The hyperarousal model of Primary Insomnia from a
neurobiological perspective.
Name: Martine Groefsema Student number: 5622735
University of Amsterdam MSc in Brain and Cognitive Science
Cognitive Neuroscience
Supervisor: dr. W. F. Hofman Co-assessor: dr L. M. Thalamini
2
Abstract
Despite the growing literature on insomnia, little is known about the underlying pathophysiology causing the disorder. Previous literature has viewed primary insomnia mainly as a psychological problem, but more recently studies began to find more evidence for physiological differences in patients with primary insomnia. The aim of this review is to discuss the concept of hyperarousal from the higher system to the molecular processes, and discuss its relevance in the pathology of primary insomnia. Electrophysiological studies demonstrate increased levels of cortical arousal in primary insomnia during different stages of sleep. Studies on regional brain metabolism are limited and are in the need for further research. I conclude that there is an imbalance between arousing and sleep-inducing brain activity in primary insomnia, resulting in hyperarousal. However, whether this is the main cause of primary insomnia can not be concluded. Further research on genetic vulnerability is needed.
3
Content
Introduction………. 4
Chapter 1: Different types and characteristics of insomnia……….6
Chapter 2: Normal sleep processes……….8
2.1 Neurophysiology of normal sleep……….9
2.2 Neurobiology of normal sleep………..10
Chapter 3: Hyperarousal model of insomnia………15
3.1 Neurophysiology of primary insomnia………15
3.2 Neurobiology of primary insomnia……….20
3.2.2 Neurotransmitters……….22 Chapter 4: Heritability………23 Chapter 5: Treatment………..25 Chapter 6: Discussion………..28 References………..31 Appendixes……….32
4
Introduction
Patients with insomnia experience inadequate nocturnal sleep in some way. They may describe difficulty falling asleep, frequent awakening, or early morning awakening with inability to return to sleep. This disorder of initiating and/or
maintaining sleep causes sleep disruption and has major daytime consequences. There are two types of insomnia; primary and secondary insomnia. Primary insomnia not attributable to medical, psychiatric, or environmental causes, while secondary insomnia is. This review will focus on primary insomnia because it has a
homogeneous character because it does not reflect effects of a comorbid disorder. Primary insomnia affects approximately 3-5% of the general population (Ohayon, 2002) and is therefore an important topic of research.
A form of hyperarousal became the main concept of primary insomnia models. The hyperarousal model assumes that an underlying state of arousal interferes with normal biological processes associated with sleep initiation and maintenance (Bonnet and Arand, 1997). Many different models were designed, mostly based on somatic (e.g. heart rate) and cognitive arousal (e.g. dysfunctional thoughts). Riemann and colleagues (2010) summarized the current literature and generated a neurocognitive model of insomnia, consisting of a cognitive-behavioural and a neurobiological domain. The authors emphasize that primary insomnia can not only be linked to alterations on psychological level, but is also associated with measurable deviations of neuroendocrine as well as electro-and neurophysiological alterations.
In this review the focus will be on the neurobiological domain, because emerging techniques allowed us to investigate cortical activity more extensively. Moreover, because primary insomnia is not related to a comorbid disorder, it can be
5 hypothesized that there is some form of genetic vulnerability. This bottom-up
approach assumes a (genetically determined) dysfunction in sleep-wake regulating neural circuitries originating from the brainstem and leading to sleep disruptions. With the use of electroencephalography (EEG) and functional magnetic resonance imaging (fMRI), cortical and even subcortical arousal could be investigated. And the hyperarousal model can be reviewed with the focus on neurobiological arousal.
One limitation of studying insomnia is the occurrence of many different types. In the first chapter of this review, the different types and characteristics of insomnia will be discussed. The second chapter will provide some necessary background information on normal sleep processes. Neurophysiological as well as neurobiological processes of normal sleep will be quickly described. The third chapter is the focus of this review. In this chapter the hyperarousal model will be described, with its evidence based on electrophysiological and MRI studies. If a hyperaroused state is correlated with the pathology of primary insomnia, it can be questioned whether this is a predisposing factor or a consequence of the disorder. Therefore in the fourth chapter heritability studies of primary insomnia will be discussed. The fifth chapter describes current treatment of primary insomnia, which is mainly based on changing cognitive arousal by cognitive behavioural therapy. However, if primary insomnia is caused by a (genetic) hyperarousal of the neuronal sleep-wake circuitry, other treatment methods based on a bottom-up approach may be more successful. In the last chapter of the thesis the evidence will be considered and the main question will be answered: is hyperarousal the model for explaining the pathology of primary insomnia, based on a neurobiological perspective? Last, some suggestions for further research will be proposed.
6
Chapter 1: Different types and characteristics of insomnia.
In research on insomnia it is important to recognize the differences between the different types (primary and secondary) of insomnia. The general definition of insomnia is: ‘repeated difficulty with sleep initiation, duration, consolidation, or quality that occurs despite adequate time and opportunity for sleep, and this results in some form of daytime impairment’ (DSM-IV, 1994). The book ‘International
classification of sleep disorders’ distinguishes 11 different types of insomnia, of
which 3 types of primary insomnia. This heterogeneity of the disorder makes it
difficult to investigate and understand the psychopathology. Because of the possibility that the different types of primary insomnia also have different underlying causes, it is important to distinguish among the different types first.
The first type of primary insomnia is idiopathic insomnia. The definition of idiopathic insomnia is: ‘a lifelong inability to obtain adequate sleep that is
presumably due to an abnormality of the neurological control of the sleep-wake system’ (DSM-IV, 1994). So idiopathic insomnia is a long-standing disorder, which
typically begins during childhood, and has no clear environmental causes. The second type of primary insomnia is called psychophysiological insomnia. The definition of psychophysiological insomnia is: 'a disorder of somatised tension and learned
sleep-preventing associations that results in a complaint of insomnia and associated decreased functioning during wakefulness' (DSM-IV, 1994). So psychophysiological
insomnia is a disorder which often begins with a period of stress and can be
developed at any time point in life. Unfortunately, while there are clear distinctions between idiopathic and psychophysiological insomnia, such as the onset period and what triggers the pathology, not many studies specified the subtypes of primary
7 insomnia, and included idiopathic insomnia as well as psychophysiological insomnia patients. And because not many studies made a distinction between these types, this review includes both idiopathic as psychophysiological primary insomnia too. The third and last type of primary insomnia is called sleep-state misperception (also known as paradoxical insomnia). This type will not be included in this review, because sleep-state misperception insomnia is referred as subjective insomnia since the dysfunctional sleep can not be objectively measured using physiological
measurements (DSM-IV, 1994). Some studies suggest that sleep-state misperception may be a transitional state of dysfunction between an objective sleep disorder
(idiopathic and psychophysiological) and normal sleep (Salin-Pascual et al, 1992; Bonnet & Arand, 1997). However, because the distinct ability of this subtype is still questionable (Edinger& Krystal, 2003), no studies including these patients are reviewed.
