Correlation Analysis between Uric Acid and Metabolic Syndrome in the Chinese Elderly Population: A Cross-Sectional Study

Background Currently, both metabolic syndrome and hyperuricaemia have attracted extensive attention in public health. The correlation between uric acid and metabolic syndrome is controversial. Research on the relationship between uric acid and metabolic syndrome in community-dwelling elderly people is relatively lacking. The purpose of this study is to explore the relationship between uric acid and metabolic syndrome in the community-dwelling elderly people. Design Cross-sectional study. Methods We collected the physical examination data of 1,267 elderly people in Gutian community in Wuhan and used SPSS IBM 25.0 for data analysis. Correlation and logistic regression analyses were performed, and ROC curves were drawn. Results The uric acid level of the nonmetabolic syndrome group was lower than that of the metabolic syndrome group (337.31 vs. 381.91 µmol/L; P < 0.05). Uric acid was positively correlated with systolic blood pressure (r = 0.177, P < 0.001), diastolic blood pressure (r = 0.135, P < 0.001), body mass index (r = 0.234, P < 0.001), waist circumference (r = 0.283, P < 0.001), and triglycerides (r = 0.217, P < 0.05). High-density lipoprotein cholesterol (r = −0.268, P < 0.001) showed the opposite trend. Logistic regression analysis results suggested that uric acid is a risk factor for metabolic syndrome. The result is described as exp (B) and 95% CI (1.003 [1.001, 1.005]). Based on the receiver operating characteristic curve, we found that the area under the curve of uric acid to diagnose metabolic syndrome was 0.64 (sensitivity: 79.3%, specificity: 45.1%). Conclusion We observed an association between uric acid levels and metabolic syndrome in the elderly Chinese population. The best threshold value for uric acid in predicting metabolic syndrome diagnosis was 314.5 μmol/l.


Introduction
Metabolic syndrome (MetS) is a multifactorial pathological condition defned by the association of several metabolic disorders [1]. It is an independent risk factor for cardiovascular diseases (CVDs) and diabetes, which are often accompanied by increasing uric acid levels [2]. Experimental data have demonstrated that high uric acid is associated with endothelial dysfunction, oxidative stress, increased platelet adhesiveness and infammation [3]. Numerous studies have shown a link between hyperuricaemia and other CVD risk factors, such as hypertension, obesity, dyslipidaemia, and diabetes [4][5][6]. Te management of MetS and its defnition in the elderly are particularly important. Whether uric acid has an efect on MetS is controversial [7].
Uric acid is an oxidation product of purine metabolism in the circulatory system. Studies have shown that high uric acid levels regulate the oxidative stress, infammation, and enzymes associated with glucose and lipid metabolism, suggesting a mechanism for the impairment of metabolic homeostasis [8]. Hyperuricaemia can cause abnormal fat accumulation in body tissues, and adipokines can induce the production of reactive oxygen species and cause the formation of free radicals, which are pro-oxidant factors [9][10][11][12]. In contrast, uric acid is also considered to be an antioxidant that can fght superoxide anions, hydroxyl radicals, and peroxynitrite [7,13]. Its antioxidant protection can be refected in cell apoptosis, cancer, and ageing [14]. Oxidative stress is inseparable from the formation of MetS [15]. Te two-way regulation of oxidative stress by uric acid might be closely related to the level of uric acid in people with MetS. Terefore, we designed a cross-sectional clinical study to explore the correlation between the two. Our research focused on the level of uric acid and MetS in the elderly and explored the relationship between them. Te optimal cut-of value of uric acid at risk of MetS was initially obtained.  (2) triglyceride (TG) ≥ 1.7 mmol/l; (3) HDL-C < 0.9 mmol/l (male) < 1.0 mmol/l (female); (4) systolic blood pressure and diastolic blood pressure (SBP/DBP) ≥ 140/90 mmHg and/or with diagnosed and treated hypertension; and (5) fasting blood glucose (FBG) ≥ 6.1 mmol/l, 2 h postprandial blood glucose (PBG) ≥ 7.8 mmol/l, or diagnosed with diabetes. MetS is diagnosed by three or more of the above items.

