Exercise and headache

Management of migraine is complex, and involves both lifestyle modification measures, and pharmacotherapy options to manage acute attacks and prevent future attacks. As part of the multidisciplinary treatment approach for migraine, exercise is frequently recommended. In a critical review published in 2008, exercise as a therapy for migraine was graded as evidence level B-C.

Individuals with migraine have been shown to be less physically active. In a large study ( n = 48,713) physical inactivity was noted in 68.4%–68.7% of patients with migraine, compared to 63.6% of patients without migraine. Patients with migraine also seem to have a reduced aerobic capacity. A survey-based study looking at patients with headache versus control patients ( n = 58) found a significant reduction in aerobic endurance in migraine patients and significantly higher body fat in female migraineurs compared to controls.

Obesity is associated with headache and migraine and is felt to be a negative prognostic marker in the development of chronic migraine. In a prospective cross-sectional study in adolescents ( n = 273), overweight females had a 4-fold greater risk of headache when compared to normal weight females. A further study in pediatrics ( n = 913) showed that body mass index (BMI) was associated with both headache frequency and disability and that weight loss was associated with a reduction in headache frequency. The association between obesity and chronic headache was duplicated in two further adolescent studies, a retrospective cross-sectional study ( n = 925) and a prospective study ( n = 3342). A retrospective cross-sectional trial in adults ( n = 181) noted that obesity was specifically a risk factor for migraine, not tension type headache, and that obesity seemed to be related to both the frequency and disability of headache. In a small trial of adolescents with migraine on the ketogenic diet and with weight loss, a significant 50-point reduction in PedMIDAS scores, a measure of disability, was observed. A further prospective cross-sectional study in adolescents ( n = 135) found that weight loss led to improvement of migraine. In obese adolescents with migraine who participated in a 12-month long intervention for weight loss, a significant reduction in both adiposity and headache was noted at the 6-month mark and maintained throughout the second 6 months. Better migraine outcomes appeared to be associated with the percent change in BMI. In this trial, the exercise ‘prescribed’ included 60 min of moderate intensity activity, preferably daily, in addition to a balanced diet and cognitive behavioral therapy (CBT). In addition to the significant changes in adiposity, BMI, and headaches frequency, there were also improvements noted in headache intensity, PedMIDAS scores and the use of acute medications.

Physical inactivity has been shown to be associated with an increased prevalence of future migraine. In a large longitudinal trial ( n = 22,397) over an 11-year period, low physical activity was associated with higher prevalence of migraine and nonmigraine headache, and there was also a linear trend showing that higher prevalence of ‘low physical activity’ was associated with increased headache frequency. A more recent study ( n = 3,124) noted that, in women only, those with < 30 min of exercise per week had a 60% higher risk of later developing migraine. A further study ( n = 148) noted that the risk of migraine was 4.4-fold higher in those with low physical activity (defined as < 30 min of moderate intensity exercise on most days) versus those with high levels of physical activity (1 h per day or more). When looking at VO2-peak by ergospirometry as an index of physical fitness in adults ( n = 3899), an inverse relationship was observed whereby a higher level of physical fitness was associated with reduced migraine prevalence. Interestingly, in the same study, those with the lowest VO2-peak quintile had the strongest association for experiencing migraine aggravated by physical activity. This might suggest that exercise is a trigger for migraine only in those with low baseline physical fitness levels. In a more recent study by the same group, a sample of 15,276 adults were followed over an 11-year period. During this study, those with 1–3 h of “low intensity” physical exercise per week had a 22% lower risk of developing migraine compared to inactivity. Those with 1–2 h of “hard intensity” physical exercise per week had a 29% lower risk of developing migraine. Finally, in adolescents ( n = 1260), a prospective study noted that in those with a low-level metabolic exercise, there was a 4.2-fold higher risk of migraine when compared to those with a higher level of metabolic exercise. Using a different study population, a 2002 study was completed in 791 division one (D1) basketball players in the USA and found that the prevalence of migraine (2.9% overall, 0.9% of men, 4.4% of women) in this very active cohort was less than that of the general population.

