Effect of antibiotic choice on the development of postoperative pancreatic fistula: a NSQIP study

Abstract

Choice of antibiotic surgical site infection for pancreatic surgery is variable, with no consensus on their optimal choice of prophylactic antibiotics. We aim to assess the influence of antibiotic choice on the development of postoperative pancreatic fistula (POPF). The American College of Surgeons - National Surgical Quality Improvement Program Pancreatectomy Targeted Participant Use Data File 2016 was queried to identify patients who underwent distal pancreatectomy (DP) or pancreatoduodenectomy (PD). The incidence of POPF was compared by genre of antibiotic given: 1st generation cephalosporin (Abx1) vs 2nd or 3rd generation cephalosporin or broad spectrum antibiotics (Abx2). A total of 1762 (32.5%) and 3659 (67.5%) patients underwent DP and PD, respectively. Abx1 was given intraoperatively as prophylaxis to 2060 (38%) patients. The rate of fistula formation was 18.4% (n=995). The rate of fistula formation was significantly different by the extent of antibiotic coverage; 19.9% (n=409) and 17.4% (n=586) for the Abx1 and Abx2 groups (P=0.026), respectively. The choice of antibiotic did not significantly change the rate of fistula formation in patients who underwent DP (P>0.05). However, a significant change was noted in PD patients; 20.2% (n=253) and 17.2% (n=414), P=0.023, for Abx1 and Abx2, respectively. The choice of antibiotic coverage did not change the severity of fistulae (P=0.77) for either DP or PD. Prophylactic antibiotic choice affects the rate of fistula formation following PD but not DP. Further prospective studies assessing the specific bacteria and its role in fistula formation is warranted, and indeed underway in the ACS-AHPBA trial.