Archives of Physical Medicine and Rehabilitation
Volume 89, Issue 5 , Pages 843-850, May 2008

Poor Sleep Quality and Changes in Objectively Recorded Sleep After Traumatic Brain Injury: A Preliminary Study

Presented in part to the Fatigue and Sleep Symposium of the International Neuropsychological Society, July 2006, Zurich, Switzerland.

  • Diane L. Parcell, DPsych

      Affiliations

    • School of Psychology, Psychiatry and Psychological Medicine, Monash University, Melbourne, Australia
    • ,
  • Jennie L. Ponsford, PhD

      Affiliations

    • School of Psychology, Psychiatry and Psychological Medicine, Monash University, Melbourne, Australia
    • Monash-Epworth Rehabilitation Research Centre, Epworth Hospital, Melbourne, Australia.
    • ,
  • Jennifer R. Redman, PhD

      Affiliations

    • School of Psychology, Psychiatry and Psychological Medicine, Monash University, Melbourne, Australia
    • ,
  • Shantha M. Rajaratnam, PhD

      Affiliations

    • School of Psychology, Psychiatry and Psychological Medicine, Monash University, Melbourne, Australia
    • Corresponding Author InformationReprint requests to Shantha M. Rajaratnam, PhD, School of Psychology, Psychiatry and Psychological Medicine, Monash University, Bldg 17, Victoria 3800, Australia

Article Outline

Abstract 

Parcell DL, Ponsford JL, Redman JR, Rajaratnam SM. Poor sleep quality and changes in objectively recorded sleep after traumatic brain injury: a preliminary study.

Objectives

To evaluate changes in sleep quality and objectively assessed sleep parameters after traumatic brain injury (TBI) and to investigate the relationship between such changes and mood state and injury characteristics.

Design

Survey and laboratory-based nocturnal polysomnography.

Setting

Sleep laboratory.

Participants

Ten community-based subjects with moderate to very severe TBI and 10 age- and sex-matched controls from the general community.

Interventions

Not applicable.

Main Outcome Measures

Pittsburgh Sleep Quality Index for self-report sleep quality, nocturnal polysomnography for objective sleep recording, and Hospital Anxiety and Depression Scales.

Results

Compared with controls, TBI patients reported significantly poorer sleep quality and higher levels of anxiety and depression. Objective sleep recording showed that TBI patients showed an increase in deep (slow wave) sleep, a reduction in rapid eye movement sleep, and more frequent nighttime awakenings. No significant relationship was observed between these changes in sleep and injury severity or time since injury. Anxiety and depression covaried with the observed changes in sleep.

Conclusions

The findings contribute to the growing body of evidence that sleep is involved in the physiologic processes underlying neural recovery. The association between anxiety and depression and the observed changes in sleep in TBI patients warrants further examination to determine whether a causative relationship exists.

Key Words: Anxiety, Brain injuries, Depression, Recovery of function, Rehabilitation, Sleep

 

SLEEP CHANGES ARE reported by the majority of survivors of traumatic brain injury (TBI).1, 2, 3 Despite this, there have been few objective studies investigating sleep changes after TBI and the small number of polysomnographic investigations reported have yielded mixed results.4, 5, 6, 7 In view of the deleterious effects of poor sleep quality and reduced sleep quantity on physical and mental health,8 the development of appropriate treatments for sleep complaints in TBI patients is warranted. Investigation of the nature and etiology of such sleep complaints is a first step in developing appropriate treatments.

There are many possible causes of sleep change after TBI, and identification of such causes has proved difficult even in relatively large samples.9 TBI may damage structures involved in the generation and maintenance of sleep, including the reticular activating system, hypothalamus, basal forebrain and pontine nuclei and their connections,10 or the circadian system, notably the suprachiasmatic nuclei of the anterior hypothalamus and its connections. Damage to hypothalamic structures may result in physiologic change, with a secondary impact on the biochemical regulation of sleep. Growth hormone deficiency and altered cortisol levels are known to have specific effects on sleep architecture, as do some other forms of endocrine dysfunction.11 Hypopituitarism, leading to altered levels of growth hormone and pituitary hormone production, has been shown to occur in approximately 40% of people after TBI.12 Compared with a sample of healthy controls, TBI patients in the acute phase showed reduced levels of hypocretin-1, a hypothalamic neuropeptide involved in sleep-wake regulation.9, 13 Finally, sleep changes may stem from concomitant changes in mood, lifestyle, and medication use after TBI. There is significantly increased incidence of anxiety and depression in the TBI population.14, 15 Anxiety and depression are associated with alterations in sleep architecture,16, 17 although a recent study reported no association between depression ratings and the presence or severity of post-traumatic sleep-wake disturbances.9 Changes in employment status and other lifestyle factors may contribute to low or anxious mood and/or alter the person's daily activity schedule, such as the need to wake for work. Some medications taken after TBI, such as antidepressants and anticonvulsants, may also affect sleep architecture.16, 18

Polysomnography is regarded as the criterion standard for objective recording of sleep. Relatively few studies have investigated sleep in TBI patients using polysomnography. The most frequently reported change to sleep architecture after TBI is an alteration in the proportion of rapid eye movement (REM) sleep. Ron et al6 found an overall decrease in the amount of REM sleep as well as fewer eye movements in REM in TBI patients (n=9) compared with controls. Conversely, Frieboes et al5 reported an increase in REM sleep in the second half of the night, as well as a tendency toward REM sleep disinhibition in the postacute stages of TBI. Busek and Faber4 reported both increases and decreases in the proportion of REM sleep, and concluded that REM sleep is the sleep stage most vulnerable to TBI. A recent study7 reported no significant difference in the proportion of REM sleep between TBI patients and healthy controls.

