Do patients with liver cirrhosis undergoing cardiac surgery have acceptable

outcomes?
Amit Modi, Hunaid A. Vohra and Clifford W. Barlow
Interact CardioVasc Thorac Surg 2010;11:630-634; originally published online Aug 25,
2010;
DOI: 10.1510/icvts.2010.241190

The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://icvts.ctsnetjournals.org/cgi/content/full/11/5/630

Interactive Cardiovascular and Thoracic Surgery is the official journal of the European Association
for Cardio-thoracic Surgery (EACTS) and the European Society for Cardiovascular Surgery
(ESCVS). Copyright © 2010 by European Association for Cardio-thoracic Surgery. Print ISSN:
1569-9293.

Downloaded from icvts.ctsnetjournals.org by on November 2, 2010

 ARTICLE IN PRESS

doi:10.1510/icvts.2010.241190

Interactive CardioVascular and Thoracic Surgery 11 (2010) 630–634

www.icvts.org

Best evidence topic - Cardiac general

Do patients with liver cirrhosis undergoing cardiac surgery have
acceptable outcomes?夞
Amit Modi, Hunaid A. Vohra, Clifford W. Barlow*
Wessex Cardiothoracic Centre, Southampton University Hospitals NHS Trust, Tremona Road, Southampton SO16 6YD, UK

Received 25 April 2010; received in revised form 30 July 2010; accepted 10 August 2010

Summary

A best evidence topic in cardiac surgery was written according to a structured protocol. The question addressed was whether patients
with liver cirrhosis have acceptable outcomes after undergoing cardiac surgery. Altogether 97 papers were found using the reported search,
of which nine presented the best evidence to answer the clinical question. The author, year, journal, country of study, study type, patient
group studied, relevant outcomes, results and study weaknesses were tabulated. One prospective and another eight retrospective studies
involving adult population of patients with liver cirrhosis undergoing various cardiac surgical procedures were selected. In these studies,
the overall mortality was 17.1% and combined mean mortality for Child–Pugh class A, B and C was 5.2%, 35.4% and 70%, respectively. The
major morbidity ranged from 20 to 60% in group A and 50 to 100% in the patients with more advanced hepatic disease. Some studies have
demonstrated that thrombocytopenia, decreased serum cholinesterase and high preoperative total bilirubin levels are significantly associated
with worse clinical outcomes. These studies, although with small samples, collectively demonstrate that patients with Child–Pugh class A
cirrhosis tolerated cardiac surgical procedures with a mild increase in mortality and morbidity. However, the risk of mortality in patients
with Child–Pugh class B and C or MELD score )13 is extremely high. Nevertheless, even if these patients underwent successful surgery,
their long-term survival was significantly poorer and their health status remains compromised even well after cardiac surgery because of
persistent liver dysfunction.
䊚 2010 Published by European Association for Cardio-Thoracic Surgery. All rights reserved.

Keywords: Cirrhosis; Liver dysfunction; Heart surgery; Outcome; Evidence-based medicine; Review

1. Introduction 4. Search strategy

A best evidence topic was constructed according to a Medline 1950 to March 2010 using the OVID interface.
structured protocol. This protocol is fully described in ICVTS wexp Thoracic SurgeryyOR heart surgery.mp. OR valve
w1 x . surgery.mp. OR exp Coronary Artery BypassyOR cardiac
surgery.mp.x AND wliver dysfunction.mp. or exp Liver Dis-
easesyOR Cirrhosis.mp.x
2. Clinical scenario

