Efficacy and safety of inhaled versus injectable aminoglycosides in Mycobacterium avium complex pulmonary disease: a network meta-analysis

Abstract Background The prevalence of Mycobacterium avium complex (MAC) pulmonary disease shows a sustained upward trend, with treatment challenged by intrinsic drug resistance and toxicity of conventional therapies. Aminoglycosides (AG) are key agents for MAC management, but injectable AGs exhibit poor pulmonary targeting and severe systemic toxicity, particularly ototoxicity. Inhaled AG formulations enhance local lung delivery and reduce systemic exposure. This network meta-analysis (NMA) compared efficacy and safety of inhaled versus injectable AG regimens for microbiologically verified MAC pulmonary disease. Methods PubMed, CNKI, and Wanfang databases were searched from inception to January 2026 for randomized controlled trials (RCTs) and cohort studies. The primary endpoint was 6-month sustained sputum culture conversion; secondary endpoints included ototoxicity, nephrotoxicity, and respiratory adverse events. Bias risk and evidence quality were assessed via RoB 2.0/ROBINS-I and GRADE, respectively. A frequentist fixed-effect network meta-analysis (NMA) was conducted (PROSPERO: CRD420251159363). Results Four studies (three RCTs and one cohort study, total n = 517) were included, with no statistical heterogeneity ( I² = 0%). Inhaled aminoglycosides significantly improved 6-month sustained sputum culture conversion compared with standard therapy (OR = 4.27, 95% CI: 2.38–7.66; moderate-quality evidence per GRADE). Injectable aminoglycosides showed no significant benefit (OR = 1.88, 95% CI: 0.37–9.54; very low-quality evidence). No ototoxicity or nephrotoxicity was observed with inhaled aminoglycosides in the three RCTs (0/302), whereas the injectable aminoglycoside group (single cohort, n = 19) reported a 78.9% ototoxicity rate. Inhaled regimens ranked highest by P-score (0.912). This review is limited by a small total study count ( n = 4), reliance on a single observational cohort for injectable regimens, lack of direct head-to-head trials, and imprecision in indirect comparisons. Conclusions Inhaled aminoglycoside regimens appear to yield greater efficacy and a more favorable safety profile than standard therapy, particularly in patients with nodular/bronchiectatic phenotype. Injectable regimens may be considered as salvage therapy for severe fibrocavitary cases due to limited high-quality evidence and substantial toxicity. Further head-to-head RCTs are warranted.