Are We in the Dark About Sleepwalking’s Dangers?

Sleepwalking generally occurs in the dark and has remained there, both literally and figuratively, for centuries. The image that comes most readily to mind is a cartoon person, amiably and aimlessly wandering the hallway with arms outstretched and eyes closed. But sleepwalking is not funny; it is a sleep disorder known to specialists as somnambulism. Many adult sleepwalkers, with eyes open, perform purposeful acts such as eating half a bag of chips and putting the rest in the microwave, taking all their shoes from the closet and lining them on the windowsill, rearranging furniture, or climbing out a window in the middle of the night—activities that are essentially benign when a person is conscious but that, when they occur during somnambulism, are potentially dangerous to the sleepwalker or other people. More frighteningly, increasing numbers of so-called “sleepdriving” cases are being reported in which somnambulists get in their cars and drive sometimes long distances, disregarding lanes, stoplights, and stationary objects, and, after waking up, having no memory of what they did.

Although these nocturnal wanderings may seem extremely odd to nonsleepwalkers, such mechanistic and automatic activities are part of the spectrum of behavior associated with somnambulism, which is estimated to affect close to 2 percent of the adult population worldwide. Sleepwalking and other sleep disorders appear to be on the rise in our demanding and fast-paced society, in which getting a good night’s sleep seems to be increasingly difficult. Many people resort to prescription (or nonprescription) drugs to induce sleep, but sometimes this only compounds the problem. Sleep deprivation, especially in combination with drugs and alcohol, is known to induce sleepwalking in some people, and behavior while sleepwalking is extremely unpredictable, particularly in a new environment. Any adult with a tendency to sleepwalk has the potential to experience an accident and can be at risk of real injury. But, until recently, published reports of injuries as a result of sleepwalking were rare, and somnambulism and other sleep disorders are frequently overlooked in the medical school curriculum.

Although I am a teacher of medical neuroscience, the dangers of sleepwalking would probably never have come to my attention had my son Stewart (who joins me in writing this article) not sleepwalked out a second-story window into an alley, sustaining serious injuries, on the night he arrived for a British Studies program at St. John’s College in Oxford, England. His potentially fatal experience with sleepwalking demanded a reexamination of this overlooked topic and raised many questions during his convalescence. Had other people been seriously injured while sleepwalking? If so, were these random and rare events, or had we encountered the tip of an unexplored iceberg?

By searching the medical literature and interviewing other sleepwalkers, we found that sleepwalking accidents and injuries, more common than usually believed, are a definite health hazard for both the sleepwalker and other people. But such accidents are not well known, because both the general public and physicians are uninformed about somnambulism.

In this article, we explore current theories about both the causes and the management of adult sleepwalking, while seeking to increase awareness of its hidden dangers. Sleep medicine needs to be an integral part of the medical school curriculum, and physicians as well as the general public should be aware that, unlike sleepwalking in children, somnambulism in adults is a potentially dangerous disorder. Both treatment of the disorder—when possible—and prevention of accidents are of paramount importance for the sleepwalker and for unsuspecting people who may find themselves in the sleepwalker’s path.

Experiences of Sleepwalkers

Stewart arrived in Oxford, sleep-deprived after the long trip from Texas, and checked into his second-floor dormitory at St. John’s College on a warm July day. The wide open windows had no screens and were surrounded by scaffolding. Even though he had been awake for more than 30 hours, he chose to postpone sleep until after dinner in order to adjust to British time. He fell asleep easily, but, when he awoke about 2:00 a.m., he was lying face down on a cobblestone street that he did not recognize. He had absolutely no recall of having left his dorm room, walking through several doorways, and stepping out a window onto the scaffolding from which he must have fallen into the alley. After unsuccessfully trying to lift himself off the cobblestone street, he dragged himself toward what appeared to be a road and was discovered by the British police. He had fractured his spine and right wrist in the accident, but thankfully had no permanent neurological damage.

