January 20, 2016
Department of Neurology,
Leiden University Medical Center, Leiden, The Netherlands
Highlights
•EDS
is a non-persistent symptom in patients with Parkinson’s disease.
•With
longer disease duration, a large proportion of PD patients develops EDS.
•A
higher dopamine agonist dose is associated with higher EDS scores at follow-up.
•The
use of antihypertensives is associated with higher EDS scores during follow-up.
•Patients
with the PIGD phenotype are at increased risk of developing EDS.
Abstract
Introduction
Excessive daytime
sleepiness (EDS) is a common feature of Parkinson’s disease (PD) that
contributes to the disease burden and increases risk of harm. The aim of this
study was to examine persistency, cross-sectional and longitudinal
associations, and risk factors for EDS in patients with PD.
Methods
Analyses were performed on
data from the SCOPA-PROPARK cohort, a 5-year hospital-based longitudinal cohort
of over 400 PD patients who were examined annually. Cross-sectional analyses
were conducted to evaluate differences between patients with and without EDS at
baseline, while linear mixed models using data of all patients were used to identify
factors associated with longitudinal changes in SCOPA-SLEEP-Daytime Sleepiness
(SCOPA-SLEEP-DS) scores. A survival analysis was done using data of patients
without EDS at baseline to identify risk factors for future EDS.
Results
EDS proved a
non-persistent symptom, although persistency and the proportion of patients
with EDS increased with longer follow-up. At baseline 43% of patients had EDS,
while 46% of patients without EDS at baseline developed this symptom during
follow-up. Male gender, poorer nighttime sleep, cognitive and autonomic
dysfunction, hallucinations, less severe dyskinesias, dose of dopamine agonists
and use of antihypertensives were associated with higher EDS scores over time,
while use of benzodiazepines was associated with lower scores. Baseline
SCOPA-SLEEP-DS score and PIGD phenotype were risk factors for future EDS.
Conclusion
With longer disease
duration a large proportion of patients develop EDS. Some risk factors are
modifiable and patients should be monitored to improve quality of life and
reduce risk of harm.
http://www.prd-journal.com/article/S1353-8020(16)30020-7/abstract
Oman Med
J. 2015 Jan; 30(1): 3–10.
PMCID: PMC4371466
Excessive
Daytime Sleepiness and Unintended Sleep Episodes Associated with Parkinson’s
Disease
This
article looks at the issues of excessive daytime sleepiness and unintended
sleep episodes in patients with Parkinson’s disease (PD) and explores the reasons
why patients might suffer from these symptoms, and what steps could be taken to
manage them. During the last decade, understanding of sleep/wake regulation has
increased. Several brainstem nuclei and their communication pathways in the
ascending arousing system through the hypothalamus and thalamus to the cortex
play key roles in sleep disorders. Insomnia is the most common sleep disorder
in PD patients, and excessive daytime sleepiness is also common. Excessive
daytime sleepiness affects up to 50% of PD patients and a growing body of
research has established this sleep disturbance as a marker of preclinical and
premotor PD. It is a frequent and highly persistent feature in PD, with
multifactorial underlying pathophysiology. Both age and disease-related
disturbances of sleep-wake regulation contribute to hypersomnia in PD.
Treatment with dopamine agonists also contribute to excessive daytime
sleepiness. Effective management of sleep disturbances and excessive daytime
sleepiness can greatly improve the quality of life for patients with PD.
