Combination of triglyceride-glucose index and waist-to-height ratio as a predictor of all-cause and cardiovascular mortality in adults with diabetes or prediabetes: a nationwide prospective cohort study

Introduction

Diabetes mellitus (DM) and its associated complications have emerged as a significant global public health concern in recent years. The International Diabetes Federation reports that currently over 589 million individuals worldwide are living with diabetes, and this figure is projected to rise to 853 million by 2050 (1). Prediabetes, characterized by blood glucose levels that exceed normal range but do not meet the criteria for DM (2), is strongly linked to the progression of diabetes. A meta-analysis of 102 prospective studies found that diabetes leads to a two-fold additional risk of vascular outcomes (coronary heart disease, ischemic stroke, and vascular death) (3). Similarly, there is also a significant association between prediabetes and cardiovascular events. The results show that during a 6.6-year period, the absolute risks of mortality and cardiovascular disease (CVD) are 7.36 per 10,000 person-years and 8.75 per 10,000 person-years respectively (4). CVD remains a leading cause of mortality among patients with DM globally (5). Additionally, epidemiological evidence indicates that DM is associated with an increased risk of cerebrovascular disease, kidney disease, and retinopathy (6,7). Moreover, prediabetes is also strongly associated with cardiovascular and renal disease (8,9).

The pathogenesis of CVD in patients with DM and prediabetes is multifactorial. Early management of traditional cardiovascular risk factors can markedly diminish the occurrence of cardiovascular events and mortality in these populations (10-12). The complex interrelationship between insulin resistance (IR), obesity, and cardiovascular disease (CVD) development is well-established in the literature. The triglyceride/high-density lipoprotein cholesterol (TG/HDL-C) ratio, as well as the triglyceride-glucose (TyG) index have emerged as valuable biomarkers for IR assessment in both diabetes and non-diabetes populations (13,14). Unlike the technically demanding insulin-glucose clamp test or the homeostasis model assessment of IR (HOMA-IR), which requires insulin measurement, the TyG index provides a simpler, more cost-effective alternative. This practical advantage has led to its widespread adoption in clinical research and practice for IR assessment across diverse patient populations (15,16). Recent research indicates that the TyG index is significantly associated with adverse clinical outcomes in patients with CVD, heart failure, stroke, or other related conditions (17-20).

DM is often accompanied by obesity, which is associated with various health risks, such as IR (21,22), which increase the risk of CVD in patients with diabetes, especially those with central obesity. Epidemiological studies have shown that waist-to-height ratio (WHtR), which reflects central obesity, is more strongly associated with significant cardiometabolic risk compared with waist circumference (WC) and body mass index (BMI) (23,24). A meta-analysis of more than 300,000 adults from various ethnic groups found that WHtR outperformed WC and BMI in detecting cardiometabolic risk factors (23).

The integration of the TyG index with central obesity indicators is anticipated to offer enhanced clinical predictive utility. These novel composite indices may provide a more precise prediction of disease outcomes than the TyG index alone. Research suggests that the TyG-WHtR index outperforms the standalone TyG index in identifying risks associated with CVD and diabetes (25-27). However, while these studies have primarily focused in the general population, the association between the combination of TyG and WHtR indices with all-cause mortality and CVD-specific mortality in a United States (U.S.) population with prediabetes and diabetes has not been extensively examined.

Therefore, we conducted a study to clarify the relationship between the TyG-WHtR index and clinical outcomes in adults with prediabetes and diabetes by analyzing a large population cohort. It’s hypothetical that elevated TyG-WHtR was positively associated with mortality in patients with prediabetes or diabetes. It indicates that TyG-WHtR may be a valuable indicator of poor prognosis in patients with prediabetes or diabetes. We present this article in accordance with the STROBE reporting checklist (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-206/rc).

Methods

Study cohort

The study population was derived from the National Health and Nutrition Examination Survey (NHANES), a comprehensive national survey conducted in the US that comprises adults and children (28). The NHANES database employs a stratified, multistage, probability-based sampling design, ensuring national representativeness. Ethics approval for the NHANES study program was granted by the Ethics Committee of the National Center for Health Statistics (NCHS), and written informed consent was obtained from all participants. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Detailed data and descriptive information from the NHANES database can be conveniently accessed through its official website (https://www.cdc.gov/nchs/nhanes/).

The data for this analysis were derived from eight waves (2003–2004 to 2017–2018) of the NHANES, with data collection spanning from January 2003 to December 2018. As the exact interview date is not provided to protect participant privacy, the study entry time was operationally defined by the examination period of each cycle. The total number of participants across these cycles was 80,312. Prediabetes was defined based on self-reported prediabetes status, a fasting blood glucose (FBG) level between 100 and 125 mg/dL, or a hemoglobin A1c (HbA1c) value ranging from 5.7% to 6.4% (29). Diabetes was diagnosed based on self-reported physician diagnosis, use of hypoglycemic agents or insulin, fasting plasma glucose levels ≥126 mg/dL, random blood glucose or 2-hour oral glucose tolerance test (OGTT) level ≥200 mg/dL, or HbA1c ≥6.5% (30). The participant inclusion flowchart is shown in Figure 1.

Figure 1 Flowchart of study participant inclusion. NHANES, National Health and Nutrition Examination Survey; TyG-WHtR, triglyceride-glucose combined with waist-to-height ratio.

Participants were excluded if they met any of the following criteria: (I) younger than 18 years of age; (II) missing TyG-WHtR data; (III) neither prediabetes and diabetes; (IV) pregnant at the interview; and (V) missing follow-up data. Ultimately, a final cohort of 19,563 participants with prediabetes or diabetes was included for analysis.

