Development of novel criteria for drain removal after pancreaticogastrostomy in pancreaticoduodenectomy: a retrospective analysis and validation study at a single institution

Postoperative pancreatic fistula (POPF) remains a major complication following pancreaticoduodenectomy (PD). Although pancreaticogastrostomy (PG) has been suggested to be associated with a lower risk of POPF in some studies, optimal criteria for postoperative drain management in PG cases remain unclear. This single-institution retrospective study analyzed 195 consecutive PD cases involving a soft pancreas and PG. New criteria for safe drain removal were developed based on independent risk factors identified through a multivariate analysis. A prospective validation study was then conducted in 70 consecutive soft pancreas cases to assess the safety and efficacy of these criteria. In the retrospective cohort, white blood cell count (WBC) > 8200/µL on postoperative day (POD) 5, drain fluid amylase (DFA) > 160 U/L on POD 5, and delayed nasogastric (NG) tube removal were identified as independent risk factors. The criteria for drain removal were defined as: WBC ≤ 8200/µL, DFA ≤ 160 U/L, and NG tube removal by POD5. In the validation cohort, 42 out of 70 patients met the criteria; none developed Clavien–Dindo grade IIIa or higher complications after drain removal. The incidence of delayed gastric emptying (DGE) was significantly lower in the post-validation cohort than in the pre-validation cohort (p < 0.01). In patients undergoing PG with a soft pancreas, application of the new criteria enables safe early drain removal and may help reduce the incidence of DGE.