Defnition of Hyperuricaemia.
Te diagnostic criteria for hyperuricaemia were based on the "China Guidelines for Diagnosis and Treatment of Hyperuricaemia and Ventilation 2019". Uric acid levels above 420 µmol/L are defned as hyperuricemia.

General Data and Biochemical
Indicators. Data for this study were obtained from the physical examination records of elderly people in the community. For example, age and sex were recorded orally in the physical examination information by the participant. Height was measured in a standing position without shoes using a stadiometer with a sensitivity of 0.1 cm. Weight was measured wearing light clothing using a scale sensitive to the nearest 0.1 kg. BMI was calculated by dividing the person's weight (kg) by their height squared (m 2 ). Waist circumference (WC) was measured using plastic tape to the nearest 0.1 cm. Te doctor used a mercury sphygmomanometer on the upper arm to measure blood pressure, systolic blood pressure and diastolic blood pressure. Te average value of two blood pressure measurements taken at least fve minutes apart was used. Biochemical indices, including FBG, blood lipids, and uric acid, were all measured using venous blood obtained from the participants on an empty stomach. Levels of serum UA, TGs, HDL-C, low-density lipoprotein cholesterol (LDL-C), and FBG were measured using Roche E602 and Roche C701 (both of which are automatic biochemical analysers).

Statistical
Analysis. IBM SPSS version 25.0 statistical software was used for data analysis in this study. Te normality of distributions was evaluated using the Kolmogorov-Smirnov test, and continuous data are presented as the mean ± SD. Categorical variables are presented as the frequency (percentage). Te statistical signifcance of the differences in clinical and biochemical values between participants with and without MetS was analysed using Student's t-test for continuous variables and the chi-squared test for categorical variables. Kendall correlation analysis was used to analyse the correlation between two categorical variables, Pearson correlation analysis was used to examine the relationship between two continuous variables, and Spearman correlation analysis was used to investigate the relationship between categorical variables and continuous variables. After drawing the receiver operating characteristic (ROC) curve, the cut-of value of the uric acid level was obtained by calculating the Youden index (sensitivity + specifcity − 1). Te uric acid level corresponding to the maximum value of the Youden index is the optimal cut-of value. Considering the efect of gender in metabolic syndrome, we also contended to plot ROC curves separately for the male and female elderly population.

Characteristics of the Research Population.
Te average age of the population was 71.64 ± 5.61 years. Te basic laboratory and clinical characteristics of 1,267 participants (556 men and 711 women) with and without MetS enrolled in this study are shown in Table 1. Te prevalence of hyperuricaemia and MetS was 28.1% (356 of 1,267) and 18.6% (128 of 1,267), respectively. Te prevalence of MetS in the hyperuricaemia population was 28.4% vs. 14.6% in the nonhyperuricaemia population. In the group diagnosed with MetS, age, WC, blood pressure, TG, TC, LDL-C, UA, and FBG were signifcantly higher than those in the non-MetS group, and there were signifcant diferences (P < 0.05). Te level of HDL-C in the MetS group was signifcantly lower than that in the non-MetS group (P < 0.05). In addition, the average serum uric acid level was higher than that in subjects without MetS, and this diference was signifcant (381.91 vs. 337.31 µmol/L; P < 0.05), as shown in Table 1.

Correlation Analysis.
We performed simple correlation analysis to explain the relationship between uric acid levels and various components of MetS. We also performed a simple correlation analysis with MetS as the dependent variable (shown in Table 2). Te results were expressed by the correlation coefcient r and P value. We found that WC (r � 0.283, P < 0.001), BMI (r � 0.234, P < 0.001), and TG (r � 0.217, P < 0.05) were positively correlated with uric acid levels, and high-density lipoprotein (r � −0.268, P < 0.001) was signifcantly negatively correlated with uric acid levels. In addition, we used metabolic syndrome as the dependent variable to perform a simple correlation analysis. Te correlation analysis results are shown in the right column of Table 2.

