Surgery for gallbladder cancer: simple versus radical cholecystectomy, re-resection and lymphadenectomy

Simple cholecystectomy removes the gallbladder alone and is adequate only for T1a cancers confined to the lamina propria. From T1b (muscle invasion) upward, radical or extended cholecystectomy adds an en bloc resection of the gallbladder fossa (a 2 to 3 cm wedge or anatomical segments IVb and V) and a regional lymphadenectomy of the cystic, pericholedochal, hilar, hepatic artery and posterosuperior pancreaticoduodenal nodes. In a 152-patient Japanese series the first- and second-echelon node groups held nearly all metastases (pericholedochal 54 percent, cystic 38 percent, second echelon 19 to 29 percent) and more distant groups 5 percent or less, which defines the rational extent of the dissection; five-year survival after R0 resection was 80 percent when nodes were negative and 43 percent when positive. The number of positive nodes (0, 1 to 3, 4 or more) predicts outcome better than their location.

Most Western patients present as incidental cancers found after laparoscopic cholecystectomy for presumed stones, and the second operation is where the evidence concentrates. In a 115-patient multicentre series residual disease was found at re-resection in 46 percent, with liver residual disease in 0 percent of T1, 10 percent of T2 and 36 percent of T3 tumours and nodal disease in 13, 31 and 46 percent; a positive cystic duct margin predicted residual disease in the common bile duct (42 percent versus 4 percent). A Memorial Sloan Kettering series found residual disease in 54 percent of re-resected patients (T1b 36 percent, T2 48 percent, T3 70 percent), and residual disease at any site cut median disease-free survival from 93 to 11 months. Timing matters: across ten US centres, re-resection between 4 and 8 weeks after the index cholecystectomy gave a median survival of 40 months against 17 months before 4 weeks and 22 months after 8 weeks. Two meta-analyses of T1b disease favour extended over simple cholecystectomy (five-year overall survival hazard ratio 0.48 in 26 cohorts of 1,316 patients; 0.73 in 8 studies of 2,097 patients), all from non-randomised cohorts.

Three debates remain. Routine excision of the extrahepatic bile duct does not improve survival or node yield and adds morbidity, so it is reserved for a positive cystic duct margin or direct duct involvement. Port-site excision is no longer mandatory: port-site metastases fell from 19 percent before 2000 to 10 percent after, occur only with T2 or T3 tumours, mark peritoneal disease, and removing the ports did not change survival or recurrence once T and N stage were accounted for. For T2 tumours the side matters: hepatic-side T2 tumours did worse than peritoneal-side ones (five-year survival 72 versus 85 percent) and needed liver resection (80 percent five-year survival with liver resection and node dissection versus 30 percent with node dissection alone), whereas the extent of liver resection did not change survival in peritoneal-side tumours. Staging laparoscopy before laparotomy found disseminated disease in 23 percent of 409 patients at a high-volume Indian centre and avoided a non-therapeutic laparotomy in 56 percent of those with unresectable disease. Laparoscopic re-resection is oncologically equivalent to open in selected patients (three-year survival 87 versus 62 percent, not significant) with a shorter stay.