Other changes in sleep parameters have been reported in TBI patients. Electroencephalographic sleep spindle activity in nonrapid eye movement (NREM) sleep was reported to increase in TBI patients.6 Although the functional significance of sleep spindle activity is presently unclear, recent evidence suggests a role in memory consolidation.19 Prigatano et al20 reported reduced sleep efficiency and increased nighttime awakenings. Frieboes et al5 reported reduced stage 1 and stage 2 sleep after TBI, while Ouellet and Morin7 reported a higher proportion of stage 1 sleep and no difference in the other sleep stages.

In addition to changes in sleep after TBI, hypersomnia is frequently reported.9, 21, 22 Masel et al23 found that excessive hypersomnia, measured by the multiple sleep latency test, affected 47% of a sample of 71 TBI patients. Polysomnography revealed that the hypersomnolent subgroup had significantly less stage 2 sleep than the nonhypersomnolent subgroup. Although it is presently unclear whether the hypersomnolence reported by TBI patients is a consequence of nocturnal sleep changes, its high prevalence in TBI patients has significant implications for their quality of life.

In summary, studies to date have shown vulnerability of REM sleep and decreased sleep efficiency to be the most consistent changes to sleep architecture in the postacute phase of TBI. However, these studies have used varied methodologies and results might have been affected by differences in participant injury severity and location, living situation (institution or independent), and psychologic state. Furthermore, some previous studies have been limited by the lack of an appropriate control group, lack of an adaptation night in the laboratory prior to sleep evaluation, and failure to screen for pre-existing sleep disorders.4, 23

The primary aim of this preliminary study was to examine sleep efficiency and sleep architecture in TBI patients using polysomnography, and to compare these data with those of an age- and sex-matched control group. The secondary aim was to examine self-reported sleep quality in TBI patients compared with controls.

Back to Article Outline

Methods 

Participants 

Participants were persons with TBI recruited from the head injury rehabilitation program at Epworth Hospital (Melbourne, Australia) and age- and sex-matched controls recruited from the general community. We matched participants for age because changes in sleep such as a reduction in slow-wave sleep (SWS) can occur with advancing age,24, 25 and for sex, in light of increased reporting of insomnia in women across all age groups as well as sex-specific vulnerabilities and differences in sleep architecture in the presence of mood disturbance.26, 27 The first 10 willing participants were recruited, with no preference for patients with sleep complaints. The sample size is consistent with that used in previous studies of TBI patients.5, 6, 7, 20 The study protocol was approved by the Monash University Standing Committee on Ethics in Research Involving Humans (study approval no. 2000/534) and the Epworth Hospital Human Research and Ethics Committee, Melbourne, Australia (study approval no. 18501).

We included TBI participants if they met the following criteria: aged between 16 and 65; sufficient English-language ability to complete the questionnaires, normal range of body mass index (18.5−25.0kg/m2); no transmeridian travel across more than 1 time zone in the preceding 3 months and no shift work in the previous 12 months; not taking benzodiazepines or other sleeping medications; and no previous head injury, neurologic disorder, or major psychiatric disorder. The presence of sleep disorders can elevate the risk of TBI through accidents.28 Therefore, to determine whether participants showed evidence of sleep disorders prior to injury, a significant other, living with the participant, completed a questionnaire screening for classic symptoms of periodic limb movement disorder (PLMD), sleep-disordered breathing, and insomnia. Participants showing evidence of these symptoms were excluded. Control participants met the same criteria and had never sustained a head injury. Some TBI participants had suffered additional physical injuries; however, these injuries were largely recovered at the time of the current investigation.

Procedure 

The polysomnographic study took place over 2 nights within 1 week. Polysomnography is the standard system of sleep recording, and consists of electroencephalogram, electro-oculogram, and electromyogram.29 Saliva samples were collected during these 2 nights and later assayed for melatonin. Salivary melatonin data have been published elsewhere.30

Adaptation night 

Participants were studied in pairs, such that a TBI participant and their matched control attended the sleep laboratory (Monash University, Melbourne, Australia) at the same time. Participants remained seated in dim light conditions (<10 lux) from 5:30 pm to 12:30 am for the purposes of assessing salivary melatonin levels.30 At approximately 7:00 pm on the adaptation night, participants were fitted with electroencephalography electrodes. Bipolar electroencephalographic activity was derived from C3, C4, and OZ according to the International 10-20 System of Electrode Placement.31, 32 Electroencephalography leads were referenced to the left mastoid (C3/OZ:A1) and the right mastoid (C4:A2). When the measured electroencephalographic position corresponded with an area of scar tissue or was placed directly over a metal plate so that recording was not clear, the electrode was displaced by the least amount necessary to clear the lesion. The homologous electrode was relocated as well to correspond to the location on the altered side in accordance with standard recommendations.31 Eye movements were recorded by electro-oculography, with electrodes applied to the outer canthi of each eye; the right electro-oculography electrode was placed slightly above the cantomeatal plane and the left electro-oculography electrode was placed slightly below.29 Both leads were referenced to nasion (Nz). Muscle activity was measured with electromyography and external electrodes were placed at the mentalis.

