Abstract
Objective
CVDs remain a major global public health concern, highlighting the need for simple and effective screening tools. Current evidence regarding the association between TyHGB and CVDs were limited. This study aims to investigate the association between TyHGB and CVDs among the elderly using data from a large community-based cohort study.
Method
This study utilized data from the BaHLS involving elderly individuals in Shenzhen, China. The TyHGB value was calculated using the following formula: TyHGB = TG/HDL-C + 0.7 × FBG (mmol/L) + 0.1 × BMI (kg/m²). A multivariate Cox proportional hazards model and a RCS model were employed to assess the longitudinal association between TyHGB and CVDs.
Results
A total of 26,603 subjects were included in the study, of whom 1,968 (7.40%) experienced CVD events during the follow-up period. After adjusting for confounding factors, we found that for per 1-SD increase in TyHGB, the HRs for overall CVDs, stroke, and AMI were 1.12 (95% CI: 1.07–1.16), 1.09 (95% CI: 1.04–1.15), and 1.30 (95% CI: 1.24–1.37), respectively. After stratification by baseline TyHGB quartiles, higher TyHGB levels were associated with increased risks of overall CVDs, stroke, and AMI, with HRs of 1.35 (95% CI: 1.15–1.56), 1.23 (95% CI: 1.03–1.47), and 1.72 (95% CI: 1.33–2.21), respectively; these significant associations were consistent in subgroup analysis. RCS models revealed a linear association between TyHGB and risks of overall CVDs, stroke, and AMI (nonlinear test, P > 0.05).
Conclusion
TyHGB demonstrates a significant longitudinal association with CVDs events among the elderly. Derived from routine clinical laboratory parameters, TyHGB represents a promising tool for CVDs risk stratification and prevention in this population.
Graphical Abstract
Supplementary Information
The online version contains supplementary material available at 10.1186/s12877-026-07749-4.
Keywords: TyHGB, Cardiovascular disease, Elderly, Llongitudinal study
Introduction
Cardiovascular diseases (CVDs) have become the leading cause of morbidity and mortality both in China and globally [1, 2]. Reports indicate that global deaths from CVDs reached 19 million in 2019 [3]. Insulin resistance (IR), a key metabolic factor, plays a pivotal role in the development of diabetes and atherosclerotic CVDs [4, 5]. It is characterized by reduced sensitivity of peripheral tissues to insulin, resulting in impaired glucose uptake and utilization. Identifying IR is crucial for assessing the risk of CVDs [6]. The current gold standard for evaluating IR is the hyperinsulinemic-euglycemic clamp test (HIEC) [7, 8]. However, this method is operationally complex and costly, limiting its clinical use [9, 10]. Recently, the triglyceride-glucose index (TyG), derived from fasting triglyceride and glucose levels, has been proposed as an alternative indicator of IR, demonstrating higher sensitivity and specificity [11, 12]. Nevertheless, the TyG index primarily reflects the dynamic interplay between glucose and triglycerides, potentially overlooking other critical factors such as high-density lipoprotein cholesterol (HDL-C) levels and body composition, which are also closely associated with CVD pathogenesis.
To enhance predictive capabilities for IR-related diseases, Triglyceride-High-Density Lipoprotein Cholesterol-Glucose Body (TyHGB) index was recently proposed by Song et al. in a large prospective cohort study, offering advantages such as high efficiency, simplicity, and cost-effectiveness [13]. The TyHGB index innovatively integrates HDL-C levels and body mass index (BMI) beyond the traditional TyG index, aiming to more accurately identify IR status [13, 14]. Compared to the conventional TyG index, TyHGB offers the following advantages: (1) incorporation of the triglyceride (TG)/HDL-C ratio—studies confirm that this indicator demonstrates a stronger association with IR than any single parameter [15]; (2) inclusion of BMI values, reflecting the critical role of fat deposition in CVD pathophysiology [16]; (3) comprehensive metabolic assessment through integration of multiple parameters. Furthermore, all components of TyHGB are routinely measured in clinical practice, making it a widely applicable tool. Preliminary studies indicate that TyHGB demonstrates superior predictive value for IR in early pregnancy compared to TyG [17]. Given the shared metabolic basis between CVDs and gestational diabetes mellitus (GDM)—including insulin resistance, dyslipidemia, and fat deposition—TyHGB holds promise as a novel tool for assessing CVD risk.
Currently, the relationship between TyHGB levels and CVDs risk remains unexamined. The study utilized large-scale cohort data from an elderly population to evaluate the association between TyHGB and CVD risk among older adults. The findings are expected to contribute to the identification of simple, effective, and practical clinical indicators for CVDs risk assessment. Furthermore, this research will advance understanding in the field and facilitate the implementation of early, comprehensive interventions aimed at risk reduction.
Methods
Study design and participants
This study utilized data from the Bao’an Health Longitudinal Study (BaHLS), a prospective cohort study focusing on the health of elderly individuals in Bao’an District, Shenzhen, China. The BaHLS recruited 30,753 participants aged 65 years and older during its baseline period (Wave 0), conducted from January to December 2019. Participants underwent annual structured interviews and health examinations, with five follow-up surveys completed to date (2020–2024; Waves 1–5) [18]. After applying exclusion criteria, a total of 26,603 participants with complete data were included in the final analysis.
Participants were excluded from the analysis if they had prevalent CVDs at baseline (n = 3,036), were aged <65 years (n = 251), had missing data on variables such as age and sex (n = 670), or the participants could not be contacted, refused to continue participating, or moved out of the study area (n = 193). During the follow-up period, 1,968 participants were diagnosed with CVDs, including 1,399 with stroke and 643 with acute myocardial infarction (AMI). The study adhered to the principles outlined in the Helsinki Declaration. Permission was obtained from the ethics committee of Shenzhen Bao’an Centre for Chronic Disease Control (approval number: 2024-28). Written informed consent was obtained from all study participants. A flowchart detailing the participant inclusion and exclusion criteria is presented in Fig. 1.
Fig. 1.
