Patients taking warfarin who also use ginkgo biloba supplements face a measurably higher chance of dangerous bleeding events, according to a body of clinical evidence spanning case reports, observational data, and systematic reviews. A large observational study of Veterans Administration health records found that concurrent use of ginkgo and warfarin was linked to a hazard ratio of 1.38 for bleeding adverse events. That signal, combined with documented cases of intracerebral hemorrhage and surgical bleeding in ginkgo users, has prompted the National Center for Complementary and Integrative Health to warn that ginkgo may increase bleeding risk in people on anticoagulants.
Why the ginkgo–warfarin bleeding signal matters right now
Warfarin remains one of the most widely prescribed blood thinners in the United States, and ginkgo biloba is among the most popular herbal supplements sold over the counter. The overlap between these two products creates a specific clinical hazard: ginkgo contains compounds that affect platelet aggregation, and warfarin already suppresses the clotting cascade. When both are taken together, the combined pressure on hemostasis can tip patients toward uncontrolled bleeding.
The risk is especially acute for older adults. Elderly patients often take multiple medications, and their baseline bleeding risk is already elevated by age-related changes in liver metabolism and vascular fragility. Most controlled pharmacokinetic studies of ginkgo and warfarin have been conducted in younger, healthy volunteers. A controlled study in healthy subjects assessed ginkgo’s effects on warfarin pharmacokinetics and pharmacodynamics, but those results may not capture what happens in a 75-year-old patient on five other medications with impaired kidney function. The gap between trial conditions and real-world patient profiles is where the danger concentrates.
The federal integrative health agency advises patients to tell their healthcare providers about any supplements they use, specifically flagging ginkgo as a concern for those on anticoagulants. Yet supplement use often goes undisclosed during clinical visits, leaving prescribers unaware of the interaction risk. Patients may view herbal products as “natural” and therefore harmless, or simply forget to mention them when listing medications. In a drug such as warfarin, where small changes in bleeding risk can have life-threatening consequences, that communication gap becomes critical.
Case reports and VA data quantify the bleeding hazard
The strongest real-world signal comes from a large observational study that mined Veterans Administration electronic health records. That analysis, published in PLOS ONE, found that concomitant ginkgo and warfarin exposure was associated with a hazard ratio of 1.38 for bleeding adverse events, with a 95% confidence interval of 1.20 to 1.58. In practical terms, veterans who used both ginkgo and warfarin experienced roughly 38% more bleeding events than those on warfarin alone, after adjusting for other variables. The size of the VA dataset and the consistency of the confidence interval give this finding substantial weight and argue against it being a chance signal.
Individual case reports add clinical texture to that statistical picture. A systematic review of published cases examined bleeding episodes temporally associated with ginkgo and cataloged several serious outcomes. Among them was a case of intracerebral hemorrhage in a patient on long-term warfarin who was also taking ginkgo, a sentinel event that has been widely cited in pharmacology literature. In that case, no alternative clear cause for the brain bleed was identified, strengthening the suspicion of an interaction. Separately, a case report in a surgical journal documented postoperative bleeding in a patient who had undergone laparoscopic cholecystectomy and had been using ginkgo without disclosing it to the surgical team. The patient experienced more bleeding than expected, prompting a retrospective review that uncovered the supplement use.
A broader systematic review of warfarin interactions with herbal products, including ginkgo, synthesized evidence across pharmacokinetic and pharmacodynamic studies, clinical trials, and case reports. That review mapped conflicting findings: some controlled data showed no meaningful change in warfarin levels or coagulation markers, while case-level evidence pointed to clinically significant bleeding. The discrepancy suggests that standard pharmacokinetic endpoints may not fully capture the interaction’s danger in vulnerable patients, particularly those with comorbidities, fluctuating nutritional status, or polypharmacy.
Conflicting trial data and the limits of current evidence
The evidence is not unanimous. A meta-analysis of 18 randomized controlled trials, covering approximately 1,985 participants, found no indication of higher bleeding risk with standardized ginkgo extracts based on hemostasis-related outcomes. That finding, according to a review published in Pharmacotherapy, represents the strongest controlled-trial counterpoint to the bleeding signal seen in observational and case-report data.
The contradiction has a likely explanation rooted in study design. Randomized trials typically enroll relatively healthy participants, exclude those on anticoagulants, and use standardized ginkgo preparations at fixed doses for limited periods. The VA observational study, by contrast, captured patients in routine clinical care, many of whom were older, sicker, and taking warfarin alongside ginkgo for extended stretches. The populations are fundamentally different, and the risks that emerge in one setting may remain invisible in the other.
In addition, trial endpoints often focus on laboratory measures such as prothrombin time, international normalized ratio (INR), or platelet function tests. These markers are important but may not fully predict real-world outcomes like gastrointestinal bleeding, intracranial hemorrhage, or postoperative oozing. Case reports, though anecdotal and vulnerable to bias, directly describe such outcomes. When multiple independent reports converge on similar patterns-warfarin plus ginkgo followed by unexpected bleeding-the signal warrants attention even if mechanistic studies are equivocal.
Another limitation is variability in ginkgo products. Over-the-counter supplements differ widely in composition, standardization, and potential contaminants. Clinical trials usually employ a specific, well-characterized extract, whereas real-world patients may take products of uncertain potency or purity. This variability could magnify bleeding risk in practice compared with the controlled environment of a research study.
Practical implications for clinicians and patients
For clinicians managing patients on warfarin, the emerging picture argues for a cautious, proactive approach. Asking specifically about herbal supplements-and documenting ginkgo use as carefully as any prescription drug-should be part of routine anticoagulation visits. If a patient on warfarin is taking ginkgo, prescribers should discuss the potential for increased bleeding risk, consider closer INR monitoring, and weigh whether the perceived benefits of ginkgo justify that added risk.
Patients, for their part, should recognize that “natural” does not mean “risk-free.” Ginkgo has pharmacologically active constituents that can interact with prescription drugs. People taking warfarin, other anticoagulants, or antiplatelet agents should not start ginkgo on their own. Any decision to use the supplement should involve a conversation with their healthcare team, including pharmacists who can help review interaction profiles.
In surgical settings, preoperative assessments should include direct questions about herbal products. As the postoperative bleeding case illustrates, undisclosed ginkgo use can complicate both intraoperative hemostasis and recovery. Surgeons and anesthesiologists may reasonably advise patients on warfarin to avoid ginkgo in the weeks leading up to elective procedures, in line with broader recommendations on holding agents that affect coagulation.
Where the evidence leaves us
Taken together, the VA observational data, case reports, and systematic reviews do not prove that ginkgo inevitably causes dangerous bleeding in every warfarin user. They do, however, establish a plausible, clinically meaningful risk signal in precisely the population most likely to be harmed by additional anticoagulant effects. The absence of excess bleeding in controlled trials conducted in healthier cohorts does not erase that concern.
Until more definitive research directly examines older, comorbid patients taking both agents, the most defensible stance is one of prudence. For many warfarin users, especially those at baseline high risk of hemorrhage, avoiding ginkgo or using it only under close medical supervision is a low-cost strategy to reduce the chance of a catastrophic bleed. In the balance between uncertain cognitive or circulatory benefits and a documented potential for harm, caution is a reasonable prescription.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
