Hormonal Interventions for Menstrual Migraines

Kathleen J. Chavanu, PharmD, Dannielle C. O'Donnell, PharmD


Pharmacotherapy. 2002;22(11) 

In This Article

Hormone-Related Therapies For Menstrual Migraines

Oral Contraceptives

Despite the controversy concerning the therapeutic role of oral contraceptives and the various effects of these agents on migraines and migraineurs, some women may improve with this therapy.[26,27,28] In keeping with the theory that estrogen withdrawal and fluctuations in estrogen levels during the luteal phase are the major contributing factors to menstrual migraines, maintaining a low level of estrogen for an extended duration of time may alleviate or improve menstrual migraines. Studies have addressed the acceptance of extended-duration oral contraception use in women and the need for regimens that delay the pill-free week.[15,16,17,18,19]

An open-label, randomized, crossover study investigated the effect of oral contraceptives on migraine headaches.[15] The study involved 40 women with histories of moderate or severe migraine with associated nausea, vomiting, photophobia, sensory symptoms, or a combination of these symptoms, which usually are not related to true menstrual migraine. Twenty patients were randomized to receive an oral contraceptive for 2 months, whereas the second group of 20 patients did not receive hormonal treatment. After 2 months, the second group received hormonal treatment and the first group received no hormonal treatment. Hormonal treatment consisted of a monophasic combination oral contraceptive pill (norgestrel 0.5 mg-ethinyl estradiol 0.05 mg [Ovral]). Patients kept a self-rating card to record the number and severity of headaches, and the amount, kind, and success of therapy for each attack.

The results showed an increased frequency of moderate and severe headaches during the 2 months in which patients received oral contraceptives.[15] Frequency of mild headaches also increased, but to a lesser extent than moderate and severe headaches. Twelve of the 40 patients reported improvement of their headaches while taking the oral contraceptive. Twenty-eight patients reported worsening of their headaches while receiving therapy. This study identified three distinct groups of patients: those with migraine headaches aggravated by oral contraceptives, those whose migraines decreased in severity or frequency when taking oral contraceptives, and those whose migraines were unaffected by oral contraceptives.

Another study, which involved 262 women receiving oral contraceptives, focused on hormone withdrawal symptoms, including menstrual migraines, over a 9-month period.[16] Three groups of participants were recognized: 26 women who previously had not taken oral contraceptives, 43 women who had taken oral contraceptives but had not done so during the past 3 months, and 193 women who had been taking oral contraceptives for at least the past 12 months. Participants recorded their symptoms in daily diaries. All women were started or maintained on a combination low-dose oral contraceptive. This consisted of 21 active pills containing ethinyl estradiol 35 µg or less plus a progestin, and seven hormone-free pills. The women involved in the study were similar in age, height, weight, body mass index, race, and education. Participants were primarily Caucasian (84%) with high school education or greater (97%). The most common reasons for taking oral contraceptives were birth control and menstrual disorders. Outcomes investigated were headaches, pelvic pain, consumption of drugs for pain, menstrual flow, and symptoms such as nausea, vomiting, bloating, swelling, and breast tenderness. Patients subjectively scored their headaches on a scale of 0-10 (0 = no headache at all; 10 = worst headache). Headache data were split into two measures for analysis: any headache (pain score 1-10), and moderate-to-severe headache (pain score 5-10).

Women who had been taking oral contraceptives before the start of the study complained of more symptoms during the pill-free interval than during the active-pill interval.[16] Among this group significant differences emerged with regard to pelvic pain (70% vs 21%, p<0.001), headaches (70% vs 53%, p<0.001), consumption of drugs for pain (69% vs 43%, p<0.001), bloating or swelling (58% vs 19%, p<0.001), and breast tenderness (38% vs 16%, p<0.001) for the pill-free interval compared with the active-pill interval. The women who previously had not taken oral contraceptives experienced an increased number of headaches during the hormone-free interval of the second monitored cycle. During cycle 1, 62% of these women experienced headaches during the 3 active-pill weeks, compared with 71% during the hormone-free week (p=0.93). In cycle 2, the respective percentages were 49% and 71% (p<0.001), and in cycle 3, 51% and 60% (p=0.13).

