Abstract
Background: Hypnic headache (HH) is a rare primary headache that occurs exclusively during sleep. Drug therapy of HH is often associated with side effects. In a few cases, antiepileptic drugs have been useful in preventing HH attacks.
Cases: We describe two patients, a man aged 57 and a woman aged 63, both with a history of duodenal ulcer and both suffering from several months of frequent and moderate headaches (at least 15 days/month), which occurred only at night and resisted treatment. Physical examination and imaging were unremarkable. Sleep polysomnography was normal in the man. In both patients, the diagnosis of HH was made on the basis of the International Classification of Headache Disorders, 2nd edition criteria. Given the prior history of duodenal ulcer and the unavailability of some drugs in Morocco, we opted for the administration of moderate doses of lamotrigine. The outcome was favourable at one year follow-up and beyond.
Conclusion: To our knowledge, the response to lamotrigine in HH cases has never been reported. Because of its safety profile and availability, lamotrigine can be an effective alternative treatment in HH.
Introduction
Hypnic headache (HH) is a primary headache disorder, first described by Raskin in 1988 (1) as a sleep-related headache, usually manifesting after the age of 50. The pain is diffuse, waking sufferers at the same time each night. HH has been classified in section 4.5 of the second edition of the International Classification of Headache Disorders (ICHD-II), under ‘Hypnic headache’ (2).
Lithium is the most frequently used drug in the treatment of HH, showing the highest efficacy. Indometacin, amitriptyline, and caffeine have also been found to be effective (3). In several studies, antiepileptic drugs, such as topiramate, gabapentin, and pregabalin, have been found helpful (4).
ICHD-II diagnostic criteria for hypnic headache
ICHD-II: International Classification of Headache Disorders, 2nd edition.
Case studies
Case 1
A 57-year-old man with a history of duodenal ulcer, treated in 2000, and well-controlled diabetes mellitus diagnosed 3 years previously, developed headaches in April 2009. These occurred exclusively during sleep and woke the patient up systematically for 3 months. The pain was localized bilaterally to the frontal area, was moderate in intensity, and usually lasted for about 60 min. The attacks occurred twice nightly on average, for at least 20 nights per month, and resisted repeated doses of paracetamol. The attacks were often accompanied by nausea and a pronounced need to drink or read, but were not accompanied by other autonomic symptoms. No motor agitation was reported by either the patient or his wife. There was no snoring or apnoea during sleep.
Physical examination was unremarkable and there were no clinical signs of peripheral neuropathy or restless legs syndrome. Laboratory tests [glucose levels and glycated haemoglobin (HbA1c); serum iron; ferritin, full blood count; kidney and thyroid function; erythrocyte sedimentation rate] were normal. A brain MRI and 24-h ambulatory blood pressure monitoring (ABPM) were unremarkable. A polysomnography (PSG) study did not reveal sleep apnoea syndrome, snoring, or periodic leg movements. Sleep architecture was preserved and a headache crisis occurred during stage II of the second sleep cycle without oxygen desaturation.
Given the history of duodenal ulcer, the low safety margin of lithium, and the unavailability of melatonin and topiramate in Morocco, we opted for increasing doses of lamotrigine: 25 mg/day for 15 days, increased to 50 mg/day for 2 months. The outcome was favourable, with the complete disappearance of the headaches after 3 weeks of treatment. We extended the treatment for another 6 months before gradually tapering it off. The patient remained asymptomatic during the 18-month follow-up.
Case 2
A woman aged 63 years and with a history of duodenal ulcer treated in 1995 was referred to our department in December 2009 because of migraine headache resistant to treatment. The attacks occurred every night and beta blockers were given for 2 months with no response, after that amitriptylin were given for 2 months with no good response. The headaches were diffused and irritating, occurring only during sleep. They happened almost every night and woke the patient up. They lasted between 30 min and 1 hour and were not associated with autonomic symptoms. The patient had to get out of bed, drink, and watch television until the crisis was over. No motor restlessness occurred. They had an occasional cup of coffee which relieved the headache but made it difficult for them to get back to sleep and caused fatigue in the morning.
The clinical examination was normal. All laboratory tests were normal. A brain MRI and 24-h ABPM were unremarkable. The diagnosis of HH was made.
As in case study 1, we opted for lamotrigine in increasing doses: 25 mg/day for 15 days, increased to 50 mg/day for 2 months. The persistence of some weekly, though moderate, headaches led us to increase the dose to 75 mg/day. The patient reported complete disappearance of HH in the second week. We extended the treatment for 6 months before gradually tapering it off. The patient is still asymptomatic after the first year of follow-up.
