Abstract
Case report
A 72-year-old retired woman was evaluated for recurrent bouts of severe left temporal-orbital headaches that began about 32 years earlier. She described the headaches as throbbing with an occasional stabbing quality. Her headaches reached their maximum intensity within a few minutes. Each attack lasted from 30 to 45 min, then subsided, only to be followed by another attack within 30–60 min, totalling up to eight attacks per day. The bouts of headache persisted for several weeks only to remit for several months. Each of the headaches was accompanied by lacrimation and injection in the left eye, rhinorrhoea in the left nostril, osmophobia and occasional photophobia. She denied nausea, vomiting, sensory or motor symptoms. She denied any personal or family history of headaches suggestive of migraine. However, she reported a first cousin with cluster headache (CH).
For over 30 years, approximately 50% of the headache cycles were preceded by a brief visual disturbance consisting of ‘flashing lights’ or ‘zigzagging lines’ that occurred 3–7 days before the onset of the first headache in the bout. These visual symptoms lasted from 30 s to about 3 min and never occurred without a subsequent headache cycle. Furthermore, her visual symptoms never occurred after the headache cycle had started or within an hour prior to an individual headache. In addition to the visual symptoms, she described a feeling of scalp tenderness on the left side of her head that occurred either with or soon after the visual symptoms and preceded the headache cycle's onset by several days. Initially, headache bouts were sporadic and occurred unpredictably at any time of the year, but during the 5 years prior to presentation, her headache bouts became seasonal (autumn and spring only) and became more excruciating. She could recall neither any triggering factors nor any relationship between the headaches and her menstrual cycle prior to menopause. The patient had tried a number of treatments including O2, indomethacin and low-dose verapamil without success. She obtained some relief from her headaches by applying ice pack to the painful side of her head. Magnetic resonance imaging (MRI) of the head with and without contrast showed small vessel white matter changes in the frontal subcortical white matter. No abnormalities were present in the occipital cortices. There was a right frontal arachnoid cyst and an ovoid 1–2-cm area of bright T1 signal in the anterior parasagittal frontal region that may represent calcified meningioma. Magnetic resonance angiography (MRA) showed no significant carotid or vertebrobasilar stenosis or arteriovenous malformation (AVM). EEG was performed and was normal and without epileptiform discharge or focal dysrhythmia. A transoesophageal echocardiogram showed no patent foramen ovale.
The patient was treated with verapamil 80 mg three times a day because the initial trial of verapamil was inadequate in dose and duration. The patient was interviewed over the phone after 3 months and seen in follow-up 9 months after her initial visit. She denied having headaches or visual symptoms since her initial visit even at the expected times, during the autumn or in the spring.
Discussion
To our knowledge, this is the first report of stereotypical neurological symptoms occurring exclusively with onset of a CH bout. The intensity, cyclicity, duration and associated ipsilateral autonomic symptoms of this patient's attacks are consistent with CH (1). In general, transient focal neurological symptoms do not commonly accompany CH, unlike migraine in which over 20% of sufferers experience one or more of these symptoms (2). In migraine, a complex of symptoms known as the aura precedes or accompanies the headache and is generally separated by no more than 1 h from the onset of pain. Most aura symptoms develop over 5–20 min and usually last less than 60 min. Typical migraine auras can involve visual, sensory or language dysfunction (3). There are two reported series in which aura or aura-like symptoms have been associated with the onset of individual cluster headaches. One study described five patients with visual and one patient with olfactory symptoms, lasting from 5 to 120 min followed shortly thereafter by a CH (3). In this series, only one of the five cases reported that the aura preceded the onset of the headache by more than 1 h (4). In a larger series of CH patients (n = 230), 33 patients were reported to have experienced aura (visual, hemisensory or hemimotor) symptoms either during or within 1 h before or after the CH (5).
In this case, the visual disturbance does not meet the criteria for typical migraine aura. It differs from aura in two ways: first, it is very brief, lasting from seconds to 3 min. Second, the interval between the aura and the onset of the CH symptoms is prolonged, from 3 days to 1 week. In this patient, the symptoms signal the beginning of the cluster bout rather than the onset of an individual headache. However, the ‘zigzagging lines’ and ‘flashing lights’ are symptoms strikingly similar to those reported in migraine visual aura. Several investigations including brain MRI, MRA, EEG and transoesophageal echocardiogram failed to provide evidence of an AVM, epileptiform discharge or any potential occipital epileptic focus. The areas of increased signal in the frontal subcortical white matter are fairly common in patients in their eighth decade of life and given their frontal location are probably unrelated to the visual symptoms. The head soreness that the patient experienced in the left temporal area in connection with the visual symptoms a few days prior to the onset of headache is premonitory (6) and cannot be ascribed to peripheral or central sensitization caused by the CH. The patient's brisk therapeutic response to verapamil (240 mg/day) is interesting, although it does not clarify the relationship of the visual symptoms to the cluster headaches. It was chosen based chiefly on the clinical considerations that verapamil when adequately dosed is effective in CH (7) and that it is a frequently used prophylactic medication in migraine with prominent neurological symptoms.
There is increasing debate over the relationship between headache and the transient focal neurological symptoms experienced during primary headache disorders. This case highlights the variability and potential complexity of this relationship. Increasing experimental evidence suggests that there are mechanisms by which the trigeminocervical pain system may be activated by cortical phenomena such as the migraine aura (8, 9). The relationship (if any) between the scalp tenderness reported by our patient and her visual symptoms is not clearly understood. It is also difficult in this particular case to attribute a direct activating role in the CH to the short-lived visual symptoms given the long interval of temporal separation. Yet these symptoms never occurred without the subsequent onset of a cluster bout. Although the cause of her visual symptoms is not known, it is possible that this patient is vulnerable to the occurrence of a cortical phenomenon that underlies her short-lived visual symptoms and which shares some characteristics with the phenomenon underlying the migraine aura. Given this vulnerability, it may be that in this particular case, the brain events resulting in the onset of the cluster bout were sufficient to generate visual symptoms long before the activation of the headaches.