RightRep EN · PT Open the app
BLOG

Can You Train With Migraines? The Storm-Front Barbell Audit: Where 40 Minutes of Exercise Three Times a Week Tied Topiramate Dead at −0.93 vs −0.97 Attacks Per Month at p=0.95, the 21-Trial Network Meta Ranks Strength Training First at MD −3.55 Monthly Attacks Ahead of High-Intensity Aerobic at −3.13 and Topiramate Itself at −0.98, the 1,207-Patient Trigger Ladder Puts Stress at 79.7% and Exercise Down at 22.1%, Falling Stress Prices the Next Attack at OR 1.5–1.9 Over the Next 6–18 Hours, the Dose Curve Peaks at 900–950 Cumulative Minutes, and Brazil's Nationwide File Carries 1.43× the Migraine Odds in Those Who Don't Exercise — the Program That Stabilizes the Top of the Trigger Ladder Is the Prophylaxis

By RightRep AI Assistant · · 21 min read

Migraine is the condition this library has already opened from the treatment side — does exercise help migraines is the general file, and this door answers the diagnosed lifter's version: what happens to a barbell program when the brain runs a storm-front that arrives on its own schedule. The scale is among the largest this library carries: 14.0% of the world (95% CI 12.9–15.2) pooled across the 357-publication Stovner update [1], the Steiner–Stovner policy review holding the best estimate at 14–15% and 4.9% of global population ill health in years lived with disability — a burden the authors call probably under-quantified [2]. The default scripts fail in opposite directions. The scared script reads the diagnosis as a fragility verdict — every hard set becomes a loaded question ("will this summon the storm?"), the trigger list gets treated as a cage, training is quit "until the neurologist clears it," and the deconditioning spiral feeds the exact stress-sleep-rhythm machinery the trigger ladder is stacked on. The reckless script trains straight through everything — through the prodrome, through skipped meals, through broken sleep, through the aura — because "a headache never stopped a set," and never learns that the first-severe-headache-under-load is a medical event, not a PR attempt. The reframe that carries the whole file is one line from the trigger ladder itself: the top of the list is not the barbell — stress sits at 79.7%, missed meals at 57.3%, sleep disturbance at 49.8%, sleeping late at 32.0%, and exercise all the way down at 22.1% [8] — which means a well-built program is not the thing the diagnosed lifter must survive; it is the delivery vehicle for the three stabilizers the top of that ladder demands. Which puts the fences in order. First: the neurologist names it. This file assumes migraine diagnosed by a neurology or headache team, secondary causes excluded — the diagnosis is a clinical pattern, not a self-diagnosis from a symptom checklist. Second: a new headache under the bar is not a migraine until medicine says so. The first-ever severe exertional headache, or one that is explosively abrupt, fundamentally different, or accompanied by neurological symptoms, is a same-day medical evaluation — primary exercise headache exists and is usually benign [10], but it is a diagnosis of exclusion, and the exclusion is the neurologist's job, not the lifter's. Third: the drug shelf is the prescriber's. No session, set or "natural protocol" edits a topiramate titration, a beta-blocker (which quietly blunts the heart-rate numbers a program trains by), a triptan schedule or a CGRP antibody — and the timing questions around training and acute medication belong to the prescriber too. With the fences standing, the operational file is unusually barbell-shaped for a brain condition: the permission question has a three-arm randomized trial that ran exercise against an actual drug [3], the modality question has a 21-trial network meta-analysis that ranks strength training first [4], and the dose question has a U-shaped curve with a named prescription [6] — while the trigger ledger turns program design itself into prophylaxis.

The short answer: yes — prophylaxis-grade, with a trigger ledger to manage

Strip the question down and it splits into three checkable parts. One: does regular training reduce migraine frequency? Yes — the pooled aerobic estimate is 0.6 ± 0.3 fewer migraine days per month of moderate-quality evidence [5], the network meta puts the best-studied regimens between −2.18 and −3.55 monthly attacks [4], and the dose-response meta peaks at SMD −1.55 on frequency at ~950 cumulative minutes [6]. Two: does it match medication? In the only three-arm head-to-head, yes — a dead tie at p=0.95, exercise −0.93 attacks/month against topiramate −0.97, with the drug arm titrated to each patient's highest tolerable dose and the exercise arm exercising 40 minutes three times a week [3]. Three: can a single session trigger an attack? Yes, for a minority — 22.1% of 1,207 patients named exercise among their triggers [8] — and that number is the file's real subject: not "is training allowed" but "how do you build sessions and weeks that don't volunteer for that 22.1%." The short answer stands on the first two questions and spends the rest of this file on the third, because the third is the one the diagnosed lifter actually lives.

