Association between uric acid and depressive symptoms in older adults: The Korea National Health and Nutrition Examination Survey

Uric acid is known for its antioxidant capacity; however, whether uric acid is independently associated with depression in the older population remains controversial. The current study aimed to investigate the association between uric acid and depressive symptoms in older adults according to sex, using a large national sample population.


| INTRODUCTION
Uric acid is a product of purine metabolism, and it is best known for its pathophysiology of gout, kidney stones, and metabolic syndrome, 1 as well as its association with mortality risk. 2 However, some observational studies found a U-shaped association between urate and mortality, presenting the possible adverse effects of very low serum urate levels. 3,4 The most well-known advantage of uric acid is its antioxidant capacity as a free-radical scavenger, which may have neuroprotective effects. 5,6 Previous findings suggest that higher uric acid levels are associated with a lower risk of developing neurological disorders such as Parkinson's disease 7 and multiple sclerosis 8 and a slow decline in cognitive impairment. 9 Among the many markers of oxidative stress, uric acid is considered promising for a few reasons. First, uric acid acts as a potent antioxidant in the extracellular environment and contributes to over 50% of total serum antioxidant activity. 6 Second, plasma uric acid levels are highly correlated with cerebrospinal fluid levels (r = 0.669, p = 0.001). 10 The prevalence of depression in older adults in South Korea has been increasing annually 11 and has had a huge impact on public mental health. Late-life depression, which occurs in persons aged 60 years or older, is common and often associated with coexisting medical illness, cognitive dysfunction, and increased risk of suicide. 12 The difference in the symptomatology of depressive disorders in older people from younger people may be due to the pathological effects of the aging brain and biochemical changes. 12,13 As oxidative stress pathways have been hypothesized to be involved in the pathophysiological mechanisms of depression, 14 there have been several studies on the relationship between uric acid levels and depression. A large-scale study found that plasma uric acid levels are lower in current, but not remitted, major depressive disorder and anxiety disorder, according to the doseresponse gradient. 15 A recent meta-analysis and meta-regression also found that participants with major depressive disorder had lower uric acid levels than those in the control group. 16 However, only a few sex-related studies have been conducted on this association. In middle-aged and older Chinese participants, a negative association between uric acid levels and the prevalence of depression was found only in men, but not in women. 17 A cross-sectional study on the Korean population found an inverse relationship between uric acid levels and the prevalence of depressive symptoms in the Korea National Health and Nutrition Examination Survey (KNHANES) 2016 dataset. 18 Notably, compared with groups with high levels of serum urate, groups with low serum urate levels were significantly associated with the overall burden of depression in older adults of both sexes. However, because it was a 1-year survey conducted in 2016, it was difficult to determine whether it adequately represented the Korean population. In this study, we aimed to obtain representativeness by using multi-year KNHANES data from 2016, 2018, and 2020 to study the association between serum uric acid levels and the prevalence of depressive symptoms in older adults in Korea. We also investigated the sex differences through subgroup analysis.

| Study population and data
This study was conducted using data from the 2016, 2018, and 2020 KNHANES-a nationwide population-based cross-sectional survey of the health and nutritional status of Koreans conducted by the Korea Disease Control and Prevention Agency. KNHANES is a continuous monitoring system in Korea that analyzes Koreans' health and nutritional status, as well as trends in health risk factors and the prevalence of major chronic diseases. It also provides data for the creation and assessment of Korean health policies and programs. 19 The survey included an established questionnaire to determine the demographic and socioeconomic characteristics of the participants, including age, education level, occupation, income, marital status, smoking habits, alcohol consumption, and previous and current diseases. The survey also collected anthropometric measures and biochemical profiles including uric acid levels using fasting blood serum.
The KNHANES is conducted every year in 20 households from each of the 192 regions, and 10,000 individuals aged ≥1 year were targeted. Survey participants were divided into three age groups: children, adolescents, and adults. Adult participants completed the health interview survey, which included the Patient Health Questionnaire-9 (PHQ-9), a depression screening scale. In this study, we screened 23,501 respondents from 2016, 2018, and 2020 KNHANES data. Exclusion criteria included age under 60 years (n = 16,699), missing serum uric acid data (n = 488), missing valid PHQ-9 scores (n = 603), and missing control variables (n = 102).
Finally, the remaining 5609 participants (2483 men and 3126 women) were found to be eligible for the analysis ( Figure 1).
All participants sign an informed consent form and KNHANES is conducted in accordance with the ethical standards of the Declaration of Helsinki. Because the KNHANES database provides publicly available, de-identified secondary data, the institutional review board of Yongin Severance Hospital waived the requirements for approval for this study.

