Review
Sleep-wake disturbances after traumatic brain injury Marie-Christine Ouellet, Simon Beaulieu-Bonneau, Charles M Morin Lancet Neurol 2015; 14: 746–57 Centre Interdisciplinaire de Recherche en Réadaptation et Intégration Sociale (CIRRIS), Québec, QC, Canada (M-C Ouellet PhD, S Beaulieu-Bonneau PhD); École de Psychologie, Université Laval, Québec, QC, Canada (M-C Ouellet, S Beaulieu-Bonneau, C M Morin PhD); and Centre de Recherche de l’Institut Universitaire en Santé Mentale de Québec, Québec, QC, Canada (S Beaulieu-Bonneau, C M Morin) Correspondence to: Dr Marie-Christine Ouellet, Centre Interdisciplinaire de Recherche en Réadaptation et Intégration Sociale (CIRRIS), Institut de Réadaptation en Déficience Physique de Québec (IRDPQ), Québec, QC, G1M 2S8, Canada marie-christine.ouellet@psy. ulaval.ca
Sleep-wake disturbances are extremely common after a traumatic brain injury (TBI). The most common disturbances are insomnia (difficulties falling or staying asleep), increased sleep need, and excessive daytime sleepiness that can be due to the TBI or other sleep disorders associated with TBI, such as sleep-related breathing disorder or posttraumatic hypersomnia. Sleep-wake disturbances can have a major effect on functional outcomes and on the recovery process after TBI. These negative effects can exacerbate other common sequelae of TBI—such as fatigue, pain, cognitive impairments, and psychological disorders (eg, depression and anxiety). Sleep-wake disturbances associated with TBI warrant treatment. Although evidence specific to patients with TBI is still scarce, cognitive-behavioural therapy and medication could prove helpful to alleviate sleep-wake disturbances in patients with a TBI.
Introduction About 1·7 million individuals sustain a traumatic brain injury (TBI) of varying severity every year in the USA (500–600 per 100 000), and 16·3% of those injuries lead to people being admitted to hospital.1 TBI can lead to persisting sequelae in the physical, cognitive, emotional, and social spheres that substantially affect not only the injured people, but also their family and community. Of the persisting sequelae of TBI, sleep-wake disturbances are very common and present as difficulties falling or staying asleep, excessive daytime sleepiness, an increased need for sleep, and changes in sleep-wake rhythms. The causes of sleep-wake disturbances after TBI include a complex interplay between pathophysiological processes (ie, at the structural, neuroelectrical, or neurochemical levels), psychological factors (eg, sleep-related habits or TBI-related psychopathology), environmental factors (eg, noises, light, or pain), and social factors (eg, social roles related to work or family).2,3 Sleep-wake disturbances are present in the whole spectrum of TBI severity4 and can arise early after the injury, during treatment in hospital, or during inpatient rehabilitation,5–7 and can often persist for years after the injury.8,9 Sleep-wake disturbances can exacerbate other outcomes of TBI, particularly cognitive deficits, pain, fatigue, and mood disturbances.10–13 In a meta-analysis14 of 21 studies, about 50% (95% CI 49–51) of people with TBI were found to have sleep disturbances. In this Review, we assess the most recent scientific literature and address the nature, prevalence, and pathophysiology of post-TBI sleep-wake disturbances and their effect on recovery and interaction with other sequelae of TBI, in addition to assessing treatment options. Findings included in this Review should be interpreted cautiously in view of the limitations of the existing literature; great variability exists in the definitions for sleep disorders or difficulties, assessment methods, sample characteristics (such as severity of TBI and time since injury), and control of potential confounders (such as comorbidities [eg, pain or depression] or medications).
