Association between the triglyceride-glucose index and albuminuria in hypertensive individuals

ABSTRACT

Background

Previous studies have demonstrated that the triglyceride-glucose (TyG) index is significantly associated with vascular damage. Albuminuria is a marker of hypertension-mediated organ damage (HMOD) and has been linked to a greater risk of cardiovascular disease (CVD). However, the association between the TyG index and albuminuria in patients with hypertension is not clear. This population research focused on subjects with hypertension to investigate the association between an elevated TyG index and albuminuria.

Methods

From September 2019 to November 2019, 789 hypertensive participants were involved in our research. Logistic regression models were performed to estimate the odds ratios (ORs) and 95% confidence intervals (CIs) for albuminuria according to the quartiles of the TyG index.

Results

Multivariate logistic regression analysis revealed that the TyG index was significantly associated with albuminuria. Using the lowest TyG index quartile as the reference, the fully adjusted ORs (95% CIs) for albuminuria for TyG index quartile II, quartile III, and quartile IV were 1.90 (1.17–3.12), 1.81 (1.07–3.07), and 3.46 (2.06–5.91), respectively. The results in the subgroup analysis were similar to the main analyses except for the smokers. Restricted cubic spline curves based on logistic regression models evaluated the linear association between the TyG index and albuminuria (P for nonlinear = 0.831).

Conclusion

The TyG index was positively associated with albuminuria among hypertensive participants.

KEYWORDS:

• The triglyceride-glucose index
• albuminuria
• hypertension
• elderly
• insulin resistance

Introduction

Cardiovascular disease (CVD) is the leading cause of human death and loss of healthy life, ranking first in the global disease burden (1). The prevalence of CVD is on the rise in China, with an estimated 330 million patients suffering from CVD (2). Hypertension is probably the most important modifiable risk factor for cardiovascular disease. Hypertension-mediated organ damage (HMOD) is an intermediate endpoint of cardiovascular events in hypertensive patients (3). With early recognition and timely treatment, HMOD can be reversed, thereby reducing the incidence of cardiovascular events (4).

Albuminuria is a marker of HMOD and has been linked to a greater risk of further CVD, renal function loss, atherosclerosis, and all-cause mortality (5,6). Albuminuria is mainly caused by glomerular damage and is therefore considered an early marker of systemic vascular damage or microvascular disease in addition to renal disease.

Insulin resistance (IR) has long been recognized as a risk factor for both micro-and macroangiopathies (7). The triglyceride-glucose (TyG) index has been suggested as a reliable and simple substitute for IR, and it has been reported to have a close relationship with the hyperinsulinemic-euglycemic clamp (8–10). Previous research has proven that the TyG index is significantly associated with vascular damage (11–15). Furthermore, Zhao et al. reported that an elevated TyG index is associated with renal microvascular damage (16). However, the association between the TyG index and albuminuria in hypertensive individuals is not clear.

Considering that this high-risk population is more likely to develop CVD, it is critical to understand the precise association between the TyG index and albuminuria in hypertensive individuals. Therefore, our population research focused on individuals with hypertension to investigate the association between an elevated TyG index and albuminuria.

Materials and methods

Study subjects

This cross-sectional study was conducted in southeast China from September 2019 to November 2019 in Zhangjiagang city. The study participants were recruited if they (1) were 45 years or older, (2) resided in the Nanfeng or Daxin towns in Zhangjiagang city, (3) were diagnosed with hypertension by a trained physician based on systolic blood pressure (SBP) ≥140 mmHg and/or diastolic blood pressure (DBP) ≥90 mmHg without antihypertensive medications or currently taking antihypertensive medication, and (4) were willing to participate in this research and able to sign the informed consent. The exclusion criteria were (1) self-reported diabetes, (2) myocardial infarction, unstable angina pectoris or cerebrovascular disease within the last 6 months and (3) uncontrolled thyroid dysfunction, coagulopathy, or malignancy. Finally, 789 participants were enrolled in this study. This study was approved by the First Affiliated Hospital of Nanjing Medical University Ethics Committee in China.