| Assessment of depressive symptoms
Presence of depressive symptoms was assessed using the PHQ-9, a nine-item self-administered version that has been validated as a reliable depression screening tool and a measure of depression severity. It is based on the criteria of the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, 20 which is used to diagnose major depressive disorder (MDD). Each item on the PHQ-9 is scored on a scale ranging from 0 to 3, after which the scores are added to yield a total score ranging from 0 to 27. PHQ-9 scores of 5, 10, 15, and 20 represent valid thresholds for mild, moderate, moderately severe, and severe depression, respectively.
In an initial validation study, a score of 10 or higher had a sensitivity of 88% and a specificity of 88% for detecting MDD. 21 However, a validation study of the Korean version of the PHQ-9 suggested a score of 5 as the cutoff for screening MDD in the senior population, with a sensitivity and specificity of 81.8% and 89.9%, respectively. 22 This cutoff score is lower than that used in the original validation study, but these differences could be related to the decreased expression of depressed mood reported in Korean patients with MDD. 23 Following previous studies on depressive symptoms using KNHANES data, 24 the presence of depressive symptoms was defined as a PHQ-9 of ≥5 in the current study. Using this cutoff score, the participants were divided according to the presence of depressive symptoms.

| Assessment of uric acid levels
In the KNHANES, after ≥8 h of fasting, blood samples were collected to assess the biochemical marker levels. Venous serum samples were collected in serum separating tubes, centrifuged at 3000 rpm for 15 min, and stored at 2-8°C. Serum uric acid levels were measured within 24 h of sample collection using the colorimetric enzymatic (uricase) method with a Hitachi Automatic Analyzer 7600-210 (Hitachi, Japan).

| Covariates
Sex, Demographic (age), socioeconomic (educational attainment, household income, working status, and marital status), and healthrelated (body mass index (BMI), alcohol consumption status, smoking status, chronic medical diseases, and menopausal status in women) characteristics were included in this study. Medical history was assessed for the presence or absence of hypertension, diabetes mellitus, cardiovascular disease, and stroke, which were included as chronic medical diseases.

| Statistical analyses
We divided the total study participants (aged ≥60 years) by gender, and within each gender, we categorized them into low (Tertile 1), middle (Tertile 2), and high (Tertile 3) groups based on their serum uric acid levels. Statistical analysis was performed using the SAS software (SAS version 9.4, SAS Institute, Cary, North Carolina, USA).
Differences between the general characteristics of the study population were tested using the chi-squared test. To examine the association between uric acid levels and depression, we used multivariable logistic regression analysis. A subgroup analysis initially stratified participants based on each covariate and then conducted univariable logistic regression analyses between uric acid levels and depressive symptoms. Through this approach, the subgroup analysis explored the combined effects of each covariate and uric acid levels on depressive symptoms. Odds ratios (ORs) and 95% confidence intervals (CIs) were computed, and a p-value of <0.05 was considered significant.

| General characteristics
The general characteristics of the study population are summarized in Table 1. Among the 5609 included participants (2483 men and 3126 women), 303 (12.2%) men and 703 (22.5%) women had depressive symptoms (i.e., PHQ-9 ≥5). The prevalence of depressive symptoms was not significantly different among the uric acid tertile groups in men. However, the prevalence of depressive symptoms in women was significantly higher in the lowest uric acid tertile than in the highest tertile (p = 0.023). This association is presented visually in Figure 2. In general, for both sexes, participants with depressive symptoms tended to have low income, be unmarried, be smokers, and have chronic medical diseases. Men with depressive symptoms tended to reside in urban areas. The uric acid levels according to gender is shown in Table 2, and the Mann-Whitney U test showed that the uric acid levels was significantly higher in men compared to women (p < 0.001). Table 3 shows the results of the multivariable logistic regression analysis of the association between uric acid levels and depressive symptoms. In men, there was no statistically significant relationship between depressive symptoms and serum uric acid levels in the lowest group (tertile 1) compared to the highest group (tertile 3) after adjustment for covariates. Low educational level, low income, unmarried status, smoking, and two or more chronic medical diseases were associated with depressive symptoms in men. However, age, alcohol consumption, residential area, and BMI were not significantly associated with depressive symptoms in men.

