A Mendelian randomization study of serum uric acid in metabolic and cardiovascular risk in East Asian populations

Abstract

Serum uric acid (SUA) levels are linked to increased disease vulnerability and higher recurrence rates; however, the exact causal relationships are elusive. Despite the prevalent hyperuricemia in East Asian populations, comprehensive research on the intricate association between SUA levels and disease is lacking. To address this, a study utilizing a 2-sample Mendelian randomization (MR) approach was conducted in East Asian populations. This study utilized MR to explore the correlation between SUA levels and various disorders, employing data from genome-wide association studies and multiple independent single-nucleotide polymorphism. Multiple single-nucleotide polymorphisms were applied to assess the causal relationship between SUA and other diseases. Methodologies encompassed inverse-variance weighting, MR-Egger regression, and weighted median analysis. This study revealed that SUA increases the risk of coronary artery disease (β = 0.197 mm, 95% CI: 0.084–0.31 mm, P = .001) but decreases the risk of rheumatoid arthritis (β = −0.172 mm, 95% CI: −0.302 to −0.043 mm, P = .009). It also increases diastolic blood pressure, serum creatinine, eosinophil count, relative wall thickness, posterior wall thickness, and interventricular septum thickness and decreases high-density lipoprotein cholesterol and the estimated glomerular filtration rate. These findings suggest that SUA may be a potential risk factor for certain diseases. Research indicates a strong correlation between SUA and illnesses, particularly metabolism and rheumatoid arthritis, in East Asians. This study underscores the necessity of monitoring SUA levels to prevent further illnesses and prompt action to address the growing burden of SUA in the East Asian populations.

1. Introduction

Serum uric acid (SUA) is a final result of purine metabolism and is produced by the breakdown of nucleic acids (DNA and RNA) and adenosine triphosphate.^[1] A substance called uricase (urate oxidase) changes SUA into 5-hydroxyisourates and allantoin. These are then flushed out of the body through urine.^[2] Throughout the process of primate evolution, there has been a gradual decline in the efficacy of the enzyme uricase, ultimately leading to its inhibition in humans. This genetic anomaly results in heightened concentrations of SUA in the bloodstream, hence impeding the regulation of SUA levels.^[3]

Research findings indicate that apes, lacking the enzyme uricase, can exhibit elevated SUA levels of approximately 3 to 4 mg/dL.^[4] An association has been established between increased concentrations of SUA and vulnerability to a range of medical conditions, such as gout, renal calculi, adiposity, metabolic syndrome, diabetes mellitus, hepatic steatosis, and hypertension (HTN), as well as cardiovascular and renal disorders.^[5] Another study revealed that male patients diagnosed with amyotrophic lateral sclerosis exhibited a survival benefit when SUA levels were higher.^[6] One study indicated a potential correlation between SUA and increased dopamine levels within the brain.^[7,8] People with hyperuricemia exhibit a reduced susceptibility to Parkinson disease.^[9] In addition, the results were not clear-cut because observational research has its own problems, such as the presence of unmeasured confounding variables and the chance of reverse causality. These limitations hindered our ability to establish a causal relationship. Thus far, the biological advantages and disadvantages of uric acid, as well as the evolutionary rationales underlying the natural selection of this genetic variation in humans, are unknown.

Mendelian randomization (MR), an analytic tool that has gained prominence in recent years, has become commonly used to infer causal effects resulting from exposures on outcomes. To obtain dependable evidence about the link between exposures and risk of disease, the MR method can be used when randomized controlled trials are unavailable or when it is not possible to conduct new randomized controlled trials.^[10–12] In light of the considerable incidence of hyperuricemia in East Asia, it is imperative to investigate the potential ramifications of this illness in populations of East Asian descent.^[13] Therefore, the main goal of this investigation was to determine the strength of the correlation between SUA levels and disease risk in a group of East Asian people who were meant to be representative. Additionally, this research sought to determine the relationship of SUA levels on disease and clinical factor outcomes through the utilization of MR methodology. The results of this study could help scientists learn more about the positive and negative biological effects of uric acid.

