Docstoc

Menopause and Migraine

Document Sample
Menopause and Migraine Powered By Docstoc
					                                                         MM




Menopause and Migraine
        Lynne O. Geweke, MD                                                       our therapeutic options can encom-
                                                                                  pass usual migraine treatments.
        Migraine is an extremely common disorder in primary care
                                                                                  Migraine, Gender and Age
        clinics, and a major source of medical morbidity in the                   The prevalence of migraine in
        productive years of adulthood. Headaches are carefully                    adulthood by age and gender is
        defined in the International Headache Society’s (IHS)                     shown in the Figure. The preva-
                                                                                  lence of migraine in childhood is
        International Classification of Headache Disorders (ICHD-II,              roughly equal for boys and girls, but
        2004).1 Migraine is a particular type of headache, as defined             right around the time of menarche
                                                                                  the prevalence in females increases
        by established criteria of the IHS. The criteria for “migraine
                                                                                  rapidly and never again drops to par
        without aura,” the most common headache for which patients                with the male curve. The highest
        seek medical care, are shown in the Table.                                prevalence in both males and fe-
                                                                                  males occurs during the thirties and
          Although these criteria were developed initially for research           forties, and the female-to-male gen-
        purposes, they can be used clinically. Application of these               der prevalence ratio is roughly 3:1.4
        criteria has undoubtedly helped increase the recognition of                  It is tempting to ascribe the in-
                                                                                  creased prevalence of migraine in
        migraine over the last two decades. Clearly, the majority of              women to hormonal influences, with
        headaches that are sufficiently troublesome to come to                    the drop in migraines after age 40
                                                                                  due to a decrease in hormonal trig-
        clinical attention are migraine.
                                                                                  gers. However, if female hormones
                                                                                  were solely responsible for the gender
   More detailed diagnostic catego-      the significant majority of self-        differential, one would expect a more
rization may not make a great deal       diagnosed “sinus headaches” are ac-      precipitous drop during the post-
of practical difference in the typical   tually migraines.3 These diagnostic      menopausal age range, and a decrease
clinic setting. Many headache spe-       points have significant implications,    in the gender ratio. Also, although
cialists subscribe to the “spectrum      meaning that we may be somewhat          male prevalence is lower throughout
hypothesis” of migraine,2 believing      inclusive of other headache disorders    adulthood, the pattern varies with age
that migraine and tension-type           when we talk about hormonal influ-       in much the same way as female
headaches merely represent the two       ences. In the clinic, the spectrum hy-   prevalence. Clearly, there is also an
ends of a continuous spectrum of         pothesis means that even when a          age effect on migraine in both sexes,
headache disorders. Furthermore,         headache is not a “perfect” migraine,    independent of hormones.


