Aim: The modified Blumgart mattress (BM) and conventional interrupted suture (IS) methods are currently the most widely adopted pancreaticojejunostomy (PJ) techniques utilized during minimally invasive pancreaticoduodenectomy (MIPD). This study aimed to evaluate the postoperative outcomes between the two PJ techniques using robotic and laparoscopic approaches.

Methods: This was a retrospective study involving patients who underwent robotic or laparoscopic pancreaticoduodenectomy (PD) performed by two surgeons from two institutions. Surgical outcomes of the patients were compared according to the PJ techniques of robotic BM (Rob-BM), robotic IS (Rob-IS), and laparoscopic IS (Lap-IS), which were further analyzed among patients who had a soft pancreas and small pancreatic duct, while those with pancreatic ductal adenocarcinoma were excluded from the study.

Results: A total of 230 patients underwent MIPD with 63 Rob-BM, 48 Rob-IS, and 119 Lap-IS for PJ. Within the study population, clinically relevant-postoperative pancreatic fistula (CR-POPF) rates were comparable between Rob-BM and Rob-IS (6.3% vs. 10.4%, P = 0.283) and between Rob-IS and Lap-IS (10.4% vs. 7.6%, P = 0.661). Comparing patients with soft pancreas and small pancreatic duct, CR-POPF rates were not statistically different among the groups [16.0% (Rob-BM) vs. 10.5% (Rob-IS), P = 0.055, and 10.5% (Rob-IS) vs. 10.1% (Lap-IS), P = 0.543]. In the multivariable analysis for risk factors of POPF, soft pancreatic textures and periampullary pathology other than pancreatic cancer were found to be risk factors.

Conclusion: POPF rates after MIPD were not different according to the PJ methods of BM and IS when performed by skilled surgeons.