Among women who had been taking oral contraceptives before the start of the study, the number of headaches increased during all hormone-free intervals.[16] In cycle 1, 53% experienced any headache during the 3 active hormone weeks compared with 70% during the hormone-free week (p<0.01). In cycle 2, 45% experienced any headache during the 3 active hormone weeks, compared with 62% during the hormone-free week (p<0.01). As for the other outcomes, increased symptomatology occurred during the hormone-free interval. These data support the hypothesis of estrogen withdrawal and the association between the fall of estrogen and exacerbation of headaches.

Extended Dosing. Another research group studied the cycle control and tolerance of an extended 7-week cycle of low-dose oral contraceptives in 55 healthy women from the Netherlands.[17] Participants ranged in age from 18-40 years (average 27 yrs), and all had received a standard 3-week cycle of low-dose oral contraceptives for the past 6 months. All patients received gestodene 75 µg-ethinyl estradiol 30 µg (Minulet) for 7 weeks and were given a diary to record occurrences of spotting, breakthrough bleeding, time of dosing, and side effects. Eight women reported the following side effects: breast tenderness (4 patients), nausea (2), bloating (2), headache (1), and dysmenorrhea (1). No association was found between time of day, side effects, and bleeding, although the study population was small. Analysis of patient satisfaction with extended-duration oral contra-ceptive therapy revealed that 66% of participants were extremely satisfied, 26% were satisfied, 2% were moderately satisfied, and 4% were not satisfied at all.

A 12-month study of a low-dose oral contraceptive (ethinyl estradiol 30 µg-desogestrel 150 µg) compared extended therapy (9 wks on, 1 wk off) with traditional therapy (3 wks on, 1 wk off).[18] This was a randomized, multicenter study involving 300 women, aged 18-39, with 200 women receiving extended treatment and 100 women receiving traditional treatment. All patients received a diary to record tablet intake, uterine bleeding, and side effects. Patients were seen after months 3 and 12, and patients completed a questionnaire after month 12. The treatment and control groups were similar with regard to age distribution, smoking habits, and oral contraceptive history.

One hundred fifteen women completed the extended treatment, and 64 women completed the traditional treatment.[18] All experienced withdrawal bleeding, defined as bleeding during the pill-free interval. Withdrawal bleeding in the extended-treatment group was unaltered from baseline, whereas in the traditional group, withdrawal bleeding seemed to diminish. The mean withdrawal bleeding time was shorter in the traditional group than in the extended-treatment group, but as expected, the total number of bleeding days during the study period was significantly less (p<0.001) in the extended-treatment group than in the traditional group. Breakthrough bleeding (bleeding occurring while taking the active pill) was significantly more frequent in the extended-treatment group than in the traditional group. Within the extended-treatment group, significantly more breakthrough bleeding occurred among women who were continuing their use of oral contraceptive than in new starters. No changes in blood pressure, hemoglobin, or body weight were reported between the two groups. No serious side effects or pregnancy occurred.

Eighty-three women, however, withdrew from the extended-treatment group, and 32 women withdrew from the traditional group.[18] The most common medical reasons for leaving the study were bleeding problems, weight changes, mood changes, and headache. Bleeding problems leading to withdrawal from the study were more prominent in the extended-treatment group than in the traditional group (26 vs 2 women, p<0.01). Conversely, more women from the traditional group than the extended-treatment group discontinued the study due to headache (9 vs 3 women, p<0.01). Other reasons for leaving the study were planning a pregnancy, end of need for contraception, loss of patient to follow-up, and noncompliance. Results from the questionnaire indicated that the only symptom that significantly differed between the two groups was headache. In the extended-treatment group, 9.7% of women complained of increased headache symptoms compared with 17.3% in the traditional group (p<0.05). Of the patients who had been taking oral contraceptives before enrollment in the study, 79% stated that they preferred the extended treatment to the traditional dosing regimen they had used before entering the study.

Building on the studies described above, another research group examined whether extended duration of oral contraceptives was effective and safe in a subset of patients with documented hormone-related symptoms, which included menstrual migraines.[19] This was a prospective analysis of 50 women taking oral contraceptives and experiencing menstrual-related problems. Patients ranged in age from 19-50 years. To be included in the study, women must have complained of at least one symptom during the pill-free week. Eighty-two percent of patients experienced two symptoms, and 38% experienced three symptoms. Complaints were categorized as migraines (76% of patients), dysmenorrhea (78%), menorrhagia (36%), premenstrual syndrome (32%), and other (6%). Menstrual migraine was ranked as the most severe symptom by 48% of patients.

