Composite Dietary Antioxidant Index Is Negatively Associated with Hyperuricemia in US Adults: An Analysis of NHANES 2007–2018

Hyperuricemia and its complications are severe risks to human health. Dietary intervention is considered an essential part of the management of hyperuricemia. Studies have reported that the intake of antioxidants has a positive effect on hyperuricemia. Here, we collected data from 8761 participants of the National Health and Nutrition Examination Survey for this analysis. Daily intakes of vitamins A, C, and E; manganese; selenium; and zinc were calculated as the composite dietary antioxidant index (CDAI). The participants were divided into four groups (Q1, Q2, Q3, and Q4) according to the CDAI. Univariate analysis was used to assess the association of covariates with hyperuricemia. The association between the CDAI and hyperuricemia was evaluated using multinomial logistic regression, and its stability was determined by stratified analysis. Our results revealed that the CDAI has a significant negative association with hyperuricemia (Q2: 0.81 (0.69, 0.95); Q3: 0.75 (0.62, 0.90); Q4: 0.65 (0.51, 0.82); P < 0.01). The results of stratified analysis emphasize that this association between CDAI and hyperuricemia is stable. In conclusion, this study suggested a negative association between the CDAI and hyperuricemia.


Introduction
Hyperuricemia (HUA) is clinically defned as high levels of serum uric acid in the body (>7 mg/dL for men and >6 mg/dL for women). Excess uric acid is often deposited in the joint, causing gout [1]. In addition, numerous complications associated with hyperuricemia have been reported, such as chronic kidney disease, cardiovascular disease, type 2 diabetes, and hypertension, and are considered critical burdens on human health [2][3][4][5]. Research data have shown that the prevalence of HUA among adults in the US is 20.2% for men and 20% for women [6]. Even worse, there is no curative treatment available [7]. In general, dietary intervention is considered to play a critical role in the management of HUA.
Recent studies have shown that excess uric acid in HUA patients can locally activate oxidative stress [8]. Oxidative stress produces oxidants, and the accumulation of oxidants leads to DNA oxidation, causing abnormal apoptosis and organ dysfunction and consequently leading to the abovementioned complications [8,9]. Tus, there may be an interaction mechanism between uric acid and antioxidants, and antioxidants may mitigate the damage caused by uric acid.
Vitamin (vit) C, a common antioxidant, has been shown to have a negative association with HUA. Sun et al. also reported that supplementation with vit C might delay the development of hyperuricemic nephropathy [10,11]. In addition, vit A, vit E, and zinc (Zn) have been shown to reduce the level of serum uric acid [12][13][14][15][16]. Although numerous studies have reported the efect of a single antioxidant on HUA, a study evaluating the association between comprehensive antioxidants and HUA still needs to be completed.
According to previous studies, we proposed a hypothesis that the intake of antioxidants and the risk of hyperacidity are inversely associated. We constructed a composite dietary antioxidant index (CDAI) consisting of food parameters for vit A, C, and E; manganese (Mn); selenium (Se); and Zn to represent an individual's antioxidant intake status and to enhance the credibility of disease risk assessment [17]. A cross-sectional study including 8761 participants was conducted based on the National Health and Nutrition Examination Survey (NHANES) from 2007 to 2018 to investigate the association between the CDAI and HUA.

Data Sources and Study
Design. Subject data for this study were obtained from the NHANES 2007-2018. NHANES is a representative cross-sectional survey of the American population designed to collect nutrition and health information from noninstitutionalized people. Te NHANES website provides comprehensive and detailed information on study design, demographics, dietary assessments, health interviews, physical examinations, and laboratory data. Demographic and health-related information was obtained through questionnaires. Health interviews were conducted in the participants' homes. Dietary assessments were obtained through 24-hour dietary memories. Te Mobile Examination Centre was responsible for the physical examination and the collection of blood samples, which were then sent to the laboratory for testing.
Data marked as missing, refused, and did not know were considered missing data and manually excluded by the researcher. Only participants over 20 years of age were considered for inclusion in the study. After excluding missing data for uric acid levels, CDAI scores and covariates, a total of 8761 participants were included in the fnal study. Te fow graph for inclusion and exclusion is shown in Figure 1. All participants provided written informed consent, and the NCHS Research Ethics Review Committee approved the study (https://wwwn.cdc.gov/ nchs/nhanes/default.aspx).
Te CDAI proposed by Wright et al. was obtained by a composite calculation of the intake of multiple dietary antioxidants [17][18][19][20]. For all participants included in this study, the CDAI contained six dietary antioxidants: vit A, C, and E; Mn; Se; and Zn. Te formula for CDAI is CDAI � 6 I�1 X i − μ i /S i [21]. In the formula, X i is the daily intake of antioxidants, μ i is the mean of antioxidants in the study population, and S i is the standard deviation of μ i . In brief, the CDAI is a scoring algorithm based on 24-hour dietary recall data designed to assess the level of antioxidant intake of participants. We averaged participants into four groups based on CDAI, Q1 (−7.177 to −1.178), Q2 (−1.177 to −0.796), Q3 (0.799 to 3.202), and Q4 (3.203 to 88.502).

