Abstract
The causal temporal association between serum uric acid (SUA) and muscle strength among elderly adults in China remains unclear. This study aimed to investigate both cross-sectional and longitudinal associations between SUA and muscle strength. Cross-sectional and longitudinal analyses were conducted using data from the China Health and Retirement Longitudinal Study (CHARLS). The cross-sectional cohort included 8706 participants from the 2015 wave. Muscle strength was assessed using grip strength and chair rising time. Generalized linear models evaluated associations between SUA and muscle strength. The longitudinal cohort included 3,404 participants from the 2011 wave, followed up in 2015. Cross-lagged panel models were applied to assess causal relationships. Cross-sectionally, SUA was positively associated with grip strength (β = 0.247, 95% CI 0.107–0.387) and negatively with chair rising time (β = − 0.166, 95% CI − 0.228 to − 0.105). No significant associations were found in participants with hyperuricemia. Longitudinally, baseline SUA predicted future grip strength (β = 0.038, 95% CI 0.006–0.072) and chair rising time (β = − 0.063, 95% CI − 0.093 to − 0.032), while baseline muscle strength did not predict future SUA. Higher SUA within the normal range is associated with better muscle strength in elderly adults, but this association is not observed in those with hyperuricemia.
Supplementary Information
The online version contains supplementary material available at 10.1038/s41598-025-29832-3.
Keywords: Aging, Chair rising time, Cross-lagged panel model, Grip strength serum uric acid
Subject terms: Diseases, Health care, Medical research, Rheumatology, Risk factors
Introduction
As the aging process intensifies, muscle strength begins to decline1. Low muscle strength has a significant negative impact on the quality of life and is linked to higher risks of falls, hospitalizations, morbidity, and mortality in older adults2–4. Some researchers have suggested that poor muscle strength in older adults is associated with higher oxidative stress5. Excess reactive oxygen and nitrogen species can affect muscle size, fiber activation, and excitation–contraction coupling6. Serum uric acid (SUA), the end product of purine metabolism7, is a compound that has evolved to sustain higher concentrations, conferring particular biological advantages8. SUA possesses important antioxidant properties, contributing to two-thirds of the total plasma antioxidant capacity9,10. Consequently, some scholars propose that SUA’s antioxidant properties may help prevent the decline in muscle strength11. However, numerous epidemiological and clinical studies have consistently demonstrated that hyperuricemia (HUA) significantly increases the risk of metabolic diseases such as gout and diabetes, which can result in declining muscle strength12,13. Given the growing proportion of the aging population, a comprehensive investigation of the potential effects of SUA on age-related muscle strength decline is imperative.
Some studies conducted in various countries, such as the United States1, Korea14, and Japan15, have shown a positive correlation between SUA levels and muscle strength in older adults. However, there remains a significant gap in nationwide, large-scale investigations exploring this association among middle-aged and elderly individuals in China. Given the diverse lifestyle, dietary patterns, and genetic factors in China, conducting research in this demographic could offer valuable insights into the relationship between SUA and muscle strength. With China’s rapidly aging population and increasing prevalence of related health conditions, it presents an opportune and relevant setting for such research. Furthermore, these previous studies predominantly focused on handgrip strength, which may reflect the impact of SUA on upper limb strength but may not comprehensively represent its effect on other body parts. Evaluating lower body strength is also critical for assessing the functional performance of older adults and could provide more reliable associations16–18. Chair rising time, also known as the Five Times Sit to Stand Test, is a valuable measure for assessing lower limb muscle strength, balance, and mobility in the elderly19. Therefore, we propose a combined evaluation of handgrip strength and chair rising time to provide a more comprehensive assessment of muscle strength. Despite recent advancements in cross-sectional studies and related hypotheses, there remains a dearth of large-scale longitudinal studies. Such studies are essential for establishing a causal link between SUA and muscle strength. This research aims to fill this gap by exploring both cross-sectional and longitudinal associations between SUA and muscle strength using a nationally representative dataset.
To bridge this gap in knowledge, our study utilized data from the China Health and Retirement Longitudinal Study (CHARLS), a comprehensive survey conducted among the elderly population in 150 counties across 28 provinces of mainland China. Our objective was to investigate the cross-sectional and longitudinal association between SUA levels and muscle strength. We hypothesized that elevated SUA levels are positively associated with increased muscle strength in this population.
Method
Study population
The CHARLS, initiated in 2011, is a nationally representative longitudinal survey targeting Chinese adults aged 45 years or older. The study employed a stratified four-stage cluster random sampling method to select participants from 150 counties across 28 mainland Chinese provinces. Participants engage in face-to-face interviews using structured questionnaires, capturing data on sociodemographic characteristics, lifestyle factors, and health-related information. Following the baseline survey, all participants are re-evaluated biennially. To date, CHARLS has completed four waves of data collection: wave 1 (2011–2012), wave 2 (2013), wave 3 (2015), and wave 4 (2018). Among these, venous blood samples were obtained and analyzed for complete blood count (CBC) and blood biochemistry in waves 1 and 3. Additionally, grip strength and chair rising time were assessed in waves 1, 2, and 3. For our analysis, we selected data from waves 1 and 3 due to the availability of both blood-based examinations and physical assessments.