Patients with documented complaints specific to the pill-free week started a continuous active pill for 6 weeks, concluding with a 7-day pill-free interval.[19] Those tolerating this extended regimen (absence of breakthrough bleeding, spotting, or other effects perceived as intolerable) were allowed to increase the interval of active pills by 3 weeks after each cycle. A stabilized patient on extended-duration oral contraceptives was defined as achieving maximum extended-cycle length (i.e., number of wks) without problems for at least two extended cycles. At each study visit, patients had the choice of continuing with extended cycles, returning to the standard 3-week cycle, or discontinuing treatment.

All patients were taking a combination low-dose oral contraceptive before study enrollment and continued on the same agent throughout the study.[19] Most subjects (58%) were taking a desogestrel-containing oral contraceptive, whereas 32% were taking a levonorgestrel-containing oral contraceptive, and 10% were taking a norethindrone-containing oral contraceptive. Twenty-six percent of patients discontinued oral contraceptives or returned to the standard method before becoming stabilized on an extended regimen. Thirty-seven women were stabilized on one of three extended regimens: a 6-week regimen (8 women), a 9-week regimen (13), and a 12-week regimen (16). All patients reported a delay of onset and decreased severity of their reported symptoms. After study completion, 54% continued on the extended-duration regimen for an average of 17.2 months.

These five studies[15,16,17,18,19] support further investigation into the efficacy, tolerance, patient acceptance, and need for extended-duration oral contraceptive pills, particularly to minimize menstrual migraine. As expressed in these studies, not all patients will respond to oral contraceptives, and not all patients are candidates for this type of hormonal treatment. Indeed, women new to oral contraceptives may experience worsening of headaches, as supported by findings from a retrospective study involving 451 women with histories of migraine headaches.[29] The study involved 295 participants who previously had taken oral contraceptives and had been diagnosed with migraine headaches. Of these patients, 201 reported no change in migraine status when taking oral contraceptives, whereas 74 patients reported worsening symptoms, and 24 patients reported improvement in migraine symptoms. The comparative analysis between those who worsened and those who improved indicated that these outcomes are unpredictable. This reaffirms that some women will benefit from oral contraceptive therapy, whereas others may not. However, patients who fail nonmenstrual migraine therapy and are naïve to oral contraceptives or are stable on oral contraceptive therapy may improve their menstrual migraine symptoms or reduce the number of attacks/year by starting low-dose monophasic estrogen oral contraceptive pills with an extended-duration regimen.

Estrogen Transdermal Patches

For women with menstrual migraines who cannot tolerate oral contraceptives or have a history of venous thromboembolism, estrogen transdermal patches may be an alternative choice of therapy. Estrogen transdermal patches can be applied to the skin to provide a stable level of plasma estrogen. Serum levels of estrogen rise within 4 hours after application of the patch and are proportional to the dose.

In contrast to the large amount of literature available on oral contraceptive therapy and migraine headaches, the literature on the role of estrogen transdermal patches in treatment of menstrual migraines is limited and not well documented. Two studies reported that estrogen transdermal patches provided minimal or no benefit.[20,21] A research group randomly assigned 20 patients (aged 38-48 yrs) with true menstrual migraines to treatment with either estradiol 50 µg (Estraderm TTS) patches or placebo patches for three consecutive menstrual cycles.[20] The treatment sequences were either estrogen-placebo-estrogen or placebo-estrogen-placebo. All eligible patients had a history of true menstrual migraine with no other attacks during the rest of the cycle, and all had regular menstrual cycles. Patients were allowed to consume aspirin or ergotamines as permitted, and prophylactic drugs were allowed if the patient had been taking them for at least 3 months before entering the study. Oral contraceptives had to be stopped 3 months before enrollment in the study. Patients maintained diaries to record the presence, duration, and severity of migraine attacks; consumption of analgesics; and adverse effects. Patients applied the patches 48 hours before the anticipated start of menstruation and again 4 days later.

