Ginkgo biloba is one of the most famous "brain herbs" in the world, sold for memory, focus, circulation, and ringing ears. It is also a rare case where a single large, well-funded trial gave us an unusually clear answer, and that answer was mostly no. This guide walks through what the best research actually found, why the memory reputation outran the evidence, the honest dose if you still want to try it, and the safety issue that matters most: bleeding risk.
The short version
- It does not prevent dementia. The roughly 3,000-person GEM trial run over about 6 years was negative.
- No memory boost in healthy adults. Controlled trials found no meaningful effect over placebo.
- Existing dementia: results are mixed and inconsistent, with at most a modest, unproven possible benefit.
- Tinnitus, vertigo, and leg circulation: evidence ranges from weak to negative.
- Safety first: ginkgo may raise bleeding risk, so use caution with blood thinners, stop before surgery, and never eat the raw seeds.
The dementia verdict: a clear no
The single most important finding about ginkgo is a negative one, and it comes from an unusually strong study. The NIH-funded Ginkgo Evaluation of Memory (GEM) trial enrolled more than 3,000 older adults and gave them 240 mg a day of standardized ginkgo extract or placebo for about six years. Ginkgo did not prevent dementia or Alzheimer disease (DeKosky 2008), and a companion analysis found it did not slow cognitive decline either (Snitz 2009). A second large trial, GuidAge, gave 240 mg a day of EGb 761 to 2,854 French adults aged 70 and older with memory complaints for five years and also found no significant reduction in Alzheimer diagnoses (Vellas 2012). This is exactly the kind of large, long, well-run trial that settles a question, and it settled this one against the popular claim. If your reason for taking ginkgo is to ward off memory loss as you age, the best evidence says it does not work.
Memory in healthy adults
The everyday version of the claim, that ginkgo sharpens memory or focus in healthy people, fares no better. A well-run 6 week trial in adults over 60 (Solomon 2002) found no meaningful improvement in memory, attention, or other measures compared with placebo, and the multi-year GEM data pointed the same way. So the "smart pill" reputation is not supported: in people without dementia, ginkgo has not been shown to boost cognition. That does not make it dangerous, but it does make the memory marketing misleading.
The evidence scorecard
Every headline ginkgo claim, matched to what the research shows:
| Claim | Evidence | What the research shows |
|---|---|---|
| Prevents dementia (healthy older adults) | Strong evidence of no benefit | GEM trial, ~3,000 people, 6 years: no effect |
| Boosts memory (healthy adults) | No established benefit | Controlled trials found no meaningful effect |
| Eases symptoms of existing dementia | Mixed and inconsistent | 2026 Cochrane: low-certainty, highly variable gains |
| Tinnitus (ringing in ears) | Not proven | 2022 Cochrane: little to no effect |
| Leg circulation (claudication) | No clinically meaningful benefit | Cochrane: not significant, likely inflated by publication bias |
| Vertigo and dizziness | Limited and mixed | Low-quality trials, benefit not established |
Existing dementia, tinnitus, and circulation
Where ginkgo has any defenders, it is around the edges, and the picture is murky.
- Symptoms of existing dementia. This is the one area with genuine disagreement. A 2026 Cochrane review of 82 trials found little or no effect in mild cognitive impairment and only low-certainty, highly variable gains in people already diagnosed with dementia (Wieland 2026), while an industry-linked meta-analysis of the EGb 761 extract reported modest improvements in cognition and daily function (Gauthier 2014). The results are not consistent enough to call a proven benefit, and this is a decision for a clinician, not a supplement aisle.
- Tinnitus. A 2022 Cochrane review of 12 trials found ginkgo may have little to no effect on tinnitus, on very uncertain evidence (Sereda 2022), and tinnitus has a strong placebo effect that can fool people.
- Leg pain from poor circulation (claudication). The most recent Cochrane review found a small gain in walking distance that was not statistically significant; the authors judged that publication bias likely inflated it and found no clinically significant benefit (Nicolai 2013).
- Vertigo and dizziness. A few small or industry-sponsored trials hint at relief, but the studies are low quality and often lack a placebo group.