8
Chapter 2: Normal sleep processes
In order to say something about processes causing sleep disorders, we first need to consider normal sleep processes. Many different models on normal sleep regulation have been developed in the last decade. A very common model is the two-process model by Borbély (1982). The two-two-process model postulates that two processes are important in the regulation of sleep, a homeostatic factor (process S) and a circadian factor (process C). The homeostatic factor rises during wakefulness and declines during sleep. So this is an appetitive process that is determined by the amount of time since the last bout of sleep. The longer the time spent awake, the greater the drive for sleep. On the other hand, the circadian factor (process C) varies with a 24-hour periodicity and is independent of the amount of preceding sleep or wakefulness. Together these processes modulate the need for sleep and influence the balance between central alertness and sleepiness (see figure 1).
A study of Perlis and colleagues (2001) suggested that this two-process model should be expanded with a second homeostatic ‘process W’. Where the 'process S' increases with extended wakefulness, the 'process W' is thought to increase with extended sleep (Perlis et al., 2001). These two processes interact with each other and deficits can cause sleep initiation and maintenance difficulties. A sleep disorder can be caused by a decreased sleep propensity (so a low 'process S') or an increased wake propensity (a hyperarousal or high 'process W'). In this case, primary insomnia can be hypothesized as a sleep disorder mainly due to an abnormal or increased 'process W'.
9 Figuur 1: Two-process model for normal sleep (from Borbély, 1982). The C refers to the
circadian factor and the S to the homeostatic factor.
2.1 Neurophysiology of normal sleep
The beginning of sleep research dates back to 1930s, where the studies mainly used electroencephalogram (EEG) techniques. Kleitman and Dement (1957) were the first to specify rapid-eye-movement sleep (REM sleep) and defined the cyclic pattern of sleep as one cycle of REM-nonREM that lasts 90-100 minutes, with 4-5 cycles repeating during one night. The non-REM sleep period is more specified by a distinction of 3 different stages, based on EEG wave patterns. Every stage has a characteristic EEG frequency and waveform. The first stage is characterized by a transition from alpha waves (8-13 Hz) to theta waves (4-7 Hz). During stage 2 sleep spindles (11-16 Hz) and K-complexes can occur. Sleep spindles and K-complexes are rapid bursts of activity. Going into detail on this phenomenon will be beyond the scope of this review. The third stage, also called slow-wave-sleep (SWS), is characterized by an extreme amount of delta waves (0-4 Hz) (Carskadon & Rechtschaffen, 2000).
For primary insomnia, the transition from wakefulness to sleep, also called the sleep-onset-period (SOP), is extremely interesting because the inability of patients to initiate sleep. In normal sleepers, the SOP is characterized by a decrease of high frequency and an increase of slow frequency EEG activity (De Gennaro et al., 2001). These results show global differences in EEG patterns. A more precise and localized
10 image of the processes during sleep can be made with the use of functional resonance imaging (fMRI).
2.2 Neurobiology of normal sleep
Explaining the history of sleep research and describing all neuroimaging studies on normal sleep processes would be beyond the scope of this review. However, to get some background knowledge I will briefly summarize the current literature of brain activation patterns during NREM and REM sleep in healthy participants.
One main finding of early functional neuroimaging studies of sleep is a global decrease of cerebral activity during NREM sleep compared to wakefulness, similar to the EEG measurements (Buchsbaum et al., 1989; Maquet et al., 1992 ; Nofzinger et al., 2002). Other studies found evidence for regional differences in activity patterns during sleep. Decreased activity in thalamic structures during NREM sleep compared to wakefulness has been found (Braun et al., 1997 ; Andersson et al., 1998; Kajimura et al., 1999; Kjaer et al., 2002; Nofzinger et al., 2002). Second, a decrease in basal ganglia structures (e.g. the caudate/ putamen) has been found repeatedly (Braun et al., 1997; Kajimura et al., 1999). Third, a decrease in several frontal cortex areas (Braun et al., 1997; Kajimura et al., 1999; Nofzinger et al., 2002). And fourth, REM sleep can be characterized by an increased metabolism within the limbic system, pons and the visual cortex (Maquet et al., 1996).
If we even zoom in more closely to the midbrain structures, a network that promotes wakefulness can be defined, the ascending arousal reticular system (ARAS). Evidence for the existence of an arousal system came from early work of Von
11 excessively, had lesions at the junction of the midbrain and the diencephalon. This area was thought to be the beginning of an arousal system, which ascends to the forebrain and promotes wakefulness. Two decades later, a study of Moruzzi and Mogoun (1949) clarified the ARAS more, using EEG techniques. Moruzzi and Mogoun were the first to divide the ARAS into two branches, which were confirmed in 2005 by Saper and colleagues (2005). One pathway, from the laterodorsal
tegmental nuclei (LDT) and pedunculopontine (PPT) to the thalamus, is the branch that activates neurons crucial for the transmission of information to the cerebral cortex. Neurons in this branch fire rapidly during wakefulness and REM sleep (figure 2, the yellow branch). The second important branch in the ARAS originates from the upper brainstem and caudal hypothalamus (including the locus coeruleus (LC), the raphe nuclei and the tuberomammillary nucleus (TMN)) and projects to the basal forebrain. Neurons in this branch fire fastest during wakefulness, slowing down during NREM sleep and stopping during REM sleep (figure 2, the red branch).