Data collection

Baseline demographic characteristics, lifestyle-related risk factors, history of disease, and family history of CVD were collected by trained investigations using a standardized questionnaire. A professional investigator assessed the subjects’ height, body weight, waist circumference, and hip circumference. After the participants rested for at least 5 minutes, trained staff took three sitting blood pressure measures for each participant using a sphygmomanometer. Blood pressure was recorded three times at intervals of 1 min, and the difference between any two readings of systolic or diastolic blood pressure should not exceed 10 mmHg. Among the participants, SBP<140 mmHg and DBP<90 mmHg were defined as well-controlled BP and, conversely, uncontrolled BP. Body mass index (BMI) was calculated as body weight in kilograms divided by height in square meters (kg/m²). We collected blood and urine samples in the morning after overnight (at least 8 hours) fasting. An automated biochemical analyzer at a local laboratory was used to examine serum glucose, lipid levels, and other biochemical parameters. The TyG index was calculated as ln[fasting triglycerides (mg/dL)×fasting glucose (mg/dL)/2]. Albuminuria was defined as a urinary albumin-to-creatinine ratio (ACR) ≥ 30 mg/g creatinine.

Statistical analysis

Participants were categorized into four groups according to the quartiles of the TyG index. Quantitative data are presented as the mean (standard deviation) or median (25th-75th percentile), and qualitative data are presented as frequencies (%). Tests for linear trends across TyG index quartiles were conducted with the use of generalized linear regression analysis for quantitative variables and the Cochran-Armitage trend χ2 for qualitative variables.

Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated for albuminuria associated with the TyG index. Logistic regression models were first adjusted for age and sex (Model a). Model b further adjusted for lifestyle-related risk factors: current cigarette smoking and current alcohol consumption. Logistic models were then further adjusted for BMI, SBP, DBP, HDL-C, LDL-C, uric acid, and the duration of diagnosed hypertension in Model c. Restricted cubic spline curves based on logistic regression models with four knots at the 5th, 35th, 65th, and 95th centiles were performed to explore the dose–response association with albuminuria. Additionally, we performed subgroup and interaction analyses to investigate whether the relationships between the TyG index and albuminuria varied according to the status of the potential covariates (age, sex, current cigarette smoking, current alcohol consumption, BMI, and blood pressure control).Table 1

Table 1. Characteristics of 789 participants according to the category of TyG index.

	TyG index	Quartile I	Quartile II	Quartile III	Quartile IV	P for trend
	Total	<8.38	8.38–8.75	8.75–9.12	≥9.12
N	789	198	197	197	197
Age, years	62.00(55.00–67.00)	63.50(56.75–69.00)	62.00(56.00–68.00)	63.00(56.00–67.00)	60.00(54.00–66.00)	<0.001
Male	411(52.09)	119(60.10)	100(50.76)	87(44.16)	105(53.30)	0.135
Current Cigarette Smoking	196(24.84)	50(25.25)	47(23.86)	49(24.87)	50(25.38)	0.929
Current Alcohol Consumption	283(35.87)	78(39.39)	62(31.47)	60(30.46)	83(42.13)	0.682
WC, cm	88.50(82.35–94.50)	83.55(78.50–91.92)	88.20(81.85–94.00)	91.00(85.20–96.10)	90.10(85.05–96.50)	<0.001
BMI, kg/m2	25.32(23.30–27.38)	23.77(21.86–26.44)	25.26(23.19–27.02)	25.67(24.07–28.08)	25.69(24.23–27.84)	<0.001
SBP, mmHg	139.67(129.33–150.33)	139.50(126.58–150.67)	138.33(127.00–149.33)	139.67(128.83–151.00)	140.33(132.42–150.25)	0.383
DBP, mmHg	86.00(80.42–92.67)	85.50(78.58–91.33)	85.00(80.67–91.67)	86.67(79.33–94.00)	87.17(82.67–93.92)	0.002
HR, bpm	71.00(65.00–80.00)	69.00(63.75–79.00)	72.00(65.00–80.00)	70.00(65.00–80.00)	73.00(67.00–80.00)	<0.001
Duration of diagnosed hypertension, years	6.00(4.00–10.00)	7.00(4.00–10.00)	7.00(4.00–10.00)	6.00(4.00–9.50)	7.00(5.00–10.00)	0.667
Fasting glucose, mmol/L	5.40(4.99–5.93)	5.13(4.78–5.52)	5.36(5.06–5.86)	5.42(4.98–5.94)	5.77(5.17–6.60)	<0.001
Total cholesterol, mmol/L	4.79(4.19–5.40)	4.55(3.92–5.14	4.74(4.08–5.20)	4.92(4.39–5.44)	5.12(4.35–5.79)	<0.001
Triglyceride, mmol/L	1.43(1.02–2.14)	0.82(0.68–0.95)	1.21(1.08–1.33)	1.71(1.52–1.95)	2.83(2.31–3.79)	<0.001
HDL-C, mmol/L	1.40(1.21–1.58)	1.51(1.33–1.72)	1.43(1.26–1.62)	1.37(1.20–1.53)	1.26(1.11–1.44)	<0.001
LDL-C, mmol/L	2.79(2.38–3.31)	2.59(2.18–3.01)	2.79(2.38–3.27)	3.00(2.58–3.49)	2.86(2.40–3.43)	<0.001
Creatinine, μmol/L	61.00(54.00–72.00)	64.00(56.00–72.00)	62.00(56.00–75.00)	60.00(53.00–71.00)	60.00(53.00–69.00)	0.112
Uric Acid, μmol/L	316.00(260.00–381.00)	294.00(242.75–357.75)	307.00(244.00–365.50)	324.00(270.00–380.50)	348.00(286.00–412.50)	<0.001
ACR (mg/g creatinine)	17.42(9.07–37.84)	13.59(7.89–26.97)	17.28(8.54–36.32)	17.42(9.59–38.70)	25.36(11.38–58.30)	0.018