At the same time that Stewart was recovering at the John Radcliffe Hospital, CNN News reported the story of a London teenaged girl who was rescued from the top of a crane, which she had climbed while sleepwalking and then gone back to sleep on the support beam—fortunately catching the attention of a pedestrian who notified the police. The hospital staff caring for Stewart found this amusing and suggested that the two young sleepwalkers get to know each other. Although Stewart and the girl never met, their similar stories motivated us to start looking further into the prevalence and possible causes of adult sleepwalking.

We did not have to look far for stories. The paramedic who helped take Stewart to Gatwick Airport had an adult sister who had injured herself while sleepwalking, and we encountered a woman on the flight home whose son had repeatedly sleepwalked onto a balcony. An acquaintance called to relate how her son not only regularly disassembled bedside lamps while sleepwalking but was recently found, in his pajamas, pumping gas after sleepdriving to the gas station. In 2003, several cases involving mysterious nighttime accidents, some of which were fatal and initially ruled as suicide, were reported in the Journal of Forensic Science by Mark Mahawald, M.D., director of the Minnesota Regional Sleep Disorders Center. These deaths, referred to as “parasomnia pseudo-suicides,” were later attributed to complex motor behaviors that can take place during sleepwalking, such as running, climbing, or jumping. One involved a 21-year-old college student who was hit by a semitrailer truck after he ran onto a highway at 4:30 a.m., clad only in his boxer shorts. He had no history of drug abuse or depression, but he and several family members had a history of frequent, complex sleepwalking. A formal review of his case requested by his family resulted in a recommendation by the medical examiner that the cause of death be changed from “suicide” to “accidental death due to sleepwalking.”

Other cases described by Mahawald involved falls from balconies, defenestration (jumping from windows), and self-inflicted gunshot wounds by people with a past history of complex sleepwalking behavior and no history of depression. Distinguishing between accidental death and suicide has profound religious, societal, and insurance implications, of course, and many of the families of these victims requested that these pseudo-suicides be reevaluated as accidental death as a result of a sleep disorder. Some cases of homicidal behavior during sleepwalking have also been reported. The legal defense in these cases has usually been to claim that the action was a “non-insane automatism,” meaning that the brain’s motor system was fully aroused but consciousness was clouded. In all of these types of cases, a correct diagnosis or verdict can be made only if the family, police, and medical examiners are willing to consider alternative scenarios.

Prescription Drugs as a Risk Factor

Early in 2006, a surge of news reports described complex sleepwalking behaviors that involved binge eating, violent outbursts, and sleepdriving in people who took the medication Ambien (zolpidem).

Ambien, the best-selling prescription sleeping pill in the United States, is a popular alternative to the more traditional benzodiazepines because of its general lack of serious side effects. Since it was introduced in the early 1990s, mild side effects such as nausea, dizziness, and nightmares have been reported in people taking the drug as prescribed. But in 1994 and 1995, the first two cases of sleepwalking attributed to Ambien appeared in the medical literature, followed by an additional six cases during the next 10 years. The most recent case, reported in the Archives of Physical Medicine in June 2005, involved a middle-aged man with no history of sleepwalking or previous Ambien use who began sleepwalking after taking Ambien while hospitalized for hip surgery. His somnambulism stopped as soon as the drug was discontinued.

A recent article in the New York Times reported that Ambien is one of the top 10 drugs identified in the blood of impaired drivers, and, in Wisconsin alone, Ambien was identified in 187 drivers arrested from 1999 to 2004. In a presentation for the American Academy of Forensic Sciences, Laura J. Liddicott and Patrick Harding of the Wisconsin State Laboratory discussed six cases now set for trial. They reported that all of the drivers tested negative for ethanol and other drugs but had serum levels of Ambien well above the therapeutic range of 29 to 272 ng/mL (nanograms per milliliter). Each of the drivers displayed extremely bizarre behavior, such as wide deviations from the marked lanes and near collisions with stationary objects, and all of them, when stopped by police, appeared confused, disoriented, and somnambulant, with no memory of what they had just done. A class-action lawsuit has been filed by New York attorney Susan Chana Lask against Sanofi-Aventis, the company that makes Ambien, on behalf of anyone nationwide who has experienced such side effects as sleepwalking, sleepeating, sleepdriving, or memory loss while taking Ambien.

