Abstract
Objective. Risk factors for complications after endoscopic retrograde cholangiopancreatography (ERCP) with emphasis on the potential advantage of the use of prophylactic antibiotics were studied in a national population-based study cohort. Materials and methods. All ERCP procedures registered in the Swedish Registry of Gallstone Surgery and ERCP (GallRiks) between May 2005 and June 2013 were analyzed. Patients with ongoing antibiotic treatment, incomplete registration or those who had not undergone an index ERCP were excluded. Risk factors for adverse events were analyzed. Results. Data from 47,950 ERCPs were collected, but after applying the exclusion criteria, 31,188 examinations were analyzed. In the group receiving prophylactic antibiotics, the postoperative adverse event rate was 11.6% compared with 14.2% in the group without antibiotics. The odds ratio (OR) for the risk of postoperative adverse events in patients receiving prophylactic antibiotics was 0.74 (95% confidence interval [CI]: 0.69–0.79). When analyzing a subgroup of 21,893 ERCPs for the three most common indications (common bile duct stones, malignancy, and obstructive jaundice), the beneficial effect of prophylactic antibiotics on adverse events remained (OR = 0.76; 95% CI: 0.70–0.82). Further, in the subgroup of patients with obstructive jaundice, the administration of prophylactic antibiotics had a beneficial effect on septic complications (OR = 0.76; 95% CI: 0.58–0.97). Conclusion. The risk of adverse events after ERCP is reduced 26% if antibiotics are given prophylactically during ERCP investigations, as suggested by data gained from this national population-based study. However, in absolute terms, the reduction in adverse events by prophylactic antibiotics is modest (2.6%).
Acknowledgments
The authors thank Professor Greger Lindberg for his guidance on the statistical analysis. Different parts of this study were presented during the 19th United European Gastroenterology Week, Stockholm, Sweden. This research was supported by unconditional research funds at the Karolinska Institutet and from Futurum, Jönköpings Läns Landsting.
Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.