Keywords:
Parkinson
Disease, Sleep, Dopamine Agonists, Pedunculopontine Nucleus
Introduction
Sleep
disturbances in the late stages of Parkinson’s disease (PD) were recognized by
James Parkinson1 in his classic monograph
noting that: "The sleep becomes much disturbed. The tremulous motions of
the limbs occur during sleep, and augment until they awaken the patient, and
frequently with much agitation and alarm…..and at the last, constant
sleepiness, with slight delirium". The diagnosis of PD requires the
identification of its cardinal features, which are motor symptoms. Diagnosis is
impossible without them, but recognition of the importance of non-motor
features has increased over the past years.2
Non-motor features (which include autonomic nervous system dysfunction,
disorders of cognition and mood, psychosis, pain, loss of smell, and fatigue)
affect nearly all PD patients, appear early in the course of PD, and contribute
to excessive daytime sleepiness (EDS). All of these symptoms have significant
adverse effects on the quality of life (QoL) of both patients and caregivers
and require proper identification and treatment.3-7
Clinical presentation of sleep disturbance
Sleep-related
problems in PD can be divided into disturbances of sleep and disturbances of
wakefulness. Disturbances of sleep include insomnia, restless leg syndrome
(RLS), rapid eye movement sleep behavior disorder (RBD), sleep apnea, and
parasomnias. Disturbances of wakefulness include EDS, and sleep attacks. With
normal aging, there is disruption of normal sleep architecture and alterations
in the normal circadian rhythm leading to impaired nocturnal sleep and EDS.8,9 These problems are
accentuated in PD patients, with 60% to 90% having some form of sleep
disturbance, particularly in the advanced stages of the disease.3,6,10-12
Epidemiology of sleep disturbance in Parkison’s disease
The
prevalence of sleep disturbance in PD is difficult to ascertain due to the
heterogeneity of patients and different criteria used to categorize sleep
disturbances. There is paucity of data on the role of gender in sleep
disturbances. Smith and colleagues,13 studied 153 patients
and their spouses, and reported that sleep disturbances occurred more frequently
in females with PD (41%) than in men (25%). However, there was no sex
difference for difficulty initiating sleep. Van Hilten and colleagues,14
observed that female patients experienced more difficulty maintaining sleep
(87.5%) and excessive dreaming (68.4%) than males (64% and 31.6%,
respectively). Sleep dysfunction in PD usually manifests by difficulty in
initiating sleep, fragmented sleep, reversal of the sleep cycle, and EDS.15,16 EDS was assessed using
the Epworth scale in 101 patients with PD and 100 age-matched controls.17
EDS was detected in 76% of patients with PD compared to 47% of controls (p<0.050).
Nearly a quarter (24%) of patients with PD had scores in the diagnostic range
of narcolepsy, compared to only 5% of controls (p<0.001).
Sleep
disturbances in PD are numerous and there may be different combinations.6,14,18-23 The cause of the
disturbances are multifactorial and may be related to aging, Parkinsonian motor
dysfunction, dyskinesia, pain, nocturia, nightmares, dopaminergic and non-dopaminergic
medications, cognitive impairment, and a variety of specific sleep disorders,
including RLS, periodic limb movements of sleep (PLMS), RBD, and sleep apnea.
Collectively, they contribute to the increase in daytime sleepiness frequently
found in PD patients.24 EDS and RBD may be
harbingers of PD and other synucleinopathies, such as multiple system atrophy,25
and thus already present in the premotor phase of the disease. It is also clear
that dopaminergic medications and particularly dopamine agonists can have a
complex effect on sleep. Sometimes these medications cause insomnia, and their
sedative properties may contribute to daytime sleepiness.17,26-30 In other situations,
they improve the quality of sleep by improving nocturnal immobility.31,32Therefore, dopaminergic
medications can either improve or worsen sleep in PD patients.
Neuroanatomy of sleep in Parkinson’s disease
The
anatomical basis of sleep disturbances in PD is not fully understood, but it
likely involves degeneration of both dopaminergic and non-dopaminergic systems.
Sleep disturbances are primarily due to the progressive disease process
impairing thalamocortical arousal and affecting sleep-regulating centers in the
brainstem. Secondary causes are nocturnal disease manifestations, and side
effects of pharmacological treatment. Mesocorticolimbic dopamine neurons that
project from the ventral tegmental area (VTA) targeting the thalamus,
hippocampus, and cerebral cortex are thought to be involved in the arousal
mechanism.33 Dopamine plays a
complex role in state control, specifically maintains the wake state, and
regulates sleep homeostasis.34 These dopamine-mediated
arousal functions are independent from the nigrostriatal dopaminergic system.