A 66-year-old man with alcohol-related liver cirrhosis 5. Search outcome

presents with dyspnoea (NYHA II). Echocardiogram demon-
strated severe mitral regurgitation due to posterior mitral
leaflet prolapse, good left ventricular systolic function and
increased pulmonary artery pressures. Coronary angiogram
is pending your decision-making in favour of surgery. You
routinely use EuroSCORE to risk stratify but this does not
account for liver cirrhosis. Therefore, you resolve to search
the literature to find the best evidence.
3. Three-part question
Do wpatients with liver cirrhosisx, undergoing wcardiac
surgeryx have acceptable wclinical outcomesx?
夞 Details of Child–Pugh Classification and MELD score are available from
the author upon request.
*Corresponding author. Tel.: q44 2380 777222; fax: q44 2380 798508.
E-mail address: clifford.barlow@suht.swest.nhs.uk (C.W. Barlow).
䊚 2010 Published by European Association for Cardio-Thoracic Surgery
Four hundred and forty papers were found, of which nine
were included in the BET analysis reported below. The
relevant papers are presented in Table 1.
6. Comments
Nine clinical studies, one prospective and eight retrospec-
tive, involving 210 adult patients were found to be suitable.
The studies looked at the clinical outcomes of patients
with liver cirrhosis undergoing a wide range of cardiac
surgical procedures.
An et al. w2x studied 24 patients with liver cirrhosis who
underwent cardiac surgery to evaluate morbidity, mortality
and predictors of outcome. Patients were divided into three
groups based on Child–Pugh classification (CPC). They
observed that 53% patients with class A cirrhosis and 100%
with class B and C cirrhosis suffered postoperative compli-

Downloaded from icvts.ctsnetjournals.org by on November 2, 2010

 ARTICLE IN PRESS
A. Modi et al. / Interactive CardioVascular and Thoracic Surgery 11 (2010) 630–634 631

Table 1. Best evidence papers

Author, year, journal Patient group Outcome Key results Comments

country, and
Study type

An et al., (2007), 24 patients with liver cirrhosis Mortality Unintentional statistical

Eur J Cardiothorac Surg, underwent cardiac surgery Overall 25% bias possible
China, w2x Child–Pugh A 6%
May 1996–June 2005 Child–Pugh B 67% Mortality and morbidity
Retrospective study Child–Pugh C 100% not linked to the aetiology
(level 2b) Child–Pugh As17 of cirrhosis
Child–Pugh Bs6 Major morbidity
Child–Pugh Cs1 Child–Pugh A 53%
Child–Pugh B 100%
Mean ages53"13 years Child–Pugh C 100%

M:Fs10:14 Mean ventilation time 32"22 hours

Aetiology of cirrhosis: Mean ITU stay 11"8 days

Post-hepatics15
Cardiacs7 Re-opening rates
Alcohols1 Child–Pugh A 18%
Child–Pugh B 50%

Predictors of poor High preoperative

outcome serum bilirubin

Low preoperative serum

cholinesterase

Prolonged CPB time

Murashita et al., (2009), 12 patients with liver cirrhosis Mortality Inherent biases
Gen Thorac Cardiovasc underwent cardiac surgery Overall 33%
Surg, Japan, w3x Child–Pugh A 50% Relatively small sample
January 2002–December 2006 Child–Pugh B 17%
Retrospective study Surprisingly, mortality is
(level 2b) Child–Pugh As6 Major morbidity lower in stage B, although
Child–Pugh Bs6 Child–Pugh A 50% morbidity is higher
Child–Pugh B 100%

Predictors of poor Thrombocytopenia

outcome low preoperative serum
cholinesterase

Filsoufi et al., (2007), 27 patients with liver cirrhosis Mortality Early and late outcomes
Liver Transpl, underwent cardiac surgery Overall 26% predicted
USA, w4x Child–Pugh A 10%
January 1998–December 2004 Child–Pugh B 18% Operative mortality not
Retrospective study Child–Pugh C 67% significantly correlated

Best Evidence
(level 2b) Child–Pugh As10 with MELD scores

Topic
Child–Pugh Bs11 Major morbidity
Child–Pugh Cs6 Child–Pugh A 20% No mortality observed in
Child–Pugh B 56% OPCAB patients
Mean ages58"10 years Child–Pugh C 100%