In this study, we constructed two cohorts. The cross-sectional cohort, derived from wave 3, was used to investigate the association between SUA and muscle strength, because it provided more recent and updated information than wave 1. The exclusion criteria for this cohort were: (1) missing muscle strength data, (2) missing SUA data, (3) age below 45 years or missing age data, (4) cancer diagnosis or indeterminate cancer status, and (5) missing data on sex, BMI, or other covariates. After applying these criteria, a total of 8,706 participants were included in the cross-sectional analysis. The second cohort, designed to examine longitudinal relationships between SUA and muscle strength, comprised respondents who participated in both waves 1 and 3. The exclusion criteria were identical to those for the first cohort. Detailed eligibility and participant selection procedures for both cohorts are illustrated in Fig. S1.
This study is a secondary analysis of anonymized, publicly available data from CHARLS. The original CHARLS fieldwork obtained ethical approval from the Ethical Review Committee of Peking University (approval number: IRB00001052-11015), and all participants provided informed consent. Our analysis used only de-identified data in accordance with the CHARLS data-use agreement. No additional ethical approval was sought for this specific analysis. All procedures performed in studies involving human participants adhered to the ethical standards of the institutional and/or national research committee, in accordance with the 1964 Helsinki declaration and its subsequent amendments or comparable ethical standards. In addition, this study was designed and reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines(Table S1).
Assessment of SUA
The blood sample collection, processing, transportation, and storage procedures were described in detail on the CHARLS website (https://charls.charlsdata.com). In brief, local CDC staff obtained fasting venous blood samples and delivered them to the local CDC and Beijing CDC laboratories under low-temperature storage conditions for analysis. The UA Plus method was used to measure the SUA level. We defined HUA as SUA ≥ 7.0 mg/dL in this study.
Assessment of muscle strength
The assessment of muscle strength encompassed both upper and lower limbs20. Handgrip strength, a reliable indicator of upper limb strength, was measured using a Yuejian™ WL-1000 dynamometer. Participants were instructed to squeeze the dynamometer with their dominant hand as forcefully as possible, with the highest value recorded from two attempts21. For evaluating lower limb strength, we employed the chair rising time test. Specifically, the Five Times Sit to Stand Test was utilized, in which participants were guided to sit down and stand up from a chair five times as quickly as possible. The total time taken to complete these repetitions was recorded using a stopwatch22.
Covariates
Some covariates were selected in our study based on previously published research23,24. Sociodemographic variables included age (45–54, 55–64, and ≥ 65), sex (male and female), education level (illiterate, junior high school or below, and high school or above), marital status[married/cohabitated, other(separated/divorced/widowed/never married)], and residence(rural and urban). Health behaviors (the history of smoking and drinking) and medical history [including arthritis, cardiovascular diseases, pulmonary diseases, metabolic diseases, and other disease (digestive diseases, kidney diseases, emotional, nervous, or psychiatric problems, memory-related disease)] were collected by well-trained interviewers. Anthropometric parameters, including body weight, height, waist circumference (WC), grip strength, and chair rising time, were measured based on standard protocols. BMI (kg/m2) was calculated as body weight/height2. Blood test data included white blood cell counts (WBC), hemoglobin (Hb), platelet counts (PLT), total cholesterol (TC), triglyceride (TG), high density lipoprotein-cholesterol (HDL-c), low-density lipoprotein-cholesterol (LDL-c), uric acid, blood glucose (BG), high sensitivity C-reactive protein (hsCRP), and cystatin C (Cys C).
Statistical analysis
In the cross-sectional analysis, generalized linear models (GLM) were employed to assess the association between SUA levels and muscle strength, including grip strength and chair-rising time. Three models were constructed. Model 1 was unadjusted. Model 2 adjusted for demographic characteristics (age, sex, BMI, educational level, residence, smoking, drinking, and waist circumference) and blood parameters (WBC, Hb, PLT, BG, TC, TG, HDL-C, LDL-C, and Cys C). Model 3 further included chronic diseases (arthritis, respiratory, cardiovascular, metabolic, and other chronic conditions).
Exploratory subgroup analyses were conducted to evaluate potential effect modification of the association between SUA and muscle strength. Interaction analyses were performed to investigate whether the association between SUA and muscle strength varied according to sex, age, or other covariates. Coefficients (β), 95% confidence intervals (CI), and p-values were reported for all models.
