Background/purpose: Carcinoma of the gallbladder shows diverse patterns of spread. The most appropriate surgical procedures according to the depth and extent of the spread of the tumor are still controversial.
Methods: We investigated this surgical problem clinicopathologically, especially regarding the indications for pancreatoduodenectomy (PD), by retrospectively reviewing the clinical records of 216 patients who were surgically treated for advanced gallbladder carcinoma. Detailed studies of resected specimens, in our department, showed that preservation of the pancreas head carried an increased risk of residual microscopic metastases in small peripancreatic lymphatic nodes and ducts. Some patients with curative operations had shown recurrence in lymph nodes around the head of the pancreas within a few years after PD. Therefore, we performed hepatopancreato duodenectomy (HPD) in 93 patients, some of whom underwent the HPD as a prophylactic dissection of peripancreatic lymph nodes and some of whom underwent the HPD as a curative resection due to invasion to the peripancreatic lymph nodes, duodenum, and pancreas, observed macroscopically. The surgical outcome after PD was compared with that obtained after other curative resections (non-HPD).
Results: In patients with microscopically negative lymph node metastasis without hepatoduodenal ligament invasion, PD was not necessary for a complete resection of lymphatic metastases around the pancreas head. The 5-year survival rate of these patients who had HPD was not significantly higher than that for non-HPD resections with curative intent (73% vs 63%), and lymphatic recurrence was not marked. On the other hand, in patients with positive lymph node metastases without hepatoduodenal ligament invasion, PD was necessary for a complete resection. The 5-year survival rate after HPD in these patients was significantly improved compared to that after non-HPD resections with curative intent (87% vs 17%), because lymphatic recurrence was reduced (0% vs 80%).
Conclusions: In advanced carcinomas with positive hepatoduodenal ligament invasion, although the basic operative strategy, which inevitably includes right lobectomy, should be curative, PD should not be performed for prophylactic lymphatic resection, because local recurrence in the hepatoduodenal ligament cannot be controlled by PD, and the 5-year survival rate of curative HPD was only 4% (non-HPD resection with curative intent, 18%); moreover, frequent, lethal, major postoperative complications (hepatic failure) occurred after combined right lobectomy.