The Effects of Aerobic Exercise on Migraine
Donna-Marie C. Lockett, M.A. and J.F. Campbell, Ph.D.
Department of Psychology, Carleton University, Ottawa, Ont. Canada. Reprint requests to: Donna-Marie C. Lockett, Department of Psychology, Carleton University, Colonel By Drive, Ottawa, Ontario, K1S 5 B6. Accepted for Publication: October 1, 1991. SYNOPSIS
A six-week cardiovascular exercise program was provided to 11 subjects classified as experiencing classical migraines, while 9 similarly-classified subjects served as waiting-list controls. Measures included the Canadian Aerobic Fitness test, a headache diary to record the Frequency, Intensity, and Duration of migraine episodes and the Pain-Severity, Affective-Distress, and Support scales of the West Haven-Yale Multidimensional Pain Inventory (MPI). Measures were taken on both treatment and control subjects before, mid-way through, and upon termination of the first aerobic program, as well as after a two week follow-up. The aerobic classes were effective in significantly improving cardiovascular fitness. Pain Severity decreased significantly for those receiving aerobic training, who also showed (nonsignificant) trends, over the measurement periods, toward reductions in Affective Distress as well as the Frequency, Intensity and Duration of migraines, but these trends failed to reach statistical significance. Control subjects demonstrated no systematic changes in any of the dependent measures. These results suggest possible long-term benefits of aerobic fitness in the management of classical migraines. Key words: aerobic exercise, classical migraine Abbreviations: MPI West Haven-Yale Multidimensional Pain Iventory PAR-Q Physical Activity Readiness Questionnaire CAFT Canadian Aerobic Fitness Test (Headache 1992, 32:50-54) Work by Atkinson1 and Ekbom and Lindahl2 suggested a therapeutic role in the management of chronic headaches for aerobic exercise. In contrast, Thompson3 has reported that strenuous exercise may trigger a migraine prodrome. Others4,5 have found that vigorous activities such as lifting or bending may precipitate an exertional headache. Physiological changes initiated by heavy exertion, including a primary increase in oxygen consumption and an abrupt increase in oxygen demand' may have cerebrovascular consequences which contribute to the migraine episode. However, Lambert and Burnet7 have found that a proper 'warm up' prior to exercise may prevent exercise-related migraines. Given the generally negative impact of migraines, including their effects on the psychological functions,8-12 interpersonal relationships,13-15 as well as other features of living,16-18 and the association of aerobic exercise with various positive health benefits, including physiological,19-22 psychological,23-27 interpersonal,28,29 and others,30-32 a positive link between exercise and migraines seemed likely. More specifically, we proposed that those with migraines who demonstrated a positive effect from an aerobic exercise program might be expected to experience less physical and psychological distress, to have more satisfying interpersonal relationships, and to show improvements in general functioning. In addition, the potential for a physiological link between the cardiovascular effects of aerobic exercise and the possible cerebrovascular involvement in migraine etiology33-38 provided a rationale for an hypothesis suggesting that cardiovascular fitness training would reduce pain intensity, as well as the frequency and duration of migraine episodes. METHOD
Subjects. Female subjects were recruited from the Ottawa-Carleton region of Eastern Ontario through advertisements which were printed in various newspapers and posted in various stores throughout the region. Women were used exclusively in order to enhance the homogeneity of subject variables. Of the 33 women who responded, 26 met the criteria for classical migraines, as outlined below. Material. A modified version of the Questionnaire for One Headache39 was administered to confirm the self-reported experience of classical migraine obtained in a telephone interview. Subjects were presented with a list of criteria, which were based on a synthesis of the literature,39,40 and the recommended criteria as suggested by the Headache Classification Committee of the International Headache Society.41 In order to be included in the study, subjects were required to experience, during their migraines, one or more fully reversible aura symptoms, accompanied, usually, by a headache. A checklist of typical sensory and cognitive disturbances was provided for them, and included, among others, visual disturbances, hearing disturbances, gastrointestinal disturbances, mood disturbances, and thought disturbances. Individuals were included only if such symptoms were not the product of a known physical trauma or illness. The Physical Activity Readiness Questionnaire (PAR-Q)42 form is designed to identify those individuals for whom certain physical activities might be inappropriate, i.e those with heart conditions, high blood pressure. None of the subjects reported any physical conditions which might preclude their participating in the study.
