DOI: https://doi.org/10.21203/rs.3.rs-1977307/v2
Background
We aimed to investigate the interaction between serum uric acid levels with estimated glomerular filtration rate (eGFR) to low muscle strength (LMS) in a large Chinese elderly population.
Methods
Cohort data were obtained from China Health and Retirement Longitudinal Study (CHARLS) in 2011 and 2015. Two thousand seven hundred forty-five community-dwelling older participants were enrolled for the follow-up. Serum uric acid was collected after 8 hours of fasting, and handgrip strength was measured with a dynamometer. eGFR was calculated with an equation based on the Chinese population. A generalized additive model was employed for interaction analysis and progressively adjusted confounders.
Results
In this study, we found that men with a low eGFR (<60 mL/min/1.73 m2) reported higher SUA levels (5.91 ± 1.27) and older (72.53 ± 6.38) than those who had a high eGFR while women share the same difference with a lower eGFR in higher SUA levels (5.00 ± 1.34) and older (72.81 ± 6.83). After progressively adjusting covariates, in females, the OR for higher eGFR with higher SUA level remained significantly with low muscle strength (OR=0.80 95%CI=0.68-0.95 P=0.0102). This correlation, however, was not observed in men.
Conclusions
This population-based cohort study in Chinese revealed that high serum uric acid level with higher eGFR seems to be significantly associated with a lower risk of low muscle strength in the elderly, especially in females.
With advancing age, the human body is accompanied by a series of physiological changes, including the loss of skeletal muscle mass and strength, which is defined as sarcopenia [1]. Muscle strength, especially handgrip strength, is a crucial parameter for assessing and diagnosing sarcopenia [2]. The declined strength as a predicted factor is also associated with an increased risk of falls [3], fracture [4], cancer [5], and even mortality [6, 7]. Despite many factors affecting the loss of muscle strength in the elderly, the accumulation of reactive oxygen species (ROS) is one of the reasons for the age-related functional losses [8], causing oxidative protein damage and diminishing muscular strength [9].
Serum uric acid (SUA) is an end-product of purine metabolism and is suggested to have both oxidation and antioxidant effects [10], which is also a reliable marker of oxidative stress [11]. To date, the previous studies that the association between SUA and low muscle strength (LMS) have provided conflicting results. Precisely, whether the relationship between SUA and LMS is still vague of linear or non-linear. Some studies demonstrated that a specific range of SUA may be associated with better grip strength [12, 13]. Others argued that the linear relationship is either positive or negative. A cross-sectional study using data from the WCHAT suggested that SUA level was positively correlated with muscle strength [14]. Moreover, A NHANES study supported the same result that SUA level could be a protective factor for muscle strength in the elderly [15]. Contrastingly, a cohort PRO.V.A. study demonstrated that hyperuricemia corresponded to poor physical performance in older people, especially handgrip in men [16].
While all these studies have expanded our knowledge of this field, there are still some limitations. For example, most of them were cross-sectional designs that could not determine the causal relationship between SUA and LMS. Furthermore, uric acid excretion of approximately 70% is accounted for renal mechanisms, and the different stages of kidney functions related to SUA levels when considering muscle strength should be considered. Because muscle strength decreases substantially in the lower estimated glomerular filtration rate (eGFR) [17], we hypothesized that the risk of low muscle strength related to serum uric acid might increase with decreasing eGFR in the elderly population.
The China Health and Retirement Longitudinal Study (CHARLS) survey recruited from 150 counties or districts and 450 villages in 28 provinces in China, including 17,708 participants from 10,257 households, generally representing China's elderly annual population. In short, the CHARLS is a nationally representative longitudinal survey of mid-aged and elderly community-dwellers in China. All participants underwent an assessment using a standardized questionnaire interviewed by well-trained staff to collect data on demographic, lifestyle, and health-related information. Detailed information on CHARLS has been published previously [18]. In the present study, we retrospectively analyzed data from the CHARLS 2011 and 2015. Of the 17,708 participants aged over 45 who were enrolled in 2011, we excluded 14,009 participants and left 3,699 participants for follow-up. The exclusion criteria were participants with incomplete demographic data (N = 62), participants under 60 years old (N = 9982), participants with incomplete biochemical parameters (N = 3,336), participants with missing data from a lifestyle survey (N = 16), and participants with a lack of anthropometric parameters (N = 613). In 2015, we excluded 954 participants with handgrip strength data missing, leaving 2,745 participants for final analysis (Fig. 1). All participants provided informed consent; the Ethical Review Committee of Peking University approved the study protocol (IRB00001052–11015).