Assessment of the TyG-WHtR index

This study measured the TyG-WHtR index, including the TyG and WHtR indices (14,31,32). The TyG index is calculated as follows: TyG index = Ln [TG (mg/dL) × FBG (mg/dL)/2]. Meanwhile, WHtR is calculated as follows: WHtR = WC (cm)/height (cm). Finally, TyG-WHtR is calculated as follows: TyG-WHtR = TyG × WHtR. In this study, participants were divided into four groups according to the TyG-WHtR quartile: Q1, Q2, Q3, and Q4.

Assessment of covariates

Basic information about the participants was collected through standardized questionnaires, including age, sex, race, education, household income, smoking status, height, weight, WC, drug use (antihypertensive, hypoglycemic, or insulin), and disease status. Race categories included non-Hispanic White, non-Hispanic Black, Mexican American, and other. Education level categories included less than a high school diploma, high school/equivalent, college, and above college. The family income-to-poverty ratio was defined as ≤1.30, 1.31–3.50, and >3.50. BMI was calculated by dividing weight in kilograms by the square of height in meters. Smoking status categories included never smoking, past smoking, and current smoking. Hypertension was considered present with a self-reported diagnosed hypertension, a mean systolic/mean diastolic blood pressure ≥140/90 mmHg, or the intake of oral antihypertensive medication. Drinking status categories included never drinking, moderate drinking (0.1–13.9 g/day for women and 0.1–27.9 g/day for men), and heavy drinking (≥14 g/day for women and ≥28 g/day for men). The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation was used to calculate the estimated glomerular filtration rate (eGFR) from serum creatinine (33). In addition, laboratory tests of the participants were collected, including blood urea nitrogen (BUN), HbA1c, total cholesterol (TC), high-density lipoprotein cholesterol (HDL-c), FBG, and triglyceride (TG) level.

Outcomes

In order to confirm the mortality status of participants during follow-up, we used relevant mortality files publicly available via NHANES as of December 31, 2019. The document correlates the NCHS with the National Death Index (NDI) through a probability matching algorithm. We used the International Statistical Classification of Diseases, 10^th Edition (ICD-10), to determine disease-specific deaths. Cardiovascular mortality rates include heart disease (ICD-10 codes I00–I09, I11, I13, and I20–I51) and cerebrovascular disease (ICD-10 codes I60–I69).

Statistical analysis

All analyses were conducted using R version 4.3.2 (The R Foundation for Statistical Computing, Vienna, Austria) Given the complex sampling design of the NHANES, all statistical analyses incorporated sample weights, clustering, and stratification to accurately represent the population. Participants were categorized into four groups based on the quartiles of the TyG-WHtR index (Q1–Q4). Continuous variables are presented as the mean ± standard deviation, while categorical variables are presented as the frequency (weighted percentage).

Baseline characteristics were compared using Student t-test for two-group comparisons, with analysis of variance (ANOVA) for comparisons involving more than two groups, the Mann-Whitney U test and Kruskal-Wallis test for nonparametric data, and the Chi-squared test for categorical variables. Comparisons among the quartile groups of TyG-WHtR were adjusted using the Bonferroni test for multiple comparisons (P<0.05/6=0.0083 was used as the Bonferroni-corrected P value threshold). A Bonferroni-corrected P value <0.05 was considered statistically significant.

Kaplan-Meier curve analysis and the log-rank test were used to assess the overall survival probability among the prediabetes and diabetes populations across the different TyG-WHtR index quartiles. Cox proportional hazard regression models were used to estimate the hazard ratios (HR) and 95% confidence intervals (CI) for the association between the TyG-WHtR index and all-cause mortality and CVD-related mortality. The proportional hazards assumption was evaluated through Schoenfeld residuals, and the results indicated that this assumption was not violated. Model 1 was unadjusted, while Model 2 was adjusted for age, sex, race, education level, and family income-to-poverty ratio. In the fully adjusted model (Model 3), additional adjustments were made for smoking status, drinking status, hypertension, BMI, eGFR, HbA1c, TC, HDL-C, uric acid, and BUN. The median values of each category were used as continuous variables to analyze linear trends. The dose-response relationship between the TyG-WHtR index and mortality was visualized using restricted cubic splines (RCS). The 5^th, 35^th, 65^th, and 95^th percentiles of the TyG-WHtR distribution were analyzed with RCS to capture the nonlinear relationship. We conducted subgroup analyses stratified by sex, age (≤60 or >60 years), race (non-Hispanic White or other), BMI (≤30 or >30 kg/m²), drinking status (never or moderate/heavy), smoking status (never or former/current), hypertension (no or yes), diabetes status (prediabetes or diabetes), education level (less than high school or college), and family income-to-poverty ratio (≤3.5 or >3.5). Potential modification effects were assessed by evaluating the corresponding multiplicative interaction terms.

Furthermore, several sensitivity analyses were performed to validate the robustness of our findings. Initially, we excluded participants who died within 2 years of follow-up to minimize the influence of early mortality on the results. Additionally, participants with a history of CVD or malignancy at baseline were excluded. Finally, we excluded individuals who used hypoglycemic agents or insulin at baseline to avoid confounding by medication effects.

We used the ‘mice’ package in R language to perform multiple imputation through chained equations. Statistical significance was determined using a two-sided P value <0.05.