2. Methods

The current study used MR and followed the STROBE-MR guidelines for reporting observational studies in epidemiology.^[14] In earlier investigations, subjects provided written informed consent, and these trials were approved by the respective ethics review boards.^[15]

2.1. Data sources for SUA and instrumental variable (IV) selection

The genome-wide association study (GWAS) data on SUA levels used in this study were made public. The study looked at 12,716 people from 2 Japanese cohorts, the Japan Multi-institutional Collaborative Cohort and the Kita-Nagoya Genomic Epidemiology cohort. SUA levels in the the Japan Multi-institutional Collaborative Cohort study were assessed using 2 methods, with exclusion criteria for individuals on medication for gout or hyperuricemia. The Kita-Nagoya Genomic Epidemiology study involved 3975 Japanese individuals aged 50 to 80, undergoing annual health assessments from 2005 to 2007 (Tables S1, Supplemental Digital Content, https://links.lww.com/MD/P843).^[15]

2.2. Data sources for other diseases

GWAS data for additional diseases and clinical factors were obtained from the Biobank of Japan, a substantial repository of biological samples primarily comprising individuals of East Asian descent. The dataset includes over 200,000 samples collected specifically from Japanese individuals aged 20 to 89, who provided written informed consent and were followed up in twelve medical facilities in Japan from 2003 to 2018. Ethical approval for Biobank of Japan 16-related GWASs was granted by the ethical committees of the RIKEN Yokohama Institute and the Institute of Medical Science at the University of Tokyo. The GWAS data on these diseases were obtained from the MR-Base database (https://gwas.mrcieu.ac.uk/).^[16]

2.3. Instrumental variable (IV) selection

This investigation utilized multiple independent single-nucleotide polymorphisms (SNPs) to assess the causal association between SUA levels and various diseases. The selection of IVs was guided by 3 hypotheses: the IVs should demonstrate a strong association with the exposure, they must remain independent of any confounding factors in the exposure–outcome relationship, and they should exert an influence on the outcome solely through the exposure.^[17] SNPs exhibiting significant associations with the exposure across the entire genome (P < 5 × 10^-8) were considered, while those exhibiting linkage disequilibrium (LD) (r² < 0.001, distance < 1000 kb) were excluded. The F-statistic is a widely used measure to assess the strength of the relationship between IVs and exposures. A higher F-statistic indicates a more robust instrument. Instrumental variables are classified as strong if their F-statistic exceeds 10.^[18] The remaining SNPs were aggregated into the GWAS database for outcome evaluation, with each SNP scrutinized for potential violations of assumptions 2 and 3 using PhenoScanner (www.phenoscanner.medschl.cam.ac.uk).^[19]

2.4. MR analyses

This study investigated several methodologies, including inverse-variance weighting (IVW), MR-Egger, and the weighted median, to assess the association between exposure and outcomes.^[20] The IVW method, commonly used to evaluate MR estimates for causal effects assuming balanced pleiotropy, was employed.^[21] Relevant SNPs and outliers were extracted from the dataset, and the primary analysis utilized the multiplicative random-effects IVW model to mitigate heterogeneity bias.^[22] MR-Egger and weighted median methodologies were employed to improve the accuracy of IVW estimates, ensuring more robust estimations across various scenarios. MR-Egger allows for pleiotropic effects across all genetic variants but necessitates independence of effects from the variant–exposure association.^[23] The weighted median method allows for the inclusion of invalid instruments, assuming at least 50% of the IVs used in IV analysis are valid. In instances of significant inconsistencies in IVW estimates, a stricter threshold for the P value of the IV was applied.^[24]

In our investigation, Cochran Q test served to identify heterogeneity in MR analysis.^[25] The study aimed to test bias from a horizontal pleiotropy using methods such as the MR-Egger intercept, which captures the average pleiotropic effect across genetic variants. To identify horizontal pleiotropic outliers, the MR pleiotropy residual sum and outlier (MR-PRESSO) test was employed, which removed the variant under scrutiny and repeated IVW regression.^[26] The automated detection of outliers was accomplished using radial MR analysis.^[27] Additionally, a leave-one-out study was conducted to examine the impact of potentially pleiotropic SNPs on causal estimations.^[28] The research flow chart is depicted in Figure 1.

Figure 1.

Flow chart of the Mendelian randomization study. GWAS = genome-wide association study, MR = Mendelian randomization, MR-PRESSO = MR pleiotropy residual sum and outlier, SNP = single-nucleotide polymorphism.

