Association of Outpatient Oral Macrolide Use With Sensorineural Hearing Loss in Children, Adolescents, and Young Adults

IMPORTANCE Prior publications have reported the sporadic development of sensorineural hearing loss (SNHL) after intravenous or high-dose macrolide therapy for adults with comorbid conditions, but investigations of the auditory effect of oral outpatient dosing for children, adolescents, and young adults have been limited. OBJECTIVE To determine whether broad-based outpatient use of oral macrolide therapy is associated with increased risk of pediatric SNHL through nationally representative analyses. DESIGN, SETTING, AND PARTICIPANTS A retrospective case-control study of 875 matched pairs of children, adolescents, and young adults was performed, matching on age, sex, and the time elapsed since prescription date. All eligible pediatric patients were included, with matched control participants from the TRICARE US military health insurance system who were evaluated between October 1, 2009, and September 30, 2014. EXPOSURES Oral outpatient macrolide treatment compared with penicillin use among pediatric patients. MAIN OUTCOMES AND MEASURES The clinical outcome of interest was SNHL in children, adolescents, and young adults. Multivariable conditional logistic regression was used to compare the risk of prior macrolide exposure with penicillin exposure, adjusted for other risk factors and potential confounders. Four time frames between exposure and diagnosis were additionally assessed. RESULTS There were 875 eligible matched pairs of children, adolescents, and young adults included. The mean (SD) age of the participants was 5.7 (4.9) years; 1082 participants were male (62%), 58 were Asian (3%), 254 were Black (15%), 1152 were White (66%), and 286 were of Native American and other (no further breakdown was available in the TRICARE database) race and ethnicity (16%). In multivariable analysis, participants who had SNHL had increased odds of having received a macrolide prescription compared with a penicillin prescription when all time frames from exposure were included (adjusted odds ratio, 1.31; 95% CI, 1.05-1.64). There were significantly higher odds of macrolide exposure than penicillin exposure when diagnosis and testing occurred more than 180 days after antibiotic exposure (adjusted odds ratio, 1.79; 95% CI, 1.23-2.60). CONCLUSIONS AND RELEVANCE In this case-control study of a nationally representative patient population, findings suggest that children, adolescents, and young adults with SNHL had increased odds of outpatient oral macrolide use compared with penicillin use, particularly when having received a diagnosis more than 180 days after exposure. Further study of the association of macrolides with SNHL in children, adolescents, and young adults is warranted.