Risk factors for clinically relevant postoperative pancreatic fistula after pancreatectomy: a retrospective observational study

Background Clinically relevant postoperative pancreatic fistula (CR-POPF) is a common complication following pancreatectomy. However, the pathophysiological mechanisms underlying CR-POPF remain unclear. This study aimed to identify risk factors for CR-POPF after pancreatoduodenectomy or distal pancreatectomy in the hope of developing preventive strategies and improving postoperative patient management. Methods We retrospectively identified 180 patients who underwent pancreatectomy at Huzhou Central Hospital between January 2019 and August 2023. After excluding two cases that underwent total pancreatectomy, one with a history of pancreatectomy, two with missing information on the amylase level in drainage fluid, and one that underwent distal pancreatectomy with combined adrenal, ureteral, and splenic resection, 174 patients were enrolled. Univariate and multivariate logistic regression analyses were performed to identify independent risk factors for CR-POPF. Predictive performance was evaluated using ROC curves. Results Thirty-five of the 174 patients (20.1%) developed CR-POPF. Multivariate analysis revealed that a positive drainage fluid culture (P = 0.005), an elevated perioperative white blood cell (WBC) count (P = 0.06), and an elevated perioperative C-reactive protein (CRP) level (P = 0.001) were independent risk factors for CR-POPF. For pancreaticoduodenectomy, univariate analysis identified a positive drainage fluid culture (P = 0.019), lower serum calcium on postoperative day (POD) 1 (P = 0.009), elevated perioperative WBC count (P < 0.001) and CRP (P = 0.014), and a prolonged operation time (P = 0.001) to be risk factors. Multivariate analysis confirmed a serum calcium level of < 2.0 mmol/L on POD 1, a perioperative WBC count ≥ 10.5×10⁹/L, and an operation time of ≥ 352 min to be independent risk factors. For distal pancreatectomy, univariate analysis identified a lower preoperative blood glucose level (P = 0.034) as a risk factor, but no independent factors emerged in multivariate analysis. Conclusions A positive drainage fluid culture, a perioperative WBC count ≥ 10.7×10⁹/L, and a perioperative CRP ≥ 52 mg/L were independent risk factors for CR-POPF. For pancreaticoduodenectomy, serum calcium < 2.0 mmol/L on POD 1, a perioperative WBC count ≥ 10.5 ×10 ⁹ /L, and operation time ≥ 352 min were independent risk factors for CR-POPF. An elevated WBC count on POD 5/6 had predictive value for CR-POPF. These findings underscore the importance of postoperative monitoring for intra-abdominal infections and inflammatory markers to mitigate the risk of CR-POPF.