Abstract
Purpose of review: Sleep and the chronobiological disease cluster headache are believed to be interconnected. Despite efforts, the precise nature of the relationship remains obscured. A better understanding of this relation may lead to more effective therapeutic regimes for patients suffering from this debilitating disease. This review aims to evaluate the existing literature on the subject of cluster headache and sleep.
Latest findings: Several previous studies describe an association between episodic cluster headache and distinct macrostructural sleep phases. This association was not confirmed in a recent study of seven episodic cluster headache patients, but it was suggested that further studies into the correlation between cluster headache attacks and the microstructure of sleep are relevant. The connection between cluster headache and the hypocretins is currently under investigation.
Summary: There is evidence in favour of an association between episodic cluster headache and REM sleep whereas no such relation to chronic cluster headache has been reported. Particular features in the microstructure of sleep and arousal mechanisms could play a role in the pathogenesis of cluster headache. Reports indicate that cluster headache and obstructive sleep apnoea are associated. Single cases show improvement upon treatment of sleep apnoea, but the causal relationship remains in question.
Introduction
A complex relationship exists between the chronobiological disease cluster headache (CH) and sleep. There is some evidence that REM sleep in particular may precipitate CH attacks in certain patients where the attacks occur at specific times and where clusters of these occur in specific circannual rhythms (1–5). The hypothalamus is highly involved in the generation of circannual and circadian rhythms and neuroimaging shows hypothalamic activation during CH attacks (6–8).
The relationship between CH and sleep is intriguing yet complex, and studying sleep and phenomena related hereto is notoriously difficult. CH is also a difficult disease to study, in part due to its low prevalence but mainly due to the very severe but periodic attacks with agitation. Furthermore, conducting polysomnography (PSG) recordings during the active period of the disease has been associated with some difficulty, and the present literature is based mainly on older cases or series of a restricted number of patients. However, investigating headache and sleep disorders has the advantage that in both cases thorough, international classification systems exist and new technology is making it easier to obtain and analyse data from sleeping patients.
Sleep has properties that seem to make it a precipitating factor for headache. The International Classification of Sleep Disorders classifies hypnic headache, cluster headache, chronic paroxysmal hemicrania and migraine as so-called sleep-related headaches (9). A high prevalence of nocturnal sleep-related attacks is reported in patients suffering from both chronic cluster headache (CCH) and episodic cluster headache (ECH) (10). Current knowledge about the function of the hypothalamus in both CH and sleep regulation suggests an unknown, underlying pathway for trigeminal pain perception and sleep induction. Against this background we aimed to evaluate the existing literature on the subject of CH and sleep.
About CH
CH is a primary headache syndrome belonging to the trigeminal autonomic cephalalgias (TACs). It is characterised by strictly unilateral pain with accompanying autonomic symptoms. Of the TACs, CH is the most prevalent and occurs in about 0.1% of the population with a male to female ratio of around 4:1 (11,12). A population-based study found a peak in the incidence for males at 40–49 years and for women at 60–69 (13). Dramatic descriptions of the pain severity in CH are abundant in literature. CH exists in a chronic and episodic form with the latter being the most prevalent (80–90%) (14). The pain is located unilaterally, primarily around the orbit of the eye, but may also affect the temporal region of the head. Attacks last from 15 to 180 minutes. A distinguishing feature of CH is the circannual and especially the circadian rhythm with which the clusters and individual attacks occur. The attacks may occur at all times of the day, although there is a close association with sleep. The classical clinical picture shows the cluster attack beginning 90 minutes after sleep onset, typically coinciding with the first REM phase (5).
Sleep and headache: proposed classification of the relation between sleep and headache. Not exhaustive, modified after Paiva and Herin-Hanit (17)
PLMS: periodic limb movement during sleep. Question mark indicates that the association is still subject to investigation.
Subjectively and objectively the structure of sleep is changed in CH patients (18,19). During bouts the patients complain of reduced total duration of nocturnal sleep, increased time to fall asleep, high frequency of nocturnal awakenings and poor sleep quality (18,19). Objectively there is a reduced total duration of sleep, increased time to fall asleep, increased latent periods of all sleep stages, increased percentage representation of superficial sleep stages (1 and 2), reduced durations of REM stage, increased number of awakening episodes during the night and increased movement activity (18). Transient recurrent situational insomnia during the active bout has been described in a single subject (19). This is a common finding, so it may be coincidental.
