Fig.3.1).
Oxygen and carbon dioxide pressure gradients between the sweep gas flow and
the blood are the fundamental determinants of the gas transfer. In turn, these pressure gradients are dependent on metabolism and blood transport of oxygen and
carbon dioxide.
[email protected]
22
3.3
Oxygen
Normal oxygen consumption for a healthy adult at rest is about 250mL/min

(58mL/kg/min). Oxygen consumption may significantly increase during exercise,
shivering and fever, but also, less noticeably, with increased level of catecholamine
(restlessness, pain, exogenous therapeutic administration), increased work of
breathing and increased thyroid hormones. Conversely oxygen consumption is
reduced by hypothermia, sedation, paralysis and hypothyroidism [7].
Oxygen is used in mitochondria for substrate oxidation, which leads to production
of energy and carbon dioxide. Therefore, the oxidative metabolism generates the partial
pressure gradient that drives the oxygen from the outer ambient to the cell mitochondria. Respiratory and cardiovascular systems are adaptive mechanisms developed by
multicellular organisms to guarantee the oxygen supply to any single cell of the body.
We will follow the path of oxygen from the ambient air to mitochondria to elucidate the main physiologic aspects of oxygenation [8].
Partial pressure of oxygen of inspired gases (pO2insp) is determined by the
inspired oxygen concentration (FiO2) and the barometric pressure (pB):

pO 2 insp = FiO 2 pB
The alveolar partial pressure of oxygen (pO2alv) is lower compared to pO2insp

due to the added water vapor and the balance between oxygen removal by pulmonary capillary and oxygen replacement by alveolar ventilation.
Subsequently oxygen passes from the alveolar gas into the blood, mainly into the
erythrocytes, by a passive diffusion process. In healthy lungs, diffusion is very efficient due to the extremely limited thickness of the alveolar-capillary barrier; therefore such equilibrium is easily reached.
The oxygen solubility in plasma is minimal (the coefficient of solubility is
0.003mL/mmHg per 100mL of blood); therefore with a normal arterial pO2 of
100mmHg, the oxygen dissolved in plasma is only 0.3mL/dL, corresponding to an
oxygen delivery of 15mL/min, assuming a cardiac output of 5L/min. Without
hemoglobin, a cardiac output of 80L/min would have been necessary to provide an
oxygen delivery of 250mL/min!
Fortunately, evolution has provided hemoglobin, which raises blood oxygen content exponentially, binding 1.39mL of oxygen per gram if fully saturated (SatO2).
Hence, total oxygen content can be calculated as

O 2 content = ( Hb SatO 2 1.39 ) + ( pO 2 0.0031)
At a normal hemoglobin concentration, the arterial blood oxygen content is

about 20mL/dL.
Once the pulmonary capillary blood is loaded with oxygen, the cardiac output
(CO) is regulated to maintain the systemic oxygen delivery (DO2) at four to five times
the consumption. Oxygen delivery is the arterial oxygen content times cardiac output,
which is the oxygen delivered to the tissues each minute. DO2 depends on cardiac
output, hemoglobin concentration, hemoglobin saturation, and dissolved oxygen.
[email protected]
23
In the peripheral arterial capillary, hemoglobin releases oxygen. Following a

pressure gradient, it diffuses through the endothelium, the intracellular space, and
the cellular membrane, reaching its final destination, the mitochondria. In the mitochondria pO2 ranges between 3.8 and 22.5mmHg, but it varies between tissue, cells,
and even regions of the same cell.
After transferring oxygen to the tissues, the capillary blood flows into the venous
district where the oxygen content (CvO2) may be computed as
VO 2 NL
CvO 2 = CaO 2
CO
The right heart then drives the venous blood into the pulmonary circulation.
There, venous blood is loaded of an amount of oxygen corresponding to that consumed by the tissues.
From the pathophysiologic point of view, the most important cause of acute
hypoxic respiratory failure is maldistribution of ventilation (VA) and perfusion (Q).
Following the three compartment lung model developed by Riley [9], the lung can
be imagined as divided in three functional units characterized by different VA/Q
ratios:
1. Ideal lung, without alteration of the natural coupling between ventilation and
perfusion, having VA/Q~1
2. Dead space, ventilated but not perfused alveoli, having VA/Q=
3. Intrapulmonary shunt, perfused but not ventilated alveoli, having VA/Q=0
In this model, gas exchange can happen only in the ideal alveoli. Dead space has
important effects on carbon dioxide elimination (see later). Conversely, as an effect
of shunt, the blood flowing through a pulmonary parenchyma with VA/Q=0 does
not participate in the gas exchanges and is not oxygenated. Hence, part of the venous
blood mixes with arterial oxygen content and determines hypoxemia. The shunt is
usually characterized as a ratio (Qs/Qt) between the shunted blood (Qs) and the total
pulmonary perfusion (Qt).
When the Qs/Qt is higher than 0.4, even providing inspiratory fraction of oxygen
up to 100% is not sufficient to ensure an adequate oxygenation, and some degrees
of hypoxia should be expected. If Qs/Qt is higher than 0.4, oxygenation provided by
the native lung cannot sustain vital oxygen delivery. In these extreme clinical conditions, ECMO may prove to be the only clinical solution (Fig.3.2).
3.3.1 Oxygenation During VV ECMO

When the natural lung is ineffective in oxygenating the blood, a VV ECMO support
may be employed.
The global physiology of oxygenation during VV ECMO may be elucidated by
following the changes in blood O2 content along the circulatory system. Figure3.3
is a schematic representation of a patient connected to a VV ECMO (panel a) and
blood oxygen contents along the circulatory system (panel b).
[email protected]
24
Fig. 3.2 Partial pressure of

oxygen in the arterial blood
(PaO2) as a function of
pulmonary shunt (Qs/Qt), at
various fractions of oxygen at
the ventilator. If pulmonary
shunt is higher than 0.4, vital
arterial oxygenation may not
be achieved even by 100%
oxygen supplementation at
the ventilator
500
FiO2 responders
PaO 2 (mmHg)
400
FiO2 non-responders
FiO2 1
300
FiO2 0.6
200
FiO 0.21
2
100
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
Qs/Qt
O2
CO2
Cc
15
Cout
14
[O2] mL/dL
ML
O2
Cc
BP
Cin
Cv
Panel a
CO2
Ca
O2
13
12
11
10
9
Ca
Cvmix
CO2
Cout
16
Cvmix
Cin
Cv
Recirculation
effect
Panel b
CO/BF
effect
Cv
VO2tot and CO
effect
Fig. 3.3 Oxygen delivery and consumption during VV ECMO. Panel (a) The main determinants
of blood oxygen content are represented. Oxygen content in the various blood compartments of the
patient and ECMO circuit: Ca arterial, Cv venous, Cin circuit inlet, Cout circuit outlet, Cvmix
mixed venous, Cc ideal unshunted pulmonary capillary. BP blood pump, ML membrane lung.
Panel (b) Diagram representing the blood oxygen content of a patient (Hb 10g/dL) connected to a
VV ECMO support. The different sections of the venous, arterial, and extracorporeal circulation
are distinguished by color and thickness. Blue lines represent deoxygenated blood, while red well-
oxygenated blood. Thicker lines correspond to higher blood flows. Arrows represent the effects of
recirculation ratio, cardiac output, and oxygen consumption on oxygen delivery during VV ECMO
Blood leaves the peripheral tissue with low oxygen content (CvO2). The blood
pump generates the extracorporeal blood flow (BF) diverging part of the venous
return towards the ML. In this process, part of the already-bypassed flow is drained
back to the extracorporeal circuit. This results in a recirculating blood flow (R),
[email protected]

250
BF 5 L/min
200
PaO 2 (mmHg)

oxygen in arterial blood
(PaO2) as a function of the
fraction of the recirculating
blood flow (R/BF) at different
blood flows (BF). High
recirculation fractions have
extreme detrimental effects
on the membrane lung
oxygen delivery
25
150
BF 4 L/min
100
BF 3 L/min
50
0
0.1
0.3
0.2
0.4
0.5
R/BF
which has detrimental effects on the oxygenation efficiency of VV ECMO (Fig.3.4).

Moreover, the R accounts for the difference in oxygen content between CvO2 and
the oxygen content of the blood entering the ML (CinO2).
Then, the ML loads the extracorporeal BF with oxygen (VO2ML) and raises the
oxygen content in the outlet blood (CoutO2), as follows:

VO 2 ML = BF ( CoutO 2 CinO 2 )
Subsequently, during VV ECMO, the blood returning to the right heart (the
mixed venous blood, with oxygen content CvmixO2) is an admixture of the deoxygenated venous return and the well-oxygenated extracorporeal blood. The important role of the interaction between CO, BF, and R in determining the respective
influence of these two components will be described later.
The resulting effect of VV ECMO application is the increase of the oxygen content of the blood returning to the lung. Fundamentally, VV ECMO improves arterial
oxygenation increasing oxygen content of mixed venous blood.
CvO2 is eventually increased to CaO2 by the residual oxygenating capacity of the
NL. The oxygen added to the blood by the natural lungs (VO2NL) is calculated as

VO 2 NL = CO ( CaO 2 CvmixO 2 )
The sum of VO2NL and VO2ML is equal to the total oxygen consumption of the
patient:

VO 2 Tot = VO 2 ML + VO 2 NL
Since most of the oxygen is transported bound to hemoglobin, the oxygen content of blood is highly dependent on the concentration and saturation of hemoglobin. An effective strategy in increasing DO2, after optimizing the ML and NL
function, is increasing the hemoglobin concentration [10].
We will now address the roles of VO2ML, SvmixO2, and VO2NL in determining
arterial oxygenation during VV ECMO.
[email protected]
26
3.3.1.1 VO2ML
The capacity of the ML of transferring oxygen is mainly determined by three
factors:
1. The intrinsic properties of the ML affect the oxygen passive diffusion from the
sweep gases into the blood.
2. The partial pressure gradient of oxygen between blood and sweep gases. The
oxygen partial pressure in the sweep gases is determined by the FiO2. The oxygen transfer through the ML is affected by the ventilation/perfusion matching,
the hemoglobin concentration, and the transit time. When the oxygen partial
pressure increases in the blood crossing the ML, hemoglobin becomes fully saturated, and little additional oxygen, the physically dissolved, can be further
loaded; therefore also an additional increase in GF will determine a minimal
increase in the ML oxygen delivery. On the blood side, PinO2 and the resulting
SinO2 highly affect the VO2ML. Recirculation, increasing PinO2, may vastly
reduce VO2ML (Fig.3.4).
3. The extracorporeal blood flow is the major determinant of VCO2ML. Indeed,
increasing the extracorporeal BF determines linear augmentation of VO2ML.
3.3.1.2 SvmixO2
During VV ECMO, SvmixO2 is mainly determined by the oxygen saturation of the
blood leaving the tissues (SvO2), the ratio between BF and CO, and the presence of
recirculation (Fig.3.5).
As previously said, an increase in BF at constant CO and R always determines an
improvement in arterial oxygenation and tissue oxygen delivery.
More complicated are the effects of CO changes on oxygen delivery and arterial
oxygenation. Assuming stable BF and R, patients with elevated CO necessitate
higher BF to achieve normal arterial PaO2 levels (Fig.3.6, panel a). This does not
necessarily means that a lower cardiac output is desirable. Indeed, DO2 depends on
CaO2 and tissue perfusion. During arterial hypoxemia, more than ever, an adequate
100
80
SvmixO (mmHg)
90
Fig. 3.5 Mixed venous

oxygen saturation (SvmixO2)
as a function of blood flow/
cardiac output ratio (BF/CO),
at different oxygen saturation
in the blood entering the ML
(SinO2)
70
60
SinO 70 %
2
SinO 60 %
2
SinO2 50 %
50
0.1
0.2
0.3
BF/CO
[email protected]
0.4
0.5
0.6
a
200
CO: 11 L/min
CO: 8 L/min
50
CO: 11 L/min
80
40
PvO2 (mmHg)
CO: 5 L/min
120
60
CO: 8 L/min
160
PaO2 (mmHg)
27
40
CO: 5 L/min
30
20
10
0
BF (L/min)
BF (L/min)
Fig. 3.6 The interaction between cardiac output (CO) and blood flow (BF). Panel (a) Partial pressure of oxygen in the arterial blood (PaO2) as a function of blood flow (BF), at different cardiac
output (CO). Panel (b) Partial pressure of oxygen in the venous blood (PvO2) as a function of blood
flow (BF), at different cardiac output (CO)
CO is essential to assure adequate oxygen delivery. Thus, increases in CO are usually associated with an increase oxygen delivery by the natural lung, assumed that
the Qs/Qt is not changing. In this situation, a higher cardiac output provides higherVO2NL, CvO2, and consequently PvO2 (Fig.3.6, panel b).
However, a rise in CO is usually a sign of augmented tissue oxygen requirements
(e.g., fever, agitation, sepsis) and may change Qs/Qt and the R flow. During
VV ECMO support, sudden changes of hemodynamic status are accompanied by
complete alteration of the oxygenation steady state. Indeed, clinical experience
teaches that in these scenarios CvmixO2 and CaO2 may unpredictably improve or
worsen. Considering that many tissues have various oxygen requirements and vasculature, it is particularly difficult to predict the effect of a change in CO on oxygenation of different peripheral organs. Still, on this topic, scientific evidence is poor,
and more research trials are needed [11] to better understand the complex pathophysiology of the effects of CO modulation on oxygen delivery.
3.3.1.3 VO2NL
The mixed venous blood is oxygenated according to gas exchange capability of the
NL, which depends on the severity of the lung disease (mainly the intrapulmonary
shunt fraction) (Fig.3.7) and the ventilator setup (Fig.3.8).
As clearly visible from Fig.3.7, the worse the residual gas exchange capability
of the NL, the higher BF and consequently cannula size are necessary. Indeed, if the
intrapulmonary shunt is over 0.7, vital blood oxygenation can be obtained only by
applying BF over 4L/min. The use of adequately sized drainage cannulas is of paramount importance in this situation.
VV ECMO replaces, partially or completely, the function of the NL and allows
reducing all the risk factors contributing to the onset of ventilator-induced lung injury:
high ventilation volumes and pressures and high FiO2 level. However, an extremely
protective ventilator strategy, based on low levels of FiO2, PEEP and minute ventilation, may temporarily worsen the gas exchange function of the NL and therefore an
increase in extracorporeal support is often required. The ventilatory strategy and the
[email protected]
28
500
Qs/Qt: 0.4
400
PaO (mmHg)
Qs/Qt: 0.7
300

oxygen in arterial blood
(PaO2) as a function of blood
flow (BF), at different
pulmonary shunts (Qs/Qt)
200
Qs/Qt: 1
100
0
3
BF (L/min)
400
FiO NL1
2
300
FiO NL 0.6
2
200
PaO (mmHg)
Fig. 3.8 Partial arterial

pressure of oxygen (PaO2)
as a function of blood flow
(BF), at various fractions of
oxygen at the ventilator
(FiO2NL)
100
FiO NL 0.21
2
BF (L/min)
target oxygenation highly influence the required BF level and the choice of ECMO
equipment and cannula sizes.
3.3.2 Oxygenation Support During VA ECMO

Particular attention has to be paid on oxygen delivery during VA ECMO support.
No particular consideration is necessary until cardiac output is severely compromised, being all the oxygen delivery provided by the VA ECMO support.
Complications may arise when lung function is impaired and a residual cardiac
output is present, especially if a femoral arterial reinfusion cannula is used. Indeed,
in this particular clinical condition, blood ejected from the left ventricle (not oxygenated by the compromised native lung) may perfuse the aortic arch and proximal
aortic branches, determining coronary and cerebral hypoxia. Contrarily, the lower
extremities will appear well perfused.
[email protected]
29
This hypoxic Harlequin syndrome [12] may be overlooked if the arterial blood
sampling catheter is positioned in the femoral artery or in the left arm, while it is promptly
recognized when the samples are collected from the right arm. It is hence important to
position a right radial artery catheter to monitor heart and brain perfusion. If this is not
feasible, at least an oximetry probe should be positioned on the upper right arm.
Potential solutions are increasing BF to limit LV ejection and conversion to a
veno-venoarterial ECMO, by addiction of an extra venous reinfusion cannula. The
increase in BF may also paradoxically worsen lung function further, as discussed
below. A low threshold for early direct LV venting should be maintained to protect
not only the heart but also the lungs.
3.4
Carbon Dioxide
In resting condition, an adult healthy man produces approximately 250mL/min of

carbon dioxide value that is influenced by metabolic activity, core body temperature
and caloric intake.
Carbon dioxide is the end product of aerobic metabolism. After being generated
in mitochondria, following a series of partial pressure gradients, carbon dioxide
passes through cytoplasm and extracellular fluid to the venous blood stream. Then,
carbon dioxide is carried to the lung alveoli where it is released to the outer
ambient.
The normal venous blood carries at least 55mL of CO2/100mL, in three different forms: dissolved, as bicarbonate ions and in combination with proteins as carbamino compounds.
Dissolved CO2 obeys Henrys law, as follows:

Dissolved CO 2 content = pCO 2 coeff
CO 2 + H 2 O H 2 CO3
where coeff=solubility coefficient=0.03 mmol/LmmHg.

With a normal venous pCO2 of 45mmHg, at a temperature of 37 C, dissolved
CO2 has a concentration of only 3mL/100mL of blood. Hence, dissolved CO2 represents only 5% of total CO2 content.
Dissolved CO2 reacts with water in blood to form carbonic acid, according to the
following reaction:
This reaction is extremely slow in the plasma (T~1min), while its speed is
greatly increased in whole blood by carbonic anhydrase (also of 10,000 times),
which is contained in red blood cells. Thanks to this enzyme, the conversion of CO2
to H2CO3 requires less than 2ms.
Subsequently, most of the carbonic acid formed in the red blood cells further
dissociates into hydrogen and bicarbonate ions, according to the following not-
enzymatic reaction:

H 2 CO3 H + + HCO3
[email protected]
30
About 70% of CO2 blood content is in bicarbonate ion form; indeed about 50mL
of CO2 is carried in this form in 100mL of blood.
Moreover, carbon dioxide also reacts with amino end groups of hemoglobin,
forming carbaminic compounds as follows:

HbNH 2 + CO 2 Hb NH COOH
Carbon dioxide carried by carbamino compounds has a concentration of

3mL/dL in venous blood.
After being transported by venous blood to the pulmonary capillary, it readily diffuses to the alveolar space, passing through the alveolar-capillary membrane. Diffusion of carbon dioxide through this membrane is extremely efficient,
and hypercapnia is indeed virtually never caused by altered diffusion capacity
of the lungs. Consequently, alveolar partial pressure of carbon dioxide (alv
pCO2) can be considered usually equal to pulmonary end-capillary blood (ven
pCO2).
Hence, alv pCO2 is the most important determinant of carbon dioxide removal
from native lungs. Carbon dioxide is constantly added to alveolar gas by venous
circulation and removed by alveolar ventilation, as follows:
alv pCO 2 =
VCO 2 NL
Alveolar ventilation
Alveolar ventilation is only the fraction of the inspired tidal volume that actually
participates to the gas exchange, hence:

Alveolar ventilation = Respiratory frequency ( Tidal volume - Dead space )
Contrarily, the dead space is the component of the tidal volume not leading to
effective gas exchange. Dead space may be further distinguished in apparatus, anatomical and alveolar dead space, which are respectively due to the presence of any
external breathing machine, the patient airways and the unperfused alveoli. Alveolar
dead space is defined as the part of the inspired volume that reaches the alveoli, but
cannot take part to the gas exchange as consequence of altered ventilation/perfusion
matching. Many clinical conditions are characterized by various levels of increased
alveolar dead space, in particular pulmonary embolism, chronic obstructive pulmonary disease (COPD) and acute respiratory distress syndrome (ARDS). In these
syndromes, the efficiency of carbon dioxide removal is altered. Consequently, the
necessary CO2 elimination is possible only at the cost of augmented alv pCO2 or
higher minute ventilation.
Both of these alternatives may not be desirable. Indeed, arterial pCO2 is
dependent on alv pCO2, and an indiscriminate rise in pCO2 has a strong impact
on patient pH and may not be clinically sustainable. Moreover, augmenting
minute ventilation is a well-known cause of ventilatory-induced lung injury
(VILI).
[email protected]
31
3.4.1 Carbon Dioxide Removal During ECMO Support

All the clinical situations in which it is advisable to reduce alveolar ventilation or to
avoid hypercapnia represent a potential indication for extracorporeal CO2 removal.
Hence, ECMO as a mean to remove carbon dioxide finds its clinical rationale to:
1. Reduce ventilatory needs and avoid VILI during ARDS [13]
2. Mitigate dynamic hyperinflation and hypercapnia during acute reacutization of
COPD [14] and status asthmaticus [15]
3. Bridge to lung transplant [16]
Removal of CO2 during VV ECMO is much easier to achieve than oxygenation.
With any type of ML, clearance of CO2 is always more efficient than oxygen delivery (Fig.3.2). As mentioned, most of the CO2 content is transported in blood in the
form of bicarbonate ion, with a total concentration of about 55mL of CO2 per
100mL of blood. This means that 500mL of venous blood contains an amount of
CO2 correspondent to the entire minute CO2 production of an adult male (approximately 250mL/min). Hence, during ECMO support, the entire patient CO2 production may be removed with low BF. As an example, a conventional ML can readily
remove 250mL/min from an extracorporeal BF of only 1.5L/min. This can happen
only if high sweep gas flows (e.g. 815L/min, according to the oxygenator characteristics) are used. Indeed, the gas flow is responsible for the removal of the carbon
dioxide from the lumen of the hollow fibers of the membrane lung. Rising sweep
gas flow of a membrane lung reduces partial pressure of CO2 inside the hollow
fibers, augments the partial pressure gradient in between blood and gas phase, and
consequently augments CO2 removal.
Hence, the amount of CO2 transfer is relatively independent of blood flow while
the sweep gas flow rate is its major determinant [17]. This strict relationship between
carbon dioxide removal and GF dissociates membrane lung CO2 removal from
extracorporeal blood flow and has important clinical consequences.
First, during VV ECMO the clinician may selectively change carbon dioxide
removal of the artificial lung by altering sweep gas flow, maintaining oxygen delivery unaltered. Through this intervention, ventilatory drive of the patient can be mastered and finely titrated to the desired level. As an example, in the most severe
clinical conditions, extremely low tidal volume ventilation (and even apnea) can be
achieved through elevated sweep gas flows [18]. Conversely, during weaning, a
residual ventilatory drive can be maintained and spontaneous breathing guaranteed
using a more moderate gas flow [19].
Second, several new devices have been recently implemented to perform low-
flow extracorporeal CO2 removal with less invasiveness and side effects compared
to ECMO aimed at oxygenation support [20, 21]. These new technical approaches
can be employed in all those clinical situations where controlling ventilation is necessary, while oxygenation is not an issue.
Third, new techniques aimed at augmenting ML CO2 removal capabilities by
extracorporeal loco-regional acidification are under evaluation [22, 23]. These may
permit to reduce required blood flows, minimize associated complications and permit a safer and broader use of extracorporeal CO2 removal.
[email protected]
32
3.4.2 Carbon Dioxide Removal During VA ECMO

Particular considerations are necessary for the management of carbon dioxide
removal and ventilation during VA ECMO. Currently VA ECMO is mainly used to
sustain circulation during cardiac failure. Namely, VA ECMO is applied in patients
whose cardiac output is severely impaired and consequently have very limited lung
perfusion, if any, from the pulmonary artery. In this condition, to conventionally
ventilate the lungs has hence no physiologic rationale, since during VA ECMO the
whole lung parenchyma is an alveolar dead space. Moreover, various studies suggest potential detrimental consequences of ventilation of unperfused lungs [24].
Contrarily, during VA ECMO, application of an adequate PEEP level and cyclically
recruitment are generally suggested to avoid lung atelectasis. Continuous end tidal
CO2 monitoring is warranted to ensure adequate ventilation when a certain degree
of venous return and right ventricular ejection is present.
3.5
Hemodynamics
The impact of ECMO on cardiovascular function depends mainly on two factors:

Type of support (venovenous versus venoarterial)
Site of vascular access (peripheral versus central, femoral artery versus axillary
artery)
The main difference lays in the type of support, with a much less hemodynamic
impact in VV as compared to VA ECMO. In VV ECMO the extracorporeal pump is
functionally in series with the patients heart, whereas in VA ECMO it is in parallel
(Fig.3.9).
c
BP
ML
ML
BP
BP
ML
Fig. 3.9 Schematics of possible venoarterial ECMO circuits. (a) Peripheral, femorofemoral cannulation. (b) Peripheral, femoro-axillary cannulation. (c) Central cannulation. BP blood pump, ML
membrane lung
[email protected]
33
3.5.1 Hemodynamics During VV ECMO

Irrespective of the site of cannulation, which can make a difference on blood gas
dynamics, blood is drained from and returned into the venous system. This makes
VV ECMO an in-series system with the natural circulation. No acute volemic
changes happen, since the same amount of blood is continuously drawn and replaced.
The preload remains unchanged, as does the afterload of the left ventricle (LV).
The energy balance of the myocardium is hence unaffected by the extracorporeal
support. Even better, blood reaches the left heart much more oxygenated than normally, with a further possible advantage on myocardial perfusion. This venous
hyperoxia may also help in reducing pulmonary vascular resistance (and hence
pulmonary arterial pressure) by partially relieving hypoxic vasoconstriction. Hypoxia
is a known risk factor for myocardial injury [25], and ECMO could act as a protective
measure to prevent such injury. Despite this, sudden hyperoxia after hypoxia was
associated with myocardial reperfusion damage in experimental models [26].
All these previous considerations stand for the conclusion that VV ECMO is a
hemodynamically neutral support. This is largely true, but it is two faceted as
every coin. VV ECMO is generally applied to very hypoxemic patients, usually
with elevated pulmonary vascular resistances and often with high cardiac output
due to infection/sepsis. It is hence essential that the patient has a competent cardiovascular function, since the extracorporeal support does not provide any circulatory
assistance. This must be assessed at the time of initiation and regularly assessed
throughout the course of ECLS, to detect any worsening due to sepsis or to any
other cause that may require conversion to a circulatory support.
3.5.2 Hemodynamics During VA ECMO

Things become more challenging from a hemodynamic point of view when it comes
to VA ECMO. Here the extracorporeal pump works in parallel with the patients
heart. Systemic flow, therefore, is a combination of that established by the extracorporeal circuit plus the amount of blood passing through the native heart and lungs.
This leads to a number of hemodynamic changes. Most of those occur irrespective
of the cannulation site, while different arterial cannulation sites carry some peculiar
advantages or disadvantages.
3.5.2.1 General Hemodynamic Changes

Loss of arterial flow pulsatility. Centrifugal pumps used for ECMO provide continuous flow. When the patients circulation is fully supported, natural ejection is
almost abolished and the arterial pressure line becomes flattened as more blood is
routed through the extracorporeal circuit (Fig.3.10). The effect of continuous versus pulsatile flow on organ perfusion has been extensively investigated, and no
definitive conclusion has been drawn to date on its potential negative effects. Both
clinical and laboratory parameters have shown a possible advantage of pulsatile
[email protected]
34
Fig. 3.10 Reduction in LV

ejection and loss of arterial
pressure pulsatility at the
increase of ECMO blood
flow
compared to non-pulsatile perfusion on cardiac, renal, and pulmonary function [27].

However, no clear advantage has been demonstrated to date on mortality [28]. A
certain degree of pulsatility index of LV ejection should however always be
promoted and maintained to prevent stagnation of blood in the left chambers and the
formation of intracardiac clots and the ensuing embolic risk. When this is not
accomplished with low-dose inotropes or with a slight reduction in pump flow, venting is required. These aspects are widely discussed in the relevant chapters. A
potential benefit in this regard has been postulated for the concomitant use of IABP
[29, 30]. Together with reduced LV afterload and increased coronary perfusion [31],
the preservation of a pulsatile flow waveform represents the rationale for aortic
counterpulsation during VA ECMO. Despite potential relevant benefits, the evidence for the use of IABP in this setting is still weak, and at the moment potential
disadvantages limit its widespread use.
Reduction of preload. VA ECMO diverts most of the venous return to the right
heart into the extracorporeal circuit. This effect is beneficial for resting the right
ventricle. The drawback of this reduction in pulmonary blood flow is that the coronary arteries are perfused mostly with desaturated blood coming from the bronchial
circulation, leading to potential myocardial ischemia.
Increase in LV afterload. The reintroduction of blood into the arterial system
leads to an increase in LV afterload that is directly related to the EC blood flow and
only marginally influenced by the cannulation site. This may lead to inadequate
drainage of the LV that may become distended. This in turn increases myocardial
energy demands and may worsen ischemia and increase pulmonary congestion with
deleterious effect on the lung. In an attempt to reduce such distention, pump flow is
frequently increased to optimize venous drainage. This might paradoxically worsen
LV distention due to a further increase in afterload. Strategies to reduce LV afterload and distention are discussed elsewhere in the book and aim at increasing natural ejection with inotropes and counterpulsation or when this is unfeasible to
directly unload the LV via either a percutaneous or surgical approach.
3.5.2.2 Peculiar Hemodynamic Changes Depending

ontheCannulation Site
Peripheral ECMO represents by far the most common configuration, and the femoral artery is the preferred cannulation site in adults. Nevertheless, other possibilities
exist. Namely, central cannulation may be needed, with the arterial cannula placed
[email protected]
35
in the ascending aorta. In peripheral ECMO, the axillary artery might also be cannulated, while the carotid artery is not used in adults.
With respect to hemodynamic perturbations, the only difference from femoral
access is that both aortic and axillary cannulations provide antegrade flow and that
the coronary arteries are perfused by well-oxygenated blood coming from the extracorporeal circuit, so avoiding the risk for myocardial ischemia. This may also constitute a theoretical advantage over femoral cannulation when IABP is used during
ECMO, but this is still uncertain.
Specific advantages and disadvantages of the different cannulation sites not pertaining to hemodynamics are discussed in the relevant chapters.
References
1. Dworking M, Rosenberg E, Schleifer K, Stackebrandt E (2006) The prokaryotes. Springer,
New York
2. Lumb A (2010) Nunns applied respiratory physiology. Elsevier, London
3. Combes A, Leprince P, Luyt C-E, Bonnet N, Trouillet J-L, Lger P, Pavie A, Chastre J (2008)
Outcomes and long-term quality-of-life of patients supported by extracorporeal membrane
oxygenation for refractory cardiogenic shock. Crit Care Med 36(5):14041411
4. Brodie D, Bacchetta M (2011) Extracorporeal membrane oxygenation for ARDS in adults.
NEngl J Med 365(20):19051914
5. Wnek G, Bowlin G (2008) Encyclopedia of biomaterials and biomedical engineering. Taylor
& Francis, London
6. OToole ML, Douglas PS, Hiller WD (1989) Applied physiology of a triathlon. Sports Med
8(4):201225
7. Bartlett RH (1996) Critical care physiology. Little, Brown/Boston
8. West JB (2008) Respiratory physiology: the essentials. Lippincott Williams & Wilkins,
Philadelphia
9. Riley RL, Cournand A (1949) Ideal alveolar air and the analysis of ventilation-perfusion relationships in the lungs. J Appl Physiol 1(12):825847
10. Schmidt M, Tachon G, Devilliers C etal (2013) Blood oxygenation and decarboxylation determinants during venovenous ECMO for respiratory failure in adults. Intensive Care Med
39(5):838846
11. Guarracino F, Zangrillo A, Ruggeri L, Pieri M, Calabr MG, Landoni G, Stefani M, Doroni L,
Pappalardo F (2012) -Blockers to optimize peripheral oxygenation during extracorporeal
membrane oxygenation: a case series. J Cardiothorac Vasc Anesth 26(1):5863
12. Sidebotham D, McGeorge A, McGuinness S, Edwards M, Willcox T, Beca J (2010)

Extracorporeal membrane oxygenation for treating severe cardiac and respiratory failure in
adults: part 2-technical considerations. J Cardiothorac Vasc Anesth 24(1):164172
13. Peek GJ, Mugford M, Tiruvoipati R etal (2009) Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet 374(9698):
13511363
14. Burki NK, Mani RK, Herth FJF etal (2013) A novel extracorporeal CO(2) removal system:
results of a pilot study of hypercapnic respiratory failure in patients with COPD. Chest
143(3):678686
15. Brenner K, Abrams D, Agerstrand C, Brodie D (2014) Extracorporeal carbon dioxide removal
for refractory status asthmaticus: experience in distinct exacerbation phenotypes. Perfusion
29(1):2628
16. Javidfar J, Bacchetta M (2012) Bridge to lung transplantation with extracorporeal membrane
oxygenation support. Curr Opin Organ Transplant 17(5):496502
17. Kolobow T, Gattinoni L, Tomlinson T, White D, Pierce J, Iapichino G (1977) The carbon
dioxide membrane lung (CDML): a new concept. ASAIO Trans 23:1721
[email protected]
36
18. Terragni PP, Del Sorbo L, Mascia L, Urbino R, Martin EL, Birocco A, Faggiano C, Quintel M,
Gattinoni L, Ranieri VM (2009) Tidal volume lower than 6ml/kg enhances lung protection:
role of extracorporeal carbon dioxide removal. Anesthesiology 111(4):826835
19. Mauri T, Bellani G, Grasselli G, Confalonieri A, Rona R, Patroniti N, Pesenti A (2013) Patient-
ventilator interaction in ARDS patients with extremely low compliance undergoing ECMO: a
novel approach based on diaphragm electrical activity. Intensive Care Med 39(2):282291
20. Bonin F, Sommerwerck U, Lund LW, Teschler H (2013) Avoidance of intubation during acute
exacerbation of chronic obstructive pulmonary disease for a lung transplant candidate using
extracorporeal carbon dioxide removal with the Hemolung. J Thorac Cardiovasc Surg
145(5):e43e44
21. Ruberto F, Pugliese F, DAlio A, Perrella S, DAuria B, Lanni S, Anile M, Venuta F, Coloni
GF, Pietropaoli P (2009) Extracorporeal removal CO2 using a venovenous, low-flow system
(Decapsmart) in a lung transplanted patient: a case report. Transplant Proc 41(4):14121414
22. Zanella A, Patroniti N, Isgr S, Albertini M, Costanzi M, Pirrone F, Scaravilli V, Vergnano B,
Pesenti A (2009) Blood acidification enhances carbon dioxide removal of membrane lung: an
experimental study. Intensive Care Med 35(8):14841487
23. Zanella A, Mangili P, Redaelli S et al (2014) Regional blood acidification enhances extracorporeal carbon dioxide removal: A 48-hour animal study. Anesthesiology 120(2):416424
24. Kolobow T, Spragg RG, Pierce JE (1981) Massive pulmonary infarction during total cardiopulmonary bypass in unanesthetized spontaneously breathing lambs. Int J Artif Organs
4(2):7681
25. Bajwa EK, Boyce PD, Januzzi JL, Gong MN, Thompson BT, Christiani DC (2007) Biomarker
evidence of myocardial cell injury is associated with mortality in acute respiratory distress
syndrome. Critical Care Med 35(11):24842490
26. Trittenwein G, Rotta AT, Gunnarsson B, Steinhorn DM (1999) Lipid peroxidation during initiation of extracorporeal membrane oxygenation after hypoxia in endotoxemic rabbits.
Perfusion 14(1):4957
27. Haines N, Wang S, Undar A, Alkan T, Akcevin A (2009) Clinical outcomes of pulsatile and
non-pulsatile mode of perfusion. J Extra Corpor Technol 41(1):P26P29
28. Alghamdi AA, Latter DA (2006) Pulsatile versus nonpulsatile cardiopulmonary bypass flow:
an evidence-based approach. J Card Surg 21(4):347354
29. Madershahian N, Wippermann J, Liakopoulos O, Wittwer T, Kuhn E, Er F, Hoppe U, Wahlers
T (2011) The acute effect of IABP-induced pulsatility on coronary vascular resistance and
graft flow in critical ill patients during ECMO. JCardiovasc Surg 52(3):411418
30. Jung C, Lauten A, Roediger C, Fritzenwanger M, Schumm J, Figulla HR, Ferrari M (2009) In
vivo evaluation of tissue microflow under combined therapy with extracorporeal life support
and intra-aortic balloon counterpulsation. Anaesth Intensive Care 37(5):833835
31. Madershahian N, Liakopoulos OJ, Wippermann J, Salehi-Gilani S, Wittwer T, Choi Y-H,
Naraghi H, Wahlers T (2009) The impact of intraaortic balloon counterpulsation on bypass
graft flow in patients with peripheral ECMO. J Card Surg 24(3):265268
[email protected]
Percutaneous Cannulation: Indication,

Technique, and Complications
Maurizio Migliari, Roberto Marcolin, Leonello Avalli,
and Michela Bombino
4.1
Introduction
Achieving proper vascular access is a fundamental step in implementing extracorporeal support, both for cardiac (venoarterial, VA ECMO) and respiratory (venovenous, VV ECMO) assistance. Vessel choices, ECMO cannula types and sizes,
and cannulation techniques are mainly dictated by the anatomical features of the
vascular tree and the skills of the implanting personnel. In the early 90s, with the
availability of thin-walled cannulas, cannulation techniques moved from surgical to
percutaneous in almost all the cases. A historical perspective outlining the steps of
this move will be given.
The percutaneous cannulation technique for venous and arterial ECMO accesses
and its indications and complications will be outlined in this chapter, while the surgical cannulation technique will be discussed in Chap. 5.
M. Migliari (*)
Cardiac Anesthesia and Intensive Care Unit, Department of Emergency Medicine,
San Gerardo Hospital, Via Pergolesi 33, Monza (MB) 20900, Italy
R. Marcolin M. Bombino
General Intensive Care Unit, Department of Emergency Medicine,
L. Avalli
Cardiac Anesthesia and Intensive Care Unit, Department of Urgency and Emergency,

[email protected]
37
38
4.2
M. Migliari et al.
History of ECMO Cannulation Techniques
Prolonged ECMO support had its move from the cardiothoracic scenario; therefore
surgical cannulation with VA configuration was implemented in the early years also in
patients with pure respiratory failure [1, 2]. Thereafter, the VV configuration was preferred in the extracorporeal support of patients with acute respiratory failure. The technique described by Gattinoni et al. in 1979 [3] involved a femoro-jugular approach: the
common femoral and internal jugular veins were surgically cannulated both centrally
and distally; the blood was drained from the two catheters in the common femoral vein
and from the distal catheter in the jugular vein and returned in the central internal jugular vein catheter. Continuous oozing of blood from the surgical wound was the rule, and
daily revision of vascular sites accesses was required. Patient care and motility were
difficult due to the multiple catheters and tubing at play [4]. Advances in the surgical
technique were the development of a single-vein cannulation of the femoral vein
through a double-lumen catheter [5] and the institution of sapheno-saphenous bypass
[6]. The double-lumen coaxial catheter (Fig. 4.1) allowed the institution of an inferior
vena cava-inferior vena cava bypass through a single surgical cutdown: the blood was
drained from the external lumen of the double catheter and from the distal drainage
from the leg and returned to the patients through the inner lumen. In the saphenosaphenous bypass, the surgical plans involved were more superficial, and the distal
venous drainage from the legs was maintained eliminating the need of distal cannulation; at ECMO termination the cannulas were removed with the ligation of the two
saphena magna veins; no reconstructive surgery of the veins was needed.
In the 90s, with the advent of thin-walled spring-wire-reinforced catheters on the
market, the first reports on percutaneous cannulation for cardiac and respiratory
ECMO support were published [7, 8]. Since then, the percutaneous cannulation
technique became the first choice in establishing vascular access for ECMO. The
benefits reported were a shorter procedure time, almost null bleeding at insertion
site if a coagulopathy was not present, a reduced risk of cannula-site infection, and
a very simple decannulation [9].
Fig. 4.1 Historical picture of

a handmade double-lumen
coaxial catheter for institution
of an inferior vena cavainferior vena cava bypass
through a single surgical
cutdown
[email protected]
Percutaneous Cannulation: Indication, Technique, and Complications
4.3
39
General Considerations for the Percutaneous

Placement of ECMO Cannulas
The establishment and maintenance of adequate vascular access is essential for any
type of ECMO support and can be achieved by percutaneous approach outside the
operating room by trained personnel (intensivists, emergency department physicians, cath-lab cardiologists). Since a failure in cannulation would be critical for
ECMO institution, a cardiothoracic surgeon or a vascular surgeon must be available
on site to perform a surgical approach via a semi-Seldinger or a cutdown technique
if difficulties in cannulation would arise. These problems are mainly related in the
literature to the placement of the arterial cannula in VA ECMO for cardiac assist.
Patient size and ECMO configuration mainly dictate the choice of the ECMO
cannulas. Different cannulas are available on the market for ECMO cannulation
purpose. Recently a comprehensive review on this argument was published [10],
and we will only summarize some cannula features that are important to know when
choosing a cannulation site.
The manufacturer provide the specifications about the pressure drops generated
at different flows to help choosing the right cannula for the specific clinical need
[11]. For an adult patient of >70 kg weight in a VV configuration, venous drainage
cannula ranges from 23 to 25 F, while reimmission cannula from 19 to 21 F.
Vascular access for VV ECMO can be challenging in patients with a high BMI;
in most of these patients, the femoro-jugular approach must be considered as the
first choice. The same would apply to pregnant woman: in late pregnancy a 1530
left lateral tilt position has been proposed to facilitate the insertion of the femoral
cannula [12].
Vascular ultrasound has become invaluable for the localization of the vessels and
measurements of their diameters [13, 14]. As a generic rule the size of the cannula
must be no more than two-thirds of the vessel diameter, so that blood coming from
the leg can flow freely around the cannula and venous drainage from the limb is not
impaired. This is of greater importance when cannulation of the femoral artery is
needed.
Some recent papers [1519] describe the decision process in choosing a peripheral percutaneous approach, the cannulation techniques available, the equipment
needed, and the procedure itself.
Choosing the vessels for access in VV ECMO must take into account:
The maximum ECMO flow needed for the support of the patient
The maximum recirculation tolerable
The patients comfort
The anatomical difficulties or the presence of some obstructed vein
The size of the drainage cannula is the main determinant of ECMO blood flow,
being flow directly related to the fourth power of cannula radius; the best position
would be in the intrahepatic portion of the inferior vena cava or in the right atrium.
Multiple holes are distributed along the cannulas to enhance blood drainage (multiplestage drainage cannulas). The reimmission cannulas normally have holes only in a
short portion near to their extremity. There is no problem in choosing multistage
[email protected]
40
M. Migliari et al.
Femor-jugular configuration
Reimmission area
Femor-femoral configuration
T12
L1
Dreinage area
Vertebral level correspondence
T11
Hepatic veins
Renal veins
L2
L3
L4
L5
Fig. 4.2 Possible VV cannula configuration according to the type of drainage cannula (multistage
side holes or cannula with side holes close to the tip) and to VV configuration (femoro-femoral or
femoro-jugular). Correspondence between inferior vena cava main branches (hepatic and renal
veins) and vertebral bodies is drawn. Independently from the reimmission cannula, tip of the drainage cannula should be positioned above renal veins, possibly in the intrahepatic portion of inferior
vena cava. Multistage cannula should be used for drainage only especially in the femoro-femoral
approach to minimize blood flow recirculation
cannulas both for drainage and reinfusion if a femoro-jugular approach is applied

(Fig. 4.2), but if a femoro-femoral approach is chosen, a different kind of venous
cannula must be used, because the side holes will generate a very high recirculation
of oxygenated blood from the reimmission cannula to the drainage one (Fig. 4.2).
The configuration with the minimal recirculation is the femoro-jugular one [20].
Recirculation in the femoro-femoral approach can be minimized maintaining the
drainage cannula below the diaphragm, above the renal veins, and the return cannula
in the atrium or just below it (Fig. 4.2).
4.4
Preparation of the Patient for Percutaneous

Cannulation
Before starting the ECMO cannulation procedure, in a nonemergent situation, an

arterial cannula (a right radial arterial catheter if a peripheral VA femoro-femoral
ECMO must be instituted) and a central venous catheter must be inserted for monitoring. Blood specimens are drawn to assess complete blood cells count, basal coagulation profile (PT, aPTT, fibrinogen, d-dimers, and ATIII if available), blood
chemistry, and blood gases. According to the results, a request for packed red blood
cells, platelets, or plasma can be forwarded to the transfusion service.
[email protected]
41
The nurse in charge for the patient prepares the chosen vascular sites according
to the procedure normally used for other central accesses; hair removal is performed
if necessary with a hair clipper maintaining the integrity of the skin.
The cannulation procedure must be accomplished with complete aseptic technique; thus at least two operators will perform surgical hand washing and will be
dressed with maximal sterile barrier precautions, cap, mask, sterile gown, and sterile gloves. The skin at insertion sites will be prepared with chlorhexidine 2 % and
the surgical field prepared with large drapes covering the entire bed allowing space
for cannulas and tubing during the procedures [21]. At this time a shot of antibiotic
(first- or second-generation cephalosporin) is advisable as prophylaxis during the
procedure [22].
The material for cannulation is prepared on a serving trolley and comprises:
A needle for venipuncture
A J-tipped guidewire
Dilatators (multiple or tapered)
Surgical tools
Sutures
4.5
VV ECMO Percutaneous Cannulation
Since the end of the 1980s, the percutaneous approach was introduced and can now
be considered the first-choice technique [8, 1519, 23, 24] for VV bypass. The surgical procedure has been almost completely abandoned, since this technique is
more time-consuming and burdened with complications, uncontrollable bleeding
representing the main one. The main advantage of percutaneous cannulation is a
reduced risk of bleeding, but this technique also allows shorter operative time and a
much easier mobilization and nursing of the patient.
Percutaneous cannulation of the femoral, jugular, and rarely subclavian vein is
described in the literature. Axillary vein cannulation requires always a surgical
technique.
The technique used for percutaneous cannulation is similar to the one introduced
by Seldinger almost 60 years ago; for a detailed description, refer to the Lancet
review published in 2005 [25].
4.5.1
Femoro-Femoral Approach
The two operators will localize the femoral veins below the inguinal ligament, and
the procedure starts with the puncture of the vessels with an 18 G needle under
ultrasound vision. We are used to introduce first an 8 French catheter sheath introducer into the femoral veins using the Seldinger technique and prepare a concentric
purse-string suture around the insertion point in order to limit blood loss during the
multiple dilatations of the vessels. A stainless-steel J-shaped guidewire
(0.038 in. 150180 cm) is passed through the 8 Fr introducer; the guidewire must
[email protected]
42
M. Migliari et al.
be long enough to reach the inferior vena cava. After wire placement, a 2,500
5,000 unit heparin bolus is administered to prevent thrombosis in the cannulas. At
this point vessel dilators of increasing caliber are passed subsequently over the
guidewire in order to obtain the right dilatation for the chosen cannula. To avoid
kinking of the guidewire, it is important that the wire moves freely within the dilator, one operator will maneuver the dilators while the other will maintain the guidewire aligned with the dilator and with a slight tension. To facilitate the dilatation of
the vessel and minimize the risks of guidewire kinking, our group introduced some
years ago a modified technique in which three guidewires were inserted in the
same vessel [26]; a dilator was passed over each wire to obtain proper dilatation for
the chosen cannula (e.g., if a 24 Fr cannula has to be inserted, an 8 Fr dilator was
passed over each guidewire). The development of a single progressive tapered dilator (Dilator Coons Taper 422 Fr, Cook Medical, Bloomington, USA) allows now
to reduce the dilatation step to the passage of a single dilator if a cannula up to
21 Fr must be inserted or a two-step dilatation if a larger cannula is needed. The
quality of the guidewire is also crucial for the success of the maneuver; if the
guidewire is too soft, the risk of kinking while passing the dilators is very high; we
have good results with the use of the Amplatz Super StiffTM Boston Scientific
Guidewire.
After the proper dilatation is achieved, the cannula is inserted over its introducer;
when the right position is achieved, the introducer and the guidewire are removed
and a controlled filling of the cannula with blood is allowed by maintaining the
extremity of the cannula slightly above the bed plane. The drainage cannula is
inserted first and flushed with saline. Then the two operators move to the contralateral site, and the reimmission catheter is inserted with the same technique. Both
cannulas are then secured to the skin at least in two points.
The VV femoro-femoral approach carries a higher risk of blood recirculation
compared to the femoro-jugular access; for this reason it is important to put the tip
of the drainage cannula at the level of L1L2, in order to receive the blood contribution of the renal veins, while the tip of the reimmission cannula should be placed
close to the junction between the inferior vena cava and the right atrium (e.g., at the
level of T10T11). In this way blood recirculation should be acceptably low, just
around 1015 % (Fig. 4.2).
A bedside imaging technique is therefore advisable to control the guidewire
position and its shape during dilatations of the vessel and to guide the correct cannula position. Chest and abdominal x-rays are static and dont allow the rapid
correction of cannula position during the procedure [27]. Ultrasound and fluoroscopy can be used during cannulation to optimize catheter placement. Ultrasounds
are easily accessible at the bedside; patients characteristics and expertise of the
operator are the main determinants of adequate imaging [28]. Fluoroscopy would
be the best imaging technique to visualize guidewire misplacements during cannulation [29] and therefore avoid ECMO cannula malposition, but is rarely available at the bedside and carries the risk of x-ray exposure, and with the new
technological ICU beds, fluoroscopic vision of the entire procedure is sometimes
very difficult.
[email protected]
4.5.2
43
Femoro-Jugular Approach
Another option to perform VV bypass is represented by the femoral-jugular

approach. In this case one operator proceeds with the cannulation of the femoral
vein as described above, and the other acts on the internal jugular vein. Its advisable to drain blood from the femoral cannula positioned in the inferior vena cava in
order to minimize recirculation. The return cannula is positioned through the internal jugular vein proximal to the right atrium. Cannulation of the jugular vein carries
the risk of pneumothorax, and this must be taken into account in choosing this site.
A shorter cannula, if available, must be chosen to allow better fixation. The increasing implementation of VV ECMO in awake spontaneously breathing patients arises
the warning about the risk of air embolism during jugular vein cannulation.
Therefore some groups advocate elective intubation before the procedure and extubation thereafter [30].
4.5.3
Double-Lumen (Avalon) Cannula
The two vessel approaches, femoro-femoral and femoro-jugular, are not comfortable for the patients. Movements are limited, and an increased need of sedatives is
reported.
A single-vessel approach has been recently applied also in the adult population
through a double-lumen cannula available in different sizes ranging from 13 to
31 Fr [3134]. This type of cannula allows both drainage and reinfusion. The cannula has to be introduced through the internal jugular vein, and the placement should
be guided using fluoroscopy and ultrasounds [3537]. The position is crucial; the
cannula must cross the right atrium with the tip in the inferior vena cava. Blood is
drained from both the superior and inferior vena cava while the reinfusion occurs
through a separate lumen into the right atrium just facing the tricuspid valve. This
cannulation seems to have good results [33, 34] and allows physiotherapy with a
walking patient [38].
4.6
VA ECMO Percutaneous Cannulation
Percutaneous femoral cannulation for venoarterial VA ECMO is mainly an emergent

procedure and can be performed everywhere in the hospital [39] and was recently
performed also outside the hospital [40, 41]. While VV cannulation for respiratory
assistance generally allows time for a safe procedure, VA cannulation for cardiac rescue requires to be accomplished in the shortest possible time, exposing to potentially
fatal difficulties and complications. Nevertheless, the percentage of successful cannulation is very high, hovering in many studies over 90 % [42, 43]. We already stated
above that the presence of a cardiothoracic or vascular surgeon is advisable on site
during the procedure; after one or two unsuccessful attempts to locate and puncture
percutaneously the vessels, a switch to an open technique is mandatory.
[email protected]
44
4.6.1
M. Migliari et al.
Implantation Technique
The preliminary identification of the femoral vessels using ultrasound can facilitate
the task and allows a more careful selection of the cannula diameter according to the
size of the vessel. The procedure is best performed with two operators, to control the
cannulas and wires. The placement of femoral catheters is performed under aseptic
technique and begins with percutaneous puncture of the femoral vessels. If the puncture is performed during CPR, both players may act simultaneously, trying to locate
the femoral artery and vein. It is preferable to use both sides for cannulation whenever
possible to minimize the chance of impaired limb perfusion. If time allows it is preferable to perform an ultrasound-guided procedure. Following the Seldinger technique,
a flexible J-tip guidewire (0.038 in. 150180 cm) is advanced from the femoral vein
into the inferior vena cava (IVC) directed toward the right atrium, and an Amplatz
ultra-stiff J-tip guidewire (0.038 in. 180 cm) is advanced from the femoral artery
toward the aortic valve. After wire placement, a 2,5005,000 unit heparin bolus is
administered to prevent thrombosis in the cannulas. Using a single progressive dilator
(Coons Taper 422 F, Cook Medical, Bloomington, USA), the venous and arterial
accesses are progressively dilated and cannulas are subsequently introduced over the
wire. The venous cannula is advanced till the cannula tip is in the mid-right atrium.
The arterial cannula is advanced for its entire length into the iliac artery. The wires are
removed and the extremities of the cannulas are clamped. The lines are de-aired and
connected to the ECMO circuit. Finally the cannulas are secured to the skin with
sutures. The mean cannulation time is around 30 min in our series with a learning
curve that determined a reduction of the time from 46 min for the first 5 patients in
2008 to 29 min for the last 15 patients in 2012. If the procedure is nonemergent, a
distal perfusion catheter, or at least the guidewire, is positioned in the superficial femoral artery before the insertion of the arterial cannula. Distal perfusion of the leg is a
simple and well-accepted method to increase the circulation of the cannulated leg, and
several criteria to detect leg ischemia were developed [4449]. In an elective
VA ECMO procedure, the distal perfusion catheter is inserted before arterial ECMO
cannula placement, since residual pulsation of the distal artery allows its easier location. In an emergent situation, like during CPR, there is no time to insert the distal
perfusion catheter electively and the maneuver is deferred starting a strict control of
the limb perfusion. In our experience a distal perfusion catheter to prevent leg ischemia is placed when a Doppler examination of the arteries of the leg does not detect
the presence of an adequate flow downstream of the arterial cannula. We choose a
68 F 11 cm introducer, (Avanti+, Cordis, LJ Roden, Netherlands) as the distal perfusion catheter. This can be placed percutaneously under US guidance and connected
via suitable connectors to the arterial limb of the ECMO circuit.
4.7
Explantation Technique (VA and VV)
Explantation of the cannulas is performed in a standardized procedure at the bedside.

If an arterial cannula is in place, heparin infusion is withdrawn and coagulation assessed
to ensure the return to baseline values; in the case of VV ECMO, heparin can be
[email protected]
45
reduced, but there is no need to have a perfectly normal coagulation before deconnection. After the placement of a purse-string suture around the insertion sites, the cannulas are removed allowing a small leakage of blood to clear small thrombi from the
distal portion of the leg involved. The venous site is manually compressed for 10 min,
and then a slightly compressive medication is applied on cannulation site. On the arterial side, after 3045 min of manual compression, a femoral compression system
(Safeguard 24 cm. Maquet, Hirrlingen, Germany) is applied to ensure a pneumatic
compression over the vessel puncture to induce hemostasis. Some ECMO centers
advocate the open repair of the artery also if it was percutaneously cannulated. Strict
control of the groin is implemented to recognize the development of hematoma.
A follow-up vascular ultrasound of the involved vessel is recommended; the possibility of a retained fibrin sleeve after cannula removal [50] is described and can
modify the anticoagulation requirements of the patient after ECMO decannulation.
4.8
Complications of ECMO Percutaneous Cannulation
Different complications can ensue with the cannulation of the vessel for ECMO.
Early complications are directly related to the implantation procedure. Guidewire
kinking, losing the vessel after full dilatation is achieved, vascular tears, not being
able to advance the guidewire to the correct position due to anomalous bifurcation
of the vessel, intimal dissection, and perforation are described.
Right ventricular rupture with cardiac tamponade [51] and myocardial infarction
[52] are reported as complications with the insertion of the double-lumen cannula.
Bleeding from cannulation sites is still the most common complication reported
in the literature [53]; with its occurrence techniques to improve vascular site hemostasis are implemented [54].
Ischemic alteration of the leg distal to arterial cannula is well described in
VA ECMO, and a reperfusion cannula can be inserted electively to reduce the risk
of limb ischemia. Strict monitoring of perfusion also in the venous side must be
performed because a compartment syndrome can develop if the venous drainage
from the distal leg is impaired by the big drainage cannula and edema develops due
to a shock state. The frequency of short-term complications, including groin hematoma, pseudoaneurysm, artero-venous fistulae, and acute thromboembolism, varies
between 2 % in VV ECMO and 8 % in VA ECMO,
Late complications affect around 12 % of VA ECMO patients and are mainly due
to stenosis of the femoral artery at the former cannulation site, in particular with the
surgical approach. Limb compartment syndrome occurs in about 1 % of VA ECMO
patients and represents a very severe complication that can lead to amputation if not
promptly recognized and treated [55, 56].
References
1. Hill JD, OBrien TG, Murray JJ et al (1972) Prolonged extracorporeal oxygenation for acute
post-traumatic respiratory failure (shock-lung syndrome). Use of the Bramson membrane
lung. N Engl J Med 286:629634
[email protected]
46
M. Migliari et al.
2. Zapol WM, Snider MT, Hill JD et al (1979) Extracorporeal membrane oxygenation in severe
acute respiratory failure. A randomized prospective study. JAMA 242:21932196
3. Gattinoni L, Kolobow T, Agostoni A et al (1979) Clinical application of low frequency positive
pressure ventilation with extracorporeal CO2 removal (LFPPV-ECCO2R) in treatment of adult
respiratory distress syndrome (ARDS). Int J Artif Organs 2:2822833
4. Gattinoni L, Pesenti A, Bombino M et al (1993) Role of extracorporeal circulation in adult
respiratory distress syndrome management. New Horiz 1:603612
5. Pesenti A, Kolobow T, Riboni A (1982) Single vein cannulation for extracorporeal respiratory
support. In: ESAO proceedings, Bruxelles, pp 6567
6. Pesenti A, Romagnoli G, Fox U (1983) Sapheno-saphenous cannulation for LFPPV-ECCO2R.
In: 10th congress of the European Society of Artificial Organs, Bologna
7. Maif P, Hoermann C, Moertl M et al (1996) Percutaneous venoarterial extracorporeal membrane oxygenation for emergency mechanical circulatory support. Resuscitation 33:2934
8. Pranikoff T, Hirschl R, Remenapp R et al (1999) Venovenous extracorporeal life support via
percutaneous cannulation in 94 patients. Chest 115:818822
9. Pesenti A, Gattinoni M, Bombino M (2013) Extracorporeal carbon dioxide removal. In: Tobin
MJ (ed) Principles and practice of mechanical ventilation, 3rd edn. McGraw-Hill Companies
Inc, New York, pp 543554
10. Kohler K, Valchanov K, Nias G, Vuylsteke A (2013) ECMO cannula review. Perfusion
28:114124
11. Paulsen MJ, Orizondo R, Le D et al (2012) A simple, standard method to characterize pressure/flow performance of vascular access cannulas. ASAIO J 59:2429
12. Ngatchou W, Ramadan ASE, Van Nooten G, Antoine M (2012) Left tilt position for easy
extracorporeal membrane oxygenation cannula insertion in late pregnancy patients. Interact
Cardiovasc Thorac Surg 15:285287
13. Weiner MM, Geldard P, Mittnacht AJ (2013) Ultrasound-guided vascular access: a comprehensive review. J Cardiothorac Vasc Anesth 27:345360
14. Troianos CA, Hartman GS, Glas KE et al (2011) Guidelines for performing ultrasound guided
vascular cannulation: recommendations of the American Society of Echocardiography and the
Society of Cardiovascular Anesthesiologists. J Am Soc Echocardiogr 24:12911318
15. Field ML, Al-Alao B, Mediratta N, Sosnowski A (2006) Open and closed chest extrathoracic
cannulation for cardiopulmonary bypass and extracorporeal life support: methods, indications,
and outcomes. Postgrad Med J 82:323331
16. Stulak JM, Dearani JA, Burkhart HM et al (2009) ECMO cannulation controversies and complications. Semin Cardiothorac Vasc Anesth 13:176182
17. Sidebotham D, McGeorge A, McGuinness S et al (2010) Extracorporeal membrane oxygenation for treating severe cardiac and respiratory failure in adults: part 2-technical considerations. J Cardiothorac Vasc Anesth 24:164172
18. Ganslmeier P, Philipp A, Rupprecht L et al (2011) Percutaneous cannulation for extracorporeal
life support. Thorac Cardiovasc Surg 59:103107
19. Sidebotham D, Allen SJ, McGeorge A et al (2012) Venovenous extracorporeal membrane
oxygenation in adults: practical aspects of circuits, cannulae, and procedures. J Cardiothorac
Vasc Anesth 26:893909
20. Rich PB, Awad SS, Crotti S et al (1988) A prospective comparison of atrio-femoral and
femoro-atrial flow in adult venovenous extracorporeal life support. J Thorac Cardiovasc Surg
116:628632
21. OGrady NP, Alexander M, Burns LA et al (2011) Guidelines for the prevention of intravascular catheter-related infections. Clin Infect Dis 52:e162e193
22. Kao LS, Fleming GM, Escamilla RJ et al (2011) Antimicrobial prophylaxis and infection
surveillance in extracorporeal membrane oxygenation patients: a multi-institutional survey of
practice patterns. ASAIO J 57:231238
23. Annich G, Lynch W, MacLaren G, Wilson J, Bartlett R (eds) (2012) ECMO extracorporeal
cardiopulmonary support in critical care, 4th edn. Extracorporeal Life Support Organization,
Ann Arbor
[email protected]
47
24. Brodie D, Bacchetta M (2011) Extracorporeal membrane oxygenation for ARDS in adults. N
Engl J Med 365:19051924
25. Higgs ZC, Macafee DA, Braithwaite BD, Maxwell-Armstrong CA (2003) The Seldinger technique: 50 years on. Lancet 366:14071409
26. Grasselli G, Pesenti A, Marcolin R et al (2010) Percutaneous vascular cannulation for extracorporeal life support (ECLS): a modified technique. Int J Artif Organs 33:553557
27. Barnacle AM, Smith LC, Hiorns MP (2006) The role of imaging during extracorporeal membrane oxygenation in pediatric respiratory failure. AJR Am J Roentgenol 186:5866
28. Platts DG, Sedgwick JF, Burstow DJ et al (2012) The role of echocardiography in the management of patients supported by extracorporeal membrane oxygenation. J Am Soc Echocardiogr
25:131141
29. El-Kayali A (2004) Insertion of haemodialysis catheters: fluoroscopy guided placement technique for malpositioned wires. Internet J Nephrol. doi:10.5580/2606
30. Extracorporeal Life Support Organization (ELSO) (2009) General guidelines for all ECLS
cases. http://www.elso.med.umich.edu/Guidelines.html. Accessed 10 Sep 2013
31. Wang D, Zhou X, Liu X et al (2008) Wang-Zwische double lumen cannula-toward a percutaneous and ambulatory paracorporeal artificial lung. ASAIO J 54:606661
32. Javidfar J, Brodie D, Wang D et al (2011) Use of bicaval dual-lumen catheter for adult venovenous extracorporeal membrane oxygenation. Ann Thorac Surg 91:17631769
33. Bermudez CA, Rocha RV, Sappington PL et al (2010) Initial experience with single cannulation for venovenous extracorporeal oxygenation in adults. Ann Thorac Surg 90:991995
34. Camboni D, Philipp A, Lubnow M et al (2012) Extracorporeal membrane oxygenation by
single-vessel access in adults: advantages and limitations. ASAIO J 58:616621
35. Trimlett RH, Cordingley JJ, Griffiths MJ et al (2011) A modified technique for insertion of
dual lumen bicaval cannulae for venovenous extracorporeal membrane oxygenation. Intensive
Care Med 37:10361037
36. Javidfar J, Wang D, Zwischenberger JB et al (2011) Insertion of bicaval dual lumen
extracorporeal membrane oxygenation catheter with image guidance. ASAIO J 57:
203205
37. Dolch ME, Frey L, Buerkle MA et al (2011) Transesophageal echocardiography-guided technique for extracorporeal membrane oxygenation dual-lumen catheter placement. ASAIO J
57:341343
38. Turner DA, Cheifetz IM, Rehder KJ et al (2011) Active rehabilitation and physical therapy
during extracorporeal membrane oxygenation while awaiting lung transplantation: a practical
approach. Crit Care Med 39:25932598
39. Feindt P, Benk C, Boeken U et al (2011) Use of extracorporeal circulation (ECC) outside the
cardiac operating room: indications, requirements and recommendations for routine practice.
Thorac Cardiovasc Surg 59:6668
40. Lebreton G, Pozzi M, Luyt CE et al (2011) Out-of-hospital extra-corporeal life support
implantation during refractory cardiac arrest in a half-marathon runner. Resuscitation
82:12391242
41. Artl M, Philipp A, Voelkel S et al (2011) Out-of-hospital extracorporeal life support for cardiac
arrest-A case report. Resuscitation 82:12431245
42. Kagawa E, Inoue I, Kawagoe T et al (2010) Assessment of outcome and differences between
in- and out-of-hospital cardiac arrest patients treated with cardiopulmonary resuscitation using
extracorporeal life support. Resuscitation 81:968973
43. Avalli L, Maggioni E, Formica F et al (2012) Favourable survival of in-hospital compared to
out-of-hospital refractory cardiac arrest patient treated with extracorporeal membrane oxygenation: an Italian tertiary care centre experience. Resuscitation 83:579583
44. Huang SC, Yu HY, Wj K et al (2004) Pressure criterion for placement of distal perfusion catheter to prevent limb ischemia during adult extracorporeal life support. J Thorac Cardiovasc
Surg 128:776777
45. Wong JK, Smith TN, Pitcher HT et al (2012) Cerebral and lower limb near-infrared spectroscopy in adults on extracorporeal membrane oxygenation. Artif Organs 36:659667
[email protected]
48
M. Migliari et al.
46. Rao AS, Pellegrini RV, Speziali G, Marone LK (2010) A novel percutaneous solution to limb
ischemia due to arterial occlusion from a femoral artery ECMO cannula. J Endovasc Ther
17:5154
47. Madershahian N, Nagib R, Wippermann J et al (2006) A simple technique of distal limb perfusion during prolonged femoro-femoral cannulation. J Card Surg 21:168169
48. Lamb KM, Hirose H, Cavarocchi NC (2013) Preparation and technical considerations for percutaneous cannulation for veno-arterial extracorporeal membrane oxygenation. J Card Surg
28:190192
49. Schwarz B, Mair P, Margreiter J et al (2003) Experience with percutaneous venoarterial cardiopulmonary bypass for emergency circulatory support. Crit Care Med 31:758764
50. Bouchez S, Mackensen GB, De Somer F et al (2012) Transesophageal echocardiographic
image of a retained fibrin sleeve after removal of a venous extracorporeal membrane oxygenation cannula. J Cardiothorac Vasc Anesth 26:883886
51. Hirose H, Yamane K, Marhefka G, Cavarocchi N (2012) Right ventricular rupture and tamponade caused by malposition of the Avalon cannula for venovenous extracorporeal membrane
oxygenation. J Cardiothorac Surg 7:36
52. Reis Miranda D, Dabiri Abkenari L, Nieman K et al (2012) Myocardial infarction due to malposition of ECMO cannula. Intensive Care Med 38:12331234
53. Paden ML, Conrad SA, Rycus PT et al (2013) Extracorporeal life support organization registry
report 2012. ASAIO J 59:202210
54. Lamb KM, Pitcher HT, Cavarocchi NC, Hirose H (2012) Vascular site hemostasis in percutaneous extracorporeal membrane oxygenation therapy. Open Cardiovasc Thorac Surg
J 5:810
55. Bisdas T, Beutel G, Warnecke G et al (2011) Vascular complications in patients undergoing
femoral cannulation for extracorporeal membrane oxygenation support. Ann Thorac Surg
92:626631
56. Zimpfer D, Heinisch B, Czerny M et al (2006) Late vascular complications after extracorporeal membrane oxygenation support. Ann Thorac Surg 81:892895
[email protected]
Surgical Cannulation: Indication,

Technique, and Complications
Francesco Formica, Silvia Mariani, and Giovanni Paolini
5.1
Introduction
5.1.1
Surgical Methods and Vascular Access:

The Decision-Making Process
In the modern era of extracorporeal life support, ECMO has been increasingly
applied in several situations such as cardiac emergencies, cardiac surgery complications, or respiratory failure. The establishment of ECMO support could be achieved
through intrathoracic or extrathoracic percutaneous or surgical cannulation strategies. The best cannulation technique should be chosen on the basis of patients and
the clinical settings. First, it is necessary to define needs and goals of the ECMO and
review the access options for the specific support and clinical setting. Such a decision process will lead the medical team to choose between a veno-venous and venoarterial extracorporeal circulation and between a central and peripheral cannulation.
The following step requires a rapid assessment of benefits and risks of the selected
options to pick the best site and strategy of the cannulation. In case of central cannulation, the surgical approach through a sternotomy is required. As far as peripheral cannulation, it is necessary to choose among several sites: femoral vessels,
axillary vessels, and cervical vessels are the most used ones. The cannulation could
be achieved through a percutaneous, a semi-open, or a surgical approach. In the end,
it is necessary to foresee and prevent complications of every chosen strategy.
F. Formica (*) S. Mariani G. Paolini

Department of Surgical Science and Translational Medicine,
Cardiac Surgery Clinic, San Gerardo Hospital,
University of Milano-Bicocca, Via Pergolesi, 33, Monza 20900, Italy
e-mail: [email protected]; [email protected]; [email protected]
[email protected]
49
50
5.1.2
F. Formica et al.
Open Versus Percutaneous Cannulation:

When Surgical Approach Is Recommended
Peripheral vessel cannulation could be achieved either by percutaneous procedures

or surgical incisions through an open approach, a semi-Seldinger, or a full Seldinger
method. The open approach allows the visualization of the vessels to guarantee
their adequate size, direct placement of a purse-string suture on the chosen vessel,
confirmation of proper cannula placement, and good hemostasis. In addition, the
vessels can be assessed for atherosclerosis, calcific disease, aneurysmal disease,
and thrombosis, allowing to choose an alternative cannulation site. Furthermore,
the surgeon has the option to anastomose a polytetrafluoroethylene (PTFE) or
Dacron graft to obtain a safer and more efficacious cannulation with lower arterial
line pressures and lower chance of malperfusion of the distal limb. The semiSeldinger technique has the advantage of guidewires and dilators with no need of
surgical arteriotomy or venotomy; at the same time it allows the assessment of the
vessels, the placement of purse-string sutures, and a better hemostasis through a
small incision. It is particularly attractive because it has the main features of the
open cannulation and the Seldinger methodology, though it maintains a minimally
invasive approach. For the venous cannulation, an entirely percutaneous Seldinger
approach is generally possible. Axillary vessels represent an exception because
they always need an open access. However, peripheral cannulation might be difficult during external cardiac massage or emergencies; so if arterial or venous percutaneous access fails or leads to complications, it is recommended to surgically
expose the vessels and cannulate them in an open or semi-Seldinger approach.
Moreover, in cases of surgical emergencies, where the peripheral vessels are
already exposed, the vessels might be considered for an open or semi-Seldinger
cannulation. In addition, if a vascular disease is suspected, the open evaluation of
the vessels is advised.
5.1.3
Peripheral Versus Central Cannulation: Differences

in Blood Flow and Mixed Cannulations
ECMO cannulation could be undertaken through peripheral vessel cannulation or

central cannulation where the ascending aorta and right atrium are directly cannulated. Such an approach allows a better drainage and flow because of the use of a
cannula with a larger diameter, that is why it is recommended in patients with a
larger body surface area (>2.0 m2). The other advantage of central cannulation is
that it provides anterograde flow to the arch vessels, coronaries, and the total body.
On the other hand, the peripheral cannulation on femoral vessels provides a
retrograde aortic flow that blends in the arch with the blood that comes from the
heart. During central cannulation, if there is no native lung function, this mixing
[email protected]
Surgical Cannulation: Indication, Technique, and Complications
51
results in PCO2 40.5, PO2 100, and sat 98 %. Throughout severe respiratory failures, the blood out of the heart is desaturated and perfuses the aortic arch and coronaries. On the other hand, the saturated inflow blood perfuses the lower 2/3 of the
body [1]. This is the reason why the monitoring of oxygenation from a right radial
arterial line would be highly suggested. Moreover, such a slow flow in the ascending
aorta could lead to clot development in the aortic root, especially if the heart is not
pumping or if it is pumping only a small amount of blood. Cannulation on axillary
artery or cervical vessels might provide anterograde flow with the advantages of a
central cannulation and a peripheral approach. A mixed cannulation approach is
possible. The medical team could choose to put both a peripheral cannula and a
central one if needed.
5.2
Central Cannulation
5.2.1
Indications
Central cannulation for ECMO could be achieved through intrathoracic cannulation

of the right atrium and the ascending aorta (Fig. 5.1). Such an approach requires an
accurate surgical field, an open-chest procedure, a sternotomy and a surgical preparation of cannulation sites. The features listed above make the central cannulation
the best approach as far as cases that require intraoperative support because of cardiopulmonary bypass wean failure. During open-chest cardiac surgery, central cannulation is often already set up, and the switch from cardiopulmonary bypass to
ECMO with central cannulation is an easy and safe procedure. In addition, during
the early postoperative period, chest cannulation after re-thoracotomy could ensure
a rapid start of cardiopulmonary support, especially in patients who show with cardiac arrest, cardiogenic shock, or emergent chest reexploration in the intensive care
unit. Central cannulation could be helpful in patients with severe peripheral disease
requiring temporary circulatory support too. If no peripheral percutaneous or surgical cannulation could be obtained, an open-chest approach might be considered as
well. Peripheral ECMO could be switched to central ECMO when insufficient left
ventricular unloading, insufficient blood flow, or peripheral complications are
detected. The central cannulation offers an optimal arterial anterograde flow and the
opportunity to use cannulas with larger internal diameters so as to obtain a lower
blood flow resistance and a better venous drainage. The choice of cannula size is
based on the patients body surface area and the expected blood flow. The arterial
cannula usually has a 2022-Fr diameter, whereas the venous one is often a 5052Fr cannula. The single-stage venous cannula is preferred to a cavo-atrial venous one
because it avoids ulcers and injuries to the inferior vena cava whenever ECMO is
maintained for several days. Bicaval cannulation does not have any indication in
ECMO patients.
[email protected]
52
F. Formica et al.
Fig. 5.1 Central cannulation.

Ao aorta, PV pulmonary vein,
LA left atrium, RV right
ventricle, LV left ventricle
Arterial
cannula
AO
PV
LA
LV
RV
Venous
cannula
5.2.2
Apical
vent
Surgical Technique
The site for the cannulation of the aorta is proximal to the origin of the innominate
artery on the anterior aortic surface (Fig. 5.1). Two purse-string sutures are placed,
a stab wound is made within the sutures, and the cannula is then placed in the distal
ascending aorta. The two purse-string sutures are secured through two tourniquets
that will be kept in the patients thorax until decannulation (Fig. 5.2). The cannula
tip must be completely within the lumen and positioned in order to direct the flow
to the mid transverse aorta. It is suggested to place the cannula in an area where the
atherosclerotic disease is minimal or absent.
The venous cannula is introduced through the right atrial appendage so as to
allow its tip to rest in the mid-right atrium (Fig. 5.1). Before the cannula placement,
a single purse-string suture is positioned, the right atrial appendage is opened, and
the cannula is then inserted. Again, the suture is secured with a tourniquet. During
venous cannulation, the surgeon should be careful about displacement or damage of
central venous or pulmonary arterial monitoring catheters; conversely, catheters
could compromise the function of venous cannula.
If an ECMO support is required after a case of difficult weaning from cardiopulmonary bypass, arterial and venous cannulas are already placed. The surgeon should
[email protected]
53
Fig. 5.2 Central cannulation:

the arterial cannula. The two
purse-string sutures are
secured through two
tourniquets that will be kept
in the patients thorax
Fig. 5.3 Central cannulation.

An occlusive dressing is
applied to the anterior chest
wall to guarantee the
protection of the mediastinic
structures
maintain the cardiopulmonary bypass cannulas, but he should change the circuit.
First, it is necessary to stop the extracorporeal circulation, and after that to clamp
the cannulas, deconnect them from the cardiopulmonary bypass circuit, connect
them to the ECMO circuit, remove the clamps from cannulas, and start the new
extracorporeal support.
In the end, the surgeon could introduce an apical or pulmonary vent (Fig. 5.1) to
unload the left ventricle and could insert a left atrial pressure monitoring line
through the right superior pulmonary vein.
At the end of a cardiac surgery intervention, after an accurate hemostasis of the
heart and other tissues, mediastinal and pleural drainage tubes are positioned and
the sternum is usually closed. In patients undergoing ECMO support, chest closure
is not always possible. In most of such cases, it might happen that the sternum could
not be closed because of the depressed cardiac function and the presence of the cannulas. When the sternum, subcutaneous tissue, and skin are left unsutured, an occlusive dressing is applied to the anterior chest wall (Fig. 5.3). Whenever possible, the
skin can be sutured with Donati stitches, or a sheet of artificial tissue can be sutured
[email protected]
54
F. Formica et al.
to the skin edges so as to protect mediastinal tissues and prevent infections. However,
new cannulas have been designed so as to be tunneled to the subcostal abdominal
wall allowing the chest to be completely closed [2].
When decannulation is required, the patient is conducted to the operating room
where the chest will be opened again. The cannulas are removed while the pursestring sutures are tied with a hemostatic effect. Primary chest closure is determined
at the discretion of the surgeon.
5.2.3
Perioperative Management
The proper position of the cannula is essential to provide an adequate ECMO blood
flow. After central cannulation, intraoperative transesophageal echocardiography is
the first technique used to confirm appropriate cannula positions and ventricular
decompression before chest closure. During postoperative days, routine chest radiograph could be used to detect any change in the positions of the cannulas.
Nursing care for patients with ECMO central cannulation is more complex than
the one that deals with peripherally cannulated patients. A special care must be
provided to the chest dressing and cannula handling. Chest drainage tubes must be
monitored very often. Every change of the position of the patient must be carefully
accomplished, and patients transport is more difficult than peripherally cannulated
patients.
5.2.4
Complications and Disadvantages
Despite the advantages of the central cannulation, it is though related to important

complications during cannulation and postoperative period. Major complications of
ascending aorta cannulation include injury or dissection of the aortic anterior and
posterior wall; misplacement of the cannula tip against the aortic wall, towards the
valve, or in an arch vessel; emboli; and inadequate or excessive cerebral flow.
Otherwise, venous atrial cannulation is associated to atrial arrhythmias, atrial or
caval tears and bleeding, and air embolization. Local complications that may occur
during and after decannulation also include bleeding, atrial or aortic injuries, and
pseudoaneurysm.
Limb ischemia is generally associated to peripheral cannulation, but it can also
be observed with the central cannulation. In such a situation, peripheral ischemia is
no local complication but more likely an embolic phenomenon associated with the
presence of aortic atheromas.
The main drawbacks of central cannulation are the open chest and the high risk
of bleeding. Patients with central cannulation, compared to patients with peripheral
cannulation, had a 6-fold higher rate of reoperation and a 3-fold higher rate of
bleeding from cannulation sites [3]. Such conditions can be explained because the
cannulas are inserted into a constantly moving organ, and the patients had a sternotomy with large areas of raw tissue. The need for repeated operations and the
[email protected]
55
evacuation of bleeding with the central cannulation contribute to the increase in

terms of costs and risks for the patient [3]. Mediastinitis is more frequent in central
cannulation because of the difficult management of the open chest. Moreover,
patients with central cannulation could not be extubated; transports and nursing care
are more difficult as already reported.
5.3
Peripheral Cannulation
5.3.1
Indications
ECMO is a well-established treatment also when cardiac surgery is not involved.

Peripheral cannulation does not require an open-chest approach so it results in a
quicker procedure. It is useful if immediate support is needed and if ECMO is
started in different hospital situations. It is recommended in case of primary cardiogenic shock, acute myocardial infarction, cardiopulmonary arrest, high-risk PTCA,
myocarditis and cardiomyopathy, pulmonary hypertension, intractable arrhythmias,
and respiratory failure. Moreover, peripheral cannulation allows easier nursing
cares and easier and safer transports of the patients, and the patients can be extubated even if ECMO support is still ongoing. The cervical vessels and the brachiocephalic artery are the best peripheral cannulation sites in neonates and children
weighing less than 15 kg. Groin cannulation of the common femoral artery and vein
often provides efficient venous drainage and perfusion for larger children and adults.
The axillary and iliac arteries represent additional sites that could be used.
5.3.2
Femoral Vessels
5.3.2.1 Femoral Artery: Surgical Technique

Femoral artery cannulation is probably the most common approach for ECMO
implantation because of its ease and the large diameters of this vessel. Femoral
cutdown is the traditional surgical approach (Fig. 5.4a). A transverse or longitudinal
skin incision is made over the femoral vessels and below the inguinal ligament.
Dissection is undertaken so as to isolate the femoral artery and vein, proximal and
distal control of each vessel is obtained through loops, and a purse-string suture is
placed. After heparinization, the common femoral artery is clamped proximally and
distally, and a transverse arteriotomy is made, leaving the posterior one third of the
artery intact. A 1521-Fr arterial cannula is directly inserted. Some authors prefer
the use of an 810-mm PTFE or Dacron chimney graft sewn in an end-to-side
way on the arterial vessel [4] (Fig. 5.4c). Afterwards, the chimney graft can be tunneled under the skin or extended over it; the cannula is then inserted and the wound
is closed. The cannula position must be fixed to the skin with multiple ligatures.
Such a technique is recommended in patients with small vessels likely to be occluded
by arterial cannulas (children, thin adults, peripheral artery disease), patients in
whom ischemia develops after percutaneous cannulation, or patients who undergo
[email protected]
56
Fig. 5.4 Femoral vessel
surgical cannulation. (a)
Direct cannulation of the
femoral vessels after surgical
isolation of the vessels. (b)
Semi-Seldinger technique:
the cannulas are introduced
via a separate stab incision
2 cm distal to the main
incision. (c) Chimney graft
technique: a PTFE or Dacron
vascular graft is sewn in an
end-to-side way on the
femoral artery. FA femoral
artery, FV femoral vein
F. Formica et al.
FV
FA
FV
FA
c
Vascular
graft
FV
FA
cannulation in the operating room with vessels already exposed for CPB [5, 6]. The
technique lowers the risk of distal leg ischemia and vessel dissection and simplifies
decannulation. Unfortunately, this approach requires a longer preparation time, and
it cannot be used in case of emergency. In such a situation, a pre-sealed short vascular prosthesis can be bevelled at its distal end and passed over the cannula. The
cannula is then inserted into the vessel, and when ECMO is stable the vascular
prosthesis around the cannula is lowered onto the femoral artery. In this moment,
the prosthesis is anastomosed to the artery, and the arterial cannula can be
[email protected]
57
withdrawn carefully, positioning its distal end within the prosthesis at the level of
the anastomosis [7]. In the semi-Seldinger technique (Fig. 5.4b), a small transverse
incision is made over the femoral artery and deepened so as to expose a short length
of the common femoral artery. The arterial puncture needle is introduced via a separate stab incision 2 cm distal to the incision, and it is inserted into the artery under
direct vision. The following steps are the same as described for Seldinger technique.
A purse-string suture on the artery could be used to assure hemostasis, and the
wound is then closed. This method of insertion allows the arterial cannula to lie
nearly parallel to the artery and prevents over-angulation and kinking at the
entry site.
After weaning from bypass, the operative field is opened again. The cannulas are
removed and the cannulation sutures tied; running or interrupted sutures are placed
as appropriate. To ensure peripheral perfusion, the arterial vessel is palpated distal
to the former cannulation site immediately after decannulation. In case of distal
malperfusion or pulselessness, reconstructive procedures are performed. If a chimney graft is present, after the cannula removal, the side graft can be closed with
staples or a snare about 1 cm above the anastomosis and the residual part can be cut.
If ECMO is still necessary, a new graft may be sutured to the stump of the old graft
[4]. In cases of percutaneous cannulation, some authors suggest an open repair of
the vessel in order to prevent bleeding and hematoma development [8]. Open repair
of vessels also allows evaluation for potential areas of stenosis and ability to perform patch angioplasty if needed [8].
5.3.2.2 Femoral Vein: Surgical Technique

Femoral vein cannulation is more often percutaneously performed. Surgical isolation is performed when percutaneous approach fails or if femoral vessels are already
exposed. Some authors suggest the cannulation of right femoral vein because of the
easier placement of the cannula up to the right atrium, due to the relationship of the
iliac vein and inferior vena cava to the iliac crest [9]. Some others suggest the cannulation of the contralateral femoral vein as opposite to the cannulated femoral
artery to avoid a cluttered field [10]. The skin incision and the dissection techniques
are the same as described above. Venous cannulation is performed first. A circumferential purse-string suture is placed in the anterior wall of the vein; the common
femoral vein is then clamped proximally and distally. A venotomy is made within
the purse string, the venous cannula is inserted, and the proximal and distal clamps
are removed. The cannula is finally advanced up to the right atrium. Once properly
positioned, the venous cannula is fixed, the arterial cannulation is performed, and
the wound is closed. Cannula diameter size is from 19 to 25 Fr.
5.3.2.3 Limb Ischemia and Distal Perfusion: Open Access
Ischemic complications using different approaches for femoral cannulation varied
between 10 and 70 % [11]. Peripheral arterial disease is an independent predictor of
vascular complications, and assessment of anklebrachial index before ECMO
implantation is recommended whenever possible [11]. During ECMO support a
nursing protocol for early detection of vascular complication is suggested. Bisdas
[email protected]
58
F. Formica et al.
et al. [11] established the following protocol: (1) clinical examination of feet and
legs for temperature, color, capillarity, and compartment syndrome and, additionally, continuous measurement of oxygen saturation of toe and comparison with the
respective finger; (2) Doppler detection of peripheral pulses/arterial blood flow by
an experienced physician every 6 h; (3) measurement of myoglobin and creatine
kinase every 8 h; and (4) if items 13 reveal any problems, color Duplex ultrasonography is used, or in case of inconsistent results, contrast-enhanced computed tomography angiography. The same protocol is followed during the first 48 h after ECMO
explantation. Therefore, distal limb perfusion is critical, and many authors recommend its implantation at the time of ECMO start or following specific criteria.
Huang et al. [12] measured the mean arterial pressure of the superficial femoral
artery by puncturing the vessel distal to the ECMO cannula with a 23-gauge needle.
If the pressure was under 50 mmHg, a distal perfusion cannula was recommended.
Similarly, a larger cannula size-to-BSA ratio could predict which patients could
develop ischemia [13]. Surgical approach to solve such a problem may interest the
femoral artery, the tibial artery, or the dorsalis pedis artery. If the femoral vessels are
already isolated, the chimney graft anastomosed to the femoral artery could guarantee the distal perfusion of the leg. Otherwise, a cannula can be placed in the distal
femoral artery or in the superficial femoral artery with the same technique described
before [1416].
When a severe atherosclerotic lesion or anatomical problems affect the distal
common femoral artery or the superficial femoral artery, distal perfusion through
the posterior tibial artery and the dorsalis pedis artery may be an effective alternative treatment. For the posterior tibial cannulation, a longitudinal incision of 5 cm in
length is made just posterior to the medial malleolus. The fibrous flexor retinaculum
is divided, and a small self-retaining retractor is inserted. The posterior tibial artery
is easily identified by its venae comitantes, and it is circumferentially isolated. The
tibial nerve is preserved. The artery is ligated distally and a small arteriotomy is
created. A cannula is then inserted in a retrograde direction [17]. For the dorsalis
pedis artery cannulation, a small incision is made on the dorsal portion of the foot.
After isolation of the deep peroneal nerve, the artery is isolated and cut down and
the distal side is ligated. An intravascular catheter is inserted into the artery, and,
after the catheter is connected to the tubing of the arterial branch of the circuit,
blood flow to the dorsalis pedis artery is initiated [18].
Another potential complication of femoral cannulation is venous congestion of
the leg by the venous cannula [19]. Superficial femoral vein could be cannulated
percutaneously or through a surgical approach if femoral vessels are already
exposed [14].
5.3.2.4 Complications
The main complication of femoral vessel cannulation is distal limb ischaemia and
reperfusion injury even leading to limb amputation. However, other morbidities
include pseudoaneurysm, neurological injury and femoral nerve weakness, compartment syndrome, retrograde arterial dissection, arterial and venous laceration or perforation, arterial thrombosis and deep vein thrombosis, embolization of luminal
[email protected]
59
debris, arteriovenous fistula, and wound complications such as lymphocele, infection, and hematoma [20, 21]. Femoral cannulation site wound healing could be problematic in the immunocompromised, malnourished, vasculopathic, obese, or diabetic
patients [22]. Common femoral vein cannulation could be complicated by an inability to negotiate the cannula successfully across its long course to the right atrium,
pelvic venous injury, and retroperitoneal bleeding. It is also contraindicated in the
presence of an inferior vena cava filter, deep venous thrombosis, or other intrinsic or
extrinsic obstruction to the pelvic veins or inferior vena cava [22]. Cannula displacement or dislocation is possible during handling of the cannulas. Transesophageal
echocardiography and routine chest radiograph could be used to detect any change in
the positions of the cannulas. Anti-infective and graft anastomosis management
should include a sterile, graft-covering dressing with daily monitoring and skin disinfections with chlorhexidine every 2 days, concomitant antibiotic therapy extended
to day 3 after graft closure and minimization of hip flexion [5, 23]. The monitoring
of the surgical wounds and interested limbs should be extended to days that follow
the decannulation so as to verify if a surgical reexploration is needed.
5.3.3
Axillary Vessels
5.3.3.1 Indications
Even if femoral vessels are the first choice for peripheral cannulation, different sites
could be considered. The use of the right axillary artery as inflow in the ECMO
circuit has several potential advantages:
It provides an antegrade flow and excellent upper body oxygenation.
It is easy and reproducible.
It is a safe procedure with low complication rates.
It may avoid cerebral embolization, and this artery is usually free from arteriosclerotic disease.
It allows closure of the chest after postcardiotomy shock and an easy decannulation [24].
It has a lower rate of complications of arm ischemia because the axillary artery
benefits from rich collateral flow from the thyrocervical trunk to the suprascapular and transverse cervical arteries [25].
Therefore, axillary cannulation could turn to an option for the following: postcardiotomy patients, patients presenting with an important peripheral vascular disease (aortoiliac aneurysms, severe peripheral aortoiliac occlusive disease, or
arteriosclerosis of the femoral vessels), patients with limb complications related to
femoral artery cannulation, and patients under peripheral ECMO support with inadequate upper body oxygenation and perfusion [24]. Such an approach is contraindicated if there is an extension of an aortic disease into the artery or a known axillary/
subclavian stenosis or atheroma. Obesity and wall chest edema is a relative contraindication as the exposure of the artery could be difficult. The major disadvantage
of the axillary cannulation is that it is not available for emergency and it cannot be
percutaneously performed.
[email protected]
60
F. Formica et al.
5.3.3.2 Axillary Artery: Surgical Technique

There are quite a few incision techniques in order to expose the right axillary
artery. Several authors [26] prefer an 810-cm horizontal incision below the
medial third of the clavicle, because of a lower risk of brachial plexus injury.
Some others [24] prefer the deltoideo-pectoral approach with a 610-cm incision
below and parallel to the lateral two thirds of the clavicle. The pectoralis muscle
is then divided parallel to its fibers; the clavipectoralis fascia is incised, exposing
the pectoralis minor, which is divided or retracted laterally. The axillary artery is
identified superior to the vein and by palpation. The brachial plexus is carefully
dissected from the artery. Proximal and distal control of the artery is obtained
through loops. If the artery is good sized and with a good exposure, direct cannulation could be performed. A single purse-string suture is placed in the anterior
wall of the axillary artery, and, after heparinization, the artery is clamped and the
cannula is inserted. In patients with a small body surface area and small vessels,
an 8-mm Dacron side graft is best (Fig. 5.5a). The chimney graft can be extended
over the skin (Fig. 5.5 B1) or tunneled under it and exteriorized by a small second
a
Vascular
graft
AA
AV
Fig. 5.5 Axillary vessel

surgical cannulation. (a)
Side-graft cannulation
technique: an 8-mm Dacron
side graft is sewn in an
end-to-side way on the
axillary artery. The arterial
cannula is then inserted into
the vascular graft. (b) The
chimney graft can be
tunneled under the skin (B1)
or extended over it (B2). AA
axillary artery, AV axillary
vein
b
Vascular graft
Cannula
b1
[email protected]
Cannula
b2
61
incision (Fig. 5.5 B2). The cannula is then connected to the distal graft. After
accurate hemostasis, the wound is closed. After weaning from ECMO, the surgical field is revised and the axillary artery is repaired. In case of direct cannulation,
the axillary artery is directly repaired. When the primary closure narrows the
artery, a patch is recommended [27]. If a side graft is used, it could be closed as
described above.
5.3.3.3 Axillary Vein: Surgical Technique

Axillary vein cannulation is rare in ECMO patients. Axillary vein lies superficial to
the artery, and it is easily accessible when the arterial cannula is already placed. A
purse-string suture is made at the cephalic veinaxillary vein junction. The mobilized segment of the vein is clamped distally, and a longitudinal venotomy is performed within the purse string. The cannula is later introduced.
5.3.3.4 Complications
Even if limb ischemia and compartment syndrome could be avoided with axillary cannulation, other complications could occur with such a technique. Hyperperfusion syndrome occurs in nearly 20 % of the patients, and it is more frequent with the use of a
side graft [28]. This syndrome could develop because of an arterial outflow obstruction. That occurs in case of technical problems in the construction of the side graft that
could narrow the artery and cause a preferential flow down the arm. A similar pattern
could be expected with atherosclerotic aortic arch disease and/or acute type A dissections. Venous obstructive causes could include bleeding, a compressive hematoma,
and the presence of a venous cannula or deep vein thrombus. The management of
hyperperfusion syndrome is directed to solve its etiology [28]. Other further complications are stump graft and wound infection, stroke, brachial plexus injury, and axillary artery injury or dissection due to its fragility whenever directly cannulated.
5.3.4
Cervical Vessels
5.3.4.1 Indications and Complications

Cervical cannulation is widely used for ECMO in neonates and infants, and it is
rarely used in adults. The only exception occurs as far as the right internal jugular
vein that is often cannulated in veno-venous ECMO and to increase venous drainage in venoarterial ECMO. Percutaneous cannulation of the jugular vein is the most
used cannulation technique even if the semi-Seldinger technique and the open technique are possible alternative solutions when the percutaneous approach is not feasible. Carotid artery cannulation is frequently related to cerebrovascular accidents,
and that is why it is usually avoided in adult ECMO cannulation. Right internal
jugular vein could be related to neurologic complications that may result from intraventricular hemorrhage, cerebral edema, or hypoxia. Local complications are vagus
nerve injuries, arterial dissection, arterial and venous laceration or perforation,
thrombosis, infection, bleeding, and hematoma.
[email protected]
62
F. Formica et al.
5.3.4.2 Surgical Technique

The surgical technique that is often used to cannulate the right internal jugular vein
is the semi-Seldinger. The patient is placed supine while the head is turned to the left.
The right internal jugular vein is exposed through a transverse incision over the lower
third of the sternocleidomastoid muscle. The vein is dissected and proximal and distal control is obtained. A small stab incision is made cephalad to the main incision,
and, after heparinization, the introducer needle is passed through the stab into the
vein. Exposure of the vein allows to size the cannula and to monitor any injury during
cannulation. The guidewire is passed through the needle, and the following steps are
the same as described for Seldinger technique. The cannula tip is placed into the right
atrium. This approach does not require any ligation of the vein, and it facilitates the
process of decannulation. The absence of ligation of the right internal jugular vein
allows drainage of deoxygenated blood down the ipsilateral vessel and into the cannula. Such a practice reduces recirculation and could also have an impact on the
intracranial venous pressure and the incidence of intracranial hemorrhage [10].
The open approach is recommended when the cannulation of the carotid artery is
required. The vessels are exposed as described previously. The internal jugular and
carotid artery are isolated and controlled. A careful attention must be paid in order
to not injure the vagus nerve, which lies behind the neck vessels within the carotid
sheath. After the heparinization and the arteriotomy, the cannula is inserted and
fixed with the tip of the cannula placed at the orifice of the innominate artery. The
venous cannula is placed in the same way [10]. In neonates, the most widely used
cannulation technique employs the permanent ligation of the jugular vein and the
carotid artery cephalically to the cannulation sites.
References
1. ELSO Guidelines for Cardiopulmonary Extracorporeal Life Support. Extracorporeal Life
Support Organization, Version 1.3 November 2013, Ann Arbor, MI, USA. HYPERLINK
http://www.elsonet.org. Accessed on 27 jan 2014
2. Marasco SF, Lukas G, McDonald M, McMillan J, Ihle B (2008) Review of ECMO (Extra
Corporeal Membrane Oxygenation) support in critically ill adult patients. Heart Lung Circ
17S:S41S47
3. Kanji HD, Schulze CJ, Oreopoulos A, Lehr EJ, Wang W, MacArthur RM (2010) Peripheral
versus central cannulation for extracorporeal membrane oxygenation: a comparison of limb
ischemia and transfusion requirements. Thorac Cardiovasc Surg 58(8):459462
4. Vander Salm TJ (1997) Prevention of lower extremity ischemia during cardiopulmonary
bypass via femoral cannulation. Ann Thorac Surg 63:251252
5. Brkle MA, Sodian R, Kaczmarek I, Weig T, Frey L, Irlbeck M, Dolch ME (2012) Arterial
chimney graft cannulation for interventional lung assist. Ann Thorac Surg 94(4):
13351337
6. Jackson KW, Timpa J, McIlwain RB, OMeara C, Kirklin JK, Borasino S, Alten JA (2012)
Side-arm grafts for femoral extracorporeal membrane oxygenation cannulation. Ann Thorac
Surg 94(5):e111e112
7. Demertzis S, Carrel T (2011) Rapid peripheral arterial cannulation for extracorporeal life support with unimpaired distal perfusion. J Thorac Cardiovasc Surg 141:10801081
8. Lamb KM, Hirose H, Cavarocchi NC (2013) Preparation and technical considerations for percutaneous cannulation for veno-arterial extracorporeal membrane oxygenation. J Card Surg
28(2):190192
[email protected]
63
9. Stulak JM, Dearani JA, Burkhart HM, Barnes RD, Scott PD, Schears GJ (2009) ECMO cannulation controversies and complications. Semin Cardiothorac Vasc Anesth 13(3):176182
10. Field ML, Al-Alao B, Mediratta N, Sosnowski A (2006) Open and closed chest extrathoracic
cannulation for cardiopulmonary bypass and extracorporeal life support: methods, indications,
and outcomes. Postgrad Med J 82:323331
11. Bisdas T, Beutel G, Warnecke G, Hoeper MM, Kuehn C, Haverich A, Teebken OE (2011)
Vascular complications in patients undergoing femoral cannulation for extracorporeal membrane oxygenation support. Ann Thorac Surg 92(2):626631
12. Huang SC, Yu HY, Ko WJ, Chen YS (2004) Pressure criterion for placement of distal perfusion catheter to prevent limb ischemia during adult extracorporeal life support. J Thorac
Cardiovasc Surg 128:776777
13. Gander JW, Fisher JC, Reichstein AR, Gross ER, Aspelund G, Middlesworth W, Stolar CJ
(2010) Limb ischemia after common femoral artery cannulation for venoarterial extracorporeal membrane oxygenation: an unresolved problem. J Pediatr Surg 45(11):21362140
14. Kasirajan V, Simmons I, King J, Shumaker MD, DeAnda A, Higgins RS (2002) Technique to
prevent limb ischemia during peripheral cannulation for extracorporeal membrane oxygenation. Perfusion 17(6):427428
15. Russo CF, Cannata A, Vitali E, Lanfranconi M (2009) Prevention of limb ischemia and edema
during peripheral venoarterial extracorporeal membrane oxygenation in adults. J Card Surg
24:185187
16. Schachner T, Bonaros N, Bonatti J, Kolbitsch C (2008) Near infrared spectroscopy for controlling the quality of distal leg perfusion in remote access cardiopulmonary bypass. Eur J
Cardiothorac Surg 34(6):12531254
17. Spurlock DJ, Toomasian JM, Romano MA, Cooley E, Bartlett RH, Haft JW (2012) A simple
technique to prevent limb ischemia during veno-arterial ECMO using the femoral artery: the
posterior tibial approach. Perfusion 27(2):141145
18. Kimura N, Kawahito K, Ito S, Murata S, Yamaguchi A, Adachi H, Ino T (2005) Perfusion
through the dorsalis pedis artery for acute limb ischemia secondary to an occlusive arterial
cannula during percutaneous cardiopulmonary support. J Artif Organs 8(3):206209
19. Le Guyader A, Lacroix P, Ferrat P, Laskar M (2006) Venous leg congestion treated with distal
venous drainage during peripheral extracorporeal membrane oxygenation. Artif Organs
30(8):633635
20. Merin O, Silberman S, Brauner R, Munk Y, Shapira N, Falkowski G, Dzigivker I, Bitran D
(1998) Femoro-femoral bypass for repeat open-heart surgery. Perfusion 13:455459
21. Greason KL, Hemp JR, Maxwell JM, Fetter JE, Moreno-Cabral RJ (1995) Prevention of distal
limb ischemia during cardiopulmonary support via femoral cannulation. Ann Thorac Surg
60(1):209210
22. Bichell DP, Balaguer JM, Aranki SF, Couper GS, Adams DH, Rizzo RJ, Collins JJ Jr, Cohn
LH (1997) Axilloaxillary cardiopulmonary bypass: a practical alternative to femorofemoral
bypass. Ann Thorac Surg 64(3):702705
23. Schmidt M, Brchot N, Hariri S, Guiguet M, Luyt CE, Makri R, Leprince P, Trouillet JL, Pavie
A, Chastre J, Combes A (2012) Nosocomial infections in adult cardiogenic shock patients
supported by venoarterial extracorporeal membrane oxygenation. Clin Infect Dis
55(12):16331641
24. Navia JL, Atik FA, Beyer EA, Ruda Vega P (2005) Extracorporeal membrane oxygenation
with right axillary artery perfusion. Ann Thorac Surg 79(6):21632165
25. Gates JD, Bichell DP, Rizzo RJ, Couper GS, Donaldson MC (1996) Thigh ischemia complicating femoral vessel cannulation for cardiopulmonary bypass. Ann Thorac Surg 61:730733
26. Baribeau YR, Westbrook BM, Charlesworth DC (1999) Axillary cannulation: first choice for
extra-aortic cannulation and brain protection. J Thorac Cardiovasc Surg 118:11531154
27. Sabik JF, Nemeh H, Lytle BW, Blackstone EH, Gillinov AM, Rajeswaran J, Cosgrove DM
(2004) Cannulation of the axillary artery with a side graft reduces morbidity. Ann Thorac Surg
77(4):13151320
28. Chamogeorgakis T, Lima B, Shafii AE, Nagpal D, Pokersnik JA, Navia JL, Mason D,
Gonzalez-Stawinski GV (2013) Outcomes of axillary artery side graft cannulation for extracorporeal membrane oxygenation. J Thorac Cardiovasc Surg 145(4):10881092
[email protected]
Materials: Cannulas, Pumps,

Oxygenators
Umberto Borrelli and Cristina Costa
6.1
Introduction
Over the past decade, several improvements have been made to the extracorporeal
circulation (ECC) used in cardiac surgery (cardiopulmonary bypass). These include
a reduction in the surface contact between blood and air, improved biocompatibility
between blood and materials, and the use of compact circuits and the assisted
venous drainage. These improvements have made possible to optimize other ECCs,
specifically extracorporeal membrane oxygenation (ECMO) and extracorporeal life
support (ECLS) [111], and, as such, led to the introduction to the market of specific circuits validated for several weeks of continuous use (1430 days).
ECMO is a closed ECC circuit, without air/blood interfaces. Its internal surface
is completely treated by glycoproteins or by pre-heparinization with covalent and/
or ionic bondings (Fig. 6.1). These treatments lead to an improved blood compatibility, the reduction of the inflammatory response, and better patient anticoagulation
[1012].
ECMO is classified according to the connection with the blood circulation of the
patient.
6.1.1
Venovenous ECMO (VV)
Used most frequently in the treatment of severe respiratory compromise, venovenous ECMO has both the inflow and outflow cannula placed in the patients venous
circulation.
U. Borrelli, BCCP, ECCP (*)

Department of Cardiovascular Surgery, Grand Hpital de Charleroi, Gilly, Belgium
C. Costa, CCP, ECCP
Department of Perfusion, Hospital San Geraldo, Monza, Italy
[email protected]
65
U. Borrelli and C. Costa
66
a
Na
Na+ +
CI
Forming ionic
bond
Mg
bH
Sodium contributes
electron, leaving it
with a closed shell
Mg
CI
CI
Chlorine gains
electron, leaving it
with a closed shell
CI
+ CI
or H - H
CI
CI
CI
or
CI - CI
or
H - CI
Constituent atoms
share a pair of electrons,
closing the shell for each
2-
2+
Forming covalent
bond
Forming ionic
bond
Ca +
H
2+
Ca +
CI
+ CI
CI
H
Bonding
pair
CI
Lone
pair
Fig. 6.1 The techniques used to coat the circuits are classified into two categories: (a) Ionic bond:
bond in which one or more electrons from one atom are removed and attached to another atom,
resulting in positive and negative ions which attract each other. (b) Covalent bond: bond in which
one or more pairs of electrons are shared by two atoms
6.1.2 Venoarterial ECMO (VA)

Able to totally or partially bypass the lungs or heart, venoarterial ECMO allows for
full cardiac or pulmonary support. It can be used to provide support of the vital
organs, temporary circulatory support, and/or relief to the heart during myocardial
recovery. Where appropriate, it can also be a bridging therapy to transplantation or
establishment of a mechanical heart. During VA ECMO, the inflow cannula is
placed in the patients venous circulation and the outflow cannula in the arterial
circulation.
6.2
Biocompatible Components
Currently, there are several approaches to improve the biocompatibility of cardiopulmonary bypass components, based on the use of antithrombotic biomolecules
such as heparin, polymeric molecules, and new glycoprotein molecules. Heparin is
a negatively charged, hydrophilic, complex polysaccharide acid.
6.3
Example of Biocoating That Is Used During ECMO
The Carmeda coating allows heparin molecules to be attached to the biomaterial

surface by a covalent bonding.
The Duraflo II heparin coating ionically attaches heparin to a quaternary
ammonium carrier, which binds to the biomaterial surfaces.
The Rheoparin coating fixes the heparin molecules to the biomaterial surface,
due to ionic forces.
[email protected]
Materials: Cannulas, Pumps, Oxygenators
67
The Physio coating is composed of a phosphorylcholine polymer. The heparin is

not present in this type of coating and is based on a phosphorylcholine molecule.
The Bioline coating combines polypeptides and heparin. Polypeptides are
adsorbed onto the components of the CPB surface forming a steric hindering.
The heparin molecules are attached to the polypeptides via triple covalent bonds
and ionic interaction.
6.3.1 ECMO Circuit for Adult Patients

The ECMO circuit for an adult patient usually consists of an inflow cannula (venous
cannula), a tubing made of polyvinyl chloride (PVC) with 3/8 of diameter, a
centrifugal pump, a heat exchanger embedded in a membrane oxygenator made of
polymethylpentene (PMP) membrane, and an outflow cannula that transports
arterialized blood (arterial cannula).
The circuits are usually very compact, with some flexibility for transportation, mobilization, and general care of patients in the ICU. ECMO circuits have several sites for
monitoring including pressure measurement, blood sampling, and the continuous analysis of blood parameters in the inflow (venous blood) or outflow (arterial blood) lines. It
is also possible to connect a hemofiltration system or hemodialysis machine on the
ECMO system, according to the operating characteristics of these devices. Some ECMO
consoles may be equipped with a servo controller that allows control of the centrifugal
pump rotations, in relation to the negative pressure (P1) in the inflow line (Fig. 6.2).
The choice of the different components being a part of ECMO circuit is very
important because it reduces the physiological impact of the assist system on the
patient. Depending on the type of assistance to be performed, the selection of the
oxygenator is a priority, because it represents the largest contact area between blood
and materials in the circuit.
The geometry and the internal resistance of each component will induce a pressure drop that will influence the hemodynamics in the system, the management of
the system and the evolution of the patient. The optimization of the position of the
cannulas, the appropriate selection of circuit components and the use of the surface
treatment (tip to tip) can reduce the impact of ECMO on the patient. This is comparable to an artificial reduction of the assistance time [3, 6, 818].
6.3.2
Centrifugal Blood Pumps
Centrifugal pumps are nonocclusive, and as such, there is a risk of inducing a backflow of blood against the current through the pump. This is the reason that some
consoles are equipped with anti-reflux system. ECMO dedicated centrifugal pumps
are driven by electromagnetic induction motors and uses the principles of centrifugal force to generate a flow (described in terms of liters/minute), which is created by
the rotation of the cones, fins, or vanes and rotors (Fig. 6.3).
[email protected]
68
Arterial blood sample
Arterial blood gas sensor
Venous blood sample
Venous blood gas sensor
oxygenator
and
heat exchange
Preload
Afterload
P 3
P 1
P 2
Line inflow 3/8

Line outflow 3/8
Flow probe
Centrifugal pump
Venous drainage
Outflow cannula
Console
Flow
Inflow cannula
Servo regulator
negative pressure
U. Borrelli
Arterial and venous

blood gas monitor
Fig. 6.2 Characteristics of the ECMO circuit for adult patients. P1 Negative pressure suction from
the inflow line, between the inflow cannula and the entrance of the centrifugal pump. P2 Positive
ejection pressure between the outlet of centrifugal pump and the inlet of the oxygenator. P3 Positive
ejection pressure of the outflow line, between the outlet of the oxygenator and the outflow cannula.
P2 P3 Corresponding to the gradient of pre- and post-oxygenator pressure, it indicates the
pressure drop of the oxygenator
The flow is controlled by either electromagnetic or ultrasonic flow meter. Flow

depends on the speed of rotation per minute (RPM) of the centrifugal pump, the
hemodynamic conditions including the preload or afterload, and the characteristics
of the inflow or outflow cannulas that are used including their positioning. To ensure
adequate preload, the centrifugal pump and the oxygenator should preferably be
below the level of the right atrium (RA) of the patient. An increase in the negative
pressure suction upstream of the centrifugal pump (P1) will be the result of a
decrease in preload. An increase in the positive pressure ejection downstream of the
oxygenator (P3) is generated by an increase of the afterload. Both of these phenomena decrease flow despite a constant pump speed (RPM). It is important to check the
parameters of the patient, the position of the cannulas, and the absence of kinking
or clamping on the lines and of thrombi in the circuit.
A decrease in preload causing a significant increase in P1 can induce the
phenomena of cavitation (chattering) in the inflow line. This can cause major trauma
to the blood, resulting in hemolysis, gaseous microemboli, and other adverse events.
It is for this reason that some ECMO circuits use a servo controller to maintain a
safe relationship between the pump speed (RPM) and P1. This ensures that there is
time available when preload decreases to treat the cause before cavitation and its
[email protected]
69
e
f
Fig. 6.3 (a) Centrifugal pump console (SCPC) Revolution of Sorin Group. (b) SCPC Affinity CP
of Medtronic. (c) SCPC CentriMag of Levitronix. (d) SCPC Cardiohelp System of Maquet. (e)
SCPC Rotaflow of Maquet. (f) SCPC DeltaStream of Medos
consequences occur, thereby reducing the probability of trauma to the blood and the
appearance of gaseous microemboli [9, 10, 13, 14].
Prior to stopping a centrifugal pump, it first is necessary to clamp the outflow
line. Once the pump has been deactivated, the inflow line should be clamped.
6.3.3 Membrane Oxygenator

The membrane oxygenators are placed distal to the centrifugal pump. They are
generally equipped with a heat exchanger made of polyurethane, polyester, or stainless steel. Depending on the model, the surface of the heat exchanger varies from
0.14 to 0.6 m2 (Table 6.1).
A temperature gradient between the water and the blood enables temperature
adjustment (Fig. 6.4). This area forms a thermal bridge and enables the patient to be
either warmed or cooled. The oxygenator is the largest area of contact between the
blood and the ECMO circuit, and its choice will depend on both patient factors and
the underlying pathology being treated. Oxygenators used in cardiac surgery have a
microporous membrane of hollow polypropylene (PP) fibers. They intent to replace
the pulmonary alveolar function of the patient and ensure the delivery of O2 and the
removal of CO2. The use of these oxygenators is generally limited to around 8 h.
[email protected]
70
Table 6.1 Characteristics of the oxygenators
Oxygenators
Specifications
Blood flow
rate
Membrane
surface area
Membrane
type
Biomedical
coating
Static priming
volume
Heat
exchanger
surface area
Heat
exchanger
material
Size of blood
inlet and outlet
connectors
Size of water
connectors
Sorin
EOS ECMO
0.55 l/min
Maquet
Quadrox PLS
0.57 l/min
Maquet
HLS module
Integrated
Medos
centrifugal pump Hilite 7000LT
0.57 l/min
07 l/min
1.2 m2
1.8 m2
1.8 m2
1.9 m2
PMP fiber
PMP fiber
PMP fiber
PMP fiber
Phosphorylcholine
Bioline
Bioline
Rheoparin
150 ml
250 ml
273 ml
270 ml
0.14 m2
0.6 m2
0.4 m2
0.45 m2
Stainless steel
Polyurethane
Polyurethane
Polyester
3/8
3/8
3/8
3/8
1/2 Hansen coupling
1/2 Hansen
coupling
1/2 Hansen
coupling
1/2 Hansen
coupling
This usage limit is due to the microporous structure of the fiber, which may alterate
over time, causing plasma leakage of the membrane [14]. The result is a decrease in
performance of the gas fibers and/or plasma leakage situated at gas outlet of the
oxygenator (gas out) (Fig. 6.5).
Different kind of membranes are used for ECMO oxygenators. Following membrane characteristics can be observed: silicone caoutchouc, selective permeability
or polymethylpentene (PMP), depending on the models; the surface of the membrane varies from 1.2 to 1.9 m2. The PMP membrane is made of hollow fibers. The
fiber structure itself is covered with a dense but tiny outer skin, characterizing it as
a diffusion membrane, avoiding plasma leakage. The gas permeability for oxygen
and carbon dioxide is excellent and the gas exchange capability remains equivalent
to the micro porous membranes. This is one of the reasons why the membrane
remains functional during several weeks without oxygenator exchange. The gas
exchanges diffuses through the membrane due to the partial pressure gradient on
both sides of the membrane. There is consequently no direct gas/blood interface
(Fig. 6.5). Their physical characteristics are tailored to meet the demand for gaseous and thermic exchange (Table 6.1). A gas blender mixing air and O2 is connected to oxygenator permetting the adjustment of exchange O2 and CO2 selectively.
Especially, the flow of the gas mixture (the sweep) acts to determine the extraction of CO2 and the FiO2 from 21 to 100 % (the oxygen concentration of the mixture) acts to govern the transfer of oxygen into the blood.
[email protected]
71
Pore size max, < 0.2 m
Without pores
Water flow
P P fiber
Heat exchanger
P M P fiber
Polymethylpentene membrane (PMP)
Heat exchanger
Polypropylene microporous membane (P P)
7,000 nm
0,12 nm
Blood flow
Gas flow
Water flow
7,000 nm
0,12 nm
Blood flow
Gas flow
Fig. 6.4 Different parts of the oxygenator
Gas in
Water in
Water out
Arterial blood out
Fig. 6.5 Overview of the

oxygenator
Venous blood in
Venous blood sample
or
pre oxygenator pressure (P2)
Arterial blood sample

or
post oxygenator pressure (P 3)
Gas out
[email protected]
72
The pressure drop generated in the oxygenator depends on its physical characteristics or by variation of its internal resistance during ECMO use (including blood
temperature, viscosity, and thrombus formation on the membrane). It is measured
by the pre- and post-oxygenator pressure gradient (P2 P3). An increase in the pressure drop of the oxygenator may indicate deterioration of the hemodynamics of the
membrane, which in turn may impair gas transfers and should prompt consideration
regarding changing of the oxygenator [11, 16].
6.3.4
Cannulas for Adult Patients
The blood flow in the ECMO circuit is determined by the size of the cannula (internal diameter and length), the design, the pressure drop, and the positioning. The
choice of the cannulas is made according to the mechanical and fluid characteristics
(pressure, flow, etc.) (Table 6.2). It takes into account surgical needs, cannula placement (central or peripheral), the quality of the patients blood vessels, and the type
of ECMO to be used (e.g., VA or VV). Certain ECMO configurations are especially
designed to suit the type of cannulation, for example, to suit the dual-lumen single
cannulation of the jugular vein or the double femoro-jugular cannulation (Fig. 6.6).
The cannulas designed for circulatory assistance are designed for percutaneous
access into the arteries or veins of the patient. They usually have a wire enforced
body on the longest part of the cannula in order to prevent changes in their hemodynamic characteristics during the mobilization of the patient (Fig. 6.7).
Additionally, the temperature of the circulating blood in the ECMO circuit can
change the resistance of the cannula. The hemodynamics of ECMO can also be
affected by compression of one or more of the cannulas by intrathoracic and/or intraabdominal pressure (cough, pleural effusion, and/or intra-abdominal hypertension).
Table 6.2 Example of pressure drop for arterial and venous cannulas
Venous
Arterial
Pressure drop vs. flow
200
180
180
160
160
60
40
fr.
22 .
r
27 f
r.
29 f
20
100
fr.
fr.
120
19
23
80
fr.
17
fr.
100
140
15
fr.
Pressure drop (mmHg)
120
21
fr.
19
17
f
140
r.
200
15 fr.
Pressure drop (mmHg)
Pressure drop vs. flow
80
21
60
fr.
23
40
fr.
20
0
Flow rate (l/min. of water) at room temperature
Flow rate (l/min. of water) at room temperature

Outflow cannula
Inflow cannula
[email protected]
73
Fig. 6.6 (a) VV ECMO with a double-lumen cannula. (b) VV ECMO with two cannulas placed
in the jugular and femoral veins
Fig. 6.7 Example of percutaneous cannulas
[email protected]
74
Logically, the use of inflow and outflow cannulas with maximal diameter would
reduce the pressure drop in the circuit. However, like often in ECMO, a balance has
to be found between treating the pathology and minimizing trauma and discomfort
to the patient caused by the presence of the cannulas in the vascular system.
The correct selection of in- and outflow cannulas, as well as their positioning, is
very important. They help to optimize the hemodynamics of ECMO in relation to
the patient and ensure an adequate flow able to satisfy the physiological demands to
be met. This results in a reduction in the trauma to the blood cells, thereby reducing
hemolysis and gaseous microemboli [13, 14, 17], by preventing an excessive
increase in the P1 or P3 (Fig. 6.2.).
6.3.4.1 Venous Cannulas

Venous cannulas (the inflow cannula) are longer than the outflow cannulas (55 cm)
as they need to reach from their insertion point in the common femoral vein to adjacent
to the right atrium (RA). They have a larger diameter than outflow cannulas (1529 Fr)
and a larger proportion of their length is multiperforated. These features reduce the
pressure drop and limit the phenomena of chattering avoiding a significant collapse of
the wall of the right atrium (RA) or the inferior vena cava (IVC). The result of these
characteristics will allow the observation of a decreased pressure drop at in the pressure drop at P1 (Fig. 6.2.) by increasing the drainage flow while maintaining a constant
preload and the RPM of the centrifugal pump. In certain circumstances, the cannula
design allows the use of smaller inflow cannulas without causing deleterious hemodynamic consequences while maintaining the same flow (Table 6.2). During pulmonary
assistance (VV) using a femoro-jugular cannulation, the inflow cannula is placed in
the inferior vena cava with its distal tip at the level of the subhepatic veins, while the
outflow cannula is placed via the jugular vein into the superior vena cava (SVC) with
its distal tip at the level of the right atrium.
To eliminate the discomfort that is caused by the presence of the cannula in the
jugular vein, or for certain therapeutic reasons, some centers cannulate the right and
left femoral veins. This requires the use of two different cannulas with different design
and diameters. The distal tip of the inflow cannula is positioned through the left femoral vein at the level of the subhepatic veins, while the distal tip of the outflow cannula
is positioned through the right femoral vein at the level of the right atrium. This combination can reduce the effect of shunt (recirculation) between the two cannulas.
6.3.4.2 Arterial Cannulas

Arterial cannulas (outflow cannula) are smaller than venous cannulas both in terms of
diameter (1523 Fr) and overall length. They have some perforations at their distal tip,
though these are not as extensive as in the venous cannulas. According to their characteristics, a pressure drop at the outflow cannula can be observed when P3 increases
(Fig. 6.2) causing a decrease in flow despite a constant afterload and RPM of the centrifugal pump. During femoro-femoral VA ECMO, the outflow cannula is placed into
the common femoral artery. The particular position of the cannula can cause ischemia
of the ipsilateral lower limb. This can be prevented by the introduction of a 6 fr catheter
for a distal perfusion [18] connected to the Luer connector of the outflow cannula and
introduced few centimeters downstream the superficial femoral artery.
[email protected]
75
The placement of the cannulas is guided by echocardiography (TEE) and/or

fluoroscopy.
In general, the cannulas have a radiopaque marker, though the distal few centimeters are often radiotransparent.
6.4
Conclusions
The optimization of the various components of the ECMO and its management in
relation to the pathology have a significant and direct impact on the prognosis and
disease course of the patient requiring circulatory assistance.
References
1. Borrelli U, Detroux M, Nackers P et al (2003) Impact on the inflammatory reaction by optimization of the extracorporeal circulation in cardiac surgery. Biomed J 24(Suppl 1):80s.
Innovation and Technology in Biology and Medicine; Editions Scientific and Medical
Elsevier
2. Borrelli U, Detroux M, Nackers P et al (2003) Comparison of the troponin I levels during coronary artery bypass graft in cardiac surgery procedures, realised with and without extracorporeal circulation. Biomed J 24(Suppl):79s. Innovation and Technology in Biology and Medicine;
Editions Scientific and Medical Elsevier
3. Borrelli U, Al-Attar N, Detroux M, Nottin R et al (2007) Compact extracorporeal circulation: reducing of cardiopulmonary bypass to improve outcomes. Surg Technol Int
16:159166
4. Society for Advancement of blood Management (2007) 5(4)
5. Borrelli U, Al-Attar N, Detroux M, Nottin R et al (2008) La rduction de la surface de la circulation extracorporelle amliore les rsultats. Journal de la Socit Franaise de Chirurgie
Thoracique et Cardio-Vasculaire 12:4653
6. Yavari N, Becker RC (2009) Coagulation and fibrinolytic protein in cardiopulmonary bypass.
J Thromb Thrombolysis 27:95104
7. Karkouti K, Djaiani G, Borger MA et al (2005) Low hematocrit during cardiopulmomary
bypass is associated with increased risk of perioperative stroke in cardiac surgery. Ann Thorac
Surg 80:13811387
8. Society of Thoracic Surgeons Blood Conservation Guideline Task Force et al (2011) 2011
update to the Society of Surgeons and the Society of Cardiovascular Anesthesiologists blood
conservation clinical practice guidelines. Ann Thorac Surg 91:944982
9. Chalegre ST et al (2011) Vacuum-assisted venous drainage in cardiopulmonary bypass
and need of blood transfusion: experience of a service. Rev Bras Cir Cardiovasc 26(1):
122127
10. Goksedef D, Omeroglu SN, Balkanay OO, Denli Yalvac ES, Talas Z, Albayrak A, Ipek G
(2012) Hemolysis at different vacuum levels during vacuum-assisted venous drainage: a prospective randomized clinical trial. Thorac Cardiovasc Surg 60(4):262268.
doi:10.1055/s-0031-1280019. Epub 2011 Jul 25
11. Zimmermann AK, Weber N, Aebert H, Ziemer G, Wendel HP (2007) Effect of biopassive and
bioactive surface-coatings on the hemocompatibility of membrane oxygenators. J Biomed
Mater Res B Appl Biomater 80(2):433439
12. Ranucci M, Isgr G, Soro G et al (2004) Reduced systemic heparin dose with phosphorylcholine coated closed circuit in coronary operations. Int J Artif Organs 27(4):311319
13. Toomasian JM, Bartlett RH (2011) Hemolysis and ECMO pumps in the 21st century. Perfusion
26(1):56
[email protected]
76
14. Pedersen TH, Videm V, Svennevig JL, Karlsen H, Ostbakk RW, Jensen O, Mollnes TE (1997)
Extracorporeal membrane oxygenation using a centrifugal pump and a servo regulator to prevent negative inlet pressure. Ann Thorac Surg 63(5):13331339
15. Meyns B, Vercaemst L, Vandezande E, Bollen H, Vlasselaers D (2005) Plasma leakage of
oxygenators in ECMO depends on the type of oxygenator and on patient variables. Int J Artif
Organs 28(1):3034
16. Khoshbin E, Roberts N et al (2005) Polymethylpentene oxygenators have improved gas
exchange capability and reduced transfusion requirements in adult extracorporeal membrane
oxygenation. ASAIO J 51(3):281287
17. Simons AP, Ganushchak Y, Wortel P, van der Nagel T, van der Veen FH, de Jong DS, Maessen
JG (2008) Laboratory performance testing of venous cannulae during inlet obstruction. Artif
Organs 32(7):566571
18. Madershahian N, Nagib R et al (2006) A Simple technique of distal limb perfusion during
prolonged femoro-femoral cannulation prevention of lower extremity ischemia during cardiopulmonary bypass via femoral cannulation. J Card Surg 21:168169
[email protected]
Coagulation, Anticoagulation,
and Inflammatory Response
Marco Ranucci
7.1
Introduction
Extracorporeal membrane oxygenation (ECMO) represents a modified system of

cardiopulmonary bypass (CPB). Depending on the type of cannulation, it may assist
or replace both the cardiac and the pulmonary function (arteriovenous ECMO) or
the pulmonary function alone (venovenous ECMO). Regardless of this difference,
the ECMO circuit always includes a centrifugal pump, tubings, and an oxygenator.
To this respect, the pathophysiological reactions of blood coming in contact with
foreign surfaces are similar to those observed during CPB. However, major differences exist between CPB and ECMO, with different degrees of hemostasis and
coagulation and inflammatory reactions.
As a consequence, the management of anticoagulation during ECMO strongly
differs from the conventional anticoagulation regimen required by CPB.
This chapter provides an overview of the hemostatic system activation,
inflammatory reaction, and anticoagulation protocols for adult patients on
ECMO.
7.2
Hemostatic System Activation During ECMO
The main scientific information on hemostatic system activation derives from studies in the field of dialysis and, most importantly, CPB during cardiac operations.
Conventionally, the hemostatic system activation may be triggered by materialdependent and material-independent mechanisms.
M. Ranucci, MD, FESC

Department of Cardiothoracic-Vascular Anesthesia and ICU,
IRCCS Policlinico San Donato, San Donato Milanese, MI, Italy

[email protected]
77
78
7.2.1
M. Ranucci
Contact Phase: The Material-Dependent Blood Activation

and the Role of Fibrinogen and Platelets
The extensive contact between blood and foreign surfaces triggers a materialdependent blood activation through the contact phase. Given the large contact surface area represented by the hollow fiber of membrane oxygenator, ECMO is a
classical model of contact-phase activation.
The initial reaction of blood coming in contact with foreign surfaces, and basically with plasticizers, is based on the interaction between the surface and the
plasma proteins. This produces a layer of proteins on the surface of the circuit and
oxygenator, which is mainly represented by fibrinogen, albumin, and -globulins.
The initial layer is mainly represented by fibrinogen, through the interaction
between the hydrophobic foreign surface and hydrophilic sites of fibrinogen. Once
fibrinogen is bound to the surface, this will in turn trigger platelet adhesion to the
fibrinogen receptors. Other blood cells like fibroblast, leukocytes, and even red cells
may participate in this layer. Fibrinogen is the main determinant of the initial protein layer on the foreign surface.
Simultaneously, the coagulation system becomes activated through the so-called
intrinsic pathway. Factor XII (Hageman factor) is activated to factor XIIa, leading
to the subsequent activation of pre-kallikrein, high-molecular weight kininogen,
and factor XI. With the subsequent contribution of factors IX and X, prothrombin
(factor II) is activated to thrombin (factor IIa) [1].
The material-dependent blood activation is of course dependent on the nature of
the materials. Presently, all the systems available for ECMO are equipped with biocompatible surfaces, which limit the severity of this reaction.
The most commonly used biocompatible treatments are based on surface bonding of heparin molecules. The natural endothelium actually contains heparin-like
molecules, called glycosaminoglycans (GAGs), which contribute to the anticoagulant properties of the endothelium.
Heparin-bonded CPB circuits are associated with a lower activation of the hemostatic system and prevention of platelet adhesion and activation as well as a preservation of platelet count [26].
7.2.2
Tissue Factor and Thrombin Generation
Thrombin generation through the material-dependent blood activation alone is by

far less pronounced than thrombin generation during CPB [79]. Actually, the most
powerful trigger for thrombin generation is tissue factor (TF). Soluble TF is released
by the damaged endothelial surface, and this release is of course greatly enhanced
during surgery. Additionally, cell-bound TF is released by the epicardium, myocardium, adventitia, and bone. In cardiac surgery with CPB, shed blood from the mediastinum is rich of TF, and readmission of this blood into the circulation is the major
determinant of thrombin generation [1012]. In the most recent dynamic interpretation of the hemostatic system activation, small amounts of thrombin are formed
[email protected]
Coagulation, Anticoagulation, and Inflammatory Response
79
following TF release (initiation). Subsequently, thrombin activates platelets through

the PAR receptors (amplification), and large amounts of thrombin are formed on the
platelet surface (propagation), finally leading to the conversion of fibrinogen into
fibrins (monomers) which are stabilized to fibrin polymers by factor XIII [13].
Once thrombin is formed, it acts as one of the most powerful platelet activators,
resulting in platelet consumption and loss of function.
Unlike cardiac surgery, during ECMO, this chain of reactions is limited by the
absence of a continuous source of soluble and cell-bound TF. Therefore, the amount
of thrombin generated is limited, albeit existent. However, other considerations
related to thrombin generation specifically apply to ECMO.
Even if an acute pattern of thrombin generation is rarely found during ECMO,
the relatively long-lasting characteristics of ECMO treatment (usually in the range
of days or weeks) induce a condition of chronic thrombin generation. Additionally,
many possible factors may enhance TF production during an ECMO treatment.
Among them, both the inflammatory reaction and possible patterns of systemic
infections may trigger the release of blood-borne TF from leukocytes.
Thrombin generation during ECMO is the reason underlying the need for anticoagulation and is the main trigger for both hemorrhagic and thromboembolic
complications.
7.2.3
After Thrombin: The Fibrinolytic System
Once thrombin is formed in excess, a pro-thrombotic state is present. The physiological reaction to this is the activation of fibrinolytic system. Plasminogen is converted to plasmin, through the release of tissue plasminogen activator from the
endothelial cells and urokinases from circulating macrophages and fibroblasts or
streptokinases from bacteria. Plasmin in turn cleaves fibrin, releasing fibrin degradation products (FDP).
During ECMO, and consequently to chronic thrombin generation, hyperfibrinolysis may occur [14] and could be one of the factors leading to hemorrhage.
7.3
Hemorrhagic and Thrombotic Complications of ECMO
Despite the improvements in techniques and materials, hemorrhagic and thromboembolic complications remain the major threat of ECMO treatment [15, 16], being
the most frequent causes of death [17]. Apart from major thromboembolic events,
including stroke, mesenteric infarction, and peripheral arterial thrombosis, microclot formation has been identified in ECMO patients and is a determinant of ischemic organ dysfunction [18, 19].
There are many factors which contribute to the risk for bleeding and thromboembolic events; they are listed in Table 7.1.
Additionally to these factors, a major role is played by the nature of the ECMO
support.
[email protected]
80
M. Ranucci
Table 7.1 Main factors contributing to hemorrhagic and thrombotic complications in ECMO
Pro-hemorrhagic factors
Excessive heparin anticoagulation
Consumption of coagulation factors
Low fibrinogen levels
Thrombocytopenia
Platelet dysfunction
Hyperfibrinolysis
Acquired von Willebrand disease
Surgical site bleeding
Pro-thrombotic factors
Inadequate heparin anticoagulation
Acquired antithrombin deficiency
Protein C-S complex consumption
Tissue factor pathway inhibitor consumption
Endothelial dysfunction
Heparin-induced thrombocytopenia
Blood stagnation in the cardiac chambers
Endotoxins
Basically, there is a great difference between venovenous ECMO, peripheral

venoarterial ECMO, and postcardiotomy ECMO. Respiratory venovenous ECMO
is usually applied through a peripheral cannulation of the femoral veins or jugular +
femoral vein or single jugular vein cannulation with a double-stage cannula.
Peripheral venoarterial ECMO is usually applied with groin vessels cannulation.
Postcardiotomy ECMO is applied in cardiac surgery patients with difficult or impossible weaning from CPB. In this case, different kinds of cannulation may be used.
Basically, a peripheral venoarterial ECMO can be placed, or, alternatively, a central
cannulation of the right atrium and ascending aorta may be chosen. In this case,
specific cannulas can be used, allowing closure of the chest.
However, even when the chest is closed, but particularly in case of open chest,
postcardiotomy ECMO carries the greatest risk for severe bleeding. This is due to
the fact that the coagulation system is already stressed by the long CPB run and by
the great amount of TF and thrombin generated during the surgical procedure;
finally, post-surgery hyperfibrinolysis, residual effects of large heparin doses,
thrombocytopenia, and surgery itself are other determinants of bleeding. On the
other side, intraoperative consumption of natural anticoagulants, like antithrombin
(AT), tissue factor pathway inhibitor, and protein C-S complexes, determines a prothrombotic condition. As addressed in the following notes on anticoagulation, postcardiotomy ECMO requires a different approach to anticoagulation and heparin
administration, at least during the first hours after implantation.
7.4
Hemostatic System Management During ECMO
7.4.1
Anticoagulation
Systemic anticoagulation during ECMO is intended to control thrombin generation

and limit the risk for thrombotic and hemorrhagic complications. Unfractionated
heparin (UFH) is the most commonly used anticoagulant [20]. UFH acts by binding
and inactivating factor Xa and thrombin; however, heparin is not a direct thrombin
inhibitor, and its efficacy is related to the presence of AT. UFH increases the kinetic
of the natural thrombin-antithrombin binding by 2,0004,000 times. Therefore, the
efficacy of heparin as an anticoagulant is strongly dependent on the AT
[email protected]
81
concentration: AT is a sort of suicide substrate, and once it is bound to thrombin

and factor Xa, it needs to be reconstituted by the liver. As a consequence, chronic
heparin administration consumes the endothelial and circulating pool of AT.
Additionally, heparin may be bound and inactivated by plasma proteins, endothelial
surface, and most of all by circulating platelets, which scavenge heparin by releasing PF4.
Due to this complex scenario, the exact dose required to correctly blunt thrombin
generation is undefined and may greatly vary during the course of an ECMO. The
classical dose range is reported between 20 and 70 IU/kg/h [20]. However, this dose
may vary among individuals and within the same subject, depending on other conditions. Basically, chronic UFH infusion leads to AT consumption and consequently
to a lower heparin sensitivity; once AT is corrected, the required dose of UFH
decreases. Simultaneously, platelet consumption is inevitable during ECMO, and
decreased levels of platelets lead to an increased sensitivity to UFH; again, once
allogeneic platelet concentrates are administered and the platelet count recovers, the
sensitivity to UFH decreases. Overall, this leads to the need for continuous adjustments of the UFH dose, even if in general there is a trend for larger doses the longer
the ECMO system is in place [21].
Even given these limitations, heparin is still necessary while on ECMO, although
some authors advocate the possibility of a heparin-free, no-anticoagulation-based
ECMO [2224]. This strategy may be considered in case of excessive bleeding risk,
like for trauma patients.
To achieve peripheral vessels cannulation, a small (50100 IU/kg) bolus dose of
heparin is usually administered. In postcardiotomy ECMO, full heparinization is
usually already achieved. In this case, a different strategy is suggested: after cannulation and onset of ECMO, heparin should be fully antagonized with protamine
sulfate. Subsequently, given the residual effects of CPB and surgery, no heparinization is usually undertaken for the first 1224 h, to avoid massive postoperative
bleeding. Once bleeding is under control, heparin infusion should be started at a low
dose (20 IU/kg/h) and subsequently adjusted to the desired level of
anticoagulation.
7.4.2
Alternatives to Heparin
Theoretically, direct thrombin inhibitors could be used as an alternative to heparin.

