The aim of this study was to review the basic aspects of sleep disturbance in children with traumatic brain injury (TBI).
The aim of this study was to review the basic aspects of sleep disturbance in children with traumatic brain injury (TBI).
A search was performed on reports of sleep disturbances in children who had suffered TBI. Adults with TBI were also considered to anticipate the nature and significance of such disturbances in younger patients. Types of reported sleep disturbance were noted and their possible aetiology and management considered.
Sleep disturbance has consistently been associated with TBI but the literature suggests that this aspect of patient care is often inadequately considered and there has been little research on the subject, especially in relation to children. Excessive daytime sleepiness is often mentioned, less so insomnia and parasomnias, but there is little information about the specific sleep disorders underlying these problems.
Sleep disorders with potentially important developmental consequences have been neglected in the care of children with TBI. Screening for sleep problems should be routine and followed, if indicated, by a detailed diagnosis of the child's underlying specific sleep disorder, the possible aetiology of which includes neuropathology and potential medical, psychological, or psychiatric comorbidities. Appropriate assessments and modern treatment options are now well defined although generally underutilized. Further well-designed research is needed for which guidelines are available.
Excessive daytime sleepiness
International Classification of Sleep Disorders, 2nd edition
Traumatic brain injury
Adequate, good-quality sleep is essential, in a number of ways, for children's development. Sleep disturbance can adversely affect cognitive function and educational progress, and the emotional state and behaviour of a child. Increasingly, the adverse consequences for physical health resulting from persistent sleep loss or disruption are also being described, such as cardiovascular, endocrine, or metabolic effects as well as effects on growth and immunity. It is of concern, therefore, that for some time sleep disturbance has been reported to be common in children who have suffered traumatic brain injury (TBI), as poor sleep in such children may well compound developmental problems that are the result of the injury itself. Unfortunately, the sleep aspects of childhood TBI have received relatively little attention. No doubt this arises from the general neglect of sleep and its disorders in medical education and training.
Owing to the lack of adequate information, the overall prevalence rates for sleep disturbance in children with TBI are difficult to compile with any accuracy. To give some indication by analogy, a recent meta-analysis suggested that at least 50% of adult patients with TBI suffer from some type of sleep disturbance. This present review is concerned with such information (as it exists) for children and adolescents, and also issues of relevance for clinical practice and to much needed further research in the field. Given the particular neglect of the topic with regard to children, selected reports of sleep and TBI in adults are included to highlight issues that will inform clinical practice and research in younger patients with TBI.
Recent estimates from the USA and from other countries suggest that the annual overall rates of TBI are high. Children aged up to 4 years and older adolescents aged 15 to 19 years are the age groups most likely to sustain TBI, and each year almost half a million children aged 0 to 14 years present to the emergency department because of TBI. A prospective birth cohort study in New Zealand found that by 25 years of age about 30% of the population had experienced at least one TBI, and about a third who experienced a TBI went on to have one or more additional injuries. It is possible that figures such as these underestimate the prevalence as some children with mild TBI may well not receive medical attention.
As regards the prevalence of sleep disorders associated with TBI, exact figures are unavailable, although, as already indicated, there is reason to believe that such disorders are commonplace. General statements are of limited clinical value because sleep disturbance can be expected to vary with such basic factors as the type (closed head or penetrating) and severity (mild, moderate, or severe) of the injury, comorbidity, the age of the patient, and the point at which sleep is assessed and reassessed following the injury. The accuracy of reported findings will be limited in retrospective studies and the clinical value of the findings restricted where the assessment of sleep is cursory. Other methodological complexities are discussed later.
Findings in the following recent studies illustrate the frequent persistence of disturbed sleep, which clearly has implications for rehabilitation. Castriotta et al. reported the presence of sleep disturbance in 46% of adult patients at least 3 months after TBI; Baumann et al. described 72% of adults with TBI as having sleep disturbances 6 months after the injury that were not present before the accident; Fogelberg et al. found persistent sleep difficulties in 44% of rehabilitation patients 1 year after TBI, and in the subsequent study by Baumann's group such sleep problems were shown to be still present 3 years after injury in two-thirds of patients.