Logistic Regression between Uric Acid Levels and MetS.
In the logistic regressions, using MetS as the dependent variable, we identifed that uric acid was a risk factor for MetS. We included the relevant variables in the regression model to construct three regression models, as shown in Table 3. Te basis for the inclusion of the regression model was univariate correlation variables obtained from the analysis as well as clinically known and recognized variables associated with MetS. Model 1 is a rough model without any adjustment for confounding factors, Model 3 is a fully corrected model, and Model 2 is a partially corrected model. Te results are expressed as exp (B) (OR value) and 95% CI. In Model 3, the confounding of factors, such as age, sex, WC, blood pressure, blood lipids, and FBG, were adjusted, and the results showed that uric acid level might be an independent risk factor for MetS (exp (B) � 1.003, 95% CI (1.001, 1.005), P � 0.014). Obesity (obesity defned as BMI ≥ 25 kg/m 2 ) is an important part of the metabolic syndrome, and both the simple correlation studies and regression results described above suggest that BMI and uric acid levels are closely related (Table 2). Terefore, we performed a stratifed analysis according to whether or not they were obese. We could see an association between uric acid and metabolic syndrome in both obese and nonobese populations (ORs 1.002, 1.006, P values <0.05, respectively), independent of obesity (Table 4).

ROC Curves between Uric Acid Level and MetS.
To predict a threshold value for the diagnosis of MetS using uric acid, ROC curves were drawn. Figure 1 shows the ROC curves for the diagnosis of MetS by diferent variables. Te AUC of UA, TG, TC, GLU, HDL, LDL, BMI, and age were 0.641, 0.786, 0.542, 0.771, 0.276, 0.544, 0.773, and 0.487, respectively. By calculating the Youden index, we found that the cut-of value for uric acid corresponding to the maximum Youden index (0.244) was 314.5 μmol/l. Te Youden index corresponded to a sensitivity and specifcity of 79.3% and 45.1%, respectively. In this community population, the prevalence of metabolic syndrome was 19.24% and 17.86% in elderly males and elderly females, respectively. Considering the infuence of gender in this, we plotted ROC curves in the elderly male and female populations, respectively, and we found diferent results (Figures 2 and 3). Te diagnostic efcacy of uric acid for metabolic syndrome seems to be more signifcant in the elderly male population, with an area under the curve of 0.735 and a sensitivity and specifcity of 75.2% and 60%, respectively. And the cut-of value for uric acid is 314.5 μmol/l too.

