Acting out dreams...a harbinger of Parkinson's disease?

Neurodegenerative diseases are among the most difficult to manage, both for the doctor and patient. There is a slow but progressive accumulation of disability and functional decline. The qualities that make you you, start to dissolve - memories, communication, volition, cognition, even personality.

A hallmark of these conditions is irreversibility: once neurones are lost they are lost and cannot be regenerated. At best, we might be able to slow the decline. For this reason, there is a drive to identify neurodegenerative disorders at their earliest stage, when slowing the disease process can have most impact.

In Alzheimer’s disease, neuropathology accumulates from years to decades before a clinical diagnosis is made. Prior to diagnosis, a patient might start to experience subtle difficulties in memory and cognition. Interventions to delay or prevent the development of Alzheimer’s disease have been targeted towards this pre-diagnosis group who exhibit subtle symptoms, termed ‘mild cognitive impairment’.

However, there are problems with the concept of mild cognitive impairment. Although mild impairment increases the risk of Alzheimer’s disease, only 10-15% go on to be diagnosed with dementia each year. Furthermore, roughly the same amount revert to normal cognitive functioning in the same time period. Many who experienced mild cognitive impairment will never be diagnosed with Alzheimer’s disease or any other neurodegenerative disorder. It is therefore problematic to target treatments (with inevitable side-effects) at individuals who may not actually be on the road to developing the disease.

What we need, is a pre-diagnosis (prodromal) stage of illness which is early enough that the disease is mild but specific enough that we don’t include otherwise healthy people who would never actually develop the disorder. ‘REM sleep behaviour disorder’ seems to represent this for Parkinson’s disease, a genuinely prodromal stage of illness.

Stages of sleep

Sleep is an essential process for (almost) all animals. It is important for memory consolidation, growth, and removal of toxins. Sleep can be divided into REM (rapid eye movement) and non-REM stages. Non-REM sleep is further divided according to brain activity as measured by EEG:

• Stage 1. Light sleep. Characterised by the transition from alpha waves (8-13 Hz) to theta waves (4-7 Hz). This is the stage between wakefulness and sleep. There may be some awareness of the environment. Hypnagogic hallucinations may be experienced.

• Stage 2. Characterised by theta waves and sleep spindles. Muscle activity decreases and there is no awareness of the environment.

• Stage 3. Deep sleep (slow wave sleep). Characterised by delta waves. The most restful phase of sleep. Moderate muscle tone but absent eye movement. This stage is associated with sleep-walking and bedwetting in children.

REM sleep, as the name suggests, is characterised by fast ocular movements but low muscle tone elsewhere in the body. Brain activity, as measured by EEG, shows similar oscillations to awake brain activity. Most notably this is the stage in which we dream.

During REM sleep we lose muscle tone (atonia), effectively paralysing the whole body. This is accomplished through the inhibition (hyperpolarisation) of motor neurones, raising the threshold which must be overcome to excite (depolarise) them. The midbrain, and specifically the medulla oblongata, seems to be crucial in this inhibition. We need to lose muscle tone to avoid responding to our dreams and inadvertently injuring ourselves or a partner while asleep.

Sometimes this ‘atonia’ goes awry. It can happen at the wrong time (while awake) during narcolepsy - causing sudden collapse. In Parkinson’s disease, patients lose the usual atonia during REM sleep, resulting in them physically acting out their dreams - termed REM sleep behaviour disorder.

REM sleep behaviour disorder

Chronic behavioural disorders of REM sleep were first described as a case series in 1986. The report describes a new parasomnia (sleep disorder) in which the patients enacted their dreams, often with punches and kicks, and in the process injured themselves or their spouse. The behaviours were absent in non-REM sleep and were associated with various neurological conditions in four of the five patients.

Dream enactment with limb movement can also occur in healthy people, particularly during heightened emotional states or withdrawal from alcohol or sedatives. This usually takes places during the transition from REM sleep to wakefulness, with atonia maintained for most of the REM sleep stage. What distinguishes REM sleep behavioural disorder is that patients will continue REM sleep before and after their dream enactment, and will experience this chronically (for more than six months).

Sometimes, researchers make the distinction between REM sleep behavioural disorder due to drugs or another identifiable cause, and ‘idiopathic’ disorder where no obvious cause is apparent.

