Vascular Dementia — The One Most Tied to Risk Factors You Can Change

Vascular dementia is the third post in our series explaining the dementia diagnoses, after mild cognitive impairment and Alzheimer's disease.

It is the subtype where our fourteen posts on dementia risk stop being abstract, because its risk factors are the cardiovascular ones. It is also where we have to report something that sits awkwardly alongside those posts: the randomised evidence that treating those risk factors prevents dementia is weaker than the risk-factor framing implies. Both of those things are in here.

We are not doctors, and nothing here is treatment advice.

What it actually is

Vascular dementia is cognitive impairment caused by damage to the blood supply of the brain. Two broad routes:

Large-vessel disease. A stroke, or several, destroying tissue outright. If a stroke lands somewhere strategically important for memory or language, a single event can produce dementia. This is the version that progresses in visible steps — a sudden drop, a plateau, another drop — which is the classic textbook picture.

Small-vessel disease. The deep, narrow arteries get stiff and leaky over years. The result on a scan is white matter changes and small deep infarcts, and the result for the person is usually gradual rather than stepwise: slowed thinking, difficulty with planning and attention, problems with walking and balance, sometimes mood changes. This is far commoner than the stepwise version, and it looks much less like the textbook.

One thing worth knowing about how it presents: in Alzheimer's, memory typically goes first. In vascular dementia, processing speed and executive function — planning, sequencing, switching between tasks — usually go first, with memory relatively better preserved early on. Someone who has become slow and disorganised rather than forgetful is a different clinical picture, and it gets missed because it doesn't match what people expect dementia to look like.

It is almost never pure, and the evidence for that is worth seeing

This is the part most explainers skip. Vascular and Alzheimer's pathology coexist in most people who have either, and the diagnostic categories are much cleaner than the brains are. Two lines of published evidence make the point:

You cannot reliably tell them apart by inflammation. A 2026 meta-analysis of 15 observational studies and 1,728 participants tested whether inflammatory markers distinguish vascular dementia from Alzheimer's.1 Pooled across all studies, there was no significant difference in IL-6 (SMD −0.13, 95% CI −0.44 to 0.19), TNF-α (−0.22, −0.71 to 0.28) or CRP (0.73, −0.17 to 1.62). Subgroup analyses found context-dependent variation by region, sample size and study quality — which is a polite way of saying the signal moves around depending on who is looking.

And microvascular damage raises the risk of both, almost equally. Diabetic retinopathy is damage to the tiny vessels of the retina, and a useful window onto small-vessel disease elsewhere. A 2026 meta-analysis of 10 cohort studies covering 1,720,128 participants found retinopathy associated with higher risk of all-cause dementia (HR 1.24, 95% CI 1.14–1.36), vascular dementia (HR 1.20, 1.05–1.37) and Alzheimer's disease (HR 1.23, 1.11–1.37).2

Read those last two numbers together. If vascular damage caused only vascular dementia, a marker of small-vessel disease should raise vascular dementia risk and leave Alzheimer's alone. It doesn't. Vascular injury appears to contribute to the clinical syndrome we label Alzheimer's as well — which is the strongest practical argument for treating cardiovascular risk regardless of which dementia label anyone is worried about.

What the randomised evidence on blood pressure actually shows

Now the uncomfortable part, and we would rather print it than let our own risk-factor posts stand without it.

A 2026 systematic review and network meta-analysis pooled four multicentre randomised trials — Syst-Eur, PROGRESS, SCOPE and HYVET-COG — covering 16,823 hypertensive older adults with no dementia at baseline, followed for 2.0 to 3.9 years.3 What it found:

  • Cognitive function scores improved on antihypertensive treatment: standardised mean difference 0.06 (95% CI 0.03–0.09, p < 0.001), GRADE moderate certainty. Statistically solid. An SMD of 0.06 is also very small — real, and not something a person would notice.
  • Dementia incidence did not significantly fall: odds ratio 0.89 (0.76–1.05), p = 0.172, GRADE low certainty. That is a trend in the right direction, not a result.
  • Exploratory ranking put calcium channel blockers highest for dementia prevention (SUCRA 94.7%) and diuretic-plus-ACE-inhibitor combinations highest for cognitive improvement (95.3%) — but the authors state the rankings were highly uncertain because the network was sparse, with one study per node. Nobody should choose a blood pressure drug off that ranking, including us.

Why this is not a contradiction of the risk-factor evidence. The observational case against hypertension rests on midlife blood pressure measured over decades. These trials treated older adults for two to four years. That is the wrong window to detect dementia prevention in — the pathology has had thirty years' head start and the follow-up is shorter than the interval over which dementia develops. The authors say exactly this: current evidence is limited by few studies and short follow-up, and large trials with dementia as a primary endpoint are urgently needed.

So the honest position, which is narrower than either "controlling blood pressure prevents dementia" or "it doesn't": treating high blood pressure is strongly justified on cardiovascular grounds, is associated with better cognitive scores by a small margin with moderate certainty, and the direct randomised evidence that it prevents dementia over a few years in older adults is a non-significant trend at low certainty. Our fourteen factors post already warns against treating the population-level estimates as precise; this is the trial-level version of the same caution.

Treatment: a field that largely stopped

There is no licensed disease-modifying drug for vascular dementia, and the state of the trial literature is itself the headline.

A 2026 systematic review searched for every randomised controlled trial of a drug targeting cognition in vascular dementia between 1990 and 2025. It found 16 trials and 5,668 participants — and with one exception, every one was published in 2012 or earlier.4 For the second most common cause of dementia, randomised drug research has been close to dormant for over a decade.