This is mandatory in case of heparin-induced thrombocytopenia (HIT).
Bivalirudin is a direct thrombin inhibitor with a short half-life of about 25 min and
partial (20 %) kidney clearance [25, 26]. Its use during ECMO has been successfully
reported in case of HIT [2729]; recently, a relatively large series of patients without
HIT and treated with bivalirudin as the sole anticoagulant for ECMO has been
reported and compared with conventional UFH management [21], with a lower procedural bleeding and less need for allogeneic blood products transfusions.
The dose of bivalirudin is usually reported around 0.030.2 mg/kg/h, with [27
29] or without [21] an initial bolus of 0.5 mg/kg.
[email protected]
82
M. Ranucci
There are some caveats for the use of bivalirudin in ECMO. The first is that renal
clearance may be strongly impaired in case of poor renal function, leading to drug
accumulation; the second is that, given its nature, bivalirudin anticoagulation
requires no blood stagnation in the circuit or inside the circulation. The ECMO
circuit is closed and usually does not present stagnation areas; in venovenous
ECMO, there is usually no blood stagnation inside the circulation. Conversely, in
case of venoarterial ECMO for cardiac failure, some patients present large areas of
blood stagnation inside the left heart chambers, easily detectable as a smoke effect
at echocardiographic examination. In this case, the risk for cardiac thrombi formation is high, and bivalirudin should not be used [30].
Other direct thrombin inhibitors proposed for ECMO in case of HIT include
argatroban [31] (0.10.4 g/kg/min) while danaparoid and lepirudin have been used
in the past but are presently abandoned.
7.4.3
Additional Drugs
Antiplatelet drugs have been proposed as additional agents during ECMO, in the
attempt to preserve platelet function and prevent aggregation. Some authors propose the use of aspirin (1.5 mg/kg/day) for pumpless arteriovenous ECMO [32].
The use of dipyridamole, once quite popular [33], is presently rarely reported. Apart
from anecdotical reports, no evidence exists with respect to the use of antiplatelet
agents during ECMO.
Synthetic antifibrinolytics can be used when hyperfibrinolysis is suspected
(excessive increase in FDP and d-dimers levels) [34].
The role of AT for the maintenance of a correct thrombin inhibition during UFH
therapy has already been highlighted. Inevitably, AT is consumed during ECMO,
and the majority of the authors suggest purified AT supplementation aimed to maintain AT activity at the lower normal range of 70 % [35, 36]. Of notice, when bivalirudin is used, AT consumption is strongly limited, albeit present [21].
7.5
Monitoring the Hemostatic System During ECMO
7.5.1
Activated Clotting Time (ACT)
ACT remains the standard of monitoring during heparin anticoagulation in ECMO. The
ACT provides a bedside assessment of the intrinsic and common pathway integrity.
During ECMO, the ACT is usually maintained between 180 and 220 s [20].
However, it is well established that the correlation between heparin concentration
and ACT is poor during CPB [37, 38]. However, direct measurement of heparin concentration is unpractical, and the optimal level of heparin concentration while on
ECMO has not yet been established. Studies confronting heparin concentration with
ACT values during ECMO reported variable heparin concentrations between 0.1 and
0.4 IU/mL, with correspondent ACT values ranging from 110 to 220 s [3941].
[email protected]
7.5.2
83
Conventional Laboratory Tests
Activated partial thromboplastin time (APTT) explores the intrinsic and common
pathways of coagulation and is the classical measure for heparin therapy [20].
APTT poorly correlates with ACT [42]; conversely, it has an acceptable degree of
correlation with heparin concentration [43] and is therefore to be considered superior to the ACT for heparin treatment monitoring during ECMO. An APTT of 1.5
times the baseline APTT (5080 s) is considered the target value during ECMO and
corresponds to a heparin concentration of 0.20.3 IU/mL [20].
Prothrombin time (PT) is a marker of the extrinsic and common coagulation
pathways and should be performed in order to detect the level of coagulation factors
and to guide their supplementation with fresh frozen plasma (FFP), prothrombin
complex concentrates (PCC), or cryoprecipitates.
Platelet count, fibrinogen levels, and d-dimers assays should be performed daily,
since they determine the need for platelet concentrates, FFP, fibrinogen, and
antifibrinolytics.
7.5.3
Thromboelastography and Thromboelastometry
Thromboelastography (TEG) and thromboelastometry (TEM) are dynamic tests

based on the viscoelastic properties of blood during the coagulation process. In both
tests, the time to change the physical nature of blood from liquid to gel (gel point) is
represented by a straight line and defined as r time (TEG) or coagulation time (TEM).
Once the gel point is reached, other parameters represent the kinetic of clot formation
(alpha angle) and the retraction force of the clot (maximum amplitude in TEG and
maximum clot firmness in TEM). Finally, the decrease over time of the clot strength is
an index of fibrinolysis (clot lysis index in TEG and maximum lysis in TEM) (Fig. 7.1).
TEG and TEM have a number of advantages over the routine coagulation tests.
They provide a comprehensive and dynamic analysis of coagulation kinetic, can be
done at point-of-care, provide data within about 30 min, and, finally, can be used for
detecting hyperfibrinolysis. Therefore, their use for monitoring coagulation and
anticoagulation gained wide popularity in recent years.
The r time and coagulation time are surrogates for thrombin generation and may
guide the UFH infusion rate during ECMO. There is not a universally accepted
value of r time for optimal UFH dose, but the majority of the authors report an
optimal window between two and three times the upper normal limit (1625 min)
[20, 21] (Fig. 7.2).
Both TEG and TEM offer an additional number of tests. Adding heparinase, it is
possible to detect the natural underlying behavior of clot formation. This is particularly useful when the r time is excessively prolonged, to distinguish a heparin
overload from a coagulation factor deficiency and for guiding the therapy with PCC,
FFP, or cryoprecipitates. Specific tests (functional fibrinogen in TEG and Fibtem in
TEM) provide a measurement of fibrinogen concentration (Fig. 7.3). These last
values may guide the therapy with FFP and fibrinogen.
[email protected]
M. Ranucci
84
Thrombin generation
Coagulation factors
Clot lysis
Clot firmness
Fibrin + platelets
R
min
12.2
48
K
min
2.7
04
Angle
deg
54.0
4774
MA
mm
57.4
5472
PMA
G
EPL
d/sc
%
6.7K
1.2
6.0K13.2K 015
0.0
A
mm
48.7
CI
5.7
33
LY30
%
1.2
08
Fig. 7.1 Thromboelastographic tracing
2 Kaolin with heparinase

Campione: 16/11/2009 11.36.04 PM 01.37.09 AM
ECMO
10 millimetri
R
min
29.5
48
K
min
7.8
14
Angle
deg
24.1
4774
MA
mm
49.6
5573
PMA
1.0
G
d/sc
4.9K
6.0K13.2K
EPL
%
0.0
015
A
mm
52.2
CI
22.0
33
LY30
%
0.0
08
Fig. 7.2 An adequate TEG during ECMO
FIBTEM
Fig. 7.3 ROTEM analysis

for fibrinogen concentration
(FIBTEM)
20110411 14:55
CT :
77s
CFT :
A10 :
10 mm
A20 :
11 mm
[email protected]
:
MCF :

11 mm
7.5.4
85
Other Hemostasis and Coagulation Tests
Given the limitations of ACT, and the fact that APTT values may change depending
on the laboratory method used, some authors [44] suggested the use of more specific tests for the definition of the UFH infusion rate. The anti-Xa UFH assay measures the anti-Xa activity of heparin in plasma. An optimal value, corresponding to
an APTT 1.52 times the baseline, is between 0.3 and 0.7 IU/mL [43].
Platelet function analysis with point-of-care tests during ECMO is suggested by
some authors [35]. At present, there is a gap in knowledge about platelet function
and antiplatelet drugs use during ECMO. There is not a clearly defined cutoff value
suggesting platelet concentrate transfusion, and many of the available tests may be
biased by the usually low platelet count during ECMO.
7.6
Adjusting the Coagulation Profile
Given the above reported data, it is possible to define an optimal coagulation pattern during ECMO. The main key point of this pattern is shown in Table 7.2.
Guiding the patient into the framework of this optimal pattern is one of the most
tricky steps during an ECMO management. UFH or bivalirudin dose should be
adjusted based on ACT, APTT, and TEG/TEM. The other issues can be adjusted
using allogeneic blood products or substitutes.
Purified AT is available for AT supplementation. AT may be administered using
FFP, but very large doses are required.
A severe gap in plasma coagulation factors (INR >3) can be corrected with PCC
or cryoprecipitates, whereas minor gaps (INR 23) could even be treated with FFP.
In case of life-threatening bleeding due to a lack in coagulation factors, recombinant activated factor VIIa (rFVIIa) may be considered. However, this approach has
several disadvantages: to be effective, rFVIIa must find an adequate amount of
platelets and fibrinogen; additionally, rFVIIa carries the risk for thromboembolic
events [45].
Fibrinogen, being an active-phase protein, usually progressively increases during
ECMO [21]; however, especially in postcardiotomy ECMO, during the first hours
Table 7.2 The optimal hemostatic pattern for the ECMO patient
Parameter
Activated clotting time (seconds)
International normalized ratio
R time at thromboelastography (seconds)
Fibrinogen (mg/dL)
Maximum clot firmness at FibTEM (mm)
Antithrombin activity (%)
Platelet count (cells/mmc)
d-dimers (g/L)
Suggested value
180220
1.31.5
1625
>100
>10
7080
>80,000 (bleeding patients/high risk)
>45,000 (no bleeding/low risk)
<300
[email protected]
86
M. Ranucci
after ECMO implantation, the fibrinogen levels can be very low. Suggested values
of fibrinogen should be at least 100 g/dL [20], which approximately correspond to
a maximum clot firmness >10 mm at TEM [46]. Fibrinogen concentrate is available
for supplementation; alternatively, fibrinogen can be administered as cryoprecipitates or FFP, but again very large doses of FFP are needed.
Antifibrinolytic therapy with epsilon-aminocaproic acid or tranexamic acid
should be initiated in presence of signs of ongoing hyperfibrinolysis at TEG/TEM
or conventional tests. A certain degree of fibrinolysis is always present during
ECMO; values of d-dimers around 300 g/L are acceptable, but signs of progressive
increase suggest a prompt intervention.
Platelet count should be maintained above 80,000 cells/mmc in a patient with
active bleeding or at high risk for bleeding, with platelet concentrate transfusions.
Conversely, lower values (however >45,000 cells/mmc) may be accepted in nonbleeding patients or patients at low risk for bleeding [20].
All the above figures should however be included within a management based on
the actual patients conditions: a bleeding patient requires a prompt and aggressive
approach, with allogeneic blood products and substitutes therapy guided by the
whole set of coagulation tests. Conversely, a non-bleeding patient should be treated
more conservatively, trying not to treat numbers instead of the patient.
Finally, red blood cells should be administered to maintain a hemoglobin level at
a minimal value of 8 g/dL; however, depending on the patients clinical situation,
higher target values may be necessary.
7.7
Particular Conditions During ECMO
7.7.1
Heparin-Induced Thrombocytopenia
HIT is much more common in ECMO and ventricular assist device patients than in the
rest of the patient population, with reported rates around 15 % [47, 48]. HIT may be
particularly difficult to diagnose, given the presence of many other reasons for a low
platelet count. When suspected, HIT should be ruled out with adequate diagnostic tests
for the presence of anti-PF4-heparin complex antibodies. If confirmed, heparin should
be stopped and replaced with a direct thrombin inhibitor (bivalirudin or argatroban).
7.7.2
Acquired Von Willebrand Disease
Acquired von Willebrand disease is characterized by the loss of the large multimers
of the von Willebrand factor, with a consequent defect of the platelet adhesion to the
disrupted endothelium. High shear forces produced by centrifugal pumps are
responsible for this condition, which is quite common in ECMO patients [49].
However, it is still unclear whether or not this condition is associated with clinically
relevant bleeding [49], and the need for a therapeutic approach (factor replacement
or desmopressin) remains unclear.
[email protected]
7.8
87
Systemic Inflammatory Reaction and ECMO
ECMO induces the activation of many inflammatory pathways. Some are directly activated by the contact-phase reaction to foreign surfaces, while others are triggered by
TF release and thrombin generation. From this point of view, ECMO is a perfect model
for understanding the complex interaction between inflammation and coagulation [50].
Contact with foreign surfaces activates the complement system through the alternative pathway, with the release of the anaphylatoxins C3a (alternative pathway)
and C5a (terminal pathway) [51]. Activated complement factors induce the synthesis of cytokines, belonging to both the subgroups of proinflammatory (interleukin-6
and interleukin-8, tumor necrosis factor-) and anti-inflammatory (interleukin-10)
cytokines [5254]. Proinflammatory cytokines are involved in increased vascular
permeability and endothelial dysfunction. Another inflammatory mechanism
involves endotoxins. Bacterial lipopolysaccharide is released by gram-negative bacteria and induces TNF- release by the macrophages [55] and interleukin-6 release
by endothelial cells [56]. During ECMO, like in CPB, endotoxins may be released
mainly due to bacterial translocation from a poorly perfused gut mucosa [57, 58].
Endotoxins activate circulating monocytes, which in turn release cytokines and
blood-borne TF, subsequently activating the coagulation cascade. In turn, the activation of thrombin generation promotes inflammation, leading to a vicious circle.
As for the hemostatic activation, and even to a greater degree, biocompatible
surfaces are associated with a blunting of the complement activation, neutrophil
activation, and cytokines release [26].
References
1. Vogler EA, Siedlecki CA (2009) Contact activation of blood plasma coagulation: a contribution from the Hematology at Biomaterial Interfaces Research Group the Pennsylvania State
University. Biomaterials 30:18571869
2. Fosse E, Thelin S, Svennevig JL et al (1997) Duraflo II coating of cardiopulmonary bypass
circuits reduces complement activation, but does not affect the release of granulocyte enzymes:
a European multicentre study. Eur J Cardiothorac Surg 11:320327
3. Gu YJ, van Oeveren W, Akkerman C, Boonstra PW, Huyzen RJ, Wildevuur CR (1993)
Heparin-coated circuits reduce the inflammatory response to cardiopulmonary bypass. Ann
Thorac Surg 55:917922
4. Moen O, Hogasen K, Fosse E et al (1997) Attenuation of changes in leukocyte surface markers
and complement activation with heparin-coated cardiopulmonary bypass. Ann Thorac Surg
63:105111
5. Spiess BD, Vocelka C, Cochran RP, Soltow L, Chandler WL (1998) Heparin-coated bypass
circuits (Carmeda) suppress the release of tissue plasminogen activator during normothermic
coronary artery bypass graft surgery. J Cardiothorac Vasc Anesth 12:299304
6. te Velthuis H, Baufreton C, Jansen PG et al (1997) Heparin coating of extracorporeal circuits
inhibits contact activation during cardiac operations. J Thorac Cardiovasc Surg 114:117122
7. Boisclair SJ, Lane DA, Philippou H (1993) Mechanisms of thrombin generation during surgery and cardiopulmonary bypass. Blood 82:33503357
8. Edmunds LH, Colman RW (2006) Thrombin during cardiopulmonary bypass. Ann Thorac
Surg 82:23152322
[email protected]
88
M. Ranucci
9. Gikakis N, Khan MMH, Hiramatsu Y (1996) Effect of factor Xa inhibitors on thrombin formation and complement and neutrophil activation during in-vitro extracorporeal circulation.
Circulation 94(Suppl II):341346
10. De Somer F, Van Belleghem Y, Caes F et al (2002) Tissue factor as the main activator of the
coagulation system during cardiopulmonary bypass. J Thorac Cardiovasc Surg 123:951958
11. Albes JM, Stohr IM, Kaluza M et al (2003) Physiological coagulation can be maintained in
extracorporeal circulation by means of shed blood separation and coating. J Thorac Cardiovasc
Surg 126:15041512
12. Aldea GS, Soltow LO, Chandler WL et al (2002) Limitation of thrombin generation, platelet
activation and inflammation by elimination of cardiotomy suction in patients undergoing coronary artery bypass grafting treated with heparin-bonded circuits. J Thorac Cardiovasc Surg
123:742755
13. Hoffman M, Munroe DM (2001) A cell-based model of hemostasis. Thromb Haemost
85:958965
14. Skinner SC, Hirschl RB, Bartlett RH (2006) Extracorporeal life support. Semin Pediatr Surg
15:242250
15. Polimenakos AC, Wojtyla P, Smith PJ et al (2011) Post-cardiotomy extracorporeal cardiopulmonary resuscitation in neonates with complex single ventricle: analysis of outcomes. Eur J
Cardiothorac Surg 40:13961405
16. Bartlett RH, Gattinoni L (2010) Current status of extracorporeal life support (ECMO) for
cardiopulmonary failure. Minerva Anestesiol 76:534
17. Beiras-Fernandez A, Deutsch MA, Kainzinger S et al (2011) Extracorporeal membrane oxygenation in 108 patients with low cardiac output a single-center experience. Int J Artif
Organs 34:365373
18. Fink SM, Bockman DE, Howell CG, Falls DG, Kanto WP Jr (1989) Bypass circuits as the
source of thromboemboli during extracorporeal membrane oxygenation. J Pediatr
115:621624
19. Rastan AJ, Lachmann N, Walther T et al (2006) Autopsy findings in patients on postcardiotomy extracorporeal membrane oxygenation (ECMO). Int J Artif Organs 29:11211131
20. Oliver WC (2009) Anticoagulation and coagulation management for ECMO. Semin
Cardiothorac Vasc Anesth 13:154175
21. Ranucci M, Ballotta A, Kandil H et al (2011) Bivalirudin-based versus conventional heparin
anticoagulation for postcardiotomy extracorporeal membrane oxygenation. Crit Care 15:R275
22. Muellenbach RM, Kredel M, Kunze E et al (2012) Prolonged heparin-free extracorporeal
membrane oxygenation in multiple injured acute respiratory distress syndrome patients with
traumatic brain injury. J Trauma Acute Care Surg 72:14441447
23. Lappa A, Donfrancesco S, Contento C et al (2012) Weaning from venovenous extracorporeal
membrane oxygenation without anticoagulation: is it possible? Ann Thorac Surg 94:e1e3
24. Lamarche Y, Chow B, Bdard A et al (2010) Thromboembolic events in patients on
extracorporeal membrane oxygenation without anticoagulation. Innovations (Phila)
5:424429
25. Hirsh J, O Donnel M, Weitz JI (2005) New anticoagulants. Blood 105:453463
26. Hirsh J, O Donnel M, Eikelboom JW (2007) Beyond unfractionated heparin and warfarin:
current and future advances. Circulation 116:552560
27. Pollak U, Yacobobich J, Tamary H, Dagan O, Manor-Shulman O (2011) Heparin-induced
thrombocytopenia and extracorporeal membrane oxygenation: a case report and review of the
literature. J Extra Corpor Technol 43:512
28. Pappalardo F, Maj G, Scandroglio A, Sampietro F, Zangrillo A, Koster A (2009) Bioline
heparin-coated ECMO with bivalirudin anticoagulation in a patient with acute heparin-induced
thrombocytopenia: the immune reaction appeared to continue unabated. Perfusion
24:135137
29. Koster A, Weng Y, Bttcher W, Gromann T, Kuppe H, Hetzer R (2007) Successful use of
bivalirudin as anticoagulant for ECMO in a patient with acute HIT. Ann Thorac Surg
83:18651867
[email protected]
89
30. Ranucci M (2012) Bivalirudin and post-cardiotomy ECMO: a word of caution. Crit Care
16:427
31. Young G, Yonekawa KE, Nakagawa P, Nugent DJ (2004) Argatroban as an alternative to heparin in extracorporeal membrane oxygenation circuits. Perfusion 19:283288
32. Bein T, Zimmermann M, Philipp A et al (2011) Addition of acetylsalicylic acid to heparin for
anticoagulation management during pumpless extracorporeal lung assist. ASAIO J 57:164168
33. Glauber M, Szefner J, Senni M et al (1995) Reduction of haemorrhagic complications during
mechanically assisted circulation with the use of a multi-system anticoagulation protocol. Int
34. Downard CD, Betit P, Chang RW et al (2003) Impact of AMICAR on hemorrhagic complications of ECMO: a ten-year review. J Pediatr Surg 38:12121216
35. Grlinger K, Bergmann L, Dirkmann D (2012) Coagulation management in patients undergoing mechanical circulatory support. Best Pract Res Clin Anaesthesiol 26:179198
36. Niebler RA, Christensen M, Berens R, Wellner H, Mikhailov T, Tweddell JS (2011)
Antithrombin replacement during extracorporeal membrane oxygenation. Artif Organs
35:10241028
37. Huyzen RJ, van Oeveren W, Wei F, Stellingwerf P, Boonstra PW, Gu YJ (1996) In vitro effect
of hemodilution on activated clotting time and high-dose thrombin time during cardiopulmonary bypass. Ann Thorac Surg 62:533537
38. Koster A, Despotis G, Gruendel M et al (2002) The plasma supplemented modified activated
clotting time for monitoring of heparinization during cardiopulmonary bypass: a pilot investigation. Anesth Analg 95:2630
39. Green TP, Isham-Schopf B, Irmiter RJ, Smith C, Uden DL, Steinhorn RH (1990) Inactivation
of heparin during extracorporeal circulation in infants. Clin Pharmacol Ther 48:148154
40. Green TP, Isham-Schopf B, Steinhorn RH, Smith C, Irmiter RJ (1990) Whole blood activated
clotting time in infants during extracorporeal membrane oxygenation. Crit Care Med
18:494498
41. Urlesberger B, Zobel G, Zenz W et al (1996) Activation of the clotting system during extracorporeal membrane oxygenation in term newborn infants. J Pediatr 129:264268
42. ONeill AI, McAllister C, Corke CF, Parkin JD (1991) A comparison of five devices for the
bedside monitoring of heparin therapy. Anaesth Intensive Care 19:592596
43. Hirsh J, Raschke R, Warkentin TE, Dalen JE, Deykin D, Poller L (1995) Heparin: mechanism
of action, pharmacokinetics, dosing considerations, monitoring, efficacy, and safety. Chest
108(4 Suppl):258S275S
44. Sievert A, Uber W, Laws S, Cochran J (2010) Improvement in long-term ECMO by detailed
monitoring of anticoagulation: a case report. Perfusion 26:5964
45. Swaminathan M, Shaw AD, Greenfield RA, Grichnik KP (2008) Fatal thrombosis after factor
VII administration during extracorporeal membrane oxygenation. J Cardiothorac Vasc Anesth
22:259260
46. Solomon C, Rahe-Meyer N, Schchl H, Ranucci M, Grlinger K (2013) Effect of haematocrit
on fibrin-based clot firmness in the FIBTEM test. Blood Transfus 11:412418
47. Warkentin TE, Greinacher A, Koster A (2009) Heparin-induced thrombocytopenia in patients
with ventricular assist devices: are new prevention strategies required? Ann Thorac Surg
87:16331640
48. Koster A, Huebler S, Potapov E et al (2007) Impact of heparin-induced thrombocytopenia on
outcome in patients with ventricular assist device support: single-institution experience in 358
consecutive patients. Ann Thorac Surg 83:7276
49. Heilmann C, Geisen U, Beyersdorf F et al (2012) Acquired von Willebrand syndrome in
patients with extracorporeal life support (ECLS). Intensive Care Med 38:6268
50. Paparella D, Yau TM, Young E (2002) Cardiopulmonary bypass induced inflammation: pathophysiology and treatment. An update. Eur J Cardiothorac Surg 21:232244
51. Chenoweth DE, Cooper SW, Hugli TE, Stewart RW, Blackstone EH, Kirklin JW (1981)
Complement activation during cardiopulmonary bypass: evidence for generation of C3a and
C5a anaphylatoxins. N Engl J Med 304:497503
[email protected]
90
M. Ranucci
52. Bruins P, te Velthuis H, Yazdanbakhsh AP et al (1997) Activation of the complement system

during and after cardiopulmonary bypass surgery: postsurgery activation involves C-reactive
protein and is associated with postoperative arrhythmia. Circulation 96:35423548
53. Fischer WH, Jagels MA, Hugli TE (1999) Regulation of IL-6 synthesis in human peripheral
blood mononuclear cells by C3a and C3a(desArg). J Immunol 162:453459
54. Donnelly RP, Freeman SL, Hayes MP (1995) Inhibition of IL-10 expression by IFN-gamma
up-regulates transcription of TNF-alpha in human monocytes. J Immunol 155:14201427
55. Giroir BP (1993) Mediators of septic shock: new approaches for interrupting the endogenous
inflammatory cascade. Crit Care Med 21:780789
56. Jirik FR, Podor TJ, Hirano T et al (1989) Bacterial lipopolysaccharide and inflammatory mediators augment IL-6 secretion by human endothelial cells. J Immunol 142:144147
57. Andersen LW, Landow L, Baek L, Jansen E, Baker S (1993) Association between gastric
intramucosal ph and splanchnic endotoxin, antibody to endotoxin, and tumor necrosis factor-a
concentration in patient undergoing cardiopulmonary bypass. Crit Care Med 21:210217
58. Riddington DW, Venkatesh B, Boivin CM et al (1996) Intestinal permeability, gastric intramucosal ph, and systemic endotoxemia in patients undergoing cardiopulmonary bypass. JAMA
275:10071012
[email protected]
Part II
ECMO for Circulatory Support
[email protected]
Extracorporeal Life Support:

Interactions with Normal Circulation
Michele G. Mondino, Filippo Milazzo, Roberto Paino,
and Roberto Fumagalli
The end point of any mechanical circulatory support (MCS) is to restore adequacy
of perfusion in order to prevent organ damage or to restore normal organ function
when damage is already commenced. Reduction of left ventricular end-diastolic
pressure, cardiac wall tension, and pulmonary congestion, together with a modulation of the neurohormonal response to acute and chronic heart failure, like endogenous catecholamine, renin-angiotensin system, ANP, and cytokine release, are all
aims of mechanical circulatory support.
Myocardial recovery from reversible acute cardiogenic shock or cardiac reverse
remodeling in chronic heart failure is what we ultimately would like to achieve in
our patients. When this is not possible, but recovery of organ dysfunction has been
reached, MCS offers other options: bridge to transplant, bridge to bridge, or destination therapy. These end points, in particular destination therapy, can be obtained
with left ventricular assist devices (L-VAD). While L-VAD is a one-ventricle support system and requires a normally functioning right ventricle, extracorporeal life
support (ECLS) is a bi-ventricular and respiratory support, acting as a heart and
lung bypass, thus reproducing the complete cardiopulmonary bypass (CPB) adopted
for open-heart surgery.
ECLS, or more commonly ECMO (extracorporeal membrane oxygenation), is
nowadays widely used to support the patients respiratory system or both circulatory
and respiratory systems in life-threatening clinical conditions. The ECLS concept is
M.G. Mondino, MD F. Milazzo, MD R. Paino, MD

Department of Cardio-Thoracic-Vascular Anaesthesia and Intensive Care,
Ospedale Niguarda CaGarnda, Piazza Ospedale Maggiore 3, Milan 20162, Italy
e-mail: [email protected], [email protected];
[email protected]; [email protected]
R. Fumagalli, MD (*)
Department of Anaesthesia and Intensive Care, Ospedale Niguarda CaGranda,
Piazza Ospedale Maggiore 3, Milan 20162, Italy
Dipartimento di Scienza della Salute, Universit Milano Bicocca, Milan, Italy
e-mail: [email protected], [email protected]
[email protected]
93
94
M.G. Mondino et al.
quite straightforward: venous blood is drained from the body into the artificial lung
and pumped back into the patients circulatory system.
The relative simplicity of this circuit though can modify patients physiology in
different ways. In fact, once the patient starts to be supported by ECMO, depending
on the type and location of cannulas and according to the underlying clinical
conditions, different hemodynamic changes are foreseeable. When blood exiting
the artificial lung returns into the patients venous system, we talk about venovenous
ECMO (VV ECMO), and our system is indeed in series with the patients cardiopulmonary physiology. When the blood leaving our circuit is returned into the
patients artery, we are talking about venoarterial ECMO (VA ECMO), which is
instead parallel. These basic concepts together with the abnormal physiology of the
patient needing extracorporeal support must be taken into account when considering the physiology of extracorporeal circulation.
We will look at VA and VV ECMO separately, to analyze how these can impact
on circulatory and respiratory systems, and the possible complications that can arise.
8.1
VA ECMO
When venous blood is completely drained by the ECMO pump, whether it is an older
roller pump or a more modern centrifugal pump, 100 % of the pulmonary circulation
will be bypassed and arterial pulsatility due to residual cardiac activity will cease.
Blood flow generated by ECMO will be therefore continuous according to the flattened pulmonary and systemic traces on the monitor. However a certain amount of
blood, from the sinus venosus and from the bronchial and Thebesian circulation, will
still flow into the left ventricle, which, when adequately preloaded, may have a pulsatile beat. This will appear as an irregular pulsatility on the systemic arterial trace.
Therefore, a complete bypass is possible only when a pulmonary vent is in place,
as it happens with the heart and lung machine used in cardiac surgery. This cannot
be easily obtained by VA ECMO, especially when ECMO is achieved by peripheral
cannulation and the heart maintains some residual contractility. As a consequence
VA ECMO is considered able to provide about 80 % of the cardiac output at rest,
while the remaining 20 % will flow through the pulmonary circulation into the left
ventricle. As described below, the incomplete drainage of the cardiac output by VA
ECMO could lead to a problematic management of severely compromised patients.
More than one study agree that the level of arterial blood flow, either continuous
or pulsatile, must be above 8090 ml/kg when addressed to counteract cardiac
shock, metabolic acidosis, high level of endogenous catecholamines, and the development of low urine output [1]. The critical threshold is considered to be 4050 ml/
kg. Below this, inadequate oxygen delivery, cardiocirculatory shock, anaerobic
metabolism, and acidosis will occur independently of the type of blood flow. At
intermediate levels, pulsatile flow can partially compensate the effect of hypoperfusion and acidosis. This is because aortic and carotid baroreceptors are strongly
stimulated by non-pulsatile flow with consequent release of endogenous catecholamines and deleterious effects on the microcirculation [1].
[email protected]
Extracorporeal Life Support: Interactions with Normal Circulation
95
Usually VA ECMO is started at 2.6 l/m2, and adequate systemic blood flow and
oxygen supply are guaranteed by parameters listed in below
Pump flow >2 l/min/m2
Ht >33 %
PaO2 >100 mmHg
MAP = 6090 mmHg (vasodilators or vasopressors)
Oxygen delivery will be mostly determined by hemoglobin level and extracorporeal blood flow, while the ECMO oxygenator can easily modify PaO2.
Physiologic values of mean arterial pressure are necessary to provide adequate
tissue perfusion and can be modulated with the appropriate use of vasodilators or
vasopressors. When considering mean arterial pressure, though, the level of systemic vascular resistances must be taken into account as the efficacy of the centrifugal pump (differently from the roller pump), and therefore the resulting blood flow
depends not only on the amount of venous blood drained (preload) and the pump
power (RPM) but also on the resistances created by the circuit and the patients
vascular system (afterload).
Clinically, normal skin temperature, normal capillary refill time, arterial blood
pH normalization, reduction of lactates, and increased urine output are all signs of
improved patient perfusion. It is worth noting that non-pulsatile flow can have an
antidiuretic effect by direct stimulation of the juxtaglomerular apparatus. This is
usually easily controlled by low doses of diuretics [1].
The use of the mixed venous saturation value (SvO2) from pulmonary artery
catheter as a parameter of good peripheral perfusion deserves special mention when
considered in a patient on VA ECMO. As a matter of fact we can sometimes see
abnormally high values of SvO2, which do not reflect the real relation between DO2
and VO2 but, instead, a phenomenon of equilibrium with pulmonary capillary blood
(retrograde flow hypothesis).
This can be seen when ECMO is achieving almost complete bypass of the pulmonary circulation, and therefore, pulmonary artery blood flow is low. Paradoxically,
as the patient recovers, the SvO2 will decrease: this occurs because as the percentage of cardiac output passing through the native heart and pulmonary circuit
increases, the PaO2 will decrease [2]. The ETCO2 values can help us better understand our data. An increase in ETCO2 values as well as a reduction in alveolararterial CO2 difference are indirect indices of increased pulmonary blood flow. This
can be taken into account when dealing with weaning from ECMO. Alternatively, a
more accurate assessment of venous saturation can be obtained by analyzing blood
before entering the oxygenator or directly at the right atrium (ScVO2).
An additional fringe benefit rarely underlined is that, during ECMO, the
patients temperature is easily controlled, thus avoiding detrimental increases in
VO2, and when needed, moderate protective hypothermia can be instituted [3, 4].
During VA ECMO, the suction of blood from the central venous system
determines right ventricle unloading. However, adequate left ventricle unloading
can be problematic, and this will be considered when adjusting the level of ECMO
[email protected]
96
M.G. Mondino et al.
support or the patient therapy. The reasons behind inadequate left ventricle unloading of the patient on ECMO are different and not always obvious and are to be found
both on preload and afterload of the left ventricle. As stated previously, while
ECMO owes its origins to the heart and lung machine, its setup is indeed different.
The lack of a venous reservoir and left ventricle venting, in order to simplify the
system to a bedside support and to allow, when needed, percutaneous access, has, as
side effect, a certain amount of blood not drained. As such, the bronchial and
Thebesian blood flow will continue to fill the left-side cavities. If residual myocardial contractility is not enough to provide stroke volume, this gradual and continuous filling will eventually lead to overdistension and high pressure in the left atrium
and left ventricle. Therefore, even though the unloading of the right ventricle determines a reduction in pulmonary blood flow, left ventricle preload can still be
problematic.
Bavaria and colleagues [5] demonstrated that VA ECMO decreases left ventricle
wall stress in normal hearts. However, a progressive rise in wall stress in postischemic hearts occurs with an increase in ECMO flow rate as a result of a concomitant
increase in afterload. Left ventricle afterload increases as a consequence of blood
return via the arterial cannula, and this is of particular concern in peripheral VA
ECMO [6].
The combination of increased afterload from the arterial cannula and the underlying myocardial dysfunction can lead to high values of left atrial and left ventricular end-diastolic pressure. This will increase wall stress and myocardial oxygen
consumption, worsening left heart failure. Increased left atrial pressure, even more
relevant if associated with severe mitral regurgitation, results in pulmonary congestion, pulmonary edema, and in extreme cases, pulmonary hemorrhage which can
lead to irreversible pulmonary failure. When severe left ventricular dilatation and
dysfunction are present, the left ventricle may become unable to generate enough
pressure to actually open the aortic valve. This will appear as a loss of pulsatility on
arterial pressure tracings and can lead to stasis and thrombosis in the ascending
aorta, left ventricle cavity, and pulmonary veins. In this case, anticoagulation must
be carefully titrated, and left ventricle afterload must be reduced by optimizing
native left ventricle output, thus facilitating aortic valve opening.
As previously stated, blood flow generated by the ECMO pump is countercurrent to
that coming from the heart. It follows that the upper and lower parts of the body could
receive differently oxygenated blood. Areas closer to the outflow cannula (i.e., arterial
cannula) will receive blood straight from the oxygenator, rich in oxygen and cleared of
carbon dioxide, while those areas further away from the ECMO oxygenator, but closer
to the left ventricle outflow, will receive blood that is oxygenated and cleared of carbon
dioxide from the native lungs. There is then a third area where O2 and CO2 content will
be a flow-weighted average of the O2 and CO2 contents of two flows.
The exact location and prevalence of these three areas vary according to the type
of support (central vs peripheral) and the level of flow (high vs low ECMO flow).
In case of peripheral ECMO, where arterial flow is delivered through the femoral
artery, a low flow will determine a good oxygenation of subdiaphragmatic regions,
a mixed zone at thoracic descending aorta, while blood flowing through supra-aortic
[email protected]
97

Mixing zone
(Poorly) oxygenated
blood from Iung
Mixing zone
LV
LV
Peripheral
VA ECMO
Low flow
Peripheral
VA ECMO
Moderate flow
Fig. 8.1 Distribution of arterialized blood during peripheral VA ECMO with low (on the left) or
moderate flows (on the right) delivered to femoral artery. If there is significant pulmonary parenchymal disease or inadequate mechanical ventilation during ECMO support, hypoxic blood returning to the LV provides the sole source of myocardial and cerebral perfusion
vessels and coronary arteries will be that coming from the patients native lungs
(Fig. 8.1, left).
When a moderate level of flow is used in the setting of peripheral ECMO
(Fig. 8.1, right), the area of mixed blood is shifted towards the left subclavian artery.
In this situation, in case of concomitant impairment of pulmonary function, we
could witness the Harlequin syndrome: locoregional and asymmetric discrepancies
in blood flow distribution appearing as differences in skin color, in different parts of
the body, with the result of a patient with a blue head, red legs, and different
oxygen saturation between the left and right arm.
Possible strategies, in case of hypoxemia to the supra-aortic territories, are:
Increase oxygenator FiO2
Increase pump speed and pump flow in order to obtain maximal RV and LV
unloading, decrease blood flow through inefficient lungs, and shift competitive
blood flow before supra-aortic arteries
In the latter option, when pump speed and flow are increased, coronary flow
would be provided by the left ventricle (i.e., with blood coming from the patents
native lungs) and only if there is any appreciable ventricular ejection, while oxygenated blood from the arterial cannula may fail to reach the coronary artery [7, 8]
(Fig. 8.2, right).
Optimal coronary oxygenation can be obtained through central VA ECMO
(Fig. 8.2, left), whether this is intrathoracic (outflow cannula directly in the ascending aorta) or extra thoracic (outflow cannula in the right subclavian artery or the
right carotid artery) at the expense of a possible higher afterload when compared
with peripheral VA ECMO. Direct venting of the LV can obviate to this problem
and is easy to obtain in central intrathoracic ECMO.
[email protected]
M.G. Mondino et al.
98
(Poorly) oxygenated
blood from Iung
Mixing zone
To Asc. Aorta
Peripheral
VA ECMO
To right subclavian
or carotid artery
Central VA ECMO
High flow
Fig. 8.2 Distribution of arterialized blood during peripheral VA ECMO with high flow delivered
to femoral artery (on the right) and during central VA ECMO with flow delivered to aortic root (on
the left)
Peripheral VA ECMO can therefore worsen myocardial damage through inadequate oxygenation of the blood reaching the coronary arteries [8, 9]. When pulmonary gas exchange is severely impaired by parenchymal disease and pulmonary
edema, or when the setting of mechanical ventilation is not adequate, hypoxic blood
returning to the left ventricle may provide the sole source of coronary perfusion
with deleterious effects on ventricular function and myocardial recovery [10].
Concomitant clinical conditions like sepsis, acidosis, or hypoxia may contribute to
the decrease in cardiac performance.
Inadequate perfusion of the coronary arterial flow may also occur: the increase in
left ventricular end-diastolic pressure during VA ECMO could result in an increase
in coronary vascular resistance and consequent decrease in coronary flow.
Physiologically, the aortic root expands during systole, and it acts as a blood
reservoir for coronary perfusion in diastole. The lack of pulsation during ECMO, as
well as a decrease in cardiac output, will result in a reduction of this function. In
addition, cardiac output is decreased in inverse proportion to the ECMO flow.
Therefore, coronary arterial flow could decrease as VA ECMO flow increases so
that high-flow VA ECMO could exert undesirable hemodynamic effects on the left
ventricle. Therefore, especially when dealing with peripheral VA ECMO, extreme
care must be taken to ensure that the heart is ejecting adequately oxygenated blood
to perfuse not only the coronaries but also the cerebral circulation, thus avoiding
disastrous anoxic/hypoxic injuries. Inadequate saturation may not be immediately
obvious: peripheral arterial blood gas analysis, according to the ECMO setting,
[email protected]
99
could show fully saturated blood and may not reflect the oxygen level in the aortic
root. In peripheral ECMO, right radial arterial analysis may better reflect the level
of blood oxygenation to the heart and brain and unmask the oxygenation discrepancy between the upper and the lower half of the body. At this level it is possible to
see the actual saturation of the blood that perfuses the myocardium as well as the
brain and to realize whether there is an oxygenation discrepancy between the upper
and the lower half of the body. To correctly monitor saturation in the supra-aortic
areas, the placement of a saturation probe at the right earlobe or right hand is advisable. NIRS is another useful parameter addressed to monitor cerebral oxygenation
in a continuous fashion especially because it is not influenced by the absence of
arterial pulsation.
The combination of left ventricle distension, augmented afterload from arterial
peripheral cannula, inadequate oxygenation of the myocardium by inefficient coronary perfusion, associated with the underlying disease, metabolic acidosis and pulmonary dysfunction can together lead to progressive worsening of left ventricle
function. Insufficient left ventricle unloading can lead to pulmonary congestion and
lung edema and blood stagnation in the left ventricle with an increased risk of systemic embolic complications and, ultimately, hinder myocardial recovery.
Rhythm instability, ventricular fibrillation, and asystole may supervene when a
patient is supported on ECMO. When such rhythm alterations are present, even if
systemic perfusion is adequately maintained by the extracorporeal system, these
will further worsen inadequate left ventricle unloading. For all these reasons it is
abundantly clear why adequate unloading of the left ventricle during VA ECMO is
of extreme concern.
Different strategies can be used to avoid these possible complications. Restoration
of an adequate perfusion rhythm is essential. Then, sufficient inotropic support
should be maintained in order to improve left ventricular contractility and to reduce
left ventricular distension [11] and clot formation, even when VA bypass can provide adequate systemic pressure.
In situations of severe myocardial dysfunction, when the heart is unable to generate enough force to open the aortic valve and overcome pressure created by the AV
circuit, systemic vascular resistances need to be decreased with the appropriate use
of vasodilators [12].
The combinations of ECMO and intra-aortic balloon pump (IABP), when feasible, can be extremely beneficial. IABP improves diastolic filling and lowers coronary vascular resistances, thus improving coronary blood flow. Moreover, intra-aortic
balloon counterpulsation alone significantly reduces afterload, thus reducing myocardial wall stress and oxygen consumption, and ultimately improves contractility
and myocardial recovery. IABP has the additional advantage, like peripheral ECMO,
that it can be quickly inserted and started at the bedside, with minimal or no surgical
intervention. Because of this beneficial effect, the concomitant use of IABP and
ECMO is recommended [1315]. In those cases in which the use of catecholamines
together with vasodilators and IABP is not sufficient to adequately unload the left
ventricle, left-side venting should be considered. Direct insertion of a vent in the left
atrium or left ventricle can be accomplished through thoracotomy [16]. However,
the risk of bleeding is significant in patients on ECMO, especially if we are not in
[email protected]
100
M.G. Mondino et al.
the setting of central ECMO and the chest has not been previously opened. Other
less-invasive methods to vent the left ventricle have been described widely in the
literature: transseptal atriotomy creating a small atrial septal defect under fluoroscopic guidance [17], anterograde left ventricle unloading through a transaortic
venting catheter [18, 19] or catheterization of the pulmonary artery by a small catheter to allow retrograde decompression into the right atrium [2022] . Several recent
reports have described the successful use of an Impella Recover 2.5 as a vent for the
left ventricle [23, 24]. The use of transthoracic and transesophageal echocardiographic guidance has also been described to perform atrial septostomy [25].
In order to preserve long-term patient homeostasis and avoid or treat organ damage, maintenance of adequate level of venous pressure is important as well. Among
splanchnic organs, the liver appears to be the most vulnerable to hypoperfusion.
Cardiac failure-induced hepatic dysfunction may progress despite adequate hemodynamics with mechanical circulatory support. Systemic hypotension alone does
not account for hypoxic hepatitis, while venous congestion predisposes the liver to
injury induced by a hypotensive event [26]. Any higher pressures in the venous
system of the body will interfere with portal circulation because of the low-pressure
gradient and the absence of a valvular mechanism available in this circulation.
A reduction in portal vein flow is also caused by concomitant vasoconstriction due
to systemic neuroendocrine responses.
Diminished O2 supply to the liver due to reduced portal venous flow together
with hepatic congestion could cause hypoxia of hepatocytes (hepatic hypoxia),
which causes centrolobular damage. The biological hallmark of centrolobular liver
cell necrosis is a massive increase in serum aminotransferase levels. It is therefore
necessary to keep the venous pressure as low as possible to preserve the portal circulation. Maintaining a low CVP will affect not only the liver function but also the
renal and intestinal circulation [27].
8.2
VV ECMO
Venovenous bypass, in general, has no major hemodynamic impact, because blood

is drained from, and returned to, the right ventricle without a specific change in right
ventricle preload and with no adverse effect in case of a normally ejecting left ventricle. Despite this, hemodynamic derangement is often observed when the bypass
starts: several reasons have been proposed (dilutional hypocalcemia or catecholamines, respiratory alkalosis, hypothermia); in order to minimize this alteration, a
very slow increment of extracorporeal blood flow is recommended.
Some aspects though need to be considered, especially when the right ventricle
is failing to eject against high pulmonary pressures. RV function may be adversely
impacted by significant hypoxemia, subsequent increase in pulmonary vascular
resistance, and possible concomitant sepsis. When a patient is supported by VV
ECMO, the level of mechanical ventilation is usually lowered towards protective
[email protected]
101
lung ventilation or, at times, even suspended. The consequent reduction in intrathoracic pressures will determine a reduction in pulmonary resistances and an increase
in right ventricular preload and contractility. Venous oxygen saturation is increased
during VV ECMO, and therefore the pulmonary circulation is perfused with blood
high in oxygen content. The reduction in pulmonary vascular resistances that this
causes will positively lower right ventricle afterload. Moreover, improvement in left
ventricle contractility may also be observed secondary to an increase in oxygenation
of the myocardium through coronary arteries. The overall increase of heart oxygenation, in the absence of any hemodynamic interference, explains the possible beneficial effect of VV ECMO on the patients hemodynamics. Correction of acidosis and
clearance of carbon dioxide will also lower pulmonary pressures and improve cardiac contractility [28, 29].
Ultimately, reducing pulmonary vascular flow potentially modulates the endothelial activation and aggravation of pulmonary edema secondary to reperfusion injury.
A special consideration should be paid to the common parameters used to monitor critically ill patients: SvO2 is abnormally elevated depending mainly on the ratio
of extracorporeal blood flow/cardiac output, and it loses its meaning to reflect the
adequacy of tissue perfusion. However the arterial to venous gap gives consistent
information on the amount of gas exchange through the natural lung.
VV ECMO can alter PaO2 in completely opposite ways: most of the time the
increase of oxygen content provided by the artificial lung, increasing the PvO2, is
followed by the increase of PaO2 (i.e., provided that the amount of oxygen transfer
through the natural lungs is constant: the higher the PvO2, the higher PaO2 will be);
however, if respiratory acidosis, hypercapnia, and venous low oxygen are corrected
by VV ECMO, pulmonary vasoconstriction is relieved, and a different V/Q match
is obtained [30].
Clinical advantages of VV ECMO include relative technical ease of cannulation,
increased aortic oxygen saturation, reduced risk of systemic embolization, and the
possibility of a higher flow rate, but on the other hand, this does not provide any
circulatory support.
VA approach instead provides circulatory support to facilitate early recovery
from the ALI process and provides relief of the pulmonary circulation and reduction
of high pulmonary pressure, almost uniformly present.
The most beneficial type of ECMO, whether it is VA or VV, when dealing with
primary graft dysfunction after lung transplantation is still a matter of debate and
varies from center to center. It is clear that both techniques have their pros and cons.
8.3
Conclusion
The influence of an extracorporeal support on homeostasis is substantial and differs

depending on the type of cannulation used. The pathophysiologic derangements
should guide the clinician to the type of support required.
[email protected]
102
M.G. Mondino et al.
References
1. Bartlett RH (2005) Physiology of ECLS. In: Van Meurs K, Lally K, Peek G, Zwischenberger
J (eds) ECMO extracorporeal cardiopulmonary support in critical care, 3rd edn. ELSO, Ann
Arbor
2. Marasco SF, Lukas G, McDonald M et al (2008) Review of ECMO (extra corporeal membrane
oxygenation) support in critically ill adult patient. Heart Lung Circ 17(Suppl 4):S41S47
3. The Hypothermia after Cardiac Arrest Study Group (2002) Mild therapeutic hypothermia to
improve the neurologic outcome after cardia arrest. N Engl J Med 346:549556
4. Horan M, Ichiba F, Firmin RK et al (2004) A pilot investigation of mild hypothermia in neonates receiving extracorporeal membrane oxygenation (ECMO). J Pediatr 144:301308
5. Bavaria JE, Ratcliffe MB, Gupta KB et al (1988) Changes in left ventricular systolic wall stress
during biventricular circulatory assistance. Ann Thorac Surg 45:526532
6. Hoefer D, Ruttmann E, Poelzl G et al (2006) Outcome evaluation of the bridge to bridge concept in patients with cardiogenic shock. Ann Thorac Surg 82:2834
7. Nowlen TT, Salley SO, Whittlesey GC et al (1989) Regional blood flow distribution during
extracorporeal membrane oxygenation in rabbits. J Thorac Cardiovasc Surg 98(6):11381143
8. Kato J, Seo T, Ando H et al (1996) Coronary arterial perfusion during venoarterial extracorporeal membrane oxygenation. J Thorac Cardiovasc Surg 111:630636
9. Shen I, Levy FH, Vocelka CR et al (2001) Effect of extracorporeal membrane oxygenation on
left ventricular function of swine. Ann Thorac Surg 71:862867
10. Baldwin JT, Duncan BW (2006) Ventricular assist devices for children. Prog Pediatr Cardiol
21:173184
11. Schwarz B, Mair P, Margreiter J et al (2003) Experience with percutaneous venoarterial cardiopulmonary bypass for emergency circulatory support. Crit Care Med 31(3):758764
12. Chen YS, Yu HY, Huang SC et al (2005) Experience and result of extracorporeal membrane
oxygenation in treating fulminant myocarditis with shock: what mechanical support should be
considered first? J Heart Lung Transplant 24:8187
13. Doll N, Fabricius A, Borger MA et al (2003) Temporary extracorporeal membrane oxygenation in patients with refractory postoperative cardiogenic shocka single center experience.
J Card Surg 18(6):512518
14. Smedira NG, Blackstone EH (2001) Postcardiotomy mechanical support: risk factors and outcomes. Ann Thorac Surg 71(3 Suppl):S60S66; discussion S82S85
15. Murashita T, Eya K, Miyatake T, Kamikubo Y et al (2004) Outcome of the perioperative use
of percutaneous cardiopulmonary support for adult cardiac surgery: factors affecting hospital
mortality. Artif Organs 28(2):189195
16. Pagani FD, Aaronson KD, Dyke DB et al (2000) Assessment of extracorporeal life support to
LVAD bridge to heart transplant strategy. Ann Thorac Surg 70:19771985
17. Johnston TA, Jaggers J, McGovern JJ et al (1999) Bedside transseptal balloon dilation atrial
septostomy for decompression of the left heart during extracorporeal membrane oxygenation.
Catheter Cardiovasc Interv 46(2):197199
18. Shibuya M, Kitamura M, Kurihara H et al (1997) Significant left ventricular unloading with
transaortic catheter venting during venoarterial bypass. Artif Organs 21(7):789792
19. Fumagalli R, Bombino M, Borelli M et al (2004) Percutaneous bridge to heart transplantation
by venoarterial ECMO and transaortic left ventricular venting. Int J Artif Organs 27(5):
410413
20. Scholz KH, Figulla HR, Schrder TT et al (1995) Pulmonary and left ventricular decompression by artificial pulmonary valve incompetence during percutaneous cardiopulmonary bypass
support in cardiac arrest. Circulation 91(10):26642668
21. Foti G, Kolobow T, Rossi F et al (1997) Cardiopulmonary bypass through peripheral cannulation with percutaneous decompression of the left heart in a model of severe myocardial failure.
ASAIO J 43(6):927931
[email protected]
103
22. Avalli L, Maggioni E, Sangalli F et al (2011) Percutaneous left-heart decompression during

extracorporeal membrane oxygenation: an alternative to surgical and transeptal venting in
adult patients. ASAIO J 57:3840
23. Koeckert MS, Jorde UP, Naka Y et al (2011) Impella LP 2.5 for left ventricular unloading during venoarterial extracorporeal membrane oxygenation support. J Card Surg 26:666668
24. Chaparro SV, Badheka A, Marzouka GR et al (2012) Combined use of impella left ventricular
assist device and extracorporeal membrane oxygenation as a bridge to recovery in fulminant
myocarditis. ASAIO J 58:285287
25. Aiyagari RM, Rocchini AP, Remenapp RT et al (2006) Decompression of the left atrium during extracorporeal membrane oxygenation using a transseptal cannula incorporated into the
circuit. Crit Care Med 34:26032606
26. Seeto C, Fenn B et al (2000) Ischemic hepatitis: clinical presentation and pathogenesis. AM J
Med 1:109113
27. Nosae Y (1996) Is it necessary to use metabolic assist for multiorgan failure with left ventricular assist device? No, it should be circulatory assist for splanchnic organs. Artif Organs 20:1
28. Wigfield CH, Lindsey JD et al (2007) Early institution of extracorporeal membrane oxygenation for primary graft dysfunction after lung transplantation improves outcome. J Heart Lung
Transplant 26:331338
29. Mason DP,MD, Boffa DJ et al (2006) Extended use of extracorporeal membrane oxygenation
after lung transplantation. J Thorac Cardiovasc Surg 132:954960
30. Rossaint R, Hahn SM, Pappert D et al (1995) Influence of mixed venous PO2 and inspired O2
fraction on intrapulmonary shunt in patients with severe ARDS. J Appl Physiol 78(4):
15311536
[email protected]
ECMO for Ischemic Cardiogenic Shock

Francesco Formica, Fabio Sangalli, and Antonio Pesenti
9.1
Introduction
Cardiogenic shock (CS) is defined as a state of tissue hypoperfusion induced by

cardiac failure after correction of preload [1].
Although it can be induced by virtually every cause of myocardial dysfunction,
acute myocardial infarction (AMI) and its complications represent the leading cause
of CS. Other common causes include postcardiotomic shock, myocarditis, pulmonary embolism, and acutely decompensated chronic heart failure.
The incidence of CS in patients with AMI is of 67 % in recent datasets [2],
although these figures are very likely underestimated since they do not take into
account prehospital deaths. Despite improvements in supportive therapies and
reperfusion strategies, the mortality rate remains unacceptably high and CS represents the most common cause of death in patients hospitalized for AMI.
F. Formica (*)
Department of Science and Translational Medicine, Cardiac Surgery Clinic,
San Gerardo Hospital, University of Milano-Bicocca,
Via Pergolesi 33, Monza 20900, Italy
F. Sangalli
A. Pesenti
Department of Health Science, San Gerardo Hospital, University of Milano-Bicocca,
[email protected]
105
106
F. Formica et al.
In this dramatic setting, the prompt institution of a peripheral mechanical circulatory support is able to stabilize the patient and buy time for diagnostic and therapeutic procedures, which can be carried out while maintaining an adequate tissue
perfusion.
9.2
Epidemiology and Pathophysiology
The true incidence of CS is difficult to determine, as prehospital deaths are not

accounted for. Moreover, different definitions across the literature lead to important
variations in the reported rate.
The first reported note of cardiogenic shock in the setting of AMI dates back to
1794, while Griffith and colleagues published the first large case series in 1954. In
this report, the authors described an incidence of 19.7 % of CS among patients
admitted to the hospital for AMI, with a mortality rate of 80 % [3].
From those early days, several changes were introduced in clinical practice: the
introduction of coronary care units, the intra-aortic counterpulsation, the first experiences with mechanical circulatory support, the pulmonary artery catheter, and the
percutaneous and surgical revascularization techniques, to name the most relevant.
In the face of all these innovations and the observed reduction of CS in patients
with AMI, only a modest reduction in mortality was recorded in patients who eventually develop cardiogenic shock. In fact, Goldberg and colleagues reported a mortality rate of 65.4 % in AMI patients who develop CS, as compared to a 10.6 % in
those without CS [2].
CS is associated with elevated mortality and morbidity despite the use of conventional therapy with inotropic agents and intra-aortic balloon pump (IABP). In the
recently published IABP-SHOCK II, the use of IABP for the treatment of advanced
CS did not appear to improve the early outcome in patients with AMI complicated
by heart failure [4]. Although the results of this trial were criticized, these must be
taken into account.
CS is often a complication of ST elevation myocardial infarction particularly
anterior AMIs but non-ST elevation infarctions can also result in cardiogenic
shock. Autopsy studies suggest that the loss of more than 40 % of LV myocardium
is required to develop cardiogenic shock, in the absence of a mechanical complication [57]. CS can also be a sequel of a mechanical complication of infarction, such
as a ventricular septal defect, left ventricle free wall rupture, or acute mitral regurgitation due to papillary muscle rupture.
Risk factors for the development of CS include older age, prior myocardial
infarction, female gender, diabetes mellitus, and anterior MI.
The initial occlusion of the coronary artery initiates a vicious circle leading to a
progressive worsening in myocardial function. Myocardial ischemia impairs
myocardial contractility and leads to a reduction in ventricular performance with a
consequent reduction in arterial pressure and hence in coronary perfusion pressure.
Together with an alteration in oxygen extraction and the loss of coronary autoregulation, this leads to further extension of the ischemic insult. Preload is initially increased
[email protected]
107
by renal compensatory mechanisms leading to fluid retention, whereas the vasoconstrictor response to hypotension increases afterload and oxygen consumption.
Increased demand and inadequate perfusion further worsen myocardial ischemia and
cardiac function, and if this vicious circle is not promptly interrupted, it can lead to
irreversible shock and ultimately to death. The pathophysiological concept of combined low cardiac output and high systemic vascular resistances has been recently
challenged by the observation that post AMI, CS may be associated with relative
vasodilation rather than vasoconstriction. This is likely due to a systemic inflammatory response syndrome (SIRS) similar to that seen in sepsis [8] due to an inappropriate production and utilization of nitric oxide, which in turn leads to vasodilation with
reduced systemic and coronary perfusion pressures. Lim and colleagues found that
several patients with CS died despite normalization of their cardiac index, suggesting
a maldistribution effect with low systemic vascular resistance [9].
Another important cause of CS in AMI is ischemia remote from the infarct zone.
The typical response of the uninjured myocardium during AMI is represented by a
compensatory hyperkinesis. Patients who develop CS on the contrary generally
present with a multivessel coronary disease so that several perfusion territories
demonstrate a pressure-dependent perfusion [10].
A further impairment of the ventricular function is determined by the extension
of the ischemic area to adjacent myocytes. These are particularly susceptible to
ischemia and have a reduced reserve to face inadequate perfusion and the increase
in oxygen demand imposed by endogenous and exogenous catecholamines.
Stunned and hibernated myocardium represent an additional cause of cardiogenic shock in the acute phase, but their function can be restored if these regions are
properly managed and promptly revascularized, and contribute to recovery.
Hibernation follows the restoration of a nearly normal perfusion. Oxidative stress
leads to an altered response of myofilaments to calcium, which in turn causes damage in the contractile apparatus when the myocardium is reperfused [11]. The
recovery of oxidative lesions and the resynthesis of contractile proteins seem to be
the mechanism of hibernated myocardium recovery which is observed up to 6 weeks
after the primary insult.
Right ventricular AMI can also lead to CS with a slightly different mechanism.
The impairment in both diastolic and systolic dysfunction of the right ventricle
results in a volume-sensitive state in contrast to the pressure-sensitive state seen in
left ventricular infarction [12]. The ensuing damage may be unresponsive to fluids
and lead to a poor prognosis.
9.3
Clinical Presentation and Diagnosis
The recognition of CS is founded on clinical and hemodynamic parameters. Patients

present mostly hypotensive, with signs of inadequate peripheral perfusion (altered
sensorium, cool extremities, oliguria, acidosis).
There are no sharp cutoff values to define the hemodynamics of CS. However,
generally accepted parameters include a systolic blood pressure of less than
[email protected]
108
F. Formica et al.
Table 9.1 Cardiogenic shock criteria

Hemodynamic criteria
Systolic blood pressure (SBP) below 90 mmHg (or more than 30 mmHg below basal in
hypertensive patients) for more than 30 min
Use of vasopressors and inotropes to keep SBP above 90 mmHg
Cardiac index of less than 2.2 L/min/m2
Pulmonary artery occlusion pressure of above 15 mmHg
Signs of tissue hypoperfusion
Pale, cool, and clammy peripheries
Prolonged capillary refill times
Altered sensorium
Oliguria/anuria (less than 0.5 mL/kg/h)
Signs of pulmonary congestion
Tachycardia
Lactic acidosis
Mixed venous saturation below 65 %
90 mmHg for more than 30 min, a mean arterial blood pressure below 60 mmHg,
oliguria (less than 0.5 mL/kg/h), a cardiac index of less than 2.2 L/min/m2 with
inotropic support or less than 1.8 L/min/m2 without support, and elevated filling
pressures (left atrial pressure above 18 mmHg and/or right atrial pressure greater
than 15 mmHg, pulmonary artery occlusion pressure more than 15 mmHg).
Hypotension may in part be compensated by a marked elevation in systemic vascular resistance (SVR), mediated by increased release of endogenous vasopressors
such as norepinephrine and angiotensin II. This deadly combination of a low cardiac output and elevated SVR may result in a further reduction in tissue perfusion.
Diagnostic criteria for CS are summarized in Table 9.1.
Other causes of shock (distributive, hypovolemic, and obstructive) must be ruled
out and contributing factors such as hypovolemia, hypoxia, and acidosis must be
corrected in order to determine the etiology of shock.
Echocardiography represents a fundamental tool in the diagnostic workup of CS.
Nonischemic causes of CS can be investigated: pericardial effusions, valvular
abnormalities, and acute overload signs such as in massive pulmonary embolism
and volemic status but also mechanical complications of MI. It also allows quantification of the severity of cardiac involvement and evaluation of systolic and diastolic function and regional abnormalities.
Echocardiography should nowadays be applied as a first-line diagnostic technique together with traditional hemodynamic tools in a so-called echodynamic
approach to CS and to any cause of hemodynamic instability.
The presence of current myocardial ischemia must be assessed and appropriate
investigations must be performed. In this regard, coronary angiography should be
performed in all patients with cardiogenic shock in whom acute myocardial infarction is suspected and who are candidates for revascularization with either percutaneous coronary intervention or coronary artery bypass graft surgery. In addition, all
patients who have undergone reperfusion therapy should be evaluated for failure of
reperfusion.
[email protected]
9.4
109
Management
Prompt restoration of adequate blood flow to the affected myocardium is the key
management measure in all patients with ongoing ischemia.
Systemic thrombolysis, percutaneous coronary revascularization (PCI), and surgical revascularization represent the available alternatives.
Thrombolysis should be restricted to patients who would have otherwise no
chance of timely reperfusion, as in CS its likelihood of success is reduced by both
the low coronary blood flow and the hostile biochemical environment [13].
PCI represents the optimal treatment for patients in whom CS developed early
after myocardial infarction and the coronary anatomy makes it feasible.
Surgery may allow a more complete revascularization, but it is more invasive and
requires longer times to reperfusion. It is better reserved to patients in whom PCI is
impossible for any technical or clinical reason [14].
Of course, the need for a prompt myocardial reperfusion should not de-emphasize
the concurrent necessity to sustain the patients hemodynamics, restore an adequate
tissue perfusion, and reverse metabolic derangements.
Fluid replacement and supplemental oxygen when needed are the basic measures. If respiratory failure is severe, ventilatory support (either noninvasive or invasive) should be provided.
Optimization of myocardial performance is generally sought with the use of catecholamines. Although they are frequently needed to increase tissue perfusion, this can
be seen as a palliative therapy, as no evidence of survival benefit exists with the use
of such drugs, which might on the contrary worsen myocardial dysfunction by increasing myocardial oxygen consumption [15, 16]. The dosage of inotropic agents should
be continuously titrated to the minimum necessary dosage needed to achieve the therapeutic goals, in order to minimize oxygen consumption and arrhythmogenic effects.
An interesting alternative with regard to myocardial oxygen consumption is represented by the class of calcium sensitizers, levosimendan being the only compound
currently available on the market. Since its positive inotropic effect is based on a
reversible increase of the affinity of the myocardial contractile apparatus to calcium
and not on the increased influx of calcium, it does not increase myocardial oxygen
consumption nor has an arrhythmogenic effect. Moreover, both its peripheral vasodilatory and anti-inflammatory effects might also be useful in the setting of CS [1719].
In many patients, fluids and inotropes alone are unable to stabilize hemodynamics. In such cases, a mechanical support device is needed.
The simplest form of mechanical support is represented by intra-aortic balloon
counterpulsation (IABP). The rationale for aortic counterpulsation is particularly
strong in the setting of myocardial ischemia and infarction and in postischemic
acute mitral regurgitation, for its positive effects on coronary perfusion and afterload
reduction.
The benefit of IABP on early mortality in patients with CS has been recently
questioned by the results of the IABP-SHOCK II Trial [4]. These results were quite
surprising, but the trial raised many criticisms, and a change in the current guidelines based on this evidence seems unjustified at the moment.
[email protected]
F. Formica et al.
110
IABP may be unable to adequately support a patient with severe CS, especially
when a large portion of the myocardium (more than 40 % on average) is affected. It
is generally accepted that a cardiac output of at least 2.5 L/min is needed for the
patient to take advantage of counterpulsation.
In such a condition, a full mechanical circulatory support (MCS) must be
considered and, if indicated, implanted as early as possible.
MCS is required to rapidly improve the coronary perfusion, unload both ventricles,
decrease the oxygen myocardial demand, and maintain end-organ perfusion. Currently,
there are several MCS devices available, such as extracorporeal membrane oxygenation
(ECMO), paracorporeal or extracorporeal ventricular assist devices (VADs), percutaneous VADs, and total artificial heart (TAH). Most of them are particularly expensive and
need time and a surgical approach for implantation. ECMO represents an ideal choice
for these patients because of the quick and easy insertion of this device even during fatal
arrhythmia or cardiac arrest. With respect to surgically implanted VADs, ECMO offers
some unique advantages in that it is readily available to provide circulatory support, with
the ability to resolve organ injury in patients who present with cardiac arrest or with
severe hemodynamic instability associated with multiorgan failure.
9.5
Role of ECLS
Whenever ECLS is deemed necessary to support a patient in CS, this should be set
up without delay, as the early introduction of ECMO has been related with better
clinical outcome and hospital survival [20]. MCS can interrupt the inflammatory
cascade initiated by the onset of shock and prevent progression to irreversible endorgan damage and subsequent death; however, a window of opportunity remains
during which rescue is possible.
Each patient should be considered as a candidate for ECMO; however, not all
patients affected by refractory CS meet the criteria for ECMO institution (Table 9.2).
Several considerations must be taken into account in order to determine a
patients eligibility for ECLS. Candidates should be selected only if significant
organ recovery is expected or there is no contraindication to long-term mechanical
support or transplant.
Table 9.2 Contraindications
to ECMO institution
Advanced age (more than 7580 years old)

Disseminated malignancy
Severe degenerative brain disease
Unwitnessed cardiac arrest
Prolonged CPR time (more than 4560 min)
Aortic dissection
Severe peripheral vasculopathy
Irreversible renal disease (dialyzed patients)
Severe aortic regurgitation
Ungraftable coronary arteries
Non-eligibility to heart transplant or VAD
[email protected]
111
Up to 60 % of survivors cannot be weaned and require a ventricular assist device

(VAD) or transplantation [21, 22]. ECMO may therefore provide a bridge to decision;
it is less costly than VADs, can be initiated quickly, and offers biventricular and respiratory support, thereby stabilizing patients while their suitability for a VAD or transplant
is evaluated. Institutions that do not provide this therapy should consider referring
patients to an experienced center once IABP support has been initiated. In these situations, expert retrieval teams from the specialist center should provide transport [23, 24].
The ideal indication for ECMO institution is isolated severe heart failure in the
absence of signs and symptoms of multiorgan failure.
Factors such as age, comorbidities, and neurological, renal, and hepatic status
could preclude ECMO institution. The most common contraindications to ECMO
are based on irreversible multiorgan failure, severity of cerebral damage, and
absence of chances for recovery in patients who are not candidates for heart transplantation or long-term VAD implantation.
ECLS is able to stabilize the majority of patients, prevent organ dysfunction, and
revert metabolic derangements, provided it is implanted in a timely fashion. Its positive effect on mid- to long-term outcome appears reasonable, with reported survival
rates of 2043 % among patients who received ECLS for cardiac arrest, severe
cardiogenic shock, or failure to wean from cardiopulmonary bypass following cardiac surgery. The evidence is however quite weak due to the small numbers of
observational studies and case series and the lack of RCTs.
Our group, as well as Combes and coworkers, demonstrated a 2831 % survival
to discharge in patients supported with ECMO for postcardiotomic or post-AMI
shock refractory to conventional management including IABP [21, 25].
The pooled data from the ELSO (Extracorporeal Life Support Organization)
Registry report an average survival rate of 39 % for adult patients with cardiogenic
shock [26]. These results are consistent with the recent report from Sakamoto and
colleagues. In a population of patients with acute coronary syndromes complicated by
cardiogenic shock or cardiac arrest, they demonstrated a 32.7 % survival to hospital
discharge. The extrapolation of data on the circulatory status at the onset of ECMO
reveals a 41 % survival to discharge in patients with CS, with the circulatory status
being one of the independent predictors of in-hospital mortality at multivariate analysis, together with failed angioplasty and ECLS-related complications. Interestingly,
univariate analysis showed a significant negative impact of the time from collapse to
ECMO on mortality [27]. This was lately confirmed by Kim and colleagues, who also
noted an association between pre-ECMO lactate levels and mortality [28].
Moreover, Bermudez and colleagues recently reported what is common gut feeling: patients presenting for an acutely decompensated chronic heart failure do much
worst on ECMO than patients with acute cardiogenic shock, with 2-year survival
rates of 11 and 48 %, respectively [29].
Similar results have been reported in a number of small studies and case series.
The available evidence and clinical current practice suggest a careful selection of
potential candidates to ECLS and the prompt institution of the extracorporeal support in order to prevent progression of distal organ failure and avoid the progression
to an irreversible degree of multiple organ dysfunction [30].
[email protected]
112
9.6
F. Formica et al.
Patient Care During ECLS
As stated before, extracorporeal support should be commenced as soon as it becomes

necessary, avoiding unnecessary and deleterious delays.
Cannulation techniques are described elsewhere in this book.
Once implanted, ECMO management should follow a standardized protocol,
though individualized on each patients needs.
General issues on anticoagulation, all aspects pertaining to patient care, strategies to enhance rest and recovery of the heart, and weaning from ECLS are presented in specific chapters across the book.
We will here focus on few specific aspects of post-AMI patients.
9.6.1
Interventions to Promote Heart Rest and Recovery
Myocardial revascularization should be performed immediately in all AMI patients

who could not be revascularized before ECMO initiation, as patients in CS are particularly likely to benefit from early revascularization [31].
If an IABP is already in place, it is reasonable to keep it whenever possible;
however, no clear evidence exists on its benefit during peripheral ECMO. It is hence
not recommended at present to suggest the placement of an IAB in all patients.
Nevertheless, the use of counterpulsation has several potential advantages in this
setting: it helps in unloading the left ventricle, augments coronary perfusion pressure, and induces a pulsatile flow that may be beneficial for distal organs. Few
published data exist on this aspect; however, Doll and coworkers found a significantly higher survival rate in postcardiotomic ECMO patients in whom an IAB was
used [32].
Adequate unloading of the left ventricle must be ensured. This is of paramount
importance to prevent LV distention and, in turn, inadequate myocardial rest, elevated filling pressures, pulmonary edema, and respiratory failure that might subsequently compromise weaning from extracorporeal support. Various tools may be
used to obtain this. Increasing the pump flow rate may paradoxically worsen the
situation, as this leads to a further increase in afterload, while it is unable to drain
the bronchial circulation. Low-dose inotropes are often helpful in many patients, as
they promote a certain degree of ejection, which is sufficient to prevent LV distention and blood stagnation. Counterpulsation may also provide a certain degree of
external venting to the LV and be useful in numerous patients. The adequacy of
LV unloading should be accurately evaluated throughout the course of extracorporeal assistance, as a direct venting of the left heart may become necessary. This
might be accomplished with various techniques, which are presented in a specific
chapter. Irrespective of the method used, careful monitoring of the adequacy of
unloading is warranted.
[email protected]
9.6.2
113
Monitoring of Heart and Lung Function
Echocardiography represents a fundamental tool for the daily assessment of the

ECMO patient. Its integration with conventional hemodynamic measurements
allows optimizing and individualizing treatments based on the changing conditions
and needs of the patient. Moreover, during the phase of full extracorporeal support,
when standard hemodynamic parameters are frequently inconclusive, it gives direct
clues on the status of the heart.
Specific aspects of echocardiographic monitoring during ECLS are presented in
the relevant chapter.
Chest x-rays and chest ultrasound are useful in evaluating the lung to detect pulmonary edema and monitor the appearance of parenchymal consolidations. Besides
this, blood gas analysis is determinant in estimating the adequacy of the natural
lung before weaning from ECLS. Blood should ideally be sampled from a right
radial arterial line, as this is the closest sample site to the left ventricle and the less
likely to be influenced by extracorporeal flow. It hence represents the closest
approximation to blood coming from the natural lung.
Myocardial specific enzymes should be monitored at least on a daily basis during
the acute phase to determine the timing of weaning and to detect recurrent ischemia
and infarction. Troponins and myocardium-specific isoforms of creatine kinase
(CK-MB) are commonly used, while myoglobin is almost constantly elevated due
to peripheral muscle release and does not hence represent a reliable index of myocardial ischemia.
Natriuretic peptides are generally extremely elevated upon presentation.
Since their half-life is relatively short (the half-life for BNP being shorter than
that for NT-proBNP), serial measurements may be of value in guiding the
management [33].
9.6.3
Discontinuation of ECLS
In patients presenting with myocardial ischemia, no weaning trial should be

attempted before 48 h of ECLS or until a significant reduction in cardiac enzymes
is observed.
The suitability for a weaning attempt must be judged based on multiple
considerations:
1. Myocardial recovery should be adequate to provide a satisfactory hemodynamic
stability without extracorporeal support and with low doses of inotropes. This
can be evaluated with a staged echodynamically guided reduction in ECMO
blood flow.
2. Lung function must be sufficient to ensure acceptable gas exchange after disconnection from ECMO.
[email protected]
114
F. Formica et al.
3. Complications of ECLS must be accurately investigated and may prompt acceleration in the weaning process.
Weaning from extracorporeal support is attempted in a stepwise fashion as
described in the relevant chapter.
Not all patients will succeed their weaning trials.
In such patients, additional factors must be taken into account:
1. Time from initial insult and extension of the affected myocardium. Is a late recovery
conceivable? In the majority of patients, most of the myocardial recovery occurs in
the first few days after revascularization. However, in some patients, recovery continues for longer periods (weeks or months) [34]. ECMO presents some disadvantages in these patients who need a prolonged assistance: suboptimal LV unloading,
increased afterload, low lung perfusion, hemolysis, immobility of the patient, platelet consumption, and thromboembolic events, to name the most relevant. Patients in
whom a late recovery is likely may benefit from a VAD as a bridge-to-decision.
2. Organ failures (apart from the heart). What is the neurological status? Lung
function? Renal function? GI function? Nutritional status?
3. Presence of sepsis or septic shock.
4. Age and comorbidities.
5. Psychological and psychiatric conditions.
Answers to these questions will determine the suitability for transplantation or
long-term VADs implantation (either as a bridge-to-transplant, bridge-to-decision,
bridge-to-recovery, or as a destination therapy).
The peculiar aspects of VADs and transplant candidates will be dealt with elsewhere in the book.
9.7
Conclusion
Functional recovery of the heart in the setting of AMI largely depends on the precocity and completeness of revascularization.
In patients presenting with refractory CS complicating AMI, a mechanical circulatory support may become necessary to stabilize hemodynamics and prevent irreversible organ dysfunction. In these patients, extracorporeal support must be
initiated without delay. ECMO represents an easy-to-implant device; it can be
applied with a percutaneous approach virtually everywhere by surgeons or intensivists in a matter of minutes and provides biventricular and respiratory support.
Diagnostic and therapeutic procedures are easily performed under stable conditions
during ECLS. Patients can be supported with ECMO for a few days or weeks and
disconnected when myocardial recovery is adequate, or bridged to long-term VADs
or transplantation when indicated.
[email protected]
115
References
1. Nieminen MS, Bohm C, Cowie MR et al (2005) Executive summary of the guidelines on the
diagnosis and treatment of acute heart failure: the Task Force on Acute Heart Failure of the
European Society of Cardiology. Eur Heart J 26:384416
2. Goldberg RJ, Samad NA, Yarzdbski J et al (1999) Temporal trends in cardiogenic shock complicating acute myocardial infarction. N Engl J Med 340:11621168
3. Griffith GC, Wallace WB, Cochran B et al (1954) The treatment of shock associated with
myocardial infarction. Circulation 9:527
4. Thiele H, Zeymer U, Neumann FJ et al (2012) Intraaortic balloon support for myocardial
infarction with cardiogenic shock. N Engl J Med 367:12871296
5. Ellis TC, Lev E, Yazbek NF, Kleiman NS (2006) Therapeutic strategies for cardiogenic shock,
2006. Curr Treat Options Cardiovasc Med 8:7994
6. Page DL, Caulfield JB, Kastor JA et al (1971) Myocardial changes associated with cardiogenic
shock. N Engl J Med 285:133137
7. Alonso DR, Scheidt S, Post M, Killip T (1973) Pathophysiology of cardiogenic shock: quantification of myocardial necrosis, clinical, pathologic and electrocardiographic correlations.
Circulation 48:588596
8. Hochman JS (2003) Cardiogenic shock complicating acute myocardial infarction: expanding
the paradigm. Circulation 1(07):29983002
9. Lim N, Dubois MJ, De Backer D, Vincent JL (2003) Do all nonsurvivors of cardiogenic shock
die with a low cardiac index? Chest 124:18851891
10. Widimsky P, George P, Cervenka V et al (1988) Severe diffuse hypokinesis of the remote
myocardium. The main cause of cardiogenic shock? An echocardiographic study of 75 patients
with extremely large myocardial infarctions. Cor Vasa 30:2734
11. Bolli R, Marban E (1999) Molecular and cellular mechanisms of myocardial stunning. Physiol
Rev 79:609634
12. Lupi-Herrera E, Lasses LA, Cosio-Aranda J et al (2002) Acute right ventricular infarction:
clinical spectrum, results of reperfusion therapy and short-term prognosis. Coron Artery Dis
13:5764
13. Kennedy JW, Gensini GG, Timmis GC et al (1985) Acute myocardial infarction related with
intracoronary streptokinase: a report of the Society for Cardiac Angiography. Am J Cardiol
55:871877
14. Hochman JS, Sleeper LA, White HD et al (2001) Should we emergently revascularize occluded
coronaries for cardiogenic shock. One-year survival following early revascularization for cardiogenic shock. JAMA 285:190192
15. Havel C, Arrich J, Losert H et al (2011) Vasopressors for hypotensive shock. Cochrane
Database Syst Rev. doi:10.1002/14651858.CD003709.pub3
16. Singer M (2007) Catecholamine treatment for shockequally good or bad? Lancet 370:636
17. Lilleberg J, Nieminen MS, Akkila J et al (1998) Effects of a new calcium sensitizer, levosimendan, on haemodynamics, coronary blood flow and myocardial substrate utilization early
after coronary artery bypass grafting. Eur Heart J 19:660668
18. Nieminen MS, Akkila J, Hasenfuss G et al (2000) Hemodynamic and neurohumoral effects of
continuous infusion of levosimendan in patients with congestive heart failure. J Am Coll
Cardiol 36:19031912
19. Parissis JT, Karavidas A, Bistola V et al (2008) Effects of levosimendan on flow-mediated
vasodilation and soluble adhesion molecules in patients with advanced chronic heart failure.
Atherosclerosis 197:278282
[email protected]
116
F. Formica et al.
20. Tayara W, Starling RS, Yamani M et al (2006) Improved survival after acute myocardial
infarction complicated by cardiogenic shock with circulatory support and transplantation:
comparing aggressive intervention with conservative treatment. J Heart Lung Transplant
25:504509
21. Combes A, Leprince P, Luyt CE et al (2008) Outcomes and long-term quality-of-life of
patients supported by extracorporeal membrane oxygenation for refractory cardiogenic shock.
22. Bakhtiary F, Keller H, Dogan S et al (2008) Venoarterial extracorporeal membrane oxygenation for treatment of cardiogenic shock: clinical experiences in 45 adult patients. J Thorac
23. Wagner K, Sangolt GK, Risnes I et al (2008) Transportation of critically ill patients on extracorporeal membrane oxygenation. Perfusion 23:101106
24. Huang SC, Chen YS, Chi NH et al (2006) Out-of-center extracorporeal membrane oxygenation for adult cardiogenic shock patients. Artif Organs 30:2428
25. Formica F, Avalli L, Martino A et al (2008) Extracorporeal membrane oxygenation with a
polymethylpentene oxygenator (Quadrox D). The experience of a single Italian centre in adult
patients with refractory cardiogenic shock. ASAIO J 54:8994
26. Extracorporeal Life Support Organization (ELSO): ECLS registry report, international summary (2009). ELSO, Ann Arbor
27. Sakamoto S, Taniguchi N, Nakajima S, Takahashi A (2012) Extracorporeal life support for
cardiogenic shock or cardiac arrest due to acute coronary syndrome. Ann Thorac Surg
94:17
28. Kim H, Lim SH, Hong J et al (2012) Efficacy of veno-arterial extracorporeal membrane oxygenation in acute myocardial infarction with cardiogenic shock. Resuscitation 83:971975
29. Bermudez CA, Rocha RV, Toyoda Y et al (2011) Extracorporeal membrane oxygenation for
advanced refractory shock in acute and chronic cardiomyopathy. Ann Thorac Surg
92:21252131
30. Cove ME, MacLaren G (2010) Clinical review: mechanical circulatory support for cardiogenic
shock complicating acute myocardial infarction. Crit Care 14:235
31. The Task Force on Myocardial Revascularization of the European Society of Cardiology
(ESC) and the European Association for Cardio-Thoracic Surgery (EACTS) (2010) Guidelines
on myocardial revascularization. Eur Heart J 31:25012555
32. Doll N, Kiaii B, Borger M et al (2004) Five-year results of 219 consecutive patients treated
with extracorporeal membrane oxygenation for refractory postoperative cardiogenic shock.
Ann Thorac Surg 77:151157
33. Bhardwaj A, Januzzi JL Jr (2009) Natriuretic peptide-guided management of acutely destabilized heart failure: rationale and treatment algorithm. Crit Pathw Cardiol 8:146
34. Chalkias A, Xanthos T (2012) Pathophysiology and pathogenesis of post-resuscitation myocardial stunning. Heart Fail Rev 17:117128
[email protected]
ECMO for Refractory Cardiac Arrest
10
Leonello Avalli, Margherita Scanziani,

Elena Maggioni, and Fabio Sangalli
Sudden cardiac arrest (CA) is a complex, life-threatening event requiring a multidisciplinary approach. Many strategies have been proposed over time to achieve
the return of spontaneous circulation (ROSC) and to optimize post-resuscitation
care in order to ultimately improve survival. These include medical, organizational, and technical aspects: mild hypothermia, oxygen control, regionalization
to specialized post-resuscitation care centers, and extracorporeal membrane oxygenation (ECMO). In this setting, ECMO might represent a unique resource for
highly selected patients suffering from CA in which conventional treatment
failed.
L. Avalli (*)
M. Scanziani E. Maggioni
San Gerardo Hospital, University of Milano-Bicocca, Milan, Italy,
F. Sangalli
University of Milano-Bicocca, Via Pergolesi 33, Monza (MB) 20900, Italy

[email protected]
117
118
10.1
L. Avalli et al.
Sudden Cardiac Arrest: a Multidisciplinary

Approach for a Complex Event
Sudden CA is a complex and potentially catastrophic event that was dealt with since
1988 with a multidisciplinary approach that resulted from the famous concept of
life-support chain, coined by Peter Safar [1]. Although this approach improved
the management of CA, its high mortality and morbidity remain a problem to face.
CA is a heterogeneous scenario, which can arise from different etiologies, with different electrical rhythms of presentation, rapidly evolve to ROSC or be refractory,
and occur in hospital or out of hospital. In this regard, efforts to improve CA survival over time were directed to a broad spectrum of issues including medical, technical, and organizational aspects. Despite this, survival of both in-hospital and
out-hospital cardiac arrest (IHCA, OHCA) remains very low. Discouraging results
about IHCA survival were largely reported with a survival rate ranging from about
1022 % [24], confirmed by both the BRESUS study [5] and the National Registry
of Cardiopulmonary Resuscitation from the United States [2] in which 44 % of
14,720 patients suffering IHCA had ROSC, while only 17 % survived to hospital
discharge. As for OHCA, survival rate remains less than 10 %, even if over the past
decades efforts have been made to get more adherence to ILCOR guidelines, with
the implementation of the chain of survival and providing early CPR and defibrillation with automated external defibrillators by lay bystander and first responders [2,
6]. A systematic review by Sasson et al. [6] analyzed 80 studies involving about
143,000 OHCA of presumed cardiac origin over a period of 30 years and reported a
survival rate ranging from 6.7 to 8.4 %, almost unchanged during the three decades.
Interestingly, they showed that in witnessed CA, CPR performed early by expert
medical or paramedical personnel and presenting shockable rhythms positively
impacted survival. Other important factors affecting outcome were early defibrillation [7] and CPR quality [8].
Moreover, it should be underlined that mortality and morbidity after CA are also
affected by the so-called post-cardiac arrest syndrome, characterized by anoxic
brain injury, myocardial dysfunction, and systemic response to ischemia and reperfusion injury [9]. In fact cerebral ischemia may last for some hours after resuscitation [10], and after ROSC an additional injury occurs due to reperfusion that causes
the release of toxic metabolic products.
In this regard, together with efforts aimed at achieving ROSC, over time many
strategies for cardiac and neurological protection during resuscitation care have
been proposed.
Hypothermia was firstly proposed as a neuroprotective treatment after anoxic
brain injury. The mechanisms underlying its beneficial effects on cerebral tissue
have already been described [913], but hypothermia also has protective effects on
myocardial tissue as already shown by experimental studies [14]. A reduction of
infarct size has been described, especially when the myocardium is cooled before or
at the beginning of reperfusion [11, 12].
The use of moderate hypothermia was firstly reported in the late 1950s and early
1960s, but because of the high rate of complications and inconclusive findings, it was
[email protected]
10
119
somehow abandoned until the 1990s, when laboratory studies demonstrated beneficial effects of mild hypothermia in animal models [15], then followed by preliminary
clinical studies [16]. Ten years later, two randomized controlled trials from Australia
and Europe showed a better neurological outcome in patients suffering from CA due
to ventricular fibrillation treated with early mild hypothermia (3234 C) for 12 or
24 h [10, 13]. In the European study, a significantly lower mortality in the hypothermic group compared to the normothermic group was also found [13]. However, some
limitations of these studies warranted further investigations, especially about the possibility of extending therapeutic hypothermia to CA from non-shockable rhythm in
which this therapeutic option is recommended with a low level of evidence (Class
IIb) [17] or in IHCA patients [16]. A recent systematic review and meta-analysis [18]
evaluated the effects of hypothermia in patients after non-shockable rhythm CA
showing that the relative risk of in-hospital mortality was significantly lower in the
therapeutic hypothermia group than in the control group, whereas the beneficial
effects of hypothermia on neurological outcome appeared less evident.
Another important aspect of resuscitation care is represented by the control of
oxygenation during ventilation in patients after ROSC and the subsequent amount
of oxygen to brain and tissues. Which concentration of supplemental oxygen should
be delivered to patients is actually debated: too little oxygen may amplify anoxic
injury; too much oxygen may increase free-radical production and lead to cellular
injury and apoptosis [19, 20]. Despite previous data suggesting a correlation
between hyperoxia and in-hospital mortality in resuscitated patients [19], in Bellomo
et al. [21], this association failed. In this regard, due to the uncertainty about the
detrimental effects of hyperoxia and the certainty of the potential harmful effects of
hypoxia, a revision of the current guidelines targeting an arterial oxygen saturation
between 95 and 98 % in these settings seems actually not justified [20, 21].
Since both prehospital interventions and in-hospital post-resuscitation care affect
survival in OHCA, it has been proposed to direct comatose patients after OHCA to
specialized centers [22]. Better neurological outcomes were described when prehospital ROSC patients received specialized post-resuscitation care [23]. In a recent
retrospective observational study using a nationwide OHCA registry in South
Korea, the benefit of transporting post-ROSC patients toward high-volume centers
compared to low-volume centers was demonstrated [24]. Thus, regionalization to
specialized post-resuscitation care centers of OHCA patients seems to play an
important role in improving survival rate.
Venoarterial extracorporeal membrane oxygenation was introduced as an additional step in the chain of survival for selected RCA patients. Derived from the pioneering applications of heart-lung machines firstly applied in the 1930s by Dr. Gibbon,
advances in technology over time allowed a wider and extended use of this unconventional device, leading to progressively more encouraging results [2528].
In the University Hospital of Caen between 1997 and 2003, 40 patients with
refractory IHCA were treated with extracorporeal life support (ECLS); ECMO was
discontinued in 22 patients due to brain death or multiorgan failure, 18 patients survived to the first 24 h of support, and 8 patients were alive without any sequelae at
18-month follow-up [3]. Chen et al. [27] obtained slightly better results in a 3-year
[email protected]
120
L. Avalli et al.
prospective observational study about the use of ECLS versus conventional CPR in
92 patients suffering IHCA of cardiac origin. Patients were analyzed by a matching
process based on propensity score to equalize potential prognostic factors. Survival
rate was significantly higher in the ECLS matched group than in the conventional
treatment group at discharge, after 30 days and after 1 year. A few years later, Shin
et al. [26] confirmed these results in a retrospective study applying a similar propensity score and reviewing data collected between 2003 and 2009 on 120 IHCA
patients. Analyzing 77 patients suffering refractory CA treated with ECLS, Kagawa
et al. [29] described a weaning rate from ECLS and 30-day survival higher in the
IHCA than in the OHCA group. Similar results were reported from our group [30],
suggesting more favorable outcomes with ECLS in IHCA than in OHCA patients.
All these studies suggest the feasibility and the potential benefits of extracorporeal cardiopulmonary resuscitation in patients with CA refractory to conventional
treatment.
Thus, ECMO is a strategic option before ROSC because it promptly restores
circulation, but it also plays a pivotal role in the post-resuscitation period. In fact, it
allows leaving the heart at rest and can promote the return of spontaneous rhythm
thanks to its capability to get ventricular unloading and ensure myocardial perfusion. Lin et al. [31] compared patients who had return of spontaneous beating
(ROSB) after ECLS with those that had ROSC after conventional CPR: no different
survival rate at hospital discharge, after 30 days, 6 months, and 1 year, was found
between groups. However, the authors emphasized that ROSB was obtained by ventricular unloading and providing extracorporeal support in patients with ECLS.
Moreover, ECMO provides other advantages, such as the possibility of performing advanced radiologic investigations and definitive surgical or percutaneous treatments in refractory CA of unknown origin, even before an ROSC is obtained [25, 29,
32, 33]; in our ECMO population, 36 OHCA patients and 15 IHCA patients underwent emergency coronary angiography; among these, 25 OHCA and 11 IHCA
patients were revascularized percutaneously, while 2 OHCA and 5 IHCA patients
were directed to surgical revascularization. Secondly, ECMO could provide rapid
cooling and controlled rewarming for therapeutic hypothermia; it can also be applied
in medical intoxications [34]. Finally, when cerebral death occurs after CA anoxic
injury, ECMO could provide peripheral perfusion to make patients organ donor [25].
Thus, ECMO could be considered as the next link in the chain of survival in
selected patients suffering from refractory CA, provided each previous step of resuscitation strategy was promptly performed according to the ILCOR recommendations.
At the same time, it could play a pivotal role in the post-resuscitation period too.
10.2
Which Criteria for a Special Unconventional Therapy
Although progressively more encouraging results were described over time from the
use of ECMO in refractory CA, criteria for its positioning are still debated. To date,
studies failed to provide precise indications and contraindications to ECMO in this
setting, while it is fundamental to identify clear criteria to avoid futile treatments.
[email protected]
10
121
Refractory CA is usually defined by the lack of ROSC within a period of 30 min

of CPR [35, 36].
The first issue that must be considered in the decision whether to consider a
patient for ECMO is the no-flow time (i.e., the duration of CA without cardiac
output before CPR). The duration of no-flow time can be known precisely only in
witnessed CAs, and the best candidates to ECMO in this setting are those receiving
immediate CPR by bystanders, since the no-flow time is negligible in these patients.
Indications about the duration of no-flow time are lacking or inhomogeneous in
literature. The French guidelines [36] proposed an algorithm in which no-flow time
is matched with the rhythm of presentation. The authors suggested a no-flow time
below 5 min as a cutoff for ECMO application when patients are found asystolic,
but their algorithm proceeds to the next step (i.e., evaluation of the low-flow time)
even when this time is longer than 5 min if the presenting rhythm is different from
asystole. Le Guen et al. [25] also suggested a no-flow time below 5 min as an inclusion criterion for ECMO in their OHCA population. The duration of no-flow time
may lose its importance when vital signs such as spontaneous movements or spontaneous respirations occur during CPR. Moreover, the role of no-flow time appears
less critical during hypothermia because of its protective effects from ischemia on
cerebral and cardiac tissues [913]. Literature reports longer no-flow times in
OHCA than in IHCA patients [30].
The second important factor to assess is the low-flow time (i.e., the duration of
CA with low cardiac output during CPR). There is no definitive consensus on the
optimal low-flow time limit: the shorter the low-flow time, the better the outcome,
but its duration varies between authors and also combines with the quality of CPR.
Some studies showed a more favorable outcome in patients treated with ECMO
after IHCA compared with those after OHCA [29, 30], with longer delays between
collapse and the start of ECMO in the latter. Although a cutoff of 30 min to start
ECMO has been previously suggested [30], some evidence that ECMO allowed a
longer CPR duration than expected in conventional CPR has been described [3, 27,
29, 30, 37]. Massetti et al. reported an average of 72 min of CPR before the onset of
ECMO in surviving patients [3]. Chen et al. [37] further extended this time having
a probability of survival of about 10 % in patients in which CPR lasted 90 min.
Positive outcomes were recently reported in ECMO patients receiving about
138 min of CPR before ECLS [38] and in a subgroup of intoxicated patients alive at
1-year follow-up, in which CPR lasted up to 180 min [32].
The quality of CPR is also a critical issue for neurological outcome, not only the
duration. Massetti et al. [3] showed that effectiveness rather than the duration of
CPR has to be considered in the decision process. In 2005, Abella et al. [39] analyzed 67 patients undergone CPR after CA and showed that in-hospital CPR was
highly variable and not consistent with current guidelines, even if performed by
well-trained medical personnel. Similar results come from a study aimed at testing
the quality of CPR in 167 patients with OHCA treated by paramedics and nurse
anesthetists, where chest compression was too shallow or withheld for half of
the time of resuscitation [8]; higher chest compression variability was also found in
the prehospital setting as compared to the emergency department [40]. In the
[email protected]
122
L. Avalli et al.
prehospital phase, the introduction of automated chest compression devices may

help to overcome the difficulties correlated to CPR during transportation. Despite
initial discouraging results, promising data have been reported recently, even if the
beneficial impact on survival has not been demonstrated yet. Duchateau et al. [41]
showed that in patients with OHCA treated with an automated load-distributing
chest compression device, diastolic and mean blood pressure were increased. In a
prospective cohort study, 1,011 patients suffering prolonged CA in emergency
department were treated with load distributing band (LBD) CPR or manual CPR.
A higher tendency toward improved survival and neurological outcome at discharge
was found in the LBD-CPR group compared to manual CPR group [42].
Finally, an index that indirectly reflects the cardiac output obtained during CPR
is the end tidal carbon dioxide (ETCO2): an ETCO2 below 10 mmHg after 20 min of
ACLS maneuvers seems to accurately predict death in CA with electrical activity
without pulse [38]. Thus, patients presenting with an ETCO2 below 10 mmHg should
not be supported with ECLS.
As for exclusion criteria, severe comorbidities precluding ICU admission should
be considered as contraindications to ECMO positioning: terminal illness, acute
aortic dissection, preexisting irreversible brain damage, hepatic failure, or late stage
of respiratory distress syndrome [3, 26, 2830]. Age alone should not constitute a
limitation since it does not appear a sufficient reason to limit ICU admission,
although some authors excluded from treatment patients older than 75 years [3, 28].
Finally, the decision to discontinue ECLS support is critical and based on the
patients clinical evolution: severe neurological impairment up to brain death [3],
irreversible multiorgan failure, or intractable sepsis [32] should warrant ECMO discontinuation. However, larger studies are needed to identify negative predictive factors during the clinical course of ECMO.
10.3
Monzas Flow Chart
Our group in Monza proposed a simple flow chart that could help the attending
physician in deciding when and when not to start an ECMO support in case of
refractory CA (Fig. 10.1).
We must emphasize that this flow chart only represents a recommendation, since
it is the attending physicians responsibility to decide whether to initiate ECLS,
even in the presence of prolonged no- or low-flow time, based on clinical or anamnestic factors, e.g., the appearance of vital signs, or a good CPR performed by welltrained personnel or with an automated chest compression device.
The first point to assess in our flow chart is the presence of comorbidities precluding ECMO positioning as reported below (Fig. 10.1).
We decided to use a no-flow time of 6 min and a low-flow time of 45 min on the
basis of the literature and from our preliminary results [30]. To reduce the no-flow
[email protected]
10
123
Witnessed refractory CA (CPR>15 min)
Comortidities?*
Yes
No indication
No
Assesment of no-flow
No-flow > 6 min
Yes
* Comorbidities:
Terminal malignancy
Aortic dissection
Severe peripheral arteriopathy
Severe cardiac failure without transplant indication
Severe aortic valve failure.
No indication
No
Assement of low-flow
Low-flow > 45 min
Yes
No indication
No
Assesment of End-Tidal CO2
End-Tidal CO2 < 10 mmHg

after 20 min CPR
Yes
No indication
No
Indication for ECMO support
Fig. 10.1 Monzas flow chart for ECMO support in refractory CA
time, we promoted telephone-guided CPR performed by bystanders. To reduce the

low-flow time, our ECMO team was alerted for all CA patients when ongoing CPR
lasted more than 15 min. In OHCA, the ambulance crew was prompted to leave the
scene and begin transport after no more than 15 min of ACLS maneuvers in the
absence of an ROSC, to reduce the no-flow time to a minimum.
Finally, an ETCO2 below 10 mmHg measured after 20 min of CPR contraindicated ECMO.
10.4
Conclusion
Sudden CA is a complex event with high mortality rate. We strongly believe that
optimal state-of-the-art conventional treatment should constitute the basis for every
CA patient. ECLS represents a valuable additional therapeutic option both in
achieving ROSC and in post-resuscitation care in highly selected CA patients not
responding to the conventional approach.
[email protected]
124
L. Avalli et al.
References
1. Safar P, Bircher N (1988) History and phases and stages of cardiopulmonary cerebral resuscitation. In: Safar P, Bircher N (eds) Cardiopulmonary cerebral resuscitation, 3rd edn. WB
Saunders Co, Philadelphia
2. Peberdy MA, Kaye W, Ornato JP et al for the NRCPR Investigators (2003) Cardiopulmonary
resuscitation of adults in the hospital: a report of 14720 cardiac arrests from the National
Registry of Cardiopulmonary Resuscitation. Resuscitation 58:297308
3. Massetti M, Tasle M, Le Page O et al (2005) Back from irreversibility: extracorporeal life support for prolonged cadiac arrest. Ann Thorac Surg 79:178184
4. Girotra S, Nallamothu BK, Spertus JA et al (2012) Trends in survival after in-hospital cardiac
arrest. N Engl J Med 367:19121920
5. Tunstall-Pedoe H, Bailey L, Chamberlain DA et al (1992) Survey of 3765 cardiopulmonary
resuscitation in British hospitals (the BREUS study): methods and overall results. BMJ
304:13471351
6. Sasson C, Roger MAM, Dahl J et al (2010) Predictors of survival from out-of-hospital cardiac
arrest. A systematic review and meta-analysis. Circ Cardiovasc Qual Outcomes 3:6381
7. Simpson PM, Goodger MS, Bendall JC et al (2010) Delayed versus immediate defibrillation
for out-of-hospital cardiac arrest due to ventricular fibrillation: a systematic review and metaanalysis of randomised controlled trials. Resuscitation 81:925931
8. Wik L, Kramer-Johansen J, Myklebust H et al (2005) Quality of cardiopulmonary resuscitation during out-of-hospital cardiac arrest. JAMA 293:299304
9. Stub D, Bernard S, Duffy S, Kaye DM (2011) Post cardiac arrest syndrome. A review of therapeutic strategies. Circulation 123:14281435
10. Bernard SA, Gray TW, Buist MD, Jones BM, Silvester W, Gutteridge G, Smith K (2002)
Treatment of comatose survivors of out-of-hospital cardiac arrest with induced hypothermia.
N Engl J Med 346:557563
11. Delhaye C, Mahmoudi M, Waksman R (2012) Hypothermia therapy. J Am Coll Cardiol
59:197210
12. Scirica BM (2013) Therapeutic hypothermia after cardiac arrest. Circulation 127:244250
13. The Hypothermia after cardiac arrest study group (2002) Mild therapeutic hypothermia to
improve the neurologic outcome after cardiac arrest. N Engl J Med 346:549556
14. Lee JH, Suh GJ, Kwon WY et al (2012) Protective effects of therapeutic hypothermia in postresuscitation myocardium. Resuscitation 83:633639
15. Kuboyama K, Safar P, Radovsky A et al (1993) Delay in cooling negates the beneficial effect
of mild resuscitative cerebral hypothermia after cardiac arrest in dogs: a prospective, randomized study. Crit Care Med 21:13481358
16. Bernard SA (2009) Hypothermia after cardiac arrest: expanding the therapeutic scope. Crit
Care Med 37(Suppl):S227S233
17. Nagao K (2012) Therapeutic hypothermia following resuscitation. Curr Opin Crit Care
18:139145
18. Kim Y-M, Yim H-W, Jeong A-H et al (2012) Does therapeutic hypothermia benefit adult cardiac arrest patients presenting with non-shockable initial rhythms?:A systematic review and
meta-analysis of randomized and non-randomized studies. Resuscitation 83:188196
19. Kilgannon JH, Jones AE, Shapiro NI (2010) Association between arterial hyperoxia following
resuscitation from cardiac arrest and in-hospital mortality. JAMA 303(21):21652171
20. Hoedemaekers CW, van der Hoeven JG (2011) Hyperoxia after cardiac arrest may not increase
ischemia-reperfusion injury. Crit Care 15:166
21. Bellomo R, Bailey M, Eastwood GM et al (2012) Arterial hyperoxia and in-hospital mortality
after resuscitation from cardiac arrest. Crit Care 15:R90
22. Bentley JB, Kern KB (2009) Regionalization of postcardiac arrest care. Curr Opin Crit Care
15:221227
23. Sunde K, Pytte M, Jacobsen D (2007) Implementation of a standardised treatment protocol for
post resuscitation care after out-of-hospital cardiac arrest. Resuscitation 73:2939
[email protected]
10
125
24. Cha WC, Lee SC, Shin SD et al (2012) Regionalisation of out-of-hospital cardiac arrest care
for patients without prehospital return of spontaneous circulation. Resuscitation 83:3381342
25. Le Guen M, Nicolas-Robin A, Carreira S et al (2011) Extracorporeal life support following
out-of-hospital refractory cardiac arrest. Crit Care 15:R29
26. Shin TG, Choi J-H, Jo IJ et al (2011) Extracorporeal cardiopulmonary resuscitation in patients
with inhospital cardiac arrest: a comparison with conventional cardiopulmonary resuscitation.
Crit Care Med 39:17
27. Chen Y-S, Chao A, Yu H-Y et al (2003) Analysis and results of prolonged resuscitation in
cardiac arrest patients rescued by extracorporeal membrane oxygenation. J Am Coll Cardiol
41:197203
28. Chen Y-S, Lin J-W, Yu H-Y et al (2008) Cardiopulmonary resuscitation with assisted extracorporeal life-support versus conventional cardiopulmonary resuscitation in adults with inhospital cardiac arrest: an observational study and propensity analysis. Lancet 372:554561
29. Kagawa E, Inoue I, Kawagoe T et al (2010) Assessment of outcomes and differences between
in- and out-of-hospital cardiac arrest treated with cardiopulmonary resuscitation with extracorporeal life support. Resuscitation 81:968973
30. Avalli L, Maggioni E, Formica F et al (2012) Favourable survival of in-hospital compared to
out-of-hospital refractory cardiac arrest patients treated with extracorporeal membrane oxygenation: an Italian tertiary care centre experience. Resuscitation 83:579583
31. Lin J-W, Wang M-J, Yu H-Y et al (2010) Comparing the survival between extracorporeal rescue and conventional resuscitation in adult in-hospital cardiac arrest: propensity analysis of
three-years data. Resuscitation 81:796803
32. Mgarbane B, Leprince P, Deye N et al (2007) Emergency feasibility in medical intensive care
unit of extracorporeal life support for refractory cardiac arrest. Intensive Care Med
33:758764
33. Kjaergaard B, Frost A, Rasmussen BS et al (2011) Extracorporeal life support makes advance
radiologic examinations and cardiac interventions possible in patients with cardiac arrest.
Resuscitation 82:623626
34. Daubin C, Lehoux P, Ivascau C et al (2009) Extracorporeal life support in severe drug intoxication: a retrospective cohort study of seventeen cases. Crit Care 13:R138
35. ECC Committee, Subcommittees and Task Forces of the American Heart Association (2005)
American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 13:IV1IV203
36. Riou B et al (2009) Guidelines for indications for the use of extracorporeal life support in
refractory cardiac arrest. Ann Fr Anesth Ranim 28:187190
37. Chen Y-S, Yu H-Y, Huang S-C et al (2008) Extracorporeal membrane oxygenation support can
extend the duration of cardiopulmonary resuscitation. Crit Care Med 36:25292535
38. Mgarbane B, Deye N, Aout M (2011) Usefulness of routine laboratory parameters in the decision to treat refractory cardiac arrest with extracorporeal life support. Resuscitation
82:11541161
39. Abella BS, Alvarado JP, Myklebust H, Edelson DP, Barry A, OHearn N, Vanden Hoek TL,
Becker LB (2005) Quality of cardiopulmonary resuscitation during in-hospital cardiac arrest.
JAMA 293:305310
40. Roosa JR, Vadeboncoeur TF, Dommer PB et al (2012) CPR variability during ground ambulance transport of patients in cardiac arrest. Resuscitation. doi:10.1016/j.
resuscitation.2012.07.042
41. Duchateau FX, Gueye P, Curac S et al (2010) Effect of the AutoPulseTM automated band chest
compression device on hemodynamics in out-of-hospital cardiac arrest resuscitation. Intensive
Care Med 36:12561260
42. Hock Ong ME, Fook-Chong S, Annathurai A et al (2012) Improved neurologically intact
survival with the use of an automated, load-distributing band chest compression device for
cardiac arrest presenting to the emergency department. Crit Care 16:R144
[email protected]
ECMO for Postcardiotomic Shock
11
Massimo Baiocchi, Fabio Caramelli, and Guido Frascaroli
11.1
Definition of Postcardiotomic Shock
The definition of postcardiotomic circulatory shock (PCCS) is still not unique in

literature, but probably the most accepted reflects patients who have inadequate
cardiac performance after cardiac surgery despite inotropic and intra-aortic balloon
pump (IABP) support. This uncommon but serious complication has been reported
to occur in approximately 0.26 % of adult patients undergoing cardiac surgical
procedure [1, 2]. As defined by Rao et al. [3], PCCS includes not only the patients
who cannot be weaned from cardiopulmonary bypass in the operating room but also
those who show low cardiac output syndrome (LCOS) in the early postoperative
period. Thus, PCCS may be identified as a type of LCOS after cardiac surgery causing inadequate end-organ perfusion despite maximal medical support.
11.2
Indications and Timing for Mechanical Cardiac Support
Regardless of the definition, PCCS is a clinical condition associated with an

extremely poor prognosis, with a mortality rate greater than 70 %. It increases
over 85 % when three of high-dose inotropes are needed to separate from CPB
[4]. Therefore, in the absence of clinical improvement despite maximal inotropic
support, the benefit-risk ratio shifts in favor of mechanical cardiac supports
(MCS).
The MCS is intended to bridge the patient to recover or, if it is not possible, for
transplantation or evaluation of alternative strategies.
M. Baiocchi (*) F. Caramelli G. Frascaroli

Anesthesia and Intensive Care Unit of Cardiothoracic Department,
Policlinico S. Orsola-Malpighi, Via Massarenti, 9, Bologna 40130, Italy
e-mail: [email protected]; [email protected];
[email protected]
[email protected]
127
128
M. Baiocchi et al.
Different types of MCS systems are currently available to treat PCCS [59].
They include simple centrifugal pumps implanted as left ventricular assist systems
[10, 11] and complex and expensive ventricular assist devices (VAD) [5].
The results of these different experiences with different devices are not comparable coming usually from limited, monocentric, and retrospectively collected data
[6, 8, 11]. Nevertheless, the uncertainty of recovery and often the lack of alternatives to it have led the clinicians to opt for bridge to decision solutions at low cost.
The use of extracorporeal membrane oxygenation (ECMO), namely, a circuit with
an oxygenator in addition to a centrifugal pump, has gradually increased over time
for its simplicity, ductility, reliability, and limited costs [2, 12, 13].
Only venous-arterial VA ECMO can support both lung and heart function and
allows further evaluations at low cost when the underlying reason of postcardiotomic shock has not yet been fully clarified.
Several published experiences have underlined the capital importance of early
implants to avoid suboptimal perfusion, leading to MOSF, to increase the weaning
rate and to improve the outcome [3, 4, 11].
The ECMO support is contraindicated when the patients life expectancy is
deemed poor and unlikely to be improved by MCS (i.e., terminal illnesses, irreversible neurological injury, advanced MOF). Technical contraindications include aortic
dissection and severe aortic regurgitation [14].
11.3
Options and Modalities of Support
ECMO for PCCS opens several possibilities of action, being not a device, but a
system. Different cannulation sites allowing diverse configurations (central, peripheral, or mixed) can be employed for ECMO institution. Having distinctive physiology, characteristics, and issues, these should be selected on the basis of the specific
patients needs and features.
The central cannulation (i.e., inlet in the left and right atria and outlet in the
aorta) is perhaps the easiest to adopt in the operating theater and probably the most
physiological. It allows real biventricular unloading, lower shear stress (shorter and
bigger cannulae), higher flow rates, and a simple upgrade to mediumlong-term
ventricular assist devices in comparison to the peripheral one. The main concerns
are the risk of infection, hemorrhage, and the need of sternotomy to remove it.
The peripheral cannulation (i.e., inlet and outlet in large peripheral vessels) is
simpler, is associated with lower bleeding risks, and can be performed quickly with
percutaneous technique in emergency condition. Lastly, it does not require resternotomy. However, it supports only the right ventricle and it may be associated
with left ventricular distension, leading to increased myocardial VO2, wall tension,
and eventually myocardial ischemia. The latter can adversely affect myocardial
recovery and result in pulmonary edema. Several options have been suggested to
solve this problem due to the countercurrent flow of the pump.
The first option is the inotropic support, but it increases the myocardial oxygen
demand, interferes with the recovery, and increases the risk of malignant arrhythmias.
[email protected]
11
129
The use of IABP decreases the left ventricular afterload and increases the coronary perfusion pressure during diastolic time. This appears to be extremely relevant
in a patient with peripheral cannulation, in whom an increased left intraventricular
pressure can occur [15].
However, only venting the left heart may decrease the ventricular pressure and
really unload the ventricle. To get this goal without sternotomy, several alternatives
have been proposed. Some authors privilege a transseptal left atrial cannulation [16,
17]. Others create an atrial septal defect by a percutaneous blade or by balloon septostomy [18]. Others prefer a transaortic vent through the aortic valve [19] or associate ECMO to Impella [20]; others use a large pulmonary catheter as a vent [21].
Other complications of peripheral cannulation include the so-called harlequin
syndrome, the leg ischemia, and the formation of thrombi in the ascending aorta
during total extracorporeal support [22].
The harlequin syndrome comes from the competition between the output of the
recovering heart and the ECMO flow, when the pulmonary function is impaired.
The upper part of the body, depending on the native lung, will be hypoxic and cyanotic (blue head), while the lower part of the body will be well oxygenated (red
leg). Therefore, certain key organs, such as the brain and heart, may be compromised for local differences of flow distribution.
The acute leg ischemia may be prevented by inserting a catheter into the femoral
artery, just distally to the ECMO cannula, and connecting it to the ECMO outlet to
provide the distal perfusion.
The use of near-infrared spectroscopy (NIRS) during ECMO support has been
recently advocated [23]. It allows early detection of regional reduction of perfusion
in the development of a compartmental syndrome or neurological complications
[23]. Stressing that in this patient the risk of neurological injury is high (750 % of
the patients [24, 25]) and the difficulty to determine the neurological status of an
intubated, and often heavily sedated (if not paralyzed), patient.
Thrombosis of the ascending aorta is another rare complication of venous-arterial
ECMO via femoral artery. It is due to minimal left ventricular ejection and stagnant
flow in the aortic root.
The direct trans-apical cannulation of the left ventricle is another type of cannulation that overcomes, through a left thoracotomy, the problem of the left ventricular distension [26]. This modality, often used for VAD implantation, ensures
the best ventricular unloading. It avoids the sternotomy, usually considered more
invasive and a source of greater surgical complexity in case of transplantation.
Moreover, it allows, when the right heart recovers, to convert a partial peripheral
ECMO to a midterm LVAS, by simply removing the venous femoral cannulation
and the oxygenator [26]. It is not more traumatic than the atrial one and may be
removed without implications for the ventricular function.
The axillary artery can replace the femoral artery as the site of arterial cannulation; usually a Gore-Tex graft is end-to-side anastomosed to the axillary artery, connected to the arterial cannula, and tunneled subcutaneously to prevent infection
[27]. This cannulation reduces significantly the afterload and avoids harlequin
syndrome and thrombosis of the ascending aorta.
[email protected]
130
11.4
M. Baiocchi et al.
Management
Several preoperative risk factors of LCOS have been identified both in coronary and
in valve surgery [3, 28]. Risk stratification is useful to select patients and to provide
for assessment eligibility to VAD or transplantation, as well as to plan the surgical
procedures (preoperative vascular assessment, right radial artery placement,
Scarpas triangle, and the subclavian artery free from surgical drape, etc.).
A pulmonary-artery catheter should be positioned at the induction of the anesthesia, because its positioning after central cannulation (but often after peripheral)
may be problematic, even under fluoroscopic guidance. It may help during the
weaning of CPB and the management of ECMO. At some centers the left atrial
pressure is monitored with a catheter that may also be used to evaluate the pulmonary gas exchange during the weaning phase.
When the inability to wean the patient from CPB becomes evident, a MCS system must be chosen without wasting time and struggling with the hemodynamic.
If only the left heart is involved, several devices are to be considered (Impella,
CentriMag, Abiomed, TandemHeart, LVAD, etc.), but the discussion is beyond the
scope of this chapter.
If the failure involves only the right ventricle, a centrifugal pump with inlet in the
right atrium and outlet in the pulmonary artery may be used, but even if the right
atrium is drained, the increase of the afterload induced by the pump can seriously
affect a ventricle, physiologically accustomed to working against low pressure.
Therefore, the development of tricuspid regurgitation, as adaptive response, is common. In addition, the management may be very difficult if a left diastolic failure is
present.
In comparison the VA ECMO provides a complete unloading even with peripheral cannulation, but, shunting the pulmonary circulation; it needs the addition of an
oxygenator to the circuit.
However, a biventricular failure is often present, often associated with a severe
gas exchange impairment. In these cases, the use of VA ECMO is valuable. When
in doubt, supporting the right heart may be the right choice, since a right ventricular
failure can be swift and fatal.
In this case, since the sternum is already open, surgeons usually prefer the central
cannulation for the VA ECMO, but there are conflicting opinions, as previously
described (need of redo sternotomy and bleeding).
At our center we usually choose the central cannulation, from the left and right
atria to the aorta, because the better unloading of the left ventricle ensures a higher
probability of weaning. The peripheral cannulation is reserved to a patient with
isolated right ventricular failure, after heart transplantation or pulmonary endarterectomy. The peripheral one is preferred also in acute heart failure that occurred in
the ICU; it allows solving the dramatic situation, permitting to save the patient and,
afterward, if necessary, easy transfer to the operating room to convert it to the central one.
Echocardiography plays a key role in the decision-making and in the whole
ECMO management: firstly, in evaluating the right and left ventricular function [29]
[email protected]
11
131
and then in disclosing contraindications to implant, such as aortic valve insufficiency, or abnormalities can affect the function or position of the cannula (PFO,
septal aneurism, tricuspid pathology, etc.). It is useful for detecting complications
such as cannula malposition, displacement, or obstruction; pericardial effusion; cardiac tamponade, thrombosis; or aortic distortion.
Another challenge in the OR is understanding if the right ventricle is able to
tolerate LVAS, or biventricular support is needed, because the pre- and afterload can
change easily and quickly. Several echocardiographic parameters predictive of an
RV dysfunction after VAD implants have been described, but they have been validated only in elective conditions [30, 31].
The echocardiography can also guide the setting of the mechanical ventilation,
minimizing its effect on the right cardiac function and its influence, as confounding
factor, on the comprehensive assessment [32] or on weaning attempts.
The balance between the cardiac output of the native heart and the outflow of the
ECMO is not only the key element of hemodynamic management but also the first
step to prevent the left ventricle distension with peripheral support. The goal is
maintaining ventricular ejection and evident pulsatility on arterial pressure monitoring, to avoid the complications previously described, even at the cost of a significant
inotropic support.
Nevertheless, the sum of the ECMO flow and the cardiac output must keep the
peripheral perfusion normal and sustain the metabolic needs of the different
parenchymas.
The anticoagulation should be increased and the afterload decreased, reducing
pump flows and using inodilators judiciously, even though maintaining the left ventricular ejection may be impossible if the ventricular function is severely depressed.
In that case, if ventricular distension develops, the left ventricle must be drained.
Another important goal is to maintain an effective perfusion pressure.
Vasoconstrictor drugs are often needed, above all, in case of long-lasting CPB or
late implant. Often the autoregulatory mechanism of the kidney perfusion is lost,
and the development of acute kidney injury is frequent. It is to emphasize that AKI
is an independent risk factor of poor prognosis and can prevent the entry into a
transplant/VAD program [33, 34].
The onset of sepsis is a further condition that not only worsens the prognosis but
contraindicates the transplant. Only a diligent management of cannulae and devices,
a strict clinical and microbiologic monitoring, and the use of antibiotics in an early,
appropriate, and limited time span can reduce the probability of infection in such
high-risk patients.
At our institution, usually, the patient is extubated as soon as possible to prevent
pulmonary complications. The weaning from ventilator is assisted by ECMO, but
all contraindications to extubation, especially the neurological ones, must be absent.
The patient should be able to expectorate, to eat, and to collaborate with nursing.
The central cannulation allows good comfort for the patient and fair mobility.
An acceptable respiratory exchange should be reached even in the centrally cannulated patient, because the coronary flow is preferably supported by blood
oxygenated through the natural way [35].
[email protected]
132
M. Baiocchi et al.
Contrary to the VV ECMO, the respiratory exchange after the weaning from the
assistance is not easy to predict with a VA ECMO, and then its removal can compromise the residual respiratory function.
Moreover, weaning the patient from ECMO and then keeping him/her sedated,
intubated, ventilated, and on high inotropic support has little meaning, and it could
go against the philosophy of the system and the weaning. As an intermediate step,
medium inotropic support and noninvasive ventilation can be used.
Thus, early extubation, if viable, simplifies the medical and nursing care,
decreases the risk of infection and allows to place the patient in conditions that help
to understand who is really weanable or who should be quickly managed with different strategies (OHTx or VAD).
11.4.1 Anticoagulation
The principal causes of ECMO mortality and morbidity are bleeding and thrombosis [36]. Taking the utmost care is required for hemostasis not only of the surgical
field but also of the cannulation site.
A bleeding tendency may persist for several hours, usually until the day after,
meaning that the hemostatic status should be frequently assessed. The heparininfusion at low dose, essential also with the heparin-coated circuits, should be initiated, as soon as the coagulation system starts to recover, to avoid platelet consumption
and dangerous thrombocytopenia.
Viscoelastic tests (TEG or ROTEM) have the advantage to give information not
only about the initiation of clotting but also about the strength and dissolution of the
clot [36]. These point-of-care tests may monitor this change: the heparin infusion
should be managed comparing TEG to TEG with heparinase and looking at the PTT
ratio, because the ACT, even if widely used, does not correlate with heparin levels
[37]. The goal is PTT about 1.52.5 normal and an r time greater than 60, but it
depends on the ongoing coagulation profile, the platelet number/function, and TEG
signs of hypercoagulability.
The use of direct thrombin inhibitors and the like has been advocated, but this is
accepted worldwide only as a therapy of HIT type 2 for costs, pharmacokinetic/
dynamic reasons, and absence of antagonists [38].
11.4.2 Weaning
Several protocols have been proposed for the withdrawal of ECMO, but every
patient is a particular case with its history, original pathology, and hemodynamic,
and ventilator status. Nevertheless, the ECMO flow must be reduced gradually,
looking carefully at the symptoms and signs of unsuccessful weaning. Singular care
must be taken to obtain a sufficient heparinization at low ECMO flow.
Usually the echocardiography and the Swan-Ganz catheter monitor the cardiocirculatory response, but a comprehensive evaluation of the patient is capital. A
[email protected]
11
133
small decrease of urinary output; a trivial increase of central temperature with pale,
cold skin; a little change of the respiratory frequency; or a slight alteration of the
neuropsychological status may be the first signs of an impossible weaning.
Looking at the trend of lactates, diastolic pulmonary arterial pressure, and obviously SvO2 is useful to predict how the attempt is proceeding. Some echocardiographic parameters at ECMO flow of 1 L/min may predict a successful weaning, as
the aortic time-velocity integral 10 cm, the ejection fraction >20 %, and lateral
mitral annulus peak systolic velocity 6 cm/s [39].
Another issue that can prevent the weaning from ECMO is the respiratory gas
exchange. During the ECMO support, different lung injuries may determine a
hypoxic state that hampers the weaning. In that case many authors advise the conversion to a venous-venous ECMO if the heart function is satisfactory.
11.4.3 Outcome
Several risk factors have been reported as independently related to poor outcome of
PCCS: prior cardiac arrest followed by severe cardiogenic shock [6], age, base deficit, and emergency [8, 40]. As outlined by Rao, all are probably linked to a delay in
implant [41].
Significant predictors for death after VA ECMO include previous cardiac surgery,
older age, thoracic aortic surgery, and nonuse of IABP [1]. The advanced age alone is
not an absolute contraindication [42], but Rastan et al. [2] confirms that age older than
70 years, obesity, and diabetes are independent risk factors for in-hospital mortality.
The underlying cardiac disease and the related surgery had a significant effect
on hospital survival: CABG had better prognosis (in-hospital survival of 44 %)
than mitral valve surgery, but the worst was for aortic arch and pericardiectomy,
probably due to bleeding complications. The prognosis for patients with ECMO
after type A dissection, constrictive pericarditis, or double valve disease is
extremely poor. The pre-ECMO lactate values are strongly associated with
increased hospital mortality [2].
Predictors for in-hospital mortality during ECMO are acute liver and renal failure
[43] and persistent high lactate values, despite a whole mechanical support. Acute
myocardial ischemia is an additional adverse prognostic factor, with a close correlation between a high release of creatine kinase MB during ECMO and in-hospital
mortality [44]. ECMO is also associated with various and important morbidities.
Severe hemorrhage is the more frequent and challenging complication, leading
to reoperation and massive transfusion. The incidence of cerebrovascular events
ranges from 17.4 % [2] to 33 % [1] and is a serious complication that may preclude
further treatment. Limb ischemia on the side of the femoral artery cannulation is a
frequent complication (19.9 %) [1, 43].
The successful weaning rate of ECMO for PCCS ranges from 31 to 60 %, but the
in-hospital mortality rate is 5984 % [2, 13]. This high gap is probably related to the
lack of alternatives to weaning in patients who cannot undergo VAD or OHTx and
to the complications of treatment that increase exponentially with time.
[email protected]
134
M. Baiocchi et al.
The ECMO technology has been greatly improved over the last decade, but
these changes do not seem to affect major outcomes. A retrospective analysis,
comparing three little groups, did not find any differences in renal failure, strokes,
or mortality, but showed a statistically in significant trend towards reexploration
for new systems [45]. Other authors reported a lesser rate of complications, i.e.,
hemolysis, reduced thrombi formation, and greater interval between oxygenator
changes, as shown by Yu et al. [46]. Therefore, technological advancements seem
to facilitate patient care and bleeding control more than to reducing main complications or mortality, the latter resulting mostly from comorbidities, initial cardiac
injury, and suitability for VAD or OHTx. No less important is the delay in instituting mechanical support.
11.5
Summary
Several devices have been used to support the vital organs in case of postcardiotomic shock. The supremacy of a system has never been proved, but the VA ECMO
is being increasingly used for its simplicity, ductility, and affordability. The PCCS
has still a poor prognosis, but early ECMO insertion minimizes the complications
of a prolonged CPB and a high inotropic support and seems to be able to increase
the survival rate. Several different configurations (central or peripheral cannulation;
different cannulae, tubes, and centrifugal pumps) are possible, making the ECMO
not simply a device, but a strategy, customizable on the patients anatomy and needs.
Nevertheless, the purpose of insertion is always the same, to earn time: time for
recovery, time for transplant or VAD, or time to decide. The outcome depends
mainly on the possibilities that may open.
References
1. Smedira NG et al (2001) Clinical experience with 202 adults receiving extracorporeal membrane oxygenation for cardiac failure: survival at five years. J Thorac Cardiovasc Surg
122:99102
2. Rastan AJ, Dege A, Mohr M et al (2010) Early and late outcomes of 517 consecutive adult
patients treated with extracorporeal membrane oxygenation for refractory postcardiotomy cardiogenic shock. J Thorac Cardiovasc Surg 139:302311
3. Rao V, Ivanov J, Weisel RD, Ikonomidis JS, Christakis GT, David TE (1996) Predictors of low
cardiac output syndrome after coronary artery bypass. J Thorac Cardiovasc Surg
112(1):3851
4. Samuels LE, Kaufman MS et al (1999) Pharmacologic criteria for ventricular assist device
insertion following postcardiotomy shock: experience with the Abiomed BVS system. J Card
Surg 14:288293
5. Pae WE Jr, Miller CA, Matthews Y, Pierce WS (1992) Ventricular assist devices for postcardiotomy cardiogenic shock. A combined registry experience. J Thorac Cardiovasc Surg
104(3):541552; discussion 5253
6. Guyton RA, Schonberger JP, Everts PA, Jett GK, Gray LA Jr, Gielchinsky I, Raess DH,
Vlahakes GJ, Woolley SR, Gangahar DM (1993) Postcardiotomy shock: clinical evaluation of
the BVS 5000 biventricular support system. Ann Thorac Surg 156:346356
[email protected]
11
135
7. Jurmann MJ, Siniawski H, Erb M, Drews T, Hetzer R (2004) Initial experience with miniature
axial flow ventricular assist devices for postcardiotomy heart failure. Ann Thorac Surg
77:16421647
8. Hernandez AF, Grab JD, Gammie JS, OBrien SM, Hammill BG, Rogers JG, Camacho MT,
Dullum MK, Ferguson TB, Peterson ED (2007) A decade of short-term outcomes in post cardiac surgery ventricular assist device implantation: data from the Society of Thoracic Surgeons
National Cardiac Database. Circulation 116(6):606612
9. Griffith BP, Anderson MB, Samuels LE, Pae WE Jr, Naka Y, Frazier OH (2013) The RECOVER
I: a multicenter prospective study of Impella 5.0/LD for postcardiotomy circulatory support.
J Thorac Cardiovasc Surg 145(2):548554
10. Curtis JJ, McKenney-Knox CA, Wagner-Mann CC (2002) Postcardiotomy centrifugal assist:
a single surgeons experience. Artif Organs 26:944947
11. Akay MH, Gregoric ID, Radovancevic R, Cohn WE, Frazier OH (2011) Timely use of a
CentriMag heart assist device improves survival in postcardiotomy cardiogenic shock. J Card
Surg 26(5):548552
12. Magovern GJ Jr, Magovern JA, Benckart DH, Lazzara RR, Sakert T, Maher TD Jr, Clark RE
(1994) Extracorporeal membrane oxygenation: preliminary results in patients with postcardiotomy cardiogenic shock. Ann Thorac Surg 57(6):14621468; discussion 971
13. Hsu PS, Chen JL, Hong GJ, Tsai YT, Lin CY, Lee CY, Chen YG, Tsai CS (2010) Extracorporeal
membrane oxygenation for refractory cardiogenic shock after cardiac surgery: predictors of
early mortality and outcome from 51 adult patients. Eur J Cardiothorac Surg 37(2):328333
14. Subramaniam K, Boisen M et al (2012) Mechanical circulatory support for cardiogenic shock.
Best Pract Res Clin Anaesthesiol 26:131146
15. Collart F, Kerbaul F, Mekkaoui C, Riberi A, Gariboldi V, Rolland PH, Metras D, Mesana TG
(2004) Balloon-pump-induced pulsatility improves coronary and carotid flows in an experimental model of BioMedicus left ventricular assistance. Artif Organs 28(8):743746
circuit. Crit Care Med 34(10):26032606
17. Madershahian N, Salehi-Gilani S, Naraghi H et al (2011) Biventricular decompression by
trans-septal positioning of venous ECMO cannula through patent foramen ovale. J Cardiovasc
Surg (Torino) 52(6):900
18. Seib PM, Faulkner SC, Erickson CC et al (1999) Blade and balloon atrial septostomy for left
heart decompression in patients with severe ventricular dysfunction on extracorporeal membrane oxygenation. Catheter Cardiovasc Interv 46(2):179186
by venoarterial ECMO and transaortic left ventricular venting. Int J Artif Organs
27(5):410413
20. Jouan J, Grinda JM, Bricourt MO et al (2009) Successful left ventricular decompression following peripheral extracorporeal membrane oxygenation by percutaneous placement of a
micro-axial flow pump. J Heart Lung Transplant 29(1):135136
adult patients. ASAIO J 57(1):3840
22. Sidebotham D, McGeorge A, McGuinness S et al (2010) Extracorporeal membrane oxygenation for treating severe cardiac and respiratory failure in adults: part 2Technical considerations. J Cardiothorac Vasc Anesth 24:164172
23. Wong JK, Smith TN, Pitcher HT et al (2012) Cerebral and lower limb near-infrared spectroscopy in adults on extracorporeal membrane oxygenation. Artif Organs 36(8):659667
24. Mateen FJ, Muralidharan R, Shinohara RT et al (2011) Neurological injury in adults treated
with extracorporeal membrane oxygenation. Arch Neurol 68:15431549
25. Lan C, Tsai PR, Chen YS, Ko WJ (2010) Prognostic factors for adult patients receiving extracorporeal membrane oxygenation as mechanical circulatory support: a 14-year experience at a
medical center. Artif Organs 34:E59E64
[email protected]
136
M. Baiocchi et al.
26. Massetti M, Gaudino M, Crea F (2013) How to transform peripheral extracorporeal membrane
oxygenation in the simplest mid-term paracorporeal ventricular assist device. Int J Cardiol
66(3):551553
27. Navia JL, Atik FA, Beyer EA, Ruda VP (2005) Extracorporeal membrane oxygenation with
right axillary artery perfusion. Ann Thorac Surg 79:21632165
28. Maganti MD, Rao V, Borger MA, Ivanov J, David TE (2005) Predictors of low cardiac output
syndrome after isolated aortic valve surgery. Circulation 112(9 Suppl):I448I452
25(2):131141
30. Kato TS, Farr M, Schulze PC et al (2012) Usefulness of two-dimensional echocardiographic
parameters of the left side of the heart to predict right ventricular failure after left ventricular
assist device implantation. Am J Cardiol 109(2):246251
31. Raina A, Seetha Rammohan HR, Gertz ZM et al (2013) Postoperative right ventricular failure
after left ventricular assist device placement is predicted by preoperative echocardiographic
structural, hemodynamic, and functional parameters. J Card Fail 19(1):1624
32. Jardin F, Vieillard-Baron A (2003) Right ventricular function and positive pressure ventilation
in clinical practice: from hemodynamic subsets to respirator settings. Intensive Care Med
29(9):14261434
33. Yan X, Jia S, Meng X et al (2010) Acute kidney injury in adult postcardiotomy patients with
extracorporeal membrane oxygenation: evaluation of the RIFLE classification and the Acute
Kidney Injury Network criteria. Eur J Cardiothorac Surg 37(2):334338
34. Chen YC, Tsai FC, Chang CH et al (2011) Prognosis of patients on extracorporeal membrane
oxygenation: the impact of acute kidney injury on mortality. Ann Thorac Surg
91(1):137142
35. Kinsella JP, Gerstmann DR, Rosenberg AA (1992) The effect of extracorporeal membrane
oxygenation on coronary perfusion and regional blood flow distribution. Pediatr Res
31(1):8084
Cardiothorac Vasc Anesth 13(3):154175
37. Chan AK, Leaker M, Burrows FA et al (1997) Coagulation and fibrinolytic profile of paediatric
patients undergoing cardiopulmonary bypass. Thromb Haemost 77:270277
38. Ranucci M (2012) Bivalirudin and postcardiotomy ECMO: a word of caution. Crit Care
16(3):427
39. Aissaoui N, Luyt CE, Leprince P et al (2011) Predictors of successful extracorporeal membrane oxygenation (ECMO) weaning after assistance for refractory cardiogenic shock.
Intensive Care Med 37(11):17381745
40. Deng MC, Weyand M, Hammel D et al (1998) Selection and outcome of ventricular assist
device patients: the Muenster experience. J Heart Lung Transplant 17:817825
41. Rao V (2006) Condition critical: can mechanical support prevent death due to postcardiotomy
shock? J Card Surg 21:238239
42. Saito S, Nakatani T et al (2007) Is extracorporeal life support contraindicated in elderly
patients? Ann Thorac Surg 83:140145
43. Ko WJ, Lin CY, Chen RJ et al (2002) Extracorporeal membrane oxygenation support for adult
postcardiotomy cardiogenic shock. Ann Thorac Surg 73:538545
44. Zhang R, Kofidis T, Kamiya H et al (2006) Creatine kinase isoenzyme MB relative index as
predictor of mortality on extracorporeal membrane oxygenation support for postcardiotomy
cardiogenic shock in adult patients. Eur J Cardiothorac Surg 30:617620
45. Pokersnik JA (2012) Have change in ECMO technology impacted outcomes in adult patients
developing postcardiotomy cardiogenic shock? J Card Surg 27:246252
46. Yu K, Long C, Hei F et al (2011) Clinical evaluation of two different extracorporeal membrane
oxygenation system: a single center report. Artif Organs 35:733737
[email protected]
ECMO in Myocarditis
and Rare Cardiomyopathies
12
Barbara Cortinovis, Monica Scanziani, and Simona Celotti
12.1
Introduction
Myocarditis is an inflammatory condition of the heart muscle, mostly affecting young

individuals without significant comorbidities. Its unpredictable clinical presentation and
evolution to life-threatening arrhythmias and overt cardiogenic shock explains the major
role of left ventricular (LV) mechanical support in affecting short- and long-term prognosis. Though representing a minority of overall indications to ECMO, myocarditis is
frequently reported in literature in numerous case reports, case series, and clinical trials,
as a group of diseases that offers a wide spectrum of applications for mechanical support
for the failing heart. We will review the main features of this group and some rare cardiomyopathies, especially focusing on peculiar aspects regarding ECMO support.
12.2
Epidemiology
The true incidence and prevalence of myocarditis in general population is unknown,

mostly because reaching a definitive diagnosis might be challenging, due to the lack
of a well-established gold-standard noninvasive test and to the low sensitivity of
endomyocardial biopsy (EMB). Dallas criteria applied to conventional histology
yield a diagnosis only in 35 % of cases [1], and although multiple sampling [1],
immunohistochemical analysis [2], and viral polymerase chain reaction (PCR) have
yielded higher sensitivity [3, 4], definitive myocarditis can still be diagnosed in only
64 % of patients [5]. Incidence of myocarditis is about 3.56 % in different study
populations, including both cardiac and noncardiac deaths [68]. Similarly, the
B. Cortinovis (*) M. Scanziani S. Celotti
Cardiac Anesthesia and Intensive Care Unit, San Gerardo Hospital,
Department of Anesthesia and Intensive Care Medicine,
University of Milano-Bicocca, Milan, Italy
[email protected]
137
138
B. Cortinovis et al.
prevalence of myocarditis in idiopathic dilative cardiomyopathy (DCM) population

is 910 %, often derived from postmortem analysis [9].
12.3
Etiology and Pathogenesis
Etiology of myocarditis is classically divided into infectious causes, in which viral myocarditis represents, by far, the most common isolated species of pathogens and noninfectious causes, represented by a fairly heterogeneous group of diseases (Table 12.1).
Table 12.1 Etiology of myocarditis
Infectious causes
Viral
Adenovirus, Coxsackie A and B, echoviruses, parvovirus B19,
influenza A and B, herpes simplex, Epstein-Barr, cytomegalovirus,
varicella zoster, respiratory syncytial virus, HIV, hepatitis B and C,
polio and non-polio enteroviruses, rubeola, rubella, mumps, variola,
rabies, arbovirus, dengue, yellow fever
Vaccinia (smallpox vaccine)
Bacteria
Diphtheria, TB, Salmonella, Staphylococcus, Streptococcus spp.,
Neisseria spp., Clostridium spp., Brucella, Chlamydia spp.,
Legionella, Haemophilus, cholera, Mycoplasma
Fungal
Candida spp., Histoplasma, Coccidiomyces, Aspergillus,
Blastomyces, Cryptococcus
Others
Spirochetal (syphilis, leptospirosis, Lyme)
Rickettsial (typhus, Rocky mountain spotted fever, Q fever)
Protozoal (Toxoplasma, amebiasis, malaria, leishmaniasis,
trypanosomiasis)
Helmintic (echinococcosis, trichinosis, schistosomiasis, ascariasis,
filariasis, paragonimiasis, strongyloidiasis)
Noninfectious causes
Drug induced (direct toxicity): cocaine, alcohol, catecholamines,
arsenic, lead, cyclophosphamide, daunorubicin, Adriamycin
Drug induced (hypersensitivity): methyldopa, hydrochlorothiazide,
ampicillin, furosemide, digoxin, tetracycline, aminophylline,
phenytoin, benzodiazepines, and tricyclic antidepressants
Environmental exposure: snake, scorpion, spider, or insect bites
Collagen-vascular diseases: systemic lupus erythematosus, systemic
sclerosis, rheumatoid arthritis, dermatomyositis/polymyositis
(sarcoidosis, celiac disease, etc.)
Radiation exposure
Various: giant cell myocarditis, sarcoid, peripartum, thyrotoxicosis,
pheochromocytoma, celiac disease
[email protected]
Tests and clues

