Migraine is a common, often undiagnosed, disorder that affects an estimated 21.5 million American women aged 12 years and older. Eighty-two percent of patients who consult a physician about their headaches do not visit a neurologist or headache specialist; instead, they consult their primary care practitioner, which, for many women is her gynecologist.1 Data from a follow-up study from the American Migraine Study II suggested that an estimated 4 million women have consulted a gynecologist for headaches, and approximately 1 million women currently consult their gynecologists for headache (Glaxo-SmithKline, unpublished data, Migraine Prevalence Follow-up Study). About 60% of women notice an increase in the frequency of migraines related to their menstrual cycle (menstrually associated migraine), and this relationship may be the reason why migraineurs consult gynecologists regarding their headaches.2,3 It is therefore important for gynecologists to know about the association of migraine and the menstrual cycle, its diagnosis, and treatment.
Migraine should be given consideration in the differential diagnosis for headache occurring during the perimenstrual period. Population-based diary studies indicated that migraine occurring during the perimenstrual period is mostly migraine without aura and is significantly more likely to occur on the 2 days before and the first 3 days of menses.4–6 Only a small percentage of women have migraine attacks exclusively in this window, with most women experiencing attacks both during and outside of their menstrual cycle.3 There is a clinical impression that these migraines associated with the menstrual cycle are more severe, although the most recently published diary study showed that migraines were only slightly more painful, on average, during the first 2 days of menses than at other times during the cycle. It did not show differences in duration, number of symptoms, or disability. This suggests that these attacks may be no more difficult to treat than other migraines.5
Studies have shown that sumatriptan 6-mg injection and 100-mg tablets are effective in treating menstrually associated migraine pain when measured as a reduction of moderate or severe pain to mild or no pain. The selection of a primary endpoint in migraine treatment trials has evolved from headache relief to being pain free (ie, reduction of severe to moderate pain to no pain) as the primary endpoint.7 The International Headache Society now recommends that freedom from pain at 2 hours after treatment should be the primary measure of efficacy in clinical trials.8 Additionally, 87% of patients questioned in a recent study thought that the ability to provide freedom from pain was the most important attribute of a migraine medication.9
The pain in migraine is now thought to result from the following sequence of events: (1) primary dysfunction in the central nervous system, (2) dilation of blood vessels in the meninges, (3) activation of the trigeminal sensory nerves innervating these vessels, and (4) transmission of pain signals from the activated peripheral sensory nerves to central sensory neurons in the brain stem and upper cervical spinal cord.10,11 Neurobiologic research suggests that these central neurons may gradually become sensitized during an attack (termed central sensitization), which may be responsible for the progression of pain during an attack as well as and increased sensitivity to convergent stimuli from extracranial tissues. Such research provides a pathophysiologic rationale for the advantages of treatment at the first sign of migraine pain.
Based on the theory of central sensitization and the results of retrospective analyses of data from patients who had treated mild pain in clinical trials,12 two randomized, double-blind, placebo-controlled studies were undertaken to prospectively assess treatment at the first sign of migraine pain with sumatriptan. These studies found that sumatriptan (50-mg and 100-mg tablets), when administered at the first sign of mild pain, was significantly more effective than placebo in providing pain-free results at 2 hours after treatment (50 mg: 50% versus 29%, P < .001; 100 mg: 57% versus 29%, P < .001).12,13
Because migraines associated with menses often occur at a predictable time they are excellent targets for acute therapeutic strategies that initiate treatment during the mild pain phase. When the menstrually associated migraines in two recent studies were analyzed,13 there was evidence of pain-free results at 2 hours after treatment (50 mg: 45%; 100 mg: 50% versus placebo 24%, but the number of menstrually associated migraine attacks was too small [data on file, GlaxoSmithKline protocols SUM40274 and SUM40275] for statistical analysis). The purpose of this study was to determine the efficacy of sumatriptan 50 mg and 100 mg in menstrually associated migraine, when treatment is administered during the mild pain phase and pain-free relief is the criterion for treatment success.