The official position of Sanofi-Aventis is that “The safety profile of Ambien is well established and reported in the Ambien Prescribing Information approved by the U.S. Food and Drug Administration. Sanofi-Aventis regularly conducts thorough analyses and has not observed any significant change in that safety profile. The information currently contained in the U.S. Prescribing Information remains accurate: sleepwalking (somnambulism) is a possible rare adverse event.”

During the past 10 years, a handful of case reports were published that describe sleepdriving in non-Ambien users. These reports described behavior not unlike that seen in Ambien sleepwalkers: long-distance driving and bizarre behavior, followed by complete amnesia for the event. The unifying question that needs to be addressed is: What is happening in the brain to cause such a disassociation between being awake and being asleep?

Inside the Sleepwalking Brain

In their classic overview of normal human sleep found in Principles and Practice of Sleep Medicine (4th edition, 2005), Mary Carskadon, Ph.D., professor of psychiatry at Brown University, and William Dement, M.D., Ph.D., founder of Stanford University’s Sleep Disorders Clinic, describe sleep as “a reversible behavioral state of perceptual disengagement from and unresponsiveness to the environment.” In the Oxford Dictionary, sleep is defined as “a condition of body and mind such as that which normally recurs for several hours every night, in which the nervous system is inactive, the eyes closed, the postural muscles relaxed, and consciousness practically suspended.” By both these definitions, a sleeping person is basically out cold.

In sleepwalkers, however, the states of being awake and being asleep are not mutually exclusive; instead, they occur simultaneously. Parts of the brain are aroused, the eyes are open, and postural muscles are tensed and active, while clear, lucid consciousness remains suspended. Sleepwalkers, who are awake and asleep at the same time, have been described for centuries. Think of Shakespeare’s Lady Macbeth: “You see, her eyes are open … but their sense is shut.”

More analytical descriptions of human sleep became possible at the beginning of the 20th century, with the development of the electroencephalogram (EEG), a device that measures and records the electrical activity of the brain by using flat metal discs placed on the surface of the head. The electrodes are connected by wires to an amplifier and recording machine that convert the electrical signals from the brain into pen-and-paper tracings that resemble waves. Further technological advances have increased the ability of researchers and physicians to analyze sleep and its disorders by using all-night monitoring of not only the brain but also eye and leg movements, respiration, and heart rate through a comprehensive diagnostic test known as polysomnography, conducted in a sleep laboratory.

On the basis of EEG wave patterns, sleep can be broadly divided into rapid eye movement (REM) and non–rapid eye movement (nREM). REM sleep is associated with dreaming and with high-frequency, low-voltage brain waves, whereas nREM sleep is characterized by low-frequency, high-amplitude waves known as slow-wave sleep (SWS). nREM sleep is further divided into stages I to IV, with stages III and IV representing the deepest sleep. A normal sleeping adult will repeatedly alternate between nREM sleep and REM sleep, with each total cycle lasting on average about 90 minutes, and SWS dominating the first third of the night. A person who sleeps for eight hours will progress through four or five cycles during one night’s sleep, with REM episodes becoming longer during the course of the night.

During REM sleep, a complete and dramatic loss of muscle tone occurs. This loss is protective, because it prevents the sleeper from acting out dreams. However, sleepwalking generally occurs during stages III and IV, possibly as a result of an incomplete transition from SWS back into REM sleep, and dreams are not commonly associated with these nREM stages. What, therefore, causes sleepwalkers to leave their beds and wander into the night?