Nineteen of the 20 patients completed the program.[20] Fifty-nine percent of patients receiving treatment complained of migraine attacks compared with 69% of those who received the placebo (p=0.11). No differences were noted for duration of migraine attack, severity of attack, or consumption of analgesic drugs. The number of patients who experienced migraine headaches was 10% higher for the placebo period than for the treatment period; however, the difference did not reach statistical significance, which may be due to the small sample size coupled with a number of protocol deviations (two patches applied earlier than 48 hrs before menses).

A double-blind, placebo-controlled, crossover study assessed the efficacy and safety of estradiol 50-µg patches during six consecutive cycles in patients with menstrual migraines.[21] The study involved 2 months of baseline assessment followed by 4 months of drug treatment. Patches were applied 2 days before expected onset of menses. Patients were randomized to either estradiol-placebo or placebo-estradiol. Patients kept diaries to track headache duration, headache intensity, and impairment through mood and pain. Fifty-eight patients were enrolled in the study, and 41 patients (aged 22-53 yrs) finished the program. No significant difference between estradiol and placebo was found in any of the parameters assessed; however, patients taking estradiol did slightly better than those receiving placebo with regard to the number of days of suffering from and the duration of migraines. However, patients who took estradiol before receiving the placebo ranked the placebo as slightly better than estradiol.[21]


Tamoxifen is a selective estrogen receptor modulator that binds to estrogen receptors. It causes estrogen antagonism by downregulation of the estrogen receptor, resulting in inhibition of messenger RNA transcription in various tissues. Estrogen receptors are present within the uterus, breast, brain, and bone. Tamoxifen is indicated in prevention and treatment of some breast cancers, and it has been evaluated in osteoporosis, prevention of coronary heart disease, and as a treatment for menstrual migraines.[5,7] Because of its interaction with estrogen receptors, tamoxifen has been evaluated in the treatment of refractory menstrual migraines. Although the exact mechanism of action by which tamoxifen might prevent menstrual migraines is unknown, case studies suggest that it may benefit some women.

In a study prompted by one patient's reported menstrual migraine improvement while taking tamoxifen citrate for premenstrual syndrome, a research group[22] identified an improvement in menstrual-related migraines in seven of eight women taking tamoxifen in a short-term, open-label evaluation. Participants complained of hemicranial pain with nausea and vomiting that occurred 3-7 days before menses. Earlier unsuccessful migraine prevention and treatment measures were not mentioned; however, these patients reported that their headaches benefited at times from caffeine intake. Patients kept records of menstrual cycles, headaches, and premenstrual syndrome symptoms. Evaluations occurred monthly for four consecutive cycles. Patients received tamoxifen citrate in dosages of 10-20 mg/day for 7-14 days before menstruation; during menses, they received tamoxifen citrate 5-10 mg for 3 days. Of the eight women, five had marked improvement or disappearance of headaches, two had mild-to-moderate improvement, and one patient was unchanged. Six patients noted improvement of their premenstrual complaints. Adverse effects included minor menstrual irregularities during three of the four cycles.

Few published studies have evaluated tamoxifen citrate as a treatment for menstrual migraines, and there are no placebo-controlled studies in the literature. The practice of prescribing tamoxifen for the treatment of menstrual-related migraines was published in 1986 by a physician who was familiar with administering tamoxifen citrate for relief of tenderness, pain, and nodularity in women with benign mammary dysplasia.[23] He noticed that these women sometimes complained of headaches. He stated that six women with histories of migraine-type headaches completely or substantially reduced the frequency and severity of their headaches, to the extent of not needing their usual migraine drug therapy. The women admitted that after stopping tamoxifen citrate, their headaches returned within a few weeks.

The same physician documented improvement in a 50-year-old perimenopausal woman with a family history of breast cancer who was referred to him for painful, tender, nodular breasts in April 1984.[23] She had suffered from migraine headaches since 1981. He treated her with tamoxifen citrate 20 mg/day. Although she complained of hot flashes for 5-6 weeks, her migraines completely resolved. In addition, her breast pain and nodularity improved. In August 1985, she discontinued the tamoxifen citrate and her headaches returned. In November 1985, the breast nodularity recurred and tamoxifen was restarted. Six weeks later the patient's headaches were milder, and she could control her headaches with aspirin alone. At her last reported interaction with the physician in September 1986, the patient was still taking tamoxifen citrate and no longer required any drug therapy for her migraine headaches. She was without complaints of breast discomfort.