Measurement of SUA and eGFR
Venous blood samples were collected from each participant after 8 hours of fasting in wave 2011. SUA levels (mg/dL) were analyzed using enzymatic-colorimetric methods. The detection limits were up to 20 mg/dL, and the coefficient of variation (CV) intra-assay and inter-assay was equal to 1.10% and 1.90%, respectively. The estimated glomerular filtration rate (eGFR) was calculated based on the result of a multicenter study in Chinese populations [19]. The equation is eGFR = 173.9×CysC− 0.725×Cr− 0.184×Age− 0.193×0.89 [if female]. The eGFR was grouped into two categorical variables (cutoff 60 mL/min/1.73 m2) in men and women.
Muscle strength was assessed by handgrip strength which was measured in kg using a dynamometer [Yuejian™ WL-1000, Nantong, China]. Participants held the dynamometer in one hand with 90° elbow flexion in a sitting or standing position and then squeezed the dynamometer as hard as possible for a few seconds. The best of the two right and left measurements was recorded as handgrip strength. According to the latest study on low muscle strength in older Chinese adults, the cutoff point was 28.5 kg in men and 18.6 kg in women [20].
The present study variables included demographic information, anthropometric parameters, and blood sample. Demographic information was collected by trained staff during face-to-face interviews, including age, gender, education, smoking habits, alcohol consumption, and medical history of self-reported diagnosis. Other biomarkers, including anthropometric parameters and blood, were collected by China CDC staff. Body mass index (BMI) was calculated as weight(kg) divided by the square of height (m2). Waist circumferences (cm) were measured using soft tape around the navel. As mentioned above, blood samples were collected after 8 hours of fasting. These biochemical parameters included high-sensitivity CRP (hs-CRP), glycosylated hemoglobin (HbA1c), total cholesterol (TC), triglycerides (TG), HDL cholesterol (HDL-c), LDL cholesterol (LDL-c), creatinine (Cr), and cystatin C (Cys C).
The continuous variables were expressed as mean ± standard deviations (SDs) for normal distribution, and the categorical variables were addressed as frequency and proportion. Those covariates with skewed distribution were log-transformed, including hs-CRP and TG. The differences between groups were analyzed by t-test (normal distribution) or KruskalWallis rank sum test (skewed distribution) for continuous variables and Chi-square test for categorical variables. A generalized additive model was employed for interaction analysis and calculated the odds ratio (OR) and 95% confidence interval (CI) for the relationship between SUA per-SD (mg/dL) increased and LMS by different eGFR. We progressively adjusted for age, education levels, smoking habits, drinking consumption, BMI, waist circumferences, medical history (hypertension, diabetes, dyslipidemia, cancer, liver disease, and kidney disease), lipid profiles (TC, TG, HDL-c, and LDL-c), hemoglobin, HbA1c, and hs-CRP (Table 2,3). After adjusting for the abovementioned factors, the smoothing plots were illustrated to explore the possible non-linear association between SUA and LMS stratified by eGFR levels (Figs. 2,3). 𝑃 < 0.05 was considered statistically significant. All the statistical analyses were performed using EmpowerStats (http://empowerstats.com/en/; X&Y Solutions, Inc., Boston, MA, USA) and the R package (4.2.0 version).
We included 2745 participants (1360 men and 1385 women) in the analysis. The male participants mean SUA was 5.02 (mg/dL) ± 1.27 SD and mean eGFR was 74.44 (mL/min/1.73 m2) ± 13.78 SD, while the female participants mean SUA was 4.14 (mg/dL) ± 1.13 SD and mean eGFR was 73.88 (mL/min/1.73 m2) ± 14.62 SD.