BMI, body mass index; WC, waist circumference; SBP, blood pressure; DBP, blood pressure; HR, heart rate; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; ACR, urine albumin-to-creatinine ratio.

All statistical analyses were performed using R statistical software (version 4.1.3). P values were two-tailed, and less than 0.05 was considered statistically significant (Table 2).

Table 2. Odds ratios (95% CIs) of albuminuria according to quartiles of TyG index.

	Quartile I	Quartile II	Quartile III	Quartile IV	Per 1 unit increase	P for trend
Model a	1.00	1.65(1.04–2.62)	1.59(1.00–2.53)	3.14(2.02–4.97)	2.07(1.61–2.69)	<0.001
Model b	1.00	1.66(1.04–2.62)	1.59(1.00–2.53)	3.11(1.99–4.92)	2.05(1.59–2.67)	<0.001
Model c	1.00	1.90(1.17–3.12)	1.81(1.07–3.07)	3.46(2.06–5.91)	2.10(1.57–2.83)	<0.001

Model a: adjusted for age, gender.

Model b: adjusted for age, gender, current cigarette smoking, and current alcohol consumption.

Model c: adjusted for age, gender, current cigarette smoking, current alcohol consumption, BMI, SBP, DBP, HDL-C, LDL-C, uric acid, and the duration of diagnosed hypertension.

Results

Baseline characteristics

The 789 participants were divided into four groups according to the quartiles of the TyG index: quartile I (<8.38), quartile II (8.38–8.75), quartile III (8.75–9.12), and quartile IV (≥9.12). The baseline characteristics of the four groups are presented in Table 1. The median (Q₂₅-Q₇₅) age of the subjects was 62.00 (55.00–67.00) years; 411 were men (52.09%). Compared with participants with a lower TyG index, participants with a higher TyG index had a higher probability of being younger; having a significantly higher BMI, waist circumference, DBP, and HR; having higher TC, LDL-C, uric acid, and ACR concentrations; and having lower HDL-C concentrations (P < .05).

In Table S1, participants were divided into two groups based on the presence of albuminuria. Participants with albuminuria tended to be women; to be diagnosed with hypertension longer; to have higher SBP, DBP, and HR; to have higher creatinine, TG, and fasting glucose concentrations; and to have a higher TyG index than those without albuminuria.

Association of the TyG index and albuminuria

Both univariate and multivariate logistic regression analyses revealed that the TyG index was significantly associated with albuminuria. Using the lowest quartile of the TyG index as the reference, the fully adjusted ORs (95% CIs) for albuminuria for TyG index quartile II, quartile III, and quartile IV were 1.90 (1.17–3.12, p for trend < 0.001), 1.81 (1.07–3.07, p for trend < 0.001), and 3.46 (2.06–5.91, p for trend < 0.001), respectively.

