|Home | About | Journals | Submit | Contact Us | Français|
The development of diagnostic criteria has enabled greater recognition of menstrual migraine as a highly prevalent and disabling condition meriting specific treatment. Although few therapeutic trials have yet been undertaken in accordance with the criteria, the results of those published to date confirm the efficacy of acute migraine drugs for symptomatic treatment. If this approach is insufficient, the predictability of attacks provides the opportunity for perimenstrual prophylaxis. Continuous contraceptive strategies provide an additional option for management, although clinical trial data are limited. Future approaches to treatment could explore the genomic and nongenomic actions of sex steroids.
Four of every ten women and two of every ten men will contract migraine in their lifetime, most before age 35 years [Stewart et al. 2008]. More than 50% of women with migraine, both in the general population and presenting to specialist clinics, report an association between migraine and menstruation [MacGregor et al. 2004, 1997, 1990; Couturier et al. 2003; Dzoljic et al. 2002; Granella et al. 1993].
The peak incidence of migraine during the menstrual cycle occurs on the days directly before and after the first day of menstruation [MacGregor and Hackshaw, 2004; Dzoljic et al. 2002; Stewart et al. 2000; Johannes et al. 1995]. In a population-based study, Stewart et al.  noted a significantly elevated risk of migraine without aura on the first two days of menstruation [odds ratio (OR) 2.04; 95% confidence interval (CI) 1.49–2.81]. The lowest risk for headache was around the expected time of ovulation [OR 0.44; 95% CI 0.27–0.72]. Headache duration appeared to be significantly longer for migraine headaches in the 3 to 7 day period before onset of menses [Stewart et al. 2000]. In a clinic-based study, MacGregor and Hackshaw  noted that women were 25% (RR 1.25) more likely to have migraine in the five days leading up to menstruation increasing to 71% (RR 1.71; 95% CI 1.45–2.01 p<0.0001) in the two days before menstruation. The risk of migraine was highest on the first day of menstruation and the following two days (RR 2.50; 95% CI 2.24–2.77 p<0.0001). Similarly, in a population-based study, Wciber et al.  found the highest risk of migraine on the first three days of menses (HR 1.96; p<0.00001). Menstrual migraine is also associated with increased menstrual distress and disability [Dowson et al. 2005; Kibler et al. 2005; Granella et al. 2004; MacGregor et al. 2004; Couturier et al. 2003; Beckham et al. 1992]. As a consequence of these findings, the International Headache Society developed diagnostic criteria for menstrual migraine (Box 1).
For most women with menstrual attacks, migraine also occurs at other times of the month ('menstrually-related’ migraine) [Headache Classification Subcommittee of the International Headache Society (IHS), 2004; MacGregor et al. 1990]. Fewer than 10% of women report migraine exclusively with menstruation and at no other time of the month ('pure’ menstrual migraine) [Headache Classification Subcommittee of the IHS, 2004; MacGregor et al. 2004; Dzoljic et al. 2002; Granella et al. 1993; MacGregor et al. 1990]. To confirm a diagnosis contemporaneous diary cards covering a minimum of three menstrual cycles should be reviewed.
Acute treatment of menstrual migraine is the same as for nonmenstrual attacks and includes a combination of analgesics with or without prokinetic antiemetics, nonsteroidal anti-inflammatory drugs, ergot derivatives and triptans [Steiner et al. 2007]. The nonprescription combination of acetaminophen, aspirin, and caffeine (AAC; Excedrin Migraine, Bristol-Myers Squibb Company, New York) was assessed for the treatment of menstruation-associated migraine compared with migraine not associated with menses using data from three double-blind, randomized, placebo-controlled, single-dose trials [Silberstein et al. 1999]. Subjects with severe vomiting or disability were excluded. Menstruation-associated migraine was treated by 185 women and 781 women treated migraine not associated with menses. There was no statistically significant difference in pain response between menstruation-associated migraine and migraine not associated with menses.
Sumatriptan for the acute treatment of migraine occurring between day —2 and day +4 of the cycle was evaluated in a randomized, double-blind, placebo-controlled study [Nett et al. 2003]. A single headache was treated with oral sumatriptan 50 mg, sumatriptan 100 mg, or placebo taken within 1 hour of onset of a mild headache. At 2 hours, 51% and 61% of the sumatriptan 50 mg and 100 mg groups were pain-free compared with 29% of the placebo group (p<0.001). Sustained pain-free response from 2 to 24 hours was reported by 30% of the sumatriptan 50 mg group (p = 0.007), 31% of the sumatriptan 100 mg group (p = 0.004), and 14% of the placebo group.
A prospective, multicentre, randomized, doubleblind, placebo-controlled, two-group crossover study was carried out on patients who self-reported migraine during an 8 day window starting 3 days before the onset of menstruation in two of their last three menstrual cycles with >80% of their attacks falling within the window in the previous 6 months [Dowson et al. 2005]. Women treated all migraine attacks for 2 months with sumatriptan 100 mg and for 2 months with placebo. The primary endpoint was the proportion of patients reporting headache relief at 4 hours for the first treated attack. Significantly more women receiving sumatriptan than placebo reported headache relief for attacks occurring inside (67% versus 33%, p = 0.007) and outside (79% versus 31%, p< 0.001) the menstrual period.
