Familial hemicrania continua

Abstract

There are now three known causative genes for familial hemiplegic migraine and increasing evidence to support a genetic predisposition to the more common types of migraine with and without aura, and for cluster headache. We present the first reported case of familial hemicrania continua. A mother and daughter developed hemicrania continua at the same time of life. Both showed an absolute response to indometacin and at similar doses. Both also suffered from migraine with aura. We discuss the increasing support for a genetic predisposition to dysfunction of the pain system within the brain manifesting as primary headache.

Introduction

Hemicrania continua is defined as a persistent and strictly unilateral headache responsive to indometacin (1). It is characterised as a daily and continuous headache of moderate severity, with exacerbations of more severe pain accompanied by ipsilateral autonomic symptoms. The latter can include conjunctival injection, lacrimation, nasal congestion, rhinnorhoea, ptosis or miosis. First described in the early 1980s, it was initially considered to be a rare primary headache (2,3). Research groups undertaking a systematic search for the condition have found it to be less rare than originally thought (4). The disorder is most likely under-reported, and has been included in the most recent edition of the International Classification of Headache Disorders (ICHD) (1).

The evidence for a genetic basis for the primary headache syndromes is steadily growing. Migraine has long been recognised to have a genetic basis. There are now a number of genes associated with the development of familial hemiplegic migraine, a rare autosomal dominant form of migraine with aura (5–7). It is most likely that the more common forms of migraine with aura and migraine without aura will have a polygenic basis to inheritance (8–10). Familial hemiplegic migraine presents at one extreme as a monogenic, high-penetrance disorder, whereas migraine without aura appears to present as a polygenic and low-penetrance disorder.

Cluster headache has been reported in monozygotic twins (11–13). There has been shown to be a 13–14-fold increase in cluster headache in first-degree relatives of affected individuals (14,15), with some families exhibiting an autosomal dominant pattern of inheritance (16–18). There are recent reports of familial cases of short-lasting unilateral neuralgiform headache with conjunctival injection and tearing (SUNCT) (19) and paroxysmal hemicrania (20).

We report the cases of a mother and daughter with hemicrania continua. Both also suffer from migraine with aura. This is the first report of familial hemicrania continua in the literature. Heritability and co-existence of more than one primary headache disorder is also discussed.

Clinical cases

Case 1

MD, a 53-year-old woman, presented with a 30-year history of headaches. These were of two types. The predominant headache was a side-locked, left-sided pain which was sharp, constant and not exacerbated by movement. This would never rate less than 2/10 on a verbal pain rating scale. Several times a week she experienced exacerbations of the pain to 6–8/10. These exacerbations were associated with ipsilateral photophobia, blockage of the nasal passage and ptosis, but no other autonomic symptoms, and no nausea. The exacerbations occurred between 3 a.m. and 5 a.m. and would last for the rest of the day unless she took naratriptan. With this the pain would take about 3–4 hours before settling back to a milder background severity. There was no other precipitant to the exacerbations.

Independent of this left-sided headache, four times a year she experienced typical fortification spectra lasting 20–30 minutes, followed by right-sided throbbing headache, evolving to become global, associated with nausea and photophobia, and of up to two days’ duration.

At the time of referral the patient was taking naratriptan for the severe left-sided exacerbations, with a little relief. She had previously tried sumatriptan orally and naproxen acutely without effect. Indometacin had been tried but only to a daily dose of 75 mg. Unsuccessful attempts at prophylaxis had been tried with gabapentin, verapamil, dosulepin, sodium valproate chrono, carbamazepine and paroxetine. The infrequent migraine attacks responded reliably to Migraleve (paracetamol, codeine phosphate and buclizine hydrochloride) or Propain. Neuroimaging was normal.

A presumptive diagnosis was made of hemicrania continua, and infrequent attacks of migraine with aura. As she was taking naratriptan on 12 days a month, the possibility of triptan overuse was considered. MD was advised to discontinue all acute treatment for a period of six weeks. At this stage indometacin was re-introduced, and titrated up to a dose of 275 mg daily with complete resolution of the daily left-sided headache. Delays in taking her doses led to recurrence of this head pain, which was reliably abolished on taking the delayed dose. Over the following six months, MD gradually reduced the dose to 125 mg daily, and has since remained headache-free at this dose +/−25 mg. She has never been able to reduce the dose below 100 mg because of recurrence of pain. The indometacin renders her free from both the background daily pain and the exacerbations.

Case 2

At a clinic visit 18 months after diagnosis, MD mentioned that her daughter, HC, also suffered from similar headaches. At the time of referral, HC was a 26-year-old primary school teacher with a 15-year history of headaches. She had suffered from infrequent headaches since the age of 11. She described headaches often triggered by sleep deprivation. These were preceded by an aura consisting of a fuzzy shape in the right visual field, followed by the development of dark patches, which eventually coalesced before fading out after approximately 45 minutes. The pain was generalised and throbbing. The headaches were improved with Migraleve (paracetamol, codeine phosphate and buclizine hydrochloride) and resolved completely after sleep.

