KES
Articles
- Page Path
- HOME > Endocrinol Metab > Ahead-of print > Article
-
Original ArticleAssociation between the Triglyceride-Glucose Index and Cardiovascular Risk and Mortality across Different Diabetes Durations: A Nationwide Cohort Study
-
Jeongeun Kwak1*
, Kyung-Do Han2*, Eun Young Lee3, Seung-Hwan Lee3, Dong-Jun Lim3, Hyuk-Sang Kwon4, Jeongmin Lee1 -
DOI: https://doi.org/10.3803/EnM.2024.2205
Published online: March 5, 2025
1Division of Endocrinology and Metabolism, Department of Internal Medicine, Eunpyeong St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
2Department of Statistics and Actuarial Science, Soongsil University, Seoul, Korea
3Division of Endocrinology and Metabolism, Department of Internal Medicine, Seoul St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
4Division of Endocrinology and Metabolism, Department of Internal Medicine, Yeouido St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, Seoul, Korea
- Corresponding author: Jeongmin Lee Division of Endocrinology and Metabolism, Department of Internal Medicine, Eunpyeong St. Mary’s Hospital, College of Medicine, The Catholic University of Korea, 1021 Tongil-ro, Eunpyeong-gu, Seoul 03312, Korea Tel: +82-2-2030-4349, Fax: +82-2-2030-4641, E-mail: 082mdk45@catholic.ac.kr
- *These authors contributed equally to this work.
• Received: October 15, 2024 • Revised: November 29, 2024 • Accepted: January 3, 2025
Copyright © 2025 Korean Endocrine Society
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
- 768 Views
- 50 Download
ABSTRACT
-
Background
- We aimed to assess the association between triglyceride-glucose (TyG) index and cardiovascular disease (CVD) risk and mortality in a large cohort of diabetes patients.
-
Methods
- A retrospective cohort study of 1,090,485 participants from the Korean National Health Insurance Service database was conducted. Participants were stratified into TyG quartiles.
-
Results
- Higher TyG index quartiles were significantly associated with an increased CVD risk and mortality risk. In fully adjusted models, participants in the highest TyG quartile (Q4) had an 18% higher risk of CVD (hazard ratio [HR], 1.18; 95% confidence interval [CI], 1.13 to 1.23) and a 16% higher risk of mortality (HR, 1.16; 95% CI, 1.11 to 1.23) compared to those in the lowest quartile (Q1). The association was particularly pronounced in patients with fasting glucose ≥126 mg/dL (CVD [HR, 1.33; 95% CI, 1.29 to 1.37], mortality [HR, 1.23; 95% CI, 1.20 to 1.26]; P for interaction <0.001). Patients with a diabetes duration of ≥10 years showed the strongest association between the TyG index and CVD risk (HR, 1.44; 95% CI, 1.38 to 1.50), while the mortality risk was particularly elevated in those with a diabetes duration of less than 5 years (HR, 1.23; 95% CI, 1.18 to 1.30). Subgroup analyses revealed stronger associations between TyG index and CVD risk in younger participants, non-obese individuals, and non-smokers.
-
Conclusion
- The TyG index is a significant predictor of CVD and mortality in diabetic patients, particularly in those with poor glycemic control or longer disease duration.
- Cardiovascular disease (CVD) is the leading cause of death globally and poses an even greater threat to individuals with diabetes mellitus (DM) [1,2]. The global rise in DM prevalence has heightened the importance of identifying accurate and practical tools for CVD risk prediction in this high-risk population [3]. Traditional risk factors, such as hyperglycemia, hypertension, and dyslipidemia, are established contributors to CVD. However, these factors alone do not fully capture the underlying metabolic disturbances—particularly insulin resistance (IR)—that are intricately linked to increased cardiovascular risk in DM patients [4].
- IR is a key pathological feature of type 2 DM and plays a central role in the development of both DM and CVD. It drives a cascade of metabolic abnormalities, including hyperglycemia, dyslipidemia, and inflammation, all of which promote endothelial dysfunction and accelerate atherosclerosis [5]. In clinical practice, IR can be estimated using markers such as the homeostasis model assessment of insulin resistance (HOMA-IR), which is a reliable method used in both research and routine care [6]. Although HOMA-IR is valuable for assessing IR, its usefulness is limited in populations with preserved β-cell function or significant insulin secretion defects, where the correlation between insulin sensitivity and HOMA-IR can be less accurate [7]. Furthermore, HOMA-IR requires precise insulin measurement, which can vary depending on assay standardization, making it less robust across diverse settings.
- Given these limitations, simpler and more universally applicable surrogate markers for IR have garnered increasing attention. The triglyceride-glucose (TyG) index, derived from fasting triglycerides (TG) and fasting glucose (FG) levels, has emerged as a promising alternative [8]. It is easily calculated and correlates well with IR. Recent study has demonstrated that the TyG index not only reflects IR but also predicts CVD outcomes, such as myocardial infarction (MI) and stroke, in both DM and non-DM populations [3]. Despite this growing evidence, its prognostic utility in DM patients across varying disease durations remains underexplored.
- This study investigates the association between the TyG index and the risk of both CVD and mortality in a nationwide cohort of DM patients. Specifically, we will assess how the TyG index predicts CVD and mortality outcomes across different durations of DM.
INTRODUCTION
- Data collection
- This retrospective cohort study used data from the Korean National Health Insurance Service (NHIS) database, which is part of South Korea’s single, universal healthcare system, covering nearly the entire population. NHIS enrollees are recommended to undergo medical checkups at least once every 2 years. The NHIS database consists of a qualification database (containing information on sex, age, income, area of residence, and qualification type), a claims database (providing details on treatment specifications, consultation records, diagnosis codes according to the International Classification of Diseases 10th revision [ICD-10], and prescription records), a health checkup database (containing general health examination results and lifestyle/behavior questionnaire responses), and death information [9].
- Participants
- Of the 2,616,505 participants who underwent health examinations between January 2015 and December 2016, individuals were excluded based on the following criteria: those taking lipid-lowering drugs (n=1,334,205), those with a history of MI before the index date (n=39,535), those with a prior stroke diagnosis (n=97,027), those with missing data (n=39,444), and those who died within 1 year after the health screening (n=15,809) to reduce the risk of reverse causality. Ultimately, 1,090,485 participants were included in the analysis and followed from baseline until December 31, 2022. This study complied with the ethical standards of the Declaration of Helsinki, and the protocol using secondary data was approved by the Eunpyeong St. Mary’s Hospital Institutional Review Board of Catholic Medical Center, Catholic University of Korea (IRB approval No. PC24ZASI0127). Researchers can access the NHIS database after obtaining approval from the official review committee. Written informed consent was waived due to the use of de-identified, previously collected data.