2.5. Statistics

The present study utilized the TwoSampleMR (version 0.4.25), MR-PRESSO (version 1.0), and RadialMR (version 1.1) packages of R (version 4.3.2) software to perform MR analysis. In the context of a global-level examination, a statistically significant 2-sided P value was determined with a significance threshold of .05. In our analysis, we deemed a significance level of P < .05 to indicate statistical significance. Additionally, we sought to determine consistent directions and magnitudes across the 3 MR procedures employed. Furthermore, the analysis did not provide any evidence of heterogeneity or pleiotropy. Additionally, no influential data points were identified in the LOO analysis. The effect estimates are presented in beta, where the outcome variable was continuous, and were transformed to ORs when the outcome variable was binary.^[17]

3. Results

Following the removal of 3 out of 34 variants due to LD with other variants or absence from the LD reference panel, 31 index SNPs were chosen for genetic prediction of SUA (Table S1, Supplemental Digital Content, https://links.lww.com/MD/P843). Subsequently, we performed an extensive MR study to investigate the association between SUA and 45 diseases alongside 64 clinical factors. Several diseases or clinical factors influenced by SUA have been identified.

For diseases, SUA was found to increase the risk of coronary artery disease (β = 0.197 mm, 95% CI: 0.084–0.31 mm, P = .001) but decrease the risk of rheumatoid arthritis (RA) (β = −0.172 mm, 95% CI: −0.302 to −0.043 mm, P = .009). Heterogeneity was observed with a Cochran Q derived P value < .05. Heterogeneity was acceptable since the major result we utilized was that of the random-effects IVW analysis. The P value for the MR-Egger intercept was more than .05 (Table 1). The leave-one-out plot showed that there was little to no bias in the results when any single SNP among the genetic variations was excluded (Figure S1, Supplemental Digital Content, https://links.lww.com/MD/P842). To obtain robust findings, the outliers identified by MR-PRESSO and radial MR were eliminated. SUA was still positively associated with coronary artery disease risk (β = 0.253 mm, 95% CI: 0.157–0.349 mm, P = 2.341 × 10^-7) and negatively associated with RA (β = −0.201 mm, 95% CI: −0.327 to −0.075 mm, P = .002) (Table 2) (Figure S2, Supplemental Digital Content, https://links.lww.com/MD/P842).

Table 1.

MR analyses investigating the causal impact of SUA concentrations on disease incidence.

Name	IVW		Cochran Q derived P-value	MR-Egger intercept derived P-value
	β (95% CI)	P
Coronary artery disease	0.197 (0.084, 0.31)	.001	1.358 × 10-5	.053
Rheumatoid arthritis	−0.171 (−0.302, −0.043)	.009	.352	.297

MR = Mendelian randomization, SUA = serum uric acid.

Table 2.

MR analyses investigating the causal impact of SUA concentrations on disease incidence after MR-PRESSO and radial MR.

Name	IVW		MR-Egger		Weighted Median
	β (95% CI)	P	β (95% CI)	P	β (95% CI)	P
Coronary artery disease	0.253 (0.157, 0.349)	2.341 × 10-7	0.209 (0.438, −0.021)	.09	0.233 (0.078, 0.387)	.003
Rheumatoid arthritis	−0.201 (−0.327, −0.075)	.002	−0.215 (−0.457, 0.026)	.092	−0.174 (−0.381, 0.032)	.098

MR = Mendelian randomization, MR-PRESSO = MR-pleiotropy residual sum and outlier, SUA = serum uric acid.

