Abstract
Hypnic headache (HH), which was first described by Raskin in 1988 (1), is a rare sleep-associated headache syndrome affecting patients aged ≥ 50 years (2). To date, only approximately 80 cases of HH have been reported in the literature, most of which have recently been presented in a detailed review by Evers et al. (2). Due to its association with sleep and stereotyped occurrence, HH has also been termed alarm clock headache (3). HH occurrs only during sleep and usually leads to awakening 2–4 h after falling asleep. In most cases, the dull headache presents without accompanying autonomic disturbances and lasts for 1–2 h. About two-thirds of patients report a bilateral, holocephalic headache. There is no known association with other diseases. Diagnostic criteria for HH have been established, and HH is included in the recent International Headache Society (IHS) classification as a primary form of headache (IHS classification 4.5) (4). Pain in HH is usually moderate, but may be severe and, due to disruption of sleep, quality of life may be severely impaired. First-line treatments include lithium carbonate and indomethacin (1–3, 5).
With HH being a rare form of headache, it must be differentiated from other, more common sleep-related headache syndromes (such as migraine, cluster headache, hemicrania continua or chronic paroxysmal hemicrania) and from other sleep disorders associated with the occurrence of headache (e.g. headache related to obstrucitve sleep apnoea, to nocturnal arterial hypertension or neck pain due to nocturnal immobility in Parkinson's disease). Due to distinct therapeutic consequences, establishing a definite diagnosis is crucial. The broad spectrum of more common forms of sleep-related headache syndromes as possible differential diagnoses to HH illustrates the value of performing polysomnography (PSG) in cases of nocturnal headache syndromes associated with disruption of sleep. To date, only a few cases of HH including PSG findings have been reported (6–10). We here present a new case of HH including PSG findings.
Case report
A 58-year-old male patient had a 6-year history of idiopathic Parkinson's disease which was treated by standard dopaminergic medication (levodopa, dopamin-agonists, amantadine). For the last 9 months the patient had complained of a new onset sleep disorder with regular awakening during the night. The reason for waking up was a headache regularly occurring nearly every night about 2–3 h after falling asleep. Headache was bilateral and holocephalic, especially affecting the forehead. The pain was dull, non-throbbing and of severe intensity (8/10 on a visual analogue scale from 0 to 10), making the patient leave the bed and walk around. There was neither photo- nor phonophobia, nor nausea or vomiting. The patient did not complain of visual disturbances or other neurological symptoms. The headache regularly lasted for about 1 h, resolving spontaneously. The patient was then able to fall asleep again. Analgesics such as aspirin did not affect the headache. There was no known history of arterial hypertension or snoring. During the day, there was no headache. There had not been a previous history of headache. The patient did not complain of sleep disturbances typical of Parkinson's disease, such as daytime sleepiness or sleep attacks.
Clinical examination revealed the known Parkinson syndrome without other pathological neurological signs. There was no glaucoma or papilloedema. Cranial magnetic resonance imaging (MRI), nerve conduction studies, electromyogram (EMG) and cerebrospinal fluid (CSF) examination were unremarkable. Polysomnography (PSG, Brainlab; Schwarzer, Munich, Germany), including electroencephalogram (EEG), electro-oculogram (EOG), electrocardiogram (ECG), EMG as well as monitoring of blood oxygen saturation, arterial blood pressure and breathing parameters, was performed between approximately 22.30 h and 06.00 h. About 4 h after falling asleep, the patient woke up at about 03.00 h, following the second rapid eye movement (REM) sleep-period due to severe headache as described above (Fig. 1). As usual, headache lasted for about 1 h and resolved spontaneously. The patient then fell asleep again, quickly entering REM sleep after only approximately 20 min. There was no second awakening until the end of the PSG recording. Overall, there was a reduced sleep efficiency of only 59%. At no time during the recording was there an altered breathing pattern with hypopnoea or apnoea, nor were there periodic limb movements. There was no decrease of oxygen saturation and arterial blood pressure was normal at all times.
Polysomnography illustrating the occurrence of headache during the second rapid eye movement (REM) sleep period, approximately 4 h after falling asleep (03.00 h). Headache was of severe intensity, led to awakening for about 1 h and resolved spontaneously. Following hypnic headache, REM sleep onset latency was remarkably short.
Treatment with a single dose of 50 mg indomethacin led to complete remission of the headache during the following nights. However, during the further course of treatment the patient experienced side-effects of the medication consisting of peripheral oedema. Thus, treatment with indomethacin was discontinued, leading to immediate re-occurrence of nocturnal headache. Treatment with a reduced dosage of indomethacin finally led to a sufficient treatment response without adverse effects. Due to a prompt response to indomethacin, treatment with lithium had not been performed.
Discussion
We here present a 58-year-old male patient with known Parkinson's disease who developed a new HH promptly responding to treatment with indomethacin. PSG enabled the documentation of awakening due to a REM sleep-associated headache. There were no pathological findings suggesting other sleep disorders. As the present case illustrates, intensity of the headache may be severe, leading to marked reduction of quality of life, making effective treatment necessary. As common forms of analgesic medication used for other forms of headache are not effective in HH, differentiation of HH from other sleep-related headache syndromes and sleep disorders associated with headache is important.
Although the individual medical/headache history and the onset of headaches at an advanced age may be suggestive of HH, establishing the diagnosis may be difficult. As illustrated in our case, PSG is a valuable tool for the diagnosis of HH, allowing the exclusion of other sleep disorders possibly associated with nocturnal headache. Especially in our case, PSG allowed us to differentiate HH from sleep disturbances associated with Parkinson's disease. To date, only a few cases of HH have been reported including PSG findings (6–10). As in our case, HH was frequently associated with REM sleep, although HH not associated with REM sleep has also been reported (6, 7). Association of HH with REM sleep has led to possible pathophysiological concepts of HH, especially the involvement of the dorsal raphe and locus coeruleus nuclei, which are known to be inactivated during REM sleep and, together with the periaqueductal grey matter, play an important role in antinociception (10, 11). In most cases, occurrence of headache was associated with the first REM sleep period. In the present case, onset of the headache followed the second REM sleep period, illustrating the variability of the occurrence of HH. Interestingly, the REM sleep onset latency after falling asleep after the HH in the night was remarkably short, indicating a disturbed sleep pattern. In our case, PSG did not reveal altered breathing patterns, oxygen desaturation or elevated arterial blood pressure. The absence of sleep-disordered breathing (SDB) in the present case supports the notion that SDB is not a major pathomechanism in HH.
In conclusion, we present a new case of hypnic headache for which polysomnography was performed. The present case extends the small number of reported HH cases including PSG findings and illustrates the value of PSG as a diagnostic tool to differentiate this rare sleep-related headache syndrome from other nocturnal forms of headache, leading to an effective treatment.