Fitness levels were evaluated using the Canadian Aerobic Fitness Test (CAFT).42 This step test is designed to measure the overall functional capacity of the cardiovascular system, by predicting the greatest rate at which oxygen can be consumed during exercise (maximum oxygen uptake) from a submaximal test.43 This procedure for predicting maximum oxygen uptake has been found to be valid and reliable.44 The impact of the migraines on the participants' well being was assessed using the Pain Severity, Affective Distress, and Support scales (ranging from 0 to 6) of the West Haven-Yale Multidimensional Pain Inventory (MPI).49 The reliability and validity of the MPI have been reported by Kerns, Turk, and Rudy.45 Headache activity (Frequency, Intensity, and Duration) was monitored using a headache diary. The headache diary has been shown to be a valid indicator of treatment improvement.46 The headache diary, in this study, was kept by individual subjects on a standard record sheets provided by the researcher. The occurrence of a migraine, the severity of the pain (on a five point scale with intervals similar to those developed by Melzack47), and the duration of the pain (in hours and minutes) were recorded daily as appropriate. These measures were then averaged to yield a measure of Frequency (per week), Intensity (average intensity scores for that week), and duration (average duration of migraines for that week). Procedure. Prospective classical migraine subjects were initially screened by telephone. Screening was intended to achieve a preliminary confirmation of classical migraine and health status. Those women who claimed to suffer from classical migraines and who stated that an exercise program would pose no apparent health risk, were invited to meet with the researcher. At the initial meeting, participants began by completing the Physical Activity Readiness Questionnaire. The researcher then began by explaining the purpose of the fitness appraisal. Blood pressure and heart rate were monitored at the beginning, and every three minutes thereafter, for the duration of the appraisal. Participants were then asked to sign a consent form, indicating that they understood the fitness appraisal process and could stop the testing, and terminate their participation in the study at any time. Upon completion of the fitness appraisal, subjects were asked to complete the MPI. They were then introduced to the headache diary, which they were asked to monitor every night at bedtime, beginning the next day, for twelve weeks. Following the initial meetings subjects were randomly assigned to a treatment group or a waiting-list control group. Those in the treatment group were provided with aerobic classes which began after a four week baseline period. Those in the control group were told that, due to space limitations, they would begin their classes in 12 weeks from the initial meeting. The aerobic classes were administered and taught by a certified fitness instructor. In accordance with the recommendations of the Canadian Association of Sport Sciences and Fitness Canada,42 fitness classes were held three times a week for 45 minutes each, with each subject instructed to work within her target zone of 70-85% of maximal aerobic capacity.1 Classes focused on the development of aerobic capacity through low impact aerobic dancing and calisthenics. Included in each class was a 10 minute warm-up period, followed by a 25 minute aerobic work-out, and a ten minute cool down period.2 The aerobic program ran for six weeks, which is consistent with the length of time required to note significant changes in cardiovascular and autonomic functioning.48 Subjects were encouraged not to work out while experiencing a migraine. All subjects were asked to meet with the researcher, to submit their diaries, undergo the Canadian Aerobic Fitness Test, and to complete the MPI three weeks after the commencement of the first session of aerobic classes, three weeks following that date, and two weeks after the last class. After the final testing session, participants were debriefed and they were told that the results of the study would be provided to each of them as soon as they become available. They were each thanked for their participation, and arrangements were then made to provide aerobic fitness classes to the waiting-list control group. RESULTS
Descriptive Statistics. Among those 26 meeting the criteria for our study, one failed to complete the study due to outside commitments. Of the 25 women who completed the study, four, two in each of the treatment and control group, were excluded from the data analysis because they had not experienced any migraines during the baseline period, which precluded any measure of effect. Also excluded from the analysis was one woman, in the control group, who experienced migraines every day, making her not only an outlier, but also placing into question her status as a classical migraine sufferer. In the final analysis, then, were 20 women, 11 in the treatment group and 9 in the control group. The age of the participants ranged from 19 to 50, with a mean age of 32.5 for the treatment group, and 32.2 for the control group. Consistent with past research,49 these women were predominantly Caucasian, middle class, with some college education. Overall, participants in the treatment group had had migraines for an average of 18.3 years while the corresponding figure for the control group was 18.9 years. Both groups presented initially with average fitness levels. The treatment group's mean fitness level fell at the 44.8 percentile in relation to the average female Canadian population. Control subjects presented initially with an average fitness level falling at the 46.3 percentile. A series of t-tests revealed no significant differences between groups on age, duration of pain, and fitness level. Statistical Analysis. A series of mixed analyses of variance with two levels of the treatment variable (between) and four levels of time (within) were performed. Where the assumption of homogeneity of covariance was not met (Fitness and Support variables), the more conservative multivariate analysis of variance (MANOVA) was used. A Bonferroni correction was used to compensate for these multiple analyses. In addition, Bonferroni adjustments were made in the examination of simple and main effects. Figure I presents the results of each of the dependent variables for the treatment group (Group 1 ) and control group (Group 2), across each testing period (baseline [A], mid-treatment [B], post-treatment [C], and follow-up [D] ). Cardiovascular fitness. A MANOVA, using each
