Association between the Triglyceride Glucose Index and Hyperuricemia in Patients with Primary Hypertension: A Cross-Sectional Study

Objective The aim of this study was to investigate the association between the triglyceride glucose (TyG) index and hyperuricemia (HUA) in patients with grades 1–3 hypertension. Study Design. This is a cross-sectional study. A total of 1,707 patients from the cardiovascular department of Affiliated Hospital of Jiangxi University of Traditional Chinese Medicine were studied. In this study, 899 patients with grades 1-2 hypertension were included, of which 151 had HUA; additionally, 808 patients with grade 3 hypertension were included, of which 162 patients had HUA. This study obtained all patient data from the electronic medical record system of the Affiliated Hospital of Jiangxi University of Traditional Chinese Medicine. The TyG index was calculated as Ln (triglycerides × fasting glucose/2). Hyperuricemia was defined as uric acid ≥420 μmol/L (7 mg/dL). Multivariate logistic regression, penalized spline regression, and generalized additive models were used to evaluate the association between the TyG index and HUA. Stratified analyses were performed to assess the association in populations with different grades of hypertension. Results The average TyG index was 8.71 ± 0.58. After adjusting for correlated variables, the logistic regression analysis revealed a positive correlation between the TyG index and HUA (OR = 1.83; 95% CI: 1.40–2.39). Smooth curve fitting showed that this correlation was linear in the whole range of the TyG index. In the subgroup analysis, the TyG index more strongly associated with HUA in the grades 1-2 hypertension group (OR = 2.22; 95% CI: 1.44–3.42) compared to that in the grade 3 hypertension group (OR = 1.58; 95% CI: 1.11–2.24; P for interaction = 0.03). In addition, this association was consistent in all models. Conclusion The TyG index was positively associated with HUA in patients with hypertension, and the association was more strongly confirmed in those with grades 1-2 hypertension rather than in those with grade 3 hypertension.


Introduction
Hyperuricemia (HUA) is a metabolic syndrome caused by a purine metabolism disorder. According to the Chinese guidelines [1], HUA can be diagnosed regardless of sex when serum uric acid (SUA) is ≥420 μmmol/L (7 mg/dL). Hyperuricemia can directly cause gout and uric acid nephropathy through sodium urate crystals' deposition. Meanwhile, HUA is an independent risk factor for chronic kidney disease, cardiovascular and cerebrovascular diseases, diabetes, and other cardiovascular events [2][3][4]. Regarding its pathogenic mechanism, a series of studies have shown that HUA is closely related to insulin resistance (IR) [5][6][7].
Insulin resistance is a clinical and biochemical disorder that can lead to impaired glucose tolerance and further cause diabetes. Previous studies have proved that IR is associated with obesity, cardiovascular diseases, hypertension, and other diseases of impaired insulin sensitivity and metabolic syndrome [8]. Te triglyceride glucose (TyG) index is a cheap, simple, and reliable substitute for IR compared to the homeostasis model (HOMA-IR) index [9]. Tus, the TyG index can be used as an early diagnostic indicator of IR and is helpful in identifying the population at a risk of cerebrocardiovascular disease [10][11][12][13].
Several previous studies have proved that the TyG index is associated with HUA [14][15][16]. However, only a few studies have analyzed the association between the TyG index and HUA in patients with hypertension [16,17]. Previous studies have focused on all populations with hypertension, with few studies investigating diferent hypertension grades. Terefore, our study concentrated on the association between the TyG index and HUA in the specifc population.