M:Fs20:7 Re-opening rates

Child–Pugh A 10%
With CPBs22
Without CPBs5 One-year survival
Child–Pugh A 80%
Aetiology of cirrhosis: Child–Pugh B 45%
Infectives13 Child–Pugh C 16%
Alcohols8
Others6 Predictors of poor Mortality worse with
outcome increasing Child–Pugh
class and preoperative
thrombocytopenia

Morisaki et al., (2010),
Ann Thorac Surg,
Japan, w5x
Retrospective study
(level 2b)
42 patients with liver cirrhosis Mortality Relatively large sample
underwent cardiac surgery Overall 9.52% size
Child–Pugh A 0%
January 1991–January 2009 Child–Pugh B 33% Perioperative deaths
excluded from Kaplan–
Meier survival analysis
(Continued on next page)
Downloaded from icvts.ctsnetjournals.org by on November 2, 2010
 ARTICLE IN PRESS
632 A. Modi et al. / Interactive CardioVascular and Thoracic Surgery 11 (2010) 630–634

Table 1. (Continued)
Author, year, journal Patient group Outcome Key results Comments
country, and
Study type

Child–Pugh As30 Median survival

Child–Pugh Bs12 Morbidity group (M) 31"26.6 months
No morbidity group (N) 35.7"19.8 months
Mean ages69"8.5 years (Ps0.3146)

M:Fs31:11 Predictors of poor Significant morbidity

outcome with MELD )13
Postoperative
Morbidity: ns13 Multivariate:
No morbidity: ns29 Age
Operative time
Aetiology of cirrhosis: Postoperative low
Infectives26 platelet count
Alcohols5
Others11

Mean MELD scores11.8"6

Suman et al., (2004), 44 patients with liver cirrhosis Mortality Relatively large sample
Clin Gastroenterol underwent on-pump cardiac Overall 16%
Hepatol, USA, w6x surgery Child–Pugh A 3.2% Cut-off for MELD score
Child–Pugh B 41.6% was not established
Retrospective study January 1992–June 2002 Child–Pugh C 100%
(level 2b)
Analysed for any relationship Liver decompensation
of child classyscore and MELD Child–Pugh A 9.7%
score for postoperative hepatic Child–Pugh B 67%
decompensation and death Child–Pugh C 100%

Child–Pugh As31 CP score )7 86% sensitivity and

Child–Pugh Bs12 92% specificity for
Child–Pugh Cs1 mortality

Lin et al., (2005), 18 patients with liver cirrhosis Mortality Small cohort
Ann Thorac Surg, underwent cardiac surgery Overall 6%
China, w7x Child–Pugh A 8% Demonstrates the
January 1993–May 2004 Child–Pugh B 0% feasibility of cardiac
Retrospective study Child–Pugh C 0% surgery in patients with
(level 2b) Child–Pugh As13 advanced liver cirrhosis
Child–Pugh Bs4 Major morbidity
Child–Pugh Cs1 Child–Pugh A 39% However, no
Child–Pugh B 80% postoperative follow-up
Median ages56 years Child–Pugh C 80%
(35–76 years)
Median ITU stay 2 (1–44) days
M:Fs14:4
Median hospital length 15 (7–85) days
With CPBs16 of stay
Without CPBs2

Aetiology of cirrhosis:
Infectives13
Alcohols5

Klemperer et al., (1998), 13 patients with liver cirrhosis Mortality Small cohort
Ann Thorac Surg, underwent cardiac surgery Overall 31%
USA, w8x Child–Pugh A 0% Three times higher
1990–1996 Child–Pugh B 80% transfusion requirement
Retrospective study
(level 2b) Child–Pugh As8 Major morbidity
Child–Pugh Bs5 Overall 54%
Child–Pugh A 25%
Mean ages65"8.3 years Child–Pugh B 100%

M:Fs11:2 Re-opening rates

Child–Pugh A 12.5%
Aetiology of cirrhosis: Child–Pugh B 100%
Infectives2
Alcohols10
PBCs1

