Towards a more precise definition of postoperative pancreatic fistula (POPF): should drain position be considered alongside drain amylase levels after pancreaticoduodenectomy? - a retrospective cohort study.

BACKGROUND: The International Study Group of Pancreatic Fistula (ISGPF) defines postoperative pancreatic fistula (POPF) based on drain fluid amylase (DFA) levels. However, drain malposition can compromise the reliability of DFA in detecting POPF. This may trigger premature drain removal, delayed recognition of clinically relevant POPF (CR-POPF), and escalation of morbidity. We evaluated how drain position modifies the diagnostic performance of DFA after PD. METHOD: A retrospective analysis of 334 patients who underwent PD (June 2019-December 2023). Drain position was assessed using postoperative imaging and DFA levels were measured on postoperative day3. The primary aim was the incidence of drain malposition, while secondary aims included its impact on the diagnosis of CR-POPF, SFC. RESULT: Among 334 patients, the drain was well-positioned in 255 patients (76.4%) and malpositioned in 79 patients (23.6%). High DFA levels were strongly associated with proper drain positioning (152/164, 92.7%). However, low/normal DFA levels often reflected drain malposition, with 67/170 patients(39.4%) incorrectly classified, leading to potential false-negative results. Malpositioned drains patients had a significantly higher SFC(p < 0.001) and longer hospital stay (p < 0.001). Multivariable analysis confirmed drain malposition as an independent predictor of CR-POPF (OR 4.74, 95% CI: 1.86-12.06, p = 0.001), small pancreatic duct(OR 10.59, 95% CI: 3.92-28.62,p < 0.001) and low preoperative albumin (OR 0.07, 95% CI: 0.02-0.24, p < 0.001). CONCLUSION: Drain malposition significantly impacts DFA reliability with false-negative diagnosis. Revising the ISGPS definition to incorporate drain positioning alongside DFA levels may improve diagnostic accuracy in defining POPF.