Abstract
The hypnic headache syndrome is a rare and recurrent primary headache disorder, first described by Raskin in 1988 (1). Headaches consistently occur each night, usually between 01.00 h and 03.00 h, waking the patient from sleep. Individual headache duration varies within individuals, but is usually from 1 to 2 h. The frequency of headaches in this syndrome is high, with at least half of hypnic headache sufferers experiencing daily attacks (2). The International Classification of Headache Disorders, 2nd edition (ICHD-2) has well established diagnostic criteria for hypnic headache (3). In contradistinction to what the criteria state, in both Evers' and Dodick's series, approximately 40% of patients experience pulsatile rather than dull pain. Furthermore, patients <50 years old were seen in those series, the youngest being 36 years old (4, 5). Not uncommonly, patients carry a prior medical history of other primary headache disorders (5). Multiple pharmacological interventions have been tried in hypnic headache in an attempt to prevent frequent nocturnal attacks, with variable success rates. With the exception of melatonin, hypnotics have not been previously reported to be effective for this disorder.
Case report
A 53-year-old man presented with a 7-year history of strictly nocturnal headaches. His usual bedtime was around 22.30–23.00 h. Every night, without exception, a headache woke him up between 01.30 and 02.30 h. The headache was described as diffuse, pulsatile, and with an intensity of 6/10. Nausea, vomiting, photophobia, phonophobia, osmophobia, dysautonomic features and focal neurological deficits had never been associated with his headaches. No vivid dreams were reported to precede the headaches. Sometimes he was able to go back to sleep at around 04.30 h and when awakening later the headache was absent. Sometimes he was not able to go back to sleep and the headache would continue until he got out of bed and started his daily activities early in the morning. He found that if he took Unisom (doxylamine succinate 25 mg), Tylenol PM (paracetamol 500 mg/diphenhydramine HCl 25 mg), or Lunesta (eszopiclone 2 mg) at bedtime he was able to sleep through the night without being awoken by the nocturnal headache. In an attempt to know whether his headaches had completely disappeared, on multiple occasions he did not take a sedative. Without exception, on those nights his usual nocturnal headache recurred. He had been on these medications on a nightly basis for years successfully preventing his headaches. His past medical history was significant for infrequent tension-type headaches and preorgasmic headaches. General and neurological examinations were normal. Cerebral magnetic resonance imaging and angiography (MRI/MRA) with and without contrast were normal.
Discussion
Hypnic headache is a rare and recurrent primary headache disorder. To date, only two cases of probable secondary hypnic headache have been reported. One was associated with a posterior fossa meningioma and the other with intracranial hypotension (2). The MRI/MRA of this patient failed to reveal structural pathology. Hyperelastic joints and skin, pachymeningeal gadolinium enhancement and ‘brain sag’ on MRI, all of which could suggest intracranial hypotension, were absent. The pathophysiology of hypnic headache is not entirely understood. However, two changes in sleep physiology might be implicated. As individuals age, the cell numbers in the suprachiasmatic nucleus (SCN) decrease significantly, leading to a decrease in melatonin, whose synchronized release is governed by the SCN (6). This argument is supported by the fact that lithium, the most effective treatment for hypnic headache, indirectly increases the levels of melatonin (4). The other mechanism involves rapid eye movement (REM) sleep. In polysomnography studies, hypnic headache attacks have the reliable tendency to start during REM sleep, a period in sleep architecture where dorsal raphe and locus cerulei nuclei activity is absent. These structures are part of the brain antinociceptive system and their dysfunction during REM sleep could lead to hypnic headache (4).
Multiple medications with different mechanisms of action have been used with variable success rates in hypnic headache therapy. Although lithium seems to be the most effective, its frequent side-effects often limit its use in the elderly, who are the usual patients. Among other therapeutic options that have been reported to be effective in some cases are indomethacin, caffeine and melatonin (4). Topiramate and pregabalin have recently been reported to be successful in single case reports (7, 8). Because of lack of awareness of his diagnosis, none of these medications was tried by our patient prior to hypnotics. Instead, he was able to prevent completely, without exception, each of his nocturnal attacks by taking a weak hypnotic at bedtime. The effectiveness persisted for 7 years and he experienced nocturnal attacks only when he did not take the hypnotic.
Neither paracetamol, diphenhydramine or doxylamine have known ties with melatonin physiology. Single doses of ezopiclone, if at all, decrease rather than increase levels of melatonin (9), making this mechanism an unlikely explanation of its benefit in hypnic headache. On the other hand, ezopiclone is known to decrease total REM time and REM density (10–12). Further, antihistamines that antagonize H1 receptors such as diphenhydramine and doxylamine are known to decrease the rate of REM sleep (13). By means of decreasing the density, rate and/or total time of REM sleep, the hypnotics used by this patient may have prevented hypnic headache attacks. This remains speculative at this time. Further mechanisms may be implicated. No polysomnogram was performed in our case. Nevertheless, the response of this patient to the sedatives here discussed was impressive, durable and 100% predictable. Other hypnic headache sufferers may benefit from prophylactic nocturnal doses of weak hypnotics, particularly those individuals with frequent or daily attacks.