A novel specific prophylaxis for menstrual-associated migraine.

Objectives: Few migraine prophylactic therapies have demonstrated a 50% reduction in headaches. Even when successful, the economic burden of prophylaxis can discourage widespread usage. This article presents a pilot study of a novel, effective, specific, and inexpensive prophylactic strategy for menstrual-associated migraine.

Materials and Methods: Eleven women with menstrual-associated migraine and fewer than 14 days of headache per month were identified from prospective enrollment at a gynecology practice and retrospective chart review at a headache center. Exclusion criteria included current use of prophylactic therapy for migraine.

Methods: Patients received open-label therapy with an oral contraceptive containing 20 [micro]g ethinyl estradiol on days 1 to 21, supplemented with 0.9 mg conjugated equine estrogens on days 22 to 28. Headache intensity and bleeding were recorded in diaries that plotted headache days by oral contraceptive pill days.

Results: All of the patients achieved at least a 50% reduction in number of headache days per cycle (mean 77.9% reduction); 10 of the 11 women achieved at least a 50% reduction in weighted headache score (mean 76.3% reduction).

Conclusions: All currently available estrogen-containing oral contraceptives produce a premenstrual fall in ethinyl estradiol concentration equal to or greater than 20 [micro]g. Estrogen supplementation during the placebo week can reduce the magnitude of this fall to less than 20 [micro]g. When the decline is limited to the equivalent of 10 [micro]g ethinyl estradiol, menstrual-associated migraine is prevented. At an average cost of six dollars per headache-day prevented, this represents an effective and inexpensive strategy for a common migraine trigger.

Key Words: estrogen, hormone therapy, menstrual migraine, prophylaxis

**********

The predominance of migraine in reproductive-aged women and its associated clinical, social, and economic burden make it one of the most important medical issues in women's health. By the conclusion of reproductive years this common headache disorder has affected up to 40.9% of women. (1) With up to 70% of female migraineurs noting relationship of their migraines to menses, menstruation is arguably the most common migraine trigger of all. (2)

Few migraine prophylactic therapies demonstrate a 50% reduction in headaches. Even when successful, the economic burden of prophylaxis can be substantial, ranging as high as $138 per headache-day prevented. (3) This article presents a small open-label pilot study of a novel specific therapy for prevention of menstrual-associated migraine (MAM) that is highly effective, inexpensive, and well-tolerated.

Hormonal-associated migraine occurs in many settings. It may occur exclusively with menses and be the only migraine that a woman experiences: the so-called "true" menstrual migraine. It may be predictable, but only one of several migraines in a month: the "menstrual-associated" migraine. Migraines may occur with scheduled withdrawals from exogenous estrogen-containing products such as oral contraceptives (OCs), postmenopausal estrogen therapy, or hormone therapy. In addition, they may occur with unintentional estrogen withdrawal on these products. These withdrawals may be secondary to missed doses, delayed doses, or increased metabolism of the estrogen component of the OC via cytochrome P450 3A4 or 1A2 interactions.

Hormonal association of headaches is exceedingly common. In a population of 262 women seen in a gynecology clinic for refill or initiation of OCs, 70% experienced headache during the placebo week of their OC. (4) Reports have stated the prevalence of menstrual association in female migraineurs as high as 70%. (5)

[FIGURE 1 OMITTED]

In the natural menstrual cycle, the late luteal phase decline in estradiol concentration is equivalent to a decline of 20 to 25 [micro]g of the synthetic estrogen, ethinyl estradiol (EE) (Fig. 1). In a previous pilot study, (6) the author found that reducing the premenstrual fall in estradiol to the equivalent of 10 [micro]g EE or less was beneficial in preventing menstrual migraine. In that study, a variety of hormonal strategies were utilized, each of which accomplished the critical reduction in the magnitude of the premenstrual estradiol decline. This study investigates a single prophylactic strategy.

Materials and Methods

Subjects

Inclusion criteria included predictable migraine without aura (per diagnostic criteria of the International Headache Society) regularly associated with menses and with a total of fewer than 14 days of headaches per month. Subjects were identified by retrospective chart review at a headache center, and by prospective enrollment at a gynecology practice.

Exclusion criteria included pregnancy, lactation, current migraine prophylactic therapy, smoking, diabetes mellitus, uncontrolled hypertension, migraine with aura, or known thrombophilic state. Eleven patients were identified, with an age range of 28 to 50 years (mean age 41). Informed consent was obtained in accordance with ethical standards for human experimentation established by the Declaration of Helsinki of 1975, revised 1983.

All subjects received open-label therapy with an OC containing 20 [micro]g EE, dosed at bedtime days 1 to 21 with 0.9 mg conjugated equine estrogens (CEE) at bedtime days 22 to 28 (in lieu of the seven placebo pills in the pill pack). This dose of CEE is considered to be equivalent to 10 [micro]g of synthetic EE.

[FIGURE 2 OMITTED]

All patients kept a 28-day menstrual cycle diary in which headache intensity as well as bleeding was plotted by pill day. The total number of headache days per cycle was tabulated, as was a headache intensity score. This score represented the sum of pain ratings for the preceding 28 days, with days of mild headache pain scored as "1", moderate as "2", and severe as "3".

[FIGURE 3 OMITTED]

[FIGURE 4 OMITTED]

After the second cycle of OC/CEE, patients were reevaluated. Total headache days and weighted headache scores were calculated from diaries and compared with baseline scores.

Results

Baseline status of the study participants and posttreatment results are summarized in the Table. All patients completed the protocol and achieved at least a 50% reduction in the number of headache days. Mean number of headache days per month was 7.6 at baseline, 1.6 after treatment (76.3% reduction). The average change in weighted headache score was a 77.9% decrease, with 10 (91%) of the 11 patients achieving at least a 50% reduction from baseline scores.