In longitudinal analysis, the cross-lagged panel models were used to explore the causal temporal relationships of SUA with grip strength and chair rising time. The study’s general modeling strategy was shown in Fig. 1A, including 5 paths and their corresponding coefficients. Cross-lagged paths concurrently include 2 cross-lagged coefficients, βCL-1 and βCL-2. The coefficient βCL-1 represents the effect of SUA at T1 (time 1) on muscle strength at T2 (time 2), and βCL-2 indicates the regression of muscle strengths at T1 on SUA at T2. The temporal sequence was determined by comparing the cross-lagged coefficients. The coefficient βCS of cross-sectional paths were also calculated, which represents the baseline correlation between SUA and muscle status. In addition, two autoregressive coefficients, βAR-SUA and βAR-muscle, were obtained from the paths between T1 and T2 for SUA and muscle strength, respectively. These coefficients accounted for within-person stability in each measurement from the baseline to the subsequent point of time25.
Fig. 1.
Cross-lagged panel models applied to assess SUA and muscle strength. (A) The general modeling strategy used for the cross-lagged panel models. (B) The cross-lagged panel model of SUA and grip strength. (C) The cross-lagged panel model of SUA and chair rising time. GS: grip strength, CRT: chair rising time, AR: autoregressive; CL: cross-lagged; CS: cross sectional. All models were adjusted for age, sex, marital history, education level, WC, BMI, smoking history, and drinking history, WBC, Hb, PLT, TG, TC, HDL-c, LDL-c, BG, Cys-c, arthritis, respiratory diseases, cardiovascular diseases, metabolic diseases and other diseases. *P < 0.05.
Cross-lagged panel models were conducted using SUA, grip strength, and chair-rising time. Confounding factors were controlled for using residual analysis to construct regression equations for study variables with sex, age, BMI, and other covariates. The resulting residuals were normalized by Z conversion (mean = 0; standard deviation = 1) and included in the cross-lagged panel models. Model fit was assessed using the root mean square error of approximation (RMSEA) and the comparative fit index (CFI), with good fit indicated by RMSEA < 0.06 and CFI > 0.90.
GLM analysis was performed using IBM SPSS Statistics version 27.0, while cross-lagged panel models were conducted using IBM SPSS Amos version 28.0. Statistical significance was set at p < 0.05 (two-sided).
We conducted a series of sensitivity analyses to examine the robustness of our findings. In the cross-sectional analyses, for models where the assumptions of linear regression were not satisfied, we would apply robust regression based on MM-estimation. Participants with missing key variables such as muscle strength and SUA were excluded, and multiple imputation by chained equations (MICE) was then performed for other missing covariates, with additional adjustment for kidney disease and cognitive function. In the longitudinal analyses, missing continuous variables were imputed using nearest neighbor averaging, while missing categorical variables were coded as “unknown” prior to cross-lagged panel model analysis. Additionally, analyses were conducted separately for participants with and without HUA.
Results
Characteristics of participants in the cross-sectional and longitudinal study
The cross-sectional cohort comprised 8,706 participants, with 45.4% being men, and the group had a mean age of 61.06 ± 9.04 years, thus representing a broad spectrum of age groups. The baseline characteristics of both included and excluded participants due to missing data are presented in Table S2. The proportion of missing data for each variable is presented in Table S3. The longitudinal cohort consisted of 3,404 participants, with a mean age of 59.20 ± 8.85 years, and 47.2% of the participants were men at wave 1. The distribution of baseline SUA, grip strength, chair-rising time, and baseline covariates is provided in Table 1. We also summarized baseline characteristics according to quartiles of SUA (Table S4–6). Across the SUA quartiles, participants showed clear demographic and lifestyle differences. Higher SUA levels were associated with older age, a markedly higher proportion of men, and higher educational attainment (all P < 0.001). Participants with higher SUA were also more likely to smoke and drink alcohol. In addition, the prevalence of cardiovascular, pulmonary, and metabolic diseases was higher among those with elevated SUA.
Table 1.
Baseline characteristics of participants in the cross-sectional and longitudinal cohort.
