Ginkgo has a bigger reputation than the trials support. It is sold to keep memory sharp and hold off dementia, and this is the part the research has tested hardest and does not back: the two largest prevention trials, together enrolling nearly six thousand older adults on a standardized 240 mg dose for years, found no reduction in dementia and no slowing of normal cognitive decline. The signals that hold up are narrower. In people who already have dementia, a standardized extract gives a modest lift to cognition and daily function.
In tinnitus and in the leg pain of poor circulation the results are mixed and small. The safety point to carry through all of it is that ginkgo thins the blood, which matters if you take a blood thinner or have surgery coming. Here is where each claim stands.
Findings & Outcomes
What It Is
Ginkgo is an extract of the fan-shaped leaves of the ginkgo tree, one of the oldest tree species alive. Almost all of the Western research uses one well-defined preparation, a standardized leaf extract known as EGb761, made to a fixed recipe of about 24 percent flavone glycosides and 6 percent terpene lactones, taken as a tablet at 120 to 240 mg a day. It reaches most people as a memory supplement, and memory is the claim the trials have tested hardest.
Anatomy of the Practice
1In the small vessels
The extract mildly relaxes small blood vessels and makes red cells and platelets less sticky, so blood moves a little more easily through fine circulation. This is the property behind both the circulation research and the bleeding caution: the same effect that eases flow also thins the blood.
2In the tissues
The flavonoids in the leaf clear reactive oxygen and the terpene lactones block platelet-activating factor, a signaling molecule in clotting and inflammation. These were the mechanisms that made ginkgo a plausible candidate for protecting the aging brain and the inner ear.
3At the level of outcomes
A plausible mechanism is only a starting point. When the outcomes were measured, whether people developed dementia, whether memory held, or whether the ringing quieted, the mechanism did not translate into a large benefit.
How It Works
The case for ginkgo rests on circulation and antioxidant chemistry. The terpene lactones block platelet-activating factor and the flavonoids neutralize reactive oxygen, and together they modestly improve flow through small vessels and reduce oxidative stress in the tissues those vessels feed. That is established pharmacology, and it is the reason ginkgo was expected to help the brain, the inner ear and the circulation in the legs. A mechanism has to be checked against what happens to people, and when ginkgo was checked at scale, the hard outcomes mostly did not change.
The research uses a standardized extract at a defined dose, not whatever a bottle labeled ginkgo happens to contain. The prevention and dementia trials used 240 mg a day of EGb761; the tinnitus and leg-circulation work used similar standardized preparations. A loosely made product tested against those results is not the same intervention, which is part of why the supplement aisle and the trial record can look so different. The circulation angle connects to the wider cardiovascular picture, and the most heavily marketed ginkgo use, quieting tinnitus, has its own page where the evidence is laid out in full.
What the Trials Found
Ginkgo entered Western use with three things behind it:
- a plausible mechanism from its chemistry,
- decades of European prescribing,
- and encouraging early trials in memory clinics.
The expectation was that an herb that improved small-vessel flow would help the aging brain, and for years it was repeated as settled. It was not, and prevention is where it was tested most directly.
Two large, long, well-run trials tested prevention head-on. The Ginkgo Evaluation of Memory study gave 3,069 older adults 240 mg a day of the standardized extract and followed them about six years, and found no reduction in the rate of new dementia and no slowing of ordinary cognitive decline. GuidAge, in France, gave 2,854 older adults with memory complaints the same extract for five years and also found no reduction in progression to Alzheimer disease. Two prevention trials of this size and length landing on the same result is about as clear as this kind of question gets.
For holding off dementia in a healthy or worried older adult, ginkgo did not do it.
Below prevention, the picture narrows. In people who already have a diagnosis of dementia, the evidence is different and more favorable: a meta-analysis of the standardized extract at 240 mg a day found modest improvements in cognition and in the activities of daily living compared with placebo. A Cochrane review of the wider literature judged the overall evidence inconsistent and difficult to rely on, with the more consistent signals clustering at the higher 240 mg dose.
The two other marketed uses come out weaker. In tinnitus, a Cochrane review found the standardized extract did not clearly help people whose main complaint was the ringing itself. In the leg pain of intermittent claudication, an early meta-analysis found a small increase in pain-free walking distance, while a later Cochrane review that pooled more trials found no meaningful benefit over placebo.
