Clinical manifestations

Presentation and course

The onset of sleep-related eating disorder can be sudden, associated with major life-stresses, medical, neurologic, sleep, or psychiatric disorders, with administration of neuroleptic medications, or cessation of cigarette smoking, alcohol abuse, or marijuana use. It can also develop gradually, without any identified precipitant. The course is often progressive. In a series of 19 patients, 74% were women with a mean age at presentation of 38 years (range 18 to 54 years), and a mean age of nocturnal eating onset of 25 years (range 5 to 44 years) (45). A prevalence of women (83%) and a young age of onset (21.6 years as a mean) were likewise reported in a series of 23 cases (56). Common features of sleep-related eating disorder episodes include consumption of high-calorie foods, bingeing but not purging, elaborate but sloppy food preparation, including cooking, burns, and lacerations from careless food preparation and cooking, and idiosyncratic concoctions such as cat food or salt sandwiches. Alcohol consumption during nocturnal episodes is rare. Patients usually get out of bed to eat and drink after having normally fallen asleep. Hunger, abdominal pain, and symptoms of hypoglycemia preceding episodes of nocturnal eating do not occur. Eating is compulsive, unassociated with any feeling of hunger, and chewing starts right after awakening (eating latency is less than 30 seconds in about half of the episodes) (46). Eating may occur up to eight times nightly, although one or two episodes nightly are more common (43). During the episodes, which last an average of 3.5 minutes (46), 90% of patients report being half or fully asleep, and describe partial or complete amnesia in the morning (56; 43), although there may be associated dreamlike mentation.

In different reports impaired consciousness was not a universal finding (46; 52; 07), and based on these results the American Academy of Sleep Medicine includes a partial or complete loss of conscious awareness during the eating episode with subsequent impaired recall as a required diagnostic criterion for sleep-related eating disorder (ICDS III) (02).

Major concerns expressed by patients about sleep-related eating include weight gain, injuries, choking on food, starting fires while cooking, potential for toxic ingestion, medical and dental consequences, and loss of control. Moreover, weight gain may precipitate or aggravate preexisting diabetes mellitus, hyperlipidemia, hypercholesterolemia, hypertension, and obstructive sleep apnea. Gastrointestinal disorders, including peptic ulcer disease, chronic autoimmune hepatitis, and anorexia nervosa in remission, are rarely associated with sleep-related eating.

Polysomnography identified an associated sleep disorder (such as narcolepsy, periodic limb movement disorder, and restless legs syndrome) in half the cases (46). Higher levels of depression and dissociation have been reported (56; 43).

A series of 35 patients from two sleep centers in Italy were reported with sleep-related episodes of compulsive eating (52). These patients did not have mental confusion and exhibited an appropriate behavior during the nocturnal eating. Restless legs syndrome or increased periodic limb movement in sleep were observed on polysomnography in 27 of the 35 patients, and recurrent chewing and swallowing movements during sleep were a feature in 29 patients, associated in about half of the events with electroencephalographic arousals. In this study, it was concluded that night eating syndrome and sleep-related eating disorder may represent different entities, not to be compounded together, and that the increased periodic limb movements during sleep, the chewing activity, and the compulsory food-seeking behavior could represent features of a dopaminergic dysfunction amenable to treatment. A similar distinction between night eating syndrome and sleep-related eating disorder was drawn in another polysomnographic study of 15 women with nocturnal eating (41), in which the polysomnographic findings were those of maintenance insomnia reflecting a state of internal circadian desynchrony rather than those of a true arousal disorder.

Brion and colleagues, studying 15 patients with sleep-related eating disorder, found several commonalities with sleepwalkers, including a high frequency of past or current episodes of sleepwalking, a similar timing of parasomnia episodes during the first half of the night, and numerous arousals from sleep stage 3 (07). However, patients with sleep-related eating disorder had higher awareness during sleep episodes. Conversely, only 10% of sleepwalkers ate during their sleepwalking. These results could suggest that patients with sleep-related eating disorder have specialized a former sleepwalking behavior toward sleep-related eating.

