Abstract
Objective: To investigate the potential causes of excessive daytime sleepiness in patients with PD—poor sleep quality, abnormal sleep–wakefulness control, and treatment with dopaminergic agents.
Methods: The authors performed night-time polysomnography and daytime multiple sleep latency tests in 54 consecutive levodopa-treated patients with PD referred for sleepiness, 27 of whom were also receiving dopaminergic agonists.
Results: Sleep latency was 6.3 ± 0.6 minutes (normal >8 minutes), and the Epworth Sleepiness score was 14.3 ± 4.1 (normal <10). A narcolepsy-like phenotype (≥2 sleep-onset REM periods) was found in 39% of the patients, who were sleepier (4.6 ± 0.9 minutes) than the other 61% of patients (7.4 ± 0.7 minutes). Periodic leg movement syndromes were rare (15%, range 16 to 43/h), but obstructive sleep apnea–hypopnea syndromes were frequent (20% of patients had an apnea–hypopnea index >15/h; range 15.1 to 50.0). Severity of sleepiness was weakly correlated with Epworth Sleepiness score (r = −0.34) and daily dose of levodopa (r = 0.30) but not with dopamine-agonist treatment, age, disease duration, parkinsonian motor disability, total sleep time, periodic leg movement, apnea–hypopnea, or arousal indices.
Conclusions: In patients with PD preselected for sleepiness, severity of sleepiness was not dependent on nocturnal sleep abnormalities, motor and cognitive impairment, or antiparkinsonian treatment. The results suggest that sleepiness—sudden onset of sleep—does not result from pharmacotherapy but is related to the pathology of PD.
Excessive daytime sleepiness (EDS), which may affect 15% of patients with PD,1 had received little attention from clinicians until the recent reports of patients with PD falling asleep while driving.2-4⇓⇓ Because two-thirds of patients with PD may have night-time sleep disturbances (including insomnia, periodic leg movements, and sleep apnea),5 these may be responsible for EDS. Alternatively, abnormal sleepiness could result from treatment with levodopa and dopamine agonists.2-4,6⇓⇓⇓ Finally, the existence of a narcolepsy-like disorder, as observed recently in patients with PD,7-9⇓⇓ cannot be excluded.
Surprisingly, widespread awareness of the frequency and the severity of EDS in PD contrasts with the paucity of objective measures of sleepiness and its potential nocturnal and diurnal causes. The frequency of insomnia, sleep-disordered breathing, periodic leg movement syndrome, and narcolepsy-like patterns in somnolent patients with PD is not known. Therefore, we undertook a prospective study of sleep and daytime sleepiness in patients with PD presenting with sleep problems, using polysomnography and multiple sleep latency tests (MSLT).
Patients and methods.
Patients.
Between March 1998 and January 2001, 62 consecutive patients were prospectively examined for sleepiness. Inclusion criteria included symptoms of idiopathic PD10 and complaints of EDS. Three patients did not wish to participate in the study, two agitated and confused patients were disconnected from the polygraph, and three were excluded because they were severely demented and not responsive to levodopa. Thus, 54 patients (10 women, 44 men) completed the study. Polysomnography and MSLT were clinically indicated in sleepy patients, thereby precluding the need for submission of the trial design to the local ethics committee.
