2015 Revista Clinical Anesthesiology

CLINICAL
ANAESTHESIOLOGY
The Best Practice & Research series aims to provide a topical serial publication describing
and integrating the results from the latest original research articles into practical,
evidence-based review articles. These articles seek to address the key clinical issues of
diagnosis, treatment and patient management. Each issue follows a problem-orientated
approach which focuses on the key questions to be addressed, clearly defining what is
known and not known. Management is described in practical terms so that it can be
applied to the individual patient. The serial is aimed at the physician in both practice
and training.
Best Practice & Research Clinical Anaesthesiology is abstracted and indexed in the
following sources:
EMBASE
Excerpta Medica
Medline/Index Medicus
PubMed
Scopus
Editor-in-Chief
Hugo Van Aken, MD, PhD, FRCA, FANZCA
Department of Anesthesiology, Intensive Care Medicine and Pain Therapy
University Hospital Muenster
Albert-Schweitzer-Campus 1, Building A1
D-48149 Mnster
Germany
Editorial Board
A. Aitkenhead, UK H. Kehlet, Denmark
J. Andres, Poland P. Pelosi, Italy
D. C. H. Cheng, Canada J. Scholz, Germany
M. E. Durieux, USA S. Schug, Australia
D. Filipescu, Romania K. Shingu, Japan
A. W. Gelb, USA D. Spahn, Switzerland
S. Gelman, USA D. Story, Australia
Best Practice & Research Clinical Anaesthesiology 29 (2015) 97e98
Contents lists available at ScienceDirect
Best Practice & Research Clinical

Anaesthesiology
journal homepage: www.elsevier.com/locate/bean
Preface
All our progress is an unfolding, like a vegetable bud. You have rst an instinct, then an opinion,
then a knowledge as the plant has root, bud, and fruit. Trust the instinct to the end, though you can
render no reason. Ralph Waldo Emerson
Over the course of the last 60 years, tremendous achievements have been made in medical science.
We have developed new antibiotics to cure infections. We have new medications to successfully treat
diabetes mellitus. We have countless classes of medicines to treat hypertension. We have also made
tremendous advances in the surgical treatment of diseases including coronary artery disease, valvular
heart disease, congenital heart disease, and chronic heart failure. The initial development and success
of cardiopulmonary bypass (CPB) has been critical to this progress. Techniques and treatments once
thought inconceivable are now commonly encountered in operating theatres across the globe. Along
with the development of CPB to facilitate surgical repair, extracorporeal circulation (ECC) has pro-
gressed to provide novel options for treating and saving the lives of patients who might otherwise have
succumbed to severe, life-threatening illness such as sudden cardiac death, arrhythmias, and severe
respiratory infections such as the 2009 H1N1 inuenza pandemic [1,2].
The inception and progression of CPB and extracorporeal circulatory technology to the extent we
recognize today is detailed in the opening chapter of this issue. Despite its successful application and
use, CPB and extracorporeal technology comes at a cost to human physiology. In the last 20 years, the
effects of CPB and ECC on end organs of the human body have been studied extensively. The rst
portion of this issue of Best Practice & Research Clinical Anaesthesiology reviews the single common
denominator for complications related to CPB, the systemic inammatory response and immune
system including the pathophysiology, outcomes, and potential targets for prophylactic interventions
for the organ systems most affected by CPB and ECC including the neurological, hepatic, and renal
systems, as well as the heart and lungs themselves.
As a transition from outcomes to specialized care of patients utilizing CPB and ECC, updates are
presented on glycemic control, which is viewed by many physicians and regulatory societies across the
world as a primary quality measure for patients undergoing cardiac surgery. ECC exposes the native
blood elements to foreign material and thrombosis that must be avoided; thus, various anticoagulation
regimens along with novel anticoagulants and monitoring in patients associated with the use of
extracorporeal technology are reviewed. Advances in mechanical circulatory support are presented
including temporary and long-term support for both the left and right ventricles. Novel application of
extracorporeal membrane oxygenation (ECMO) with attention to the intensive care unit patient
population is reviewed. In what can only be viewed as the Bible for pediatric CPB, this issue's Chapter
11 provides an incredible resource and reference for anyone involved in the care of pediatric patients
undergoing heart surgery. Rounding out this issue of Best Practice & Research Clinical Anaesthesiology is
a nal chapter in which the pharmacologic approaches to weaning not only from CPB in the operating
theatre but also from ECMO in the intensive care unit are updated.
http://dx.doi.org/10.1016/j.bpa.2015.04.005
1521-6896/ 2015 Elsevier Ltd. All rights reserved.
98 Preface / Best Practice & Research Clinical Anaesthesiology 29 (2015) 97e98
Those who pioneered the beginnings of extracorporeal technology 60 years ago had an instinct, an
opinion, and gained knowledge over time. They worked tirelessly and had a vision of how their hard
work would enable patients with life-threatening heart diseases to undergo surgical repair to sustain
life. While they likely never dreamed that there would be articial hearts, ventricular assist devices, or
ECMO for tiny infants to adults, it is important to remember that those of us who practice now and in
the future have an incredible opportunity to continue to advance the practice of extracorporeal
technology. As mentioned in the opening chapter of this issue, much of what we know is based on years
of experience and not on robust evidence-based data. Our challenge for the future should be to improve
care for patients, and improve extracorporeal technology based on good data, basic research, and
appropriately powered randomized clinical trials. Hence, each chapter nishes with our ideas for
future research agendas. Imagine what we could accomplish if we could develop these ideas and
answer these important research questions.
It has been my sincere pleasure to put this issue together with the editorial staff at Elsevier as well as
the editorial board for Best Practice & Research Clinical Anaesthesiology. My genuine appreciation goes to
Gillian House (in the Elsevier Health Sciences Editorial Ofce) for her constant communication and
assistance. I have had the great fortune to invite world-renowned experts to contribute and thus
present to you, the reader, Developments in Extracorporeal Circulation.
References
[1] Peek GJ, Mugford M, Tiruvoipati R, et al. Efcacy and economic assessment of conventional ventilatory support versus
extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled
trial. Lancet 2009;374:1351e63.
[2] Combes A, Pellegrino V. Extracorporeal membrane oxygenation for 2009 inuenza A (H1N1)-associated acute respiratory
distress syndrome. Semin Respir Crit Care Med 2011;32:188e94.
Julie L. Huffmyer, MD, Assistant Professor Anesthesiology

University of Virginia, Department of Anesthesiology, Charlottesville, VA, USA
E-mail address: jh3wd@hscmail.mcc.virginia.edu

Anaesthesiology
History of cardiopulmonary bypass (CPB)

Eugene A. Hessel II, MD, FACS, Professor Anesthesiology,
Surgery (Cardiothoracic), Neurosurgery, and Pediatrics *, 1
University of Kentucky College of Medicine, Lexington, KY, USA
Keywords:
The development of cardiopulmonary bypass (CPB), thereby
extracorporeal circulation
permitting open-heart surgery, is one of the most important ad-
heartelung machines
cardiopulmonary bypass vances in medicine in the 20th century. Many currently practicing
cardiac surgery cardiac anesthesiologists, cardiac surgeons, and perfusionists are
history unaware of how recently it came into use (60 years) and how
much the practice of CPB has changed during its short existence. In
this paper, the development of CPB and the many changes and
progress that has taken place over this brief period of time, making
it a remarkably safe endeavor, are reviewed. The many as yet un-
resolved questions are also identied, which sets the stage for the
other papers in this issue of this journal.
2015 Elsevier Ltd. All rights reserved.
Introduction
The development of cardiopulmonary bypass (CPB) to permit cardiac surgery is considered one of
the greatest advances in medicine in the 20th century. Many do not appreciate how recently this has
developed. This has occurred during the lifetime of many who are still practicing. The rst successful
series of open-heart surgery utilizing heartelung (HeL) machines occurred exactly 60 years ago in the
spring and summer of 1955. In this article, the historical development and technological advancement
of extracorporeal circulation and CPB are reviewed. Much of this material is based upon or extracted
from other publications by this author [1e3], and based upon other papers and books related to the
history of cardiac surgery and CPB [4e15]. Cardiopulmonary Bypass Bibliography, covering 1667e1989,
* Tel.: 1 859 533 0756 (cell), 1 859 323 5956x80050 (ofce); Fax: 1 859 323 1080.
E-mail address: ehessel@uky.edu.
1
Administrative assistant: Charles York.
100 E.A. Hessel II / Best Practice & Research Clinical Anaesthesiology 29 (2015) 99e111
by Utley and colleagues is a particularly valuable resource [16]. Unfortunately, my review has likely
overlooked many of the important contributions that have occurred in other countries, if not published
in the English language literature.
The birth of use of mechanical HeL machines for CPB in cardiac surgery
The concept of ECC to support organ function was rst suggested and explored in animals in the
19th century, but many scientic discoveries and techniques were necessary before clinical CPB could
be achieved including blood types and transfusion, heparin and protamine, and the use of silicone
antifoam [3,11,17]. A number of surgical teams in the early 1950s attempted to develop HeL machines
and used them to perform open-heart surgery without success, but this was nally accomplished by
John Gibbon Jr. (Fig. 1A) on 6 May 1953 [18e20]. Dr. Gibbon was inspired to develop an HeL machine
while caring for a patient in 1930 as a surgical research fellow. Despite being dissuaded by his mentors,
he spent the next 20 years working alongside his wife and nally developed an HeL machine with the
assistance of the engineers at IBM (Fig. 1 B and C), with which he achieved 90% survival in dogs. This led
to his rst clinical application, which failed due to wrong diagnosis, but his second attempt, on an 18-
year-old woman with an atrial septal defect (ASD), was successful. Unfortunately, his next two patients,
died and he declared a moratorium on further application of his HeL machine. Only one other suc-
cessful open-heart surgery was performed using an HeL machine during 16 other attempts by six other
teams between 1952 and 1954, leading to an attitude of hopelessness. Some hypothesized that human
patients, as compared with animal subjects, were too sick to tolerate CPB. However, in March 1954, C.
Walton Lillehei (Fig. 2A) and his colleagues at the University of Minnesota initiated a remarkable series
of direct-vision intracardiac surgery with total CPB using another adult as the heartelung machine
(so-called controlled cross-circulation), in which the adult's femoral artery and vein were connected
to the child's arterial and venous system, respectively [21] (Fig. 2B). Over the next 16 months, they
operated on 45 seriously ill children with congenital heart disease with 28 survivors and thus clearly
demonstrated the potential for CPB to permit even complex open-heart surgery in sick patients once a
satisfactory articial HeL machine was available for use by a skillful cardiac surgical team. This was
accomplished during the spring and summer of 1955 when two groups, one led by John Kirklin
(Fig. 3A) at the Mayo Clinic starting on March 22nd [22] and the other by C. Walton Lillehei (Fig. 2A)
starting on May 13th [23], working 90 miles apart, using vastly different HeL machines and approaches
to the conduct of CPB, each operated on about 40 cases.
The Mayo Clinic group used the IBM-Gibbon machine, freely provided by Gibbon and IBM, which
they modied and called the MayoeGibbon HeL machine (Fig. 3B). Their HeL machine was capable
of oxygenating at ows of 2.4 L/min per square meter, and it included inline arterial and venous
saturation and a vaporizer for administering a volatile anesthetic. Lillehei's team employed a helical
reservoir bubble oxygenator (BO) (Figs. 2C and D) developed by a young physician, Richard A. DeWall,
who was working in the cardiac surgical research laboratory and as a perfusionist for the open-heart
cross-circulation procedures.
Lillehei and Kirklin were truly the giants of the development of open-heart surgery, with vastly
different styles and personalities. Kirklin was methodical and academic [24], whereas Lillehei was
impulsive and action oriented [14], but both made tremendous contributions to the practice of CPB.
While searching for a method to conduct open-heart surgery between 1950 and 1955, besides
inventing and successfully applying human cross-circulation, Lillehei helped his colleagues at the
University of Minnesota develop and apply clinically (but unsuccessfully) an HeL machine employing a
screen oxygenator and the rst successful use of circulatory arrest under moderate surface-induced
hypothermia to perform open-heart surgery. He also introduced epicardial pacing to treat surgically
induced heart block, and he pioneered the use of median sternotomy and femoral artery cannulation
for inow in the late 1950s. Before it was given a name, Kirklin practiced evidence-based cardiac
surgery and perfusion (show me the data), stressing the acquisition and the appropriate statistical
analysis of objective data. He authored or coauthored many papers, exploring the pathophysiology and
optimal conduct of CPB. He was the editor of The Journal of Thoracic and Cardiovascular Surgery from
1987 to 1994 [24].
E.A. Hessel II / Best Practice & Research Clinical Anaesthesiology 29 (2015) 99e111 101
Fig. 1. John Gibbon and the rst successful clinical use of the heartelung machine for open-heart surgery. 1A. John H Gibbon, Jr.
(1903e1973). Fig. 1 in Theruvath TP and Ikonomidis JS. Historical perspectives of The American Association for Thoracic Surgery:
John H. Gibbon, Jr. (1903e1973). Journal of Thoracic and Cardiovascular Surgery 2014; 147: 833-836. Used with permission from
Elsevier. 1B. Photo of Gibbon Model II HeL machine. Used by Gibbon for the rst successful clinical open-heart surgery with a
mechanical HeL machine, 6 May 1953. Note the stationary vertical screen oxygenator and roller pumps and elevation of HL machine
above the oor requiring active venous drainage. Figure from Romaine-Davis A. John Gibbon and His Heart-Lung Machine. Phila-
delphia, University of Pennsylvania Press, 1991. Used with permission. 1C. Diagram of Gibbon HeL machine. Stationary vertical
screen oxygenator (K); roller pumps for active venous drainage (D) and arterial return (P) and a third (E) for recirculation; (A) blood
reservoir, (B) venous pressure transducer, (C) automatic venous pump shutdown control, (F) oxygen input, (G) reservoir pressure
relief valve, (H) Oxygen exhaust, (J) stainless steel screens, (O) automatic electronic artery pump motor control, and (Q) reverse ow
lter. Figure 14 from Romaine-Davis A. John Gibbon and His Heart-Lung Machine. Philadelphia, University of Pennsylvania Press, 1991.
Used with permission from Wolters Kluwer Health, Inc.
In the last half of the 1950s, many groups initiated open-heart programs employing CPB, mainly to
treat congenital heart disease. Much has changed with regard to the equipment and components
employed and the conduct of CPB, which I have been fortunate to witness during my career in med-
icine, starting as a medical student in 1957, then as a cardiac surgeon until 1982, and subsequently as a
cardiac anesthesiologist.
Subsequent developments
Oxygenators (Fig. 4)
Early on, crude membrane oxygenators (MOs) and biologic oxygenators (animal lungs) (Fig. 4A)
were investigated, but most groups employed BOs of their own design or copied from DeWall's.
However, the rotating disc oxygenator developed by Kay and Cross in the late 1950s [25] became
widely used and favored by many for longer and more complex cases, but it required large priming
volume and reusable parts. Thus, many adopted the unitized plastic-bag disposable BOs (Fig. 4B),
which were introduced in the early 1960s and subsequently replaced by hard-shell disposable BO with
integrated heat exchangers (Fig. 4C). In the early 1970s, disposable but relatively inefcient MOs were
introduced separately by Lande and Kolobow; however, in the mid-1970s, more efcient and
commercially available microporous polypropylene MOs became available, and by the end of the
decade, these were being used in nearly 20% of cases. In the early 1980s, hollow-ber technology led to
the adoption of MOs, which virtually replaced all oxygenators by 1994. In the early 2000s, a new
nonporous true diffusion membrane constructed from poly-(4-methyl-1-pentene) (PMP) was intro-
duced, which is thought to be more biocompatible and more suitable for long-term perfusion (e.g.,
extracorporeal membrane oxygenation (ECMO)). In recent years, some manufactures have integrated
screen microlters into MOs, thereby perhaps eliminating the need for a separate arterial line
microlter.
Fig. 2. CW Lillehei and early CPB. 2A. C. Walton Lillehei. (1918e1999). Used with permission of the Lillehei Heart Institute at the
University of Minnesota. 2B. Lillehei's cross-circulation procedure using an adult as the HeL machine. The adult's femoral artery
was connected to the child's left subclavian artery. A single SVC cannula drained both SVC and IVC into the adult's femoral vein. A
single Sigmamotor pump controlled both the ow out of the adult into the child and the ow out of the child into the adult to keep
blood volumes in each stable. Figure 7 in Stoney WS. Evolution of Cardiopulmonary Bypass. Circulation 2009; 119; 2844-2853. Used
with permission from Wolters Kluwer Health. 2C. Photograph of HeL machine employing a DeWall helical reservoir bubble
oxygenator used by Lillehei for the rst series of open-heart surgery performed at the University of Minnesota in the summer of
1955. Note the use of Sigmamotor pumps. Fig. 2 in DeWall RA. The Evolution of the Helical Reservoir Pump- Oxygenator System at
the University of Minnesota. Ann Thorac Surg 2003; 76: S2210 e5. Used with permission from Elsevier. 2D. Diagram of early HeL
machine used at the University of Minnesota. Note the bubble oxygenator a with debubbling chamber, a helical coil debubbling
reservoir, and separate Sigmamotor venous and arterial pumps. Arterial inow into subclavian but venous drainage through separate
cannulas in IVC and SVC. Fig. 1 in DeWall RA. The Evolution of the Helical Reservoir Pump- Oxygenator System at the University of
Minnesota. Ann Thorac Surg 2003; 76: S2210 e5. Used with permission from Elsevier.
Arterial pumps
Initially, Sigmamotor (nger) pumps were commonly used, but they were quickly replaced with
roller pumps. However, in the 1980s, perhaps due to their use in ECMO and as ventricular assist de-
vices, and their purported improved safety and reduced blood trauma, centrifugal pumps began to
Fig. 3. John Kirklin and early CPB. 3A. John W Kirklin (1917e2004). Circa 1957, 2 years after accomplishing the rst successful series
of clinical open-heart surgeries using an HeL machine (MayoeGibbon). Fig. 1, from Stephenson L J. Historical perspective of The
American Association for Thoracic Surgery: John W. Kirklin, MD (1917e2004). J Thorac Cardiovasc Surg. 2007; 134: 225-8. Used with
permission from Elsevier. 3B. The MayoeGibbon HeL machine employed for open-heart surgery at the Mayo Clinic in the spring of
1955. Figure 11 in Stoney WS. Evolution of Cardiopulmonary Bypass. Circulation 2009; 119; 2844-2853. Used with permission from
Wolters Kluwer Health, Inc.
compete with roller pumps. Throughout the development of CPB, there has been a continuing debate
about the advantages of providing pulsatile ow with numerous preclinical and clinical studies
providing conicting data [26]. Pulsatile modes of CPB seem to be more popular in Europe than in the
USA.
Arterial cannulation
In the early days of cardiac surgery, which was most commonly performed through a lateral tho-
racotomy or an transverse sternal bilateral anterior thoracotomy, arterial inow was directed into the
Fig. 4. Early oxygenators. 4A. Mustard's HeL machine employing a biologic (monkey) lung. (Clinically unsuccessful.) Rhesus monkey
lungs were suspended in the glass ask and used as the oxygenator. Fig. 3 in Stoney WS. Evolution of Cardiopulmonary Bypass.
Circulation 2009; 119; 2844-2853. Used with permission from Wolters Kluwer Health, Inc. 4B. Disposable plastic sheet bubble
oxygenator with reservoir. Modied from Fig. 3 in DeWall RA. The Evolution of the Helical Reservoir Pump- Oxygenator System at
the University of Minnesota. Ann Thorac Surg 2003; 76: S2210 e5. Used with permission from Elsevier. 4C. Hard-shell disposable
bubble oxygenator with integrated heat exchanger (Bentley Temptrol). Modied from Figure 6 in DeWall RA. The Evolution of the
Helical Reservoir Pump- Oxygenator System at the University of Minnesota. Ann Thorac Surg 2003; 76: S2210 e5. Used with
permission from Elsevier.
subclavian artery, but by the end of the 1950s femoral artery cannulation became dominant. By the late
1960s and early 1970s, direct cannulation of the ascending aorta, advocated by DeWall in 1963 [27],
became standard, partly related to the more common use of median sternotomy and because of the
signicant incidence of retrograde dissection associated with femoral cannulation. This complication
reappeared with the introduction of port-access surgery in the mid-1990s. The reason why cannu-
lating the ascending aorta was not adopted for a long time remains an unanswered question to the
author. Because the role of dislodging atheroemboli and the management of ascending aortic dissec-
tions were of concern, the use of subclavian cannulation was reintroduced in the mid-1990s [28].
Venous cannulation and drainage
Initially active venous drainage accomplished and controlled by a separate venous pump was used
mainly to help balance venous and arterial ow, but soon gravity drainage became standard. This
remained true until the reintroduction of augmented venous return, with either pumps (kinetic) [29]
or vacuum [30], in the mid-1990s, initially to facilitate drainage through the long narrow femoral
venous cannulas used for port-access minimally invasive cardiac surgery, but now often used to
facilitate reduced prime volume by utilizing smaller venous lines and raising the level of the HeL
machine up from the oor level.
Temperature management
During its early days, CPB was conducted at normothermia, but in the late 1950s Sealy and others
introduced efcient heat exchangers utilizing moderate hypothermia (28e32 C). This rapidly became
standard practice both to improve the tolerance to the limited ow and oxygenation capacity of early
HeL machines and to hemodilution and for neuroprotection. Until the early 1980s, temperature-
corrected (i.e., corrected to the patient's actual temperature) pH and PaCO2 were kept at 7.40 and
40, respectively (so-called pH-stat strategy). However, in 1982, Ream and colleagues advocated the use
of alpha-stat management to produce alkalemia and hypocapnea during hypothermia [31], and alpha-
stat quickly became standard practice. The use of moderate hypothermia during CPB remained com-
mon until normothermic or tepid bypass, advocated by the group in Toronto, Canada [32], became
popular starting in the 1990s, partly to reduce the risk of cerebral hyperthermia and injury during
rewarming; this led to emphasis on monitoring nasopharyngeal instead of rectal or bladder temper-
ature [33] with the recommendation of maintaining the arterial inow temperature at 37 C.
In 1959, Drew in London, UK, introduced the technique of deep hypothermic circulatory arrest
(DHCA) [34]. DHCA was not widely adopted at the time, but it was reintroduced in the mid- and late
1960s separately by Dillard [35] and Barratt-Boyes [36] to facilitate infant heart surgery and by Griepp
in 1975 for conducting aortic arch surgery [37].
Anticoagulation and bleeding
Until 1975, heparin was administered by protocol without any objective monitoring of its effect. In
that year, Bull at Loma Linda University documented the limitations of this approach and advocated the
use of objective monitoring with the activated clotting time (ACT) being aimed at a rather arbitrary
target of 480 s [38]. The proper level of anticoagulation and the method of monitoring it continue to be
debated, but by the end of the 1970s, some type of objective monitoring and control of heparinization
was employed by most teams. In 1987, Royston [39] and other groups introduced the use of aprotinin to
reduce perioperative bleeding, and soon the use of aprotinin and other synthetic anti-brinolytics
during CPB became ubiquitous, although aprotinin was withdrawn from the market in 2007 due to
safety concerns. Heparin-coated circuits were introduced partly with the potential of reducing heparin
requirements, but this has remained controversial. In recent years, the nearly ubiquitous use of drugs
affecting platelet function and the recognition and management of heparin-induced thrombocyto-
penia have led to the development and use of new point-of-care monitoring of coagulation and platelet
function during CPB (e.g., thromboelastography (TEG)) and the introduction of alternatives to heparin
for conduct of CPB (e.g., bivalirudin). In 2007, cardiac surgeons, anesthesiologists, and perfusionists
collaborated on the development and publication of blood management guidelines which were
updated in 2011 [40].
Hemodilution, anemia, and hemoltration
Initially, whole blood (and a large volume of it!) was used to prime HeL machines. However, in the
late 1950s and the early 1960s, several groups introduced hemodilution and asanguineous priming to
reduce the adverse effects of blood, conserve this valuable resource, and improve tissue perfusion
during hypothermia. Thereafter, hemodilution, sometimes to extreme degrees, became standard
practice. This became problematic with the adoption of tepid or normothermic bypass, and since 1997
a number of observational studies have detected increased morbidity and mortality associated with
low hematocrit during CPB. Conversely, the adverse effects of RBC transfusion (beyond the infectious
risks) were becoming more apparent. Thus, interventions to reduce the severity of anemia during CPB
were instituted including the use of retrograde arterial priming (RAP) in 1998 [41], reduction of the
priming volumes of extracorporeal circuit (ECC), and the development and use of miniaturized circuits.
The use of hemoltration during CPB was rst reported by Darup and colleagues in 1979 [42] to remove
excess crystalloid, to raise hematocrit and plasma protein concentration, and to remove inammatory
mediators. In 1991, Naik et al. introduced the practice of modied ultraltration (MUF) to the practice
of pediatric CPB [43], and in 1996 Journis et al. advocated zero-balance ultraltration (Z-BUF) during
CPB [44].
Modifying the ECC and conduct of CPB to improve organ preservation
Gross air embolism was an early concern during CPB (and the cause for the failure of a number of
early attempts at using CPB); thus, most early HeL machines included a macro-bubble trap on the
arterial line. However, in the early 1970s, recognition of the presence and signicance of microemboli
including gaseous microemboli (GME), especially by Roy Swank and Russell Patterson, led to the
development and placement of microlters on both the cardiotomy suction and the arterial line. The
in-depth Dacron wool lter developed by Swank was soon supplanted by the screen lter from Pat-
terson. By the mid-1990s, the use of arterial line microlters was nearly 100%. Because activated
leukocytes are thought to be a major contributor to the inammatory response (systemic inammatory
response syndrome (SIRS)) associated with CPB, the use of leukocyte-depleting lters in various sites in
the ECC was introduced in the 1990s; however, the lack of high-level evidence for benets to clinically
important outcomes [52] has curtailed their widespread use. In the 1990s, more attention was redir-
ected at the cardiotomy suction as a major source of microemboli and of activation of other blood
components. This led to the emphasis on minimizing or eliminating the use of cardiotomy suction
when possible or the preprocessing of cardiotomy suction blood before returning it to the ECC using
cell saver/processors. The latter were rst introduced by Jack Latham and his company, Haemonetics, in
1971.
Optimal conduct of CPB for brain protection has been of concern since the early days of CPB. In the
early 1980s, the groups at University of Alabama and Copenhagen reported on the objective mea-
surement of cerebral blood ow during CPB [53,54], which partly led to the adoption of alpha-stat pH/
PaCO2 management [31,55]. In 1986, Nussmeier and colleagues [56] reported that barbiturate
administration decreased the incidence of neurologic injury during cardiac surgery. Although not
conrmed by a subsequent study [57], this observation stimulated investigation of other drugs and
methods to reduce adverse neurologic sequelae. In 1992, the Washington University group called
attention to the probable role of ascending aortic atherosclerosis and manipulation of the aorta, and
they introduced the use of epivascular ultrasound scanning to guide cannulation [58]. Recognition of
the increased incidence of neurologic dysfunction of the brain during longer periods of circulatory
arrest during deep hypothermia led Ueda et al. [59] to resurrect retrograde cerebral perfusion (rst
described by Mills and Ochsner in 1980 to treat massive cerebral air embolism [60]). Shortly thereafter,
antegrade cerebral perfusion was advocated as a superior method of cerebral protection during
induced circulatory arrest. Strategies for neurocognitive protection are reviewed in a subsequent article
in this issue.
To protect the heart, initially brief periods of ischemic arrest were often induced by administration
of high concentrations of potassium into the aortic root [61]. With the advent of aortic valve
replacement surgery in the early 1960s, continuous perfusion of oxygenated blood from the HeL
machine directly into the coronary ostia was common practice, although Shumway at Stanford Uni-
versity advocated simply topical cooling of the heart with ice and ice-cold uid. This did not always
reliably protect hypertrophied left ventricles, and in 1972 Cooley et al. described the phenomenon of
ischemic contracture of the LV, so-called stone heart. In 1973, Gay and Ebert [62] in the United States
and others in England and Europe introduced intermittent infusion of cold low-concentration potas-
sium crystalloid cardioplegia, and this was rapidly adopted to facilitate all types of cardiac surgery.
Cardioplegia was initially administered antegrade, but in the early 1980s retrograde administration
was introduced and cold blood cardioplegia was introduced in the late 1970s. Although warm blood
cardioplegia was discussed in the late 1970s, it was not until the early 1990s that it was strongly
advocated, especially by the Toronto group [32]. Myocardial protection is discussed further in a sub-
sequent article in this issue.
In 1981, Kirklin's group at University of Alabama identied the systemic inammatory reaction (SIR)
induced by CPB as a cause for much of the morbidity associated with cardiac surgery [63]. This
contributed to the introduction of drugs such as aprotinin, heparin-coated circuits, hemoltration,
leukoltration, low-surface-area minimized circuits, minimized use of cardiotomy suction, and off-
pump coronary artery bypass graft (CABG). The inammatory response related to CPB is discussed
more thoroughly later in this issue.
Monitoring
Initially, left heart lling was estimated by CVP, but this was soon recognized as being unreliable and
placement of the left atrial (LA) line for direct monitoring of the left atrial pressure (LAP) was intro-
duced by the group at the Mayo Clinic. Placement of an LA line was widely practiced until supplanted
by the use of the pulmonary artery catheter (PAC) developed by cardiologists Swan and Ganz in 1970.
The PAC was quickly adopted in cardiac surgery by the groups at MGH and Emory, and it soon became
standard practice until the introduction of transesophageal echocardiography (TEE) to cardiac surgery
in the 1980s. Perioperative TEE was introduced initially, at least in the USA, by anesthesiologist Michael
Cahalan at the University of California Medical School in San Francisco [45], and followed soon
thereafter by multiple other groups [46e48]. TEE is used not only to guide cardiac surgery but also to
assist with conduct of CPB (e.g., assessing atherosclerosis in the aorta, cannulation strategies,
decompression of the left ventricle and residual air, and possible aortic dissection.)
Although the early HeL machines used by the Mayo group included monitors of arterial and venous
oxygen saturation, these, along with low-level monitors and bubble detectors, were only slowly
adopted until they became standard [49]. In recent years, manufacturers have provided data man-
agement and computerized monitoring and alarm systems, including some that control arterial pump
ows, to HeL machine consoles. Although the Mayo Clinic group employed electroencephalographic
(EEG) monitoring in their early experience, this was not widely practiced and little cerebral monitoring
was used until the introduction of processed EEG (e.g., bispectral index) and cerebral oximetry (by
Somanetics) in the early 1990s. Murkin's trial demonstrating the benet of cerebral oximetry [50] has
led some to advocate routine cerebral monitoring [51], and although widely practiced its use remains
controversial because of the lack of high-level evidence.
Less invasive cardiac surgery
In an effort to reduce some of the adverse effects and cost associated with CPB, and to compete with
percutaneous coronary intervention, a revolution started by the Swiss cardiologist Andreas Gruntzig
with his letter to the editor of Lancet in 1978 [64], cardiac surgeons Benetti of Argentina [65] and
Buffolo of Brazil [66] began performing direct coronary revascularization without the use of CPB (off-
pump coronary artery bypass (OP-CAB)) in the late 1970s and the early 1980s. Others began adopting
this practice in the 1990s, and soon it was widely practiced by groups throughout the world. Although
there continues to be debate about the benets and proper indication for OP-CAB, one notable
observation has been that many of the complications, including stroke, neurocognitive decline, renal
failure, inammatory responses, and other morbidity and mortality, were only modestly reduced or not
reduced as compared with CABG accomplished with CPB, suggesting that CPB is not the major cause of
the complications associated with cardiac surgery.
In 1996, the port-access system was introduced for minimally invasive cardiac surgery [67]. Its use
was associated with some unanticipated morbidity (including the reappearance of retrograde aortic
dissection) and mortality, but it led to the further development of minimal-access surgery. This led to
the reintroduction of augmented venous drainage, required by the use of long and narrow venous
cannulas, yet it was also associated with some catastrophic complications. These unfortunate expe-
riences emphasized the risk of the premature adoption of new techniques prior to adequate evaluation
and the need to carefully prepare before the introduction of any changes in CPB practice. Partly in
response to the challenge of OP-CAB, beginning in 2002 [68], minimized extracorporeal circuits began
to be offered by several different manufactures and evaluated by many different groups. However, their
use posed other limitations on the conduct of CPB and demanded specialized care by the entire team,
and thus it has not been widely adopted in the USA.
Safety and accident surveys
Risk and risk containment has been a concern since the beginning of clinical CPB, but interest
became more focused with the publication of the rst national survey of perfusion accidents by Stoney
et al., in 1980 [69] followed by multiple subsequent surveys in various countries over the next 20 years
[70e72]. These have led to additional educational efforts and adoption of safe practices and safety
equipment on HeL machines, and promulgation of guidelines such as those issued by the American
Society of ExtraCorporeal Technology (AmSECT) International Consortium for Evidence-Based Perfu-
sion [49]. The Australian and New Zealand College of Perfusionists (ANZCP) has initiated a model
Perfusion Incident Reporting System (PIRS). In 2007, the Society of Cardiovascular Anesthesiologists
(SCA) started the FOCUS (Flawless Operative Cardiovascular Unied Systems) initiative through the
SCA Foundation [73], and in 2013 the American Heart Association (AHA) published a scientic state-
ment on patient safety in the cardiac operating room emphasizing human factors and teamwork [74].
Perfusionists
In the early days, HeL machines were managed by surgeons (often residents) and laboratory
technicians working in the research laboratories. Gradually, at least in the USA, this task was taken over
by technicians drawn from various elds, who received on-the-job training. A group of these
perfusionists (a designation credited to Bennett Mitchell) began meeting in 1964, and they estab-
lished the American Society of ExtraCorporeal Technology in 1968. In the late 1960s and the early
1970s, the demand for perfusionists increased dramatically with the introduction of coronary artery
surgery and more successful valve surgery. In 1974, AmSECT established the American Board of Car-
diovascular Perfusion (ABCP).The rst baccalaureate program in perfusion was established by James
Dearing at the Ohio State University in 1969 and the famed Texas Heart Institute School of Perfusion
was established by Charles Reed in 1971. In 1991, the European Board of Cardiovascular Perfusion was
established.
A survey of 811 perfusionists practicing in North America (USA and Canada) in 1980 by this author
and his colleagues provides a snapshot of the nature of perfusion practice 25 years after it began [75].
Although 68% received their training on the job, 76% were ABCP certied. Disposable BOs were used in
>90% of cases, roller pumps in 94%, and arterial lters in 64%. Low-level alarms were used in 45% of
cases, oxygen sensors in 26%, and bubble alarms in 10%. Asanguineous priming was used in 72%. Two-
thirds monitored heparinization prior to initiation of bypass, mostly with the ACT. Most employed
moderate hypothermia (26e30 ), and 84% added carbon dioxide to the oxygenator. Cold cardioplegia
was employed in 99% of cases, but 50% of the time it was administered by a pressurized bag (usually
controlled by the anesthesiologist). Changes in these practices are reected in more recent surveys of
practice in the USA and elsewhere [70e72,76e78].
Evidence-based practice and guidelines
The concept of basing medical practice on a critical review of the best evidence available in the
medical literature was developed and promulgated by David Sakett and his colleagues at the McMaster
University in Hamilton, Ontario, Canada beginning in 1981. This group included GH Guyatt who is
credited with coining the term evidence-based medicine (EBM). They posited a hierarchy of levels of
evidence, with a meta-analysis of several large well-conducted randomized controlled trials (RCTs) at
the top. In recent years, the practice of EBM has been adopted for CPB. Unfortunately, as in many other
elds of medicine, the existence of high-level evidence to guide most aspects of CPB is lacking [26,79],
and there remains a great need and opportunities to generate such evidence to guide the conduct of
CPB, especially in high-risk patients [3,26]. The development of small animal models (e.g., rabbit and
rat) of CPB in the 1990s and the early 2000s has permitted the detailed analysis of the impact of various
aspects of the conduct of CPB [80,81]. In the late 1980s, Mangano and colleagues initiated large
multicenter studies involving thousands of patients, which have generated important observational
data related to the conduct of cardiac surgery and CPB, as has the Northern New England Cardiovas-
cular Disease Study Group. To promulgate EBM practice of CPB, the ANZCP inaugurated the annual
Perfusion Down Under and AmSECT established the International Consortium for Evidence Based
Perfusion in 2007, and several EBM guidelines for conduct of CPB have been published in recent years
[26,40,49,82,83].
Education and simulation
In the early days, education mainly occurred through meetings of surgical and cardiology societies
and publications in their journals. AmSECT inaugurated the rst journal devoted to CPB (Journal of
Extracorporeal Perfusion) in 1966, and a second journal, Perfusion, was inaugurated in 1986. In 1962,
Galletti and Brecher published the rst comprehensive textbook on CPB that served as the denitive
treatise on CPB until other texts began to appear about 13 years later. Meetings devoted to CPB were
rst sponsored by AmSECT. In 1978, the SCA was established, which quickly devoted much attention to
education regarding CPB. In 1980, cardiac surgeon Joe Utley initiated the San Diego Cardiothoracic
Surgery Symposium mainly devoted to the conduct of CPB.
As in other elds of medicine and high-tech industries, the development of simulators and the use
of simulation for training and evaluation have recently expanded into the practice of CPB. However, the
use of computer simulation for basic education in perfusion was rst described by Riley and O'Kane at
the Mayo Clinic in 1977 [84]. The use of simulation was greatly enhanced by the development of the
Orpheus simulator by Morris and Pybus in 2007 [85], and in the last decade various low- and high-
delity simulators have been introduced and many groups have demonstrated their value in intro-
ductory perfusion education, maintenance of core skills, the practice of emergency situation responses,
and the development of team resource management and multidisciplinary team education. This
stimulated the AmSECT to establish a Simulation Taskforce in 2011.
Summary
Practical clinical CPB began only 60 years ago, and tremendous changes and advances have been
made over that period of time, making it a remarkably safe and effective procedure (as this author can
attest to personally, after his CABG utilizing CPB about 5 years ago.) Over this period, some accepted
practices have been discarded (e.g., pH-stat strategy for blood gas management, routine use of mod-
erate hypothermia, and prolonged DHCA), whereas previously discarded techniques have been
brought back (e.g., subclavian artery inow, potassium cardioplegia, and augmented venous return). In
recent years, enhanced attention has been directed to safety, including devices and monitors for the
ECC, establishment of guidelines, and enhanced education and evaluation including introduction of
simulation. An important lesson is that the premature use of new equipment or techniques can lead to
unexpected adverse consequences. Finally, much of our practice of CPB is not based upon a high level of
evidence, and thus there is great opportunity for future research and improvement, therefore adding to
this fascinating history of CPB.
Practice points
1. CPB has evolved over a short period of time, and it has become remarkably safe but is far
from perfect.
2. The practice of CPB including choice of equipment and conduct is largely based on expe-
rience rather than high-level evidence.
3. The entire team (surgeon, perfusionist, and anesthesiologist) should be involved in selecting
the equipment and deciding how to monitor and conduct CPB
4. Thorough evaluation of new techniques and equipment should be conducted before intro-
ducing them into one's clinical practice.
Research agenda
1. High-level evidence is still needed to guide selection of equipment and conduct of CPB and
to identify in which patients these are critical
2. Some of the many unresolved issues include desirability of pulsatile flow, benefits and use
of minimized circuits, importance of excluding cardiotomy suction, optimal hematocrit, type
of arterial pump, need for and optimal surface coating, importance of GME and optimal
filtration, and role of leukocyte-depleting filters.
Funding
The author wrote this entire manuscript and received no outside funding.
Conict of interest
None.
References
*[1] Hessel 2nd EA. History of cardiac surgery and anesthesia. In: Estafanous FG, Barash PG, Reves JG, editors. Cardiac
anesthesia: principles and clinical practice. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2001. p. 3e35.
*[2] Hessel 2nd EA. Evolution of cardiac anesthesia and surgery. In: Kaplan JA, Reich DL, Lake CL, et al., editors. Kaplan's
cardiac anesthesia. 5th ed. Philadelphia: PA Saunders/Elsevier; 2006. p. 3e32.
*[3] Hessel II EA. A brief history of cardiopulmonary bypass. Semin Cardiothorac Vasc Anesth 2014;18(2):87e100.
[4] American Society of ExtraCorporeal Technology. History and Heritage of cardiovascular surgery, perfusion and the
american society of ExtraCorporeal technology. 1885e2011. Richmond, VA: American Society of ExtraCorporeal Tech-
nology; 2012.
*[5] Shumacker Jr HB. The evolution of cardiac surgery. Bloomington, IN: Indiana University Press; 1992.
[6] Shumacker HB. The birth of an idea and the development of cardiopulmonary bypass. In: Gravlee GP, Davis RF,
Stammers AH, et al., editors. Cardiopulmonary bypass: principles and practice. 3rd ed. Philadelphia, PA: Wolters Kluwer/
Lippincott Williams & Wilkins; 2008. p. 21e32.
[7] Lillehei CW. Historical development of cardiopulmonary bypass in Minnesota. In: Gravlee GP, Davis RF, Stammers AH,
et al., editors. Cardiopulmonary bypass. principles and practice. 3rd ed. Philadelphia, PA: Wolters Kluwer/Lippincott
Williams & Wilkins; 2008. p. 3e20.
[8] Westaby S, Bosher C. Landmarks in cardiac surgery. Oxford, England: Isis Medical Media; 1997.
[9] Edmunds Jr LH. The evolution of cardiopulmonary bypass: lessons to be learned. Perfusion 2002;17:243e51.
[10] Galletti PM. Cardiopulmonary bypass: a historical perspective. Artif Organs 1993;17:675e86.
[11] Galletti PM, Mora CT. Cardiopulmonary bypass: the historical foundation, the future promise. In: Mora CT, editor. Car-
diopulmonary bypass. Berlin, Germany: Springer- Verlag; 1995. p. 3e21.
[12] Stoney WS. Evolution of cardiopulmonary bypass. Circulation 2009;119:2844e53.
[13] Cooper DKC. Open heart. The radical surgeons who revolutionized medicine. New York, NY: Kaplan; 2010.
[14] Miller GW. King of hearts. The true story of the Maverick who pioneered open heart surgery. New York, NY: Crown; 2000.
[15] Utley JR. Early development of cardiopulmonary bypass. Perfusion 1986;1:1e14.
*[16] Utley JR, Johnson HD, Morgan MS. Cardiopulmonary bypass bibliography. Spartanburg, South Carolina: Cardiothoracic
Research and Education Foundation; 1989.
[17] Comroe Jr JH, Dripps RD. Ben Franklin and open heart surgery. Circ Res 1974;35:661e9.
[18] Romaine-Davis A. John Gibbon and his heart-lung machine. Philadelphia: University of Pennsylvania Press; 1991.
*[19] Shumacker Jr HB. A dream of the heart. The life of john H. Gibbon, jr. Father of the heart-lung machine. Santa Barbara, CA:
Fithian Press; 1999.
[20] Gibbon Jr JH. Application of a mechanical heart and lung apparatus to cardiac surgery. Minn Med 1954;37(3):171e85.
[21] Lillehei CW, Cohen M, Warden HE, et al. The direct-vision intracardiac correction of congenital anomalies by controlled
cross circulation; results in thirty-two patients with ventricular septal defects, tetralogy of Fallot, and atrioventricularis
communis defects. Surgery 1955;38(1):11e29.
[22] Kirklin JW, Dushane JW, Patrick RT, et al. Intracardiac surgery with the aid of a mechanical pump-oxygenator system
(Gibbon type): report of eight cases. Proc Staff Meet Mayo Clinic 1955;30:201e6.
[23] Lillihei CW, DeWall RA, Gott VL, et al. The direct vision correction of calcic aortic stenosis by means of a pump-
oxygenator and retrograde coronary sinus perfusion. Dis Chest 1956;30(2):123e32.
[24] Stephenson LW. Historical perspective of the american association for thoracic surgery: john W. Kirklin, MD (1917-2004).
J Thorac Cardiovasc Surg 2007;134(1):225e8.
[25] Cross FS, Kay EB. Direct vision repair of intracardiac defects utilizing a rotating disc reservoir-oxygenator. Surg Gynecol
Obstet 1957 Jun;104(6):711e6.
*[26] Murphy GS, Hessel 2nd EA, Groom RC. Optimal perfusion during cardiopulmonary bypass: an evidence-based approach.
Anesth Analg 2009;108:1394e417.
[27] DeWall RA, Levy MJ. Direct cannulation of the ascending aorta for open-heart surgery. J Thorac Cardiovasc Surg 1963;45:
496e9.
[28] Sabik JF, Lytle BW, McCarthy PM, et al. Axillary artery: an alternative site of arterial cannulation for patients with
extensive aortic and peripheral vascular disease. J Thorac Cardiovasc Surg 1995;109:885e90.
[29] Toomasian JM, McCarthy JP. Total extrathoracic cardiopulmonary support with kinetic assisted venous drainage: expe-
rience in 50 patients. Perfusion 1998;13:137e43.
[30] Taketani S, Sawa Y, Masai T, et al. A novel technique of cardiopulmonary bypass using vacuum system for venous drainage
with pressure relief valve: an experimental study. Artif Organs 1998;22:337e41.
[31] Ream AK, Reitz BA, Silverberg G. Temperature correction of PCO2 and pH in estimating acid-base status: an example of
the emperor's new clothes? Anesthesiology 1982;56:41e4.
[32] Lichtenstein SV, Ashe KA, Dalati H, et al. Warm heart surgery. J Thorac Cardiovasc Surg 1991;101:269e74.
[33] Nathan HJ, Lavallee G. The management of temperature during hypothermic cardiopulmonary bypass: Canadian survey.
Can J Anaesth 1995;42:669e71.
[34] Drew CE, Anderson IM. Profound hypothermia in cardiac surgery: report of three cases. Lancet 1959;1(7076):748e50.
[35] Dillard DH, Mohri H, Hessel 2nd EA, et al. Correction of total anomalous pulmonary venous drainage in infancy utilizing
deep hypothermia with total circulatory arrest. Circulation 1967;35(4 Suppl):I105e10.
[36] Barratt-Boyes BG, Simpson M, Neutze JM. Intracardiac surgery in neonates and infants using deep hypothermia with
surface cooling and limited cardiopulmonary bypass. Circulation 1971;43(5 Suppl):I25e30.
[37] Ergin MA, O'Connor J, Guinto R, et al. Experience with profound hypothermia and circulatory arrest in the treatment of
aneurysms of the aortic arch. Aortic arch replacement for acute arch dissections. J Thorac Cardiovasc Surg 1982;84(5):
649e55.
[38] Bull BS, Korpman RA, Huse WM, et al. Heparin therapy during extracorporeal circulation. I. Problems inherent in existing
heparin protocols. J Thorac Cardiovasc Surg 1975;69(5):674e84.
[39] Royston D, Bidstrup BP, Taylor KM, et al. Effect of aprotinin on need for blood transfusion after repeat open-heart surgery.
Lancet 1987;2(8571):1289e91.
[40] Society of Thoracic Surgeons Blood Conservation Guideline Task Force, Ferraris VA, Brown JR, et al. 2011 update to the
society of thoracic surgeons and the society of cardiovascular anesthesiologists blood conservation clinical practice
guidelines. Ann Thorac Surg 2011;91:944e82.
[41] Rosengart TK, DeBois W, O'Hara M, et al. Retrograde autologous priming for cardiopulmonary bypass: a safe and effective
means of decreasing hemodilution and transfusion requirements. J Thorac Cardiovasc Surg 1998;115(2):426e38.
[42] Darup J, Bleese N, Kalmar P, et al. Hemoltration during extracorporeal circulation (ECC). Thorac Cardiovasc Surg 1979;
27(4):227e30.
[43] Naik SK, Knight A, Elliott MJ. A successful modication of ultraltration for cardiopulmonary bypass in children. Perfusion
1991;6(1):41e50.
[44] Journois D, Israel-Biet D, Pouard P, et al. High-volume, zero-balanced hemoltration to reduce delayed inammatory
response to cardiopulmonary bypass in children. Anesthesiology 1996;85(5):965e76.
[45] Cahalan MK, Kremer P, Schiller NB, et al. Intraoperative monitoring with two-dimensional transesophageal echocardi-
ography (Abst.). Anesthesiology 1982;57:A-153.
[46] Kaplan JA. Transesophageal echocardiography. Mt Sinai J Med 1984;51(5):592e4.
[47] Cahalan MK, Litt L, Botvinick EH, et al. Advances in noninvasive cardiovascular imaging: implication for the anesthesi-
ologist. Anesthesiology 1987;66:356e72.
[48] Clements FM, deBruijn NP. Perioperative evaluation of regional wall motion by transesophageal two-dimensional
echocardiography. Anesth Analg 1987;66:249e61.
*[49] Baker RA, Bronson SL, Dickinson TA, et al. Report from AmSECT's international consortium for evidence-based perfusion:
American society of ExtraCorporeal technology standards and guidelines for perfusion practice. J Extra Corpor Technol
2013;45:156e66.
[50] Murkin JM, Adams SJ, Novick RJ, et al. Monitoring brain oxygen saturation during coronary bypass surgery: a randomized,
prospective study. Anesth Analg 2007;104:51e8.
[51] Edmonds Jr HL. 2010 standard of care for central nervous system monitoring during cardiac surgery. J Cardiothorac Vasc
Anesth 2010;24:541e3.
[52] Warren O, Alexiou C, Massey R, et al. The effects of various leukocyte ltration strategies in cardiac surgery. Eur J Car-
diothorac Surg 2007;31(4):665e76.
[53] Henriksen L, Hjelms E, Lindeburgh T. Brain hyperperfusion during cardiac operations. Cerebral blood ow measured in
man by intra-arterial injection of xenon 133: evidence suggestive of intraoperative microembolism. J Thorac Cardiovasc
Surg 1983;86:202e8.
[54] Govier AV, Reves JG, McKay RD, et al. Factors and their inuence on regional cerebral blood ow during nonpulsatile
cardiopulmonary bypass. Ann Thorac Surg 1984;38:592e600.
[55] Murkin JM, Farrar JK, Tweed WA, et al. Cerebral autoregulation and ow/metabolism coupling during cardiopulmonary
bypass: the inuence of PaCO2. Anesth Analg 1987;66(9):825e32.
[56] Nussmeier NA, Arlund C, Slogoff S. Neuropsychiatric complications after cardiopulmonary bypass: cerebral protection by
a barbiturate. Anesthesiology 1986;64:165e70.
[57] Zaidan JR, Klochany A, Martin WM, et al. Effect of thiopental on neurologic outcome following coronary artery bypass
grafting. Anesthesiology 1991;74:406e11.
[58] Wareing TH, Davila-Roman VG, Barzilai B, et al. Management of the severely atherosclerotic ascending aorta during
cardiac operations. A strategy for detection and treatment. J Thorac Cardiovasc Surg 1992;103:453e62.
[59] Ueda Y, Miki S, Kusuhara K, et al. Surgical treatment of aneurysm or dissection involving the ascending aorta and aortic
arch, utilizing circulatory arrest and retrograde cerebral perfusion. J Cardiovasc Surg Torino 1990;31:553e8.
[60] Mills NL, Ochsner JL. Massive air embolism during cardiopulmonary bypass. Causes, prevention, and management.
J Thorac Cardiovasc Surg 1980;80:708e17.
[61] Melrose DG, Dreyer B, Bentall HH, et al. Elective cardiac arrest. Lancet 1955;269(6879):21e2.
[62] Gay Jr WA, Ebert PA. Functional, metabolic, and morphologic effects of potassium-induced cardioplegia. Surgery 1973;
74(2):284e90.
[63] Chenoweth DE, Cooper SW, Hugli TE, et al. Complement activation during cardiopulmonary bypass: evidence for gen-
eration of C3a and C5a anaphylatoxins. N Engl J Med 1981;304:497e503.
[64] Gruntzig A. Transluminal dilatation coronary-artery stenosis. Lancet 1978;1(8058):263.
[65] Benetti FJ, Naselli G, Wood M, et al. Direct myocardial revascularization without extracorporeal circulation: experience in
700 patients. Chest 1991;100:312e6.
[66] Buffolo E, Andrade JC, Succi J, et al. Direct myocardial revascularization without cardiopulmonary bypass. Thorac Car-
diovasc Surg 1985 Feb;33(1):26e9.
[67] Stevens JH, Burdon TA, Peters WS, et al. Port-access coronary artery bypass grafting: a proposed surgical method. J Thorac
Cardiovasc Surg 1996;111:567e73.
[68] Fromes Y, Gaillard D, Ponzio O, et al. Reduction of the inammatory response following coronary bypass grafting with
total minimal extracorporeal circulation. Eur J Cardiothorac Surg 2002;22(4):527e33.
[69] Stoney WS, Alford Jr WC, Burrus GR, et al. Air embolism and other accidents using pump oxygenators. Ann Thorac Surg
1980;29:336e40.
[70] Jenkins OF, Morris R, Simpson JM. Australasian perfusion incident survey. Perfusion 1997;12(5):279e88.
[71] Mejak BL, Stammers A, Rauch E, et al. A retrospective study on perfusion incidents and safety devices. Perfusion 2000;
15(1):51e61.
[72] Charriere JM, Pelissie
J, Verd C, et al. Survey: retrospective survey of monitoring/safety devices and incidents of car-
diopulmonary bypass for cardiac surgery in France. J Extra Corpor Technol 2007;39(3):142e57.
[73] Martinez EA, Marsteller JA, Thompson DA, et al. The society of cardiovascular anesthesiologists' FOCUS initiative: locating
errors through networked surveillance (LENS) project vision. Anesth Analg 2010;110:307e11.
*[74] Wahr JA, Prager RL, Abernathy 3rd JH, et al. Patient safety in the cardiac operating room: human factors and teamwork: a
scientic statement from the American heart association. Circulation 2013;128:1139e69.
*[75] Miller Jr DW, Binford JM, Hessel 2nd EA. Results of a survey of the professional activities of 811 cardiopulmonary per-
fusionists. J Thorac Cardiovasc Surg 1982;83(3):385e9.
[76] Silvay G, Ammar T, Reich DL, et al. Cardiopulmonary bypass for adult patients: a survey of equipment and techniques.
J Cardiothorac Vasc Anesth 1995;9(4):420e4.
[77] Tuble SC, Willcox TW, Baker RA. Australian and New Zealand perfusion survey: management and procedure. J Extra
Corpor Technol 2009;41(2):64e72.
[78] Warren OJ, Wallace S, de Wit KL, et al. Variations in the application of various perfusion technologies in Great Britain and
Irelandea national survey. Artif Organs 2010;34(3):200e5.
[79] Bartels C, Gerdes A, Babin-Ebell J, et al. Cardiopulmonary bypass: evidence or experience based? J Thorac Cardiovasc Surg
2002;124:20e7.
[80] Ballaux PK, Gourlay T, Ratnatunga CP, et al. A literature review of cardiopulmonary bypass models for rats. Perfusion
1999;14:411e7.
[81] Grocott HP, Mackensen GB, Newman MF, et al. Neurological injury during cardiopulmonary bypass in the rat. Perfusion
2001;16:75e81.
[82] Shann KG, Likosky DS, Murkin JM, et al. An evidence based review of the practice of cardiopulmonary bypass in adults: a
focus on neurologic injury, glycemic control, hemodilution, and the inammatory response. J Thorac Cardiovasc Surg
2006;132:283e90.
[83] Hogue Jr CW, Palin CA, Arrowsmith JE. Cardiopulmonary bypass management and neurologic outcomes: an evidence
based appraisal of current practices. Anesth Analg 2006;103:21e37.
[84] Riley JB, O'Kane KC. A computer simulation of maintaining total heart lung bypass for basic education. J Extra Corpor
Technol Proc 1977:42e9.
[85] Morris RW, Pybus DA. Orpheus cardiopulmonary bypass simulation system. J Extra Corpor Technol 2007;39(4):228e33.

Anaesthesiology
Inammatory response and extracorporeal

circulation
Florian Kraft, M.D., Resident Anaesthetist,
Christoph Schmidt, M.D., Ph.D., Consultant Anaesthetist,
Hugo Van Aken, M.D., Ph.D., Consultant Anaesthetist, Head of
Department,
Alexander Zarbock, M.D., Ph.D., Consultant Anaesthetist *
Department of Anaesthesiology, Intensive Care and Pain Medicine, University Hospital Muenster, Mnster,
Germany
Keywords:
Patients undergoing cardiac surgery with extracorporeal circula-
tion (EC) frequently develop a systemic inammatory response
inammation
systemic inammatory response syndrome syndrome. Surgical trauma, ischaemiaereperfusion injury, endo-
neutrophil recruitment toxaemia and blood contact to nonendothelial circuit compounds
promote the activation of coagulation pathways, complement
factors and a cellular immune response. This review discusses the
multiple pathways leading to endothelial cell activation, neutro-
phil recruitment and production of reactive oxygen species and
nitric oxide. All these factors may induce cellular damage and
subsequent organ injury. Multiple organ dysfunction after cardiac
surgery with EC is associated with an increased morbidity and
mortality. In addition to the pathogenesis of organ dysfunction
after EC, this review deals with different therapeutic interventions
aiming to alleviate the inammatory response and consequently
multiple organ dysfunction after cardiac surgery.
* Corresponding author. University of Muenster, Department of Anaesthesiology Intensive Care and Pain Medicine, Albert-
Schweitzer-Campus 1, Geba ude A1, 48149 Mnster, Germany. Tel.: 49 (251) 8347252; Fax: 49 (251) 88704.
E-mail addresses: kraftf@uni-muenster.de (F. Kraft), schmch@uni-muenster.de (C. Schmidt), hva@uni-muenster.de (H. Van
Aken), zarbock@uni-muenster.de (A. Zarbock).
114 F. Kraft et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 113e123
Introduction
During cardiac surgery, extracorporeal circulation (EC) enables surgeons to treat a wide range of
heart diseases while circulation and oxygenation of human blood are maintained by machines. While
the rst successful cardiopulmonary bypass (CBP) during a ventricular septum repair in 1953 by
Gibbon was still an experimental procedure, CPB has in the meantime emerged as a well-established
procedure, used worldwide in more than half a million people per year [1,2]. Blood trauma and the
creation of embolic particles were initially common adverse effects of EC. However, through the
development of new biomaterial technologies and implementation of pharmacologic drugs, these
potential side effects were reduced, linked to a rapid improvement of patients' outcome. Nevertheless,
despite the encouraging successes, patients undergoing cardiac surgery continue to frequently develop
post-operative complications arising from the inappropriate activation of inammatory pathways [3].
Inammation is the response of the organism to several noxious stimuli. During cardiac surgery,
different stimuli, such as blood exposition to the nonendothelial surface of the CPB, ischae-
miaereperfusion injury and endotoxaemia, may trigger an inammatory response. The activation of
humoral and cellular cascades leads to an increase of pro-inammatory cytokines in the circulating
blood and to enhanced leucocyte recruitment [4]. The resulting systemic inammatory response
syndrome (SIRS) is associated with post-operative complications including myocardial dysfunction,
respiratory failure, acute kidney injury (AKI), neurologic dysfunction, bleeding disorders and, nally,
multiple organ failure (MOF). MOF is strongly associated with increased morbidity and mortality rates
among patients undergoing cardiac surgery.
Mediators of the inammatory response
Coagulation linked to inammation
In the vascular system, endothelial cells produce pro- and anticoagulation factors, which together
maintain the equilibrium of blood uidity. The exposure of blood components to articial surfaces
within the extracorporeal circuit leads to an activation of coagulation cascades. These pathways,
classically divided into an intrinsic and extrinsic pathway, consist of a series of enzyme cascades and
lead to the generation of activated thrombin (Fig. 1) [5].
Fig. 1. Schematic overview of the intrinsic and extrinsic coagulation pathway. Coagulation factors that contribute to inammatory
cell activation are coloured blue. TF, tissue factor; Pre-Kal, prekallikrein; Kal, kallikrein; HMWK, high molecular weight kininogen;
PL, phospholipids; and coagulation factors, V,VII, VIII, IX, X, XI and XII (a indicating active form).
F. Kraft et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 113e123 115
The intrinsic pathway is initiated by the exposure of negatively charged, nonendothelial surfaces to
factor XII (Hageman factor), whose autoactivation requires the presence of prekallikrein and high-
molecular-weight kininogen (HMWK). Factor XIIa contributes to an enzyme cascade that leads to
the generation of thrombin. Thrombin subsequently cleaves brinogen to brin, which is able to form
brin strands in a polymerised form [6]. Active factor XII enables the activation of prekallikrein to
kallikrein, a serine protease-releasing bradykinin through the cleavage of HMWK. Bradykinin is a
potent vasoactive peptide that alters vascular endothelial permeability and smooth muscle tone level.
In addition, binding to the bradykinin receptor on leucocytes may alter their activation state and
cytokine production of leucocytes [7].
Tissue trauma initiates the extrinsic coagulation pathway. Injured vessel walls enable the exposure
of blood to tissue factor (TF), an integral membrane glycoprotein (GP) of perivascular cells [8]. Binding
of the circulating factor VII to TF enables its activation to factor VIIa. Depending on Ca2 and phos-
pholipids, the factor VIIaeTF complex promotes the activation of factor X to factor Xa. Production of
Factor Xa is the point of convergence of both coagulation pathways leading to thrombin generation and
cleavage of brinogen.
In addition to their pro-coagulation properties, factor Xa and thrombin have pleiotropic effects on
inammatory processes, tissue remodelling and development of atherosclerosis [9,10]. Both factors
bind and activate protease-activated receptors (PARs). These G-protein-coupled receptors (GPCRs)
activate intracellular signalling cascades leading to platelet and leucocyte activation. In vitro studies
revealed that factor Xa induces cytokine expression and surface expression of adhesion molecules on
leucocytes, thereby linking coagulation to inammation [11,12].
During cardiac surgery with the use of EC, both the intrinsic and extrinsic pathways of the
coagulation cascade are activated. To prevent intravascular clot formation, anticoagulation therapy is
indispensable [13]. Systemic heparinisation has evolved as the current standard therapy, largely
because of the easy monitoring of its anticoagulant effect. Despite full heparinisation, thrombin
generation still occurs, as indicated by elevated levels of thrombineantithrombin III complexes in
patients undergoing CPB [13]. This may cause a consumptive coagulopathy, which is responsible for
thromboembolic events and non-surgical haemorrhage after CPB [14]. Thus, increased levels of
biomarkers of activated coagulation pathways are associated with negative outcomes after cardiac
surgery [15].
Cytokines and chemokines
Cytokines and chemokines are released by different cell types in response to different stimuli [16].
These mediators activate immune cells and modulate the inammatory response of the body. CPB
induces an altered release of pro- and anti-inammatory cytokines [17]. Increased levels of the pro-
inammatory cytokines tumour necrosis factor-a (TNF-a), interleukin (IL)-6 and IL-8 were found to
be associated with adverse clinical outcomes in patients undergoing cardiac surgery [18]. Furthermore,
human myocardial contractile function was profoundly depressed after exposure to TNF-a and IL-1b
(myocardial depressant factor) [19].
Nitric oxide
In response to physiologic stimuli such as pulsatile ow and shear stress, endothelial cells regulate
the vascular tone through the release of nitric oxide (NO). The constitutive nitric oxide synthase (cNOS)
in endothelial cells produces NO from the amino acid L-arginine in response to calcium signalling.
Constitutively released NO decreases smooth muscle contraction, thereby promoting vasodilatation
and reducing shear stress [20].
In response to the release of pro-inammatory cytokines or endotoxins, endothelial cells and
smooth muscle cells express an inducible nitric oxide synthase (iNOS), which is independent of calcium
signalling [21]. Besides the benecial role of NO under physiologic conditions, iNOS-derived NO im-
pairs the mitochondrial respiratory chain. NO-mediated inhibition of the cytochrome oxidase increases
the production of reactive oxygen species (ROS) [22]. ROS alter intracellular pH levels, reduce aden-
osine triphosphate (ATP) production and induce cellular damage through lipid peroxidation, oxidation
of DNA and inactivation of specic enzymes [23]. In the presence of NO, ROS may also form highly
noxious peroxynitrite [24]. These compounds can lead to the necrosis of endothelial cells, thereby
impairing the endothelial barrier function and enabling vascular leakage. Extracellular ROS serve as
potent pro-inammatory stimuli, as they may trigger an up-regulation of cell-surface adhesion mol-
ecules and release of cytokines by activated macrophages [25].
Platelets and inammation
In addition to their coagulation function, platelets contribute to the inammatory response

following EC. Heparinisation, hypothermia and surgical trauma activate platelets. Activated platelets
change their shape and release a variety of mediators, including inammatory cytokines. Platelet
granules contain chemokines (chemokine (CeC motif) ligand 3, 5, 7, 17 and chemokine (C-X-C motif)
ligand 4, 5, 7, and 8), cytokines (IL-1b, CD 40 ligand and b-thrombomodulin), adhesion factors (P-
selectin, GP IIb/IIIa and von Willebrand factor), growth factors (platelet-derived growth factor, trans-
forming growth factor-b, epidermal growth factor and vascular endothelial growth factor) and coag-
ulation factors (factor V, factor XI, plasminogen activator inhibitor-1, plasminogen and protein S)
(reviewed in ref. [26]). These mediators act on other platelets, leucocytes and endothelial cells.
Platelets are activated by the CPB system, resulting in the up-regulation of P-selectin on their
surface membranes [27]. Platelet P-selectin binds to P-selectin glycoprotein ligand (PSGL)-1 expressed
on leucocytes, thus mediating the rst cellecell interaction [28e30]. P-selectin binding to PSGL-1
promotes monocyte activation leading to secretion of cytokines such as IL-1b, IL-8 and monocyte
chemoattractant protein (MCP)-1 [31,32]. In the course of the exocytosis of cytokines, activated
monocytes present TF on their surface, which initiates the extrinsic coagulation pathway and favours
thrombus formation [33].
The initial contact between platelets and leucocytes is further strengthened by integrins [34].
Integrins are a family of transmembrane ab heterodimeric receptors, which are expressed on almost all
cell surfaces [35]. They bind to extracellular matrix as well as immunoglobulin-like adhesion mole-
cules, but a conformational change of the molecules is required before binding to a ligand. The leu-
cocyte integrin aM2 (CD11b/CD18) recognises platelet GP Ib and brinogen, which is bound to the
platelet GP IIb/IIIa receptor. Platelet intercellular adhesion molecule (ICAM)-2 interacts with the leu-
cocyte integrin aLb2 (CD11a/CD18).
The vascular endothelium presents the adhesion molecule CD40, which is bound by the comple-
mentary CD40 ligand (CD40L) on platelets. The transmembrane CD40L is structurally related to TNF-a
and promotes the endothelial secretion of pro-inammatory cytokines (IL-8 and MCP-1). Thus,
platelets bound to the endothelium contribute to neutrophil recruitment and activation of monocytes
by direct plateleteleucocyte interaction, by release of pro-inammatory granules or by stimulation of
endothelial cells [7,36]. In patients undergoing CPB during cardiac surgery, the level of plate-
leteleucocyte aggregates was shown to increase [37].
EC activates leucocytes
Following cardiac surgery with CPB, neutrophil recruitment is involved in the pathogenesis of SIRS
[38]. Polymorphonuclear neutrophils store intracellular granules containing neutrophil elastase,
myeloperoxidase and several lysozymes [39]. After being activated, neutrophils release their granules
and undergo a respiratory burst discharging large quantities of superoxide and hydrogen peroxide.
Activated neutrophils also release nuclear extracellular traps (NETs), which are potent antibacterial
compounds, composed of chromatin fragments, histones and granular proteins [40]. Neutrophil
compounds may cause tissue damage and may increase microvascular permeability [4].
During inammation, cytokines such as TNF-a and IL-1b stimulate endothelial cells, which up-
regulate the expression of adhesion molecules [41]. Activated endothelial cells express P-selectin on
their cell surface followed by an increased expression of E-selectin, which occurs 2e4 h later [42,43].
These adhesion molecules mediate the initial interaction between leucocytes and activated endothelial
cells and they initiate the leucocyte recruitment cascade consisting of capturing, rolling, adhesion,
crawling and transmigration.
E- and P-selectins expressed on activated endothelial cells bind PSGL-1 on neutrophils. This cell-to-
cell contact induces an intracellular signalling in neutrophils, resulting in integrin activation and in a
deceleration of the rolling velocity of leucocytes [34,44]. During rolling, PSGL-1 engagement and the
activation of GPCRs by chemokines induce full integrin activation and leucocyte arrest. Neutrophils
express the b2 integrins aLb2 (function-associated antigen (LFA)-1), aMb2 (macrophage-1 antigen (Mac-
1)), axb2 and low levels of a4b1 [45]. LFA-1 is responsible for selectin-mediated slow leucocyte rolling
and chemokine-induced arrest, whereas Mac-1 is required for post-adhesion strengthening [39]. After
leucocyte adhesion, leucocytes start crawling along the endothelium to nd permissive sites for
diapedesis.
A recently published study demonstrated that cardiac surgery with CPB abrogates selectin-induced
slow leucocyte rolling on E-selectin/ICAM-1 and P-selectin/ICAM-1 [46]. By contrast, chemokine-
induced arrest and transmigration was signicantly increased. The abolishment of slow leucocyte
rolling was mechanistically linked to disturbances in intracellular signalling with reduced phosphor-
ylation of phospholipase C (PLC) g2, Akt and p38 mitogen-activated protein kinase. Furthermore, CPB
induced an elevated transmigration, which was caused by up-regulation of Mac-1 on neutrophils [46].
Ischaemiaereperfusion injury
Reperfusion injury is the tissue damage caused when blood supply returns to the tissue after a
period of ischaemia or lack of oxygen [25]. During cardiac surgery with the use of CPB, blood supply to
the heart and lungs is temporarily almost completely stopped. The lack of oxygen within these organs
promotes cells to accumulate metabolic intermediates such as adenosine monophosphate and hypo-
xanthine, resulting in ROS production. ROS-mediated cellular damage induces the up-regulation of
cell-surface adhesion molecules and expression of pro-inammatory cytokines [23]. After reperfusion,
the tissue injury even increases due to leucocyte recruitment into the inamed tissue.
Following reperfusion, the recruited neutrophils increase tissue levels of ROS, cytokines and che-
moattractants, thereby amplifying the inammatory response [25,47]. Neutrophils are the major
source of ROS production, which may damage cells even in non-hypoxic tissue [48,49]. Recruitment of
neutrophils increases vascular permeability [39]. Cellular swelling and damage due to hypoxia or
cytotoxic compounds may shift the equilibrium between pro- and anticoagulation factors produced by
the endothelium towards thrombus formation. Platelets as well as neutrophils may stick within
inamed capillaries, thereby provoking hypoperfusion of hypoxic tissues, the so-called no-reow
phenomenon [50]. Finally, the devastating neutrophil recruitment during ischaemiaereperfusion
contributes to the generation of acute respiratory distress syndrome (ARDS), SIRS and MOF causing
high morbidity and mortality [51].
Endotoxaemia
Gram-negative intestinal ora produce lipopolysaccharides and endotoxins as a part of their cell
wall. In patients undergoing cardiac surgery, hypoperfusion and hypoxia may impair the intestinal
barrier integrity, which is necessary to avoid the transition of endotoxins [52]. Intravascular endotoxins
bind to various receptors, such as Toll-like receptors on leucocytes, causing cytokine production and
leucocyte activation [53]. Clinical studies investigating endotoxin levels during cardiac surgery
demonstrated a reduced inammatory response in off-pump patients, but these studies failed to show
improvements in clinical outcomes (reviewed in ref. [52]).
Inammatory response and organ dysfunction
Each of the mediators mentioned above may inuence patients' inammatory response during and
after CPB. Inappropriate inammatory response can lead to ARDS, AKI and nally MOF.
The pathogenesis of ARDS depends on a multitude of different factors, among those a number of
vasoactive mediators released by activated leucocytes and platelets, which contribute to oedema
formation and increased pulmonary vascular resistance [23]. In a rat model, neutropaenia mitigated
ischaemiaereperfusion injury of the lung, showing that neutrophils directly contribute to tissue
damage [54]. In a mouse model, blocking of P-selectin inhibited the formation of plateleteneutrophil
aggregates, thereby attenuating the severity of ARDS in a model of acid-induced lung injury [55].
Further detailed insights into the pathogenesis and therapeutic options of organ dysfunctions
following CBP will be provided by separate reviews by Huffmyer et al. and Landoni et al. within this issue.
Therapeutic options
During CPB, a multitude of stimuli trigger the emergence of an SIRS. Enormous efforts have been
made to promote the development of drugs or technical approaches able to ameliorate inammatory
responses (Table 1). Clinical studies to evaluate the effect of single interventions are fundamentally
awed, as multiple factors contribute to the clinical picture of SIRS and to the varying degrees of in-
ammatory response [16].
Technical approaches to ameliorate the inammatory response
Several components of the CPB circuit, including type of oxygenator, pulsatile versus non-pulsatile
ow, selective leucocyte lters and different priming solutions (reviewed in ref. [56]), were tested with
respect to their inammatory properties. Until now, only few strategies were able to signicantly
ameliorate the inammatory response and to improve patients' outcome.
As soon as blood comes into contact with the CBP circuit, a multitude of stimuli contribute to the
development of SIRS. Several attempts have been made to improve the biocompatibility of circuit
compounds by coating the circuit surfaces with heparin, poly-2-methoxyethylenacrylate, hyaluronan
or phosphophorycholine [16]. A recent meta-analysis by Mahmood and colleagues provides evidence
that heparin coating does not increase the number of adverse events, but it decreases blood transfusion
requirements, re-operation rates, time of mechanical ventilation and length of stay (LOS) in the
intensive care unit (ICU) and in hospital [56]. The authors justify the usage of coated CPB circuits with
improved clinical outcome. Before heparin-coated circuits are implemented in daily practice, further
investigations are warranted.
Mini-extracorporeal circuits (MECCs) provide a small-volume circuit, which consists of a heparin-
coated system and a membrane oxygenator but lacks the venous reservoir of a standard CBP circuit
Table 1
List of clinical trials that revealed benecial effects on outcome of patients after cardiac surgery due to technical or pharma-
cologic intervention.
Approach Targeted mechanism Clinical effect Reference
Technical Coating circuit Reduce activation of blood Blood transfusions Y [56]

approaches compounds coagulation compounds due mechanical ventilation
to contact to nonendothelial time Y
surfaces LOS on ICU Y
Mini-extracorporeal Reduce haemodilution and Blood transfusions Y [59]
circuits associated impairment of oxygen levels of IL-6 and TNF-a Y
delivery capacity
avoid activation of coagulation
pathways and cellular activation
Pharmacologic High-dose versus Antioxidant and anti-inammatory Markers of neuronal injury Y [68,69]
approaches low-dose propofol effects
Volatile anaesthetics Ischaemic preconditioning Rate of myocardial [70,71]
versus TIVA infarction Y
mortality Y
Statin administration Pleiotropic effects Rate of new onset of [74]
binding to leucocyte LFA-1 receptor atrial brillation Y
LOS in hospital Y
myocardial infarction Y
mortality Y
Glucocorticoid Anti-inammatory effects Rate of new onset of atrial [80,81]
administration brillation Y
incidence of pneumonia Y
[57]. Standard CPB circuits need large priming volumes, which lead to signicant haemodilution
associated with reduced blood oxygen transport capacity and impairment of coagulation [58].
Compared to conventional CPB, MECCs lead to signicant reduction of haemodilution and red blood
transfusion requirements in a study of 199 patients undergoing coronary artery bypass grafting [59].
Moreover, levels of the inammatory cytokines IL-6 and TNF-a were signicantly decreased in patients
undergoing CABG with MECCs [60]. A similar effect of MECC on inammatory cytokines was observed
in patients undergoing aortic valve replacement [61]. These results are promising, but pulmonary
complications, neurologic events or mortality were not signicantly reduced due to usage of MECC
[62]. Despite improvements regarding haemodilution and blood transfusion requirements, MECCs still
contain a considerable potential to trigger an SIRS [63].
Pharmacologic approaches
Anaesthetic agents
In addition to their anaesthetic effect, the majority of anaesthetic agents used during general or
local anaesthesia possess immunomodulatory effects [64]. High-dose opiates are used during cardiac
anaesthesia to preserve haemodynamic stability and to attenuate the stress response to surgical stimuli
due to their potent analgesic effects [65]. In a small randomised pilot study, remifentanil adminis-
tration was associated with a shortened length of stay and a reduced level of cytokine signalling [66].
Although remifentanil has a favourable pharmacokinetic property, this substance may have immu-
nomodulatory functions in patients undergoing elective CBP surgery compared to fentanil [66].
Propofol is commonly administered for induction and maintenance of general anaesthesia, which
negatively affects haemodynamic parameters. High-dose administration of propofol (200 mg/kg/min)
decreased the mean arterial pressure and cardiac index followed by an increased heart rate [67]. This
resulted in a decreased myocardial blood ow and myocardial oxygen consumption [67]. In addition,
high-dose propofol appears to exert a brain-protective effect due to its antioxidant and anti-
inammatory properties [68,69].
Volatile anaesthetics were shown to decrease levels of the pro-inammatory cytokines IL-6, IL-8
and TNF-a [64,70]. According to a comparing body of evidence, anaesthetic preconditioning by volatile
anaesthetics attenuates the deleterious consequences of ischaemiaereperfusion and protects the heart
through a mechanism similar to ischaemic preconditioning [70]. Protective effects on the myocardium
were also investigated by randomised controlled trials comparing volatile anaesthetics and total
intravenous anaesthesia in patients undergoing cardiac surgery. A meta-analysis by Landoni and col-
leagues revealed that desurane and sevourane reduced the incidence of myocardial infarction and
all-cause mortality [71]. This indicates that the use of volatile anaesthetics for patients undergoing
cardiac surgery is associated with a reduction of adverse major events [70].
The application of thoracic epidural anaesthesia (TEA) may offer several benets including sym-
pathetic block with stable heart rate and decreased myocardial oxygen consumption [65]. One of the
main concerns against the application of TEA in this patient population undergoing cardiac surgery is
the risk of epidural haematoma, due to heparinisation during CBP. A risk assessment of TEA application
for cardiac surgery in 2012 revealed that the risk of epidural haematoma is very low, similar to the
general surgery population [72]. Providing TEA in cardiac patients is not unreasonable but it requires
protocols to safely manage these patients' coagulation proles and post-operative anti-platelet ther-
apies [65].
Pleiotropic effects of statins