repetition of the dependent variable as separate dependent variables, indicated a significant difference in fitness level over time (Pillais = .629, p £ .006). Further analysis, using t-tests, revealed that this significance could be attributed to improvements in fitness experienced by the treatment group from time A to time B (44.8 to 55.2 percent), t(10) = 4.66, p £ .008. The control group, on the other hand, remained relatively stable in their fitness levels across these testing intervals (46.3 and 46.0 respectively), t(8) = .44, p>.008. Having established that the treatment was effective in improving fitness, we turn to an analysis of the effects of cardiovascular fitness on migraine activity and impact. Migraine activity variables. Three separate mixed ANOVAS for each of the migraine activity variables, Frequency (per week), Intensity (from 1-5), and Duration (average duration of migraine), while showing some improvements for the treatment group (Figure 1 ), failed to produce statistically significant effects. Migraine impact variables. The three migraine impact measures, taken from the MPI (Pain Severity, Affective Distress, and Support), were measured on seven point scales (ranging from 0 to 6). While each variable showed some trend towards improvement over the course of the study for those who exercised, only Pain Severity demonstrated a significant main effect for time, F(3,54) = 6.79, p £ .006. While a main effect for time might seem trivial, a simple effects analysis is warranted when seeking support for an a priori hypothesis.50 Simple Analyses of Variances revealed that a significant change in Pain Severity occurred from time A to B F(1,18) < 18.69, p £ .017 and was attributed predominantly to a significant drop in Pain Severity experienced by the treatment group (t [10] = 3.20, p £ .025). In contrast, the change in Pain Severity experienced by the control group was nonsignificant (t [8] = 1.93, p >.025). DISCUSSION
This study addressed the relationship between cardiovascular fitness and several measures of migraine impact and migraine activity. Aerobic fitness classes were found to be effective in improving cardiovascular fitness. Moreover, subjects who improved their cardiovascular fitness perceived their migraines, retrospectively, as being less painful and distressing as measured by the Pain Severity scale of the West Haven-Yale Multidimensional Pain Inventory. The effects of improved fitness on the remaining impact and activity variables were not statistically significant, although some positive trends did emerge. As was hypothesized, the Affective Distress scale of the MPI did show some (albeit nonsignificant) inclination toward improvement for those in the aerobic classes, during the course of their six week program. No such pattern was detected for the control group. This is consistent with research linking cardiovascular exercise to improvements in mood,23-27 as well as that linking improved psychological functioning with a reduction of migraines.8 Counter to our expectations, the Support scale of the West Haven-Yale Multidimensional Pain Inventory showed little change for either group over the course of the study. It may be that both partners in a
relationship must partake in physical fitness for improvement in their relationship to occur. The assessment of partners' lifestyles would likely enhance the interpretation of future studies. The Frequency, Intensity, and Duration of migraine episodes were assessed using a migraine diary. Over the course of the study, the treatment group experienced systematic, though nonsignificant, improvements on all three diary measures. No such patterns were detected for the control group. It may be tentatively suggested that those who became aerobically fit experienced changes in their reaction to stressful stimuli19-22 which may have altered their migraine experiences. More specifically, cardiovascular fitness may result in an increased tolerance for stress which may precipitate a migraine, thereby reducing the incidence of migraine episodes. Equally likely, cardiovascular fitness may moderate the stress response, such that the usual and dramatic parasympathetic counter-reaction, e.g. dilation of blood vessels and increase in blood flow, will also be moderated resulting in less painful headaches. Finally, being fit may enable one to recover more quickly from both the stressor which triggered the migraine process, and the migraine itself, resulting in shorter migraine episodes. A plausible explanation for the observed improvements in several variables by the treatment group may be that they had greater expectations for improvements. However, a two-week follow-up revealed that fitness levels for these subjects remained elevated while no significant regression in any of the activity or impact variables took place. In addition, the pattern of improvement for the treatment group was not uniform, as would be expected by a generalized expectancy effect. The results of this study provide a promising rational for persons with migraines to exercise. However, caution must be exercised not to generalize the results of this study to the entire classical migraine population, as this study focused exclusively on middle class, educated women who volunteered for a study and who were required to meet a strict inclusion criteria. Future research should incorporate longer exercise programs to investigate possible trends toward reduced negative affect, frequency, intensity, and duration of migraines. REFERENCES
1.
Atkinson R: Physical fitness and headache. Headache 1977; 17:189-191.
2.
Ekbom K, Lindahl J: Effects of induced rise of blood pressure on pain in cluster headache. Acta Neurol Scand 1980; 46:585-600.
3.
Thompson JK: Exercised-induced migraine prodrome symptoms. Headache 1987; 27:250-251.
4.
Rooke ED: Benign exertional headache. Med Clin North America 1968; 52(4):801-808.
5.
Symonds C: Cough Headache. Brain 1956; 79:557-568.
6.
Braun A, Klawans HL: Headache associated with exercise and sexual activity. In Rose, FC (ed): Handbook of Clinical Neurology, 4(48). New York, Elsevier Science Publishing Co. Inc., 1986, pp 373-382.
7.
Lambert RW, Burnet DL: Prevention of exercise induced migraine by quantitative warm-up. Headache 1985; 25:317-319.
8.
Henryk-Gutt R, Rees L: Psychological aspects of migraine. Journal of Psychosomatic Research 1973; 17:141-153.
9.
Andrasik F, Blanchard E, Arena J, Teders S, Teevan R, Rodichok L: Psychological functioning in headache sufferers. Psychosomatic Medicine 1982; 44: 171-181.
10.
Friedman AP, yon Storch TJC, Merritt HH: Migraine and tension headaches: a clinical study of 2000 cases. Neurology 1954; 4:773-788.
11.
Merikangas KR, Risch NJ, Merikangas JR, Weissman MM, Kidd KK: Migraine and depression: Association and familial transmission. Journal of Psychiatric Research 1988; 22(2):119-129.
12.
Wolff HG: Personality factors and reactions of subjects with migraine. Archives of Neurological Psychiatry 1937; 37:895.
13.
Philips C, Jahanshahi M: The effects of persistent pain: the chronic pain sufferer. Pain 1985; 21:163-176.
14.
Craig KD: Ontogenetic and cultural influences on the expression of pain in man. In Kosterlitz HW, Terenius LY (eds): Pain and Society, Weinheim, Verlag Chemic, 1980, pp 37-52.
15.
Bond MR: The suffering of severe intractable pain. In Kosterlitz HW, Terenius LY (eds): Pain and Society, Weinheim, Verlag Chemic, 1980, pp 53-62.
16.