Viral cultures and titers
Swabs of rectal and
nasal mucosa
Acute and convalescent
antibody titers
History of recent
vaccination
Bacterial cultures
Early antigens (if
available)
Beta-D-glucan
Galactomannan antigen
Careful and detailed
history (travel,
exposure, tick bite, etc.)
and physical exam
Careful and detailed

history and physical
exam
Toxicologic panels
Drug dosage and

ethanol levels
Autoimmune workup
Thyroid function and

urinary catecholamines
12
ECMO in Myocarditis and Rare Cardiomyopathies
139
Bowles et al. published a large interesting study, defining the most common viral
etiology by age group [1]. Besides more common viruses, we also mention myocarditis in HIV-positive patients (for either potential toxicity of gp120 protein or
adverse reaction to antiviral agents or to opportunistic infections), in which, although
uncommon, it is associated with advanced disease and poor prognosis [10];
0.013 % incidence rates of vaccinia-associated myopericarditis are estimated,
occurring within 30 days after smallpox vaccination [11, 12]; and finally, patients
with endoscopically proven celiac disease may present a form of virus-negative
myocarditis, showing higher titer of serum anti-heart antibodies than general population (4.8 % vs. 0.3 % in control group), presenting with severely depressed ejection fraction (EF) and high NYHA class or ventricular arrhythmias, improved by a
gluten-free diet [13].
Pathogenesis of myocarditis is still largely unexplained. From experimental studies in animal models and human population, it was demonstrated that cardiac injury
depends on direct viral damage, often requiring expression of surface receptors, and
on humoral and cellular host immune response, especially to persistence and replication of viral genome within myocardial tissue [14]. In fact this is associated with a
progressive impairment of LVEF and incomplete functional recovery [4, 15], whereas
spontaneous viral elimination is associated with a significant improvement in LV
function. It is now accepted that individual susceptibility together with stronger
native immunity is able to affect both the initial inflammatory response (resulting in
shorter duration of severe symptoms) and long-term prognosis (lower rate of progression to DCM) [15]. Autoimmune mechanisms have also been suggested to
explain virus-negative myocarditis. In conclusion, initial immune response limits the
extent of early viremia, therefore protecting against myocarditis, but once the cardiac
damage ensues, the persistence of viral genome can trigger autoimmune response.
Noninfectious causes of myocarditis are summarized in Table 12.1.
12.3.1 Hypersensitivity and Eosinophilic Myocarditis

HSM is a drug-mediated autoimmune reaction toward the cardiac muscle, characterized by signs and symptoms of hypersensitivity (skin rash, fever, eosinophilia,
and malaise) and nonspecific ECG findings that may occur not necessarily early in
the course of drug use (up to 2 years from initial drug assumption). Burke et al.
analyzed postmortem histologic samples of patients with HSM to describe histologic findings, patterns of distribution of cellular infiltrates, drug associations, and
clinical-histologic correlation [16]: HSM is defined by the infiltration of eosinophils, lymphocytes, and histiocytes in the absence of fibrosis or granulation tissue.
Involvement of myocardium is often focal and may be missed in up to 50 % of
EMB. The right ventricle (RV) was involved in the majority of patients. Cardiac
arrhythmias or unexplained death occurred in 29 of 69 patients, and even worse
prognosis was reported by Fenoglio et al. (20 sudden cardiac deaths in 24 patients)
[17]. The differential diagnosis with GCM or necrotizing eosinophilic myocarditis
can be made only by EMB, but it is crucial for possible treatment; therefore,
[email protected]
140
accompanying symptoms should be carefully sought. Necrotizing eosinophilic

myocarditis associated with the hypereosinophilic syndrome typically evolves over
weeks to months, presenting with biventricular failure or arrhythmias, including
sudden cardiac deaths (SCD). It has an exceptionally poor prognosis with most
cases diagnosed at autopsy.
12.3.2 Giant Cell Myocarditis

GCM is a rare form of autoimmune myocarditis, of unknown pathogenesis, characterized by fulminant course, and grave prognosis, despite best medical treatment. It
was previously described only in postmortem samples or explanted hearts. More
recently, a multicenter international registry described 63 patients who had a
transplant-free survival of 5.5 months without immunosuppressive treatment [18].
Cooper et al. enrolled 11 patients with histologically confirmed GCM, reporting
that treatment with steroids, cyclosporine, and monoclonal antilymphocyte antibodies improves long-term survival [19]. A relapse of the disease is described after
acute discontinuation of treatment, confirming previously reported findings of histologic recurrence, in posttransplant follow-up EMB. It usually presents with acute
deterioration in LVEF, ventricular arrhythmias, and heart block.
12.3.3 Peripartum Cardiomyopathies (PP-CMP)

PP-CMP is rare dilative CMP of unknown etiology affecting women in the last
month of pregnancy or within 5 months from delivery, although similar dilative
CMP has been described also earlier in pregnancy. It presents with profound cardiogenic shock and major arrhythmias, in young healthy patients, without cardiac or
extra-cardiac comorbidities. Although prone to spontaneous recovery, it may present with severe LV dysfunction, mandating prompt mechanical support, and, like
acute myocarditis, progresses to dilative CMP, requiring VAD or heart transplantation, despite the best medical treatment. Maternal mortality rate reaches 2550 %.
In literature, several case reports advocate the use of ECMO to allow recovery or to
bridge the patients to VAD or transplantation. We report Gavaert et al.s work
describing a 10-year experience with 6 patients affected by PP-CMP, all supported
by IABP, 1 by ECMO, 4 implanted with an LVAD, 2 of which were transplanted,
and the last being still on list at time of publication. EMB were performed in 2
patients at the time of VAD implant [20].
12.3.4 Catecholamine-Induced and Takotsubo Cardiomyopathies

Pheochromocytoma is a catecholamine-secreting tumor. It originates from the chromaffin cells in the adrenal medulla or in extra-adrenal paragangliomas. It presents
with a variety of symptoms, but although catecholamine-induced CMP is a
[email protected]
12
141
well-known entity, it presents usually with more benign features and more rarely with
unexplained or intractable cardiogenic shock due to paroxysmal release of catecholamines. Grinda et al. described the first case report of successful use of VAD to rescue
catecholamine-induced CMP. Huang et al. reported a case series of 3 patients, rescued
with venoarterial peripherally inserted ECMO, two of which were under CPR [21].
Literature offers several case reports of centrally [22] and peripherally [23, 24]
inserted ECMO and VAD [25]. Septostomy was performed in some cases to decompress the LV. Functional recovery usually occurs within the first few days allowing for
further diagnosis and surgical treatment of underlying disease. Interestingly, Sheinberg
et al. reported a Takotsubo (apical ballooning appearance of LV) in a pheochromocytoma, successfully rescued by venoarterial ECMO and IABP [23].
12.3.5 Hantavirus Cardiopulmonary Syndrome

Andes virus, Bayou virus, Black Creek Canal virus, Choclo virus, Juquitiba virus,
Laguna Negra virus, and Sin Nombre virus, first isolated in the Four Corners region
of southwestern, USA, are the etiologic agents of an acute cardiac and respiratory
failure called Hantavirus cardiopulmonary syndrome (HCPS), causing deaths (43
76 % of patients) mostly due to multiorgan failure, secondary to intractable cardiogenic shock. Currently, no etiologic treatment is available, and therapy is primarily
supportive. Wernly et al. report a two thirds survival and complete recovery of 51
HCPS patients with a predicted mortality of 100 % [26], rescued by venoarterial percutaneous femoral ECMO. Dietl et al reported 38 patients with severe HCPS with
similar technique. ECMO had a mean duration of 132 h. Several reports are available
in literature [27].
12.4
Clinical Presentation
The extreme variability of histologic patterns (i.e., focality of infiltration vs. diffuse
biventricular injury) together with innate immune response could significantly affect
the presentation and course of the disease, accounting for a multiplicity of clinical
patterns. The majority of patients present with a nonspecific prodrome, mostly confined to respiratory and/or gastrointestinal systems, subsequently progressing in
overt although still aspecific cardiac involvement. Clinical features of myocarditis
are summarized in Table 12.2. Myocarditis is a major cause of DCM, presenting with
classical symptoms of heart failure (HF), with atrial and ventricular arrhythmias,
usually manageable with standard of treatment. Conduction delay is more common
with infiltrative and GCM than with lymphocytic. Myocarditis represents a cause of
SCD without structural abnormalities in up to 20 % of cases [28, 29], in young subjects with little or no prodrome and regardless activity or rest. Only a minority of
patients were reported to have SCD during physical or emotional stress [28].
Important to mention, even asymptomatic patients are at risk for SCD. Fulminant
myocarditis is well characterized by viral prodrome, acute onset, severe
[email protected]
142
Table 12.2 Clinical presentation of myocarditis

Aspecific
Cardiac
1. Gastrointestinal: nausea and vomiting, cramp, diarrhea, appetite loss,

abdominal and epigastric pain
2. Respiratory: cough, pharyngeal pain
3. General: increased fever, general fatigue, arthralgia and myalgia, headache,
and back pain
1. Chest pain or discomfort (particularly common in young patients with
coronary vasospasm), concomitant pericarditis, syncope, palpitations, dyspnea
2. Heart failure: generalized fatigue, intolerance to exercise and dyspnea.
Subsequently acute or fulminant cardiogenic shock
3. Sinus tachycardia (most common, especially out of proportion with
concomitant fever), premature atrial and ventricular contraction, atrial
fibrillation, and ventricular tachycardia. I and II degree, up to complete AV
block (more common with infiltrative and GCM), RBBB, and LBBB
cardiovascular compromise, and ventricular dysfunction that either resolves spontaneously or rapidly evolves to a fatal course. Acute myocarditis develops ventricular
dysfunction that usually progresses toward DCM. Failure of standard of treatment
for congestive HF, refractory arrhythmias, and cardiac arrest is therefore the major
indication to LV mechanical support, to bridge the patient to full recovery, or to further support (heart transplantation or long-term support devices) in selected cases
with otherwise extremely high mortality over the course of a few days.
12.5
Diagnosis
As previously discussed, diagnostic process in myocarditis is complex and articulated. It is based on standard screening routine but also on specific biomarkers and
invasive tests. The primary goals in approaching this disease are to select patients
who will candidate for further testing; to reach a definitive diagnosis, but also to
anticipate, compatibly with the clinical scenario which patients may require LV
mechanical support; and possibly to estimate the best option and the likelihood of
progression to intermediate-term assist devices and/or transplantation. A thorough
discussion of available diagnostic tests and their role in myocarditis is beyond the
aim of this chapter. Nonetheless, we will report the most relevant features, and we
grouped diagnostic tests for each specific category, in the purpose to offer an
extremely practical approach in more emergent situations (Table 12.1). Myocarditis
requires a high index of suspicion and should be considered whenever a patient
(primarily young males, with gastrointestinal and respiratory prodrome) presents
with new onset of unexplained cardiovascular abnormalities. Valvular, congenital,
ischemic, toxic (especially ethanol and cocaine related), and pulmonary heart disease should be carefully sought and excluded, before establishing a diagnosis of
myocarditis. Concomitant symptoms, such as exanthematous disease, specific
pathogen-related symptoms, co-existent pericarditis, or a triad of eosinophilia, rash,
and exposure to either a new drug or vaccine, may reveal useful to further restrict
differential diagnosis. Therefore, careful and thorough history and physical examination are essential and may help in selecting more focused testing. Some features
[email protected]
12
143
Table 12.3 Diagnostic tests results in myocarditis

Diagnostic test
Chest radiography
ECG
Cardiac enzymes
Echocardiography
Magnetic resonance
imaging
Endomyocardial
biopsy
Common features
Cardiomegaly, pulmonary vascular congestion, and/or pleural effusion
(depending on right ventricular and tricuspid valve dysfunction). No
specific abnormalities
Completely normal or with minor, aspecific abnormalities to markedly
abnormal
Brady- and tachyarrhythmias, conduction abnormalities, ST and T wave
alterations, abnormal Q wave, low voltage, and poor R wave
progression. Patterns of infarction or pericarditis. Presence of Q wave
associated with a severe course, higher early cardiac enzymes, worse LV
function, and higher incidence of cardiogenic shock, but not necessarily
with a worse long-term outcome (31)
CPK MB, TnT, and TnI may be elevated (reflecting the extent of
myocardial injury). TnI probably superior to CPK MB early in the
disease, TnT levels correlating with more extensive damage
Highly variable, from a completely normal to a markedly abnormal
Various degrees of hypokinesis most commonly (mild hypokinesis
limited to areas of focal infiltration, segmental wall motion
abnormalities or diffuse involvement, with severe hypokinesis)
LVEF or FS reduced, not necessarily with LV dilatation
Areas of inflammation and infiltration, mostly focal within the first
2 weeks, more diffuse within 4 weeks. Extent of the lesion correlates
with LV dysfunction
510 samples from RV septum. Submit 45 to light microscopic
examination. Transmission electron microscopy may be useful but
reserved to infiltrative disorders
Routine viral genome testing only for referral centers
Several patterns of infiltration (histiocytic and mononuclear), varying in
severity and structural abnormalities of myocardium
of routine test may be helpful in suggesting myocarditis (Table 12.3), but they are
usually non-conclusive.
12.5.1 Chest Radiography, ECG, and Laboratory Tests

The most common findings on chest radiography, ECG, and laboratory testing are
summarized in Table 12.3. Etiologic diagnosis may be suggested by viral cultures
and titers, swabs of rectal and nasal mucosa, and acute and convalescent antibody
titers. If EMB is performed, samples should be sent for viral and bacterial culture,
as well as for PCR analysis.
12.5.2 Transthoracic and Transesophageal

Echocardiography (TTE-TEE)
TTE-TEE is useful in myocarditis for many reasons: it is a noninvasive test useful
to support clinical suspicion, and it provides assessment of LV performance,
[email protected]
144
functional and anatomical abnormalities of cardiac chambers, valvular apparatus,

and pericardial involvement (i.e., in myopericarditis) and detection of intracavitary
thrombi. Last, position of the cannulas, assessment of LV decompression, weaning
process, and detection of specific ECMO-related complications could be easily and
consistently followed by TTE-TEE. The exam usually reveals several degrees and
extent of hypokinesis, associated with LV or biventricular systolic dysfunction, not
necessarily associated with chamber dilatation. The most interesting aspect of TTETEE is the possibility of discriminating between fulminant and acute myocarditis:
the first presents with a near-normal LV size and septal thickening, whereas the latter presents with increased diastolic dimensions and normal septal thickness. Both
the clinical presentations were associated with a decrease in FS [30]. This knowledge is particularly useful in anticipating the need of ventricular support directly at
presentation, and it allows to modify referral to ECMO centers and diagnostic and
therapeutic pathways accordingly, as well as offering prognostic value.
12.5.3 Cardiac Magnetic Resonance (MRI)

Several reports showed that inflammation is reflected by signal changes in contrastenhanced magnetic resonance imaging (MRI), due to the characteristics of gadolinium penetrating into extracellular fluid, but not into living cells: in acute
inflammation, lymphocytic infiltrate, cell damage, and interstitial edema lead to
accumulation of contrast in inflammatory lesions.
Contrast-enhanced MRI is therefore useful to document the location and extent
of inflammation.
Different studies report that within the first 2 weeks, myocardial damage is
mostly focal and became more diffuse within 4 weeks, with the extent of the lesion
correlating with LV dysfunction [31, 32]. Moreover, the sensitivity of MRI in identifying areas of inflammation and infiltration may guide EMB, thus theoretically
improving diagnostic yield, and different viral species seem to produce different
patterns of involvement of myocardium, further restricting differential diagnosis.
12.5.4 Endomyocardial Biopsy

Endomyocardial biopsy is the gold standard for a definitive diagnosis of myocarditis, and based on the clinical scenario, guidelines strongly recommend to perform
EMB in selected cases: a new-onset HF (less than 2 weeks) with hemodynamic
compromise and normal-sized or dilated LV (fulminant myocarditis), a new-onset
HF (2 weeks to 3 months) with dilated LV and new ventricular arrhythmias, II- or
III-degree AV block, and refractory to standard treatment (GCM). Cardiac tissue
samples may be obtained by standard EMB, which will be briefly illustrated, and by
open surgical approach. At the present time, high-quality biotomes allow multiple
sampling, commonly by transjugular approach (longer biotomes for transfemoral
approach), usually guided by echocardiography (less commonly) or fluoroscopy,
[email protected]
12
145
with sampling mostly in the RV. It is reported that biventricular sampling may
increase the diagnostic yield in myocarditis, compared to uni-ventricular biopsy,
with comparable complication rate. Although hypothetically contrast-enhanced
MRI could guide in choosing the ideal site to perform EMB, this was not confirmed.
Complications of EMB are derived from case reports and include immediate (heart
structure and great vessel damage, major arrhythmias, tamponade, pneumothorax,
etc.) and delayed complications (bleeding, tricuspid valve damage, tamponade).
Their relative incidence is unknown. Histologic examination reveals several patterns of infiltration (histiocytic and mononuclear), varying in severity and associating with structural abnormalities of myocardium. Despite being the actual gold
standard, Dallas criteria interpretation of EMB has a low diagnostic yield. This
depends on focal and transient nature of the disease, sampling error (either site
choice or number of samples), possible disagreement in interpretation of histology
even by expert pathologists, and most importantly issues concerning immunohistochemistry and PCR analysis, which were not routinely performed in earlier publications (hence the lower sensitivity) and the fact that some pathogens do not cause
intense inflammatory reactions, therefore limiting histologic abnormalities of the
cardiac tissue.
12.6
ECMO in Myocarditis
12.6.1 Indications
As previously discussed, myocarditis affects a group of patients with extremely
severe cardiovascular failure, who paradoxically have a good prognosis and a high
likelihood of return to normal cardiac function. This per se explains why this group
could benefit from ECMO, mostly until recovery, a minority to an intermediate- and
long-term cardiac support or to heart transplantation. Indications to ECMO could be
summarized as cardiogenic shock and life-threatening arrhythmias, including SCD.
Some patients will candidate for ECMO, during or after cardiac arrest; therefore,
assessing the quality of resuscitation is mandatory before proceeding to mechanical
support.
12.6.2 Diagnosis and ECMO

The diagnostic process needs to be adapted accordingly: noninvasive tests pose no
particular dilemma and therefore should be immediately performed as soon as the
suspicion arises. We underline once more the value of TTE-TEE in myocarditis,
especially to discriminate fulminant myocarditis [30], to attempt predicting its rapid
evolution, and to anticipate the need for ECMO. More invasive or unpractical tests
need to be discussed further. First, it is well established that patients with myocarditis should be transferred to a specialized facility where EMB is available. Given
the unpredictable nature of this disease and the extremely rapid evolution to LV
[email protected]
146
mechanical failure, we would suggest to transfer the patients to facilities where

ECMO is available, to offer them adequate cardiovascular support, while still conducting the diagnostic process. Moreover, transjugular EMB is an invasive procedure, contraindicated if the patient is receiving anticoagulation for ECMO. To our
knowledge, it is not reported in literature, nor it is the experience of either our center
or of centers we mostly collaborated with, to perform EMB while the patient is on
peripherally inserted ECMO, because anticoagulation and the impossibility of adequate hemostasis make the procedure unsafe. It is reported in literature that EMB
was performed in myocarditis patients, shortly after they were weaned from ECMO
still reaching an acceptable diagnostic yield. This would exclude patients with nonfulminant myocarditis, who would not be weaned from ECMO: most of these
patients, anyways, will have indications to intermediate- or long-term VAD, which
is usually performed with a central access through median sternotomy. A surgically
performed biopsy, by expert hands and with immediate hemostasis, seems advisable
for this category. Finally, patients with myocarditis will require diagnostic tests in
remote facilities: we recommend an ECMO standby for high-risk procedures and
for patients with high likelihood of rapid deterioration.
12.6.3 Which Support?

Literature accounts for numerous case reports or case series but a relatively small
amount of large cohort studies, some of which were performed in children or young
adults; therefore, some of the data reported here are extrapolated from pediatric
population.
Both peripherally and centrally inserted ECMO are equally feasible in this group
of patients, with slight advantage for percutaneous peripheral ECMO, for practical
reasons. Most frequent percutaneous sites are jugular and femoral veins and femoral
artery, although carotid artery is sometimes preferred, especially in children.
Transthoracic approach usually requires right atrium and aortic cannulation. In some
cases, the use of a secondary cannulation site is reported to optimize ECMO performance. Depending on the characteristics of the patients and occurrence of concomitant respiratory failure, circuit choice may be varied from standard extracorporeal
centrifugal pump with an oxygenator to a paracorporeal non-pulsatile VAD without
an oxygenator, to offer the advantages of lower heparin requirements. It is important
to underline that some pathogens of myocarditis could affect the patient primarily or
exclusively with respiratory failure: it is important to modify the approach to ECMO
considering that purely respiratory patients (discussed elsewhere in the book) could
evolve to cardiovascular collapse, therefore needing a switch from venovenous to
venoarterial ECMO or to a combination of the two, or that isolated myocarditis may
evolve to respiratory failure, therefore contraindicating the use of VAD for isolated
cardiac failure. One of the largest published studies is a retrospective ELSO registry
review, enrolling 260 ECMO runs in 255 (<18 years of age) patients, with the intent
of defining survival outcomes, trends in ECMO use, and its complications [33].
Overall survival was 61 %, in line with adult population reported elsewhere [34],
[email protected]
12
147
with 12 % patients deaths after weaning from ECMO and the remaining for ECMO
withdrawal due to irreversible organ failure. Six of the seven transplanted patients
survived to hospital discharge. Pre-cannulation variables correlating with poor outcomes are metabolic acidosis, more severe hypotension, pre-ECMO cardiac arrest,
and enterovirus infection [33]. No major ECMO technical difference was reported
between survivors and nonsurvivors, although ECMO for isolated cardiac failure
carried better prognosis. Complications during ECMO are commonly reported in the
majority of published literature, including major cardiac arrhythmias, renal failure
requiring dialysis, neurologic complications (seizures, infarction, hemorrhage, and
brain death, which in Rajagopal et al.s study [33] interestingly were not higher in
patients cannulated during CPR), metabolic acidosis, pulmonary hemorrhage (more
prominent in pediatric population and mandating LA decompression with venting),
DIC, higher cardiac enzymes [35], and hyperglycemia that are almost invariably
associated with worse outcomes. Careful attention should be made to maintain good
end-organ perfusion during ECMO: prevention and relief of abdominal compartment
syndrome, avoidance or prompt reperfusion of ischemic limb, and prevention of
organ and especially renal failure improved outcomes in many published papers [36].
Interesting to report, nearly 50 % of patients needing ECMO for refractory cardiac
arrest survived [33]. The use of IABP is associated with a better outcome in several
adult reports [35, 37].
Cannulation occurs early, usually in the first 24 h from presentation, whereas
total ECMO duration depends on underlying disease, being shorter for fulminant
lymphocytic myocarditis (<2 weeks), and reducing the need for early listing for
heart transplantation or escalation to VAD. This usually occurs in patients who do
not meet weaning criteria from ECMO after such period of time, mostly as a bridge
to subsequent transplantation. A variety of LV decompression techniques are also
described for left heart distention or lung edema [35]: LA or LV drain and ASD
creation. In pediatric population, ventriculotomy for VAD implantation or venting
is perceived as potentially detrimental for development of future arrhythmias or
ventricular dysfunction [33]. A thorough discussion of indications, implantation,
management, and weaning from VAD is beyond the scope of this chapter, and we
refer the reader to the specific section in the book.
12.7
Prognosis and ECMO
Prognosis for fulminant myocarditis is paradoxically good: survival rates are reported
up to 75 % with mechanical support and favorable long-term outcome at 6 months
with complete recovery of LV function. GCM and eosinophilic myocarditis have
overall poor prognosis, with a relapse of disease when specific immunosuppressive
treatment is discontinued and histologic recurrence post transplantation. Finally, we
report an interesting paper on outcomes and psychophysiologic assessment of 41
patients with fulminant myocarditis rescued by mechanical support (including 6
BiVAD and 35 ECMO). Compared to sex- and age-matched controls, myocarditis
patients show satisfactory mental health and vitality but persistent physical- and
[email protected]
148
psychosocial-related difficulties, with about a third of patients showing symptoms

such as PTSD, anxiety, and depression, therefore underlining the need for optimizing
strategies aimed at reducing emotional distress and its long-term sequelae [34].
References
1. Bowles NE, Ni J, Kearney DL et al (2003) Detection of viruses in myocardial tissues by polymerase chain reaction. Evidence of adenovirus as a common cause of myocarditis in children
and adults. J Am Coll Cardiol 42:466472
2. Pankuweit S, Moll R, Baandrup U et al (2003) Prevalence of the parvovirus B19 genome in
endomyocardial biopsy specimens. Hum Pathol 34:497503
3. Kindermann I, Kindermann M, Kandolf R et al (2008) Predictors of outcome in patients with
suspected myocarditis. Circulation 118:639648
4. Khl U, Pauschinger M, Seeberg B et al (2005) Viral persistence in the myocardium is associated with progressive cardiac dysfunction. Circulation 112:19651970
5. Mason JW, OConnell JB, Herskowitz A et al (1995) A clinical trial of immunosuppressive
therapy for myocarditis. The Myocarditis Treatment Trial Investigators. N Engl J Med
333(5):269275
6. Maron BJ, Doerer JJ, Haas TS et al (2009) Sudden deaths in young competitive athletes: analysis of 1866 deaths in the United States, 1980-2006. Circulation 119:1085
7. Lambert EC, Menon VA, Wagner HR et al (1974) Sudden unexpected death from cardiovascular disease in children. A cooperative international study. Am J Cardiol 34:8992
8. Stevens PJ, Ground KE (1970) Occurrence and significance of myocarditis in trauma. Aerosp
Med 41:776780
9. Felker GM, Thompson RE, Hare JM et al (2000) Underlying causes and long-term survival in
patients with initially unexplained cardiomyopathy. N Engl J Med 342:10771084
10. Herskowitz A, Willoughby SB, Baughman KL et al (1992) Cardiomyopathy associated with
antiretroviral therapy in patients with HIV infection: a report of six cases. Ann Intern Med
116(4):311313
11. Halsell JS, Riddle JR, Atwood JE et al (2003) Myopericarditis following smallpox vaccination
among vaccinia-naive US military, personnel. JAMA 289:32833289
12. Cassimatis DC, Atwood JE, Engler RM et al (2004) Smallpox vaccination and myopericarditis: a clinical review. J Am Coll Cardiol 43:15031510
13. Frustaci A, Cuoco L, Chimenti C et al (2002) Celiac disease associated with autoimmune
myocarditis. Circulation 105:26112618
14. Bergelson JM, Cunningham JA, Droguett G et al (1997) Isolation of a common receptor for
Coxsackie B viruses and adenoviruses 2 and 5. Science 275:13201323
15. Martino TA, Liu P, Sole MJ (1994) Viral infection and the pathogenesis of dilated cardiomyopathy. Circ Res 74:182188
16. Burke AP, Saenger J, Mullick F et al (1991) Hypersensitivity myocarditis. Arch Pathol Lab
Med 115:764769
17. Fenoglio JJ Jr, McAllister HA Jr, Mullick FG (1981) Drug related myocarditis. I.
Hypersensitivity myocarditis. Hum Pathol 12:900907
18. Cooper LT Jr, Berry GJ, Shabetai R et al (1997) Idiopathic giant-cell myocarditis natural
history and treatment. Multicenter Giant Cell Myocarditis Study Group Investigators. N Engl
J Med 336:18601866
19. Cooper LT Jr, Hare JM, Tazelaar HD et al (2008) Usefulness of immunosuppression for giant
cell myocarditis. Am J Cardiol 102:15351539
20. Gavaert S, Van Belleghem Y, Bouchez S et al (2011) Acute and critically ill peripartum cardiomyopathyand bridge to therapeutic options: a single center experience with intra-aortic balloon pump, extra corporeal membrane oxygenation and continuous-flow left ventricular
device. Crit Care 15:R93
[email protected]
12
149
21. Huang J, Huang S, Chou N et al (2008) Extracorporeal membrane oxygenation rescue for
cardiopulmonary collapse secondary to pheochromocytoma: report of three cases. Intensive
Care Med 34:15511552
22. Banfi C, Juthier F, Ennezat P et al (2012) Central extracorporeal life support in pheochromocytoma crisis. Ann Thorac Surg 93:13031305
23. Sheinberg R, Gao W, Wand G et al (2012) A perfect storm: fatality resulting from metoclopramide unmasking a pheochromocytoma and its management. J Cardiothorac Vasc Anesth
26(1):161165
24. Chao A, Yeh YC, Yen TS et al (2008) Phaeochromocytoma crisis a rare indication for extracorporeal membrane oxygenation. Anaesthesia 63:8688
25. Grinda JM, Bricourt MO, Salvi S et al (2006) Unusual cardiogenic shock due to pheochromocytoma: recovery after bridge-to-bridge (extracorporeal life support and DeBakey ventricular
assist device) and right surrenalectomy. J Thorac Cardiovasc Surg 131:913914
26. Wernly JA, Dietl CA, Tabel CE et al (2011) Extracorporeal membrane oxygenation support
improves survival of patients with Hantavirus cardiopulmonary syndrome refractory to medical treatment. Eur J Cardiothorac Surg 40:13341340
27. Dietl CA, Wernly JA, Pett SB et al (2008) Extracorporeal membrane oxygenation support
improves survival of patients with severe Hantavirus cardiopulmonary syndrome. J Thorac
28. Theleman KP, Kuiper JJ, Roberts WC (2001) Acute myocarditis (predominately lymphocytic)
causing sudden death without heart failure. Am J Cardiol 88:10781083
29. Drory Y, Turetz Y, Hiss Y et al (1991) Sudden unexpected death in persons less than 40 years
of age. Am J Cardiol 68:13881392
30. Felker GM, Boehmer JP, Hruban RH et al (2000) Echocardiographic findings in fulminant and
acute myocarditis. J Am Coll Cardiol 36:227232
31. Friedrich MG, Strohm O, Schulz-Menger J et al (1998) Contrast media-enhanced magnetic
resonance imaging visualizes myocardial changes in the course of viral myocarditis.
32. Mahrholdt H, Goedecke C, Wagner A et al (2004) Cardiovascular magnetic resonance assessment of human myocarditis: a comparison to histology and molecular pathology. Circulation
109:12501258
33. Rajagopal SK, Almond CS, Laussen PC et al (2010) Extracorporeal membrane oxygenation
for the support of infants, children, and young adults with acute myocarditis: a review of the
Extracorporeal Life Support Organization registry. Crit Care Med 38:382387
34. Mirabel M, Luyt CE, Leprince P et al (2011) Outcomes, long-term quality of life, and psychologic assessment of fulminant myocarditis patients rescued by mechanical circulatory support.
35. Hsu KH, Chi NH, Wang CH et al (2011) Extracorporeal membranous oxygenation support for
acute fulminant myocarditis: analysis of a single centers experience. Eur J Cardiothorac Surg
40:682688
36. Aoyama N, Izumi T, Hiramori K et al (2002) National survey of fulminant myocarditis in
Japan. Therapeutic guidelines and long-term prognosis of using percutaneous cardiopulmonary support for fulminant myocarditis. Circ J 66:133144
37. Maejima Y, Yasu T, Kubo N et al (2004) Long-term prognosis of fulminant myocarditis rescued by percutaneous cardiopulmonary support device. Circ J 68:829833
[email protected]
ECMO for High-Risk Procedures
13
Fabio Ramponi, Paul Forrest, John F. Fraser,

Korana Musicki, and Michael P. Vallely
13.1
Introduction
A few years ago, while still at medical school, I was sitting at a wedding dinner,
when a consultant cardiac surgeon challenged the table with this joke: Whats the
difference between God and a cardiac surgeon? Well, God doesnt think he is a
cardiac surgeon! Without digressing into a metaphysical discussion, it is evident
that the impressive progresses of the last few decades, in both fields of science and
technology, have dramatically changed the management of cardiac conditions previously deemed too high risk.
The use of extracorporeal membrane oxygenation (ECMO) as cardiopulmonary
support has paved the way for new operative indications for those patients who were
previously relegated to conservative medical management. This may have been
because of poor left ventricular function, cardiogenic shock with complex multivessel disease, or multiple other co-morbidities. Recent applications have shown
ECMO to be potentially effective as a temporising measure or bridge to therapeutic
intervention in the setting of myocardial dysfunction and cardiogenic shock.
F. Ramponi, MD, FEBVS (*) K. Musicki, MBBS

Department of Cardiothoracic Surgery, The Baird Research Institute for Applied Heart and
Lung Surgical Research, Royal Prince Alfred Hospital, The University of Sydney,
Missenden Road, Sydney, NSW 2050, Australia
P. Forrest, MBChB, FANZCA M.P. Vallely, PhD, FRACS
Department of Cardiothoracic Anaesthesia, The Baird Research Institute for Applied Heart
and Lung Surgical Research, Royal Prince Alfred Hospital, The University of Sydney,
Missenden Road, Sydney, NSW 2050, Australia
J.F. Fraser, PhD, MRCP, FRCA, FFARCSI, FCICM
Department of Intensive Care, The Prince Charles Hospital, The University of Queensland,
627 Rode Road, Brisbane, QLD, 4032, Australia
[email protected]
151
152
F. Ramponi et al.
The combination of a compact pump-oxygenator design, with smaller, percutaneous, flexible cannulae, makes the modern veno-arterial VA ECMO an ideal miniaturised cardiopulmonary bypass; it is easy to set up and to accommodate at the
patient bedside. The 2009 H1N1 epidemic [1] exposed our unit to a more extensive
use of veno-veno VV ECMO. The consequent increase in expertise, and level of
familiarity with extracorporeal life support, extended its use to many others scenarios. Some of these include various high-risk catheter-based procedures, like percutaneous coronary interventions (PCI) and transcatheter aortic valve implantation
(TAVI) [24], post-infarct ventricular septal defect (PI-VSD) repair, surgery on the
thoracoabdominal aorta, international retrievals for respiratory and cardiac failure
[5], and bridge therapy before urgent pulmonary thromboembolectomy (PTE) [6].
As there is a paucity of prospective, randomised controlled trials in this area, the
evidence behind the use of VA ECMO during high-risk procedures has had to rely
on retrospective series or case reports. In this chapter, we will review the main indications for the use of VA ECMO for emergent support or backup during high-risk
procedures.
13.2
ECMO Support in High-Risk Coronary Angioplasty
13.2.1 Elective PCI in High-Risk Patients

and Left Main Procedures
Patients with unstable angina and ischaemic cardiomyopathy with poor left ventricular function represent a subgroup at high risk for either percutaneous or surgical
revascularisation. In this context, a transient coronary occlusion during angioplasty
can precipitate profound instability, irreversible arrhythmia, or fatal loss of cardiac
output.
ECMO-assisted PCI was initially introduced more than 20 years ago to provide
haemodynamic backup in the event of acute coronary occlusion post angioplasty. In
1988, Vogel reported the first experience from the University of Maryland with supported PCI and valvuloplasty in high-risk patients (nine and six, respectively) [7];
patients were defined as high risk if they presented with severe left ventricular
impairment or large amounts of myocardium perfused by the index vessel. In this
early experience, all patients were weaned off bypass within 90 min (pump flow
between 3 and 5 L/min) and the initial outcome was encouraging, with only one
reported death in each group (acute mesenteric ischaemia 8 h after PCI and sudden
ventricular fibrillation 12 h after valvuloplasty).
The following year, Taub et al. reported their experience with patients for whom
ECMO-supported PCI was necessary because of low left ventricular ejection fraction (mean LVEF 31.5 %, n = 7) [8]. The mechanical support permitted safe, longer
balloon inflation time, yielding satisfactory angiographic results. However, this was
at the expense of high complication rates in other areas, such as groin haematoma
(requiring transfusion), deep vein thrombosis, and iliac artery occlusion, as well as
a death from retroperitoneal haemorrhage. Similar results were described by the
[email protected]
13 ECMO for High-Risk Procedures
153
University of California on patients with unstable angina and ischaemic cardiomyopathy (mean LVEF 24 %, n = 5) who successfully underwent ECMO-assisted PCI
[9]; the main drawback was again the high rate of access site complications. This is
also probably a reflection of old cannulation technology.
More recently, Magovern reported the Allegheny experience with 27 high-risk
patients who were revascularised with PCI under ECMO support [10]; technical
success was achieved in 26 patients (96 %), including 12 requiring left main angioplasty. Most patients (85 %) survived to discharge, with sudden cardiac arrest and
heart failure as the main causes of the death in the remainder.
Left main coronary artery (LMCA) disease is usually an indication for surgical
revascularisation. However, patients at prohibitive risk for coronary artery bypass
grafts (CABG) can be directed to the angiography suite, accepting that unprotected
LMCA stenting in the high-risk population carries a mortality of 9 % at 30 days and
11 % at 1 year [11]. Significant risks associated with worse outcomes in the treatment of unprotected LMCA include a severely reduced LVEF (<35 %), a synchronous right coronary artery (RCA) occlusion, the use of angioplasty without stents,
and the presence of significant co-morbid conditions (older age and renal and respiratory failure). ECMO-supported LMCA stenting has been successfully reported
for the first time in 1996 by Irons et al. [12]; they described the case of a 70-year-old
woman with intractable unstable angina despite heparin, nitrates, and an intra-aortic
balloon pump (IABP). She was deemed to have unacceptable surgical risk due to
end-stage COPD (FEV1 <0.46). She tolerated ECMO-supported LMCA stenting
with multiple high-pressure balloon inflations and only a transient sinus bradycardia, but no ST changes or haemodynamic instability.
Likewise, other successful cases have been reported [13]: our first patient was an
81-year-old gentleman, who presented post STEMI with a 90 % LMCA stenosis
and an estimated mortality risk by logistic EuroSCORE of 47 % [2]. The procedure
was performed under general anaesthesia, and ECMO cannulation was established
after femoral cut-down and a single bolus of 10,000 units of intravenous heparin.
ECMO flow was maintained at 2.5 L/min for the duration of the case, and weaned
off after LMCA and RCA stenting. Note, we have since moved on to percutaneous
cannulation with pre-close technique [4].
13.2.2 Emergent PCI Post Myocardial Infarction

and Cardiogenic Shock
Patients with myocardial infarction (MI) who have out-of-hospital cardiac arrest
and cardiogenic shock have a high mortality rate (510 % of all STEMIs) [14, 15].
The SHOCK trial has since reported improved survival of up to 6 years with early
revascularisation compared to initial medical stabilisation and delayed revascularisation in this subset of patients [16]. The use of peri-procedural circulatory support
has become increasingly common with the availability of percutaneous left ventricular support devices such as the Impella recover LP2.5 (Abiomed Europe GmbH,
Aachen, Germany), TandemHeart (Cardiac Assist inc., Pittsburgh, PA) [17], or the
[email protected]
154
F. Ramponi et al.
use of IABPs. However, such devices support only the left ventricle, cannot take
over the patients gas exchange function, have high disposable costs, and lately the
use of IABP in this setting has been strongly questioned [18]. If systemic hypoxia
cannot be treated rapidly and effectively, the oxygen supply to the heart, brain, and
tissues will remain poor; restoration of spontaneous circulation will then be very
unlikely.
VA ECMO provides a less expensive full cardiopulmonary support in patients who
suffer an acute and profound but potentially treatable cardiac insult, complicated by
cardiogenic shock and recurrent cardiac arrest despite inotropes and IABP [19, 20].
Patients with profound cardiogenic shock will often present with a cardiac index
below 1.5 L/min/m2 and consequent acidosis secondary to hypoperfusion leading to
multi-organ failure. Prompt initiation of ECMO support with a flow rate of 2.55 L/
min, depending on the patients afterload and intravascular volume, will quickly stabilise the haemodynamic status, providing adequate cardiac output and peripheral
perfusion [21]; once rescued from the acute insult, the patient can rest on ECMO
until complete revascularisation with PCI or coronary bypass is achieved [22].
In the case of intra-procedural cardiac arrest, rapid mechanical chest compression is initially necessary to prevent no or low blood flow episodes [23]. However,
external cardiac compression is often not compatible with successful PCI, leading
to loss of wire access in the target vessel, unsuccessful revascularisation and prolonged procedures, or complete breakdown. Mechanical chest compression devices
can be supportive but are often responsible of severe thoracic and intra-abdominal
damage [24].
Lee et al. reported two cases of VA ECMO-assisted PCI in patients suffering
cardiac arrest and cardiogenic shock post STEMI [25]: one patient was successfully
bridged to transplant after ECMO-supported PCI, while the other one survived the
initial procedure but succumbed later to a severe anoxic brain injury. In this scenario, timing is essential: door-to-balloon time should be within maximum
4560 min. If circulatory support is necessary, ECMO should be ideally established
within 1015 min. This requires all of (1) impeccable coordination between cardiac
surgeons, cardiologists, perfusionists, and anaesthetists; (2) familiarity with percutaneous cannulation; and (3) prompt availability of cannulae in different sizes and
lengths.
Despite ECMO support, the outcome in this setting remains dismal. In a retrospective review of 36 patients with post-MI cardiogenic shock necessitating extracorporeal mechanical support, PCI was attempted in 11 patients and was successful
in only seven cases. Four patients were weaned from ECMO within 48 h, but none
survived to hospital discharge [13]. Arlt et al. reported 50 % in-hospital mortality
in a group of 14 patients who developed circulatory arrest in the cath lab during
PCI or TAVI and required emergent extracorporeal life support [21]. In the PCI
group (nine post-acute MI and one pre-transplant diagnostic cath), only four
patients survived to hospital discharge. Finally, the Cleveland Clinic retrospectively analysed 138 patients who suffered post-acute MI cardiogenic shock [26];
patients who underwent revascularisation and circulatory support including ECMO
as a bridge to cardiac transplantation experienced a significant 5-year survival
benefit.
[email protected]
13.3
155
High-Risk TAVI and ECMO Support
The introduction of TAVI has revolutionised the management of aortic valve disease
in the elderly and high-risk populations. In our recently reported experience, among
the key points necessary to obtain good results, we identified the central role of the
heart team. At the Royal Prince Alfred Hospital (The University of Sydney),
TAVIs are performed as a joint approach between cardiologists and cardiac surgeons, in close collaboration with anaesthetists and perfusionists, to facilitate
appropriate bailout options in the event of complications [3]. All patients are offered
only temporary ECMO support; long-term extracorporeal assistance in the intensive
care unit (ICU) is futile and inappropriate in this elderly and frail population.
We prefer ECMO to a full cardiopulmonary bypass (CPB) circuit for reasons of
space, reduced circuit prime, reduced activated clotting time (ACT) requirement, and
ease of use. VA ECMO has been used liberally in the elective situation, in particular for
patients with poor left or right ventricular function and in patients with incompletely
treated coronary artery disease. It has also been used emergently for acute right heart
failure, cardiac decompression to manage left ventricular apex haemorrhage, and various rescue procedures; some of these include valve-in-valve sealing of aortic root rupture, managing valve embolisation and thoracotomy for tamponade decompression.
All patients have reperfusion and rest on ECMO or are transferred to an operating
theatre for rescue surgery if that is deemed appropriate before undertaking TAVI.
In 2010, Webb reported the outcome of the Canadian experience with 345 TAVI
patients at very high or prohibitive surgical risk [27]. The 30-day mortality was
10.4 %, while a further 14 patients (4.1 %) needed haemodynamic support with
IABPs (0.9 %), or extracorporeal circulation (2.9 %), or both (0.3 %); this was due
to severe maintained hypotension or haemodynamic collapse secondary to acute
severe left ventricular dysfunction (2.9 %), ventricular apical bleeding (0.9 %), or
cardiac perforation (0.3 %). The need for intra-procedural haemodynamic support
was significantly associated with higher 30-day and late mortality.
Recently, a group from the University of Regensburg in Germany updated their
experience with emergency and prophylactic use of ECMO during TAVI [28]. Initially,
they limited the use of VA ECMO to bailouts for intraoperative complications in 8 out
of 131 cases (including ventricular perforation, cardiogenic shock, and ventricular
tachycardia). VA ECMO was then used prophylactically in nine patients who were
deemed very high risk. The median logistic EuroSCORE in this subgroup was considerably higher compared to the remaining TAVI population (30 % vs 15 %, p = 0.0003),
while in the emergency ECMO subgroup it was comparable. The use of prophylactic
VA ECMO had a significant positive impact on procedural success (p = 0.03) and
30-day mortality (p = 0.02) compared to emergency extracorporeal support.
13.4
ECMO Support for Post-infarct VSD
Post-infarct VSD (PI-VSD) is a well recognised, and now rare (0.3 %), complication of MI, with very significant morbidity and mortality [29]. Despite the high risks
associated with surgery [30] or percutaneous interventions [31], medical
[email protected]
156
F. Ramponi et al.
management on its own has an almost 100 % mortality [32]. The timing remains
controversial: a careful balance between the need for early VSD closure to avoid
haemodynamic collapse and delayed closure to allow recently infarcted myocardial
tissue to organise to enable closure needs to be respected. It is not surprising that
those patients with cardiogenic shock and multi-organ failure have the highest mortality [33]; in this subset of patients, VA ECMO can restore perfusion, provide organ
support, and allow repair in a more controlled clinical situation [34, 35].
Recently, we provided mechanical cardiopulmonary support in two very high-risk
patients with PI-VSD and cardiogenic shock (logistic EuroSCORE 80 %). The first
case involved a 60-year-old man who suffered cardiogenic shock 4 days after VSDcomplicated MI, with severe left and right ventricular failure; VA ECMO (femorofemoral) was used as a bridge to definitive surgical double-patch (Daggett) repair.
Immediately after institution of ECMO, there was normalisation of metabolic parameters and significant wean from inotropes was possible. The second patient presented
with cardiogenic shock secondary to PI-VSD who continued to deteriorate despite
maximal medical therapy (IABP and inotropes), with impending multi-organ failure
and metabolic acidosis; VA ECMO (femoro-axillary) was initiated to restore organ
perfusion and delay VSD closure. After a week on mechanical support, his clinical
condition stabilised dramatically, with resolution of renal and hepatic failure. He
underwent successful VSD repair with a pericardial patch and bioprosthetic mitral
valve replacement. Both patients required post-operative VA ECMO support, and
they were eventually decannulated, surviving to hospital discharge.
Various case reports have suggested that extracorporeal support may be an option
to allow haemodynamic stability, thus allowing a delayed closure approach. Initial
reports utilised ventricular-assist devices (VAD) to bridge patients to definitive surgical repair [36]. VAD support requires a sternotomy and involves cannula placement in stunned and recently infarcted myocardial tissue in a patient with multi-organ
failure [37]. VA ECMO is less invasive, does not require a sternotomy, and provides
cardiorespiratory support, thus allowing a period of cardiac rest and restored
organ perfusion prior to attempting surgical closure. The recent literature on VA
ECMO support for cardiogenic shock does report the use of VA ECMO in cardiogenic shock secondary to PI-VSD. However, it is not clear if ECMO was initiated
prior to operative intervention [38].
13.5
ECMO and Acute Pulmonary Embolism
Patients presenting with massive acute pulmonary embolism (PE) complicated by

right ventricular (RV) failure and cardiogenic shock have a poor prognosis [39].
Urgent thrombolysis or embolectomy is mandatory; however, these therapeutic
measures might not be initiated immediately due to haemodynamic instability or
logistic reasons. Mechanical support in these circumstances is essential as a stabilising measure or bridge to further interventions [40]. The use of right VADs has been
described, but carries the risk of even further elevation of pulmonary artery (PA)
pressures to suprasystemic values [41]. VA ECMO is an optimal strategy in RV
[email protected]
157
failure cases resulting from pressure overload secondary to pulmonary obstruction.

Misawa et al. [42] reported two cases in which percutaneous VA ECMO was used
as an adjunct to thrombolytic therapy for progressive circulatory collapse secondary
to massive acute PE. In both cases, restoration of haemodynamic stability was readily achieved once ECMO flow was initiated.
When cardiogenic shock secondary to acute RV failure increases the risks of
immediate surgery, VA ECMO is a useful bridge option. In our case, a 56-year-old
man presented with acute massive PE and paradoxical embolism, with obstruction
of the infrarenal aorta and ensuing mesenteric ischaemia [6]. After emergency
aorto-mesenteric embolectomy, peripheral VA ECMO was established and continued for 2 days as a bridge to successful PTE. A similar case was described by
Deehring et al. [43]: a 17-year-old presented with acute PE and rapidly deteriorated
with cardiogenic shock secondary to acute RV failure. Despite intubation and cardiopulmonary resuscitation, she remained unstable and required VA ECMO cannulation in the emergency department prior to transfer to a paediatric hospital where
she underwent successful PTE.
Finally, some patients will persist in refractory cardiogenic shock despite operative management. Howes et al. recently presented the case of a 38-year-old man
who experienced acute deep venous thrombosis (DVT) and PE following knee
arthroscopy [44]; despite thrombolytic therapy with alteplase, the patient remained
hypotensive, and peripheral VA ECMO was initiated and continued until restoration
of RV function. Similarly, Belohlavek et al. reported a 51-year-old man who presented with recurrent PE 1 year after previous PTE [45]; surgery was not indicated
due to severe RV dilatation and systolic dysfunction, so he underwent thrombolysis
but within minutes, he experienced ventilatory and haemodynamic collapse secondary to further embolisation. Peripheral VA ECMO was immediately started at
2.53 L/min and continued until day 4 post successful redo PTE.
13.6
Cannulation and Complications
The most common arterial access sites for peripheral ECMO are the femoral and
axillary arteries with surgical exposure, direct cannulation, and purse-string control
of the arteriotomy; the use of a Dacron interposition conduit is preferred in case of
diseased vessels. Percutaneous cannulation with vessel closure device (PerClose
Proglide, Abbott Vascular, Redwood City, CA, USA) is also safe and effective in the
right patients and experienced hands (Fig. 13.1) [4]. Axillary perfusion is preferred
in patients with severe cardiopulmonary dysfunction requiring high-flow bypass
[46]. The femoral vein is almost universally used as drainage site.
The main disadvantage of ECMO over full cardiopulmonary bypass is the lack
of myocardial protection with cardioplegia and ventricular decompression.
Moreover, in case of aortic regurgitation or bradyarrhythmia, peripheral retrograde
flow can lead to ventricular overtension.
In case of heparin-induced thrombocytopenia, anticoagulation during ECMO
can be safely achieved with the use of bivalirudin [47].
[email protected]
158
F. Ramponi et al.
Fig. 13.1 Percutaneous

femoral cannulation with
Preclose technique using
PerClose Proglide vascular
closure device (Abbott
Vascular, Redwood City, CA,
USA)
Finally, a point has to be made on safe cannulation; many patients experienced

severe and catastrophic complications from iliac rupture or uncontrolled groin
bleeding. In elective situation they are avoidable with careful preoperative workup
(arterial CT angiogram to avoid calcified and tortuous vessels). Careful haemostasis is crucial especially when ECMO support follows thrombolysis post STEMI or
PE [44, 45].
References
1. Australia and New Zealand Extracorporeal Membrane Oxygenation (ANZ ECMO) Influenza
Investigators (2009) Extracorporeal membrane oxygenation for 2009 influenza A (H1N1)
acute respiratory distress syndrome. JAMA 302(17):18881895
2. Spina R, Forrest AP, Adams MR, Wilson MK, Ng MK, Vallely MP (2010) Veno-arterial extracorporeal membrane oxygenation for high-risk cardiac catheterisation procedures. Heart Lung
Circ 19(12):736741
3. Vallely MP, Wilson MK, Adams M, Ng MKC (2012) How to set up a successful TAVI program. Ann Cardiothorac Surg 1(2):185189
4. Ramponi F, Yan TD, Vallely MP, Wilson MK (2011) Total percutaneous cardiopulmonary
bypass with Perclose ProGlide. Interact Cardiovasc Thorac Surg 13(1):8688
5. Forrest P, Cheong JY, Vallely MP, Torzillo PJ, Hendel PN, Wilson MK, Bannontt PG, Bayfield
MS, Herkes R, Walker SW (2011) International retrieval of adults on extracorporeal membrane oxygenation support. Anaesth Intensive Care 39(6):10821085
6. Ramponi F, Wilson MK, Vedelago J, Bayfield MS (2011) Catastrophic pulmonary and paradoxical embolism. ANZ J Surg 81(11):843844
7. Vogel RA (1988) The Maryland experience: angioplasty and valvuloplasty using percutaneous
cardiopulmonary support. Am J Cardiol 62(18):11K14K
8. Taub JO, LHommedieu BD, Raithel SC, Vieth DG, Vieth PJ, Barner HB, Vandormael M,
Pennington DG (1989) Extracorporeal membrane oxygenation for percutaneous coronary
angioplasty in high risk patients. ASAIO Trans 35(3):664666
9. Ott RA, Mills TC, Tobis JM, Allen BJ, Dwyer ML (1990) ECMO assisted angioplasty for
cardiomyopathy patients with unstable angina. ASAIO Trans 36(3):M483M485
10. Magovern GJ Jr, Simpson KA (1999) Extracorporeal membrane oxygenation for adult cardiac
support: the Allegheny experience. Ann Thorac Surg 68(2):655661
11. Silvestri M, Barragan P, Sainsous J, Bayet G, Simeoni JB, Roquebert PO, Macaluso G, Bouvier
JL, Comet B (2000) Unprotected left main coronary artery stenting: immediate and mediumterm outcomes of 140 elective procedures. J Am Coll Cardiol 35(6):15431550
[email protected]
159
12. Irons D, Lim YL, Lefkovits J, Esmore D (1996) Left main coronary artery stenting under
extracorporeal circulatory support. Aust N Z J Med 26(6):842843
13. Shammas NW, Roberts S, Early G (2002) Extracorporeal membrane oxygenation for unprotected left main stenting in a patient with totally occluded right coronary artery and severe left
ventricular dysfunction. J Invasive Cardiol 14(12):756759
14. Hasdai D, Topol EJ, Califf RM, Berger PB, Holmes DR Jr (2000) Cardiogenic shock complicating acute coronary syndromes. Lancet 356(9231):749756
15. Zeymer U, Vogt A, Zahn R, Weber MA, Tebbe U, Gottwik M, Bonzel T, Senges J, Neuhaus
KL (2004) Predictors of in-hospital mortality in 1333 patients with acute myocardial infarction
complicated by cardiogenic shock treated with primary percutaneous coronary intervention
(PCI); Results of the primary PCI registry of the Arbeitsgemeinschaft Leitende Kardiologische
Krankenhausrzte (ALKK). Eur Heart J 25(4):322328
16. Hochman JS, Sleeper LA, Webb JG, Sanborn TA, White HD, Talley JD, Buller CE, Jacobs
AK, Slater JN, Col J, McKinlay SM, LeJemtel TH (1999) Early revascularization in acute
myocardial infarction complicated by cardiogenic shock. SHOCK Investigators. Should We
Emergently Revascularize Occluded Coronaries for Cardiogenic Shock. N Engl J Med
341(9):625634
17. Sarkar K, Kini AS (2010) Percutaneous left ventricular support devices. Cardiol Clin
28(1):169184
18. Sjauw KD, Engstrm AE, Vis MM, van der Schaaf RJ, Baan J Jr, Koch KT, de Winter RJ, Piek
JJ, Tijssen JG, Henriques JP (2009) A systematic review and meta-analysis of intra-aortic balloon pump therapy in ST-elevation myocardial infarction: should we change the guidelines?
Eur Heart J 30(4):459468
19. Anderson H 3rd, Steimle C, Shapiro M, Delius R, Chapman R, Hirschl R, Bartlett R (1993)
Extracorporeal life support for adult cardiorespiratory failure. Surgery 114(2):161172
20. Chen JS, Ko WJ, Yu HY, Lai LP, Huang SC, Chi NH, Tsai CH, Wang SS, Lin FY, Chen YS
(2006) Analysis of the outcome for patients experiencing myocardial infarction and cardiopulmonary resuscitation refractory to conventional therapies necessitating extracorporeal life support rescue. Crit Care Med 34(4):950957
21. Arlt M, Philipp A, Voelkel S, Schopka S, Husser O, Hengstenberg C, Schmid C, Hilker M
(2012) Early experiences with miniaturized extracorporeal life-support in the catheterization
laboratory. Eur J Cardiothorac Surg 42(5):858863
22. Lai CH, Chu YS, Li WL, Wang CC, Chang Y (2008) Percutaneous coronary intervention
under extracorporeal membrane oxygenation support for high-risk acute myocardial infarction
with cardiogenic shock. J Med Sci 28(1):3944
23. Larsen AI, Hjrnevik A, Bonarjee V, Barvik S, Melberg T, Nilsen DW (2010) Coronary blood
flow and perfusion pressure during coronary angiography in patients with ongoing mechanical
chest compression: a report on 6 cases. Resuscitation 81(4):493497
24. Wind J, Bekkers SC, van Hooren LJ, van Heurn LW (2009) Extensive injury after use of a
mechanical cardiopulmonary resuscitation device. Am J Emerg Med 27(8):1017.e12
25. Lee MS, Pessegueiro A, Tobis J (2008) The role of extracorporeal membrane oxygenation in
emergent percutaneous coronary intervention for myocardial infarction complicated by cardiogenic shock and cardiac arrest. J Invasive Cardiol 20(9):E269E272
26. Tayara W, Starling RC, Yamani MH, Wazni O, Jubran F, Smedira N (2006) Improved survival
after acute myocardial infarction complicated by cardiogenic shock with circulatory support
and transplantation: comparing aggressive intervention with conservative treatment. J Heart
Lung Transplant 25(5):504509
27. Rods-Cabau J, Webb JG, Cheung A, Ye J, Dumont E, Feindel CM, Osten M, Natarajan MK,
Velianou JL, Martucci G, DeVarennes B, Chisholm R, Peterson MD, Lichtenstein SV,
Nietlispach F, Doyle D, DeLarochellire R, Teoh K, Chu V, Dancea A, Lachapelle K, Cheema
A, Latter D, Horlick E (2010) Transcatheter aortic valve implantation for the treatment of
severe symptomatic aortic stenosis in patients at very high or prohibitive surgical risk: acute
and late outcomes of the multicenter Canadian experience. J Am Coll Cardiol 55(11):
10801090
[email protected]
160
F. Ramponi et al.
28. Husser O, Holzamer A, Philipp A, Nunez J, Bodi V, Mller T, Lubnow M, Luchner A, Lunz
D, Riegger GA, Schmid C, Hengstenberg C, Hilker M (2013) Emergency and prophylactic use
of miniaturized veno-arterial extracorporeal membrane oxygenation in transcatheter aortic
valve implantation. Catheter Cardiovasc Interv 82(4):E542E551
29. Jeppsson A, Liden H, Johnsson P, Hartford M, Rdegran K (2005) Surgical repair of post
infarction ventricular septal defects: a national experience. Eur J Cardiothorac Surg
27(2):216221
30. Fukushima S, Tesar PJ, Jalali H, Clarke AJ, Sharma H, Choudhary J, Bartlett H, Pohlner PG
(2010) Determinants of in-hospital and long-term surgical outcomes after repair of postinfarction ventricular septal rupture. J Thorac Cardiovasc Surg 140(1):5965
31. Thiele H, Kaulfersch C, Daehnert I, Schoenauer M, Eitel I, Borger M, Schuler G (2009)
Immediate primary transcatheter closure of postinfarction ventricular septal defects. Eur Heart
J 30(1):8188
32. Crenshaw BS, Granger CB, Birnbaum Y, Pieper KS, Morris DC, Kleiman NS, Vahanian A,
Califf RM, Topol EJ (2000) Risk factors, angiographic patterns, and outcomes in patients with
ventricular septal defect complicating acute myocardial infarction. GUSTO-I (Global
Utilization of Streptokinase and TPA for Occluded Coronary Arteries) Trial Investigators.
Circulation 101(1):2732
33. Menon V, Webb JG, Hillis LD, Sleeper LA, Abboud R, Dzavik V, Slater JN, Forman R,
Monrad ES, Talley JD, Hochman JS (2000) Outcome and profile of ventricular septal rupture
with cardiogenic shock after myocardial infarction: a report from the SHOCK Trial Registry.
J Am Coll Cardiol 36:11101116
34. Tsai M, Wu H, Chan S, Luo C (2012) Extracorporeal membrane oxygenation as a bridge to
definite surgery in recurrent postinfarction ventricular septal defect. ASAIO J 58:8889
35. Rohn V, Spacek M, Belohlavek J, Tosovsky J (2009) Cardiogenic shock in patient with posterior postinfarction septal rupture successful treatment with extracorporeal membrane oxygenation (ECMO) as a ventricular assist device. J Card Surg 24:435436
36. Pitsis AA, Kelpis TG, Visouli AN, Bobotis G, Filippatos GS, Kremastinos DT (2008) Left
ventricular assist device as a bridge to surgery in postinfarction ventricular septal defect. J
Thorac Cardiovasc Surg 135:951952
37. Alba AC, Rao V, Ivanov J, Ross HJ, Delgado DH (2009) Usefulness of the INTERMACS scale
to predict outcomes after mechanical assist device implantation. J Heart Lung Transplant
28:827833
38. Formica F, Avalli L, Colagrande L, Ferro O, Greco G, Maggioni E, Paolini G (2010)
Extracorporeal membrane oxygenation to support adults with cardiac failure: predictive factors of 30 day mortality. Interact Cardiovasc Thorac Surg 10:721726
39. Torbicki A, Perrier A, Konstantinides S, Agnelli G, Gali N, Pruszczyk P, Bengel F, Brady AJ,
Ferreira D, Janssens U, Klepetko W, Mayer E, Remy-Jardin M, Bassand JP, ESC Committee
for Practice Guidelines (CPG) (2008) Guidelines on the diagnosis and management of acute
pulmonary embolism: the Task Force for the Diagnosis and Management of Acute Pulmonary
Embolism of the European Society of Cardiology (ESC). Eur Heart J 29(18):22762315
40. Norita H, Ohteki H, Hisanou R (1994) Emergency pulmonary embolectomy for massive pulmonary embolism. J Jpn Coll Angiol 34:39
41. Rajdev S, Benza R, Misra V (2007) Use of Tandem Heart as a temporary hemodynamic support option for severe pulmonary artery hypertension complicated by cardiogenic shock. J
Invasive Cardiol 19(8):E226E229
42. Misawa Y, Fuse K, Yamaguchi T, Saito T, Konishi H (2000) Mechanical circulatory assist for
pulmonary embolism. Perfusion 15(6):527529
43. Deehring R, Kiss AB, Garrett A, Hillier AG (2006) Extracorporeal membrane oxygenation as
a bridge to surgical embolectomy in acute fulminant pulmonary embolism. Am J Emerg Med
24(7):879880
44. Howes J, Khilkin M, DeRose J, Dicpinigaitis P, Dulu A (2011) Veno-arterial extracorporeal
membrane oxygenation as a salvage therapy in massive pulmonary embolism. Chest 140(4_
MeetingAbstracts):64A. doi:10.1378/chest.1117995
[email protected]
161
45. Belohlavek J, Rohn V, Jansa P, Tosovsky J, Kunstyr J, Semrad M, Horak J, Lips M, Mlejnsky
F, Balik M, Klein A, Linhart A, Lindner J (2010) Veno-arterial ECMO in severe acute right
ventricular failure with pulmonary obstructive hemodynamic pattern. J Invasive Cardiol
22(8):365369
46. Miyamoto S, Hadama T, Mori Y et al (1995) Hemodynamic profiles during concurrent
intraaortic balloon pumping and venoarterial bypass a canine study comparing subclavian
and femoral artery perfusion sites. Jpn Circ J 59:693703
47. Koster A, Weng Y, Bttcher W, Gromann T, Kuppe H, Hetzer R (2007) Successful use of
bivalirudin as anticoagulant for ECMO in a patient with acute HIT. Ann Thorac Surg
83(5):18651867
[email protected]
ECMO for Severe Accidental

Hypothermia
14
Peter Mair and Elfriede Ruttmann
In general, emergency mechanical circulatory support for the treatment of prolonged, intractable cardiorespiratory arrest is assigned only a low-grade recommendation in current resuscitation guidelines [1]. However, for cardiorespiratory arrest
associated with severe accidental hypothermia, emergency mechanical circulatory
support and extracorporeal rewarming are widely recommended and considered the
gold standard of treatment [13]. Increasingly, venoarterial ECMO systems are
being used for extracorporeal support in hypothermic patients, as ECMO not only
offers significant advantages as compared to standard cardiopulmonary bypass
(CPB) technology but may also be associated with improved survival [4, 5].
14.1
Accidental Hypothermia and Hypothermic

Cardiorespiratory Arrest
Accidental hypothermia is defined as an involuntary drop in body core temperature

below 35 C [1, 2]. Respiratory and cardiocirculatory functions are not significantly
impaired as long as core temperature does not drop below 32 C [2, 6]. When body
core temperature falls below 32 C to 30 C, patients present with impaired consciousness, reduced respiratory rate, bradycardia, and hypotension [2]. With further
cooling deep coma with fixed, dilated pupils develops together with very slow gasping respiration and reduced cardiac output, making detection of vital signs difficult
[6]. Reduced cardiorespiratory function per se does not put the patient at risk for
P. Mair, MD (*)
Department of Anaesthesiology and Critical Care Medicine,
Innsbruck Medical University, Anichstrasse 35, 6020 Innsbruck, Tyrol, Austria
E. Ruttmann, MD
Department of Cardiac Surgery, Innsbruck Medical University,
Anichstrasse 35, 6020 Innsbruck, Tyrol, Austria
[email protected]
163
164
P. Mair and E. Ruttmann
irreversible injury, but instead is a physiological response by the human body to the
reduced oxygen demands during hypothermia [6]. If core temperature falls below
20 C, respiratory arrest and asystole will eventually occur [6].
The heart of a patient with severe accidental hypothermia and a core temperature
below 30 C is not only bradycardic but also highly irritable and prone to arrhythmias [1, 2, 6]. At a core temperature below 30 C, minimal alterations in the heart
or minor movements of the patient can result in ventricular fibrillation, the so-called
hypothermic sudden cardiac death phenomenon [1, 2, 6]. The fibrillating hypothermic heart in general does not respond to electrical or pharmacological therapy
unless rewarmed [2, 3]. On the other hand, hypothermia offers significant protection
from ischemic brain injury and enables full neurological recovery, even after resuscitation efforts lasting several hours [2, 3, 6]. Consequently, it is difficult to diagnose irreversible cardiorespiratory arrest during hypothermia. That being the case,
death in a hypothermic patient should be defined as failure to revive with rewarming, and it is widely accepted that nobody is dead unless warm and dead [6].
Asystole in arrested hypothermic patients indicates very low body core temperature, prolonged cardiac arrest, or concomitant asphyxia, and ventricular fibrillation
indicates cardiac arrest due to arrhythmias induced during rescue or initial treatment. Consequently, hypothermic cardiac arrest associated with ventricular fibrillation has a far better prognosis than does hypothermic cardiac arrest in an asystolic
patient [2, 3].
14.2
Indications for ECMO in Patients with Severe

Accidental Hypothermia
ECMO is associated with a significant risk for major complications. Most patients
with severe accidental hypothermia and a core temperature below 32 C can be successfully treated with noninvasive external rewarming or minimally invasive techniques of internal rewarming [2, 6]. Consequently, most clinicians do not use
ECMO, even in a profoundly hypothermic patient, as long as the patient can be
stabilized with medical therapy. ECMO support is usually restricted to patients with
severe hemodynamic compromise or cardiorespiratory arrest [13] (Fig. 14.1). It
has been suggested that also some non-arrested hypothermic patients (e.g., those
with a history of asphyxia or near-drowning) may benefit from more aggressive use
of ECMO support instead of medical therapy [5, 7].
ECMO is available in only a few specialized centers. Many cases of accidental
hypothermia, however, occur in remote areas, and long transfer times are necessary
to provide ECMO support. For this reason, alternative therapeutic approaches have
been used that combine prolonged external chest compression and alternative
rewarming techniques (e.g., hemofiltration, hemodialysis, peritoneal lavage, thoracic lavage) [2, 3, 6, 8]. A technique widely available, even in small hospitals, is
closed chest thoracic lavage [3, 8]. Although alternatives may be successful in some
cases, extracorporeal support is associated with higher survival rates [2]. Whether
transfer to an ECMO center is the best approach must be decided on a case-by-case
[email protected]
14 ECMO for Severe Accidental Hypothermia
165
Accidental hypothermia (core temperature 30 C)
Non-arrested
Witnessed arrest or ventricular fibrillation
Unwitnessed arrest
Systolic pressure 60 mmHG?
Plasma potassium 1012 mmoL/L?
Severe arrhythmias?
No
Medical therapy,
active rewarming
Yes
No
Yes
Systolic pressure 90 mmHG?

Yes
No
Consider ECMO support*
Stop CPR
ECMO support**
*Use in selected arrested patients, may improve outcome in non-arrested patients, ** Therapy of choice, use if available
Fig. 14.1 Indications for ECMO support in severe accidental hypothermia
basis, depending on transfer time, mode of transport available, experience of the

attending physicians, and rewarming techniques available at the referring hospital.
Outcome of unwitnessed hypothermic cardiorespiratory arrest is poor as in many
cases cardiac arrest precedes cooling, and there is insufficient protection from ischemic tissue injury. This is particularly true for hypothermic cardiorespiratory arrest
associated with avalanche accidents and near-drowning [3, 9]. Insufficient protection from ischemia results in cell autolysis, which can be detected by high plasma
potassium levels in some of these patients [2, 3]. Consequently, a plasma potassium
level exceeding 1012 mmol/L is considered the reason to withhold ECMO support
in an asystolic hypothermic patient with unwitnessed cardiorespiratory arrest [2, 3]
(Fig. 14.1). If plasma potassium is used to support the decision to terminate resuscitation, alternative reasons for severe hyperkalemia like extensive local freezing
injuries must always be excluded [2, 3].
14.3
Advantages of ECMO as Compared to Standard

CPB Technology in the Treatment of Severe
In many patients rewarmed primarily with CPB, ECMO later becomes necessary
because of the inability to wean the patient from CPB due to intractable cardiorespiratory failure [1013]. Furthermore, ECMO has significant advantages over standard CPB technology in emergency situations (Table 14.1). Thus, in an increasing
number of hospitals, venoarterial ECMO has become the method of choice for
[email protected]
166
Table 14.1 Advantages of
ECMO as compared to
cardiopulmonary bypass in
patients with severe
accidental hypothermia
Provides more rapid extracorporeal support, thanks to its

portable systems and short setup time
Cannulation and support are possible outside the OR
Lower levels of anticoagulation are needed
No systemic anticoagulation needed in case of major bleeding
Prolonged extracorporeal support is possible for hours and days
Easy patient transfer within hospital
Minimally invasive using percutaneous cannulation techniques
Negative pressure applied to venous cannulas enables higher
extracorporeal flow rates
emergency extracorporeal support in hypothermic patients [4, 5, 13]. In a retrospective study of 59 hypothermic patients using multivariate logistic regression analysis,
ECMO resuscitation was associated with improved survival as compared to standard CPB resuscitation [4]. The key factor responsible for improved survival was
the routine use of prolonged cardiorespiratory support for 2448 h in the ECMO
group, thus preventing early mortality from respiratory insufficiency which is
responsible for 64 % of fatalities after CPB resuscitation [4].
14.4
Clinical Experience with and Outcome Following ECMO

in Severe Accidental Hypothermia
ECMO techniques (Table 14.2) have been used with success over the whole range of
underlying pathologies associated with severe accidental hypothermia, including
near-drowning [13], avalanche burial [4, 10], urban hypothermia [14], and multisystem trauma [15, 16]. Survival rates in published case series vary over a wide range and
depend predominantly on the underlying pathology and preexisting co-morbidities [2,
3]. Urban hypothermia and hypothermia associated with avalanche accidents consistently produce poor survival rates [3, 6, 14], whereas hypothermia after prolonged
exposure to cold in healthy individuals suffering from intoxication or wilderness accidents is associated with survival rates of 7090 % in arrested patients [17]. Venoarterial
and venovenous ECMO support have been used after CPB rewarming when patients
cannot be weaned from extracorporeal support in the operating room because of
intractable respiratory or cardiorespiratory failure [1013, 18] (Table 14.2).
Increasingly, however, venoarterial ECMO support with femorofemoral cannulation
is used as the primary therapeutic intervention [4, 5]. Percutaneous femoral cannulation techniques have been used with high success rates and further reduce invasiveness [5]. ECMO is regularly implanted outside the operating theater, and transfer of
the ECMO team to an outside hospital to implant an ECMO system in patients with
hypothermic cardiorespiratory arrest has been reported [13, 19]. Even initiation of
ECMO resuscitation at the scene may be a therapeutic option in the near future [20].
In most centers ECMO is used only in hypothermic patients with cardiorespiratory arrest. Based on their experience with 69 profoundly hypothermic patients,
[email protected]

Table 14.2 Clinical
experience with ECMO in
patients with severe
accidental hypothermia
167
Venoarterial ECMO
Emergency mechanical circulatory support in arrested
hypothermic patients
Immediate restoration of systemic blood flow before
rewarming
Extracorporeal rewarming in profoundly hypothermic
non-arrested patients
Rapid rewarming with cardiorespiratory support
Extracorporeal lung or heart/lung support after
cardiopulmonary bypass rewarming
Inability to wean patient from bypass because of lung or
heart/lung failure
Venovenous ECMO
Lung replacement therapy after extracorporeal rewarming
Upper body hypoxemia during prolonged femoral
venoarterial ECMO
Lung failure after cardiopulmonary bypass rewarming
Extracorporeal rewarming in arrested hypothermic patients
during ongoing CPR
Rapid rewarming in patients with no peripheral arterial
access
Morita and coworkers suggested that ECMO may improve survival also in nonarrested patients [5].
Multiorgan failure, prolonged stay in the intensive care unit for weeks, and full
neurological recovery after several months in a rehabilitation unit are regularly
observed in hypothermic arrest victims after initial, successful resuscitation [10, 11,
21, 22]. Therefore, one should always be cautious not to terminate maximum therapy too early after ECMO resuscitation when the clinical course is complicated in a
hypothermic patient.
14.5
Practical Issues of ECMO Support in Severe

Almost no scientifically valid data are available on how to manage a hypothermic

patient on ECMO support. Consequently, perfusion protocols vary over a wide
range among different institutions. As an example, the perfusion protocol used at
Innsbruck University Hospital is shown in Table 14.3. Many institutions support
patients with high flow rates in the range 2.53 L/m2/min also during hypothermia
to compensate for a preexisting oxygen debt, although the optimal flow rate is in
fact unknown. High rewarming rates are regularly used, at least until successful
defibrillation of the heart. This enables early restoration of pulsatile flow and
improved left ventricular unloading. On the other hand, it has been repeatedly stated
that rapid rewarming may put the hypoxic, cold brain at risk for additional injury
[2, 3]. Vasopressors to maintain mean arterial pressure above 50 mmHg, sodium
[email protected]
168
Table 14.3 ECMO perfusion protocol* for patients with accidental hypothermia
Cannulation
Peripheral cannulation of femoral vessels using Seldinger technique
Percutaneous cannulation whenever possible
Use Seldinger technique after surgical cutdown and vessel exposure, if necessary
Always insert cannula for distal leg perfusion
Extracorporeal flow rates
High flow rates also during hypothermia
Anticoagulation
5080 U/kg heparin before cannulation (no heparin in case of major bleeding)
Continuous heparin infusion to maintain ACT 150250 s during ECMO support
Reduce (significant bleeding) or stop (life-threatening bleeding) heparin, if necessary
Temperature management
Start rewarming only after a period of hypothermic reperfusion to allow correction of abnormal
blood values and insertion of additional lines
Rewarm no faster than 46 C/h
Stop rewarming at 32 C and maintain therapeutic hypothermia for at least 24 h
Defibrillate as soon as possible
Early restoration of pulsatile flow and improved left ventricular unloading
Weaning from ECMO
Do not wean immediately after rewarming
If possible, wean with moderate circulatory and ventilatory support after 1224 h
Monitor blood gases and pulse oximetry on the right hand (upper body hypoxemia)
Transesophageal Echo Monitoring
Position of guidewire in arterial (descending aorta) and venous (right atrium) circulation during
cannulation, reduced risk of major vessel injury during cannulation
Control and optimize venous cannula position during ECMO support
Monitor left ventricular unloading before defibrillation and during ECMO support
Monitor right atrial filling to optimize volume replacement during ECMO support
Near infrared spectroscopy monitoring (bilateral cerebral, both forelegs)
Monitor oxygenation and cerebral perfusion during CPR and extracorporeal rewarming
Detect upper body hypoxemia early during prolonged support
Monitor for ischemia in the cannulated leg during prolonged support
*
Protocol used at Innsbruck University Hospital, Austria
bicarbonate to correct severe metabolic acidosis, and catecholamines to support left

ventricular unloading and restore pulsatile flow are used, although their actual value
has not been proven to date. Acid base management is normally done using an alpha
stat regime. A systemic heparin dose lower than for standard CPB is almost always
used, and the option of avoiding any systemic anticoagulation by using heparincoated ECMO systems has been repeatedly mentioned [3, 4]. Nowadays, rewarming is normally discontinued at 3234 C in patients with a history of cardiorespiratory
arrest and therapeutic hypothermia maintained for additional 1224 h [23]. This
approach is based on the extrapolation of data obtained during normothermic cardiorespiratory arrest. In some institutions prolonged ECMO support after initial
resuscitation and rewarming is routine practice [4]. When prolonged femorofemoral
ECMO support is used, upper body hypoxemia can occur in patients in whom
recovery of myocardial function is more pronounced than is recovery of pulmonary
[email protected]
169
function (Table 14.3). This phenomenon is caused by poorly oxygenated blood that
is ejected by the heart into the proximal parts of the aorta, whereas distal parts of the
aorta receive well-oxygenated blood from the ECMO system. Therefore, oxygenation of the patient on prolonged femoral venoarterial ECMO must always be monitored on the right hand to detect this problem early (Table 14.3). If hypoxemia in the
proximal aorta does not respond to alterations in mechanical ventilation, the patient
must be switched to a venovenous ECMO system. Leg ischemia is a major complication of femorofemoral ECMO repeatedly reported [24]. Incidence of leg ischemia
can be decisively reduced by inserting a separate cannula for leg perfusion. As kinking and dislocation of the cannula are possible in patients with distal leg perfusion,
in particular during prolonged ECMO support, additional near-infrared spectroscopy monitoring of foreleg muscle perfusion is advisable. An organized, preemptive approach and standardized treatment protocols (Table 14.3) are used at several
institutions [4, 5, 25] and are a good option for improving patient care.
14.6
Summary
Extracorporeal life support with ECMO is an increasingly popular alternative to

cardiopulmonary bypass rewarming in hypothermic patients, as ECMO may
improve survival and justify wider use of extracorporeal support, also in nonarrested patients.
References
1. Soar J, Perkins GD, Abbas G et al (2010) European Resuscitation Council Guidelines for
Resuscitation 2010 Section 8. Cardiac arrest in special circumstances. Resuscitation
81:140014332
2. Brown DJA, Brugger H, Boyd J et al (2012) Accidental hypothermia. N Engl J Med
367:19301938
3. Mair P, Schwarz B, Walpoth B, Silfvast T (2007) Cardiopulmonary resuscitation in hypothermic patients. In: Paradise NA, Halperin HR, Kern KB, Wenzel V, Chamberlain DA (eds)
Cardiac arrest. Cambridge University Press, Cambridge/New York/Melbourne, pp 10141027
4. Ruttmann E, Weissenbacher A, Ulmer H et al (2007) Prolonged extracorporeal membrane
oxygenation-assisted support provides improved survival in hypothermic patients with cardiocirculatory arrest. J Thorac Cardiovasc Surg 134:594600
5. Morita S, Inokuchi S, Yamagiwa T et al (2011) Efficacy of portable cardiopulmonary bypass
rewarming versus that of conventional internal rewarming for patients with accidental deep
hypothermia. Crit Care Med 39:10641068
6. Lloyd EL (1996) Accidental hypothermia. Resuscitation 32:111134
7. Kornberger E, Schwarz B, Lindner KH et al (1999) Forced air surface rewarming in patients
with severe accidental hypothermia. Resuscitation 41:105111
8. Plaisier BR (2005) Thoracic lavage in accidental hypothermia with cardiac arrest report of a
case and review of the literature. Resuscitation 66:99104
9. Brugger H, Durrer B, Adler-Kastner L (1996) On-site triage of avalanche victims with asystole
by the emergency doctor. Resuscitation 31:1116
10. Gilbert M, Busund R, Skagseth A et al (2000) Resuscitation from accidental hypothermia of
13.7 C with circulatory arrest. Lancet 355:375376
[email protected]
170
11. Eich C, Braeuer A, Kettler D (2005) Recovery of a hypothermic drowned child after resuscitation with cardiopulmonary bypass followed by prolonged extracorporeal membrane oxygenation. Resuscitation 67:145148
12. Coskun KO, Popov AF, Schmitto JD et al (2010) Extracorporeal circulation for rewarming in
drowning and near-drowning pediatric patients. Artif Organs 34:10261030
13. Wanscher M, Agersnap L, Ravn J et al (2012) Outcome of accidental hypothermia with or
without circulatory arrest Experience from the Danish Praesto Fjord boating accident.
Resuscitation 83:10781084
14. Sansone F, Flocco R, Zingarelli F et al (2011) Hypothermic cardiac arrest in the homeless:
what can we do? J Extra Corpor Technol 43:252257
15. Ruenitz K, Thornberg K, Wanscher M (2009) Resuscitation of severely hypothermic and multitraumatised female following long-term cardiac arrest. Ugeskr Laeger 171:328329
16. Firstenberg MS, Nelson K, Abel E et al (2012) Extracorporeal membrane oxygenation for
complex multiorgan system trauma. Case Rep Surg. doi:10.1155/2012/897184
17. Hohlrieder M, Kroesslhuber F, Voelckel W et al (2010) Experience with helicopter rescue missions for crevasse accidents. High Alt Med Biol 11:375379
18. Tiruvoipati R, Balasubramanian SK, Khoshbin E et al (2005) Successful use of venovenous
extracorporeal membrane oxygenation in accidental hypothermic cardiac arrest. ASAIO
51:474476
19. Kumle B, Doering B, Mertes H et al (1997) Resuscitation of a near-drowning patient by the
use of a portable extracorporeal circulation device. Anaesthesiol Intensivmed Notfallmed
Schmerzther 32:754756
20. Lebreton G, Pozzi M, Luyt CE et al (2011) Out-of-hospital extra-corporeal life support
implantation during refractory cardiac arrest in a half-marathon runner. Resuscitation
82:12391242
21. Thalmann M, Trampitsch E, Haberfellner N et al (2001) Resuscitation in near drowning with
extracorporeal membrane oxygenation. Ann Thorac Surg 72:607608
22. Walpoth BH, Walpoth-Aslan BN, Mattle HP et al (1997) Outcome of survivors of accidental
deep hypothermia and circulatory arrest treated with extracorporeal blood rewarming. N Engl
J Med 337:15001505
23. Hypothermia after Cardiac Arrest Study Group (2002) Mild therapeutic hypothermia to
improve neurological outcome after cardiac arrest. N Engl J Med 346:549556
24. Marasco SF, Lukas G, McDonald M et al (2008) Review of ECMO support in critically ill
adult patients. Heart Lung Cir 17:S41S47
25. Scaife ER, Connors RC, Morris SE et al (2007) An established extracorporeal membrane
oxygenation protocol promotes survival in extreme hypothermia. J Pediatric Surg 42:
20122016
[email protected]
ECMO in Drug Intoxication
15
Piergiorgio Bruno, Piero Farina, and Massimo Massetti
Drug intoxication (from abuse or from reactions not directly related to the dosage)
may be associated with various clinical scenarios, sometimes primarily affecting the
cardiovascular system and leading to such severe forms to provoke death. Indeed, if
drug poisoning in general carries a low mortality rate (about 1 % in the adult), mortality appears much higher if cardiotoxic drugs are involved [1]. In a report of the
American Association of Poison Control Centers, among 847,483 poisonings in
adults over 19 years of age, cardiotoxic drugs were involved in 5.8 % but accounted
for about 19 % of the total 1,261 poisoning fatalities [2]; it was also found that,
though accounting for approximately 40 % of cardiovascular drug poisonings, calcium channel blockers and beta-blockers represented more than 65 % of deaths
from cardiovascular medications [3]. Medical treatment for drug intoxication relies
mostly on supportive measures or, in selected cases, on the use of antidotes. It is
generally effective and well codified and appears to have reached high standards:
further improvements might be unlikely. Nevertheless, the most severe forms of
drug intoxication still carry a high mortality rate. The depression of cardiac function
in drug intoxication is usually temporary and reversible: mechanical support to circulation can prevent death while waiting for the heart to recover.
15.1
Commonly Involved Drugs and Epidemiology
Most of the cardiotoxic drugs are membrane stabilizing agents (MSA). Already in
the 1980s, Henry and Cassidy [1] showed that, for any pharmacological class of
drugs, the mortality rate from poisoning is significantly increased if the involved
drugs possess a MS effect in addition to their main pharmacological activity.
Membrane-stabilizing effect consists in the inhibition or total abolishing of action
potentials from being propagated across the cell membranes. Substances with such
P. Bruno (*) P. Farina M. Massetti
Cardiac Surgery, Universitary Polyclinic A. Gemelli Catholic University, Rome, Italy
[email protected]
171
172
P. Bruno et al.
effect interact with phospholipids in the cellular membranes, closing the sodium
channels and therefore preventing the cell depolarization (action potential phase 0).
For all the excitable cells (smooth or striated muscle cells, neurons, heart conduction
system cells), this translates into (a) increased excitability threshold and (b) reduced
conduction and automaticity. Cardiovascular, respiratory, and nervous system are
mostly affected. Drugs with MS effect are Vaughan Williams class I antiarrhythmics,
beta-blockers, antimalarial drugs, tricyclic antidepressants, phenothiazine, and
cocaine. Other cardiotoxic drugs (without MS effect) are digoxin and calcium channel blockers (particularly verapamil). MSAs, digoxin, and calcium channel blockers
are commonly prescribed drugs (apart from cocaine, which though illegal is a
popular recreational drug), and this explains the prevalence of their abuse, which is
almost always intentional (aiming to suicide or toxicomania related). Notably, when
used for suicidal purposes, poisoning relies on the intake of multiple MSA.
15.2
Clinical Findings
Besides symptoms and signs that are typical for each class of substances, all the
cardiotoxic drugs (MSA especially) lead to a common clinical scenario ruled by
respiratory depression, metabolic disturbances, and obviously cardiovascular
effects. The most typical presentation of severe cardiotoxicity consists in hypotension and/or severe cardiogenic shock. The electrophysiological alterations induced
by the MSA lead to alterations on the electrocardiogram, typically enlarged QRS
complexes and prolonged QT interval. In the most severe forms, disturbances of the
atrioventricular conduction and ventricular arrhythmias (from tachycardia to fibrillation) can appear.
The delay in onset of life-threatening events depends on the toxicant and its
galenic formulation, the ingested dose, the duration of QRS length on echocardiogram for the MSA, and the occurrence of mixed cardiotropic poisonings [4]. The
delay is up to two hours after ingestion for class I antiarrhythmics [5] and of about
6 h for polycyclic antidepressants [6], chloroquine [7], and beta-blockers [8]. As
reported by Baud et al. in an interesting review [4], in a non-negligible portion of
cases, drug-induced cardiovascular shock does not result from a decreased cardiac
contractility, but rather from a combination between relative hypovolemia and arterial vasodilation. This is well recognized for calcium channel blockers [2], less
known for polycyclic antidepressants and chloroquine, and can be underestimated
for labetalol poisoning. Therefore, in drug-induced cardiovascular shock with apparent refractoriness to conventional treatment, it is mandatory to perform a hemodynamic examination (using either right heart catheterization or echocardiography) to
assess the mechanisms of shock before considering indication to mechanical support. In the same article by Baud, it was reported that, in 137 consecutive cases of
severe MSA poisoning, survival rate for medically treated patients (catecholamines
support in addition to specific treatments) was 72 %. Once again, this confirms that
conventional therapy and pharmacological inotropic support are effective in most
cases, while mechanical support must be restricted to the most severe forms only.
[email protected]
15
173
When treatment fails, death is usually related to ventricular arrhythmias, electromechanical dissociation, asystolia (usually preceded by other disturbances of the
cardiac rhythm not responding to medical therapy), refractory cardiogenic shock, or
cerebral death (the latter being common in patients who were found by rescuers in
cardiocirculatory arrest).
15.3
Mechanical Circulatory Support for Drug Intoxication
Already in 2001, long before the resurrected interest in extracorporeal membrane

oxygenation (ECMO) in many intensive care units around the world (fueled by the
surprisingly good outcomes for its support in the 2009 A/H1N1 pandemic influenza), the toxicologic-oriented advanced cardiac life support (TOX-ACLS) guidelines stated that evidence supported the use of circulatory assist devices such as
intra-aortic balloon pumps (IABPs) and emergency cardiopulmonary bypass (CPB)
in the management of drug-induced cardiovascular shock refractory to maximal
therapy [9].
Evidence comes from experimental studies and small clinical series. Five experimental studies with control group have been published [1014]. In all of these studies, cardiogenic shock from MSA intoxication was induced. Drugs involved were
lidocaine in 6 dogs [10], amitriptyline in 6 dogs [11] and 9 swines [12], desipramine
in 6 dogs [13], and propranolol plus procainamide in 17 dogs [14]. The animals
were assigned either to maximal supportive measures (i.e., correction of acid-base
disturbances, fluid resuscitation, antiarrhythmics, inotropic support, or cardiopulmonary resuscitation, but no specific therapy for the intoxicant) or to mechanical
support to circulation, by means of extracorporeal life support (ECLS) [1013] or
IABP [14]. In all of the studies, mechanically supported animals were all successfully weaned, while weaning rate in control groups ranged from 10 to 25 % only.
Clinical reports of mechanical support for cardiogenic shock in drug intoxication
will be examined later in this chapter.
15.4
Options for Mechanical Circulatory Support

in Drug Intoxication
Mechanical circulatory support in cardiogenic shock due to drug intoxication may

rely on the use of IABP, CPB, or ECMO.
15.4.1 Aortic Counterpulsation

IABP is the most currently used form of mechanical support to circulation. The
alternated inflation and deflation of the balloon, synchronized with the cardiac
cycle, improve peak diastolic pressure and coronary blood flow while reducing endsystolic pressure, afterload, and myocardial oxygen consumption. It is the least
[email protected]
174
P. Bruno et al.
expensive form of mechanical circulatory support and it is easy and fast to deploy.
It has been used alone to treat cardiotoxic poisonings induced by quinidine [15],
propranolol [16], dextropropoxyphene [17], antihistamine [18], and a combination
of verapamil and atenolol [19]. It has also been used in combination with ECLS in
the case of organophosphate poisoning [20].
In this context, however, IABP shows some important limitations: in the first
place, it requires some residual cardiac function to be effective and is totally ineffective when systolic arterial pressure is less than 40 mmHg or during a cardiac
arrest. Another drawback is the impossibility of providing oxygenation, which
might be needed in severe poisonings impairing the respiratory function.
15.4.2 Cardiopulmonary Bypass via Central Cannulation

CPB provides full circulatory support and oxygenation. It requires a full median
sternotomy, so that a venous return cannula is inserted in the right atrium, while the
arterial inflow cannula is inserted in the ascending aorta. Sporadic experiences are
reported in literature [21, 22]: indeed, the strong limits to this technique are linked
to its high invasiveness, the required setting of implantation (operatory room), and,
therefore, the longer times required before ECLS can be provided. In case of cardiac
arrest, it is not possible to deliver uninterrupted chest compressions before CPB is
started. A prolonged assistance is inevitably associated with infective sequelae and
higher rates of bleeding issues (a high anticoagulation regimen is required).
15.4.3 ECMO
Features and indications of ECMO support in the setting of refractory cardiogenic
shock from drug intoxication do not differ from what has already been discussed in
the previous chapters of this book. The possibilities of a complete cardiopulmonary
support, a rapid deployment, and an implantation almost anywhere inside the hospital make ECMO the support of choice in the setting of refractory cardiogenic
shock from drug intoxication. Peripheral cannulation through the femoral vessels
allows continuation of chest compressions if the patient is in cardiac arrest.
Requirements in terms of anticoagulation are minor; if compared with CPB, the
infective risk is smaller and postoperative pain is avoided.
Nonetheless, the rate of complications is all but negligible, rising exponentially
with the duration of support. Knowing this, and remembering the outcomes reported
without ECLS [4], we once again underline that ECMO support should be reserved
to very sick patients in whom the risk of death overcomes the risk of ECMO-related
complications. On the other hand, as we will see later in this chapter, if we examine
the outcomes reported in literature for ECMO support in cardiogenic shock, we will
find that drug intoxication is one of the most favorable scenarios. Indeed, in this
context the planned strategy for ECMO support is almost inevitably bridging to
recovery.
[email protected]
15
175
The evidence on the subject relies mostly on isolated case reports and three case
series. The first reported case series by Babatasi et al. dates 2001 [23]: six patients
with cardiac arrest following acute severe self-administration of an overdose of
beta-blockers, calcium antagonists, or antiarrhythmics were supported on femorofemoral venoarterial ECMO. The first two patients died of multiorgan failure due to
a delay in the installation of the assistance, while the remaining four patients survived without sequelae.
In 2009, Daubin and colleagues published the largest case series to date [24]:
over a period of 10 years, out of 721 patients admitted for drug intoxication, 17
patients with refractory cardiogenic shock (n = 10) or cardiac arrest (n = 7) fulfilled the institutions criteria for ECMO implantation. In all of them, cannulation was achieved through the groin vessels and assistance was venoarterial.
Thirteen patients survived and were discharged without significant cardiovascular or neurological sequelae. In 2012, Masson published the first retrospective
cohort analysis comparing survival among critically ill poisoned patients treated
with or without ECLS [25]. Sixty-two patients with cardiogenic shock (n = 42) or
cardiac arrest (n = 20) following poisoning from drug intoxications were admitted in two centers over a time span of 10 years: 14 were treated with ECLS and
the remaining 48 with conventional therapies. Global survival was 56 % (35
patients): 86 % in the ECLS group and 48 % in the non-ECLS group (p = 0.02).
Notably, none of the patients with persistent cardiac arrest survived with conventional therapy.
A recent review by De Lange et al. [26] extensively examined the literature on
the subject: a total of 46 publications were found that dealt with ECMO support in
drug poisoning. The authors concluded commenting that, in the absence of contraindications, the organ support provided by ECMO makes it especially useful in
patients with severe poisoning, as the clinical impact of the intoxication is often
temporary; therefore, ECMO can be used as a bridge to recovery and is a good
salvage therapy for patients who are severely poisoned with acute respiratory distress syndrome (ARDS) or refractory circulatory shock.
The overall rate of complications (typically bleeding at the surgical entry site or
intracranial hemorrhage) did not differ from reported complications for ECMO support in general.
The heart being the most severely affected organ in this subset of patients, it is
important to remember that, though supporting the circulation, ECMO may have
detrimental effects on the left ventricle. In fact, in a not negligible proportion of
cases, the combination of severely reduced left ventricular function, blood return to
the left atrium via the bronchial circulation, and increased afterload from the arterial
cannula result in a dangerous rise in left atrial and ventricular pressures and pulmonary congestion [27]. The increase in wall stress due to left ventricular distention
decreases myocardial perfusion and increases oxygen consumption leading to ischemia and reducing the likelihood of ventricular recovery [28]. In literature, a case of
septal atriotomy to accomplish mechanical decompression of the left heart in
patients with ECMO support for drug intoxication has already been described [24],
but other methods are also available [29].
[email protected]
176
P. Bruno et al.
If the patient cannot be weaned from the ECMO support, the initial bridge to
recovery strategy can turn into a bridge to ventricular assist device (VAD) or a
bridge to transplantation strategy.
To date, no cases of switch from ECMO to VAD in drug poisoning have been
reported. Nonetheless, it is technically feasible to alter the ECMO circuit in order
to turn it into a left VAD for a potential midterm support by means of a left thoracotomy that allows the insertion of an outflow cannula in the apex of the left ventricle [29].
A case of ECMO as bridge to transplantation in a case of flecainide and betaxolol
poisoning has been reported in 2010 [30].
15.5
Conclusions
In rare cases of drug-induced cardiogenic shock refractory to conventional therapies,

the role of mechanical circulatory support is to sustain the cardiocirculatory system
until the heart recovers. Intra-aortic balloon counterpulsation plays a role in less
severe forms, in which it helps the failing heart. ECMO support is crucial when the
cardiopulmonary function is severely depressed: it allows to gain time until recovery
or until further options are available (heart transplantation, implantation of long-term
devices). In the largest reported clinical series to date, weaning rate was 76 % with
an average ECLS duration of 4.5 2.4 days [24]. The relatively short duration of support for these patients translates into an acceptable rate of ECMO-related complications. In the light of the above reported results, drug poisoning appears to be one of
the most favorable scenarios for ECMO support in cardiogenic shock.
References
1. Henry JA, Cassidy SL (1986) Membrane stabilising activity: a major cause of fatal poisoning.
Lancet 1:14141417
2. Lai MW, Klein-Schwartz W, Rodgers GC, Abrams JY, Haber DA, Bronstein AC, Wruk KM
(2006) 2005 Annual Report of the American Association of Poison Control Centers national
poisoning and exposure database. Clin Toxicol (Phila) 44:803932
3. DeWitt CR, Waksman JC (2004) Pharmacology, pathophysiology and management of calcium
channel blocker and beta-blocker toxicity. Toxicol Rev 23:223238
4. Baud FJ, Megarbane B, Deye N, Leprince P (2007) Clinical review: aggressive management
and extracorporeal support for drug-induced cardiotoxicity. Crit Care 11(2):207
5. Koppel C, Oberdisse U, Heinemeyer G (1990) Clinical course and outcome in class IC antiarrhythmic overdose. J Toxicol Clin Toxicol 28:433444
6. Boehnert MT, Lovejoy FH Jr (1985) Value of the QRS duration versus the serum drug level in
predicting seizures and ventricular arrhythmias after an acute overdose of tricyclic antidepressants. N Engl J Med 313(8):474479
7. Clemessy JL, Taboulet P, Hoffman JR, Hantson P, Barriot P, Bismuth C, Baud FJ (1996)
Treatment of acute chloroquine poisoning: a 5-year experience. Crit Care Med 24:11891195
8. Love JN, Howell JM, Litovitz TL, Klein-Schwartz W (2000) Acute beta blocker overdose:
factors associated with the development of cardiovascular morbidity. J Toxicol Clin Toxicol
38:275281
[email protected]
15
177
9. Albertson TE, Dawson A, de Latorre F, Hoffman RS, Hollander JE, Jaeger A, Kerns WR 2nd,
Martin TG, Ross MP, American Heart Association, International Liaison Committee on
Resuscitation (2001) TOX-ACLS: toxicologic-oriented advanced cardiac life support. Ann
Emerg Med 37(4 Suppl):S7890
10. Freedman MD, Gal J, Freed CR (1982) Extracorporeal pump assistance novel treatment for
acute lidocaine poisoning. Eur J Clin Pharmacol 22:129135
11. Martin TG, Klain MM, Molner RL, Michelson EA, Schneider SM (1990) Extracorporeal life
support vs thumper after lethal desipramine OD. Vet Hum Toxicol 32:349
12. Larkin GL, Graeber GM, Hollingsed MJ (1994) Experimental amitriptyline poisoning: treatment of severe cardiovascular toxicity with cardiopulmonary bypass. Ann Emerg Med
23:480486
13. Martin TG, OConnell JJ, Pentel P, Miller DL, Killer DE, Knox MA (1988) Resuscitation in
severe cyclic antidepressant toxicity using cardiopulmonary bypass. Vet Hum Toxicol 30:364
14. Grossman JI, Furman S (1971) Intraaortic balloon augmentation during drug-induced myocardial depression. Surgery 70:304310
15. Shub C, Gau GT, Sidell PM, Brennan LA Jr (1978) The management of acute quinidine intoxication. Chest 73:173178
16. Lane AS, Woodward AC, Goldman MR (1987) Massive propranolol overdose poorly responsive to pharmacologic therapy: use of the intra-aortic balloon pump. Ann Emerg Med
16:13811383
17. Gillard P, Laurent M (1999) Dextropropoxyphene-induced cardiogenic shock: treatment with
intra-aortic balloon pump and milrinone. Intensive Care Med 25:335
18. Freedberg RS, Friedman GR, Palu RN, Feit F (1987) Cardiogenic shock due to antihistamine
overdose. Reversal with intra-aortic balloon counterpulsation. JAMA 257:660661
19. Frierson J, Bailly D, Shultz T, Sund S, Dimas A (1991) Refractory cardiogenic shock and
complete heart block after unsuspected verapamil-SR and atenolol overdose. Clin Cardiol
14:933935
20. Kamijo Y, Soma K, Uchimiya H, Asari Y, Ohwada T (1999) A case of serious organophosphate
poisoning treated by percutaneus cardiopulmonary support. Vet Hum Toxicol 41:326328
21. Hendren WG, Schieber RS, Garrettson LK (1989) Extracorporeal bypass for the treatment of
verapamil poisoning. Ann Emerg Med 18:984987
22. Pasic M, Potapov E, Kuppe H, Hetzer R (2000) Prolonged cardiopulmonary bypass for severe
drug intoxication. J Thorac Cardiovasc Surg 119:379380
23. Babatasi G, Massetti M, Verrier V, Lehoux P, Le Page O, Bruno PG, Khayat A (2001) Severe
intoxication with cardiotoxic drugs: value of emergency percutaneous cardiocirculatory assistance. Arch Mal Coeur Vaiss 94(12):13861392
24. Daubin C, Lehoux P, Ivascau C, Tasle M, Bousta M, Lepage O, Quentin C, Massetti M,
Charbonneau P (2009) Extracorporeal life support in severe drug intoxication: a retrospective
color study of seventeen cases. Crit Care 13(4):R138
25. Masson R, Colas V, Parienti JJ, Lehoux P, Massetti M, Charbonneau P, Saulnier F, Daubin C
(2012) A comparison of survival with and without extracorporeal life support treatment for
severe poisoning due to drug intoxication. Resuscitation 83(11):14131417
26. De Lange DW, Sikma MA, Meulenbelt J (2013) Extracorporeal membrane oxygenation in the
treatment of poisoned patients. Clin Toxicol (Phila) 51(5):385393
27. Combes A, LePrince P, Luyt C-E, Trouillet J-L, Chastre J (2009) Extracorporeal membrane
oxygenation (ECMO) for cardiopulmonary support. Reanimation 18:420427
28. Scholz KH, Schrder T, Hering JP, Ferrari M, Figulla HR, Chemnitius JM et al (1994) Need
for active left-ventricular decompression during percutaneous cardiopulmonary support in cardiac arrest. Cardiology 84(3):222230
29. Massetti M, Gaudino M, Saplacan V, Farina P (2013) From extracorporeal membrane oxygenation to ventricular assist device without sternotomy. J Heart Lung Transplant 32(1):138139
30. Vivien B, Deye N, Mgarbane B, Marx JS, Leprince P, Bonnet N, Roussin F, Jacob L, Pavie A,
Baud FJ, Carli P (2010) Extracorporeal life support in a case of fatal flecainide and betaxolol
poisoning allowing successful cardiac allograft. Ann Emerg Med 56(4):409412
[email protected]
Newer Indications for ECMO:

Pulmonary Embolism,
Pulmonary Hypertension,
Septic Shock and Trauma
16
Michela Bombino, Sara Redaelli, and Antonio Pesenti
16.1
ECMO in Right Ventricular Failure
Massive pulmonary embolism (MPE) is a keystone in the history of ECMO, since

the invention of the heart-lung machine was urged by the disappointing death of a
patient suffering from this condition [1]. In 1931, Gibbon was assigned to monitor
a lady with pulmonary embolism after a cholecystectomy; he had to call his chief
when the situation was deteriorating at the point that a pulmonary embolectomy
could be tried as a last chance. Gibbon [2] vividly described what happened:
During that long night, helplessly watching the patient struggle for life as her blood became
darker and her veins more distended, the idea naturally occurred to me that if it were possible to remove continuously some of the blue blood from the patients swollen veins, put
oxygen into that blood and allow carbon dioxide to escape from it, and then to inject continuously the now-red blood back into the patients arteries, we might have saved her life.
We would have bypassed the obstructing embolus and performed part of the work of the
patients heart and lungs outside the body.
Then Gibbon and his wife started their experimental work on acute occlusive
pulmonary hypertension and right ventricular failure [3, 4] and finally developed the
heart-lung machine [2] reporting in 1937 the possibility to maintain circulation in
cats with experimental occlusion of the pulmonary artery [5].
M. Bombino (*) S. Redaelli

Department of Emergency and Urgency, General Intensive Care Unit,
San Gerardo Hospital, Via Pergolesi, 33, Monza (MB) 20900, Italy
A. Pesenti
Department of Health Sciences, University of Milano-Bicocca,
San Serardo Hospital, Via Pergolesi, 33, Monza (MB) 20900, Italy
[email protected]
179
180
M. Bombino et al.
Many years passed since then, but the indication for ECMO in supporting the
failing right ventricle (RV) never subsided. The rationale of ECMO support in RV
failure is to divert some blood from the right atrium to the arterial circulation, thus
unloading the RV and relieving its dilatation, which in turn will lead to increased
left ventricular output due to ventricular interdependence [6, 7]. ECMO relieves
hypoxemia due to shunt in this setting and, through the required anticoagulation,
provides also a therapeutic mean in thromboembolism.
In idiopathic pulmonary arterial hypertension, extracorporeal support is used to
bridge patients to lung transplantation when medical therapy is exhausted or as a
temporary aid in cases of increased cardiovascular requirement.
16.1.1 ECMO in Massive Pulmonary Embolism

Despite the great therapeutic armamentarium available for MPE, the reported mortality of patients presenting with RV failure and cardiogenic shock is still as high as
2050 % [8].
Davies reported the first successful use of ECMO as a temporary support in MPE
outside the operating room in 1995 [9]. It is noteworthy that the patient was kept
conscious during the 6-day ECMO support. Plenty of case reports have been published since then on the successful use of extracorporeal support in MPE leading to
cardiac arrest or cardiogenic shock [1027]; some patients were treated only with
ECMO, and others were successfully bridged to surgical or catheter thrombectomy.
Case reports are biased by the high survival rate, since they report almost always on
successful cases. Key points in the published case reports are the prompt institution
of percutaneous ECMO, the achievement of haemodynamic stabilisation, the possibility to perform diagnostic exams during ECMO and the transport of the patient
to a facility where surgical embolectomy could be feasible.
Maggio and co-workers [28] published one of the largest series, including 21
patients with MPE with profound shock and severe hypoxemia treated with V-A
(19 patients) or VV ECMO (2 patients). The overall survival rate was 62 %. They
pointed out the feasibility of the rapid institution of percutaneous ECMO support
as a rescue manoeuvre also when the patients suffered cardiac arrest due to fulminant PE (8 cases). ECLS can be part of a therapeutic strategy comprising
thrombolytics and other kinds of thrombectomies or can be curative with anticoagulation by itself. Daily follow-up with echocardiography would recognise the
few patients requiring surgical embolectomy. Unfortunately, neurological complications leading to death were high in this patient population (4 patients, 50 %
of the deaths).
Other case series [2931] with good overall results have been published. Hashiba
[30] reported 12 patients with fulminant PE in cardiac arrest at time of ECMO institution, 10 of them survived. The authors pointed out that both survival and neurological
outcomes of cardiac arrest patients with MPE were better compared to the outcomes
of 16 patients with post-acute myocardial infarction cardiac arrest (survival 83.3 % vs.
12.5 %, p < 0.001; good neurological outcome 58.3 % vs. 6.3 %, p = 0.004).
[email protected]
16
Newer Indications for ECMO
181
Recently, Sakuma [32] reported the Japanese experience with percutaneous

ECMO as adjunctive support in MPE. 193 cases were collected from the literature;
the overall survival rate was 73, 65 % in patients with cardiac arrest at the time of
ECMO institution and as high as 86 % in those with cardiogenic shock.
In conclusion, ECMO support is of established benefit in patients with MPErelated cardiac failure, and it is recognised as a fundamental step in its interventional treatment algorithm [33, 34].
16.1.2 ECMO in Pulmonary Arterial Hypertension

The term pulmonary hypertension (PH) comprises different clinical entities as
pointed out by the 2008 Dana Point classification [35]. The importance of an early
diagnosis and the recognition of common pathophysiologic mechanisms underlying
some diseases are pivotal steps to establish an optimal management [3641].
Furthermore, algorithms have been developed for the assessment and accurate classification of PH and for treatment allocation. Despite improvement in overall management for these patients, mortality is still high; poor prognostic factors are a
scleroderma-associated diagnosis and indices of RV failure.
RV failure can supervene in the setting of pulmonary arterial hypertension (PAH)
also if there is a response to pulmonary vasodilators. Its pathophysiology is largely
reviewed in the literature [7, 4247]: RV adapts better to volume overload than pressure overload, chronic pressure overload will cause RV dilatation and increase in wall
stress leading to hypertrophic remodelling, RV ischemia can ensue and cardiac output will decrease due to ventricular interdependence. Treatment of RV failure depends
on underlying disease and its stage [7, 4547] and the chronic or acute presentation.
ECMO can play a role in some conditions leading to RV failure as a temporary support in crisis deteriorations or most commonly as a bridge to transplantation.
Veno-venous ECMO has been successfully used in cases of RV failure due to
chronic pulmonary embolism hypertension after pulmonary endarterectomy to treat
reperfusion syndrome or persistent pulmonary hypertension after surgery [48, 49].
In a recent survey on surgical management and outcome of patients with chronic
thromboembolic pulmonary hypertension, Mayer reported a 9.6 % incidence of pulmonary reperfusion oedema, and 16.7 % of the patients had persistent pulmonary
hypertension. The need for ECMO in this population was 3.1 % [50].
In idiopathic PAH, ECMO has been used with different indications. Pereszlenyi
[51] reported the use of ECMO intra- and postoperatively in patients undergoing
bilateral lung transplantation: ECMO allowed the maintenance of protective ventilation and a controlled reperfusion of the allograft.
VA ECMO is the choice if the patient is haemodynamically unstable due to
global heart failure [5254], but different approaches have been advocated to unload
the RV. In the case of a patent foramen ovale, VV ECMO can succeed by itself in
this task [55]. In other cases, VV ECMO plus an atrial septostomy has been applied
[56, 57], or a shunt between the pulmonary artery and the left atrium (PA-LA) interposed with a low-resistance pumpless system [54, 58]. The introduction of ECMO
[email protected]
182
M. Bombino et al.
as bridge to transplantation in idiopathic PAH patients has been reported to

decrease the waiting list mortality and was not associated with a worse outcome
after transplantation [59].
Another ECMO indication is to overcome a pulmonary hypertensive crisis in
otherwise stable patients on pharmacological therapy [55, 60] or in cases of abrupt
RV failure due to anaesthesia as described in a parturient PAH patient [61]. In the
last case, ECMO has also been used as a pre-emptive measure [62] during elective
termination of gestation.
In conclusion, temporary support of RV failure with ECMO during hypertensive pulmonary crisis or its long-term use as a bridge to lung transplantation is
largely reported in the literature. ECMO support before transplantation is now feasible in awake, non-intubated patients with the possibility to perform physical
therapy while waiting for organ assignment and is associated with a better prognosis [6365].
16.2
ECMO in Septic Shock
Septic shock in adults is mainly hyperdynamic, characterised by vasoplegia and

unresponsive to fluid loading, and the mechanisms underlying its pathogenesis are
well described [66, 67]. Nevertheless, Dellinger [68] defined, more than 10 years
ago, septic shock as a melting pot of shock aetiologies, between which also a
cardiogenic component can be present. Pathophysiologic mechanisms [6972] and
echocardiography findings [73, 74] of septic cardiomyopathy are now well
characterised.
16.2.1 ECMO in Paediatric Septic Shock

Children and neonates with septic shock, due to their limited cardiac reserve,
present in the majority of cases with a cold shock characterised by profound
vasoconstriction, and therefore, a cardiogenic component is almost always at play
[75]. For this reason, the first applications of ECMO in septic shock have been in
neonates and children. Already in 1995, the Extracorporeal Life Support
Organization (ELSO) published the registry results on neonatal sepsis [76] and
few years after in paediatric septic shock [77]. The conclusions were that ECMO
support should not be withheld in these populations, despite a higher rate of intracranial haemorrhage compared with the non-septic patients. Experience in the use
of ECMO in septic shock children increased, with overall good results [7881],
and ECMO support is indicated in the clinical guidelines for the treatment of neonatal and paediatric septic shock [82]. Controversies exist about the utility of
ECMO in specific groups of septic shock children, like in meningococcal septicaemia [83, 84]. Indeed, other authors pointed out that the mortality increases
with age despite ECMO [85].
[email protected]
16
183
16.2.2 ECMO in Adult Septic Shock

In the early years, septic shock and bacteremia were considered relative contraindications to ECMO. The concern was that the ECMO circuit, entrapping bacteria,
could behave as a culture medium [86]. Septic coagulopathy and the related
increased risk of bleeding was also a contraindication. For these reasons, there are
scanty reports, till the recent years, about the use of ECMO in adult septic shock
[87, 88]. During the last decade, with the improvement in ECMO technology and its
safety, we observed expanding indications for ECMO use, and its absolute contraindications are continuously challenged [89]. Some case reports on ECMO use in
adult septic shock related to Hantavirus [90], malaria [91], Staphylococcus aureus
[92], Neisseria meningitides [93], H1N1 influenza [94] and necrotising soft tissue
infections [95] have been recently published.
Two series have been recently accepted for publication. Brchot [96] reports on
14 patients with cold refractory septic shock with a very low left ventricular ejection fraction and elevated systemic resistance vascular index: the survival rate was
71 %. The other series is from a single centre in Taiwan [97]: 53 patients in septic
shock were treated with ECMO in a 6-year period, 40 % of them were in cardiac
arrest at time of ECMO institution and the overall survival to discharge was only
15 %. Age greater than 60 years was associated with the worst outcome. The large
difference in the reported survival can be at least in part explained by the presence
of a cardiac arrest group in the Taiwan series; the French cohort has, indeed, a septic
cardiomyopathy characterised by profound left ventricular ejection fraction and
very high systemic resistances, like a Takotsubo-like cardiomyopathy, that can
explain the high survival observed.
16.2.3 Conundrums in ECMO for Septic Shock

In our opinion, it is essential to distinguish three groups of patients when dealing
with the use of ECMO in septic shock:
(a) Patients with cardiogenic shock and sepsis. Most of them can be treated with
peripheral VA ECMO, being the cardiac component prevalent over the respiratory one.
(b) Patients with ARDS and hyperdynamic septic shock. In this setting, VV ECMO
is preferred, and a reduction in catecholamines requirements is commonly
observed at its start. Furthermore, a worse outcome of septic shock patients
treated with VA ECMO compared with matched patients on VV ECMO has
been recently reported [98].
(c) Patients with a depressed left ventricular function and concomitant severe
ARDS. In these cases, the use of peripheral femoro-femoral VA ECMO can
result in deoxygenated blood being ejected by the left ventricle, due to the pulmonary dysfunction, and perfusing the heart, brain and upper body. A central
VA ECMO [96] or a V-V/V-A hybrid ECMO can be used to overcome this
[email protected]
184
M. Bombino et al.
problem; in the last case, the use of a bicaval double-lumen cannula can prevent
the cannulation of a new jugular venous site to increase the oxygen saturation
in the right atrium/ventricle [99]. Some authors described a better outcome in
patients on veno-venoarterial ECMO when ARDS was associated with cardiac
compromise [100, 101].
16.3
ECMO in Trauma
Major trauma is a leading cause of death, particularly among young patients, in

three different ways: (1) immediate death on stage due to intractable injuries; (2)
early death (hours to few days) related to severe haemorrhage, cardiovascular/pulmonary failure or severe brain injury; (3) late death (days to weeks) usually as a
consequence of post-traumatic ARDS and/or multi-organ failure. ECMO may be an
effective and lifesaving strategy in both early and late trauma-related deaths. Since
major causes of early deaths are haemorrhagic shock, hypoxemia, hypothermia,
metabolic acidosis and coagulopathy, by means of VA ECMO, we can restore adequate tissue perfusion and oxygenation, achieve quickly rewarming and infuse massive fluids or blood products. Conversely, in the treatment of post-traumatic ARDS,
VV ECMO ensures protective ventilation of the traumatised lung, providing adequate oxygenation and avoiding consequent multi-organ failure. The association of
ARDS and traumatic brain injury leads to higher mortality rate [102]; lung protective ventilation, permissive hypercapnia and increased intra-thoracic pressures used
in ARDS may lead to worsening brain injury. VV ECMO could resolve this therapeutic conflict allowing adequate pulmonary support and lung rest and minimising secondary brain damage [103].
The indication for VV ECMO support in the early phase may be refractory
hypoxemia and/or severe hypercapnic acidosis despite full conventional treatment.
Conversely, VA ECMO support should be instituted in case of persisting shock,
with signs of tissue hypoperfusion, despite fluid resuscitation, blood transfusion and
vasopressor support or in case of post-traumatic cardiac arrest. Overall, in the
absence of specific trauma guidelines, ECMO institution may be considered when
clinicians are convinced that standard therapies have been exhausted in patients
with potentially reversible injuries [104, 105]. Larsson and colleagues recently suggested that VA ECMO can play a possible role in the control of pulmonary bleeding
since it reduces pulmonary perfusion; furthermore, it may be helpful during surgery
since the jugular-femoral approach allows inferior vena cava clamping without
compromising systemic organ perfusion [106].
Since the first successful use of ECMO in a trauma patient performed by Hill
in 1972 [107], post-traumatic ARDS became a common indication for extracorporeal support. Furthermore, patients with post-traumatic ARDS were included in the
major randomized trials on ECMO [108111]. Early institution of ECMO (days
on mechanical ventilation 5) has been associated with a better outcome [112] in
this population. Finally, Bein and colleagues proposed the application of ECMO in
hypercapnic patients with traumatic brain injury to minimise secondary brain damage [103].
[email protected]
16
185
16.3.1 Contraindications
Only few reports clearly listed contraindications to ECMO in trauma patients: (1)
fatal cerebral lesions, (2) incontrollable major bleeding (e.g. aortic rupture), (3)
advanced age (>5570 years), (4) witnessed prolonged hypoxemia (e.g. prolonged
inefficacious resuscitation) and (5) potentially fatal pre-existing diseases [105, 113].
Prior to ECMO institution, a total-body CT scan is recommended, when feasible, to
rule out the presence of absolute contraindications [104, 105].
Active bleeding, recent surgery and brain injury are recognised contraindications
to anticoagulation and thus to ECMO, but recently several reports [104, 114116]
have been published on the successful use of ECMO in trauma patients after damage surgery, with severe brain injury or in bleeding shock.
16.3.2 Conundrums in ECMO Management in Trauma

16.3.2.1 Timing of ECMO Institution
Timing of ECMO institution depends on the clinical conditions and local resources;
thus, emergency/urgent support is often established in emergency departments or in
operating rooms during damage control surgery, while respiratory support in posttraumatic ARDS is mainly instituted in intensive care units. Emergency ECMO has
also been instituted in austere settings like war zones to allow aeromedical evacuation of combat casualties to specialised centres [117120].
16.3.2.2 Anticoagulation
Mostly in these patients, with high risk of bleeding, anticoagulation should be carefully managed with close ACT and/or aPTT monitoring. Some authors [104, 105,
115, 116] suggest heparin administration only after surgical bleeding control with a
heparin-free ECMO running time of few hours or days. Others [114, 121, 122]
apply standard anticoagulation protocol and prefer to stop or reduce heparin infusion only if bleeding occurs. Muellenbach reported one case of multiple thrombi in
inferior vena cava after 3 days on heparin-free ECMO in a patient who however
received recombinant factor VIIa (rFVIIa) for massive bleeding [115]. No other
main thrombotic events were recorded, and bleeding complication rate was similar
to traditionally ARDS patients undergoing ECMO.
Surgical procedures during ECMO are extremely frequent with low rate of
bleeding complications [103105, 112116, 121126]. Usually, if systemic anticoagulation is already established and surgery is scheduled, heparin should be interrupted few hours (46 h) before surgery.
The use of rFVIIa in massive bleeding in ECMO is controversial [127, 128]:
despite the efficacy in bleeding control and, specifically in trauma patients, in reducing the risk of ARDS, some concerns remain on the risk of thrombotic events either
in the circuit (circuit clotting or obstruction, oxygenator failure) or in the patients
(stroke or pulmonary embolism). A higher rate of thrombotic events (33 % vs. 22 %
in the adult population) was observed in the paediatric population, possibly due to
differences in coagulation system and in ECMO circuit size [127].
[email protected]
186
M. Bombino et al.
In summary, when the risk of bleeding is very high (i.e., in trauma patients withintracranial haemorrhage, severe brain injury or bleeding shock), either low-dose
heparin or temporary heparin-free ECMO could be used with some precautions:
(1) high blood flow to reduce the risk of circuit clotting; (2) tight coagulation monitoring including D-dimer or fibrin degradation product (FDP), fibrinogen and
platelet count; and (3) close membrane lung performance monitoring for early
recognition of oxygenator failure/circuit thrombosis. Factor rVIIa may be considered in massive bleeding patients after correction of coagulation parameters, hypothermia, hypocalcemia and acidosis. After bleeding control, systemic
anticoagulation should be started according to standard local protocol or with
lower ACT/aPTT target if the risk of bleeding still remains high.
16.3.2.3 Traumatic Brain Injury

