RESECTIONAL SURGERY OF HILAR CHOLANGIOCARCINOMA - A MULTIVARIATE-ANALYSIS OF PROGNOSTIC FACTORS

Purpose: To define the prognostic factors after surgical resection of bile duct carcinomas at the hepatic bifurcation. Patients and Methods: The retrospective single-center experience details 151 patients after surgical resection of central bile duct carcinoma performed between 1 971 and 1995. Tumor removal was accomplished by resection of the bile duct bifurcation either alone (group I, n = 33), in combination with h epatic resection (group II, n = 77), or combined with hepatic and vasc ular resection (group III, n = 41). Survival analysis was performed by the Kaplan-Meier method and the relationship between each of the clin icopathologic variables and survival was assessed by the log-rank test . Multivariate results were confirmed using Cox regression. Results: T he overall hospital mortality rate was 9.9% and depended on the extent of resection (group I, 6.1%; group II, 7.8%; group III, 17.1%). After exclusion of hospital deaths, the overall patient survival rate was 2 8.4% at 5 and 15.5% at 10 years, with a median survival duration of 2. 05 +/- 0.23 years. Univariate survival analysis identified tumor size, lymph node metastases, residual tumor stage, and tumor grading as fac tors with a statistically significant prognostic impact. Survival prog nosis was not influenced by the site of the tumor according to the cla ssification of Bismuth and Corlette, extent of resection, Internationa l Union Against Cancer (UICC) stage, perineural and vascular invasion, age, or sex. In a multivariate Cox analysis, only lymph node metastas es and residual tumor stage proved to be of independent prognostic sig nificance. Conclusion: Resection of central bile duct carcinoma is fea sible in many patients and a favorable outcome after resection is main ly determined by curative resection and the absence of lymph node meta stases. (C) 1997 by American Society of Clinical Oncology.