We obtained electroencephalographic, electro-oculographic, and electromyographic recordings through standard gold cap Grass-type electrodes. All signals were amplified and digitalized online (8-bit analog-to-digital converter; storage sampling rate: 250Hz for electroencephalographic recordings; 128Hz for electro-oculographic recordings; 1000Hz for electromyographic recordings) and were digitally low-pass filtered at 35Hz and high-pass filtered at 0.3Hz, on standard polysomnography hardware and software.a Raw signals were recorded directly on an IBM hard drive and later transferred to compact disk for long-term storage and analysis. The scheduled sleep opportunity commenced at approximately 12:30 am and ended at approximately 8:00 am. Prior to the sleep opportunity, participants completed a series of questionnaires, described below.

Sleep evaluation night 

The sleep evaluation night was conducted within 1 week of the adaptation night. Participants were instructed to arrive at the sleep laboratory 2 hours before their habitual bedtime, calculated from a self-report sleep diary completed beforehand. Electrodes were fitted as per the adaptation night. Participants were permitted to sit and read or watch television until the sleep opportunity commenced. Polysomnographic activity was recorded during the sleep opportunity and ended when the participant awoke. Sleep records from the sleep evaluation night were manually scored in 30-second epochs as wake, REM sleep, or NREM sleep (stages 1 to 4) according to standard criteria by an experienced and blinded rater.29 Nighttime arousals were scored according to criteria defined by the Sleep Disorders Task Force of the American Sleep Disorders Association as transient interruptions characterized by changes in electroencephalographic frequency (and in REM sleep also brief changes in electromyographic amplitude) preceded by 10 continuous seconds or more of any stage of sleep.33 Awakening was defined as any epoch with greater than 50% alpha or beta low-voltage electroencephalographic activity.

Measures 

We obtained demographic details and medication from participants at recruitment and we verified injury details and medication from medical records. A standard sleep-wake diary was completed by each participant for 7 days prior to the sleep recording night. The Epworth Sleepiness Scale (ESS)34 was used to evaluate sleepiness during wake time, and participants indicated whether their sleep had changed since their head injury (TBI group) or in the past 3 months (control group) using a general sleep questionnaire that we developed. Participants were also asked to identify any factors that restricted their sleep and to report sleep quality using the same questionnaire. The 2 sleep quality rating scales were “How do you feel on waking in the morning?”(“refreshed” to “need more sleep”); and “How rested do you feel in the morning?” (“very rested” to “not rested at all”). Information from both scales was summed to give a sleep quality score between 2 and 10, a higher score reflecting less satisfaction with sleep.

The Hospital Anxiety and Depression Scale35 measures symptoms and severity of anxiety and depression in general medical populations and has been used with TBI patients.35, 36, 37 The scale provides separate scores for anxiety and depression. For both scales, scores of 0 through 7 represent normal levels, 8 through 10 mild mood disturbance, 11 through 14 moderate, and 15 through 21 represents severe anxiety or depression.

The Pittsburgh Sleep Quality Index (PSQI)38 was used as an index of sleep quality and has been extensively validated, including in postacute patients with TBI.39

Data Analysis 

Because participants in the TBI and control groups were matched for age and sex, all t tests were conducted pairwise for greater sensitivity. Results are reported for 2-tailed analyses at the .05 significance level. No adjustments for multiple comparisons have been made in light of the limited sample size. Unless otherwise stated, data are reported as mean and standard error of mean (SEM).

Back to Article Outline

Results 

Participant Characteristics 

Participant characteristics are summarized in table 1. The groups were matched for sex and there was no significant difference in age. Based on post-traumatic amnesia (PTA) duration, 40% (n=4) sustained moderate injuries (PTA 1−7d), 40% (n=4) had severe injuries (PTA 8−28d), and 20% (n=2) had very severe injuries (PTA >4wk). PTA duration was not noted in medical records for 1 participant. Sixty percent (n=6) were injured in a motor vehicle collision, 20% (n=2) in motorbike or bicycle collision, and 20% (n=2) were pedestrians. The most common type and location of injury according to computed tomography (CT) scans obtained from medical records were frontal contusions present in 6 of 10 participants, followed by parietal contusions (4/10) and base of skull fractures (4/10). Two participants had occipital contusions and none showed diffuse axonal injury or damage to medial or limbic structures. It should be noted that many of these CT scans were taken on admission to the acute hospital and it is known that scans at this time or even a later time may not detect the full extent of injury.40

Table 1. Participant Characteristics
CharacteristicsTBIControlStatisticP
n1010
Sex (male)66
Age (y)38.80±4.34(23–61)37.80±4.38(23–63)t=1.17.27
Employed (includes part-time) (%)4070χ2=1.82†.37
Glasgow Coma Scale score10.9±1.01(6–14)
PTA duration (d)16.44±4.29(1–36)⁎
Days postinjury516±124.04(74–1194)
Medication (%)4020χ2=0.95†.63
Medication description3 x pain‡ 1 x anticonvulsant1 x pain‡ 1 x bronchodilatorχ2=1.05†1.00
Caffeinated drinks (daily number)0.29±0.13(0.00–1.14)0.79±0.43(0.00–4.43)t=1.25.24
Alcoholic drinks (daily)0.34±0.24(0.00–2.43)0.84±0.18(0.00–1.86)t=1.88.09

NOTE. Values mean ± SEM (range) or as indicated.