Nature and prevalence The sleep-wake disturbances that are mostly studied in association with TBI include insomnia, excessive daytime 746
sleepiness and disorders of hypersomnolence (eg, narcolepsy and post-traumatic hypersomnia), increased sleep need, sleep-related breathing disorders (eg, sleep apnoea), and circadian rhythm sleep-wake disorders (panel 1 [glossary of terms]; table 1). These disturbances are covered in this Review because they occur frequently and evidence is available about their association with TBI. Other sleep disorders (eg, parasomnias and sleep-related movement disorders) can also be present, but much less evidence is available about their association with TBI. Insomnia is reported in 30–60% of individuals with TBI, either presenting as symptoms (without fulfilling all diagnostic criteria) or as an insomnia disorder.3,4,14 Some studies point to more insomnia complaints in people with mild TBI than those with severe TBI, possibly because of increased awareness of their own symptoms in this subgroup.11,17,18 In a study11 of 452 patients with mild-tosevere TBI at an average of 8 years post injury, 29% fulfilled the diagnostic criteria for an insomnia disorder, with an additional 21% of patients presenting symptoms without meeting all diagnostic criteria. Of the patients with an insomnia disorder, insomnia was chronic in most cases and was not treated in 60% of cases.11 Most studies reporting insomnia after TBI use self-reported questionnaires of sleep disturbances or sleep diaries and the diagnosis is based on a subjective complaint, whereas polysomnography is regarded as the gold-standard to diagnose most other sleep disorders (eg, sleep apnoea). Excessive daytime sleepiness is a complaint reported by 14–57% of individuals with TBI,8–10,19–22 and is among the core symptoms of many primary sleep disorders (eg, sleep apnoea and post-traumatic hypersomnia). Subjective reports of sleepiness do not always correlate well with quantitative measures from daytime polysomnography. Data from a well controlled prospective study20 showed that patients with TBI underestimate their level of daytime sleepiness compared with people without TBI. In this sample,20 24 (57%) of 42 patients with TBI were defined as having objective daytime sleepiness (as assessed with daytime polysomnography) compared with eight (19%) of 42 healthy controls. These data suggest that studies relying solely on subjective reports of excessive daytime sleepiness should be interpreted with caution. Imbach and colleagues20 used 2-week continuous actigraphy completed www.thelancet.com/neurology Vol 14 July 2015
Review
Panel 1: Glossary of terms Actigraph Watch-like device measuring motor activity that yields an estimate of sleep and wake periods.
REM density Frequency of eye movements per unit of time during REM sleep.
Increased sleep need (pleiosomnia) Excessive sleep (abnormally increased need for sleep [eg, increase of 2 h per 24 h compared with pre-injury habits]) need per 24 h.
REM sleep latency Time from sleep onset to the onset of the first REM sleep episode.
Excessive daytime sleepiness Difficulty maintaining alertness or wakefulness, and unintentionally falling asleep.
Non-restorative sleep Subjective assessment of sleep as poor, light, or unrestful.
Microarousal Partial awakening from sleep, measured by polysomnography but often undetected by the individual sleeping. Non-rapid eye movement (NREM) sleep One of two main types of sleep, which is divided into three stages: N1, N2, and N3; characterised by minimal mental activity, a movable body, but rarely by dreaming. Polysomnography Assessment of sleep by electroencephalographical (brain activity), electro-oculographical (eye movements), and electromyographical (muscle movements) measures; provides comprehensive information about the continuity and architecture of sleep, and is used as a diagnostic method for several sleep disorders; it can be used during the night (eg, to assess sleep apnoea) or during the day (eg, Multiple Sleep Latency Test to measure excessive daytime sleepiness).15 Rapid eye movement (REM) sleep One of the two main types of sleep; characterised by rapid EEG, muscle atonia, dreaming, and rapid eye movements occurring in episodic bursts and when dreaming.
6 months after injury, which showed that individuals with TBI slept an average 1·2 h more per 24 h than did healthy control participants. This finding shows another very important sequelae of TBI: increased sleep need (also referred to as pleiosomnia), which can present either concomitantly or independently from excessive daytime sleepiness.19,20 Persistent fatigue, another common outcome of TBI, can generally be confused with excessive daytime sleepiness. However, fatigue is a subjective state that is mostly assessed with self-reported questionnaires,23,24 whereas sleepiness is a measurable physiological state characterised by objective signs of reduced alertness or drowsiness (such as yawning and eyelids drooping).25 Sleep disruption and daytime sleepiness can cause or exacerbate post-TBI fatigue, yet fatigue can present independently from sleep-wake disturbances.24,26 All of these disorders have major implications for daytime functioning ability and need to be carefully assessed and addressed in all patients with TBI. Panel 2 lists the assessment methods used for the most common sleep disturbances after TBI. www.thelancet.com/neurology Vol 14 July 2015
Sleep onset latency Duration of time from going to bed to sleep onset. Sleep efficiency Ratio between total time spent sleeping and time spent in bed. Sleep onset REM period REM sleep occurring earlier than normal on daytime or nighttime polysomnography (within 15 min of sleep onset rather than the typical 90 min). Sleep spindle Burst of rapid EEG activity progressively increasing and then progressively decreasing in amplitude. Sleep spindles are a hallmark feature of NREM stage N2 sleep. Slow-wave sleep NREM stage N3 sleep; characterised by slow EEG activity, also called deep sleep. Spectral analysis Method used to split sleep EEG activity into its constituting frequency bands; typical bandwidths used are: delta (15 Hz).