Studies of TBI and sleep disturbance in children are few, and the findings can be inconsistent. Using retrospective parental reports for pre-injury sleep and repeated prospective post-injury assessments up to 2 years, Beebe et al. described children aged 6 to 12 years with severe TBI as being at increased risk for post-injury sleep problems compared with those with moderate TBI and a comparison group of children with only an orthopaedic injury. In contrast, a small group of 7- to 12-year-old children with a mild TBI 6 months before assessment were said to have greater sleep disturbance than children with orthopaedic injuries. In a small-scale study, Kaufman et al. found that adolescents who had suffered a minor head injury 3 years previously still complained of severe sleep disturbances and showed polysomnographic and actometry results in keeping with their complaints. Subsequently, researchers from the same group followed a larger number of adolescents who had suffered a minor head injury up to 6 years previously, and found that over a quarter still had subjective sleep disturbance.
A more recent prospective study over 2 years of children aged up to 17 years of age who had experienced TBI confirmed a higher rate of sleep disturbance (with more prolonged duration) than comparison children who had experienced an orthopaedic injury. Risk factors included pain, psychosocial factors, and mild TBI, which has been reported to be possibly more closely correlated with an increased likelihood of sleep disturbance than severe forms of TBI. The counter-intuitive recurrent theme that mild TBI particularly seems to predispose to sleep disturbance is just one of the many current imponderables in this area of enquiry.
A fundamental distinction should be made between, on the one hand, the general terms ‘sleep disturbance’ and ‘sleep problem’ and, on the other, ‘sleep disorder’, which specifies the precise type of sleep disturbance in terms of the underlying cause of a sleep problem. Although there are just three basic sleep problems (insomnia, excessive daytime sleepiness [EDS], and parasomnias), over 80 different sleep disorders are officially listed in the International Classification of Sleep Disorders, 2nd edition (ICSD-2). These distinctions may well not be made, in which case diagnostic accuracy is lost, and there is a risk that the choice of treatment will be inappropriate, as this should be matched with the type of sleep disorder.
|Sleep problem||Responsible sleep disorder|
|Excessive daytime sleepiness||Sleep apnoea|
|Periodic limb movements in sleep|
|Delayed sleep phase syndrome|
|Irregular sleep–wake rhythm|
|Insomnia||Possibly sleep disorders associated with anxiety, depression, or pain|
|Parasomnia||Periodic limb movements in sleep|
|Rapid eye movement sleep behaviour disorder|
The available literature about sleep and TBI varies in that some reports deal with sleep problems, others specify sleep disorders, and some reports deal with sleep disturbances that may have pre-dated the TBI, rather than have been caused by it.
Excessive daytime sleepiness is the sleep problem most consistently described in adults with TBI. It is defined in the ICSD-2 as ‘the inability to stay awake and alert during the main waking episodes of the day, resulting in unintended lapses into drowsiness or sleep’. It has many possible types of underlying cause: inadequate overnight sleep, disrupted (poor-quality) sleep, and disorders in which sleep requirements are pathologically increased. The terms EDS and hypersomnia are sometimes used interchangeably.
In their detailed prospective study of sleep disturbance 6 months after TBI, Baumann et al. found that 72% of patients had developed disturbed sleep following the injury. EDS and fatigue were very common, but not insomnia or sleep–wake cycle disorders. Other recent studies of EDS and adult TBI include that by Watson et al., who emphasized the persistence of sleepiness in many patients. In the case of children, Hooper et al. reported that, in a large group aged from infancy to late teens with various degrees of TBI severity, caregivers described EDS as a common problem which persisted for up to at least 10 months post-injury.
Excessive daytime sleepiness is essentially a sleep problem rather than a sleep disorder. Of the many sleep disorders which can cause EDS, sleep apnoea has been implicated in some patients with TBI: as many as 44% in a small series described by Guilleminault et al. and 23% in a larger prospective study by Castriotta et al. Periodic limb movements in sleep (reported in the latter study to affect 7% of the patient series) can, like sleep apnoea, disrupt (‘fragment’) sleep and impair its quality and restorative value, causing sleepiness and other adverse daytime effects.