Table 1 shows the baseline characteristics of different eGFR by sex. Men with a low eGFR (< 60 mL/min/1.73 m2) reported higher SUA levels (5.91 ± 1.27) and older (72.53 ± 6.38) than those who had a high eGFR, while women share the same difference with eGFR in SUA (5.00 ± 1.34) and age (72.81 ± 6.83). Biochemical parameters about hemoglobin and hs-CRP in different eGFR levels had statistically significant (P < 0.001) both in men and women. The prevalence of smoking in men was higher than in women; further, most women had neither a drinking nor smoking history. Different eGFR levels in the prevalence of hypertension and kidney disease had statistically significant in men. There were no apparent differences in other characteristics between different eGFR levels in men and women.
| Male | Female | |||||
|---|---|---|---|---|---|---|
| Characteristics | Low eGFR | High eGFR | P-value* | Low eGFR | High eGFR | P-value* |
| Number (N) | 185 | 1175 | 207 | 1178 | ||
| eGFR (mL/min/1.73 m2) | 53.35 ± 6.63 | 77.76 ± 11.48 | < 0.001 | 53.47 ± 6.03 | 77.47 ± 12.60 | < 0.001 |
| Uric acid (mg/dL) | 5.91 ± 1.27 | 4.88 ± 1.21 | < 0.001 | 5.00 ± 1.34 | 3.99 ± 1.01 | < 0.001 |
| Age (years) | 72.53 ± 6.38 | 67.61 ± 5.55 | < 0.001 | 72.81 ± 6.83 | 66.97 ± 5.85 | < 0.001 |
| Height (cm) | 161.26 ± 6.86 | 162.15 ± 6.52 | 0.089 | 150.41 ± 6.38 | 150.73 ± 6.48 | 0.510 |
| Weight (kg) | 57.60 ± 12.31 | 59.50 ± 10.98 | 0.031 | 52.67 ± 11.50 | 53.37 ± 10.53 | 0.386 |
| BMI (kg/m2) | 22.09 ± 4.36 | 22.57 ± 3.55 | 0.099 | 23.22 ± 4.58 | 23.82 ± 13.91 | 0.540 |
| WC (cm) | 81.98 ± 13.82 | 83.63 ± 12.13 | 0.093 | 86.06 ± 13.96 | 85.03 ± 12.59 | 0.287 |
| HbA1C (mg/dL) | 5.19 ± 0.57 | 5.25 ± 0.70 | 0.274 | 5.25 ± 0.68 | 5.39 ± 0.92 | 0.044 |
| TC (mg/dL) | 181.56 ± 31.77 | 186.97 ± 37.09 | 0.060 | 199.71 ± 42.91 | 203.85 ± 39.82 | 0.172 |
| Log TG (mg/dL) | 4.55 ± 0.44 | 4.57 ± 0.52 | 0.527 | 4.78 ± 0.45 | 4.79 ± 0.53 | 0.921 |
| HDL-c (mg/dL) | 51.97 ± 14.69 | 51.63 ± 16.44 | 0.789 | 52.18 ± 13.46 | 52.21 ± 15.21 | 0.978 |
| LDL-c (mg/dL) | 110.55 ± 29.41 | 113.31 ± 33.81 | 0.295 | 122.74 ± 40.61 | 124.13 ± 36.88 | 0.622 |
| Creatinine (mg/dL) | 1.17 ± 0.32 | 0.85 ± 0.15 | < 0.001 | 0.90 ± 0.19 | 0.68 ± 0.12 | < 0.001 |
| Cys C (mg/dL) | 1.63 ± 0.33 | 1.06 ± 0.17 | < 0.001 | 1.46 ± 0.23 | 0.97 ± 0.16 | < 0.001 |
| Hemoglobin (g/dL) | 14.17 ± 2.44 | 14.96 ± 2.13 | < 0.001 | 13.09 ± 2.08 | 13.76 ± 2.06 | < 0.001 |
| Log hs-CRP (mg/dL) | 0.65 ± 1.15 | 0.23 ± 1.08 | < 0.001 | 0.53 ± 1.09 | 0.19 ± 1.06 | < 0.001 |
| Education (%) | 0.058 | 0.014 | ||||
| illiteracy | 48 (26.09%) | 218 (18.57%) | 137 (66.18%) | 656 (55.69%) | ||
| primary school | 95 (51.63%) | 670 (57.07%) | 59 (28.50%) | 418 (35.48%) | ||
| secondary school above | 41 (22.28%) | 286 (24.36%) | 11 (5.31%) | 104 (8.83%) | ||