Abbreviation: PTA, post-traumatic amnesia.

⁎n=9, no duration of PTA recorded for 1 participant.

†Chi-square analyses showed that 2 cells had an expected count of less than 5. An exact significance test was selected for Pearson chi-square.

‡Paracetamol.

The TBI group was less likely to be employed and more likely to be taking medication although these differences were not significant. There was no significant difference in caffeine or alcohol consumption between the groups, although the TBI participants tended to have a lower intake of alcohol, possibly due to their clinician's recommendation to cease drinking for 12 months postinjury. Further analyses of possible factors that may restrict sleep, such as work schedules and other daily activities and also pain, showed no significant differences between groups.

Sleep Parameters 

TBI group reported significantly poorer sleep quality (PSQI) than controls (t8=6.61, P<.01; partial η2=.85) (fig 1C). PSQI scores greater than 5, which can be used clinically to define poor sleep quality,37 were observed significantly more frequently in TBI patients (89%) than in controls (30%) (χ12=6.74, P<.01). Preferred sleep-wake time (Morningness-Eveningness Questionnaire) and ESS did not differ between the groups (P>.05). Self-reported sleep data in TBI patients are described in detail elsewhere.3

  • View full-size image.
  • Fig 1. 

    Changes in sleep in TBI patients. Values are mean ± SEM for TBI patients and controls. Significant differences between groups are shown (*P<.05, †P<.01). (A, B) Data derived from polysomnography; and (C) data derived from a self-report questionnaire. (A) The percentage of time spent in each sleep stage—stage 1, stage 2, SWS, and REM. (B) Total sleep time. (C) Sleep quality ratings (PSQI), with a horizontal arrow indicating the cutoff above which people are defined as having poor sleep quality.37

Polysomnographic data revealed that the observed number of awakenings was significantly greater in the TBI group compared with the control group (t9=2.70, P=.03; partial η2=.45) (table 2). The total number of arousals as well as the mean number of arousals per hour did not differ between groups.

Table 2. Polysomnographic Sleep Parameters
ParametersTBIControltP
Sleep onset latency (h)0.26±0.05(0.02–0.52)0.24±0.06(0.17–0.58)0.34.74
Total sleep time (h)6.74±0.42(5.13–9.49)6.73±0.22(5.48–7.50)0.02.98
Sleep efficiency (%)81.89±2.97(62.95–93.66)84.82±3.65(62.27–94.69)0.74.48
Awakenings (n)15.20±1.69(6–22)9.50±1.72(2–18)2.70.03⁎
Arousals (/h)13.60±1.65(6.60–26.40)13.00±1.42(6.80–22.60)0.29.78
Arousals (total)95.10±18.49(39–251)87.90±10.41(39–161)0.36.73

NOTE. Values mean ± SEM (range) or as indicated.

⁎P<.05.

The percentage of time spent in each stage of sleep was calculated (fig 1A). While there was no difference in total sleep time between groups (fig 1B), a substantial increase in SWS was observed in the TBI group (26.62%) compared with the control group (20.18%) (t9=4.21, P=.002; partial η2=.66). The TBI group also showed a significantly lower percentage of REM sleep (21%) compared with the control group (26%) (t9=2.92, P=.02; partial η2=.49). Other sleep stages did not vary significantly between groups.

Association Between Sleep Changes, Anxiety, and Depression 

The TBI group had significantly higher anxiety scores (mean ± SEM, 8.90±1.41; range, 2−14) than the control group (mean, 3.40±0.58; range, 1−7) (t8=3.67, P<.01), and significantly higher levels of depression (mean, 7.60±1.36; range, 1−15) compared with controls (mean, 1.70±0.56; range, 0−5) (t8=5.36, P<.01). Within the TBI group 3 participants showed mild anxiety (scores 8−10) and 4 showed moderate anxiety (scores 11−14). Five TBI participants exhibited mild depression (scores 8−10), 2 showed moderate depression (scores 11−14), and only one had a severe depression rating (score 15−21). No control participants showed clinically significant anxiety or depression.

In light of these differences, and because of the established association between anxiety, depression, and sleep, sleep parameters were reanalyzed by analysis of covariance with anxiety and depression as separate covariates, to examine the possible association between the observed changes in sleep parameters in TBI patients and anxiety and depression (table 3). After adjusting for anxiety and depression, no significant difference in REM sleep was observed between the groups (see table 3), indicating that these psychologic variables covaried significantly with the observed differences in REM. For SWS, a trend toward a difference between groups persisted after adjusting for anxiety (P=.08) and depression (P=.05), suggesting that the observed difference in SWS was at least in part modulated by anxiety and depression. The difference in PSQI between TBI patients and controls persisted after adjusting for anxiety and depression.

