Factors for the Initial Regimen Modification in Mycobacterium Avium Complex Pulmonary Disease: A Single-center Retrospective Cohort Study

Abstract Rationale: Current guidelines recommend a multidrug regimen including macrolides, ethambutol, and rifamycin for the treatment of Mycobacterium avium complex pulmonary disease. Although antibiotic treatment requires a prolonged duration, this regimen is poorly tolerated and frequently results in adverse events, leading to regimen change or discontinuation. However, the factors that may be associated with tolerability to multidrug treatments remain unclear. Therefore, we aimed to investigate the patient- and treatment-related factors associated with tolerance to multidrug treatment. Methods: This single-center retrospective cohort study included patients with MAC-PD who underwent guideline-based multidrug therapy at the Kameda Medical Center between November 2000 and July 2024. As several missing values were observed, we used a multiple imputation with R (var.4.1.0, mice package, m = 50, iteration = 50) to complement the missing values. A logistic model was applied to the imputed data, and the results were integrated using Rubin's rule. The primary outcome was the proportion of patients with initial regimen because of adverse events. Regimen modification was defined as the interruption or dose reduction of ≥ 1 drugs and failure to return to the initial regimen after > 28 days. We estimated the adjusted odds ratios (aORs) and 95% confidence intervals (CIs) of patient characteristics for initial regimen modification using multivariable logistic regression analysis. Results: We evaluated 225 patients with MAC-PD treated with guideline-based multidrug therapy. The cohort had a mean age of 69.4 years at the start of treatment and was mostly women (75.9%) patients. Ninety-seven patients required initial regimen modification. A total of 161 (71.6%) and 64 (28.4%) patients received clarithromycin- and azithromycin-based regimens, respectively. The aORs for initial regimen modification were as follows; age 1.05 (95% CI; 1.01-1.10), women 0.77 (95% CI; 0.34-1.74), body mass index 1.06 (95% CI; 0.96-1.18), smear-positive status 0.85 (95% CI; 0.43-1.68), azithromycin-based regimen 0.43 (95% CI; 0.21-0.89), aminoglycoside use 0.39 (95% CI; 0.11-1.44), and high-dose ethambutol (≥ 12.5 mg/kg/day) 2.84 (95% CI; 1.38-5.83). Conclusion: An azithromycin-based regimen was less likely to undergo initial modification than a clarithromycin-based regimen, whereas older age and higher doses of ethambutol were more frequently associated with regimen changes or discontinuation. This study provided insights into the tolerability of specific multidrug antibiotic regimens in clinical settings, particularly in patients undergoing MAC-PD.