CH attacks show periodicity
Overview of published results on CH and sleep. Grey rows indicate studies in SDB
In Italy, Manzoni et al. found that there was cyclicity but that it was not necessarily related to the months of the year (3). Focusing on the daily occurrence, Manzoni found that onsets of the attacks peaked around 01.00 and 02.00 and 13.00, 15.00 and 21.00 hours (13.00 and 15.00 hours incidentally being the time when most of the patients stopped working). Some of the first papers published on CH also note that CH exhibits a relationship with the sleep–wake cycle and there is evidence that attacks are influenced by activity and relaxation (20–22). In a study of 77 spontaneous attacks Russell found that 75% of these began between 21.00 and 10.00 hours. Of the daytime attacks, 71% occurred when the patients were physically relaxed (a few while napping) (4).
Lithium has been shown to be efficacious in the prophylactic treatment of some, but not all, CH patients (23). Interestingly lithium is also effective in treating other syndromes that exhibit chronobiological characteristics such as hypnic headache, bipolar disorder and perhaps in some cases of migraine (24). Lithium as a drug shows high concentrations in the hypothalamus (25) and it affects serotonergic systems in the CNS (26). The observable effects on sleep include a reduction in REM sleep and alteration of the circadian rhythm (27). The observed therapeutic effect of lithium may be mediated partly by an increase in the concentration of melatonin. In rodent animal studies, lithium enhances the absorption of tryptophan and promotes its transformation to serotonin (26), which both serve as precursors in the synthesis of melatonin.
Is CH associated with REM sleep?
It is not unusual for CH patients, awakened at night by an attack, to be able to vividly recall their dreams. In concordance with this observation, these attacks have been shown to occur in association with REM sleep, the sleep stage where dreams occur. In one CH patient in active bout PSG showed altered sleep architecture, including prolonged REM latency and increased REM fragmentation (1). Early studies of sleep and CH concluded that attacks in ECH in particular are related to REM sleep and that apparently there is no association between CCH and REM sleep (28) (Table 2). It is reported in some cases that sleep deprivation can temporarily prevent or delay the onset of CH (29). It is noteworthy that this chronotherapy also shows efficacy in the treatment of periodic depression (30), but the relation is otherwise unclear.
Some of the first PSG recordings from CH patients were made more than 40 years ago by Dexter and Weitzman (5) (Table 2). They showed that in three patients with ECH all nine of the recorded nocturnal attacks occurred during REM sleep or within 9 minutes of this sleep stage terminating. Further strength is added to this observation by the finding that many attacks occur between 04.00 and 07.00, the time of night where REM sleep is most abundant (4). In one series of recorded attacks almost 60% of CH attacks followed REM sleep, even though REM sleep only comprised 20% of total sleeping time (31).
It has been proposed (32,33) that the switching between different states of relative arousal may precipitate the attack, specifically the transfer of dominance of the parasympathetic cholinergic system (mediators of REM sleep) to a primary sympathetic serotonergic system (mediators of non-REM sleep) (32). In the acute cluster, there may be a condition of hypoarousability during REM sleep (1). This was shown in one patient by the use of actigraphy (a method for measuring gross motor activity). Hypoarousability is a possible factor in the pathogenesis of migraine (34,35) but the role in CH has not been investigated. The previously mentioned activation of the posterior hypothalamus during CH attacks is especially interesting in this context, because this CNS structure plays a significant role in arousal mechanisms (36).
It is noteworthy that one recent study of limited size (seven patients, of whom four had attacks during recording), failed to find any association between ECH attacks and REM sleep (37). Consequently, there is still debate on the association of CH attacks and REM sleep. As proposed by the aforementioned study it seems logical to focus future studies of sleep not only on the macrostructure of sleep but also the microstructure.
CH and the hypothalamus
There is substantial evidence suggesting that the hypothalamus plays a part in the pathogenesis of CH. The hypothalamus is arguably the most important structure in maintaining body homeostasis. This is accomplished by regulating the secretion of several hormones in a cyclic fashion. It is well established that in most but not all CH patients, hormones secreted in a cyclic manner are affected (38). These include melatonin, cortisol, testosterone, luteinising hormone, follicle-stimulating hormone, prolactin, growth hormone, thyrotropin and beta-lipotropin. Some of these changes occur only during a cluster period, returning to normal once the patient is no longer in active bout; others remain outside normal limits even when the patient is in remission (38). This is suggestive of hypothalamic dysfunction. Furthermore, studies in functional neuroimaging (positron emission tomography and functional MR) report neuronal activation in the ipsilateral posterior inferior hypothalamic gray during the pain state (6–8). Neurons located specifically in this area have been found to secrete the neuropeptides hypocretin-1 and hypocretin-2.