| Smoking history (%) | 0.434 | 0.320 | ||||
| No | 45 (24.32%) | 318 (27.06%) | 180 (86.96%) | 1052 (89.30%) | ||
| Yes | 140 (75.68%) | 857 (72.94%) | 27 (13.04%) | 126 (10.70%) | ||
| Drinking history (%) | < 0.001 | 0.937 | ||||
| No | 116 (62.70%) | 579 (49.28%) | 182 (87.92%) | 1038 (88.12%) | ||
| Yes | 69 (37.30%) | 596 (50.72%) | 25 (12.08%) | 140 (11.88%) | ||
| Hypertension | < 0.001 | 0.066 | ||||
| No | 112 (60.54%) | 867 (74.10%) | 125 (60.98%) | 793 (67.55%) | ||
| Yes | 73 (39.46%) | 303 (25.90%) | 80 (39.02%) | 381 (32.45%) | ||
| Dyslipidemia | 0.678 | 0.456 | ||||
| No | 168 (92.31%) | 1050 91.38%) | 181 (89.60%) | 1011 (87.76%) | ||
| Yes | 14 (7.69%) | 99 (8.62%) | 21 (10.40%) | 141 (12.24%) | ||
| Diabetes | 0.859 | 0.957 | ||||
| No | 173 (94.02%) | 1102 94.35%) | 190 (92.68%) | 1081 (92.79%) | ||
| Yes | 11 (5.98%) | 66 (5.65%) | 15 (7.32%) | 84 (7.21%) | ||
| Cancer | 0.823 | 0.129 | ||||
| No | 184 (99.46%) | 1161 99.32%) | 205 (100.00%) | 1157 (98.89%) | ||
| Yes | 1 (0.54%) | 8 (0.68%) | 0 (0.00%) | 13 (1.11%) | ||
| Liver disease | 0.470 | 0.493 | ||||
| No | 180 (97.30%) | 1123 96.23%) | 198 (97.06%) | 1123 (96.07%) | ||
| Yes | 5 (2.70%) | 44 (3.77%) | 6 (2.94%) | 46 (3.93%) | ||
| Kidney disease | 0.003 | 0.507 | ||||
| No | 159 (86.41%) | 1080 92.78%) | 192 (93.66%) | 1110 (94.79%) | ||
| Yes | 25 (13.59%) | 84 (7.22%) | 13 (6.34%) | 61 (5.21%) | ||
| High eGFR, eGFR ≥ 60 mL/min/1.73 m2. eGFR, estimated glomerular filtration rate; BMI, body mass index; WC, waist circumferences; HbA1c, glycosylated hemoglobin; TC, total cholesterol; log TG, log triglycerides; HDL-c, high-density lipoprotein cholesterol; LDL-c, low-density lipoprotein cholesterol; Cys C, cystatin C; log hs-CRP, log high-sensitivity CRP. *P < 0.05 was considered statistically significant. | ||||||
Table 1 is at the end of the document.
Tables 2,3 shows the eGFR stratified into the high and low groups to explore its interaction with SUA for LMS and adjusted progressively several risk factors. For women (Table 2), in analyses adjusted for age and education level, a significant association between SUA and low muscle strength in the high eGFR populations (OR = 0.79 95%CI = 0.68–0.91). Furthermore, after progressive adjustment for various risk factors, the OR for high eGFR level with SUA remained significantly with LMS (OR = 0.80 95%CI = 0.68–0.95 P = 0.0102). This correlation, however, was not observed in men (Table 3). There was no strong evidence for an increase in odds of low muscle strength with SUA in the high eGFR population (OR:1.03 95%CI = 0.88–1.20 P = 0.7305). Additionally, a significant interaction between different eGFR levels was observed (P = 0.0075) in women compared with interaction analysis in men (P = 0.5492).