- Demographic variables and measurement
- Demographic and lifestyle data were collected using a standardized self-reported questionnaire, covering variables such as smoking status (current smoker, defined as having consumed at least five packs or 100 cigarettes and currently smoking), alcohol consumption (≥30 g/day for males and ≥20 g/day for females), and regular exercise (≥20 minutes of vigorous-intensity physical activity at least 3 days per week, or ≥20 minutes of moderate-intensity physical activity at least 5 days per week). Income level was dichotomized, with the lower 25% representing the lower-income group. Systolic blood pressure (SBP) and diastolic blood pressure (DBP) were measured with participants seated after at least 5 minutes of rest. Body weight, height, and waist circumference (WC) were directly measured at each visit, and body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters (kg/m2). Laboratory tests included serum FG, alanine aminotransferase, aspartate aminotransferase, and γ-glutamyl transpeptidase, total cholesterol (TC), low-density lipoprotein cholesterol, high-density lipoprotein cholesterol (HDL-C), and TG which were measured after an overnight fast of at least 8 hours. The presence of DM was defined as FG ≥126 mg/dL at least one claim per year under ICD-10 codes E10–14 and the prescription of anti-diabetic medication. Hypertension was defined as SBP/DBP ≥140/90 mm Hg or at least one claim per year under ICD-10 codes I10–13 or I15 and the prescription of antihypertensive agents. Dyslipidemia was defined as TC ≥240 mg/dL or at least one claim per year under ICD-10 code E78 and the prescription of lipid-lowering drugs.
- Chronic kidney disease (CKD) was defined by an estimated glomerular filtration rate of <60 mL/min/1.73 m2, calculated using the Modification of Diet in Renal Disease formula: 186×(serum creatinine)−1.154×age−0.203×0.742 (if female) [10].
- Definition of TyG
- The TyG index was calculated using the following equation:
-
- Definition of study outcomes
- The primary outcomes of this study were CVD diagnosis and death. CVD was defined as the first occurrence of either MI (ICD-10 codes I21–I22) or stroke (ICD-10 codes I63–I64). These diagnoses were confirmed through claims for brain magnetic resonance imaging or brain computed tomography scans during hospitalization.
- Statistical analysis
- Continuous variables are presented as mean±standard deviation, while categorical variables are expressed as numbers and percentages. An independent t test was used to compare continuous variables, and the chi-squared test was employed for categorical variables. Annual trends in the number of cases and incidence were analyzed using the chi-square test for trends within age-gender strata. The incidence rate of thyroid cancer was calculated as the number of events divided by the total follow-up duration, expressed per 1,000 person-years. Cox proportional hazard models were used to estimate hazard ratios (HRs) and their corresponding 95% confidence intervals (CIs) to evaluate the risk of CVD and death across different TyG index quartiles. Significance was assessed using the log-rank test. Model 1 was a crude model providing unadjusted HR estimates for CVD and death. Model 2 was adjusted for age and sex, while Model 3 further adjusted for age, sex, BMI, smoking status, alcohol consumption, regular exercise, income level, hypertension, CKD, the use of three or more oral anti-diabetic agents, and insulin therapy. Model 4 was the fully adjusted model, which included all factors from Model 3 as well as TC. Subgroup analyses were performed by stratifying participants based on sex, age, obesity, smoking status, and alcohol consumption, with interaction testing conducted using the likelihood ratio test. All statistical analyses were conducted using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA), and a P value of <0.05 was considered statistically significant.
METHODS
- The baseline characteristics according to TyG quartiles
- A total of 1,090,485 participants were analyzed with a mean follow-up of 5.64 years. The mean age was 56.96±12.47 years in total participants, with younger participants (<40 years) more prevalent in higher TyG quartiles (6.37% in Q1, 13.86% in Q4) (Table 1). Conversely, participants aged ≥50 years were concentrated in lower quartiles (78.85% in Q1, 55.94% in Q4). Men comprised 68.55% of the cohort. Most participants had new-onset DM or disease duration of less than 5 years. The proportion of participants using three or more oral anti-diabetic agents was 15.06%, decreasing from 16.32% in Q1 to 13.9% in Q4. Insulin use also decreased from 8.34% in Q1 to 4.35% in Q4.
- Obesity prevalence increased from 34.5% in Q1 to 60.48% in Q4, along with increases in body weight, BMI, and WC. SBP and DBP also increased, while hypertension prevalence remained similar across quartiles. The prevalence of hypertension and chronic CKD was lower in Q4. Regular exercise was more common in Q1, while smoking and high-risk alcohol consumption were more frequent in Q4. Lipid profiles and FG levels changed with higher TyG quartiles; HDL-C levels decreased from 56.83 mg/dL in Q1 to 46.3 mg/dL in Q4, while FG, TC, and TG levels increased.
- Association between TyG quartiles and CVD risk or mortality
- The relationship between the TyG index and CVD outcomes, including MI, stroke, and all-cause mortality, was analyzed across quartiles in the overall cohort (Table 2). A higher TyG index was consistently associated with an increased risk of CVD. In the fully adjusted model (model 4), individuals in the highest quartile (Q4) had a HR of 1.28 (95% CI, 1.25 to 1.31) for CVD compared to the lowest quartile (Q1), with a clear upward trend observed across quartiles.
- A higher TyG index also showed a significant association with increased risk of MI or stroke after adjustment for all covariates. The HR for MI in Q4 was 1.22 (95% CI, 1.17 to 1.26), and for stroke, it was 1.34 (95% CI, 1.29 to 1.38), compared to Q1.
- CVD mortality also increased progressively across TyG quartiles. However, a significant association with CVD mortality was observed only in the highest quartile (Q4), with an HR of 1.19 (95% CI, 1.16 to 1.22) in the fully adjusted model.
- Association between TyG quartiles and CVD risk according to DM duration
- The association between TyG index quartiles and CVD incidence was analyzed across different durations of DM, using Q1 as the reference group (Table 3, Fig. 1A). The incidence rate of CVD per 1,000 person-years varied across TyG quartiles and DM duration categories. Among patients with new-onset DM, the incidence rate was 6.03 in Q1 and 6.25 in Q4. For patients with a DM duration of less than 5 years, the incidence rate ranged from 9.21 in Q1 to 9.06 in Q4, while for those with a DM duration of 5 to 9 years, it increased from 11.83 in Q1 to 13.14 in Q4. In patients with a DM duration of 10 years or more, the IR was 14.91 in Q1 and 19.47 in Q4.
- CVD risk increased with both higher TyG quartiles and longer DM duration. In fully adjusted model with age, sex, BMI, income, smoking status, alcohol consumption, physical activity, hypertension, CKD, and TC level (model 4), CVD risk in new-onset DM was highest in TyG quartile (Q4) (HR, 1.18; 95% CI, 1.13 to 1.23). In patients with a DM duration of less than 5 years, the adjusted HR (aHR) for CVD risk was 1.23 (95% CI, 1.16 to 1.29) in Q4. Among those with a DM duration of 5 to 10 years, aHR for CVD risk was 1.34 (95% CI, 1.27 to 1.41) in Q4. For patients with a DM duration of 10 years or more, aHR for CVD risk was highest in Q4 at 1.44 (95% CI, 1.38 to 1.50).
- Additional analyses, using the lowest quartile of the TyG index in individuals with new-onset DM as the reference group, further supported these findings (Supplemental Table S1). In these analyses, the HR for CVD in the highest TyG quartile among patients with a DM duration of 10 years or more increased to 3.24 (95% CI, 3.09 to 3.39). Even after full adjustment, the HR in this group remained significantly elevated at 1.87 (95% CI, 1.78 to 1.97; P for interaction <0.001).