Demographic and clinical characteristics of the patients studied are summarized in table 1. Mean body mass index (body weight/height2) was 25.0 ± 4.7 kg/m2; 15% of the patients were considered to be obese. Hoehn and Yahr Disability scores11 were 1 (n = 1), 1.5 (n = 4), 2 (n = 16), 2.5 (n = 7), 3 (n = 11), and 4 (n = 15) when treated. Thirteen patients (24%) reported vivid dreams, 9 (17%) benign hallucinations, 13 (24%) frequent hallucinations, and 9 (17%) had delusions. Thirty-three patients (61%) had motor fluctuations, and 31 patients (57%) had dyskinesias. All patients were receiving long-term treatment with levodopa. Twenty-seven patients (50%) were also receiving dopamine agonists as adjunct treatment: bromocriptine (n = 19), piribedil (n = 5), pergolide (n = 3), apomorphine (n = 2), lisuride (n = 1), and ropinirole (n = 1). Four patients were treated concomitantly with two dopamine agonists. Other antiparkinsonian drugs included selegiline (n = 1), amantadine (n = 1), anticholinergics (n = 1), and entacapone (n = 10). Potential sedatives included benzodiazepines (n = 15), zopiclone (n = 4), antidepressant drugs (n = 13), and codeine (n = 2). Eight of the 28 patients taking psychotropic medication took a sedative drug during the day. All drugs had been taken at stable doses for at least 1 month before the sleep study.
Clinical profile of the 54 levodopa-treated patients with PD referred for somnolence
Methods.
Each patient was studied for 24 hours. Assessments included the Unified PD Rating Scale score for motor disability, evaluated when motor performance under treatment was optimal (“on” condition);11 interviews about sleep disorders; Epworth Sleepiness score;12 Mini-Mental State Examination score;13 night-time sleep recordings from lights off (ad lib) to 6:30 am; and MSLT.14 The class-II human leukocyte antigen phenotype was determined in 22 patients to rule out primary narcolepsy.
Night-time polysomnographic recordings included fronto-central and occipito-central bipolar electroencephalography, ocular electro-oculogram, chin and bilateral tibialis anterior surface electromyography, nasal pressure cannula (n = 10), naso-oral thermistance (n = 44), respiratory movements (using thoracic and abdominal belts), pulse rate, and pulse oximetry. After the night study, MSLT were performed, during which the patients were allowed to nap for 20 minutes at 2-hour intervals (at 8 am, 10 am, noon, 2 pm and 4 pm). Each test was terminated 20 minutes after lights off.
Sleep stages, arousals, periodic leg movements, and respiratory events were scored visually during 20-second periods according to standard criteria.15-18⇓⇓⇓ Because defining sleep onset, REM sleep, and sleep-onset REM may be difficult in parkinsonism, recordings were scored by physicians experienced in sleep neurology (E.K., I.A., M.M.A.), with findings and interrater reliability similar to those of other experienced teams.19 REM sleep without atonia was defined as periods with REM and saw-tooth waves or theta activity on the EEG. Occurrence of the first REM was used to determine the onset of an REM sleep period. Termination of an REM sleep period was identified either by the occurrence of an EEG profile characteristic of another stage (K complex, spindle, awakening) or by the absence of REM for 3 consecutive minutes. Stage 2 with REM was identified by the presence of K complexes and spindles on a background theta activity combined with REM.
Data analyses.
The main criterion was the mean daytime sleep latency (MSL). Unpaired Student’s t-tests were used to compare qualitative variables and correlation coefficients for quantitative variables. Multivariate analysis was performed using stepwise multiple regression. All computations were performed with SAS 6.12 software (SAS Institute, Cary, NC).
Results.
Daytime sleepiness.
The MSL (± SE) during the day, measured by MSLT, was 6.3 ± 0.6 minutes (range 0 to 19 minutes). MSL was less than 5 minutes in 27 patients (50%), 5 to 10 minutes in 16 patients (30%), and greater than 10 minutes (considered normal) in 11 patients (20%). Sleepiness was similar in women (8.2 ± 1.2 minutes) and men (5.9 ± 0.7 minutes; p = 0.16). The mean (± SD) Epworth Sleepiness score was 14.3 ± 4.1 (n = 51; range: 6 to 22; normal <10).20 MSL of the seven patients with “normal” Epworth Sleepiness scores was 6.1 ± 0.7 minutes; this is similar to the MSL of the other 47 patients (6.9 ± 1.2 minutes). There was a weak correlation between Epworth Sleepiness score and MSL (r = −0.34; p = 0.02). The false negative rate was 6% (three patients had an Epworth Sleepiness score <10 and an MSL <8 minutes), and the false positive rate was 18% (nine patients had an Epworth Sleepiness score ≥10 and an MSL >10 minutes).
