Female hormones are thought to be of importance in the aetiology of migraine, which is more common in women than in men. Little attention has been paid to androgens. This study investigates the associations between migraine and serum levels of androgens in postmenopausal women not taking oestrogens. A case-control study was carried out among women participating in a mammography screening program. A neurologist clinically assessed the participants. Headache criteria proposed by the International Headache Society were used. Each of the 15 women with migraine was matched to three controls by time since menopause and by body mass index. Serum levels of androstenedione and total testosterone were measured by radioimmunoassays. Free testosterone was calculated from total testosterone, immunoassayed sex hormone-binding globulin and albumin. The mean±S.D. serum level of androstenedione was 2.7±1.1nmol/l and 3.4±1.9nmol/l in cases and controls respectively. The mean serum level of testosterone was 0.8±0.4nmol/l and 1.0±0.5nmol/l in cases and controls respectively. The mean serum level of free testosterone was 12.5±8.5pmol/l and 14.0±7.9pmol/l in cases and controls respectively. Conditional logistic regression analysis was used to test for differences in serum levels of androstenedione, total testosterone and free testosterone between cases and controls. No statistically significant differences were found. To conclude, there was no evidence in this study that differences within normal levels of androgens play an important role in the aetiology of migraine in postmenopausal women not taking oestrogens.
The pathophysiology theory of migraine postulates a local, neurogenic inflammation and the possible involvement of oxidative stress. We analysed the levels of 15-oxo-dihydro-prostaglandin F 2α (a metabolite of prostaglandin F 2α ) and 8-iso-prostaglandin F 2α (a major isoprostane), which are biomarkers for inflammation and oxidative stress respectively, in urine from 21 patients with migraine, with and without aura. Urine samples from migraine patients were collected during a migraine attack, and control samples were collected from the same subjects on a migraine-free morning. The mean basal levels of 15-oxo-dihydro-prostaglandin F 2α and 8-iso-prostaglandin F 2α in the morning control urine samples were 0.54±0.11 and 0.31±0.13nmol/mmol of creatinine respectively. The mean levels of 15-oxo-dihydro-prostaglandin F 2α and 8-iso-prostaglandin F 2α in the urine samples collected during the migraine attack in the 21 patients were 0.53±0.13 and 0.32±0.11nmol/mmol of creatinine respectively. Thus there were no differences in the 15-oxo-dihydro-prostaglandin F 2α and 8-iso-prostaglandin F 2α excretion rates during the migraine attack compared with on the migraine-free day. However, the basal 8-iso-prostaglandin F 2α excretion levels on the migraine-free day were significantly lower in pre-menopausal women (0.24±0.08nmol/mmol of creatinine, n = 11) compared with post-menopausal women (0.39±0.14nmol/mmol of creatinine; n = 7; P = 0.009). In conclusion, in this study we found no support for the involvement of inflammation and oxidative stress in migraine pathophysiology. Our results indicate, however, a lower level of oxidative stress in pre-menopausal compared with post-menopausal women.