PATIENTS AND METHODS
Women aged 18–65 years were eligible for participation in this study. They were required to have a minimum 1-year history of migraine with or without aura as defined by International Headache Society criteria, the ability to distinguish between migraine attacks and other types of headaches, and a minimum 6-month history of regularly occurring menstrually associated migraines. Menstrually associated migraine was defined as any patient-identified migraine beginning on day 32 through day 4, with day 1 being the first day of flow; this was termed the perimenstrual period. To increase the likelihood of patients having a menstrually associated migraine during the study, patients were eligible only if they reported regularly occurring menstrually associated migraines and reported menstrually associated migraine in at least two of their last three perimenstrual periods before screening. Patients had to report that they typically had moderate to severe menstrually associated migraine pain preceded by a mild pain phase. Patients reporting consistent day 32, day 31, or day 1 menstrually associated migraine were required to report the ability to predict onset of menstrual flow to within 1 day. This requirement was intended to ensure that a high percentage of attacks treated were menstrually associated. Patients who had used triptans, as well as patients who had never treated a migraine with a serotonin agonist, were eligible. Women of childbearing potential were required to use an adequate method of birth control.
Patients were not eligible for this study if they had confirmed or suspected ischemic heart disease (angina pectoris, history of myocardial infarction, documented silent ischemia), Prinzmetal angina, cardiac arrhythmia requiring medication or history of a clinically significant electrocardiogram abnormality, or congenital heart disease. Patients with a history of cerebrovascular pathology, including stroke, ischemic abdominal syndromes, peripheral vascular disease, or Raynaud syndrome, as well as patients with hypertension at screening (systolic blood pressure of at least 160 mm Hg or diastolic pressure of at least 95 mm Hg) were excluded. Patients with epilepsy or a brain lesion that might lower seizure threshold, basilar or hemiplegic migraine, or impaired hepatic or renal function could not enter the study. Patients could not have a tension-type headache for more than 15 days per month or more than six migraine attacks per month in either of the 2 months before screening. Patients were excluded if they were taking a monoamine oxidase inhibitor or an ergotamine-containing or ergot-type migraine prophylactic medication. Patients were not eligible if they were pregnant, breast-feeding, or sexually active and not using adequate contraceptive measures. Patients who had participated in an investigational drug trial within 4 weeks of screening could not enter the study.
We conducted a randomized, double-blind, placebo-controlled, parallel-group, single-attack outpatient study at 39 centers (28 in the United States, eight in Canada, and three in Puerto Rico). A central Institutional Review Board reviewed the protocol unless local Institutional Review Board review was required. The study was conducted in accordance with good clinical practice and all applicable regulations, including the 1996 version of the Declaration of Helsinki.
Patients who met inclusion and exclusion criteria and gave informed consent were randomly assigned in a 1:1:1 ratio to receive sumatriptan 50-mg tablets, sumatriptan 100-mg tablets, or placebo. A computer-generated randomization schedule was provided by the Medical Data Sciences Division of GlaxoSmithKline. All tablets were visually indistinguishable. Bottles containing study drug tablets were identified with the subjects' treatment numbers. These numbers were linked to the treatment identity through the randomization code, which was maintained by the sponsor and not shared with investigators or subjects. The bottle labels contained a hidden portion which identified the study drug if the information was needed in an emergency. Investigators were instructed to inform the sponsor if the blind was broken but not to reveal the treatment identity to the sponsor representative.