- Risk of MI or stroke according to TyG index and DM duration
- The impact of TyG index quartiles on MI incidence varied according to DM duration. In new-onset diabetes, the incidence rate increased from 2.86 to 3.23 per 1,000 person-years, with an aHR of 1.18 (95% CI, 1.11 to 1.26) for Q4 vs. Q1. In patients with less than 5 years of DM, the incidence rate remained stable (4.44 vs. 4.35), but the aHR rose to 1.12 (95% CI, 1.04 to 1.21). In the 5- to 9-year group, the incidence rate showed an increase (5.66 to 5.93), with an aHR of 1.17 (95% CI, 1.08 to 1.26). For those with a DM duration of ≥10 years, the incidence rate increased from 6.80 to 9.14, with a corresponding aHR of 1.40 (95% CI, 1.32 to 1.50).
- Stroke incidence showed a similar pattern. In new-onset DM, the incidence rate was stable (3.40 in Q1 vs. 3.23 in Q4), but the aHR for Q4 was 1.17 (95% CI, 1.02 to 1.24). For patients with less than 5 years of DM, the incidence rate remained steady (5.16 vs. 5.07), but the aHR rose to 1.32 (95% CI, 1.23 to 1.42). In the 5- to 9-year group, the incidence rate increased from 6.68 to 7.73, with an aHR of 1.49 (95% CI, 1.40 to 1.60). For those with ≥10 years of DM, the IR rose from 8.78 to 11.30, and the aHR reached 1.49 (95% CI, 1.40 to 1.57) (Supplemental Table S2).
- Association between TyG index and CVD mortality according to DM duration
- The association between TyG index quartiles and mortality was evaluated across different diabetes durations, using the lowest quartile (Q1) as the reference (Table 4, Fig. 1B). In new-onset DM, the IR dropped from 8.75 in Q1 to 5.11 in Q4. Similarly, for DM duration less than 5 years, the IR fell from 13.67 to 8.03, for 5–9 years from 17.64 to 12.02, and for ≥10 years from 22.90 to 18.97, indicating a clear downward trend in incidence rates as TyG index quartiles increased.
- Despite these reductions in incidence, the aHR for mortality consistently increased with higher TyG quartiles in all DM duration groups. In new-onset DM, Q4 had the highest mortality risk (aHR, 1.09; 95% CI, 1.04 to 1.13). The aHR increased to 1.17 (95% CI, 1.11 to 1.23) for those with DM duration <5 years, 1.23 (95% CI, 1.18 to 1.30) for 5–9 years, and 1.29 (95% CI, 1.24 to 1.34) for ≥10 years, indicating a clear trend of increasing risk with longer disease duration.
- Further analysis confirmed these findings, showing that in patients with ≥10 years of DM, the aHR for CVD mortality in Q4 was markedly higher (aHR, 2.17; 95% CI, 2.08 to 2.27), and remained elevated even after adjusting for all covariates (aHR, 1.15; 95% CI, 1.10 to 1.20; P interaction <0.001) (Supplemental Table S3).
- Association between TyG index and CVD or CVD mortality according to glycemic control
- The association between the TyG index and CVD or CVD mortality was assessed across subgroups stratified by glycemic control, including FG levels, insulin use, and the number of oral anti-diabetic drugs (OADs) (Supplemental Table S4).
- There was no significant difference according to insulin use or number of OADs. Non-insulin users had an aHR for CVD in Q4 vs. Q1 of 1.27 (95% CI, 1.24 to 1.31), which was comparable to insulin users (aHR, 1.31; 95% CI, 1.22 to 1.41; P for interaction=0.596). Similarly, patients on fewer than three OADs showed an aHR of 1.27 (95% CI, 1.24 to 1.31), which was consistent with those on three or more agents (aHR, 1.29; 95% CI, 1.22 to 1.35; P for interaction=0.832). Patients with FG levels ≥126 mg/dL showed a stronger association between the TyG index and CVD (aHR for Q4 vs. Q1, 1.33; 95% CI, 1.29 to 1.37) compared to those with FG levels <126 mg/dL (aHR, 1.13; 95% CI, 1.05 to 1.22; P for interaction <0.001).
- The association between the TyG index and CVD mortality showed consistent results across subgroups stratified by glycemic control. There was no significant difference based on insulin use or the number of OADs. In those with FG levels ≥126 mg/dL, the aHR for Q4 vs. Q1 was 1.23 (95% CI, 1.20 to 1.26), while no significant association was observed in the <126 mg/dL group (aHR, 0.97; 95% CI, 0.90 to 1.05; P for interaction <0.001).
- Subgroup analyses
- Subgroup analyses were performed based on age, sex, obesity, smoking status, alcohol consumption, and regular exercise. Across all subgroups, a higher TyG index was consistently associated with an increased CVD risk.
- In sex-specific analyses, both men and women with longer DM duration in the highest TyG quartile had elevated CVD risks (men: aHR, 1.35; women: aHR, 1.51; P interaction=0.085) (Supplemental Fig. S1). Age-stratified analysis showed that younger patients (20–39 years) had a significantly higher CVD risk in Q4 compared to older patients (40–49 or ≥50 years) regardless to DM duration (Supplemental Fig. S2). When stratified by obesity status, non-obese individuals showed a stronger association between the TyG index and CVD risk. The highest TyG quartile was associated with a higher CVD risk in non-obese individuals (aHR, 1.24) and those with obesity (aHR, 1.09; P for interaction=0.009) (Supplemental Fig. S3). Non-smokers and non-drinkers exhibited a stronger association between higher TyG index levels and CVD risk. For individuals in the highest quartile (Q4) with longer DM duration, non-smokers (aHR, 1.45), non-drinkers (aHR, 1.42) and those who engaged in regular exercise (aHR, 1.51) showed a greater increase in CVD risk compared to smokers and drinkers (Supplemental Figs. S4-S6).
RESULTS
CVD risk
CVD mortality
- This large, population-based cohort study demonstrates a significant association between the TyG index and the risk of CVD and mortality in patients with DM. Our findings reveal that higher TyG index quartiles are strongly predictive of increased risk for MI, stroke, and mortality. The results show that the TyG index is a robust predictor of adverse outcomes across various stages of DM, including new-onset and long-standing DM cases. The association between the TyG index and these outcomes was stronger in patients with longer DM duration, higher FG levels (≥126 mg/dL), younger individuals, and non-obese populations. These results support the TyG index as a simple and effective tool for assessing metabolic risk and CVD outcomes, independent of traditional CVD risk factors.
- The TyG index, derived from TG and FG, serves as a surrogate marker for IR. IR is a key pathophysiological process in type 2 DM and contributes directly to the development of atherosclerosis through multiple pathways [12]. IR contributes to atherosclerosis development through dyslipidemia, endothelial dysfunction, inflammation, and oxidative stress [13-15]. Elevated TG levels and hyperglycemia reflect an advanced metabolic state characterized by impaired lipid metabolism, chronic inflammation, and increased production of reactive oxygen species, which together promote endothelial damage and atherosclerosis progression [16,17].
- Previous studies have consistently demonstrated a strong association between an elevated TyG index and increased risks of DM and CVD [18-20]. However, the role of the TyG index in predicting CVD outcomes in patients with established DM remains controversial. Our study shows that the highest TyG quartile (Q4) was associated with the greatest risk of CVD outcomes, including MI and stroke. This aligns with findings from previous Taiwan study, which demonstrated that a high TyG index was associated with a 1.52-fold increased risk of CVD risk, including MI and stroke in patients with DM [21]. Conversely, Sanchez-Inigo et al. [22] reported no significant association between TyG index and 10-year CVD risk. In those with a longer duration of DM, the cumulative burden of chronic hyperglycemia and IR exacerbates cardiovascular risk [23]. Prolonged exposure to elevated glucose and TG levels leads to persistent vascular inflammation, endothelial dysfunction, and arterial stiffness, promoting plaque formation and rupture, thus increasing the risk of MI and stroke [24]. The progressive nature of IR in long-standing DM further compounds this metabolic dysfunction, as reflected in the higher HRs for CVD in patients with elevated TyG index levels.