Flowchart of participants selction from BaHLS 2019–2024
Data collection
All data collection in the BaHLS (from 2019 to 2024) was conducted by trained medical personnel from local community health centers. Primary data sources included standardized questionnaires, physical measurements, and biological sample collection. The questionnaire gathered information on lifestyle behaviors (e.g., smoking, alcohol consumption, physical activity), medical history (e.g., hypertension, diabetes), and medication use. Physical measurements were performed according to the World Health Organization (WHO) standard protocol [19] toassess participants’ height, weight, waist circumference, and blood pressure. Height and weight measurements required participants to remove shoes and hats while wearing light clothing, with height recorded to a precision of 0.1 cm and weight to 0.1 kg. Waist circumference was measured using a 1.5-meter tape at the midpoint between the lower edge of the rib cage and the iliac crest along the midaxillary line [19]. Blood pressure was measured twice on the right upper arm using a calibrated mercury sphygmomanometer by trained medical personnel, after subjects had rested in a seated position for at least 5 min in a quiet room [20]. Measurements were taken 1–2 min apart, with the arm supported at heart level and the cuff appropriately sized. The average of the two readings was used for subsequent analysis. If the two measurements differed by > 5 mmHg, a third measurement was obtained, and the average of the two closest readings was used [18]; For biological sample collection, subjects were required to fast for at least 8 h (typically overnight) prior to venous blood sampling. Fasting status was verbally confirmed by trained nurses before phlebotomy. Blood samples (10 mL) were drawn from the antecubital vein using vacuum tubes containing EDTA-K2 anticoagulant for hemoglobin measurement and serum separator tubes for lipid and glucose assays [18, 21]. Samples were immediately placed on ice and transported to the laboratory within 2 h of collection. Serum was separated by centrifugation at 3,000 rpm for 10 min at 4 °C within 4 h of collection [18]. Hemoglobin was measured immediately using the cyanmethemoglobin method on an automated hematology analyzer. Serum samples were aliquoted and stored at -80 °C for batch analysis of lipid profiles within 72 h to minimize degradation. The following biomarkers were measured using standardized enzymatic methods on an automated biochemistry analyzer: Total Cholesterol (TC) and TG were estimated using enzymatic colorimetric methods with commercially available reagents [18, 21]. HDL-C and Low-Density Lipoprotein Cholesterol (LDL-C) were measured using direct timed-endpoint colorimetric methods [18]. Fasting blood glucose (FBG) was determined using glucose oxidase-peroxidase assays [18, 21].
TyHGB and outcomes
Based on the subjects’ weight, height, TG, HDL-C, and FBG levels, relevant indicators were calculated, including BMI [22], TyG, and TyHGB [23]. The specific calculation formulas were as follows:
 |
 |
 |
Annual face-to-face follow-up visits with participants were conducted by staff from the same medical institution, covering the topics outlined previously. The primary outcome variable was CVD diagnosis of during follow-up, including stroke and AMI as subtypes. These conditions were defined as self-reported diagnoses confirmed by a physician or current use of medication for such conditions. Participants who reported a prior history of AMI or stroke in earlier surveys were required to reconfirm this information in subsequent surveys. If participants denied a previously self-reported diagnosis of stroke or AMI, these conflicting reports were retrospectively corrected.
Covariates
The study adjusted for baseline sociodemographic characteristics, lifestyle behaviors, physical measurements, and selected biological indicators potentially influencing CVD risk as covariates. Sociodemographic variables included gender, age, marital status (married or others), educational attainment (illiterate, primary school, middle school, high school or above). Clinical comorbidities included history of diabetes and hypertension. A history of hypertension was defined as a self-reported physician-diagnosed of hypertension or current use of antihypertensive medication. A history of diabetes was defined as a self-reported physician-diagnosed of type 2 diabetes or current use of hypoglycemic medication. Lifestyle behaviors encompassed smoking status (never smoked, current smoker, or former smoker), alcohol consumption (current drinker or non-drinker), and physical activity (0 times per week,1–4 times per week, or ≥5times per week). Physical measurements included systolic blood pressure (SBP), diastolic blood pressure (DBP) and BMI. Biological indicators primarily comprised hemoglobin (Hb), TC, LDL-C, aspartate aminotransferase (AST), alanine aminotransferase (ALT) and FBG.
Statistical analyses
TyHGB was divided into quartiles based on its distribution: Q1 (≤ 25th percentile), Q2 (> 25th to 50th percentile), Q3 (> 50th to 75th percentile), and Q4 (> 75th percentile). As a sensitivity analysis, TyHGB was also categorized into tertiles (T1: ≤33.3%rd, T2: >33.3%rd to 66.7%th, T3: >66.7%th) to evaluate the robustness of findings across different categorization approaches. Continuous baseline variables with a normal distribution were expressed as mean ± standard deviation, and intergroup comparisons were conducted using one-way analysis of variance (ANOVA), and the post hoc Bonferroni test was utilized for subgroup comparisons after ANOVA. Continuous variables that were not normally distributed were reported as medians with interquartile ranges (IQRs), and group comparisons were performed using the Kruskal-Wallis test. Categorical variables were presented as frequencies and percentages, with comparisons between groups carried out using chi-square tests.
For longitudinal follow-up data analysis, we calculated the incidence rate of CVD events per 1,000 person-years based on the time of CVD diagnosis and compared incidence rates using the Kaplan-Meier method with the log-rank test. A Cox proportional hazards model was employed to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) to assess the association between TyHGB and CVD events. Model 1 included no adjustments; Model 2 adjusted for age, gender, education level, marital status, smoking status, alcohol consumption, and physical activity; Model 3 further incorporated Hb, hypertension, diabetes, SBP, DBP, TC, LDL-C, FBG, AST, and ALT based on Model 2. An RCS model was used to assess the linearity of the relationship between TyHGB and CVD events. Subgroup analyses examined the association between TyHGB and CVDs across populations stratified by age (65–75 years / ≥75 years), gender (male / female), educational attainment (illiterate, primary school, middle school, high school or above), hypertension (yes/no), and diabetes status (yes/no). Two sensitivity analyses were conducted: (1) excluding participants who died during follow-up (n = 330) to assess whether mortality influenced the observed association; and (2) treating CVD deaths as cardiovascular events and non-CVDs deaths as a competing risk. To account for death as a competing risk, the Fine-Gray sub-distribution hazard model was applied to estimate the sub-distribution hazard ratio (SHR) and its 95% confidence interval (CI) for the association between TyHGB and CVD, with death treated as a competing event.