Discussion

If we accept the current assumption that MAM is directly or indirectly triggered by the premenstrual decline in estrogen, there is no currently available OC that would be expected to help relieve menstrual migraine. Today's lowest-dose formulations produce a 20 [micro]g decline in EE concentration between pill days 21 and 22--equivalent to the physiologic fall experienced in the natural menstrual cycle (Fig. 1). The majority of available formulations allow supraphysiologic falls (30-50 [micro]g EE) and would thus be expected to worsen MAM (Fig. 2). (7)

Commercial formulations are available that add back 10 [micro]g EE for 5 days during the placebo week. However, they do so only after 2 days of inert pills (Fig. 3). With this formulation, patients experience the same premenstrual decline in estradiol concentration on pill day 22 as with the other 20 [micro]g formulations. Clinical experience shows these pills offer no advantage over traditional 20 [micro]g OCs, because they show equal potential for precipitating MAM.

It is only with the supplementation of estrogen from the beginning of the placebo week that the magnitude of the fall in estradiol is diminished adequately to avoid the MAM trigger. (Fig. 4)

This strategy was not only clinically effective in this small pilot study, but relatively inexpensive. At an average cost of $32 per cycle treated and an average reduction in total headache days from 7.6 to 1.6 per month, the mean cost per headache day prevented was six dollars. Although this alone might argue cost-effectiveness, it is important to note that 8 of the 11 women were already taking hormonal products of equivalent or greater cost at baseline. Five of the 6 women who were not on OCs had menstrual comorbidities (short follicular phase, menometrorrhagia, dysmenorrhea, and irregular cycles) which were successfully cotreated with the prescribed hormonal therapy.

Conclusion

A larger, placebo-controlled, double-blind extension of this pilot is warranted, as well as trials of this strategy in the patients more commonly seen in headache centers: women whose MAM is complicated by chronic daily headache.

A lie gets halfway around the world before the truth has a chance to get
its pants on.
--Sir Winston Churchill

Table. Baseline status and posttreatment results (a)

                             Baseline
             Hormone                  Headache       Menstrual
ID     Age   therapy          Score     days          factors

 1     48    None              7         3         23-day cycles
 2     42    Progesterone     18         8         Metrorrhagia,
               cream                                 dysmenorrhea
 3     42    30 [micro]g      21         7
               OC/0.625 CEE
 4     45    20 [micro]g OC   17        10
 5     39    30 [micro]g OC   25        12         Dysmenorrhea
 6     29    30 [micro]g OC   14         8
 7     48    CC-HRT           18        10         Bleeding cyclically
                                                     on CC-HRT
 8     35    None             14         9         Menorrhagia
 9     45    20 [micro]g OC   16        10
10     50    None              6         2         Menometrorrhagia,
                                                     dysmenorrhea
11     28    MPA injections   10         4
Mean   41                     15.1       7.6

                            Posttreatment
                      %       Headache        %          Menstrual
ID     Score        change      days        change        factors

 1      0            100         0          100        28 day cycles
 2      0            100         0          100        Normal menses
 3      2             90.5       1           85.7
 4      2             88.2       1           90
 5      5             80         2           83.3      Dysmenorrhea
                                                         resolved
 6      3             78.6       3           62.5
 7      5             72.2       3           70        Normal menses
 8      4             71.4       2           77.8      Normal menses
 9      5             68.8       3           70
10      2             66.7       1           50        Normal menses
11      6             40         2           50
Mean    3.1 (b)       77.9       1.6 (b)     76.3

(a) OC, oral contraceptive: CEE, conjugated equine estrogen; CC-HRT,
combined/continuous hormone replacement therapy: MPA,
medroxyprogesterone acetate.
(b) T < 0.0001.

Accepted March 11, 2004.

References

1. Launer LJ, Terwindt GM, Ferrari MD. The prevalence and characteristics of migraine in a population-based cohort: the GEM study. Neurology. 1999;53:537-542.

2. MacGregor A. Migraine associated with menstruation. Funct Neurol. 2000;15 Suppl 3:143-153.

3. Adelman JU, Adelman LC, Von Seggern R. Cost-effectiveness of anti-epileptic drugs in migraine prophylaxis. Headache. 2002;42:978-983.

4. Sulak PJ, Scow RD, Preece C, et al. Hormone withdrawal symptoms in oral contraceptive users. Obstet Gynecol. 2000;95:261-266.

5. Silberstein SD. Menstrual migraine. J Womens Health Gend Based Med. 1999;8:919-931.

6. Calhoun A. Adjusting estradiol concentrations reduces headache frequency and severity in female migraineurs (Abst). Cephalalgia 2001;21:448-449.

7. Calhoun A. Women's issues in headache, in Loder E (ed): Headache. American College of Physicians/American Society of Internal Medicine. In press.

RELATED ARTICLE: Key Points

* Menstrual-associated migraine (MAM) is common and is related to the premenstrual decline in estradiol concentration.

* No currently available combined oral contraceptive will benefit MAM if a woman has migraine with her natural cycle (as the lowest-dose pills produce a 20 [micro]g ethinyl estradiol-equivalent decline).

* Reducing the premenstrual decline in estradiol to a 10 [micro]g ethinyl estradiol-equivalent prevents MAM.

Anne H. Calhoun, MD

From the Department of Neurology, University Headache Clinic, University of North Carolina at Chapel Hill, Chapel Hill, NC.

Reprint requests to Anne H. Calhoun, MD, Clinical Associate Professor, Neurology, University of North Carolina at Chapel Hill, 3114 Bioinformatics Bldg., Campus Box 7025, Chapel Hill, NC 27599-7025.