Among middle-aged and older Chinese adults who remained free of high blood pressure through 2015, those carrying a heavier long-term combined burden of insulin resistance and central obesity were substantially more likely to develop hypertension by 2020, according to a prospective cohort study published in Cardiovascular Diabetology – Endocrinology Reports. The study, based on the nationwide China Health and Retirement Longitudinal Study (CHARLS), tracked 2,571 participants aged 45 and older and found that each interquartile-range increase in the cumulative average triglyceride-glucose-waist circumference index, or TyG-WC, was associated with a 40 percent higher hazard of new-onset hypertension in fully adjusted models. Participants in the highest quartile of the index faced roughly double the risk of those in the lowest quartile. The researchers emphasize that the findings are associative rather than causal, but they argue the composite marker, which requires only fasting triglycerides, fasting glucose, and a tape measure, could offer communities an inexpensive way to identify people who might benefit from earlier lifestyle counseling and closer blood pressure surveillance.
The TyG-WC index is built from two well-established components. The triglyceride-glucose index, or TyG, is calculated as the natural logarithm of the product of fasting triglycerides and fasting blood glucose divided by two, and has been proposed as a simple surrogate for insulin resistance, a metabolic state that the hyperinsulinemic-euglycemic clamp and HOMA-IR capture more precisely but that is difficult to measure in large community samples. Insulin resistance and the resulting hyperinsulinemia may raise blood pressure by activating the sympathetic nervous system and the renin-angiotensin-aldosterone system. The waist circumference component adds a direct measure of abdominal obesity, which reflects visceral fat more accurately than body mass index and has been repeatedly linked to hypertension. By multiplying the TyG value by waist circumference in centimeters, the index couples the metabolic base of glucose and lipid metabolism with the visceral fat burden in a single figure.
What distinguishes the new analysis from most previous work is its use of a cumulative average rather than a single measurement. The researchers calculated TyG-WC separately from the 2011 baseline survey and the 2015 follow-up, then averaged the two values to characterize long-term exposure. Single measurements are vulnerable to short-term dietary and physiological fluctuations, seasonal variation, laboratory error, and regression dilution bias, which can weaken observed associations with outcomes that develop over years. Averaging two measurements can, in principle, reduce the influence of random error and better approximate the sustained metabolic background that chronic disease development requires, although the authors note that this assumption was not directly validated within their dataset. Most earlier studies of TyG-derived indices relied on one-time exposure assessments.
To establish a clear temporal sequence, the analytic cohort was constructed with considerable care. The researchers began with CHARLS participants and excluded anyone with hypertension at the 2011 baseline, then excluded those who met hypertension criteria at the 2015 visit or lacked hypertension information in 2020, along with participants missing exposure or covariate data. The final cohort of 2,571 participants, including 1,177 men and 1,394 women, therefore contained no one who had hypertension before or at the second exposure measurement. New-onset hypertension was ascertained at the 2020 follow-up and defined by self-reported physician diagnosis, antihypertensive treatment, or measured blood pressure at or above 140/90 mmHg. By the end of follow-up, 226 participants, or 8.8 percent, had developed hypertension, while 2,345 remained free of the condition.
The statistical analysis was designed to probe the association from several angles. Cox proportional hazards models served as the primary analysis, with logistic regression as a complementary check on cumulative incidence, because exact calendar dates of hypertension onset were unavailable and onset timing was treated as interval-censored at the survey-wave level. The fully adjusted model accounted for age, sex, marital status, education, residence, smoking, drinking, sleep duration, diabetes, stroke, heart disease, dyslipidemia, body mass index, cholesterol fractions, C-reactive protein, white blood cell count, platelet count, creatinine, and glycated hemoglobin. Baseline systolic and diastolic blood pressure were deliberately not included, because high-normal blood pressure may be a preclinical intermediate state on the pathway from metabolic burden to hypertension, although the authors acknowledge that residual confounding by baseline blood pressure severity cannot be ruled out. Waist circumference, triglycerides, and fasting glucose were likewise excluded from the adjustment set because they are mathematical components of the exposure index itself.
The results showed a consistent gradient. In the fully adjusted Cox model, each 1-IQR increase in cumulative average TyG-WC was associated with a hazard ratio of 1.40 (95 percent confidence interval 1.13 to 1.73), and participants in the highest quartile had a hazard ratio of 2.00 (95 percent confidence interval 1.27 to 3.14) compared with the lowest quartile, with a significant trend across quartiles. Logistic regression produced directionally consistent odds ratios of 1.44 per IQR increase and 2.12 for the highest versus lowest quartile, which the authors interpreted cautiously as odds rather than risk ratios given the 8.8 percent cumulative incidence. Restricted cubic spline analysis indicated a significant overall association with no statistically significant departure from linearity, supporting an approximately linear positive dose-response pattern. Participants who developed hypertension were older, more likely to live in rural areas, and had higher baseline body mass index, waist circumference, fasting glucose, and blood pressure than those who did not.
