Lewy Body Dementia and Acting Out Dreams: What the Link Means

01What dementia with Lewy bodies actually is

The name comes from a person. In 1912 the German neurologist Friedrich Lewy described abnormal protein deposits inside brain cells, and, as the National Institute on Aging explains in its guide to Lewy body dementia, those deposits are made of a protein called alpha-synuclein. Alpha-synuclein has ordinary jobs in a healthy brain, mostly at the junctions where nerve cells pass signals to each other. In this family of conditions it clumps together inside neurons, starting in particular regions, and the cells it collects in gradually stop working well.

Which regions get involved is the whole story of the illness. The NIA lists the areas affected in Lewy body dementia, and among them are the parts of the cortex used for thinking and planning, the areas involved in movement, the olfactory pathways used for recognizing smells, and the brainstem, which it describes as important in regulating sleep and maintaining alertness. Nerves outside the brain can be affected too, in the gut, the heart and the salivary glands, which is why constipation, dizziness on standing and drooling turn up in the same picture.

Clinically, the diagnosis rests on four core features, set out in the fourth consensus report of the DLB Consortium, published in Neurology in 2017 by McKeith and colleagues. Those features are fluctuating cognition, with visible variation in attention and alertness across a day; recurrent visual hallucinations that are typically well formed, often of people, children or animals; parkinsonism, meaning at least one of slowness, tremor or rigidity; and REM sleep behavior disorder. Dreaming is not on that list. Acting a dream out physically is.

One detail from that report is worth carrying with you, because it explains a boundary that confuses almost everybody. Dementia with Lewy bodies and Parkinson's disease dementia are separated by what clinicians call the one-year rule: which came first, the dementia or the movement symptoms, and by how long. The consensus group remained supportive of the rule while stating plainly that the cutoff is arbitrary, based on expert opinion, and useful mainly in practice rather than settled biology. Two names, one underlying protein, and a line drawn where the evidence has not yet drawn one itself.

02Why sleep is one of the first places it shows up

During REM sleep the brain sends out movement signals while a small brainstem circuit holds the voluntary muscles quiet. When that circuit stops holding, the dream reaches the body. Our companion guide to acting out dreams while asleep works through the mechanism and the bedroom safety measures in detail, and the opposite failure, where the same muscle shutoff lingers a few seconds too long after waking, is covered in what sleep paralysis is and why you cannot move.

The reason that circuit matters here is anatomical. Boeve and colleagues set out the pathophysiology in the journal Brain in 2007, proposing that the human structure equivalent to the subcoeruleus region in cats and the sublaterodorsal nucleus in rats, along with the pathways running into and out of it, is the crucial site. They then placed that anatomy alongside the Braak staging scheme, in which the location and order of alpha-synuclein deposits across the brain could account for the sequence in which parkinsonism and dementia appear. In plainer terms: the equipment that keeps you still while you dream sits low in the brainstem, and in some people it is affected before the equipment that controls walking or remembering. The nights change first because of where the trouble starts, not because dreams are a warning system.

That is also why the 2017 consensus report moved dream enactment up in diagnostic weight. In autopsy-confirmed cases, the report notes, it was present in 76 percent of people with dementia with Lewy bodies against 4 percent of those without, which is an unusually sharp separation for a symptom you can ask about in a clinic. The same passage adds two things that rarely make it into short summaries. Dream enactment often begins many years before other symptoms. And it may become less vigorous, or even fall quiet, as time passes, so a calmer year is not by itself reassurance.

03What a prodrome means in plain language

Prodrome is a borrowed word for the stretch of time when a condition has begun to affect the body but has not yet declared itself in the way the diagnosis requires. It is not a mild version of the illness and it is not a countdown. It is a period that is usually only visible in hindsight, which is exactly why researchers are interested in it and exactly why it is so hard to sit with from the inside.

In 2020, McKeith and colleagues published research criteria for prodromal dementia with Lewy bodies in Neurology, and the shape of what they proposed is more informative than any single number. They described three ways the prodromal phase presents: as mild cognitive impairment, as a delirium-onset picture, and as a psychiatric-onset picture. They then set out operational criteria for only the first of those. Their own words about the status of the criteria are the part to hold onto: they are intended for use in research settings, pending validation for use in clinical practice.

Read that carefully, because it answers a question people rarely think to ask. There is no diagnosis called prodromal dementia with Lewy bodies that a doctor can give somebody today. The framework exists so that research groups can enroll comparable people into studies. It does not exist as a verdict, and nobody is being denied one by an unhelpful clinician.