12 Contrary to the ARAS, there is a network that promotes sleep. A key role in this network is assigned to the ventrolateral preoptic nuclei (VLPO), a group of neurons located in the hypothalamus. The VLPO appears to be strategically well placed because of its robust connections with hypothalamic regions (Chou et al., 2002). Moreover, major descending projections from the VLPO to the TMN and more regions of the ARAS were found, using tracing methods (Sherin et al., 1998). The neurotransmitters present in the VLPO are mainly gamma-aminobutyric acid (GABA). GABA is known as the most important and ubiquitous inhibitory
neurotransmitters of the central nervous system. This suggests that the VLPO may provide inhibitory GABAergic inputs to these cell bodies in the ARAS, and therefore may play a key role in silencing the ARAS and promoting sleep (see figure 4). Figuur 2: A schematic drawing showing key components of the ARAS. Red branch: neurons that fire during wakefulness and slow down during NREM/REM. Yellow branch: neurons that fire during wakefulness and REM sleep (from Saper et al., 2005, pg1258)
13 The arousal and sleep promoting networks interact with each other creating the balance between sleep and wake processes. The transition from wake to sleep occurs very rapidly and it is therefore called a flip-flop switch, you are either awake or asleep, but not much time is spent in the transitional state. The question that rises is: what triggers this switch? Apparently, something makes the balance towards
activation of the one network, or the other. A neuropeptide ‘orexin’ seems to be highly important. This neuropeptide is produced only in the posterior half of the lateral hypothalamus (LH). The orexin neurons are mainly active during wakefulness and can directly reinforce the arousal system (Lee et al., 2005). These orexin neurons can have an indirect effect on the VLPO (see figure 3). During wakefulness the VLPO is inhibited through the regions of the ARAS, which are directly reinforced by the orexin neurons. On the other hand, during sleep, the VLPO inhibits the orexin neurons Figuur 3: A schematic drawing showing primary projections of the VLPO to the main
14 and the regions of the ARAS. This inhibition is strengthened by the fact that the orexin neurons inhibit the ARAS regions even more.
Figuur 4: A schematic drawing of the flip-flop switch model (from Saper, 2005). a) the balance during wakefulness; b) the balance during sleep.
Taken together, these studies suggest that normal sleep can be characterized by global and regional deactivations. During the SOP, the VLPO and several basal ganglia areas are key factors in the regulation of sleep. During NREM sleep, the thalamus, basal ganglia, pons and some prefrontal cortex areas show great deactivation, and therefore may be important.
15
Chapter 3: Hyperarousal model of insomnia
A hyperaroused state became the main concept of different models on primary insomnia. This hyperarousal can be found in three different dimensions. For instance somatic arousal causes you to have heightened sensory response, which can be seen by increased automatic activity such as elevated heart rate/respiration and the blood levels of stress hormones increase dramatically as has been found in poor sleepers repeatedly (Monroe, 1967; Adam et al., 1986; Bonnet, 1995).
Another form of arousal is cognitive arousal. Cognitive arousal is reflected by intrusive and negative cognitions and their associations with sleep complaints
(Wicklow & Epsie, 2000; Lichstein and Rosenthal, 1980). The last proposed form of arousal is defined as cortical arousal. In this review I will investigate the current literature on cortical (hyper)arousal in primary insomnia, this can also be called the neurobiological aspects of insomnia. This research field has expended dramatically the last decade, due to new techniques such as EEG and fMRI, which allow us to have a closer look at the biological basis of hyperarousal.
3.1 Neurophysiology of primary insomnia (cortical arousal)
Research using quantitative EEG (QEEG) techniques provides insights in cortical arousal in insomnia. For example, cortical electrophysiological signals in the beta and especially gamma band have been hypothesized as a main feature of coherent cortical processing of sensory information (Llinás & Ribary, 1993), where an EEG power increase in this frequency range can be interpreted as a sign of cortical hyperactivity or hyperarousal. Therefore, analysing EEG patterns in insomniacs can provide meaningful information, in favour or against the hyperarousal hypothesis.
16 As noted above, the transition from wakefulness to sleep, also called the sleep onset period (SOP), entails a shift in the frequency power spectrum of the EEG from high to low frequencies in healthy individuals (Wright et al., 1995). This period is of specific interest in the light of primary insomnia. Merica and Gaillard (1992) were one of the first to study the SOP in insomniac patients. They confirmed the idea that the SOP of normal sleepers can be characterized by a rapid change in beta and delta activities. Beta activity decreases and delta activity increases, both in insomniacs as in healthy controls. However, insomniacs had higher beta and lower delta activity
compared to controls in the SOP and the first minutes of the first NREM cycle. The differences in delta activity between insomniacs and healthy controls were significant. The authors of this article suggest that this increased high frequency activity may explain the tendency of insomnia patients to overestimate their sleep latencies.
Lamarche and Ogilvie (1997) performed a similar study, looking at the SOP in psychophysiological insomniacs, non-specified psychiatric insomniac and control sleepers. They extended previous findings by defining the SOP into four quartiles. The results showed that insomniac exhibit less delta power in the SOP, specifically in the last quartile. Reduced alpha activity was found during the first quartile of the SOP in insomniacs, and it did not show a dramatic drop, which is common during normal sleep. Moreover, they found that increased relative beta power during wakefulness in primary insomnia patients.
In line with previous findings, Staner and colleagues (2003) found that patients with primary insomnia did not experience a decrease of their alpha and beta power during the SOP, and showed lower delta activity in the first 5 minutes of sleep. They concluded that hyperarousal mainly affects the sleep initiation of primary
17 insomnia patients whereas the homeostatic sleep regulation process seems to be partially maintained.
In contrast to the study of Staner and colleagues, Merica and Guillard (1998) questioned whether primary insomnia affected only the SOP, or if this dysregulation can be seen in other sleep stage too. They were the first who did not limit their EEG data to an interval at/around sleep onset and also examined the first four NREM/REM cycles. Consistent with previous studies, increased beta power in insomnia patient compared to healthy controls was found during the SOP. However, elevated beta activity was found during the whole night in primary insomnia patients, emphasizing the importance to investigate the whole sleep pattern instead of focussing only on the SOP. Moreover, they found increased power in fast frequency bands and a decreased power in the delta and theta bands in insomnia patients during REM sleep. One major disadvantage of this study is that the authors did not remove all episodes of
wakefulness within each NREM/REM cycle before performing the power analysis. They only excluded episodes that contained periods of wakefulness lasting 5 minutes or more. This may have resulted in group differences for high frequencies, that can be explained by differences in short periods of wakefulness. However, these results were the first to show that primary insomnia also affects the REM sleep. Moreover, they suggest that not only the sleep initiation is dysregulated in primary insomnia, but also the maintenance of sleep.