All the statistical analyses in the study were conducted using IBM SPSS (Version 26.0) and R version 4.4.1 (R Foundation for Statistical Computing). Statistical significance was defined as P < 0.05 across all analyses.
Results
Baseline characteristics and outcomes across TyHGB quartiles
The study ultimately included 26,603 subjects without diagnosed CVD at baseline, of whom 15,186 (57.08%) subjects were females. The mean age of participants was 70.22 ± 4.59 years (ranging from 65 to 99 years). 2,861 (10.75%) subjects were current smokers, 4,351 (16.36%) subjects were current drinkers, and 41.21% of participants and 17.15% of participants had previously diagnosed hypertension and diabetes, respectively. Participants were divided into four groups based on baseline TyHGB quartile levels. Baseline characteristics of participants are detailed in Table 1. Significant differences were observed between groups in BMI, waist circumference, education level, smoking status, alcohol consumption, hypertension, diabetes, SBP, DBP, TC, TG, HDL-C, LDL-C, FBG, Hb, TyG, and TyHGB levels (all p < 0.05). Subjects across different TyHGB quartiles exhibited significant differences in the occurrence of CVD events, stroke and AMI during follow-up. No significant differences were found in marital atatus and physical activity. The results of the post hoc Bonferroni test were shown in Supplementary Table S1.
Table 1.
Baseline characteristics and outcomes of participants across TyHGB quartiles at baseline in 2019
| Characteristics | Total (n = 26,603) | Q1 (n = 14,182) |
Q2 (n = 4,069) |
Q3 (n = 16,901) |
Q4 (n = 1,350) |
P value |
|---|---|---|---|---|---|---|
| Age a, years | 70.22 ± 4.59 | 70.34 ± 4.70 | 70.12 ± 4.57 | 70.21 ± 4.59 | 70.21 ± 4.51 | 0.051 |
| BMI a, kg/m2 | 23.68 ± 3.19 | 21.56 ± 2.66 | 23.55 ± 2.76 | 24.54 ± 2.96 | 25.09 ± 3.15 | < 0.001 |
| Waistline a, cm | 85.68 ± 8.61 | 80.34 ± 7.80 | 85.27 ± 7.73 | 87.74 ± 7.86 | 89.36 ± 8.24 | < 0.001 |
| Female c, n(%) | 15,186 (57.08) | 3557 (53.73) | 3730 (56.12) | 3946 (59.38) | 3933 (59.11) | < 0.001 |
| Married c, n(%) | 25,738 (96.75) | 6431 (96.61) | 6452 (97.07) | 6425 (96.69) | 6430 (96.63) | 0.404 |
| Education level c, n(%) | ||||||
| Illiterate | 3987 (14.99) | 1091 (16.39) | 975 (14.67) | 977 (14.70) | 944 (14.19) | 0.001 |
| Primary school | 12,446 (46.78) | 3173 (47.66) | 3075 (46.26) | 3069 (46.19) | 3129 (47.02) | |
| Middle school | 6205 (23.32) | 1464 (21.66) | 1566 (23.56) | 1576 (23.72) | 1599 (24.03) | |
| High School or above | 3965 (14.90) | 929 (13.96) | 1031 (15.51) | 1023 (15.40) | 982 (14.76) | |
| Smoking c, n(%) | ||||||
| No | 21,538 (80.96) | 5288 (79.44) | 5372 (80.82) | 5433 (81.76) | 5445 (81.83) | < 0.001 |
| Yes | 2861 (10.75) | 814 (12.23) | 684 (10.29) | 669 (10.07) | 694 (10.43) | |
| Quit | 2204 (8.28) | 555 (8.34) | 591 (8.89) | 543 (8.17) | 515 (7.74) | |
| Alcohol consumption c, n(%) | ||||||
| No | 22,252 (83.64) | 5460 (82.02) | 5525 (83.12) | 5624 (84.64) | 5643 (84.81) | < 0.001 |
| Yes | 4351 (16.36) | 1197 (17.98) | 1122 (16.88) | 1021 (15.36) | 1011 (15.19) | |
| Physical activity c, n(%) | ||||||
| 0 times per week | 5704 (21.44) | 1459 (21.92) | 1406 (21.15) | 1429 (21.50) | 1410 (21.19) | 0.558 |
| 1–4 times per week | 3177 (11.94) | 789 (11.85) | 759 (11.42) | 805 (12.11) | 824 (12.38) | |
| ≥ 5times per week | 17,703 (66.62) | 4409 (66.23) | 4482 (67.43) | 4411 (66.38) | 4420 (66.34) | |
| With hypertension c, n(%) | 10,963 (41.21) | 1917 (28.80) | 2628 (39.54) | 3042 (45.78) | 3376 (50.74) | < 0.001 |
| With diabetes c, n(%) | 4563 (17.15) | 227 (3.41) | 636 (9.57) | 1204 (18.12) | 2496 (37.51) | < 0.001 |
| Clinical characteristics | ||||||
| SBP a, mmHg | 133.50 ± 17.72 | 129.76 ± 18.09 | 132.79 ± 17.53 | 135.09 ± 17.28 | 136.35 ± 17.26 | < 0.001 |
| DBP a, mmHg | 77.90 ± 10.19 | 76.29 ± 10.36 | 77.76 ± 10.18 | 78.43 ± 10.07 | 79.10 ± 9.93 | < 0.001 |
| Hb a, g/L | 135.36 ± 15.96 | 132.76 ± 15.47 | 135.24 ± 15.59 | 135.96 ± 15.95 | 137.46 ± 16.46 | < 0.001 |
| FBG a, mmol/L | 5.97 ± 1.76 | 5.02 ± 0.59 | 5.51 ± 0.71 | 5.97 ± 1.04 | 7.39 ± 2.12 | < 0.001 |
| TC a, mmol/L | 5.13 ± 1.04 | 5.04 ± 0.97 | 5.13 ± 1.04 | 5.17 ± 1.05 | 5.18 ± 1.07 | < 0.001 |
| TG a, mmol/L | 1.55 ± 0.89 | 0.87 ± 0.26 | 1.20 ± 0.33 | 1.58 ± 0.44 | 2.55 ± 1.09 | < 0.001 |
| HDL-C a, mmol/L | 1.39 ± 0.40 | 1.67 ± 0.43 | 1.44 ± 0.37 | 1.29 ± 0.30 | 1.15 ± 0.27 | < 0.001 |
| LDL-C a, mmol/L | 3.00 ± 0.90 | 2.85 ± 0.87 | 3.04 ± 0.91 | 3.10 ± 0.91 | 3.02 ± 0.91 | < 0.001 |
| ALT b, U/L |
19.00 (14.00, 26.70) |
17.00 (13.00,23.00) | 18.30 (14.00,25.00) | 20.00 (14.90,27.00) | 22.00 (16.00,31.00) | < 0.001 |
