Emergence of fluoroquinolones as the predominant risk factor for Clostridium difficile-associated diarrhea: a cohort study during an epidemic in Quebec.

Since 2002, an epidemic of Clostridium difficile-associated-diarrhea (CDAD) associated with a high case-fatality rate has involved >30 hospitals in the province of Quebec, Canada. In 2003, a total of 55% of patients with CDAD at our hospital had received fluoroquinolones in the preceding 2 months. It has been suggested that massive use of proton pump inhibitors might have facilitated this epidemic. To delineate the risk of CDAD associated with specific classes of antibiotics and whether this is modulated by concomitant use of proton pump inhibitors and other drugs altering gastric acidity or gastrointestinal motility, we conducted a retrospective cohort study of patients hospitalized in a teaching hospital in Sherbrooke, Canada, during the period of January 2003 through June 2004. We obtained data on 7421 episodes of care corresponding to 5619 individuals. Patients were observed until they either developed CDAD or died or for 60 days after discharge from the hospital. Adjusted hazard ratios (AHRs) were calculated using Cox regression. CDAD occurred in 293 patients. Fluoroquinolones were the antibiotics most strongly associated with CDAD (AHR, 3.44; 95% confidence interval [CI], 2.65-4.47). Almost one-fourth of all inpatients received quinolones, for which the population-attributable fraction of CDAD was 35.9%. All 3 generations of cephalosporins, macrolides, clindamycin, and intravenous beta-lactam/beta-lactamase inhibitors were intermediate-risk antibiotics, with similar AHRs (1.56-1.89). Proton pump inhibitors (AHR, 1.00, 95% CI, 0.79-1.28) were not associated with CDAD. Administration of fluoroquinolones emerged as the most important risk factor for CDAD in Quebec during an epidemic caused by a hypervirulent strain of C. difficile.