The same paper explains why sleep carries so much weight inside that framework. It cites a multicenter autopsy study in which, of 80 people with sleep-study-confirmed dream enactment who also had a neurodegenerative disorder, only two had a condition unrelated to alpha-synuclein. That is a very specific association, and it is the reason confirmed enactment is treated as meaningful rather than incidental. It is also, precisely, an association measured in people whose disorder was confirmed.

Even the boundary of confirmation is still being negotiated among the specialists themselves. The International RBD Study Group's guidelines on video-polysomnography procedures, published in Sleep in 2022, were written because several aspects of the recording and scoring needed harmonizing across sleep centers, and they state directly that the boundary between prodromal and definite REM sleep behavior disorder remains unclear. If that boundary is unsettled in a room full of sleep neurologists, it cannot be settled at two in the morning by somebody replaying a memory of last night.

04The numbers, and exactly who they describe

The study nearly every alarming video is built on is Postuma and colleagues' 2019 multicenter cohort in the journal Brain, and it deserves to be read properly rather than quoted in a caption.

Twenty-four sleep centers pooled their prospective follow-up data. Everyone included had isolated REM sleep behavior disorder confirmed on a polysomnogram to American Academy of Sleep Medicine criteria, and was free of parkinsonism and dementia when they entered. There were 1,280 of them. Their average age was 66.3, and 82.5 percent were men. Average follow-up was 4.6 years, with individual follow-up ranging from one year to nineteen.

During that follow-up, 352 people, 28 percent of the cohort, developed an overt neurodegenerative syndrome. The median time from assessment to that point was 8.0 years. Projected forward with survival analysis, the risk was 10.6 percent at two years, 17.9 percent at three, 31.3 percent at five, 51.4 percent at eight, 60.2 percent at ten and 73.5 percent at twelve. The much-repeated 6.3 percent per year is the overall rate across the whole group, and the 73.5 percent is a twelve-year projection, not a count of people who actually converted.

Two further details from that paper matter a great deal to anybody searching the words Lewy body. Of the 352 who converted, 199 developed parkinsonism as the first manifestation and 153 developed dementia first. So, the frightening headline number is not a dementia number. It is a number for a group of related conditions, and dementia was the opening presentation in fewer than half of the cases. For a household where Parkinson's and dream enactment is already the situation rather than the risk, that page takes over from this one.

The authors also drew a boundary around their own findings. Discussing the study's limits, they warned that the size of the risk associations they measured should not be extrapolated to the general population, because in a group already selected for confirmed dream enactment the baseline risk is so high that the statistics behave differently. That caution is in the paper. It rarely survives the trip to a caption.

Set against it, the population picture looks different. Haba-Rubio and colleagues reported in Sleep in 2018 on nearly two thousand adults from a population-based study in Lausanne, average age 59, who each had a full sleep recording at home. Three hundred and sixty-eight of them said yes to questionnaire items about acting out dreams. Twenty-one met the polysomnographic criteria. The estimated prevalence in that population was 1.06 percent, and there was no difference between men and women.

That last clause is worth sitting with, since the clinic cohorts are more than four-fifths male. It cuts both ways. It means a man who thrashes is not automatically the case the textbooks describe, and it means a woman whose enactment is real may be less likely to reach a sleep center at all.

05What that does not mean tonight

One night establishes nothing. Neither does a phone recording, however alarming it looks on replay. The 2017 consensus report lists the conditions that imitate dream enactment in older people, naming confusional awakenings, severe obstructive sleep apnea and periodic limb movements, and it says each must be excluded by careful questioning to avoid a false-positive diagnosis. That advice is aimed at clinicians who see this every week. It applies with more force to a family member watching from three feet away in the dark.

And here is the harder half, because the honest version of this includes removing a comfort as well as an alarm. If the enactment turns out to be real and confirmed, there is at present no treatment that changes the long-term course. The American Academy of Sleep Medicine's 2023 clinical practice guideline on managing REM sleep behavior disorder says so directly when it discusses how to talk to patients about risk, referring to the current lack of neuroprotective treatments to slow or halt disease progression. What treatment does address is the nights themselves and the injuries. The guideline's recommendations for isolated dream enactment in adults are conditional ones for clonazepam and for immediate-release melatonin, and for drug-induced cases, a conditional recommendation to discontinue the drug involved. Those are real options with real trade-offs, and none of them is a shield.