Inhibitors of 3-hydroxy-3-methyl-glutaryl coenzyme A reductase (statins) are well established in
the treatment of hypercholesterolaemia and have benecial effects on morbidity and mortality due to
the prevention of major cardiovascular events [73]. Beyond their lipid-lowering effect, statins were
found to have anti-inammatory properties, the so-called pleiotropic effects [74]. These may
particularly be explained by the binding of statins to the leucocyte integrin LFA-1, which is involved in
the neutrophil recruitment cascade [75]. Statins counteract pro-inammatory transcriptional path-
ways and suppress the activity of pro-oxidant enzymes within endothelial cells [74].
Several randomised controlled trials investigated the effect of perioperative statin treatment in
patients undergoing cardiac surgery. A recent review and meta-analysis provided by de Waal and
colleagues demonstrated that preoperative initiation of statin administration reduces mortality,
myocardial infarction, perioperative new-onset atrial brillation and length of hospital stay [76].
Administration of glucocorticoids
The administration of glucocorticoids aims at attenuating the inammatory response following
CBP. The rst positive effects of glucocorticoid administration were obtained by Moses et al. as early
as 1966. The authors investigated dogs undergoing prolonged CPB [77]. Over decades, several
clinical trials scrutinised the anti-inammatory effect of glucocorticoids on major and minor
outcome parameters in human patients. After several promising results in single-centre trials, a
thorough meta-analysis revealed that the positive effects of glucocorticoid application were minor
than initially thought [78]. There was no difference in major outcome parameters such as major
cardiac events, pulmonary complications or mortality after 30 days. Secondary outcomes such as
length of stay in ICU and hospital, as well as new-onset atrial brillation, showed a slight trend
towards steroid administration [79]. Although glucocorticoid administration is believed to raise the
incidence of infectious complication, Dieleman and colleagues observed less pneumonia in the
dexamethasone group [80]. The Steroids in Cardiac Surgery (SIRS) trial is a large international
randomised controlled study that investigates the effects on AKI and other outcome parameters
[81]. Nevertheless, until now, there has been no evidence for an effect of glucocorticoid adminis-
tration on mortality or severe cardiac complications [78].
Conclusion
EC can induce an SIRS response in patients undergoing cardiac surgery. This syndrome is
frequently associated with organ injury and MOF. The understanding of the inammatory processes
and the interplay of humoral factors and cellular immune response increased rapidly during the last
decade. Multiple anti-inammatory strategies were applied in the past, which signicantly reduced
cytokine levels without improving clinical outcome. This implies that the amplitude of inammatory
cytokines does not directly predict patients' outcome. The lack of evidence may also be due to poor
study design or large variations in covariants, such as different transfusion regiments, CBP circuits
and temperature management. Future studies should aim at maximising the standardisation of peri-
and post-operative patient management and addressing multiple targets as anti-inammatory
strategy.
Practice points
The use of CPB in patients undergoing cardiac surgery is frequently associated with the develop-
ment of an SIRS. A variety of pro-inammatory stimuli contributes to the activation of endothelial
cells, neutrophil activation and recruitment and production of cytotoxic compounds.
While there is no doubt that the administration of glucocorticoids and statins is effective in
reducing the cytokine levels, improvements in major clinical outcome parameters have not been
evidenced.
Volatile and intravenous anaesthetic agents possess multiple immunomodulatory effects. The
choice of anaesthetic agent was shown to have the potential to positively inuence the outcome of
patients undergoing cardiac surgery with CPB.
Research agenda
The knowledge of inammatory processes and the interplay between humoral factors and cellular
immune response has been rapidly increasing over the last few decades. Future studies should aim
at targeting multiple inammatory mediators simultaneously to attenuate the evolution of SIRS.
Early diagnosis of organ dysfunction such as AKI is an indispensable precondition for the early
initiation of adequate treatment. Large, randomised controlled trials are needed to provide evidence
for pharmaceutical and technical approaches that aim to reduce post-operative complications
following cardiac surgery.
Conict of interest
The authors have no conicts of interest to declare.
Acknowledgements
The work was supported by the German Research Foundation (ZA428/6-1 to AZ).
References
[1] Gibbon JH. Application of a mechanical heart and lung apparatus to cardiac surgery. Minn Med 1954;37(3):171e85.
passim.
*[2] Warren OJ, Smith AJ, Alexiou C, et al. The inammatory response to cardiopulmonary bypass: part 1mechanisms of
pathogenesis. J Cardiothorac Vasc Anesth 2009;23(2):223e31.
[3] Landis RC. Redening the systemic inammatory response. Semin Cardiothorac Vasc Anesth 2009;13(2):87e94.
[4] Paparella D, Yau TM, Young E. Cardiopulmonary bypass induced inammation: pathophysiology and treatment. An
update. Eur J Cardiothorac Surg 2002;21(2):232e44.
[5] Davie EW, Fujikawa K, Kisiel W. The coagulation cascade: initiation, maintenance, and regulation. Biochemistry 1991;
30(43):10363e70.
[6] Day JR, Taylor KM. The systemic inammatory response syndrome and cardiopulmonary bypass. Int J Surg 2005;3(2):
129e40.
[7] Bockmann S, Paegelow I. Kinins and kinin receptors: importance for the activation of leukocytes. J Leukoc Biol 2000;
68(5):587e92.
[8] Mackman N, Tilley RE, Key NS. Role of the extrinsic pathway of blood coagulation in hemostasis and thrombosis. Arte-
rioscler Thromb Vasc Biol 2007;27(8):1687e93.
[9] Spronk HM, de Jong AM, Crijns HJ, et al. Pleiotropic effects of factor Xa and thrombin: what to expect from novel an-
ticoagulants. Cardiovasc Res 2014;101(3):344e51.
[10] Egorina EM, Sovershaev MA, Hansen JB. The role of tissue factor in systemic inammatory response syndrome. Blood
Coagul Fibrinolysis 2011;22(6):451e6.
[11] Edmunds Jr LH, Colman RW. Thrombin during cardiopulmonary bypass. Ann Thorac Surg 2006;82(6):2315e22.
[12] Borensztajn K, Peppelenbosch MP, Spek CA. Factor Xa: at the crossroads between coagulation and signaling in physiology
and disease. Trends Mol Med 2008;14(10):429e40.
[13] Hunt BJ, Parratt RN, Segal HC, et al. Activation of coagulation and brinolysis during cardiothoracic operations. Ann
Thorac Surg 1998;65(3):712e8.
[14] Philippou H, Davidson SJ, Mole MT, et al. Two-chain factor VIIa generated in the pericardium during surgery with car-
diopulmonary bypass : relationship to increased thrombin generation and heparin concentration. Arterioscler Thromb
Vasc Biol 1999;19(2):248e54.
[15] Dixon B, Santamaria J, Campbell D. Coagulation activation and organ dysfunction following cardiac surgery. Chest 2005;
128(1):229e36.
*[16] Hall R. Identication of inammatory mediators and their modulation by strategies for the management of the systemic
inammatory response during cardiac surgery. J Cardiothorac Vasc Anesth 2013;27(5):983e1033.
[17] Wan S, LeClerc JL, Vincent JL. Inammatory response to cardiopulmonary bypass: mechanisms involved and possible
therapeutic strategies. Chest 1997;112(3):676e92.
[18] Teoh KH, Bradley CA, Gauldie J, et al. Steroid inhibition of cytokine-mediated vasodilation after warm heart surgery.
Circulation 1995;92(9 Suppl.):Ii347e353.
[19] Cain BS, Meldrum DR, Dinarello CA, et al. Tumor necrosis factor-alpha and interleukin-1beta synergistically depress
human myocardial function. Crit Care Med 1999;27(7):1309e18.
[20] Kubes P. Polymorphonuclear leukocyteeendothelium interactions: a role for pro-inammatory and anti-inammatory
molecules. Can J Physiol Pharmacol 1993;71(1):88e97.
[21] Moncada S, Higgs A. The L-arginine-nitric oxide pathway. N Engl J Med 1993;329(27):2002e12.
[22] Kadenbach B. Intrinsic and extrinsic uncoupling of oxidative phosphorylation. Biochim Biophys Acta 2003;1604(2):
77e94.
*[23] den Hengst WA, Gielis JF, Lin JY, et al. Lung ischemia-reperfusion injury: a molecular and clinical view on a complex
pathophysiological process. Am J Physiol Heart Circ Physiol 2010;299(5):H1283e99.
[24] Ovechkin AV, Lominadze D, Sedoris KC, et al. Lung ischemia-reperfusion injury: implications of oxidative stress and
platelet-arteriolar wall interactions. Arch Physiol Biochem 2007;113(1):1e12.
[25] Schoeld ZV, Woodruff TM, Halai R, et al. Neutrophilsea key component of ischemia-reperfusion injury. Shock 2013;
40(6):463e70.
[26] Zarbock A, Polanowska-Grabowska RK, Ley K. Platelet-neutrophil-interactions: linking hemostasis and inammation.
Blood Rev 2007;21(2):99e111.
[27] Kondo C, Tanaka K, Takagi K, et al. Platelet dysfunction during cardiopulmonary bypass surgery. With special reference to
platelet membrane glycoproteins. ASAIO J 1993;39(3):M550e3.
[28] Yang J, Furie BC, Furie B. The biology of P-selectin glycoprotein ligand-1: its role as a selectin counterreceptor in
leukocyte-endothelial and leukocyte-platelet interaction. Thrombosis Haemostasis 1999;81(1):1e7.
[29] Evangelista V, Manarini S, Sideri R, et al. Platelet/polymorphonuclear leukocyte interaction: P-selectin triggers protein-
tyrosine phosphorylation-dependent CD11b/CD18 adhesion: role of PSGL-1 as a signaling molecule. Blood 1999;93(3):
876e85.
[30] Ley K, Kansas GS. Selectins in T-cell recruitment to non-lymphoid tissues and sites of inammation. Nature reviews.
Immunology 2004;4(5):325e35.
[31] Neumann FJ, Marx N, Gawaz M, et al. Induction of cytokine expression in leukocytes by binding of thrombin-stimulated
platelets. Circulation 1997;95(10):2387e94.
[32] Weyrich AS, Elstad MR, McEver RP, et al. Activated platelets signal chemokine synthesis by human monocytes. J Clin
investigation 1996;97(6):1525e34.
[33] Maugeri N, Brambilla M, Camera M, et al. Human polymorphonuclear leukocytes produce and express functional tissue
factor upon stimulation. J Thromb Haemost 2006;4(6):1323e30.
[34] McEver RP, Cummings RD. Role of PSGL-1 binding to selectins in leukocyte recruitment. J Clin investigation 1997;100(11
Suppl.):S97e103.
[35] Herter J, Zarbock A. Integrin regulation during leukocyte recruitment. J Immunol 2013;190(9):4451e7.
[36] Henn V, Slupsky JR, Grafe M, et al. CD40 ligand on activated platelets triggers an inammatory reaction of endothelial
cells. Nature 1998;391(6667):591e4.
[37] Butler J, Parker D, Pillai R, et al. Effect of cardiopulmonary bypass on systemic release of neutrophil elastase and tumor
necrosis factor. J Thorac Cardiovasc Surg 1993;105(1):25e30.
*[38] Esper SA, Subramaniam K, Tanaka KA. Pathophysiology of cardiopulmonary bypass: current strategies for the prevention
and treatment of Anemia, coagulopathy, and organ dysfunction. Semin Cardiothorac Vasc Anesth 2014;18(2):161e76.
[39] Kolaczkowska E, Kubes P. Neutrophil recruitment and function in health and inammation. Nat Rev Immunol 2013;13(3):
159e75.
[40] Brinkmann V, Reichard U, Goosmann C, et al. Neutrophil extracellular traps kill bacteria. Science 2004;303(5663):
1532e5.
[41] Swerlick RA, Lee KH, Li LJ, et al. Regulation of vascular cell adhesion molecule 1 on human dermal microvascular
endothelial cells. J Immunol 1992;149(2):698e705.
[42] Ley K, Laudanna C, Cybulsky MI, et al. Getting to the site of inammation: the leukocyte adhesion cascade updated.
Nature reviews. Immunology 2007;7(9):678e89.
[43] Leeuwenberg JF, Smeets EF, Neefjes JJ, et al. E-selectin and intercellular adhesion molecule-1 are released by activated
human endothelial cells in vitro. Immunology 1992;77(4):543e9.
[44] Stadtmann A, Germena G, Block H, et al. The PSGL-1-L-selectin signaling complex regulates neutrophil adhesion under
ow. J Exp Med 2013;210(11):2171e80.
[45] Zarbock A, Kempf T, Wollert KC, et al. Leukocyte integrin activation and deactivation: novel mechanisms of balancing
inammation. J Mol Med Berl 2012;90(4):353e9.
[46] Rossaint J, Berger C, Van Aken H, et al. Cardiopulmonary bypass during cardiac surgery modulates systemic inflammation
by affecting different steps of the leukocyte recruitment cascade. PloS one 2012;7(9):e45738.
[47] Widgerow AD. Ischemia-reperfusion injury: inuencing the microcirculatory and cellular environment. Ann Plast Surg
2014;72(2):253e60.
[48] Fialkow L, Wang Y, Downey GP. Reactive oxygen and nitrogen species as signaling molecules regulating neutrophil
function. Free Radic Biol Med 2007;42(2):153e64.
[49] Duilio C, Ambrosio G, Kuppusamy P, et al. Neutrophils are primary source of O2 radicals during reperfusion after pro-
longed myocardial ischemia. Am J Physiol Heart Circ Physiol 2001;280(6):H2649e57.
[50] El-Benna J, Dang PM, Gougerot-Pocidalo MA. Priming of the neutrophil NADPH oxidase activation: role of p47phox
phosphorylation and NOX2 mobilization to the plasma membrane. Semin Immunopathol 2008;30(3):279e89.
[51] Angus DC, Barnato AE, Linde-Zwirble WT, et al. Use of intensive care at the end of life in the United States: an epide-
miologic study. Crit Care Med 2004;32(3):638e43.
[52] Kats S, Schonberger JP, Brands R, et al. Endotoxin release in cardiac surgery with cardiopulmonary bypass: pathophys-
iology and possible therapeutic strategies. an update. Eur J Cardiothorac Surg 2011;39(4):451e8.
[53] Chow JC, Young DW, Golenbock DT, et al. Toll-like receptor-4 mediates lipopolysaccharide-induced signal transduction.
J Biol Chem 1999;274(16):10689e92.
[54] Seekamp A, Mulligan MS, Till GO, et al. Requirements for neutrophil products and L-arginine in ischemia-reperfusion
injury. Am J Pathol 1993;142(4):1217e26.
[55] Zarbock A, Singbartl K, Ley K. Complete reversal of acid-induced acute lung injury by blocking of platelet-neutrophil
aggregation. J Clin investigation 2006;116(12):3211e9.
[56] Mahmood S, Bilal H, Zaman M, et al. Is a fully heparin-bonded cardiopulmonary bypass circuit superior to a standard
cardiopulmonary bypass circuit? Interact Cardiovasc Thorac Surg 2012;14(4):406e14.
[57] Baikoussis NG, Papakonstantinou NA, Apostolakis E. The benets of the mini-extracorporeal circulation in the minimal
invasive cardiac surgery era. J Cardiol 2014;63(6):391e6.
[58] Shapira OM, Aldea GS, Treanor PR, et al. Reduction of allogeneic blood transfusions after open heart operations by
lowering cardiopulmonary bypass prime volume. Ann Thorac Surg 1998;65(3):724e30.
[59] Sakwa MP, Emery RW, Shannon FL, et al. Coronary artery bypass grafting with a minimized cardiopulmonary bypass
circuit: a prospective, randomized trial. J Thorac Cardiovasc Surg 2009;137(2):481e5.
total minimal extracorporeal circulation. Eur J Cardiothorac Surg 2002;22(4):527e33.
[61] Bical OM, Fromes Y, Gaillard D, et al. Comparison of the inammatory response between miniaturized and standard CPB
circuits in aortic valve surgery. Eur J Cardiothorac Surg 2006;29(5):699e702.
[62] Yilmaz A, Sjatskig J, van Boven WJ, et al. Combined coronary artery bypass grafting and aortic valve replacement with
minimal extracorporeal closed circuit circulation versus standard cardiopulmonary bypass. Interact Cardiovasc Thorac
Surg 2010;11(6):754e7.
[63] Curtis N, Vohra HA, Ohri SK. Mini extracorporeal circuit cardiopulmonary bypass system: a review. Perfusion 2010;25(3):
115e24.
[64] Pagel PS. Myocardial protection by volatile anesthetics in patients undergoing cardiac surgery: a critical review of the
laboratory and clinical evidence. J Cardiothorac Vasc Anesth 2013;27(5):972e82.
[65] Bechtel A, Huffmyer J. Anesthetic management for cardiopulmonary bypass: update for 2014. Semin Cardiothorac Vasc
Anesth 2014;18(2):101e16.
[66] von Dossow V, Luetz A, Haas A, et al. Effects of remifentanil and fentanyl on the cell-mediated immune response in
patients undergoing elective coronary artery bypass graft surgery. J Int Med Res 2008;36(6):1235e47.
[67] Stephan H, Sonntag H, Schenk HD, et al. Effects of propofol on cardiovascular dynamics, myocardial blood ow and
myocardial metabolism in patients with coronary artery disease. Br J Anaesth 1986;58(9):969e75.
[68] Ma G, Chen J, Meng X, et al. High-dose propofol reduces S-100beta protein and neuron-specic enolase levels in patients
undergoing cardiac surgery. J Cardiothorac Vasc Anesth 2013;27(3):510e5.
[69] Vasileiou I, Xanthos T, Koudouna E, et al. Propofol: a review of its non-anaesthetic effects. Eur J Pharmacol 2009;
605(1e3):1e8.
[70] McMullan V, Alston R, Tyrrell J. Volatile anaesthesia during cardiopulmonary bypass. Perfusion 2015;30(1):6e16.
[71] Landoni G, Biondi-Zoccai GG, Zangrillo A, et al. Desurane and sevourane in cardiac surgery: a meta-analysis of ran-
domized clinical trials. J Cardiothorac Vasc Anesth 2007;21(4):502e11.
[72] Hemmerling TM, Cyr S, Terrasini N. Epidural catheterization in cardiac surgery: the 2012 risk assessment. Ann Card
Anaesth 2013;16(3):169e77.
[73] Taylor F, Huffman MD, Macedo AF, et al. Statins for the primary prevention of cardiovascular disease. Cochrane Database
Syst Rev 2013;1. Cd004816.
[74] Margaritis M, Channon KM, Antoniades C. Statins as regulators of redox state in the vascular endothelium: beyond lipid
lowering. Antioxid Redox Signal 2014;20(8):1198e215.
[75] Weitz-Schmidt G, Welzenbach K, Brinkmann V, et al. Statins selectively inhibit leukocyte function antigen-1 by binding
to a novel regulatory integrin site. Nat Med 2001;7(6):687e92.
*[76] de Waal BA, Buise MP, van Zundert AA. Perioperative statin therapy in patients at high risk for cardiovascular morbidity
undergoing surgery: a review. Br J Anaesth 2014;114(1):44e52.
[77] Moses ML, Camishion RC, Tokunaga K, et al. Effect of corticosteroid on the acidosis of prolonged cardiopulmonary bypass.
J Surg Res 1966;6(8):354e60.
[78] Dieleman JM, van Paassen J, van Dijk D, et al. Prophylactic corticosteroids for cardiopulmonary bypass in adults. Cochrane
Database Syst Rev 2011;5. Cd005566.
[79] Nebelsiek T, Beiras-Fernandez A, Kilger E, et al. Routine use of corticosteroids to prevent inammation response in
cardiac surgery. Recent Pat Cardiovasc Drug Discov 2012;7(3):170e4.
*[80] Dieleman JM, Nierich AP, Rosseel PM, et al. Intraoperative high-dose dexamethasone for cardiac surgery: a randomized
controlled trial. Jama 2012;308(17):1761e7.
[81] Garg AX, Vincent J, Cuerden M, et al. Steroids in caRdiac Ssssurgery (SIRS) trial: acute kidney injury substudy protocol of
an international randomised controlled trial. BMJ Open 2014;4(3):e004842.

Anaesthesiology
Neurocognitive outcomes after extracorporeal

membrane oxygenation
Lyndsey C. Graber, M.D., Resident in Anesthesiology a,
Nidia Quillinan, Ph.D., Assistant Professor of Anesthesiology a,
Eric J. Marrotte, D.O., Ph.D., Fellow in Critical Care Medicine b,
David L. McDonagh, M.D., Professor of Anesthesiology c,
Karsten Bartels, M.D., Assistant Professor of Anesthesiology a, *
a
Department of Anesthesiology, University of Colorado Denver, Aurora, CO, USA
b
Department of Neurology, Duke University, Durham, NC, USA
c
Department of Anesthesiology and Pain Management, UT Southwestern, Dallas, TX, USA
Keywords:
Extracorporeal membrane oxygenation (ECMO) has been a therapy
ECMO
of last resort for the treatment of severe cardiorespiratory failure
stroke
neurocognitive dysfunction since the 1970s [1]. In recent years, ECMO has seen a resurgence in
neuroprotection its use in adults. Recent work examining rates of ECMO use in the
US adult population, using Nationwide Inpatient Sample data,
quotes an increase in use of 433% from 2006 to 2011 [2]. While
much research has focused on neurologic injury after cardiac
surgery and cardiopulmonary bypass (CPB), the effects of ECMO on
neurocognitive function are less well described. This review aims
to summarize recent ndings as they pertain to pathophysiology,
monitoring techniques, prevention, therapy, and emerging exper-
imental concepts in the context of ECMO for adult patients. Given
that neurocognitive outcomes after cardiac surgery have been
recently reviewed [3,4], we will limit the discussion of ndings
from the cardiac surgery/CPB literature to those especially relevant
for ECMO.
* Corresponding author. University of Colorado Denver, 12401 E, 17th Avenue, MS B-113, Aurora, CO 80045, USA. Tel.: 1 303
724 4747; Fax: 1 303 848 7375.
E-mail address: karsten.bartels@ucdenver.edu (K. Bartels).
126 L.C. Graber et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 125e135
Veno-venous versus veno-arterial ECMO
Extracorporeal membrane oxygenation (ECMO) can be applied with gas exchange as the primary
goal: providing oxygen (O2) to and removing carbon dioxide (CO2) from the blood. In this case, venous
blood is drained from a large vein and returned to the right atrium (or another large vein): veno-
venous (VV) ECMO (Fig. 1). Many cannulation approaches exist and include femoral, internal jugu-
lar, subclavian, and superior/inferior vena cava access points. Especially during the 2009 H1N1
inuenza pandemic, VV ECMO use surged for the therapy of acute respiratory distress syndrome [5].
The highly cited CESAR study found that patients with severe acute respiratory distress syndrome
(ARDS) utilizing ECMO had a higher disability-free survival at 6 months as compared to conventional
care [6].
If a patient requires support of heart function, venous blood is routed to an oxygenator and then
pumped into the arterial circulation, thereby bypassing the heart. This conguration is termed
veno-arterial (VA) ECMO. Cannulation can be central, for example, from the right atrium to the
ascending aorta, or peripheral, for example, from the femoral vein to the femoral artery. VA ECMO is
used for the treatment of acute cardiac failure, including cardiogenic shock from acute myocardial
infarction, ischemic and non-ischemic cardiomyopathy, acute myocarditis, postcardiotomy syn-
drome, or failure to wean from cardiopulmonary bypass (CPB) postoperatively (to name a few)
[7,8]. ECMO is being increasingly utilized as a bridge to a durable ventricular assist device,
heartelung transplant, or as a rescue modality for primary graft dysfunction following cardiac or
lung transplantation.
Denition of neurocognitive dysfunction
Neurocognitive dysfunction ranges from subtle neurologic or neuropsychologic impairment to

overtly symptomatic stroke or even brain death [9]. As described by the World Health Organization
(WHO), stroke is dened as follows: rapidly developing clinical signs of focal (or global) disturbance of
Fig. 1. Veno-venous extracorporeal membrane oxygenation (VV ECMO). A bicaval, double-lumen central venous cannula is placed in
the right internal jugular vein. Deoxygenated blood is collected from both the inferior and the superior vena cava. After passing
through a centrifugal pump and a membrane oxygenator, the oxygenated blood is then returned to the right atrium through the
cannula's second lumen's orice. Various congurations are currently in use, including pumpless systems and alternative cannu-
lation techniques. Reproduced with permission from: Bartels K, et al. Perioperative organ injury. Anesthesiology. Dec 2013; 119
(6):1474e1489. Copyright Wolters Kluwer Health 2013 [22].
L.C. Graber et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 125e135 127
cerebral function, with symptoms lasting 24 h or longer or leading to death, with no apparent cause
other than of vascular origin. [10] The subtlety of a less apparent neurologic injury makes it much
harder to categorize and dene. To illustrate this, Mahanna and colleagues studied 232 patients who
underwent coronary artery bypass grafting (CABG) surgery and cognitive testing at different time
points [11]. When applying ve separate denitions of cognitive dysfunction to this cohort, the results
were quite disparate. For example, at 6 months following surgery, cognitive decline was present in
3.4e19.4% of patients, depending on which set of criteria was applied.
Preexisting (baseline) central nervous system dysfunction can make the differentiation of new
impairments from existing abnormalities challenging. This holds especially true for ECMO patients
who are commonly started on ECMO in extremis. Hence, obtaining a meaningful baseline examination
is often impossible. In addition, differentiating adverse effects attributable solely to ECMO versus those
due to coexisting pathologies is not straightforward.
Neurocognitive outcomes and ECMO
Development of chronic neurocognitive dysfunction is common after intensive care unit (ICU)
discharge even without ECMO therapy, and this has been reported in 25e78% of patients [12]. Similarly,
studies focused on ICU patients having had ECMO therapy show a broad range of incidences of neu-
rocognitive dysfunction (Table 1). In an analysis of 87 patients aged 15 years undergoing an average
duration of 91 h of ECMO, 50% were found to have suffered neurologic injury [13]. Diagnoses included
subarachnoid hemorrhage, ischemic infarctions, hypoxiceischemic encephalopathy, unexplained
coma, and death. This study also reported cerebral autopsy results on some of the included patients.
Here, nine out of 10 brains examined demonstrated ndings consistent with either hypoxic/ischemic or
hemorrhagic damage. Another single-center retrospective study reported discharge outcomes in 212
patients receiving ECMO support (69% with VA and 31% with VV conguration) [7]. The combined
complication rate for neurological problems was 47%, which included intracranial hemorrhage,
embolic or ischemic stroke, seizure, or generalized encephalopathy. These data contrast with an
Australian single-center cohort of 151 patients receiving ECMO (66.5% VA and 33.5% VV), where only
two separate incidences of neurologic complications were noted [14]. Gray et al. studied 2000 adult
and pediatric patients managed with either VA or VV ECMO from 1973 to 2010 [15]. In this sample, the
complication rate for intracranial bleeding or cerebral infarction was <10% in all groups except the
extracorporeal cardiopulmonary resuscitation group.
Specic ECMO approaches and indications may have different outcomes. A meta-analysis of 20
studies by Cheng et al. incorporated 1866 patients after ECMO for the treatment of cardiac arrest or
cardiogenic shock [16]. They found a stroke rate of 5.9% and a broader neurological complication rate
of 13.3%. Focusing on long-term neurological outcomes, a Norwegian study described 28 patients
who had been treated with ECMO (12 patients with VV ECMO and 16 patients with VA ECMO) [17].
After a follow-up of 0.5e12 years post treatment, 43% of patients were without clinically apparent
Table 1
Summary of select analyses of neurologic outcomes after ECMO therapy.
Study Population n Findings
Mateen FJ Prospective single-center data collection of ECMO 87 50% had a neurological event
et al., 2011 [13] patients 15 years with ECMO treatment for 12 h
Guttendorf J Retrospective single-center data collection of adult 212 47% had neurological problems
et al., 2014 [7] ECMO patients
Aubron C Prospective single-center data collection of ECMO 151 Two patients had neurologic
et al., 2013 [14] patients >16 years of age complications
Gray BW Single-center review of prospectively collected adult 2000 8% had intracranial bleeding or
et al., 2014 [18] and pediatric ECMO patients infarction
Cheng R Meta-analysis of ECMO patients treated for cardiogenic 1866 5.9% had a stroke and 13.3% had
et al., 2014 [16] shock and cardiac arrest a neurological complication
Risnes I Single-center study to determine cerebral status in 28 41% had neuropsychologic impairment
et al., 2006 [17] surviving ECMO patients and 52% had neuroradiologic lesions
ndings on examination, yet 41% demonstrated neuropsychological performance impairment, and

52% had cerebral radiologic ndings. Of note, neuroradiologic ndings consistent with cerebral injury
were more frequent after VA ECMO (75%) compared to VV ECMO (17%). In summary, these data
suggest that adverse neurologic outcomes are common after ECMO therapy. Given the limited
sample sizes, high severity of illness and level of comorbidities, and the retrospective nature of many
available studies, a causeeeffect relationship of ECMO and neurologic injury remains probable but
unproven.
Etiology and prevention
Emboli to the brain occur commonly during cardiac surgery, and they account for up to 62% of
strokes after CABG [18]. Similarly, embolization of thrombi from the circuit to the brain is a risk with
ECMO. This is more obvious in cases of VA ECMO, as the lung can act as a lter for embolic material
originating from a VV ECMO circuit. However, in the case of a shunt at the atrial level (Fig. 2), embo-
lization to the brain may still occur.
A genetic predisposition for less favorable neurologic outcomes 5 years after cardiac surgery has
been described [19]. These ndings suggest that a patient's genetic makeup rather than injury
occurring at the time of surgery may be responsible for long-term cognitive trajectories.
Hypoperfusion is a known cause of neurologic injury, with a mean arterial pressure (MAP) <30%
from baseline associated with postoperative stroke after general surgery [20]. In the setting of VA
ECMO, a femoralefemoral access strategy can lead to cerebral hypoperfusion if retrograde ow into the
thoracic aorta is unable to reach the aortic arch due to the competing ejection of deoxygenated blood
from the heart (Fig. 3). In this scenario, perfusion of the coronary and cerebral vessels with oxygenated
blood may be compromised [21].
Inammatory mediators, systemically circulating as well as locally acting in the central nervous
system, can result in neuronal damage. These are seen in various critical conditions such as infection,
major illness, traumatic injury, and the postsurgical period [22]. The cerebral injury likely results in
part from the disruption of the endothelium causing dysfunction of the bloodebrain barrier [23].
The effects of hemodilution and glycemic control have been studied in CPB and in patients who
underwent cardiac surgery. In a study by Mathew et al., extreme hemodilution during CPB in cardiac
surgery (hematocrit on CPB, 15e18%) was associated with increased cognitive decline [24]. In addition,
in a study by Karkouti et al., a low hematocrit during CPB was associated with an increased risk of
Fig. 2. A patent foramen ovale (red arrow), as diagnosed here with color-ow Doppler using perioperative transesophageal echo-
cardiography, can permit embolic material (thrombus, air, etc.) to bypass the lungs and ow directly from the right atrium (RA) to
the left atrium (LA) with the potential to cause embolic stroke.
Fig. 3. Venoarterial extracorporeal membrane oxygenation (VA ECMO) with peripheral femoral vein and femoral artery cannulation.
The remaining native cardiac output can deliver deoxygenated blood into the ascending aorta. Here, myocardial and cerebral
oxygenation is dependent on competition between the ow of oxygenated blood from the ECMO circuit via the femoral artery and
aorta and the ow of deoxygenated blood from the heart.
perioperative stroke [25]. The etiology of stroke from hemodilution is unclear, but it may be related to
an increased number of emboli and/or decreased cerebral oxygenation during CPB [26,27]. While
hyperglycemia (glucose 200 mg/dL) has been associated with cognitive dysfunction after cardiac
surgery [28], Lazar and colleagues showed that aggressive glycemic control in diabetic patients un-
dergoing CPB increased the incidence of hypoglycemic events, but it did not result in any signicant
improvement in outcomes [29]. Additionally, two separate studies were unable to demonstrate that
tight glycemic control during cardiopulmonary bypass improved neurocognitive outcomes [30,31].
Strict glucose control (81e108 mg/dL) in critically ill patients has actually been shown to increase
mortality [32]. Although there is a striking lack of evidence [78], current recommendations are that
blood glucose should be controlled and maintained (180 mg/dL) after cardiac surgery [33]. Optimal
glucose levels in ECMO patients are not yet dened.
Therapeutic hypothermia is currently used for neuroprotection in patients who have undergone
return to spontaneous circulation after cardiac arrest, but remain comatose. Risks associated with
therapeutic hypothermia include coagulopathy and immunosuppression [34]. Therapeutic hypother-
mia has not been shown to change the rate of intraoperative emboli, or the rate of stroke or cognitive
decline in patients undergoing CPB [35e37]. Based on current data, therefore, therapeutic hypothermia
appears safe but not effective in decreasing the rate of stroke or cognitive decline in patients after CPB.
Protocols involving combined mechanical cardiopulmonary resuscitation, hypothermia, ECMO, and
early reperfusion have been successfully implemented for the treatment of refractory cardiac arrest
[38], but neurocognitive outcomes remain incompletely described.
Monitoring
The National Institutes of Health Stroke Scale (NIHSS) is a standardized, graded neurological ex-
amination that can be utilized to detect focal neurological impairments [39]. The examination focuses
on rating speech and language, cognition, visual elds, motor and sensory function, and coordination/
ataxia in a standardized, validated, and easily reproducible manner with minimal training. Neuro-
cognitive testing, although a frequently used tool in research, is not easily applied to routine clinical
evaluation. Commonly used tests include the Rey auditory verbal learning test, the Trail-making A test,
the Trail-making B test, and the Grooved pegboard test [9]. These tests assess different cognitive do-
mains, such as attention, concentration, memory, language, perception, and motor functioning (Table
2). Computed tomography (CT) is the primary imaging modality to assess acute intracranial compli-
cations in ECMO patients [40]. While magnetic resonance imaging offers higher resolution and options
for functional analysis, at present it cannot be used in ECMO patients due to ferromagnetic equipment/
pumps.
Currently, the MAP to achieve adequate cerebral perfusion pressure (CPP) during CPB and ECMO is
often not tailored to the individual patient, particularly in the ICU. Novel and more sophisticated ap-
proaches to determine individual cerebral autoregulation and pressure limits, for example, using near-
infrared spectroscopy (NIRS) or transcranial Doppler sonography, are on the horizon. In one study, CPP
above the upper cerebral autoregulation limit during CPB was found to be associated with post-
operative delirium [41]. Alternatively, CPP below the cerebral autoregulation threshold during CPB has
been related to increased morbidity and postoperative mortality [42]. Certainly, individualizing blood-
pressure targets in ECMO patients is conceptually appealing.
Cerebral NIRS permits continuous, noninvasive measurement of regional O2 saturation in the brain
[43,44]. Brain hypoperfusion, for example, from competitive ow of hypoxemic blood ejected through
the heart, is a major concern in VA ECMO patients (Fig. 3). Wong and colleagues used bilateral cerebral
oximetry to guide hemodynamic interventions during ECMO [45]. In 16 of 20 patients studied, altering
parameters such as pressure, oxygenation, or ECMO blood ow resulted in the resolution of brain
hypoxia.
Continuous electroencephalographic (EEG) monitoring is common in pediatric ECMO patients [46].
Specically, it can be used to detect nonconvulsive seizures and nonconvulsive status epilepticus.
Transcranial Doppler sonography can be used to detect macro-emboli and micro-emboli within the
cerebral circulation (middle cerebral arteries via temporal bone windows). Micro-embolic signals are
Table 2
Neurocognitive testing modalities utilized in cognitive dysfunction studies. The Rey auditory verbal learning test, Trail-making
tests A and B, and the Grooved pegboard test are included in consensus recommendations for cognitive testing in cardiac surgery.
WAIS-R Wechsler Adult Intelligence Scale revised [9,11,76,77].
Neurocognitive test Cognitive domain Description
Rey auditory Verbal memory Immediate recall of 15 verbally presented, unrelated

verbal learning words over three attempts
Trail-making A Attention/concentration Connecting a series of numbers, in sequence, as fast
as possible
Trail-making B Attention/concentration Connecting a series of numbers and letters, in
sequence, as fast as possible
Grooved pegboard Motor skills Insertion of pegs, with a key along one side, into
25 holes in sequence as quickly as possible with
dominant and then nondominant hand.
Randt memory Verbal memory Recall of details of a short story immediately after
hearing it and 30 min after hearing it
WAIS-R digit symbol Visual-motor performance/perception/speed, Reproduction of as many coded symbols as possible
subtest visual memory, learning according to a coding-scheme pairing digits and
symbols, in 90 s
WAIS-R digit span Verbal memory, attention/concentration Repeating a series of digits presented orally,
subtest ordered forward, and then in reverse
Benton-revised visual Visual perception/memory Drawing reproductions of geometric shapes from
retention memory after a 10-s viewing
independently associated with poor outcomes in patients with ischemic stroke [47]. However, in pa-
tients during CABG using CPB, it has not been proven to be of any prognostic value (perhaps because
they are more often gaseous rather than particulate) [48]. More research is required to determine its
usefulness in ECMO therapy.
Therapeutic approaches
Several clinical trials have attempted to reduce postoperative cognitive dysfunction after cardiac
surgery. Neuroprotection has been attempted with 17 beta-estradiol [49], magnesium [50], lidocaine
[51], piracetam [52], steroids [53], ischemic preconditioning [54], and off-pump cardiopulmonary
bypass [55,56]. Unfortunately, none of these approaches has provided any tangible improvements in
relevant clinical outcomes.
Treatment options for postoperative stroke are limited. In the updated 2013 American Heart As-
sociation/American Stroke Association (AHA/ASA) guidelines, recent intracranial or intraspinal surgery
continues to be an absolute exclusion to intravenous recombinant tissue plasminogen activator (IV
rtPA) [57]. In these recommendations, major surgery is a relative exclusion criteria. Systemic throm-
bolysis is generally not an option in the VA or VV ECMO patient. Catheter-directed intra-arterial tPA (IA
tPA) and, more often, mechanical thrombectomy can be considered in patients with ischemic stroke
after cardiac surgery [58e61]. A previous study has shown that IV tPA and IA tPA can provide equivalent
therapeutic outcomes [62]. Hence, intra-arterial approaches for mechanical embolectomy and possibly
catheter-directed thrombolysis may be feasible in patients after cardiac surgery or ECMO, when sys-
temic thrombolysis is contraindicated. Therapy is time limited, so immediate diagnosis and inter-
vention is essential. Therefore, minimizing patient sedation to allow for clinical neurologic assessment
is desirable.
Emerging experimental concepts
The failure of human clinical trials to successfully translate preclinically proven pharmacologic
neuroprotectants suggests that advances are needed in both preclinical modeling of hypoxiceischemic
brain injury and the approach to translational research [63]. Strategies to protect the brain from
hypoxic or ischemic injury have primarily focused on preventing excitotoxicity, oxidative stress, and
inammation. There is increasing evidence from animal studies and humans suggesting that, in
addition to neuronal death, ischemia causes neurophysiological and connectivity changes that
contribute to the pathophysiology. Cortical excitability and interhemispheric connectivity are altered
following ischemic stroke [64e66]. In addition, there is increased tonic inhibition of the motor cortex
that, when reversed, improves motor function [67]. Synaptic plasticity, a cellular process that underlies
learning and memory, is impaired in ischemia-sensitive brain regions following global and focal
ischemia [68,69]. Therefore, in addition to preventing neuronal death, therapeutic interventions
should also aim to prevent more subtle neurophysiological changes. Preclinical studies to test therapies
to prevent ischemic brain injury should include analyses of neuronal death and synaptic and network
functions.
Due to the lack of therapies available for ischemic brain injury, recent research has focused on
improving recovery and plasticity in the post-ischemic brain. Human and animal studies have utilized
pharmacological therapies to enhance plasticity and functional connectivity. For example, the gamma-
aminobutyric acid (GABA) a5 antagonist L655,708 blocks tonic inhibition and improves motor recovery
following stroke [70]. Similar enhancements in recovery have been observed by stimulating norad-
renergic signaling or by enhancing N-methyl D-aspartate (NMDA) receptor function [71,72]. Human
and animal studies have also demonstrated that deep-brain stimulation and transcranial magnetic
stimulation in areas remote to the injury, such as the cerebellum, the contralateral cortex, and even the
vagus nerve, can enhance functional connectivity and recovery following stroke [73e75]. These studies
demonstrate that the stimulation of areas outside of the infarcted area can potentially enhance
function in post-ischemic patients, and can emphasize the importance of understanding network al-
terations caused by cerebral ischemia.
Summary
Neurologic injury in the setting of ECMO is common. High patient acuity, severity of comorbidities,
and lack of baseline examinations complicate the determination and full understanding of a causal link
between ECMO and neurocognitive injury. Factors that likely contribute to neurocognitive dysfunction
after ECMO include genetic predisposition, hypoperfusion, embolism, hemodilution, and inamma-
tion. Point-of-care cerebral perfusion monitors will allow for the optimization of pressure and ow in
ECMO patients. Future clinical studies should use standardized denitions of neurocognitive
dysfunction, and they should aim to report serial neurologic outcomes prior to and after ECMO therapy.
Finally, ECMO is an ideal setting in which to apply future neuroprotectant therapeutics as the inter-
vention may be delivered simultaneously to the neurologic insult. The expanding role of ECMO in
critical illness demands that we seek greater insight into the pathophysiology of ECMO-associated
neurologic injury and interventions to ameliorate such injury.
Practice points
Extracorporeal membrane oxygenation (ECMO) use in adults for both cardiac and pulmonary
support has increased in recent years.
ECMO patients commonly exhibit post treatment cognitive dysfunction. However, whether
this finding is due to ECMO therapy or other comorbidities remains unclear.
Neurological complications during ECMO therapy can occur secondary to embolism
(thrombus, particulate, and gas), hemorrhage, inflammation, hypoperfusion, and possibly
hemodilution.
Monitoring neurological function in ECMO patients currently relies on the physical exami-
nation. Other neuromonitoring technologies include brain imaging, cerebral near-infrared
spectroscopy, electroencephalographic (EEG) monitoring, and transcranial Doppler.
While systemic therapy with tissue plasminogen activator (tPA) is usually contraindicated in
ECMO patients with acute ischemic stroke, mechanical thrombectomy may be considered in
select patients.
Research agenda
Standardized definitions of neurocognitive dysfunction should be used for clinical trials.

Further work is needed to define the mechanisms of extracorporeal membrane oxygenation
(ECMO)-induced neurocognitive injury.
Experimental work should aim to better understand the impact of ischemic injury on network
connectivity as opposed to solely focus on neuronal death.
Stimulation of areas outside of the infarcted area should be considered as a promising target
to enhance recovery after stroke.
Conict of interest statement
Lyndsey C. Graber, M.D., Nidia Quillinan, Ph.D., Eric J. Marrotte, D.O., Ph.D., and Karsten Bartels, M.D.,
report no potential conicts of interest. David L. McDonagh, M.D., reports being a consultant for
Cephalogics Corporation, LLC.
References
[1] Zapol WM, Snider MT, Schneider RC. Extracorporeal membrane oxygenation for acute respiratory-failure. Anesthesiology
1977;46(4):272e85.
[2] Sauer CM, Yuh DD, Bonde P. Extracorporeal membrane oxygenation use has increased by 433% in adults in the United
States from 2006 to 2011. ASAIO J JaneFeb 2015;61(1):31e6.
[3] Bartels K, McDonagh DL, Newman MF, et al. Neurocognitive outcomes after cardiac surgery. Curr Opin Anaesthesiol Feb
2013;26(1):91e7.
[4] Uysal S, Reich DL. Neurocognitive outcomes of cardiac surgery: a review. J Cardiothorac Vasc Anesth May 3 2013;27(5):
958e71.
*[5] Pham T, Combes A, Roze H, et al. Extracorporeal membrane oxygenation for pandemic inuenza A(H1N1)-induced acute
respiratory distress syndrome: a cohort study and propensity-matched analysis. Am J Respir Crit Care Med Feb 1 2013;
187(3):276e85.
*[6] Peek GJ, Mugford M, Tiruvoipati R, et al. Efcacy and economic assessment of conventional ventilatory support versus
trial. Lancet Oct 17 2009;374(9698):1351e63.
*[7] Guttendorf J, Boujoukos AJ, Ren D, et al. Discharge outcome in adults treated with extracorporeal membrane oxygenation.
Am J Crit Care Sep 2014;23(5):365e77.
[8] Ventetuolo CE, Muratore CS. Extracorporeal life support in critically ill adults. Am J Respir Crit Care Med Sep 1 2014;
190(5):497e508.
[9] Murkin JM, Newman SP, Stump DA, et al. Statement of consensus on assessment of neurobehavioral outcomes after
cardiac surgery. Ann Thorac Surg May 1995;59(5):1289e95.
[10] The World Health Organization MONICA project (monitoring trends and determinants in cardiovascular disease): a major
international collaboration. WHO MONICA project principal investigators. J Clin Epidemiol 1988;41(2):105e14.
[11] Mahanna EP, Blumenthal JA, White WD, et al. Dening neuropsychological dysfunction after coronary artery bypass
grafting. Ann Thorac Surg May 1996;61(5):1342e7.
[12] Hopkins RO, Jackson JC. Long-term neurocognitive function after critical illness. Chest Sep 2006;130(3):869e78.
[13] Mateen FJ, Muralidharan R, Shinohara RT, et al. Neurological injury in adults treated with extracorporeal membrane
oxygenation. Arch Neurol Dec 2011;68(12):1543e9.
*[14] Aubron C, Cheng AC, Pilcher D, et al. Factors associated with outcomes of patients on extracorporeal membrane
oxygenation support: a 5-year cohort study. Crit Care 2013;17(2):R73.
[15] Gray BW, Haft JW, Hirsch JC, et al. Extracorporeal life support: experience with 2,000 patients. ASAIO J JaneFeb 2015;
61(1):2e7.
*[16] Cheng R, Hachamovitch R, Kittleson M, et al. Complications of extracorporeal membrane oxygenation for treatment of
cardiogenic shock and cardiac arrest: a meta-analysis of 1,866 adult patients. Ann Thorac Surg Feb 2014;97(2):610e6.
[17] Risnes I, Wagner K, Nome T, et al. Cerebral outcome in adult patients treated with extracorporeal membrane oxygenation.
Ann Thorac Surg Apr 2006;81(4):1401e6.
[18] Likosky DS, Marrin CA, Caplan LR, et al. Determination of etiologic mechanisms of strokes secondary to coronary artery
bypass graft surgery. Stroke Dec 2003;34(12):2830e4.
[19] Bartels K, Li YJ, Li YW, et al. Apolipoprotein E4 genotype is associated with less improvement in cognitive function ve
years after cardiac surgery: a retrospective cohort study. Can J Anaesth. 2015 Mar 6 [Epub ahead of print].
[20] Bijker JB, Persoon S, Peelen LM, et al. Intraoperative hypotension and perioperative ischemic stroke after general surgery:
a nested case-control study. Anesthesiology Mar 2012;116(3):658e64.
[21] Abrams D, Combes A, Brodie D. Extracorporeal membrane oxygenation in cardiopulmonary disease in adults. J Am Coll
Cardiol Jul 1 2014;63(25 Pt A):2769e78.
[22] Bartels K, Karhausen J, Clambey ET, et al. Perioperative organ injury. Anesthesiology Dec 2013;119(6):1474e89.
[23] Bartels K, Ma Q, NV T, et al. Effects of deep hypothermic circulatory arrest on the blood brain barrier in a cardiopulmonary
bypass model e a pilot study. Heart Lung Circ Oct 2014;23(10):981e4.
[24] Mathew JP, Mackensen GB, Phillips-Bute B, et al. Effects of extreme hemodilution during cardiac surgery on cognitive
function in the elderly. Anesthesiology Oct 2007;107(4):577e84.
[25] Karkouti K, Djaiani G, Borger MA, et al. Low hematocrit during cardiopulmonary bypass is associated with increased risk
of perioperative stroke in cardiac surgery. Ann Thorac Surg Oct 2005;80(4):1381e7.
[26] Jonas RA. Optimal hematocrit for adult cardiopulmonary bypass. J Cardiothorac Vasc Anesth Oct 2001;15(5):672.
[27] Nollert G, Reichart B. Cardiopulmonary bypass and cerebral injury in adults. Shock 2001;16(Suppl. 1):16e9.
[28] Puskas F, Grocott HP, White WD, et al. Intraoperative hyperglycemia and cognitive decline after CABG. Ann Thorac Surg
Nov 2007;84(5):1467e73.
[29] Lazar HL, McDonnell MM, Chipkin S, et al. Effects of aggressive versus moderate glycemic control on clinical outcomes in
diabetic coronary artery bypass graft patients. Ann Surg Sep 2011;254(3):458e63. discussion 463e454.
[30] Butterworth J, Wagenknecht LE, Legault C, et al. Attempted control of hyperglycemia during cardiopulmonary bypass fails
to improve neurologic or neurobehavioral outcomes in patients without diabetes mellitus undergoing coronary artery
bypass grafting. J Thorac Cardiovasc Surg Nov 2005;130(5):1319.
*[31] Gandhi GY, Nuttall GA, Abel MD, et al. Intensive intraoperative insulin therapy versus conventional glucose management
during cardiac surgery: a randomized trial. Ann Intern Med Feb 20 2007;146(4):233e43.
*[32] Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med Mar
26 2009;360(13):1283e97.
[33] Hillis LD, Smith PK, Anderson JL, et al. 2011 ACCF/AHA guideline for coronary artery bypass graft surgery: executive
summary: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice
Guidelines. J Thorac Cardiovasc Surg Jan 2012;143(1):4e34.
[34] Mackensen GB, McDonagh DL, Warner DS. Perioperative hypothermia: use and therapeutic implications. J Neurotrauma
Mar 2009;26(3):342e58.
[35] Grigore AM, Mathew J, Grocott HP, et al. Prospective randomized trial of normothermic versus hypothermic cardiopul-
monary bypass on cognitive function after coronary artery bypass graft surgery. Anesthesiology Nov 2001;95(5):1110e9.
[36] Ho KM, Tan JA. Benets and risks of maintaining normothermia during cardiopulmonary bypass in adult cardiac surgery:
a systematic review. Cardiovasc Ther Aug 2011;29(4):260e79.
[37] Boodhwani M, Rubens F, Wozny D, et al. Effects of sustained mild hypothermia on neurocognitive function after coronary
artery bypass surgery: a randomized, double-blind study. J Thorac Cardiovasc Surg Dec 2007;134(6):1443e50. discussion
1451e1442.
*[38] Stub D, Bernard S, Pellegrino V, et al. Refractory cardiac arrest treated with mechanical CPR, hypothermia, ECMO and
early reperfusion (the CHEER trial). Resucitation. 2014 Oct 2 [Epub ahead of print].
[39] Goldstein LB, Samsa GP. Reliability of the National Institutes of Health Stroke Scale. Extension to non-neurologists in the
context of a clinical trial. Stroke Feb 1997;28(2):307e10.
[40] Lidegran MK, Mosskin M, Ringertz HG, et al. Cranial CT for diagnosis of intracranial complications in adult and pediatric
patients during ECMO: clinical benets in diagnosis and treatment. Acad Radiol Jan 2007;14(1):62e71.
*[41] Hori D, Brown C, Ono M, et al. Arterial pressure above the upper cerebral autoregulation limit during cardiopulmonary
bypass is associated with postoperative delirium. Br J Anaesth Dec 2014;113(6):1009e17.
*[42] Ono M, Brady K, Easley RB, et al. Duration and magnitude of blood pressure below cerebral autoregulation threshold
during cardiopulmonary bypass is associated with major morbidity and operative mortality. J Thorac Cardiovasc Surg Jan
2014;147(1):483e9.
[43] Maldonado Y, Singh S, Taylor MA. Cerebral near-infrared spectroscopy in perioperative management of left ventricular
assist device and extracorporeal membrane oxygenation patients. Curr Opin Anaesthesiol Feb 2014;27(1):81e8.
[44] Tyree K, Tyree M, DiGeronimo R. Correlation of brain tissue oxygen tension with cerebral near-infrared spectroscopy and
mixed venous oxygen saturation during extracorporeal membrane oxygenation. Perfusion Sep 2009;24(5):325e31.
[45] Wong JK, Smith TN, Pitcher HT, et al. Cerebral and lower limb near-infrared spectroscopy in adults on extracorporeal
membrane oxygenation. Artif Organs Aug 2012;36(8):659e67.
[46] Abend NS, Dlugos DJ, Clancy RR. A review of long-term EEG monitoring in critically ill children with hypoxic-ischemic
encephalopathy, congenital heart disease, ECMO, and stroke. J Clin Neurophysiol Apr 2013;30(2):134e42.
[47] Serena J, Segura T, Castellanos M, et al. Microembolic signal monitoring in hemispheric acute ischaemic stroke: a pro-
spective study. Cerebrovasc Dis JuleAug 2000;10(4):278e82.
[48] Rodriguez RA, Rubens FD, Wozny D, et al. Cerebral emboli detected by transcranial Doppler during cardiopulmonary
bypass are not correlated with postoperative cognitive decits. Stroke Oct 2010;41(10):2229e35.
[49] Hogue Jr CW, Freedland K, Hershey T, et al. Neurocognitive outcomes are not improved by 17beta-estradiol in post-
menopausal women undergoing cardiac surgery. Stroke Jul 2007;38(7):2048e54.
[50] Mathew JP, White WD, Schinderle DB, et al. Intraoperative magnesium administration does not improve neurocognitive
function after cardiac surgery. Stroke Dec 2013;44(12):3407e13.
[51] Mathew JP, Mackensen GB, Phillips-Bute B, et al. Randomized, double-blinded, placebo controlled study of neuro-
protection with lidocaine in cardiac surgery. Stroke Mar 2009;40(3):880e7.
[52] Holinski S, Claus B, Alaaraj N, et al. Cerebroprotective effect of piracetam in patients undergoing open heart surgery. Ann
Thorac Cardiovasc Surg 2011;17(2):137e42.
[53] Ottens TH, Dieleman JM, Sauer AC, et al., Dexamethasone for Cardiac Surgery Study G. Effects of dexamethasone on
cognitive decline after cardiac surgery: a randomized controlled trial. Anesthesiology 2014;121(3):492e500.
[54] Meybohm P, Renner J, Broch O, et al. Postoperative neurocognitive dysfunction in patients undergoing cardiac surgery
after remote ischemic preconditioning: a double-blind randomized controlled pilot study. PloS One 2013;8(5):e64743.
[55] Shroyer AL, Grover FL, Hattler B, et al. On-pump versus off-pump coronary-artery bypass surgery. N Engl J Med Nov 5
2009;361(19):1827e37.
[56] Lamy A, Devereaux PJ, Prabhakaran D, et al. Effects of off-pump and on-pump coronary-artery bypass grafting at 1 year.
N Engl J Med Mar 28 2013;368(13):1179e88.
[57] Jauch EC, Saver JL, Adams Jr HP, et al. Guidelines for the early management of patients with acute ischemic stroke: a
guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke Mar
2013;44(3):870e947.
[58] Fukuda I, Imazuru T, Osaka M, et al. Thrombolytic therapy for delayed, in-hospital stroke after cardiac surgery. Ann Thorac
Surg Oct 2003;76(4):1293e5.
[59] Katzan IL, Masaryk TJ, Furlan AJ, et al. Intra-arterial thrombolysis for perioperative stroke after open heart surgery.
Neurology Mar 23 1999;52(5):1081e4.
[60] Moazami N, Smedira NG, McCarthy PM, et al. Safety and efcacy of intraarterial thrombolysis for perioperative stroke
after cardiac operation. Ann Thorac Surg Dec 2001;72(6):1933e7. discussion 1937e1939.
[61] Chalela JA, Katzan I, Liebeskind DS, et al. Safety of intra-arterial thrombolysis in the postoperative period. Stroke Jun
2001;32(6):1365e9.
[62] Broderick JP, Palesch YY, Demchuk AM, et al. Endovascular therapy after intravenous t-PA versus t-PA alone for stroke.
N Engl J Med Mar 7 2013;368(10):893e903.
[63] Warner DS, James ML, Laskowitz DT, et al. Translational research in acute central nervous system injury: lessons learned
and the future. JAMA Neurol Oct 2014;71(10):1311e8.
[64] Grefkes C, Fink GR. Connectivity-based approaches in stroke and recovery of function. Lancet Neurol Feb 2014;13(2):
206e16.
[65] Grefkes C, Fink GR. Reorganization of cerebral networks after stroke: new insights from neuroimaging with connectivity
approaches. Brain May 2011;134(Pt 5):1264e76.
[66] Rehme AK, Fink GR, von Cramon DY, et al. The role of the contralesional motor cortex for motor recovery in the early days
after stroke assessed with longitudinal FMRI. Cereb Cortex Apr 2011;21(4):756e68.
[67] Carmichael ST. Brain excitability in stroke: the yin and yang of stroke progression. Arch Neurol Feb 2012;69(2):161e7.
[68] Orla JE, Shimizu K, Garske AK, et al. Increasing small conductance Ca2-activated potassium channel activity reverses
ischemia-induced impairment of long-term potentiation. Eur J Neurosci Oct 2014;40(8):3179e88.
[69] Mori K, Yoshioka M, Suda N, et al. An incomplete cerebral ischemia produced a delayed dysfunction in the rat hippo-
campal system. Brain Res Jun 8 1998;795(1e2):221e6.
*[70] Clarkson AN, Huang BS, Macisaac SE, et al. Reducing excessive GABA-mediated tonic inhibition promotes functional
recovery after stroke. Nature Nov 11 2010;468(7321):305e9.
[71] Wang LE, Fink GR, Diekhoff S, et al. Noradrenergic enhancement improves motor network connectivity in stroke patients.
Ann Neurol Feb 2011;69(2):375e88.
[72] Dhawan J, Benveniste H, Luo Z, et al. A new look at glutamate and ischemia: NMDA agonist improves long-term func-
tional outcome in a rat model of stroke. Future Neurol Nov 1 2011;6(6):823e34.
[73] Hays SA, Khodaparast N, Hulsey DR, et al. Vagus nerve stimulation during rehabilitative training improves functional
recovery after intracerebral hemorrhage. Stroke Oct 2014;45(10):3097e100.
[74] Machado AG, Baker KB, Schuster D, et al. Chronic electrical stimulation of the contralesional lateral cerebellar nucleus
enhances recovery of motor function after cerebral ischemia in rats. Brain Res Jul 14 2009;1280:107e16.
[75] Xiang Y, Liu H, Yan T, et al. Functional electrical stimulation-facilitated proliferation and regeneration of neural precursor
cells in the brains of rats with cerebral infarction. Neural Regen Res Feb 1 2014;9(3):243e51.
[76] Hoyer WJ, Stawski RS, Wasylyshyn C, et al. Adult age and digit symbol substitution performance: a meta-analysis. Psychol
Aging Mar 2004;19(1):211e4.
[77] Demakis GJ, Sawyer TP, Fritz D, et al. Incidental recall on WAIS-R digit symbol discriminates Alzheimer's and Parkinson's
diseases. J Clin Psychol Mar 2001;57(3):387e94.
[78] Fullerton DA, Sundt TM, Fritz D. Process versus outcome: the sugar window. Ann Thorac Surg. Dec 2014;98(6):1902e4.

Anaesthesiology
Myocardial injury and protection related to

cardiopulmonary bypass
Stefan De Hert, MD, PhD, Anesthesiologist *,
Anneliese Moerman, MD, PhD, Anesthesiologist 1
Department of Anaesthesiology, Ghent University Hospital, Ghent University, Ghent B-9000, Belgium
Keywords:
During cardiac surgery with cardiopulmonary bypass, the heart is
heart
isolated from the circulation. This inevitably induces myocardial
myocardium
ischemiaereperfusion injury ischemia. In addition to this ischemic insult, an additional hit will
cardiopulmonary bypass occur upon reperfusion, which may worsen the extent of tissue
conditioning damage and organ dysfunction. Over the years, several strategies
cardioprotection have been developed that aim to attenuate and/or modulate the
extent of this ischemiaereperfusion injury related to the episode
of cardiopulmonary bypass. This article reviews the pathophysi-
ology of myocardial injury related to cardiopulmonary bypass and
summarizes potential therapeutic strategies that may modulate
the extent of this myocardial injury.
Introduction
The introduction of cardiopulmonary bypass (CPB) has allowed for the development of cardiac
surgery. Indeed, the separation of the lung and heart for surgery, while maintaining adequate ow of
oxygenated blood to all the other organs and tissues of the body, has allowed for the surgical correction
of even the most complex cardiac conditions. Since its introduction in 1953, CPB techniques have
immensely improved and nowadays most patients tolerate the procedure relatively well. Nevertheless,
despite major advances in technologies and scientic insights, and despite the implementation of
strategies to reduce the pro-inammatory effects of CPB on the myocardium, the fact remains that
during cardiac surgery with CPB the heart is isolated from the circulation and as a consequence suffers
* Corresponding author. Tel.: 32 (0)9 332 32 81; Fax: 32 (0)9 332 49 87.
E-mail addresses: stefan.dehert@ugent.be (S. De Hert), annelies.moerman@ugent.be (A. Moerman).
1
Tel.: 32 (0)9 332 32 81; Fax: 32 (0)9 332 49 87.
138 S. De Hert, A. Moerman / Best Practice & Research Clinical Anaesthesiology 29 (2015) 137e149
from an ischemic insult followed by an additional hit upon reperfusion, the so-called ische-
miaereperfusion injury.
Mechanisms of myocardial injury related to CPB
Apart from mechanical trauma, both CPB and aortic cross-clamping trigger myocardial injury. CPB
provokes a vigorous systemic inammatory response, induced by the exposure of blood elements to
non-physiological surfaces, resulting e among others e in myocardial damage. Furthermore, excluding
the heart from the systemic circulation renders the myocardium ischemic, and upon reperfusion,
triggers post-ischemic myocardial dysfunction. Central in the pathogenesis of ischemic myocardial
injury is the depletion of high-energy phosphates and the disturbance of normal intracellular calcium
homeostasis [1,2] (Fig. 1). Upon this period of myocardial ischemia of variable duration, reperfusion
may lead to additional injury beyond that generated by the period of ischemia and may manifest as the
occurrence of arrhythmias, reversible contractile dysfunction (myocardial stunning), endothelial
dysfunction, and ultimately irreversible reperfusion injury with myocardial cell death [3]. The most
crucial in the pathophysiology of reperfusion injury is the extent of damage to the mitochondrion,
related to the degree of opening of the mitochondrial permeability transition pore (MPTP) [4,5] (Fig. 2).
As a consequence, prevention and treatment of CPB-related myocardial ischemia should not only be
directed towards minimizing energy consumption and cellular protection during the period of
ischemia but should also include measures to prevent or minimize the extent of reperfusion injury.
Cardioprotective strategies during CPB: key approaches
Preparing the myocardium for CPB
It is a generally accepted concept that the myocardium should be prepared as much as possible for
the period of myocardial ischemia occurring during the aortic cross-clamp. Over the years, different
strategies have been explored, some of which seem very logical and straightforward such as the
maintenance of hemodynamic stability with avoidance of tachycardia and hypo- or hypertension.
Other strategies have been proposed and then were again abandoned because of insufcient proof of
clinical efcacy. It is beyond the scope of this review to discuss all these proposed strategies. Table 1
summarizes the principles of the strategies that have been proposed over the years to prepare the
myocardium for the induced arrest.
Fig. 1. Schematic representation of the different factors contributing to myocardial dysfunction and/or damage with
ischemiaereperfusion.
S. De Hert, A. Moerman / Best Practice & Research Clinical Anaesthesiology 29 (2015) 137e149 139
Fig. 2. Schematic representation of the presumed role of the mitochondrial permeability transition pore (MPTP) in the pathogenesis
of ischemiaereperfusion injury. During ischemia, the hypoxic conditions result in a decrease and depletion of the energy stores and
an accumulation of metabolites resulted in tissue acidosis. At the level of the mitochondriae, the MPTP remains closed keeping
mitochondrial integrity intact at this level. Upon reperfusion, the tissue re-oxygenation triggers a burst of reactive oxygen species
release and the MPTP opens, leading to disturbance of the electrochemical gradients over the mitochondrial membrane, swelling of
the mitochondrial intermembrane space, and disruption of the supramolecular complex containing the proton pump, ATP synthase,
the adenine nucleotide transporter, and mitochondrial creatine kinase. If the MPTP opening is minimal, full functional recovery of
the mitochondrion may occur. When MPTP opening occurs in 10e50% of the mitochondriae, either recovery or apoptosis may occur
depending on the extent of damage. However, if MPTP opening occurs in more than 50%, necrosis of cell will occur leading to
irreversible damage of the myocardium.
Modulation of the inammatory response
The use of CPB is associated with the induction of a systemic inammatory response, which may
result in damage and dysfunction, not only of the heart but also of all other organ systems. Over the
years, several strategies have been proposed that aim to modulate and minimize the deleterious effects
of this inammatory response. These include pharmacological interventions, better design of the
components of the extracorporeal circuit, use of leukocyte lters, the use of miniaturized CPB circuits,
etc [6e9].
The use of corticosteroids in this respect has been a point of heavy debate during the last decades
[10]. In a recent large, randomized, placebo-controlled trial in 4494 cardiac surgery patients, Dieleman
et al. observed no effect of a single administration of 1 mg/kg dexamethasone on 30-day mortality and
organ dysfunction, although respiratory complications seemed to be lower [11]. There was also no
difference in the incidence of postoperative atrial brillation [12]. The ongoing prospective randomized
SIRS trial (Steroids In caRdiac Surgery trial; NCT00427388) involving 7507 patients evaluated the
potential benet of methylprednisolone (250 mg at anesthetic induction and 250 mg on CPB) versus
placebo. Preliminary results seem to indicate that methylprednisolone did not improve 30-day
morbidity and mortality but was associated with an increased incidence of myocardial infarction
(13. vs. 11.2%, p 0.001) (http://www.Cardiosource.org/Science-And-Quality/Clinical-Trials/S/SIRS.
aspx).
Modulation of myocardial oxygen consumption
The traditional approach for minimizing myocardial dysfunction after CPB is directed towards
modulation of the factors affecting myocardial energy consumption. Determinants of myocardial ox-
ygen consumption have already been identied many years ago [13] and are summarized in Table 2.
Starting from this perspective, it is logical that the rst step in the protection against myocardial
ischemia during CPB is by decreasing the three main determinants of myocellular oxygen utilization:
Table 1
Strategies for preparing the myocardium for cardiopulmonary bypass.
Hemodynamic preparation
Prevent tachycardia
Prevent hypotension
Prevent hypertension
Metabolic preparation
Hydration
Glucose homeostasis
Substrate administration

