By Regina Patrick, RPSGT

In 1955, Albert Stunkard(1) and colleagues first described a phenomenon in obese patients whose obesity was treatment-resistant: the patients awoke one or more times from sleep to eat excessively (hyperphagia) but had no appetite in the morning (morning anorexia). Stunkard termed this behavior “nocturnal eating syndrome” (NES).

In 2008, Stunkard and colleague Kelly Allison compared the characteristics of NES in obese and nonobese people who had maintained food and sleep diaries for 7 days.(2) They found that obese people with NES had a significantly delayed circadian food intake pattern. Other researchers have noted alterations in the circadian production of melatonin and leptin (a hormone that promotes weight loss). The levels of both hormones normally rise at night; however, this rise was attenuated in people with NES.(3)

In recent years, the following criteria have been proposed to describe NES: the person has evening or nocturnal hyperphagia; nocturnal awakenings; morning anorexia; and no other eating disorder (bulimia).(4)

A related disorder, sleep-related eating disorder (SRED), shares the features of NES. However, a distinguishing feature of SRED may be that the person often has no memory of having eaten during the night. The person may awaken to evidence that food has been prepared: food left haphazardly on a table or counter or crumbs in the bed. Sleep-related eating disorder has been included as a parasomnia in the International Classification of Sleep Disorders 2 (ICSD-2) and has the following criteria(4):

The person has recurrent episodes of involuntary eating and one or more of the following:

> Consumption of peculiar foods or combinations of foods or the consumption of inedible or toxic substances
> Insomnia complaints (because of the repeated episodes of nocturnal awakenings to eat) and complaints of nonrestorative sleep, daytime fatigue, or somnolence
> Sleep-related injury
> Dangerous behavior performed while seeking food or cooking food
> Morning anorexia
> Adverse health consequences from bingeing on high caloric foods

Despite these criteria for NES and SRED, scientists remain uncertain whether these represent two separate eating disorders or are different aspects of the same disorder.(5)

For some scientists, a differentiating feature between NES and SRED is that people with NES remember the episode the next day, whereas people with SRED often do not remember eating. However, some research indicates that the use of sedative or hypnotic drugs (eg, zolpidem or the benzodiazepines) may contribute to the amnesic phenomenon noted in people with SRED who are taking these drugs for insomnia or other reasons.6 In one study, patients with SRED were all drug-free and all had full consciousness during the SRED episode and remembered the episode the next day.(5)

Zolpidem and some benzodiazepine drugs (eg, triazolam and midazolam) are known for their amnesic effect. These drugs bind to the benzodiazepine (BDZ) receptor. The BDZ receptor can be found in many areas of the brain, such as the sensorimotor regions of the cortex of the brain, the basal ganglia, and the substantia nigra (a dark layer of cells in the midbrain that plays a role in movement, addiction, and reward; the substantia nigra contains many dopaminergic neurons).

Certain sites on the BDZ receptor may be involved in memory. In an animal study, researchers Savic and colleagues administered to rats different BDZ drugs and BDZ antagonist drugs. The BDZ drugs were midazolam (which nonselectively binds to several sites on the BDZ receptor), zolpidem (which binds to a specific site [called the alpha1 subunit] on the BDZ receptor), and methyl 6,7-dimethoxy-4-ethyl-beta-carboline-3-carboxylate (DMCM, which nonselectively binds to several sites on the BDZ receptor but decreases BDZ activity). The BDZ antagonists were flumazenil (which nonselectively binds to several sites on the BDZ receptor) and beta-carboline-3-carboxylate-t-butyl ester (beta-CCt, which selectively binds to the alpha1 subunit on the BDZ receptor).

After the administration of each of the drugs, the rats were assessed on their performance of a task. Savic found that midazolam and zolpidem induced amnesia, but the antagonists flumazenil and beta-CCt did not. Savic further found that the amnesic effect of zolpidem was attenuated by flumazenil and by beta-CCt. By contrast, only flumazenil could attenuate the amnesic effect of midazolam. DMCM promoted memory in the rats, but the administration of flumazenil or beta-CCt abolished this memory-promoting effect. From these results, Savic concluded that the alpha1 subunit and possibly other alpha subunits on the BDZ receptor may contribute to the amnesic effect of BDZ drugs.