At the age of 18, she developed a second type of headache. This was a constant side-locked, right-sided headache. At the time of referral, this could be as mild as simply a feeling of pressure (1/10). On 3–4 days each week the pain was more noticeable (3–4/10) and on one day each week, severe (8/10). The moderate and severe pain was associated with nausea and ipsilateral photophobia. The severe exacerbations were associated with ipsilateral nasal stuffiness, ptosis and occasional tearing. There was no precipitant to the exacerbations.

Attempts at acute treatment with aspirin, paracetamol, ibuprofen, Propain, codeine and Co-codamol had all been ineffective for the more recent-onset headache. Indometacin had been tried, but again only to a dose of 75 mg daily. At the time of referral, she was taking a long-acting formulation of Etodolac, which reduced, but did not abolish, the pain. Neuroimaging was normal.

HC received a presumptive diagnosis of hemicrania continua, and a diagnosis of migraine with aura. Etodolac was discontinued and indometacin was restarted. The dose was titrated up to 325 mg daily, at which stage she became headache-free. Unfortunately, despite taking lansoprazole 30 mg daily, after three weeks she experienced a significant upper gastrointestinal haemorrhage. The indometacin had to be discontinued, leading to a recurrence of the headache.

Discussion

There are a number of interesting issues presented by this report. Most reported patients with phenotypic hemicrania continua have an indometacin response at lower doses than both patients in this report. Some patients do require initial higher dosing, as did both patients reported. However, in most published cohorts the maximum dosing of indometacin trialled has been between 150–225 mg total daily dose (4). This is not unexpected given the known adverse event profile and licensed maximum dosing of 200 mg daily. The questions arises whether the published non-responders may have responded to higher dosing, given the opportunity. Of the primary headache disorders, the genetic predisposition of migraine has been the most extensively investigated by virtue of high prevalence and significant disability. The advent of clinical classification criteria has aided the collection of patient cohorts with more uncommon and thus less well-recognised primary headache disorders. It follows that with time familial cases are more likely to be encountered, and potentially adequately sized patient populations may allow heritability to be further defined.

Both our patients suffered from hemicrania continua and migraine with aura. Further enquiries led to the establishment of a more extensive pedigree, including other family members with migraine, with and without aura (Figure 1). The co-existence of more than one primary headache disorder within an individual is reported for most of the primary headaches (21,22); this includes reports of up to three primary headache phenotypes in one patient (23). Hemicrania continua has been reported to co-exist with migraine with aura (24,25) and cluster. The co-existence of more than one primary headache disorder in an individual would not necessarily be unexpected, albeit uncommon. However, co-existence of more than one disorder in a patient population, greater than would be expected based on population prevalence data, raises the possibility of shared pathophysiology at some level. Shared physiology is reported for cluster headache and migraine (22); it remains to be determined whether this exists for the more uncommon primary headache disorders. It is possible that there may be a genetic predisposition which results in dysfunction of the pain system within the brain and renders such individuals predisposed to one or more phenotypes of spontaneous head pain. It is likely that this predisposition in the majority is polygenically determined.

Figure 1.

Patients’ family history for headache. The headache history of deceased family members is unknown.

Current functional imaging studies suggest different structures integral to pain mechanisms are active during, and specific to, different primary head pain phenotypes. The dorsal rostral pons has been implicated in migraine headache (26–28). The syndromes of cluster headache, paroxysmal hemicrania and SUNCT have been phenotypically categorised as the trigeminal autonomic cephalalgias in the most recent edition of the ICHD (1). The clinical categorisation has since been strengthened mechanistically by functional imaging studies implicating distinctive hypothalamic activation (29–31). Hemicrania continua has straddled clinical boundaries. The strict unilaterality, autonomic features and indometacin responsiveness have favoured an inclination towards the trigeminal autonomic cephalalgias. However, the disorder also shares clinical features and peculiarities typical of migraine (4,32,33). The latter includes the tendency towards the development of medication overuse headache (34). Consistent with the clinical characteristics, recent functional imaging has shown activation of both the dorsal pons and hypothalamus in hemicrania continua (35). Ultimately, we await functional imaging paradigms which allow visualisation of pain dynamics during different headache phenotypes in the same individual.

The current definition of hemicrania continua requires not only the defined phenotype but an absolute response to indometacin. Both patients responded, albeit at higher doses. Many patients do not achieve an adequate trial of indometacin as determined by limiting side effects. Some patients, despite trying high doses, do not respond. Do these patients require even higher doses or are there typical phenotypes which are non–indometacin responsive? How does the pathophysiology of these indometacin-responsive and non-responsive phenotypes differ? Adequately sized patient populations remain imperative for the acquisition of reliable epidemiological data and to provide further insights into putative pathophysiological mechanisms and effective therapeutics.