Antibiotic prophylaxis for patients undergoing elective endoscopic retrograde cholangiopancreatography

Abstract

The use of prophylactic antibiotics before endoscopic retrograde cholangiopancreatography (ERCP) is recommended by all major international gastroenterological societies, especially in the presence of an obstructed biliary system. Their use is intended to decrease or eliminate the incidence of complications following the procedure, namely cholangitis, cholecystitis, septicaemia, and pancreatitis. To assess the benefits and harms of antibiotics before elective ERCP in patients without evidence of acute or chronic cholecystitis, or acute or chronic cholangitis, or severe acute pancreatitis. We searched The Cochrane Hepato-Biliary Group Controlled Trials Register, the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library, MEDLINE, EMBASE, Science Citation Index Expanded, and LILACS until March 2010. Relevant medical and surgical international conference proceedings were also searched. Only randomised clinical trials were included in the analyses, irrespective of blinding, language, or publication status. Participants were patients that underwent elective ERCP that were not on antibiotics, without evidence of acute or chronic cholecystitis, cholangitis, or severe acute pancreatitis before the procedure. We compared patients that received prophylactic antibiotics before the procedure with patients that were given placebo or no intervention before the procedure. The review was conducted according to the recommendations of The Cochrane Collaboration as well as the Cochrane Hepato-Biliary Group. Review Manager 5 was used employing fixed-effect and random-effects models meta-analyses. Nine randomised clinical trials (1573 patients) were included in the analyses. The majority of the trials had risks of bias. When all patients providing data for a certain outcome were included, the fixed-effect meta-analyses significantly favoured the use of prophylactic antibiotics in preventing cholangitis (relative risk (RR) 0.54, 95% CI 0.33 to 0.91), septicaemia (RR 0.35, 95% CI 0.11 to 1.11), bacteriaemia (RR 0.50, 95% CI 0.33 to 0.78), and pancreatitis (RR 0.54, 95% CI 0.29 to 1.00). In random-effects meta-analyses, only the effect on bacteriaemia remained significant. Overall mortality was not reduced (RR 1.33, 95% CI 0.32 to 5.44). If one selects patients in whom the ERCP resolved the biliary obstruction at the first procedure, there seem to be no significant benefit in using prophylactic antibiotics to prevent cholangitis (RR 0.98, 95% CI 0.35 to 2.69, only three trials). Prophylactic antibiotics reduce bacteriaemia and seem to prevent cholangitis and septicaemia in patients undergoing elective ERCP. In the subgroup of patients with uncomplicated ERCP, the effect of antibiotics may be less evident. Further research is required to determine whether antibiotics can be given during or after an ERCP if it becomes apparent that biliary obstruction cannot be relieved during that procedure.