Abstract
BACKGROUND: This study aimed to compare the efficacy and safety of different mucolytic agents in patients with COPD. METHODS: This study enrolled adults, regardless of the medication regimen. The control groups received placebo, usual care, or a medication other than the intervention group. We reported interleukin-6 (IL-6), interleukin-8 (IL-8), ROS, COPD Assessment Test (CAT) scores, forced expiratory volume in 1 s (FEV1), adverse effects, and exacerbation rates. Statistical analyses were performed using Stata MP15, R4.4.3, and Bayesian network models. The quality of the included studies was assessed using RoB 2.0. PubMed, Embase, Cochrane Library, and Web of Science were searched from database inception up to January 28, 2026 to identify English-language randomized controlled trials (RCTs) reporting outcomes related to inflammatory markers and pulmonary function parameters. The study protocol was registered with PROSPERO (CRD42024622223). RESULTS: Data from 4408 COPD patients across 18 RCTs were analyzed. For reducing 8-isoprostane (SUCRA was 97.06%) and ROS (SUCRA was75.26%), erdosteine 900 mg was more effective. Regarding inflammatory markers, tyloxapol outperformed in reducing IL-6 (SUCRA was 67.16%) and IL-8 levels (SUCRA was 69.44%). For improving FEV1, erdosteine 900 mg (SUCRA: 89.40%) ranked first according to the SUCRA ranking. For reducing exacerbation rates of COPD (SUCRA: 87.72%) and the incidence of adverse reactions (SUCRA: 79.04%), cineole was the most efficacious. These results demonstrated that cineole was the most favorable for lowering the incidence of adverse reactions and exacerbation. CONCLUSION: Erdosteine 900 mg is most effective for improving lung function and reducing oxidative stress, while cineole is safer with a lower incidence of adverse reaction and acute exacerbation. Future multicenter, large-sample RCTs are warranted to collect data on long-term safety, explore combination therapies, and systematically monitor adverse drug reactions. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12890-026-04166-6.