Observational Study of the Relation between Parkinson’s Disease and Sleep Apnea

Abstract

Background:

Parkinson’s disease is characterized by numerous non-motor symptoms, including sleep disorders. Sleep apnea has been reported in a substantial proportion of patients with Parkinson’s disease, but it is unclear whether it has significant consequences for the quality of life of those affected or whether it is associated with other manifestations of the disease.

Objective:

To verify whether sleep apnea is associated with more severe motor and non-motor clinical features in Parkinson’s disease.

Methods:

Parkinson’s disease patients underwent polysomnography to diagnose the presence of sleep apnea (apnea-hypopnoea index >10). Participants also underwent an extensive assessment, blinded to sleep apnea status, to determine disease severity, quantitative motor indices, motor subtypes, treatment complications, and sleep, autonomic, psychiatric, and sensory dysfunctions. Cognitive status was also determined with a complete neuropsychological assessment. Results were assessed using regression analysis adjusted for age, sex, and disease duration.

Results:

Of 92 patients examined, 19 had sleep apnea (21%) and 73 did not. We found no significant differences in motor and non-motor symptoms or signs between apneic and non-apneic Parkinson’s disease patients. The use of different apnea-hypopnoea index cut-offs (>5 and >15) produced similar results.

Conclusions:

Our results show that sleep apnea is not associated with more severe motor or non-motor manifestations in Parkinson’s disease. More studies including control groups are needed to confirm the implications of those results.

INTRODUCTION

One of the most common sleep problems in the general population is sleep apnea, characterized by episodes of arrested breathing during sleep, often associated with reduced oxygen saturation [1]. In the general population, sleep apnea is associated with excessive daytime sleepiness, cognitive impairment, and vascular disease [1]. Patients with Parkinson’s disease (PD) also commonly suffer from sleep symptoms, including excessive daytime sleepiness, insomnia, and dream-enactment behavior [2]. These manifestations by themselves are often disabling. Moreover, some of them are associated with complications such as cognitive decline in PD [3, 4].

However, the relationship between sleep apnea and PD is controversial. It is clear that, like the general population, individuals with PD and other neurodegenerative diseases can suffer from sleep apnea [2, 5]. However, estimated prevalences of sleep apnea in PD vary widely, and it is unclear whether PD actually predisposes to sleep apnea [2, 7]. Regardless of prevalence, sleep apnea symptoms overlap considerably with motor and non-motor manifestations of PD. Therefore, in some patients, these symptoms could be caused or exacerbated by sleep apnea. Moreover, given the effects of sleep apnea on overall health, some disease manifestations may be more severe in patients with concomitant sleep apnea.

We conducted a comprehensive cohort study of sleep variables in PD, focusing on relationships between sleep disorders and PD manifestations. This has provided an opportunity to determine the associations between sleep apnea and PD manifestations.

The objective of this study was to describe the clinical, motor, sleep, autonomic, sensory, mood, and cognitive manifestations of PD patients in association with the presence or absence of sleep apnea.

METHODS

Participant selection

PD patients were recruited from two specialized movement disorder units: the McGill University Health Centre (Department of Neurology, Montreal General Hospital, Quebec, Canada) and the Unité des troubles du mouvement André Barbeau, Centre Hospitalier de l’Université de Montréal (Montreal, Quebec, Canada). These two units have participated in previous works focusing on rapid eye movement sleep behavior disorder (RBD) [8–11]. Patients were included in the present study if they had parkinsonism according to the UK Parkinson Disease Society Brain Bank criteria and if idiopathic PD was determined to be the likeliest cause [12]. The exclusion criteria were: 1) dementia, based on neuropsychological assessment and MDS criteria; [13] and 2) an alternate cause of parkinsonism. Patients were assessed at the Centre for Advanced Research in Sleep Medicine, Hôpital du Sacré-Cœur de Montréal (Montreal, Quebec, Canada). The study was approved by the hospital’s research ethics board, and all participants gave their written informed consent.

