METHODS: Using a UK general practice database (1995-2015), we studied a matched cohort of adults without prior MRSA or C.difficile. Patients with incident penicillin allergy were matched with up to five penicillin users without allergy by age, sex, and index date. We calculated relative risks (RR) for the association of penicillin allergy with incident MRSA and C.difficile, adjusting for potential confounders. We also examined beta-lactam alternative antibiotic use and whether it was a mediator for MRSA/C.difficile incidence.
RESULTS: Among 64,141 penicillin allergy patients and 237,258 matched comparators, 1,345 developed MRSA and 1,688 developed C.difficile over 6 years of mean follow-up. The adjusted RRs among penicillin allergy patients were 1.62 (95%CI 1.42-1.85) for MRSA and 1.27 (95%CI 1.13-1.43) for C.difficile. The adjusted RRs for antibiotic use among penicillin allergy patients were 4.08 (95%CI 4.05-4.10) for macrolides, 3.73 (95%CI 3.51-3.97) for clindamycin, and 2.13 (95%CI 2.10-2.16) for fluoroquinolones. Increased beta-lactam alternative antibiotic use accounted for 53% of the increased MRSA risk and 25% of the increased C.difficile risk.
CONCLUSIONS: In this population-based cohort study, documented penicillin allergy was associated with an increased risk of MRSA and C.difficile that was mediated by the increased use of beta-lactam alternative antibiotics. Systematically addressing penicillin allergies may substantially reduce MRSA and C.difficile incidence.