After randomization, patients had up to 3 months to treat a single menstrually associated migraine. They were instructed that the onset of migraine pain should only be considered a new attack if they had been pain-free for at least 24 hours. They were asked to treat the attack within 1 hour of the onset of pain but only if the pain was mild at onset and only if the pain was still mild at time of treatment. Pain was rated by subjects on a four-point scale that is the standard used in migraine trials (0 = no pain, 1 = mild pain, 2 = moderate pain, and 3 = severe pain). Patients who treated migraines that were not menstrually associated because of a late menstrual period, for example, were not allowed to remain in the study to attempt to treat a second migraine. After treatment of the migraine (whether menstrually associated or not), patients returned to the clinic within approximately 2 weeks for a final visit.
Patients were allowed to continue to take migraine prophylactic medications (other than ergotamine-containing or ergot-type medications such as dihydroergotamine or methysergide) they were taking at screening provided they had been on a constant regimen for at least 1 month before screening. The regimen was to remain constant throughout the study. Patients were also restricted from taking analgesics, antiemetics, or non-serotonin-agonist acute migraine medications within 6 hours before and 2 hours after the first dose of study medication. Rescue medication, including a second double-blind dose of study medication, was allowed if patients did not achieve freedom from pain at 2 hours or did achieve freedom from pain at 2 hours but then had a recurrence of moderate or severe pain 2–24 hours after the initial dose of study drug.
Predose evaluations included patient history, vital signs (heart rate and blood pressure), physical examination, and HIT-6 (QualityMetric, Inc, Lincoln, RI) disability questionnaire.14–16 After dosing, patients recorded the following information in a diary: date and time of study drug administration, including a second double-blind dose, if applicable; first day of flow for the perimenstrual period in which they treated a migraine with study drug; date and time of onset of pain or time of awakening if pain was present upon awakening; migraine pain severity (0 = none, 1 = mild, 2 = moderate, 3 = severe) at onset, immediately before initial dose of study drug and at 30 minutes, 1 hour, and 2 hours after this dose; whether there was a return of pain 2–24 hours after the initial dose of study drug and the date and time of the recurrence as well as severity of pain at time of recurrence; presence of aura with the migraine; presence or absence of nausea, vomiting, photophobia, or phonophobia at all times for which pain severity was assessed; ability to perform work or usual activities (normal, mildly impaired, moderately impaired, severely impaired, required bed rest) at all times for which pain severity was assessed; and use of additional medications. Adverse events were monitored from the time of administration of study drug until discharge from the study.
The primary efficacy measure was the percentage of patients who had pain-free relief 2 hours after treatment of a menstrually associated migraine with sumatriptan 100 mg compared with placebo. The same comparison using sumatriptan 50 mg was a secondary efficacy measure. Sample size was based on the assumed response to the 50-mg tablet. Based on post-hoc analysis of data from two other studies (data on file, GlaxoSmithKline Protocols SUM40274 and SUM40275), it was estimated that 50% of subjects treating a menstrually associated migraine in the sumatriptan 50-mg treatment group would achieve freedom from pain at 2 hours versus 25–30% of subjects in the placebo group. With alpha = 0.05 and 90% power to detect a difference, approximately 100 subjects treating a menstrually associated migraine in each of the three treatment groups were needed. The study was not powered to detect a difference between sumatriptan 50 mg and sumatriptan 100 mg.
The population of patients who treated a menstrually associated migraine with study drug and provided an evaluation of treatment (menstrually associated migraine population) comprised the primary population for assessment of efficacy, although the analysis was also performed using the patients who treated any migraine and provided a post-treatment evaluation (intent-to-treat population). Tolerability was analyzed using the population of all patients who took study drug (safety population).
The overall alpha was controlled at 5% for the following comparisons of sumatriptan 100 mg versus placebo using a closed, sequential approach in the order listed: (1) freedom from pain at 2 hours, (2) photophobia and phonophobia at 2 hours, (3) freedom from pain and associated symptoms at 2 hours, (4) sustained pain-free response 2–24 hours after treatment, and (5) ability to perform work or usual functions at 2 hours. If at any point statistical significance was not achieved, subsequent comparisons could not be made for inferential purposes. For any statistically significant endpoint, sumatriptan 50 mg was also compared with placebo. All sumatriptan versus placebo comparisons were made using the Cochran-Mantel-Haenzel test controlling for center.
