Psychiatrists are waking up to the world of sleep and behavior. Just as the biological revolution witnessed a collapse of barriers between the physical and the mental, so the boundary between sleep and wakefulness is crumbling. The emerging discipline of sleep medicine, baptized in the discovery of REM sleep in 1953 and by the characterization of sleep cycles a few years later, initially focused on adult physiology and sleep disorders.
Sleep, cognition, and behavior
Serious sleep problems usually imply either too much or too little sleep. Too little sleep can mean abbreviated sleep episodes, as might be imposed by external demands or by insomnia, or longer episodes punctuated by arousals that fragment sleep. Pilcher and Huffcutt1 examined the effects of prolonged sleep deprivation (more than 45 hours), brief sleep deprivation (45 hours or less), or partial sleep deprivation (less than 5 hours of sleep in a 24-hour period) on measures of cognitive and motor performance or mood. The overall performance decrement for all measures was 1.37 standard deviations (SDs). This is comparable to a drop in IQ from 100 to about 80. Under the conditions of partial sleep deprivation—not unlike a typical Monday morning for many of us—mood scores dropped by over 4 SDs.
Although children are not suitable subjects for such experiments, Sadeh and colleagues2 demonstrated that very modest changes in sleep duration can substantially affect the neurobehavioral function of children. His team monitored 77 fourth and sixth graders in regular classrooms with actigraphy for 5 nights. For the first 2 nights, children slept as usual; on the remaining 3 nights, they were asked to either extend or restrict their sleep time by 1 hour. Children who failed to change their sleep duration by at least 30 minutes were analyzed as “no change.” On a simple reaction time test, performance of both the sleep restriction and no-change groups deteriorated, whereas on measures of digit span and continuous reaction time, children in the extended-sleep group performed significantly better than the others.
When Gruber and associates3 applied a single hour’s sleep restriction to 11 children with ADHD and to 32 controls over 6 days, neurobehavioral function in both groups deteriorated, as assessed by a continuous performance test. For children with ADHD, two-thirds of the continuous performance test measures dropped from the subclinical to the clinical range of impairment.
If the slight manipulations in sleep duration have measurable effects after just 3 nights, what might the effects be when sleep is shortened or fragmented long-term by busy family schedules, sleep-related breathing disorders (SRBD), behavioral insomnia of childhood, or other sleep problems? Would it be possible to produce the full clinical picture of ADHD under such circumstances?
Prevalence of sleep disorders
Sleep problems described in children with ADHD include behavioral problems that often manifest as resistance to bedtime or difficulty in going back to sleep in the middle of the night without caregiver presence, parasomnias, sleep-disordered breathing, restless legs and periodic leg movements, and circadian rhythm disorders.4 Similarly, children with ADHD experience alterations in sleep architecture, such as delayed onset of REM, reduced time in REM, and arousals with sleep fragmentation.5 Inattention, hyperactivity, neuropsychological deficits, syndromic ADHD, and behavior disorders are also common among children who present with sleep problems. Yet despite abundant evidence of these comorbidities, the extent and nature of possible causal relationships between disturbances in sleep and behavior remain largely a mystery.
Prevalence estimates vary because of the way sleep disorders are operationally defined. Subjective parent complaints about resistance to bedtime, a behavior that is not unusual in children, may be counted as a significant problem or disorder by one investigator but not by another. Laboratories may use different procedures or scoring schemes that produce somewhat inconsistent counts of polysomnographic events, such as leg movements and hypopneas. Similarly, heterogeneous diagnostic rules may yield disparate rates of disorder, ie, if 5 rather than 1 apnea or hypopnea episodes per hour define obstructive sleep apnea, fewer cases with more severe symptoms on average will result. Finally, deviations in sleep architecture detected on polysomnography may be consequential yet may not be indicative of a specific sleep diagnosis. Similarly, diagnoses of ADHD depend on the criteria set, subjective thresholds for counting a behavior as symptomatic, and whether diagnoses are based on rating scales, clinical interviews, or structured and comprehensive clinical assessments.
Rates of sleep disorder in ADHD also depend, sometimes dramatically, on how and where data are obtained. Parent checklists almost always generate high sleep symptom rates, exceeding those that can be demonstrated during formal sleep studies. Research subjects recruited in specialty clinics (such as psychiatry or sleep disorders clinics) often display both sleep and behavior problems, whereas children assessed in regular school settings exhibit fewer problems from either domain.