Prognosis and complications

Spontaneous remissions are rarely reported, and most cases appear to be progressive. Complications include obesity, injuries (cuts, burns), toxic ingestions, and psychological distress associated with excessive weight gain and repeated loss of control during sleep. Adverse outcomes of night eating behaviors may be particularly relevant to patients already affected with diabetes mellitus.

Clinical vignette

A 74-year-old man who worked night shifts from the ages of 60 to 70 years old began having problems with his nocturnal sleep when he retired at the age of 70. He was able to fall asleep without difficulty, but would then wake up in the night four to seven times due to an inability to keep his legs still. At the same time, he would feel an urge to get out of bed and eat a few biscuits, or some yogurt or crackers, even without feeling any hunger. He would usually stay up for a few minutes, wander around and eat, and then go straight to bed and fall asleep, only to wake up again in the same night and go through the same process all over again. He reported feeling confused and complained of sleepiness and difficulties with mental concentration in the morning. According to his wife, the patient had recurrent movements during sleep of his feet and legs and was an occasional snorer. Physical examination, neuropsychological evaluation, blood chemistry with glycated hemoglobin, and brain CT were all normal. A multiple sleep latency test showed pathologic daytime sleepiness (mean latency 3’ 20”) and a 24-hour polysomnography showed periodic limb movements in sleep, also involving the upper limbs. On one occasion during the recording, the patient was compelled to leave the bed and walk around in the lab. The patient underwent a 2-day Vitaport ® recording under audio-visual control, which showed a total of seven NREM sleep episodes during daytime, periodic limb movements in sleep during light sleep stages, and several awakenings during the night, one (lasting 10 minutes) characterized by chewing and eating a snack placed near the bed. The patient was diagnosed as affected with sleep-related eating disorder and with restless legs syndrome with associated periodic limb movements in sleep. He was given pramipexole 0.18 mg at bedtime. One month later, on follow-up, he reported a striking amelioration of both restless legs and the eating behavior, which had now almost completely disappeared.

Biological basis

Etiology and pathogenesis

The etiology of sleep-related eating disorder is still unclear. An indeterminate etiology is not uncommon. Sleepwalking, a disorder of arousal arising from NREM sleep, is, however, frequently found in sleep-related eating disorder. Periodic limb movement disorder (with or without restless legs syndrome), obstructive sleep apnea, and triazolam abuse may be contributing factors in some cases. Problematic amnestic nocturnal eating has also been reported to be associated with treatment with zolpidem, a nonbenzodiazepine hypnotic used for complaints of insomnia. Some cases of somnambulism or SRED have been reported in patients treated with antipsychotics and lithium (16).

Nocturnal eating syndrome is also associated with a distinct psychopathology (anxiety, depression, and low self-esteem) (14). A relationship with unclear self-definition, faulty interpersonal communication, low frustration threshold, or with undreamt or avoided dreams as a defense against perception of the unconscious have been proposed as psychodynamic explanations. Depression is a frequent accompaniment (41).

Dalle Grave and colleagues investigated the personality characteristics of 568 obese women consecutively seeking treatment at eight Italian medical centers and 185 age-matched, normal weight women without symptoms of eating disorders (10). The authors concluded that the presence of disordered eating is associated with specific personality characteristics: personality traits of obese individuals included significantly lower self-directedness and cooperativeness on Temperament and Character Inventory (TCI). Binge Eating Scales (BES) and Night Eating Questionnaire scores were higher in obese women, and high BES values were associated with high novelty seeking and harm avoidance and low self-directedness.

On the other hand, it is unclear whether neuroendocrine dysfunctions are a cause or effect of night eating. Marked blunting of the nocturnal rise in melatonin and absent nighttime increase in leptin hormones associated with 24-hour increased levels of cortisolemia characterize nocturnal eating syndrome. In this setting, the attenuation in melatonin rise at nighttime has been suggested to contribute to sleep maintenance insomnia, and the failure of leptin to impair suppression of appetite, the increased cortisolemia would indicate increased levels of stress. The relevance of such psychogenic hypotheses and hormonal findings in night eating syndrome to sleep-related eating disorder is uncertain. In this context, it is worthwhile remarking that patients with daytime bulimia nervosa have substantially normal sleep structure (11; 01).