Vertigo is the one peripheral use with a firmer signal. In a 12-week randomized trial of 160 adults with vertigo, EGb 761 at 240 mg a day worked about as well as betahistine, a standard prescription for dizziness, with 79 percent of the ginkgo group rated much or very much improved against 70 percent on the drug, and fewer side effects. The trial had no placebo arm, and the maker of the extract sponsored it, so it shows ginkgo matching an accepted treatment rather than beating an inactive one.
Read together, the trials place ginkgo like this:
- It does not prevent dementia or sharpen a healthy memory.
- It gives a modest lift to the symptoms of dementia already present.
- It performs about as well as a standard drug for vertigo, in a single trial with no placebo arm.
- It does little to nothing dependable for tinnitus or the leg pain of claudication.
The Research & Studies
Everything here is based on the research we have collected and checked, sorted into groups and ordered with the strongest evidence first. Click any claim to open the studies behind it.
Cognition
No reduction in dementia on 240 mg a day over six years in 3,069 older adults
Taking a standardized ginkgo extract every day for about six years did not lower the chance of developing dementia compared with a dummy pill.
The Ginkgo Evaluation of Memory (GEM) study randomized 3,069 community-dwelling adults aged 75 and older, with normal cognition or mild cognitive impairment at entry, to 120 mg of standardized EGb761 twice daily (240 mg total) or placebo, and followed them a median of 6.1 years. There was no difference in the rate of incident dementia or of Alzheimer-type dementia between groups. This is the largest and longest prevention trial of ginkgo.
Who this may not transfer to:Tested in adults 75 and older; it measured whether ginkgo prevents dementia, not whether it treats established dementia.
The study · 1
DeKosky 2008, JAMA · JAMA
Memory and thinking declined at the same pace as placebo in 3,069 older adults
The everyday decline in memory and thinking that comes with age went at the same pace whether older adults took standardized ginkgo or a dummy pill.
A pre-specified analysis of the GEM cohort (3,069 adults aged 72 to 96) compared the rate of cognitive decline between 240 mg a day of standardized EGb761 and placebo across a battery of neuropsychological tests. There was no difference in the trajectory of any cognitive domain over a median 6.1 years of follow-up. This addresses the popular claim that ginkgo keeps a healthy aging mind sharp.
Who this may not transfer to:Adults 72 and older with healthy or mildly impaired cognition; it does not test younger users.
The study · 1
Snitz 2009, JAMA · JAMA
No fewer Alzheimer's diagnoses over 5 years in 2,854 adults with memory complaints
In older adults who had noticed memory problems, five years of standardized ginkgo did not reduce how many went on to develop Alzheimer disease.
The GuidAge trial randomized 2,854 French adults aged 70 and older who had spontaneously reported memory complaints to their doctor to 120 mg of standardized EGb761 twice daily or placebo for five years. Conversion to probable Alzheimer disease did not differ significantly between groups. The result corroborates the GEM prevention finding in a different population and country.
Who this may not transfer to:Older European adults who already noticed memory changes; it repeats the US prevention result in a different country.
The study · 1
Vellas 2012, Lancet Neurol · Lancet Neurol
A small gain in cognition and daily function at 240 mg a day in people who already have dementia
For people who already have dementia, a standardized ginkgo extract at 240 mg a day gave a modest improvement in thinking and in managing daily tasks compared with a dummy pill.
A systematic review and meta-analysis pooling nine randomized placebo-controlled trials (2,561 patients) of ginkgo in cognitive impairment and dementia found that EGb761 at 240 mg a day produced statistically significant improvements in cognition and in activities of daily living relative to placebo, with the clearer effects at the higher dose. The benefit is in treating symptoms of established dementia, a different question from the prevention trials, and the effect is modest.
Who this may not transfer to:Applies to people who already carry a dementia diagnosis, not to healthy adults hoping to prevent it.
The study · 1
Tan 2015, J Alzheimers Dis · J Alzheimers Dis
Pooled across 36 trials, the dementia evidence is inconsistent and unreliable
When all the ginkgo dementia trials are read together, the results disagree with one another enough that a firm conclusion cannot be drawn.