Restless Legs Syndrome

In a person with restless legs syndrome (RLS), creepy, crawly, uncomfortable sensations in the legs emerge in the late evening. To relieve the sensations, the person has an irresistible urge to move the legs. The person may walk, move the legs around, and rub or massage the legs to gain relief, but these actions only temporarily relieve the sensations. The sensations ultimately subside with the onset of sleep, but the efforts to relieve the sensations can delay sleep onset and shorten a person’s sleep period. If they awaken during the night, the sensations can return and the person may have to walk, massage the legs, and move the legs around again to relieve the sensations before finally being able to go back to sleep.

A disorder that is related to RLS is periodic limb movement disorder (PLMD) in which one or both legs (or arms) contract rhythmically and intermittently during sleep. The contractions last from 0.5 seconds to 10 seconds and can occur 5 to 90 seconds apart. The contractions may cause momentary arousals, and thereby disrupt sleep. Some people with RLS also have PLMD.

Dysregulation in the dopamine system has been implicated in the compulsive nature of nocturnal eating disorder and in the abnormal limb movements in RLS and PLMD. These disorders often improve with treatment by dopaminergic drugs (eg, L-dopa, carbidopa) or drugs that enhance dopaminergic transmission (eg, benzodiazepines such as clonazepam).(4,7)

The connection between nocturnal eating and RLS was first noted in 2006 by Roberto Vetrugno(5) and colleagues who noted a high prevalence of RLS and PLMs among 35 patients who had nocturnal eating episodes. One colleague, Federica Provini, later performed a study to determine whether the association noted in the Vetrugno study was significant or the result of population bias.(4)

The Provini study involved 100 people with RLS and 100 controls without RLS. The study participants were all from the same region in Italy to ensure genetic homogeneity. Telephone interviews and several questionnaires (ie, Beck Depression Inventory, Maudsley Obsessive-Compulsive Inventory, Bologna Questionnaire on Sleepiness Related Symptoms) were used to collect information regarding the frequency of nocturnal eating episodes; daytime sleepiness; sleep characteristics (ie, subjective sleep latency); and whether the person had eating disorders (ie, bulimia, anorexia), depression, and/or obsessive-compulsive traits.

Provini found that among the participants with RLS, 33% also had SRED. By contrast, only 1% of the controls (ie, no RLS) had SRED. This difference was statistically significant. Participants with RLS and SRED tended to have higher scores on the Maudsley Obsessive-Compulsive Inventory (MOCI), compared to participants with RLS only. For example, 15.2% of the participants with RLS and SRED scored as “pathological” on the MOCI, compared to 4.5% of the participants with RLS only. This suggests that neuropathological pathways involved in obsessive-compulsive disorders also may be involved in the compulsion to eat often reported by people who experience NES or SRED.

Some participants with RLS were using dopaminergic drugs or benzodiazepine/hypnotic drugs to treat their RLS symptoms; however, the use of these drugs showed no significant difference in the prevalence of SRED between the treated and untreated participants. From these results, Provini concluded that there was indeed an association between RLS and SRED.


Narcolepsy is a sleep disorder that comprises four symptoms (the narcoleptic tetrad): excessive daytime somnolence; cataplexy; sleep paralysis; and hypnagogic hallucinations. Not all people with narcolepsy have all four symptoms. The most serious symptom in narcolepsy is cataplexy because it can endanger the life of the person with narcolepsy or others.

In recent years, scientists have determined that people with narcolepsy-cataplexy have low-to-undetectable cerebrospinal fluid levels of the hormone hypocretin, which plays a role in hunger and appetite.(6) This reduction in the hypocretin level results from the loss of hypocretin-producing neurons in the hypothalamus (a brain structure that plays a role in hunger and in the transition between wake and sleep). Based on this, scientists believe that low hypocretin levels may play a role in cataplexy.

