Derivation and validation of the incremental prognostic value of increased triglyceride glucose index

Background Triglyceride-glucose index (TyG), a simple and cost-effective marker of insulin resistance, has been recognized as an emerging prognostic biomarker in cardiovascular disease (CVD). Validation of clinically relevant cut-off values in primary and secondary prevention and of its incremental value over established risk scores, is needed to define its applicability in clinical practice. Purpose We aimed to explore the associations of TyG with traits of increased cardiometabolic risk, and externally validate the incremental value of a derived optimal cut-off value over established risk scores. Methods In an ongoing registry aiming to stratify CVD risk (Athens Angiometabolic Study), we analyzed 1,170 consecutively recruited patients (mean age 58.9 years, 57.1% male) (derivation cohort). TyG was calculated by the formula: ln[fasting triglycerides (mg/dL) x fasting glucose (mg/dL)/2]. Indices of subclinical arterial injury (i.e. carotid wall thickness for subclinical atherosclerosis, pulse wave velocity (PWV) for arterial stiffness), the progression of atherosclerosis and the incidence of all-cause mortality were used as endpoints. Because we found increased TyG to incrementally predict mortality only in our primary prevention cohort, for external validation, we included a general population cohort after excluding patients with established CVD (Attica cohort, n=1,879 patients). Patients were followed for a median of 4 years and 10 years for all-cause mortality in derivation and validation cohorts, respectively. Results Elevated TyG index was independently associated with higher number of diseased vascular beds [odds ratio (OR)=2.00], maximal carotid wall thickness (maxWT)>2mm (OR=1.57) and PWV>10m/s (OR=1.51), peripheral and aortic diastolic blood pressure and with increased preperitoneal and subcutaneous fat thickness (P<0.05 for all). Increased TyG associated with accelerated progression of the number of carotid plaques (OR=2.99) and of maxWT (i.e OR=4.37 for >50% increase of maxWT) at follow-up (P<0.05 for both). A TyG index≥8.53 was independently associated with increased risk for all-cause mortality only in the primary prevention subgroup (n=666) of the derivation cohort [hazard ratio (HR)=10.84, P=0.003]. This derived cut-off conferred incremental prognostic value over Systematic Coronary Risk Evaluation 2 (SCORE2) both in the derivation [delta area under the curve (deltaAUC=0.067, P=0.003)] and validation cohort (deltaAUC=0.065, P<0.001). Conclusion In a population with a wide range of cardiometabolic risk, TyG was associated with increased prevalence and progression of subclinical arterial injury and an adverse metabolic profile. In individuals assessed for primary prevention, a TyG≥8.53 conferred incremental value over a well-established guidelines-recommended score, supporting its clinical applicability in this setting.