| AST b, U/L |
23.00 (19.30, 29.00) |
23.40 (20.00,28.10) | 23.00 (19.60,28.00) | 23.00 (19.00,29.00) | 23.00 (19.00,30.00) | 0.001 |
| TyG a | 8.75 ± 0.59 | 8.11 ± 0.30 | 8.52 ± 0.24 | 8.87 ± 0.24 | 9.48 ± 0.41 | < 0.001 |
| TyHGB a | 9.42 ± 2.69 | 6.92 ± 0.50 | 8.18 ± 0.33 | 9.52 ± 0.48 | 13.05 ± 2.71 | < 0.001 |
| Outcomes | ||||||
| With CVDs c, n(%) | 1968 (7.40) | 353 (5.30) | 446 (6.71) | 551 (8.29) | 618 (9.29) | < 0.001 |
| With stroke c, n(%) | 1399 (5.26) | 264 (3.97) | 320 (4.81) | 390 (5.87) | 425 (6.39) | < 0.001 |
| With AMI c, n(%) | 643 (2.42) | 99 (1.49) | 139 (2.09) | 181 (2.72) | 224 (3.37) | < 0.001 |
a ANOVA test; b Wilcoxon rank-sum test; c chi-square test
Characteristics associated with incident CVD
During the follow-up period, 1,968 participants (7.40%) developed CVD events (Table 1). Table 2 presents a comparison of baseline characteristics and clinical indicators between participants with incident CVD and those without. Subjects with incident CVD exhibited significantly higher baseline values for age, BMI, waist circumference, SBP, DBP, Hb, FBG, TC, TG, LDL-C, ALT, TyG, and TyHGB compared to those without incident CVD (p < 0.05). Furthermore, a statistically significant difference (p < 0.05) was observed in the proportion of gender (50.71% vs. 42.29%), smokers (9.50% vs. 8.19%), and individuals with hypertension (56.10% vs. 40.02%) and diabetes (26.63% vs. 16.40%) in the CVD group compared to the non-CVD group. Conversely, HDL-C levels were significantly lower in participants with incident CVDs (p < 0.05).
Table 2.
Characteristics of participants that developed CVDs and those without
| Characteristics | Total (n = 26,603) |
Without CVDs (n = 24,635) |
With CVDs (n = 1,968) |
P value |
|---|---|---|---|---|
| Age a, years | 70.22 ± 4.59 | 70.11 ± 4.52 | 71.57 ± 5.23 | < 0.001 |
| BMI a, kg/m2 | 23.68 ± 3.19 | 23.64 ± 3.18 | 24.21 ± 3.22 | < 0.001 |
| Waistline a, cm | 85.68 ± 8.61 | 85.53 ± 8.59 | 87.47 ± 8.65 | < 0.001 |
| Female c, n(%) | 15,186 (57.08) | 14,216 (57.71) | 970 (49.29) | < 0.001 |
| Married c, n(%) | 25,738 (96.75) | 23,833 (96.74) | 1905 (96.80) | 0.896 |
| Education level c, n(%) | ||||
| Illiterate | 3987 (14.99) | 3170 (15.06) | 277 (14.08) | < 0.001 |
| Primary school | 12,446 (46.78) | 11,612 (47.14) | 834 (42.38) | |
| Middle school | 6205 (23.32) | 5727 (23.25) | 478 (24.29) | |
| High School or above | 3965 (14.90) | 3586 (14.56) | 379 (19.26) | |
| Smoking c, n(%) | ||||
| No | 21,538 (80.96) | 20,004 (81.20) | 1534 (77.95) | 0.002 |
| Yes | 2861 (10.75) | 2614 (10.61) | 247 (12.55) | |
| Quit | 2204 (8.28) | 2017 (8.19) | 187 (9.50) | |
| Alcohol consumption c, n(%) | ||||
| No | 22,252 (83.64) | 20,586 (83.56) | 1666 (84.65) | 0.208 |
| Yes | 4351 (16.36) | 4049 (16.44) | 302 (15.35) | |
| Physical activity c, n(%) | ||||
| 0 times per week | 5704 (21.44) | 5307 (21.54) | 397 (20.17) | 0.346 |
| 1–4 times per week | 3177 (11.94) | 2934 (11.91) | 243(12.35) | |
| ≥ 5times per week | 17,703 (66.62) | 16,394 (66.55) | 1328 (67.48) | |
| With hypertension c, n(%) | 10,963 (41.21) | 9859 (40.02) | 1104 (56.10) | < 0.001 |
| With diabetes c, n(%) | 4563 (17.15) | 4039 (16.40) | 524 (26.63) | < 0.001 |
| Clinical characteristics | ||||
| SBP a, mmHg | 133.50 ± 17.72 | 133.31 ± 17.67 | 136.62 ± 18.13 | < 0.001 |
| DBP a, mmHg | 77.90 ± 10.19 | 77.84 ± 10.17 | 78.63 ± 10.35 | 0.001 |
| Hb a, g/L | 135.36 ± 15.96 | 135.32 ± 15.91 | 136.49 ± 16.53 | 0.001 |
| FBG a, mmol/L | 5.97 ± 1.76 | 5.94 ± 1.72 | 6.35 ± 2.16 | < 0.001 |
| TC a, mmol/L | 5.13 ± 1.04 | 5.13 ± 1.03 | 5.08 ± 1.13 | 0.031 |
| TG a, mmol/L | 1.55 ± 0.89 | 1.55 ± 0.89 | 1.61 ± 0.88 | 0.003 |
| HDL-C a, mmol/L | 1.39 ± 0.40 | 1.39 ± 0.40 | 1.33 ± 0.37 | < 0.001 |
| LDL-C a, mmol/L | 3.00 ± 0.90 | 3.00 ± 0.90 | 2.99 ± 0.94 | 0.468 |
| ALT b, U/L | 19.00 (14.00, 26.70) | 19.00 (14.00,26.60) | 20.00 (15.00, 27.10) | 0.001 |
| AST b, U/L | 23.00 (19.30, 29.00) | 23.00 (19.30,29.00) | 23.00 (19.00, 29.00) | 0.166 |
| TyG a | 8.75 ± 0.59 | 8.74 ± 0.59 | 8.84 ± 0.60 | < 0.001 |
| TyHGB a | 9.42 ± 2.69 | 9.375 ± 2.66 | 9.95 ± 2.91 | < 0.001 |
a t test; b Wilcoxon rank-sum test; c chi-square test
Risk of TyHGB and CVD events
The results of Kaplan-Meier curves are presented in Fig. 2, which illustrate the incidence of CVD events across quartiles of TyHGB. A statistically significant association between elevated TyHGB levels and an increased risk of CVD events was indicated (all log-rank tests, P < 0.05). Concurrently, significant increases were also observed in the risk of stroke and AMI events (log-rank tests, P < 0.05).