The hypocretins are relatively newly discovered neuropeptides involved in the regulation of sleep and arousal, amongst other things (39,40). Neurons secreting hypocretin are located almost exclusively in the posterior, inferior hypothalamus and project widely to CNS structures involved in pain processing, sleep and arousal (39–42). In two samples a polymorphism in the hypocretin-2 receptor (1246 G > A) was found to be associated with CH (43–45), although this did not seem to affect treatment response (46). The result on the molecular structure of the receptor could interfere with its dimerisation, thus affecting normal function (43). One northern European study (47) did not find this association, suggesting that genetic factors may be heterogeneous in different populations. Interestingly, the pathogenesis of narcolepsy-cataplexy revolves around a deficiency in the hypocretin system (48) and sufferers exhibit reduced autonomic response during sleep (49) (decreased heart rate increase during periodic limb movement during sleep). At present it seems that there are at least two forms of narcolepsy-cataplexy: one form where the hypocretin levels in cerebrospinal fluid are low or unmeasurable and another form where the levels are normal. In the latter form it is speculated that the receptor might be dysfunctional, possibly as a result of a polymorphism. A recent study of the hypocretin concentration in the cerebrospinal fluid of ECH patients did not report abnormal findings (50).
A well-known substance regulating sleep and circadian rhythms is melatonin, which is secreted by the pineal gland receiving input from the suprachiasmatic nucleus of the hypothalamus. This secretion is inhibited by exposure to light. ECH patients exhibit a delay in peak melatonin during cluster periods, blunted nocturnal melatonin responses and reduced 24-hour melatonin production in active bouts (51). In a small study administering melatonin proved effective in the prophylactic treatment of CH (52), but larger confirmatory studies are lacking.
In a review, Dodick et al. (53) assess the possible ways in which melatonin could play a role in the pathogenesis of CH: the physiological effect of melatonin is in part mediated by a short-term potentiation of the inhibitory effect of GABA, resulting in a lower threshold for pain-sensing neurons normally inhibited by this neurotransmitter. On a cellular level melatonin regulates Ca influx and it may alter the tone or vasoreactivity of cerebral blood vessels by modulating 5-HT2 receptors on these vessels. Drugs that are antagonistic in action on this receptor are used to prevent both CH and migraine (flunarizine, cyproheptadine, methysergide). Melatonin also inhibits synthesis of prostaglandin E2 which mediates sterile perivascular inflammation and activates trigeminovascular nociceptive afferents (53).
Deep brain stimulation affects sleep in CH patients
A small study on the effect on sleep of deep brain stimulation (DBS) of the posterior hypothalamus, a fairly novel therapy for treating refractory CCH, has been conducted recently (54). The study was of very limited size (three patients) but the results do seem to confirm existing theories. In patients receiving DBS nocturnal CH attacks disappeared, and sleep architecture and sleep quality were improved. PSG showed increased total sleep time, sleep efficiency, slow-wave sleep stages and decreased PSG indexes of fragmented sleep. The authors also found that DBS decreased periodic limb movement during sleep (PLMS), a condition often associated with arousals. They state that this cannot be ascribed solely to decreased light sleep stages (where PLMS is more frequent) and it suggests that the areas affected by DBS also affect the relative state of arousal. Stimulation of the subthalamic nucleus in Parkinson’s disease has not been reported to curtail PLMS, rather it seems to provoke restless legs syndrome (55,56).
These findings support the known fact that the posterior hypothalamus plays an important part in regulating sleep–wake states and relative arousal. Furthermore it is important in pain modulation, as evidenced by its connections to certain brainstem regions (dorsal raphe nuclei, the locus coeruleus, periaqueductal gray). A H215O PET study performed in CCH patients with implanted DBS devices showed increased blood flow in the ipsilateral posterior inferior hypothalamic gray (site of stimulator tip) and the ipsilateral trigeminal system in the brainstem (57). It is important to note that activation of the trigeminal system in this case did not provoke pain or accompanying autonomic symptoms.