                                                                                           SEPTEMBER /OCTOBER 2007    13
                                                                                    MM




      Table. International Classification of Headache Disorders (ICHD)
      Criteria for the Diagnosis of Migraine Without Aura1
                                                                                                                 which implicated a drop in estro-
      Diagnostic criteria                                                                                        gen levels as the major trigger for
                                                                                                                 menstrual migraine. More recent
      A. At least 5 attacks fulfilling criteria B-D
                                                                                                                 work and clinical observations have
      B. Headache attacks lasting 4-72 hours (untreated or unsuccessfully treated)                               verified this.7-10 Clinical experience
      C. Headache has at least two of the following characteristics:                                             also suggests that almost any hor-
                                                                                                                 monal fluctuation can be linked to
          a.   Unilateral location
                                                                                                                 migraine in certain individuals, al-
          b. Pulsating quality                                                                                   though statistical support is less
          c. Moderate or severe pain intensity                                                                   compelling.
                                                                                                                    This is most understandable if we
          d. Aggravation by or causing avoidance of routine physical activity
                                                                                                                 consider that the primary trigger for
             (e.g., walking or climbing stairs)
                                                                                                                 migraine is probably change itself.
      D. During headache at least one of the following:                                                          This is not limited to hormonal
          a.   Nausea and/or vomiting                                                                            changes. Change in weather, stress,
                                                                                                                 sleep, eating patterns and much
          b. Photophobia and phonophobia
                                                                                                                 more can trigger migraine. However,
      E. Not attributed to another disorder                                                                      the unpredictable hormonal changes
      Source: International Headache Society
                                                                                                                 of perimenopause and menopause
                                                                                                                 can be related to noticeable difficul-
                                                                                                                 ties with regard to headache fre-
                                                                                                                 quency, severity and predictability.
                                                                                                                    Studies that have looked at the
                                                                                                                 patterns of migraine during peri-
                                                                                                                 menopause have suggested that
                                                                                                                 some women are indeed more
                                                                                                                 prone to migraine at this time.11
                                                                                                                 There is considerable individual
                                                                                                                 variation, and there is some sugges-
                                                                                                                 tion that “hormonally sensitive”
                                                                                                                 women are more likely to have dif-
                                                                                                                 ficulty with migraine at this stage.
                                                                                                                 No formal definition of “hormonally
                                                                                                                 sensitive” has been put forth, but the
                                                                                                                 term generally refers to women who
                                                                                                                 have experienced migraine changes
                                                                                                                 in conjunction with previous hor-
     Figure. Adjusted age-specific prevalence of migraine by sex, 1999.4
                                                                                                                 monal events such as menstruation,
     Adapted from Lipton RB, Diamond S, Reed M, et al. Prevalence and burden of migraine in the United States:
     data from the American Migraine Study II. Headache 2001;41:646-57.
                                                                                                                 pregnancy and use of oral contra-
                                                                                                                 ceptives (OCs). On the other hand,
                                                                                                                 the link between menopausal com-
Hormones and Migraine                                        particularly troublesome. The tra-                  plaints and migraine is weak at best,12
at Menopause                                                 ditional theory of menstrual mi-                    and this is consistent with a similarly
For the woman with migraine, the                             graine is an estrogen-withdrawal                    poor correlation between migraine
hormonal irregularities of peri-                             theory, based on the work of                        and premenstrual symptoms in
menopause and menopause are                                  Somerville in the early 1970s,5,6                   younger women. A few cautions are

14         MENOPAUSE MANAGEMENT
                                                       MM




also in order. Women may over- or
underestimate the hormonal con-
nections to migraine, and headache
diary data can be imperfect.
                                           T
                                           T         here is also
                                          little hard evidence
                                                                               (under the age of 45). On the other
                                                                               hand, migraine without aura has not
                                                                               clearly been associated with in-
                                                                               creased stroke risk. There is also lit-
   As for menopause itself, it is im-                                          tle hard evidence for a correlation
portant to distinguish between natu-        for a correlation                  between stroke and migraine over
ral and surgical menopause. In                                                 the age of 50, except perhaps for a
natural menopause, the likelihood
                                             between stroke                    small increase among women with
that the pattern of preexisting mi-                                            migraine with aura. Smoking ap-
                                          and migraine over
graine will improve has been re-                                               pears to compound the stroke risk in
ported as high as 50%-60%,13,14         the age of 50, except                  migraine with aura, as does the use
although there is much variability                                             of estrogen-containing OCs. Cur-
between studies. Some women do           perhaps for a small                   rently, the World Health Organi-
worsen, many experience no change                                              zation (WHO), the American
in their migraines, and some even           increase among                     College of Obstetricians and Gyne-
develop new-onset migraine around                                              cologists (ACOG) and the IHS all
the time of menopause. There is a              women with                      recommend that women who have
logical suggestion that women with                                             migraine with aura not use combi-
menstrual migraine would be most
                                         migraine with aura.                   nation OCs. Concurrent smoking
likely to improve after menopause,                                             or history of thrombosis, such as
but supporting data for this notion     tempted in the prevention of re-       deep vein thrombosis, are also
are limited.                            fractory migraine,16 and this has      WHO and ACOG contraindications
   Surgically induced menopause         been suggested as a way of selecting   to OC use. Since OCs are often used
(that is, oophorectomy with or          women for surgical menopause.          to control menstrual problems in the
without hysterectomy) appears to        However, this remains very much        perimenopausal years, women who
be associated with worsening of mi-     unproven and unreliable. Migraine      have migraine with aura should
graine in a substantial percentage of   is not an indication for total hys-    be warned against using estrogen-
women;13 again, however, the num-       terectomy. It is not a contraindi-     containing OCs for any indication.
bers vary widely in studies. This       cation if oophorectomy with or            Imaging studies have suggested
worsening occurs even with reten-       without hysterectomy needs to be       an increased number of clinically
tion of the ovaries or postsurgical     done for another compelling medical    silent white matter lesions, some-
estrogen therapy.15 The fraction of     reason, but it is important to make    times interpreted as ischemic events,
women who worsen is definitely          sure the patient knows that the sur-   specifically localized to the posterior
sufficiently high (often cited infor-   gery is unlikely to benefit migraine   circulation in migraineurs.20,21 The
mally as two-thirds) for surgical       and may, in fact, make it worse.       relationship of age and hormonal
menopause not to be considered a                                               status to these lesions is unclear, al-
treatment for migraine at this time,    Migraine and the Risk of Stroke        though in general, small white mat-
no matter how hormonally sensi-         The risk of stroke is of increasing    ter hyperintensities visible on
tive the woman appears to be. The       concern as women age through           magnetic resonance imaging in-
reason for this worsening is unclear,   menopause. It is an area that has      crease with age. At this point, there
but it may be related to the abrupt,    been controversial for years, but      is no indication for systemic an-
large change in hormones and/           some patterns and recommendations      tithrombotic treatment in mi-
or the use of supplementary             have emerged from studies.17-19        graineurs, even in patients with such
estrogen following surgery. Chem-       Migraine with aura does seem to        cerebellar lesions.
ical oophorectomy with estrogen         carry a roughly 3-fold increase in        It does appear from the Women’s
add-back therapy has been at-           the risk of stroke in young women                         (continued on page 19)


