Comparison of pancreaticoduodenectomy and bile duct resection for middle bile duct cancer: A multi-center collaborating study of Japan and Korea

Background: It is currently unknown whether bile duct segmental resection (BDSR) is an acceptable method for localized middle bile duct cancer (mid-BDC) when R0 resection can be achieved. This study aimed to investigate the short- and long-term outcomes of mid-BDC patients treated with pancreaticoduodenectomy (PD) compared to those for BDSR. Methods: This was a retrospective, Japanese and Korean multi-center collaboration study based on patients' medical records. Results: A total of 663 patients, including 245 BDSR and 418 PD cases, were enrolled. The incidence of postoperative pancreatic fistula (3.3% vs 44.1%, P <.0001), surgical site infection in the organ space (6.1% vs 17.7%, P <.0001) and clinically problematic morbidities (15.9% vs 32.8%, P <.0001) was significantly higher in the PD group. There was no difference in the mortality rate (0.8% vs 1.7%, P =.3566). Local (33.9% vs 14.4%, P <.0001) and lymph node (22.4% vs 11.0%, P <.0001) recurrence rates were significantly higher in the BDSR group. Relapse-free survival (25.0 vs 34.0 months, P =.0184) and overall survival (41.2 vs 60.1 months, P =.0019) were significantly longer in the PD group. The PD group had significantly better prognosis in stage IA/IB cases (58.3 vs 111.5 months, P =.0067), which were the best indicators for BDSR, even when R0 resection was achieved. In multivariate analysis, BDSR was an independent poor prognostic factor. Conclusion: Despite the inferior perioperative short-term outcomes, our data advocate that PD should be the standard procedure for mid-BDCs and that BDSR should be avoided even if R0 resection can be achieved. (UMIN000017914).