A case of remitting hemicrania continua with seasonal variation and clustering: a diagnostic confusion with cluster headache

Abstract

Hemicrania continua (HC) is an indomethacin responsive primary headache that is characterised by a continuous strictly unilateral headache with periodic exacerbations. About 15% may have a remitting subtype of HC. Herein, we are reporting a 36-year-old man who had a 5-year history of episodic right-sided headaches. The headaches used to occur in a discrete series lasting 4–6 weeks, separated by pain-free remissions of 10–11 months. In each relapse, he had continuous background pain with superimposed exacerbations. The superimposed exacerbations were 1–2 attacks per day, lasting for 2–5 hours, and were associated with ipsilateral cranial autonomic symptoms. However, the patient did not respond to usual therapies of custer headache (CH). He had a complete response to indomethacin. We suggest that remitting subtype of HC may mimic CH. A therapeutic trial of indomethacin should be done in all strictly unilateral headaches who are not responding to other drugs.

Background

Hemicrania continua (HC) is an indomethacin-responsive trigeminal autonomic cephalalgias (TACs) that is characterised by a continuous, strictly unilateral headache with periodic exacerbations.¹ Exacerbations are usually associated with ipsilateral cranial autonomic symptoms (CAS) and agitation. There are two forms of HC: remitting and unremitting subtype. Unremitting HC is recognised by continuous pain, without any remissions, for at least 1 year. In remitting HC, headache is not continuous but is interrupted by symptom-free period of at least 1 day.¹ Herein, we are reporting a case with remitting subtype HC in which the patient had strictly unilateral continuous pain for about 4–6 weeks in a year.

Case presentation

A 36-year-old man had a 5-year history of episodic right-sided headaches. The headaches used to occur in a discrete series lasting 4–6 weeks, separated by pain-free remissions of 10–11 months. The headache attacks occurred with a seasonal variation and all relapses occurred in autumn only (September to November). In each relapse, he had continuous background pain with superimposed exacerbations. The continuous background pain was dull, moderately severe and maximal in the orbital, supraorbital and temporal areas. The intensity of background pain was about 4–5 in Visual Analogue Scale (VAS) and it did not hamper any physical activities. The superimposed exacerbations were 1–2 attacks per day, lasting for 2–5 hours, and were associated with ipsilateral conjunctival injection, lacrimation and rhinorrhoea. The exacerbations were more severe (VAS 8–10) and the patient had restlessness or agitation in about 50% attacks. The pain never occurred on the left side. He had nocturnal attacks but, no circadian periodicity was noted. The patient noted nausea in about two-thirds exacerbations. However, there were no vomiting, photophobia, phonophobia and auras during exacerbations.

He was a non-smoker and a non-alcoholic. There was no family history for similar headache attacks. Physical and neurological examinations revealed no abnormality. Routine biochemical screening was normal. MRI of the brain and MR angiography of the intracranial and extracranial vessels were reported as normal.

He tried a number of drugs (amitriptyline, sodium valproate, propranolol, flunarizine, lithium, verapamil and topiramate), without getting any significant positive effects in every relapse. Cluster headache (CH) and migraine were the two most common diagnoses he received over the years. Oxygen inhalation, oral sumatriptan, oral rizatriptan and injectable zolmitriptan did not provide any significant relief during exacerbations.

Outcome and follow-up

The patient visited our neurology outpatient clinic in one of the relapses, which had been going for about 8–9 days. We made a diagnosis of HC, and indomethacin was started at a dose of 25 mg three times a day. The patient had a complete response within 24 hours. We asked the patient to skip the drug for 1 day to see the placebo response of the drug. The headache reappeared within 12 hours after skipping indomethacin. Reinstitution of indomethacin provided complete relief. As his relapse used to persist for 4–6 weeks, we advised the patient to continue the drug for about 2 months to cover the active phase of relapse. The patient could taper the drug comfortably without any immediate relapse after 2 months. We followed the patients for about 14 months. He had one more relapse after 10 months that again responded completely to indomethacin.

