In patients undergoing cardiovascular operations, advanced

Transcription

1 Risk Factor Analysis in Patients With Liver Cirrhosis Undergoing Cardiovascular Operations Akimasa Morisaki, MD, Mitsuharu Hosono, MD, Yasuyuki Sasaki, MD, Shoji Kubo, MD, Hidekazu Hirai, MD, Shigefumi Suehiro, MD, and Toshihiko Shibata, MD Departments of Cardiovascular Surgery and Hepato-Biliary-Pancreatic Surgery, Osaka City University Graduate School of Medicine, and Department of Cardiovascular Surgery, Osaka City General Hospital, Osaka, Japan Background. Variable outcomes of cardiac operations have been reported in cirrhotic patients, but no definitive predictive prognostic factors have been established. This retrospective study assessed operative results to identify risk factors associated with morbidity after cardiovascular operations in cirrhotic patients. Methods. The study comprised 42 cirrhotic patients who underwent cardiovascular operations from January 1991 to January Thirty patients were Child-Turcotte-Pugh class A, and 12 were class B. Hospital morbidity occurred in 13 patients (31.0%; M group), including 4 who died in-hospital. Patients without severe complications (N group) were compared with the M group patients. The Model for End-Stage Liver Disease (MELD) score was evaluated in 25 patients. Results. Significant differences in hospital morbidity between the M vs N groups were identified for platelet count ( vs / L), MELD score ( vs ), operation time ( vs minutes), and cardiopulmonary bypass time ( vs minutes) in univariate analyses (p < 0.005). Platelet count, operation time, and age were significantly associated with hospital morbidity in multivariate analyses (p < 0.005). Platelet count of / L and MELD score of 13 were cutoff values for hospital morbidity. Conclusions. Careful consideration of operative indications and methods are necessary in cirrhotic patients with low platelet counts or high MELD scores. A high incidence of hospital morbidity is predicted in patients with platelet counts of less than / L or MELD scores exceeding 13. (Ann Thorac Surg 2010;89:811 8) 2010 by The Society of Thoracic Surgeons In patients undergoing cardiovascular operations, advanced liver dysfunction increases the risk of complications, including infections and impairment of various organs such as the heart, kidney, and lungs, thereby increasing the burden of postoperative management [1, 2]. Previous studies that have analyzed the risk factors in cirrhotic patients undergoing cardiac operations [3 11] found preoperatively elevated serum levels of total bilirubin and low serum concentrations of cholinesterase were associated with higher rates of operative morbidity and mortality [3 5]. The Child-Turcotte-Pugh (CTP) classification was also strongly related to morbidity and death, particularly in cirrhotic patients with CTP class B or C [6 11]. Although empirical evidence shows cardiopulmonary bypass (CPB) is contraindicated in cirrhotic patients with CTP class B or C [12], recent case reports have described successful cardiac procedures using CPB in such patients with severe liver cirrhosis [13 15]. Thus, various outcomes of cardiac operations in cirrhotic patients have been reported. A new scoring system, the Model for End-Stage Liver Disease (MELD) score, has recently been applied to Accepted for publication Dec 7, Address correspondence to Dr Morisaki, Department of Cardiovascular Surgery, Osaka City University Graduate School of Medicine, Asahimachi, Abeno-ku, Osaka, , Japan; m @ med.osaka-cu.ac.jp. evaluate liver dysfunction and to predict death after general operations. Several studies have reported that the MELD score was as good as or better than the CTP classification in patients undergoing abdominal procedures [16 19]. In patients undergoing cardiovascular operations, however, there have been conflicting results about the MELD score, which has not yet been established as a predictive factor for hepatic decompensation after cardiovascular operations. No definitive prognostic factors for cirrhotic patients undergoing cardiovascular operations have been identified because the numbers of patients and studies remain small. To allow accurate prediction of significant postoperative complications, and to determine the indications for cardiovascular operation, it is mandatory to demonstrate reliable predictive factors, including the CTP classification and the MELD score, in cirrhotic patients. The purpose of this retrospective study was to identify risk factors for fatal morbidity after cardiac surgical procedures by assessing the operative results in a relatively large sample size of cirrhotic patients undergoing cardiovascular procedures. Patients and Methods This study was approved by the Institutional Review Board of Osaka City Medical School Hospital and complied with the current ethical guidelines according to the standards of the Declaration of Helsinki. Individual con by The Society of Thoracic Surgeons /10/$36.00 Published by Elsevier Inc doi: /j.athoracsur