Pharmacological preparation
Statins
b-blockers
Calcium channel blockers
Lidoazine

heart rate, contractility, and intramyocardial tension development. An empty heart at rest (i.e., non-
brillating) not only greatly facilitates the surgical procedure but also reduces myocardial oxygen
consumption to the basal state, which reects the energy necessary for basal cellular processes. It was
shown that normothermic arrest decreased myocardial oxygen demand by 90% to 1 mL O2.
100 g1 min1 and that additional cooling to 22 C further reduced myocardial oxygen consumption to
0.3 mL O2. 100 g1 min1 [14]. These goals could be obtained by the administration of all kinds of cold
cardioplegic solutions. The rst decades of research in myocardial protection during CPB mainly
focused on the search for the exact composition of solutions to meet the ideal requirements of car-
dioplegic solutions: induce metabolic arrest, and maintain a favorable metabolic environment (pre-
vention of interstitial and intracellular edema, prevention of loss of cellular metabolites, maintenance
of an appropriate acidebase balance, and providing metabolic substrate, mainly oxygen and glucose
but also other compounds) without inducing cellular toxicity. Numerous studies have been published
comparing different compositions of cardioplegic solutions including the addition of different meta-
bolic substrates, scavenging agents, mannitol, etc [15e18]. Also the use of blood cardioplegia has been
evaluated in an attempt to further reduce post-ischemic damage after CPB. Finally, different admin-
istration techniques (single vs. intermittent vs. continuous; antegrade, retrograde, or combined) have
been evaluated [15e20].
Despite all these advances, post-CPB ischemic myocardial dysfunction continues to occur, thereby
underscoring the fact that this entity represents a complex phenomenon that cannot simply be
addressed by modulation of myocardial metabolism and by adapting the composition of the car-
dioplegic solution.
Table 2
Determinants of myocardial oxygen consumption.
Intramyocardial tension
Ventricular pressure
Ventricular volume
Myocardial mass
Heart rate
Contractile state
Forceevelocity relation
Maximal velocity of contraction
Basal metabolism
Energy associated with shortening against a load
External work (load and ber shortening)
Internal work (series elastic component shortening)
Conditioning the myocardium
Basic principles
During the past decades, it has become increasingly obvious that important cellular damage may
occur upon reperfusion after a period of myocardial ischemia. Interestingly, the extent of ische-
miaereperfusion injury can be modulated by an intrinsic defense mechanism which is conditioning of
the organ. In 1986, Murry et al. observed that in an in vivo dog model, four 5-min occlusions of the
circumex coronary artery, each separated by 5 min of reperfusion, followed by a sustained 40-min
occlusion was associated with a 25% lower infarct size of that seen in a control group that received a
single 40-min occlusion [21]. This phenomenon has been termed ischemic preconditioning. This
study initiated extensive research e both experimentally and clinically e with regard to underlying
mechanisms and potential clinical relevance.
It was only in 2003 that Zhao et al. introduced the concept of ischemic postconditioning. In an
open-chest dog model, three cycles of 30-s reocclusion were applied at the start of reperfusion after a
60-min occlusion of the left anterior descending artery [22]. Compared to a control group without
reocclusion, a similar reduction in infarct size was observed as with an ischemic preconditioning pro-
tocol. This protective mechanism has since then been widely studied in order to determine both the
pathways involved and its potential clinical applications.
This concept of endogenous cardioprotection was further rened by the intriguing observation that
similar degrees of cardioprotection could be achieved by applying brief episodes of nonlethal ischemia
and reperfusion to an organ or tissue remote from the heart. Skeletal muscle has been investigated as
the most appropriate remote site for generation of such cardioprotection. In experimental studies, the
application of an ischemic stimulus to skeletal muscle both before (remote ischemic preconditioning)
and immediately after (remote ischemic postconditioning) myocardial ischemia has been shown to
induce cardioprotective effects [23,24].
The underlying mechanisms involved in ischemic preconditioning and postconditioning have been
the subject of intensive research in the last few years. It is beyond the scope of this article to discuss
these mechanisms. The interested reader is referred to some excellent review articles on the topic
[25e36]. The main signaling pathways involved in the protective mechanisms of ischemic pre- and
postconditioning are summarized in Figs. 3 and 4.
Clinical application of conditioning in cardiac surgery
Ischemic conditioning. The concept of ischemic conditioning has been applied in the clinical environ-
ment, but its use in the cardiac surgical setting has been limited to a few studies. Although both
ischemic preconditioning [37e39] and postconditioning protocols [40] have been associated with less
myocardial damage and better outcome after surgery, they have not been introduced in daily clinical
practice. This is probably related to the fact that no clinician is tempted to put an additional ischemic
burden to an organ that specically needs treatment for ischemia. Of note, one study has reported that
CPB itself may trigger activation of one of the pathways involved in the protective mechanisms of
ischemic preconditioning [41]. However, this observation also did not translate into a clinically relevant
strategy.
More than direct ischemic conditioning, the concept of remote ischemic conditioning may constitute
an attractive protective strategy to reduce the extent of ischemiaereperfusion injury in the setting of
cardiac surgery. Several studies have shown that such protocol was associated with a decreased
postoperative troponin T release [42,43], although not all studies seem to show an unequivocal pro-
tective effect [44]. The concept however remains interesting and clinical trials are ongoing that aim to
clearly identify the effects of remote pre- and postconditioning protocols on postoperative outcome
after cardiac surgery [45].
Pharmacological conditioning. Although several studies have indicated that ischemic conditioning may
confer clinically relevant cardioprotective effects, its widespread application is hampered by the simple
fact that it may be hazardous to apply additional ischemic insults to an already jeopardized organ.
Fig. 3. Central in the cardioprotection induced by pre- and postconditioning stimuli is the prevention of the opening of the
mitochondrial permeability transition pore (MPTP). These effects can be divided in an indirect and a direct inhibition of the MPTP
opening. The indirect inhibition relates to a protective modulation by pre- and postconditioning of those factors that are responsible
for MPTP opening upon reperfusion such as attenuating the amount and effects of reactive oxygen species (ROS), the preservation of
intracellular ATP levels, the delay of abrupt pH correction at reperfusion, and the reduction of mitochondrial and cytosolic calcium
loading. The direct inhibition of MPTP opening occurs as a consequence of those mechanisms that depend on a signal transduction
pathway beginning at the plasma membrane of the cardiomyocyte (summarized in Fig. 4). ROS has a dual role in ische-
miaereperfusion injury: a benecial one as a critical mediator of the cardioprotective effects by pre- and postconditioning and a
detrimental role in the occurrence of lethal myocardial injury at the time of reperfusion. Both pre- and postconditioning are capable
of attenuating detrimental ROS activity at reperfusion, though the underlying mechanisms remain to be established. Conditioning
has been shown to reduce the ATP consumption during ischemia, thereby preventing ATP depletion at the time of reperfusion,
which may attenuate the MPTP opening at reperfusion. The intracellular acidic conditions generated during a sustained episode of
myocardial ischemia maintain the MPTP in a closed state and prevent cardiomyocyte hypercontracture. On myocardial reperfusion,
the intracellular pH is rapidly corrected to normal due to the washout of lactate and the actions of the Na/H exchanger and the
Na/HCOe 3 symporter. The rapid restoration of intracellular pH results in opening of the MPTP and hypercontracture, leading to
myocardial cell death. It has been reported that activation of components of the RISK (reperfusion injury salvage kinase) pathway
may inuence the Na/H exchanger, the Na/HCO
3 symporter, and the Na /Ca
2
exchanger, resulting in a prolonged acidosis and
less cellular calcium intake. Components of the RISK pathway may also inuence intracellular calcium regulation by promoting the
uptake of calcium via sarcocoplasmatic Ca2 ATPase into the sarcoplasmatic reticulum (SR).
The alternative is to mimic the effects of ischemic conditioning with the use of pharmacological
agents that modulate one or more of the different steps involved in ischemic pre- and post-
conditioning. Over the years, different pharmacological compounds have been tested that may either
inhibit or activate different steps in the pathways involved in ischemic conditioning, such as adenosine
Fig. 4. Proven and putative signaling pathways involved in the cellular protective actions of pre- and postconditioning stimuli.
Adenosine, bradykinin, and opioid peptides are known to play an essential role via G-protein-coupled receptors. Additional putative
triggers include natriuretic peptides (ANP and BNP), peptide growth factors (IGF-1 and IGF-2), and TNF-a. Bindings of these different
substances to the respective receptors trigger the activation of multiple intracellular signaling pathways. The exact nature of these
intracellular signaling pathways and their possible interactions and connections are still not denitively elucidated and remain an
area of intensive research. A rst pathway that was identied was based on an activation of PKC via activation of the phospholipases
C and D (PLC and PLD). The PKC activation was then thought to open both the sarcolemmal and the mitochondrial KATP channel,
resulting in a cytoprotective effect against IR injury. In the meantime, additional pathways have been explored. The RISK (reper-
fusion injury salvage kinase) pathway involves the pharmacological activation of pro-survival kinases such as phosphatidylinositol
3-kinase (PI3K), the mitogen-activated protein kinase MEK1/2, the serine/threonineeprotein kinase family (Akt), and the extra-
cellular signal-regulated kinases 1 and 2 (ERK1/2). These kinases phosphorylate downstream targets such as glycogen synthase 3b
(GSK3b), endothelial nitric oxide synthase (eNOS), and also confer protection by inactivating the Bcl-2-associated death promoter
(BAD), thereby preventing opening of the mPTP but also by exerting a direct anti-apoptotic effect. The RISK pathway also promotes
an increased Ca2 uptake in the sarcoplasmatic reticulum (SR). The eNOS pathway may prevent mPTP opening either through the
protein kinase (PK) GePKCemitochondrial KATP signaling pathway or through the generation of NO. It is believed that the different
signal transduction pathways culminate in the opening of the mKATP channel with the generation of reactive oxygen species (ROS)
via the electron transport chain (ETC). The ROS will then on its turn initiate the activation of a number of survival kinases which will
result in a cardioprotective action. The result is then a chain of events that will protect against the consequences of ische-
miaereperfusion injury. The major components of the SAFE pathway are TNF-a, The Janus kinase signal and the STAT3 signaling
pathway. It is proposed that after translocation to the nucleus, STAT3 controls a transcriptional upregulation of factors that confer
cardioprotection such as iNOS (inducible nitric oxide synthase), COX2 (cyclo-oxygenase 2), NFkB (nuclear factor k B), and heat shock
proteins (HSP).
receptor agonists, KATP channel openers, activators of protein kinases including protein kinase C, p38
mitogen-activated protein kinases and tyrosine kinases, free radical scavengers, and others. However,
with the exception of the KATP opener nicorandil [46,47], none of these compounds has entered clinical
practice up to now, either because of the occurrence of substantial side effects or because of the lack of
measurable clinical benets. Interestingly, there is now some recent evidence suggesting that direct
modulation of the MPTP by inhibition of its opening with cyclosporine A is capable of substantially
reducing myocardial infarction size [48e50].
Anesthetic conditioning. In the laboratory setting, it has been extensively shown that volatile anesthetics
have both pre- and postconditioning effects and are capable of substantially reducing the deleterious
effects of myocardial ischemiaereperfusion injury. Numerous experimental papers have been pub-
lished further characterizing this phenomenon of anesthetic cardioprotection and its underlying
mechanisms. It became evident that ischemic and anesthetic pre- and postconditioning share many
common pathways. A detailed discussion on these mechanisms is beyond the scope of this article and
the interested reader is referred to several excellent review articles on the topic [4,51e55].
Anesthetic preconditioning. In recent years, the potential protective properties of volatile anesthetic
agents have been explored during cardiac surgery. Application of an anesthetic preconditioning pro-
tocol yielded various results with some studies showing substantial protective effects with less
myocardial damage and better preserved post-CPB myocardial function, but these effects were not
conrmed in other trials (reviewed in refs. 53 and 55). It seems therefore that the cardioprotective
properties of an anesthetic preconditioning protocol, which seem very straightforward in the exper-
imental setting, are less evident when applied in the clinical setting of cardiac surgery. Different
reasons can be invoked for this discrepancy, but recently, two studies have indicated that the pre-
conditioning protocol used might be of crucial importance in the extent of cardioprotection. Fr abdorf
et al. randomly allocated 30 coronary surgery patients to one of three groups [56]. All patients received
total intravenous anesthesia with sufentanil and propofol. The control group had no further inter-
vention. Ten minutes prior to the start of CPB, patients in the rst study group received 1 MAC of
sevourane for 5 min. Patients in the second study group received two times 5 min of sevourane
interspersed by a 5-min washout. Only in group 2, and not in group 1, postoperative troponin I levels
were signicantly lower than in the control group. The authors attributed this to the intermittent
administration protocol, although a longer administration time (two times 5 min vs. only one 5-min
run) might have been a plausible alternative explanation. This later issue was solved by the ndings
of Bein et al. [57] In this study, 42 coronary surgery patients were randomly assigned to one of three
following groups: propofol only, sevourane continuously at 1 MAC from after induction to initiation of
CPB, and sevourane intermittently, with an interruption of the sevourane administration for 10 min
between both sevourane (1 MAC) runs. Interestingly, a protective effect as evidenced by a lower
postoperative troponin T release and a better myocardial performance index was only observed with
the intermittent administration. These data indicate that perhaps not the total amount of sevourane
but instead the intermittent administration protocol seems to be essential to yield a cardioprotective
effect.
Such cardioprotective effects of administration of drugs before the actual ischemia/reperfusion
injury may not be conned to the volatile anesthetics. It has been shown in coronary surgery
patients, that the administration of 40 mg morphine before CPB (as part of a standardized
opioideisourane anesthesia) was associated with a better postoperative myocardial performance
index and a lower release of inammatory markers than with fentanyl 1000 mg. Postoperative
levels of troponin I and brain natriuretic peptide however did not differ between both treatments
[58,59].
Anesthetic drugs during CPB. Nader et al. applied a cardioplegia mixture that was vaporized with
2% sevourane and oxygen in coronary surgery patients and observed a decreased inammatory
response after CPB together with a better left ventricular function and lower postoperative troponin I
levels [60,61]. Recently, Kortekaas et al. observed that in mitral valve repair patients selective
myocardial sevourane administration during aortic cross-clamping more strongly attenuated the
inammatory response compared to systemic administration, however, without reducing post-
operative markers of myocardial cell damage [62]. Similarly, aortic root infusion of sufentanil 5 min
before aortic unclamping attenuated myocardial ischemiaereperfusion injury in mitral valve surgery
patients [63].
Of note, Xia et al. reported that a high dose of propofol on CPB was also associated with a car-
dioprotective effect. Fifty-four coronary artery surgery patients were randomly assigned to receive
either a small dose of propofol (60 mg kg1 min1 throughout), a large-dose of propofol (the dose of
propofol was increased to 120 mg kg1 min1 for 10 min before the onset of CPB until 15 min after aortic
unclamping and then decreased to 60 mg kg1 min1 until the end of surgery), or isourane (1e3.5%
end-tidal concentration). Troponin I was lower at 24 h after CPB and mean cardiac index was higher in
the high-dose propofol group than in the other two groups [64]. The potential benecial effects of
addition of propofol when used during aortic cross-clamping is subject of ongoing research and further
data will help to identify the exact role of this substance in modulating the extent of ische-
miaereperfusion injury after CPB [65].
Anesthetic postconditioning. Clinical reports on the potential cardioprotective effects of volatile
anesthetic agents when administered only during the reperfusion period are scarce. In one study, in
coronary artery surgery patients De Hert et al. observed that the administration of sevourane (0.5e1
MAC) at early reperfusion until the end of the operation did not signicantly reduce postoperative
troponin I release. However, recovery of ventricular function after CPB occurred earlier in the sevo-
urane postconditioning group compared to the control group [66]. Recently, Huang et al. observed
that a combination of isourane preconditioning and propofol postconditioning decreased post-
operative creatine kinase MB isoenzyme (CKMB) and troponin I release and facilitated postoperative
myocardial functional recovery compared to the control group anesthetized with fentanyl and mid-
azolam [67].
Clinical cardioprotection: does the anesthetic agent make a difference?. The absence of an unequivocal
and reproducible clinical protective effect on the different variables in the clinical preconditioning and
postconditioning protocols has initiated the question whether the choice of the anesthetic regimen is
of any importance to postoperative cardiac function and outcome.
This issue was rst addressed by De Hert et al. who compared the effects of sevourane and pro-
pofol on myocardial function during and after coronary artery surgery [68]. Before CPB, all hemody-
namic variables were comparable between the two anesthetic treatment groups. However, after CPB,
patients who received the volatile anesthetic regimen for anesthesia had preserved cardiac perfor-
mance, which was evident from a preserved stroke volume and dP/dtmax, and the preservation of the
length-dependent regulation of myocardial function. In addition, the need for inotropic support in the
early postoperative period was signicantly less with the volatile anesthetic, and postoperative plasma
concentrations of cardiac troponin I were consistently lower when compared with patients who
received the total intravenous anesthetic regimen. These data therefore suggested that volatile anes-
thetics provided a cardioprotective effect that was not observed with the intravenous anesthetic
regimen. Such cardioprotective effects of volatile anesthetic agents against the consequences of
ischemia/reperfusion injury with cardiac surgery have subsequently been reported in other studies
(reviewed in refs. 53 and 55)
Although the majority of the clinical observations indicate that a volatile anesthetic regimen
throughout the entire procedure confers myocardial protection during coronary surgery, the impact of
this phenomenon on postoperative morbidity and clinical recovery remains to be established. In a meta-
analysis of 22 randomized controlled trials including 1922 patients, Landoni et al. found a reduction of in-
hospital postoperative myocardial infarctions (volatile anesthetics 2.4% vs. intravenous 5.1%, odds ratio
(OR) 0.51; condence interval (CI) [0.32e0.84]; p 0.008) and a reduced in-hospital mortality (volatile
anesthetics 0.4% vs. intravenous anesthetic 1.6%, OR 0.31; CI [0.12e0.80], p 0.02) [69].
Data on longer-term outcome after cardiac surgery and possible inuence of the choice of the
anesthetic regimen are scarce. In a follow-up study of a randomized trial in coronary surgery patients
comparing a sevourane preconditioning protocol to a total intravenous anesthetic regimen [70],
Garcia et al. observed a signicantly lower incidence of new cardiovascular events (six vs. one;
p 0.038) [71]. The authors hypothetized that sevourane preconditioning may directly modify the
plateleteneutrophileendothelial interaction by altering the synthesis of various protective and anti-
protective proteins and thereby prevent the development of coronary endothelial dysfunction. A recent
multicenter study observed a lower 1-year mortality in coronary surgery patients treated with a
volatile anesthetic regimen compared to those treated with an intravenous anesthetic regimen [72].
However, not all studies come to the same conclusion [73e75], and it seems that the issue of post-CPB
ischemiaereperfusion injury is a very complex phenomenon to be targeted by one single intervention,
such as for instance choice of the anesthetic regimen.
Summary
During cardiac surgery with CPB, the heart is isolated from the circulation, and as a consequence
suffers from an ischemic insult followed by an additional hit upon reperfusion, the so-called ische-
miaereperfusion injury.
Key approaches for cardioprotection during CPB include (1) metabolic preparation of the myocar-
dium for CPB, (2) modulation of the inammatory response to CPB, (3) modulation of myocardial
oxygen consumption during CPB, and (4) conditioning of the myocardium.
Conditioning of the myocardium in the setting of cardiac surgery can be obtained by either direct or
remote ischemic preconditioning, pharmacological conditioning, and anesthetic conditioning.
Despite the fact that CPB inevitably induces a huge hit to the normal homeostasis of the body, a vast
majority of cardiac surgery patients survive surgery without evidence of serious harm. This indicates
that contemporary strategies for CPB are sufciently well developed to provide safe care. This is
probably a reason why attempts to implement specic therapeutic strategies targeting only one small
aspect of CPB do not seem to result in a consistent signicant clinical benet.
The time has come to redirect the search for CPB-related cardioprotective strategies from studies in
large patient populations, from which most will have an uneventful perioperative course, to targeting
those patients who specically represent a high risk of CPB-related complications. Tailored application
of specic therapeutic strategies may then result in less myocardial ischemiaereperfusion injury after
CPB with enhanced recovery. The main challenge now is to link increased risk to specic tailored
therapy.
Disclosures
SDH has received speaker's fees from Abbott/Abbvie and Baxter.

AM has no disclosures to declare.
Practice points
Key approaches for cardioprotection during CPB
* Metabolic preparation of the myocardium

* Modulation of the inflammatory response
* Modulation of myocardial oxygen consumption
* Conditioning of the myocardium
Research agenda
* Contemporary strategies for CPB seem sufficiently well developed to provide safe care for
the majority of patients.
* To date, no single new therapeutic strategy has resulted in a consistent significant better
clinical outcome in the average cardiac patient.
* The time has come to redirect the research from studies in large patient populations, from
which most will have an uneventful perioperative course, to targeting those patients who
specifically represent a high risk of CPB-related complications.
* Tailored application of specific therapeutic strategies described above may then result in less
myocardial ischemiaereperfusion injury after CPB with enhanced recovery.
* The main challenge now is to link increased risk to specific tailored therapy.
Acknowledgments
None.
References
[1] Forman MB, Puett DW, Virmani R. Endothelial and myocardial injury during ischemia and reperfusion: pathogenesis and
therapeutic implications. J Am Coll Cardiol 1989;13:450e9.
[2] De Hert SG. Myocardial protection from ischemia and reperfusion injury. In: Mebazaa A, Gheorghiade M, Zannad FM,
et al., editors. Acute heart failure. 1st ed. London: Springer-Verlag; 2008. p. 70e86.
*[3] Collard CD, Gelman S. Pathophysiology, clinical manifestations, and prevention of ischemia-reperfusion injury. Anes-
thesiology 2001;94:1133e8.
*[4] Zaugg M, Lucchinetti E, Uecker M, et al. Anaesthetics and cardiac preconditioning. Part I. Signalling and cytoprotective
mechanisms. Br J Anaesth 2003;91:551e65.
[5] Honda HM, Korge P, Weiss JN. Mitochondria and ischemia/reperfusion injury. Ann N Y Acad Sci 2005;1047:248e58.
[6] Laffey JG, Boylan JF, Cheng DCH. The systemic inammatory response to cardiac surgery. Anesthesiology 2002;97:215e52.
*[7] Warren OJ, Smith AJ, Alexiou C, et al. The inammatory response to cardiopulmonary bypass: part 1 e mechanisms of
pathogenesis. J Cardiothorac Vasc Anesth 2009;23:223e31.
*[8] Warren OJ, Watret AL, de Wit KL, et al. The inammatory response to cardiopulmonary bypass: part 2 e anti-
inammatory therapeutic strategies. J Cardiothorac Vasc Anesth 2009;23:384e93.
[9] Ng CSH, Wan S. Limiting inammatory response to cardiopulmonary bypass: pharmaceutical strategies. Curr Opin
Pharmacol 2012;12:155e9.
[10] Augoustides JG. The inammatory response to cardiac surgery with cardiopulmonary bypass: should steroid prophylaxis
be routine? J Cardiothorac Vasc Anesth 2012;26:952e8.
[11] Dieleman JM, Nierich AP, Rosseel PM, et al. Intraoperative high-dose dexamethasone for cardiac surgery: a randomized
controlled trial. JAMA 2012;308:1761e7.
[12] van Osch D, Dieleman JM, van Dijk D, et al. Dexamethasone for the prevention of postoperative atrial brillation. Int J
Cardiol 2015;182:431e7.
[13] Sonnenblick EH, Ross Jr J, Braunwald E. Oxygen consumption of the heart. Newer concepts of its multifactorial deter-
mination. Am J Cardiol 1968;22:328e36.
[14] Buckberg GD, Brazier JR, Nelson RL, et al. Studies of the effects of hypothermia on regional myocardial blood ow and
metabolism during cardiopulmonary bypass. I. The adequately perfused beating, brillating, and arrested heart. J Thorac
[15] Cordell AR. Milestones in the development of cardioplegia. Ann Thorac Surg 1995;60:793e6.
[16] Buckberg GD. Update on current techniques on myocardial protection. Ann Thorac Surg 1995;60:805e14.
[17] Mauney MC, Kron IL. The physiologic basis of warm cardioplegia. Ann Thorac Surg 1995;60:819e23.
[18] Cleveland Jr JC, Meldrum DR, Rowland RT, et al. Optimal myocardial preservation: cooling, cardioplegia, and conditioning.
Ann Thorac Surg 1996;61:760e8.
[19] Guru V, Omura J, Alghamdi AA, et al. Is blood superior to crystalloid cardioplegia? A meta-analysis of randomized clinical
trials. Circulation 2006;114:I331e8.
[20] Jacob S, Kallikourdis A, Sellke F, et al. Is blood cardioplegia superior to crystalloid cardioplegia? Interact Cardiovasc Thorac
Surg 2008;7:491e9.
[21] Murry CE, Jennings RB, Reimer KA. Preconditioning with ischemia: a delay of lethal cell injury in ischemic myocardium.
Circulation 1986;74:1124e36.
[22] Zhao Z-Q, Corvera JS, Halkos ME, et al. Inhibition of myocardial injury by ischemic postconditioning during reperfusion:
comparison with ischemic preconditioning. Am J Physiol Heart Circ Physiol 2003;285:H579e88.
[23] Birnbaum Y, Hale SL, Kloner RA. Ischemic preconditioning at a distance: reduction of myocardial infarct size by partial
reduction of blood supply combined with rapid stimulation of the gastrocnemius muscle in the rabbit. Circulation 1997;
96:1641e6.
[24] Andreka G, Vertesaljai M, Szantho G, et al. Remote ischaemic postconditioning protects the heart during acute myocardial
infarction in pigs. Heart 2007;93:749e52.
[25] Cohen MV, Baines CP, Downey JM. Ischemic preconditioning: from adenosine receptor to KATP channel. Annu Rev Physiol
2000;62:79e109.
[26] Hausenloy DJ, Yellon DM. Survival kinases in ischemic preconditioning and postconditioning. Cardiovasc Res 2006;70:
240e53.
[27] Hausenloy DJ, Yellon DM. Reperfusion injury salvage kinase signaling: taking a RISK for cardioprotection. Heart Fail Rev
2007;12:217e34.
*[28] Hausenloy DJ, Yellon DM. Preconditioning and postconditioning: united at reperfusion. Pharmacol Ther 2007;116:
173e91.
[29] Ferdinandy P, Schulz R, Baxter GF. Interaction of cardiovascular risk factors with myocardial ischemia/reperfusion injury,
preconditioning, and postconditioning. Pharmacol Rev 2007;59:418e58.
*[30] Hausenloy DJ, Ong S-B, Yellon DM. The mitochondrial permeability transition pore as a target for preconditioning and
postconditioning. Basic Res Cardiol 2009;104:189e202.
*[31] Hausenloy DJ, Yellon DM. Preconditioning and postconditioning: underlying mechanisms and clinical application.
Atherosclerosis 2009;204:334e41.
[32] Hausenloy DJ, Yellon DM. The second window of preconditioning (SWOP): where are we now? Cardiovasc Drug Ther
2010;24:235e54.
[33] Ovize M, Baxter GF, Di Lisa F, et al. Postconditioning and protection from reperfusion injury: where do we stand? Car-
diovasc Res 2010;87:406e23.
*[34] Lavi S, Lavi R. Conditioning of the heart: from pharmacological interventions to local and remote protection. Possible
implications for clinical practice. Int J Cardiol 2011;146:311e8.
[35] Kharbanda RK, Nielsen TT, Redington AN. Translation of remote ischaemic preconditioning into clinical practice. Lancet
2009;374:1557e65.
[36] Pac-Soo CK, Mathew H, Ma D. Ischaemic conditioning strategies reduce ischaemia/reperfusion-induced organ injury. Br J
Anaesth 2015;114:204e16.
[37] Yellon DM, Alkhulai AM, Pugsley WB. Preconditioning the human myocardium. Lancet 1993;342:276e7.
[38] Jenkins DP, Pugsley WB, Alkhulai AM. Ischaemic preconditioning reduces troponin T release in patients undergoing
coronary artery bypass surgery. Heart 1997;77:314e8.
[39] Wu ZK, Iivainen T, Pehkonen E, et al. Ischemic preconditioning suppresses ventricular tachyarrhythmias after myocardial
revascularization. Circulation 2002;106:3091e6.
[40] Durdu S, Sirlak M, Cetintas D, et al. The efcacies of modied mechanical post conditioning on myocardial protection for
patients undergoing coronary artery bypass grafting. J Cardiothorac Surg 2012;7:73.
[41] Pouzet B, Lecharny JB, Dehoux M, et al. Is there a place for preconditioning during cardiac operations in humans? Ann
Thorac Surg 2002;73:843e8.
[42] Hausenloy DJ, Mwamure PK, Venugopal V, et al. Effect of remote ischaemic preconditioning on myocardial injury
in patients undergoing coronary artery bypass graft surgery: a randomized controlled trial. Lancet 2007;370:
575e9.
[43] Venugopal V, Hausenloy DJ, Ludman A, et al. Remote ischaemic preconditioning reduces myocardial injury in patients
undergoing cardiac surgery with cold-blood cardioplegia: a randomized controlled trial. Heart 2009;95:1567e71.
[44] Rahman IA, Mascaro JG, Steeds RP, et al. Remote ischemic preconditioning in human coronary artery bypass surgery:
from promise to disappointment? Circulation 2010;122:S53e9.
[45] Brevoord D, Hollmann MW, De Hert SG, et al. Effect of remote ischemic conditioning on atrial brillation and outcome
after coronary artery bypass grafting (RICO-trial). BMC Anesthesiol 2011;11:11.
[46] Kaneko T, Saito Y, Hikawa Y, et al. Dose-dependent prophylactic effect of nicorandil, an ATP sensitive potassium channel
opener, on intra-operative myocardial ischaemia in patients undergoing major abdominal surgery. Br J Anaesth 2001;86:
332e7.
[47] Chinnan NK, Puri GD, Thingnam SK. Myocardial protection by nicorandil during open heart surgery under cardiopul-
monary bypass. Eur J Anaesth 2007;24:26e32.
[48] Piot C, Croisille P, Staat P, et al. Effect of cyclosporine on reperfusion injury in acute myocardial infarction. N Engl J Med
2008;359:473e81.
[49] Mewton N, Croisille P, Gahide G, et al. Effect of cyclosporine on left ventricular remodeling after reperfused myocardial
infarction. J Am Coll Cardiol 2010;55:1200e5.
[50] Hausenloy DJ, Kunst G, Boston-Grifths E, et al. The effect of cyclosporine-A on perioperative myocardial injury in adult
patients undergoing coronary artery bypass graft surgery: a randomized controlled clinical trial. Heart 2014;100:544e9.
[51] Zaugg M, Lucchinetti E, Garcia C, et al. Part 2. Clinical implications. Br J Anaesth 2003;91:556e76.
[52] Bienengraeber MW, Weihrauch D, Kersten JR, et al. Cardioprotection by volatile anesthetics. Vasc Pharmacol 2005;42:
243e52.
[53] De Hert SG, Turani F, Mathur S, Stowe DF. Cardioprotection with volatile anesthetics: mechanisms and clinical impli-
cations. Anesth Analg 2005;100:1584e93.
[54] Pagel PS. Postconditioning by volatile anesthetics: salvaging ischemic myocardium at reperfusion by activation of pro-
survival signaling. J Cardiothorac Vasc Anesth 2008;22:753e65.
abdorf J, De Hert S, Schlack W. Anaesthesia and myocardial ischaemia/reperfusion injury. Br J Anaesth 2009;103:89e98.
*[55] Fr
bdorf J, Borowski A, Edel D, et al. Impact of preconditioning protocol on anesthetic-induced cardioprotection in pa-
[56] Fra
tients having coronary artery bypass surgery. J Thorac Cardiovasc Surg 2009;137:1436e42.
[57] Bein B, Renner J, Caliebe D, et al. The effects of interrupted or continuous administration of sevourane on pre-
conditioning before cardiopulmonary bypass in coronary artery surgery: comparison with continuous propofol. Anaes-
thesia 2008;63:1046e55.
[58] Murphy GS, Szokol JW, Marymont JH, et al. Opioids and cardioprotection: the impact of morphine and fentanyl on re-
covery of ventricular function after cardiopulmonary bypass. J Cardiothorac Vasc Anesth 2006;20:493e502.
[59] Murphy GS, Szokol JW, Marymont JH, et al. The effects of morphine and fentanyl on the inammatory response to
cardiopulmonary bypass in patients undergoing elective coronary artery bypass graft surgery. Anesth Analg 2007;104:
1334e42.
[60] Nader ND, Li CM, Khadra WZ, et al. Anesthetic myocardial protection with sevourane. J Cardiothorac Vasc Anesth 2004;
18:269e74.
[61] Nader ND, Karamanoukian HL, Reedy RL, et al. Inclusion of sevourane in cardioplegia reduces neutrophil activity during
cardiopulmonary bypass. J Cardiothorac Vasc Anesth 2006;20:57e62.
[62] Kortekaas KA, van der Baan A, Aarts LPHJ, et al. Cardiospecic sevourane treatment quenches inammation but does not
attenuate myocardial cell damage markers: a proof-of-concept study in patients undergoing mitral valve repair. Br J
Anaesth 2014;112:1005e14.
[63] Zuo Y, Cheng X, Gu E, et al. Effect of aortic root infusion of sufentanil on ischemia-reperfusion injury in patients un-
dergoing mitral valve replacement. J Cardiothorac Vasc Anesth 2014;28:1474e8.
[64] Xia Z, Huang Z, Ansley DM. Large-dose propofol during cardiopulmonary bypass decreases biochemical markers of
myocardial injury in coronary surgery patients: a comparison with isourane. Anesth Analg 2006;103:527e32.
[65] Plummer ZE, Baos S, Rogers CA, et al. The effects of propofol cardioplegia on blood and myocardial biomarkers of stress
and injury in patients with isolated coronary artery bypass grafting or aortic valve replacement using cardiopulmonary
bypass: protocol for a single-center randomized controlled trial. JMIR Res Protoc 2014;3:e35.
*[66] De Hert SG, Van der Linden PJ, Cromheecke S, et al. Cardioprotective properties of sevourane in patients undergoing
coronary surgery with cardiopulmonary bypass are related to the modalities of its administration. Anesthesiology 2004;
101:299e310.
[67] Huang Z, Zhong X, Irwin MG, et al. Synergy of isourane preconditioning and propofol postconditioning reduces
myocardial reperfusion injury in patients. Clin Sci 2011;121:57e69.
[68] De Hert SG, ten Broecke PW, Mertens E, et al. Sevourane but not propofol preserves myocardial function in coronary
surgery patients. Anesthesiology 2002;97:42e9.
[69] Landoni G, Biondi-Zoccai GG, Zangrilla A, et al. Desurane and sevourane in cardiac surgery: a meta-analysis of ran-
domized clinical trials. J Cardiothorac Vasc Anesth 2007;21:502e11.
[70] Julier K, da Silva R, Garcia C, et al. Preconditioning by sevourane decreases biochemical markers for myocardial and renal
dysfunction in coronary artery bypass graft surgery: a double-blinded, placebo-controlled, multicenter study. Anesthe-
siology 2003;98:1315e27.
[71] Garcia C, Julier K, Bestmann L, et al. Preconditioning with sevourane decreases PECAM-1 expression and improves one-
year cardiovascular outcome in coronary artery bypass graft surgery. Br J Anaesth 2005;94:159e65.
[72] De Hert S, Vlasselaers D, Barbe R, et al. A comparison of volatile and non volatile agents for cardioprotection during on-
pump coronary surgery. Anaesthesia 2009;64:953e60.
[73] De Hert SG. Is anaesthetic cardioprotection clinically relevant ? Another futile search for a magic bullet. Eur J Anaesth
2011;28:616e7.
[74] Bein B. Clinical application of the cardioprotective effects of volatile anaesthetics -PRO: get an extra benet from a proven
anaesthetic free of charge. Eur J Anaesth 2011;28:620e2.
[75] Van Rompaey N, Barvais L. Clinical application of the cardioprotective effects ofvolatile anaesthetics e CON: TIVA or not
TIVA to anaesthetise a cardiac patient? Eur J Anaesth 2011;28:623e7.

Anaesthesiology
Hepatic and renal effects of cardiopulmonary

bypass
Nora Di Tomasso, MD, Consultant a, 1,
Fabrizio Monaco, MD, Consultant a, 1,
Giovanni Landoni, MD, Consultant & Associate Professor a, b, *
a
IRCCS San Raffaele Scientic Institute, Milan, Italy
b
Vita-Salute San Raffaele University, Milan, Italy
Keywords:
Although associated with low morbidity and mortality, cardio-
pulmonary bypass remains a non-physiologic device that carries
hepatic dysfunction
shock liver a set of complications. Hepatic and renal impairment are associ-
renal injury ated with a poor outcome. The knowledge of pathophysiology, risk
anaesthesia factors and therapeutic interventions can help the anaesthesiolo-
cardiac anaesthesia gist in preventing these complications in daily practice. The pre-
intensive care sent narrative review provides an update of the literature on the
acute kidney injury effects of cardiopulmonary bypass on hepatic and renal functions,
focussing on markers of hepatic and renal injuries, perioperative
strategies in preserving organ function and replacement therapies.
The haemodynamic, inammatory and organic responses to cardiac surgery with cardiopulmonary
bypass (CPB) are well known, and they can lead to multiorgan dysfunction. Although multiorgan
dysfunction after CPB is generally subclinical in nature due to the physiologic reserve and resilience of
the liver and kidneys, cardiac surgery requiring CPB is being carried out for an extended patient
population who are older and undergoing complex surgical procedures [1], and thus it places them at
an increased risk of hepatic and renal impairment.
* Corresponding author. San Raffaele Scientic Institute, Via Olgettina 60, 20132 Milano, Italy. Tel.: 39 02 26436151; Fax:
39 02 26436152.
E-mail addresses: ditomasso.nora@hsr.it (N. Di Tomasso), monaco.fabrizio@hsr.it (F. Monaco), landoni.giovanni@hsr.it
(G. Landoni).
1
San Raffaele Scientic Institute, Via Olgettina 60, Milano 20132, Italy. Tel.: 39 02 26436151; Fax: 39 02 26436152.
URL: http://www.unisr.it/persona.asp?id=8713&linguacv=english
152 N. Di Tomasso et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 151e161
Hepatic effects of CPB
The hepatic integrity is affected during cardiac surgery in particular when CPB is adopted [2e4],
even in uncomplicated elective surgery [5].
Microembolism, free radicals generation, inadequate tissue perfusion [6], dilutional anaemia [7]
and haemodynamic changes play a major role in the development of hepatic dysfunction. Hyper-
bilirubinaemia and transient elevation of hepatic enzymes are commonly observed after cardiac sur-
gery with CPB, but in the vast majority of patients, the hepatocellular function recovers a few days after
surgery without developing perioperative clinically relevant hepatic failure.
In high-risk patients with a reduced physiological reserve, a severe liver dysfunction after CPB can
occur. Pre-operative risk factors are right-side heart failure, moderate-to-severe tricuspid regurgita-
tion, pulmonary hypertension (systolic pulmonary pressure above 45 mm Hg), high preload (central
venous pressure (CVP) above 8 mm Hg) chronic heart failure, New York Heart Association (NYHA) class
IIeIV and low ejection fraction [8] are at a higher risk to develop liver dysfunction after CPB [9].
Interestingly, as reported by Van Deursen et al. in 323 patients with heart failure, the haemodynamic
prole can affect the liver function tests (LFTs). In fact, elevated LFTs mainly indicate a higher CVP,
whereas only the presence of elevated aspartate aminotransferase (AST), alanine aminotransferase
(ALT) or direct bilirubinaemia may indicate a low cardiac index (CI) [10].
Once established, liver dysfunction is associated with high mortality, morbidity [11] and costs. An
early detection and a prompt treatment of the causative factors can signicantly improve outcome.
However, there is no agreement on which strategy signicantly improves the perioperative hepatic
dysfunction, reducing consequentially morbidity and mortality. In particular, the literature fails to
clarify the pathophysiology of liver dysfunction related to the extracorporeal circulation (CPB and
extracorporeal membrane oxygenator (ECMO)), and to suggest specic biochemical markers or im-
aging techniques to detect the occurrence of hepatic dysfunction.
Pathophysiology and markers of liver dysfunction
In the setting of cardiac surgery, organ hypoperfusion has been reported in a percentage ranging
between 1.2% and 2.3%, with an increased risk in on-pump cases [2,11]. The incidence of hepatic failure
after cardiac surgery is as low as 0.1%, but it is associated with high mortality (74%) [11].
CPB-related liver impairment can be attributed both to a pro-inammatory syndrome, with the
release of hepatotoxic cytokines, and to the haemodynamic changes related to surgery and CPB (i.e.,
hypotension, low cardiac output syndrome (LCOS), hypoxia and right ventricular (RV) dysfunction).
These factors lead to two different clinical syndromes that can occur together or separately, named
ischaemic hepatitis (shock liver) [12,13] and passive liver congestion, respectively (Table 1) [14]. For
both, risk factors are prior history of congestive heart failure (CHF), inadequate hepatosplanchnic blood
ow and difcult weaning from CPB.
Ischaemic hepatitis can be found after a period of relatively profound hypotension and haemody-
namic instability, and it is often associated with left ventricular dysfunction. The reduction in the
hepatic blood ow leads to a consequent hypoxia/anoxia of hepatocytes histologically characterised by
the centrilobular necrosis of zone 3 hepatocytes [15]. Biochemical markers of ischaemic hepatitis are an
increase of serum AST and serum ALT 10e20 times the normal value, a rise of lactate dehydrogenase
(LDH), total bilirubin and a deciency of hepatic coagulation factors with a consequent prolongation of
prothrombin time (PT), 1e3 days after the ischaemic injury. Usually, these biochemical indices return to
normal within 5e10 days. If the hepatic biomarkers remain persistently high and other organs are
affected from the perioperative systemic hypoperfusion, multiorgan failure (MOF) can occur. MOF leads
to death in the majority of cases [16,17].
Congestive hepatopathy, also known as nutmeg liver [17], is a liver dysfunction usually associated
to right heart failure [18]. Venous congestion couples with high CVP, uid retention and blood stasis,
and it is related to the deoxygenation of hepatocytes surrounding the central venule of the liver.
Typically, the histological pattern shows hyperaemia and congestion of the hepatic lobule central zone.
As hepatocytes, biliary epithelium and bile ducts are the most sensitive to lobular congestion (Fig. 1),
cholestatic LFTs, such as total serum bilirubin, gamma-glutamyl transpeptidase (GGT), alkaline
N. Di Tomasso et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 151e161 153
Table 1
Clinical and biochemical markers of congestive hepatopathy versus ischaemic hepatitis.
Congestive hepatopathy Ischaemic hepatitis
Co-morbidity RH failure; high CVP LV dysfunction;

Hypoperfusion; ATN
Symptoms Usually none; if present: Usually none; if present:
jaundice, RUQ discomfort jaundice, RUQ discomfort
nausea, anorexia, vomiting
Hepatomegaly Yes No
Hepatojugular reux Yes No
Bilirubinaemia >3 mg/dL (unconjugated) >3 mg/dL
ALP Raised twice UNL
GGT Raised Normal
Aminotransferase 2e3 times UNL 10e20 times UNL
Serum albumin 2.5 g/dL normal
PT Mildly raised Raised
Ammonia Usually normal Usually normal
LDH Usually normal Massively raised
ALT/LDH ratio e <1.5
RH right heart; CVP central venous pressure; LV left ventricular; ATN acute tubular necrosis; RUQ right upper
quadrant; ALP alkaline phosphatase; UNL upper normal limit; GGT gamma-glutamyl transpeptidase; PT prothrombin
time; LDH lactate dehydrogenase; ALT alanine transaminase.
Fig. 1. Hepatic lobule and mechanisms of liver injury. Centrilobular zone, sensitive to ischaemia; bile ducts, sensitive to lobular
congestion.
phosphatase (ALP) and albuminaemia, are severely abnormal. Conversely, AST and ALT are normal or
only moderately altered [19]. A mix of ischaemic and congestive hepatopathy pattern can be observed
when RV dysfunction is not promptly treated. A reduced left ventricle preload, related to RV failure,
results in an LCOS and impaired perfusion of hepatosplanchnic circulation [12].
CPB is probably one of the determinants of post-operative hepatic failure, but, as demonstrated by
Ascione et al. [2], patients undergoing off-pump cardiac surgery can develop this detrimental
complication as well. Probably, haemodynamic control, rather than CPB management, plays a pivotal
role in preventing hepatic failure in cardiac surgery.
However, predisposing factors can be identied: female gender, history of CHF, complex cardiac
surgery with prolonged CPB and aortic cross-clamp time, infective endocarditis and transfusions [11].
Post-operative liver dysfunction can be detected as an increase in both liver enzymes and bilir-
ubinaemia (>3 mg/dL or 51 mmol/L) [17]. The latter, which can be observed in a setting of normal LFTs,
is an independent predictor of worse outcome after cardiac surgery accounting for 25% mortality
[20,21]. The risk factors for hyperbilirubinaemia are the same as the ones for hepatic failure [22].
Intrahepatic cholestasis rather than cell necrosis [23] seems related to post-CPB hyper-
bilirubinaemia and jaundice. Post-operative jaundice due to conjugated bilirubin has been associated
with a defect of the hepatic excretion of bilirubin [24]. This mainly occurs following gaseous micro-
emboli and debris during CPB.
On the contrary, post-operative jaundice due to unconjugated bilirubin is related to hepatocellular
damage, liver congestion, haemolysis or blood transfusion. Therefore, the increase of unconjugated
bilirubin is more frequently transient and less associated to mortality.
The biochemical markers routinely adopted in the post-operative period are tardive and affected by
low specicity. Thus, new markers have been proposed.
Sander et al. have observed that the plasma clearance of indocyanine green is an excellent method
for measuring the hepatic function and perfusion in 60 patients undergoing normothermic coronary
artery bypass grafting (CABG) [25,26]. Another precocious index of subclinical liver impairment is
monoethylglycinexylidide (MEGX)/lidocaine levels ratio [2]. MEGX is measured 15 and 30 min after
intravenous injection of 1 mg/kg bolus of lidocaine. Finally, Theodorakis et al. showed that the ultra-
sonographic evaluation of the portal vein and hepatic artery is a good predictor of post-operative
elevation of ALT [27]. However, further investigations on direct hepatic haemodynamic ultrasound
approach are needed.
Patient with pre-operative liver disease
Pre-operative liver dysfunction is a rare (0.5%) and detrimental condition in patients undergoing
cardiac surgery [28]. Two scores named ChildeTurcotteePugh (ChildePugh) [29] and model for end-
stage liver disease (MELD) have been proposed to stratify the risk of the patients with pre-operative
liver dysfunction undergoing cardiac surgery. Neither of the two has been prospectively validated in
this setting (Table 2).
The ChildePugh score considers two qualitative variables (ascites and encephalopathy), and three
laboratory tests, serum bilirubin, serum albumin and international normalised ratio (INR). The major
limitations are the subjective classication of the ascites and encephalopathy, and the ceiling effect of
the variables considered in the score [29].
The MELD score includes three standardised laboratory tests: INR, serum creatinine (sCrea) and
serum bilirubin, arranged in a mathematical formula. It has been validated to allocate organs for liver
transplantation. Although MELD score overcomes some Child score limitations, it fails to predict
perioperative mortality; thus, the American Society of Anaesthesiologists (ASA) physical status class,
the surgical severity score and cardiopulmonary co-morbidity have to be included [30].
Several studies have investigated perioperative mortality of patients who underwent cardiac sur-
gery with pre-operative liver disease [31,32]. Filsou et al. [28] reported, in a retrospective study, an
overall mortality of 26% after cardiac surgery in patients with cirrhosis, and an adjusted mortality of
Table 2
Model for end-stage liver disease (MELD) and ChildeTurcotteePugh (ChildePugh) scores to stratify the risk of patients with pre-
operative liver dysfunction undergoing cardiac surgery.
Objective variables Subjective variables Score Mortality
INR s-Bilirubin mg/dL s-Creatinine s-Alb Ascites Encephalopathy

mg/dL g/dL
MELD 11.2 3.78 9.57 loge(sCr) NO NO NO <9 2%

formula loge(INR) loge(sBil) 6.4 >30 >50%
Child 1 <1.7 points <2 NO >3.5 Absent Absent 5e6 Child-A

points 2 1.7e2.3 2e3 2.8e3.5 Mild 1e2 Treatable 7e9 0%

3 >2.3 >3 <2.8 Mod-Severe 3e4 Refractory 10e15 Child-B
20%
Child-C
55%
11%, 18% and 67% in patients classied as ChildePugh A, B and C, respectively. Other authors reported a
mortality of 100% for patients with advanced liver disease (ChildePugh C) [33]. Morimoto et al. have
not conrmed these ndings: they have observed, in a small retrospective study, that MELD rather than
Child predicts the operative mortality after cardiac surgery [34]. Similarly, Arif et al. in a population of
106 patients with cirrhosis undergoing cardiac surgery have reported a signicantly higher area under
the curve for the MELD score and Euroscore than Child in predicting a 30-day mortality [35]. In a recent
retrospective paper led on a small cohort of 55 patients undergoing cardiac surgery with mild or severe
cirrhosis, neither MELD nor Child predicts early and late mortalities [36].
According to current data and our experience, it is important not to preclude patients with
advanced cirrhosis, on the basis of the MELD and Child classication, from cardiac surgery.
The use of CPB is related to a higher rate of complications. For instance, Hayashida et al. observed a
lower incidence of complications in patients undergoing off-pump coronary artery bypass surgery
compared to those undergoing on-pump surgery [37]. The major post-operative complications related
to pre-operative liver impairment are refractory coagulopathy, an increased susceptibility to infections,
a higher rate of RV failure, portopulmonary hypertension and hepatorenal syndrome (HRS). Therefore,
perioperative therapy optimisation is mandatory to minimise morbidity, mortality, hospital stay and
costs [38,39].
In conclusion, the surgical risk in hepatopathic patients undergoing cardiac surgery requiring CPB
remains high, and the standard surgical option has to be reserved to patients in whom other ap-
proaches (i.e., percutaneous cardiac surgery) are not feasible. Pre-operative therapy optimisation, strict
intraoperative monitoring of both haemodynamics (i.e., transoesophageal echocardiography and/or
continuous cardiac output monitoring) and coagulation (i.e., thromboelastography) [38] are manda-
tory to improve outcome.
Therapeutic intervention
Pre-operative
The rst-line therapy of hepatic dysfunction is prevention. The identication of risk factors allows
prompt support of liver function, especially in patients with pre-operative liver dysfunction.
Usually, hepatic failure is related to LCOS and perioperative haemodynamic instability; thus, the
precocious use of inotropes/vasopressor or extracorporeal circulatory supports (intra-aortic balloon
pump (IABP), ECMO and ventricular assist device (VAD)) can reduce hepatic insults by optimising
organ perfusion. Pre-operative therapy and haemodynamic support of hepatopathic patients have to
be optimised before the surgery.
Intraoperative
The mean arterial pressure (MAP) is the circulatory system-driving pressure; thus, it should be
maintained above 60 mm Hg during surgery. However, the best MAP able to guarantee an adequate
liver perfusion remains unknown [40]. On the contrary, CVP should be as low as possible, compatibly
with a satisfactory stroke volume, in order to avoid liver congestion.
The use of uids should be strictly monitored, and a dry approach preferred. In case of blood
transfusion, the benets have to overcome the risks.
Metabolism and systemic effects of medications chronically administered have to be taken into
account during general anaesthesia (GA). The hypotensive effect of GA can reduce liver perfusion
pressure below a critical threshold. In case of acute liver dysfunction, the metabolism of anaesthetic
drugs is decreased, causing toxic effects and accumulation.
Although the haemodynamic stability rather than the anaesthesia per se is related to liver
dysfunction, the following considerations have to be taken into account. Most of the anaesthetic
medications undergo hepatic metabolism, which may be profoundly altered during liver disease. The
way hepatic metabolism acts on anaesthetic drugs is biphasic. The rst step, usually altered in liver
disease, is cytochrome-p450-mediated, while the second step, often normal at the beginning of liver
dysfunction, is glucuronide conjugation. The distribution volume of drugs can signicantly be affected
when hypoalbuminaemia coexists with liver dysfunction. These changes result in increased concen-
trations and reduced plasma clearance of drugs, which are often difcult to predict. For example,
propofol may be more manageable than benzodiazepines because of its rapid redistribution in patients
with cirrhosis, also when administered in continuous perfusion [41]. In the same way, the metabolism
of opioids (morphine, fentatienil and fentanyl) is decreased during hepatic impairment, leading to a
slower elimination and a higher risk of toxicity [42]. Conversely, remifentanil, a newer opioid, is
metabolised by esterases that are widespread throughout the plasma, red blood cells and interstitial
tissues. For this favourable pharmacokinetic prole, we suggest administering continuous-perfusion
remifentanil rather than other opioids in patients with advanced liver dysfunction undergoing car-
diac surgery. Rocuronium and vecuronium are neuromuscular blocking agents (NBAs) metabolised by
the liver. Khalil et al. reported slower onset and longer duration of action when these agents were
administered in patients with liver disease [43,44]. On the contrary, atracurium and cisatracurium, as
their metabolism is not dependent on the hepatic function, have been suggested as the NBA of choice
during hepatic disease [45]. The use of halogenated anaesthetics in the context of liver dysfunction is
still debated because of their hepatotoxic effect. The metabolism of halothane, enurane, isourane
and desurane is dependent on cytochrome p450, and it may produce triuoroacetylated components.
Besides, especially for halothane (no longer available in Western countries) and more rarely for the
newer halogenated agents, an autoimmune-mediated liver dysfunction can occur [46]. Among the
halogenated agents, sevourane, which is not metabolised to triuoroacetyl compounds, has shown
less liver toxicity [45].
Post-operative
Once liver dysfunction has occurred, it is mandatory to precociously treat it by avoiding further
organ impairment. Kidney dysfunction, in particular, is commonly related to the end-stage liver disease
in the setting of HRS. As this fatal complication is associated with a very poor prognosis, during the last
decades therapeutic interventions aiming to reduce mortality have been investigated.
The optimisation of systemic haemodynamics and the improvement of renal perfusion are deter-
minant to reverse HRS, as conrmed by Solanki et al. [47], who demonstrated an improved survival at
15 days in patients treated with terlipressin.
Terlipressin, an analogue of arginine vasopressin, is a potent vasoconstrictor of splanchnic arterial
system. It increases MAP and urine output, and it decreases sCrea, plasma renin activity and plasma
aldosterone. A recent meta-analysis [48] of randomised trials conrmed that terlipressin improves HRS
reversal with no signicant differences compared to noradrenaline. Moreover, Nguyen-Tat et al. re-
ported that terlipressin in combination with albumin improves survival in patients affected by HRS
[49]. Besides, albumin, compared to crystalloid and other colloids, has shown a nephroprotective effect
in patients with advanced liver disease who need volume expansion [50].
Liver replacement therapy, in which the detoxifying procedure is carried out with a molecular-
adsorbent recirculating system (MARS), has been extensively tested in patients with acute and
chronic liver diseases. Although MARS therapy is well tolerated and reduces hepatic toxins, it remains
unclear whether this liver dialysis improves the outcome, in the context of HRS [51].
Renal effects of CPB
Renal dysfunction after cardiac surgery remains a common complication and an independent pre-
dictor of post-operative morbidity and mortality [52]. The incidence is still unclear, and it ranges between
7% and 28% depending on the denitions of acute kidney injury (AKI). Post-operative renal replacement
therapy (RRT) is required in 1.4% of patients, and this percentage is higher in case of complex surgical
procedures (1% for isolated CABG; 5.1% for combined CABG and mitral valve surgery) [53].
The pathogenesis of AKI after cardiac surgery is multifactorial. As in the case of hepatic failure, the
release of pro-inammatory agents, haemodynamic changes during CPB and patient-related predis-
posing factors may trigger AKI [54]. In particular, activated complement system, up-regulation of cy-
tokines, activation of platelets, oxygen-free radicals, among the others, are related to leucocyte
migration and uid overload in the renal cellular interstitium [55]. The nal result is a cytotoxic tubular
injury. Kidneys are sensitive to haemodynamic changes occurring in cardiac surgery. Renal medulla is
more prone to ischaemia than other organs, due to its peculiar blood circulation characterized by low
oxygen tension with limited reserve. Other intraoperative factors related to acute tubular necrosis
(ATN) are [1] perioperative ischaemic-reperfusion injury [2], low cardiac output syndrome (LCOS) [3],
vasoconstriction [4], hypovolmia [5] and haemodilution/anaemia [56].
The optimisation of intraoperative haemodynamic can prevent AKI. There is no target MAP or
haematocrit during CPB able to reduce the risk of AKI.
For example, Azau et al. [57] reported that a higher MAP does not reduce the incidence of AKI, length
of hospital stay and mortality rate compared to a lower MAP. Ranucci et al. [58] observed, in a pro-
spective observational study, that a high degree of haemodilution, with consequent low haematocrit, is
an independent risk factor for post-operative renal dysfunction, particularly when associated with a
low oxygen delivery (DO2). In fact, an adequate pump blood-ow rate could counteract the negative
effects of a low haematocrit, improving the DO2.
The loss of a pulsatile ow can add further insult to the kidney. On the contrary, a pulsatile ow,
during CPB, reduces vasoconstrictive reexes, optimises oxygen consumption and decreases acidosis
[59]. Microemboli and hypothermia can be involved signicantly in the genesis of AKI.
Because of the clinical impact of even mild pre-operative renal dysfunction on the post-operative
outcome, some authors have suggested to include mild renal dysfunction in the EuroSCORE system
[60,61].
In summary, the major risk factors for AKI during cardiac surgery are the duration of CPB and the
cross-clamp time [62,63]. Further studies are necessary to investigate a more reliable method for risk
stratication and perioperative haemodynamic optimisation.
Biomarkers of AKI
An early diagnosis of AKI is decisive to prevent any further progression of renal damage. The RIFLE
score is based on creatinine increase and reduction of urinary output [64] as markers of renal
impairment. Its major limitation is the use of sCrea as a surrogate for glomerular ow rate (GFR). The
sCrea is not a specic marker of renal impairment; in fact, it can be altered by age, trauma, fever and
immobilisation. Moreover, sCrea is a late marker of AKI, and the trend, rather than the single value, is
more reliable [65].
Several experimental biomarkers of AKI have been tested, but none of them has been adopted into
clinical practice [66].
Cystatin-C showed a higher sensitivity than creatinine in identifying precociously AKI in cardiac
surgery [67].
Neutrophil gelatinase-associated lipocalin (N-GAL) is another marker of renal proximal tubular
injury [68]. Several observational studies have reported that urinary N-GAL is specic and sensitive in
predicting AKI 1 h after CPB [69]. This biomarker has also been tested to discriminate prerenal from
intrinsic AKI [70].
The interleukin (IL)-6 [71] and IL-18 [70], expression of systemic inammatory response, produced
by proximal renal tubules, have also been investigated. The TRIBE (Translational Research Involving
Biomarkers and Endpoints) study [72] recently validated the clinical predictivity of plasma N-GAL and
urine IL-18 in estimating postcardiac surgery AKI and relative outcomes, with an area under the curve
of 0.75 and 0.76, respectively.
Further studies are needed to determine which biochemical markers of AKI are able to guide
therapeutic interventions and to signicantly change patients' outcome.
Therapeutic interventions
To date, no specic therapeutic strategies are recommended to prevent AKI in patients undergoing
cardiac surgery with or without CPB. Obviously, CPB adding cross-clamp, cardioplegia, haemodilution
and hypothermia are detrimental for the kidney. As observed by several authors, cardiac surgery
without CBP (off-pump CPB) carries less post-operative AKI compared to on-pump surgery [73,74].
However, neither a decrease of mortality [75] nor a better preservation of long-term kidney function
has been reported [76]. A recent systematic review and international survey identied several drugs,
techniques or strategies that could reduce mortality in patients with or at risk of AKI, as suggested by at
least one peer-reviewed manuscript showing statistically signicant effect on survival [77].
Perioperative haemodynamic optimisation, albumin administration in cirrhotic patients, peri-

angiography haemoltration, plasma exchange in multiple-myeloma-associated AKI, increased in-
tensity and early use of renal replacement therapy, vasopressin in septic shock and furosemide by
continuous infusion, among others, have at least one manuscript supporting their benecial effect on
survival in patients with, or at risk of, AKI.
In daily practice, the following approaches can be suggested:
1. The maintenance of MAP above 60 mm Hg during CPB. This target is not specic; in patients with
hypertension or diabetes, this threshold should probably be higher, to guarantee good renal
perfusion and consequent urinary output.
2. The maintenance of euvolaemia. In the setting of cardiac surgery with CPB, cardioplegia is a key
point. The risk of uid overload is higher with crystalloid cardioplegia than with blood cardioplegia.
In the former, an adequate urinary output must be guaranteed.
3. The avoidance of low haematocrit. The kidney is sensitive to anaemia. Ranucci et al. have recom-
mended not tolerating haematocrit levels <25% during CPB.
4. Cardiac output support. The treatment of heart dysfunction is a cornerstone. A prompt use of
inotropes, vasopressors and paracorporeal devices provides sufcient renal perfusion pressure,
minimising renal damage.
Levosimendan, a calcium-sensitising agent increasing the contractility of cardiomyocytes without

increasing oxygen demand, is the most promising agent for AKI reduction, by acting on cardiac per-
formance [78]. Obviously, a pre-operative low ejection fraction is a risk factor for AKI.
5. The avoidance of nephrotoxic agents.

6. The administration of renoprotective drugs. A recent multicentre study has reported that fenol-
dopam does not reduce post-operative AKI and mortality in the setting of cardiac surgery [79].
Similarly, dopamine has shown no benets in preventing renal dysfunction [80].
7. The type of administered uid. A recent meta-analysis [81] of randomised trials supported the hy-
pothesis that albumin is nephroprotective, and hydroxyethyl starch is nephrotoxic. Albumin reduces
renal dysfunction and death, particularly in cirrhotic patients with spontaneous peritonitis [50].
8. The administration of terlipressin in extreme splanchnic vasodilatation syndrome.
HRS, indeed, as previously mentioned, is a common clinical state of renal impairment occurring in
cirrhotic patients, characterised by an intense renal vasoconstriction and reduced glomerular ltration.
This is probably due to the extreme splanchnic vasodilation and the relative reduced arterial volume.
Terlipressin, as a specic vasoconstrictor for splanchnic vessels, seems to improve circulatory function
and renal perfusion [47] in HRS.
Similarly, some studies showed that vasopressin [82,83] may reduce the progression of renal failure
and mortality in patients with septic shock, which is characterized by vasodilation and hyperdynamic
circulation. However, current clinical data are insufcient to conclude that one vasoactive agent is
better than another in preventing renal injury [84].
Conclusions
Post-operative hepatic and renal impairment are two of the most dreadful complications following
cardiac surgery requiring CPB. Nowadays, there are no conclusive studies on the prevention of hepatic
and renal perioperative failure. Further well-designed studies on the management of hepatic and renal
dysfunctions after cardiac surgery are necessary.
Conict of interest statement
None.
Research agenda
There is an urgent need to find more specific scores to classify patients at risk of renal and
hepatic dysfunction after CPB
There is a need for stronger evidence about the negative effects of CPB on renal and hepatic
functions (e.g., comparison with off-pump surgery)
Further works on novel hepatic and renal markers are warranted
We need urgent well-designed randomised trials on the effectiveness of haemodynamic
optimisation in cardiac surgery for the prevention of organ failure and death
Practice points
Every patient undergoing cardiac surgery should be clinically assessed and managed to
reduce the risk to develop renal or liver dysfunction
In selected cases, multispecialty collaboration between cardiac surgeons, anaesthesiologists
and gastroenterologists or nephrologists is indispensable to correctly formulate the indica-
tion for surgery in patients with advanced liver (Child B or C) or kidney disease
Perioperative haemodynamic optimisation is the most important factor to prevent both liver
and renal failure
Acknowledgements
The authors gratefully acknowledge the technical contribution of Alberto Castella and Valentina
Tarzia (San Raffaele Scientic Institute). None received compensation for their contributions.
This work was supported by departmental funds only. The department (sponsor) had no role in the
design of the study; the analysis and interpretation of the data; preparation, review or approval of the
manuscript; and decision to submit the manuscript for publication.
References
[1] Scandroglio AM, Finco G, Pieri M, et al. Cardiac surgery in 260 octogenarians : a case series. BMC Anesthesiol 2015;15:
1e8.
[2] Ascione R, Talpahewa S, Rajakaruna C, et al. Splanchnic organ injury during coronary surgery with or without cardio-
pulmonary bypass: a randomized, controlled trial. Ann Thorac Surg 2006 Jan;81(1):97e103.
[3] Raja SG, Haider Z, Ahmad M. Predictors of gastrointestinal complications after conventional and beating heart coronary
surgery. J R Coll Surg Edinb Irel 2003;1:221e8.
*[4] Yamada T, Ochiai R, Takeda J, et al. Off-pump coronary artery bypass attenuates transient hepatocellular damage after
myocardial revascularization. J Cardiothorac Vasc Anesth 2005;19:603e7.
[5] Shahbazi S, Panah A, Sahmeddini M. Evaluation of factors inuencing liver function test in on-pump coronary artery
bypass graft surgery. Iran J Med Sci 2013;38(4):308e13.
[6] Mustafa I, Roth H, Hanaah A, et al. Effect of cardiopulmonary bypass on lactate metabolism. Int Care Med 2003 Aug;
29(8):1279e85.
[7] DeFoe GR, Ross CS, Olmstead Elaine M, et al. Lowest hematocrit on bypass and adverse outcomes associated with cor-
onary artery bypass grafting. Ann Thorac Surg 2001;71(00):769e76.
[8] Poelzl G, Ess M, Mussner-Seeber C, et al. Liver dysfunction in chronic heart failure: prevalence, characteristics and
prognostic signicance. Eur J Clin Invest 2012;42:153e63.
*[9] Lau GT, Tan HC, Kritharides L. Type of liver dysfunction in heart failure and its relation to the severity of tricuspid
regurgitation. Am J Cardiol 2002;90:1405e9.
[10] Deursen Van, Damman K, Hillege HL, et al. J Cardiac Fail 2010;16(no. 1).
[11] Hessel EA. Abdominal organ injury after cardiac surgery. Semin Cardiothorac Vasc Anesth 2004 Sep 1;8(3):243e63.
[12] Seeto RK, Fenn B, Rockey DC. Ischemic hepatitis: clinical presentation and pathogenesis. Am J Med 2000;109:109e13.
[13] Alvarez AM, Mukherjee D. Liver abnormalities in cardiac diseases and heart failure. Int J Angiol 2011;1(212):135e42.
*[14] Birgens HS, Henriksen J, Matzen P, et al. The shock liver. Clinical and biochemical ndings in patients with centrilobular
liver necrosis following cardiogenic shock. Acta Med Scand 1978;204:417e21.
[15] Ackland G, Grocott MP, Mythen MG. Understanding gastrointestinal perfusion in critical care: so near, and yet so far. Crit
Care 2000;4:269e81.
[16] Fitzgerald T, Kim D, Karakozis S, et al. Visceral ischemia after cardiopulmonary bypass. Am Surg 2000;66:623e6.
[17] Sherlock S. The liver in heart failure; relation of anatomical, functional and circulatory changes. Br Hear J 1951;13(3):
273e93.
[18] Denault AY, Deschamps A, Couture P. Intraoperative hemodynamic instability during and after separation from cardio-
pulmonary bypass. Semin Cardiothorac Vasc Anesth 2010;14(3):165e82.
[19] Alvarez AM, Mukherjee D. Liver abnormalities in cardiac diseases and heart failure. Int J Angiol 2011;20:135e42.
*[20] An Y, Xiao Y, Zhong Q. Hyperbilirubinemia after extracorporeal circulation surgery: a recent and prospective study. World
J Gastroenterol 2006;12(41):6722e6.
[21] Hsu RB, Lin FY, Chen RJ, et al. Incidence, risk factors, and prognosis of postoperative hyperbilirubinemia after heart
transplantation. Eur J Cardio Thoracic Surg 2007;32:917e22.
[22] Nishi H, Sakaguchi T, Shigeru M, et al. Frequency, risk factors and prognosis of postoperative hyperbilirubinemia after
heart valve surgery. Cardiology 2012;122:12e9.
[23] Lockey E, McIntyre N, Ross DN, et al. Early jaundice after open-heart surgery. Thorax 1967;22:165e9.
[24] Collins JD, Bassendine MF, Ferner R, et al. Incidence and prognostic importance of jaundice after cardiopulmonary bypass
surgery. Lancet 1983;1(8334):1119e23.
[25] Sander M, Spies CD, Foer A, et al. Peri-operative plasma disappearance rate of indocyanine green after coronary artery
bypass surgery. Cardiovasc J Afr 2007;18(6):375e9.
[26] Sander M, Spies CD, Berger K, et al. Perioperative indocyanine green clearance is predictive for prolonged intensive care
unit stay after coronary artery bypass grafting: an observational study. Crit Care 2009;13(5):R149.
[27] Theodorakis K, Theodorakis I, Chatzimichael K, et al. Ultrasonographic evaluation of abdominal organs after cardiac
surgery. J Surg Res 2014:1e10.
[28] Filsou F, Salzberg SP, Rahmanian PB, et al. Early and late outcome of cardiac surgery in patients with liver cirrhosis. Liver
Transpl 2007;13:990e5.
[29] Kim HJ, Lee HW. Important predictor of mortality in patients with end-stage liver disease. Clin Mol Hepatol 2013;19:
105e15.
[30] Ziser A, Plevak DJ, Wiesner RH, et al. Morbidity and mortality in cirrhotic patients undergoing anesthesia and surgery.
Anesthesiology 1999;90:42e53.
[31] Klemperer JD, Ko W, Krieger KH, et al. Cardiac operations in patients with cirrhosis 1998;4975(97):8e10.
[32] Bizouarn P, Ausseur A, Desseigne P, et al. Early and late outcome after elective cardiac surgery in patients with cirrhosis.
Ann Thorac Surg 1999;67(99):1334e8.
[33] Nobuhiko H, Takahiro S, Hideki T, et al. Clinical outcome after cardiac operations in patients with cirrhosis. Ann Thorac
Surg 2004 Feb;77(2):500e5.
*[34] Morimoto N, Okada K, Okita Y. The model for end-stage liver disease (MELD) predicts early and late outcomes of car-
diovascular operations in patients with liver cirrhosis. Ann Thorac Surg 2013;96:1672e8.
[35] Arif R, Seppelt P, Schwill S, et al. Predictive risk factors for patients with cirrhosis undergoing heart surgery. Ann Thorac
Surg 2012;94:1947e52.
[36] Lin Cheng-Hsin, Hsu Ron-Bin. Cardiac surgery in patients with liver cirrhosis: risk factors for predicting mortality. World J
Gastroenterol 2014;20(35):12608e14. 37.
[37] Hayashida N, Teshima H, Chihara S, et al. Does off-pump coronary artery bypass grafting really preserve renal function?
Circ J 2002;66:921e5.
[38] Laurence W, Irene K, Clarissa T, et al. Haemostatic management for aortic valve replacement in a patient with advanced
liver disease. World J Clin Cases 2014 Oct 16;2(10):596e603.
[39] Kubota Y, Sakaguchi T, Miyagawa S, et al. Successful management of complex open heart surgery in a patient with child-
pugh class C liver cirrhosis: report of a case. Surg Today 2013 Mar;43(3):335e8.
[40] Sirvinskas E, Benetis R, Raliene L, et al. The inuence of mean arterial blood pressure during cardiopulmonary bypass on
postoperative renal dysfunction in elderly patients. Perfusion 2012;27:193e8.
[41] Servin F, Cockshott ID, Farinotti R, et al. Pharmacokinetics of propofol infusions in patients with cirrhosis. Br J Anaesth
1990;65:177e83.
[42] Delco F, Tchambaz L, Schlienger R, et al. Dose adjustment in patients with liver disease. Drug Saf 2005;28:529e45.
[43] Khalil M, D'Honneur G, Duvaldestin P, et al. Pharmacokinetics and pharmacodynamics of rocuronium in patients with
cirrhosis. Anesthesiology 1994;80:1241e7. 51.
[44] Lebrault C, Berger JL, D'Hollander AA, et al. Pharmacokinetics and pharmacodynamics of vecuronium (ORG NC 45).
[45] Hoetzel A, Ryan H, Schmidt R. Anesthetic considerations for the patient with liver disease. Curr Opin Anaesthesiol 2012;
25(3):340e7.
[46] Safari S, Motavaf M, Seyed Siamdoust SA, et al. Hepatotoxicity of halogenated inhalational anesthetics. Iran Red Crescent
Med J 2014;16(9):e20153.
[47] Solanki P, Chawla A, Garg R, et al. Benecial effects of terlipressin in hepatorenal syndrome: a prospective, randomized
placebo-controlled clinical trial. J Gastroenterol Hepatol 2003;18:152e6.
[48] Dobre M, Demirjian S, Sehgal AR, et al. Terlipressin in hepatorenal syndrome: a systematic review and meta-analysis. Int
Urol Nephrol 2011;43(1):997e1003.
[49] Nguyen-Tat M, Go tz E, Ahrens J, et al. Response to terlipressin and albumin is associated with improved outcome in
patients with cirrhosis and hepatorenal syndrome. Dtsch Med Wochenschr 2015;140. e21e6.
*[50] Sort P, Navasa P, Arroyo V, et al. Effect of intravenous albumin on renal impairment and mortality in patients with
cirrhosis and spontaneous bacterial peritonitis. N Engl J Med 1999;341:403e9.
[51] Gonwa T. Should mars and prometheus be used in patients with liver disease? Semin Dial 2014;27(10):228e31.
*[52] Bove T, Monaco F, Covello RD, et al. Acute renal failure and cardiac surgery. HSR Proc Intensive Care Cardiovasc Anesth
2009;1:13e21.
[53] Mehta RH, Grab JD, O'Brien SM, et al. Bedside tool for predicting the risk of postoperative dialysis in patients undergoing
cardiac surgery. Circulation 2006;114:2208e16.
[54] Mariscalco G, Lorusso R, Dominici C, et al. Acute kidney injury: a relevant complication after cardiac surgery. Elsevier Inc.
Ann Thorac Surg 2011;92(4):1539e47.
[55] Haddad R, El-hassan D, Araj A, et al. Some inammation-related parameters in patients following normo and hypo-
thermic cardiopulmonary bypass. Immunopharmacol Immunotoxicol 2001;23(2):291e302.
[56] Abu-Omar Y, Ratnatunga C. Cardiopulmonary bypass and renal injury. Perfusion 2006;21:209e13.
[57] Azau A, Markowicz P, Cottineau C, et al. Increasing mean arterial pressure during cardiac surgery does not reduce the rate
of postoperative acute kidney injury. Perfusion 2014;29(6):486e504.
*[58] Ranucci M, Romitti F, Isgro G, et al. Oxygen delivery during cardiopulmonary bypass and acute renal failure after coronary
operations. Ann Thorac Surg 2005;80:2213e20.
[59] O'Neil MP, Fleming JC, Badhwar A, et al. Pulsatile versus nonpulsatile ow during cardiopulmonary bypass: microcir-
culatory and systemic effects. Ann Thorac Surg 2012;94:2046e53.
[60] Geissler HJ, Holzl P, Marohl S, et al. Risk stratication in heart surgery: comparison of six score systems. Eur J Car-
diothorac Surg 2000;17:400e6.
[61] Miceli A, Bruno VD, Capoun R, et al. Mild renal dysfunction in patients undergoing cardiac surgery as a new risk factor for
EuroSCORE. Heart 2011;97:362e5.
[62] Urzua J, Troncoso S, Bugedo G, et al. Renal function and cardiopulmonary bypass: effect of perfusion pressure.
[63] Slogoff S, Reul GJ, Keats AS, et al. Role of perfusion pressure and ow in major organ dysfunction after cardiopulmonary
bypass. Ann Thorac Surg 1990;50:911e8.
*[64] Bellomo R, Ronco C, Kellum J a, et al. Acute renal failure e denition, outcome measures, animal models, uid therapy and
information technology needs: the second international consensus conference of the acute dialysis quality initiative
(ADQI) group. Crit Care 2004;8(4):R204e12.
*[65] Bellomo R, Kellum J a, Ronco C. Dening acute renal failure: physiological principles. Intensive Care Med 2004;30:73e7.
[66] Vanmassenhove J, Vanholder R, Nagler E, et al. Urinary and serum biomarkers for the diagnosis of acute kidney injury: an
in-depth review of the literature. Nephrol Dial Transpl 2013;28:254e73.
[67] Haase-Fielitz A, Bellomo R, Devarajan P, et al. Novel and conventional serum biomarkers predicting acute kidney injury in
adult cardiac surgeryea prospective cohort study. Crit Care Med 2009;37:553e60.
[68] Cowland JB, Borregaard N. Molecular characterization and pattern of tissue expression of the gene for neutrophil
gelatinase-associated lipocalin from humans. Genomics 1997;45:17e23.
[69] Wagener G, Jan M, Kim M, et al. Association between increases in urinary neutrophil gelatinase-associated lipocalin and
acute renal dysfunction after adult cardiac surgery. Anesthesiology 2006;105:485e91.
[70] Jayaraman R, Sunder S, Sathi S, et al. Post cardiac surgery acute kidney injury: a woebegone status rejuvenated by the
novel biomarkers. Nephrourol Mon 2014;6(4):1e6.
[71] Gueret G, Lion F, Guriec N, et al. Acute renal dysfunction after cardiac surgery with cardiopulmonary bypass is associated
with plasmatic IL6 increase. Cytokine 2009;45(2):92e8.
[72] Parikh CR, Coca SG, Thiessen-Philbrook H, et al. Postoperative biomarkers predict acute kidney injury and poor outcomes
after adult cardiac surgery. J Am Soc Nephrol 2011;22:1748e57.
[73] Seabra VF, Alobaidi S, Balk EM, et al. Off-pump coronary artery bypass surgery and acute kidney injury: a meta-analysis of
randomized controlled trials. Clin J Am Soc Nephrol 2010;5:1734e44.
[74] Sellke FW, DiMaio JM, Caplan LR, et al. Comparing on-pump and off-pump coronary artery bypass grafting: numerous
studies but few conclusions. Circulation 2005;111:2858e64.
[75] Crescenzi G, Landoni G, Romano A, et al. A propensity score analysis on the effect of eliminating cardiopulmonary bypass
for coronary artery bypass grafting. Minerva Anestesiol 2007;73(3):135e41.
[76] Garg A, Devereaux P, Yusuf S, et al. Kidney function after off-pump or on-pump coronary artery bypass graft surgery a
randomized clinical trial. JAMA 2014;311(21):2191e8.
[77] Landoni G, Bove T, Sze kely A, et al. Reducing mortality in acute kidney injury patients: systematic review and interna-
tional web-based survey. Elsevier J Cardiothorac Vasc Anesth 2013;27(6):1384e98.
[78] Pagel PS, Hettrick D a, Warltier DC. Inuence of levosimendan, pimobendan, and milrinone on the regional distribution of
cardiac output in anaesthetized dogs. Br J Pharmacol 1996;119:609e15.
[79] Bove T, Zangrillo A, Guarracino F, et al. Effect of fenoldopam on use of renal replacement therapy among patients with
acute kidney injury after cardiac surgery. JAMA 2014;312(21):2244.
[80] Woo EB, Tang AT, El-Gamel A, et al. Dopamine therapy for patients at risk of renal dysfunction following cardiac surgery:
science or ction? Eur J Cardiothorac Surg 2002;22:106e11.
[81] Wiedermann CJ, Dunzendorfer S, Gaioni LU, et al. Hyperoncotic colloids and acute kidney injury: a meta-analysis of
randomized trials. BioMed Central Ltd Crit Care 2010;14(5):R191.
[82] Gordon AC, Russell J a, Walley KR, et al. The effects of vasopressin on acute kidney injury in septic shock. Intensive Care
Med 2010;36:83e91.
[83] Gordon AC. The cardiopulmonary effects of vasopressin compared with norepinephrine in septic shock. Cardiopulmonary
effects of vasopressin. Chest J 2012;142:593.
[84] Lee RW, Di GD, May C, et al. Vasoactive drugs and the kidney. Best Pract Res Clin Anaesthesiol 2004;18:53e74.