Dose and what EGb 761 means
If you still want to try ginkgo, buy it right. Most quality research uses a specific standardized leaf extract called EGb 761, defined by its manufacturing process plus 22 to 27 percent flavonol glycosides, 5.4 to 6.6 percent terpene lactones and under 5 ppm ginkgolic acids. That standardization matters because ginkgo products vary widely, and a random "ginkgo" capsule may not match what was studied. The studied dose is 120 to 240 mg a day, usually split into two or three doses, and any effect builds over several weeks rather than right away. No US supplement label we found names EGb 761, so look for 120 mg of extract standardized to 24 percent flavone glycosides and 6 percent terpene lactones that also states a ginkgolic acid limit of 5 ppm or less. Our ginkgo rankings compare current labels.
Safety and the bleeding-risk warning
The most important practical point about ginkgo is not about whether it works, but about safety:
- Bleeding risk. In a pooled analysis of 18 randomized trials, ginkgo did not significantly change platelet aggregation or standard clotting tests (Kellermann 2011), but case reports have linked it to serious bleeding, including inside the skull. Use caution or avoid it if you take anticoagulant or antiplatelet drugs such as warfarin, aspirin, or clopidogrel, or regular NSAIDs, and stop it before surgery or dental procedures: the EU herbal monograph advises 3 to 4 days beforehand, and your surgeon may ask for longer.
- Never eat raw or roasted ginkgo seeds or nuts. They contain a natural toxin (ginkgotoxin) that can cause poisoning and seizures. Only the standardized leaf extract is used as a supplement.
- Drug interactions. Ginkgo may interact with some seizure medications (it may lower the seizure threshold), efavirenz (avoid), nifedipine and dabigatran, and possibly other drugs processed by liver enzymes, so check with a pharmacist before combining it with prescriptions.
- Other notes. Side effects are usually mild (headache, upset stomach, dizziness), it can rarely trigger allergic skin reactions, and it is best avoided in pregnancy and breastfeeding since safety has not been established.
Frequently asked questions
Does ginkgo biloba actually improve memory?
In healthy adults, a memory boost is not established. Well-run trials, including a 6 week study in adults over 60, found no meaningful benefit over placebo.
Can ginkgo prevent Alzheimer disease or dementia?
No. The large NIH-funded GEM trial of more than 3,000 older adults over about 6 years found ginkgo did not prevent dementia or Alzheimer disease.
How long does ginkgo take to work?
If it helps at all, it works slowly. Studies use it for at least 6 to 12 weeks at 120 to 240 mg per day, and a benefit is not guaranteed.
Is it safe to take ginkgo with blood thinners or before surgery?
Use caution. Controlled trials found no significant effect on standard clotting tests, but case reports have linked ginkgo to serious bleeding, so talk with your doctor before combining it with warfarin, aspirin, or clopidogrel. The EU herbal monograph advises stopping it 3 to 4 days before surgery; follow your surgeon if they ask for longer.
What is EGb 761 and what dose was studied?
EGb 761 is a standardized ginkgo leaf extract made by a defined process, with 22 to 27 percent flavonol glycosides, 5.4 to 6.6 percent terpene lactones and under 5 ppm ginkgolic acids. Most large ginkgo trials used it, typically at 120 to 240 mg per day. No US supplement label we found names it.
Does ginkgo help tinnitus or dizziness?
The evidence is weak. A 2022 Cochrane review found little to no effect on tinnitus, and the evidence for vertigo is limited and mixed.
The bottom line
Ginkgo biloba is a case where a strong study did the public a favor. The large GEM trial showed it does not prevent dementia or slow cognitive decline, and it does not boost memory in healthy adults either, so the reputation that sells it is not backed by the evidence. Its only areas of possible benefit, symptoms of existing dementia, are mixed and belong in a doctor's hands, while tinnitus and circulation claims are weak to negative. If you still choose to try it, use a standardized 24/6 extract with a stated ginkgolic acid limit at 120 to 240 mg a day, and take the bleeding-risk warning seriously: avoid it with blood thinners, stop it before surgery, and never eat the seeds.