In the same study by Castriotta et al., narcolepsy is mentioned in 6% of the series, and low hypocretin-1 levels (which characterize narcolepsy) were described in almost all patients in the Baumann et al. series in the first few days after TBI; however, such low levels were found in only a small minority 6 months later. Verma et al. included narcolepsy along with sleep apnoea and periodic limb movement disorder as a main cause of excessive sleepiness complaints in their series of patients with TBI of various degrees of severity. On the other hand, cases of post-traumatic narcolepsy, with compatible polysomnographic findings, have been considered to be relatively few and diverse in various ways including human leukocyte antigen type.
There is also uncertainty about the possibility that circadian sleep–wake disorders (also a cause of both EDS and insomnia) are associated with TBI. Such disorders, in the form of delayed sleep phase syndrome and irregular sleep–wake patterns, were described in 36% of patients with minor TBI, and Shekleton et al. reported a reduction in evening melatonin production in such patients, raising the possibility of a traumatic disruption of the circadian regulation of melatonin synthesis. However, it has been considered that in such patients evidence of a shift in the circadian timing of sleep (judged in terms of the timing of the onset of melatonin production) is inconclusive.
Fatigue (characterized by feelings of lack of energy, exhaustion, physical and mental tiredness, and apathy with slowness in performing tasks) has also been described as a consequence of TBI. However, despite some similarities, fatigue should not to be confused with EDS as it consists of daytime tiredness without either subjective or objective signs of excessive sleepiness. Nevertheless, it deserves attention, if only as a possible accompaniment to sleep problems.
Fatigue was described in 17% of patients in the study by Baumann et al. and elsewhere it has also been reported to be common in adult patients with TBI. For example, Kempf et al. described both fatigue and EDS persisting for up to 3 years after TBI. Findings by Ponsford et al. associated both sleep disturbance and fatigue with anxiety, depression, and pain (themselves likely to contribute to poor sleep). However, it has been considered that fatigue associated with TBI may be related to brain injury itself (irrespective of severity), rather than comorbid conditions alone such as depression, pain, or sleep problems.
A number of studies in adults with TBI have focused specifically on the problem of insomnia, in the form of difficulty falling asleep and/or remaining asleep at night. All have reported this as a common problem, although its origins are not necessarily directly neurophysiological. Particular points emphasized in these reports include the fact that insomnia can be severe and long-standing and remain untreated in many cases, and that risk factors associated with insomnia are milder degrees of injury, anxiety, depression, and pain; although largely separable entities, insomnia is closely related to fatigue and both are associated with poor quality of life; and the diagnosis and management of insomnia needs to be a prominent part of rehabilitation.[33, 34]
Information specifically on childhood insomnia and TBI is lacking, although there is extensive discussion of insomnia in children with other neurodisabilities, which has implications for TBI.
Parasomnias are unusual behaviours, experiences, or movements associated with sleep. There is little mention of them in the literature about TBI and sleep disturbance. This is surprising because, collectively, parasomnias are common in the general population (many different types are described in the ICSD-2), and some might well be part of a reaction to injury, notably parasomnias associated with post-traumatic stress disorder, especially nightmares. Reference has already been made to periodic limb movements in sleep, which are part of the sleep-related movement disorders subset of the parasomnias in the ICSD-2. In their meta-analysis of adults with TBI, Mathias and Alvaro referred to nightmares and sleepwalking, reported in 27% and 9% of patients respectively. Parasomnias were said to be a presenting complaint in 25% of the series of patients with chronic TBI described by Verma et al., who listed rapid eye movement sleep behaviour disorder and sleepwalking as topping the list, followed by nightmares, nocturnal enuresis, sleep paralysis, cataplexy, and nocturnal eating. The possibility exists that such parasomnias as these pre-dated the brain injury.
A tangential point is that sleep disturbance may be the cause rather than an effect of TBI. For example, many motor vehicle accidents are caused by the driver of the vehicle falling asleep at the wheel for various reasons, including sleep loss, sleep apnoea, or narcolepsy. Furthermore, some parasomnias (especially those of a violent or otherwise dramatic nature) can result in accidental injury. Examples include sleepwalking, sleep terrors, rapid eye movement sleep behaviour disorder, and some sleep-related epilepsies.
