Risk of sensorineural hearing loss with macrolide antibiotics: A nested case-control study.

OBJECTIVES To determine the association between a diagnosis of sensorineural hearing loss (SNHL) and the prescription of a macrolide antibiotic. STUDY DESIGN Retrospective nested case-control study. METHODS From the LifeLink (IMS, Danbury, CT) health claims database, we randomly selected a cohort of subjects 15 to 60 years old from 2006 to 2014. Cases were identified as patients diagnosed with SNHL, each matched by age and calendar time to 10 controls selected from the same cohort. All macrolide prescriptions (erythromycin, azithromycin, clarithromycin, and telithromycin) were identified, and statistical comparison of usage was compared between cases and controls. Amoxicillin and fluoroquinolone antibiotics were used as positive controls to further investigate confounding by infection. Albuterol was used as a negative control because this is a drug class not expected to be associated with SNHL or with a confounding condition potentially causing SNHL. RESULTS From a cohort of 6,110,723 subjects, we identified 5,989 cases of SNHL and 59,890 corresponding controls. The rate ratio for one prescription of a macrolide was 1.36 (95% confidence inteval [CI]: 1.24-1.49) and for multiple prescriptions was 1.66 (95% CI: 1.42-1.94). Similar rate ratios were observed with multiple prescriptions of amoxicillin and fluoroquinolones. CONCLUSION A significant association between SNHL and macrolide use was likely due to confounding by indication for antibiotic treatment because the risk was also observed with fluoroquinolones and amoxicillin, antibiotics with no known ototoxic potential. Therefore, there does not appear to be an increased risk of SNHL in patients treated with macrolide antibiotics. LEVEL OF EVIDENCE 3b. Laryngoscope, 127:229-232, 2017.