The other thing an assessment cannot do is put a date on anything. Median time to conversion in the Brain cohort was eight years, and individual follow-up in that study ranged from one year to nineteen. A median is the middle of a wide spread, not a schedule.

06What else can produce the same night

The confounders here are not footnotes. They are the likeliest explanations for most people, and they are the reason an assessment is worth the trouble rather than being an overreaction.

Medicines come first, and antidepressants lead the list. A 2023 review of diagnosis and management by Sobreira-Neto and colleagues in Arquivos de Neuro-Psiquiatria tabulates the drugs associated with dream enactment and with the underlying muscle-activity finding, naming the selective serotonin reuptake inhibitors citalopram, fluoxetine, paroxetine and sertraline, the serotonin-norepinephrine reuptake inhibitors duloxetine and venlafaxine, and mirtazapine. The population study above found the same thing from the other direction, reporting that confirmed cases were associated with antidepressant and antipsychotic use. What the published evidence cannot cleanly settle is whether such a medicine causes the enactment or reveals a tendency that was already there, and the AASM guideline treats both readings as live. Write down when the prescription started and when the nights changed, and take the dates to the prescriber. Do not stop or adjust anything to test the theory.

Obstructive sleep apnea is the second, and it is the most hopeful item on the list, because the behavior can look identical from across the room and the treatment is entirely different. Untreated apnea fragments the night into repeated arousals, and the resulting movements and shouts are sometimes described as pseudo-enactment. A sleep study separates the two by showing whether the muscle shutoff during REM was normal.

Beyond those, the same review lists narcolepsy type 1, which is a genuinely different thing, structural brainstem lesions, and post-traumatic stress disorder among the conditions associated with dream enactment, and notes that alcohol and substance withdrawal belong in the picture too. Non-REM parasomnias such as sleepwalking and night terrors, and nocturnal seizures, complete the set of things an overnight recording is used to rule out.

There is one more explanation that has nothing to do with the body at all. Sometimes what changed is not the dreaming but the noticing. Broken sleep produces more awakenings, each awakening is another chance to catch a dream mid-scene, and a household that has started paying attention to the nights will remember far more of them. Our guide to why dreams suddenly become so vivid works through that loop, and whether vivid dreams are a sign of dementia separates intensity of dreaming from enactment, which is the distinction the whole subject turns on.

07What an assessment actually involves

People put this off because they do not know what they are agreeing to, so here is the night in concrete terms.

The test is a video polysomnogram, and the technical standards Sobreira-Neto and colleagues reproduce describe what is set up. The recording is time-synchronized and audio-equipped, so that the video, the sound and the physiological traces line up second for second and any behavior can be matched to what the body was doing at that moment. Surface electrodes go on the chin. More go on the legs, placed lengthwise along the middle of the shin muscle a couple of centimeters apart, and on the forearms over the muscles that flex the fingers. Both legs and both arms are monitored, with a separate channel for each limb wherever possible. Nothing goes under the skin, and none of it is painful.

What happens afterward is the part that explains why a home video cannot substitute. The night is divided into thirty-second epochs, and each epoch of REM sleep is divided again into ten three-second segments. A burst of muscle activity counts if it lasts between a tenth of a second and five seconds and is at least twice the height of that person's own baseline muscle tone during REM. An epoch is marked as showing REM sleep without atonia if the chin muscle is continuously active, or if enough of those small segments contain qualifying bursts. The report then states the percentage of REM epochs that met the criteria.

That is the number the diagnosis rests on. It is a measurement of muscle tone against your own baseline, tallied across a whole night, and it is why a dramatic clip settles nothing while a quiet recorded night can still be positive. It is also why the study is worth doing when the nights are frightening: it can find apnea instead, or find nothing abnormal at all, and both of those are answers.

Two practical notes. Recording and scoring practices vary between centers, which is what the International RBD Study Group's guidelines were written to address, so it is fair to ask which criteria a center uses. And what you bring shapes the appointment more than anything else: the date the episodes started, roughly how often they happen, what time of night, what the sleeper did, any injury or near miss, whether they remembered a dream and whether it matched, every medicine and supplement with dates of any change, alcohol use, and whether there is snoring, gasping or witnessed pauses in breathing. The person asleep cannot report most of this. Whoever was awake for it can.

08Whether it helps to know

This is a genuine argument, and it is being had by the people it lands on rather than only by clinicians.