Results

Baseline characteristics of study participants

Table 1 presents baseline characteristics of study participants stratified by TyG-WHtR quartiles. The cohort (n=19,563) had a mean age of 53.78±16.02 years, with 47.69% male participants and a mean TyG-WHtR index of 6.01±1.14. A distinct demographic and clinical profile emerged across TyG-WHtR quartiles. Participants in higher TyG-WHtR quartiles were predominantly female (56.12% in Q4 vs. 42.17% in Q1, P<0.001), older, more likely to be non-Hispanic White, and had higher rates of obesity (94.26% in Q4 vs. 4.94% in Q1, P<0.001). These individuals also demonstrated lower educational attainment and income levels. Clinically, higher TyG-WHtR was associated with increased prevalence of hypertension (67.27% in Q4 vs. 33.69% in Q1, P<0.001), longer diabetes duration, and greater use of hypoglycemic medications and insulin. Laboratory parameters revealed that participants with higher TyG-WHtR had significantly elevated HbA1c, uric acid, BUN, triglycerides, and total cholesterol, alongside reduced HDL-C and eGFR levels (all P<0.001).

During a median follow-up of 7.6 years, 2,949 (15.1%) all-cause deaths were recorded, of which 969 (32.86%) were CVD-related deaths. Compared with survivors, nonsurvivors were more likely to be male, older, non-Hispanic White, current smokers, and never alcohol drinkers and have a lower education level, lower BMI, and lower income; additionally, they were more likely to have high blood pressure, diabetes, and a higher TyG-WHtR index (Table 2).

Association of the TyG-WHtR index with adult prediabetes and diabetes mortality

Compared with the prediabetes and diabetes participants in the lowest quartile for the TyG-WHtR index, those in the highest levels had significantly higher all-cause mortality and CVD-related mortality (log-rank test P<0.01) (Figure 2).

Figure 2 Kaplan-Meier curves for the survival patterns of prediabetes and diabetes adults with different quartile levels of TyG-WHtR indices. (A) All-cause mortality of prediabetes and diabetes adults at different TyG-WHtR quartile levels. (B) CVD-related mortality of prediabetes and diabetes adults at different TyG-WHtR quartile levels. TyG-WHtR quartile levels: Q1, <5.23; Q2, 5.23–5.93; Q3, 5.94–6.70; Q4, >6.70. CVD, cardiovascular disease; TyG-WHtR, triglyceride-glucose combined with waist-to-height ratio.

We constructed three Cox regression models to examine the association between the TyG-WHtR index and mortality (Table 3). In the fully adjusted model, the multivariable-adjusted HRs and 95% CIs for all-cause mortality from the lowest to highest TyG-WHtR index quartiles were 1.00 (reference), 1.02 (0.84–1.23), 1.16 (0.95–1.42), and 1.39 (1.06–1.81), respectively; meanwhile for CVD-related mortality the corresponding HRs were 1.00 (reference), 1 (0.73–1.37), 1.26 (0.9–1.77), and 1.36 (0.91–2.03), respectively.

Nonlinear associations in the TyG-WHtR index and mortality in adults with prediabetes and diabetes

As shown in Figure 3, we performed multivariate-adjusted RCS analysis and found that the relationship between the TyG-WHtR index and all-cause mortality was nonlinear (P for nonlinearity <0.001). However, the TyG-WHtR index was linearly correlated with CVD-related mortality (P for nonlinearity =0.57). We determined that the inflection points for all-cause and cardiovascular mortality were 5.9, respectively. We further investigated the dose-response relationship between the TyG-WHtR index and mortality in individuals with prediabetes and diabetes (Figure 4). The TyG-WHtR index of diabetes patients was nonlinearly associated with all-cause mortality and CVD-related mortality (P for nonlinearity =0.15 and 0.70, respectively). The TyG-WHtR index of prediabetes patients was nonlinearly associated with all-cause mortality (P for non-linearity =0.002) but there was an approximately linear association with CVD-related mortality (P for nonlinearity =0.51).

Figure 3 Association between TyG-WHtR index and (A) all-cause mortality and (B) CVD-related mortality in patients with diabetes or prediabetes. Adjustments were made for age, sex, race, education level, family income-to-poverty ratio, smoking status, drinking status, hypertension, BMI, eGFR, HbA1c, TC, HDL-C, uric acid, and BUN. The solid line and green/blue area represent the estimated values and their corresponding 95% CIs, respectively. BMI, body mass index; BUN, blood urea nitrogen; CI, confidence interval; CVD, cardiovascular disease; eGFR, estimated glomerular filtration rate; HbA1c, hemoglobin A1c; HDL-C, high-density lipoprotein cholesterol; HR, hazard ratio; TC, total cholesterol; TyG-WHtR, triglyceride-glucose combined with waist-to-height ratio.

Figure 4 Kaplan-Meier survival analysis curves for mortality. Association between TyG-WHtR index and (A) all-cause mortality and (B) CVD-related mortality in patients with prediabetes. Association between TyG-WHtR index and (C) all-cause mortality and (D) CVD-related mortality in patients with diabetes. The solid line and green/blue area represent the estimated values and their corresponding 95% CIs, respectively. CI, confidence interval; CVD, cardiovascular disease; HR, hazard ratio; TyG-WHtR, triglyceride-glucose combined with waist-to-height ratio.

Subgroup analysis and sensitivity analysis

We conducted subgroup analyses based on various demographic and clinical characteristics to assess the consistency of the association between the TyG-WHtR index and mortality across different populations (Figures 5,6). The subgroups included sex, age (≤60 or >60 years), race (non-Hispanic White or other), BMI (≤30 or >30 kg/m²), drinking status (never or moderate/heavy), smoking status (never or former/current), hypertension (no or yes), diabetes status (prediabetes or diabetes), education level (less than high school or college), and family income-to-poverty ratio (≤3.5 or >3.5). The results from the subgroup analysis showed that the TyG-WHtR index was significantly associated with all-cause mortality risk across all subgroups. However, the association with CVD-related mortality was weaker and less consistent. Notably, the presence of a few significant interaction effects (P<0.05) indicated that the association between the TyG-WHtR index and all-cause mortality and CVD-related mortality differed significantly across specific subgroups, particularly those defined by BMI and household income.