In table 1 all the results are summarized. First, beta activity (14-35Hz) is increased (or drops more slowly) in primary insomnia relative to good sleepers, both around the SOP as during NREM sleep (Merica et al., 1992;1998; Nofzinger et al., 1999; Perlis et al., 2001; Lamarche and Ogilvie, 1997). Second, insomnia patients exhibit reduced levels of alpha power during NREM, wakefulness and sleep onset
18 period (Merica et al., 1998; Lamarche and Ogilvie, 1997), and increased levels of alpha activity during the SOP (Lamarche and Ogilvie, 1997; Staner et al., 2001). Third, theta power during NREM and REM sleep was decreased in insomniacs (Merica et al., 1998). Fourth, delta power appears to be lower in insomniacs during the SOP, NREM and REM sleep (Lamarche and Ogilvie, 1997; Merica et al., 1998; Staner et al.,2003). And fifth, higher gamma power was observed during NREM sleep (Krystal et al., 2002). In conclusion, insomniacs show elevated levels of high
frequency and a reduction of slow frequency EEG activity during all phases of sleep compared to normal sleepers, which is evidence in favour of the hyperarousal hypothesis (see table 1).
Table 1: EEG differences in PI compared to healthy sleepers (during first NREM cycle, except for the study of Merica and Gaillard (1998) who investigated the first four NREM/REM cycles).
Stage EEG frequency Effect Article Wakefulness Alpha (8-12 Hz)
Beta (14-35 Hz)
Decreased in PI Increased in PI
Lamarche and Ogilvie (1997)
SOP Delta (0-4 Hz) Decreased in PI Merica and Gaillard (1992),
Lamarche and Ogilvie (1997)
Alpha (8-12 Hz) Increased in PI Lamarche and Ogilvie (1997)
Staner et al., (2003) Beta (14-35 Hz) Dropped more slowly/
increased in PI
Merica and Gaillard (1992 ‘ns’; 1998) Staner et al., (2003) La Marche and Ogilvie (1997)
19
NREM Delta (0-4 Hz) Decreased in PI Staner et al., (2003), Merica and Gaillard (2008)
Theta (4-7 Hz) Decreased in PI Merica and Gaillard (2008)
Alpha (8-12 Hz) Decreased in PI Merica and Gaillard (2008)
Beta (14-35 Hz) Increased in PI Merica and Gaillard (1998)
Nofzinger et al., (1999) Gamma (25-100 Hz) Increased in PI Krystal et al., 2002
REM Delta (0-4 Hz) Decreased in PI Merica and Gaillard (2008)
Theta (4-7 Hz) Decreased in PI Merica and Gaillard (2008)
Alpha (8-12 Hz) Increased in PI Merica and Gaillard (2008)
Beta (14-35 Hz) Increased in PI Merica and Gaillard (1998)
EEG: electroencephalography, SOP: sleep onset period, NREM: non-rapid-eye movement sleep, REM: rapid-eye-movement sleep
There are several limitations of sleep research using QEEG measurements. First, some studies focused on a limited portion of the sleep EEG spectrum (e.g. Staner et al,, 2001) while others focused on a specific sleep period (e.g. Merica and Gaillard, 1992; Lamarche and Ogilvile 1997). Second, the measurement is susceptible to confounding artefacts from electromyography (EMG) signals (Goncharova et al., 2003). And last, in contrast to the excellent temporal resolution of EEG, the spatial resolution of is very limited. EEG is only able to measure cortical activity (and no
20 subcortical brain structures) and therefore can not give information about more
subcortical brain activity. As in normal sleep, new techniques as MRI allow researchers to measure brain activity at lower levels, including subcortical brain circuits and areas. This may give new insights in the underlying brain mechanisms controlling sleep.
3.2 Neurobiology of primary insomnia (sub-cortical arousal)
If abnormal sleep is a result of a failure of sleep promoting areas, we would expect to find a decreased activity of the VLPO and several basal ganglia areas. If abnormal sleep is a result of a failure of wake promoting areas, we would expect to find an excessive ascending activation through increased activity areas located in the ARAS (thalamus, basal ganglia, pons and prefrontal cortex). This last expectation is the main idea of the hyperarousal hypothesis.
To my knowledge only two studies investigated functional abnormalities of insomniacs, directly looking at brain metabolism with the use of MRI (Nofzinger et al., 2004; Smith et al., 2002).
Smith et al., (2002) used a single photon emission computed tomography (SPECT) method, to investigate the brain metabolism in insomniacs. Their results showed a significant decreased regional cerebral blood flow in the frontal medial cortex, frontal lateral cortex, basal ganglia, thalamus, pons and temporal, occipital and parietal cortices during the first cycle of NREM sleep in insomniacs compared to good sleepers.
And a recent neuroimaging of Nofzinger and colleagues (2004), who used fluordeoxyglucose positron emission tomography (PET) to assess the regional cerebral glucose metabolism (CMRglu). Contrary to the results of Smith and
21 colleagues (2002), Nofzinger and colleagues (2004) showed that in the transition from wakefulness to NREM sleep insomniacs showed increased global glucose metabolism compared to healthy controls. Secondly, insomniacs exhibited less reduction of relative CMRglu in the ARAS, hypothalamus, insular cortex, amygdala,
hippocampus, anterior cingulate cortex and the medial prefrontal cortex during NREM sleep. Third, PI patients show an increased metabolism in the frontal cortex, superior temporal cortex, parietal and occipital cortices, and in the thalamus,
hypothalamus, and brainstem reticular formation during wakefulness. These results were the first to directly show a dysfunctional deactivation or an inhibition of the arousal mechanisms at the transition from wakefulness to (NREM) sleep in humans with primary insomnia. The differences in metabolism during wakefulness may be responsible for the daytime impairments of insomnia, and are shown to be a related to sleep deprivation (Thomas et al., 2000).