Traumatic brain injury is traditionally a relative contraindication to ECMO; however, several reports have been published on ECMO in patients with traumatic brain
injury [103, 115, 116, 122, 123, 125, 129]. In all cases except one [122], intracranial
pressure monitoring was established as soon as possible. Anticoagulation management was different among the reports: in some cases, heparin was started only after
hours or days from ECMO institutions, with only one case of multiple thrombi
[115]; in others, standard anticoagulation was performed, with a case of possible
worsening in intracranial bleeding after ECMO institution [125].
Available data are not sufficient to make definitive recommendations; thus, each
case should be carefully evaluated for benefits and risks of ECMO institution; in
high-risk patients, ECMO should be run without anticoagulation and certainly with
intracranial monitoring to promptly recognise and treat bleeding complications.
References
1. Gibbon JH Jr, Hill JD (1982) Part I. The development of the first successful heart-lung
machine. Ann Thorac Surg 134:337341
2. Gibbon JH Jr (1968) Development of the artificial heart and lung extracorporeal blood circuit.
JAMA 206:19831986
3. Gibbon JH Jr, Hopkinson M, Churchill ED (1932) Changes in the circulation produced by
gradual occlusion of the pulmonary artery. J Clin Invest 11:543553
4. Gibbon JH Jr, Churchill ED (1936) The physiology of massive pulmonary embolism. An
experimental study of the changes produced by obstruction to the flow of blood through the
pulmonary artery and its lobar branches. Ann Surg 104:811822
5. Gibbon JH Jr (1937) Artificial maintenance of circulation during experimental occlusion of
pulmonary artery. Arch Surg 34:11051131
6. Lahm T, McCaslin CA, Wozniak TC et al (2010) Medical and surgical treatment of acute right
ventricular failure. J Am Coll Cardiol 56:14351446
7. Haddad F, Doyle R, Murphy DJ, Hunt SA (2008) Right ventricular function in cardiovascular
disease, part II: pathophysiology, clinical importance, and management of right ventricular
failure. Circulation 117:17171731
8. Goldhaber SZ, Visani L, De Rosa M (1999) Acute pulmonary embolism: clinical outcomes in the
International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet 353:13861389
[email protected]
16
187
9. Davies MJ, Arsiwala SS, Moore HM et al (1995) Extracorporeal membrane oxygenation for
the treatment of massive pulmonary embolism. Ann Thorac Surg 60:18011803
10. Ohteki H, Norita H, Sakai M, Narita Y (1997) Emergency pulmonary embolectomy with percutaneous cardiopulmonary bypass. Ann Thorac Surg 63:15841586
11. Murata S, Adachi H, Ino T, Yamaguchi A et al (1997) An emergent surgical case of acute massive pulmonary embolism supported by antithrombotic percutaneous cardiopulmonary support
system. J Jpn Assoc Thorac Surg 45:11591164
12. Kolvekar SK, Peek GJ, Sosnowski AW, Firmin RK (1997) Extracorporeal membrane oxygenation for pulmonary embolism. Ann Thorac Surg 64:883884
13. Kawahito K, Murata S, Ino T, Fuse K (1998) Angioscopic pulmonary embolectomy and
ECMO. Ann Thorac Surg 66(3):982983
14. Sudo K, Ide H, Fujiki T et al (1999) Pulmonary embolectomy for acute massive pulmonary
embolism under percutaneous cardiopulmonary support. J Cardiovasc Surg (Torino)
40:165167
15. Misawa Y, Fuse K, Yamaguchi T et al (2000) Mechanical circulatory assist for pulmonary
embolism. Perfusion 15:527529
16. Hsieh PC, Wang SS, Ko WJ et al (2001) Successful resuscitation of acute massive pulmonary
embolism with extracorporeal membrane oxygenation and open embolectomy. Ann Thorac
Surg 72:266267
17. Deehring R, Kiss AB, Garrett A, Hillier AG (2006) Extracorporeal membrane oxygenation as
a bridge to surgical embolectomy in acute fulminant pulmonary embolism. Am J Emerg Med
24:879880
18. Haller I, Kofler A, Lederer W et al (2008) Acute pulmonary artery embolism during transcatheter embolization: successful resuscitation with veno-arterial extracorporeal membrane oxygenation. Anesth Analg 107:945947
19. Frickey N, Kraincuk P, Zhilla I et al (2008) Fulminant pulmonary embolism treated by extracorporeal membrane oxygenation in a patient with traumatic brain injury. J Trauma
64:E41E43
20. Griffith KE, Jenkins E, Haft J (2009) Treatment of massive pulmonary embolism utilizing a
multidisciplinary approach: a case study. Perfusion 24:169172
21. Arlt M, Philipp A, Iesalnieks I et al (2009) Successful use of a new hand-held ECMO system
in cardiopulmonary failure and bleeding shock after thrombolysis in massive post-partal pulmonary embolism. Perfusion 24:4950
22. Hori D, Tanaka M, Kohinata T et al (2010) Successful usage of extracorporeal membrane
oxygenation as a bridge therapy for acute pulmonary embolism between hospitals. Gen Thorac
23. Mydin M, Berman M, Klein A et al (2011) Extracorporeal membrane oxygenation as a bridge
to pulmonary endarterectomy. Ann Thorac Surg 92:e101e103
24. Malekan R, Saunders PC, Yu CJ et al (2012) Peripheral extracorporeal membrane oxygenation: comprehensive therapy for high-risk massive pulmonary embolism. Ann Thorac Surg
94:104108
25. Ko CH, Forrest P, DSouza R, Qasabian R (2012) Case report: successful use of extracorporeal
membrane oxygenation in a patient with combined pulmonary and systemic embolisation.
Perfusion 28:138140
26. Leick J, Liebetrau C, Szardien S et al (2012) Percutaneous circulatory support in a patient with
cardiac arrest due to acute pulmonary embolism. Clin Res Cardiol 101:10171020
27. Zhong M, Tan L, Xue Z et al (2014) Extracorporeal membrane oxygenation as a bridge therapy
for massive pulmonary embolism after esophagectomy. J Cardiothorac Vasc Anesth.
doi:10.1053/j.jvca.2012.08.010 [Epub ahead of print]
28. Maggio P, Hemmila M, Haft J, Bartlett R (2007) Extracorporeal life support for massive pulmonary embolism. J Trauma 62:570576
29. Kawahito K, Murata S, Adachi H et al (2000) Resuscitation and circulatory support using
extracorporeal membrane oxygenation for fulminant pulmonary embolism. Artif Organs
24:427430
[email protected]
188
M. Bombino et al.
30. Hashiba K, Okuda J, Maejima N et al (2012) Percutaneous cardiopulmonary support in pulmonary embolism with cardiac arrest. Resuscitation 83:183187
31. Munakata R, Yamamoto T, Hosokawa Y et al (2012) Massive pulmonary embolism requiring extracorporeal life support treated with catheter-based interventions. Int Heart J 53:
370374
32. Sakuma M, Nakamura M, Yamada N et al (2009) Percutaneous cardiopulmonary support for
the treatment of acute pulmonary embolism: summarized review of the literature in Japan
including our own experience. Ann Vasc Dis 2:716
33. Imberti D, Ageno W, Manfredini R (2012) Interventional treatment of venous thromboembolism: a review. Thromb Res 129:418425
34. Wu MY, Liu YC, Tseng YH et al (2013) Pulmonary embolectomy in high-risk acute pulmonary
embolism: the effectiveness of a comprehensive therapeutic algorithm including extracorporeal
life support. Resuscitation 84:13651370
35. Simonneau G, Robbins IM, Beghetti M et al (2009) Updated clinical classification of pulmonary hypertension. J Am Coll Cardiol 54(1 Suppl):S43S54
36. McLaughlin VV, Archer SL, Badesch DB et al (2009) ACCF/AHA 2009 expert consensus
document on pulmonary hypertension: a report of the American College of Cardiology
Foundation Task Force on Expert Consensus Documents and the American Heart Association:
developed in collaboration with the American College of Chest Physicians, American Thoracic
Society, Inc., and the Pulmonary Hypertension Association. Circulation 119:22502294
37. Badesch DB, Champion HC, Sanchez MA et al (2009) Diagnosis and assessment of pulmonary arterial hypertension. J Am Coll Cardiol 54(1 Suppl):S55S66
38. Vachiry JL, Gaine S (2012) Challenges in the diagnosis and treatment of pulmonary arterial
hypertension. Eur Respir Rev 21:313320
39. Peacock A (2013) Pulmonary hypertension. Eur Respir Rev 22:2025
40. Poor HD, Ventetuolo CE (2012) Pulmonary hypertension in the intensive care unit. Prog
Cardiovasc Dis 55:187198
41. Price LC, McAuley DF, Marino PS et al (2012) Pathophysiology of pulmonary hypertension
in acute lung injury. Am J Physiol Lung Cell Mol Physiol 302:L803L815
42. Voelkel NF, Quaife RA, Leinwand LA et al (2006) Right ventricular function and failure:
report of a National Heart, Lung, and Blood Institute working group on cellular and molecular
mechanisms of right heart failure. Circulation 114:18831891
43. Greyson CR (2008) Pathophysiology of right ventricular failure. Crit Care Med 36(1 Suppl):
S57S65
44. Simon MA, Pinsky MR (2011) Right ventricular dysfunction and failure in chronic pressure
overload. Cardiol Res Pract. doi:10.4061/2011/568095
45. Price LC, Wort SJ, Finney SJ et al (2010) Pulmonary vascular and right ventricular dysfunction in adult critical care: current and emerging options for management: a systematic literature review. Crit Care 14:R169
46. Green EM, Givertz MM (2012) Management of acute right ventricular failure in the intensive
care unit. Curr Heart Fail Rep 9:228235
47. Keogh AM, Mayer E, Benza RL et al (2009) Interventional and surgical modalities of treatment in pulmonary hypertension. J Am Coll Cardiol 54(1 Suppl):S67S77
48. Thistlethwaite PA, Madani MM, Kemp AD et al (2006) Venovenous extracorporeal life support after pulmonary endarterectomy: indications, techniques, and outcomes. Ann Thorac Surg
82:21392145
49. Berman M, Tsui S, Vuylsteke A et al (2008) Successful extracorporeal membrane oxygenation
support after pulmonary thromboendarterectomy. Ann Thorac Surg 86:12611267
50. Mayer E, Jenkins D, Lindner J (2011) Surgical management and outcome of patients with
chronic thromboembolic pulmonary hypertension: results from an international prospective
registry. J Thorac Cardiovasc Surg 141:702710
51. Pereszlenyi A, Lang G, Steltzer H et al (2002) Bilateral lung transplantation with intra- and
postoperatively prolonged ECMO support in patients with pulmonary hypertension. Eur J
Cardiothorac Surg 21:858863
[email protected]
16
189
52. Gregoric ID, Chandra D, Myers TJ et al (2008) Extracorporeal membrane oxygenation as a

bridge to emergency heart-lung transplantation in a patient with idiopathic pulmonary arterial
hypertension. J Heart Lung Transplant 27:466468
53. Hoeper MM (2011) Treat-to-target in pulmonary arterial hypertension and the use of extracorporeal membrane oxygenation as a bridge to transplantation. Eur Respir Rev 20:297300
54. Cypel M, Keshavjee S (2011) Extracorporeal life support as a bridge to lung transplantation.
Clin Chest Med 32:245251
55. Srivastava MC, Ramani GV, Garcia JP et al (2010) Veno-venous extracorporeal membrane
oxygenation bridging to pharmacotherapy in pulmonary arterial hypertensive crisis. J Heart
Lung Transplant 29:811813
56. Camboni D, Akay B, Sassalos P et al (2011) Use of venovenous extracorporeal membrane
oxygenation and an atrial septostomy for pulmonary and right ventricular failure. Ann Thorac
Surg 91:144149
57. Hoopes CW, Gurley JC, Zwischenberger JB, Diaz-Guzman E (2012) Mechanical support for
pulmonary veno-occlusive disease: combined atrial septostomy and venovenous extracorporeal membrane oxygenation. Semin Thorac Cardiovasc Surg 24:232234
58. Strueber M, Hoeper MM, Fischer S et al (2009) Bridge to thoracic organ transplantation in
patients with pulmonary arterial hypertension using a pumpless lung assist device. Am J
Transplant 9:853857
59. de Perrot M, Granton JT, McRae K et al (2011) Impact of extracorporeal life support on outcome in patients with idiopathic pulmonary arterial hypertension awaiting lung transplantation. J Heart Lung Transplant 30:9971002
60. Hsu HH, Ko WJ, Chen JS et al (2008) Extracorporeal membrane oxygenation in pulmonary
crisis and primary graft dysfunction. J Heart Lung Transplant 27(2):233237
61. Hhn L, Schweizer A, Morel DR (1999) Circulatory failure after anesthesia induction in a
patient with severe primary pulmonary hypertension. Anesthesiology 91(6):19431945
62. Satoh H, Masuda Y, Izuta S et al (2002) Pregnant patient with primary pulmonary hypertension: general anesthesia and extracorporeal membrane oxygenation support for termination of
pregnancy. Anesthesiology 97:16381640
63. Olsson KM, Simon A, Strueber M et al (2010) Extracorporeal membrane oxygenation in nonintubated patients as bridge to lung transplantation. Am J Transplant 10:21732178
64. Fuehner T, Kuehn C, Hadem J et al (2012) Extracorporeal membrane oxygenation in awake
patients as bridge to lung transplantation. Am J Respir Crit Care Med 185:763768
65. Hoopes CW, Kukreja J, Golden J et al (2013) Extracorporeal membrane oxygenation as a
bridge to pulmonary transplantation. J Thorac Cardiovasc Surg 145:862867
66. Annane D, Bellissant E, Cavaillon JM (2005) Septic shock. Lancet 365:6378
67. Landry DW, Oliver JA (2001) The pathogenesis of vasodilatory shock. N Engl J Med 345:588595
68. Dellinger RP (2003) Cardiovascular management of septic shock. Crit Care Med 31:946955
69. Hochstadt A, Meroz Y, Landesberg G (2011) Myocardial dysfunction in severe sepsis and
septic shock: more questions than answers? J Cardiothorac Vasc Anesth 25:526535
70. Hunter JD, Doddi M (2010) Sepsis and the heart. Br J Anaesth 104:311
71. Merx MW, Weber C (2007) Sepsis and the heart. Circulation 116:793802
72. Flynn A, Chokkalingam Mani B, Mather PJ (2010) Sepsis-induced cardiomyopathy: a review
of pathophysiologic mechanisms. Heart Fail Rev 15:605611
73. Vieillard-Baron A (2011) Septic cardiomyopathy. Ann Intensive Care 1:6
74. Pulido JN, Afessa B, Masaki M et al (2012) Clinical spectrum, frequency, and significance of
myocardial dysfunction in severe sepsis and septic shock. Mayo Clin Proc 87:620628
75. Aneja R, Carcillo J (2011) Differences between adult and pediatric septic shock. Minerva
Anestesiol 77:986992
76. Meyer DM, Jessen ME (1995) Results of extracorporeal membrane oxygenation in neonates
with sepsis. The Extracorporeal Life Support Organization experience. J Thorac Cardiovasc
Surg 109:419425
77. Meyer DM, Jessen ME (1997) Results of extracorporeal membrane oxygenation in children
with sepsis. The Extracorporeal Life Support Organization. Ann Thorac Surg 63:756761
[email protected]
190
M. Bombino et al.
78. Maclaren G, Butt W, Best D et al (2007) Extracorporeal membrane oxygenation for

refractory septic shock in children: one institutions experience. Pediatr Crit Care Med
8:447451
79. Bartlett RH (2007) Extracorporeal support for septic shock. Pediatr Crit Care Med
8:498499
80. Keckler SJ, Laituri CA, Ostlie DJ, St Peter SD (2010) A review of venovenous and venoarterial extracorporeal membrane oxygenation in neonates and children. Eur J Pediatr Surg
20:14
81. Fortenberry JD, Paden ML (2006) Extracorporeal therapies in the treatment of sepsis: experience and promise. Semin Pediatr Infect Dis 17:7279
82. Brierley J, Carcillo JA, Choong K et al (2009) Clinical practice parameters for hemodynamic
support of pediatric and neonatal septic shock: 2007 update from the American College of
Critical Care Medicine. Crit Care Med 37:666688
83. Goldman AP, Kerr SJ, Butt W et al (1997) Extracorporeal support for intractable cardiorespiratory failure due to meningococcal disease. Lancet 349:466469
84. Luyt DK, Pridgeon J, Brown J et al (2004) Extracorporeal life support for children with meningococcal septicaemia. Acta Paediatr 93:16081611
85. Creech CB, Johnson BG, Bartilson RE et al (2007) Increasing use of extracorporeal life support in methicillin-resistant Staphylococcus aureus sepsis in children. Pediatr Crit Care Med
8:231235
86. Maclaren G, Butt W (2007) Extracorporeal membrane oxygenation and sepsis. Crit Care
Resusc 9:7680
87. Rich PB, Younger JG, Soldes OS et al (1998) Use of extracorporeal life support for adult
patients with respiratory failure and sepsis. ASAIO J 44:263266
88. MacLaren G, Pellegrino V, Butt W et al (2004) Successful use of ECMO in adults with lifethreatening infections. Anaesth Intensive Care 32:707710
89. Firstenberg MS (2012) Contraindications to extracorporeal membrane oxygenation: are there
any absolutes? J Am Soc Echocardiogr 25:698
90. Dietl CA, Wernly JA, Pett SB et al (2008) Extracorporeal membrane oxygenation support
improves survival of patients with severe Hantavirus cardiopulmonary syndrome. J Thorac
91. Descheemaeker PN, Mira JP, Bruneel F et al (2009) Near-fatal multiple organ dysfunction
syndrome induced by Plasmodium malariae. Emerg Infect Dis 15:832834
92. Vohra HA, Adamson L, Weeden DF et al (2009) Use of extracorporeal membrane oxygenation
in the management of septic shock with severe cardiac dysfunction after Ravitch procedure.
Ann Thorac Surg 87:e4e5
93. Firstenberg MS, Blais D, Abel E et al (2010) Fulminant Neisseria meningitidis: role for extracorporeal membrane oxygenation. Heart Surg Forum 13:E376E378
94. MacLaren G, Cove M, Kofidis T (2010) Central extracorporeal membrane oxygenation for
septic shock in an adult with H1N1 influenza. Ann Thorac Surg 90:e34e35
95. Firstenberg MS, Abel E, Blais D et al (2010) The use of extracorporeal membrane oxygenation
in severe necrotizing soft tissue infections complicated by septic shock. Am Surg
76:12871289
96. Brchot N, Luyt CE, Schmidt M et al (2013) Venoarterial extracorporeal membrane oxygenation support for refractory cardiovascular dysfunction during severe bacterial septic shock.
97. Huang CT, Tsai YJ, Tsai PR, Ko WJ (2013) Extracorporeal membrane oxygenation resuscitation in adult patients with refractory septic shock. J Thorac Cardiovasc Surg 146:
10411046
98. Cheng A, Sun HY, Lee CW et al (2013) Survival of septic adults compared with nonseptic
adults receiving extracorporeal membrane oxygenation for cardiopulmonary failure: a
propensity-matched analysis. J Crit Care 28:532.e1e10
99. Zhao J, Wang D, Zhou X et al (2013) Hybrid ECMO using AvalonElite DLC for circulatory
support guarantees adequate heart/brain oxygen supply. J Heart Lung Transplant 32(4 suppl):
S117S118
[email protected]
16
191
100. Chou NK, Chen YS, Ko WJ et al (2001) Application of extracorporeal membrane oxygenation in adult burn patients. Artif Organs 25:622626
101. Sthr F, Emmert MY, Lachat ML et al (2011) Extracorporeal membrane oxygenation for
acute respiratory distress syndrome: is the configuration mode an important predictor for the
outcome? Interact Cardiovasc Thorac Surg 12:676680
102. Bratton SL, Davis RL (1997) Acute lung injury in isolated traumatic brain injury.
Neurosurgery 40:707712
103. Bein T, Scherer MN, Philipp A et al (2005) Pumpless extracorporeal lung assist (pECLA) in
patients with acute respiratory distress syndrome and severe brain injury. J Trauma
58:12941297
104. Arlt M, Philipp A, Voelkel S et al (2010) Extracorporeal membrane oxygenation in severe
trauma patients with bleeding shock. Resuscitation 81:804809
105. Bonacchi M, Spina R, Torracchi L et al (2013) Extracorporeal life support in patients with
severe trauma: an advanced treatment strategy for refractory clinical settings. J Thorac
Cardiovasc Surg. 145:16171626
106. Larsson M, Talving P, Palmr K et al (2010) Experimental extracorporeal membrane oxygenation reduces central venous pressure: an adjunct to control of venous hemorrhage? Perfusion
25:217223
107. Hill JD, OBrien TG, Murray JJ et al (1972) Prolonged extracorporeal oxygenation for acute
post-traumatic respiratory failure (shock-lung syndrome): use of Bramson membrane lung. N
Engl J Med 286:629634
108. Michaels A, Schriener RJ, Kolla S et al (1999) Extracorporeal life support in pulmonary
failure after trauma. J Trauma 46:638645
109. Perchinsky MJ, Long WB, Hill JG et al (1995) Extracorporeal cardiopulmonary life support
with heparin-bonded circuitry in the resuscitation of massively injured trauma patients. Am J
Surg 169:488491
110. Voelckei W, Wenzel V, Rieger M et al (1998) Temporary extracorporeal membrane oxygenation in the treatment of acute traumatic lung injury. Can J Anaesth 45:10971102
111. Zapol WM, Snider MT, Hill JD et al (1979) Extracorporeal membrane oxygenation in severe
acute respiratory failure. A randomized prospective study. JAMA 242:21932196
112. Morris AH, Wallace CJ, Menlove RL et al (1994) Randomized clinical trial of pressurecontrolled inverse ratio ventilation and extracorporeal CO2 removal for adult respiratory distress syndrome. Am J Respir Crit Care Med 149:295305
113. Peek GJ, Clemens F, Elbourne D et al (2006) CESAR: conventional ventilatory support vs
extracorporeal membrane oxygenation for severe adult respiratory failure. BMC Health Serv
Res 6:163
114. Brogan TV, Thiagarajan RR, Rycus PT et al (2009) Extracorporeal membrane oxygenation in
adults with severe respiratory failure: a multi-center database. Intensive Care Med
35:21052114
115. Muellenbach RM, Kredel M, Kunze E et al (2011) Prolonged heparin-free extracorporeal
membrane oxygenation in multiple injured acute respiratory distress syndrome patients with
traumatic brain injury. J Trauma 72:14441447
116. Sasadeusz KJ, Long WB, Kemalyan N et al (2000) Successful treatment of a patient with
multiple injuries using extracorporeal membrane oxygenation and inhaled nitric oxide.
J Trauma 49:11261128
117. Zimmermann M, Philipp A, Schmid FX et al (2007) From Baghdad to Germany: use of a new
pumpless extracorporeal lung assist system in two severely injured US soldiers. ASAIO J
53:e4e6
118. Bein T, Osborn E, Hofmann HS et al (2010) Successful treatment of a severely injured soldier
from Afghanistan with pumpless extracorporeal lung assist and neurally adjusted ventilatory
support. Int J Emerg Med 13:177179
119. Allan PF, Osborn EC, Bloom BB et al (2011) The introduction of extracorporeal membrane
oxygenation to aeromedical evacuation. Mil Med 176:932937
120. Bein T, Zonies D, Philipp A et al (2012) Transportable extracorporeal lung support for rescue
of severe respiratory failure in combat casualties. J Trauma Acute Care Surg 73:14481454
[email protected]
192
M. Bombino et al.
121. Huang YK, Liu KS, Lu MS et al (2009) Extracorporeal life support in post-traumatic respiratory distress patients. Resuscitation 80:535539
122. Madershahian N, Wittwer T, Strauch J et al (2007) Application of ECMO in multitrauma
patients with ARDS as rescue therapy. J Card Surg 22:180184
123. Yen TS, Liau CC, Chen YS, Chao A (2007) Extracorporeal membrane oxygenation resuscitation for traumatic brain injury after decompressive craniotomy. Clin Neurol Neurosurg
110:295297
124. Maif P, Hoermann C, Moertl M et al (1996) Percutaneous venoarterial extracorporeal membrane oxygenation for emergency mechanical circulatory support. Resuscitation 33:2934
125. Friesenecker BE, Peer R, Rieder J et al (2005) Craniotomy during ECMO in a severely traumatized patient. Acta Neurochir (Wien) 147:993996
126. Lisagor P, Cohen D, McDonnell B et al (1997) Irreversible shock revisited: mechanical support of the cardiovascular system: a case report and review. J Trauma 42:11821186
127. Repesse X, Au SM, Brechot N et al (2013) Recombinant factor VIIa for uncontrollable bleeding in patients with extracorporeal membrane oxygenation: report on 15 cases and literature
review. Crit Care 17:R55
128. Yank V, Tuohy CV, Logan AC et al (2011) Systematic review: benefits and harms of inhospital use of recombinant factor VIIa for off-label indications. Ann Intern Med
154:529540
129. Firstenberg MS, Nelson K, Abel E et al (2012) Extracorporeal membrane oxygenation for
complex multiorgan system trauma. Case Rep Surg. doi:10.1155/2012/897184
[email protected]
Left Ventricular Rest and Unloading

During VA ECMO
17
Gianluca Greco, Barbara Cortinovis, and Leonello Avalli
17.1
Introduction
End-diastolic volume of the left ventricle (LV) is closely related to overall ventricular performance, as described by a landmark of cardiac physiology, such as FrankStarling law. In the normal heart, myocardial fibers reach, at the end of diastole, a
predetermined length, which is a balance between the filling pressure from the left
atrium (LA) and the cardiac compliance. This is defined as preload of the LV. In the
failing heart, this relationship alters, causing progressive dilatation of the LV and
resulting in remodeling and worse systolic performance.
When LV contractility is profoundly impaired, inadequate right ventricular
drainage and bronchial circulation can lead to left ventricular distension, increasing
end-diastolic pressure and risk of pulmonary edema.
The purpose of ECMO support is, indeed, not only to achieve valid tissue perfusion indexes but also to provide ventricular unloading and wall tension control,
providing rest for the failing heart. Mechanical circulation of blood through the
cannulated vessels, per se, may not be sufficient to achieve this purpose, especially
if the ventricles are enlarged and stimulated with high-dose inotropic drugs.
G. Greco (*) B. Cortinovis

Cardiac Anesthesia and Intensive Care Unit,
Department of Emergency Medicine,
Milan, Italy, Via Pergolesi 33, 20900 Monza, Italy
L. Avalli
Cardiac Anesthesia and Intensive Care Unit,
Department of Urgency and Emergency,
San Gerardo Hospital, Via Pergolesi 33,
Monza (MB) 20900, Italy
[email protected]
193
194
G. Greco et al.
Moreover, increased ventricular end-diastolic pressure results in increased wall

stress and myocardial oxygen consumption, creating a vicious circle progressively
worsening left heart failure. When mechanical support is indicated for cardiogenic
shock of any etiology, it should be considered that inadequate right ventricular
drainage and bronchial circulation could also contribute to LV distension. It is
therefore of primary relevance to establish whether patients on mechanical support
experience adequate LV decompression. This is particularly important in peripherally inserted ECMO (rather than orthograde flowing centrally inserted mechanical
supports), in which it has been hypothesized that the combination of severe myocardial dysfunction and the adjunctive afterload, due to retrograde flow from the
arterial cannula, could dangerously increase LV end-diastolic, LA, and pulmonary
pressure (PA).
During VA ECMO, LV output must be carefully monitored to identify early
changes of outflow and to guarantee effective opening of the aortic valve.
Even brief periods of overdistention will lead to worsening of myocardial damage. First-line interventions that can improve LV output include inotropes (e.g.,
dobutamine) to increase contractility and intra-aortic balloon counterpulsation
(IABP) to reduce afterload and facilitate LV output.
Positive inotropic drugs act by different mechanisms to increase heart contractility and, as such, have been proposed for use in patients with cardiac failure
to increase myocardial contractility and hence cardiac output. Indeed, in the
short-term, acute phase of heart failure, inotropic drugs can be useful to increase
cardiac output. On the other hand, prolonged stimulation with long-term inotropic therapy has been shown to increase mortality rates in patients with chronic
heart failure.
Cardiac remodeling is manifested as a change in cardiac size, shape, and function. These modifications take place in response to either tissue injury or increase
load, and it is thought to be an important factor in disease progression. The main
clinical consequences of inadequate LV drainage could be summarized as
follows.
First, mechanical supports are aimed mostly to ensure organ perfusion during
profound cardiogenic shock but also adequate LV rest, allowing recovery of stunned
myocardium and relieving shear stress from the necrotic area. Intuitively, if the support fails to provide adequate decompression, this results in direct dilatation of heart
chambers, increasing oxygen consumption and ischemic damage but also adding
mechanical damage to infarct area that could result in severe complications such as
ventricular free wall rupture. Secondly, an enlarged, severely hypokinetic or akinetic cardiac chamber that doesnt provoke valve opening could cause blood stasis
and intracardiac clot formation.
Hence, it is of paramount importance to ensure an adequate LV drainage, both to
allow rest and recovery as well as to prevent severe life-threatening complications.
Several techniques have been described to drain LV, including IABP, surgical or
percutaneous venting, and axial impeller pumps. We will review the most relevant
techniques to decompress the LV in ECMO patients.
[email protected]
17
Left Ventricular Rest and Unloading During VA ECMO
17.2
195
Myocardial Dysfunction
Three major determinants of LV stroke volume and performance are the preload
(venous return and end-diastolic volume), myocardial contractility (the force generated at any given end-diastolic volume), and the afterload (aortic impedance and
wall stress). A thorough discussion of detailed physiology is beyond the scope of
this chapter. During myocardial ischemia it is possible to recognize different pathophysiological entities, determined by the extent and duration in coronary flow alteration. The presence of malfunction is fairly common and clinically manifested as an
alteration of the normal contractility of the heart wall.
Persistent, asymptomatic ischemia produces LV dysfunction that can mimic nonischemic causes of heart failure and is defined as hibernating myocardium.
Hibernation can be partially or completely restored to normal, either by improving
blood flow or by reducing oxygen demand [1].
Transient ischemia can lead as well to a period of persistent dysfunction, even
after the restoration of normal flow; this phenomenon is referred to as myocardial
stunning.
Postischemic dysfunction or myocardial stunning appears to develop in various
conditions involving the occurence of transient ischemia, like unstable angina, acute
myocardial infarction with early reperfusion, and cardiac surgery. Previous studies
investigating myocardial perfusion and systolic function have noticed a close relation between reduction of blood flow and failing of contractile performance.
Regional LV wall motion can persist for hours or days following reperfusion, despite
the absence of irreversible damage and despite restoration of normal coronary flow.
This tissue is still viable and the contractile abnormality is supposed to be reversible. The amount of flow which is normal at rest may not be adequate during exercise and there may be transmural variations in myocardial blood flow. Thus, it is
possible that areas of dysfunction secondary to stunning and hypoperfusion may
coexist within the same contractile area.
It has been observed in animals that a period of ischemia <3 h causes infarction
of subendocardial portion of the interested region, whereas quantities of subendocardial tissue remained viable. The severity of stunning was greater in the inner
layers of the left ventricular wall than in the outer layers, and this subendocardial
tissue salvaged by reperfusion may require days or weeks to recover its contractile
function.
Early reperfusion during acute myocardial infarction results in an admixture of
thickened and stunned subendocardium. Factors that determine severity of cardiac
dysfunction include size of ischemic region and loading condition of the heart. The
process probably involves multiple factors, e.g., abnormal calcium homeostasis and
oxidative stress among others [24]. Myocardial stunning is an important cause of
post-resuscitation circulatory failure. Transitional global myocardial ischemia and
profound depression of LV function are common after resuscitation maneuvers, and
the resulting myocardial dysfunction has been documented in both animal and clinical
studies. The compromise in systolic LV function is manifested by the decreased LV
[email protected]
196
G. Greco et al.
ejection fraction, decreased fractional shortening, and decreased peak systolic LV

pressure/end-systolic volume ratio. This kind of stunning can take place with no evidence of infarction, and both systolic contraction and diastolic relaxation remain
impaired after myocardial blood flow is restored [510]. In some patients, areas of
persistent ischemia can produce as well LV dysfunction, creating a condition of
hibernating myocardium where coronary blood flow is chronically reduced. As
blood flow is reduced, there is a corresponding reduction in contractile performance,
and this coupling of low perfusion and diminished performance can take place with no
ischemic symptoms or necrosis. In particular, this response of the heart coping with a
reduced myocardial blood flow has been considered as an act of self-preservation. If
blood flow remains low, the myocardium may be able to reduce its metabolic requirements still further, by undergoing a more chronic form of adaptation involving alterations in the morphology and protein content of the myocardium [1113].
Whatever the cause, stunning and hibernation may cause significant cardiac failure. Structural remodeling would be necessary to restore contractility and adequate
support is often necessary to achieve a satisfactory organ perfusion. As a result,
chronically impaired but viable myocardium may take weeks to months to recover
once flow is restored [14, 15].
17.3
Unloading in Chronic Heart Failure
Data collected in patients with chronic heart failure and left ventricular assist device
(LVAD) show that myocardial unloading with an LVAD results in long-term recovery. The underlying process of myocardial recovery is still not completely understood, but previous clinical observations have shown that even cardiac patients with
idiopathic dilated cardiomyopathy (ICD) subjected to strict measures of bed rest for
long time showed improvement in symptoms [16]. Different theories suggest influences on the microvasculature, fibrosis, inflammation, and structural and cardiac
remodeling. Furthermore, the combination of unloading and optimal coronary perfusion could facilitate a reduction in myocardial cytokines and a decrease in neurohumoral activity [1719]. It has been observed that the heart can grow hypertrophic
cardiomyocytes to reduce stress on the failing ventricular wall. Hetzer et al. showed
that lasting recovery can be reached by ventricular unloading in a subset of patients
with IDC, where LVAD unloading has been shown to induce regression of cardiomyocyte hypertrophy [20]. Klotz noted that LVAD support induced reverse structural remodeling of the heart, reducing LV size and myocyte dimensions and
improving chamber stiffness [21]. Drakos collected hemodynamic data and LV tissue with digital microscopy coupled with ultrastructural and functional evaluation,
speculating that pulsatile mechanical unloading of the failing heart increases microvascular density. The vascular changes were accompanied by increased fibrosis and
reduced cardiomyocyte hypertrophy without any structural or metabolic evidence
of outright degeneration and atrophy [22]. Even the beneficial effects of drug therapy with vasodilator therapy and angiotensin-converting enzyme inhibitor in afterload reduction on IDC hearts may be seen in the same context of LV unloading,
[email protected]
17
197
interrupting the links between decreasing in cardiac performance and increasing in

systemic vascular resistance and LV filling pressure [23].
Despite previous observation, direct effect of mechanical unloading on myocardial endothelium and microvasculature is unknown and its effect on the degree of
hypertrophy regression is controversial. Pathophysiological models have focalized
attention on the role of excess load in driving a vicious circle of cardiac remodeling.
Mechanical unloading would interrupt this cycle improving function of the failing
heart. Recovery, if it occurs, is reached within a few weeks. Beyond this time, a
process of gradual loss of ventricular function seems to take place [2426].
17.4
Intra-aortic Balloon Counterpulsation
IABP is the most widely used form of mechanical hemodynamic support. AHA and
European guidelines, respectively, give class IB and class IC recommendation to the
use of an intra-aortic balloon in the treatment of cardiogenic shock [27, 28].
Nonetheless, the evidence provided is based mostly on registry data, whereas adequately powered randomized trials demonstrating the efficacy of IABP, beyond its
physiological theoretical benefits, are still lacking.
Indication to clinical use of IABP includes [29, 30] mostly cardiogenic shock of
any etiology, especially in complicating acute coronary syndrome, intractable
arrhythmias, and adjunctive therapy in high-risk procedures. There is no sufficient
literature, to our knowledge, to extrapolate definitive recommendations about the
use of IABP in ECMO patients, and the controversies divide authors as supporters
and nonsupporters, based on a single centers experience and clinical protocols.
Well briefly summarize pros and cons of IABP in ECMO and its physiological
effects in terms of mechanical benefits and endothelial function, and well make an
effort to provide some suggestions based on the presented literature.
Cyclic inflation and deflation of the balloon that provided significant interindividual variability (due to balloon size and position and physiological variables, such
as heart rate and rhythm, compliance of the aorta, and systemic resistance) carry
two major consequences: a displacement of blood through the proximal aorta during diastole and a reduction of afterload during systole through a vacuum effect due
to rapid deflation of the balloon [31, 32]. Experimental and clinical studies suggest
that afterload reduction and diastolic augmentation improve antegrade flow in coronary arteries, thus resulting in increased blood supply to territory perfused by a critically stenotic vessel [33], whereas hemodynamic effects on cardiac output are
modest and do not impact on overall mortality.
The effect of IABP on coronary flow is still largely debatable: some studies have
found little or no change in coronary blood flow, while others noted a significant
augmentation [3436].
Improved blood supply is higher where maximized autoregulation determines
pressure-dependent perfusion and coronary vessel is fully dilated by ischemia. At
lower perfusion pressures, IABP could increase blood supply even when coronary
flow to the territory affected by the stenosis cannot be maintained. It is important to
[email protected]
198
G. Greco et al.
mention that no improvement in coronary perfusion could be found distally to critical stenoses (>95 %) [37].
As far as mechanical effects, though, despite evidence, a theoretical benefit may
be hypothesized, especially in acute mechanical abnormalities, such as mitral regurgitation (MR), ventricular septal defect (VSD), and non-pulsatile flow, resulting
from ECMO. In fact, reported beneficial effects of IABP are a reduction of heart
rate and mean pulmonary capillary wedge pressure, increase in cardiac output
(especially when mechanical complications arise), and better perfusion of large territory of refractory ischemia. More interestingly, IABP reduces mean systemic
impedance and systolic pressure, resulting in 14 % decline in calculated peak left
ventricular wall stress [38]. This may represent the exact goal of afterload and wall
stress reductions that is desirable to achieve in ECMO patients with otherwise suboptimal LV protection.
Despite intuitive physiology, IABP does not, at present days, encounter global
consensus, and no randomized trial is available to our knowledge to specifically
address the issue of using IABP and ECMO. The most relevant literature supporting
combined use of IABP and ECMO includes the following works.
Madershahian and colleagues showed that IABP in refractory cardiogenic shock,
during non-pulsatile ECMO flow, may be beneficial in terms of coronary flow, graft
patency in the early postoperative period, and compensation for lower ECMO pump
flow to maintain equivalent bypass graft flow. However, the small number of patients
does not allow conclusive evidence. Phillips et al. reported that a combination of
peripherally inserted ECMO and IABP, in 16 cardiogenic shock patients, provided
greater hemodynamic support and pulsatile flow during diastole, increasing coronary blood flow and allowing LV systolic decompression [39]. Lazar and associates
[40] demonstrated a reduction of infarct size and reversal of hemodynamic deterioration, less tissue acidosis, higher wall motion scores, and the least amount of
necrosis.
The last two groups of authors encounter our favor in supporting the use of IABP
during ECMO: they suggest that application of these modalities can be readily instituted in emergency situations and produces optimal recovery of acutely ischemic
myocardium, concluding that ECMO should always be used in conjunction with
IABP support. They also proposed a staged protocol in mechanical support and
weaning, by inserting first IABP and subsequently ECMO, and the opposite in
weaning. Moreover, IABP may accelerate weaning from ECMO resulting in less
heparinization and potential bleeding.
Finally, ONeil et al. reported that pulsatile is superior to non-pulsatile perfusion
in preserving the microcirculation and decreasing systemic inflammatory response
during cardiopulmonary bypass (CPB), thus potentially improving outcomes in
high-risk cardiac surgical procedures requiring prolonged CPB time [41]. This
could be applied to ECMO patients, who are both in need of prolonged mechanical
support, even if less pro-inflammatory than standard CPB, and are in the high-risk
category.
Potential complications of IABP are well known to all physicians dealing with
mechanical support for cardiogenic shock, and despite careful management, adverse
[email protected]
17
199
events are to some extent still unavoidable. Vascular complications (625 % of

cases) are the major risk, including limb and visceral ischemia, vascular laceration,
and major hemorrhage. It is mandatory to promptly address the issue of surgical
repair as soon as the complication arises. Ultrasonography assessment of arterial
vessels and guidance during placement, sheathless insertion, and adequate sizing of
the balloon could help in minimizing complications.
If we do not support widespread use of IABP in all ECMO patients, we favor its
placement for mechanical complications (MR, VSD, etc.) and for patients in which
even a minimally pulsatile flow could not be achieved. IABP for weaning is discussed elsewhere.
17.5
Venting
Surgical Vent. Surgical left heart decompression is very common especially in valvular surgery and can be accomplished through insertion of an LA or LV vent. The
cannula can easily be placed by the surgeon under direct vision, either in the LA or
more commonly in the right superior pulmonary vein: this serves mostly to improve
surgical field vision, by suctioning blood, and to prevent excessive ventricular distention. If longer-term support is anticipated, LV apical cannulation can also be a
valuable alternative, and the need for sternal diastasis with central cannulation may
influence the decision of whether or not to tunnel the vent.
If ECMO is initiated peripherally, decompression of the left ventricle is, to some
extent, more difficult. In settings where the chest is not already open, decision for
sternotomy/thoracotomy should carefully weigh the risk-benefit of a centrally
inserted vent versus the risk of associated bleeding complications due to systemic
anticoagulation. For this reason, a number of percutaneous techniques have been
described: balloon and combined blade and balloon atrial septostomy under transthoracic (TTE) or transesophageal echocardiographic (TEE) guidance, transseptal
sheath placement, transaortic cannula, percutaneous transjugular pulmonary artery
venting, and impeller pumps.
17.6
Balloon and Combined Blade and Balloon

Atrial Septostomy
Koenig et al. [42] described transvenous balloon atrial septostomy in four pediatric
patients with cardiogenic shock from myocarditis. The technique is historically a
modification of Rashkind procedure used to create atrial septal defects in children
with transposition of great arteries.
Although theoretically beneficial, the procedure may pose some serious challenges: technically, if balloon septostomy is relatively easy to create in infants and
toddlers due to soft atrial septum, this is far more challenging in older patients, in
which septal thickening requires often the use of a blade technique, to achieve unrestricted left-to-right atrial flow. Moreover, blade septostomy, which is a valuable
[email protected]
200
G. Greco et al.
alternative in difficult cases, often requires the use of fluoroscopic guidance and
therefore exposes patients to potential adverse effects due to transfer and, especially
in children, to radiation exposure. In fact, in Koenigs report, one procedure did not
result in a hemodynamically significant atrial septal defect, and the older patient
required blade septostomy to be performed under fluoroscopic guidance. Moreover,
this procedure is technically feasible at the bedside only if a patent foramen ovale
exists.
Later Johnston et al. [43] reported a 10 year-old patient, requiring decompression
for pulmonary hemorrhage due to cardiogenic shock with severely dilated LV with
an intracavitary thrombus. Drainage was achieved by creating an atrial septal defect,
with transseptal puncture followed by progressive balloon dilatation, under transesophageal echocardiographic guidance. The advantages presented by Johnston
et al. include the feasibility of this technique at the bedside, under TEE guidance,
without risks and disadvantages due to fluoroscopy in catheterization laboratory and
the fact that the procedure can be performed regardless the atrial anatomy. TEE
provides adequate visualization of the needle and intracavitary thrombi, if present,
and positioning can be precisely assessed through different projections by an expert
echocardiographer. The transseptal balloon technique is relatively safe even in anticoagulated patients. Patency of the newly created septal defect was followed up to
a week after procedure, showing no decrease in size.
Supporters of balloon or blade and balloon techniques underline the importance
of achieving adequate results, without the need of additional cannulae and complications of the ECMO circuit.
17.7
Transseptal Cannulation
Balloon and blade and balloon techniques have been used to effectively decompress
the left heart chambers and to relieve hypertension from pulmonary circulation. Due
to technical difficulties in achieving unrestricted left-to-right atrial flow, several
options have been subsequently presented by different authors. In particular, addition of a transseptal cannula provides not only optimal decompression but also a
potential benefit during weaning, when test-occluding the left heart cannula first
allows observation of heart function.
Ward et al. [44] described a 7-Fr long introducer into the left atrium using TEEguided transseptal puncture. The introducer was subsequently connected to the
venous circuit to achieve decompression. The following techniques are very similar
to this, adding an ECMO cannula to provide even better drainage.
Aiygari and colleagues [45] reviewed seven patients undergoing LA drain procedures on ECMO, focusing on procedural feasibility, complications, and success in
alleviating LA hypertension. Their hypothesis was to test if percutaneous insertion
of a transseptal sheath incorporated into the ECMO venous circuit was a feasible
alternative to surgical venting, in draining LA. Femoral vein access was used in all
patients using modified Seldinger technique. Under fluoroscopic guidance in cardiac catheterization laboratory, transseptal puncture was performed and a
[email protected]
17
201
transseptal sheath was positioned to be exchanged over super-stiff guidewire with a

larger cannula, connected in-line into the venous limb of the ECMO circuit.
Cannulae were chosen based on size and availability. Decannulation was performed
bedside in the intensive care unit with manual pressure applied to site until adequate
hemostasis. This resulted in prompt resolution of radiographic and echocardiographic findings of LA hypertension and left heart chamber dilation. Flow rate
depends on cannula diameter and length and on LA pressure, of which only the first
is to some extent modifiable. In fact adequate sizing of LA drain to the largest possible not only allows to achieve satisfactory decompression but also seems to provide a more rapid and profound resolution of LA hypertension. Higher maximum
indexed LA cannula flow over a 96 h period seemed to correlate with procedural
success. Moreover, the authors suggest a minimum 8-Fr sheath in infants and toddlers, 1012-Fr sheaths in larger children, and 1415-Fr sheaths in adult-sized
patients. This procedure represents a reasonable alternative to blade or balloon atrial
septostomy, which is an independent risk factor for mortality in the creation of atrial
defects [46].
Schwartz et al. [47] reported a similar technique, by advancing a percutaneous
femoral transseptal cannula in the left atrium, under transesophageal echocardiography guidance in a 13 year-old patient with cardiogenic shock rescued by
peripherally inserted ECMO. The transseptal cannula was connected with the
venous limb of ECMO circuit as previously described by Aiyagari and colleagues.
In this report, authors demonstrated an immediate benefit in terms of LA and LV
decompression, with a modified technique, feasible at the bedside with transesophageal echocardiographic guidance, which further simplifies the transseptal
cannulation already described under fluoroscopic guidance, without the need of
transferring a potentially unstable patient. In conclusion, this technique could be
considered (either fluoroscopy or TEE guided, according to expertise of single
centers) in the management of any ECMO patient, showing signs of progressive
LV dilatation and LA and PA hypertension, in which thoracotomy/sternotomy is
not anticipated.
17.8
Percutaneous Pulmonary Artery Venting
In 1988 Kolobow et al. [48] reported, in a sheep model, effective left heart decompression with a modified Swan-Ganz catheter by causing pulmonary valve
insufficiency.
Avalli and coworkers [49] described a different technique, to some extent, based
on the same physiological principles, by using a 6-Fr angiographic catheter that was
introduced through the right jugular vein with a modified Seldinger technique and
advanced in the right pulmonary artery (Fig. 17.1). The catheter was subsequently
changed over a super-stiff angiographic guidewire with a 15-Fr venous cannula,
which was advanced and positioned in the common pulmonary artery. At the end of
the procedure, the guidewire was removed and the cannula connected to the venous
limb of the ECMO circuit. The procedure was fluoroscopy guided at the bedside.
[email protected]
202
G. Greco et al.
Fig. 17.1 Transpulmonary venting
Hemodynamics showed a decrease in the PA and LA pressures. At the time of

weaning, the cannula was withdrawn to the superior vena cava and subsequently
removed. No issues were reported in maintaining proper positioning of the venting
cannula in place. Modified pulmonary cannulation offers several advantages over
the previously mentioned methods: in fact surgical or even percutaneous left heart
decompression by septostomy may carry a high risk of bleeding (37 %) and damages due to the manipulation of the heart. It can be accomplished at bedside and it
is far less expensive and easier to place and manage than axial flow pumps, which
will be discussed further in the appropriate section. Our group privileges this technique over the others, due to its relatively lower invasiveness and the practical
advantages in performing the venting at bedside anytime with minimal patient preparation, lower cost, and steep learning curve of the trained personnel.
17.9
Transaortic Catheter Venting
Possibility exists also to directly drain left heart chambers. Fumagalli et al. [50]
reported draining blood from the LV, through a percutaneously placed transaortic cannula, pumping directly into the femoral artery, with a normalization of left heart filling
pressures, and resolution of pulmonary edema, as a bridge to heart transplantation.
Several experimental animal models reported a significant reduction of LV preload, in peripherally inserted ECMO, comparing pre- and post-transaortic cannula
insertion in one study [51] and a significant reduction in LV total energy and work
in a second publication. The LV energetic charge was significantly increased by a
combination of transaortic cannula and peripheral ECMO. A third study compared
four different conditions: baseline, during isolated ECMO, ECMO with transaortic
venting cannula, and a combination to the previous two with IABP, showing that
venting reduced LV energy and work, compared with other techniques alone [52].
[email protected]
17
203
17.10 Impeller Pumps

The Impella is a minimal intra-aortic impeller blood pump, a form of minimally invasive LV assist device (LVAD) that can be used to support the cardiogenic shock patient
in the short and medium term. It can be positioned in the LV, via open chest, but also
advanced through the femoral artery over a guidewire until reaching its definitive position. It is designed as a support device, but several works report its use as an adjunctive
mechanical support for ECMO patients, in which LV decompression is indicated. Its
main effects can be summarized as follows: it directly unloads the LV and it reduces
myocardial workload and oxygen consumption while increasing cardiac output and
coronary and end-organ perfusion. Chaparro and colleagues [53] reported for the first
time the use of combined Impella and ECMO for biventricular and respiratory failure,
as a bridge to recovery, in a myocarditis patient. Beuthered et al. [54] also reported the
case of a 34-year-old woman with fulminant myocarditis needing ECMO support and
subsequently an Impella device to decompress the LV, for acute pulmonary edema.
Although reporting effective unloading of the LV, the paper also underlines the occurrence of Impella pump failure, to underline technical challenges in managing complex
mechanical assistance. Koeckert et al. [55] reported successful use of Impella device
for ECMO complicated by pulmonary edema, with weaning for myocardial recovery.
Vlaesselars [56] also reported similar combination in a pediatric patient with congenital cardiomyopathy. Interestingly, the procedure was echocardiography guided.
17.11 How to Assess Effective Decompression?

There is no established gold standard to evaluate optimal LV decompression.
In literature, the most frequently assessed parameters include echocardiographic
inspection of the heart chamber size, aortic valve opening, and Doppler evaluation
of flow velocities. Most often, published literature reports effective LV decompression clinical resolution of symptoms, such as disappearance of pulmonary edema
and hemoptysis and preload pressure reduction. This represents the clinical goal
that is desirable to achieve. We will not further discuss hemodynamic monitoring
but refer the reader to the specific section.
Nonetheless, it is worthwhile to mention the possibility of TEE evaluation of the
proximal part of left coronary arteries and estimation of blood flow velocity in the
left anterior descending (LAD) coronary artery, by means of pulsed Doppler. The
left main is visualized as an echo-free space, by placing the transducer just above
the aortic leaflets, with adjustments needed to fully visualize the length of the vessel. Once the Y-shaped bifurcation between LAD and circumflex artery is visualized, coronary blood flow velocity can be evaluated by pulsed Doppler. There are no
reports, to our knowledge, of this technique applied to ECMO patients. Although
coronary arteries might be more challenging to visualize due to chamber decompression, still an attempt to visualize non-pulsatile or pulsatile (especially in combination with IABP or native valve opening) flow on the left main coronary artery
may be useful to assess perfusion of a recovering heart.
[email protected]
204
G. Greco et al.
Although the measurement of cardiac biomarkers may seem appealing to evaluate LV rest and unloading as well as to prognosticate outcome, Luyt et al. [57]
reported that serial measurement of N-terminal fragment of the B-type natriuretic
peptide and troponin I-C and midregional fragment of the proatrial natriuretic peptide, proadrenomedullin, and copeptin have no role as prognostic markers in refractory cardiogenic shock patients rescued by ECMO.
In conclusion, a combination of clinical and radiographic resolution, hemodynamic parameters, and standard echocardiography is up to the present days the standard to evaluate effective LV rest and decompression.
References
1. Braunwald E, Rutherford JD (1986) Reversible ischemic left ventricular dysfunction: evidence
for the hibernating myocardium. J Am Coll Cardiol 8:146
2. Rahimtoola SH et al (1993) The hibernating myocardium in ischaemia and congestive heart
failure. Eur Heart J 14(Suppl A):2226
3. Ito H, Tomooka T, Sakai N (1993) Time-course of functional improvement in stunned myocardium in risk area in patients with reperfused anterior infarction. Circulation 87:355362
4. Bolli R (1992) Myocardial stunning in man. Circulation 86:16711691
5. Zia A, Kern B (2011) Management of postcardiac arrest myocardial dysfunction. Curr Opin
Crit Care 17:241246
6. Gazmuri RJ, Weil MH, Bisera J (1996) Myocardial dysfunction after successful resuscitation
from cardiac arrest. Crit Care Med 24:9921000
7. Laurent I, Monchi M, Chiche JD et al (2002) Reversible myocardial dysfunction in survivors
of out-of-hospital cardiac arrest. J Am Coll Cardiol 40:21102116
8. Peatfield RC, Sillet RW, Taylor D (1977) Survival after cardiac arrest in the hospital. Lancet
1:12231225
9. Deantonio HJ, Kaul S, Lerman BB (1990) Reversible myocardial depression in survivors of
cardiac arrest. Pacing Clin Electrophysiol 13:982985
10. Kern KB et al (1996) Myocardial dysfunction after resuscitation from cardiac arrest: an example of global myocardial stunning. J Am Coll Cardiol 28(1):232240
11. Ross J Jr (1991) Myocardial perfusion-contraction matching. Implications for coronary heart
disease and hibernation. Circulation 83:10761083
12. Gallagher KP (1995) Myocardial hibernation in terms of the flow-function relationship. Basic
Res Cardiol 90:1215
13. Braunwald E (1986) Reversible ischemic left ventricular dysfunction: evidence for the hibernating myocardium. J Am Coll Cardiol 8:14671470
14. Schinkel AF, Bax JJ, Delgado V (2010) Clinical relevance of hibernating myocardium in ischemic left ventricular dysfunction. Am J Med 123:978986
15. Heusch G, Schulz R, Rahimtoola SH (2005) A myocardial hibernation: a delicate balance. Am
J Physiol Heart Circ Physiol 288(3):H984H999
16. Burch GE, McDonald CD (1971) Prolonged bed rest in treatment of ischemic cardiomyopathy.
Chest 60(5):424430
17. Hummel M (1994) Interleukin-6 and interleukin-8 concentrations as predictors of outcome in
ventricular assist device patients before heart transplantation. Crit Care Med 22:448454
18. Hosenpud JD (1989) Interleukin-1 induced myocardial depression in an isolated perfused
beating heart preparation. Heart Transplant 8:460464
19. Hill JA, Olson EN (2008) Cardiac plasticity. N Engl J Med 358:13701380
20. Hetzer R et al (1999) Cardiac recovery in dilated cardiomyopathy by unloading with a left
ventricular assist device. Ann Thorac Surg 68:742749
[email protected]
17
205
21. Klotz S, Foronjy RF, Dickstein ML (2005) Mechanical unloading during left ventricular assist
device support increases left ventricular collagen cross-linking and myocardial stiffness.
22. Drakos SG, Kfoury AG, Hammond EH et al (2010) Impact of mechanical unloading on microvasculature and associated central remodeling features of the failing human heart. J Am Coll
Cardiol 56:382391
23. Cohn JN (1986) Effect of vasodilator therapy on mortality in chronic congestive heart failure.
Results of a Veterans Administration Cooperative Study. N Engl J Med 314:15471552
24. Scheinin SA (1992) The effect of prolonged left ventricular support on myocardial histopathology in patients with end- stage cardiomyopathy. ASAIO J 38:M271M274
25. Kinoshita M (1996) Influence of prolonged ventricular assistance on myocardial histopathology in intact heart. Ann Thorac Surg 61:640645
26. Gerdes AM (2002) Cardiac myocyte remodeling in hypertrophy and progression to failure. J
Card Fail 8:S24S268
27. Kushner FG, Hand M, Smith SC Jr et al (2009) 2009 focused updates: ACC/AHA guidelines
for the management of patients with ST-elevation myocardial infarction (updating the 2004
guideline and 2007 focused update) and ACC/AHA/SCAI guidelines on percutaneous coronary intervention (updating the 2005 guideline and 2007 focused update) a report of the
American College of Cardiology Foundation/American Heart Association Task Force on
Practice Guidelines. J Am Coll Cardiol 54:22052241
28. Steg G, James SK, Atar D et al (2012) ESC guidelines for the management of acute myocardial
infarction in patients presenting with ST-segment elevation. Eur Heart J 33(20):25692619
29. Santa-Cruz RA, Cohen MG, Ohman EM (2006) Aortic counterpulsation: a review of the
hemodynamic effects and indications for use. Catheter Cardiovasc Interv 67:6877
30. Ferguson JJ 3rd, Cohen M, Freedman RJ Jr et al (2001) The current practice of intra-aortic balloon counterpulsation: results from the Benchmark Registry. J Am Coll Cardiol 38:14561462
31. Weber KT, Janicki JS (1974) Intraaortic balloon counterpulsation. A review of physiological
principles, clinical results, and device safety. Ann Thorac Surg 17:602636
32. Marchionni N, Fumagalli S, Baldereschi G et al (1998) Effective arterial elastance and the
hemodynamic effects of intraaortic balloon counterpulsation in patients with coronary heart
disease. Am Heart J 135:855861
33. Port SC, Patel S, Schmidt DH (1984) Effects of intraaortic balloon counterpulsation on myocardial blood flow in patients with severe coronary artery disease. J Am Coll Cardiol
3:13671374
34. Williams DO, Korr KS, Gewirtz H, Most AS (1982) The effect of intraaortic balloon counterpulsation on regional myocardial blood flow and oxygen consumption in the presence of coronary artery stenosis in patients with unstable angina. Circulation 66:593597
35. Mueller H, Ayres SM, Conklin EF et al (1971) The effects of intra-aortic counterpulsation on
cardiac performance and metabolism in shock associated with acute myocardial infarction. J
Clin Invest 50:18851900
36. Kern MJ, Aguirre FV, Tatineni S et al (1993) Enhanced coronary blood flow velocity during
intraaortic balloon counterpulsation in critically ill patients. J Am Coll Cardiol 21:359368
37. Kern MJ, Aguirre F, Bach R et al (1993) Augmentation of coronary blood flow by intra-aortic
balloon pumping in patients after coronary angioplasty. Circulation 87:500511
38. Urschel CW, Eber L, Forrester J et al (1970) Alteration of mechanical performance of the
ventricle by intraaortic balloon counterpulsation. Am J Cardiol 25:546551
39. Phillips SJ, Zeff RH, Kongtahworn C et al (1992) Benefits of combined balloon pumping and
percutaneous cardiopulmonary bypass. Ann Thorac Surg 54:908910
40. Lazar HL, Treanor P, Yang M et al (1994) Enhanced recovery of ischemic myocardium by
combining percutaneous bypass with intraaortic balloon pump support. Ann Thorac Surg
57:663668
41. ONeil MP, Fleming JC, Badhwar A et al (2012) Pulsatile versus nonpulsatile flow during
cardiopulmonary bypass: microcirculatory and systemic effects. Ann Thorac Surg 94:
20462053
[email protected]
206
G. Greco et al.
42. Koenig P, Ralston M, Kimball T, Meyer R, Daniels S, Schwartz D (1993) Balloon atrial septostomy for left ventricular decompression in patients receiving extracorporeal membrane
oxygenation for myocardial failure. J Pediatr 122:S95S99
43. Johnston TA, Jaggers J, McGovern JJ et al (1999) Bedside transseptal balloon dilation atrial
septostomy for decompression of the left heart during extracorporeal membrane oxygenation.
Catheter Cardiovasc Interv 46:197199
44. Ward KE, Tuggle DW, Gessouroun MR et al (1995) Transseptal decompression of the left
heart during ECMO for severe myocarditis. Ann Thorac Surg 59:749751
circuit. Crit Care Med 34:26032606
46. Veldtman GR, Norgard G, Whlander H et al (2005) Creation and enlargement of atrial defects
in congenital heart disease. Pediatr Cardiol 26:162168
47. Swartz MF, Smith F, Byrum CJ et al (2012) Transseptal catheter decompression of the left
ventricle during extracorporeal membrane oxygenation. Pediatr Cardiol 33:185187
48. Kolobow T, Rossi F, Borelli M, Foti G (1988) Long-term closed chest partial and total cardiopulmonary bypass by peripheral cannulation for severe right and/or left ventricular failure,
including ventricular fibrillation. The use of a percutaneous spring in the pulmonary artery
position to decompress the left heart. ASAIO Trans 34:485489
adult patients. ASAIO J 57:3840
by venoarterial ECMO and transaortic left ventricular venting. Int J Artif Organs
27(5):410413
51. Kitamura M, Hanzawa K, Takekubo M et al (2004) Preclinical assessment of a transaortic
venting catheter for percutaneous cardiopulmonary support. Artif Organs 28(3):298302
52. Morishita A, Kitamura M, Shibuya M et al (1999) Effectiveness of transaortic venting from a
failing left ventricle during venoarterial bypass. ASAIO J 45(1):6973
53. Chaparro SV, Badheka AA, Marzouka GR et al (2012) Combined use of Impella left ventricular assist device and extracorporeal membrane oxygenation as a bridge to recovery in fulminant myocarditis. ASAIO J 58(3):285287
54. Beurtheret S, Mordant P, Pavie A et al (2012) Impella and extracorporeal membrane oxygenation: a demanding combination. ASAIO J 58:291293
55. Koeckert MS, Jorde UP, Naka Y et al (2011) Impella LP 2.5 for left ventricular unloading during venoarterial extracorporeal membrane oxygenation support. J Card Surg 26(6):666668
56. Vlasselaers D, Desmet M, Desmet L et al (2006) Ventricular unloading with a miniature axial
flow pump in combination with extracorporeal membrane oxygenation. Intensive Care Med
32(2):329333
57. Luyt CE, Landivier A, Leprince P et al (2012) Usefulness of cardiac biomarkers to predict
cardiac recovery in patients on extracorporeal membrane oxygenation support for refractory
cardiogenic shock. J Crit Care 27(5):524.e7524.e14
[email protected]
Weaning from Extracorporeal

Circulatory Support
18
Anna Coppo, Lucia Galbiati, and Gianluigi Redaelli
18.1
Introduction
Extracorporeal membrane oxygenation provides effective circulatory support while

waiting for cardiac recovery in patients with potentially reversible heart disease, or
for heart transplantation, or for implantation of a ventricular assist device (VAD) in
patients with terminal heart disease [14].
Finding the optimal time for weaning patients from extracorporeal support is one
of the greatest challenges in ECMO management. Given the potential risk of
ECMO-related complications (bleeding, limbs ischemia, neurological complications, organ hypoperfusion, infections), early identification of weanable/notweanable patients is a target of the clinician in order to achieve decannulation or
indication to long-term assistance as soon as possible.
Many studies report that ECMO duration and any related complication are predicting factors of postweaning mortality [5, 6]. To identify potentially weanable
patients soon would be of the greatest importance in order to limit the duration of
futile assistances and direct patients to transplant lists or long-term assistance program without other additional failures or infective complications. Clear strategies to
optimize outcomes and minimize futile support have not been established.
18.2
Predictors of Successful Weaning
The original indication to ECMO and the individual patients conditions should
guide the strategy and optimal time for weaning. Clinical, hemodynamic, and
Doppler-echocardiographic parameters are evaluated at least on a daily basis and
A. Coppo (*) L. Galbiati G. Redaelli
San Gerardo Hospital, University of Milano-Bicocca, Milan, Italy,
[email protected]
207
208
A. Coppo et al.
contribute to the decision of whether and when to perform a weaning trial. The
search for predictors of successful weaning from ECMO combines different
strategies.
18.2.1 Biomarkers
No prognostic cardiac markers of myocardial recovery are established in patients
with refractory cardiogenic shock requiring circulatory support. Luyt et al. [7]
recently found that NT-proBNP, TnIc, MR-proANP, proADM, and copeptin levels
are high in patients with refractory cardiogenic shock requiring ECMO support, but
their kinetics during the first week of ECMO support are not predictive of cardiac
recovery. Moreover, there are many noncardiac reasons for cardiac biomarkers concentrations rise in ICU patients: sepsis with or without shock, other non-cardiogenic
shock, and multiorgan failure (MOF) with cardiac involvement.
Blood lactate levels are considered a good indicator of organ perfusion. With
regard to ECMO in the literature, different and sometimes contradictory interpretations of the values of blood lactate are reported [3, 810]. Low arterial blood pH
value and high lactates are described as independent risk factors for mortality after
extracorporeal CPR using ECMO, which most likely mirrors a longer duration of
low cardiac output before ECMO initiation [11].
However, we judge reasonable to monitor the kinetic of cardiac biomarkers in
patients requiring ECMO for ischemic heart disease or myocarditis and start the
weaning trial after the drop of the cardiac biomarkers. Besides this we routinely
monitor blood lactate levels and SvO2 during the weaning trial as indexes of the
adequacy of oxygen delivery (DO2) related to oxygen uptake (VO2).
18.2.2 Echocardiography
Echocardiography plays a fundamental role throughout the entire journey of a
patient on ECMO and helps in monitoring cardiac recovery and the feasibility of
weaning from ECMO support [12].
Daily evaluation by echography provides the physician with information about
myocardial contractility, diastolic function, valvular abnormalities, and pericardial effusions; this information helps in identifying the appropriate time for a
weaning trial and may alert on problems that must be faced and solved before
the attempt of flow reduction such as tamponading effusions or valvular
abnormalities.
During a weaning trial, echocardiography can confirm hemodynamic and clinical parameters that indicate a good response; even more important, echo may be the
only monitoring method able to explain the mechanisms of weaning failure and to
evaluate cardiac response to therapeutic interventions (Table 18.1).
As shown by Combes and colleagues, some Doppler echocardiography parameters discriminate weaned and not-weaned patients better than any other parameter
[email protected]
18 Weaning from Extracorporeal Circulatory Support

Table 18.1 Echocardiographic
findings in difficult weaning
209
Right vs left ventricular failure

Systolic vs diastolic dysfunction
Regional wall motion abnormalities
Severe postischemic mitral regurgitation
Dynamic outflow tract obstruction
Pericardial tamponade
Pulmonary hypertension
Hypovolemia
tested. In their study, all weaned patients had a left ventricular ejection fraction
greater than 2530 %, an aortic velocity time interval above 12 cm, and a lateral
mitral annulus peak systolic velocity above 6 cm under minimal ECMO flow support [13].
In Table 18.2 we report recent studies describing the outcome of ECMO-assisted
refractory cardiac shock.
18.3
Technique
Criteria for weaning depend on the indication for ECMO. Obviously, the initial
cause of cardiogenic shock should have been solved, and adequate time for myocardial rest and recovery must be guaranteed.
In postcardiotomy patients, suffering from postoperative myocardial stunning,
some improvement of ventricular function should be evident within 7296 h of support. In patients with myocarditis, ventricular recovery sufficient for a weaning trial
from mechanical support may necessitate a longer period (23 weeks, depending on
the etiology).
A lot of authors agree that the weaning strategy should be individualized for each
patient and generally not attempted before 2448 h of support.
Frequent assessments of clinical status and hemodynamic parameters are mandatory while on ECMO.
When a patient has stable hemodynamics with or without inotropes or IABP for
more than 24 h without the need for relevant interventions and echocardiography
shows a sufficient ventricular recovery, a weaning attempt can be made.
Weaning is achieved by progressively reducing pump blood flow; doing so,
hemodynamic conditions change with an increase in preload and a decrease in afterload, thus resulting in rising stroke volume and cardiac output.
General criteria that must be fulfilled to start a weaning trial are the following:
Mean arterial pressure >70 mmHg
Low vasopressor requirement (inotropic score less than 10)
SpO2 >95 %
ScvO2 >70 %
Adequate natural lung oxygenating ability (chest X-ray improving after acute
pulmonary edema)
Improving 2D echo with EF >2530 %
[email protected]
40 %
(+12 pts bridged to VAD/transplant)
(only weaned pts)
69 %
Mean ECMO duration 190 127 h
63 %
60 %
(+5 pts bridged to transplant)
Chang et al. [14]
Formica et al. [15]
Rastan et al. [16]
Luo et al. [10]
62.3 %
% of successful weaning
61.7 %
Aissaoui et al. [13]
Slottosch et al. [6]
Unosawa et al. [5]
[email protected]
75.2 % in-hospital
82.4 % at 6 months
83.5 % at 1 years
86.3 % at 5 years
(20 pts bridged to VAD/
transplant)
42 % in-hospital
(5 pts bridged to transplant)
47.6 % at 30 days
61.9 % in-hospital
26 % in-hospital
70 % at 30 days (tot)
(52.1 % in ECMO weaned)
Mortality
(51.7 % in ECMO weaned)
70.2 % at 1 year
82.4 % at 5 years
Table 18.2 Recent studies describing the outcome of ECMO-assisted refractory cardiac shock
CRRT on ECMO is a predictor of mortality

No significant difference between ECMO and
ECMO + IABP
Conclusions/comments
Incomplete sternum closure predicts mortality during
ECMO; intraoperative CPB time is significantly different
among W/NW
ECMO >48 h is a predictor of mortality post weaning;
age, preop-LVEF, EuroSCORE, duration of ECMO, and
peak creatine level during ECMO are significantly
different among WS/WNS
Predictors of mortality: age, lactates at 24-h ECMO,
duration of ECMO support, GI complications, any
ECMO-related complication
Echographic predictors of successful weaning: LVEF
>2025 %, aortic VTI 10 cm, mitral annulus peak
systolic velocity TDSa 6 cm/s at minimal ECMO flow
Predictors of in-hospital mortality: MAP and SOFA
score (cutoff value 13) on the day of ECMO removal,
daily urine amount on the second day after weaning
Blood lactate levels at 48 h of ECMO support and
number of PRBCs transfused are associated with 30-day
mortality
Predictors of in-hospital mortality: age, diabetes,
preoperative chronic kidney disease, obesity, lactates,
EuroSCORE >20 %
210
A. Coppo et al.
43.7 %
61 %
35 %
(+48 pts bridged to transplant)
Zhang et al. [17]
Doll et al. [18]
Smedira et al. [4]
[email protected]
24 % at 3 days
(48 % in ECMO weaned)
76 % at 5 years (tot)
(48 pts bridged to transplant)
76 % at 30 days
82 % at 5 years
68.75 % at 30 days
75 % in-hospital
53 % at 30 days
71 % in-hospital
78 % at 3 years
66.7 % in-hospital (tot)
S vs NS have lower inotropic score, reduced blood
lactate level, shorter CPR duration, surgical
revascularization, reduced SOFA score
Preop-LVEF and lactates, CK-MB, and CK-MB/TOT
CK at 48-h ECMO are significantly different among
W/NW pts
CK-MB/TOT CK at 48-h ECMO predicts mortality on
ECMO
Higher mortality for CABG + aortic valve replacement vs
other surgery
Predictors of in-hospital survival are younger age,
absence of preoperative AMI, absence of DM, use of
IABP
Risk factors for mortality: age, thoracic aorta surgery,
reoperation, nonuse of IABP
Predictors of hospital survival: absence of pulmonary

hypertension, absence of diabetes, use of IABP
W weaned patients, NW not-weaned patients, S survivors, NS nonsurvivors, WS weaned and survived, WNS weaned but not survived (died after ECMO)
Chen et al. [8]
55 %
69.4 %
Bakhtiary et al. [3]

211
212
A. Coppo et al.
Our weaning strategy follows a step-by-step approach. Doppler echocardiography is repeated at each ECMO flow level and cardiac function is continuously monitored through a Swan-Ganz catheter.
Ventilatory support is augmented progressively during the weaning to face the
progressive rise in pulmonary blood flow.
The weaning protocol consists in the reduction of ECMO pump flow in steps of
0.5 L down to 1.0 L/min. This flow is maintained for about 4060 min after having
obtained an ACT of 180200 s. A failure in cardiac output rise, with high filling
pressures and signs of inadequate peripheral perfusion (rise in blood lactate levels,
significant reduction in SvO2), associated with echographic findings of ventricular
insufficiency leads to restore full assistance waiting for further possible recovery.
When hemodynamic parameters remain stable without the addition of inotropes
or with low level of inotropes and the patients LVEF is greater than 2025 % with
normal right ventricular TAPSE, adequate cardiac index (CI > 2.2 L/min/m2), wedge
pressure <18 mmHg, and central venous pressure <15 mmHg, ECMO removal is
considered. Heparin infusion is stopped and pump flow is raised in order to avoid clot
accumulation on the membrane. At normal coagulation and platelet count, ECMO
circuit is clamped and decannulation is performed at the bedside except for central or
surgical cannulation where the procedure is performed in the operating room.
Levosimendan is frequently administered at our institution to facilitate the weaning
process. Its peculiar characteristics make it a promising drug in this setting: it improves
both systolic and diastolic function without altering the myocardial oxygen balance
and reduces afterload and inflammatory response [1922]. Although no prospective
observations have been published to date to our knowledge on the use of levosimendan in this specific condition, our preliminary results suggest a potential beneficial role of this drug in restoring cardiac output and improving endothelial function
[23]. Levosimendan has hence become part of our weaning protocol and is nowadays
employed in most patients with an alteration in cardiac function. It is infused at an
average dose of 0.1 mcg/kg/min for 24 h, and weaning is attempted thereafter.
When intracerebral hemorrhage or brain death occurs, extracorporeal support
can be sustained if the patient is a potential organ donor to allow organ support and
is withdrawn in other cases.
18.4
After Weaning
After successful weaning from ECMO support, patients need strict monitoring of
vital parameters to ensure cardiac output adequacy in the long term.
Continuous monitoring of cardiac output, SvO2, lactate levels, acidosis, urine output, and peripheral perfusion is mandatory for the first 2448 h after decannulation.
Signs of low cardiac output syndrome must be quickly identified and prompt
therapeutic interventions carried out. Inotropic support may need to be optimized;
again, echography is fundamental in this phase.
In case of extreme hemodynamic instability, with new onset of cardiogenic
shock, reimplantation of extracorporeal support must be considered.
[email protected]
213
In the first hours after decannulation, a systemic inflammatory response may be

evident, with tachycardia, fever, and hypotension; volume expansion and vasopressors are often required.
In the subsequent days, if the course in the immediate after-weaning period was
regular, optimization of cardiologic therapy is the challenge. Periodic echographic
assessments of post-recovery LVEF, diastolic function, valvular defects, and pericardial effusions are performed.
When left ventricular function remains severely impaired after successful ECMO
weaning, cardiac resynchronization therapy (CRT) must be considered [24, 25].
In peripheral ECMO, post-decannulation complications related to the presence
of cannulas may be evident or may need appropriate investigation by Doppler
echography. We also suggest to routinely inspect cannula insertion sites to uncover
wound infectious complications.
18.5
Failure to Wean
Sometimes ECMO weaning is deemed impossible. When in the course of extracorporeal support, the impossibility to wean the patient becomes clear, prompt identification of patients amenable to long-term support or transplantation is mandatory,
in order to prevent further organ dysfunction and infections and to direct patients to
these treatments. In these patients VADs can be considered either as bridge to
transplantation or as destination therapy.
In the last years, mechanical circulatory support devices have evolved into
advanced easy-to-implant and easy-to-use devices, capable of reversing low cardiac
output syndrome in an exit strategy tailored specifically to each patient. Selection
of the appropriate device should take into account residual cardiac function and the
presence of left/right or bi-ventricular failure and underlying comorbidities. These
specific issues are addressed elsewhere in the book.
References
1. Tayara W, Starling RC, Yamani MH, Wazni O, Jubran F, Smedira N (2006) Improved survival
after acute myocardial infarction complicated by cardiogenic shock with circulatory support
and transplantation: comparing aggressive intervention with conservative treatment. J Heart
Lung Transplant 25(5):504509
2. Combes A, Leprince P, Luyt CE, Bonnet N, Trouillet JL, Lger P, Pavie A, Chastre J (2008)
Outcomes and long-term quality-of-life of patients supported by extracorporeal membrane
oxygenation for refractory cardiogenic shock. Crit Care Med 36(5):14041411
3. Bakhtiary F, Keller H, Dogan S, Dzemali O, Oezaslan F, Meininger D, Ackermann H, Zwissler
B, Kleine P, Moritz A (2008) Venoarterial extracorporeal membrane oxygenation for treatment
of cardiogenic shock: clinical experiences in 45 adult patients. J Thorac Cardiovasc Surg
135(2):382388
4. Smedira NG, Moazami N, Golding CM, McCarthy PM, Apperson-Hansen C, Blackstone EH,
Cosgrove DM 3rd (2001) Clinical experience with 202 adults receiving extracorporeal membrane
oxygenation for cardiac failure: survival at five years. J Thorac Cardiovasc Surg 122(1):92102
[email protected]
214
A. Coppo et al.
5. Unosawa S, Sezai A, Hata M, Nakata K, Yoshitake I, Wakui S, Kimura H, Takahashi K, Hata

H, Shiono M (2013) Long-term outcomes of patients undergoing extracorporeal membrane
oxygenation for refractory postcardiotomy cardiogenic shock. Surg Today 43(3):264270
6. Slottosch I, Liakopoulos O, Kuhn E, Deppe AC, Scherner M, Madershahian N, Choi YH,
Wahlers T (2013) Outcomes after peripheral extracorporeal membrane oxygenation therapy for
postcardiotomy cardiogenic shock: a single-center experience. J Surg Res 181(2):e47e55.
doi:10.1016/j.jss.2012.07.030. Epub 2012
7. Luyt CE, Landivier A, Leprince P, Bernard M, Pavie A, Chastre J, Combes A (2012) Usefulness
of cardiac biomarkers to predict cardiac recovery in patients on extracorporeal membrane oxygenation support for refractory cardiogenic shock. J Crit Care 27(5):524.e7e14
8. Chen JS, Ko WJ, Yu HY, Lai LP, Huang SC, Chi NH, Tsai CH, Wang SS, Lin FY, Chen YS
(2006) Analysis of the outcome for patients experiencing myocardial infarction and cardiopulmonary resuscitation refractory to conventional therapies necessitating extracorporeal life support rescue. Crit Care Med 34(4):950957
9. Ko WJ, Lin CY, Chen RJ, Wang SS, Lin FY, Chen YS (2002) Extracorporeal membrane oxygenation support for adult postcardiotomy cardiogenic shock. Ann Thorac Surg 73(2):
538545
10. Luo XJ, Wang W, Hu SS, Sun HS, Gao HW, Long C, Song YH, Xu JP (2009) Extracorporeal
membrane oxygenation for treatment of cardiac failure in adult patients. Interact Cardiovasc
Thorac Surg 9(2):296300
11. Haneya A, Philipp A, Diez C, Schopka S, Bein T, Zimmermann M, Lubnow M, Luchner A,
Agha A, Hilker M, Hirt S, Schmid C, Mller T (2012) A 5-year experience with cardiopulmonary resuscitation using extracorporeal life support in non-postcardiotomy patients with cardiac arrest. Resuscitation 83(11):13311337
12. Platts DG, Sedgwick JF, Burstow DJ, Mullany DV, Fraser JF (2012) The role of echocardiography in the management of patients supported by extracorporeal membrane oxygenation.
J Am Soc Echocardiogr 25(2):131141
13. Aissaoui N, Luyt CE, Leprince P, Trouillet JL, Lger P, Pavie A, Diebold B, Chastre J, Combes
A (2011) Predictors of successful extracorporeal membrane oxygenation (ECMO) weaning
after assistance for refractory cardiogenic shock. Intensive Care Med 37(11):17381745
14. Chang WW, Tsai FC, Tsai TY, Chang CH, Jenq CC, Chang MY, Tian YC, Hung CC, Fang JT,
Yang CW, Chen YC (2012) Predictors of mortality in patients successfully weaned from extracorporeal membrane oxygenation. PLoS One 7(8):e42687
15. Formica F, Avalli L, Colagrande L, Ferro O, Greco G, Maggioni E, Paolini G (2010)
Extracorporeal membrane oxygenation to support adult patients with cardiac failure: predictive factors of 30-day mortality. Interact Cardiovasc Thorac Surg 10(5):721726
16. Rastan AJ, Dege A, Mohr M, Doll N, Falk V, Walther T, Mohr FW (2010) Early and late outcomes of 517 consecutive adult patients treated with extracorporeal membrane oxygenation
for refractory postcardiotomy cardiogenic shock. J Thorac Cardiovasc Surg 139(2):302311,
311.e1
17. Zhang R, Kofidis T, Kamiya H, Shrestha M, Tessmann R, Haverich A, Klima U (2006)
Creatine kinase isoenzyme MB relative index as predictor of mortality on extracorporeal membrane oxygenation support for postcardiotomy cardiogenic shock in adult patients. Eur J
Cardiothorac Surg 30(4):617620
18. Doll N, Kiaii B, Borger M, Bucerius J, Krmer K, Schmitt DV, Walther T, Mohr FW (2004)
Five-year results of 219 consecutive patients treated with extracorporeal membrane oxygenation for refractory postoperative cardiogenic shock. Ann Thorac Surg 77(1):151157; discussion 157
19. Givertz MM, Andreou C, Conrad CH, Colucci WS (2007) Direct myocardial effects of levosimendan in humans with left ventricular dysfunction: alteration of force-frequency and
relaxation-frequency relationships. Circulation 115(10):12181224
20. Lilleberg J, Nieminen MS, Akkila J, Heikkila L, Kuitunen A, Lehtonen L, Verkkala K, Mattila
S, Salmenpera M (1998) Effects of a new calcium sensitizer, levosimendan, on haemodynamics, coronary blood flow and myocardial substrate utilization early after coronary artery bypass
grafting. Eur Heart J 19(4):660668
[email protected]
215
21. Nieminen MS, Akkila J, Hasenfuss G, Kleber FX, Lehtonen LA, Mitrovic V, Nyquist O,
Remme WJ (2000) Hemodynamic and neurohumoral effects of continuous infusion of levosimendan in patients with congestive heart failure. J Am Coll Cardiol 36(6):19031912
22. Parissis JT, Karavidas A, Bistola V, Paraskevaidis IA, Farmakis D, Korres D, Filippatos G,
Matsakas E, Kremastinos DT (2008) Effects of levosimendan on flow-mediated vasodilation
and soluble adhesion molecules in patients with advanced chronic heart failure. Atherosclerosis
197(1):278282
23. Marzorati C, Erba L, Cortinovis B, Pagan de Paganis C, Avalli L, Sangalli F (2013)
Levosimendan infusion during ECMO weaning: effect on endothelial function and haemodynamics. Appl Cardiopulm Pathophysiol 7:170171
24. Pecha S, Yildirim Y, Reichenspurner H, Deuse T (2012) Successful extracorporeal membrane
oxygenation weaning after cardiac resynchronization therapy device implantation in a patient
with end-stage heart failure. Interact Cardiovasc Thorac Surg 15(5):922923
25. Milliez P, Thomas O, Haggui A, Schurando P, Squara P, Cohen-Solal A, Mebazaa A, Leenhardt
A (2008) Cardiac resynchronisation as a rescue therapy in patients with catecholaminedependent overt heart failure: results from a short and mid-term study. Eur J Heart Fail
10(3):291297
[email protected]
Treatment Options for End-Stage

Cardiac Failure
19
Gurmeet Singh
19.1
Introduction
Preceding chapters have reviewed extracorporeal membrane oxygenation (ECMO)

support of cardiogenic shock in detail. Cardiogenic shock is associated with significant morbidity and mortality. Rapid intervention and resuscitation are essential.
Expeditious management requires suitable support technology and a clearly delineated plan of action. This section will discuss alternative mechanical strategies and
technologies that have emerged to support cardiac failure. Mechanical circulatory
support (MCS) augurs a maturing frontier with the potential to inexorably alter the
future of heart failure management.
Rapid evolution and adoption of novel devices, along with global variations in
practice and device availability, limit any review from being comprehensive and
current; instead, this chapter will focus on general principles, with selected illustrative examples.
19.2
INTERMACS Classification
The New York Heart Association functional classification insufficiently discriminates between varying degrees of advanced heart failure (HF). Accordingly, the
Interagency Registry for Mechanical Assisted Circulatory Support (INTERMACS)
has proposed distinct patient profiles to differentiate stages of advanced HF [1].
INTERMACS is sponsored by the National Heart, Lung, and Blood Institute
(NHLBI) and represents a collaborative database effort between the NHLBI, the
US Food and Drug Administration (FDA), and the Centers for Medicare &
G. Singh, BMedSc, MD, MSc, FRCSC

Critical Care Medicine and Cardiac Surgery, Mazankowski Alberta Heart Institute,
University of Alberta, Room 3A6.074, 8440-112 Street, Edmonton, AB T6G 2B7, Canada
[email protected]
217
G. Singh
218
Table 19.1 INTERMACS
patient profiles (after
Stevenson) [1]
INTERMACS profile
1
2
3
4
5
6
7
Short description
Critical cardiogenic shock
Progressive decline on inotropes
Stable, but inotrope dependent
Symptoms at rest; home on oral
therapy
Exertion intolerant
Exertion limited
Advanced NYHA class III
symptoms
Medicaid Services [2, 3]. This North American registry accounted VAD implants
in nearly 6,900 patients from 145 participating sites in the fifth INTERMACS
annual report [2].
Hospitalized patients are categorized as follows: INTERMACS 1 profile patients
(Table 19.1) are the most severely decompensated, in critical cardiogenic shock;
level 2 patients are those experiencing progressive decline on inotropes; and profile
level 3 is defined as hemodynamically stable, but inotrope dependent.
Not surprisingly, preoperative severity of cardiac decompensation correlates
with outcomes following MCS. INTERMACS profile levels 1 and 2 have the poorest survival approximately 58 % lower. Overall, actuarial survival is 80 and 70 %
at 1 and 2 years, respectively [2]. With recognition of suboptimal outcomes for
profile level 1 patients, combined with improving technology, increasingly earlier
VAD implantation is being performed. At present, INTERMACS level 1 patients
account for 16.6 % of new implants [2].
It is anticipated that as more devices receive approval and wider adoption, the
ability to track trends in MCS will provide further insights and guide future care.
Efforts are underway to consolidate data into a single registry the International
Society for Heart & Lung Transplantation (ISHLT) Mechanical Assisted Circulatory
Support (IMACS) Registry [4].
19.3
Intra-aortic Balloon Pump (IABP)

Counterpulsation Therapy
IABP therapy has been employed since Kantrowitzs initial publication in 1968 [5].
The success, combined with relative speed of insertion and simplicity of the IABP,
has led to it becoming the most widely used initial assist device.
The IABP is a helium-filled balloon affixed to the tip of a catheter, generally
inserted retrograde through the femoral artery. The tip of the balloon is optimally
positioned 12 cm distal to the origin of the left subclavian artery. The IABP operates on a volume-displacement counterpulsation principle to exert its hemodynamic
effect. Thus, the balloon inflates in diastole and deflates in systole.
Physiologically, the IABP augments coronary perfusion, reduces left ventricular
afterload, and reduces left ventricular wall tension. Diastolic inflation increases
[email protected]
19
Treatment Options for End-Stage Cardiac Failure
219
coronary perfusion pressure and coronary blood (and hence oxygen) supply.
Increased diastolic pressure may improve coronary collateral perfusion, as well as
systemic perfusion.
Systolic deflation just prior to isovolumetric contraction results in afterload
reduction, reduced LV wall tension, increased stroke volume, and cardiac output
augmentation. Management with IABP requires appropriate timing to adjust inflation and deflation, guided by the arterial pressure waveform. Modern devices are
substantially automated, simplifying monitoring and timing.
Indications for IABP include cardiogenic shock, coronary ischemia, and dysrhythmias. Complications post-myocardial infarction can also supported, such as
ventricular septal defects (VSDs), acute severe mitral regurgitation secondary to
papillary muscle rupture, and left ventricular aneurysms.
IABP use is absolutely contraindicated in aortic insufficiency and aortic dissection. Caution must be advised in patients with severe atherosclerotic disease, significant peripheral vascular disease, abdominal aortic aneurysm, and graft replacement
of the iliac or femoral arteries.
For patients in extremis, IABP may be insufficient to support severe dysrhythmias or the degree of hemodynamic embarrassment. Additionally, immobilization
of the patient, combined with peripheral arterial access, limit the effective duration
of support. Hemolysis and platelet consumption may also occur.
The hemodynamic benefits of IABP support have been reported to improve survival in cardiogenic shock following acute myocardial infarction (MI) [6]. The joint
American College of Cardiology (ACC) and American Heart Association (AHA)
guidelines have assigned IABP a class IIa recommendation as a management option
in this scenario [7]. European guidelines suggest consideration of IABP in this situation as a class IIb recommendation [8]. Previous international guidelines had supported IABP usage as a class I indication for post-MI shock, but recently IABP
efficacy has been questioned in the IABP-SHOCK II trial [9, 10]. Nevertheless,
previous widespread adoption and clinician comfort with IABP mean it currently
remains a first-line management tool for hemodynamic support.
19.4
Percutaneous Mechanical Circulatory Support
IABP limitations, combined with invasiveness of traditional surgical ventricular

assist device placement, have stimulated the development of percutaneous mechanical circulatory support (MCS). Impella (Abiomed, Danvers, Massachusetts, USA)
and TandemHeart (CardiacAssist, Inc., Pittsburgh, Pennsylvania, USA) are prototypical devices in this category.
Advantages of percutaneous MCS include deployment without the traditional
surgical approach, thus sparing operating room resources and time, relative simplicity of insertion, and ability to rapidly institute mechanical assistance, thereby expediting resuscitation.
The Impella consists of a micro-axial pump mounted on a 9 French catheter.
The device is situated in the left ventricle, having crossed the aortic valve. Blood is
[email protected]
220
G. Singh
drawn into the pump through an inlet, and then ejected beyond the aortic valve
through an outlet, into the ascending aorta. An external console controls and
monitors speed and pressure measurements, ensuring appropriate pump function.
Multiple configurations of the Impella are available. Currently, the 2.5 (2.5 lpm,
12 French pump) and CP (14 French pump) are percutaneously insertable. The 5.0
(5.0 lpm, 21 French pump) and LD (left direct 5 lpm, 21 French pump) require a
surgical approach. The 5.0 is inserted via a graft anastomosed to the femoral or axillary artery, while the LD is placed through a graft sewn on the ascending aorta and
directly inserted in LV.
The Impella 2.5 is unlikely to provide sufficient decompression and cardiac output for the severest cases of cardiogenic shock and postcardiotomy shock (PCS). The
Impella-EUROSHOCK Registry found a greater than 64 % 30-day mortality in postmyocardial infarct cardiogenic shock supported with Impella 2.5 [11]. The Impella
5.0 and LD, however, have shown some potential utility in the PCS cohort [12].
Contraindications to insertion of this device include presence of a mechanical
prosthetic aortic valve, significant aortic stenosis, aortic regurgitation, atherosclerotic aortic, and left ventricular (LV) thrombus [13]. Complications particular to this
pump to consider include arrhythmias, aortic insufficiency, LV perforation, lower
extremity ischemia, and pump migration. Hemolysis and intraventricular thrombosis have also been reported with Impella usage [1416].
The TandemHeart system consists of a paracorporeal centrifugal pump, requiring only 10 ml priming volume, a transseptal left atrial cannula, and a femoral arterial cannula. Thus, left atrial to femoral artery bypass is accomplished, entirely
percutaneously. This system is able to deliver up to 45 lpm cardiac output while
decompressing the left heart [17, 18]. Utility of this system has been reported in a
variety of clinical scenarios, achieving satisfactory hemodynamic support [19]. The
interatrial septal puncture does not require closure although it is surgically repaired
if the patient is transitioned to long-term VAD. While device insertion is performed
in the cardiac catheterization lab, the TandemHeart is removable at the bedside.
Significant aortic insufficiency is a contraindication to specifically employing the
TandemHeart. Device-specific complications include lower extremity ischemia,
cannulae migration, persistent atrial septal defect, and left atrial perforation [13].
19.5
Short-Term Ventricular Assist Devices (VADs)
Decisions regarding MCS implantation can be challenging for INTERMACS profile

1 or 2. Time limitations based on critical clinical conditions may not permit detailed
evaluation to determine long-term MCS or transplantation suitability. Short-term
VADs are an appropriate bridge to decision strategy, whereby a trial of MCS may
provide an opportunity to ascertain end-organ dysfunction reversibility, especially if
neurologic status is unclear. This is particularly applicable for postcardiotomy shock
(PCS), cardiac arrest scenarios, or cardiac catheterization lab support. Historically
dismal PCS outcomes have improved substantially with VAD support [20].
[email protected]
19
221
The ideal short-term VAD should be relatively inexpensive and capable of rapid,
easy deployment. Simplicity of management is also desirable. Percutaneous devices
fulfill these requirements. Currently, these devices have limitations with duration of
support. Patient immobility is another consideration. Most importantly, however,
the option to readily provide biventricular support is desirable. Right ventricular
percutaneous support systems are still under development and evolution.
Additionally, percutaneous VAD support systems are suboptimal choices for more
intermediate durations of support.
At the Mazankowski Alberta Heart Institute, we employ the CentriMag
(Thoratec, Pleasanton, California, USA) paracorporeal support system. This
device provides the option to provide isolated left or right ventricular assistance or
biventricular support [2124]. Takayama et al. have described a percutaneous strategy for deploying the CentriMag as an RVAD [25].
When cannulating the patient centrally, the cannulae are tunneled and exit
through the anterior abdominal wall. The actual pump rests within a bearingless
motor, connected to the drive console. We have also successfully employed a temporary in-line oxygenator when hypoxemia is not manageable with mechanical ventilation alone. As pulmonary edema resolves, and hypoxemia improves, the
oxygenator can be readily removed from the circuit.
The CentriMag system is magnetically levitated, bearingless, and capable of
generating up to 10 l/min of flow at a maximum of 5,500 rpm. Without bearings,
regions of blood stasis and friction, thermal damage, hemolysis, and thrombus formation are reduced [22].
Temporary VAD implantation is recommended (class IIa) for patients in
cardiogenic shock with end-organ compromise or unclear transplant eligibility
status, who have a reasonable expectation to improve with restoration of good
hemodynamics [26].
19.6
Long-Term VADs
For long term destination may be considered as a bridge to transplant, bridge to

candidacy, bridge to recovery, or for long-term/destination therapy. Increasingly,
patients may migrate between these broad categories, based on medical status, personal preference, and technological advances. Early referral for VAD is always
preferable. Suitable candidate selection requires recognition of patients too ill to
benefit, balanced against situations where patients are not ill enough.
Currently, implantable continuous-flow pumps have supplanted volumedisplacement (pulsatile) devices in most adult assist device programs. Continuousflow devices are quieter, have greater ease of implantation, and are of significantly
smaller size. Continuous-flow ventricular assist devices (CF-VADs) still require
anticoagulation, typically with warfarin and ASA. Patients bridged to heart transplantation with CF-VADs have similar posttransplant survival at 1 and 3 years (87
and 82 %, respectively) compared to non-LVAD-bridged recipients not on inotropic
support (88 and 82 %) [27].
[email protected]
222
G. Singh
With respect to long-term implantation, when compared to pulsatile devices, continuous-flow LVADs, CF-VADs, have been shown to have better outcomes with
respect to stroke and 2-year survival [28]. Besides better device durability, CF-VADs
also have 50 % fewer device-related infections [28]. CF-VADs improve both functional capacity and quality of life based on heart failure metrics [29]. Battery technology continues to improve, permitting increasing freedom for these patients. While
pulsatile devices provide greater left ventricular volume unloading, there is no difference in hemodynamic support or exercise capacity based on VAD design alone [30].
The HeartMate II (Thoratec, Pleasanton, California, USA) axial-flow rotary
blood pump is currently the most popular durable continuous-flow implantable left
ventricular assist device (LVAD) and has been effective for bridge to transplant
(BTT) and permanent or destination therapy (DT) [2, 31, 32]. The device currently
has US Food and Drug Administration (FDA) approval for bridge to transplantation, as well as destination therapy.
An inflow cannula is inserted into the left ventricular apex, with an outflow graft
anastomosed to the ascending aorta. The device is implantable and rests subdiaphragmatically either intra-abdominally or in the pre-peritoneal space of the left
upper quadrant. Blood leaves the left ventricle and enters the pump through an
inflow conduit. An electric motor drives a permanent magnet, the rotor. As the rotor
spins, blades propel blood through the outflow graft back into the ascending aorta.
The HeartMate II is an axial-flow pump; that is, blood flow enters and exits parallel
to the pump axis.
The rotor spins on bearings, capable of generating as much as 10 l per min blood
flow, functioning in parallel with the patients circulation. Clinicians set a fixed
speed for the pump, generally between 8,000 and 10,000 revolutions per minute
(RPM), and actual flow depends upon various factors, including patient afterload,
pump speed, and power provided to the motor. A system controller, worn around the
patients waist, is connected to the pump by a transcutaneous driveline and regulates
device function. Portable batteries allow patients to mobilize untethered [33].
Long-term implantation with this rotary device has been shown to have fewer
complication, improved survival, better quality of life, and improved functional
capacity compared to a pulsatile VAD [29]. The European experience with this
device has demonstrated similar excellent outcomes and durability for long-term
support [34]. Destination therapy patients have improving 1-year survival, around
74 % [32].
Our center also employs the HeartWare ventricular assist system (HVAD)
(Framingham, Massachusetts, USA). The HeartWare device currently has FDA
approval for bridge to transplant indications. This pump sits within the pericardial
space. Ease of implantation is enhanced not only by eliminating the need to dissect
below the diaphragm but also by simplicity of actual implant technique. Centrifugal
in design, the rotor (often referred to as the impeller), is suspended by magnets and
hydrodynamic thrust bearings. There are no points of mechanical contact within the
pump. The pattern of blood flow is similar to that described above for the HeartMate
II: blood enters the device through an inflow cannula integrated within the pump.
The suspended impeller drives blood forward, exiting via the outflow cannula and
[email protected]
19
223
through a graft anastomosed onto the ascending aorta. A percutaneous driveline is

tunneled from the device to a controller worn around the patients waist. Portable
batteries permit patient mobility and freedom from tethering.
Blood flow is determined by impeller speed (RPMs), current (power), and blood
viscosity. Typically, the device is set to operate between 2,400 and 3,300 RPM.
Additionally, clinicians enter the patients hematocrit, and blood viscosity is calculated from this value. As with the HeartMate II, blood flows are estimated.
Typically employed as a left-sided VAD, the HeartWare HVAD has been
implanted as an isolated RVAD [35] as well as a permanent implanted biventricular
support system (BiVAD) [36, 37]. Potential use as a BiVAD is advantageous; however, quality of life is significantly different with two devices, two controllers, and
two sets of batteries to manage, compared to a simple LVAD.
Potential disadvantages with the HeartWare HVAD device include higher recommendations for anticoagulation, along with thrombosis concerns [3840]. Lower
rates of thrombosis have been described more recently, possibly attributable to an
improved LV coring tool and a sintered inflow cannula [39, 40]. HeartWare recommends targeting a PT INR of 23 for the HVAD [41], compared to a PT INR of
1.52.5 for the HeartMate II [42]. Additionally, although there are considerable
differences between centers and local protocols, recommendations for antiplatelet
therapy are generally greater with the HVAD than the HeartMate II [43].
There are notable physiological differences between axial and centrifugal pumps [44].
While both types of rotary pumps are afterload sensitive, centrifugal pumps have greater
afterload sensitivity compared to axial-flow devices. Centrifugal pumps also have
greater flow pulsatility and higher estimated flow accuracy. During lower flow conditions, centrifugal flow devices typically demonstrate lower inlet suction.
Beyond survival, end-organ optimization and functional recovery are the goals of
MCS. Patients with established renal failure who are unlikely to recover function
despite improved cardiac output are not recommended as candidates for long-term
devices [26]. CF-VADs enhanced functional capacity and quality of life [28] compared to pulsatile devices [45]. Health-related quality of life and functional capacity
assessment, rather than survival, are increasingly pertinent. Improved understanding of these factors may be useful in determining patient-device suitability and lifestyle modifications, and perhaps refining implantation indications [46].
Neurologic events are still reported in nearly 10 % of HeartMate II recipients
[47]. Ischemic strokes occur in 8 % of recipients, while 11 % may suffer a hemorrhagic stroke [28]. Subsequently, less aggressive anticoagulation regimens have
been advocated.
Development of aortic insufficiency (AI) has been identified as an issue with
continuous-flow long-term VADs. Nearly half of patients demonstrate moderate or
greater AI by 18 months after CF-VAD implantation [48]. For this reason, if possible, maintaining aortic valve opening by permitting some left ventricular loading
may be desirable.
Use of CF-VADs has uncovered a specific hemodynamic and hematologic
constellation that can result in hemorrhage in 25 % of patients [49]. Significant
epistaxis is an etiology in one-fifth of cases [49]. Gastrointestinal bleeding is
[email protected]
224
G. Singh
common over 20 % following CF-VAD implantation [50, 51]. Angiodysplasia

and arteriovenous malformations (AVM), coupled with anticoagulation, acquired
von Willebrand factor (vWF) deficiency, fibrinolysis, and reduced platelet numbers
with impaired function, all contribute to bleeding [5255].
Two mechanisms are proposed to explain AVM. Firstly, it is postulated that
increased intraluminal pressure and vascular smooth muscle contraction results in
increased smooth muscle tone and vessel dilation, with ensuing AVM formation.
The second mechanism supposes that reduced pulse pressure results in hypoperfusion, vascular dilation, and ultimately angiodysplasia [56].
Analogous to Heydes syndrome, the shear stress produced during CF-VAD therapy
results in reduction of high molecular weight (HMW) vWF multimers [52, 57]. The
pump itself may directly contribute to HMW vWF deformation and proteolysis. vWF
binding to collagen or platelet gp1b receptor binding is thus impaired. Multiple mechanisms of acquired vWF deficiency have been elucidated from in vitro studies [58].
Driveline infections remain problematic: the Mayo Clinic reported a 12 % driveline infection rate, with prolonged duration of support increasing the risk [59].
Based on INTERMACS data, nearly one-fifth of patients experience a driveline
infection within a year of LVAD implantation. Interestingly, younger age represents
the only identifiable risk factor. Most concerning is that driveline infections may be
adversely associated with survival [60].
Patient selection is becoming increasingly refined with greater clinical experience. Various scoring systems have emerged to assist clinical decision-making.
Recently, the HeartMate II Risk Score (HMRS) has been proposed as a mortality
risk stratification tool for CF-VADs [61]. Patient age, serum albumin, serum creatinine, INR, and center volume are predictive factors. The Destination Therapy Risk
Score (DTRS) [62] was developed during the pulsatile device era and is more complex to calculate. DTRS utility in the CF-VAD era is likely limited. MELD scoring
has also been successful at predicting mortality for CF-VADs [6365].
VAD usage as a bridge to transplantation is currently a class I recommendation
for patients who have failed maximal therapy and have a high risk for mortality
prior to allograft availability [26]. Early referral for VAD implantation is currently
a class IIa recommended approach, as outcomes are superior prior to the patients
developing hypotension, hyponatremia, renal dysfunction, and the need for recurrent hospitalizations [26]. Increasingly outpatients are being evaluated for appropriateness of mechanical circulatory support (MCS) [66].
Patients deemed ineligible for cardiac transplantation due to pulmonary hypertension may benefit from hemodynamic unloading of the left ventricle. Coupled
with aggressive medical therapy, long-term VAD often successfully bridges this
patient group to transplant candidacy by reducing pulmonary artery pressures,
transpulmonary gradient, and pulmonary vascular resistance [67]. The current recommendation for bridge to candidacy with long-term VAD for pulmonary hypertension related to HF is class IIa [26].
Permanent or destination therapy (DT) with a pulsatile LVAD was first demonstrated to be a feasible alternative to medical therapy for end-stage heart failure in
the REMATCH trial [45]. Subsequently, it was advocated that LVAD therapy for
[email protected]
19
225
DT yielded better results if implantation was performed prior to development of

major complications [62]. Currently, durable LVAD placement is advised (class I)
for transplant-ineligible advanced heart failure patients with a high 1-year mortality
risk, without irreversible end-organ dysfunction [26]. Elective implantation is also
advocated over urgent VAD (class IIa) [26].
Prior to acceptance for long-term VAD implantation, a multidisciplinary team
assessment (surgical, medical, nutritional, and psychosocial) is highly recommended (class I) [26]. In our experience, participation of cardiology, cardiac surgery, critical care, and a specialized VAD team has greatly improved assessment,
decision-making, communication, and management.
19.7
Right Ventricular Failure (RVF)
Medical management of RVF is beyond the scope of this chapter. However, RV

function is a critical consideration during MCS application. RVF is associated with
higher earlier morbidity and mortality [2, 68]. Unlike venoarterial ECMO, LVADs
do not directly unload the right heart. Nevertheless, following LVAD implantation,
objective improvements in RV function are detectable [69]. Conversely, CF-VADs
may exacerbate RVF by causing interventricular septal shift, RV distortion, and
worsening tricuspid regurgitation, combined with increasing RV preload. RVF that
requires temporary RVAD support may occur in up to 9 % of LVAD recipients, with
an associated significantly increased mortality [70].
Predicting RV failure, therefore, has important clinical implications. The right ventricular failure risk score relies upon four variables (vasopressor usage, creatinine,
bilirubin, and aspartate aminotransferase) as predictors of post-LVAD implantation
RV failure [71]. A higher RVFRS was also associated with greater mortality [71].
Other investigators have reported preoperative tricuspid regurgitation as predictive of
RV failure [68]. Raina and colleagues describe an echocardiographic scoring system
consisting of RV fractional area change, estimated right atrial pressure, and left atrial
volume index [72]. Central venous pressure (CVP) to pulmonary capillary wedge
pressure ratio greater than 0.63, elevated blood urea nitrogen, and preoperative
mechanical ventilation have been shown to be independent predictors of RVF following CF-VAD insertion [73]. These investigators also found elevated CVP (>15 cm H2O)
and right ventricular stroke work index <300 mmHg ml m2 to be predictive.
19.8
Total Artificial Heart (TAH)
TAH technology offers biventricular support in the form of an orthotopically placed

device, which necessitates complete biventricular and heart valve excision [74]. The
SynCardia temporary TAH (TAH-t) (SynCardia Systems, Inc., Tucson, Arizona,
USA) is a pneumatically driven, volume-displacement device, typically delivering
between 7 and 9 l/min of cardiac output [75]. It has been used as a bridge to transplant with 6187 % success rate [7477].
[email protected]
226
G. Singh
There is limited clinical experience with the AbioCor (Abiomed, Massachusetts,

USA), an entirely self-contained TAH. The internal battery is rechargeable using
transcutaneous energy transmission (TET). In the USA, the device has FDA
approval as Humanitarian Use Device. A next generation device is under
development.
TAH is an attractive option for MCS in the setting of cardiac allograft failure, as
donor graft ventricles are excised, eliminating the need for immunosuppression, and
thus the associated risks of wound healing and sepsis [78, 79]. Other unique clinical
scenarios where TAH support is potentially advantageous include post-infarct ventricular septal defect [75, 80] and massive myocardial infarction/necrosis.
Furthermore, in patients requiring biventricular support as bridge to cardiac transplantation, TAH had a significantly lower stroke rate and a trend toward improved
survival when supported for greater than 90 days [81].
Currently, limitations of TAH technology include device size and complication
rates.
19.9
Transplantation
The gold standard treatment for end-stage heart failure remains cardiac transplantation. Graft durability, however, remains limited by accelerated cardiac allograft vasculopathy (CAV). Median recipient survival remains 10 years [27]. Additionally,
the organ supply-demand discrepancy continues to grow. Since 1998, less than
4,000 heart transplants have been reported per annum to the International Society
for Heart & Lung Transplantation (ISHLT) Registry, and these data probably represent two-thirds of heart transplants performed globally [27]. Interestingly, 36 % of
heart transplant recipients are currently bridged with MCS [27].
Besides limited donor organ availability, post-transplantation challenges
exist. Prolonged immunosuppressive therapy is required. Donor right heart failure following transplantation remains a significant concern due to its frequency
and association with poor outcomes. In the first 3 years following cardiac transplant, graft failure and infection represent the most common causes of death.
Beyond 3 years, malignancy and cardiac allograft vasculopathy contribute to
mortality [27].
Refractory, end-stage heart failure patients are recommended for cardiac transplant referral [81]. Patients in cardiogenic shock with documented inotropic dependence, peak VO2 (oxygen consumption) less than 10 ml/kg/min, severe symptomatic
ischemic heart disease not amenable to revascularization, and refractory ventricular
arrhythmias are all suitable cardiac allograft candidates [82]. Additionally, patients
on MCS with device-related complications should be advanced for transplantation.
Canadian Cardiovascular Society guidelines recommend MCS for cardiac transplant candidates who clinically deteriorate or are unlikely to survive until a suitable
donor organ becomes available [83].
Decisions to proceed with listing for transplantation may be aided by an estimation of mortality risk. Adjudging heart transplant candidacy in ambulatory patients
[email protected]
19
227
includes measuring VO2, calculating the Heart Failure Survival Score, and employing the Seattle Heart Failure Model [84].
Heart transplantation is contraindicated for patients with active malignancy or
infection, as both conditions may be exacerbated by immunosuppression. Pulmonary
hypertension heralds poor outcomes following transplantation. Pulmonary vascular
resistance (PVR) >5 Wood units or a trans pulmonary gradient (TPG) 15 mmHG
are contraindication to cardiac transplantation [85]. European guidelines suggest
that a PVR >45 Wood units and a transpulmonary gradient >15 mmHg are contraindications [86].
Recently, the Columbia group has published the CARRS prognostic scoring system to predict survival in high-risk transplant candidates. CARRS incorporates
cerebral vascular accident, serum albumin, retransplantation, renal dysfunction, and
>2 prior sternotomies as risk factors. They found that a high score was predictive of
poorer survival [87].
Currently, over one-third of patients are bridged by MCS to heart transplant
[27]. Of the current MCS options, LVAD support as a bridge to transplant provides
the best outcomes [88]. It is recognized that cardiac transplantation for
INTERMACS 1 and 2 patients is associated with poorer outcomes than bridging
with MCS. Attisani and colleagues reported 42.3 % early mortality for INTERMACS
profile 1 and 2 patients undergoing urgent cardiac transplantation versus 4.3 % for
emergent MCS insertion [89]. The Spanish National Heart Transplant Registry
database was recently examined and the investigators demonstrated that
INTERMACS profile correlated with outcomes following emergency heart transplantation: postoperative mortality was 43 % in profile 1 patients and 26.8 % in
profile two recipients [90].
Enthusiasm for a possible future with limitless donor organs was fostered by the
clinical case of baboon-to-human cardiac xenotransplantation in 1985 Baby Fae
[91]. The eagerness for cardiac xenotransplantation, however, abated following recognition of the potential for xenozoonoses, with an unknown actual transmission
risk [92]. Furthermore, immunologic barriers have not been overcome, and primary
graft dysfunction following cardiac xenograft remains a challenge [93].
Globally, insufficient access to donor organs exists to meet the demand for
heart transplantation. Additionally, in the MCS era, some patients decline the
opportunity for a heart transplant once they have become accustomed to improved
quality of life compared to their previous existence. The evolution of VAD technology in the continuous-flow era has led to the suggestion that long-term mechanical
assist device support outcomes are rapidly becoming on par with heart transplantation [94].
As patient selection and technology are refined, risks and complications will be
further reduced, and MCS may become preferred to heart transplantation in certain
scenarios. Since cardiac allografts have a finite lifespan median recipient survival
of 10 years it may be more desirable to perform VAD implantation in young
patients, with device replacement as required, reserving the limited donor cardiac
grafts until later in the course of the disease process or until needed to overcome
device-related complications.
[email protected]
G. Singh
228
INTERMACS 1-2
INTERMACS 2-7
Post-cardiotomy shock
Cardiac arrest
Cath lab
HeartMate II
HeartWare
Total artificial heart
Thoratec pVAD
Berlin heart
CentriMag
Venoarterial ECMO
Impella
TandemHeart
Bridge to Decision
Bridge to Recovery
Bridge to Transplant
Bridge to Candidacy
Long Term VAD
Bridge to Recovery
Patients commonly migrate between bridge categories based on changing clinical status
Fig. 19.1 Mechanical circulatory support (MCS) algorithm based on INTERMACS profile level
19.10 Algorithm
The key to salvageability is early, expeditious intervention. The Minnesota program
has described their bridge to decision approach for refractory cardiogenic shock
patient with multiple organ dysfunction [18]. They employ CentriMag BiVAD
support and reevaluate future decisions based on end-organ recovery, neurologic
status, and cardiac recovery.
Figure 19.1 outlines a proposed algorithm for device selection based on INTERMACS
profile level and clinical scenario. Since clinical status determines patient categorization,
migration between categories is common. Accordingly, a bridge to bridge strategy may
also be employed. For example, VA ECMO may be used as bridge to a short-term device,
which could in turn serve as a bridge to decision.
19.11 Future Considerations

Improving devices with fewer complications may make VAD therapy more attractive to medically managed advanced HF patients. Accordingly, the Medical Arm of
INTERMACS (MEDAMACS) is collecting data on medically treated patients who
have the potential to require VAD therapy [2]. The registry was launched in January
2013, with an intention to focus on INTERMACS profile levels 46.
[email protected]
19
229
HLA sensitization has become an increasingly important challenge facing VAD

recipients. Donor availability and posttransplant outcomes may both be affected.
Induction of sensitization is complex, with various mechanisms proposed [95].
Highly anticipated future developments include further improved quality of life
with greater battery life and less tethering. Transcutaneous energy transfer (TET)
would substantially improve quality of life for patients by eliminating tethering.
Pocket-sized controllers are already available. Additionally, new convenient
approaches to biventricular support are expected to emerge.
19.12 Conclusions
The breadth of MCS options provides clinicians the opportunity to tailor therapy
and management strategies to a patients clinical status and unique needs.
Understanding advantages and limitations of various devices, combined with algorithms, such as an approach based on INTERMACS status, may assist in optimizing
decision-making. Future developments in assist device technology portend exciting
frontiers for advanced HF support.
References
1. Stevenson LW, Pagani FD, Young JB, Jessup M, Miller L, Kormos RL et al (2009)
INTERMACS profiles of advanced heart failure: the current picture. J Heart Lung Transplant
28(6):535541. doi:10.1016/j.healun.2009.02.015
2. Kirklin JK, Naftel DC, Kormos RL, Stevenson LW, Pagani FD, Miller MA et al (2013) Fifth
INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory
support patients. J Heart Lung Transplant 32(2):141156. doi:10.1016/j.healun.2012.12.004
3. Kirklin JK, Naftel DC, Stevenson LW, Kormos RL, Pagani FD, Miller MA et al (2008)
INTERMACS database for durable devices for circulatory support: first annual report. J Heart
Lung Transplant 27(10):10651072. doi:10.1016/j.healun.2008.07.021
4. Holman WL (2012) Interagency Registry for Mechanically Assisted Circulatory Support
(INTERMACS): what have we learned and what will we learn? Circulation 126(11):1401
1406. doi:10.1161/CIRCULATIONAHA.112.097816
5. Kantrowitz A, Tjonneland S, Freed PS, Phillips SJ, Butner AN, Sherman JL (1968) Initial clinical experience with intraaortic balloon pumping in cardiogenic shock. JAMA 203(2):113118
6. Ramanathan K, Farkouh ME, Cosmi JE, French JK, Harkness SM, Davk V et al (2011)
Rapid complete reversal of systemic hypoperfusion after intra-aortic balloon pump counterpulsation and survival in cardiogenic shock complicating an acute myocardial infarction. Am
Heart J 162(2):268275. doi:10.1016/j.ahj.2011.04.025
7. OGara PT, Kushner FG, Ascheim DD, Casey DE, Chung MK, de Lemos JA et al (2013) 2013
ACCF/AHA guideline for the management of ST-elevation myocardial infarction: executive
summary: a report of the American College of Cardiology Foundation/American Heart
Association Task Force on Practice Guidelines. Circulation 127(4):529555. doi:10.1161/
CIR.0b013e3182742c84
8. Authors/Task Force Members, Steg PG, James SK, Atar D, Badano LP, Lundqvist CB et al
(2012) ESC guidelines for the management of acute myocardial infarction in patients presenting with ST-segment elevation: the Task Force on the management of ST-segment elevation
acute myocardial infarction of the European Society of Cardiology (ESC). Eur Heart J
33(20):25692619. doi:10.1093/eurheartj/ehs215
[email protected]
230
G. Singh
9. Thiele H, Schuler G, Neumann F-J, Hausleiter J, Olbrich H-G, Schwarz B et al (2012)

Intraaortic balloon counterpulsation in acute myocardial infarction complicated by cardiogenic shock: design and rationale of the Intraaortic Balloon Pump in Cardiogenic Shock II
(IABP-SHOCK II) trial. Am Heart J 163(6):938945. doi:10.1016/j.ahj.2012.03.012
10. Zeymer U, Hochadel M, Hauptmann K-E, Wiegand K, Schuhmacher B, Brachmann J et al
(2013) Intra-aortic balloon pump in patients with acute myocardial infarction complicated by
cardiogenic shock: results of the ALKK-PCI registry. Clin Res Cardiol 102(3):223227.
doi:10.1007/s00392-012-0523-4
11. Lauten A, Engstrm AE, Jung C, Empen K, Erne P, Cook S et al (2013) Percutaneous
left-ventricular support with the Impella-2.5-assist device in acute cardiogenic shock: results of
the Impella-EUROSHOCK-registry. Circ Heart Fail 6(1):2330. doi:10.1161/
CIRCHEARTFAILURE.112.967224
12. Griffith BP, Anderson MB, Samuels LE, Pae WE, Naka Y, Frazier OH (2013) The RECOVER
I: a multicenter prospective study of Impella 5.0/LD for postcardiotomy circulatory support.
J Thorac Cardiovasc Surg 145(2):548554. doi:10.1016/j.jtcvs.2012.01.067
13. Kapur NK, Jumean MF (2013) Defining the role for percutaneous mechanical circulatory support devices for medically refractory heart failure. Curr Heart Fail Rep. doi:10.1007/
s11897-013-0132-1
14. Kummerfeldt CE, Toma A, Badheka AO, Azzam I, Andrews D, Alfonso C, Chaparro SV (2011)
Severe hemolytic anemia and acute kidney injury after percutaneous continuous-flow ventricular assistance. Circ Heart Fail 4(6):e20e22. doi:10.1161/CIRCHEARTFAILURE.111.964023
15. Ranc S, Sibellas F, Green L (2013) Acute intraventricular thrombosis of an impella LP 5.0
device in an ST-elevated myocardial infarction complicated by cardiogenic shock. J Invasive
Cardiol 25(1):E1E3
16. Tanawuttiwat T, Chaparro SV (2013) An unexpected cause of massive hemolysis in percutaneous left ventricular assist device. Cardiovasc Revasc Med 14(1):6667. doi:10.1016/j.
carrev.2012.10.011
17. Kar B, Adkins LE, Civitello AB, Loyalka P, Palanichamy N, Gemmato CJ et al (2006) Clinical
experience with the TandemHeart percutaneous ventricular assist device. Tex Heart Inst J
33(2):111115
18. Ziemba EA, John R (2010) Mechanical circulatory support for bridge to decision: which
device and when to decide. J Card Surg 25(4):425433. doi:10.1111/j.1540-8191.2010.01038.x
19. Tempelhof MW, Klein L, Cotts WG, Benzuly KH, Davidson CJ, Meyers SN, McCarthy PM,
Malaisrie CS, McGee EC, Beohar N (2011) Clinical experience and patient outcomes associated with the tanden heart percutaneous transsepital assist device among a heterogeneous
patient population. A Saio Journal 57:254261
20. Hernandez AF, Grab JD, Gammie JS, OBrien SM, Hammill BG, Rogers JG et al (2007) A
decade of short-term outcomes in post cardiac surgery ventricular assist device implantation:
data from the Society of Thoracic Surgeons National Cardiac Database. Circulation
116(6):606612. doi:10.1161/CIRCULATIONAHA.106.666289
21. Akay MH, Gregoric ID, Radovancevic R, Cohn WE, Frazier OH (2011) Timely use of a
CentriMag heart assist device improves survival in postcardiotomy cardiogenic shock. J Card
Surg 26(5):548552. doi:10.1111/j.1540-8191.2011.01305.x
22. John R, Long JW, Massey HT, Griffith BP, Sun BC, Tector AJ et al (2011) Outcomes of a
multicenter trial of the Levitronix CentriMag ventricular assist system for short-term circulatory support. J Thorac Cardiovasc Surg 141(4):932939. doi:10.1016/j.jtcvs.2010.03.046
23. Shuhaiber JH, Jenkins D, Berman M, Parameshwar J, Dhital K, Tsui S, Large SR (2008) The
Papworth experience with the Levitronix CentriMag ventricular assist device. J Heart Lung
Transplant 27(2):158164. doi:10.1016/j.healun.2007.10.015
24. Worku B, Pak S-W, van Patten D, Housman B, Uriel N, Colombo P et al (2012) The CentriMag
ventricular assist device in acute heart failure refractory to medical management. J Heart Lung
Transplant 31(6):611617. doi:10.1016/j.healun.2011.12.016
25. Takayama H, Naka Y, Kodali SK, Vincent JA, Addonizio LJ, Jorde UP, Williams MR (2012)
A novel approach to percutaneous right-ventricular mechanical support. Eur J Cardiothorac
Surg 41(2):423426. doi:10.1016/j.ejcts.2011.05.041
[email protected]
19
231
26. Peura JL, Colvin-Adams M, Francis GS, Grady KL, Hoffman TM, Jessup M et al (2012)
Recommendations for the use of mechanical circulatory support: device strategies and patient
selection: a scientific statement from the American Heart Association. Circulation. doi:10.1161/
CIR.0b013e3182769a54
27. Stehlik J, Edwards LB, Kucheryavaya AY, Benden C, Christie JD, Dipchand AI et al (2012)
The registry of the International Society for Heart and Lung Transplantation: 29th official
adult heart transplant report2012. J Heart Lung Transplant 31(10):10521064. doi:10.1016/j.
healun.2012.08.002
28. Slaughter MS, Rogers JG, Milano CA, Russell SD, Conte JV, Feldman D et al (2009) Advanced
heart failure treated with continuous-flow left ventricular assist device. N Engl J Med
361(23):22412251. doi:10.1056/NEJMoa0909938
29. Rogers JG, Aaronson KD, Boyle AJ, Russell SD, Milano CA, Pagani FD et al (2010)
Continuous flow left ventricular assist device improves functional capacity and quality of life
of advanced heart failure patients. J Am Coll Cardiol 55(17):18261834. doi:10.1016/j.
jacc.2009.12.052
30. Haft J, Armstrong W, Dyke DB, Aaronson KD, Koelling TM, Farrar DJ, Pagani FD (2007)
Hemodynamic and exercise performance with pulsatile and continuous-flow left ventricular
assist devices. Circulation 116(11 Suppl):I8I15. doi:10.1161/CIRCULATIONAHA.106.
677898
31. Pagani FD, Miller LW, Russell SD, Aaronson KD, John R, Boyle AJ et al (2009) Extended
mechanical circulatory support with a continuous-flow rotary left ventricular assist device.
J Am Coll Cardiol 54(4):312321. doi:10.1016/j.jacc.2009.03.055
32. Slaughter MS, Meyer AL, Birks EJ (2011) Destination therapy with left ventricular assist
devices: patient selection and outcomes. Curr Opin Cardiol 26(3):232236. doi:10.1097/
HCO.0b013e328345aff4
33. HeartMate II LVAS, Operating Manual (2007)
34. Lahpor J, Khaghani A, Hetzer R, Pavie A, Friedrich I, Sander K, Strber M (2010) European
results with a continuous-flow ventricular assist device for advanced heart-failure patients. Eur
J Cardiothorac Surg 37(2):357361. doi:10.1016/j.ejcts.2009.05.043
35. Deuse T, Schirmer J, Kubik M, Reichenspurner H (2013) Isolated permanent right ventricular
assistance using the HVAD continuous-flow pump. Ann Thorac Surg 95(4):14341436.
doi:10.1016/j.athoracsur.2012.08.090
36. Krabatsch T, Potapov E, Stepanenko A, Schweiger M, Kukucka M, Huebler M et al (2011)
Biventricular circulatory support with two miniaturized implantable assist devices. Circulation
124(11 Suppl):S179S186. doi:10.1161/CIRCULATIONAHA.110.011502
37. Wu L, Weng Y-G, Dong N-G, Krabatsch T, Stepanenko A, Hennig E, Hetzer R (2013)
Outcomes of HeartWare Ventricular Assist System support in 141 patients: a single-centre
experience. Eur J Cardiothorac Surg 44(1):139145. doi:10.1093/ejcts/ezs263
38. Aissaoui N, Brgermann J, Gummert J, Morshuis M (2012) HeartWare continuous-flow ventricular assist device thrombosis: the Bad Oeynhausen experience. J Thorac Cardiovasc Surg
143(4):e37e39. doi:10.1016/j.jtcvs.2011.12.035
39. Siddique A, Wrightson N, Macgowan GA, Schueler S (2013) Device thrombosis in the HeartWare
left ventricular assist device. Ann Thorac Surg 95(4):1508. doi:10.1016/j.athoracsur.2012.10.011
40. Aaronson KD, Slaughter MS, Miller LW, McGee EC, Cotts WG, Acker MA et al (2012) Use of
an intrapericardial, continuous-flow, centrifugal pump in patients awaiting heart transplantation.
Circulation 125(25):31913200
41. Swartz M (2012) HeartWare IFU 1120 pdf, pp 1104
42. Boyle AJ, Russell SD, Teuteberg JJ, Slaughter MS, Moazami N, Pagani FD et al (2009) Low
thromboembolism and pump thrombosis with the HeartMate II left ventricular assist device:
analysis of outpatient anti-coagulation. J Heart Lung Transplant 28(9):881887. doi:10.1016/j.
healun.2009.05.018
43. Menon AK, Gtzenich A, Sassmannshausen H, Haushofer M, Autschbach R, Spillner JW
(2012) Low stroke rate and few thrombo-embolic events after HeartMate II implantation under
mild anticoagulation. Eur J Cardiothorac Surg 42(2):319323. doi:10.1093/ejcts/ezr312; discussion 323
[email protected]
232
G. Singh
44. Moazami N, Fukamachi K, Kobayashi M, Smedira NG, Hoercher KJ, Massiello A et al

(2013) Axial and centrifugal continuous-flow rotary pumps: a translation from pump
mechanics to clinical practice. J Heart Lung Transplant 32(1):111. doi:10.1016/j.
healun.2012.10.001
45. Rose EA, Gelijns AC, Moskowitz AJ, Heitjan DF, Stevenson LW, Dembitsky W et al (2001)
Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med
345(20):14351443. doi:10.1056/NEJMoa012175
46. Grady KL, Warner Stevenson L, Pagani FD, Teuteberg J, Pamboukian SV, Birks E et al (2012)
Beyond survival: recommendations from INTERMACS for assessing function and quality of
life with mechanical circulatory support. J Heart Lung Transplant 31(11):11581164.
doi:10.1016/j.healun.2012.08.020
47. John R, Kamdar F, Eckman P, Colvin-Adams M, Boyle A, Shumway S et al (2011) Lessons
learned from experience with over 100 consecutive HeartMate II left ventricular assist devices.
Ann Thorac Surg 92(5):15931599. doi:10.1016/j.athoracsur.2011.06.081; discussion
1599600
48. Cowger J, Pagani FD, Haft JW, Romano MA, Aaronson KD, Kolias TJ (2010) The development of aortic insufficiency in left ventricular assist device-supported patients. Circ Heart Fail
3(6):668674. doi:10.1161/CIRCHEARTFAILURE.109.917765
49. Wever-Pinzon O, Selzman CH, Drakos SG, Saidi A, Stoddard GJ, Gilbert EM et al (2013)
Pulsatility and the risk of non-surgical bleeding in patients supported with the continuous-flow
left ventricular assist device HeartMate II. Circ Heart Fail. doi:10.1161/
50. Islam S, Cevik C, Madonna R, Frandah W, Islam E, Islam S, Nugent K (2013) Left ventricular
assist devices and gastrointestinal bleeding: a narrative review of case reports and case series.
Clin Cardiol 36(4):190200. doi:10.1002/clc.22096
51. Morgan JA, Paone G, Nemeh HW, Henry SE, Patel R, Vavra J et al (2012) Gastrointestinal
bleeding with the HeartMate II left ventricular assist device. J Heart Lung Transplant
31(7):715718. doi:10.1016/j.healun.2012.02.015
52. Slaughter MS (2010) Hematologic effects of continuous flow left ventricular assist devices.
J Cardiovasc Transl Res 3(6):618624. doi:10.1007/s12265-010-9222-6
53. Crow S, Chen D, Milano C, Thomas W, Joyce L, Piacentino V et al (2010) Acquired von
Willebrand syndrome in continuous-flow ventricular assist device recipients. Ann Thorac Surg
90(4):12631269. doi:10.1016/j.athoracsur.2010.04.099; discussion 1269
54. Demirozu ZT, Radovancevic R, Hochman LF, Gregoric ID, Letsou GV, Kar B et al (2011)
Arteriovenous malformation and gastrointestinal bleeding in patients with the HeartMate II
left ventricular assist device. J Heart Lung Transplant 30(8):849853. doi:10.1016/j.
healun.2011.03.008
55. Meyer AL, Malehsa D, Bara C, Budde U, Slaughter MS, Haverich A, Strueber M (2010)
Acquired von Willebrand syndrome in patients with an axial flow left ventricular assist device.
Circ Heart Fail 3(6):675681. doi:10.1161/CIRCHEARTFAILURE.109.877597
56. Suarez J, Patel CB, Felker GM, Becker R, Hernandez AF, Rogers JG (2011) Mechanisms of
bleeding and approach to patients with axial-flow left ventricular assist devices. Circ Heart Fail
4(6):779784. doi:10.1161/CIRCHEARTFAILURE.111.962613
57. Loscalzo J (2012) From clinical observation to mechanismHeydes syndrome. N Engl J Med
367(20):19541956. doi:10.1056/NEJMcibr1205363
58. Dassanayaka S, Slaughter MS, Bartoli CR (2013) Mechanistic pathway(s) of acquired Von
Willebrand syndrome with a continuous-flow ventricular assist device. ASAIO J 59(2):123
129. doi:10.1097/MAT.0b013e318283815c
59. Sharma V, Deo SV, Stulak JM, Durham LA, Daly RC, Park SJ et al (2012) Driveline infections
in left ventricular assist devices: implications for destination therapy. Ann Thorac Surg
94(5):13811386. doi:10.1016/j.athoracsur.2012.05.074
60. Goldstein DJ, Naftel D, Holman W, Bellumkonda L, Pamboukian SV, Pagani FD, Kirklin J
(2012) Continuous-flow devices and percutaneous site infections: clinical outcomes. J Heart
[email protected]
19
233
61. Cowger J, Sundareswaran K, Rogers JG, Park SJ, Pagani FD, Bhat G et al (2013) Predicting
survival in patients receiving continuous flow left ventricular assist devices: the HeartMate II
risk score. J Am Coll Cardiol 61(3):313321. doi:10.1016/j.jacc.2012.09.055
62. Lietz K, Long JW, Kfoury AG, Slaughter MS, Silver MA, Milano CA et al (2007)
Outcomes of left ventricular assist device implantation as destination therapy in the postREMATCH era: implications for patient selection. Circulation 116(5):497505. doi:10.1161/
CIRCULATIONAHA.107.691972
63. Bonde P, Ku NC, Genovese EA, Bermudez CA, Bhama JK, Ciarleglio MM et al (2012) Model
for end-stage liver disease score predicts adverse events related to ventricular assist device
therapy. Ann Thorac Surg 93(5):15411547. doi:10.1016/j.athoracsur.2012.02.008; discussion
15471548
64. Matthews JC, Pagani FD, Haft JW, Koelling TM, Naftel DC, Aaronson KD (2010) Model for
end-stage liver disease score predicts left ventricular assist device operative transfusion
requirements, morbidity, and mortality. Circulation 121(2):214220. doi:10.1161/
CIRCULATIONAHA.108.838656
65. Yang JA, Kato TS, Shulman BP, Takayama H, Farr M, Jorde UP et al (2012) Liver dysfunction
as a predictor of outcomes in patients with advanced heart failure requiring ventricular assist
device support: use of the model of end-stage liver disease (MELD) and MELD eXcluding
INR (MELD-XI) scoring system. J Heart Lung Transplant 31(6):601610. doi:10.1016/j.
healun.2012.02.027
66. Kato TS, Stevens GR, Jiang J, Christian Schulze P, Gukasyan N, Lippel M et al (2013) Risk
stratification of ambulatory patients with advanced heart failure undergoing evaluation for heart
transplantation. J Heart Lung Transplant 32(3):333340. doi:10.1016/j.healun.2012.11.026
67. Kutty RS, Parameshwar J, Lewis C, Catarino PA, Sudarshan CD, Jenkins DP et al (2013) Use
of centrifugal left ventricular assist device as a bridge to candidacy in severe heart failure with
secondary pulmonary hypertension. Eur J Cardiothorac. doi:10.1093/ejcts/ezs678
68. Baumwol J, Macdonald PS, Keogh AM, Kotlyar E, Spratt P, Jansz P, Hayward CS (2011)
Right heart failure and failure to thrive after left ventricular assist device: clinical predictors and outcomes. J Heart Lung Transplant 30(8):888895. doi:10.1016/j.healun.2011.
03.006
69. Morgan JA, Paone G, Nemeh HW, Murthy R, Williams CT, Lanfear DE et al (2013) Impact of
continuous-flow left ventricular assist device support on right ventricular function. J Heart
70. Aissaoui N, Morshuis M, Schoenbrodt M, Hakim Meibodi K, Kizner L, Brgermann J,
Gummert J (2013) Temporary right ventricular mechanical circulatory support for the
management of right ventricular failure in critically ill patients. J Thorac Cardiovasc Surg
146(1):186191. doi:10.1016/j.jtcvs.2013.01.044
71. Matthews JC, Koelling TM, Pagani FD, Aaronson KD (2008) The right ventricular failure risk
score a pre-operative tool for assessing the risk of right ventricular failure in left ventricular
assist device candidates. J Am Coll Cardiol 51(22):21632172. doi:10.1016/j.jacc.2008.03.009
72. Raina A, Seetha Rammohan HR, Gertz ZM, Rame JE, Woo YJ, Kirkpatrick JN (2013)
Postoperative right ventricular failure after left ventricular assist device placement is predicted
by preoperative echocardiographic structural, hemodynamic, and functional parameters.
J Card Fail 19(1):1624. doi:10.1016/j.cardfail.2012.11.001
73. Kormos RL, Teuteberg JJ, Pagani FD, Russell SD, John R, Miller LW, Massey T, Milano CA,
Moazami N, Sundareswaran KS, Farrar DJ, HeartMate II, Investigators C (2010) Right ventricular failure in patients with the HeartMate II continuous-flow left ventricular assist device:
incidence, risk factors, and effect on outcomes. J Thorac Cardiovasc Surg 139(5):13161324.
doi:10.1016/j.jtcvs.2009.11.020
74. Kasirajan V, Tang DG, Katlaps GJ, Shah KB (2012) The total artificial heart for biventricular
heart failure and beyond. Curr Opin Cardiol 27(3):301307. doi:10.1097/HCO.0b013e32835220c9
75. Copeland JG, Copeland H, Gustafson M, Mineburg N, Covington D, Smith RG, Friedman M
(2012) Experience with more than 100 total artificial heart implants. J Thorac Cardiovasc Surg
143(3):727734. doi:10.1016/j.jtcvs.2011.12.002
[email protected]
234
G. Singh
76. Copeland JG, Smith RG, Arabia FA, Nolan PE, Sethi GK, Tsau PH et al (2004) Cardiac
replacement with a total artificial heart as a bridge to transplantation. N Engl J Med
351(9):859867
77. Kirsch MEW, Nguyen A, Mastroianni C, Pozzi M, Lger P, Nicolescu M et al (2013) SynCardia
temporary total artificial heart as bridge to transplantation: current results at la piti hospital.
Ann Thorac Surg 95(5):16401646. doi:10.1016/j.athoracsur.2013.02.036
78. Kalya A, Jaroszewski D, Pajaro O, Scott R, Gopalan R, Kasper D, Arabia F (2013) Role of
total artificial heart in the management of heart transplant rejection and retransplantation: case
report and review. Clin Transplant 27(4):E348E350. doi:10.1111/ctr.12146
79. Quader MA, Tang D, Katlaps G, Shah KB, Kasirajan V (2013) Total artificial heart for patients
with allograft failure. J Thorac Cardiovasc Surg 145(2):e21e23. doi:10.1016/j.jtcvs.2012.10.050
80. Ashfaq A, Jaroszewski DE, Pajaro OE, Arabia FA (2013) The role of the total artificial heart
in the treatment of post-myocardial infarction ventricular septal defect. J Thorac Cardiovasc
Surg 145(2):e25e26. doi:10.1016/j.jtcvs.2012.11.018
81. Kirsch M, Mazzucotelli J-P, Roussel J-C, Bouchot O, Nloga J, Leprince P et al (2012) Survival
after biventricular mechanical circulatory support: does the type of device matter? J Heart
82. Hunt SA, Abraham WT, Chin MH, Feldman AM, Francis GS, Ganiats TG et al (2009) 2009
Focused update incorporated into the ACC/AHA 2005 Guidelines for the Diagnosis and
Management of Heart Failure in Adults: a report of the American College of Cardiology
Foundation/American Heart Association Task Force on Practice Guidelines: developed in collaboration with the International Society for Heart and Lung Transplantation. Circulation.
doi:10.1161/CIRCULATIONAHA.109.192065
83. McKelvie RS, Moe GW, Cheung A, Costigan J, Ducharme A, Estrella-Holder E et al (2011)
The 2011 Canadian Cardiovascular Society heart failure management guidelines update: focus
on sleep apnea, renal dysfunction, mechanical circulatory support, and palliative care. Can J
Cardiol 27(3):319338. doi:10.1016/j.cjca.2011.03.011
84. Mancini D, Lietz K (2010) Selection of cardiac transplantation candidates in 2010. Circulation
122(2):173183. doi:10.1161/CIRCULATIONAHA.109.858076
85. Mehra M, Kobashigawa J, Starling RC, Russell S, Uber P, Parameshwar J et al (2006) Listing
criteria for heart transplantation: International Society for Heart and Lung Transplantation
guidelines for the care of cardiac transplant candidates2006. J Heart Lung Transplant
25(9):10241042
86. McMurray JJV, Adamopoulos S, Anker SD, Auricchio A, Bhm M, Dickstein K et al (2012)
ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The
Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the
European Society of Cardiology. Developed in collaboration with the Heart Failure Association
(HFA) of the ESC. Eur Heart J 33(14):17871847. doi:10.1093/eurheartj/ehs104
87. Schulze PC, Jiang J, Yang J, Cheema FH, Schaeffle K, Kato TS et al (2013) Preoperative
assessment of high-risk candidates to predict survival after heart transplantation. Circ Heart
Fail 6(3):527534. doi:10.1161/CIRCHEARTFAILURE.112.000092
88. Karamlou T, Gelow J, Diggs BS, Tibayan FA, Mudd JM, Guyton SW et al (2013) Mechanical
circulatory support pathways that maximize post-heart transplant survival. Ann Thorac Surg
95(2):480485. doi:10.1016/j.athoracsur.2012.05.108; discussion 485
89. Attisani M, Centofanti P, La Torre M, Boffini M, Ricci D, Ribezzo M et al (2012) Advanced heart
failure in critical patients (INTERMACS 1 and 2 levels): ventricular assist devices or emergency
transplantation? Interact Cardiovasc Thorac Surg 15(4):678684. doi:10.1093/icvts/ivs256
90. Barge-Caballero E, Segovia-Cubero J, Almenar-Bonet L, Gonzalez-Vilchez F, Villa-Arranz A,
Delgado-Jimenez J et al (2013) Preoperative INTERMACS profiles determine postoperative
outcomes in critically Ill patients undergoing emergency heart transplantation: analysis of the
Spanish National Heart Transplant Registry. Circ Heart Fail 6(4):763772. doi:10.1161/
[email protected]
19
235
91. Bailey LL, Nehlsen-Cannarella SL, Concepcion W, Jolley WB (1985) Baboon-to-human cardiac xenotransplantation in a neonate. JAMA 254(23):33213329
92. Fishman JA, Scobie L, Takeuchi Y (2012) Xenotransplantation-associated infectious risk: a
WHO consultation. Xenotransplantation 19(2):7281. doi:10.1111/j.1399-3089.2012.00693.x
93. Postrach J, Bauer A, Schmoeckel M, Reichart B, Brenner P (2012) Heart xenotransplantation
in primate models. Methods Mol Biol 885:155168. doi:10.1007/978-1-61779-845-0_10
94. Kirklin JK, Naftel DC, Pagani FD, Kormos RL, Stevenson L, Miller M, Young JB (2012)
Long-term mechanical circulatory support (destination therapy): on track to compete with
heart transplantation? J Thorac Cardiovasc Surg 144(3):584603. doi:10.1016/j.jtcvs.
2012.05.044; discussion 597598
95. Itescu S, John R (2003) Interactions between the recipient immune system and the left ventricular assist device surface: immunological and clinical implications. Ann Thorac Surg
75(6):s58s65
[email protected]
Part III
ECMO for Respiratory Support
[email protected]
Ventilatory Management of ARDS

Before and During ECMO
20
Giacomo Bellani, Giacomo Grasselli, and Antonio Pesenti
20.1
What Is Acute Respiratory Distress Syndrome (ARDS)?
20.1.1 Definition
The first report of adult respiratory distress syndrome (ARDS, the term adult was
later replaced with acute acknowledging that the syndrome could also occur in children) was published by Ashbaugh and coworkers in 1967 [1]. ARDS is characterized
by an acute onset of respiratory failure with arterial hypoxemia and lung stiffening, not
arising from cardiac failure, but rather due to a massive lesional (i.e., caused by an
increased permeability of the alveolar-capillary membrane) pulmonary edema. While
the main features of the syndrome are well recognized, its formal definition remains
more elusive. In 1994 the American/European consensus conference gave the first definition of ARDS [2], which was updated almost 30 years later by the ARDS Definition
Task Force with the so-called Berlin definition [3], which is reported in Table 20.1.
20.1.2 Pathophysiology
ARDS can be seen as the result of a stereotyped response of the lungs to one or more
inflammatory stimuli, originating primarily from the lungs (e.g., pneumonia) or from
another organ (e.g., sepsis). The inflammatory response causes the recruitment of
G. Bellani, MD, PhD (*) A. Pesenti, MD

Department of Health Science, University of Milan-Bicocca,
Via Cadore 48, Monza (MB) 20900, Italy
Department of Emergency, San Gerardo Hospital, Via Pergolesi 33, Monza (MB) 20900, Italy
G. Grasselli, MD
Department of Emergency, San Gerardo Hospital, Via Pergolesi 33, Monza (MB) 20900, Italy
[email protected]
239
G. Bellani et al.
240
Table 20.1 Berlin definition of ARDS
Timing
Within 1 week of a known clinical insult or new or worsening respiratory

symptoms
Bilateral opacities not fully explained by effusions, lobar/lung collapse, or
nodules
Respiratory failure not fully explained by cardiac failure or fluid overload
Need objective assessment (e.g., echocardiography) to exclude hydrostatic
edema if no risk factor is present
Chest
imaginga
Origin of
edema
Oxygenation
Mild
Moderate
Severe
200 mmHg < PaO2/FiO2 300 mmHg with PEEP or CPAP 5 cmH2Ob
100 mmHg < PaO2/FiO2 200 mmHg with PEEP 5 cmH2Ob
PaO2/FiO2 100 mmHg with PEEP 5 cmH2O
Chest radiograph or computed tomography scan

This may be delivered noninvasively in the mild acute respiratory distress group
neutrophils and macrophages, which extravasate from the capillaries to the alveoli,
releasing inflammatory mediators and thereby further amplifying the inflammatory
reaction [4]. This causes an increased permeability of the alveolar-capillary membrane, with the formation of a protein-rich alveolar edema which dramatically
increases lung weight and stiffness [5]. The clinical consequence is the development
of severe arterial hypoxemia and of a dramatic increase in the work of breathing,
frequently requiring patients intubation and the use of mechanical ventilation.
In the mid-1980s, computed tomography studies demonstrated that the lungs of
ARDS patients were profoundly inhomogeneous, with areas of complete aeration
loss, mostly located in the dorsal regions, and other zones of reduced or normal
aeration. These observations led to the fundamental concept of baby lung [6]: the
increased stiffness of ARDS lungs is mainly due to the low pulmonary volume
available for tidal ventilation. Moreover, it was shown that the baby lung is a
functional rather than an anatomical concept, and that a variable fraction of lung
tissue can be regained to ventilation if an appropriate pressure is applied to the airways. In parallel with this findings, it was unveiled that mechanical ventilation per
se could exacerbate the lung damage by means of the so-called ventilator-induced
lung injury (VILI) [7], which is caused by two main pathogenetic mechanisms.
First, the small baby lung is exposed to the risk of an exaggerated distension from
tidal ventilation (overinflation), which is associated with excessive alveolar stretching. Second, the unstable alveoli, collapsed at end expiration, can be reopened during tidal insufflation: the cyclic alveolar opening and closing enhances the
inflammatory response and further amplifies the lung damage.
20.2
How to Manage ARDS Before ECMO?
20.2.1 Ventilatory Strategies

The growing body of evidence regarding the aforementioned mechanisms of VILI
stimulated the research of ventilatory strategies aimed at decreasing the stretch of
[email protected]
20
Ventilatory Management of ARDS Before and During ECMO
241
ventilated tissue and at promoting recruitment and ventilation of atelectatic

lung, while minimizing the amount of cyclic intra-tidal recruitment/
derecruitment.
20.2.1.1 Tidal Volume (Vt)

Tidal volume (Vt) is the amount of gas that is periodically insufflated in the lungs,
and it is thus a major determinant of the cyclic distension of aerated parenchyma
and of the possible genesis of intra-tidal recruitment/derecruitment: for this reason
the use of low Vt is a cornerstone of ARDS treatment. A large study conducted by
the National Institute of Health (USA) on 861 ARDS patients demonstrated a significant reduction of mortality when Vt was reduced from 12 to 6 ml/kg of ideal
body weight [for men, 50 kg + 2.3 kg * (0.39 * height (cm) 60); for women, 45.5 k
g + 2.3 kg * (0.39 * height (cm) 60)], and plateau airway pressure (Pplat) was kept
below 30 cmH2O [8]. Since the publication of this study, it is recommended in all
patients with ARDS to use a Vt of 6 ml/kg and to further reduce Vt or PEEP, if
necessary, to keep Pplat below 30 cmH2O. Interestingly, the beneficial effects of
using low Vt are evident even if plateau pressures are not high [9]. Although this
approach might appear simplistic (Does one size fits all?) and other investigators
suggested that other parameters such as end-expiratory lung volume and transpulmonary pressure should be taken into account, it is important to underline that the
reduction of Vt is still the (almost) only univocally accepted beneficial strategy in
ARDS.
20.2.1.2 Respiratory Rate

An inevitable consequence of the reduction of Vt is the need to increase respiratory
rate to values often considered high, way above 20 breaths per minute: for example, in the seminal ARDSnet study, the average respiratory rate was 29 7 breaths
per minute [8]. The need of elevated respiratory rates is further exacerbated by the
increased alveolar dead space characteristic of ARDS, which translates in a very
high ventilatory demand. The respiratory rate should be adjusted targeting a physiologic pH range (say between 7.3 and 7.45), rather than toward normal PaCO2 values: unless contraindicated for intracranial problems, hypercapnia should be
expected and tolerated (permissive hypercapnia [10]).
20.2.1.3 Setting of Positive End-Expiratory Pressure (PEEP)
Along with Vt, PEEP is probably the most relevant ventilatory parameter. The beneficial effect of PEEP derives mainly from the avoidance of alveolar derecruitment
at end expiration, with consequent improvement in oxygenation. At the same time,
PEEP expands also the aerated alveoli, with the risk of overdistension and VILI
[11]. The right PEEP for each patient should ideally represent a balance between
these two opposite phenomena. Recruitment and overdistension, however, cannot
be measured routinely at the bedside but only inferred from indirect measures such
as oxygenation and compliance. For these reasons how to set PEEP in ARDS
patients remains a matter of debate. A widely diffuse approach is based on the use
of tables reporting pairs of PEEP and FiO2, which are raised (or decreased) in
[email protected]
242
G. Bellani et al.
parallel to reach a given oxygenation target [12]. Other authors proposed to set
PEEP as high as possible to achieve a Pplat of 30 cmH2O with a Vt of 6 ml/kg [13].
Other more physiological approaches select PEEP levels by simultaneously taking into account the effect of PEEP on compliance and gas exchange, particularly
aiming to identify the PEEP level associated with the best compliance [14].
However, independently of the approach used, there is quite strong evidence in the
literature suggesting that higher PEEP levels are beneficial in the most severe subcategory of ARDS patients [15].
Finally, since low Vt and high respiratory rate ventilatory strategies can easily
lead to the development of intrinsic PEEP [16], this parameter should be periodically measured by end-expiratory holds.
20.2.1.4 Recruitment Maneuvers

It is well known that the inspiratory pressure required to reopen derecruited alveoli
is by far higher than that necessary to keep the same alveoli open. On this concept
relies the rationale of recruitment maneuvers, in which airway pressure is transiently increased to levels much higher than those achieved during regular tidal
ventilation, typically equal or above 40 cmH2O. The most common effect observed
is an improvement of arterial oxygenation, which usually indicates that alveolar
recruitment has occurred. If the oxygenation improvement is only transient and vanishing, this likely indicates that the set PEEP level is not adequate to avoid derecruitment of the recruited alveoli, prompting the need for a new recruitment
maneuver followed by an increase in PEEP [17]. Several types of recruitment
maneuvers have been described in the literature, including the 40-by-40 (airway
pressure held for 40 at 40 cmH2O) and periodic SIGHs [18]. Recruitment maneuvers tend to be more efficacious during the early course of the disease [19]. A metaanalysis showed that recruitment maneuvers are safe and that the most common side
effect is a transient and self-resolving hypotension [20]; at the same time, it is
unknown if a systematic use of recruitment maneuvers has any impact on patients
outcome.
20.2.1.5 High-Frequency Oscillatory Ventilation (HFOV)

HFOV is a ventilatory technique, widely applied in the neonatal population, in
which very small tidal volumes (12 ml/kg) are delivered at very high respiratory
rates (510 breaths per seconds), while mean airway pressure is kept at levels much
higher than those achieved during conventional mechanical ventilation to promote
alveolar recruitment. By these mechanisms, HFOV is known to improve oxygenation, and for this reason, it is often regarded as a rescue treatment in refractory
hypoxemic ARDS. Following some preliminary encouraging data suggesting a
potential benefit from the systematic application of HFOV in ARDS patients, two
large multicenter randomized trials [21, 22] were conducted to compare HFOV with
a conventional lung protective strategy in patients with moderate/severe ARDS:
both failed to demonstrate any survival benefit of HFOV and quite surprisingly a
mortality reduction was not observed even in the subgroup of more severely hypoxemic patients.
[email protected]
20
243
20.2.1.6 The Role of Spontaneous Breathing

Spontaneous assisted ventilatory modalities (aimed at maintaining a variable
amount of diaphragmatic activity supported by the mechanical ventilator) are
widely employed during weaning from mechanical ventilation, but the preservation
of spontaneous breathing also during the acute phase of ARDS bears some advantages, including improved ventilation-perfusion matching, alveolar recruitment,
decreased sedation needs, lower hemodynamic impact, and reduced diaphragmatic
dysfunction [23]. A large recent randomized controlled trial, however, showed a
significant benefit on mortality when patients with more severe forms of ARDS
(with a PaO2/FiO2 <150 mmHg) received neuromuscular blockade with cisatracurium during the first 48 h of ventilation, while neuromuscular paralysis was promptly
ceased after this period [24]. The implications of this work should be carefully
considered: the paper does not show that spontaneous breathing should be completely avoided during ARDS, but it underlines that neuromuscular blockade should
be thoughtfully considered in the early phases of most severe disease since the pressure generated by spontaneously breathing patients can be very high with potentially injurious consequences.
20.2.2 Non-ventilatory Strategies

20.2.2.1 Pharmacologic Strategies
In recent years several pharmacological strategies aimed at modifying the clinical
course of ARDS have been proposed and tested in the clinical field [25], including
(only to quote some) pulmonary surfactant, activated protein C, salbutamol, omega-3
fatty acids, and caloric restriction. Unfortunately none of these drugs seems to have
any beneficial effect on patient survival, and in some cases, a higher incidence of
serious adverse events has been observed. Thus, to date, no drug is currently
approved for specific ARDS treatment. Nitric oxide is a powerful pulmonary vasodilator: for this reason, when given by inhalation, it provides selective vasodilation
of ventilated lung regions, thus improving gas exchange in the majority of ARDS
patients, but again without a positive effect on survival [26]. However, given its
powerful, but transient, effect on oxygenation, inhaled nitric oxide remains a useful
rescue tool for refractory hypoxemia.
20.2.2.2 Prone Position

The use of prone position has been proposed in the 1980s as a tool to improve oxygenation: indeed when ARDS patients are turned prone, arterial oxygenation
improves in about 6070 % of the cases [27]. Different mechanisms are involved:
more homogeneous distribution of ventilation, recruitment of dorsal lung segments,
and improved ventilation-perfusion matching [28]. This may ultimately lead to an
optimal recruitment at a given level of PEEP with a reduced risk of VILI [29]. As
opposed to other rescue therapies, the effect of prone position on mortality is becoming more clear. A recent meta-analysis [30] of confirmed the beneficial effect of
prone position on oxygenation and revealed a survival benefit in the population of
[email protected]
244
G. Bellani et al.
more severely hypoxemic patients (those having a PaO2/FiO2 <100 mmHg).

