Although sodium-glucose transporter 2 inhibitors (SGLT2is) are recommended medications for the treatment of type 2 diabetes mellitus (T2DM), conclusive evidence suggesting that they could alleviate the common complication of obstructive sleep apnea (OSA) in T2DM is lacking at present. Our objective is to systematically summarize the clinical evidence of SGLT2is for OSA in patients with T2DM by searching 12 databases and registers from the establishment to July 25, 2025. We included randomized controlled trials (RCTs) and non-randomized studies evaluating the use of SGLT2is for OSA in participants with T2DM. This analysis was registered on the PROSPERO website (CRD42024576637) and performed in accordance of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. Data were extracted by two investigators separately, and the meta-analysis of sleep-disordered breathing parameters, metabolic parameters, and adverse events outcomes was conducted using Review Manager 5.4 and Stata 15.0. Nine studies, including six RCTs and three non-RCTs, were included. The results demonstrated that the efficacy of SGLT2is in reducing the apnea-hypopnea index (AHI) (mean difference [MD] = -12.57, 95 % confidence interval [CI]: [-21.47, -3.66], P = 0.006, I = 87 %) and increasing the lowest oxygen saturation (lowest SpO) was superior to that of the control interventions with other hypoglycemic drugs. Patients with T2DM showed a 50 % relative risk reduction for incident OSA when receiving SGLT2is in comparison with placebo. Furthermore, the adverse event rate in the SGLT2i group was comparable to that in the control group. Considering of the risk reduction for incident OSA and the enhanced metabolic parameters observed with SGLT2i administration, these drugs may be recommended as useful medication for the management of T2DM and OSA. However, Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) assessments rated the evidence quality as low to moderate because of inconsistency and indirectness. Given the weak strength of the evidence, further trials with sufficient statistical power are still needed to confirm efficacy and safety. 2 2