- Our study investigated the relationship between the TyG index and cardiovascular outcomes by stratifying patients based on glycemic control. Unlike previous studies, which primarily focused on overall diabetes populations, our analysis highlights the stronger association of the TyG index with CVD and mortality risk in patients with FG levels ≥126 mg/dL, compared to those with levels <126 mg/dL. These findings indicate the compounding effects of hyperglycemia on IR and its downstream consequences on CVD risk. In contrast, the predictive value of the TyG index remained consistent across subgroups defined by insulin use or the number of OAD, suggesting its reliability as a marker of metabolic risk regardless of treatment intensity.
- The impact of age on the association between the TyG index and CVD risk remains uncertain. A study from China highlighted the predictive value of the TyG index for CVD risk among adults aged 45 years or above [25]. Additionally, research conducted on individuals aged 65 and older demonstrated strong associations between the TyG index and both all-cause and CVD mortality [26]. In comparison, the association between a higher TyG index and CVD risk or mortality was particularly strong in younger individuals in our study. Mechanistically, younger DM patients in the higher TyG quartiles may experience more pronounced metabolic disturbances at an earlier age, leading to accelerated vascular damage [27]. The early onset of IR could result in longer exposure to elevated glucose and lipid levels, which accelerates the development of atherosclerotic plaques and increases CVD risk [28]. This is reflected in our finding that younger individuals (aged 20 to 39 years) with the highest TyG index had nearly double the risk of CVD compared to those in the lowest quartile. These results were consistent with those of a recent study based on United States population data [29]. Liu and Liang [29] demonstrated a positive association between the TyG index and CVD risk for individuals under 65 years of age with prediabetes or DM (odds ratio,1.65; 95% CI, 1.20 to 2.25). Although our study focused on a DM population, our results align with previous research in non-DM cohorts, which also found a stronger association between the TyG index and CVD incidence in younger individuals [30]. These findings suggest that the TyG index may serve as an early indicator of metabolic disturbances in younger individuals, where traditional risk factors such as hypertension and hyperlipidemia are less prevalent. In older populations, the cumulative effect of these conventional risk factors may reduce the sensitivity of TyG index in predicting cardiovascular risk associated with IR. Additionally, when used to assess IR in both DM and non-DM individuals, the TyG index was more reliable than the HOMA-IR [8].
- Both men and women exhibited a consistent association between higher TyG index levels and CVD risk, with a slightly stronger trend observed in women, aligning with findings from a previous Chinese study that focused on new-onset DM [31]. This suggests that while the metabolic disturbances reflected by the TyG index impact both sexes, women may experience a more pronounced effect due to factors like increased hepatocellular lipid content contributing to higher IR [32]. However, unlike the previous study, our analysis included individuals with varying durations of DM. Despite the broader range of DM duration in our study, the TyG index remained a strong predictor of CVD risk across both sexes.
- The previous research utilizing National Health and Nutrition Examination Survey data demonstrated that a strong association between the TyG index and CVD events was particularly pronounced in obese individuals, where a high TyG index increased the risk of CVD by 63% and all-cause mortality by 32%, primarily in obese young and middle-aged adults [33]. In contrast, our study demonstrates that the TyG index is a strong predictor, particularly in non-obese DM individuals. Our finding that the TyG index was a stronger predictor of CVD risk in non-obese individuals may be partly explained by the complex metabolic roles of adipose tissue and the phenomenon known as the ‘obesity paradox,’ where adipose tissue, particularly through adipokines, exerts protective effects on glucose and lipid metabolism, reducing oxidative stress and inflammation [34,35]. In non-obese individuals, lower levels of adiponectin may result in reduced protection against IR and metabolic disturbances, making them more vulnerable to CVD risk [36]. In new-onset and early-stage DM patients, non-obese individuals in the highest TyG quartile exhibited a higher cardiovascular risk compared to the obese individuals. For patients with a longer duration of DM (≥ 10 years), the predictive power of the TyG index was consistent across both obese and non-obese individuals, indicating that prolonged IR and hyperglycemia lead to similarly elevated CVD risk, regardless of BMI.
- Our findings also demonstrated that non-smokers and non-drinkers exhibited a stronger relationship between higher TyG index levels and CVD risk. This could be due to the fact that smoking and alcohol consumption independently contribute to CVD risk, potentially diluting the relative contribution of IR.
- A major strength of our study is that it is the first to investigate the association between the TyG index and CVD outcomes across different durations of DM. By stratifying our cohort based on DM duration, we were able to demonstrate that the TyG index remains a strong predictor of CVD events not only in individuals with new-onset DM but also in those with long-standing DM. The large sample size and the use of a nationwide cohort provide a robust, representative dataset that minimizes selection bias and allows for the generalization of our results to a broader DM population. The longitudinal nature of the study, with a mean follow-up period of 5.64 years, offers a comprehensive view of how the TyG index predicts long-term CVD outcomes. Additionally, by incorporating subgroup analyses based on age, sex, obesity, and lifestyle factors, our study adds nuance to the understanding of how IR contributes to CVD risk across different demographic and clinical subgroups.
- Despite these strengths, our study also has certain limitations. One significant limitation is the reliance on a single point measurement of the TyG index and other clinical variables, which limits our ability to assess changes in metabolic status over time. Longitudinal measurements could provide more dynamic insights into the progression of IR and CVD risk. The absence of glycosylated hemoglobin data, while mitigated by FG levels, remains a limitation for assessing overall glycemic control. Additionally, the study population is composed primarily of South Korean individuals, which may limit the generalizability of the findings to other ethnic groups with different genetic backgrounds and environmental exposures. Another limitation of our study is that we focused exclusively on MI and stroke as primary CVD outcomes, without including other important CVD events such as heart failure, peripheral artery disease, or arrhythmias. While MI and stroke represent significant and clinically relevant endpoints, expanding the scope to include other CVD events could provide a more comprehensive assessment of the TyG index’s predictive value for broader CVD risk.
- In conclusion, our study highlights the TyG index as a strong predictor of cardiovascular events in both new-onset and long-standing DM patients with poor glycemic control. The TyG index effectively identifies cardiovascular risk across various subgroups, including age, sex, obesity, and lifestyle factors, emphasizing its relevance beyond traditional risk factors. The TyG index should be considered a valuable tool for identifying high-risk diabetic patients and guiding early interventions. Longitudinal studies should further investigate its prognostic utility across broader populations and explore potential interventions aimed at reducing TyG index levels to mitigate cardiovascular risk in DM individuals.
DISCUSSION
Supplementary Material
Supplemental Table S1.
Supplemental Table S2.
Supplemental Table S3.
Supplemental Table S4.
Supplemental Fig. S1.
Supplemental Fig. S2.
Supplemental Fig. S3.
Supplemental Fig. S4.
Supplemental Fig. S5.
Supplemental Fig. S6.
-
CONFLICTS OF INTEREST
No potential conflict of interest relevant to this article was reported.
-
ACKNOWLEDGMENTS
This study was performed using the database from the National Health Insurance System and the results do not necessarily represent the opinion of the National Health Insurance Corporation.