Discussion

Back to back headache attacks over subsequent days or weeks in a year is called ‘clustering’ of headache, highly suggestive of CH.² However, the treatment directed for CH was largely ineffective in this patient. He did not show any responses to lithium, verapamil, topiramate and sodium valproate. There was no response even to oxygen inhalation. It raised a possibility of other headache disorders. A careful history taking pointed out continuous background pain. We also noted that he had several attacks of more than 3 hours (ie, beyond the upper limit of CH attack’s duration). Secondary headaches were already ruled out by several investigations over the years. An episodic nature and a very long history (>5 years) suggested a possibility of some primary headache disorder. A continuous background pain with a response to indomethacin is highly suggestive of HC. A minimum 3-month duration is required to fulfil the diagnostic criteria of HC.¹ At least 1-day remission in a year is required to label it as remitting subtype of HC. However, International Classification of Headache disorders-Third edition (ICHD-3) is silent regarding the duration of relapse or active phase of headache in remitting HC. In Cittadini and Goadsby series, six patients (15%) had a remitting form of HC. The mean duration of the episode was 47 months (range 13–72 months).³ Peres et al reported a 52-year-old woman with a 19-year history of remitting form of HC. Here, the patient had relapses for about 2–3 months and that relapses had a seasonal predilection (between May and July).⁴

A seasonal pattern in headache is the hallmark of CH. However, a seasonal variability may also occur occasionally in other headaches, including migraine, paroxysmal hemicrania (PH) and short-lasting unilateral neuralgiform headache with conjunctival injection and tearing (SUNCT).^5–7 A seasonal pattern with clustering is a diagnostic clue for CH.² Duration and frequency of attacks and therapeutic response to specific drugs help in differentiating CH with other primary headaches.¹

To the best of our literature search, this is the second case of HC (remitting subtype) showing a seasonal pattern. All relapses occurred between September and November. Our case was different from Peres et al’s case in the sense that the patient had a relatively shorter relapse phase (4–6 weeks). Peres et al did not mention the details of superimposed exacerbations of headaches. The literature review suggests that the duration of superimposed exacerbations may vary from a few seconds to a few weeks. Therefore, superimposed exacerbations may mimic several headache disorders, including CH and migraine.⁸ In our patient, the superimposed exacerbations varied between 2 and 5 hours. Therefore, the duration of several attacks fulfilled the criteria of CH. However, many attacks were of more than 3 hours.

The central feature of HC is continuous background pain. However, patients may not volunteer continuous pain if they are not specifically questioned about it.⁹ A response to indomethacin is a must for HC. Ipsilateral CAS and agitation are the specific features for all TACs, including HC. All these features were noted in our patient. The immediate reappearance of the headache on skipping indomethacin is another striking feature of HC.^{8 9} Headache reappeared on skipping indomethacin during the relapse phase of HC. So, we are confident that our case represents the remitting subtype of HC. However, a possibility of CH with interictal pain and responsive to indomethacin should also be considered here. A response to indomethacin has been reported in a few case reports in the literature.¹⁰ However, interictal pain is largely a feature of chronic CH, rather than episodic CH. Moreover, interictal pain is very mild in CH and usually does not present throughout the day.¹¹ The duration of severe attacks/exacerbations does not match with defined criteria of CH. So, a possibility of HC is more likely than CH. A response to indomethacin also raises a possibility of episodic PH with interictal pain. However, the duration of severe painful attacks (2–5 hours) and lesser frequency (1–2/day) are not according to ICHD-3 criteria for PH. Moreover, interictal pain is a feature of chronic PH, rather than episodic PH.

The circannual periodicity of CH is explained by pathophysiological disturbances in the hypothalamus. Positron emission tomography (PET) study has demonstrated an activation of the contralateral posterior hypothalamus in patients with HC.¹² So there is a theoretical possibility of circannual periodicity of HC, especially in remitting type. Besides Peres et al case report, a few patients in Cittadini and Goadsby series have also some circannual periodicity.³ This case highlights the importance of a background headache in all side-locked headaches to clinch the diagnosis of HC. A therapeutic trial of indomethacin is required in all such cases.

Conclusion

A remitting subtype of HC may have variable presentation and may mimic several primary headache disorders, including CH.

Patient’s perspective.

Case-1: I am happy that I got diagnosed.

Learning points.

• Remitting subtype of hemicrania continua may mimic cluster headache.

• All patients with strictly unilateral headache should be inquired about the presence of continuous background pain.

• A therapeutic trial of indomethacin should be done in all strictly unilateral headaches who are not responding to other drugs.