Sage Journals: Discover world-class research

Abstract

This pilot study investigated the effect of menstrual cycle phase (late luteal and mid-follicular) on cerebral perfusion changes during photic stimulation in both controls (n = 5) and true menstrual migraine patients (n = 5). No significant differences in resting baseline perfusion were observed between the two groups during either phase of the menstrual cycle. During the late luteal phase, changes in perfusion within the occipital lobe due to photic stimulation were similar for both groups. However, during the mid-follicular phase, occipital perfusion during visual stimulation decreased for controls but significantly increased for true menstrual migraine patients (P < 0.05). A two way repeated measures ANOVA also demonstrated a significant difference between menstrual migraine patients and controls for photic activation (P < 0.05).

Keywords

True menstrual migraines perfusion imaging oestrogen withdrawal

Introduction

Migraine is a common disorder which predominantly affects women with considerable evidence suggesting a link between oestrogen and progesterone, circulating female sex hormones, and migraine (1–5). Women outnumber men with migraine by at a ratio of at least 2 : 1 with this gender imbalance starting at puberty (6). In women migraines may vary in a predictable pattern in relation to the reproductive cycle, with changes related to ovulation, menses, pregnancy, menopause and exogenous hormone medication (7).

Migraines can occur before or during menstruation (5, 8). Patients may experience headaches, nausea, vomiting, backache, and breast tenderness preceding menstruation and occurring with a premenstrual syndrome (8). In a review of 32 articles using the definition of menstrual migraine MacGregor (9) defined true menstrual migraine as ‘occurring exclusively within 1 or 2 days of the onset of menses’ while menstrual-associated migraines ‘occur at the menses as well as at other times in the menstrual cycle’. Overall, 60% of affected women have menstrual-associated migraines while only 16% have true menstrual migraines (1). The exact mechanism for menstrual migraines remains unknown, but both oestrogen and progesterone have been postulated to be involved as both have potent effects on serotonergic and opioid neurones and can modulate neuronal activity and receptor density (5, 8). Withdrawal of oestrogen, as proposed by the ‘oestrogen withdrawal hypothesis’ has also been suggested to precipitate menstrual migraines (10, 11).

Migraine symptoms are often referable to the visual cortex and in some patients visual stimulation by LED goggles can induce migraines and/or migraine-like symptoms (12, 13). Only a limited number of neuroimaging studies have utilized functional stimulation paradigms to measure noninvasively the effects of menstrual cycle phase on the degree of activation (14, 15). To date, no studies have investigated the degree of activation during the menstrual cycle within a pathophysiological condition such as true menstrual migraine. In this pilot study perfusion measurements were obtained at baseline and with visual stimulation using LED goggles during the late luteal and mid-follicular states of the menstrual cycles of controls (n = 5) and medicated true menstrual migraine patients with (n = 2) and without aura (n = 3). We hypothesized that hormonal variations due to oestrogen withdrawal within the pathophysiological condition of menstrual migraine may affect the degree of functional activation observed during the different phases of the menstrual cycle.

Methods

Subjects

Five right handed female subjects (age 21–35 years) with a normal menstrual cycle were recruited from the University of Pennsylvania community. Five right handed true menstrual migraine patients (age 28–39 years) were recruited from the clinical population served by the Department of Neurology, University of Pennsylvania Medical Center. International Headache Society diagnostic criteria (16–18) were used to diagnose patients with (n = 2) and without aura (n = 3), with all patients allowed to continue to take their medications (Table 1). All patients met criteria of the International Headache Society for menstrual migraine.

