Introduction

Multiple system atrophy (MSA) is a sporadic and adult-onset neurodegenerative disease characterized by parkinsonism, cerebellar ataxia, autonomic dysfunction and pyramidal dysfunction in any combination [1]. The main neuropathological features of MSA are abnormal α-synuclein positive cytoplasmic inclusions presented in oligodendrocytes, neuronal loss, gliosis, myelin pallor and axonal degeneration in striatonigral and/or olivopontocerebellar systems [1]. The etiology of MSA is not yet fully understood. Recently, some epidemiological studies found that occupational exposure to some chemicals that interfere with the mitochondrial electron transport chain and increase oxidative stress may increase the risk of MSA [2, 3]. An animal study found that a compound 3-nitropropionic acid (3-NP) with a function of inhibition mitochondrial complex II could induce MSA in an animal model [4–6]. These findings suggest that oxidative stress is involved in the pathogenesis of MSA [2, 4, 6–9].

Uric acid, a natural antioxidant, has been found to play a key role in the risk and progression of some neurodegenerative diseases with the pathogenesis of oxidative stress [10–13]. For example, many clinical studies have found that high levels of serum uric acid decreased the risk of Parkinson’s disease (PD) [14–16] and amyotrophic lateral sclerosis (ALS) [17, 18]. Furthermore, some studies showed that high levels of serum uric acid decreased the progression of PD [19] and ALS [17]. However, other studies found that high levels of serum uric acid decreased the risk [20, 21] and progression [22–24] of PD and ALS in male subjects, but not in females.

At present, the associations between serum uric acid and MSA remain largely unknown. There were two studies from Korea focused on the association of uric acid and the progression of MSA. However, the results were not consistent with each other. Lee JE et al. [25] studied 52 MSA patients, most of whom had MSA type C (48 patients), and found that high uric acid slowed down the progression of MSA. Another large sample Korean study including 455 MSA patients found that uric acid was not associated with survival of MSA [26]. However, the study did not consider the influence of metabolism factors, such as BMI, hypertension and diabetic mellitus, which could have an impact on the level of uric acid levels. To further verify the relationship between serum uric acid and MSA, we conducted this prospective study in a Chinese population.

Patients and methods

A total of 234 Chinese MSA patients admitted in the Department of Neurology, West China Hospital, Sichuan University from November 2006 to May 2013 were enrolled in the study. All the MSA patients met the probable MSA clinical diagnostic criteria [1]. The control group was composed of 240 age- and gender-matched subjects from the Medical Examination Center (MEC), West China Hospital of Sichuan University. None of the controls had any neurological diseases. The fasting serum uric acid concentrations of the MSA patients and controls were measured in the clinical laboratory of the West China Hospital of Sichuan University. Clinical information, including gender, age, BMI, histories of hypertension and diabetes mellitus, were collected. The severity of MSA was assessed by the Unified MSA Rating Scale (UMSARS), which had been thought to have a high criteria-related validity [27, 28]. The progression of MSA was assessed by the annualized rate of changes in the UMSARS scores, which was calculated by the following formula: [(total UMSARS score at the last visit − total UMSARS score at the baseline visit)/number of days between the assessments] × 365 day [25]. The study was approved by the Sichuan University Ethics Committee.

Data analyses

All continuous data including mean age, disease duration, BMI and serum uric acid level were presented as mean ± standard deviation and all categorical variables including gender, subtype (C-type or P-type), histories of hypertension and diabetes mellitus were presented as percentages. The differences in continuous data and categorical variables between MSA patients and controls were studied by student’s t test and Chi-square test respectively. The levels of serum uric acid were assigned into quartiles based on its distribution in the controls and the highest quartile was set as a reference for analysis. Odds ratios (ORs) and 95 % confidence intervals (CIs) were measured by Chi-square tests. We employed the mean rate of annualized changes of the UMSARS score as a dependent variable, and the uric acid level, age, disease duration at initial visit, gender, BMI and subtype of MSA as independent factors. All the statistical analyses were performed using commercially available software (SPSS, version 19.0), and a p < 0.05 was deemed statistically significant.

Results

The demographic characteristics and serum uric acid levels of included MSA patients and controls are listed in Table 1. The mean disease duration of MSA patients was 2.60 ± 1.75 years. There were no significant differences in mean age, gender distribution, BMI and histories of hypertension and diabetes mellitus between MSA patients and controls (Table 1). Except for disease duration at initial visit and the mean interval between baseline and last assessment, the rest of the continuous data in our study were normally distributed by the Kolmogorov–Smirnov test. Serum uric acid levels were lower in all MSA patients than that in controls (5.01 ± 1.36 vs. 5.31 ± 1.33, p = 0.01, Table 1). However, in a gender-specific analysis, this difference was only found in male patients compared with controls (5.50 ± 1.30 vs. 6.15 ± 1.19, p = 0.0001, Table 1).

Table 1 Demographic characteristics and serum uric acid levels of patients with MSA and control subjects
Full size table

The ORs of MSA patients according to the uric acid profile quartiles of the controls are listed in Table 2. The univariant analysis of uric acid profiles found that subjects with the lowest uric acid quartiles had higher occurrences of MSA compared with those with the highest uric acid quartiles (OR = 2.10, CI = 1.24–3.56, p = 0.005, Table 2). In a gender-specific analysis, a significant higher occurrence of MSA was found in the second and lowest quartiles compared with those with the highest quartiles in males (OR = 3.63, CI = 1.77–7.46, p = 0.0001; OR = 4.67, CI = 1.84–11.85, p = 0.001, respectively, Table 2). However, the serum uric acid levels were not associated with either increased or decreased occurrence of MSA in females (Table 2).