In a randomized, placebo-controlled trial of acute treatment of migraine occurring between day —3 and day +5 of the menstrual cycle zolmitriptan 1.25 mg was used for mild pain, 2.5 mg for moderate pain, and 5 mg for severe headache pain [Loder et al. 2004]. Zolmitriptan (all doses) significantly increased 2-hour headache response compared with placebo (48% versus 27%, respectively; p<0.0001). Pain relief was statistically superior (p = 0.03) with zolmitriptan treatment, as early as 30 minutes after dosing.
Naratriptan for the acute treatment of migraine occurring on day —2 to day +4 of the menstrual cycle was assessed in a randomized, doubleblind, placebo-controlled trial [Massiou et al. 2005]. A significantly greater percentage of the naratriptan group were pain free at 4 hours compared with placebo (58% versus 30% with placebo; p<0.001).
Almotriptan and zolmitriptan for acute treatment of menstrual migraine were evaluated in a double-blind, randomized trial of 255 women [Allais et al. 2006]. Pain-free response at 2 hours was achieved in 44.9% of patients receiving almotriptan and 41.2% of those receiving zolmitriptan. The 2-hour pain-free status was sustained for at least 24 hours in 29.3% and 27.1% of patients treated with almotriptan and zolmitriptan, respectively. In a post hoc analysis of the AXERT Early miGraine Intervention Study (AEGIS), 275 women treated 506 migraine attacks. Almotriptan treatment efficacy outcomes were not significantly different for menstrual and nonmenstrual attacks: 2-hour pain relief, 77.4% versus 68.3%; 2-hour pain free, 35.4% versus 35.9%; and sustained pain free, 22.9% versus 23.8% [Diamond et al. 2008].
Rizatriptan 10 mg was effective for the treatment of ICHD-II menstrual migraine in two prospective, randomised, double-blind, placebo-controlled trials, as measured by 2 hour pain relief and 24 hour sustained pain relief in 707 women [Nett et al. 2008]. In an early intervention model using rizatriptan 10 mg, 2 hour pain-free rates were comparable for 94 women treating menstrual and nonmenstrual migraine attacks [Martin et al. 2008].
When acute therapy is insufficient to reduce disability from menstrual migraine, there is the option for preventing attacks using perimenstrual or continuous prophylaxis. The choice depends on the individual woman's type of migraine, regularity of menstruation, other menstrual problems and need for contraception (Figure 1).
Short-term prophylactic strategies have the advantage that treatment is only used at the time of need, potentially reducing the risk of adverse events compared with continuous prophylaxis. However, results from RCTs are limited (Table 1). None of the drugs and hormones recommended for perimenstrual prophylaxis are licensed for management of menstrual migraine.
Studies using 550 mg naproxen once or twice daily perimenstrually have shown limited efficacy [Nattero et al. 1991; Sances et al. 1990; Szekely et al. 1989; Sargent et al. 1985]. A recent open-label study of perimenstrual naproxen 550 mg daily has shown efficacy for prevention of menstrual migraine [Allais et al. 2007]. NSAIDs are useful as first-line agents for migraine associated with dysmenorrhoea and/or menorrhagia.
Maintaining luteal phase oestrogen levels can prevent menstrual attacks [MacGregor et al. 2006; Somerville, 1975a, 1975b, 1972]. Doses equivalent to 1.5 mg estradiol gel allow a mean estradiol plasma level of 80pg/ml to be reached. Lower doses of oestrogen are not effective [Smits et al. 1994; Pradalier et al. 1994; Pfaffenrath, 1993].
There is evidence that some women responding to oestrogen supplements experience delayed attacks when the supplements are discontinued [MacGregor et al. 2006] Oestrogen ‘withdrawal’ migraine may occur if treatment is not continued until the rise in endogenous oestrogen. Although there are no trial data, clinical practice suggests that for these women the duration of supplement use can be extended until day 7 of the cycle, tapering the dose over the last 2 days.
Trials using frovatriptan, naratriptan, sumatriptan and zolmitriptan for perimenstrual prophylaxis have suggested efficacy [Tuchman et al. 2008; Moschiano et al. 2005; Silberstein et al. 2004; Newman et al. 2001, 1998].
Perimenstrual triptan prophylaxis is well tolerated and the high completion rates in the clinical trials are notable. Post-treatment migraine has been reported following naratriptan but not frovatriptan [Mannix et al. 2007; Brandes et al. (in press)]
A small pilot, open-label, nonrandomized, parallel group study assessed the efficacy of 2.5 mg frovatriptan against 25 mg transdermal oestrogen or 50 mg naproxen sodium each taken once daily for 6 days, beginning 2 days before the expected onset of menstrual headache [Guidotti et al. 2007]. The baseline median headache severity score severity was 4.6, 4.2 and 4.3 in the group subsequently treated with frovatriptan, transdermal oestrogen and naproxen sodium, respectively (p = 0.819) compared with scores of 2.5, 3.0 and 3.0 during treatment (p = 0.049). Although these results suggest that short-term prophylaxis of menstrual migraine with frovatriptan may be more effective than transdermal oestrogen or naproxen sodium, the drug doses used in the study were suboptimal.
