Characteristics of menstrual vs nonmenstrual migraine: a post hoc, within-woman analysis of the usual-care phase of a nonrandomized menstrual migraine clinical trial.

OBJECTIVE: To compare, using a within-woman analysis, the severity, duration, and relapse of menstrual vs nonmenstrual episodes of migraine during treatment with usual migraine therapy.
BACKGROUND: Studies comparing the clinical characteristics of menstrual and nonmenstrual migraine attacks have yielded conflicting results, contributing to disagreement regarding whether menstrual migraine attacks are clinically more problematic than nonmenstrual migraine attacks.
METHODS: Post hoc within-woman analysis of the usual-care phase (month 1) of a 2-month, multicenter, prospective, open-label study at 21 US medical practices (predominantly primary care). Participants were women > or =18 years of age with regular predictable menstrual cycles (28 +/- 4 days) who self-reported a > or =1-year history of migraine attacks occurring between days -2 and +3 (menses onset = day +1) and > or =8 such attacks within the previous 12 cycles. Migraine treatment episodes were categorized as menstrual (occurring on days -2 to +3 of menses) or nonmenstrual (occurring on days +4 to -3 of menses). Pain severity, functional impairment, duration, relapse in 24 hours, and use of rescue medication were compared. Sources of variability (within- or between-patient) were determined using mathematical modeling. The http://www.clinicaltrial.gov code for trial is NCT00904098.
RESULTS: Women (n = 153; intent to treat) reported 212 menstrual (59.2%) and 146 nonmenstrual (40.8%) migraine treatment episodes. Compared with nonmenstrual treatment episodes, menstrual episodes were more likely to cause impairment (unadjusted odds ratio, 1.65, 95% CI, 1.05-2.60; P = .03), were longer (unadjusted hazard ratio 1.68; 95% CI, 1.31-2.16; P < .001), and were more likely to relapse within 24 hours (unadjusted odds ratio, 2.66; 95% CI, 1.25-5.68; P = .01). Within-patient effects accounted for only 18-33% of the total variance in these outcomes.
CONCLUSIONS: Post hoc, within-woman analysis of migraine treatment episodes categorized based on International Headache Society criteria showed that menstrual treatment episodes were more impairing, longer lasting, and more likely to relapse than nonmenstrual treatment episodes in this selected population of women with frequent menstrual migraine. The current analysis indicates that most of the variability in these outcomes is due to differences between headache types and not within-patient differences for a given type of headache, suggesting that menstrual episodes are potentially treatable. These findings underscore the differences between menstrual and nonmenstrual episodes of migraine and the need to offer effective migraine treatment to women.