People with narcolepsy often are overweight or obese and some people with narcolepsy-cataplexy struggle with food craving. Some research indicates an increased prevalence of eating disorders such as bulimia in people with narcolepsy-cataplexy.(8)

Researchers Vincenzo Palaia(8) and colleagues recently examined the prevalence of SRED in people with narcolepsy-cataplexy. The participants were narcolepsy patients and age- and sex-matched unrelated non-narcoleptic controls from the same geographical area in Italy. All underwent interviews that collected demographic information (eg, age; weight; body mass index; habits such as smoking and consumption of alcohol; medical disorders; medications; and subjective sleep characteristics, such as sleep latency and the number of nocturnal awakenings). All participants also filled out questionnaires that assessed, among other factors, the presence of RLS (International Restless Legs Syndrome Study Rating Scale); eating disorders (Eating Disorder Inventory-2); obsessive-compulsive traits (MOCI); and sleepiness (Epworth Sleepiness Scale).

They found that the prevalence of SRED was higher in participants with narcolepsy-cataplexy (32%) than in the controls (3%) and that participants with narcolepsy-cataplexy had higher scores for obsessive-compulsive traits. From these results, Palaia proposes that hypocretin deficiency may impair impulse control through its effect on dopaminergic neurons, and thereby contribute to SRED.

During the day, hypocretin normally promotes wakefulness, locomotion, and food seeking. Therefore, Palaia further proposes that the loss of hypocretin neurons may allow dysregulation between wake and sleep and set the stage for the nocturnal eating behavior.

New Avenues for Treatment?

A better understanding of the association between SRED/NES, RLS, and narcolepsy may result in improved treatment for these sleep disorders, as well as other disorders. For example, if it is determined that the amnesia noted in SRED is a drug-induced effect, then avoiding the amnesic drugs may reduce or eliminate SRED and its consequences (eg, weight gain, sleep-related injury).

In one study,(7) a female patient had developed SRED after abusing the drug triazolam, which she had been taking for depression; after withdrawal from the drug, the SRED resolved. However, studies are needed to determine if SRED is a drug-induced effect and to what extent using other types of sedative-hypnotic drugs can avoid this effect. 

People with RLS are often misdiagnosed as having insomnia because of complaints of frequent arousals during the night; they are consequently treated with BDZ to improve sleep. For some, this may set the stage for the development of SRED. Using a drug that does not act on the BDZ receptor may be more beneficial. However, studies are needed to determine which patients with RLS are at greatest risk of developing SRED. If the loss of hypocretinergic neurons is proven to play a role in SRED in people with narcolepsy, then treatment may focus on altering the function of the hypocretinergic system (ie, by stimulating the remaining neurons to release more hypocretin or by using a drug that mimics the actions of hypocretin).

These questions potentially offer new avenues for treating SRED/NES, RLS, and narcolepsy. For now, scientists remain intrigued by the apparent connection between these sleep disorders.

Regina Patrick, RPSGT, is a contributing writer for Sleep Review.


1. Stunkard AJ, Grace WJ, Wolff HG. The night-eating syndrome; A pattern of food intake among certain obese patients. American Journal of Medicine. 1955;19(1):78-86.
2. Lundgren JD, Allison KC, O’Reardon JP, Stunkard AJ. A descriptive study of non-obese persons with night eating syndrome and a weight-matched comparison group. Eat Behav. 2008;9(3):343-351.
3. Birketvedt GS, Florholmen J, Sundsfjord J, et al. Behavioral and neuroendocrine characteristics of the night-eating syndrome. JAMA. 1999;282(7):657-663.
4. Provini F, Antelmi E, Vignatelli L, et al. Association of restless legs syndrome with nocturnal eating: a case-control study. Mov Disord. 2009;24(6):871-877.
5. Vetrugno R, Manconi M, Ferini-Strambi L, Provini F, Plazzi G, Montagna P. Nocturnal eating: sleep-related eating disorder or night eating syndrome? A videopolysomnographic study. Sleep. 2006;29(7):949-954.
6. Howell MJ, Schenck CH. Restless nocturnal eating: a common feature of Willis-Ekbom syndrome (RLS). J Clin Sleep Med. 2012;8(4):413-419.
7. Schenck CH, Hurwitz TD, Bundlie SR, Mahowald MW. Sleep-related eating disorders: polysomnographic correlates of a heterogeneous syndrome distinct from daytime eating disorders. Sleep. 1991;14(5):419-431.
8. Palaia V, Poli F, Pizza F, et al. Narcolepsy with cataplexy associated with nocturnal compulsive behaviors: a case-control study. Sleep. 2011;34(10):1365-1371.