Finding the answer to this question has been the goal of much research. Currently, several theories exist about what precipitates sleepwalking, although the underlying pathophysiology is still not well understood. An August 2000 research letter to the British medical journal The Lancet described the results of an experiment on a sleeping subject by using a brain imaging technique called single photon emission computed tomography (SPECT), which showed how continued deactivation of the prefrontal cortex of the brain, in combination with abnormal activation of the cingulate cortex and the thalamus, may lead to the dissociation between “body sleep” and “mind sleep” characteristic of somnambulism. In addition, twin studies have shown a possible genetic association, and in 2003 a specific genetic marker, called the HLA subtype (DQB1*0501), was implicated in sleep-associated motor disorders by Michel Lecendreux, M.D., of the Robert Dobre Hospital in Paris, France.

nREM Sleep Instability

Sleep researchers have found that sleepwalking occurs against a background of nREM sleep instability characterized by a particular kind of high-voltage brain wave called hypersynchronous slow delta (HSD). These HSD waves were first described in a May 1965 article in the journal Science describing a University of California, Los Angeles (UCLA), study of a group of sleepwalkers who underwent all-night EEG recording using special techniques that allowed them to get up and move around. HSD waves were recorded both before and during sleepwalking episodes in the UCLA study, and subsequent studies have consistently identified HSD waves associated with sleepwalking. However, the significance and specificity of these waves for sleepwalking have been questioned in subsequent research, because HSD waves have also been identified in the EEGs of people with other sleep disorders, as well as in normal sleep.

According to Christian Guilleminault, M.D., Ph.D., of the Stanford University Sleep Disorders Clinic, HSD waves can be seen normally in the nightly cyclical passage from stable to unstable sleep as the billions of neurons in the human brain are recruited during early stages of nREM sleep into the orderly SWS rhythms of stages III and IV. However, when the HSD wave pattern persists throughout stages III and IV, it is indicative of an interruption in the normal progression of nREM sleep. This disturbance can be interpreted by sleep researchers using a technique called cyclic alternating pattern (CAP) analysis.

CAP analysis is particularly valuable in evaluating nREM disorders of arousal, because these disorders are not generally accompanied by significant changes in the brain processes that can be detected by all-night sleep monitoring using EEG and the other tools of polysomnography. CAP provides a measure of the microstructure of brain waves in sleep instability through analysis of sequences of EEG patterns. If the CAP rate is indicative of abnormal sleep, then it is imperative to search for the instability’s cause, which is generally a subtle associated sleep disorder. In a January 2006 article in the journal Sleep Medicine, Guilleminault stated that he did not find a “pure” sleepwalker in the most recent 100 cases he studied and that the identification of the underlying cause of sleep instability often led to treatment and elimination of sleepwalking in his patients.

What Causes Sleep Instability?

In chronic sleepwalkers, respiratory syndromes are the most frequently diagnosed accompanying disorders. As a result of the close relation between abnormal retention of carbon dioxide in the blood (a condition known as hypercapnia) and activation of neurons within the brain stem that control sleep and waking, an inability to breathe normally affects neural control of the progression through sleep. Specific respiratory syndromes, including upper airway resistance syndrome, mild obstructive sleep apnea, and sleep-disordered breathing, have been diagnosed as underlying causes of sleep instability. Through CAP analysis of chronic sleepwalkers, researchers have learned that the basic nREM instability accompanying the breathing disorder is present even on nights when no sleepwalking occurs. But the instability almost always completely vanishes when the respiratory problem is successfully treated, usually through the delivery of air to the upper respiratory tract through a specially designed mask (continuous positive airway pressure, or CPAP), or through surgery.

The Stanford University study of chronic sleepwalkers reported by Guilleminault found that sleepwalking was much more likely to be eradicated in patients with treatable respiratory disorders, so it is important to seek an underlying cause of sleep instability for chronic sleepwalkers. The rare cases of “pure” sleepwalking, which appear to have no associated disorder, may represent a subgroup of sleepwalkers in whom nREM sleep instability is the result of genetic factors. Benzodiazepines, the most commonly prescribed drugs for sleepwalking, are only partially effective in eliminating sleepwalking in these patients, so attention must be focused on maintaining a safe sleeping environment to prevent accidents.

Living Safely with Adult Somnambulism

When sleepwalking behavior persists or reemerges in adulthood, it is no longer a relatively benign disorder of childhood, even though the same underlying nREM sleep instability is present at all ages. Occasional injuries have been reported in childhood sleepwalkers, but by the time a child is about 12 years old, when the central nervous system matures, episodes usually disappear—before most children are able to drive or have access to alcohol and guns. The most common automatic, unconscious behavior in young children who sleepwalk is to seek their parents, which is what Stewart did during his childhood sleepwalking episodes.