Table 1

Demographics of subjects

Subject	Age (years)	Symptoms of migraine	Medications
Patient no. 1	31	Dull and throbbing migraines without auras. Patient usually experiences nausea/vomiting with symptoms with migraines always at menses.	Sumatriptan 200 mg PO qD Verapamil SR 120 mg PO BID
Patient no. 1	31	Patient experiences at least 6 menstrual migraines a year and has a positive family history	Sumatriptan 200 mg PO qD Verapamil SR 120 mg PO BID
Patient no. 2	33	Temporal regions migraines without aura with radiation to neck. Symptoms usually occur 2–3 days prior to menses. Patient experiences at most 5 menstrual migraines a year and has a negative family history	Propranolol LA 100 mg PO qD Desipramine 50 mg PO qD Sumatriptan 100 mg PO qD Acetaminophen 2 Tabs PO q 6
Patient no. 3	28	Migraines with aura and often blurry vision within one visual field (usually left). Patient may sometimes experience nausea/vomiting/photophobia with symptoms usually occurring approximately 1–2 days prior to menses. Patient experiences at least 6 menstrual migraines a year and has a negative family history	Atenolol 50 mg PO qD Sumatriptan 50 mg PO qD
Patient no. 4	31	Throbbing temporal migraine without aura with radiation to neck Patient often experience nausea/vomiting/photophobia with symptoms occurring only at menses. Patient experiences ‘a couple’ menstrual migraines a year and has a postive family history	Sumatriptan 50 mg PO qD
			Docusate 100 mg PO qD
			Nortryptyline 10 mg PO QID
Patient no. 5	39	Migraine with aura and central scotoma/and often unilateral numbness and experiences symptoms only at menses. Patient experiences several menstrual migraines a year and has a + family history	Naproxen 500 mg PO BID
Patient no. 5	39		Sumatriptan 100 mg PO BID
Control no. 1	21	None	None
Control no. 2	31	None	None
Control no. 3	35	None	None
Control no. 4	28	None	None
Control no. 5	24	None	None

All subjects consented to two perfusion magnetic resonance imaging scans with a photic activation paradigm. All subjects had an initial scan at the late luteal phase (approximately one to two days prior to the start of menses), when oestrogen levels are low. For all subjects a subsequent scan was then performed approximately four to five days after completion of the menses (the mid-follicular period), where oestrogen levels are higher. Overall, approximately 8–12 days elapsed between the two scans for each subject. Blood oestrogen levels were not obtained from any of the subjects with all participants questioned if they experienced migraines or migraine-like symptoms during or between the scans. Exclusion criteria for all subjects included menorrhagia, metorrhagia, menometorrhagia, infertility, dyspareunia, pregnancy, history of hypertension, diabetes, smoking, alcohol or drug abuse, anaemia, hyperviscosity states, Raynaud's phenomenon, collagen-vascular disease, epilepsy, stroke, significant head trauma, claustrophobia or the presence of a pacemaker or other metallic implants such as metal fragments or aneurysm clips.

Perfusion imaging procedures

Both resting perfusion measurements and changes in perfusion due to photic stimulation within the occipital cortex were obtained at two phases of the menstrual cycle. Functional magnetic resonance imaging was carried out at 1.5T GE Signa MRI scanner capable of echo-planar imaging, using a standard head RF coil. Routine sagittal and axial T1-weighted images were initially obtained for each subject. Quantitative perfusion imaging was obtained noninvasively using continuous arterial spin labelling (19). A multislice continuous spin-labelling protocol with transit time insensitivity and a postlabelling delay of 1.2 s was used (20, 21). Eight contiguous 8 mm thick slices with an interslice interval of 2 mm were obtained, including slices through the calcarine cortex. The field of view was 24 × 15 cm and the acquisition matrix was 64 × 40. Arterial labelling was performed to the level of the cervical medullary junction. Data acquisition took 420 s for the eight slices with a total of 90 repetitions performed. A 10 minute imaging protocol was used to quantify cerebral perfusion during the off period (rest) and during the on period (photic stimulation using LED goggles flashing at 10 Hz). Each period alternated for 40 s with total imaging time approximately 60 min per session.

Data analysis of perfusion imaging

Perfusion images were calculated from MRI scans acquired with arterial spin tagging as previously described (19). Raw data from the scanner was processed using Interactive Data Language. Automatic phase and distortion corrections for static magnetic field inhomogenieties were applied. Perfusion images were corrected for motion artifact and physiological noise using an algorithm based on principal component analysis. Images were then transformed into standard stereotactic space corresponding to the atlas of Talairach et al. (22), using linear spatial deformations. All perfusion images were inspected for the presence of artifact.

Global perfusion measurements were obtained at baseline conditions during the two phases of the menstrual cycle with a t-test performed both within and between groups to determine if differences were significant. Condition specific effects for photic activation due to LED goggles were assessed by subtracting the off period (rest) from the corresponding on period (photic stimulation). Changes in perfusion were assessed both within a region known to be involved in visual processing (the occipital lobe) as well within a control region not thought to be not involved in visual processing (the superior frontal gyrus). Standard errors were determined for all the above conditions. A two way repeated measures ANOVA was performed with group (normals vs. true menstrual migraine patients) being the between subject variable and cycle (late luteal phase vs. mid-follicular) being the within subject variable.

Results

Behavioural effects of photic stimulation in controls and true menstrual migraine patients

All subjects successfully completed the scans without difficulty and were not lost to follow-up. No subjects experienced migraine or migraine-like symptoms either between the scans or during the scans due to photic stimulation. Subjects did not require medications during either of their scans.