2 812 MORISAKI ET AL Ann Thorac Surg OPERATIVE RISK FACTORS FOR LIVER CIRRHOSIS 2010;89:811 8 Abbreviations and Acronyms ALT alanine aminotransferase AST aspartate aminotransferase CABG coronary artery bypass grafting CI confidence interval CPB cardiopulmonary bypass CTP Child-Turcotte-Pugh ICGR 15 indocyanine green retention rate at 15 minutes MELD Model for End-Stage Liver Disease NYHA New York Heart Association OR odds ratio PT prothrombin time PT-INR international normalized ratio of prothrombin time SD standard deviation Table 2. Surgical Procedures Variable No. (%) Emergency 7 (16.7) Reoperation 3 (7.1) CABG 11 (26.2) On-pump 6 (14.3) Off-pump 5 (11.9) Valve operation 20 (47.6) Single valve operation 18 (42.9) Thoracic aneurysm repair 2 (4.8) Atrial septal defect closure 3 (7.1) Combined operation 5 (11.9) Extirpation of myxoma 1 (2.4) CABG coronary artery bypass grafting; Off-pump CABG without cardiopulmonary bypass; On-pump CABG with cardiopulmonary bypass. sent for later retrospective studies was obtained at the same time as the necessity for undergoing the cardiovascular operation was explained. Patients We retrospectively reviewed patients who underwent cardiovascular operations in our institution from January 1991 to January These patients included 194 with Table 1. Patient Characteristics Variable a No (%) or Mean SD Patients, total 42 (100) Sex Female 11 Male 31 Age, y NYHA class III, IV 27 (64.3) Renal dysfunction b 10 (23.8) Hemodialysis 3 (7.1) Etiology Chronic hepatitis B 1 (2.4) Chronic hepatitis C 26 (61.9) Alcoholic liver disease 5 (11.9) Biliary cirrhosis 1 (2.4) Autoimmune hepatitis 1 (2.4) Cryptogenic 7 (16.7) Hepatocellular carcinoma 8 (19.0) CTP classification A 30 (71.4) B 12 (28.6) MELD score (n 25) a Categoric data are presented as number (%); continuous data are presented as the mean standard deviation. b Renal dysfunction; serum creatinine level 1.5 mg/dl. CTP Child-Turcotte-Pugh; MELD Model for End-Stage Liver Disease; NYHA New York Heart Association; SD standard deviation. liver function disorders, and among these, 42 patients (11 women; 31 men) with liver cirrhosis were identified. After 1996, 38 patients underwent surgical procedures and their characteristics are summarized in Table 1. The mean age was years. Of 10 patients with renal dysfunction (serum creatinine level 1.5 mg/dl), 3 were dialysis dependent. Liver cirrhosis was diagnosed by comprehensive examination of the clinical features, laboratory data, and findings from abdominal ultrasound imaging and computed tomography scans in the absence of liver biopsy. The main etiologies of the liver cirrhosis were chronic hepatitis C virus in 26 and alcoholic liver disease in 5. Among the 42 patients with liver cirrhosis, 30 were CTP class A and 12 were CTP class B. The MELD scores were measured in 25 patients, and the mean score was (range, 6 to 27). Surgical procedures (Table 2) consisted of coronary artery bypass grafting (CABG) in 11, comprising 6 with CPB (on-pump) and 5 without CPB (off-pump), and 20 patients underwent isolated valvular operations. A median sternotomy was performed in all patients. Conventional techniques of CPB and cardiac protection were used, as previously described [20]. Selective cerebral perfusion was applied in aortic arch operations. These techniques for CPB did not change during the course of this study. Blood components were transfused when required during and after the operation. The criteria for platelet transfusion were a platelet count of less than / L or a bleeding tendency after weaning from CPB. Hospital morbidity occurred in 13 patients (31%), consisting of renal failure in 6, respiratory failure in 6 (1 tracheotomy), cardiac tamponade in 4, mediastinal infection in 3; reexploration for mediastinal bleeding, hepatic decompensation, gastrointestinal bleeding, and low output syndrome in 2 patients each; and cerebral infarction in 1. Four patients (9.5%) died in the hospital of mediastinal infection (n 3) or pneumonia (n 1). Patients with hospital morbidity were defined as the M group and the patients without hospital morbidity as the N