(Continued on next page)

Downloaded from icvts.ctsnetjournals.org by on November 2, 2010
 ARTICLE IN PRESS
A. Modi et al. / Interactive CardioVascular and Thoracic Surgery 11 (2010) 630–634 633

Table 1. (Continued)
Author, year, journal Patient group Outcome Key results Comments
country, and
Study type

Hayashida et al., (2004), 18 patients with liver cirrhosis Mortality Small cohort
Ann Thorac Surg, underwent cardiac surgery Overall 17%
Japan, w9x Child–Pugh A 0% Patients operated without
February 1989–January 2003 Child–Pugh B 28.5% CPB had significantly
Retrospective study Child–Pugh C 100% less morbidity and no
(level 2b) Child–Pugh As10 mortality
Child–Pugh Bs7 Major morbidity
Child–Pugh Cs1 Child–Pugh A 60%
Child–Pugh B (CPB) 100%
Mean ages64"12 years Child–Pugh B (no CPB) 33%
Child–Pugh C 100%
M:Fs11:7
Mean ventilation time 28"20 hours
With CPBs15
Without CPBs3 Mean ITU stay 13"24 days

Aetiology of cirrhosis:
Infectives12
Alcohols3
PBCs1
Idiopathics1

Bizouarn et al., (1999), 12 patients with liver cirrhosis Mortality The only prospective
Ann Thorac Surg, undergoing cardiac surgery Overall 8% study
France, w10x Child–Pugh A 0%
1995–1997 Child–Pugh B 50% Small sample
Prospective
observational study Child–Pugh As10 Major morbidity All patients were operated
(level 2a) Child–Pugh Bs2 Child–Pugh A 50% electively and high doses
Child–Pugh B 100% of aprotinin were used
Mean ages58.8"13.9 years
Median ITU stay Three (two to 10) days
M:Fs8:4
Median hospital length 15 (7–36) days
Aetiology of cirrhosis: of stay
Infectives2
PBCs2
Alcohols7
Unknowns1

MELD, model for end-stage liver disease; CPB, cardiopulmonary bypass; ITU, intensive treatment unit; PBC, primary biliary cirrhosis; OPCAB, off-pump coronary
artery bypass grafting.

cations. Postoperative mortality of patients with class A, mortality for 12 cirrhotic patients who underwent cardiac
class B and class C cirrhosis were 6%, 67%, and 100%, surgery. Six patients each, with class A and B cirrhosis had

Best Evidence
respectively. Increased total serum bilirubin levels, low a mortality of 50% and 17%, and morbidity of 50% and 100%,

serum cholinesterase levels preoperatively and prolonged
cardiopulmonary bypass (CPB) times were found to be
predictors of worse outcome. Murashita et al. w3x reviewed
clinical outcomes and aimed to determine risk factors for
Topic
respectively. Patients who experienced major morbidities
had markedly lower levels of serum cholinesterase
(Ps0.02) and lower platelet level (Ps0.04). An excellent
retrospective review by Filsoufi and colleagues w4x including

Table 2. Combined mortality analysis

Child–Pugh A Child–Pugh B Child–Pugh C Total

Deaths Patients Deaths Patients Deaths Patients Deaths Patients

Murashita et al. w3x 3 6 1 6 0 0 4 12

Filsoufi et al. w4x 1 10 2 11 4 6 7 27
Morisaki et al. w5x 0 30 4 12 0 0 4 42
Suman et al. w6x 1 31 5 12 1 1 6 44
Lin et al. w7x 1 13 0 4 0 1 1 18
Klemperer et al. w8x 0 8 4 5 0 0 4 13
Hayashida et al. w9x 0 10 2 7 1 1 3 18
Bizouarn et al. w10x 0 10 1 2 0 0 1 12
Total 6 18 19 59 6 9 30 186
Mortality risk 5.08% 32.2% 66.6% 16.1%