Narcolepsy, which is characterised predominantly by frequent and recurrent sleep attacks, is another potential cause of excessive daytime sleepiness after TBI.9,28 However, evidence that this disorder might develop after TBI is sparse. If excessive sleepiness is present almost daily, and both sleep apnoea and narcolepsy are ruled out, then excessive daytime sleepiness might be classified as post-traumatic hypersomnia, provided patients had no premorbid sleep-wake issues.16 A diagnosis of posttraumatic hypersomnia could possibly apply to 10–30% of patients with TBI.10,19,29 Sleep-related breathing disorders, for which daytime sleepiness is one of the core symptoms, are more often reported after TBI than in the general population. Precise prevalence estimates are difficult to obtain because most studies investigating these disorders have used clinical samples of patients who were referred for polysomnographic assessment after excessive sleepiness. For example, in a study of 60 patients who had poor sleep in the first 2 years after injury, the incidence of abnormal apnoea or hypopnea was reported in 30% of patients. 747
Review
Prevalence
Clinical features and symptoms
Insomnia
29%
Dissatisfaction with sleep quality or quantity accompanied by subjective difficulty falling asleep or staying asleep, or waking up too early in the morning; substantial distress or daytime functional impairment is associated with the night-time sleep disturbances
Sleep apnoea
25%
Obstructive sleep apnoea can be complete (apnoea) or a partial (hypopnoea) obstruction of the upper airway when sleeping; central sleep apnoea is a temporary loss of ventilatory effort. Episodes of altered respiration are often associated with snoring and followed by arousals to resume breathing, resulting in non-restorative sleep and excessive daytime sleepiness
Narcolepsy
4%
A disorder of hypersomnolence, characterised by irrepressible need to sleep with recurrent involuntary sleep attacks. Symptoms include cataplexy (sudden muscle weakness triggered by strong emotions), sleep paralysis (temporary inability to move or talk right before falling asleep or upon awakening) and hypnagogic or hypnopompic hallucinations (vivid dreamlike experiences occurring right before falling asleep or upon awakening from sleep)
Post-traumatic hypersomnia
28%
Hypersomnia due to TBI, characterised by excessive daytime sleepiness, increased sleep need, or both; other primary sleep disorders causing sleepiness must be ruled out
Circadian rhythm sleep-wake disorders
Unknown
Disruption of the sleep-wake rhythm causing a disparity with the conventional 24 h sleep and wake schedule; melatonin secretion and body temperature rhythms might also be disrupted; associated with insomnia at night and daytime sleepiness; most common subtypes include greatly delayed bedtime and rising time (delayed sleep phase disorder), greatly advanced bedtime and rising time (advanced sleep phase disorder), and high variability in bedtime and arising time (irregular sleep-wake rhythm)
Prevalence estimates are based on the meta-analysis by Mathias and Alvaro.14 These estimates should be considered in light of the different definitions of sleep-wake disorders used in the scientific literature. The International Classification of Sleep Disorder third edition has more information about the definitions and diagnostic criteria of sleep disorders.16 TBI=traumatic brain injury.