The disposition of subjects and composition of populations for statistical analysis is shown in Figure 1. Unless otherwise noted, the reference population is the menstrually associated migraine population because this group was predetermined for analysis of efficacy. Only 19 of 368 patients in the intent-to-treat population were excluded from the menstrually associated migraine population, with the result that the data are very similar for these populations (Figure 1). The menstrually associated migraine population consisted of 349 subjects (300 was the protocol-specified target), and these subjects were evenly distributed among treatment groups.
Table 1 shows a summary of demographic and baseline characteristics, which are representative of a general migraine population. Treatment groups were similar with respect to age and race. Most patients had migraine without aura. The fewest patients had a diagnosis of migraine with aura, and an intermediate number had migraine both with and without aura. Treatment groups were similar with respect to average number of migraines and menstrually associated migraines per month. Treatment groups consisted of similar percentages of patients currently (within the past 3 months) using serotonin-type drugs to treat menstrually associated migraines, reducing the possibility of an imbalance in triptan responders between groups. Approximately half the patients in each treatment group had used serotonin-type drugs for acute treatment of their menstrually associated migraines in the 3 months before screening, and approximately 30% of patients in each treatment group had used sumatriptan specifically. Sumatriptan responsiveness was similar between the groups. Consistent mean HIT-6 scores greater than or equal to 60 across treatment groups suggest that migraine was having a similar and significant impact on these patients' ability to function at work, school, home, and in social situations.
In the intent-to-treat population, 97% (118 of 122), 94% (115 of 122), and 94% (116 of 124) of patients in the placebo, sumatriptan 100-mg, and sumatriptan 50-mg treatment groups, respectively, treated a menstrually associated migraine. Compliance with protocol treatment requirements within the menstrually associated migraine population was very good with 84% (99 of 118), 85% (98 of 115), and 81% (94 of 116) of patients in the placebo, sumatriptan 100-mg, and sumatriptan 50-mg treatment groups treating a menstrually associated migraine that was mild at onset and at the time of treatment and treating the attack within 1 hour of onset. The median time to treatment was 26, 30, and 30 minutes for the placebo, sumatriptan 100-mg, and sumatriptan 50-mg treatment groups, respectively.
Sumatriptan was superior to placebo in providing patients with pain-free relief at 2 hours as shown by the results presented in Table 2. Significantly more patients had pain-free relief at 2 hours in the sumatriptan 100-mg group (61%) and sumatriptan 50-mg group (51%) compared with the placebo group (29%) (both P < .001). Beginning at 1 hour after treatment, sumatriptan 100 mg was more effective than placebo at providing pain-free relief (30% versus 14%, P = .006). The results for the intent-to-treat and menstrually associated migraine populations were similar and are also presented in Table 2.
Significantly more patients in the sumatriptan 100-mg (31%) and 50-mg (30%) treatment groups had a sustained pain-free response compared with 14% of patients in the placebo group (100 mg versus placebo P = .004; 50 mg versus placebo P = .007). This means that the patients were pain free at 2 hours, remained pain free until at least 24 hours after treatment, and did not take a second dose of study drug or other medication for migraine pain.
Sumatriptan was also significantly better than placebo at eliminating other symptoms associated with migraine. Phonophobia and photophobia were absent in significantly more patients at 2 hours after taking sumatriptan 100 mg (73%) and sumatriptan 50 mg (69%) compared with the placebo group (53%) (100 mg versus placebo P = .002; 50 mg versus placebo P = .025). Nausea was absent 2 hours after taking sumatriptan 100 mg and 50 mg in 80% and 75% of subjects, respectively, compared with 71% of subjects in the placebo group. No statistically significant differences were observed for nausea, but nausea was reported by few patients at baseline. Also, less than 4% of subjects reported vomiting at baseline, so there were too few subjects with vomiting to analyze the results.