Figure 5 Stratified analyses of the associations between TyG-WHtR and all-cause mortality. BMI, body mass index; CI, confidence interval; HR, hazard ratio; TyG-WHtR, triglyceride-glucose combined with waist-to-height ratio.

Figure 6 Stratified analyses of the associations between TyG-WHtR and CVD-related mortality. BMI, body mass index; CI, confidence interval; CVD, cardiovascular disease; HR, hazard ratio; TyG-WHtR, triglyceride-glucose combined with waist-to-height ratio.

We performed several sensitivity analyses to further test the robustness of our findings. First, we excluded participants who died within the first 2 years of follow-up to minimize the impact of undiagnosed or severe conditions at baseline (Table S1). Even after this exclusion, the association between the TyG-WHtR index and all-cause mortality and CVD-related mortality remained consistent. Second, we excluded participants who had a history of CVD or malignancy at baseline to ensure that the associations were not confounded by pre-existing severe illnesses. The results from this analysis were also robust and consistent with our primary findings (Tables S2,S3). Finally, we excluded participants who were taking hypoglycemic agents or insulin at baseline to avoid confounding due to pharmacological interventions for diabetes management. After this exclusion, the associations between the TyG-WHtR index and all-cause mortality and CVD mortality remained unchanged, except for some minor variations (Table S4). Overall, these subgroup and sensitivity analyses support the generalizability and robustness of our findings across different subpopulations and under various analytical scenarios.

Discussion

In this nationally representative cohort study, we observed that an elevated TyG-WHtR index was associated with increased all-cause mortality in both prediabetes and diabetes populations, irrespective of whether TyG-WHtR was treated as a continuous or categorical variable. This association persisted in the fully adjusted models. Specifically, participants in the highest quartile (Q4) for the TyG-WHtR index had an approximately 1.39-fold increased risk of all-cause death compared to those in the lowest quartile (Q1). Concurrently, higher levels of the TyG-WHtR index significantly predicted CVD mortality in these populations. The RCS curves indicated both linear and nonlinear relationships between the TyG-WHtR index and all-cause and CVD-related mortality across the entire cohort, as well as in the subgroups of prediabetes and diabetes populations. To our knowledge, this is the first study to evaluate the dose–response association between the TyG-WHtR index and mortality risk in adults with prediabetes and diabetes in the US. These findings suggest that the TyG-WHtR index could serve as an effective alternative prognostic biomarker for mortality risk in individuals with diabetes and prediabetes. Furthermore, sensitivity analyses confirmed the robustness of these results.

Notably, our study revealed a distinct inflection point at a TyG-WHtR value of 5.9, beyond which the risk of all-cause mortality increased more steeply. This threshold effect suggests a potential biological mechanism whereby the combined impact of insulin resistance and central adiposity reaches a critical point that significantly compromises physiological function. The identification of such thresholds could have important clinical implications for risk stratification and intervention timing in prediabetes and diabetes populations. Interestingly, the relationship between TyG-WHtR and CVD-related mortality appeared more linear, potentially reflecting different pathophysiological pathways by which insulin resistance and central obesity contribute to cardiovascular versus other causes of mortality.

Previous studies have demonstrated that a high TyG index is an effective early predictor of CVD events (34,35). Additionally, the association between the TyG index and conditions such as heart failure, stroke, renal insufficiency, and nonalcoholic fatty liver disease (NAFLD) has been well-documented (18,36-38). The TyG index, which integrates triglyceride levels and FBG, is a biomarker noted for its ease of measurement and high specificity and sensitivity. Consequently, it has been validated as a surrogate indicator of IR (14,39) although the gold standard for diagnosing IR—the hyperinsulinemic-euglycemic clamp—is infrequently employed in clinical practice due to challenges related to data consistency and reproducibility. The TyG index has exhibited superior predictive capabilities in assessing the prevalence of metabolic disorders, including diabetes and metabolic syndrome (MetS) (40-43). It has also been shown to maintain high accuracy in predicting CVD progression and cardiovascular mortality among adults with diabetes or prediabetes (44,45). A study involving 555 patients with diabetes-related foot disease reported a robust positive correlation between the TyG index and all-cause mortality (46). However, it is crucial to highlight that the relationship between the TyG index and mortality risk remains understudied, with the related findings being inconsistent. For instance, the Prospective Urban Rural Epidemiology (PURE) study, which tracked 141,243 participants across 22 countries over 13.2 years, did not identify a clear association between the TyG index and all-cause or non–cardiovascular-related mortality (47). Similarly, a meta-analysis found no definitive link between the TyG index and all-cause mortality (48). Given the moderate predictive accuracy of the TyG index, we conducted this study to determine whether adjusted TyG-related indices could be used to improve mortality outcome prediction in adults with prediabetes or diabetes. Epidemiological and genetic studies have demonstrated that anthropometric measures indicative of central obesity, such as WHtR, have a stronger association with significant cardiometabolic risk than do general obesity measures such as BMI (23,24). Consequently, integrating the TyG index with these metrics (WHtR and WC) may improve the predictive performance. Other research has established that central obesity is strongly associated with all-cause and CVD-related mortality (49,50). Moreover, the WHtR index is a superior predictor of cardiometabolic health and mortality as compared to other obesity-related measures, particularly among individuals with diabetes (51,52). A recently introduced index that combines the TyG index with WHtR has exhibited higher accuracy in identifying the risk of cardiometabolic disease onset and death than has the TyG index alone (53). Li et al. identified a positive J-shaped correlation between the TyG-WHtR index and all-cause and cause-specific mortality in the general population, with the risk inflection point ranging between 3.55 and 4.25, with the TyG-WHtR index exhibiting a stronger statistical correlation than the TyG index alone (54). Notably, in this study, we observed a weak association between TyG-WHtR and CVD mortality, with significant differences only observed in the unadjusted model and Model 2. Controversy still remains regarding the relationship between TyG-WHtR and CVD mortality. A study focusing solely on prediabetic patients showed that the association between TyG-WHtR and CVD mortality varied with age, and was only significant among prediabetic patients aged <65 years (55). Another study on patients with diabetes and prediabetes found no significant association between TyG-WHtR and CVD mortality, despite adjusting different models (56).

