Abstract
Background
To investigate the associations between TyG-related indices and hypertension-diabetes comorbidity (HTN-DM) among middle-aged and older adults.
Methods
A total of 2854 participants (aged ≥ 45 years) living in a suburban community were included in this study. The collected information comprised three parts: baseline information, physical examination, and laboratory testing. Multivariable logistic regression was used to estimate the odds ratios (ORs) and 95% confidence intervals (CIs) for the associations between TyG-related indices (TyG, TyG-BMI, and TyG-BRI) and HTN-DM. Restricted cubic splines (RCS) were used to model the dose-response relationships. Receiver operating characteristic (ROC) curves were applied to evaluate the discriminative performance of these indices.
Results
Among 2,854 participants, each 1-SD increment in TyG, TyG-BMI, and TyG-BRI conferred multivariable-adjusted ORs (95% CIs) of 3.16(2.75–3.64), 1.39(1.15–1.66), and 1.37(1.22–1.54), respectively (all P < 0.001). Corresponding ORs (95% CIs) comparing the highest versus lowest quartile were 30.99(16.97–62.97), 6.42(4.20-10.14), and 6.43(4.21–10.17), with significant dose-response trends (all P for trend < 0.001). RCS analysis revealed a linear relationship for TyG and nonlinear associations for TyG-BMI and TyG-BRI with HTN-DM. TyG showed the highest observed AUC (0.786) among the evaluated indices, followed by TyG-BMI (0.679) and TyG-BRI (0.633).Across all subgroup analyses, TyG remained consistently and positively associated with HTN-DM prevalence, and the interaction with sex was significant (P = 0.045).
Conclusions
This study revealed that TyG-related indices were associated with HTN-DM. RCS analysis indicated a linear association for TyG but nonlinear associations for TyG-BMI and TyG-BRI. Among the three indices, TyG showed the highest observed AUC. This finding suggests that composite indices incorporating BMI or BRI may not necessarily improve discriminative ability beyond TyG alone.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s12872-026-06313-7.
Keywords: TyG-related indices, HTN-DM, Suburban community
Background
Noncommunicable diseases (NCDs) cause more than 50% of global disease burden [1–3]. HTN (226 million DALYs) and DM (notably type 2 diabetes, with 3.67 million DALYs from disability) are major contributors [4]. One-third of adults worldwide have multimorbidity [5]. Globally, the rising burden of cardiometabolic multimorbidity, encompassing HTN, DM, and related conditions, presents major health and economic challenges. In China, NCDs account for 91% of deaths and 86.7% of DALYs, with higher burdens in poorer provinces [6]. Among adults aged ≥ 45 years, the prevalence of HTN, DM, and HTN-DM comorbidity is 46%, 19.5%, and 12.3%, respectively [7]. During 2019–2020, the comorbidity detection rate reached 13.78% among adults aged ≥ 60 years [8]. These findings highlight the need to identify relevant risk factors and implementing preventive measures are crucial in this population [9].
The TyG index, a reliable surrogate marker for insulin resistance, has been associated with cardiometabolic risk factors in epidemiological studies [10]. Numerous prospective studies have examined the predictive value of the TyG index—alone or in combination with other biomarkers—for diabetes mellitus and cardiovascular diseases [11]. The TyG index, TyG-BMI index, and TyG-BRI index are emerging indicators in this field. These indices have demonstrated predictive utility for diabetes and ischemic stroke in longitudinal cohorts [12–14]. Nevertheless, research on these indices has been largely confined to a single chronic disease [15], and evidence regarding their association with coexisting conditions remains scarce. Moreover, the available findings remain somewhat inconsistent [16]. This suggests that such associations may be context-dependent, varying by population, age, sex, or even geographical setting.
Communities constitute a strategic locus for NCD prevention and control. Situated at the urban–rural interface, suburban communities face limited healthcare access and exhibit mixed lifestyle patterns reflecting both urban and rural influences. These features collectively elevate chronic disease risk [17, 18]. Residents of suburban communities often experience a “middle-ground” lifestyle, characterized by access to basic healthcare services but limited availability of specialized resources typically found in urban centers. Although suburban residents may be more physically active than their urban counterparts, environmental factors such as car-dependent commuting, sprawling built environments with limited walkability, and widespread availability of processed convenience foods may still increase the burden of metabolic risk factors [19]. Furthermore, suburban populations may exhibit distinct demographic profiles relative to urban or rural areas, such as a higher proportion of middle-class families, specific age distributions, and heterogeneous levels of health literacy [20, 21]. Previous studies have demonstrated that residents of the Qinghai region tend to have viscous and hypercoagulable bloodle [22]. This is attributable to distinctive ethnic compositions, geographical environments, and dietary habits. Acute hypoxic exposure reduces plasma triglycerides of specific carbon chain lengths and increases free fatty acids [23]. However, long-term native residents with adaptive genetic backgrounds (e.g., EPAS1 variants) may have developed compensatory networks. These networks actively offset hypoxia-induced lipid disturbances [24]. The TyG index is fundamentally linked to triglyceride metabolism. However, whether its association with cardiometabolic diseases persists in this specific environment remains unclear. Moreover, suburban communities in this region function as transitional zones between urban and rural areas. They face limited healthcare accessibility and exhibit mixed lifestyles. These factors may further increase metabolic phenotypic complexity. Therefore, this cross-sectional study aimed to examine the associations between TyG-related indices and HTN-DM co-occurrence among middle-aged and elderly individuals in a suburban community of Qinghai. The goal was to clarify whether these indices retain their metabolic relevance in this unique setting.
Methods
Study population and data content
In a cross-sectional study from February to July 2023, a total of 2854 participants aged ≥ 45 years from a suburban community in Qinghai were included after excluding those with missing information. The study was approved by the Ethics Committee of Qinghai University (LOT NO, 2022-48) and was conducted following the Declaration of Helsinki. All the participants signed informed consent forms. The information was collected by uniformly trained staff and was divided into three parts: baseline information, physical examination and laboratory testing. Baseline information included demographic characteristics (age, sex, ethnicity, job type, marital status and educational level) and lifestyle characteristics (frequency of smoking, drinking and consuming red and white meat; frequency of drinking salted tea or sweetened tea). Physical examination included height, weight and waist circumference. Height and waist circumference were measured to the nearest 0.1 cm, and weight to the nearest 0.1 kg. Blood pressure was measured three times with brief intervals between measurements on the left upper arm using a calibrated electronic sphygmomanometer after resting in a seated position for at least 5 min, and the mean of the three readings was used for analysis. Blood samples were collected in the morning after an overnight fast of 8–12 h, consistent with the National Health Commission of China WS/T 661–2020 Guideline for Venous Blood Specimen Collection. The following parameters were measured: fasting plasma glucose (FPG), total cholesterol (TC), fasting triglyceride (TG), high-density lipoprotein cholesterol (HDL-c), white blood cell count (WBC), platelet count (PLT), alanine aminotransferase (ALT), aspartate aminotransferase (AST) and low-density lipoprotein cholesterol (LDL-c). Detailed information on medication type, dosage, duration and adherence was not systematically recorded in this community-based screening program.
