Abstract
Dear Sir We report a 25-year-old female patient whose maternal grandmother suffered migraines. She had no notable medical antecedents. There was no history of epistaxis or bleeding disorders. She was referred to neurology from shortly after menarche and had had holocranial pulsating cephalalgia attacks every 2–3 months with nausea, phono- and photophobia lasting 72 h. The patient treated migraines with 1000 mg paracetamol, and had never taken non-steroidal anti-inflammatory drugs (NSAIDs). These episodes were not preceded by neurological symptoms and prevented her from following her daily routine. They became worse with physical activity. Pain receded coinciding with bilateral epistaxis. These nose bleeds occurred with every attack and never without such an attack. General physical (including blood pressure) and neurological examinations were normal. On one occasion, she had been to the emergency room (ER) for evaluation by otorhinolaryngology (ORL), ruling out organic pathology; a coagulation study was done as well as cranial magnetic resonance imaging, both of which were normal.
Epistaxis frequently occurs in the general population and along with local lesions, and is usually due to coagulation disorders or blood pressure (1). In 1967, Ikonomoff (2) studied 24 patients with this problem, and more than half of them were migraine related. Since then, a relation between migraine aetiopathogenesis and this disorder has been considered (3). Ikonomoff's initial observation has recently been confirmed by a study of two groups of paediatric patients (migraine and control group), which concluded that the former were more predisposed to suffer repeated nose bleeds compared with the control group. Their emergence in infancy could be a precursor of future migraine (4).
The characteristics of cephalalgia in our patient were compatible with migraine without aura according to the International Headache Society 2004 classification (5). In cephalalgia, and migraines in particular, the occurrence of epistaxis during an attack is very infrequent. In fact, we have found only one case described in the literature. It relates to a patient who, like ours, had normal blood pressure with migraine attacks that ended in nose bleeding, with no other medical explanation except the temporary coexistence of migraine (6). In our case, no other possible explanation was found because neuroimaging, coagulation tests and ORL evaluation were normal. Sacks (7) has stated that at least one-quarter of migraine sufferers develop engorged and purple turbinates in the course of an attack. In our patient, ORL examination did not show engorged mucosa and she did not complain of blocked or running nostrils before the nose bleeds. Epistaxis also could not be ascribed to treatment, as our patient took no NSAIDs or triptans. Repeated episodes ruled out a chance connection. Moreover, coincidence supports a record that migrainous women have a higher frequency of other bleeding disorders such as menorrhagia (8). Epistaxis during a migraine attack could probably be explained by a common origin, trigeminovascular system (TVS) activation. Nostril irrigation depends on external and internal carotid artery terminal branches, which coalesce and form an arterial area that relies on the TVS. The main branch supplying nostrils from the external carotid artery is the internal maxillary artery; ethmoidal arteries (branches of ophthalmic artery) from the internal carotid artery. TVS activation would provoke vasodilation and subsequent bleeding (9). Several facts support this theory. Trigeminal activation increasing extracerebral flow has been proven (10) and the implications of this system in the physiopathology of migraine are not in dispute (11). Furthermore, the presence of facial flushing and vasodilation of extracranial arteries prior to pain have been described in migraine (12) along with their appearance in other trigeminal cephalalgias, such as cluster headache (13). Although the possibility of epistaxis during migraine attacks is infrequent, it is possible that both conditions are due to TVS activation.
Acknowledgements
We thank Michelle Sigrid Kremser, from Virgin Macarena University Hospital, for the English translation.