Does low-dose daily tadalafil reduce dementia risk or protect the brain?

The short answer

Low-dose daily tadalafil has weaker brain evidence than its popularity suggests. The observational signal that started this belongs mostly to sildenafil, not tadalafil. The one randomised trial of tadalafil in vascular cognitive impairment missed significance, and the largest study of daily low-dose tadalafil found no dementia reduction at all.

Evidence tier: Tier 1–2 for the human data (one RCT, large propensity-matched cohorts, two Mendelian randomisation studies); Tier 3 for the mechanistic reasoning about cerebral blood flow. Educational content, not medical advice. Tadalafil is a prescription medicine.

The key points:

• The strongest observational signal is sildenafil-specific — tadalafil did not reach significance in the same analyses.
• The direct test failed. PASTIS, the randomised trial of tadalafil in small vessel disease, found a non-significant increase in cerebral blood flow.
• The best-matched real-world study found nothing. 5,204 daily low-dose tadalafil users: HR 0.99.
• Genetic evidence points the other way for Alzheimer's — one study found higher odds of Alzheimer's and Lewy body dementia, alongside a possible vascular-dementia benefit.
• Tadalafil plus nitrates can be fatal. This is a real drug with a real contraindication.

Why is everyone suddenly talking about tadalafil for the brain?

Evidence tier: 3 — context, not evidence.

In August 2026, longevity researcher David Sinclair posted that he takes low-dose daily tadalafil "for long-term organ health, especially the brain," adding that vascular dementia is "the second most common and most preventable cause of dementia." He was amplifying a cardiologist who described PDE5 inhibitors as "vascular drugs that happen to work below the belt." The post reached over 250,000 views within hours.

The framing is appealing because the mechanism is real. Tadalafil inhibits phosphodiesterase-5, which raises cyclic GMP, relaxes vascular smooth muscle and increases blood flow. Cerebral small vessel disease — the chronic damage to the brain's smallest arteries — is genuinely the main driver of vascular cognitive impairment. A drug that improves small-vessel function might plausibly slow it. That is a coherent hypothesis, and it is why serious researchers have tested it.

The problem is what happened when they did.

Does tadalafil actually reduce dementia risk?

Evidence tier: 1–2 — one RCT plus large propensity-matched cohorts.

The best-matched study found nothing. A 2025 Israeli analysis using Clalit's database examined precisely the regimen being discussed: new users of low-dose daily tadalafil prescribed for benign prostatic hypertrophy. It compared 5,204 tadalafil initiators against 18,565 propensity-score-matched alpha-1 blocker initiators, with a two-year lag to reduce reverse causation, and followed them for dementia (Gronich et al., PMID 38952132).

The result was HR 0.99 (95% CI 0.88–1.12, p = 0.927). Not a weak benefit — no signal whatsoever. A second cohort of 133,336 men with erectile dysfunction, followed a mean of 7.9 years with 8,631 dementia diagnoses, gave HR 0.95 (0.86–1.04). Stratifying by which PDE5 inhibitor people took did not change it.

This matters more than any other single study here, because it tested the actual intervention — daily low-dose tadalafil, in older men, over years — rather than a proxy.

The signal that generated the excitement is sildenafil's. The widely-cited 2024 Neurology cohort of men with erectile dysfunction found reduced Alzheimer's risk in those initiated on sildenafil (adjusted HR 0.81, 95% CI 0.71–0.93). For tadalafil and vardenafil, effect estimates were similar in direction but did not reach significance (Adesuyan et al., PMID 38324745). A 2025 systematic review and meta-analysis found the same split: a clear sildenafil association, non-significant findings for tadalafil (Chua et al., PMID 40096550).

So the drug with the evidence is not the drug being recommended. If the class effect were real and simple, that distinction should not exist.

What did the actual tadalafil trial find?

Evidence tier: 1 — randomised, placebo-controlled, crossover.

PASTIS is the trial designed to answer this question directly. It tested whether tadalafil increases cerebral blood flow in older people with symptomatic cerebral small vessel disease — the exact population in whom the vascular hypothesis should pay off — in a double-blind, placebo-controlled crossover design (Pauls et al., PMID 35135037).

Tadalafil increased cerebral blood flow in all subcortical areas, but the increase was not statistically significant. The largest treatment effect was within white matter hyperintensities, at +9.8% — directionally encouraging, and the reason the question stayed open, but not a positive trial.