Anaesthesiology
Pulmonary complications of cardiopulmonary

bypass
Julie L. Huffmyer, MD, Assistant Professor of Anesthesiology *,
Danja S. Groves, MD, PhD, Associate Professor of
Anesthesiology 1
Department of Anesthesiology, University of Virginia, PO Box 800710, Charlottesville, VA 22908-0710, USA
Keywords:
Pulmonary complications after the use of extracorporeal circulation
are common, and they range from transient hypoxemia with altered
pulmonary complications gas exchange to acute respiratory distress syndrome (ARDS), with
acute lung injury variable severity. Similar to other end-organ dysfunction after car-
acute respiratory distress syndrome diac surgery with extracorporeal circulation, pulmonary compli-
cations are attributed to the inammatory response, ischemia
ereperfusion injury, and reactive oxygen species liberated as a
result of cardiopulmonary bypass. Several factors common in car-
diac surgery with extracorporeal circulation may worsen the risk of
pulmonary complications including atelectasis, transfusion
requirement, older age, heart failure, emergency surgery, and pro-
longed duration of bypass. There is no magic bullet to prevent or
treat pulmonary complications, but supportive care with protective
ventilation is important. Targets for the prevention of pulmonary
complications include mechanical, surgical, and anesthetic in-
terventions that aim to reduce the contact activation, systemic in-
ammatory response, leukocyte sequestration, and hemodilution
associated with extracorporeal circulation.
Introduction
Postoperative pulmonary complications after cardiac surgery with cardiopulmonary bypass (CPB)
are common, and hypoxemia after bypass may be related to cardiac failure or pulmonary failure, or a
* Corresponding author. Tel.: 1 434 924 2283; Fax: 1 434 982 0019.
E-mail addresses: jh3wd@virginia.edu (J.L. Huffmyer), dgroves@virginia.edu (D.S. Groves).
1
Tel.: 1 434 924 2283; Fax: 1 434 982 0019.
164 J.L. Huffmyer, D.S. Groves / Best Practice & Research Clinical Anaesthesiology 29 (2015) 163e175
combination of both [1e4]. This review focuses on hypoxemia after CPB, and it denes the etiologies,
risk factors, and some prevention strategies for pulmonary complications. Historically, lung injury
associated with CPB was thought to be due to antiquated bubble oxygenation and pump circuitry. With
the advent of membrane oxygenation and improvement in technological advancements for the use of
extracorporeal circulation, the incidence of lung injury should have declined.
Altered postoperative pulmonary mechanics and impaired gas exchange contribute to pulmonary
complications after the use of extracorporeal circulation. Manifestations range from hypoxemia and
atelectasis to acute lung injury (ALI), respiratory failure, and acute respiratory distress syndrome
(ARDS), described originally in 1994 with the publication of the AmericaneEuropean Consensus
Conference [5]. In 2012, as a result of the need to describe changes based on ventilator settings, the
ARDS Denition Task Force convened and proposed the Berlin Denition for ARDS. This denes ARDS
based on three severities and several other ancillary variables [6] (see Table 1).
Surgical factors, including hemidiaphragmatic paresis leading to atelectasis, pain from sternal
wound and surgical drains, and residual pleural effusions, contribute to some of the additional pul-
monary risks after CPB and cardiac surgery. Atelectasis, shunt, alterations in the chest wall and lung
mechanics, changes in the capillary bed permeability, and in the lung tissue itself participate in the risk
of lung dysfunction. Most importantly, inammation related to the use of extracorporeal circulation is
postulated to be responsible for many of the other gas exchange issues and eventual ARDS. While
postoperative pulmonary complications after cardiac surgery leading to hypoxemia are thought to be
common, on the order of 5e7% in some studies, lung injury and ARDS are more rare complications
[7e11]. Weiss et al. performed a study on post-CPB hypoxemia, and they found that 1.3% of patients had
transient reduction in PaO2/FiO2 ratio, but these patients had prompt recovery and were extubated
within 12 h after surgery [12]. ARDS is thought to occur in as many as 2e3% of patients who underwent
cardiac surgery post operatively, but it carries a very poor prognosis with mortality in the 15e50%
range [3,4,13,14].
Etiology
Lung ischemia
During the course of CPB, the lungs are excluded from the systemic circulation as blood is diverted
from the right side of the heart to the venous reservoir, through the membrane oxygenator, and nally
oxygenated blood is returned to the aorta and systemic circulation. Metabolic demands of the lungs,
excluded from the systemic circulation for the duration of the CPB period, are dependent on the blood
ow from the bronchial arteries. The bronchial arteries supply about 3e5% of the pulmonary blood
ow in normal physiologic conditions, but upon experimental conditions of CPB in pigs, this ow may
decrease as much as 10-fold [15e17]. Thus, there is some period of ischemia to the pulmonary bed,
which is followed by reperfusion after CPB. This ischemiaereperfusion is thought to aggravate the
inammatory response initiated by CPB in general. During CPB, ATP in the lungs is reduced [18], and
lactate in the pulmonary blood is increased even to 6 h after the use of CPB [19]. In addition to the
effects of pulmonary ischemia during CPB with reperfusion and reoxygenation, oxygen-free radicals
are generated, and they contribute to this process of lung injury [20]. Oxygen-free radicals augment
inammation and activate neutrophils, macrophages, and endothelial cells [1,21]. Pulmonary
Table 1
Updated Berlin Denition for acute respiratory distress syndrome.
Oxygenation defect Mild: 200 mm Hg < PaO2/FIO2 300 mm Hg with PEEP or CPAP 5 cm H2O
Moderate: 100 mm Hg < PaO2/FIO2 200 mm Hg with PEEP 5 cm H2O
Severe: PaO2/FIO2 100 mm Hg with PEEP 5 cm H2O
Timing Occurs within 7 days of known clinical insult, new or worsening respiratory symptoms
Imaging Bilateral pulmonary opacities, not otherwise explained by pleural effusions, lobar/lung
collapse, or pulmonary nodules
Characteristics of Noncardiogenic, nonoverload. Echocardiography to exclude hydrostatic edema with
pulmonary edema no risk factors
J.L. Huffmyer, D.S. Groves / Best Practice & Research Clinical Anaesthesiology 29 (2015) 163e175 165
ischemiaereperfusion injury causes an increase in microvascular permeability, increased pulmonary

vascular resistance, pulmonary edema, impairment in oxygenation, and pulmonary hypertension [22].
Inammatory response
The inammatory reaction related to CPB has been hypothesized to contribute to lung injury and
postoperative pulmonary dysfunction. A review of the inammatory response related to extracorpo-
real circulation is the sole topic of a chapter by Zarbock et al. in this issue of Best Practice and Research
Clinical Anaesthesiology. The systemic inammatory response syndrome (SIRS) initiates this process
with important contributions from pro-inammatory and anti-inammatory cytokines, the comple-
ment system, and neutrophils [23e25].
Complement activation in combination with an imbalance of pro-inammatory and anti-
inammatory cytokines affects the magnitude of the SIRS response. Complement activation involves
approximately 20 proteins, activated in a sequence by classic and alternative pathways. The alternative
pathway is activated by the exposure of patient blood to extracorporeal (foreign material) circuitry,
which leads to the formation of C3a and C5a. The classic pathway is activated after heparin reversal
with protamine, and it causes an increase in C4a and subsequent C3a levels [23]. Endotoxin release in
the systemic circulation may activate both the alternative and classic complement pathways, and it
augments/amplies the inammatory response by increasing other pro-inammatory cytokines such
as tumor necrosis factor-a (TNF-a), procalcitonin, and interleukins (ILs) IL-1, IL-2, IL-6, and IL-8
[26e29]. This amplication leads to endothelial cell swelling, plasma and protein extravasation into
the interstitium, the release of enzymes, and the congestion of the alveolar air sacs with plasma, red
blood cells, and other debris [26].
The lungs are especially vulnerable to the inammatory effects ascribed to the use of CPB and
extracorporeal circulation. Apostolakis et al. described three reasons the lungs seem to be so vulnerable
to injury [30]. The inammatory process and ischemia affect endothelial cells of the lungs, the lungs
receive the full cardiac output with each ventricular systole, and the neutrophils that become
entrapped in the lung vasculature lead to congestion in the lung bed. These neutrophils become
aggregated via thromboxane A2, and they have been shown to be at their peak 2e4 h after CPB is
weaned [31,32]. Activated leukocytes increase capillary permeability and cause accumulation of uid in
the interstitial space and increase in the extravascular lung water [18]. This edema causes worsening of
gas exchange and mechanics of the chest wall and lungs. Edema, in addition to constrictive mediators,
leads to obstructive process in the airways, which serves to worsen bronchospasm and atelectasis [33].
Relating the inammatory response with the use of high concentration of oxygen and high PaO2
levels, Fujii et al. evaluated the effect of hyperoxia versus normoxia in the rat model of CPB, and they
found increased pro-inammatory cytokines such as TNF-a, IL-6, and IL-10, and biochemical markers
such as lactate dehydrogenase, alanine transaminase (ALT), and aspartate aminotransferase (AST) in
the hyperoxia group (PaO2 >400 mm Hg). There was also suppression in the increase of anti-
inammatory cytokines in the hyperoxia group as compared to the normoxia group. The wetedry
weight of the rat lungs was increased in the hyperoxia group, thus indicating an increase in the
extravascular lung water and edema [34]. In human patients, high oxygen concentrations have been
studied during CPB, and these were found to exacerbate lung injury. Pizov et al. divided 30 patients
undergoing coronary artery bypass graft (CABG) surgery with CPB into high oxygen (FiO2 1.0) and low
oxygen (FiO2 0.5) groups. Both groups experienced a decrease in oxygenation in the early period
following CPB, but patients treated with FiO2 1.0 throughout surgery had a signicant delay in the
recovery of oxygenation, and they had signicantly higher levels of TNF-a in the bronchoalveolar
lavage uid [35].
Atelectasis
Atelectasis occurs commonly after cardiac surgery with CPB, and it can contribute not only to
hypoxemia but also to other postoperative risks such as the development of pneumonia. In a
radiologic-computed tomography study by Rodrigues et al., there was a 31% reduction in pulmonary
gas volume but a 19% increase in tissue volume in patients after CABG with CPB. Nonaerated spaces
increased signicantly from 3% to 27%, poorly aerated spaces increased from 24% to 27%, and normally
aerated spaces decreased signicantly from 72% to 46%, preoperatively and post operatively [36].
Hutschala et al. found that perioperative atelectasis signicantly impairs antibiotic penetration in
patients undergoing coronary artery bypass surgery using CPB as compared to those having off-pump
coronary artery bypass surgery (OPCAB) [37].
Transfusion
Transfusion as a contributor to pulmonary dysfunction is difcult to cover comprehensively within

a review such as this. Transfusion results in complications such as transfusion-related acute lung injury
(TRALI) and transfusion-associated circulatory overload (TACO), and it contributes to immune and
inammatory complications. Even if transfusion does not lead to TRALI, it increases the risk of post-
operative pulmonary complications. In a prospective multicenter cohort study, patients who received
transfusion during cardiac surgery with CPB had signicantly greater increase in pulmonary capillary
permeability than those not transfused. The amount of packed red blood cell transfusion, not fresh-
frozen plasma or platelets, was associated with this increase in the pulmonary leak index [38]. In a
study of nearly 12,000 patients undergoing CABG surgery, the transfusion of red blood cells was
associated with a signicantly increased risk of prolonged ventilatory support, among other post-
operative morbidities [39]. Additionally, the use of older blood as compared to newer blood ap-
pears to be signicantly associated with higher complications including intubation beyond 3 days post
operatively from cardiac surgery [40].
TRALI is characterized by the onset of acute respiratory distress, hypoxemia as evidenced by PaO2/
FiO2 ratio <300, bilateral inltrates on chest X-ray, fever, and hypotension, temporally related to
transfusion within 6 h of symptoms. These criteria make it difcult to diagnose TRALI in patients who
underwent postoperative cardiac surgery. Patients are often hypothermic, intubated, and sedated after
bypass, and they receive blood products in the intraoperative and postoperative phases of cardiac
surgery procedures. Koch et al. evaluated >16,000 patients undergoing cardiac surgery, and they found
that patients receiving packed red blood cells and/or fresh-frozen plasma had signicantly more risk-
adjusted respiratory complications including respiratory distress, respiratory failure, longer intubation
times, ARDS, and requirement for reintubation [41].
Patient risk factors
Despite all of pathophysiological changes associated with extracorporeal circulation and the lungs,
most patients do not develop lung injury and ARDS after bypass. Thus, aside from the inammatory
response, lung ischemia and reperfusion, and oxygen-free radicals, there must be a patient component.
Response to the inammatory changes of CPB may be variable among patients. Recently, gene poly-
morphisms have been shown to be associated with elevated levels of pro-inammatory cytokines such
as TNF-a, IL-6, and IL-8, all of which are signicantly associated with postoperative pulmonary
dysfunction and prolonged intubation [42e44].
Risk factors for hypoxemia in previous studies included advanced age, obesity, reduced cardiac
function, previous myocardial infarction (MI), the emergency nature of surgery, baseline chest X-ray
with alveolar edema, a high creatinine level, prolonged CPB time, a decreased baseline P/F ratio, the
need for dopamine after CPB, an increased Hgb or protein content, persistent hypothermia, and the
need for reexploration [12]. Ji et al. found independent patient risk factors for postoperative pulmonary
complications after cardiac surgery with CPB to include age >65 years, preoperative congestive heart
failure, low preoperative oxygenation (PaO2), prolonged CPB, intraoperative phrenic nerve injury, and
postoperative acute kidney injury [2]. In a study of >5000 patients who underwent cardiac surgery,
Cislaghi et al. similarly found age >65 years, chronic renal failure, chronic obstructive pulmonary
disease (COPD), repeat surgery, emergency surgery, New York Heart Association (NYHA) Class 2 or
greater heart failure symptoms, left ventricular ejection fraction 30%, red blood cell or fresh-frozen
plasma transfusion more than four units, and CPB time >77 min to predict mechanical ventilation
time beyond 12 h [45]. By contrast, patients who were extubated early had shorter intensive care unit
(ICU) and hospital stays. A more recent study found that the body mass index (BMI) >30 was an
independent risk factor for postoperative hypoxia dened as PaO2/FiO2 ratio <200 at arrival in the ICU
[46].
Allou et al. developed a preoperative risk score for the development of postoperative pneumonia in
patients undergoing cardiac surgery. Upon multivariate analysis of 5582 patients, four risk factors were
found to be signicantly associated with postoperative pneumonia: advanced age >70 years, COPD,
preoperative left ventricular ejection fraction <60%, and the interaction between the transfusion of red
blood cells and the duration of CPB. In their study, Allou et al. found postoperative pneumonia to be a
frequent complication after cardiac surgery with high mortality (>40%) [47].
Prevention
As the end point lung injury after cardiac surgery utilizing CPB is a result of a multifactorial
process, efforts to ameliorate lung injury must address these factors in a multimodal approach. The
following section highlights some selected options (see Table 2). Although some studies were able to
demonstrate a benecial effect either on a cellular or on a short-term clinical level, most studies failed
to show a long-term benet. It must be kept in mind that these end points are inuenced by a number
of factors, and therefore they rarely reect only pulmonary function.
Mechanical factors
The CPB circuit

The articial surface of the CPB circuit acts as a contact-activating agent leading to an SIR [48]. The
reduction in the surface area to improve inammation has been achieved by the introduction of
smaller, newer oxygenators with smaller reservoirs, and arterialevenous loops. The development of a
Table 2
Prevention strategies for postoperative pulmonary complications.
Options Proposed mechanisms
Mechanical factors Miniaturized CPB circuit Decrease contact activation, decreased hemodilution
Coated CPB circuit (heparin, PMEA) Decrease contact activation
Leukocyte ltration Remove activated leukocytes, especially for longer
CPB duration
Retrograde autologous priming (RAP) Decrease hemodilution
Ultraltration (modied, zero-balance) Hemoconcentration, ltration of mediators
Normoxia on CPB Reduce reperfusion injury with free oxygen radicals
Surgical technique Avoid CPB if possible
Reduce the duration of time on CPB Reduction in time available for inammation/ischemic
injury to lungs
Cardioprotection with cardioplegia Prevent ischemiaereperfusion injury
Minimize cardiotomy suction Decrease bloodeair contact; decrease activation of
inammatory response
Transfusion-sparing techniques Decrease inammation/immune responses to transfusion
Partial lung perfusion Decrease ischemia to lungs, more complicated
surgical procedure
Anesthesia factors Intermittent ventilation Prevent atelectasis
Recruitment maneuver Reduce atelectasis, improve respiratory mechanics,
reduce volutrauma
Low tidal volume ventilation Prevent shear stress, various types of trauma to lungs:
volutrauma, barotraumas, and atelectatrauma
Volatile anesthesia-based
Medications Steroids Modulate immune response, potential negative impact
on glucose control and wound healing
Neutrophil elastase inhibitors Inhibit neutrophil elastase and reduce leukocyte
sequestration in the lungs
Hypertonic saline Decrease extravascular lung water, improve oxygenation
Aprotonin Decrease extravascular lung water, improve oxygenation,
and reduce neutrophil sequestration
miniaturized, closed-volume circuit system is the latest emergence in this effort. The miniaturized CPB
circuit reduces priming volume, the degree of hemodilution and its negative effects on organ perfusion,
and reduces transfusion requirements [49e52]. The miniaturized circuit eliminates the venous car-
diotomy suction reservoir, thereby decreasing the bloodeair interface and stasis in the reservoir. All
salvaged blood is cell-saved and washed before reinfusion, reducing inammation and dilution of
coagulation factors. This system is more complex than the standard CPB system, and it requires
communication between a surgeon, a perfusionist, and an anesthesiologist. While it is not suitable for
all types of surgeries, it has been shown to be benecial in complex procedures requiring longer
perfusion times [50].
Another approach in reducing the blood-circuit surface is by biocompatible coating, which mimics
the endothelial surface and thereby reduces the activation of the inammatory system. Newer gen-
eration materials are coated with heparin [53,54] or other biocompatible materials [55e57]. Benecial
effects of reduced cellular immune response (platelets, leukocytes, and endothelial cells) as well as
diminished release of inammatory markers (IL-6, IL-8, E-selectin, lactoferin, myeloperoxidase,
integrin, selectin, platelet b-thromboglobulin, and oxygen-free radicals) have been observed in mul-
tiple studies [58e61]. Positive clinical effects on postoperative lung compliance, pulmonary shunt
fraction, pulmonary vascular resistance index, and PaO2/FiO2 ratio as well as reduced pulmonary
capillary endothelial cell activation after CPB by using heparin-coated circuits have been observed.
However, this did not inuence the intubation time or the length of ICU stay in most studies
[58,62e64].
Benecial effects on clinical outcomes were observed especially in procedures requiring long cross-
clamp times [54,65].
Leukocyte ltration
Activated leukocytes and oxygen-free radicals have been implicated in the pathogenesis of lung
injury associated with CPB. Some experimental studies demonstrated a substantial reduction in
pulmonary injury after leukocyte depletion [66], but the results of clinical studies have been mixed
[67e70]. Various positions of the lter within the CPB circuit as well as strategic timing and
temperature of the ltration have been studied. Warren et al. summarized the current literature for
leukocyte ltration in their review [71]. The observed improvements in the early postoperative
lung function in patients receiving systemic leukodepletion did not lead to a reduced hospital stay
or to a decreased mortality. There is substantial evidence that cardioplegic leukocyte ltration
attenuates the reperfusion injury at a cellular level, but this has not translated into clinical
improvements.
Ultraltration
Upon institution of CPB, there is mixing of the patient's blood with the acellular CPB prime that
leads to hemodilution. Hemodilution may be helpful to facilitate tissue perfusion, yet hematocrit levels
below 23% have been associated with increased interstitial edema and dysfunction of end organs such
as brain, heart, and lungs [72]. Ultraltration during CPB removes uid volume from the pump prime,
thus increasing hematocrit and colloid osmotic pressure. The combined effect of the reduced total body
water and the increased intravascular colloid osmotic pressure reduces postoperative total body water
and edema. Indeed, benecial effects on the postoperative lung function have been demonstrated in
numerous experimental and clinical studies [73e76]. Furthermore, the ability of ultraltration to
remove inammatory substances from the circulation is a focus of interest. Most inammatory me-
diators have a molecular weight that is below the membrane pore size of commonly used ultralters,
which should allow them to be freely ltered [77,78]. A number of variants of the conventional ul-
traltration have been developed and applied in the pediatric population where positive effects on the
total body water content [79] and edema formation, improved pulmonary function [73,75], reduced
inammatory response, decreased coagulopathy [80,81], and blood transfusion requirements [82] have
been demonstrated. Ziyaeifard et al. summarized the current literature in a review article [83].
Retrograde autologous priming (RAP) of the CPB circuit prior to the institution of CPB helps to remove
some of the crystalloid prime uid by allowing priming of the circuit with some of the patient's own
circulating blood after cannulae have been inserted for bypass. A study by Hwang et al. showed that
patients who underwent RAP prior to bypass had signicantly higher hematocrit as well as cerebral
oxygenation saturation levels during the bypass period [84].
Surgical technique
While this issue of Best Practice is a review of extracorporeal circulation and its effects and
outcomes, the avoidance of CPB when possible may be one of the only options for the amelioration of
ARDS and lung injury [85]. A meta-analysis of studies evaluating effects of cardiac surgery with CPB
versus OPCAB on various postoperative complications for patients with a reduced left ventricular
function (preoperative ejection fraction <40%) revealed signicant reduction in intubation time in
the ICU but comparable postoperative pulmonary complication rates between the groups [86].
However, this is only possible in CABG surgery and not applicable to most valve procedures.
Therefore, maintaining the duration of CPB as short as possible is a key factor in the prevention of ALI
[87].
During surgery, ensuring excellent cardioprotection during aortic cross-clamp time is a mainstay in
protecting the lung from the release of mediators during reperfusion. The type (crystalloid, blood, cold,
and warm) and route (antegrade, retrograde, or both) of administered cardioplegia seem to play a role
in cytokine release [88]. Partial-lung and full-lung perfusion techniques during CPB include partial
continuous low-ow perfusion, intermittent perfusion with a protective solution (pneumoplegia), and
biventricular CPB [89e93]. The so-called DreweAnderson technique uses the patient's own lungs in a
biventricular CPB as an oxygenator, excluding the articial oxygenator as an inammatory stimulator
from the CPB circuit [94]. Although some favorable results have been demonstrated, this technique has
not made it into clinical routine [95e97]. A review by Suzuki summarizes the current knowledge on
this topic [98].
Anesthetic management
Ventilation
Blood ow to the lungs during total CPB is limited to the bronchial artery ow, and ischemic
pulmonary injury causes the bronchial ow to become insufcient to meet the metabolic demands
[16]. The concept of ventilating the lungs during CPB is mainly to prevent atelectasis. Oxygenation will
only occur through passive diffusion as the pulmonary vasculature is excluded during CPB. Various
regimens of ventilatory settings (i.e., intermittent, continuous, CPAP, low tidal volume, and positive
end-expiratory pressure (PEEP)) have been studied, during and after separation from CPB, with mixed
results in regard to inammatory markers as well as to clinical outcomes [99e101]. Ventilation during
CPB might not always be feasible as the inated lung could interfere with surgical exposure. Apnea
not only promotes atelectasis but also promotes the activation of enzymes in the pulmonary circu-
lation that may lead to postoperative lung dysfunction [102]. A small study of patients who under-
went valve surgery found that patients managed with beating heart, on CPB, with low tidal volume
ventilation throughout had lower levels of inammatory and oxidative stress markers such as
malondialdehyde, lactic acid, and myeloperoxidase [103]. Vital capacity maneuvers one to three times
with a pressure of 35e40 cm H2O at the end of the CPB period improves oxygenation in the early
postoperative period, but there are no sustainable effects on oxygenation or lung function once into
the ICU [104].
Choice of anesthesia
Balanced anesthesia including a volatile agent plus opiate is a common choice for the maintenance
of anesthesia during cardiac surgery and CPB. Total intravenous anesthesia (TIVA) versus balanced
anesthesia has been studied by a variety of authors, and it has not been found to signicantly reduce
the pro-inammatory cytokine milieu [105,106]. While the myocardium may be a source of pro-
inammatory cytokines during surgery with CPB, the lungs are often the consumers of those cyto-
kines [107,108]. Some studies have demonstrated benet in the reduction of pro-inammatory cyto-
kines as well as reduction in the pulmonary sequestration of cytokines with the use of volatile
anesthetic agents such as sevourane [109,110].
Steroids
There have been several large trials in patients undergoing cardiac surgery to evaluate the impact of
steroids on the inammatory response and many other outcomes after the use of extracorporeal cir-
culation. Study results including small studies [111], Cochrane review [112], and large, multicenter
international trials [113,114] have yielded mixed results; thus, steroids are not currently indicated to
reduce pulmonary complications.
Alternative treatments
As neutrophil and leukocyte activation and sequestration in the lungs contribute to postoperative
pulmonary dysfunction, one potential target for prevention is white blood cells. Neutrophil elastase
inhibitors such as sivelestat and ulinastatin have been used to competitively inhibit neutrophil elastase
activity and improved respiratory and oxygenation indices, as well as shorter time to extubation
[115,116]. A small study by Lomivorotov et al. randomized patients to receive hypertonic saline infusion
(7.2% NaCl/hydroxylethyl starch) versus 0.9% NaCl infusion, and they found decreases in extravascular
lung-water index measurements and improvements in arterial oxygenation in the early time period
after weaning from CPB in the hypertonic saline group [117]. While it is currently not available in the
United States, aprotinin, a serine protease inhibitor used for hemostasis and for the prevention of
brinolysis in cardiac surgery, has been shown to have a variety of positive effects on multiple organ
systems, including the lungs. Aprotinin use has been associated with a reduction in the ventilation time
and pulmonary complications as well as decreased extravascular lung water and improvement in
oxygenation indices [118,119].
Conclusion
Postoperative pulmonary complications after cardiac surgery with extracorporeal circulation are
common, and they range from transient hypoxemia to the more severe ALI and ARDS, which are
difcult to treat. There are multiple etiologies and patient risk factors, but the inammatory response
associated with extracorporeal circulation and CPB is thought to play a major role in lung injury and
pulmonary complications after cardiac surgery. The prevention techniques reviewed here aim to
modulate the inammatory immune response as well as the coagulation system. It is important to
remember that pulmonary complications after extracorporeal circulation have a multifactorial
component; thus, prevention and treatment must be directed toward multiple targets.
Practice points
Pulmonary complications including hypoxemia, pneumonia, ALI, and ARDS continue to

occur with frequency, and they are associated with significant morbidity and mortality risks
after cardiac surgery utilizing extracorporeal circulation despite advances in technology.
Much of the risk associated with the use of CPB for ALI and ARDS is postulated to occur due to
inflammation, ischemia, and reperfusion of the lung bed that occurs during the bypass
period.
Atelectasis occurs commonly during general anesthesia, and it is worsened with the use of
extracorporeal membrane oxygenation. Atelectasis worsens hypoxemia and increases the
risk of pneumonia. Lung recruitment maneuvers are important to reduce atelectasis.
Transfusion increases risks of pulmonary complications through increased inflammation,
TRALI, and TACO. Avoid transfusion if possible, and utilize newer blood as compared to
older blood to ameliorate inflammatory reaction.
Optimize and treat aggressively conditions such as heart failure and COPD that impair gas
exchange and predict pulmonary complications in the postoperative period.
Consider mechanical, surgical, and anesthetic techniques to ameliorate lung injury such as
shortest CPB as possible, coated circuits, hemoconcentration, volatile anesthesia, and
ventilatory recruitment maneuvers.
Research agenda
Many of the studies are small and underpowered to detect differences in the severe pul-
monary complications after extracorporeal circulation. Large studies are needed to evaluate
the effect of interventions on the more rare, but severe pulmonary complications such as ALI
and ARDS.
Interventions to reduce inflammation and transfusion risk should be sought, as these seem to
be major complicating risk factors for pulmonary complications.
Alternative ventilation and perfusion techniques for the lungs during CPB have shown var-
iable results; thus, larger trials should aim to define which of these strategies are helpful to
reduce pulmonary complications post operatively.
Conict of interest
The authors do not have any conict of interest.
Acknowledgment
The preparation of this manuscript was solely supported by departmental funding.
References
[1] Ng CS, Wan S, Yim AP, et al. Pulmonary dysfunction after cardiac surgery. Chest 2002;121(4):1269e77.
[2] Ji Q, Mei Y, Wang X, et al. Risk factors for pulmonary complications following cardiac surgery with cardiopulmonary
bypass. Int J Med Sci 2013;10(11):1578e83.
[3] Fowler AA, Hamman RF, Good JT, et al. Adult respiratory distress syndrome: risk with common predispositions. Ann
Intern Med 1983;98(5 Pt 1):593e7.
[4] Messent M, Sullivan K, Keogh BF, et al. Adult respiratory distress syndrome following cardiopulmonary bypass:
incidence and prediction. Anaesthesia 1992;47(3):267e8.
[5] Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus Conference on ARDS. Denitions,
mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994;149(3 Pt 1):818e24.
*[6] Ranieri VM, Rubeneld GD, Thompson BT, et al. Acute respiratory distress syndrome: the Berlin Denition. JAMA 2012;
307(23):2526e33.
[7] Rock P, Rich PB. Postoperative pulmonary complications. Curr Opin Anaesthesiol 2003;16(2):123e31.
[8] Weissman C. Pulmonary complications after cardiac surgery. Semin Cardiothorac Vasc Anesth 2004;8(3):185e211.
[9] Younossian AB, Adler D, Bridevaux PO, et al. Postoperative pulmonary complications: how to anticipate and prevent
the risk? Rev Med Suisse 2011;7(317):6e9. 2214.
[10] Sachdev G, Napolitano LM. Postoperative pulmonary complications: pneumonia and acute respiratory failure. Surg
Clin N Am 2012;92(2):321e44. ix.
[11] Akdur H, Yigit Z, Arabaci UO, et al. Investigation of the relationship between the duration of postoperative mechanical
ventilation and complication incidence following coronary artery bypass graft. Med Sci Monit Int Med J Exp Clin Res
2007;13(2). CR105-10.
*[12] Weiss YG, Merin G, Koganov E, et al. Postcardiopulmonary bypass hypoxemia: a prospective study on incidence, risk
factors, and clinical signicance. J Cardiothorac Vasc Anesth 2000;14(5):506e13.
[13] Milot J, Perron J, Lacasse Y, et al. Incidence and predictors of ARDS after cardiac surgery. Chest 2001;119(3):884e8.
[14] Christenson JT, Aeberhard JM, Badel P, et al. Adult respiratory distress syndrome after cardiac surgery. Cardiovasc Surg
1996;4(1):15e21.
[15] Schlensak C, Doenst T, Preusser S, et al. Bronchial artery perfusion during cardiopulmonary bypass does not prevent
ischemia of the lung in piglets: assessment of bronchial artery blood ow with uorescent microspheres. Eur J Cardio-
thoracic Surg Off J Eur Assoc Cardio-thoracic Surg 2001;19(3):326e31. disciussion 31e2.
[16] Schlensak C, Doenst T, Preusser S, et al. Cardiopulmonary bypass reduction of bronchial blood ow: a potential
mechanism for lung injury in a neonatal pig model. J Thorac Cardiovasc Surg 2002;123(6):1199e205.
[17] Schlensak C, Doenst T, Beyersdorf F. Lung ischemia during cardiopulmonary bypass. Ann Thorac Surg 2000;70(1):
337e8.
[18] Serraf A, Robotin M, Bonnet N, et al. Alteration of the neonatal pulmonary physiology after total cardiopulmonary
bypass. J Thorac Cardiovasc Surg 1997;114(6):1061e9.
[19] Gasparovic H, Plestina S, Sutlic Z, et al. Pulmonary lactate release following cardiopulmonary bypass. Eur J Cardio-
thoracic Surg Off J Eur Assoc Cardio-thoracic Surg 2007;32(6):882e7.
[20] Friedman M, Sellke FW, Wang SY, et al. Parameters of pulmonary injury after total or partial cardiopulmonary bypass.
Circulation 1994;90(5 Pt 2):II262e8.
[21] Partrick DA, Moore EE, Fullerton DA, et al. Cardiopulmonary bypass renders patients at risk for multiple organ failure
via early neutrophil priming and late neutrophil disability. J Surg Res 1999;86(1):42e9.
*[22] den Hengst WA, Gielis JF, Lin JY, et al. Lung ischemia-reperfusion injury: a molecular and clinical view on a complex
pathophysiological process. Am J Physiol Heart Circ Physiol 2010;299(5):H1283e99.
[23] Laffey JG, Boylan JF, Cheng DC. The systemic inammatory response to cardiac surgery: implications for the anes-
thesiologist. Anesthesiology 2002;97(1):215e52.
[24] Davies MG, Hagen PO. Systemic inammatory response syndrome. Br J Surg 1997;84(7):920e35.
[25] Paparella D, Yau TM, Young E. Cardiopulmonary bypass induced inammation: pathophysiology and treatment. An
update. Eur J Cardio-thoracic Surg Off J Eur Assoc Cardio-thoracic Surg 2002;21(2):232e44.
[26] Edmunds Jr LH. Why cardiopulmonary bypass makes patients sick: strategies to control the blood-synthetic surface
interface. Adv Cardiac Surg 1995;6:131e67.
[27] Finn A, Naik S, Klein N, et al. Interleukin-8 release and neutrophil degranulation after pediatric cardiopulmonary
bypass. J Thorac Cardiovasc Surg 1993;105(2):234e41.
[28] Chen H, Cheng ZB, Yu RG. Procalcitonin as a predictor of moderate to severe acute respiratory distress syndrome after
cardiac surgery with cardiopulmonary bypass: a study protocol for a prospective cohort study. BMJ Open 2014;4(10):
e006344.
[29] Li T, Luo N, Du L, et al. Early and marked up-regulation of TNF-alpha in acute respiratory distress syndrome after
cardiopulmonary bypass. Front Med 2012;6(3):296e301.
*[30] Apostolakis E, Filos KS, Koletsis E, et al. Lung dysfunction following cardiopulmonary bypass. J Cardiac Surg 2010;
25(1):47e55.
[31] Cave AC, Manche A, Derias NW, et al. Thromboxane A2 mediates pulmonary hypertension after cardiopulmonary
bypass in the rabbit. J Thorac Cardiovasc Surg 1993;106(6):959e67.
[32] Abdullah F, Ovadia P, Feuerstein G, et al. The novel chemokine mob-1: involvement in adult respiratory distress
syndrome. Surgery 1997;122(2):303e12.
[33] Babik B, Asztalos T, Petak F, et al. Changes in respiratory mechanics during cardiac surgery. Anesth Analg 2003;96(5):
1280e7. table of contents.
[34] Fujii Y, Shirai M, Tsuchimochi H, et al. Hyperoxic condition promotes an inammatory response during cardiopul-
monary bypass in a rat model. Artif Organs 2013;37(12):1034e40.
[35] Pizov R, Weiss YG, Oppenheim-Eden A, et al. High oxygen concentration exacerbates cardiopulmonary bypass-
induced lung injury. J Cardiothorac Vasc Anesth 2000;14(5):519e23.
[36] Rodrigues RR, Sawada AY, Rouby JJ, et al. Computed tomography assessment of lung structure in patients undergoing
cardiac surgery with cardiopulmonary bypass. Braz J Med Biol Res Revista brasileira de pesquisas medicas e bio-
logicas/Sociedade Brasileira de Biosica 2011;44(6):598e605.
[37] Hutschala D, Kinstner C, Skhirtladze K, et al. The impact of perioperative atelectasis on antibiotic penetration into lung
tissue: an in vivo microdialysis study. Intensive Care Med 2008;34(10):1827e34.
[38] Vlaar AP, Cornet AD, Hofstra JJ, et al. The effect of blood transfusion on pulmonary permeability in cardiac surgery
patients: a prospective multicenter cohort study. Transfusion 2012;52(1):82e90.
[39] Koch CG, Li L, Duncan AI, et al. Morbidity and mortality risk associated with red blood cell and blood-component
transfusion in isolated coronary artery bypass grafting. Crit Care Med 2006;34(6):1608e16.
[40] Koch CG, Li L, Sessler DI, et al. Duration of red-cell storage and complications after cardiac surgery. New Engl J Med
2008;358(12):1229e39.
*[41] Koch C, Li L, Figueroa P, et al. Transfusion and pulmonary morbidity after cardiac surgery. Ann Thorac Surg 2009;88(5):
1410e8.
[42] Tomasdottir H, Hjartarson H, Ricksten A, et al. Tumor necrosis factor gene polymorphism is associated with enhanced
systemic inammatory response and increased cardiopulmonary morbidity after cardiac surgery. Anesth Analg 2003;
97(4):944e9. table of contents.
[43] Grunenfelder J, Umbehr M, Plass A, et al. Genetic polymorphisms of apolipoprotein E4 and tumor necrosis factor beta
as predisposing factors for increased inammatory cytokines after cardiopulmonary bypass. J Thorac Cardiovasc Surg
2004;128(1):92e7.
[44] Wacharasint P, Nakada TA, Boyd JH, et al. AA genotype of IL-8 -251A/T is associated with low PaO(2)/FiO(2) in critically
ill patients and with increased IL-8 expression. Respirology 2012;17(8):1253e60.
*[45] Cislaghi F, Condemi AM, Corona A. Predictors of prolonged mechanical ventilation in a cohort of 5123 cardiac surgical
patients. Eur J Anaesthesiol 2009;26(5):396e403.
[46] Ranucci M, Ballotta A, La Rovere MT, et al., Surgical, Clinical Outcome Research G. Postoperative hypoxia and length of
intensive care unit stay after cardiac surgery: the underweight paradox? PLoS One 2014;9(4):e93992.
[47] Allou N, Bronchard R, Guglielminotti J, et al. Risk factors for postoperative pneumonia after cardiac surgery and
development of a preoperative risk score. Crit Care Med 2014;42(5):1150e6.
[48] Butler J, Rocker GM, Westaby S. Inammatory response to cardiopulmonary bypass. Ann Thorac Surg 1993;55(2):
552e9.
total minimal extracorporeal circulation. Eur J Cardio-thoracic Surg Off J Eur Assoc Cardio-thoracic Surg 2002;22(4):
527e33.
[50] Rimpilainen R, Biancari F, Wistbacka JO, et al. Outcome after coronary artery bypass surgery with miniaturized versus
conventional cardiopulmonary bypass. Perfusion 2008;23(6):361e7.
[51] El-Essawi A, Hajek T, Skorpil J, et al. Are minimized perfusion circuits the better heart lung machines? Final results of a
prospective randomized multicentre study. Perfusion 2011;26(6):470e8.
[52] Remadi JP, Rakotoarivelo Z, Marticho P, et al. Prospective randomized study comparing coronary artery bypass grafting
with the new mini-extracorporeal circulation Jostra System or with a standard cardiopulmonary bypass. Am Heart J
2006;151(1):198.
[53] Hall RI, Smith MS, Rocker G. The systemic inammatory response to cardiopulmonary bypass: pathophysiological,
therapeutic, and pharmacological considerations. Anesth Analg 1997;85(4):766e82.
[54] Jansen PG, te Velthuis H, Huybregts RA, et al. Reduced complement activation and improved postoperative perfor-
mance after cardiopulmonary bypass with heparin-coated circuits. J Thorac Cardiovasc Surg 1995;110(3):829e34.
[55] Suhara H, Sawa Y, Nishimura M, et al. Efcacy of a new coating material, PMEA, for cardiopulmonary bypass circuits in
a porcine model. Ann Thorac Surg 2001;71(5):1603e8.
[56] Wimmer-Greinecker G, Matheis G, Martens S, et al. Synthetic protein treated versus heparin coated cardiopulmonary
bypass surfaces: similar clinical results and minor biochemical differences. Eur J Cardio-thoracic Surg Off J Eur Assoc
Cardio-thoracic Surg 1999;16(2):211e7.
[57] De Somer F, Francois K, van Oeveren W, et al. Phosphorylcholine coating of extracorporeal circuits provides natural
protection against blood activation by the material surface. Eur J Cardio-thoracic Surg Off J Eur Assoc Cardio-thoracic
Surg 2000;18(5):602e6.
*[58] de Vroege R, van Oeveren W, van Klarenbosch J, et al. The impact of heparin-coated cardiopulmonary bypass circuits
on pulmonary function and the release of inammatory mediators. Anesth Analg 2004;98(6):1586e94. table of
contents.
[59] Weerwind PW, Maessen JG, van Tits LJ, et al. Inuence of Durao II heparin-treated extracorporeal circuits on the
systemic inammatory response in patients having coronary bypass. J Thorac Cardiovasc Surg 1995;110(6):1633e41.
[60] Bozdayi M, Borowiec J, Nilsson L, et al. Effects of heparin coating of cardiopulmonary bypass circuits on in vitro oxygen
free radical production during coronary bypass surgery. Artif Organs 1996;20(9):1008e16.
[61] Fukutomi M, Kobayashi S, Niwaya K, et al. Changes in platelet, granulocyte, and complement activation during car-
diopulmonary bypass using heparin-coated equipment. Artif Organs 1996;20(7):767e76.
[62] Ranucci M, Cirri S, Conti D, et al. Benecial effects of Durao II heparin-coated circuits on postperfusion lung
dysfunction. Ann Thorac Surg 1996;61(1):76e81.
[63] Wan S, LeClerc JL, Antoine M, et al. Heparin-coated circuits reduce myocardial injury in heart or heart-lung trans-
plantation: a prospective, randomized study. Ann Thorac Surg 1999;68(4):1230e5.
[64] te Velthuis H, Baufreton C, Jansen PG, et al. Heparin coating of extracorporeal circuits inhibits contact activation
during cardiac operations. J Thorac Cardiovasc Surg 1997;114(1):117e22.
[65] Wildevuur CR, Jansen PG, Bezemer PD, et al. Clinical evaluation of Durao II heparin treated extracorporeal circulation
circuits (2nd version). The European Working Group on heparin coated extracorporeal circulation circuits. Eur J
Cardio-thoracic Surg Off J Eur Assoc Cardio-thoracic Surg 1997;11(4):616e23. discussion 24e5.
[66] Bando K, Pillai R, Cameron DE, et al. Leukocyte depletion ameliorates free radical-mediated lung injury after car-
diopulmonary bypass. J Thorac Cardiovasc Surg 1990;99(5):873e7.
[67] Morioka K, Muraoka R, Chiba Y, et al. Leukocyte and platelet depletion with a blood cell separator: effects on lung
injury after cardiac surgery with cardiopulmonary bypass. J Thorac Cardiovasc Surg 1996;111(1):45e54.
[68] Gu YJ, de Vries AJ, Vos P, et al. Leukocyte depletion during cardiac operation: a new approach through the venous
bypass circuit. Ann Thorac Surg 1999;67(3):604e9.
[69] Mihaljevic T, Tonz M, von Segesser LK, et al. The inuence of leukocyte ltration during cardiopulmonary bypass on
postoperative lung function. A clinical study. J Thorac Cardiovasc Surg 1995;109(6):1138e45.
[70] Sheppard SV, Gibbs RV, Smith DC. Does leucocyte depletion during cardiopulmonary bypass improve oxygenation
indices in patients with mild lung dysfunction? Br J Anaesth 2004;93(6):789e92.
[71] Warren O, Alexiou C, Massey R, et al. The effects of various leukocyte ltration strategies in cardiac surgery. Eur J
Cardio-thoracic Surg Off J Eur Assoc Cardio-thoracic Surg 2007;31(4):665e76.
[72] DeFoe GR, Ross CS, Olmstead EM, et al. Lowest hematocrit on bypass and adverse outcomes associated with coronary
artery bypass grafting. Northern New England Cardiovascular Disease Study Group. Ann Thorac Surg 2001;71(3):
769e76.
[73] Huang H, Yao T, Wang W, et al. Continuous ultraltration attenuates the pulmonary injury that follows open heart
surgery with cardiopulmonary bypass. Ann Thorac Surg 2003;76(1):136e40.
[74] Keenan HT, Thiagarajan R, Stephens KE, et al. Pulmonary function after modied venovenous ultraltration in infants:
a prospective, randomized trial. J Thorac Cardiovasc Surg 2000;119(3):501e5. discussion 6e7.
[75] Mahmoud AB, Burhani MS, Hannef AA, et al. Effect of modied ultraltration on pulmonary function after cardio-
pulmonary bypass. Chest 2005;128(5):3447e53.
[76] Nagashima M, Shin'oka T, Nollert G, et al. High-volume continuous hemoltration during cardiopulmonary bypass
attenuates pulmonary dysfunction in neonatal lambs after deep hypothermic circulatory arrest. Circulation 1998;
98(19 Suppl.):II378e84.
[77] Tallman RD, Dumond M, Brown D. Inammatory mediator removal by zero-balance ultraltration during cardio-
pulmonary bypass. Perfusion 2002;17(2):111e5.
[78] Berdat PA, Eichenberger E, Ebell J, et al. Elimination of proinammatory cytokines in pediatric cardiac surgery:
analysis of ultraltration method and lter type. J Thorac Cardiovasc Surg 2004;127(6):1688e96.
[79] Gurbuz AT, Novick WM, Pierce CA, et al. Impact of ultraltration on blood use for atrial septal defect closure in infants
and children. Ann Thorac Surg 1998;65(4):1105e8. discussion 8e9.
[80] Buchholz BJ, Bert AA, Price DR, et al. Veno-arterial modied ultraltration in children after cardiopulmonary bypass.
J Extra-corporeal Technol 1999;31(1):47e9.
[81] Ootaki Y, Yamaguchi M, Oshima Y, et al. Effects of modied ultraltration on coagulation factors in pediatric cardiac
surgery. Surg Today 2002;32(3):203e6.
[82] Friesen RH, Campbell DN, Clarke DR, et al. Modied ultraltration attenuates dilutional coagulopathy in pediatric open
heart operations. Ann Thorac Surg 1997;64(6):1787e9.
[83] Ziyaeifard M, Alizadehasl A, Massoumi G. Modied ultraltration during cardiopulmonary bypass and postoperative
course of pediatric cardiac surgery. Res Cardiovasc Med 2014;3(2):e17830.
[84] Hwang J, Huh J, Kim J, et al. The effect of retrograde autologous priming of the cardiopulmonary bypass circuit on
cerebral oxygenation. J Cardiothorac Vasc Anesth 2011;25(6):995e9.
[85] Wan S, Izzat MB, Lee TW, et al. Avoiding cardiopulmonary bypass in multivessel CABG reduces cytokine response and
myocardial injury. Ann Thorac Surg 1999;68(1):52e6. discussion 6e7.
[86] Jarral OA, Saso S, Harling L, et al. Organ dysfunction in patients with left ventricular impairment: what is the effect of
cardiopulmonary bypass? Heart Lung Circ 2014;23(9):852e62.
[87] Diegeler A, Doll N, Rauch T, et al. Humoral immune response during coronary artery bypass grafting: a comparison of
limited approach, off-pump technique, and conventional cardiopulmonary bypass. Circulation 2000;102(19 Suppl.
3):III95e100.
[88] Wan S, Yim AP, Ari AA, et al. Can cardioplegia management inuence cytokine responses during clinical cardio-
pulmonary bypass? Ann Thorac Cardiovasc Surg Off J Assoc Thorac Cardiovasc Surg Asia 1999;5(2):81e5.
[89] Sievers HH, Freund-Kaas C, Eleftheriadis S, et al. Lung protection during total cardiopulmonary bypass by isolated lung
perfusion: preliminary results of a novel perfusion strategy. Ann Thorac Surg 2002;74(4):1167e72. discussion 72.
[90] Liu Y, Wang Q, Zhu X, et al. Pulmonary artery perfusion with protective solution reduces lung injury after cardio-
pulmonary bypass. Ann Thorac Surg 2000;69(5):1402e7.
[91] Zheng JH, Xu ZW, Wang W, et al. Lung perfusion with oxygenated blood during aortic clamping prevents lung injury.
Asian Cardiovasc Thorac Ann 2004;12(1):58e60.
[92] Goebel U, Siepe M, Mecklenburg A, et al. Reduced pulmonary inammatory response during cardiopulmonary bypass:
effects of combined pulmonary perfusion and carbon monoxide inhalation. Eur J Cardio-thoracic Surg Off J Eur Assoc
Cardio-thoracic Surg 2008;34(6):1165e72.
[93] Siepe M, Goebel U, Mecklenburg A, et al. Pulsatile pulmonary perfusion during cardiopulmonary bypass reduces the
pulmonary inammatory response. Ann Thorac Surg 2008;86(1):115e22.
[94] Drew CE, Anderson IM. Profound hypothermia in cardiac surgery: report of three cases. Lancet 1959;1(7076):748e50.
[95] Glenville B, Ross D. Coronary artery surgery with patient's lungs as oxygenator. Lancet 1986;2(8514):1005e6.
[96] Mendler N, Heimisch W, Schad H. Pulmonary function after biventricular bypass for autologous lung oxygenation. Eur
J Cardio-thoracic Surg Off J Eur Assoc Cardio-thoracic Surg 2000;17(3):325e30.
[97] Richter JA, Meisner H, Tassani P, et al. Drew-Anderson technique attenuates systemic inammatory response syn-
drome and improves respiratory function after coronary artery bypass grafting. Ann Thorac Surg 2000;69(1):77e83.
[98] Suzuki T. Additional lung-protective perfusion techniques during cardiopulmonary bypass. Ann Thorac Cardiovasc
Surg Off J Assoc Thorac Cardiovasc Surg Asia 2010;16(3):150e5.
[99] Ng CS, Ari AA, Wan S, et al. Ventilation during cardiopulmonary bypass: impact on cytokine response and cardio-
pulmonary function. Ann Thorac Surg 2008;85(1):154e62.
*[100] Ng CS, Wan S, Wan IY, et al. Ventilation during cardiopulmonary bypass: impact on neutrophil activation and pul-
monary sequestration. J Invest Surg Off J Acad Surg Res 2009;22(5):333e9.
[101] Durukan AB, Gurbuz HA, Salman N, et al. Ventilation during cardiopulmonary bypass did not attenuate inammatory
response or affect postoperative outcomes. Cardiovasc J Afr 2013;24(6):224e30.
*[102] Apostolakis EE, Koletsis EN, Baikoussis NG, et al. Strategies to prevent intraoperative lung injury during cardiopul-
monary bypass. J Cardiothorac Surg 2010;5:1.
[103] Tutun U, Parlar AI, Altinay L, et al. Does on-pump normothermic beating-heart valve surgery with low tidal volume
ventilation protect the lungs? Heart Surg Forum 2011;14(5):E297e301.
*[104] Schreiber JU, Lance MD, de Korte M, et al. The effect of different lung-protective strategies in patients during car-
diopulmonary bypass: a meta-analysis and semiquantitative review of randomized trials. J Cardiothorac Vasc Anesth
2012;26(3):448e54.
[105] El Azab SR, Rosseel PM, De Lange JJ, et al. Effect of VIMA with sevourane versus TIVA with propofol or midazolam-
sufentanil on the cytokine response during CABG surgery. Eur J Anaesthesiol 2002;19(4):276e82.
[106] Brix-Christensen V, Tonnesen E, Sorensen IJ, et al. Effects of anaesthesia based on high versus low doses of opioids on
the cytokine and acute-phase protein responses in patients undergoing cardiac surgery. Acta Anaesthesiol Scand 1998;
42(1):63e70.
[107] Nader ND, Li CM, Khadra WZ, et al. Anesthetic myocardial protection with sevourane. J Cardiothorac Vasc Anesth
2004;18(3):269e74.
[108] Guarracino F, Landoni G, Tritapepe L, et al. Myocardial damage prevented by volatile anesthetics: a multicenter
randomized controlled study. J Cardiothorac Vasc Anesth 2006;20(4):477e83.
[109] Landoni G, Biondi-Zoccai GG, Zangrillo A, et al. Desurane and sevourane in cardiac surgery: a meta-analysis of
randomized clinical trials. J Cardiothorac Vasc Anesth 2007;21(4):502e11.
[110] Cho EJ, Yoon JH, Hong SJ, et al. The effects of sevourane on systemic and pulmonary inammatory responses after
cardiopulmonary bypass. J Cardiothorac Vasc Anesth 2009;23(5):639e45.
[111] Murphy GS, Sherwani SS, Szokol JW, et al. Small-dose dexamethasone improves quality of recovery scores after
elective cardiac surgery: a randomized, double-blind, placebo-controlled study. J Cardiothorac Vasc Anesth 2011;
25(6):950e60.
[112] Dieleman JM, van Paassen J, van Dijk D, et al. Prophylactic corticosteroids for cardiopulmonary bypass in adults.
Cochrane Database Syst Rev 2011;(5):CD005566.
[113] Dieleman JM, Nierich AP, Rosseel PM, et al. Intraoperative high-dose dexamethasone for cardiac surgery: a randomized
controlled trial. JAMA 2012;308(17):1761e7.
[114] Whitlock R, Teoh K, Vincent J, et al. Rationale and design of the steroids in cardiac surgery trial. Am Heart J 2014;
167(5):660e5.
[115] Morimoto N, Morimoto K, Morimoto Y, et al. Sivelestat attenuates postoperative pulmonary dysfunction after total
arch replacement under deep hypothermia. Eur J Cardio-thoracic Surg Off J Eur Assoc Cardio-thoracic Surg 2008;34(4):
798e804.
[116] Xu CE, Zou CW, Zhang MY, et al. Effects of high-dose ulinastatin on inammatory response and pulmonary function in
patients with type-A aortic dissection after cardiopulmonary bypass under deep hypothermic circulatory arrest.
[117] Lomivorotov VV, Fominskiy EV, Efremov SM, et al. Hypertonic solution decreases extravascular lung water in cardiac
patients undergoing cardiopulmonary bypass surgery. J Cardiothorac Vasc Anesth 2013;27(2):273e82.
[118] Sedrakyan A, Wu A, Sedrakyan G, et al. Aprotinin use in thoracic aortic surgery: safety and outcomes. J Thorac Car-
diovasc Surg 2006;132(4):909e17.
[119] Mathias MA, Tribble CG, Dietz JF, et al. Aprotinin improves pulmonary function during reperfusion in an isolated lung
model. Ann Thorac Surg 2000;70(5):1671e4.

Anaesthesiology
Glycemic control and outcome related to

Steven Thiessen, M.D., Medical Doctor, Anesthesiologist in
training,
Ilse Vanhorebeek, Ph.D., Associate Professor,
Greet Van den Berghe, M.D, Ph.D., Prof. Full Professor *
Clinical Division and Laboratory of Intensive Care Medicine, Department Cellular and Molecular Medicine,
KU Leuven, B-3000 Leuven, Belgium
Keywords:
Perioperative hyperglycemia, aggravated by cardiopulmonary
hyperglycemia
bypass, is associated with adverse outcome in adult and pediatric
critical care
cardiac surgery patients. Whereas hyperglycemia was originally perceived as an
cardiopulmonary bypass adaptive response to surgical stress, it is now clear that glycemic
glycemic control control is a strategy to reduce adverse outcomes after cardiac sur-
intensive insulin therapy gery and cardiopulmonary bypass. The optimal blood glucose target,
hypoglycemia whether or not glycemic control should be initiated already intra-
glucose variability operatively, and whether or not perioperative glucose administra-
tion affects the impact of glycemic control on ischemiaereperfusion
damage remain open questions. Hypoglycemia, the risk of which is
increased with glycemic control, is also associated with adverse
outcomes. However, it remains controversial whether brief episodes
of hypoglycemia, rapidly corrected during glycemic control, have
adverse effects on outcome. This review gives an overview of the
currently available literature on glycemic control during and after
cardiac surgery and focuses on the indicated open questions about
this intervention for this specic patient population.
Hyperglycemia in cardiac surgery patients
Hyperglycemia is a commonly occurring metabolic disturbance in patients who undergo cardiac

surgery, whether or not suffering from preexisting diabetes mellitus [1]. The use of cardiopulmonary
* Corresponding author. Tel.: 32 16 34 40 21; Fax: 32 16 34 40 15.

E-mail address: greet.vandenberghe@med.kuleuven.be (G. Van den Berghe).
178 S. Thiessen et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 177e187
bypass (CPB) for cardiac surgery further disturbs glucose homeostasis and aggravates the hypergly-
cemic response [2]. Although any blood glucose value higher than the normal healthy range during
cardiac surgery could be labeled as hyperglycemia, a clear denition of perioperative hyperglycemia is
currently lacking [3]. However, the American Diabetes Association suggests dening any blood glucose
value above 140 mg/dl (>7.8 mmol/l) in the perioperative phase as perioperative hyperglycemia [4].
Several factors contribute to the hyperglycemic response evoked by cardiac surgery, including the
patient's predisposition, metabolic alterations induced by the injury of the surgery, and concomitant
treatments. Patient's characteristics that predispose to a more severe hyperglycemic response include
the presence of obesity, insulin resistance, hypertension, and an atherogenic lipid prole, which are all
highly prevalent among cardiac surgery patients [5]. The development of a sort of diabetes of injury or
stress diabetes during cardiac surgery is characterized by insulin resistance, which contributes to the
degree of perioperative hyperglycemia [6]. The exact pathophysiology of this stress hyperglycemia is
complex and reviewed extensively elsewhere [6]. Furthermore, several commonly used anesthesiologic
interventions during cardiac surgery may disturb glucose metabolism. These include the administration
of inotropes, vasopressors, glucocorticoids, heparin, beta-blockers, and the infusion of glucose [7].
The hyperglycemic response to surgery was long considered to be adaptive and benecial. This was
based on the assumption that high levels of circulating glucose would fuel the high glucose need of
cells that predominantly rely on glucose as metabolic substrate and that can take up glucose inde-
pendent of insulin, such as neurons, hepatocytes, endothelial cells, and blood cells. This viewpoint,
however, contrasted with the clear association between hyperglycemia and increased morbidity and
mortality as observed for critically ill patients [8e10] and cardiac surgery patients [11e13]. This evi-
dence has recently been summarized in a systematic review [3]. These ndings suggested that cardiac
surgery-induced hyperglycemia, rather than being an adaptive response and reecting the severity of
the surgical procedure or pharmacological interventions, may contribute to complications. This
alternative interpretation provided the rationale for lowering blood glucose concentrations during the
perioperative period in cardiac surgery patients.
Glycemic control in cardiac surgery patients
Several studies have investigated the impact of lowering blood glucose concentrations in cardiac
surgery patients. These are discussed separately for adult and pediatric patients in the following
sections and summarized in Table 1.
Adult cardiac surgery patients
The rst randomized controlled trial (RCT) to investigate the effect of lowering glycemia in the
perioperative period was published in 2001 [14]. In this single-center study, 1548 adult patients were
randomized to receive either intensive insulin therapy (IIT) in which insulin was infused to target blood
glucose to the normal range of 80e110 mg/dl (4.4e6.1 mmol/l) or to conventional therapy with
infusion of insulin only when glucose concentrations exceeded 215 mg/dl (11.9 mmol/l). The latter was
the standard of care at that time for patients admitted to the surgical intensive care unit (ICU). In this
study, 63% of the patients were admitted to the ICU after cardiac surgery. IIT reduced mortality of the
total group of ICU patients, as well as of the cardiac surgery subgroup [15]. In the latter subgroup, the
mortality benet was maintained up to 4 years after hospital discharge. Morbidity of the cardiac
surgery patients was also reduced, most pronounced for patients who stayed more than 3 days in the
ICU, as indicated by earlier weaning from mechanical ventilation, less acute renal failure, reduced
incidence of critical illness polyneuropathy, reduced inammation, and shorter ICU stay [15]. Similar
ndings have been reported in observational studies by Furnary et al. [16] In these studies, 3554 pa-
tients with preexisting diabetes mellitus who underwent coronary artery bypass grafting (CABG) be-
tween 1987 and 2001 were described. These reports showed a 50% decrease in perioperative mortality
and a signicant decrease in the incidence of deep sternal wound infections in patients receiving
continuous insulin infusion with a blood glucose target of 100e150 mg/dl (5.6e8.3 mmol/l). The au-
thors further expanded the patient cohort between 2001 and 2005 to include an additional 1980
patients [17]. In this study, a tighter glycemic control resulted in a decreased incidence of deep sternal
Table 1
Impact of glycemic control in studies that included cardiac surgery patients.
Glycemic control in adult cardiac surgery patients
Van den Berghe et al. Furnary et al. 2006 [17] Lazar et al. 2009 [21]
2001 [14]
Type of study Randomized controlled trial Observational study Randomized controlled trial
Patient population Surgical ICU Cardiac surgery patients Cardiac surgery patients
No. of cardiac 970 (of 1548) 5534 141
surgery patients
% of patients with 16 100 100
history of diabetes
Intervention arm
Blood glucose target 80e110 mg/dl 100e150 mg/dl 125e200 mg/dl
(4.4e6.1 mmol/l) (5.6e8.3 mmol/l) (6.9e11.1 mmol/l)
Therapy i.v. insulin administration i.v. insulin administration s.c. insulin administration
Blood glucose reached$ 103 mg/dl (5.7 mmol/l) 121 mg/dl (6.7 mmol/l) 138 mg/dl (7.7 mmol/l)
Control arm
Blood glucose target Tolerating hyperglycemia up Historical controls Tolerating hyperglycemia
to 215 mg/dl (11.9 mmol/l) up to 250 mg/dl
(13.9 mmol/l)
Blood glucose reached$ 153 mg/dl (8.5 mmol/l) >200 mg/dl (11.1 mmol/l) 260 mg/dl (14.4 mmol/l)
Effect of intervention - Decreased mortality* - Decreased mortality - Decreased mortality
on outcome - Fewer bloodstream - Less atrial brillation - Less atrial brillation
infections - Fewer wound infections - Fewer wound infections
- Decrease in need for renal - Shorter length of hospital - Shorter length of hospital
replacement therapy* stay stay
- Reduced inammatory - Fewer blood transfusions
response
- Fewer blood transfusions
- Shorter length of ICU stay
- Shorter duration of
ventilatory support
- Less hyperbilirubinemia*
- Less polyneuropathy*
Glycemic control in pediatric cardiac surgery patients
Vlasselaers et al. [40] Agus et al. [43] Macrae et al. [46]
Type of study Randomized controlled trial Randomized controlled Randomized controlled

trial trial
Patient population Mixed PICU Cardiac surgery patients Mixed PICU
No. of cardiac 526 (of 700) 980 837 (of 1369)
surgery patients
Intervention arm
Blood glucose target Age-adjusted normoglycemia 80e110 mg/dl 72e126 mg/dl
* Infants: 50e80 mg/dl (4.4e6.1 mmol/l) (4e7 mmol/l)
(2.8e4.4 mmol/l)
* Children: 70e100 mg/dl
(3.9e5.6 mmol/l)
Therapy i.v. insulin administration i.v. insulin administration i.v. insulin administration
Blood glucose reached$ * Infants: 86.4 mg/dl 109 mg/dl (6 mmol/l) 105 mg/dl (5.8 mmol/l)
(4.8 mmol/l)
* Children: 95.4 mg/dl
(5.3 mmol/l)
Control arm
Blood glucose target Tolerating hyperglycemia Standard care (no target Tolerating hyperglycemia
up to 215 mg/dl range, treated according up to 215 mg/dl
(11.9 mmol/l) to the discretion of the (11.9 mmol/l)
physician)
113 mg/dl (6.3 mmol/l)
Blood glucose reached$ * Infants: 115.2 mg/dl 112 mg/dl (6.2 mmol/l)
(6.4 mmol/l)
* Children: 147.6 mg/dl
(8.2 mmol/l)
Effect of intervention - Decreased mortality - No effect on mortality - No effect on mortality*
on outcome - Reduced inammatory - No effect on morbidity - Decrease in need for
response renal replacement
(continued on next page)
Table 1 (continued )
Glycemic control in pediatric cardiac surgery patients
Vlasselaers et al. [40] Agus et al. [43] Macrae et al. [46]
- Lower postoperative therapy*

levels of troponin* - Lower total health-care
- Reduced rate of costs
secondary infections
- Shorter ICU stay
*Effect of the intervention on outcome was also reported for the cardiac surgery patient subgroup. Effect of the intervention on
outcome was also reported for cardiac surgery patients who stayed in ICU for at least a third day. $ Mean blood glucose con-
centration reached; if mean blood glucose concentration was not stated in the article, then mean blood glucose concentration
was estimated from the provided gures. Mean achieved blood glucose concentration during the calendar year 2005.
wound infections, hospital length of stay, blood transfusions, new-onset atrial brillation, and low
cardiac output syndrome [17]. A study by Lazar et al. conrmed these observations [18]. In this study,
141 diabetic patients undergoing CABG were randomized to receive glycemic control with a blood
glucose target of 125e200 mg/dl (6.9e11.1 mmol/l), or standard therapy (blood glucose target
<250 mg/dl or <13.9 mmol/l). The patients randomly allocated to the tighter glycemic control achieved
lower blood glucose levels (138 4 vs. 260 6 mg/dl), which resulted in a reduced incidence of atrial
brillation and wound infections, a shorter postoperative length of hospital stay, and a survival
advantage 2 years after surgery.
However, the enthusiasm about the use of glycemic control in the perioperative period was tempered
by the ndings of the NICE-SUGAR (Normoglycemia in Intensive Care Evaluation and Surviving Using
Glucose Algorithm Regulation) trial [19]. This prospective, multicenter RCT included 6104 general ICU
patients who were randomized to a blood glucose target of 81e108 mg/dl (4.5e6.0 mmol/l) (IIT) or
144e180 mg/dl (8e10 mmol/l). The control group of NICE-SUGAR was thus treated quite differently
from those in the earlier proof-of-concept studies as most patients also received a form of glycemic
control. IIT, as compared with the control group of this study, increased mortality from 24.9% to 27.5%. It
should be noted, however, that only 37% of the studied patients in NICE-SUGAR were surgical ICU pa-
tients, and only few, if any, cardiac surgery patients were included. Hence, the relevance of the NICE-
SUGAR ndings for the cardiac surgery population remains debated. Moreover, this study was further
confounded by a signicant overlap between the glucose concentrations that were achieved in the two
groups and the use of inaccurate blood glucose measurement tools, making it difcult to correctly
interpret the results [20]. Therefore, the standard of care for adult patients who undergo cardiac surgery
with or without CPB still includes careful glycemic control, as advised by current guidelines [21].
Timing and duration of the intervention

Glycemic control in cardiac surgery patients was rst applied only in the postoperative period, that
is, in the ICU, where it showed to reduce morbidity and mortality [14]. However, also the presence of
intraoperative hyperglycemia has been shown to correlate with higher morbidity and mortality
[11,12,22], suggesting that additional intraoperative glycemic control could possibly be benecial.
Several observational studies reported protocols for glycemic control in the intraoperative and post-
operative period, with benecial effects on mortality and morbidity, suggesting that intraoperative
glycemic control is safe and feasible [17,18,23]. Interestingly, the effect of IIT appeared to depend on the
length of time it was applied, with a minimum of 3 days apparently required for an appreciable benet
[14e16]. Hence, it remained an open question whether extending glycemic control to the intra-
operative period, for an additional 3e4 h, provides any additional benet as compared with only
postoperative glycemic control. This question was addressed in a study by Gandhi et al. [24]. This study
showed no additional benet of intraoperative glycemic control, when postoperative glycemic control
is maintained. However, as the study was underpowered to detect a small additional effect, the
question remains unanswered [25]. Therefore, the addition of intraoperative glycemic control, when
postoperative glycemic control is provided, remains experimental and should only be used in centers
with sufcient experience in targeting and achieving normoglycemia during cardiac surgery. Further
studies are needed to establish any role of intraoperative glycemic control during cardiac surgery.
S. Thiessen et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 177e187 181
Blood glucose target range