Definition of the variables
Ethnic groups were classified as Han or ethnic minorities, and occupations were dichotomized as mental versus manual labour. Educational attainment was divided into three categories: illiterate, primary school, and junior secondary school or higher. Household registration types were categorized as agricultural or non-agricultural. Marital status was classified as married or divorced/widowed. Living status was dichotomized as living alone or living with others. Smoking status was categorized as “yes” (current smoker) or “no” (current nonsmoker). Drinking status was categorized similarly.
Diagnostic criteria and definition of the index
HTN was defined as SBP ≥ 140 mm Hg and/or DBP ≥ 90 mm Hg and/or the use of antihypertensive medications [25]. Diabetes was defined as FPG ≥ 7.0 mmol/L and/or current treatment with antidiabetic medication according to the Chinese Guidelines for T2DM. HTN-DM comorbidity was defined as having both HTN and DM [26].
The formulas for TyG, BMI, BRI, and the composite indices (TyG-BMI and TyG-BRI) are as follows [2, 3, 27]:
 |
where TG and FPG both in mg/dL, respectively.
 |
where WC and height were measured in cm.
 |
The composite indices TyG-BMI and TyG-BRI were subsequently calculated as the product of the respective components.
The TyG index was treated as a continuous surrogate marker of insulin resistance, not a diabetes diagnostic tool. Despite the shared dependence on fasting glucose, we examined its association with HTN-DM independent of diabetes status to explore the role of insulin resistance in chronic disease comorbidity.
Statistical analysis
The baseline characteristics were summarised according to TyG quartile groups. Data were presented as medians with interquartile ranges (IQRs) for non-normally distributed continuous variables, and as means ± standard deviations (SDs) for normally distributed data. Frequencies and percentages were summarised for categorical variables. Differences among TyG quartile groups were analysed using the Kruskal-Wallis test or one-way ANOVA, as appropriate. TyG-related indices were analysed both as continuous variables (per 1-SD increase) and as quartiles based on the sample distribution. The median value of each quartile was assigned to the corresponding group as a continuous score to test for linear trend (P for trend) across quartiles. Multivariable logistic regression was performed to estimate the ORs and 95% CIs for the associations of TyG-related indices with HTN-DM. RCS were fitted to examine potential nonlinear associations between TyG-related indices and HTN-DM. ROC curve analysis was used to evaluate the discriminative ability of TyG-related indices. Pairwise AUC comparisons were performed using the DeLong test with Bonferroni correction for multiple comparisons. Subgroup analyses were performed by sex, age group, and other relevant strata. Multiplicative interaction terms were included to test for potential effect modification. All statistical analyses were performed using R software (version 4.2.3). A two-sided P value < 0.05 was considered statistically significant.
Results
Characteristics of the study participants
The participants in the study were divided into four groups (Q1: ≤8.20, 8.20 < Q2≤8.56, 8.56 < Q3≤8.96, and Q4:>8.96) on the basis of TyG quartile. The mean age of the participants were 70.24 ± 5.39 years, and 54.8% were female. The detection rates of HTN, DM, and the comorbidity of HTN and DM was 60.7%, 15.8%, and 10.7% respectively. The following baseline variables differed significantly among the four quartile groups: age, sex, registration, HTN, DM, HTN-DM, sweet, SBP, ALT, PLT, WBC, TB, TG, TC, HDL-c, and LDL-c (P < 0.05) (Table 1).
Table 1.
Baseline characteristics of the study population
| TyG Quartiles | |||||||
|---|---|---|---|---|---|---|---|
| Variables | Total | Q1 (<8. 20) |
Q2(8.20–8.56) | Q3(8.56–8.9) | Q4(>8.96) | P Value | |
| n = 2854 | n = 716 | n = 711 | n = 706 | n = 721 | |||
|
Sex (%) |
Female | 1565 (54. 8) | 297 (41. 5) | 388 (54. 6) | 420 (59. 5) | 460 (63. 8) | < 0. 001 |
| Male | 1289 (45. 2) | 419 (58. 5) | 323 (45. 4) | 286 (40. 5) | 261 (36. 2) | ||
|
age_group (%) |
<65 | 258 ( 9. 0) | 52 ( 7. 3) | 60 ( 8. 4) | 62 ( 8. 8) | 84 (11. 7) | 0. 028 |
| ≥ 65 | 2596 (91. 0) | 664 (92. 7) | 651 (91. 6) | 644 (91. 2) | 637 (88. 3) | ||
|
Ethnicity (%) |
Han | 2523 (88. 4) | 647 (90. 4) | 632 (88. 9) | 623 (88. 2) | 621 (86. 1) | 0. 09 |
| Minority | 331 (11. 6) | 69 ( 9. 6) | 79 (11. 1) | 83 (11. 8) | 100 (13. 9) | ||
|
Registration (%) |
Agricultural | 2650 (92. 9) | 674 (94. 1) | 670 (94. 2) | 649 (91. 9) | 657 (91. 1) | 0. 047 |
| Non-agricultural | 204 ( 7. 1) | 42 ( 5. 9) | 41 ( 5. 8) | 57 ( 8. 1) | 64 ( 8. 9) | ||
|
marital (%) |
Married | 2009 (70. 4) | 532 (74. 3) | 509 (71. 6) | 474 (67. 1) | 494 (68. 5) | 0. 014 |
| Divorced/Widowed | 845 (29. 6) | 184 (25. 7) | 202 (28. 4) | 232 (32. 9) | 227 (31. 5) | ||
|
Medical (%) |
1urban employee medical insurance | 105 ( 3. 7) | 24 ( 3. 4) | 21 ( 3. 0) | 36 ( 5. 1) | 24 ( 3. 3) | 0. 167 |
| other | 10 ( 0. 4) | 5 ( 0. 7) | 1 ( 0. 1) | 2 ( 0. 3) | 2 ( 0. 3) | ||
| Resident medical | 2739 (96. 0) | 687 (95. 9) | 689 (96. 9) | 668 (94. 6) | 695 (96. 4) | ||
|
Education (%) |
Illiteracy | 1674 (58. 7) | 396 (55. 3) | 408 (57. 4) | 433 (61. 3) | 437 (60. 6) | 0. 11 |
| Primary and junior high school | 1071 (37. 5) | 292 (40. 8) | 278 (39. 1) | 240 (34. 0) | 261 (36. 2) | ||
| High school level above | 109 ( 3. 8) | 28 ( 3. 9) | 25 ( 3. 5) | 33 ( 4. 7) | 23 ( 3. 2) | ||
| worker | 148 ( 5. 2) | 33 ( 4. 6) | 26 ( 3. 7) | 48 ( 6. 8) | 41 ( 5. 7) | ||