For clinical factors, SUA was associated with increased diastolic blood pressure (OR = 1.078 mm, 95% CI: 1.033–1.126 mm, P = .001); eosinophil count (OR = 1.061 mm, 95% CI: 1.017–1.107 mm, P = .006); high-density lipoprotein cholesterol (HDL-C) (OR = 0.916 mm, 95% CI: 0.873–0.962 mm, P = .0004); interventricular septum thickness (OR = 1.093 mm, 95% CI: 1.007–1.187 mm, P = .034); posterior wall thickness (OR = 1.095 mm, 95% CI: 1.013–1.183 mm, P = .023); relative wall thickness (OR = 1.087 mm, 95% CI: 1.016–1.164 mm, P = .016); estimated glomerular filtration rate (eGFR) (OR = 0.902 mm, 95% CI: 0.851–0.957 mm, P = .001), and serum creatinine (OR = 1.092 mm, 95% CI: 1.032–1.157 mm, P = .002) (Table 3) (Figure S3, Supplemental Digital Content, https://links.lww.com/MD/P842). The P value for the MR-Egger intercept was more than .05. The leave-one-out plot demonstrated that when any individual SNP of the genetic variants was eliminated, the findings were very rarely or never biased (Figure S1, Supplemental Digital Content, https://links.lww.com/MD/P842). To obtain robust findings, the MR-PRESSO and radial MR outliers were eliminated, SUA was still positively associated with diastolic blood pressure (OR = 1.065 mm, 95% CI: 1.035–1.095 mm, P = 1.297 × 10^-5); eosinophil count (OR = 1.067 mm, 95% CI: 1.026–1.109 mm, P = .0011); HDL-C (OR = 0.878 mm, 95% CI: 0.836–0.922 mm, P = 1.702 × 10^-7); interventricular septum thickness (OR = 1.116 mm, 95% CI: 1.041–1.197 mm, P = .002); posterior wall thickness (OR = 1.116 mm, 95% CI: 1.04–1.197 mm, P = .002), relative wall thickness (OR = 1.103 mm, 95% CI: 1.029–1.182 mm, P = .005); eGFR (OR = 0.87 mm, 95% CI: 0.81–0.933 mm, P = .00011), and serum creatinine (OR = 1.139 mm, 95% CI: 1.064–1.218 mm, P = .00017) (Table 4). The details are presented in Figures S4 to S7, Supplemental Digital Content, https://links.lww.com/MD/P842.

Table 3.

MR analyses investigating the causal impact of SUA concentrations on clinical factors.

Name	IVW		Cochran Q derived P value	MR-Egger intercept derived P value
	β (95% CI)	P
Diastolic blood pressure	1.078 (1.033, 1.126)	.001	9.062 × 10-5	.155
Eosinophil count	1.061 (1.017, 1.107)	.006	.185	.275
High-density lipoprotein cholesterol	0.916 (0.873, 0.962)	.0004	.022	.027
Interventricular septum thickness	1.093 (1.007, 1.187)	.034	.034	.824
Posterior wall thickness	1.095 (1.013, 1.183)	.023	.099	.693
Relative wall thickness	1.087 (1.016, 1.164)	.016	.693	.324
Estimated glomerular filtration rate	0.902 (0.851, 0.957)	.001	2.586 × 10-11	.237
Serum creatinine	1.092 (1.032, 1.157)	.002	1.444 × 10-10	.292

IVW = inverse-variance weighting, MR = Mendelian randomization, SUA = serum uric acid.

Table 4.

MR analyses investigating the causal impact of SUA concentrations on clinical factors after MR-PRESSO and radial MR.

Name	IVW		MR-Egger		Weighted median
	β (95% CI)	P	β (95% CI)	P	β (95% CI)	P
Diastolic blood pressure	1.065 (1.035, 1.095)	1.197 × 10-5	1.034 (0.985, 1.085)	.195	1.059 (1.017, 1.103)	.006
Eosinophil count	1.067 (1.026, 1.109)	.0011	1.015 (0.948, 1.086)	.682	1.022 (0.965, 1.082)	.461
High-density lipoprotein cholesterol	0.878 (0.836, 0.922)	1.702 × 10-7	0.921 (0.818, 1.037)	.19	0.881 (0.812, 0.957)	.003
Interventricular septum thickness	1.116 (1.041, 1.197)	.002	1.118 (0.989, 1.263)	.086	1.113 (1.001, 1.238)	.049
Posterior wall thickness	1.116 (1.04, 1.197)	.002	1.075 (0.951, 1.115)	.253	1.08 (0.979, 1.193)	.115
Relative wall thickness	1.103 (1.019, 1.182)	.005	1.037 (0.919, 1.171)	.559	1.059 (0.954, 1.175)	.185
Estimated glomerular filtration rate	0.87 (0.81, 0.933)	.00011	0.846 (0.719, 0.995)	.059	0.887 (0.819, 0.96)	.003
Serum creatinine	1.139 (1.064, 1.218)	.00017	1.23 (1.051, 1.441)	.02	1.126 (1.042, 2.116)	.003

IVW = inverse-variance weighting, MR = Mendelian randomization, MR-PRESSO = MR-pleiotropy residual sum and outlier, SUA = serum uric acid.

