Accuracy and limitations of preoperative assessment of longitudinal spread of perihilar cholangiocarcinoma

Although surgical resection offers the only chance of cure of perihilar cholangiocarcinoma and R1 resection has a poor prognosis, there is no consensus on optimal preoperative assessment of its longitudinal spread. We aimed to establish the optimal means of achieving this goal. This was a retrospective, single-center study of 61 patients who had undergone multi-detector row computed tomography, endoscopic retrograde cholangiography, intraductal ultrasonography, and mapping biopsy prior to resection of perihilar cholangiocarcinomas in our institute from January 2010 and December 2021. The most accurate single methods for assessing longitudinal spread were intraductal ultrasonography and mapping biopsy (both 72.1%). A combination of all four assessment methods was accurate in 51 (83.6%) of our patients. Independent risk factors for inaccuracy were Bismuth–Corlette Type IV and high histologic-grade tumors. The R0 resection rate was higher with accurate than inaccurate assessments (90.2% vs. 30.0%, P