Category: Herbs & Ingredients
Ginkgo contains a compound that blocks platelet-activating factor, giving it a genuine mechanism for affecting how blood clots. Bleeding case reports raised alarm for decades, but controlled trials on warfarin and platelet function have often found nothing. Here is what the published literature actually shows.
Category: Herbs and Ingredients | Reading time: ~13 min | Level: Intermediate
Search "ginkgo and blood thinners" and you will find two confident answers pointing in opposite directions. One camp says ginkgo is a dangerous blood thinner that should never be combined with warfarin or aspirin, citing case reports of spontaneous bleeding. The other says that is overblown, pointing to clinical trials that found no measurable effect at all. Both sides are quoting real published research. That is the part most articles skip over.
Ginkgo is not a case where the evidence is simply thin and everyone is guessing. It is a case where good quality studies genuinely disagree, and the honest answer sits in the tension between them rather than on either side. If you take warfarin, a DOAC, aspirin or clopidogrel, or you are due for surgery, that tension is exactly what you need to understand before deciding what to do with a bottle of ginkgo capsules.
One thing to be clear about from the start. The interaction described here is drawn from the published clinical and regulatory literature. It is not a claim manufactured by any single product or platform, and the primary studies are named below so you can look them up yourself.
Ginkgo biloba leaf extract contains two groups of active compounds: flavonoid glycosides and terpene lactones, the best known of which are the ginkgolides and bilobalide [1][2]. Ginkgolide B is the compound at the centre of this interaction. It blocks platelet-activating factor, a signalling molecule platelets use to clump together during normal clotting [3]. In simple terms, that gives ginkgo a real, mechanistically plausible antiplatelet action, the same broad category of effect as low dose aspirin, just through a different pathway.
That mechanism is the reason ginkgo appears on almost every list of supplements to check before starting a blood thinner. A compound that can, in principle, make platelets stick together less readily is a compound worth asking about if you are already taking a medicine whose entire job is to reduce clotting. Whether that theoretical mechanism produces a meaningful bleeding risk in practice is where the evidence gets genuinely mixed, which is the subject of the rest of this article.
The original alarm came from individual case reports. A systematic review published in 2005 collected 15 published cases of bleeding temporally associated with ginkgo use, including eight episodes of intracranial haemorrhage, bleeding inside the skull [3]. The authors concluded that the pattern suggested "a possible causal association" between ginkgo and bleeding events and recommended that people with known bleeding risks be counselled about it. That conclusion should be read with its own caveat attached: 13 of the 15 cases had other bleeding risk factors present, such as being on warfarin or aspirin already, or having an underlying condition that raises bleeding risk on its own [3]. Case reports cannot prove causation, but a cluster of severe events, several of them inside the skull, is not nothing either.
Set against those case reports is a body of controlled research that has generally not found what the case reports implied. A meta-analysis of 18 randomised controlled trials, covering 1,985 adults, most with dementia, peripheral artery disease or diabetes, found that standardised ginkgo extract did not indicate a higher bleeding risk, with no clinically relevant effect on platelet aggregation or on standard clotting tests such as aPTT and prothrombin time [4]. A separate review of the controlled evidence reached a similar conclusion, stating that controlled studies consistently show ginkgo does not significantly affect blood clotting nor compromise the safety of aspirin or warfarin taken alongside it [5].
The most direct test of the warfarin question is a small but purpose built crossover trial. Twelve healthy volunteers took a single dose of warfarin alone, and again after a week of pretreatment with ginkgo at a standard recommended dose. The result: ginkgo did not significantly affect warfarin's pharmacokinetics, its pharmacodynamics, or the volunteers' clotting status [6]. That is a genuinely reassuring, purpose built result, and it is one of the reasons careful reviewers describe the interaction as theoretical rather than proven.
If the story stopped there, the honest conclusion would lean towards reassurance. It does not stop there. A 2015 study used a large Veterans Administration medical record database to compare bleeding rates in warfarin patients who were and were not also taking ginkgo. It found that concurrent ginkgo use was associated with a significantly higher risk of a bleeding event, with a hazard ratio of 1.38, meaning roughly a 38% higher rate of bleeding events in the ginkgo group [7]. This was one of the first attempts to test a suspected herb-drug interaction using real world medical record data at scale, and unlike the small crossover trial, it captured years of ordinary use in a large population rather than a single controlled week.