-
AUTHOR CONTRIBUTIONS
Conception or design: J.K., K.D.H., E.Y.L., S.H.L., D.J.L., H.S.K., J.L. Acquisition, analysis, or interpretation of data: J.L, K.D.H., E.Y.L., S.H.L., D.J.L., H.S.K., J.L. Drafting the work or revising: J.K., J.L. Final approval of the manuscript: J.L.
Article information
Fig. 1.(A) Association between triglyceride-glucose (TyG) quartiles and cardiovascular disease (CVD) risk according to diabetes mellitus (DM) duration. (B) Association between TyG quartiles and CVD mortality according to DM duration.
Table 1.The Baseline Characteristics according to TyG Quartiles
Values are expressed as mean±standard deviation, number (%), or geometric mean (95% confidence interval). P values for the trend were <0.001 for all variables due to large size of the study population.
TyG, triglyceride-glucose; DM, diabetes mellitus; BMI, body mass index; CKD, chronic kidney disease; SBP, systolic blood pressure; DBP, diastolic blood pressure; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate.
Table 2.Association between TyG Index Quartiles and CVD Outcomes
Model 1: Unadjusted; Model 2: Adjusted for age and sex; Model 3: Adjusted for model 2 plus body mass index, income, smoking, alcohol consumption, physical activity, hypertension, chronic kidney disease, oral anti-diabetic drugs at least three or more, insulin therapy; Model 4: Adjusted for model 3 plus total cholesterol.
TyG, triglyceride-glucose; CVD, cardiovascular disease; IR, incidence rate; PY, person-year; MI, myocardial infarction.
Table 3.Association between TyG Index Quartiles and CVD Risk according to DM Duration Categories
Model 1: Unadjusted; Model 2: Adjusted for age and sex; Model 3: Adjusted for model 2 plus body mass index, income, smoking, alcohol consumption, physical activity, hypertension, chronic kidney disease, oral anti-diabetic drugs at least three or more, insulin therapy; Model 4: Adjusted for model 3 plus total cholesterol.
TyG, triglyceride-glucose; CVD, cardiovascular disease; DM, diabetes mellitus; IR, incidence rate; PY, person-year.
Table 4.Association between TyG Index Quartiles and CVD Mortality according to DM Duration Categories
Model 1: Unadjusted; Model 2: Adjusted for age and sex; Model 3: Adjusted for model 2 plus body mass index, income, smoking, alcohol consumption, physical activity, hypertension, chronic kidney disease, oral anti-diabetic drugs at least three or more, insulin therapy; Model 4: Adjusted for model 3 plus total cholesterol.
TyG, triglyceride-glucose; CVD, cardiovascular disease; DM, diabetes mellitus; IR, incidence rate; PY, person-year.
- 1. Fan W. Epidemiology in diabetes mellitus and cardiovascular disease. Cardiovasc Endocrinol 2017;6:8–16.ArticlePubMedPMC
- 2. Yun JS, Ko SH. Current trends in epidemiology of cardiovascular disease and cardiovascular risk management in type 2 diabetes. Metabolism 2021;123:154838.ArticlePubMed
- 3. Hippisley-Cox J, Coupland CA, Bafadhel M, Russell RE, Sheikh A, Brindle P, et al. Development and validation of a new algorithm for improved cardiovascular risk prediction. Nat Med 2024;30:1440–7.ArticlePubMedPMCPDF
- 4. Clemente-Suarez VJ, Martin-Rodriguez A, Redondo-Florez L, Lopez-Mora C, Yanez-Sepulveda R, Tornero-Aguilera JF. New insights and potential therapeutic interventions in metabolic diseases. Int J Mol Sci 2023;24:10672.ArticlePubMedPMC
- 5. Zhao X, An X, Yang C, Sun W, Ji H, Lian F. The crucial role and mechanism of insulin resistance in metabolic disease. Front Endocrinol (Lausanne) 2023;14:1149239.ArticlePubMedPMC
- 6. Park SY, Gautier JF, Chon S. Assessment of insulin secretion and insulin resistance in human. Diabetes Metab J 2021;45:641–54.ArticlePubMedPMCPDF
- 7. Chang AM, Smith MJ, Bloem CJ, Galecki AT, Halter JB, Supiano MA. Limitation of the homeostasis model assessment to predict insulin resistance and beta-cell dysfunction in older people. J Clin Endocrinol Metab 2006;91:629–34.PubMed
- 8. Simental-Mendia LE, Rodriguez-Moran M, Guerrero-Romero F. The product of fasting glucose and triglycerides as surrogate for identifying insulin resistance in apparently healthy subjects. Metab Syndr Relat Disord 2008;6:299–304.ArticlePubMed
- 9. Kim MK, Han K, Lee SH. Current trends of big data research using the Korean National Health Information Database. Diabetes Metab J 2022;46:552–63.ArticlePubMedPMCPDF
- 10. Levey AS, Eckardt KU, Tsukamoto Y, Levin A, Coresh J, Rossert J, et al. Definition and classification of chronic kidney disease: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2005;67:2089–100.ArticlePubMed
- 11. Guerrero-Romero F, Simental-Mendia LE, Gonzalez-Ortiz M, Martinez-Abundis E, Ramos-Zavala MG, Hernandez-Gonzalez SO, 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.ArticlePubMed
- 12. Di Pino A, DeFronzo RA. Insulin resistance and atherosclerosis: implications for insulin-sensitizing agents. Endocr Rev 2019;40:1447–67.ArticlePubMedPMCPDF
- 13. DeFronzo RA. Insulin resistance, lipotoxicity, type 2 diabetes and atherosclerosis: the missing links: the Claude Bernard Lecture 2009. Diabetologia 2010;53:1270–87.ArticlePubMedPMC
- 14. Ye J, Li L, Wang M, Ma Q, Tian Y, Zhang Q, et al. Diabetes mellitus promotes the development of atherosclerosis: the role of NLRP3. Front Immunol 2022;13:900254.ArticlePubMedPMC
- 15. Li Y, Liu Y, Liu S, Gao M, Wang W, Chen K, et al. Diabetic vascular diseases: molecular mechanisms and therapeutic strategies. Signal Transduct Target Ther 2023;8:152.ArticlePubMedPMCPDF