This is the honest state of play: the one randomised test of the specific mechanism, in the specific disease, did not demonstrate what it set out to demonstrate. A small uncontrolled study of 25 men with erectile dysfunction and mild cognitive impairment reported a 2.88-point MoCA improvement on 5 mg daily tadalafil over eight weeks, but it had no placebo group, and eight weeks of repeated cognitive testing produces practice effects that comfortably exceed three points.

What does the genetic evidence say?

Evidence tier: 2 — Mendelian randomisation, two independent studies.

Mendelian randomisation uses genetic variants as natural experiments to estimate causal effects, sidestepping much of the confounding that plagues observational drug research. It is the right tool here, and two independent groups have applied it.

The first, from Oxford, performed cis-MR across four separate Alzheimer's GWAS datasets plus a sex-stratified analysis in UK Biobank. It found no evidence of a causal association between PDE5 inhibition and Alzheimer's risk, concluding that PDE5 inhibitors are "unlikely to decrease the risk" (Alcalde-Herraiz et al., PMID 41078087).

The second is more nuanced, and considerably more interesting. A drug-target MR study found that genetically proxied PDE5 inhibition was associated with higher odds of Alzheimer's disease (OR 1.09, 95% CI 1.07–1.11) and higher odds of Lewy body dementia (OR 1.32, 95% CI 1.23–1.41) — but a trend towards lower odds of vascular dementia across all instruments. On imaging it cut both ways: lower white matter hyperintensity volume (−0.035 SD), which is a genuine marker of small vessel disease, but also reduced volume of other structures including the thalamus. The authors concluded that PDE5 inhibition "may increase risk of Alzheimer's disease and Lewy body dementia, warranting further investigation before clinical trials" (Brennan & Tinworth, PMID 39951190).

Read carefully, this partially vindicates the vascular half of the argument — reduced white matter hyperintensities and a vascular-dementia trend are exactly what the mechanism predicts — while raising a signal in the opposite direction for the two most common non-vascular dementias. That is not the story being told on social media, in either direction.

Why do the observational studies and the causal studies disagree?

Evidence tier: 2 — methodological.

The gap between "men who took PDE5 inhibitors got less Alzheimer's" and "genetic proxies for PDE5 inhibition show no protection" has an unglamorous likely explanation: the men are different.

Almost all the observational data comes from men prescribed these drugs for erectile dysfunction. Men who seek treatment for ED, obtain a prescription, and keep refilling it are systematically healthier, more engaged with healthcare, more likely to be in a relationship and more likely to be sexually active than men who do not. Every one of those is independently associated with lower dementia risk. This is textbook healthy-user bias, and it is the standard reason observational drug-repurposing signals evaporate under scrutiny.

The Israeli study is instructive precisely because it was designed to avoid this. By comparing daily tadalafil users against alpha-1 blocker users — men with the same condition, seeking the same treatment, from the same system — it removed most of that confounding. The signal went to zero.

When observational data, a randomised trial and Mendelian randomisation all point in different directions, the observational data is usually the one that is wrong.

Is low-dose daily tadalafil safe to take indefinitely?

Evidence tier: 2 — established pharmacology and label warnings.

Tadalafil is a well-characterised prescription medicine with decades of use, and daily 2.5–5 mg dosing is an approved regimen for benign prostatic hypertrophy and erectile dysfunction. It is not an exotic compound. But "well-tolerated in trials" is not "consequence-free for anyone who wants it":

• Nitrates are an absolute contraindication. Combining tadalafil with nitrate medication — including recreational "poppers" — can cause catastrophic hypotension. This is the reason it is prescription-only.
• Alpha-blockers and some antihypertensives require dose care for the same reason. Relevant, given many people in this demographic take them.
• Non-arteritic anterior ischaemic optic neuropathy (NAION), a rare form of sudden vision loss, and sudden hearing loss have both been reported with PDE5 inhibitors.
• Tadalafil's long half-life (~17.5 hours) is what makes daily dosing attractive, but it also means interactions persist far longer than with sildenafil.
• Common effects — headache, back pain, nasal congestion, dyspepsia, flushing — are dose-dependent and often what causes people to stop.