Differences between these two studies can be caused by several aspects. Despite the advantages of MRI techniques, studying sleep parameters with the use of a MRI scanner does not come without disadvantages. Two major problems of
studying sleep in the MRI scanner are: difficulties with co registration of the polysomnography (a measurement of brain waves (EEG), eye movement, muscle activity, of heart rhythm, used to define sleep stages) and functional imaging data. And secondly, it is hard to capture tomography during sleep from uniform sleep stages. Due to the magnetic field, the EEG signal can easily be disrupted, if the subjects were able to sleep at all in the noisy MRI environment (Smith et al., 2002). These general limitations of using the MRI scanner in sleep research may have influenced the results of both studies in different ways.
22 Another possible explanation of the different results may due to different measurement techniques. Nofzinger and colleagues (2004) used PET recordings, where Smith and colleagues (2002) used SPECT recordings. Both techniques use an injection of a different short-lived radioactive tracer isotope, which decays by emitting a positron. On the one hand, SPECT imaging is a fast scan completed, but lacks the spatial resolution of PET imaging. Where on the other hand PET imaging allows a more refined analysis at the neuroanatomical level, but involves a much longer uptake window, because of the slow tracer. This results in much longer scanning times (Meikle et al., 2006). Since blood flow is known to decline with increasing duration of NREM sleep, these small differences due to different
techniques makes it difficult to compare the studies and to get a uniform picture of the alternations in brain metabolism in primary insomnia.
3.2.2 Neurotransmitters
As pointed out before, sleep is a consequence of several brain areas connecting with each other in a network. This occurs under the influence of neurotransmitters, hormones, receptors and connections. The neurotransmitter GABA may have a key role in promoting sleep by inhibiting midbrain areas of the ARAS. If insomnia patients exhibit hyperarousal, this may be caused by decreased GABA to inhibit the brain areas located in the ARAS. In patients with primary insomnia, reduced daytime overall average brain GABA levels, averaged from basal ganglia, thalamus, and parietal, occipital, and temporal white matter and cortical regions, were observed (Winkelman et al., 2008). Remarkably, GABA levels correlated with both subjective and objective sleep measures. Longer wake time after sleep onset was associated with
23 lower levels of GABA. To our knowledge, this is the only investigation focussing on neurotransmitters in primary insomnia.
Chapter 4: Heritability
If evidence is found for the occurrence of hyperarousal in primary insomnia, it can be questioned if hyperarousal is a contribution to insomnia or a consequence of it. This leads us to the question: are there predisposing factors in insomnia?
There are a couple studies who investigated this topic. Bastien and colleagues (2000) evaluated the familial incidence of sleep disturbances among individuals with sleep complaints. They found that 35% of patients consulting for insomnia had a positive family history of sleep disturbances, with trends suggesting a higher frequency among patients with primary insomnia, an early age of onset and sleep onset insomnia. Particularly the mother was the most afflicted family member. Similar results were found by Beaulieu-Bonneau and colleagues (2007), who reported that 34.9% of their participants had at least one first-degree relative, with past or current insomnia. And again, the mother was the most frequently afflicted (19.7%).
Two major disadvantages of the study of Bastien and colleagues (2000) were that the data were obtained by self-report and the absence of a control group. This way the results are based on subjective judgements instead of direct assessments and individuals with past or current insomnia are more likely to report a positive family history of insomnia than individuals without any history of insomnia (Beaulieu-Bonneau et al., 2007). A study by Dauvilliers and colleagues (2005) did include a control group. They compared familial insomnia in three groups, primary insomnia,
24 psychiatric insomnia and a non-insomnia control groups. They results showed reports of familial insomnia in 72.7%, 43.3% and 24.1% respectively. Similar to previous studies, the mother was the relative most frequent affected and there was a tendency to younger age at onset observed in familial and primary insomnia. From these studies it can be concluded that there is strong evidence for heritability in primary insomnia if we look at behavioural scores. This leads us to the question: ‘What is the genetic component of primary insomnia?’.
Ambrosius and colleagues (2008) performed a classical twin study in monozygotic and dizygotic twin pairs on the genetic effect of sleep EEG in healthy participants. They found that the relative contributions of the delta, theta, alpha and sigma frequency bands were significantly correlated to the genetic background. In other words, the monozygotic twin pairs showed significantly higher within-pair concordance in spectral power of these frequencies, compared to dizygotic twin pairs. These results showed that the spectral composition of non-REM sleep in humans is significantly influenced by genetics. And because it has been found that insomnia is characterized by specific EEG pattern, these results suggest that a hyperaroused EEG pattern may be the main cause of primary insomnia.
Sadly, no studies have investigated a possible genetic component in brain alternations of the neuronal sleep-wake circuitry. This would be of specific interest in the light of the hyperarousal hypothesis.
25
Chapter 5: Treatment
Current treatment options for primary insomnia are mainly based on behavioural aspects, including stimulus control therapy, sleep restriction therapy, relaxation therapy, cognitive behaviour therapy, sleep hygiene education (Principles and practice of sleep medicine, pp. 726-737). The aim of cognitive behaviour therapy for insomnia (CBT-I), which is considered the treatment of choice for primary
insomnia, is to change maladaptive sleep habits, reduce autonomic arousal and alter dysfunctional beliefs as for instance rumination (Morin et al., 1994; Murtagh et al., 1995). However, there is a significant group of patients (about 20%) that does not respond to cognitive behaviour therapy (CBT) (Morin et al., 1994, Harvey et al., 2003) and the patients that do respond, show an average improvement but not
sufficient to categorize them as ‘good sleepers’ (Harvey et al., 2003). I suggest that in the light of the hyperarousal view it would be of interest to think about a treatment approach that works from bottom-up instead of normal CBT, that is based on a top-down view. Regulating the specific neurobiological alternations in insomnia patients may result in a more successful treatment approach.
There are two different methods to achieve a bottom-up approach in treatment. One of them is by using pharmacological interventions. Current pharmacological treatment interferes with the function of neurotransmitters in the ARAS by
counteracting the result of wake-promoting systems. For instance, specific targets can be regulated by adding a histamine H1-antagonist (e.g. Doxepin or Diphenhydramine) to counteract histamine from the TMN or a serotonin 5-HT2a-receptor antagonist (e.g. Trazodone) to block serotonin in the DR (Szabadi, 2006). Another method can be to inhibit the wake-promoting system itself, instead reacting on the results of an active
26 ARAS. A benzodiazepine receptor agonist (BZRA) mimics the action of the VLPO by inhibiting the ARAS through increased activity of GABA (Gottesman, 2002)
The main disadvantage of pharmacological treatment is that some drugs are not prescribed for long-term use, because of their side effects (Ringdahl et al., 2004; Kripke, 2006). Moreover, there is a risk for substance dependence. Even with relative save drugs such as Zolpidem and Zopiclone, which are both prescribed for treatment of insomnia (Hakak et al., 2003). Furthermore, not all sedatives show greater
improvement compared to CBT in older adults, who are more vulnerable for side effects (Sivertson et al., 2006).