Moreover, an elegant randomized controlled trial was recently published: severe
ARDS patients (PaO2/FiO2 <150 mmHg after 1224 h of MV) were randomized to
standard treatment in supine position or to undergo prone position session lasting
16 h, which were interrupted if oxygenation worsened (as compared to supine position) in two consecutive sessions. By this approach the authors were able to demonstrate a reduction of both 28- and 90-day mortality (from 33 to 16 % and from 41 to
24 %, respectively) [31], suggesting that prone position is likely to decreased mortality if applied early, in severe ARDS patients and interrupted if no response is
seen.
20.3
How to Manage ARDS During Extracorporeal

Membrane Oxygenation (ECMO)?
The classical indications for ECMO in ARDS patients are as follows: a) refractory
impairment of gas exchanges despite an optimized ventilatory strategy (i.e., use of
protective mechanical ventilation with low Vt and high PEEP and no response to
less invasive rescue therapies) and b) the need to apply unacceptably high Vt and/or
inspiratory pressures to support oxygenation. An additional indication is represented by interhospital transportation of unstable patients [27].
However, the optimal setting of the mechanical ventilator during ECMO is still a
matter of debate, and no specific studies have been conducted to address this issue.
20.3.1 Controlled Mechanical Ventilation During ECMO

With few exceptions, in the first days after ECMO connection, patients are heavily
sedated, if necessary paralyzed and kept on controlled mechanical ventilation.
The main goal of ECMO is to support gas exchanges while minimizing the risk of
VILI. ECMO is an extremely efficient tool for CO2 removal, and its efficiency increases
with increasing pCO2 values of the blood entering the artificial lung: for this reason, at
the moment of ECMO institution, great caution must be exerted to avoid excessive
shifts in blood pCO2 and pH. To do this, a very slow increase of the sweep gas flow is
coupled with a parallel reduction of the minute ventilation of the native lung.
The latter is usually obtained through a reduction of both tidal volume and respiratory rate, which in most cases translates into a reduction of more than 50 % of the
pre-ECMO minute ventilation [32]. Most centers suggest to decrease respiratory
rate to 1015 breaths/min and to increase the inspiratory time. All the experts agree
on the importance of keeping lung rest settings, by limiting plateau inspiratory
pressure to minimize the risk of barotrauma. The guidelines of the Extracorporeal
Life Support Organization (ELSO) recommend a Pplat limit of 25 cmH2O [33];
the same target was adopted in the largest randomized trial on ECMO published to
date (the CESAR trial) [34], while in other case series, the Pplat limit was fixed at
30 cmH2O.
[email protected]
20
245
A recent study on the cohort of patients treated with ECMO for H1N1-associated
ARDS in French ICUs (the national REVA registry) showed that after ECMO connection, the mean Vt was reduced from 6.7 to 3.9 ml/kg and the mean Pplat from 32
to 26 cmH2O. Interestingly, the authors of the study concluded that under ECMO
an ultraprotective ventilation strategy minimizing Pplat may be required to improve
outcome [35]. Based on these observations, some clinical trials adopting an ultraprotective Pplat target of 20 cmH2O have been designed.
An open issue is how to set PEEP once the patient is connected to ECMO. In the
CESAR trial, PEEP was abruptly reduced to 1015 cmH2O [34]; by contrast, other
experts suggest to keep PEEP unchanged or even to increase it, with the aim of
avoiding a sudden reduction of mean airway pressure that could lead to lung collapse and pulmonary flooding.
Indeed an important dilemma in these patients is as follows: Should we let the
lungs collapse or should we try to keep them open? Supporters of the open lung
approach claim that recruited (ventilated) lungs have a lower risk of superinfection, better surfactant function, and better secretion clearance [36]. If this is true, it
remains unclear how to recruit the lungs while keeping protective ventilatory settings. An option may be the periodic application of recruitment maneuvers, but no
data on their safety and efficacy in ECMO patients are available. Some referral
centers (e.g., the University of Michigan) routinely use prone position during
ECMO to improve the ventilation-perfusion matching [37], and some reports in the
literature have shown that proning patients while on ECMO is feasible and safe [38,
39]. It must be underlined, however, that prone positioning during ECMO may be
associated with potentially dramatic complications, such as compression or inadvertent removal of vascular cannulas: for this reason, it should be performed only in
centers with extensive experience in the field [40].
Finally, after ECMO institution it is recommended to reduce the ventilator FiO2
to the lowest level compatible with a target arterial pO2 of about 5560 mmHg, to
reduce oxygen toxicity and the risk of resorption atelectasis [33]. Some authors
recommend to abruptly reduce the ventilator FiO2 to very low levels: in the CESAR
trial, for example, FiO2 was set to 30 % [34]. However, it must be remembered that
if the patient is severely hypoxemic and intrapulmonary shunt is lower than 100 %
(which means that the native lung still contributes to arterial oxygenation), such a
marked decrease of the ventilator FiO2 may lead to a significant worsening of oxygenation that could be compensated only with the use of extremely high extracorporeal blood flows.
20.3.2 Assisted Spontaneous Breathing During ECMO

When the native lung function and the patients clinical conditions improve, a shift
from controlled mechanical ventilation to an assisted spontaneous breathing mode
can be considered. The potential advantages of the preservation of a spontaneous
breathing activity in ARDS patients have already been described in a previous paragraph. During ECMO, there is a complex interplay between the patients own
[email protected]
246
G. Bellani et al.
respiratory drive, the level of sedation and the ventilation of the artificial lung: by
varying the level of sedation and the sweep gas flow, we can change the relative
fraction of total CO2 production which is removed by the native and the artificial
lung. In other words, modulating the level of extracorporeal assist may facilitate the
switch to an assisted modality of ventilation by controlling the patients respiratory
drive.
The most commonly used mode of assisted ventilation is pressure-support ventilation (PSV), but recent data seem to indicate that, at least in certain subgroups of
patients, the use of neurally adjusted ventilatory assist (NAVA) may have some
additional benefits.
A detailed description of NAVA is beyond the scope of this paragraph: briefly,
during NAVA the ventilator inspiratory assist is delivered in synchrony and in proportion to the diaphragm electromyogram (EAdi), which is acquired through a specialized nasogastric tube [41].
Literature data on the use of NAVA during ECMO are limited. Karagiannidis
et al. studied the effect of different sweep gas flows on gas exchange and ventilation
in a small sample of six patients: after gas flow reduction, the patients rapidly
increased their minute ventilation to restore a physiological pH value but tended to
maintain a protective Vt. The authors concluded that the combination of NAVA
and ECMO may permit a closed-loop ventilation with automated protected ventilation [42].
More recently, Mauri et al. compared PSV and NAVA in 10 patients undergoing
ECMO for severe primary ARDS with very low respiratory system compliance:
they observed a better patient-ventilator interaction and a reduction of asynchronies
with NAVA [43], but further studies are needed to confirm these findings.
References
1. Ashbaugh DG, Bigelow DB, Petty TL et al (1967) Acute respiratory distress in adults. Lancet
2:319323
2. Bernard GR, Artigas A, Brigham KL et al (1994) The American-European Consensus
Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 149:818824
3. Ranieri VM, Rubenfeld GD, Thompson BT et al (2012) Acute respiratory distress syndrome:
the Berlin definition. JAMA 307:25262533
4. Bellani G, Messa C, Guerra L et al (2009) Lungs of patients with acute respiratory distress
syndrome show diffuse inflammation in normally aerated regions: a [18F]-fluoro-2-deoxy-Dglucose PET/CT study. Crit Care Med 37:22162222
5. Ware LB, Matthay MA (2000) The acute respiratory distress syndrome. N Engl J Med
342:13341349
7. Del Sorbo L, Goffi A, Ranieri VM (2011) Mechanical ventilation during acute lung injury:
current recommendations and new concepts. Presse Med 40:e569e583
8. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung
injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome
Network (2000) N Engl J Med 342:13011308
[email protected]
20
247
9. Hager DN, Krishnan JA, Hayden DL et al (2005) Tidal volume reduction in patients with acute
lung injury when plateau pressures are not high. Am J Respir Crit Care Med 172:12411245
10. Curley G, Hayes M, Laffey JG (2011) Can permissive hypercapnia modulate the severity of
sepsis-induced ALI/ARDS? Crit Care 15:212
11. Zanella A, Bellani G, Pesenti A (2010) Airway pressure and flow monitoring. Curr Opin Crit
Care 16:255260
12. Brower RG, Lanken PN, MacIntyre N et al (2004) Higher versus lower positive end-expiratory
pressures in patients with the acute respiratory distress syndrome. N Engl J Med 351:
327336
13. Mercat A, Richard JC, Vielle B et al (2008) Positive end-expiratory pressure setting in adults
with acute lung injury and acute respiratory distress syndrome: a randomized controlled trial.
JAMA 299:646655
14. Terragni PP, Rosboch GL, Lisi A et al (2003) How respiratory system mechanics may help in
minimising ventilator-induced lung injury in ARDS patients. Eur Respir J Suppl 42:15s21s
15. Briel M, Meade M, Mercat A et al (2010) Higher vs lower positive end-expiratory pressure in
patients with acute lung injury and acute respiratory distress syndrome: systematic review and
meta-analysis. JAMA 303:865873
16. de Durante G, del Turco M, Rustichini L et al (2002) ARDSNet lower tidal volume ventilatory
strategy may generate intrinsic positive end-expiratory pressure in patients with acute respiratory distress syndrome. Am J Respir Crit Care Med 165:12711274
17. Lapinsky SE, Aubin M, Mehta S et al (1999) Safety and efficacy of a sustained inflation for
alveolar recruitment in adults with respiratory failure. Intensive Care Med 25:12971301
18. Patroniti N, Foti G, Cortinovis B et al (2002) Sigh improves gas exchange and lung volume in
patients with acute respiratory distress syndrome undergoing pressure support ventilation.
Anesthesiology 96:788794
19. Grasso S, Mascia L, Del Turco M et al (2002) Effects of recruiting maneuvers in patients with
acute respiratory distress syndrome ventilated with protective ventilatory strategy.
Anesthesiology 96:795802
20. Fan E, Wilcox ME, Brower RG et al (2008) Recruitment maneuvers for acute lung injury: a
systematic review. Am J Respir Crit Care Med 178:11561163
21. Young D, Lamb SE, Shah S et al (2013) High-frequency oscillation for acute respiratory distress syndrome. N Engl J Med 368:806813
22. Ferguson ND, Cook DJ, Guyatt GH et al (2013) High-frequency oscillation in early acute
respiratory distress syndrome. N Engl J Med 368:795805
23. Marini JJ (2011) Spontaneously regulated vs. controlled ventilation of acute lung injury/acute
respiratory distress syndrome. Curr Opin Crit Care 17:2429
24. Papazian L, Forel JM, Gacouin A et al (2010) Neuromuscular blockers in early acute respiratory distress syndrome. N Engl J Med 363:11071116
25. Hooper M, Bernard G (2011) Pharmacogenetic treatment of acute respiratory distress syndrome. Minerva Anestesiol 77:624636
26. Taylor RW, Zimmerman JL, Dellinger RP et al (2004) Low-dose inhaled nitric oxide in patients
with acute lung injury: a randomized controlled trial. JAMA 291:16031609
27. Patroniti N, Bellani G, Pesenti A (2011) Nonconventional support of respiration. Curr Opin
Crit Care 17:527532
28. Pelosi P, Brazzi L, Gattinoni L (2002) Prone position in acute respiratory distress syndrome.
Eur Respir J 20:10171028
29. Galiatsou E, Kostanti E, Svarna E et al (2006) Prone position augments recruitment and
prevents alveolar overinflation in acute lung injury. Am J Respir Crit Care Med 174:187197
30. Sud S, Friedrich JO, Taccone P et al (2010) Prone ventilation reduces mortality in patients with
acute respiratory failure and severe hypoxemia: systematic review and meta-analysis. Intensive
Care Med 36:585599
31. Guerin C, Reignier J, Richard JC et al (2013) Prone positioning in severe acute respiratory
distress syndrome. N Engl J Med 368:21592168
[email protected]
248
G. Bellani et al.
32. Terragni PP, Del Sorbo L, Mascia L et al (2009) Tidal volume lower than 6 ml/kg enhances
lung protection: role of extracorporeal carbon dioxide removal. Anesthesiology 111:826835
33. ELSO Guidelines for Cardiopulmonary Extracorporeal Life Support Extracorporeal Life
Support Organization, Version 1.3 November 2013. Ann Arbor, MI, USA. www.elsonet.org.
Accessed 16 May 2013
34. Peek GJ, Mugford M, Tiruvoipati R et al (2009) Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet 374:13511363
35. Pham T, Combes A, Roze H et al (2013) Extracorporeal membrane oxygenation for pandemic
influenza A(H1N1)-induced acute respiratory distress syndrome: a cohort study and
propensity-matched analysis. Am J Respir Crit Care Med 187:276285
36. Haitsma JJ, Lachmann B (2006) Lung protective ventilation in ARDS: the open lung maneuver. Minerva Anestesiol 72:117132
37. Hemmila MR, Rowe SA, Boules TN et al (2004) Extracorporeal life support for severe acute
respiratory distress syndrome in adults. Ann Surg 240:595605, discussion 605597
38. Goettler CE, Pryor JP, Hoey BA et al (2002) Prone positioning does not affect cannula function
during extracorporeal membrane oxygenation or continuous renal replacement therapy. Crit
Care 6:452455
39. Haefner SM, Bratton SL, Annich GM et al (2003) Complications of intermittent prone positioning in pediatric patients receiving extracorporeal membrane oxygenation for respiratory
failure. Chest 123:15891594
40. Litmathe J, Sucker C, Easo J et al (2012) Prone and ECMO a contradiction per se? Perfusion
27:7882
41. Sinderby C, Navalesi P, Beck J et al (1999) Neural control of mechanical ventilation in respiratory failure. Nat Med 5:14331436
42. Karagiannidis C, Lubnow M, Philipp A et al (2010) Autoregulation of ventilation with neurally adjusted ventilatory assist on extracorporeal lung support. Intensive Care Med
36:20382044
43. Mauri T, Bellani G, Grasselli G et al (2013) Patient-ventilator interaction in ARDS patients
with extremely low compliance undergoing ECMO: a novel approach based on diaphragm
electrical activity. Intensive Care Med 39:282291
[email protected]
Respiratory Monitoring
of the ECMO Patient
21
Alberto Zanella, Francesco Mojoli, Luigi Castagna,

and Nicol Patroniti
21.1 Introduction
Extracorporeal membrane oxygenation is a highly effective technique for cardiopulmonary support, but it is not devoid of complications. Hence, proper monitoring is essential
before, during, and after ECMO application. In this chapter, we will discuss respiratory
monitoring during venovenous ECMO (VV ECMO), while the indications for ECMO
institution and weaning from ECMO will be reviewed in specific chapters 16, 20, 26, 27, 40.
Respiratory monitoring during venoarterial ECMO is elucidated in Chap. 33.
ECMO can totally or partially substitute the gas exchange functions of patient
lungs: therefore, during VV ECMO, the interpretation of parameters usually
employed to monitor respiratory function, such as arterial partial pressure of oxygen
(PaO2) and carbon dioxide (PaCO2), must take into account the contribution of
extracorporeal gas exchange. Also other parameters, such as intrapulmonary shunt
fraction (natural lung shunt, Qs/Qt), may be significantly affected by extracorporeal
gas exchange. Moreover, VV ECMO substantially affects mixed venous blood gas
composition, increasing oxygen and reducing carbon dioxide content, and may
A. Zanella (*) L. Castagna

Dipartimento di Scienze della Salute, Universit di Milano-Bicocca, Ospedale San Gerardo
Nuovo dei Tintori, Via Donizetti 106, Monza, Milan 20900, Italy
F. Mojoli
SC Anestesia e Rianimazione 1, Fondazione IRCCS Policlinico S. Matteo,
Dipartimento di Scienze Clinico-chirurgiche, Diagnostiche e Pediatriche,
Sezione di Anestesia Rianimazione e Terapia Antalgica, Universit degli Studi di Pavia,
V.le Golgi 19, Pavia 27100, Italy
N. Patroniti, MD
Department of Health Sciences, Department of Urgency and Emergency, Milano-Bicocca
University, San Gerardo Hospital, Via Pergolesi 33, Monza (MB) 20900, Italy
[email protected]
249
250
A. Zanella et al.
Table 21.1 Facsimile of the electronic spreadsheet used at the San Gerardo Hospital, Monza,
Italy, to monitor patients on VV ECMO; the blue boxes show parameters and measures performed
with the clinical FiO2, while orange boxes require FiO2 100% in order to calculate the true
shunt. We report all the formulas used to compute each parameter
itself affect lung function, for example, altering the hypoxic vasoconstriction. Also
respiratory mechanics (respiratory system compliance, airway resistance, lung volume) and ventilatory parameters (mean airways pressure, plateau pressure, transpulmonary pressure) need to be closely monitored, to assess the severity of lung
disease and prevent ventilator-induced lung injury (VILI).
To date very little data exist on respiratory monitoring during ECMO support,
therefore we will start from available data in patients with ARDS, and, subsequently,
we will discuss the topic during VV ECMO with a physiological approach supported
by data from literature and the experience of our institution. Table21.1 shows a facsimile of the electronic spreadsheet used in our institution to monitor patients on
VVECMO. Our daily evaluation of patients on VV ECMO is performed with four
blood gas analyses (arterial, mixed venous, ECMO inlet, and ECMO outlet) and
[email protected]
21 Respiratory Monitoring of the ECMO Patient
251
assessment of hemodynamic, ventilatory, and ECMO parameters. To understand the

respiratory function of the natural lung (NL), we need to evaluate the role of
VVECMO and the hemodynamic status. If no major contraindications are present,
all patients treated with VV ECMO are monitored with a SwanGanz catheter with
continuous measurement of mixed venous oxygen saturation (SvO2). The pulmonary
artery catheter not only allows the measurement of pulmonary arterial pressure and
cardiac output but also allows a precise assessment of the mixed venous oxygen content; moreover, it provides continuous monitoring of core temperature which is
essential during ECMO institution or ECMO circuit replacement. From the recorded
data, we can calculate the oxygen added to the blood by the NL (VO2 NL) and by the
membrane lung (ML) (VO2 ML), the carbon dioxide removed by the NL (VCO2
NL) and by the ML (VCO2 ML), and the intrapulmonary shunt fraction (shunt NL);
it is also possible to estimate the extracorporeal blood flow recirculation fraction (R/
BF) and other useful parameters to monitor the ML function as reported in Chap. 35.
Since the measurement of intrapulmonary shunt requires ventilation with pure
oxygen, we usually perform this global daily assessment, setting the ventilator and
sweep gas FiO2 to 100%. On the contrary, during the day, we prefer to assess arterial gases at the clinical FiO2 and to continuously monitor the SvO2.
All these parameters are critical to understand and monitor the contribution of
patient lungs to oxygenation and CO2 removal: we will discuss these two functions
separately, since they have different underlying physiological mechanisms, as suggested by Kolobow and Gattinoni in the late 1970s [1, 2].
21.2 Oxygenation
Hypoxemia, deficiency of oxygen in arterial blood, is a hallmark of ARDS; quantification of the degree of hypoxemia is critical to assess the severity of the disease and prevent tissue hypoxia. It is difficult to define hypoxemia based on a single PaO2 threshold
value. However, when PaO2 is above 60mmHg, the hemoglobin dissociation curve is
almost flat, arterial hemoglobin oxygen saturation (aO2Hb) is higher than 90%, and the
oxygen content of arterial blood is close to the maximum for a given hemoglobin content. Otherwise, PaO2 levels below 40mmHg, corresponding to an aO2Hb lower than
75%, invariably result in tissue hypoxia [3]. Hence, several authors [4, 5] suggest a
target PaO2 value between 50 and 60mmHg or a target aO2Hb between 85 and 95%.
Since in ARDS patients severe hypoxemia is mainly caused by intrapulmonary
shunt (see below), VV ECMO, increasing mixed venous oxygen content, may efficiently improve arterial oxygenation if the extracorporeal blood flow is adequate. If the
native lung function is completely compromised (i.e., if the intrapulmonary shunt fraction approaches 100%), an extracorporeal blood flow higher than 44.5l/min (in the
absence of extracorporeal blood flow recirculation) will be required to support oxygenation: in such scenario, the arterial oxygen content would be close to the mixed venous
one, since NL does not contribute to gas exchange, and therefore oxygenation is completely dependent on ML function (see Fig.21.1). When the native lung function
improves, the difference between arterial and mixed venous oxygen content increases,
which further stresses the importance of continuously monitoring mixed venous blood.
[email protected]
252
A. Zanella et al.
100
2.8
Arterial O2 Hb
Mixed venous O2 Hb
2.6
Blood flow
90
2.4
85
2.2
80
2.0
75
1.8
70
10
15
Days
20
25
30
Blood flow (l/min)
O2Hb (%)
95
1.6
Fig. 21.1 Arterial and mixed venous O2Hb saturation during days in a patient successfully treated
with VV ECMO. After approximately 18 days, the patient improved, and, despite a reduction in
extracorporeal blood flow, both arterial and mixed venous O2Hb saturation ameliorated
21.2.1 PaO2 and PaO2/FiO2

A variety of indices have been introduced to evaluate oxygenation: the PaO2/FiO2
ratio, the difference between alveolar and arterial PO2, the intrapulmonary shunt
fraction, and the oxygenation index. Since the definition of ARDS proposed in 1994
by the American-European Consensus Conference (AECC) [6], PaO2/FiO2 ratio is
the more commonly employed index and has been used to differentiate patients with
acute lung injury (ALI) (PaO2/FiO2 <300mmHg) from those with acute respiratory
distress syndrome (ARDS) (PaO2/FiO2 <200mmHg). The PaO2/FiO2 ratio also
appears in the new ARDS definition (proposed in 2011 as the Berlin definition),
where it is used to define three categories of respiratory insufficiency: mild
(200mmHg <PaO2/FiO2 300mmHg), moderate (100mmHg <PaO2/FiO2
200mmHg), and severe (PaO2/FiO2 100mmHg) [7]. Such degrees of severity
were associated with different mortality and duration of mechanical ventilation,
although the prognostic role of PaO2/FiO2 ratio in ARDS patients remains to be
investigated since previous publications did not show any association between
severity of hypoxemia and patient outcome [8, 9]. Such conflicting results can be
explained by the fact that the PaO2/FiO2 ratio is strongly influenced by the FiO2 at
which it is measured. In patients with moderate (<30%) intrapulmonary shunt, the
PaO2 is greatly affected by FiO2 and their relationship appears to be nonlinear: in
particular, the PaO2/FiO2 ratio is greater at both the extremities of the FiO2 range
[email protected]
253

350
300
PaO2 (mmHg)
250
200
150
100
50
0
0
10
15
20
25
30
Days on ECMO
Fig. 21.2PaO2 over time of two patients on VV ECMO for severe respiratory failure, FiO2 100%.
Open circles: the patient was successfully disconnected from the ECMO. Solid circles: the patient
did not survive, his intrapulmonary shunt was always above 95%, and the PaO2 was always lower
than 50mmHg in spite of an extracorporeal BF always higher than 3l/min
compared to intermediate FiO2 values [10, 11]. Important variations of FiO2 may
therefore significantly alter the PaO2/FiO2 ratio, leading to different classifications
of the disease [12]. On the other hand, in patients with intrapulmonary shunt higher
than 30%, who may require VV ECMO support, PaO2 is relatively independent
from FiO2 and may be a good indicator of lung function; however, also the PaO2/
FiO2 ratio appears almost constant, since the commonly employed FiO2 is generally
elevated. The average PaO2/FiO2 ratio at enrolment in the CESAR ECMO trial
was 76mmHg [13], just slightly higher than that reported by the Italian ECMO
network (ECMOnet) in 153 critically ill patients before ECMO institution
(63mmHg) [14].
When the patient has a considerable intrapulmonary shunt and remains hypoxemic even during VV ECMO, PaO2 itself is a good indicator of the native lung
condition. Figure21.2 depicts the variation of PaO2 in two patients undergoing
VV ECMO, recorded during the daily assessment with FiO2 100%. In one patient
(solid circles), PaO2 remained always below 50mmHg despite the maximization
of the extracorporeal support (blood flow ranged between 3 and 3.5l/min),
because the intrapulmonary shunt was almost 100%; the second patient (open
circles) showed a significant PaO2 improvement which allowed disconnection
from the bypass.
[email protected]
254
A. Zanella et al.
VO2 NL
Panel a
250
VO2 NL
Panel b
VO2 ML
Blood Flow
4.4
VO2 ML
Blood Flow
250
3.8
4.2
3.2
3.0
2.8
50
3.6
150
3.4
100
3.2
Blood flow (l/min)
3.4
100
VO2 (ml/min)
3.6
150
Blood flow (l/min)
3.8
VO2 (ml/min)
200
4.0
200
50
3.0
2.6
2.4
0
0
6
Days
10
12
2.8
50
0
8
10
Days
12
14
16
Fig. 21.3 Panel (a): VO2ML and VO2NL in a patient disconnected from the VV ECMO after 12
days. Panel (b): VO2ML and VO2NL in a patient who died during VV ECMO. Measuring the
relative contribution of the NL and the ML allows to monitor the evolution of the native lung disease over time. In panel (b), VO2NL eventually became negative, since NL shunt was 100% and
the lungs were extracting oxygen from the blood
21.2.2 Oxygenation Index

Although PaO2 and possibly also the PaO2/FiO2 seem good indicators of oxygenation, they do not take into account the cost in terms of positive pressures that must
be applied by the ventilator. The oxygenation index (OI) overcomes this limitation,
including the mean airway pressure (PAW) value in its calculation as shown in
Table.21.1 [15, 16]. The OI has gained widespread popularity in neonatal and pediatric ECMO patients, but it has recently been proposed also in adults: values above
30 are usually considered an indication for ECMO institution. For example, among
the 60 critically ill patients admitted to the Italian ECMOnet ICUs and treated with
ECMO support, the median OI (cmH2O/mmHg) was 36.3 and 33.9, respectively, for
patients with or without influenza A H1N1 [14].
21.2.3 Membrane Lung and Natural Lung Oxygen Supply

PaO2, PaO2/FiO2 ratio, and the OI do not take into account the oxygen supplied by the
ECMO support: indeed, during VV ECMO arterial oxygenation depends on the oxygen
added to the blood by the ML (VO2ML) and by the NL (VO2NL). At equilibrium, the
sum of VO2ML and VO2NL equals the total body oxygen consumption. Figure21.3
shows an example of trends of VO2ML and VO2NL in two patients. Panel a: an
increase in VO2NL was observed after 8 days and eventually allowed to reduce the
extracorporeal support and successfully disconnect the patient from the ECMO. In panel
b, VO2NL showed a constant decline, and eventually after 15 days of ECMO, the lungs
were actually extracting oxygen from the blood resulting in a negative VO2NL.
[email protected]

1.2
1.0
Shunt natural lung
Fig. 21.4Intrapulmonary
shunt of the same two patients
reported in Fig.21.3. Solid
circles: the NL shunt starts to
improve after 8 days of ECMO
support, and the patient was
successfully disconnected
from the ECMO when NL
shunt was lower than 30%.
Empty circles: the patient died
during ECMO; after 12 days,
the natural lung function
worsened and eventually the
shunt was 100%
255
0.8
0.6
0.4
0.2
8
Days
10
12
14
16
21.2.4 Intrapulmonary Shunt

Intrapulmonary shunt has been preferred by some authors [11, 1719] to describe
oxygenation in ARDS patients, even if its calculation requires the placement of a
pulmonary arterial catheter for mixed venous blood sampling. Intrapulmonary
shunt is described as the fraction of cardiac output perfusing nonventilated alveoli
and therefore not participating to gas exchange. In healthy subjects, only a small
fraction of the cardiac output is shunted, due to perfusion of bronchial tissue and a
small amount of coronary venous blood. An intrapulmonary shunt fraction higher
than 10% of the cardiac output results in hypoxemia, since the non-shunted blood,
due to the peculiar form of the oxyhemoglobin dissociation curve, is not able to
load the extra amount of oxygen required to fully saturate the shunted blood. The
nomenclature is not strictly defined, but usually the term true shunt is referred to
the areas with zero ventilation/perfusion ratio and is calculated as showed in
Table.21.1 while breathing pure oxygen. Venous admixture is obtained from the
same equation while breathing an oxygen concentration lower than 100% and
embeds three components: ventilation/perfusion mismatch, diffusion limitation,
and true shunt; for this reason, it is also referred as physiologic shunt. In our
experience, ECMO support is deemed necessary in patients with an intrapulmonary shunt fraction exceeding 5060%, while disconnection is attempted when
shunt becomes lower than 40% (see Fig.21.4). It is commonly believed that intrapulmonary shunt increases when FiO2 or mixed venous oxygen tension (PvO2)
increases, which is exactly what ECMO does. However, a well-designed study on
severe ARDS patients undergoing VV ECMO examined the specific role of PvO2
and FiO2 and showed that reducing PvO2 by decreasing the FiO2 of the ML caused
a dramatic fall in PaO2, but the intrapulmonary shunt fraction did not substantially
change; similarly, increasing the ventilator FiO2 from 60 to 100% did not modify
the intrapulmonary shunt [20]. These results, partially in contrast with other literature data, may indicate that in those ARDS patients, the hypoxic vasoconstriction
was blunted.
[email protected]
256
A. Zanella et al.
250
VCO2 (ml/min)
PO2, PCO2, EtCO2 (mmHg)
Native lungs
Artificial lungs
200
Total
150
100
50
0
Panel a
Gas flow (I/min)
100
90
80
70
60
50
40
30
20
10
0
Panel c
pO2
pCO2
EtCO2
Gas flow (I/min)
Minute Ventilation
35
Plateau pressure
35
30
Paw (cmH2O)
MV (Imin), Pplat (cmH2O)
40
40
25
20
15
10
25
20
15
10
GF 6
TV 190
Pplat 22
GF 0.5
TV 320
Pplat 29
GF 0
TV 425
Pplat 39
0
6
Panel b
30
Gas Flow (I/min)
Time (s)
0
Panel d
Fig. 21.5 Example of VV ECMO patients global assessment. Relationship between native and
extracorporeal VCO2 (panel a), plateau pressure, minute ventilation (panel b), arterial and exhaled
gases (panel c) and airways pressure waveforms (panel d) during a progressive reduction of sweep
gas flow of the membrane lung in a patient undergoing VV ECMO (BF 3.5 l/min) on spontaneous
ventilation (mode: pressure support, support level 12 cmH2O, PEEP 12 cmH2O, FIO2 55%). In
panel d, GF and TV refer to the sweep gas flow (l/min) of artificial ML and to the tidal volume (ml)
of NL, respectively
21.3 Carbon Dioxide

Also when analyzing arterial PCO2 and pH, we need to consider the relative role of
NL and ML. Extracorporeal CO2 removal, differently from extracorporeal oxygenation, which mainly depends on extracorporeal blood flow, is primarily related to
sweep gas flow: therefore, VO2 and VCO2 are independent and need to be measured separately. A decrease in sweep gas flow reduces the artificial lung CO2
extraction (nonlinearly; see Fig.21.5 panel a), and consequently if the ventilation of
the natural lung is not adjusted, PaCO2 will increase. In a spontaneously breathing
patient, a reduction of the sweep gas flow will be associated with an increased respiratory drive, leading to an increase in minute ventilation and VCO2NL (Fig.21.5,
panels a and b). In the patient shown in Fig.21.5, the progressive reduction of
sweep gas flow from 6 to 0.5 l/min (and of VCO2ML from 150 to 40 ml/min) was
associated with increased spontaneous patients efforts, which may enhance lung
recruitment and actually resulted in a higher PaO2 (panel c). However, when the gas
flow was further reduced, the patients respiratory efforts became excessive, leading
both to a dangerous elevation of plateau pressures (panels b and d) and to a not
endurable increase of work of breathing and total CO2 production (panel a).
Furthermore, the higher respiratory muscles O2 consumption and the concurrent
reduction of the extracorporeal support caused a critical decrease in venous oxygen
[email protected]
257
saturation and ultimately a drop in PaO2 (panel c). This example strongly supports
the need for a global assessment of the patient, with simultaneous evaluation of gas
exchanges (PaO2, PaCO2) and of the ventilatory load imposed to the native lung
(Pplat, minute ventilation). Monitoring the behavior to progressive re-loading of the
native lung is helpful in indicating whether or not, and in the case how much, the
patient is still dependent on the extracorporeal support (see Chap. 27 for further
information on weaning from VV ECMO).
21.4 Respiratory Mechanics

The elastic characteristics of the respiratory system are described by the compliance
of the respiratory system (CplRS, ml/cmH2O) which is the change in volume (V)
for any applied pressure variation (P):
CplRS =

DV ( ml )
D P ( cmH 2 O )
A normal value of CplRS in healthy adults is around 100mL/cmH2O, but during

invasive mechanical ventilation, the expected normal value is lower, around
5060mL/cmH2O [21], and it further decreases down to 3040mL/cmH2O or less
in patients with ALI/ARDS [21, 22].
Some authors prefer to calculate the system elastance, which is the reciprocal of
compliance and is therefore expressed as P/V.
The concept of baby lung, introduced by Gattinoni and Pesenti [23], states that
the lungs in ARDS patients are not stiff per se, since the intrinsic elasticity of the
aerated lung parenchyma appears nearly normal. Therefore, the reduction in CplRS,
always seen in patients with ARDS, is caused by the reduced dimension of the aerated lungs, the baby lung. Consequently, the CplRS is linearly related to the lung
volumes and reflects the degree of lung volume loss [22, 24, 25].
Although the CplRS value was not included in the final version of the Berlin definition of ARDS, since it did not improve the patient outcome prediction, the panel
of experts emphasized the importance of daily evaluation of CplRS while managing
ARDS patients [24]. The utility of CplRS in patients on ECMO is even higher
because it is not affected by the extracorporeal circulation and is an excellent indicator of the severity of the lung disease.
Generally, in adult patients, values of CplRS above 30ml/cmH2O indicate clinical conditions that can be treated with conventional therapies, while values lower than 2025ml/
cmH2O are found in extremely severe medical conditions that often require extracorporeal support. Between these two extremes, there are a multitude of clinical situations that
can be difficult to manage with conventional supports: the average CplRS value of patients
considered for ECMO treatment at the time of inclusion in the CESAR trial was 26ml/
cmH2O [13]. Almost 30 years ago, Gattinoni demonstrated in a group of 36 patients with
severe ARDS that CplRS was the most useful measurement to guide the clinical management: all the patients, but one, with CplRS below 25 needed ECMO support [24].
[email protected]
258
A. Zanella et al.
The measurement of CplRS in intubated patient during mechanical ventilation is

calculated with the following formula:
CplRS =

TV ( ml )
Pplat PEEPtot ( cmH 2 O )
where TV is the tidal volume, Pplat is the airway pressure during an end-inspiratory
pause of at least 3s, and PEEPtot=set PEEP+auto-PEEP (or intrinsic PEEP). The
PEEPtot is measured during an expiratory pause.
Until now we have considered the compliance of the respiratory system as a
single entity, but it is well known that the respiratory system is constituted by two
mechanical structures in series: the lung and the chest wall, therefore the applied
pressure (P) required to introduce an air volume (V) in the system is spent in part
to expand the lungs (transpulmonary pressure=airway pressure minus pleural pressure) and in part to expand the chest wall. The pleural pressure is clinically estimated by measuring the pressure inside the esophagus with a special nasogastric
tube equipped with a balloon. In our institution the esophageal pressure is not routinely monitored, nevertheless some authors emphasize the critical role of transpulmonary pressure, also referred as lung stress, as the primary determinant of
ventilator-induced lung injury [21]. Recently, G. Cortes and J.J. Marini suggested
that monitoring transpulmonary pressure and functional residual capacity (FRC) at
the bedside can improve the interpretation of conventional parameters of lung
mechanics based on airway pressures alone and may help to develop a ventilator
strategy tailored to the specifics of a given patient [26].
During the last influenza A (H1N1) pandemic, Grasso etal. applied these principles in a cohort of 14 patients with severe ARDS referred for extracorporeal membrane oxygenation [27]. In seven cases, the transpulmonary pressure was above
25cmH2O, and all these patients underwent ECMO. In the other seven cases, the
transpulmonary pressure was lower than 25cmH2O and PEEP was raised (from
17.91.2 to 22.31.4 cmH2O) to obtain a target transpulmonary pressure of
25cmH2O: this strategy improved oxygenation, allowing these patients to be successfully treated with conventional ventilation. In other words, the transpulmonary
pressure measurement allows an estimation of the real driving pressure across the
lung parenchyma, regardless of peculiar characteristics of the chest wall [26].
However, when the transpulmonary pressure is not available, clinicians may use
its surrogate, the plateau pressure (Pplat) obtained with an inspiratory pause. In the
year 2000, the ARDSnet seminal trial suggested not to exceed a Pplat of 30cmH2O
in order to reduce VILI [28]. Recently two different studies, one performed by
Bellani etal. with PET images [29] and the other by Terragni etal. with pulmonary
computed tomography and pulmonary cytokines [30], suggested a lower safety
limit for plateau pressure around 27cmH2O.
Also during spontaneous breathing, Pplat plays a critical role in the evaluation of
patient effort and of the safety of native lung ventilation. Panel d in Fig.21.5
shows inspiratory occlusions during pressure support ventilation at different extracorporeal supports. At high sweep gas flow, Pplat is 22cmH2O, 2cmH2O lower than
the sum of PEEP (12cmH2O) plus pressure support (12cmH2O), indicating that
[email protected]
259
patient effort is negligible. At lower gas flows, the patient effort increases as demonstrated by the pressure waveform (deep deflections preceding the ventilatory trigger and during the end-inspiratory pause), and the Pplat value much greater than the
sum of PEEP plus pressure support [31], clearly exceeding safety limits.
Another strategy frequently employed in our ICU to monitor patient effort and to
detect patientventilator asynchronies, which are common in patients with extremely
low CplRS, is through the analysis of the diaphragm electrical activity (EAdi): this is
obtained with a specific nasogastric tube and a mechanical ventilator employed to
perform neurally adjusted ventilatory assist (NAVA). During pressure support ventilation, the EAdi signal may help to set the cycling criteria so as to improve patient
ventilator interaction or, when necessary, may allow to switch to the NAVA
ventilatory mode, which supports the patient proportionally and in sync to the EAdi
signal [32]. Furthermore, since EAdi is directly related to patient effort, its continuous recording may enable a real-time estimate of patients inspiratory effort through
the assessment of a coefficient termed Pmusc/EAdi index [33], which may be of
extreme interest, especially during the weaning phase from ECMO.
The aerated lung volume in patients requiring VV ECMO can be particularly
reduced, so VILI prevention strategies should be based on pressure limits rather
than tidal volume indexed to the ideal body weight; alternatively, tidal volume may
be indexed to the functional residual capacity (FRC), which is the volume of gas
contained in the lungs at the end of a normal exhalation or, when a PEEP level is
applied, to the end-expiratory lung volume (EELV), which is the sum of FRC and
the volume of gas in the lung due to the application of PEEP [21, 34]. There are
several methods to measure, directly or indirectly, the lung volume in critically ill
patients: the most used are CT scan, helium dilution technique, and washin/washout
of a tracer gas (N2 or O2). The first two techniques can also be applied to patients on
ECMO, while the last one would not be reliable. Today the CT scan is still considered the gold standard, but the dilution technique with helium, which showed an
accuracy comparable to that of CT, has the great advantage that it can be performed
also at the bedside but requires a brief disconnection of the patient from the ventilator, which may enhance lung derecruitment. The FRC in healthy adults is around
33.5l while in ARDS patients may be lower than 700ml [22]. Lung volume reduction correlates to the severity of lung disease [21]. Measures of lung volumes may
help to evaluate the course of lung disease, to assess the efficacy of recruitment
maneuvers and of different levels of PEEP, and to tailor the ventilatory strategy to
the real size of the baby lung [26, 34].
Recently, lung ultrasound (LUS) is becoming a new tool for daily respiratory
monitoring [35]. This bedside, noninvasive and easy repeatable technique offers
accurate information that may help the clinician to deal with several different scenarios, such as the diagnosis of pneumothorax, pleural effusion, airways obstruction, parenchymal consolidation and alveolar-interstitial syndromes [35]. Moreover,
LUS effectively describes progressive lung de-aeration and re-expansion [36], thus
can easily and repeatedly check lung disease progression/resolution [37] and closely
monitor any respiratory maneuver, eventually suggested by LUS, aimed at improve
lung recruitment [35, 36, 38]. Fig.21.6 reports the use of lung ultrasound in a patient
undergoing VV ECMO in which pressure-volume curves were also obtained.
[email protected]
260
A. Zanella et al.
d 1,000
Day 1
Day 5
Day 13
900
800
Volume (ml)
700
600
500
400
300
200
100
0
10
15
20
25
30
Paw (cmH2O)
35
40
Fig. 21.6 Monitoring of VV ECMO-assisted ARDS varicella pneumonia patient by means of

lung ultrasound. (ac) Quad panels show LUS changes over time in representative ventral (upper
pictures of each quad panel) and dorsal (lower pictures of each quad panel) areas of the lung (transversal scans through the same intercostal spaces, respectively, in the most nondependent and
dependent areas, on both lungs; R right, L left). Panel (d) shows PV curves performed simultaneously to LUS images acquisition. All observations performed on day 1, day 5, and day 13 of ICU
stay (ARDS onset occurred 2 days before ICU admission). Panel (a) (day 1): ventral areas show a
typical sonographic interstitial syndrome pattern with multiple longitudinal artifacts (B lines) arising from the pleural line (horizontal hyperechoic line, arrow) and spreading to the edge of the
screen; zones of major crowding, especially on the left scan, can be observed; additional features
are represented by pleural thickenings and irregularities and scattered subpleural consolidations
(all consistent with an inflammatory origin of the edema). Dorsal areas show complete hepatization of the lung, i.e., an LUS pattern of consolidation consistent with complete loss of aeration of
the most dependent regions (dashed areas highlight the extension of these consolidations). The
corresponding PV curve in panel (d) shows a very low compliance (CplRS=20 ml/cmH2O) and a
limited recruitability, as demonstrated by the small hysteresis of the curve. Panel (b) (day 5): ventral areas now show an increase in B lines crowding, even reaching on the left scan the pattern of
white lung (complete coalescence). Dorsal areas show an increase in the size of consolidations
(note the dashed areas size) and the appearance of likely abundant pleural effusions (black,
anechoic, layer surrounding the consolidated lung). The corresponding PV curve in panel (d)
shows now an even smaller compliance (CplRS=13 ml/cmH2O) and further reduction of recruitability (very small hysteresis of the curve). Panel (c) (day 13): ventral areas finally show a relevant
reduction in the number and crowding of B lines, while dorsal ones show a dramatic reduction in
consolidated areas size, despite an increase in the pleural effusion amount (thicker anechoic layer).
The corresponding PV curve in panel (d) confirms the relevant improvement of the lungs, as
demonstrated by a markedly increased compliance (CplRS=43ml/H2O) and a much higher recruitability (wide hysteresis). Patient was successfully weaned from VV ECMO on day 15
[email protected]
261
21.5 Conclusions
Clinical management of severe ARDS patient undergoing VV ECMO for respiratory support requires a specific monitoring to discriminate the role of extracorporeal
gas exchange from the native lung function. This challenge is even more complex
since extremely scarce data are present in the literature, and therefore anecdotic data
became relevant. Experience of the clinical staff becomes invaluable to understand
and interpret the numerous information obtained. We always monitor these patients
with a SwanGanz catheter that gives us valuable information and a continuous
measurement of mixed venous blood saturation. Important technological developments are revolutionizing the way we monitor our patients, especially those undergoing ECMO. The bigger the challenge, the greater the commitment.
References
1. Kolobow T, Gattinoni L, Tomlinson TA, Pierce JE (1977) Control of breathing using an extracorporeal membrane lung. Anesthesiology 46:138141
2. Gattinoni L, Pesenti A, Kolobow T, Damia G (1983) A new look at therapy of the adult respiratory distress syndrome: motionless lungs. Int Anesthesiol Clin 21:97117
3. Gross R (2007) Arterial blood gas measurement. In: Parrillo J (ed) Critical care medicine,
Elsevier, Philadelphia, PA, USA
4. ELSO guidelines. http://www.elso.med.umich.edu/Guidelines.html
5. Bartlett RH (2012) Physiology of extracorporeal life support. In: ECMO: extracorporeal cardiopulmonary support in critical care. Extracorporeal Life Support Organization, Ann Arbor, MI, USA
6. Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L, Lamy M, Legall JR, Morris
A, Spragg R (1994) The American-European Consensus Conference on ARDS. Definitions,
mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med
149:818824
7. Ranieri VM, Rubenfeld GD, Thompson BT, Ferguson ND, Caldwell E, Fan E, Camporota L,
Slutsky AS (2012) Acute respiratory distress syndrome: the Berlin definition. JAMA
307:25262533
8. Luhr OR, Karlsson M, Thorsteinsson A, Rylander C, Frostell CG (2000) The impact of respiratory variables on mortality in non-ARDS and ARDS patients requiring mechanical ventilation. Intensive Care Med 26:508517
9. Nuckton TJ, Alonso JA, Kallet RH, Daniel BM, Pittet JF, Eisner MD, Matthay MA (2002)
Pulmonary dead-space fraction as a risk factor for death in the acute respiratory distress syndrome. N Engl J Med 346:12811286
10. Aboab J, Louis B, Jonson B, Brochard L (2006) Relation between PaO2/FIO2 ratio and FIO2:
a mathematical description. Intensive Care Med 32:14941497
11. Gowda MS, Klocke RA (1997) Variability of indices of hypoxemia in adult respiratory distress
syndrome. Crit Care Med 25:4145
12. Ferguson ND, Kacmarek RM, Chiche JD, Singh JM, Hallett DC, Mehta S, Stewart TE (2004)
Screening of ARDS patients using standardized ventilator settings: influence on enrollment in
a clinical trial. Intensive Care Med 30:11111116
13. Peek GJ, Mugford M, Tiruvoipati R, Wilson A, Allen E, Thalanany MM, Hibbert CL,
Truesdale A, Clemens F, Cooper N, Firmin RK, Elbourne D (2009) Efficacy and economic
assessment of conventional ventilatory support versus extracorporeal membrane oxygenation
for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet
374:13511363
[email protected]
262
A. Zanella et al.
14. Patroniti N, Zangrillo A, Pappalardo F, Peris A, Cianchi G, Braschi A, Iotti GA, Arcadipane
A, Panarello G, Ranieri VM, Terragni P, Antonelli M, Gattinoni L, Oleari F, Pesenti A (2011)
The Italian ECMO network experience during the 2009 influenza A(H1N1) pandemic: preparation for severe respiratory emergency outbreaks. Intensive Care Med 37:14471457
15. Bayrakci B, Josephson C, Fackler J (2007) Oxygenation index for extracorporeal membrane
oxygenation: is there predictive significance? J Artif Organs 10:69
16. Durand M, Snyder JR, Gangitano E, Wu PY (1990) Oxygenation index in patients with meconium aspiration: conventional and extracorporeal membrane oxygenation therapy. Crit Care
Med 18:373377
17. Covelli HD, Nessan VJ, Tuttle WK 3rd (1983) Oxygen derived variables in acute respiratory
failure. Crit Care Med 11:646649
18. Oliven A, Abinader E, Bursztein S (1980) Influence of varying inspired oxygen tensions on the
pulmonary venous admixture (shunt) of mechanically ventilated patients. Crit Care Med
8:99101
19. Rasanen J, Downs JB, Malec DJ, Oates K (1987) Oxygen tensions and oxyhemoglobin saturations in the assessment of pulmonary gas exchange. Crit Care Med 15:10581061
20. Rossaint R, Hahn SM, Pappert D, Falke KJ, Radermacher P (1995) Influence of mixed venous
PO2 and inspired O2 fraction on intrapulmonary shunt in patients with severe ARDS. J Appl
Physiol 78:15311536
21. Chiumello D, Carlesso E, Cadringher P, Caironi P, Valenza F, Polli F, Tallarini F, Cozzi P,
Cressoni M, Colombo A, Marini JJ, Gattinoni L (2008) Lung stress and strain during mechanical
ventilation for acute respiratory distress syndrome. Am J Respir Crit Care Med 178:346355
22. Patroniti N, Bellani G, Cortinovis B, Foti G, Maggioni E, Manfio A, Pesenti A (2010) Role of
absolute lung volume to assess alveolar recruitment in acute respiratory distress syndrome
patients. Crit Care Med 38:13001307
24. Ferguson ND, Fan E, Camporota L, Antonelli M, Anzueto A, Beale R, Brochard L, Brower R,
Esteban A, Gattinoni L, Rhodes A, Slutsky AS, Vincent JL, Rubenfeld GD, Thompson BT,
Ranieri VM (2012) The Berlin definition of ARDS: an expanded rationale, justification, and
supplementary material. Intensive Care Med 38:15731582
25. Henzler D, Pelosi P, Dembinski R, Ullmann A, Mahnken AH, Rossaint R, Kuhlen R (2005)
Respiratory compliance but not gas exchange correlates with changes in lung aeration after a
recruitment maneuver: an experimental study in pigs with saline lavage lung injury. Crit Care
9:R471R482
26. Cortes GA, Marini JJ (2013) Two steps forward in bedside monitoring of lung mechanics:
transpulmonary pressure and lung volume. Crit Care 17:219
27. Grasso S, Terragni P, Birocco A, Urbino R, Del Sorbo L, Filippini C, Mascia L, Pesenti A,
Zangrillo A, Gattinoni L, Ranieri VM (2012) ECMO criteria for influenza A (H1N1)associated ARDS: role of transpulmonary pressure. Intensive Care Med 38:395403
28. The ARDS network (2000) Ventilation with lower tidal volumes as compared with traditional
tidal volumes for acute lung injury and the acute respiratory distress syndrome. The Acute
Respiratory Distress Syndrome Network. N Engl J Med 342:13011308
29. Bellani G, Messa C, Guerra L, Spagnolli E, Foti G, Patroniti N, Fumagalli R, Musch G, Fazio
F, Pesenti A (2009) Lungs of patients with acute respiratory distress syndrome show diffuse
inflammation in normally aerated regions: a [18F]-fluoro-2-deoxy-D-glucose PET/CT study.
30. Terragni PP, Del Sorbo L, Mascia L, Urbino R, Martin EL, Birocco A, Faggiano C, Quintel M,
Gattinoni L, Ranieri VM (2009) Tidal volume lower than 6ml/kg enhances lung protection:
role of extracorporeal carbon dioxide removal. Anesthesiology 111:826835
31. Foti G, Cereda M, Banfi G, Pelosi P, Fumagalli R, Pesenti A (1997) End-inspiratory airway
occlusion: a method to assess the pressure developed by inspiratory muscles in patients with
acute lung injury undergoing pressure support. Am J Respir Crit Care Med 156:12101216
[email protected]
263
32. Mauri T, Bellani G, Grasselli G, Confalonieri A, Rona R, Patroniti N, Pesenti A (2013)

Patient-ventilator interaction in ARDS patients with extremely low compliance undergoing ECMO:
a novel approach based on diaphragm electrical activity. Intensive Care Med 39:282291
33. Bellani G, Mauri T, Coppadoro A, Grasselli G, Patroniti N, Spadaro S, Sala V, Foti G, Pesenti
A (2013) Estimation of patients inspiratory effort from the electrical activity of the diaphragm.
Crit Care Med 41(6):14831491
34. Protti A, Cressoni M, Santini A, Langer T, Mietto C, Febres D, Chierichetti M, Coppola S,
Conte G, Gatti S, Leopardi O, Masson S, Lombardi L, Lazzerini M, Rampoldi E, Cadringher
P, Gattinoni L (2011) Lung stress and strain during mechanical ventilation: any safe threshold?
Am J Respir Crit Care Med 183:13541362
35. Via G, Storti E, Gulati G et al (2012) Lung ultrasound in the ICU: from diagnostic instrument
to respiratory monitoring tool. Minerva Anestesiol 78:12821296
36. Via G, Lichtenstein D, Mojoli F et al (2010) Whole lung lavage: a unique model for ultrasound
assessment of lung aeration changes. Intensive Care Med 36:9991007
37. Bouhemad B, Liu ZH Arbelot C et al (2010) Ultrasound assessment of antibiotic-induced
pulmonary reaeration in ventilator-associated pneumonia. Crit Care Med 38:8492
38. Bouhemad B, Brisson H, Le-Guen M et al (2011) Bedside ultrasound assessment of positive
end-expiratory pressure-induced lung recruitment. Am J Resp Crit Care Med 183:341347
[email protected]
Structure of an ECMO Network

for Respiratory Support
22
Maria Grazia Calabr, Federico Pappalardo,

and Alberto Zangrillo
22.1
Introduction
In adult patients with respiratory failure refractory to conventional treatment,

ECMO represents a potentially lifesaving option, and the CESAR trial indeed indicated that significantly more patients with severe ARDS survived without severe
disability if they were transferred to a single ECMO center compared with patients
who were managed conventionally at remote hospitals [1, 2]. Nevertheless, several
questions still remain to be considered when figuring out a national ECMO network
with a structured interhospital transport.
ECMO is a supportive therapy that ensures gas exchange and systemic perfusion,
sustains the life of the patient when lung function and the native heart are dangerously compromised, and therefore should be considered to facilitate safe transfer
from outlying hospitals to referral centers.
Interhospital transportation of critically ill patients to referral centers is required
when local resources and technology are insufficient for adequate management.
Moreover, many patients requiring transfer are often too unstable to undergo conventional transport [3]. Cannulation is performed on site by the retrieving team, and
the patient is stabilized before transportation. The process requires a specialized
team and resources dedicated to retrieval. Therefore, as ECMO is an invasive, intensive form of support, it requires considerable institutional commitment.
Consequently, its use is advocated only in those patients believed to be at substantial
risk of death.
In Australia and New Zealand, during the 2009 influenza A(H1N1) winter pandemic, there was a large increase in the use of ECMO for ARDS in patients compared with the winter of 2008, which was predominantly explained by the high
M.G. Calabr (*) F. Pappalardo A. Zangrillo

Cardiothoracic and Vascular Intensive Care, San Raffaele Scientific Institute,
Via Olgettina 60, Milan 20132, Italy
[email protected]
265
266
M.G. Calabr et al.
number of patients who were transported on ECMO. Despite their illness severity
and the prolonged use of life support, most of these patients survived [4].
Establishing explicit criteria for patient selection, timing of ECMO initiation,
and optimal and safe application are first steps toward the validity of ECMO for
adults with ARDS.
Could a network organization based on preemptive patient centralization allow a
higher survival rate of patients with severe ARDS?
22.1.1 Network Organization

The recent epidemics, severe acute respiratory syndrome (SARS) and pandemic
influenza A H1N1, have highlighted the potential for respiratory viral infections to
cause severe disease with a significant risk of mortality. However, several other
viruses cause significant respiratory morbidity annually and have the potential to
produce epidemics [5].
Mortality can be reduced by adequate preparation, preventive measures, and specific plans for the organization of ICU services. The European Society of Intensive
Care Medicine Task Force suggested recommendations and standard operating procedures for the ICUs [6].
In 2009, the Italian Ministry of Health established a national network of selected
ICU centers, the Extracorporeal Membrane Oxygenation Network (ECMOnet), and
ensured economical, human, and technological resources. Two competent physicians guided ECMOnet organization and development. The ECMOnet organization
is officially operational since November 5, 2009. The Italian network was set up to
centralize all potentially severe patients in a limited number of tertiary hospitals to
provide advanced treatment options including ECMO and identify predictors of
mortality in order to define the best timing of ECMO institution.
The network consisted of 14 ICUs with ECMO capability and a national call center.
The ICU centers were selected based on their (1) experience in treating ARDS
patients, (2) experience in respiratory ECMO or presence of a cardiac surgery team
expert in ECMO, and (3) territorial distribution. Five centers ensured the interhospital transport through the whole Italian territory whenever the nearest ECMOnet
center could not handle a case. The national ECMOnet Call Center Service screened
all requests from any hospital in Italy and directed them to the closest ECMOnet
center and/or to the transportation ECMO team.
Sessions of ECMO training course, open to physicians, perfusionists, and nurses
of the ECMOnet, were organized [7].
22.1.2 Patient Selection and Referral to the ECMOnet

National recommendations and procedures for patient referral to the ECMOnet
(Table 22.1 and Fig. 22.1) and ECMO eligibility criteria (Table 22.2) were enacted
by the Italian Ministry of Health and communicated to all local sanitary authorities
and to the administration of all Italian hospitals.
[email protected]
22
Structure of an ECMO Network for Respiratory Support
Table 22.1 Recommended

national clinical criteria for
early patient centralization
267
From primary and secondary hospitals to tertiary hospitals

with ARDS treatment experience
Suspected H1N1 infection with one of the following:
1. Need for invasive mechanical ventilation with PEEP
2. FiO2 >0.6
From any non-ECMO center to ECMOnet centers
Suspected H1N1 infection with one of the following:
1. HbO2 <85 %
2. OI >25
3. PaO2/FiO2 <100 with PEEP 10 cm H2O
4. Hypercapnia and respiratory acidosis with pH <7.25
5. SvO2 or SvcO2 <65 % despite Ht >30 and administration
of vasoactive drugs
PEEP positive end-expiratory pressure, FiO2 inspired oxygen
fraction, HbO2 oxygenated hemoglobin, PaO2/FiO2 arterial partial
pressure of oxygen to FiO2 ratio, OI oxygenation index (computed
as FiO2 mean airway pressure 100/PaO2)
Originating
Hospital (OH)
Closest available
ECMO center (EC)
Need to transfer?
No
Maintain daily contact

for consultation
Yes
Transportable by OH?
No
Transportable by EC?
No
Yes
Yes
Transfer to
ECMO center
On site assessment
Yes
On-Call National
transpor team
Transfer safe
without ECMO?
No
Daily assessment
for ECMO criteria
Yes
Start ECMO
Fig. 22.1 Management algorithm for the referrals to the Italian ECMOnet system
[email protected]
M.G. Calabr et al.
268
Table 22.2 Recommended national clinical criteria for ECMO eligibility
ECMO inclusion criteria

All adult and pediatric patients with severe ARDS related to suspected influenza A(H1N1)
presenting with at least one of the following criteria despite the use of available rescue
therapies:
1. OI >30
2. PaO2/FiO2 <70 with PEEP 15 cm H2O (in patient already admitted to one of the
ECMOnet centers)
3. PaO2/FiO2 <100 with PEEP 10 cm H2O (in patients still to be transferred)
4. pH <7.25 for at least 2 h
5. Hemodynamic instability
ECMO exclusion criteria
Absolute
1. Intracranial bleeding or other major contraindication to anticoagulation
2. Previous severe disability
3. Poor prognosis because of the underlying disease (i.e., unresolved malignancy)
Relative
1. MV >7 days
PEEP positive end-expiratory pressure, FiO2 inspired oxygen fraction, PaO2/FiO2 arterial partial
pressure of oxygen to FiO2 ratio, OI oxygenation index (computed as FiO2 mean airway pressure
100/PaO2), MV mechanical ventilation
If required, an ECMOnet team (2 ICU physicians, 1 perfusionist, 1 ICU nurse)

traveled to the referring hospital to take care of the transfer. After an attempt to stabilize/improve the status of the patient, the ECMOnet team would decide to either
transport the patient conventionally or establish ECMO at the referring hospital.
Transportation was carried out via ambulance, helicopter, or fixed-wing aircraft,
depending on distance, weather conditions, and ECMOnet center resources [7].
22.1.3 ECMO Team, Ventilator Management, and Safety

The ECMO retrieval team must be very skilled and equipped for both venovenous
VV ECMO and venoarterial VA ECMO. Percutaneous peripheral VV ECMO is preferred when cardiac function is adequate or mildly depressed. Patients should always
be initiated on VV and eventually transitioned to VA ECMO if cardiac support is
required. Vessel cannulation for VV ECMO can be configurated in several ways:
dual-site or single-site approach. Beyond the hemodynamic instability, the ECMO
team also will have to assess other conditions such as severe obesity and bleeding.
Some basic facilities are required in peripheral hospitals to ensure safety: echocardiography, fluoroscopy, surgery, and blood bank. The use of a bicaval dual-lumen
cannula is recommended only if a safe environment is available [8]. Settings of
mechanical ventilation for patients on VV ECMO should minimize ventilator-associated
lung injury and permit higher degrees of protective lung ventilation.
Initiation and target for anticoagulation during transport depend on the availability of ACT. Bleeding should be thoroughly assessed before leaving the remote
hospital.
[email protected]
22
269
22.1.4 ECMOnet Activity

Between August 2009 and March 2010, 153 critically ill patients with suspected H1N1
were admitted to the ICUs of the 14 ECMOnet centers, of which 81 patients were
referred from other hospitals. 71 patients were transferred by ambulance (19 on ECMO),
8 by helicopter (all on ECMO), and 2 by fixed-wing aircraft (1 on ECMO). All patients
were transported successfully and without complications to the referral hospital.
Sixty patients (median age 39.7 12, 60 % were male) received ECMO (59
VV ECMO and 1 VA ECMO) according to ECMO eligibility criteria. All patients
fulfilled criteria for ARDS. Median duration of MV (mechanical ventilation) before
ECMO was 2 (15) days in patients with confirmed H1N1 (ARDSH1N1) and 8 (114)
days in patients with other causes of ARDS (ARDSother). Before ECMO, 42 patients
(70 %) had received at least one rescue therapy (recruitment maneuvers, prone
positioning, high-frequency oscillatory ventilation, inhaled nitric oxide, vasoactive
drugs, steroid therapy). There were no statistically significant differences between
ARDSH1N1 and ARDSother in terms of severity of respiratory failure, treatment, and
nonrespiratory organ function before ECMO.
Survival to hospital discharge in patients receiving ECMO was 68 %. Survival of
patients receiving ECMO within 7 days from the onset of mechanical ventilation
was 77 %. Survival rate in patients transported on ECMO was 81 %. There were no
statistically significant differences between patients transported on ECMO and
patients starting ECMO at the ECMOnet center in terms of severity of respiratory
failure, treatment, and outcomes. The length of MV prior to ECMO was an independent predictor of mortality.
Among the 60 patients who received ECMO, 49 (82 %) ARDSH1N1 presented a
survival rate of 71 %, the remaining 11 (18 %) ARDSother presented a survival rate
of 54 %. The median duration of ECMO support was 10 (717) days in ARDSH1N1
and 8 (321) days in ARDSother.
Sixteen patients had hemorrhagic complications, and in 10 of them, a major
bleeding event occurred, requiring blood transfusions and temporary reduction or
suspension of anticoagulation. One patient died of cerebral hemorrhage diagnosed
2 days after cannulation. Blood components were transfused in 47 (78 %) patients.
Multiple organ failure associated with sepsis was the most common cause of
death (53 %), followed by septic shock (26 %). All nonsurvivors were still on
ECMO at the time of death [7].
22.1.5 ECMOnet Score

All baseline patient characteristics, clinical parameters, and vital signs before
ECMO initiation were tested by univariate analysis. Using multivariate analysis, we
identified five statistically significant predictors of death: preECMO hospital length
of stay, bilirubin value, creatinine level, hematocrit value, and systemic mean arterial pressure.
The ECMOnet score was developed based on these variables. With the aim to be
as intuitive as possible, the score was constructed to give a result between 0 and 10
[email protected]
270
M.G. Calabr et al.
Table 22.3 The ECMOnet

score
Parameter
PreECMO hospital length of stay (days)
3
47
811
>11
Bilirubin (mg/dl)
0.15
0.160.65
0.661.15
1.161.65
1.662.15
>2.15
Creatinine (mg/dl)
0.5
0.510.8
0.811.10
1.111.14
1.411.7
1.712.0
2.012.3
> 2.3
Hematocrit (%)
>40
3640
3135
30
Mean arterial pressure (mmHg)
>90
6190
60
Partial score
0.5
1
1.5
2
0
0.5
1
1.5
2
2.5
0
0.5
1
1.5
2
2.5
3
3.5
0.5
1
1.5
2
0
0.5
1
ECMO extracorporeal membrane oxygenation
(Table 22.3). Thus, the number resulting from score calculation can be easily associated with the mortality risk. A score of 4.5 was found to be the most appropriate cutoff
for mortality risk prediction. The high accuracy of the ECMOnet score was further
confirmed by ROC analysis and by an independent external validation analysis [9].
22.2
Comment
The role of ECMO in ARDS is now well-defined: ECMO support should be considered in patients with respiratory failure refractory to conventional therapy not only
to ensure gas exchange but also to minimize ventilator-associated lung injury and its
associated multiple organ dysfunction, both crucial determinants of survival for
patients with ARDS. Several reports demonstrated that ECMO can be undertaken
without the prohibitive morbidity and adverse events seen in the 1970s.
[email protected]
22
271
To be effective, ECMO must be applied to the appropriate patient (indications,

contraindications), timing of ECMO initiation must be the most correct (not too
early, not too late), and ECMO must be correctly and safely applied (patient considerations: age, obesity, VV or VA ECMO, dual-site or single-site venovenous cannulation; monitoring and general measuresultrasound, fluoroscopy, surgery, and
blood bank).
The centralization of patients to a few selected, specifically equipped centers can
improve patient outcome, but the risks associated with patient transportation could
exceed the benefits of centralization [10]. To reduce these risks, we planned some
strategies: transferring in advance the largest proportion of patients potentially at
risk of severe respiratory deterioration according to clinical criteria and assigning
the patients to expert transportation teams, able to institute ECMO at the referring
hospital and provide safe transportation with ECMO according to precise criteria.
Some of the patients transported with ECMO might not have needed ECMO if
treated from the beginning with other rescue therapies at the referral centers, where
more therapeutic options were available. However, most of these patients were considered to be not safely transportable without ECMO.
The Italian ECMO network ensured a high survival rate of patients with severe
ARDS due to H1N1 infection treated by ECMO, a safe centralization, and created an
organization ready to challenge future possible epidemics with high demand for critical
care units with advanced respiratory support [7]. CESAR randomized trial indicated that
significantly more patients with severe ARDS survived without severe disability if they
were transferred to a single ECMO center compared with patients who were managed
conventionally. Moreover, this trial showed that patients referred for ECMO had roughly
two times longer hospital stays and twice the medical costs of those treated in the conventional management group [1]. One or two patients with influenza A (H1N1)-related
ARDS can strain the capacity of any intensive care unit (ICU) and ECMO team, particularly when ECMO availability is needed for other patients. ECMO centers should make
plans for allocation of resources: identification of early predictors of adverse outcome
could allow optimization of criteria for ECMO eligibility and referral.
The Italian ECMOnet activity showed that mortality of adult patients suffering
from influenza A (H1N1)-related ARDS undergoing VV ECMO is related to extrapulmonary organ function at the time of cannulation. PreECMO hospital length of
stay; bilirubin, creatinine, hematocrit values; and systemic mean arterial pressure
were significantly associated with mortality as assessed by multivariate analysis,
while respiratory parameters were not associated with survival.
To improve risk stratification and prediction of mortality risk at time of
VV ECMO initiation, we developed a multifactorial scoring systemthe ECMOnet
score [9].
These data provide new perspectives concerning the allocation of resources for
VV ECMO. We confirm the strong clinical perception that survival is strongly correlated to extrapulmonary organ function at the time of ECMO initiation. This
knowledge may help to identify potential candidates for ECMO support according
to their mortality risk and provides guidance to solve crucial economic and ethical
issues.
[email protected]
272
M.G. Calabr et al.
References
1. Peek GJ, Mugford M, Tiruvoipati R et al (2009) Efficacy and economic assessment of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial. Lancet 374:13511363
2. Noah MA, Peek GJ, Finney SJ et al (2011) Referral to an extracorporeal membrane oxygenation center and mortality among patients with severe 2009 influenza A(H1N1). JAMA
306(15):16591668
3. Isgr S, Patroniti N, Bombino M et al (2011) Extracorporeal membrane oxygenation for interhospital transfer of severe acute respiratory distress syndrome patients: a 5-year experience. Int
J Artif Organs 34(11):10521060
Influenza Investigators (2009) Extracorporeal membrane oxygenation for 2009 influenza
5. Lapinsky SE (2010) Epidemic viral pneumonia. Curr Opin Infect Dis 23:139144
6. Sprung CL, Zimmerman JL, Christian MD, et al, European Society of Intensive Care Medicine
Task Force for Intensive Care Unit Triage during an Influenza Epidemic or Mass Disaster
(2010) Recommendations for intensive care unit and hospital preparations for an influenza
epidemic or mass disaster: summary report of the European Society of Intensive Care
Medicines Task Force for intensive care unit triage during an influenza epidemic or mass
disaster. Intensive Care Med 36:428443
7. Patroniti N, Zangrillo A, Pappalardo F et al (2011) The italian ECMO network experience
during the 2009 influenza A(H1N1) pandemic: preparation for severe respiratory emergency
outbreaks. Intensive Care Med 37:14471457
9. Pappalardo F, Pieri M, Greco T, et al, on behalf of the Italian ECMOnet (2013) Predicting
mortality risk in patients undergoing venovenous ECMO for ARDS due to influenza A (H1N1)
pneumonia: the ECMOnet score. Intensive Care Med 39:275281
10. Singh JM, MacDonald RD (2009) Pro/con debate: do the benefits of regionalized critical care
delivery outweigh the risks of interfacility patient transport? Crit Care 13:219
[email protected]
ECMO and Thoracic Surgery
23
Alia Noorani and Alain Vuylsteke
23.1
Introduction
Extracorporeal cardiopulmonary support was conceptually born in 1930. John

Gibbon, a recently qualified doctor, saw a young patient die of a massive pulmonary
embolus following cholecystectomy. He understood that temporary mechanical
support of the cardiopulmonary system may have possibly saved her life during the
embolectomy. It is in 1953 that he successfully pioneered the use of an artificial
oxygenator and perfusion system to perform the first successful open heart procedure: an atrial septal defect repair in an 18-year-old female [1].
A further decade lapsed before it was apparent that cardiopulmonary bypass
(CPB) was not ideal to support patients for the long term. In 1971, Donald Hill and
colleagues successfully supported for 75 h a 24-year-old patient with acute posttraumatic respiratory failure following blunt trauma [2].
In 1974, a thoracic surgeon named Robert Bartlett and his colleagues developed
the concept of extracorporeal membrane oxygenation (ECMO) whilst treating neonates with respiratory distress syndrome. They successfully supported a neonate
with meconium aspiration syndrome with an entirely experimental extracorporeal
membrane oxygenator. The support lasted for 3 days and the neonate made a full
recovery [3].
With variations in cannulation, complete cardiopulmonary (veno-arterial,
VA ECMO) or respiratory support alone (veno-venous VV ECMO) can be provided. The versatility and cardiorespiratory safety provided by ECMO has made
A. Noorani
Cardiothoracic Surgery, Papworth Hospital,
Cambridge, CB23 3RE, UK
A. Vuylsteke (*)
Anaesthesia and Intensive Care, Papworth Hospital,
[email protected]
273
274
A. Noorani and A. Vuylsteke
bridging to recovery, decision or transplantation in respiratory failure or cardiogenic

shock possible. Its use has made complicated thoracic surgical procedures feasible
by providing complete pulmonary isolation even in patients with poor respiratory
reserve. Thoracic procedure that can be supported by ECMO ranges from lung
resections to airway surgery, including bridge to lung transplantation and rescue
therapy if required after transplantation.
23.2
General Principles for Cannulation
23.2.1 In Adult Patients

VV ECMO is preferred for adults with respiratory failure with minimal cardiac
depression.
VA ECMO is the preferred option in the face of moderately or severely depressed
myocardial function. Recent advances in technology and equipment include the
development and use of the Avalon (Avalon Laboratories, LLC, California) cannula, which comprises a single cannula for both venous drainage and return of oxygenated blood to be placed under fluoroscopic control into the right atrium, via a
Seldinger technique in the internal jugular vein. Alternately, two or more venous
sites can be used. Peripheral VA ECMO can be achieved via the femoral artery and
vein route [4].
23.2.2 In Paediatric Patients

VV access can be achieved via a double-lumen catheter placed into the right atrium
via the internal jugular vein for patients weighing less than 15 kg or via either the
femoral vein or the right internal jugular vein. VA cannulation is performed via a
femoral artery and the right internal jugular route with other venous cannulas being
inserted into an unused femoral or internal jugular vein [4].
23.3
Thoracic Surgery Facilitated by ECMO
23.3.1 Congenital Diaphragmatic Hernia

Congenital diaphragmatic hernia is a life-threatening condition that often requires
ECMO support for repair. Intrathoracic hypertension and subsequent cardiorespiratory compromise caused by displaced abdominal viscera can leave the neonate susceptible to cardiopulmonary collapse. The use of VA or VV ECMO in these cases
enables safe conduct of operative repair.
Although the optimal timing of ECMO (pre-repair versus during repair) has been
widely speculated, with varying results, a recent study by Bryner et al. in 2009
showed improved outcomes for patients who underwent surgery after a period of
[email protected]
23
275
stabilisation followed by liberation from ECMO [5]. Regardless of timing, studies

have repeatedly shown that ECMO support improves survival compared to no support [6]. Guner et al. compared cannulation options and type of support offered and
found that VA and VV ECMO were comparable with the added benefit of preserving the carotid artery in the VV group [7].
23.3.2 Airway Surgery

Airway surgery can pose significant challenges and obstacles for both the surgeon
and the anaesthetist, as patients undergoing tracheal resections are at a real risk of
cardiovascular collapse. Indications for surgical resection include tracheal stenosis
commonly caused by endotracheal intubation, burns, trauma as well as tracheal
malignancies. Several cases of successful institution of VV ECMO have been
described, with cannulation via the femoral vessels. These cases have included
resection of tracheal papillomas as well as for tracheal rupture following endotracheal intubation and blunt trauma [810]. In addition, ECMO may facilitate healing
without the risk of anastomotic injury from positive pressure ventilation.
23.3.3 Mediastinal Masses

ECMO may be beneficial in cases where large anterior mediastinal masses (e.g.
thyroid goitres) can cause airway compression and a significant risk of death when
the patient is paralysed to facilitate endotracheal intubation. ECMO may be instituted under light sedation or local anaesthesia, even in the upright position [1].
23.3.4 Surgery for Pulmonary Embolism

Cardiorespiratory failure due to massive pulmonary embolism can be successfully
managed with ECMO if this is instituted expeditiously, before neurological injury
ensues. Usually 4872 h of support is sufficient, by which time the emboli have
moved into segmental branches. At this time point, successful liberation can be
achieved and patients managed with thromboprophylaxis and a vena caval filter.
Conversely if there is little or no cardiopulmonary recovery after this time, pulmonary embolectomy may be performed, with ongoing ECMO support postoperatively
until normal lung function returns [1].
23.3.5 Pulmonary Infections

Posttraumatic empyema following bony fractures, subsequent pulmonary contusions and haemothorax has been successfully treated with thoracotomy, lung isolation via single-lung ventilation and VV ECMO. In a case similar to the very first
[email protected]
276
ECMO case, Brenner et al. reported a successful case of a 45-year-old male involved
in a motorcycle collision, who presented with bilateral pulmonary contusions and a
left haemothorax. Intensive care stay was protracted due to ventilator-associated
pneumonia and the development of a left-sided empyema. Successful thoracotomy
and decortication with concurrent VV ECMO support was possible with vascular
access via the femoral and internal jugular veins. This demonstrates the feasibility
of surgery in high-risk patients [11].
Souilamas et al. reported a case of pulmonary aspergilloma resistant to medical
treatment and embolisation therapy, presenting with recurrent haemoptysis. The
patients preoperative lung function was borderline, with an FEV1 of 42 % and
left lung perfusion of 75 %. Single-lung ventilation was considered difficult, and
the risk of postoperative respiratory failure was high. VV ECMO was instituted
via single-site cannulation using the Avalon cannula, and uneventful segmentectomy was performed with ECMO support being weaned after 12 h following
surgery [12].
23.3.6 Cancer
An unusual application of ECMO has been used as support for lung cancer surgery.
Kondo and colleagues in Japan described two cases of left sleeve pneumonectomy
for adenocarcinoma of the left main bronchus successfully resected with ECMO
support [13]. Lei and colleagues reported a case of a 55-year-old man who presented with haemoptysis 10 months following left pneumonectomy for squamous
cell carcinoma. Bronchoscopic findings demonstrated blood in the left bronchial
stump, and biopsy samples confirmed adenosquamous carcinoma. His risk of
imminent asphyxiation encouraged them to consider carinal resection and reconstruction under ECMO support. Cannulation was via an ipsilateral femoral artery
and vein, and the patient was successfully discharged home 10 days following
surgery [14].
23.3.7 Trauma
ECMO has been instrumental in successful salvage of a case of penetrating chest
trauma. Massive haemorrhage from a laceration to the lung parenchyma was controlled with massive transfusions and surgical repair. Subsequent transfusionassociated lung injury was managed with VV ECMO support [15].
23.3.8 Bridge to Lung Transplant (BTT)

Pre-transplant mechanical ventilation has been shown to be a significant predictor
of post-mortality, and as patients on ECMO may fall into this category, many have
been denied transplantation and have subsequently died. Recent results, however,
[email protected]
23
277
from many centres have demonstrated that patients on ECMO as a BTT have excellent survival rates, comparable to non-ECMO patients. For this reason in general, it
is advocated that carefully selected patients be offered a chance of transplant whilst
on ECMO preoperatively [16].
23.3.9 Post-lung Transplant Support

Primary graft dysfunction (PGD) following lung transplantation may occur due to a
number of factors ranging from surgical technique and trauma, borderline donor
organs or donor organ preservation, denervation injury or rejection. Despite being
subject to much research and study, survival remains poor in this subgroup of
patients, with registry data from the International Heart and Lung Transplantation
Society demonstrating that PGD is responsible for up to a third of all deaths within
90 days following transplantation [17].
Several advantages for ECMO may be evident in these cases. Firstly, high FiO2
is not necessary, thereby limiting toxicity and barotrauma to the implanted lung;
secondly, ECMO can produce a lowered capillary hydrostatic pressure thereby limiting oedema. Finally, if re-transplantation is considered, ECMO can provide support to the patient as a bridge to re-transplantation. Survival figures of up to 88 %
have been reported with the use of VV ECMO for PGD, and these compare favourably with the 94 % survival rate for lung recipients within 30 days of surgery [17].
Advancements in technology and therapy mean that this form of support is no longer considered as salvage intervention.
23.3.10 Bronchopleural Fistula or Massive Air Leak

Massive air leak in which less than half the inspired volume is expired out can be
an indication for VV ECMO. The first principle of any air leak management is to
evacuate air from the pleural space to allow approximation of lung to the chest
wall, creating adhesions. With ECMO, chest tube placement and application of
suction is necessary followed by limitation of inspiratory pressure and volume.
Larger air leaks may be initially managed in the same way, followed by videoassisted or open surgery to aid in closure. In an example of this scenario, Oey et al.
describe a case of air leak management in a difficult case whereby a 55-year-old
man with a previous right pneumonectomy for squamous cell carcinoma developed
emphysema of his remaining lung. He presented acutely to the emergency department with increasing dyspnoea, whereby a large bulla in the only lung was mistaken for a pneumothorax. Chest drain placement into this bulla led to impending
respiratory failure due to the massive air leak. In view of his poor respiratory
reserve and previous surgery, single-lung ventilation was not an option and VV
ECMO support was instituted, with concomitant VATS repair of the air leak.
ECMO support was successfully discontinued within hours postoperatively, the
patient being discharged in 5 days [18].
[email protected]
278
23.3.11 Pulmonary Thromboendarterectomy

ECMO has been used successfully to bridge and support patients undergoing pulmonary endarterectomy [19] and in patients recovering from pulmonary endarterectomy and suffering from reperfusion injury [20].
23.4
Thoracic Surgery on the ECMO Patient
Thoracic surgical procedures are sometimes necessary in the ECMO patient, and
these can be particularly hazardous due to ongoing need for anticoagulation and
detrimental effects of the circuit on platelets and the coagulation cascade. In the
largest series of ECMO patients investigated over a 16-year period, the Leicester
group retrospectively reviewed 569 patients on ECMO and determined that the need
for thoracotomy whilst on ECMO was 3.2 %, with 40 thoracotomies performed in
18 patients (19 were primary operations and 21 were reexplorations). The commonest indication for thoracotomy was bleeding post chest drain insertion (58 %), followed by uncontrolled air leak in 47 % and pleural effusion in 21 %. The commonest
primary operation was evacuation of haemothorax in 63 % of patients. The authors
noted that although the overall need for thoracotomy was 3.2 %, the in-hospital
mortality was considerable, at 39 %. For this reason they advocated that ECMO
specialists either have thoracic surgical experience or have thoracic surgeons present on-site in ECMO centres [21].
23.5
Complications of ECMO Used to Facilitate Surgery
These will be covered in more detail in other chapters in this book. Essentially,
however, they can be categorised into the following:
1. Vascular injury from cannulation
2. Air embolus
3. Excessive bleeding due to platelet dysfunction or coagulation deficiencies
4. Haemolysis from the mechanical effects of the pump on red blood cells
23.6
Conclusion
ECMO has come a long way from Gibbons first thoughts in 1930, and significant
technological advances have enabled its versatile use in the adult and paediatric
population in supporting the cardiopulmonary circulation. From a thoracic surgical
perspective, the safety provided by the level of support has encouraged more and
more diverse use of this technology and has made many more patients eligible for
surgery than ever before.
[email protected]
23
279
References
1. Stoney WS (2009) Evolution of cardiopulmonary bypass. Circulation 119(21):28442853.
doi:10.1161/CIRCULATIONAHA.108.830174
2. Hill D et al (1972) Extracorporeal oxygenation for shock lung. N Engl J Med 286:
629634
3. Bartlett RH (2009) Artificial organs: basic science meets critical care. J Am Coll Surg
196(2):171179. doi:10.1016/S1072-7515(02)01605-8
4. Extracorporeal Life Support Organization (ELSO) (2009) Patient Specific Supplements to the
ELSO General Guidelines. 124.
5. Bryner BS, West BT, Hirschl RB, Drongowski R, Lally KP, Lally P, Mychaliska GB (2009)
Congenital diaphragmatic hernia requiring extracorporeal membrane oxygenation: does timing of repair matter? J Pediatr Surg 44(6):11651171. doi:10.1016/j.jpedsurg.2009.02.022;
discussion 11711172
6. Kattan J, Godoy L, Zavala A, Faunes M, Becker P, Estay A, Fabres J et al (2010) Improvement
of survival in infants with congenital diaphragmatic hernia in recent years: effect of ECMO
availability and associated factors. Pediatr Surg Int 26(7):671676. doi:10.1007/
s00383-010-2624-3
7. Guner YS, Khemani RG, Qureshi FG, Wee CP, Austin MT, Dorey F, Rycus PT et al (2009)
Outcome analysis of neonates with congenital diaphragmatic hernia treated with veno-venous
vs. veno-arterial extracorporeal membrane oxygenation. J Pediatr Surg 44(9):16911701.
doi:10.1016/j.jpedsurg.2009.01.017
8. Smith IJ, Sidebotham D, McGeorge AD, Dorman EB, Wilsher ML, Kolbe J (2009) Use of
extracorporeal membrane oxygenation during resection of tracheal papillomatosis.
Anesthesiology 110(2):427429. doi:10.1097/ALN.0b013e3181943288
9. Roman PEF, Battafarano RJ, Grigore AM (2013) Anesthesia for tracheal reconstruction and
transplantation. Curr Opin Anaesthesiol 26(1):15. doi:10.1097/ACO.0b013e32835bd0dc
10. Korvenoja P, Pitknen O, Berg E, Berg L (2008) Veno-venous extracorporeal membrane oxygenation in surgery for bronchial repair. Ann Thorac Surg 86(4):13481349. doi:10.1016/j.
athoracsur.2008.04.018
11. Brenner M, OConnor JV, Scalea TM (2010) Use of ECMO for resection of post-traumatic
ruptured lung abscess with empyema. Ann Thorac Surg 90(6):20392041. doi:10.1016/j.
athoracsur.2010.01.085
12. Souilamas R, Souilamas JI, Alkhamees K, Hubsch J-P, Boucherie J-C, Kanaan R, Ollivier Y
et al (2011) Extra corporal membrane oxygenation in general thoracic surgery: a new single
veno-venous cannulation. J Cardiothorac Surg 6(1):52. doi:10.1186/1749-8090-6-52
13. Kondo T et al (1999) Left sleeve pneumonectomy performed through a clamshell incision with
extracorporeal membrane oxygenation for bronchogenic carcinoma: report of two cases. Surg
Today 29(8):807810
14. Lei J, Su K, Li XF, Zhou Y, Han Y, Huang LJ, Wang XP (2010) ECMO-assisted carinal resection and reconstruction after left pneumonectomy. J Cardiothorac Surg 5(1):89.
doi:10.1186/1749-8090-5-89
15. Incagnoli P, Blaise H, Mathey C, Vinclair M, Albaladejo P (2012) Pulmonary resection and
ECMO: a salvage therapy for penetrating lung trauma. Ann Fr Anesth Reanim 31(78):641
643. doi:10.1016/j.annfar.2012.03.010
16. Toyoda Y, Bhama JK, Shigemura N, Zaldonis D, Pilewski J, Crespo M, Bermudez C (2013)
Efficacy of extracorporeal membrane oxygenation as a bridge to lung transplantation. J Thorac
Cardiovasc Surg 145(4):10651071. doi:10.1016/j.jtcvs.2012.12.067
17. Hartwig MG, Walczak R, Lin SS, Davis RD (2012) Improved survival but marginal allograft
function in patients treated with extracorporeal membrane oxygenation after lung transplantation. Ann Thorac Surg 93(2):366371. doi:10.1016/j.athoracsur.2011.05.017
[email protected]
280
18. Oey IF, Peek GJ, Firmin RK, Waller DA (2001) Post-pneumonectomy video-assisted thoracoscopic bullectomy using extra-corporeal membrane oxygenation. Eur J Cardiothorac Surg
20(4):874876. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/21590658
19. Mydin M, Berman M, Klein A, Tsui S, Dunning J, Valchanov K et al (2011) Extracorporeal
membrane oxygenation as a bridge to pulmonary endarterectomy. Ann Thorac Surg
92(5):e101e103
20. Berman M, Tsui S, Vuylsteke A et al (2008) Successful extracorporeal membrane oxygenation
support after pulmonary thromboendarterectomy. Ann Thorac Surg 86:12611267
21. Joshi V, Harvey C, Nakas A, Waller D, Peek G, Firmin R (2013) The need for thoracic surgery
in adult patients receiving extracorporeal membrane oxygenation: a 16-year experience.
Perfusion. doi:10.1177/0267659113480401
[email protected]
ECMO in the Awake/Extubated Patient
24
Giorgio A. Iotti, Francesco Mojoli, and Mirko Belliato
24.1
Awake ECMO: Why? On Which Patients?
Extracorporeal membrane oxygenation (ECMO) could be the more reasonable supportive treatment of acute respiratory. By using an artificial organ to temporarily
replace the failing lung, not only we can favor lung recovery by allowing organ rest,
but in some patients we can even avoid intubation, thus preventing the well-known
complications associated with invasive mechanical ventilation (IMV).
However, for a long time the approach of replacing the failing native lung with
an artificial organ has not been widely applied, due to the invasiveness, costs, and
complexity of the extracorporeal technique. ECMO was used only as a complement
to IMV, in those relatively rare patients with persistent refractory impairment of gas
exchanges despite the optimization of mechanical ventilation settings [1]. However,
in the last decade ECMO has evolved into a less invasive, less dangerous, and userfriendly technique [2], and for this reason some groups began to consider ECMO as
a reasonable alternative to IMV in selected groups of patients, thus breaking the
paradigm of IMV as the inevitable treatment of severe respiratory failure. Awake
ECMO is the name frequently used to indicate this alternative approach of using
ECMO without IMV.
G.A. Iotti (*) M. Belliato

F. Mojoli
[email protected]
281
282
G.A. Iotti et al.
Up to 2010, published reports of ECMO in awake, non-intubated patients

were very limited: of note, only two pediatric cases of malignant mediastinal
masses treated with femorofemoral venoarterial ECMO (vaECMO) [3] and one
case of a papillomatous carinal mass treated with femoro-jugular venovenous
ECMO (vvECMO) [4] were reported. In those cases, ECMO was used as a primary respiratory support without intubation in order to avoid the severe risk of
total airway collapse associated with induction of anesthesia, intubation, and
paralysis.
In the last years, the most important indication for awake ECMO has been as
bridge to lung transplantation [57]. The application of awake ECMO as bridge to
lung transplantation aims at avoiding, if possible, the unfavorable cascade that,
starting with intubation and sedation, involves muscle deconditioning, neuromuscular complications, hospital-acquired infections, and hence poor outcome even when
the patient eventually receives the graft.
Recently, awake ECMO has been successfully used also as a bridge to recovery
from acute respiratory failure [8], especially in patients in whom the risk associated
with IMV was considered particularly high, like immunocompromised or COPD
patients [9].
24.2
Awake ECMO: When?
Awake ECMO, the differently invasive alternative to intubation and IMV, can
be used:
As a planned early choice, in patients in which the respiratory support provided
by IMV alone is predicted as insufficient, or severely injurious for the lungs, or
involving a severe risk due to intubation.
In established or resolving respiratory failure, with IMV already supplemented
by ECMO; in some patients, if ECMO was uneventful and is running smoothly,
it may be questioned whether the endotracheal tube is still really needed or if it
is advisable to maintain the extracorporeal support to help weaning from
mechanical ventilation first and only later proceed to weaning from ECMO.
24.3
Anesthesia During Cannulation for

Planned Awake ECMO
Cannulation for planned awake ECMO can be performed under local or general
anesthesia. The choice should be tailored on each individual patient.
If the clinical condition is reasonably stable and the patient is cooperative, cannulation can be performed under local anesthesia and on noninvasive ventilation
(NIV). Awake cannulation can be a judicious choice also in very severe patients in
which muscle relaxation and intubation are expected to be potentially catastrophic,
like in some patients with advanced lung fibrosis or cystic fibrosis, totally dependent
on NIV.
[email protected]
24
283
However, planned awake ECMO does not necessarily mean that the patient must
be awake during cannulation. In most cases, a safe and practical approach consists
in cannulating the patient under general anesthesia, followed by extubation within
few hours, after stabilization.
24.4
By Which ECMO Mode?
Awake ECMO has been applied in several different configurations, as pumpless

arteriovenous (av) bypass, as vvECMO, as vaECMO, and even as pumpless bypass
on the pulmonary circulation.
The pumpless av approach requires the femorofemoral implantation of a specific
low-resistance artificial lung, the Novalung interventional lung assist (ILA). This
is the easiest way to provide an extracorporeal respiratory support in an awake
patient and has been used as a bridge to lung transplantation in patients with good
cardiac function [10]. A 15-F arterial cannula combined with a 17-F venous cannula
can provide a blood flow to the ILA of approximately 1.5 l/min. This blood flow,
combined with hyperventilation of the artificial lung with up to 10 l/min of oxygen,
can provide a powerful carbon dioxide (CO2) removal of 200250 ml/min. However,
the oxygenation capacity of this approach is minimal [11]. Therefore, the indications are limited to patients in which CO2 elimination is the main problem, while
oxygenation can be effectively compensated by inhalation of supplemental oxygen,
with or without NIV or noninvasive continuous positive airway pressure (CPAP).
vvECMO is the most widely used mode for extracorporeal support in nonintubated patients. vvECMO can easily obtain a CO2 removal equal to the entire
metabolic production (Fig. 24.1), with significant reduction of respiratory drive in
terms of both frequency and amplitude [9]; this translates into a reduction of the
work of breathing and oxygen consumption and thus may also indirectly improve
oxygenation. The direct effect of vvECMO on oxygenation depends mainly on the
balance among cardiac output, extracorporeal blood flow rate, and recirculation rate
and hence on size, type, and position of the cannulas. With a 23-F jugular bi-caval
double-lumen Avalon Elite cannula and an extracorporeal blood flow of 2.5 l/min,
it is possible to obtain an oxygen transfer of approximately 100 ml/min in a spontaneously breathing adult [12]. A much higher oxygen transfer, even approaching the
level of the metabolic oxygen consumption, can be obtained with a cannulation
setup enabling a higher extracorporeal blood flow with limited recirculation. The
choice of the cannulation approach strongly affects the oxygenation power of
vvECMO; as a consequence, the patient might either need just an inhaled oxygen
supplementation or require positive pressure to improve alveolar recruitment and
oxygen transfer by the native lung.
The awake vaECMO approach has been used when lung failure was associated
with severe pulmonary hypertension and right heart failure. In a recent report of
vaECMO as a bridge to recovery or to heart transplantation for primary cardiogenic
shock or cardiac arrest [13], the treatment protocol recommended awake ECMO, if
possible, especially when a need for prolonged extracorporeal support was
[email protected]
G.A. Iotti et al.
284
450
CO2 removal (mI/min)
400
Total
350
Airway
300
250
200
150
100
ECMO
50
0
0
3
4
5
6
Sweep gas flow (I/min)
Fig. 24.1 CO2 removal at different levels of sweep gas flow (SGF) during awake vvECMO.
Increasing levels of SGF were associated with an increase of extracorporeal CO2 removal and a
simultaneous decrease of the CO2 removed by the lungs. The decrease in total CO2 removal
observed between 1 and 3 l/min of SGF resulted from a decrease in oxygen consumption (and
metabolic CO2 production) due to the drop of a high work of breathing. At an SGF of 8 l/min, the
metabolic CO2 production was nearly totally removed by the artificial lung
foreseen. In this report of 16 patients with a 50 % survival at hospital discharge,

29 % of 3,514 h on ECMO was spent as awake ECMO, and 4 patients were never
intubated. This report shows that a strategy of awake ECMO, with the purpose of
reducing the risk of infection and with the advantage of easy and prompt neurologic
assessment, is feasible in several patients with cardiogenic shock.
Finally, in cases of severe chronic pulmonary hypertension, a bridge to lung
transplantation has been obtained by surgical central implantation of a Novalung
ILA in parallel to the pulmonary circulation, between the pulmonary artery and
either the left atrium [14] or a pulmonary vein [15]. This pumpless approach,
denoted as paracorporeal artificial lung (PAL), is made possible by the combination of a powerful hypertrophic right ventricle with the low resistance typical of the
ILA oxygenator. Once stabilized after PAL implantation, patients were extubated
and maintained awake up to lung transplantation [14, 15].
24.5
Cannulation Sites
The choice about cannulation sites partly depends on whether or not the clinical
staff is oriented to implement a program of active physical therapy including patient
ambulation. If this is the case, the femoral vessels cannot be used, and vvECMO has
been commonly applied through the right internal jugular vein with a bi-caval
double-lumen Avalon Elite cannula [8, 12, 1618]. In patients needing a substantial oxygenation support and hence a high blood flow, a very large cannula (27
31 F) must be implanted, but its prolonged use is frequently complicated by deep
[email protected]
24
285
vein thrombosis of the upper extremities [19]. The cannula can be stabilized by
tunnelization of the right jugular access [12] or by using a left subclavian access
[20]. An upper-extremity approach suitable for ambulation has also been used in
one case of vaECMO, by using the left axillary vein and artery [21].
The alternative approach with two femoral cannulas, although not compatible
with ambulation, does not impede active physical therapy in bed, even including the
lower extremities. Usually total leg immobility is unnecessary, and the long wirereinforced cannulas used for femorofemoral ECMO do not kink with moderate
thigh flexion. In our practice, we recommend leg rest only in case of bleeding
through the cannula insertion points. The bi-femoral approach is used for both
vvECMO and vaECMO and is very practical in patients who need CPAP to improve
oxygenation; the free neck allows application of helmet CPAP [22], which is much
better tolerated than masks and is the perfect interface for a prolonged treatment.
Whether or not the awake patient on ECMO should also be enabled to ambulate,
it is questionable. Walking while on ECMO may expose the patient to additional
risks and requires assistance by extra personnel. In a recent Italian report of ECMO
as bridge to lung transplantation, patients bridged with awake ECMO had much
lower morbidity and an easier clinical course, with less need of postoperative IMV,
shorter posttransplant length of stay both in ICU and in hospital, and a lower incidence of critically ill polyneuropathy/myopathy than intubated patients [7]. In this
group of patients, awake ECMO was always applied with femoral cannulation;
therefore, it can be inferred that the advantages of awake ECMO are maintained, at
least partly, even if the ambulation option is excluded.
As already mentioned, the femoral approach is the most practical way to apply a
pumpless av bypass, while the PAL in parallel to the pulmonary circulation obviously requires a central cannulation.
24.6
Management of Awake ECMO
24.6.1 Setup and Start

Power and stability of the extracorporeal system are the two keystones for effectiveness and good tolerance of the treatment, and this is particularly important in awake
patients, who have the ability to react to the changes in the artificial lung function.
The amount of extracorporeal CO2 removal can be easily regulated by changing
the sweep gas flow of the artificial lung (Fig. 24.1). Extracorporeal CO2 removal and
oxygen transfer decrease patients need of alveolar ventilation and hence respiratory
drive and work of breathing [9]. If ECMO is working properly, shortly after start,
the patient feels much better, perceiving the function of the implanted artificial lung
like an apparent rapid improvement of the impaired function of his/her native lung.
In turn, in case of drop of extracorporeal support performance due to technical problems, the patient will suddenly feel and react like suffocating.
In order to achieve a favorable interaction, it is important to choose the extracorporeal support according to the failing function of the native lung. If the
[email protected]
G.A. Iotti et al.
286
oxygenation impairment is very severe, it is fundamental to implant large-size cannulas, in order to allow the setting of a high extracorporeal blood flow with good
stability and without excessive negative/positive pressures. When CO2 elimination
is the major problem, especially if lung compliance is very low, it is important to set
an adequately high ventilation of the artificial lung in order to deeply depress the
respiratory drive. The resulting reduction of a previously extremely high work of
breathing will involve a decrease of the patients oxygen consumption, thus rising
venous oxygen saturation and hence contributing to improve arterial oxygenation.
Moreover, the reduction of metabolic CO2 production (Fig. 24.1) will further contribute to unload the patient.
However, unless the patient is breathing pure oxygen with adequate CPAP,
excessive CO2 removal with vvECMO may result in oxygen desaturation due to
alveolar hypoventilation and derecruitment, as a consequence of excessive depression of the respiratory drive. In the clinical case of Fig. 24.2, this is what happened
on day 1 with a gas flow of 6 l/min and on day 7 with a gas flow of 8 l/min (corresponding to an extracorporeal CO2 removal of 192 and 208 ml/min,
respectively).
In some patients bridged to lung transplantation for advanced lung fibrosis, lung
compliance can be so low that every breath is associated with deep inspiratory
breathing efforts: in these cases, extracorporeal CO2 removal can be virtually complete, and the patient will remain nearly apneic in an atmosphere of pure oxygen
obtained by a CPAP helmet; his/her will breathe just when his/her needs to speak
or to cough.
CO2 removal / Blood flow (mI/I/min)
100
90
80
70
Day 1
60
Day 7
50
Day 14
40
30
20
10
0
0
4
6
8
Sweep gas flow (I/min)
10
12
Fig. 24.2 Relationship between sweep gas flow (SGF) and extracorporeal CO2 removal (expressed
as ml/min per liter of blood flow) in a clinical case of awake vvECMO. Blood flow on days 1, 7,
and 14 was 2.5, 2.7, and 4.2 l/min, respectively. Compared to day 1, the CO2 removal function
slightly deteriorated on day 7, while it greatly deteriorated on day 14. Oxygen desaturation (values
not showed) associated with critical alveolar hypoventilation due to excessive extracorporeal CO2
removal occurred at an SGF of 6 l/min on day 1 and 8 l/min on day 7 (arrows); it never occurred
on day 14, due to the limited CO2 removal even at high SGF
[email protected]
24
287
24.6.2 Adaptation to Changes in Extracorporeal

Support Function
Once a good compensation of the failing respiratory function is achieved, the system may become unstable because of an imbalance between the extracorporeal support and the native respiratory function.
Thanks to its pump, vvECMO is typically more stable than an av pumpless respiratory support, during which the extracorporeal blood flow (that affects CO2
removal, to a given extent) strongly depends on arterial blood pressure [11]. With an
av pumpless support, blood flow and CO2 removal can also be greatly affected by
movements of the lower extremities, with increased resistance of the cannulas due
to bending and position changes.
Although much more rarely, also during vvECMO blood flow can become unstable, due to bending of cannulas or tubes or to poor blood drainage because of hypovolemia. However, when cannulation and volemic status are adequate, the main
technical factor requiring an adjustment of ECMO settings is represented by the
progressive deterioration of the artificial lung function. In Fig. 24.2 it can be noted
how, at any level of sweep gas flow, the CO2 removal efficiency was much lower on
day 14 than on ECMO start or on day 7; moreover, on day 14 no further increase of
CO2 removal could be obtained by increasing the gas flow above 4 l/min, indicating
the need for oxygenator replacement. The progressive deterioration of the artificial
lung may affect first CO2 removal and only later oxygenation (Fig. 24.3). Up to a
given extent, the deterioration of the extracorporeal CO2 removal function can be
compensated by increasing the ventilation of the artificial lung. Manual adjustment
of ECMO gas flow to compensate for progressive deterioration is particularly
important in the awake patient, who otherwise would react, thus becoming
unstable.
24.6.3 Adaptation to Changes in Patient State

Modern ECMO systems are designed to be safe, simple, and stable, but not to adapt
to changes in patients needs.
Fever is a particularly challenging condition during awake ECMO. Except for
the very-low-resistance Novalung ILA, ECMO circuits are combined with a temperature-regulated heat exchanger that effectively compensates the heat loss continuously taking place through the extracorporeal system. The heat exchanger may
also display the power, expressed in watts, instantaneously applied in order to warm
the blood at the set temperature.
In sedated patients on ECMO, a rise of internal body temperature above the heat
exchanger set point (usually set at 37 C) is rare, even when the heat exchanger is
not provided with an active cooling function. When fever tends to rise, the heat
exchanger decreases or even stops to heat the blood, and so the entire ECMO circuit
becomes a passive cooler. Therefore, unless patients thermogenesis is extremely
high, in sedated patients on high-flow ECMO, fever may have little evidence: any
[email protected]
G.A. Iotti et al.
288
700
1st oxygenator
2nd oxygenator
12
3rd
600
10
8
400
6
300
4
Flow (I/min)
PO2 mmHg
500
200
PO2, out
Sweep gas flow
Blood flow
100
0
0
10
20
30
40
50
60
70
80
90 100 110 120 130 140 150
Time (days)
Fig. 24.3 Prolonged awake vvECMO. The first oxygenator was replaced after 90 days, when the
oxygenation performance was still satisfactory (see PO2 in blood leaving the oxygenator PO2,
out), while CO2 removal had deteriorated, requiring a progressive increase of sweep gas flow
(SGF) up to 10 l/min (corresponding to a progressive increase of extracorporeal ventilation/perfusion ratio from 1:1 to 1:2.5). With the second oxygenator, the slow deterioration involved simultaneously both CO2 removal and oxygenation
potential increase of body temperature is prevented by extracorporeal circulation,

while shivering may be blocked by sedation. Interestingly, information about this
not expressed fever can be provided by the ECMO heat exchanger that will show
that the set blood temperature (for instance, 37 C) is achieved with zero or nearzero watts.
During awake ECMO the issue of fever is totally different compared to sedation, because the awake patient has much more power for fighting against the
thermostatic action of the ECMO system. Typically, patients on awake ECMO
react by strong shivering, which involves a major increase of oxygen consumption and tissue oxygen extraction, with relevant drop of venous oxygen saturation. In this framework, the consequent deterioration of oxygenation could be
wrongly attributed to a malfunction of the vvECMO system (Fig. 24.4).
Therefore, the typical picture of an awake patient on ECMO with an attack of
fever includes shivering, respiratory distress with worsening of arterial blood
gases, apparent worsening of the oxygen output from the oxygenator, no rise (or
minimum rise) in body temperature, and zero watts on the display of the
ECMO heat exchanger.
In such cases, a fast and relevant clinical improvement can be obtained by administering an antipyretic, by increasing the ventilation of the artificial lung, but above
all by temporarily increasing the set point of the ECMO heat exchanger (from 37 to
38 C or even more), aiming at a better coupling between artificial thermoregulation
and patients thermoregulation. A kind of artificial fever thus matches the bodys
attempt to rise its temperature, instead of opposing to that. Usually this simple
[email protected]

PO2/FO2,out
SO2,in
Intraoxygenator shunt
450
PO2/FO2
400
100
90
80
350
70
300
60
250
50
200
40
150
30
100
20
50
10
0
0
SO2 (%)
500
289
Shunt (%)
24
0
10 20 30 40 50 60 70 80 90 100 110
Time (h)
Fig. 24.4 Awake vvECMO with sepsis and not expressed fever: instability due to high oxygen
consumption. Five events of apparent loss of performance by the oxygenator: the oxygen output,
evaluated in terms of output blood PO2 during ventilation with 100 % oxygen (PO2/FO2), considerably dropped. A complete analysis showed that, on the contrary, the oxygenator performance
(evaluated in terms of intra-oxygenator shunt) was quite stable, while each event was marked by a
significant drop of oxygen saturation of the venous blood entering the oxygenator (SO2, in). During
each event, the patient was shivering, while the internal body temperature was stable at 37 C all
the time
change stops shivering and thus greatly limits the increase in patients oxygen consumption and CO2 production. After some hours, this artificial fever should be
progressively reduced while checking patients tolerance.
24.6.4 Ventilatory Support

The ventilatory support of patients on awake ECMO ranges from simple oxygen
therapy to NIV. As discussed above, the choice is wider when the oxygenating
power of ECMO is large. On the contrary, when extracorporeal oxygenation is limited, it is necessary to maximally exploit the residual oxygenation function of the
native lung by applying CPAP or NIV.
A positive pressure treatment with CPAP or NIV can be chosen also with the aim
of preventing a progressive derecruitment of the lungs. Actually, also during awake
ECMO as a bridge to lung transplantation, it is wise and recommendable to maintain the native lung open and working, if possible, while respecting its functional
and mechanical limits and avoiding injuries due to mechanical ventilation.
Helmet CPAP [22] with active humidification is our preferred option for ventilatory support during awake ECMO. For a prolonged continuous application, helmet
CPAP has less complications and is tolerated much better than mask CPAP or NIV.
The good tightness that is typical of the helmet allows to maintain easily a high level
of CPAP and/or a high oxygen concentration. High-flow humidified nasal oxygen is
an interesting option whenever patients need to remove the helmet.
[email protected]
G.A. Iotti et al.
290
24.6.5 Circuit Replacement

The replacement of the ECMO circuit is more critical during awake treatments that
in intubated patients. Patients on high-flow vvECMO have a particularly high risk
for severe deterioration of oxygenation, as well as patients on vaECMO. It must be
considered that, during the time of ECMO stop, a patient highly dependent on the
extracorporeal support will suddenly feel like suffocating, if awake.
Therefore, circuit replacement must be performed at least under sedation and
NIV. In some cases, temporary invasive ventilation can be considered. In rare cases,
the replacement has been performed under hypothermia.
All the clinical staff must be well trained to perform a fast and smooth replacement, as well as ready to face the possible complications, including cardiac arrest.
Maximum reduction of the time of ECMO stop is important; nonetheless, when
restarting the bypass, a gradual progression in increasing the pump speed must be
observed.
24.6.6 Psychological Issues

Compared to life support treatments delivered under sedation, awake ECMO may
expose the patient to relevant psychological stress. Patients on awake ECMO share
with their relatives the awareness that their own life depends on the ECMO machine
as well as on the education and skill of the ICU staff. In bridge to lung transplantation, this is added to the fear that donors organs may not be made available in time.
Patients may differently express, or hide, their fear of dying, despair, and depression.
Patients can be relieved by family presence, psychological assistance, anxiolytics, and antidepressants. It is very important that the entire ICU staff gains patient
trust.
24.7
Conclusion
By improving safety and ease of use, the latest ECMO technology allows to consider ECMO even before intubation in selected cases of lung failure. Up to now,
bridge to lung transplantation has been the wider field of application of ECMO in
non-intubated patients. Using ECMO in awake patients raises several specific issues
that must be well known to warrant a successful clinical outcome.
References
1. The Acute Respiratory Distress Syndrome Network (2000) Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory
distress syndrome. N Engl J Med 342:13011308
2. MacLaren G, Combes A, Bartlett RH (2012) Contemporary extracorporeal membrane oxygenation for adult respiratory failure: life support in the new era. Intensive Care Med
38:210220
[email protected]
24
291
3. Wickiser JE, Thompson M, Leavey PJ et al (2007) Extracorporeal membrane oxygenation

(ECMO) initiation without intubation in two children with mediastinal malignancy. Pediatr
Blood Cancer 49:751754
4. Collar RM, Taylor JC, Hogikyan ND et al (2010) Awake extracorporeal membrane oxygenation for management of critical distal tracheal obstruction. Otolaryngol Head Neck Surg
142:618620
7. Crotti S, Iotti GA, Lissoni A et al (2013) The organ allocation waiting time during extracorporeal bridge to lung transplantation affects outcomes. Chest. doi:10.1378/chest.12-1141
8. Garcia JP, Kon ZN, Evans C et al (2011) Ambulatory veno-venous extracorporeal membrane
oxygenation: innovation and pitfalls. J Thorac Cardiovasc Surg 142:755761
9. Crotti S, Lissoni A, Tubiolo D et al (2012) Artificial lung as an alternative to mechanical ventilation in COPD exacerbation. Eur Respir J 39:212215
10. Fischer S, Hoeper MM, Bein T et al (2008) Interventional lung assist: a new concept of protective ventilation in bridge to lung transplantation. ASAIO J 54:310
11. Mller T, Lubnow M, Philipp A et al (2009) Extracorporeal pumpless interventional lung
assist in clinical practice: determinants of efficacy. Eur Respir J 33:551558
12. Garcia JP, Iacono A, Kon ZN et al (2010) Ambulatory extracorporeal membrane oxygenation:
a new approach for bridge-to-lung transplantation. J Thorac Cardiovasc Surg 139:e137e139
13. Mojoli F, Venti A, Pellegrini C et al (2013) Hospital survival and long term quality of life after
emergency institution of venoarterial ECMO for refractory circulatory collapse. Minerva
Anestesiol 79:11471155
14. Camboni D, Philipp A, Arlt M et al (2009) First experience with a paracorporeal artificial lung
in humans. ASAIO J 55:304307
15. Taylor K, Holtby H (2009) Emergency interventional lung assist for pulmonary hypertension.
Anesth Analg 109:382385
16. Turner DA, Cheifetz IM, Rehder KJ et al (2011) Active rehabilitation and physical therapy
during extracorporeal membrane oxygenation while awaiting lung transplantation: a practical
approach. Crit Care Med 39:25932598
17. Hayes D Jr, Kukreja J, Tobias JD et al (2012) Ambulatory venovenous extracorporeal respiratory support as a bridge for cystic fibrosis patients to emergent lung transplantation. J Cyst
Fibros 11:4045
18. Hoopes CW, Gurley JC, Zwischenberger JB et al (2012) Mechanical support for pulmonary
veno-occlusive disease: combined atrial septostomy and venovenous extracorporeal membrane oxygenation. Semin Thorac Cardiovasc Surg 24:232234
19. Shafii AE, Brown CR, Murthy SC et al (2012) High incidence of upper-extremity deep vein
thrombosis with dual-lumen venovenous extracorporeal membrane oxygenation. J Thorac
20. Shafii AE, McCurry KR (2012) Subclavian insertion of the bicaval dual lumen cannula for
venovenous extracorporeal membrane oxygenation. Ann Thorac Surg 94:663665
21. Mangi AA, Mason DP, Yun JJ et al (2010) Bridge to lung transplantation using short-term
ambulatory extracorporeal membrane oxygenation. J Thorac Cardiovasc Surg 140:713715
22. Bellani G, Patroniti N, Greco M et al (2008) The use of helmets to deliver non-invasive continuous positive airway pressure in hypoxemic acute respiratory failure. Minerva Anestesiol
74:651656
[email protected]
ECMO as a Bridge to Lung Transplant
25
Stefania Crotti and Alfredo Lissoni
25.1
Introduction
Lung transplant (LTx) has now become a therapeutic tool for many end-stage respiratory failures. In the last 10 years, the number of lung transplants has doubled, and
the 1-year survival rate has significantly increased from 75 % to more than 80 % [1].
However, the mortality rate of patients on waiting lists for lung transplant is still
high, due to the scarcity of lung donors and the difficulty in managing patients with
end-stage respiratory failure.
The progression of respiratory disease can cause different clinical scenarios,
ranging from hypercapnia alone to both hypercapnia and hypoxia, with eventual
associated pulmonary hypertension. In some cases, medical therapy is unable to
control the worsening of respiratory failure, and an artificial respiratory support is
needed until a suitable organ becomes available.
Until a few years ago, when the noninvasive ventilation (NIV) failed, the invasive mechanical ventilation (IMV) was the only artificial respiratory support for
bridging patients to lung transplant. The mortality rate of the patients awaiting LTx
on mechanical ventilation varies from 13 % [2] up to 90 % [3]. These data reflect
the extreme variability of the pathophysiologic and mechanic characteristics of the
lung affected by the transplantable disease. The invasive mechanical ventilation
of the emphysematous lung increases the risk of pneumothorax, which could lead to
a high-flow air fistula, thus making the mechanical ventilation itself ineffective.
Furthermore, positive pressure IMV can worsen a severe case of pulmonary hypertension. Lastly, patients with end-stage respiratory failure (i.e., cystic fibrosis, CF)
are particularly exposed to the ventilator-induced lung injury (VILI) and the
S. Crotti (*) A. Lissoni