Table 1: Clinical features and symptoms of sleep disorders frequently noted in patients after traumatic brain injury
Circadian rhythm sleep-wake disorders have also been documented after TBI,30–32 and these were mainly delayed sleep-wake phase disorders. Some patients with TBI can develop very irregular sleep-wake schedules and present with severe insomnia at night and excessive daytime sleepiness. Data about the type and number of injuries that affect sleep patterns after TBI are scarce. Imbach and colleagues20 showed that increased sleep need was associated with increased severity of TBI and with the presence of an intracranial haemorrhage. Preliminary data from 116 individuals with mild-to-moderate TBI sustained in a military context suggests that blunt injuries are associated with obstructive sleep apnoea (odds ratio 3·09, 95% CI 1·02–9·38, p=0·047).33 In another military sample, Bryan34 noted that one (6%) of 18 individuals without TBI reported having insomnia, whereas 11 (20%) of 54 patients with a first TBI and 39 (50%) of 78 patients who had multiple TBIs reported having this disorder (p
Sleep-wake disturbances after traumatic brain injury Marie-Christine Ouellet, Simon Beaulieu-Bonneau, Charles M Morin Lancet Neurol 2015; 14: 746–57 Centre Interdisciplinaire de Recherche en Réadaptation et Intégration Sociale (CIRRIS), Québec, QC, Canada (M-C Ouellet PhD, S Beaulieu-Bonneau PhD); École de Psychologie, Université Laval, Québec, QC, Canada (M-C Ouellet, S Beaulieu-Bonneau, C M Morin PhD); and Centre de Recherche de l’Institut Universitaire en Santé Mentale de Québec, Québec, QC, Canada (S Beaulieu-Bonneau, C M Morin) Correspondence to: Dr Marie-Christine Ouellet, Centre Interdisciplinaire de Recherche en Réadaptation et Intégration Sociale (CIRRIS), Institut de Réadaptation en Déficience Physique de Québec (IRDPQ), Québec, QC, G1M 2S8, Canada marie-christine.ouellet@psy. ulaval.ca
Sleep-wake disturbances are extremely common after a traumatic brain injury (TBI). The most common disturbances are insomnia (difficulties falling or staying asleep), increased sleep need, and excessive daytime sleepiness that can be due to the TBI or other sleep disorders associated with TBI, such as sleep-related breathing disorder or posttraumatic hypersomnia. Sleep-wake disturbances can have a major effect on functional outcomes and on the recovery process after TBI. These negative effects can exacerbate other common sequelae of TBI—such as fatigue, pain, cognitive impairments, and psychological disorders (eg, depression and anxiety). Sleep-wake disturbances associated with TBI warrant treatment. Although evidence specific to patients with TBI is still scarce, cognitive-behavioural therapy and medication could prove helpful to alleviate sleep-wake disturbances in patients with a TBI.
Introduction About 1·7 million individuals sustain a traumatic brain injury (TBI) of varying severity every year in the USA (500–600 per 100 000), and 16·3% of those injuries lead to people being admitted to hospital.1 TBI can lead to persisting sequelae in the physical, cognitive, emotional, and social spheres that substantially affect not only the injured people, but also their family and community. Of the persisting sequelae of TBI, sleep-wake disturbances are very common and present as difficulties falling or staying asleep, excessive daytime sleepiness, an increased need for sleep, and changes in sleep-wake rhythms. The causes of sleep-wake disturbances after TBI include a complex interplay between pathophysiological processes (ie, at the structural, neuroelectrical, or neurochemical levels), psychological factors (eg, sleep-related habits or TBI-related psychopathology), environmental factors (eg, noises, light, or pain), and social factors (eg, social roles related to work or family).2,3 Sleep-wake disturbances are present in the whole spectrum of TBI severity4 and can arise early after the injury, during treatment in hospital, or during inpatient rehabilitation,5–7 and can often persist for years after the injury.8,9 Sleep-wake disturbances can exacerbate other outcomes of TBI, particularly cognitive deficits, pain, fatigue, and mood disturbances.10–13 In a meta-analysis14 of 21 studies, about 50% (95% CI 49–51) of people with TBI were found to have sleep disturbances. In this Review, we assess the most recent scientific literature and address the nature, prevalence, and pathophysiology of post-TBI sleep-wake disturbances and their effect on recovery and interaction with other sequelae of TBI, in addition to assessing treatment options. Findings included in this Review should be interpreted cautiously in view of the limitations of the existing literature; great variability exists in the definitions for sleep disorders or difficulties, assessment methods, sample characteristics (such as severity of TBI and time since injury), and control of potential confounders (such as comorbidities [eg, pain or depression] or medications).