Subgroup analyses found the positive association generally consistent across most population strata, including age, sex, residence, smoking, drinking, and metabolic disease status. In sex-stratified analyses the association was statistically significant in women, with an odds ratio of 1.64, and directionally positive but not significant in men, with an odds ratio of 1.17, and no significant sex interaction was detected. The only statistically significant interaction was observed for heart disease status, a finding the authors urge readers to interpret cautiously because of the small number of participants with heart disease. An exploratory receiver operating characteristic analysis compared a base model of conventional risk factors with an enhanced model that added cumulative average TyG-WC; the addition raised the area under the curve only modestly, from 0.628 to 0.642, and the analysis did not include calibration or reclassification metrics.
The authors situate their findings within a broader evidence base linking TyG-derived indices to cardiometabolic outcomes. Previous studies have reported associations between TyG and its derivatives and hypertension, diabetes, coronary heart disease, acute ischemic stroke, fatty liver disease, and myocardial infarction, and other composite indices such as TyG-BMI have been examined in cardiovascular populations. The mechanistic rationale is that the index integrates two pathways: insulin resistance and hyperinsulinemia, which can promote blood pressure elevation through sympathetic and renin-angiotensin-aldosterone activation, renal sodium retention, vascular smooth muscle proliferation, and endothelial dysfunction; and central obesity, which is associated with adverse lipid profiles, arterial stiffness, and impaired cardiovascular health. The approximately linear dose-response pattern observed here is consistent with a gradient in which higher long-term metabolic burden corresponds to higher blood pressure risk, although the study design cannot confirm that the index itself lies on the causal pathway.
Several limitations temper the conclusions. The complete-case design may introduce selection bias if missingness was related to exposure or outcome, and time-varying initiation, dose, adherence, and duration of antihypertensive, lipid-lowering, and glucose-lowering treatments during follow-up were not incorporated into the main models, even though such treatments may influence both TyG-WC values and hypertension ascertainment. The incremental value of the composite index was not compared directly against cumulative TyG alone or cumulative waist circumference alone, and the ROC analysis was exploratory. Because the cohort comprised Chinese adults aged 45 and older, generalizability to other populations requires external validation. The authors also note that the modest AUC gain means the index should be considered risk-stratification support rather than a validated prediction model.
Even so, the study’s practical appeal lies in its simplicity. Fasting triglycerides, fasting glucose, and waist circumference are routine items in community health examinations, the formula is explicit enough to be embedded in information systems, and repeated measurement is inexpensive. For individuals with persistently high cumulative average TyG-WC, the authors suggest that priority could be given to weight and waist management, diet and exercise interventions, education on glucose and lipid metabolic risk, and more intensive blood pressure follow-up, an approach they frame as earlier and stratified primary prevention rather than clinical prediction. They caution that the observational design cannot establish causality, and that future work should incorporate more survey waves, time-varying medication information, and formal causal inference methods, along with discrimination, calibration, reclassification, and decision-curve metrics to determine whether the composite index truly adds value beyond its individual components.
Subject of Research: The association between cumulative average TyG-WC and new-onset hypertension in Chinese middle-aged and older adults
Article Title: Association between the cumulative average triglyceride-glucose-waist circumference index and new-onset hypertension among Chinese middle-aged and older adults-a national prospective study
Article References: Kuang, Z., Zhang, B., Lin, S., Wang, C., Gao, Y., Zheng, J., & Fan, G. (2026). Association between the cumulative average triglyceride-glucose-waist circumference index and new-onset hypertension among Chinese middle-aged and older adults-a national prospective study. Cardiovascular Diabetology – Endocrinology Reports, 12(1), Article 65. https://doi.org/10.1186/s40842-026-00334-z
Image Credits: AI Generated
DOI: 10.1186/s40842-026-00334-z
Keywords: hypertension, TyG-WC, insulin resistance, central obesity, CHARLS, cohort study, risk stratification, metabolic syndrome, blood pressure, China, dose-response, primary prevention
News Source: Daisy Hatcher. (October 28, 2026). Long-Term Metabolic Burden Index Flags New Hypertension Risk in Chinese Adults. Scienmag.
