Reviewed for medical accuracy by Dr Evelyne Jonniaux, Medical Advisor
Last updated 17 September 2026
Migraine with no red flags: MRI almost never changes what happens next
In a patient whose headaches meet migraine criteria and whose neurological examination is normal, the probability of finding a significant lesion is 0.9%, with an interval from 0.5 to 1.4 (Sempere, Cephalalgia, 2005, across 1,876 consecutive patients). In an asymptomatic population, 37 people have to be scanned to find one incidental abnormality (Morris, BMJ, 2009). The American guideline, graded A, concludes that imaging these patients is not necessary. These figures describe populations, not one patient. Noria Health arranges a full neurological work-up near Brussels within 24 to 48 hours.
What imaging finds, and what it finds by chance
The reference meta-analysis covers 41 studies and 15,760 headache patients with a normal neurological examination.
| What is found | Frequency |
|---|---|
| Unexpected abnormalities, all modalities | 17.5% (13.1 to 22.3) |
| Same, MRI only | 26.6% (15.5 to 39.4) |
| Neoplastic findings | 1.4% |
| Glioma | 0.2% |
| Meningioma | 0.1% |
| Aneurysm | 1.8% |
The 17.5% frightens people, which is why it needs its denominator. In a strictly asymptomatic population, incidental neoplastic findings run at 0.70%, with an interval from 0.47 to 0.98, and non-neoplastic ones at 2.0%. High-resolution MRI finds 4.3% against 1.7% for standard sequences. In other words, most of what is found in a migraine patient would be found in anyone of the same age.
The American Headache Society guideline is explicit: clinically significant abnormalities in migraine patients without atypical features are no more frequent than in the general population. The remaining indications are grade C, and the authors write that most rest on consensus with little or no support in the literature.
How many people have migraine, and how many are diagnosed
| What is measured | The figure | Source |
|---|---|---|
| One-year prevalence, women | 17.1% | Lipton, Neurology, 2007, n = 162,576 |
| One-year prevalence, men | 5.6% | Lipton, 2007 |
| People with migraine who have had a medical diagnosis | 56.2% | Diamond, Headache, 2007 |
| People with migraine on preventive treatment | 12.4% | Diamond, 2007 |
| Severe impairment or need for bed rest | 53.7% | Lipton, 2007 |
Almost four people in ten with migraine have never been diagnosed. The often quoted claim that only one in three is diagnosed is wrong: the primary source gives 56.2%.
The CGRP drugs, and the figure brochures leave out
These treatments work. The question is by how much, and against what.
| Trial | Reduction on treatment | Reduction on placebo | Drug effect alone |
|---|---|---|---|
| STRIVE, erenumab, episodic migraine | -3.7 days | -1.8 days | 1.9 days |
| HALO-CM, fremanezumab, chronic migraine | -4.6 days | -2.5 days | 2.1 days |
| EVOLVE-1, galcanezumab | -4.7 days | -2.8 days | 1.9 days |
| PROMISE-2, eptinezumab | -8.2 days | -5.6 days | 2.6 days |
Placebo accounts for 49% to 68% of the total reduction depending on the trial. The drug’s own effect is one to three fewer migraine days a month. That can change a life when you have fifteen, and it is modest when you have four.
The claim that CGRP drugs halve migraine days is misleading. The 50% refers to the proportion of responders, not the mean reduction. And in the same STRIVE trial, 26.6% of patients on placebo were also 50% responders.
On acute treatment, the reference meta-analysis across 53 trials and 24,089 patients gives, for sumatriptan 100 mg, 59% headache relief at two hours and 29% complete freedom from pain. The often quoted 70% to 80% is an overstatement.
Medication-overuse headache, and an admission from the classification
It affects 1% to 2% of the general population. The thresholds are ten days a month for triptans and ergotamine, and fifteen days for non-aspirin anti-inflammatories.
The international classification itself states that these day counts rest on expert opinion rather than formal evidence. That is unusual, and it deserves saying to a patient being told they are overusing their treatment.
After withdrawal, 73.1% of patients are free of overuse at six months with electronic monitoring, against 64.1% with paper follow-up, odds ratio 1.45 and a p of 0.046. The relapse rate at one year is 20.5%.
Red flags, and what is not known about them
The reference list has fifteen items: systemic symptoms including fever, history of cancer, neurological deficit, sudden thunderclap onset, onset after 65, change in pattern or a new headache, positional character, precipitation by coughing or exertion, papilloedema, progressive worsening, pregnancy or the puerperium, painful eye with autonomic signs, post-traumatic headache, immune compromise, and painkiller overuse or a new drug.
Its authors write that the absence of prospective epidemiological studies on these flags leaves many questions unanswered. There is therefore no validated sensitivity or predictive value for most of them. Any claim that a given flag has a sensitivity of some percentage is unsourced.
The only clinical element whose discriminating power has been quantified prospectively is the neurological examination itself. It was the only variable associated with the probability of an intracranial abnormality in Sempere’s study.
What the studies do not allow us to claim
That migraine is the second leading global cause of years lived with disability is not found as such in the source. What is sourced is that migraine accounts for 16.3% of neurological disability-adjusted life years, second among neurological disorders behind stroke.
The overuse thresholds for opioids, combination analgesics and paracetamol could not be verified line by line in the official text, and we do not give them.