                                                                                        SEPTEMBER /OCTOBER 2007       15
                                                           MM




Menopause and Migraine
(continued from page 15)

Health Initiative (WHI) reports that
                                           T
                                           T         riptans are the
                                                most effective
                                                                                     volving multiple brainstem signs and
                                                                                     symptoms, or motor weakness).
                                                                                        Nonsteroidal anti-inflammatory
                                                                                     drugs in prescription doses have
hormone therapy (HT) increases the                                                   been successful in some women for
risk of stroke.22,23 However, there are           abortives for                      hormonally related migraines. Other
no specific data suggesting that mi-                                                 abortives are also acceptable, or at
graine and HT interact; thus, mi-
                                           migraine, and there                       least as acceptable at menopause as
graine without aura in itself is not                                                 at other ages and life stages. Overuse
                                          is no reason to think
currently a contraindication to the                                                  and rebound headaches can occur
use of HT in menopausal women.              that they should be                      at any age, and potentially habit-
                                                                                     forming medications—particularly
Treatment and Prevention of                 less effective at or                     barbiturates and narcotics—should
Migraine in the Menopausal Years                                                     be used extremely sparingly, if at all.
First-line treatment of hormonally            after menopause                           Preventive therapy. The usual
triggered migraine does not involve                                                  rules for preventive therapy hold
hormonal manipulation. Rather, it         than in earlier years.                     during the perimenopausal and
involves the same abortive and pre-                                                  menopausal years as well. Indica-
ventive agents used to treat any          menopause than in earlier years.           tions for preventive treatment in-
other migraine. For some time now,        Safety is not a large concern, al-         clude (but are not limited to) overly
specialists have debated whether          though menopause can serve as a            frequent migraine, incompletely ef-
menstrual migraines (the prototype        signal to review cardiac and cere-         fective abortive therapy and cardio-
for hormonally triggered migraines)       brovascular risk factors. A 10-year        vascular or other contraindications
are truly different, more intractable     retrospective assessment of suma-          to effective abortive therapy. The
and harder to treat than other mi-        triptan (Imitrex) in clinical use failed   term “overly frequent migraine” has
graines. Still, several standard med-     to show an increase in myocardial          a range of possible interpretations,
ications, such as triptans, have been     infarction or stroke, although it was      but any time the migraine frequency
shown to be quite successful in           acknowledged that sumatriptan has          averages twice per week or more,
menstrual migraine, whereas hor-          generally not been given to patients       the potential for rebound becomes
mone manipulation is less pre-            with significant risk factors.24 A dif-    significant and prevention should
dictable and may actually worsen          ferent statistical examination of the      be seriously considered. Standard
migraine in a significant percentage      question of triptan safety asked           preventives, including beta block-
of patients. Also, hormone changes        whether a cardiac workup should            ers, tricyclic antidepressants and
are likely best viewed as potent mi-      be considered before prescribing           some of the antiepileptic drugs, are
graine triggers—akin to red wine          triptans; the conclusion was that po-      all useful in this age range, and
and monosodium glutamate—                 tential triptan use did not add to the     should be prescribed based on the
rather than as fundamental causes of      need for a cardiac workup.25 In other      patient’s individual profile.
migraine (integral parts of a univer-     words, a cardiac workup should be
sal pathophysiology).                     undertaken if there are cardiac con-       Menopause and Medication Overuse
   Abortive therapy. For significant      cerns, but should not be done solely       Rebound (medication-overuse) head-
(moderate to severe) migraines, the       because a triptan is being pre-            ache is an important clinical prob-
current first-line abortive agents are    scribed. Current contraindications         lem in this age group. It is critical to
the triptans. Triptans are the most       for triptan use remain known car-          remember that medication overuse
effective abortives for migraine, and     diovascular disease, uncontrolled hy-      actively prolongs chronic headaches,
there is no reason to think that they     pertension and basilar or hemiplegic       sometimes for years. There is, as
should be less effective at or after      migraine (migraine with auras in-          previously noted, a tendency for


