Surgical outcomes after cardiac surgery in liver transplant recipients

Abstract

Objective: This was a single-center retrospective study to assess the surgical outcomes and predictors of mortality of liver transplant recipients undergoing cardiac surgery.

Methods: From 2000 to 2010, 61 patients with a functioning liver allograft underwent cardiac surgery. The mean interval between liver transplantation and cardiac surgery was 5.4 ± 4.4 years. Of the 61 patients, 33 (54%) were in Child-Pugh class A and 28 in class B. The preoperative and postoperative data were reviewed.

Results: The overall in-hospital mortality was 6.6%. The survival rate was 82.4% ± 5.1% at 1 year and 50.2% ± 8.2% at 5 years. Cox regression analysis identified preoperative encephalopathy (odds ratio, 5.2; 95% confidence interval, 1.8-15.5; P = .003) and pulmonary hypertension (odds ratio, 3.5; 95% confidence interval, 1.3-9.4; P = .045) as independent predictors of late mortality. The preoperative Model for End-Stage Liver Disease (MELD) scores of patients who died in-hospital or late postoperatively were significantly greater statistically than the scores of the others (in-hospital death, 23.7 ± 7.8 vs 13.1 ± 4.5, P < .001; late death, 15.2 ± 6.1 vs 12.3 ± 4.1, P = .038). The Youden index identified an optimal MELD score cutoff value of 13.5 (sensitivity, 56.0%; specificity, 67.6%). Kaplan-Meier survival analysis successfully demonstrated that the survival rate of the MELD score less than 13.5 (MELD <13.5) group was significantly greater than that of the MELD >13.5 group (MELD <13.5 group, 93.8% ± 4.2% at 1 year and 52.4% ± 11.8% at 5 years; MELD >13.5 group, 66.9% ± 9.6% at 1 year and 46.1% ± 11.1% at 5 years; P = .027). In contrast, the survival rate when stratified by Child-Pugh class (class A vs B) was not significantly different.

Conclusions: Cardiac surgery in the liver allograft recipients was associated with acceptable surgical outcomes. Preoperative encephalopathy and pulmonary hypertension were independent predictors of late mortality. The cutoff value of 13.5 in the MELD score might be useful for predicting surgical mortality in cardiac surgery.