As shown in Figure 2, significantly more patients in the sumatriptan 100-mg (51%) and sumatriptan 50-mg (47%) treatment groups were migraine free (free of pain and associated symptoms) at 2 hours compared with patients in the placebo group (25%) (both, P < .001).
The extent of exposure to study drug was similar between treatment groups, with each group taking a mean of 1.4 ± 0.5 doses. The number of adverse events was low for sumatriptan 100 mg and 50 mg, and both doses were generally well tolerated. The percentage of patients in the safety population who had at least one adverse event was low (placebo group 7%, sumatriptan 50-mg group 8%, sumatriptan 100-mg group 16%). Even fewer subjects had an adverse event, as assessed by the investigator, that was related to the study drug (placebo group 3%, sumatriptan 50-mg group 3%, sumatriptan 100-mg group 11%). Few individual drug-related adverse events occurred at a rate of 2% or higher and with greater frequency in the sumatriptan treatment groups than in placebo group, as shown in Table 3. Only 2% of subjects in each of the three treatment groups reported an adverse event of severe intensity. No serious adverse events, eg, life-threatening or fatal events or events resulting in hospitalization, were reported. A single patient in the sumatriptan 50-mg treatment group had a cardiovascular adverse event (tachycardia) that was judged by the investigator as not related to study drug.
The results of this prospective study indicate that sumatriptan 50 mg and 100 mg effectively treat menstrually associated migraine when treatment is administered during the mild pain phase, and pain-free relief at 2 hours is the criterion for treatment success. Results of this study suggest that sumatriptan 100 mg may be more efficacious than 50 mg when used in the early treatment paradigm. Because this study was not powered to detect statistical differences between active doses, further studies to investigate this finding are warranted. The results of this study are in close agreement with those of a recent clinical trial with similar design assessing the efficacy of sumatriptan in menstrually associated migraine (SUM40282); it was completed at 53 centers in 12 countries, not including the United States (Savani N, Loftus J, Boswell D. A randomized, double blind, placebo controlled, parallel group evaluation of sumatriptan 50 mg and 100 mg tablets administered during the mild pain phase of a menstrually associated migraine attack (Abstract PB. 30). Cephalalgia 2002;22:606). In that study, the percentage of patients in the menstrually associated migraine population who had freedom from pain at 2 hours was 51% (77 of 133) and 58% (71 of 138) for the sumatriptan 50-mg and 100-mg groups, respectively, versus 22% (29 of 132) for the placebo group. The difference was statistically significant for both comparisons (P < .001).
The magnitude of pain-free relief at 2 hours in patients who treated menstrually associated migraine (50 mg 51%, 100 mg 61%, placebo 29%) in this study is similar to that obtained in the two prospective trials for patients who treated non-menstrually associated migraines at the mild pain stage. In those studies, 50% (n = 233; 50 mg) and 57% (n = 222; 100 mg) of sumatriptan-treated patients were pain-free at 2 hours compared with 29% (n = 236) of placebo-treated patients.13 This comparison suggests that sumatriptan is as effective in producing freedom from pain in menstrually associated migraine as it is in migraine occurring outside the perimenstrual window.
Of the 417 patients enrolled, only 22 did not have the opportunity to treat a menstrually associated migraine within 3 months of screening, and of the 369 patients who did treat one, only 20 did not treat a menstrually associated migraine. This indicates, at least in this sample of patients, that menstrually associated migraine is a regular, predictable event. The predictable nature of menstrually associated migraines makes them excellent candidates for an acute treatment strategy initiated at the first sign of migraine pain. All patients who were enrolled in this study had to report that they typically had moderate to severe menstrually associated migraine pain preceded by a mild pain phase. If this progression can be generalized to the population of women who have menstrually associated migraine, treatment at the first sign of migraine pain is a strategy particularly well suited to these attacks.