A different study using the NHANES database confirmed the predictive value of the TyG-WHtR index in assessing long-term mortality risk in the general population (25). The findings indicated that the TyG-WHtR index has a critical predictive range of 5.7 to 6.5 for all-cause and CVD-related mortality risk among individuals with prediabetes and diabetes. This suggests significant variability in the TyG-WHtR index’s predictive efficacy across different disease populations.

Although epidemiological studies have demonstrated that the TyG-WHtR index is a robust predictor of all-cause and CVD-related mortality, the underlying biological mechanisms remain unclear. Elevated TyG-WHtR index is generally associated with IR (57,58). As an early sign of IR, a high TyG-WHtR index may influence the body’s proinflammatory status and indicate vulnerability to disease severity and progression. There is extensive evidence showing that IR is associated with chronic low-grade inflammation. In addition, IR has been shown to increase the risk CVD, even in non-diabetes patients. Furthermore, IR is currently considered as a “nontraditional” risk factor contributing to CVD by promoting hypertension, oxidative stress, endothelial dysfunction, dyslipidemia, and type 2 diabetes mellitus. CKD is also considered an IR state (59).

The TyG index, which reflects FBG and lipid levels, is closely linked to systemic inflammation, oxidative stress, and endothelial dysfunction, aligning with the pathophysiological process of IR (60). One study on insulin mediation showed that the indirect effect between insulin-mediated TyG-WHtR and CVD mortality was −23.8% (25), supporting the potential association between TyG-WHtR index levels and CVD-related mortality. Insulin has a protective effect on the heart and its function (61-63). In individuals with diabetes, insulin is widely used to control blood glucose levels. Notably, our study found that the inflection point range of mortality risk in prediabetes and diabetes populations was higher than in the general population, indicating IR in these groups, which is consistent with the increased inflection point range of the TyG-WHtR index. Further analysis of prediabetes and diabetes populations revealed that in the diabetes population, the TyG-WHtR index showed a linear relationship with all-cause and CVD-related mortality. In contrast, in the prediabetes population, the TyG-WHtR index demonstrated a significant nonlinear relationship with all-cause mortality and a linear relationship with CVD-related mortality. Although this dose-response relationship varied across different stages of diabetes, the trend was consistent. These differences may be related to sample size and adjustment covariates, necessitating further research with larger sample sizes.

Our findings have important implications for clinical intervention. This study is the first to reveal an association between the TyG-WHtR index and all-cause and CVD-related mortality in the US population with prediabetes or diabetes. Notably, similar results were obtained in the sensitivity analyses after patients with baseline CVD were excluded, suggesting that the prognostic value of the TyG-WHtR index in prediabetes and diabetes populations is not solely attributable to baseline CVD. We also excluded participants using hypoglycemic drugs and insulin, as their blood glucose levels directly affect the TyG index. Other studies have found that the association between the TyG-WHtR index and CVD-related mortality is higher in individuals without diabetes (64,65), which may be related to the use of hypoglycemic drugs in patients with diabetes influencing TyG levels. Our findings suggest that hypoglycemic drugs may exert an effect on the TyG-WHtR index and CVD mortality. Overall, TyG-WHtR emerges as an effective, direct, easily accessible, and low-cost reliable alternative indicator for poor prognosis. It holds significant clinical significance for the early identification of long-term adverse prognostic risks in adults with prediabetes and diabetes. For clinicians, TyG-WHtR can be incorporated into risk assessment. A threshold of 5.9 has been identified in our study; values above this level are strongly associated with all-cause mortality. When assessing a patient’s risk, clinicians can calculate the TyG-WHtR index. If the TyG-WHtR value exceeds 5.9, it serves as a clear warning sign of a significantly elevated risk of all-cause mortality in patients with prediabetes and diabetes. Considering that TyG-WHtR is calculated using key physiological parameters such as blood glucose, triglycerides, waist circumference, and height, this elevated value reflects the complex interplay between insulin resistance and central obesity, thus guiding clinicians to develop more proactive and precise intervention plans. For instance, they can strengthen the management of blood glucose, lipid levels, and body weight, increase the frequency of follow-up, and promptly adjust treatment strategies to reduce the patients’ risk of all-cause mortality.

However, several limitations in this study should be considered. First, we only used the baseline TyG-WHtR index and were unable to assess the long-term exposure trajectory of fasting glucose, lipids, WC, and height. Second, as we employed an observation design, our findings cannot be used to establish causality. Although we excluded participants who died within 2 years of follow-up, those with baseline malignancy, and those with baseline CVD, reverse causality cannot be entirely ruled out. Third, although various population, socioeconomic, and other relevant factors were accounted for, other potential covariates were not, potentially biasing the results. Fourth, the study population was drawn from the US, which may limit the generalizability of the findings to other populations, especially non-prediabetes or non-diabetes populations. Additionally, while we adjusted for numerous potential confounders, residual confounding factors may persist. The TyG-WHtR index itself, while serving as a practical proxy for insulin resistance and central obesity, does not directly measure insulin action or adipose tissue distribution and functionality. Inflammatory markers and adipokines, which play crucial roles in the pathophysiology linking obesity to cardiometabolic outcomes, were not available for inclusion in our analyses. Furthermore, the generalizability of our findings to non-U.S. populations requires confirmation through studies in diverse ethnic and geographical contexts, particularly given known ethnic variations in body composition and fat distribution patterns.