- 16. Jia G, DeMarco VG, Sowers JR. Insulin resistance and hyperinsulinaemia in diabetic cardiomyopathy. Nat Rev Endocrinol 2016;12:144–53.ArticlePubMedPDF
- 17. Garcia-Diez E, Lopez-Oliva ME, Caro-Vadillo A, Perez-Vizcaino F, Perez-Jimenez J, Ramos S, et al. Supplementation with a cocoa-carob blend, alone or in combination with metformin, attenuates diabetic cardiomyopathy, cardiac oxidative stress and inflammation in Zucker diabetic rats. Antioxidants (Basel) 2022;11:432.ArticlePubMedPMC
- 18. Moon JH, Kim Y, Oh TJ, Moon JH, Kwak SH, Park KS, et al. Triglyceride-glucose index predicts future atherosclerotic cardiovascular diseases: a 16-year follow-up in a prospective, community-dwelling cohort study. Endocrinol Metab (Seoul) 2023;38:406–17.ArticlePubMedPMCPDF
- 19. Luo XY, Fan FF, Jia J, Zheng B, Zhang Y. The associations of triglyceride glucose and its derived indexes with 10-year cardiovascular disease risk in a Chinese community-based population. Eur Heart J 2024;45(Supplement_1):ehae666.2752.ArticlePDF
- 20. Won KB, Choi SY, Chun EJ, Park SH, Sung J, Jung HO, et al. Different associations of atherogenic index of plasma, triglyceride glucose index, and hemoglobin A1C levels with the risk of coronary artery calcification progression according to established diabetes. Cardiovasc Diabetol 2024;23:418.ArticlePubMedPMCPDF
- 21. Su WY, Chen SC, Huang YT, Huang JC, Wu PY, Hsu WH, et al. Comparison of the effects of fasting glucose, hemoglobin A1c, and triglyceride-glucose index on cardiovascular events in type 2 diabetes mellitus. Nutrients 2019;11:2838.ArticlePubMedPMC
- 22. Sanchez-Inigo L, Navarro-Gonzalez D, Fernandez-Montero A, Pastrana-Delgado J, Martinez JA. The TyG index may predict the development of cardiovascular events. Eur J Clin Invest 2016;46:189–97.ArticlePubMed
- 23. Yao X, Zhang J, Zhang X, Jiang T, Zhang Y, Dai F, et al. Age at diagnosis, diabetes duration and the risk of cardiovascular disease in patients with diabetes mellitus: a cross-sectional study. Front Endocrinol (Lausanne) 2023;14:1131395.ArticlePubMedPMC
- 24. Giri B, Dey S, Das T, Sarkar M, Banerjee J, Dash SK. Chronic hyperglycemia mediated physiological alteration and metabolic distortion leads to organ dysfunction, infection, cancer progression and other pathophysiological consequences: an update on glucose toxicity. Biomed Pharmacother 2018;107:306–28.ArticlePubMed
- 25. Ye Z, Xie E, Jiao S, Gao Y, Li P, Tu Y, et al. Triglyceride glucose index exacerbates the risk of future cardiovascular disease due to diabetes: evidence from the China Health and Retirement Longitudinal Survey (CHARLS). BMC Cardiovasc Disord 2022;22:236.ArticlePubMedPMCPDF
- 26. Zhao M, Xiao M, Tan Q, Lu F. Triglyceride glucose index as a predictor of mortality in middle-aged and elderly patients with type 2 diabetes in the US. Sci Rep 2023;13:16478.ArticlePubMedPMCPDF
- 27. Yi Q, Hu H, Zeng Q. Association of triglycerides to high density lipoprotein cholesterol ratio with hypertension in Chinese adults: a cross-sectional study. Clin Exp Hypertens 2023;45:2195996.ArticlePubMed
- 28. Ormazabal V, Nair S, Elfeky O, Aguayo C, Salomon C, Zuniga FA. Association between insulin resistance and the development of cardiovascular disease. Cardiovasc Diabetol 2018;17:122.ArticlePubMedPMCPDF
- 29. 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.ArticlePubMedPMCPDF
- 30. Liu L, Wu Z, Zhuang Y, Zhang Y, Cui H, Lu F, 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.ArticlePubMedPMCPDF
- 31. Chen CL, Liu L, Lo K, Huang JY, Yu YL, Huang YQ, et al. Association between triglyceride glucose index and risk of new-onset diabetes among Chinese adults: findings from the China Health and Retirement Longitudinal Study. Front Cardiovasc Med 2020;7:610322.ArticlePubMedPMC
- 32. Greenman Y, Golani N, Gilad S, Yaron M, Limor R, Stern N. Ghrelin secretion is modulated in a nutrient- and gender-specific manner. Clin Endocrinol (Oxf) 2004;60:382–8.ArticlePubMed
- 33. Chen W, Ding S, Tu J, Xiao G, Chen K, Zhang Y, et al. Association between the insulin resistance marker TyG index and subsequent adverse long-term cardiovascular events in young and middle-aged US adults based on obesity status. Lipids Health Dis 2023;22:65.ArticlePubMedPMCPDF
- 34. Tutor AW, Lavie CJ, Kachur S, Milani RV, Ventura HO. Updates on obesity and the obesity paradox in cardiovascular diseases. Prog Cardiovasc Dis 2023;78:2–10.ArticlePubMed
- 35. Parto P, Lavie CJ, Arena R, Bond S, Popovic D, Ventura HO. Body habitus in heart failure: understanding the mechanisms and clinical significance of the obesity paradox. Future Cardiol 2016;12:639–53.ArticlePubMed
- 36. Horwich TB, Fonarow GC, Clark AL. Obesity and the obesity paradox in heart failure. Prog Cardiovasc Dis 2018;61:151–6.ArticlePubMed
References
Figure & Data
References
Citations
Citations to this article as recorded by 
PubReader
ePub Link
Cite
XML DownloadAssociation between the Triglyceride-Glucose Index and Cardiovascular Risk and Mortality across Different Diabetes Durations: A Nationwide Cohort Study
Fig. 1. (A) Association between triglyceride-glucose (TyG) quartiles and cardiovascular disease (CVD) risk according to diabetes mellitus (DM) duration. (B) Association between TyG quartiles and CVD mortality according to DM duration.
Fig. 1.
Association between the Triglyceride-Glucose Index and Cardiovascular Risk and Mortality across Different Diabetes Durations: A Nationwide Cohort Study
| Characteristic | Total (n=1,090,485) | TyG index |
|||
|---|---|---|---|---|---|
| Q1 (n=272,523) | Q2 (n=272,763) | Q3 (n=272,554) | Q4 (n=272,645) | ||
