Background: Variations in the definition of bile leakage after hepatic resection have prevented the identification of risk factors for early diagnosis and efficient management. The International Study Group of Liver Surgery (ISGLS) definition standardizes reporting of this complication. It was our aim in the present study to prospectively validate the ISGLS definition of bile leakage after hepatic resection. Furthermore, we sought to identify early predictors of clinically relevant bile leakage. Methods: A total of 265 patients who underwent elective hepatic resection were enrolled prospectively. Bilirubin concentrations were determined in the serum and drainage fluid until postoperative day 5. Risk factors of Grade B/C bile leakage were assessed by the use of univariate and multivariate analyses. Results: Grade A, B, and C bile leakage was diagnosed in 23 (8.7%), 38 (14.3%), and 11 (4.1%) patients, respectively. The definition as well as severity grading of bile leakage correlated with the duration of drainage and intensive care unit and hospital stay. Perioperative mortality was 0% for Grade A, 5.2% for Grade B, and 45.4% for Grade C bile leakage (P <.0001). Multivariate analysis confirmed bilirubin concentration in the drainage fluid ≥2.4 mg/dL on postoperative day 2 (odds ratio 11.88; 95% confidence interval 5.33-26.49; P <.0001) and anatomic resection (odds ratio 3.59; 95% CI 1.08-11.97; P =.04) as independent predictors of clinically relevant bile leakage. Conclusion: The ISGLS definition and severity grading of bile leakage after hepatic resection is clinically meaningful. Bilirubin concentration in the drainage fluid on postoperative day 2 is a strong predictor of clinically relevant bile leakage.