For each unit increase in the TyG index, the OR for albuminuria was 2.10(1.57–2.83) in fully adjusted Model c. As shown in Figure 1, we used restricted cubic spline curves based on a logistic regression model to visualize the relationships between the TyG index and albuminuria. The TyG index showed a linear relationship with albuminuria in the multiple-adjusted models, and the p for nonlinearity was 0.831.

Figure 1. Association of the TyG index and albuminuria in hypertensive patients. Restricted cubic spline regression was used to calculate the odds ratios expressed as solid lines and 95% confidence intervals expressed as shaded areas. The reference point is the median of the TyG index (8.75 mg/dL), with knots placed at the 5th, 35th, 65th, and 95th centiles of the TyG index distribution. P for nonlinear is 0.831.

Subgroup analyses for the association between albuminuria and the TyG index

The association between albuminuria and the TyG index in different subgroups was investigated using stratified analysis. As shown in the forest plot (Figure 2), the subgroup analyses were consistent with the major findings except in the subgroups of smokers. There was no significant association of the TyG index with albuminuria in smokers. Meanwhile, the following subgroups did not have significant interactions: age, sex, current cigarette smoking, current alcohol consumption, BMI, and blood pressure control subgroups (all P values > .05).

Figure 2. Subgroup analyses of the association between the TyG index and albuminuria. ORs were adjusted for age, sex, current cigarette smoking, current alcohol consumption, BMI, SBP, DBP, HDL-C, LDL-C, uric acid, and the duration of diagnosed hypertension.

Discussion

In our population-based study, we observed an association of the TyG index with albuminuria in participants with hypertension. We discovered that an elevated TyG index was independently positively associated with albuminuria after adjusting for confounding variables. The results in the subgroup analysis were similar to the main analyses except for the smokers, highlighting the soundness of these relationships in hypertensive individuals. Restricted cubic spline curves based on logistic regression models were used to evaluate the linear association between the TyG index and albuminuria.

Hypertension-mediated organ damage (HMOD) is of great significance for cardiovascular risk stratification and drug therapy selection for hypertensive patients. The MONICA study included a total of 1968 participants with HMOD (including left ventricular hypertrophy (LVH), carotid plaque, arterial sclerosis, increased ACR) and showed that after adjustment for age and sex, the greater number of combined target organ damage, the higher the risk of cardiovascular events (17). Therefore, attention should be given to target organ damage in hypertensive patients. Kidney damage may occur in the early stage of hypertension, and urine microalbumin is often used clinically to detect the progress of renal function. In addition, a meta-analysis of 32 studies (including 80,812 hypertensive patients with or without diabetes) found that every 10% reduction in urinary protein was associated with a 13%, 29%, and 14% reduction in the risk of myocardial infarction, stroke, and cardiovascular death, respectively, suggesting that urinary protein may represent an intermediate endpoint of cardiovascular events in hypertensive patients (18).

For the assessment of renal damage in the early stage of hypertension, ACR has few influencing factors and is relatively stable in individuals. Therefore, in our study, we used ACR as a parameter for the evaluation of renal damage. Epidemiological data showed that hypertensive patients had a high prevalence of albuminuria (19). Compared with a previous study, the prevalence of albuminuria in our analysis was approximately 37.83% in women and 26.76% in men.

Several prospective studies have shown that the TyG index is associated with new-onset diabetes, hypertension and CVD (11,20–22), which indicates that the TyG index can be used to assess vascular injury. A Pakistani study of 277 participants showed a positive correlation between the TyG index and urine ACR (23), although the correlation was weak, and no further analysis was performed. Meanwhile, Zhu et al. observed that the TyG index is significantly associated with hypertension and shows the superior discriminative ability for hypertension compared with lipid and glycemic parameters in the Chinese elderly population (24).