Conclusions

In this nationally representative large sample of American adults with prediabetes and diabetes, the TyG-WHtR index showed a significant and strong association with all-cause mortality, especially above the threshold of 5.9. These insights have added an increasing amount of evidence to support the clinical application of TyG-WHtR in predicting poor prognosis, providing valuable perspectives for the early risk stratification and intervention strategy development of prediabetes and diabetes populations.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-206/rc

Peer Review File: Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-206/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2025-206/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Ethics approval for the NHANES study program was granted by the Ethics Committee of the National Center for Health Statistics (NCHS), and all participants provided written informed consent. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. International Diabetes Federation. IDF Diabetes Atlas, 11th edn. Brussels, Belgium: 2025. Available online: https://diabetesatlas.org
2. Rooney MR, Fang M, Ogurtsova K, et al. Global Prevalence of Prediabetes. Diabetes Care 2023;46:1388-94. [Crossref] [PubMed]
3. Emerging Risk Factors Collaboration. Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. Lancet 2010;375:2215-22. [Crossref] [PubMed]
4. Echouffo-Tcheugui JB, Perreault L, Ji L, et al. Diagnosis and Management of Prediabetes: A Review. JAMA 2023;329:1206-16. [Crossref] [PubMed]
5. Zheng Y, Ley SH, Hu FB. Global aetiology and epidemiology of type 2 diabetes mellitus and its complications. Nat Rev Endocrinol 2018;14:88-98. [Crossref] [PubMed]
6. Chi ZS, Lee ET, Lu M, et al. Vascular disease prevalence in diabetic patients in China: standardised comparison with the 14 centres in the WHO Multinational Study of Vascular Disease in Diabetes. Diabetologia 2001;44:S82-6. [Crossref] [PubMed]
7. Forouhi NG, Sattar N, Tillin T, et al. Do known risk factors explain the higher coronary heart disease mortality in South Asian compared with European men? Prospective follow-up of the Southall and Brent studies, UK. Diabetologia 2006;49:2580-8. [Crossref] [PubMed]
8. Brannick B, Dagogo-Jack S. Prediabetes and Cardiovascular Disease: Pathophysiology and Interventions for Prevention and Risk Reduction. Endocrinol Metab Clin North Am 2018;47:33-50. [Crossref] [PubMed]
9. Tabák AG, Herder C, Rathmann W, et al. Prediabetes: a high-risk state for diabetes development. Lancet 2012;379:2279-90. [Crossref] [PubMed]
10. Low Wang CC, Hess CN, Hiatt WR, et al. Clinical Update: Cardiovascular Disease in Diabetes Mellitus: Atherosclerotic Cardiovascular Disease and Heart Failure in Type 2 Diabetes Mellitus - Mechanisms, Management, and Clinical Considerations. Circulation 2016;133:2459-502. [Crossref] [PubMed]
11. Joseph JJ, Deedwania P, Acharya T, et al. Comprehensive Management of Cardiovascular Risk Factors for Adults With Type 2 Diabetes: A Scientific Statement From the American Heart Association. Circulation 2022;145:e722-59. [Crossref] [PubMed]
12. Wang M, Sperrin M, Rutter MK, et al. Cancer is becoming the leading cause of death in diabetes. Lancet 2023;401:1849. [Crossref] [PubMed]
13. Kosmas CE, Silverio D, Tsomidou C, et al. The Impact of Insulin Resistance and Chronic Kidney Disease on Inflammation and Cardiovascular Disease. Clin Med Insights Endocrinol Diabetes 2018;11:1179551418792257. [Crossref] [PubMed]
14. Khan SH, Sobia F, Niazi NK, et al. Metabolic clustering of risk factors: evaluation of Triglyceride-glucose index (TyG index) for evaluation of insulin resistance. Diabetol Metab Syndr 2018;10:74. [Crossref] [PubMed]
15. Tura A, Chemello G, Szendroedi J, et al. Prediction of clamp-derived insulin sensitivity from the oral glucose insulin sensitivity index. Diabetologia 2018;61:1135-41. [Crossref] [PubMed]
16. Park HM, Lee HS, Lee YJ, et al. The triglyceride-glucose index is a more powerful surrogate marker for predicting the prevalence and incidence of type 2 diabetes mellitus than the homeostatic model assessment of insulin resistance. Diabetes Res Clin Pract 2021;180:109042. [Crossref] [PubMed]
17. Xiong S, Chen Q, Zhang Z, et al. A synergistic effect of the triglyceride-glucose index and the residual SYNTAX score on the prediction of intermediate-term major adverse cardiac events in patients with type 2 diabetes mellitus undergoing percutaneous coronary intervention. Cardiovasc Diabetol 2022;21:115. [Crossref] [PubMed]
18. Shi W, Xing L, Jing L, et al. Value of triglyceride-glucose index for the estimation of ischemic stroke risk: Insights from a general population. Nutr Metab Cardiovasc Dis 2020;30:245-53. [Crossref] [PubMed]
19. Won KB, Park EJ, Han D, et al. Triglyceride glucose index is an independent predictor for the progression of coronary artery calcification in the absence of heavy coronary artery calcification at baseline. Cardiovasc Diabetol 2020;19:34. [Crossref] [PubMed]
