Pancreatic Fistula after Pancreatectomy – What Else Can We Consider Other than Drain Amylase in the Early Postoperative Period?

Introduction: Drain amylase is often relied on to determine the timing of drain removal after pancreatectomy. Although early drain removal is becoming more common, other predictors of clinically relevant postoperative pancreatic fistula (CR-POPF) when drain amylase is low in the early postoperative period have yet to be established. Methods: Patients who underwent distal pancreatectomy (DP) or pancreatoduodenectomy (PD) between 2013-2020 were queried from a prospectively maintained NSQIP database at a single, high-volume institution. Those whose drain amylase on postoperative day 1 (POD1) was less than three times the laboratory upper limit for normal were selected. CR-POPF was defined as grade B or C fistulas according to the 2016 International Study Group on Pancreatic Surgery guidelines. Results: 499 of 1549 (32.2%) of DPs and PDs demonstrated low drain amylase on POD1. Of these, 17 (3.4%) developed CR-POPF (11 grade B, 6 grade C). Surgery type (DP vs. PD) was not associated with CR-POPF (3.6% vs 3.5%, p=0.95). Age, BMI, and other historical medical conditions including diabetes, ascites, and dialysis dependence also showed no association. Patients who were male (5.7% vs. 0.8%, RR=7.1, p<0.01), had disseminated cancer (13.3% vs 3.1%, RR=4.3, p=0.03), and soft pancreas gland texture (7.6% vs. 1.6% intermediate vs. 1.8% hard, p=0.014) were more likely to develop CR-POPF. Blood transfusion also correlated with CR-POPF (6.3% vs. 2.4%, RR=2.6, p=0.04). Conclusions: Patient and perioperative factors other than drain amylase, such as blood transfusion, may be predictive of CR-POPF in the early postoperative period and should be considered in postsurgical care.