Downloaded from icvts.ctsnetjournals.org by on November 2, 2010

 ARTICLE IN PRESS
634 A. Modi et al. / Interactive CardioVascular and Thoracic Surgery 11 (2010) 630–634
27 patients with cirrhosis who underwent cardiac surgery
demonstrated that stratified mortality according to CPC
was 11%, 18%, and 67% for class A, B, and C, respectively.
No mortality occurred in patients who had ‘off’ pump
revascularisation (ns5). Major postoperative complications
occurred in 22%, 56%, and 100% for CPC class A, B, and C,
respectively. One-year survival was 80%, 45%, and 16%,
respectively (Ps0.02) and the long-term survival was sig-
nificantly lower in cirrhotic patients compared to overall
patient population (Ps0.001). Further, they deduced that
CPC was a better predictor of hospital mortality (Ps0.02)
compared to model for end-stage liver disease (MELD) score
(Ps0.065). Morisaki et al. w5x reviewed 42 cirrhotic patients
who underwent cardiovascular operations (CPC A, ns30;
CPC B, ns12). They reported 33% mortality in class B with
31% (ns13) overall morbidity. Lower platelet counts and
higher MELD scores were preoperative risk factors and
operation time, CPB time and aortic cross-clamp time were
intraoperative risk factors for hospital morbidity. A higher
incidence of hospital morbidity was predicted by platelet
counts of -96,000yml or MELD scores )13. Suman et al.
w6x, conducted a retrospective study to quantify the risk of
cardiac surgery in patients with cirrhosis and reported that
the association of hepatic decompensation and mortality
with Child–Pugh class, Child–Pugh score, and MELD score
was significant (P-0.005). Twenty-seven percent devel-
oped hepatic decompensation and 16% patients died. A
‘cut-off’ Child–Pugh score )7 was found to have a sensi-
tivity and specificity of 86% and 92% for mortality, with a
negative predictive value of 97% and positive predictive
value of 67%, respectively. A series of 18 patients with
cirrhosis undergoing cardiac procedures was analysed by
Lin et al. w7x, and found an overall in-hospital mortality
rate of 6% with a major morbidity rate of 39% in CPC A and
80% in CPC B and C. Thirteen similar patients were retro-
spectively reviewed by Klemperer and colleagues w8x. With
eight patients in CPC A and five in CPC B, they observed
major postoperative complications in 25% and 100%, respec-
tively. There was no mortality in CPC A but 80% of patients
in CPC B died. They concluded that patients with minimal
clinical evidence of cirrhosis can tolerate CPB and cardiac
surgical procedures, whereas those with more advanced
liver disease should not be offered operation. Hayashida et
al. w9x assessed perioperative mortality and major organ
morbidity in 18 cardiac surgical cirrhotic patients. They
observed that the hospital mortality of patients with class
A, class B, and class C cirrhosis undergoing CPB was 0%,
50%, and 100%. No deaths were observed in ‘off’ pump
patients belonging to CPC B. Similarly, major complications
were observed in 60% of those with class A cirrhosis and
100% of those with class B cirrhosis and class C cirrhosis
Downloaded from icvts.ctsnetjournals.org by on November 2, 2010
undergoing CPB but only 33% of the class B patients without
CPB. They concluded saying that ‘off’ pump surgery can be
an alternative therapeutic strategy for patients with
advanced cirrhosis requiring surgical revascularisation.
Bizouarn et al. w10x performed a prospective study to
evaluate the early and late outcome after elective cardiac
surgery in patients with cirrhosis with 10 patients in CPC
A and two in CPC B. No mortality was observed in CPC A
and 50% in CPC B. However, the complications remained as