The reframe: the trigger ladder is a training-program audit

Kelman's 1,207-patient trigger file is usually quoted as a list of things to avoid. Read as a lifter, it is an audit of what a training program is: stress 79.7%, hormones 65.1% (in women), not eating 57.3%, weather 53.2%, sleep disturbance 49.8%, perfume or odour 43.7%, neck pain 38.4%, light 38.1%, alcohol 37.8%, smoke 35.7%, sleeping late 32.0%, heat 30.3%, food 26.9%, exercise 22.1%, sexual activity 5.2% — with 75.9% of patients reporting triggers at all [8]. Now score a badly-run bulk against that ladder: skipped meals while fasting through morning sessions (57.3%), sleep sacrificed to early alarms and late protein feeds (49.8%), weekend sleep-in destroying the weekday rhythm (32.0%), a neck that hurts because the program never trained it and the desk job did (38.4%), alcohol as the weekend's reward (37.8%), and the stress of a life run around an unmanaged program (79.7%). The top four trainable entries of the ladder are all things a sane program stabilizes — regular meals, defended sleep, a consistent wake time, a neck that is trained rather than neglected — which is why Brazil's nationwide file prices migraine at 1.43× higher odds in people who do not exercise, independent of body mass index [14]. The barbell is not competing with the trigger list; done right, it is the compliance mechanism for the top of it.

The let-down file sharpens this into a scheduling rule. In the Lipton cohort's 2,011 diary entries, it was not the level of stress that predicted attacks — level was flat — but the decline: a fall in perceived stress from one evening to the next raised the odds of migraine onset over the following 6, 12 and 18 hours at OR 1.5–1.9, all p<0.05 [9]. This is the Friday-after-the-delivery-deadline attack, the first-day-of-vacation attack, the day-after-the-meet attack — the storm that arrives precisely when the pressure lifts. The honest footnote: 17 completers carried the whole file, so it is directional, not gospel [9]. But the operational takeaway survives the sample size: hard weeks should taper into rest rather than fall off a cliff, and the days after a peak block — not the peak block itself — are when the attack calendar should expect traffic.

The mechanism file: what honestly moves

The Brazilian mechanism RCT gives the file its cleanest dose-response artifact. Fifty episodic-migraine patients completed a 12-week aerobic program at Universidade Federal de São Paulo; in the migraine exercise group, migraine days fell (p<0.01), attacks fell (p<0.05), and abortive medication use fell (p<0.05), while cardiorespiratory fitness rose — and the reduction in abortive medication correlated with the fitness gain at r=−0.81, p<0.001 [13]. That correlation is the finding: the patients whose engines actually improved were the ones who reached for the rescue med least. Plasma anandamide shifted with training, and the authors read the endocannabinoid column as partly linked to the clinical benefit [13] — a mechanism candidate, not a mechanism verdict, and this file carries it at exactly that weight. Mood moved with it — anxiety, depression, anger and fatigue scores all improved (p<0.05) [13] — which matters less as neurology than as adherence: the program that improves the mood column is the program that survives the calendar.

The barbell file: the crown RCT and the network-meta ladder

The crown is the Swedish three-arm trial, and its design is the point. Ninety-one patients were randomized to exercise — 40 minutes, three times a week — or relaxation via recorded program, or daily topiramate titrated to each individual's highest tolerable dose, maximum 200 mg/day, for three months [3]. The primary endpoint, mean reduction in migraine attacks during the final month versus baseline: exercise −0.93 (95% CI 0.31–1.54), relaxation −0.83 (0.22–1.45), topiramate −0.97 (0.36–1.58) — no significant difference between groups, p=0.95 [3]. Read that sentence the way a formulator would: an intervention delivered by three 40-minute sessions a week matched a dose-titrated preventive drug, dead even, in the trial designed to ask exactly that question. The trial's own conclusion is the permission sentence this door exists to carry — exercise may be an option for patients who do not benefit from or do not want daily medication [3].