| The cross-sectional cohort | The longitudinal cohort | ||
|---|---|---|---|
| No. of participant | Wave 3 (N = 8706) |
Wave 1 (N = 3404) |
Wave 3 (N = 3404) |
| Age, n (%) | |||
| 45–54 | 2414(27.7) | 1130(33.2) | 659(19.4) |
| 55–64 | 3338(38.3) | 1329(39.0) | 1324(38.9) |
| ≥ 65 | 2954(33.9) | 945(27.8) | 1421(41.7) |
| Male, n (%) | 4115(47.3) | 1606(47.2) | 1606(47.2) |
| Married, n (%) | 8570(98.4) | 3354(98.5) | 3364(98.8) |
| Educational level, n (%) | |||
| Illiterate | 2193(25.2) | 956(28.1) | 956(28.1) |
| Junior high school or below | 5555(63.8) | 2141(62.9) | 2141(62.0) |
| High school or above | 958(11.0) | 307(9.0) | 307(9.0) |
| Residence, n (%) | |||
| Rural | 6977(80.1) | 2884(84.7) | 2847(83.6) |
| Urban | 1729(19.9) | 520(15.3) | 557(16.4) |
| Smoking history, n (%) | |||
| Yes | 3855(44.3) | 1349(39.6) | 1498(44.0) |
| NO | 4851(55.7) | 2055(60.4) | 1906(56.0) |
| Drinking history, n (%) | |||
| Yes | 4030(46.3) | 1332(39.1) | 1525(44.8) |
| NO | 4676(53.7) | 2071(60.9) | 1879(55.2) |
| BMI, mean (SD) | 24.62(17.74) | 23.53(3.90) | 24.38(18.33) |
| WC, mean (SD) | 85.51(12.92) | 84.19(12.65) | 84.93(13.57) |
| Blood, mean (SD), (mg/dL) | |||
| WBC | 5.97(1.80) | 6.21(1.87) | 5.94(1.80) |
| Hb | 13.71(1.89) | 14.40(2.13) | 13.64(1.82) |
| PLT | 203.85(75.54) | 211.85(71.95) | 203.69(76.48) |
| TG | 142.92(90.74) | 130.11(90.49) | 142.25(88.8) |
| TC | 184.65(36.51) | 193.34(37.70) | 185.40(37.12) |
| HDL-c | 51.39(11.71) | 51.12(15.01) | 51.36(11.80) |
| LDL-c | 102.74(29.01) | 116.65(34.90) | 103.39(29.23) |
| BG | 103.49(35.05) | 109.30(33.28) | 103.22(33.80) |
| SUA | 4.95(1.40) | 4.40(1.22) | 4.95(1.40) |
| Cys-c | 0.86(0.24) | 1.00(0.23) | 0.88(0.22) |
| CRP | 2.64(5.69) | 2.34(5.80) | 2.68(6.06) |
| Chronic diseases, n (%) | |||
| Arthritis | 3691(42.4) | 1153(33.9) | 1566(46.0) |
| Cardiovascular diseases | 3680(42.3) | 1143(33.6) | 1536(45.1) |
| Pulmonary diseases | 1351(15.5) | 400(11.8) | 592(17.4) |
| Metabolic diseases | 2148(24.7) | 479(14.1) | 875(25.7) |
| Other diseases | 3501(40.2) | 986(29.0) | 1439(42.3) |
| Muscle strength, mean (SD) | |||
| Grip strength(kg) | 30.09(10.43) | 31.89(10.41) | 29.30(10.15) |
| Chair rising time(s) | 9.45(3.64) | 10.75(4.09) | 9.68(3.61) |
SD, standard deviation; BMI, body mass index; WC, waist circumference; WBC, white blood cell; Hb, hemoglobin; PLT, platelet; BG, blood glucose; TC, total cholesterol; TG, triglyceride; HDL-c, high-density lipoprotein-cholesterol; LDL-c, low density lipoprotein-cholesterol; SUA, serum uric acid; Cys-c, cystatin c.
The cross-sectional association between SUA and muscle status
We observed that higher SUA levels were associated with a lower risk of poor muscle strength across three adjusted models (Table 2). Upon adjusting for all covariates in model 3, higher SUA levels exhibited a significant positive association with grip strength (β = 0.247, 95%CI 0.107–0.387) and a negative association with chair rising time (β = − 0.166, 95%CI − 0.228 to − 0.105). Additionally, SUA was categorized into quartiles. Compared to the first quartile (Q1), individuals in the third (Q3) and fourth (Q4) quartiles demonstrated stronger grip strength and shorter chair rising time. We also performed diagnostic analyses to assess the normality and homoscedasticity of residuals (Figure S2–S3). In models with chair rising time as the dependent variable, we observed potential heteroscedasticity and non-normality of residuals; therefore, robust regression was applied in the sensitivity analyses.
Table 2.
Associations between SUA and muscle strength by generalized linear models.
| Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|
| β (95%CI) | P | β (95%CI) | P | β (95%CI) | P | |
| By continuous, + 1 mg/dl | ||||||
| Grip strength | 1.768(1.6161.921) | < 0.001 | 0.242(0.102–0.383) | < 0.001 | 0.247(0.107–0.387) | < 0.001 |
| Chair rising time | − 0.126(− 0.181 to − 0.072) | < 0.001 | − 0.156(− 0.218 to − 0.095) | < 0.001 | − 0.166(-0.228 to − 0.105) | < 0.001 |
| By quartile | ||||||
| Grip strength | ||||||
| Q1 | Ref | Ref | Ref | |||
| Q2 | 2.470(1.876–3.063) | < 0.001 | 0.487(0.022–0.952) | 0.040 | 0.517(0.054–0.979) | 0.029 |
| Q3 | 4.612(4.017 -5.207) | < 0.001 | 0.669(0.183–1.155) | 0.007 | 0.705(0.221–1.188) | 0.004 |
| Q4 | 6.569(5.979- 7.159) | < 0.001 | 0.779(0.249- 1.309) | 0.004 | 0.800(0.272–1.328) | 0.003 |
| P for trend | < 0.001 | 0.003 | 0.002 | |||
| Chair rising time | ||||||
| Q1 | Ref | Ref | Ref | |||
| Q2 | − 0.162(− 0.374– 0.051) | 0.136 | − 0.136(− 0.340–0.067) | 0.189 | − 0.173(− 0.376 − 0.030) | 0.095 |
| Q3 | − 0.465(− 0.679 to − 0.252) | < 0.001 | − 0.416(− 0.629 to − 0.204) | < 0.001 | − 0.439(-0.651 to − 0.228) | < 0.001 |
| Q4 | − 0.455(− 0.667 to − 0.244) | < 0.001 | − 0.511(− 0.743 to − 0.279) | < 0.001 | − 0.549(− 0.780 to − 0.318) | < 0.001 |
| P for trend | < 0.001 | < 0.001 | < 0.001 | |||
β and (95%CI) are presented. Model 1: unadjusted model, Model 2: adjusted for age, sex, marital history, education level, residence, WC, BMI, smoking history, and drinking history, WBC, Hb, PLT, TG, TC, HDL-c, LDL-c, BG, Cysc and CRP. Model 3: additionally adjusted arthritis, respiratory diseases, cardiovascular diseases, metabolic diseases and other diseases. SUA, serum uric acid; ref., reference.