Subsequently, the responsible mesolimbic dopaminergic system may also
degenerate later than the nigrostriatal system.34
Aetiology of sleep disturbance in Parkinson’s disease
Non-dopaminergic
neurons have also been implicated in sleep dysfunction in PD. Reduced levels of
hypocretin in the cerebrospinal fluid are an established biomarker in
narcolepsy, and PD patients show narcolepsy-like sudden onset sleeps during the
daytime, suggesting similar hypocretin action in these patients. Braak and his
colleagues35 hypothesis of ascending
brainstem degeneration proposes early disease involvement of several other
non-dopaminergic brainstem nuclei, such as the cholinergic pedunculopontine
nucleus (PPN), serotonergic tegmental area, nucleus magnocellularis, and
noradrenergic locus cereleus (LC). Degeneration of neurons in these sleep-wake
related pathways (the flip-flop switch), which are associated with
thalamocortical arousal, could contribute to the development of sleep
dysfunction in PD.36 The PPN has attracted
particular attention because it is intimately related to the anatomic control
of sleep, and is thought to play a critical role in mediating inhibition of
voluntary muscles during REM.16,37 This hypothesis has
propelled our understanding of sleep dysfunction in PD. Interestingly, direct
evidence of a beneficial effect of a normally functioning PPN has been given by
deep brain stimulation of this nucleus, confirming that PPN promotes REM sleep
and plays a role in switching from one state to another. Thus, low-frequency
stimulation of the PPN increases alertness and high-frequency stimulation
induces non-rapid eye movement sleep (NREM) sleep, while sudden withdrawal of
the stimulation elucidates REM sleep.38-40
Excessive daytime sleepiness
EDS
is defined as a chronic state of inability to stay awake during the day. A
score greater than 10 on the Epworth Sleepiness Scale (ESS), or a mean sleep
latency less than eight minutes on the Multiple Sleep Latency Test (MSLT)30,41,42 is considered
inappropriate sleepiness during waking hours and has been under-recognized in
PD. EDS was initially considered a side effect of non-ergot dopamine D2-D3
agonists,43 but it is not
restricted to a specific class of dopaminomimetic agents and may have other
causes. Because of the many potential problems that can interfere with nocturnal
sleep in patients with PD and the tendency of dopaminergic medications to
induce sedation, EDS is a common problem.44,45
Epidemiology
One
study found no increase in the prevalence of EDS in untreated PD patients
compared with an age-matched healthy control group. EDS was more frequent in
treated patients, suggesting that either the progression of the disease, the
treatment, or a combination of both, may be critical in the development of this
symptom.46 Another study found
that progression of the disease, before initiation of dopaminergic treatment,
was associated with increased sleepiness.47 Polysomnographic
recordings indicate that the average patient with PD obtains only four to five
hours of documented sleep per night instead of the approximately eight hours
that are normally required.19,48 In one study, 76% of
consecutive PD patients reported EDS, compared with 47% of age-matched controls
(p<0.050) and 24% had sleep scores in the range of patients with
narcolepsy, compared to only 5% of controls (p<0.001).17
EDS is common in PD,6,14 however, it is a
multifaceted phenomenon not solely related to dopaminergic medication. Next to
dopaminergics, disease severity, "wearing-off", and sleep disordered
breathing have been shown to influence PD-related EDS.41
Putative biological markers
The
notion of PD-related EDS is supported by the fact that magnetic resonance
imaging (MRI) brain morphometry demonstrated that in PD patients EDS was
related to atrophy of the medial cerebellar peduncle (PD with EDS (mean+SD)
16.08+0.93mm vs. PD without EDS 17.82+0.80mm; p=0.010), leading the
authors to suggest the involvement of degeneration of the pontomedullary
respiratory centers in the development of PD-linked EDS.49
In one study,14 no significant
difference was found in the degeneration of the pontomedullary respiratory
centers between PD patients (44.4%) and control patients (31%). The diurnal
pattern was similar with a peak in the early afternoon. The authors concluded
that no relationship existed between PD and EDS, and that EDS was probably a
consequence of aging, as reported previously by Carskadon50
and Morewitz.51 EDS was noted in
patients with Parkinsonian syndromes in early descriptions52
and spontaneous dozing during the daytime occurred in nearly half of PD patients
in one study.6 However, EDS has only
received increasing attention since the controversially discussed report of
"sleep attacks" in PD patients on dopaminergic therapy. These case
reports first involved patients taking non-ergoline dopamine agonists43
and were subsequently supplemented by case reports for virtually all other
dopamine agonists and levodopa.