Then the modality ladder, and it is the most barbell-friendly finding in this condition's file. The 2022 network meta-analysis pooled 21 clinical trials, 1,195 migraine patients, mean age 35, female-to-male ratio 6.7, and ranked the interventions on monthly migraine frequency [4]: strength training first at MD −3.55 (95% CI −6.15 to −0.95), high-intensity aerobic exercise second at −3.13 (−5.28 to −0.97), moderate-intensity aerobic third at −2.18 (−3.25 to −1.11), topiramate fourth at −0.98 (−4.16 to 2.20), placebo at null — and amitriptyline worse than placebo at +3.82 (−1.03 to 8.68) [4]. The confidence intervals overlap and this is network, not head-to-head, evidence — but the direction is consistent: in the only condition file where the drug was run against the barbell's rivals directly, the barbell's shelf did not embarrass itself. The intensity question got its own randomized file: in 37 episodic migraineurs training twice a week for 12 weeks, high-intensity intervals cut migraine days more than moderate continuous training (SMD 1.05 vs 0.43), and only the interval arm reduced pulse-wave reflection, with central systolic pressure falling 118→110 mmHg [7] — the intensity that the zone 2 crowd fears is, at least in this trial, the one that bought the bigger migraine column.

The dose file: the U-curve and the weekly template

The 2026 dose-response meta-analysis turned fifteen studies into a prescription. Total aerobic minutes plotted against pain intensity and attack frequency both traced statistically significant U-shapes (p<0.001 for both splines): a minimum of 200 cumulative minutes before pain intensity moves meaningfully and 300 minutes before frequency does, a maximum effect at ~900 minutes for intensity (SMD −2.4) and ~950 minutes for frequency (SMD −1.55) — and beyond the peak, more minutes buy nothing and may buy less [6]. Translated to a calendar, the authors' preliminary recommendation reads like a beginner template: three 30-minute sessions a week for 10–11 weeks [6]. That is the aerobic column. The strength column runs on the network meta's shelf [4] and the crown trial's proven dosing — 40-minute sessions, three non-consecutive days, at least 12 weeks before judging the migraine ledger [3] — which lands on ordinary, sustainable training frequency and set budgets anyway. The U-curve's descending limb is the file's quiet warning for exactly this readership: the migraine brain does not reward the volume-maximal bulk, and "more" past the peak is not conservative programming, it is a dose overrun.

The honest counterfile: what training does not do

Three nulls keep this file honest. The pooled aerobic effect is modest: six studies, −0.6 ± 0.3 migraine days per month, moderate quality, with duration (−20–27%) and intensity (−20–54%) improvements too heterogeneous to pool [5] — a real column, not a cure. The neck-fix is not a file: the São Paulo trial randomized migraine patients to eight weeks of neck-specific strengthening against sham ultrasound plus home stretching, and the primary outcomes — headache frequency and intensity — did not move; craniocervical endurance electromyography improved, the storms did not [11]. "Train your neck and the migraines stop" is a pitch, not a result; neck training belongs in the program because the trigger ladder carries neck pain at 38.4% [8], not because it is prophylaxis. The weight-loss shortcut is not either: in the 110-woman behavioral weight-loss RCT, the weight arm lost 3.8 kg against a 0.9 kg gain in the education arm (P<0.001) — and migraine days fell in both arms with no difference between them (−3.0 vs −4.0, P=0.185), sustained at follow-up [12]. Whatever moved those headaches, it was not the kilograms; the attention, the diary and the structure did the work. The counterfile's combined message: training is a real preventive column with real fences around its claims — and the claims it cannot make are already made by things that work better (acute meds, prescribed preventives).

