Interactive association of baseline and changes in serum uric acid on renal dysfunction among community-dwelling persons.

Abstract Background Chronic kidney disease (CKD) is a major public health concern. Baseline serum uric acid (SUA) levels were independently associated with incident renal dysfunction, but whether baseline and changes in SUA produce an interactive effect on renal dysfunction remains unclear. Methods The subjects comprised 460 men aged 68 ± 10 (mean ± standard deviation) years and 635 women aged 68 ± 9 years from a rural village. We have found participants who underwent a similar examination 3 years later, and analyzed the relationship between baseline SUA, changes in SUA, and a 3‐year follow‐up renal function evaluated by estimated glomerular filtration rate (eGFR). Results A total of 93 (20.2%) men and 76 (12.0%) women had hyperuricemia (men: SUA ≥ 7.0 mg/dL and women: SUA ≥ 6.0 mg/dL) at baseline. Multiple regression analysis using changes in eGFR as objective variable, adjusted for risk factors as explanatory variables, showed that the baseline SUA and changes in SUA were linearly associated with changes in eGFR (β = －0.115, P < .001 and β = －0.431, P < .001, respectively). In both normal SUA group and hyperuricemia group, changes in SUA significantly associated with changes in eGFR (β = －0.473, P < .001 and β = －0.197, P = .009, respectively). Participants with increased SUA from normal to hyperuricemia group had greater eGFR decline over the follow‐up period, and their multivariate‐adjusted 3‐year follow‐up eGFR was significantly lower than in other groups (P < .001). Conclusion Our data demonstrated that baseline and longitudinal changes in SUA were independently and interactively associated with the renal function decline among community‐dwelling persons.


| INTRODUC TI ON
The increasing prevalence of chronic kidney disease (CKD) is a common public health concern. 1 In Japan, the prevalence of CKD increased significantly with time, 2 and similar increases are being observed worldwide. There is now convincing evidence that persons with CKD and hyperuricemia have high rates of morbidity (eg, cardiovascular disease [CVD]), mortality, [3][4][5] healthcare utilization, 6,7 and progression to end-stage renal disease, 8 for which systemic inflammation and oxidant stress are associated with a higher risk. 9 For decades, high serum uric acid (SUA) levels were mainly considered a result rather than a cause of renal dysfunction. 10 However, several experimental and epidemiological studies have demonstrated that increased SUA in humans is associated with systemic inflammation, 11 hypertension, 12 and progression to end-stage renal disease. 13 The main pathophysiological mechanisms of these deleterious effects caused by uric acid are endothelial dysfunction, activation of local renin-angiotensin system, increased oxidative stress, and proinflammatory and proliferative actions. 14 Moreover, changes in SUA from baseline to year 1 were also correlated with the decreased changes in the estimated glomerular filtration rate (eGFR). 15 These studies provide direct evidence that SUA may be a true mediator of renal disease and progression. However, to our acknowledgment, few studies have examined whether baseline and changes in SUA have an interactive effect on the risk of renal dysfunction in the general Japanese population.
Thus, the aim of this study was to evaluate the relationship be-

| Subjects
The present study was a retrospective cohort designed as part of the Nomura study. 16 We have found and compared participants who underwent a similar examination 3 years later to this study. The study population was selected from those who received a commu-

| Evaluation of risk factors
Information on demographic characteristics and risk factors was collected using the clinical files of the subjects. BMI was calculated by dividing weight (in kilograms) by the square of height (in meters).
Smoking status was defined as the number of cigarette packs per day multiplied by the number of years smoked (pack-year), and the participants were classified into never smokers, past smokers, light smokers (<20 pack-year), and heavy smokers (≥20 pack-year).