Finally, there is no sex breakdown of migraine prevalence for France in the available sources. We do not transpose American ratios.
In practice, if this concerns you
Three questions structure the discussion. Do your headaches meet migraine criteria and is your neurological examination normal, since that is what makes imaging unnecessary. How many days a month do you take an acute treatment, since the overuse threshold is lower than people think. And if a preventive treatment is proposed, how many migraine days a month are you starting from, since the drug’s own effect is one to three days.
Noria Health arranges a full neurological work-up near Brussels within 24 to 48 hours: structured history against the diagnostic criteria, complete neurological examination, headache and medication diary, and imaging only where the examination or history justifies it. The report is issued within 24 hours, follow-up is structured at D14, D30 and D90.
What Noria Health does not do: order an MRI for reassurance. A thunderclap headache, a headache with fever and neck stiffness, a rapidly developing neurological deficit or a first headache after 65 are matters for the emergency services in your country, immediately, not for an appointment.
Further reading
- Neurological diagnosis, what is at stake in the delay
- Multiple sclerosis: migraine is the commonest wrong diagnosis
- Memory: 86% of complaints do not become dementia
Sources
- Lipton RB, Bigal ME, Diamond M, et al. Migraine prevalence, disease burden, and the need for preventive therapy. Neurology, 2007;68:343-349. https://pubmed.ncbi.nlm.nih.gov/17261680/
- Diamond S, Bigal ME, Silberstein S, et al. Patterns of diagnosis and acute and preventive treatment for migraine. Headache, 2007;47:355-363. https://pubmed.ncbi.nlm.nih.gov/17371352/
- Kamtchum-Tatuene J, Kalra A, Gwynn LC, et al. Diagnostic yield of brain imaging in non-acute headache. J Neurol Sci, 2020;416:116997. https://pubmed.ncbi.nlm.nih.gov/32683173/
- Sempere AP, Porta-Etessam J, Medrano V, et al. Neuroimaging in the evaluation of patients with non-acute headache. Cephalalgia, 2005;25:30-35. https://pubmed.ncbi.nlm.nih.gov/15606567/
- Morris Z, Whiteley WN, Longstreth WT, et al. Incidental findings on brain magnetic resonance imaging. BMJ, 2009;339:b3016. https://pubmed.ncbi.nlm.nih.gov/19687093/
- Vernooij MW, Ikram MA, Tanghe HL, et al. Incidental findings on brain MRI in the general population. N Engl J Med, 2007;357:1821-1828. https://pubmed.ncbi.nlm.nih.gov/17978290/
- Evans RW, Burch RC, Frishberg BM, et al. Neuroimaging for migraine, the American Headache Society evidence-based guideline. Headache, 2020;60:318-336. https://pubmed.ncbi.nlm.nih.gov/31891197/
- Loder E, Weizenbaum E, Frishberg B, Silberstein S. Choosing wisely in headache medicine. Headache, 2013;53:1651-1659. https://pubmed.ncbi.nlm.nih.gov/24118318/
- Goadsby PJ, Reuter U, Hallstrom Y, et al. A controlled trial of erenumab for episodic migraine, STRIVE. N Engl J Med, 2017;377:2123-2132. https://pubmed.ncbi.nlm.nih.gov/29171821/
- Silberstein SD, Dodick DW, Bigal ME, et al. Fremanezumab for the preventive treatment of chronic migraine, HALO-CM. N Engl J Med, 2017;377:2113-2122. https://pubmed.ncbi.nlm.nih.gov/29171818/
- Stauffer VL, Dodick DW, Zhang Q, et al. Evaluation of galcanezumab for the prevention of episodic migraine, EVOLVE-1. JAMA Neurol, 2018;75:1080-1088. https://pubmed.ncbi.nlm.nih.gov/29813147/
- Lipton RB, Goadsby PJ, Smith J, et al. Efficacy and safety of eptinezumab in chronic migraine, PROMISE-2. Neurology, 2020;94:e1365-e1377. https://pubmed.ncbi.nlm.nih.gov/32209650/
- Ferrari MD, Roon KI, Lipton RB, Goadsby PJ. Oral triptans in acute migraine treatment, a meta-analysis of 53 trials. The Lancet, 2001;358:1668-1675. https://pubmed.ncbi.nlm.nih.gov/11728541/
- Diener HC, Holle D, Dresler T, Gaul C. Medication-overuse headache. Lancet Neurol, 2019;18:891-902. https://pubmed.ncbi.nlm.nih.gov/31174999/
- Headache Classification Committee of the International Headache Society. The International Classification of Headache Disorders, 3rd edition. Cephalalgia, 2018;38:1-211. https://pubmed.ncbi.nlm.nih.gov/29368949/
- Tassorelli C, Jensen R, Allena M, et al. The added value of an electronic monitoring and alerting system, COMOESTAS. Cephalalgia, 2017;37:1115-1125. https://pubmed.ncbi.nlm.nih.gov/27638937/
- Do TP, Remmers A, Schytz HW, et al. Red and orange flags for secondary headaches, SNNOOP10. Neurology, 2019;92:134-144. https://pubmed.ncbi.nlm.nih.gov/30587518/
See also: all our articles on neurology.
Have your case reviewed. Noria Health arranges a specialist consultation near Brussels and confirms feasibility within 24 to 48 hours. Open a case.