20. Park K, Ahn CW, Lee SB, et al. Elevated TyG Index Predicts Progression of Coronary Artery Calcification. Diabetes Care 2019;42:1569-73. [Crossref] [PubMed]
21. Huang DQ, Mathurin P, Cortez-Pinto H, et al. Global epidemiology of alcohol-associated cirrhosis and HCC: trends, projections and risk factors. Nat Rev Gastroenterol Hepatol 2023;20:37-49. [Crossref] [PubMed]
22. Lavie CJ, Laddu D, Arena R, et al. Healthy Weight and Obesity Prevention: JACC Health Promotion Series. J Am Coll Cardiol 2018;72:1506-31. [Crossref] [PubMed]
23. Ashwell M, Gunn P, Gibson S. Waist-to-height ratio is a better screening tool than waist circumference and BMI for adult cardiometabolic risk factors: systematic review and meta-analysis. Obes Rev 2012;13:275-86. [Crossref] [PubMed]
24. Ross R, Neeland IJ, Yamashita S, et al. Waist circumference as a vital sign in clinical practice: a Consensus Statement from the IAS and ICCR Working Group on Visceral Obesity. Nat Rev Endocrinol 2020;16:177-89. [Crossref] [PubMed]
25. Dang K, Wang X, Hu J, et al. The association between triglyceride-glucose index and its combination with obesity indicators and cardiovascular disease: NHANES 2003-2018. Cardiovasc Diabetol 2024;23:8. [Crossref] [PubMed]
26. Xuan W, Liu D, Zhong J, et al. Impacts of Triglyceride Glucose-Waist to Height Ratio on Diabetes Incidence: A Secondary Analysis of A Population-Based Longitudinal Data. Front Endocrinol (Lausanne) 2022;13:949831. [Crossref] [PubMed]
27. Li X, Sun M, Yang Y, et al. Predictive Effect of Triglyceride Glucose-Related Parameters, Obesity Indices, and Lipid Ratios for Diabetes in a Chinese Population: A Prospective Cohort Study. Front Endocrinol (Lausanne) 2022;13:862919. [Crossref] [PubMed]
28. Johnson CL, Dohrmann SM, Burt VL, et al. National health and nutrition examination survey: sample design, 2011-2014. Vital Health Stat 2 2014;1-33. [PubMed]
29. Zou X, Zhou X, Zhu Z, et al. Novel subgroups of patients with adult-onset diabetes in Chinese and US populations. Lancet Diabetes Endocrinol 2019;7:9-11. [Crossref] [PubMed]
30. American Diabetes Association. 2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2021. Diabetes Care 2021;44:S15-S33. [Crossref] [PubMed]
31. Lim J, Kim J, Koo SH, et al. Comparison of triglyceride glucose index, and related parameters to predict insulin resistance in Korean adults: An analysis of the 2007-2010 Korean National Health and Nutrition Examination Survey. PLoS One 2019;14:e0212963. [Crossref] [PubMed]
32. Alizargar J, Bai CH, Hsieh NC, et al. Use of the triglyceride-glucose index (TyG) in cardiovascular disease patients. Cardiovasc Diabetol 2020;19:8. [Crossref] [PubMed]
33. Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med 2009;150:604-12. [Crossref] [PubMed]
34. Shao Y, Hu H, Li Q, et al. Link between triglyceride-glucose-body mass index and future stroke risk in middle-aged and elderly chinese: a nationwide prospective cohort study. Cardiovasc Diabetol 2024;23:81. [Crossref] [PubMed]
35. Hong S, Han K, Park CY. The triglyceride glucose index is a simple and low-cost marker associated with atherosclerotic cardiovascular disease: a population-based study. BMC Med 2020;18:361. [Crossref] [PubMed]
36. Zhang S, Du T, Zhang J, et al. The triglyceride and glucose index (TyG) is an effective biomarker to identify nonalcoholic fatty liver disease. Lipids Health Dis 2017;16:15. [Crossref] [PubMed]
37. Zhao Q, Zhang TY, Cheng YJ, et al. Impacts of triglyceride-glucose index on prognosis of patients with type 2 diabetes mellitus and non-ST-segment elevation acute coronary syndrome: results from an observational cohort study in China. Cardiovasc Diabetol 2020;19:108. [Crossref] [PubMed]
38. Huang R, Wang Z, Chen J, et al. Prognostic value of triglyceride glucose (TyG) index in patients with acute decompensated heart failure. Cardiovasc Diabetol 2022;21:88. [Crossref] [PubMed]
39. Guerrero-Romero F, Simental-Mendía LE, González-Ortiz M, et al. The product of triglycerides and glucose, a simple measure of insulin sensitivity. Comparison with the euglycemic-hyperinsulinemic clamp. J Clin Endocrinol Metab 2010;95:3347-51. [Crossref] [PubMed]
40. Duan Y, Zhang W, Li Z, et al. Predictive ability of obesity- and lipid-related indicators for metabolic syndrome in relatively healthy Chinese adults. Front Endocrinol (Lausanne) 2022;13:1016581. [Crossref] [PubMed]
41. Kang SW, Kim SK, Kim YS, et al. Risk prediction of the metabolic syndrome using TyG Index and SNPs: a 10-year longitudinal prospective cohort study. Mol Cell Biochem 2023;478:39-45. [Crossref] [PubMed]
42. Moon S, Park JS, Ahn Y. The Cut-off Values of Triglycerides and Glucose Index for Metabolic Syndrome in American and Korean Adolescents. J Korean Med Sci 2017;32:427-33. [Crossref] [PubMed]