Discussion
Global ageing is becoming increasingly serious, and achieving healthy "ageing" can reduce the burden on national fnances and improve the quality of life of the elderly. According to the  International Journal of Endocrinology data of the World Health Organization, the prevalence of MetS in the elderly ranges from 11% to 43% (median 21%) and NCEP ranges 23% to 55% (median 31%) [16][17][18]. We conducted a study in a community-based elderly population and found that the uric acid level might be independently associated with the risk of MetS and determined that a uric acid level of 314.5 μmol/l was the optimal cut-of value.
In fact, there are diferent opinions about the correlation between uric acid and MetS. An increase of 65% in the risk of MetS per standard deviation increase in uric acid was found using unadjusted observational analyses. Tis association attenuated upon adjustment for potential confounders. Mendelian randomization analyses showed no evidence of a causal association between uric acid and MetS and MetS components    International Journal of Endocrinology [19]. Adnan et al. investigated 102 outpatients. In subjects with MetS, the average serum uric acid level was higher than that in subjects without MetS, but this diference was not signifcant (6.62 vs. 6.28 mg/dL; P � 0.556) [20]. However, some studies even showed that uric acid is an independent risk factor for MetS [21,22]. Cibičková     International Journal of Endocrinology metabolism showed moderate correlations (correlation coefcient r � 0.4 -0.6) with uric acid levels (positive correlation with TAG and AIP and negative correlation with HDL cholesterol), whereas parameters of insulin resistance (glycaemia, insulin, C-peptide, and HOMA-IR) showed low positive correlations (correlation coefcient r � 0.1 -0.3) with uric acid levels [23]. Te above conclusions are consistent with our fndings. A 5-year retrospective cohort study of healthy Japanese adults reported that elevated SUA increased the risk of developing high LDL cholesterol as well as hypertriglyceridemia [24]. In this cross-sectional study, we came to similar conclusions with a signifcant correlation between uric acid levels and blood lipids (triglycerides and HDL-C). A study suggested that higher intracellular uric acid levels can induce mitochondrial translocation of the nicotinamide adenine dinucleotide phosphate oxidase subunit nicotinamide adenine dinucleotide phosphate oxidase 4, further leading to increased mitochondrial oxidative stress, mitochondrial dysfunction, and citrate release to the cytosol, ultimately promoting the synthesis of lipids and TG [25]. In addition, both soluble and crystalline uric acid inhibit AMPkinase, leading to reduced fatty acid oxidation and triglyceride accumulation [26]. Hyperinsulinaemia in the body increases the reabsorption of uric acid in the renal tubules, thereby forming hyperuricaemia. In addition, the enzyme activity that catalyses the decomposition of TG is afected by high levels of uric acid, which inhibits the decomposition of serum TG, leading to the incidence of hypertriglyceridaemia [27].
Experimental studies have shown that hyperuricaemia may mediate insulin resistance in models of fructose-dependent and fructose-independent metabolic syndrome [7,28]. MetS is defned as IR syndrome, which can lead to the occurrence of MetS [29]. Studies have suggested that hyperuricaemia and IR have a bidirectional relationship [30]. Increasing serum uric acid can lead to IR through the low-pressure bioavailability of nitric oxide (NO) and ultimately produce oxidative stress in the mitochondria [20]. IR can also cause hyperuricaemia by increasing the sodium reabsorption mechanism and increasing the absorption of uric acid. Te increase in serum uric acid is negatively correlated with insulin sensitivity [31]. In people with hyperuricaemia, the use of allopurinol or benzbromarone may improve insulin resistance in patients with the metabolic syndrome [32]. Overall, insulin resistance may be the mechanism explaining the association between uric acid and metabolic syndrome. Women with metabolic syndrome and general obesity are at higher risk of developing severe hyperuricaemia compared to men [33]. In a large prospective cohort study of middle-aged and elderly Chinese, researchers also found that the association between uric acid and triglycerides was more pronounced in women [34]. When we plotted the ROC curves, we found diferences in the diagnostic efcacy of uric acid for metabolic syndrome by gender, which seemed to be more signifcant in the female population.
Several limitations of this study should be mentioned. First, cross-sectional research cannot draw a causal relationship between MetS and uric acid afected by the type of research, and the degree of risk cannot be measured in this study design.
Second, we could not obtain some useful data, such as information on drug use. Terefore, we hope to conduct multicentre and prospective research in the future. In summary, we hope to bring new content to the management of MetS for the elderly in China through our research. Te common pathogenesis of hyperuricaemia and MetS is also worthy of further exploration.

MetS:
Metabolic syndrome BMI: Body mass index BP: Blood pressure CVD: Cardiovascular disease FBG: Fasting blood glucose HDL-C: High-density lipoprotein cholesterol LDL-C: Low-density lipoprotein cholesterol TC: Total cholesterol TG: Triglyceride UA: Uric acid WC: Waist circumstance.

Data Availability
Te data used for this paper are available from the corresponding author upon reasonable request.

Ethical Approval
Tis study was conducted in accordance with the contents of the Declaration of Helsinki. Te participants' information was kept anonymous, was noninvasive, and will not cause harm to them. Te Institutional Review Board of Tongji Medical College, Huazhong University of Science and Technology approved the request for a waiver of informed consent. Te approval number is 2020-S273.

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