A possible new treatment to focus on the hyperarousal instead of cognitive arousal, but without clear pharmacological side effects, is EEG neurofeedback. Neurofeedback is a new method based on training a patients’ self regulation (Othmer et al., 1999). The patient receives feedback of his EEG recordings and is trained to normalize the functioning of the brain by inhibiting or reinforcing specific frequency bands. In the light of hyperarousal as a leading cause of primary insomnia, and significant EEG evidence for this, this treatment may be successful by training the patients to regulate their cortical arousal. This may result in a decreased arousal and therefore positive effect on primary insomnia. Because primary insomnia is
considered a 24-h disorder (Bonnet & Arand, 1995) training of wake EEG can result in changes of the sleep EEG (Sterman et al., 1970). Thus neurofeedback may be a successful new treatment possibility. Cartoos and colleagues (2010) performed an exploratory study on the effects of neurofeedback and showed a significant
improvement in total sleep time after neurofeedback training. Further research on this treatment method is needed.
27 Concluding, neurofeedback may be in interesting new treatment technique by reducing the cortical hyperarousal in primary insomnia. However one shortcoming of this technique is that no study investigated the effects of neurofeedback on subcortical hyperarousal and the regulation of metabolism in for instance the ARAS system. Based on the current evidence I suggest a combined treatment of neurofeedback and pharmacological treatment for primary insomnia.
28
Chapter 6: Discussion
In this review, I summarized the current literature on the hyperarousal hypothesis of sleep in primary insomnia, with the use of EEG or MRI methods. I questioned whether there is evidence for a hyperarousal model on neurobiological basis. I discussed the literature from cortical to subcortical levels of arousal. First EEG studies were described, and after that the scarce literature on MRI studies was described.
In EEG studies, elevated levels of high frequency and a reduction of slow frequency EEG activity has been repeatedly found during all phases of sleep, in insomnia patients compared to normal sleepers. Technical limitations result in an inability to adequately localize regional activity throughout the brain. This resulted in the need for another, more advanced, measurement technique. However, despite this limitation, these results clearly support the hyperarousal hypothesis.
From MRI studies no clear conclusions can be drawn because currently available data are limited and inconsistent. The results of the two current studies (Smith et al, 2002; Nofzinger et al, 2004) are contradictive. A possible explanation for this clear difference may lie within the changing of brain metabolism over time. To date, no fMRI study recorded more than the first part of the first NREM cycle. Perlis and colleagues (2001) showed that the heightened beta activity in insomniacs during the SOP rapidly attenuates across the sleep-wake transition. This decline was found to be significantly greater in patients with primary insomnia and leads to a brief period of hypoarousal. With this difference in arousal over time, it is possible that the 2-min time window of the study of Smith accidently recorded this temporary hypoarousal. This emphasises the importance of future research on a whole night cycle with fMRI.
29 Such research can give more insight in the question whether primary insomnia
patients are generally hyperaroused during NREM sleep, as found by Nofzinger and colleagues (2004) or if they show periods or hyperarousal alternating with periods of hypoarousal.
A dysfunction of the sleep-wake circuitry may be a result of difference in neurotransmitters. To date, only one study investigated differences in the amount of neurotransmitters in primary insomnia. Winkelman and colleagues (2008) found global decreased GABA in primary insomnia patients, suggesting a failure to inhibit wake-promoting areas.
In the light of treatment possibilities, we propose a new treatment method. The current behavioural treatments (Principles and practice of sleep medicine, pp. 726-737) have proven to be successful in reducing arousal by changing mostly thoughts and rumination. However, because primary insomnia can be caused by (sub)cortical instead of cognitive arousal, other treatment methods may be more successful.
Pharmacological treatment has proven to be successful, and a new proposed method is neurofeedback, which focuses on reducing cortical arousal. I suggest a combination of these two treatment modalities. When combined, pharmacological treatment is used to change the arousal on neurotransmitter level, where simultaneously neurofeedback can result in reducement of cortical arousal. Together these two therapies may be the most successful treatment method of primary insomnia.
The available data generally support the hyperarousal theory of insomnia with increased cortical activity during NREM sleep. However, the key brain areas
responsible for the hyperarousal remain unclear. Further research on the
identification of regional brain activation should allow development of increasingly specific medical treatments, to control abnormal activation.
30 Moreover, differences on neurotransmitter level and the temporal
characteristics of hyperarousal should be investigated. As said before, the study of Perlis and colleagues (2001) suggest that hyperarousal may be a dynamic, instead of static process. Therefore, a whole night fMRI study looking at the hyperarousal over time would be very useful.
In this review I conclude that hyperarousal is key factor in the pathology of primary insomnia, based on neurophysiological evidence. Moreover, the literature suggests a hyperarousal of the ARAS as underlying factor of primary insomnia. But due to limited research this hypothesis can not be confirmed (yet).
31
Appendix 1: Definitions of subtypes of insomnia, according to the ICSD.
Psychophysiological insomnia (307.42-0)
Learned insomnia, conditioned insomnia, functionally autonomous insomnia, psychophysiological arousal, chronic somatised tension, internal arousal without psychopathology.
Psychophysiological insomnia is a disorder of somatised tension and learned sleep-preventing associations that results in a complaint of insomnia and associated decreased functioning during wakefulness. Psychophysiological insomnia is an objectively verifiable insomnia that develops as a consequence of two mutually reinforcing factors: (a) somatised tension and (b) learned sleep-preventing
associations. Individuals who have psychophysiological insomnia typically react to stress with somatised tension an agitation. The meaning of stressful events (other than insomnia) is typically denied an represses, but manifests itself as increased
physiological arousal (e.g., increased muscle tension, increased vasoconstriction, etc.)
Idiopathic insomnia (780.52-7)
Childhood insomnia, lifelong insomnia, insomnia associated with problems within the sleep-wake system, excessive arousal, inadequately developed sleep system.