The exact target for blood glucose concentrations in ICU patients has been the subject of consid-
erable debate. Recent results of large RCTs have clearly shown that a one-size-ts-all policy may not
be applicable for glycemic control in ICUs worldwide. For cardiac surgery patients, current guidelines
advocate keeping blood glucose concentrations below 180 mg/dl (10 mmol/l) in the perioperative
period [21]. However, the randomized controlled study that provided the rst evidence in favor of
blood glucose control after cardiac surgery targeted strict normoglycemia (80e110 mg/dl or
4.4e6.1 mmol/l) and found this to be effective [14]. Whether targeting higher blood glucose levels
(110e180 mg/dl) in cardiac surgery patients is equally effective for survival is unclear, as no adequately
powered RCTs have been done to address this specic question.
Recently, it also has become clear that targeting strict normoglycemia (80e110 mg/dl) in ICU patients
with preexisting diabetes mellitus is not associated with similar benets as was observed for patients
without a history of diabetes and may even be harmful in this population [26]. These ndings suggest that
patients suffering from diabetes who are not perfectly treated to normoglycemia have developed a certain
tolerance or adaptation to a moderate degree of hyperglycemia. Such adaptation may occur at the level of
expression of glucose transporters in different cell types [27,28]. When acutely targeting strict normo-
glycemia in the perioperative period for such patients, this could evoke an acute derangement of this new
homeostasis, with insufcient cellular glucose uptake as a consequence, which could be harmful. Results
of several studies support this possibility for diabetic patients undergoing cardiac surgery. Furnary et al.
reported a benet from targeting mild hyperglycemia instead of strict normoglycemia for patients with
diabetes mellitus [16], and in the Leuven surgical ICU study, there was a clear benet when targeting strict
normoglycemia in predominantly nondiabetic patients, whereas those with preexisting diabetes mellitus
did not appear to benet [14]. In the light of these ndings, it may be useful to obtain hemoglobin A1c
(HbA1c), a glycosylated hemoglobin, prior to cardiac surgery [21]. HbA1c, an indicator of the average level
of glycemia over the previous 2e3 months, has been shown to be frequently elevated in patients, with or
without known diabetes mellitus, undergoing cardiac surgery [29]. HbA1c has been shown to tightly
correlate with average blood glucose levels and may therefore be used to decide on the optimal target for
perioperative blood glucose control [30]. Further research is warranted to investigate these open ques-
tions, in a prospective randomized controlled way. Since such studies are currently not available yet, it
may be wise to tolerate mild hyperglycemia in patients with a history of diabetes mellitus, more spe-
cically to the level of blood glucose these patients were used to have before they underwent surgery, and
to only target strict normoglycemia for patients without a history of diabetes.
Nutritional support
One of the major risks of cardiac surgery, with or without the use of CPB, is the impact of transient
ischemiaereperfusion on the myocardium. More than 50 years ago, Dr. Sodi-Pallares suggested a
metabolic strategy to protect the myocardium from the deleterious effects of ischemiaereperfusion
[31]. This strategy comprised the infusion of glucose, insulin, and potassium, which was called GIK
therapy. The rationale for this intervention was to shift substrate utilization in the ischemic
myocardium from fatty acids to glucose for anaerobic glycolysis, whereby reducing oxygen con-
sumption in turn reduces ischemiaereperfusion damage in the myocardium. Furthermore, the
increased inux of potassium in the myocardium by means of insulin and glucose infusion would
reduce the risk of malignant arrhythmias and insulin could mediate cardioprotection [32].
Several large trials addressing the use of GIK therapy during cardiac surgery were conducted,
yielding contradictory results [33]. These studies were, however, obscured as different GIK cocktails
were used in the different studies and blood glucose was unaffected, resulting in profound hyper-
glycemia in some studies. Therefore, a negative effect of the concomitant hyperglycemia may have
overruled any possible protective effect of GIK therapy [34]. This question was addressed by an elegant
study by Carvalho et al. [35]. In this study, patients were randomly allocated to a group receiving GIK
therapy while targeting normoglycemia, called GIN therapy (N stands for normoglycemia), or to a
group receiving standard care. GIN therapy clearly protected the myocardium during cardiac surgery,
suggesting that normoglycemia may be the crucial factor of GIK/GIN therapy.
One could question whether the infusion of glucose is in fact required to obtain the cardioprotection
from targeting normoglycemia with insulin infusion [34]. The infusion of glucose might even be
deleterious, as macronutrients and insulin are potent inhibitors of autophagy, a crucial cellular
housekeeping machinery responsible for the removal of toxic protein aggregates and damaged or-
ganelles [36]. Recent ndings of a large RCT investigating the effect of early parenteral nutrition in
patients admitted to the ICU, referred to as the EPaNIC (Early Parenteral Nutrition Completing Enteral
Nutrition in Adult Critically Ill Patients) study, may shed some light on this question [37]. In this study,
the effect of early parenteral supplementation of insufcient enteral feeding on morbidity and mortality
of ICU patients was examined. In one study arm, a 20% glucose infusion was administered on the
admission day and the day after, resulting in a total energy intake of about 400 kcal on day 1 and 800 kcal
on day 2. From the morning of day 3 onwards, all-in-one parenteral nutrition was started, when
necessary, to reach the caloric goal. Patients who were randomized to the other study arm only received
5% glucose solution in an equal volume to that of the early-initiation group to provide adequate hy-
dration, and for them the severe macronutrient decit that accumulated over the rst week was
accepted. In this group, the all-in-one parenteral nutrition was only initiated beyond day 8 and only if
enteral nutrition was still insufcient at that time. In both study arms, normoglycemia was maintained.
Of the 4640 included patients, 63% were cardiac surgery patients. Withholding early parenteral nutri-
tion had no effect on mortality, but decreased the rate of new ICU infections, lowered the incidence of
cholestastis, and reduced the duration of mechanical ventilation, of renal replacement therapy and of
ICU stay, as well as the incidence of clinically relevant muscle weakness [37,38]. This study clearly
showed that early provision of macronutrients while maintaining normoglycemia has no benecial
effects and even causes excess harm. Therefore, it appears wise not to provide high doses of glucose
while preserving normoglycemia with insulin infusion in cardiac surgery patients. Further research is
required to investigate specically whether or not perioperative administration of glucose, when nor-
moglycemia is maintained, results in myocardial protection against ischemiaereperfusion injury.
Pediatric cardiac surgery patients
Hyperglycemia is also prevalent in children undergoing cardiac surgery, and also in this patient
population the degree of hyperglycemia has been associated with adverse outcome [39]. Whether
glycemic control in pediatric cardiac surgery patients during the perioperative period is benecial, as
shown in the adult population, was rst studied by Vlasselaers et al. [40]. In this study, 700 critically ill
patients were randomized to receive IIT to obtain age-adjusted normoglycemia (50e80 mg/dl
(2.8e4.4 mmol/l) in infants and 70e100 mg/dl (3.9e5.6 mmol/l) in children) or to tolerating hyper-
glycemia up to 215 mg/dl (11.9 mmol/l). The studied population comprised a predominantly surgical
population, with 75% of the patients being included after cardiac surgery for congenital heart defects. IIT
in this pediatric ICU population reduced the inammatory response, lowered the postoperative levels of
troponin and heart-type fatty acid binding protein, reduced the rate of secondary infections, shortened
ICU stay, and improved ICU survival. IIT also increased the occurrence of hypoglycemia, but this did not
have a negative effect on the acute outcomes nor on the neurocognitive development at 4 years follow-
up [40,41]. In fact, IIT had a positive effect on cognitive executive functions after 4 years, such as motor
coordination and cognitive exibility [41]. The same investigators showed that targeting age-adjusted
normoglycemia during and after cardiac surgery in neonates protected the myocardium and reduced
the inammatory response [42]. Recently, two other RCTs on glycemic control in pediatric cardiac
surgery patients were performed. The rst study, the SPECS (Safe Pediatric Euglycemia in Cardiac Sur-
gery) study, was a two-center RCT which included 980 children after cardiac surgery with CPB, most of
them younger than 1 year [43]. The RCT targeted blood glucose concentrations of 80e110 mg/dl
(4.4e6.1 mmol/l), a level much higher than the age-adjusted normoglycemia range for the studied
children, in one group as compared with virtually no glucose management in the other group and did
not nd an effect on infections or mortality. However, important differences between this study and the
rst pediatric RCT complicate interpretation of these ndings [44]. First, the blood glucose target in the
intervention arm (80e110 mg/dl, reached in only about 50% of the patients) of the SPECS study was not
normal for age and therefore age-adjusted normoglycemia was not achieved. In fact, the majority of
the children in the control group spontaneously reached the targeted blood glucose level of the inter-
vention arm, which reected hyperglycemia for that age group, and therefore blood glucose concen-
trations in the two study arms were only slightly and very transiently different, a difference that was not
clinically relevant. Interestingly, in a post hoc analysis the investigators of the SPECS trial did nd a
reduced incidence of infections with targeting lower, but not normal for age, glycemia in older patients
[45]. The second study, the CHiP trial (Control of Hyperglycaemia in Pediatric Intensive Care), was a
multicenter study that included 1369 critically ill children of which 60% had undergone cardiac surgery.
Patients were randomized to achieve glycemic control targeting blood glucose to 72e126 mg/dl
(4e7 mmol/l), a blood glucose range that was again higher than the normal range for the age group, or to
tolerating glycemia up to 215 mg/dl (11.9 mmol/l) [46]. This study again showed very minor effects on
blood glucose concentrations, with transient differences being smaller than the error of the measure-
ment tools, and, not unexpectedly, found no effect on mortality. Nevertheless, randomization to gly-
cemic control lowered the length of stay in the hospital and reduced the incidence of kidney failure.
Again, due to the chosen target for blood glucose in this trial, the expected effect size was an over-
estimation and hence the study was not statistically powered to detect any benet from such a small
difference in blood glucose [47]. In conclusion, current evidence still supports careful targeting of blood
glucose levels that are normal for age in the perioperative setting in pediatric cardiac surgery patients.
Further research is needed to investigate several other aspects of glycemic control in pediatric patients.
These comprise, among others, the optimal duration of treatment needed to obtain benets as well as
the role of concomitant nutritional support.
Hypoglycemia and glycemic variability
Treating hyperglycemia in the ICU with insulin inherently increases the incidence of hypoglycemia,
as was uniformly shown in all RCTs that studied this intervention [14,19,48]. Whether this increased
incidence of hypoglycemia is detrimental has been a topic of considerable debate. Whereas it is
commonly accepted that severe and prolonged hypoglycemia increases morbidity and mortality, this is
not clear for short-lasting, iatrogenic hypoglycemia in the ICU [4]. In the rst Leuven study of surgical
ICU patients, including many cardiac surgery patients, the incidence of hypoglycemia in the IIT group
rose from 0.8% to 5.1% [14]. However, hypoglycemia was never associated with immediate mortality in
this study. Interestingly, a recent retrospective cohort study showed that only spontaneous hypogly-
cemia, but not iatrogenic hypoglycemia, is associated with an increased mortality risk [49]. These
ndings suggest that hypoglycemia may be more a sign of severity of illness, but not necessarily an
inducer of harm, as long as hypoglycemia is short lasting. This viewpoint is supported by other studies.
Quantication of circulating markers of neuronal and astrocyte damage in the Leuven pediatric ICU
study showed that patients experiencing a brief hypoglycemic episode had elevated levels of these
markers already before the hypoglycemic event, with no increase in these markers evoked by the
hypoglycemic event, making it unlikely that hypoglycemia caused damage to these cells [50]. A recent
retrospective study of cardiac surgery patients receiving IIT showed that patients who developed
hypoglycemia also had an increased risk of respiratory complications, and prolonged ICU and hospital
stay, which again might indicate that hypoglycemia occurs in the sicker patients [51]. Interestingly, in
this study hypoglycemia was not associated with mortality. The most convincing evidence that iat-
rogenic short-lasting hypoglycemia did not induce excess harm was generated by the long-term
follow-up of the children included in the Leuven pediatric ICU study [41]. In this study, randomiza-
tion to IIT resulted in an increased incidence of hypoglycemia below 40 mg/dl from 1% to 25%, but this
did not result in a worse score on any measure of cognitive function at 4 years follow-up. Actually, the
IIT group scored better for motor coordination and cognitive exibility. Even though short-lasting
hypoglycemia most likely did not induce harm in this study, this may be due to the fact that all
measures were taken to quickly and adequately correct it. In order to prevent hypoglycemia when
targeting normoglycemia, and thereby increasing the safety of IIT, it is obligatory to use an intravenous
insulin drip and to measure the glycemia frequently with appropriate measuring tools, as was recently
acknowledged in a consensus statement [52]. Extending intervals for blood glucose measurement
beyond 2 h increases the risk of serious hypoglycemic events. Therefore, caution is advocated when
increasing the interval. Recently, several risk factors contributing to hypoglycemia during cardiac
surgery have been identied. These included female gender, preexisting diabetes mellitus, the appli-
cation of hemodialysis, and intraoperative blood product transfusion [51]. Therefore, it may be wise to
monitor glycemia more frequently in these patients undergoing cardiac surgery.
Another point of interest is glucose variability. Irrespective of blood glucose control, glucose vari-
ability is closely associated with the risk of death in critically ill patients [53e56] and in cardiac surgery
patients [11]. Factors contributing to glycemic variability include intrinsic and extrinsic patient factors.
For example, patients with a premorbid deranged glucose homeostasis are more at risk of experiencing
uctuating glucose levels [57]. Examples of extrinsic factors are the mode of insulin administration
(higher glycemic variability with the use of subcutaneous insulin administration), nutritional support,
and the applied glycemic control algorithm. It is possible that glycemic variability could have
contributed to the different outcomes of the studies on glycemic control during/after cardiac surgery,
as it has been shown that protocols differ signicantly in their ability to keep patients within the
desired range [58]. Furthermore, recent evidence points out that hyperglycemia after hypoglycemia
may be more detrimental rather than the hypoglycemic event itself [59]. The fact that in the Leuven
studies, also in the critically ill children, rebound hyperglycemia after hypoglycemia was carefully
avoided may be a very important safety aspect in this regard [40].
In conclusion, whereas severe and prolonged hypoglycemia is considered to be detrimental and
should be avoided, current evidence points out that the occurrence of mild and short-lasting hypo-
glycemia during glycemic control in the ICU does not have a detrimental effect on outcome in the
context of careful and frequent monitoring and with use of performant control algorithms. Further
research is needed to identify the role of glycemic variability in cardiac surgery patients and to identify
new techniques to minimize the rate of hypoglycemic events and high glycemic variability, such as
continuous glucose measurement tools, the use of a validated blood glucose control algorithm [60], and
possibly also the use of drugs other than insulin to lower blood glucose.
Summary
Based on the available evidence, careful and effective glycemic control, in combination with
delaying any parenteral nutrition to beyond the rst postoperative week, can be advised to prevent
additional metabolic damage to patients undergoing cardiac surgery, with or without CPB. Currently,
evidence is lacking for additional intraoperative glycemic control if postoperative glycemic control is
provided. For patients without a history of diabetes mellitus, age-adjusted normoglycemia is probably
the most effective blood glucose target range. However, when the logistics and the experience to
achieve safe glycemic control are not available, tolerating mild hyperglycemia up to 180 mg/dl and
delaying any parenteral nutrition may be a defendable option, based on common sense. For patients
with a history of diabetes mellitus, it may be better to tolerate mild hyperglycemia, instead of targeting
strict normoglycemia. More research on glycemic control for cardiac surgery patients, with or without
CPB, is necessary in order to further optimize the care for these patients.
Practice points
Pronounced hyperglycemia in the perioperative period of patients undergoing cardiac sur-

gery is detrimental and should be prevented with insulin treatment.
In order to apply glycemic control in cardiac surgery patients, one should measure blood
glucose concentrations frequently with appropriate measuring tools and use a well-validated
guideline/algorithm for insulin titration.
A one-size-fits-all blood glucose target range may not apply for all cardiac surgery patients.
For patients without a history of diabetes mellitus, current evidence supports targeting
normoglycemia (80e110 mg/dl) without the use of early parenteral nutrition. If appropriate
logistics are not available to do this in a safe way, accepting somewhat higher blood glucose
levels is to be advised.
For patients with diabetes mellitus, current evidence supports a somewhat higher target
range of blood glucose control in the perioperative setting.
Research agenda
The role of glucose supplementation, when normoglycemia is provided, in preventing

myocardial ischemiaereperfusion damage during cardiac surgery should be further
investigated.
Several aspects of glycemic control in the pediatric population, such as optimal timing,
duration, and the effect of combination with nutritional support, should be examined.
The role of techniques to minimize the rate of hypoglycemic events and high glycemic
variability, such as continuous glucose measurement tools, the use of validated blood
glucose control algorithms, and possibly also the use of drugs other than insulin to lower
blood glucose, should be defined.
Conict of interest
None.
Acknowledgments
ST received an FWO Research Assistant Fellowship (Aspirant). GVdB and IV, through the University
of Leuven, receive structural research nancing via the Methusalem program, funded by the Flemish
Government (METH14/06). GVdB holds a European Research Council Advanced Grant (AdvG-2012-
321670) from the Ideas Programme of the EU FP7. The study sponsors had no involvement in literature
selection and writing of the manuscript.
References
[1] Carvalho G, Moore A, Qizilbash B, et al. Maintenance of normoglycemia during cardiac surgery. Anesth Analg 2004;99(2):
319e24.
[2] Knapik P, Nadziakiewicz P, Urbanska E, et al. Cardiopulmonary bypass increases postoperative glycemia and insulin
consumption after coronary surgery. Ann Thorac Surg 2009;87(6):1859e65.
[3] Giakoumidakis K, Nenekidis I, Brokalaki H. The correlation between peri-operative hyperglycemia and mortality in
cardiac surgery patients: a systematic review. Eur J Cardiovasc Nurs 2012;11(1):105e13.
[4] Moghissi ES, Korytkowski MT, DiNardo M, et al. American Association of Clinical Endocrinologists and American Diabetes
Association consensus statement on inpatient glycemic control. Diabetes Care 2009;32(6):1119e31.
[5] Donatelli F, Cavagna P, Di DG, et al. Correlation between pre-operative metabolic syndrome and persistent blood glucose
elevation during cardiac surgery in non-diabetic patients. Acta Anaesthesiol Scand 2008;52(8):1103e10.
[6] Vanhorebeek I, Langouche L. Molecular mechanisms behind clinical benets of intensive insulin therapy during critical
illness: glucose versus insulin. Best Pract Res Clin Anaesthesiol 2009;23(4):449e59.
[7] Barth E, Albuszies G, Baumgart K, et al. Glucose metabolism and catecholamines. Crit Care Med 2007;35(9):508e18.
[8] Whitcomb BW, Pradhan EK, Pittas AG, et al. Impact of admission hyperglycemia on hospital mortality in various intensive
care unit populations. Crit Care Med 2005;33(12):2772e7.
[9] Umpierrez GE, Isaacs SD, Bazargan N, et al. Hyperglycemia: an independent marker of in-hospital mortality in patients
with undiagnosed diabetes. J Clin Endocrinol Metab 2002;87(3):978e82.
[10] Krinsley JS. Association between hyperglycemia and increased hospital mortality in a heterogeneous population of
critically ill patients. Mayo Clin Proc 2003;78(12):1471e8.
[11] Duncan AE, Abd-Elsayed A, Maheshwari A, et al. Role of intraoperative and postoperative blood glucose concentrations in
predicting outcomes after cardiac surgery. Anesthesiology 2010;112(4):860e71.
[12] Ouattara A, Lecomte P, Le MY, et al. Poor intraoperative blood glucose control is associated with a worsened hospital
outcome after cardiac surgery in diabetic patients. Anesthesiology 2005;103(4):687e94.
[13] Furnary AP, Wu Y. Eliminating the diabetic disadvantage: the Portland Diabetic Project. Semin Thorac Cardiovasc Surg
2006;18(4):302e8.
*[14] Van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy in critically ill patients. N Engl J Med 2001;
345(19):1359e67.
*[15] Ingels C, Debaveye Y, Milants I, et al. Strict blood glucose control with insulin during intensive care after cardiac surgery:
impact on 4-years survival, dependency on medical care, and quality-of-life. Eur Heart J 2006;27(22):2716e24.
[16] Furnary AP, Gao G, Grunkemeier GL, et al. Continuous insulin infusion reduces mortality in patients with diabetes un-
dergoing coronary artery bypass grafting. J Thorac Cardiovasc Surg 2003;125(5):1007e21.
*[17] Furnary AP, Wu Y. Clinical effects of hyperglycemia in the cardiac surgery population: the Portland Diabetic Project.
Endocr Pract 2006;12:22e6.
[18] Lazar HL, Chipkin SR, Fitzgerald CA, et al. Tight glycemic control in diabetic coronary artery bypass graft patients im-
proves perioperative outcomes and decreases recurrent ischemic events. Circulation 2004;109(12):1497e502.
[19] Finfer S, Chittock DR, Su SY, et al. Intensive versus conventional glucose control in critically ill patients. N Engl J Med
2009;360(13):1283e97.
[20] Van den Berghe G, Schetz M, Vlasselaers D, et al. Clinical review: intensive insulin therapy in critically ill patients: NICE-
SUGAR or Leuven blood glucose target? J Clin Endocrinol Metab 2009;94(9):3163e70.
*[21] Lazar HL, McDonnell M, Chipkin SR, et al. The Society of Thoracic Surgeons practice guideline series: blood glucose
management during adult cardiac surgery. Ann Thorac Surg 2009;87(2):663e9.
[22] Doenst T, Wijeysundera D, Karkouti K, et al. Hyperglycemia during cardiopulmonary bypass is an independent risk factor
for mortality in patients undergoing cardiac surgery. J Thorac Cardiovasc Surg 2005;130(4):1144.
[23] Lecomte P, Foubert L, Coddens J, et al. Management of tight intraoperative glycemic control during off-pump coronary
artery bypass surgery in diabetic and nondiabetic patients. J Cardiothorac Vasc Anesth 2011;25(6):937e42.
[24] Gandhi GY, Nuttall GA, Abel MD, et al. Intensive intraoperative insulin therapy versus conventional glucose management
during cardiac surgery: a randomized trial. Ann Intern Med 2007;146(4):233e43.
[25] Van den Berghe G. Does tight blood glucose control during cardiac surgery improve patient outcome? Ann Intern Med
2007;146(4):307e8.
[26] Krinsley JS, Egi M, Kiss A, et al. Diabetic status and the relation of the three domains of glycemic control to mortality in
critically ill patients: an international multicenter cohort study. Crit Care 2013;17(2):R37.
[27] Brosius FC, Heilig CW. Glucose transporters in diabetic nephropathy. Pediatr Nephrol 2005;20(4):447e51.
[28] Kahn BB, Rosen AS, Bak JF, et al. Expression of GLUT1 and GLUT4 glucose transporters in skeletal muscle of humans with
insulin-dependent diabetes mellitus: regulatory effects of metabolic factors. J Clin Endocrinol Metab 1992;74(5):1101e9.
[29] Engoren M, Habib R, Zacharias A, et al. The prevalence of elevated hemoglobin A1c in patients undergoing coronary
artery bypass surgery. J Cardiothorac Surg 2008;3:63.
[30] Nathan D, Kuenen J, Borg R, et al. Translating the A1c assay into estimated average glucose values. Diabetes Care 2008;31:
1473e8.
[31] Sodi-Pallares D, Testelli MR, Fischleder BL, et al. Effects of an intravenous infusion of a potassium-glucose-insulin so-
lution on the electrocardiographic signs of myocardial infarction. A preliminary clinical report. Am J Cardiol 1962;9:
166e81.
[32] Sack MN, Yellon DM. Insulin therapy as an adjunct to reperfusion after acute coronary ischemia: a proposed direct
myocardial cell survival effect independent of metabolic modulation. J Am Coll Cardiol 2003;41(8):1404e7.
[33] Fan Y, Zhang AM, Xiao YB, et al. Glucose-insulin-potassium therapy in adult patients undergoing cardiac surgery: a meta-
analysis. Eur J Cardiothorac Surg 2011;40(1):192e9.
[34] Van den Berghe G. Coronary bypass surgery: protective effects of insulin or of prevention of hyperglycemia, or both?
J Clin Endocrinol Metab 2011;96(5):1272e5.
*[35] Carvalho G, Pelletier P, Albacker T, et al. Cardioprotective effects of glucose and insulin administration while maintaining
normoglycemia (GIN therapy) in patients undergoing coronary artery bypass grafting. J Clin Endocrinol Metab 2011;
96(5):1469e77.
[36] Yang Z, Klionsky DJ. Eaten alive: a history of macroautophagy. Nat Cell Biol 2010;12(9):814e22.
*[37] Casaer MP, Mesotten D, Hermans G, et al. Early versus late parenteral nutrition in critically ill adults. N Engl J Med 2011;
365(6):506e17.
[38] Hermans G, Casaer MP, Clerckx B, et al. Effect of tolerating macronutrient decit on the development of intensive-care
unit acquired weakness: a subanalysis of the EPaNIC trial. Lancet Respir Med 2013;1(8):621e9.
[39] Srinivasan V, Spinella PC, Drott HR, et al. Association of timing, duration, and intensity of hyperglycemia with intensive
care unit mortality in critically ill children. Pediatr Crit Care Med 2004;5(4):329e36.
*[40] Vlasselaers D, Milants I, Desmet L, et al. Intensive insulin therapy for patients in paediatric intensive care: a prospective,
randomised controlled study. Lancet 2009;373(9663):547e56.
*[41] Mesotten D, Gielen M, Sterken C, et al. Neurocognitive development of children 4 years after critical illness and treatment
with tight glucose control: a randomized controlled trial. JAMA 2012;308(16):1641e50.
[42] Vlasselaers D, Mesotten D, Langouche L, et al. Tight glycemic control protects the myocardium and reduces inammation
in neonatal heart surgery. Ann Thorac Surg 2010;90(1):22e9.
*[43] Agus MS, Steil GM, Wypij D, et al. Tight glycemic control versus standard care after pediatric cardiac surgery. N Engl J Med
2012;367(13):1208e19.
[44] Gielen M, Vlasselaers D, Van den Berghe G. Glucose in the ICU-evidence, guidelines, and outcomes. N Engl J Med 2012;
367(25):2451e2.
[45] Agus MS, Asaro LA, Steil GM, et al. Tight glycemic control after pediatric cardiac surgery in high-risk patient populations:
a secondary analysis of the safe pediatric euglycemia after cardiac surgery trial. Circulation 2014;129(22):2297e304.
*[46] Macrae D, Tasker RC, Elbourne D. A trial of hyperglycemic control in pediatric intensive care. N Engl J Med 2014;370(14):
1355e6.
[47] Van den Berghe G, Mesotten D. A trial of hyperglycemic control in pediatric intensive care. N Engl J Med 2014;370(14):
1354e5.
[48] Brunkhorst FM, Engel C, Bloos F, et al. Intensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J
Med 2008;358(2):125e39.
[49] Kosiborod M, Inzucchi SE, Goyal A, et al. Relationship between spontaneous and iatrogenic hypoglycemia and mortality
in patients hospitalized with acute myocardial infarction. JAMA 2009;301(15):1556e64.
[50] Vanhorebeek I, Gielen M, Boussemaere M, et al. Glucose dysregulation and neurological injury biomarkers in critically ill
children. J Clin Endocrinol Metab 2010;95(10):4669e79.
[51] Stamou SC, Nussbaum M, Carew JD, et al. Hypoglycemia with intensive insulin therapy after cardiac surgery: predisposing
factors and association with mortality. J Thorac Cardiovasc Surg 2011;142(1):166e73.
[52] Finfer S, Wernerman J, Preiser JC, et al. Clinical review: consensus recommendations on measurement of blood glucose
and reporting glycemic control in critically ill adults. Crit Care 2013;17(3):229.
[53] Al-Dorzi HM, Tamim HM, Arabi YM. Glycaemic uctuation predicts mortality in critically ill patients. Anaesth Intensive
Care 2010;38(4):695e702.
[54] Ali NA, O'Brien JM, Dungan K, et al. Glucose variability and mortality in patients with sepsis. Crit Care Med 2008;36(8):
2316e21.
[55] Dossett LA, Cao H, Mowery NT, et al. Blood glucose variability is associated with mortality in the surgical intensive care
unit. Am Surg 2008;74(8):679e85.
[56] Hermanides J, Vriesendorp TM, Bosman RJ, et al. Glucose variability is associated with intensive care unit mortality. Crit
Care Med 2010;38(3):838e42.
[57] Subramaniam B, Lerner A, Novack V, et al. Increased glycemic variability in patients with elevated preoperative HbA1C
predicts adverse outcomes following coronary artery bypass grafting surgery. Anesth Analg 2014;118(2):277e87.
[58] Chase JG, Le Compte AJ, Suhaimi F, et al. Tight glycemic control in critical careethe leading role of insulin sensitivity and
patient variability: a review and model-based analysis. Comput Methods Programs Biomed 2011;102(2):156e71.
[59] Suh SW, Hamby AM, Swanson RA. Hypoglycemia, brain energetics, and hypoglycemic neuronal death. Glia 2007;55(12):
1280e6.
[60] Van Herpe T, Mesotten D, Wouters PJ, et al. LOGIC-insulin algorithm-guided versus nurse-directed blood glucose control
during critical illness: the LOGIC-1 single-center, randomized, controlled clinical trial. Diabetes Care 2013;36(2):188e94.

Anaesthesiology
Anticoagulation management associated with

Roman M. Sniecinski, MD, Associate Professor of
Anesthesiology, Cardiothoracic Anesthesia Fellowship
Director a, *,
Jerrold H. Levy, MD, Professor of Anesthesiology/Associate
Professor of Surgery, CoDirector Cardiothoracic ICU b, 1
a
Emory University School of Medicine, Department of Anesthesiology, 1364 Clifton Rd, NE, Atlanta, GA
30322, USA
b
Cardiothoracic Anesthesia and Critical Care, Duke University Medical Center, 2301 Erwin Road, 5691H
HAFS, Durham, NC 27710, USA
Keywords:
The use of extracorporeal circulation requires anticoagulation to
maintain blood uidity throughout the circuit, and to prevent
unfractionated heparin
whole-blood-activated coagulation time thrombotic complications. Additionally, adequate suppression of
heparin resistance hemostatic activation avoids the unnecessary consumption of
antithrombin III coagulation factors caused by the contact of blood with foreign
surfaces. Cardiopulmonary bypass represents the greatest chal-
lenge in this regard, necessitating profound levels of anti-
coagulation during its conduct, but also quick, efcient reversal of
this state once the surgical procedure is completed. Although
extracorporeal circulation has been around for more than half a
century, many questions remain regarding how to best achieve
anticoagulation for it. Although unfractionated heparin is the
predominant agent used for cardiopulmonary bypass, the amount
required and how best to monitor its effects are still unresolved.
This review discusses the use of heparin, novel anticoagulants, and
the monitoring of anticoagulation during the conduct of cardio-
pulmonary bypass.
* Corresponding author. Tel.: 1 404 558 7307; Fax: 1 678 620 2634.
E-mail addresses: roman.sniecinski@emoryhealthcare.org (R.M. Sniecinski), JERROLD.LEVY@duke.edu (J.H. Levy).
1
Tel.: 1 919 684 0862.
190 R.M. Sniecinski, J.H. Levy / Best Practice & Research Clinical Anaesthesiology 29 (2015) 189e202
The development of anticoagulation for cardiopulmonary bypass
While the birth of cardiopulmonary bypass (CPB) is widely celebrated as 6 May 1953, when Dr. John
Gibbon successfully utilized it during the closure of an atrial septal defect, it is less appreciated that
extracorporeal circulation for organs was actually conceived >100 years earlier by the French physi-
ologist Julien-Jean Cesar le Gallois [1]. One of the greatest obstacles faced by le Gallois and other early
developers of CPB was the profound activation of the coagulation system, resulting in clotting of cir-
cuits and thrombosis of vessels. Debrinated blood was used in conjunction with bubble-type oxy-
genators as early as 1856, but reliable systemic anticoagulation remained elusive [2]. It was not until
the discovery of heparin by Howell and McLean in 1916, and really only following the development of
high-yield manufacturing processes in the late 1930s, that anticoagulation for CPB became feasible [3].
Early protocols for heparin administration were simply weight-based with little attempt at moni-
toring its effects. In 1957, the Mayo Clinic described prevention of the coagulation of blood for the
Gibbon-type pump oxygenator, prescribing 3 mg of heparin (i.e., 300 units) per kilogram of patient
bodyweight [4]. The activated clotting time (ACT) of whole blood was reported by Hattersley in 1966
[5], and this eventually led to a practical point-of-care testing that could be imported into the operating
room for monitoring the effects of heparin. Bull et al. popularized the concept of the safe zone for
anticoagulation during CPB in the mid-1970s, dening this as an ACT between 300 and 600 s [6].
It is intuitively evident that clot formation in CPB circuits can lead to devastating complications, but
preventing thrombi is actually only one of the goals of anticoagulation. A variety of mechanisms lead to
thrombin production during CPB, which can activate and consume critical hemostatic components
including platelets, brinogen, and other coagulation factors [7]. Somewhat paradoxically, inadequate
anticoagulation on CPB can lead to a profound coagulopathy, likely due to disseminated intravascular
coagulation (DIC), upon its termination. This review attempts to summarize the knowledge gained over
the past 40 years, and to guide clinicians with respect to the practice of anticoagulation for adults on
CPB.
Anticoagulation agents
Despite its age, which is older than the US Food and Drug Administration (FDA), heparin remains
the anticoagulant of choice for extracorporeal circulation. It can achieve profound levels of anti-
coagulation, is inexpensive, and it has a readily available antidote in protamine. Unfortunately, it also
has signicant limitations that have led to the search for alternate agents. Clinicians today are faced
with an increasing array of choices for providing anticoagulation for CPB. While most are reserved for
situations where heparin cannot be used, it is important to understand their potential benets and
limitations. Table 1 summarizes the dosing information of the below agents that have been used in the
setting of CPB.
Unfractionated heparin
Unfractionated heparin (UFH) is a heterogenous mixture of sulfated polysaccharides prepared from

animal tissues. The primary source has changed over the years from canine liver to bovine lung, and it is
now predominantly porcine intestine. The extraction and purication processes result in a mixture of
polysaccharide chains of 5e30 kDa in weight, of which 20e50% contain the specic pentasaccharide-
binding site for the protein antithrombin (AT) [8]. The primary mechanism of UFH's anticoagulant
effect is largely indirect, in that it accelerates AT's endogenous inhibition of factor IIa (thrombin) and
factor Xa by >1000-fold each. The ratio of thrombin to Xa inhibition for UFH is assumed to be 1:1.
Importantly, the neutralization of thrombin by AT requires heparin chains of higher molecular weight,
containing at least 18 saccharide units. Low molecular weight heparins (LMWHs), provided they
contain the specic pentasaccharide sequence, can still bind to AT and accelerate its inhibition of Xa,
but inhibition of thrombin is severely reduced [8]. Factors IXa, XIa, and XIIa are also inactivated by the
heparineAT complex, although to a much lesser degree [9]. Once AT binds to the coagulation factor, the
heparin molecule is released and available to bind to other targets.
R.M. Sniecinski, J.H. Levy / Best Practice & Research Clinical Anaesthesiology 29 (2015) 189e202 191
Table 1
Summary of anticoagulants for cardiopulmonary bypass.
Agent Dosing Monitoring Comments
Unfractionated Bolus: 300e500 U/kg ACT Wide variety of dosing and monitoring used
heparin CPB Prime: up to Heparin concentration
10,000 units (Hepcon)
High-dose thrombin
time
Danaparoid (27) Bolus: 125 U/kg Anti-Xa activity Lower incidence of HIT cross-reactivity
CPB Prime: 3 U/ml Bleeding complications due to long T1/2
of priming uid Not recommended if other alternates available
Infusion: 7 U/kg/h
Lepirudin (32) Bolus: 0.25 mg/kg ECT No longer manufactured
CPB Prime: 0.2 mg/kg Desirudin similar, but CPB data not available
in priming uid
Infusion: 0.5 mg/min
Bivalirudin (35) Bolus: 1 mg/kg ECT ACT target 2.5 times baseline measurement
CPB Prime: 50 mg ACT Additional boluses 0.1e0.5 mg/kg as needed
in priming uid Hemoconcentration reduces drug levels on CPB
Infusion: 2.5 mg/kg/h
Argatroban (38) Bolus: 0.1 mg/kg ACT Infusion adjusted to keep ACT >400 s
Infusion: 5 mg/kg/min ECT Extensive blood product use when ACT > 600 s
ACT, activated clotting time; HIT, heparin-induced thrombocytopenia; ECT, ecarin clotting time.
Although the majority of UFH's anticoagulation effect comes from its interaction with AT, there are
some AT-independent effects as well. These interactions are particularly important with respect to clot-
bound thrombin, as AT-mediated inhibition is severely reduced when thrombin is attached to a brin
molecule [10]. This is the primary reason that low-dose heparin infusions fail to stop the propagation of
venous clot. In 1981, Tollefsen et al. isolated a second heparin-dependent thrombin inhibitor that, since
AT was occasionally referred to as heparin cofactor, they termed heparin cofactor II (HCII) [11]. The
exact function of HCII remains largely unknown, although it has been postulated to play a role in
hemostasis during pregnancy [12]. The dermatan sulfate Intimatan, an HCII agonist, has been shown to
provide anticoagulation independent of AT [13], but clinical studies in the setting of CPB are lacking.
Heparin has also been shown to induce endothelial cells to secrete tissue factor pathway inhibitor
(TFPI) [14]. TFPI inhibits tissue factorefactor VIIa complexes formed as part of the extrinsic pathway.
The majority of TFPI becomes bound to plasma proteins, and its signicance in cardiac surgery is
unclear. The mechanisms of UFH's inhibitory actions against thrombin are summarized in Fig. 1.
The pharmacokinetics of UFH is complex and nonlinear [15]. There are two mechanisms for its
elimination: one is quick, but saturable, and the other is non-saturable, but prolonged. Following an
intravenous (IV) bolus, a portion of UFH binds to various plasma proteins, is taken up by macrophages,
and is internalized by endothelial cells. This represents the rapid clearance of the drug, but it is
overwhelmed by doses exceeding 100 U/kg [16,17]. Renal elimination is the non-saturable pathway,
with larger molecular weight chains ltered more quickly than smaller ones. IV administration of a
bolus of UFH results in peak onset of anticoagulant effects within 2e3 min [18], and the half-life in-
creases signicantly with larger doses. A dose of 100 U/kg has a half-life of approximately 1 h, whereas
a 400 U/kg dose has one of 2.5 h [19].
In addition to anticoagulant effects on thrombin, heparin is also well known for activating platelets.
The most feared complication of this is heparin-induced thrombocytopenia (HIT), which has been
extensively reviewed elsewhere [20,21]. Briey, the disorder is a result of antibody production against
heparin complexed to platelet factor 4 (PF4) on the surface of platelets. These antibodies then cross-
link platelets via the Fc receptor, activate them, and begin the coagulation process. However, not all
heparin effects on platelets are immune-mediated, but rather caused by the heparin molecule's
negative charge. Nonimmune interactions (the so-called HIT type I) are thought to occur via platelet
IIb/IIIa receptors, decreasing the platelet's threshold for activation [22,23]. This often results in a small
decrease in platelet count during the rst few days of an infusion of UFH, which is not typically clin-
ically relevant. As the doses of UFH used for CPB are signicantly higher, it is possible that these
Fig. 1. Anticoagulant effects of UFH. This gure shows a condensed version of the coagulation cascade, including the tissue factor
(TF) pathway (i.e., extrinsic pathway) and the contact activation pathway (i.e., intrinsic pathway). The major actions of UFH consist of
forming a complex with antithrombin (AT) to inhibit factor Xa and factor IIa (thrombin) as denoted by the large dotted lines. The
smaller dashed lines indicate other anticoagulant effects including heparineAT complex inhibition of factors IXa, XIa and XIIa. UFH
effects independent of AT include complexing with heparin cofactor II (HCII) to inhibit IIa, as well as promoting the release of tissue
factor pathway inhibitor (TFPI) from the endothelium. HMWK, high molecular weight kininogen.
nonimmune interactions could contribute to coagulopathy seen following cardiac surgery. However,
given that the conduct of CPB itself causes platelet dysfunction, it is difcult to isolate these effects in
the operating room.
Low molecular weight heparins
As their name suggests, LWMHs include the smaller chains of UFH e typically <8 kDa. Rather than
actually fractionating UFH, however, most are produced by depolymerization and cleavage of the
longer chains [24]. As previously mentioned, these shorter chains combine with AT to inhibit factor Xa,
but they have a reduced capacity to cause AT-mediated thrombin inhibition. LMWHs are often char-
acterized by a ratio of anti-Xa to anti-IIa activity, which can range from about 1.5 up to 10 [24]. LWMHs
have fewer platelet interactions than UFH, although they can still produce HIT antibodies. Because they
undergo predominantly renal elimination, rather than via macrophages, LWMHs tend to have a more
predictable, albeit signicantly longer, duration of action than UFH.
LMWHs bind to fewer plasma proteins, and they have better bioavailability than UFH, but the main
drawback to their use during CPB is that they are only partially neutralized with protamine. In the
largest case series substituting LMWH for UFH during CPB, Massonnet-Castel et al. reported signicant
bleeding in almost half of the patients [25]. The major reason was that protamine administration failed
to fully reverse the LMWH anti-Xa effects. Given the possibility of signicant postoperative bleeding,
combined with the fact that there is still notable cross-reactivity with HIT antibodies, the use of LMWH
does not seem to offer any signicant benet over UFH for use during CPB.
Danaparoid (heparinoids)
Some glycosaminoglycans have heparin-like activities, and they are termed heparinoids since
they do not contain the actual heparin chains. Danaparoid (Orgaran, Merck, QC, Canada) is a mixture
of such molecules prepared from porcine intestine with an average molecular weight of 6 kDa [26]. It is
composed mainly of heparan sulfate, with smaller amounts of dermatan and chondroitin sulfates. It is
the heparan sulfate molecules that have afnity for AT and that produce anti-Xa activity. Thrombin
inhibition also does occur, as do other anticoagulation effects, but the mechanism is not well
understood. Danaparoid is primarily eliminated via the kidneys, and it exerts anti-Xa effects for >24 h.
A potential advantage of danaparoid is that it has a much lower, although not zero, cross-reactivity
with HIT antibodies compared to LMWHs [26]. As such, in the past it was used in patients with HIT for
anticoagulation on CPB. However, since it is not neutralized with protamine and has a long half-life, the
incidence of major postoperative bleeding complications approaches 50% [27]. Danaparoid is no longer
available in most countries.
Fondaparinux (pentasaccharides)
Unlike UFH, LMWHs, or heparinoid compounds, fondaparinux (Arixtra, GlaxcoSmithKline, NC,

USA) is a completely homogenous mixture of an oligosaccharide [28]. It contains the same critical
pentasaccharide sequence as heparin which interacts with AT and produces only anti-Xa inhibition.
Although fondaparinux can initiate the formation of anti-PF4 antibodies, it does not seem to support
platelet activation [29]. Despite this, its long half-life (~20 h) and the inability to be neutralized by
protamine preclude it from being useful for anticoagulation for CPB. However, its use has been reported
in a patient on postoperative extracorporeal membrane oxygenation (ECMO) [30].
Direct thrombin inhibitors
Hirudin is a small (<7-kDa) peptide found in the salivary glands of Hirudo medicinalis, the medicinal
leech, which directly binds to thrombin and inhibits its ability to convert brinogen to insoluble brin.
It can bind plasma or clot-bound thrombin, and it does not require AT or any other intermediary to
produce its anticoagulant effects. Another advantage of hirudin is that it does not induce the pro-
duction of HIT antibodies. Unfortunately, there is no known antidote to reverse its effects. All direct
thrombin inhibitors (DTIs) are either recombinant derivatives or synthetic analogs of hirudin [31].
Lepirudin (Reudan, Bayer Healthcare, NJ, USA) was the rst recombinant hirudin to be used for
the purposes of CPB [32]. It is a very potent inhibitor of thrombin and binds it irreversibly. Lepirudin is
cleared via renal excretion with a half-life following IV administration of about 90 min. This is obvi-
ously signicantly prolonged in renal failure. Hemodialysis and plasmapharesis can be used to reduce
plasma levels to a limited extent [33]. Although the production of lepirudin was stopped in 2012,
desirudin (Iprivask, Marathon Pharmaceuticals, IL, USA) is a very similar recombinant molecule. It is
available for venous thromboembolism prophylaxis, but experience in percutaneous coronary in-
terventions is limited, and its during CPB is non-existent.
Bivalirudin (Angiomax, The Medicines Company, NJ, USA) is a synthetic DTI which, unlike lepir-
udin or desirudin, binds reversibly to thrombin. It has a half-life of 25 min, but since 80% is eliminated
via renal excretion, this is prolonged in kidney disease [34]. With two multicenter trials to evaluate its
safety and efcacy for anticoagulation during CPB (EVOLUTION-ON and CHOOSE-ON) [35,36], biva-
lirudin has the best-studied dosing regimen, albeit off-label, of any alternate anticoagulant. Because of
its reversible binding to thrombin, however, special care must be used to avoid stagnant blood in the
surgical eld and CPB circuit in order to avoid clot formation.
Argatroban is a synthetic DTI, which, like bivalirudin, binds reversibly to thrombin, but has less
potency [37]. Unlike bivalirudin, argatroban's primary route of elimination is via the liver, making it
potentially attractive in patients with renal dysfunction. Experience with argatroban and CPB is limited,
but dosing schedules have been proposed [38]. Unfortunately, levels seem to have a particularly
inconsistent relationship with the ACT, and signicant clots within the CPB circuit have been reported
even with values >480 s [39]. The American College of Chest Physician recommendations for HIT
management do not include argatroban in conjunction with CPB [40].
Monitoring anticoagulation during CPB
Biological variability and a rapidly changing clinical environment dictate the need for anti-
coagulation monitoring during the conduct of CPB. Conceptually, the ideal test would have a direct
relationship with the plasma level of the drug and an appropriate sensitivity. Although the activated
partial thromboplastin time (aPTT) has long been a standard for monitoring infusions of UFH, it is not
useful in the setting of CPB. This is because the aPTT has a semi-logarithmic relationship with UFH, and
the test becomes unclottable at the high doses of UFH employed during cardiac surgery and other
interventions. Currently used monitors fall into two categories: those based upon measuring clotting
times, and those based upon measuring heparin levels. Each approach has its own limitations.
The ACT
Hattersley rst described the ACT in 1966 as a screening test for coagulation defects in presurgical
patients [5]. It is interesting that, although he reported it appeared to reect the degree of heparin-
emia, the test was not originally developed as an UFH monitoring tool. However, the ACT was rela-
tively easy to perform, and thus it formed the cornerstone of Bull's heparin monitoring strategy that
was popularized in the mid-1970s [6]. Four decades later, instead of visually inspecting glass tubes,
photo-optical and electromagnetic systems are employed to detect the presence of clot, but the un-
derlying principles remain the same.
The ACT begins by taking whole blood and stimulating the contact system with an activator e
typically kaolin, but some systems use celite or glass beads [41]. The resulting stimulation of the
intrinsic pathway of the coagulation cascade leads to thrombin production, brinogen cleavage,
platelet aggregation, and nally thrombus formation. Ideally, the length of time for clot formation is a
direct reection of the effects of UFH. By using whole blood, the invitro conditions should reect the
invivo environment, taking into account the various biological variables such as AT levels and the
amount of heparin-binding proteins. This is the key advantage of clotting-time-based tests. Unfortu-
nately, because the end point of the ACT is the formation of clot, there are several confounders that will
cause it to be prolonged independent of UFH effects (Fig. 2).
Although the ACT's correlation with the actual plasma heparin concentration is good following the
initial bolus of UFH, once CPB has commenced, there is signicant divergence [42]. This disconnect is
likely due to hemodilution, hypothermia, and platelet dysfunction, and this may worsen with longer
CPB times. Supplementing the patient sample with exogenous normal plasma in a 1:1 ratio has been
shown to provide a better correlation with heparin levels [43]. However, the practice of supplementing
ACTs has not been widely adopted, and it is typically reserved for patients with special factor
deciencies.
High-dose thrombin time
Unlike the ACT and aPTT, which both begin coagulation by stimulating the intrinsic pathway, the
thrombin time (TT) begins further down the cascade by directly adding thrombin to platelet-poor
plasma. Test values reect the time required for brinogen to be directly cleaved to form a brin
clot. It is extremely sensitive to inhibitors of thrombin (i.e., UFH). Much like the partial thromboplastin
time (PTT), however, the TT becomes unclottable at UFH doses used during CPB. The high-dose
Fig. 2. Factors prolonging the ACT independent of heparin effects. Low coagulation factors, low platelets, and low brinogen can all
prolong the ACT. Additional confounders that cause the ACT to not correlate well with heparin levels include hypothermia (less a
problem when small sample sizes are utilized), excess protamine, and the presence of anti-phospholipid antibodies (typically found
in patients with lupus).
thrombin time (HiTT) follows the same principle as the TT, but uses a much higher dose of thrombin. It
has good correlation with plasma heparin concentrations in the 1.5e4.8-U/ml range, and it is less
affected by hemodilution than the ACT [44]. Because the test clots too quickly when there is little
inhibition of thrombin present, the HiTT is not useful for baseline measurements, nor for detecting low
levels of unneutralized heparin after CPB (i.e., heparin rebound).
Ecarin clotting time
Ecarin is an enzyme isolated from the snake Echis carinatus that converts prothrombin to meizo-
thrombin. Like thrombin, meizothrombin converts brinogen to brin, although it is less potent.
Meizothrombin is rapidly inhibited by DTIs, but not UFH. The ecarin clotting time (ECT) works similarly
to the TT, and it has a relatively linear relationship with DTIs that has been found to correlate better
with plasma DTI concentration than the ACT [45]. Unfortunately, the point-of-care ECT tests are not
widely available in the United States. The addition of ecarin to thromboelastometry devices has been
reported, and this may represent an alternate method for monitoring DTIs in the operating room
[46,47].
Heparin concentration
In order to inhibit clot-bound thrombin, heparin plasma levels of at least 2.0 U/ml are required [10].
This forms the basis of the argument for monitoring heparin concentration, as the ACT does not
correlate well with heparin levels during hypothermic CPB. Despite ACTs > 400 s, the heparin plasma
levels can fall below critical levels due to hemodilution, binding to articial surfaces, or to a variety of
other factors, thus leading to unnecessary consumption of coagulation proteins.
The Hepcon HMS system (Medtronic Inc., Minneapolis, MN, USA) is currently the most widely
available point-of-care system for measuring heparin levels. The device uses the protamine titration
method, automatically dispensing a sample of the patient's whole blood into a cartridge containing
different channels with thromboplastin and a known concentration of protamine. The channel with
the amount of protamine that most closely matches the amount of heparin in the sample will clot the
quickest. As protamine neutralizes heparin at approximately 1:1 ratio (1 mg protamine per 100 U
heparin), the amount of heparin in the sample can be calculated from the known quantity of prot-
amine. The levels obtained correlate well with laboratory anti-Xa values (below text) [48]. Critics are
quick to point out, however, that the protamine titration method is not a functional test, and this may
not accurately reect the specic anticoagulation happening in vivo. The Hepcon system also includes a
test to estimate heparin bolus dose requirements, the heparin dose response (HDR), but this has been
shown to correlate poorly with achieving target ACTs [49].
Although not used in the operating room, it is worth mentioning that chromogenic assays for
heparin concentration are widely used as the gold standard in laboratory monitoring of heparin levels.
These tests rely on the chromophore p-nitroanaline (p-NA) for photometric detection at 405 nm [50].
Factor Xa is the reagent used to cleave the substrate molecule, releasing p-NA for detection. The reagent
is rst added to the patient sample; the more heparin (and AT) in the sample, the less factor Xa is active.
When the substrate is then added to the patient sample, the optical density will be inversely pro-
portional to the heparin level. Unlike protamine titration, this method does provide a functional
assessment of heparin-mediated inhibition, provided exogenous AT is not added. Unfortunately, the
current chromogenic anti-Xa assays require platelet poor plasma, which limits its ability to be used as a
point-of-care test.
Altered heparin responsiveness
Altered heparin responsiveness is essentially a function of using clotting times e almost always the
ACT e to determine the adequacy of anticoagulation. Loosely dened, it means that the test values
following heparin administration are signicantly different from those normally obtained for any given
dose (Fig. 3A). Reduced heparin sensitivity, commonly termed heparin resistance, refers to a lower
than expected ACT following an IV bolus, and it is particularly problematic when it is uncovered
immediately prior to instituting CPB. A universal denition for heparin resistance in cardiac surgery is
lacking, in part because there is no widespread agreement upon what a normal ACT for CPB should
be.
The minimum ACT target
As previously stated, it was not until almost 20 years after the birth of CPB that the concept of an
adequate level of anticoagulation was quantied. The safe zone was originally dened by Bull et al. as
an ACT between 300 and 600 s [6]. The suggested oor of adequate anticoagulation was raised to an
ACT of 400 s in 1978 by Young et al. based upon the electron micrographs of oxygenators from CPB
circuits used in rhesus monkeys [51]. The issue of the lowest acceptable ACT became further clouded in
the 1990s with the widespread use of the antibrinolytic aprotinin, which had variable effects on ACT
values depending on the activator used [52]. While aprotinin stopped being marketed in the United
States in 2007, the US Pharmacopoeia changed its heparin monograph in 2008 [53]. The result was a
decrease in UFH potency by about 10%, which resulted in lower ACT values seen by some clinicians [54].
It is not surprising, then, that the minimal ACT deemed acceptable to initiate CPB varies widely among
North American institutions, and it ranges from <350 s to >500 s [55].
Fig. 3. Doseeresponse graph for initial heparin bolus as measured by the ACT. (A) Prior to commencing cardiopulmonary bypass, the
relationship of an IV heparin bolus and the resulting ACT is relatively linear. As discussed in the text, this relationship does not
continue during CPB. (B) The heparin sensitivity index (HSI) is the doseeresponse slope. It is calculated by subtracting the baseline
ACT measurement from the ACT obtained after the loading dose (both measured in seconds), and dividing that by the loading dose of
heparin given (in units/kg).
In truth, the ideal ACT for CPB initiation is not known. For constructing individual heparin
doseeresponse curves, Bull et al. recommended targeting 480 s [56]. Similarly, the majority of North
American centers continue to target a value between 400 and 480 s [55]. Unfortunately, the evidence
for this practice is based upon limited studies using older CPB circuits with much larger prime volumes.
Proponents of targeting a lower ACT will point to the development of heparin-bonded circuits and
increased biocompatibility as reasons to lower the need for such profound systemic anticoagulation.
However, the activation of the hemostatic system occurs beyond the CPB circuit; it also takes place in
the surgical eld, with the use of cardiotomy suction, and inside the venous reservoir [7]. In a 2009
meta-analysis of biocompatible CPB circuits, Ranucci et al. concluded that they had little impact on
clinical outcomes, and this should probably not be relied upon by themselves to reduce thrombin
generation on CPB [57]. Since there are no controlled studies comparing one ACT target to another, and
likely never will be, the required ACT value prior to commencing CPB will likely remain an institutional
decision.
Heparin resistance
The overall incidence of heparin resistance in cardiac surgery is 4e26%, depending on the heparin-
loading dose and the delineated target ACT [58]. Although a common denition is the failure to achieve
an ACT 480 s after 500 U/kg of UFH [59,60], these values are somewhat arbitrary, as discussed in the
preceding section. One way to account for institutional variability in loading doses and ACT targets is to
look at the heparin-sensitivity index (Fig. 3B). Although there is no widely agreed upon cutoff for a
normal heparin sensitivity index (HSI), the vast majority of studies regarding heparin resistance
treatment include patients with an HSI < 0.75e0.9 s/U kg [61]. A key question that remains to be
answered is whether a low HSI is clinically relevant, or whether this can simply be overcome by
extreme doses of UFH. Given that large doses of UFH have the undesirable side effects of platelet in-
hibition [62], heparin rebound [63], and the formation of HIT antibodies [21], it is probably worthwhile
to consider a ceiling dose for UFH. Unfortunately, there is no consensus as to what that value might
be.
A variety of risk factors for heparin resistance have been reported (Table 2) [59,64e66]. As heparin
exerts its anticoagulant effects via AT, most studies on treating heparin resistance have focused on
administering exogenous AT with either fresh-frozen plasma or AT concentrate. There are, however,
other causes, and low preoperative AT levels by themselves do not always accurately predict when
heparin resistance will be encountered [67]. Nevertheless, replacing AT to physiological or supra-
physiological levels has consistently been shown to help restore normal heparin responsiveness
Table 2
Risk factors for heparin resistance.
Clinical factors Hematological factors
Prior heparin administration (UFH or LMWH) Low AT levels (<60%)

Preoperative use of intra-aortic balloon pump Thrombocytosis (platelet count 300,000/ml)
Preoperative use of nitroglycerine infusions High levels of platelet factor 4 (PF4)
Disseminated intravascular coagulation (DIC) Elevated factor VIII levels
Endocarditis Elevated binding proteins (e.g., bronectin, lipoproteins, etc.)
Impaired liver synthetic function
Age 65 years
[60,68e70]. Other approaches that have been utilized to achieve an adequate ACT are adding a DTI, a
platelet inhibitor such as tiroban, or nafamostatmesilate (a non-specic protease inhibitor) [71,72].
There is limited experience with these alternate approaches, however, and current guidelines
recommend AT supplementation [73].
Management of UFH during CPB
Following the institution of CPB, plasma heparin concentration falls due to hemodilution from
1500 ml or so within the circuit. To overcome this, it is common practice to add 5000e10,000 units of
UFH to the pump prime. As the CPB run progresses, the amount of heparin further decreases as it is
metabolized and eliminated. The rate of this decrease is heavily dependent upon the initial loading
dose, with larger boluses having a longer duration of action than the smaller ones. It is the management
of this fall in plasma heparin concentration that remains a matter of much debate.
In 1995, Despotis et al. randomly assigned 254 patients to either conventional management of
heparin dosing on CPB using ACTs or the maintenance of a target heparin concentration as determined
by the Hepcon system [74]. Targeting the heparin concentration resulted in both larger doses of UFH
being administered and the need for fewer hemostatic blood products. One explanation for this is that
higher heparin concentrations on CPB more effectively suppress the hemostatic response, leading to
better preservation of coagulation factors [75]. This concept is not universally accepted, however, and
other studies have shown no difference between using ACTs and heparin levels to guide UFH re-dosing
on CPB [76,77].
A key difference in studies regarding UFH management during CPB is time. For CPB runs in the range
of 75e90 min, a typical 300e400-U/kg UFH bolus will still have signicant activity throughout the
procedure. As the run goes beyond 2 h (CPB time averaged 145 min in the above Despotis study), there
is a much greater chance that the remaining amount of UFH will be inadequate to achieve full thrombin
suppression. Additionally, the standard ACT becomes a poor reection of heparin activity during long,
complex cardiac surgical cases. Clinicians need to keep these factors in mind when planning their
anticoagulation strategy. If plasma heparin levels are not monitored, empiric re-dosing of UFH should
be based upon UFH pharmacokinetics. Fig. 4 provides such an example of the overall UFH management
for CPB.
Summary
The use of CPB requires profound anticoagulation, in order to both prevent thrombus formation
within the circuit and patient, and preserve coagulation factors to prevent life-threatening bleeding.
The ideal agent to achieve this would have a rapid onset, a predictable response, and be readily
reversed following the end of CPB. Such an agent does not currently exist. UFH comes the closest to this
ideal, but its need for cofactors and its biological variability make it a complex drug. More studies are
required to determine optimal dosing and monitoring techniques. Alternatively, new anticoagulants
are available which have the potential to overcome many of the limitations of UFH. Experience with
them has been limited, however, and until a reversal agent is discovered, they are unlikely to supplant
UFH as the mainstay for CPB use.
Fig. 4. Sample heparinization protocol utilizing ACT. The actual UFH doses and ACT targets may be adjusted, but the algorithm
contained in the chart recommends treatment for heparin resistance when HSI is <0.75. Additionally, the maintenance of plasma
heparin levels is advocated in CPB runs exceeding 2.5 h. The time to additional boluses may need to be shortened if the initial UFH
bolus is <400 U/kg since the expected half-life of smaller doses is shorter.
Practice points
UFH is the standard anticoagulant for CPB, but it has limitations including complex phar-
macokinetics, the need for AT as a cofactor, and platelet-activating properties.
The ACT is the standard test used to monitor the effects of UFH while on CPB, but it is
affected by a number of factors, and this can lead clinicians to believe that UFH plasma levels
are higher than they actually are.
While no universal definition for altered heparin responsiveness (aka heparin resistance)
exists, clinicians should consider adjuvants for anticoagulation, including AT supplemen-
tation, when it is encountered in order to better preserve coagulation factors while on CPB.
Alternatives to UFH are available, but clinicians need to be aware of their lack of any reversal
agent and the problems associated with monitoring their effects.
Research agenda
The ideal test (i.e., a functional clotting time vs. a heparin plasma level) to monitor anti-
coagulation during CPB remains controversial; it may vary depending upon the patient
population and the length of surgical procedure.
The clinical consequences of altered heparin responsiveness need to be further investigated,
and best practices for treatment need to be developed.
Reversal agents and point-of-care monitoring tests need to be developed for alternate an-
ticoagulants, particularly DTIs.
Funding
No funds were provided for this review.
Disclosures
Dr. Sniecinski receives research support from Grifols and Shire-Viropharma. He is a member of an
advisory board for Grifols.
Dr. Levy is a member of the research steering committees for CSL Behring, Boehringer-Ingelheim,
Grifols, and Janssen.
References
[1] Hessel 2nd EA. A brief history of cardiopulmonary bypass. Semin Cardiothorac Vasc Anesth 2014;18:87e100.
[2] Lim MW. The history of extracorporeal oxygenators. Anaesthesia 2006;61:984e95.
[3] Wardrop D, Keeling D. The story of the discovery of heparin and warfarin. Br J Haematol 2008;141:757e63.
[4] Patrick RT, Theye RA, Moftt EA. Studies in extracorporeal circulation. V. Anesthesia and supportive care during intra-
cardiac surgery with the Gibbon-type pump-oxygenator. Anesthesiology 1957;18:673e85.
[5] Hattersley PG. Activated coagulation time of whole blood. JAMA 1966;196:436e40.
*[6] Bull BS, Korpman RA, Huse WM, et al. Heparin therapy during extracorporeal circulation. I. Problems inherent in existing
heparin protocols. J Thorac Cardiovasc Surg 1975;69:674e84.
[7] Sniecinski RM, Chandler WL. Activation of the hemostatic system during cardiopulmonary bypass. Anesth Analg 2011;
113:1319e33.
[8] Linhardt RJ, Gunay NS. Production and chemical processing of low molecular weight heparins. Semin Thromb Hemost
1999;25(Suppl. 3):5e16.
*[9] Hirsh J, Anand SS, Halperin JL, et al. Guide to anticoagulant therapy: heparin: a statement for healthcare professionals
from the American Heart Association. Circulation 2001;103:2994e3018.
[10] Weitz JI, Hudoba M, Massel D, et al. Clot-bound thrombin is protected from inhibition by heparineantithrombin III but is
susceptible to inactivation by antithrombin III-independent inhibitors. J Clin Investig 1990;86:385e91.
[11] Tollefsen DM, Majerus DW, Blank MK. Heparin cofactor II. Purication and properties of a heparin-dependent inhibitor of
thrombin in human plasma. J Biol Chem 1982;257:2162e9.
[12] Giri TK, Tollefsen DM. Placental dermatan sulfate: isolation, anticoagulant activity, and association with heparin cofactor
II. Blood 2006;107:2753e8.
[13] Tanaka KA, Szlam F, Vinten-Johansen J, et al. Effects of antithrombin and heparin cofactor II levels on anticoagulation with
Intimatan. Thromb Haemost 2005;94:808e13.
[14] Hansen JB, Svensson B, Olsen R, et al. Heparin induces synthesis and secretion of tissue factor pathway inhibitor from
endothelial cells in vitro. Thromb Haemost 2000;83:937e43.
[15] de Swart CA, Nijmeyer B, Roelofs JM, et al. Kinetics of intravenously administered heparin in normal humans. Blood 1982;
60:1251e8.
[16] Lijnen HR, Hoylaerts M, Collen D. Heparin binding properties of human histidine-rich glycoprotein. Mechanism and role
in the neutralization of heparin in plasma. J Biol Chem 1983;258:3803e8.
[17] de Swart CA, Nijmeyer B, Andersson LO, et al. Elimination of high afnity heparin fractions and their anticoagulant and
lipase activity. Blood 1984;63:836e42.
[18] Gravlee GP, Angert KC, Tucker WY, et al. Early anticoagulation peak and rapid distribution after intravenous heparin.
[19] Boneu B, Caranobe C, Sie P. Pharmacokinetics of heparin and low molecular weight heparin. Bailliere's Clin Haematol
1990;3:531e44.
*[20] Warkentin TE, Greinacher A. Heparin-induced thrombocytopenia and cardiac surgery. Ann Thorac Surg 2003;76:638e48.
[21] Levy JH, Winkler AM. Heparin-induced thrombocytopenia and cardiac surgery. Curr Opin Anaesthesiol 2010;23:74e9.
[22] Gao C, Boylan B, Fang J, et al. Heparin promotes platelet responsiveness by potentiating alphaIIbbeta3-mediated outside-
in signaling. Blood 2011;117:4946e52.
[23] Xiao Z, Theroux P. Platelet activation with unfractionated heparin at therapeutic concentrations and comparisons with a
low-molecular-weight heparin and with a direct thrombin inhibitor. Circulation 1998;97:251e6.
[24] Gray E, Mulloy B, Barrowcliffe TW. Heparin and low-molecular-weight heparin. Thromb Haemost 2008;99:807e18.
[25] Massonnet-Castel S, Pelissier E, Bara L, et al. Partial reversal of low molecular weight heparin (PK 10169) anti-Xa activity
by protamine sulfate: in vitro and in vivo study during cardiac surgery with extracorporeal circulation. Haemostasis
1986;16:139e46.
[26] Wilhelm MJ, Schmid C, Kececioglu D, et al. Cardiopulmonary bypass in patients with heparin-induced thrombocytopenia
using Org 10172. Ann Thorac Surg 1996;61:920e4.
[27] Magnani HN, Gallus A. Heparin-induced thrombocytopenia (HIT). A report of 1,478 clinical outcomes of patients treated
with danaparoid (Orgaran) from 1982 to mid-2004. Thromb Haemost 2006;95:967e81.
*[28] Hoppensteadt D, Walenga JM, Fareed J, et al. Heparin, low-molecular-weight heparins, and heparin pentasaccharide:
basic and clinical differentiation. Hematology/Oncol Clin N Am 2003;17:313e41.
[29] Warkentin TE, Cook RJ, Marder VJ, et al. Anti-platelet factor 4/heparin antibodies in orthopedic surgery patients receiving
antithrombotic prophylaxis with fondaparinux or enoxaparin. Blood 2005;106:3791e6.
[30] Parlar AI, Sayar U, Cevirme D, et al. Successful use of fondaparinux in a patient with heparin-induced thrombocytopenia
while on extracorporeal membrane oxygenation after mitral valve redo surgery. Int J Artif Organs 2014;37:344e7.
*[31] Winkler AM, Tormey CA. Pathology consultation on monitoring direct thrombin inhibitors and overcoming their effects
in bleeding patients. Am J Clin Pathol 2013;140:610e22.
[32] Becker RC. Hirudin-based anticoagulant strategies for patients with suspected heparin-induced thrombocytopenia un-
dergoing percutaneous coronary interventions and bypass grafting. J Thromb Thrombolysis 2000;10(Suppl. 1):59e68.
[33] Willey ML, de Denus S, Spinler SA. Removal of lepirudin, a recombinant hirudin, by hemodialysis, hemoltration, or
plasmapheresis. Pharmacotherapy 2002;22:492e9.
[34] Van De Car DA, Rao SV, Ohman EM. Bivalirudin: a review of the pharmacology and clinical application. Expert Rev
Cardiovasc Ther 2010;8:1673e81.
[35] Dyke CM, Smedira NG, Koster A, et al. A comparison of bivalirudin to heparin with protamine reversal in patients un-
dergoing cardiac surgery with cardiopulmonary bypass: the EVOLUTION-ON study. J Thorac Cardiovasc Surg 2006;131:
533e9.
[36] Koster A, Dyke CM, Aldea G, et al. Bivalirudin during cardiopulmonary bypass in patients with previous or acute heparin-
induced thrombocytopenia and heparin antibodies: results of the CHOOSE-ON trial. Ann Thorac Surg 2007;83:572e7.
[37] Dang CH, Durkalski VL, Nappi JM. Evaluation of treatment with direct thrombin inhibitors in patients with heparin-
induced thrombocytopenia. Pharmacotherapy 2006;26:461e8.
[38] Martin ME, Kloecker GH, Laber DA. Argatroban for anticoagulation during cardiac surgery. Eur J Haematol 2007;78:
161e6.
[39] Follis F, Filippone G, Montalbano G, et al. Argatroban as a substitute of heparin during cardiopulmonary bypass: a safe
alternative? Interact Cardiovasc Thorac Surg 2010;10:592e6.
*[40] Warkentin TE, Greinacher A, Koster A, et al. Treatment and prevention of heparin-induced thrombocytopenia: American
College of Chest Physicians evidence-based clinical practice guidelines (8th edition). Chest 2008;133:340Se80S.
[41] Spinler SA, Wittkowsky AK, Nutescu EA, et al. Anticoagulation monitoring part 2: unfractionated heparin and low-
molecular-weight heparin. Ann Pharmacother 2005;39:1275e85.
[42] Despotis GJ, Summereld AL, Joist JH, et al. Comparison of activated coagulation time and whole blood heparin mea-
surements with laboratory plasma anti-Xa heparin concentration in patients having cardiac operations. J Thorac Car-
diovasc Surg 1994;108:1076e82.
[43] Koster A, Despotis G, Gruendel M, et al. The plasma supplemented modied activated clotting time for monitoring of
heparinization during cardiopulmonary bypass: a pilot investigation. Anesth Analg 2002;95:26e30. table of contents.
[44] Wang JS, Lin CY, Karp RB. Comparison of high-dose thrombin time with activated clotting time for monitoring of anti-
coagulant effects of heparin in cardiac surgical patients. Anesth Analg 1994;79:9e13.
[45] Casserly IP, Kereiakes DJ, Gray WA, et al. Point-of-care ecarin clotting time versus activated clotting time in correlation
with bivalirudin concentration. Thromb Res 2004;113:115e21.
[46] Koster A, Buz S, Krabatsch T, et al. Monitoring of bivalirudin anticoagulation during and after cardiopulmonary bypass
using an ecarin-activated TEG system. J Cardiac Surg 2008;23:321e3.
[47] Schaden E, Schober A, Hacker S, et al. Ecarin modied rotational thromboelastometry: a point-of-care applicable alter-
native to monitor the direct thrombin inhibitor argatroban. Wien Klin 2013;125:156e9.
[48] Despotis GJ, Joist JH, Goodnough LT, et al. Whole blood heparin concentration measurements by automated protamine
titration agree with plasma anti-Xa measurements. J Thorac Cardiovasc Surg 1997;113:611e3.
[49] Garvin S, FitzGerald DC, Despotis G, et al. Heparin concentration-based anticoagulation for cardiac surgery fails to reliably
predict heparin bolus dose requirements. Anesth Analg 2010;111:849e55.
[50] Wool GD, Lu CM. Pathology consultation on anticoagulation monitoring: factor X-related assays. Am J Clin Pathol 2013;
140:623e34.
[51] Young JA, Kisker CT, Doty DB. Adequate anticoagulation during cardiopulmonary bypass determined by activated clotting
time and the appearance of brin monomer. Ann Thorac Surg 1978;26:231e40.
[52] Wang JS, Lin CY, Hung WT, et al. Monitoring of heparin-induced anticoagulation with kaolin-activated clotting time in
cardiac surgical patients treated with aprotinin. Anesthesiology 1992;77:1080e4.
[53] Smythe MA, Nutescu EA, Wittkowsky AK. Changes in the USP heparin monograph and implications for clinicians.
Pharmacotherapy 2010;30:428e31.
[54] Anderson DA, Holt DW. Has the new USP assay for heparin affected dosage for patients undergoing cardiopulmonary
bypass? J Extra-Corpor Technol 2013;45:112e5.
[55] Lobato RL, Despotis GJ, Levy JH, et al. Anticoagulation management during cardiopulmonary bypass: a survey of 54 North
American institutions. J Thorac Cardiovasc Surg 2010;139:1665e6.
*[56] Bull BS, Huse WM, Brauer FS, et al. Heparin therapy during extracorporeal circulation. II. The use of a doseeresponse
curve to individualize heparin and protamine dosage. J Thorac Cardiovasc Surg 1975;69:685e9.
[57] Ranucci M, Balduini A, Ditta A, et al. A systematic review of biocompatible cardiopulmonary bypass circuits and clinical
outcome. Ann Thorac Surg 2009;87:1311e9.
*[58] Finley A, Greenberg C. Review article: heparin sensitivity and resistance: management during cardiopulmonary bypass.
Anesth Analg 2013;116:1210e22.
[59] Staples MH, Dunton RF, Karlson KJ, et al. Heparin resistance after preoperative heparin therapy or intraaortic balloon
pumping. Ann Thorac Surg 1994;57:1211e6.
[60] Williams MR, D'Ambra AB, Beck JR, et al. A randomized trial of antithrombin concentrate for treatment of heparin
resistance. Ann Thorac Surg 2000;70:873e7.
*[61] Spiess BD. Treating heparin resistance with antithrombin or fresh frozen plasma. Ann Thorac Surg 2008;85:2153e60.
[62] John LC, Rees GM, Kovacs IB. Inhibition of platelet function by heparin. An etiologic factor in postbypass hemorrhage.
J Thorac Cardiovasc Surg 1993;105:816e22.
[63] Teoh KH, Young E, Bradley CA, et al. Heparin binding proteins. Contribution to heparin rebound after cardiopulmonary
bypass. Circulation 1993;88:II420e5.
[64] Ranucci M, Isgro G, Cazzaniga A, et al. Predictors for heparin resistance in patients undergoing coronary artery bypass
grafting. Perfusion 1999;14:437e42.
[65] Bharadwaj J, Jayaraman C, Shrivastava R. Heparin resistance. Lab Hematol Off Publ Int Soc Lab Hematol 2003;9:125e31.
[66] Chan T, Hwang NC, Lim CH. A statistical analysis of factors predisposing patients to heparin resistance. Perfusion 2006;21:
99e103.
[67] Garvin S, Fitzgerald D, Muehlschlegel JD, et al. Heparin dose response is independent of preoperative antithrombin
activity in patients undergoing coronary artery bypass graft surgery using low heparin concentrations. Anesth Analg
2010;111:856e61.
[68] Lemmer Jr JH, Despotis GJ. Antithrombin III concentrate to treat heparin resistance in patients undergoing cardiac sur-
gery. J Thorac Cardiovasc Surg 2002;123:213e7.
[69] Avidan MS, Levy JH, van Aken H, et al. Recombinant human antithrombin III restores heparin responsiveness and de-
creases activation of coagulation in heparin-resistant patients during cardiopulmonary bypass. J Thorac Cardiovasc Surg
2005;130:107e13.
[70] Sabbagh AH, Chung GK, Shuttleworth P, et al. Fresh frozen plasma: a solution to heparin resistance during cardiopul-
monary bypass. Ann Thorac Surg 1984;37:466e8.
[71] Koster A, Fischer T, Gruendel M, et al. Management of heparin resistance during cardiopulmonary bypass: the effect of
ve different anticoagulation strategies on hemostatic activation. J Cardiothorac Vasc Anesth 2003;17:171e5.
[72] Kikura M, Tanaka K, Hiraiwa T. Nafamostatmesilate, as a treatment for heparin resistance, is not associated with peri-
operative ischemic stroke in patients undergoing cardiac surgery with cardiopulmonary bypass. J Cardiothorac Vasc
Anesth 2012;26:239e44.
*[73] Ferraris VA, Brown JR, Despotis GJ, et al. 2011 Update to the Society of Thoracic Surgeons and the Society of Cardio-
vascular Anesthesiologists blood conservation clinical practice guidelines. Ann Thorac Surg 2011;91:944e82.
[74] Despotis GJ, Joist JH, Hogue Jr CW, et al. The impact of heparin concentration and activated clotting time monitoring on
blood conservation. A prospective, randomized evaluation in patients undergoing cardiac operation. J Thorac Cardiovasc
Surg 1995;110:46e54.
[75] Koster A, Fischer T, Praus M, et al. Hemostatic activation and inammatory response during cardiopulmonary bypass:
impact of heparin management. Anesthesiology 2002;97:837e41.
[76] Ovrum E, Holen EA, Tangen G, et al. Completely heparinized cardiopulmonary bypass and reduced systemic heparin:
clinical and hemostatic effects. Ann Thorac Surg 1995;60:365e71.
[77] Slight RD, Buell R, Nzewi OC, et al. A comparison of activated coagulation time-based techniques for anticoagulation
during cardiac surgery with cardiopulmonary bypass. J Cardiothorac Vasc Anesth 2008;22:47e52.

Anaesthesiology
Mechanical circulatory support

Kathirvel Subramaniam, MD, MPH, Visiting Associate
Professor
Department of Anesthesiology, University of Pittsburgh, Pittsburgh, PA 15213, USA
Keywords:
Heart failure (HF) is a condition in which the heart is not able to
ventricular assist device
pump enough blood and oxygen required for organ systems to
mechanical circulatory support
function. According to recent statistics from the American Heart
Association (AHA), about 5.1 million people in the nation suffer
from HF; one in nine deaths in 2009 included HF as a contributing
cause. About half of people who develop HF die within 5 years of
diagnosis. HF costs the nation an estimated $32 billion each year.
This total includes the cost of health-care services, medications to
treat HF, and missed days of work [1]. Despite several recent
promising developments in medical therapy for HF, many patients
still progress to advanced stages of HF. The annual mortality rate
for patients with advanced HF remains high [2].
Heart transplantation (HT) is the denitive therapy for advanced
HF, but it is limited by the availability of donors and strict recipient
criteria applied to avoid poor outcomes. Therefore, the alternate
treatment of mechanically supporting the ventricles, ventricular
assist device (VAD) therapy, has gained an important role in the
management of advanced HF (stage D). This chapter discusses the
indications, contraindications, and various classications of me-
chanical circulatory support (MCS) and individual features of
commonly used VADs. Perioperative management of patients un-
dergoing MCS will also be described in detail.
Indications and categories of mechanical circulatory support
Heart failure (HF) is a well-recognized costly public health problem with high patient mortality
[1,2]. Patients with heart failure symptoms refractory to optimum medical treatment are referred for
mechanical circulatory support therapy. According to Interagency Registry for Mechanically Assisted
E-mail address: subramaniamk@upmc.edu.
204 K. Subramaniam / Best Practice & Research Clinical Anaesthesiology 29 (2015) 203e227
Table 1
Intermacs patient proles for end-stage heart failure.
Prole Clinical picture
1 Critical cardiogenic shock. Life-threatening hypotension, increasing requirement for inotropes with evidence
of organ hypoperfusion (elevated lactate levels and acidosis). Patients need MCS in hours.
2 Progressive decline on inotropes with evidence of volume overload, poor tolerance to increasing doses of
inotropes and impaired organ perfusion. Patient will need VAD support in few days.
3 Clinical stability on moderate dose of inotropes or temporary circulatory support, can be hospitalized or at home.
VAD support indicated in weeks.
4 Recurrent advanced heart failure e resting symptoms on home oral therapy. VAD support is indicated in weeks
or months.
5 Exertion intolerant e comfortable at rest, but ADL limited by symptoms. VAD support requirement is variable.
6 Exertion limited e ADL not limited, but all meaningful exertion limited. VAD support requirement is variable.
7 History of decompensation but currently stable at reasonable level of activity (advanced NYHA Class 3). VAD
implantation has been shown to improve survival.
ADL, Activity of Daily Living.