Multiple studies have shown that the regular practice of exercise is associated with a reduction in headache frequency, duration, intensity, and improvement in overall wellbeing, quality of life, cognitive outcomes, and improved mood. A few early trials noted that migraines decreased as VO2 max, a measure of fitness, increased, that pain severity improved as VO2 max increased and that all of frequency, intensity, and duration of headaches improved as cardiovascular fitness increased. Using chemiluminescence to analyze plasma nitric oxide, adult women with migraine ( n = 40) were nonrandomly placed in the exercise group (1 h of aerobic exercise three times weekly) or the standard of care group. Those in the exercise group showed a significant reduction in pain severity scores (measured using the Visual Analogue Score, VAS), migraine frequency, duration, and improved quality of life. The authors postulated that the improvement might be mediated by the increased production of nitric oxide with regular exercise. Further work looking at beta endorphin levels in patients with migraine ( n = 40) found a reduction in migraine frequency, duration, and intensity in association with an increase beta endorphin levels both acutely and after 6 weeks of exercise. The correlation between beta endorphins and the headache parameters was difficult to ascertain from the data, but the authors note that, curiously, those with low preexercise beta endorphin levels appeared to have the greatest improvement in headache parameters post exercise, and also that those with lower beta endorphin levels seem to have longer duration of headaches. In adults with episodic migraine ( n = 48) randomized into either high intensity training (HIT), moderate continuous exercise training (MCT) or usual activities (control group) for a 12-week program, it was shown that both types of exercise lead to a reduction of migraines. HIT seemed to be most effective (− 63% of attacks) vs MCT (− 26% of attacks). HIT also seemed to be more beneficial for retinal vasculature outcomes, which was used as a measure of cerebrovascular health.

Other trials have emphasized improvement in measures beyond headache parameters. In adult patients ( n = 30) randomized to aerobic exercise in addition to standard medical care or standard medical care only, a significant reduction in migraine pain intensity was noted after the 6-week intervention and also a trend towards improved depression-related symptoms was found with the use of exercise. Others have noted that in adult patients with migraine ( n = 26), an exercise program consisting of aerobic exercise for 40 min, 3 times a week for 12 weeks not only led to a reduction in migraine days (7.5 at baseline to 5.4 days per month in month 3), migraine intensity and the use of acute medications, but also to an increase in quality of life. Most importantly, in this specific trial, the exercise program was well tolerated, and no patients noted deterioration in their migraine status after commencing the exercise program. In medical students ( n = 480), a survey study noted that those who were physically active had less functional disability, as measured by the PedMIDAS, from their migraine headaches. In migraineurs with coexisting tension type headache (TTH) and/or neck pain ( n = 52) randomized into aerobic exercise (45 min 3 days a week for 3 months) versus controls, the within group outcomes at 6 months revealed a reduction in migraine days in the exercise group in those with migraine. More specifically, in those with chronic migraine, headache frequency reduced from 16 to 9 days and in those with episodic migraine, from 7 to 5 days. There was also a within group reduction for migraine duration, pain intensity and neck pain. Quality of life and ability to engage in activities also improved between groups leading to a significant reduction in ‘migraine burden’ in the exercise group. Finally, in addition, to migraine frequency reduction, a further trial showed that exercise therapy also led to improved cognitive function (information processing and attention) in migraineurs versus controls.

When compared to standard of care therapeutics for migraine, exercise appears to be at least noninferior to pharmaceutical treatments and potentially additive. In a 3-arm trial looking at adults with episodic migraine ( n = 91), submaximal exercise (3 times a week for 3 months) was compared to topiramate (to the highest tolerated dose, with a maximum dose of 200 mg/day) or relaxation. All interventions were found to be equivalent with regards to their ability to reduce headache frequency, but adverse events only occurred in the topiramate group. In a subsequent treatment trial in adults with chronic migraine ( n = 60), amitriptyline (25 mg/day) used in combination with aerobic exercise therapy lead to a superior reduction in headache frequency, duration, intensity, and depression and anxiety scores when compared to amitriptyline used alone.

There are, however, some studies that either have shown negative or nonconclusive results with regards to the interaction between exercise and migraine and have also noted that exercise may trigger pain in some patients. The lifetime prevalence of exercise triggering a migraine in patients with migraine was 38% in one study ( n = 108) and 22% in another large prospective study ( n = 1207). Interestingly, a further study seemed to find an association between the baseline attack frequency of migraine and the risk of developing an attack post exercise in patients with migraine ( n = 14). It also appears that migraine may co-exist with primary exertional headache (PEH) in 6.2% 30% of patients with migraine. Importantly, despite the possible triggering of attacks in patients with migraine doing exercise initially, this effect appears to lessen with time, and one author has noted that a ‘tolerance’ to the possible pain-inducing effects of moderate exercise may develop with time.