That study has its own limits. It is retrospective and observational, not a randomised trial, so it cannot rule out that people who chose to take ginkgo alongside warfarin differed in other ways, such as age, other medications or underlying health, that independently raised their bleeding risk. A more recent 2025 pharmacovigilance analysis of over 2,600 prescriptions found something in between: the overall severity of a flagged ginkgo drug interaction did not statistically correlate with bleeding risk, but specific co-prescribed drugs, including clopidogrel, aspirin, celecoxib and nifedipine, were significantly associated with bleeding and abnormal coagulation test results [8]. Read together, the observational evidence has not gone away just because the small controlled trials came back null.
The defensible summary is this. Small, tightly controlled trials measuring platelet function and clotting tests directly have generally found no significant effect. Real world observational data and case reports point towards a genuine, if inconsistent, bleeding signal, concentrated in people already on another drug that raises bleeding risk. Regulators have not waited for the disagreement to resolve. The European Medicines Agency's herbal monograph formally lists organ bleeding, including cerebral and gastrointestinal haemorrhage, as an undesirable effect of ginkgo, advises stopping it three to four days before surgery, and states that anyone with a pathologically increased bleeding tendency who is also on an anticoagulant or antiplatelet medicine should only use ginkgo after consulting a doctor [2]. That is the position of a body reading the same mixed evidence base described here, and it lands on caution rather than either extreme.
| Combination | What the evidence shows | Practical caution | |---|---|---| | Ginkgo with warfarin | A dedicated crossover trial and controlled meta-analyses found no significant change in INR or clotting tests [4][6]. A large retrospective database study found a higher rate of bleeding events in warfarin patients also taking ginkgo [7]. | Disclose to your anticoagulation clinic; do not start or stop without them knowing; watch for bruising or bleeding. | | Ginkgo with aspirin | Mechanistically additive, since both act on platelets. Controlled reviews found no significant added bleeding effect [5], but pharmacovigilance data flag aspirin as a commonly co-prescribed interacting drug [8]. | Mention it to whoever prescribes your aspirin, particularly if aspirin is for secondary prevention after a cardiac event. | | Ginkgo with clopidogrel | Same additive antiplatelet logic as aspirin. No dedicated randomised bleeding-outcome trial exists for this specific pair. Pharmacovigilance data list clopidogrel among the most frequently flagged interacting drugs [8]. | Caution and disclosure, especially around any planned procedure. | | Ginkgo with a DOAC (apixaban, rivaroxaban, dabigatran, edoxaban) | Direct bleeding-outcome data are limited. The EMA monograph specifically flags a possible interaction with dabigatran through a gut transport protein ginkgo may inhibit [2]. | Treat as an open question; ask a prescriber or pharmacist rather than assuming safety by omission. | | Ginkgo before surgery | EMA advises stopping three to four days beforehand as a bleeding precaution [2], regardless of which trial result you find more persuasive. | Stop as advised and tell your surgical team you take it. |
If you are on any blood thinning medicine, warfarin, a DOAC, aspirin or clopidogrel, tell whoever prescribes it that you take ginkgo, or are thinking about starting it. This is not a demand to stop; it is a request to be seen. The controlled trial evidence is genuinely reassuring for the average person, but the case reports and the large retrospective study describe real bleeding events that happened to real people, and you cannot know in advance whether you resemble the reassuring average or the unlucky outlier.
If you and your prescriber decide ginkgo is reasonable to try, watch for the practical signs of a bleeding problem: bruising that appears without explanation, nosebleeds that will not stop quickly, blood in urine or stool, unusually heavy periods, or a sudden severe headache, which needs emergency assessment rather than a wait and see approach. If you are on warfarin specifically, a repeat INR after starting ginkgo is a reasonable, low effort way to check nothing has shifted, even though the trial evidence suggests it likely will not.