Table 3. Sleep Quality Ratings and Polysomnographic Sleep Parameters After Adjustment for Anxiety and Depression Levels
ParametersPreadjustmentAnxiety⁎Depression⁎
Sleep quality (PSQI)
TBI11.33±1.5310.13±1.2710.80±1.39
Control3.90±0.624.99±1.194.38±1.30
F1,16=6.96, P=.02†F1,16=8.83, P=.01†
Awakenings (n)
TBI15.20±1.6915.63±2.0414.12±2.06
Control9.50±3.169.07±2.0410.58±2.06
F1,17=4.09, P=.06F1,17=1.13, P=.30
Arousals
TBI87.90±10.4192.12±17.9078.50±18.21
Control95.10±18.4990.88±17.90103.50±18.21
F1,17=0.02, P=.97F1,17=0.66, P=.43
Sleep efficiency (%)
TBI81.89±2.9782.38±3.9984.27±3.99
Control84.82±3.6584.38±3.9982.44±3.99
F1,17=1.05, P=.75F1,17=0.08, P=.78
Stage 1 sleep (%)
TBI3.93±0.653.15±0.772.77±0.61
Control3.16±0.643.97±0.774.35±0.61
F1,17=0.44, P=.52F1,17=2.56, P=.13
Stage 2 sleep (%)
TBI48.41±2.7348.70±2.797.01±2.81
Control50.26±1.8249.98±2.7951.67±2.81
F1,17=0.08, P=.78F1,17=1.05, P=.32
SWS (%)
TBI26.62±1.4725.92±1.6726.33±1.75
Control20.18±1.3620.88±1.6720.47±1.75
F1,17=3.60, P=.08F1,17=4.27, P=.05
REM sleep (%)
TBI21.04±2.4721.42±2.5523.90±2.33
Control26.37±1.7325.99±2.5523.52±2.33
F1,17=1.27, P=.28F1,17=0.01, P=.92

NOTE. Values mean ± SEM, with main effect of group.

⁎Mean and SEM adjusted.

†P<.05.

Association Between Time Since Injury, Injury Severity, and Sleep Parameters 

Correlation analyses were conducted to explore the relationships between injury severity and time postinjury with those sleep variables that differed significantly in the TBI group. For the TBI group, injury severity (PTA days) and time since injury were positively associated with sleep efficiency although these correlations did not reach statistical significance (PTA days r=.60, P=.09; time postinjury r=.62, P=.06). That is, more severe injuries and a longer time since injury tended to be associated with better sleep efficiency. Injury severity and time since injury showed positive relationships with PSQI scores, although again these relationships were not significant (PTA days r=.54, P=.17; time postinjury r=.60, P=.09). There was no significant relationship between injury severity and SWS or REM sleep (P>.05). Similarly there was no significant relationship between time postinjury and SWS or REM sleep (P>.05).

Back to Article Outline

Discussion 

TBI patients reported significantly poorer sleep quality than the control group. Polysomnography revealed that the TBI group had a higher percentage of SWS, a reduction in REM sleep, and more frequent nocturnal awakenings. This study has important clinical implications, because reduced sleep quality and increased nocturnal awakenings are reported to be associated with neurocognitive functioning at least in mild TBI patients.41 In view of the significant safety and health consequences associated with sleep disruption,8 results of the present study indicate that appropriate treatment of sleep complaints in TBI patients should be undertaken.

Our findings broadly support the previous study by Ouellet and Morin,7 which used TBI patients with comparable characteristics (ie, age group, TBI severity, time since injury) and age- and sex-matched controls, in that we also found self-report evidence of sleep disturbance in TBI patients. However, in contrast to this previous study, we found significant differences in REM sleep and SWS between TBI patients and controls. The divergent findings may at least in part be explained by the requirement in the previous study for patients to present with an insomnia syndrome, whereas in the present study there was no preference for patients with sleep complaints.

The observed change in self-reported sleep quality is consistent with previous studies that report that 50% of TBI survivors report sleep disturbance.1, 2 In the present study, 89% of the TBI patients reported poor sleep quality compared with 30% of controls. While previous studies have reported relationships between changes in sleep quality and injury severity39, 41 we did not observe such associations in our sample. Our results also indicate that the observed difference in self-reported sleep quality between TBI patients and controls cannot be explained entirely by differences in anxiety and depression levels.

Previous studies of subjects with TBI have indicated that REM sleep was the most vulnerable sleep stage after TBI.4 The present study showed a significant reduction in REM sleep. One lifestyle factor known to influence REM sleep is alcohol consumption, which may suppress REM sleep.39, 42 However, in the present study the control group tended to have a higher level of alcohol intake than the TBI group so the direction of change was opposite to that expected. Indeed the difference in alcohol intake may have masked even greater differences in REM sleep.

The displacement of REM may have been secondary to the increased proportion of SWS. However, generally it is thought unlikely that REM sleep would be the sleep stage displaced by the SWS increase, because previous studies of SWS recovery have shown the proportion of REM sleep to remain stable and stage 2 NREM sleep to be most prone to displacement.43 The functional significance of the observed difference in REM sleep therefore remains unclear.

Previous studies examining polysomnographic sleep have also reported nonsignificant increases in SWS in subjects with TBI.5, 20 The observed increase in SWS in TBI patients may be explained by a number of possible mechanisms.

Increased SWS after TBI may be the result of changes to the homeostatic mechanisms of sleep.44, 45 Such mechanisms are thought to involve diffuse cerebral projections and the accumulation during wakefulness of somnogens such as adenosine, cytokine, and oleamide among others.46 These somnogens in turn affect the activity of the thalamus, reticular activating system (RAS), and basal forebrain, to enable the transition from wake to sleep by inhibiting the RAS activation of the cortex.44, 46 Diffuse injury could disrupt the balance of this complex homeostatic sleep system. For example, the production of some somnogens may increase after TBI due to alterations in their synthesis pathways. In turn, increased synthesis of somnogens could result in increased sleep drive and hence an increase in SWS.47 Alternatively, physical damage to the structures involved in sleep, such as the brainstem, thalamus, hypothalamus, and corticothalamic pathways, may alter their function and disrupt the balance of NREM and REM sleep, giving rise to more deep sleep. The mesopontine junction is thought to control the regular cycling of NREM and REM sleep once sleep is initiated.46 Damage to this region could therefore also contribute to an increase in NREM sleep, or to changes in the cycling between sleep stages.