Table 2 Odds ratios (OR) of patients with MSA based on uric acid profile quartiles
Full size table

A total of 133 patients were followed-up; however 26 of these patients have since died. The longitudinal data of 107 patients were analyzed for the progression of MSA. Among 107 MSA patients, including 56 males and 51 females, 67 had MSA-C and 30 had MSA-P. All patients were assessed by the UMSARS at initial visit and at the last visit. The mean age of these followed-up MSA patients was 58.83 ± 8.88 years and the mean duration at the initial visit was 4.34 ± 1.84 years. The mean level of serum uric acid was 5.02 ± 1.37 mg/dl and mean BMI was 23.29 ± 2.70. The mean interval between baseline and last assessment of UMSARS-1 was 1.05 ± 0.70 years, and the mean rate of annualized changes in the UMSARS score was 18.08 ± 10.40. Through multiple regression analysis, we found that the level of uric acid, age, disease duration at the initial visit, BMI, gender, and subtype of MSA did not significantly correlate with the mean rate of annualized changes in the UMSARS.

Discussion

To our knowledge, this is the first study exploring the association of serum uric acid levels and the prevalence of MSA. Our research found that serum uric acid was significantly lower than that of healthy controls. High levels of serum uric acid may be associated with a low prevalence of MSA in the Chinese population, especially in males. Uric acid levels neither worsened nor ameliorated the progression of MSA.

Significantly low levels of uric acid observed in our Chinese MSA patients compared to that in controls was consistent with our findings that low levels of serum uric acid increases the occurrence of MSA, and is similar to the results of previous studies on other neurodegenerative diseases such as PD [14, 15, 19, 21] and ALS [17]. Accumulated epidemiological data and animal-based evidence have highlighted the notion that oxidative stress plays a key role in the pathogenesis of MSA [2, 4, 6–9]. For example, some epidemiological studies [2, 7] found that occupational exposure to pesticides, insecticides, or chemicals which interfere with the mitochondrial electron transport chain and induce oxidative stress may increase the risk of MSA. 3-NP as an environment toxin, which can inhibit mitochondrial complex II and enhance oxidative modification of α-synuclein, has been reported to induce MSA in an animal model [4–6]. In vitro experiments have been revealed that uric acid is a scavenger of peroxyl radicals (RO2), hydroxyl radicals (OH) and singlet oxygen [10]. Uric acid has been reported to inhibit the radicals generated by the decomposition of peroxynitrite, a strong oxidizing agent which is able to interact with almost all important cell constituents inducing cell injuries [29]. Besides its action as a radical scavenger, uric acid can also chelate metal ions such as iron and copper, and convert them to poorly reactive forms, which are unable to catalyze free-radical reactions [30–32]. Although the exact mechanism of low uric acid levels increasing the occurrence of MSA remains unclear, the antioxidative function of uric acid may contribute to such an association.

Our finding of the lower serum uric acid levels increasing the occurrence of MSA only in males may be due to the significantly higher uric acid levels in males than in females. Further, we speculate that a biological interaction between gender specific hormones and uric acid might play a key role in the strong association between uric acid level and the occurrence of MSA in men.

Our study found that the level of uric acid did not worsen or ameliorate the progression of MSA. Our finding was supported by the Korean study [26], although some metabolism factors such as BMI, hypertension and diabetic mellitus, which could have an impact on uric acid levels, were not taken into consideration in that study. In the current study, we excluded the effect of BMI, hypertension and diabetic mellitus on uric acid. Our finding was not consistent with the positive finding of another small sample Korean study [25]. The sample selection bias may contribute to such a difference since most of the included patients in that study were patients with MSA subtype C. Whether uric acid has an important effect on the occurrence but not the progression of MSA remains largely unknown. With our findings, we understand that uric acid may have an important function in the pathogenesis but not in the progression of MSA. Considering the complications of high uric acid in aging individuals, appropriate medicines should be provided to treat high serum uric acid level in patients with MSA.

Some limitations of our study should be considered. First, part of our study involved a case-control design, and the results did not reflect the longitudinal effects of uric acid. The control subjects were from the Medical Examination Center, and they may contribute to the occurrence of bias because these subjects were from the urban population and may have had a better economic situation. Additionally, we have to consider the selection bias of patients in our follow-up study. The disease severity was variable among our patients, and patients with less severe symptoms had a higher rate of return visits while patients with severe symptoms had a lower rate of return visit. Moreover, other factors that may influence the risk of MSA, such as smoking, were not adjusted in the statistical analysis [33, 34]. Although our study is a prospective study, the small sample size and the short follow-up period limited the significance of our finding. Future studies with a larger sample size and longer follow-up period should be considered.

Conclusion

MSA patients have lower levels of serum uric acid than healthy controls. Our study found that low levels of uric acid may be associated with the prevalence of MSA in the Chinese population, especially in males. However, uric acid did not contribute to the progression of MSA. Further prospective studies with larger sample sizes should be conducted to confirm such a relationship.