Increased Serum Cystatin C in Early Parkinson's Disease with Objective Sleep Disturbances.

BACKGROUND: Sleep disturbance is one of the major non-motor symptoms which cause the disability of Parkinson's disease (PD) patients. Cystatin C (CysC) is a more sensitive biomarker than serum creatinine or estimated glomerular filtration rate. Previous studies have reported altered CysC levels in neurodegenerative disorders and sleep disorders. This study aimed to explore the correlations of serum CysC levels and objective sleep disturbances in early PD.
METHODS: We recruited 106 early PD patients and 146 age- and sex-matched controls. All participants underwent clinical investigation and video-polysomnography. Sleep parameters and serum levels of CysC were measured. Then, we investigated the relationships between CysC and clinical variables and objective sleep disturbances in early PD patients.
RESULTS: The mean serum level of CysC was significantly higher in patients with early PD (1.03 ± 0.19 mg/L) compared to controls (0.96 ± 0.15 mg/L, P = 0.009). There were significantly positive correlations between serum CysC levels and age (r = 0.334, P < 0.001), gender (r = 0.264, P = 0.013), and creatinine levels (r = 0.302, P = 0.018) in early PD patients. Increased serum CysC levels in early PD patients were significantly associated with higher apnea and hypopnea index (AHI) (r = 0.231, P = 0.017), especially hypopnea index (r = 0.333, P < 0.001). In early PD patients, elevated serum CysC levels were positively correlated with oxygen desaturation index (r = 0.223, P = 0.021), percentage of time spent at oxygen saturation (SaO2) <90% (r = 0.644, P < 0.001), arousal with respiratory event during sleep (r = 0.247, P = 0.013). On the contrary, the elevated serum CysC levels were negatively correlated with mean and minimal SaO2(r = -0.323, -0.315, both P = 0.001) in PD patients.
CONCLUSIONS: The level of serum CysC was higher in early PD patients. PD patients with elevated serum CysC levels had more respiratory events and more severe oxygen desaturation. Therefore, the serum CysC levels may predict the severities of sleep-disordered breathing problems in early PD patients.

INTRODUCTION

Parkinson's disease (PD) is a common neurodegenerative disease which is identified by typical motor symptoms such as resting tremor, rigidity, bradykinesia, and postural instability.[1] It has been recognized recently that a variety of non-motor symptoms accompanied with PD such as sleep disorders, olfactory dysfunction, depression, autonomic changes, and dementia. Sleep disturbances, a major non-motor symptom of PD, particularly degrades PD patients’ quality of life and even cause the disability of PD patients.[2] Although the authentic pathophysiology of PD is still unknown, the pathogenesis of PD has been linked to several mechanisms, including inflammation, oxidative stress, mitochondrial dysfunction, abnormal protein aggregation, and hyperactivation of N-methyl-D-aspartic acid receptors.[34] Recent studies which focused on the mechanisms of neurodegenerative diseases have involved homocysteine, uric acid, cystatin C (CysC), and C-reactive protein.[567]

CysC, an inhibitor of cysteine proteases, belongs to the cystatin type 2 superfamily. As a biomarker of kidney dysfunction, it is more sensitive than serum creatinine or estimated glomerular filtration rate.[8] CysC shows a broad spectrum of biological activities in numerous cellular systems, including effects on growth promotion, inhibition of inflammation, and antimicrobial activity.[9] Even though previous studies have demonstrated that CysC has been got involved in neurodegenerative disease, its potential pathologic role is not completely understood and remains controversial. CysC induces autophagy in vivo as a protective mechanism in brain injury and in neurodegenerative disorders.[10] Enhanced CysC gene expression and higher CysC protein levels were also shown in dopaminergic-depleted rat striatum following a 6-hydroxydopamine (6-OHDA)-induced lesion in nigrostriatal neurons, astrocytes, and microglia cells.[11] Several studies have confirmed an association between CysC and sleep disorders especially obstructive sleep apnea (OSA).[121314] In the study presented here, we compared CysC levels between early PD patients and healthy controls and explored the correlations between CysC levels and clinical variables in early PD patients. Moreover, we demonstrated the correlations between CysC and objective sleep disturbances in early PD patients. We seek to clarify whether CysC could predict objective sleep disturbances especially sleep-disordered breathing (SDB) problems in early PD patients.

METHODS

Ethical approval

This study was approved by the Ethics Committee of the Second Affiliated Hospital of Soochow University, and each participant provided written informed consent.

Participants

We recruited 106 early PD patients from the Department of Neurology, The Second Affiliated Hospital of Soochow University from July 2011 to May 2016. All the PD patients met the United Kingdom PD Brain Bank diagnosis criteria.[15] We had excluded individuals with atypical parkinsonism (e.g., multiple system atrophy or progressive supranuclear palsy), serious heart disease, renal disease, liver disease, hematologic system disorders, cancer, and infective or inflammatory disorders. During the same enrollment period, we assessed 146 age- and sex-matched controls that we screened from the medical healthy examination system of our hospital.