| Statistical analysis
All values are expressed as the mean ± standard deviation (SD), unless otherwise specified, and in the cases of parameters with non-normal distribution (such as TG and HbA1c), the data are shown as median (interquartile range) values. In all the analyses, parameters with non-normal distributions were used after logtransformation. Statistical analysis was performed using IBM SPSS Statistics Version 21 (Statistical Package for Social Science Japan, Inc). Hyperuricemia was defined as the level of SUA ≥ 7.0 mg/ dL in men and ≥ 6.0 mg/dL in women. 18 Three-year changes in SUA and eGFR were calculated by subtracting the baseline values from those at 3 years. Participants were divided into four groups according to baseline and 3-year follow-up SUA (normal SUA throughout the follow-up period, N → N group; hyperuricemia at baseline and normal SUA later, H → N group; normal SUA at baseline and hyperuricemia later, N → H group; and hyperuricemia throughout the follow-up period, H → H group). Differences in means and prevalence among baseline and follow-up findings were analyzed by paired t test for continuous data and chi-square test for categorical data, respectively. Pearson's correlations were calculated in order to characterize the associations between various characteristics and changes in eGFR. Multiple linear regression analysis was used to evaluate the contribution of each confounding factor to changes in eGFR. ANCOVA was performed using a general linear model approach to determine the association between confounding factors and eGFR. In these analyses, eGFR was the dependent variable, the four categories according to baseline and a 3-year follow-up SUA were the fixed factors, and each confounding factor was added as covariates. The synergistic effect of baseline SUA category and changes in SUA on changes in eGFR was evaluated using a general linear model. A value of P < .05 was considered significant. SUA were significantly higher in the hyperuricemia group than in the normal SUA group, and HDL-C, the presence of antilipidemic medication, and eGFR were significantly lower in the hyperuricemia group.

| Baseline characteristics of participants categorized by baseline SUA
There was no inter-group difference regarding age, SBP, LDL-C, and HbA1c.

| Multivariate-adjusted relationship between baseline characteristics including changes in SUA and changes in eGFR categorized by baseline SUA
Multiple regression analysis using changes in eGFR as an objective variable, adjusted for confounding factors as explanatory variables, showed that in the normal SUA group, changes in SUA as well as the presence of CVD, and HbA1c were significantly and independently associated with changes in eGFR (Table 3), and in the hyperuricemia group, changes in SUA as well age and eGFR were significantly associated with changes in eGFR. While in total group, both SUA and changes in SUA as well as age, HbA1c, and eGFR were significantly and negatively associated with changes in eGFR.

| Comparison between baseline and the 3-year follow-up eGFR categorized by changes in SUA
Furthermore, participants were divided into four groups for further analysis in Table 4. Eight hundred and fifty-one participants had normal SUA throughout the follow-up period (4.8 ± 1.0 mg/ dL and 4.8 ± 1.0 mg/dL, respectively, defined as N-N group); 77 participants had hyperuricemia at baseline and normal SUA later In subjects with normal renal function, an elevated SUA has almost uniformly been found to independently predict the development of CKD. 19 Obermayr et al 20    Third, confounding factors and eGFR are based on a single assessment of blood, which may introduce a misclassification bias. Fourth, we could not eliminate the possible effects of underlying diseases and medications for hypertension, diabetes, and dyslipidemia on the present findings. Therefore, the demographics and referral source may limit generalizability.
In conclusion, this study showed that both baseline and increased changes in SUA contributed to eGFR decline, independent of baseline eGFR. The underlying mechanism behind this relationship is unknown, and these factors seem to be independent of confounding factors, such as age, gender, drinking status, smoking status, blood pressure, lipids, HbA1c, and medication. Further prospective population-based studies are needed to investigate the changes in SUA metabolism and eGFR by lifestyle interventions.

ACK N OWLED G M ENT
This work was supported in part by a grant-in-aid from the Foundation for Development of Community (2019).

CO N FLI C T O F I NTE R E S T
The authors declare that they have no competing interests.

AUTH O R CO NTR I B UTI O N S
RK participated in the design of the study, performed the statistical analysis, and drafted the article. RK DN, TA, AK, and TK contributed to the acquisition and interpretation of data. RK, DN, and TK contributed to the conception and design of the statistical analysis. All authors read and approved the article.

E TH I C A L A PPROVA L
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee at which the studies conducted (IRB Approval Number: 1903018).

I N FO R M ED CO N S ENT
We obtained consent through opt-out procedure from all individual participants included in the study.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets analyzed in this study are available from the corresponding author (Ryuichi Kawamoto, rykawamo@m.ehime-u.ac.jp) on reasonable request.

Ryuichi Kawamoto
https://orcid.org/0000-0003-4674-0787 F I G U R E 2 Multivariate-adjusted 3-y follow-up eGFR categorized by changes in SUA. N-N group participants with normal SUA throughout the follow-up period, N-H group those with normal SUA at baseline and hyperuricemia 3 y later, H-N group those with hyperuricemia at baseline and normal SUA 3 y later, and H-H group those with hyperuricemia throughout the follow-up period