Nature and prevalence The sleep-wake disturbances that are mostly studied in association with TBI include insomnia, excessive daytime 746
sleepiness and disorders of hypersomnolence (eg, narcolepsy and post-traumatic hypersomnia), increased sleep need, sleep-related breathing disorders (eg, sleep apnoea), and circadian rhythm sleep-wake disorders (panel 1 [glossary of terms]; table 1). These disturbances are covered in this Review because they occur frequently and evidence is available about their association with TBI. Other sleep disorders (eg, parasomnias and sleep-related movement disorders) can also be present, but much less evidence is available about their association with TBI. Insomnia is reported in 30–60% of individuals with TBI, either presenting as symptoms (without fulfilling all diagnostic criteria) or as an insomnia disorder.3,4,14 Some studies point to more insomnia complaints in people with mild TBI than those with severe TBI, possibly because of increased awareness of their own symptoms in this subgroup.11,17,18 In a study11 of 452 patients with mild-tosevere TBI at an average of 8 years post injury, 29% fulfilled the diagnostic criteria for an insomnia disorder, with an additional 21% of patients presenting symptoms without meeting all diagnostic criteria. Of the patients with an insomnia disorder, insomnia was chronic in most cases and was not treated in 60% of cases.11 Most studies reporting insomnia after TBI use self-reported questionnaires of sleep disturbances or sleep diaries and the diagnosis is based on a subjective complaint, whereas polysomnography is regarded as the gold-standard to diagnose most other sleep disorders (eg, sleep apnoea). Excessive daytime sleepiness is a complaint reported by 14–57% of individuals with TBI,8–10,19–22 and is among the core symptoms of many primary sleep disorders (eg, sleep apnoea and post-traumatic hypersomnia). Subjective reports of sleepiness do not always correlate well with quantitative measures from daytime polysomnography. Data from a well controlled prospective study20 showed that patients with TBI underestimate their level of daytime sleepiness compared with people without TBI. In this sample,20 24 (57%) of 42 patients with TBI were defined as having objective daytime sleepiness (as assessed with daytime polysomnography) compared with eight (19%) of 42 healthy controls. These data suggest that studies relying solely on subjective reports of excessive daytime sleepiness should be interpreted with caution. Imbach and colleagues20 used 2-week continuous actigraphy completed www.thelancet.com/neurology Vol 14 July 2015
Review
Panel 1: Glossary of terms Actigraph Watch-like device measuring motor activity that yields an estimate of sleep and wake periods.
REM density Frequency of eye movements per unit of time during REM sleep.
Increased sleep need (pleiosomnia) Excessive sleep (abnormally increased need for sleep [eg, increase of 2 h per 24 h compared with pre-injury habits]) need per 24 h.
REM sleep latency Time from sleep onset to the onset of the first REM sleep episode.
Excessive daytime sleepiness Difficulty maintaining alertness or wakefulness, and unintentionally falling asleep.
Non-restorative sleep Subjective assessment of sleep as poor, light, or unrestful.
Microarousal Partial awakening from sleep, measured by polysomnography but often undetected by the individual sleeping. Non-rapid eye movement (NREM) sleep One of two main types of sleep, which is divided into three stages: N1, N2, and N3; characterised by minimal mental activity, a movable body, but rarely by dreaming. Polysomnography Assessment of sleep by electroencephalographical (brain activity), electro-oculographical (eye movements), and electromyographical (muscle movements) measures; provides comprehensive information about the continuity and architecture of sleep, and is used as a diagnostic method for several sleep disorders; it can be used during the night (eg, to assess sleep apnoea) or during the day (eg, Multiple Sleep Latency Test to measure excessive daytime sleepiness).15 Rapid eye movement (REM) sleep One of the two main types of sleep; characterised by rapid EEG, muscle atonia, dreaming, and rapid eye movements occurring in episodic bursts and when dreaming.
6 months after injury, which showed that individuals with TBI slept an average 1·2 h more per 24 h than did healthy control participants. This finding shows another very important sequelae of TBI: increased sleep need (also referred to as pleiosomnia), which can present either concomitantly or independently from excessive daytime sleepiness.19,20 Persistent fatigue, another common outcome of TBI, can generally be confused with excessive daytime sleepiness. However, fatigue is a subjective state that is mostly assessed with self-reported questionnaires,23,24 whereas sleepiness is a measurable physiological state characterised by objective signs of reduced alertness or drowsiness (such as yawning and eyelids drooping).25 Sleep disruption and daytime sleepiness can cause or exacerbate post-TBI fatigue, yet fatigue can present independently from sleep-wake disturbances.24,26 All of these disorders have major implications for daytime functioning ability and need to be carefully assessed and addressed in all patients with TBI. Panel 2 lists the assessment methods used for the most common sleep disturbances after TBI. www.thelancet.com/neurology Vol 14 July 2015
Sleep onset latency Duration of time from going to bed to sleep onset. Sleep efficiency Ratio between total time spent sleeping and time spent in bed. Sleep onset REM period REM sleep occurring earlier than normal on daytime or nighttime polysomnography (within 15 min of sleep onset rather than the typical 90 min). Sleep spindle Burst of rapid EEG activity progressively increasing and then progressively decreasing in amplitude. Sleep spindles are a hallmark feature of NREM stage N2 sleep. Slow-wave sleep NREM stage N3 sleep; characterised by slow EEG activity, also called deep sleep. Spectral analysis Method used to split sleep EEG activity into its constituting frequency bands; typical bandwidths used are: delta (15 Hz).