Narcolepsy-like phenotype.
REM sleep was observed during one test in eight patients, two tests in four patients, three tests in eight patients, four tests in six patients, and five tests in three patients. Thus, 21 patients (39% of the group) met the polysomnographic criterion used to define narcolepsy (i.e., two or more sleep-onset REM periods during the five MSLT). These patients formed the narcolepsy-like group. MSL was shorter in the narcolepsy-like group (MSL ± SE: 4.6 ± 0.9 minutes) than in the non–narcolepsy-like group (MSL ± SE: 7.4 ± 0.7 minutes; p = 0.02). Patients in the narcolepsy-like group had hallucinations (62%) as often as the patients in the non–narcolepsy-like group (55%). There were no differences between narcolepsy and non–narcolepsy-like groups for Epworth Sleepiness score, age, duration of disease, Mini-Mental State Examination score, Unified PD Rating Scale Motor Disability score, night-time sleep (total sleep time, REM sleep duration, wakefulness after sleep onset, arousal index, apnea–hypopnea index), daily dose of levodopa, or use and dose of dopamine agonists (table 2). There was a trend for REM sleep latency during night-time sleep to be shorter in the narcolepsy-like group (84 ± 83 minutes) compared with the non–narcolepsy-like group (146 ± 141 minutes; p = 0.07). Human leukocyte antigen DQB1 602 was found in three of the 22 patients in whom the examination was performed (two of the patients in the narcolepsy-like group).
Comparison of clinical, treatment, and sleep characteristics of patients with and without a narcolepsy-like phenotype
Night-time sleep.
Night-time sleep recordings showed continuous bursts of ocular movements during stage 2 in four patients in the narcolepsy-like group and extensive slow-wave sleep (>140 minutes) in 13 patients. Periodic leg movements were observed in eight patients (15%) with indices of 16 to 43 and related arousal indexes of 16 to 23. Although central or mixed apneas were extremely rare, obstructive apnea and hypopnea were frequently observed, with apnea–hypopnea indices of 0 to 5 (n = 27), 5.1 to 15 (mild, n = 15), 15.1 to 30 (moderate, n = 6), and greater than 30 (severe, n = 5). Overall, 11 patients (20%) had moderate to severe obstructive sleep apnea syndromes. These patients had body mass indices (25.3 ± 5.2 kg/m2) and an MSL (5.3 ± 0.8 minutes) similar to those of the other 80% of patients with mild or no sleep apnea syndromes (24.8 ± 4.6 kg/m2; MSL: 7.3 ± 0.9 minutes). Treatment with continuous positive airway pressure was proposed for the five patients with severe sleep apnea syndromes.
Relationships between objective daytime sleepiness and clinical status, treatment, and sleep characteristics.
There was no correlation between MSL and age, disease duration, Mini-Mental State Examination score, Hoehn and Yahr Disability score, or Unified PD Rating Scale Motor Disability score (table 3), or between MSL and total sleep time, sleep efficiency, arousal index, stage 1, stage 2, slow-wave sleep, REM sleep percentage and duration, apnea–hypopnea, and periodic leg movement indices (see table 3). MSL was 5.8 ± 0.8 minutes in the group of patients treated with levodopa and dopamine agonists, and 6.8 ± 0.9 minutes in the group of patients treated with levodopa alone (p = 0.44, figure). There was no correlation between MSL and daily dose of dopamine agonists (bromocriptine equivalent; r = −0.17, p = 0.20) or total daily dose of levodopa equivalent (r = 0.17, p = 0.20, see table 3). However, there was a weak positive correlation between MSL and daily dose of levodopa (r = 0.3, p = 0.03), which was also identified as a factor by multiple regression analysis (p = 0.03).