The pathogenesis and pathophysiology of sleep-related eating is unknown. Night eating syndrome is considered a circadian disorder, as borne out by circadian studies that document that night eaters suffer a phase delay in energy consumption relative to sleep-wake times, which would involve a dissociation of the circadian control of eating relative to sleep (36; 41).

Alternatively, night eating syndrome has been considered to be consistently related to mood disorders and anxiety (54). Howell and colleagues conclude that sleep-related eating disorder and night eating syndrome are situated at the opposite end points of a disordered eating spectrum: although night eating syndrome may be considered an abnormality in the circadian rhythm of meal timing with a normal circadian timing of sleep onset, the feeding behavior observed in sleep-related eating disorder is characterized by recurrent episodes of eating after an arousal from nighttime sleep with or without amnesia (20). Moreover, sleep-related eating disorder is often associated with other sleep disorders, in particular the parasomnias.

In night eating syndrome, the abnormal leptin and melatonin rhythms, together with evidence that sleep loss reduces diurnal rhythm amplitude of leptin and that nocturnal lifestyle is associated with impairment of insulin response to glucose and, hence, to obesity, suggest a hypothalamic dysfunction. There is growing evidence that night eating syndrome and obesity may be involved in a cyclic relationship: circadian behavior misalignment and obesity may be involved in a cyclic relationship driven by behavior and circadian genetic variations, although no studies have focused on specific genes involved in night eating syndrome. An abnormality of circadian rhythms was suggested by the finding that patients with night eating syndrome show a phase delay in the timing of meals and delayed circadian rhythms for total caloric, fat, and carbohydrate intake. In addition, phase delays of 1.0 to 2.8 h were found in the food-regulatory rhythms of leptin and insulin and in the circadian melatonin rhythm, whereas circulating levels of ghrelin were phase advanced by 5.2 h (15). Birketvedt reported variable attenuated circadian secretion of ghrelin and increased circadian secretion of prolactin in night eating syndrome independent of being overweight (06). It is remarkable in this context that mice fed a high-fat diet only during the 12-h light phase gain significantly more weight than mice fed only during the 12-h dark phase, findings that emphasize the role of the circadian phase of food consumption in the control of weight gain. Orexins (from the Greek orexis, meaning appetite), also known as hypocretins, are neuropeptides secreted by the lateral hypothalamic areas and implicated in the regulation of both feeding and wakefulness. Orexins seem, therefore, interesting candidate substances in the mechanisms of sleep-related eating disorder, but their pathogenic role has not been examined yet. On the basis of the favorable response to serotonin agonists and serotonin reuptake inhibitors, inadequate serotonin levels have been implicated in the genesis of night eating disorders. Night eating behavior was shown to develop in EP3R-deficient mice. Moreover, mice deficient in the clock protein mPer2 do not have a glucocorticoid rhythm; show an abnormal rhythm of peptide alphaMSH, a major effector of appetite control in the hypothalamus; and lack a diurnal feeding rhythm. On high-fat diet, they eat as much during the light period as they do during the dark period and develop significant obesity (60). These animal models may provide a genetic background of disordered circadian rhythms in the eating disorders.