Global baseline perfusion changes in controls and true menstrual migraine patients during the menstrual cycle

Global baseline perfusion changes for both controls and true menstrual migraine patients during the late luteal and mid-follicular phases of the menstrual cycle are shown in Fig. 1. Significant differences in resting baseline perfusion measurements were not observed within groups nor across groups for these two phases of the menstrual cycle.

Figure 1

Global baseline perfusion changes for both controls (n = 5, □) and true menstrual migraine patients (n = 5, ▪) during the late luteal and mid-follicular phases of the menstrual cycle. All error bars indicate standard error.

Perfusion changes within the superior frontal gyrus and the occipital cortex in controls and true menstrual migraine patients during photic stimulation for the two phases of the menstrual cycle

We further investigated the affect of different phases of the menstrual cycle on perfusion changes due to photic stimulation within a control region (superior frontal gyrus) and a region involved in visual processing (occipital lobe) for both groups. Figure 2 shows representative perfusion changes (ml/100 mg/min) for rest, photic stimulation, and photic-rest for a true menstrual migraine patient and a control subject during the mid-follicular phase. An increase in activity can be observed during within the occipital cortex for photic stimulation compared to rest (white arrows). A region of interest analysis for photic stimulation was performed on controls and menstrual migraine patients during the different phases of the menstrual cycle. Figure 3 shows changes in perfusion within the superior frontal gyrus, a region known to not be involved in visual processing, during photic stimulation in controls and true menstrual migraine patients across the menstrual cycle. For both groups no significant changes in perfusion were observed across menstrual cycle. No hemispheric differences were observed within controls or menstrual migraine patients during the different phases of the menstrual cycle (data not shown). There was no correlation between the side of symptoms and changes in perfusion within a particular hemisphere for the menstrual migraine patients (data not shown).

Figure 2

Changes in perfusion (ml/100 g/min) within a representative true menstrual migraine patient and control for rest, photic stimulation, and photic-rest activation paradigm during the mid-follicular phases of the menstrual cycle. An increase in activity can be observed during within the occipital cortex for photic-rest paradigm (white arrows) for the menstrual migraine patient compared to the control.

Figure 3

Changes in perfusion (ml/100 g/min) within the superior frontal gyrus during photic stimulation in controls (□) and true menstrual migraine patients (▪) during the late luteal and mid-follicular phases of the menstrual cycle. All error bars indicate standard error.

Changes in perfusion in the occipital lobe due to photic stimulation in controls and menstrual migraine patients across the menstrual cycle are shown in Fig. 4. During the late luteal phase similar perfusion changes due to photic stimulation were observed for controls compared to true menstrual migraine patients. There was no difference in the percentage change in perfusion for controls due to photic stimulation during the mid-follicular phase compared to the late luteal phase. Within true menstrual migraine patients there existed a trend (P = 0.06) towards an increase in the percentage change in perfusion due to photic stimulation during the mid-follicular phase compared to the late luteal phase. When the percentage changes in perfusion during photic stimulation in menstrual migraine patients was compared to controls a significant increase was observed in menstrual migraine patients (P < 0.05). This increase in perfusion for photic stimulation during the mid-follicular phase compared to the late luteal phase was observed in four of the five menstrual migraine patients (Fig. 5). A two way repeated measures ANOVA demonstrated a main effect for true menstrual migraine patients. A difference greater than chance (P < 0.05) was observed in the mean values among the true menstrual migraine patients compared to normals after allowing for effects of differences in cycle (late luteal phase vs. mid-follicular).

Figure 4

Changes in perfusion (ml/100 g/min) within the occipital lobe during photic stimulation in controls (□) and true menstrual migraine patients (▪) during the late luteal and mid-follicular phases of the menstrual cycle. All error bars indicate standard error. ∗P < 0.5 for controls vs. true menstrual migraine patients during the mid-follicular phase of the menstrual cycle.

Figure 5

Changes in perfusion (ml/100 g/min) within the occipital lobe during photic stimulation in true menstrual migraine patients during the late luteal and mid-follicular phases of the menstrual cycle.

