HOME
Return to Table of Contents
PROPHYLACTIC ANTIBIOTIC COVERAGE FOR ENDOSCOPIC PROCEDURES – ARE GASTROENTEROLOGISTS COMPLIANT OR CONFUSED?
M MacMillan, D Leddin
Division of Gastroenterology. Dalhousie University. Halifax, Nova Scotia
Decisions frequently need to be made with regard to antibiotic coverage for endoscopic procedures. We wished to assess whether practicing Gastroenterologists used the recommendations of either the Canadian Association of Gastroenterology (CAG) or the American Society for Gastrointestinal Endoscopy (ASGE) in their clinical practice.
METHODS: We developed 20 short clinical case scenarios based on the guidelines of the CAG or ASGE. They were designed to test physician utilization of prophylactic antibiotics. The cases consisted of five colonoscopy, 12 upper endoscopy and 3 ERCP scenarios where antibiotics may, may not, or may possibly have been indicated. Three variables were altered in the scenarios: the procedure, the intervention and the underlying risk factor for infection. The correct utilization of antibiotics in the scenarios was described in either the CAG or ASGE guidelines. We distributed 30 surveys to Gastroenterologists attending the Atlantic Provinces Gastroenterology Conference.
RESULTS: Ten surveys were fully completed, 2 were 80% completed. There were eight community based and 2 hospital based Gastroenterologists with a mean age of 53.5. All respondents were male. Overall, the median correct response rate of the scenarios was 45% with a range of correct answer by scenario from 0 (no one answering correctly) to 91%. answering correctly. The upper endoscopy median correct response rate was 46% (range 0 – 91%). Colonoscopy median correct response rate was 64% (range 0 – 83%). ERCP median correct response rate was 30% (range 30 – 55%). We further evaluated the data based upon an intervention being completed. Upper endoscopy with and without intervention had a median correct response rate of 50% (range 0 – 91%) and 50% (range 8 – 91%), respectively. Colonoscopy with and without intervention had a median correct response rate of 32% (range 0 – 64%) and 32% (range 27 – 83%), respectively.
The sample size was not large enough to permit sub analysis. The low correct response rate regardless of whether the procedure was upper endoscopy, colonoscopy, or ERCP indicates that the clinical application of the guidelines is deficient. Patients who need antibiotics may not get them and those who do not may be exposed to unnecessary antibiotics with subsequent risk of antibiotic related complications. There is a need either for guidelines, which can more easily be translated into clinical practice, or the development of a teaching tool, which will allow clinicians to check appropriateness of use in clinical practice.