The relevant question is not whether tadalafil is dangerous. It is whether the evidence supports taking a prescription vasodilator indefinitely for an indication where the best-designed study found no benefit. That is a conversation with a prescriber, not a decision to make from a tweet.

Limitations

This is educational content, not medical advice, and tadalafil is a prescription medicine.

• Absence of evidence is not evidence of absence. The null findings show no detectable effect at studied doses and durations, not proof that none exists. A true effect could be small, slow, or confined to a subgroup.
• The vascular-dementia question is genuinely unresolved. Both the WMH imaging data and the MR vascular trend are consistent with the mechanism. Vascular dementia is under-diagnosed and under-studied, and most cohorts here were powered for Alzheimer's.
• Almost all human data is in men, largely via ED and BPH prescribing. Generalisation to women is not supported.
• The Mendelian randomisation studies conflict with each other on direction for Alzheimer's — one null, one suggesting increased risk. Neither is definitive, and MR has its own assumptions.
• PASTIS measured cerebral blood flow, not cognition or dementia incidence. A larger or longer trial could still find a clinical effect.
• Nothing here evaluates tadalafil for its approved indications, where it works well and the evidence is strong.
• Marko Maal, MSc Pharmacy reviewed this article. Reviewer attribution does not constitute a doctor-patient relationship.

The bottom line

The vascular argument for tadalafil is mechanistically reasonable and worth taking seriously — which is exactly why it has been tested properly, and why the results deserve attention rather than dismissal. When they came in, the drug with the observational signal turned out to be sildenafil rather than tadalafil; the randomised trial of tadalafil in small vessel disease missed significance; and the largest study of daily low-dose tadalafil, designed specifically to strip out healthy-user bias, returned a hazard ratio of 0.99. Genetic evidence found no protection against Alzheimer's, with one analysis suggesting possible harm for Alzheimer's and Lewy body dementia alongside possible benefit for vascular dementia specifically.

The honest summary is that this is an open question with mostly negative human answers so far, not an established longevity intervention. If you already take daily tadalafil for an approved indication, none of this is a reason to stop. If you are considering starting it for your brain, the strongest thing that can be said is that the direct evidence has not supported the hypothesis, and that a prescription vasodilator with an absolute nitrate contraindication deserves a prescriber's involvement rather than a social-media protocol. Meanwhile the modifiable risk factors that actually drive vascular dementia — blood pressure, ApoB, smoking, diabetes, sleep — remain unglamorous, well-evidenced, and largely unaddressed by a pill.

Related on this site

• Longevity biomarkers worth tracking in 2026
• Why most longevity peptide stacks don't survive evidence scrutiny
• Peptides for cognitive performance (2026)
• Cerebrolysin: neurotrophic protocol for stroke, TBI and dementia
• Peptides for longevity (2026): what's backed by evidence
• Our evidence-tier framework

References

• Gronich N, Stein N, Saliba W. Phosphodiesterase-5 Inhibitors and Dementia Risk: A Real-World Study. Neuroepidemiology 2025;59(3):193–202. PMID 38952132 — 5,204 daily low-dose tadalafil users, HR 0.99.
• Adesuyan M, et al. Phosphodiesterase Type 5 Inhibitors in Men With Erectile Dysfunction and the Risk of Alzheimer Disease: A Cohort Study. Neurology 2024;102(4):e209131. PMID 38324745 — sildenafil HR 0.81; tadalafil not significant.
• Pauls MMH, et al. The PASTIS trial: Testing tadalafil for possible use in vascular cognitive impairment. Alzheimers Dement 2022;18(12):2393–2402. PMID 35135037 — RCT; non-significant cerebral blood flow increase.
• Alcalde-Herraiz M, et al. Phosphodiesterase-5 Inhibition and Alzheimer's Disease Risk: A Mendelian Randomisation Study. Aging Cell 2025;24(12):e70265. PMID 41078087 — no causal association.
• Brennan SO, Tinworth AC. Genetically Proxied PDE5 Inhibition and Risk of Dementia: A Drug Target Mendelian Randomization Study. Mol Neurobiol 2025;62(6):7864–7874. PMID 39951190 — higher AD and LBD odds; vascular dementia trend downward.
• Chua WY, et al. Sildenafil and risk of Alzheimer disease: a systematic review and meta-analysis. Aging (Albany NY) 2025;17(3):726–739. PMID 40096550 — sildenafil-specific association.