Electroencephalographic slow wave activity, and by implication visually scored SWS, may be increased after TBI due to increased neural plasticity or reorganization.48, 49, 50 The experience of wakefulness is inextricably linked to learning about one's environment, which is thought to result in changes in the strength of synaptic connections and an increase in synaptic weight.50 These authors postulate that slow wave activity serves to return the synaptic weight of neurons to a baseline level, a process known as synaptic downscaling. The findings of increased slow wave activity in brain regions where metabolic and cognitive demands are high lend support to this postulation.51, 52, 53, 54 In other words, differing levels of metabolic or cognitive activity in specific brain regions may lead to differing demands for synaptic plasticity in an experience-dependent manner.48 Kolb49 presents evidence that synaptic plasticity occurs after injury and in response to therapy. People with TBI are likely to be in a continuous state of learning and adaptation in the months and even years after injury, leading to neuronal reorganization, with an associated increase in synaptic potentiation. According to the theory proposed by Tononi and Cirelli,50 during recovery persons with TBI may have a higher need for synaptic downscaling at night to correct neuronal weight and therefore experience greater slow wave activity in sleep.

Levels of anxiety and depression were found (at least in part) to covary with the observed changes in SWS in the present sample. Anxiety was found to be associated with increased SWS. This finding is in contrast with the previously reported relationship between anxiety and SWS, where anxiety and apprehension were found to be associated with a decrease in SWS.55 Kajimura et al56 reported that healthy participants with high anxiety tended to show an increase in SWS on the first night of polysomnographic recording compared with a low anxiety group; however, the pattern was reversed on subsequent recording nights. Therefore, on the basis of these previous studies, the opposite relationship would be predicted to that observed in the present study between anxiety levels and SWS. Depression is associated with changes to REM sleep such as shortened latency to REM, increased REM duration and increased REM density.57 There is scant evidence of changes to SWS in depression; however, again in contrast with the present findings, some reporting of decreased SWS secondary to an increase in REM has been noted.58 Therefore, as with anxiety, the observed associations between depression and SWS and REM sleep may represent independent outcomes of TBI rather than causal mechanisms.

Consistent with previous studies the TBI group showed an increased number of nighttime awakenings.20, 23 Increased nighttime awakenings are a common symptom of hyperarousal, anxiety, and lighter sleep.17 Arousals and subsequent awakenings from sleep may be triggered by environmental stimuli (eg, noise, light, temperature), endogenous stimuli (eg, pain, hunger), or stimulant medications.59, 60 Anxiety and depression levels were found to be associated with night time awakenings in the present sample. Physical discomfort may also substantially contribute to the more frequent awakenings in the TBI group.

Study Limitations 

It should be noted that although we excluded participants who were overweight and obese (a major risk factor for sleep-disordered breathing61) and who, based on the report of a significant other person living with the participant, showed history of sleep disorder (PLMD, sleep-disordered breathing, insomnia), participants were not screened for sleep disorders by polysomnography. Therefore, the possibility remains that the observed disruption of sleep in TBI patients may be accounted for, at least in part, by sleep disorders.

Back to Article Outline

Conclusions 

In this preliminary study, we report that TBI patients show poorer subjective sleep quality, decreased REM sleep, increased SWS, and increased nighttime awakenings. The TBI group also reported higher levels of anxiety and depression. Future studies may use imaging technologies to examine the association between structural changes and sleep characteristics in TBI patients to elucidate the mechanisms responsible for sleep changes and also the anatomic regulation of sleep-wakefulness. Changes to sleep after TBI are of interest not only from a clinical perspective, but also because of the implications that may be drawn from such studies about the relationship between sleep and neural recovery.

Supplier

Back to Article Outline

Acknowledgment 

We thank Jo Phipps Nelson, BBehSci, for her assistance in conducting this research.