Sleep disturbance can be caused in several ways. This has serious implications for the extent of assessment required initially and subsequently in patients with TBI. The same applies regarding rehabilitation and later care. The most fundamental consideration is the physiological effects of the damage inflicted on the brain by the injury, but other possible contributory influences, namely comorbid conditions, medication effects, and parenting practices, also need to be considered.
Although there are indications that the traumatized brain in children reacts differently in some respects to the adult brain, with implications for management following trauma, certain basic aspects apply at any age. ‘Immediate (primary) brain injury’ results from initial linear forces when a moving object strikes the head; angular forces produced by cranial acceleration–deceleration can cause immediate widespread shearing of axons. Following primary brain injury ‘secondary brain injury’ can occur which may result from, for example, hypoxaemia, hypotension, raised intracranial pressure, haematomas, or seizures, which are considered potentially avoidable or treatable. In contrast, various cellular and biochemical events in the brain can occur soon after the injury and can continue for months afterwards causing axonal injury and neuronal cell damage and death (‘delayed brain injury’) for which preventative measures have not yet been developed.
Both primary and secondary brain injury can cause diffuse damage to the brain including the widespread complex structures involved in sleep. That said, precise correlations between specific sleep disorders and localization of brain injury, or indeed neuropathophysiological systems, are difficult to discern. An exception is the report by Baumann et al., which found that the number of hypocretin neurons in the hypothalamus is significantly reduced in patients with severe TBI, presumably because of hypothalamic damage. As loss of such neurons causes the sleepiness of narcolepsy and possibly other disorders, this observation seems to provide an explanation of the chronic sleepiness associated with TBI.
Various problems and conditions, each capable in their own right of disturbing sleep, are very likely to coexist with the basic neuropathological effects in people with TBI, whatever their age. Details of the ways in which various comorbidities can affect children's sleep are provided elsewhere.
|Traumatic brain injury comorbidity|
|Pain including headache|
|Sleep-related breathing problems|
|Mobility problems at night|
|Impaired overall intelligence|
|Attention and memory|
|Behavioural and psychiatric comorbidity|
|Anxiety states including post-traumatic stress disorder|
|Attention-deficit–hyperactivity disorder symptoms|
Of the many possible comorbidities of this type in patients with TBI, a number are capable of disturbing sleep in different ways. Examples include pain (including headache), which is reported to be a common complication of TBI even among patients with minor injuries, sleep-related breathing problems, post-traumatic epilepsy, mobility problems causing discomfort in bed, blindness (giving rise to sleep–wake cycle disorders from lack of visual input to the suprachiasmatic nucleus), and nocturnal incontinence. Other medical conditions to consider that may affect sleep include endocrine disorders such as hypothyroidism.
The effects of epilepsy on sleep in children, in general, vary with the type of seizure disorder: the more severe the epilepsy, the more serious the effects. The direct effects of epilepsy include a reduction of the total time spent asleep, and fragmentation of sleep architecture by frequent brief arousals which impair the quality and restorative value of sleep, affecting daytime function and behaviour. On the other hand, some antiepileptic drugs such as phenytoin (insomnia) or barbiturates and benzodiazepines (oversedation) can indirectly affect sleep. In their different ways, benzodiazepines and valproate may predispose to obstructive sleep apnoea (or its worsening) by causing respiratory depression or weight gain respectively. Newer antiepileptic drugs tend to have less detrimental effects on sleep, and better seizure control may well improve sleep. A further way in which epilepsy may indirectly disturb sleep is through the psychosocial consequences of suffering from a seizure disorder, including the development of anxiety or depression.
Cognitive impairment following TBI has been documented extensively in both adults and children. Generally speaking, in proportion to the severity of the TBI, overall intelligence, attention, and memory, as well as other aspects of cognitive function such as the ability to communicate effectively, can be impaired to the extent of interfering with the learning of good sleep habits, or, for example, their reacquisition by means of cognitive behaviour treatment. Adult survivors of severe childhood TBI are reported to be particularly liable to global (including cognitive) impairments in association with poor quality of life and an increased risk of mental health problems.