Polysomnographic recording and sleep apnea diagnosis

All participants underwent one night of polysomnographic (PSG) recording in a sleep laboratory. The recording montage included central (C3-A2) and occipital (O2-A1) EEG to assess sleep stages, left and right electrooculograms, and submental EMG. Participants were also monitored with infrared video to detect movements during REM sleep. REM sleep stage and REM sleep EMG activity (tonic and phasic) were scored according to a method described elsewhere [14]. Sleep stages 1 to 4 were scored according to a modified version of Rechtschaffen and Kales’ method using 20-second epochs [15]. The following PSG variables were measured: sleep latency, total sleep time, sleep efficiency, REM efficiency, REM latency, number of awakenings during sleep, microarousal index, and percentage of stage 1, stage 2, slow-wave, and REM sleep. RBD diagnosis was based on the International Classification of Sleep Disorders, Second Edition (ICSD-2), including PSG [16].

Thoracoabdominal strain gauges and oronasal canula were used to monitor respiration, and a transcutaneous finger pulse oximeter was used to measure oxygen saturation. An apnea was defined as a reduction by at least 90% of the baseline airflow signal lasting 10 s or more, and a hypopnea as at least a 30% reduction (from baseline) in airflow lasting 10 s or more accompanied by an oxygen desaturation of 4% or more or an arousal [17]. The apnea-hypopnea index (AHI) was obtained by summing the number of apneas and hypopneas and then dividing by the number of sleep hours.

Patient assessment

A neurologist (RBP) and a neuropsychologist (JFG) specialized in movement disorders performed a comprehensive assessment of each patient’s disease status “on medication” state and blinded to the PSG assessment. Outcomes were broad, and included overall disease severity, all sections of the Unified Parkinson Disease Rating Scale (UPDRS) [18], quantitative motor testing (Purdue Pegboard, Alternate Tap Test, and Timed Up and Go), definitions of motor subtypes, motor complications of therapy, autonomic symptoms and signs, and olfaction (University of Pennsylvania Smell Identification Test) and color discrimination (Farnsworth–Munsell 100 hue test). These assessment procedures are described in detail elsewhere [8]. The neuropsychological assessment included cognitive tests and mild cognitive impairment (MCI) diagnostic criteria described in detail elsewhere [9]. Questionnaires for mood (Beck Depression Inventory-II [19] and Beck Anxiety Inventory [20]), cognition (Cognitive Failure Questionnaire [21]), and sleep complaints (Epworth Sleepiness Scale [22] and Insomnia Severity Index [23]) were also administered.

Statistical analysis

Characteristics were compared between groups (PD patients with and without sleep apnea) using two-sided chi-square tests, t-tests, or Mann-Whitney tests as appropriate. Analysis of variance was used to compare continuous measures between groups, adjusted for known potential confounders, including age, sex, and disease duration. Because the criteria for apnea definition vary widely in the literature, and to compare with the results in the literature, three different analyses were performed: with apnea defined as AHI >10 (primary analysis), AHI >5, and AHI >15 (secondary analyses).

RESULTS

Demographic, clinical, and sleep variables

Ninety-two PD patients were included in the study (35 women; mean age: 65.6 ± 8.7 years; mean disease duration: 5.9 ± 4.1 years; mean Hoehn and Yahr stage: 2.4 ± 0.9; mean “on” UPDRS-III: 23.22 ± 10.09). Using AHI >10, 19 PD patients had sleep apnea (21%). The prevalence of sleep apnea with AHI >5 was 33% , and 11% using AHI >15. No statistical differences in age or sex were found between the two groups (Table 1). PD patients without sleep apnea had slightly higher body mass index. Disease duration, Hoehn and Yarh stage, UPDRS part II and part III, levodopa equivalent dose, and percentage use of levodopa, dopamine agonist, or other antiparkinsonian medications did not differ between groups.

Concerning the sleep variables (Table 2), PD patients with sleep apnea had higher sleep latency and microarousal index and lower slow-wave sleep percentage than PD patients without sleep apnea. Except for nocturnal respiratory abnormalities no other sleep variables, including PSG and REM sleep EMG measures, insomnia symptom severity, excessive daytime sleepiness, and RBD frequency, differed statistically between groups. Results were similar using AHI >5 and AHI >15 (Supplementary Tables 1 and 2).

