Early prophylactic antibiotics for severe acute pancreatitis: A population-based cohort study using a nationwide database in Japan.

Abstract

Previous studies evaluating the clinical benefits of prophylactic antibiotics for severe acute pancreatitis (SAP) have generated inconsistent results due to heterogeneities among the study settings. We determined if early prophylactic antibiotics improved the outcomes of SAP patients using a study designed to overcome these previous methodological weaknesses. We conducted a retrospective cohort study of SAP patients discharged between July 2010 and March 2016, using a Japanese nationwide inpatient database. We divided patients into those with early prophylactic carbapenem use (prophylaxis group) and those without prophylaxis (control group). The primary outcome was in-hospital mortality and the secondary outcomes included oral vancomycin use and others. We identified a total of 3354 eligible patients, including 2493 in the prophylaxis group and 861 in the control group. The overall in-hospital mortality was 12.8%. Prophylactic antibiotics were not significantly associated reduced in-hospital mortality according to Cox regression analysis (hazard ratio (HR), 0.88; 95% confidence interval (CI), 0.62-1.23) or instrumental variable analysis (risk difference,-1.2%; 95% CI,-9.8%-7.4%). However, prophylactic antibiotic use was significantly associated with oral vancomycin use during hospitalization in a competing-risk model (subdistribution HR, 1.91; 95% CI, 1.02-3.56). The present study suggests that routine early prophylactic antibiotic use have no significant clinical benefit in SAP patients but may increase the risk of hospital-acquired infections.