Correlations between objective daytime sleepiness and clinical, treatment, and sleep characteristics
Figure. Effect of dopamine-agonist treatment on excessive daytime sleepiness. Patients treated with levodopa alone had mean daytime sleep latencies similar to those of patients treated with levodopa and dopamine agonists.
Discussion.
This is the first systematic study of sleepiness in patients with PD. Importantly, the patients studied were representative in all respects (including the relative proportions of antiparkinsonian treatments used) of the PD population referred for sleep problems, except for a predominance of men and patients with advanced PD. Potential causes of EDS—such as age, cognitive impairment, motor disability, use and dose of dopamine agonists, and short, fragmented nocturnal sleep (caused by sleep apnea syndrome or periodic leg movements)—were not associated with severity of daytime sleepiness.
As a group, the patients in this study, who were preselected for clinical sleepiness, were more somnolent (sleep latency, 6 minutes) than the group of 27 unselected patients with PD (sleep latency, 11 minutes) and age-matched controls (sleep latency, 12 minutes) studied by Rye et al.8 Because patients with a short MSL may be more prone to accidents when driving, it is important to identify short MSL as early as possible. Sleepiness questionnaires, such as the Epworth Sleepiness scale, could be used.12 In the current study, Epworth Sleepiness score was correlated with MSL, but the correlation was weak and there were false negatives. This suggests that a normal Epworth Sleepiness score does not exclude the presence of clinically significant sleepiness.
Potential causes of EDS include poor sleep quality, abnormal sleep-wakefulness control innate to PD, and perhaps induction with dopaminergic antiparkinsonian medications. However, in our series none of the usual measures of sleep disturbance (total sleep time, wakefulness after sleep onset, arousal index, duration of slow-wave sleep) correlated with severity of sleepiness as measured by MSL. These observations are consistent with those of Rye et al.8 in a nonselected population of patients with PD and contrast with the accepted notion that insomnia and sleep fragmentation are responsible for daytime sleepiness.21
Sleep fragmentation secondary to periodic leg movement syndrome is another potential cause of abnormal sleepiness. It has been suggested that it affects as many as one-third of patients with PD.22 This was not the case in our study, in which only 15% of the sleepy patients presented with periodic leg movement syndrome and related arousal indices were not severe. In addition, there was no correlation between periodic leg movement index and MSL. Consequently, periodic leg movements do not appear to be a significant cause of sleepiness in our group. It still remains to be proved that arresting leg movements will alleviate sleepiness.
Sleep-disordered breathing has been reported as another cause of sleepiness in the elderly.23 Based on the international definition,18 20% of the patients in this study had a moderate to severe obstructive sleep apnea syndrome. This is, to our knowledge, the first time that the prevalence of sleep-disordered breathing has been measured in sleepy patients with PD. This prevalence is greater than that found in an elderly American population (2.5 to 4.4%).24 Obesity, which is a classic risk factor for sleep-disordered breathing, was rare in our group and cannot in itself account for the high prevalence of sleep apnea in somnolent patients with PD. Upper airway dysfunction may occur in as many as 24% of patients with PD25,26⇓ and could cause obstructive apneas in our patients, possibly because of nocturnal akinesia of upper airway muscles. In contrast to previous studies in patients with obstructive sleep apnea syndrome,27 there was no correlation between apnea–hypopnea index and MSL in our group. This suggests that obstructive apnea syndrome itself does not contribute to severity of sleepiness in patients with PD. However, the alleviation of sleepiness with positive airway pressure would help to establish a causal relationship between these phenomena. This should encourage neurologists to perform polysomnography in sleepy patients with PD.