Beckstead J: An examination of the perceived impact of migraine and chronic headaches on personal project system. Unpublished Thesis, Carleton University, Ottawa, Canada, 1988.
17.
Levor R, Cohen M, Naliboff B, McArthur D: Psychosocial precursors and correlates of migraine headaches. Journal of Consulting and Clinical Psychology 1986; 54:347-353.
18.
Zenz H: Social aspects of the management of headache patients. In Holroyd KA, Schlote B, Zenz H (eds): Perspectives in research on headache, Lewiston, NY, C. J. Hogrefe, Inc., 1983, pp 105-114.
19.
Gal R, Lazarus RS: The role of activity in anticipating and confronting stressful situations. Journal of Human Stress 1975; 1:4-20.
20.
Hollander BJ, Seraganian P: Aerobic fitness and psychophysiological reactivity. Canadian Journal of Behaviour Science 1984; 16(3):257-261.
21.
Keller S, Saraganian P: Physical fitness level and autonomic reactivity to psychosocial stress. Journal of Psychosomatic Research 1984; 28:279-287.
22.
Sinyor D, Schwartz S, Peronnet F, Brisson G, Seraganian P: Aerobic fitness level and reactivity to psychosocial stress: Physiological, biochemical, and subjective measures. Psychosomatic Medicine 1983; 45(3):205-217.
23.
Brown RS, Ramirez DE, Taub JM: The prescription of exercise for depression. Phys Sportsmed 1978; 6:34-45.
24.
Greist JH, Klein MH, Eichens RR, Farris J, Goffman A, Morgan WP: Running through your mind. Journal of Psychosomatic Research 1978; 22:259-294.
25.
Ledwidge B: Run for your mind: Aerobic exercise as a means of alleviating anxiety and depression. Canadian Journal of Behaviour Science Review 1980; 12:126-140.
26.
Morgan WP: Psychological benefits of physical activity. In Nagle F, Montage H (eds): Exercise in Health and Disease. Springfield, Illinois, Charles C. Thomas, 1981.
27.
Greist JH, Klein MH, Eichens RR, Farris J, Gurman AS: Running as a treatment for depression. Comprehensive Psychiatry 1979; 20:41-54.
28.
Harris TG, Gurin J: Look who's getting it all together. American Health 1985; March: 42-47.
29.
Harris TG, Kagan J: The fitness advantage. American Health 1985; March: 12-15.
30.
Donoghue S: The correlation between physical fitness, absenteeism, and work performance. Canadian Journal of Public Health 1977; 68:201-203.
31.
Astrand PO: Health & Fitness. New York, Barron's, 1977.
32.
Blumenthal JA, Williams RS, Needles TL, Wallace AG: Psychological changes accompany aerobic exercise in healthy middle-aged adults. Psychosomatic Medicine 1982; 44(6):529-536.
33.
Ellertson B, Hammerborg D: Psychophysiological response patterns in migraine patients. Cephalalgia 1982; 2:19-24.
34.
Johnson ES: A basis for migraine therapy-The autonomic theory reappraised. Postgraduate Medical Journal 1978; 54:231-242.
35.
Sicuteri F: Vaso-neuroactive substances and their implications in vascular pain. Research and Clinical Studies in Headache 1967; 1:6-45.
36.
Anderson CD, Stoyva JM, Vaughn LJ: A test of delayed recovery following stressful stimulation in four psychosomatic disorders. Journal of Psychosomatic Research 1982; 26:571-580.
37.
Anthony M: Biochemical indices of sympathetic activity in migraine. Cephalalgia 1981; 1:83-89.
38.
Drummond PD: Extracranial and cardiovascular reactivity in migrainous subjects, Journal of Psychosomatic Research 1982; 26:317-331.
39.
Campbell J: Carleton University Pain Project: Questionnaire for One Headache. Unpublished Manuscript, Carleton University, Ottawa, Canada, 1988.
40.
Adams HE, Feuerstein M, Fowler JL: Migraine headache: review of parameters, etiology, and intervention. Psychological Bulletin 1980; 87:217-237.
41.
Headache Classification Committee of the International Headache Society. Cephalalgia 1988; 8(7):20-25.
42.
Canadian Association of Sports Sciences and Fitness Canada: Canadian Standardized Test of Fitness (CSTF), Ottawa, Minister of Supply and Services Canada, 1986.
43.
Jette M: A comparison between predicted V02 max. from the Astrand Procedure and the Canadian Home Fitness Test. Canadian Journal of Applied Sport Sciences 1979; 4:214-218.
44.
Shephard RJ: Development of the Canadian Home Fitness Test. Canadian Medical Association Journal 1976; 114: 675-679.
45.
Kerns RD, Turk DC, Rudy TE: The West Haven-Yale Multidimensional Pain Inventory. Pain 1985; 23:345-356.
46.
Follick MJ, Ahern DK, Laser-Wolston N: Evaluation of a daily activity diary for chronic pain patients, Pain 1984; 19:373-382.
47.
Melzack R: The McGill Pain Questionnaire: Major properties and scoring methods. Pain 1975; 1:277-299.
48.
Hickson R: Time course of the adaptation response of aerobic power and heart rate to training. Med Sci Sports Ex 1981; 13:17-20.
49.
Mitchell KR, White RG: Behaviourial self-management: An application to the problem of migraine headaches. Behaviour Therapy 1977; 8:213-221.
50.
Howell DC: Statistical Methods for Psychology. Boston, Duxbury Press, 1987.