                                                                                              SEPTEMBER /OCTOBER 2007      19
                                                           MM




migraine to improve with age, and
some women do improve with
menopause. If a patient is in rebound,
however, any natural tendency to im-
                                            H
                                            H                igh doses
                                              of HT, even given
                                                                                     tion of migraine with aura and stroke,
                                                                                     new-onset aura may be considered a
                                                                                     contraindication to the continued use
                                                                                     of HT. HT that is given cyclically
prove with age or hormonal status                                                    may trigger estrogen-withdrawal mi-
may well be masked. In other words,         transdermally, have                      graines, much like OCs.
menopause will not “cure” rebound                                                        Oral HT formulations are asso-
headaches, and medication overuse
                                              been associated                        ciated wtih wide variations in blood
must be treated aggressively at any                                                  levels of estrogen, and studies sug-
                                             with development
age. This author has treated many                                                    gest that non-oral estrogen delivery
women in their 60s, 70s and even               of auras as well                      is associated with better outcomes in
80s who have medication-overuse                                                      migraine sufferers.27 Theoretically,
headaches and who have been tak-                  as increased                       transdermal patches probably pro-
ing abortive medications (often bu-                                                  vide the smoothest release form of
talbital or codeine) daily for 40 years      migraine severity.                      HT. If HT is being used solely for
or more. Some of these women,                                                        relief of hot flashes and other vaso-
if successfully weaned from their                                                    motor symptoms, dosing should be
abortives, become virtually migraine-     week of the pills, when estrogen           started as low as possible and in-
free, making one wonder how many          levels fall. Continuous administra-        creased only as necessary in order to
years ago they would have naturally       tion regimens of OCs are available,        prevent aggravating migraine. High
improved had they not continued to        and may theoretically decrease the         doses of HT, even given transder-
overuse medications. Hope for a           number of estrogen-withdrawal              mally, have been associated with
“miraculous menopause” should not         headaches. In practice, however, mi-       development of auras as well as in-
delay appropriate treatment. Gener-       graine-prone women often report            creased migraine severity. If this
ally, overuse is defined as regular use   irregular headache occurrence re-          happens, reduction or withdrawal
of abortive therapy on more than          gardless. The low estrogen content         of estrogens is recommended.
about 2 days per week on average.         of these pills may not fully override          Many women now are using
                                          natural hormone fluctuations in            over-the-counter preparations to
Hormonal Manipulation and                 some perimenopausal women.26 If            treat menopausal symptoms such as
Migraine at Menopause                     a woman who has migraine with              hot flashes. Most typically, these
Just a few years ago, a large per-        aura asks about use of low-dose OCs        preparations contain soy and/or
centage of menopausal women               at menopause, it would be wise to          black cohosh, sometimes in fairly
were using HT at menopause.               advise against the OCs because of          complex herbal combinations. They
Since the release of the WHI re-          the stroke risk. If she does not have      may have some estrogenic activity,
ports, however, many more women           auras, another consideration is            although they probably offer little
are going through menopause               whether she has ever had trouble           symptom relief.31 Since these prepa-
without HT, although some have            tolerating OCs in the past. If so, it is   rations are not regulated as drugs,
returned to HT because of unac-           probably best to avoid them now.           the amount of potentially active
ceptable menopausal symptoms.                The effect of HT on migraine is         components may vary significantly.
   Low-dose OCs, most typically           likely highly individualized, but it       This means that hormonally sensi-
containing 20-30 mcg of ethinyl           appears that current users are more        tive women may find that these treat-
estradiol, are frequently given to        likely to report migraine than are         ments actually trigger migraines.
women in perimenopause. Women             nonusers.27-29 Case reports also in-       Migraine patients should be encour-
who are prone to migraine may ex-         dicate that HT can trigger migraine        aged to always report all medications
perience what is essentially a men-       aura, particularly at higher doses.30      they are taking, including any herbal
strual migraine during the placebo        Given the concern about the correla-       and vitamin supplements, and they

