Intraoperative frozen section analysis of the proximal bile ducts in hilar cholangiocarcinoma is of limited value

Frozen section analysis (FS) during cancer surgery is widely used to assess resection margins. However, in hilar cholangiocarcinoma (HCCA), FS may be less reliable because of the specific growth characteristics of the tumor. The aim of this study was to determine the accuracy and consequences of intraoperative FS of the proximal bile duct margins in HCCA. Between 1990 and 2014, 67 patients underwent combined extrahepatic bile duct resection and partial liver resection for HCCA with the use of FS. Sensitivity and specificity of FS was 68% and 97%, respectively. Seventeen of 67 patients (25%) displayed a positive bile duct margin at FS. The false-negative rate was 16% (eight patients). Ten patients (15%) with a positive bile duct margin underwent an additional resection in an attempt to achieve negative margins, which succeeded in three patients (4%). However, only one of these three patients did not have concomitant lymph node metastases, which are associated with a poor prognosis by itself. The use of FS of the proximal bile duct is of limited clinical value because of the relatively low sensitivity, high risk of false-negative results, and the low rate of secondary obtained tumor-free resection margins. Supported by the literature, a new approach to the use of FS in HCCA should be adopted, reserving the technique only for cases in which a substantial additional resection is possible.