| Low eGFR | High eGFR | ||||
|---|---|---|---|---|---|
| model | OR (95%CI) | P-value* | OR (95%CI) | P-value* | P for interaction* |
| Not adjusted | 1.16 (0.91, 1.47) | 0.2391 | 0.80 (0.69, 0.92) | 0.0023 | 0.0097 |
| Plus age and education | 1.17 (0.91, 1.51) | 0.2255 | 0.79 (0.68, 0.91) | 0.0018 | 0.0084 |
| Plus smoking and drinking | 1.17 (0.90, 1.51) | 0.2356 | 0.79 (0.68, 0.92) | 0.0021 | 0.0094 |
| Plus BMI and WC | 1.20 (0.93, 1.55) | 0.1605 | 0.81 (0.70, 0.95) | 0.0091 | 0.0099 |
| Plus medical history | 1.17 (0.90, 1.53) | 0.2407 | 0.81 (0.69, 0.95) | 0.0094 | 0.0176 |
| Plus HbA1c | 1.17 (0.90, 1.53) | 0.2484 | 0.81 (0.69, 0.95) | 0.0087 | 0.0175 |
| Plus lipid profiles | 1.20 (0.92, 1.58) | 0.1816 | 0.83 (0.70, 0.97) | 0.0229 | 0.0175 |
| Plus Hemoglobin | 1.24 (0.94, 1.64) | 0.1322 | 0.82 (0.69, 0.97) | 0.0188 | 0.0108 |
| Plus Log hs-CRP | 1.24 (0.94, 1.64) | 0.1309 | 0.80 (0.68, 0.95) | 0.0102 | 0.0075 |
| A generalized additive model progressively adjusted risk factors and smoothly adjusted for non-linear factors in the last model. High eGFR, eGFR ≥ 60 mL/min/1.73 m2. Medical history including hypertension, diabetes, dyslipidemia, cancer, liver disease, and kidney disease; Lipid profiles including total cholesterol, log triglycerides, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol; BMI, body mass index; WC, waist circumferences; HbA1c, glycosylated hemoglobin; log hs-CRP, log high-sensitivity CRP; SUA, serum uric acid; eGFR, estimated glomerular filtration rate; SD, standard deviation; OR, odds ratio; CI, confidence interval. *P < 0.05 was considered statistically significant. | |||||
| Low eGFR | High eGFR | ||||
|---|---|---|---|---|---|
| model | OR (95%CI) | P-value* | OR (95%CI) | P-value* | P for interaction* |
| Not adjusted | 1.03 (0.77, 1.38) | 0.8455 | 0.95 (0.83, 1.08) | 0.4484 | 0.6222 |
| Plus age and education | 1.13 (0.82, 1.55) | 0.4625 | 0.98 (0.85, 1.13) | 0.7655 | 0.4267 |
| Plus smoking and drinking | 1.14 (0.83, 1.58) | 0.4123 | 0.99 (0.86, 1.15) | 0.9370 | 0.4334 |
| Plus BMI and waist | 1.17 (0.85, 1.61) | 0.3442 | 1.03 (0.89, 1.20) | 0.6532 | 0.4982 |
| Plus medical history | 1.17 (0.84, 1.63) | 0.3511 | 1.05 (0.90, 1.22) | 0.5667 | 0.5412 |
| Plus HbA1c | 1.15 (0.83, 1.61) | 0.3985 | 1.04 (0.90, 1.22) | 0.5765 | 0.5915 |
| Plus lipid profiles | 1.17 (0.84, 1.63) | 0.3658 | 1.05 (0.90, 1.22) | 0.5626 | 0.5618 |
| Plus Hemoglobin | 1.14 (0.81, 1.61) | 0.4581 | 1.05 (0.90, 1.23) | 0.5203 | 0.6801 |
| Plus Log hs-CRP | 1.15 (0.82, 1.63) | 0.4179 | 1.03 (0.88, 1.20) | 0.7305 | 0.5492 |
| The generalized additive model progressively adjusted risk factors and smoothly adjusted for non-linear factors in the last model. High eGFR, eGFR ≥ 60 mL/min/1.73 m2. Medical history including hypertension, diabetes, dyslipidemia, cancer, liver disease, and kidney disease; Lipid profiles including total cholesterol, log triglycerides, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol; BMI, body mass index; WC, waist circumferences; HbA1c, glycosylated hemoglobin; log hs-CRP, log high-sensitivity CRP; SUA, serum uric acid; SD, standard deviation; OR, odds ratios; CI, confidence interval. *P < 0.05 was considered statistically significant. | |||||
Figures 2,3 show the relationship between serum uric acid and the risk of low muscle strength by the smoothing plot, with an adjustment for the risk mentioned above and stratification by different eGFR levels. The plots indicated that a higher eGFR level with SUA positively correlates to a lower risk of LMS in women. In men, however, this correlation was not observed.
To the best of our knowledge, this was the first retrospective cohort study that investigated the causal relationship between serum uric acid and the risk of low muscle strength at different eGFR levels above 60yrs population. We revealed that in subjects with a high eGFR level, higher SUA levels were significantly associated with low muscle strength events independent of several risk factors, especially in females. At the same time, this correlation was no longer present when the eGFR level decreased.