What those older trials showed:

  • Galantamine reduced ADAS-Cog/11 scores by 2.01 points (2 trials, n = 1,332; 95% CI −3.18 to −0.85) and ADAS-Cog/13 by 2.51 (−3.87 to −1.15). Moderate certainty.
  • Memantine reduced ADAS-Cog/11 by 2.20 points (2 trials, n = 752; −3.24 to −1.15). Moderate certainty.
  • Neither helped the outcomes beyond cognitive scores: galantamine showed no clear benefit on neuropsychiatric symptoms (MD −0.13, −4.05 to 3.79 — a confidence interval straddling zero in both directions), and memantine none on global function.
  • Seven other interventions — nimodipine, rivastigmine, pentoxifylline, citicoline, sulodexide, donepezil and a herbal granule — were tested in single trials and could not be pooled.

The authors' conclusion: small short-term cognitive effects, and patient-important benefits remain uncertain. A two-point move on a 70-point cognitive scale, with no accompanying improvement in how someone functions or feels, is not nothing — but it is not what anyone means by treatment either.

The ginkgo result, and why we are reporting it instead of selling it

The same review assessed EGb 761, a standardised Ginkgo biloba extract, across three trials. It reduced SKT cognitive scores by 2.65 points (n = 283; 95% CI −5.17 to −0.12).

Now look at what sits underneath that number. Heterogeneity was I² = 92.8% — the three trials barely agree with each other. Certainty was rated very low. The upper bound of the confidence interval is −0.12, which is to say almost exactly zero. Total sample: 283 people. The authors' own words are a "fragile, very low-certainty signal."

We do not sell ginkgo, and this is a good illustration of why. A statistically significant result in 283 people with 92.8% heterogeneity and a confidence interval nearly touching zero is exactly the kind of finding a supplement gets built on — it is technically positive, it is published, and it mentions a real cognitive scale. It is also the kind of finding that does not survive a larger, better trial more often than it does. We would rather print the I² than the point estimate, and our standing argument for that is in why we don't sell brain blends.

Where we come in, which is nowhere again

We sell nothing that treats or prevents vascular dementia. No ingredient in our range has been tested for it. The honest list of what does have evidence behind it is the cardiovascular one: blood pressure, diabetes, LDL cholesterol, smoking, physical activity, weight — and, per the retinopathy finding above, those matter for the Alzheimer's label too, not just this one.

That is covered with the actual effect sizes in modifiable dementia risk, the cardiovascular thread specifically in the fourteen factors, cholesterol in LDL cholesterol and dementia risk, and inactivity in physical inactivity and dementia risk. The comparison between all of that and what we sell is in exercise versus supplements.

The short version

  • Vascular dementia is cognitive impairment from damaged brain blood supply — stepwise after strokes, or gradual from small-vessel disease, which is commoner.
  • It typically takes processing speed and executive function first, with memory better preserved early. That is why it gets missed.
  • It is rarely pure. Inflammatory markers can't reliably separate it from Alzheimer's, and diabetic retinopathy raises Alzheimer's risk (HR 1.23) about as much as vascular dementia risk (HR 1.20) across 1.72 million people.
  • Across four randomised trials and 16,823 older adults, antihypertensives improved cognitive scores by SMD 0.06 (moderate certainty) but reduced dementia incidence only as a non-significant trend, OR 0.89, p = 0.172, low certainty.
  • That is a short-follow-up trial in older adults, not a refutation of the midlife observational evidence. Both can be true.
  • Of 16 randomised drug trials in vascular dementia, all but one predate 2012. Galantamine and memantine move cognitive scores about 2 points; neither improves function or neuropsychiatric symptoms.
  • Ginkgo's signal rests on 283 people with 92.8% heterogeneity and very low certainty. We don't sell it.
  • We sell nothing that treats or prevents this.

The rest of this series

Six diagnoses, each written the same way: what it is, what the evidence actually shows, and what we do and do not sell for it.

This article is for informational purposes only and has not been evaluated by the FDA. It is not intended to diagnose, treat, cure, or prevent any disease. We are not doctors. Vascular dementia requires specialist diagnosis, and blood pressure, diabetes and cholesterol management belong with the clinician who prescribes for you — nothing here is a reason to start, stop or change any medication, including a blood pressure medication. If you have had a stroke or a TIA, or you notice sudden changes in thinking, speech or movement, that is urgent medical attention rather than a reading matter.

References

  1. Ling Y, Sun J, Hu L, Zhao M, Chen J, Zheng Y. "Diagnostic Value of Inflammatory Biomarkers in Differentiating Vascular Dementia From Alzheimer's Disease: A Systematic Review and Meta-Analysis." Brain and Behavior, 2026. doi:10.1002/brb3.71341
  2. Hu H, Wang Y, Xu Y, Chen L. "Diabetic retinopathy and the risk of all-cause dementia, Alzheimer's disease, and vascular dementia: a systematic review and meta-analysis." Frontiers in Medicine, 2026. doi:10.3389/fmed.2026.1716648
  3. Ding L, Zhang M, Wu H, Wang X, Zhu X, Chen Z, Zhang Y, Liu Z. "Impact of Different Antihypertensive Drug Classes on Incident Dementia in Older Adults: A Systematic Review and Meta-Analysis." Journal of the American Medical Directors Association, 2026, 27(6):106185. doi:10.1016/j.jamda.2026.106185
  4. Cheng T, Zhang L, Zhang Y, Liang H, Wang J, Chen G, Huo Y. "Pharmacological interventions targeting cognition and related clinical outcomes in vascular dementia: a systematic review and meta-analysis of randomized controlled trials." European Journal of Clinical Pharmacology, 2026, 82(9):240. doi:10.1007/s00228-026-04158-9