Participants.
Our study included a total of 5,153 inpatients with primary hypertension from January, 2020, to December, 2021, who attended the Afliated Hospital of Jiangxi University of Traditional Chinese Medicine. All participants were adults (aged 18 years or over) and hospitalized in the Department of Cardiology. Primary hypertension was diagnosed by referring to the 2018 ESC/ESH [18] guidelines for the management of arterial hypertension, which defnes hypertension as ofce systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg, when blood pressure value are measured in a resting and sitting position. Final blood pressure values were obtained by averaging three consecutive measurements, following a protocol that required at least 5 minutes of seated rest and two repeated measurements with 5-minute intervals. Inpatients who were taking antihypertensive medications or self-reported hypertensive diagnosis were also included. Te exclusion criteria were as follows: (1) those with incomplete SUA, triglyceride (TG), and glucose data; (2) those with secondary hypertension, diabetes, acute myocardial infarction, renal insufciency (estimated glomerular fltration rate (eGFR) <60 mL/min per 1.73 m 2 ), or malignant tumor; (3) those taking lipid-lowering drugs in the past 1 month; and (4) those with a history of using diuretics and other drugs that may afect the metabolism of uric acid within 2 months of inclusion. Finally, a total of 1,707 consecutive patients with grades 1-3 primary hypertension were included in our study, of which 899 had grades 1-2 hypertension and 808 patients had grade 3 hypertension. Figure 1 describes the initial sample population and exclusion criteria. Ethics approval was obtained from the Ethics Committee of the Afliated Hospital of Jiangxi University of Traditional Chinese Medicine (no.: JZFYLL20220727034). Te data were anonymous; thus, the requirement for informed consent was waived.

Data Collection and Defnition.
Clinical data, such as sex, age, medical history, as well as drug and blood test data, were collected from the electronic medical record system. To obtain blood samples, patients fasted overnight for 10-12 h, and then 4-5 mL venous blood was drawn the next morning. All samples were tested in the Department of Laboratory Medicine, Afliated Hospital of Jiangxi University of Traditional Chinese Medicine. Te following lab methods were used: fast blood glucose was assessed using the hexokinase method; SUA and blood urea nitrogen (BUN) were measured using the enzyme method; blood creatinine (Scr) was measured using the picric acid method; TG and total cholesterol (TC) were tested using the oxidase method; biochemical indicators, such as lipoprotein a (LPa), high density lipoprotein cholesterol (HDL-C), low density lipoprotein cholesterol (LDL-C), d-dimer, and high-sensitivity C-reactive protein (hs-CRP), were measured using immunoturbidimetry; the international normalized ratio (INR) was calculated using its formula; serum albumin (ALB) was analyzed using the bromide green method; alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were assessed using the enzymatic rate method; and all of the previous biomarkers were analyzed using an automatic biochemical analyzer (Siemens advia2400). Te TyG index � Ln (TG [mg/dL] × fasting glucose [mg/dL]/2) [19]. Te modifed modifcation of diet in renal disease (MDRD) equation was used to estimate eGFR [20]. Hyperuricemia was diagnosed as SUA ≥ 420 μmol/L in men and women using the 2019 Chinese guidelines for the diagnosis and treatment of HUA and gout [1].

Statistical Analysis.
Continuous variables were presented as the mean ± standard deviation or as the median (interquartile range), and categorical variables were presented as absolute values and/or frequency (%). Te baseline characteristics among the groups organized according to grades of hypertension (grades 1-2 or grade 3) were compared using chi-square tests (categorical variables), one-way analysis of variance (normal distribution), and the Kruskal-Wallis (skewed distribution) tests, respectively.
Multivariate logistic regression analyses were performed to assess the independent association between the TyG index and HUA after adjusting for correlated variables in the three models. Te variables were selected based on the following: clinical importance, statistical signifcance in the univariable analyses, and an estimated variable change of at least 10% of the potential confounding efects. Te restricted cubic spline model (a ftted smooth curve) was used to determine the dose-response relationship of the TyG index with SUA and HUA.