|
Occupation (%) |
Farmers and herders | 2617 (91. 7) | 666 (93. 0) | 666 (93. 7) | 634 (89. 8) | 651 (90. 3) | 0.069 |
| technical personnel | 42 ( 1. 5) | 8 ( 1. 1) | 11 ( 1. 5) | 13 ( 1. 8) | 10 ( 1. 4) | ||
| other | 47 ( 1. 6) | 9 ( 1. 3) | 8 ( 1. 1) | 11 ( 1. 6) | 41 ( 5. 7) | ||
| Live with others | 2626 (92.0) | 661 (92.3) | 655 (92.1) | 641 (90.8) | 669 (92.8) | ||
|
Living_Status (%) |
Live alone | 228 ( 8. 0) | 55 ( 7. 7) | 56 ( 7. 9) | 65 ( 9. 2) | 52 ( 7. 2) | 0. 548 |
|
Total_Income (CNY, %) |
<1000 | 2213 (77. 5) | 566 (79. 1) | 553 (77. 8) | 536 (75. 9) | 558 (77. 4) | 0. 302 |
| 1000–3000 | 505 (17. 7) | 116 (16. 2) | 133 (18. 7) | 128 (18. 1) | 128 (17. 8) | ||
| 3001–5000 | 76 ( 2. 7) | 23 ( 3. 2) | 10 ( 1. 4) | 25 ( 3. 5) | 18 ( 2. 5) | ||
| > 5000 | 60 ( 2. 1) | 11 ( 1. 5) | 15 ( 2. 1) | 17 ( 2. 4) | 17 ( 2. 4) | ||
|
Smoking (%) |
yes | 417 (14. 6) | 108 (15. 1) | 98 (13. 8) | 110 (15. 6) | 101 (14. 0) | 0. 742 |
|
drinking (%) |
yes | 437 (15. 3) | 99 (13. 8) | 107 (15. 0) | 105 (14. 9) | 126 (17. 5) | 0. 265 |
|
HTN (%) |
yes | 1731 (60. 7) | 382 (53. 4) | 408 (57. 4) | 460 (65. 2) | 481 (66. 7) | < 0. 001 |
|
diabetes (%) |
yes | 451 (15. 8) | 15 ( 2. 1) | 59 ( 8. 3) | 97 (13. 7) | 280 (38. 8) | < 0. 001 |
|
HTN_DM (%) |
yes | 304 (10. 7) | 10 ( 1. 4) | 37 ( 5. 2) | 66 ( 9. 3) | 191 (26. 5) | < 0. 001 |
|
Salty (%) |
often | 1399 (49. 0) | 364 (50. 8) | 360 (50. 6) | 337 (47. 7) | 338 (46. 9) | 0. 627 |
| occasionally | 480 (16. 8) | 110 (15. 4) | 121 (17. 0) | 124 (17. 6) | 125 (17. 3) | ||
| Hardly ever eat | 975 (34. 2) | 242 (33. 8) | 230 (32. 3) | 245 (34. 7) | 258 (35. 8) | ||
| n = 2854 | n = 716 | n = 711 | n = 706 | n = 721 | |||
| often | 451 (15.8) | 131 (18.3) | 105 (14.8) | 111 (15.7) | 104 (14.4) | 0. 001 | |
| sweet (%) | occasionally | 726 (25. 4) | 195 (27.2) | 195 (27.4) | 191 (27. 1) |
145 (20. 1) |
|
| Hardly ever eat | 1677 (58.8) | 390 (54.5) | 411 (57.8) | 404 (57. 2) | 472 (65.5) | ||
| Red meat (%) | often | 2564 (89. 8) | 654 (91.3) | 645 (90.7) | 625 (88. 5) | 640 (88.8) | |
| occasionally | 247 ( 8.7) | 56 ( 7.8) | 53 ( 7.5) | 72 (10.2) | 66 ( 9.2) | 0. 184 | |
| Hardly ever eat | 43 ( 1.5) | 6 ( 0.8) | 13 ( 1.8) | 9 ( 1.3) | 15 ( 2.1) | ||
|
FPG (median [IQR]) |
5. 30 (4. 9, 6. 1) |
5.00 (4.5,5.3) |
5. 20 (4. 8,5.7) |
5.40 (5.0,6.0) |
6.2 (5.4,7.9) |
< 0. 001 | |
|
Waist (median [IQR]) |
86. 0 (80. 0,93. 0) |
83. 0 (76. 0,90.0) |
84.00 (78, 91) |
87.0 (81.0,94. 0) |
89. 0 (82.0,95.0) |
< 0. 001 | |
|
SBP (median [IQR]) |
140. 0 (129. 0,155. 0) |
138.0 (126.0,152.0) |
140. 0 (129,153.5) |
142. 0 (129,157) |
144.0 (131.0,157.0) |
< 0. 001 | |
|
DBP (median [IQR]) |
84. 0 ( 76. 0,90. 0) |
83. 0 (74. 0,90. 00) |
83. 0 (76. 0,90. 0) |
84.0 (77,91.0) |
85.0 (77. 0,92. 0) |
0. 042 | |
|
ALT (median [IQR]) |
11. 68 (8. 93,15. 95) |
10. 39 (8. 32,13. 98) |
11.14 (8. 64,15. 25) |
12.45 (9.30,16.29) |
12.98 (9. 96,18. 29) |
< 0. 001 | |
|
AST (median [IQR]) |
21.37 (17. 91, 26. 10) |
21.38 (18.28,26.04) |
21.33 (17.84,25.56) |
21.25 (18.08,26.2) |
21.40 (17.62,26.5) |
0. 816 | |
|
TB (median [IQR]) |
11. 68 (9.03,15.48) |
12. 59 (9. 74,16. 71) |
11. 89 (9.26,15.86) |
11.40 (8.74,15.09) |
10. 86 (8.58,13.81) |
< 0. 001 | |
|
TC (median [IQR]) |
4. 32 (3.76, 4.96) |
3. 89 (3. 45,4.35) |
4.28 (3.75,4.85) |
4.52 (3. 91,5.09) |
4.73 (4.14,5.43) |
< 0. 001 | |
|
TG (median [IQR]) |
1. 19 (0. 87,1.69) |
0. 72 (0. 63,0.82) |
1. 05 (0. 94,1.17) |
1. 42 (1.26,1. 62) |
2.15 (1.78,2.73) |
< 0. 001 | |
|
HDL-c (median [IQR]) |
98 (0.81,1.16) |
1. 08 (0.91,1. 28) |
1. 03 (0.87,1.21) |
0.93 (0. 80,1.09) |
0. 85 (0.71,1.03) |
< 0. 001 | |
|
LDL-c (median [IQR]) |
2.29 (1. 92,2. 69) |
2.02 (1.73,2. 33) |
2.28 (1. 92,2. 63) |
2.43 (2.04,2.83) |
2. 49 (2. 10,2. 91) |
< 0. 001 | |
|
Urea (median [IQR]) |
5. 53 (4.59, 6.74) |
5. 56 (4. 58,6.76) |
5. 43 (4. 51,6. 56) |
5.56 (4.70,6.76) |
5. 63 (4. 58,6. 82) |
0. 097 | |
|
SCR (median [IQR]) |
62.02 (54.09,72.12) |
62.38 (54.54,72.8) |
62.03 (53. 60,71. 90) |
62.94 (54.49,72.7) |
61.29 (53.94,71. 98) |
0. 538 | |
|
WBC (median [IQR]) |
5. 69 (4.74,6. 69) |
5.31(4.48,6.26) |
5. 57 (4. 62,6. 40) |
5.84 (4 85,7.02) |
6. 06 (5. 11,7.09) |
< 0. 001 | |
|
Hb (median [IQR]) |
148. 0 (137.0,161.0) |
148.0 (137.0,161.0) |
0 (137.0,160.0) |
148.0 (137.0,161.0) |
150.0 (139.0,161. 0) |
0. 764 | |
|
PLT (median [IQR]) |
188.0 (150,231.0 ) |
179.0 (142.7,223.5) |
189. 0 (152. 0,233) |
189. 0 (155.0,231) |
193.0 (153,237) |
< 0. 001 | |
|
Happiness (median [IQR]) |
8.0(7.9) | 8.0(8.9) | 8. 0(7. 9) | 8. 0(7.9) | 8. 0(7.9) | 0. 195 | |
Association between the TyG-related indices and HTN- DM
Table 2 presents the multivariable associations. To avoid overadjustment, Model 3 excluded TG and FPG—the two components of the TyG index—while adjusting for all remaining covariates.In continuous analyses, higher TyG was associated with increased odds of HTN-DM 3.16(95% CI:2.75–3.64) in the fully adjusted model. The derived indices showed attenuated but significant associations: TyG-BMI 1.39(95% CI:1.15–1.66) and TyG-BRI 1.37(95% CI:1.22–1.54). In categorical analyses, participants in the highest quartile (Q4) exhibited higher odds of HTN-DM than those in the lowest quartile (Q1) across all indices (all P for trend < 0.001). The adjusted ORs (95% CIs) for Q4 were 30.99 (16.97–62.97) for TyG, 6.42 (4.20-10.14) for TyG-BMI, and 6.43 (4.21–10.17) for TyG-BRI. In sensitivity analyses, TyG index ORs ranged from 2.85 to 3.21, with all estimates statistically significant, indicating robustness across model specifications(Supplementary Table 2).