Fig. 2.
Kaplan - Meier incidence rate of CVDs, stroke and AMI according to quartiles of TyHGB. a, CVD, b, Stroke, c, AMI
Figure 3 presented the results of the Cox proportional hazards regression analysis. After adjusting for relevant covariates, compared with the lowest quartile of TyHGB levels, HRs for the risk of CVD events in subjects in the second to highest quartiles were 1.16 (95% CI: 1.01–1.34), 1.34 (95% CI: 1.17–1.54), and 1.35 (95% CI: 1.15–1.56), respectively. The adjusted HRs for stroke were 1.12 (95% CI: 0.95–1.32), 1.27 (95% CI: 1.08–1.50), and 1.23 (95% CI: 1.03–1.47), respectively. The HRs for AMI were 1.27 (95% CI: 0.98–1.65), 1.55 (95% CI: 1.20–1.99), and 1.72 (95% CI: 1.33–2.21), respectively. When treating TyHGB as a continuous variable in Model 3, each 1-SD increased in TyHGB was associated with elevated risks of CVD, stroke, and AMI, with HRs of 1.12 (95% CI: 1.07–1.16), 1.09 (95% CI: 1.04–1.15), and 1.17 (95% CI: 1.09–1.25), respectively.
Fig. 3.
Longitudinal association between TyHGB and risk of CVD. Model 1, no confounders were included; model 2, age, gender, educational level, marital status, smoking, alcohol consumption and physical activities were included. Model 3, Hb, hypertension, diabetes, SBP, DBP, FBG, TC, LDL-C, AST and ALT based on Model 2
Linear relationship between TyHGB and CVD
RCS modeling analysis revealed a linear association between TyHGB levels and the risk of CVDs, stroke, and AMI (p for nonlinear > 0.05). These results are presented in Fig. 4.
Fig. 4.
The nonlinear relationship between TyHGB and CVD, stroke and AMI. Adjustment factors included age, gender, education level, marital status, smoking, alcohol consumption, physical activities, Hb, hypertension, diabetes, SBP, DBP, FBG, TC, LDL-C, AST and ALT. a, CVD, b, Stroke, c, AMI
Subgroup analysis and sensitivity analysis
To examine the relationship between TyHGB and CVD incidence across different populations, we conducted subgroup analyses based on age (65–74 years and ≥ 75 years), gender (female and male), education level (illiterate, primary school, middle school, and high school or above), hypertension status (yes or no) and diabetes status (yes or no). The results demonstrated that the associations between TyHGB and the risk of overall CVD, stroke, and AMI remained significant within these subgroups after adjusting for other confounding factors (Fig. 5). No significant interactions were observed between TyHGB and any subgroup variables (all p-values for interaction > 0.05).
Fig. 5.
Association between TyHGB and overall CVD, stroke, AMI in subgroups
To evaluate the robustness of our findings, we conducted three sensitivity analyses. First, we excluded 330 participants who died during follow-up, among whom 175 died from causes of CVD. The association between TyHGB and CVD remained consistent with the primary analysis (Supplementary Table S2). Second, to account for the competing risk of death, we applied Fine-Gray subdistribution hazard models, the results were materially unchanged, indicating that our findings were not substantially affected by competing mortality risks (Supplementary Table S2). Third, we categorized TyHGB into tertiles instead of quartiles and performed Kaplan-Meier curve analysis. The results showed that the risk of CVD in the highest tertiles was higher than in the lowest tertiles (Supplementary Figure S1).
Discussion
Emerging evidence highlights the importance of metabolic markers in CVD [23–27]. Given that all components of TyHGB are routinely measured in clinical practice, monitoring TyHGB levels could offer additional value in identifying populations at high risk for CVD. Utilizing large-scale, community-based cohort data from elderly individuals in China, this study represented the first investigation into the association between TyHGB and the risk of CVD in the population. Longitudinal analyses revealed that elevated baseline TyHGB levels were associated with an increased risk of CVD events, with similar elevations observed for stroke and AMI risks. These findings suggest that TyHGB could serve as a valuable indicator for monitoring and preventing the risk of CVD events.
Accumulating evidence from diverse research methodologies has implicated insulin resistance (IR) as a critical factor in the progression of CVD [28–30]. From an epidemiological perspective, studies have consistently confirmed the association between IR—frequently estimated via the Homeostatic Model Assessment for Insulin Resistance (HOMA-IR)—and an increased risk of CVD [31, 32]. Complementing these observational findings, genetic data have established a causal link between genetically induced IR and various cardiovascular pathologies, including coronary heart disease [33, 34] and ischemic stroke [35, 36]. Mechanistically, IR is proposed to drive CVD progression primarily through its impact on vascular structure and function. It contributes significantly to increased vascular stiffness, which accelerates arterial aging [29].