A Cochrane systematic review pooling 36 trials of ginkgo for cognitive impairment and dementia concluded that the evidence was inconsistent and unreliable, noting that while some trials, particularly at higher doses, reported benefits, the body of evidence as a whole did not support a predictable clinical effect. This is the reason not to rely on any single positive meta-analysis.
Who this may not transfer to:A judgment on the whole trial base for cognitive impairment and dementia, not one study.
The study · 1
Birks 2009, Cochrane Database Syst Rev · Cochrane Database Syst Rev
Hearing And Tinnitus
No clear relief when ringing in the ears is the main complaint, across 12 trials
For people whose main problem is ringing in the ears, standardized ginkgo did not clearly reduce the ringing or the distress it causes.
A Cochrane review of ginkgo for tinnitus pooled 12 randomized placebo-controlled trials (1,915 participants) and found no clear evidence that ginkgo reduces tinnitus severity when tinnitus is the primary presenting symptom, rating the certainty of the evidence as low. Some older trials in which tinnitus was a secondary symptom of another condition are a separate and weaker literature.
Who this may not transfer to:People whose main complaint is the ringing itself, not tinnitus that is a symptom of another condition.
The study · 1
Sereda 2022, Cochrane Database Syst Rev · Cochrane Database Syst Rev
Heart And Vascular
No dependable gain in walking distance across 14 pooled trials
When more trials were added, standardized ginkgo did not meaningfully help people with poor-circulation leg pain walk further than a dummy pill did.
A Cochrane review of ginkgo for intermittent claudication pooled 14 randomized placebo-controlled trials (739 participants) and concluded there was no evidence of a clinically significant benefit on pain-free or maximum walking distance. This later, larger review supersedes the earlier positive meta-analysis and points toward no dependable effect.
Who this may not transfer to:Adults with peripheral arterial disease; the larger, later pooling of the walking-distance trials.
The study · 1
Nicolai 2013, Cochrane Database Syst Rev · Cochrane Database Syst Rev
About 112 feet (34 meters) more pain-free walking in an early pool of 8 trials
In people with leg pain from poor circulation, early pooled trials suggested ginkgo let them walk a little further before the pain started.
A meta-analysis of 8 randomized placebo-controlled trials found ginkgo produced a modest, statistically significant increase in pain-free walking distance in intermittent claudication compared with placebo, a weighted mean difference of 112 feet (34 meters) (95% CI 26 to 43). The absolute gain was small and the authors noted the trials were modest in size and quality, so they treated it as a preliminary signal that later work would need to confirm.
Who this may not transfer to:Adults with peripheral arterial disease; an early signal the larger later review did not confirm.
The study · 1
Pittler 2000, Am J Med · Am J Med
Vestibular
Matched betahistine, a standard vertigo drug, over 12 weeks in a 160-patient trial
For vertigo, standardized ginkgo held its own against betahistine, a common prescription for dizziness. In a 12-week trial the two worked about equally, and ginkgo caused fewer side effects.
A randomized, double-blind trial (Sokolova 2014, Int J Otolaryngol) gave 160 adults with vertigo, mean age 58, either EGb 761 at 240 mg a day or betahistine at 32 mg a day for 12 weeks. Both groups improved on all outcome measures, including a vertigo symptom scale and clinician global impression, with no significant difference between them, and the ginkgo group's gains were numerically slightly larger. Clinicians rated 79% of the EGb 761 group and 70% of the betahistine group much or very much improved, and EGb 761 caused fewer adverse events. The limit that matters is that the trial had no placebo arm, so the improvement in both groups also includes the natural settling of vertigo over time; what it shows is that ginkgo matched an accepted drug, not that it beat placebo.
Who this may not transfer to:Middle-aged and older adults with vertigo; the comparison is against an active drug, not placebo.
For vertigo specifically, standardized EGb 761 is one of the better-supported ginkgo uses, but the cause of the dizziness should be assessed first, and the bleeding cautions still apply.
The study · 1
Sokolova 2014, Int J Otolaryngol · Int J Otolaryngol 2014;2014:682439
Ways to Do It
Match what you do to what you want. If the goal is protecting your memory for the years ahead, the strongest levers are not ginkgo, so the first rung is free. If the goal is trying a standardized extract for symptoms that already exist, the later rungs cover how to do that sensibly.