When looking at large scale reviews, systematic reviews, and metanalyses, it appears, however, that the overall effect of exercise in migraine is favorable. In a recent review paper looking at exercise and migraine, the conclusion was given that “ although exercise can trigger migraines, regular exercise may have a prophylactic effect on migraine frequency .” Furthermore, a systematic review concluded that there was strong evidence for the absence of adverse events following exercise and strong evidence that exercise can lower headache intensity, frequency, and duration of pain in those with tension type headache. A further systematic review and meta- analysis concluded that despite having a “low” level of evidence, aerobic exercise had a significant effect on reducing migraine intensity and frequency after removal of studies with so called “high risk of bias domains.” The grade of evidence given for exercise as a therapy in migraine was B C in an earlier critical review. But, in a recent 2018 review, aerobic exercise of moderate intensity for > 40 min, 3 times a week for use in migraine prevention was recommended, as it can lead to considerable improvement in migraine frequency, severity, duration, and improvement of quality of life. Finally, a systematic review and meta-analysis from 2019 found moderate quality evidence that aerobic exercise treatment in migraine leads to on average a reduction of 0.6 ± 0.3 migraine days/month.

How exercise improves headache is poorly understood, but postulated mechanisms include decreased peripheral sensitization and activation of the descending inhibitory pain pathway; increased production of beta-endorphins, which is known to be lower in those with migraine, specifically chronic migraine; changes in blood nitric oxide (NO) levels; modulation of the endocannabinoid system, known to be dysfunctional in migraine patients; and finally through up-regulation brain-derived neurotrophic factor (BDNF) after exercise. Likewise, the mechanisms for why pain may be triggered by exercise in migraine are also poorly understood, but might include the release of hypocretin from the hypothalamus post exercise, the increase in lactate with intense exercise or the possible release of CGRP with exercise.

Yoga and headache

Yoga has been shown to be an effective therapy for headache and migraine. In a randomized control trial in adults ( n = 72) assigned to 1 h of yoga 5 days a week versus self-care, a significant reduction in migraine frequency was noted in those practicing yoga (10.22 ± 2.59 to 4.56 ± 1.79 days per month) vs controls (9.82 ± 2.31 to 10.18 ± 2.14, P < 0.001). There is also a significant reduction in pain intensity and duration. In a trial in adults with migraine ( n = 60) looking at the combination of a bio-purification process known as “Ayurvedic” followed by yoga for 3 months compared to standard symptomatic treatment, it was noted that both quality of life and headache intensity significantly improved. In a further randomized control trial in adults with chronic tension type headache (CTTH) ( n = 23), temporalis EMG muscle recordings were analyzed 4 weeks after a 7 week therapeutic course of yoga versus symptomatic acute treatment with NSAIDs and versus botulinum toxin treatment (50–60 unites divided into 10–12 equal doses over “tender points”). Pain score were noted to be reduced using the VAS in those using NSAID acute treatment and yoga, but not in the Botox group. Temporalis EMG recordings, that were noted to be higher at rest in those with CTTH versus controls, showed a significant reduction after the yoga intervention. In a trial of female migraineurs ( n = 32), a 12-week program of yoga (3 sessions a week for 75 min each) was compared to medication only. Headache frequency, severity, and functional impact decreased in the yoga group as compared to the medication only group. There is limited data for the use of yoga in pediatrics, but a pilot program of 8 weekly 75-min yoga sessions was performed and importantly concluded feasibility and acceptability of such a paradigm in this age group. Finally, in a meta-analysis on yoga for pain, a moderate effect was noted for reducing chronic pain intensity.

Other trials have looked to assess both efficacy and the possible mechanism for why yoga might be effective in migraine and headache. In adult patients with and without migraine aura ( n = 60), yoga therapy (5 days a week for 6 weeks) versus conventional care not only reduced headache frequency, intensity, and disability (using the HIT-6 score), but it also led to enhanced vagal tone and decreased sympathetic tone. Regulation of autonomic tone is postulated to be one of the reasons by yoga is effective in patients with migraine. In a further trial, of adults with migraine ( n = 42), practicing yoga was associated with a significant decrease in plasma vascular cell adhesion molecule (VCAM), which is possibly a marker of improved cerebrovascular health.

In conclusion, both aerobic exercise and yoga seem be well tolerated for headache, specifically migraine. Data supports that both may offer preventive benefit with regards to headache frequency, intensity, and duration and may improve measures of quality of life. Data also supports that there is limited evidence to suggest that headache will worsen with the commencement of a regular exercise program and if it does, this worsening may habituate over time.