ABSTRACT
BACKGROUND: There is a paucity of information available regarding the impact of cardiac surgical procedures on patients who have undergone previous liver transplantation. The primary purpose of this study was to ascertain the survival rate and predictors of death in this specific patient population. METHODS: This retrospective cohort study consisted of a consecutive series of patients with a functioning liver allograft who subsequently underwent cardiac surgical procedures between January 1991 and December 2012. The optimal Model for End-Stage Liver Disease (MELD) score for predicting late death was identified using receiver operating characteristic curve analysis. Risk of postoperative death was determined by parametric hazard analysis. RESULTS: Between January 1991 and December 2012, 43 patients (median age, 60 years) underwent cardiac surgical procedures after liver transplantation. The median interval between liver transplant and cardiac operation was 63 months (range, 1.1 to 217 months). Three operative deaths and 24 late deaths occurred. Receiver operating characteristic curve analysis identified the optimal preoperative and postoperative MELD score cut points for predicting late death as greater than 13.8 (area under the curve = 0.674) and greater than 17 (area under the curve = 0.633), respectively. Patients with a preoperative MELD score of 13.8 or less had significantly greater survival rates than those with a MELD score greater than 13.8 (p = 0.028); patients with a postoperative MELD score of 17 of less had significantly greater survival rates than those with a MELD score greater than 17 (p < 0.001). Multivariate parametric hazard analysis identified postoperative peak creatinine level as a statistically significant predictor of death (relative risk, 1.8; p = 0.01). The 1-, 5-, and 10-year Kaplan-Meier survival rates were 90%, 51%, and 35%, respectively; postoperative mortality rates followed a constant phase model with a hazard of death of 10% per year. CONCLUSIONS: Cardiac surgical procedures can be performed with acceptable short-term and long-term outcomes in liver transplant recipients. Elevated preoperative and postoperative MELD scores and postoperative peak creatinine level may portend death in this cohort. There is a constant hazard of death of 10% per year.