Idiopathic insomnia is a lifelong inability to obtain adequate sleep that is presumably due to an abnormality of the neurological control of the sleep-wake system.
32
References
Adam, K., Tomeny, M., Oswald, I. (1986). Physiological and psychological
differences between good and poor sleepers. Journal of psychiatric research, 20(4), 301-316.
Ambrosius, U., Lietzenmaier, S., Wehrle, R., Wichniak, A., Kalus, S., Winkelmann, J., Bettecken, T., Holsboer, F., Yassouridis, A., Friess, E. (2008). Heritability of sleep electroencephalogram. Biological Psychiatry, 64, 344-348.
American Psychiatric Association (APA). Diagnostic and statistical manual of mental disorder. 4th ed. Washington, DC: APA; 1994.
American Sleep Disorders Association, The international classification of sleep disorders: diagnostic and coding manual-revised. Rochester, M.N: Americal Sleep Disorders Association 1997.
Andersson, J.L., Onoe, H., Hetta, J., Lidstrom, K., Valind, S., Lilja, A., et al. (1998). Brain networks affected by synchronized sleep visualized by positron emission tomography. Journal of Cerebral Blood Flow & Metabolism, 18 (7), 701-715. Beaulieu-Bonneau, S., LeBlanc, M., Mérette, C., Dauvilliers, Y., Morin, C.M. (2007).
Family history of insomnia in a population-based sample. Sleep, 30(12), 1739-1745.
Bonnet, M.H., Arand, D.L. (1995). 24-hour metabolic rate in insomniacs and matched normal sleepers. Sleep, 8, 581–8.
Bonnet, M.H., Arand, D.L. (1997). Hyperarousal and insomnia. Sleep medicine reviews, 1(2), 97-108.
Bonnet, M.H., Arand, D.L. (1997). Psysiological activation in patients with sleep state misperception. Psychosomatic medicine, 59(5), 533-540.
Borbély, A.A., (1982). A two process model of sleep regulation. Human Neurobiology, 1: 195-204.
Braun, A.R., et al. (1997). Regional cerebral blood flow throughout the sleep-wake cycle. An H2(15)O PET study. Brain, 120(Pt7), 1173-1197.
33 Buchsbaum, M.S., Gillin, J.C., Wu, J., Hazlett, E., Sicotte, N., Dupont, R.M. et al.
1986. Regional cerebral glucose metabolic rate in human sleep assessed by positron emission tomography. Life Sciences, 45(15), 1349-1356.
Cartoos, A., De Valck, E., Arns, M., Breteler, M.H.M., Cluydts, R. (2010). An exploratory study on the effects of tele-neurofeedback and tele-biofeedback on objective and subjective sleep in patients with primary insomnia. Applied Psychophysiological Biofeedback, 35, 125-134.
Chou, T.C., Bjorkum, A.A., Gaus, S.E., Lu, J., Scammell, T.E., Saper, C.B. (2002). Afferents to the vertrolateral preoptic nucleus. The Journal of Neuroscience, 22(3), 977-990.
Dauvilliers, Y., Morin, C., Cervena, K., Carlander, B., Touchon, J., Besset, A., Billiard, M. (2005). Family studies in insomnia. Journal of Psychosomatic Research, 58, 271-278.
Dement, W.C., Kleitman, N. (1957). Cyclic variations in EEG during sleep and their relation to eye movements, body motility and dreaming. Clinical Neurophysiology,
9:673–390.
De Gennaro, L., Ferrara, M., Bertini, M., (2001). The boundary between wakefulness and sleep : quantitative electroencephalographic changes during the sleep onset period. Neuroscience, 107, 1-11.
Chairman, T.M.J. (1990). ICSD – International classification of sleep disorders: Diagnostic and coding manual. Diagnostic Classification Steering Committee, Rochester, Minnesota: American Sleep Disorders Association.
Edinger, J.D., Krystal, A.D. (2003). Subtyping primary insomnia: is sleep state misperception a distinct clinical entity? Sleep Medicinereviews, 7(3), 203-214. Goncharova, I.I., McFarland, D.J., Vaughan, T.M., Wolpaw ,J.R.. EMG
contamination of EEG: spectral and topographical characteristics (2003). Clinical Neurophysiology, 114(9), 1580-1593.
Gottesman, C. (2002). GABA mechanisms in sleep. Neuroscience, 111, 231-239. Hakak, G., Müller, G.W., Wittchen, H.U., Pittrow, D., Kirch, W. (2003). Abuse and
dependence potential for the non-benzodiazepine hypnotics zolpidem and
zoplicone: a review of case reports and epidemiological data. Addiction, 98, 1371-1378.
Harvey, A.G., Tang, N.K.Y. (2003). Cognitive behavior therapy for primary insomnia: can we rest yet? Sleep Medicine Review, 7, 237-62.
34 Kajimura, N., Uchiyama, M., Takayama, Y., Uchida, S., Uema, T., Kato, M., et al.
(1999). Activity of midbrain reticular formation and neocortex during the progression of human non-rapid eye movement sleep. Journal of Neuroscience, 19(22), 10065-10073.
Kjaer, T.W., Law, I., Wiltschiotz, G., Paulson, O.B., Madsen, P.L. (2002). Regional cerebral blood flow during light sleep-a H(2)(15)O-PET study. Journal of Sleep Research, 11, 201-207.
Kripke, D.F,. Risks of chronic hypnotic use. In: Lader M, Cardinali DP, Pandi- Perumal SR, editors. Sleep and sleep disorders – a neuropsychopharmacological approach. New York: Springer; 2006. p. 141–5.
Krystal, A.D., Edinger, J.D., Wohlgemuth, W.K., Marsh, G.R., (2002). NREM sleep EEG frequency spectral correlates of sleep complaints in primary insomnia subtypes. Sleep, 25(6), 630-40.
Lamarche, C.H., Ogilvie, R.D. (1997). Electrophysiological changes during the sleep onset period of psychological insomniacs, psychiatric insomniacs, and normal sleepers. Sleep, 20, 724-33.
Lee., M.G., Hassani, O.K., Jones, B.E. (2005). Discharge of identified
orexin/hypocretin neurons across the wake-sleep cycle. Journal of Neuroscience, 25, 6716-6720.