| Age, yr | 55.96±12.47 | 59.05±12.55 | 57.65±12.37 | 55.50±12.11 | 51.64±11.54 |
| Age group, yr | |||||
| <40 | 97,943 (8.98) | 17,359 (6.37) | 18,974 (6.96) | 23,830 (8.74) | 37,780 (13.86) |
| 40–49 | 233,373 (21.40) | 40,293 (14.79) | 49,537 (18.16) | 61,209 (22.46) | 82,334 (30.20) |
| ≥50 | 759,169 (69.62) | 214,871 (78.85) | 204,252 (74.88) | 187,515 (68.8) | 152,531 (55.94) |
| Sex | |||||
| Male | 747,558 (68.55) | 166,373 (61.05) | 175,866 (64.48) | 190,340 (69.84) | 214,979 (78.85) |
| Female | 342,927 (31.45) | 106,150 (38.95) | 96,897 (35.52) | 82,214 (30.16) | 57,666 (21.15) |
| DM duration, yr | 99,879 (36.65) | 126,497 (46.38) | 143,373 (52.60) | 159,372 (58.45) | |
| New-onset | 529,121 (48.52) | ||||
| <5 | 206,752 (18.96) | 61,927 (22.72) | 53,833 (19.74) | 47,660 (17.49) | 43,332 (15.89) |
| 5–9 | 171,006 (15.68) | 50,921 (18.69) | 44,521 (16.32) | 39,936 (14.65) | 35,628 (13.07) |
| ≥10 | 183,606 (16.84) | 59,796 (21.94) | 47,912 (17.57) | 41,585 (15.26) | 34,313 (12.59) |
| DM treatment | |||||
| Taking three or more oral anti-diabetic agents | 164,239 (15.06) | 44,485 (16.32) | 41,503 (15.22) | 40,354 (14.81) | 37,897 (13.90) |
| Insulin | 61,870 (5.67) | 22,738 (8.34) | 14,677 (5.38) | 12,607 (4.63) | 11,848 (4.35) |
| Height, cm | 164.7±9.08 | 162.82±8.95 | 163.77±9.14 | 165.06±9.03 | 167.15±8.60 |
| Weigh, kg | 68.62±13.03 | 63.62±11.39 | 67.41±12.35 | 70.13±12.87 | 73.31±13.42 |
| BMI, kg/m2 | 25.18±3.65 | 23.93±3.43 | 25.04±3.57 | 25.64±3.59 | 26.12±3.63 |
| Waist circumference, cm | 85.64±9.12 | 82.34±9.01 | 85.24±8.92 | 86.83±8.77 | 88.15±8.72 |
| Obesity | 536,322 (49.18) | 94,027 (34.5) | 128,740 (47.2) | 148,663 (54.54) | 164,892 (60.48) |
| CKD | 63,775 (5.85) | 17,516 (6.43) | 17,489 (6.41) | 15,977 (5.86) | 12,793 (4.69) |
| Low income | 230,501 (21.14) | 60,300 (22.13) | 58,081 (21.29) | 56,609 (20.77) | 55,511 (20.36) |
| Smoking status | |||||
| Never smoker | 532,745 (48.85) | 159,995 (58.71) | 145,120 (53.20) | 127,571 (46.81) | 100,059 (36.70) |
| Ex-smoker | 251,324 (23.05) | 61,989 (22.75) | 62,962 (23.08) | 64,064 (23.51) | 62,309 (22.85) |
| Current smoker | 306,416 (28.10) | 50,539 (18.54) | 64,681 (23.71) | 80,919 (29.69) | 110,277 (40.45) |
| Alcohol consumption | 127,595 (11.70) | 19,593 (7.19) | 25,763 (9.45) | 33,679 (12.36) | 48,560 (17.81) |
| Regular exercise | 235,562 (21.60) | 69,855 (25.63) | 61,676 (22.61) | 55,354 (20.31) | 48,677 (17.85) |
| SBP, mm Hg | 128.50±15.10 | 125.96±14.90 | 128.15±14.93 | 129.21±14.93 | 130.67±15.25 |
| DBP, mm Hg | 79.08±10.13 | 76.54±9.71 | 78.47±9.83 | 79.76±9.97 | 81.54±10.34 |
| LDL-C, mg/dL | 115.36±34.26 | 110.37±30.27 | 118.05±32.16 | 118.8±34.48 | 114.21±38.85 |
| HDL-C, mg/dL | 51.20±15.42 | 56.83±16.83 | 52.33±15.13 | 49.35±14.00 | 46.30±13.52 |
| Fasting glucose, mg/dL | 149.92±46.56 | 125.71±24.95 | 139.28±29.14 | 151.55±37.59 | 183.13±63.09 |
| Total cholesterol, mg/dL | 199.16±38.79 | 182.47±33.58 | 194.65±34.91 | 202.87±36.64 | 216.64±41.36 |
| eGFR, mL/min/1.73 m2 | 93.06±56.60 | 92.07±51.32 | 91.99±55.62 | 92.54±55.68 | 95.64±63.06 |
| Triglyceride, mg/dL | 143.72 (143.56–143.88) | 72.58 (72.50–72.67) | 118.72 (118.62–118.82) | 169.76 (169.61–169.91) | 291.64 (291.21–292.08) |
| Disease | TyG | Number | CVD events | IR, /1,000 PY | Hazard ratio (95% confidence interval) |
|||
|---|---|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | Model 4 | |||||
| CVD | ||||||||
| Q1 | 272,523 | 14,913 | 9.74 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 272,763 | 14,691 | 9.56 | 0.98 (0.96–1.01) | 1.06 (1.04–1.08) | 1.06 (1.03–1.08) | 1.04 (1.01–1.06) | |
| Q3 | 272,554 | 14,373 | 9.34 | 0.96 (0.94–0.98) | 1.16 (1.14–1.19) | 1.15 (1.12–1.18) | 1.11 (1.08–1.14) | |
| Q4 | 272,645 | 14,146 | 9.19 | 0.95 (0.93–0.97) | 1.42 (1.38–1.45) | 1.36 (1.32–1.39) | 1.28 (1.25–1.31) | |
| MI | ||||||||
| Q1 | 272,523 | 7,118 | 4.59 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 272,763 | 6,995 | 4.50 | 0.98 (0.95–1.01) | 1.04 (1.01–1.08) | 1.04 (1.01–1.08) | 1.02 (0.98–1.02) | |
| Q3 | 272,554 | 6,919 | 4.44 | 0.97 (0.94–1.00) | 1.13 (1.10–1.17) | 1.12 (1.09–1.16) | 1.07 (1.04–1.11) | |
| Q4 | 272,645 | 6,987 | 4.49 | 0.98 (0.95–1.01) | 1.6 (1.32–1.41) | 1.31 (1.27–1.36) | 1.22 (1.17–1.26) | |
| Stroke | ||||||||
| Q1 | 272,523 | 8,608 | 5.57 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 272,763 | 8,475 | 5.46 | 0.98 (0.95–1.01) | 1.07 (1.04–1.10) | 1.07 (1.04–1.10) | 1.05 (1.02–1.09) | |
| Q3 | 272,554 | 8,240 | 5.30 | 0.95 (0.93–0.98) | 1.19 (1.16–1.23) | 1.18 (1.14–1.21) | 1.14 (1.11–1.18) | |
| Q4 | 272,645 | 7,918 | 5.09 | 0.92 (0.89–0.95) | 1.47 (1.43–1.52) | 1.40 (1.36–1.45) | 1.34 (1.29–1.38) | |
| Mortality | ||||||||
| Q1 | 272,523 | 22,861 | 14.60 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 272,763 | 18,598 | 11.83 | 0.81 (0.80–0.83) | 0.91 (0.90–0.93) | 0.97 (0.95–0.99) | 1.00 (0.98–1.02) | |
| Q3 | 272,554 | 15,333 | 9.74 | 0.67 (0.66–0.68) | 0.90 (0.88–0.92) | 0.96 (0.94–0.98) | 1.02 (0.99–1.04) | |
| Q4 | 272,645 | 12,911 | 8.20 | 0.57 (0.55–0.58) | 1.05 (1.03–1.08) | 1.09 (1.06–1.11) | 1.19 (1.16–1.22) | |
| DM duration | TyG | Number | CVD events | IR, /1,000 PY | Hazard ratio (95% confidence interval) |
|||
|---|---|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | Model 4 | |||||
| New-onset | ||||||||
| Q1 | 99,879 | 3,428 | 6.03 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 126,497 | 4,577 | 6.37 | 1.06 (1.01–1.10) | 1.08 (1.03–1.13) | 1.06 (1.02–1.11) | 1.04 (0.99–1.09) | |