Further complexity arises from potential nonlinearity in the impact of sleep problems on both behavior and behavioral diagnosis. Thus, mild to moderate levels of sleepiness could produce excessive activity and reduced attention, while very high levels of sleep disturbance may cause torpor and inattention so obviously related to sleep that a separate diagnosis of ADHD would be improvident even if criteria were technically fulfilled.
For all of these reasons, the nature and extent of the relationship between ADHD and sleep remains unsettled. One widely quoted estimate for the prevalence of sleep disorders in ADHD places the rate of parent-reported sleep problems at 25% to 50% in the absence of medication.6 Spruyt and Gozal,7citing this study, give a guess-estimated rate of sleep problems in children with ADHD as 5-fold greater than that in healthy controls.
High-end estimates of sleep disorder prevalence in ADHD populations include a report from an Italian clinic that yielded polysomnography-confirmed sleep disorders, such as periodic limb movements in 40%, restless legs syndrome in 26%, SRBD in 18%, and confusional arousals in 36%.8 Findings from one study showed that 73% of children with ADHD had sleep problems and 45% of those had moderate to severe problems.9
Inconsistencies among reports have clouded interpretation of findings relative to sleep morbidity and ADHD. A recent meta-analysis produced compelling findings for both subjective and objective sleep measures in children with ADHD. In their analysis, Cortese and associates10 minimized the likelihood that comorbid conditions could account for sleep problems by excluding studies in which children were receiving pharmacological treatment or had comorbid depressive or anxiety disorders. Compared with controls, children with ADHD had significantly higher bedtime resistance and more sleep onset difficulties, night awakenings, morning wakening difficulty, sleep-disordered breathing, and daytime sleepiness. Among objective measures, sleep onset latency, stage shift frequency, and apnea-hypopnea index were increased in children with ADHD, while sleep efficiency, true sleep time, and average time to fall asleep were lower than in controls.
Notwithstanding our inability to identify true prevalence rates, studies purporting to show frequent sleep disorders are probably a fair reflection of the rate at which clinicians who see children with ADHD can expect to hear complaints of sleep problems after a careful inquiry using some combination of rating scale or checklist and interview. Children referred for diagnosis and treatment of sleep disorders or for adenotonsillectomy are likely to display psychiatric problems in which inattention, hyperactivity, or both are prominent.
Restless legs syndrome and periodic limb movements
In their meta-analyses, Sadeh and colleagues11 found that the frequency of periodic limb movements was consistently elevated in children with ADHD. The comorbidity of restless legs syndrome symptoms in ADHD has been reported to be as high as 24%.12,13 Frequent parent reports of prominent periodic limb movements, restless legs syndrome, and growing pains in their children with ADHD offer a plausible mechanistic link between sleep movement disorders and ADHD, namely, a functional dopamine(Drug information on dopamine) deficit.12 Both restless legs syndrome and periodic limb movement disorder can be successfully treated with dopamine agonists, much as ADHD usually responds to stimulant medication.
The diagnosis of restless legs syndrome is based on history and requires that the child offer some description of subjective symptoms. Children describe the discomfort that compels movement in restless legs syndrome in idiosyncratic ways; patience and creativity may be needed to understand such communications.
Most patients who experience restless legs syndrome also have periodic limb movements, although the converse is not necessarily true. In contrast to the voluntary movements of restless legs syndrome, periodic limb movements are involuntary and have a distinctively “neurological” appearance. The movements last from 0.5 to 5 seconds and recur in cycles of 5 to 90 seconds, typically in clusters, during non-REM sleep.
Both restless legs syndrome and periodic limb movement disorder can reduce the amount or quality of sleep. Restless legs syndrome prevents sleep onset and the return to sleep after awakenings at night, whereas periodic limb movement disorder may cause arousal and sleep fragmentation. Occasional limb movements are not necessarily problematic. Treatment recommendations for periodic limb movement disorder and restless legs syndrome are sleep hygiene, iron supplementation, and medications.