43. Chamroonkiadtikun P, Ananchaisarp T, Wanichanon W. The triglyceride-glucose index, a predictor of type 2 diabetes development: A retrospective cohort study. Prim Care Diabetes 2020;14:161-7. [Crossref] [PubMed]
44. Zhang Q, Xiao S, Jiao X, et al. The triglyceride-glucose index is a predictor for cardiovascular and all-cause mortality in CVD patients with diabetes or pre-diabetes: evidence from NHANES 2001-2018. Cardiovasc Diabetol 2023;22:279. [Crossref] [PubMed]
45. Liu C, Liang D. The association between the triglyceride-glucose index and the risk of cardiovascular disease in US population aged ≤ 65 years with prediabetes or diabetes: a population-based study. Cardiovasc Diabetol 2024;23:168. [Crossref] [PubMed]
46. Huang X, Han J, Nong Y, et al. Triglyceride-glucose index is strongly associated with all-cause mortality in elderly females with diabetic foot ulcers: A 9-year follow-up study. Int Wound J 2024;21:e14344. [Crossref] [PubMed]
47. Lopez-Jaramillo P, Gomez-Arbelaez D, Martinez-Bello D, et al. Association of the triglyceride glucose index as a measure of insulin resistance with mortality and cardiovascular disease in populations from five continents (PURE study): a prospective cohort study. Lancet Healthy Longev 2023;4:e23-33. [Crossref] [PubMed]
48. Liu X, Tan Z, Huang Y, et al. Relationship between the triglyceride-glucose index and risk of cardiovascular diseases and mortality in the general population: a systematic review and meta-analysis. Cardiovasc Diabetol 2022;21:124. [Crossref] [PubMed]
49. Mehran L, Amouzegar A, Fanaei SM, et al. Anthropometric measures and risk of all-cause and cardiovascular mortality: An 18 years follow-up. Obes Res Clin Pract 2022;16:63-71. [Crossref] [PubMed]
50. Min YI, Gao Y, Anugu P, et al. Obesity and overall mortality: findings from the Jackson Heart Study. BMC Public Health 2021;21:50. [Crossref] [PubMed]
51. Parente EB, Mutter S, Harjutsalo V, et al. Waist-height ratio and waist are the best estimators of visceral fat in type 1 diabetes. Sci Rep 2020;10:18575. [Crossref] [PubMed]
52. Schneider HJ, Friedrich N, Klotsche J, et al. The predictive value of different measures of obesity for incident cardiovascular events and mortality. J Clin Endocrinol Metab 2010;95:1777-85. [Crossref] [PubMed]
53. Raimi TH, Dele-Ojo BF, Dada SA, et al. Triglyceride-Glucose Index and Related Parameters Predicted Metabolic Syndrome in Nigerians. Metab Syndr Relat Disord 2021;19:76-82. [Crossref] [PubMed]
54. Li S, An L, Fu Z, et al. Association between triglyceride-glucose related indices and all-cause and cause-specific mortality in the general population: a cohort study. Cardiovasc Diabetol 2024;23:286. [Crossref] [PubMed]
55. Zhang L, Zhao W, Zheng P, et al. Association of triglyceride-glucose-related indices with all-cause and cause-specific mortality in individuals with prediabetesss. Cardiovasc Diabetol 2025;24:330. [Crossref] [PubMed]
56. Yan Z, Chang X, Liu Z, et al. The association of obesity and lipid-related indicators with all-cause and cardiovascular mortality risks in patients with diabetes or prediabetes: a cross-sectional study based on machine learning algorithms. Front Endocrinol (Lausanne) 2025;16:1492082. [Crossref] [PubMed]
57. Ormazabal V, Nair S, Elfeky O, et al. Association between insulin resistance and the development of cardiovascular disease. Cardiovasc Diabetol 2018;17:122. [Crossref] [PubMed]
58. Ren J, Wu NN, Wang S, et al. Obesity cardiomyopathy: evidence, mechanisms, and therapeutic implications. Physiol Rev 2021;101:1745-807. [Crossref] [PubMed]
59. Kosmas CE, Rodriguez Polanco S, Bousvarou MD, et al. The Triglyceride/High-Density Lipoprotein Cholesterol (TG/HDL-C) Ratio as a Risk Marker for Metabolic Syndrome and Cardiovascular Disease. Diagnostics (Basel) 2023;13:929. [Crossref] [PubMed]
60. Li T, Wang P, Wang X, et al. Inflammation and Insulin Resistance in Diabetic Chronic Coronary Syndrome Patients. Nutrients 2023;15:2808. [Crossref] [PubMed]
61. Yazar A, Polat G, Un I, et al. Effects of glibenclamide, metformin and insulin on the incidence and latency of death by oubain-induced arrhythmias in mice. Pharmacol Res 2002;45:183-7. [Crossref] [PubMed]
62. Sinha AK, Bhattacharya S, Acharya K, et al. Stimulation of nitric oxide synthesis and protective role of insulin in acute thrombosis in vivo. Life Sci 1999;65:2687-96. [Crossref] [PubMed]
63. Xi L, Kukreja RC. Pivotal role of nitric oxide in delayed pharmacological preconditioning against myocardial infarction. Toxicology 2000;155:37-44. [Crossref] [PubMed]
64. Liu L, Wu Z, Zhuang Y, et al. Association of triglyceride-glucose index and traditional risk factors with cardiovascular disease among non-diabetic population: a 10-year prospective cohort study. Cardiovasc Diabetol 2022;21:256. [Crossref] [PubMed]
65. Sánchez-Íñigo L, Navarro-González D, Fernández-Montero A, et al. The TyG index may predict the development of cardiovascular events. Eur J Clin Invest 2016;46:189-97. [Crossref] [PubMed]

(English Language Editor: J. Gray)