20    MENOPAUSE MANAGEMENT
                                                                  MM




should be aware that “natural” is not    tinue to be useful in healthy women                          based study of women aged 40 to 74 years. Headache
                                                                                                      2003;43:27-35.
a synonym for “safe.”                    in this age group. HT may cause
                                                                                                      13. Neri I, Granella F, Nappi R, et al. Characteristics of
   For the woman with migraine,          increased difficulty with migraines                          headache at menopause: a clinico-eipidemiologic study.
caution and individualization of         in the hormonally sensitive woman,                           Maturitas 1993;17:31-7.
                                                                                                      14. Bousser MG. Estrogens, migraine, and stroke. Stroke
treatment is the bottom line when it     and should be used cautiously. Con-                          2004;35(11 Suppl 1):2652-56.
comes to use of HT in any form.          trol of migraine via hormonal ma-                            15. MacGregor EA. Migraine and the menopause. J Br
For most female migraineurs, going       nipulation is difficult at best, and                         Menopause Soc 2006;12:104-08.
                                                                                                      16. Martin V, Wernke S, Mandell K, et al. Medical
through menopause without addi-          there is no role for oophorectomy                            oophorectomy with and without estrogen add-back
tional hormones is likely to be the      with or without hysterectomy for                             therapy in the prevention of migraine. Headache 2003;
                                                                                                      43:309-21.
most beneficial course for their         migraine treatment. I
                                                                                                      17. Tzourio C, Iglesias S, Hubert JB, et al. Migraine and
headaches. If they need HT for                                                                        risk of ischaemic stroke: a case-control study. Br Med J
other reasons, the above recom-                                                                       1993;308:289-92.
                                         Lynne O. Geweke, MD, is Clinical Assistant                   18. Tzourio C, Tehindrazanarivelo A, Iglesias S, et al.
mendations may be considered. It’s                                                                    Case-control study of migraine and risk of ischemic stroke
                                         Professor and Director of the Headache
worth keeping in mind, though, that      Clinic, Department of Neurology, Univer-
                                                                                                      in young women. Br Med J 1995;310:830-33.

standard nonhormonal preventives         sity of Iowa, Iowa City, IA.
                                                                                                      19. Kurth T, Slomke MA, Kase CS, et al. Migraine,
                                                                                                      headache, and the risk of stroke in women: a prospective
for migraine help a significant per-                                                                  study. Neurology 2005;64:1020-26.
                                             Dr. Geweke reports no potential conflicts
centage of patients, and only rarely                                                                  20. Kruit MC, van Buchem MA, Hofman PA, et al. Migraine
                                         related to the content of this article.                      as a risk factor for subclinical brain lesions. JAMA 2004;
worsen headaches. Hormones, on                                                                        291:427-34.
the other hand, may make some               This article includes discussion of off-
                                                                                                      21. Kruit MC, Launer LJ, Ferrari MD, et al. Infarcts in the
                                         label use of medication.                                     posterior circulation territory in migraine. The population-
women better, but worsen signifi-                                                                     based MRI CAMERA study. Brain 2005;128:2068-77.
cant numbers as well. The odds fa-          Submitted: August 1, 2006. Accepted:
                                                                                                      22. Writing Group for the Women’s Health Initiative Inves-
                                         April 15, 2007.
vor the standard preventives, and                                                                     tigators. Risks and benefits of estrogen plus progestin in
                                                                                                      healthy postmenopausal women: principal results from the
when HT worsens migraines but is         References                                                   Women’s Health Initiative randomized controlled trial.
otherwise deemed necessary, adding                                                                    JAMA 2002;288:321-33.
                                         1. Headache Classification Subcommittee of the Inter-
                                         national Headache Society. The International Classifica-     23. Writing Group for the Women’s Health Initiative
a preventive may be the most ra-                                                                      Investigators. Effects of conjugated equine estrogen in
                                         tion of Headache Disorders (2nd ed). Cephalalgia
tional route to take.                    2004;24(Suppl 1):1-160.                                      postmenopausal women with hysterectomy. JAMA 2004;
                                                                                                      291:1701-12.
                                         2. Lipton RB, Cady RK, Stewart WF, et al. Diagnostic
                                         lessons from the Spectrum Study. Neurology 2002;58           24. Hall GC, Brown MM, Mo J, et al. Triptans in migraine.
Summary and Conclusions                  (9 Suppl 6):S27-S31.                                         The risks of stroke, cardiovascular disease, and death in
                                                                                                      practice. Neurology 2004;62:563-68.
Hormonal changes in the peri-            3. Eross E, Dodick D, Eross M. The sinus, allergy, and
                                         migraine study (SAMS). Headache 2007:47:213-24.              25. Orlando LA, Matchar DB. When to stress over trip-
menopausal and menopausal years                                                                       tans: a Markov analysis of cardiovascular risk in migraine
                                         4. Lipton RB, Diamond S, Reed M, et al. Prevalence and
can be potent triggers for migraine.     burden of migraine in the United States: data from the
                                                                                                      treatment. Headache 2004;44:652-60.

