Abstract
Hemicrania continua (HC) is a primary headache syndrome marked by a continuous one-sided headache typically of low-grade intensity with exacerbation periods of increased pain intensity and associated migrainous and autonomic symptoms (1). HC is one of the indomethacin-responsive headache disorders and is part of the trigeminal–autonomic spectrum of headaches. The response to indomethacin is fast and dramatic and helps to define the syndrome. Very few other agents have demonstrated any response in HC. Recently, positron emission tomography (PET) imaging has shed some light on the possible pathogenesis of HC (2). Activation was noted in the contralateral posterior hypothalamus and ipsilateral dorsal rostral pons in seven patients with indomethacin-sensitive HC. In several recent case reports HC has been observed to occur in patients with a concomitant history of migraine (3) or cluster headache (4). As both the hypothalamus (possible cluster headache generator) and pons (possible migraine generator or modulator) are activated in HC, it would seem likely that we should be seeing more of these dual headache presentations.
A patient is presented who had a history of verapamil-responsive episodic cluster headache and who began to develop a new continuous headache contralateral to her cluster headache side. This headache was indomethacin responsive and a diagnosis of HC was made. Verapamil was tried as a HC preventive and the patient became pain free on verapamil with a very small dose of concomitant daily indomethacin. This case suggests that HC may respond to non-indomethacin treatments when it occurs in individuals with other trigeminal-autonomic cephalgias (TAC). This is an important observation, when the long-term possible side-effects of daily indomethacin are considered.
Case report
A 24-year-old woman presented with a 6-year history of episodic cluster headache. The headaches always occurred on the left side. They were severe in intensity, caused retro-orbital pain, lasted 45 min and were associated with unilateral ptosis, conjunctival injection, lacrimation, nasal congestion and rhinorrhoea. During the headache she would pace, press her hand into her left temple region and shake her legs. She would experience two attacks per day and her typical cluster period lasted 2–3 months. She had some response to valproic acid as a preventive but was never pain free on medication during a cluster cycle. At her first visit (1 month into a cluster headache cycle), verapamil was begun and at a dose of 480 mg/day she became pain free. Since that time with every subsequent cluster headache cycle verapamil has allowed the patient to become pain free during a cluster period. While in a cluster headache remission period she began to develop a new type of headache. This was right-sided, behind her eye with some occipito-nuchal discomfort. The pain was constantly present with no pain-free moments but was of low-grade intensity. Almost daily around noontime the pain intensity would rise to severe and was associated with migrainous features (photophobia, phonophobia) but no true autonomic symptoms and no sense of agitation. The period of pain exacerbation would last 12–18 h. According to the patient, this new type of headache pain was different from her cluster headache pain. A probable diagnosis of HC was made and indomethacin was started. She initially was dosed up to 150 mg/day with a significant decrease in headache intensity with pain-free periods and at a dose of 250 mg/day she became pain free and remained so as long as she continued this dose. Any decline in indomethacin even by 25 mg led to pain recurrence. The patient tolerated the indomethacin without any complications and was content and happy to remain on it, as she was pain free. A brain magnetic resonance imaging/angiography (to rule out a secondary cause) was ordered because of the new headache and the large dose of indomethacin required to treat it (5) and was negative. Three months into the right-sided headache, verapamil was prescribed by the author to see if the patient would have any response and to see if she could at least lower her indomethacin dosage. Verapamil was started at 80 mg bid and increased every 5–7 days by 80 mg. The patient was encouraged to taper down on her indomethacin as the verapamil dose was raised. At a verapamil dose of 480 mg/day (160 mg tid) the patient was completely pain free and was able to taper off all of her indomethacin with no pain recurrence for 1 week. Her pain then recurred but with the addition of only indomethacin 25 mg/day she became pain free again and has been so through 8 months’ follow-up. If she stops the indomethacin the headache will return within 10 h and if she tries to lower the verapamil dose the pain will recur within 24 h.
Discussion
This is the second documented case of HC occurring in a patient with cluster headache. As with the case documented by Lisotto et al. (4), the headache of HC occurred contralateral to the side of the cluster headache. This makes sense on a pathogenesis basis since the publication of the PET data by Matharu et al. (2). In cluster headache hypothalamic activation occurs ipsilateral to the side of the headache while in HC hypothalamic activation occurs contralateral to the side of HC. Thus, depending on what side of the hypothalamus is activated, different clinical headache presentations could evolve in the same individual.
In the presented case, the patient had HC based on her headache symptoms and her complete response to indomethacin. She did not, though, have any autonomic symptoms during her pain exacerbation periods. The new International Headache Society criteria require at least one autonomic symptom during the pain exacerbation period to make a diagnosis of HC. However, Peres et al. (1), in one of the largest clinical descriptive studies of HC, found that only 74% of their HC patients developed autonomic symptoms with their headache, thus 26% never had autonomic symptoms with diagnosed HC.
In many individuals with HC indomethacin treatment may be required for a decade or longer. HC patients are usually very content to stay on treatment as they are pain free, but for the physician there should be significant worry about possible long-term complications. There are great risks with continuous treatment with indomethacin, including gastrointestinal haemorrhage and renal papillary necrosis, both of which could lead to mortality. Verapamil was chosen as a possible preventive agent for HC in this patient based on the response to verapamil with her cluster headaches. In addition, her HC had a unique cyclic/circadian flavour to it with the almost clock-like pain exacerbation periods at noontime each day. Initially, verapamil alone at a dose of 480 mg/day completely and unexpectedly alleviated her HC pain. Subsequently, the added verapamil allowed the patient to be pain free on only 25 mg/day of indomethacin vs. a previous dose of 250 mg/day when she was on indomethacin alone. Verapamil, as well as other typical cluster headache and migraine preventives, has never shown much benefit in HC. Perhaps there are different subforms of TACs, those that occur in isolation and thus respond to distinct therapies (e.g. paroxysmal hemicrania, hemicrania continua to indomethacin) and those that occur in combination in the same individual and thus have a shared treatment response. Evers and Husstedt (6) documented partial response to verapamil in several patients with indomethacin-responsive chronic paroxysmal hemicrania (CPH). There was no documentation if these patients had CPH in isolation or had experienced CPH in combination with another TAC subform.
One could also hypothesize that there are hypothalamic-influenced forms of HC and thus they will respond to cluster headache (a hypothalamic-generated headache) preventives, whereas typical HC is non-hypothalmic influenced and thus responds only to indomethacin. HC normally lacks the circadian features of cluster headache; however, Peres et al. (7) have described a case of seasonal HC suggesting a hypothalamic subtype. In the Lisotto et al. (4) case patient with probable verapamil-responsive cluster headache there was complete response of the HC to indomethacin but not to verapamil, although the verpamil dose was never pushed above 360 mg/day. Of note, this patient was started on verapamil for cluster headache and 1 week later his headaches ceased, questioning if this was truly verapamil-responsive cluster headache or just the natural end of a cycle.