These results help to clarify conflicts among previous studies. A cross-sectional study reported that association of higher uric acid levels with better grip strength [14], and the study by Molino-Love et al. concluded the same results that high uric acid levels remained independent positive predictors of grip strength after adjusting confounders [21]. These two studies, however, did not control for the contribution of eGFR in the model, neither ruled out the lower eGFR level nor adjusted eGFR as confounders, which may be unable to present the effect of uric acid on muscle strength properly. Another notable point is that some studies included populations under 60 years to assess sarcopenia, an age-related disease. The researchers suggested that a specific range of SUA positively correlated with better grip strength, whose results could be misleading due to including other non-specific populations [12, 13]. On the other hand, consistent with our findings, a Korean study excluding eGFR < 60 populations extracted data from KNHANES 2016 indicated that SUA is positively associated with increased grip strength in the elderly [22]. Nahas et al., adjusting eGFR and other confounders, showed that older men and women could take advantage of a high SUA level with better handgrip strength [15]. Thus, the present study and the abovementioned studies suggested that a better eGFR level may be a protective factor for serum uric acid levels to improve muscular strength.
We describe that different eGFR levels may cause the opposite effects of SUA to the risk of LMS in both men and women (Fig. 2,3). The mechanism by which a high serum level of UA is associated with a decreased risk of LMS in different eGFR remains unclear. Most studies regarding the relationship between circulating uric acid and handgrip strength believed that SUA had an antioxidant effect on lowering oxidative stress by scavenging reactive oxygen species (ROS) [12–16, 21, 22]. This hypothesis, however, may need more tests when it comes to the lower eGFR. The accumulation of SUA due to declined kidney function may be easily influenced by alteration in a particular chemical milieu in the organism, which tends to affect the antioxidant ability of uric acid [23]. Accumulation of bicarbonate, commonly seen in electrolyte disturbances in renal insufficiency, might cause the uric acid its ability to against tyrosine nitrosylation, a crucial mechanism of oxidative damage [24]. Furthermore, according to a recent study, some researchers believed that uric acid might contribute to oxidative stress rather than being an antioxidant in physiological conditions [11]. The authors demonstrated that uric acid might directly lead to the production of ROS, and the antioxidant properties of uric acid might be neutralized by ROS generated from xanthine oxidoreductase catalyzed reactions, which implicated that only exogenous uric acid administration may have the anti-oxidative stress effect. For instance, the administration of uric acid improved the clinical outcomes in patients with acute ischemic stroke [25], or uric acid injection in Parkinson’s disease mice showed a neuroprotective effect [26]. Thus, oxidative stress accumulation could diminish the age-related loss of muscle strength [27] and kidney function [8]. However, since the method of detecting accurate circulating xanthine oxidoreductase activity was not pervasive in clinical and few studies with exogenous uric acid to improve muscle strength, whether the uric acid act as an antioxidant in muscle wasting with aging requires further investigation.
Our results also showed that sex-specific impacts might affect SUA to LMS in different eGFR levels, indicating that women rather than men had a high eGFR level with SUA per-SD (3.99 ± 1.01) benefit from a 20% lower risk of LMS. This result was opposite to the previous study. A cohort study adjusted eGFR and other confounders by Veronese et al. demonstrated that hyperuricemia in men was associated with lower handgrip strength, while this relationship was not observed in women [16]. Besides the small sample size included in the studies, which compromised a study's statistical power, both studies did not include gender-specific biochemical parameters as confounders, which require further investigation in the future. Gender difference is a critical factor in the associations between physical activity and muscle strength during aging [28]. Additionally, AWGS 2019 consensus showed that sarcopenia was more prevalent in men than women, suggesting that sexual dimorphism may be present in the pathogenesis of the disease [2]. Sex hormones could be part of the differences between genders due to their effect on circulating uric acid, eGFR level, and muscle strength. During adolescence, higher testosterone and lower sex hormone-binding globulin were reported as the gender difference in circulating uric acid [29]. Testosterone levels were decreased with aging, which was associated with body composition, including low muscle mass, decreased strength, and increased muscular fatigue [30]. The high follicle-stimulating hormone was associated with declined eGFR in post-menopausal women [31]. Correspondingly, a mendelian randomization study from the United Kingdom Biobank population suggested that high sex hormone binding globulin was beneficial to better kidney function and lower risk of CKD in men [32]. In a prospective cohort study, Tsai et al. using data from MJ Health Screening Database, indicated that a low serum testosterone level (< 400 ng/dL) was significantly associated with a high SUA level (> 7 mg/dL) in males [33]. Additionally, testosterone therapy could increase serum uric acid pharmacologically [34]. The possible mechanism showed that testosterone exposure caused elevated UA production by simulating xanthine oxidase [35]. Further, the administration of testosterone could increase muscle mass [36], muscle strength, and muscle power [37]. All these studies suggested that testosterone may cause a gender-specific difference in the relationship between SUA and LMS. Therefore, the interaction of sex differences in both muscular strength and circulating uric acid in kidney function should gain more attention.