Covariates.
To exclude other factors interfering with the results, age, sex, race, education, household income to poverty ratio, BMI, dietary capacity and protein intake, hypertension status, diabetes status, smoking status, alcohol consumption status, physical activity, and biochemical indicators, including gamma glutamyl transferase, triglycerides, total cholesterol, HDL cholesterol, and creatinine, were selected as covariates for the analysis. Race was categorized as Mexican American, Other Hispanic, Non-Hispanic White, Non-Hispanic Black, and Other race-including multiracial. Educational attainment included three levels: below high school, high school, and above high school. Household income and poverty rates were used to measure participants' household economic status and were categorized into three levels: less than 1, between 1 and 3, and more than 3. Hypertension was defned as participants taking medication for hypertension or having a past/current diagnosis of hypertension. Diabetes was classifed into four categories: no, impaired fasting glucose (IFG), impaired glucose tolerance (IGT), and yes. Smoking status was categorized as never, former, and now. Participants who consumed at least 12 alcoholic beverages in a year were considered to have drinking behavior. Physical activity included two categories, work activity status and recreational activity status, with four ratings of no, vigorous, moderate, and both. In addition, specifc data for biochemical indicators were provided by the NHANES laboratory. All covariate data for this study can be viewed in detail on the NHANES website (https://www.cdc.gov/nchs/ nhanes/index.htm).

Statistical Analyses.
Te statistical packages R (Te R Foundation; https://www.r-project.org; version 3.6.3) and Empower Stats (https://www.empowerstats.net, X&Y solutions, Inc., Boston, Massachusetts) were used to process the data. In the analysis of participant characteristics, continuous variables were expressed as the "mean ± standard deviation," and categorical variables were expressed as weighted percentages (%). χ 2 tests and Kruskal-Wallis tests were used to assess the signifcance of categorical and continuous variables, respectively. Univariate analysis of variance was used to assess the relation between each covariate and HUA. Multinomial logistic regression analysis with fve adjusted models was used to investigate the association between CDAI and HUA, and the stability of the association was assessed by stratifed analysis. Te 95% confdence intervals were calculated. P < 0.05 was considered statistically signifcant in this study.  with HUA are more likely to be older, to have a high BMI, and to have hypertension and glucose-related disorders. Table 1 describes the participants' characteristics in detail.

Analysis of Factors Associated with HUA.
In the univariate analysis, several covariates were selected as independent exposure variables in this study to determine the factors that interfered with the association between CDAI and HUA. Te results of the univariate analysis indicated that age, sex, race, household income to poverty ratio, marital status, BMI, gamma glutamyl transferase, triglycerides, total cholesterol, HDL cholesterol, creatinine, dietary energy intake, hypertensive status, diabetes status, smoking status, and leisure activity status were statistically signifcant (P < 0.05), demonstrating that they may be potential confounding factors. Te results of the univariate analysis are described in detail in Table 2. Figure 2 describes the association between the CDAI and HUA. Participants in the study were equally allocated into four groups according to the CDAI. Te multinomial logistic regression included one crude and fve adjusted models. A negative and statistically signifcant association of CDAI with HUA was observed in all six models. In model 5, which adjusted for all confounders, participants in the CDAI Q4 group (highest) had a 35% lower risk of sufering from HUA than those in the Q1 group (lowest) (OR � 0.65, 95% CI 0.51, 0.82, P < 0.05). To visualize the association between the CDAI and HUA, smoothed curve fts were plotted according to adjusted model 5. Te results are shown in Figure 3, where the CDAI is negatively associated with HUA.

Stratifcation Analysis.
Statistically signifcant covariates from the univariate analysis were included in the stratifed analysis to assess the stability of the association between CDAI and HUA in diferent populations. All covariates in the stratifed analysis except for the stratifed variables were adjusted. CDAI shows a negative association with HUA in the vast majority of subgroups, except in the few subgroups where a positive association is observed. In addition, no statistically signifcant results are observed for any subgroups with positive associations. Te results suggest that the CDAI has stability in its negative association with HUA and may be a valid protective factor for HUA. Table 3 describes the detailed results of the stratifed analysis.