| Q3 | 143,373 | 5,118 | 6.27 | 1.04 (1.00–1.09) | 1.15 (1.10–1.20) | 1.12 (1.07–1.67) | 1.07 (1.02–1.12) | |
| Q4 | 159,372 | 5,669 | 6.25 | 1.04 (1.00–1.08) | 1.34 (1.28–1.40) | 1.26 (1.21–1.32) | 1.18 (1.13–1.23) | |
| <5 years | ||||||||
| Q1 | 61,927 | 3,235 | 9.21 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 53,833 | 2,890 | 9.44 | 1.03 (0.98–1.08) | 1.08 (1.02–1.13) | 1.07 (1.02–1.12) | 1.04 (0.99–1.09) | |
| Q3 | 47,660 | 2,486 | 9.15 | 0.99 (0.94–1.05) | 1.15 (1.09–1.21) | 1.12 (1.07–1.18) | 1.08 (1.02–1.13) | |
| Q4 | 43,332 | 2,237 | 9.06 | 0.99 (0.93–1.04) | 1.39 (1.32–1.47) | 1.33 (1.26–1.40) | 1.23 (1.16–1.30) | |
| 5–9 years | ||||||||
| Q1 | 50,921 | 3,374 | 11.83 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 44,521 | 2,952 | 11.80 | 1.00 (0.95–1.05) | 1.04 (0.99–1.09) | 1.02 (0.98–1.08) | 1.00 (0.95–1.05) | |
| Q3 | 39,936 | 2,801 | 12.46 | 1.05 (1.00–1.11) | 1.20 (1.14–1.26) | 1.16 (1.11–1.22) | 1.12 (1.06–1.17) | |
| Q4 | 35,628 | 2,621 | 13.14 | 1.11 (1.06–1.17) | 1.53 (1.45–1.61) | 1.44 (1.37–1.51) | 1.34 (1.27–1.41) | |
| ≥10 years | ||||||||
| Q1 | 59,796 | 4,876 | 14.91 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 47,912 | 4,272 | 16.30 | 1.09 (1.05–1.14) | 1.12 (1.07–1.16) | 1.10 (1.05–1.14) | 1.07 (1.03–1.12) | |
| Q3 | 41,585 | 3,968 | 17.53 | 1.18 (1.13–1.23) | 1.28 (1.23–1.34) | 1.25 (1.19–1.30) | 1.20 (1.15–1.25) | |
| Q4 | 34,313 | 3,619 | 19.47 | 1.31 (1.25–1.35) | 1.65 (1.58–1.72) | 1.54 (1.48–1.61) | 1.44 (1.38–1.50) | |
| P for interaction | <0.001 | <0.001 | <0.001 | <0.001 | ||||
| DM duration | TyG | Number | CVD events | IR, /1,000 PY | Hazard ratio (95% confidence interval) |
|||
|---|---|---|---|---|---|---|---|---|
| Model 1 | Model 2 | Model 3 | Model 4 | |||||
| New-onset | ||||||||
| Q1 | 99,879 | 5,043 | 8.75 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 126,497 | 5,564 | 7.63 | 0.87 (0.84–0.91) | 0.92 (0.88–0.95) | 0.96 (0.94–1.01) | 1.01 (0.97–1.05) | |
| Q3 | 143,373 | 5,073 | 6.12 | 0.70 (0.68–0.73) | 0.85 (0.82–0.89) | 0.92 (0.88–0.96) | 0.97 (0.93–1.01) | |
| Q4 | 159,372 | 4,708 | 5.11 | 0.59 (0.56–0.61) | 0.95 (0.91–0.98) | 1.00 (0.96–1.04) | 1.09 (1.04–1.13) | |
| <5 years | ||||||||
| Q1 | 61,927 | 4,907 | 13.67 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 53,833 | 3,582 | 11.44 | 0.84 (0.80–0.88) | 0.91 (0.87–0.95) | 0.96 (0.92–1.00) | 1.00 (0.95–1.04) | |
| Q3 | 47,660 | 2,652 | 9.54 | 0.70 (0.67–0.73) | 0.88 (0.84–0.93) | 0.94 (0.90–0.99) | 1.00 (0.95–1.05) | |
| Q4 | 43,332 | 2,030 | 8.03 | 0.59 (0.56–0.62) | 1.03 (0.98–1.09) | 1.06 (1.01–1.12) | 1.17 (1.11–1.23) | |
| 5–9 years | ||||||||
| Q1 | 50,921 | 5,171 | 17.64 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 44,521 | 3,821 | 14.84 | 0.84 (0.81–0.88) | 0.90 (0.87–0.94) | 0.95 (0.91–0.99) | 0.98 (0.94–1.02) | |
| Q3 | 39,936 | 2,994 | 12.92 | 0.73 (0.70–0.77) | 0.90 (0.86–0.95) | 0.94 (0.90–0.99) | 0.99 (0.95–1.04) | |
| Q4 | 35,628 | 2,478 | 12.02 | 0.68 (0.65–0.71) | 1.14 (1.08–1.19) | 1.13 (1.08–1.19) | 1.23 (1.18–1.30) | |
| ≥10 years | ||||||||
| Q1 | 59,796 | 7,740 | 22.90 | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | 1.0 (reference) | |
| Q2 | 47,912 | 5,631 | 20.69 | 0.90 (0.87–0.94) | 0.95 (0.92–0.98) | 0.98 (0.95–1.01) | 1.01 (0.98–1.05) | |
| Q3 | 41,585 | 4,614 | 19.53 | 0.85 (0.82–0.89) | 1.00 (0.97–1.04) | 1.03 (0.99–1.07) | 1.09 (1.05–1.13) | |
| Q4 | 99,879 | 5,043 | 18.97 | 0.83 (0.80–0.86) | 1.22 (1.17–1.27) | 1.18 (1.14–1.23) | 1.29 (1.24–1.34) | |
| P for interaction | <0.001 | <0.001 | <0.001 | <0.001 | ||||
Table 1. The Baseline Characteristics according to TyG Quartiles
Values are expressed as mean±standard deviation, number (%), or geometric mean (95% confidence interval). TyG, triglyceride-glucose; DM, diabetes mellitus; BMI, body mass index; CKD, chronic kidney disease; SBP, systolic blood pressure; DBP, diastolic blood pressure; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; eGFR, estimated glomerular filtration rate.
Table 2. Association between TyG Index Quartiles and CVD Outcomes
Model 1: Unadjusted; Model 2: Adjusted for age and sex; Model 3: Adjusted for model 2 plus body mass index, income, smoking, alcohol consumption, physical activity, hypertension, chronic kidney disease, oral anti-diabetic drugs at least three or more, insulin therapy; Model 4: Adjusted for model 3 plus total cholesterol. TyG, triglyceride-glucose; CVD, cardiovascular disease; IR, incidence rate; PY, person-year; MI, myocardial infarction.
Table 3. Association between TyG Index Quartiles and CVD Risk according to DM Duration Categories
Model 1: Unadjusted; Model 2: Adjusted for age and sex; Model 3: Adjusted for model 2 plus body mass index, income, smoking, alcohol consumption, physical activity, hypertension, chronic kidney disease, oral anti-diabetic drugs at least three or more, insulin therapy; Model 4: Adjusted for model 3 plus total cholesterol. TyG, triglyceride-glucose; CVD, cardiovascular disease; DM, diabetes mellitus; IR, incidence rate; PY, person-year.
Table 4. Association between TyG Index Quartiles and CVD Mortality according to DM Duration Categories
Model 1: Unadjusted; Model 2: Adjusted for age and sex; Model 3: Adjusted for model 2 plus body mass index, income, smoking, alcohol consumption, physical activity, hypertension, chronic kidney disease, oral anti-diabetic drugs at least three or more, insulin therapy; Model 4: Adjusted for model 3 plus total cholesterol. TyG, triglyceride-glucose; CVD, cardiovascular disease; DM, diabetes mellitus; IR, incidence rate; PY, person-year.