The elements of good sleep hygiene are presented in the Table. Common over-the-counter drugs such as antihistamines, cold preparations, and antiemetics aggravate restless legs syndrome symptoms. Recently, SSRIs have been implicated as a cause of periodic limb movements in nearly one-third of the children who receive these drugs.14
Iron deficiency, which is usually diagnosed in children as a ferritin level of 50 ng/mL or less, is associated with restless legs syndrome and periodic limb movement disorder. Small trials of iron supplementation have shown promise in children.15
The literature contains limited guidance for treating children with medications commonly recommended for adults with restless legs syndrome, such as the first-line, nonergot dopamine agonists (ropinirole and pramipexole(Drug information on pramipexole)); levodopa/carbidopa; and other agents, including clonidine(Drug information on clonidine), clonazepam(Drug information on clonazepam), opioids, and gabapentin(Drug information on gabapentin). Ropinirole(Drug information on ropinirole) and gabapentin enacarbil, a gabapentin prodrug, are the only agents approved for the restless legs syndrome indication in adults. None of these drugs have been approved or systematically studied in children. Use of dopaminergic agents may be complicated by the phenomenon of augmentation, in which symptoms begin to develop earlier in the day or spread to other parts of the body, giving rise to a condition suspiciously similar to akathisia.
Sleep-related breathing disorders
Obstructive sleep apnea is the prototypical SRBD that includes but is not limited to conditions characterized by conspicuous apneas. Most psychiatrists are familiar with the obstructive sleep apnea syndrome in adults, but the SRBD of children differs mechanistically from that of adult sleep apnea and the treatment is usually different as well. In many children with obstructive sleep apnea, enlarged adenoids and palatine tonsils are the underlying anatomic problem.
Described in vivid detail by Sir William Osler16 in 1892, SRBD can have dramatic neurobehavioral consequences: “Chronic enlargement of the tonsillar tissues is an affection of great importance, and may influence in an extraordinary way the mental and bodily development of children. . . .”
Patients with obstructive sleep apnea present with snoring, sleepiness during the day, difficulty in rising in the morning, unsatisfying sleep, nasal speech, mouth breathing, a damp pillow from drooling while asleep, and a distinctive appearance known as adenoid facies. Sleep-disordered breathing is more likely in the presence of developmental anomalies that affect the size or tone of the upper airway, such as deviated nasal septum, retrognathia, midline clefts, or Down syndrome. As Osler emphasizes, severe cases predictably cause mental dullness and typically languid, sluggish behavior rather than locomotor hyperactivity. The treatment is adenotonsillectomy.
Numerous studies have examined the impact of sleep-disordered breathing on inattention, hyperactivity, syndromal ADHD, and other disruptive behavior and on neuropsychological dysfunction, with remarkably inconsistent findings. Compared with controls, children referred for adenotonsillectomy had greater neuropsychological impairment and higher rates of ADHD and oppositional-defiant behavior.17,18 Unexpectedly, postsurgical improvement in these neurobehavioral indices was no greater among the children whose presurgical sleep studies confirmed SRBD.
Other studies of adenotonsillectomy confirm improvement in preoperative behavior and cognitive problems postoperatively, but it is not clear whether changes in sleep or sleep apnea measures explain the effects of surgery.19-21 In most correlative studies of sleep-disordered breathing, behavior, and cognitive problems, the relationships among sleep and behavioral variables have been difficult to demonstrate. For example, in a community sample of 571 school-aged children who underwent overnight polysomnography, no relationship between mild sleep apnea and multiple measures of neuropsychological functioning were detected.22 In contrast, another group of children who underwent similar examinations displayed neurocognitive deficits at all levels of sleep-disordered breathing.23
It is reasonable to say that there is an association between sleep-disordered breathing and neurobehavioral problems. Adenotonsillectomy often remediates these problems, but the nature of the association and of the treatment effects has been elusive. For children in whom adenotonsillectomy is ineffective or inappropriate, nasal continuous positive airway pressure may be effective. For especially severe cases, tracheostomy is an option.
The effects of medications
It was long supposed that medication treatment accounted for most sleep problems associated with ADHD that could not be attributed to parenting. Consistent with this view, insomnia is a common adverse effect of stimulant medications. Similarly, the medication rebound effect—the return of baseline symptoms, perhaps even heightened symptoms, at day’s end following 2 or 3 doses of short-acting medication—often coincides with bedtime. Furthermore, some children fall asleep more readily after receiving an evening dose of stimulant medication.
Sleep-onset insomnia is not always evident when sleep latency is assessed objectively under controlled conditions. O’Brien and colleagues24 found no differences in sleep characteristics between medicated and unmedicated children with ADHD, although both groups had reduced REM and more sleep disturbances than children without ADHD. Sleep latencies for each of the groups were close to 30 minutes, which suggests that some aspect of the setting might have kept them all awake, thereby reducing any contrast that could be attributed to either ADHD or medication.