Dipartimento di Anestesia, Rianimazione (Intensiva e Subintensiva) e Terapia del Dolore,
Fondazione IRCCS Ca Granda Ospedale Maggiore Policlinico of Milan,
Via F. Sforza 35, Milan 20122, Italy
[email protected]
293
294
S. Crotti and A. Lissoni
ventilator-associated pneumonia (VAP) due to the pulmonary chronic inflammation

caused by the recurrent infections [4].
Although the use of the extracorporeal respiratory support as bridge to lung
transplant was first reported more than 30 years ago [5], the increasing number of
transplant centers that have started ECMO programs to bridge critically ill patients
to lung transplant is only recent, thanks to the technological improvements of the
extracorporeal systems.
The ECMO support as bridge to lung transplant can provide the following:
1. Extending the bridge option to the patient with a high risk of complications
related to IMV (i.e., bullous emphysema and severe pulmonary hypertension)
2. Decreasing VILI when associated to IMV
3. Avoiding the risk of VAP when patient is maintained conscious and not
intubated
4. Reducing respiratory distress in conscious patients both on NIV or IMV
Nevertheless, the bridge to lung transplant still remains a controversial topic. The
main reason is that the patients who need an artificial respiratory support have a
higher perioperative risk of death. Thus, considering the scarcity of the lung donors,
the use of organs in patients with lower postoperative survival could be considered
a waste of resources. However, health economists have proposed the concept of
benefit of survival to improve the effectiveness of health budgets [6]. This concept
introduces the idea that, in a contest of scarcity of resources, these resources should
be reserved to patients with higher survival benefit. This idea has been also applied
to organ donations. The lung allocation score (LAS) has been introduced in the
United States in 2005 [7]. According to this score system, the organ is assigned to
the recipient with the higher predicted survival at 1 year, hence balancing the risk of
death on waiting list with the survival after the transplant. However, larger and longterm survival studies are needed to determine if the LAS system improves overall
allocation and survival for patients interested in lung transplant. In other countries,
such as Italy, critically ill patients are prioritized through urgent lists.
Although patients undergoing ECMO bridging to LTx have to be considered critically ill, recent larger case series (Table 25.1) report a promising short- and mediumterm survival, which justifies the increasing use of the ECMO tool by many centers.
25.2
Indications, Timing, and Patient Selection
The main indications for the ECMO bridging to lung transplant include all the irreversible end-stage respiratory diseases, which have a rapid worsening of the respiratory
function, as well as severe pulmonary hypertension with right-sided ventricular failure.
So far, there is no evidence regarding the correct timing of the artificial respiratory support. Some centers start the ECMO bridging when the clinical condition
deteriorates to the point that the patients life expectancy could be considered less
than 2448 h without intubation and/or extracorporeal support.
Careful patient selection is needed to maximize the results of ECMO bridging
and to avoid a waste of viable donor lungs. In our center we have applied ECMO
bridging in LTx candidates in whom respiratory failure is the sole relevant organ
[email protected]
25
295
Table 25.1 Experiences of the use of ECMO as bridge to LTx (series with more than ten patients)
Bridge
duration
(days)
15 8
(432)
5 (125)
Reference
Fischer et al. [8]
Year
2006
No. of
patients
12
Cypel et al. [9]
2010
10
Ricci et al. [10]
2010
12
Hammainen
et al. [11]
Bermudez et al.
[12]
Fuehner et al.
[13]
Lang et al. [14]
2011
16
2011
17
VA (3), VV
(2), AV (4),
PA-LA (4)
13.5 14.2 AV (6), Decap
(448)
(6)
16.8 19.2 VV, VA
(159)
3.2 (149) VV (8), VA (9)
2012
26
9 (145)
2012
34
Javidar et al. [15] 2012
18
Shafii et al. [16]
2012
19
Toyoda et al.
[17]
Hoopes et al.
[18]
2012
31
2013
31
Crotti et al. [19]
2013
25
Type of
ECMO
AV
VV (14), VA
(12)
4.5 (163) VV (18), VA
(14), AV (1),
comb (4)
11.5
VV (13), VA
(618)
(5)
6.5 (116) VV (11), VA
(8)
7.1 10.1 VV (15), VA
(0.146)
(9), 7 NA
11 (253) VV (13), VA
(12) PA-LA
(3), comb (4)
24 31
VV (19), VA
(1157)
(2), AV (4)
Successful
bridge (%)
83
1-year
survival (%)
80
100
70
25
NA
81
92
NA
74
77
80
(6 months)
60
76
72
74
100
(3 months)
75
77
74
NA
92
68
76
AV arteriovenous, VV venovenous, VA venoarterial, PA-LA pulmonary atrium-left atrium, comb

combination
failure (aside from right-sided heart failure) and in whom no exclusion criterion for
LTx or ECMO is present.
The most common behavior is to use this therapeutic option for patients: already
candidates for lung transplant, young, free from other organ failures, and with a
good expectancy of physical recovery after transplant. This is because factors, such
as age, organ dysfunction, some infections, and physical status, are risks for postoperative death, even for patients on the standard list [1]. It is worth noting that the
postoperative survival of patients is affected by the recipients age. In fact recipients
older than 55 have a significantly higher risk of death at 1 year after transplant [1].
The coexistence of other organ dysfunctions alongside the end-stage respiratory
failure and/or pulmonary hypertension decreases the 1-year survival after LTx.
Particularly the need of hemofiltration and the use of inotropic drugs in the perioperative period are risk factors for the 1-year mortality [1].
Septic shock is a contraindication for lung transplants, whereas the presence of
leukocytosis and fever in the immediate preoperative period slightly increases the
[email protected]
296
risk of death for postoperative sepsis [20]. In CF patients, the pre-transplant pulmonary colonization with Burkholderia cepacia genomovar III increases postoperative
mortality, whereas the colonization with other B. cepacia strains, or multi- or panresistant Pseudomonas aeruginosa, or methicillin-resistant Staphylococcus aureus,
or Aspergillus fumigatus, does not affect postoperative survival [20].
Investigations of muscle function in lung transplant recipients reveal decreased
muscle mass and strength with a persistent limitation in exercise capacity at 1 year
after LTx. Pre-existing peripheral muscle dysfunction in chronic lung disease is one
of the determinants of the postoperative impairment in physical status, suggesting a
need for physical therapy to optimize muscle strength and functional capacity during the pre-transplant period [21].
25.3
ECMO Configuration
Deciding the type of extracorporeal support must take into account the characteristics of the respiratory failure, the presence of pulmonary hypertension, and the concomitant right-sided heart failure. The ECMO approach and the specific device
chosen will fit with the clinical patients condition (Fig. 25.1).
Most end-stage respiratory diseases, requiring lung transplant as a unique therapeutic option, lead to a mainly hypercapnic respiratory failure. When the noninvasive trial fails, the extracorporeal support becomes a valid option to bridge these
Candidate selection
Hypercapnic
respiratory failure
Hypoxic/hypercapnic
respiratory failure
PAH
+ pulmonary
hypertension
le
ab
St
Pumpless AV ECMO
(MV or SB)
If hypoxemia
VV ECMO
(MV or SB)
Un
Unstable
sta
ble
VA ECMO
(MV or SB)
Lung transplant/recovery/withdrawal or death
Fig. 25.1 Algorithm for selection of ECMO support configuration
[email protected]
VA ECMO
BAS,
or PA-LA
pumpless
25
297
patients to LTx. In the hypercapnic patients, the two configurations most often used
are the pumpless arteriovenous (AV) and the venovenous VV ECMO.
The pumpless arteriovenous (AV) approach has been recently utilized successfully. The first report is by Fisher et al. who described the use of the novel pumpless device in 12 patients between 2003 and 2005 [8]. They reported a very
successful bridge to LTx (10 of 12 patients underwent a transplant) and an 80 %
1-year survival. Other groups experienced the low blood flow CO2 removal devices
reporting different successful rate. Ricci et al. described 12 patients treated with
pumpless AV or decap navigation system Decap [10]. They were able to reverse the
respiratory acidosis, but 8 of 12 patients died prior to transplant. Cypel et al. recently
published four patients successfully bridged to LTx with the AV mode and another
four patients that required a conversion to a VV or venoarterial VA ECMO during
the bridging period [9].
In the AV ECMO setting, the blood is driven through the circuit by the difference
between the femoral arterial pressure and the venous reinfusion pressure. This
requires an adequate patients mean arterial pressure. Moreover, in AV mode the
ECMO blood flow cannot be actively changed (maximum value of 11.5 L/min),
limiting the extracorporeal oxygen supply. If the patients oxygenation drops during
the bridge period, a switch to a VV configuration becomes necessary, with possible
bleeding problems at the arterial cannula removal site. In AV mode, heart performance must be good enough to increase the patients cardiac output as requested by
the high-flow fistula, as the AV ECMO could be considered. Nevertheless, AV
mode offers an optimum and effective CO2 removal that can be titrated by changing
the sweep gas flow from a minimum level up to 12 L/min.
Some end-stage respiratory failure could be both hypercapnic and hypoxic. The
VV configuration permits controlling respiratory acidosis through an adequate CO2
removal and provides oxygen supply by varying the ECMO blood flow up to 45 L/
min. If recirculation is minimized, even in the presence of severe hypoxic disease,
the VV mode could be feasible in managing the patients until a suitable organ
becomes available. Most of the transplant centers have recently increased the use of
this approach for all the patients without right-sided heart failure. Up until 2007,
only single-center case reports had been published [22]. In the last 45 years, many
centers around the world have reported an increased experience [819]. As shown
in Table 25.1, in which the case series of more than ten patients are listed, the VV
approach is the more frequently used in each center. The VV configuration could
also support blood gasses during the intraoperative management of the patient, and,
whenever possible, it should be preferred to a central VA bypass. This offers some
advantages, such as less need of intraoperative anticoagulation, simpler technical
management, and better evaluation of graft performance after LTx.
Moreover, the VV configuration could be chosen to bridge to LTx hypercapnic and
hypoxic patients even in the presence of secondary moderate to severe pulmonary
hypertension, if they are hemodynamically stable without signs of right-sided heart
failure. In fact, as previously described in 11 severe ARDS patients, the increase of the
mixed venous oxygen tension reduces the pulmonary vascular resistance as a
consequence of the decreasing hypoxic pulmonary vasoconstriction [23]. Furthermore,
[email protected]
298
the normalization of the arterial carbon dioxide tension and consequently of the pH
could reduce the pulmonary pressure itself. We observed the decrease of the pulmonary vascular resistance in many ARDS patients just after the ECMO onset, and we
recently report a case of a bridge to LTx with VV ECMO in severe pulmonary hypertension secondary to acute on chronic respiratory failure Oral communication at Euro
Elso meeting Stockholm 2013.
A different clinical scenario is the primary pulmonary arterial hypertension
(PAH). In these patients, the pathophysiology of the underlying disease is difficult
to be reversed and, in the end-stage of the disease, often leads to a severe right-sided
ventricular failure. The ECMO configuration used in this condition is the VA mode,
which can offer a complete support of gas exchange and hemodynamics. If peripheral VA approach is performed, the upper body and the coronary vessels could be
poorly oxygenated (Harlequin syndrome). The improvement of the upper body oxygenation can be achieved by the insertion of an additional cannula into the internal
jugular vein, modifying the circuit in the hybrid VAV [24]. The VA ECMO is the
former approach described to bridge patients to lung transplantation even in the
exclusively respiratory disease. Now, this support is usually reserved to the cardiacrespiratory failure. In sheep suffering from respiratory failure and right ventricle
dysfunction, Camboni D et al. have recently described the use of VV ECMO associated to the balloon atrial septostomy (BAS) and so creating a right to left shunt and
then unloading the right ventricle [25]. However, the right to left shunt drives an
amount of desaturated blood in the left side, and this could worsen the hypoxia.
A recent approach described by some authors in few patients is pulmonary artery to
left atrium (PA-LA) configuration [26]. This is a pumpless system, which takes
advantage of the higher PA pressure to drive blood through the artificial lung
(NovalungR) into the left circulation. This configuration needs a sternotomy and a
central cannulation to be performed.
The cannulation site often depends on the ECMO configuration chosen and for
each configuration on the centers experience and preference.
The pumpless AV ECMO is always performed with a peripheral cannulation draining from the femoral artery and reinfusing into the femoral vein. Instead, VV configuration can be obtain with four different cannulation approaches. The femoro-femoral
approach, which is our centers preference, is safe, easy to perform (even in the
awake patients) (Fig. 25.2), but does not permit patients standing out of bed and
their ambulation. The femoro-jugular cannulation has been shown to be more effective than the reverse approach (i.e., jugular-femoral), reaching higher venous oxygen
saturation at the same blood flow [27]. The use of the double-lumen cannula has
recently been implemented thanks to the technological improvement of the cannulas
manufacture and its use in the awake patient [28]. This cannulation type follows the
sedation and intubation of the patient; some adverse events have been recently
described during the cannula insertion; the correct position is difficult to be achieved
and needs frequent echocardiographic or fluoroscopic daily checks. However, with
this cannulation patients can stand up, ambulate, and follow an active physical
therapy.
The VA ECMO can be performed with a peripheral or a central cannulation.
Some centers in the peripheral cannulation routinely insert additional distal perfusion catheter in place to prevent leg ischemia. The central cannulation, which
[email protected]
25
299
Fig. 25.2 Awake ECMO as bridge to lung transplant
follows a sternotomy, is the more frequent intraoperative approach, and it is required

when the PA to LA configuration is performed.
25.4
Patient Management
The main goal in patient management during ECMO as bridge to lung transplant is to
avoid the development of further organ dysfunctions aside from respiratory failure
that increases morbidity and mortality during bridge and after transplant. We recently
reported a retrospective on two Italian centers and their study showing the effects of
the duration of the extracorporeal bridge to lung transplant [19]. In this study, we
observed that the patients who awaited organ allocation less than 2 weeks had a higher
survival and a better postoperative course compared to the patients with longer bridge
duration. This was related to the clinical impairment during the bridge course as
shown by the difference in the pre-transplant SOFA score, higher in the group that
awaited the organ allocation longer. And so, the objective to decrease the ECMO
bridge duration justifies the use of high-priority lists for these critically ill patients.
In mechanically ventilated patients, the protective ventilation of the native lungs,
although they may not recover their function, could reduce adverse events such as
right cardiac dysfunction, sepsis, and multiple organ failure. However, mechanical
ventilation per se can worsen the patients clinical condition. The Hannover group
has recently described the awake ECMO approach, maintaining the patient nonintubated, on spontaneous breathing [13]. This awake ECMO strategy reduces the
[email protected]
300
well-known drawbacks related to sedation and mechanical ventilation. Furthermore,

awake patients can follow an active physical therapy in order to reduce neuromuscular dysfunction. This approach can be used regardless of the underlying disease
and the ECMO configuration chosen, as reported by the same group during AV or
VV approach and also in the VA ECMO-supported patients [13, 29]. In this ECMO
setting, attention has to be made to titrate the CO2 removal, both seeking respiratory
distress relief and avoiding pulmonary hypoventilation, which can occur by increasing the ECMO CO2 removal too much. The use of the awake strategy could be
more complicated in the CF patients due to the huge amount of pulmonary secretions that are more difficult to be cleared if the patient reduces his own ventilation
as well as the ability to cough. The need to stand out of bed and ambulate patients
is better satisfied by the use of the double-lumen cannula, recently implemented in
some transplant centers.
If the other organ functions deteriorate, intensive care is needed for organ support as hemofiltration for renal dysfunction or inotropic drugs for cardiovascular
failure. When the patient deteriorates during the ECMO bridge, support withdrawal
has to be taken into account and the patient can be deemed unfit for transplant.
25.5
Outcome
Technological advances allowing a safer application of the ECMO circuits, as well

as changes in choosing the bridge configuration, involving the widest use of the VV
support (instead of the more invasive and complicated VA approach), have lead to
considering ECMO as a valid tool to bridge the critically ill patients to lung transplant. The data so far available on postoperative survival, in some experienced centers quite equal to the patients on the standard waiting lists, are satisfactory enough
to justify its use. The main effort in increasing bridge success and the postoperative
course is to improve the preoperative clinical condition through the physical exercise and the maintenance of multiple organ function integrity. The awake strategy
seems to obtain better results in terms of postoperative morbidity, but not enough
data on differences in survival rate are yet available. Many centers believe in this
strategy, following an active physical therapy in awake and ambulatory patients.
A wider experience, a larger case series, and long-term follow-ups are needed to
confirm the use of the ECMO to bridge critically ill patients to lifesaving transplants. Furthermore, a careful evaluation of each individual patient is necessary to
correctly choose candidates for lung organ allocation, so as to avoid a waste of very
scarce resources.
References
1. Christie JD, Edwards LB, Kucheryavaya AY et al (2012) The registry of the International
Society for Heart and Lung Transplantation: 29th adult lung and heart-lung transplant
report-2012. J Heart Lung Transplant 10:10731086
[email protected]
25
301
2. Vermeijden JW, Zijlstra JG, Erasmus ME et al (2009) Lung transplantation for ventilatordependent respiratory failure. J Heart Lung Transplant 28:247351
3. Stern JB, Mal H, Groussard O et al (2001) Prognosis of patients with advanced idiopathic
pulmonary fibrosis requiring mechanical ventilation for acute respiratory failure. Chest
120:213219
4. Del Sorbo L, Boffini M, Rinaldi M et al (2012) Bridging to lung transplantation by extracorporeal support. Minerva Anestesiol 78:243250
5. Veith F (1977) Lung transplantation. Transplant Proc 9:203208
6. Chandra A, Jena AB, Skinner JS (2011) The pragmatists guide to comparative effectiveness
research. J Econ Perspect 25:2746
7. Egan TM, Kotloff RM (2005) Pro/Con debate: lung allocation should be based on medical
urgency and transplant survival and not on waiting time. Chest 128:407415
8. Fischer S, Simon AR, Welte T et al (2006) Bridge to lung transplantation with the novel pumpless interventional lung assist device NovaLung. J Thorac Cardiovasc Surg 131:719723
9. Cypel M, Waddel TH, de Perrot M et al (2010) Safety and efficacy of the NovaLung
Interventional Lung Assist (iLA) device as a bridge to lung transplantation. J Heart Lung
Transplant 29:S88
10. Ricci D, Boffini M, Del Sorbo L et al (2010) The use of CO2 removal devices in patients awaiting lung transplantation: an initial experience. Transplant Proc 42:12551258
11. Hammainen P, Schersten H, Lemstrom K et al (2011) Usefulness of extracorporeal membrane
oxygenation as a bridge to lung transplantation: a descriptive study. J Heart Lung Transplant
30:103107
12. Bermudez CA, Rocha RV, Zaldonis D et al (2011) Extracorporeal membrane oxygenation as a
bridge to lung transplant: midterm outcomes. Ann Thorac Surg 92:12261231
14. Lang G, Taghavi S, Aigner C et al (2012) Primary lung transplantation after bridge with extracorporeal membrane oxygenation: a plea for a shift in our paradigms for indications.
Transplantation 93:729736
15. Javidar J, Brodie D, Iribarne A et al (2012) Extracorporeal membrane oxygenation as a bridge
to lung transplantation and recovery. J Thorac Cardiovasc Surg 144:716721
16. Shafii AE, Mason DP, Brown CR et al (2012) Growing experience with extracorporeal membrane oxygenation as a bridge to lung transplantation. ASAIO J 58:526529
17. Toyoda Y, Bhama JK, Shigemura N et al (2013) Efficacy of extracorporeal membrane oxygenation as a bridge to lung transplantation. J Thorac Cardiovasc Surg 145:10651071
18. Hoopes CW, Kukreja J, Golden J et al (2013) Extracorporeal membrane oxygenation as a
bridge to pulmonary transplantation. J Thorac Cardiovasc Surg 145:862868
19. Crotti S, Iotti GA, Lissoni A et al (2013) The organ allocation waiting time during extracorporeal bridge to lung transplantation affects outcomes. Chest 144(3):10181025
20. Orens JB, Estenne M, Arcasoy S et al (2006) International guidelines for the selection of lung
transplant candidates: 2006 updatea consensus report from the pulmonary scientific council
of the International Society for Heart and Lung Transplantation. J Heart Lung Transplant
25:745755
21. Wickerson L, Mathur S, Brooks D (2010) Exercise training after lung transplantation: a systematic review. J Heart Lung Transplant 29:497503
22. Jackson A, Cropper J, Pye R et al (2008) Use of extracorporeal membrane oxygenation as a
bridge to primary lung transplant: 3 consecutive, successful cases and a review of the literature. J Heart Lung Transplant 27:348352
23. Benzing A, Mols G, Brieschal T et al (1997) Hypoxic pulmonary vasoconstriction in nonventilated lung areas contributes to difference in hemodynamic and gas exchange responses to
inhalation of nitric oxide. Anesthesiology 86:12541261
24. Stohr F, Emmert MY, Lachat ML et al (2011) Extracorporeal membrane oxygenation for acute
respiratory distress syndrome: is the configuration mode an important predictor for the outcome? Interact Cardiovasc Thorac Surg 12:676680
[email protected]
302
25. Camboni D, Akay B, Sassalos P et al (2011) Use of venovenous extracorporeal membrane

oxygenation and an atrial septostomy for pulmonary and right ventricular failure. Ann Thorac
Surg 91:144149
26. Strueber M, Hoeper MM, Fischer S et al (2009) Bridge to thoracic organ transplantation in
patients with pulmonary arterial hypertension using a pumpless lung assist device. Am J
Transplant 9:853857
27. Rich PB, Awad SS, Crotti S et al (1998) A prospective comparison of atrio-femoral and
femoro-atrial flow in adult venovenous extracorporeal life support. J Thorac Cardiovasc Surg
116:628632
28. Diaz-Guzman E, Hoopes CW, Zwischenberger JB (2013) The evolution of extracorporeal life
support as a bridge to lung transplantation. ASAIO J 59:310
[email protected]
Low-Flow ECMO and CO2 Removal
26
Vito Fanelli, Andrea Costamagna, Pierpaolo P. Terragni,

and V. Marco Ranieri
26.1 Introduction
In 1952, the outbreak of poliomyelitis in Copenhagen recalls the first clinical scenario in which positive pressure mechanical ventilation (MV) was extensively
applied to patients with acute respiratory failure. MV was able to restore the balance
between weak respiratory muscles and inspiratory workload [1]. Paradoxically,
depending on the impairment of respiratory system mechanics, the ventilator setting, per se, may cause macroscopic and microscopic lung injury, negatively affecting patient outcome [2, 3].
Thereafterfrom the theoretical consideration that resting the lung could be the
optimal solution to avoid ventilator-induced lung injuryfinding a non-ventilatory
strategy consisting of an extracorporeal support therapy that is able to clear CO2 and
ensure oxygenation is needed. In fact, there are different extracorporeal life support
techniques (ECLS) ranging from low blood flow extracorporeal membrane oxygenation (low-flow ECMO) for extracorporeal CO2 removal up to full high flow for
oxygenation support (high-flow ECMO). In addition, less invasive systems designed
only for extracorporeal CO2 removal (ECCO2R) have been developed. The objectives of this chapter are to review current concepts of low-flow ECMO and ECCO2R
systemsremoving only carbon dioxide, with little to no impact on oxygenation
and provide the rationale for their application in patients with ARDS and COPD and
thoracic surgery candidates and for its use as bridge to lung transplantation.
V. Fanelli, MD, PhD A. Costamagna, MD P.P. Terragni, MD V.M. Ranieri, MD (*)

Department of Anesthesia and Critical Care, University of Turin, Turin, Italy
Citt della Salute e della Scienza, Ospedale S. Giovanni
Battista-Molinette, corso Dogliotti 14, Turin 10126, Italy
[email protected]
303
304
V. Fanelli et al.
26.2 P
hysiology of CO2 Removal During Extracorporeal
Support
CO2 produced within aerobic cellular respiration is transported from tissues to the
lungs through blood in three main ways: 90% of the total CO2 is conveyed as bicarbonate ion (HCO3) that originates both from carbonic acid dissociation and CO2
hydration according to the following equation:

H + + HCO3- H 2 O + CO 2
The remaining 10% is transported as CO2 freely dissolved in blood (5%) and as
carbaminic compounds, arising from the interaction of CO2 and the aminic groups
of circulating proteins, like hemoglobin.
Under physiological conditions, the amount of CO2 produced during systemic
metabolism (VCO2) is 200ml/min, which can increase to a value of 30% higher in
pathological condition. As a consequence, the concentration of CO2 in arterial blood
is about 480ml/l, and in mixed venous blood it rises 10% to a value of 520ml/l,
corresponding to a partial pressure of CO2 (PCO2) of 40 and 45mmHg, respectively.
It is possible to argue that 1l of blood contains two times the amount of CO2 produced in the whole body per minute. Theoretically, it is possible to remove 250ml/
min of CO2 by the filtration of 500ml/min of blood during extracorporeal support
depending on specific efficiency of the systems. In light of these considerations,
CO2 removal may be achieved with blood flow rate lower than 1l/min and ventilating the membrane lung with a fresh gas flow (sweep gas) to maintain the CO2 gradient across the artificial lung [4, 5].
26.3 C
linical Applications of Low-Flow ECMO
and ECCO2R Systems
Currently, extracorporeal life therapies are indicated for the temporary support of
the pulmonary function to gain time waiting for full recovery of organ function or
as a bridge to organ transplantation. In the following section, several diseases, in
which low-flow ECMO and ECCO2R systems are used as therapeutic interventions,
will be discussed (Table26.1).
26.4 Acute Respiratory Distress Syndrome (ARDS)

In patients with ARDS, extracorporeal life support techniques are currently applied
for two main reasons: first, as rescue therapy for life-threatening hypoxemia that is
refractory to conventional mechanical ventilation and, second, as adjunctive therapy
to conventional mechanical ventilation to allow ultra-protective ventilation that
minimizes VILI, limiting end-inspiratory lung stretch. As previously mentioned,
these two applications of ECLS are coupled with different complexities and
[email protected]
[email protected]
VT=430 (360450) ml
RCT
Retrospective
Prospective
observational
Morris etal.
[9] (n=40)
Bein etal.*
[10] (n=90)
Florchinger
etal. [11]
(n=159)
PIP=34.57.2cmH2O
RR=2917/min
RR=3715/min
A-V ECCO2R
pumpless
Blood flow
1.92.1l/min
Sweep gas
613l/min
A-V ECCO2R
pumpless
Blood flow
1.92.5l/min
Blood Flow 2.4 l/min
Pressure control
inverse ratio ventilation
VT=380
(320470) ml
PIP=35
(3139)cmH2O
RR=23(1739)/min
VT=402144 ml
VV ECMO-LFPPV
VV ECMO-LFPPV
Blood flow 1.3l/min
Sweep gas 16l/min
VV ECMO-LFPPV
Blood flow 1l/min
Sweep gas 15l/min
VV ECMO-LFPPV
ECCO2R settings
Int arm (n=19)
PIP=37.76.3 cmH2O
RR=27 (2143)/min
VT=453134 ml
PIP=38 (3540)cmH2O
VT=720150 ml
PIP=519 cmH2O
PEEP=132 cmH2O
RR=1624/min
C arm (n=19):
conventional
Int arm (n=21)
VT=8.90.6 ml/kg
PIP=553 cmH2O
Prospective
interventional
RR=3/min
(minute ventilation
0.71.5l/min)
PIP<3545 cmH2O
PEEP 1525cmH2O
RR=35/min
VT=29035 ml
PIP=366 cmH2O
RR=4/min
MV settings
After extracorporeal
support
Brunet etal.
[8] (n=23)
RR=1622/min
(minute ventilation
1425l/min)
Baseline
Prospective
interventional
Case report
Type of study
ARDS
Gattinoni
etal. [6]
(n=3)
Gattinoni [7]
(n=43)
Author
CI-coated
system
Single
bolus+CI
CI
Single
bolus+CI
Single bolus
Single
bolus+CI
Heparin
Table 26.1 Clinical studies that evaluate the efficacy of ECMO and ECCO2R systems in different diseases
a, b
a, b
A, B
c, d
(continued)
Complications
No benefit in a, b
terms of
survival,
hospital, and
ICU length
of stay
A, B
A, B
A, B
Outcome
26 Low-Flow ECMO and CO2 Removal

305
Prospective
interventional
Int arm:
VT=6.30.2 ml/kg
Int arm:
VT=4.20.3 ml/kg
Hommel etal. Postsurgical

Case report
[15] (n=4)
bronchial fistulae
with ARDS
Wiebe etal.
Apneic
Observational
[16] (n=10)
oxygenation
during thoracic
surgery
Thoracic surgery
Author
Clinical setting Type of study
Plateau pressure
Plateau pressure
29.11.2 cmH2O
25.01.2 cmH2O
RR=31.22.3/min
RR=37.01.9/min
Bein etal.
Multicenter
C arm (n=39)
Int arm (n=40)
[13] (n=76)
RCT
and
VT=3 ml/kg
Int arm (n=40)
VT=68 ml/kg
Chronic obstructive pulmonary disease (COPD)
Author
Type of study
MV settings
Burki etal.
Prospective
Noninvasive MV at risk for intubation and
[14] (n=20)
observational
invasive MV after failing weaning trial
Terragni et al.
[12] (n=32)
Table 26.1(continued)
VT<4 ml/kg PBW

RR=2322/min
PIP<30 cmH2O
Apneic oxygenation
MV settings
ECCO2R settings
VV ECCO2R
Blood flow
0.4374 l/min
A-V ECCO2R
pumpless
Blood flow 1.3l/min
ECCO2R
settings
A-V
ECCO2R
pumpless
A-V
ECCO2R
Blood flow
1.580.3 l/
min
Sweep gas
612l/min
Heparin
Single
bolus+CI
VV ECCO2R
Single
Blood flow 0.20.4l/ bolus+CI
min
Sweep gas 8l/min
[email protected]
Single bolus A, B
+ CI
A, B
CI
Hypothermia
None
Outcome Complications
Complications
A, thrombocytopenia
Heparin
Outcome
A
Low rate of
noninvasive
MV failure
A, B, C
A, B
306
V. Fanelli et al.
MV arm (n=34) conventional MV settings

ECMO arm (n=26) in awake patients
MV settings
ECCO2R settings
A-V ECCO2R (n=6)
VV-ECCO2R (n=6)
Int arm (n=26)
VV ECMO group
Blood flow
2.8(2.33.8) l/min
Sweep gas
4.0(3.05.5) l/min
[VA ECMO group
Blood flow
3.3(3.34.1) l/min
Sweep gas
3.3(2.54.0) l/min]
Heparin
Single
bolus+CI
Single
bolus+CI
Complications
a, b, e, sepsis-like
syndrome multiorgan
failure and need of renal
replacement therapy
Outcome
A, B
ECMO was
effective as
bridge
strategy to
LT and may
result in
better
survival
Abbreviations: A Improvement in CO2 removal in patients treated with ECCO2R, a Hemorrhageminor bleeding, B Improvement in ventilatory management,
allowing protective MV thus reducing risk of VILI in patients treated with ECCO2R, b Hemorrhagemajor bleeding, C Improvement in ventilator-free days
(VFD) at 60 days in patients with PaO2/FiO2 <150 in patients treated with ECCO2R, c Complications related to ECCO2R device such as plasma leak or circuit
clotting, C arm Control arm, CI Continuous infusion, CO Cardiac output, d Lower limb ischemia, e Hemodynamic impairment, Flow Blood flow through the
ECCO2R circuit, Int arm Interventional arm, MV Minute ventilation (l/min), P max Maximum peak pressure allowed, PEEP Positive end-expiratory pressure,
PIP Peak inspiratory pressure, RCT Randomized controlled trial, RR Respiratory rate, VILI Ventilator-induced lung injury, VT Tidal volume expressed in ml or
in ml/kg of predicted body weight
*
Values are expressed as median (interquartile range)
Bridge to lung transplantation

Author
Type of study
Ricci etal.
Observational
[17] (n=12)
Fuehner etal.* Retrospective
[18]
(n=60)

307
[email protected]
308
V. Fanelli et al.
invasiveness of ECLS apparatus. In fact, high blood flow (36l/min) ECMO provides full oxygenation and CO2 removal in severe ARDS patients with life-threatening hypoxemia. On the other hand, low blood flow (0.41l/min) extracorporeal CO2
removal (ECCO2 R) apparatus removes all CO2 produced by metabolism with minimal effect on oxygenation. In the following paragraphs, the current application of
ECCO2 R in patients with ARDS will be reviewed. The reader is referred to other
chapters for high-flow ECMO in ARDS.
Seminal animal observations which showed that extracorporeal CO2 removal
was able to progressively reduce the respiratory drive [19] were the rationale to
apply an ultra-protective ventilation strategy to humans, with the objective of keeping the lung at rest and avoiding the deleterious consequences of mechanical ventilation. In 1980, Gattinoni etal. described a small series of three patients in whom
ECMO with a blood flow of around 1.3l/min reduced the needs of inspiratory flow
from 15 to 20l/min to only 0.71.5l/min, thus avoiding lung overdistension and
barotrauma [6]. Six years later, the same authors showed convincing evidence that
low-flow ECMO was a feasible and safe adjunctive therapy to conventional mechanical ventilation that was confined to merely support oxygenation [7]. In fact,
43patients with early (1week) and late (23weeks) ARDS were treated with a
veno-venous low-flow ECMO as last therapy after they failed conventional mechanical ventilation and other rescue therapies as high-frequency jet ventilation and
inverse ratio ventilation. In particular, normal PaCO2 values, a blood flow of
2030% of cardiac output on ECMO, a sweep gas of 15l/min, and only five positive pressure limited breaths at 3545cmH2O were obtained. Oxygenation was
completely dissociated from CO2 clearance; in fact, it was achieved through an
apneic oxygenation that consisted of an oxygen flow rate of 23l/min delivered at
the level of the carina and PEEP similar to the value of mean airway pressure before
the beginning of bypass. These patients had a mortality of 52.1%, and this rate was
lower than a reported value of 80% in a previous NIH trial [7]. Of note, conflicting
results of the above mentioned studies could be explained by the fact that despite
patients having similar baseline characteristics, patients in the NIH trial were treated
with artero-venous bypass and, more importantly, with full conventional mechanical ventilation that did not prevent barotrauma and VILI.
Moreover, these results were confirmed in a subsequent observational study of
23 patients with severe ARDS [8]. In this study, Brunet and colleagues reported a
mortality rate of 52% in patients with severe ARDS who were treated with low-
flow ECMO and 5 breaths of conventional MV. The ECLS strategy reduced PaCO2
from 56 to 41mmHg (p<0.0001) and tidal volume from 730 to 284ml. Of note,
four of 23 patients died during ECMO because of major bleeding complications [8].
In the same period, Morris and colleagues performed a single-center RCT, in which
40 patients with severe ARDS were randomized to receive conventional mechanical
ventilation or inverse ratio ventilation (IPRV) and ECMO. No significant difference
in survival (42% conventional vs. 32% ECMO group) was demonstrated between
the two arms of study [9].
Despite the disappointing results of the two RCTs, the H1N1 influenza pandemic
in 2009 dramatically prompted physicians to support young patients with severe
[email protected]
309
ARDS by viral pneumonia with ECMO. However, there has been a growing interest
to look for simpler and less invasive ECCO2R devices to assist conventional MV in
order to minimize VILI without all of the risks associated with ECMO.
Toward this end, two new devices, pumpless extracorporeal lung assist (PECLA)
(iLA Membrane Ventilator, Novalung GmbH, Hechingen, Germany) and Decap
(Hemodec, Salerno, Italy), have been proposed. Novalung is a low-resistance A-V
bypass (of approximately 15mmHg at 2.5l/min blood flow) with a diffusion membrane of poly-4-methyl-1-penten (surface of 1.3m2), through which an oxygen flow
of 112l/min can be administered. It requires relatively small cannulae (1519F
arterial and 1719F venous) and a priming volume of 200ml of crystalloids.
Florchinger and colleagues published their 10-year experience using Novalung
as a life-support device for patients with acute respiratory failure. In total, 159
patients were treated for 7.06.2days and 70% of them had acute respiratory failure. Both PaO2 (from 7237mmHg to 20361mmHg) and PaCO2 (from
6724mmHg to 39 17mmHg) improved at the end of treatment with a significant
reduction of minute ventilation (from 13.84.8L/min to 9.84.8l/min) [11]. These
data established Novalung as a reliable method of supporting ARDS patients. These
results were confirmed in a retrospective study of 92 patients with ARDS in whom
hypoxia and hypercapnia were promptly corrected; however, lower limb ischemia
was reported in 24% of cases [10]. More recently, in a multicenter RCT involving
10 hospitals, the same authors evaluated whether a Vt of 3ml/kg PBW enhances
lung protection. In the treatment arm, patients with ARDS were ventilated with a Vt
of 3ml/kg and PECLA support to obviate respiratory acidosis. In the control arm,
patients were ventilated according to ARDSnet strategy (6ml/kg PBW) without an
extracorporeal device. The primary outcomeventilator-free days at 30 and 60
dayswas not different between the study groups. Unfortunately, the trial was
stopped after 3years, after 79 out of 106 patients were enrolled, as previously
planned. However, a post hoc analysis showed that patients with severe hypoxemia
(PaO2/FiO2<150) had improved ventilator-free days compared to controls [13].
A minimally invasive system that removes CO2, Decap, has been proposed as an
efficacious system that provides ultra-protective mechanical ventilation [12]. It consists of a modification of the continuous VV hemodialysis machine. Access is
accomplished through a single double-lumen catheter inserted in the femoral vein.
Blood flow is via a nonocclusive roller pump. Blood circulates through a membrane
oxygenator (total membrane surface is 0.33m2) then through a hemofilter. The ultrafiltrate from the hemofilter is recirculated into the pre-gas exchanger blood, increasing CO2 removal. In an observational study, Terragni and colleagues demonstrated
that Decap treatment associated with ultra-protective mechanical ventilation (VT
<6ml/kg PBW) may mitigate VILI. In 32 patients with ARDS ventilated with a VT
of 6ml/kg PBW, those with plateau pressures between 28 and 30cmH2O had their
VT reduced to achieve plateau pressures between 25 and 28cmH2O. Respiratory
acidosis (pH7.25) derived from VT reduction was managed with Decap for at least
72h. Alternatively, patients who already had plateau pressures between 25 and
28cmH2O continued to receive protective MV (VT of 6ml/kg PBW). In the ECCO2R
group (ten patients), PaCO2 (mean 50mmHg) and pH (mean 7.32) were normalized,
[email protected]
310
V. Fanelli et al.
and VT was reduced from 6 to 4ml/kg PBW, and plateau pressure decreased from 29
to 25cmH2O (p<0.001). Moreover, there was a significant reduction in the percentage of lung hyperinflation in the treatment group at 72h, as demonstrated by CT scan
and pulmonary cytokines (p<0.01). No patient-related complications occurred in
patients receiving Decap treatment [12]. Of note, Zanella and colleagues showed a
new method to improve the CO2 removal in the system mentioned above. In six pigs,
blood acidification through continuous infusion of 0.5N lactic acid increased the
CO2 removal capacity of the membrane lung up to 70% [20].
26.5 Chronic Obstructive Pulmonary Disease (COPD)

COPD represents the third cause of death in the USA [21], and hypercapnic respiratory failure worsens prognosis and increases mortality [22, 23].
Noninvasive positive pressure ventilation (NIPPV) is able to reduce the number
of patients requiring intubation, and it represents the standard of treatment of
patients with acute exacerbation of COPD [24]. However, NIPPV fails in 2550%
of COPD patients who still require invasive positive pressure ventilation (IPPV)
[25] with a poor prognosis in terms of hospital survival [26].
During acute exacerbation of COPD, dynamic hyperinflation (due to shorter
expiratory time) and expiratory flow limitations (due to compression of small airways) increase the value of intrinsic PEEP (PEEPi), thus worsening respiratory acidosis and work of breathing.
As previously mentioned, the ECCO2R techniques were applied first in patients
with hypoxic respiratory failure, but their role in hypercapnic respiratory failure has
not been properly addressed.
In a proof of concept study, Burki etal. examined the application of a novel,
single venous catheter ECCO2R system in patients with acute exacerbation of
COPD. The authors demonstrated that this single-catheter, low-flow ECCO2R system was able to partially remove CO2, improving some aspects of this complicated
disease. The study involved 6 ICUs (1 in India and 5 in Germany), and 20 hypercapnic patients with COPD were treated with ECCO2R in three different ways. In the
first group (7 patients), patients failing NIV were treated with a minimally invasive
extracorporeal CO2 removal system with the aim of avoiding tracheal intubation. In
the second group (2 patients), patients who could not be weaned from NIV were
treated. In the third group (11 patients), patients who underwent invasive ventilation
received ECCO2R to improve the process of liberation from mechanical ventilation.
Blood flow through the system ranged between 117 and 587ml/min (meanSD of
430.573.7ml/min) with an elimination of CO2 ranging between 14 and 121ml/
min (meanSD of 82.016.3ml/min). The rate of complications was similar to
those observed with central venous catheterization. Of interest, this minimally invasive ECCO2R strategy prevented intubation in all patients who were failing NIV,
serving as feasible adjunctive therapy during acute exacerbation of COPD [14].
Similar strategies are the subject of an ongoing observational phase 2 Italian
study (Extracorporeal CO2 Removal in COPD ExacerbationDECOPD,
[email protected]
311
NCT01422681), in which the efficacy of the Decap Smart in reducing the intubation
rate or the duration of invasive mechanical ventilation in patients with COPD,
treated either with NIV or invasive mechanical ventilation (IMV), will be addressed.
Thus, pending results will hopefully expand and confirm the efficacy of minimally
invasive extracorporeal CO2 removal strategy in COPD exacerbation.
26.6 Thoracic Surgery

More recently, extracorporeal CO2 removal support has been successfully applied to
patients who underwent elective or emergent thoracic surgery. In this scenario,
extracorporeal CO2 removal support makes one lung ventilation feasible in patients
with severe impairment of alveolar ventilation in whom prolonged apneic ventilation is also required.
In an observational study, Wiebe etal. showed that the pumpless extracorporeal
CO2 removal deviceNovalungwas a safe and feasible tool in patients with poor
respiratory function who underwent major thoracic surgery. In fact, ten patient candidates for elective or emergent thoracic surgery were supported by pumpless
artero-venous bypass. Indications for the intraoperative application of the Novalung
were hypoxemic respiratory failure (defined by PaO2/FiO2 <150) that impeded one
lung ventilation (6 patients) and prolonged periods of apnea required during surgery
(four patients). The Novalung produced only small oxygen transfer (49.24.4ml/
min) and remarkable clearance of CO2 (12118ml/min). In fact, paCO2 decreased
significantly from 58.427 to 379mmHg and pH increased from 7.240.2 to
7.450.04. Blood flow across the Novalung A-V was 1.580.3l/min (1.22.2l/
min) and the blood pressure was maintained with a low dose of norepinephrine [16].
The only device-related complications observed were a retroperitoneal hematoma
after percutaneous removal of an arterial cannula and a delayed extubation due to
hypothermia.
These results were expanded in a small series of patients with ARDS following
the repair of the bronchopleural fistula. In fact, Hommel and colleagues showed that
ultra-protective mechanical ventilation, consisting of low inspiratory pressure plus
extracorporeal assistance, minimized the risk of barotrauma and VILI. Moreover, at
4days of treatment, there was a significant reduction of tidal volume (from 5.1ml
pre-bypass to 2.8ml PBW) and plateau pressure (from 32.4 to 27.6cmH2O), while
PaCO2 decreased from 73.6 to 54mmHg and pH was normalized [15].
26.7 Bridge to Lung Transplantation

Extracorporeal carbon dioxide removal systems have been recently used in patients
with advanced lung diseases waiting for lung transplantation. In fact, lung transplant waiting lists hold a high mortality rate, which is the highest compared to other
solid organ transplants. The progression of the underlying disease, or the occurrence of supra-imposed lung infections, often requires the use of invasive
[email protected]
312
V. Fanelli et al.
mechanical ventilation as a support measure [27]. Evidence exists that mechanical

ventilation before lung transplantation can further enhance the initial lung damage
and lead to multiorgan dysfunction, resulting in clinical unsuitability for lung transplantation (too sick to be transplanted) [28]. In addition, MV is a significant risk
factor for mortality after lung transplantation and certainly needs to be avoided.
In several reports at single centers, extracorporeal supports ranging from minimally invasive CO2 removal system to full ECMO have been used to support patients
with advanced respiratory failure unresponsive to maximal MV support, as defined
by high concentration of inspired oxygen, severe hypercapnia, use of nitric oxide,
and high levels of PEEP [17]. Of interest, midterm patient outcomes and allograft
function of patients treated with ECMO were similar to those who did not receive
this support before lung transplant [29, 30].
Recently, in a non-matched case control study, Fuehner and colleagues showed
that ECMO is a valuable supportive therapy as a bridge to lung transplantation for
non-intubated patients with end-stage lung disease. Compared to 34 historical control patients supported with invasive mechanical ventilation, 26 awake patients
treated with ECMO had better outcome in terms of graft survival. In fact, survival
at 6months after lung transplantation was significantly higher in the ECMO group
(80% vs. 50%, p=0.02). Moreover, duration of support (9 vs. 15days) and percentage of patients that were transplanted (23% vs. 29%) were similar in both
groups [18]. These data suggest that supporting patients with extracorporeal support
instead of invasive mechanical ventilation may improve patients homeostasis and,
consequently, graft function.
26.8 Complications of ECMO and ECCO2R Support

Since 1976, a constant decrease in ARDS mortality in patients treated with extracorporeal devices was reported [31, 32]. In recent years, crucial progresses have
been made in the conception and construction of ECMO circuits, heparin-bonded
cannulae, rotary pumps, and small efficient long-lasting oxygenators. They are now
simpler and safer and require less anticoagulants, thus being associated with fewer
bleeding complications. These improvements in ECMO technology, along with an
intensive training of the ICU staff, lead to safer and more effective ECMO
application.
The ECMO experience maturated during the past H1N1 pandemic spread, producing a wide array of data on ECMO complications. Hemorrhagic complications
occurred in half of the patients during ECMO therapy, with the most common
sources being ECMO cannulation sites in up to one-fifth of all patients. Less bleeding, observed in a tenth of the patients, was found in the gastrointestinal tract, respiratory tract, vaginal site, and brain [33]. Moreover, in a retrospective study of the
ELSO registry from 1986 to 2006, complications of ECMO were analyzed: over
time, circuit rupture and pneumothorax became less frequent, but circuit clots, renal
insufficiency, renal replacement therapies, pulmonary hemorrhage, inotropic medications, hyperglycemia, extremes of pH, arrhythmias, and hypertension became
[email protected]
313
more common [34]. Another widespread complication during ECMO is represented

by infection. A retrospective study of post-surgery ECMO showed that infection
complications occurred in 42 patients (62%), with the most common sites being the
respiratory tract in 30 patients (44%) and bloodstream in 14 patients (21%).
Moreover, non-ECMO catheter-related infections occurred in 13 patients (19%)
and ECMO cannula-related infections in 7 patients (10%) [35]. Another cannula-
related adverse event was reported during the positioning of a bicaval dual-lumen
cannula for VV ECMO via the internal jugular vein: the cannula migration into the
right ventricle, resulting in right ventricular rupture and cardiac tamponade [36].
PECLA device can remove up to 50% of the total body CO2 production, with a
blood flow of around 1l/min, and can transfer 2060ml/min of oxygen. Moreover,
these devices can also be utilized in patients with high risk of bleeding because the
cannula and device are heparin coated and also because they require the same dose
of anticoagulation therapy used in immobilized critically ill patients [37].
In a retrospective experience published by Bein etal. in 2006, it was reported that
the incidence of serious malfunctioning was at 24.4%. The complications associated were bleeding during cannulation and formation of hematoma at the insertion
site. An episode of ischemia and compartmental syndrome of a lower limb were
also reported [33]. As recently reported by Sanchez etal. during the PECLA period,
several plasma markers increased significantly. Even though these changes were
statistically significant, none of the parameters reached pathologic levels within or
after the PECLA period. Moreover, alterations showed no clinical relevance and
normalized within 48h after its removal [10].
Many fewer complications have been described with new veno-venous
ECCO2R systems because of their less invasiveness. In a prospective study on
patients with ARDS, in which the ECCO2R system Decap was used to allow protective mechanical ventilation with tidal volume less than 6ml/kg, no patientrelated complications were observed. In three cases, the 14-French double-lumen
catheter had to be replaced by two 8-French simple-lumen catheters (one for each
femoral vein) due to recirculation issues (two cases) and catheter kinking (one
case). The membrane clotting, which has been observed in three patients, did not
result in additional transfusion [12]. Moreover, in a cohort of 20 COPD patients
treated with the ECCO2R systemHemolungsingle death due to internal hemorrhage from vessel perforation was described, although it is not directly attributable to this device. In addition, heparin-induced thrombocytopenia was described
as a more common side effect, although this was not associated with clinically
significant bleeding [14].
26.9 Conclusions
The continuous technological improvement of ECMO and ECCO2R systems caused
a widespread use of this technique despite the lack of solid clinical data assessing
safety, efficacy, and cost-effectiveness. Based on the goal of treatment to achieve,
modulation of treatment invasiveness seems to be prudent; in fact, as early in the
[email protected]
314
V. Fanelli et al.
story of extracorporeal support, this powerful life-sustaining therapy may be

obscured whether risks as serious hemorrhagic complications will stand above benefits. Clinical trials assessing the balance between risk and benefit of this powerful
therapy are needed.
References
1. Tobin MJ (2001) Advances in mechanical ventilation. N Engl J Med 344(26):19861996
2. Tremblay LN, Slutsky AS (1998) Ventilator-induced injury: from barotrauma to biotrauma.
Proc Assoc Am Physicians 110(6):482488
3. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung
injury and the acute respiratory distress syndrome. The Acute Respiratory Distress Syndrome
Network (2000) N Engl J Med 342(18):13011308
4. Gattinoni L etal (1983) A new look at therapy of the adult respiratory distress syndrome:
motionless lungs. Int Anesthesiol Clin 21(2):97117
5. Pesenti A, Patroniti N, Fumagalli R (2010) Carbon dioxide dialysis will save the lung. Crit
Care Med 38:S549S554
6. Gattinoni L etal (1980) Treatment of acute respiratory failure with low-frequency positive-
pressure ventilation and extracorporeal removal of CO2. Lancet 2(8189):292294
7. Gattinoni L etal (1986) Low-frequency positive-pressure ventilation with extracorporeal CO2
removal in severe acute respiratory failure. JAMA 256(7):881886
8. Brunet F etal (1993) Extracorporeal carbon dioxide removal and low-frequency positive-
pressure ventilation. Improvement in arterial oxygenation with reduction of risk of pulmonary
barotrauma in patients with adult respiratory distress syndrome. Chest 104(3):889898
9. Morris AH etal (1994) Randomized clinical trial of pressure-controlled inverse ratio ventilation and extracorporeal CO2 removal for adult respiratory distress syndrome. Am J Respir Crit
Care Med 149(2):295305
10. Bein T etal (2006) A new pumpless extracorporeal interventional lung assist in critical hypoxemia/hypercapnia. Crit Care Med 34(5):13721377
11. Florchinger B etal (2008) Pumpless extracorporeal lung assist: a 10-year institutional experience. Ann Thorac Surg 86(2):410417
12. Terragni PP etal (2009) Tidal volume lower than 6 ml/kg enhances lung protection: role of
extracorporeal carbon dioxide removal. Anesthesiology 111(4):826835
13. Bein T etal (2013) Lower tidal volume strategy (approximately 3 ml/kg) combined with extracorporeal CO2 removal versus conventional protective ventilation (6 ml/kg) in severe ARDS:
the prospective randomized Xtravent-study. Intensive Care Med 39(5):847856
14. Burki NK etal (2013) A novel extracorporeal CO(2) removal system: results of a pilot study
of hypercapnic respiratory failure in patients with COPD. Chest 143(3):678686
15. Hommel M etal (2008) Bronchial fistulae in ARDS patients: management with an extracorporeal lung assist device. Eur Respir J 32(6):16521655
16. Wiebe K etal (2010) Thoracic surgical procedures supported by a pumpless interventional
lung assist. Ann Thorac Surg 89(6):17821788
17. Ricci D etal (2010) The use of CO2 removal devices in patients awaiting lung transplantation:
an initial experience. Transplant Proc 42(4):12551258
18. Fuehner T etal (2012) Extracorporeal membrane oxygenation in awake patients as bridge to
lung transplantation. Am J Respir Crit Care Med 185(7):763768
19. Kolobow T etal (1977) Control of breathing using an extracorporeal membrane lung.

Anesthesiology 46(2):138141
20. Zanella A etal (2009) Blood acidification enhances carbon dioxide removal of membrane
lung: an experimental study. Intensive Care Med 35(8):14841487
[email protected]
315
21. Qaseem A etal (2011) Diagnosis and management of stable chronic obstructive pulmonary
disease: a clinical practice guideline update from the American College of Physicians,
American College of Chest Physicians, American Thoracic Society, and European Respiratory
Society. Ann Intern Med 155(3):179191
22. Connors AF Jr etal (1996) Outcomes following acute exacerbation of severe chronic
obstructive lung disease. The SUPPORT investigators (Study to Understand Prognoses and
Preferences for Outcomes and Risks of Treatments). Am J Respir Crit Care Med 154(4 Pt 1):
959967
23. Hoogendoorn M etal (2010) Case fatality of COPD exacerbations: a meta-analysis and statistical modelling approach. Eur Respir J 37(3):508515
24. Chandra D etal (2012) Outcomes of noninvasive ventilation for acute exacerbations of chronic
obstructive pulmonary disease in the United States, 1998-2008. Am J Respir Crit Care Med
185(2):152159
25. Hoo GW, Hakimian N, Santiago SM (2000) Hypercapnic respiratory failure in COPD patients:
response to therapy. Chest 117(1):169177
26. Menzies R, Gibbons W, Goldberg P (1989) Determinants of weaning and survival among
patients with COPD who require mechanical ventilation for acute respiratory failure. Chest
95(2):398405
27. Del Sorbo L etal (2012) Bridging to lung transplantation by extracorporeal support. Minerva
Anestesiol 78(2):243250
28. Del Sorbo L, Ranieri VM, Keshavjee S (2012) Extracorporeal membrane oxygenation as
bridge to lung transplantation: what remains in order to make it standard of care? Am
J Respir Crit Care Med 185(7):699701
29. Bermudez CA etal (2011) Extracorporeal membrane oxygenation as a bridge to lung

transplant: midterm outcomes. Ann Thorac Surg 92(4):12261231; discussion 12311232
30. Jackson A etal (2008) Use of extracorporeal membrane oxygenation as a bridge to primary
lung transplant: 3 consecutive, successful cases and a review of the literature. J Heart Lung
Transplant 27(3):348352
31. Gille JP, Bagniewski AM (1976) Ten years of use of extracorporeal membrane oxygenation
(ECMO) in the treatment of acute respiratory insufficiency (ARI). Trans Am Soc Artif Intern
Organs 22:102109
32. Zapol Wm SMT (1979) Extracorporeal membrane oxygenation in severe acute respiratory
failure: a randomized prospective study. JAMA 242(20):21932196
33. Davies A etal (2009) Extracorporeal membrane oxygenation for 2009 influenza a(H1N1)
acute respiratory distress syndrome. JAMA 302(17):18881895
34. Brogan TV etal (2009) Extracorporeal membrane oxygenation in adults with severe respiratory failure: a multi-center database. Intensive Care Med 35(12):21052114
35. Aubron C etal (2013) Infections acquired by adults who receive extracorporeal membrane
oxygenation: risk factors and outcome. Infect Control Hosp Epidemiol 34(1):2430
36. Hirose H etal (2012) Right ventricular rupture and tamponade caused by malposition of the Avalon
cannula for venovenous extracorporeal membrane oxygenation. J Cardiothorac Surg 7:36
37. Moerer O, Quintel M (2011) Protective and ultra-protective ventilation: using pumpless interventional lung assist (iLA). Minerva Anestesiol 77(5):537544
[email protected]
Weaning from VV ECMO
27
Giacomo Grasselli, Paolo Mangili, Simone Sosio,

and Nicol Patroniti
27.1
Introduction
Discontinuing extracorporeal respiratory support is a crucial step in patient management, but deciding when and how to wean patients represents a significant challenge
for ICU physicians. Due to the lack of definite criteria, the decision is usually based
on the personal experience and clinical judgment of attending physicians; however,
some indications come from guidelines of scientific societies, local hospital protocols, or published case series.
In the following paragraphs, we will discuss the following aspects of the weaning process: (a) When is a patient ready for weaning? (b) How is weaning performed? (c) How is ventilation managed during the weaning process?
27.2
Assessment of Patient Readiness for Weaning
As described above, the extracorporeal support is progressively reduced as the native

lung function improves. To establish if a patient is ready for weaning off ECMO, we
recommend a thorough assessment of his respiratory function, based on the variations of
the following functional parameters during the course of the disease:
G. Grasselli (*) P. Mangili S. Sosio

Department of Emergency Medicine,
N. Patroniti, MD
Department of Health Sciences, Department of Urgency and Emergency,
Milano-Bicocca University, San Gerardo Hospital,
[email protected]
317
318
G. Grasselli et al.
(a) Increase of the fraction of oxygen delivery provided by the native lung compared to that provided by the artificial lung. According to the ELSO Guidelines
[1], ECMO discontinuation can be considered when the native lung is supporting 5080 % of total gas exchange. In a large cohort of ARDS patients treated
with ECMO, Mols et al. reported weaning off ECMO when at least 80 % of
total oxygen delivery was supplied by the patients own lung [2].
(b) Improvement of respiratory mechanics, e.g., increase of static compliance of
the respiratory system (in ARDS patients) and/or reduction of airway resistance
(in patients with severe asthma).
(c) Improvement of gas exchange: most authors indicate that the patient can be
considered for weaning when arterial pO2 and pCO2 are adequate at moderate
ventilator settings (i.e., FiO2 0.60.5 and relatively low PEEP), but do not
provide clear-cut threshold levels of these parameters.
27.3
Discontinuation Procedure
The extracorporeal support is reduced in parallel with the improvement of native

lung function.
Severely hypoxemic patients require high extracorporeal blood flow: once the
native lung starts to heal and to contribute significantly to arterial oxygenation, the
extracorporeal blood flow can be progressively reduced (Fig. 27.1).
On the other hand, in purely hypercapnic patients, the extracorporeal blood flow
is low from the beginning and the magnitude of the assist will depend on the sweep
gas flow.
In the first phases, the FiO2 of the sweep gas is usually kept at 100 %, especially in
hypoxemic patients. As gas exchanges improve, it is suggested to reduce the FiO2 of the
ventilator before that of the sweep gas, to avoid oxygen-related toxicity on the native lung.
During the weaning phase, different strategies of mechanical ventilation can be
adopted.
In some patients, for example those with serious bleeding complications, the
priority is to discontinue ECMO as soon as possible; in these cases, patients will be
disconnected from ECMO while still on controlled mechanical ventilation.
More commonly, ECMO is used to facilitate the switch from controlled mechanical ventilation to an assisted spontaneous breathing mode, such as Pressure-Support
Ventilation (PSV) or Neurally Adjusted Ventilatory Assist (NAVA) [3, 4]. In this
case, there is a complex interplay between sedation, respiratory drive, and ventilation of the artificial lung; for example, increasing the sweep gas is a very efficient
mode to control the respiratory drive of the patient and may allow a reduction of
sedative drug dosage. In other words, modulating the extracorporeal assist becomes
a strategy to facilitate the patients weaning from the ventilator.
Finally, there are situations where it is desirable to reduce as much as possible the
duration of invasive mechanical ventilation, such as in presence of severe immunocompromise: in these cases, patients can be extubated while still on ECMO, and the extracorporeal support is discontinued only after separation from the ventilator.
[email protected]
319
27 Weaning from VV ECMO
Tidal volume, ml
Assisted ventilation
50
50
40
40
30
30
20
20
10
10
500
50
400
40
300
30
200
20
100
10
300
ml/min
200
Respiratory
system
compliance
Tidal
volume
Respiratory
rate
VO2 membrane
lung
150
100
VO2 natural
lung
50
Blood flow
After-ECMO
ECMO day 8
ECMO day 7
ECMO day 6
ECMO day 5
ECMO day 4
ECMO day 3
ECMO day 2
ECMO day 1
Gas flow
Pre-ECMO
l/min
Shunt
fraction
Total VO 2
250
0
6
5
4
3
2
1
0
Shunt fraction % Respiratory rate breaths/min
Compliance ml/cmH2O
Control ventilation
Fig. 27.1 Example of daily ventilatory management in an ARDS patient. Main parameters of
ventilator and ECMO setting, gas exchange, and hemodynamics are shown day by day. As soon as
the native lung improves, the patient is switched to assisted mechanical ventilation (dotted line),
and mechanical and ECMO GF support is progressively reduced
[email protected]
320
27.4
G. Grasselli et al.
Trial of ECMO Discontinuation
Once the patient is judged as ready for weaning according to the criteria listed
above, it is recommended to perform a trial of temporary discontinuation of the
extracorporeal support. By definition, venovenous ECMO does not provide
hemodynamic support: for this reason, unlike with venoarterial ECMO, there is
no need for stopping or reducing extracorporeal blood flow at the time of the
trial off.
The trial of venovenous ECMO discontinuation should be performed as
follows:
If the patient is on controlled mechanical ventilation, the ventilator settings
(respiratory rate, plateau pressure, FiO2, and PEEP) should be adjusted to values
that are considered acceptable off ECMO. If the patient is on assisted spontaneous ventilation (e.g., PSV, ACV, NAVA), an adjustment of the level of inspiratory
assist and a careful modulation of the level of sedation may be required.
Once the ventilator settings have been adjusted as described before, the sweep
gas to the oxygenator is turned off. It should be remembered that it is not enough
to turn the flowmeter to zero, but it is necessary to clamp the gas tubes, since
oxygen can leak around the flowmeter even when it appears to be off. Once the
sweep gas flow is stopped, the oxygen will be fully consumed after about 20 min:
monitoring of venous oxygen saturation on the extracorporeal circuit will indicate when the excess oxygen in the circuit has been used up.
The extracorporeal blood flow is continued, and no adjustment of heparin dose is
required.
There are no clear indications on the duration of the trial: some centers suggest a
trial off for 16 h, but if needed the duration of the trial can be prolonged up to several hours. During this period, the patient should be closely monitored, paying particular attention to the following aspects:
Hemodynamic stability: besides standard hemodynamic parameters (heart rate,
arterial blood pressure, cardiac filling pressures), continuous monitoring of
mixed venous oxygen saturation (if available) is recommended to evaluate the
adequacy of oxygen delivery during ECMO discontinuation.
Adequacy of gas exchanges (serial monitoring of arterial blood gas analysis).
If the patient is on an assisted spontaneous mode of ventilation, respiratory pattern (tidal volume, respiratory rate, minute ventilation) and mechanics (signs of
distress, use of accessory muscles) should be carefully assessed.
If the patient remains stable during the trial and, most importantly, his ventilatory
load is acceptable, the extracorporeal support can be definitively discontinued and
the cannulas removed as described below (Fig. 27.2).
In particularly unstable patients, some centers tend to disconnect the circuit leaving the cannulas in place to allow a prompt reinstitution of the extracorporeal support in case of sudden deterioration of the patients conditions (Fig. 27.3). Venous
cannulas can be left in place for up to 48 h: to avoid clotting, they should be flushed
with a drip of heparinized solution and systemic anticoagulation continued at
unchanged dosage.
[email protected]
Heart rate, beat/min
GF
321
120
110
100
90
80
70
50
40
150
30
100
20
50
SBP
PAPm
10
0
100
Hb saturation, %
PAPm mmHg
SBP, mmHg
60
200
95
90
SaO2
85
SvO2
80
minute venilation, l/min
75
70
12
10
8
6
4
2
0
Time, h
Fig. 27.2 Example of a successful trial of ECMO discontinuation. ECMO gas flow (GF) is
decreased from 5 to 3 to 0 l/min. In spite of the fall in mixed venous oxygen saturation (SvO2),
arterial oxygen saturation (SaO2) remains stable with a reasonable increase in minute ventilation
and hemodynamic drive. The patient was successfully decannulated
27.5
Decannulation
Cannulas placed percutaneously can be removed directly. Some centers propose to

turn off heparin for 3060 min before decannulation. Before cannula removal,
a purse-string suture is inserted around the cannulation site. Immediately after
decannulation, the suture is tightened and local pressure is applied for at least
[email protected]
322
G. Grasselli et al.
Heart rate, beat/min
GF
120
100
80
60
60
200
50
40
150
30
100
20
50
SBP
PAPm
10
0
SaO2
PAPm mmHg
SBP, mmHg
40
250
100
98
96
94
92
minute ventilation, l/min
90
88
86
84
12
10
8
6
4
2
0
0
60
30
90
Time, min
Fig. 27.3 Example of a failed trial of ECMO discontinuation. ECMO gas flow (GF) is decreased
from 3 to 0 l/min. As soon as GF is turned off, minute ventilation doubles and arterial oxygen saturation decreases while the heart rate and systemic blood pressure increase. The increase in minute
ventilation and the decrease in SaO2 were considered unacceptable. GF was returned to 8 l/min to
allow patient rest
30 min. It is advised to check regularly the cannulation site for signs of bleeding or
hematoma formation.
When removing venous cannulas (especially jugular catheters) in spontaneously
breathing patients, there is a potential risk of air aspiration through the catheters
side holes: to avoid this, a Valsalva maneuver on the ventilator should be performed
at the time of cannula removal.
[email protected]
323
After decannulation, we perform a venous Doppler of the lower limbs and of the
cannulated vessels to exclude thrombotic events.
27.6
Discontinuation for Futility
The extracorporeal support should be discontinued for futility if the patients conditions evolve toward a permanent and irreversible damage of brain, lung, and/or heart
function, and there is no hope of recovery or organ replacement. According to the
ELSO Guidelines [1], this possibility should be explained to the family before
ECMO institution.
References
1. ELSO guidelines. http://www.elso.med.umich.edu/guidelines.html
2. Mols G, Loop T, Geiger K, Farthmann E, Benzing A (2000) Extracorporeal membrane oxygenation: a ten-year experience. Am J Surg 180:144154
3. Karagiannidis C, Lubnow M, Philipp A et al (2010) Autoregulation of ventilation with neurally
adjusted ventilatory assist on extracorporeal lung support. Intensive Care Med 36:20382044
4. Mauri T, Bellani G, Grasselli G et al (2013) Patient-ventilator interaction in ARDS patients
with extremely low compliance undergoing ECMO: a novel approach based on diaphragm
electrical activity. Intensive Care Med 39:282291
[email protected]
Part IV
ECMO for Organ Procurement
[email protected]
Heart-Beating and Non-Heart-Beating

Donors
28
Marinella Zanierato, Francesco Mojoli,

and Antonio Braschi
28.1
Introduction
Transplantation is currently considered to be an effective therapy for treating endstage organ diseases. However, the widespread application of organ transplantation
is limited by the shortage of viable donor organs. This shortage has been addressed
by various measures: improvements in brain death donor (BDD) management by
means of aggressive support, extension of the acceptance criteria for marginal
donors, and the implementation of protocols that accept donation after cardiac death
(DCD). This latter measure has created a donor group that is frequently referred to
as non-heart-beating donors (NHBDs) [1]. In countries where legal or societal barriers discourage the use of brain death donors, DCD donors are the only alternative
source of organs for transplantation when no living donors are available [2]. NHBDs
are classified, in accordance with the Maastricht criteria [3], into the following four
categories: donors who are declared dead outside the hospital and are brought into
the hospital without any attempt at resuscitation (Type 1), donors in whom cardiac
arrest occurs unexpectedly and for whom resuscitation attempts are unsuccessful
(Type 2), donors for whom cardiac arrest is expected after withdrawal of treatment
(Type 3), and donors in whom cardiac arrest occurs during or after brain death
diagnostic procedures (Type 4). Types 1 and 2 are defined as uncontrolled donors on
M. Zanierato (*)
F. Mojoli A. Braschi
Sezione di Anestesia Rianimazione e Terapia Antalgica,
Universit degli Studi di Pavia, V.le Golgi 19, Pavia 27100, Italy
[email protected]
327
328
M. Zanierato et al.
Table 28.1 Maastricht

NHBD categories
Category Description
I
Dead on arrival
II
Unsuccessful
resuscitation
III
Awaiting cardiac arrest
IV
Cardiac arrest while
brain death
Alternative categorization
Uncontrolled
Uncontrolled
Controlled
Controlled
the bases of unexpected cardiac arrest and unsuccessful resuscitation. In contrast,

Types 3 and 4 are defined as controlled donors. Type 3 donors occur after a decision
that care is futile and the consequent removal of life-sustaining treatment, such as
mechanical ventilation or organ perfusion support (Table 28.1).
The main impediment to NHBD is the increased incidence of complications or
of impaired graft function as a result of extended periods of warm ischemia. A univocal definition of warm ischemia time (WIT) has yet to be agreed. In controlled
DCD, WIT can be defined as the interval between support withdrawal and cold
perfusion onset [4] or as that between the withdrawal of life support and cardiac
arrest; this interval is also known as the agonal phase [5]. It is characterized by
sustained hypotension and may impact substantially on warm ischemic damage to
DCD organs. Research has shown that agonal periods of more than 2 h duration are
not acceptable for transplantation purposes [6]. In uncontrolled DCD, WIT can
impact even more significantly, since the exact period of circulatory arrest is usually
not known [7]. Furthermore, between cardiac arrest and cold perfusion start, there
is a period of cardiopulmonary resuscitation (CPR) of variable duration and efficacy. It has recently been shown that the use of automated chest compression
devices may improve organ perfusion before the organ preservation measures are
initiated [8]. An apparently clear finding is that static hypothermic storage may not
be the most appropriate strategy for DCD graft preservation, because hypoxia and
reperfusion can exacerbate ischemic tissue damage. Research and clinical practice
have accordingly pursued the development of several strategies to prevent warm
ischemic injury and improve organ preservation in DCD donors.
28.2
Preservation Strategies for DCD Donors
There are two current perfusion techniques:

In situ perfusion cooling: Introduced in the early 1970s and subsequently modified, this technique effects cold perfusion (4 C) by means of a double-balloon
triple-lumen (DBTL) catheter that is inserted through the femoral artery into the
aorta; the thoracic and abdominal balloons are then, respectively, inflated immediately above the renal arteries and the aortic carrefour. An infusion system is then
connected to the catheter to enable flushing with a total of 1520 l of cold preservation solution, such that the kidneys cool to 1015 C. Blood drainage is enabled
by a large Foley catheter that is introduced into the femoral vein [9]. Donor
nephrectomy can be performed within 2 h of flush perfusion. This is an easy technique with which to preserve kidneys both in controlled DCD and in specific
[email protected]
28
Heart-Beating and Non-Heart-Beating Donors
329
cases of uncontrolled DCD. However, this procedure is affected by technical

complications, and grafts harvested from these donors show an increased incidence of delayed graft function (DGF) and acute tubular necrosis (ATN) [10].
Extracorporeal membrane oxygenation (ECMO): ECMO is widely applied as
systemic supply for circulatory and respiratory support. In DCD, it can be used
to restore warm and oxygenated blood flow through the organs and thus to provide adequate tissue perfusion between death and organ procurement. Two different perfusion methods have been described:
Total body cooling (TBC) or core cooling at 18 C through cardiopulmonary bypass (CPB). Originally used in BDD for heart and lung procurement
[11], this method was applied in the early 1990s to NHBD and provided
acceptable results for kidney and liver transplantation. However, an increased
incidence of DGF has been observed in kidney transplantation [12, 13].
Selective ECMO, exclusively for abdominal organs perfusion. There are
two variants of the technique:
Hypothermic ECMO: The technical problems encountered using TBC to
obtain profound hypothermia have prompted many centers to develop a
selective hypothermic form of ECMO [11]. In combination with a temperature-controlled heater-cooler, the ECMO system is placed into the femoral venoarterial route. An occlusion balloon catheter is inserted through the
contralateral artery to occlude the thoracic aorta, and femoral arteries are
ligated bilaterally. With this system, progressive cooling enables the maintenance of a steady 4 C temperature in abdominal organs perfused with
oxygenated blood. Unfortunately, hypothermic ECMO has a limited
capacity to improve cellular function because metabolic activity is almost
completely shut down at such reduced temperature [14]. For this reason all
the cellular repair processes are impaired by this organ preservation
strategy.
Normothermic ECMO (NECMO): This more recent technique consists in
normothermic perfusion of the abdominal organs at 37 C with oxygenated
blood [15]. The current evidence suggests the NECMO reliably serves to
reverse ischemic lesions, improve tissue microcirculation, and ultimately
improve the viability of suboptimal graft [16].
28.3
The NECMO Technique
The femoral artery and vein are, respectively, cannulated (surgically or percutaneously) with 1519 Fr and 2124 Fr perfusion catheters. The cannulae are connected
to the tubing of an ECMO circuit. The ECMO circuit consists in a reservoir, a centrifugal pump, and a blood oxygenator connected to a heat exchanger and a gas (O2/
air) source (Fig. 28.1). The circuit is primed with saline solution. The contralateral
femoral artery is cannulated with a Fogarty balloon catheter, which is advanced into
the supraceliac aorta. The balloon is inflated to prevent cardiac and brain perfusion.
Proper positioning of the balloon is confirmed by chest radiography. Pump flow is
maintained at between 1.7 and 3 l/min, temperature at 35.537.5 C, and pH at
[email protected]
330
M. Zanierato et al.
Oxygenator
Heater
Pump
Arterial
Cannula
Occlusion
Balloon
Venous
Cannula
Fig. 28.1 Normothermic extracorporeal membrane oxygenation (NECMO)
7.07.4 [17, 18]. Post-oxygenator arterial blood gas is sampled at baseline and
throughout NECMO to determine oxygenation parameters and acid-base status.
The circuit sweep gas levels (FiO2 and flow rate) are adjusted to keep PaCO2 at
between 30 and 45 mmHg and SaO2 at about 98100 %. Sodium bicarbonate should
be added to the circuit to correct metabolic acidosis. Anticoagulation is started with
full heparinization (3 mg/kg) prior to cannulation and is subsequently maintained
by heparin bolus of 1.5 mg/kg, in accordance with ACT values.
NECMO is continued until cold perfusion is performed at organ retrieval. The
abdomen is thoroughly explored, and retrograde in situ flush is performed with ice
cold preservation solution via the arterial cannula, to perfuse the liver and the kidneys. It is only necessary to perform one additional venous cannulation in order to
perfuse the portal vein. At this point, NECMO is discontinued, the arterial line is
clamped distally to the oxygenator and proximally to the preservation solution perfusion line, and venous return is collected by means of field suction [17, 19].
28.4
The Physiology of NECMO
Experimental normothermic ECMO studies started around 1997 in Spain and in

Japan [11, 19]. They described this technique as playing a protective role in organ
function preservation by maintaining tissue perfusion under controlled conditions.
The protection consists in the ability of normothermic organs to recover in situ from
warm ischemic damage and in the provision of oxygen and nutrients at
[email protected]
28
331
physiological temperature to restore metabolic processes, which in turn enable the

repair of damaged cells, the correction of acidosis prior to cold ischemia, the restoration of depleted ATP, the regulation of calcium homeostasis, and the removal of
free radicals. The avoidance of cold ischemia by continuous normothermic perfusion is the other obvious advantage of NECMO [20]. It enables improvement in the
quality of abdominal graft (kidneys and livers) and the reversal of ischemic damage.
Blood has been shown to be better than crystalloids for the maintenance of membrane integrity and hence for recovery from tissue damage. On the other hand, since
lower temperatures rapidly reduce metabolism, normothermia is the most favorable
condition for the restoration of metabolic processes, for the repair of damaged cells,
and for the reinvigoration of metabolite energy levels. According to Net et al. [16],
NECMO could shift warm ischemia time toward an ischemic preconditioning
period and thus provide DCD donors with greater viability. This technique was first
applied to clinical practice in the early 2000s in the USA, Spain, and Japan.
28.5
Normothermic ECMO in Uncontrolled NHBD
Normothermic ECMO is the preferred approach for uncontrolled NHBDs in

Spain, in France, and, more recently, in Italy too [2123]. Patients with witnessed out-of-hospital refractory cardiac arrest (CA) are considered eligible.
The emergency medical service is mobilized to the scene of witnessed CA,
where it starts advanced life support in accordance with the international standard guidelines and using an automated chest compression system [24]. If the
asystolic period persists for at least 2030 min and no reversible cause is identified, the CA may be considered to be irreversible and further attempts at resuscitation futile [25]. If an out-of-hospital patient fulfills Type 2 donor criteria, he
is maintained on an automated chest compression system and transferred to
hospital. The declaration of death is made in the hospital, on the basis of the
absence of ECG and spontaneous respiratory activity for at least 5 min.
Subsequently, heparin is administered and external cardiac massage is restarted
with automated chest compression. At the same time, femoral cannulations are
performed, NECMO is started, and mechanical ventilation and the chest compressor are removed. A Fogarty balloon catheter is inserted through the contralateral femoral artery and inflated in the supraceliac aorta. NECMO continues to
be applied if prior consent for donation is in hand. During this period, the donor
is evaluated for possible contraindications to donation. The criteria for donation
include the following: witnessed arrest; no-flow period without CPR limited to
a maximum 1520 min; age below 60 years for kidneys, 5055 for lungs and
liver; cause of death known or presumed; and non-bleeding abdominal injuries
[21, 22]. The time limits for the interval between cardiac arrest and organ perfusion, and between cardiac arrest and organ retrieval, have, respectively, been set
at 120 and 240 min. These limits offer two major benefits: They give the next of
kin more time (as much as 4 h) to accept the death and to consider donation and
they prolong organ perfusion under controlled physiological conditions.
[email protected]
332
M. Zanierato et al.
Accordingly, liver and kidney tests can be performed, and their functions can be
biochemically evaluated [26]. It has been reported that DCD Type 2 kidneys
undergoing NECMO showed a lower rate of DGF and primary graft nonfunction
(PGNF) than did kidneys used after hypothermic preservation [27]. Although
the DGF levels reported for NHBD are consistently high (1620 %) even under
NECMO, this effect appears to be transient and not to impact on the posttransplant outcome. In kidneys from UNHBD undergoing NECMO, the 1-year graft
survival rate has been reported to be as high as 87.4 % [24]. The use of liver
from these donors is somewhat problematic because immediate organ function
is mandatory for the recipients survival. Major concerns in this respect initially
emerged from initial series to use uncontrolled DCD liver grafts; specifically, an
increased incidence of ischemic cholangiopathy and of subsequent primary
graft nonfunction was observed [25]. Despite the encouraging results they
obtained in liver transplantation from uncontrolled DCD, Fondevila et al.
reported respective 1-year recipient and graft survival rates of 82 and 70 %,
even though only 10 % of grafts from Type 2 DCD donors had been used [28].
More recently, in Madrid, a lung transplantation program that uses uncontrolled
DCD donors was launched [29]; the said program developed a new multiorgan
preservation methodology that is also called bithermia preservation. In this
methodology, abdominal organs are preserved in NECMO, and the thoracic
organs are kept in hypothermia by means of continuous pneumoplegia perfusion
[29, 30].
28.6
Normothermic ECMO in Controlled NHBD
In the USA and Northern Europe, NECMO is applied in those patients for whom a
futility of care decision has been made, usually by intensive care staff and other
treating physicians [31, 32]. When the planned withdrawal of life-sustaining support is discussed with the relatives of patients who have a nonreversible neurological injury, organ donation can be an appropriate consideration. Having used
NECMO for abdominal organ perfusion in controlled DCD donors, Migliocca et al.
reported very low DGF levels (8 %) and proposed that this procedure could expand
the kidney donor pool by one third.
Before withdrawal of treatment and after the familys consent, the cannulae for
NECMO are placed at the bedside in ICU. Life support is withdrawn and comfort
measures are continued. If death occurs, it is declared after a no touch period
(generally around 5 min) and NECMO is started. A balloon catheter is placed in the
thoracic aorta, to prevent the oxygenated blood returning to the heart, thus avoiding
functional recovery. In some centers, if cardiac death does not occur within 60 min,
the patient is no longer considered a candidate for DCD donation [17, 33]. Survival
rates for kidney transplant from ECMO-controlled DCD are comparable with those
for kidneys from brain-dead donors [34]. Despite favorable outcomes reported by
certain centers, graft survival is consistently lower in liver than in kidney
[email protected]
28
333
transplantation. More rigorous selection is mandatory for liver transplantation as a

more precise definition of donor-specific risk factors. Increasing experience in DCD
liver transplantation has enabled the definition of the donor-specific risk factors that
are associated with poor graft survival, namely, donor WIT >2030 min, cold ischemia time >810 h, and donor age >4060 years [35]. However, single centers have
reported favorable DCD outcomes and equivalence in DCD-BDD outcomes for
liver transplantation [32].
28.7
ECMO Assistance for Brain-Dead Donors (BDD)
In some countries, ECMO has been used to support BD multiorgan donors in whom
it is not possible to complete death assessment for cardiac or respiratory failure [13,
36]. ECMO support is an option in hemodynamically unstable donors, who require
three different inotropes and vasopressors or with an inotrope score >30 to maintain
mean arterial pressure >60 mmHg. Other candidates for ECMO support are donors
with a PaO2/FiO2 ratio <200 mmHg (at FiO2 100 % and high level of PEEP) and in
whom hypoxemia precludes the apnea test [37, 38]. In these cases, systemic ECMO
begins during brain-death assessment. Despite this being an uncommon use for
ECMO, single centers have demonstrated that early ECMO support for unstable BDD
is a feasible strategy to increase the donor pool and to preserve donor organs. It
reduces vasoactive drug doses, which in turn may lower the incidence of primary graft
dysfunction, especially after heart transplantation. Under ECMO, apnea test should be
performed with the gas flow to the oxygenator lowered to zero, so that oxygen supply
and ventilation are completely dependent on the ventilator [36]. In selected BD
donors, prompt ECMO could thus play a role in recovering viable organs which otherwise would have been lost.
28.8
The Future
Organ procurement strategies for the future will center on exact evaluation of the viability of DCD-sourced organs for transplantation purposes. To this end, the underlying
procedures will necessarily include the option of an adjunctive period of normothermic
ex vivo recirculation [38]. In theory, the use either of normothermic machine perfusion (NMP) after kidney and liver procurement or of ex vivo lung perfusion (EVLP)
after lung procurement will allow physiological aerobic metabolism to continue, which
in turn will provide affected organs with specific substrates and thus enable the reversal
of warm ischemic injury [39, 40]. Analogously, time will have to be invested in, and in
the long term saved by, exact evaluation of the quality of the organs themselves. Periods
of 36 h of ex vivo perfusion, depending on which organs are involved, seem to be necessary to reverse ischemic damage and to enable pre-transplant assessment of organs.
Furthermore, normothermic in vivo and ex vivo recirculation offer an invaluable platform upon which to introduce potential therapies that target ischemia/reperfusion injury
and acute rejection after transplantation.
[email protected]
334
M. Zanierato et al.
References
1. Howard RJ (2007) The challenging triangle: balancing outcomes, transplant numbers and
costs. Am J Transplant 7:24432445
2. Bernat JL, DAlessandro AM, Port FK et al (2006) Report of a national conference on donation after cardiac death. Am J Transplant 2006:281291
3. Koostra G, Daemen JH, Oomen AP (1995) Categories of nonheart-beating donors. Transplant
Proc 27:28932894
4. Reich DJ, Mulligan DC, Pl A et al (2009) ASTS recommended practice guidelines for controlled donation after cardiac death organ procurement and transplantation. Am J Transplant
9:20042011
5. Levvey BJ, Westall GP, Kotsimbos T et al (2008) Definitions of warm ischemic time when
using controlled donation after cardiac lung doors. Transplantation 86:17021706
6. Sohrabi S, Navarro A, Asher J (2006) Agonal period in potential non-heart beating donors.
Transplant Proc 38:26292630
7. Hoogland ERP, Snoeijs MGJ, Winkens B et al (2011) Kidney transplantation from donors after
cardiac death: uncontrolled versus controlled donation. Am J Transplant 11:14271434
8. Wigginton JG, Miller AH, Benitez FL et al (2005) Mechanical devices for cardiopulmonary
resuscitation. Curr Opin Crit Care 11:219223
9. Wind J, Hoogland ERP, van Heurn LWE (2011) Preservation techniques for donors after cardiac death kidneys. Curr Opin Organ Transplant 16:157161
10. Snoeijs MG, Dekkers AJ, Buurman WA et al (2007) In situ preservation of kidneys from
donors after cardiac death: results and complications. Ann Surg 246:844852
11. Kyoma I, Hoshino T, Nagashima N et al (1989) A new approach to kidney procurement from
non-heart beating donors: core cooling on cardiopulmonary bypass. Transplant Proc 21:
12031205
12. Hoshino T, Maley WR, Stump KC et al (1987) Evaluation of core cooling technique for liver
and kidney procurement. Transplant Proc 19:41234128
13. Ko WJ, Chen YS, Tsai PR et al (2000) Extracorporeal membrane oxygenation support of
donor abdominal organs in non-heart-beating donors. Clin Transplant 14:152156
14. Lee CY, Tsai MK, Ko WJ et al (2005) Expanding the donor pool: use of renal transplants from
nonheart-beating donors supported with extracorporeal membrane oxygenation. Clin
15. Rojas-Pena A, Reoma JL, Krause E et al (2010) Extracorporeal support: improves donor renal
graft function after cardiac death. Am J Transplant 10:1361374
16. Net M, Valero R, Almenara R et al (2005) The effect of normothermic recirculation is mediated by ischemic preconditioning in NHBD liver transplantation. Am J Transplant 5:
23852392
17. Magliocca JF, Magee JC, Rowe SA et al (2005) Extracorporeal support for organ donation
after cardiac death effectively expands the donor pool. J Trauma 58:10951101
18. Fondevilla C, Hessheimer AJ, Maathius MHJ et al (2011) Superior preservation of DCD livers
with continuous normothermic perfusion. Ann Surg 254(6):10001007
19. Valero R, Cabrer C, Oppenheimer F et al (2000) Normothermic recirculation reduces primary
graft dysfunction of kidneys obtained from nonheart-beating donors. Transplant Int 13:
303310
20. Farney AC, Singh RP, Hines MH et al (2008) Experience in renal and extrarenal transplantation
with donation after cardiac death donors with selective use of extracorporeal support. J A Coll
Surg 206(5):10281037
21. Garca-Valdecasas JC, Fondevila C (2010) In-vivo normothermic recirculation: an update.
Curr Opin Organ Transplant 15:173176
22. Abboud I, Viglietti D, Antoine C et al (2012) Preliminary results of transplantation with kidneys donated after cardiocirculatory determination of death: a French single-center experience. Nephrol Dial Transpl 27:25832587
[email protected]
28
335
23. Geraci PM, Sepe V (2011) Non-heart-beating organ donation in Italy. Minerva Anestesiol
77:613623
24. Sanchez-Fructuoso AI, Marques M, Prats D et al (2006) Victims of cardiac arrest occurring
outside the hospital: a source of transplantable kidneys. Ann Intern Med 145:157164
25. Baskett PJ, Steen PA, Bossaert L et al (2005) European Resuscitation Council guidelines for
resuscitation 2005. Section 8. The ethics of resuscitation and end-of-life decisions.
Resuscitation 67:S171S180
26. Fondevila C, Hessheimer AJ, Ruiz A et al (2007) Liver transplant using donors after unexpected cardiac death: novel preservation protocol and acceptance criteria. Am J Transplant
7:18491855
27. Jimnez-Galanes Marchn S, Meneu-Diaz JC, Elola-Olaso A et al (2009) Liver transplantation
using uncontrolled nonheart-beating donors under normothermic extracorporeal membrane
oxygenation. Liver Transplant 15:11101118
28. Fondevila C, Hessheimer AJ, Flores E et al (2012) Applicability and results of Maastricht type
II donation after cardiac death liver transplantation. Am J Transplant 12:162170
29. Gmez P, Crdoba M, Ussetti U et al (2005) Lung Transplant Group of the Puerta de Hierro
Hospital: lung transplantation from out-of-hospital non-heart-beating lung donors. J Heart
Lung Transplant 24:10982005
30. Meneses JC, Gmez P, Mariscal A et al (2012) Development of a non-heart-beating donor
program and results after the first year. Transplant Proc 44:20472049
31. Abt PL, Fisher CA, Singhal AK (2006) Donation after cardiac death in the US: history and use.
J Am Coll Surg 203:208225
32. Detry O, Seydel B, Delbouille MH et al (2009) Liver transplant donation after cardiac death:
experience at the University of Liege. Transplant Proc 41:582584
33. Sohrabi S, Navarro C, Wilson C et al (2006) Renal graft function after prolonged agonal time
in non-heart-beating donors. Transplant Proc 38:34003401
34. Gravel MT, Arenas JD, Chenault R et al (2004) Kidney transplantation from organ donors following cardiopulmonary death using extracorporeal membrane oxygenation support. Ann
Transplant 9:5758
35. Monbaliu D, Pirenne J, Talbot T (2012) Liver transplantation using Donation after Cardiac
Death donors. J Hepatol 56:474485
36. Yang HY, Lin CY, Tsai YT et al (2012) Experience of heart transplantation from hemodynamically unstable brain-dead donors with extracorporeal support. Clin Transplant 26:792796
37. Hsieh CE, Lin HC, Tsui YC et al (2011) Extracorporeal membrane oxygenation support in
potential organ donors for brain death determination. Transplant Proc 43:24952498
38. Hosgood SA, Nicholson ML (2011) Normothermic kidney perfusion. Curr Opin Organ
39. Brockmann J, Reddy S, Coussios C et al (2009) Normothermic perfusion a new paradigm for
organ preservation. Ann Surg 20:16
40. Oto T (2008) Lung transplantation from donation after cardiac death (non-heart-beating)
donors. Gen Thorac Cardiovasc Surg 56:533538
[email protected]
Lung Reconditioning
29
Franco Valenza, Jacopo Fumagalli, Valentina Salice,

and Luciano Gattinoni
29.1
Introduction
Transplantation is considered a valuable option in the treatment of end-stage lung

disease. However, organs from multiorgan donors available for transplantation are
far less than the number of potential recipients, so that as many as 1520 % of them
die while on a waitlist [18].
Over the years, a number of ways to overcome the discrepancy between the need
and the availability of organs have been explored, including the use of lung allocation scores and the implementation of standardized donor management protocol [7].
In the last decade, several authors have extended lung donor criteria to increase the
pool of organs, unfortunately with controversial results.
Recently, the feasibility and safety of transplanting high-risk donor lungs that
have undergone ex vivo lung perfusion (EVLP) have been successfully documented
[15, 16]. The technique has brought into a new era of lung transplantation. In fact,
EVLP not only has allowed for expansion of the pool of lungs available for transplantation, but it has also deeply challenged the concept of lung suitability itself. In
fact, organs previously not considered for transplantation are now safely used with
outcomes similar to those of standard donor lungs [1, 4, 5, 8, 2022].
The concept of extracorporeal evaluation and treatment of lung function before
transplantation date back to 1970 [9]. However, a renewed interest on
F. Valenza (*) L. Gattinoni

Dipartimento di Anestesia Rianimazione (Intensiva e Subintensiva) e Terapia del dolore,
Fondazione IRCCS Ca Granda-Ospedale Maggiore Policlinico, Milano, Italy
Dipartimento di Fisiopatologica Medico-Chirurgica e dei Trapianti,
Universit degli Studi di Milano, Milano, Italy
J. Fumagalli V. Salice
Dipartimento di Anestesia Rianimazione (Intensiva e Subintensiva) e Terapia del dolore,
Fondazione IRCCS Ca Granda-Ospedale Maggiore Policlinico, Milano, Italy
[email protected]
337
338
F. Valenza et al.
Table 29.1 Published clinical experiences on EVLP
Ingemansson
Pego-Fernandez
Cypel
Madeiros
Sedaria
Cypel
Aigner
Valenza
Zych
Wallinder
Wallinder
Ann Thor Surg

Rev Bras Cir Cardiovasc
New Engl J Med
J Heart Lung Transpl
Ann Thor Surg
J Thorac Cardiovasc Surg
Am J Transplantation
Transp Proc
J Heart Lung Transpl
J Thorac Cardiovasc Surg
Eur J Card-Thor Surg
EVLP donors
Year DBD
2009
2010
2011
2011
2011
2012
2012
2012
2012
2012
2013
DCD EVLP Lung transplant
Yes
No
Yes
No
No
Yes
Yes
Yes
Yes
Yes
Yes
extracorporeal lung perfusion has started with the clinical experience of Ingemansson
and colleagues who showed initially rejected donor lungs were reconditioned to
acceptable function, and in six recipients, double lung transplantation was performed with a 3-month survival of 100 % [8]. The pivotal role of EVLP has been
further fostered by the study of Cypel et al. that showed no inferiority of clinical
transplantation of lungs retrieved from high-risk donors and reconditioned by EVLP
[4]. A number of lung transplantation centers have since then taken the challenge of
implementing clinical EVLP programs (Table 29.1).
29.2
Technique
Figure 29.1 shows the circuit used to perfuse the isolated lungs. It consists of a
blood reservoir (1 in the figure) connected to a gas oxygenator with a built-in heat
exchanger (2), a centrifugal pump (3), a leukocyte arterial filter (4), and a nonheparin-coated polyvinyl tubing. The system is primed with Steen solutionTM
(Vitrolife, Gothenburg, Sweden). This is a specifically designed buffered solution
with an extracellular-type composition and with an optimized albumin-based colloid osmotic pressure. Methylprednisolone, antibiotics, and heparin are also added
to the perfusate. To run the EVLP, the lungs procured from donors and cold stored
on ice are contained in a specifically designed chamber (XVIVO, Vitrolife).
Temperature of the perfusate is gradually increased to a target temperature of 37 C
over approximately 30 min. Once the lung outflow temperature exceeds 32 C,
mechanical ventilation of the lungs is started. The circuit oxygenator is used unconventionally during EVLP; in fact, gas flow through the artificial lung is composed
of CO2 and air and is intended to add CO2 and remove O2 so that the perfusate composition is similar to that of the pulmonary artery. The lungs are ventilated and
perfused up to 4 h in most protocols, at the end of which, a final evaluation of lung
function is performed. This takes into account parameters of lung perfusion (perfusate flow, temperature, and pulmonary artery pressure, pulmonary vascular resistance) and ventilation (tidal volume, airway pressure, dynamic compliance,
[email protected]
29
339
Lung Reconditioning
Fig. 29.1 Figure shows the

circuit used to perfuse the
isolated lungs. It consists of
a blood reservoir (1)
connected to a gas
oxygenator with a built-in
heat exchanger (2),
a centrifugal pump (3), a
leukocyte arterial filter (4),
and a non-heparin-coated
polyvinyl tubing
4
1
3
Table 29.2 Comparison

between the Lund and
Toronto EVLP protocols
Duration, hours
Perfusate flow, % donor CO
Pulmonary artery pressure, mmHg
Left atrium
FiO2, %
Tidal volume, mL/kg donors weight
Respiratory rate, bpm
Perfusate composition
Lund
1.5
100
<20
Open
50
57
1520
Cellular
Toronto
4
40
1015
Closed
21
7
7
Acellular
respiratory rate, PEEP and FiO2), together with analysis of partial pressures of oxygen (PO2) and carbon dioxide (PCO2). Chest X-ray and fibrobronchoscopy are also
added to the final evaluation of lung suitability. If deemed suitable for transplantation, the lungs are flushed with preservation solution and cold stored on ice, ready
to be used for transplantation.
While EVLP protocols all account for reperfusion, reconditioning, evaluation, and
cooling periods, two main philosophies have been diversified over time, summarized in
the Toronto and the Lund protocols. The main differences are shown in Table 29.2.
[email protected]
F. Valenza et al.
340
29.3
Clinical Application
Ex vivo lung perfusion technique is used to evaluate and/or recondition the function
of lungs procured from marginal donors.
The prototypical use of EVLP for lung evaluation is in donation after cardiocirculatory determination of death (DCDD). When blood flow through the lung is
absent, PaO2/FiO2 ratio, which is the main determinant of lung suitability for transplantation, cannot be assessed. EVLP allows to restore blood flow and render evaluation of the lungs possible. The first application of clinical EVLP was in fact to
evaluate lungs in a case of donation after cardiocirculatory determination of death
[16]. The donor was a patient dying of acute myocardial infarction in a cardiac
intensive care unit after failed cardiopulmonary resuscitation. The concept of EVLP
as an evaluation tool in DCD is of culprit importance, given the lack of organs and
the need to explore new pools of lung donors. However, there might also be cases of
donation after brain death determination (DBD) when lung function is doubtful. In
these cases, EVLP may extend assessment comfort zone by improving the ability to
perform a physiologic and objective evaluation of lung function. In this sense,
EVLP has dramatically changed the scenario of organ suitability and procurement,
as shown in the EVLP diagram flow that we adopted at our institution (Fig. 29.2).
EVLP also allows to recondition the function of previously rather normal donor
lungs that worsened over the donation process to a point that rendered them
Potential lung donor
Lung evaluation
Classical criteria
Yes
No
Not retrieved
No
Suitable for EVLP

evaluation/reconditioning
Suitable
for transplantation
Yes
EVLP
Not transplanted
No
Lung evaluation
EVLP criteria
Yes
Lung transplant
Fig. 29.2 Clinical EVLP diagram flow. The lungs offered for transplantation are first evaluated by
the classic criteria (oxygenation, history of smoking, age, presence of secretions, chest X-ray): If
they are deemed suitable, transplantation is performed. If the lungs do not satisfy the classic criteria, they may undergo EVLP after which a second organ evaluation is performed using the EVLP
criteria (pulmonary arterial pressure, pulmonary vascular resistances, oxygenation, secretions,
chest X-ray, surgeons judgment). If, after EVLP treatment, the lungs are suitable, transplant is
performed
[email protected]
29
Lung Reconditioning
341
unacceptable for transplantation. Deterioration over time of donors lung function

is much relevant; international data report that only 1525 % of lungs from multiorgan donors are used for transplantation [6, 12]. In a recent report from the Nord
Italia Transplant program (NITp), we observed a rate of lung donation of 29 % in
a cohort of 201 potential lung donors [13]. In fact, low PaO2/FiO2, alteration on the
chest X-ray, neurogenic inflammatory lung edema, inhalation injury, and complication related to the treatment in the intensive care unit (e.g., barotrauma and pulmonary edema) often occur in brain death multiorgan donors [2, 14, 19] and
contribute to lung rejection. Ingemansson et al. were the first to show that lungs
from DBD donors, previously rejected by other centers, could be safely transplanted after EVLP reconditioning [8]. This triggered a number of clinical applications of the technique (Table 29.1), fostered by the study of Cypel et al. that showed
no inferiority of clinical transplantation of lungs retrieved from high-risk donors
and reconditioned by EVLP [4].
Lung reconditioning has in fact a number of advantages: EVLP allows to dehydrate edematous lungs by the hyperoncotic solution used to perfuse the lung during
the procedure. It allows to finely tune ischemiareperfusion injury by gentle reperfusion and ventilation of the ischemic lungs. A thorough clearance of thrombi from
the pulmonary vasculature is also an indirect benefit of EVLP. Optimization of ventilation during the procedure and at the time of cooling is of great importance for
organ preservation. Finally, there might also be a preconditioning effect induced by
the double cooling period that the EVLP lungs undergo.
Overall, provided evaluation and reconditioning often coexist, EVLP allows
safer transplantation of the lungs from marginal donors.
29.4
Future Directions
We deem that the true potential of EVLP is to repair previously injured lungs.
The time frame of EVLP, potentially as long as 12 h [3], allows to repair diseased
organs. In fact, gene, cellular, or pharmacological therapies may be applied during
EVLP. Martins et al. have shown that the administration of adenoviral-mediated
human IL-10 to the donor lung reduced ischemiareperfusion injury and improved
graft function after lung transplantation in a pig lung transplantation model; transfection prevented the release of inflammatory cytokines such as IL-6 in the lung
tissue and plasma [11]. Lee et al. showed in an ex vivo perfused human lung injured
by E. coli endotoxin that treatment with allogeneic human mesenchymal stem cells
or the conditioned medium restored normal fluid balance [10]. We have documented
that salbutamol reduces pulmonary artery pressure and improves respiratory
mechanics during EVLP [17].
Many investigations on EVLP are ongoing. These will contribute to improve
technical aspects of the procedure and will perhaps open new possibilities for EVLP.
Overall, these efforts underline the breakthrough of organ reconditioning before
transplantation.
[email protected]
342
F. Valenza et al.
References
1. Aigner C, Slama A, Hotzenecker K et al (2012) Clinical ex vivo lung perfusionpushing the
limits. Am J Transplant 12:18391847
2. Avlonitis VS, Fisher AJ, Kirby JA et al (2003) Pulmonary transplantation: the role of brain
death in donor lung injury. Transplantation 75:19281933
3. Cypel M, Rubacha M, Yeung J et al (2009) Normothermic ex vivo perfusion prevents lung
injury compared to extended cold preservation for transplantation. Am J Transplant
9:22622269
4. Cypel M, Yeung JC, Liu M et al (2011) Normothermic ex vivo lung perfusion in clinical lung
transplantation. N Engl J Med 364:14311440
5. Cypel M, Yeung JC, Machuca T et al (2012) Experience with the first 50 ex vivo lung perfusions in clinical transplantation. J Thorac Cardiovasc Surg 144:12001206
6. de Perrot M, Snell GI, Babcock WD et al (2004) Strategies to optimize the use of currently
available lung donors. J Heart Lung Transplant 23:11271134
7. Egan TM, Murray S, Bustami RT et al (2006) Development of the new lung allocation system
in the United States. Am J Transplant 6:12121227
8. Ingemansson R, Eyjolfsson A, Mared L et al (2009) Clinical transplantation of initially
rejected donor lungs after reconditioning ex vivo. Ann Thorac Surg 87:255260
9. Jirsch DW, Fisk RL, Couves CM (1970) Ex vivo evaluation of stored lungs. Ann Thorac Surg
10:163168
10. Lee JW, Fang X, Gupta N et al (2009) Allogeneic human mesenchymal stem cells for treatment of E. coli endotoxin-induced acute lung injury in the ex vivo perfused human lung. Proc
Natl Acad Sci U S A 106:1635716362
11. Martins S, de Perrot M, Imai Y et al (2004) Transbronchial administration of adenoviralmediated interleukin-10 gene to the donor improves function in a pig lung transplant model.
Gene Ther 11:17861796
12. Oto T, Levvey BJ, Whitford H et al (2007) Feasibility and utility of a lung donor score: correlation with early post-transplant outcomes. Ann Thorac Surg 83:257263
13. Porro GA, Valenza F, Coppola S et al (2012) Use of the Oto lung donor score to analyze the
2010 donor pool of the Nord Italia Transplant program. Transplant Proc 44:18301834
14. Snell GI, Paraskeva M, Westall GP (2013) Donor selection and management. Semin Respir
15. Steen S, Ingemansson R, Eriksson L et al (2007) First human transplantation of a nonacceptable donor lung after reconditioning ex vivo. Ann Thorac Surg 83:21912194
16. Steen S, Sjoberg T, Pierre L et al (2001) Transplantation of lungs from a non-heart-beating
donor. Lancet 357:825829
17. Valenza F, Rosso L, Coppola S et al (2012) beta-adrenergic agonist infusion during extracorporeal lung perfusion: effects on glucose concentration in the perfusion fluid and on lung function. J Heart Lung Transplant 31:524530
18. Van Raemdonck D, Neyrinck A, Verleden GM et al (2009) Lung donor selection and management. Proc Am Thorac Soc 6:2838
19. Venkateswaran RV, Dronavalli V, Patchell V et al (2013) Measurement of extravascular lung
water following human brain death: implications for lung donor assessment and transplantation. Eur J Cardiothorac Surg 43:12271232
20. Wallinder A, Ricksten SE, Hansson C et al (2012) Transplantation of initially rejected donor
lungs after ex vivo lung perfusion. J Thorac Cardiovasc Surg 144:12221228
21. Wallinder A, Ricksten SE, Silverborn M et al (2014) Early results in transplantation of initially
rejected donor lungs after ex vivo lung perfusion: a case-control study. Eur J Cardiothorac Surg
45:4044
22. Zych B, Popov AF, Stavri G et al (2012) Early outcomes of bilateral sequential single lung
transplantation after ex-vivo lung evaluation and reconditioning. J Heart Lung Transplant
31:274281
[email protected]
Part V
Monitoring the ECMO Patient
[email protected]
Patient Care During ECMO
30
Michela Bombino, Sara Redaelli,

and Nicol Patroniti
30.1
Introduction
The daily care of an ECMO patient is a complex multidisciplinary task. As other

critically ill patients, the ECMO-supported one must be re-evaluated for the primary
disease that leads to ICU admission and for the possible development and treatment
of other organ failures. The complex relationship between the patient and the ECMO
circuit makes sometimes difficult to understand how the patient is doing. Other
tasks refer to the monitoring of the ECMO circuit and the prevention of complications. Some other chapters in this book will address specifically some of the pitfalls
in the management of these patients (complications, monitoring, ventilation); we
will here focus on global daily care and its relationship with the ECMO equipment.
The day of our ECMO patient will start with nursing care, followed by a careful
head-to-toe assessment of organ functions to reveal actual problems and draw the
daily care plan.
M. Bombino (*) S. Redaelli

General Intensive Care Unit, Department of Emergency and Urgency,
N. Patroniti, MD
University of Milano-Bicocca, San Gerardo Hospital,
[email protected]
345
346
30.2
M. Bombino et al.
Daily Nursing of the ECMO Patient
Daily nursing in the intensive care unit (ICU) is a fundamental therapeutic intervention, generally performed to improve patient hygiene, ameliorate patient comfort,
prevent iatrogenic infections and assess skin integrity [1]. Patients in ECMO are
generally unstable, and nursing care may be more challenging and hazardous due to
the strict dependence of patients oxygenation on the maintenance of ECMO blood
flow, anticoagulation and possible decannulation. It consists in complete bed bath,
oral hygiene and sheets and dressing replacement as in other ICU patients [2], with
some precautions related to the risk of bleeding in anticoagulated patients and to the
specific management of ECMO circuit.
During mouth care, trauma to the oral mucosa should be avoided with the use of
swabs without or controlled low aspiration [3]; the position of the endotracheal tube
should be changed two to three times in a day to preserve the oral mucosa and labial
commissures; when fixing the ET tube, the commisures must be protected from the
beginning interposing anti-decubitus dressing; the same applies to protect the nostrils in the case of nasotracheal or nasogastric tubes. If hairs need to be removed, an
electrical razor is recommended.
Cannulation sites must be examined, as all the other catheters in place, to rule out
signs of infection or bleeding; therefore, a transparent semipermeable dressing is
preferred if there is no bleeding. Routine cannula dressing changes must follow the
same rules stated by the Infectious Diseases Society of America (IDSA) guidelines
[4]. When renewal of dressing is deemed necessary, it must be pulled off towards the
insertion site to minimise the risk of cannula displacement. The distance between the
insertion site and the end of the wire-wound cannula must be checked and recorded
at the beginning of the nurse shift in order to recognise cannula dislodgement; at the
same time the integrity of the cannula fixing system is checked (Fig. 30.1). The cannula and the tubing must be maintained along the leg axis for at least 40 cm in the
femorofemoral veno-arterial (VA) or venovenous (VV) configuration; the skin in
contact with the tubing must be protected; if an internal jugular cannula is in place, it
should be fixed to the patients head with a bandage. Never leave the weight of the
ECMO tubing exerting traction to the cannula. Furthermore, during VA ECMO, frequent evaluation of peripheral pulses, skin temperature and colour should be performed to avoid lower limb ischaemia; in VV ECMO the efficacy of venous return
from the distal leg and possible haematoma formation near the insertion site should
be assessed by means of daily measurement of thigh circumference.
Mobilisation for back hygiene, sheets replacement and assessment of skin integrity may be performed either by logrolling the patient or by means of a scooping
stretcher attached to a lift (Fig. 30.2), to minimise tube kinking and alteration in
blood flow [3]. In our experience, daily nursing may be associated with several
alterations of vital signs, such as tachycardia, hypertension and arterial desaturation
(Fig. 30.3). These adverse events are mainly related to a neurovegetative response
to stimuli, especially at the beginning of the procedures, during oral care and mobilisation; additional pulse sedation may be required to accomplish such nursing tasks.
[email protected]
30
347
Fig. 30.1 ECMO cannulas care. Panel (a): medication and fixation of the ECMO cannulas and
tubing along the leg axis. Panel (b): measurements to assess cannula dislodgement
In relatively hypovolaemic patients, both elevation of the stretcher and mobilisation

on the side may affect the position of the ECMO cannulas, mainly the drainage one,
into the vessel with a reduction in effective ECMO blood flow and therefore
desaturation.
[email protected]
M. Bombino et al.
348
Fig. 30.2 Linen change with a scooping stretcher attached to a lift
Change position of
endotracheal tube
Nursing start
Mobilization with
scooping stretcher
Systolic blood pressure [mmHg]
190
100
170
95
150
90
130
Arterial saturation [%]
105
210
110
85
90
SpO2
BF = 2.4 L/min
70
50
5
SBP
BF = 0.8 L/min
80
75
0
10
15
20
25
Nursing time [min]
30
35
40
45
Fig. 30.3 Typical course of vital signs (sBP systolic blood pressure) and arterial oxygenation
(SpO2 pulse oximeter saturation) during daily nursing care
[email protected]
30
349
Finally, when the patient is lightly sedated, as in the weaning phase, a brisk
increase in the work of breathing can ensue during nursing care (elevation in respiratory rate and in minute ventilation); a higher level of ventilatory support may
therefore be required during daily nursing.
Hence, we suggest that daily nursing should be performed by two or three nurses
(at least one experienced in ECMO management) with an extra staff member designated to monitor tubing and circuit, ensuring there is no kinking or tension. It should
only be undertaken when medical personnel are present or readily available to manage circuit problems, to administer additional sedation if deemed necessary or to
adjust ventilatory support. It should be scheduled during the day shift, avoiding
nocturnal turning unless essential [3]. Finally, if the patient is greatly unstable, for
example, in ongoing active bleeding, daily nursing must be deferred or not be performed at all, according to clinical judgement.
30.3
Daily Assessment of the ECMO Patient
Physicians and nurses will evaluate completely the patient at least three times in a
day. A head-to-toe approach is applied for simplification.
30.3.1 Head and Sedation

We know that it is possible, with the new ECMO technology, to have a patient
walking around while on ECMO support (see sect. 30.5), but heavy sedation, and
sometimes muscle paralysis, is needed in the most critical one, as a tool to decrease
oxygen consumption (VO2) and therefore improve global oxygenation. In these
cases, adequacy of sedation is difficult to evaluate; vital signs (heart rate and systemic arterial pressure) plus objective measures of brain function, such as bispectral index (BIS), are recommended in the recent published guidelines [5].
Critically ill patients receiving ECMO have been reported to have increased
sedation requirements; the literature is mainly related to neonatal and paediatric
patients; scanty data are available for the adult one [6]. The possible interference of
artificial oxygenators with some sedative and analgesic drugs has also been reported
[79], and two studies to understand the pharmacokinetics of antibiotics, sedatives
and analgesics during ECMO are under way [10, 11].
To have a calm and cooperative patient is a very difficult task. Due to the safety
concerns about possible cannula dislodgments and even decannulations, agitation
and delirium must be promptly treated. Never leave the lightly sedated, apparently
calm patient unattended, especially during the night shifts: A brisk arousal or a
nightmare can result in a disaster.
The pupils reactivity is our neurological window in the sedated and paralysed
patient and must be evaluated several times during the day, since the rate of neurological complications (stroke and bleeding) is still relatively high ranging from 4 to
9 % in the most recent series [12, 13].
[email protected]
350
M. Bombino et al.
30.3.2 Airways and Ventilation

As stated above the oral or nasal endotracheal tubes carry a high risk of mucosal
bleeding; if the ECMO run is prolonged, a tracheostomy must be planned for patient
comfort and as a weaning facilitator [14]. Percutaneous techniques are preferred
over the surgical ones; having an otolaryngologist specialist involved can assure
surgical assistance if a bleeding problem will ensue in the following days. As for
other planned surgical procedure, anticoagulation is withheld 46 h before and after
the procedure, and platelets are transfused as needed; strict monitoring of ECMO
circuit for clot formation is performed; and if necessary, the ECMO circuit will be
changed.
Several questions need to be answered daily during the ECMO run:
Is the ventilatory setting protective?
This question is surely the principal one at the beginning of the ECMO course,
but it remains valid for every day thereafter. Keep in mind that spontaneous
breathing with huge transpulmonary pressure swings carries the hidden danger
of harming the lung, mainly if pursued too early in the setting of increased capillary leak. On the other side, maintaining the patient for an unnecessary prolonged time heavily sedated and even paralysed carries the risk of diaphragmatic
dysfunction and polyneuropathy. What we have considered as lung rest ventilatory settings is of debate; Extracorporeal Life Support Organization (ELSO)
guidelines [15] give targets of PEEP, peak airway pressure and respiratory rate
(PEEP 10 cmH2O, PIP 20 cmH2O, RR 510), that are very difficult to achieve, in
our experience, in the early phases when capillary leak is an issue, due to frothy
oedema coming up in the endotracheal tube and drowning the patients. Also
other ECMO centres would decrease PEEP with caution only when the high
permeability acute phase subsides and in the absence of alveolar bleeding [3, 16,
17]. If respiratory acidosis is an issue, it must be corrected slowly to avoid brisk
changes in acidbase balance and its deleterious effect on cerebral circulation
[15].
Can the patient start a weaning process leading to ECMO disconnection?
We need to evaluate every day the possibility of weaning the patient from ECMO
support. In VA ECMO, a key role in the weaning decision is played by daily echocardiographic evaluation of the patients heart function during trials of ECMO
blood flow decrease [1820] (see Chap. 18). In the respiratory patient in VV
ECMO, we need to evaluate if gas exchange and respiratory mechanics will allow
weaning. Improvement in oxygenation through the natural lung is easy to assess
if the principles of O2 delivery are applied to the patient-ECMO relationship, partitioning the relative contribution of natural and artificial lung to oxygenation. We
like to monitor the ECMO patient with a pulmonary artery catheter, and therefore
we can evaluate shunt fraction and the contribution of the natural lung to VO2 by
simply drawing blood gases from the arterial and pulmonary artery catheter and
the contribution of ECMO by drawing blood gases from the inlet and outlet of the
artificial lung. The real challenge is to evaluate the ventilatory endurance of the
patient and to understand if the ventilatory burden can be accomplished without
[email protected]
30
351
ECMO. This process normally takes some days or even weeks. If there are important bleeding complications that put the patients at risk of death and ECMO disconnection is warranted, we must indeed evaluate if we can handle gas exchange
with protective settings also in controlled mechanical ventilation.
Are there complications?
Huge pleural effusions are not characteristic of ARDS at the beginning but can
develop during the ECMO run due to fluid overload or a decompensating heart
function. In these cases we must evaluate the introduction of forced diuresis or
even CRRT or add inotropes before attempting a thoracentesis, because this
manoeuvre can lead to a haemothorax in the ECMO anticoagulated patient.
The same reasoning applies to pneumothorax. The ELSO guidelines [15]
suggest that a conservative approach must be implemented even for pneumothoraces up to 50 %, without haemodynamic compromise and not enlarging,
due to the high bleeding complications related to chest tubes placement, often
leading to thoracotomies [2123]. Lung ultrasound and chest x-rays are used to
monitor the pneumothorax, while an important reduction of airway pressure is
attempted with an increase in ECMO blood flow to support oxygenation [24]
(Fig. 30.4).
Is gas exchange deteriorating?
Oxygen desaturation can be patient or ECMO equipment related. First, as a temporary manoeuvre, increase the ventilator and ECMO FiO2 to 100 %. If the
patient is in full ECMO support and the native lung is not participating in gas
exchange, carefully check the ECMO equipment: Is blood flow decreased? Is the
oxygen supply to the artificial lung ok? Is the outlet blood bright red and significantly different from the inlet blood?
Fig. 30.4 Bilateral pneumothorax in a patient with severe pulmonary tuberculosis. Chest x-rays
showing left (panel a) and right (panel b) spontaneous pneumothorax. Both were left undrained
and ventilation pressures were decreased while oxygenation was guaranteed by extracorporeal
membrane oxygenation (ECMO). The patient eventually survived. Arrows indicate air presence in
the pleural spaces (Reprinted by permission of Edizioni Minerva Medica from Mauri et al. [24])
[email protected]
352
M. Bombino et al.
If the native lung is consistently participating in gas exchange, we must assess

the patient and the ventilator: Is chest expansion normal, with bilateral respiratory sound? Are there airway secretions with increased peak inspiratory pressures? Is there a ventilator malfunction? Is the patients metabolic demand
increased? We must also evaluate the haemodynamics of the patient, since oxygen desaturation can ensue from an increased cardiac output with a proportional
decreased contribution of ECMO to oxygenation.
30.3.3 Haemodynamic and Volume Status

Assessing volume status can be challenging in ECMO patients and is affected by cannula position. In the VV femorofemoral approach, central haemodynamics is not
really disturbed by the ECMO circuit, and we like to monitor it with a pulmonary
artery catheter since it gives us tools to fully understand the gas exchange of the patient
and gives information about pulmonary hypertension. Central venous pressure and
pulmonary capillary pressure in association with echocardiography can help to evaluate the volume status of the patient. Another parameter to follow is the negative pressure in the drainage cannula; if relative hypovolaemia will ensue, sometimes the first
sign will be chattering of the drainage line due to collapse of the inferior vena cava
around the cannula. Cardiac output by thermodilution has known limitations [25], but
the trends are still important to follow. If the drainage cannula is in the right atrium, we
use the ventricular port as injection site for cardiac output measurements [26].
On VV ECMO, central venous saturation (SvcO2) is contaminated by arterialised blood coming from the artificial lung and therefore cannot be used to assess
the adequacy of cardiac output to metabolic needs. The saturation of blood at the
inlet of the artificial lung can be monitored indeed, low saturation being surely a
sign of hypoperfusion.
Pulmonary hypertension is quite common in patients with ARDS treated in
ECMO, and we need to understand its relationship with cardiac output, oxygenation
and CO2. A fixed pulmonary artery pressure increasing over days is a poor prognostic factor. Echocardiography and continuous monitoring through a pulmonary artery
catheter are our tools to assess if right ventricular failure is in development and the
needs for some specific treatments (inhaled nitric oxide, sildenafil, bosentan).
30.3.4 Infection Evaluation and Workup

If the cause of ARDS was infective, we need to evaluate its resolution and recognise
if new infections are ensuing. We know that the risk of infection is high in the ECMO
patient and is mainly related to the length of the ECMO run, patients age, VA access
and immunosuppression [2731]. Diagnosis of a new infection can be difficult since
some of the clinical clues we normally use in other ICU patients can be affected by
the presence of ECMO [32]. The inflammatory response of the patient to the ECMO
foreign surfaces can influence white blood cell count, and chest x-ray is almost never
diagnostic of a new infiltrate in the setting of ARDS. The ECMO heater controls
[email protected]
30
353
the patients body temperature, and fever response is therefore blunted; the feverish
patient on ECMO must face the cooling effect of the ECMO circuit, and this will result
in high oxygen demands. The suspicion of a new infection should arise from small
signals, like a difference between patients and ECMO temperature of 0.30.5 C,
presence of mottled skin and purulent secretions and variations in haemodynamics,
which need to be integrated with biomarkers of infection like C-reactive protein and
procalcitonin [33]. The most common infections during ECMO runs are secondary
VAP and bacteraemia. We monitor the colonisation of the airways with surveillance
cultures drawn at least once in a week, also if their role is controversial. Colonisation
with MDR microorganism by rectal swabs will be searched if the patient has been in
the ICU for more than 15 days. If a new infection is suspected (purulent secretions,
brisk increase in biomarkers with hyperdynamic haemodynamics, etc.), blood cultures and other specimens will be sent, and the antimicrobial therapy will be changed
empirically and then adjusted on microbiological results. An infectious disease task
force was established by ELSO in 2008 to address issues on diagnosis, treatment
and prevention of infections during ECMO. They concluded that there is no reason
for antibiotic prophylaxis solely for ECMO after the initial bolus at cannulation and
remarked on the importance of infection prevention (hand washing, no circuit break,
limiting manipulation of the ECMO circuit and other catheters, head of bed elevation, oral hygiene and decontamination) [32, 34].
30.3.5 The Abdomen and Nutrition

The abdomen must be evaluated for liver and splenic enlargement and bowel function. Liver enlargement is quite common in these patients as the result of stasis but
can become a real problem with the development of secondary hepatic insufficiency
in cases of severe right ventricular failure; in such cases decompression of the right
ventricle can be necessary.
Abdominal compartment syndrome (ACS) has been described in the adult ECMO
literature [35, 36], mainly in the veno-arterial mode, as a result of fluid overload.
In our experience another possible cause is spontaneous retroperitoneal bleeding
(Fig. 30.5). Measurements of intra-abdominal pressure are therefore recommended.
The development of ACS in both VA and VV femorofemoral approaches will compromise the ECMO blood flow, and a new drainage cannula must be inserted in the
jugular vein. Surgical decompression can be needed if bowel, renal and respiratory
compromise is ensuing.
Early in 1983, Iapichino and co-workers pointed out that effective nutritional
support could be provided safely to patients undergoing ECMO, even if the patients
are heavily anaesthetised and curarised [37]. The same findings have been reported
by other ECMO centres [3840]. The data on nutritional requirement and tolerance
are scanty in the adult ECMO population [41], and the Australian and New Zealand
Intensive Care Research Centre already started a survey to answer some question
about this topic [42].
In our experience the tolerance to enteral feeding is good also in the most severe
ARDS patients since the beginning. We needed to shift to parenteral nutrition only
[email protected]
354
M. Bombino et al.
Fig. 30.5 Abdominal

compartment syndrome due
to a spontaneous lumbar
haematoma in a patient on
VV ECMO
in few cases due to important gastric bleeding or development of abdominal compartment syndrome. Severe diarrhoea can be a problem mainly in the femorofemoral ECMO, due to the possible faecal contamination of access sites. Faecal
management devices must be avoided or kept in place for a very short time due to
the high risk of rectal lesions in these patients. In these cases, changing the diet or
decreasing the velocity of enteral nutrition by adding a parenteral integration can be
the right choice.
30.3.6 Hepatic and Renal Function

Hepatic and renal functions must be carefully evaluated and monitored since these
have been demonstrated to be related to a poor prognosis [13]. Liver function can be
impaired for many reasons (overwhelming sepsis, medications, cardiogenic shock,
right ventricular failure, haemolysis); the treatment of the primary insult is almost
always the cure. In few cases extracorporeal liver support (MARS) [43] or plasma
exchange or adsorption (CPFA) can be of value.
Severe renal impairment must be treated with continuous renal replacement therapies (CRRT) early in its course due to the risk of fluid overload. If the patient on
ECMO requires a CRRT system, some key points must be taken into account [13]:
[email protected]
30
355
1. The CRRT circuit can be kept independent from the ECMO one by inserting a
new dialysis catheter. If this is the choice, the risk associated with a new catheter
placement in a fully anticoagulated patient must be faced.
2. If the CRRT system is connected to the ECMO circuit:
a. You must know the pressures regimen inside the ECMO circuit and the pressure alarms, both inflow and outflow, that will stop the CRRT machine. In
some CRRT systems the alarms can be changed to allow the connection to the
ECMO circuit.
b. Try to avoid the negative pressure compartment of the ECMO circuit, i.e. the
drainage line from the skin insertion to the centrifugal pump head. The risk of
air entrance in the ECMO circuit during CRRT manipulation is higher in the
negative pressure drainage line.
c. The CRRT machine can be connected to the side-ports across the artificial
lung used for monitoring the pressure drop (resistance) and withdrawing
blood gases to assess artificial lung performance. The post-pump pre-oxygenator port and the post-oxygenator port are both at positive pressure; both
sides can be used as inlet and outlet for the CRRT system, if this allows
positive pressure on both inflow and outflow lines. If the CRRT system in
use would not allow positive pressure in the inlet, some resistance to the
inflow line can be added but this will increase the risk of clotting and
haemolysis.
d. We prefer to connect the inflow line of the CRRT machine to the post-oxygenator port and the outflow line to the pre-oxygenator one (post-pre configuration). In this way a recirculation of oxygenated blood will result but the high
efficiency of the oxygenator would accomplish for it. We think this is safer
than having a shunted venous blood, as it will be the case if the inflow of the
CRRT will be attached to the pre-oxygenator and the outflow to the postoxygenator port. With the post-pre configuration, the oxygenator will also act
as an additional bubble-trap if air would enter the CRRT circuit.
e. The CRRT system can increase the heparin requirements. Regional anticoagulation with sodium citrate can be used in the CRRT machine while the
patient is maintained anticoagulated with heparin, or with other anticoagulants if heparin induced thrombocytopenia (HIT) is an issue.
30.3.7 Bleeding Management and Transfusion Targets

Critically ill patients normally develop anaemia after 3 days of ICU stay. The
ARDS patient on ECMO needs to have a good haematocrit level to maintain good
oxygen-carrying capacity and delivery. High volumes of blood, ranging between
70 and 100 mL/day, are drawn every day to monitor blood gases (patient and
ECMO), coagulation and routine biochemistry; therefore, the normal transfusion
requirement is 1 PRBC every 2 days. If bleeding will ensue, the transfusion requirements will increase dramatically, and this will impact the prognosis of our ECMO
patient [44, 45].
[email protected]
356
M. Bombino et al.
In order to prevent bleeding complications, a do no harm approach must be

used: no intramuscular or subcutaneous shots, no thoracentesis, no chest tubes and
no changes of nasogastric or urinary catheters.
Bleeding can be localised or generalised. In the first case manoeuvres at the
bleeding sites, like compressive gauzes on cannula insertion site or packing of nasal
or oral cavities, can per se be effective. If bleeding is generalised and persists also
after temporary heparin discontinuation, thromboelastography can be used in addition to normal coagulation assay, to understand if there is an underlying pathological
derangement in coagulation. In case of major haemorrhages, a massive transfusion
protocol must be followed trying to transfuse matched components (PRBC, FFP,
platelets, fibrinogen, tranexamic acid). Surgeons or interventional radiologists must
be involved early and useful imaging gathered [46]. If heparin is stopped, a strict
monitoring of the ECMO circuit for clots and increase in drop pressure across the
artificial membrane must be implemented.
30.4
Imaging in ECMO Patients
Chest x-rays and ultrasounds are our routine bedside imaging in the ECMO patient
[47]. They provide a lot of information and only in few cases we need other special
imaging. A CT-scan study can become necessary when we are searching for an
occult haemorrhage or an infection source and, if the ECMO course of the patient is
prolonged, to assess the evolution of lung parenchyma [48]. The transport of the
patient on ECMO outside the ICU is demanding but can be safely accomplished
with a careful ICU preparation (see chap. 39); therefore, we can weigh the risk of
transportation and the benefit we aspect from the study.
30.5
Physical Therapy and Mobilisation
Muscle dysfunction is common in patients in the intensive care unit (ICU) due to inactivity, inflammation, use of pharmacologic agents (corticosteroids, muscle relaxants,
neuromuscular blockers, antibiotics) and the presence of neuromuscular syndromes
associated with critical illness [4951]. Hence, it is important to prevent or attenuate
muscle deconditioning as early as possible in patients with expected prolonged bed rest,
and appropriate interventions (passive or active physiotherapy, mobilisation or walking) should be chosen according to clinical conditions [52]. There are several reports
in the literature on active physiotherapy and walking during ECMO [5359] which
mainly refer to patients candidate to lung transplantation; in this specific category of
patients, physiotherapy allows a better conditioning for transplantation, resulting in a
reduction in mechanical ventilation (MV) days after transplantation and in higher survival rates compared to those on MV [53]. Ambulatory ECMO and active physiotherapy require jugular cannulation with a double-lumen cannula, which allows an easier
management during the manoeuvre. However, according to the European Society of
Intensive Care Medicine (ESICM) recommendations [52], active physiotherapy should
[email protected]
30
357
not be performed in patients with haemodynamic instability, high FiO2 or high levels
of ventilatory support; hence, despite the published promising results, physiotherapy in
the acute severe ARDS patients on ECMO should be carefully evaluated. Major concerns are related to the risk of bleeding or decannulation during the manoeuvres and
to the inability of patients to tolerate physical efforts that may lead to non-protective
ventilation, sustained cough and bronchial bleeding or haemodynamic failure. Thus, in
acute ARDS patients, especially in those with femoral cannulation, physical therapy
should consist of mobilisation of the upper limbs progressively implemented in the
last part of respiratory weaning and fully performed only after ECMO disconnection.
References
1. Fulbrook P, Grealy B (2007) Essential nursing care of the critically ill patient. In: Elliott D,
Aitken L, Chaboyer W (eds) Critical care nursing. Mosby/Elsevier, Sydney
2. Coyer FM, OSullivan J, Cadman N (2011) The provision of patient personal hygiene in the
intensive care unit: a descriptive exploratory study of bedbathing practice. Aust Crit Care
24:198209
3. Strickland R, Buttery J, Frantzis P (2009) Royal Adelaide Hospital General ICU ECMO
guidelines. www.icuadelaide.com.au/files/manual_ecmo.pdf. Accessed 29 Apr 2013
4. Happ MB, Tate AJ, Hoffman LA et al (2010) Wash and wean: bathing patients undergoing
weaning trials during prolonged mechanical ventilation. Heart Lung 39:S47S56
5. Barr J, Fraser GL, Puntilllo K et al (2013) Clinical practice guidelines for the management of
pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med
41:263306
6. Shekar K, Roberts JA, Mullany DV et al (2012) Increased sedation requirements in patients
receiving extracorporeal membrane oxygenation for respiratory and cardiorespiratory failure.
Anaesth Intensive Care 40:648655
7. Wildschut ED, Ahsman MJ, Allegaert K et al (2010) Determinants of drug absorption in
different ECMO circuits. Intensive Care Med 36:21092116
8. Mousavi S, Levcovich B, Mojtahedzadeh M (2011) A systematic review on pharmacokinetic
changes in critically ill patients: role of extracorporeal membrane oxygenation. Daru
19:312321
9. Shekar K, Roberts JA, McDonald CI et al (2012) Sequestration of drugs in the circuit may lead
to therapeutic failure during extracorporeal membrane oxygenation. Crit Care 16:R194
10. Shekar K, Roberts JA, Welch S et al (2012) ASAP ECMO: Antibiotic, Sedative and Analgesic
Pharmacokinetics during Extracorporeal Membrane Oxygenation: a multi-centre study to optimise drug therapy during ECMO. BMC Anesthesiol 12:29
11. Shekar K, Roberts JA, Smith MT et al (2013) The ECMO PK Project: an incremental research
approach to advance understanding of the pharmacokinetic alterations and improve patient
outcomes during extracorporeal membrane oxygenation. BMC Anesthesiol 13:7
Influenza Investigators (2009) Extracorporeal membrane oxygenation for 2009 influenza
13. Brogan TV, Thiagarajan RR, Rycus PT et al (2009) Extracorporeal membrane oxygenation in
adults with severe respiratory failure: a multi-center database. Intensive Care Med 35:21052114
14. Pierson DJ (2005) Tracheostomy and Weaning. Respir Care 50:526533
15. Extracorporeal Life Support Organization (ELSO) (2009) Patient specific supplements to
the ELSO general guidelines. http://www.elsonet.org/index.php/resources/guidelines.html.
Accessed 29 Apr 2013
[email protected]
358
M. Bombino et al.
16. Camboni D, Philipp A, Lubnow M et al (2011) Support time-dependent outcome analysis for
veno-venous extracorporeal membrane oxygenation. Eur J Cardiothorac Surg 40:13411346
17. Combes A, Bacchetta M, Brodie D et al (2012) Extracorporeal membrane oxygenation for
respiratory failure in adults. Curr Opin Crit Care 18:99104
18. Aissaoui N, Luyt CE, Leprince P et al (2011) Predictors of successful extracorporeal membrane oxygenation (ECMO) weaning after assistance for refractory cardiogenic shock.
Intensive Care Med 37:17381745
25:131141
20. Firstenberg MS, Orsinelli DA (2012) ECMO and ECHO: the evolving role of quantitative
echocardiography in the management of patients requiring extracorporeal membrane oxygenation. J Am Soc Echocardiogr 25:641643
21. Marasco SF, Preovolos A, Lim K, Salamonsen RF (2007) Thoracotomy in adults while on
ECMO is associated with uncontrollable bleeding. Perfusion 22:2326
22. Huang PM, Ko WJ, Tsai PR et al (2012) Aggressive management of massive hemothorax in
patients on extracorporeal membrane oxygenation. Asian J Surg 35:1622
23. Joshi V, Harvey C, Nakas A et al (2013) The need for thoracic surgery in adult patients receiving extracorporeal membrane oxygenation: a 16-year experience. Perfusion 28(4):328332
24. Mauri T, Foti G, Zanella A et al (2012) Long-term extracorporeal membrane oxygenation with
minimal ventilatory support: a new paradigm for severe ARDS? Minerva Anestesiol 78:
385389
25. Reuter DA, Huang C, Edrich T et al (2010) Cardiac output monitoring using indicator-dilution
techniques: basics, limits, and perspectives. Anesth Analg 110:799811
26. Jansen JRC (1995) The thermodilution method for the clinical assessment of cardiac output.
27. Schmidt M, Brchot N, Hariri S et al (2012) Nosocomial infections in adult cardiogenic shock
patients supported by venoarterial extracorporeal membrane oxygenation. Clin Infect Dis
55:16331641
28. Conrick-Martin I, O'Gorman J, Lenehan D et al (2012) Nosocomial infections in a cohort of
extracorporeal life support patients. Crit Care Resusc 14:198201
29. Bizzarro MJ, Conrad SA, Kaufman DA et al (2011) Infections acquired during extracorporeal
membrane oxygenation in neonates, children, and adults. Pediatr Crit Care Med 12:277281
30. Sun HY, Ko WJ, Tsai PR et al (2010) Infections occurring during extracorporeal membrane
oxygenation use in adult patients. J Thorac Cardiovasc Surg 140:11251132
31. Aubron C, Cheng AC, Pilcher D et al (2013) Infections acquired by adults who receive extracorporeal membrane oxygenation: risk factors and outcome. Infect Control Hosp Epidemiol
34:2430
32. Lynch W (2012) Infections and ECMO. In: Annich G, Lynch W, MacLaren G, Wilson J,
Bartlett R (eds) ECMO extracorporeal cardiopulmonary support in critical care, 4th edn.
Extracorporeal Life Support Organization, Ann Arbor
33. Pieri M, Greco T, De Bonis M et al (2012) Diagnosis of infection in patients undergoing extracorporeal membrane oxygenation: a casecontrol study. J Thorac Cardiovasc Surg 143:14111416
34. Kao LS, Fleming GM, Escamilla RJ et al (2011) Antimicrobial prophylaxis and infection
surveillance in extracorporeal membrane oxygenation patients: a multi-institutional survey of
practice patterns. ASAIO J 57:231238
35. Augustin P, Lasocki S, Dufour G et al (2010) Abdominal compartment syndrome due to extracorporeal membrane oxygenation in adults. Ann Thorac Surg 90:e40e41
36. Maj G, Calabr MG, Pieri M et al (2012) Abdominal compartment syndrome during extracorporeal membrane oxygenation. J Cardiothorac Vasc Anesth 26:890892
37. Iapichino G, Pesenti A, Radrizzani D et al (1983) Nutritional support to long-term anesthetized and curarized patients under extracorporeal respiratory assist for terminal pulmonary
failure. JPEN J Parenter Enteral Nutr 7:5054
[email protected]
30
359
38. Scott KL, Boudreaux K, Thaljeh F et al (2004) Early enteral feedings in adults receiving venovenous extracorporeal membrane oxygenation. JPEN J Parenter Enteral Nutr 28:295300
39. Lukas G, Davies AR, Hilton AK et al (2010) Nutritional support in adult patients receiving
extracorporeal membrane oxygenation. Crit Care Resusc 12:230234
40. Umezawa Makikado LD, Flordels Lasierra JL, Prez-Vela JL et al (2013) Early enteral nutrition in adults receiving venoarterial extracorporeal membrane oxygenation: an observational
case series. JPEN J Parenter Enteral Nutr 37:281284
41. Kagan I, Singer P (2013) Nutritional imbalances during extracorporeal life support. In: Singer
P (ed) Nutrition in intensive care medicine: beyond physiology. World Rev Nutr Diet. Karger,
Basel. vol 105, pp 154159
42. Nutrition therapy in adult patients requiring Extracorporeal Membrane Oxygenation. http://
www.anzicrc.monash.org/nutrition-ecmo.html. Accessed 30 Apr 2013
43. Peek GJ, Killer HM, Sosnowski MA, Firmin RK (2002) Modular extracorporeal life support
for multiorgan failure patients. Liver 22:6971
44. Lamb KM, Cowan SW, Evans N (2012) Successful management of bleeding complications in
patients supported with extracorporeal membrane oxygenation with primary respiratory failure. Perfusion 28:125131
45. Aubron C, Cheng AC, Pilcher D et al (2013) Factors associated with outcomes of patients on
extracorporeal membrane oxygenation support: a 5-year cohort study. Crit Care 17:R73
46. Smith AH, Hardison DC, Bridges BC, Pietsch JB (2012) Red blood cell transfusion volume
and mortality among patients receiving extracorporeal membrane oxygenation. Perfusion
28:5460
47. Barnacle AM, Smith LC, Hiorns MP (2006) The role of imaging during extracorporeal membrane oxygenation in pediatric respiratory failure. AJR Am J Roentgenol 186:5866
48. Lidegran MK, Ringertz HG, Frenckner BP, Lindn VB (2005) Chest and abdominal CT during
extracorporeal membrane oxygenation: clinical benefits in diagnosis and treatment. Acad
Radiol 12:276285
49. Deem S (2006) Intensive-care-unit-acquired muscle weakness. Respir Care 51:10421052
50. De Jonghe B, Lacherade JC, Durand MC et al (2007) Critical illness neuromuscular syndromes.
Crit Care Clin 23:5569
51. Schweickert WD, Hall J (2007) ICU-acquired weakness. Chest 131:15411549
52. Gosselink R, Bott J, Johnson M et al (2008) Physiotherapy for adult patients with critical illness: recommendations of the European Respiratory Society and European Society of Intensive
Care Medicine Task Force on Physiotherapy for Critically III Patients. Intensive Care Med
34:11881199
53. Olsson KM, Simon A, Strueber M et al (2010) Extracorporeal membrane oxygenation in non
intubated patients as bridge to lung transplantation. Am J Transplant 10:21732178
55. Hayes D Jr, Kukreja J, Tobias JD et al (2012) Ambulatory venovenous extracorporeal respiratory support as a bridge for cystic fibrosis patients to emergent lung transplantation. J Cyst
Fibros 11:4045
56. Garcia JP, Iacono A, Kon ZN, Griffith BP (2010) Ambulatory extracorporeal membrane oxygenation: a new approach for bridge-to-lung transplantation. J Thorac Cardiovasc Surg
139:e137e139
58. Nosotti M, Rosso L, Tosi D et al (2013) Extracorporeal membrane oxygenation with spontaneous breathing as a bridge to lung transplantation. Interact Cardiovasc Thorac Surg 61:
5559
59. Rahimi RA, Skrzat J, Reddy DR et al (2013) Physical rehabilitation of patients in the intensive
care unit requiring extracorporeal membrane oxygenation: a small case series. Phys Ther
93:248255
[email protected]
Echocardiography in Venoarterial
and Venovenous ECMO
31
Nicola Bianco, Leonello Avalli, and Fabio Sangalli
31.1
Introduction
After worldwide successes with ECMO for the 2009 pandemic influenza H1N1
[1, 2], and publication of the results of the Conventional Ventilatory Support Versus
Extracorporeal Membrane Oxygenation for Severe Adult Respiratory Failure trial
that showed a survival advantage for ECMO over conventional ventilator management for severe respiratory failure [3], there has been renewed enthusiasm for venovenous ECMO (VV ECMO) for pulmonary support. Similarly, venoarterial ECMO
(VA ECMO) for acute cardiogenic shock is also becoming a more commonly used
tool in the early management of critically ill patients as part of a treatment strategy
to bridge to recovery, ventricular assist device implantation, or transplantation.
Algorithms that incorporate early VA ECMO as part of the treatment for witnessed
cardiopulmonary arrest (ECMO-assisted cardiopulmonary resuscitation) are also
becoming more common and have resulted in a twofold increase in neurologically
intact patients surviving to discharge [4] (Table 31.1).
At the same time, the intensivist has become more familiar with echocardiography and ultrasound techniques even outside the traditional field of cardiac surgery,
making it possible an ultrasound-guided approach to the indication, correct positioning, and patient monitoring during extracorporeal assistance. The targets of
ultrasound monitoring during extracorporeal circulation will be different depending
N. Bianco F. Sangalli (*)