For keeping a mind sharp into later life, the best-supported levers are keeping blood pressure controlled, moving most days, sleeping well, staying socially and mentally engaged, and not smoking. None of these is ginkgo, and together they carry far more evidence than any supplement for brain aging. This is the fair first answer for anyone whose goal is prevention.
If you are going to try it, the only version with a research record is a standardized leaf extract of the EGb761 type, made to about 24 percent flavone glycosides and 6 percent terpene lactones. A product simply labeled ginkgo, with no standardization stated, was not what the trials tested. Look for that specification and an independent testing mark on the label.
The modest benefit in people who already have dementia showed up at 240 mg a day of the standardized extract, taken for at least several weeks before judging it. This belongs alongside a proper medical assessment of the memory change, not instead of one, and it pairs with the bleeding cautions below, which is why a clinician should know you are taking it.
If you are healthy and hoping to hold off decline, the two large trials point away from ginkgo doing that. Spending the effort on the first rung will do more for the same goal. Trying ginkgo anyway is a reasonable personal choice, as long as you go in knowing prevention is the part the research did not find.
The trials that saw any symptom benefit ran for weeks to months, so a few days tells you nothing. If you try it, give it eight to twelve weeks at the standardized 240 mg dose, decide with the person managing your care whether anything has changed, and stop if it has not.
Go Deeper
- Tinnitus: the most heavily marketed use of ginkgo, with the full evidence on what does and does not quiet the ringing.
- Omega-3 and fish oil: another supplement where the large trials revised a confident popular claim, and a useful comparison for reading past a headline.
The Chinese Medicine View
Chinese medicine and the Western supplement use two different parts of the same tree. The Western product is the leaf extract. The classical Chinese drug is the seed inside the fruit, Bai Guo, and it is used for something else entirely.
Bai Guo, the ginkgo seed, is a longstanding item in the materia medica, classed as astringent and used mainly to bind the Lung and calm wheezing, which places it among the herbs for cough and asthma; it has no classical link to memory or circulation. It appears in that role in the classic wheezing formula Ding Chuan Tang, and it is also used for leukorrhea and for frequent urination. The tradition treats it with a specific caution of its own: the raw seed is considered toxic and is used only in small, prepared amounts, a caution consistent with modern chemistry, since the seed contains a compound that can provoke seizures in quantity. The ginkgo leaf is a much more recent addition to Chinese herbal use and does not carry the deep classical record the seed does.
So the two views do not converge, and reading one into the other would be an invention. The tradition reasoned about a seed that astringes the Lung and must be dosed carefully because it is toxic raw. Western research studied a leaf extract for the brain and the circulation and found the memory claim mostly did not hold. They share a plant and little else. If you know Bai Guo from Chinese medicine, the leaf supplement on the shelf is a different medicine, from a different part of the tree, used for a different purpose.
Cautions For This Practice
Extra restraintEverything to be aware of is here, in one place. This practice suits most healthy people; a few situations call for real care.
Serious spontaneous bleeding in 15 reported cases, especially with blood thinners
A case report and systematic review of the literature identified 15 published cases of bleeding associated with ginkgo, including 8 episodes of intracranial hemorrhage and bleeding into the eye, several in people also taking antiplatelet or anticoagulant drugs. The mechanism is ginkgo inhibition of platelet-activating factor, which reduces platelet aggregation. Controlled trials of platelet function have been less consistent, but the case evidence and the mechanism together warrant caution around blood thinners and surgery.Bent 2005, J Gen Intern Med
Bleeding, blood thinners and surgery
Ginkgo makes platelets less sticky and blocks platelet-activating factor, so it thins the blood, and there are documented cases of spontaneous bleeding, including into the eye and the brain, in people taking it. Take it seriously if you use warfarin, aspirin, clopidogrel or any other blood thinner, if you have a bleeding disorder, or if you have surgery or a dental procedure coming up. Tell the clinician who manages those, and it is standard to stop ginkgo one to two weeks before planned surgery.
