OBJECTIVE; Morbidity and mortality involved in the resection of hilar cholangiocarcinoma were reviewed retrospectively. The clinicopathologic and laboratory parameters that might influence the patient's survival also were re-evaluated. BACKGROUND Although much progress has been made in the diagnosis and management of hilar cholangiocarcinoma, long-term outlook for most patients remains poor. Surgical resection is usually prohibited because of its local invasiveness, and most patients can only be managed by palliative drainage. Recently, many surgeons have adopted a more aggressive resection with varying degrees of success. Several prognostic factors in bile duct carcinoma have been proposed; however, no reports have specifically focused on resected hilar cholangiocarcinoma and its prognostic survival factors using multivariate analysis. METHODS The clinical records and pathologic slides of 49 cases with resected hilar cholangiocarcinoma were reviewed retrospectively. Twenty clinical and laboratory parameters were evaluated for their correlation with postoperative morbidity and mortality, whereas 31 variables were evaluated for their significance with postoperative survival. Variables showing statistical significance in the first univariate analysis were included in the following multivariate analysis using stepwise logistic regression test for factors affecting morbidity and mortality and Cox stepwise proportional hazard model for factors influencing survival. RESULTS There were 5 in-hospital deaths, and the cumulative 5-year survival rate in 44 patients who survived was 14.9%, with a median survival of 14.0 months. Multivariate analysis disclosed that co-existent hepatolithiasis and lower serum asparate aminotransferase levels (<90 U/L) had a significant low incidence of postoperative morbidity, whereas a serum albumin of less than 3 g/dL was the only significant factor affecting mortality. Regarding survival, univariate analysis identified eight significant factors: 1) total bilirubin > or = 10 mg/dL, 2) curative resection, 3) histologic type, 4) perineural invasion, 5) liver invasion, 6) depth of cancer invasion, 7) positive proximal resected margin, and 8) positive surgical margin. However, multivariate analysis disclosed total bilirubin > or = 10 mg/dL, curative resection, and histologic type as the three most significant independent variables. CONCLUSIONS Surgical resection provides the best survival for hilar cholangiocarcinoma. An adequate nutritional support to increase serum albumin over 3 g/dL is the most important factor to decrease postoperative mortality. Moreover, preoperative biliary drainage to decrease jaundice and a curative resection with adequate surgical margin are recommended if longer survival is anticipated. Patients with well-differentiated adenocarcinoma seem to survive longer compared to those with moderately or poorly differentiated tumors.