Subgroup analysis in the cross-sectional cohort
The majority of subgroup analysis findings aligned with the main analysis results, particularly regarding chair rising time, as depicted in Fig. 2. Regarding grip strength, significant interaction with SUA levels was only evident across age subgroups (p = 0.016). The positive association between SUA and grip strength varied among different age groups. Specifically, the 55–64 age group displayed the strongest association (β = 0.400, 95%CI 0.176–0.624), followed by the group aged 65 and above (β = 0.225, 95%CI − 0.004–0.453), while no association was observed in the 45–54 age group (β = − 0.056, 95%CI − 0.343–0.232). Conversely, no interaction effects were noted between the variables and SUA for chair rising time.
Fig. 2.
The subgroup analysis in cross-sectional cohort. Graph shows β and 95%CIs for (A) grip strength, (B) chair rising time adjusted for age, sex, marital history, education level, WC, BMI, smoking history, and drinking history, WBC, Hb, PLT, TG, TC, HDL-c, LDL-c, BG, Cys-c, arthritis, respiratory diseases, cardiovascular diseases, metabolic diseases and other diseases.
The longitudinal association between SUA and muscle strength
The cross-lagged path analysis of SUA and grip strength, controlling for confounding factors, is depicted in Fig. 1B–C. Good model fit was indicated by fitting indicators (CFI = 0.952 and SRMR = 0.034). In the autoregressive model, baseline SUA positively correlated with SUA in 2015 (βAR = 0.563, 95%CI 0.527–0.596), while baseline grip strength was positively associated with grip strength in 2015 (βAR = 0.307, 95%CI 0.271–0.347). These positive associations supported the autoregressive cross-lagged design hypothesis. The cross-lagged path from baseline SUA to grip strength in 2015 (βCL = 0.038, 95%CI 0.006–0.072) was significant. However, the path from baseline grip strength to SUA in 2015 (βCL = 0.015, 95%CI -0.014–0.042) was not significant, indicating that baseline SUA could prospectively predict grip strength at further time points after controlling all synchronous relations and stability coefficients, while baseline grip strength could not predict SUA levels at further time points (Table 3). A similar unidirectional association was found between SUA and chair rising time. Baseline SUA was significantly associated with chair rising time (βCL = − 0.063, 95%CI − 0.093 to − 0.032), whereas baseline chair rising time was not related to SUA (βCL = − 0.028, 95%CI − 0.058 to 0.003) (Table 3).
Table 3.
The cross-lagged panel model.
| Path | Grip strength | Chair rising time | ||
|---|---|---|---|---|
| β(95%CI)b | P-value | β(95%CI)b | P-value | |
| Autoregressive paths | ||||
| SUA_T1 → SUA_T2 | 0.563(0.527–0.596) | < 0.001 | 0.561(0.525–0.595) | < 0.001 |
| Strengtha_T1 → Strengtha_T2 | 0.307(0.271–0.347) | < 0.001 | 0.305(0.252–0.362) | < 0.001 |
| Correlations | ||||
| SUA_T1 with Strengtha_T1 | 0.025(− 0.008–0.059) | 0.140 | − 0.063(− 0.097 to − 0.030) | < 0.001 |
| Cross-lagged paths | ||||
| UA_T1 → Strengtha_T2 | 0.038(0.006–0.072) | 0.019 | − 0.063(− 0.093 to − 0.032) | < 0.001 |
| Strengtha_T1 → UA_T2 | 0.015(− 0.014–0.042) | 0.277 | − 0.028(− 0.058 to 0.003) | 0.055 |
| Model fit | ||||
| CFI | 0.996 | 0.999 | ||
| TLI | 0.975 | 0.995 | ||
| GFI | 0.999 | 1.000 | ||
| IFI | 0.996 | 0.999 | ||
| RMR | 0.012 | 0.006 | ||
| RMSEA | 0.045 | 0.019 | ||
aStrength represented grip strength or chair rising time
b95%CI were calculated by bias-corrected percentile method
CFI, Comparative Fit Index; TLI, Tucker–Lewis Fit Index; GFI, Goodness of Fit Index; IFI, Incremental Fit Index; RMR, Root Mean square Residual (RMR); RMSEA, Root Mean Square Error of Approximation
Sensitivity analyses
We reanalyzed chair rising time using robust regression based on MM-estimation. The direction and interpretation of the estimates remained consistent with those from ordinary linear regression (Table S7). After imputing missing data and adjusting for kidney disease and cognitive function, results from both the cross-sectional analysis (n = 11,982) and the longitudinal analysis (n = 5,839) were consistent with the main findings (Table S8–9). In the imputed dataset, robust regression of chair rising time produced results consistent with linear regression (Table S10). Additionally, analyses were conducted according to HUA status. The results indicated that, in the presence of HUA, SUA showed no clear association with muscle strength in either the crude model or after adjusting for confounders (Fig. 2).