Narcolepsy, which is characterised predominantly by frequent and recurrent sleep attacks, is another potential cause of excessive daytime sleepiness after TBI.9,28 However, evidence that this disorder might develop after TBI is sparse. If excessive sleepiness is present almost daily, and both sleep apnoea and narcolepsy are ruled out, then excessive daytime sleepiness might be classified as post-traumatic hypersomnia, provided patients had no premorbid sleep-wake issues.16 A diagnosis of posttraumatic hypersomnia could possibly apply to 10–30% of patients with TBI.10,19,29 Sleep-related breathing disorders, for which daytime sleepiness is one of the core symptoms, are more often reported after TBI than in the general population. Precise prevalence estimates are difficult to obtain because most studies investigating these disorders have used clinical samples of patients who were referred for polysomnographic assessment after excessive sleepiness. For example, in a study of 60 patients who had poor sleep in the first 2 years after injury, the incidence of abnormal apnoea or hypopnea was reported in 30% of patients. 747
Review
Prevalence
Clinical features and symptoms
Insomnia
29%
Dissatisfaction with sleep quality or quantity accompanied by subjective difficulty falling asleep or staying asleep, or waking up too early in the morning; substantial distress or daytime functional impairment is associated with the night-time sleep disturbances
Sleep apnoea
25%
Obstructive sleep apnoea can be complete (apnoea) or a partial (hypopnoea) obstruction of the upper airway when sleeping; central sleep apnoea is a temporary loss of ventilatory effort. Episodes of altered respiration are often associated with snoring and followed by arousals to resume breathing, resulting in non-restorative sleep and excessive daytime sleepiness
Narcolepsy
4%
A disorder of hypersomnolence, characterised by irrepressible need to sleep with recurrent involuntary sleep attacks. Symptoms include cataplexy (sudden muscle weakness triggered by strong emotions), sleep paralysis (temporary inability to move or talk right before falling asleep or upon awakening) and hypnagogic or hypnopompic hallucinations (vivid dreamlike experiences occurring right before falling asleep or upon awakening from sleep)
Post-traumatic hypersomnia
28%
Hypersomnia due to TBI, characterised by excessive daytime sleepiness, increased sleep need, or both; other primary sleep disorders causing sleepiness must be ruled out
Circadian rhythm sleep-wake disorders
Unknown
Disruption of the sleep-wake rhythm causing a disparity with the conventional 24 h sleep and wake schedule; melatonin secretion and body temperature rhythms might also be disrupted; associated with insomnia at night and daytime sleepiness; most common subtypes include greatly delayed bedtime and rising time (delayed sleep phase disorder), greatly advanced bedtime and rising time (advanced sleep phase disorder), and high variability in bedtime and arising time (irregular sleep-wake rhythm)
Prevalence estimates are based on the meta-analysis by Mathias and Alvaro.14 These estimates should be considered in light of the different definitions of sleep-wake disorders used in the scientific literature. The International Classification of Sleep Disorder third edition has more information about the definitions and diagnostic criteria of sleep disorders.16 TBI=traumatic brain injury.
Table 1: Clinical features and symptoms of sleep disorders frequently noted in patients after traumatic brain injury
Circadian rhythm sleep-wake disorders have also been documented after TBI,30–32 and these were mainly delayed sleep-wake phase disorders. Some patients with TBI can develop very irregular sleep-wake schedules and present with severe insomnia at night and excessive daytime sleepiness. Data about the type and number of injuries that affect sleep patterns after TBI are scarce. Imbach and colleagues20 showed that increased sleep need was associated with increased severity of TBI and with the presence of an intracranial haemorrhage. Preliminary data from 116 individuals with mild-to-moderate TBI sustained in a military context suggests that blunt injuries are associated with obstructive sleep apnoea (odds ratio 3·09, 95% CI 1·02–9·38, p=0·047).33 In another military sample, Bryan34 noted that one (6%) of 18 individuals without TBI reported having insomnia, whereas 11 (20%) of 54 patients with a first TBI and 39 (50%) of 78 patients who had multiple TBIs reported having this disorder (p