Table 2.
Multivariable associations of TyG and its derived indices with HTN-DM across progressive adjustment models
| Variables | Model 1 | Model 2 | Model 3 | |||
|---|---|---|---|---|---|---|
| OR (95% CI) | P value | OR (95% CI) | P value | OR (95% CI) | P value | |
| TyG index | 2.85 (2.51–3.25) | < 0.001 | 2.85 (2.51–3.25) | < 0.001 | 3.16 (2.75–3.64) | < 0.001 |
|
TyG index quartile Q1 |
Reference | Reference | Reference | |||
| Q2 | 3.75 (1.95–7.90) | 3.78 (1.96–7.95) | < 0.001 | 4.20 (2.17–8.86) | < 0.001 | |
| Q3 | 7.02 (3.78–14.39) | < 0.001 | 7.04 (3.79–14.44) | < 0.001 | 8.27 (4.42–17.05) | < 0.001 |
| Q4 | 24.58 (13.70–49.33) | < 0.001 | 24.70 (13.76–49.59) | < 0.001 | 30.99 (16.97–62.97) | < 0.001 |
| P for trend | < 0.001 | < 0.001 | < 0.001 | |||
| TyG-BMI index | 1.44 (1.20–1.71) | < 0.001 | 1.44 (1.20–1.71) | < 0.001 | 1.39 (1.15–1.66) | < 0.001 |
|
quartile Q1 |
Reference | Reference | Reference | |||
| Q2 | 2.52 (1.60–4.07) | < 0.001 | 2.55 (1.62–4.13) | < 0.001 | 2.51 (1.59–4.07) | < 0.001 |
| Q3 | 3.02 (1.94–4.85) | < 0.001 | 3.07 (1.97–4.93) | < 0.001 | 2.97 (1.89–4.78) | < 0.001 |
| Q4 | 6.40 (4.23–10.04) | < 0.001 | 6.46 (4.26–10.14) | < 0.001 | 6.42 (4.20–10.14) | |
| P for trend | < 0.001 | < 0.001 | < 0.001 | |||
| TyG-BRI index | 1.39 (1.25–1.55) | < 0.001 | 1.39 (1.24–1.55) | < 0.001 | 1.37 (1.22–1.54) | < 0.001 |
|
quartile Q1 |
Reference | Reference | Reference | < 0.001 | ||
| Q2 | 2.52 (1.60–4.09) | < 0.001 | 2.56 (1.62–4.14) | < 0.001 | 2.52 (1.60–4.09) | < 0.001 |
| Q3 | 3.03 (1.94–4.86) | < 0.001 | 3.07 (1.97–4.94) | < 0.001 | 2.97 (1.90–4.80) | |
| Q4 | 6.41 (4.23–10.06) | < 0.001 | 6.47 (4.27–10.15) | < 0.001 | 6.43 (4.21–10.17) | < 0.001 |
| P for trend | < 0.001 | < 0.001 | < 0.001 |
Model 1: Adjusted for age, sex;
Model 2: Adjusted for age, sex, education level and marital status
Model 3:Adjusted for age, sex, education level and marital status, current smoking, current drinkings, Sweet frequency, TC, HDL-c,
LDL-c, ALT, WBC, and PLT
Abbreviations: OR Odds ratio, CI Confidence interval
The RCS analysis showed a linear association between TyG and HTN-DM (P for nonlinear > 0.05). Nonlinear associations were observed for TyG-BMI and TyG-BRI (all P for nonlinear < 0.05) (Fig. 1).As shown in Table 3, all three TyG-related indices demonstrated significant discriminatory performance for identifying HTN-DM (all P < 0.001). DeLong tests further confirmed that the TyG index had a significantly higher discriminatory ability than both the TyG-BMI and TyG-BRI indices (all Bonferroni-adjusted pairwise comparisons, P < 0.05).TyG had the highest AUC (0.786) among the three indices. This value exceeded those of TyG-BMI (0.679) and TyG-BRI (0.633). TyG therefore demonstrated better discriminative ability for HTN-DM than the other two indices (Fig. 2).
Fig. 1.
Restricted cubic spline curve for the association of TyG related indices and HTN -DM A RCS analysis of the TyG index for HTN-diabetes comorbidity B RCS analysis of the TyG-BMI index for HTN-diabetes comorbidity C RCS analysis of the TyG-BRI index for HTN-diabetes comorbidity. The solid red lines represent the odds ratios, and the shaded areas indicate 95% confidence intervals
Table 3.
Discriminative performance and optimal cut-off points of TyG-related indices for identifying HTN-DM
| Index | AUC (95% CI) | Cut-off | Sensitivity | Specificity | Youden’s Index | PPV | NPV | P |
|---|---|---|---|---|---|---|---|---|
| TyG index | 0.786 (0.759–0.813) | 8.871 | 67.76 | 74.78 | 0.4255 | 24.26 | 95.11 | < 0.001 |
| TyG-BMI index | 0.679(0.647–0.710) | 225.674 | 67.11 | 58.43 | 0.2554 | 16.14 | 93.71 | < 0.001 |
| TyG-BRI index | 0.632 (0.601–0.664) | 37.327 | 62.50 | 57.61 | 0.2011 | 14.95 | 92.80 | < 0.001 |
All pairwise differences were statistically significant (P < 0.001 for all comparisons)
P values were Bonferroni-adjusted for three pairwise comparisons
※A Pairwise comparisons of AUCs were performed using the DeLong test
Fig. 2.
ROC curves of TyG-related indices for identifying HTN-DM in the study population
Subgroup analyses
Interaction tests showed that sex was statistically significant (P for interaction = 0.045). In all subgroups, TyG values had a consistent positive correlation with risk of HTN-DM (Fig. 3).
