The association between triglyceride-glucose index and the incidence of type 2 diabetes mellitus—a systematic review and dose–response meta-analysis of cohort studies

Abstract

We aimed to assess the dose–response relationship between triglyceride–glucose (TyG) index and the incidence of type 2 diabetes mellitus (T2DM). We performed a comprehensive systematic literature search using PubMed, Scopus, and Embase for records published from inception until 9 February 2021. The effect estimates were reported as relative risks (RRs). 270,229 subjects from 14 studies were included in this systematic review and meta-analysis. The pooled incidence of T2DM was 9%. Meta-regression analysis indicates that baseline age (coefficient: 0.67, p\u2009=\u20090.026), drinking (coefficient: 0.03, p\u2009=\u20090.035), and HDL (coefficient: −0.89, p\u2009=\u20090.035) affected the incidence of T2DM in future. High TyG index was associated with increased incidence of T2DM in pooled unadjusted (RR 4.68 [3.01, 7.29], p\u2009<\u20090.001; I2: 96.6%) and adjusted model (adjusted RR 3.54 [2.75, 4.54], p\u2009<\u20090.001; I2: 83.7%). Dose–response meta-analysis for the adjusted RR showed that the linear association analysis was not significant per 0.1 increase in TyG index (RR 1.01 [0.99, 1.03], p\u2009=\u20090.223). There is a non-linear trend (p\u2009<\u20090.001) for the association between TyG index and incidence of T2DM. The dose–response curve became increasingly steeper at TyG index above 8.6. TyG index was associated with the incidence of T2DM in a non-linear fashion.