Biomarkers and Personalized Medicine in Cardiovascular Disease Symposium

Association between Triglyceride- Glucose Index and Heart Failure Risk by Body Mass  Index in a Multi-Ethnic Cohort

Nandan Thakkar, MD, MSc¹; Kimberly Hong, MD, MHSA²; Michael D. Shapiro, DO, MCR³; Alain G. Bertoni, MD, MPH⁴; Harpreet S. Bhatia, MD, MAS, FACC²¹ Department of Medicine, University of California San Diego, La Jolla, CA² Division of Cardiovascular Medicine, Department of Medicine, University of California San Diego, La Jolla, CA³ Center for Preventive Cardiology, Wake Forest University School of Medicine, Winston-Salem, NC⁴ Department of Internal Medicine and Public Health Sciences, Wake Forest University School of Medicine, Winston-Salem, NCBackground: Insulin resistance contributes to heart failure (HF) through metabolic dysregulation, systemic inflammation, and adverse myocardial remodeling. However, commonly used measures of cardiometabolic risk, such as body-mass-index (BMI), may not capture underlying metabolic dysfunction in all populations. The triglyceride-glucose index (TyG) is a validated surrogate marker of insulin resistance that may better reflect pathophysiologic processes associated withHF, particularly in non-obese individuals.

Methods: We analyzed 6,683 participants from the Multi-Ethnic Study of Atherosclerosis (MESA) cohort who were free of baseline HF. TyG was calculated as ln(fasting triglycerides × fasting glucose/2). Multivariable cox proportional hazards models were used to evaluate the association between TyG (per 1-unit increase) and incident HF, adjusting for age, sex, race, systolic blood pressure, total and HDL cholesterol, eGFR, diabetes, C-reactive protein, antihypertensive and lipid-lowering medication use, and smoking status. Analyses were stratified by BMI (obese ≥30 kg/m² vs. non-obese <30 kg/m²) and HF subtype (HFpEF ≥50%, HFrEF <50%).

Results: Over a median follow-up of 15.2 years, 502 HF events occurred. In fully adjusted models, higher TyG was associated with increased HF risk (HR 1.24, 95% CI 1.02–1.50, p=0.032). This association was significantly modified by BMI, with a BMI-by-TyG interaction (p = 0.010), andrestricted to non-obese participants (n=4,544; 302 events; HR 1.35, 95% CI 1.05–1.74, p=0.018), with no significant association among obese individuals (n=2,139; 200 events; HR 1.09, 95% CI 0.80–1.48, p=0.60). In HF subtype analyses, TyG demonstrated a trend towardassociation with HFpEF (HR 1.32, 95% CI 0.96–1.82, p=0.085) but not HFrEF (HR 1.03, 95% CI 0.78–1.37, p=0.81).

Conclusions: The TyG index is associated with incident HF independent of traditional risk factors – a relationship that appears to be modified by BMI and most pronounced in non-obese individuals. These findings suggest that insulin resistance—reflecting underlying metabolic dysfunction even in the absence of elevated BMI—may be a key contributor to HF risk. Incorporating markers of metabolic dysregulation such as TyG might improve identification of at-risk individuals not captured by conventional anthropometric measures.