This study aims to systematically evaluate which single-inhaler triple therapy (inhaled corticosteroids ICS, long-acting β2-agonists LABA, and long-acting muscarinic antagonists LAMA) is safer and more effective for treating chronic obstructive pulmonary disease (COPD). A comprehensive search was performed in PubMed, Embase, Ovid, Cochrane library and Google Scholar from database establishment to November 2025. Searches were limited to English articles. The RCTs that compared single-inhaler triple therapy (ICS/LABA/LAMA) with triple therapy (ICS/LABA+LAMA) or dual therapy (ICS/LABA or LABA/LAMA) for COPD were included. Minimum duration ≥ 12 weeks and minimum participant numbers ≥ 300 patients. Outcomes included forced expiratory volume in 1 s (FEV1), moderate and severe exacerbations, St George’s Respiratory Questionnaire (SGRQ) total score and SGRQ responders, transition dyspnea index (TDI) focal score and safety. 12 RCTs (n = 28930 patients) were included in this network meta-analysis. Fluticasone furoate/vilanterol/umeclidinium (FF/VIL/UMEC) was statistically significantly more effective at increasing trough FEV1 (based on change from baseline) than dual therapies (ICS/LABA or LABA/LAMA), free triple therapy (ICS/LABA+LAMA), budesonide/formoterol fumarate/glycopyrronium bromide (BUD/FOR/GLY) and beclomethasone dipropionate/formoterol fumarate/glycopyrronium bromide (BDP/FOR/GLY). In addition, FF/VIL/UMEC, BUD/FOR/GLY, and free triple therapy (ICS/LABA+LAMA) showed significant improvement in the total SGRQ score to dual therapies (ICS/LABA or LABA/LAMA). FF/VIL/UMEC and free triple therapy (ICS/LABA+LAMA) showed borderline significant improvement in the total SGRQ score to BDP/FOR/GLY. The four available single-inhaler triple therapies and free triple therapy appear to have similar effectiveness and safety. Given the absence of direct comparison studies differences cannot be found with a sufficient level of certainty. Further analysis is needed, as additional evidence becomes available.
Zhang et al. (2026) studied this question.