Clinical assessment and video-polysomnography

All PD patients underwent comprehensive clinical investigation including general characters, disease history, and comorbid diseases. The Hoehn and Yahr (H-Y) stage of all PD patients were lower than or equal to 2.5.[16] Each PD patient was performed an overnight video-polysomnography (vPSG) (Compumedics-E series, Australia) in the sleep center. The basic recordings included electroencephalogram (F3A2, F4A1, C3A2, C4A1, O1A2, and O2A1), electrooculogram (LOCA2, and ROCA1), chin EMG, electrocardiogram, nasaloral pressure transducer airflow, thermal oronasal airflow, thoracic and abdominal respiratory efforts, oxyhemoglobin saturation, snoring sound, body position, and leg movement. Sleep architecture such as awakenings (n); total sleep time (TST)(min); sleep efficiency (SE)(%); sleep latency (SL)(min); percentage of time spent in non rapid eye movement sleep stage (NREM) 1(%), NREM2(%), NREM3(%), and rapid eye movement sleep stage (REM)(%), the index of periodic leg movements during sleep (PLMSI)(/h), and sleep apnea parameters including apnea hypopnea index (AHI)(/h), apnea index (AI), hypopnea index (HI)(/h), oxygen desaturation index (ODI)(/h), and minimal oxygen saturation (SaO2)(%), the percentage of time spent at SaO2<90% (%), mean SaO2 (%), and arousal with respiratory events (ARO RES)(/h) during sleep were assessed. The scoring of vPSG was determined by experienced PSG technologists and clinicians, according to the American Academy of Sleep Medicine (AASM).[1718]

Laboratory measurements

All patients and controls underwent routine laboratory tests in our clinical laboratory center after they fasted overnight. The blood samples were drawn from a peripheral vein and centrifuged for up to one hour. Serum levels of CysC were measured using an immunoturbidimetry assay (Cystatin C Kit, Shanghai Jingyuan Company, China). Concentrations of serum creatinine, urea, and uric acid were determined using an enzymatic assay and different diagnostic reagents (Kyowa Medex Company, Japan). All laboratory tests were performed with an AU5400 random access analyzer (Olympus Corporation, Japan).

Statistical analysis

SPSS software 17.0 (Chicago, IL, USA) was used for the statistical analysis. Continuous variables were expressed as mean ± standard deviation (SD). Categorical variables are expressed as frequency (percent). Comparisons of continuous variables between two groups were conducted using independent Student's t-test. Relationships between CysC and clinical variables and sleep parameters in PD patients were assessed by Pearson's correlations. Statistical significance was defined as P < 0.05.

RESULTS

Demographics and clinical characteristics

Demographic and clinical characteristics of individuals are presented in Table 1. Among the 106 early PD patients, 70 were male. For the PD group, the mean age was 65.5 ± 9.6 years, mean body mass index was 23.06 ± 3.28 kg/m2, mean disease duration was 4.4 ± 3.2 years, and mean H-Y stage was 2.07 ± 0.36. We found OSA in 47 (44.9%) PD patients, with 25 (53.19%) demonstrating mild OSA (5< AHI ≤15), 12 (25.53%) demonstrating moderate OSA (15< AHI ≤30), and 10 (21.28%) demonstrating severe OSA (AHI >30).[19]

Table 1

Comparison between CysC and clinical variables in PD patients and controls

Characteristics	PD	Control	P
Number of subjects	106	146
Demographics
Age (years)	65.5 ± 9.6	65.7 ± 8.4	0.964
Male gender	70 (66.0)	84 (57.5)	0.204
CysC (mg/L)	1.03 ± 0.19	0.96 ± 0.15	0.009*
Creatinine (μmol/L)	64.5 ± 11.5	65.5 ± 12.3	0.591
Urea (mmol/L)	5.3 ± 1.5	5.4 ± 1.4	0.594
Uric acid (μmol/L)	291.5 ± 64.9	303.5 ± 75.7	0.277

Continuous variables are expressed as mean ± SD. Categorical variables are expressed as n (%). P<0.05 was considered statistically significant. *P<0.05. SD: Standard deviation; CysC: Cystatin C; PD: Parkinson’s disease.

As shown in Table 1, the PD patients and controls were similar in age and gender, but the PD group had a slightly higher percentage of men which is consistent with the male predominance of PD.[20] The mean serum level of CysC was significantly higher in patients with early PD (1.03 ± 0.19 mg/L) compared to controls (0.96 ± 0.15 mg/L, P = 0.009), while the creatinine, urea, and uric acid levels were very approximate of two groups.