Donna-Marie C. Lockett, M.A. and J.F. Campbell, Ph.D.
Department of Psychology, Carleton University, Ottawa, Ont. Canada. Reprint requests to: Donna-Marie C. Lockett, Department of Psychology, Carleton University, Colonel By Drive, Ottawa, Ontario, K1S 5 B6. Accepted for Publication: October 1, 1991. SYNOPSIS
A six-week cardiovascular exercise program was provided to 11 subjects classified as experiencing classical migraines, while 9 similarly-classified subjects served as waiting-list controls. Measures included the Canadian Aerobic Fitness test, a headache diary to record the Frequency, Intensity, and Duration of migraine episodes and the Pain-Severity, Affective-Distress, and Support scales of the West Haven-Yale Multidimensional Pain Inventory (MPI). Measures were taken on both treatment and control subjects before, mid-way through, and upon termination of the first aerobic program, as well as after a two week follow-up. The aerobic classes were effective in significantly improving cardiovascular fitness. Pain Severity decreased significantly for those receiving aerobic training, who also showed (nonsignificant) trends, over the measurement periods, toward reductions in Affective Distress as well as the Frequency, Intensity and Duration of migraines, but these trends failed to reach statistical significance. Control subjects demonstrated no systematic changes in any of the dependent measures. These results suggest possible long-term benefits of aerobic fitness in the management of classical migraines. Key words: aerobic exercise, classical migraine Abbreviations: MPI West Haven-Yale Multidimensional Pain Iventory PAR-Q Physical Activity Readiness Questionnaire CAFT Canadian Aerobic Fitness Test (Headache 1992, 32:50-54) Work by Atkinson1 and Ekbom and Lindahl2 suggested a therapeutic role in the management of chronic headaches for aerobic exercise. In contrast, Thompson3 has reported that strenuous exercise may trigger a migraine prodrome. Others4,5 have found that vigorous activities such as lifting or bending may precipitate an exertional headache. Physiological changes initiated by heavy exertion, including a primary increase in oxygen consumption and an abrupt increase in oxygen demand' may have cerebrovascular consequences which contribute to the migraine episode. However, Lambert and Burnet7 have found that a proper 'warm up' prior to exercise may prevent exercise-related migraines. Given the generally negative impact of migraines, including their effects on the psychological functions,8-12 interpersonal relationships,13-15 as well as other features of living,16-18 and the association of aerobic exercise with various positive health benefits, including physiological,19-22 psychological,23-27 interpersonal,28,29 and others,30-32 a positive link between exercise and migraines seemed likely. More specifically, we proposed that those with migraines who demonstrated a positive effect from an aerobic exercise program might be expected to experience less physical and psychological distress, to have more satisfying interpersonal relationships, and to show improvements in general functioning. In addition, the potential for a physiological link between the cardiovascular effects of aerobic exercise and the possible cerebrovascular involvement in migraine etiology33-38 provided a rationale for an hypothesis suggesting that cardiovascular fitness training would reduce pain intensity, as well as the frequency and duration of migraine episodes. METHOD
Subjects. Female subjects were recruited from the Ottawa-Carleton region of Eastern Ontario through advertisements which were printed in various newspapers and posted in various stores throughout the region. Women were used exclusively in order to enhance the homogeneity of subject variables. Of the 33 women who responded, 26 met the criteria for classical migraines, as outlined below. Material. A modified version of the Questionnaire for One Headache39 was administered to confirm the self-reported experience of classical migraine obtained in a telephone interview. Subjects were presented with a list of criteria, which were based on a synthesis of the literature,39,40 and the recommended criteria as suggested by the Headache Classification Committee of the International Headache Society.41 In order to be included in the study, subjects were required to experience, during their migraines, one or more fully reversible aura symptoms, accompanied, usually, by a headache. A checklist of typical sensory and cognitive disturbances was provided for them, and included, among others, visual disturbances, hearing disturbances, gastrointestinal disturbances, mood disturbances, and thought disturbances. Individuals were included only if such symptoms were not the product of a known physical trauma or illness. The Physical Activity Readiness Questionnaire (PAR-Q)42 form is designed to identify those individuals for whom certain physical activities might be inappropriate, i.e those with heart conditions, high blood pressure. None of the subjects reported any physical conditions which might preclude their participating in the study.