Raloxifene, a selective estrogen receptor modulator that is newer than tamoxifen, is available for prevention and treatment of osteoporosis in postmenopausal women. To date, there have been no studies or case reports pertaining to raloxifene as a treatment for either true menstrual migraines or menstrual-related migraines. Both tamoxifen and raloxifene are selective estrogen receptor modulators, but they vary in their affinity for these receptors in different tissues. Indeed, unlike tamoxifen, raloxifene is not associated with endometrial hyperplasia. Evaluation of raloxifene in women with menstrual migraines is warranted to determine whether a benefit exists. Controlled studies of tamoxifen in menstrual migraines are also needed.


Unlike the therapies described above, danazol is an androgen derivative that downregulates estrogen receptors, thereby suppressing the pituitary-ovarian axis, inhibiting ovarian steroidogenesis, and preventing the rise of both estrogen and progesterone during the luteal phase of the menstrual cycle. Averting the rise in estrogen levels is of key importance in menstrual migraines; for that reason, danazol has been evaluated as a possible treatment modality. The agent is indicated for treatment of endometriosis and cystic diseases of the breast.

Only a single prospective study has addressed the efficacy of danazol in the control of "hormonal migraine."[24] The study's investigators defined hormonal migraine as a headache occurring from 7-10 days before menstruation through the second day of menses. The study involved 131 women (aged 20-51 yrs) with hormonally related migraines that were unresponsive to standard drug therapy. Participants maintained daily diaries to verify headache occurrence, frequency, and intensity (as rated on a 4-point scale), and to document drug therapy to relieve headaches. The study consisted of four phases; each of the first three were 2 months long, and the fourth phase lasted 6 months. Phase I consisted of dietary restriction with addition of acetazolamide 125 mg/day after the first month. In phase II, danazol 200 mg twice/day was added to the treatment regimen for 25 days/month, beginning on the third day of menses. In phase III, danazol was discontinued, although diet and acetazolamide were continued. In phase IV, danazol 200 mg twice/day was restarted for 25 days/month in those who had improved while receiving danazol in phase II and whose headaches had worsened while in phase III (diet and acetazolamide alone).

The first 6 months of the study were completed by 131 patients.[24] Eighty-three patients reported a 75% decrease (which was the authors' a priori definition of treatment success) in headache index (the monthly sum of daily headaches as graded on a 1-4 severity scale). Twenty-seven patients had no improvement, and 21 patients withdrew from the study due to side effects from danazol. The percentage of women obtaining headache relief was greater among women over 40 years compared with women younger than 30 (75% vs 31.8%, p=0.0165). According to patients' reports of "helpfulness," danazol was most effective for women whose migraines occurred 1 week before menses or during menstruation. All 83 women whose migraines were improved by danazol reported a return of migraines during phase III. Sixty-seven (82.6%) reported positive outcomes (headaches relieved and no side effects) during phase IV. Fourteen patients experienced side effects, two of which were severe and included joint pain and acne.

This study included women with migraines associated with ovulation and cessation of menses; that is, it was not limited to true menstrual migraineurs.[24] Therefore, the results may not be directly comparable to those of studies using the more narrow definition of menstrual migraine.

This prospective study[24] was preceded by two case reports describing the responsiveness of migraines to treatment with danazol. One report described the successful prevention of migraines with danazol in a 38-year-old woman with a 17-year history of migraine with aura.[30] She had failed treatment that had consisted of diet limitation, chlordiazepoxide, amitriptyline, propranolol, isometheptene, cyproheptadine, ergotamine tartrate, and biofeedback. She was instructed to take danazol 200 mg every 1-3 hours as needed, up to a maximum of 600 mg/day during her menses or when her premonitory signs of attack occurred. The patient reported decreased emotional lability and edema, and she had no migrainous headache pain or side effects other than decreased menstrual flow during her three treatment cycles. She then discontinued danazol and experienced a severe migraine attack. Such attacks did not occur when she restarted the regimen the next month.