Women with irregular periods or who require contraception may benefit from specific strategies that prevent migraine in the hormone-free interval of combined hormonal contraceptives, although to date evidence is currently based more on clinical practice than on robust clinical trial data [Calhoun and Ford, 2008; MacGregor, 2007].
Continuous hormones, in place of the usual regimen of 3 weeks of active followed by 1 week of inactive pills or no therapy, have been recommended based on evidence that oestrogen withdrawal provokes headache in susceptible women. Compared with the usual 21/7-day regimen of combined hormonal contraceptives, a 168-day extended placebo-free regimen led to a decrease in headache severity along with improvement in work productivity and involvement in activities [Sulak et al. 2007]. Similarly, an extended 84-day regimen of a transdermal contraceptive reduced the total incidence of mean headache days compared with a 21/7-day regimen [Laguardia et al. 2005]. No double-blind, placebo-controlled trials, or even open-label trials, of this strategy in menstrual migraine have been performed. However, there is increasing clinical experience of their use in this way [Edelman et al. 2006]. Combined hormonal contraceptives should not be used by women with migraine with aura because of the synergistic increased risk of ischaemic stroke [World Health Organization, 2004; MacGregor and Guillebaud, 1998].
There are no studies assessing anovulatory progestogens such as the intramuscular depot medroxyprogesterone acetate, subdermal etonogestrel and oral desogestrel, which inhibit ovulation. In general, standard contraceptive oral progestogens have little place in the management of menstrual migraine since most do not inhibit ovulation and are associated with a disrupted menstrual cycle [Chumnijaraki et al. 1984]. In contrast, unlicensed higher doses of oral progestogen, sufficient to inhibit ovulation, have shown benefit [Davies et al. 2003].
Although effective, adverse effects of oestrogen deficiency; for example, hot flushes, restrict their use [Holdaway et al. 1991]. The hormones are also associated with a marked reduction in bone density and should not usually be used for longer than 6 months without regular monitoring and bone densitometry. ‘Add-back’ continuous combined oestrogen and progestogen can be given to counter these difficulties [Martin et al. 2003; Murray and Muse, 1997]. Given these limitations, in addition to increased cost, such treatment should be instigated only in specialist departments.
The association between sex steroids and migraine warrants further investigation and has the potential to result in more targeted diagnosis and treatment. Sex steroid activity is not confined to reproductive tissues, having activity in both the peripheral and central nervous systems through genomic and nongenomic effects. Studies investigating the role of the oestrogen receptor 1 (ESR1) gene in migraine show a significant association of the A allele of the G594A SNP with migraine [Colson et al. 2004]. The progesterone receptor (PGR) PROGINS insert has also been implicated [Colson et al. 2005]. Women who carry a copy of both PR and ESR1 risk alleles were 3.2 times more likely to suffer from migraine, an effect that is greater than the independent effects of these genetic variants on disease susceptibility. It is anticipated that this association will be stronger in women with menstrual migraine, who have a strong hormonal trigger for attacks.
If the genes that play a role in this subtype of migraine can be identified, it should be possible to develop objective ways of testing for these genes as a diagnostic tool. Accurate diagnosis of menstrual migraine can aid the selection of currently available treatments. Further, identification of the genes involved could ultimately lead to the development of more effective treatments for this disabling condition, targeted to the specific genes.
Understanding how the genetics translates into a clinical outcome could also provide a more targeted therapeutic approach. Animal models suggest that abnormalities in how oestrogen modulates neuronal function in migraine are due to a mismatch between its gene-regulation and membrane effects [Welch et al. 2006]. The hypothesis is that during phases of high oestrogen levels, increased neuronal excitability is balanced by homeostatic gene regulation in brain cortex, and nociceptive systems. When oestrogen is ‘withdrawn’ around menstruation, mismatch in homeostatic gene regulation by oestrogen unmasks non-nuclear mitogen-activated hyperexcitability of cell membranes, sensitizing neurons to triggers that activate migraine attacks. At the trough of oestrogen levels, the downregulating effect on inflammatory genes is lost and peptide modulated central sensitization is increased as is pain and disability of the migraine attack.
Other potential lines of research could enable a better understanding of the interplay between oestrogen and serotonin. The mechanisms underlying the benefit of perimenstrual triptan prophylaxis are as yet unexplained. Sex steroids modulate neurotransmission in the brain, spinal cord and peripheral nerves, and influence receptor activity of other neurotransmitters, including the serotonergic system. Oestrogen is associated with increased production of serotonin, reduced serotonin reuptake and decreased serotonin degradation.
The role of progesterone also merits more research. Although oestrogen ‘withdrawal’ can trigger migraine in the absence of progesterone, the GABAergic actions of progestrogen are likely to modulate pain and pain perception.
Dr MacGregor has acted as a paid consultant to, and/or her department has received research funding from, Addex, AstraZeneca, BTG, Endo Pharmaceuticals, GlaxoSmithKline, Menarini, Merck and Pozen.