For years I carried the same picture of a stroke that most people do: a pipe clogs, blood stops, the tissue downstream dies. It is the model that put low-dose aspirin in millions of medicine cabinets, mine eventually included. So I did a genuine double take at a finding that turns it inside out. One of the most common strokes may not be a clog at all. The arteries feeding it were not too narrow. They were too wide.

The stroke in question is the lacunar stroke, and it is not some rare curiosity. It accounts for about a quarter of all ischemic strokes, the small deep-brain strokes that quietly stack up over the years and drag people toward vascular dementia and unsteady gait. The working assumption for decades was that a lacunar stroke was just a miniature of the big kind: fatty plaque narrows a vessel, starves the tissue behind it. Treat the plaque, treat the stroke. That is the logic aspirin and the other antiplatelet drugs were built on.

So look at what a team led by Fei Han and Joanna Wardlaw at the University of Edinburgh did with that assumption. They took 229 patients who had recently had either a lacunar stroke or a mild non-lacunar one, scanned their brains with MRI at the start and again a year later, and tracked cognition and vessel structure the whole way. What jumped out was not narrowing. It was the reverse. People whose deep-brain arteries were dilated and elongated, a condition with the clinical name dolichoectasia, were more than 4 times as likely to have had a lacunar stroke. The wider and more stretched the vessels, the worse the small-vessel disease, and the faster the brain damage moved.

Wait, why would a stretched, ballooned artery be more dangerous than a pinched one? That question is the whole reason this study got under my skin, because it flips the plumbing picture on its head. These tiny deep vessels were never really behaving like pipes. When the wall of a small arteriole loses its structure, the muscle and elastic layers that let it clench and relax on cue, it does not so much clog as give up the job. It stiffens in some segments, balloons in others, stops fine-tuning its own blood flow, and the lining that keeps the whole system responsive goes quiet. The Edinburgh group’s name for it is segmental arteriolar disorganization, part of what they describe as a non-atheromatous, intrinsic microvascular pathology. In plainer terms, the vessel is sick from the inside, and the plaque the old model kept chasing was never where the trouble started.

They did not rest it on their own cohort, either. They pulled together a meta-analysis of 27 studies and more than 9,500 lacunar-stroke patients, and across all of it there was little evidence that large-artery narrowing actually drives these infarcts. The dolichoectasia-and-small-vessel-disease link held up. The mechanism most of us were taught, and that most patients are still treated for, has been standing on surprisingly thin evidence.

You can see the cost of that inside the same year of scans. More than a quarter of participants developed silent strokes, the covert little infarcts that never announce themselves with a dramatic collapse but keep chipping away at memory and balance. These were people already on standard secondary-prevention treatment, a toolkit built around the large-artery model, while the damage kept accruing in vessel walls that toolkit does not reach. “This study provides strong evidence that lacunar stroke is not caused by fatty blockage of larger arteries,” Wardlaw said, “but by disease of the small vessels within the brain itself.”

I want to be straight about what this can and cannot carry. It is an observational cohort, so a widened vessel is a powerful marker of small-vessel disease, not yet proof that the widening itself causes the stroke. Nobody has shown that dilation is the trigger rather than the fingerprint. But the association is large, it agrees with the meta-analysis, and it finally explains the thing the plumbing model never could: why the standard drugs keep underperforming in exactly these patients.

So what would actually help, and here is where the story turns quietly infuriating. Drugs that work on the vessel wall instead of the clot already exist, and they are cheap. In the LACI-2 trial, a phase 2 feasibility study out of the same Edinburgh group, 363 lacunar-stroke patients were given cilostazol, isosorbide mononitrate, both, or neither, two old generic medicines that act on endothelial function and blood-vessel tone rather than on platelets. The drugs were well tolerated and 98.6 percent of participants completed the full year. The signals were worth taking seriously: isosorbide mononitrate reduced recurrent stroke and cognitive impairment, cilostazol reduced dependence, and taken together they cut a composite of vascular, cognitive, and dependency outcomes to 48.6 percent of patients versus 69.6 percent on neither. Both drugs are off-patent. Neither makes anyone a fortune. And the larger, definitive trial that would move them into routine care has been slow to arrive.

That is the pattern I keep noticing, and I will name it as analysis rather than proof: the model was probably wrong for a long time, the standard of care was aimed at the wrong target, and the correction points toward two generics no company has much reason to push. The funding here tells its own story. It came from public and charitable sources, the UK Dementia Research Institute, the Stroke Association, the British Heart Foundation and others, not a drugmaker with a molecule to sell. Findings that cut against a lucrative model tend to come from the people who have no model to protect.

I am not going to tell you to stop a medication a doctor prescribed after a stroke, and I would not do that myself. But if someone I loved had a lacunar stroke, I would not accept “here is your aspirin, see you in a year” as the end of it. I would ask the neurologist directly whether the MRI shows small-vessel disease or dolichoectasia, and whether cilostazol or isosorbide mononitrate belong in the plan. The vessels were never a simple clog, and I would want the treatment to finally match the biology inside them.

Sources

  1. ScienceDaily – “The real cause of a common stroke may have been missed for decades” (2026)
  2. medRxiv – Han, Wardlaw et al., “Implications of Cranial Arterial Stenosis and Dolichoectasia for Cerebral Small Vessel Disease Etiopathogenesis” (2026)
  3. SciTechDaily – “New Stroke Study Challenges Decades-Old Medical Beliefs” (2026)
  4. PMC – “Isosorbide Mononitrate and Cilostazol Treatment in Symptomatic Cerebral Small Vessel Disease: The LACI-2 Randomized Clinical Trial” (2023)
  5. tctmd – “Cilostazol, Isosorbide Mononitrate May Help After Lacunar Stroke: LACI-2”