Fitness levels were evaluated using the Canadian Aerobic Fitness Test (CAFT).42 This step test is designed to measure the overall functional capacity of the cardiovascular system, by predicting the greatest rate at which oxygen can be consumed during exercise (maximum oxygen uptake) from a submaximal test.43 This procedure for predicting maximum oxygen uptake has been found to be valid and reliable.44 The impact of the migraines on the participants' well being was assessed using the Pain Severity, Affective Distress, and Support scales (ranging from 0 to 6) of the West Haven-Yale Multidimensional Pain Inventory (MPI).49 The reliability and validity of the MPI have been reported by Kerns, Turk, and Rudy.45 Headache activity (Frequency, Intensity, and Duration) was monitored using a headache diary. The headache diary has been shown to be a valid indicator of treatment improvement.46 The headache diary, in this study, was kept by individual subjects on a standard record sheets provided by the researcher. The occurrence of a migraine, the severity of the pain (on a five point scale with intervals similar to those developed by Melzack47), and the duration of the pain (in hours and minutes) were recorded daily as appropriate. These measures were then averaged to yield a measure of Frequency (per week), Intensity (average intensity scores for that week), and duration (average duration of migraines for that week). Procedure. Prospective classical migraine subjects were initially screened by telephone. Screening was intended to achieve a preliminary confirmation of classical migraine and health status. Those women who claimed to suffer from classical migraines and who stated that an exercise program would pose no apparent health risk, were invited to meet with the researcher. At the initial meeting, participants began by completing the Physical Activity Readiness Questionnaire. The researcher then began by explaining the purpose of the fitness appraisal. Blood pressure and heart rate were monitored at the beginning, and every three minutes thereafter, for the duration of the appraisal. Participants were then asked to sign a consent form, indicating that they understood the fitness appraisal process and could stop the testing, and terminate their participation in the study at any time. Upon completion of the fitness appraisal, subjects were asked to complete the MPI. They were then introduced to the headache diary, which they were asked to monitor every night at bedtime, beginning the next day, for twelve weeks. Following the initial meetings subjects were randomly assigned to a treatment group or a waiting-list control group. Those in the treatment group were provided with aerobic classes which began after a four week baseline period. Those in the control group were told that, due to space limitations, they would begin their classes in 12 weeks from the initial meeting. The aerobic classes were administered and taught by a certified fitness instructor. In accordance with the recommendations of the Canadian Association of Sport Sciences and Fitness Canada,42 fitness classes were held three times a week for 45 minutes each, with each subject instructed to work within her target zone of 70-85% of maximal aerobic capacity.1 Classes focused on the development of aerobic capacity through low impact aerobic dancing and calisthenics. Included in each class was a 10 minute warm-up period, followed by a 25 minute aerobic work-out, and a ten minute cool down period.2 The aerobic program ran for six weeks, which is consistent with the length of time required to note significant changes in cardiovascular and autonomic functioning.48 Subjects were encouraged not to work out while experiencing a migraine. All subjects were asked to meet with the researcher, to submit their diaries, undergo the Canadian Aerobic Fitness Test, and to complete the MPI three weeks after the commencement of the first session of aerobic classes, three weeks following that date, and two weeks after the last class. After the final testing session, participants were debriefed and they were told that the results of the study would be provided to each of them as soon as they become available. They were each thanked for their participation, and arrangements were then made to provide aerobic fitness classes to the waiting-list control group. RESULTS
Descriptive Statistics. Among those 26 meeting the criteria for our study, one failed to complete the study due to outside commitments. Of the 25 women who completed the study, four, two in each of the treatment and control group, were excluded from the data analysis because they had not experienced any migraines during the baseline period, which precluded any measure of effect. Also excluded from the analysis was one woman, in the control group, who experienced migraines every day, making her not only an outlier, but also placing into question her status as a classical migraine sufferer. In the final analysis, then, were 20 women, 11 in the treatment group and 9 in the control group. The age of the participants ranged from 19 to 50, with a mean age of 32.5 for the treatment group, and 32.2 for the control group. Consistent with past research,49 these women were predominantly Caucasian, middle class, with some college education. Overall, participants in the treatment group had had migraines for an average of 18.3 years while the corresponding figure for the control group was 18.9 years. Both groups presented initially with average fitness levels. The treatment group's mean fitness level fell at the 44.8 percentile in relation to the average female Canadian population. Control subjects presented initially with an average fitness level falling at the 46.3 percentile. A series of t-tests revealed no significant differences between groups on age, duration of pain, and fitness level. Statistical Analysis. A series of mixed analyses of variance with two levels of the treatment variable (between) and four levels of time (within) were performed. Where the assumption of homogeneity of covariance was not met (Fitness and Support variables), the more conservative multivariate analysis of variance (MANOVA) was used. A Bonferroni correction was used to compensate for these multiple analyses. In addition, Bonferroni adjustments were made in the examination of simple and main effects. Figure I presents the results of each of the dependent variables for the treatment group (Group 1 ) and control group (Group 2), across each testing period (baseline [A], mid-treatment [B], post-treatment [C], and follow-up [D] ). Cardiovascular fitness. A MANOVA, using each