Epidemiology

Night eating in the general population is most common during the weekend and prevalence is greatest among young adults 18 to 30 years of age; it is not associated with obesity (47; 42). Nocturnal eating, defined as an episode appearing at least once a week, with a self-reported level of awareness, occurs in about 5% of young adults, and in 9% to 17% of patients with a variety of eating disorders (57; 40). A familial basis has also been reported for sleep-related eating disorder (12). The overlap between binge eating disorder and nighttime eating problems was indeed modest according to de Zwaan and colleagues (13), arguing for the distinctiveness of night eating syndrome from binge eating disorder. However, night eating syndrome was found to be associated with binge eating, increased body mass index, and male gender, and elevated psychological distress only in those who consumed nocturnal snacks. Night eating syndrome occurs more frequently in patients with bipolar disorder. Bipolar disorder patients with night eating syndrome present more disease-related manifestations, such as more anxiety, poorer functioning, and worse sleep parameters (32). Yamada and colleagues reported a night eating prevalence of 0.8% in a population of 1169 outpatients with diabetes (59). Nocturnal eating was, in particular, related to nocturnal anxiety among night eating syndrome patients, whereas it was related to diurnal anxiety among patients affected by binge eating disorder (44). In a study including 909 university students sampled from three provinces in China, the prevalence of night eating syndrome was 2.8% (2.4% after excluding those with binge eating), with males being more likely to have night eating syndrome (17). Moreover, night eating behaviors were closely related to higher psychological distress. Controversial data were reported on the association between nocturnal eating disorder and increased body mass index. Nocturnal eating in the form of night eating syndrome occurred in 14% of patients with obstructive sleep apnea, independently of whether obese or not. According to Olbrich and colleagues, however, night eating syndrome did not appear to be closely linked to obstructive sleep apnea, even though a significant comorbidity with psychiatric disorders in these patients should be expected (34). In an epidemiological study in Italy, the prevalence of nocturnal eating (defined as sleep-related eating disorder) was found to be increased significantly in patients with restless legs syndrome compared to controls (33% vs. 1%, P < 0.001). Medication use and pathologic obsessive-compulsive scores were more prevalent in restless legs syndrome patients with nocturnal eating than among restless legs patients without nocturnal eating. An association between the restless legs syndrome and sleep-related eating disorder was indicated by these findings and tentatively attributed to an underlying common abnormality in dopaminergic metabolism (39). Indeed, eating as a compulsive symptom was found to be induced by dopaminergic medications in patients with restless legs syndrome and psychiatric comorbidity (37). However, in a study by Howell and Schenck, patients with restless legs syndrome reported more frequently nocturnal eating than patients with psychophysiological insomnia (61% vs. 12%, respectively) and amnestic sleep-related eating disorder was more common among patients on sedative-hypnotics (19). The authors failed to find any cases of dopaminergic agents inducing novel “nocturnal eating” or sleep-related eating disorder. Patients with restless legs syndrome demonstrated a substantial drop (68% to 34%) in the frequency of “nocturnal eating” after dopamine therapy. All these findings suggest that “nocturnal eating” could be viewed as a “non-motor” feature of restless legs syndrome (04; 22).

Prevention

Risk factors tentatively identified for sleep-related eating include irregular sleeping habits; especially nocturnal lifestyle; irregular daytime eating pattern; periodic leg movements in sleep and restless legs syndrome; stress; cessation of cigarette smoking and of drug abuse; administration of benzodiazepines, neuroleptics, and anticholinergics; premenstrual state; autoimmune hepatitis; narcolepsy; encephalitis; and travel (26; 50; 23). Notably, amnestic sleep-related eating behavior has been observed to occur after zolpidem administration for insomnia (18; 51) or for restless legs syndrome, and particularly after extended-release but not immediate-release formulations of zolpidem, suggesting that even small pharmacotherapeutical differences may affect the development of parasomnias. According to one study, sleep-related eating disorder due to zolpidem usually disappears after withdrawal of the drug (51). Nocturnal eating syndrome also has been described as a possible adverse effect of zaleplon and of antipsychotic medications in patients with schizophrenia (25; 16). Pramipexole and sodium oxybate have been reported to trigger sleep-related eating disorder (55).

Differential diagnosis

Confusing conditions

Differential diagnosis with the daytime eating behavior disorder (bulimia nervosa and binge eating disorder) may be difficult, especially because daytime bulimia nervosa at times may extend to nocturnal hours (01).

Clear similarities but also differences between night eating syndrome and sleep-related eating disorder were found by several authors (24). It is essential to study night eating syndrome as an independent disorder to further develop its diagnostic criteria and treatment options, thereby, increasing the quality of life of the patients suffering from this syndrome. Patients with night eating syndrome or binge eating disorder may indeed exhibit awakenings with automatic eating and drinking and amnesia on awakening in the morning, similar to sleep-related eating disorder. Frequent purging and induced vomiting are, however, not typical features of sleep-related eating disorder. Absence of morning anorexia (skipping breakfast) also distinguishes sleep-related eating disorder from night eating syndrome. According to Stunkard and colleagues, the main feature differentiating night eating syndrome from sleep-related eating disorder is the lack of any consciousness impairment during eating in the former (48). Wakeful and appropriate behavior was indeed possible (46) or even the rule (41; 52) on polysomnographic recordings of the eating episodes. Furthermore, the effects of medications on vigilance levels during the eating episodes and those of sleep inertia, which is physiological when awakening from deep sleep, should always be taken into consideration when evaluating the level of consciousness during any nocturnal eating episode.

