Simple insulin resistance scores linked to heart disease in early metabolic syndrome

Twelve simple, low-cost measures of insulin resistance were all linked to the risk of developing cardiovascular disease in adults with early-stage cardiovascular-kidney-metabolic (CKM) syndrome, according to a prospective cohort study published in Frontiers in Cardiovascular Medicine. The strongest association was seen for a composite index that combines cholesterol, high-density lipoprotein cholesterol, fasting glucose and a measure of visceral adiposity, known as the CHG-CVAI index.

The study drew on the Qingdao Diabetes Program, a population-based research effort in Qingdao, China, that conducted standardized diabetes surveys in 2006 and 2009. From a pooled sample of participants aged 35 to 74 years, the researchers identified 5,625 adults who met criteria for stages 0 to 3 of CKM syndrome, a framework proposed by the American Heart Association in 2023 that describes the continuum from no risk factors through excess or dysfunctional adiposity, metabolic risk factors such as diabetes, hypertension or chronic kidney disease, and finally subclinical cardiovascular disease. Participants with prevalent cardiovascular disease at baseline were excluded, leaving 2,192 men and 3,433 women with a mean age of about 52 years.

CKM syndrome stages 0 to 3 are considered a critical window for prevention, because people in these early, largely asymptomatic phases may still avoid progression to overt clinical events if risk factors are addressed. Yet systematic comparisons of which insulin resistance surrogates work best in this specific population have been lacking. Insulin resistance, a core mechanism linking metabolic dysfunction to atherosclerotic disease, is usually estimated in practice with surrogate indices because the gold-standard measurement, the hyperinsulinemic-euglycemic clamp, is invasive and unsuitable for routine clinical use.

The researchers evaluated twelve surrogates: the triglyceride-glucose (TyG) index, the atherogenic index of plasma, the cardiometabolic index, the metabolic score for insulin resistance (METS-IR), the cholesterol-HDL-glucose (CHG) index, and seven composite indices that combine CHG with an adiposity measure, including body mass index, waist circumference, waist-to-height ratio, body roundness index, A body shape index, weight-adjusted waist index and the Chinese visceral adiposity index (CVAI). All indices were calculated from baseline fasting blood samples and standardized anthropometric measurements collected during the 2009 examination.

Participants were followed through the Qingdao Cardiovascular Disease Surveillance System, with outcomes defined as the first major cardiac event or stroke and coded according to the International Classification of Diseases, Tenth Revision. Over a median follow-up of 15.2 years, 1,407 participants, or 25 percent of the cohort, experienced an incident cardiovascular event. In multivariable-adjusted Cox models that accounted for age, sex, residence, smoking, drinking, marital status, education, hypertension, hyperuricemia and chronic kidney disease, all twelve indices were significantly associated with cardiovascular risk, with all quartile trend tests remaining significant after false discovery rate correction.

CHG-CVAI showed the strongest association: participants in the highest quartile had 2.10 times the risk of cardiovascular disease compared with those in the lowest quartile (hazard ratio 2.10, 95 percent confidence interval 1.74 to 2.52). METS-IR followed with a hazard ratio of 1.87 (95 percent CI 1.58 to 2.21) for the highest versus lowest quartile, and CHG-BMI showed a hazard ratio of 1.81 (95 percent CI 1.54 to 2.14). When analyzed as continuous variables, every index showed a significant positive association, with the per-standard-deviation hazard ratio ranging from 1.07 for the cardiometabolic index to 1.24 for CHG-CVAI.

Restricted cubic spline analyses revealed that most of the indices had nonlinear dose-response relationships with cardiovascular risk. Nine of the twelve surrogates, including CHG, AIP, TyG, METS-IR, CMI, CHG-CVAI, CHG-BMI, CHG-BRI and CHG-WWI, showed statistically significant nonlinearity, and threshold analyses identified inflection points for all twelve indices. For most nonlinear indices, the association followed a threshold pattern in which risk rose above a certain level and then attenuated at higher values, a pattern the authors note may partly reflect the obesity paradox. The researchers cautioned that bootstrap resampling confirmed stable inflection points only for CHG, TyG and METS-IR, while the threshold estimates for the remaining nine indices had wide confidence intervals and should be regarded as exploratory.

Subgroup analyses suggested that the associations were stronger among adults younger than 60 years and among urban residents compared with their counterparts, with age significantly moderating the associations of CHG-CVAI, CHG-WWI and CHG-ABSI, and residence moderating those of CHG, CHG-WWI, CHG-ABSI and AIP. No consistent effect modification was observed for sex, CKM stage, hypertension or hyperuricemia. Sensitivity analyses, including restricting follow-up to at least five years, excluding participants taking lipid-lowering or glucose-lowering medications, using tertiles instead of quartiles, and competing-risk regression treating non-cardiovascular death as a competing event, all confirmed the robustness of the primary findings.

The predictive performance of the indices was more modest. The basic model, adjusted for conventional risk factors, achieved a Harrell’s C-index of 0.667, and adding any single surrogate raised it only slightly, to between 0.669 and 0.676. In time-dependent area-under-the-curve analyses at 5, 10 and 15 years, adding an individual index increased the AUC by less than 0.02 at every time point, and none of these increments was statistically significant. However, the indices did significantly improve risk reclassification: at 10 years, the net reclassification improvement ranged from 0.089 to 0.205 and the integrated discrimination improvement ranged from 0.0004 to 0.0032, all statistically significant. The authors interpret this as evidence that the surrogates are intended to complement, not replace, existing risk assessment tools, helping to refine risk categorization in early-stage CKM syndrome even though their incremental discriminative value in absolute terms is limited.

The study has several limitations that the authors acknowledge. Only single baseline measurements of the indices were used, so changes in insulin resistance over time could not be assessed. As an observational study, residual confounding cannot be fully excluded despite extensive adjustment, and the cohort consisted solely of residents of Qingdao, China, which may limit generalizability to other populations. The authors also note possible selection bias from excluded participants, although baseline characteristics of included and excluded individuals did not differ significantly. They conclude that all twelve insulin resistance surrogates, particularly CHG-CVAI, were independently associated with incident cardiovascular disease in early-stage CKM syndrome, and that longer-term studies in more diverse cohorts are needed to validate the findings and explore the mechanisms underlying the link between insulin resistance and cardiovascular risk.

Subject of Research: Insulin resistance surrogate indices and incident cardiovascular disease risk in early-stage cardiovascular-kidney-metabolic syndrome

Article Title: Associations of insulin resistance surrogates with cardiovascular disease in early stage cardiovascular–kidney–metabolic syndrome: a prospective cohort study

Article References: Li, A., Tian, S., Zhang, J., Zhang, D., Xu, J., Wang, X., Yan, H., Sun, X., Ning, F., & Zhang, L. (2026). Associations of insulin resistance surrogates with cardiovascular disease in early stage cardiovascular–kidney–metabolic syndrome: a prospective cohort study. Frontiers in Cardiovascular Medicine, 13, Article 1876121. https://doi.org/10.3389/fcvm.2026.1876121

Image Credits: AI Generated

DOI: 10.3389/fcvm.2026.1876121

Keywords: insulin resistance, cardiovascular disease, cardiovascular-kidney-metabolic syndrome, CHG-CVAI, TyG index, prospective cohort study, risk prediction, Qingdao Diabetes Program, Cox regression, risk stratification, METS-IR, visceral adiposity