3 Ann Thorac Surg MORISAKI ET AL 2010;89:811 8 OPERATIVE RISK FACTORS FOR LIVER CIRRHOSIS Table 3. Child-Turcotte-Pugh Classification a Variable 1 Point 2 Points 3 Points Encephalopathy None Grade 1 2 (or precipitant-induced) Grade 3 4 (or chronic) Ascites None Mild/moderate (diuretic-responsive) Severe (diuretic-refractory) Total bilirubin, mg/dl Albumin, g/dl PT (seconds prolonged) or PT-INR a The Child-Turcotte-Pugh score was obtained by adding the scores for each variable. Child-Turcotte-Pugh classes: A 5 6 points; B 7 9 points; C points. PT prothrombin time; PT-INR international normalized ratio of prothrombin time. group. The 4 dead patients were included in the M group. The risk factors were compared between the two groups. Severity of Liver Dysfunction The CTP classification is calculated based on five clinical variables: hepatic encephalopathy, ascites, serum total bilirubin, serum albumin, and prolonged prothrombin time (PT) or international normalized ratio of prothrombin time (PT-INR; Table 3). The PT-INR has been used in our hospital since November 2000, and the PT in seconds was used before then. The CTP classification corresponds to one of three groups, according to the sum of the scores. The MELD score was calculated using the standard formula published by Mayo Clinic College of Medicine, which adds multiples of the natural logarithms of the values for PT-INR, creatinine, and total bilirubin, as follows: 11.2 log e (PT-INR) 9.57 log e (serum creatinine [mg/dl]) 3.78 log e (serum total bilirubin [mg/dl]) 6.43 (an intercept) [21]. Values of less than 1 were replaced by 1 for all variables, and values exceeding 4 for creatinine were replaced by 4 (if the patients had dialysis at least twice in the past week, creatinine was set at 4). The MELD score (rounded to the nearest integer) ranges from 6 to 40, with higher values indicating more severe disease. Because the PT-INR was not adopted in our hospital until November 2000, the MELD scores could only be calculated for 25 patients. Statistical Analysis Data were analyzed using Stat View 5.0 software (SAS Institute, Cary, NC) and SPSS II (SPSS Inc, Troy, NY). Numeric variables were expressed as the mean standard deviation and analyzed with the parametric t test or nonparametric Mann-Whitney U test, as appropriate. Categoric data and the numbers or percentages of patients were compared using the 2 test and the Fisher exact test. Receiver-operating characteristic curves were then designed to identify cutoff values for predicting the risks of morbidity. The specificity and sensitivity were calculated, as well as the positive and negative predictive values. The best possible cutoff point was defined as the highest Youden Index [(specificity sensitivity) 1]. Multivariate logistic regression analyses were performed to detect factors influencing morbidity. Survival curves excluding hospital deaths were generated using the Kaplan-Meier method. The logrank test was used to compare the differences in the survival rates between the groups. Values of p 0.05 were considered statistically significant. Results Comparisons of the N and M Groups in Univariate Analyses The preoperative and operative risk factors associated with severe complications after cardiovascular operations in cirrhotic patients are reported in Tables 4 and 5, respectively. Univariate analysis identified significant differences between the M and N groups for preoperative platelet count, preoperative MELD score, operation time, CPB time, and aortic cross-clamp time. There were no significant differences in the surgical interventions between the M and N groups. The postoperative minimal platelet count was significantly lower in the M group than in the N group, although the platelet transfusion volume was higher in the M group than in the N group. The differences between the two groups for the values of the maximal serum total bilirubin and minimal total protein were not significant. Cutoff Values Receiver-operating characteristic curves were used to evaluate the cutoff values for the platelet count and MELD score as preoperative predictive factors influencing hospital morbidity in patients with liver cirrhosis undergoing cardiovascular operations (Fig 1). A platelet count of / L had a sensitivity of 69.2% and a specificity of 72.4% for predicting hospital morbidity based on all 42 cirrhotic patients. A MELD score of 13 had a sensitivity of 83.3% and a specificity of 84.2% for predicting hospital morbidity based on the 25 available cirrhotic patients. The area under the curve for the platelet count was and the area under the curve for the MELD score was Both values were statistically significant. Multivariate Analyses of Hospital Morbidity Platelet count, age, and operation time were significantly associated with hospital morbidity (Table 6). The MELD score could not be calculated for all patients and was therefore excluded from the multivariate analyses.