Finally, sleep-related eating disorder should be differentiated from the so-called Kleine-Levin syndrome, in which recurrent hyperphagia is associated with hypersomnia, hypersexuality, and irritability.

Diagnostic workup

Evaluation includes a sleep-wake, medical, and psychiatric history, review of systems, examination, and video-polysomnographic monitoring. Psychiatric assessment may be indicated if a waking eating disorder is suspected. Particular attention should be paid to symptoms suggestive of other parasomnias, particularly sleepwalking without eating, periodic limb movements during sleep and restless legs, and symptoms potentially related to daytime eating disorders. Video-polysomnographic monitoring should comprise additional channels devoted to monitoring EEG and limb movements. On polysomnography, the episodes of sleep-related eating disorder usually originate from NREM sleep, especially stages 1-2, and occasionally from REM sleep, with a short latency between awakening and the beginning of the episode (less than 30 seconds in half of the cases) (46; 52). Usually, except for the increased awakenings, there are no gross alterations in sleep structure such as those in patients affected by night eating (53). Sleep latency may be normal or increased; REM sleep latency is normal. Sleep efficiency is, however, decreased (66% to 80%). Chewing and swallowing automatisms may be recorded throughout the various sleep stages (46) often in the form of rhythmic masticatory motor activity associated with cortical arousals (52). Periodic limb movements during sleep are typically found in those cases with concomitant restless legs syndrome.

Management

Treatment of an underlying sleep disorder, if present, is often effective in controlling the nocturnal eating, sleep-related eating disorder (08). Hypnotics appear to be ineffective. In a retrospective study including 49 patients with sleep-related eating disorder/night eating syndrome, ramelteon significantly reduced the mean frequency of eating behavior (per week) (30). In single case reports, melatonergic drugs lead to relief from sleep-related eating, even when the condition occurs in the context of physical and mental disorders as well as psychopharmacological treatment (61). For patients with sleepwalking, periodic limb movement disorder, and restless legs syndrome, clonazepam, codeine, and carbidopa/L-dopa or bromocriptine are often effective (22). Low doses of pramipexole were reported effective in a randomized, double-blind, placebo-controlled trial (38). In other studies, the use of dopaminergic agents has been implicated in triggering impulse control disorders such as pathological gambling, hypersexuality, compulsive shopping, compulsive eating, and punding in patients with Parkinson disease or restless legs syndrome (33; 09). Remission or reduction of compulsive behaviors is reported with reduction or cessation of dopaminergic therapy. Fluoxetine may also help, either alone or associated with dopaminergic agonists, especially in patients with depression and a history of substance abuse. The latter may respond to a combination of bupropion, levodopa, and trazodone. Serotonin reuptake inhibitors have been shown effective in the treatment of both sleep-related eating disorder and night eating syndrome (35; 48). Topiramate, an antiepileptic drug, given at a dose range of 100 to 400 mg at night, reduced night eating, improved nocturnal sleep, and produced weight loss in patients with night eating syndrome or sleep-related eating disorder (58). Other antiepileptic medications such as zonisamide, valproate, carbamazepine, and phenytoin might have a role in the treatment of eating disorders, but controlled trials are still lacking. Cognitive-behavioral strategies, hypnotherapy, and psychotherapy are reported as usually ineffective. An abbreviated progressive muscle relaxation therapy and/or bright light exposure were reported to provide relief of stress and lower hunger and nighttime eating, improving subjective sleep quality (05; 31).

Outcomes

Night eating syndrome is a clinical entity that requires further investigation, especially concerning adolescents and the development of symptoms during the transition into adulthood (27).

Special considerations

Data from the literature showed that parasomnias are found in patients that consume or are going through withdrawal from several different drugs, suggesting a complex relationship between different factors, such as changes in neural system functioning, the presence of comorbid sleep disorders, pharmacological treatment, stress, and antecedents of parasomnias in childhood or adolescence (23).