In a Parkinson's community, someone posted a question that named sleep behavior disorder in a list of early non-motor signs and then made the case against watching for any of them. "We're told to watch for these because they can appear years/decades before motor symptoms. But honestly, what real good does knowing this do right now?" They pointed out that a diagnosis is not available before motor symptoms, that no therapy has been proven to halt progression, and that most of the early signs have plenty of ordinary causes.

The replies did not tell them they were wrong. They offered a different accounting. One person, writing about their own years before diagnosis, listed what they would have done: "I would've absolutely worked out more, quit alcohol sooner, bought life insurance, found a PCP, developed a better diet." They allowed openly that it might have changed nothing, and said the value was in not being blindsided. Another reply deflated the panic without dismissing the question, observing that "most people without Parkinson's are not paying attention to what these early signs are or worrying about them", and naming earlier exercise as the one concrete advantage of knowing sooner.

The clinical guideline lands in almost the same place, and is unusually candid about it. The AASM task force describes disclosure of neurodegenerative risk as an ethical dilemma with weight on both sides. Telling somebody can help them plan, arrange monitoring for change over time, and take part in research. Not telling them, given the absence of a treatment that alters the course, may spare anxiety, depression and worse over a condition that could take years to appear and might not appear in that person's lifetime. Then it names the thing that brought you here: withholding the information may itself damage the relationship, because patients often discover the association anyway through an internet search. The task force notes there are no data at all on what patients with isolated dream enactment want in this situation, and sets out two approaches, a patient-centered risk disclosure and a watchful waiting that revisits the subject at later visits.

For someone who does end up with a confirmed diagnosis, the guideline also names the practical middle ground. It suggests clinicians point people toward research groups such as the North American Prodromal Synucleinopathy consortium and the International RBD Study Group, and observes that many people find being part of that search steadying rather than frightening. It also notes which other features change a clinician's read: reduced sense of smell, slowed bowel motility and blood pressure that drops on standing, occurring alongside confirmed enactment, are strong predictors of conversion within five years, while their absence, particularly in someone taking a serotonergic antidepressant, is associated with lower five-year risk. That is a judgment for a clinician who has examined the person. It is not a list to score at home, and it is not offered here as one.

09If the person you are worried about is not you

Most people reading this are the one who was awake. That position comes with a particular kind of helplessness, because you are the only witness to something the other person genuinely cannot report.

Three things are worth doing, and they are small. Write down what you actually saw, with dates, in plain words rather than medical ones. Make the bed area safer tonight, which costs nothing and takes twenty minutes; the specifics are set out in our guide to acting out dreams while asleep, and the short version is that hard and sharp things move away from the bed, something soft goes on the floor beside it, and a pillow between the two of you is a reasonable interim measure. Then ask for a referral to a sleep specialist rather than for a verdict.

What is worth avoiding is turning the person into a case study. Watching somebody for signs is heavy on both of you, it changes how a household talks at breakfast, and it produces nothing that a dated note and one appointment would not. If they have started sleeping badly because they know they are being observed, that is a real cost with nothing to show for it.

It is also worth saying that dream enactment carries no message about the sleeper. The behavior comes from a muscle switch that is not holding, and the dreams that go with it lean toward threat and defense, which is why the movements so often look like a fight. Nobody is expressing anything about you at four in the morning. Nor is every noise at night the same thing. Sleep talking on its own is very common in people with nothing wrong with them, and the pattern that gets studied involves repeated physical enactment, which is a different animal from muttering.

And if the person reading this is the sleeper, being told what you did while you were unconscious is its own particular experience. It is ordinary not to remember, since Neikrug and Ancoli-Israel, reviewing the diagnostic tools in Sleep Medicine Reviews, cite earlier research finding that up to 35 percent of people with the disorder are not aware of their own dream-enactment behaviors, and it is ordinary to feel exposed by something you had no part in choosing. It is also reasonable to want the question settled rather than carried around. An appointment is the only route to that, and it is a shorter route than it looks from here.

10Common questions about Lewy body dementia and acting out dreams

Does acting out dreams mean Lewy body dementia is coming?

No, and the association is narrower than it sounds. It has been measured in people whose dream enactment was confirmed by an overnight recording and who were then followed for years at sleep centers. In that group, 6.3 percent per year developed a neurodegenerative condition, and of those who did, fewer than half presented with dementia first. Somebody who has not had a sleep study is not in the group those numbers describe.

How many years before dementia does dream enactment start?