Fig. 3.
Stratified analyses of the associations between TyG index and HTN-DM. Adjusted for age, sex, education level and marital status, current smoking, current drinkings, Sweet frequency, TC, HDL-c, LDL-c, ALT, WBC, and PLT
Discussion
DM and HTN are major global public health issues posing a rapidly increasing burden [28–30]. HTN-DM comorbidity further elevates the risk of stroke, kidney disease, and retinopathy [29, 30]. Age-specific data reveal escalating burdens: the prevalence of HTN, DM, and their co-prevalence among Chinese adults aged 65–75 years was 59.2%, 25%, and 17.7%, respectively, which rose to 67.6%, 26.7%, and 20.6% in those over 75 years [31].
HTN imposed a particularly heavy burden in this cohort, with a prevalence of 60.7% that markedly surpassed the 2018 national average for middle-aged and elderly populations (46.0%). The prevalence of DM and HTN-DM comorbidity was 15.8% and 10.7%, respectively [7]. The unique climate and dietary practices of Qinghai—characterized by hypoxia, cold temperatures, a high-fat and high-salt diet, and alcohol consumption—may contribute to a significant disease burden [32, 33]. However, suburban communities face a unique convergence of risk factors. Residents in these settings often maintain traditional dietary habits while experiencing declines in physical activity and increases in psychosocial stress amid rapid urbanization [34, 35]. Moreover, primary healthcare access is frequently fragmented—community health centers may lack adequate screening equipment or trained personnel for early HTN and DM detection. Residents in these transitional zones also demonstrate lower health literacy and poorer medication adherence than their urban counterparts, partly attributable to limited health education and higher out-of-pocket costs [36]. Given the rapidly aging population, the rising prevalence of HTN and DM in suburban communities in Qinghai underscores the urgent need for early detection and intervention.
In this study, the TyG index, TyG-BMI, and TyG-BRI were each significantly associated with HTN-DM after adjustment for covariates. This finding aligns with a growing body of evidence supporting the role of IR surrogates in identifying individuals with concurrent cardiometabolic conditions [10, 11]. Notably, although all indices demonstrated discriminative ability, the TyG index alone exhibited the highest observed AUC among the evaluated indices. This suggests that integrating anthropometric parameters with metabolic indices may not consistently improve discriminative performance beyond that achieved by the metabolic parameter alone. This finding is noteworthy, as several previous studies have advocated for TyG-BMI or TyG-WC as superior surrogate markers of IR and related cardiometabolic outcomes [37, 38].This findings align with those study in Iraq [39], Ke et al. [40], Vedika Rathore [39], and Selvi et al. [41] ependent performance of TyG indices. However, contrasting evidence exist [38, 42, 43]. The proliferation of TyG-derived indices—particularly those combining TyG with anthropometric measures—presents an epidemiological puzzle that merits critical examination. If the basic TyG index already serves as a robust surrogate for IR [44] and integrates fasting glucose and triglycerides—two core metabolic parameters that directly reflect the pathophysiological drivers of both hypertension and diabetes [44, 45]—why do researchers persitently combine it with various anthropometric measures? We suspect this trend reveals more about our methodological anxieties than about metabolic physiology.It reflects a belief that adding complexity (e.g., TyG-BMI or TyG-BRI) will inevitably enhance discriminative performance, an assumption that needs scrutiny.
The RCS analysis revealed distinct dose-response patterns across the three indices. TyG demonstrated a linear relationship with HTN-DM, suggesting a consistent proportional increase in odds across its range. In contrast, both TyG-BMI and TyG-BRI showed significant nonlinear relationships with HTN-DM, indicating that the strength of association varied at different levels of these composite indices. In categorical analyses, the odds of HTN-DM increased across quartiles, with a marked elevation in Q4 for all indices. The OR for TyG in Q4 was much higher than those for TyG-BMI and TyG-BRI, indicating that combined indices did not strengthen the association. The high OR for TyG in Q4 may be due to mathematical overlap as TyG includes fasting glucose and diabetes mellitus definition relies on glucose thresholds. However, per1-SD continuous estimates are less affected by this overlap. All indices had significant positive associations in the fully adjusted model, with TyG having the strongest effect. Biologically, TyG reflects IR, which is the shared pathophysiological link between hypertension and type2 diabetes [46]. IR-driven hyperinsulinemia activates the sympathetic nervous system and the renin-angiotensin-aldosterone system [47]. This promotes vasoconstriction and renal sodium retention. IR also impairs endothelial nitric oxide bioavailability and amplifies oxidative stress through selective impairment of the PI3K/Akt pathway [48]. These molecular events create a common cardiometabolic soil. Both hyperglycemia and elevated blood pressure arise from this soil. It is possible that this biological gradient is not merely a mathematical artifact, though further investigation is warranted. A meta-analysis of cohort studies demonstrated that elevated TyG predicted new-onset hypertension in the general population [49]. This association remained robust across sensitivity and subgroup analyses. Furthermore, a retrospective study of Japanese normoglycemic adults reported a significant association between TyG and hypertension even in the absence of overt glucose dysregulation [50].These convergent findings are consistent with the hypothesis that the association between TyG and HTN-DM may reflect a metabolic continuum, though this interpretation requires prospective validation and formal assessment of clinically useful thresholds.
The observed nonlinear pattern is compatible with the hypothesis that mild-to-moderate insulin resistance may be accompanied by compensatory hyperinsulinemia, whereas beyond a certain threshold, decompensation may coincide with concurrent disturbances in glycemic and blood pressure regulation [51]. Importantly, this threshold effect may be more accurately captured by the TyG index alone, whereas multiplying the TyG with BMI or BRI (e.g., TyG-BMI) tends to smooth out or mask this critical turning point and may distort the dose-response curve.
A particularly intriguing finding was that sex was statistically significant as an effect modifier, suggesting that the association between TyG and HTN-DM may differ between males and females. This aligns with the findings of a previous study [52]. Although our findings show a stronger association in certain sex subgroups, the question of whether sex-specific TyG cut-offs would be clinically useful requires prospective validation and formal threshold evaluation before any practical recommendation can be made. Future studies should explore the mechanisms underlying this interaction.
The coexistence of hypertension and diabetes confers a synergistic cardiovascular burden that exceeds the sum of individual risks. In a 7-year CHARLS cohort, comorbid patients demonstrated a 2.04-fold risk of all-cause mortality and a 2.01-fold risk of CVD, with significant synergistic interaction beyond additive effects (RERI = 0.567) [53]. More importantly, the blood pressure control rate in comorbid patients is only 28% of that in patients with hypertension alone, and the rate of pulse pressure widening is significantly accelerated, leading to progressive pulse pressure widening [54]. This means that even when blood pressure and blood glucose levels are “within the target range,” the vascular risk in comorbid patients continues to accumulate; monitoring traditional indicators alone cannot capture this hidden process.