Clinical presentation
Patients
with EDS have a tendency to fall asleep in unintended situations. Typically,
these occur in relatively benign situations that are conducive to falling
asleep such as while watching television or reading. However, in extreme
situations patients may fall asleep during a meal, while in conversation, and
in potentially dangerous situations such as while driving. Previous studies
reported sleepiness with varying frequencies (42%14
and 49%6).
Diagnostic tools
To
identify sleepiness in an individual patient, it may be necessary to use sleep
questionnaires such as the Epworth Sleepiness Scale (ESS),5
which do not rely on subjective estimates of sleepiness, but rather on a
measure of the propensity of the patient to fall asleep. The ESS is a set of
eight questions, quick and easy to use for the patient and carer, and does not
require technical measurements or the involvement of a sleep laboratory. The
ESS has been shown to correlate with more cumbersome, expensive, and
time-consuming tests such as the Multiple Sleep Latency Test (MSLT) in patients
with sleep apnea.5 It is the sum of eight
items that ask for ratings on the tendency to doze in a variety of situations.
The ratings are scaled from zero (no chance of dozing) to three (high chance of
dozing) for each item. Higher scores indicate greater sleepiness as indicated
by a higher likelihood to fall asleep during daytime activities. The ESS has
been translated into different languages throughout the world. A validated
Arabic ESS questionnaire was just as good as its English counterpart.53
The
importance of addressing EDS in PD was highlighted by a report of eight
patients who suddenly fell asleep while driving a motor vehicle.43
These episodes were termed "sleep attacks" by the author because they
seemed to have occurred without warning, and were attributed to dopamine
agonists because they disappeared when the drugs were withdrawn. This report
generated intense interest in the nature and frequency of sleep disturbances in
PD and a debate as to how these episodes are related to the use of dopamine
agonists. It is generally thought that EDS in PD patients results from impaired
nocturnal sleep. However, not all studies confirm this concept. The FAST TRACK
study evaluated daytime sleepiness using the MSLT. In 27 PD patients, the MSLT
scores did not correlate with the quantity and quality of the previous night’s
sleep or other sleep architecture measures, such as sleep stage percentage, and
total sleep time.54 Similarly, in another
study, no correlation was found between MSLT score and total sleep time, sleep
efficiency, arousal index, apnea-hypopnea, or periodic leg movement indices.55
These studies suggest that the quality of nighttime sleep may not be the only
factor responsible for daytime sleepiness. Whatever the mechanism, EDS (defined
as being sleepy most of the day) is present in a large number of PD patients.
Varying estimates have been reported, ranging from 15% to 75%.17,56-63 The most widely used
tools are the ESS, MSLT, Scales for Outcomes in PD (SCOPA-SLEEP), Parkinson’s
Disease Sleep Scale (PDSS), and Polysomnography (PSG). The possibility that
dopaminergic medications, and especially dopamine agonists, may aggravate EDS
has attracted considerable attention, again driven by the observation by Frucht
and colleagues43 that all patients who
fell asleep while driving were receiving high doses of dopamine agonists. PSG
studies have similarly demonstrated that total dopaminergic dose, rather than
the specific dopaminergic agent, was the best predictor of EDS, as MSLT scores
of patients on different dopaminergic therapies were similar to one another.55
EDS may occur with use of other PD medications, including levodopa and
carbidopa. Seventy percent of dysautonomic patients with PD reported sleep
attacks compared to 17.8% of nondysautonomic patients with PD.29
Sleep attacks (unintended sleep episodes)
A
sleep attack is described as "an event of overwhelming sleepiness that
occurs without warning or with a prodrome that is sufficiently short or
overpowered to prevent the patient from taking appropriate protective
measure".43 Others have suggested
that sleep attacks in PD patients are more likely to represent an extreme form
of EDS due to the combination of a sleep disturbance and the sedative effects
of dopaminergic medication.28 Sleep attacks in which
patients fall asleep without an antecedent warning of sleepiness are not known
to occur either physiologically or in association with pathologic conditions.5
For this reason, the concept of a sleep attack has been abandoned even in
narcolepsy.64 It was proposed that
sleep attacks represent an extreme form of sedation in patients who were sleep
deprived and on sedative medications, and would be better termed unintended
sleep episodes (USE). The term sleep attack re-emerged in 1999 when Frucht et
al,43 described sudden
episodes of falling asleep that caused driving accidents. Some experts have
suggested that the term USE is a more appropriate description of these events,
arguing that the word "attack" fails to recognize the background of
sedation that may precede the onset of sleep.28,65 Patients experiencing
sleep attacks may fall asleep because they are continuously sleepy, and fall
asleep in situations where resistance to sleep is decreased.66
The concept of a sleep attack implies that the events are inevitable and occur
without any warning whatsoever. The notion of USE implies that at-risk
individuals can be identified and the episodes prevented by instituting
appropriate treatment measures. Prodromes of sleepiness include yawning,
blinking, or tearing.
