Cholangiocarcinoma. A spectrum of intrahepatic, perihilar, and distal tumors

Ann Surg. 1996 Oct;224(4):463-73; discussion 473-5. doi: 10.1097/00000658-199610000-00005.


Objective: The objective of this article is to introduce a simple method for classifying cholangiocarcinomas and to apply this system to analyze a large number of patients from a single institution.

Summary background data: For the past 2 decades, most western reports on cholangiocarcinoma have separated intrahepatic from extrahepatic tumors and have subclassified this latter group into proximal, middle, and distal subgroups. However, "middle" lesions are uncommon and are managed most often either with hilar resection or with pancreatoduodenectomy. The spectrum of cholangiocarcinoma, therefore, is best classified into three broad groups: 1) intrahepatic, 2) perihilar, and 3) distal tumors. These categories correlate with anatomic distribution and imply preferred treatment.

Methods: The records of all patients with histologically confirmed cholangiocarcinoma who underwent surgical exploration at The Johns Hopkins Hospital over a 23-year period were reviewed.

Results: Of 294 patients with cholangiocarcinoma, 18 (6%) had intrahepatic, 196 (67%) had perihilar, and 80 (27%) had distal tumors. Age, gender, race, and associated diseases were similar among the three groups. Patients with intrahepatic tumors, by definition, were less likely (p < 0.01) to be jaundiced and more likely (p < 0.05) to present with abdominal pain. The resectability rate increased with a more distal location (50% vs. 56% vs. 91%), and resection improved survival at each site. Five-year survival rates for resected intrahepatic, perihilar, and distal tumors were 44%, 11%, and 28%, and median survival rates were 26, 19, and 22 months, respectively. Postoperative radiation therapy did not improve survival. In a multivariate analysis resection (p < 0.001. hazard ratio 2.80), negative microscopic margins (p < 0.01, hazard ratio 1.79), preoperative serum albumin (p < 0.04, hazard ratio 0.82), and postoperative sepsis (p < 0.001, hard ratio 0.27) were the best predictors of outcome.

Conclusions: Cholangiocarcinoma is best classified into three broad categories. Resection remains the primary treatment, whereas postoperative adjuvant radiation has no influence on survival. Therefore, new agents or strategies to deliver adjuvant therapy are needed to improve survival.

MeSH terms

  • Bile Duct Neoplasms* / classification
  • Bile Duct Neoplasms* / diagnosis
  • Bile Duct Neoplasms* / mortality
  • Bile Duct Neoplasms* / surgery
  • Bile Ducts, Intrahepatic*
  • Cholangiocarcinoma* / classification
  • Cholangiocarcinoma* / diagnosis
  • Cholangiocarcinoma* / mortality
  • Cholangiocarcinoma* / surgery
  • Female
  • Humans
  • Male
  • Middle Aged
  • Survival Rate