repetition of the dependent variable as separate dependent variables, indicated a significant difference in fitness level over time (Pillais = .629, p £ .006). Further analysis, using t-tests, revealed that this significance could be attributed to improvements in fitness experienced by the treatment group from time A to time B (44.8 to 55.2 percent), t(10) = 4.66, p £ .008. The control group, on the other hand, remained relatively stable in their fitness levels across these testing intervals (46.3 and 46.0 respectively), t(8) = .44, p>.008. Having established that the treatment was effective in improving fitness, we turn to an analysis of the effects of cardiovascular fitness on migraine activity and impact. Migraine activity variables. Three separate mixed ANOVAS for each of the migraine activity variables, Frequency (per week), Intensity (from 1-5), and Duration (average duration of migraine), while showing some improvements for the treatment group (Figure 1 ), failed to produce statistically significant effects. Migraine impact variables. The three migraine impact measures, taken from the MPI (Pain Severity, Affective Distress, and Support), were measured on seven point scales (ranging from 0 to 6). While each variable showed some trend towards improvement over the course of the study for those who exercised, only Pain Severity demonstrated a significant main effect for time, F(3,54) = 6.79, p £ .006. While a main effect for time might seem trivial, a simple effects analysis is warranted when seeking support for an a priori hypothesis.50 Simple Analyses of Variances revealed that a significant change in Pain Severity occurred from time A to B F(1,18) < 18.69, p £ .017 and was attributed predominantly to a significant drop in Pain Severity experienced by the treatment group (t [10] = 3.20, p £ .025). In contrast, the change in Pain Severity experienced by the control group was nonsignificant (t [8] = 1.93, p >.025). DISCUSSION
This study addressed the relationship between cardiovascular fitness and several measures of migraine impact and migraine activity. Aerobic fitness classes were found to be effective in improving cardiovascular fitness. Moreover, subjects who improved their cardiovascular fitness perceived their migraines, retrospectively, as being less painful and distressing as measured by the Pain Severity scale of the West Haven-Yale Multidimensional Pain Inventory. The effects of improved fitness on the remaining impact and activity variables were not statistically significant, although some positive trends did emerge. As was hypothesized, the Affective Distress scale of the MPI did show some (albeit nonsignificant) inclination toward improvement for those in the aerobic classes, during the course of their six week program. No such pattern was detected for the control group. This is consistent with research linking cardiovascular exercise to improvements in mood,23-27 as well as that linking improved psychological functioning with a reduction of migraines.8 Counter to our expectations, the Support scale of the West Haven-Yale Multidimensional Pain Inventory showed little change for either group over the course of the study. It may be that both partners in a
relationship must partake in physical fitness for improvement in their relationship to occur. The assessment of partners' lifestyles would likely enhance the interpretation of future studies. The Frequency, Intensity, and Duration of migraine episodes were assessed using a migraine diary. Over the course of the study, the treatment group experienced systematic, though nonsignificant, improvements on all three diary measures. No such patterns were detected for the control group. It may be tentatively suggested that those who became aerobically fit experienced changes in their reaction to stressful stimuli19-22 which may have altered their migraine experiences. More specifically, cardiovascular fitness may result in an increased tolerance for stress which may precipitate a migraine, thereby reducing the incidence of migraine episodes. Equally likely, cardiovascular fitness may moderate the stress response, such that the usual and dramatic parasympathetic counter-reaction, e.g. dilation of blood vessels and increase in blood flow, will also be moderated resulting in less painful headaches. Finally, being fit may enable one to recover more quickly from both the stressor which triggered the migraine process, and the migraine itself, resulting in shorter migraine episodes. A plausible explanation for the observed improvements in several variables by the treatment group may be that they had greater expectations for improvements. However, a two-week follow-up revealed that fitness levels for these subjects remained elevated while no significant regression in any of the activity or impact variables took place. In addition, the pattern of improvement for the treatment group was not uniform, as would be expected by a generalized expectancy effect. The results of this study provide a promising rational for persons with migraines to exercise. However, caution must be exercised not to generalize the results of this study to the entire classical migraine population, as this study focused exclusively on middle class, educated women who volunteered for a study and who were required to meet a strict inclusion criteria. Future research should incorporate longer exercise programs to investigate possible trends toward reduced negative affect, frequency, intensity, and duration of migraines. REFERENCES
1.
Atkinson R: Physical fitness and headache. Headache 1977; 17:189-191.
2.
Ekbom K, Lindahl J: Effects of induced rise of blood pressure on pain in cluster headache. Acta Neurol Scand 1980; 46:585-600.
3.
Thompson JK: Exercised-induced migraine prodrome symptoms. Headache 1987; 27:250-251.
4.
Rooke ED: Benign exertional headache. Med Clin North America 1968; 52(4):801-808.
5.
Symonds C: Cough Headache. Brain 1956; 79:557-568.
6.
Braun A, Klawans HL: Headache associated with exercise and sexual activity. In Rose, FC (ed): Handbook of Clinical Neurology, 4(48). New York, Elsevier Science Publishing Co. Inc., 1986, pp 373-382.
7.
Lambert RW, Burnet DL: Prevention of exercise induced migraine by quantitative warm-up. Headache 1985; 25:317-319.
8.
Henryk-Gutt R, Rees L: Psychological aspects of migraine. Journal of Psychosomatic Research 1973; 17:141-153.
9.
Andrasik F, Blanchard E, Arena J, Teders S, Teevan R, Rodichok L: Psychological functioning in headache sufferers. Psychosomatic Medicine 1982; 44: 171-181.
10.
Friedman AP, yon Storch TJC, Merritt HH: Migraine and tension headaches: a clinical study of 2000 cases. Neurology 1954; 4:773-788.
11.
Merikangas KR, Risch NJ, Merikangas JR, Weissman MM, Kidd KK: Migraine and depression: Association and familial transmission. Journal of Psychiatric Research 1988; 22(2):119-129.
12.
Wolff HG: Personality factors and reactions of subjects with migraine. Archives of Neurological Psychiatry 1937; 37:895.
13.
Philips C, Jahanshahi M: The effects of persistent pain: the chronic pain sufferer. Pain 1985; 21:163-176.
14.
Craig KD: Ontogenetic and cultural influences on the expression of pain in man. In Kosterlitz HW, Terenius LY (eds): Pain and Society, Weinheim, Verlag Chemic, 1980, pp 37-52.
15.
Bond MR: The suffering of severe intractable pain. In Kosterlitz HW, Terenius LY (eds): Pain and Society, Weinheim, Verlag Chemic, 1980, pp 53-62.
16.
Beckstead J: An examination of the perceived impact of migraine and chronic headaches on personal project system. Unpublished Thesis, Carleton University, Ottawa, Canada, 1988.