NYHA, New York Heart Association.
Circulatory Support (Intermacs) statistics, about 12,474 ventricular assist devices (VADs) were
implanted across 149 participating programs in the United States between June 2006 and September
2014 [3]. Intermacs classied severe heart failure (HF) into seven subcategories to describe patient
status before VAD implantation (Table 1). Potential modiers for the clinical proles include frequent
shocks (>2 per week) by a debrillator, requirement for any form of temporary circulatory support
(TCS) (applies to proles one, two, and three), and frequent hospitalizations (three visits in <6 months
applies to proles four, ve, and six). Patient proles two and three accounted for 65% of VAD patients
in the recent report; 17.7% of patients were prole one and 13% were prole four.
Once a patient receives VAD support for severe HF, they either wait for a suitable donor (bridge to
transplant (BTT)) or live with VAD life support (destination therapy (DT)). About 5e10% of patients can
be successfully weaned off VAD support (bridge to recovery (BTR)) [4]. The device strategies are
classied in Table 2. When mechanical circulatory support (MCS) became an established therapy for HF,
most of the VADs were inserted with the hope of getting heart transplantations (HTs) for the patients.
The outlook has changed over the past 3e4 years because of evidence that DT improves survival and
quality of life in New York Heart Association (NYHA) class IV and class IIB patients [29]. DT accounted
for most VAD implantations, outnumbering bridging indications from 2011 onwards, when the US Food
and Drug Administration (FDA) approved the Heartmate II (HM II) for DT [3].
Besides the indications dened by Intermacs patient proles (acute and chronic HFs from various
causes), several other clinical circumstances require MCS, including high-risk percutaneous trans-
luminal coronary angioplasty and complicated valvular/revascularization procedures. MCS is used to
Table 2
Device strategy at the time of VAD implantation.
1. Bridge to Transplant (BTT) Listed e patient already listed for transplant or listed within 24 h before device implantation.
2. Bridge to Transplant (BTT) Likely e patient in whom the transplant evaluation has not been completed, but no
contraindications are anticipated, or in whom a current contraindication is anticipated to resolve rapidly.
3. Bridge to Transplant (BTT) Moderate e patient in whom the transplant evaluation has not been completed, but with some
potential concerns that might prevent eligibility.
4. Bridge to Transplant (BTT) Unlikely e patient in whom major concerns that might prevent eligibility have already been
identied.
5. Destination Therapy (DT) e the patient is denitely not eligible for transplant.
6. Bridge to Recovery (BTR) e the use of a durable device to allow recovery from chronic cardiac failure (at least 3 months
in duration).
7. Rescue Therapy e the use of a durable device to support resolution from an acute event without major previous cardiac
dysfunction.
8. Other.
K. Subramaniam / Best Practice & Research Clinical Anaesthesiology 29 (2015) 203e227 205
Table 3
Indications and contraindications for mechanical circulatory support.
Indications by patients condition

Chronic heart failure from dilated cardiomyopathy (ischemic/nonischemic)
Hypertrophic obstructive cardiomyopathy
Restrictive cardiomyopathy
Acute cardiogenic shock (includes acute myocardial infarction and myocarditis)
Procedure-related cardiogenic shock (PTCA, ventricular tachycardia ablation, CABG, valve and other cardiac surgery)
Early graft failure after heart transplantations
Indication by patient parameters
Left ventricular ejection fraction <25%
Inotropic dependency
Cardiac index <2 l/min/m2, pulmonary capillary wedge pressure >20 mm Hg, Systolic blood pressure 80e90 mm Hg
VO2 < 12 ml/kg/min
Right ventricular dysfunction (cardiac index <2 l/min/m2 with central venous pressure >20 mm Hg: PCWP <10 mm Hg in
isolated RV failure or >20 mm Hg in biventricular dysfunction)
Progressive renal/liver dysfunction attributed to cardiac failure
Absolute contraindications
Irreversible neurological damage
Irreversible organ dysfunction not likely to be reversed with MCS
Incurable malignancy
Septic shock/active infective endocarditis
Hemostatic disorders (bleeding or hypercoaguable disorders not amenable to treatment)
Psychiatric disorders
Noncompliance and lack of psychosocial and nancial support
Relative contraindications
Sepsis treatable with antibiotics
Recent stroke
Persistent vasodilatory state
Treatable psychiatric diseases
Extreme body size (cachexia and morbid obesity)
support the myocardium that suffers ischemiaereperfusion injury during these procedures. In-
dications and contraindications for MCS are dened in Table 3.
Classication methods for MCS devices
MCS devices are classied based on the ventricles they support (right, left, or both ventricles), their
ow characteristics (continuous and pulsatile), pump design (axial and centrifugal), drive mechanism
(pneumatic and electric), location of the pump in relation to the body (intracorporeal, extracorporeal,
or paracorporeal), duration of support they can provide (short term: <7 days, intermediate: 7 days to 1
year, and long term: >1 year), and their generation (rst-generation, second-generation, and third-
generation devices). Total articial heart (TAH) is another special category of circulatory support,
which will be described in a separate section. Commonly used MCS devices and their characteristics are
described in Table 4.
Temporary circulatory support
The aim of TCS is to temporarily stabilize patients presenting with acute cardiogenic shock (ACS)
with MCS before denitive therapy can be employed. An ideal TCS device should increase the coronary
blood ow and unload the ventricles (decrease preload and afterload) to give a much needed rest to the
myocardium at risk. It is anticipated that the period of myocardial rest will augment the recovery of the
myocardium. Percutaneous VADs (pVADs) (Impella, TandemHeart) are available for use in patients
with ACS [5]. Intra-aortic balloon counterpulsation (IABP) is readily available, easily inserted, and is
quite a familiar device used to treat the clinical syndrome of ACS. However, randomized controlled
clinical trials consistently showed no survival benet with IABP use in cardiogenic shock [6,7].
206
K. Subramaniam / Best Practice & Research Clinical Anaesthesiology 29 (2015) 203e227
Table 4
Commonly used ventricular assist devices at the University of Pittsburgh Medical Center.
Device Duration of Ventricle Purpose of Generation Location of the Drive Pump Flow
support supported support device
Thoratec Short to RV, LV, BV BTT, BTR First Paracorporeal Pneumatic Vacuum-assisted lling and
PVAD intermediate pneumatic compression of blood
chamber pumping blood
producing pulsatile ow
Levotronix Short RV, LV, BV BTR, BTD, BTB Second Extracorporeal Electric Centrifugal Magnetically levitated Continuous
Centrimag
Heartmate II Intermediate, LV BTT, DT Second Intracorporeal Electric Axial Continuous
Long
Heartware Intermediate, LV BTT, DT Third Intracorporeal Electric Centrifugal, Magnetically Continuous
Long levitated
Impella Short LV, RV BTR, BTD, BTB Third Intracorporeal Electric Impellar driven-microaxial Continuous
blood pump
TandemHeart Short LV, RV BTR, BTD, BTB Third Extracorporeal Electric Centrifugal pump Continuous
Extracorporeal life support
Extracorporeal circulatory life support (ECLS) or extracorporeal membrane oxygenation (ECMO) are
reliable TCS methods that can be easily established in any emergency situation [8]. ECMO can also
ensure adequate oxygenation in patients with ACS presenting with acute pulmonary edema. The
essential components of ECMO circuit include cannulae, a centrifugal pump, a hollow-ber oxygenator,
and a warmer. ECMO support is usually achieved through peripheral cannulation in emergency situ-
ations (usually through the femoral vein and artery). In patients with postcardiotomy shock, central
cannulation is feasible through sternotomy. Many cardiac surgical procedures are performed nowadays
through thoracotomy and mini-sternotomy. In these patients, peripheral cannulation is once again
preferred. Patients who are unlikely to recover their cardiac and neurologic function or patients who
are not candidates for HT or long-term MCS should not be considered for TCS using ECMO. Therapeutic
anticoagulation aimed for a partial thromboplastin time of 1.5 times normal is required during ECMO
support. Left ventricular (LV) venting or the insertion of a pVAD (Impella) may be required if LV
distension persists during ECMO support. Weaning trials are attempted after 24 h of myocardial rest.
After ensuring adequate anticoagulation and moderate levels of inotropic support, ows through the
pump can be reduced while monitoring mixed venous saturations, pulsatility in arterial waveform, and
LV contractility with echocardiography. Once myocardial recovery is conrmed, patients are dec-
annulated. Common complications include bleeding and thrombosis [8]. Survival after ECMO depends
on the etiology of cardiac failure and severity of injury before initiation of ECMO. In patients with in-
hospital cardiac arrest where ECMO is used to support the circulation during and after cardiopulmo-
nary resuscitation, the survival to discharge rate was 27%, while postcardiotomy patients has better
survival rates (36e42%) [9,10]. Patients with acute fulminant myocarditis had the best survival rate
(71%), while patients with out-of-hospital cardiac arrest had the worst outcomes (4% survival to
discharge) [11,12].
Impella
The Impella left ventricular assist device (LVAD) operates via a microaxial pump that displaces
blood from the LV into the aorta to provide decompression of the LV with an increase in the cardiac
output (Fig. 1A and B). Impella 2.5 and Impella cardiac power (CP) are inserted percutaneously from the
femoral artery retrograde through the aortic valve into the LV, and they can deliver up to 2.5 l/min of
Fig. 1. A. Impella device inserted through the aortic valve into the left ventricle. (Reproduced from Werdan K, Gielen S, Ebelt H,
Hochman JS. Mechanical circulatory support in cardiogenic shock. Eur Heart J. 2014; 35:156e67 by permission from Oxford Uni-
versity Press.). B. Impella 2.5 design showing inlet, outlet, and the pump.
ow rate. Impella 5.0 and Impella LD operate via similar mechanisms (ow rates 5 l/min), but they
require a direct cut-down of the femoral or axillary artery. This device is approved for circulatory
support for 5 h, but longer use has been reported. The use of Impella should be avoided in aortic
insufciency (AI), severe calcic aortic stenosis (AS), and in patients with LV thrombus and prosthetic
aortic valves. Transesophageal echocardiography (TEE) is used to guide placement in the operating
room. The inlet should be positioned 4 cm away from the aortic valve, and it should be clear from any
mitral valve apparatus or papillary muscles. A mosaic ow pattern can be demonstrated with color
ow Doppler [13] (Fig. 2A and B). Impella right percutaneous (RP) is a similar axial pump recently
approved for right ventricular (RV) support. This is inserted through the femoral vein, and it pumps
blood from the inferior vena cava into the pulmonary artery (PA) to decompress the RV [14].
TandemHeart
TandemHeart is another percutaneously inserted LVAD used in ACS. The inow cannula is inserted
through the femoral vein, and it is guided into the left atrium (LA) using a transseptal approach. TEE can
be used to guide the placement of the inow cannula. The blood removed from the LA ows through an
extracorporeal pump and inow cannula into the femoral artery (Fig. 3). Flow rates up to 5 l/min can be
achieved using TandemHeart. An oxygenator can be added to the circuit if necessary. Systemic anti-
coagulation is required, and complications are mainly vascular, as with any other TCS method. Iatro-
genic atrial septal defect closes spontaneously after the removal of the device in 4e6 weeks. Recently,
the use of TandemHeart for RV support has been described [15,16]. A dual lumen cannula is inserted
through the internal jugular vein, and blood is removed through the outer cannula from the right
atrium (RA) and pumped back by an extracorporeal pump through the inner cannula into the PA [17]
(Fig. 4A and B). Oxygenator support can also be provided if necessary. Insertion can be completed only
in specialized centers, and TandemHeart is not easy to insert in a crashing patient. The complication
rate is also higher than with IABP.
In studies comparing IABP with other pVAD devices, pVAD increased the cardiac index (CI) and the
mean arterial blood pressure (MAP), but decreased the pulmonary capillary wedge pressure and lactate
levels. No difference in mortality was found between the treatment groups. Complications, especially
bleeding, were slightly higher in patients receiving pVAD [18e20]. The major characteristics and he-
modynamic effects of commonly used TCS devices are described in Tables 5 and 6 [21].
Levotronix Centrimag VAD
Centrimag is a magnetically levitated impeller technology that can support both the RV and LV. This
continuous-ow centrifugal extracorporeal pump has no mechanical bearings or seals, and thus it
operates with minimal hemolysis and thrombogenicity (Fig. 5). An oxygenator can be attached to
support oxygenation if required. System implantation requires a sternotomy, although implantation
through thoracotomy has been described. For LVAD, the inow cannula is placed through the LV apex
Fig. 2. A. Mid-esophageal long-axis view conrms the placement of Impella device. B. Mid-esophageal aortic valve long-axis color
ow Doppler showing the mosaic pattern at Impella outow.
Fig. 3. TandomHeart LVAD inow is from the left atrium (cannulation through the femoral vein with interatrial septal puncture to
reach the left atrium) and outow into the femoral artery. (Reproduced from Werdan K, Gielen S, Ebelt H, Hochman JS. Mechanical
circulatory support in cardiogenic shock. Eur Heart J. 2014; 35:156e67 by permission from Oxford University Press.)
or the LA and the outow graft is attached to the aorta. For right ventricular assist device (RVAD),
cannulation is performed to direct the ow from the RA to the main PA. John et al. reported 12 patients
with ACS treated with levotronix as a bridge to decision. Eight patients received permanent LVAD, and
two had levotronix explanted [22]. The overall survival at 1 year was 62.5%. Several other studies re-
ported successful use of levotronix for ACS, RV failure after LVAD insertion and HT, and in post-
cardiotomy patients [23,24]. Levotronix is approved for RVAD use for 30 days.
Other biventricular assist devices such as Abiomed BVS 5000 and Thoratec paracorporeal VAD
(Thoratec PVAD) are used less often in sick patients with ACS, as they require sternotomy, cardiopul-
monary bypass (CPB), and partial aortic cross-clamping. The Thoratec PVAD is still used in some pa-
tients as BTT or BTR after an initial period of stabilization with TCS devices, usually ECMO.
Thoratec PVAD
The Thoratec PVAD is a pneumatically driven, pulsatile, and paracorporeal VAD that can be used to
assist both the ventricles (LVAD, RVAD, and biventricular VAD (BiVAD)). Thoratec PVAD has been used
for short-term and long-term support. It uses unidirectional disk valves to maintain a unidirectional
blood ow. A blood sac in a rigid casing is compressed by air to achieve ejection (Fig. 6). Cannulation is
placed from the RA to the PA for RVAD and from the LV apex to the aorta for LVAD. The pump can
Fig. 4. A. TandemHeart used for right ventricular support with a dual lumen cannula (yellow arrow) inserted through the internal
jugular vein. TandemHeart pump is indicated by a red arrow. B. Oxygenator can be attached to the TandemHeart RVAD circuit to
provide venovenous extracorporeal oxygenation. The gure also shows the TandemHeart controller.
operate in a ll-to-empty mode, a xed-rate mode, and a synchronous-to-electrocardiographic (EKG)

mode. A ll-to-empty mode is commonly used. The device is programmed to eject when the VAD is
lled with blood during diastole. The rate of LVAD ejection will vary with preload. A higher preload will
increase the rate. In a xed-rate mode, the VAD empties at a xed rate, and stroke volume varies with
each ejection. In a synchronous mode, VAD ejection is triggered by the R wave of EKG. This will prevent
VAD ejection from occurring during native ventricular contraction. Thoratec PVAD can also be used in
obese and pediatric patients.
Long-term LVAD devices
HM II and Heartware (HVAD) are the two continuous-ow LVAD devices used for long-term support
at the University of Pittsburgh Medical Center (UPMC). The FDA approved HM II for use as a BTT and DT.
This device consists of a valveless, axial ow rotary pump with an inow conduit inserted into the LV
apex and the outow graft anastomosed to the ascending aorta. The pump weighs 281 g with a volume
of 63 ml, and it is implanted in the pre-peritoneal pocket with the exiting driveline connected to the
wearable controller (Fig. 7). Flow rates up to 10 l/min can be achieved, and the displayed ows are
estimated from pump-power utilization. The ow varies with speed, preload, and afterload. Speed
refers to how fast the impeller can spin, which is measured in revolutions per minute (RPM). Increasing
the speed increases the ow and decreasing the speed decreases the ow. Power is utilized to run the
pump at a set speed, and a persistent increase in power may indicate pump thrombosis. Pulsatility
index (PI) indicates how well the pump unloads the LV. PI is dened as the difference between the
maximum and minimum pump ows divided by an average pump ow multiplied by 10. A low PI
indicates a high amount of assistance from the pump, whereas a high PI indicates an increased
contribution from native LV contractility.
HVAD is a third-generation continuous ow, centrifugal LVAD that is approved for use as a BTT.
Unlike HM II, the pump is implanted intrapericardially above the diaphragm. The pump has a displaced
volume of 50 cc and weighs 160 g. The inow cannula is sutured to the LV apex and the shorter outow
graft to the ascending aorta (Fig. 8A and B). Pump controls are quite similar to those in the HM II, except
that PI is not displayed through the waveform to give a clear idea about pulsatility.
Table 5
K. Subramaniam / Best Practice & Research Clinical Anaesthesiology 29 (2015) 203e227

Comparison of Temporary Circulatory Support devices.
IABP ECMO TandemHeart Impella 2.5 Impella 5.0
Pump mechanism Pneumatic Centrifugal Centrifugal Axial ow Axial ow

Cannula size 7.9 Fr 18e21 Fr inow; 21 Fr inow; 15e17 Fr 13 Fr 22 Fr
15e22 Fr outow outow
Insertion technique Descending aorta Inow cannula into 21 Fr inow cannula 12 Fr catheter 21 Fr catheter placed
via the femoral the right atrium via into the left atrium via placed retrogradely retrogradely across the
artery the femoral vein, the femoral vein and across the aortic aortic valve via a surgical
outow cannula into transseptal puncture valve via the femoral cut-down of the femoral artery
the descending aorta and 15e17 Fr outow artery
via the femoral artery cannula into the femoral
artery
Hemodynamic support 0.5e1.0 l/min >4.5 l/min 4 l/min 2.5 l/min 5.0 l/min
Implantation time
Risk of limb ischemia
Anticoagulation
Hemolysis
Post-implantation
management complexity
Optional active cooling in No Yes (Yes) No No
postcardiopulmonary
resuscitation patients
ECMO, extracorporeal membrane oxygenation; IABP, intra-aortic balloon pump; , , , , relative qualitative grading concerning time (implantation time), risk (risk of limb
ischemia), intensity (anticoagulation, postimplantation management complexity), and severity (hemolysis).
Reproduced from Werdan K, Gielen S, Ebelt H, Hochman JS. Mechanical circulatory support in cardiogenic shock. Eur Heart J. 2014; 35:156e67 by permission from Oxford University Press &
Permission also obtained from original source Ouweneel DM, Henriques JP. Percutaneous cardiac support devices for cardiogenic shock: current indications and recommendations. Heart.
2012 Aug; 98(16):1246e54. Reproduced with permission from BMJ Publishing Group Ltd.
211
Table 6
Proposed hemodynamic effects of the mechanical circulatory support devices.
IABP ECMO TandemHeart Impella
Afterload Reduced Increased Increased Neutral

LV stroke volume Slight increase Reduced Reduced Reduced
Coronary perfusion Slight increase Unknown Unknown Unknown
LV preload Slightly reduced Reduced Reduced Slightly reduced
PCW pressure Slightly reduced Reduced Reduced Slightly reduced
Peripheral tissue perfusion No signicant increase Improved Improved Improved
Reproduced from Werdan K, Gielen S, Ebelt H, Hochman JS. Mechanical circulatory support in cardiogenic shock. Eur Heart J.
2014; 35:156e67 by permission from Oxford University Press.
Clinical trials
Miller et al. studied 133 patients with HF who underwent HM II implantation while waiting for HT.
The principal outcome measure was survival to recovery, survival to HT, or living with LVAD at 180 days
after HM II implantation. Seventy-ve percent of patients achieved the principal outcome at 6 months
and 68% at 1 year [25]. In a further study on patients waiting for HT, Starling et al. compared the effect of
HM II (n 169) to that of other pulsatile LVAD devices (n 169). Survival to transplant, recovery, and
ongoing support at 6 months occurred in 91% of patients with HM II versus 80% of patients with other
LVADs. A 1-year survival was 85% in the HM II group compared to 70% with other LVADs [26]. ADVANCE
clinical trial investigators evaluated the use of HVAD for BTT, and they compared the successful survival
to the contemporary LVADs. HVAD was not inferior to other LVADs, and the adverse event prole was
favorable [27].
Patients ineligible for HT were evaluated in the REMATCH clinical trial, and survival with LVAD
therapy was compared with medical therapy. The investigators found a 48% reduction in the risk of
death from all causes at 1 year but increased adverse events related to the device (infection, bleeding,
and malfunction). The rst-generation pulsatile LVAD Heartmate XVE was used in the clinical trial [28].
Slaughter et al. studied HM II for DT and showed improved survival free of disabling stroke and pump
replacement at 2 years (46%) compared to pulsatile LVADs (11%). Pump replacement, rehospitaliza-
tions, infections, renal failure, respiratory failure, cardiac arrhythmias, and RV failure were signicantly
reduced with continuous ow LVADs (CF-LVADs) [29]. The ENDURANCE clinical trial examining the
role of HVADs for DT is currently ongoing. Whether the newer third-generation LVADs provide better
outcomes compared to the standard HM II is also being investigated. Sabashnikov et al. reported that
Fig. 5. Centrimag levotronix continuous ow centrifugal pump. (Reprinted with the permission of Thoratec Corporation.)
Fig. 6. Thoratec paracorporeal biventricular support device. (Reprinted with the permission of Thoratec Corporation.)
HVAD and HM II produced similar outcomes, but HM II patients reported more transfusion and
driveline infections compared to patients with HVAD [30].
From the above clinical trials, it is clear that LVAD support is better than medical therapy alone for
end-stage HF patients. HM II and other CF-LVADs offer less surgical trauma, better durability and
reliability, improved quality of life, and decreased adverse events. However, the importance of the
preservation of pulsatility is being recognized. Pulsatile perfusion may have advantages in the long
term, including improved LV recovery, improved compliance of small arteries, and decreased gradients
across the aortic valve [31]. Active research on generating pulsatility within CF-LVADs to negate some
of the known negative consequences of CF such as AI, acquired von Willebrand factor (VWF) deciency,
arteriovenous (A-V) malformations, and gastrointestinal (GI) bleeding is underway [32].
Preoperative management
Preoperative preparation and optimization may not be feasible in patients presenting with ACS.
ECMO represents the best modality of hemodynamic support in these circumstances, and patients
should be reevaluated after 24e48 h for further course of action (bridge to decision). Surgical or
percutaneous revascularization while on MCS is possible in some patients to facilitate myocardial
recovery. MCS should be terminated in patients with no neurological recovery. While few patients
completely recover for ECMO or TCS to be weaned off completely, others will require long-term VAD
placement as BTT or bridge to candidacy (BTC). This strategy is called bridge to bridge. BTC is more
likely in this scenario because the patients cannot be listed for transplantation at the time of VAD
implantation because they are critically ill, have a contraindication for transplantation, or have not
been evaluated for HT. Pulsatile devices such as Thoractec PVAD should be considered in patients with
ACS, as CF-LVADs may not achieve satisfactory ventricular emptying in ventricles without sufcient
remodeling (nondilated ventricles). Long-term VADs should not be inserted in patients with acute
infarction of the LV apex [33].
Fig. 7. Heartmate II continuous ow pump with driveline and controller. (Reprinted with the permission of Thoratec Corporation.)
Patients who develop ACS in the catheterization laboratory or in the operating room (post-
cardiotomy) have more options for TCS. Central ECMO, peripheral ECMO, Impella, TandemHeart, and
Centrimag are possible options. The requirement for very high doses of inotropes and vasopressors to
maintain blood pressure and the presence of metabolic acidosis after the termination of CPB should
Fig. 8. Heartware HVAD (A). The inow into the left ventricular apex and the outow to the ascending aorta (B). The pump is housed
intrapericardially. (Reproduced with permission from Heartware. Inc.)
alert the team regarding the need for TCS. CPB is reinitiated while preparations are made for assist
device insertion. TEE is vital for decision making in the operating room. Patients with ACS who received
thrombolysis or anticoagulation and antiplatelet therapy for percutaneous or surgical revascularization
are at a very high risk of bleeding, and the team should be prepared to monitor coagulation and
administer blood products accordingly.
Patients with decompensated HF but not presenting with ACS are prepared in the intensive care
unit (ICU) before surgery. PA catheterization, aggressive diuresis or mechanical uid removal, inotropes
with or without inodilator therapy, and IABP are employed to prepare patients before surgery. Elec-
trolyte and metabolic parameters should be monitored and optimized before surgery [33].
An extensive cardiac and systematic workup is required before deciding to implant long-term VAD.
All results of preoperative workup investigations, including metabolic prole, arterial blood gas, he-
matology and coagulation prole, echocardiography, and cardiac catheterization should be reviewed
before surgery.
Patients with coronary artery disease (CAD) and previous bypass surgery should have computerized
tomography (CT) of the chest to identify the bypass grafts and guide the surgical approach. Patients
with atrial and ventricular arrhythmias should undergo ablation and automatic implantable car-
dioverter debrillator (AICD) treatment before LVAD implantation. AICD is turned off in the operating
room by the electrophysiology cardiology team before surgical incision and external pads are attached
to treat any arrhythmias in the operating room. BiVAD insertion should be planned in patients with
intractable ventricular arrhythmias who cannot have AICD insertion before VAD placement. Preoper-
ative screening and appropriate management of peripheral vascular disease (angioplasty and stenting)
are recommended before the initiation of MCS. Patients with pulmonary hypertension due to cardiac
disease can be treated with long-term LVAD support to make them suitable for HT. Preoperative PA
pressure and RV function are important factors in determining the need for RVAD support after LVAD
implantation. These patients with a poor LV ejection fraction are prone to develop intracardiac
thrombus, so they are placed on anticoagulation. Coumadin is transitioned to heparin and stopped 4 h
before surgery. Antithrombin III should be available for the treatment of heparin resistance in the
operating room.
Patients with advanced age, diabetes mellitus, neurological disease, malignancy, hematological and
coagulation disorders, psychiatric disease, alcohol and substance abuse, and morbid obesity require
careful evaluation before VAD implantation [33]. Active and severe diseases are contraindications for
VAD support (Table 3). Nutritional status should be optimized before VAD implantation.
Patients with end-stage HF also develop end-organ dysfunction (chronic renal insufciency and
liver dysfunction) from inadequate cardiac output and venous congestion. Appropriate precautions
should be taken with regard to anesthesia drug titration, uid administration, and treatment of anemia
and coagulopathy. Milder forms of hepatic and renal dysfunction (serum creatinine <2 mg/dl) have
been shown to improve at 6 months after HM II insertion [34]. Patients with severe renal dysfunction
and patients on permanent dialysis should not be considered for MCS, as they are associated with high
perioperative mortality and morbidity. Worsening renal dysfunction correlates with decreased sur-
vival, with a nearly 20% reduction in the 2-year survival rate and a decrease from low to severe
dysfunction [35]. ChildePugh class and model for end-stage liver disease (MELD) score are used to
preoperatively evaluate liver dysfunction. Patients with ChildePugh class B and C and MELD score
>13e17 were associated with increased surgical and long-term morbidity, so these patients are poor
candidates for MCS therapy [33,36].
LVAD recipients should be evaluated for pulmonary risk factors such as asthma, chronic obstructive
pulmonary disease, smoking, obesity, obstructive sleep apnea, and respiratory infections. Chest X-ray,
pulmonary function tests, and CT imaging of the chest should be used to evaluate preoperative lung
function whenever appropriate. Preoperative medical optimization with cessation of smoking, bron-
chodilators, antibiotics, and incentive spirometers is recommended. Lung protective ventilation,
avoidance of transfusion, and adequate postoperative pain relief are a few perioperative strategies that
can be applied to decrease pulmonary morbidity. Gas-exchange abnormalities not only prolong the
duration of mechanical ventilation but also can increase pulmonary vascular resistance (PVR) and
precipitate RV failure. If intraoperative hypoxemia is problematic after LVAD insertion, venovenous
ECMO can be used to stabilize the situation. RVAD with oxygenator should be used in patients with
hypoxemia and RV failure.
Intraoperative management
Anesthetic management of patients undergoing LVAD insertion is driven by specic institutional

practice protocols rather than by evidence-based guidelines. Judicious administration of Midazolam
premedication is advised to provide anxiolysis but to avoid oversedation, hypoxemia, hypercarbia, and
increased PVR. Inotropes (dobutamine, milrinone) are continued in the pre-bypass period. Proper
intravenous and invasive arterial access should be established before the induction of anesthesia.
Central venous access with a PA catheter can be established before the induction of anesthesia in
patients prone to developing hemodynamic instability and RV failure during anesthesia induction.
Ultrasound-guided cannulation is advisable, as difcult vascular access is not uncommon in patients
presenting for VAD implantation. Etomidate is a preferred induction agent because of its lack of sig-
nicant effects on myocardial contractility or vascular tone. Hypotension after induction is common,
and it is contributed to by preoperative angiotensin-converting enzyme inhibitor (ACEI) therapy,
autonomic dysfunction, poor myocardial reserve, and positive pressure ventilation. These patients
maintain their cardiac output by increased sympathetic drive, which should not be suppressed
completely by beta-blockers or high-dose narcotics. Vasopressors and inotropes should be ready in the
infusion pump if required to treat hemodynamic instability. The surgeon and perfusion team should be
available in the operating room during induction. Femoral arterial and venous access can be obtained
under local anesthesia in patients at risk of hemodynamic compromise during the induction of
anesthesia to facilitate emergency initiation of CPB.
Choice of narcotics, inhalational anesthetics, and muscle relaxants are not crucial to patient man-
agement as long as they are titrated to effect. Fentanyl, rocuronium, and isourane are used to maintain
anesthesia at our institution with good effect. Besides the standard American Society of Anesthesiol-
ogists (ASA) monitors, invasive arterial pressure, PA catheter, TEE, point-of-care coagulation (POC)
monitoring, and urine output are also used for VAD implantation. PA catheters with continuous cardiac
output and mixed venous saturation (SVO2) capabilities are preferred at our institution. Cardiac output
may not be reliable in the presence of tricuspid regurgitation (TR), but continuous SVO2 monitoring
ensures adequate tissue perfusion and can guide red cell transfusions. Goal-directed therapy for pul-
monary hypertension can be used intraoperatively and postoperatively with the use of PA catheters.
Transesophageal echocardiography
TEE plays a major role in the management of VAD patients. RV function is evaluated pre-bypass to
predict the requirement for RVAD after LVAD implantation. The tricuspid annular plane systolic
excursion of <7.5 mm, the presence of severe TR, the RV fractional area change of <20%, the RV-to-LV
end diastolic ratio >0.72, and the RV short axis-to-long axis ratio of >0.6 are the echocardiographic
parameters known to predict RV failure (RVF) after LVAD placement [37e40]. Kato et al. investigated
the utility of tissue Doppler RV systolic/diastolic velocities and RV strain by speckle track echocardi-
ography in predicting RVF after LVAD placement. Preoperative S0 <4.4 cm/s, RV-E/E0 >10, and RV strain
<14% predicted patients who would develop RVF at day 14 with a 76.5% accuracy [41]. Kiernan et al.
showed that preoperative three-dimensional estimation of RV end-diastolic and end-systolic volumes
can independently predict RVF after LVAD placement [42].
Intraventricular thrombus, especially apical thrombus, should be evaluated; apical thrombectomy is
required before the placement of an inow cannula into the apex. If not recognized, apical thrombus
can lead to pump thrombosis, hemolysis, pump dysfunction, peripheral embolus, and stroke [33]. The
aorta should be carefully examined for the presence of atheromas for the purpose of aortic cannulation
and outow graft anastomosis. Valvular pathology and its management during LVAD placement are
summarized in Table 7 [43]. Patent foramen ovale (PFO) and atrial and ventricular septal defects should
be carefully evaluated using a two-dimensional (2D) color ow Doppler and bubble contrast study
(with Valsalva maneuver to raise right-sided pressures) before the onset of bypass.
Table 7
Management of valvular heart disease in VAD patients.
Valve lesion Importance Intervention
Aortic regurgitation (AR) Recirculation of blood into LV resulting Mild AR e No correction required
in reduced forward ow into the aorta. Moderate and severe AR e Leaet
approximation by closure in a linear fashion or with
autologous pericardium or aortic valve replacement
Aortic stenosis Patients with partial support need Patients with AS are not candidates for partial support
intermittent AV opening
Mitral regurgitation LV decompression by LVAD results MR does not require surgical correction
in decrease in the severity of MR
Mitral stenosis MS will result in inadequate LVAD Mitral commisurotomy or bioprosthetic valve
lling replacement for mean gradients >10 mm Hg
Tricuspid regurgitation The effect on RV function and overall STS database study has shown worse outcomes
survival/outcome unclear after TV repair with LVAD insertion. Few other
studies have shown benet. No consensus yet.
Pulmonary regurgitation Increased recirculation leading to More than mild may be addressed with planned
decreased forward ow into RVAD RVAD support? limited literature
Mechanical aortic valve Risk of xed open position Replaced by bioprosthesis or Cohn Sandwich
Plug technique to exclude the valve from affecting
the blood ow
Prior to stopping CPB and starting the LVAD pump, TEE-guided de-airing is essential to avoid RV and
neurological dysfunction. The LA, pulmonary vein, and LV apex around the cannula and septum should
be examined carefully for microbubble and macrobubble. LVAD activation with a subsequent fall in the
left-sided pressures may uncover an unsealed PFO, so the bubble study may have to be repeated after
the termination of CPB.
After the activation of LVAD, cannula inow and outow should be evaluated with a 2D color and
spectral continuous wave Doppler (Fig. 9A and B). The inow cannula should be angled toward the
mitral valve, aligned with LV outow tract, and should not abut on any ventricular walls. A smooth
laminar blood ow with inow velocities between 60 and 120 cm/s and outow velocities between
100 and 200 cm/s is considered normal [13]. Because of LV decompression, TEE should be reevaluated
for AI severity.
LV lling and LVAD preload should be evaluated using TEE. Two common reasons for LVAD
underlling include hypovolemia and RV dysfunction. If both the RV and LV are small, hypovolemia is
diagnosed. Ballooning of the RV, leftward shift of the interatrial and ventricular septum, and empty
collapsed LV indicate RV dysfunction.
TEE also helps with the placement of RVAD inow and outow cannulae. An RVAD inow cannula
should be placed in the middle of the RA without abutting on the interatrial septum, while the outow
cannula should be placed 1.5e2 cm distal to the pulmonic valve in the main PA.
Fig. 9. A. LVAD inow position conrmed by transesophageal echocardiography. B. LVAD color inow pattern smooth, laminar color
pattern.
Table 8
Factors contributing to RV dysfunction after LVAD placement.
Preexisting increased transpulmonary gradient

RV ischemic injury (hypotension, prolonged CPB, and air)
Increased cardiac output by LVAD leading to an increased RV preload
Altered ventricular geometry leading to the loss of septal contribution to RV output
Administration of blood products
Chest closure
Bleeding and RV cardiac tamponade
Cardiac tamponade and uid overload are two other conditions in the postoperative period that can
be promptly diagnosed with TEE [44]. A persistent increase in RA pressure along with low pump ows
should alert the clinician to perform TEE in the ICU. Regional tamponade and compression of the RA,
RV, and LA are common; 20e30% of patients require reexploration in the operating room. Fluid
overload is diagnosed when there is radiological evidence of pulmonary congestion, moderate-to-
severe mitral regurgitation, LV dilatation, increased aortic valve opening, or rightward septal shift
along with normal cannular position and ows. Diuretics and ultraltration are required to treat this
situation.
RV dysfunction and management
RV dysfunction is dened as the requirement for RVAD or the need for inotropes for >14 days after
LVAD implantation [45]. RVAD is required in 50% of patients undergoing VAD placement for acute
shock and 4e10% of patients undergoing LVAD for chronic HF [46,47]. RV dysfunction after LVAD is
associated with increased mortality and morbidity [45]. LVAD recipients with RV dysfunction also have
higher reexplorations, longer hospital stay, and lower survival to HT. Several perioperative factors
contribute to RVF after LVAD placement [48] (Table 8). Therefore, preoperative prediction of RV
function using clinical, laboratory, hemodynamic, and echocardiographic parameters is of paramount
importance in improving the prognosis of LVAD recipients (Table 9).
RV function is assessed intraoperatively by the correlation of several parameters: MAP, central
venous pressure (CVP), mean PA pressure, CI, SVO2, ows through the LVAD, direct visualization of the
RV through surgical eld, and TEE evaluation of RV function. Elevated CVP, lower PA pressure/MAP
ratio, lower CI and SVO2, lower ows through the LVAD, RV dilatation, and worsening TR all indicate the
need to support RV function.
Maintenance of adequate preload to the LVAD without causing RV dysfunction can be challenging.
Fluid infusion and transfusion are titrated to PA diastolic pressures between 15 and 18 mm Hg and CVP
between 10 and 15 mm Hg while observing the interventricular septum (IVS) and ventricular lling via
TEE. Midline IVS position should be maintained by adjusting LVAD speed (RPM). Higher speed and
ows can cause excessive LV unloading and negative pressure around the inow cannula, resulting in
LV collapse with leftward shifting of the IVS. This shift will further worsen RV function, as the septal
Table 9
Preoperative predictors of post-LVAD RV dysfunction.
Clinical Laboratory Hemodynamic
Female gender Elevated liver enzymes CVP/PCWP ratio > 0.64

Old age Hyperbilirubinemia Increased RAP
Poor nutritional status Hypoalbuminemia RVSWI < 250 mm Hg. Ml/m2
Obesity Elevated BUN and creatinine CI < 2.2 l/m2
Nonischemic cardiomyopathy Elevated NT-proBNP, C-reactive protein
Redocardiac surgery Thrombocytopenia
Preoperative inotropes Elevated prothrombin time
Preoperative mechanical ventilation
Preoperative IABP/ECMO
Emergency surgery
contribution to overall RV contractility is lost. Decreasing the speed will enable the situation to revert
back to normal. Factors causing elevation of PVR (hypoxemia, hypercarbia, acidosis, hypothermia,
inadequate anesthesia, and excessive positive end expiratory pressure) should be avoided. MAP should
be normalized to optimize RV function, keeping in mind systemic hypertension; the resultant
increased afterload can cause a decrease in LVAD ows. Maintenance of sinus rhythm is advised to
optimize hemodynamics. Sinus rhythm is important to preserve RV function and prevent clot for-
mation and ventricular recovery. Patients with ischemic cardiomyopathy and signicant right CAD may
benet from coronary artery grafting. The effect of continuous ow on the native coronary and graft
blood ows deserves further investigation.
Management of RV dysfunction is initially pharmacological with inotropes, vasopressin, and pul-
monary vasodilators. Among the pulmonary vasodilators, inhaled nitric oxide (iNO) is used in the
operating room and ICU to improve LVAD ows. However, a multicenter, placebo controlled, double-
blinded and prospective study by Potapov et al. failed to show any signicant effect of iNO on the
incidence of RV dysfunction after LVAD placement. Secondary end points such as requirement for
RVAD, mechanical ventilation, and length of ICU and hospital stay were not improved [49]. Few studies
using inhaled prostaglandins have shown a decrease in PA pressures after LVAD placement, but
denitive conclusions on benets for RV function cannot be drawn from available studies [50,51]. The
phosphodiesterase 5 inhibitor sildenal has been shown to improve persistent pulmonary hyperten-
sion after LVAD placement and to prevent the development of RVF [52,53]. Any benecial effect of
pulmonary vasodilators on PVR and RV function is probably masked by the effective unloading of
pulmonary circulation by LVAD [54].
Inotropes are the mainstay of RV dysfunction. Milrinone 0.25e0.75 mg/kg/min is initiated while the
patient is on CPB for inotropy and pulmonary vasodilatation. A bolus of milrinone can be given before
the termination of bypass if MAP is not low. Epinephrine of 0.05 mg/kg/min is added before the
termination of bypass, and it is increased to 0.1 mg/kg/min as needed. Dobutamine and dopamine are
not commonly used intraoperatively at our institution, but practices may vary. Preoperative optimi-
zation with levosimendan, a calcium-sensitizing inodilator, has also been investigated in two reports,
and it was shown to improve hemodynamic performance and identify patients who will develop RVF
after LVAD insertion [55,56]. This drug is not approved by the US FDA, and more controlled clinical
trials are required before its routine use is adopted for patients who underwent cardiac surgery.
Vasoplegic syndrome after LVAD implantation occurs in up to 40% of patients [57,58]. The mech-
anisms include excessive endogenous production of nitric oxide, the lack of vasopressin, systemic
inammatory response, and the use of pulmonary vasodilators (e.g., milrinone). Preoperative low LV
ejection fraction, prolonged CPB, and preoperative ACEI intake are risk factors. Vasoplegic syndrome is
best treated by the infusion of vasopressin, which increases MAP with negligible effects on PVR [59,60].
Severe vasoplegia may require the use of multiple vasopressors such as dopamine, norepinephrine, and
methylene blue (in refractory vasoplegia). Failure to maintain adequate systemic blood pressure will
lead to LV over-decompression, negative suction around the cannula on the LV, septal shift toward LV,
and worsening RV dysfunction.
If RV dysfunction persists despite maximum doses of inotropes (epinephrine of 0.1 mg/kg/min and
milrinone of 0.75 mg/kg/min) and vasopressor support, RVAD should be inserted early to avoid venous
congestion and end-organ dysfunction. Centrimag is the preferred RVAD device at our institution,
although percutaneous RVAD devices are increasingly used for RV dysfunction nowadays [61,62]. The
only long-term RVAD support device in use is paracorporeal Thoratec RVAD, which can provide sup-
port for up to 12e24 months as BTT [63]. HVAD has been used as BiVAD in a limited case series with
good results [64]. Research is needed to pursue more options for long-term RV mechanical support
after LVAD insertion, as a signicant number of patients need continued RV support after failure to
wean from temporary RVAD support [65].
Perfusion and surgical considerations
Implantation of LVAD using the off-pump method is possible but hemodynamically can be very
challenging, so most implantations are done with the use of normothermic or mild hypothermic CPB.
Bicaval cannulation is recommended for tricuspid valve procedures and PFO closure. Aortic cross-
clamping and cardioplegic arrest are avoided, except when aortic or mitral valve procedures are
necessary. CPB is utilized to remove excessive volume through ultraltration. CO2 in the eld is used to
decrease air embolism, and TEE-guided de-airing is performed before coming off CPB [33]. Ventricles
are lled with volume, and Valsalva breaths are given to get the air out of the pulmonary circulation.
Patients are placed in the Trendelenburg position, the outow is clamped, and the needle inserted into
the outow graft proximal to the clamp to de-air the ventricles. Vigorous shaking of the atrium and
ventricles may be needed to get the air out of the poorly contracting ventricles. Sometimes, air may
appear in the ascending aorta even after thorough de-airing. If this happens or if RV function is poor,
CPB should be reinitiated and de-airing maneuvers continued. The LVAD pump is started slowly at
lower speeds, and MAP is maintained (afterload to LVAD) to avoid excessive LV decompression, septal
shift, and consequent RV dysfunction. Pump speed can be increased slowly while monitoring TEE for
septal midline position.
Bleeding and transfusions
Bleeding and transfusions are major concerns after LVAD insertion. Bleeding could be related to
inadequate surgical hemostasis or medical coagulopathy. Common sites of surgical bleeding include
the RA cannulation site, sternal wiring, pre-peritoneal pump pocket, and around the inow cannula
and outow-aorta graft anastomotic site. Absolute surgical hemostasis should be ensured before
closure of the chest. The chest can be kept open with packing if bleeding is signicant or hemodynamic
instability prevents closure. The sternum can be closed after removing clots when hemostasis is
restored in 24 h. Delayed chest closure after LVAD insertion increases the ICU length of stay, but it does
not increase infection rates [66].
Coagulopathy in VAD patients is multifactorial. Preexisting antiplatelet/anticoagulant therapy, liver
dysfunction, hypothermia, prolonged CPB, platelet dysfunction, dilutional coagulopathy, and the effect
of contact between mechanical device and blood all contribute to bleeding after LVAD insertion.
Transfusion can be associated with three major problems in patients with VADs: transfusion-
induced lung injury (TRALI), transfusion-induced circulatory overload (TACO), and transfusion-
induced immunomodulation (TAIM) have implications for VAD patients [33]. TRALI may cause hyp-
oxemia and increased PVR, thus necessitating venovenous ECMO support or RVAD support in some
patients. TACO can cause RV dysfunction and increase morbidity. TAIM is associated with immuno-
suppression and increased infection rate. Transfusion-associated allosensitization and the formation of
panel reactive antibodies can complicate a patient's listing for HT.
Care to avoid transfusions in these patients is important. Normalizing coagulation preoperatively,
the use of antibrinolytics and retrograde autologous priming, the use of miniature circuits, and POC
testing can reduce the need for transfusions. Transfusion cannot be avoided completely in all patients,
and prompt treatment of coagulopathy with appropriate blood products is indicated. Desmopressin
can be used in chronic renal insufciency. The safety of factor concentrates in LVAD patients has not
been established. Bruckner et al. have shown a high incidence of thromboembolic events (37%) with a
high-dose factor VII (30e70 mg/kg) compared with a low-dose regimen (9.4% with 10e20 mg/kg) during
LVAD procedures [67]. The successful use of prothrombin complex concentrate (PCC) has been reported
in two patients with LVAD undergoing noncardiac surgery (NCS) without any thromboembolic
sequelae [68]. Further research is required to establish the safety of factor concentrates in LVAD pa-
tients as they are given in smaller volumes than fresh-frozen plasma, thereby possibly avoiding RV
dysfunction.
Postoperative management
Postoperative care of VAD patients is complex, and a multidisciplinary team approach involving
intensivists, cardiac anesthesiologists, cardiac surgeons, HF cardiologists, perfusionists, VAD engineers,
nutritionists, physical therapists, VAD nurses, and social workers is essential to improve outcomes.
Teams should meet regularly to evaluate progress and future care. Patients who had MCS placed for
ACS will need more intense care and experience longer ICU stays. These patients have a higher
incidence of bleeding, reoperations, RVAD support, end-organ dysfunction, and mortality compared to
those who had elective long-term VADs inserted for chronic HF. Postoperative hemodynamic man-
agement is a continuum of intraoperative intensive management. The usefulness of ImaCor hTEE has
been shown in hemodynamic management after high-risk cardiac surgery and MCS [69,70]. Patients
are extubated once the criteria are fullled, invasive monitors are removed, and they are taken off
vasopressors/inotropes. Anticoagulation with heparin and Coumadin is started once bleeding is stable.
Enteral nutrition is started as soon as possible, and renal replacement therapy may be required in
patients with volume overload. Oral antihypertensives are restarted once the patients can orally take
medication.
Arrhythmias may indicate or may result in RV dysfunction. In one case series of 23 patients, ven-
tricular arrhythmias occurred in 52% of patients. Eight patients received debrillation and three of
them became hemodynamically unstable [71]. AICD should be reactivated in the postoperative period
or some VAD patients may require AICD insertion or ablation for intractable new onset arrhythmias.
Reoperations are common, with bleeding and tamponade accounting for the majority of reoperations.
Other reasons for reoperations include RV failure, pump failure, and infections [72]. Several other early
and late complications are listed in Table 10 [73].
Among late complications, bleeding deserves special mention. Acquired VWF deciency is reported
with only CF-LVAD, which may result in bleeding episodes or excessive bleeding after HT [74e76].
Device explantation results in the resolution of this deciency. Desmopressin and cryoprecipitate are
recommended therapies. Another unique complication is the development of A-V malformations in the
GI tract and bleeding in patients supported with CF devices [77,78].
Thrombotic complications are less common than bleeding. Altered blood ow in the chambers and
cannulae and the activation of inammatory/coagulation cascade are suggested mechanisms for VAD-
related thrombosis. Development of thrombosis within the LVAD pump is the most devastating
complication, and it has been increasing in frequency in the last 3 years. The diagnosis of VAD
thrombosis requires symptoms and signs of HF, elevated lactate dehydrogenase from hemolysis, and
echocardiography. VAD monitoring will show increasing power with a low PI. Thrombolysis, anti-
coagulation, and urgent surgical pump replacement are currently recommended therapies. More
research is needed to understand this fatal complication [79]. Clinical heparin-induced thrombocy-
topenia (HIT) occurs in 10% of patients despite positive antibodies in 60% of patients [80]. Device
malfunction and failure are uncommon with the third-generation CF devices nowadays.
Infections are common, and they can be classied as device and nondevice related [73]. Infections
are associated with signicant morbidity, reoperations, and mortality. Device-related infections
include driveline and pump-pocket infections, pump endocarditis, and sternal wound infections, and
common systemic infections reported in LVAD recipients include respiratory, urinary tract, and
catheter-related infections. CF devices with small parts produce fewer infections compared to pulsatile
devices. The administration of prophylactic antibiotics is recommended before surgical incision and
continued postoperatively for 48 h. Once an infection is diagnosed, it is treated with broad-spectrum
antibiotics, surgical exploration/drainage, and, in refractory cases, VAD exchange or HT.
Table 10
Early and late complications after VAD insertion.
Early (<60 days) Late (>60 days)
Bleeding requiring transfusion and reoperation (31e81%), Bleeding complications (17.5%)

Tamponade 28%
Atrial and ventricular arrhythmias (30e60%) GI bleeding 10e40%
RV failure (10e30%) Thromboembolism 6%
Pump thrombosis 2e9%
Reoperations (50%) Aortic valve degeneration (90% in long term)
Respiratory failure (20e30%) Cannula malposition/occlusion
Neurologic dysfunction stroke 15%, TIA 12%, seizures 4% Infections (94% at 1 year)
Renal failure 3e28% Hemolysis (0e3%)
Hepatic failure 2e8%
Hemolysis Psychiatric problems
Infections (42%) Device malfunction and failure (rare to 2%)
Total articial heart
TAH devices can provide biventricular support for critically ill end-stage HF patients. They are useful
in patients with severe biventricular failure, ventricular rupture, signicant ventricular septal defect,
and intractable arrhythmias. TAH offers an advantage in situations in which ventricular drainage may
be inadequate (restrictive, hypertrophic cardiomyopathy).
Currently, two TAH devices are available for human use: Cardiowest TAH and Abiocor implantable
replacement heart. Cardiowest TAH is a pneumatically driven, pulsatile biventricular support device
approved for use as BTT. It has two prosthetic ventricles that connect to the native atria (Fig. 10). The
stroke volume is 70 cc, and it can deliver a maximum ow rate of 8 l/min. It has been implanted in 930
patients worldwide. Copeland et al. described its use in 101 patients between 1993 and 2009 [81].
Ninety-one percent belonged to Intermacs prole one patients and 63.8% survived to HT. Survival after
transplantation at 1, 5, and 10 years was 76.8%, 60.5%, and 41.2%, respectively. Infection, bleeding,
neurological events, and device malfunction are common complications. Copeland et al. reported a
7.9% incidence of stroke and reexploration for bleeding in 24% of patients in their series. Current
versions of TAH are limited by their size, durability, and complications. Continuous ow TAH pumps are
being developed and evaluated in preclinical research studies; these pumps may offer hope in the
future [82].
VAD patients for noncardiac surgery
Hessel II et al. recently reviewed the management of NCS in patients with VADs [83]. As patients
with CF-VADs survive longer with good quality of life, they more often present for NCS. Several
Fig. 10. Total articial heart support using Syncardia Systems. (Courtesy: SynCardia Systems, Inc. Reproduced with permission from
Syncardia Systems Inc.)
series and case reports on VAD patients undergoing NCS have been published [84e87]. Outcomes
after NCS in patients with VAD are favorable [88,89]. Anesthesia care should be managed at
specialized centers by cardiac anesthesiologists, and patients should recover in the ICU. A VAD
engineer should be available to manage the system, and a cardiac surgeon who implanted the device
should be contacted to discuss the management. In many centers, HF cardiologists take over the care
of VAD patients once they are stable from surgical perspective, and they can also serve as a useful
resource in perioperative management. RV dysfunction and the need for RVAD at the time of LVAD
implantation should alert the anesthesiologist to monitor RV function closely. The status of anti-
coagulation (last dose of Coumadin, ASA physical status, prothrombin time, and INR) should be
checked. Reversal of Coumadin's effect and holding anticoagulation during the perioperative period
out of concern for bleeding are well tolerated in HM II patients. AICD can be inactivated depending
on the procedure needs, and external debrillator pads should be applied in these patients until
AICD can be reactivated. Special attention should be paid to the administration of broad-spectrum
antibiotics required for the particular NCS, as infectious complications may result in signicant
morbidity and mortality. It is not possible to obtain either reliable blood pressure with a noninvasive
blood pressure cuff or oxygen saturation with pulse oximetry in patients with CF-LVADs. Invasive
arterial access under ultrasound guidance is advised in surgeries with anticipated uid shifts and
blood loss. Minor surgical procedures can be performed using a blood pressure cuff and Doppler
ultrasound of the radial artery. A cerebral oximeter can be used to get an idea about oxygenation, as
this does not depend on pulsatility. Central line placement will aid in the administration of ino-
tropes/vasoactive drugs, and it can be used to monitor CVP. Any elevation in CVP (>15 mm Hg) with
lower ows and blood pressure may indicate RV dysfunction. Diagnosis should be conrmed by TEE
and appropriate management initiated. The use of a PA catheter is not unreasonable in patients with
pulmonary hypertension and borderline RV function who are undergoing major surgery. In the
event of cardiac standstill, external cardiac massage is avoided. Epinephrine is administered, and
ECMO is initiated for TCS.
Conict of interest
None.
Practice points
LVADs improve survival and quality of life compared to medical therapy alone in HF patients
CF-LVADs offer durability, reliability, and better survival compared to pulsatile LVAD devices
CF-LVAD devices also provide similar 2-year survival rates (80%) when compared to HT
Patients with ACS should be stabilized on TCS before definite decisions can be made
regarding long-term support, transplantation, and recovery. Several options are currently
available for temporary support of both ventricles: Impella, TandemHeart, and ECMO.
RV dysfunction occurs in 20% of patients after LVAD insertion. Preoperative prediction is
possible using clinical, hemodynamic, and echocardiographic parameters so that early
management with biventricular support can be initiated
Intraoperative echocardiography plays an important role in the guidance of VAD procedures
Early and late complication rates are quite significant, and they are the limiting factors in
making mechanical support an alternative treatment to HT
NCS is common in patients with VADs, and definitive guidelines and protocols should be
established at each institution to improve outcomes after NCS
Research points
In the era of cost containment and the Affordable Care Act, the cost effectiveness of LVAD
implantation and maintenance should be proven before strong recommendations can be
made about their use, especially in NYHA class III patients.
Pulsatile perfusion may have long-term advantages, including improved LV recovery,
improved compliance of small arteries, and decreased gradients across the aortic valve.
Therefore, the creation of pulsatility within CF-LVADs is being explored.
Safe and effective RVADs are needed for patients who require long-term RV support
Large prospective multicenter studies comparing the safety and effectiveness of various
percutaneous temporary VADs are needed, as they are very promising in the management of
ACS.
TAHs with continuous flow technology are being developed; these will play a significant role
in the management of severe biventricular failure in the future.
Reliable, noninvasive blood pressure measurement devices, if developed, will be useful in
the perioperative management of patients with CF-LVADs undergoing NCS.
References
[1] Go AS, Mozaffarian D, Roger VL, et al. Heart disease and stroke statisticsd2013 update: a report from the American Heart
Association. Circulation 2013;127. e6ee245.
[2] Konstam M. Progress in heart failure management? Lessons from the real world. Circulation 2000;102:1076e8.
[3] http://www.uab.edu/medicine/intermacs/stat-summaries/intermacs-stat-summaries/intermacs-descriptive-slides
[accessed 28.02.15.].
[4] Hernanadez AF, Shea AM, Milano CA, et al. Long term outcomes and costs of ventricular assist devices among medicare
beneciaries. J Am Med Assoc 2008;300:2398e406.
*[5] Gilotra NA, Stevens GR. Temporary mechanical circulatory support: a review of the options, indications, and outcomes.
Clin Med Insights Cardiol 2015;8(Suppl. 1):75e85.
[6] Thiele H, Zeymer U, Neumann F-J, et al., for the IABP-SHOCK II Trial Investigators. Intraaortic balloon support for
myocardial infarction with cardiogenic shock. N Engl J Med 2012;367:1287e96.
[7] Prondzinsky R, Lemm H, Swyter M, et al. Intra-aortic balloon counterpulsation in patients with acute myocardial
infarction complicated by cardiogenic shock: the prospective, randomized IABP SHOCK Trial for attenuation of multi-
organ dysfunction syndrome. Crit Care Med 2010;38:152e60.
[8] Lawler PR, Silver DA, Scirica BM, et al. Extracorporeal membrane oxygenation in adults with cardiogenic shock. Circu-
lation 2015;131:676e80.
[9] Rastan AJ, Dege A, Mohr M, et al. Early and late outcomes of 517 consecutive adult patients treated with extracorporeal
membrane oxygenation for refractory postcardiotomy cardiogenic shock. J Thorac Cardiovasc Surg 2010;139:302e11.
[10] Thiagarajan RR, Brogan TV, Scheurer MA, et al. Extracorporeal membrane oxygenation to support cardiopulmonary
resuscitation in adults. Ann Thorac Surg 2009;87:778e85.
[11] Le Guen M, Nicolas-Robin A, Carreira S, et al. Extracorporeal life support following out-of-hospital refractory cardiac
arrest. Crit Care 2011;15:R2.
[12] Tsuneyoshi H, Rao V. The role of extracorporeal membrane oxygenation (ECMO) therapy in acute heart failure. Int
Anesthesiol Clin 2012;50:114e22.
*[13] Catena E, Tasca G. Role of echocardiography in the perioperative management of mechanical circulatory assistance. Best
Pract Res Clin Anaesthesiol 2012;26:199e216.
[14] Cheung AW, White CW, Davis MK, et al. Short-term mechanical circulatory support for recovery from acute right ven-
tricular failure: clinical outcomes. J Heart Lung Transpl 2014;33:794e9.
[15] Weiss S, Jolly N, Shah AP. Multivessel intervention and placement of a percutaneous right ventricular assist device in a
patient with acute myocardial infarction complicated by cardiac arrest. J Invasive Cardiol 2011;23:248e51.
[16] Kiernan MS, Krishnamurthy B, Kapur NK. Percutaneous right ventricular assist via the internal jugular vein in cardiogenic
shock complicating an acute inferior myocardial infarction. J Invasive Cardiol 2010;22:E23e6.
[17] Kowalczyk AK, Mizuguchi KA, Couper GS, et al. Use of intraoperative transesophageal echocardiography to evaluate
positioning of TandemHeart percutaneous right ventricular assist device cannulae. Anesth Analg 2014;118:72e5.
[18] Thiele H, Sick P, Boudriot E, et al. Randomized comparison of intra-aortic balloon support with a percutaneous left
ventricular assist device in patients with revascularized acute myocardial infarction complicated by cardiogenic shock.
Eur Heart J 2005;26:1276e83.
[19] Burkhoff D, Cohen H, Brunckhorst C, et al., TandemHeart Investigators Group. A randomized multicenter clinical study to
evaluate the safety and efcacy of the TandemHeart percutaneous ventricular assist device vs. conventional therapy with
intraaortic balloon pumping for treatment of cardiogenic shock. Am Heart J 2006;152:469.e1e8.
[20] Seyfarth M, Sibbing D, Bauer I, et al. A randomized clinical trial to evaluate the safety and efficacy of a percutaneous left
ventricular assist device vs. intra-aortic balloon pumping for treatment of cardiogenic shock caused by myocardial
infarction. J Am Coll Cardiol 2008;52:1584e8.
*[21] Werdan K, Gielen S, Ebelt H, et al. Mechanical circulatory support in cardiogenic shock. Eur Heart J 2014;35:156e67.
[22] John R, Liao K, Lietz K, et al. Experience with the Levitronix CentriMag circulatory support system as a bridge to decision
in patients with refractory acute cardiogenic shock and multisystem organ failure. J Thorac Cardiovasc Surg 2007;134:
351e8.
[23] John R, Long JW, Massey HT, et al. Outcomes of a multicenter trial of the Levitronix CentriMag ventricular assist system
for short-term circulatory support. J Thorac Cardiovasc Surg 2011;141:932e9.
[24] Thomas HL, Dronavalli VB, Parameshwar J, et al., Steering Group of the UK Cardiothoracic Transplant Audit. Incidence and
outcome of Levitronix CentriMag support as rescue therapy for early cardiac allograft failure: a United Kingdom national
study. Eur J Cardiothorac Surg 2011;40:1348e54.
*[25] Miller LW, Pagani FD, Russell SD, et al., HeartMate II Clinical Investigators. Use of a continuous-ow device in patients
awaiting heart transplantation. N Engl J Med 2007;357:885e96.
[26] Starling RC, Naka Y, Boyle AJ, et al. Results of the post-U.S. Food and Drug Administration-approval study with a
continuous ow left ventricular assist device as a bridge to heart transplantation: a prospective study using the
INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support). J Am Coll Cardiol 2011;57:1890e8.
*[27] Aaronson KD, Slaughter MS, Miller LW, et al., HeartWare Ventricular Assist Device (HVAD) Bridge to Transplant ADVANCE
Trial Investigators. Use of an intrapericardial, continuous-ow, centrifugal pump in patients awaiting heart trans-
plantation. Circulation 2012;125:3191e200.
*[28] Rose EA, Gelijns AC, Moskowitz AJ, et al., Randomized Evaluation of Mechanical Assistance for the Treatment of
Congestive Heart Failure (REMATCH) Study Group. Long-term use of a left ventricular assist device for end-stage heart
failure. N Eng J Med 2001;345:1435e43.
*[29] Slaughter MS, Rogers JG, Milano CA, et al. Advanced heart failure treated with continuous-ow left ventricular assist
device. N Engl J Med 2009;361:2241e51.
[30] Sabashnikov A, Mohite PN, Weymann A, et al. Outcomes after implantation of 139 full-support continuous-ow left
ventricular assist devices as a bridge to transplantation. Eur J Cardiothorac Surg 2014;46:e59e66.
[31] Baric D. Why pulsatility still matters: a review of current knowledge. Croat Med J 2014;55:609e20.
*[32] Cheng A, Williamitis CA, Slaughter MS. Comparison of continuous-ow and pulsatile-ow left ventricular assist devices:
is there an advantage to pulsatility? Ann Cardiothorac Surg 2014;3:573e81.
*[33] Feldman D, Pamboukian SV, Teuteberg JJ, et al., International Society for Heart and Lung Transplantation. The 2013 In-
ternational Society for Heart and Lung Transplantation Guidelines for mechanical circulatory support: executive sum-
mary. J Heart Lung Transpl 2013;32:157e87.
[34] Russell SD, Rogers JG, Milano CA, et al., HeartMate II Clinical Investigators. Renal and hepatic function improve in
advanced heart failure patients during continuous-ow support with the HeartMate II left ventricular assist device.
[35] Kirklin JK, Naftel DC, Kormos RL, et al. Quantifying the effect of cardiorenal syndrome on mortality after left ventricular
assist device implant. J Heart Lung Transpl 2013;32:1205e13.
[36] Matthews JC, Pagani FD, Haft JW, et al. Model for end-stage liver disease score predicts left ventricular assist device
operative transfusion requirements, morbidity, and mortality. Circulation 2010;121:214e20.
[37] Vivo RP, Cordero-Reyes AM, Qamar U, et al. Increased right-to-left ventricle diameter ratio is a strong predictor of right
ventricular failure after left ventricular assist device. J Heart Lung Transpl 2013;32:792e9.
[38] Topilsky Y, Oh JK, Shah DK, et al. Echocardiographic predictors of adverse outcomes after continuous left ventricular assist
device implantation. JACC Cardiovasc Imaging 2011;4:211e22.
[39] Potapov EV, Stepanenko A, Dandel M, et al. Tricuspid incompetence and geometry of the right ventricle as predictors of
right ventricular function after implantation of a left ventricular assist device. J Heart Lung Transpl 2008;27:1275e81.
[40] Puwanant S, Hamilton KK, Klodell CT, et al. Tricuspid annular motion as a predictor of severe right ventricular failure after
left ventricular assist device implantation. J Heart Lung Transpl 2008;27:1102e7.
[41] Kato TS, Jiang J, Schulze PC, et al. Serial echocardiography using tissue Doppler and speckle tracking imaging to monitor
right ventricular failure before and after left ventricular assist device surgery. JACC Heart Fail 2013;1:216e22.
[42] Kiernan MS, French AL, DeNofrio D, et al. Preoperative three dimensional echocardiography to assess risk of right ven-
tricular failure following left ventricular assist device surgery. J Card Fail 2015;21:189e97.
[43] Wang TS, Hernandez AF, Felker GM, et al. Valvular heart disease in patients supported with left ventricular assist devices.
Circ Heart Fail 2014;7:215e22.
[44] Chumnanvej S, Wood MJ, MacGillivray TE, et al. Perioperative echocardiographic examination for ventricular assist de-
vice implantation. Anesth Analg 2007;105:583e601.
[45] Meineri M, Van Rensburg AE, Vegas A. Right ventricular failure after LVAD implantation: prevention and treatment. Best
Pract Res Clin Anaesthesiol 2012;26:217e29.
[46] Ziemba EA, John R. Mechanical circulatory support for bridge to decision: which device and when to decide. J Card Surg
2010;25:425e33.
[47] Kormos RL, Teuteberg JJ, Pagani FD, et al., HeartMate II Clinical Investigators. Right ventricular failure in patients with the
HeartMate II continuous-ow left ventricular assist device: incidence, risk factors, and effect on outcomes. J Thorac
[48] Marzec LN, Ambardekar AV. Preoperative evaluation and perioperative management of right ventricular failure after left
ventricular assist device implantation. Semin Cardiothorac Vasc Anesth 2013;17:249e61.
[49] Potapov E, Meyer D, Swaminathan M, et al. Inhaled nitric oxide after left ventricular assist device implantation: a pro-
spective, randomized, double-blind, multicenter, placebo-controlled trial. J Heart Lung Transpl 2011;30:870e8.
[50] Groves DS, Blum FE, Huffmyer JL, et al. Effects of early inhaled epoprostenol therapy on pulmonary artery pressure and
blood loss during LVAD placement. J Cardiothorac Vasc Anesth 2014;28:652e60.
[51] Antoniou T, Prokakis C, Athanasopoulos G, et al. Inhaled nitric oxide plus iloprost in the setting of post-left assist device
right heart dysfunction. Ann Thorac Surg 2012;94:792e8.
[52] Hamdan R, Mansour H, Nassar P, et al. Prevention of right heart failure after left ventricular assist device implantation by
phosphodiesterase 5 inhibitor. Artif Organs 2014;38:963e7.
[53] Tedford RJ, Hemnes AR, Russell SD, et al. PDE5A inhibitor treatment of persistent pulmonary hypertension after me-
chanical circulatory support. Circ Heart Fail 2008;1:213e9.
[54] Kukucka M, Potapov E, Stepanenko A, et al. Acute impact of left ventricular unloading by left ventricular assist device on
the right ventricle geometry and function: effect of nitric oxide inhalation. J Thorac Cardiovasc Surg 2011;141:1009e14.
[55] Theiss HD, Grabmaier U, Kreissl N, et al. Preconditioning with levosimendan before implantation of left ventricular assist
devices. Artif Organs 2014;38:231e4.
[56] Sponga S, Ivanitskaia E, Potapov E, et al. Preoperative treatment with levosimendan in candidates for mechanical cir-
culatory support. ASAIO J 2012;58:6e11.
[57] Omar S, Zedan A, Nugent K. Cardiac vasoplegia syndrome: pathophysiology, risk factors and treatment. Am J Med Sci
2015;349:80e8.
[58] Levin MA, Lin HM, Castillo JG, et al. Early on-cardiopulmonary bypass hypotension and other factors associated with
vasoplegic syndrome. Circulation 2009;120:1664e71.
[59] Gilbert M, Lema G. [Vasoplegic syndrome and its treatment with vasopressin during cardiac surgery with cardiopul-
monary bypass]. Rev Med Chil 2011;139:368e72.
[60] Currigan DA, Hughes RJ, Wright CE, et al. Vasoconstrictor responses to vasopressor agents in human pulmonary and
radial arteries: an in vitro study. Anesthesiology 2014;121:930e6.
[61] Shuhaiber JH, Jenkins D, Berman M, et al. The Papworth experience with the Levitronix CentriMag ventricular assist
device. J Heart Lung Transpl 2008;27:158e64.
[62] Haneya A, Philipp A, Puehler T, et al. Temporary percutaneous right ventricular support using a centrifugal pump in
patients with postoperative acute refractory right ventricular failure after left ventricular assist device implantation. Eur J
Cardiothorac Surg 2012;41:219e23.
[63] Farrar DJ. The thoratec ventricular assist device: a paracorporeal pump for treating acute and chronic heart failure. Semin
Thorac Cardiovasc Surg 2000;12:243e50.
[64] Krabatsch T, Potapov E, Stepanenko A, et al. Biventricular circulatory support with two miniaturized implantable assist
devices. Circulation 2011;13(124):S179e86.
[65] Takeda K, Naka Y, Yang JA, et al. Outcome of unplanned right ventricular assist device support for severe right heart
failure after implantable left ventricular assist device insertion. J Heart Lung Transpl 2014;33:141e8.
[66] Stulak JM, Romans T, Cowger J, et al. Delayed sternal closure does not increase late infection risk in patients undergoing
left ventricular assist device implantation. J Heart Lung Transpl 2012;31:1115e9.
[67] Bruckner BA, DiBardino DJ, Ning Q, et al. High incidence of thromboembolic events in left ventricular assist device pa-
tients treated with recombinant activated factor VII. J Heart Lung Transpl 2009;28:785e90.
[68] Hurlburt L, Roscoe A, van Rensburg A. The use of prothrombin complex concentrates in two patients with non-pulsatile
left ventricular assist devices. J Cardiothorac Vasc Anesth 2014;28:345e6.
[69] Sarosiek K, Kang CY, Johnson CM, et al. Perioperative use of the imacor hemodynamic transesophageal echocardiography
probe in cardiac surgery patients: initial experience. ASAIO J 2014;60:553e8.
[70] Cavarocchi NC, Pitcher HT, Yang Q, et al. Weaning of extracorporeal membrane oxygenation using continuous hemo-
dynamic transesophageal echocardiography. J Thorac Cardiovasc Surg 2013;146:1474e9.
[71] Andersen M, Videbaek R, Boesgaard S, et al. Incidence of ventricular arrhythmias in patients on long-term support with a
continuous-ow assist device (HeartMate II). J Heart Lung Transpl 2009;28:733e5.
[72] Genovese EA, Dew MA, Teuteberg JJ, et al. Incidence and patterns of adverse event onset during the rst 60 days after
ventricular assist device implantation. Ann Thorac Surg 2009;88:1162e70.
[73] Allen SJ, Sidebotham D. Postoperative care and complications after ventricular assist device implantation. Best Pract Res
Clin Anaesthesiol 2012;26:231e46.
[74] Meyer AL, Malehsa D, Budde U, et al. Acquired von Willebrand syndrome in patients with a centrifugal or axial
continuous ow left ventricular assist device. JACC Heart Fail 2014;2:141e5.
[75] Crow S, Chen D, Milano C, et al. Acquired von Willebrand syndrome in continuous-ow ventricular assist device re-
cipients. Ann Thorac Surg 2010;90:1263e9.
[76] Uriel N, Pak SW, Jorde UP, et al. Acquired von Willebrand syndrome after continuous-ow mechanical device support
contributes to a high prevalence of bleeding during long-term support and at the time of transplantation. J Am Coll
Cardiol 2010;56:1207e13.
[77] Demirozu ZT, Radovancevic R, Hochman LF, et al. Arteriovenous malformation and gastrointestinal bleeding in patients
with the HeartMate II left ventricular assist device. J Heart Lung Transpl 2011;30:849e53.
[78] Stern DR, Kazam J, Edwards P, et al. Increased incidence of gastrointestinal bleeding following implantation of the
HeartMate II LVAD. J Card Surg 2010;25:352e6.
[79] Birati EY, Rame JE. Diagnosis and management of LVAD. Thromb Curr Treat Options Cardiovasc Med 2015;17:361.
[80] Warkentin TE, Greinacher A, Koster A. Heparin-induced thrombocytopenia in patients with ventricular assist devices: are
new prevention strategies required? Ann Thorac Surg 2009;87:1633e40.
[81] Copeland JG, Copeland H, Gustafson M, et al. Experience with more than 100 total articial heart implants. J Thorac
[82] Sale SM, Smedira NG. Total articial heart. Best Pract Res Clin Anaesthesiol 2012;26:147e65.
*[83] Hessel 2nd EA. Management of patients with implanted ventricular assistdevices for noncardiac surgery: a clinical re-
view. Semin Cardiothorac Vasc Anesth 2014;18:57e70.
[84] Chacon MM, Hattrup EA, Shillcutt SK. Perioperative management of two patients with left ventricular assist devices
presenting for noncardiac surgery in the prone position. A A Case Rep 2014;2:70e3.
[85] Chestovich PJ, Kwon MH, Cryer HG, et al. Surgical procedures for patients receiving mechanical cardiac support. Am Surg
2011;77:1314e7.
[86] Riha H, Netuka I, Kotulak T, et al. Anesthesia management of a patient with a ventricular assist device for noncardiac
surgery. Semin Cardiothorac Vasc Anesth 2010;14:29e31.
[87] Stone ME, Soong W, Krol M, et al. The anesthetic considerations in patients with ventricular assist devices presenting for
noncardiac surgery: a review of eight cases. Anesth Analg 2002;95:42e9.
[88] Brown JB, Hallinan WM, Massey HT, et al. Does the need for noncardiac surgery during ventricular assist device therapy
impact clinical outcome? Surgery 2009;146:627e33.
[89] Stehlik J, Nelson DM, Kfoury AG, et al. Outcome of noncardiac surgery in patients with ventricular assist devices. Am J
Cardiol 2009;103:709e12.