There are wide individual differ-        American Migraine Study II. Headache 2001;41:646-57.         26. SulakPJ, Kuehl TJ, Ortiz M, et al. Acceptance of alter-
                                                                                                      ing the standard 21-day/7-day oral contraceptive regimen
ences, but migraine tends to improve     5. Somerville BW. The role of progesterone in menstrual      to delay menses and reduce hormone withdrawal symp-
                                         migraine. Neurology 1971;21:853-59.                          toms. Am J Obstet Gynecol 2002;186:1142-49.
into old age. How much of this is re-    6. Somerville BW. The role of estradiol withdrawal in the    27. Nappi RE, Cagnacci A, Granella F, et al. Course of
lated to hormone change is debat-        etiology of menstrual migraine. Neurology 1972;22:355-65.    primary headaches during hormone replacement therapy.
able. With increasing age comes          7. MacGregor EA, Frith A, Ellis J, et al. Incidence of       Maturitas 2001;38:157-63.
                                         migraine relative to menstrual cycle phases of rising and    28. Misakian AL, Langer RD, Bensenor IM, et al. Post-
increasing concern about stroke risk     falling estrogen. Neurology 2006;67:2154-58.                 menopausal hormone therapy and migraine headache.
and the safety of hormone use, but       8. Loder E, Rizzoli P, Golub J. Hormonal management of       J Women’s Health 2003;12:1027-36.
                                         migraine associated with menses and the menopause: a
little is known about how migraine       clinical review. Headache 2007;47:329-40.
                                                                                                      29. Kelman L. Women’s issues of migraine in tertiary
                                                                                                      care. Headache 2004;44:2-7.
factors into this equation.              9. Martin VT, Behbehani M. Ovarian hormones and              30. MacGregor A. Estrogen replacement and migraine
    In the menopausal woman, treat-      migraine headache: understanding mechanisms and              aura. Headache 1999;39:674-78.
                                         pathogenesis—part 1. Headache 2006;46:3-23.
ment of migraine should initially        10. Martin VT, Behbehani M. Ovarian hormones and
                                                                                                      31. Newton KM, Reed SD, LaCroix AZ, et al. Treatment
                                                                                                      of vasomotor symptoms of menopause with black cohosh,
proceed along standard migraine          migraine headache: understanding mechanisms and              multibotanicals, soy, hormone, therapy, or placebo: a ran-
                                         pathogenesis—part 2. Headache 2006;46:365-86.                domized trial. Ann Intern Med 2006;145:869-79.
abortive and preventive protocols.
                                         11. Wang SJ, Fuh JL, Lu SR, et al. Migraine prevalence
Most medications used in younger         during menopausal transition. Headache 2003;43:470-78.
patients, including the triptans, con-   12. Mattson P. Hormonal factors in migraine: a population-



                                                                                                                    SEPTEMBER /OCTOBER 2007                      21