17.
Levor R, Cohen M, Naliboff B, McArthur D: Psychosocial precursors and correlates of migraine headaches. Journal of Consulting and Clinical Psychology 1986; 54:347-353.
18.
Zenz H: Social aspects of the management of headache patients. In Holroyd KA, Schlote B, Zenz H (eds): Perspectives in research on headache, Lewiston, NY, C. J. Hogrefe, Inc., 1983, pp 105-114.
19.
Gal R, Lazarus RS: The role of activity in anticipating and confronting stressful situations. Journal of Human Stress 1975; 1:4-20.
20.
Hollander BJ, Seraganian P: Aerobic fitness and psychophysiological reactivity. Canadian Journal of Behaviour Science 1984; 16(3):257-261.
21.
Keller S, Saraganian P: Physical fitness level and autonomic reactivity to psychosocial stress. Journal of Psychosomatic Research 1984; 28:279-287.
22.
Sinyor D, Schwartz S, Peronnet F, Brisson G, Seraganian P: Aerobic fitness level and reactivity to psychosocial stress: Physiological, biochemical, and subjective measures. Psychosomatic Medicine 1983; 45(3):205-217.
23.
Brown RS, Ramirez DE, Taub JM: The prescription of exercise for depression. Phys Sportsmed 1978; 6:34-45.
24.
Greist JH, Klein MH, Eichens RR, Farris J, Goffman A, Morgan WP: Running through your mind. Journal of Psychosomatic Research 1978; 22:259-294.
25.
Ledwidge B: Run for your mind: Aerobic exercise as a means of alleviating anxiety and depression. Canadian Journal of Behaviour Science Review 1980; 12:126-140.
26.
Morgan WP: Psychological benefits of physical activity. In Nagle F, Montage H (eds): Exercise in Health and Disease. Springfield, Illinois, Charles C. Thomas, 1981.
27.
Greist JH, Klein MH, Eichens RR, Farris J, Gurman AS: Running as a treatment for depression. Comprehensive Psychiatry 1979; 20:41-54.
28.
Harris TG, Gurin J: Look who's getting it all together. American Health 1985; March: 42-47.
29.
Harris TG, Kagan J: The fitness advantage. American Health 1985; March: 12-15.
30.
Donoghue S: The correlation between physical fitness, absenteeism, and work performance. Canadian Journal of Public Health 1977; 68:201-203.
31.
Astrand PO: Health & Fitness. New York, Barron's, 1977.
32.
Blumenthal JA, Williams RS, Needles TL, Wallace AG: Psychological changes accompany aerobic exercise in healthy middle-aged adults. Psychosomatic Medicine 1982; 44(6):529-536.
33.
Ellertson B, Hammerborg D: Psychophysiological response patterns in migraine patients. Cephalalgia 1982; 2:19-24.
34.
Johnson ES: A basis for migraine therapy-The autonomic theory reappraised. Postgraduate Medical Journal 1978; 54:231-242.
35.
Sicuteri F: Vaso-neuroactive substances and their implications in vascular pain. Research and Clinical Studies in Headache 1967; 1:6-45.
36.
Anderson CD, Stoyva JM, Vaughn LJ: A test of delayed recovery following stressful stimulation in four psychosomatic disorders. Journal of Psychosomatic Research 1982; 26:571-580.
37.
Anthony M: Biochemical indices of sympathetic activity in migraine. Cephalalgia 1981; 1:83-89.
38.
Drummond PD: Extracranial and cardiovascular reactivity in migrainous subjects, Journal of Psychosomatic Research 1982; 26:317-331.
39.
Campbell J: Carleton University Pain Project: Questionnaire for One Headache. Unpublished Manuscript, Carleton University, Ottawa, Canada, 1988.
40.
Adams HE, Feuerstein M, Fowler JL: Migraine headache: review of parameters, etiology, and intervention. Psychological Bulletin 1980; 87:217-237.
41.
Headache Classification Committee of the International Headache Society. Cephalalgia 1988; 8(7):20-25.
42.
Canadian Association of Sports Sciences and Fitness Canada: Canadian Standardized Test of Fitness (CSTF), Ottawa, Minister of Supply and Services Canada, 1986.
43.
Jette M: A comparison between predicted V02 max. from the Astrand Procedure and the Canadian Home Fitness Test. Canadian Journal of Applied Sport Sciences 1979; 4:214-218.
44.
Shephard RJ: Development of the Canadian Home Fitness Test. Canadian Medical Association Journal 1976; 114: 675-679.
45.
Kerns RD, Turk DC, Rudy TE: The West Haven-Yale Multidimensional Pain Inventory. Pain 1985; 23:345-356.
46.
Follick MJ, Ahern DK, Laser-Wolston N: Evaluation of a daily activity diary for chronic pain patients, Pain 1984; 19:373-382.
47.
Melzack R: The McGill Pain Questionnaire: Major properties and scoring methods. Pain 1975; 1:277-299.
48.
Hickson R: Time course of the adaptation response of aerobic power and heart rate to training. Med Sci Sports Ex 1981; 13:17-20.
49.
Mitchell KR, White RG: Behaviourial self-management: An application to the problem of migraine headaches. Behaviour Therapy 1977; 8:213-221.
50.
Howell DC: Statistical Methods for Psychology. Boston, Duxbury Press, 1987.