Discussion
To the best of our knowledge, this study represents the first examination of the associations between SUA and muscle strength among middle-aged and older Chinese individuals, incorporating both cross-sectional and longitudinal perspectives. In our cross-sectional study, we observed a positive correlation between SUA levels and muscle strength in middle-aged and older adults from China, as indicated by both strength tests. These findings are consistent with prior research conducted on older adults using the NHANES database. Nahas et al. reported a positive association between UA and muscle strength in older adults from NHANES (1999–2002) 26. Additionally, García-Esquinas and Rodríguez-Artalejo observed a positive association between UA and handgrip strength in older adults from NHANES (2011–2012) 1. Subgroup analysis revealed that the association between SUA and muscle strength remained relatively stable across different genders, BMI categories, educational levels, residential areas, and among patients with chronic diseases such as arthritis.
Of particular interest was our observation that higher SUA levels exhibited a stronger positive correlation with muscle strength compared to lower levels when SUA levels were categorized into quartiles. Furthermore, our study explored the association between HUA and grip strength. We found that while higher SUA levels within the normal range had a beneficial effect on muscle strength, this positive effect diminished as SUA levels increased and HUA occurred. Our findings suggest that the optimal beneficial role of SUA on muscle strength may be limited to a specific range.
In longitudinal study, we used cross-lagged panel models, a practical statistical approach to assess directional, temporal relationships and allow evaluation of potential causal effects between variables over time27,28. In contrast, linear mixed-effects models estimate only overall associations and cannot disentangle the temporal ordering needed to address the “chicken-and-egg” question between SUA and muscle strength29. Our results revealed a unidirectional relationship between SUA and muscle strength, which remained relatively stable over time. However, the mechanisms underlying the observed stronger muscle strength in middle-aged and older individuals with optimal SUA levels remain unclear. The prevailing theory suggests that SUA acts as a potent antioxidant, scavenging singlet oxygen and free radicals in the body, thereby protecting skeletal muscle function from oxidative damage induced by reactive oxygen species (ROS) 5,30. Clinical trials in previous studies have demonstrated that increasing SUA concentration in vivo can enhance serum antioxidant capacity and reduce oxidative stress during acute physical exercise in healthy subjects 31. However, it is important to note that when SUA levels exceed 7 mg/dl, the risk of urate crystal formation/precipitation increases, potentially triggering inflammatory reactions through the release of pro-inflammatory mediators32. These reactions may negate the antioxidant effect exerted by SUA. Supporting this notion, Rafaela Nehme et al. found a positive association between SUA and muscle strength in older adults without a diagnosis of gout, while this association was not significant in adults with gout11. Another hypothesis posits that under conditions of oxidative stress; abnormally high concentrations of SUA may switch from exerting an antioxidant effect to a pro-oxidant effect. This shift in function could potentially negate the protective effect of SUA on muscle strength or even exacerbate muscle strength deterioration13,33. However, further in vivo and in vitro experiments are needed to elucidate the specific mechanisms underlying these hypotheses.
Finally, we are pleased to offer some suggestions to help middle-aged and elderly community members maintain their SUA levels in the higher normal range. we encourage dietary adjustments that include foods rich in purines, such as lean meats, seafood, and legumes, balanced with adequate hydration34. Promote regular moderate-intensity physical activities like walking or swimming. Implement routine health screenings to monitor SUA levels and provide personalized advice. Organize educational workshops on the benefits of optimal SUA levels for muscle health. Establish partnerships with local healthcare providers to create tailored health plans. By adopting these strategies, communities can enhance the muscle strength and overall well-being of their middle-aged and elderly members.