Anaesthesiology
10
Extracorporeal life support for adult

cardiopulmonary failure
Basil W. Schaheen, MD, Resident a,
Robert H. Thiele, MD, Assistant Professor of Anesthesiology
and Biomedical Engineering, Director, Technology in
Anesthesia and Critical Care Group, Co-Director, Enhanced
Recovery after Surgery (ERAS) Program b, 1,
James M. Isbell, MD, MSCI, Assistant Professor of Surgery,
Director, Adult Extracorporeal Life Support Program,
Co-Director, Thoracic and Cardiovascular Critical Care c, *
a
General Surgery, University of Virginia, P.O. Box 800681, Charlottesville, VA 22908, USA
b
Department of Anesthesiology, University of Virginia Health System, P.O. Box 800710, Charlottesville, VA
22908-0710, USA
c
Department of Surgery, Division of Thoracic and Cardiovascular Surgery, University of Virginia Health
System, P.O. Box 800679, Charlottesville, VA 22908, USA
Keywords:
The use of extracorporeal life support (ECLS) or extracorporeal
extracorporeal life support/extracorporeal
membrane oxygenation (ECMO), as it is also known, has rapidly
acute respiratory distress syndrome/acute expanded over the past decade. The increase in ECMO use is a
lung injury consequence of multiple factors including signicant advance-
cardiogenic shock/postcardiotomy shock ments in extracorporeal technology, the emergence of data sup-
pumpless extracorporeal lung assist porting its use, and a growing number of potential clinical
applications. This review focuses on the various modes of ECLS as
well as the clinical indications and available evidence for the use of
extracorporeal support.
* Corresponding author. Tel.: 1 (434) 243 6443; Fax: 1 (434) 244 9429.
E-mail addresses: Bs2cd@virginia.edu (B.W. Schaheen), thiele@virginia.edu (R.H. Thiele), james.isbell@virginia.edu (J.M.
Isbell).
1
Tel.: 1 (434) 243 9412.
230 B.W. Schaheen et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 229e239
Introduction
Extracorporeal life support (ECLS), also known as extracorporeal membrane oxygenation (ECMO), is
a lifesaving technology that employs a modied form of cardiopulmonary bypass for patients with
severe cardiac and/or pulmonary failure. The report of the rst ECMO survivor in 1972 generated
signicant excitement for the use of this technology [1]. However, this enthusiasm was tempered after
the US National Institutes of Health (NIH) sponsored a randomized trial comparing ECMO to con-
ventional mechanical ventilation in the 1970s that demonstrated 90% mortality in both treatment
groups [2]. Although multiple studies have since established ECMO as an effective rescue strategy for
circulatory and pulmonary failure in the neonatal and pediatric populations, until recently, the
application of ECMO in adults had been slow to gather broad support. Advancements in ECMO tech-
nology along with multiple reports of its successful use in patients with recalcitrant respiratory failure
during the 2009e2010 H1N1 inuenza pandemic have led to a signicant rise in its use. The adult
patient populations treated with ECLS and their overall outcomes as reported in the International
Registry for the Extracorporeal Life Support Organization (ELSO) are provided in Table 1. Additionally,
with modern ECMO technology's improved safety prole, the potential clinical applications for
extracorporeal support are rapidly expanding. In this review, we will discuss the indications and po-
tential cannulation sites available for each of the more common modes of ECLS in adults. We will also
highlight the evidence for ECLS and briey discuss newer applications for the technology.
Basic circuit conguration
ECLS involves the removal of circulating venous blood that is then delivered by a pump to a
membrane oxygenator, which introduces oxygen and removes carbon dioxide (CO2) from the blood.
Oxygenated blood can then be brought to the desired temperature and returned to the patient's sys-
temic circulation. In patients with primary respiratory failure with adequate cardiac function, a
venovenous (VV) approach is preferred. With the VV technique, blood is both withdrawn and returned
to the patient via the venous circulation. A venoarterial (VA) approach is typically employed in patients
with primary circulatory failure or in patients with both circulatory and pulmonary failure. With this
arrangement, blood is withdrawn from the venous circulation and returned to the arterial circulation,
thereby bypassing the heart and lungs. Specic cannulation strategies for each form of ECLS will be
discussed in more detail below.
Centrifugal pump
Most adult ECMO centers have transitioned to centrifugal pumps and away from older roller pumps.
The newest centrifugal pump design employs a magnetically suspended, rotating impeller that propels
blood to the oxygenator while also minimizing blood stasis, thrombosis, and hemolysis. Another
benet of centrifugal pumps is the low priming volume, decreasing the need for red cell transfusion.
Membrane oxygenator
Contemporary ECLS circuits designed for longer-term support (i.e., greater than a few hours) typi-
cally incorporate hollow-core ber membrane oxygenators. Although most are only approved by the US
Food and Drug Administration for up to 6 hours of use, the newer oxygenators can often last for days to
Table 1
Overall Outcomes of Adult Patients in the International Extracorporeal Life Support Organization Registry Adapted from Paden
ML, Rycus PT, Thiagarajan RR. Update and outcomes in extracorporeal life support. Seminars in perinatology. 2014; 38 (2):65e70.
Indication Number of patients Survived ECLS (%) Survival to discharge or transfer (%)
Respiratory 7008 4587 (65%) 4026 (57%)

Cardiac 5603 3129 (56%) 2294 (41%)
ECPR 1657 639 (39%) 471 (28%)
B.W. Schaheen et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 229e239 231
weeks (off label use) when anticoagulation is managed appropriately. These oxygenators work by
allowing the so-called sweep gas to ow through the hollow bers while blood ows external to the
bers. The composition of the sweep gas is regulated by using an aireoxygen blender, which mixes
ambient air with oxygen to control the fraction of oxygen delivered to the oxygenator. Despite their
name, oxygenators are far less efcient at oxygenating blood than removing CO2 due to the increased
solubility of the latter. The central determinant of PaCO2 clearance in patients on ECMO is sweep gas
ow, while the main driver of PaO2 is circuit ow. ECMO ow rates are typically set based on the pa-
tient's cardiac output and overall oxygen consumption. It should be noted that ow rates are more often
limited by the size of the venous (inow) drainage cannula than the arterial (outow) cannula.
Anticoagulation
Most modern ECMO circuit components are coated with heparin or biocompatible materials, which
has markedly reduced the level of anticoagulation required. Nonetheless, some anticoagulation is still
necessary in most circumstances. Unfractionated heparin is most commonly used. In patients who are
not bleeding, heparin is dosed to achieve an anti-Xa activity level in the range of 0.3e0.5 IU/mL,
although it should be noted that there is wide variation across centers in target goals for anti-
coagulation. Relatively recent innovations in circuit components have made ECMO safer and easier to
implement with far fewer bleeding complications.
The clinical applications of ECMO have expanded considerably over the past decade, and they are
listed in Table 2. The remainder of this review will detail many of these applications along with their
indications and cannulation strategies.
Indications and clinical evidence
ECMO for adult respiratory failure
In general, VV ECMO should be considered for patients with reversible hypoxemic or uncompen-
sated hypercarbic respiratory failure that is refractory to optimal mechanical ventilation strategies.
Hypoxemic respiratory failure

The rst successful use of ECMO was for the treatment of severe acute respiratory failure in a 24-
year-old trauma victim in 1971 [1]. This early success in the treatment of severe respiratory failure
led to a NIH-sponsored randomized trial in the 1970s, which demonstrated no survival benet in those
patients receiving ECMO support [2]. However, the trial design was awed in many ways. The study
used only the VA mode of support along with rst-generation ECMO technology requiring high levels
of anticoagulation, which led to excessive bleeding complications. In addition, both the treatment and
control arms used what would be considered today injurious mechanical ventilator settings with high
tidal volumes and pressures. Ultimately, survival in each group was less than 10%. Despite these
abysmal results, several centers continued to use ECMO for the treatment of refractory respiratory
failure, and they have reported generally favorable outcomes [3e7].
Table 2
Indications for venovenous and venoarterial ECMO.
Venovenous Venoarterial
Hypoxemic respiratory failure Refractory cardiogenic shock

Hypercarbic respiratory failure Fulminant myocarditis
Bridge to lung transplantation Postcardiotomy shock
Surgery necessitating apnea Sepsis-induced cardiomyopathy
Severe air leak Massive pulmonary embolism
Refractory cardiac arrest
Bridge to heart transplantation or ventricular assist device
Primary graft failure following heart transplantation
The true tipping point for the use of ECMO in the setting of severe adult respiratory failure came in
2009 in the midst of the worldwide H1N1 inuenza pandemic, which disproportionately affected
younger rather than older adults in contrast to typical seasonal inuenza patterns. A signicant pro-
portion of these younger H1N1 patients failed to improve with advanced mechanical ventilation
strategies, prompting the rapid expansion of ECMO utilization in many intensive care units (ICUs)
across the world. Several retrospective cohorts from the 2009e2010 H1N1 inuenza experience
demonstrated outcomes favoring the use of ECMO in patients with severe acute respiratory distress
syndrome (ARDS) (median PaO2 to FiO2 ratio <60) [8e11]. However, the efcacy of ECMO in H1N1-
related ARDS remains uncertain as patients with similar disease severity managed without ECMO
had comparable outcomes in some studies [12e14]. Even after the H1N1 pandemic subsided, the use of
ECMO support continued to rise, possibly spurred by the availability of a new dual-lumen cannula
(Avalon Elite, Maquet, Wayne, NJ, USA) as well as the results of the multicenter, randomized
controlled trial comparing conventional ventilatory support versus ECMO for severe adult respiratory
failure (CESAR) [15,16].
The CESAR trial randomized 180 adults with severe respiratory failure to either continued con-
ventional management or referral to a specialized center where patients were managed under a
standardized ARDS treatment protocol, which included consideration for treatment with VV ECMO.
The primary outcome measure was death or severe disability at 6 months, which occurred in 37% of the
patients referred for ECMO consideration and 53% of the patients in the conventional treatment group
(RR: 0.96, 95% CI: 0.05e0.97; P 0.03) [16]. Of note, among the 90 patients randomly assigned for
transfer to an ECMO center, only 68 (76%) actually received ECMO support. The trial had two principal
limitations that should be considered. Of the 22 patients who were randomized to the ECMO group but
did not receive ECMO, three died before transfer and two died in transit, while the others either
improved with conventional management or had a contraindication to anticoagulation. The vast
majority of patients (93%) cared for at-the-specialized ECMO center were managed with lung-
protective ventilation, whereas only 70% of those in the conventional management group received
protective ventilation. Despite the methodological issues with the CESAR trial, its results support the
development of specialized treatment centers with expertise in the care of patients with severe ARDS,
including the use of ECMO.
Due to the limitations of the CESAR trial, additional randomized trials are needed to evaluate the
efcacy of ECMO in severe respiratory failure. Toward this goal, there is an ongoing international,
multicenter, randomized ECMO for severe ARDS (EOLIA) trial, which will compare early VV ECMO in
ARDS to a control (non-ECMO) group managed under a strict, protocol-driven protective lung-
ventilation strategy [17,18]. Unlike the CESAR trial, all patients assigned to the ECMO treatment arm
will be placed on ECMO before being transported to specialized centers. If this trial design proves
successful in achieving early implementation of ECMO as intended (within 3e6 h of worsening ARDS),
it should lend important insights into the utility of early ECMO, which is thought to be superior to later
or rescue implementation. The EOLIA trial is approaching completion of enrollment with results ex-
pected in early 2016.
In the meantime, according to the ELSO guidelines for the management of adult respiratory failure,
ECMO should be considered when the risk of mortality is 50%, and it is indicated when the risk of
mortality is 80% [19]. A PaO2/FiO2 <150 mm Hg on a FiO2 of 90% or a Murray score (Table 3) of 2e3 is
associated with a 50% mortality risk, whereas a PaO2/FiO2 <100 mm Hg on a FiO2 of 90% or a Murray
score of 3e4 despite optimal management of the mechanical ventilator is associated with 80%
mortality.
Hypercarbic respiratory failure

Extracorporeal support should also be considered in patients with uncompensated hypercarbia
with acidemia (pH < 7.15) or those with end-inspiratory plateau pressures above 30 cm of water [20].
The extracorporeal removal of CO2 in this setting permits the use of low tidal volume ventilation, and
this in turn minimizes barotrauma and atelectrauma [4,21e23]. Extracorporeal CO2 removal (ECCO2R)
has been successfully deployed to wean patients with hypercarbic respiratory failure from mechanical
ventilation [24e26]. Aside from traditional VV ECMO, there are now several devices available in Europe
and Canada (but none yet approved by the US Food and Drug Administration) for ECCO2R. One such
Table 3
Lung Injury Score- Murray Score Adapted from Murray JF,
Matthay MA, Luce JM, Flick MR. An expanded denition of
the adult respiratory distress syndrome. The American re-
view of respiratory disease. 988; 138 (3):720e3.
1-Chest X-Ray (Quadrants of alveolar consolidation)

No alveolar consolidation 0
1 quadrant 1
2 quadrant 2
3 quadrant 3
All 4 quadrants 4
2-Hypoxemia (PaO2/FiO2)
300 mmHg 0
225-299 mmHg 1
175-224 mmHg 2
100-174 mmHg 3
<100 mmHg 4
3-PEEP (cm H2O)

PEEP 5 0
PEEP 6-8 1
PEEP 9-11 2
PEEP 12-14 3
PEEP 15 4
4-Respiratory Compliance (mL/cm H2O)

80 0
60-79 1
40-59 2
20-39 3
19 4
The score is obtained by addition of the individual com-

ponents.
0 no lung injury; 1e2.5 mild to moderate injury; >2.5
severe injury.
device is the Novalung iLA Membrane Ventilator (Hechingen, Germany), the only pumpless extra-
corporeal lung assist (PECLA) system available. It uses the patient's own systemic blood pressure via
arteriovenous cannulation to propel blood through a low-resistance hollow ber membrane lung.
Other systems including Hemolung (ALung, Pittsburgh, PA, USA) and Decap (Hemodec, Salerno,
Italy) use a pump, but they are more similar to continuous VV hemoltration devices than to traditional
VV ECMO. Although these systems are very efcient at removing CO2, they are poor at oxygenating the
blood given their relatively low ow rates (<25% of cardiac output).
Bridge to lung transplantation

In the past, ECMO was considered a contraindication to lung transplantation, but with the improved
safety prole of modern ECMO, several groups have demonstrated excellent outcomes with the use of
pre-transplant extracorporeal support [27e31]. In some circumstances, VV ECMO via a dual-lumen
cannula can provide sufcient gas exchange to allow for the removal of mechanical ventilatory sup-
port for patients awaiting lung transplantation. By avoiding femoral cannulation, such patients can also
actively participate in physical therapy while on ECMO [32].
Cannulation strategies
Until recently, VV ECMO was commonly performed through a femoral venous drainage cannula and
a jugular return cannula. These cannulae are typically placed percutaneously using the Seldinger
technique. The introduction of a new bicaval dual-lumen cannula (Avalon Elite, Maquet, Wayne, NJ,
USA) facilitates single-site, right internal jugular percutaneous access. For placement of the dual-lumen
cannula, the right internal jugular vein is accessed using the Seldinger technique with the catheter
positioned over a wire. Fluoroscopic and/or transoesophageal echocardiographic (TOE) guidance is

used to position the catheter such that the distal end of the catheter sits in the proximal inferior vena
cava, and the outow jet is directed across the tricuspid valve. With the proximal and distal inow
ports positioned in the superior vena cava and the inferior vena cava, respectively, this cannula
arrangement minimizes the recirculation of oxygenated blood within the ECMO circuit. While place-
ment of the dual-lumen cannula can be accomplished safely with TOE guidance alone, the use of
uoroscopy, when available, makes the placement of the cannula simpler. Several centers have
reported excellent results using the dual-lumen cannula [33e35]. Aside from the ease of percutaneous
placement, another benet of the cannula is that its right internal jugular location allows for the
ambulation of appropriate patients on ECMO.
ECMO for circulatory failure
VA ECMO plays an increasing role in the management of various forms of circulatory shock. While
VA ECMO has been used for some time in postcardiotomy cardiogenic shock as rescue therapy, newer
applications for its use are rapidly emerging with advancements in technology.
Cardiogenic shock
Patients with severe cardiogenic shock, regardless of the underlying etiology, progress to multi-
system organ failure and death if measures are not quickly undertaken to ensure adequate end-organ
tissue perfusion. Despite improvements in the management of cardiogenic shock, mortality remains
over 40% [36]. Vasopressor and inotropic support are the mainstays of therapy to increase cardiac
output; however, these medications also increase myocardial oxygen demand and decrease tissue
microcirculation [37]. To counter the deleterious effects of these medications and to further augment
cardiac output, the use of intra-aortic balloon pump (IABP) support was routinely advocated until the
recent publication of the IABP-SHOCK II trial [38]. This randomized study enrolled 600 patients who
were undergoing early revascularization for AMI complicated by cardiogenic shock to IABP or con-
ventional therapy. There was no difference between the two groups in a 30-day mortality, the primary
end point, or any of the secondary end points (serum lactate, catecholamine use, renal function or ICU
length of stay). The results of the IABP-SHOCK II trial have engendered interest in other forms of
temporary, percutaneous mechanical circulatory support (MCS). Aside from VA ECMO, there are other
percutaneous MCS devices available including the TandemHeart left ventricular assist device (LVAD)
(CardiacAssist, Inc., Pittsburgh, PA, USA), the 2.5 and 5.0 L Impella devices (Abiomed, Danvers, MA,
USA), and the new pulsatile iVAC 2L (PulseCath BV, The Netherlands) device [37]. Among the short-
term MCS devices currently available, VA ECMO may offer some advantages in the setting of cardio-
genic shock including relatively easy percutaneous VA access; full circulatory support for right ven-
tricular, left ventricular, or biventricular failure; and the ability to provide adequate gas exchange in the
setting of respiratory dysfunction [39]. Without the availability of randomized controlled trials
comparing VA ECMO to other MCS devices in the treatment of cardiogenic shock complicating AMI, it is
difcult to assess the efcacy of one MCS device over another. However, several observational studies
have shown possible survival benet among patients with AMI and profound cardiogenic shock when
VA ECMO was combined with coronary revascularization [40e42]. Importantly, two of these studies
used control groups over consecutive rather than concurrent time periods, potentially introducing
selection bias [40,42]. More research is needed to determine the efcacy of VA ECMO and the other
available short-term MCS devices in the treatment of cardiogenic shock.
Acute fulminant myocarditis

Based on the results of several observational studies, there appears to be a role for VA ECMO support
in the treatment of cardiogenic shock due to fulminant myocarditis [43e47]. A recent retrospective
review of the ELSO registry database reported on 147 patients with acute myocarditis from 1995 to
2011, and it found a survival-to-hospital discharge of 61%, similar to the results of other published
multicenter studies [44,48,49]. A case series of 11 patients with fulminant myocarditis was reviewed to
compare the efcacy of VA ECMO versus temporary biventricular support. Although signicantly
limited by small sample size, this study demonstrated no difference in transplant-free survival at
discharge (83% vs. 80%). Interestingly, the patients receiving VA ECMO support had a more rapid
resolution of kidney and liver dysfunction, despite having a lower ejection fraction and a higher
severity of illness prior to the initiation of mechanical support [46].
Postcardiotomy cardiogenic shock

As a distinct etiology of cardiogenic shock, postcardiotomy cardiogenic shock occurs in 0.5e1.5% of
patients who underwent postoperative adult cardiac surgery [50,51]. The application of VA ECMO in
this setting can provide necessary circulatory support until myocardial recovery or bridge to me-
chanical assist device implantation. Mortality among patients who require ECMO support following
cardiac surgery remains unacceptably high, ranging from 67% to 75% in published series [50,51].
Sepsis-induced cardiomyopathy
Evaluations of cardiac function by echocardiography in septic adult patients demonstrate that
30e60% will have some degree of myocardial dysfunction [52e54]. Historically, sepsis was considered
a contraindication to ECMO, but there is emerging observational evidence that potentially supports a
role for VA ECMO in septic patients with concomitant refractory cardiac failure [55,56]. In one cohort of
14 septic patients with severe myocardial dysfunction (median ejection fraction 16%) supported on VA
ECMO, 10 (71%) survived to discharge with good health-related quality of life at long-term follow-up
(median 13 months) [55].
Pulmonary embolus
Despite appropriate and timely intervention for patients with pulmonary embolism who present
with hemodynamic compromise, 90-day mortality rates approach 51% [57]. Outcomes of surgical
embolectomy continue to improve; however, hemodynamic compromise makes these patients very
high-risk surgical candidates. ECMO can provide hemodynamic support during the institution of
anticoagulation or until mechanical intervention can be undertaken. A retrospective review of a single
center's experience with the use of ECMO in 21 patients with massive pulmonary embolus, including
eight patients in active cardiac arrest, revealed a 62% overall survival [58]. ECMO has also been suc-
cessfully used to hemodynamically support patients while they undergo thrombolysis and catheter-
based interventions [59e61].
Extracorporeal cardiopulmonary resuscitation

With over 200,000 in-hospital cardiac arrests in the United States each year and a survival-to-
discharge rate of only 18e20%, many are exploring alternative strategies to conventional resuscitation
[62e64]. One such strategy is extracorporeal cardiopulmonary resuscitation (ECPR), which employs VA
ECMO to restore circulation in patients with ongoing cardiac arrest. Although it has yet to be studied in a
randomized fashion, ECPR has shown encouraging results in early observational studies [65e67]. One
recent prospective, propensity-matched study of in-hospital cardiac arrests comparing 46 patients who
received standard cardiopulmonary resuscitation (CPR) to 46 patients who received ECPR demonstrated
almost twice the survival among the ECPR cohort (32.6% vs. 17.4%, p < 0.0001). A 1-year mortality was
also signicantly reduced in the ECPR group (hazard ratio of 0.53, p < 0.006) [65]. In a more recent
propensity-matched study evaluating longer-term outcomes among in-hospital cardiac arrest patients,
those receiving ECPR were four times more likely to survive to 2 years with minimal neurological
impairment than those treated with conventional CPR (20% vs. 5%, p < 0.002) [67]. Currently, the ELSO
registry lists 1657 adult ECPR cases with a survival to discharge of 28% (Table 1) [15]. CPR duration of
<30 min prior to ECMO initiation is associated with improved survival [68,69]. The 2010 American Heart
Association guidelines for cardiopulmonary resuscitation indicate that there is inadequate evidence to
recommend the routine use of ECPR; however, ECPR may be considered when it is readily available, the
arrest time is short, and the underlying process that led to the cardiac arrest is likely reversible [70].
Bridge to VAD implantation or heart transplantation

VA ECMO can be used to bridge patients with severe cardiac failure to ventricular assist device
(VAD) implantation or heart transplantation. Alternatively, in patients in whom the prognosis is un-
clear, ECMO can be employed as a bridge to decision [71].
Cannulation strategies
Usually, the underlying pathology dictates the cannulation strategy to be employed in VA ECMO
patients. The level of urgency, patient body habitus, and the location in the hospital can inuence this
decision. In the setting of postcardiotomy cardiogenic shock, patients who fail to wean from cardio-
pulmonary bypass are often initiated on VA ECMO via the central cannulae placed during their index
operation.
Patients with circulatory failure unrelated to cardiac surgical procedures and lacking a median
sternotomy are most easily cannulated peripherally. Peripheral arterial cannulation sites for VA ECMO
include the femoral, axillary, and subclavian arteries. Choices for peripheral venous cannulation
include the femoral and internal jugular veins. In the setting of femoral cannulation for ECMO, vascular
complications are reported to occur in roughly 10% of patients, with 88% of vascular complications
occurring in VA ECMO patients. A clinical history of peripheral arterial disease is a strong predictor of
vascular complications following femoral cannulation [72e75].
Femoral venous drainage (inow) and femoral arterial return (outow) cannulation can both be
accomplished percutaneously using the Seldinger technique, or, alternatively, a surgical cut-down may
be employed. When femoral arterial cannulation is chosen, the placement of a distal perfusion cannula
is recommended to prevent distal limb ischemia [73,76].
Axillary or subclavian arterial cannulation and internal jugular venous cannulation are
preferred over femoral cannulation to minimize the risk of infectious complications and to allow
for patient ambulation, when appropriate. Axillary and subclavian arterial cannulation are usually
accomplished by rst sewing on a polyethylene terephthalate or polytetrauoroethylene graft in
an end-to-side fashion to the vessel. The arterial cannula can then be inserted into the end-to-side
graft.
It is important to note that VA ECMO increases left ventricular afterload. Serial echocardiography is
recommended to monitor for left ventricle distention. An early sign of decreased left ventricular
ejection is absent pulsatility in the arterial line tracing, which should prompt further evaluation with
echocardiography. If the aortic valve fails to open or the left ventricle appears distended, the decom-
pression of the left heart should be undertaken in an effort to reduce strain and myocardial oxygen
demand. An initial step that can be taken to ofoad the left ventricle is to decrease the ECMO ows,
provided adequate perfusion will be maintained. This simple maneuver is sometimes all that is needed
to allow for left ventricular ejection. When this is unsuccessful, left ventricular venting can be
accomplished surgically or via a transseptal catheter-based approach. Alternatively, some have re-
ported success with the placement of a 2.5-L Impella device to unload the left ventricle, while others
have used an IABP for this purpose with mixed results [77e79].
Summary
The use of extracorporeal support for severe cardiac and respiratory failure in adults has expanded
considerably over the past decade. The growth in ECMO is due in part to advancements in extracor-
poreal technology that have made it easier and safer to use. Additionally, there is a growing body of
evidence, albeit mostly observational, which has supported the recent rise in the use of ECMO for a
variety of evolving applications. Although the CESAR trial results do suggest a role for ECMO in patients
with severe respiratory failure, the conclusions that can be drawn from this trial are limited. Hopefully,
the upcoming results of the randomized EOLIA trial will answer some of the questions that still remain
regarding the efcacy of ECMO in severe respiratory failure. Randomized controlled trials are also
needed to further dene the role of ECMO support in refractory cardiac failure and other emerging
indications. Until these data are available, ECMO should remain a rescue therapy used only in selected
patients with presumably reversible disease processes.
Conict of interest
The authors have no conict of interest related to the content of this manuscript.
Practice Points
ECMO is a potentially lifesaving intervention that should be considered in refractory cardiac

or pulmonary failure due to a reversible cause
ECMO should be considered when conventional management strategies to support gas
exchange or circulation are failing
Venovenous ECMO is the preferred mode of support for patients with respiratory failure
refractory to conventional critical care
Venoarterial ECMO should be considered in patients with severe cardiogenic shock
Clinical applications for ECMO continue to expand; however, additional research is needed to
gauge the benefits of any novel uses for this expensive technology
Research Agenda
Additional randomized controlled trials are warranted to assess the efficacy of ECMO in the
management of specific forms of cardiopulmonary failure (e.g., ARDS, cardiogenic shock,
and massive pulmonary embolus)
Studies evaluating long-term quality of life following ECMO need to be undertaken
Acknowledgments
We would like to thank Dr. Emily A. Downs and Dr. John P. Davis for their assistance in reviewing
and editing this manuscript.
References
[1] Hill JD, O'Brien TG, Murray JJ, et al. Prolonged extracorporeal oxygenation for acute post-traumatic respiratory failure
(shock-lung syndrome). Use of the Bramson membrane lung. N Engl J Med 1972;286(12):629e34.
[2] Zapol WM, Snider MT, Hill JD, et al. Extracorporeal membrane oxygenation in severe acute respiratory failure. A ran-
domized prospective study. JAMA 1979;242(20):2193e6.
[3] Brunet F, Mira JP, Belghith M, et al. Extracorporeal carbon dioxide removal technique improves oxygenation without
causing overination. Am J Respir Crit Care Med 1994;149(6):1557e62.
[4] Gattinoni L, Pesenti A, Mascheroni D, et al. Low-frequency positive-pressure ventilation with extracorporeal CO2 removal
in severe acute respiratory failure. JAMA 1986;256(7):881e6.
[5] Hemmila MR, Rowe SA, Boules TN, et al. Extracorporeal life support for severe acute respiratory distress syndrome in
adults. Ann Surg 2004;240(4):595e605. discussion-7.
[6] Lewandowski K, Rossaint R, Pappert D, et al. High survival rate in 122 ARDS patients managed according to a clinical
algorithm including extracorporeal membrane oxygenation. Intensive Care Med 1997;23(8):819e35.
[7] Mols G, Loop T, Geiger K, et al. Extracorporeal membrane oxygenation: a ten-year experience. Am J Surg 2000;180(2):
144e54.
*[8] Davies A, Jones D, Bailey M, et al. Extracorporeal membrane oxygenation for 2009 inuenza A(H1N1) acute respiratory
distress syndrome. JAMA 2009;302(17):1888e95.
[9] Freed DH, Henzler D, White CW, et al. Extracorporeal lung support for patients who had severe respiratory failure
secondary to inuenza A (H1N1) 2009 infection in Canada. Can J Anaesth 2010;57(3):240e7.
*[10] Noah MA, Peek GJ, Finney SJ, et al. Referral to an extracorporeal membrane oxygenation center and mortality among
patients with severe 2009 inuenza A(H1N1). JAMA 2011;306(15):1659e68.
[11] Holzgraefe B, Broome M, Kalzen H, et al. Extracorporeal membrane oxygenation for pandemic H1N1 2009 respiratory
failure. Minerva Anestesiol 2010;76(12):1043e51.
[12] Miller 3rd RR, Markewitz BA, Rolfs RT, et al. Clinical ndings and demographic factors associated with ICU admission in
Utah due to novel 2009 inuenza A(H1N1) infection. Chest 2010;137(4):752e8.
[13] Morris AH, Hirshberg E, Miller 3rd RR, et al. Counterpoint: efcacy of extracorporeal membrane oxygenation in 2009
inuenza A(H1N1): sufcient evidence? Chest 2010;138(4):778e81. discussion 82e4.
[14] Pham T, Combes A, Roze H, et al. Extracorporeal membrane oxygenation for pandemic inuenza A(H1N1)-induced acute
respiratory distress syndrome: a cohort study and propensity-matched analysis. Am J Respir Crit Care Med 2013;187(3):
276e85.
[15] Paden ML, Rycus PT, Thiagarajan RR. Update and outcomes in extracorporeal life support. Semin Perinatology 2014;38(2):
65e70.
*[16] Peek GJ, Mugford M, Tiruvoipati R, et al. Efcacy and economic assessment of conventional ventilatory support versus
trial. Lancet 2009;374(9698):1351e63.
[17] Combes A, Bacchetta M, Brodie D, et al. Extracorporeal membrane oxygenation for respiratory failure in adults. Curr Opin
Crit Care 2012;18(1):99e104.
[18] Combes A. Extracorporeal membrane oxygenation for severe acute respiratory distress syndrome (EOLIA). National In-
stitutes of Health; 2015 [updated December 7, 2014; cited 2015 January 17]. Available from: https://clinicaltrials.gov/ct2/
show/NCT01470703.
*[19] ELSO guidelines for adult respiratory failure (v. 1.3). https://www.elso.org/Portals/0/IGD/Archive/FileManager/
989d4d4d14cusersshyerdocumentselsoguidelinesforadultrespiratoryfailure1.3.pdf [accessed 29.03.15].
*[20] Brodie D, Bacchetta M. Extracorporeal membrane oxygenation for ARDS in adults. New Engl J Med 2011;365(20):
1905e14.
[21] Florchinger B, Philipp A, Klose A, et al. Pumpless extracorporeal lung assist: a 10-year institutional experience. Ann
Thorac Surg 2008;86(2):410e7. discussion 7.
[22] Zimmermann M, Bein T, Arlt M, et al. Pumpless extracorporeal interventional lung assist in patients with acute respi-
ratory distress syndrome: a prospective pilot study. Crit Care 2009;13(1):R10.
*[23] Terragni PP, Del Sorbo L, Mascia L, et al. Tidal volume lower than 6 ml/kg enhances lung protection: role of extracorporeal
carbon dioxide removal. Anesthesiology 2009;111(4):826e35.
[24] Abrams DC, Brenner K, Burkart KM, et al. Pilot study of extracorporeal carbon dioxide removal to facilitate extubation and
ambulation in exacerbations of chronic obstructive pulmonary disease. Ann Am Thorac Soc 2013;10(4):307e14.
[25] Burki NK, Mani RK, Herth FJ, et al. A novel extracorporeal CO(2) removal system: results of a pilot study of hypercapnic
respiratory failure in patients with COPD. Chest 2013;143(3):678e86.
[26] Kluge S, Braune SA, Engel M, et al. Avoiding invasive mechanical ventilation by extracorporeal carbon dioxide removal in
patients failing noninvasive ventilation. Intensive Care Med 2012;38(10):1632e9.
[27] Biscotti M, Sonett J, Bacchetta M. ECMO as bridge to lung transplant. Thorac Surg Clin 2015;25(1):17e25.
[28] Chiumello D, Coppola S, Froio S, et al. Extracorporeal life support as bridge to lung transplantation: a systematic review.
Crit Care 2015;19(1):19.
[29] Hayanga AJ, Aboagye J, Esper S, et al. Extracorporeal membrane oxygenation as a bridge to lung transplantation in the
United States: an evolving strategy in the management of rapidly advancing pulmonary disease. J Thorac Cardiovasc Surg
2015;149(1):291e6.
[30] Gulack BC, Hirji SA, Hartwig MG. Bridge to lung transplantation and rescue post-transplant: the expanding role of
extracorporeal membrane oxygenation. J Thorac Dis 2014;6(8):1070e9.
[31] Javidfar J, Bacchetta M. Bridge to lung transplantation with extracorporeal membrane oxygenation support. Curr Opin
Organ Transplant 2012;17(5):496e502.
[32] Abrams D, Javidfar J, Farrand E, et al. Early mobilization of patients receiving extracorporeal membrane oxygenation: a
retrospective cohort study. Crit Care 2014;18(1):R38.
[33] Bermudez CA, Rocha RV, Sappington PL, et al. Initial experience with single cannulation for venovenous extracorporeal
oxygenation in adults. Ann Thorac Surg 2010;90(3):991e5.
[34] Javidfar J, Brodie D, Wang D, et al. Use of bicaval dual-lumen catheter for adult venovenous extracorporeal membrane
oxygenation. Ann Thorac Surg 2011;91(6):1763e8. discussion 9.
[35] Javidfar J, Wang D, Zwischenberger JB, et al. Insertion of bicaval dual lumen extracorporeal membrane oxygenation
catheter with image guidance. ASAIO J Am Soc Artif Intern Organs 1992 2011;57(3):203e5.
[36] Goldberg RJ, Spencer FA, Gore JM, et al. Thirty-year trends (1975 to 2005) in the magnitude of, management of, and
hospital death rates associated with cardiogenic shock in patients with acute myocardial infarction: a population-based
perspective. Circulation 2009;119(9):1211e9.
[37] Thiele H, Ohman EM, Desch S, et al. Management of cardiogenic shock. Eur Heart J 2015 [Epub ahead of print].
[38] Thiele H, Zeymer U, Neumann FJ, et al. Intraaortic balloon support for myocardial infarction with cardiogenic shock.
N Engl J Med 2012;367(14):1287e96.
[39] Werdan K, Gielen S, Ebelt H, et al. Mechanical circulatory support in cardiogenic shock. Eur Heart J 2014;35(3):156e67.
[40] Tsao NW, Shih CM, Yeh JS, et al. Extracorporeal membrane oxygenation-assisted primary percutaneous coronary inter-
vention may improve survival of patients with acute myocardial infarction complicated by profound cardiogenic shock.
J Crit Care 2012;27(5). 530.e1-11.
[41] Sakamoto S, Taniguchi N, Nakajima S, et al. Extracorporeal life support for cardiogenic shock or cardiac arrest due to acute
coronary syndrome. Ann Thorac Surg 2012;94(1):1e7.
[42] Sheu JJ, Tsai TH, Lee FY, et al. Early extracorporeal membrane oxygenator-assisted primary percutaneous coronary
intervention improved 30-day clinical outcomes in patients with ST-segment elevation myocardial infarction compli-
cated with profound cardiogenic shock. Crit Care Med 2010;38(9):1810e7.
[43] Mirabel M, Luyt CE, Leprince P, et al. Outcomes, long-term quality of life, and psychologic assessment of fulminant
myocarditis patients rescued by mechanical circulatory support. Crit Care Med 2011;39(5):1029e35.
[44] Diddle JW, Almodovar MC, Rajagopal SK, et al. Extracorporeal membrane oxygenation for the support of adults with acute
myocarditis. Crit Care Med 2015;43(5):1016e25.
[45] Fayssoil A, Nardi O, Orlikowski D, et al. Percutaneous extracorporeal membrane oxygenation for cardiogenic shock due to
acute fulminant myocarditis. Ann Thorac Surg 2010;89(2):614e6.
[46] Pages ON, Aubert S, Combes A, et al. Paracorporeal pulsatile biventricular assist device versus extracorporal membrane
oxygenation-extracorporal life support in adult fulminant myocarditis. J Thorac Cardiovasc Surg 2009;137(1):194e7.
[47] Asaumi Y, Yasuda S, Morii I, et al. Favourable clinical outcome in patients with cardiogenic shock due to fulminant
myocarditis supported by percutaneous extracorporeal membrane oxygenation. Eur Heart J 2005;26(20):2185e92.
[48] Acker MA. Mechanical circulatory support for patients with acute-fulminant myocarditis. Ann Thorac Surg 2001;71(3
Suppl):S73e6. discussion S82e5.
[49] Aoyama N, Izumi T, Hiramori K, et al. National survey of fulminant myocarditis in Japan: therapeutic guidelines and long-
term prognosis of using percutaneous cardiopulmonary support for fulminant myocarditis (special report from a sci-
entic committee). Circulation J Ofcial J Jpn Circulation Soc 2002;66(2):133e44.
[50] Pokersnik JA, Buda T, Bashour CA, et al. Have changes in ECMO technology impacted outcomes in adult patients
developing postcardiotomy cardiogenic shock? J Card Surg 2012;27(2):246e52.
*[51] Rastan AJ, Dege A, Mohr M, et al. Early and late outcomes of 517 consecutive adult patients treated with extracorporeal
membrane oxygenation for refractory postcardiotomy cardiogenic shock. J Thorac Cardiovasc Surg 2010;139(2):302e11.
11 e1.
[52] Pulido JN, Afessa B, Masaki M, et al. Clinical spectrum, frequency, and signicance of myocardial dysfunction in severe
sepsis and septic shock. Mayo Clin Proc 2012;87(7):620e8.
[53] Vieillard-Baron A, Caille V, Charron C, et al. Actual incidence of global left ventricular hypokinesia in adult septic shock.
Crit Care Med 2008;36(6):1701e6.
[54] Parker MM, Shelhamer JH, Bacharach SL, et al. Profound but reversible myocardial depression in patients with septic
shock. Ann Intern Med 1984;100(4):483e90.
*[55] Brechot N, Luyt C-E, Schmidt M, et al. Venoarterial extracorporeal membrane oxygenation support for refractory car-
diovascular dysfunction during severe bacterial septic shock. Crit Care Med 2013;41(7):1616e26.
[56] Huang CT, Tsai YJ, Tsai PR, et al. Extracorporeal membrane oxygenation resuscitation in adult patients with refractory
septic shock. J Thorac Cardiovasc Surg 2013;146(5):1041e6.
[57] Jaff MR, McMurtry MS, Archer SL, et al. Management of massive and submassive pulmonary embolism, iliofemoral deep
vein thrombosis, and chronic thromboembolic pulmonary hypertension: a scientic statement from the American Heart
Association. Circulation 2011;123(16):1788e830.
[58] Maggio P, Hemmila M, Haft J, et al. Extracorporeal life support for massive pulmonary embolism. J Trauma 2007;62(3):
570e6.
[59] Munakata R, Yamamoto T, Hosokawa Y, et al. Massive pulmonary embolism requiring extracorporeal life support treated
with catheter-based interventions. Int Heart J 2012;53(6):370e4.
[60] Harika R, Wells C. Successful treatment of an acute intraoperative pulmonary embolus with emergent ecmo and directed
tissue plasminogen activator. Chest 2013;144 (4_MeetingAbstracts):136A-A.
[61] Weinberg L, Kay C, Liskaser F, et al. Successful treatment of peripartum massive pulmonary embolism with extracorporeal
membrane oxygenation and catheter-directed pulmonary thrombolytic therapy. Anaesth Intensive Care 2011;39(3):
486e91.
[62] Ehlenbach WJ, Barnato AE, Curtis JR, et al. Epidemiologic study of in-hospital cardiopulmonary resuscitation in the
elderly. N Engl J Med 2009;361(1):22e31.
[63] Merchant RM, Yang L, Becker LB, et al. Incidence of treated cardiac arrest in hospitalized patients in the United States. Crit
Care Med 2011;39(11):2401e6.
[64] Kazaure HS, Roman SA, Sosa JA. Epidemiology and outcomes of in-hospital cardiopulmonary resuscitation in the United
States, 2000e2009. Resuscitation 2013;84(9):1255e60.
[65] Chen Y-S, Lin J-W, Yu H-Y, et al. Cardiopulmonary resuscitation with assisted extracorporeal life-support versus con-
ventional cardiopulmonary resuscitation in adults with in-hospital cardiac arrest: an observational study and propensity
analysis. Lancet 372(9638):554e561.
*[66] Shin TG, Choi J-H, Jo IJ, et al. Extracorporeal cardiopulmonary resuscitation in patients with inhospital cardiac arrest: a
comparison with conventional cardiopulmonary resuscitation. Crit Care Med 2011;39(1):1e17.
[67] Shin TG, Jo IJ, Sim MS, et al. Two-year survival and neurological outcome of in-hospital cardiac arrest patients rescued by
extracorporeal cardiopulmonary resuscitation. Int J Cardiol 2013;168(4):3424e30.
[68] Cardarelli MG, Young AJ, Grifth B. Use of extracorporeal membrane oxygenation for adults in cardiac arrest (E-CPR): a
meta-analysis of observational studies. ASAIO J 2009;55(6):581e6. 10.1097/MAT.0b013e3181bad907.
[69] Chen Y-S, Chao A, Yu H-Y, et al. Analysis and results of prolonged resuscitation in cardiac arrest patients rescued by
extracorporeal membrane oxygenation. J Am Coll Cardiol 2003;41(2):197e203.
[70] Cave DM, Gazmuri RJ, Otto CW, et al. Part 7: CPR techniques and devices: 2010 american heart association guidelines for
cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2010;122(18 Suppl 3):S720e8.
[71] Barth E, Durand M, Heylbroeck C, et al. Extracorporeal life support as a bridge to high-urgency heart transplantation. Clin
Transpl 2012;26(3):484e8.
[72] Bisdas T, Beutel G, Warnecke G, et al. Vascular complications in patients undergoing femoral cannulation for extracor-
poreal membrane oxygenation support. Ann Thorac Surg 2011;92(2):626e31.
[73] Stulak JM, Dearani JA, Burkhart HM, et al. ECMO cannulation controversies and complications. Semin Cardiothorac Vasc
Anesth 2009;13(3):176e82.
[74] Aziz F, Brehm CE, El-Banyosy A, et al. Arterial complications in patients undergoing extracorporeal membrane oxygen-
ation via femoral cannulation. Ann Vasc Surg 2014;28(1):178e83.
[75] Roussel A, Al-Attar N, Alkhoder S, et al. Outcomes of percutaneous femoral cannulation for venoarterial extracorporeal
membrane oxygenation support. Eur Heart J Acute Cardiovasc Care 2012;1(2):111e4.
[76] Mohite PN, Fatullayev J, Maunz O, et al. Distal limb perfusion: Achilles' heel in peripheral venoarterial extracorporeal
membrane oxygenation. Artif Organs 2014;38(11):940e4.
*[77] Petroni T, Harrois A, Amour J, et al. Intra-aortic balloon pump effects on macrocirculation and microcirculation in
cardiogenic shock patients supported by venoarterial extracorporeal membrane oxygenation. Crit Care Med 2014;42(9):
2075e82.
[78] Ma P, Zhang Z, Song T, et al. Combining ECMO with IABP for the treatment of critically Ill adult heart failure patients.
Heart Lung Circulation 2014;23(4):363e8.
[79] Cheng A, Swartz MF, Massey HT. Impella to unload the left ventricle during peripheral extracorporeal membrane
oxygenation. ASAIO J Am Soc Artif Intern Organs 1992 2013;59(5):533e6.

Anaesthesiology
11
Cardiopulmonary bypass in the pediatric

population
David Whiting, M.D., Instructor in Anaesthesia a, b, *,
Koichi Yuki, M.D., Assistant Professor in Anaesthesia a, b,
James A. DiNardo, M.D., Professor of Anaesthesia a, b
a
Department of Anesthesiology, Perioperative and Pain Medicine, Division of Cardiac Anesthesia, Boston
Children's Hospital, Boston, MA, USA
b
Department of Anaesthesia, Harvard Medical School, Boston, MA, USA
Keywords:
Cardiopulmonary bypass (CPB) facilitates the repair of cardiac le-
sions in adults and children. Surgical mortality has decreased with
congenital heart disease
ultraltration improvements in technique allowing for the successful repair of
complex heart defects in neonates of increasingly low body weight
and gestational age. The physiological effects of CPB are more
signicant in children. The presence of intracardiac shunts and
other anatomic variants further complicates CPB in patients with
congenital heart disease. Special techniques and monitors are
often necessary. Protocols are often established within individual
institutions to standardize the approach to CPB. The anesthesiol-
ogist caring for the patient must understand the physiology of CPB
to facilitate the initiation and separation from bypass, and to be
able to treat complications. Evidence supporting a particular
technique of CPB in pediatric population is still largely from un-
controlled or nonrandomized trials, observational studies,
extrapolation from adult studies, and expert opinion. The hetero-
geneity of congenital heart disease makes randomized controlled
trials or meta-analyses challenging, and thus they are limited in
the literature.
* Corresponding author. Department of Anesthesiology, Perioperative and Pain Medicine, Division of Cardiac Anesthesia,
Boston Children's Hospital, 300 Longwood Avenue, Boston, MA 02115, USA. Tel.: 1 1617 355 6225; Fax: 1 1617 730 0894.
E-mail address: david.whiting@childrens.harvard.edu (D. Whiting).
242 D. Whiting et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 241e256
History of pediatric cardiopulmonary bypass
Cardiopulmonary bypass (CPB) isolates the cardiopulmonary system to facilitate surgical repair,
oxygenates blood, and removes carbon dioxide without causing permanent damage to the blood,
cardiopulmonary system or end organs. Although the use of CPB is a routine clinical practice in cardiac
surgery, the history of CPB is relatively short. In the 1940s, cardiac operations were limited to
extracardiac procedures such as closure of patent ductus arteriosus, repair of aorta coarctation, and
creation of the BlalockeTaussig shunt [1,2]. Techniques for mechanical CPB were being investigated to
repair intracardiac lesions. The most signicant challenge in the development of CPB was the creation
of an effective oxygenator. In 1951, Dennis et al. repaired the rst endocardial cushion defect in a 6-
year-old patient using CPB with a rotating disk oxygenator; unfortunately, the patient could not be
weaned from CPB and died on the operating table [3]. In 1953, Gibbon performed the rst successful
closure of an atrial septal defect in an 18-year-old patient using CPB with a lm oxygenator [4]. This was
the rst successful cardiac repair in a human patient; however, the operation was hurried because of
clot formation in the oxygenator. One year later in 1955, Lillehei et al. performed a ventricular septal
defect repair in a 1-year-old child using cross-circulation with the patient's father who functioned as
the oxygenator in the CPB circuit.
Oxygenators further improved in the 1950s and 1960s with the development of the bubble and
membrane oxygenators, making the repair of complex congenital heart disease (CHD) possible. Mor-
tality for small infants and neonates remained high at that time, but survival improved markedly in the
1970s with the development of deep hypothermic circulatory arrest (DHCA). However, neuro-
developmental abnormalities were observed among survivors, and this concern persists today. Im-
provements in CPB have allowed for a signicant reduction in morbidity in cardiac surgery. Current
efforts seek to reduce hemodilution and the systemic inammatory response (SIR) by modifying the
technique of CPB and reducing the size of the pediatric perfusion circuit.
Components of CPB
Bypass circuits
A weight-specic bypass circuit minimizes prime volumes and hemodilution, and it reduces
exposure of the patient's blood to articial surfaces (Fig. 1). Kinetic energy to move blood through the
bypass circuit is either from a roller pump or from a centrifugal impeller pump. A double-headed roller
pump is a load-independent positive-volume displacement pump. A centrifugal impeller pumps blood
using the constrained vortex principle, and it is a load-dependent pump because shear forces increase
in the vortex and decrease forward displacement when load is increased. An inline microporous lter
with pore sizes of 30e40 mm removes particulate contaminants and prevents micro-air and macro-air
emboli from entering the arterial cannula. A bypass line ensures systemic blood ow in the case of a
lter occlusion from particulate matter or from air trapping. Cardiotomy suction returns blood to the
venous reservoir after passing through a lter. This is important for blood conservation on CPB, but it is
also the major source of blood trauma and hemolysis during CPB due to the aspiration of air and blood.
Cardiotomy blood is rich in tissue factor, pro-coagulant cellular and platelet-derived microparticles,
and it is thus a major source of inammatory mediator generation and of thrombin, neutrophil, and
platelet activation.
Oxygenator and heat exchanger
Oxygenators perform the gas-exchange functions of the lung in the CPB circuit. Gas exchange occurs
at atmospheric pressure through a thin membrane that is either solid or microporous. Microporous
membranes are arranged in hollow bers or sheets with small hydrophobic pores (0.03e0.07 mm) that
are covered with a thin proteinaceous layer on exposure to blood or albumin. The membrane
oxygenator has a reference ow rate that is determined by the manufacturer, and the smallest one is
chosen that is capable of meeting the anticipated CPB ow rates. Examples and performance charac-
teristics of membranes are listed in Table 1. Solid membranes are constructed of methyl-silicon rubber
D. Whiting et al. / Best Practice & Research Clinical Anaesthesiology 29 (2015) 241e256 243
Fig. 1. Schematic of cardiopulmonary bypass circuit.
that is <25 mm thick to facilitate the diffusion of gas. These membranes limit carbon dioxide and oxygen
diffusion, and they require a larger priming volume and a larger surface area for gas exchange than
microporous membranes. Solid nonporous membranes are more resistant to clot formation, and they
are used primarily for extracorporeal membrane oxygenation units.
A heat exchanger is often part of a combined unit with the oxygenator. Countercurrent heat ex-
changers actively cool and rewarm patient's blood to facilitate hypothermic and deep hypothermic
CPB. The temperature gradient is limited to 8e10 C to prevent large changes in gas solubility and
bubble formation, and it is limited to 40 C to avoid damage to blood. The efciency of heat exchangers
is rated as the performance factor (PF), and it is a fraction of 1.
Ventricular vents
Ventricular vents return blood collecting in the heart to the bypass circuit preventing distension of
the ventricles, facilitating surgical exposure, and preventing warming of the heart. Ventricular vents
are of particular importance in congenital heart surgery because the anatomy is variable, and intra-
cardiac shunts are common. In a structurally normal heart, coronary artery blood returns to the right
heart through the coronary sinus, and it is captured by the venous cannula, and blood from the The-
besian veins drains directly to the left ventricle and atrium, and the bronchial veins drain directly to the
pulmonary veins. In CHD, additional shunts may be present that must be identied and drained or
ligated to facilitate surgical exposure. Extracardiac left-to-right shunts must be compensated for by
increasing ows equal to the rate of shunted blood to maintain adequate systemic perfusion. Anatomic
left-to-right shunts include patent ductus arteriosus and major aortopulmonary collaterals as well as
the BlalockeTaussig, Waterston, and Potts surgical shunts. When appropriate, these shunts can be
Table 1
Characteristics of various CAPIOX FX Oxygenators (Terumo Cardiovascular Group).
Oxygenator BSA for 3.0 L/min CI Prime volume (mL) Max ow rate (mL/min) Oxygen/m2
FX-05 <0.40 43 1500 0.5

FX 15-30 144 144 4000 1.5
FX-25 250 250 7000 2.5
occluded preoperatively in the cardiac catheterization laboratory or ligated once CPB is initiated. The
vent line returns blood either to the ltered cardiotomy reservoir and then to the venous reservoir or
directly to a venous reservoir that contains a lter. The vent line may be either active or passive.
Continued active venting may result in the entrainment of air into the ventricular cavity after the
ventricle is emptied of blood.
Ultraltrators
Ultraltrators are devices that concentrate blood by removing water by convective currents across a
semipermeable membrane. Volume is removed from the bypass circuit and thus from the patient. This
technique is used with volume-overloaded patients and those with renal insufciency. An ultraltrate
is produced through a hydrostatic pressure gradient across a semipermeable membrane. The pore size
is 0.3e0.4 mm. The ultraltrate is similar in composition to glomerular ltrate. The rate of ultraltrate
production is proportional to the transmembrane pressure gradient (TMP), where TMP is determined
by the arterial inlet pressure (Pa), the venous outlet pressure (Pv), the absolute value of the applied
outlet suction (Pn), the oncotic pressure at the inlet (Pi), and the oncotic pressure at the outlet (Po):
TMP Pa Pv=2 Pn Pi Po=2
Blood ow into these devices is provided by the diversion of blood from the arterial line. Blood ow
out of the device is directed to the venous reservoir. A regulated vacuum source connected to the outlet
of the ultraltrator can increase Pn and thus the TMP.
Physiological effects of CPB: differences between adult and pediatric patients
CPB in the pediatric population has special challenges in part related to the small size of neonates
and infants relative to the size of the bypass circuit. The degree of hemodilution is signicant, and it
often requires priming with blood to maintain an acceptable hematocrit (Hct) on CPB. Pump ow rates
are higher in pediatric circuits, and bypass times are often longer, increasing mechanical trauma to the
blood. Hypothermic CPB and DHCA are common in the pediatric population, and they require acid-
ebase management at the extremes of temperature on CPB and rewarming.
Patients with congenital heart disease often have anatomic variations that complicate CPB including
atresia or duplication of vessels required for cannulation and the presence of aortopulmonary collat-
erals that can signicantly reduce effective CPB output by shunting blood into the pulmonary circu-
lation. The presence of anatomic variations may mandate alternative cannulation sites, pH
management strategies, and DHCA.
Prime volume and hemodilution
The size of the CPB circuit cannot be reduced proportionally to the size of the patient. Prime volume
depends on the size of the arterial and venous tubing, venous reservoir, and oxygenator. In our
institution, the prime volume of the smallest circuit has been reduced to 175 mL by using the smallest
possible venous reservoir, oxygenator, and reducing the length of the smallest available bypass boot.
Kinetic-assisted or vacuum-assisted venous drainage can also be used to reduce circuit volume, and
blood can be re-concentrated prior to protamine administration by ultraltration or by modied

ultraltration.
Weight-based patient blood volume (PBV) formulas are listed in Table 2. The ratio of the pump-
prime volume/PBV is large particularly in neonates. The CPB circuit can be primed with crystalloids,
colloids, or blood products. Blood-less priming has the advantage of potentially avoiding transfusion-
related complications; however, lower Hct's may be inadequate to maintain oxygen delivery. Hemo-
dilution decreases blood viscosity and systemic vascular resistance, reduces plasma oncotic pressure
leading to tissue edema, and exaggerates the release of stress hormones and complement.
The dilutional Hct on CPB is calculated by the formula:
PBV mL pre bypass Hct 0:XX

Dilutional Hct
PBV mL circuit prime volume mL
The quantity of blood needed in the CPB prime to reach a target Hct is calculated by the formula:
PBVcircuit primevolumedesired bypassHctPBV patient Hct

Prime blood needed
Primeblood productHct 0:XX
In our institution, infant CPB circuits are primed with whole blood reconstituted from packed red
blood cells (PRBCs) and fresh-frozen plasma, as studies have advocated reconstituted whole blood over
fresh whole blood [5]. A protocol for the management of Hct on CPB is described in Table 3. The goals
for the protocol are to minimize red-cell transfusion while ensuring that an Hct of at least 24 is
maintained on bypass. This is particularly important in the pediatric population because even small-
volume cardioplegia and crystalloid-based solutions given during CPB can reduce the Hct signi-
cantly [6]. Electrolytes are balanced by the addition of electrolyte solutions to the reconstituted whole
blood. In our institution, plasma-lyte A (526-mg NaCl, 502-mg sodium gluconate, 386-mg sodium
acetate trihydrate, 37-mg KCl and 30-mg MgCl2/100 mL) with sodium bicarbonate and 300-mg calcium
gluconate per 100 mL whole blood is used. The addition of lactate and dextrose is avoided.
Ultraltration
The techniques of pre-bypass ultraltration (PBUF), zero balance ultraltration (ZBUF), conven-
tional ultraltration (CUF), and modied ultraltration (MUF) are commonly used during the conduct
of pediatric CPB. All techniques can be utilized for a bypass case with the same ultraltrator. The two
most commonly employed methods are CUF and MUF.
Excess volume in the cardiotomy venous reservoir can be removed on bypass with CUF. Excess
venous-reservoir volume can be caused by the crystalloid component of cardioplegia, crystalloid valve-
testing solutions, or when a patient has an abnormally elevated circulating blood volume. During CUF
on CPB, the perfusionist directs blood actively or passively through the ultraltrator. CUF is utilized in
nearly all pediatric bypass cases.
MUF allows the ultraltration of CPB circuit volume as well as the patient's blood volume after
weaning from CPB. The major advantage of MUF over CUF is that it allows hemoconcentration to
continue once CPB has been terminated. As a result, MUF normally allows for a greater degree of
Table 2
Estimated blood volume.
Patient weight (kg) Blood volume (mL/kg)
<10 85
10e20 80
20e30 75
30e40 70
>40 65
Table 3
Hematocrit management.
Primary repair with Complex two ventricle Other Neonates with

two ventricle anatomy anatomy/older single neonates shunted single
ventricle palliation ventricles
Predicted hematocrit after prime 24 28 35 35

Minimum hematocrit coming off CPB 25 30 35e40 40e45
Minimum HCT for transfer to ICU 25 30 37 40e50
hemoconcentration than can be obtained with CUF alone, particularly in small children. Some in-
stitutions utilize both CUF and MUF, as the techniques are not mutually exclusive.
The MUF circuit is in parallel to the bypass circuit. The total body water is decreased as the blood is
ultraltrated, and the blood is concentrated and salvaged from the bypass circuit. In the traditional
arterialevenous MUF system, inow to the ultraltrator is directly from the aorta with the aid of a
roller pump, and blood is returned to the patient by the atrial cannula (Fig. 2). In venoarterial MUF, or
simplied MUF (SMUF), a roller pump is used to pump blood from the right atrium or from the venous
line through an ultraltrator in parallel to the bypass circuit. Blood is returned to the patient via the
aortic cannula (Fig. 3). Intravascular volume of the patient is maintained during SMUF by transfusing
blood via the aortic cannula. Once the circuit-reservoir volume is reduced, crystalloid is ushed
through the reservoir, bypass tubing, and oxygenator. In the venovenous system, the intravenous
cannula (IVC) provides inow to the ultraltrator before returning to the superior vena cava (SVC). The
end point for the termination of MUF following CPB varies from institution to institution, and it may
depend on either a set time interval (15e20 min), a set target Hct (40%), or a set volume removed
(750 mL/m2). In both systems, the CPB circuit remains primed.
MUF reliably increases the Hct while salvaging bypass circuit blood after CPB; however, improve-
ments in clinical outcomes are equivocal in the literature [7]. Myocardial function may be improved by
reducing myocardial edema or by removing anesthetic agents, vasodilating agents, and myocardial
depressing factors, or by inducing sympathetic stimulation during volume removal [8]. Reducing the
total body water and edema is thought to improve the alveolarearterial oxygen gradient with a trend
toward improved postoperative hemodynamics with no change in markers for levels of inammation
[9,10]. However, all of these studies utilize a CPB technique that consists of a very large asanguineous
pump prime (400e900 mL) with PRBCs added to reach a Hct of 15e20%. This is precisely the group who
would need and benet from aggressive hemoconcentration [11].
Temperature control
In our institution, hypothermia on CPB is dened as mild (30e34 C), moderate (25e30 C), and
deep (15e22 C). The optimal temperature for CPB in pediatric patients has not been determined.
Systemic hypothermia reduces global metabolic demand, and it can allow for reduced pump ow rates,
better myocardial protection, less blood trauma, and improved organ protection. A recent systematic
review and meta-analysis of normothermic (core temperature: >34 C) and hypothermic CPB (core
temperature: <34 C) showed that normothermic CPB is as safe as hypothermic CPB in pediatric
populations with simple CHD [12], and it was in line with studies in adults where normothermic CPB
(35e37 C) had less incidence of organ dysfunction when compared to hypothermic CPB [13e15].
Children with complex CHD, those with lesions requiring DHCA, and those presenting with congestive
heart failure were specically excluded. Most pediatric centers continue to use hypothermic CPB in
pediatric congenital patients because of the absence of randomized controlled trials advocating
normothermic CPB in children and the large variety of complex congenital lesions in children [16].
Hypothermia reduces cellular metabolic rates and oxygen consumption linearly, and it is an
effective way to protect organ function [17]. Metabolic rates are determined by temperature-sensitive
enzyme-controlled chemical reactions. The Q10 temperature coefcient is the ratio of metabolic rates at
two temperatures 10 C apart, and it is higher in the pediatric population (infant 3.6 vs. adult 2.6). This
suggests that hypothermia more effectively reduces metabolic rates in children versus adults. The brain
Fig. 2. Schematic of arterialevenous MUF system.
is most at risk of ischemia because of its high metabolic demand (7-mg glucose/100-g brain tissue/min
in infants and 5-mg glucose/100-g brain tissue/min in adults) and requirement for a continuous supply
of glucose since the brain cannot store glucose. The basal metabolic rate of a neuron is approximately
30e40% of the total cellular energy expenditure. Additional effects of hypothermia include preserva-
tion of high-energy phosphate stores, reduction of excitatory neurotransmitters release by limiting the
degree of ischemia to neuronal cells, reduction of microglial activation, and attenuation of neutrophil
migration into ischemic tissue [18]. Attenuation of the ischemia-mediated release of excitatory neu-
rotransmitters such as glutamate reduces the downstream activation of lytic enzymes that lead to
further tissue injury [19]. Hypothermia reduces apoptosis and tissue necrosis by decreasing
cytochrome-c release and caspase activation.
The negative effects of hypothermia include the development of cytotoxic and vasogenic systemic
edema. Hypothermia induces dysfunction of the bloodebrain barrier, which increases cerebral hy-
drostatic pressure leading to cerebral edema. Oxygen and glucose uptake in the pediatric brain as well
as the function of the coagulation cascade is impaired [20]. Systemic vascular resistance is increased by
hypothermia, and it reduces blood ow to all organs of the body. Blood ow is reduced to the greatest
extent in the skeletal muscle and the extremities, followed by the kidneys, the splanchnic bed, the
heart, and the brain. Cerebral blood ow autoregulation is preserved in the pediatric population during
normothermic CPB, decreased during moderate hypothermic CPB, and it is absent during deep hy-
pothermia [21]. However, this study was conducted using alpha-stat pH management, where many
pediatric centers use pH-stat management to improved cerebral blood ow during hypothermic CPB.
Cannula ow
The ideal ow rate for CPB in the neonatal, infant, and pediatric patient is unknown. Flow rates have
been extrapolated from adult CPB data with compensation for the higher metabolic demand of
Fig. 3. Schematic of venousearterial or simplied MUF system.
pediatric patients (7.5e9-mL O2/kg at 37 C in neonates and infants vs. 4-mL O2/kg/min at 37 C in
adults). Oxygen consumption under general anesthesia and hypothermia is reduced. In adults, it is
known that oxygen consumption is reduced to 2e3-mL O2/kg/min under general anesthesia at 37 C,
and it is further reduced to 1e2-mL O2/kg/min at 28 C. Lower oxygen consumption under general
anesthesia and hypothermic CPB allows for a reduction in ow rates on CPB. In adults, increasing
lactate levels on CPB were noted with CPB ows of <1.6 L/m2/min [22,23], but not with ow rates of
2.0e2.4 L/m2/min, which is comparable to the expected oxygen consumption in an anesthetized adult
at 37 C. The data are less clear in the pediatric population. Furthermore, the calculated systemic blood
ow on CPB may be inadequate in pediatric patients who often have anatomic shunts, left-to-right
shunts, which reduce the amount of systemic blood ow [24]. Current recommendations for ow
rates in the absence of shunts are listed in Table 4. Serum lactate is measured during CPB to ensure
adequate systemic blood ow.
Assessment of CPB in CHD
Adequate monitoring of metabolism on CPB is imperative in pediatrics given the anatomic het-
erogeneity of patients with congenital heart disease and the differences in age-based and weight-based
ows. Mixed venous oxygen saturation is an indicator of adequate systemic blood ow, but it does not
directly measure the adequacy of cellular oxygenation [25]. Serum lactate indicates the presence of
anaerobic metabolism, and it may better reect the adequacy of cellular oxygenation. Serum-lactate
levels are routinely measured in our institution 10 min after the initiation of CPB and then every
30 min. Mixed venous oxygen saturation and serum lactate are surrogate indicators of global metabolic
supply and demand. Assessing individual organ function is challenging. Renal function is assessed by
Table 4
Recommended ow rates.
Patient temperature ( C) Flow rate (L/min/m2)
35 2.5
32 2.2
30 2.0
28 1.8
26 1.6
24 1.4
22 1.2
20 1.0
18 0.7
urine output on CPB. This can be difcult because the volume of urine produced by infants is relatively
low compared to the length of the urinary catheter and tubing. Perfusion of the heart is assessed by
visual inspection and noting changes in the electrocardiogram.
Neurological monitoring
The reported incidence of neurological complications after pediatric cardiac surgery ranges from
2% to 25% [26]. The etiology is usually hypoperfusion of the brain, and it may be related to preexisting
brain malformations, hypoxemia, and low cardiac output states, as well as sequelae of CPB. The brain
can be assessed with electroencephalographic monitoring, near infrared spectroscopy (NIRS), and
transcranial Doppler (TCD) ultrasound, but the type of monitoring that correlates with improved
patient outcome is not known [27]. DHCA method used early in the history of congenital heart
surgery was associated with poor neurological outcomes. The technique of DHCA has improved with
strategies including maintaining the Hct >30% during DHCA and improving the uniformity of brain
cooling by actively cooling for >20 min, using pH-stat acidebase management, and placing ice on the
head. Currently, DHCA is used to repair congenital heart lesions that cannot be repaired with CPB
cannula in place, and when the presence of large aortopulmonary collaterals precludes adequate
CPB.
Alternatives to DHCA include intermittent reperfusion and regional cerebral perfusion (RCP).
RCP refers to the technique of directing the arterial pump blood ow to the innominate artery, thus
selectively perfusing the right subclavian and the right common carotid arteries. The ow rates
necessary to provide adequate cerebral and somatic perfusion during regional low-ow cerebral
perfusion (RLFP) have yet to be completely elucidated although ow rates >30 mL/kg/min are
generally used. An average ow rate of 63 mL/kg/min was required to maintain cerebral blood ow
velocity (as measured by TCD) and cerebral oxygen saturation (as measured by NIRS) within 10% of
baseline during RLFP in a group of 34 infants [26]. It is not clear if this strategy is superior to DHCA.
There was no difference in neurodevelopmental outcomes in single ventricle patients undergoing
the Norwood procedure who received RCP as compared to DHCA [28]. Systematic reviews have not
shown a signicant improvement in neurological outcomes with any specic cooling strategy or
duration, the use of DHCA or regional perfusion, or the use of TCD or NIRS monitors, suggesting
that much work is needed to understand the pathophysiology of these brain injuries [27].
Acidebase management
Acidebase management on CPB in CHD is challenging because of the frequent use of hypo-
thermia. The solubility of ions in solution is temperature dependent, and so the pH is expected to
change with temperature. All acids and bases including water exist in a temperature-dependent
equilibrium with the ionized component of the parent molecule. At normal body temperature
(37 C), blood and tissue uids are alkaline relative to water (normal range pH: 7.36e7.44). Multiple
buffer systems create and maintain this relative alkalinity so that the pH and thus the ratio of [OH]
to [H] remain constant at 16:1 in humans despite temperature variations. This is the concept of
constant relative alkalinity. [29] The imidazole moiety of histidine (an amino acid common in
body tissues) acts as a major buffer. Two strategies are used clinically to address acidebase
management at different temperatures on CPB.
In alpha-stat pH management, the partial pressure of carbon dioxide on CPB is permitted to
decrease with temperature. Carbon dioxide is not added to the CPB circuit to compensate for the
decreased partial pressure of carbon dioxide at lower temperatures. Intracellular pH and enzyme ac-
tivity are preserved, and metabolic demand is reduced at low temperature. Cerebral autoregulation is
maintained, and cerebral blood ow decreases linearly with the decrease in temperature. Cerebral
metabolic demand (CMRO2) drops exponentially with decreasing temperature, and so the metabolic
demands of the brain are met. At normothermia, the mean ratio of CBF to CMRO2 is 20:1, and at deep
hypothermia, the ratio increases to 75:1. The efciency and homogeneity of cerebral cooling with
alpha-stat pH management is less than with pH-stat management, so there is less of a reduction in
CMRO2 than expected. Trivedi et al. showed that alpha-stat management was associated with a
reduction in CBF compared to pH-stat management [30]. Cold-blooded animals utilize alpha-stat
management.
In pH-stat management, carbon dioxide is added to the CPB circuit to maintain a constant partial
pressure of carbon dioxide and neutral pH at decreasing temperatures. In hypothermic CPB, 3e5%
carbon dioxide is added to the oxygenator sweep gas to maintain a temperature-corrected blood
partial pressure of carbon dioxide of 40 mm Hg and a pH of 7.40. As a result, tissue carbon dioxide
stores increase, and intracellular pH decreases. Tissue acidosis further depresses metabolism. Hiber-
nating mammals use a pH-stat acidebase management strategy when body temperature decreases.
These animals hypoventilate as they become hypothermic, thus increasing body carbon dioxide con-
tent. Tissue acidosis depresses metabolism in nonfunctioning tissues in hibernating animals (skeletal
muscle, gastrointestinal tract, and higher brain centers). Interestingly, active tissues in hibernating
animals such as the heart and liver do not adopt pH-stat acidebase management. These cells actively
extrude H across their cell membranes to maintain intracellular pH at or near the values predicted by
the alpha-stat method.
Advantages of the pH-stat approach include cerebral vasodilation and an increased cerebral blood
ow, a more homogeneous cooling, and a greater reduction in cerebral oxygen consumption compared
to alpha-stat management. An increased proportion of cerebral blood ow is distributed to deep brain
structures (thalamus, brainstem, and cerebellum) in pH-stat management though cerebral metabolic
recovery is faster with alpha-stat for unclear reasons. Tissue oxygen availability is increased secondary
to tissue acidosis and the subsequent right shift of the hemoglobineoxygen dissociation curve. Cere-
bral autoregulation is lost with pH-stat management, which may increase the likelihood of throm-
boembolic strokes. However, thromboembolic events are less frequent causes of CPB-associated brain
injuries in children than in adults. CPB-associated cerebral vascular events in children are more
frequently associated with hypoperfusion. pH-stat management is thought to decrease the incidence of
these events. These vascular events usually present as seizures and cognitive dysfunction. In our
institution, pH-stat management is used when CPB is maintained <28 C, and alpha-stat management
is used when CPB is maintained >28 C.
pH-stat management was popular in the 1960s and 1970s until studies of cold-blooded verte-
brates suggested that alpha-stat acidebase management might be more physiological. Alpha-stat
management is technically easier to manage because carbon dioxide is not added to the sweep
gas. Alpha-stat and pH-stat acidebase managements are comparable under moderate hypothermia.
The cerebral vasodilation with pH-stat management is advantageous under deep hypothermia
because cerebral cooling is more homogeneous. Patients having cyanosis with aortopulmonary col-
laterals might benet from pH-stat management because less blood is shunted to the pulmonary
circulation.
Patients who underwent surgical correction of d-transposition of the great arteries who were
managed with pH-stat management under DHCA were shown to have shorter endotracheal
intubation times, shorter length of intensive care unit (ICU) stay, and lower mortality [31].
Another study examined the developmental and neurological outcomes of 168 children at 1 and
2e4 years who underwent reparative surgeries using either alpha-stat or pH-stat during DHCA
with no detectable differences in early neurodevelopmental outcomes [32]. Further study is