The direct measurement of insulin levels, while clinically valuable, is often hindered by operational complexity and high costs. Consequently, there is a significant need for accessible surrogate markers, leading to the development of indicators such as the triglyceride-glucose (TyG) index. Previous research has linked the TyG index to various adverse cardiometabolic outcomes, including hypertension [37], diabetes [38], and CVD [39]. However, the TyG index has notable limitations that may restrict its predictive accuracy. Specifically, it excludes high-density lipoprotein cholesterol (HDL-C), a component known for its antioxidant and anti-inflammatory properties that protect against CVD [40, 41]. This omission may limit the TyG index’s ability to comprehensively capture an individual’s metabolic risk profile. Additionally, the TyG index does not incorporate body composition metrics such as BMI, despite the established independent associations between these metrics and cardiovascular health [42]. In contrast, TyHGB index provides a more comprehensive assessment of metabolic health by integrating HDL-C levels and BMI into the traditional TyG framework. TyHGB does not merely add HDL-C as an independent factor, it incorporates HDL-C into the TG/HDL-C ratio framework. Researchs have indicated that the ratio effectively reflects the balance between potentially harmful lipids and protective lipid, providing deeper insights into lipid metabolism compared to TG alone. Consistent with prior research demonstrating a strong association between TyG index and CVD [43–45], the study revealed a strong association between TyHGB index and CVD incidence. Furthermore, RCS indicated a clear linear relationship between TyHGB levels and the risk of CVDs events. TyHGB index is derived from routine laboratory parameters and requires no specialized equipment, facilitating its seamless integration into low-cost diagnostic protocols. This ease of implementation makes the index especially valuable for remote areas with limited medical resources. The study further expands the range of indicators for predicting CVD risk and provides a basis for strengthening public health screening and preventive counseling for high-risk populations.
The study utilized a longitudinal survey data from elderly populations in Chinese communities, provides a systematic and comprehensive analysis of the association between TyHGB indicators and CVD. It effectively supplements the existing evidence base, with conclusions demonstrating causal relationships and representativeness. Additionally, the measurement methods for the indicators examined in the study are operationally simple and readily applicable in clinical practice. Furthermore, all surveys were conducted by trained professionals using structured questionnaires administered individually, while physical examinations and laboratory test indicators were collected by medical personnel, ensuring data consistency and reliability.
Several limitations of the study warrant consideration. Several limitations of this study warrant consideration. First, the assessment of TyHGB components at a single time point failed to capture potential temporal variations in metabolic status during the follow-up period. This single-measurement approach may have introduced regression dilution bias, potentially leading to an underestimation of the true effect sizes. Second, data on certain covariates were self-reported through standardized questionnaires, which may have introduced information bias, specifically recall or social desirability bias. Third, although we adjusted for a range of potential confounders based on prior knowledge, the possibility of residual confounding remains, as factors such as dietary patterns, physical fitness, inflammatory biomarkers, subclinical atherosclerosis, and genetic predisposition were not accounted for. Fourth, while sensitivity analyses excluding deaths and the use of Fine-Gray competing risk models yielded consistent results, unmeasured differences between survivors and non-survivors may have influenced the risk estimates. Finally, the observational nature of this study limits the ability to establish definitive causality, and the observed associations may still be affected by sample selection or information collection biases. Given these methodological constraints, future investigations—specifically randomized controlled trials or Mendelian randomization studies—are warranted to clarify causal inferences and confirm the mechanistic association between TyHGB and CVD risk.
Conclusion
This longitudinal study demonstrates a positive correlation between TyHGB and the risk of CVD in older adults, providing new evidence supporting a causal relationship between the two. These findings suggest that incorporating TyHGB assessment into clinical practice may be crucial for preventing the onset of cardiovascular disease in this population.
Supplementary information
Supplementary Material 1. Table S1:Post hoc Bonferroni test of demographic and clinical characteristics of participants across TyHGB quartiles.
Supplementary Material 2. Table S2. Sensitivity Analyses for the association between TyHGB and CVDs.
Supplementary Material 3. Figure S1. Kaplan-Meier incidence rate of CVDs, stroke and AMI according to tertiles of TyHGB. a, CVD, b, Stroke, c, AMI.
Acknowledgements
We are grateful to all the volunteers for their participation inthe present study and to all the investigators for their support and hard work duringthis survey.
Authors’ contributions
XY-H: Conceptualization, Data curation, Formal analysis, Software, Writing–original draft, Writing – review & editing. Z-L: Data curation, Software, Visualization, Writing – review & editing. YF-G: Funding acquisition, Writing – review & editing. PP-S: Data curation, Visualization, Writing – review & editing. RC-Z: Data curation, Software, Writing – review & editing, Supervision. XL-Z: Funding acquisition, Writing – review & editing, Supervision.
Funding
This study received support from the Medical Scientific Research Foundation of Guangdong Province of China (C2023107), Bao’an District Medical and Health Research Projects (2024JD232) and the Bao’an Key Discipline of Chronic Noncommunicable Disease Prevention and Control, Shenzhen, Guangdong Province, China.
Data availability
The data used in this study have been uploaded to the “Zenodo” database and are available using the target URL: 10.5281/zenodo.18104963.
Declarations
Ethics approval and consent to participate
The study adhered to the principles outlined in the Helsinki Declaration. Permission was obtained from the ethics committee of Shenzhen Bao’an Centre for Chronic Disease Control (approval number: 2024-28). The participants were informed about the study objectives, their voluntary participation, and their ability to withdraw from the study at any point. Written informed consent was obtained from all study participants.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Ren-cheng Zhao, Email: 394105366@qq.com.
Xing-lin Zhong, Email: 3314369279@qq.com.
