Background/Aim: Differentiation between proximal, mid and distal extrahepatic cholangiocarcinoma (CCA) is primarily based on surgical approach, rather than differences in tumor biology. The aim of the current study was to compare disease specific survival (DSS) between proximal, mid and distal CCA after resection and to identify prognostic factors.Methods: Clinicalpathological data of 175 patients with extrahepatic CCA who had undergone resection with curative intent between 1992 and 2007 were reviewed. Treatment consisted of proximal bile duct resection with concomitant hepatectomy, local bile duct resection (mid CCA, and radiologically defined Bismuth type I and II lesions), and distal resection as part of pancreatoduodenectomy (PD). Follow-up data was analyzed for DSS. Results: Of 175 patients, 48 (28%) had proximal CCA, 37 (21%) local, and 90 (51%) distal CCA. Median length of follow-up was 19 months (IQR 10-37) for the deceased patients and 54 months (IQR 34-99) for the 36 patients (21%) alive at last follow-up. Kaplan Meier estimate of 5-year DSS (overall 26%) was 42% in proximal CCA, 23% in local CCA and 19% in distal CCA (P 0.055). Tumor positive nodes were present in 8 (17%) patients with proximal CCA as compared to 13 (35%) and 45 (50%) patients with local and distal CCA, respectively (P <.001). Proximal CCA was also the least often associated with moderate to poor tumor differentiation; ie 24 (50%) patients, compared to 22 (59%) and 80 (89%) in local and distal CCA, respectively (P <.001). Nine patients (24%) with local CCA had tumor free resection margins compared to 16 (33%) and 51 (57%) in proximal and distal CCA, respectively (P .001). Forward-entry multiple regression analysis demonstrated that lymph node status (Hazard Ratio [HR] 1.75; 95%CI 1.19-2.58), tumor differentiation (HR 2.14; 95% CI 1.32-3.47), and status of resection margin (HR 1.60; 95% CI 1.08-2.37) were independent prognosticators for worse DSS in the entire group.Conclusions: Patients with distal CCA were associated with the worst DSS after resection, however, location was not an independent prognostic factor for survival. A margin negative resection was the only prognostic factor determined by the surgical procedure. In order to improve survival of local lesions the surgical procedure should be extended either with liver resection or with pancreatoduodenectomy, depending on site of the lesion.