Triglyceride Glucose Index Predicts Risk of Adverse Cardio-metabolic and Mortality Outcomes Among Chinese Adults: A Territory-Wide Longitudinal Study

BackgroundThe triglyceride glucose (TyG) index has been proposed to be a surrogate of insulin resistance. In the study, we aimed to examine the relationship between TyG index and the risk of new onset DM and secondary outcomes included atrial fibrillation (AF), heart failure (HF), acute myocardial infarction (AMI), ventricular tachycardia/fibrillation (VTF), cardiovascular mortality (CAD) and all-cause mortality. MethodsThe retrospective observational study analyzed patients recruited from 1st January 2000 to 31st December 2003 and followed up until 31st December 2019. Demographics, past comorbidities, medications and laboratory tests were extracted. At baseline and follow-up, DM was defined as 1) any occasion Hb1Ac [&ge;]14 g/dL, 2) fasting glucose[&ge;]7 mmol/L on two occasions, or 3) with DM diagnosis. We excluded 1) patients with prior DM or with the use of antidiabetic medications; 2) patients with prior AMI/HF/AF or with the use of diuretic/beta blockers for HF were excluded. Univariate analysis and multivariate Cox analysis with adjustments on demographics, past comorbidities and medications were conducted to identify the significant risk predictors of primary and secondary outcomes. Optimal cutoffs of TyG index for the primary and secondary outcomes were found with maximally selected rank statistics approach. ResultLager TyG index is significantly associated with new onset DM (HR: 1.51, 95% CI: [1.47, 1.55], P vlaue<0.0001), new onset HF (HR: 1.27, 95% CI: [1.2, 1.34], P vlaue<0.0001), new onset AF (HR: 2.36, 95% CI: [2.26, 2.46], P vlaue<0.0001), new onset AMI (HR: 1.51, 95% CI: [1.42, 1.6], P vlaue<0.0001), new onset VTF (HR: 1.22, 95% CI: [1.13, 1.31], P vlaue<0.0001), new onset CAD (HR: 1.56, 95% CI: [1.45, 1.69], P value<0.0001) and all-cause mortality (HR: 1.21, 95% CI: [1.18, 1.25], P vlaue<0.0001). TyG index and its 3rd tertile remained significant after being adjusted with significant demographics, past comorbidities and medicactions in multivariate cox models (HR>1, P value<0.05). Optimal cut-off values of baseline TyG index and adjusted multivariate restricted cubic spline models further uncovered detailed associations of larger baseline TyG index with the primary and secondary outcome. ConclusionHigher TyG index remained significantly associated with the elevated risk of new onset DM, AF, HF, AMI, VTF, CAD and all-cause mortality after adjustments on demographics, past comorbidities, and medications.