Treatment was generally well tolerated in this study. The incidence of adverse events was low, and the percentage of patients who had adverse events judged by investigators to be related to the study drug was similar for placebo and sumatriptan 50 mg. The 100-mg dose was also generally well tolerated. Few individual adverse events occurred at a rate of 2% or greater and with greater frequency in the sumatriptan-treatment groups than in the placebo group.
Given the results of this study, physicians should consider recommending that their patients treat menstrually associated migraines with sumatriptan while the pain is mild. Sumatriptan 50-mg and 100-mg tablets were generally well tolerated and effective in providing pain-free relief and relief of the associated symptoms of menstrually associated migraine, if administered when the pain was still mild.
1. Lipton RB, Diamond S, Reed M, Diamond M, Stewart W. Migraine diagnosis and treatment: Results from the American Migraine Study II. Headache 2001;41:638–45.
2. Kornstein SG, Parker AJ. Menstrual migraines: Etiology, treatment, and relationship to premenstrual syndrome. Curr Opin Obstet Gynecol 1997;9:154–9.
3. MacGregor EA. “Menstrual” migraine: Towards a definition. Cephalalgia 1996;16:11–21.
4. Johannes C, Linet M, Stewart W, Celentano DD, Lipton RB, Szklo M, et al. Relationship of headache to phase of the menstrual cycle among young women: A daily diary study. Neurology 1995;45:1076–82.
5. Stewart WF, Lipton RB, Chee E, Sawyer J, Silberstein SD. Menstrual cycle and headache in a population sample of migraineurs. Neurology 2000;55:1517–23.
6. MacGregor EA. Menstruation, sex hormones, and migraine. Neurol Clin 1997;15:125–41.
7. Salonen R, Saiers J. Sumatriptan is effective in the treatment of menstrual migraine. Cephalalgia 1999;19:16–9.
8. Tfelt-Hansen P, Block G, Dahlof C, Diener HC, Ferrari MD, Goadsby PJ, et al. Guidelines for controlled trials in migraine: Second edition. Cephalalgia 2000;20:765–86.
9. Lipton RB, Stewart WF. Acute migraine therapy: Do doctors understand what patients with migraine want from therapy? Headache 1999;39:S20–6.
10. Burstein R, Cutrer MF, Yarnitsky D. The development of cutaneous allodynia during a migraine attack: Clinical evidence for the sequential recruitment of spinal and supraspinal nociceptive neurons in migraine. Brain 2000; 123:1703–9.
11. Burstein R, Yarnitsky D, Goor-Aryeh I, Ransil BJ, Bajwa ZH. An association between migraine and cutaneous allodynia. Ann Neurol 2000;47:614–24.
12. Cady RK, Sheftell F, Lipton RB, O'Quinn S, Jones M, Putnam DG, et al. Effect of early intervention with sumatriptan on migraine pain: Retrospective analyses of data from three clinical trials. Clin Ther 2000;22:1035–48.
13. Winner PW, Mannix LK, McNeal S, O'Quinn S, Metz A. Treatment of migraine at the first sign of pain: prospective, double-blind, placebo-controlled, multicenter studies of sumatriptan 50mg and 100mg versus placebo. Neurology 2002; Suppl 3:A415.
14. Garber WH, Kosinski M, Dahlof C, Tepper S, Kujawski SC, Ware J, Batenhorst A. HIT-6 reliably measures the impact of headache. Cephalalgia 2001;21:333.
15. Ware J, Kosinski M, Dahlof C, Dowson A, Garber WH, Bayliss MS, et al. Validity of HIT-6, a paper-based short form for measuring headache impact. Cephalalgia 2001; 21:333.
16. Ware JE, Bjorner JB, Kosinski M. Practical implications of item-response theory and computerized adaptive testing. Med Care 2000;38:73–82.