Changes in behaviour (including those of psychiatric proportions) that are likely to cause sleep problems are, similarly, often reported following TBI (even of mild degree) in adults.[26, 47] Rao et al. found that anxiety disorder secondary to TBI was the most significant predictor of worsening sleep disturbance in the acute injury period. Fogelberg et al. emphasized anxiety and depression (as well as pain) as common accompaniments of TBI.
A similar picture regarding children whose psychological problems follow TBI has been of interest for some time. The following social and behavioural effects of TBI were claimed by Andrews et al.: low self-esteem, loneliness, maladaptive behaviour, and aggressive/antisocial behaviour. More recently, Hooper et al. described, in a large proportion of children with TBI of various degrees of severity, reports by caregivers not only of persistent neurocognitive symptoms, headaches, and school problems, but also personality changes and low frustration intolerance. Hawley et al. described frequent behavioural and emotional problems, including personality change as well as attention and memory difficulties, in 5- to 15-year-olds at the time of their TBI. The occurrence of these problems increased with injury severity, while improvement over an average of 2 years was inversely related to severity; however, only 30% of the group received adequate follow-up after the injury. Anxiety and depression have also been emphasized as psychiatric effects of childhood TBI.
The need for adequate follow-up was also stressed by Li and Liu in view of their findings that behavioural, neurocognitive, and social difficulties could emerge from shortly to several years after injury and persist or even worsen with time. Karver et al. found that higher levels of attention-deficit–hyperactivity disorder symptoms and anxiety were more marked in young children with severe TBI than in older children. The need for prolonged treatment was emphasized in view of the persistence of such problems. Based on their findings in a similar study (including children and adolescents), the same point was made by Kenardy et al., who emphasized the enduring difficulties of children with TBI and post-traumatic stress disorder.
Various paediatric medications which might be used in children with TBI have been associated with the occurrence of sleep disturbance of one type or another. Examples of medications that may cause sleeplessness include pseudoephedrine, theophylline and stimulant drugs for attention-deficit-hyperactivity disorder. Sedative-hypnotic drugs causing excessive sleepiness include benzodiazepines and some antihistamines, major tranquillizers, and some antiepileptic drugs such as barbiturates, valproate, and carbamazepine. Seemingly less well known is that some drugs (e.g. zolpidem and some antidepressants) have been linked with the occurrence of parasomnias, especially sleepwalking, although some doubt has been expressed about the strength of this association. The introduction or increased dosage of such medications should be monitored for the possible development or worsening of sleep disturbance.
Parenting practices can profoundly influence any child's sleep patterns: lack of routine, poor limit setting, and reinforcement by paying too much attention to a child's reluctance to settle to sleep can cause or maintain sleep problems – so-called ‘behavioural insomnia of childhood’ as distinct from insomnia due to other causes. In various and changing ways, parenting factors continue to be influential throughout childhood and into adolescence regarding the origin, severity, or the maintenance of childhood sleep problems.
Parents' ability to promote their child's good sleep patterns is likely to be impaired if they are stressed, if they themselves are sleep deprived, anxious, and depressed because of their child's sleep problem, and if their knowledge about children's sleep is inadequate, as may well be the case without informed help and advice. These problems are likely to be intensified if their child has a chronic illness. Caregivers of children with TBI experience substantial distress when their child's health care needs are not met.
A survey of such children revealed that a high proportion had unmet or unrecognized health care needs during the post-injury period. This shortcoming was said to be particularly associated with abnormal family functioning. The authors of the survey emphasize that, because the health care needs of children with TBI change over time, it is important for paediatricians to monitor their recovery to ensure that they receive the service that they need. This should include correcting Parents' mistaken belief that treatment for the sleep disorders of children with a neurodevelopmental disorder is unlikely to be effective.
A thorough analysis of the cause(s) of a child's sleep problem is essential. Treatment of a sleep disturbance should not precede diagnosis of the underlying cause of which there are many possibilities, especially in children with neurodevelopmental disorders.