Similarly, another case report described positive results with danazol in a 21-year-old woman with menstrual migraines and endometriosis.[31] The patient's migraine headaches, which occurred 2-3 times/month, began when she started oral contraceptives. She discontinued the oral contraceptives due to the headaches, and their frequency decreased to only one/month before her menses for the next 2 years. During this time, she developed menstrual difficulties and was diagnosed with pelvic endometriosis. As a result, she was started on danazol 200 mg 4 times/day. For the next 2 years, her menstrual problems and her headaches disappeared. When the patient discontinued danazol, her migraine headaches returned. The headaches then were treated successfully with ibuprofen and a caffeine-ergotamine preparation.

These two case reports laid the foundation for the prospective, open-label study described above.[24] However, the case reports[30,31] involved higher doses of danazol. In one of the case reports,[30] the patient experienced migraine with aura, which is generally not encountered in menstrual migraine. The patient also was instructed to take the drug during menses or ovulation, whenever her aura occurred. No details were provided about the timing of her migraines with regard to her cycle or dosing, but certainly, these may not have been true menstrual migraines as they are more narrowly defined. The other case report[31] suggests that danazol in dosages up to 200 mg 4 times/day was efficacious and well tolerated for an extended duration of 2 years in one patient.

Gonadotropin-Releasing Hormone Agonists

A gonadotropin-releasing hormone agonist, such as leuprolide, may be an alternative for menstrual migraineurs who have failed the more conservative hormonal interventions discussed above. These agents work by inducing a reversible, medical ovariectomy. Initially, these agents are stimulatory, causing release of large amounts of luteinizing hormone and follicle-stimulating hormone. At first, this results in a surge of estradiol and worsening of symptoms. However, with continued exposure to the gonadotropin-releasing hormone agonist, pituitary gonadotropin-releasing hormone receptors are downregulated, levels of luteinizing hormone decrease, and follicle-stimulating hormone and estradiol levels begin to fall as well. This is otherwise known as hypogonadotropic hypogonadism. Leuprolide is indicated for the treatment of prostrate cancer, endometriosis, uterine leiomyomata, and precocious puberty.

A nonrandomized, prospective outpatient treatment study assessed the efficacy of gonadotropin-releasing hormone therapy, both alone and combined with continuous estrogen-progestin add-back therapy, for women with severe menstrual migraine headaches.[25] Participants were five women with regular menstrual cycles and a history of menstrual migraines (90% of attacks occurring during menses or 2 days before menses) for more than 1 year. All had experienced severe migraines that they felt had limited their quality of life. Their headaches had failed to respond to other medical treatments (analgesics, abortive migraine preparations, and drugs for migraine prophylaxis). They were not taking any exogenous hormones. The women had no underlying gynecologic complaints and agreed to nonhormonal contraception for the duration of the study. Participants kept a diary throughout the trial to rate each day's headache experience using a score of 0 (absent) to 3 (severe). They were allowed to take previously prescribed migraine drugs. A cumulative monthly headache score was calculated for each patient, which was the sum of all daily scores during a 28-day interval. After 2 months of baseline charting, the women received intramuscular leuprolide acetate 3.75 mg for 2 months, followed by leuprolide plus add-back hormonal therapy (transdermal estrogen 0.1 mg/day and oral medroxyprogesterone acetate 2.5 mg/day) for an additional 6 months.

All patients completed the study, and all experienced a hypogonadal state. A marked reduction in mean cumulative monthly headache score was reported with treatment, and their scores did not change with the addition of add-back therapy. The mean monthly cumulative scores were 15.3 ± 2.4 at baseline, 4.0 ± 1.5 during treatment with leuprolide alone (p<0.001), and 3.0 ± 0.6 during treatment with leuprolide plus add-back therapy. The only month that did not show a significant decrease in headache score was the first month of treatment. During that month, two patients had increases in their scores, one was unchanged, and two had decreases in their scores. This was expected, considering that these agents are initially stimulatory. Patients' global assessments of therapy were positive. All described dramatic improvements in daily functioning and quality of life, as well as decreased consumption of analgesic drugs. Two patients stopped their daily prophylaxis for migraine during the study period. Three of the patients remained on the therapy after the study's end date. Of note, the other two patients reported an increased frequency of headaches when they discontinued therapy.[25]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.