If a procedure or surgery is on the calendar, stop ginkgo three to four days beforehand, in line with the regulatory advice, and mention it on your pre-operative checklist the same way you would mention any other supplement.
Ginkgo has a real, plausible mechanism for affecting how platelets behave, and that mechanism is not in dispute. What is genuinely unresolved is how much that mechanism matters in practice. Small controlled trials, including one built specifically to test warfarin, have mostly found no significant effect on bleeding, platelet function or INR. Case reports of serious, sometimes intracranial bleeding exist, and a large retrospective study of warfarin patients found a higher bleeding rate associated with ginkgo use. Neither side of that evidence cancels the other out.
The responsible reading is mechanistically plausible, flagged by real world and case report data, not robustly confirmed in the tightest controlled trials. That combination supports prudent caution, particularly for anyone already on an anticoagulant or antiplatelet medicine, anyone with a bleeding disorder, and anyone approaching surgery, without supporting a claim that ginkgo is a proven, potent blood thinner in its own right. This is exactly the kind of interaction where the honest answer is "talk to your prescriber," not because the article is dodging the question, but because the published evidence itself has not settled it.
Ginkgo sits in the PlantRx catalogue as a cognitive support ingredient, and its bleeding caution is presented alongside that use rather than left out of the picture. Structure and function only: ginkgo is offered for circulatory and cognitive support; it is not a blood thinning medicine, and it must not be used to self-manage a clotting condition or substitute for anticoagulant therapy.
If you are unsure whether your medication list makes ginkgo a poor fit, Remy, the PlantRx assistant, can surface the published bleeding and interaction cautions relevant to your situation and point you back to your prescriber for the calls that depend on your full medical history.
Ginkgo's bleeding caution is the headline here, but a few other points belong in the same decision. Raw or roasted ginkgo seeds are toxic and are an entirely separate hazard from the standardised leaf extract sold as a supplement; they should never be eaten [1]. Ginkgo carries a seizure caution in people with epilepsy, since the European monograph notes that further seizures cannot be excluded in this group [2]. It is not recommended in pregnancy, where regulators list it as a contraindication because of the same platelet effect discussed above, or during breastfeeding [2]. The liver safety profile is reassuring: the US LiverTox database gives ginkgo its lowest likelihood score for causing liver injury, with no clearly attributed cases identified [9]. Common side effects include headache, dizziness and mild gastrointestinal upset [1][2].
Pregnant, breastfeeding, or on medication? Check with a healthcare professional first.
1. National Center for Complementary and Integrative Health (NCCIH). Ginkgo: Usefulness and Safety. Government evidence and safety review. nccih.nih.gov/health/ginkgo. 2. European Medicines Agency, Committee on Herbal Medicinal Products (HMPC). European Union herbal monograph on Ginkgo biloba L., folium. EMA/HMPC/321097/2012, Final, adopted 28 January 2015. Regulatory herbal monograph. 3. Bent S, Goldberg H, Padula A, Avins AL (2005). Spontaneous bleeding associated with Ginkgo biloba: a case report and systematic review of the literature. Journal of General Internal Medicine. Systematic review of 15 published case reports, including 8 of intracranial haemorrhage; 13 of 15 cases had other bleeding risk factors. 4. Kellermann AJ, Kloft C (2011). Is there a risk of bleeding associated with standardized Ginkgo biloba extract therapy? A systematic review and meta-analysis. Pharmacotherapy 31(5):490 to 502. Meta-analysis of 18 RCTs, 1,985 adults. 5. Bone KM (2008). Potential interaction of Ginkgo biloba leaf with antiplatelet or anticoagulant drugs: what is the evidence? Molecular Nutrition & Food Research 52:764 to 71. Narrative review of controlled studies. 6. Jiang X, Williams KM, Liauw WS, et al. (2005). Effect of ginkgo and ginger on the pharmacokinetics and pharmacodynamics of warfarin in healthy subjects. British Journal of Clinical Pharmacology 59(4):425 to 32. Open label, three way crossover RCT, 12 healthy volunteers. 7. Stoddard GJ, Archer M, et al. (2015). Ginkgo and Warfarin Interaction in a Large Veterans Administration Population. AMIA Annual Symposium Proceedings 2015:1174 to 1183. Retrospective database study using electronic health record text mining; hazard ratio 1.38 (95% CI 1.20 to 1.58) for bleeding events. 8. Mai NTQ, Hieu NV, Ngan TTT, et al. (2025). Impact of Ginkgo biloba drug interactions on bleeding risk and coagulation profiles: a comprehensive analysis. PLOS ONE. Retrospective observational pharmacovigilance study, 2,647 prescriptions. 9. National Institute of Diabetes and Digestive and Kidney Diseases, LiverTox. Ginkgo (NBK548847). Government safety monograph, likelihood score E (unlikely cause of clinically apparent liver injury).