Severity of sleepiness in patients treated with levodopa alone was similar to that in patients treated with a dopamine agonist (mainly bromocriptine) and levodopa together. Because most patients were treated with D1–D2 agonists (bromocriptine, pergolide, apomorphine, lisuride, and piribedil, a D1–D2–D3 mixed agonist), a causal role of D2–D3-specific agonists such as ropinirole (pramipexole is unavailable in France) in excessive sleepiness cannot be completely excluded. Although ropinirole was widely prescribed in our center during the period of the study, only one patient on ropinirole had EDS, suggesting that this drug per se does not cause more (and may even cause less) sleepiness than other dopamine agonists. In addition, there was no correlation between sleepiness and dopamine agonist or levodopa-equivalent daily doses. Absence of dose-related sleepiness is a strong argument against an effect of this class of drugs on the mechanisms of sleepiness and suggests that individual characteristics of the patients predominate. However, daily dose of levodopa was inversely (and weakly) correlated with severity of sleepiness: The higher the dose of levodopa, the more alert the patient. This suggests that levodopa could have, at least in some patients of the group, vigilance-enhancing properties.
A narcolepsy-like phenotype was described previously in 60% of the patients with PD who had severe hallucinations.9 In this new series of patients, preselected for presence of significant clinical sleepiness rather than hallucinations, secondary narcolepsy was present in 40% of the group. These patients were sleepier than the patients without secondary narcolepsy and, except for two patients, they were human leukocyte antigen–negative. This pattern may be secondary to night-time REM sleep deprivation, sleep apnea syndrome, or drug treatment. However, we have shown that REM sleep duration, apnea–hypopnea index, and type and doses of drugs were similar in narcolepsy-like patients and in other patients. Other REM sleep abnormalities observed included the presence of REM during stage 2 sleep in some patients in the narcolepsy-like group, a pattern also described in primary narcolepsy. Hallucinations were no more frequent in this subgroup, suggesting that REM sleep control was abnormal in patients with PD (particularly in those with EDS or with severe hallucinations). This abnormal control would include increased daytime REM sleep pressure, responsible for sleep attacks, and dissociation between REM sleep and wakefulness, responsible for REM sleep behavior disorder and hypnagogic hallucinations. Abnormal REM sleep pressure during the day may be inherent in the disease. This is consistent with a previous case report of an untreated juvenile patient with PD and narco-lepsy7; with lesions in the subcoeruleus nucleus, the REM sleep generator, found in a patient with PD and secondary narcolepsy9; and with narcolepsy-like phenotypes unrelated to any identifiable cause. Finally, the subgroup of patients with PD and secondary narcolepsy, which unfortunately cannot be identified without sleep recordings, may be more prone to sudden sleep onset during the day because they were sleepier than the others.
When analyzing the potential causes of sleepiness in our series, we found none of the expected cause of sleepiness, except sleep-disordered breathing. The results challenge the idea that, in clinical practice, increasing the length of night-time sleep (for example using benzodiazepines or antidepressive drugs) will alleviate daytime sleepiness in patients with PD. They also suggest that individual characteristics of each patient predominate over other potential factors that might cause sleepiness (aging, poor sleep at night, features of the disease, antiparkinsonian therapy) and result from lesions effecting expression of thalamocortical arousal states, such as wakefulness and REM sleep. Finally, results of this study indicate, at least in this series of patients selected for sleepiness, that the reestablishment of normal dopaminergic transmission with levodopa or dopamine agonists does not make a major contribution to sleepiness—the sudden onset of sleep—in PD. Rather, agents acting to enhance dopaminergic transmission in surviving dopaminergic circuits may, in fact, permit greater alertness.
Acknowledgments
Supported by an unrestricted grant from Inserm–SmithKline–Beecham Pharmaceuticals 00/146A (M.M.-A.).
- Received September 12, 2001.
- Accepted December 24, 2001.
References
Letters: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
You May Also be Interested in
Hemiplegic Migraine Associated With PRRT2 Variations A Clinical and Genetic Study
Dr. Robert Shapiro and Dr. Amynah Pradhan