There is no fixed interval. In the largest long-term cohort the median time from assessment to any overt neurodegenerative diagnosis was eight years, but individual follow-up in that study ran from one year to nineteen, and the outcome counted included Parkinson's disease and multiple system atrophy as well as dementia. The 2017 consensus report says only that enactment often begins many years before other symptoms.

Is REM sleep behavior disorder always a sign of something neurological?

Not always. It is also associated with antidepressants and antipsychotics, with narcolepsy type 1, with structural lesions in the brainstem, with post-traumatic stress disorder, and with alcohol or substance withdrawal. Obstructive sleep apnea, sleepwalking, night terrors and nocturnal seizures can look the same from outside the bed without being it at all.

Can a doctor diagnose prodromal Lewy body dementia?

Not as things stand. Research criteria for prodromal dementia with Lewy bodies were published in 2020, describing presentations that begin with mild cognitive impairment, with delirium, or with psychiatric symptoms, and their authors state that the criteria are intended for research use pending validation for clinical practice. There is no prodromal diagnosis available in a clinic today.

Why does a sleep study matter if there is no treatment that stops the disease?

Because most of what an assessment finds is treatable. It can identify sleep apnea, a medication effect, or a non-REM parasomnia, each of which is managed differently. It gives access to injury prevention that works, and to options a sleep specialist can weigh with you. It also replaces an open question with a real answer, which is worth something at three in the morning.

Does the sleep study hurt, and what is actually attached?

Nothing goes under the skin. Surface electrodes are placed on the chin, on both shins and on both forearms, with a separate channel for each limb, alongside the standard sleep recording, plus synchronized video and audio so any behavior can be matched to the traces. The scoring afterward counts brief muscle bursts against your own baseline muscle tone during REM sleep.

My parent has Lewy body dementia and still acts out dreams. Is it getting worse?

Not necessarily, and the pattern can run the other way. The 2017 consensus report notes that dream enactment may become less vigorous over time, or fall quiet altogether, so a change in the nights is not a reliable measure of anything. Changes in memory, attention, movement or waking perception are what a clinician will want to hear about.

Should we sleep in separate rooms?

Sometimes, and it is a safety measure rather than a verdict on a relationship. The AASM guideline's good practice statement suggests separate sleeping for severe or uncontrolled episodes, and at minimum a pillow placed between the two people. Plenty of couples do this while an assessment is underway and revisit it afterward.

If you came here frightened, the shape of the answer is this. There is a real association, it was measured carefully, and it was measured in people who had spent a night in a sleep laboratory having their muscle tone recorded. Somebody who has not had that night is not described by those numbers, and the way to find out which situation you are in is an appointment rather than another search. DreamTold treats dreams as a window into your own mind rather than a warning system, and once the medical question is with someone qualified, whatever image stayed with you from a rough night is still yours to think about in the dream dictionary. This guide is educational and meant for reflection, not medical advice, and it cannot diagnose anything. If dream enactment is repeating, if anyone has been hurt, or if new changes in memory, movement or waking perception are worrying your family, speak with a qualified clinician.

11Sources

Every claim on this page that comes from somewhere else, with the somewhere else. Discussion quoted from people living with these conditions links to the comment it was written in.

• nia.nih.gov: National Institute on Aging explains in its guide to Lewy body dementia
• PubMed: Boeve and colleagues set out the pathophysiology in the journal Brain in 2007
• PubMed: Haba-Rubio and colleagues reported in Sleep in 2018 on nearly two thousand adults
• PubMed: International RBD Study Group's guidelines on video-polysomnography procedures, published in Sleep in 2022
• PubMed: McKeith and colleagues published research criteria for prodromal dementia with Lewy bodies in Neurology
• PubMed Central: 1,280 people who already had that laboratory diagnosis
• PubMed Central: Neikrug and Ancoli-Israel, reviewing the diagnostic tools in Sleep Medicine Reviews
• PubMed Central: American Academy of Sleep Medicine's 2023 clinical practice guideline on managing REM sleep behavior disorder
• PubMed Central: fourth consensus report of the DLB Consortium, published in Neurology in 2017 by McKeith and colleagues
• PubMed Central: 2023 review of diagnosis and management by Sobreira-Neto and colleagues in Arquivos de Neuro-Psiquiatria
• Reddit: We're told to watch for these because they can appear years/decades before motor symptoms
• Reddit: I would've absolutely worked out more, quit alcohol sooner, bought life insurance, found a PCP, developed a better diet
• Reddit: most people without Parkinson's are not paying attention to what these early signs are or worrying about them