Discussion

This pilot study examined changes in resting perfusion and changes in activation within the occipital cortex due to photic stimulation within true menstrual migraine patients (n = 5) and control subjects (n = 5). Neither group showed cyclical variations in resting perfusion, but the menstrual migraine group showed an overall increase in occipital activation with photic activation. This increase was particularly prominent during the mid-follicular phase of the menstrual cycle for menstrual migraine patients. The concentration of circulating in sex steroids may affect brain and behaviour (23). Changes in cognitive performance during neuropsychological tests correlate with natural fluctuations in endogenous oestrogen levels that occur with the menstrual cycle (24). Rode et al. (25) have demonstrated a modulation of cerebral lateralization during the menstrual cycle for figural comparison tasks with hemispheric asymmetry reduced during the luteal phase but more pronounced during the early follicular phase. Similar cerebral asymmetric changes have been shown for verbal tasks, which generated a more pronounced left hemispheric dominance during the luteal phase, and musical tasks, which showed the greatest asymmetry during the follicular phase (26).

To date only two longitudinal neuroimaging studies using functional magnetic resonance imaging (fMRI) have investigated the effects of menstrual cycle phase on the degree of activation. A recent fMRI study using T2∗ examined brain activation changes for a verbal rhyming decision paradigm during high and low oestrogen and progesterone states of the menstrual cycle. No differences in the degree of activation were observed during the various hormonal states (15). Dietrich et al. (14) used Blood Oxygen Level Dependent (BOLD) fMRI to image cortical activation patterns associated with motor and cognitive tasks for six normal females during various phases of their menstrual cycle. No significant differences in areas activated areas for a finger motor opposition task were observed for different phases of the menstrual cycle. However, larger areas of activation were observed during the luteal phase compared to the follicular phase for both word stem completion and mental rotation tasks. This variation in the size of activation for the two tasks was postulated to be due to differences in the distributive network patterns required for higher cognitive tasks such as word stem completion. The authors further hypothesize that the differences observed between the motor and cognitive tasks may be due to the effects of oestrogen on denditric spine density on the hippocampus (23). The results for the control subjects within this pilot study for a simple photic stimulation task involving the occipital cortex (Fig. 4) are in agreement with the above studies and suggest that phase of the menstrual cycle does not affect perfusion changes observed for simple tasks for normals.

However, variations in hormonal levels during the menstrual cycle may be abnormal in pathophysiological conditions such as menstrual migraine (2, 3, 5, 27). Migraine headaches are in general 2–3 times more common in females than males with frequency increasing after menarche (28, 29). True menstrual migraines, migraines that only occur at or around the time of menses, are a less common form of migraine (30). While the physiological basis of female sex hormone interactions in migraine has not been elucidated, oestrogen has direct vasodilatory effects and increases metabolites, such as nitric oxide, that can lead to vasodilation (31, 32).

Estrogen and progesterone levels decline at the menstrual phase of the cycle. Hormonal levels throughout the cycle have been compared to clinical symptoms in both normals and menstrual migraine patients (1). Prolonged oestrogen exposure followed by withdrawal at menses may reduce necessary factors for vasodilation and could precipitate the pathological events observed in menstrual migraine (11, 33, 34). This ‘oestrogen withdrawal hypothesis’ was further confirmed by Sommerville (10, 35) who demonstrated that a sustained level of oestrogen but not progesterone postponed migraines. Results from this preliminary pilot study are in good agreement with the ‘oestrogen withdrawal hypothesis’ as significantly higher changes in perfusion for photic stimulation were observed in true menstrual migraine patients compared to normals at the mid-follicular phase of the menstrual cycle (Fig. 4). Baseline levels of oestrogen may be higher during the mid-follicular phase for menstrual migraine patients compared to normals with the extent of oestrogen level decline during menses responsible for migraines within these patients (7). Although these results are promising, this pilot study is limited as no oestrogen blood levels were obtained and only a limited number of patients on medication were studied. Future larger studies measuring blood oestrogen levels at a number of periods of the menstrual cycle in true menstrual migraine patients on and off medication should be performed to validate these results.

References

1. Epstein

Hockaday

. Migraine and reproductive hormones throughout the menstrual cycle. Lancet 1975; 1: 543–8.

2. Silberstein

. Hormone-related headache. Med Clin North Am 2001; 85: 1017–35.

3. Silberstein

. Sex hormones and headache. Rev Neurol (Paris) 2000; 156: 4S30–41.

4. Silberstein

Merriam

. Menstrual migraine. Del Med J 1991; 63: 477–84.

5. Silberstein

Merriam

. Physiology of the menstrual cycle. Cephalalgia 2000; 20: 148–54.

6. Waters

O'Connor

. Epidemiology of headache and migraine in women, J Neurol Neurosurg Psychiatry 1971; 34: 148–53.

7. Marcus

. Interrelationships of neurochemicals, estrogen, and recurring headache. Pain 1995; 62: 129–39.