Meanwhile, the core value of the TyG index lies in risk stratification rather than disease diagnosis. Calculated from routine fasting glucose and triglycerides, it requires no additional cost or specialized testing. Evidence from a recent NHANES-based cohort study of patients with comorbid hypertension and diabetes revealed U-shaped associations between the TyG index and all-cause and cardiovascular mortality, with thresholds at 9.37 and 8.87, respectively [55]. This allows patients to be stratified into low-risk, intermediate-risk, and very-high-risk categories [55], guiding precise referral and tiered management in suburban communities. This information cannot be provided by blood pressure and blood glucose values alone.
Conventional monitoring of BP and glucose alone may be insufficient to capture the metabolic drivers of poor control. In a large hypertensive cohort (n = 99,336), the TyG index and HbA1c were independently associated with inadequate BP control, and their interaction showed significant synergistic effects on control failure (RERI = 1.13, AP = 0.69) [56]. As a cost-free marker derived from routine fasting glucose and triglyceride measurements, the TyG index reflects insulin resistance and metabolic dysregulation that blood pressure and glucose values alone cannot fully capture. These preliminary findings raise the possibility that incorporating this composite metabolic indicator into risk assessment may offer a potential approach for stratifying and managing HTN-DM comorbidity, particularly in resource-limited community settings, although this still requires prospective validation.
In summary, blood pressure and glucose are used to identify the presence of diseases, while the TyG index quantifies metabolic risk. These complementary tools support a shift from diagnosis to risk stratification, especially in resource-limited suburban settings. However, due to the cross - sectional design, these findings serve to inform risk stratification rather than predict future HTN-DM, and prospective validation is still necessary. The study population was drawn from a suburban community, and the conclusions are limited to this particular suburban community. Future prospective cohort studies and multi-region comparative studies are required to further validate these findings.
Limitations
Firstly, the cross-sectional design precludes the establishment of causality between the TyG-related indices and HTN-DM. Prospective longitudinal studies are needed to address this issue. Secondly, although multiple confounding factors were adjusted, potential unmeasured confounders (e.g., genetic and environmental factors, medication histories) may still influence the results. Third, the non-random selection of participants and the limited ethnic diversity (predominantly Han, with only Tibetan and Hui minorities represented) restrict the generalizability of our findings to broader populations. Fourth, blood pressure was recorded as the mean of three consecutive readings at a single visit. This method may not reflect white-coat effects, diurnal variation, or long-term blood pressure trajectories. Fifth, the TyG index shares mathematical overlap with the glucose threshold used to define HTN-DM, which may inflate ORs in the highest quartile; while per 1-SD estimates and sensitivity analyses were provided, the observed association may partly reflect this overlap, and any inference regarding a biological gradient requires prospective validation. Sixth, this study was limited to suburban community, a transitional demographic between urban and rural settings. Findings should be generalized cautiously to purely urban or rural populations. Future prospective cohort studies and multi-region comparative studies are needed to further validate these findings.
Conclusions
This study revealed that TyG-related indices were associated with HTN-DM co-occurrence. RCS analysis indicated a linear association for TyG, whereas nonlinear associations were observed for TyG-BMI and TyG-BRI. The TyG index alone exhibited the highest observed AUC among the evaluated indices, exceeding that of TyG-BMI and TyG-BRI, suggesting that the incorporation of obesity measures does not necessarily strengthen the observed association. These findings indicate that the TyG index may serve as a simple and accessible marker associated with HTN-DM co-occurrence in community-based settings, although prospective validation is warranted.
Supplementary Information
Acknowledgements
Thanks to all the participants who understood and supported this study and to all the staff who helped to collect the questionnaires and data.
Abbreviations
- TyG
index Triglyceride-glucose index
- HTN DM
Hypertensive-diabetes comorbidity
- TyG BMI
Triglyceride glucose-body mass index
- TyG BRI
Triglyceride glucose-Body roundness index
- SBP
Systolic blood pressure
- DBP
Diastolic blood pressure
- FPG
Fasting plasma glucose
- TC
Total cholesterol
- TG
Triglyceride
- HDL-c
High density lipoprotein cholesterol
- LDL-c
Low density lipoprotein cholesterol
- ROC
Receiver Operating Characteristic curves
- AUC
Area Under the Curve
- RCS
Restricted Cubic Spline
- WBC
White blood cell
- PLT
Platelet
- Hb
Hemoglobin
- ALT
Alanine aminotransferase, aspartate aminotransferase
- AST
Aspartate aminotransferase
- PPV
Positive Predictive Value
- NPV
Negative Predictive Value
Authors’ contributions
Zhancui Dang contributed to the study conception, design, review and edit the manuscript. Shou Liu contribute to collect data, review and edit the manuscript. All authors read and approved the final manuscript. Fanlian Kong performed the statistical analyses, interpreted the data and wrote the paper.
Funding
This work was supported by Plateau Big Health Science and Technology Backyard (qdyjd-2508).
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Declarations
Ethics approval and consent to participate
The study was approved by the Ethics Committee of Qinghai University (LOT NO, 2022-48) and was conducted following the Declaration of Helsinki. All individual participants included in the study gave their informed consent.
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
Shou Liu, Email: liushou2004@aliyun.com.
Zhancui Dang, Email: yxyyy2019@163.com.
References
- 1.Zhang X, Ma N, Lin Q, Chen K, Zheng F, Wu J, Dong X, Niu W. Body roundness index and all-cause mortality among US adults. JAMA Netw Open. 2024;7(6):e2415051. 10.1001/jamanetworkopen.2024.15051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Er LK, Wu S, Chou HH, Hsu LA, Teng MS, Sun YC, Ko YL. Triglyceride glucose-body mass index is a simple and clinically useful surrogate marker for insulin resistance in nondiabetic individuals. PLoS ONE. 2016;11(3):e0149731. 10.1371/journal.pone.0149731. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chen X, Du X, Lu F, Zhang J, Xu C, Liang M, Chen L, Zhong J. The association between the triglyceride-glucose index, its combination with the body roundness index, and chronic kidney disease in patients with type 2 diabetes in eastern China: a preliminary study. Nutrients. 2025;17(3):492. 10.3390/nu17030492. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zhang H, et al. Global burden of metabolic diseases, 1990–2021. Metab Clin Exp. 2024;160:155999. 10.1016/j.metabol.2024.155999. [DOI] [PubMed] [Google Scholar]
