Abstract
Paroxysmal hemicrania (PH) is an uncommon primary headache disorder of which about 80 cases have been well described in the literature (1). Its prevalence has been estimated to be 0.021% in the general population (2), corresponding to about 2% of cluster headache prevalence (3). The female
Case report
A 43-year-old woman first presented to our Headache Centre in January 2004. She had a negative pathological history and a familial history positive for migraine. She had suffered from MoA, ICHD-II criteria, since puberty. The attacks were described as pulsating, generally moderate in intensity, aggravated by physical effort, lasting about 6 h, with photophobia and phonophobia, nausea without vomiting, occurring once or twice a month. The localization was always on the left side, when lateralized. The headaches were diffuse in about 10% of cases, but always arose from the left side of the head. Her attacks were frequently, but not invariably, related to menstruation. At the beginning for her headaches she had used non-steroidal anti-inflammatory drugs (NSAIDs) with a good response after a few hours, but during the last 5 years she had obtained relief only with triptans. About 4 years before presentation a new type of headache had developed, characterized by a stabbing, burning, very intense pain, lasting about 3–4 min, accompanied by tearing, rhinorrhoea and ptosis, even if not all together in the same attack. The pain was periorbital, strictly unilateral on the left side. These attacks were strictly related to the patient's menstrual cycle; they usually began on day −2 from the onset of menses and recurred for 6–8 days until the end of the bleeding. The frequency was 4–10 episodes/day, usually during the night time. The patient reported that such paroxysmal headache episodes did not occur outside the menstrual period, as demonstrated by the prospective evaluation of the diaries she has filled out for about 3 years after admission to our Centre. Her neurological examination was negative. Routine examinations (blood, urine analysis, ECG, chest X-ray) were negative. Brain magnetic resonance imaging (MRI) with gadolinium showed a cyst of 1 cm in diameter, with liquor-like content over the left quadrigeminal lamina, with slight compression on the left superior colliculus, without modification after enhancement, interpreted as a pineal gland cyst (Fig. 1). The patient did not usually treat her attacks, because they were too short and, furthermore, NSAIDs (not indomethacin) and triptans did not significantly affect the course of attacks. The patient was advised to take indomethacin at a dosage of 50 mg bid at the time of the onset of the headache attacks and thereafter during menstruation until the end of menstruation. This therapy was able to stop the ‘monthly attack clusters’ or prevent them completely. She repeated the therapy monthly, as the first attack occurred with a resolution of the attacks in the same day or on occasion of the beginning of the menstruation. This therapy did not prevent the attacks of MoA, which were responsive only to the triptans. She presented MoA attacks with the same frequency as before the onset of EPH. The patient has been followed by our Headache Centre for the last 3 years and the efficacy of therapy has persisted during this period. Eight months and then 2 years later MRI controls were performed, without modification of the lesion.
Saggital enhanced T1-weighted magnetic resonance image of the brain shows a pineal cyst (black arrows) with a maximum dimension of 10.5 mm.
Discussion
PH is a rare but well-characterized clinical entity, so single case reports no longer appear in the literature. Nevertheless, to the best of our knowledge our patient represents the first case of PH strictly related to the menstrual cycle. Although in this case the free interval between episodes was usually <30 days, we believe that EPH is the more appropriate diagnosis.
The brain MRI performed in this patient disclosed a pineal gland cyst, which remained unchanged in diameter during the follow-up period. Although secondary cases of PH were not mentioned in a review examining 84 cases described in the literature, of which 17 were considered to be EPH (3), a more recent review of symptomatic PH has highlighted the presence of 22 cases associated with an organic pathology, including vascular (aneurysm, arteriovenous malformation, artery infarction), tumoral (cavernous or petrous ridge meningiomas, pituitary adenoma, parotid metastases, Pancoast tumour) and miscellaneous diseases (maxillary cyst, ophthalmic herpes zoster, intracranial hypertension, post-traumatic) (9). Pineal gland cysts are thought to be common findings in neuroimaging investigations, as they have been reported in 1.4–4.3% of brain MRIs (10). Therefore, the association of PH episodes temporally related to the menstrual cycle and the pineal cyst appears to be merely coincidental in this case. The association between migraine and EPH is an uncommon feature and again reflects a probable coincidence, as previously described in CPH (11). The correlations between these two forms of headache in our case are limited to the side of pain occurrence and to the menses as trigger factor, invariable for EPH and frequent for MoA. On the other hand, the headaches responded differently to therapy, in that EPH attacks were completely prevented by use of indomethacin, whereas episodes of MoA were responsive to triptans, but not to indomethacin or NSAIDs. Furthemore, the course of MoA does not appear to have been modified in the 4 years since the onset of EPH. The pathogenesis of these two forms of primary headache is considered different in the literature, even if not sufficiently explained for both conditions, particularly for PH (8). Female hormones may play an important role in both headaches, perhaps more so in PH than in migraine, considering the male/female ratio. The mechanism by which hormones act in migraine is unknown, but central sensizition, fluid and salt retention, platelet aggregation or change in serotonin and prostaglandin levels may play a role. To date, there is no evidence to extend such explanations to the pathogenesis of PH because of the lack of studies on this issue. As a clinical observation, Boes and Dodick have reported a case of PH in which the patient had worsening of headache during menses without other details (12). The correlation with menses has been considered in the review by Antonaci and Sjaastad (3) of 84 cases. They highlight the benefit from menstruation in four cases and worsening in 11 cases.
From a clinical standpoint, the correlation between migraine and menses has been clearly established and there is a proposal for introducing a specific entity in the appendix of ICHD-II. As regards the other primary headaches, tension-type headache seem to be more frequent during the menses than at other times, and menstruation has been reported to be an important trigger factor (13–15). Indeed, a recent paper failed to demonstrate such an association in cluster headache (16). To the best of our knowledge, there are no specific studies on the relationship between the other trigeminal autonomic cephalalgias and menses or, in particular, on menses-related EPH. Considering the clear female preponderance in PH and the strong link with menses in our patient, further studies are needed to elucidate the complex linkage between hormones and PH.