Efficacy and safety of fluoroquinolone-containing regimens in treating pulmonaryMycobacterium aviumcomplex disease: A propensity score analysis

Background Although combination therapy using clarithromycin, rifampicin, and ethambutol is recommended for patients with pulmonaryMycobacterium aviumcomplex (MAC) disease, some patients do not tolerate it because of adverse effects or underlying diseases. The efficacy and safety of fluoroquinolone-containing combination regimens as an alternative remain uncertain. This study aimed to compare the efficacy and safety of fluoroquinolone-containing regimens with those of the standard regimens for treating pulmonary MAC disease. Methods We retrospectively included consecutive MAC patients who were treated in our hospital between January 2011 and May 2019. Patients treated with fluoroquinolone-containing regimens who had relapsed after treatment with standard regimens were excluded. A propensity score analysis was conducted to reduce selection bias, and the proportions of clinical improvement, defined by chest imaging findings and sputum conversion, were compared between the fluoroquinolone-containing regimen and standard regimen groups. Results We analyzed 28 patients who received fluoroquinolone-containing regimens and 46 who received the standard regimen. Fluoroquinolone-containing regimens were more likely selected for patients with cavitary lesions, diabetes mellitus, culture negativity, a low daily physical activity level, a decreased lymphocyte count and an increased CRP level. The propensity score was calculated using these variables (C-statistic of the area under the receiver operating characteristic curve of the propensity score: 0.807, p < 0.0001). The fluoroquinolone-containing regimens were significantly inferior to the standard regimen in clinical improvements (p = 0.002, Log-rank test) in the univariate analysis, but the significance was lost after adjusting for the propensity score (HR 0.553, 95% CI 0.285-1.074, p = 0.080). Six (21%) patients in the fluoroquinolone-containing regimen group and ten (22%) patients in the standard regimen group experienced low-grade adverse effects. Conclusions There was no significant difference in clinical improvement between these regimens after propensity score adjustment. A large-scale prospective study is required to validate these results.