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.

Endoscopic retrograde cholangiopancreatography (ERCP) involves cannulation of the ampulla of Vater and has diagnostic as well as therapeutic capabilities, but the number of non-therapeutic ERCPs is decreasing with time. 1 Endoscopic sphincterotomy, stone extraction and stenting are not without complications. The most widely recognised of these include bleeding, which occurs in 0.7 -2% of patients, perforation (0.3 -0.6%), pancreatitis (7%), cholangitis (1%) and cholecystitis (0.2 -0.5%). Procedure-related mortality is approximately 0.2%. 2 Review of international guidelines regarding the use of prophylactic antibiotics with ERCP shows that routine use of antimicrobials is recommended for biliary obstruction and pancreatic pseudocysts. However, several studies, including a meta-analysis, fail to show any benefit. [3][4][5][6] We set out to assess the current antibiotic prescribing practice among South African endoscopists who perform ERCPs, and then review international guidelines and relevant studies. MethodsOur audit of South African endoscopists who perform ERCPs took the form of a questionnaire. This was distributed at the Hepato-Pancreatico-Biliary Association of South Africa Congress held during October 2007 in Johannesburg, and was also sent to all members of the South African Gastro-Enterology Society via email. The questionnaire was anonymous. Endoscopists were questioned regarding their years of experience, the monthly volume of ERCPs they perform, and their indications for antibiotic prophylaxis (for diagnostic biliary ERCP, diagnostic pancreatic ERCP, therapeutic biliary ERCP and therapeutic pancreatic ERCP). Respondents were also asked to indicate their antibiotic of preference and the number of doses administered. The results were then tabulated for comparison, and the chisquared test was used to calculate p-values. A p-value of 0.05 was considered significant.A Pubmed search was performed from 1978 to March 2008 using the search terms Cholangiopancreatography-Endoscopic-Retrograde Antibiotic-Prophylaxis, random* or control* or blind* or meta-analys*, all subheadings. An Internet search was also performed to identify recommendations from various international gastrointestinal societies. ResultsThirty-nine endoscopists (22 surgeons, 16 medical gastroenterologists and 1 radiologist) responded to our questionnaire. Most had more than 6 years of experience (30/39) and performed more than 10 ERCPs per month (22/39). Approximately half of the endoscopists (19/39) were aware of ERCP antibiotic protocols, either the American Society of Gastro-Enterology (ASGE) or UK National Health Service (NHS) recommendations. The results are depicted in Table I. 'Always' implied that the endoscopist used antibiotic prophylaxis with each patient, 'selected' implied specific indications, and 'never' implied no use of antibiotic prophylaxis.