L. Avalli
[email protected]
361
N. Bianco et al.
362
Table 31.1 Common indications and purposes of venoarterial ECMO
Common indications
Cardiogenic shock
Refractory cardiac arrest
Impossible weaning from cardiopulmonary bypass during cardiac surgery
Myocarditis
Intoxication or sepsis with severe myocardial depression
Primary graft failure after heart or lung transplant
Purposes of extracorporeal support
Bridge to recovery assistance until recovery and weaning from support
Bridge to bridge until the implantation of medium- or long-term mechanical support (VADs)
Bridge to transplant until heart transplant
Bridge to decision to evaluate the possible cardiac recovery and the most appropriate
therapy
on which approach (VV versus VA) has been used. During circulatory assistance,
after the patient has been selected and the cannulation sites have been evaluated,
the attention will be focused on the correct functioning of the mechanical support
(position of the cannulae, thrombus formation, etc.) and on the cardiac function,
monitoring its recovery and deciding when the patient is ready to be disconnected
from support. During respiratory support, after assessing the proper positioning of
the cannulae, it is important to evaluate the cardiocirculatory status of the patient
before and during the assistance.
The echocardiographic approach, transthoracic (TTE) versus transesophageal
(TEE), will depend on the patients characteristics (acoustic windows and echogenicity) and on the region of the heart and vessels of interest. Ultrasound monitoring during ECMO is not only important for cardiac evaluation but also for choosing
the best sites for cannulation (especially during percutaneous procedures) and for
monitoring the peripheral perfusion of the cannulated limbs.
Despite its important role in the management of critically ill patients, there are few
published data outlining the use and experience of echocardiography and ultrasound
in critically ill adults requiring extracorporeal support. Because ECMO is based on
the principles of oxygenation and hemodynamic support via blood flow within largebore cannulas placed in or near the heart in patients with cardiorespiratory failure,
echocardiography would be expected to have a fundamental role throughout the care
of patients supported on ECMO. In this chapter, we outline the targets and the modalities of ultrasound evaluation in patients requiring extracorporeal assistance.
31.2
Venoarterial ECMO
As pointed out in the previous paragraph, the primary aim of VA ECMO is to support the circulatory function in patients with cardiac failure (acute or chronic,
acutely decompensated) of different etiologies. Besides this traditional indication,
in the last decade, venoarterial ECMO has emerged as a circulatory and respiratory
support for refractory cardiac arrest, allowing diagnostic evaluation and treatment
in selected patients that would have a mortality rate close to 100 %.
[email protected]
31 Echocardiography in Venoarterial and Venovenous ECMO
363
Compared to other mechanical cardiac assistance devices, ECMO has the advantage of reduced costs, and the possibility of being easily and rapidly instituted outside
the operating room (critical care units, cardiac catheterization suites, or emergency
departments) even during cardiopulmonary resuscitation maneuvers. It also has limitations: it is a short-term support, with important infectious, thrombotic, and hemorrhagic complications, and causes an increase in the left ventricle afterload.
In the next paragraphs, we are going to analyze in details how a comprehensive
ultrasound evaluation should be performed during the different phases of ECMO support. For cannulation procedures, we refer to the peripheral cannulation, which is
employed in the majority of patients. Central cannulation (right atrium to ascending
aorta) represents a specific condition, reserved almost exclusively to the intraoperative
cardiac surgery period (impossible weaning from extracorporeal circulation). In these
cases, the ultrasound evaluation is represented by the intraoperative transesophageal
examination performed to assist the weaning from cardiopulmonary bypass.
31.3
Indications to VA ECMO Support
31.3.1 Cardiogenic Shock

Patients with cardiogenic shock should be carefully evaluated: A comprehensive
clinical history and examination together with a complete echocardiographic exam
are necessary to determine their functional status as well as the presence of clinical
conditions which can contraindicate the positioning of a circulatory assistance
(Table 31.2). Echocardiography helps to exclude new reversible pathologies, which
Table 31.2 Contraindications to ECMO support

General
Severe, not reversible cardiac insufficiency with no indication to transplant or VAD as
destination therapy
Severe, not reversible neurological injury
Terminal malignancy
Intracranial bleeding
Agea
Venoarterial ECMO
Aortic dissection
Severe aortic valve regurgitation
Not witnessed cardiac arrest
Prolonged CPRb
Venovenous ECMO
Cardiac arrest
Cardiogenic shock
Severe pulmonary hypertension
a
The age limit is not uniform in different centers. It is usually considered excessive to assist patients
older than 80 years old, even if the results are not significantly worse in this subpopulation
b
Also the duration of CPR in patients supported for refractory cardiac arrest is not uniform in various protocols. In our center, the time limit is 45 minutes, but limits of 90 minutes are also reported
in literature
[email protected]
364
N. Bianco et al.
may account for a patients hemodynamic instability (such as cardiac tamponade,

undiagnosed cardiac valve pathology, and LV dysfunction), avoiding the need for
ECMO support. During the exam, it is important to study both the systolic and the
diastolic function of the left ventricle, the function and dimensions of right ventricle, the valves (searching for functional or anatomic anomalies), and the pericardium, focusing the attention to the presence of pericardial effusion. It is of utmost
importance to accurately evaluate the aortic valve and the aorta (ascending and
descending): The presence of severe aortic valve regurgitation may have a detrimental impact on left ventricle unloading once a VA ECMO (in which LV afterload is
increased) is positioned, and the detection of an aortic dissection is an absolute
contraindication for VA ECMO positioning. The positioning of a venous cannula in
the right atrium also dictates that right-heart anatomy be evaluated for any structural
abnormality that may adversely affect the function and positioning of the cannula.
Notable findings include a prominent patent foramen ovale, atrial septal defect,
interatrial septal aneurysm, prominent Chiari network, presence of a pacemaker or
implantable cardioverter-defibrillator leads, and tricuspid valve pathology (such as
tricuspid stenosis or a tricuspid valve replacement).
31.3.2 Refractory Cardiac Arrest (Fig. 31.1)

The criteria to establish the indication for ECMO assistance in patients with cardiac
arrest are primarily based on the clinical history and on the timing of CPR start (i.e.,
the duration of no-flow or non-assisted circulatory arrest) (Table 31.2). In this category of patients, the echocardiographic evaluation is limited to the anatomic aspects
of cardiac chambers and valves, the study of cardiac function being impossible. In
PEA (pulseless electrical activity) patients, it is important to distinguish a true PEA
Fig. 31.1 ECMO cannulae

positioning during CPR
[email protected]
365
(the heart has no contractions, electromechanical dissociation) from a pseudo-PEA

(the heart has contractions which are not sufficient to generate a significant pulse
because of a severe dysfunction or extrinsic compression). Especially in these
patients, it is important to look for clear and gross anomalies which can explain the
cause of the arrest (the 4 Hs and 4 Ts of the ALS algorithm) focusing the attention on
the dimension of the cardiac chambers especially of the right ventricle to exclude
pulmonary embolism and on the presence of significant pericardial effusion. It is
also possible to evaluate the efficacy of the thoracic compressions estimating the
transaortic or transpulmonary flow. In these patients, it is also important to exclude
clinical conditions aortic dissection in primis which contraindicate cannulation.
31.4
Cannulation
Ultrasounds also have a key role during ECMO cannulation, as they assist in the
correct placement of ECMO cannulas. The evaluation of arterial and venous vessels
is important for choosing the best site for cannulation. With ultrasounds, it is possible to determine the position of the veins and arteries, the potential anatomical
variants, as well as the presence of significant pathologies of the venous (thrombosis) or arterial system (occlusion, significant stenosis, aneurysms, atheromas). If a
significant pathology is found in the examined vessels, alternative sites can be evaluated to determine the most appropriate for cannulation. This aspect has to be investigated as quickly as possible, especially in cardiac arrest patients.
The evaluation of vessels size, especially of the arteries, is important for choosing the dimension of the cannulae; the size of the arterial cannula is the major determinant of the maximal output achievable with ECMO support, as larger cannulae
permit a greater output. In our experience with adult patients, it is possible to achieve
a total circulatory assistance with cannulae ranging from 15 to 19 French depending
on the size of the patient. It is easy to obtain the approximate dimensions of the cannula in millimeters, dividing the size in French by 3. From this calculation, it is clear
that we can safely cannulate arteries with a diameter greater than 55.5 mm. The
ultrasound examination of the artery we intend to cannulate can tell if its diameter
is sufficient to accommodate the smallest cannula able to provide an adequate blood
flow to the patient. The presence of a vessel that is close in size to the cannula
dimension warrants the positioning of a peripheral reperfusion cannula to prevent
ischemia of the cannulated limb.
When commencing cannulation, guidewires are initially inserted percutaneously
and positioned within the heart or great vessels, before the advancement of cannulas
over these wires. Because of the strong echocardiographic artifacts that can be generated from these wires and cannulas, close attention must be paid to their placement.
In peripheral VA ECMO, the venous cannula is optimally located in the mid right
atrium to provide unobstructed flow of central venous blood into the circuit.
Transesophageal echocardiography (TEE) is useful to guide positioning. The return
cannula is usually placed in the contralateral femoral artery and the tip located in the
iliac artery or abdominal aorta. This region cannot be visualized with TEE. However,
[email protected]
366
N. Bianco et al.
imaging for placement of this cannula is usually not required. TEE can confirm that
the guidewire used in percutaneous arterial cannulation is in the lumen of the aorta
before dilatation, reducing the risk for extra-arterial cannula placement.
31.5
Monitoring ECMO Support
During ECMO support, the echographic evaluation performed at least on a daily

basis is of utmost importance for the study of cardiac function as well as for monitoring the assistance and managing the complications. All these aspects are investigated during all the period of assistance that can be divided into three stages: the
initial phase after connection, the intermediate of maintaining support, and the final
of weaning and de-connection.
31.5.1 Initial Stage (After Connection)

After ECMO connection and the beginning of assistance, it is important to verify if
the venous drainage is sufficient to maintain an adequate cardiac index. If it is not,
echocardiography can help identifying the causes and verifying the efficacy of corrective maneuvers: It is possible to identify hypovolemia and resolve it, check the
correct positioning of the cannulae, and detect masses (especially clots) which can
obstruct the venous drainage (Fig. 31.2). In case of low-flow states during ECMO,
it is also important to rule out problems to the arterial side of the circuit, for example, aortic dissection (not detected during pre-assistance evaluation or iatrogenic
after-cannulation) or severe aortic regurgitation.
In patients who need ECMO support for refractory cardiac arrest, it is possible to
perform the first cardiac evaluation soon after the beginning of the assistance to evaluate the recovery of mechanical activity that can take place spontaneously or after
therapeutic interventions (revascularization, evacuation of pericardial effusion, surgical, or pharmacological resolution of pulmonary embolism). At this stage, it is also
possible to detect possible abnormalities in contractility or significant valvular defects.
Fig. 31.2 Drainage cannula

completely obstructed by a
right atrial thrombus
[email protected]
367
31.5.2 Intermediate Stage (Maintaining Support)

The duration of the assistance is variable and depends on the initial causes that indicated it. ECMO support can be needed for few hours or days (e.g., in cases of
arrhythmic cardiac arrest, accidental hypothermia, intoxications) or for considerably longer periods in some other cases (e.g., for myocarditis or for patients being
bridged to long-term ventricular assistance device or transplantation).
The echocardiographic evaluation is essential during the subsequent days or
weeks not only to monitor the recovery of cardiac function but also to avoid and treat
potential problems and complications inherent to the mechanical assistance itself.
Together with the constant monitoring of global and segmentary contractility, it
is important to verify the adequate drainage of the cardiac chambers and avoid their
distention, particularly of the left ventricle. During peripheral VA ECMO, LV preload usually decreases (because of decreased pulmonary blood flow), but LV afterload increases (because of the pressurized return of blood via the arterial return
cannula). In cases of very severe LV dysfunction, especially when associated with
severe mitral regurgitation, the left ventricle may become more distended, and the
aortic valve may not open. This will be indicated by a loss of pulsatility on the arterial pressure waveform. This can lead to stasis and thrombosis in the ascending
aorta, LV cavity, and pulmonary veins and finally to problems to the mechanical
assistance itself and to its weaning together with a very high embolic risk (Figs. 31.3
and 31.4). Echocardiographic aspects are the following: a dilated and impaired left
ventricle, minimal aortic valve opening, severe spontaneous echo contrast in the
ascending aorta, and severe mitral regurgitation. Failure of the aortic valve to open
during peripheral VA ECMO support is a significant concern. In this situation, anticoagulation is increased and afterload is decreased while optimizing native LV output (by reducing ECMO flows and judicious use of inodilators) to facilitate aortic
valve opening and resolution of spontaneous echo contrast in the left ventricle.
Fig. 31.3 Complete

thrombosis of the left
ventricle
[email protected]
368
N. Bianco et al.
Fig. 31.4 Thrombus in the

ascending aorta, adherent to
the left coronary aortic cusp
Fig. 31.5 Flow variation sampled at the left main left coronary artery when IABP support is
suspended
Our strategy of cardiac rest is based upon the use of low-dose inotropes (e.g.,
dobutamine) to permit a constant opening and closure of the cardiac valves, together
with aortic counterpulsation when it is not contraindicated. The echocardiographic
control has a pivotal role in monitoring the cardiac rest; the use of aortic counterpulsation during ECMO support with peripheral cannulation is controversial, and
strong evidences to support or discourage it are lacking. In our experience, we found
a better left ventricle unloading, a preserved pulsatility to ECMO blood flow, and a
positive effect on coronary, splanchnic, and renal circulation (Fig. 31.5) when an
intra-aortic balloon pump (IABP) was in place. If the left ventricle is not sufficiently
drained and signs of pulmonary circulation overload are evident despite all these
[email protected]
369
Fig. 31.6 Flow variation in anterior tibial artery before (a) and after (b) positioning of a distal
reperfusion cannula in common femoral artery
measures, an active drainage of the left ventricle is mandatory. Various techniques

have been described in literature [5, 6], but the most common technique in adult
patients is represented by apical venting of the left ventricle; in this case, echocardiography can be used to verify the correct placement and function of the venting
cannula. We also recently described an alternative technique of transpulmonary
drainage, in which the echocardiographic control is important during positioning of
the drainage cannula and, still, to verify the efficacy of venting [7].
Echocardiography, through the evaluation of transaortic flow, represents a valuable tool to measure cardiac output during ECMO support: The thermodilution
techniques in fact appear largely unreliable because of the variable blood steal
caused by the venous drainage cannula, and the methods based on pulse contour
cannot always be used because of the scarce pulsatility in the peripheral arteries
when the ECMO flow is high.
During all the course of support, it is important to constantly monitor the perfusion of the cannulated limbs, to detect early when a peripheral reperfusion is needed,
and to control its effectiveness (Fig. 31.6).
31.5.3 Intermediate Stage (Detecting Complications)

Patients supported on ECMO are critically unwell and thus at increased risk for
complications due to the underlying disease process, critical illness, anticoagulation, or the device itself, particularly after cardiac surgery. Echocardiography can
[email protected]
370
N. Bianco et al.
assist in the detection and management of specific complications that may arise
during ECMO support. It is usually the first investigation requested when there is
a suspicion of ECMO malfunction, particularly for thrombosis, cannula displacement, or pericardial tamponade. A significant number of transthoracic and transesophageal echocardiographic studies are performed on patients who are on ECMO
[8, 9]. Because of the limitations of spatial resolution with TTE, TEE is usually
required to detect these complications. It enables rapid assessment of cannula positioning, cardiac filling and function, chamber compression from tamponade, and
cannula-associated thrombi.
The detection of cardiac tamponade and the significance of pericardial effusions
or collections can be difficult in patients supported on ECMO, as the heart is in a
partially bypassed state. There may be significant compression of a cardiac chamber
from a pericardial hematoma, but if this does not adversely affect cannula flow, it
may be of no hemodynamic significance.
The presence of a significant pericardial collection, which may even result in
cardiac chamber compression and may not necessarily affect hemodynamics or
ECMO flow while on support, may become a significant factor when contemplating
weaning from ECMO.
The cannulas used in ECMO can be large, which predisposes them to being a
common cause of complications, especially thrombosis or venous and arterial
obstruction. Thrombus formation associated with venous cannulas may either
reduce ECMO flow or complicate the clinical course by causing a pulmonary embolism. Additionally, on removal of a venous ECMO cannula, organized thrombi that
had formed around the cannula may be left behind in the heart. If the venous cannulas are removed at the time of surgery (such as during VAD insertion or cardiac
transplantation), it is recommended that intraoperative TEE of the inferior vena
cava be performed to assess for the presence of a venous cannula cast. If missed, this
may subsequently cause pulmonary embolization.
31.5.4 Final Stage (Recovery and Weaning)

The final area in which echocardiography plays an important role is the determination of recovery and readiness for weaning from ECMO support. For VA ECMO,
this may be performed under direct echocardiographic visualization and guidance
with or without a pulmonary artery catheter. The decision to wean ECMO support
and its timing are complex. Cardiac recovery is often marked by increasing pulsatility seen on the patients arterial line tracing. For VA ECMO, it would be unusual to
attempt to wean ECMO in the first 72 h [10]. As a general rule, it is possible to
evaluate the possibility of weaning a patient from ECMO support when the problems that led to the assistance (ischemic, arrhythmic, septic, etc.) are under control
and the hemodynamic status is stable.
The level of recovery and the likelihood for weaning ECMO support are based
on a multitude of clinical, hemodynamic, and echocardiographic variables which
are extensively addressed in a specific chapter. Echocardiographic parameters that
may suggest an attempt to cease ECMO support include an adequate LV ejection
[email protected]
371
fraction, an LV outflow tract velocity-time integral >10 cm, and the absence of LV
dilation and of signs of cardiac tamponade [11, 12]. Few data have been published
outlining methods and findings during ECMO weaning. In a study by Konishi et al.
[13], Doppler evaluation of flow in the descending aorta was used to help determine
cardiac recovery after viral myocarditis treated with peripheral VA ECMO. The
authors commented that the level at which the two flows mix may be of benefit in
determining whether adequate cardiac output is being generated by the native heart.
In another study [14], to evaluate hemodynamic and functional changes of the failing left ventricle by velocity vector imaging (VVI) and tissue Doppler, 22 patients
with cardiogenic shock supported by extracorporeal life support (ECLS) were
imaged during ECMO output variations inducing severe load manipulations. Load
variations were documented by a significant decrease in afterload (mean arterial
pressure), an increase in preload (left ventricular end-diastolic volume, E, E/Ea ratio
all increased), and an increase in the velocity-time integral. VVI parameters
increased, unlike tissue Doppler systolic velocities. The authors concluded that VVI
parameters are load dependent, like conventional Doppler echocardiographic data,
while the systolic velocities of the mitral annulus measured by Doppler tissue imaging (Sa) were found to be load independent and to have significant prognostic value
for predicting ECMO weaning. Sa was higher (>6 cm/s) in patients who survived.
When weaning VA ECMO, a common approach is to reduce the ECMO flow in
0.51.0 L/min increments and assess the clinical and hemodynamic parameters
(including heart rate, blood pressure, arterial waveform pulsatility, oxygen tension
level in a right radial arterial line, and changes in central venous pressure and pulmonary artery pressure) and echocardiographic parameters (stroke volume, ventricular dimensions, ventricular volumes, and ejection fraction) [15]. ECMO flows
are usually not reduced below 12 L/min, because of the increased risk for circuit
thrombosis at low-flow rates. If the patient remains hemodynamically stable at
flows as low as 1 L/min, we can suppose that the patient is ready to be disconnected
from support. The ECMO flow is raised up to the pre-test level so that anticoagulant
drugs can be stopped to facilitate the hemostasis during the phase of decannulation.
A slow and progressive weaning from support enables to accurately predict when a
patient is ready to be disconnected; however, because the pump cannot be completely stopped during the assistance for thromboembolic reasons, decannulation
represents a delicate phase, and a careful hemodynamic and echocardiographic
evaluation is needed to promptly identify and treat contingent problems.
After decannulation and hemostasis of the vascular access sites, ultrasounds are
extremely useful to verify possible damage to the cannulated vessels and to tissues
around them (laceration, pseudoaneurysms, relevant hematomas).
31.6
Venovenous ECMO
In patients with severe respiratory failure who need an extracorporeal assistance, it

is important to provide the appropriate support. Venovenous ECMO (VV ECMO)
usually represents the best option: It has fewer vascular complications, keeps the
lung perfusion intact allowing a better pulmonary penetration of antibiotics, and
[email protected]
372
N. Bianco et al.
provides a better oxygen delivery to the upper part of the body (especially brain and
heart) which may be impaired during a VA ECMO.
During a venovenous assistance, it is of paramount importance to ensure an adequate cardiac function, sufficient to completely sustain the circulation. Therefore,
before establishing a VV support, a complete evaluation of both echocardiographic
and hemodynamic parameters of the patient is essential. As for VA ECMO, the
exam has to be complete, paying particular attention on the estimation of pulmonary
pressure derived from tricuspidal regurgitation, and on the function and dimensions
of the right ventricle, which is one of the major determinants, with sepsis, of cardiac
insufficiency in patients with ARDS.
The venovenous assistance has generally no significant impact on cardiac performance. In fact as blood is taken from and then returned to the right heart, there is no
significant change in RV preload, and there is no adverse effect in hemodynamics in
the normal left heart. During VV ECMO, the pulmonary circulation receives blood
with increased oxygen content and increases the mixed venous oxygen saturation.
This may have two beneficial effects. First, it may decrease pulmonary vascular
resistance, leading to lower RV afterload. Second, it may indirectly improve LV
function by increasing oxygen delivery to the left heart and hence coronary arterial
circulation. Sepsis and increased pulmonary vascular resistance in response to significant hypoxemia may adversely affect RV function. Echocardiography can be
useful in documenting the effects of improving oxygenation and acidbase status on
the adequacy of RV function when VV ECMO is applied.
As for VA ECMO echocardiography is useful to guide the correct placement of
the cannulae. Venous cannulae may be placed through a femoral or internal jugular
approach. As a general rule, the tip of the return cannula should be placed in the mid
right atrium, clear from the interatrial septum and tricuspid valve, i.e., more proximally than the tip of the drainage cannula. If the access cannula is placed more
proximally than the return cannula, or if the two cannulae ends are too close, recirculation will occur, resulting in a reduction in the efficiency of the system and in the
amount of oxygenated blood passing into the pulmonary and systemic circulation.
Echocardiographic assistance can also help in detecting abnormal positioning of a
cannula against the interatrial septum, through a patent foramen ovale and into the
left atrium, in the coronary sinus, and across the tricuspid valve or subvalvular apparatus. Cannula malpositioning may also result in vascular or cardiac injury and
inadequate flows.
Imaging is recommended for placement of the Avalon Elite cannula [16]. These
dual-lumen cannulas are inserted via the right internal jugular vein. One lumen has
specifically located holes that drain blood from the inferior vena cava and superior
vena cava. The other lumen returns blood to the right atrium through a side hole that
is positioned to face the tricuspid valve. Meticulous positioning is required so that the
cannula tip is located in the inferior vena cava just below the cavo-atrial junction and
the return side hole is positioned to enable return flow across the tricuspid valve [17].
Even though VV ECMO is the assistance of choice for patients with respiratory failure, in some cases the cardiocirculatory status is inadequate to sustain
the patients hemodynamics. Also, in septic patients or in patients with severe
[email protected]
373
pulmonary hypertension, an acute cardiac insufficiency can develop during the

period of support. For this reason, even in patients needing mostly respiratory support, an echo-dynamic approach (based on the combination of echographic and
hemodynamic data) is advisable to identify and treat early a potential worsening
in cardiac function and to predict the need for a change to a circulatory assistance
(venoarterial or veno-artero-venous ECMO).
31.7
Conclusions
Ultrasounds play a fundamental role in managing patients supported with ECMO

during all the different stages of assistance, from indication to cannulation, monitoring, and weaning. Either during circulatory or respiratory assistance, ultrasounds
are fundamental to evaluate the cardiac function of the patients, providing information that determines appropriate patient selection. They are also needed to choose
the best vascular access sites, guide the insertion of cannulas, monitor progress,
detect complications, and help in determining recovery and weaning of support.
References
1. Davies A, Jones D, Bailey M, Beca J, Bellomo R, Blackwell N et al, Australia and New
Zealand Extracorporeal Membrane Oxygenation (ANZ ECMO) Influenza Investigators (2009)
Extracorporeal membrane oxygenation for 2009 influenza A(H1N1) acute respiratory distress
syndrome. JAMA 302:18881895
2. Firstenberg MS, Blais D, Louis LB, Stevenson KB, Sun B, Mangino JE (2009) Extracorporeal
membrane oxygenation for pandemic (H1N1) 2009. Emerg Infect Dis 15:20592060
3. Peek GJ, Mugford M, Tiruvoipati R, Wilson A, Allen E, Thalanany MM et al, CESAR Trial
Collaboration (2009) Efficacy and economic assessment of conventional ventilatory support
multicentre randomised controlled trial. Lancet 374:13511363
4. Cardarelli MG, Young AJ, Griffith B (2009) Use of extracorporeal membrane oxygenation for
adults in cardiac arrest (E-CPR): a meta-analysis of observational studies. ASAIO J
55:581586
5. Koenig PR, Ralston MA, Kimball TR, Meyer RA, Daniels SR, Schwartz DC (1993) Balloon
atrial septostomy for left ventricular decompression in patients receiving extracorporeal membrane oxygenation for myocardial failure. J Pediatr 122:S95S99
6. OConnor TA, Downing GJ, Ewing LL, Gowdamarajan R (1993) Echocardiographically
guided balloon atrial septostomy during extracorporeal membrane oxygenation (ECMO).
Pediatr Cardiol 14:167168
7. Avalli L, Maggioni E, Sangalli F, Favini G, Formica F, Fumagalli R (2011) Extracorporeal
Membrane Oxygenation: An Alternative to Surgical and Transeptal Venting in Adult Patients.
ASAIO J 57:3840
8. Sedgwick JF, Burstow DJ, Platts DG (2010) The role of echocardiography in the management
of patients supported by extracorporeal membranous oxygenation (ECMO). Int J Cardiol
147(Suppl):S16
9. Platts DG, Sedgwick JF, Burstow DJ, Mullany DV, Fraser JF (2012) The role of echocardiography in the management of patients supported by extracorporeal membrane oxygenation.
J Am Soc Echocardiogr 25:131141
[email protected]
374
N. Bianco et al.
10. Chen Y-S, Lin J-W, Yu H-Y, Jerng J-S, Ko W-J, Chang W-T et al (2008) Cardiopulmonary
resuscitation with assisted extracorporeal life-support versus conventional cardiopulmonary
resuscitation in adults with in-hospital cardiac arrest: an observational study and propensity
analysis. Lancet 372:554561
11. Scherer M, Sirat AS, Moritz A, Martens S (2011) Extracorporeal membrane oxygenation as
perioperative right ventricular support in patients with biventricular failure undergoing left
ventricular assist device implantation. Eur J Cardiothorac Surg 39:939944
12. Santelices LC, Wang Y, Severyn D, Druzdzel MJ, Kormos RL, Antaki JF (2010) Development
of a hybrid decision support model for optimal ventricular assist device weaning. Ann Thorac
Surg 90:713720
13. Konishi H, Misawa Y, Nakagawa Y, Fuse K (1999) Doppler aortic flow pattern in the recovering heart treated by cardiac extracorporeal membrane oxygenation. Artif Organs 23:367369
14. Aissaoui N, Guerot E, Combes A, Delouche A, Chastre J, Leprince P et al (2012) Twodimensional strain rate and Doppler tissue myocardial velocities: analysis by echocardiography of hemodynamic and functional changes of the failed left ventricle during different degrees
of extracorporeal life support. J Am Soc Echocardiogr 25:632640
15. Marasco SF, Lukas G, McDonald M, McMillan J, Ihle B (2008) Review of ECMO (extra corporeal membrane oxygenation) support in critically ill adult patients. Heart Lung Circ
17(Suppl):S41S47
16. Avalon Laboratories. Avalon Elite_ bi-caval dual lumen catheter. Available at: http://www.
avalonlabs.com/html/pulmonary_support.html. Accessed 25 Nov 2011
17. Javidfar J, Wang D, Zwischenberger J, Costa J, Mongero L, Sonett J et al (2011) Insertion of
bicaval dual lumen extracorporeal membrane oxygenation catheter with image guidance.
ASAIO J 57:203205
[email protected]
Haemodynamic Monitoring
32
Fabio Guarracino and Rubia Baldassarri
32.1 Introduction
Critically ill patients submitted to ECMO (extracorporeal membrane oxygenation)
routinely require an advanced haemodynamic monitoring to evaluate either the cardiovascular function or the effectiveness of the cardiopulmonary bypass. Adequate
monitoring should be focused on the evaluation of the underlying life-threatening
cardiac and/or respiratory disease in order to assess both the severity of the cardiac
and/or respiratory failure and the eventual recovery from organ failure. Proper haemodynamic monitoring can also help decide the timing of weaning from the
mechanical circulatory support.
Veno-arterial (VA ECMO) is generally employed for the treatment of refractory cardiogenic shock due to different causes (post-cardiotomy, post-heart transplant, cardiomyopathy, myocarditis, acute coronary syndrome). Veno-venous
(VV ECMO) is used for the treatment of severe respiratory failure primarily due
to adult respiratory distress syndrome (ARDS), pneumonia, trauma or primary
graft failure following lung transplantation [1, 2]. Despite VA and VV ECMO
being generally employed in different pathological populations, cardiac failure
can be present in patients with severe respiratory failure requiring extracorporeal
mechanical device [3].
32.2 Haemodynamic Monitoring

The goal of the haemodynamic monitoring in the ECMO patients is to provide
information about the adequacy of the organs and the peripheral perfusion [4].
F. Guarracino (*) R. Baldassarri

Cardiothoracic Anaesthesia and Intensive Care Medicine,
Azienda Ospedaliero Universitaria Pisana, Pisa, Italy
[email protected]
375
376
F. Guarracino and R. Baldassarri
32.2.1 VA ECMO

Patients with severe cardiac dysfunction, submitted to the VA extracorporeal
mechanical support, can easily suffer from low cardiac output (CO) syndrome
because of either the underlying cardiogenic shock or the presence of the cardiopulmonary bypass (CPB). Adequate haemodynamic monitoring should provide information about either the underlying cardiac disease evaluating the patients cardiac
performance or the effectiveness of the ECMO support [5].
ECMO should adequately support the patients baseline cardiac performance in
order to achieve an optimal tissue perfusion, providing adequate flow rates and oxygen supply.
In this contest, the main target of the haemodynamic monitoring is the evaluation
of the global tissue perfusion. In addition, a good cardiopulmonary monitoring
should correctly define the haemodynamic profile of the patient in order to guide the
therapy and follow up the progression of the underlying disease. The optimization
of either the oxygen delivery or the haemodynamic status can result from the information acquired by the correct haemodynamic analysis. For all these considerations,
the most appropriate haemodynamic monitoring for the patients in VA ECMO
should be focused on the evaluation of the patients cardiac function (throughout the
echocardiographic assessment of the heart function at the various steps of the
mechanical support and the measure of both the CO and the mixed venous oxygen
saturation (SvO2)) and on the efficiency of the mechanical device to sustain the
cardiocirculatory function and the organ perfusion (throughout the monitoring of
the SvO2, the urinary output and the blood lactate values).
32.2.1.1 Cardiac Output

Although echocardiography remains the best tool to evaluate the cardiac function,
many monitoring systems that provide the continuous assessment of the CO are
actually available. Among them the pulmonary artery catheterization (PAC) with
the Swan-Ganz catheter is still considered the gold standard of haemodynamic
monitoring [6]. Although the role of the PAC is still the object of discussion among
the scientific community, nevertheless its limitations and contraindications, the pulmonary artery catheterization provides many information about the haemodynamic
profile of the critically ill patient [7, 8]. Pulmonary artery pressures and left and
right cardiac chamber pressures can be directly measured by pulmonary artery catheterization, and both CO and mixed venous oxygen saturation (SvO2) can be continuously displayed by the most recent monitoring instrumentations. In addition,
PAC allows the calculation of other derivate haemodynamic parameters including
the systemic and pulmonary vascular resistances [4].
Because PAC is an invasive technique and the haemodynamic data collected by
the pulmonary artery catheterization can be misinterpreted in some clinical contests
(e.g. tricuspid regurgitation), several either minimally invasive or absolutely noninvasive CO monitoring systems have been recently introduced in the clinical scenario. The most common techniques employed for the measure of the CO are listed
in Table32.1.
[email protected]
377
32 Haemodynamic Monitoring
Table 32.1Minimally
invasive technique for CO
monitoring
Fick principle
Pulsed Doppler technology
Echocardiography
Pulse contour analysis
Bioimpedance
Bioreactance
32.2.1.2 SvO2
The maintenance of the adequate tissue oxygen supplydemand balance is fundamental in the critically ill patients on ECMO. The conduct of the extracorporeal mechanical device should provide an optimal tissue perfusion throughout the management of
the different pathophysiological variables encountered during the ECMO assistance.
The SvO2, measured in the main pulmonary artery with the Swan-Ganz catheter,
has been largely recognized as a good indicator of the global tissue perfusion provided by both the patients cardiac performance and the extracorporeal assistance.
According to the Fick principle, SvO2 value results from the combination of five
major variables (Fig.32.1):

SvO 2 = SaO 2 ( VO 2 / CI Hb PO )
where SaO2 arterial, O2 saturation, CI cardiac index, PO O2 affinity, Hb oxyhaemoglobin content.

SvO2 does not always correspond to the effective tissue O2 tension (PvO2) that
depends on the position on the oxyhaemoglobin dissociation curve. During ECMO
the continuous variation of either the haemodynamic parameters or the haemoglobin
content and the arterial blood oxygenation makes the trend of the SvO2 values a better indicator than an absolute value of mixed venous oxygen saturation of the matching between the oxygen delivery (DO2) and the oxygen consumption (VO2) [9]. As it
is clearly showed by the Fick equation, the variation of any of the determinants of the
SvO2 can lead to significant changes of the SvO2 values during its continuous monitoring. Because the relationships among the five components of the equation are
mathematically different and it is uncommon that only one of the variables changes
independently from the others in a clinical contest, a normal value of SvO2 is quite
difficult to define; it has been demonstrated that a SvO2 range between 60 and 80%
suggests an appropriate peripheral perfusion.
Although the measurement of the SvO2 during ECMO generally correctly reflects
the effectiveness of the tissue perfusion, sometimes SvO2 value could not adequately
indicate the global perfusion.
Although the SvO2 expresses the match between DO2 and VO2, the value of the
SvO2 calculated in the main pulmonary artery could not reflect the eventual regional
distribution of the blood flow and the different perfusion of the different body districts. It should be considered that during VA ECMO, low coronary and/or cerebral
flow can occur because of a certain degree of deoxygenated and oxygenated blood
mixture at level of the descending thoracic aorta.
[email protected]
378
Blood from the LV
Mixing
Blood
Blood from the ECMO

arterial inflow cannula
Fig. 32.1 VA ECMO with femoral approach. Left panel: the arrows indicate the venous (blue
arrow) and arterial (red arrow) cannulas. Right panel: the black double arrow indicates the proximal
part of the descending aorta where the blood mixture of the oxygenated blood coming from the arterial cannula and the less oxygenated blood running from the patients LV can occur. The location of
the arterial cannula inside the descending thoracic aorta leads to a mixture between the oxygenated
blood coming from the ECMO circuits and the less oxygenated blood ejected by the LV
32.2.2 VV ECMO

Patients submitted to VV ECMO do not generally present cardiac function impairment, since severe respiratory dysfunction requiring extracorporeal support because
of being a nonresponder to the conventional therapy can occur in patients with
chronic pulmonary hypertension and right heart decompensation. Therefore, cardiac dysfunction can coexist with respiratory failure even when a cardiocirculatory
support is not required. Because VV ECMO do not provide circulatory support, it
has generally few effects on the systemic haemodynamic. This is not always true
because it should be considered that the high blood flow in the pulmonary artery
would increase the pulmonary vascular resistances (PVR) with consequent increase
of the right ventricle (RV) afterload that can lead to RV failure. For this reason the
patients in VV ECMO require an advanced haemodynamic monitoring too.
VV ECMO supports the respiratory function providing an adequate arterial oxygenation and CO2 removal. The adequate monitoring during VV ECMO should be
[email protected]
379
focused on the effectiveness of the respiratory support provided by both the extracorporeal help and the patients breathing [10]. The arterial oxygen saturation and
the SvO2 are the best indicators of the respiratory profile. According to the guidelines, the respiratory support provided by the VV ECMO is considered adequate
when the SaO2 >80% and SvO2 >70% [11]. As reported by some recent literature,
low tissue oxygenation and low perfusion are associated with worse outcome. For
this reason higher values of SaO2: could be required in patients under VV ECMO to
maintain an adequate organ perfusion [5]. In VV ECMO the arterial oxygenation
strictly depends on the blood flow, so the higher is the flow through the circuit, the
better is the oxygenation. Conversely, because the CO2 is highly diffusive, low
blood flows are generally sufficient to provide decarboxylation. Blood gases can be
easily measured by blood samples obtained from the patients arterial line.
Hypoxaemia, and consequently low SvO2, can frequently occur in VV ECMO
because of different mechanisms. One of them is the recirculation. Recirculation is an
unavoidable effect of the VV ECMO depending on the aspiration back into the extracorporeal circuit of a variable portion of the oxygenated blood previously infused in
the right atrium (RA) [12]. This phenomena is strictly correlated to the physical characteristic of the venous cannula and to the position of the proximal and distal lumen
on the same cannula [13, 14]. The main effect of the recirculation is that the blood in
the RA can be poorly oxygenated and the oxygen delivery can be decreased, leading
to global and/or regional hypoperfusion. For the presence of the recirculation, the
value of the mixed venous saturation measured at the venous line of the ECMO circuit is not always appropriate. SvO2 should be directly measured in the pulmonary
artery by the Swan-Ganz catheter which can be considered a better indicator of the
global perfusion. Another significant cause of hypoxaemia during VV ECMO is the
mixture between the oxygenated blood coming from the circulatory support and the
low oxygenated blood running from the patients venae cavae. In the RA the blood
oxygen content decreases, and the tissue oxygenation is lower than that required.
32.2.2.1 Harlequin Syndrome

Patients submitted to percutaneous VA ECMO can suffer from regional low
perfusion despite adequate oxygenation and pump flows rate being preserved [2].
The proximal cannula is generally located in the upper part of the descending thoracic aorta. Because ECMO provides up to 80 % of the global blood flow and a
native flow through the patients aortic valve is preserved, a mixture of the oxygenated blood coming from the arterial cannula and the less oxygenated blood running
from the patients left ventricle (LV) can occur (Fig.32.2). Therefore, the O2 content
of the blood in the ascending aorta can decrease. Because of the anatomic position
of the coronary arteries and of the epiaortic vessels, the myocardial and the cerebral
flow are at main risk to receive low oxygenated blood [15]. The coronary arteries
hypoperfusion can result in myocardial ischaemia with consequent worsening of the
heart function, limiting the organ recovery. The cerebral hypoperfusion can lead to
moderate to severe neurologic complications that can negatively impact on the
patients outcome [16].
The alteration of the regional perfusion can lead to severe organ dysfunction; the
absolute value of the SvO2 as well as its variations in the range of normality measures
[email protected]
380
Fig. 32.2 The figure shows a case of ECMO in which a multimodal monitoring was applied: note
the transesophageal echo probe in site and the NIRS monitoring displaying the cerebral
saturation
Table 32.2Information
on ECMO flow obtained
by coupling NIRS and
SvO2 monitoring
NIRS
Normal
Reduced
Reduced
SvO2
Normal
Reduced
Normal
Flow
Adequate
Reduced
Normal but not adequate
cannot represent the mismatch between the DO2 and VO2 occurring in some of the
body districts (mainly in the coronary and cerebral flow) (Table32.2). For this reason
a multimodal monitoring approach should be considered. The coronary flow may be
investigated with echocardiography, and the cerebral flow should be assessed with
neuromonitoring systems like the near-infrared spectroscopy (NIRS) (Fig.32.2).
The double check between the global SvO2 and the regional SrO2 provided by the
NIRS monitoring can lead to a better interpretation of the haemodynamic assessment, revealing any significant variation of the tissue perfusion [17].
32.3 Conclusions
1. The haemodynamic monitoring in the ECMO patients should basically investigate the effectiveness of the cardiocirculatory and/or respiratory support.
[email protected]
381
2. Any information about the global tissue perfusion in VA ECMO and about the gas
exchanges in VV ECMO should be promptly acquired by the appropriate tools.
3. The most adequate monitoring system should be chosen according either to the
underlying disease or to the type of mechanical support provided.
4. The haemodynamic monitoring should provide information about the progression of the disease and the eventual recovery from organ failure.
5. The haemodynamic monitoring should guide the therapy evaluating the response
of the patient to the treatment.
6. The appropriate haemodynamic monitoring should allow the optimization of the
clinical and mechanical support.
References
1. Sidebotham D, McGeorge A, McGuinness S etal (2009) Extracorporeal membrane oxygenation for treating severe cardiac and respiratory disease in adults: part 1overview of extracorporeal membrane oxygenation. J Cardiothorac Vasc Anesth 23:886892
2. Sidebotham D, McGeorge A, McGuinness S etal (2010) Extracorporeal membrane oxygenation for treating severe cardiac and respiratory failure in adults: part 2-technical considerations. J Cardiothorac Vasc Anesth 24:164172
3. Marasco SF, Lukas G, McDonald M, McMillan J, Ihle B (2008) Review of ECMO (extra corporeal membrane oxygenation) support in critically ill adult patients. Heart Lung Circ 17(Suppl
4):S41S47
4. Porhomayon J, El-Solh A, Papadakos P, Djalal N (2012) Cardiac output monitoring devices:
an analytic review. Intern Emerg Med 7:163171
5. Guarracino F, Zangrillo A, Ruggeri L, Pieri M, Calabr MG, Landoni G, Stefani M, Doroni L,
Pappalardo F (2012) Beta-blockers to optimize peripheral oxygenation during extracorporeal
membrane oxygenation: a case series. J Cardiothorac Vasc Anesth 26(1):5863
6. Chatterjee K (2009) The Swan-Ganz catheters: past, present, and future. A viewpoint.
Circulation 119(1):147152
7. Pinsky MR, Vincent JL (2005) Let us use the pulmonary artery catheter correctly and only
when we need it. Crit Care Med 33:11191122
8. Barnett CF, Vaduganathan M, Lan G, Butler J, Gheorghiade M (2013) Critical reappraisal of
pulmonary artery catheterization and invasive hemodynamic assessment in acute heart failure.
Expert Rev Cardiovasc Ther 11(4):417424
9. Chauhan S, Subin S (2011) Extracorporeal membrane oxygenation, an anesthesiologists perspective: physiology and principles. Part 1. Ann Card Anaesth 14(3):218229
10. Schmidt M, Tachon G, Devilliers C, Muller G, Hekimian G, Brchot N, Merceron S, Luyt CE,
Trouillet JL, Chastre J, Leprince P, Combes A (2013) Blood oxygenation and decarboxylation
determinants during venovenous ECMO for respiratory failure in adults. Intensive Care Med
39(5):838846
11. ELSO Guidelines for Cardiopulmonary Extracorporeal Life Support, Extracorporeal Life
Support Organization, Version 1:1. April 2009 Ann Arbor, MI. Available at www.elso.med.
umich.edu
12. Walker JL, Gelfond J, Zarzabal LA, Darling E (2009) Calculating mixed venous saturation during
veno-venous extracorporeal membrane oxygenation. Perfusion. doi:10.1177/0267659109354790
13. Bonacchi M, Harmelin G, Peris A, Sani G (2011) A novel strategy to improve systemic
oxygenation in venovenous extracorporeal membrane oxygenation: the -configuration.
JThorac Cardiovasc Surg 142(5):11971204
14. Mennen MT, Rosenfeldt FL, Salmonsen RF (2012) Veno-right ventricular cannulation reduces
recirculation in extracorporeal membrane oxygenation. Perfusion 27(6):464469
[email protected]
382
15. Wong JK, Smith TN, Pitcher HT, Hirose H, Cavarocchi NC (2012) Cerebral and lower limb
near-infrared spectroscopy in adults on extracorporeal membrane oxygenation. Artif Organs
36(8):659667
16. Slater JP, Guarino T, Stack J, Vinod K, Bustami RT, Brown JM 3rd, Rodriguez AL, Magovern
CJ, Zaubler T, Freundlich K, Parr GV (2009) Cerebral oxygen desaturation predicts cognitive
decline and longer hospital stay after cardiac surgery. Ann Thorac Surg 87(1):3644
17. Hoffman GM (2006) Pro: near-infrared spectroscopy should be used for All cardiopulmonary
bypass. J Cardiothorac Vasc Anesth 20(4):606612
[email protected]
Respiratory Monitoring
During VA ECMO
33
Daniela Pasero, Pietro Persico, Tommaso Tenaglia,

and Vito Marco Ranieri
Venoarterial extracorporeal membrane oxygenation (VA ECMO) represents an

excellent support in severe cardiogenic shock. Arterial oxygenation is largely provided by the oxygenator, while the flow through the pulmonary circulation is dramatically reduced, because blood is drained from the right atrium by the venous
cannula to the circuit and returns to the patient through the arterial line [1, 2].
Therefore, it is difficult to evaluate the lung function during ECMO support, even
though the lung might not be initially affected.
Why do we need to know lung function during VA ECMO? The main reason is
to allow VA ECMO weaning when heart function has recovered.
33.1
How Can Lung Function Deteriorate During VA ECMO?
One of the main causes is left ventricular congestion. During VA ECMO, if the left
ventricular function is completely depressed, the blood coming from the bronchial
arteries to the left atrium cannot be ejected easily, resulting in a left ventricular distension with deleterious effects on heart and lung. In turn, left ventricular congestion increases the venous pulmonary pressure and induces pulmonary oedema.
Although this might be clinically irrelevant during VA ECMO because the oxygenator provides gas exchange, it becomes pivotal when the cardiac function recovers
D. Pasero P. Persico T. Tenaglia

Anaesthesiology and Critical Care Department, Citt della Salute e della Scienza,
Ospedale S. Giovanni Battista-Molinette, Corso Bramante, 88, Turin 10126, Italy
V.M. Ranieri, MD (*)
Citt della Salute e della Scienza, Ospedale S. Giovanni Battista-Molinette,
Corso Bramante, 88, Turin 10126, Italy
Department of Anesthesia and Critical Care, University of Turin, Turin, Italy
[email protected]
383
384
D. Pasero et al.
and the VA ECMO might be removed. Hence, a persistently congested lung might
be compromised in its function and drive to a severe respiratory failure, which
blocks weaning from the extracorporeal device [3]. An experimental study showed
the histological damage of the lung after prolonged VA ECMO: morphological
changes of the lungs varied from frank pulmonary oedema to alveolar haemorrhage
and pulmonary parenchymal necrosis. This histological phenomenon could explain
the reduction of compliance after weaning from the VA ECMO [4]. A recent observational study, performed on patients undergoing VA ECMO for refractory cardiogenic shock, who required long-term mechanical circulatory support (MCS),
showed that 27 % developed acute lung injury during extracorporeal support,
increasing the rate of weaning failure for hypoxaemia and the rate of early mortality
(87 %). This complication might be explained by different mechanisms, such as a
pre-existing unrecognized lung injury at VA ECMO implantation, or it might be
related to persistency of pulmonary oedema, or it might be due to a systemic inflammatory response syndrome and multi-organ dysfunction [5]. Furthermore, Chen
et al. observed during the first 96 h after VA ECMO implantation that one of the
main factors predicting weaning failure and a poor outcome might be lung dysfunction [6]. Ensuring early an adequate left ventricular unload might prevent or reduce
this phenomenon.
33.2
How Do We Monitor the Lung Function?
The chest X-ray might be useful to evaluate the level of oedema, but it does not give
us a clear idea of the real lung function and it could be misleading. Haemodynamic
monitoring with the pulmonary catheter might be useful to evaluate and follow the
unloading of the left atrial and ventricular pressure, which might represent an indirect measurement of pulmonary congestion. Furthermore, the haemodynamic monitoring with PiCCO has been used during veno-venous ECMO, and it might be
useful to calculate the extravascular lung water index (ELWI) and giving an estimation of the degree of the pulmonary oedema [7]. During VA ECMO this kind of
monitoring is useless because the pump function of the heart is largely supported by
the extracorporeal device; therefore, cardiac output through the lung is markedly
reduced, and it does not allow the measurement of ELWI. A more recent and interesting alternative method to monitor lung parenchyma might be chest ultrasound
examination. This method is easy to perform at bedside, and it is inexpensive and in
certain case might be more sensitive than chest X-ray: in fact interstitial oedema can
be diagnosed with 97 % of sensitivity and 95 % of specificity, when three or more
B lines, which are typical hyperechoic artifacts (ring-down artifacts), are present.
Furthermore, it is highly sensitive for the evaluation of pleural effusion and parenchymal consolidation, even superior to chest X-ray, and some authors proposed to
use ultrasound to differentiate acute respiratory distress syndrome from pulmonary
oedema [810].
[email protected]
33
Respiratory Monitoring During VA ECMO
33.3
385
Lung Management
Mechanical ventilation during VA ECMO should be as protective as possible to

avoid ventilatory-induced lung injury, although lung injury is not usually present
when mechanical circulation support (MCS) is positioned. Since the lung does not
contribute to a relevant extent to gas exchange when VA ECMO provides total
assistance, tidal volume and minute ventilation should be reduced and positive
end-expiratory pressure (PEEP) should be optimized to keep the lung open and to
reduce atelectasis. Several authors showed that injurious mechanical ventilation
with high tidal volumes and high peak inspiratory pressures might be associated
with an increased risk of developing acute lung injury in ventilated patients without lung injury [11, 12]. Usually, to ensure lung protection, some authors suggest
to maintain a peak inspiratory pressure lower than 2025 cmH2O, PEEP at 1015
cmH2O, and respiratory rate lower than 10 per minute with a FiO2 = 30 % [13].
During extracorporeal support, respiratory mechanics should be routinely monitored, because compliance might be modified by the increase in oedema or the
development of acute lung injury, due to inflammatory processes or flogistic infiltrations [4, 14]. However, respiratory mechanics do not give us a real estimation of
the lung function in term of gas exchange: in fact in some cases, compliance could
be normal, but the oxygenation capability could be extremely low, because of acute
pulmonary oedema. Therefore, during VA ECMO, it is extremely important to
ventilate the lung gently and to avoid atelectasis, while the unloading of the left
ventricle should be the priority to reduce the risk of pulmonary congestion and
oedema.
How can the left ventricle be unloaded? Different methods have been proposed:
1. The most common technique in adult patients is represented by the use of intraaortic balloon counterpulsation with the aim of reducing afterload and increasing
coronary perfusion [5].
2. Another possibility is the decompression of the left atrium with a transseptal
cannula incorporated in the circuit, which allows the unloading of the pulmonary
venous circulation and might reduce the pulmonary congestion [3].
3. An alternative percutaneous method might be performed by a pulmonary arterial
catheter, which drains the residual blood in the pulmonary circulation and the
left atrium to the venous line of the ECMO circuit [15].
4. A further option, by using the percutaneous approach, is the Impella microaxial
flow pump, which permits the direct drainage of the left ventricle [16].
Recently a new method to unload the left ventricle was implemented. The novel
approach is performed by a mini-thoracotomy with a Seldinger technique using an
arterial high-flow cannula, inserted in the apex of the left ventricle (transapical left
ventricle vent, TLVV). This system allowed the drainage of blood and might reduce
the wall tension of the ventricle and the pulmonary congestion. TLVV is connected
to the venous inflow line and contributed to total flow, before the oxygenator. This
new system might allow switching from a VA ECMO to a short-term left
[email protected]
386
D. Pasero et al.
ventricular assist device (L-VAD), if the right ventricular function is maintained

and the lung function is good. The great advantage on the standard L-VAD is that
an oxygenator is available in the circuit and gives the possibility to evaluate the
pulmonary function gradually [17]. In our centre, from January 2010 to June 2012,
16 patients supported by peripheral VA ECMO for cardiogenic shock underwent
TLVV implantation. All patients were switched to a short-term L-VAD in two
phases: first, patients were weaned from the right ventricle support (arterio-arterial
ECMO), and then they were weaned from the oxygenator. During this second
phase, blood gases were analyzed before the oxygenator and from the arterial line
of the patient, with the same FiO2 both at the oxygenator and at the mechanical
ventilator, to obtain a baseline value. Then, the FiO2 of the gas blender of the oxygenator was set at 21 % and other two samples were obtained, to decide if the
oxygenator could be removed [18]. Additionally, continuous monitoring of endtidal CO2 provides a fundamental guide to tailor ventilation in order to ensure an
adequate PCO2 and pH in the lungs. It becomes often necessary to add dead space
on the ventilator circuit to optimize local pH when this remains too high despite a
reduction in minute ventilation.
Furthermore, during VA ECMO, patients can also breathe spontaneously in
the absence of primary respiratory failure. However, ECMO could be a specific
challenge for the respiratory control system because it reduces the venous return
to the pulmonary circulation dramatically and with variable proportion, while
aortic blood flow, mean arterial pressure, and CO2 remain unchanged. A group of
authors reported a case where the increased of PaCO2 induced by the reduction
of the gas flow to the oxygenator did not change the breathing pattern of the
patient, while an increase of the venous return to the right atrium and ventricle
obtained with the reduction of the blood flow on the ECMO circuit did, with an
increase in the respiratory rate [19]. This mechanism has been described only in
an experimental model of VA ECMO, where the acute rise in pressure in the
pulmonary artery and right ventricle or in the left atrium might modify the
breathing pattern, independently from neural control [20, 21]. Further clinical
studies are needed to better understand the mechanism of control of breathing
during VA ECMO.
33.4
Conclusions
In conclusion during VA ECMO, a proper respiratory monitoring is difficult to perform, but the keys to preserve the lung function are the unloading of the left atrium
and ventricle and a protective lung ventilation strategy. Further studies are needed
to evaluate the neural control of breathing during VA ECMO, when the blood
diverges from the right atrium and ventricle and the pulmonary circulation to the
ECMO circuit.
[email protected]
33
Respiratory Monitoring During VA ECMO
387
References
1. Sayer GT, Baker JN, Parks KA (2012) Heart rescue: the role of mechanical circulatory support
in the management of severe refractory cardiogenic shock. Curr Opin Crit Care 18:409416
2. Bakhtiary F, Keller H, Dogan S, Dzemali O, Oezaslan F, Meininger D, Ackermann H, Zwissler
B, Kleine P, Moritz A (2008) Venoarterial extracorporeal membrane oxygenation for treatment
of cardiogenic shock: clinical experiences in 45 adult patients. J Thorac Cardiovasc Surg
135:382388
3. Aiyagari RM, Rocchini AP, Remenapp RT, Graziano JN (2006) Decompression of the left
atrium during extracorporeal membrane oxygenation using a transseptal cannula incorporated
into the circuit. Crit Care Med 34:26032606
4. Koul B, Willen H, Sjoberg T, Wetterberg T, Kugelberg J, Steen S (1991) Pulmonary sequelae
of prolonged total venoarterial bypass: evaluation with a new experimental model. Ann Thorac
Surg 51:794799
5. Boulate D, Luyt CE, Pozzi M, Niculescu M, Combes A, Leprince P, Kirsch M (2013) Acute
lung injury after mechanical circulatory support implantation in patients on extracorporeal life
support: an unrecognized problem. Eur J Cardiothorac Surg 44(3):544549
6. Chen YS, Ko WJ, Chi NH, Wu IH, Huang SC, Chen RJ, Chou NK, Hsu RB, Lin FY, Wang SS,
Chu SH, Yu HY (2004) Risk factor screening scale to optimize treatment for potential heart transplant candidates under extracorporeal membrane oxygenation. Am J Transplant 4:18181825
7. Banach M, Soukup J, Bucher M, Andres J (2010) High frequency oscillation, extracorporeal
membrane oxygenation and pumpless arteriovenous lung assist in the management of severe
ARDS. Anestezjol Intens Ter 42:201205
8. Gardelli G, Feletti F, Nanni A, Mughetti M, Piraccini A, Zompatori M (2012) Chest ultrasonography in the ICU. Respir Care 57:773781
9. Lichtenstein DA, Meziere GA (2008) Relevance of lung ultrasound in the diagnosis of acute
respiratory failure: the BLUE protocol. Chest 134:117125
10. Copetti R, Soldati G, Copetti P (2008) Chest sonography: a useful tool to differentiate acute
cardiogenic pulmonary edema from acute respiratory distress syndrome. Cardiovasc
Ultrasound 6:16
11. Gajic O, Dara SI, Mendez JL, Adesanya AO, Festic E, Caples SM, Rana R, St Sauver JL,
Lymp JF, Afessa B, Hubmayr RD (2004) Ventilator-associated lung injury in patients without
acute lung injury at the onset of mechanical ventilation. Crit Care Med 32:18171824
12. Gajic O, Frutos-Vivar F, Esteban A, Hubmayr RD, Anzueto A (2005) Ventilator settings as a
risk factor for acute respiratory distress syndrome in mechanically ventilated patients. Intensive
Care Med 31:922926
13. Peek GJ, Elbourne D, Mugford M, Tiruvoipati R, Wilson A, Allen E, Clemens F, Firmin R,
Hardy P, Hibbert C, Jones N, Killer H, Thalanany M, Truesdale A (2010) Randomised controlled trial and parallel economic evaluation of conventional ventilatory support versus extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR). Health Technol
Assess 14:146
14. Mendler N, Heimisch W, Schad H (2000) Pulmonary function after biventricular bypass for
autologous lung oxygenation. Eur J Cardiothorac Surg 17:325330
15. Avalli L, Maggioni E, Sangalli F, Favini G, Formica F, Fumagalli R (2011) Percutaneous leftheart decompression during extracorporeal membrane oxygenation: an alternative to surgical
and transeptal venting in adult patients. ASAIO J 57:3840
16. Kawashima D, Gojo S, Nishimura T, Itoda Y, Kitahori K, Motomura N, Morota T, Murakami
A, Takamoto S, Kyo S, Ono M (2011) Left ventricular mechanical support with Impella provides more ventricular unloading in heart failure than extracorporeal membrane oxygenation.
ASAIO J 57:169176
[email protected]
388
D. Pasero et al.
17. Massetti M, Gaudino M, Saplacan V, Farina P (2013) From extracorporeal membrane oxygenation to ventricular assist device oxygenation without sternotomy. J Heart Lung Transplant
32:138139
18. Ricci D, Boffini M, Barbero C, El Qarra S, Marchetto G, Rinaldi M (2013) Minimally invasive
tricuspid valve surgery in patients at high risk. J Thorac Cardiovasc Surg. doi:10.1016/j.
jtcvs.2013.03.018. [pii: S0022-5223(13)00320-6] [Epub ahead of print]
19. Bekteshi E, Bell HJ, Haouzi A, El-Banayosy A, Haouzi P (2010) Control of breathing during
acute change in cardiac preload in a patient with partial cardiopulmonary bypass. Respir
Physiol Neurobiol 170:3743
20. Lloyd TC Jr (1990) Effect of increased left atrial pressure on breathing frequency in anesthetized dog. J Appl Physiol 69:19731980
21. Lloyd TC Jr (1984) Effect on breathing of acute pressure rise in pulmonary artery and right
ventricle. J Appl Physiol 57:110116
[email protected]
Neurological Monitoring During ECMO
34
Paolo Zanatta, Enrico Bosco, Alessandro Forti,

Elvio Polesel, and Carlo Sorbara
34.1
Introduction
Neurological monitoring during extracorporeal membrane oxygenation (ECMO) is

one of the most challenging tasks in an intensive care unit because of the complexity
of the clinical scenario and the instability of the patient, who is most often in a
comatose condition just before the onset of extracorporeal circulation.
In this context, neuromonitoring is an extension of clinical examination that is
often unfeasible because of multi-organ failure, sedation, and hypothermic
P. Zanatta (*)
Department of Anesthesia and Intensive Care, Unit of Cardiac Anesthesia
and Intensive Care, Treviso Regional Hospital, Treviso, Italy
Intraoperative and Intensive Care Neuromonitoring in Cardiac Surgery,
Anesthesia and Intensive Care Department, Treviso Regional Hospital,
Piazzale Ospedale 1, Treviso 31100, Italy
E. Bosco
Department of Anesthesia and Intensive Care, Unit of Neuro Surgery Anesthesia
and Intensive Care, Treviso Regional Hospital, Treviso, Italy
Intraoperative and Intensive Care Neuromonitoring in Neuro Surgery, Anesthesia and Intensive
Care Department, Treviso Regional Hospital, Piazzale Ospedale 1, Treviso 31100, Italy
A. Forti
Department of Anesthesia and Intensive Care, Unit of Cardiac Anesthesia and Intensive Care,
Treviso Regional Hospital, Treviso, Italy
Cardiac Anesthesia, Anesthesia and Intensive Care Department, Treviso Regional Hospital,
E. Polesel
Director of Cardiac Surgery Unit, Treviso Regional Hospital, Piazzale Ospedale 1,
Treviso 31100, Italy
[email protected]
389
390
P. Zanatta et al.
C. Sorbara
Director of the Department of Anesthesia and Intensive Care, Treviso Regional Hospital,
Treviso, Italy
Anesthesia and Intensive Care Department, Treviso Regional Hospital,
treatment. In particular, after the advent of therapeutic hypothermia, clinical evaluation has been called into question because of its higher rate of producing false positives. A multimodal neurophysiological strategy can overcome this limitation and
provide additional information on brain function, blood-flow velocity, and brain
oxygenation. This strategy allows targeting of not only the level of neuroprotection
but also the haemodynamic and respiratory parameters to be maintained during the
extracorporeal treatment.
Neurological monitoring can be extemporaneous or continuous based on available miniaturised technologies and the severity of the patients condition. Moreover,
it plays a critical role in the diagnosis and prognosis of the neurological
dysfunction.
34.2
The Value of a Multimodal Strategy
Because the pathogenesis of brain damage during ECMO can be multifactorial,

monitoring instruments should be used to obtain information on the status of
brain function, like EEG and somatosensory evoked potential. Circulatory
pathogenetic mechanisms that predispose the patient to functional damage
should also be checked, such as embolisation and hypoperfusion, through methods, such as transcranial Doppler (TCD), near-infrared spectroscopy (NIRS), or
jugular oxygen saturation (SjO2), and biochemical markers of brain injury [13]
(Fig. 34.1).
34.3
Neurophysiological Monitoring of Brain Function
Brain function can be monitored directly by EEG and SEPs, the spectrograms of
which correlate directly with CBF (cerebral blood flow) [4, 5]. A gradual reduction
in cerebral perfusion is associated initially with changes in synaptic transmission
and at the end with neuron death due to the impossibility of maintaining the electrochemical gradient of the cell membrane. Normally, CBF is approximately
5080 ml/100 g/min. Moderate hypoperfusion up to values of 30 ml/100 g/min is
well tolerated and does not cause neuron dysfunction; when the flow decreases
below the functional limit (25 ml/100 g/min), both the EEG and SEPs begin to
change. The functional limit is not time dependent, but the injury limit is. The
necrosis area increases with time beyond the injury limit.
EEG and SEPs disappear at flow values of about 1215 ml/100 g/min, although
some authors consider that cortical SEPs disappear at a flow that is 20 % less than
the flow necessary to bring about an isoelectric EEG [6].
[email protected]
34
391
Fig. 34.1 Continuous multimodal neuromonitoring during postcardiotomy ECMO; on the upper
side, the neurophysiological, NIRS, and TCD monitoring are highlighted
Between the functional and injury limit is the ischaemic penumbra: the brain tissue in this area is electrically silent and not functional but still vital [7]. The ischaemic penumbra is a dynamic area extending from the periphery of an infarcted
area, and its progression from functional damage to structural damage depends on
the timing and efficacy of therapy.
The ischaemic penumbra can evolve to functional recovery if the flow increases or
to necrosis if the ischaemia persists. Thus, the potential regression of the ischaemic
penumbra area introduces a basic concept for brain recovery that is directed at preventing or minimising a secondary lesion by improving DO2, reducing CMRO2 (brain
metabolism), and reducing reperfusion damage. There is no doubt that the duration
and size of the ischaemia influence the degree of brain damage. There are regional
differences in resistance to ischaemic damage that depend on metabolic activity, differing susceptibility to ischaemia, availability of collateral pathways, and capillary
density. A reduction in EEG and SEP amplitude, slowing of the EEG, and increased
latency of the SEPs indicate that the functional limit of CBF has been reached.
[email protected]
392
P. Zanatta et al.
These changes are extremely important during extracorporeal perfusion of the

patient because they provide an alarm signal that goes off before the brain lesion
becomes irreversible. If the proper steps are taken to increase the CBF (increasing
the pump rate or administering vasoconstrictors) or in general the DO2 and reduce
brain metabolism (increasing the depth of sedation or level of hypothermia), the
functional limit will not be reached.
34.3.1 EEG
Much of the literature agrees that gradual reduction in CBF produces an attenuation
in the EEG with amplitude reduction and slowing at a frequency expressed by a
reduction in alpha rhythm (, 813 Hz) and beta rhythm (, >14 Hz) and an increase
in theta rhythm (, 57 Hz) and delta rhythm (, 0.54 Hz). These changes can be
generalised (global ischaemia) or regional (focal ischaemia). The degree of ischaemia is associated with the severity of the electroencephalographic changes; the
EEG cannot evaluate the entire cerebral cortex and is less effective in distinguishing
changes in subcortical structures. In addition, the asymmetry between the hemispheres is considered significant in patients given stable sedation or anaesthesia to
induce electroencephalographic suppression.
The EEG changes may have latency from 10 s after a cardiac arrest [8] and up to
3 min after an embolic injury [9]. In general, the EEG has had a secondary role with
respect to the somatosensory evoked potentials in given information about the neurological prognosis; recently, some studies have pointed out the high predictive
value of an accurate neurological prognosis from a good and reactive EEG in the
acute phase of post-anoxic coma. A good EEG is considered a continuous pattern
without a period of suppression, while isoelectric or low voltage and burst suppression are considered EEG patterns associated with a negative outcome [10].
Moreover, one of the most important EEG patterns that can be detected is nonconvulsive status epilepticus; rapid recognition guarantees early treatment.
Fourier spectral analysis is an established method of quantitative analysis of the
EEG and is useful for providing data in tighter groups for easier interpretation and
early identification of brain ischaemia [11]. The data from the Fourier analysis can
be presented in the form of CSA (compressed spectral analysis) or DSA (modulateddensity spectral analysis); CSA and DSA are both graphic representations based on
finding a total reduction in the power of the tracing and/or an increase in the power
of the components in the - and -slow bands (Figs. 34.2 and 34.3).
34.3.2 Somatosensory Evoked Potentials (SEPs)

SEPs are used more than BAEPs (brainstem auditory evoked potentials) and VEPs
(visual evoked potentials) to find cerebral ischaemia because they provide a direct
correlation between function and perfusion of the territory of the middle cerebral
artery. They are less sensitive to pharmacological depression brought about by the
anaesthetic agents and are also easier to monitor because the cortical somatosensory
projections are located on the convexity of the hemispheres.
[email protected]
34
393
Fig. 34.2 TCD-EEG coupling during ECMO. Left side: a nonconvulsive status epilepticus at
EEG (lower) induced an increase of blood-flow velocity at TCD (upper). Right side: showers of
gas microemboli (TCD upper) induced a further decrease in amplitude of EEG signals (lower)
b
Fig. 34.3 Two possible neurophysiological scenarios of patients with good outcomes. Bilateral
somatosensory evoked potentials of normal amplitude and with both middle- and long-latency
waves; nonconvulsive status epilepticus on the upper side (a) and a low-voltage theta and delta
rhythm at the lower side (b)
[email protected]
394
P. Zanatta et al.
The cortical generators of N20 are located in the territory of the middle cerebral
artery. A flow reduction below the electrical compromise threshold produces desynchronisation of the activity of the cortical neurons and reduction in the number of
functional neurons with a reduction in amplitude.
Numerous studies propose using the N20 parietal amplitude as the diagnostic
criterion of cerebral hypoperfusion [5, 12]. It is believed that the latency of SEPs is
less sensitive with respect to amplitude in the case of reduced CBF because the
metabolic requirement of the white substance (axonal) is less than that of the grey
substance (neuronal bodies) [13]. Nonetheless, the CCT (central conduction time)
is always measured: this variable reflects electrical conduction between the spinal
medulla, the trunk, and the other subcortical hemispherical areas. The significant
values are a reduction in N20 amplitude of greater than 50 % and an increase in the
CCT of 1 ms [14] or 20 % relative to the baseline value [15].
Both anaesthesia and hypothermia induce a reduction of cerebral metabolism
during ECMO and influence brain function. Hypothermia and anaesthesia bring
about a gradual, symmetrical change in the EEG and, at temperatures below 30, in
SEP as well. Moderate and mild hypothermia (body temperature over 30 C) is
associated with a reduction in EEG amplitude and frequency. The EEG becomes
isoelectric at 2225 C [16].
The SEPs react to steady-state hypothermia with an increase in latency and
amplitudes of the peripheral and cortical potentials at temperatures over 30. Below
2830, the SEP amplitude reduces [17]. The N20 and P14 components of the SEP
can disappear between 1526 C and 1520 C nasopharyngeal temperature,
respectively [16, 17].
Most of the anaesthetic agents cause similar effects in the cortical EEG and
SEPs, with the exception of ketamine. These anaesthetic drugs initially cause an
increase in the speed of the EEG tracing, which consists of disappearance of the
-rhythm and appearance of a -rhythm, followed by gradual synchronisation and
postero-frontal slowing with subsequent appearance of - and -rhythms. Increasing
the dose of anaesthesia appears as a burst suppression pattern followed by a gradual
increase in the isoelectric period until the isoelectric pattern emerges.
Generally the effects of total intravenous sedation on the cortical SEPs, expressed
by increased latency and reduction in amplitude, are irrelevant. Opiates may cause
slowing of the EEG and the appearance of -waves at high amplitudes; the effect on
the SEPs, however, is negligible.
Finally, curarisation does not affect the neurophysiological electrical signals; on
the contrary, it may bring about an increase in SEP amplitude due to a better signalto-noise ratio caused by the removal of muscular artefacts.
In addition to standard clinical examination, SEPs improve the accuracy of neurological prognosis of comatose patients from multifactorial etiopathogenetics [18].
Regarding cardiac arrest and therapeutic hypothermia, it has been shown that a multidisciplinary approach composed of the Glasgow Coma Scale (GCS), patients
pupil light reactivity, corneal reflexes, serum neuron-specific enolase, and shortlatency somatosensory evoked potentials (N20/P25) improves the accuracy of neurological prognosis [19]. Indeed, the bilateral absence of the early cortical SEP (i.e.
N20/P25) has a high predictive value for adverse outcomes such as death or survival
in a vegetative state. However, the presence of N20/P25 is not sensitive enough to
[email protected]
34
395
predict a good neurological outcome [18]. It is well established that only eventrelated evoked potentials (i.e. P300) and the presence of middle-latency cortical
somatosensory evoked potentials (MLCEP) strongly correlate with a favourable
neurological prognosis in patients affected by severe brain injury [20, 21].
The appearance of MLCEP can be produced by an electrical stimulation on the
median nerve, because the SEP responses can reflect a more integrated cerebral
processing of pain that includes primary and secondary somatosensory cortices: the
insular, anterior cingulate, and prefrontal cortices [22].
Pain-related MLCEP may be a measure to predict good neurological outcomes in
comatose patients, as demonstrated by previous studies without a pain-related
method [23]. Given that only high-intensity stimulation would generate MLCEP, this
method allows the detection of the quiescent brain network in the ischaemic penumbra and represents a dynamic test of brain availability; it can be considered a sort of
neurophysiological GCS with which the clinicians can evaluate the brains reactivity
to painful stimulation [24]. Interestingly, this method might also be a useful tool to
evaluate brain connectivity in patients without normal EEG patterns (e.g. the NCSE).
SEP is a stable signal, is reproducible, and is more resistant to temperature
changes, anaesthesia changes, and electrical interference. It also gives information
on subcortical structures. SEP provides information on the function of the somatosensory area and, indirectly, on the rest of the parenchyma perfused by the middle
cerebral artery. Moreover, exploring neurophysiological reactivity with the painful
stimulation gives the physician the chance to gain information on brain function
with an isoelectric EEG trace.
The neurophysiological evaluation also provides the possibility of testing other
brain functions such as motor, acoustic, and visual, which can contribute to a better
definition of the neurological prognosis.
34.3.3 Continuous Brain Function Monitoring

Simultaneous and continuous monitoring of EEG and SEPs has considerable value
because their modifications rapidly direct the physician towards more aggressive
clinical management [25]. For example, continuous EEG is useful for the diagnosis
and treatment of nonconvulsive seizures and status epilepticus, whereas continuous
SEPs are more able to indicate the occurrence of loss of somatosensory integrity
and brain connectivity in short, middle, and long latency. Continuous EEG/SEP
monitoring is unfortunately still lacking in the armamentarium of anaesthesiologists
and intensive care physicians, although the majority of the interventional and anaesthesiological procedures performed have an impact on the nervous system.
34.4
Neurosonology Monitoring of Cerebral Blood-Flow

Velocity and Microembolic Signals
Transcranial Doppler is the only available method for continuous evaluation of

changes in brain haemodynamics in real time, allowing supplementary detection
of brain emboli. A sudden drop in peak velocity or loss of the spectrum indicates
[email protected]
396
P. Zanatta et al.
conditions of altered perfusion, with a direct causality relationship. An increase in

cerebrovascular resistance during ECMO detected with a reduction in diastolic velocity of brain blood-flow velocity may indicate increased intracranial pressure [26, 27].
Ultrasounds are very sensitive in identifying the presence of emboli in the vascular lumen. Patients treated with ECMO can suffer from brain and systemic gas
microembolism because air can enter the circuit through the central venous lines;
this phenomenon occurs at high flow when all blood is shunting from the lung [28].
The current extracorporeal oxygenator and mininvasive circuits are less efficient in
removing air microbubbles from the extracorporeal circulation, as also experienced
during extracorporeal circulation during cardiac surgery [29]. For this reason, maximum care is required in setting and managing the infusion lines during ECMO. The
interposition of an air filter device between the intravenous catheter and the infusion
lines can prevent this dangerous occurrence.
34.5
Monitoring of Cerebral Metabolism
Brain oxygen saturation can be carried out with either invasive or non-invasive
methods, both of which call on continuous infrared spectroscopy; the difference in
the absorption spectrum between total haemoglobin and oxygenated haemoglobin
provides an estimate of oxygen saturation. With the invasive approach, this
measurement is obtained from a sensor located at the end of a catheter positioned in
the jugular bulb. With the non-invasive method, an infrared source is positioned at
the surface of two adhesive patches placed on the scalp. The sensor proximal to the
source picks up the infrared light reflected primarily from the skin and bone, while
the more distal sensor receives the light reflected from a sample of brain tissue measuring 1 cm3 [30, 31]. Because the blood in the brain tissue is 75 % venous, the
differential signal provides an estimate of the venous amount of brain oxygen saturation. NIRS has a possible limitation in old patients with brain atrophy because of
the low-resolution depth.
While NIRS is bilateral on the frontal lobe, the SjO2 is monolateral, and it is still
an open question to identify the dominant internal jugular vein (which in most individuals is the right jugular). Among the methods proposed for identifying this is the
analysis of the size of the jugular foramen lacerum by a CT scan. Moreover, cerebral
venous dominance can be obtained by evaluating the highest decrement in the diastolic velocity of the brain blood flow caused by selective compression of the right
and left jugular veins.
The metabolic methods (NIRS and SjO2) for cerebral oxygen saturation reflect
the balance between brain oxygen availability and demand. Brain oxygen availability depends on CBF and arterial oxygen level. Because the uncoupling
between cerebral oxygen demand and supply is one of the causes of neurological
damage, cerebral oxygen saturation measurement is an essential tool for evaluating proper perfusion of the brain tissue; this methodology is also very feasible
during ECMO. Brain ischaemia conditions may come about from the combined
presence of the following situations: hypotension below the self-regulation limit,
[email protected]
34
397
anaemia, arterial desaturation, low PaCO2 levels, hyperthermia, and cerebral

vasoconstriction.
Because the absolute values for regional cerebral oxygen saturation are influenced by many variables, the normal range has not been defined; it is viewed as a
cut-off level of 40 % or a 25 % change from baseline values [32]. The finding of
cerebral desaturation of less than 40 % with NIRS [33] or less than 50 % with SjO2
[34] is associated with brain ischaemia. A normal SjO2 does not guarantee that there
are no regional cerebral ischaemias, but a low SjO2 definitely indicates global ischaemia, focal ischaemia, or both.
34.6
Conclusions
Neurological monitoring during ECMO allows for maintaining the brains haemostasis. Its functional integrity is vital to patient outcomes and quality of life. EEG
and SEP are the most informative neurophysiological tests, and they represent the
natural extensions of the neurological clinical evaluation; both have a major prognostic role in acute neurological dysfunction. Their combined use is a unique example of dynamic brain function monitoring.
Transcranial Doppler visualised brain blood-flow velocity and brain vascular
resistance provide indirect information on intracranial pressure. TCD is also important in detecting the possible risk of microembolic injury during total extracorporeal
supply. Metabolic methods like NIRS and SjO2 give the chance to monitor the oxygen extraction, targeting the systemic blood flow, oxygenation, and systemic vascular resistance.
The complexity of the ECMO patient and the capabilities of current technologies
often make it unfeasible to evaluate brain function, blood flow, and oxygenation
simultaneously. In current practice, it is more feasible to perform continuous monitoring with NIRS, while SEP, EEG, and TCD should, in our opinion, be performed
at fixed intervals or when the patients clinical condition changes.
References
1. Zanatta P, Messerotti Benvenuti S, Bosco E, Baldanzi F, Palomba D, Valfr C (2011)
Multimodal brain monitoring reduces major neurologic complications in cardiac surgery.
J Cardiothorac Vasc Anesth 25(6):10761085
2. Edmonds HL Jr (2002) Multimodality neurophysiologic monitoring for cardiac surgery. Heart
Surg Forum 5:225228
3. Luyt CE, Landivier A, Leprince P, Bernard M, Pavie A, Chastre J, Combes A (2012) Usefulness
of cardiac biomarkers to predict cardiac recovery in patients on extracorporeal membrane oxygenation support for refractory cardiogenic shock. J Crit Care 27(5):524.e7524.e14
4. Sundt TH Jr, Sharbrough FW, Plepgras DG et al (1981) Correlation of cerebral blood flow and
electroencephalographic changes during carotid endarterectomy with results of surgery and
hemodynamics of cerebral ischemia. Mayo Clin Proc 56:533543
5. Florence G, Guerit JM, Gueguen B (2004) Electroencephalography and somatosensory evoked
potentials to prevent cerebral ischemia in the operating room. Neurophysiol Clin 34:1732
[email protected]
398
P. Zanatta et al.
6. Prior PF (1985) EEG monitoring and evoked potentials in brain ischemia. Br J Anaesth
57:6381
7. Astrup J (1982) Energy requiring cell functions in the ischemic brain. J Neurosurg 56:482
8. de Vries JW, Bakker PF, Visser GH, Diephuis JC, van Huffelen AC (1998) Changes in cerebral
oxygen uptake and cerebral electrical activity during defibrillation threshold testing. Anesth
Analg 87(1):1620
9. McGrail KM (1996) Intraoperative use of electroencephalography as an assessment of cerebral blood flow. Neurosurg Clin N Am 7(4):685692, Review
10. Cloostermans MC, van Meulen FB, Eertman CJ, Hom HW, van Putten MJ (2012) Continuous
electroencephalography monitoring for early prediction of neurological outcome in postanoxic
patients after cardiac arrest: a prospective cohort study. Crit Care Med 40(10):28672875
11. Isley MR, Edmonds HL Jr, Stecker M, American Society of Neurophysiological Monitoring
(2009) Guidelines for intraoperative neuromonitoring using raw (analog or digital waveforms)
and quantitative electroencephalography: a position statement by the American Society of
Neurophysiological Monitoring. J Clin Monit Comput 23(6):369390
12. Horsch S, De Vleeschauwer P, Ktenidis K (1990) Intraoperative assessment of cerebral ischemia during carotid surgery. J Cardiovasc Surg 31:599602
13. Prior P (1996) The rationale and utility of neurophysiological investigations in clinical monitoring for brain and spinal cord ischaemia during surgery and intensive care. Comput Methods
Programs Biomed 51(12):1327, Review
14. Guerit JM, Witdoeckt C, de Tourtchaninoff M et al (1997) Somatosensory evoked potential
monitoring in carotid surgery. I. Relationships between qualitative sep alterations and intraoperative events. Electroencephalogr Clin Neurophysiol 104:459469
15. Thiel A, Russ W, Zeiler D et al (1990) Transcranial Doppler sonography and somatosensory
evoked potential monitoring in carotid surgery. Eur J Vasc Surg 4:597602
16. Kochs E (1995) Electrophysiological monitoring and mild hypothermia. J Neurosurg
Anesthesiol 7:222228
17. Stecker MM, Cheung AT, Pochettino A, Kent GP, Patterson T, Weiss SJ, Bavaria JE (2001)
Deep hypothermic circulatory arrest: I. Effects of cooling on electroencephalogram and
evoked potentials. Ann Thorac Surg 71(1):1421
18. Robinson LR, Micklesen PJ, Tirschwell DL, Lew HL (2003) Predictive value of somatosensory evoked potentials for awakening from coma. Crit Care Med 31:960967
19. Oddo M, Rossetti AO (2011) Predicting neurological outcome after cardiac arrest. Curr Opin
Crit Care 17(3):254259
20. Daltrozzo J, Wioland N, Mutschler V, Kotchoubey B (2007) Predicting coma and other low
responsive patients outcome using event-related brain potentials: a meta-analysis. Clin
Neurophysiol 118:606614
21. Zhang Y, Su YY, Ye H, Xiao SY, Chen WB, Zhao JW (2011) Predicting comatose patients with
acute stroke outcome using middle-latency somatosensory evoked potentials. Clin
Neurophysiol 122:16451649
22. Zanatta P, Messerotti Benvenuti S, Bosco E, Baldanzi F, Longo C, Palomba D, Salandin V,
Sorbara C (2011) Intraoperative neurophysiological monitoring of the afferent pain pathway in
cardiac surgery patients. Clin Neurophysiol 122:20932099
23. Madl C, Kramer L, Domanovits H, Woolard RH, Gervais H, Gendo A, Eisenhuber E, Grimm
G, Sterz F (2000) Improved outcome prediction in unconscious cardiac arrest survivors with
sensory evoked potentials compared with clinical assessment. Crit Care Med 28(3):721726
24. Farag E, Abd-Elsayed A, Manno EM (2012) Sensory evoked potentials and the search for the
Holy Grail method to predict the outcome after hypoxic-ischemic coma. Minerva Anestesiol
78(7):741742
25. Bosco E, Marton E, Feletti A, Scarpa B, Longatti P, Zanatta P, Giorgi E, Sorbara C (2011)
Dynamic monitors of brain function: a new target in neurointensive care unit. Crit Care
15(4):R170
[email protected]
34
399
26. Newell DW, Aaslid R (1992) Transcranial Doppler clinical and experimental uses. Cerebrovasc
Brain Metab Rev 4:122143
27. Burrows FA (1993) Transcranial Doppler monitoring of cerebral perfusion during cardiopulmonary bypass. Ann Thorac Surg 56:14821484
28. Zanatta P, Forti A, Bosco E, Salvador L, Borsato M, Baldanzi F, Longo C, Sorbara C, Longatti
P, Valfr C (2010) Microembolic signals and strategy to prevent gas embolism during extracorporeal membrane oxygenation. J Cardiothorac Surg 5:5
29. Zanatta P, Forti A, Minniti G, Comin A, Mazzarolo AP, Chilufya M, Baldanzi F, Bosco E,
Sorbara C, Polesel E (2013) Brain emboli distribution and differentiation during cardiopulmonary bypass. J Cardiothorac Vasc Anesth 27(5):865875
30. Cui W, Kumar C, Chance B (1991) Experimental study of migration depth for the photons
measured at sample surface. Proc SPIE 1431:180191
31. Edmonds HL (2005) Multimodality neuromonitoring for perioperative brain protection. Semin
Anesth Perioper Med Pain 24:186194
32. Daubeney PEF, Pilkington SN, Janke E et al (1996) Cerebral oxygenation measured by nearinfrared spectroscopy: comparison with jugular bulb oximetry. Ann Thorac Surg 61:930934
33. Yao FF, Chia-Chih A (2004) Cerebral oxygen desaturation is associated with early postoperative neuropsychological dysfunction in patients undergoing cardiac surgery. J Cardiothorac
Vasc Anesth 18(5):552558
34. Kadoi Y, Saito S, Goto F et al (2001) Decrease in jugular venous oxygen saturation during
normothermic cardiopulmonary bypass predicts short-term postoperative neurologic dysfunction in elderly patients. J Am Coll Cardiol 38(5):14501455
[email protected]
Monitoring the ECMO Patient:

The Extracorporeal Circuit
35
Stefano Isgr, Francesco Mojoli, and Leonello Avalli
35.1 Introduction
During ECMO, extracorporeal circuit-related adverse events must be carefully
monitored, being acute ECMO failure a potentially life-threatening event. ECMO
circuit failure requires emergency change-out [13], which is possible only by temporarily stopping the treatment, thus exposing patients to further harm [4, 5].
Extracorporeal circuit is composed of cannulae, tubing, a pump, and an artificial
membrane lung (ML). Each part of the circuit can break or fail, thus potentially
exposing patient to serious side effects. In this chapter we will review major ECMO
technical complications and monitoring methods. For a full explanation of different
ECMO techniques and materials, we remand to specific chapters in this book.
S. Isgr, MD (*)
Urgency and Emergency Department, San Gerardo Hospital,
F. Mojoli, MD
L. Avalli, MD
[email protected]
401
402
S. Isgr et al.
35.2 ECMO Failure

ECMO circuit failure, especially during prolonged use, is not an uncommon occurrence in ECMO centres ICUs [3, 6]. Although every part of the circuit may break
or fail, critical parts of the circuit are certainly ML membranes and circuit/pump
coupling sites.
35.2.1 ML Failure

Typically modern ECMO treatment fails due to critical narrowing or occlusion of
artificial exchanging sites (i.e., embolisms or thrombosis). As ECMO technique
exposes blood to contact on artificial surfaces, this interaction leads to the recognition of foreign material surfaces with protein body inflammatory and coagulation
system activation [3, 7]. Depending on the charge and composition of the exposed
surface, protein deposition and denaturation, as well as thrombin production, occur.
Parenteral continuous anticoagulation is thus required to prevent and delay coagulation system activation, being difficult to find the equilibrium between excessive and
insufficient anticoagulation (i.e., bleeding vs. circuit or patient thrombosis) [7].
When anticoagulation is not optimal, circuit coagulation activation occurs, precipitating consumption coagulopathy with blood coagulation factor level drop
(especially fibrinogen and platelets), fibrinolytic system hyperactivation, and bleeding (Fig.35.1) [7].
650
150
600
550
125
500
450
100
400
350
75
300
250
PLT (U*10^3)
200
50
150
100
25
50
Circuit day
0
50
% DELTA
dimers
FNG (Mg/dL)
0 1 1 1 2 2 2 3 33 4 4 4 5 5 5 6 6 6 7 70 1 112 22 3
Fig. 35.1 Time course of platelets, fibrinogen, D-dimer % change from baseline between ML
change-outs (black vertical lines) during a prolonged VV ECMO treatment, showing circuit
coagulation activation
[email protected]
35 Monitoring the ECMO Patient: The Extracorporeal Circuit
403
A significant technological improvement in MLs has been achieved in recent

years, with the transition from silicon non-microporous membrane MLs to microporous hollow-fiber MLs. The latter allows highly efficient gas exchange, reduction
in transmembrane pressures, and low priming volume, but plasma leakage through
microporous fibers frequently occurs, reducing gas transfer rate. More recently
polymethylpentene hollow-fiber MLs have eliminated this problem [4], allowing
more longer treatments, reducing the need for circuit change-out. Anyway, even
with polymethylpentene MLs, progressive clot formation with partial to total occlusion of blood path and loss of gas-exchanging surface is still an unresolved problem
(Fig.35.2) [2, 6].
35.2.2 Circuit/Pump Failure

Though more resistant materials and more biologically compatible inner circuit
coatings have been developed, hemocompatibility is still limited, and careful monitoring of the ECMO system is required. Each part of the circuit must be inspected
frequently in search of blood clots and deformities of tubing, especially where
blood flow is more turbulent as at fittings, stopcocks, and cannulae. Expected complications may be different depending on the pump technology employed and on
patient population (adult vs. pediatric); the most popular technologies are the roller
pump, actually more diffused for children [5], and the centrifugal pump. The overall
complication rate is not significantly different [8], with a significative difference in
hemolysis (but not in in-hospital survival) in favor of roller pumps among children
[5]. Roller pump specific complications are raceway or circuit break and, consequently, air embolism; centrifugal pumps instead are specifically exposed to pump
head/pump engine decoupling, consequently stopping treatment and thrombosis.
Alternative technologies such as nonocclusive and miniaturized pumps have been
Fig. 35.2 Black arrows

showing clots in the pump
head after a circuit
change-out
[email protected]
404
S. Isgr et al.
developed[8]. Some centres use two pumps/MLs running in parallel: in case of

failure of an ECMO system, the second ECMO allows a full support to the patient.
Other centers have a primed ML circuit always ready to be implemented [8]. Ideally
whole circuit or ML change-out should be performed electively to decrease morbidity due to circulatory lack of assistance (VA ECMO) or oxygenation/CO2 removal
(VV ECMO) during change-out procedure. Up to now circuits are changed
according to clinical nonstandardized criteria. Major clinical criteria may be
summarizedin:
1. Circuit coagulation activation
2. Gross thrombosis of circuit components
3. Technical failure of the circuit/ML
35.3 Monitoring Techniques

35.3.1 Blood Flow and RPM
In order to monitor energy/mechanical failure of the ECMO pump, rotation per
minute (RPM, available both for roller and centrifugal pumps) is always continuously displayed and alarmed. Monitoring RPM is of particular importance on roller
pumps, RPM being directly related to blood flow (when occlusion is accurately set
and circuit and raceway are intact); when a centrifugal pump is employed, for a set
RPM, blood flow is mainly driven by flow resistance of the ECMO circuit (cannulae, tubing, ML) and by patient fluid status: therefore, in case of low-volume state
and high RPM, blood flow may abruptly fall, eventually leading to suctioning of
venous wall and/or ECMO treatment interruption. This occurrence often leads to
bloodpump head decoupling and requires a special emergency procedure to restart
ECMO. Close flow monitoring and alarming are thus needed. According to different technologies, companies developed specially modified fittings where flowmeters may be applied on the circuit.
35.3.2 Anticoagulation
As seen above ECMO circuits still require adjunctive anticoagulation to avoid circuit
component thrombosis. Adequate anticoagulation level during ECMO is the balance
between circuit/patient thrombosis and patient bleeding; finding the optimal level
could be challenging as it is not absolute and varies according to circuit and patient
status [9] (i.e., preexisting coagulation disorders, ECMO circuit and treatment duration, sepsis). As bleeding and thrombosis are still the principal causes of morbidity
and mortality during ECMO [7], strict anticoagulation monitoring is essential.
ECMO anticoagulation is generally obtained with unfractionated heparin (UFH),
which is the drug of choice because of a favorable pharmacological profile: in fact
it has a parenteral, fast-onset, rapidly reversible effect and is an economical and a
[email protected]
405
widely diffused drug. Despite this, heparin is not ideal, as it has a non-predictable
anticoagulant effect, thus requiring close monitoring and frequent variations in dosage. Moreover, anticoagulant effect could not be accurately monitored, and consequently each test requires careful interpretation (see below), sometimes making
clinical management difficult [7, 10]. Unfractionated heparin acts by binding with
antithrombin III (ATIII); hence, as hepatic ATIII production may be impaired, ATIII
level must be frequently monitored and kept constant. Finally, it is associated with
heparin-induced thrombocytopenia (HIT) [11, 12].
Heparin anticoagulation can be monitored at bedside with activated clotting time
(ACT), point-of-care activated partial thromboplastin time (aPTT), and viscoelastic
tests (VET) or can be monitored with routine (aPTT, D-dimers) or special (heparin
plasma level antiXa activity) coagulation lab tests.
aPTT is the standard test to measure heparin anticoagulation; it measures
plasma coagulation pathway in-vitro at 37C temperature. Accordingly, interactions between platelets and other blood cells with soluble factors are not tested,
so aPTT test reflects only a small fraction of the thrombin produced during the
coagulation process [13].
ACT [14] is a widely diffused test among ECMO centers [9], being classically
employed in the cardiopulmonary bypass setting, where high dose of heparin and
point-of-care test are required. ACT can also be used for lower dose of heparin: a
sample of fresh whole blood is inserted into a warmed cuvette where coagulation is
activated by a negatively charged activator (celite or kaolin). ACT measures the
seconds needed for the blood to start clotting. Anticoagulation is kept to obtain a
value between 180 and 220s. It is affected by sepsis, D-dimers,platelet dysfunction, thrombocytopenia, hypofibrinogenemia, and hypothermia.
Correlation between ACT and aPTT is poor [9, 14]; therefore, in clinical decision making, factors influencing both tests must be considered.
Some centers employ viscoelastic test as an adjunct to anticoagulant therapy
monitoring, being a test extremely sensitive of heparin activity variation.
Nevertheless, experience in this application of thromboelastography/thromboelastometry is still limited [7].
This is the anticoagulation monitoring protocol that we follow in our centre:
1. Platelets
(a) Platelet levels are checked every 1224h.
(b) Platelets are administered when lower than 50,000units/mm3.
2. ATIII
(a) Plasmatic ATIII level checked daily until stable value with (or without)
replacement and then every 23days.
(b) Replace as needed to obtain a value >100%. Dosage is decided according to
the formula: unit to be administered=weight (kg)[ATIII desired (%)ATIII
measured (%)]/1.4.
3. ACT
(a) Checked after 1 and 3h after heparin boluses and then every 68h. Target
result is decided by integration of ACT with other coagulation tests.
[email protected]
406
S. Isgr et al.
4. Standard coagulation tests

(a) aPTT, fibrinogen, and D-dimers checked three times a day. aPTT value is
kept between 50 and 60s (ideally aPTT 1.52 times the patients baseline),
and FNG above 200mg/dL.
(b) Ionized calcium is obtained with ABG analysis and is kept within physiological range.
5. Thromboelastography
(a) Upon clinical decision and simultaneously with standard coagulation tests to
address baseline coagulation (whole kaolin+heparinase TEG tracing shape)
and R time value in the kaolin tracing.
Case series with new parenteral direct thrombin inhibitors argatroban [1517]
and bivalirudin [18] have been published, but experience with these drugs during
ECMO is still limited.
35.3.3 Pressures
Circuit pressure monitoring may be mounted onto the ECMO circuit at different
sites, representing an easy and effective alarming and information system about
both roller pump and centrifugal pump ECMO performances. To monitor pressures
and to draw blood, manufacturers provide preassembled or spare fittings with a side
port (Luer-lock connectors). Monitoring sites vary through institutions according to
the age of the ECMO patient [4] and to local practices and protocols. Commonly
used monitored sites are:
1. Two sites on either side of the ML (pre-ML and post-ML), providing information
about variation in pressure drop and blood flow resistances through the oxygenator.
An increase in pressure drop suggests ML thrombosis (Fig.35.3) [2]. Resistance of
blood flowing through the oxygenator may be computed according to the following
formula: R=(pin-pout)/BF, where pressures are expressed in millimeters of mercury and BF in liters per minute (R=mmHg/L/min). Each manufacturer provides
product-specific performance charts comparing resistances at different blood flows.
2. Inlet venous line, giving negative pressure readings of blood suctioning from the
patient, thus monitoring oversuction (centrifugal pump ECMO) [4]. Excessive
negative pressure applied to the venous system may actually result in bubble
formation (cavitation) and hemolysis [5], suctioning of vena cavaright atrium
wall (endothelial damage), and air embolism [3]. To avoid this occurrence,
apump speed control system is employed on roller pump ECMO, servo-regulated
by a transducer mounted onto the aspiration line.
3. Pressure measured at the patient inlet, system pressure (roller pump ECMO).
35.3.4 ML Gas Transfer

The primary task of artificial membrane lungs (MLs) is the total or partial support
of the native lung function by O2 transfer to and CO2 removal from blood.
TheAssociation for the Advancement of Medical Instrumentation (AAMI)
[email protected]
407

120
1,200
PIN
DELTA
100
1,000
RPM/BF
80
Pressure (mmHg)
800
60
40
600
20
400
0
18
24
30
48
200
20
40
Circuit working hours
Fig. 35.3 Time course of circuit pressures (Pin pressure measured before centrifugal pump,
DELTA difference of pressures measured before and after ML blood line, R resistances across ML,
RPM/BF centrifugal pump rotations per minute/ECMO blood flow) showing a case of urgent ML
change-out due to a rapid increase in resistances across ML. D-Dimers and other coagulation index
of circuit activation were all in the normal reference range
standards defined the minimum requirements for clinical use of MLs: 45ml of O2
transferred and 38ml of CO2 removed per liter of blood flowing through the ML.
This means that MLs are able to turn normal venous blood into normal arterial
blood in terms of both O2 and CO2 content, whatever the blood flowing through (up
to the maximum flow rate of single devices).
Anyway, the performance of ML can deteriorate over time, and sometimes it
happens abruptly. As in the native lungs, abnormalities of gas exchange in the artificial membrane lungs can be described and quantified by Rileys three-compartment
model: ideal, shunt, and dead space compartment [19]. Displacement of hollow
fibers in MLs is studied in order to maximize gas exchange by optimal matching of
gas and blood flows. Anyway, in hollow-fiber MLs the ideal compartment is always
associated with a variable shunt compartment (perfused but not ventilated) and dead
space compartment (ventilated but not perfused). When the shunt compartment was
evaluated during cardiopulmonary bypass in the cardiac surgery setting, it ranged
1030% in modern hollow-fiber MLs [2022]. In Figs.35.4 and 35.5 shunt and
dead space compartments in two cases of extracorporeal lung assistance with a high
sweep gas flow (10l/min) are shown: early after the start of the extracorporeal
[email protected]
408
30
550
450
400
15
10
20
p0 /FO (mmHg)
500
25
Shunt (%)
Fig. 35.4Monitoring
oxygen transfer in the
artificial membrane lung:
time course of shunt and pO2/
FO2 ratio. Shunt is already
present at the very beginning
of the extracorporeal
assistance and increases with
time, indicating progressive
loss of O2 transfer
performance in the ML. In
this particular case, pO2/FO2
ratio mirrors almost perfectly
the behavior of shunt
S. Isgr et al.
350
Shunt
po2/F02
10
300
55
2.00
50
1.75
45
1.50
40
1.25
35
1.00
30
25
V'd
V'/Q
5
V '/Q
Fig. 35.5 Monitoring carbon

dioxide removal in the
artificial membrane lung:
time course of dead space
and V/Q ratio. Dead space is
already present at the very
beginning of the
extracorporeal assistance and
increases with time,
indicating progressive loss of
CO2 removal performance in
the ML. In this particular
case, V/Q ratio mirrors
almost perfectly the behavior
of dead space
Dead space (%)
Time (days)
0.75
0.50
Time (days)
assistance, shunt and dead space were already present and accounted for 15% and
35% of blood and gas flowing through the ML, respectively.
The further loss of ML performance during prolonged use can be due to both
fluid accumulation in the hollow fiber lumen and clot formation on the blood side of
the hollow fiber surface.
The availability of polymethylpentene (PMP) hollow-fiber MLs has dramatically
reduced the occurrence of significant plasma leakage [4]. Anyway, slow filtration of
plasma water through the PMP microporous membrane is still present, and sometimes fluid accumulation in the fiber lumen occurs. When the gas flow is impeded
by fluid accumulation in a significant proportion of fibers, ML performance accordingly decreases due to a shunt effect. In these cases, patency of hollow fibers can
be reestablished by gently purging them with a brief period of high sweep gas flow.
[email protected]
409
To avoid this problem, recently developed devices for extracorporeal lung support
are provided with a periodic purge function.
Despite the use of biocompatible inner circuit coating and systemic anticoagulation, thrombotic and cellular deposits on the outward surface of fibers occur, with a
rate that is not easily predictable [23]. These deposits increase the resistance to
blood flow and the diffusion path of the gas exchanger fibers, leading both to shunt
and to dead space formation. Actually, pseudomembranous deposits whose thickness can almost reach the thickness of the hollow-fiber wall may completely
impede gas exchange while blood flow is still maintained (shunt effect), whereas
extensive clot formation around fibers may completely stop blood flow while gas is
still flowing in fibers lumen (dead space effect).
Monitoring the ML gas exchange function is therefore mandatory in prolonged
respiratory and/or circulatory extracorporeal assistances: patients metabolic
demands, especially in case of large/awake/critically ill subjects, could eventually
not be assured by an ML with a significant loss of performance. Monitoring ML may
accordingly help in identifying the timing for a scheduled, elective replacement of
the extracorporeal circuit, thus avoiding emergency and dangerous procedures.
Finally, monitoring the ML provides useful information for the management of the
extracorporeal assistance: the knowledge of the actual contribution of the ML to gas
exchange may help in identifying and correctly treating the cause of blood gas abnormalities and, in the venovenous ECMO setting, may guide the weaning procedure.
Both O2 transfer and CO2 removal should be monitored, because both the functions can deteriorate with time in MLs and, to note, not always with the same rate.
An example of this is Fig. 24.3: the first displayed ML was replaced because of
progressive and significant deterioration of CO2 removal while O2 transfer was still
satisfactory; in the second ML, decay of CO2 removal and of O2 transfer went on
instead in parallel.
In clinical practice, the simplest way to monitor oxygen transfer is to measure the
O2 partial pressure (pO2) in the blood coming out the device (usually at the same
outlet site of Pout measurement) in relation to the O2 fraction of sweep gas flow
(FO2): this way, the pO2/FO2 ratio of the ML can be calculated. Alternatively, once
two simultaneous blood samples at the ML inlet and outlet are obtained, the percentage of shunt can be formally computed as shown in Fig.35.6.
Generally, pO2/FO2 and shunt % provide the same information on ML status: in
Fig. 35.4, time course of these parameters is almost specular, the progressive
decrease of pO2/FO2 corresponding to a simultaneous increase of shunt % in the
ML. But this is not always the case, because pO2/FO2 depends also on inlet blood O2
content. An example of abrupt and significant drops of pO2/FO2 not associated with
increased shunt % is displayed in Fig. 24.4. High metabolic demands, as during
shivering, may favor low venous (inlet) blood O2 saturation and low pO2/FO2 despite
stable shunt %, i.e., without any evidence of ML loss of performance.
In order to counteract the decay of CO2 removal capacity in the ML, the sweep
gas flow (SGF) should often be progressively increased up to the maximum value
indicated by the manufacturer (usually 1012l/min). Accordingly, the ventilation/
perfusion (V/Q) ratio of the ML (i.e., the sweep gas to blood flow ratio)
[email protected]
410
S. Isgr et al.
Qs
Shunt membrane lung

(CcapO2 COUTO2)
QCEC
(CcapO2 CINO2)
Ccap O2 =0.0031 PAO2 + 1.39 Hb SatTC

PAO2 = (PATM PH2O) FiO2 PaCO2/RQ
SatTC = X (HbCO + MetHb)
Fig. 35.6 Shunt definition: CEC means extracorporeal circulation, C means concentration,
CcapO2 means capillary (ideal) O2 concentration, and RQ means respiratory quotient, x is 1, 0.99,
or 0.98 if PaO2 is above 250mmHg, between 150 and 250, or below 150, respectively
progressively increases: this parameter can be used in clinical practice to easily and
continuously (even if grossly) monitor the CO2 removal function. To note, CO2
removal depends on SGF in a nonlinear manner [24]: when the gas flow is progressively increased, the gain in CO2 removal progressively decreases and, at gas flows
greater than 10l/min, almost vanishes (see also Figs. 24.1 and 24.2 for the relationship between SGF and CO2 removal). Figure35.5 shows the time course of the ML
V/Q ratio during prolonged extracorporeal support: the increase of dead space in
the ML is coupled with and counterbalanced by a corresponding increase of V/Q.
Anyway, it should be underlined that V/Q ratio is directly influenced by an ECMO
setting the sweep gas flow that in turn depends also on clinical targets (arterial
blood CO2 and pH) and on patients CO2 production. This means that the V/Q ratio
can change for reasons other than impaired CO2 removal in the ML.
Alternatively, formal computation of dead space % in the ML can be obtained as
follows:

Dead space % = 100 ( pCO 2 , blood pCO 2 , gas ) / pCO 2 , blood
where pCO2,blood and pCO2,gas are the CO2 partial pressures in the ML blood and
gas outlet, respectively. Recently developed devices for extracorporeal lung support
are provided with an integrated capnometer at the gas outlet port, thus permitting
continuous monitoring of pCO2, gas and extracorporeal CO2 removal.
In conclusion, for monitoring gas exchange in the MLs, pO2/FO2 and V/Q ratio
are easy to obtain and can be used for trends, whereas shunt and dead space computations are more cumbersome but reliable parameters. At our institutions, we measure shunt and dead space of the ML at least daily and whenever clinics dictate the
need for an accurate assessment of the ML performance.
References
1. Da Broi U, Adami V, Falasca E etal (2006) A new oxygenator change-out system and
procedure [Internet]. Perfusion 21:297303. Available from: http://prf.sagepub.com/cgi/
doi/10.1177/0267659106074771
[email protected]
411
2. Schaadt J (1999) Oxygenator thrombosis: an international phenomenon. Perfusion 14:425435

3. Annich GM (2012) ECMO: extracorporeal cardiopulmonary support in critical care, Red
book. Extracorporeal Life Support Organization. Ann Arbor, Michigan, USA
4. Palanzo D, Qiu F, Baer L etal (2010) Evolution of the extracorporeal life support circuitry.
Artif Organs 34:869873
5. Barrett CS, Jaggers JJ, Cook EF etal (2013) Pediatric ECMO outcomes: comparison of centrifugal versus roller blood pumps using propensity score matching. ASAIO J 59:145151
6. Cornelissen CG, Dietrich M, Gromann K etal (2013) Fibronectin coating of oxygenator membranes enhances endothelial cell attachment. Biomed Eng Online 12:7. Available from: http://
eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=pubmed&id=23356939&retmode=
ref&cmd=prlinks
Cardiothorac Vasc Anesth 13:154175
8. Mielck F, Quintel M (2005) Extracorporeal membrane oxygenation. Curr Opin Crit Care
11(1):8793
9. Muntean W (1999) Coagulation and anticoagulation in extracorporeal membrane oxygenation.
Artif Organs 23:979983
10. Baird CW, Zurakowski D, Robinson B etal (2007) Anticoagulation and pediatric extracorporeal membrane oxygenation: impact of activated clotting time and heparin dose on survival
[Internet]. Ann Thorac Surg 83:912919; discussion 919920. Available from: http://linkinghub.elsevier.com/retrieve/pii/S0003497506018790
11. Arepally GM, Ortel TL (2006) Clinical practice. Heparin-induced thrombocytopenia. N Engl
J Med 355:809817
12. Sakr Y (2011) Heparin-induced thrombocytopenia in the ICU: an overview [Internet]. Crit
Care 15:211. Available from: http://ccforum.com/series/annualupdate
13. Favaloro EJ, Lippi G (2011) Coagulation update: whats new in hemostasis testing? Thromb
Res 127(Suppl 2):S13S16
14. De Waele JJ, Van Cauwenberghe S, Hoste EAJ etal (2003) The use of the activated clotting
time for monitoring heparin therapy in critically ill patients. Intensive Care Med 29:325328
15. Young G, Yonekawa KE, Nakagawa P etal (2004) Argatroban as an alternative to heparin in
extracorporeal membrane oxygenation circuits. Perfusion 19:283288
16. Beiderlinden M, Treschan T, Grlinger K etal (2007) Argatroban in extracorporeal membrane
oxygenation. Artif Organs 31:461465
17. Zayac EA, Pivalizza EG, Levine RL (2008) Thrombelastography in a patient with heparin-
induced thrombocytopenia treated with argatroban. Anesth Analg 106:351352
18. Pappalardo F, Maj G, Scandroglio A etal (2009) Bioline(R) heparin-coated ECMO with
bivalirudin anticoagulation in a patient with acute heparin-induced thrombocytopenia: the
immune reaction appeared to continue unabated [Internet]. Perfusion 24:135137. Available
from: http://prf.sagepub.com/cgi/doi/10.1177/0267659109106773
19. Riley RL, Cournand A (1951) Analysis of factors affecting partial pressures of oxygen and
carbon dioxide in gas and blood of lungs: theory. J Appl Physiol 4:77101
20. Fried DW, Zombolas TL, Leo JJ etal (1998) Clinical oxygen transfer comparison of the
Terumo Capiox SX18 and SX25 membrane oxygenators. Perfusion 13:119127
21. Segers PAM, Heida JF, de Vries I etal (2001) Clinical evaluation of nine hollow-fibre
membrane oxygenators. Perfusion 16:95106
22. Jegger D, Tevaearai HT, Mallabiabarrena I etal (2007) Comparing oxygen transfer performance between three membrane oxygenators: effect of temperature changes during cardiopulmonary bypass. Artif Organs 31:290300
23. Lehle C, Philipp A, Gleich O etal (2008) Efficiency in extracorporeal membrane oxygenation-
cellular deposits on polymethypentene membranes increase resistance to blood flow and
reduce gas exchange capacity. ASAIO J 54:612617
24. Zhou X, Loran DB, Wang D etal (2005) Seventy-two hour gas exchange performance and
hemodynamic properties of NOVALUNGiLA as a gas exchanger for arteriovenous carbon
dioxide removal. Perfusion 20:303308
[email protected]
Part VI
Complications of ECMO
[email protected]
Complications of Extracorporeal
Support and Their Management
36
Antonio Rubino, Richard Haddon,

Fabrizio Corti, and Fabio Sangalli
Due to its technical complexity and the critical illness of patients suited for its use,
ECMO has a high potential for complications [1]. Most of these can result in lifethreatening conditions able to change the patients outcome. For this reason, a correct prevention and an early recognition of symptoms and signs can help to reduce
the incidence of adverse events.
Complications can be related to either circuit components or patient conditions.
36.1
Circuit-Related Complications
36.1.1 Blood Clots in the Circuit and Thromboembolism

Because of blood-surface interaction, clots can form in the circuit and originate
embolic events with potentially devastating consequences [2].
Thrombi can appear at almost any point in the circuit (oxygenator, pump head,
tubing) (Figs. 36.1 and 36.2).
A. Rubino R. Haddon
Anaesthesia and Intensive Care,
Papworth Hospital NHS Trust, Papworth Everard CB23 3RE, UK
F. Corti
Department of Cardiac Surgery, San Gerardo Hospital,
F. Sangalli (*)
[email protected]
415
416
A. Rubino et al.
Fig. 36.1 Clot formation on

the venous side of the
membrane lung
Fig. 36.2 Clots in the tubing
36.1.2 Gas Embolism

Gas embolism is related to the capacity of the centrifugal pump to generate high
negative pressure between the drainage cannula and the pump head. In case of air
entrainment in this part of the circuit, a massive gas embolism can occur.
36.1.3 Circuit Fractures

Fissures or breakage can occur in virtually every component of the extracorporeal
circuit. Those may be large, causing major blood loss, or unapparent and difficult to
detect. When on the venous site of the circuit, massive air aspiration and consequent
embolism can occur as a result of high negative pressure generated by the centrifugal pump (Fig. 36.3). All fractures prompt substitution of the part or of the entire
circuit depending on the site of rupture.
[email protected]
36
Complications of Extracorporeal Support and Their Management
417
Fig. 36.3 Massive air

embolism in the left ventricle
36.2
Patient-Related Complications
36.2.1 Cannulation Vascular Complications

Cannulation may pose difficulties related to both anatomical reasons (size and stenosis of vessels, particularly arteries, anatomical variations, previous surgeries,
morbid obesity) and clinical conditions of the patient (absence of pulsatility during
low-flow states or cardiac arrest, vasoconstriction). These may even if this is quite
rare in our experience make cannulation impossible unless a central approach is
readily attainable.
Vascular complications can occur at the time of cannulation and are more frequent with the percutaneous approach as compared to the surgical approach.
36.2.1.1 Vascular Access Complications

Arterial cannulation, particularly when performed with a percutaneous approach,
may lead to serious complications.
Perforation of the posterior wall of the vessel may cause uncontrollable bleeding
with ensuing inadequate perfusion. This may in turn lead to compartmental syndrome or retroperitoneal hematoma, depending on the site of vascular injury. This
may also result in arteriovenous fistula or pseudoaneurysm formation, which may
require surgical repair immediately or at a later time (Fig. 36.4).
Guidewires and dilators can also cause arterial dissection or the creation of false
ways leading to extravascular cannula positioning and consequent inability to circulate blood.
All these complications require close monitoring and prompt interventions.
Some tricks to prevent such complications are outlined in Chap. 37.
36.2.1.2 Leg Ischemia

Leg ischemia is a risk with femoral arterial cannulation (Fig. 36.5). In case of
peripheral VA ECMO, the return cannula in fact is typically placed into the femoral
artery and the drainage cannula into the contralateral femoral vein. The insertion of
a distal perfusion cannula, which is connected to the return tubing via a T-connector,
[email protected]
418
A. Rubino et al.
Fig. 36.4 Iatrogenic arteriovenous fistula of the femoral vessels with voluminous pseudoaneurysm of the femoral artery
should always be considered (Fig. 36.6). It is easier to place the distal perfusion
cannula before the return cannula is inserted in the femoral artery to avoid the insertion in condition of reduced pulsatility. In case of emergency VA ECMO insertion
while the return cannula insertion has been prioritized, a surgical cutdown could be
performed in order to place the perfusion cannula instead of percutaneous
insertion.
36.2.2 Bleeding
Bleeding is one of the most common complications during ECMO because of systemic anticoagulation and platelet dysfunction, which results from contact and
shear stress associated activation. For this reason even conventional routine procedures (i.e., endotracheal suctioning, nasogastric tube positioning, urinary catheterization) can lead to uncontrollable bleeding, requiring further intervention and
alteration in the anticoagulation regimen.
[email protected]
36
419
Fig. 36.5 Peripheral

ischemia in the cannulated
leg
Fig. 36.6 Peripheral

perfusion cannula in the
cannulated leg
A continuous monitoring of coagulatory status (ACT, aPTT, PT, and PLT count,
TEG/ROTEM) is essential especially prior to invasive procedures to reduce the
bleeding events.
[email protected]
420
A. Rubino et al.
36.2.2.1 Surgical and Cannulation Site Bleeding

Cannulation sites represent the most frequent source of bleeding, especially in the
setting of surgical cannulation. In this case a slow oozing around the cannula can be
related to disruption of small vessels and managed with compression or redressing.
Sometimes a resuturing or modification of coagulation protocol is required. The
extent of bleeding is generally less in case of percutaneous cannulation with
Seldinger technique. Compared to venous cannulation, arterial cannulation is
related to higher risk of bleeding. Whenever bleeding occurs from cannulation sites,
it is good practice to assess the correct positioning of the cannula before taking any
intervention, in order to exclude malposition and prevent accidental decannulation.
In one series of more than 400 adults placed on ECLS for severe ARDS refractory to all other treatment, cannulation site bleeding occurred in 31.4 % and surgical
site bleeding in 26.7 % of patients [3].
In case of central ECMO, cannula site bleeding represents a surgical problem
and requires careful monitoring and prompt surgical revision in order to avoid serious complications such as tamponade.
36.2.2.2 GI Bleeding and Airway Bleeding
In this delicate scenario of hemostasis and anticoagulation, bleeding from the
mucous membrane as in the nasopharynx, trachea, stomach, rectum, and bladder
commonly occurs even with minor trauma. Routine procedures related to patient
care such as suctioning and bronchoscopy or urinary catheter insertion can trigger
bleeding often difficult to control in these areas.
GI bleeding can occur from esophagitis, gastritis, duodenal ulcer, or other
sources. Endoscopy is often able to recognize the precise source of bleeding, and
angiography should be considered if further investigation is needed. ECLS patients
should be treated in the same way of any patient with active GI bleeding ensuring
the correction of the coagulopathy and considering endoscopic procedure when feasible or surgical operation in case of uncontrolled bleeding.
36.2.3 Coagulopathy (Thrombocytopenia, HITT, and DIC)

One of the key points in the ECMO management is the delicate balance between
hemostasis and thrombosis that requires constant clinical and laboratory monitoring
with replacement of coagulation factors, fibrinogen, and platelets. These aspects are
dealt with in Chap. 7.
An expected dilution of blood cells, platelets, and proteins due to the circuit
priming with crystalloids occurs when ECLS is established with increased fluid
component of the blood and equilibration into the extracellular space causing
edema. An anemic state requires furthermore higher pump flow to achieve adequate perfusion and gas exchange, resulting in higher post-pump pressures. For
these reasons achievement of adequate hematocrit levels is particularly important
in ECLS [4].
Thrombocytopenia is common in ECLS patients. It may be a consequence of the
primary disease, drug induced, or caused by blood exposure to the circuit surface.
[email protected]
36
421
Fig. 36.7 Thrombus (arrow)

in the ascending aorta
The interaction between the blood and the artificial surface causes protein
absorption such as fibrinogen and albumin. As this occurs, platelets adhere and create a cascade of events that results in thrombus formation (Fig. 36.7) and activation
of the coagulation cascade by the intrinsic pathway, resulting in the release of
inflammatory mediators and the production of thrombin [5].
Platelet count can drop as low as 40 % from baseline within the first 4 h of ECLS
[6]. The usual practice and the ELSO Guidelines indicate the threshold for platelet
transfusions of 80,000 despite different centers indicating a minimum platelet count
of 100,000.
Even though the platelet count is over the minimum level (80,000100,000), the
platelet function may be impaired. In this case a kallikrein inhibitor (tranexamic acid
or aprotinin) is suggested to improve platelet function if bleeding is a problem [4].
Systemic heparinization has been demonstrated as the gold standard for anticoagulation in these patients, as it is inexpensive, easily titrated, easily monitored at
the bedside, and immediately reversed by protamine. Despite these advantages,
heparin does not prevent platelet-surface interaction, and furthermore, it can itself
cause further platelet activation, dysfunction, and consumption [6].
A rare condition associated with heparin anticoagulation is heparin-induced
thrombotic thrombocytopenia (HITT) characterized by multiple white arterial
thrombi and platelet count less than 10,000. The assay available for HITT has a very
high false-positive rate. If an ECLS patient has true HITT, the platelet count will be
consistently less than 10,000 despite platelet infusions. In this case, if there are no
other reasons to explain the thrombocytopenia, it is reasonable to use different anticoagulant such as Argatroban.
Another aspect that can trigger the coagulopathy related to ECLS is hemolysis,
suspected if the urine has a pink tinge and verified by plasma Hb measurement
(normal plasma hemoglobin should be less than 10 mg/dl). High risk of hemolysis
occurs if the pump suctions exceeds the blood drainage (high inlet pressures) as a
result of high flow rate through a very small orifice or if there is a high level of
occlusion in the post-pump circuit. The presence of clots in the pump chamber may
enhance this phenomenon [5].
[email protected]
422
A. Rubino et al.
36.2.4 Neurological Complications

The most devastating of these causes is the intracranial hemorrhage, usually fatal.
The reported incidence of intracranial hemorrhage varies between 1.6 and 18.9 %.
A careful management of anticoagulation with a prompt correction of thrombocytopenia and prevention of renal failure seems to be the factor that can reduce the
incidence of this fatal complication, while the duration of ECMO support has not
been shown as an independent risk factor [6]. The Extracorporeal Life Support
Organization (ELSO) registry reported 2 % of CNS hemorrhage in adult patients
requiring cardiac support with an overall survival of 8 % and a rate of 4 % in patients
requiring VV ECMO with a higher survival rate (21 %).
In addition to hemorrhage, infarction (18 %) and seizures (210 %) are common complications. Thromboembolic events can progress to ischemic stroke in different portions of the brain, including frontal lobe, occipital lobe, basal ganglia, and
parietal lobe. Clinical seizures can be associated with radiographic evidence of
cerebral edema.
36.2.5 Cardiac Complications

Complications within the chest such as cardiac tamponade, LV distention, or pneumothorax can be the cause of further hemodynamic instability and ECMO flow
disturbances because of cannula compression or reduced atrial volume. Hypotension
and concomitant high (negative) inlet pressure can be related to hypovolemia or
suboptimal patient sedation itself, but it is fundamental to exclude any of these complications. Any sudden change of intrathoracic pressure such as tension pneumothorax may constrict the pericardium leading to decreased venous return and decrease
of the ECMO flow similar to tamponade physiology. An initial fluid challenge can
assess the volume status of the patient as well as an optimization in sedation and
paralysis. If no improvement in ECMO flow is noted, imaging such as chest radiograph or echocardiogram is recommended.
The ECMO flow disturbances related to significant intrathoracic pathology have
been well described in pediatric population; however, these have not been clearly
described in the adult ECMO patients.
A particular problem associated with VA ECMO is the LV distention. An evidence of pulmonary edema on the chest radiograph or edema fluid frothing up the
endotracheal tube can be the first manifestation of this problem. A transesophageal
echocardiography can confirm the diagnosis identifying a severely dilated LV. The
presence of mitral or aortic regurgitation can exacerbate the problem, and an increase
in pump flow helps to reduce the pulmonary flow, ameliorating the problem.
LV distention is not a problem exclusively related to central VA ECMO; in
patient with peripheral VA ECMO, despite adequate left ventricular unloading,
there is still returning blood flow to the left atrium, principally due to the bronchial
[email protected]
36
423
circulation. Consequently, if LV contractility is profoundly reduced, we can assist to

an increase in left heart pressures resulting in LV distention [7].
The increase in wall stress associated with LV distention not only increases myocardial energy consumption resulting in ischemia but also reduces the likelihood of
ventricular recovery.
A surgical or percutaneous LV vent insertion must be performed in these cases.
36.2.6 Infection/Sepsis
ECMO patients are at increased risk of nosocomial infection when compared with
other patients in the surgical ICU setting [8].
ECMO patients generally have multiple indwelling catheters, such as pulmonary
artery catheters and radial artery catheters, in addition to the cannulae used for the
ECMO circuit, with consequentially increased risk of bloodstream infections (BSI).
Considering their prolonged intubation, invasive catheters, and frequent antibiotic therapy, these patients are exposed as well at high risk of ventilator-associated
pneumonia.
Usual clinical signs and symptoms associated with nosocomial infections may
not be present in these patients making the diagnosis difficult. In particular, fever
may be absent due to servo-control of body temperature by the heat exchanger.
Broad-spectrum empiric antimicrobial therapy should be instituted early until the
results of microbiological cultures become known.
References
1. Smedira NG, Moazami N, Golding CM, McCarthy JF, Apperson-Hansen C, Blackstone EH,
Cosgrove DM III (2001) Clinical experience with 202 adults receiving extracorporeal membrane oxygenation for cardiac failure: survival at five years. J Thorac Cardiovasc Surg
122(1):92102
2. Gaffney AM, Wildhirt SM, Griffin MJ, Annich GM, Radomski MW (2010) Extracorporeal life
support. BMJ 341(2):c5317c5317
3. Hemmila MR, Rowe SA, Boules TN, Miskulin J, McGillicuddy JW, Schuerer DJ, Haft JW,
Swaniker F, Arbabi S, Hirschl RB, Bartlett RH (2004) Extracorporeal life support for severe acute
respiratory distress syndrome in adults. Trans Meet Am Surg Assoc CXXII & NA:193205
4. ELSO Guidelines for Cardiopulmonary Extracorporeal Life Support. Extracorporeal Life
Support Organization, Version 1.3 (2013) Ann Arbor. Accessed on Jan 2014. www.elsonet.org
5. Reynolds MM, Annich GM (2011) The artificial endothelium. Organogenesis 7(1):4249
6. Kasirajan V, Smedira NG, McCarthy JF, Casselman F, Boparai N, McCarthy PM (1999) Risk
factors for intracranial hemorrhage in adults on extracorporeal membrane oxygenation.
Eur J Cardiothorac Surg 15(4):508514
7. Guirgis M, Kumar K, Menkis AH, Freed DH (2010) Minimally invasive left-heart decompression during venoarterial extracorporeal membrane oxygenation: an alternative to a percutaneous approach. Interact Cardiovasc Thorac Surg 10(5):672674
8. Burket JS, Bartlett RH, Vander Hyde K, Chenoweth CE (1999) Nosocomial infections in adult
patients undergoing extracorporeal membrane oxygenation. Clin Infect Dis 28(4):828833
[email protected]
Troubleshooting Common and Less

Common Problems
37
Lisen Hockings and Alain Vuylsteke
37.1
Patient Selection
No absolute or definitive criteria for patient selection or exclusion exist, and institutions have developed lists of indications and contraindications for ECMO based on
their own experiences.
Published criteria vary based on the mode of support [13].
When doubt exists about the need for and/or suitability of a patient for ECMO a
consensus opinion between experienced ECMO clinicians should be sought.
Issues can arise when ECMO has been started without considering future plans.
Sometimes used as a bridge to decision, it is most often a bridge to recovery or
another therapy such as an organ transplant.
37.2
Type of Support
It is tempting to consider the use of VA ECMO as a solution to all problems.

