Prescription CBD and Warfarin: An Unexpected Interaction in Medical Case Reports

Medical cannabis collection · Expanded 11 October 2026

Medical cannabis case studies · Literature checked 2 October 2026

A medicine introduced for epilepsy can unexpectedly change the management of an unrelated heart or clotting condition. Published cases involving prescription cannabidiol and warfarin illustrate this clearly. The signal was a change in a blood test, not necessarily an obvious new symptom. [1] [2]

Evidence at a glance: medically supervised cases suggesting a clinically important interaction, with another medical-cannabis report showing stable results. The size of the effect cannot be predicted from these cases alone.

On this page

1. The first epilepsy case: a previously stable INR rose
2. A second prescription-CBD report
3. The contrasting case matters too
4. What INR tells us
5. What the cases establish, and what they cannot
6. What this means for a medication review
7. Bring a medicine record, not just the words “CBD oil”
8. References

The first epilepsy case: a previously stable INR rose

Grayson and colleagues described a 44-year-old man with a mechanical mitral valve and treatment-resistant epilepsy. His warfarin control had been stable before he entered a compassionate-use programme for pharmaceutical CBD. During CBD titration his INR rose, reaching 6.86 at one visit. His clinician adjusted warfarin, eventually reducing the dose by about 30%. No bleeding complication was reported. The authors proposed interference with drug metabolism as a possible explanation and recommended INR monitoring when cannabinoids are introduced. [1]

A second prescription-CBD report

A 2020 report described a 46-year-old man receiving warfarin for deep-vein thrombosis who started the prescription CBD medicine Epidiolex for intractable epilepsy. Close monitoring accompanied initiation and titration. Maintaining the target INR required a warfarin reduction of nearly 20%. This supports the earlier interaction signal, but the reported percentage is not an adjustment rule for other patients. [2]

The contrasting case matters too

Thomas and colleagues reported an 85-year-old man using a relatively low-dose THC–CBD oromucosal product for chronic pain while receiving warfarin. His INR remained broadly stable. The authors suggested that exposure and route might help explain why their findings differed from other reports. One uneventful course does not establish a universally safe cannabinoid dose, just as two interactions do not mean every patient will respond identically. [3]

What INR tells us

Warfarin reduces the blood’s ability to form clots. Its effect is monitored using the international normalised ratio, or INR, interpreted against an individual target. Excessive anticoagulation increases bleeding risk. Regular blood testing is therefore part of treatment even when a person feels well. [4]

The unusual feature of these reports is the separation between treatment purpose and treatment consequence. A person could be concentrating on seizure control while the clinically important new change appears in anticoagulation monitoring. Looking only for sedation, dizziness or a noticeable “cannabis effect” would miss that possibility.

What the cases establish, and what they cannot

Repeated observations are a reason to take an interaction seriously. They do not provide an incidence rate or a dependable formula linking a particular CBD amount to a particular warfarin change. In an individual patient, illness, other medicines, dietary changes and adherence also require review before an abnormal result is attributed to one cause.

It is also important to preserve the distinction between a laboratory signal and an observed clinical injury. A high INR is clinically significant, but it should not be rewritten as a haemorrhage if the report did not describe one. Conversely, the absence of bleeding during careful management does not make the interaction harmless.

The contrasting case is useful precisely because it prevents overstatement. It encourages questions about the actual product, exposure and clinical setting. It cannot be used to reassure another person that their combination is safe without monitoring.

What this means for a medication review

Anyone prescribed warfarin should tell the anticoagulation team about a proposed cannabinoid medicine and any subsequent product or dose change. NHS Specialist Pharmacy Service guidance recommends additional INR testing when a potentially interacting medicine is prescribed. The clinical team should arrange the timing and make any dose adjustment. Do not copy the 20% or 30% reductions from these reports or stop warfarin yourself. [5]

A useful medication list records the product name, THC and CBD content, route, frequency, recent changes and who prescribed it. This allows the cannabis prescriber, pharmacist and anticoagulation service to work from the same information rather than from the vague description “CBD oil”.

Severe or recurrent bleeding, blood in vomit, black stools, or a significant head injury while anticoagulated require urgent medical assessment. These are general anticoagulant precautions, not a prediction that a CBD interaction will cause them. [4]

This article explains published clinical evidence. Individual cases cannot predict personal treatment response. Decisions about prescription medicines belong with the treating clinician.

Bring a medicine record, not just the words “CBD oil”

A normal result during one combination is reassuring for that moment; it is not a guarantee after a new formulation or exposure. Conversely, an abnormal INR needs clinical interpretation rather than an automatic assumption that CBD is the only cause. Monitoring allows the teams to respond to the actual result, without copying adjustments from another person’s case report.

References

External links open in a new tab. Journal links lead to the original report, indexed abstract or official guidance.

1. Grayson L, Vines B, Nichol K, Szaflarski JP. An interaction between warfarin and cannabidiol, a case report. Epilepsy & Behavior Case Reports. 2018;9:10–11 (published online 2017). doi:10.1016/j.ebcr.2017.10.001.
2. Cortopassi J. Warfarin dose adjustment required after cannabidiol initiation and titration. American Journal of Health-System Pharmacy. 2020;77:1846–1851. doi:10.1093/ajhp/zxaa268.
3. Thomas TF, et al. Case report: Medical cannabis—warfarin drug-drug interaction. Journal of Cannabis Research. 2022;4:6. doi:10.1186/s42238-021-00112-x.
4. NHS. Anticoagulant medicines: side effects. Accessed 2 October 2026.
5. NHS Specialist Pharmacy Service. Warfarin monitoring. Accessed 2 October 2026.

Image sources

• Harbin, with revisions by Mykhal, via Wikimedia Commons. Public-domain chemical structure.