Discussion
Tis research analysed the most representative US population data (4244 males and 4571 females) and found a negative association between CDAI and the incidence of HUA in the population, which confrmed our hypothesis. Tese results suggest that appropriate modifcations in the level and proportion of antioxidants in the diet may facilitate the prevention and treatment of HUA.
Tis is the frst large-scale study to consider the association between a composite of dietary antioxidants and HUA. Although previous studies have researched the efects of dietary antioxidants such as vit C, vit E, and Zn on HUA, they have not been considered in combination. In the average person's daily diet, it is clear that a single antioxidant intake is difcult to achieve, and it is more likely that a variety of foods and multiple antioxidants are absorbed. In addition, there is no single antioxidant component in food, for example, tomatoes, are rich in vit A, vit E, vit C, Mn, and many other antioxidants [22,23]. Terefore, a comprehensive study on the efects of dietary antioxidants on HUA is necessary.
Te composition of dietary antioxidants includes vit A, C, and E; Mn; Se; and Zn. Tere is a large amount of research in the feld reporting the association, causality, and mechanism of the efect of individual nutrients on HUA. A study of 1387 males showed a negative association between vit C intake and HUA [24]. Studies by Huang et al. also showed that vit C intake can reduce serum uric acid levels and prevent the development of gout [25,26]. Current reports in the feld of vit C reducing serum uric acid focus on the renal excretion mechanism. Two vit C transport proteins, SLC23A1 and SLC23A2, existing in proximal renal tubular epithelial cells, can alter the activity of URAT1 in renal tubular cells, thus promoting uric acid excretion and reducing serum uric acid levels [27][28][29][30][31]. Te mechanism of xanthine oxidase (XO) inhibition is also of concern. An in vitro study reported that vit C and vit E could inhibit XO activity. Studies have also suggested a negative association between vit E and HUA [12,13,32]. Similar associations and mechanisms have also been reported for dietary Zn [14,[33][34][35]. Te role of Se in HUA is currently controversial, with some studies claiming a negative association between serum selenium and uric acid levels [15,36]. However, other studies have suggested the opposite. Te diference may be related to the source of Se [37]. Tis suggests that we consider the role of other substances in food while increasing the dietary intake of antioxidants. Ma et al. showed a negative association between Mn and uric acid levels [16].
Tere is a near consensus that XO plays a crucial role in uric acid metabolism. Uric acid is generated during the XOcatalyzed conversion of hypoxanthine and xanthine and is accompanied by the production of reactive oxygen species (ROS) [38]. Furthermore, uric acid increases the production of infammatory factors and reduces the amount of free radical nitric oxide (NO) in cells, which can lead to an increase in ROS levels and oxidative stress, with the contribution of XO [39,40]. Tis process can activate osteoclasts and inhibit osteoblasts, resulting in increased bone loss and triggering osteoporosis and bone destruction [39,[41][42][43]. Several antioxidants have been reported to function as XO inhibitors, reducing serum uric acid levels and scavenging intracellular oxygen radicals, thereby ameliorating the impairment of hyperuricemia. Zeng et al. showed that baicalein and baicalein, two antioxidants widely present in plants, could bind to the FAD center of XO and inhibit XO activity, suppressing uric acid production and oxidative stress levels [38]. Ellagic acid, another natural antioxidant, has also been 4 International Journal of Endocrinology International Journal of Endocrinology 5        [44]. Moreover, the inhibition of XO also contributes to increased bone formation by promoting osteoblast diferentiation, which may be associated with the inhibition of ROS [39,[45][46][47]. We hypothesize that the association of a high CDAI with a low risk of hyperuricemia may be attributable to the inhibition of dietary antioxidants on XO. In fact, Li et al. showed that vit C supplementation reduced XO activity in hyperuricemic rats and decreased ROS levels by inhibiting TGF-beta [11]. Similar fndings have also been reported in other studies [12,44,48,49]. However, studies exploring the inhibitory capacity of these antioxidants against XO and their mechanisms are still insufcient. Our fndings may provide valuable insights for further research. In this study, the association between the CDAI and hyperuricemia was explored through a cross-sectional study that included 8761 participants. Diferent models were constructed to exclude the efect of confounding factors on the results. Te results showed that the negative association between CDAI and hyperuricemia was stable across the diferent models. In the fnal adjusted model 5, the Q4 group (highest) had a 35% lower risk of sufering from HUA than the Q1 group (lowest) (OR � 0.65, 95% CI 0.51, 0.82, P < 0.05). Tis suggests that the treatment of hyperuricemia may be facilitated by adjusting the ratio of antioxidants in the diet.
To summarize, the CDAI is negatively associated with HUA. However, there are still some limitations to our study.
First, we cannot determine the causal relation between the CDAI and HUA, which is limited because our research is a cross-sectional study. Although a large number of studies have shown that dietary antioxidants reduce serum uric acid levels, further cohort studies and clinical trials are needed to determine the causal relationship between CDAI and HUA and the therapeutic value of CDAI. Second, although a large number of participants were included in this study, it was limited to US residents only. Considering the diferences between a variety of factors, such as body composition, lifestyle, and dietary habits of residents in diferent countries and regions, multicenter controlled trials are needed to validate our fndings.

Conclusion
Our study suggests that there is a signifcant and stable negative association between CDAI and HUA in the population. Tis suggests that the CDAI may be an efective protective factor for HUA and holds promise as a preventive and therapeutic tool for HUA.

Ethical Approval
All procedures performed in studies involving human participants followed the ethical standards of the Institutional and National Research Committee and the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Consent
Informed consent was obtained from all participants included in the study.

Disclosure
Zhenzong Lin and Haokai Chen should be regarded as cofrst authors.