Background: In recent years, the profile for patients undergoing complex liver resections has changed, with mortality rates remaining generally stable. With these factors in mind, the objective of this study was to evaluate the variables associated with surgical outcomes after hepatectomy and identify groups at high risk for postoperative mortality.Methods: The records of 1,796 patients who underwent liver resection of more than one liver segment at the Department of General and Transplantation Surgery, University Hospital Heidelberg, Germany, were analyzed. The primary end point was a 90-day in-hospital mortality. Logistic regression analyses were performed to identify risk factors associated with mortality. A risk score was created in accordance with weighted points based on the odds ratios obtained from multivariate logistic regression analyses. External validation of the score was performed, using data derived from 281 patients at the board-certified center for liver surgery in Karlsruhe, Germany.Results: The overall patient morbidity rate (Clavien-Dindo Grade II or greater) was 32%. The 30- and 90-day mortality rates were 3.0% and 4.5%, respectively. In multivariate analysis, factors independently associated with risk for 90-day in-hospital mortality were age >= 60 years (OR 3.71), ASA classification III (OR 2.94), ASA IV (15.66), perihilar cholangiocarcinoma (OR 5.65), intrahepatic cholangiocarcinoma (OR 3.08), INR >= 1.1 (OR 2.43), g-GT >= 60 U/L (OR 2.86), platelet count <= 120/nL (OR 5.52), creatinine >= 2 mg/dL (OR 9.85), and right trisectionectomy (OR 2.88). The 90-day mortality-risk score that was created based on these factors effectively stratified patients into very low risk (0-1 points, 0.2% mortality rate in 662 patients), low risk (2-3 points, 2.9% mortality rate in 769 patients), medium risk (4-5 points, 14.7% mortality rate in 232 patients), and high risk (>= 6 points, 33% mortality rate in 57 patients) groups (P < .0001). As a performance metric, the C-index for the proposed risk score for 90-day mortality was 0.86; whereas external validation revealed that this C-index was 0.89 (P = .0002).Conclusion: Based on patient-related factors and procedure-specific variables, the proposed preoperative risk score can be used to identify high-risk patients to determine 90-day mortality after liver resection. (C) 2018 Elsevier Inc. All rights reserved.