Quinolone Ear Drops and Achilles Tendon Rupture

Abstract

Abstract Background Delayed eardrum healing has been observed in the ear opposite to the ear treated with otic quinolones (OQ) in rats. Case reports describe tendinopathies after OQ treatment, suggesting adverse systemic effects. Methods We studied patients aged 19 to 64 years with diagnosis of otitis externa or media in private insurance between 2005 and 2015. We compared OQ treatment against otic neomycin, oral amoxicillin, or azithromycin. Outcomes included Achilles tendon rupture (ATR), Achilles tendinitis (AT), and all-type tendon rupture (ATTR). We applied an active comparator, new-user design with 1-year look-back and ceased follow-up at initiation of systemic steroids or oral quinolones, external injury, hospitalization, and after 35 days. We used trimmed stabilized inverse probability of treatment weights to balance comparison groups in a survival framework. Negative outcomes (clavicle fractures or sports injuries) were examined to rule out differences from varied physical activity (unmeasured confounding). Results We examined 1 501 009 treated otitis episodes. Hazard ratios (HR) for OQ exposure associated with ATR were 4.49 (95% confidence interval [CI], 1.83–11.02), AT 1.04 (95% CI, 0.73–1.50), and ATTR 1.71 (95% CI, 1.21–2.41). Weighted risk differences (RD) per 100 000 episodes for OQ exposure were ATR 7.80 (95% CI, 0.72–14.89), AT 1.01 (95% CI, −12.80 to 14.81), and ATTR 18.57 (95% CI, 3.60–33.53). Corresponding HRs for clavicle fractures and sports injuries were HR,1.71 (95% CI, 0.55–5.27) and HR,1.45 (95% CI, 0.64–3.30), suggesting limited residual confounding. Conclusions OQ exposure may lead to systemic consequences. Clinicians should consider this potential risk and counsel patients accordingly. Risk factors and mechanisms for this rare, adverse effect deserve further evaluation. Mechanistic and other clinical studies are warranted to corroborate this finding.