| Sex differences in the association between uric acid levels and depressive symptoms
In contrast, a significant inverse association was observed between serum uric acid levels and depressive symptoms in women.
Compared with the reference group of serum uric acid levels (tertile 3), the lowest serum uric acid levels group (tertile 1) was approximately 36% more likely to have depressive symptoms (adjusted OR:1.36, 95% CI:1.10-1.68, p = 0.005). Low educational level, low income, smoking, chronic medical disease morbidity, and low BMI were associated with depressive symptoms in women. Age, marital status, alcohol consumption, residential area, and menopause were not associated with depressive symptoms in women. The detailed results of the multivariate logistic regression analysis are shown in Table 3.

| Combined effects of uric acid levels and each covariate on depressive symptoms
The results of the subgroup analysis showing the combined effects of uric acid levels and the study variables on the presence of depressive symptoms in women and men are presented in Table 4. There was no significant combined effect of uric acid levels and each covariate on depressive symptoms in men, except for age and non-alcohol consumption status. For age, the opposite combined effect of uric acid levels and age was observed on depressive symptoms according to sex. In men, the older group (age >80) with the lowest uric acid levels (tertile 1) was 2.93 times more likely to have depressive symptoms than those with the highest uric acid levels (tertile 3) (OR:2.93, 95% CI:1.09-7.88, p = 0.034). However, the younger group (age 60-69) with lower uric acid levels (tertile 1, tertile 2) was more likely to have depressive symptoms than those in the reference uric acid levels group (tertile 3) in women (OR:1.53, 95% CI:1.14-2.05, p = 0.005 for tertile 1, OR:1.41, 95% CI:1.04-1.92, p = 0.029 for tertile 2). In the group without alcohol intake, lower uric acid groups had higher depressive symptoms than the reference group (tertile 3) in both sexes (OR:1.74, 95% CI:1.04-2.91, p = 0.035 for tertile 2 in men, OR:1.37, 95% CI:1.09-1.72, p = 0.007 for tertile 1 in women). In women, high educational level, high-income level, separated/ divorced/widowed status, non-smoking, chronic medical disease, normal weight, and menopausal state showed a combined effect with uric acid. With these covariates, low uric acid levels were significantly associated with the presence of depressive symptoms. The detailed results of the subgroup analyses are presented in Table 4.

| DISCUSSION
In this study, we demonstrated an inverse association between serum uric acid levels and depressive symptoms in women aged over 60 years in large, national-level databases of the Korean population collected in 2016, 2018, and 2020. However, this association was not statistically significant in men.
Depression in older adults is associated with disability and cognitive and functional impairment. 25 Moreover, subthreshold T A B L E 1 Sociodemographic characteristics and clinical features of participants by presence of depression (PHQ-9 ≥5) in older adults.  depressive symptoms also affect the quality of life of older adults. 25 The difference between the symptomatology of depressive disorders of older people and younger people may be due to the pathoplastic effects of the aging brain and biochemical changes 12 ; moreover, there have been many approaches to screening depression in older adults. 26 Oxidative stress has been linked to aging 27 and has been implicated in several age-associated neurodegenerative diseases, including Alzheimer's disease, 28 Parkinson's disease, 7 and nonneurodegenerative diseases. 27 Uric acid is a breakdown product of purine metabolism and is known for its antioxidant effects. 10 In previous research, uric acid has been shown to contribute to over 50% of antioxidant capacity as a free-radical scavenger. 29 As plasma T A B L E 3 Results of the multivariable logistic regression analysis for the association between uric acid and depressive symptoms in older adults.  A previous study has suggested that uric acid levels are negatively associated with depressive symptoms, anxiety symptoms, and symptom duration. 15 Lower uric acid levels are also related to other psychiatric symptoms, such as higher suicide risk. 30 In the current study, the prevalence of depressive symptoms was significantly higher in women with relatively lower uric acid levels than in those with high uric acid levels. This finding is in agreement with previous studies that have shown that women with depression have lower plasma levels of uric acid. 17,31 In particular, the results of the study on older adults in the Korean population are consistent with the findings that used KNHANES data from 2016. 18 This sex-specific difference may be attributed to the susceptibility of women to oxidative stress, which is a major factor in depression. 14 The underlying mechanisms explaining this relationship have been investigated; in particular, inflammatory response induced by oxidative stress has been comprehensively studied as a potential mechanism of mood disorders, including depression. 32,33 A study examining gender differences in oxidative stress and assessing protein carbonyl content, an oxidative stress indicator, indicated that women were more vulnerable to oxidative stress. 34 It has also been revealed that as women age, they become more sensitive to oxidative stress, and some of these mechanisms are known to be related to a decline in estrogen. 35 Another possibility is that, as shown in Table 2, women have relatively lower serum uric acid levels than those in men (p < 0.001), which may explain why women are more prone to depressive symptoms related to uric acid deficiency.