Although this study benefits from a large prospective cohort and the use of cross-lagged panel models to examine temporal relationships, several limitations should be noted. CHARLS lacks detailed data on dietary intake, long-term medication history, physical activity, and nutritional status (e.g., serum albumin), preventing adjustment for these potential confounders. Selection bias (e.g., loss to follow-up or exclusion due to missing data) and information bias (e.g., measurement error or misclassification) may also affect the results. In addition, multiple comparisons raise the possibility of chance findings (type I error). Finally, while cross-lagged models are useful for assessing temporal associations, they cannot fully rule out reverse causality—for example, healthier individuals may have higher SUA due to dietary habits, which could influence muscle strength. These limitations warrant cautious interpretation and replication in cohorts with more comprehensive covariate data.
In conclusion, our study provides evidence that SUA is positively associated with muscle strength and can influence muscle strength over time. However, this positive association diminishes when HUA occurs. Therefore, monitoring and sustaining appropriate SUA levels may offer valuable insights into protecting against muscle strength loss in middle-aged and older adults in community-dwelling populations.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We would like to thank the CHARLS research team, the field team, and every respondent for their time and efforts that they have devoted to the CHARLS project.
Author contributions
P.H. conceived and designed the study, and performed the investigation; Q.W. conducted the formal analysis; L.C. provided resources and supervised the study; H.L. and Q.W. curated the data; H.Y. wrote the main manuscript text; H.Y. and L.C. reviewed and edited the manuscript. All authors reviewed the manuscript.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Data availability
The datasets generated for this study are available on request to the corresponding author and with permission of CHARLS.
Declarations
Ethics approval
This study used anonymized, publicly available CHARLS data, which were collected with prior ethical approval (Peking University, IRB00001052-11015). No additional ethics approval was required for this secondary analysis.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.García-Esquinas, E. & Rodríguez-Artalejo, F. Association between serum uric acid concentrations and grip strength: Is there effect modification by age?. Clin. Nutr.37(2), 566–572. 10.1016/j.clnu.2017.01.008 (2018). [DOI] [PubMed] [Google Scholar]
- 2.Wu, M. et al. Associations of muscle mass, strength, and quality with all-cause mortality in China: A population-based cohort study. Chin. Med. J.135(11), 1358–1368. 10.1097/cm9.0000000000002193 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Rantanen, T. et al. Muscle strength and body mass index as long-term predictors of mortality in initially healthy men. J. Gerontol. A Biol. Sci. Med. Sci.55(3), M168–M173. 10.1093/gerona/55.3.m168 (2000). [DOI] [PubMed] [Google Scholar]
- 4.Wang, D. X. M. et al. Muscle mass, strength, and physical performance predicting activities of daily living: A meta-analysis. J. Cachexia Sarcopenia Muscle11(1), 3–25. 10.1002/jcsm.12502 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Howard, C. et al. Oxidative protein damage is associated with poor grip strength among older women living in the community. J. Appl. Physiol.103(1), 17–20. 10.1152/japplphysiol.00133.2007 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Fulle, S. et al. The contribution of reactive oxygen species to sarcopenia and muscle ageing. Exp. Gerontol.39(1), 17–24. 10.1016/j.exger.2003.09.012 (2004). [DOI] [PubMed] [Google Scholar]
- 7.Kanellis, J. & Kang, D. H. Uric acid as a mediator of endothelial dysfunction, inflammation, and vascular disease. Semin. Nephrol.25(1), 39–42. 10.1016/j.semnephrol.2004.09.007 (2005). [DOI] [PubMed] [Google Scholar]
- 8.Wu, Y. et al. Association of serum uric acid level with muscle strength and cognitive function among Chinese aged 50–74 years. Geriatr. Gerontol. Int.13(3), 672–677. 10.1111/j.1447-0594.2012.00962.x (2013). [DOI] [PubMed] [Google Scholar]
- 9.Ames, B. N. et al. Uric acid provides an antioxidant defense in humans against oxidant- and radical-caused aging and cancer: A hypothesis. Proc. Natl. Acad. Sci. USA78(11), 6858–6862. 10.1073/pnas.78.11.6858 (1981). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.De Oliveira, E. P. & Burini, R. C. High plasma uric acid concentration: Causes and consequences. Diabetol. Metab. Syndr.4, 12. 10.1186/1758-5996-4-12 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Nehme, R., Nahas, P. C. & De Oliveira, E. P. Association between serum uric acid and muscle strength in older adults with or without gout diagnosis: NHANES 2011–2014. Aging Clin. Exp. Res.35(9), 1967–1972. 10.1007/s40520-023-02462-3 (2023). [DOI] [PubMed] [Google Scholar]