Baseline Characteristics.
Of the 1,707 patients, the mean age was 62.97 ± 12.87 years and 786 patients were men. In the grades' 1-2 hypertension group, 403 (44.7%) patients were men. In the grade 3 hypertension group, 384 (47.5%) patients were men. Te average TyG index was 8.71 ± 0.58, and there was no signifcant diference between the two groups. Of the 1,707 patients, 313 (18.3%) had HUA, including 151 (16.8%) with grades 1-2 hypertension and 162 (20%) with grade 3 hypertension. Missing data for ALB (0.1%), d-dimer (4.0%), and INR (4.0%) were imputed using the mean value. In the grades' 1-2 hypertension group, the levels of SUA, glucose, and hs-CRP were lower than those in the grade 3 hypertension group (all P < 0.05). Tere was no signifcant diference in the age and sex of the participants between the groups (P > 0.05, Table 1). In addition, there was no signifcant diference in the value of ALB, LDL-C, HDL-C, AST, Scr, BUN, TC, INR, eGFR, ALT, homocysteine (HCY), LPa, d-dimer, and TG between the two groups (P > 0.05, Table 1).

Te Positive Association of the TyG Index with HUA.
As shown in Table 2, the univariate analysis revealed that the TyG index, sex, age, ALB, HDL-C, AST, Scr, BUN, glucose, eGFR, ALT, LPa, and TG were associated with HUA in patients with primary hypertension (all P < 0.05). Furthermore, the multivariate analysis also revealed that the TyG index was positively associated with SUA and HUA after adjusting for potential confounding factors, as demonstrated in Tables 2 and 3. Figure 2 shows the dose-response relationship among the TyG index, SUA, and HUA levels in patients with primary hypertension. We found a linear relationship between the TyG index and HUA in patients with primary hypertension and an increasing trend in the incidence of HUA with the increasing TyG index. When the TyG index was assessed as tertiles, the incidence risks of HUA in the 2 nd and 3 rd tertiles were 1.45 (95% CI: 0.98-2.15) and 2.40 (95% CI: 1.60-3.60) times greater than that in the lowest tertile (P for trend <0.001), as shown in Table 4.