Motor measures

PD patients with sleep apnea reported fewer tremors than PD patients without apnea, and PD patients with apnea reported tremors as a first symptom more often than PD patients without sleep apnea did. No other motor-related variables differed between groups. Moreover, there were no between-group differences in any quantitative measure of motor functions (Table 3).

Autonomic, sensory, mood, and cognitive measures

The only significant difference in autonomic variables was higher supine diastolic pressure in PD patients with sleep apnea. No other autonomic signs differed between groups. No between-group differences were found in olfaction, color discrimination, or severity of depressive and anxiety symptoms. No cognitive measures, including the proportion of PD patients with MCI, differed between groups (Table 4).

DISCUSSION

The objective of the study was to compare motor and non-motor manifestations of PD patients with and without sleep apnea. Despite few minor impacts on some sleep parameters, tremor symptoms and diastolic blood pressure, our study was striking in its absence of relationships between sleep apnea and all major outcomes in PD, including motor, autonomic, sensory, mood and cognitive symptoms. Our results confirm those reported recently that sleep apnea seems to be not associated with PD features [24]. However, our study is the first to assess the relationship between sleep apnea and PD manifestations using a comprehensive assessment of a wide range of motor and non-motor symptoms. It allowed us to draw a relatively complete portrait of PD patients suffering from sleep apnea, which led a relatively clear conclusion; PD people suffering from sleep apnea are not notably different than those without sleep apnea.

Considering the well-documented effects of sleep apnea on sleep quality and health in the general population, the lack of similar relationships in PD is surprising. This could suggest that sleep apnea in PD is predominantly an epiphenomenon, and does not warrant further investigation or treatment, unless very severe. On the other hand, this is an observational study, and it is possible that an unmeasured variable could confound the relationship. For example, one could propose that if severe airway rigidity protects against apnea, then those with more severe rigidity would normally have more severe manifestations, which were not evident because of an otherwise beneficial absence of sleep apnea - if so, those without sleep apnea had a more benign outcome, which would be evident if the sleep apnea were treated. However, such confounds would be highly speculative and would require a strong relationship with a variable that was completely unmeasured in an otherwise very comprehensive evaluation.

The prevalence of sleep apnea in PD varies widely depending of the PD population studied and the diagnosis criteria used for sleep apnea [2, 7]. We found a sleep apnea prevalence from 33% (AHI >5) to 11% (AHI >15) in our PD cohort. Other studies in PD found a sleep apnea prevalence from 27 to 44% (AHI >5) [7, 24–26] or from 11 to 22% (AHI >15) [7, 24–27]. When considering confidence interval, our sample is similar to what is found in the literature. It is now well recognized that the prevalence of sleep apnea in PD is similar to the one found in the general population [1, 24–27]. These results confirm that sleep apnea is not more prevalent in PD, unlike other sleep disorders such as RBD, excessive daytime sleepiness or insomnia [2, 6].

Our study has some limitations. Although our sample was reasonably large, the sample size would not be powered to find subtle differences in some variables. Moreover, we do not have neuroimaging measures on all PD patients, so could not assess potential effects of sleep apnea on asymptomatic vascular brain disease or patterns of metabolic cerebral activity. One of the limitation that is important to highlight in that study is the fact that we do not have control group which limit us to compare with general population and to conclude about the predisposition to apnea in patients. Only hypothesis can be done.

In conclusion, despite the known impact of sleep apnea in the general population, sleep apnea in PD does not appear to be associated with differences in motor and non-motor manifestations.

CONFLICTS OF INTEREST

The authors declare that there are no potential conflicts of interest concerning this research article.

FUNDING SOURCES FOR THIS STUDY

This study was supported by grants from the Canadian Institutes of Health Research (J-F. Gagnon, R.B. Postuma) and the Fonds de Recherche du Québec – Santé (J-F. Gagnon, R.B. Postuma). S-G. Béland, J-A Bertrand, and V. Latreille were supported by a scholarship from the Canadian Institutes of Health Research. J-F. Gagnon holds a Canadian Research Chair on Cognitive Decline in Pathological Aging.