4 814 MORISAKI ET AL Ann Thorac Surg OPERATIVE RISK FACTORS FOR LIVER CIRRHOSIS 2010;89:811 8 Table 4. Preoperative Data Variable a M Group (n 13) N Group (n 28) No. or Mean SD No. or Mean SD p Value Female sex Age, y NYHA class III, IV Ejection fraction Creatinine, mg/dl Platelet count, 10 4 / L PT, sec Prothrombin activity, % PT-INR Albumin, g/dl Total bilirubin, mg/dl AST, IU/L ALT, IU/L Cholinesterase, IU/L ICGR 15,% CTP score CTP class B MELD score Ascites a Categoric data are presented as the number; continuous data are presented as the mean standard deviation. ALT alanine aminotransferase; AST aspartate aminotransferase; CTP Child-Turcotte-Pugh; ICGR 15 indocyanine green retention rate at 15 minutes; MELD Model for End-Stage Liver Disease; NYHA New York Heart Association; PT prothrombin time; PT-INR international normalized ratio of prothrombin time. Long-Term Outcomes After Cardiovascular Operations Four patients died in the hospital. All the hospital deaths were cirrhotic patients with CTP class B and MELD Table 5. Operative Data and Postoperative Data Variable a M Group (n 13) N Group (n 28) p Value Operation time, min CPB time, b min Cross-clamp time, b min CABG With CPB Without CPB Valve operation Combined operation Minimal platelet count, / L Minimal total protein, g/dl Maximal total bilirubin, mg/dl Platelet transfusion volume, ml a Categoric data are presented as the number; continuous data are presented as the mean standard deviation. b Patients undergoing CABG without CPB are not included. CABG coronary artery bypass grafting; CPB cardiopulmonary bypass. scores exceeding 14 who underwent emergency or urgent operations. Figure 2 shows Kaplan-Meier survival curves according to morbidity, which excluded the hospital deaths. The median survival times of the M and N groups were and months, respectively. The 1-, 3-, and 5-year survival rates were 72.9%, 43.8%, and 43.8% in the M group, and 92.7%, 74.0%, and 62.1% in the N group, respectively. The difference between the survival rates of the M and N groups was not significant (p ). Comment The results of this study demonstrated that platelet count, MELD score, operation time, CPB time, and aortic cross-clamp time were significantly correlated with hospital morbidity in patients with liver cirrhosis undergoing cardiovascular operations. Preoperative platelet count and MELD score were indicated as possible predictive prognostic factors for hospital morbidity after cardiovascular operations, and operation time and age could also be significant risk factors in cirrhotic patients. No definitive recommendations have yet been made for the indications for cardiovascular operations in cirrhotic patients owing to the small numbers of patients studied. However, the current study analyzed the predictive prognostic factors in a relatively large sample size compared with previous studies.

5 Ann Thorac Surg MORISAKI ET AL 2010;89:811 8 OPERATIVE RISK FACTORS FOR LIVER CIRRHOSIS Thrombocytopenia is an empiric indicator of progressive liver dysfunction, inducing a secondary phase of portal hypertension that causes splenomegaly, esophageal varices, hemorrhoids, and ascites. Several recent studies showed that the platelet count was associated with the degree of fibrosis in cirrhotic patients and with the presence of esophageal varices [22, 23]. Interestingly, both these studies showed that a platelet count of less than / L was a good cutoff value for predicting the presence of large varices and severe liver fibrosis. Table 6. Multivariate Analyses of Hospital Morbidity 815 Variable Adjusted OR (95% CI) p Value Female sex ( ) Age ( ) NYHA class III, IV ( ) CTP class B ( ) Platelet count, 10 4 / L ( ) Prothrombin time, sec ( ) Albumin, g/dl 0.33 ( ) Total bilirubin, mg/dl ( ) Cholinesterase, IU/L ( ) Creatinine, mg/dl ( ) Operation time, min ( ) CI confidence interval; CTP Child-Turcotte-Pugh; NYHA New York Heart Association; OR odds ratio. Filsoufi and colleagues [6] initially showed an association between preoperative platelet count and operative death in cirrhotic patients undergoing cardiac operations [6]. However, previous studies did not define a clear cutoff value that was correlated with postoperative morbidity in cirrhotic patients undergoing cardiovascular operations. The present study demonstrated that a preoperative platelet count of less than / L was a significant risk factor predicting postoperative morbidity. Our cutoff value therefore indicates advanced liver fibrosis or cirrhosis. In this study, the postoperative