8. Silberstein

. Migraine and women. The link between headache and hormones. Postgrad Med 1995; 97: 147–53.

9. MacGregor

. ‘Menstrual’ migraine: towards a definition. Cephalalgia 1996; 16: 11–21.

10.

10. Somerville

. Estrogen-withdrawal migraine. I. Duration of exposure required and attempted prophylaxis by premenstrual estrogen administration. Neurology 1975; 25: 239–44.

11.

11. Somerville

. The role of estradiol withdrawal in the etiology of menstrual migraine. Neurology 1972; 22: 355–65.

12.

12. Cao

Welch

Aurora

Vikingstad

. Functional MRI-BOLD of visually triggered headache in patients with migraine. Arch Neurol 1999; 56: 548–54.

13.

Hadjikhani

Sanchez Del Rio

Schwartz

Bakker

Fischl

Mechanisms of migraine aura revealed by functional MRI in human visual cortex. Proc Natl Acad Sci USA 2001; 98: 4687–92.

14.

Dietrich

Krings

Neulen

Willmes

Erberich

Thron

Effects of blood estrogen level on cortical activation patterns during cognitive activation as measured by functional MRI. Neuroimage 2001; 13: 425–32.

15.

15. Veltman

Friston

Sanders

Price

. Regionally specific sensitivity differences in fMRI and PET. where do they come from? Neuroimage 2000; 11: 575–88.

16.

16. Olesen

. The International Headache Society classification and diagnostic criteria are valid and extremely useful. Cephalalgia 1996; 16: 293–6.

17.

Olesen

Lipton RB. Migraine classification and diagnosis. International Headache Society criteria. Neurology 1994; 44: S6–10.

18.

18. Olesen

Rasmussen

. The International Headache Society classification of chronic daily and near-daily headaches: a critique of the criticism, Cephalalgia 1996; 16: 407–11.

19.

19. Alsop

Detre

. Reduced transit-time sensitivity in noninvasive magnetic resonance imaging of human cerebral blood flow. J Cereb Blood Flow Metab 1996; 16: 1236–49.

20.

20. Detre

Alsop

. Perfusion magnetic resonance imaging with continuous arterial spin labeling. Methods and clinical applications in the central nervous system. Eur J Radiol 1999; 30: 115–24.

21.

21. Detre

Floyd

. Functional MRI and its applications to the clinical neurosciences. Neuroscientist 2001; 7: 64–79.

22.

22. Talairach

Tournoux

Musolino

Missir

. Stereotaxic exploration in frontal epilepsy. Adv Neurol 1992; 57: 651–88.

23.

23. McEwen

Alves

Bulloch

Weiland

. Ovarian steroids and the brain: implications for cognition and aging, Neurology 1997; 48: S8–15.

24.

24. Maki

Resnick

. Effects of estrogen on patterns of brain activity at rest and during cognitive activity: a review of neuroimaging studies, Neuroimage 2001; 14: 789–801.

25.

25. Rode

Wagner

Gunturkun

. Menstrual cycle affects functional cerebral asymmetries. Neuropsychologia 1995; 33: 855–65.

26.

26. Sanders

Wenmoth

. Verbal and music dichotic listening tasks reveal variations in functional cerebral asymmetry across the menstrual cycle that are phase and task dependent. Neuropsychologia 1998; 36: 869–74.

27.

27. Silberstein

. Headache and female hormones: what you need to know. Curr Opin Neurol 2001; 14: 323–33.

28.

28. Case

Reid

. Effects of the menstrual cycle on medical disorders. Arch Intern Med 1998; 158: 1405–12.

29.

29. Case

Reid

. Menstrual cycle effects on common medical conditions. Compr Ther 2001; 27: 65–71.

30.

30. Digre

Damasio

. Menstrual migraine: differential diagnosis, evaluation and treatment. Clin Obstet Gynecol 1987; 30: 417–30.

31.

31. Hurn

Littleton-Kearney

Kirsch

Dharmarajan

Traystman

. Postischemic cerebral blood flow recovery in the female: effect of 17 beta-estradiol. J Cereb Blood Flow Metab 1995; 15: 666–72.

32.

Salas

Lopez

Villarroya

Sanchez-Garcia

De Pascual

Dixon

Endothelium-independent relaxation by 17-alpha-estradiol of pig coronary arteries. Eur J Pharmacol 1994; 258: 47–55.

33.

Nattero

Allais

De Lorenzo

Benedetto

Zonca

Melzi

Relevance of prostaglandins in true menstrual migraine. Headache 1989; 29: 233–8.

34.