However, it cannot be assumed that the child's sleep problem has come to medical attention; as mentioned, parents may not seek help for even grossly disturbed sleep patterns in the mistaken belief that it is an inevitable and untreatable part of their child's condition. In addition, paediatricians may overlook sleep problems, resulting in missed opportunities for successful treatment. For that reason, as part of initial clinical assessment, it is important to screen all children with TBI and to repeat such screening periodically as sleep disturbance might arise or change with time.
In addition, it is essential to review each child's overall condition with the various other factors in mind, including possible comorbid conditions and medication effects that might be affecting sleep. Special care is needed in assessing the effects of TBI in children who were intellectually impaired before their injury to avoid mistakenly attributing TBI symptoms to their pre-existing problems.
As far as screening for sleep problems is concerned, history taking should routinely include basic enquiries about bedtime difficulties or settling to sleep, waking during the night, breathing problems while asleep, unusual behaviours, experiences, or movements at night, difficulty waking up in the morning, and being sleepy or ‘overtired’ during the day. Further details can be obtained by means of a brief screening questionnaire. Spruyt and Gozal have reviewed available paediatric sleep questionnaires. Of the various possibilities, the Children's Sleep Habits Questionnaire can be considered an appropriate choice for use in clinical practice because of the range of sleep problems that it covers, despite its relative brevity, and its versions for both school-age children and toddlers to children of preschool age. This questionnaire has been used with typically developing children and also children of different ages with neurological problems. Scales are available for somewhat closer scrutiny of EDS (and also fatigue) in children and adolescents.[69-71]
However, screening for sleep symptoms simply aims at highlighting the possibility of a sleep disorder and does not constitute a diagnosis. Adequate identification of a sleep disorder requires a comprehensive clinical enquiry, the main components of which are detailed histories, especially about the sleep problem; the child's 24-hour sleep–wake pattern, including parenting practices; developmental details; and family history and circumstances. Both physical and behavioural examination may well be appropriate, and possibly further assessment in the form of sleep diary records and objective sleep recordings such as actigraphy or polysomnography (including the Multiple Sleep Latency Test for assessing EDS, or combined with audio-visual recording in the case of possible parasomnias).
The value of an objective assessment of sleep is that the findings can be at variance with subjective reports: Ouellet and Morin found that patients post-TBI who suffered from insomnia tended to overestimate their sleep disturbance compared with polysomnographic measures, whereas Baumann et al. reported that in their patients post-TBI, excessive sleepiness complaints appeared to be less than expected from objective recordings.
Depending on the nature and complexity of the sleep disturbance, referral for assessment at a specialist paediatric service (such as the ear, nose, and throat service) or a sleep disorders clinic may be needed.
In the case that physical and behavioural or psychiatric comorbidities mentioned earlier have been overlooked, the child's overall condition may merit screening for these conditions, with a comprehensive review initially and at intervals (the Strengths and Difficulties Questionnaire is an appropriate means of screening mainly for behavioural and psychiatric problems). By the same token, possible medication effects should be monitored.
Some guidance about the treatment of particular types of sleep problems or disorders appropriate for children with TBI is available from the limited reports concerning adults.[29, 74, 75] The use (and relative merits) of both pharmacological and non-pharmacological treatments are mentioned such as stimulant drugs, including modafinil for EDS (reported to be convincingly effective for EDS but not for fatigue), sedative/hypnotic drugs, melatonin, or cognitive behaviour therapy for insomnia, continuous positive airway pressure for obstructive sleep apnoea, and dopamine agonists for periodic limb movements (management of other parasomnias is not discussed).
Clearly, controlled trials of the use of these various treatments in the TBI context for both adults and children are required. In the meantime, use can be made of the various types of treatment recommended for children's sleep disorders in general, although formal evidence for their efficacy varies. Behavioural methods, mainly for insomnia, are generally preferred to pharmacological medication. The place of melatonin in the treatment of children's insomnia still needs to be clarified because of the relatively few methodologically sound studies of its efficacy and inconsistent findings. However, some recent reports provide more convincing evidence of its usefulness, especially in children with various neurodevelopmental disorders.[81-83] Other treatments such as chronotherapy for circadian sleep–wake cycle disorders (by rescheduling the sleep period or the use of melatonin or bright light), and physical methods such as adenotonsillectomy, continuous positive airway pressure, and weight reduction for obstructive sleep apnoea are described elsewhere.