Back to Article Outline

References 

  1. Beetar JT, Guilmette TJ, Sparadeo FR. Sleep and pain complaints in symptomatic traumatic brain injury and neurologic populations. Arch Phys Med Rehabil. 1996;77:1298–1302
  2. Cohen M, Oksenberg A, Snir D, Stern MJ, Groswasser Z. Temporally related changes of sleep complaints in traumatic brain injured patients. J Neurol Neurosurg Psychiatry. 1992;55:313–315
  3. Parcell DL, Ponsford JL, Rajaratnam SM, Redman JR. Self-reported changes to nighttime sleep after traumatic brain injury. Arch Phys Med Rehabil. 2006;87:278–285
  4. Busek P, Faber J. The influence of traumatic brain lesion on sleep architecture. Sb Lek. 2000;101:233–239
  5. Frieboes RM, Muller U, Murck H, von Cramon DY, Holsboer F, Steiger A. Nocturnal hormone secretion and the sleep EEG in patients several months after traumatic brain injury. J Neuropsychiatry Clin Neurosci. 1999;11:354–360
  6. Ron S, Algom D, Hary D, Cohen M. Time-related changes in the distribution of sleep stages in brain injured patients. Electroencephalogr Clin Neurophysiol. 1980;48:432–441
  7. Ouellet MC, Morin CM. Subjective and objective measures of insomnia in the context of traumatic brain injury: a preliminary study. Sleep Med. 2006;7:486–497
  8. Committee on Sleep Medicine and Research, Board on Health Sciences Policy, Institute of Medicine, Institute of Medicine. Sleep disorders and sleep deprivation: an unmet public health problem. Washington (DC): Natl Acad Pr; 2006;
  9. Baumann CR, Werth E, Stocker R, Ludwig S, Bassetti CL. Sleep-wake disturbances 6 months after traumatic brain injury: a prospective study. Brain. 2007;130(Pt 7):1873–1883
  10. Coenen AM. Neuronal phenomena associated with vigilance and consciousness: from cellular mechanisms to electroencephalographic patterns. Conscious Cogn. 1998;7:42–53
  11. Van Cauter E, Leproult R, Plat L. Age-related changes in slow wave sleep and REM sleep and relationship with growth hormone and cortisol levels in healthy men. JAMA. 2000;284:861–868
  12. Kelly DF, Gonzalo IT, Cohan P, Berman N, Swerdloff R, Wang C. Hypopituitarism following traumatic brain injury and aneurysmal subarachnoid hemorrhage: a preliminary report. J Neurosurg. 2000;93:743–752
  13. Baumann CR, Stocker R, Imhof HG, et al. Hypocretin-1 (orexin A) deficiency in acute traumatic brain injury. Neurology. 2005;65:147–149
  14. Hibbard MR, Uysal S, Kepler K, Bogdany J, Silver J. Axis I psychopathology in individuals with traumatic brain injury. J Head Trauma Rehabil. 1998;13(4):24–39
  15. Kreutzer JS, Seel RT, Gourley E. The prevalence and symptom rates of depression after traumatic brain injury: a comprehensive examination. Brain Inj. 2001;15:563–576
  16. Benca RM. Mood disorders. In:  Kryger MH,  Roth T,  Dement WC editor. Principles and practice of sleep medicine. 3rd ed.. Philadelphia: WB Saunders; 2000;p. 1140–1158
  17. Uhde TW. Anxiety disorders. In:  Kryger MH,  Roth T,  Dement WC editor. Principles and practice of sleep medicine. 3rd ed.. Philadelphia: WB Saunders; 2000;p. 1123–1139
  18. Legros B, Bazil CW. Effects of antiepileptic drugs on sleep architecture: a pilot study. Sleep Med. 2003;4:51–55
  19. Schabus M, Gruber G, Parapatics S, et al. Sleep spindles and their significance for declarative memory consolidation. Sleep. 2004;27:1479–1485
  20. Prigatano GP, Stahl ML, Orr WC, Zeiner HK. Sleep and dreaming disturbances in closed head injury patients. J Neurol Neurosurg Psychiatry. 1982;45:78–80
  21. Castriotta RJ, Lai JM. Sleep disorders associated with traumatic brain injury. Arch Phys Med Rehabil. 2001;82:1403–1406
  22. Guilleminault C, Yuen KM, Gulevich MG, Karadeniz D, Leger D, Philip P. Hypersomnia after head-neck trauma: a medicolegal dilemma. Neurology. 2000;54:653–659
  23. Masel BE, Scheibel RS, Kimbark T, Kuna ST. Excessive daytime sleepiness in adults with brain injuries. Arch Phys Med Rehabil. 2001;82:1526–1532
  24. Bliwise BL. Normal aging. In:  Kryger MH,  Roth T,  Dement WC editor. Principles and practice of sleep medicine. 3rd ed.. Philadelphia: WB Saunders; 2000;p. 26–42
  25. Feinberg I. Changes in sleep cycle patterns with age. J Psychiatr Res. 1974;10:283–306
  26. Armitage R, Hoffmann RF. Sleep EEG, depression and gender. Sleep Med Rev. 2001;5:237–246
  27. Voderholzer U, Al-Shajlawi A, Weske G, Feige B, Riemann D. Are there gender differences in objective and subjective sleep measures? (A study of insomniacs and healthy controls). Depress Anxiety. 2003;17:162–172
  28. Barbe , Pericas J, Munoz A, Findley L, Anto JM, Agusti AG. Automobile accidents in patients with sleep apnea syndrome (An epidemiological and mechanistic study). Am J Respir Crit Care Med. 1998;158:18–22
  29. Rechtschaffen A, Kales A. A manual of standardized terminology, techniques, and scoring system for sleep states of human subjects. Washington (DC): Superintendent of Documents, US Government Printing Office; 1968;
  30. Steele DL, Rajaratnam SM, Redman JR, Ponsford JL. The effect of traumatic brain injury on the timing of sleep. Chronobiol Int. 2005;22:89–105
  31. Harner PF, Sannit T. A review of the international ten-twenty system of electrode placement. Warwick: Grass Instrument; 1974;