For example, patients with respiratory failure can present with profound hemodynamic compromise, perhaps because of hypoxia, acidosis, or extreme settings on
the mechanical ventilator. The clinician might be tempted to initiate VA ECMO,
either peripherally or centrally.
Central VA ECMO should be very carefully considered, as it requires sternotomy. The positioning and anchorage of the ECMO tubing is critical to allow mobilization of the patient and avoid infection to set in (mediastinitis). A sternotomy is a
L. Hockings
Anaesthesia and Intensive Care, Papworth Hospital NHS Foundation Trust,
A. Vuylsteke (*)
Anaesthesia and Intensive Care, Papworth Hospital,
[email protected]
425
426
L. Hockings and A. Vuylsteke
major aggression when other options are possible. It will render future surgery difficult and may influence decision to progress to other form of support.
Peripheral VA ECMO (femoro-femoral) can lead to differential oxygenation,
with the ECMO circuit in effect oxygenating part of the body and the native circulation oxygenating another (or not if the lung ability to exchange gas is affected by the
disease process). In this situation, a patient can be pink but have profound coronary
ischemia! Peripheral VA ECMO is primarily a bridge to decision (such as during
e-CPR) or a relatively more permanent form of cardiac support (such as a ventricular assist device or transplant). Complications related to the arterial cannulation are
common, including ischemia and dislodgment.
Other configuration of peripheral VA ECMO (such as with the return cannula
grafted onto the axillary/subclavian artery) is more likely to ensure adequate cerebral oxygenation (but not necessarily myocardial) [4]. Grafting can in itself cause
an issue with disconnection due to continuous shear stress on the suture lines.
The hemodynamic disturbance observed in patients with respiratory failure can
usually be sorted with the restoration of the physiological acidbase equilibrium,
oxygenation, and decrease in intrathoracic pressures. This can be achieved by both
VA and VV ECMO [5].
VV ECMO is relatively less invasive and allows patient mobilization.
Single-stage dual-lumen cannulation will suit some patients and offers the
advantages of single-site cannulation. Support of some patients may be hindered by
the limitation of blood flow rates when using this cannula [1, 6, 7].
37.3
Insertion
37.3.1 Location of Other Lines

The location of all central venous access (such as central venous lines, dialysis catheters, and pulmonary artery catheters) and arterial monitoring lines needs to be
reviewed when contemplating ECMO.
Sufficient time should be made available to resite lines and change infusions
prior to establishing ECMO whenever possible.
Team can select to use the ECMO circuit to connect dialysis systems or infusion
pumps. This increases the risk of air embolism or infections due to repeated manipulation. Strict protocol and training program must be in place for such practices to
be safe.
In VV ECMO, the right internal jugular vein (RIJV) is best situated to allow easy
insertion of large cannula. It is the only site that can be used safely with single-stage
dual-lumen in the adult patient. We advocate having the RIJV available from the
outset as it is safer to resite lines prior to instituting ECMO. If a line is already in
place, it should be left in and used as a guide to insert the cannula guidewire,
decreasing the risk of puncture.
Concerns have been raised in relation to obstructing cerebral venous drainage
when both internal jugular veins are used with large cannula. Positioning the patient
[email protected]
37
Troubleshooting Common and Less Common Problems
427
head carefully may help improve this. Assessing patency of the jugular veins before
establishing VV ECMO is warranted.
Thrombosis in any of the jugular vein seen prior to cannulation should force the
clinician to reassess the risk/benefit of VV ECMO and/or the cannulation location.
Solution would be to limit the VV ECMO support to femoro-femoral cannulation, but efficiency of such system will be lower.
Subclavian insertion of multiple-lumen central venous line should be considered,
accounting for the risk of pneumothorax in often unstable patients and subsequent
thrombosis. Real-time ultrasound to cannulate the axillary/subclavian vein increases
the safety of this approach in patients [8, 9].
A pulmonary artery catheter can measure the pulmonary artery pressure.
Thermodilution measurements will be inaccurate as part of the injected solution is
likely to end in the ECMO circuit and mixed venous oxygen saturation is not meaningful. Visualization of the pulmonary artery trace allows identification of pulmonary blood flow on those patients on VA ECMO. The usefulness of a pulmonary
artery catheter during VV ECMO is dubious.
The presence of end-tidal CO2 is an alternative means of assessing the presence
of pulmonary flow when pulmonary artery catheter is not in situ and the lung can be
ventilated.
Cannulation of an artery may lead to distal ischemia, and reperfusion cannula
should be considered and inserted in most cases.
Note that the relative venous obstruction from the presence of a femoral venous
line can leave the limb at risk of hyperperfusion (if the reperfusion flow exceeds
venous drainage). Venous cannulation can be the cause of arterial ischemia by
compression.
The decision to rewire an existing line for ECMO cannulation should be undertaken on a case-by-case basis. De novo puncture under sterile conditions will
decrease the infectious risk.
A right upper limb arterial line is preferred in patients on peripheral VA ECMO
(alternative arterial monitoring sites may miss the early signs of differential
hypoxia). Where a right upper limb arterial line is not possible, the saturation probe
should be placed on the right upper limb.
If time allows, it is easier to resite the arterial lines prior to commencing VA
ECMO support even in the setting of cardiogenic shock (pulsatile flow).
In our view, the use of ultrasounds is mandatory.
Where central veins are being accessed during ECMO support, it is important to
remember that the continuous access from the vena cava(e) generates negative pressures that increase the potential for air entrainment and air embolus.
37.3.2 Cannulation Location and Techniques

The portable size of ECMO circuits, associated to increased experience and range
of indications, has led to ECMO being used in different hospital locations. Some
countries have even experimented with out-of-hospital insertion.
[email protected]
428
Operator skill set, urgency of support, and available equipment will influence the
technique.
Operating rooms, interventional radiology laboratories, and angiography suites
offer the advantage of real-time imaging to prevent accidental cannulation of inappropriate vessels (particularly as they occur more distally e.g., the hepatic vein)
and early identification of wire kinking during dilatation.
Combined with real-time ultrasound for percutaneous vessel cannulation, we believe
that image intensification is the gold standard technique for ECMO cannulation.
In urgent and emergent situations, cannulation can be performed safely with realtime ultrasound for percutaneous cannulation and transthoracic or transoesophageal
echocardiography. This is used to confirm and optimize cannula location and position (VV and VA ECMO).
Transthoracic ultrasounds (particularly subcostal views) are often adequate for
confirming guidewire location (correct vessel) and positioning cannula.
The presence of skilled operator in sufficient number allows continued visualization
of the J-loop of the guidewire using echocardiography. Careful communication with
the primary operator(s) can alert the team to inadvertent wire kinking or migration.
Parasternal/apical imaging and Doppler can be used to confirm the correct orientation of dual-lumen catheters (return jet directed across tricuspid valve).
The length of the wires and catheters used in ECMO means that a 2-person team
technique is better used for percutaneous cannulation: the primary operator is in
charge of sequential dilatation of the skin/subcutaneous tissue/vessel, while the second (and potentially more important operator) controls the guidewire at all times.
Excellent teamwork skills and clear, concise communication are paramount to
safe, timely, and efficient percutaneous ECMO cannulation.
These technical and nontechnical skills are particularly important when the
potentially stressful situations in which ECMO can be inserted are considered.
Cannulation in extreme conditions, without the use of imaging, can be attempted.
New cannulae allow insertion without dilatation. These techniques increase substantially the risk and should only be used in exceptional circumstance.
37.3.3 Ultrasound Tips

Ultrasound can be invaluable during ECMO cannulation optimizing access location choice, confirming vessel patency, and guiding the choice of cannula size.
We advocate sterile, real-time US cannulation to minimize short- and longerterm cannulation issues.
Femoral vessels should be imaged distal to the inguinal ligament.
Femoral veins: the femoral vein should be cannulated away from the insertion of
the long saphenous vein to avoid shearing it away from the FV during serial dilatation and cannulation.
Femoral arteries: the femoral artery should be imaged in both the short and long
axes. The radius of the vessel should be measured in the long axis to avoid overestimation of size and attempts at inserting excessively large arterial cannula.
[email protected]
37
429
The diameter of an ECMO catheter in mm is one-third of its size in French gauge

(i.e., a 21 Fr catheter has an external diameter of 7 mm).
The catheter should be inserted into the common femoral artery, distal to the
inguinal ligament but proximal to the bifurcation into the superficial femoral (SFA)
and profunda femoris (PFA) arteries.
Inserting the cannula close to the bifurcation may complicate later surgical repair
(after decannulation).
It is much easier to insert the backflow or reperfusion catheter guidewire
anterograde down the common femoral artery prior to inserting the ECMO cannula
retrograde into the distal aorta. It is possible to do it subsequent to the cannula insertion (in case of emergency) but will require skills and patience.
37.3.4 Peripheral VA: Backflow Cannula Insertion

The backflow or reperfusion cannula guidewire should be inserted into the common femoral artery, but the guidewire should be seen to be directed down the superficial femoral artery and not the profunda femoris artery.
In-plane real-time US can be used to guide this procedure.
The guidewire has a tendency to feed down the PFA due to the angle of insertion.
This can be overcome by rotating the needle 180 just as the guidewire approaches
the bifurcation.
37.3.5 Difficult Cannulation

Cannulation difficulties can arise for even the most experienced operators.
A two-person technique may alleviate frustration when difficulties are
encountered.
Wherever possible, the Seldinger serial dilatation technique is preferable to a
surgical approach in terms of minimizing longer-term bleeding complications.
When an open surgical approach is required, it is preferable for this to occur in an
operating theatre with the procedure performed by an experienced cardiovascular
surgeon.
Dilators with a very long taper assist in dilatation of even the most difficult subcutaneous tissues.
When the subcutaneous dilators are advanced and meet resistance, we advocate
rotation of the dilator such that a significant amount of torque is applied to the skin
and subcutaneous tissue. With the torque maintained the dilator is withdrawn and
the soft tissues tend to give around the dilator. The dilator is then readvanced.
At all times the operator should ensure free movement of the guidewire.
If a small surgical skin incision is required, a single insertion of the scalpel blade
with the blunt end of the blade run along the guidewire may limit bleeding complications. Aggressive and multiple skin incisions are associated with increased bleeding rates at the cannulation site.
[email protected]
430
37.3.6 Kinking the Wire

The guidewire may kink during cannulation, and it is important to recognize this complication quickly to avoid extravascular insertion of a larger dilator or the cannula.
The guidewire should be withdrawn until the kink is external to the patient (while
making sure not to completely withdraw the guidewire).
If the kink cannot be straightened (unfortunately common), it can often be
exchanged by inserting a smaller dilator than the one that kinked the wire back into
the vessel, removing the kinked wire and replacing it with a new one.
Spare guidewire kits are available and should always be available.
When performing multiple site cannulation, it might be possible to use the guidewire from the other cannula kit if one guidewire kinks.
The guidewire should not be discarded after the first cannula is in!
It is important to be aware of the length of the guidewire being inserted a long
multistage venous cannula cannot be used with a short arterial cannula guidewire.
The amount of guidewire external to the patient when using a longer guidewire
for an arterial cannula must be kept in mind (problems with this can be overcome by
real-time echo imaging of the J-loop).
37.3.7 Right Ventricular Perforation

Right ventricular perforation is a recognized complication of jugular cannulation
particularly when the Avalon cannula is being used as it needs to traverse the right
atrium [10].
Multiple premature ventricular ectopic beats may indicate that the guidewire is
in the right ventricle.
Tamponade is the near-inevitable result of perforation of the right ventricle. This
can occur slowly (when the perforation is due to the wire) or very rapidly (when the
perforation is with the cannula). If recognized and the cannula is the cause of the
perforation, it should be clamped immediately but not removed.
The risks of this procedure can be decreased by employing the imaging techniques described above.
On the rare occasions it does occur, it is best managed with urgent surgical repair.
Undertaking cannulation in the operating theater with a cardiothoracic surgeons
performing the cannulation procedure or readily available increases the safety of the
procedure but is not always possible
Other vessels can be damaged, such as the inferior vena cava, coronary sinus,
and hepatic veins [10, 11].
The interatrial septum may be perforated even by a femoral venous line inserted
too far.
Some advocates the use of a long sheath to exchange the initial guidewire for a
stiffer wire to minimize the risk of kinking with serial dilatation and cannula insertion. This prolongs cannulation time and increases its complexity.
[email protected]
37
431
37.3.8 Awake Cannulation

Patients requiring central ECMO and most patients requiring VV ECMO (after failure of conventional treatments for respiratory failure and de novo cannulation e.g.,
in the hope of bridge to transplantation) are cannulated while anesthetized and
ventilated.
There are a number of situations where patients may be cannulated awake.
This should be explained to them as much as practicable (and consent
obtained).
A designated, experienced clinician or critical care physician should manage the
patients analgesia and hemodynamics during cannulation and be prepared for and
capable of anesthetizing and managing safely the airways if the procedure is poorly
tolerated or complications occur.
A generous amount of local anesthesia is required. Local anesthetic toxicity must
be considered.
37.3.9 ECMO CPR

The decision to perform ECMO CPR should be considered early as there is a necessary lag time until support is established.
An ECMO CPR team and equipment should be readily available where this is
being considered [12].
A pre-primed ECMO circuit will reduce the time to establish support
[12, 13].
The primary resuscitation team should continue resuscitation in line with international guidelines. Interruptions to CPR should be kept to a minimum and therapeutic hypothermia should be considered [14, 15].
To facilitate cannulation of the vessel with the needle, passing the guidewire and
minimizing sharp injuries, it is advisable to interrupt chest compressions until the
wire has been fed into the vessel [16].
An automated cardiopulmonary resuscitation device may free additional personnel to focus on ECMO CPR and reduce fatigue from continued chest compression.
It can cause occult intrathoracic or intra-abdominal organ damages that will cause
subsequent problems. A team member will be required to manage/troubleshoot this
device during the resuscitation.
Ultrasound should be used for cannulation, whenever possible.
During CPR it can be difficult to identify venous from arterial blood on
aspiration. The venous blood may paradoxically appear more pulsatile during
CPR.
Defibrillation safety is paramount, and defibrillation should be delayed until all
operators are able to stand clear of the patient.
When ECMO has been established, definitive diagnosis and treatment of the
underlying cause of arrest can be addressed.
[email protected]
432
37.4
Maintenance
37.4.1 Low-Flow Alarms

This is perhaps the most common alarm heard while patients are on ECMO. It must
prompt an immediate search for possible causes.
It may simply identify that the alarm limit is set above the current flow rates (e.g.,
immediately after changing the pump settings). It is advised to set the alarm at
500 mL/min below the current or desired flow reading.
Cause of low-flow alarm includes:
Access insufficiency. It is unusual, but not impossible, for circuit flows to be
limited by return cannula size.
Kinks or occlusion of the circuit tubing or cannula.
Oxygenator thrombosis. This can be suggested by a rising transmembrane pressure gradient and falling ECMO flows. The oxygenator may have visible thrombus within it (but this can occur without obvious thrombus). Although the
transmembrane pressure gradient will fall slightly when the pump speed is turned
down, the flows will not improve.
Systemic hypertension or vasoconstriction in patients on VA ECMO may cause
low flow (an increased afterload to the pump). This is usually only seen to activate the alarm when the changes occur suddenly, for example, coughing/
waking.
Air embolus. Low-flow or no-flow alarms may be the first indication of an air
embolus when the presence of gas bubbles prevents the ultrasonic flow probe
from estimating flow. More dramatically, an air lock in the pump head will stop
flow entirely. This complication is discussed in greater depth in Chap. 36.
37.4.2 Access Insufficiency

Access insufficiency or suck down occurs when the venous return to the access
cannula(e) is insufficient for the degree of pump suction at the set pump speed [3].
In circuits with a negative pressure monitor on the access side, this will be suggested by the access pressures falling below 100 mmHg.
Where the access pressures are not measured, the first signs of may be variations
in the flow, swinging of the access cannula, and subsequently kicking of the
access line with activation of the low-flow alarms on the pump. This usually indicates a negative pressure much greater than 100 mmHg.
The excessively negative pump suction (relative to venous return) results in the
vessel wall being sucked onto the access ports of the cannula and obstructing flow
into the pump.
Swinging and kicking of the ECMO lines, the most obvious manifestations of
access insufficiency, are rarely seen when using dual-lumen catheters.
Immediate interventions are to reduce the pump speed (to disengage the vessel
wall from the access cannula) paradoxically, the flows will tend to increase with
this maneuver and to correct any underlying volume disturbance.
[email protected]
37
433
Access insufficiency is a common problem when the access cannula is positioned too low (in the more collapsible extrathoracic inferior vena cava).
A careful examination of secondary causes for reduced venous return to the heart
should be considered:
Bleeding (both obvious and covert)
Excessive diuresis
Other forms of hypovolemia
Coughing, straining, and increased intra-abdominal pressure
Access insufficiency can be suggestive of tamponade or tension pneumothorax,
and these should be looked for with echocardiography/CXR/US if simple interventions fail to resolve the issues.
Where access insufficiency continues to be problematic, the patient still needs
high flows (if not higher flows), and secondary causes have been eliminated; a second access cannula can resolve the issues and allow high-flow ECMO.
37.4.3 Recirculation (VV ECMO)

Recirculation can be identified when blood in the access cannula appears better
oxygenated than the patients saturations would suggest that is, the blood being
drawn into the ECMO circuit is oxygenated blood from the return cannula, not
deoxygenated blood from the systemic venous system.
It can be confirmed by taking a preoxygenator blood gas. If the access and return
cannulae are too close within the venous system, recirculation is more likely to
occur and their position(s) should be adjusted.
Recirculation increases as flow increases and the system will become less efficient.
Recirculation occurs less often with dual-lumen catheters [1], and this often
compensates the lower flow obtained with this cannula, in relation to the overall
efficiency of a circuit.
37.4.4 Inadequate Support

37.4.4.1 VV ECMO: Persistent Hypoxemia
When a circuit can deliver 35 L of oxygenated blood to the right atrium but the
patients native cardiac output is significantly higher than this, they will remain hypoxic
(although less hypoxic than in the absence of VV ECMO). This is effectively a shunt.
Options to improve tissue oxygen delivery matching include increasing the
ECMO flow (reducing the size of the shunt), increasing the oxygen-carrying capacity of the blood (transfusion), and reducing systemic oxygen requirements (aggressive treatment of sepsis, preventing fever, consider active cooling, and even moderate
therapeutic hypothermia).
This is analogous to the approaches to addressing inadequate tissue oxygenation
in the non-ECMO patient.
Circuit flows can be increased by optimizing cannula(e) position and volume
state or by introducing a second access cannula (e.g., high-flow VV ECMO).
[email protected]
434
37.4.4.2 VA ECMO: Vasoplegia

Patients can remain hypotensive even with VA ECMO flows of 46 L in the setting
of vasoplegia.
When seemingly adequate ECMO flows have been established, vasoplegia
should be managed aggressively with potent vasoconstriction.
Causes of vasoplegia should be looked for and treated (sepsis is most likely but
other forms of high-output cardiac failure should be considered, as well as acute
spinal cord injury).
37.4.4.3 Peripheral VA ECMO: Pulmonary Edema

This can occur in the setting of aortic valve regurgitation where the left ventricle
gradually fills and dilates if the regurgitant volume exceeds the amount of blood
ejected.
Echocardiography can confirm the diagnosis.
Suggested interventions include increasing PEEP, reducing the ECMO flows as
permitted by systemic flow requirements, and increasing inotropy to assist left ventricular ejection.
If these interventions fail, then consideration should be given to proceeding to
central VA ECMO, left ventricular assist device (LVAD), or biventricular assist
device (BiVAD) insertion with aortic valve replacement/repair and/or a left ventricular vent.
An alternative cause for pulmonary edema on ECMO can be in the setting of
isolated left ventricular failure and residual adequate right ventricular function.
All of the venous return not captured by the access cannula of the ECMO circuit
is pumped by the right ventricle into the pulmonary circulation, and it is not able to
be ejected by the failing left ventricle.
Suggested interventions include increasing the ECMO flows, increasing the
PEEP, and reducing inotropy (any exogenous inotropes are most likely to further
augment RV function and accelerate the pulmonary edema).
These interventions should be carefully monitored with echo and reassessment.
If the patient remains in pulmonary edema and hypoxic, then a high-flow VA
ECMO configuration with a second venous access cannula can be considered.
37.4.4.4 Peripheral VA ECMO: Differential Hypoxia

Differential hypoxia occurs when the native cardiac function returns but the patient
has hypoxemic respiratory failure.
The deoxygenated blood that reaches the left ventricle via the pulmonary circulation is ejected into the ascending aorta, and the patients native cardiac output supplies the first branches of the aorta (the coronary arteries, the brachiocephalic trunk,
and the left common carotid artery).
This can lead to myocardial and cerebral ischemia with potentially fatal
consequences.
Suggested interventions include early diagnosis and management of the primary
cause of the respiratory failure (particularly if it relates to pulmonary edema and
identifying the mechanism of the pulmonary edema) and temporarily attempting to
[email protected]
37
435
reduce the output of the left ventricle to prevent hypoxemic injury to the myocardium and brain:
Increasing ECMO flows (and considering conversion to high-flow VA ECMO
configuration)
Reducing exogenous inotropes
Considering simultaneous VV and VA ECMO (VVA ECMO) or, where myocardial recovery is sufficient, conversion from VA to VV ECMO
Considering conversion to central VA ECMO or peripheral VA ECMO with
the return cannula moved to be a side graft onto the axillary or subclavian
artery [4]
37.5
Cardiac Arrest on ECMO
37.5.1 VV ECMO
Patients who arrest on VV ECMO need CPR as per standard protocols with careful
attention to securing the cannula(e) [14, 15]. Defibrillation should occur as per published guidelines.
If the cause of the cardiac arrest is hypoxemia and the patient had previously
been well supported on VV ECMO, then the circuit should be interrogated for
potential issues confirm power supply, oxygen supply connected to the oxygenator, ensure FiO2 is at 1.0, and sweep gas at least equal to blood flow rates.
Consider a fluid bolus and blood transfusion to improve ECMO flows and tissue
oxygen delivery.
Alternative ECMO-related causes of cardiac arrest include hemorrhage and air
embolus (see Chap. 36).
37.5.2 VA ECMO
Cardiac arrest on VA ECMO is often not hemodynamically significant certainly if
the patient has very little native cardiac function at the time of the arrest.
In some cases, the arrest may only be recognized by dysrhythmia on the monitors
(e.g., asystole or ventricular fibrillation), when there is loss of pulsatility in a previously pulsatile patient (if there is not an IABP in situ), or when the triggering on the
IABP becomes irregular.
It is not necessary to perform CPR immediately in a patient who suffers a cardiac
arrest on VA ECMO as the circuit (as long as it is functioning) will sustain cardiac
output.
Look for and treat reversible causes (patient and circuit).
In a patient who was only partially dependent on ECMO, increase the ECMO
flows as tolerated by the circuit.
In the setting of ventricular fibrillation/ventricular tachycardia, it is advisable to
defibrillate, as prolonged dysrhythmia will have significant implications for
[email protected]
436
weaning from ECMO when this is suitable and ventricular distension can lead to
cardiac ischemia.
Blood stasis in the cardiac chambers might lead to clot formation and increase
the risk of stroke or peripheral embolization if pulsatility returns.
ECMO-related causes again include pump failure, oxygenator failure, hemorrhage, and air embolus.
37.6
Awake ECMO and Mobilizing the ECMO Patient
Awake ECMO is now possible and has been described for both peripheral VA and
VV ECMO [7, 1719].
Benefits include prevention of deconditioning, pressure areas, and other side
effects from prolonged sedation in the ICU. It decreases the occurrence of delirium
and the need for high doses of sedatives. Patient can exercise and enjoy the presence
of their families. We routinely provide a game console for such patients.
Potential risks are those from cannula(e) dislodgement with excessive movement
making the awake patient more difficult to manage.
Patients who are awake and on ECMO for extended periods of time may become
depressed.
Careful patient selection, explanation and reinforcement of the need for cooperation with the treating team from the outset, as well as clear and explicit plans for
future management may help to alleviate problems.
Mobilizing ECMO patients should be undertaken with appropriate staffing levels
and equipment available.
Securing the cannula during mobilization can be difficult, and many institutions
have developed their own systems [4, 19].
At our own institution, mobilization is undertaken in the presence of a dedicated
ECMO specialist and at least one additional ECMO-trained staff member specifically managing the cannula(e) and lines, in addition to those team members required
to mobilize a non-ECMO ICU patient.
Difficulties arise when no solution to the medical condition having triggered the
use of ECMO is found. The patient can then take part in end-of-life decisions and
planning.
37.7
Miscellaneous Laboratory Issues
37.7.1 Unable to Anticoagulate

Some patients demonstrate relative heparin resistance while on ECMO, and it can
be difficult to achieve preset anticoagulation targets.
Investigations should include confirming that heparin is being administered,
checking clotting factor, anti-Xa and anti-thrombin (AT) levels, and careful liaison
with a consultant hematologist.
[email protected]
37
437
Consideration may need to be given to antithrombin replacement where patients

are found to be deficient.
Coagulation issues are complex and still poorly understood.
37.7.2 Fever and Elevated White Cell Count

Infection is a common complication in patients treated with ECMO.
Prolonged duration of ECMO support is consistently associated with an increased
risk of nosocomial infectious complications [2024].
The role of prophylactic antimicrobial therapy in patients on ECMO is controversial and varies between institutions.
Patients should not receive antimicrobials therapy simply because they are supported by ECMO.
Infection can be difficult to diagnose in the patient on ECMO:
Fever may be masked by the extracorporeal circuit.
A systemic inflammatory response may occur in the setting of extracorporeal life
support, in the post-cardiac surgical patient, and in the face of massive
transfusion.
The use of steroids as part of the treatment for respiratory and/or cardiovascular
failure may be associated with leukocytosis.
The patient may have required ECMO for support of sepsis (respiratory or systemic), and secondary infection may be difficult to distinguish from progression
of underlying disease.
Sepsis, severe sepsis, and septic shock may occur in the absence of an elevated
white cell count, particularly in the immunosuppressed patient.
There are no standard diagnostic criteria for infection occurring on ECMO.
New onset of pressor dependence may be the only indication of secondary infection in some circumstances.
In the setting of rising temperature and white cell count, a thorough search for
infection must be undertaken and appropriate cultures sent.
Where possible, lines should be changed and empiric antimicrobial treatment
instituted based on likely source of infection, local antibiograms, and patient
allergies.
Empiric antifungal treatment should be considered as Candida spp. are
consistently identified as a leading cause of infection in adult patients on ECMO
[20, 21, 23].
The kinetics of antimicrobial therapies during ECMO is under ongoing
investigation.
Where possible, drug levels should be monitored.
It is not feasible to change the ECMO cannula(e) on the suspicion of infection.
However, when it is clear that the cannula site is the cause of infection (e.g., abscess
at cannulation site), changing the cannulation strategy should be considered.
Where this involves VV ECMO, consideration should be given to changing from
femoro-femoral cannulation to dual-lumen jugular cannulation or vice versa.
[email protected]
438
37.7.3 Thrombocytopenia
Thrombocytopenia is common in patients on ECMO.
Transfusion thresholds vary between institutions and will vary according to the
type of ECMO and initial disease process.
In bleeding patients, the platelet count should be maintained in the normal range
(>50 100 109/L) [19, 24].
Where thrombocytopenia is thought to be due to heparin-induced thrombocytopenia (HIT), heparin should be stopped immediately, HIT-screen sent according to
institutional protocols, and alternative anticoagulation therapy instituted. Heparincoated circuit should be changed to non-coated ones. Argatroban and bivalirudin
have been used successfully [25, 26].
37.7.4 Increased Plasma-Free Hemoglobin

Plasma-free hemoglobin (plasma-free Hb) can be monitored as a measure of
hemolysis.
Potential ECMO-related causes of increased plasma-free hemoglobin can include
thrombus within the extracorporeal circuit, excessive suction from the centrifugal
pump (potentially generating subtle access insufficiency and red cell trauma), and
overt access insufficiency.
The plasma-free hemoglobin can be elevated by damage to the red cells during
sampling so a single isolated result should be repeated with careful attention to how
the sample is measured.
When the plasma-free hemoglobin is consistently elevated, a comprehensive
assessment of the circuit and cannula, pump speed, and anticoagulation strategy
should be undertaken and consideration given to changing the circuit [19].
37.8
The X-Ray
37.8.1 Cannula(e) Position

X-ray can monitor ECMO cannulae position.
The type of cannula in situ will affect the interpretation of the X-ray.
The clear plastic access ports at the tip of many single-lumen cannulae currently
available are radiolucent. The catheter is more proximal than would be suggested by
the appearance of the radiopaque reinforced area on X-ray. An exception is the
multiple-stage access cannula where the reinforced area extends all the way to the
tip of the cannula.
Ultrasound and echocardiography can visualize the tips of the cannula regardless
of the material.
It is important to monitor the cannula position regularly to avoid the potentially
disastrous complication of accidental decannulation.
[email protected]
37
439
Small movements in cannula position may limit the efficacy of the ECMO support (e.g., recirculation).
Clinical monitoring of the cannula position can include nursing-led checks of
measurements of cannula landmarks from skin entry.
37.8.2 Pneumothorax Management

Despite ventilatory strategies aimed at minimizing barotrauma and ventilatorinduced lung injury, pneumothoraxes are relatively common in patients on ECMO.
Intercostal drain insertion is often associated with a much greater risk of bleeding in patients on ECMO. Thoracotomy can be performed in these situations but
substantially increase the overall morbidity and mortality.
The ELSO guidelines recommend a conservative approach to pneumothorax
management in ECMO patients:
A small pneumothorax (estimated 50 % or less with no hemodynamic compromise and no enlargement over time) is best managed by waiting for absorption
with no specific treatment.
A symptomatic pneumothorax (>50 %, enlarging, or causing hemodynamic
compromise) should be treated by external drainage.
There are occasions when we would advocate draining a relatively asymptomatic
pneumothorax (e.g., patient almost weaned from ECMO in the hope that reexpansion may facilitate weaning).
Any procedure undertaken on a patient on ECMO should be performed by an
experienced operator and with the coagulation profile optimized.
Consideration should be given to stopping systemic heparin, ensuring platelet
count is high enough and the use of antifibrinolytics such as tranexamic acid.
References
1. Javidfar J, Brodie D, Wang D, Ibrahimiye AN, Yang J, Zwischenberger JB et al (2011) Use of
Bicaval Dual-Lumen Catheter for Adult Venovenous Extracorporeal Membrane Oxygenation.
Ann Thorac Surg 91(6):17631769, Elsevier Inc
2. Brchot N, Luyt C-E, Schmidt M, Leprince P, Trouillet J-L, Lger P et al (2013) Venoarterial
extracorporeal membrane oxygenation support for refractory cardiovascular dysfunction during severe bacterial septic shock. Crit Care Med 41:16161626
3. Combes A, Bacchetta M, Brodie D, Mller T, Pellegrino V (2012) Extracorporeal membrane
oxygenation for respiratory failure in adults. Curr Opin Crit Care 18(1):99104
4. Javidfar J, Brodie D, Iribarne A, Jurado J, Lavelle M, Brenner K et al (2012) Extracorporeal
membrane oxygenation as a bridge to lung transplantation and recovery. J Thorac Cardiovasc
Surg 144(3):716721
5. Peek GJ (2012) Adult respiratory ECMO. In: Annich GM, Lynch WR, MacLaren G, Wilson
JM, Bartlett RH (eds) ECMO: extracorporeal cardiopulmonary support in critical care, 4th
edn. Extracorporeal Life Support Organisation, Ann Arbor, USA
6. Bermudez CA, Rocha RV, Sappington PL, Toyoda Y, Murray HN, Boujoukos AJ (2010) Initial
experience with single cannulation for venovenous extracorporeal oxygenation in adults. Ann
Thorac Surg 90(3):991995, Elsevier Inc
[email protected]
440
7. Garcia JP, Kon ZN, Evans C, Wu Z, Iacono AT, McCormick B et al (2011) Ambulatory venovenous extracorporeal membrane oxygenation: innovation and pitfalls. J Thorac Cardiovasc
Surg 142(4):755761
8. Fragou M, Gravvanis A, Dimitriou V, Papalois A, Kouraklis G, Karabinis A et al (2011) Realtime ultrasound-guided subclavian vein cannulation versus the landmark method in critical
care patients: a prospective randomized study*. Crit Care Med 39(7):16071612
9. Troianos CA, Hartman GS, Glas KE, Skubas NJ, Eberhardt RT, Walker JD et al (2012)
Guidelines for performing ultrasound guided vascular cannulation. Anesth Analg
114(1):4672
10. Hirose H, Yamane K, Marhefka G, Cavarocchi N (2012) Right ventricular rupture and tamponade caused by malposition of the Avalon cannula for venovenous extracorporeal membrane
oxygenation. J Cardiothorac Surg 7:36
11. Javidfar J, Wang D, Zwischenberger JB, Costa J, Mongero L, Sonett J et al (2011) Insertion of
bicaval dual lumen extracorporeal membrane oxygenation catheter with image guidance.
ASAIO J 57(3):203205
12. Chen Y-S, Yu H-Y, Huang S-C, Lin J-W, Chi N-H, Wang C-H et al (2008) Extracorporeal
membrane oxygenation support can extend the duration of cardiopulmonary resuscitation*.
Crit Care Med 36(9):25292535
13. Varon J, Acosta P (2008) Extracorporeal membrane oxygenation in cardiopulmonary resuscitation: are we there yet?*. Crit Care Med 36(9):26852686
14. Field JM, Hazinski MF, Sayre MR, Chameides L, Schexnayder SM, Hemphill R et al (2010)
Part 1: executive summary: 2010 American Heart Association guidelines for cardiopulmonary
resuscitation and emergency cardiovascular care. Circulation 122(18 suppl 3):S640S656
15. Nolan JP, Soar J, Zideman DA, Biarent D, Bossaert LL, Deakin C et al (2010) European resuscitation council guidelines for resuscitation 2010 section 1. Executive summary. Resuscitation
81(10):12191276
16. Stub D, Bernard S, Pellegrino V, Smith K, Walker T, Stephenson M et al (2012) Issues in
establishing the refractory Out-of-hospital cardiac arrest treated with mechanical CPR, hypothermia, ECMO and early reperfusion (CHEER) study. Heart Lung Circ 21:S163
17. Olsson KM, Simon A, Strueber M, Hadem J, Wiesner O, Gottlieb J et al (2010) Extracorporeal
membrane oxygenation in nonintubated patients as bridge to lung transplantation. Am J
Transplant 10(9):21732178
18. Fuehner T, Kuehn C, Hadem J, Wiesner O, Gottlieb J, Tudorache I et al (2012) Extracorporeal
membrane oxygenation in awake patients as bridge to lung transplantation. Am J Respir Crit
Care Med 185(7):763768
19. MacLaren G, Combes A, Bartlett RH (2012) Contemporary extracorporeal membrane oxygenation for adult respiratory failure: life support in the new era. Intensive Care Med
38(2):210220
20. Bizzarro MJ, Conrad SA, Kaufman DA, Rycus P (2011) Infections acquired during extracorporeal membrane oxygenation in neonates, children, and adults*. Pediatr Crit Care Med
12(3):277281
21. Aubron C, Cheng AC, Pilcher D, Leong T, Magrin G, Cooper DJ et al (2013) Infections
acquired by adults who receive extracorporeal membrane oxygenation: risk factors and outcome. Infect Control Hosp Epidemiol 34(1):2430
22. Hsu MS, Chiu KM, Huang YT, Kao KL, Chu SH, Liao CH (2009) Risk factors for nosocomial
infection during extracorporeal membrane oxygenation. J Hosp Infect 73(3):210216, Elsevier
Ltd
23. Sun H-Y, Ko W-J, Tsai P-R, Sun C-C, Chang Y-Y, Lee C-W et al (2010) Infections occurring
during extracorporeal membrane oxygenation use in adult patients. J Thorac Cardiovasc Surg
140(5):1125.e21132.e2
24. Brodie D, Bacchetta M (2011) Extracorporeal membrane oxygenation for ARDS in adults. N
Engl J Med 365(20):19051914
[email protected]
37
441
25. Scott LK, Grier LR, Conrad SA (2006) Heparin-induced thrombocytopenia in a pediatric
patient receiving extracorporeal support and treated with argatroban. Pediatr Crit Care Med
7(3):255257
26. Ranucci M, Ballotta A, Kandil H, Isgr G, Carlucci C, Baryshnikova E et al (2011) Bivalirudinbased versus conventional heparin anticoagulation for postcardiotomy extracorporeal membrane oxygenation. Crit Care 15(6):R275, BioMed Central Ltd
[email protected]
Part VII
Transport of the ECMO Patient
[email protected]
Air Transport: Fixed-Wing

and Helicopter
38
Antonio F. Arcadipane and Gennaro Martucci
38.1
Introduction
Extracorporeal life support (ECLS) for cardiopulmonary or respiratory failure is

one of the most specialized medical and technological activities and is available
only in experienced centers. Without mobile support, it might be impossible to
transport patients needing extracorporeal membrane oxygenation (ECMO), and
transportation of a patient requiring ECMO is often unavoidable if that patient
requires prolonged treatment in a referral center [1].
The increasing cooperation between medical institutions and specialists, broader
indications for ECLS, and progressive miniaturization of ECLS equipment, as well
as the increasing recognition that high volume centers achieve better outcomes, are
likely to increase the need for patient transfer in the near future [2].
Aeromedical transport is used to transfer critically ill patients over large distances or in areas with inadequate road networks or geographical and infrastructural
barriers. It comprises fixed-wing aircraft and rotary-wing aircraft (helicopter). By
definition, ECMO transport is principally a secondary transport: from a referring
hospital to a referral hospital for specialized treatment.
The goal when transporting critically ill patients is to continue current ICU treatment as safely as possible throughout the journey and to be in a position to provide
additional emergency treatment if necessary. This goal must be achieved in the
minimal and unfriendly space of aircrafts, helicopters, and ambulances. So, in a
sense, ECMO transport teams are a mobile extension of ECMO centers.
In this chapter we will address some general concepts and practical aspects to
prepare and conduct a safe ECMO transport, beginning with some history, and
A.F. Arcadipane (*) G. Martucci

Department of Anesthesia and Critical Care, ISMETT
(Mediterranean Institute for Transplantation and Advanced
Specialized Therapies), Via Tricomi 5, Palermo 90127, Italy
[email protected]
445
446
A.F. Arcadipane and G. Martucci
address basic concepts of physiology and crew resource management that are fundamental for all kinds of transport, and even more so in the complex experience of
an in-flight ECMO patient referral.
38.2
History of Interhospital Aeromedical Transport
Aeromedical transport grew out of basic military models into integrated civilian
systems of care [3]. Airplanes began transporting casualties during World War I
when Serbian patients were carried in an unmodified French fighter plane. In 1916,
Dr. Eugene Chassaing suggested to the French government that it modify an aircraft
to allow it to carry two stretchers. The first documented case of helicopter transport
of an injured patient was in Burma during World War II, on April 23, 1944, and the
first dedicated use of helicopters by US forces occurred during the Korean War,
between 1950 and 1953.
The Vietnam War was the definitive showcase for demonstrating the efficacy of
helicopter medical rescue transport: over 400,000 patients were airlifted to hospitals
during this conflict.
Expertise in the use of air ambulances evolved in parallel with the evolution of
aircraft design. The use of military aircraft as battlefield ambulances continues to
grow and develop today in a number of countries (e.g., UH-60 Black Hawk helicopters recently employed by US military during the Iraq War).
From their use in the military, air ambulances were introduced into the civilian
environment, where they are now used for primary scene retrieval and secondary
inter-facility transport in modern health systems.
The initial civilian uses of aircraft as ambulances were probably incidental. In
northern Canada, Australia, and in Scandinavian countries, remote, sparsely populated settlements are often inaccessible by road for months. So air ambulances
quickly established their usefulness in remote locations, while their role in developed areas increased more slowly.
In Los Angeles in 1947, J. Walter Schaefer founded the first air ambulance service in the United States.
The first permanent civil air ambulance helicopter was the Christoph 1, at the
Harlaching Hospital in Munich, Germany, in 1970. The success of this initiative led
to a quick expansion of the concept across Germany to roughly 80 helicopters at
present.
The US Air Force began ECMO transport in late 1985 [4], and in 1994 Critical
Care Air Transport Teams (CCATs) were implemented as an extension of the ICU
resuscitation phase of combat casualty care. In 2005, emerging need for specialized
treatment of ARDS led to the creation of the Acute Lung Injury Response Team
(ALIRT) [5, 6].
In the last few years, a number of international centers have reported increased
experience in interhospital ECMO transport [7, 8].
There are a variety of helicopters used for civilian health emergency medical
systems (HEMS). The most commonly used types are the Bell 206, 407, and 412;
[email protected]
38
Air Transport: Fixed-Wing and Helicopter
447
Eurocopter AS350, BK117, EC130, EC135, and EC145; Sykorsky S76; and the
AgustaWestland 109 and 139. These aircraft are normally configured to transport
one patient and three to five medical personnel, but some can be configured to transport two patients if necessary.
In our ECMO program, at ISMETT, we currently employ an AgustaWestland
139, which can carry one patient, everything needed for ECMO function (Fig. 38.1),
and our ECMO team [9], composed of an anesthesiologist, a perfusionist, and either
a cardiac or thoracic surgeon. In addition, there is room for two other people (e.g.,
fellow and nurse) (Fig. 38.2).
Many aircraft have been used for medical transport: small executive jets or turboprops, though listing of all of them would be beyond the purpose of this chapter.
The Lockheed C-130 Hercules, a military cargo plane, may be considered the best
choice because it allows direct transport of an ambulance, without need for the loading and unloading of the patient or stretcher, and care is provided in the ambulance
Fig. 38.1 ISMETT setup for
control unit and oxygenator
pump unit (Maquet Rotaflow
with PLS circuit) fixed by a
cerified plate on the
helicopter floor
Fig. 38.2 Agusta Westaland

139 cabine view
[email protected]
448
environment. On the other hand, this poses several organizational concerns. Because
it is a military means of transport, there needs to be close collaboration between the
national Air Force and the medical system. It is also time-consuming because the
C-130 has to reach the airport nearest to the hospital, and one has to also consider the
distance from the landing area to the referral center. Finally, C-130 transport imposes
extremely high costs, which are affordable only in select conditions.
38.3
How to Choose the Right Means for the ECMO Patient
The use of air medical transport in Europe, the United States, and Australia differs
dramatically and is influenced by distances between hospitals, the history and
development of medical services, cooperation with the military, geography, and
insurance policies.
The principal advantage of air transport is the shorter journey time, and it is generally
considered one of the preferred modes of secondary transfer [10, 11]. Air transport
should be considered for journeys longer than approximately 80 km, though the velocity
of the craft needs to be adjusted for potential organizational delays and the need for
transfer between vehicles. A useful benchmark would be to use air transport when it
would save at least roughly 2 h compared with ground transport.
Fixed-wing flights constitute a significant portion of aeromedical transport and
are preferred when distances are more than 350400 km, with the considerable
advantage of being able to travel in inclement weather. The biggest disadvantages
of conventional aircrafts are that they require airports to collect and deliver their
human cargo and hospitals are rarely near airports, so this inevitably requires ground
transfers. Takeoff and landing times should also be taken into account.
Medium range aircraft provide adequate working space for patient care and can
fly within a range of 5001,500 km and reach speeds of up to 250 knots. Small jets
have the longest range, up to 5,000 km, and the fastest speed (up to 450 knots).
However, their streamlined shape severely limits cabin space, interfering with the
care for the critically ill patient.
Helicopters typically cruise at 120150 knots (220280 km/h), with a useful
radius of 50400 km. The advantage over fixed-wing aircraft is the ability to operate
from a range of surfaces, and helipads are more diffuse than airports [12]. The maximum advantage is found in areas with underdeveloped roadway systems [13].
As demonstrated during the H1N1 pandemic, delivery of advanced medical technology can be achieved even in remote and underserved areas that present a variety
of geographical barriers. Adoption of an appropriate means of transportation [14]
and use of a multidisciplinary team are the key elements for the success of the
ECMO rescue mission and improve the chances that critically ill patients will not
only arrive safely at the referral hospital, but will also be discharged well enough to
enjoy the trip home [15, 16].
The choice between rotary-wing aircraft (helicopters) or fixed-wing aircraft
should be made following a cost versus benefit analysis, using criteria such as travel
distance, landing infrastructure, weather conditions, and clinical emergency.
[email protected]
38
38.3.1
449
Preparing for the Transport
Principles for avoiding critical events during transport are based on extensive anticipation, effective communication, and patient assessment/stabilization [17]. Though
stabilizing the patient before transport is mandatory, it can be time-consuming, so
the degree of stabilization should be dictated by the estimated time of arrival at the
referral hospital [18].
It is not redundant to say that transferring a patient on ECMO support from one
facility to another should not be viewed as an extraordinary event, but as an integral
part of an ECMO transportation program.
Before transport, all therapeutic options have to be explored with the referring
hospital. In some cases an ECMO team from the referral center could be sent to
personally evaluate the patient and determine the requirements for transport. The
transport then involves a complex interplay between the referring hospital, the
transport team, and the receiving ECMO Center.
In general, early referral is preferred so that the ECMO center can apply rescue
therapies not yet considered impossible in that context before ECMO support starts.
During initial communication between the hospitals, the ECMO team leader
must confirm the availability, at the referring hospital, of some required equipment
and supplies: 24 units of packed red blood cells, an echography machine, a large
sterile surgical drape to completely cover the patient, and a small surgical light for
possible surgical isolation of vessels in case of impossible percutaneous cannulation. These requests are part of our protocol of ECMO placement. As far as requesting blood products, some centers ask for a large quantity of packed red blood cells
to be available, while our protocol calls for only a safely adequate supply because
we operate in an area with a limited availability of blood.
Before starting operations at the referring center, the transport team should
always meet with the patients family to describe the nature of the support and the
type of transport.
38.3.2
Equipment
ECMO devices have become increasingly smaller thanks to centrifugal pump-based

systems, shorter circuits, and miniaturized monitors [19]. It is important to bring a
limited set of disposables instead of storing all possible items in a bulky emergency
kit. Take into account that if stabilization prior to transfer is adequate, this equipment is seldom used. The kit should be inspected prior to each departure. Clear
labeling increases awareness of the site of the material and avoids time-consuming
searches.
The number of oxygen cylinders to be transported depends on the duration of transfer and the size/capacity of the cylinders. Ensure that all cylinders are full before transport. Hoses and adapters must be compatible with those of the facility, ambulance, and
aircraft. The monitor has to be compatible with all equipment available or in use (e.g.,
saturation cable and arterial pressure transducer connection cable).
[email protected]
450
All the ECMO equipment must be certified by the national or international flight
safety agency, or the pilot will not allow you to bring it onboard. The use of a dedicated checklist just before departure is part of a systematic safety check of equipment and drugs.
38.3.3
Before Departure
Stabilization of the ECMO patient involves optimization of DO2, though hemodynamics should be carefully assessed.
The ECMO circuit right function has to be tested, and whenever a doubt arises,
a chest x-ray should be done to check the cannulae position and possible need for
repositioning. Any necessary intervention and/or pharmacologic treatment needs to
be done prior to leaving the referring hospital [20, 21].
All IV access, infusion lines, monitoring equipment, and drains must be checked,
rechecked, and secured. Spare IV access is mandatory, and all unnecessary medication must be discontinued before the transfer. Particular attention must be given to
monitoring the ECMO cannulae positions and to preventing any kinking within the
ECMO circuit, along with a careful connection of the membrane lung to the oxygen
supply.
38.3.3.1 What to Check

Whether there are enough spare syringes to cover the patients needs during
transport
Which medications are indispensable and which can be interrupted temporarily
Whether there is special and adequate tubing, depending on the type of pump
Battery power and the number of wall sockets available
Preparation of medications and syringe pumps will be easier in the secure environment of an ICU than in the close confines of a shaky helicopter or a small aircraft, with possibly poor lighting. In many helicopters, and certainly in small
airplanes, the ceiling is not high enough to guarantee a sufficient spontaneous flow
(because of gravity) of IV fluids. Therefore, pressure bags or additional infusion
pumps may be needed. Glass bottles should be avoided. Beware of air embolism
from pressurized intravenous lines.
A so-called time out for the transport team before departure is now successfully
being used as a security check. Items on a checklist should include the following:
Patients name and diagnosis.
Comprehensive documentation, including radiography and laboratory exams: the
patients medical chart is a legal obligation and an integral part of patient care.
Destination unit, shortest route from the landing pad/ambulance point of
departure.
Final check of weather conditions.
Name and phone numbers of physician responsible for the transfer.
Duration of the transfer and subsequent calculation of required quantities of
medical gases, medications, and other equipment.
Possible Plan B for transfer.
[email protected]
38
38.3.4
451
During Transport
In helicopters, internal noise levels are frequently >95 dB, so that normal conversation is impossible. Auscultation is impracticable, as is reliance on audible monitor
alarms. Earplugs have to be used for all patients, regardless of conscious state, to
prevent hearing damage. Cabin lighting may be poor because strong lighting can
distract the aircrew and adversely affect pilot night vision. All this may make monitors, cyanosis, veins, and patient movement difficult to see. Vibration is greatest at
takeoff and landing and may induce pain in unstable fractures; it also makes accurate adjustment of fluid infusion rates difficult [18].
Often fixed-wing aircraft suffer from less noise but are characterized by stronger
acceleration/deceleration forces during landing/takeoff.
Incidents related to transport are mainly associated with equipment-related problems, vascular line management, inadequate monitoring, and inadequate communication among staffs.
38.3.4.1 What to Monitor

Three-lead (or more) ECG with heart rhythm and ST segment monitoring.
Invasive pressures (if too time-consuming or not available, noninvasive blood
pressure is acceptable in case of emergency cannulation in life-threatening conditions; but before transport the patient should be fully monitored, and it is
worthwhile to wait for invasive arterial pressure monitoring).
Hemoglobin oxygen saturation. For long-term transport, SvO2 is mandatory either by
full optioned Cardiohelp System (Maquet) or portable point of care ABG machine.
End-tidal capnography, if available.
Temperature: airplanes that reach higher altitudes have greater temperature
drops, with risk of hypothermia for a critically ill patient.
Complete assessment of the patient and equipment is mandatory throughout the
transport. In addition, there are some particularly crucial things to consider:
Patient movement to and from the transport vehicle, including to/from the
stretchers and helicopter
Securing of the patient, the circuit, and oxygen supply during takeoff and landing
Effects of altitude on membrane gas exchanges
Recognition and management of in-transport emergencies
38.3.4.2 Effects of Altitude
With an increase in altitude, temperature and barometric pressure decrease, and the
partial pressure of oxygen drops. This effect also influences the gas exchange
capacity of the membrane.
Furthermore, according to Boyles Law, as the atmospheric pressure falls, gas
volume increases. At a typical helicopter altitude of 2,000 ft, this increase is about
8 %. In the case of fixed-wing transport, the problem is less essential but still present. Pneumothoraces, pneumoencephaly, bubble emboli, and air trapped in the
abdominal cavity will all expand in proportion to the drop in cabin pressure. All
surgical drains should be unclamped and patent. Air in the endotracheal cuff also
expands, with possible tracheal injury. ETT cuffs should be filled with saline [22] or
[email protected]
452
have their pressure rigorously checked and adjusted in-flight. Pulmonary artery
catheter balloons should be fully deflated.
These are just some of the considerations that dictate complete stabilization and
assessment of the potential ECMO patient before transport, though this should be
done for all transported patients.
38.3.4.3 Acceleration/Deceleration Forces

When a supine patient is exposed to the force of sustained acceleration or deceleration in the longitudinal plane, blood flow is diverted to either the feet or the head
depending on the patients orientation. Theoretically, this blood flow diversion can
cause cerebral underperfusion or excess perfusion (if the force exceeds cerebral
autoregulation), with a risk of deterioration in patients with an intracranial pathology. Cardiac preload and afterload can also be influenced by such forces, which can
cause further deterioration in hemodynamically compromised patients.
The main problem, in any event, is related to circuit problems. During air transport the circuit and patient must be well secured to the aircraft and watched carefully during takeoff and landing to avoid shifting in the position of the cannula or
kinking and accidental dislodgment of the circuit. Movement of the patient is recognized as one of the more dangerous practices in critical care.
A safety procedure in an unfriendly environment is to always check all the monitored parameters without losing any of them during loading from one means of
transport to another.
Prevention of hypothermia, or restoration of normothermia, must be ensured in
all patients through the use of the so-called space blankets, electrical blankets, and/
or infusion warming systems.
38.3.4.4 Safety
Safety of the patient and aeromedical team is paramount. Staff should undergo training
in aeromedical transport. For interhospital helicopter transport, additional training is
required so the physician can adequately provide critical care support outside the ICU.
No pressure must be put on the aircrew to alter their normal safety procedures. The
captain has the final say on whether the conditions are suitable for flight, independent of
the patients condition. On the other hand, a patient who has just been put on ECMO is
usually stable enough to wait until easier and safer transport can be carried out.
Acknowledgment We are indebted to Warren Blumberg, science editor in ISMETTs Language
Services Department, for his help in editing the text.
References
1. Wagner K et al (2008) Transportation of critically ill patients on extracorporeal membrane
oxygenation. Perfusion 23:101106
2. Patroniti N et al (2011) The Italian ECMO network experience during the 2009 influenza
A(H1N1) pandemic: preparation for severe respiratory emergency outbreaks. Intensive Care
Med 37:14471457
[email protected]
38
453
3. Martin T (2006) Aeromedical transportation. A clinical guide, 2nd edn. Ashgate Company,
Aldershot/Hampshire
4. Cannon JW et al (2012) Transport of the ECMO patient: from concept to implementation. In:
Annich G (ed) ECMO: extracorporeal cardiopulmonary support in critical care, 4th edn.
Extracorporeal Life Support Organization, Ann Arbor, pp 451478
5. Midla GS (2007) Extracorporeal circulatory systems and their role in military medicine: a
clinical review. Mil Med 172(5):523526
6. Dorlac GR et al (2009) Air transport of patients with severe lung injury: development and
utilization of the acute lung rescue team. J Trauma 66:S164S171
7. Schaible T et al (2010) A 20-year experience on neonatal extracorporeal membrane oxygenation in a referral center. Intensive Care Med 36:12291234
8. Forrest P et al (2011) Retrieval of critically ill adults using extracorporeal membrane oxygenation: an Australian experience. Intensive Care Med 37:824830
9. DAncona et al (2011) Extracorporeal membrane oxygenator rescue and airborne transportation of patients with influenza A (H1N1) acute respiratory distress syndrome in a Mediterranean
undeserved area. Interact Cardiovasc Thorac Surg 12:935937
10. Hinds CJ et al (2008) Principles of safe secondary transport. In: Intensive care: a concise textbook, 3rd edn. Saunders, Edinburgh/New York, pp 543545
11. Michaels AJ et al (2013) Pandemic flu and the sudden demand for ECMO resources: a mature
trauma program can provide surge capacity in acute critical care crises. J Trauma Acute Care
Surg 74(6):14931497
12. McVey J et al (2010) Air versus ground transport of the major trauma patient: a natural experiment. Prehosp Emerg Care 14:4550
13. Diaz MA et al (2005) When is the helicopter faster? A comparison of helicopter and ground
ambulance transport times. J Trauma 58:148153
14. Taylor CB et al (2010) A systematic review of the costs and benefits of helicopter emergency
medical services. Injury 41:1020
15. Noah MA et al (2011) Referral to an extracorporeal membrane oxygenation center and mortality among patients with severe 2009 influenza A (H1N1). JAMA 306(15):16591668
16. Michaels AJ et al (2013) Adult refractory hypoxemic acute respiratory distress syndrome
treated with extracorporeal membrane oxygenation: the role of a regional referral center. Am J
Surg 205:492499
17. Beckmann U et al (2004) Incidents relating to the intra-hospital transfer of critically ill patients.
An anlaysis of the reports submitted to the Australian Incident Monitoring Study in Intensive
Care. Intensive Care Med 30:15791585
18. Waldmann C et al (eds) (2008) Oxford desk reference: critical care. Oxford University Press,
Oxford/New York, pp 580581
19. Arlt M et al (2008) First experience with a new miniaturized life support system for mobile
percutaneous cardiopulmonary bypass. Resuscitation 77:345350
20. Linden V et al (2001) Inter-hospital transportation of patients with severe acute respiratory
failure on extracorporeal membrane oxygenation national and international experience.
21. Cornish JD et al (1986) Inflight use of extracorporeal membrane oxygenation for severe
neonatal respiratory failure. Perfusion 1:281287
22. Bassi M et al (2010) Endotracheal tube intracuff pressure during helicopter transport. Ann
Emerg Med 56(2):8993
[email protected]
Ground Transport: Ambulance
39
Stefano Isgr, Roberto Rona, and Nicol Patroniti
39.1
Introduction
Interhospital transportation of severe critically ill adult/paediatric patients is a highrisk procedure due to unstable clinical condition, lack of diagnostic/therapeutic
tools, progression of the severity of disease and the consequences related to the
equipment malfunction. These patients often require technology, staff and supplies
that outweigh local resources and need to be centralized to regional tertiary care
centres. The decision to set up transportation of such patients needs careful evaluation of risks and benefits [13].
ECMO instituted at referral centres allows stabilization of an unstable patient,
otherwise unmovable, and thus a safer transportation to the targeted destination.
Nevertheless, adding an ECMO system makes transportation more complex, requiring a specialized multidisciplinary ECMO team, trained and equipped to stabilize
patients at the referring hospital, to apply and manage ECMO and to assist patients
during transportation up to the tertiary care facility. Several large ECMO ground
transportation international case series have been published in recent years, both for
adult and paediatric patients [315]. Finally this method proved to be of utmost
importance during the recent virus AH1N1 pandemic flu outbreak in Italy [16].
S. Isgr, MD (*) R. Rona, MD

Urgency and Emergency Department, San Gerardo Hospital,
via Pergolesi 33, Monza 20900, Italy
N. Patroniti, MD
Milano-Bicocca University, San Gerardo Hospital,
[email protected]
455
456
S. Isgr et al.
Table 39.1 ECMO patient ground (ambulance) transportation: characteristics

Advantages
Not influenced by weather
Available night and day
Airports/heliport and secondary
transportation not required
No need for a flight-trained medical
emergency team
More units available at the same time
Disadvantages
Shorter maximum distance coverage
Safety of transportation negatively affected by street
roughness, other vehicles, traffic
Slower
Ambulance ground transportation presents specific issues which will be addressed

in this chapter. For a comprehensive analysis of air transportation issues, we remand
to specific chapters.
39.2
Ground vs Air Transportation
When planning the logistics of transport, the choice of the means of transport is the
first decision to take several factors must be taken into account when considering
the use ambulance instead of fixed wing/helicopter air transportation:
1. Distance from referral to tertiary care institution: the average distance covered
varies among centres, often being on average less than 100 km. Nevertheless,
patients have been safely moved up to 300500 km ground distance [5, 8, 13].
2. Air or ground trained crew availability.
3. Geographical obstacles and/or availability of adequate roads.
4. Local resource availability.
5. Specific patient issues contraindicating air transportation.
6. Weather conditions [6, 7, 15].
The ambulance transportation advantages/disadvantages are summarized in
Table 39.1. The vehicle employed is a mobile ICU unit, often customized to be used
for this type of transportation and equipped with an ICU ventilator and an adequate
monitoring system. Ambulance is adapted to provide extra space with enlarged
power, oxygen and fuel supplies. Ambulance special technical requirements are
summarized in Table 39.2. When the patient is moved by air transportation, ambulance is needed to cover short distances between the airport/heliport and the targeted
hospital; alternatively ambulance can be loaded on large fixed-wing aircraft avoiding the time-consuming and risky procedure of on-/offloading the patient [6].
39.3
Equipment
Custom-made mobile steel carts or cradles have been developed by each ECMO
centre to easily load all the equipment on the patient and to allow safe loading/
unloading on the ambulance. Our custom-made steel cart (Fig. 39.1) extends in
[email protected]
39
457
Table 39.2 ECMO transportation ambulance technical requirements

Requirement
Adequate space (enlarged)
Reinforced ambulance stretcher
and stretcher holders
Reinforced shock adsorbers
Increased compressed oxygen
storage
Increased battery capacity
Increased fuel tanks
Reason
To support increased weight (ECMO pump, ECMO heater,
ICU ventilator)
To mitigate the effect of street roughness on the devices
employed (especially ECMO pump)
To support both ventilator MV and ECMO oxygenator
sweep gas at iFO2 100 %
To support air heater/cooler, ICU ventilator, ICU monitor,
infusional pumps, ECMO pump, ECMO heater
To avoid the need for fuel supplying during transportation
Fig. 39.1 ECMO

transportation custom-made
steel cart at the end of patient
preparation. Special attention
should be paid to weight
distribution on the stretcher
height (100 cm) and provides two supports on two levels. It mounts on the spinal
board over the patients feet and allows transportation of the ECMO unit (driving pump, heater and console), an ICU ventilator and an ICU monitor. Moreover,
a especially designed iron post allows holding of infusional pumps, pressure
bags and fluids. A power strip is fixed on a post to organize electrical cables.
Patient-cart unit is then secured and loaded on ambulance stretcher (maximum
tolerance 250 kg). When loaded on the ambulance (Fig. 39.2), it may be secured
with additional ropes to avoid rolling of the patient-cart unit and displacement
of the equipment.
All the equipment needed for cannulation, priming of ECMO circuit and emergency procedures are listed in checklists and organized into backpacks. Backpacks
and electrical equipment must be checked before departure, and materials must be
replaced immediately after use.
[email protected]
458
S. Isgr et al.
Fig. 39.2 Patient-cart unit

loaded on the special ICU
unit and secured. Of note, up
to three ECMO team
members can sit on the right
of the patient during transport
(ECMO pump, ventilator and
monitoring), while other two
members can sit at the head
of the patient (drug infusions
and artificial airway
management)
39.4
Pre-transport Preparation
On arrival the ECMO team gets clinical informations from the referring medical staff
and takes charge of the patient. Its important to avoid unnecessary delays of referral;
the ECMO team will attempt to get the most stable clinical condition after ECMO is
started and the patient is positioned supine on the stretcher. The cart will be placed
on the stretcher and carefully loaded with special attention to weight-balance the
[email protected]
39
459
cart. All the equipment will be checked again while on battery energy supply and
tightly secured. Only after these procedures the patient will be moved to the ambulance. Moving the patient to the ambulance, oxygen to both the ECMO membrane
lung and the ventilator will be provided by at least two small oxygen tanks. Power
and gas supplies will be replaced by ambulance power inverter and gas tanks once
loaded.
39.5
Team
The ECMO team is multidisciplinary and should be staffed by experienced and

trained personnel. The team must be able to apply and run ECMO via peripheral
venous-venous or venous-arterial cannulation and to front serious adverse events
(see below). The number and composition of the group may vary from a minimum
of three [5, 17] up to 15 [7, 10, 18], depending on the distance of the referring hospital (i.e. need for two drivers, support vehicles) and the presence of trainees, logistics and local practices. More frequently, the team includes at least a perfusionist to
provide ECMO circuit set-up and management, two ECMO skilled physicians (surgeons and/or critical care physicians or both), one or two critical care nurses and
drivers. Although Seldinger cannulation technique is preferable, many groups
include a cardiac surgeon in order to provide, when needed, direct vascular cutdown
or rescue thoracotomy.
39.6
Adverse Events
The main reported adverse events may be grouped into equipment/vehicle-related

technical failure or patient-related complications. To our knowledge no fatal adverse
events have been reported during ground ECMO patient transportation. In 2002
Foley et al. [7] reported an incidence rate of minor adverse events of 17 % during
100 ECMO transportation (80 % ground transportation). The most frequent problem reported is due to acute loss of power supply due to battery or vehicle power
supply failure [8, 10, 17, 18], even in the presence of back-up battery pack [7].
Other frequent complications are ML damage [5, 7], circuit tubing leakage [7] and
pump/circuit uncoupling. Linden et al. [6] reported also a failure of an ambulance
shock adsorber. Inadvertent ECMO cannula removal/dislocation is certainly the
most feared accident, but, to our knowledge, it has never been reported in out-ofhospital setting. Also tracheal tube, IV lines and thoracic/abdominal drainages may
be dislodged, requiring a stopover and dangerous emergency manoeuvres.
Additional length of the ECMO circuit may be of help in positioning the patient.
Correct management necessitate experience in transport medicine and careful
patient/equipment loading/unloading, manipulation and handling [3].
Technological improvements in monitoring, ventilators and ECMO (lighter,
shock/vibration-proof transportation centrifugal pumps, improved capacity battery
packs) may provide safer transportation of these patients.
[email protected]
460
S. Isgr et al.
References
1. Vincent J-L, Abraham E, Kochanek P et al (2011) Textbook of critical care. In: Chapter 225:
Transport medicine. Elsevier; Saunders, USA
2. Warren J, Fromm RE, Orr RA et al (2004) Guidelines for the inter- and intrahospital transport
of critically ill patients. Crit Care Med 32:256262
3. Annich GM, Gail M, Annich E (2012) ECMO: extracorporeal cardiopulmonary support in
critical care, Red book. Extracorporeal Life Support Organization, USA
4. Bennett JB, Hill JG, Long WB et al (1994) Interhospital transport of the patient on extracorporeal cardiopulmonary support. Ann Thorac Surg 57:107111
5. Rossaint R, Pappert D, Gerlach H et al (1997) Extracorporeal membrane oxygenation for
transport of hypoxaemic patients with severe ARDS. Br J Anaesth 78:241246
6. Lindn V, Palmr K, Reinhard J et al (2001) Inter-hospital transportation of patients with
severe acute respiratory failure on extracorporeal membrane oxygenationnational and international experience. Intensive Care Med 27:16431648
7. Foley DS, Pranikoff T, Younger JG et al (2002) A review of 100 patients transported on extracorporeal life support. ASAIO J 48:612619
8. Huang S-C, Chen Y-S, Chi N-H et al (2006) Out-of-center extracorporeal membrane oxygenation for adult cardiogenic shock patients. Artif Organs 30:2428
9. Zimmermann M, Bein T, Philipp A et al (2006) Interhospital transportation of patients with
severe lung failure on pumpless extracorporeal lung assist. Br J Anaesth 96:6366
10. Coppola CP, Tyree M, Larry K et al (2008) A 22-year experience in global transport extracorporeal membrane oxygenation. J Pediatr Surg 43:4652; discussion 52
11. Haneya A, Philipp A, Foltan M et al (2009) Extracorporeal circulatory systems in the interhospital transfer of critically ill patients: experience of a single institution. Ann Saudi Med
29:110114
12. Javidfar J, Brodie D, Takayama H et al (2011) Safe transport of critically ill adult patients on
extracorporeal membrane oxygenation support to a regional extracorporeal membrane oxygenation center. ASAIO J 57:421425
13. Isgr S, Patroniti N, Bombino M et al (2011) Extracorporeal membrane oxygenation for interhospital transfer of severe acute respiratory distress syndrome patients: 5-year experience. Int
14. Chenaitia H, Massa H, Toesca R et al (2011) Mobile cardio-respiratory support in prehospital
emergency medicine. Eur J Emerg Med 18:99101
15. Clement KC, Fiser RT, Fiser WP et al (2010) Single-institution experience with inter-hospital
extracorporeal membrane oxygenation transport: a descriptive study*. Pediatr Crit Care Med
11:509513
16. Patroniti N, Zangrillo A, Pappalardo F et al (2011) The Italian ECMO network experience
during the 2009 influenza A(H1N1) pandemic: preparation for severe respiratory emergency
outbreaks. Intensive Care Med 37:14471457
17. Bulpa P, Evrard P, Dive A et al (2002) Inter-hospital transportation of patients with severe
acute respiratory failure on extracorporeal membrane oxygenation. Intensive Care Med 28:802
18. Rosengarten A, Elmore P, Epstein J (2002) Long distance road transport of a patient with
Wegeners Granulomatosis and respiratory failure using extracorporeal membrane oxygenation. Emerg Med (Fremantle) 14:181187
[email protected]
Part VIII
Conclusion
[email protected]
Newer Indications and Challenges
40
Marco Giani, Alberto Zanella, Fabio Sangalli,

and Antonio Pesenti
Extracorporeal gas exchange was first developed in the 1940s to replace heart and
lung function in the context of cardiac surgery. Ever since, a great technical development occurred allowing a widespread use of this technique as cardiac and respiratory support therapy. In this chapter we will present some possible future perspectives
of extracorporeal gas exchange. Technical aspects regarding materials and coatings
of extracorporeal circuits will not be discussed in detail.
Since the 1970s Kolobow and Gattinoni have conceptually separated the function of gas exchange of the lung into oxygenation and CO2 removal [1]. The same
physiology also applies to extracorporeal gas exchange: blood passing through
the membrane lung directly absorbs oxygen and hemoglobin is rapidly saturated;
hence, oxygenation requires only a small oxygen flow, whereas to completely
substitute the patient lung function, a conspicuous blood flow is needed.
Oxygenated blood contains 150200ml of oxygen per liter; if venous blood is
70% saturated, an ECMO system can approximately transfer 4060ml of oxygen
per liter of blood flow. On the contrary, CO2 removal requires a substantial gas
flow in order to maintain the maximum blood/air CO2 gradient to quickly transfer
CO2 from the extracorporeal blood into the open air. Since a great amount of CO2
is available (550600ml of CO2 per liter of blood, most as bicarbonate ions) in a
small volume of blood, even a relatively low blood flow is adequate. For these
reasons a high extracorporeal blood flow is mandatory for a hypoxic patient,
whereas a low-flow extracorporeal system perfectly suits hypercapnic patients.
M. iani A. anella (* A. Pesenti

Dipartimento di Scienze della Salute, University of Milano-Bicocca,
Ospedale San Gerardo Nuovo dei Tintori, via Donizetti 106, Monza 20900, Italy
F. Sangalli
De artment o Anaesthesia an ntensi e Care Me i ine, an erar o os ital,
ni ersit o Milano i o a, Via Pergolesi , Mon a
, tal
[email protected]
463
464
M. Giani et al.
Technological development of extracorporeal systems addressed both the highflow systems (venoarterial or venovenous extracorporeal systems for refractory
hypoxia) and the less invasive low-flow systems.
40.1 Extracorporeal Oxygenation and Future Perspectives

The amount of oxygen transferred through a membrane lung (VO2ML) is described
by the following equation:
VO 2 ML = BF * ( COUT O 2 C IN O 2 )

where BF is expressed as liters per minute, COUTO2 is the oxygen content (ml O2 per
liter of blood) in the reinfusion limb of the circuit, and CINO2 is the oxygen content
of the blood before entering the membrane lung.
BF: The main determinant of VO2ML is the blood flow (BF); in the absence of
CINO2 variations (i.e., oxygenated blood recirculation or increase in venous oxygen content), the VO2ML is proportional to the extracorporeal blood flow.
Therefore, the choice of the size of the cannulae should be carefully evaluated to
match the patient needs (degree of hypoxia, cardiac output, total O2 consumption, etc.).
COUTO2: The performance of modern membrane lungs is nearly optimal: blood in
the reinfusion limb of the circuit is fully saturated, and when high FiO2 is used,
the o gen artial ress re (PoutO2 rea hes er high al es (
mm g .
By increasing the FiO2 at the membrane lung, we often obtain a negligible
in rease o hemoglo in sat ration on the ontrar the s stantial rise o PoutO2
determines a significant increase of the fraction of the oxygen found in the dissol e orm an in rease o the PoutO2 from 100 to 600 determines an increase of
VO2ML of 15ml O2 per liter of blood flow; at high blood flows (i.e., 4l/min) this
ma a o nt or
o the total o o gen ons m tion. en i the
oxygenation performance of modern MLs is already satisfying, a modest
decrease in gas exchange efficiency is commonly seen after some days of use;
any further development in membrane coating might allow to maintain the initial
performance over days.
CINO2 A hara teristi ro lem o eno eno s CM (VV CM s stems is
recirculation, which is defined as the fraction of the oxygenated blood flow that
reenters the rainage line o the ir it. A high re ir lation ra tion s all
manifests with a high CINO2; as it is shown in the formula above, any increase
in CINO2 may significantly impair the VO2ML. For example, a recirculation
fraction of 50%, by reducing the difference COUTO2CINO2, will approximately halve the VO2ML, functionally wasting half of the blood flow of our
ECMO system. The main parameters that may affect recirculation are the position of the cannulae, their shape and features, the ratio BF/cardiac output, and
the vena cava diameter.
[email protected]
40 Newer Indications and Challenges
465
The recirculation fraction during VV ECMO has always been difficult to

quantify; methods based on oxygen as a tracer have proved to be unreliable as
they are based on the rough assumption that, in the absence of recirculation, the
oxygen content of the venous blood in the drainage cannula equals the theoretical mixed venous content of O2. Most of the experience for the calculation of
recirculation comes from nephrology where, in the early 1990s, besides the classical two-/three-needle methods based on blood urea nitrogen dilution, new techniques based on saline bolus dilution were introduced [2, ]. These techniques
proved to be reliable and were successfully applied to the VV ECMO [46], but
they have not been largely introduced in clinical practice yet.
n the earl
s, ol ro lene mem rane l ngs (PPM re la e the ol silicon rubber membrane lungs (SRML) because of their lower resistance to blood
flow, easier priming with low volumes, and higher efficiency in gas exchange.
o e er, hen PPM
ere se or more than h, lasma lea age o rre . The
te hnologi al e elo ment le to ne ol meth l entene (PMP mem rane l ngs
[7], which maintained the superior performance compared to SRML [8] without the
plasma leakage problem.
Moreover, the development of heparin-coated ECMO circuit surfaces allowed to
reduce the need of anticoagulants and to avoid most of the systemic hypercoagulability secondary to the activation of the clotting cascade. The possibility of performing short-term ECMO treatments without anticoagulants extended the use of this
technique to the trauma patient with acute cardiorespiratory compromise. Recently
new ultra-compact ECMO systems have been developed [9]; these systems, besides
improved portability, represented a significant safety improvement as they feature
continuous monitoring of arterial and venous line pressure, blood temperature,
hemoglobin, and saturation of the blood entering the venous limb of the circuit, useful to quickly detect high recirculation fractions during cannulae positioning.
40.2 Extracorporeal CO2 Removal and Future Perspectives

Almost all ARD atient re ire h er entilation 10] because the severe hypoxia
is most frequently accompanied by an inefficient CO2 remo al. A min te entilation
(VE) as high as 15l/min is commonly employed, and since the functional residual
capacity (FRC) estimated for these patients is around 0.50.7l [11], the ratio
et een V an en FRC is higher than , hereas in a health man (FRC . l,
V
l min s h ratio is lo er than . This ele ate me hani al entilation nee is
e to the in rease ea s a e o ARD atients, ro ing that ARD is also a
microvascular disease: pulmonary pressures indeed are often increased and the
development of right heart failure is common. Dead space is so significant that it is
probably the strongest predictor of mortality [12 in ARD .
igh me hani al entilation means also higher han e to in e entilation
induced lung injury (VILI); despite the proven efficacy in terms of patient outcome
[email protected]
466
M. Giani et al.
of a reduction of the tidal volume (TV) from 12 to 6ml/kg [10], Terragni etal.
recently demonstrated that also a 6ml/kg TV can induce severe lung hyperdistension [ . A rther re tion in TV, itho t a on omitant rise in C 2, can be
achieved by means of extracorporeal CO2 removal technology [14 . As e laine in
the introductory paragraph, CO2 removal can be performed with blood flows far
lower than the ones used for supporting oxygenation. Since lower blood flow also
means smaller cannulae, hence lower invasiveness, today several extracorporeal
CO2 removal (ECCO2R) systems are available and many efforts are made to achieve
a substantial CO2 removal from an extracorporeal blood flow close to the one commonl em lo e
ring ontin o s eno eno s hemo ltration (CVV
1417].
o e er, es ite ma or im ro ements in e tra or oreal te hnolog , mo erate high
extracorporeal blood flow rates (5001,000ml/min) [11, 12] are still required to
remove a significant fraction (e.g., 50%) of the total CO2 production of an adult
patient. This implies the use of large caliber vascular catheters and specific technical
e ertise an e i ment that limits the i es rea i sion o this te hni e. An
ideal ECCO2R system should remove about 50% of the total CO2 production of an
adult patient from a blood flow of 200250ml/min, achievable with standard
double-lumen dialysis catheters; this target can only be attained by drawing from
the considerable fraction of CO2 present under the form of bicarbonate ions.
The CO2 transfer in the ML is driven by the transmembrane delta pressure of
CO2, which is related to the dissolved CO2: only a minor portion (about 5%) of the
blood total CO2 content. The majority of CO2 (~90%) is under the form of bicarbonate ions, which are in equilibrium with the dissolved fraction as follows:

CO 2 + H 2 O H 2 CO3 HCO3 + H +
The extracorporeal CO2 removal has been attempted with two different strategies
[18]: the first is through hemodialysis, which removes CO2 both as dissolved CO2
and as bicarbonate ions; the other is through ventilation of blood, which removes
CO2 in gas form. The first solution requires a hemofilter and implies the replacement o the remo e i ar onate ith other anions (a etate, h ro i e, T AM to
prevent electrolyte and metabolic imbalance; the latter requires only a membrane
lung. Both these strategies can be enhanced by shifting the equilibrium between the
CO2 forms: alkalization increases the bicarbonate form and may enhance hemofiltration o C ; acidification increases the gaseous form and may enhance the ML
performance. In the past both these strategies have been attempted [19 ] as short-
term treatment in animal models and proved to be effective in increasing the CO2
removal, but none proved to be safe; hence, for more than 20 years, these strategies
have been locked in a drawer. In the recent times, with new technologies, these
enhanced CO2 removal techniques have been rediscovered and showed a great
potential as a tool to improve extracorporeal CO2 removal.
Cressoni etal. [24] achieved a 50% reduction of mechanical ventilation through
loo ltra ltration. A onsistent ra tion o loo ltra ltrate as remo e at the
same time the lost fluid was replaced by a solution containing no bicarbonate. By
this arrangement, the hemofilter removes Na+ C , whereas Na+ is reinfused
with the replacement solution; the net balance is the CO2 remo al. anella et al. 25]
[email protected]
467
5mEq/min
pH
Lactic acid infusion
VCO2ML
104 ml/min (no acid)
>> 171 ml/min (5mEQ/min)
6.91 0.05
pCO2
137 19
HCO3
27 2.1
ML
No acid
pH
5mEq/min
No acid
7.36 0.003 7.36 0.03
5mEq/min
pCO2
57 7
57 7
pH
HCO3
32 4
32 4
pCO2
14 5
21 12
HCO3
23 4
16 3
7.85 0.09 7.51 0.14
Fig. 40.1 Blood gas parameters and CO2 removal in the extracorporeal circuit (BF 500ml/min,
GF 10l/min) with and without acidification (5mEq/min of l-lactic acid) [25]
showed that acidification with a concentrated lactic acid solution may determine a
substantial increase of the extracorporeal CO2 removal of the membrane lung
(VCO2ML). In a low-flow extracorporeal system (500ml/min), the infusion of lactic
acid at 125mEq/min increased the VCO2M
, , an
, res e ti el .
At the highest rate o a i i ation, the VC 2ML reached 171ml/min, a value close
to the total CO2 ro tion o an a lt man. An e am le o loo gas arameters in
the extracorporeal circuit during acidification is shown in Fig.40.1. Preliminar ata
[26] in the animal model showed that a 48-h infusion of 2.5mEq of lactic acid in a
low-flow ECCO2R system (BF 250ml/min) was safe and efficient in determining a
stable increase (6080%) of VCO2ML compared to the same setting without acid
infusion. Since lactate is a metabolizable compound and a source of calories, to
avoid overfeeding and a consequent increase in total CO2 production, the calories
in se as la tate sho l e onsi ere in the ail total alorie inta e. An a
ring lactic acid infusion, when the total caloric input was maintained constant, the
resulting increase of total VCO2 was negligible [27 . Promising res lts ha e also
been achieved with a system based on ventilation of acidified dialysate [28]. In this
series of experiments, no complication ascribable to lactic acid infusion was reported.
Future development may imply the use of other metabolizable acid such as citric
acid, which may play a dual role of blood acidification and regional anticoagulation.
Another o tion ma e the se o non meta oli a le a i (e.g., h ro hlori a i ,
which however requires more complex dialysis system to remove the strong anions
and to prevent the resulting progressive metabolic acidosis.
Recently new ECCO2R systems have been developed [29]; new features include
an integrated gas exchange membrane and centrifugal blood pump and the possibility to continuously monitor the VCO2ML. These systems however extract only
ml o ar on io i e er min te rom a loo o o
ml min. The
development of ultra low-flow-enhanced ECCO2R systems may allow a widespread
[email protected]
468
M. Giani et al.
diffusion of this technique, further reducing invasiveness and allowing its application
even outside the intensive care unit setting [ ].
40.3 Clinical Management Challenges

40.3.1 ECMO for Refractory Cardiac Arrest (RCA)
ECLS represents a valuable tool for the support of cardiac and respiratory functions.
Obviously this is true only when top-level standard care is provided together with
the extracorporeal support.
In particular, standardized protocols of care should be implemented in patients
s ering rom RCA
timal stan ar A
are sho l e g arantee an onstit tes the on itio
sine qua non for the institution of ECLS. Of utmost importance is the adequacy of chest compressions, with the minimization of interruptions from the
moment of collapse to the start of extracorporeal support. The use of mechanical chest compression devices seems a promising option to achieve this objective (1).
Strict protocols should indicate the inclusion and exclusion criteria, leaving to
the attending clinician the possibility to deviate in specific circumstances on a
ase
ase asis. n arti lar, o t o hos ital RCA i tims sho a ar orst
rognosis hen om are to in hos ital RCA atients ( . A rate riteria are
fundamental to avoid futile implantations.
o o time (i.e., the time rom olla se to the eginning o CPR is a n amental determinant of neurologic outcome. Low-flow time (from the beginning
o CPR to R C or C start is also rele ant an sho l e e t to a minim m
(ideally below 45). Strategies should be implemented to pursue this objective.
Implantation in places of the hospital other than theaters and ICUs is common,
while only few reports have been published on the out-of-hospital implantation
and at the moment this seems, in our opinion, not advisable for several clinical
an organi ational reasons. An e e tion o l e re resente
mass e ents
where a cardiac event is likely (e.g., marathons), where an ECMO-capable
advanced medical facility might be implemented.
Neurologic damage represents a frequent cause of morbidity and mortality in
this setting. Secondary ischemic insults represent an important cause of permanent neurologic damage. It is hence fundamental to standardize care (therapeutic
hypothermia, seizures management, sedation, etc.) and neurologic monitoring
(clinical, radiological, EEG, somatosensory evoked potentials, as discussed in
the specific chapter in this book).
A itional re i tors o ne rologi o t ome sho l e so ght an in estigate
in order to minimize futile assistances and give proper indications for sustaining
support.
Early coronary angiography should be performed in all patients with an ischemic
C ost R C an in atients ith a ossi le ar ia a se o RCA hen no
other cause is evident. It should be considered and weighed against possible
[email protected]
469
contraindications in all patients with an apparently normal ECG post-ROSC, as

several studies revealed a considerable amount of patients demonstrating
occluded coronary arteries despite a silent post-ROSC ECG.
40.3.2 Ischemia/Reperfusion Injury

Reperfusion injury represents an important cause of morbidity, particularly in cariogeni sho an RCA atients. n othelial an mito hon rial amage an res lt
into a s stemi in ammator res onse lea ing to ro o n istri ti e sho . A art
from optimizing pre-ECMO perfusion as possible, many strategies have being
investigated to minimize such damage, but none, apart from therapeutic hypothermia, is now in clinical practice.
The administration of NO donors constitutes a promising strategy to minimize
ischemia/reperfusion injury in many organs and is under investigation in different
settings ( ho
.
Centri gal CM
m s ro i e a ontin o s o attern. Arterial lsatilit
is hen e lost hen nati e ar ia o t t is ramati all re e . P lsatile o
showed a protective effect on endothelial function and inflammation in several studies. Various strategies are commonly implemented to maintain a certain degree of
pulsatility (see Chap. 9 for further details), but their impact on outcome has still to
be determined.
40.3.3 Evaluation and Reconditioning of Organs for Transplant

rain ea CM atients ( D, heart eating onors re entl e ome organ
donors, and in these patients the extracorporeal support guarantees adequate organ
perfusion until the harvesting of organs.
In recent years two additional techniques have been realized for organ prorement. A ominal organs rom non heart eating onors (
D an e
er se ith emoro emoral VA CM a ter eath has een e lare an en ovascular clamp is inflated in the descending thoracic aorta to avoid brain and
heart perfusion. The second promising technique is represented by ex vivo lung
perfusion; marginal lungs are extracted from the donor and reconditioned for a
few hours with an ECMO machine (see the specific chapter for details) and subsequently implanted in the recipient if they match implantation criteria. Both
techniques are described in the specific section of this book and represent a
potentially relevant resource for organ procurement.
40.4 Organizational Challenges

ECMO represents a typical low-volume, high-risk technique. Organizational aspects
are hence of paramount importance to minimize complications and optimize management, in order to ultimately improve patients outcome.
[email protected]
470
M. Giani et al.
A n m er o iss es still nee to e im ro e in this regar

Referral to high-volume centers has been shown to improve outcome. The implementation o h an s o e mo els is essential or oth VV an VA CM .
Standardized protocols for referrals to the ECMO center are critical, and so is the
development of retrieval programs. Every established ECMO center should perform at least 1015 runs per year in order to maintain an adequate performance
and should provide retrieval service for regional spokes.
ECMO-related complications markedly affect patients outcome. These
should be accurately recorded, and a systematic appraisal carried out. Regular
debriefings should be carried out with all the team to critically examine every
CM r n. Perio i re ie o ra ti e an roto ols is i tate on the asis
of audit results.
Due to the number and extent of complications, ECMO cannulation and management should always be provided from the most experienced staff. This on the
other hand constitutes a limit to the training of new staff. On-the-field training
represents nowadays the standard in most centers. Simulation techniques are
being developed in several institutions and may represent the future of ECMO
education. They could also become an invaluable tool for continuous training
and refresh and for the revision of critical incidents.
References
.
attinoni , Pesenti A, olo o T, Damia (

A ne loo at thera o the a lt res irator istress s n rome motionless l ngs. nt Anesthesiol Clin
. ester R , Ash ra t D, C rr , o er (
on in asi e etermination o re ir lation
in the atient on ial sis. A A
M
M
. in sa RM, ra el , Rothera C, ian ar C, Male P, la e P (
A om arison o
methods for the measurement of hemodialysis access recirculation and access blood flow rate.
A A
. Clements D, Primmer , R man P, Marr , earles , Darling (
Meas rements o
recirculation during neonatal veno-venous extracorporeal membrane oxygenation: clinical
a li ation o the ltraso n il tion te hni e.
tra Cor or Te hnol
5. Darling EM, Crowell T, Searles BE (2006) Use of dilutional ultrasound monitoring to detect
changes in recirculation during venovenous extracorporeal membrane oxygenation in swine.
A A
.
r er P, an sh ha
M, imons AP, Don er D , Maessen , eer in P (
Quantification of recirculation as an adjuvant to transthoracic echocardiography for optimization o
al l men e tra or oreal li e s
ort. ntensi e Care Me
. Pee
, iller M, Ree es R, osno s i A , Firmin R (
arl e erien e ith
a ol meth l entene o genator or a lt e tra or oreal li e s
ort. A A
48:480482
. Toomasian M, hreiner R , Me er D , hmi t M , agan , ri th
, artlett R ,
Coo
(
A ol meth l entene er gas e hanger or long term e tra or oreal li e
s
ort. A A
. Arlt M, Phili A, Voel el , Cam oni D, R
re ht , ra M, hmi C, il er M (
an hel minimise e tra or oreal mem rane o genation a ne
ri ge to re o er in
atients ith o t o entre ar iogeni sho . r Car iothora
rg
[email protected]
471
. The A te Res irator Distress n rome et or (

Ventilation ith lo er ti al olumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory
istress s n rome.
ngl Me
. Patroniti , ellani , Cortino is , Foti , Maggioni , Man o A, Pesenti A (
Role o
absolute lung volume to assess alveolar recruitment in acute respiratory distress syndrome
atients. Crit Care Me
.
ton T , Alonso A, allet R , Daniel M, Pittet F, isner MD, Mattha MA (
P lmonar ea s a e ra tion as a ris a tor or eath in the a te res irator istress s nrome.
ngl Me
. Terragni PP, Ros o h , Teal i A, Corno , Menal o , Da ini , an ini , errmann P,
Mas ia , intel M, l ts A , attinoni , Ranieri VM (
Ti al h erin ation ring
lo ti al ol me entilation in a te res irator istress s n rome. Am Res ir Crit Care
Med 175:160166
. Terragni PP, Del or o , Mas ia , r ino R, Martin , iro o A, Faggiano C, intel M,
Gattinoni L, Ranieri VM (2009) Tidal volume lower than 6 ml/kg enhances lung protection:
role o e tra or oreal ar on io i e remo al. Anesthesiolog
. i igni , Maio M, Ferretti , ongo ar o A, Poten a R, Ri alta , el aggi P, Vergano M,
Bertolini G (2006) Efficacy and safety of a low-flow veno-venous carbon dioxide removal
device: results of an experimental study in adult sheep. Crit Care 10:R151
. at hins A , or an , Regn D, e soi C, Fe ers iel
, Morris M , Can io C (
Respiratory dialysis: reduction in dependence on mechanical ventilation by venovenous extracorporeal CO2 remo al. Crit Care Me
. l ge , ra ne A, ngel M, ierha s A, Frings D,
elt , hrig A, Mets h e M,
egs hei er , ttor
, Ro ssea
(
A oi ing in asi e me hani al entilation
extracorporeal carbon dioxide removal in patients failing noninvasive ventilation. Intensive
Care Me
. ille P, a tier A, To sse l (
C C 2R o genator or hemo ial er An in itro
st . nt Arti rgans
. ni er MT, Cha hari
, Ri har R , hit om DR, R ssell
(
A gmentation o
CO2 trans er in mem rane l ngs the in sion o a meta oli a le organi a i . A A Trans
.
.
.
.
.
.
.
.
olte
, en er R , ra
(
tra or oreal C 2 removal with hemodialysis
(ECBicCO2R ho to ma e
or the i ar onate loss nt Arti rgans
olte
, onit
, ra , Roth , Assen a m R (
emo ial sis or e tra orporeal bicarbonate/CO2 removal (ECBicCO2R) and apneic oxygenation for respiratory
ail re in the ne orn. Theor an reliminar res lts in animal e eriments. A A
Trans
ille P, a er P, ollaert P , To sse l , a hani Manso r R, M ns h
(
C 2
remo al ith hemo ial sis an ontrol o lasma on oti ress re. A A
Trans
654657
ille P, a nier C, hri en F, artemann D, To sse l (
Meta oli C 2 removal by
ial sis T AM s a
in sion. nt Arti rgans
Cressoni M, anella A,
M, Corti , Patroniti , olo o T, Pesenti A (
De reasing
pulmonary ventilation through bicarbonate ultrafiltration: an experimental study. Crit Care
Me
anella A, Patroniti , sgr , Al ertini M, Costan i M, Pirrone F, ara illi V, Vergnano ,
Pesenti A (
loo a i i ation enhan es ar on io i e remo al o mem rane l ng an
e erimental st . ntensi e Care Me
anella A, Mangili P, Re aelli , ara illi V, iani M, Ferli a D, a a aro i D, Pirrone F,
Al ertini M, Patroniti , Pesenti A (
Regional loo a i i ation enhan es e tra or oreal
ar on io i e remo al a
ho r animal st . Anesthesiolog .
(
anella A, iani M, Re aelli , Mangili P, ara illi V, rmas V, Costan i M, Al ertini M,
ellani , Patroniti , Pesenti A (
n sion o . me min o la ti a i minimall
[email protected]
472
M. Giani et al.
increases CO2 production compared to an isocaloric glucose infusion in healthy anesthetized,

mechanically ventilated pigs. Crit Care 11;17(6):R268. [Epub ahead of print]
. anella A, Mangili P, iani M, Re aelli , ara illi V, Castagna , osio , Pirrone F,
Al ertini M, Patroniti , Pesenti A (
tra or oreal ar on io i e remo al thro gh
entilation o a i i e ial sate An e erimental st .
eart ng Trans lant. ii
(
. oi
.
.heal n.
. .
[Epub ahead of print]
. ear en PD, Fe ers iel
, Morle
, Rosen erg M, ienie PD, n
,
hs D
(2012) Respiratory dialysis with an active-mixing extracorporeal carbon dioxide removal system in a hroni shee st . ntensi e Care Me
. ang D, i
D, Cam ell M, oran D , Al ar
,
is hen erger , Cham ers D
(
De elo ment o am lator arterio eno s ar on io i e remo al (AVC 2R): the
downsized gas exchanger prototype for ambulation removes enough CO2 with low blood resistan e. A A
[email protected]
Index
A
AAMI. See Association for the Advancement
of Medical Instrumentation (AAMI)
Abdominal compartment syndrome (ACS), 353
Abella, B.S., 121
Access insufficiency, 432433
Accidental hypothermia
and cardiorespiratory arrest, 163164
CPB, 165166
ECMO, 164
emergency mechanical circulatory support,
163
Acquired von Willebrand disease, 86
ACS. See Abdominal compartment syndrome
(ACS)
Activated clotting time (ACT), 82, 83, 85, 405
Activated partial thromboplastin time (APTT)
and ACT, 405
heparin anticoagulation, 405
heparin therapy, 83
Acute myocardial infarction (AMI)
and complications, 105
CS, 106, 107
revascularization, 112
Acute respiratory distress syndrome (ARDS)
Berlin definition, 239, 240
ECLS techniques, 304, 308
ECMO (see Extracorporeal membrane
oxygenation (ECMO))
H1N1 infection, 271
hyperventilation, 465
minimally invasive system, 309
non-ventilatory strategies
prone position, 243244
pulmonary vasodilator, 243
pathophysiology
inflammatory response, 239
lungs, 240
PECLA, 309
rescue therapy, 269
survival rate, 269

ultra-protective ventilation strategy, 308
ventilator-associated lung injury, 270
ventilatory strategies
HFOV, 242
oxygenation improvement, 242
PEEP, 241242
respiratory rate, 241
spontaneous breathing, 243
Tidal volume (Vt), 241
Adult respiratory distress syndrome (ARDS)
and hyperdynamic septic shock, 183
and traumatic brain injury, 184
Aeromedical transport
acceleration/deceleration forces, 452
advantages, 448
altitude effects, 451452
before departure, 450
ECLS, 445
ECMO, 445
equipment, 449450
fixed-wing aircraft, 445
helicopters, 448
interhospital aeromedical transport, 446448
preparation, 449
and rotary-wing aircraft (helicopter), 445
safety, 452
segment monitoring, 451
Aigner, C., 338
Aiyagari, R.M., 200, 201
Ambulance. See Ground transportation
AMI. See Acute myocardial infarction (AMI)
Angeles, L., 446
Anticoagulation
ACT [14], 405
aPTT, 405
ECMO, 404
platelets and heparin, 405
standard coagulation tests, 406
thromboelastography, 406

DOI 10.1007/978-88-470-5427-1, Springer-Verlag Italia 2014
[email protected]
473
474
Index
Anticoagulation
bivalirudin, 81
efficacy, heparin, 8081
thrombotic and hemorrhagic
complications, 80
Antifibrinolytic therapy, 86
APTT. See Activated partial thromboplastin
time (APTT)
ARDS. See Acute respiratory distress
syndrome (ARDS); Adult
respiratory distress syndrome
(ARDS)
Arlt, M., 154
Arteriovenous (AV)
blood, 297
fistulae, 45
myocardial dysfunction, 99
pumpless, 297
Arteriovenous malformations (AVM), 224
Artificial lung. See Membrane lungs (MLs)
Ashbaugh, D.G., 239
Assessment
abdomen and nutrition, 353354
airways and ventilation
bilateral pneumothorax, pulmonary
tuberculosis, 351
gas exchange deteriorating, 351352
ventilatory setting, protective, 350
weaning process, ECMO
disconnection, 350351
bleeding management and transfusion
targets, 355356
haemodynamic and volume status, 352
head and sedation, 349
hepatic and renal functions, 354355
infection evaluation and workup, 352353
Assisted mechanical ventilation, 319
Association for the Advancement of Medical
Instrumentation (AAMI), 406407
Attisani, M., 227
AV ECMO. See Arteriovenous (AV)
AVM. See Arteriovenous malformations
(AVM)
Awake ECMO approach
as BTT, 298, 299
reduction, sedation and mechanical
ventilation drawbacks, 299300
Axillary vessels cannulation
advantages, 59
artery, 6061
complications, 61
disadvantages, 59
obesity and wall chest edema, 59
post-cardiotomy patients, 59
vein, 61
B
Bartlett, R.H., 5, 273
Baud, F.J., 172
Bavaria, J.E., 96
BDD. See Brain death donor (BDD)
Bein, T., 184, 305, 306, 313
Bellomo, R., 118
Belohlavek, J., 157
Bermudez, C.A., 111
Beurtheret, S., 203
Bicarbonate ultrafiltration, 466
Biocoating
cannulas
arterial, 7475
design, 72, 73
percutaneous, 72, 73
selection, arterial and venous, 72
size, 72
venous, 74
Carmeda, 66
centrifugal blood pumps, 6769
circuit, adult patients, 67, 68
Duraflo II heparin, 66
membrane oxygenators
characteristics, 69, 70
PMP, 70
pressure, 72
structure, 70, 71
temperature gradient, water and the
blood, 69, 71
physio, 67
polypeptides and heparin, bioline, 67
Rheoparin, 66
Biocompatibility, 65, 66
Biocompatible components, 67
Biomarkers, 208
Bisdas, T., 57
Bithermia preservation, 332
Bleeding
coagulatory status, 419
ECMO, 418
GI and airway, 420
surgical and cannulation site, 420
Blood acidification, 467
Blood clotting, 415416
Blood flow, 404
Bosco, E., 389397
Bowles, N.E., 139
Boyles law, 451
Brain death determination (DBD), 340, 341
Brain death donor (BDD), 327, 329, 333
Brain function, neurophysiological monitoring
continuous EEG/SEP, 395
EEG, 392393
ischaemic penumbra, 391
[email protected]
Index
475
SEPs (see Somatosensory voked

potentials (SEPs))
Brchot, N., 183
Brenner, M., 276
Bridge to lung transplantation (BTT). See also
Extracorporeal membrane
oxygenation (ECMO)
extracorporeal carbon dioxide removal
systems, 311
extracorporeal respiratory support, 283
intubation and sedation, 282
invasive mechanical ventilation, 311312
mechanical ventilation, 289
polyneuropathy/myopathy, 285
pre-transplant mechanical ventilation, 276
pulmonary circulation, 284
Bronchopleural fistula, 277
Brukhonenko, S., 4
Brunet, F., 305, 308
Bryner, B.S., 274
BTT. See Bridge to lung transplant (BTT)
Burke, A.P., 139
Burki, N.K., 306, 310
C
CA. See Cardiac arrest (CA)
Camboni, D., 298
Cancer, 276
Cannulation
ECMO, 427
location and techniques, 427428
RIJV, 426
thrombosis, 427
vascular complications
leg ischemia, 417419
vascular access complications, 417, 418
Carbon dioxide (CO2)
aerobic cellular respiration, 304
aerobic metabolism, 29
alveolar ventilation, 30
bicarbonate ion form, 2930
carbamino compounds, 30
carbonic acid, 29
COPD and ARDS, 30
dissolved, 29
lung diseases, 311
metabolism, 308
minimally invasive system, 309
oxygenation, 308
removal
ARDS, 465
artificial lung, 304
and blood gas parameters, 467
citric acid, 467

CVVH, 466
ECCO2R systems, 467468
ECMO, 312
extracorporeal, 256
hemofilter, 309
lungs, 30
oxygenation, 308
VA-ECMO, 32
VV-ECMO, 31
transporation, 30
ventilatory-induced lung injury, 30
Cardiac arrest (CA)
controlled DCD, WIT, 328
donors, 327
mobilization, 436
VA ECMO, 435436
VV ECMO, 435
witnessed out-of-hospital refractory, 331
Cardiac complications, 422423
Cardiac failure
algorithm, 228
HLA sensitization, 229
IABP, 218219
INTERMACS classification, 217218
MCS, 219220
refractory cardiac arrest, 7
RVF, 225
surgery, 3
TAH, 225226
transplantation, 226228
VADs, 220225
Cardiac output (CO) syndrome
minimally invasive technique, 376, 377
PAC, 376
Cardiac recovery, 207, 208
Cardiac support. See Physiology
Cardiogenic shock (CS)
AMI, 105
description, 105
diagnostic criteria, 108
echocardiography, 108
ECLS (see Extracorporeal life support (ECLS))
epidemiology and pathophysiology, 106
hemodynamics, 107108
low cardiac output, 106107
management
IABP, 109110
MSC, 110
myocardial performance, 109
PCI, 109
mechanical circulatory support, 106
SIRS, 107
stunned and hibernated myocardium, 107
ventricular function, 107
[email protected]
476
Index
Cardiopulmonary bypass (CPB)

cardiac operations, 77
circulatory support and oxygenation, 174
ECMO support, 166
heparin, 168
and IABPs, 173
Cardiopulmonary resuscitation (CPR)
duration, 121
and ECMO, 431
and VV ECMO, 435
Cardiorespiratory arrest
accidental hypothermia and hypothermic,
163164
hypothermia, 164
CBF. See Cerebral blood-flow (CBF)
Central ECMO cannulation
complications and disadvantages, 5455
indications, 5152
perioperative management, 54
surgical technique
aorta, 52
decannulation, 54
left ventricle and atrial pressure
monitoring, 53
occlusive dressing, 53
primary chest closure, 54
two purse-string sutures, arterial, 52, 53
venous cannula, 52
Centrifugal pump
ECMO circuit, 67
electromagnetic induction motors, 67
hemodynamic conditions, 68
membrane oxygenator, 69
Cerebral blood-flow (CBF)
and arterial oxygen level, 396
neurosonology monitoring, velocity and
microembolic signals, 395396
Cervical vessels cannulation
complications, 61
jugular vein, 61
neonates and infants, 61
surgical technique, 62
CF-VADs. See Continuous-flow ventricular
assist devices (CF-VADs)
Chaparro, S.V., 203
Chassaing, E., 446
Chen, Y.-S., 121, 384
Chronic obstructive pulmonary disease
(COPD)
NIPPV, 310
NIV/IMV, 310311
Circuit fractures, 416417
Circuit pressure monitoring, 406, 407

Circuit/pump failure, 403404
Circuit-related complications
blood clots, 415416
fractures, 416, 417
gas embolism, 416
Clarke, A.J., 4
CO. See Cardiac output (CO) syndrome
Coagulopathy, 420421
Combe, A., 208
Combes, A., 111
Congenital diaphragmatic hernia, 274275
Continuous-flow ventricular assist devices
(CF-VADs)
anticoagulation, 221
heart failure metrics, 222
hemodynamic and hematologic
constellation, 223
RVF, 225
Continuous renal replacement therapies
(CRRT)
increase, heparin requirement, 355
manipulation, 355
renal impairment, 354
Cooper, L.T., 140
COPD. See Chronic obstructive pulmonary
disease (COPD)
Coronary perfusion
left ventricle, 98
systole, 98
Cortes, G.A., 258
CPB. See Cardiopulmonary bypass (CPB)
CPR. See Cardiopulmonary resuscitation
(CPR)
Cressoni, M., 466
Critically ill transportation, 455
CRRT. See Continuous renal replacement
therapies (CRRT)
CVVH, 466
Cypel, M., 295, 297, 338, 341
D
Daily care
assessment (see Assessment)
imaging, 356
nursing (see Nursing)
physical therapy and mobilisation,
356357
Daubin, C., 175
DCD. See Donation after cardiac death (DCD)
DCM. See Dilative cardiomyopathy (DCM)
[email protected]
Index
477
Decarboxylation, 20
Deehring, R., 157
Deep venous thrombosis (DVT), 157
De Lange, D.W., 175
Delayed graft function (DGF), 329, 332
Dellinger, R.P., 182
DGF. See Delayed graft function (DGF)
Dietl, C.A., 141
Dilative cardiomyopathy (DCM)
myocarditis, 141
ventricular dysfunction, 142
Distal perfusion, 44
Doll, N., 112
Donald Hill, J., 4
Donation after cardiac death (DCD)
BDD management, 327
controlled, 332
preservation strategies, 328329
uncontrolled, 328, 332
Drakos, S.G., 196
Drug intoxication
life-threatening events, 172
mechanical circulatory support
aortic counterpulsation, 173174
CPB, 174
ECMO, 173176
medical treatment, 171
membrane-stabilizing effect, 171172
DVT. See Deep venous thrombosis (DVT)
E
Echocardiography
algorithms, 361, 362
cannulation, 365366
coronary flow, 380
ECMO-assisted refractory cardiac shock,
209211
haemodynamic monitoring, 376
monitoring (see Monitoring)
tamponading effusions/valvular
abnormalities, 208
transthoracic (TTE) vs. transesophageal
(TEE), 361
ultrasound-guided approach, 361
VA ECMO (see Veno-arterial ECMO
(VA-ECMO))
VV ECMO (see Veno-venous ECMO
(VV-ECMO))
Echo-dynamic approach, 373
ECLS. See Extracorporeal life support
(ECLS)
ECMO. See Extracorporeal membrane

oxygenation (ECMO)
ECMOnet. See Extracorporeal membrane
oxygenation network (ECMOnet)
ECMO program
architectural and infrastructural features,
1718
artificial heart/lung apparatus, 11
description, 11
ELSO, 11, 12
equipment selection, 16
financial support and cost-benefit ratio, 16
health care, 11
identification, steering group, 13
manageable patient population, 13
planning, 12
staff (see ECMO service)
training, 1617
transplant centers, 294
ECMO referral, 455, 458
ECMO service
biomedical engineering department, 16
coordinator, 1314
supportive personnel, 13, 15
team, 1415
ECMO specialist
consultants and rehabilitation, 13
intensive care physicians and nurses, 14
training, 17
EEG. See Electroencephalography (EEG)
Electroencephalography (EEG)
changes, 392
Fourier spectral analysis, 392
low-voltage theta and delta rhythm, 392,
393
TCD-EEG coupling, 392, 393
ELSO guidelines. See Extracorporeal Life
Support Organization (ELSO)
guidelines
EMB. See Endomyocardial biopsy (EMB)
Endomyocardial biopsy (EMB), 144145
End-stage respiratory failures
LTx, 293
organ dysfunctions, 295
EVLP. See Ex vivo lung perfusion (EVLP)
Extracorporeal carbon dioxide removal
(ECCO2R)
development, ultra low-flow-enhanced,
467468
and low-flow ECMO (see Low-flow
ECMO and ECCO2R systems)
Extracorporeal circulation. See Monitoring
[email protected]
478
Index
Extracorporeal circulatory support

CO, 212
description, 207
peripheral ECMO, 213
technique
hemodynamics, 209
levosimendan, 212
postcardiotomy patients, 209
ventilatory support, 212
weaning protocol, 212
weaning, predictors
biomarkers, 208
ECMO, 207208
Extracorporeal gas exchange
CO2 removal (see Carbon dioxide (CO2))
description, 463
evaluation and reconditioning, 469
ischemia/reperfusion injury, 469
organizational challenges, 469470
oxygenation, 464465
RCA, 468469
Extracorporeal life support (ECLS)
ARDS, 304, 308
artificial oxygenation, 34
bubble and surface-type oxygenators, 4
cardiopulmonary bypass, 4
cardiopulmonary/respiratory failure, 445
development and spread, 5
ECCO2R, 56
ECMO (see Extracorporeal membrane
oxygenation (ECMO))
evaluation, left ventricle failure, 371
evolution, 3, 9
heart-lung machines, 45
interaction, normal circulation (see Normal
circulation)
patient care
discontinuation, 113114
heart and lung function, 113
interventions, heart rest and recovery,
112
stem, cardiac surgery, 3
techniques, 4
VAD, 111
whole-body extracorporeal perfusion, 4
Extracorporeal Life Support Organization
(ELSO) guidelines, 12, 17
Extracorporeal membrane oxygenation
(ECMO). See also Cardiogenic
shock (CS); High-risk procedures;
left ventricle (LV); Weaning,
VV-ECMO
accidental hypothermia, 164165, 167168
acquired von Willebrand disease, 86

antifibrinolytic therapy, 86
assisted spontaneous breath, 245246
biocoating, 6667
biocompatible components, 67
bridging therapy, 66
BTT, 282
CA, 175, 435436
cannulation sites, 284285, 427
centers, 6
circuit replacement, 290
circulation, cardiac surgery, 65
clinical experience and outcome, 166167
configuration
algorithm, selection, 296
AV and VV, 297
awake ECMO, 298, 299
PAH, 298
vs. conventional therapies, 5
CPB technology, 77, 165166
and CPR, 431
CS, 174
daily care (see Daily care)
drug intoxication, 175
echocardiography (see Echocardiography)
ECLS, 175
effects, 300
experiences, 294, 295
failure
circuit/pump, 403404
and ML, 402403
features and indications, 174
fibrinogen, 8586
hemorrhagic and thrombotic
complications, 7980
hemostatic system (see Hemostatic system)
HIT, 86
H1N1 infection, 271
hyperfibrinolysis, 86
IMV, 281, 293
inflow cannula, 66
Italian network, 266
LAS, 294
life-threatening bleeding, 85
mobilization, 436
MPE, 180181
MV control, 244245
myocarditis
diagnosis, 145146
indications, 145
support, 146147
Neurological monitoring (see Neurological
monitoring)
[email protected]
Index
479
nonrespiratory organ, 269

optimal hemostatic pattern, 85
patient state changes
artificial fever, 288289
fever, 287
heat exchanger, 287
oxygenation, 288, 289
PH, 181182
physiology (see Physiology)
planned awake, 282283
platelet count, 86
and prognosis, 147148
psychological issues, 290
respiratory deterioration, 271
resuscitation, 166
right ventricular failure, 179
septic shock, 182, 295296
severe chronic pulmonary hypertension,
284
support function, 287, 288
survival rate, 269
and systemic inflammatory reaction, 87
team, ventilator management and safety,
268
techniques, 65, 66
trauma, 184
traumatic brain injury, 186
treatment, 65
VA (see Veno-arterial ECMO
(VA-ECMO))
VAD, 176
vaECMO, 283
venovenous (VV), 67
ventilatory support, 289
vvECMO, 283
network (ECMOnet)
ARDS, 270
cannulation, 265
extrapulmonary organ, 271
gas exchange and systemic perfusion, 265
hemorrhagic complications, 269
H1N1, 269
ICU centers selection, 266
mortality, 271
multivariate analysis, 269
patients centralization, 271
patient selection and referral
clinical criteria, 266268
management algorithm, 266, 267
transportation, 268
rescue therapy, 269
respiratory viral infections, 266
score calculation, 270
team, ventilator management and safety, 268

(ECMO) transportation
equipment, 450
and ISMETT, 447
Extracorporeal oxygenation
PPML, 465
SRML, 465
VO2ML, 464
Ex vivo lung perfusion (EVLP)
advantages, 341
clinical diagram flow, 340
Lund vs. Toronto EVLP protocols, 339
published clinical experiences, 338
time frame, 341
F
Failure to wean, 208
Femoral vessels cannulation
artery, 5557
vein, 57
Fenoglio, J.J., 139
Fever
oxygen consumption and CO2 production,
289
temperature-regulated heat exchanger, 287
thermostatic action, 288
zero watts, 288
Fischer, S., 295, 297
Florchinger, B., 305, 309
Foley, D.S., 459
Fondevila, C., 332
FRC. See Functional residual capacity (FRC)
Frey, M., 3
Fuehner, T., 295, 307, 312
Fulminant, 140
Functional residual capacity (FRC), 258, 259,
465
G
Gas embolism, 416
Gas exchange
adequacy, 320
artificial lung, 20
capability, VO2NL, 27
ECMO discontinuation, 318
intrapulmonary shunt fraction, 249
native lung, 257
nonventilated alveoli, 255
oxygen, 23
VV-ECMO, 249
[email protected]
480
Index
Gas transfer
carbon dioxide (CO2), 407, 408
MLs, 406
monitoring, 407, 408
GCM. See Giant cell myocarditis (GCM)
Giant cell myocarditis (GCM)
differential diagnosis, 139140
and eosinophilic myocarditis, 147
relapse, disease, 140
Gibbon, J., 4, 11, 273
Gibbon, J.H. Jr., 179
Gollan, 4
Grasso, S., 258
Griffith, G.C., 106
Grinda, J.M., 141
Ground transportation
adverse events, 459
vs. air, 456, 457
ECMO, 455
equipment, 456458
interhospital, 455
pre-transport preparation, 458459
team, 459
Gruber, M., 3
Guner, Y.S., 275
H
Haemodynamic monitoring
life-threatening cardiac/respiratory disease,
375
(VA-ECMO))
(VV-ECMO))
Hammainen, P., 295
Hantavirus cardiopulmonary syndrome
(HCPS), 141
Harlequin syndrome, 129, 379380
HCPS. See Hantavirus cardiopulmonary
syndrome (HCPS)
Heart and lung function
chest x-rays and ultrasound, 113
myocardial specific enzymes, 113
natriuretic peptides, 113
Heart assist device
L-VAD (see Left ventricular assist device
(L-VAD))
VADs (see Ventricular assist devices
(VADs))
Heart-beating and NHBDs
ECMO assistance, BDD, 333
Maastricht, categorization, 328

NECMO (see Normothermic ECMO)
organ procurement strategies, 333
preservation strategies, DCD donors,
328329
WIT, 328
Helicopter transport, 446, 452
Hemodynamics
cannulation site, 3435
VA ECMO
cannulation sites, 33
increment, LV afterload, 34
loss of arterial flow pulsatility, 3334
preload reduction, 34
systemic flow, 33
venoarterial ECMO circuits, 32
VV ECMO, 33
Hemorrhagic complications, 7980
Hemostatic system
activation
dialysis, 77
fibrinolytic system, 79
material-dependent, 78
tissue factor and thrombin generation,
7879
and coagulation tests, 85
management
antiplatelet drug, 82
aspirin, 82
HIT, 8182
synthetic antifibrinolytics, 82
monitoring
ACT, 82
APTT, 83
TEG and TEM, 8384
Heparin coating, 66
Heparin-induced thrombocytopenia (HIT)
direct thrombin inhibitors, 81
ECMO and ventricular assist device, 86
HIT-screen, 438
Hetzer, R., 196
HFOV. See High-frequency oscillatory
ventilation (HFOV)
High-frequency oscillatory ventilation
(HFOV), 242
High-risk procedures
cannulation and complications, 157158
compact pump-oxygenator design, 152
description, 151
ECMO, 151
PCI, 152154
[email protected]
Index
481
PE, 156157
TAVI, 155
VSD, 155156
Hill, D., 273
History. See Extracorporeal life support
(ECLS)
HIT. See Heparin-induced thrombocytopenia
(HIT)
Hommel, M., 306, 311
Hoopes, C.W., 295
HSM. See Hypersensitivity and eosinophilic
myocarditis (HSM)
Huang, S.C., 58, 141
Hypersensitivity and eosinophilic myocarditis
(HSM), 139140
Hypothermia, 118, 329
I
IABP. See Intra-aortic balloon
counterpulsation (IABP); Intraaortic balloon pump (IABP)
Iapichino, G., 353
ICU. See Intensive care unit (ICU)
In situ perfusion cooling, 328329
IMV. See Invasive mechanical ventilation
(IMV)
Infections, 423
Influenza A (H1N1)
ECMOnet, 271
respiratory support, 271
respiratory viral infections, 266
Ingemansson, R., 338, 341
Inotropes, 209, 212
Insertion, cannula
awake cannulation, 431
cannulation location and techniques,
427428
central venous access, 426
difficult cannulation, 429
ECMO CPR, 431
kinking wire, 430
peripheral VA, 429
pulmonary artery catheter and cannulation,
427
right ventricular perforation, 430
RIJV, 426
subclavian insertion, 427
thrombosis, 427
ultrasound tips, 428429
Intensive care unit (ICU)
daily nursing, 346
diagnosis, new infection, 352
muscle dysfunction, 356

preparation, 356
Interhospital aeromedical transport
Agusta Westaland, 447
description, 446
ISMETT, 447
Lockheed C-130 Hercules, 447448
Vietnam War, 446
INTERMACS classification
description, 217
profile patients, 218
Intra-aortic balloon counterpulsation (IABP)
blood supply, 197198
clinical use, 197
complications, 198199
coronary flow, 197
and ECMO, 198
Intra-aortic balloon pump (IABP)
diastolic inflation, 218219
helium-filled balloon, 218
post-myocardial infarction, 218
systolic deflation, 219
Invasive mechanical ventilation (IMV)
alternative to intubation, 282
complications, 294
postoperative, 285
severe respiratory failure, 281
Irons, D., 153
J
Javidar, J., 295
Johnston, T.A., 200
K
Kantrowitz, A., 218
Karagiannidis, C., 246
Kirklin, J.W., 4
Klotz, S., 196
Koeckert, M.S., 203
Koenig, P., 199, 200
Kolobow, T., 5, 201, 251, 463
Kondo, T., 276
Konishi, H., 371
L
Laboratory issues
anticoagulate, 436437
fever and elevated white cell count, 437
plasma-free hemoglobin, 438
thrombocytopenia, 438
[email protected]
482
Index
Lactic acid, 467

Lang, G., 295
Larsson, M., 184
LAS. See Lung allocation score (LAS)
Lazar, H.L., 198
LBD. See Load distributing band (LBD)
LCOS. See Low cardiac output syndrome
(LCOS)
Lee, J.W., 341
Lee, M.S., 154
Left main coronary artery (LMCA), 153
Left ventricle (LV)
blade and balloon atrial septostomy,
199200
cardiac remodeling, 194
decompression, 194, 203204
ECMO support, 193
heart failure, 196197
IABP, 197199
impeller pumps, 203
inotropic drugs, 194
mechanical supports, 194
percutaneous pulmonary artery venting,
201202
rest, 194, 204
transaortic catheter venting, 202
transseptal cannulation, 200201
unloading, 196197
V-A ECMO, 194
venting, 199
Left ventricular assist device (L-VAD)
advantage, 386
destination therapy, 93, 224
implantation, 224
placement, 225
RVAD support, 225
Left ventricular unloading, 384386
Leg ischemia
femoral arterial cannulation, 417, 419
peripheral perfusion cannula, 417419
Le Guen, M., 121
Lei, J., 276
Lim, N., 107
Linden, V., 459
Lin, J-W., 120
LMCA. See Left main coronary artery
(LMCA)
Load distributing band (LBD), 122
Long-term ventricular assist devices
axial and centrifugal pumps, 223
CF-VADs, 221
electric motor drives, 222
gastrointestinal bleeding, 223224

Heydes syndrome, 224
implantation, 225
inflow cannula, 222
rotor spins, 222
thrombosis, 223
Low cardiac output syndrome (LCOS)
description, 127
risk factors, 130
Low-flow alarms, 432
Low-flow ECMO and ECCO2R systems
ARDS (see Acute respiratory distress
syndrome (ARDS))
BTT, 311312
clinical studies, efficacy, 304307
complications, 312313
COPD (see Chronic obstructive pulmonary
disease (COPD))
MV, 303
physiology, 304
thoracic surgery, 311
Ludwig, C., 3
Lung
donor criteria, 337
EVLP, evaluation, 340
extracorporeal perfusion, 337338
function, 337, 338, 340, 341
perfuse, isolated, 338, 339
reconditioning, advantages, 341
rejection, 341
Lung allocation score (LAS), 294
Lung reconditioning
EVLP (see Ex vivo lung perfusion (EVLP))
perfuse, isolated lungs, 338, 339
Lung transplant (LTx). See also Bridge to lung
transplant (BTT); Extracorporeal
membrane oxygenation (ECMO)
centers, clinical EVLP programs
implementation, 338
double, 338
pig model, 341
Lung ultrasound (LUS), 259
LUS. See Lung ultrasound (LUS)
Luyt, C.E., 204, 208
L-VAD. See Left ventricular assist device
(L-VAD)
M
Machine perfusion. See Normothermic
machine perfusion (NMP)
[email protected]
Index
483
Maclean, J., 4
Madershahian, N., 198
Maggio, P., 180
Magnetic resonance imaging (MRI), 144, 145
Magovern, G.J., 153
Management
avoidance, faecal devices, 354
bleeding and transfusion targets, 355356
ECCO2R, 284, 285
inspiratory breathing efforts, 286
objectives, 299
oxygenation impairment, 286
oxygen desaturation, 286
Marini, J.J., 258
Martins, S., 341
Massive pulmonary embolism (MPE)
cardiac failure, 181
ECMO, 180
Masson, R., 175
Mauri, T., 246, 351
MCS. See Mechanical circulatory support
(MCS)
Mechanical cardiac supports (MCS)
aortic regurgitation, 128
ECMO, 128
types, 128
Mechanical circulatory support (MCS)
advantages, 219
hemolysis and intraventricular thrombosis,
220
impella, 219
TandemHeart system, 220
Mechanical ventilation (MV)
hypoxemic and intrapulmonary shunt, 244
lung rest, 244
positive pressure, 303
protective, 309310
support gas exchanges, 244
ultra-protective, 309
Membrane lungs (MLs)
failure, 402403
gas transfer
and AAMI, 406407
displacement, hollow fibers, 407
management, extracorporeal assistance,
409
partial pressure (pO2), 409
and PMP, 408
and SGF, 409410
shunt and dead space, 407408
shunt definition, 409, 410
Membrane oxygenators
characteristics, 69, 70
PMP, 70
pressure, 72
structure, 70, 71
temperature gradient, water and the blood,
69, 71
Membrane stabilizing agents (MSA)
cardiogenic shock, 173
electrophysiological alterations, 172
Middle-latency cortical somatosensory evoked
potentials (MLCEP), 395
Migliocca, 332
Misawa, Y., 157
Mixed venous oxygen saturation (SvO2)
pulmonary artery catheter, 95
SwanGanz catheter, 377
VA ECMO, femoral approach, 377, 378
MLCEP. See Middle-latency cortical
somatosensory evoked potentials
(MLCEP)
MLs. See Membrane lungs (MLs)
Mobile ECMO support, 445
Mols, G., 318
Monitoring
blood flow and RPM, 404
cardiac rest, 368
detection, complications, 369370
drainage cannula, 366
echographic evaluation, 366
ECMO failure (see Extracorporeal
membrane oxygenation (ECMO))
extracorporeal circuit, 401
flow variation, 368
ML gas transfer, 406410
perfusion, cannulated limbs, 369
pressures, 406, 407
recovery and weaning, 370371
recovery, cardiac function, 367
thrombosis, left ventricle, 367
thrombus, ascending aorta, 367, 368
Monzas flow chart, 122123
Morris, A.H., 305, 308
Mortality
economic and ethical issues, 271
ICU services, 266
systemic mean arterial pressure, 271
MPE. See Massive pulmonary embolism
(MPE)
Multiorgan preservation methodology. See
Bithermia preservation
MV. See Mechanical ventilation (MV)
Myocardial dysfunction, 195196
Myocardial infarction (MI), 153
[email protected]
484
Index
Myocardial recovery
acute cardiogenic shock, 93
IABP, 99
and ventricular function, 98
Myocarditis
clinical presentation, 141142
diagnosis
chest radiography, ECG and laboratory
tests, 143
description, 142143
EMB, 144145
MRI, 144
TTE-EEE, 143144
ECMO, 137
epidemiology, 137138
etiology, 138
GCM, 140
HCPS, 141
HSM, 139140
pathogenesis, 139
pheochromocytoma, 140141
PP-CMP, 140
N
Near-infrared spectroscopy (NIRS), 129, 391,
396, 397
NECMO. See Normothermic ECMO
(NECMO)
Net, M., 331
Neurological complications, 422
Neurological monitoring
cerebral metabolism, 396397
ICU, 389
multimodal strategy, 390, 391
neurophysiological, brain function (see
Brain function, neurophysiological
monitoring)
neurosonology, CBF velocity and
microembolic signals, 395396
NIPPV. See Noninvasive positive pressure
ventilation (NIPPV)
NIRS. See Near-infrared spectroscopy (NIRS)
NIV. See Noninvasive ventilation (NIV)
NMP. See Normothermic machine perfusion
(NMP)
Non-heart-beating donors (NHBDs). See
Heart-beating and NHBDs
Noninvasive positive pressure ventilation
(NIPPV), 310
Noninvasive ventilation (NIV)
circuit replacement, 290
lung fibrosis/cystic fibrosis, 282
oxygen therapy, 289

pressure treatment, 289
Normal circulation
ECMO, 9394
L-VAD, 93
MCS, 93
myocardial recovery, 93
(VA-ECMO))
VV ECMO, 100101
Normothermic ECMO (NECMO)
controlled NHBD, 332333
femoral artery and vein, 329
physiology, 330331
post-oxygenator arterial blood gas, 330
UNHBD, 331332
Normothermic machine perfusion (NMP), 333
Nursing
alterations, vital signs, 346, 348
ECMO cannulas care, 346, 347
hypovolaemic patients, 347
ICU, 346
mouth care, 346
skin integrity, 346, 348
O
Oey, I.R., 277
OHCA. See Out-hospital cardiac arrest
(OHCA)
OI. See Oxygenation index (OI)
ONeil, M.P., 198
Organizational challenges, 469470
Outcomes
accidental hypothermia, 166
hypothermic cardiorespiratory arrest, 165
Out-hospital cardiac arrest (OHCA), 118
Oxygenation
arterial and mixed venous O2Hb saturation,
251, 252
arterial blood, 251
hypoxemia, 251
PPML, 465
SRML, 465
VA-ECMO, 2829
VO2ML, 464
VV-ECMO support
arterial, 25
calcualtion, 25
delivery and consumption, 23, 24
efficiency, 25
extracorporeal blood flow, 24, 25
SvmixO2, 2627
[email protected]
Index
485
VO2ML, 26
VO2NL, 2728
Oxygenation index (OI), 254
P
PAC. See Pulmonary artery catheterization
(PAC)
PAH. See Pulmonary arterial hypertension
(PAH)
Patient-related complications
bleeding (see Bleeding)
cannulation vascular, 417418
cardiac, 422423
coagulopathy, 420421
infection/sepsis, 423
neurological, 422
Patroniti, P., 1935
PCI. See Percutaneous coronary interventions
(PCI)
PECLA. See Pumpless extracorporeal lung
assist (PECLA)
PEEP. See Positive end-expiratory pressure
(PEEP)
Percutaneous cannulation
complications, 45
development, 38
placement
configuration, 40
drainage and reinfusion, 3940
establishment and maintenance, 39
femoro-jugular approach, 40
handmade double-lumen coaxial
catheter, 38
reimmission, 39
size, 39
VA ECMO, 39
vascular ultrasound, 39
VV ECMO, 39
preparation, 4041
(VA-ECMO))
(VV-ECMO))
Percutaneous coronary interventions (PCI)
ECMO, 154
IABP, 153154
and LMCA, 153
MI, 153
and valvuplasty, 152
Percutaneous mechanical circulatory support,
219220
Percutaneous pulmonary artery venting, 201202
Pereszlenyi, A., 181

Peripartum cardiomyopathies (PP-CMP), 140
Peripheral cannulation
vs. central, 5051
complications, 5859
description, 55
femoral artery, 5557
femoral vein, 57
limb ischemia and distal perfusion, open
access, 5758
Peripheral VA ECMO
backflow cannula insertion, 429
differential hypoxia, 434435
pulmonary edema, 434
Persistent hypoxemia, 433434
PGD. See Primary graft dysfunction (PGD)
PH. See Pulmonary hypertension (PH)
Phillips, S.J., 198
Phosphorylcholine coating, 67, 70
Physiology
artificial lung, 2021
carbon dioxide (see Carbon dioxide (CO2))
hemodynamics (see Hemodynamics)
life support technique, 20
management, 20
modern intensive care medicine,
cardiorespiratory system, 20
oxygen
acute hypoxic respiratory failure, 23
cascade, 19
consumption, 22
content, calculation, 22
hemoglobin, 22
mitochondria, 22
oxygenation (see Oxygenation)
partial and alveolar pressure, 2224
plasma, 22
pulmonary capillary blood, 22
release, 23
respiratory and cardiovascular
systems, 22
systemic delivery, 22
transferring, 23
VA/Q ratio, 23
Plasma-free hemoglobin (plasma-free Hb),
438
PMP. See Polymethylpentene (PMP)
Pneumothorax management, 439
Polymethylpentene (PMP) membrane, 67, 70,
403, 408
Polypropylene membrane lungs (PPML), 465
Polypropylene (PP) microporous
membrane, 71
[email protected]
486
Index
Positive end-expiratory pressure (PEEP),

241242
Postcardiotomic shock (PCS)
anticoagulation, 131, 132
ascending aorta, 129
axillary artery, 129
central cannulation, 128
definition, 127
harlequin syndrome, 129
inotropic support, 128
LCOS, 130
MCS, 127128
NIRS, 129
outcomes, 133134
peripheral cannulation, 128
sepsis, 131
weaning, 132133
Post-infarct ventricular septal defect
((PI-VSD)
description, 155156
ECMO, 156
PP-CMP. See Peripartum cardiomyopathies
(PP-CMP)
PPML. See Polypropylene membrane lungs
(PPML)
Primary graft dysfunction (PGD), 277
Pseudoaneurysm, 45
Pulmonary arterial hypertension (PAH)
lung transplantation, 182
RV failure, 181
V-A ECMO, 181182
Pulmonary artery catheterization (PAC), 376
Pulmonary edema
acute, 203
endothelial activation and aggravation, 101
and hypoxic, 434
and parenchymal disease, 98
peripheral VA ECMO, 434
Pulmonary embolism (PE)
DVT, 157
right ventricular, 156157
surgery, 275
Pulmonary hypertension (PH), 180181
Pulmonary infections, 275276
Pulmonary thromboendarterectomy, 278
Pumpless extracorporeal lung assist (PECLA),
309, 313
R
Raina, A., 225
Rajagopal, S.K., 147
Rao, V., 127
RCA. See Refractory cardiac arrest (RCA)
Recirculation (VV ECMO), 433

Redaelli, G., 207213
Refractory cardiac arrest (RCA)
CPR, 121122
description, 117
LBD, 122
low-flow time, 121
Monzas flow chart, 122123
multidisciplinary approach
ECLS, 120
ECMO, 120
hypothermia, 118
OHCA, 118
resuscitation care, 119
venoarterial extracorporeal membrane
oxygenation, 119
no-flow time, 121
Respiratory failure
acute, 282
IMV, 281
severe, 3
support, heart-lung machine, 4
therapies, 5
Respiratory mechanics
baby lung, 257
CplRS value, 257
FRC, 259
LUS, 259
MV, 258
pleural pressure, 258
transpulmonary pressure, 258
Respiratory monitoring
carbon dioxide, 256257
ECMO, 249
electronic spreadsheet, 250
gas exchange, 249
hypoxic vasoconstriction, 250
intrapulmonary shunt, 255256
mechanical ventilation, 252
membrane lung and natural lung oxygen
supply, 254255
OI, 254
oxygenation, 251252
pulmonary arterial pressure and cardiac
output, 251
respiratory mechanics, 257260
VA-ECMO
deterioration, lung function, 383384
lung function, 384
management, lung, 385386
VV-ECMO, 253
Respiratory support, 273
Resuscitation
CPB, 166
ECMO, 167
[email protected]
Index
487
and rewarming, 168

Return to spontaneous beating (ROSB), 120
Return to spontaneous circulation (ROSC),
117
Ricci, D., 295, 297, 307
Right internal jugular vein (RIJV), 426
Right ventricular failure (RVF), 225
RIJV. See Right internal jugular vein (RIJV)
Riley, R.L., 23, 407
ROSB. See Return to spontaneous beating
(ROSB)
ROSC. See Return to spontaneous circulation
(ROSC)
Rotation per minute (RPM), 95, 404, 407
RVF. See Right ventricular failure (RVF)
Surgical ECMO cannulation

axillary vessels (see Axillary vessels
cannulation)
central (see Central ECMO cannulation)
cervical vessels, 6162
methods and vascular access, 49
open vs. percutaneous, 50
peripheral (see Peripheral cannulation)
SwanGanz catheter, 212, 376, 377, 379
Sweep gas flow (SGF)
CO2 removal capacity, 409
and ECMO, 410
Systemic inflammatory reaction, 87
Systemic inflammatory response syndrome
(SIRS), 107
S
Sakamoto, S., 111
Sakuma, M., 181
Sanchez, M.A., 313
Sasson, C., 118
Schaefer, J.W., 446
Schmidt, A., 3
Schwartz, D., 201
Seldinger technique, 41, 44, 385
SEPs. See Somatosensory voked potentials
(SEPs)
Sepsis, 423
Septic shock
adults, 182
children, 182
description, 182
ECMO use, 183184
Severe respiratory failure
ECMO, 375
treatment, 281
Shafii, A.E., 295
Sheinberg, R., 141
Shin, T.G., 120
Short-term ventricuar assist devices
bridge to decision strategy, 220
ventricular assistance/biventricular support,
221
Silicon rubber membrane lungs (SRML), 465
SIRS. See Systemic inflammatory response
syndrome (SIRS)
Somatosensory voked potentials (SEPs)
anaesthetic agents, 394
cerebral hypoperfusion, 394
determination, cerebral ischaemia, 392
MLCEP, 395
Souilamas, R., 276
SRML. See Silicon rubber membrane lungs
(SRML)
T
TAH. See Total artificial heart (TAH)
Takayama, H., 221
Taub, J.O., 152
TAVI. See Transcatheter aortic valve
implantation (TAVI)
TBC. See Total body cooling (TBC)
TCD. See Transcranial Doppler (TCD)
Tchetchuline, S., 4
TEG. See Thromboelastography (TEG)
TEM. See Thromboelastometry (TEM)
Thoracic surgery
airway, 275
bronchopleural fistula, 277
BTT, 276277
cancer, 276
cannulation
adult, 274
paediatric patients, 274
cardiopulmonary system, 273
complications, 278
congenital diaphragmatic hernia, 274275
mediastinal masses, 275
PGD, 277
pulmonary
embolism, 275
infections, 275276
thromboendarterectomy, 278
thoracotomy, 278
trauma, 276
Thoracic surgery, 311
Thoracotomy
chest drain insertion, 278
femoral and internal jugular veins, 276
pulmonary contusions and haemothorax,
275
Thrombocytopenia, 420421, 438
Thromboelastography (TEG), 8384
[email protected]
488
Index
Thromboelastometry (TEM), 8384

Thromboembolism, 415416
Thrombotic complications, 7980
Tidal volume (Vt), 241
Total artificial heart (TAH)
biventricular and heart valve excision, 225
cardiac allograft failure, 226
Total body cooling (TBC), 329
Toyoda, Y., 295
Training, EUMO program, 1617
Transcatheter aortic valve implantation
(TAVI), 155
Transcranial Doppler (TCD), 390, 391, 393,
395, 397
Transfer
conventional transport, 265
H1N1, 269
hospitals to referral center, 265
Transplantation, cardiac failure
allografts, 227
cardiac xenotransplantation, 227
Heart, 227
immunosuppressive therapy, 226
Transthoracic and transesophageal
echocardiography (TTE-TEE),
143144
Trauma
ARDS, 184
timing, institution, 185
traumatic brain injury, 186
V-V ECMO support, 184
Traumatic brain injury, 186
TTE-TEE. See Transthoracic and
transesophageal echocardiography
(TTE-TEE)
U
Ultra low-flow CO2 removal, 467468
Ultrasound
ECMO cannulation, 365, 428
evaluation, 362, 363
monitoring, 361362
V
VADs. See Ventricular assist devices (VADs)
VA ECMO. See Veno-arterial ECMO
(VA-ECMO)
Vascular complications
leg ischemia, 417419
vascular access complications, 417, 418
Veno-arterial ECMO (VA-ECMO). See also

Respiratory monitoring, VA-ECMO
axillary vein and artery, 285
bi-femoral approach, 285
CA, 435
cannulation procedures, 363
cardiopulmonary resuscitation maneuvers,
363
central, 425
CO (see Cardiac output (CO) syndrome)
ECC components, 66
facilities, 7, 8
lung failure, 283
normal circulation
aortic and carotid baroreceptors,
non-pulsatile flow, 94
arterial blood flow, 94
artial spetostomy, 100
bypass, 94
cardiac failure, 100
coronary arterial flow, 97, 98
and IABP, 99100
left ventricle load and wall stress, 96,
99
optimal coronary oxygenation, 97
oxygen sypply, 100
parameters, blood flow and oxygen, 95
peripheral, 9697
pulmonary circulation, 94
RPM, 95
strategies, 97
SvO2 values, 95
venuous saturation, 95
percutaneous cannulation
accesses, 37
explantation, 4445
femoral, 43
implantation, 44
placement, 39
safe procedure, 43
peripheral, 426
pulmonary
circulation, 283
support, heart/lung, 66
refractory cardiac arrest, 364365
respiratory support, 7
retrieval, acute cardiogenic shock patient,
7, 8
survival and neurological outcome, 7
SvO2, 377378
vasoplegia, 434
Veno-venous ECMO (VV-ECMO)
[email protected]
Index
489
aerated lung volume, 259

arterial blood pressure, 287
bi-femoral approach, 285
blood gas analyses, 250251
configuration, 297, 298
creation, ARDS network, 65
ECC components, 65
echo-dynamic approach, 373
guidance, cannulae placement, 372
Harlequin syndrome, 379380
hemodynamic status, 251
hypoxaemia and SvO2, 379
imaging, 372
intrapulmonary shunt, 253
monitor respiratory function, 249
oxygenation, 283
oxygen desaturation, 286
papillomatous carinal mass, 282
percutaneous cannulation
accesses, 37
advantages, 41
avalon, 43
explanation technique, 4445
femoro-femoral approach, 4142
femoro-jugular, 43
placement, 39
persistent hypoxemia, 433
pulmonary circulation, 283, 372
pumpless, 297
recirculation, 433
respiratory dysfunction, 378
treatment, 65
venous oxygen content, 251
Venovenous (VV) femorofemoral approach,
352, 353
Ventilator-induced lung injury (VILI)
alveoli, 241
baby lung, 240
ECMO, 244
ventilatory strategies, 240
Ventricular assist devices (VADs)
AVM, 224
axial and centrifugal pumps, 223
bridge to decision strategy, 220
CF-VADs, 221
and ECMO, 141
electric motor drives, 222
gastrointestinal bleeding, 223224
Heydes syndrome, 224
implantation, 225
inflow cannula, 222
percutaneous devices, 221
rotor spins, 222

RV dysfunction, 131
thrombosis, 223
ventricular assistance/biventricular support,
221
Ventricular septal defect (VSD)
complications, post-myocardial infarction,
219
PI-VSD, 152, 155156
VILI. See Ventilator-induced lung injury
(VILI)
Viral etiology, 139
Vlasselaers, D., 203
Vogel, R.A., 152
von Willebrand disease, 86
VSD. See Ventricular septal defect (VSD)
VV-ECMO. See Veno-venous ECMO
(VV-ECMO)
W
Wallinder, A., 338
Ward, K.E., 200
Warm ischemia time (WIT), 328, 333
Weaning
biomarkers, 208
ECMO, 207208
VV-ECMO
assessment, patient readiness, 317318
decannulation, 321323
discontinuation, futility, 323
discontinuation procedure, 318319
failed trial, discontinuation, 320, 322
trial, discontinuation, 320, 321
Webb, J.G., 155
Wernly, J.A., 141
Wiebe, K., 306, 311
WIT. See Warm ischemia time (WIT)
X
X-Ray
cannula(e) position, 438439
pneumothorax management, 439
Z
Zapol, W., 3
Zych, B., 338
[email protected]

ECMO-Extracorporeal+Life+Support+in+Adul (Estudio)

Uploaded by

Copyright:

Available Formats

ECMO-Extracorporeal+Life+Support+in+Adul (Estudio)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

ECMO-Extracorporeal+Life+Support+in+Adul (Estudio)

Uploaded by

Copyright:

Available Formats

What are some of the main topics discussed in the document?

What are some of the main topics discussed in the document?

What are some of the main procedures and techniques mentioned?

What are some of the main procedures and techniques mentioned?

ECMOExtracorporeal Life

ECMO-Extracorporeal Life Support in Adults

Fabio Sangalli Nicol Patroniti

Extracorporeal membrane oxygenation (ECMO) is not a new technique. It has been

History and Technical Aspects

History of Extracorporeal Life Support . . . . . . . . . . . . . . . . . . . . . . .

Developing a New ECMO Program . . . . . . . . . . . . . . . . . . . . . . . . . .

Basic Aspects of Physiology During ECMO Support . . . . . . . . . . . .

Percutaneous Cannulation: Indication, Technique,

Surgical Cannulation: Indication, Technique, and Complications .

Materials: Cannulas, Pumps, Oxygenators . . . . . . . . . . . . . . . . . . . .

Coagulation, Anticoagulation, and Inflammatory Response . . . . . .

ECMO for Circulatory Support

Extracorporeal Life Support: Interactions

ECMO for Refractory Cardiac Arrest . . . . . . . . . . . . . . . . . . . . . . . .

ECMO for Postcardiotomic Shock . . . . . . . . . . . . . . . . . . . . . . . . . . .

ECMO in Myocarditis and Rare Cardiomyopathies . . . . . . . . . . . . .

ECMO for High-Risk Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ECMO for Severe Accidental Hypothermia . . . . . . . . . . . . . . . . . . . .

ECMO in Drug Intoxication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Newer Indications for ECMO: Pulmonary Embolism,

Left Ventricular Rest and Unloading During VA ECMO . . . . . . . . .

Weaning from Extracorporeal Circulatory Support . . . . . . . . . . . . .

Treatment Options for End-Stage Cardiac Failure . . . . . . . . . . . . . .

ECMO for Respiratory Support

Ventilatory Management of ARDS Before and During ECMO . . . .

Respiratory Monitoring of the ECMO Patient . . . . . . . . . . . . . . . . .

Structure of an ECMO Network for Respiratory Support . . . . . . . .

ECMO and Thoracic Surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ECMO in the Awake/Extubated Patient . . . . . . . . . . . . . . . . . . . . . . .

ECMO as a Bridge to Lung Transplant . . . . . . . . . . . . . . . . . . . . . . .

Low-Flow ECMO and CO2 Removal . . . . . . . . . . . . . . . . . . . . . . . . .

Weaning from VV ECMO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

ECMO for Organ Procurement

Heart-Beating and Non-Heart-Beating Donors . . . . . . . . . . . . . . . . .

Monitoring the ECMO Patient

Patient Care During ECMO . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Echocardiography in Venoarterial and Venovenous ECMO . . . . . .

Respiratory Monitoring During VA ECMO. . . . . . . . . . . . . . . . . . . .

Neurological Monitoring During ECMO . . . . . . . . . . . . . . . . . . . . . .

Monitoring the ECMO Patient: The Extracorporeal Circuit . . . . . .

Complications of Extracorporeal Support

Troubleshooting Common and Less Common Problems . . . . . . . . .

Transport of the ECMO Patient

Air Transport: Fixed-Wing and Helicopter . . . . . . . . . . . . . . . . . . . .

Ground Transport: Ambulance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Newer Indications and Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . .

History of Extracorporeal Life Support

ECMO (extracorporeal membrane oxygenation), also called ECLS (extracorporeal

F. Sangalli (*) C. Marzorati

direct-contact extracorporeal oxygenator that exposed a thin film of blood to air

History of Extracorporeal Life Support

Fig. 1.1 The first successful ECMO patient

Fig. 1.2 VV ECMO in Monza, early 1990s

History of Extracorporeal Life Support

The Artificial Lung

3.3.1 Oxygenation During VV ECMO

3.3.2 Oxygenation Support During VA ECMO

3.4.1 Carbon Dioxide Removal During ECMO Support

3.4.2 Carbon Dioxide Removal During VA ECMO

3.5.1 Hemodynamics During VV ECMO

3.5.2 Hemodynamics During VA ECMO

3.5.2.1 General Hemodynamic Changes

3.5.2.2 Peculiar Hemodynamic Changes Depending