Galland and colleagues25 reached different conclusions in a study that included both within-subject comparisons (children with ADHD on and off methylphenidate(Drug information on methylphenidate)) and across-subjects comparisons (children with ADHD versus controls). Children with ADHD who were taking the drug had an increase in sleep latency of 29 minutes and a decrement in overall sleep time by 1.2 hours, while children with ADHD who had not received medication were indistinguishable from controls.
Sangal and colleagues26 obtained mixed results, finding that sleep latency increased by 39 minutes in children who received methylphenidate 3 times a day, while total sleep interval and the frequency and duration of interruptions decreased.
Longer-acting methylphenidate preparations reportedly have a minimal effect on sleep. Huang and colleagues27 recently reported that “randomized, double-blind, placebocontrolled trials have demonstrated that CNS stimulants do not cause a statistically significant increase in sleep problems, but showed that some beneficial effects were noted on the sleep of both children and adults with ADHD.” This would seem a very optimistic appraisal of the literature, which might be better characterized as showing mixed results: the most well-documented effect of stimulants is initial insomnia, the impact of which may be counterbalanced by fewer interruptions once sleep has been attained.
Atomoxetine, which for many clinicians is the preferred ADHD agent when trials of methylphenidate and amphetamine have failed, commonly produces somnolence, especially when used in higher doses or titrated rapidly. Compared with children who were being treated with methylphenidate, children treated with atomoxetine(Drug information on atomoxetine) fell asleep faster, arose more easily, and slept better.26
Various sedating drugs are popular adjuncts to stimulant treatment. Patients are likely to receive clonidine or guanfacine(Drug information on guanfacine), both of which are available in long- or short-acting preparations, for target symptoms of insomnia, hyperactivity/impulsivity, and aggression.28 Clonidine reduces REM sleep and may be effective in treating bruxism, but the drug may be difficult to discontinue. Morning sedation is also a common effect of taking clonidine at bedtime.
Neither clonidine nor guanfacine is approved for sleep or for ADHD, although extended-release formulations of both, which presumably are less soporific than their shorter-life parent drugs, have FDA indications for ADHD. Melatonin(Drug information on melatonin) can be used to promote sleep and may be especially useful with a delayed sleep phase disorder.
Although treatment recommendations always include simple nostrums about sleep hygiene, these amount to mere tautologies absent a focused therapeutic effort to help parents establish a realistic bedtime routine. In adolescents, these disrupted routines may be aggravated by stimulating beverages, urgent social matters, daytime naps, and other lifestyle changes. An Australian project has developed a protocol designed to offer guidance to parents that is based on the nature of sleep problems in children who have ADHD. For example, in sleep-onset association disorder, parents are instructed to gradually fade their presence at bedtime. For limit-setting disorder, children are rewarded for adherence to bedtime routines and are ignored when protesting the bedtime rules. Delayed sleep phase disorder is managed by a combination of fixed morning awakening time, gradually earlier bedtime, and morning phototherapy.29
There are several parent questionnaires for sleep disorder screening, including the Pediatric Sleep Questionnaire.30 Mindell and Owens31 offer a simple and popular acronym to guide screening: BEARS reminds us to inquire about bedtime problems, excessive daytime sleepiness, awakenings during the night, regular bedtime and awakening time, and snoring or difficulty in breath-ing during sleep. The University of Chicago’s detailed “Pediatric Sleep Medicine Questionnaire,” along with a 2-week pediatric sleep diary, can be downloaded from the University of Chicago, Comer Children’s Hospital.32
Sleep changes associated with psychotropic drugs are common enough to justify routinely obtaining a baseline sleep diary before beginning treatment, even when the initial screening for sleep disorders indicates that no further investigation is needed.
1. Pilcher JJ, Huffcutt AI. Effects of sleep deprivation on performance: a meta-analysis. Sleep. 1996;19:318-326.
2. Sadeh A, Gruber R, Raviv A. Effects of sleep restriction and extension on school-age children: what a difference an hour makes. Child Dev. 2003;74:444-455.
3. Gruber R, Wiebe S, Montecalvo L, et al. Impact of sleep restriction on neurobehavioral functioning of children with attention deficit hyperactivity disorder. Sleep. 2011;34:315-323.
4. Cortese S, Konofal E, Yateman N, et al. Sleep and alertness in children with attention-deficit/hyperactivity disorder: a systematic review of the literature. Sleep. 2006;29:504-511.
Click here for full reference list.