Abstract
In this issue, Prakash and colleagues present this thought-provoking report on four patients each with cluster headaches which responded favourably within 2 weeks to a relatively high dosage of indomethacin (1). After a thorough literature review, they identified three relevant groups of such patients: (i) those with cluster headache responsive to indomethacin; (ii) those with cluster headache wrongly diagnosed as paroxysmal hemicrania (PH) because of apparent response to indomethacin; and (iii) patients with co-existent cluster headache and PH. They also note that physicians tended to diagnose patients as having PH if patients were responsive to indomethacin even though the clinical profiles of the patients were compatible with cluster headache (1). This paper inspires us to re-think the diagnostic criteria of trigeminal autonomic cephalalgias (TACs) in the International Classification of Headache Disorders, 2nd edn (ICHD-2) (2).
The overlaps of clinical manifestations
Cluster headache and PH are both characterised by attacks of unilateral headache associated with ipsilateral craniofacial autonomic symptoms, the so-called TAC. Of note, there is another headache disorder, hemicrania continua, which is also, and recently, considered as one type of TAC, although it is grouped into ‘other primary headache’ in the ICHD-2 (2,3). The overlaps in duration and frequency of attacks of cluster headache and PH are inherent in the diagnostic criteria of the ICHD-2 (2) and make borderline cases difficult to classify (3). Therefore, response to indomethacin therapy becomes an important differential criterion between these two headache disorders.
Drug response as one criterion for headache disorders
The ICHD-2 lists indomethacin response as one criterion for both PH and hemicrania continua but does not specify either dose range or treatment duration. It is controversial to use therapeutic response as diagnostic criteria because ‘grey zone’ patients do exist, such as: (i) cluster headache patients responsive to indomethacin (1); (ii) putative hemicrania continua patients non-responsive to indomethacin (4); and (iii) PH patients responsive to common cluster headache prophylactics including verapamil and topiramate (5). In fact, many headache disorders are also found to be indomethacin responsive including cough headache, exertional headache, stabbing headache, hypnic headache and idiopathic intracranial hypertension (6).
A common pathophysiology?
Some clues suggest that cluster headache, PH and hemicrania continua might share a common pathophysiology. Observational studies have reported patients with co-existent cluster headache and PH (1) and patients with hemicrania continua evolving from cluster headache (7). Modern neuro-imaging studies have documented hypothalamic activation in all three of these headache disorders (5). The different headache duration and frequency of these disorders might result from different levels of hypothalamic regulation (5).
We fully agree with one very practical point suggested by the authors: indomethacin should be tried in patients with refractory cluster headache before any surgical intervention. Be sure of adequate dosing and treatment duration.