Prevalence and risk factors
A
prospective survey of 236 patients with PD found that 72 (30.5%) reported
sudden sleep episodes.29 Another study, which
used structured telephone interviews in 2,952 patients with PD, found that 177
patients (6%) had sleep attacks.67 Ninety-one patients had
at least one sleep attack without a warning sign, while 86 patients always had
a warning sign prior to a sleep attack. Although sleep attacks were initially
described in patients receiving pramipexole and ropinirole, it is clear that
sedative effects and USE can be seen with any dopaminergic agents, including
levodopa,68-71 and that these effects
are dose related, occurring with greater frequency in patients taking
relatively high doses. Thus, somnolence is more likely to occur in patients
taking higher doses of dopaminergic medications and is greatest when a given
dose reaches its maximal concentration. The package insert for pramipexole in
the US recommends that patients must be informed that they should not drive a
vehicle or engage in potentially dangerous activities until they have enough
experience to determine whether pramipexole adversely affects their mental
performance.72 Sleep experts have
criticized the term sleep attacks saying it is inappropriate as sleepiness is
not adequately perceived, specifically in chronically sleepy patients, and
electrophysiological signs of sleepiness precede sleep onset even in patients
who are not aware of it.73,74 This was confirmed when
47 somnolent PD patients underwent MSLT, and after each nap they were asked
whether they had slept or dozed. Thirty-eight percent of the PD patients did
not perceive at least one PSG-confirmed nap. These patients also showed a lower
score on the subjective ESS. However, sleep-state misperception was no more
frequent than in control subjects who had other hypersomnias or sleep apnoea.75
These findings demonstrate that sleep misperception is a factum in PD patients
as it is in individuals with chronic sleepiness due to other conditions.73
Management approaches
Recent
years have seen several disappointments in neuroprotective therapy and it
remains an important challenge to understand why very promising drugs have
failed in human trials. Overall, sleep problems in PD remain a major
therapeutic challenge.
Management
of PD is complex, has to target underlying mechanisms, and comprises of
non-pharmacological and pharmacological approaches. The first step is to
identify at-risk patients.76 To accomplish this, the
physician or ancillary personnel must inquire about EDS from both the patient
and carer, who might provide a more objective assessment of the patient’s sleep
habits as patients may not recognize that they are sleepy having become
tolerant to the sensation of chronic tiredness. Therefore, management options
for EDS and unintended sleep episodes include ensuring correct diagnosis by
ruling out syncope, seizures, and cardiac disorder.