References
- 1.Kaminsky LA, German C, Imboden M, Ozemek C, Peterman JE, Brubaker PH. The importance of healthy lifestyle behaviors in the prevention of cardiovascular disease. Prog Cardiovasc Dis. 2022;70:8–15. 10.1016/j.pcad.2021.12.001. [DOI] [PubMed] [Google Scholar]
- 2.US Preventive Services Task Force, Bibbins-Domingo K, Grossman DC, Curry SJ, Davidson KW, Epling JW, et al. Statin Use for the Primary Prevention of Cardiovascular Disease in Adults: US Preventive Services Task Force Recommendation Statement. JAMA. 2016;316:1997–2007. 10.1001/jama.2016.15450. [DOI] [PubMed] [Google Scholar]
- 3.Roth GA, Mensah GA, Johnson CO, Addolorato G, Ammirati E, Baddour LM, et al. Global Burden of Cardiovascular Diseases and Risk Factors, 1990–2019: Update From the GBD 2019 Study. J Am Coll Cardiol. 2020;76:2982–3021. 10.1016/j.jacc.2020.11.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Louie JZ, Shiffman D, McPhaul MJ, Melander O. Insulin resistance probability score and incident cardiovascular disease. J Intern Med. 2023;294:531–5. 10.1111/joim.13687. [DOI] [PubMed] [Google Scholar]
- 5.Abdul-Ghani M, Maffei P, DeFronzo RA. Managing insulin resistance: the forgotten pathophysiological component of type 2 diabetes. Lancet Diabetes Endocrinol. 2024;12:674–80. 10.1016/S2213-8587(24)00127-X. [DOI] [PubMed] [Google Scholar]
- 6.Wu X, Li X, Xu M, Zhang Z, He L, Li Y. Sarcopenia prevalence and associated factors among older Chinese population: Findings from the China Health and Retirement Longitudinal Study. PLoS ONE. 2021;16:e0247617. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wallace TM, Matthews DR. The assessment of insulin resistance in man. Diabet Med. 2002;19:527–34. 10.1046/j.1464-5491.2002.00745.x. [DOI] [PubMed] [Google Scholar]
- 8.Tahapary DL, Pratisthita LB, Fitri NA, Marcella C, Wafa S, Kurniawan F, et al. Challenges in the diagnosis of insulin resistance: Focusing on the role of HOMA-IR and Tryglyceride/glucose index. Diabetes Metab Syndr. 2022;16:102581. 10.1016/j.dsx.2022.102581. [DOI] [PubMed] [Google Scholar]
- 9.Jerobin J, Ramanjaneya M, Bettahi I, Parammal R, Siveen KS, Alkasem M, et al. Regulation of circulating CTRP-2/CTRP-9 and GDF-8/GDF-15 by intralipids and insulin in healthy control and polycystic ovary syndrome women following chronic exercise training. Lipids Health Dis. 2021;20:34. 10.1186/s12944-021-01463-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Cai H, Que Z, Chen J, Chen D, Rui G, Lan W. Association between different insulin resistance surrogates and osteoarthritis: a cross-sectional study from NHANES 1999–2018. BMC Musculoskelet Disord. 2024;25:901. 10.1186/s12891-024-08027-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Simental-Mendía LE, Rodríguez-Morán 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. 10.1089/met.2008.0034. [DOI] [PubMed] [Google Scholar]
- 12.Guerrero-Romero F, Simental-Mendía LE, González-Ortiz M, Martínez-Abundis E, Ramos-Zavala MG, Hernández-González 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. 10.1210/jc.2010-0288. [DOI] [PubMed] [Google Scholar]
- 13.Song S, Zhang Y, Qiao X, Duo Y, Xu J, Peng Z, et al. HOMA-IR as a risk factor of gestational diabetes mellitus and a novel simple surrogate index in early pregnancy. Int J Gynaecol Obstet. 2022;157:694–701. 10.1002/ijgo.13905. [DOI] [PubMed] [Google Scholar]
- 14.Hu Z, Song X, Hu K, Ruan Y, Zeng F. Association between sleep duration and asthma in different weight statuses (CHNS 2009–2015). Sleep Breath. 2021;25:493–502. 10.1007/s11325-020-02081-6. [DOI] [PubMed] [Google Scholar]
- 15.Baratta F, Pastori D, Polimeni L, Bucci T, Ceci F, Calabrese C, et al. Adherence to Mediterranean Diet and Non-Alcoholic Fatty Liver Disease: Effect on Insulin Resistance. Am J Gastroenterol. 2017;112:1832–9. 10.1038/ajg.2017.371. [DOI] [PubMed] [Google Scholar]
- 16.Rana MN, Neeland IJ. Adipose Tissue Inflammation and Cardiovascular Disease: An Update. Curr Diab Rep. 2022;22:27–37. 10.1007/s11892-021-01446-9. [DOI] [PubMed] [Google Scholar]
- 17.Xu Z, Li C, Wang N, Song G. Association between the new TyG indicator-TyHGB and gestational diabetes mellitus: results from the case-control and prospective cohort studies. J Transl Med. 2025;23:190. 10.1186/s12967-025-06166-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.He X-Y, Lu Z-W, Guo Y-F, Zhao R-C, Liu Z. Association between atherogenic index of plasma and new onset of type 2 diabetes among elderly in China: a longitudinal study. Front Endocrinol (Lausanne). 2025;16:1632400. 10.3389/fendo.2025.1632400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Bondyra-Wiśniewska B, Harton A. Effect of the Nutritional Intervention Program on Body Weight and Selected Cardiometabolic Factors in Children and Adolescents with Excess Body Weight and Dyslipidemia: Study Protocol and Baseline Data. Nutrients. 2023;15:3646. 10.3390/nu15163646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Ni W, Yuan X, Zhang J, Li P, Zhang HM, Zhang Y, et al. Factors associated with treatment and control of hypertension among elderly adults in Shenzhen, China: a large-scale cross-sectional study. BMJ Open. 2021;11:e044892. 10.1136/bmjopen-2020-044892. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ali N, Samadder M, Kathak RR, Islam F. Prevalence and factors associated with dyslipidemia in Bangladeshi adults. PLoS ONE. 2023;18:e0280672. 10.1371/journal.pone.0280672. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Zhang J-S, Yeh W-C, Tsai Y-W, Chen J-Y. The Relationship between Atherogenic Index of Plasma and Obesity among Adults in Taiwan. Int J Environ Res Public Health. 2022;19:14864. 10.3390/ijerph192214864. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Zhao E, Wang R, Chen Y, Xie H, Gao Y, Dong B, et al. Evaluation of a new TyG indicator, TyHGB, in predicting non-alcoholic fatty liver disease: evidence from two independent populations. Lipids Health Dis. 2025;24:342. 10.1186/s12944-025-02742-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Li W, Shen C, Kong W, Zhou X, Fan H, Zhang Y, et al. Association between the triglyceride glucose-body mass index and future cardiovascular disease risk in a population with Cardiovascular-Kidney-Metabolic syndrome stage 0–3: a nationwide prospective cohort study. Cardiovasc Diabetol. 2024;23:292. 10.1186/s12933-024-02352-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Zhang Q, Xiao S, Jiao X, Shen Y. 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. 10.1186/s12933-023-02030-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Mo D, Zhang P, Zhang M, Dai H, Guan J. Cholesterol, high-density lipoprotein, and glucose index versus triglyceride-glucose index in predicting cardiovascular disease risk: a cohort study. Cardiovasc Diabetol. 2025;24:116. 10.1186/s12933-025-02675-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Sun Y, Ji H, Sun W, An X, Lian F. Triglyceride glucose (TyG) index: A promising biomarker for diagnosis and treatment of different diseases. Eur J Intern Med. 2025;131:3–14. 10.1016/j.ejim.2024.08.026. [DOI] [PubMed] [Google Scholar]