For a confirmed diagnosis of REM sleep behaviour disorder, four criteria must be met:

• Repeated episodes of sleep-related vocalisation or complex motor behaviours

• Episodes being present during REM sleep according to polysomnography or based on clinical history of dream enactment

• Polysomnographic recording demonstrating REM sleep without atonia

• Episodes not better explained by another sleep disorder, mental disorder, medication or substance abuse

An example of a patient experiencing REM sleep behavioural disorder is shown in this video. It is estimated to have a prevalence of ~1%, mainly in the over 50s. It poses risk of injury to the patient and their partner, with the disorder worsening over time.

What makes REM sleep behavioural disorder particularly interesting is its strong link with Parkinson’s disease and related syndromes such as Lewy body dementia, and multiple system atrophy. These are disorders caused by the accumulation of an abnormal protein, α-synuclein (they are collectively known as synucleinopathies).

The link with synucleinopathies

Pathological examination of the brains of people affected by REM sleep behaviour disorder on autopsy find evidence of synucleinopathy in around 95% of cases.

At baseline, 3/4 of people with REM sleep behaviour disorder meet criteria for prodromal Parkinson’s disease, for example showing subtle impairments in motor and cognitive domains.

Even more strikingly, at follow-up of five years, up to 1/3 will have a defined neurodegenerative disease, mostly Parkinson’s disease or Lewy body dementia. When follow-up is extended to 14 years, up to 90% of patients will be diagnosed with a neurodegenerative disease. The average time to diagnosis of a Parkinson’s syndrome is between four to nine years.

Why does the loss of muscle paralysis during dreaming precede the diagnosis of a Parkinson’s syndrome? The answer comes from the neuropathological staging of synucleinopathies. The brain regions where the abnormal α-synuclein accumulates tends to follow a sequential pattern, known as Braak staging:

Braak staging of Parkinson’s pathology. Pink/purple represent Stages 1 and 2. Blue represents Stages 3 and 4. Orange represents Stage 5. Reproduced under CC-BY.

As shown, the earliest stages of pathology tend to affect parts of the brainstem such as the medulla oblongata which would normally inhibit muscle movement during REM sleep. Later, midbrain structures (basal ganglia) that control movement become affected resulting in motor symptoms, as well as the cortex, resulting in cognitive problems.

Clinical implications

Making a confirmed diagnosis of REM sleep disorder poses some difficult clinical questions. How should the risk of neurodegenerative disorder be communicated? Whether the patient should be fully informed of the risk has been discussed. There are currently no proven treatments to delay or prevent a neurodegenerative disorder after the diagnosis of REM sleep disorder. Some neurologists argue that all patients should be fully informed and encouraged to participate in clinical research.

The mainstay of current pharmacological management is symptomatic rather than disease-modifying. First-line treatment is clonazepam, a long-acting benzodiazepine that acts on the GABA-A receptor, increasing inhibition of motor activity.

In the future, synucleinopathy treatments that target the disease pathology (such as anti-α-synuclein antibodies) during the REM sleep behaviour disorder may modify disease progression. Any slowing of neurodegeneration could delay the development of later motor or cognitive symptoms associated with Parkinson’s syndromes.

Whether improving the sleep cycle itself could be a disease-modifying target is another prospect. Greater amplitude of slow waves during sleep is associated with slower motor progression in Parkinson’s disease.

In a mouse model, enhancing slow wave sleep lead to reduced accumulation of the α-synuclein protein. Glymphatic clearance of these proteins during deep sleep is one plausible mechanism.

Drugs such as sodium oxybate that promote slow-wave sleep are under investigation, with some promising initial results.

Final thoughts

REM sleep behaviour disorder is a true prodrome for Parkinson’s disease and related conditions. It offers an opportunity to detect these syndromes at an early stage and the best chance to slow disease progression. This feels like real progress in neurodegenerative disorders, which may lead to improved treatment and patient outcomes.

I can’t help but compare this progress to that of psychosis research, where much of the focus of the last 20 years has been in identifying individuals in the earliest stages of psychosis (putatively, ‘prodromal’). In contrast to REM sleep behaviour disorder, the transition rate from those at ‘Clinical High Risk for Psychosis’ to frank psychosis is still around 1/3 at 10 years follow-up.

Clearly, as in psychiatry generally, our ‘at risk’ syndromes are more heterogenous, less specific, and lack well-defined pathological correlates.