Non-pharmacological
approaches can be the mainstay of treatment for mild to moderate EDS. As
described, the use of a validated sleepiness scale, such as the ESS, provides a
quick and reliable assessment of sleepiness based on the propensity of the
patient to fall asleep in unintended situations, and does not rely on the
patient’s subjective awareness of whether or not they are sleepy. ESS scores
greater than 10 are considered to be in the sleepy range, and such patients are
at higher risk for experiencing unintended episodes of falling asleep.76
Further management options include introducing proper sleep hygiene,
eliminating unnecessary sedative medications, using the lowest dose of
dopaminergic medication that provides satisfactory clinical control,
identifying and treating sleep disorders, and counselling patients on risks of
daytime sleepiness and sudden sleep episodes.75
Patients with EDS should not drive a motor vehicle until this problem has been
corrected. Indeed, European agencies have suggested that patients with PD
taking dopamine agonists should not drive at all, although some experts believe
that this recommendation is too harsh and that patients may safely drive,
subject to specific treatment guidelines. PSG is the "gold standard"
method used to evaluate sleep disorders and provides detailed information about
actual sleep status. It can detect the co-occurrence of sleep apnea, restless
leg syndrome-periodic leg movement in sleep (RLS-PLMS), and RBD. MSLT is
helpful to quantify the severity of EDS. Good sleep hygiene is the cornerstone
of effective management of any sleep disorder.
The
management of Parkinsonian motor symptoms can be improved with the use of
dopaminergic agents. If alterations in dopaminergic medications fail to help
EDS, one can consider adding a wakefulness promoting agent like modafinil.
Usually, mechanisms are multiple and treatment multimodal. Modafinil is a
non-amphetamine drug well-established as a first-line, symptomatic treatment
for EDS associated with narcolepsy and, more recently, is proving to be a
useful agent in other medical conditions where EDS is a symptom. Whilst its
mode of action is yet to be explained, modafinil appears to exert its effects
specifically on the hypothalamus sleep-wake system, increasing wake promoting
neuronal activity in the tuberomammillary nucleus (TMN) and decreasing
sleep-promoting neuronal activity in the ventrolateral preoptic area (VLPO),
thus inducing "calm wakefulness".77
Early open-label reports were promising,78,79 but double-blind
controlled studies showed only modest80,81 or no benefit.82
In the clinic, the drug may be useful in treating EDS in individual patients
with PD. It should be started at a dose of 100mg and increased to 200–400mg per
day as necessary. Side effects include insomnia, head pains, and depression.
Depression should be evaluated and treated accordingly. Sometimes reassurance
with or without supplementary psychotherapy is sufficient, but most often
antidepressant medications are needed.
Further
areas of research are now also focusing on adenosine A2A receptor antagonistsodium
oxybate and caffeine to promote wakefulness. In the pursuit of improved
treatments for PD, the adenosine A2A receptor was used as an attractive
non-dopaminergic target.83 This was based on
compelling behavioral pharmacology and selective basal ganglia expression of
this G-protein-coupled receptor. Its antagonists crossed the threshold of
clinical development as adjunctive symptomatic treatment for relatively
advanced PD. Adenosine derived from the degradation of adenosine triphosphate
(ATP) or adenosine monophosphate (AMP) function as a signaling molecule in the
nervous system through the occupation of A1, A2, and A3 adenosine.84
Adenosine A2A receptors have a selective localization to the basal ganglia and
specifically to the indirect output pathway, and as a consequence offer a
unique opportunity to modulate the output from the striatum that is believed to
be critical to the occurrence of motor components of PD. Several studies
conducted worldwide report an inverse association between caffeine/coffee consumption
and the risk of developing PD.85 The association is
strong and consistent in men, but uncertain in women possibly because of an
interaction with hormone replacement therapy.86
Palacios et al,86 found that consumption
of decaffeinated coffee was not associated with PD risk.
Conclusion
Improving
patients’ QoL is a key factor to consider when reviewing PD treatment plans
since more than a half of PD patients report problems with sleep disturbances
more than the motor symptoms of the disease, and EDS and unintended sleepiness
has a large impact on the QoL of PD patients as well as their carers. There is
no doubt that non-motor aspects of PD are of unquestionable relevance. As of
today; however, options for their management are very limited. Prompt diagnosis
should become standard in clinical practice, and management a research
priority. Advice on good sleep hygiene is instrumental, as pharmacological
approaches have yet to provide consistent and reliable results without
significant adverse effects. The efficacy of pharmacological treatment of EDS
in PD using wakefulness-promoting drugs, such as modafinil, remains
controversial. Sleep attacks have been reported in patients taking dopamine
agonists for conditions other than PD.87
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4371466/