For adults in whom sleepwalking has become chronic or dangerous, it is thus imperative to address both issues of safety and the eventual elimination of the behavior. Since repeated episodes of somnambulism indicate an underlying nREM sleep instability, physicians must try to identify any associated sleep disorder that could be causing abnormal progression through the stages of sleep. But even if, as is most often the case, a sleep-related breathing disorder is identified, treatment such as CPAP or surgery is not instantly successful, and the inherent dangers of sleepwalking will persist until the cause of nREM instability is completely eliminated. Therefore, safety remains of paramount importance in managing chronic sleepwalking even after diagnosing and starting to treat the underlying disorder. In some sleepwalkers, no treatable cause will be found; for them attention to sleep practices and safeguarding the environment are lifetime challenges.

The real key to sleepwalking safety is knowledge, knowledge of whether a person is a sleepwalker and awareness of the conditions or drugs that increase the possibility of a sleepwalking incident. All prescription sleep medications should be taken exactly as directed. Ambien may not be a good choice for someone with a history of sleepwalking. If taken with alcohol, Ambien has the potential to induce sleepwalking even in people with no previous history of the disorder. According to Laura Liddicoat, the forensic toxicologist who is investigating the Wisconsin sleepdriving incidents, the only tolerable blood alcohol level for someone who is taking Ambien is 0.0 percent.

Sleep deprivation is also known to trigger sleepwalking in susceptible persons, possibly as a result of the extremely deep nREM sleep, known as rebound or recovery sleep, that often occurs after long periods without sleep. Stewart had been awake for more than 35 hours when he finally fell asleep in his Oxford dormitory room, an amount of sleep deprivation that sleep laboratory studies have shown is sufficient to increase the frequency and complexity of somnambulistic episodes during recovery sleep. In some laboratories, artificially inducing sleepwalking by sleep deprivation has been used as a tool in the diagnosis of somnambulism. Known sleepwalkers should therefore do everything possible not to become sleep deprived, particularly when they cross time zones and sleep the first night in a new environment—all factors in Stewart’s Oxford accident. In addition, neither alcohol nor sleep drugs such as Ambien should be taken under these conditions.

Both at home and when traveling, safeguarding the environment should be a top priority for sleepwalkers. Appropriate precautions include choosing lower bunks and ground-floor rooms and bolting shut doors and windows, possibly with a chair placed in front of them (after first locating and not blocking the fire exits.) Beds should be pushed to the wall, and a sleeping partner should sleep on the outside, so the sleepwalker would have to climb over the partner to get out of bed and wander into the night. If possible, bedroom and outer doors should be equipped with alarms and buzzers that are loud enough to awaken the sleepwalker or the family, particularly when traveling by boat because of the risk of falling overboard. Power tools or guns should be stored in locked cabinets with combinations or key entry not amenable to being unlocked in an unconscious state. A sleepwalker should never be allowed to drive while somnambulant, and car keys as well as the car should be made inaccessible at night if there is any tendency to sleepdrive. Contrary to what most people think, it is not dangerous to awaken a sleepwalker, and he or she will probably thank you the next morning upon waking up safely in bed.

For the most part, sleepwalkers have earned their amiable reputation as, in the words of Shakespeare, just “merry wanderers of the night.” Whether conducting imaginary orchestras, climbing trees, or taking walks outside clad only in pajamas, they give us a glimpse of the incredible intricacy and complexity of a human brain that is capable of being awake and asleep at the same time. That most people successfully make the journey through the many stages of sleep several times each night is a testimony to the ability of billions of neurons to synchronize themselves into the fundamental biological process required by all organisms —the need to rest. In somnambulism this process has somehow been subverted, but, with a growing public awareness about the hidden dangers of sleepwalking and increasing coverage of sleep medicine in neuroscience textbooks and medical school curricula, we hope that the world is becoming a safer place for all night-time wanderers.