A woman in her late thirties sat down in my office holding a printed MRI report, and she had already read the words out loud to herself so many times that she recited them without looking: multiple scattered T2/FLAIR hyperintensities in the subcortical and deep white matter. Her primary care doctor had ordered the scan for worsening migraines, then told her it “looked like it could be early MS” and referred her to me. She had spent eleven days assuming she was going to lose the ability to walk.
She did not have multiple sclerosis. She had migraine with aura, and the spots on her scan are among the most common incidental findings in all of neurology. Here is the short answer before the details: white matter lesions are found more often in people with migraine than in people without it, they are usually small and clinically silent, and in the great majority of cases they neither predict multiple sclerosis nor mean your brain is being damaged by your headaches. What matters far more than the presence of the spots is their pattern — and that is something a radiology report alone often cannot settle.
What those spots actually are
On T2-weighted and FLAIR sequences, certain areas of white matter show up brighter than the surrounding tissue. Radiologists describe them as hyperintensities; older neurologists still call them UBOs — unidentified bright objects — which tells you how nonspecific the finding is. Pathologically they represent small regions where the tissue holds more water than normal: mild loss of myelin, widened perivascular spaces, gliosis, or tiny areas of prior ischemia in the territory of a single small perforating vessel.
They are common. By the seventh decade of life the majority of people have at least a few, migraine or not. The strongest drivers across the whole population are age and vascular risk — hypertension, smoking, diabetes, sleep apnea. This is why, when I see a scan like the one my patient brought me, my first questions are not about her headaches. They are about her blood pressure, her cigarettes, and whether she snores.
Migraine does add its own independent signal. The best data come from the Dutch CAMERA population studies, which imaged people with migraine and matched controls and then re-scanned them roughly nine years later. Migraine — particularly migraine with aura, and particularly in women with frequent attacks — was associated with more deep white matter hyperintensities than in controls, and with a modest increase in new lesions over that decade. That sounds alarming until you read the second half of the finding: the increase in lesion volume was small, it did not correlate with attack frequency during the follow-up period, and the participants with more lesions did not perform worse on cognitive testing. A statistical difference between groups is not the same thing as a disease in an individual.
Migraine lesions versus MS lesions: the pattern is everything
This is where most of the anxiety I see is manufactured, because “white matter lesions” is the phrase that appears in both reports. The distribution is what separates them, and it usually separates them cleanly.
Migraine-associated lesions tend to be small — often just a few millimetres — round or punctate, scattered in the deep and subcortical white matter with a preference for the frontal lobes. They typically spare the corpus callosum, do not touch the juxtacortical U-fibers, do not sit perpendicular to the ventricles, and do not enhance with gadolinium.
Demyelinating lesions behave differently. Multiple sclerosis favours locations that carry diagnostic weight: periventricular lesions oriented at right angles to the ventricle (the classic Dawson’s fingers), juxtacortical lesions abutting the cortex, infratentorial lesions in the brainstem and cerebellum, and lesions in the spinal cord. They are frequently ovoid rather than round, often larger, sometimes involve the corpus callosum, and an active lesion will enhance after contrast.
Clinical history does the rest of the work. Multiple sclerosis announces itself with discrete neurological episodes lasting days to weeks — optic neuritis with painful vision loss in one eye, numbness climbing up from the feet, double vision, a limb that stops obeying. Migraine aura is fundamentally different in tempo: it spreads over five to twenty minutes, resolves within an hour, and is usually positive before it is negative — you see the shimmering zigzag before you notice the blind spot. When the history is genuinely ambiguous, spinal cord imaging, contrast, and occasionally spinal fluid for oligoclonal bands settle it. In my practice, a request for those tests is not a sign I think you have MS; it is how I earn the right to tell you that you don’t.
The one pattern that does change my thinking
There is a rare inherited small-vessel disease called CADASIL, caused by mutations in the NOTCH3 gene, and it deserves mention because migraine with aura is often its first symptom — appearing decades before anything else. It should cross your doctor’s mind when the white matter changes are extensive and confluent for the patient’s age, and especially when they involve the anterior temporal poles and the external capsule, which is an unusual signature. The other clue is the family tree: relatives with migraine plus strokes in their forties or fifties, or early cognitive decline, or a parent labelled with “early dementia” that nobody could explain.
CADASIL is genuinely uncommon, and I want to be careful not to trade one anxiety for another. Most people reading this with a handful of scattered spots do not have it. But it is the reason I ask about family history in every migraine patient whose scan shows more lesions than their age and vascular profile can account for.
What I actually do about the spots
Almost never anything directly. There is no treatment that erases white matter hyperintensities, and no evidence that suppressing migraine attacks reverses the ones already there. What I do instead is take the finding as an invitation to be more serious about the things that are modifiable.
I check blood pressure properly, more than once. I push hard on smoking, because it is the single largest reversible contributor to small-vessel white matter change and it also makes migraine harder to treat. I screen for obstructive sleep apnea, particularly in patients whose headaches are worst on waking. If attacks are frequent, I treat them properly with a preventive rather than letting the person absorb them with escalating painkillers — not because I can promise it protects the white matter, but because frequent untreated migraine is worth treating on its own terms. And in women with migraine with aura I revisit contraception, since combined estrogen-containing pills carry a stroke risk in that group that the spots on the scan should remind us to take seriously.
Repeat imaging is the question I get most often. My usual answer is no. If the clinical picture is straightforward migraine and the lesion pattern is typical, rescanning a year later tends to generate anxiety rather than information. I repeat the scan when the neurological examination changes, when new symptoms appear that migraine does not explain, or when the original pattern was atypical enough that I wanted a baseline in the first place.
What I told my patient — after reviewing the actual images rather than the report — was that her scan looked exactly like the scan of a woman in her thirties with a long history of migraine with aura, and that nothing on it required treatment or follow-up. She cried, which happens more often in my office than people imagine. The lesions had been there, silently, for years. The suffering started when a sentence got written about them without a conversation attached.
If your MRI report has left you searching for answers at midnight, the more useful question is not “what are these spots” but “does the whole picture hang together” — pattern, history, examination, and risk factors read as one thing rather than four. That is what a headache-focused review is for. I’ve written more about why a stalled headache diagnosis is usually a naming problem rather than a mystery in Headache Not Getting Better? What Doctors Miss, and about the opposite scenario — a completely clean scan that still leaves you in pain — in Normal MRI but Still Headaches?. If the fear driving your search is something more ominous, Is My Headache a Brain Tumor? covers what actually makes a neurologist worry, and Migraine With Aura and Stroke Risk puts the vascular numbers in proportion.
Free download: The Headache Red Flags Checklist — the warning signs I teach patients to recognize, in one page.
Still unsure about your scan or your diagnosis? Talk to a specialist.
Written by Dr. Tae-Ho Guak, MD, board-certified neurologist. This article is for general education and is not a substitute for individual medical advice. It cannot account for your full history, examination, or imaging. Please discuss your own MRI findings and symptoms with the physician who ordered the scan, and seek emergency care for any sudden, severe, or rapidly worsening headache or new neurological symptoms.
From the textbook. Chapter 5, “Red Flags, Investigations and When to Image,” of my clinical textbook Headache Medicine covers this topic in full for doctors, with the evidence and decision pathways behind it. About the book →