Discussion
We analyzed the correlation between the TyG index and HUA in patients with grades 1-3 hypertension. Te results showed that the TyG index was positively associated with SUA and HUA. Te restricted cubic spline indicated that the association was linear across the TyG range. Te risk of HUA gradually increased with an increase in the TyG index. Moreover, the subgroup analyses suggested that the positive association seemed to be strong among grades 1-2 hypertension and women (P for interaction � 0.03).
In recent years, there are some studies that have drawn conclusions on the association among the TyG index, SUA, and HUA. Kahaer et al. [21] found that the TyG index had a signifcant association with HUA and could be used as a risk screening indicator of HUA in the Xinjiang population in China. Meanwhile, da Silva et al. [11] conducted a linear relationship analysis between the TyG index and HUA in the general Chinese population by using the TyG index values to predict HUA. Te results are helpful in providing a simple and cost-efective method for the prevention and control of HUA. Furthermore, Zhou et al. [13]     hypertension) with HUA. tTyG, total triglyceride-glucose index; other abbreviations are listed in Table 3. Statistically signifcant P-values in bold for a better follow-up of the results. Te bold values mean that the diferences are signifcant. And in the 1st -3rd tertiles, the TyG index had a strong signifcant association with HUA in the grades 1 -2 hypertension group compared to that in the grade 3 hypertension group (P for interaction <0.05).  International Journal of Endocrinology hypertensive patients in a cross-sectional study to analyze the association between the TyG index and HUA. Te research divided the TyG index into four quartiles and found that the risk of HUA in the highest tertile was 2.79 times higher than that of the lowest tertile, again confrming a strong linear relationship between the TyG index and HUA in a hypertensive population. Our study revealed that when the TyG index was divided into three tertiles, and the risk of HUA was similar to the previous results. We further conducted subgroup analyses for grades 1-2 hypertension and grade 3 hypertension after performing a stratifed analysis of the TyG index. Te results showed that the HUA risk odds ratio (OR) and regression coefcient of grades 1-2 hypertension were higher than those of grade 3 hypertension (P value for interaction <0.05). Te diference was consistent in all models, indicating that the diference was steady for these risk factors; thus, the study concluded that the TyG index was positively associated with HUA, and the association was more strongly confrmed in those with grades 1-2 hypertension rather than in those with grade 3 hypertension. However, the mechanisms underlying these diferences are not yet fully understood.
Previous studies have confrmed that the TyG index identifed HUA diferently in women and men [17]. Similarly, our study found that the OR and regression coefcient values of the TyG index were higher in women, suggesting that the TyG index had a strong association with HUA in female populations with hypertension. Tis may be explained by estrogen being related to complex endocrine factors as well as being a uric acid producing agent. Estrogen also accounts for some diferences in lipid metabolism between women and men.
Although our study and previous research have confrmed the association between the TyG index and HUA, the specifc mechanisms remain unclear. Te most commonly recognized mechanism is related to IR. Previous epidemiological studies have revealed a signifcant relationship between IR and SUA [22,23]. After IR, compensatory hyperinsulinemia occurs, leading to decreased uric acid excretion through renal tubular sodium reabsorption, which further causes HUA. In addition, one study proved that early β-cell dysfunction was mainly associated with elevated uric acid levels [24]. Insulin resistance also causes HUA through several other pathways, including by inducing systemic infammation, causing kidney damage, decreasing renal uric acid excretion, and afecting lipid metabolism [25]. Conversely, HUA can lead to IR and infammation by afecting adipocytes and reducing mitochondrial oxidative stress and nitric oxide bioavailability [26]. HUA and IR are both important risk factors for hypertension. Te specifc mechanisms are as follows: (1) human studies have shown that HUA may be related to impaired endothelial function [27,28], (2) animal studies shown that HUA can activate the renin-angiotensin system, which leads to increased systemic BP and vascular resistance [29], and (3) activating the immune system and its related hemodynamic efects [30]. Tese changes can aggravate systemic hypertension and lead to end-organ damage. Wang et al. [31] conducted a prospective cohort study included 21,999 subjects without hypertension or gout at the baseline. Te results showed that the elevated SUA is associated with an increased risk of hypertension, and IR may play an intermediary role in the relationship between SUA and hypertension.
Studies have shown that the TyG index is a more sensitive and specifc indicator for IR detection compared to other IR indicators [32]. In addition, some researchers found an association between the TyG index and HUA in populations from diferent regions.
Our study aimed to explore the association of the TyG index and HUA in hospitalized patients with grades 1-3 hypertension. One main strength of our research was that we conducted stratifed analyses in populations with grades 1-2 hypertension and grade 3 hypertension. Tese results are helpful to show the association between the TyG index and HUA in these specifed populations.
Our study had some limitations. First, this was a crosssectional study, and therefore, the association between the TyG index and HUA in patients with grades 1-3 hypertension is insufcient to draw causal conclusions. In the future, prospective randomized studies could help obtain more evidence to support the associations found in our study. Second, we included only a subset of hospitalized patients in Jiangxi, China, which was not representative of the entire population. In addition, most in-patients had grade 2 and grade 3 hypertension and fewer in-patients had grade 1 hypertension. Terefore, we combined the grade 1 and grade 2 hypertension cases for further analysis. Te target population of our study was hospitalized hypertensive patients, most of whom had grades 2-3 hypertension; thus, our study is consistent with clinical reality. Tird, data derived from the electronic medical record system lacked information on waist circumference (WC), height, weight, alcohol consumption, and smoking history. Tese unrecorded risk factors for HUA could be residual confounding variables, potentially causing further bias in our results. However, most studies found that the positive association between the TyG index and HUA was still signifcant after adjusting for these risk factors [16,17,21], and therefore, we consider our conclusions to be reliable.

Conclusion
In conclusion, our study confrmed the positive association between the TyG index and the risk of HUA in hospitalized patients with hypertension. Te association was more strongly confrmed in women than in men and in grades 1-2 hypertension than in grade 3 hypertension.

Data Availability
Te datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.

Ethical Approval
Ethical approval was obtained from the Ethics Committee of the Afliated Hospital of Jiangxi University of Traditional International Journal of Endocrinology Chinese Medicine (no.: JZFYLL20220727034). Te data were anonymized, and the requirement for informed consent was waived.

Consent
All subjects signed the informed consent form.

Conflicts of Interest
Te authors declare that they have no conficts of interest.