high as 50% and 100% in the two groups, respectively. They
inferred that the incidence of significant complications was
high even after elective cardiac surgery and the health
status remained compromised even well after the operation
because of persistent liver dysfunction.
7. Clinical bottom line
These studies demonstrate that cardiac surgery could be
offered to patients with Child–Pugh A and a low MELD score
with mild increase in the risk of mortality (Table 2).
However, for patients with more advanced cirrhosis, the
risk of mortality is unacceptably high. Revascularisation
without the use of CPB may be lower risk but in these
patients long-term survival is significantly poor and health
status is compromised even well after cardiac surgery.
References
w1x Dunning J, Prendergast B, Mackway-Jones K. Towards evidence-based
medicine in cardiothoracic surgery: best BETS. Interact CardioVasc
Thorac Surg 2003;2:405–409.
w2x An Y, Xiao YB, Zhong QJ. Open-heart surgery in patients with liver
cirrhosis. Eur J Cardiothorac Surg 2007;31:1094–1098.
w3x Murashita T, Komiya T, Tamura N, Sakaguchi G, Kobayashi T, Furukawa
T, Matsushita A, Sunagawa G. Preoperative evaluation of patients with
liver cirrhosis undergoing open heart surgery. Gen Thorac Cardiovasc
Surg 2009;57:293–297.
w4x Filsoufi F, Salzberg SP, Rahmanian PB, Schiano TD, Elsiesy H, Squire A,
Adams DH. Early and late outcome of cardiac surgery in patients with
liver cirrhosis. Liver Transpl 2007;13:990–995.
w5x Morisaki A, Hosono M, Sasaki Y, Kubo S, Hirai H, Suehiro S, Shibata T.
Risk factor analysis in patients with liver cirrhosis undergoing cardio-
vascular operations. Ann Thorac Surg 2010;89:811–817.
w6x Suman A, Barnes DS, Zein NN, Levinthal GN, Connor JT, Carey WD.
Predicting outcome after cardiac surgery in patients with cirrhosis: a
comparison of Child-Pugh and MELD scores. Clin Gastroenterol Hepatol
2004;2:719–723.
w7x Lin CH, Lin FY, Wang SS, Yu HY, Hsu RB. Cardiac surgery in patients
with liver cirrhosis. Ann Thorac Surg 2005;79:1551–1554.
w8x Klemperer JD, Ko W, Krieger KH, Connolly M, Rosengart TK, Altorki NK,
Lang S, Isom OW. Cardiac operations in patients with cirrhosis. Ann
Thorac Surg 1998;65:85–87.
w9x Hayashida N, Shoujima T, Teshima H, Yokokura Y, Takagi K, Tomoeda H,
Aoyagi S. Clinical outcome after cardiac operations in patients with
cirrhosis. Ann Thorac Surg 2004;77:500–505.
w10x Bizouarn P, Ausseur A, Desseigne P, Le Teurnier Y, Nougarede B, Train
M, Michaud JL. Early and late outcome after elective cardiac surgery
in patients with cirrhosis. Ann Thorac Surg 1999;67:1334–1338.
 Do patients with liver cirrhosis undergoing cardiac surgery have acceptable
outcomes?
Amit Modi, Hunaid A. Vohra and Clifford W. Barlow
Interact CardioVasc Thorac Surg 2010;11:630-634; originally published online Aug 25,
2010;
DOI: 10.1510/icvts.2010.241190
This information is current as of November 2, 2010

Updated Information including high-resolution figures, can be found at:

& Services http://icvts.ctsnetjournals.org/cgi/content/full/11/5/630
References This article cites 10 articles, 7 of which you can access for free at:
http://icvts.ctsnetjournals.org/cgi/content/full/11/5/630#BIBL
Subspecialty Collections This article, along with others on similar topics, appears in the
following collection(s):
Cardiac - other
http://icvts.ctsnetjournals.org/cgi/collection/cardiac_other
Permissions & Licensing Requests to reproducing this article in parts (figures, tables) or in
its entirety should be submitted to: icvts@ejcts.ch
Reprints For information about ordering reprints, please email:
icvts@ejcts.ch

Downloaded from icvts.ctsnetjournals.org by on November 2, 2010