Not without telling the clinic that manages your warfarin first. Ginkgo has a plausible mechanism for affecting platelets, and case reports of bleeding exist alongside a large retrospective study linking ginkgo to more bleeding events in warfarin patients. Set against that, several controlled trials, including one built specifically to test ginkgo with warfarin, found no measurable change in INR or clotting. The honest position is caution and disclosure, not a flat yes or no.
It has a real mechanism for doing so. Ginkgo contains ginkgolide B, which blocks platelet-activating factor, a signal platelets use to clump together. That gives ginkgo a genuine antiplatelet action in the laboratory. Whether that translates into a meaningful bleeding risk in most people taking a standard dose is the part the evidence disagrees on, with several controlled studies finding no measurable effect on platelet function or standard clotting tests.
There is no dedicated large trial answering this directly, which is itself a gap worth knowing about. Because ginkgo and both drugs act on platelets, an additive bleeding risk is mechanistically reasonable, and pharmacovigilance data list aspirin and clopidogrel among the medicines most often flagged alongside ginkgo. The sensible approach is to mention ginkgo to whoever prescribes your aspirin or clopidogrel rather than assuming it is inert.
Direct clinical bleeding data for ginkgo with the newer anticoagulants are limited. The European regulator's herbal monograph specifically flags a possible interaction with dabigatran through a different mechanism, ginkgo's effect on a gut transport protein, and advises caution. For apixaban and rivaroxaban the data are thinner still. Treat any DOAC and ginkgo combination as a prescriber conversation, not a self-directed decision.
The European herbal monograph advises stopping ginkgo three to four days before a planned procedure as a precaution against bleeding, and this is sound practice regardless of which side of the evidence debate you find more convincing. Tell your surgical team you take it, the same as you would any supplement with a theoretical bleeding link.
Because the controlled trial evidence and the case report evidence point in somewhat different directions, and that is the honest state of the science. Randomised trials and meta-analyses measuring platelet function, clotting tests and warfarin levels directly have mostly found no significant change. Case reports and one large observational database study point the other way. Both kinds of evidence are real; they just answer slightly different questions, and neither should be ignored.
Unusual or easy bruising, nosebleeds that are hard to stop, blood in urine or stool, unusually heavy periods, or any sudden severe headache, which can be a sign of bleeding inside the skull and needs emergency care. Anyone on a blood thinner who also takes ginkgo should know these signs regardless of which way the statistics point for the average person.
A stable INR means your warfarin dose and your usual habits are in balance, and any new variable, including ginkgo, is worth flagging even if the average trial found little effect. The controlled data are reassuring for most people, but the case reports and the retrospective bleeding signal exist precisely because average results do not describe every individual. A quick call to your anticoagulation clinic before starting costs little and closes the gap between what a average trial shows and what applies to you.
The case reports are older, but the underlying mechanism, ginkgolide B's action on platelet-activating factor, has not changed, and a large retrospective study published in 2015 using modern medical records still found more bleeding events in warfarin patients also taking ginkgo. Age of a case report does not by itself weaken a plausible mechanism. It is one more reason caution has held up rather than faded.
That is worth raising directly with your prescriber or pharmacist rather than swapping one unverified option for another, because most botanicals marketed for memory carry their own evidence gaps and some carry their own bleeding or interaction cautions. The more useful question for someone on a blood thinner is often not which supplement to add, but whether the underlying memory concern needs a clinical work up first.