Lichstein, K.L., Rosenthal, T.L. (1980). Insomniacs’ perceptions of cognitive versus somatic determinants of sleep disturbance. Journal of Abnormal Psychology, 89(1), 105-107.
Maquet P, Dive D, Salmon E, Sadzot B, Franco G, Poirrier R, et al. (1992). Cerebral glucose utilization during stage 2 sleep in man. Brain Research, 571(1), 149-153. Maquet, P., Peters, J., Aerts, J., Delfiore, G., Degueledre, C., Luxen, A., et al. (1996).
Functional neuroanatomy of human reapid-eye-movement sleep and dreaming. Nature 383(6596), 163-166.
Meikle, S.R., Beekman, F.J., Rose, S.E. (2006). Complementary molecular imaging technologies: High resolution SPECT, PET and MRI. Drugs Discovery Today: Techonolgies, 3(2), 178-194.
Merica, H., Gaillard, J. (1992). The EEG of the sleep onset period in insomnia: a discriminant analysis. Physiology and Behavior, 52, 199-204.
Merica, H., Blois, R., Gaillard, J. (1998). Spectral characteristics of sleep EEG in chronic insomnia. European Journal of Neuroscience, 10, 826-34.
35 Monroe, L.J. (1967). Psychological and physiological differences between good and
poor sleepers. Journal Abnormal Psychology, 72, 255-64.
Morin, C.M., Culbert, J.P., Schwartz, S.M. (1994). Nonpharmacological interventions for insomnia: A meta-analysis of treatment efficacy. American Journal of
Psychiatry, 151, 1172-1180.
Moruzzi, G., Magoun, H.W. (1949). Brain stem reticular formation and activation of the EEG. Electroenceohalography and Neurophysiology, 1(4), 455-73.
Murtagh, D.R.R., Greenwood, K.M. (1995). Identifying effective psychological treatments for insomnia: A meta-analysis. Journal of Consulting and Clinical Psychology, 63, 79-89.
Nofzinger, E.A., Buysee, D.J., Miewald, J.M., Meltzer, C.C., Price, J.C., Sembrat, R.C., et al. (2002). Human regional cerebral glucose metabolism during non-rapid eye movement sleep in relation to waking. Brain, 125(part 5), 1105-1115.
Nofzinger, E.A., Buysse, D.J., Germain, A., Price, J.C., Miewald, J.M., Kupfer, D.J. (2004). Functional neurimaging evidence for hyperarousal in insomnia. American Journal of Psychiatry, 161: 2126-2129.
Nofzinger, E.A., Mintun, M.A., Wiseman, M., Kupfer, D.J., Moore, R.Y. (1997) Forebrain activation in REM sleep: an FDG PET study. Brain Research, 770, 192-201.
Nofzinger, E.A., Nowell, P.D., Buysee, D.J., Vasco, R.C., Thase, M.E., Frank, E., Kupfer, D.J., Reynolds, C.F. (1999). Towards a neurobiology of sleep disturbance in primary insomnia and depression: a comparison of subjective, visually scored, period amplitude, and power spectral densisity sleep measures. Sleep, 22 (supp. 1), S99.
Othmer, S., Othmer, S.F., Kaiser, D.A. (1999). EEG biofeedback: an emerging model for its global efficiency. In: Evans, J.R., Abarbanel, A., editors. Introduction to quantitive EEG and neurofeedback. San diego: Academic press, 3-27.
Perlis, M.L., Merica, H., Smith, M.T., Giles, D.E., (2001). Beta EEG activity and insomnia. Sleep Medicine Reviews, 5 (5), 365-376.
Perlis, M.L., Smith, M.T., Andrews, P.J., Orff, H., Giles, D.E. (2001). Beta/gamma EEG activity in patients with primary and secondary insomnia and good sleeper controls. Sleep, 24, 10-7.
36 Riemann, D., Spiegelhalder, K., Feige, B., Voderholzer, U., Berger, M., Perlis, M.,
Nissen, C. (2010). The hyperarousal model of insomnia: a review of the concept and its evidence. Sleep Medicine Reviews, 14, 19-31.
Ringdahl, E.N., Pereira, S.L., Delzell, E. (2004). Treatment of primary insomnia. JABFM, 17(3), 212-219.
Salin-Pascual, R.J., Roehrs, T.A., Merlotti, L.A., Zorick, F., Roth, T. (1992). Long-term study of the sleep of insomnia patients with sleep state misperception and other insomnia patients. American Journal of Psychiatry, 149, 904-908.
Saper., C.B., Scammell, T.E., Lu, J. (2005). Hypothalamic regulation of sleep and circadian rhythms. Nature, 437, 1257-1263.
Sherin, J.E., Elmguist, J.K., Torrealba, F., Saper, C.B., (1998). Innervation of
histaminergic tuberomammillary neurons by GABAergic and galaninergic neurons in the verntrolateral preoptic nucleus of the rat. Journal of Neuroscience, 18(12), 4705-21.
Staner, L., Cornette, F., Maurice, D., Viardot, G., Le Bon, O., Haba, J., et al (2003). Sleep microstructure around sleep onset differentiates major depressive insomnia from primary insomni. Journal of Sleep Research, 12, 319-30.
Sterman, M.B., Howe, R.C., MacDonald, L.R. (1970). Facilitation of spindle-burst sleep by conditioning of electroencephalographic activity while awake. Science, 167, 1146-8.
Szabadi, E. (2006). Drugs for sleep disorders: mechanism and therapeutic prospects.
Brain Journal of Clinical Pharmacology, 61, 761-6.
Thomas M, Sing H, Belenky G, Holcomb H, Mayberg H, Dannals R, Wagner H, Thorne, D., Popp, K., Rowland, L., Welsh, A., Balwinski, S., Redmond, D. (2000). Neural basis of alertness and cognitive performance impairments during sleepiness. 1. Effects of 24 h of sleep deprivation on waking human regional brain activity. Journal of Sleep Research, 9(4), 335-52.
Von Economo, C., (1930). Sleep as a problem of localization. Journal of Nervous and Mental Disease, 71, 249-259.
Wicklow, A., Epsie, C.A., (2000). Intrusive thoughts and their relationship to actigraphic measurement of sleep: towards a cognitive model of insomnia. Behavioural Research and Therapy, 38, 679-93.