The raw seed and seizures
The traditional Chinese drug is the seed, Bai Guo, and the raw or undercooked seed contains a compound that can trigger seizures when eaten in quantity, with children most at risk. This is a caution about eating ginkgo seeds, not about the standardized leaf extract, which is processed to remove most of that compound. The plant and the word are the same even though the products differ, which is why the seed caution gets misapplied to the supplement.
Product quality and adulteration
Ginkgo is among the more frequently adulterated botanicals on the market, sometimes bulked with cheaper flavonoids so a lab test looks right. A product that does not state EGb761-type standardization, or carry an independent testing mark, may not be what the trials studied or what the label claims. This is the quality variable most within your control.
Allergy and ginkgolic acids
Poorly processed extracts can retain ginkgolic acids, which are related to the compounds in poison ivy and can cause allergic skin and mucous-membrane reactions. People sensitive to those plants, and anyone who reacts to a ginkgo product, should avoid it. A properly standardized extract keeps these acids low.
Pregnancy, and when a memory change needs a doctor
There is little safety data in pregnancy, and the blood-thinning effect is a further reason to avoid ginkgo then. Separately, a new or worsening memory problem should be assessed medically; the treatable causes are worth finding, and ginkgo did not prevent dementia in the trials that tested it.
Start slow, be smart, read the research, and consult a professional if you have any concerns. This is here to inform your choice, not make it for you.
Common Questions
Does ginkgo improve memory or prevent dementia?
For a healthy person, the large trials say no. The Ginkgo Evaluation of Memory study gave 3,069 older adults a standardized 240 mg a day for about six years and found no reduction in new dementia and no slowing of ordinary cognitive decline, and the French GuidAge trial reached the same result over five years in people with memory complaints. Two prevention trials of that size landing together is a clear answer: ginkgo did not keep memory sharp or hold off dementia in people who did not already have it.
So is ginkgo useless?
Not quite. In people who already have dementia, a standardized extract at 240 mg a day gave a modest improvement in thinking and daily function compared with placebo, pooled across trials. In tinnitus, a Cochrane review found it did not clearly help. In the leg pain of poor circulation, an early meta-analysis found a small gain in walking distance while a later, larger review found none. The picture is small benefits in one specific group and mixed-to-flat results elsewhere, a long way from the memory-herb reputation.
Does ginkgo help with dizziness or vertigo?
This is ginkgo's better peripheral result. In a 12-week trial of 160 adults with vertigo, standardized EGb 761 at 240 mg a day matched betahistine, a common prescription for dizziness, with clinicians rating 79 percent of the ginkgo group much or very much improved against 70 percent on the drug, and ginkgo caused fewer side effects. The catch is that the trial had no placebo group and was funded by the extract's maker, so it shows ginkgo keeping pace with an accepted drug rather than beating placebo. Have the cause of the dizziness checked first, since vertigo has many causes, and the bleeding cautions above still apply.
Is ginkgo safe to take with blood thinners?
This is the safety point to raise with a clinician. Ginkgo reduces platelet stickiness and thins the blood, and there are case reports of serious spontaneous bleeding in people taking it, so combining it with warfarin, aspirin, clopidogrel or another blood thinner adds risk. The same applies before surgery or a dental procedure, where stopping ginkgo one to two weeks ahead is standard. If you take any blood thinner or have a bleeding disorder, clear ginkgo with the clinician who manages that before you use it.
Is ginkgo the same as the Chinese herb Bai Guo?
They come from the same tree but are not the same medicine. Bai Guo is the ginkgo seed, a longstanding Chinese herb used mainly to calm wheezing and for the Lung, and the raw seed is treated as toxic and dosed carefully. The Western supplement is an extract of the leaf, a much more recent use, aimed at the brain and circulation. Same plant, different part, different purpose.
Explore Related
Other pages this one connects to, by the evidence they share, the outcomes they touch, and the ground they cover.
All 10 sources on this page independently checked and cross-referenced.
Thomas Dehli, Founder & Editor, Sacred Lotus
Sacred Lotus has published Chinese medicine reference material since 2001. Integrative pages are held to the same standard as the herb and formula library: cite the source, grade the claim at its real strength, and say where the research has not looked. This page is educational and it is not medical advice. Last reviewed and updated August 9, 2026.
Evidence strength
How confidently the research supports a claim. Strength describes the evidence, not our endorsement.