CH and sleep disordered breathing (SDB)
The association between CH and SDB has been studied for several decades but many questions remain unanswered. The most common form of SDB is obstructive sleep apnoea (OSA). In one study approximately 4% of middle-aged males were estimated to have OSA with associated daytime hypersomnolence (58), and up to 24% of males may have sleep apnoea, defined as AHI ≥ 5 (58). Although sleep apnoea seems to be a very common finding there may be a positive association with CH as the prevalence in these patients may be as high as 6 in 10 (31). Another study concluded that CH patients are 8.4 times more likely to exhibit OSA than normal individuals (59).
The tendency for attacks of CH to occur during the night has led to speculation that the changes invoked by SDB, in particular OSA, might somehow trigger the cluster attack. OSA leads to recurrent hypoxemia, hypercapnia, excessive negative intrathoracic pressure and increased intracranial pressure, amongst other findings. It may also be accompanied by abrupt changes in sympathetic tone. All of these changes could potentially serve as the trigger of CH attacks. Further strengthening the association between the two disorders is the fact that OSA severity is higher during REM sleep (60), the sleep stage that seems to be associated with an increased likelihood of CH attacks.
More than two decades ago Kudrow et al. suggested hypoxemia to be a potential trigger of CH attacks (31). Hypoxemia has been observed prior to CH attacks in wakefulness and if induced by nitroglycerine it can trigger attacks if the patient is in an active bout (61). The fact that inhaling supplementary oxygen is a very effective treatment (through an unknown mechanism) for an acute attack also lends support to this theory. However, in one study induced hypoxemia failed to provoke attacks at all (62). It seems therefore that there is conflicting evidence. There is now very effective treatment available for OSA, and the accompanying hypoxemia, in the form of nasal CPAP. A few cases exist where effective treatment of OSA resulted in reduced severity and frequency of otherwise refractory CH attacks (63,64) but larger controlled series are lacking.
One could speculate that there may be an overlap between the underlying factors that lead to OSA and CH. However, it seems clear that in some patients OSA can affect the severity of CH, as evidenced by the case reports mentioned above, but the actual causal relationship has not been clarified. In recent publications Graff-Radford et al. theorise that CH and OSA are not in a causal relationship but are epiphenomena both generated in the hypothalamus (65,66). This theory is based on two findings: activation of temperature-sensitive neurons in the pre-optic/anterior hypothalamus in cats leads to suppression of airway dilator and diaphragmatic muscle activity during non-Rapid Eye Movement (NREM) sleep (67) and during CH attacks there is activation of the posterior hypothalamus (6). Thus the two disorders both seem to have their pathophysiology related to the hypothalamus, but to different parts of this structure. The pathology of the sleep disorder may also be related to other regions altogether (e.g. the brainstem, thalamus and/or basal forebrain). In conclusion it must be said that the connection between CH and OSA remains elusive.
Conclusion
In conclusion the prospect of uncovering the mechanism behind CH attacks occurring during sleep could provide clues to the underlying pathology. This could bring potentially more effective treatment, and also offer a chance of better understanding sleep- and wake-promoting mechanisms. A role where the hypothalamus serves as a gating or switching mechanism has been suggested in the primary headaches (68). A disorder of the hypothalamic system could result in destabilisation of pro- and anti-nociceptive input and the activation of the trigeminal autonomic reflex also affecting sleep. It has been suggested that the gradual progression from ECH to CCH that occurs in some patients may be associated with such a change, resulting in a loss of association with REM sleep. In order to get a clearer understanding of the causality between changes in sleep patterns and CH a sleep study of the time leading up to the cluster period could offer clues about this relationship.
The posterior hypothalamus plays a significant role in arousal mechanisms, because it projects to the cerebral cortex. Its stimulation results in cortical activation, motor activity and sympathetic responses (the last two accompanying symptoms in CH). Impaired arousal in CH has been considered (19). However, the connection between this consideration of arousal mechanisms and the chronobiological nature of CH, including the interaction with sleep, still offers much to be elucidated. A possible association with the hypocretin system is interesting. The 1246 G > A polymorphism in CH could suggest involvement of a central mechanism for arousal, the regulation of sleep and pain processing.
The true role of the hypothalamus in CH is still unknown. There is a possibility that the symptoms and findings pointing towards its involvement are simply epiphenomena. Many questions remain unanswered. How are attacks, both in ECH and CCH, related to the structure of sleep? Is there a role for hypocretin, considering its involvement in pain sensing and arousal processes? Research into this area is still very much warranted.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest
The authors declare that there is no conflict of interest.