minimal platelet count was lower in the M group than in the N group. A reduced platelet count is considered to be a sensitive reflection of liver fibrosis and may thus be a significant risk factor predicting postoperative morbidity after cardiovascular operations. Various scoring systems for severity of liver dysfunction have been developed to predict death after general surgical procedures. The CTP score has been developed as a predictive formula for patients with liver disease undergoing portosystemic operations and has proven to Fig 1. Receiver operating characteristic curves show (A) platelet count and (B) the Model for End-Stage Liver Disease (MELD) score as predictors of complications in patients with liver cirrhosis undergoing cardiovascular operations. A platelet count of / L had a sensitivity of 69.2% and a specificity of 72.4% for predicting hospital morbidity, based on all 42 patients. A MELD score of 13 had a sensitivity of 83.3% and a specificity of 84.2% for predicting hospital morbidity, based on the 25 available cirrhotic patients. The area under the curve was for the platelet count and for the MELD score. Both values were statistically significant. Fig 2. Survival rates were compared after cardiovascular operations between the M group (with morbidity, black circles) and the N group (without morbidity, white circles).

6 816 MORISAKI ET AL Ann Thorac Surg OPERATIVE RISK FACTORS FOR LIVER CIRRHOSIS 2010;89:811 8 be a useful tool in hepatic and nonhepatic operations [24]. The MELD score has been used to stratify patients awaiting liver transplantation [25]. Several studies have compared the predictive values of the CTP classification and the MELD score in patients undergoing general surgical procedures [16 19]. The results showed that the MELD score was as good as, or better than, the CTP classification. However, there have been conflicting results for cardiac operations: Suman and colleagues [8] showed that the MELD score and the CTP score were both able to predict hepatic decompensation and death after cardiac operations in cirrhotic patients [8]. Ailawadi and colleagues [26] reported that the MELD score predicted death after tricuspid valve operations and that patients with MELD scores of 15 or higher were at particularly high risk. Filsoufi and colleagues [6], on the other hand, reported that the CTP classification was a better predictor of hospital death than the MELD score. Our present study found that the MELD score was a more significant risk factor than the CTP classification. All of the hospital deaths in this study were in CTP class B patients, which suggests the usefulness of the CTP classification as a predictive factor for death. The MELD score, however, may be a more reliable predictive factor for fatal morbidity after cardiovascular operations than the CTP classification. The MELD score comprises three objective and commonly available laboratory values: PT-INR, total bilirubin, and creatinine. Meanwhile, the CTP classification uses subjective indicators, such as ascites and encephalopathy. The MELD score therefore has the advantages of simplicity and objectivity. In addition, an evaluation of renal function (serum creatinine level) is included in the MELD score but not in the CTP classification. Renal dysfunction is a common comorbidity in cirrhotic patients and a risk factor in cardiac operations because advanced liver cirrhosis can induce significant renal complications (hepatorenal syndrome). We therefore believe that the MELD score is a useful scoring system in cirrhotic patients undergoing cardiovascular operations, and our present results indicated that patients with MELD scores exceeding 13 were more likely to experience complications. This finding was similar to those in a previous study [8]. A MELD score exceeding 13 may therefore be a reliable predictive value for morbidity in patients with liver cirrhosis undergoing cardiovascular operations. The results of this study demonstrated that prolongation of the operation, CPB, and aortic cross-clamping were strongly associated with hospital morbidity and death, showing that decisions about the surgical indications and operative method are important in cirrhotic patients with low platelet counts or high MELD scores. Surgical procedures and use of CPB induce activation of inflammatory cytokines and endotoxins, which influence the hemodynamics, coagulation system, immune system, vascular resistance and permeability, and platelet concentration and function [27, 28]. In addition, hypothermia