4. Discussion

In this MR study, we carefully examined the associations between disease incidence and clinical indicators and between disease incidence and gene-predicted SUA levels in East Asian populations. The primary finding was that there were strong correlations between SUA and metabolic parameters, such as reduced HDL-C, increased diastolic blood pressure, impaired eGFR, elevated serum creatinine levels, abnormal eosinophil count, altered relative wall thickness, increased posterior wall thickness, thickened interventricular septum, and the presence of coronary artery disease. For a long time, SUA was believed to be an unwanted product essential for the removal of nitrogen in the human body, and its concentration in the blood was considered only indicative of kidney function. Nevertheless, the notion that SUA is merely an inert, neutral metabolite has faced significant scrutiny. Multiple foundational and observational studies, together with certain intervention studies, have established a direct causal relationship between SUA and metabolic dysfunction. SUA has also been acknowledged as a danger contributor to cardiovascular disease (CVD) and renal diseases.^[29] The relationships between hyperuricemia and arteriosclerosis, HTN, chronic renal disease, and CVD are outlined in Figure 2.

Figure 2.

The mechanisms underlying the relationships between hyperuricemia and arteriosclerosis, hypertension, chronic renal disease, and cardiovascular disease. *Note: The breakdown of ATP generates uric acid, the accumulation of which leads to endothelial dysfunction, triggers inflammation, and increases the risk of atherosclerosis, hypertension, chronic kidney disease, and cardiovascular disease. At the same time, the oxidative stress generated during uric acid production can exacerbate endothelial damage. Drugs such as allopurinol and febuxostat can reduce these effects by inhibiting uric acid production. ATP = adenosine triphosphate, CKD = chronic kidney disease, CVD = cardiovascular disease, HTN = hypertension.

SUA, through the action of xanthine oxidase, produces reactive oxygen species such as superoxide. In addition, while SUA has antioxidant effects in the bloodstream, it functions as a prooxidant within cells to activate a specific catalytic subunit of the NADPH oxidase NOX4.^[30] In the end, the presence of reactive oxygen species and NOX4 leads to impaired mitochondrial performance and decreased oxidation of glucose and fatty acids. SUA in the endothelial decreases the accessibility of nitric oxide, which is a substance that dilates blood vessels. This leads to damage of the blood vessel lining and impaired function, ultimately contributing to high blood pressure.^[31] Gill team explored the relationship between uric acid levels, blood pressure, and CVD risk using a MR approach. They combined data from GWAS, such as the UK Biobank and the Million Veterans Program. The results showed that genetically predicted elevated SUA levels were associated with a significant increase in systolic blood pressure. Additionally, blood pressure partially mediated the effect of uric acid on CVD risk.^[32] An analysis of health examination data unveiled that in a cohort not undergoing any pharmacological interventions to mitigate blood pressure or SUA levels, the risk of HTN surged by approximately 20% with each 1 mg/dL rise in blood SUA.^[33]

Elevated levels of SUA can impact the process of lipid metabolism in the liver. A recent study that followed a group of people over a long period of time revealed a strong link between triglycerides and HDL-C levels, dyslipidemia, and the development of hyperuricemia.^[34] HDL cholesterol, a blood lipid, has a positive impact on human health by eliminating cholesterol from the walls of blood vessels and reducing the development of atherosclerosis. Another study analyzed UK Biobank data using MR to explore the causal relationship between SUA levels and HDL-C. The results showed that elevated uric acid levels may lead to lower HDL-C levels, suggesting that hyperuricemia may increase the risk of metabolic syndrome by reducing HDL-C levels.^[35] Several clinical investigations have indicated that elevated SUA can lower HDL-C levels, thereby increasing the likelihood of developing CVD.^[36]

Increasing levels of SUA can elevate the concentration of cyclooxygenase II in vascular smooth muscle. This may contribute to coronary artery atherosclerosis and endothelial cell lesion formation, leading to coronary artery disease. One study examined the positive correlation between SUA levels and the Gensini score, an indicator of coronary artery disease extent in patients.^[37] Additionally, the investigation found correlations between SUA levels and measures such as relative wall thickness, posterior wall thickness, and ventricular septal thickness. Another study systematically analyzed the causal relationship between SUA levels and 6 major CVDs, such as coronary heart disease, HTN, myocardial infarction, heart failure, angina pectoris, and coronary artery disease, using a 2-sample MR method combined with data from the UK Biobank. The results showed that genetically predicted high uric acid levels were associated with a significantly increased risk of CVDs: coronary heart disease risk increased by 15.5%, HTN by 31.8%, myocardial infarction by 18.4%, heart failure by 15.8%, angina by 15.0%, and coronary artery disease by 17.0%. Sensitivity analyses verified the robustness of these findings, further supporting the causal role of high uric acid levels in the development of CVD.^[38] RWT, an indicator of heart anatomy and function, serves as a prognosis marker in chronic heart failure. Conditions like HTN and aortic valve stenosis or HTN and hypertrophic cardiomyopathy affect the left ventricular wall thickness and interventricular septal thickness.^[39] Consequently, an elevated level of SUA may heighten the likelihood of developing CVD.

