GLP-1 Receptor Agonists and Risk of Adverse Cerebrovascular Outcomes in Type 2 Diabetes: A Systematic Review and Meta-Analysis of Randomized Controlled Trials

CONTEXT: The effect of glucagon-like peptide-1 receptor agonists (GLP-1RAs) on ischemic/hemorrhagic stroke and transient ischemic attacks (TIA) in type 2 diabetes mellitus (T2DM) remains undetermined. OBJECTIVE: To pool effects of GLP-1RAs on adverse cerebrovascular outcomes and investigate impact of baseline variables on these effects. METHODS: PubMed, Embase, Web of Science, Cochrane Library, and clinical trial registry websites were searched for randomized controlled trials (RCTs) ≥24 weeks duration in adults with T2DM (PROSPERO: CRD42022331547). Adjudicated cerebrovascular events in GLP-1RA treatment vs control arms were pooled together to calculate risk ratios (RR) using fixed-effects model. Subgroup analysis was performed based on individual drugs, treatment duration, and baseline patient characteristics. Quality of evidence was assessed using GRADE framework. RESULTS: We identified 28 RCTs involving 74 148 patients (57% male; median [range], age 58 [52-67] years, BMI 32 [25.4-37.2] kg/m2, T2DM duration 9 [3.5-15.4] years, treatment duration 52 [24-259] weeks). GLP-1RA use in T2DM was associated with significantly decreased risk of adverse cerebrovascular outcomes vs placebo/active comparator (RR, 0.83; 95% CI, 0.76-0.91; I2 = 0%). Pooling data from cardiovascular outcome trials (n = 8), GLP-1RA treatment vs placebo was associated with reduced risk of nonfatal stroke (RR, 0.85; 95% CI, 0.76-0.94; I2 = 0%) but not fatal stroke (RR, 0.80; 95% CI, 0.61-1.05; I2 = 0%). GLP-1RA use was associated with reduced risk of ischemic stroke (RCTs = 12; RR, 0.73; 95% CI, 0.60-0.89; I2 = 0%), composite of ischemic stroke/TIA (RCTs = 16; RR, 0.76; 95% CI, 0.65-0.90; I2 = 0%), but not hemorrhagic stroke (RCTs = 3; RR, 0.92; 95% CI, 0.51-1.64; I2 = 0%). Treatment benefits differed according to baseline eGFR and diabetes duration (P interaction < .1). Benefits were statistically significant for dulaglutide, subcutaneous/oral semaglutide (P < .05). Sensitivity analysis, excluding shorter-acting lixisenatide, eliminated the heterogeneity between individual GLP-1RA effects. CONCLUSION: GLP-1RAs, particularly longer-acting formulations, reduced ischemic cerebrovascular events in T2DM. Observed benefits were significantly higher in patients with shorter T2DM duration and higher eGFR.