and nonpulsatile circulation during CPB are thought to contribute to hemodynamic deterioration and synergistically exacerbate the cirrhotic state. Thus, previous studies have contraindicated the use of CPB in patients with severe liver cirrhosis, such as CTP class B or C [4 6, 9]. Recent case reports, however, have described successful on-pump cardiac operations in patients with severe liver cirrhosis. The two main problems caused by CPB in severely cirrhotic patients, coagulopathy and inflammatory responses, have been overcome. Perioperative treatments, including preoperative antibiotic therapy, sufficient perioperative transfusion of platelet concentrate and fresh frozen plasma, and intraoperative use of dilutional ultrafiltration to remove chemical mediators were used in these patients [13 15]. Careful consideration of the surgical strategy and careful perioperative management are necessary to improve operative results in cirrhotic patients with low platelet counts or high MELD scores requiring CPB. Off-pump CABG can be a good alternative strategy in patients with ischemic heart disease, taking into account the disadvantages of CPB [9]. This study had several limitations. Although the number of patients was relatively large compared with other studies, the sample size was still small. The types of operations were very mixed, ranging from low surgical interventions such as off-pump CABG and closure of an atrial septal defect to operations for valvular disease and aneurysms. It is therefore difficult to evaluate the values for stratified surgical procedures because of the very small number for each surgical procedure. The MELD scores could only be calculated for 25 patients because the PT-INR was not adopted in our hospital until November For this reason, the MELD score could not be evaluated in the multivariate analyses. It is also possible that patients considered to have severe cirrhosis did not undergo cardiovascular operations. In conclusion, the results of our study suggest that platelet count and MELD score are preoperative risk factors and that operation time, CPB time, and aortic cross-clamp time are intraoperative risk factors for hospital morbidity in cirrhotic patients undergoing cardiovascular operations. Careful consideration of the indications for cardiovascular operations and the operative methods used in cirrhotic patients with low platelet counts or high MELD scores is therefore required. A high incidence of hospital morbidity is predicted in patients with platelet counts of less than / L or MELD scores exceeding 13. We would like to thank Dr Mitsuru Fukui (Laboratory of Statistics, Osaka City University Medical School, Osaka, Japan) for assistance with the statistical analyses. References 1. Schuppan D, Afdhal NH. Liver cirrhosis. Lancet 2008;371: Moller S, Henriksen JH. Cardiovascular complications of cirrhosis. Postgrad Med J 2009;85:44 54.

7 Ann Thorac Surg MORISAKI ET AL 2010;89:811 8 OPERATIVE RISK FACTORS FOR LIVER CIRRHOSIS 3. Reinhartz O, Farrar DJ, Hershon JH, Avery GJ Jr, Haeusslein EA, Hill JD. Importance of preoperative liver function as a predictor of survival in patients supported with Thoratec ventricular assist devices as a bridge to transplantation. J Thorac Cardiovasc Surg 1998;116: Hirata N, Sawa Y, Matsuda H. Predictive value of preoperative serum cholinesterase concentration in patients with liver dysfunction undergoing cardiac surgery. J Card Surg 1999;14: An Y, Xiao YB, Zhong QJ. Open-heart surgery in patients with liver cirrhosis. Eur J Cardiothorac Surg 2007;31: Filsoufi F, Salzberg SP, Rahmanian PB, et al. Early and late outcome of cardiac surgery in patients with liver cirrhosis. Liver Transpl 2007;13: Lin CH, Lin FY, Wang SS, Yu HY, Hsu RB. Cardiac surgery in patients with liver cirrhosis. Ann Thorac Surg 2005;79: 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: Hayashida N, Shoujima T, Teshima H, et al. Clinical outcome after cardiac operations in patients with cirrhosis. Ann Thorac Surg 2004;77: 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: Klemperer JD, Ko W, Krieger KH, et al. Cardiac operations in patients with cirrhosis. Ann Thorac Surg 1998;65: Hayashida N, Aoyagi S. Cardiac operations in cirrhotic patients. Ann Thorac Cardiovasc Surg 2004;10: Iino K, Tomita S, Yamaguchi S, Watanabe G. Successful aortic valve replacement using dilutional ultrafiltration during cardiopulmonary bypass in a patient with Child-Pugh class C cirrhosis. Interact Cardiovasc Thorac Surg 2008;7: Nemati MH, Astaneh