Association between electronic cigarette use and prediabetes, diabetes, and insulin resistance: a systematic review and meta-analysis
摘要
Electronic cigarette (e-cigarette) use has grown rapidly worldwide, yet metabolic health consequences, particularly glycaemic outcomes, remain poorly understood. This systematic review and meta-analysis synthesises the available epidemiological evidence on the association between e-cigarette use and diabetes mellitus, prediabetes, and insulin resistance. We conducted a systematic search of PubMed, Scopus and Embase from inception through December 2025, following PRISMA 2020 guidelines. Studies reporting effect estimates for the association between e-cigarette use and diabetes, prediabetes, or insulin resistance in adult populations were eligible. Random-effects meta-analyses were performed, stratified by outcome and exposure category (current exclusive e-cigarette users, dual users, and former e-cigarette users versus never users). Heterogeneity was assessed using I2 statistics and Cochran’s Q test. Ten studies (9 cross-sectional, 1 prospective cohort; combined participants exceeding 2.8 million) met inclusion criteria; 8 cross-sectional studies contributed to quantitative pooling, while one cross-sectional study and the prospective cohort were narratively synthesised. For prediabetes, pooled ORs versus never users were 1.34 (95% CI 1.01–1.77; I2 = 74.2%) for current exclusive users, 1.26 (1.06–1.49; I2 = 75.4%) for dual users, and 1.13 (1.06–1.20; I2 = 0%) for former users. For diabetes, only dual users showed a significant elevation (OR 1.23; 1.01–1.49; I2 = 77.4%); current exclusive (OR 1.04; 0.93–1.16; I2 = 45.1%) and former users (OR 1.11; 0.80–1.54; I2 = 31.6%) did not differ from never users. For insulin resistance, current exclusive e-cigarette users versus never users had an OR of 1.37 (95% CI 1.06–1.77; I2 = 0%). The sole prospective cohort study found no significant association between exclusive e-cigarette use and incident type 2 diabetes (HR 0.88; 95% CI 0.66–1.16). Neither cross-sectional nor prospective evidence linked exclusive e-cigarette use to diabetes; significant associations were confined to dual users. Elevated odds of prediabetes and insulin resistance among exclusive users were observed cross-sectionally but cannot support causal claims. Large prospective studies with longer follow-up, objective verification of product exposure status (for example, biomarkers distinguishing exclusive vaping from combustible tobacco use), and standardised laboratory-based glycaemic endpoints are needed.