  32. Jasper HH. The ten-twenty electrode system of the international federation. EEG Clin Neurophysiol. 1958;10:371–375
  33. EEG arousals: scoring rules and examples: a preliminary report from the Sleep Disorders Atlas Task Force of the American Sleep Disorders Association. Sleep. 1992;15:173–184
  34. Johns MW. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;14:540–545
  35. Zigmond AS, Snaith RP. The hospital anxiety and depression scale. Acta Psychiatr Scand. 1983;67:361–370
  36. Hawley CA. Reported problems and their resolution following mild, moderate and severe traumatic brain injury amongst children and adolescents in the UK. Brain Inj. 2003;17:105–129
  37. Herrmann C. International experiences with the Hospital Anxiety and Depression Scale—a review of validation data and clinical results. J Psychosom Res. 1997;42:17–41
  38. Buysse DJ, Reynolds CF, Monk TH, Berman SR, Kupfer DJ. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res. 1989;28:193–213
  39. Fichtenberg NL, Millis SR, Mann NR, Zafonte RD, Millard AE. Factors associated with insomnia among post-acute traumatic brain injury survivors. Brain Inj. 2000;14:659–667
  40. Bigler ED. Distinguished Neuropsychologist Award Lecture 1999 (The lesion(s) in traumatic brain injury: implications for clinical neuropsychology). Arch Clin Neuropsychol. 2001;16:95–131
  41. Mahmood O, Rapport LJ, Hanks RA, Fichtenberg NL. Neuropsychological performance and sleep disturbance following traumatic brain injury. J Head Trauma Rehabil. 2004;19:378–390
  42. Roehrs T, Roth T. Sleep, sleepiness, sleep disorders and alcohol use and abuse. Sleep Med Rev. 2001;5:287–297
  43. Ferrara M, De Gennaro L, Bertini M. Selective slow-wave sleep (SWS) deprivation and SWS rebound: do we need a fixed SWS amount per night?. Sleep Res Online. 1999;2(1):15–19
  44. Evans BM. What does brain damage tell us about the mechanisms of sleep?. J R Soc Med. 2002;95:591–597
  45. Graham DI. Pathophysiological aspects of injury and mechanisms of recovery. In:  Rosenthal M,  Griffith ER,  Kreutzer JS,  Pentland B editor. Rehabilitation of the adult and child with traumatic brain injury. 3rd ed.. Philadelphia: FA Davis; 1999;p. 19–41
  46. Pace-Schott EF, Hobson JA. The neurobiology of sleep: genetics, cellular physiology and subcortical networks. Nat Rev Neurosci. 2002;3:591–605
  47. Porkka-Heiskanen T, Strecker RE, Thakkar M, Bjorkum AA, Greene RW, McCarley RW. Adenosine: a mediator of the sleep-inducing effects of prolonged wakefulness. Science. 1997;276:1265–1268
  48. Benington JH, Frank MG. Cellular and molecular connections between sleep and synaptic plasticity. Prog Neurobiol. 2003;69:71–101
  49. Kolb B. Mechanisms of cortical plasticity after neuronal injury. In:  Ponsford JL editors. Cognitive and behavioural rehabilitation: from neurobiology to clinical practice. New York: Guildford Pr; 2004;p. 30–58
  50. Tononi G, Cirelli C. Sleep and synaptic homeostasis: a hypothesis. Brain Res Bull. 2003;62:143–150
  51. Achermann P, Finelli LA, Borbely AA. Unihemispheric enhancement of delta power in human frontal sleep EEG by prolonged wakefulness. Brain Res. 2001;913:220–223
  52. Cajochen C, Foy R, Dijk DJ. Frontal predominance of a relative increase in sleep delta and theta EEG activity after sleep loss in humans. Sleep Res Online. 1999;2(3):65–69
  53. Huber R, Deboer T, Tobler I. Topography of EEG dynamics after sleep deprivation in mice. J Neurophysiol. 2000;84:1888–1893
  54. Huber R, Ghilardi MF, Massimini M, Tononi G. Local sleep and learning. Nature. 2004;430:78–81
  55. Kecklund G, Akerstedt T. Apprehension of the subsequent working day is associated with a low amount of slow wave sleep. Biol Psychol. 2004;66:169–176
  56. Kajimura N, Kato M, Sekimoto M, et al. A polysomnographic study of sleep patterns in normal humans with low- or high-anxiety personality traits. Psychiatry Clin Neurosci. 1998;52:317–320
  57. Lauer CJ, Riemann D, Wiegand M, Berger M. From early to late adulthood (Changes in EEG sleep of depressed patients and healthy volunteers). Biol Psychiatry. 1991;29:979–993
  58. Riemann D, Berger M, Voderholzer U. Sleep and depression—results from psychobiological studies: an overview. Biol Psychol. 2001;57:67–103
  59. Bourdet C, Goldenberg F. Insomnia in anxiety: sleep EEG changes. J Psychosom Res. 1994;38(Suppl 1):93–104
  60. Broughton RJ. NREM arousal parasomnias. In:  Kryger MH,  Roth T,  Dement WC editor. Principles and practice of sleep medicine. 3rd ed.. Philadelphia: WB Saunders; 2000;p. 693–706
  61. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med. 1993;328:1230–1235
  • a S-Series sleep monitoring system; Compumedics Ltd, 30-40 Flockhart St, Abbotsford 3067, Victoria, Australia.

 Supported in part by the National Health and Medical Research Council (project no. 334002).

 No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit upon the authors or upon any organization with which the authors are associated.

PII: S0003-9993(08)00105-6

doi:10.1016/j.apmr.2007.09.057

Archives of Physical Medicine and Rehabilitation
Volume 89, Issue 5 , Pages 843-850, May 2008

Article Tools

* Image Usage & Resolution