Despite the theory of sex hormones and oxidative stress, other possibilities could be affecting muscular strength with SUA level. First, chronic low-grade inflammation was pivotal in sarcopenia [38]. It is known that high SUA levels caused by decreased renal excretion may produce urate crystals, which would induce gout and nephrolithiasis [39], repeatedly leading to high circulating inflammatory mediators. A meta-analysis has shown that higher circulating inflammatory markers, such as CRP and IL-6, were associated with a decline in muscular strength [40]. In the present study, however, the results are still unfluctuating after solely adjusted hs-CRP, suggesting that inflammation is not the primary driver of muscle strength loss. Consistent with our results, a Danish cohort study demonstrated that only high levels of hs-CRP had weakly relationship with low muscular mass [41]. Second, metabolic syndrome may inversely influence muscular strength [27]. A prospective longitudinal study showed that high waist circumference was positively associated with lower handgrip strength [42]. Our results showed that the results remained robust after progressive adjusting BMI and waist circumference, lipid profiles, HbA1c, and medical history separately. Thus, metabolic syndrome could lead to decreased muscular strength, which may not be the crucial factor in the relationship between SUA and LMS by different eGFR. Third, the decreased hemoglobin level could affect muscle mass and strength through several mechanisms associated with sarcopenia in non-dialysis chronic kidney disease patients [43] and kidney transplant recipients [45]. In the present study, we also adjusted the hemoglobin alone, which did not change the results in women and men.
Although our hypotheses were supported statistically, our study's results should be interpreted within its limitations. First, we did not evaluate whether the SUA levels changed during the follow-up and investigated the potential confounders of SUA levels at the baseline, including allopurinol and diuretics. Second, despite adjusting significant covariates separately, we cannot exclude the residual confounding of unmeasured factors in an observational study, including sex hormones, biomarkers of oxidative stress, nutritional status, and daily activity. Third, we did not have any information about gout history in this cohort. Moreover, the results should not be extrapolated to the presence of gout, which was related to SUA level and may affect the performance of muscle strength testing. Fourth, we focused on the population above 60yrs, which means that the results should not be extrapolated to those under 60 years, as they were “healthier” than the elderly.
This population-based cohort study in Chinese revealed that high serum uric acid level with higher eGFR seems to be significantly associated with a lower risk of low muscle strength in the elderly, especially in females.
Serum uric acid
estimated glomerular filtration rate
Low muscle strength
China Health and Retirement Longitudinal Study ORs:odds ratio
confidence interval
coefficient of variation
high-sensitivity CRP
glycosylated hemoglobin
total cholesterol
triglycerides
HDL cholesterol
LDL cholesterol
creatinine
cystatin C.
Ethics approval and consent to participate
The study was approved by the ethics committee of Peking University (IRB00001052–11015).
Consent for publication
Not applicable
Availability of data and materials
The datasets generated and analyzed during the current study are available in the CHARLS database, which is publicly available on the CHARLS website (http://charls.pku.edu.cn/).
Competing interests
The authors declare that they have no competing interests.
Funding
This work was supported by the Shanghai Municipal Administrator of Traditional Chinese Medicine [ZY[2021-2023]-0209-02].
Authors' contributions
YCH was responsible for the study conception and design, data acquisition, statistical analysis, and manuscript writing. SLC was responsible for acquiring data and writing and revising the manuscript. YD was responsible for statistical analysis and interpretation of data. YS was responsible for the study conception and design, data interpretation, and manuscript revision. All authors read and approved the final manuscript.
Acknowledgments
This work is based on China Health and Retirement Longitudinal Study [CHARLS]. We especially thank the institute of social science survey and the national school of development of Peking University for providing data. We also gratefully acknowledge the suggestions from Jian Pang in manuscript revision.