Relationships between cystatin C level and clinical variables

Correlation analysis between CysC and clinical variables in early PD patients was presented in Table 2. There were significant positive correlations between serum CysC levels and age (r = 0.334, P < 0.001), gender (r = 0.264, P = 0.013), and creatinine levels (r = 0.302, P = 0.018) in early PD patients. No correlation was observed between CysC levels and urea (r = 0.224, P = 0.083) or uric acid (r = 0.000, P = 0.998) in PD cohort.

Table 2

Correlation analysis between CysC and clinical variables in PD patients

Variables	r	P
Age	0.334	<0.001†
Gender	0.264	0.013*
BMI	0.139	0.161
Creatinine	0.302	0.018*
Urea	0.224	0.083
Uric acid	0.000	0.998

P<0.05 was considered statistically significant. *P<0.05; †P<0.01. BMI: Body mass index.

Relationships between cystatin C level and objective sleep disturbances

The objective sleep parameters of this early PD cohort were similar to our studies before[2122] as TST was 352.6 ± 108.3 min, SE was 60.9% ± 18.3%, SL was 30.6 ± 52.0 min, awakenings (n) was 20.5 ± 9.4, NREM1 was 29.6% ± 18.5%, NREM2 was 42.9% ± 16.0%, NREM3 was 13.1% ± 11.6%, REM was 14.4% ± 9.1%, PLMSI was 31.5 ± 50.1/h, AHI was 9.2 ± 13.5/h, AI was 6.2 ± 11.1/h, HI was 3.0 ± 5.4/h, ODI was 8.6 ± 12.5/h, minimal SaO2 (%) was 89.3% ± 3.9%, the percentage of time spent at SaO2<90% (%) was 1.7% ± 4.2%, and arousal with respiratory events during sleep was 2.5 ± 6.3/h.

The relationships between CysC level and objective sleep parameters were displayed in Table 3. Increased serum CysC levels in early PD patients were significantly associated with higher AHI (r = 0.231, P = 0.017) especially HI (r = 0.333, P < 0.001). Therefore, elevated serum CysC levels were positively correlated with ODI (r = 0.223, P = 0.021), percentage of time spent at SaO2 90% (r = 0.644, P < 0.001), arousal with respiratory event during sleep (r = 0.247, P = 0.013). On the contrary, the elevated serum CysC levels were negatively correlated with mean and minimal SaO2 (r = −0.323, −0.315, both P = 0.001). There were no significant relationships among serum CysC levels and sleep architecture parameters including TST, SL, SE, awakenings, percentage of NREM1, percentage of NREM2, percentage of NREM3, percentage of REM, and PLMSI (r = 0.047, 0.060, −0.066, 0.149, 0.138, −0.157, −0.063, 0.077, 0.016, all P > 0.05).

Table 3

Correlation analysis between CysC and sleep parameters in PD patients

Variables	r	P
TST (min)	0.047	0.634
SL (min)	0.060	0.539
SE (%)	−0.066	0.498
Awakenings (n)	0.149	0.125
NREM1 (%)	0.138	0.155
NREM2 (%)	−0.157	0.107
NREM3 (%)	−0.063	0.519
REM (%)	0.077	0.429
PLMSI (/h)	0.016	0.878
AHI (/h)	0.231	0.017*
AI (/h)	0.102	0.295
HI (/h)	0.333	<0.001†
ODI (/h)	0.223	0.021*
Mean SaO2 (%)	−0.323	0.001†
Minimal SaO2 (%)	−0.315	0.001†
Percentage of time spent at SaO2 <90% (%)	0.644	<0.001†
ARO RES (/h)	0.247	0.013*

P<0.05 was considered statistically significant. *P<0.05; †P<0.01. TST: Total sleep time; SE: Sleep efficiency; SL: Sleep latency; NREM: Non rapid eye movement sleep stage; REM: Rapid eye movement sleep stage; PLMSI: Index of periodic leg movements during sleep; AHI: Apnea hypopnea index; AI: Apnea index; HI: Hypopnea index; ODI: Oxygen desaturation index; ARO RES: Arousal with respiratory events during sleep; SaO2: Oxygen saturation.

DISCUSSION

This study was to investigate the relationships between CysC and objective sleep disturbances in early PD patients. We found that patients with early PD had higher serum CysC levels than healthy controls. We found significant positive correlations between CysC level and age, sex, and creatinine level in early PD patients. More importantly, elevated serum CysC levels in early PD patients were found in correlation with the severity of sleep apnea, hypopnea, and oxygen desaturation. These findings suggest that serum CysC level could predict the situation of OSA in early PD patients.