Men
Through subgroup analysis, we determined which covariates had a compounding effect with uric acid levels on the prevalence of depressive symptoms. Regarding alcohol use status, uric acid levels and depressive symptoms showed an inverse relationship only in participants who did not use alcohol. Considering that serum uric acid levels are lower in the non-alcohol use group than in the alcohol use group, 36 a shortage of antioxidant uric acid in the non-alcohol use group might be exacerbated, leading to depression. The results also showed that in the non-smoking group, lower serum uric acid levels were associated with a greater burden of depressive symptoms.
Because smoking is associated with elevated uric acid levels in women, but not in men in the Korean population, 37 so the nonsmoker group had relatively lower uric acid levels and might be more susceptible to shortage of antioxidant effect. During disease states, reactive oxygen species levels can increase exponentially and may damage cellular structures. 38 Therefore, the prevalence of depressive symptoms was high in the group with relatively low uric acid levels among women suffering from one or more chronic diseases because the group suffering from chronic diseases is already vulnerable to oxidative stress. As estrogen functions as an antioxidant by scavenging free radicals, women might be more vulnerable to oxidative stress after menopause, such as an increased risk of cardiovascular disease. 35 This could explain why women with low uric acid levels in the menopausal group had a higher prevalence of depression in this subgroup analysis.
The strength of this study is that the correlation between serum uric acid levels and the prevalence of depressive symptoms in older adults was investigated using a large sample and homogeneous nationwide data. Although there was a similar study using the KNHANES data, 18 single-year data might not be sufficient to adequately represent the Korean population. In this study, representativeness was achieved by integrating multi-year KNHANES data. We also performed subgroup analysis to determine which specific subgroups were more affected by uric acid in their depressive symptoms. However, this study had some limitations. Since this cross-sectional study provides a single perspective of the relationship between serum uric acid levels and depressive symptoms, no causal relationship can be determined. Additional research is needed to determine whether uric acid acts as a protective factor against depressive symptoms. In addition, it is well known that gout is caused by high serum uric acid levels and that the prevalence of depression and gout are positively correlated. 39 However, the impact of gout on the depressive symptoms could not be adjusted because KNHANES data used in this study lacked gout-related information. For the PHQ-9 scale, which was the main measurement scale for detecting depressive symptoms in this study, it is also necessary to consider Abbreviations: BMI, body mass index; CI, confidence interval; OR, odds ratio; PHQ-9, Patient Health Questionnaire-9.

of 11
-KIM ET AL.
T A B L E 4 Subgroup analysis of the association between uric acid and depressive symptoms (PHQ-9 ≥5) stratified by sociodemographic variables. that the use of the scale may be limited to adequately detect whether depressive symptoms can vary depending on gender. Lastly, diet, supplements, and medications, such as antihypertensive drugs and diuretics, that can affect serum uric acid levels were not considered.
In conclusion, our study indicates that low serum uric acid levels are associated with a higher prevalence of depressive symptoms in older women. Relatively lower serum uric acid levels in women compared to men and sex differences in oxidative stress may explain the significant association of uric acid levels and depressive symptoms in older women. The findings may be used to screen for depressive symptoms in the older adult population, and measuring serum uric acid levels in older female patients with depressive symptoms is anticipated to be clinically beneficial. Further studies are necessary to determine the specific mechanisms of the sex difference in the association between uric acid levels and depressive symptoms revealed in this study. Note: Some ORs and 95% CIs are presented as '-' because the related case numbers were not sufficient to be analyzed. Statistically significant values with a p-value of less than 0.05 are bolded.

of 11
-KIM ET AL.