The trigger-management file: sessions that don't volunteer for the 22.1%

This is the section that separates this door from the treatment-side file. The operational question is not whether to train but how to build sessions and weeks that don't stand in the trigger line-up, and every rule below prices against the ladder [8]:

  1. Warm up like the diagnosis depends on it — because the abrupt-start hypothesis does. The sudden-onset hard set is the classic shape of exertional headache [10]; a graded warm-up before lifting — plus the cardio primer if the first working set is heavy — costs ten minutes and removes the single most migraine-shaped loading pattern in the gym. In hot, bright boxes, heat (30.3%) and light (38.1%) are on the ladder too [8]: shade the platform, hydrate per the dehydration file.
  2. Never train fasted on a migraine program. "Not eating" is the ladder's third entry at 57.3% [8] — above exercise itself. The pre-session meal is migraine medicine in this file; schedule it like a set. The fasted training debate is a fat-loss conversation this diagnosis opts out of.
  3. Treat caffeine as a timed drug, not a vibe. The pre-workout at irregular hours is a withdrawal lottery on a brain that notices schedule drift; fix the caffeine clock and keep it identical on rest days.
  4. Defend sleep like a lift. Sleep disturbance (49.8%) and sleeping late (32.0%) flank the ladder's top [8]; the sleep requirement file and the migraine calendar want the same thing — a fixed wake time, including weekends.
  5. Taper out of hard blocks, don't cliff them. The let-down file prices the post-stress window at OR 1.5–1.9 for 6–18 hours [9]; after a peak week or a competition, descend on purpose — a lighter week, not a zero week.
  6. During the prodrome, downgrade, don't push. The aura is a stop-work order for the session in progress; a pre-headache phase without aura earns at most a light technical session, and the full attack means the sick rules logic applies: the session is rescheduled, the program is not abandoned.

And the fence under all six: the first severe headache under load, the worst-ever, the explosively abrupt, the one with neurological accompaniment — same-day medicine, not a second attempt at the set [10]. Primary exercise headache is real, usually benign, usually self-limiting — and it is diagnosed after the dangerous mimics are excluded, never before [10].

The fences, in order

  1. The diagnosis fence. Migraine is a neurologist's pattern-match against secondary causes; this file programs for the diagnosed, not the self-diagnosed. New headache patterns after 50, abrupt-onset ("thunderclap") headaches, headaches with fever/stiff neck/focal neurology/eye pain are medicine-first events.
  2. The exertional-headache fence. New or different headache under load gets evaluated before it gets programmed around [10]; returning to the bar after an exertional headache is a medical timing call, not a toughness call.
  3. The drug fence. Beta-blockers reshape the heart-rate targets a program trains by; topiramate's cognitive and fatigue sides distort perceived exertion; triptans and CGRP antibodies are the prescriber's to time against sessions. The training log documents what happened; the prescriber adjusts the shelf.
  4. The hormonal column. Menstrual migraine rides the ladder's 65.1% hormones entry [8]; program around the vulnerable window rather than through it, with the prescriber in the loop for perimenopause and estrogen-containing decisions — the same coordination the high blood pressure door teaches for its shelf.
  5. The red-flag list outranks every training decision. Worst-ever headache, headache with confusion or neurological signs, new daily persistent headache, headache after head trauma — same-day medicine, program paused without apology.

The Brazil anchors

The pt-BR lane's reader holds a specific position in this condition, and the Brazilian file is unusually complete. The nationwide population-based study — 3,848 telephone interviews across all 27 states, adults 18–79 — puts one-year migraine prevalence at 15.2%, 2.2× higher in women, and 1.43× higher in those who do not exercise, independent of BMI [14]; migraine also ran 1.59× higher under five minimum wages and 1.5× higher beyond eleven years of education, a socioeconomic gradient the authors read as real and unexplained [14]. The fresh PNS2019 lifestyle file scales it nationally: among 88,531 Brazilian adults, headache-associated activity loss ran at 5.3% — the fifth most prevalent disease-related disability in the country — and every additional healthy-lifestyle pillar carried OR 0.86 (95% CI 0.82–0.91) against it [15]. And the mechanism RCT on this file's shelf is Brazilian too — the Unifesp anandamide trial with its r=−0.81 fitness-to-medication correlation [13]. For the Brazilian lifter, the file's numbers are not imported: 15.2% of the adults in the gym's postcode run this storm-front, and the ones not training run it 43% harder [14].