- 5.Chowdhury SR, Chandra Das D, Sunna TC, Beyene J, Hossain A. Global and regional prevalence of multimorbidity in the adult population in community settings: a systematic review and meta-analysis. EClinicalMedicine. 2023;57:101860. 10.1016/j.eclinm.2023.101860. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Liu H, Yin P, Qi J, Zhou M. Burden of non-communicable diseases in China and its provinces, 1990–2021: results from the Global Burden of Disease Study 2021. Chin Med J. 2024;137(19):2325–33. 10.1097/CM9.0000000000003270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Yu N, Zhang M, Zhang X, Zhao ZP, Li C, Huang ZJ, Gao XX, Zhang WR, Yu MT, Zhang YS, Deng XQ, Wang LM. Prevalence and influencing factors of comorbidities of hypertension, diabetes and dyslipidemia among middle-aged and older adults in China. Chin J Epidemiol. 2023;44(2):196–204. 10.3760/cma.j.cn112338-20220523-00451. [DOI] [Google Scholar]
- 8.Ren Y, Cheng L, Qie R, Han M, Kong L, Yan W, Li Z, Li Y, Lei Y. Dose-response association of Chinese visceral adiposity index with comorbidity of hypertension and diabetes mellitus among elderly people. Front Endocrinol. 2023;14:1187381. 10.3389/fendo.2023.1187381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zhou YF, Zhang XT, Zeng QT, He HB. The association of the atherogenic index of plasma with hypertension, diabetes, and their comorbidities in Chinese middle-aged and elderly people: a cross-sectional study from CHARLS. Front Nutr. 2025;12:1607601. 10.3389/fnut.2025.1607601. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Su W, Wang J, Chen K, Yan W, Gao Z, Tang X, Wan Q, Luo Z, Ning G, Mu Y. A higher TyG index level is more likely to have enhanced incidence of T2DM and HTN comorbidity in elderly Chinese people: a prospective observational study from the REACTION study. Diabetol Metabolic Syndrome. 2024;16(1):29. 10.1186/s13098-024-01258-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Li X, Sun M, Yang Y, Yao N, Yan S, Wang L, Hu W, Guo R, Wang Y, Li B. Predictive effect of triglyceride glucose-related parameters, obesity indices, and lipid ratios for diabetes in a Chinese population: a prospective cohort study. Front Endocrinol. 2022;13:862919. 10.3389/fendo.2022.862919. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wang B, Li L, Tang Y, Ran X. Joint association of triglyceride glucose index (TyG) and body roundness index (BRI) with stroke incidence: a national cohort study. Cardiovasc Diabetol. 2025;24(1):164. 10.1186/s12933-025-02724-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lim J, Kim J, Koo SH, Kwon GC. Comparison of triglyceride glucose index, and related parameters to predict insulin resistance in Korean adults: an analysis of the 2007–2010 Korean National Health and Nutrition Examination Survey. PLoS ONE. 2019;14(3):e0212963. 10.1371/journal.pone.0212963. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ramdas Nayak VK, Nayak KR, Vidyasagar S, Rekha P. Predictive performance of traditional and novel lipid combined anthropometric indices to identify prediabetes. Diabetes Metabolic Syndrome. 2020;14(5):1265–72. 10.1016/j.dsx.2020.06.045. [DOI] [PubMed] [Google Scholar]
- 15.Song K, Xu Y, Wu S, Zhang X, Wang Y, Pan S. Research status of triglyceride glucose-body mass index (TyG-BMI index). Front Cardiovasc Med. 2025;12:1597112. 10.3389/fcvm.2025.1597112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Du Z, Xing L, Lin M, Sun Y. Estimate of prevalent ischemic stroke from triglyceride glucose-body mass index in the general population. BMC Cardiovasc Disord. 2020;20(1):483. 10.1186/s12872-020-01768-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Diez Roux AV, Mair C. Neighborhoods and health. Annals of the New York Academy of Sciences. 2010;1186:125–45. 10.1111/j.1749-6632.2009.05333.x [DOI] [PubMed]
- 18.Zhang F, Li W. Uncovering the subtle relationship between vitamin D and kidney stones: a cross-sectional NHANES-based study. Eur J Med Res. 2025;30(1):202. 10.1186/s40001-025-02474-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Iamtrakul P, Chayphong S. Analyzing the link between built environment and physical activity: a spatial study in suburban area. Front Built Environ. 2024;10:1420020. 10.3389/fbuil.2024.1420020. [DOI] [Google Scholar]
- 20.Mshelia Y, Adekeye O, Zirra A. One-health KAP measures and their influence on health outcomes in suburban population: a study of Karu, Nasarawa State in Nigeria. Res Square. 2025:1–26. 10.21203/rs.3.rs-6600624/v1 [DOI]
- 21.Velmurugan G, Mohanraj S, Dhivakar M, Veerasekar G, Brag-Gresham J, He K, Alexander T, Cherian M, Saran R, Pradeep T, Swaminathan K. Differential risk factor profile of diabetes and atherosclerosis in rural, suburban and urban regions of South India: the KMCH-non-communicable disease studies. Diabet Med. 2021;38(6):e14466. 10.1111/dme.14466. [DOI] [PubMed] [Google Scholar]
- 22.Peng W, Wang YX, Wang HJ, Li K, Sun XM, Wang YF. The prevalence and associated factors of metabolic syndrome among Tibetan pastoralists in transition from nomadic to settled urban environment. Chin J Epidemiol. 2022;43(4):533–40. 10.3760/cma.j.cn112338-20211118-00900. [DOI] [PubMed] [Google Scholar]
- 23.O’Brien KA, Atkinson RA, Richardson L, Koulman A, Murray AJ, Harridge SDR, Martin DS, Levett DZH, Mitchell K, Mythen MG, Montgomery HE, Grocott MPW, Griffin JL, Edwards LM. Metabolomic and lipidomic plasma profile changes in human participants ascending to Everest Base Camp. Sci Rep. 2019;9(1):2297. 10.1038/s41598-019-38832-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lin Y, Xu X, Xie H, et al. Homeostasis of glucose and lipid metabolism during physiological responses to a simulated hypoxic high altitude environment. Nat Commun. 2025;16(1):9406. 10.1038/s41467-025-64110-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Mills KT, Stefanescu A, He J. The global epidemiology of hypertension. Nat Rev Nephrol. 2020;16(4):223–37. 10.1038/s41581-019-0244-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Yimam M, Desse TA, Hebo HJ. Glycemic control among ambulatory type 2 diabetes patients with hypertension comorbidity in a developing country: a cross-sectional study. Heliyon. 2020;6(12):e05671. 10.1016/j.heliyon.2020.e05671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Weir CB, Jan A. BMI classification percentile and cut off points. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2025. Available from: https://www.ncbi.nlm.nih.gov/books/NBK541070/
- 28.Khan MAB, Hashim MJ, King JK, Govender RD, Mustafa H, Al Kaabi J. Epidemiology of type 2 diabetes - global burden of disease and forecasted trends. J Epidemiol Global Health. 2020;10(1):107–11. 10.2991/jegh.k.191028.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Oliveros E, et al. Hypertension in older adults: assessment, management, and challenges. Clin Cardiol. 2020;43(2):99–107. 10.1002/clc.23303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.GBD 2017 Risk Factor Collaborators. Global, regional, and national comparative risk assessment of 84 behavioural, environmental and occupational, and metabolic risks or clusters of risks for 195 countries and territories, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet. 2018;392(10159):1923–94. 10.1016/S0140-6736(18)32225-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Fasehun OO, Adjei-Mensah J, Ugorji WS, Titus VO, Asade OO, Adeyemo DA, Okobi OE. Trends and patterns in hypertension-related deaths: a comprehensive analysis using Center for Disease Control and Prevention’s Wide-ranging Online Data for Epidemiologic Research (CDC WONDER) data. Cureus. 2024;16(10):e70754. 10.7759/cureus.70754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Bai YM, Gao RT, Guo SL, Chen XR, Jiang YY, Yang ZX, Zhao WH. Dietary characteristics of rural residents in Qinghai Province and its significance for chronic disease prevention and control. Chin J Prev Control Chronic Dis. 2013;21(3):267–70. 10.16386/j.cjpccd.issn.1004-6194.2013.03.017. [DOI] [Google Scholar]
