Polysomnographic and quantitative EEG analysis of subjects with long-term insomnia complaints associated with mild traumatic brain injury
Introduction
Mild traumatic brain injury (MTBI) affects approximately one in two hundred people every year (Bazarian et al., 2005) and is associated with various cognitive, affective and physical difficulties, which can persist well after the injury, and are often subtle and go unrecognised (Gronwall, 1989, Eide and Tysnes, 1992, Segalowitz and Brown, 1991, Segalowitz and Lawson, 1995, Wang et al., 2006). While there is some disagreement in terminology, mild head injury is defined as a loss of consciousness of 20 min or less with a Glasgow Coma Scale rating of 13–15, and a period of hospitalisation not exceeding 48 h (Levin et al., 1987). MTBI is defined herein as a blow to the head in the mild head injury range of severity, resulting in symptoms associated with post-concussion syndrome (PCS) at the time of injury. These symptoms include dizziness, fatigue, headaches, impairments in attention, poor concentration, sensitivity to noise, memory problems, depression, sleep difficulties, and others (Bigler, 1990).
Difficulty sleeping is the fourth most common symptom reported by MTBI patients, occurring in 43.9% of cases (Levin et al., 1987), and is still a complaint years after the injury in a subset of patients (Eide and Tysnes, 1992, Bigler, 1990, Masson et al., 1996, Beetar et al., 1996). In general, individuals with MTBI report a variety of difficulties with initiating and maintaining sleep (Parsons and Ver Beek, 1982, Perlis et al., 1997, Mahmood et al., 2004, Ouellet et al., 2006, Parcell et al., 2006, Stulemeijer et al., 2006, Baumann et al., 2007). A number of polysomnographic (PSG) studies have examined the relationship between traumatic brain injury (TBI) or MTBI and sleep – with mixed results. These include more stage 2 (Lenard and Pennigstorff, 1970), less slow wave sleep (Lenard and Pennigstorff, 1970), less stage 1 (Prigatano et al., 1982), greater stage 1 (Ouellet and Morin, 2006), more awakenings (Prigatano et al., 1982, George and Landau-Ferey, 1986, Kaufman et al., 2001), and less REM (Ron et al., 1980, George and Landau-Ferey, 1986). Another study found no differences in overall sleep architecture, but found lower power in theta and alpha frequency bands during the first cycle of non-rapid eye movement (NREM) sleep, and lower delta, theta, and alpha power during the second cycle of NREM sleep (Parsons et al., 1997). These variable results may be partly attributed to the different age of subjects in each study, the length of time after the injury, and the severity and type of injury. Furthermore, sleep difficulties after MTBI may be due to variety of causes such as neurological damage to the sleep/wake system, learned associations to stressful or anxiety-inducing events associated with the injury that counteract sleep (psychophysiological insomnia), the result of chronic pain, or secondary to psychiatric symptoms such as anxiety or depression, which often develop after MTBI in both humans (Busch and Alpern, 1998) and animal models (Milman et al., 2005).
While some of the above studies included subjects with TBI in the mild range, none have specifically examined long-term polysomnographic measures subsequent to MTBI. In addition, the majority of studies have focused on overall sleep architecture and have not analysed micro aspects of sleep, which can discriminate between various types of sleep disorders and controls. For example, Lamarche and Ogilvie found that it was possible to differentiate psychophysiological insomniacs, psychiatric insomniacs, and controls using power spectral analysis of the sleep onset period (SOP) (Lamarche and Ogilvie, 1997). During wakefulness, the psychophysiological insomniacs had more beta (indicating higher physiological arousal) and less alpha power. In addition, alpha power did not show as dramatic a drop during the descent into sleep. Psychophysiological insomniacs also had less delta power during the latter stages of the SOP compared to psychiatric insomniacs and controls. In addition to examining mean power, the variability in power across the SOP has been shown to differentiate between normal and insomniac populations, with insomniacs having greater variability over the first 5 min of stage two sleep (McCartney et al., 1998). Such analyses have also been used to distinguish normal controls from subjects with narcolepsy (Alloway et al., 1999), and depression (Armitage et al., 1994), and were therefore included in this study. The FFT results were also used to determine whether sleep difficulties were more characteristic of psychophysiological or psychiatric insomnia, while standard polysomnographic measures were used to determine if MTBI subjects exhibit characteristics of idiopathic insomnia. This might provide clues to the aetiology of the sleep disturbance as well as suggest treatment strategies.
Section snippets
Overview of procedure
Subjects were recruited from a first year undergraduate psychology class, posters placed around the university campus, and an article in the local newspaper. After an initial telephone interview, subjects, who met the inclusion criteria (see below) for either the MTBI or control group, were scheduled for a 2.5–3 h interview and questionnaire session in the laboratory. A sleep log was given to subjects to take home for a two-week period. During overnight sessions, subjects were scheduled for
Questionnaire data
Data on depression, anxiety, adaptive functioning, and current and pre-injury sleep (MTBI group only) were collected via the PAI, BAFQ, and multiple self-report measures of sleep and sleep difficulty. The MTBI group had higher total depression scores on the PAI (t(16) = 3.07, p = 0.007), as well as higher scores on the physiological (t(16) = 1.32, p < 0.001) and cognitive (W = 16, p = 0.030) subscales, while there was no significant difference between groups on the affective depression subscale (Table 1).
Questionnaire data
Consistent with previous studies, MTBI patients with persistent sleep complaints had impairments on a variety of psychometric measures, including, depression, anxiety, attention, and memory, which serves to validate the selection of subjects. These scores were only slightly elevated, and a number of patients were within the normal range. Patients also reported difficulties with initiating and maintaining sleep, which is not surprising since subjects in the MTBI group were selected based on the
Acknowledgements
B.R.W. was supported by a Natural Sciences and Engineering Research Council (NSERC) postgraduate scholarship (PGS-A) and S.E.L. was supported by an NSERC Undergraduate Student Research Award. The research was supported by a grant from the Rick Hansen Institute/Ontario Neurotrauma Foundation as well as an NSERC grant to R.D.O.
References (45)
- et al.
Sleep and pain complaints in symptomatic traumatic brain injury and neurologic populations
Arch Phys Med Rehabil
(1996) - et al.
The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research
Psychiatry Res
(1989) - et al.
Long-term sleep disturbances in adolescents after minor head injury
Pediatr Neurol
(2001) - et al.
Subjective and objective measures of insomnia in the context of traumatic brain injury: a preliminary study
Sleep Med
(2006) - et al.
Self-reported changes to nighttime sleep after traumatic brain injury
Arch Phys Med Rehabil
(2006) - et al.
Time-related changes in the distribution of sleep stages in brain injured patients
Electroencephalogr Clin Neurophysiol
(1980) - et al.
Examination of postconcussion-like symptoms in healthy university students: relationships to subjective and objective neuropsychological function performance
Arch Clin Neuropsychol
(2006) - et al.
EEG spectral analysis of the sleep-onset period in narcoleptics and normal sleepers
Sleep
(1999) - American Sleep Disorders Association, 1990. International Classification of Sleep Disorders. American Academy of Sleep...
- et al.
Sleep disturbances in depression and the effects of antidepressants
Int Rev Psychiatry
(2005)