There were rare studies about the correlation of CysC and PD. In our previous study, Hu et al.[23] identified the elevated CysC levels and their positive correlations with age, sex, and creatinine in PD patients, which were consistent with our results. On the other hand, it demonstrated that the PD patients with higher CysC levels had lower scores on cognitive tests while we identified that increased CysC had a significant correlation with objective sleep disturbances especially OSA characters. Therefore, CysC may predict cognitive dysfunction and SDB in patients with PD. In addition, it showed PD patients with middle- and late-stage disease had significantly higher CysC levels than those in the early stage and CysC level correlated significantly with H-Y stage and with disease duration.[23] Since all the PD patients we enrolled in this cohort were at early stages with H-Y stage less than or equal to 2.5, we did not repeat the analysis of the relationship between serum CysC level and H-Y stage. The recruitment strategy of this study would be more critical for biomarker research in early PD patients even though we did not comprehensively enroll PD participants with all H-Y stages.

CysC, a 13-kD protein and a member of a family of competitive inhibitors of lysosomal cysteine protease, was originally identified in human cerebrospinal fluid and was later found to be expressed in most mammalian tissues as well as in blood.[24] CysC has an excellent ability to detect clinically latent chronic kidney disease (CKD) and to reflect cardiovascular diseases mortality.[25] Enhanced CysC expression in neurodegeneration diseases caused a debate as to whether CysC up-regulation facilitates neurodegeneration or it is an endogenous neuroprotective attempt to prevent the progression of the pathology. Nagai et al.[26] reported that direct injection of CysC caused the death of rat hippocampal neurons and that CysC released from toxin-injured rat dopaminergic neurons induced microglial activation and exacerbated neurotoxicity. However, recent in vitro and in vivo data have demonstrated that CysC played protective roles via pathways that were dependent on inhibition of cysteine proteases, such as cathepsin B, or by induction of autophagy, induction of proliferation, and inhibition of amyloid-beta aggregation.[9] For instance, Xu et al.[11] reported that loss of dopaminergic neurons in cultures of rat fetal mesencephalic cells exposed to 6-OHDA could be partially reversed by treatment with human CysC. This result suggested that the neuroprotective effect of CysC in PD might result from the regrowth of dopaminergic neurons.

OSA is a common condition characterized by repetitive obstruction episodes of the upper airway during sleep, resulting in oxygen desaturation and arousal from sleep. The associations are believed to be in part based on adverse effects of intermittent hypoxia[27] and sympathetic nervous system activity on oxidative stress, insulin resistance, and endothelial dysfunction.[28] Epidemiological studies suggested that OSA is not an uncommon disorder in PD as well.[2930] Previous studies reported a frequency of OSA between 20% and 66% in PD.[19] Kato et al.[12] performed a cross-sectional study with 267 consecutive OSA patients without CKD who had an AHI ≥5 events per hour and demonstrated CysC was significantly correlated with AHI (r = 0.17), severe OSA defined by an AHI ≥30 events per hour was an independent variable for the highest quartiles of serum CysC levels (≥0.88 mg/L), indicated that severe OSA independently increases serum CysC levels in patients without CKD. Zhang et al.[13] detected that serum CysC levels were associated with the severity of OSA in younger men without comorbidities.

Strengths of our study include its enrollment of early PD patients with H-Y stage lower than or equal to 2.5, a relatively precise cohort in comparison with prior investigations, comprehensive measures of objective sleep disturbances, and concurrent identification of OSA using vPSG and AASM scoring manual. Nevertheless, our study has several limitations. The sample size was not sufficient, and all participants were recruited from a single center. This was a retrospective study, precluding any conclusions regarding causality. Furthermore, the challenge of identifying true associations between CysC and sleep disturbances in the early PD patients related to multiple comorbidities that may confound or modify hypothesized associations.

In conclusion, the level of serum CysC was higher in early PD patients than in controls. PD patients with elevated serum CysC levels had more respiratory events and more severe oxygen desaturation. Therefore, the serum CysC levels may predict the severities of SDB problems in early PD patients. Further studies which should be performed in multicenter with prospective and mechanic design are needed to clarify the predictive value of CysC in early PD and understand the pathologic mechanisms of altered CysC in early PD patients with SDB problems.

Financial support and sponsorship

This work was supported by grants from National Key R and D Program of China (No. 2017YFC0909100); National Natural Science Foundation of China (No. 91649114); Jiangsu Provincial Medical Key Discipline Project (No. ZDXKB2016022); Jiangsu Provincial social development projects (No. BE2017653); Jiangsu Key Laboratory of Neuropsychiatric Diseases (No. BM2013003); Suzhou Clinical Research Center of Neurological Disease (No. Szzx201503). This work was also partly supported by the Priority Academic Program Development of Jiangsu Higher Education Institutions.

Conflicts of interest

There are no conflicts of interest.