A further aspect of treatment for sleep disturbance includes the education of parents about children's sleep and the importance of the principles of sleep hygiene to encourage good sleep habits. These principles are applicable to both adults and children, including those with a neurodisability. Attention also needs to be given to the possible contributions to sleep disturbance of the comorbidities and possible medication effects mentioned earlier.
In addition to the accurate identification of sleep disorders and the correct choice of treatment, success (hopefully demonstrated to be long-lasting) will depend on Parents' preference, capabilities, and commitment, the child's willingness and ability to comply, and an adequate trial of treatment. Sustained support, explanation, and guidance for parents are an essential part of the comprehensive care of any child with a sleep disturbance. In principle, attention to these treatment issues can be expected to improve the child's condition and general circumstances, as well as parental well-being. However, it is unfortunate that, although much has been written about the rehabilitation of patients with TBI with otherwise detailed design of rehabilitation programmes, often sleep problems and their consequences and management receive little mention, despite their likely interference with participation in the rehabilitation process. Similarly, measures for assessing outcome in children with head injury generally ignore sleep. More fundamentally, as mentioned earlier, a significant proportion of patients are not followed-up at all from any point of view.
It is clinically important for the relationship between sleep and TBI to be researched further. Apart from broad generalisations, the current information (including its inconsistencies and various degrees of methodological rigour) leaves many important issues unresolved. For that reason, further well-designed systematic research is needed on the nature, origins, consequences, and management of sleep disturbance in adults and, especially, in children with TBI.
A number of writers have suggested guidance for the scope and design of further studies. Orff et al. suggested that research should focus on uncovering the specific types, causes, and severity of TBI that most often lead to sleep problems, the specific consequences of sleep disturbance such as impaired physical or cognitive recovery, and the most effective strategies for the treatment of sleep–wake abnormalities. In the light of the limitations that current information imposed on their meta-analysis, Mathias and Alvaro make a number of recommendations for future studies, largely to overcome the present varied nature of the groups on which reports have been based, and the different study designs. These apply to future studies of children as well as adults. They include (1) providing clear information about recruitment criteria such as whether patients were known to be symptomatic or whether they were recruited prospectively together with details of the criteria used for identifying sleep disturbance; (2) using larger samples with appropriate use of comparisons in order to more accurately estimate prevalence rates; (3) providing more detailed descriptions of their samples including injury severity and time since injury, as well as information about use of medication or experience of pain; and (4) establishing whether patients had any pre-existing disorder or other neurological, psychiatric, or medical disorders that could affect sleep.
A further recommendation would be to pay close attention to outcome measures such as those adopted comprehensively by the Pediatric TBI Outcomes Workgroup for characterizing the course of recovery from childhood TBI, predicting later outcome, measuring treatment effects, and comparing outcomes across studies. Taylor has also discussed research issues concerning outcome of childhood TBI, including the need for further study of the relationships between specific forms of neuropathology and outcome, environmental influences, and effects of TBI on later development such as socio-emotional functioning.
There is good reason to believe that disturbed sleep and its potential developmental consequences are often associated with childhood TBI. However, this is generally not adequately considered as part of the overall care including rehabilitation, nor has the subject been researched sufficiently.
Screening for sleep disorders and their potential origins should be performed routinely as part of initial and repeated assessment, followed by detailed diagnosis in keeping with modern practice in the sleep disorders field. This will indicate the choice of appropriate treatment, which can be expected to be to the benefit of both the children and their families.
Much more needs to be discovered about the precise sleep disorders associated with TBI and their multifactorial aetiology, taking into account the contributions of neuropathological factors and possible contributions from comorbid conditions of a physical, psychological, or pharmacological nature, as well as parenting issues.