Antibiotic therapy for prophylaxis against infection of pancreatic necrosis in acute pancreatitis

Background Pancreatic necrosis may complicate severe acute pancreatitis, and is detectable by computed tomography (CT). If it becomes infected mortality increases, but the use of prophylactic antibiotics raises concerns about antibiotic resistance and fungal infection. Objectives To determine the efficacy and safety of prophylactic antibiotics in acute pancreatitis complicated by CT proven pancreatic necrosis. Search methods Searches were updated in November 2008, in The Cochrane Library (Issue 2, 2008),MEDLINE, EMBASE, and CINAHL. Conference proceedings and references from found articles were also searched. Selection criteria Randomised controlled trials (RCTs) comparing antibiotics versus placebo in acute pancreatitis with CT proven necrosis. Data collection and analysis Primary outcomes were mortality and pancreatic infection rates. Secondary end-points included non pancreatic infection, all sites infection, operative rates, fungal infections, and antibiotic resistance. Subgroup analyses were performed for antibiotic regimen (betalactam, quinolone, and imipenem). Main results Seven evaluable studies randomised 404 patients. There was no statistically significant effect on reduction of mortality with therapy: 8.4% versus controls 14.4%, and infected pancreatic necrosis rates: 19.7% versus controls 24.4%. Non-pancreatic infection rates and the incidence of overall infections were not significantly reduced with antibiotics: 23.7% versus 36%; 37.5% versus 51.9% respectively. Operative treatment and fungal infections were not significantly different. Insufficient data were provided concerning antibiotic resistance. With beta-lactam antibiotic prophylaxis there was less mortality (9.4% treatment, 15% controls), and less infected pancreatic necrosis (16.8% treatment group, 24.2% controls) but this was not statistically significant. The incidence of non-pancreatic infections was non significantly different (21% versus 32.5%), as was the incidence of overall infections (34.4% versus 52.8%), and operative treatment rates. No significant differences were seen with quinolone plus imidazole in any of the end points measured. Imipenem on its own showed no difference in the incidence of mortality, but there was a significant reduction in the rate of pancreatic infection (p=0.02; RR 0.34, 95% CI 0.13 to 0.84)