Discussion
We have described two patients suffering from HH for a period of several months. This was a dull headache that manifested itself only during sleep, waking each patient up. It occurred more than15 times a month and lasted for longer than 15 min after waking. The headache was accompanied by nausea, but not by autonomic symptoms. Despite these symptoms fitting the criteria of the ICHD-II, motor behaviours during each headache episode were observed in both patients (getting up, eating, drinking coffee, reading, watching television). These motor activities have been reported in several HH studies (6,7,8). Other clinical features associated with HH include: trigemino–autonomic symptoms (7), migraine-like headaches (pulsating, unilateral, stabbing), and symptoms onset before the age of 50 (8).
The pathophysiology of HH remains unclear; an anatomical remodelling of the hypothalamus may be responsible for HH. A voxel-based morphometry study has shown grey matter volume loss in the posterior hypothalamus (9).
Another study has suggested a relationship between HH and rapid eye movement (REM) sleep. This has been supported by the description of vivid dreams by many patients before being woken by an HH attack and confirmed by PSG studies, showing that about 69% of HH attacks occur during REM sleep (9,10). However, recent studies have suggested that attacks might also arise from non-REM sleep (11,12). After recording 22 HH attacks during PSG in six patients, Holle et al. (13) showed that different sleep stages can precede HH and that HH attacks are in fact not dependent on REM sleep.
Sleep-disordered breathing has also been proposed as a possible trigger factor for HH; obstructive respiratory events leading to a decrease in nocturnal oxygen saturation were monitored in several PSG studies (10,12). However, Holle et al. (13) did not detect any temporal correlation between headache onset and oxygen desaturation in any patient. In our first case study, HH occurred during stage II of the second sleep cycle without oxygen desaturation.
Alterations of trigeminal processing with regard to central facilitation or sensitization are commonly observed in various headache disorders. However, the nociceptive blink reflex and trigeminal pain-related evoked potentials in patients with HH do not support this hypothesis with regard to the pathophysiology of HH (14).
Because of the very low prevalence of HH in the population, effective treatment options are yet to become widely available. For the acute phase, non-steroidal anti-inflammatory drugs (NSAIDs) and paracetamol are often taken, but are not always effective. Triptans have been shown to be moderately effective. Oxygen inhalation, metamizole and opiates are not effective. However, one study has shown caffeine as being effective in 78% of cases (7). The most commonly used prophylactic treatment is lithium; it is tried most frequently by patients and has shown response rates of up to 88% (9,12). This data contradicts a recent report that shows a success rate of only 36%; one possible explanation is that more than half of the patients stop lithium treatment because of its side effects (7). Dodick (10) previously reported that melatonin might be effective in some cases, but noted that indometacin may prove more helpful in patients with unilateral HH (10).
A preventive effect has been observed when patients drank a cup of coffee before going to sleep; however, many patients stopped drinking coffee because of sleep disturbance. In several reported case studies, indometacin and amitriptyline have shown moderate to good efficacy, but did not completely eliminate HH and were not well tolerated (9). In a few cases, antiepileptic drugs such as topiramate, gabapentin and pregabalin have been suggested as being helpful in preventing attack recurrence (3,4).
Due to the unavailability of drugs such as melatonin and topiramate in Morocco, our options were limited. Options were further limited by the history of duodenal ulcer in both patients which did not allow the use of indometacin. Additionally, the use of lithium in patients over the age of 50 is particularly problematic.
All of these factors led us to looking for an alternative treatment and we chose lamotrigine. Lamotrigine is an antiepileptic drug which blocks voltage-dependent sodium channels. Although it has been reported to be an effective prophylactic treatment for migraine with aura (5), its role in the prevention of migraine headaches is still being debated. It is also well tolerated and has good performance and safety profiles, providing it is slowly titrated so as to prevent skin rashes. We achieved a very dramatic outcome with moderate doses of lamotrigine (50–75 mg per day) in our patients, who are still asymptomatic after more than 1 year of follow-up.
Conclusion
To our knowledge, the use of lamotrigine in the treatment of HH has never been reported. Lamotrigine may indeed constitute an effective alternative in the preventive treatment of HH due to its good safety profile, particularly in older patients in whom the use of other drugs can be harmful.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
Conflict of interest declaration
The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.