So what do you actually do? The practical file

Six rules, every one sourced from the files above:

  1. Get the diagnosis and the exclusion first, then program. The neurologist names the pattern and rules out the mimics; the first exertional headache gets the same evaluation before it gets a workaround [10].
  2. Run the strength column as the prophylaxis backbone: 3 non-consecutive days, ~40-minute sessions, at least 12 weeks before judging the ledger [3] — the modality the network meta ranked first [4] — on ordinary progressive overload, sustainable set budgets, and loads climbing honestly week over week.
  3. Add the aerobic column at the U-curve's prescription: 3 × 30 minutes a week for 10–11 weeks, ~900–950 cumulative minutes total, then hold [6]; if the engine allows, bias it interval-shaped [7]. That dose is deliberately ordinary — the migraine brain does not reward volume heroics.
  4. Audit the week against the trigger ladder, not the mirror: meals never skipped, sleep defended at a fixed wake time, caffeine on a clock, alcohol budgeted, neck trained as a column not a cure [8][11].
  5. Taper out of stress peaks and schedule around the let-down window [9]; after the peak week, descend for a week — the storm-front statistically prefers the day after the deadline.
  6. Keep the fences standing: the drug shelf is the prescriber's, the aura stops the session, the new-or-worst headache is same-day medicine [10] — the program runs with the treatment plan, never instead of it.

How RightRep fits into this

The migraine program runs on one currency this app counts better than any notebook — the dated, dosed, camera-verified session:

  • The 12-week verdict needs dated evidence, not memory. The crown trial's endpoint was the final month against baseline [3] — a session-by-session log is how a 12-week dose actually gets delivered and how the next neurology visit reads the trend.
  • The trigger audit needs a training-side diary to correlate against the attack calendar. Meals, sleep and caffeine are on the lifter; the dated session log is the column that proves the program's rhythm held — or didn't — in the weeks the calendar went quiet or loud.
  • The fitness-to-medication correlation needs both columns counted. The Unifesp trial's r=−0.81 linked rising fitness to falling rescue-med use [13]; reps counted and load climbed is the fitness column's ledger.
  • Progressive overload honesty at U-curve doses. The file says build the strength column [4] — but the dose ceiling is real [6]; the camera keeps the climb honest without the volume drift the descending limb penalizes.
  • The neurology-visit artifact. "What have you been doing since the last visit?" — the question every headache consult asks is answered by the camera-counted log you hand over [2].

The bottom line

  • The crown: exercise 40 min × 3/week tied dose-titrated topiramate dead on attack reduction — −0.93 vs −0.97 attacks/month, p=0.95, across 91 randomized patients [3].
  • The modality ladder: strength training first at MD −3.55 monthly attacks, high-intensity aerobic −3.13, moderate aerobic −2.18, topiramate −0.98, amitriptyline worse than placebo at +3.82 — 21 trials, 1,195 patients [4].
  • The dose: U-shaped — ~200–300 cumulative minutes minimum, peak effect ~900–950 minutes (intensity SMD −2.4, frequency −1.55), prescription 3 × 30 min/week over 10–11 weeks [6]; intervals beat continuous for migraine days (SMD 1.05 vs 0.43) with central systolic pressure 118→110 mmHg [7].
  • The trigger ladder: stress 79.7%, not eating 57.3%, sleep disturbance 49.8%, neck pain 38.4%, sleeping late 32.0% — exercise down at 22.1% [8]; the let-down: falling stress prices the next 6–18 hours at OR 1.5–1.9 [9].
  • The honest counterfile: pooled aerobic −0.6 ± 0.3 days/month [5]; neck-specific strengthening a clean null on frequency and intensity [11]; weight loss bought −3.8 kg and no migraine advantage over education (−3.0 vs −4.0 days, P=0.185) [12].
  • The mechanism: migraine days p<0.01, attacks p<0.05, rescue meds p<0.05 down, fitness up — rescue-med reduction correlating with fitness at r=−0.81 [13].
  • The scale: 14.0% of the world (12.9–15.2), 4.9% of global YLDs, under-quantified [1][2]; Brazil: 15.2% nationwide, 1.43× in non-exercisers independent of BMI [14], OR 0.86 per lifestyle pillar across 88,531 adults [15].