- 28.Cho H, Lai C-C, Bonnavion R, Alnouri MW, Wang S, Roquid KA, et al. Endothelial insulin resistance induced by adrenomedullin mediates obesity-associated diabetes. Science. 2025;387:674–82. 10.1126/science.adr4731. [DOI] [PubMed] [Google Scholar]
- 29.Hill MA, Yang Y, Zhang L, Sun Z, Jia G, Parrish AR, et al. Insulin resistance, cardiovascular stiffening and cardiovascular disease. Metabolism. 2021;119:154766. 10.1016/j.metabol.2021.154766. [DOI] [PubMed] [Google Scholar]
- 30.Stewart AJ, Tuncay E, Pitt SJ, Rainbow RD. Editorial: Insulin resistance and cardiovascular disease. Front Endocrinol (Lausanne). 2023;14:1266173. 10.3389/fendo.2023.1266173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Wang Y, Zhang X, Li Y, Gui J, Mei Y, Yang X, et al. Obesity- and lipid-related indices as a predictor of type 2 diabetes in a national cohort study. Front Endocrinol (Lausanne). 2023;14:1331739. 10.3389/fendo.2023.1331739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Duan M, Zhao X, Li S, Miao G, Bai L, Zhang Q, et al. Metabolic score for insulin resistance (METS-IR) predicts all-cause and cardiovascular mortality in the general population: evidence from NHANES 2001–2018. Cardiovasc Diabetol. 2024;23:243. 10.1186/s12933-024-02334-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Erqou S, Adler AI, Challa AA, Fonarow GC, Echouffo-Tcheugui JB. Insulin resistance and incident heart failure: a meta-analysis. Eur J Heart Fail. 2022;24:1139–41. 10.1002/ejhf.2531. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hou X-Z, Lv Y-F, Li Y-S, Wu Q, Lv Q-Y, Yang Y-T, et al. Association between different insulin resistance surrogates and all-cause mortality in patients with coronary heart disease and hypertension: NHANES longitudinal cohort study. Cardiovasc Diabetol. 2024;23:86. 10.1186/s12933-024-02173-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Georgakis MK, Harshfield EL, Malik R, Franceschini N, Langenberg C, Wareham NJ, et al. Diabetes Mellitus, Glycemic Traits, and Cerebrovascular Disease: A Mendelian Randomization Study. Neurology. 2021;96:e1732–42. 10.1212/WNL.0000000000011555. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Ding P-F, Zhang H-S, Wang J, Gao Y-Y, Mao J-N, Hang C-H, et al. Insulin resistance in ischemic stroke: Mechanisms and therapeutic approaches. Front Endocrinol (Lausanne). 2022;13:1092431. 10.3389/fendo.2022.1092431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Sun M, Huang Y, Luo N, Qiu J, Lin Y, Huang Y, et al. TyG Index and Related Indices Predicting Hypertension: Mediation by Neutrophil-to-Lymphocyte Ratio in Multiple Chinese Cohorts. Nutrients. 2025;17:2859. 10.3390/nu17172859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Selvi NMK, Nandhini S, Sakthivadivel V, Lokesh S, Srinivasan AR, Sumathi S. Association of Triglyceride-Glucose Index (TyG index) with HbA1c and Insulin Resistance in Type 2 Diabetes Mellitus. Maedica (Bucur). 2021;16:375–81. 10.26574/maedica.2021.16.3.375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Ding X, Wang X, Wu J, Zhang M, Cui M. Triglyceride-glucose index and the incidence of atherosclerotic cardiovascular diseases: a meta-analysis of cohort studies. Cardiovasc Diabetol. 2021;20:76. 10.1186/s12933-021-01268-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Nosrati M, Safari M, Alizadeh A, Ahmadi M, Mahrooz A. The Atherogenic Index Log (Triglyceride/HDL-Cholesterol) as a Biomarker to Identify Type 2 Diabetes Patients with Poor Glycemic Control. Int J Prev Med. 2021;12:160. 10.4103/ijpvm.IJPVM_357_20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Welty FK. How Do Elevated Triglycerides and Low HDL-Cholesterol Affect Inflammation and Atherothrombosis? Curr Cardiol Rep. 2013;15:400. 10.1007/s11886-013-0400-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Weeldreyer NR, De Guzman JC, Paterson C, Allen JD, Gaesser GA, Angadi SS. Cardiorespiratory fitness, body mass index and mortality: a systematic review and meta-analysis. Br J Sports Med. 2025;59:339–46. 10.1136/bjsports-2024-108748. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Tao L-C, Xu J-N, Wang T-T, Hua F, Li J-J. Triglyceride-glucose index as a marker in cardiovascular diseases: landscape and limitations. Cardiovasc Diabetol. 2022;21:68. 10.1186/s12933-022-01511-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Qiao J, Chen X, Pang J, Fei H, Liu Z, Cheng F, et al. Relationship between triglyceride glucose-body mass index baselines and variation with future cardiovascular diseases risk in the middle-aged and elderly individuals. Front Endocrinol (Lausanne). 2025;16:1514660. 10.3389/fendo.2025.1514660. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Tang X, Zhang K, He R. The association of triglyceride-glucose and triglyceride-glucose related indices with the risk of heart disease in a national cohort study. Cardiovasc Diabetol. 2025;24:54. 10.1186/s12933-025-02621-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Material 1. Table S1:Post hoc Bonferroni test of demographic and clinical characteristics of participants across TyHGB quartiles.
Supplementary Material 2. Table S2. Sensitivity Analyses for the association between TyHGB and CVDs.
Supplementary Material 3. Figure S1. Kaplan-Meier incidence rate of CVDs, stroke and AMI according to tertiles of TyHGB. a, CVD, b, Stroke, c, AMI.
Data Availability Statement
The data used in this study have been uploaded to the “Zenodo” database and are available using the target URL: 10.5281/zenodo.18104963.