BACKGROUND: Farrar's criteria for cystic duct carcinoma (histopathological diagnosis of a carcinoma strictly limited to the cystic duct) have practical limitations. We propose new "working definition": a gallbladder tumor, the center of which is located in the cystic duct. PATIENTS AND METHODS: Between 1980 and 2000 we diagnosed cystic duct carcinoma in 31 patients, 28 of whom (90%) had increased serum bilirubin concentrations. Extrahepatic bile duct resection and cholecystectomy were performed in 10 patients; in others, extended right hepatectomy (16), right hepatectomy (3), and liver bed resection (2) were necessary as well for a potentially curative resection. Additional procedures were portal vein resection (10) and pancreatoduodenectomy (7). RESULTS: All tumors were adenocarcinomas. Depth of invasion was T2 in 3 patients, T3 in 12, and T4 in 16. Thirteen patients (42%) had lymph node metastasis. Curative resection was performed in 24 patients (77%). Hospital mortality was 5 of 31 (16%). Actuarial 5-year survival rate excluding hospital deaths was 22%. CONCLUSIONS: The proposed "working definition" avoids the problems associated with Farrar's criteria and describes a distinct patient group with an approximately equal proportion of men and women, advanced T stage, but a lower than expected frequency of lymph node metastasis. It establishes a basis for standard reporting of results.
BACKGROUND: Farrar's criteria for cystic duct carcinoma (histopathological diagnosis of a carcinoma strictly limited to the cystic duct) have practical limitations. We propose new "working definition": a gallbladder tumor, the center of which is located in the cystic duct. PATIENTS AND METHODS: Between 1980 and 2000 we diagnosed cystic duct carcinoma in 31 patients, 28 of whom (90%) had increased serum bilirubin concentrations. Extrahepatic bile duct resection and cholecystectomy were performed in 10 patients; in others, extended right hepatectomy (16), right hepatectomy (3), and liver bed resection (2) were necessary as well for a potentially curative resection. Additional procedures were portal vein resection (10) and pancreatoduodenectomy (7). RESULTS: All tumors were adenocarcinomas. Depth of invasion was T2 in 3 patients, T3 in 12, and T4 in 16. Thirteen patients (42%) had lymph node metastasis. Curative resection was performed in 24 patients (77%). Hospital mortality was 5 of 31 (16%). Actuarial 5-year survival rate excluding hospital deaths was 22%. CONCLUSIONS: The proposed "working definition" avoids the problems associated with Farrar's criteria and describes a distinct patient group with an approximately equal proportion of men and women, advanced T stage, but a lower than expected frequency of lymph node metastasis. It establishes a basis for standard reporting of results.