Train like the trigger ladder is a program audit — the top of the list is stress, meals and sleep, the barbell is the delivery vehicle for all three, and the fences are the neurologist's: at rightrep.app.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Stovner LJ, Hagen K, Linde M, Steiner TJ. The global prevalence of headache: an update, with analysis of the influences of methodological factors on prevalence estimates. The Journal of Headache and Pain, 2022;23(1):34. pubmed.ncbi.nlm.nih.gov
  2. Steiner TJ, Stovner LJ. Global epidemiology of migraine and its implications for public health and health policy. Nature Reviews Neurology, 2023;19(2):109–117. pubmed.ncbi.nlm.nih.gov
  3. Varkey E, Cider Å, Carlsson J, Linde M. Exercise as migraine prophylaxis: a randomized study using relaxation and topiramate as controls. Cephalalgia, 2011;31(14):1428–1438. pubmed.ncbi.nlm.nih.gov
  4. Woldeamanuel YW, Oliveira ABD. What is the efficacy of aerobic exercise versus strength training in the treatment of migraine? A systematic review and network meta-analysis of clinical trials. The Journal of Headache and Pain, 2022;23(1):134. pubmed.ncbi.nlm.nih.gov
  5. Lemmens J, De Pauw J, Van Soom T, Michiels S, Versijpt J, van Breda E, Castien R, De Hertogh W. The effect of aerobic exercise on the number of migraine days, duration and pain intensity in migraine: a systematic literature review and meta-analysis. The Journal of Headache and Pain, 2019;20(1):16. pubmed.ncbi.nlm.nih.gov
  6. Ogrezeanu DC, Núñez-Cortés R, Salazar-Méndez J, Cuyul-Vásquez I, López-Bueno R, Ferrer-Sargues FJ, Andersen LL, Calatayud J, Suso-Martí L. How much aerobic exercise is needed to reduce migraine? A dose-response meta-analysis of pain intensity and frequency. Headache, 2026;66(1):40–52. pubmed.ncbi.nlm.nih.gov
  7. Hanssen H, Minghetti A, Magon S, Rossmeissl A, Papadopoulou A, Klenk C, Schmidt-Trucksäss A, Faude O, Zahner L, Sprenger T, Donath L. Superior effects of high-intensity interval training vs. moderate continuous training on arterial stiffness in episodic migraine: a randomized controlled trial. Frontiers in Physiology, 2017;8:1086. pubmed.ncbi.nlm.nih.gov
  8. Kelman L. The triggers or precipitants of the acute migraine attack. Cephalalgia, 2007;27(5):394–402. pubmed.ncbi.nlm.nih.gov
  9. Lipton RB, Buse DC, Hall CB, Tennen H, Defreitas TA, Borkowski TM, Grosberg BM, Haut SR. Reduction in perceived stress as a migraine trigger: testing the "let-down headache" hypothesis. Neurology, 2014;82(16):1395–1401. pubmed.ncbi.nlm.nih.gov
  10. Upadhyaya P, Nandyala A, Ailani J. Primary exercise headache. Current Neurology and Neuroscience Reports, 2020;20(5):9. pubmed.ncbi.nlm.nih.gov
  11. Benatto MT, Florencio LL, Bragatto MM, Dach F, Fernández-de-las-Peñas C, Bevilaqua-Grossi D. Neck-specific strengthening exercise compared with placebo sham ultrasound in patients with migraine: a randomized controlled trial. BMC Neurology, 2022;22(1):131. pubmed.ncbi.nlm.nih.gov
  12. Bond DS, Thomas JG, Lipton RB, Roth J, Pavlovic JM, Rathier L, O'Leary KC, Evans EW, Wing RR. Behavioral weight loss intervention for migraine: a randomized controlled trial. Obesity (Silver Spring), 2018;26(1):81–87. pubmed.ncbi.nlm.nih.gov
  13. Oliveira AB, Ribeiro RT, Mello MT, Tufik S, Peres MFP. Anandamide is related to clinical and cardiorespiratory benefits of aerobic exercise training in migraine patients. Cannabis and Cannabinoid Research, 2019;4(4):275–284. pubmed.ncbi.nlm.nih.gov
  14. Queiroz LP, Peres MF, Piovesan EJ, Kowacs F, Ciciarelli MC, Souza JA, Zukerman E. A nationwide population-based study of migraine in Brazil. Cephalalgia, 2009;29(6):642–649. pubmed.ncbi.nlm.nih.gov
  15. de Oliveira AB, Abib S, Dantas MIO, Souza MNP, Hech G, Papassidero P, Blood MRY; Brazilian Headache Society's Lifestyle Medicine Committee. The 6 pillars of lifestyle medicine and headache disability in Brazil: a cross-sectional, population-based study. The Journal of Headache and Pain, 2025;26(1):214. pubmed.ncbi.nlm.nih.gov