B, Zamirian M. Aortic valve replacement in a patient with liver cirrhosis and coagulopathy. Gen Thorac Cardiovasc Surg 2008;56: Takahashi M, Li TS, Ikeda Y, Ito H, Mikamo A, Hamano K. Successful aortic valve replacement for infective endocarditis in a patient with severe liver cirrhosis. Ann Thorac Cardiovasc Surg 2006;12: Befeler AS, Palmer DE, Hoffman M, Longo W, Solomon H, Di Bisceglie AM. The safety of intra-abdominal surgery in patients with cirrhosis: model for end-stage liver disease 817 score is superior to Child-Turcotte-Pugh classification in predicting outcome. Arch Surg 2005;140:650 4; discussion Northup PG, Wanamaker RC, Lee VD, Adams RB, Berg CL. Model for End-Stage Liver Disease (MELD) predicts nontransplant surgical mortality in patients with cirrhosis. Ann Surg 2005;242: Farnsworth N, Fagan SP, Berger DH, Awad SS. Child- Turcotte-Pugh versus MELD score as a predictor of outcome after elective and emergent surgery in cirrhotic patients. Am J Surg 2004;188: Hoteit MA, Ghazale AH, Bain AJ, et al. Model for end-stage liver disease score versus Child score in predicting the outcome of surgical procedures in patients with cirrhosis. World J Gastroenterol 2008;14: Fujii H, Suehiro S, Kumano H, et al. Relationship between granulocyte elastase and C3a under protamine dosing in on-pump cardiac surgery. Eur J Cardiothorac Surg 2005;28: Malinchoc M, Kamath PS, Gordon FD, Peine CJ, Rank J, ter Borg PC. A model to predict poor survival in patients undergoing transjugular intrahepatic portosystemic shunts. Hepatology 2000;31: Pohl A, Behling C, Oliver D, Kilani M, Monson P, Hassanein T. Serum aminotransferase levels and platelet counts as predictors of degree of fibrosis in chronic hepatitis C virus infection. Am J Gastroenterol 2001;96: Sanyal AJ, Fontana RJ, Di Bisceglie AM, et al. The prevalence and risk factors associated with esophageal varices in subjects with hepatitis C and advanced fibrosis. Gastrointest Endosc 2006;64: Friedman LS. The risk of surgery in patients with liver disease. Hepatology 1999;29: Martin AP, Bartels M, Hauss J, Fangmann J. Overview of the MELD score and the UNOS adult liver allocation system. Transplant Proc 2007;39: Ailawadi G, Lapar DJ, Swenson BR, et al. Model for endstage liver disease predicts mortality for tricuspid valve surgery. Ann Thorac Surg 2009;87:1460 7; discussion Paparella D, Yau TM, Young E. Cardiopulmonary bypass induced inflammation: pathophysiology and treatment. An update. Eur J Cardiothorac Surg 2002;21: Addonizio VP, Colman RW. Platelets and extracorporeal circulation. Biomaterials 1982;3:9 15. INVITED COMMENTARY Liver cirrhosis is recognized as a critical comorbidity that limits the indications for operation and worsens the prognosis of major surgical procedures. Recently, the Mayo End-Stage Liver Disease (MELD) score was introduced as an adjunct to the Child-Turcotte-Pugh classification, and positive data have accumulated from various operations compromised by liver dysfunction [1, 2]. The timely report by Morisaki and coworkers [3] indicates an important role of the MELD score. The authors analyzed the morbidity and outcome of 42 patients with liver cirrhosis who underwent cardiovascular operations to elucidate the risk factors for prediction of postoperative outcomes. After analysis of the morbidity (31%) and mortality (9.5%), they conclude that the MELD score is a reliable predictor in addition to low platelet count, operating and cardiopulmonary bypass times, and other clinical variables. Very low hospital mortality in such critical patients should be possible. Specifically, a MELD score exceeding 13 and a platelet count of less than were critical risk factors. Besides the MELD score, additional variables such as platelet count and serum cholinesterase level should be considered. Cholinesterase levels strongly correlate with the protein synthesis activity in the liver in addition to serum albumin and prothrombin time. We have found serum cholinesterase has a predictive role in advanced valvular disease when congestive liver cirrhosis or dysfunction is a crucial preoperative risk factor [4]. Morisaki and colleagues have adopted several strategies, including platelet and fresh frozen plasma transfusions as well as perioperative dilutional ultrafiltration. To reduce morbidity in liver disease, the inflammatory response and subsequent cytokine cascade activation disease should be minimized. As mentioned, long pump 2010 by The Society of Thoracic Surgeons /10/$36.00 Published by Elsevier Inc doi: /j.athoracsur