necessary.
SIR and use of corticosteroids
The SIR to CPB is exaggerated in pediatric patients compared to adult patients. Contact acti-
vation by the bypass circuit is increased because of the proportionately larger circuit in children.
Higher ow rates and smaller-sized cannula increase shear stress of the blood. SIR leads to organ
injury secondary to the activation of the innate immune response and apoptosis, tissue edema due
to increased vascular permeability, and endocrine dysfunction. The lungs are particularly sus-
ceptible to inammatory injury. The lungs become edematous secondary to capillary leak, and
compliance is decreased. The kidneys have a reduced blood ow and a reduced glomerular
ltration rate, possibly due to increased serum vasopressin levels for the rst 24e48 h after CPB.
The administration of corticosteroids on CPB has been advocated to reduce the incidence of SIR
despite questionable efcacy. In our institution, neonatal and infant bypass circuits are primed
with 30-mg/kg methylprednisolone. A Cochrane systematic review in 2007 found only weak ev-
idence in favor of prophylactic corticosteroid use in pediatric patients, with a reduced duration of
mechanical ventilation and ICU stay of 1 day. The data review by Pasquali et al. did not show any
benet to prophylactic steroid administration in patients undergoing neonatal cardiac surgery, but
it did show a higher incidence of infection in those patients receiving corticosteroids [33].
Furthermore, the study by Graham et al. did not see a signicant attenuation of proinammatory
cytokine levels by the prophylactic administration of corticosteroids [34]. However, a study by
Clarizia et al. demonstrated that corticosteroid use was associated with a lower chest tube volume
output and a shorter ICU and hospital stay without an increased incidence of infection in patients
undergoing high-risk cardiac surgery [35]. Heterogeneity in study patient populations as well as
differences in dosing and timing of the corticosteroids administration makes comparison of out-
comes difcult.
Anticoagulation, reversal of anticoagulation, and antibrinolytic agents
Heparin anticoagulation
Unfractionated heparin (UFH) is the most common anticoagulant used for CPB. It is generally
accepted that an activated clotting time (ACT) in excess of 400 s is necessary to ensure adequate
anticoagulation and anti-Xa suppression for the safe conduct of CPB.
The ACT is commonly used to assess the therapeutic effects of anticoagulant medications prior to
the initiation of CPB. The ACT is also prolonged by hypothermia, hemodilution, platelet dysfunction,
and low coagulation factor levels [36]. As such, the ACT in children will overestimate the antifactor IIa
and Xa effects of UFH. It has been suggested that the Hepcon Hemostasis Management System (HMS)
(Medtronic Inc., Minneapolis, MN, USA), which is less dependent on variations in dilutional throm-
bocytopenia than the ACT, is a better assessment of anticoagulation than the ACT in children [37]. The
Hepcon heparin doseeresponse test determines a patient's ACT responsiveness to heparin, and using
an estimate of PBV determines a UFH dose necessary to reach the target UFH concentration, which will
result in an ACT 480 s.
Most institutions use an age-based or weight-based protocol to administer the initial pre-CPB
dose of UFH, and additional UFH is added to the CPB prime. The circuit prime volume to blood
volume ratio would be expected to decrease plasma UFH levels with the initiation of CPB unless an
appropriate quantity of UFH is added to the CPB prime. Recent evidence suggests that this approach
results in inferior clinical outcomes as compared to the use of the Hepcon system with a protocol
modied for use in infants [38]. Heparin should always be given into a central line through which
venous return can be demonstrated easily or as is more common in infants/neonates directly into
the heart (usually the right atrium) by the surgeon. This is necessary to ensure that the UFH dose
has reached the central circulation. An ACT can be drawn within minutes of UFH administration as
peak arterial ACT prolongation occurs within 30 s and peak venous ACT prolongation within 60 s
[39].
Protamine reversal of heparin
Protamine is given after CPB once stable hemodynamics are exhibited, and the adequacy of the
surgical repair is veried by echocardiography. Protamine should not be administered until the like-
lihood that having to reinstitute CPB is minimal. After protamine neutralization of heparin begins, the
cardiotomy suction should not be used. This practice maintains the usefulness of the bypass circuit
should reinstitution of bypass be required. The venous cannulae are usually removed before protamine
administration. The arterial cannula remains until adequate hemostasis is achieved, and circuit blood is
no longer required.
There are several approaches to the neutralization of UFH with protamine. Some centers use
1.0e1.3 mg of protamine for each 100 units of UFH determined to exist at the termination of CPB. This
ratio is based on the in vitro protamineeUFH neutralization ratio of 1.3:1.0. The amount of UFH present
is determined by obtaining an ACT when CPB terminates, and using reverse extrapolation of the pa-
tient's UFH doseeresponse curve to correlate ACT and UFH dose. Some centers simply administer a
xed dose of protamine based on the patient's weight (3e4 mg/kg) regardless of the UFH dose
administered, whereas others administer 1.0e1.3 mg of protamine for each 100 units of UFH admin-
istered. Obviously, these methods do not rely on any post-CPB assessment of residual UFH effect, such
as the ACT, to determine the protamine dose. Some centers use UFH assays and then calculate the
protamine dose based on the patient's blood volume and a protamineeUFH neutralization ratio
ranging from 1:1 to 1.3:1. This method has been shown to provide adequate UFH reversal with lower
doses of protamine even though all UFH present in blood does not exert an anticoagulant effect and
does not need to be neutralized. Other centers use Hepcon automated UFHeprotamine titration. In
theory, this method should allow the determination of the appropriate dose of protamine independent
of the non-heparin parameters that prolong ACT.
The ACT should be checked after administration of the selected protamine dose keeping in mind
that, particularly in infants, ACT prolongation may be due to factors other than the presence of residual
UFH. Comparison of a heparinase or protamine titration ACT and native ACT will allow the UFH effect to
be excluded from these other causes. Administration of additional protamine is not benign, and it tends
to delay detection and treatment of thrombocytopenia, platelet dysfunction, and coagulation factor
deciencies by the surgical team.
The incidence of protamine reactions in children following cardiac surgery is substantially lower
than that in adults. A recent retrospective analysis of 1249 children revealed the incidence of hypo-
tension (dened as at least 25% decrease in the mean arterial pressure) following protamine admin-
istration to be 1.76e2.88% depending on the stringency of criteria linking the episode to protamine
administration [40].
Antibrinolytics
The continued generation of thrombin during CPB despite heparinization is largely responsible for
the induction of ongoing brinolysis. Adjuvant therapy to improve hemostasis post CPB is directed
toward the use of antibrinolytic agents. Sequential cleavage of brin by the serine protease plasmin is
responsible for brinolysis. Plasmin is the activated form of plasminogen. Plasminogen contains ve
lysine-binding domains or kringles that allow it to bind to the lysine residues on partially lysed brin.
Fibrin-bound plasminogen is subsequently cleaved to plasmin by a tissue plasminogen activator (t-PA).
t-PA also contains lysine-binding domains that allow binding to brin. Free t-PA is capable of con-
verting brin-bound plasminogen to plasmin, but t-PA bound to brin lysine residues stimulates the
activation of plasminogen to plasmin by two orders of magnitude.
Due to the removal of aprotinin from the clinical market based on its association between morbidity
and mortality in adult cardiac surgeries, available antibrinolytic agents are currently limited to the
lysine analogs -aminocaproic acid (EACA) and tranexamic acid (TXA). Both competitively bind to
lysine-binding sites on plasminogen, thus rendering it incapable of binding to the lysine residues on
brin. The recent study by Wesley et al. addresses the pharmacokinetics of TXA [41], but the efcacy of
reducing blood loss and transfusion requirements was not addressed.
Several studies have examined the efcacy of EACA and TXA with various dosing regimens during
pediatric cardiac surgery [42e50]. The majority of these studies demonstrate that lysine analogs
reduce intraoperative blood loss and transfusion requirements during pediatric cardiac surgery.
However, it is extremely difcult to compare and interpret these studies due to heterogeneity in both
patient populations (both age and diagnosis) and dosing regimens (both dose and timing of doses). In
the future, studies to examine the efcacy of antibrinolytics using dose regimens based on known
pharmacokinetics are required.
Examples of complications during CPB and management as anesthesiologists
CPB is used routinely in cardiac surgery with few immediate adverse events. Acute dissection of the
aorta during cannulation prevents the initiation of CPB. Failure of the bypass circuit and massive air
embolism requires immediate separation from CPB and an alternate source of cardiopulmonary
function. Critical events during CPB are recognized by vigilance of the anesthesiologist, cardiac sur-
geon, and perfusionist.
Inadequate gas exchange on CPB
Inadequate gas exchange on CPB is secondary to failure of the oxygenator or gas delivery system to
the oxygenator, and it can be recognized by observing dark-colored blood exiting the oxygenator.
Proper gas supply to the oxygenator must be conrmed. Inspection of the oxygenator may reveal blood
clots or air locks in the oxygenator, requiring its replacement. Causes for inadequate anticoagulation
should be investigated, including heparin resistance, anti-thrombin III deciency, or the inadvertent
administration of protamine. Careful coordination with the perfusionist is imperative, as the patient
will need to be separated from CPB during the exchange.
Inadequate blood ow
Inadequate blood ow is usually recognized during the initiation of CPB, and it can be related to
cannula size and position and intravascular volume. This is often encountered at the initiation of CPB.
An aortic cannula that is too small or improperly positioned or an acute aortic dissection will cause
elevated arterial cannula pressures. An unrecognized aortic dissection is devastating, as arterial blood
ow is inadequate and distal perfusion might be compromised by the evolving dissection.
Massive air embolism
Massive air embolism is a rare event because of technological improvements in the bypass circuit,
automatic/fail safe emboli detection systems, perfusion safety checklists, and vigilance of the perfu-
sionists. Emboli can be particularly devastating in pediatric patients who frequently have shunts be-
tween the systemic and pulmonary circulations, potentially resulting in severe neurological injury.
Severe neurological injury can occur due to hypoperfusion distal to the emboli. Common causes of air
emboli on CPB are defects in the oxygenator, reversal of ventricular vent line, improper removal of air
from arterial tubing, and a large introduction of air in the venous cannula during cannulation [51]. In
our institution, an air embolus occurred when a non-vented cap was accidentally placed on the car-
diotomy reservoir, which pressurized it resulting in an embolus [52]. Prevention is a key by establishing
perfusion checklists. An example of a massive air embolus treatment protocol is below (Table 5).
Retrograde cerebral perfusion is used to mechanically push air from the cerebral arterial tree. Hy-
perbaric oxygen therapy has also been used. Postoperatively, it is important to rewarm slowly over
24 h.
Table 5
Massive air embolus treatment protocol.
Perfusionist Surgeon Anesthesiologist
- Discontinue bypass (clamp - Clamp and remove aortic cannula - Place patient in a steep Trendelenburg position
arterial and venous lines) - Cannulate SVC or connect to - Compress carotid arteries
- Identify origin of problem SVC cannula - Consider administering steroids/mannitol
- Reprime CPB circuit and - Retrieve blood/air existing in - Ventilate patient on 100% oxygen
arterial cannula the aorta via vent - Avoid hyperglycemia and hyponatremia.
- Retrograde SVC perfusion - When no more air is visible at - Maintain patient at 34 C.
1e2 L/min aortotomy, recannulate
- Re-institute CPB (decrease aorta and reinstitute CPB
temperature to 22e30 C, - Let bleed air from coronary arteries
increased systemic pressure) - Complete surgical procedures.
- Off bypass at 34 C.
Conclusion
CPB is particularly challenging in pediatric patients. The small blood volumes in pediatric patients
magnify the negative effects of hemodilution during bypass. Current efforts seek to reduce the size of
the bypass circuit, and to provide adjuvant therapies to attenuate these negative effects. The hetero-
geneity of the pediatric patient further complicates the management of CHD. Improved strategies for
perfusion have reduced mortality of cardiac surgery in children. Continued research is needed, as most
data comparing one technique to another have not been derived from randomized controlled trials.
Research agenda
The ideal hematocrit for hypothermic cardiopulmonary bypass (CPB) in congenital heart
disease is not known
Neurological monitors and neuroprotective strategies need to be improved
Further miniaturization of bypass circuit components will reduce the physiological effects of
CPB
The ideal acidebase management strategy is not known, and additional research is needed
to improve oxygen delivery while on CPB
Additional research is needed to determine the ideal method to monitor and reverse heparin
anticoagulation after CPB
Practice points
The anesthesiologist must be continuously vigilant during CPB to ensure patient safety.
In neonates, a hematocrit of >25% during CPB has been shown to improve neurological
outcome at 1 year
Small-volume CPB circuits require vigilance to ensure that a hematocrit of at least 25% is
maintained while on CPB
The cardiac surgeon, anesthesiologist, and perfusionist work together to ensure the safe
conduct of CPB
Conicts of interest
None.
References
[1] Baum VC. Pediatric cardiac surgery: an historical appreciation. Paediatr Anaesth 2006;16:1213e25.
[2] Stoney WS. Evolution of cardiopulmonary bypass. Circulation 2009;119:2844e53.
[3] Dennis C, Spreng Jr DS, Nelson GE, et al. Development of a pump-oxygenator to replace the heart and lungs; an apparatus
applicable to human patients, and application to one case. Ann Surg 1951;134:709e21.
[4] Gibbon Jr JH. Application of a mechanical heart and lung apparatus to cardiac surgery. Minn Med 1954;37:171e85.
passim.
*[5] Mou SS, Giroir BP, Molitor-Kirsch EA, et al. Fresh whole blood versus reconstituted blood for pump priming in heart
surgery in infants. N Engl J Med 2004;351:1635e44.
*[6] Wypij D, Jonas RA, Bellinger DC, et al. The effect of hematocrit during hypothermic cardiopulmonary bypass in infant
heart surgery: results from the combined Boston hematocrit trials. J Thorac Cardiovasc Surg 2008;135:355e60.
[7] Williams GD, Ramamoorthy C, Chu L, et al. Modied and conventional ultraltration during pediatric cardiac surgery:
clinical outcomes compared. J Thorac Cardiovasc Surg 2006;132:1291e8.
[8] Naik SK, Knight A, Elliott M. A prospective randomized study of a modied technique of ultraltration during pediatric
open-heart surgery. Circulation 1991;84:III422e31.
*[9] Chew MS, Brix-Christensen V, Ravn HB, et al. Effect of modied ultraltration on the inammatory response in paediatric
open-heart surgery: a prospective, randomized study. Perfusion 2002;17:327e33.
[10] Nagashima M, Imai Y, Seo K, et al. Effect of hemoltrated whole blood pump priming on hemodynamics and respiratory
function after the arterial switch operation in neonates. Ann Thorac Surg 2000;70:1901e6.
[11] Mate G. Perfusion for congenital heart surgery: notes on cardiopulmonary bypass for a complex patient population. 1st
ed. John Wiley & Sons, Inc; 2015.
*[12] Xiong Y, Sun Y, Ji B, et al. Systematic review and meta-analysis of benets and risks between normothermia and hy-
pothermia during cardiopulmonary bypass in pediatric cardiac surgery. Paediatr Anaesth 2015;25:135e42.
[13] Caputo M, Bays S, Rogers CA, et al. Randomized comparison between normothermic and hypothermic cardiopulmonary
bypass in pediatric open-heart surgery. Ann Thorac Surg 2005;80:982e8.
[14] Birdi I, Caputo M, Underwood M, et al. Inuence of normothermic systemic perfusion temperature on cold myocardial
protection during coronary artery bypass surgery. Cardiovasc Surg 1999;7:369e74.
[15] Birdi I, Caputo M, Underwood M, et al. The effects of cardiopulmonary bypass temperature on inammatory response
following cardiopulmonary bypass. Eur J Cardio Thorac Surg e Off J Eur Assoc Cardio Thorac Surg 1999;16:540e5.
[16] DiNardo JA. Normothermic CPB for pediatric cardiac surgery, not ready for prime time. Paediatr Anaesth 2015;25:111e2.
[17] Gravlee G. Cardiopumonary bypass: principles and practice. Lippincott Williams & Wilkins; 2007.
[18] Busto R, Globus MY, Dietrich WD, et al. Effect of mild hypothermia on ischemia-induced release of neurotransmitters and
free fatty acids in rat brain. Stroke J Cereb Circ 1989;20:904e10.
[19] Rothman SM, Olney JW. Glutamate and the pathophysiology of hypoxiceischemic brain damage. Ann Neurol 1986;19:
105e11.
[20] Pigula FA, Siewers RD, Nemoto EM. Hypothermic cardiopulmonary bypass alters oxygen/glucose uptake in the pediatric
brain. J Thorac Cardiovasc Surg 2001;121:366e73.
[21] Taylor RH, Burrows FA, Bissonnette B. Cerebral pressure-ow velocity relationship during hypothermic cardiopulmonary
bypass in neonates and infants. Anesth Analg 1992;74:636e42.
[22] Diesh G, Flynn PJ, Marable SA, et al. Comparison of low (azygous) ow and high ow principles of extracorporeal cir-
culation employing a bubble oxygenator. Surgery 1957;42:67e72. discussion, e 5.
[23] Clowes Jr GH, Neville WE, Sabga G, et al. The relationship of oxygen consumption, perfusion rate, and temperature to the
acidosis associated with cardiopulmonary circulatory bypass. Surgery 1958;44:220e39.
[24] Kirshbom PM, Skaryak LA, DiBernardo LR, et al. Effects of aortopulmonary collaterals on cerebral cooling and cerebral
metabolic recovery after circulatory arrest. Circulation 1995;92:II490e4.
[25] Harris EA, Seelye ER, Squire AW. Oxygen consumption during cardiopulmonary bypass with moderate hypothermia in
man. Br J Anaesth 1971;43:1113e20.
*[26] Andropoulos DB, Stayer SA, Diaz LK, et al. Neurological monitoring for congenital heart surgery. Anesth Analg 2004;99:
1365e75. table of contents.
*[27] Hirsch JC, Jacobs ML, Andropoulos D, et al. Protecting the infant brain during cardiac surgery: a systematic review. Ann
Thorac Surg 2012;94:1365e73. discussion 73.
[28] Goldberg CS, Bove EL, Devaney EJ, et al. A randomized clinical trial of regional cerebral perfusion versus deep hypo-
thermic circulatory arrest: outcomes for infants with functional single ventricle. J Thorac Cardiovasc Surg 2007;133:
880e7.
[29] Rahn H. Body temperature and acid-base regulation. (Review article). Pneumonol Pneumonol 1974;151:87e94.
[30] Trivedi UH, Patel RL, Turtle MR, et al. Relative changes in cerebral blood ow during cardiac operations using xenon-133
clearance versus transcranial Doppler sonography. Ann Thorac Surg 1997;63:167e74.
*[31] du Plessis AJ, Jonas RA, Wypij D, et al. Perioperative effects of alpha-stat versus pH-stat strategies for deep hypothermic
cardiopulmonary bypass in infants. J Thorac Cardiovasc Surg 1997;114:991e1000. discussion e 1.
*[32] Bellinger DC, Wypij D, duPlessis AJ, et al. Neurodevelopmental status at eight years in children with dextro-transposition
of the great arteries: the Boston circulatory arrest trial. J Thorac Cardiovasc Surg 2003;126:1385e96.
[33] Pasquali SK, Li JS, He X, et al. Perioperative methylprednisolone and outcome in neonates undergoing heart surgery.
Pediatrics 2012;129:e385e91.
[34] Graham EM, Atz AM, McHugh KE, et al. Preoperative steroid treatment does not improve markers of inammation after
cardiac surgery in neonates: results from a randomized trial. J Thorac Cardiovasc Surg 2014;147:902e8.
[35] Clarizia NA, Manlhiot C, Schwartz SM, et al. Improved outcomes associated with intraoperative steroid use in high-risk
pediatric cardiac surgery. Ann Thorac Surg 2011;91:1222e7.
[36] Huyzen RJ, van Oeveren W, Wei F, et al. In vitro effect of hemodilution on activated clotting time and high-dose thrombin
time during cardiopulmonary bypass. Ann Thorac Surg 1996;62:533e7.
[37] Guzzetta NA, Bajaj T, Fazlollah T, et al. A comparison of heparin management strategies in infants undergoing cardio-
pulmonary bypass. Anesth Analg 2008;106:419e25. table of contents.
[38] Gruenwald CE, Manlhiot C, Chan AK, et al. Randomized, controlled trial of individualized heparin and protamine man-
agement in infants undergoing cardiac surgery with cardiopulmonary bypass. J Am Coll Cardiol 2010;56:1794e802.
[39] Heres EK, Speight K, Benckart D, et al. The clinical onset of heparin is rapid. Anesth Analg 2001;92:1391e5.
[40] Seifert HA, Jobes DR, Ten Have T, et al. Adverse events after protamine administration following cardiopulmonary bypass
in infants and children. Anesth Analg 2003;97:383e9. table of contents.
*[41] Wesley MC, Pereira LM, Scharp LA, et al. Pharmacokinetics of tranexamic acid in neonates, infants, and children un-
dergoing cardiac surgery with cardiopulmonary bypass. Anesthesiology 2015;122(4):746e58.
[42] Williams GD, Bratton SL, Riley EC, et al. Efcacy of epsilon-aminocaproic acid in children undergoing cardiac surgery.
J Cardiothorac Vasc Anesth 1999;13:304e8.
[43] Rao BH, Saxena N, Chauhan S, et al. Epsilon aminocaproic acid in paediatric cardiac surgery to reduce postoperative blood
loss. Indian J Med Res 2000;111:57e61.
[44] Chauhan S, Kumar BA, Rao BH, et al. Efcacy of aprotinin, epsilon aminocaproic acid, or combination in cyanotic heart
disease. Ann Thorac Surg 2000;70:1308e12.
[45] Chauhan S, Das SN, Bisoi A, et al. Comparison of epsilon aminocaproic acid and tranexamic acid in pediatric cardiac
surgery. J Cardiothorac Vasc Anesth 2004;18:141e3.
[46] Zonis Z, Seear M, Reichert C, et al. The effect of preoperative tranexamic acid on blood loss after cardiac operations in
children. J Thorac Cardiovasc Surg 1996;111:982e7.
*[47] Reid RW, Zimmerman AA, Laussen PC, et al. The efcacy of tranexamic acid versus placebo in decreasing blood loss in
pediatric patients undergoing repeat cardiac surgery. Anesth Analg 1997;84:990e6.
[48] Levin E, Wu J, Devine DV, et al. Hemostatic parameters and platelet activation marker expression in cyanotic and
acyanotic pediatric patients undergoing cardiac surgery in the presence of tranexamic acid. Thromb Haemost 2000;83:
54e9.
[49] Chauhan S, Bisoi A, Kumar N, et al. Dose comparison of tranexamic acid in pediatric cardiac surgery. Asian Cardiovasc
Thorac Ann 2004;12:121e4.
[50] Bulutcu FS, Ozbek U, Polat B, et al. Which may be effective to reduce blood loss after cardiac operations in cyanotic
children: tranexamic acid, aprotinin or a combination? Paediatr Anaesth 2005;15:41e6.
[51] Mukherji J, Hood RR, Edelstein SB. Overcoming challenges in the management of critical events during cardiopulmonary
bypass. Semin Cardiothorac Vasc Anesth 2014;18:190e207.
[52] Matte GS, Kussman BD, Wagner JW, et al. Massive air embolism in a Fontan patient. J Extra Corpor Technol 2011;43:
79e83.

Anaesthesiology
12
Pharmacologic approaches to weaning from

cardiopulmonary bypass and extracorporeal
Wilson W. Cui, MD, PhD, Health Science Assistant Clinical
Professor *,
James G. Ramsay, MD, Professor of Clinical Anesthesia
Department of Anesthesia and Perioperative Care, University of California, San Francisco Medical Center, San
Francisco, CA 94143, USA
Keywords:
Cardiopulmonary bypass (CBP) and extracorporeal membrane
oxygenation (ECMO) are two modalities of mechanical circulatory
extracorporeal membrane oxygenation
extracorporeal life support support. They provide hemodynamic stability for patients under-
mechanical circulatory support going invasive cardiothoracic interventions, and they can be life-
left ventricular dysfunction saving in emergencies resulting from cardiogenic shock or
right ventricular dysfunction respiratory failure. Unlike implantable ventricular assist devices,
vasoplegic syndrome CPB and ECMO are short-term solutions meant to last from hours
pulmonary hypertension to days, and the patient will need to be weaned from the me-
inotropic agents chanical support once the intervention has completed or when the
vasoconstrictor agents
underlying condition has improved. Weaning imposes major
physiological strain upon the recovering cardiovascular and pul-
monary systems, and it usually requires pharmacological support.
This article focuses on the proper diagnosis of the underlying
pathophysiology, an understanding of the pharmacology of avail-
able agents, and a rational approach to the management of pa-
tients weaning from CPB and ECMO.
* Corresponding author. Department of Anesthesia and Perioperative Care, University of California, San Francisco Medical
Center, 521 Parnassus Ave, C450, San Francisco, CA 94143, USA. Tel.: 1 415 476 2131; Fax: 1 415 476 9516.
E-mail address: Wilson.cui@ucsf.edu (W.W. Cui).
258 W.W. Cui, J.G. Ramsay / Best Practice & Research Clinical Anaesthesiology 29 (2015) 257e270
Introduction
Mechanical cardiopulmonary support (MCS) is deployed when the native heart and lung functions
have to be interrupted, whether this is necessitated by an invasive procedure or due to intrinsic organ
failure. The two most common MCS techniques are cardiopulmonary bypass (CPB) and extracorporeal
membrane oxygenation (ECMO). CPB is an essential component of cardiac surgery, and successful
separation from CPB constitutes a critical step in the intraoperative course. ECMO is more appropriately
called extracorporeal life support (ECLS), as it can provide both cardiac and pulmonary support, but by
convention, it will be referred to as ECMO in this article. It is deployed in a variety of peri-procedural
and critical care settings ranging from an elective, invasive interventional cardiac procedure, to a
lifesaving intervention for someone in life-threatening cardiac or respiratory distress. While a patient
can remain on ECMO support for an extended period, up to several weeks, separation from ECMO
support is ultimately inevitable, and it remains a crucial and challenging step.
Pharmacologic agents are usually necessary during weaning from MCS, as the native organs, likely
still recovering from the recent intervention or injury, must resume their full physiologic function. The
choice of agents depends on the underlying pathophysiology likely to persist after separation. Common
causes of hemodynamic instability include systemic vasodilation, left ventricular (LV) dysfunction,
pulmonary hypertension (PH) and associated right heart failure (RHF), respiratory failure, acidosis, and
disturbance in normal hemostasis. Unsuccessful or premature weaning attempts may result in organ
injury or dysfunction. Weaning of CPB takes place in the operating room; weaning of ECMO may be
done in the intensive care unit (ICU) or in an operating room, and this requires the close cooperation of
cardiothoracic surgeons, interventional cardiologists, anesthetists, cardiologists, intensivists, perfu-
sionists, and support staff.
This article is organized in three parts. The rst part will focus on the common clinical scenarios and
indications that require the use of CPB and ECMO. The indication for support helps predict the likely
pathophysiology to emerge during the weaning process. This is followed by a brief discussion of the
preparation before weaning from CPB or ECMO. Finally, common problems encountered during
weaning and appropriate pharmacological treatments are discussed (Table 1). Attention is paid to each
agent's mechanism, relevant clinical indications, and available evidence. There is a signicant overlap
in the clinical problems encountered in CPB and ECMO, and the readers should be mindful that
treatments should be dictated by the problems rather than by the particular modality of MCS.
Clinical scenarios
Cardiopulmonary bypass
Today, CPB remains the most commonly used method of providing extracorporeal circulatory
support in the operating room, and it makes cardiac surgery possible. CPB decompresses the heart,
minimizes bleeding in the surgical eld, provides circulatory stability in the absence of cardiac func-
tion, and helps in the oxygenation and carbon dioxide removal in the absence of lung function. Full CPB
includes a reservoir, an oxygenator, and a circulatory pump, and separate pumps to permit the salvage
of blood from the operative eld, and it decompresses the heart (vent) as well as provides car-
dioplegia when needed. CPB inevitably elicits a profound inammatory response in the body [1], and
either prolonged or insufcient administration of cardioplegia may lead to ischemic injury of the
myocardium. The ease of separation from CPB is largely determined by the pre-bypass cardiac function
and medical comorbidities, the length of bypass and ischemic times, the quality of myocardium pro-
tection during aortic cross-clamp, and the success of the surgical intervention.
In some instances, CPB is deployed urgently or semi-emergently. One example is the use of CPB
during lung transplantation. Modern bilateral lung transplantation involves the sequential implanta-
tion of the lungs. While acceptable gas exchange is often feasible with the ventilation of the contra-
lateral lung, it is not always possible due to hypoxia, respiratory acidosis, or exacerbation of PH and
right heart strain. The surgeon may opt to place the patient on CPB either preemptively prior to
explanting the native lung or do so emergently at any point during the operation when support is
needed. Another example would be acute hemorrhage and instability during intrathoracic procedure
W.W. Cui, J.G. Ramsay / Best Practice & Research Clinical Anaesthesiology 29 (2015) 257e270 259
Table 1
Potential problems during weaning from CPB or ECMO.
Problem Characteristics Pharmacological treatment
Systemic vasodilation CI: normal to high Vasoconstrictors (adrenergic)

LVEF: normal to high - Phenylephrine (10e100 mg/min)
CVP: low - Norepinephrine (0.02e0.2 mg/kg/min)
PAOP: low Vasoconstrictors (other)
- Vasopressin (0.01e0.04 mg/kg/min)
- Methylene blue (2 mg/kg over 30e60 min)
LV dysfunction CI: low Positive inotropes (adrenergic)
LVEF: low - Epinephrine (0.02e0.2 mg/kg/min)
CVP: normal to high - Dobutamine (2e20 mg/kg/min)
PAOP: high - Dopamine (2e20 mg/kg/min)
Positive inotropes (PDE-3 inhibitors)
- Milrinone (0.1e0.75 mg/kg/min)
Calcium sensitizer
- Levosimendan (0.05e0.2 mg/kg/min)
LVOT obstruction CI: low Volume resuscitation
LVEF: high Beta adrenergic antagonists
CVP: low - Esmolol (50e300 mg/kg/min)
PAOP: low Vasoconstrictor
- Phenylephrine
Anesthesia
Analgesia
RV dysfunction CI: low Preload reduction
LVEF: high Pulmonary vasodilators
CVP: high - iNO (10e20 ppm)
PAOP: variable - Epoprostenol (inhaled nebulizer 2e50 ng/kg/min)
- Milrinone
Positive inotropes
- Dobutamine
- Milrinone
- Levosimendan
where the heart or one of the great vessels is injured. The use of CPB will divert the blood away from the
heart, and it will allow the surgeon to inspect the source of bleed while the patient can be stabilized
and resuscitated on the CPB circuit.
Extracorporeal membrane oxygenation
Modern ECMO is a versatile technology that is relatively simple to deploy, can restore tissue
perfusion and oxygenation in an unstable patient, and is much more portable than the traditional CPB
machine. Unlike CPB with a reservoir and multiple pump options, ECMO is a closed circuit without a
reservoir or secondary pumps. ECMO has found its usefulness in a growing number of clinical settings
and patient populations. Most patients are placed on ECMO emergently due either to life-threatening
respiratory failure or to cardiogenic shock. In such a setting, it is potentially lifesaving or life-extending.
The goal is to temporarily relieve the physiological strain on the overwhelmed cardiopulmonary sys-
tems, buy time to initiate other therapies, and allow the patient to recover. Depending on the clinical
situation, different ECMO modalities may be appropriate. In venoarterial (VA) ECMO, blood is diverted
from the right heart and, after oxygenation and CO2 removal, it is returned to the patient's systemic
circulation via a large artery, similar to CPB. This has the advantage of providing both cardiac and
pulmonary support, and it is equally effective in those with cardiogenic shock or respiratory failure. In
general, VA ECMO is the preferred choice in an acutely decompensating patient. Alternatively, veno-
venous (VV) ECMO removes deoxygenated blood, but it returns the oxygenated blood to the right heart.
Because of this, it does not provide cardiac support, and it should only be used in those who do not have
signicant cardiac, especially right ventricle (RV), dysfunction. A patient with solely respiratory failure
may be electively transitioned from VA to VV ECMO. On the other hand, someone on VV ECMO showing
evidences of RHF may require emergent transition to VA ECMO.
Some patients may not, however, recover sufciently to be successfully weaned off ECMO. One
example is in the setting of acutely decompensated end-stage lung disease where ECMO may stabilize
the patient while waiting for a suitable donor lung, or to complete the evaluation for lung transplant. It
is not uncommon at our institution that a potential recipient is placed on ECMO and optimized in the
ICU, then undergoes lung transplantation either on intraoperative ECMO or on CPB, returns to the ICU
on ECMO due to concerns of RV or graft function, and recovers before weaning. Other examples include
those with postcardiotomy heart failure or acute decompensated patients awaiting left ventricular
assist device (LVAD) or heart transplantation.
Lastly, interventional cardiac procedures are increasingly being performed in high-risk patients
with limited cardiopulmonary reserve. The use of ECMO in this population can provide hemodynamic
stability during critical portions of the procedure. These procedures could vary in urgency, from an
elective transcatheter aortic valve replacement to emergency percutaneous coronary angioplasty
during an acute myocardial infarction. Upon completion of the intervention, the underlying pathology
leading to the need for ECMO usually is unchanged, and signicant pharmacologic support may be
needed to enable weaning the mechanical support.
Preparation
Cardiopulmonary bypass
Preparation should be made to ensure that the initial separation from CPB is successful as an un-
successful attempt can lead to ischemic injury to vital organs. Planning for separation requires good
communication between the surgeon, the anesthetist, and the perfusionist. Most measures before
weaning from bypass are routine and often standardized, such as correction of acid/base status,
normalization of electrolytes, ensuring the return of organized cardiac rhythm, and normothermia
(Table 2) [2]. The airway is suctioned, atelectatic lungs are re-expanded, and ventilation is resumed. The
reversal agent for anticoagulation, protamine, and blood products and factors to correct any potential
coagulopathy should be readily available. It is not infrequent for the heart to brillate after the aorta is
unclamped and the heart is reperfused. Anti-arrhythmic drugs such as lidocaine and magnesium are
routinely given, and internal debrillation may be delivered. Temporary epicardial wires are often
placed, and external pacing may be necessary to ensure adequate heart rate (HR). Hyperbrinolysis is
common on bypass, and aminocaproic acid and tranexamic acid are two lysine-analog antibrinolytics
that are currently available. Their use has been shown to reduce blood transfusion requirement in
cardiac surgery [3]. Both agents are dependent on renal excretion, and the dosage should be adjusted
accordingly. Based on the 2011 update from the Society of Thoracic Surgeons (STS) and the Saskatch-
ewan Cancer Agency (SCA) clinical practice guidelines on blood conservation, the treatment of
nonsurgical bleeding once heparin has been reversed with protamine should be directed by docu-
mented deciency or dysfunction of platelets or coagulation factors [4]. Fibrinogen deciency
(<200 mg/dL) can be replaced with a cryoprecipitate or a brinogen concentrate. Fresh-frozen plasma
can be transfused to replace the deciency of other procoagulant factors, and recombinant factor VII
Table 2
Routine checks and preparation before weaning from CPB.
Parameter Target goal Treatment

Temperature >36 C Warming
Heart rate 70e100 bpm Epicardial pacing, atropine, inotrope
Heart rhythm Sinus (preferred) Debrillation, cardioversion, lidocaine, magnesium, amiodarone
Ventilation Airway suction, recruitment, resume ventilation
Electrolytes K: 3.5e5 mEq Ultraltration, diuresis
iCa > 1.0 mEq Ca repletion
Acidebase pH > 7.3 Sodium bicarbonate
and prothrombin complex concentrates (PCCs) can be used for refractory coagulopathy. Prior to the
initiation of weaning, the team should discuss any concern and likely problems during and after
weaning from CPB. Weaning from CPB commences with the gradual reduction in venous return into
the pump, and with the decrease in pump ow. The management strategies during weaning at most
institutions are often not standardized, and they are largely based on the clinical judgment of the
surgeons and the anesthetists.
Extracorporeal membrane oxygenation
For elective, intraoperative ECMO, deployed for high-risk procedures, preparation before weaning is
similar to that of CPB. However, when ECMO support is prolonged (days to weeks), the weaning process
most likely begins in the ICU. Similar to CPB, prolonged ECMO induces a signicant inammatory
response in the body, and it can cause signicant morbidity. Exposure to the ECMO circuit can lead to
thromboembolic complications with inadequate anticoagulation, while anticoagulation, consumption
of factors, platelet activation, and depletion can lead to excessive bleeding. The data on the use of rFVIIa
and PCC in patients on ECMO support are limited to case reports [5]. Given that they carry a high
thrombotic risk and the thrombosis of the ECMO circuit is catastrophic, many practitioners advise
against the use of rFVIIa and PCC while the patient is on ECMO. Prolonged ECMO may cause other
systemic complications. Hypovolemia from bleeding and volume overload, and venous congestion due
to resuscitation can both occur. Metabolic acidosis is common, and it can be due to malperfusion,
excessive vasoconstrictor therapy, or renal failure. Renal replacement therapy (RRT) may be necessary.
Patients on ECMO often suffer from acute lung injury, and ECMO offers a resting period for recovery by
lowering oxygen exposure and airway pressure to avoid further injury. However, agitation and
ventilator dyssynchrony are common, requiring sedation, tracheostomy, and even neuromuscular
paralysis, which may mask new neurological injury. Limb ischemia is another complication of pe-
ripheral cannulation, and it can cause compartment syndrome. Multiorgan failure carries very poor
prognosis. Correction of these issues is paramount, and it remains a formidable challenge for the
surgical and critical care team.
Unlike that of CPB, the end point of ECMO is often not predetermined, and weaning can be a
prolonged process. For those on extended ECMO support in the ICU, the recovery of the patient's
cardiac and pulmonary function can be monitored by a temporary decrease of ECMO ow rate based on
hemodynamics and echocardiographic assessment. This may be done on a daily basis to assess the
readiness to stop ECMO support. The entire weaning process may last days to even weeks. Nonetheless,
the cardiovascular and pulmonary systems remain fragile and can be easily overwhelmed, which
means prolonged pharmacologic support after separation is often necessary.
Clinical problems during weaning
Systemic vasodilation
The most common cause of hypotension during weaning from CPB is excessive systemic vasodi-
lation. Ohm's law states that perfusion pressure is dependent on cardiac output (CO) and systemic
vascular resistance (SVR). A number of factors may contribute to excessive vasodilation after CPB or
ECMO. These include the duration of nonpulsatile ow, the degree of hypothermia during CPB, the
ongoing systemic inammatory states induced by the CPB itself and possibly sepsis, and the use of
certain preoperative medications [1]. SVR is readily determined by using Ohm's law. Hypotension
dened as the low mean arterial pressure (<60 mm Hg) in the setting of normal to high cardiac index
(CI 2.4 L/m2/min) would suggest a low SVR (<1000 dyn/s/cm5) that requires treatment with vaso-
constrictors. Prior to weaning from CPB, communication with the perfusionist can be useful to
ascertain the overall vasoconstrictor requirement of the patient. On the other hand, if the patient has
been on an extended ECMO support, a review of the ICU course and discussion with the critical care
team regarding the ongoing vasoconstrictor use will provide a good starting point on the choice of
agent and dosage.
Catecholamines with alpha 1-adrenergic agonist activity are the rst-line vasoconstrictors. In North
America, the principal agents are phenylephrine and norepinephrine; in higher doses, the alpha-1
effect of epinephrine is also apparent. Alpha-1 receptors are members of the G-protein-coupled re-
ceptors that when activated start a cascade of intracellular signaling via the phospholipase-C (PLC)
pathway, which results in the inux of cytoplasmic calcium ions, smooth muscle contraction, and
increased vascular tone [6]. Norepinephrine is an endogenous catecholamine, released by the adrenal
medulla as a stress hormone and by the adrenergic neurons in the central nervous system. In addition
to peripheral vasoconstriction via alpha-1 activation, it also activates beta-1 adrenergic receptors, but it
has a minimal beta-2 effect. Phenylephrine is a synthetic alpha-1 agonist that induces vasoconstriction,
but it is less potent than norepinephrine, and tachyphylaxis is not uncommon. The main adverse effect
of alpha-1 agonists is excessive vasoconstriction resulting in hypoperfusion and ischemia of organs
such as the kidney, intestines, and skin.
Refractory hypotension post CPB not responsive to traditional (adrenergic) vasoconstrictor treat-
ment is sometimes termed vasoplegic syndrome (VS) [7]. It is associated with high CI, low SVR, low
central venous pressure (CVP), and increased uid requirement. The reported incidence ranges from
8% to 20% after on-pump cardiac surgery [7e9]. Preoperative predictors include low LV ejection
fraction (LVEF), recent coronary ischemia, intravenous (IV) heparin, angiotensin-converting-enzyme
inhibitors (ACEIs), and calcium-channel blockers. Vasopressin, a nonadrenergic vasoconstrictor, has
been used as an alternative vasoconstrictor in VS. Otherwise known as antidiuretic hormone, it is a
neurohormone that plays a signicant homeostatic role in the control of body water and vascular
tone. Similar to catecholamines, vasopressin receptor subtype-1 activation in the vascular smooth
muscle cells is mediated via G-protein signal transduction, and it results in an increased intracellular
calcium and vascular tone. Studies have shown that while endogenous vasopressin level usually
increases 100-fold with the commencement of CPB and remains elevated for hours in the post-bypass
period [10], this phenomenon is absent in a subset of patients [7]. In addition, endogenous vaso-
pressin stores may be depleted in prolonged stress and vasodilated states [11]. In three randomized
trials in patients on ACEI therapy, prophylactic treatment with a low-dose vasopressin (0.03 unit/min)
prior to CPB weaning resulted in fewer hypotensive episodes and lower norepinephrine requirement
[12e14]. Similarly, the use of vasopressin signicantly improved blood pressure in patients after LVAD
implantation [15]. Combined vasopressin and norepinephrine treatment were shown to have a higher
blood pressure, lower norepinephrine dosage, lower HR, lesser arrhythmia, and a better cardiac
performance index than norepinephrine infusion alone [16]. Based on these studies, the clinical dose
range of vasopressin for its vasoconstrictor effect is 0.03e0.04 unit/min, rarely up to 0.1 unit/min in
severe refractory VS. An advantage of vasopressin is its minimal effect (in low dosage) on the pul-
monary vascular resistance (PVR) and on glomerular ltration rate compared to that of phenylephrine
and norepinephrine. With prolonged use, especially at higher doses than 0.04 unit/min, vasopressin
can cause excessive vasoconstriction and organ hypoperfusion such as mesenteric ischemia and skin
or digit necrosis. Excessive hypertension and increased afterload can also increase myocardium de-
mand and reduce CI.
The common intracellular signaling pathway that leads to vasodilation is the activation of guanylate
cyclase (GC) that generates cyclic guanosine monophosphate (cGMP) in smooth muscle cells. Nitric
oxide (NO) activates GC by binding to the ironeheme moiety in the enzyme complex. An increased NO
level has been implicated in refractory vasodilatory shock [17]. Methylene blue (MB) is a competitive
inhibitor of NO-dependent GC activation, and thus it can prevent the vasodilatory response of the
vascular smooth muscle to proinammatory mediators. In one study, 92.4% of patients suffering from
VS after CPB responded to MB with an increased arterial pressure and a decreased norepinephrine
requirement [18]. In one multicenter, randomized trial, patients with vasoplegia receiving MB showed
faster recovery and lower mortality and morbidity than those receiving placebo [19]. The prophylactic
use of MB infusion in high-risk patients either preoperatively or during CPB resulted in a lesser inci-
dence of vasoplegia, lower vasoconstrictor requirement, and a shorter ICU stay [20,21]. However, a
recent propensity-score-matched study showed that the use of MB as rescue therapy post-bypass is
associated with a signicant increase in morbidity and a trend in higher morbidity [22]. The thera-
peutic effect may be transient, and refractory vasoplegia may require either repeated dosing or
continuous infusion. Urine and skin discoloration is a unique side effect of the MB administration, but it
is transient and not clinically relevant. MB has signicant light absorption at 660 nm, and it will
interfere with the proper functioning of pulse oximeters.
LV dysfunction
LV dysfunction is another common nding during weaning from CPB. For open cardiac surgery, the
cross-clamping of the ascending aorta interrupts coronary perfusion, and the myocardium is protected
from ischemia with a combination of cardioplegia-induced electricalemechanical quiescence, ven-
tricular decompression, and myocardium cooling. Inadequate myocardial protection, prolonged cross-
clamp duration, and reperfusion can result in myocardial injury and reduced ventricular function after
bypass. In some patients, there is a preexisting LV dysfunction, which may be the result of myocardial
infarction, hibernating myocardium due to chronic ischemia, or progressive valvular disease. The
benet of surgical intervention, whether it is coronary revascularization or valve repair, may not be
immediately apparent after CPB. The patient's preoperative systolic and diastolic functions, advanced
age, chronic kidney disease, reoperation, the need for inotropic agents prior to CPB, the length and
complexity of procedure, the length of bypass and cross-clamp times, and the effectiveness of the
surgical intervention can predict the likelihood of LV failure during weaning [2]. Therefore, post-bypass
low cardiac output syndrome (LCOS) is common, usually manifests as elevated left atrial or pulmonary
arterial occlusive pressure (PAOP), reduced CI (dened by various groups as <2e2.4 L/m2/min), and
normal or elevated SVR. On transesophageal echocardiography (TEE), the LV may have regional or
global hypokinesis, and it appears distended with a reduced LVEF. Similarly, LV dysfunction is common
for patients on ECMO as cardiogenic shock and low LVEF are common indications for ECMO.
Positive inotropic agents are the treatment of choice in the setting of LV dysfunction. The rst-line
agents include epinephrine and other catecholamines with a signicant beta-1 adrenergic agonist
activity. The activation of beta adrenergic receptors increases intracellular cyclic adenosine-
monophosphate (cAMP) concentration via the adenyl cyclase (AC) pathway, and it leads to an
increased cytoplasmic calcium concentration [6]. In cardiomyocytes, this results in the activation of
voltage-gated calcium channels, and it increases the rate of depolarization and automaticity in pace-
maker cells, and thus HR (positive chronotropy). Calcium ions also bind to troponin, and these enhance
muscle contraction (positive inotropy). Epinephrine, similar to norepinephrine, is a hormone released
by the adrenal medulla, and it is a neurotransmitter in the sympathetic nervous system. Epinephrine is
a potent alpha-1 and beta-1/beta-2 agonist, which is frequently used in patients with post-bypass
LCOS. In one randomized study, epinephrine at 0.1 mg/kg/min caused a signicant increase in MAP
without changes in SVR [23], suggesting that the effect was due to an increased beta-related inotropy
rather than to an alpha-related vasoconstriction. Epinephrine can, however, signicantly increase the
myocardial work and oxygen demand. As the prototypical ght-or-ight stress hormone, it has sig-
nicant effect on most organs in addition to the heart due to its relative nonselectivity for adrenergic
receptors. It can cause bronchial dilation via beta-2 receptor activation. Its effect on vascular tone
depends on the relative ratio of alpha-1 to beta-2 receptors in the particular vascular bed and drug
concentration. Its major metabolic effect is the mobilization of glucose storage via glycogenolysis and
lipolysis.
Other less potent positive inotropic catecholamines include dobutamine and dopamine. Dobut-
amine is a synthetic analog of isopreterenol with predominantly beta-1 agonist effect. The combination
of increased CO and HR (beta-1) and secondary vasodilation has led to the term ino-dilator. Recall
Ohm's law, where an increased CO in the setting of severe vasodilation may result in a drop in perfusion
pressure. In practice, the use of ino-dilators often requires the concomitant use of vasoconstrictors,
such as norepinephrine. Both observational and randomized studies have consistently shown that
dobutamine increases CI, stroke volume index, and HR, accompanied by reduction in SVR and PVR
[24e28]. HR increase with dobutamine is dose dependent. At doses that produce similar increase in CI,
dobutamine causes signicantly more tachycardia than epinephrine, suggesting that HR increase may
be the primary mechanism of CO augmentation [24,26]. Unlike alpha agonists, dobutamine infusion
increases splanchnic ow and maintains intestinal oxygenation and lactate extraction [29].
Dopamine has an agonist activity not only on the adrenergic receptors (beta-1 and alpha-1) but also
on its own endogenous receptors. The cardiovascular action of exogenous dopamine is determined by
the plasma concentration of the drug and its relative afnity to the various receptors. At a low dose
(<3 mg/kg/min), dopamine causes renal and mesenteric arteriole vasodilation. However, a recent meta-
analysis review of the available data failed to nd any positive renal protective effect of dopamine in
CPB [30]. An intermediate dose of dopamine (2.5e5 mg/kg/min) predominantly activates the beta-1
receptors resulting in positive chronotropic and inotropic effects. In one study, dopamine at 7 mg/kg/
min dose produced the same degree of CI augmentation as that of epinephrine at 0.1 mg/kg/min [31].
However, at doses >5 mg/kg/min, alpha-1 receptor activation and vasoconstriction becomes signicant
and increases in HR as well [32,33]. One study found that while dopamine (6.2 mg/kg/min) and
dobutamine (6.7 mg/kg/min) produced similar CI and HR increase, there was a signicant off-loading of
the heart with dobutamine with lower LV and RV lling pressures and SVR [34]. A randomized study in
high-risk patients with mitral stenosis showed that dobutamine is more efcacious than dopamine in
weaning [25]. Dopexamine is a synthetic analog of dopamine, with beta-1, beta-2, and dopamine re-
ceptor agonist effects but with little alpha activity. It is not available in North America.
Milrinone is a positive inotropic agent not structurally related to catecholamines. It is an inhibitor of
type-3 phosphodiesterase (PDE-3), and it prevents cAMP and cGMP degradation by PDE-3 [35].
Because cAMP is the common signaling molecule in adrenergic receptor activation, PDE-3 inhibitors
work synergistically with the catecholamine agents. In patients with LCOS and suboptimal response to
conventional catecholamines, milrinone is a good choice. In three randomized trials with high-risk
patients with preexisting low LVEF, milrinone increased post-bypass CI, decreased the need for
other inotropes, and it was more successful in weaning from CPB than from placebo [36e38]. In a large
randomized trial, involving 120 patients, milrinone (0.5 mg/kg bolus followed by 0.5 mg/kg/min
infusion) and dobutamine (10e20 mg/kg/min infusion) were similarly effective in treating post-bypass
LCOS [27]. However, dobutamine group had a higher incidence of atrial brillation, which may be dose-
related, while milrinone group had a lower PAOP [27,28]. One observational study using multivariable
logistic analysis appears to suggest an increased risk of postsurgical AF with milrinone [39], which is in
disagreement with the previously mentioned randomized study comparing milrinone and dobutamine
[27]. PDE-3 inhibitors are also ino-dilators because they induce protein kinase-A activation of sarco-
plasmic calcium pumps that remove cytoplasmic calcium in the vascular smooth muscle. Milrinone has
a much longer half-life, 30e60 min than catecholamines, and its hypotensive effect can be profound
and lingers even after discontinuation. The dosage of milrinone is 50 mg/kg, typically administered as a
load while on CPB, followed by infusion 0.1e0.75 mg/kg/h. Some anesthetists choose to reduce or forego
the loading dose to avoid excessive vasodilation. Enoximone is another PDE-3 inhibitor, but it is not
currently available in North America. The rst clinically available PDE-3 inhibitor was amrinone, but it
has fallen out of favor due to the side effect of thrombocytopenia.
Levosimendan is a novel agent, termed calcium-sensitizer, that enhances the sensitivity of car-
diomyocytes to intracellular calcium without increasing calcium concentration. Unlike catecholamines
and PDE-3 inhibitors, which increase myocardial oxygen consumption and the risk of tachyarrhythmia,
levosimendan was originally thought to exert its positive inotropic effect by binding and stabilizing the
calciumetroponin complex to enhance muscle excitationecontraction [40]. However, recent evidence
suggests that levosimendan may also inhibit PDE-3 and enhance contractility via the cAMP-dependent
pathway [40]. Additionally, it activates ATP-dependent potassium channels, causes relaxation of
vascular smooth muscle cells, and exerts anti-ischemic effect in cardiomyocytes [40]. A number of
prospective, randomized trials investigating its utility in CPB have been published in the past decade,
most of which were reviewed in a recent meta-analysis [41]. In the largest trial involving patients with
normal LVEF, intraoperative administration of levosimendan was found to decrease the need for rescue
inotrope, the use of intra-aortic balloon pump, and the incidence of postoperative heart failure
compared to placebo [42]. Two randomized studies found that, in those with reduced LVEF, preoper-
ative levosimendan increased the chance of successful primary weaning from CPB [43,44]. A head-to-
head comparison with dobutamine in patients with post-bypass LCOS showed that levosimendan was
as effective in improving hemodynamics, but results were limited by small sample sizes and difference
in dosing [45,46]. Levosimendan's clinical effects are most similar to milrinone; however, two small
studies comparing the two agents yielded inconclusive results [47,48]. Meta-analysis by Harrison et al.
concluded that perioperative use of levosimendan in patients with reduced LVEF has benet in
reducing postoperative mortality, RRT, atrial brillation, and myocardial injury [41]. The dosing
regimens for levosimendan varied among the studies with bolus doses ranging from 10 to 24 mg/kg
followed by infusion at rates from 0.05 to 0.2 mg/kg/min for typically 24 h, though some omitted the
bolus dose altogether. Unlike catecholamines and milrinone, one metabolite of levosimendan (rep-
resenting 6% of the dose) is pharmacologically active with a long elimination half-life providing a
prolonged therapeutic effect of more than a week. Levosimendan is not currently available in North
America outside of an ongoing multicenter trial.
LV outow tract obstruction
Another cause of hemodynamic instability during weaning from CPB is LV outow tract obstruction
(LVOTO). LVOTO can be present in patients with hypertrophic cardiomyopathy, also known as idio-
pathic hypertrophic subaortic stenosis, and it may be sufciently severe to warrant surgical inter-
vention. However, LVOTO can rst appear intraoperatively most commonly after mitral valve surgery.
The combination of systolic anterior motion (SAM) of the mitral valve leaet, LV hypertrophy, and
contact between the mitral leaet and the septum during systole can lead to dynamic ow obstruction.
The incidence of SAM after mitral valve surgeries is at least 4e6.6% [49,50]. Prerepair echocardio-
graphic ndings of a redundant anterior leaet, and anterior malpositioning of the coaptation point
(anterior/posterior leaet length ratio < 1.3, and coaptation point to a septum distance < 2.5 cm) can be
used to identify those at an increased risk of postrepair SAM [51].
Dynamic LVOTO post-bypass is the result of an increased LV contractility, a reduced diastolic vol-
ume, and an abnormal mitral leaet mechanics. Medical management targeting these underlying
mechanisms is usually successful. The hyperdynamic LV can be the result of improved cardiac function
after repair and/or inotrope use. Tachycardia e shortened diastolic time e exacerbates LV under-lling.
The reduced LV internal diameter potentiates the likelihood of SAM and leaeteseptum contact during
systole. Management includes volume resuscitation to increase LV size, the use of negative inotropic
and chronotropic agents such as beta-blockers, and an increase in anesthetic depth. Hypotension in the
setting of SAM is due to obstruction and not LV failure, so the use of a positive inotrope is counter-
productive. Rather, phenylephrine is useful because it maintains systemic pressure to avoid myocar-
dium ischemia, and it may induce reex bradycardia. TEE is required to make the diagnosis of SAM and
to assess the treatment response. Rarely, SAM is so severe that a return to CPB for surgical intervention
is needed [49].
RV dysfunction
During weaning from CPB, some patients may suffer from RV dysfunction. There are a number of
culprits. Due to its crescent shape and anterior anatomy, hypothermia of the RV is challenging to
maintain with usual cooling techniques. Gas emboli tend to enter the more anteriorly positioned right
coronary artery, and they cause acute RV ischemia. Post-bypass atelectasis and ventilationeperfusion
mismatch can lead to an acute increase in PVR. Lastly, protamine administered to reverse anti-
coagulation can cause pulmonary vasoconstriction, rarely precipitating RHF requiring emergent return
to CPB. Patients with preexisting PH and right heart dysfunction are more likely to experience
persistent RV dysfunction post bypass. The incidence of RHF after LVAD implantation e dened as the
need for an extended inotrope support (>14 days) or a rescue RVAD e is between 13% and 32% [52e54].
RHF is common in patients on ECMO support as it is often initiated for right heart decompensation
in the setting of severe lung disease. The RV, compared to LV, is less muscular, more compliant, with a
lower EF, and it is normally exposed to the pulmonary vascular bed with one-fth of the systemic
resistance. With chronic PH, the RV becomes hypertrophic to compensate for a higher afterload. Unlike
the LV, coronary perfusion of RV normally occurs throughout the cardiac cycle due to the favorable
pressure gradient in the coronary arteries and the endocardium, but this normal perfusion pattern is
interrupted when the RV systolic pressure approaches systemic pressure, and RV ischemia may occur
with relatively mild systemic hypotension. Echocardiographic evidence of RV dysfunction includes RV
dilation, decreased free-wall movement, decreased tricuspid annular plane systolic excursion (<2 cm),
worsening tricuspid regurgitation, and septal dyskinesia. The septum attens, even bows leftward (D-
sign), and it is pathognomonic for RHF. The principle of ventricular interdependence dictates that LV
lling is maintained by an adequate RV output, and RHF leads to decreased CO and systemic hypo-
tension, which causes further RV ischemia and sustains the undesirable cycle. Hemodynamically, RHF
is usually characterized by the combination of tachycardia, low CO, systemic hypotension, and elevated
CVP. While an acute increase in pulmonary artery (PA) pressure may exacerbate RV dysfunction,
pseudonormalization (i.e., decrease) of PA pressure is a late sign of RHF. Hemodynamic measure-
ments, echocardiographic ndings, and laboratory evidence of organ dysfunction have been investi-
gated as risk factors and prognostic tools; however, these are based on single-center experiences, and
they have not been validated by others [55]. The management of RV dysfunction is especially chal-
lenging, and it usually falls into four categories: optimize preload, minimize afterload, enhance ino-
tropy, and maintain perfusion by avoiding systemic hypotension [56].
Accurate assessment of RV volume and function status is based on direct visualization (intra-
operatively), echocardiography, and the CVP trend. The motion and behavior of the RV free wall during
lling and contracting phases can be directly observed. TEE can visualize any dyskinetic septum
movement and signs of volume overload. Transient manipulation of venous return with patient
positioning can be used to test RV response to varying preload conditions. In the setting of volume
overload, venodilation with nitrates such as nitroglycerin may be used albeit cautiously due to the risk
of systemic hypotension. Persistent volume overload may warrant the use of loop diuretic or RRT.
Avoiding additional increase in PVR is another important aspect of RV protection. This includes
pulmonary toilet, minimizing airway pressures, and avoidance of hypoxia and respiratory or metabolic
acidosis. This is especially important for patients on ECMO in the ICU. The weaning process usually
starts there where the ventilation strategy is optimized. ECMO ow rate is then slowly lowered to test
the tolerance of the patient. The ability to maintain adequate oxygen saturation and blood pressure on
a minimal ECMO support (1 L/min) suggests that the patient may be ready for separation from ECMO.
In patients with PH, pharmacological vasodilation of pulmonary vasculature may be necessary; this
is most effective in those with primary PH. The underlying cause of PH e endothelial dysfunction e
results from a pathological imbalance between vasodilatory molecules such as NO and prostacyclins
(PGI2) and vasoconstrictors such as endothelin receptors and thromboxanes. These molecules have
been investigated as therapeutic targets for the treatment of PH and RHF (Table 3).
NO is a potent endogenous vasodilator due to its activation of GC and the cGMP pathway. NO is the
active metabolite of medical nitrates such as nitroglycerin and sodium nitroprusside, and it is the
basis of their vasodilatory effect. However, unlike IV nitrates, which can cause signicant systemic
vasodilation, inhaled NO (iNO) is delivered directly to the pulmonary alveoli and vasculature, and it
exerts its vasodilatory action locally without systemic effect because it is rapidly inactivated by
binding to hemoglobin. In acute lung injury, iNO has the added benet of improving ven-
tilationeperfusion matching as it improves blood ow to the ventilated alveoli. In lambs, iNO has
Table 3
Pulmonary hypertension therapies.
Class Mechanism Agents Side effects
Nitric oxide cGMP-dependent vasodilation iNO Rebound PH

Reversal of right-to-left shunt
Pulmonary edema
Prostacyclin cAMP-dependent vasodilation Epoprostenol (IV, lnh) Flushing
Treprostinil (IV, SC, lnh) Headache
Iloprost (lnh) Hypotension
Endothelin Blocks endothelin-induced Bosentan Hepatotoxicity
antagonists PLC-dependent vasoconstriction Macitentan Peripheral edema
Ambrisentan
PDE-3 inhibitors cAMP-dependent vasodilation Milrinone Hypotension

PDE-5 inhibitors cGMP-dependent vasodilation Sildenal Hypotension
Tadalal Vision change
Vardenal Hearing loss
Priapism
been shown to decrease PVR at a dosage of up to 80 parts per million (ppm), but the effect plateaus
signicantly above 20 ppm [57]. iNO is only approved in the USA for the treatment of neonatal
hypoxic respiratory failure, but it is frequently used for pulmonary vasodilation in both critical care
and surgical settings, especially in heart transplantation, lung transplantation, and LVAD implantation
[58]. The starting concentration of iNO is usually between 10 and 20 ppm. As a free radical, NO is
further oxidized to nitrogen dioxide, NO2, which can cause pulmonary irritation and edema. At higher
doses, NO binding of hemoglobin can generate a signicant level of methemoglobin. For these rea-
sons, the levels of NO and NO2 need to be closely measured by a calibrated delivery system, and fresh
gas ow with ventilators needs to be sufciently high to avoid NO2 buildup. Furthermore, post-
operative weaning of iNO should be done gradually to avoid rebound PH and acute RHF from abrupt
withdrawal.
PGI2 analogs are a class of medication with potent vasodilatory and anti-platelet aggregation ac-
tivities. PGI2 and prostaglandin E2 (PGE2) are produced in the endothelial cells by cyclooxygenase. They
activate AC, increase intracellular cAMP, which lead to vascular smooth muscle relaxation, and inhibit
platelet aggregation. PGI2 analogs include epoprostenol (IV and inhaled), treprostinil (IV, subcutane-
ous, and inhaled), and iloprost (inhaled). Chronic IV epoprostenol has been shown in a randomized
controlled trial to have symptomatic and survival benets in patients with primary PH [59]. Further-
more, inhaled PGI2 has been investigated as a pulmonary vasodilator in cardiac surgery. Inhaled PGI2
increased the success of weaning from CPB in patients with severe mitral stenosis and PH [60]. It is as
effective as iNO in decreasing PA pressures and increasing CI in patients after heart transplantation
[61]. The delivery of iNO through proprietary systems is simple and easily adjustable, whereas the
delivery of PGI2 requires a nebulizer and a pump, with complex dosing. Nonetheless, due to the cost of
iNO, some centers have had success in transitioning postoperative patients on iNO to inhaled PGI2.
Unlike iNO, systemic hypotension can occur with inhaled PGI2 due to slower elimination. Those pa-
tients who are on a chronic stable IV PGI2 therapy should continue therapy in the perioperative or peri-
procedural setting without interruption.
In addition to PGI2, two other classes of agents are frequently used in chronic management of
primary PH: endothelin receptor antagonists and type 5 PDE (PDE-5) inhibitors. The initiation of these
oral agents should be cautious, and it is reserved for those patients with stable pulmonary arterial
hypertension [62]. Endothelins are a class of potent vasoconstrictor molecules that activate PLC
pathway to increase intracellular calcium in vascular smooth muscle cells that result in contraction and
hypertrophy, and they play an important role in PH. Three oral agents that block endothelin receptors
have been approved e bosentan, macitentan, and ambrisentan e as a part of advance therapy for
patients with PH. All of them have the risk of hepatotoxicity, and they can cause peripheral edema.
PDE-5 degrades cGMP, thereby limiting vasodilation induced by NO-dependent GC pathway. Approved
PDE-5 inhibitors e sildenal, tadalal, and vardenal e promote pulmonary vasodilation, and they are
used as a stand-alone therapy or as a part of combination therapy for PH [62]. Systemic hypotension
and ush are common side effects, and their use is cautioned when the patient is on nitrate therapy.
Patients who are on these chronic therapies may need to be transitioned to PGI2 or to iNO therapy
during the perioperative setting.
Finally, positive inotropic agents mentioned in the Section LV dysfunction e dobutamine,
milrinone, and levosimendan e can enhance RV inotropy and reduce PVR and afterload. However,
all of these IV agents, unlike iNO, also cause signicant systemic hypotension and possible RV
ischemia. Recently, nebulized milrinone has been investigated as a pulmonary vasodilator, as a less
expensive alternative to iNO. In a randomized study, inhaled milrinone is as effective as its IV
counterpart in reducing PAP while preserving systemic arterial pressure and SVR [63]. As
mentioned earlier, vasoconstrictors are often necessary to counteract the systemic hypotension
associated with the use of ino-dilator such as milrinone and dobutamine. The choice of vaso-
constrictor agent is of some debate as catecholamines such as norepinephrine may increase PVR as
well as SVR. In low doses, vasopressin appears to have less effect on PVR. One small study in off-
pump coronary bypass surgery showed that vasopressin compared to norepinephrine e used to
restore milrinone-induced systemic hypotension e caused a small but favorable decrease in PVR/
SVR ratio [64].
Summary
Successful weaning from CPB or ECMO poses a signicant challenge to the cardiothoracic surgeons
and the anesthetists. The withdrawal of MCS places strain on the patient's native cardiopulmonary
systems, and hemodynamic instability due to failed weaning attempt can be injurious. Diagnosing the
cause of instability should be based on preoperative risk factors, perioperative course, echocardio-
graphic examination, and clinician assessment. Pharmacological treatment should be directed at the
underlying pathophysiology while remaining mindful of the therapeutic and side-effect proles of
each agent. While prolonged ECMO support may lead to other morbidities, maintaining the patient on
ECMO or transitioning one who has failed weaning from CPB to ECMO may be appropriate by allowing
time for cardiopulmonary recovery.
Practice points
1. Preoperative cardiopulmonary function and clinical history before the initiation of CPB and/
or ECMO have a predictive value in the ease of weaning.
2. Echocardiography is invaluable in the assessment of cardiac function and the diagnosis of
problems during weaning from CPB and ECMO.
3. Common problems during weaning from CPB and ECMO include vasodilation, LV
dysfunction, LVOTO, and RV dysfunction. Multiple problems may coexist.
4. Vasoconstrictors are used to treat systemic vasodilation, but these may cause tissue mal-
perfusion and ischemia.
6. Positive inotropic agents are useful in the treatment of LCOS, but these may cause
arrhythmia and increase myocardial oxygen demand.
7. LVOTO is usually managed medically, involving volume resuscitation, HR control, and
avoidance of positive inotropic agents.
8. The management of RV dysfunction includes optimization of RV preload, decreasing RV
afterload with pulmonary vasodilators, inotropic support, and maintaining RV perfusion by
avoiding systemic hypotension.
9. Multiple agents may be used either working synergistically or to compensate for another's
deleterious side effects.
Research agenda
1. Development and trials of novel agents for the treatment of VS, LCOS, and RHF.
2. Validation of prognostic tools to predict RHF and development of new therapeutic strategies.
3. Improvement in MCS technology and the management of patients on extended ECMO to
decrease associated morbidities.
Conict of interest
None.
Acknowledgment
None.
References
[1] Warren OJ, Smith AJ, Alexiou C, et al. The inammatory response to cardiopulmonary bypass: part 1emechanisms of
pathogenesis. J Cardiothorac Vasc Anesth 2009;23:223e31.
*[2] Licker M, Diaper J, Cartier V, et al. Clinical review: management of weaning from cardiopulmonary bypass after cardiac
surgery. Ann Card Anaesth 2012;15:206e23.
[3] Fergusson DA, He bert PC, Mazer CD, et al. A comparison of aprotinin and lysine analogues in high-risk cardiac surgery.
N Engl J Med 2008;358:2319e31.
[4] Ferraris VA, Brown JR, Despotis GJ, et al. 2011 update to the Society of Thoracic Surgeons and the Society of Cardiovascular
anesthesiologists blood conservation clinical practice guidelines. Ann Thorac Surg 2011;91:944e82.
[5] Bui JD, Despotis GD, Trulock EP, et al. Fatal thrombosis after administration of activated prothrombin complex concen-
trates in a patient supported by extracorporeal membrane oxygenation who had received activated recombinant factor
VII. J Thorac Cardiovasc Surg 2002;124:852e4.
*[6] Gillies M, Bellomo R, Doolan L, et al. Bench-to-bedside review: inotropic drug therapy after adult cardiac surgeryea
systemic literature review. Crit Care 2005;3:266e79.
*[7] Argenziano M, Chen JM, Choudhri AF, et al. Management of vasodilatory shock after cardiac surgery: identication of
predisposing factors and use of a novel pressor agent. J Thorac Cardiovasc Surg 1998;116:973e80.
[8] Carrel T, Englberger L, Mohacsi P, et al. Low systemic vascular resistance after cardiopulmonary bypass: incidence, eti-
ology, and clinical importance. J Card Surg 2000;15:347e53.
[9] Levin MA, Lin H-M, Castillo JG, et al. Early on-cardiopulmonary bypass hypotension and other factors associated with
vasoplegic syndrome. Circulation 2009;120:1664e71.
[10] Philbin DM, Coggins CH, Wilson N, et al. Antidiuretic hormone levels during cardiopulmonary bypass. J Thorac Cardiovasc
Surg 1977;73:145e8.
[11] Landry DW, Oliver JA. The pathogenesis of vasodilatory shock. N Engl J Med 2001;345:588e95.
[12] Morales DLS, Garrido MJ, Madigan JD, et al. A double-blind randomized trial: prophylactic vasopressin reduces hypo-
tension after cardiopulmonary bypass. Ann Thorac Surg 2003;75:926e30.
[13] Hasija S, Makhija N, Choudhury M, et al. Prophylactic vasopressin in patients receiving the angiotensin-converting
enzyme inhibitor ramipril undergoing coronary artery bypass graft surgery. J Cardiothorac Vasc Anesth 2010;24:230e8.
[14] Papadopoulos G, Sintou E, Siminelakis S, et al. Perioperative infusion of low- dose of vasopressin for prevention and
management of vasodilatory vasoplegic syndrome in patients undergoing coronary artery bypass grafting-a double-blind
randomized study. J Cardiothorac Surg 2010;5:17.
[15] Argenziano M, Choudhri AF, Oz MC, et al. A prospective randomized trial of arginine vasopressin in the treatment of
vasodilatory shock after left ventricular assist device placement. Circulation 1997;96:II:286e90.
[16] Dnser MW, Mayr AJ, Ulmer H, et al. Arginine vasopressin in advanced vasodilatory shock: a prospective, randomized,
controlled study. Circulation 2003;107:2313e9.
[17] Kirkeboen1 KA, Strand OA. The role of nitric oxide in sepsis e an overview. Acta Anaesthesiol Scand 1999;43:275e88.
[18] Leyh RG, Kodis T, Strber M, et al. Methylene blue: the drug of choice for catecholamine-refractory vasoplegia after
cardiopulmonary bypass. J Thorac Cardiovasc Surg 2003;125:1426e31.
*[19] Levin RL, Degrange MA, Bruno GF, et al. Methylene blue reduces mortality and morbidity in vasoplegic patients after
cardiac surgery. Ann Thorac Surg 2004;77:496e9.
[20] Ozal E, Kuralay E, Yildirim V, et al. Preoperative methylene blue administration in patients at high risk for vasoplegic
syndrome during cardiac surgery. Ann Thorac Surg 2005;79:1615e9.
[21] Maslow AD, Stearns G, Butala P, et al. The hemodynamic effects of methylene blue when administered at the onset of
cardiopulmonary bypass. Anesth Analg 2006;103:2e8. table of contents.
[22] Weiner MM, Lin H-MM, Danforth D, et al. Methylene blue is associated with poor outcomes in vasoplegic shock.
J Cardiothorac Vasc Anesth 2013;27:1233e8.
[23] Gnnicker M, Brinkmann M, Donovan TJ, et al. The efcacy of amrinone or adrenaline on low cardiac output following
cardiopulmonary bypass in patients with coronary artery disease undergoing preoperative beta-blockade. Thorac Car-
diovasc Surg 1995;43:153e60.
[24] Butterworth JF, Prielipp RC, Royster RL, et al. Dobutamine increases heart rate more than epinephrine in patients
recovering from aortocoronary bypass surgery. J Cardiothorac Vasc Anesth 1992;6:535e41.
[25] Tarr TJ, Moore NA, Frazer RS, et al. Haemodynamic effects and comparison of enoximone, dobutamine and dopamine
following mitral valve surgery. Eur J Anaesthesiol Suppl 1993;8:15e24.
[26] Romson JL, Leung JM, Bellows WH, et al. Effects of dobutamine on hemodynamics and left ventricular performance after
cardiopulmonary bypass in cardiac surgical patients. Anesthesiology 1999;91:1318e28.
*[27] Feneck RO, Sherry KM, Withington PS, et al. Comparison of the hemodynamic effects of milrinone with dobutamine in
patients after cardiac surgery. J Cardiothorac Vasc Anesth 2001;15:306e15.
[28] Carmona MJC, Martins LM, Vane MF, et al. Comparison of the effects of dobutamine and milrinone on hemodynamic
parameters and oxygen supply in patients undergoing cardiac surgery with low cardiac output after anesthetic induction.
Rev Bras Anestesiol 2010;60:237e46.
[29] Thore n A, Elam M, Ricksten SE. Differential effects of dopamine, dopexamine, and dobutamine on jejunal mucosal
perfusion early after cardiac surgery. Crit Care Med 2000;28:2338e43.
[30] Patel NN, Rogers CA, Angelini GD, et al. Pharmacological therapies for the prevention of acute kidney injury following
cardiac surgery: a systematic review. Heart Fail Rev 2011;16:553e67.
[31] Gattiker R, Schmid E. Haemodynamic effects of dopamine, epinephrine and orciprenaline (Alupent) in patients early after
cardiac surgery. Intensive Care Med 1978;4:55e61.
[32] Salomon NW, Plachetka JR, Copeland JG. Comparison of dopamine and dobutamine following coronary artery bypass
grafting. Ann Thorac Surg 1982;33:48e54.
[33] Rosseel PM, Santman FW, Bouter H, et al. Postcardiac surgery low cardiac output syndrome: dopexamine or dopamine?
Intensive Care Med 1997;23:962e8.
[34] DiSesa VJ, Gold JP, Shemin RJ, et al. Comparison of dopamine and dobutamine in patients requiring postoperative cir-
culatory support. Clin Cardiol 1986;9:253e6.
[35] Janelle GM, Urdaneta F, Blas ML, et al. Inhibition of phosphodiesterase type III before aortic cross-clamping preserves
intramyocardial cyclic adenosine monophosphate during cardiopulmonary bypass. Anesth Analg 2001;92:1377e83.
[36] Doolan LA, Jones EF, Kalman J, et al. A placebo-controlled trial verifying the efcacy of milrinone in weaning high-risk
patients from cardiopulmonary bypass. J Cardiothorac Vasc Anesth 1997;11:37e41.
[37] Lobato EB, Florete O, Bingham HL. A single dose of milrinone facilitates separation from cardiopulmonary bypass in
patients with pre-existing left ventricular dysfunction. Br J Anaesth 1998;81:782e4.
[38] Kikura M, Sato S. The efcacy of preemptive Milrinone or amrinone therapy in patients undergoing coronary artery
bypass grafting. Anesth Analg 2002;94:22e30. table of contents.
[39] Fleming GA, Murray KT, Yu C, et al. Milrinone use is associated with postoperative atrial brillation after cardiac surgery.

[40] Papp Z, Edes I, Fruhwald S, et al. Levosimendan: molecular mechanisms and clinical implications: consensus of experts
on the mechanisms of action of levosimendan. Int J Cardiol 2012;159:82e7.
*[41] Harrison RW, Hasselblad V, Mehta RH, et al. Effect of levosimendan on survival and adverse events after cardiac surgery: a
meta-analysis. J Cardiothorac Vasc Anesth 2013;27:1224e32.
[42] Lahtinen P, Pitka nen O, Po
lo
nen P, et al. Levosimendan reduces heart failure after cardiac surgery: a prospective, ran-
domized, placebo-controlled trial. Crit Care Med 2011;39:2263e70.
[43] Eriksson HI, Jalonen JR, Heikkinen LO, et al. Levosimendan facilitates weaning from cardiopulmonary bypass in patients
undergoing coronary artery bypass grafting with impaired left ventricular function. Ann Thorac Surg 2009;87:448e54.
[44] Levin R, Degrange M, Del Mazo C, et al. Preoperative levosimendan decreases mortality and the development of low
cardiac output in high-risk patients with severe left ventricular dysfunction undergoing coronary artery bypass grafting
with cardiopulmonary bypass. Exp Clin Cardiol 2012;17:125e30.
[45] Alvarez J, Taboada M, Rodrguez J, et al. Hemodynamic effects of levosimendan following cardiac surgery. Rev Esp
Anestesiol Reanim 2005;52:389e94.
[46] Alvarez J, Bouzada M, Fern andez AL, et al. [Hemodynamic effects of levosimendan compared with dobutamine in patients
with low cardiac output after cardiac surgery. Rev Espan ~ ola Cardiol 2006;59:338e45.
[47] Al-Shawaf E, Ayed A, Vislocky I, et al. Levosimendan or milrinone in the type 2 diabetic patient with low ejection fraction
undergoing elective coronary artery surgery. J Cardiothorac Vasc Anesth 2006;20:353e7.
[48] De Hert SG, Lorsomradee S, Cromheecke S, et al. The effects of levosimendan in cardiac surgery patients with poor left
ventricular function. Anesth Analg 2007;104:766e73.
*[49] Loulmet DF, Yaffee DW, Ursomanno PA, et al. Systolic anterior motion of the mitral valve: a 30-year perspective. J Thorac
[50] Varghese R, Anyanwu AC, Itagaki S, et al. Management of systolic anterior motion after mitral valve repair: an algorithm.
J Thorac Cardiovasc Surg 2012;143:S2e7.
[51] Maslow AD, Regan MM, Haering JM, et al. Echocardiographic predictors of left ventricular outow tract obstruction and
systolic anterior motion of the mitral valve after mitral valve reconstruction for myxomatous valve disease. J Am Coll
Cardiol 1999;34:2096e104.
[52] Kato TS, Farr M, Schulze PC, et al. 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 2012;109:246e51.
[53] Kormos RL, Teuteberg JJ, Pagani FD, et al. Right ventricular failure in patients with the HeartMate II continuous-ow left
ventricular assist device: incidence, risk factors, and effect on outcomes. J Thorac Cardiovasc Surg 2010;139:1316e24.
[54] Slaughter MS, Rogers JG, Milano C a, et al. Advanced heart failure treated with continuous-ow left ventricular assist
device. N Engl J Med 2009;361:2241e51.
[55] Pettinari M, Jacobs S, Rega F, et al. Are right ventricular risk scores useful? Eur J Cardiothorac Surg 2012;42:621e6.
*[56] Denault AY, Haddad F, Jacobsohn E, et al. Perioperative right ventricular dysfunction. Curr Opin Anaesthesiol 2013;26:
71e81.
[57] Frostell C, Fratacci MD, Wain JC, et al. Inhaled nitric oxide. A selective pulmonary vasodilator reversing hypoxic pul-
monary vasoconstriction. Circulation 1991;83:2038e47.
[58] George I, Xydas S, Topkara VK, et al. Clinical indication for use and outcomes after inhaled nitric oxide therapy. Ann
Thorac Surg 2006;82:2161e9.
[59] Barst RJ, Rubin LJ, Long WA, et al. A comparison of continuous intravenous epoprostenol (prostacyclin) with conventional
therapy for primary pulmonary hypertension. N Engl J Med 1996;334:296e301.
[60] Fattouch K, Sbraga F, Sampognaro R, et al. Treatment of pulmonary hypertension in patients undergoing cardiac surgery
with cardiopulmonary bypass: a randomized, prospective, double-blind study. J Cardiovasc Med Hagerst 2006;7:119e23.
[61] Khan TA, Schnickel G, Ross D, et al. A prospective, randomized, crossover pilot study of inhaled nitric oxide versus inhaled
prostacyclin in heart transplant and lung transplant recipients. J Thorac Cardiovasc Surg 2009;138:1417e24.
*[62] Gordon C, Collard CD, Pan W. Intraoperative management of pulmonary hypertension and associated right heart failure.
Curr Opin Anaesthesiol 2010;23:49e56.
[63] Wang H, Gong M, Zhou B, et al. Comparison of inhaled and intravenous milrinone in patients with pulmonary hyper-
tension undergoing mitral valve surgery. Adv Ther 2009;26:462e8.
[64] Jeon Y, Ryu JH, Lim YJ, et al. Comparative hemodynamic effects of vasopressin and norepinephrine after milrinone-
induced hypotension in off-pump coronary artery bypass surgical patients. Eur J Cardiothorac Surg 2006;29:952e6.
Best Practice & Research Clinical Anaesthesiology
Vol. 29, No. 2, p. I1, 2015
doi:10.1016/S1521-6896(15)00047-6
available online at http://www.sciencedirect.com
Keyword index
acute kidney injury, 151 inflammation, 113
acute lung injury, 163 inotropic agents, 257
acute respiratory distress syndrome, 163 intensive care, 151
acute respiratory distress syndrome/acute lung intensive insulin therapy, 177
injury, 229 ischemiareperfusion injury, 137
anaesthesia, 151
antithrombin III, 189 left ventricular dysfunction, 257
cardiac anaesthesia, 151 mechanical circulatory support, 203, 257

cardiac surgery, 99, 177 myocardium, 137
cardiogenic shock/postcardiotomy shock, 229
cardioprotection, 137 neurocognitive dysfunction, 125
cardiopulmonary bypass, 99, 137, 151, 163, 177, neuroprotection, 125
189, 241, 257 neutrophil recruitment, 113
conditioning, 137
congenital heart disease, 241 pulmonary complications, 163
critical care, 177 pulmonary hypertension, 257
pumpless extracorporeal lung assist, 229
ECMO, 125
extracorporeal circulation, 99, 113, 163 renal injury, 151
extracorporeal life support, 257 right ventricular dysfunction, 257
extracorporeal life support/extracorporeal
membrane oxygenation, 229 shock liver, 151
extracorporeal membrane oxygenation, 257 stroke, 125
systemic inflammatory response syndrome,
glucose variability, 177 113
glycemic control, 177
ultrafiltration, 241
heart, 137 unfractionated heparin, 189
heartlung machines, 99
heparin resistance, 189 vasoconstrictor agents, 257
hepatic dysfunction, 151 vasoplegic syndrome, 257
history, 99 ventricular assist device, 203
hyperglycemia, 177
hypoglycemia, 177 whole-blood-activated coagulation time, 189

2015 Revista Clinical Anesthesiology

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

2015 Revista Clinical Anesthesiology

Загружено:

Авторское право:

Доступные форматы

CLINICAL

Contents lists available at ScienceDirect

Best Practice & Research Clinical

Julie L. Huffmyer, MD, Assistant Professor Anesthesiology

Contents lists available at ScienceDirect

Best Practice & Research Clinical

History of cardiopulmonary bypass (CPB)

Venous cannulation and drainage

Anticoagulation and bleeding

Hemodilution, anemia, and hemoltration

Modifying the ECC and conduct of CPB to improve organ preservation

Less invasive cardiac surgery

Safety and accident surveys

Evidence-based practice and guidelines

Education and simulation

Contents lists available at ScienceDirect

Best Practice & Research Clinical

Inammatory response and extracorporeal

Mediators of the inammatory response

Coagulation linked to inammation

Cytokines and chemokines

Platelets and inammation

In addition to their coagulation function, platelets contribute to the inammatory response

Inammatory response and organ dysfunction

Technical approaches to ameliorate the inammatory response

Approach Targeted mechanism Clinical effect Reference

Technical Coating circuit Reduce activation of blood Blood transfusions Y [56]

Pleiotropic effects of statins

The authors have no conicts of interest to declare.

Contents lists available at ScienceDirect

Best Practice & Research Clinical

Neurocognitive outcomes after extracorporeal

Veno-venous versus veno-arterial ECMO

Denition of neurocognitive dysfunction

Neurocognitive dysfunction ranges from subtle neurologic or neuropsychologic impairment to

Neurocognitive outcomes and ECMO

Study Population n Findings

ndings on examination, yet 41% demonstrated neuropsychological performance impairment, and

Etiology and prevention

Neurocognitive test Cognitive domain Description

Rey auditory Verbal memory Immediate recall of 15 verbally presented, unrelated

Emerging experimental concepts

Standardized definitions of neurocognitive dysfunction should be used for clinical trials.

Conict of interest statement

Contents lists available at ScienceDirect

Best Practice & Research Clinical

Myocardial injury and protection related to

Mechanisms of myocardial injury related to CPB

Cardioprotective strategies during CPB: key approaches

Preparing the myocardium for CPB

Modulation of the inammatory response

Modulation of myocardial oxygen consumption

Conditioning the myocardium

Clinical application of conditioning in cardiac surgery

SDH has received speaker's fees from Abbott/Abbvie and Baxter.

* Metabolic preparation of the myocardium

Contents lists available at ScienceDirect

Best Practice & Research Clinical

Hepatic and renal effects of cardiopulmonary

Hepatic effects of CPB

Pathophysiology and markers of liver dysfunction

Congestive hepatopathy Ischaemic hepatitis

Co-morbidity RH failure; high CVP LV dysfunction;

Patient with pre-operative liver disease