To the best of our knowledge, only two previous studies showed an association between the TyG index and albuminuria. The Northern Shanghai Study investigated the relationship between macrovascular and microvascular injury and the TyG index in 2830 elderly Chinese community-dwelling individuals, and they found that an elevated TyG index was significantly associated with a higher risk of microalbuminuria (16). Similarly, Chiu et al. observed that the TyG index could predict the occurrence of microalbuminuria among Japanese patients with type 2 diabetes mellitus (20). The study populations of these two studies were elderly population and diabetic population. Our research enriches the limited current evidence on the relationship between the TyG index and albuminuria, extending this association to individuals with hypertension. At the same time, slightly different from the results of previous studies, our study showed that quartile II and quartile III were still significantly associated with albuminuria in the multiple-adjusted Model c. We also discovered a positive dose–response relationship between the TyG index and albuminuria in hypertensive individuals, which is consistent with the conclusion of previous reports. However, none of the previous studies conducted subgroup analysis. The purpose of subgroup analysis is to study the interactions or effect modifications, that is, whether the magnitude of the association is different in different populations or conditions. In our study, the participants were stratified according to age, sex, current cigarette smoking, current alcohol consumption, BMI, and blood pressure control. Subgroup analyses were consistent with the major findings except in the subgroups of smokers. Notably, the TyG index was associated with albuminuria in both the well-controlled BP and uncontrolled BP groups, which suggests that the TyG index can be used to reflect albuminuria regardless of blood pressure control.

Although the underlying mechanisms of the relationship between the TyG index and albuminuria in hypertensive individuals have not been elucidated, they may be related to insulin resistance. In the present, the hyperinsulinemic-euglycemic clamp test is used to measure IR. While the tests are time-consuming, invasive, and complex, they are difficult to apply to large populations in research and clinical environment. Guerrero et al. demonstrated in 2010 that the TyG index, which is the product of TG and glucose in plasma, could be a reasonable estimate of IR (8). In addition to serving as a simple and reliable surrogate marker of IR, the TyG index is also associated with metabolic abnormalities. This index was compared to the hyperinsulinemic-euglycemic clamp, and it was shown to have high sensitivity and specificity for IR detection (8). Insulin resistance can induce an imbalance in glucose metabolism, resulting in chronic hyperglycemia, which can trigger oxidative stress (25) and cause inflammation, which can lead to damage to vascular endothelial cells, increase microvascular permeability, and moderately increase glomerular urinary albumin. According to our findings, the TyG index may assess renal microvascular injury.

To the best of our knowledge, our study is the first population study to investigate the relationship between the TyG index and albuminuria among hypertensive individuals. We collected as many confounding factors as possible for the baseline investigation. Multivariable-adjusted models in the present analyses made our results more accurate. The strong positive association between the TyG index and albuminuria supports the validity of our study. For IR prediction, the TyG index is easier to calculate by TG and fasting glucose without additional cost. Therefore, the TyG index may be expected to find wide application in clinical practice to identify hypertensive patients at a higher risk of albuminuria early. At the same time, several limitations exist in our study. First, in our cross-sectional study, we were unable to establish a causal relationship. Therefore, additional prospective research could provide further insight into the potential causal associations of the TyG index with the development of albuminuria in hypertensive individuals. Second, the effects of the TyG index and HOMA-IR on albuminuria were not compared because insulin and HbA1c levels were not detected in our investigation. However, the TyG index is certainly simpler and less expensive to measure in clinical settings. Third, urine specimens were collected only once for all study subjects, so the influence of experimental error or specimen contamination cannot be completely excluded. Last, our investigation population comprised only Chinese adults with hypertension. Hence, our results should be interpreted with caution when extended to other racial or ethnic populations.

In conclusion, the TyG index were discovered to be positively associated with albuminuria among hypertensive participants. Our results suggests that educational interventions about lifestyle changes should be provided to patients with high TyG indexes.

Author contributions

Yunfan Tian performed the data analyses and wrote the manuscript. Jingyu Sun contributed significantly to analysis and manuscript preparation. Ming Qiu and Yan Lu helped perform the analysis with constructive discussions. Xuesong Qian and Wei Sun contributed to refining the ideas and finalizing this paper. Xiangqing Kong contributed to the conception of the study.

Ethical approval

Our study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Nanjing Medical University (2019-SR-097). Informed consent was obtained from all individual participants included in our study.

Acknowledgments

None.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/10641963.2022.2150204.

Additional information

Funding

This work was supported by grants from the National Natural Science Foundation of China (No. 82170425), China International Medical Foundation (Z-2019-42-2908).