- 33.Ma Y, Li Y, Zhang Z, Du G, Huang T, Zhao ZZ, Liu S, Dang Z. Establishment of a risk prediction model for metabolic syndrome in high altitude areas in Qinghai Province, China: a cross-sectional study. Diabetes Metabolic Syndrome Obes. 2024;17:2041–52. 10.2147/DMSO.S445650. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Luo Y, Wang S. Urban living and chronic diseases in the presence of economic growth: evidence from a long-term study in southeastern China. Front Public Health. 2022;10:1042413. 10.3389/fpubh.2022.1042413. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Cyr-Scully A, Howard AG, Sanzone E, Meyer KA, Du S, Zhang B, Wang H, Gordon-Larsen P. Characterizing the urban diet: development of an urbanized diet index. Nutr J. 2022;21(1):55. 10.1186/s12937-022-00807-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Guo Y, Hong Z, Cao C, Cao W, Chen R, Yan J, Hu Z, Bai Z. Urban-rural differences in the association of eHealth literacy with medication adherence among older people with frailty and prefrailty: cross-sectional study. JMIR Public Health Surveillance. 2024;10:e54467. 10.2196/54467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Huang X, He J, Wu G, Peng Z, Yang B, Ye L. TyG-BMI and hypertension in normoglycemic subjects in Japan: a cross-sectional study. Diabetes Vascular Disease Res. 2023;20(3):14791641231173617. 10.1177/14791641231173617. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Qiu J, Li J, Xu S, Yang J, Zeng H, Zhang Y, Yang S, Fang L, Huang J, Zhou H, Feng J, Zan Y, Zhan J, Liu J. Triglyceride glucose-weight-adjusted waist index as a cardiovascular mortality predictor: incremental value beyond the establishment of TyG-related indices. Cardiovasc Diabetol. 2025;24(1):306. 10.1186/s12933-025-02873-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Hameed EK. TyG index: a promising biomarker for glycemic control in type 2 diabetes mellitus. Diabetes Metabolic Syndrome. 2019;13(1):561–3. 10.1016/j.dsx.2018.11.030. [DOI] [PubMed] [Google Scholar]
- 40.Ke P, Wu X, Xu M, Feng J, Xu H, Gan Y, Wang C, Deng Z, Liu X, Fu W, Tian Q, He Y, Zhong L, Jiang H, Lu Z. Comparison of obesity indices and triglyceride glucose-related parameters to predict type 2 diabetes mellitus among normal-weight elderly in China. Eat Weight Disorders. 2022;27(3):1181–91. 10.1007/s40519-021-01238-w. [DOI] [PubMed] [Google Scholar]
- 41.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. 2021;16(3):375–81. 10.26574/maedica.2021.16.3.375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Fan C, Guo M, Chang S, Wang Z, An T. Elevated TyG-BMI index predicts incidence of chronic kidney disease. Clin Experimental Med. 2024;24(1):203. 10.1007/s10238-024-01472-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Wei C, Zhang G. Association between body roundness index (BRI) and gallstones: results of the 2017–2020 National Health and Nutrition Examination Survey (NHANES). BMC Gastroenterol. 2024;24(1):192. 10.1186/s12876-024-03280-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.DeFronzo RA. Insulin resistance, lipotoxicity, type 2 diabetes and atherosclerosis: the missing links. The Claude Bernard Lecture 2009. Diabetologia. 2010;53(7):1271–87. 10.1007/s00125-010-1684-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Guerrero-Romero F, Simental-Mendía LE, González-Ortiz M, Martínez-Abundis E, Ramos-Zavala MG, Hernández-González SO, Jacques-Camarena O, Rodríguez-Morán M. The product of triglycerides and glucose, a simple measure of insulin sensitivity. Comparison with the euglycemic-hyperinsulinemic clamp. J Clin Endocrinol Metab. 2010;95(7):3347–51. 10.1210/jc.2010-0288. [DOI] [PubMed] [Google Scholar]
- 46.Mancusi C, Izzo R, di Gioia G, Losi MA, Barbato E, Morisco C. Insulin resistance: the hinge between hypertension and type 2 diabetes. High Blood Press Cardiovasc Prev. 2020;27(6):515–26. 10.1007/s40292-020-00408-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Zamolodchikova TS, Tolpygo SM, Kotov AV. Insulin in the regulation of the renin-angiotensin system: a new perspective on the mechanism of insulin resistance and diabetic complications. Front Endocrinol. 2024;15:1293221. 10.3389/fendo.2024.1293221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Liu Z, Lu J, Sha W, Lei T. Comprehensive treatment of diabetic endothelial dysfunction based on pathophysiological mechanism. Front Med. 2025;12:1509884. 10.3389/fmed.2025.1509884. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Yang C, Song Y, Wang P. Relationship between triglyceride-glucose index and new-onset hypertension in general population: a systematic review and meta-analysis of cohort studies. Clin Exp Hypertens. 2024;46(1):2341631. 10.1080/10641963.2024.2341631. [DOI] [PubMed] [Google Scholar]
- 50.Xu J, Xu W, Chen G, Hu Q, Jiang J. Association of TyG index with prehypertension or hypertension: a retrospective study in Japanese normoglycemic subjects. Front Endocrinol. 2023;14:1288693. 10.3389/fendo.2023.1288693. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Samuel VT, Shulman GI. The pathogenesis of insulin resistance: integrating signaling pathways and substrate flux. J Clin Investig. 2016;126(1):12–22. 10.1172/JCI77812. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Zeng Q, Zhong Q, Zhao L, An Z, Li S. Combined effect of triglyceride-glucose index and atherogenic index of plasma on cardiovascular disease: a national cohort study. Sci Rep. 2024;14(1):31092. 10.1038/s41598-024-82305-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Tu Q, Hyun K, Lin S, Hafiz N, Manandi D, Li E, Wang X, Wu H, Redfern J. Impacts of hypertension and diabetes on the incidence of cardiovascular diseases and all-cause mortality: findings from the China Health and Retirement Longitudinal Study cohort. J Hypertens. 2025;43(4):623–30. 10.1097/HJH.0000000000003946. [DOI] [PubMed] [Google Scholar]
- 54.Xu L, Wen X, Yang Y, Cui D. Trends and comparisons of blood pressure and fasting plasma glucose in patients with hypertension, diabetes, and comorbidity: 4-year follow-up data. Risk Manage Healthc Policy. 2022;15:2221–32. 10.2147/RMHP.S385815. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Zou J, Lu S, Cao K, He R, Feng X, Yang S, Liu W, Liu H, Wang Z, Liao R, Chen J, Jiang X, Peng X. Association of the triglyceride-glucose index with all-cause and cause-specific mortality in patients with comorbid hypertension and diabetes: a population-based cohort study. Am Heart J Plus: Cardiol Res Pract. 2025;60:100657. 10.1016/j.ahjo.2025.100657. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56.Han Y, Hai J, Yang X, Lu D, Li J, Yan X, Bu P, Ti Y, Li X. The synergistic effect of triglyceride-glucose index and HbA1c on blood pressure control in patients with hypertension: a retrospective cohort study. Sci Rep. 2024;14(1):20038. 10.1038/s41598-024-70213-z. [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
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.