Association of Metformin Use with Asthma Exacerbation in Patients with Concurrent Asthma and Diabetes: A Systematic Review and Meta-Analysis of Observational Studies

BACKGROUND: Asthma and diabetes are both diseases that affect a wide range of people worldwide. As a common treatment for diabetes, metformin has also been reported to be effective in improving asthma outcomes. We conducted a combined analysis to examine the efficacy of metformin in reducing asthma exacerbation in patients with concurrent asthma and diabetes. METHODS: We searched the PubMed, Embase, and CENTRAL databases for articles published prior to April 2020 to find observational studies of individuals with concurrent asthma and diabetes that compared the risk of asthma exacerbation between metformin users and nonusers. Two researchers separately screened the studies, extracted data, and evaluated the risk of bias. The primary outcome was the adjusted risk of asthma exacerbation. The secondary outcomes were the adjusted risk of asthma-related hospitalization and emergency room visits. Review Manager was used for data analysis and plotting. I(2) and χ(2) tests were used to estimate heterogeneity. A random effects or fixed effects model was used depending on the heterogeneity. Odds ratios were calculated for dichotomous variables. RESULTS: We included two studies with a total of 25252 patients. The pooled effect size showed that metformin was inversely associated with a risk of asthma exacerbation (OR = 0.65, 95% CI 0.28–1.48; χ(2) = 5.42, P=0.02; I(2) = 82%), asthma-related emergency department visits (OR = 0.81, 95% CI 0.74–0.89; χ(2) = 0.36, P=0.55; I(2) = 0%), and hospitalizations (OR = 0.43, 95% CI 0.14–1.29; χ(2) = 4.01, P=0.05; I(2) = 75%). CONCLUSION: This meta-analysis suggested that metformin decreased the risk of asthma-related emergency room visits for patients with concurrent asthma and diabetes. Metformin reduced the risk of asthma-related hospitalization and exacerbation but was not statistically significant. More randomized trials involving larger samples should be considered, and the mechanisms of these effects need to be fully elucidated.