Abstract
BACKGROUND: Current guidelines recommend ethambutol as a first-line agent for treating Mycobacterium avium complex pulmonary disease (MAC-PD) but can induce optic neuropathy. Previous studies suggest that low-dose ethambutol may reduce the incidence of optic neuropathy without compromising its effectiveness, but have not adequately adjusted for risk factors for optic neuropathy. We investigated the association between low-dose ethambutol therapy and optic neuropathy in patients with MAC-PD. METHODS: This retrospective cohort study included patients treated for MAC-PD at Kameda Medical Center between April 2003 and July 2024. Patients were categorized into low-dose (<12.5 mg/kg/day) and high-dose (≥12.5 mg/kg/day) ethambutol groups. The primary outcome was the incidence of optic neuropathy. Secondary outcomes included failure of negative culture conversion and macrolide resistance. Propensity score-based overlap weighting was used to adjust for patient characteristics and outcomes were compared using datasets generated through bootstrap resampling. RESULTS: Of 223 patients, 79 received low-dose (10.7 mg/kg/day) and 144 received high-dose ethambutol (15.4 mg/kg/day). None of the patients in the low-dose group and 4.8% in the high-dose group developed optic neuropathy. After adjustment, the risk of optic neuropathy was significantly lower in the low-dose ethambutol group (risk difference: -17.1%, 95% CI: -32.9% to -5.4%), but the risk difference for failure of negative culture conversion and macrolide resistance did not differ significantly between groups (-20.0%, 95% CI: -45.6% to 2.5%; and -4.9%, 95% CI: -13.5% to 0.0%, respectively). CONCLUSIONS: Low-dose ethambutol therapy may reduce the risk of optic neuropathy without compromising treatment outcomes, offering a safer option for MAC-PD.