Abstract
Introduction
Cluster headache has a life time prevalence between 0.07% and 0.14% in population based studies (1–4) and predominantly begins at an age between 20 and 40 years with a statistical average of 28 years (3, 5, 6). In large sample studies, the age of onset varied, for instance, between 12 and 67 (7) or between 15 and 68 (8). However, the first cluster attack might occur in childhood (9–12) and at higher ages (5, 13, 14). The youngest patient with cluster headache reported in detail in the literature, to our knowledge, was 3 years old (9) and the oldest was 72 (14). Manzoni (5) reported patients with the age of one and with the age of 66, Torelli et al. (15) reported an age of onset of 71 years. However, no further data are available for these latter patients (in particular not how the diagnosis was established in the age of one). The longest remission periods for episodic cluster headache reported in the literature, as far as we are aware, were 11 years (8) and even 25 years (16).
We report three cases of cluster headache who were admitted to our supraregional headache clinic and which present the extreme variability in the possible age of onset in cluster headache.
Case reports
Case 1
A 9-year-old boy came to our headache outpatient clinic because of extremely severe headache attacks every other day lasting for 15–60 min (average 45 min). The attacks were strongly left-sided and periorbital, the left eye was tearing and red, there was a nasal congestion. Analgesics were without any effect, we gave him oxygen (100%, 7 l/min) which stopped the attacks within 5–10 min. His parents did not want a prophylactic medication. An MRI of the skull was unremarkable as were an EEG and the neurological and ophthalmological examinations. The development of the boy was completely normal. His father reported a migraine during his childhood and adolescence; otherwise, the family history was unremarkable. The mother of the boy presented a diary which showed regular headache attacks of this type since his fourth birthday (i.e. age 4.0 years) for five years until presentation in our clinic. The boy reported no other types of headache.
Case 2
An 83-year-old-woman was admitted to our headache outpatient clinic by the department of internal medicine. She complained of very severe headache attacks which occurred one to three times every night for about nine months. The attacks had started during a stay in a rehabilitation centre which she visited after a replacement of a cardiac valve because of mitral insufficiency. She was on anticoagulation because of left cardiac insufficiency. A CT and an MRI scan of her skull were normal, the neurological examination showed symptoms of a mild distal symmetric polyneuropathy and was otherwise normal. She was very fit for her age and did not show any mental disturbances at all.
The attacks lasted between one and two hours. They were strongly left-sided, accompanied by tearing of the left eye, and forced her to move. She never had attacks like this or other types of headache before in her life. Simple analgesics were without any effect. A prophylactic treatment with verapamil (40 mg per day) did not lead to a reduced frequency of attacks. We gave her oxygen (100%, 7 l/min) which stopped the attacks after about 15 min, we also increased verapamil up to 240 mg per day which gave her relief of attack frequency. We did not try sumatriptan because of her age.
Case 3
A 75-year-old man presented to the neurological inpatient clinic because of severe headache attacks for two weeks. The attacks were strongly left-sided and occurred about three to four times a day. The duration of a single attack was up to three hours. One attack always occurred at 2 o'clock a.m. every night. Every attack was accompanied by redding and tearing of the left eye, there was a nasal congestion on the left side. No trigger factors were recognized by the patient. His general practitioner thought of trigeminal neuralgia and treated him with carbamazepine (600 mg per day) without any effect. The patient was given oxygen (100%, 7 l/min) which stopped the attacks after about 20 min Because of an atrioventricular block I°, we did not give the patient verapamil. A prophylactic treatment with prednisolone (100 mg per day for three days, then tapering off for about two weeks) lead to a significant reduction of attack frequency (only one attack every other day) and intensity. The diagnosis of cluster headache was made based on a total observation period of five weeks.
The neurological history of the patient was unremarkable. Laboratory examination was normal as well as an EEG and an ophthalmological evaluation. An MRI of the skull showed a mild microangiopathic encephalopathy which was most probably due to a mild, hitherto untreated, hypertension. He had no history of migraine or other primary headache types, the family history was free of headache disorders.
The patient remembered that he had a similar episode 5 years ago which lasted a few weeks and was considered as trigeminal neuralgia. He did not receive a successful prophylactic treatment at that time and suffered at home. In 1945, at the end of world war II in which he took part as a young soldier, he remembered another episode of similar strongly left-sided headache attacks with the same time features which were also considered as trigeminal neuralgia. He could, however, not remember any further details of treatment at that time.
Discussion
We present three cases of a very unusual age of onset of cluster headache. According to the IHS criteria (17), patient 1 had a chronic cluster headache unremitting from onset. Patient 2 must be classified as cluster headache periodicity undetermined but most probably suffers from chronic cluster headache. Patient 3 fulfils the criteria for episodic cluster headache although he had such a long remission period; the criteria of the IHS do not define how long the remission periods in episodic cluster headache may last. The diagnostic criteria for the single cluster headache attacks were fulfilled by all three patients and diagnosis is further supported by the efficacy of treatment with oxygen, steroids and verapamil.
The cases show that cluster headache can start both very early and very late in life. Sometimes, as in patient 3, the age of onset cannot clearly be determined since this patient most probably had a remission period of 50 years between his first and his second episode of cluster headache attacks. Probably, a number of patients with cluster headache have very long remission periods undetected because of a recall bias which is well known in headache patients. This might also be true for patient 2.
Our observations do no contradict the recent findings that possibly structural or functional changes in the hypothalamus are responsible for the liability to cluster headache (18, 19). However, they show that the onset of cluster headache is independent from the life period of the patients. This supports the hypothesis of an inherited liability which can be triggered by external factors during the whole life.
To our knowledge, we report the patient with the oldest onset of cluster headache and the patient with the longest remission period published in the literature to date. Our report on the child with cluster headache does not present the youngest age of onset published in the literature to date. However, he is one of the youngest patients reported in detail and with a diagnosis based on a diary protocol.
All patients were free of other neurological or psychiatric disorders which might explain the cluster headache. There was even no family history for cluster headache in all patients although epidemiological studies have recently shown an increased genetic risk for cluster headache (20, 21). Future studies on this issue must take into account that cluster headache can start the whole life long and that the distinction between affected and nonaffected individuals must be clearly defined as provisional.
The treatment of cluster headache in very young and very old patients has some limitations. Fortunately, all our patients responded to oxygen which sufficiently controlled their attacks. Sumatriptan subcutaneously is contraindicated for these age groups and steroids might have severe side-effects, in particular in older patients with chronic cluster headache (diabetes, hypertension, osteoporosis). Verapamil was effective in the old patient with probable chronic cluster headache, and steroids could be given to the other old patient with episodic cluster headache.