- 12.Veronese, N. et al. Results of an observational cohort study of hyperuricemia as a predictor of poor physical performance in the elderly. Arthritis Care Res.69(8), 1238–1244. 10.1002/acr.23118 (2017). [DOI] [PubMed] [Google Scholar]
- 13.Huang, C. et al. An inverted J-shaped association of serum uric acid with muscle strength among Japanese adult men: A cross-sectional study. BMC Musculoskelet. Disord.14, 258. 10.1186/1471-2474-14-258 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lee, J. et al. High serum uric acid level is associated with greater handgrip strength in the aged population. Arthritis Res. Ther.21(1), 73. 10.1186/s13075-019-1858-2 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Kuriyama, S. et al. Relationship between serum uric acid levels and muscle strength/volume: A new insight from a large-scale survey. Nihon. Jinzo Gakkai Shi.56(8), 1260–1269 (2014). [PubMed] [Google Scholar]
- 16.Alcazar, J. et al. The sit-to-stand muscle power test: An easy, inexpensive and portable procedure to assess muscle power in older people. Exp. Gerontol.112, 38–43. 10.1016/j.exger.2018.08.006 (2018). [DOI] [PubMed] [Google Scholar]
- 17.Mentiplay, B. F. et al. Five times sit-to-stand following stroke: Relationship with strength and balance. Gait. Posture78, 35–39. 10.1016/j.gaitpost.2020.03.005 (2020). [DOI] [PubMed] [Google Scholar]
- 18.Bohannon, R. W. et al. Sit-to-stand test: Performance and determinants across the age-span. Isokinet Exerc. Sci.18(4), 235–240. 10.3233/ies-2010-0389 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Meretta, B. M. et al. The five times sit to stand test: Responsiveness to change and concurrent validity in adults undergoing vestibular rehabilitation. J. Vestib. Res.16(4–5), 233–243 (2006). [PubMed] [Google Scholar]
- 20.Chen, L. K. et al. Asian working group for sarcopenia: 2019 consensus update on sarcopenia diagnosis and treatment. J. Am. Med. Dir. Assoc.21(3), 300-307.e2. 10.1016/j.jamda.2019.12.012 (2020). [DOI] [PubMed] [Google Scholar]
- 21.Zhao, Y. et al. Cohort profile: The China health and retirement longitudinal study (CHARLS). Int. J. Epidemiol.43(1), 61–68. 10.1093/ije/dys203 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Qiu, S. et al. Muscle strength and prediabetes progression and regression in middle-aged and older adults: A prospective cohort study. J. Cachexia Sarcopenia Muscle13(2), 909–918. 10.1002/jcsm.12905 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Liu J, Liu W, Wang L, et al. Association of Visceral Adiposity Index and Handgrip Strength with Cardiometabolic Multimorbidity among Middle-Aged and Older Adults: Findings from Charls 2011–2020. Nutrients, 2024, 16(14), 10.3390/nu16142277. [DOI] [PMC free article] [PubMed]
- 24.Luo, J. H. et al. Association between relative muscle strength and hypertension in middle-aged and older Chinese adults. BMC Public Health23(1), 2087. 10.1186/s12889-023-17007-6 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Li, H. et al. The bidirectional association between cognitive function and gait speed in Chinese older adults: Longitudinal observational study. JMIR Public Health Surveill9, e44274. 10.2196/44274 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Nahas, P. C. et al. Serum uric acid is positively associated with muscle strength in older men and women: Findings from NHANES 1999–2002. Clin. Nutr.40(6), 4386–4393. 10.1016/j.clnu.2020.12.043 (2021). [DOI] [PubMed] [Google Scholar]
- 27.Zheng, J., Jiang, M. & Xie, Y. Influence of uric acid on the correlation between waist circumference and triglyceride glucose index: an analysis from CHARLS. Lipids Health Dis20(1), 42. 10.1186/s12944-021-01474-0 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Gauld, C., Grasman, R. & Bailly, S. Usefulness of cross-lagged panel models for clinical research. Chest167(6), 1537–1540. 10.1016/j.chest.2024.12.013 (2025). [DOI] [PubMed] [Google Scholar]
- 29.Laird, N. M. & Ware, J. H. Random-effects models for longitudinal data. Biometrics38(4), 963–974 (1982). [PubMed] [Google Scholar]
- 30.Kawamoto, R. et al. Serum uric acid is positively associated with handgrip strength among Japanese community-dwelling elderly women. PLoS ONE11(4), e0151044. 10.1371/journal.pone.0151044 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Waring, W. S. et al. Uric acid reduces exercise-induced oxidative stress in healthy adults. Clin. Sci.105(4), 425–430. 10.1042/cs20030149 (2003). [DOI] [PubMed] [Google Scholar]
- 32.Richette P, Bardin T. Gout. Lancet, 2010, 375(9711): 318–28, 10.1016/s0140-6736(09)60883-7. [DOI] [PubMed]
- 33.Hayden, M. R. & Tyagi, S. C. Uric acid: A new look at an old risk marker for cardiovascular disease, metabolic syndrome, and type 2 diabetes mellitus: The urate redox shuttle. Nutr. Metab.1(1), 10. 10.1186/1743-7075-1-10 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Schlesinger, N. Dietary factors and hyperuricaemia. Curr. Pharm. Des.11(32), 4133–4138. 10.2174/138161205774913273 (2005). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated for this study are available on request to the corresponding author and with permission of CHARLS.