Serum creatinine and eGFR are pivotal indicators of renal function, exhibiting a reciprocal relationship. eGFR serves as a quantitative measure of the kidneys’ ability to eliminate waste substances from the bloodstream. A reduced eGFR leads to inefficient removal of waste products, such as serum creatinine, from the blood, resulting in elevated quantities of these compounds in the bloodstream. In the event of glomerular filtration dysfunction, there will be a gradual increase in the serum creatinine level. Elevated serum creatinine levels are indicative of impaired glomerular filtration. The presence of SUA can have negative physiological consequences for the serum creatinine concentration and glomerular filtration rate through proliferation of vascular calm muscle cells, malfunction of endothelial cells, and oxidative stress.^[40] A study reveals the causal relationship between SUA levels, renal function, and racial differences through cross-ethnic MR analyses of GWAS data from the UK BioBank, the Japan BioBank, the CKDGen Consortium, and the Global Uric Acid Genetics Consortium. In European populations, elevated uric acid contributed to decreased eGFR, increased BUN, and a heightened risk of chronic kidney disease. Conversely, in African populations, uric acid levels were not significantly and causally related to eGFR but were associated with increased BUN. Meanwhile, in Asian populations, elevated uric acid was associated with a significant decrease in eGFR. These results reveal racial differences in the impact of uric acid on renal function and highlight the need for race-specific clinical interventions regarding uric acid levels in different races.^[41] A study was performed to estimate the prevalence of hyperuricemia and chronic kidney disease. As the SUA level increased, there was a significant decrease in the eGFR, whereas serum creatinine levels increased.^[42]

However, an analysis comparing SUA concentrations from RA patients and healthy controls in the present study revealed no significant changes in uric acid concentrations between RA patients and controls.^[43] Traditional observational studies on the impact of SUA on RA remain unclear due to small sample sizes, residual confounding factors, and reverse causality. Our study provides strong evidence of a causal association between genetic liability for SUA and increased RA risk using various MR methods, supporting causal inference.

The genetic instruments we used for our study were robust IVs, with F values surpassing the widely recognized threshold of 10. Furthermore, we conducted an examination of the PhenoScanner to verify that our results are unaffected by any possible confounding variables. We conducted a thorough assessment of directional pleiotropy by examining the intercept derived from the MR-Egger analysis. Additionally, we conducted a leave-one-out test to determine whether any individual SNP had an impact on the MR estimations.

The results of our study establish a basis for researchers to investigate the correlations between SUA and other medical conditions, with particular emphasis on metabolism and RA. However, it is important to acknowledge that our findings are derived from East Asian populations, and the external validity of these results for other ethnic groups, regions, and healthcare settings remains uncertain. Differences in genetic backgrounds, environmental exposures, dietary habits, and healthcare systems may influence the observed associations. Future investigations should seek to replicate and validate these findings using diverse cohorts from various regions and ethnicities, thus enhancing the external validity of our conclusions. Moreover, our study relied on a single GWAS dataset, which may limit the robustness and reliability of the findings. To address this, future research should incorporate independent validation using additional datasets, ideally drawn from similar ethnic backgrounds as well as more diverse populations. Such replication efforts will help confirm the stability of our results and strengthen the evidence for a causal relationship between SUA levels and various metabolic and cardiovascular conditions. Further research should aim to clarify the mechanism that links SUA levels with metabolism and RA. Additionally, new therapeutic approaches for preventing or managing complications associated with elevated or decreased SUA levels should be investigated.

Acknowledgments

All authors would like to acknowledge J-MICC, KING, and BBJ for their selfless public sharing of GWAS summary data.

Author contributions

Conceptualization: Yanan Li.

Funding acquisition: He Shi.

Methodology: Jianing Li, He Shi.

Validation: He Shi.

Writing – original draft: Jianing Li.

Writing – review & editing: Shuang Zhang, He Shi.