Cannabidiol Has a Proven Role in Treating Seizures in Certain Severe Epilepsy Syndromes
Evidence for cannabis-based epilepsy treatments has become more precise since a 2017 review found that the data were too limited to support firm conclusions. Research now shows that pharmaceutical-grade cannabidiol (CBD) can reduce seizures in people with two rare, treatment-resistant epilepsy syndromes—Dravet syndrome and Lennox–Gastaut syndrome. Those findings do not establish that marijuana, THC-containing oils, or over-the-counter CBD products are effective treatments for epilepsy in general.
The distinction matters. Cannabis contains many compounds, including CBD and tetrahydrocannabinol (THC), the psychoactive component associated with a “high.” The strongest clinical evidence concerns a standardized, purified oral CBD solution used as an add-on to conventional antiseizure medication—not smoked cannabis or unregulated cannabis oils.
How the evidence changed
The 2017 report from the National Academies of Sciences, Engineering, and Medicine reviewed thousands of studies on cannabis and cannabinoids. For epilepsy, it concluded that there was insufficient evidence at the time to determine whether cannabinoids were effective. That assessment reflected the shortage of high-quality controlled trials, rather than evidence that cannabis had been shown not to work.
Subsequent randomized trials provided stronger evidence for purified CBD in specific syndromes. In a 2017 trial published in the New England Journal of Medicine, 120 children and young adults with drug-resistant Dravet syndrome received either oral CBD or placebo in addition to their usual medications. Median monthly convulsive seizures fell from 12.4 to 5.9 in the CBD group, compared with a decline from 14.9 to 14.1 among those receiving placebo. CBD was associated with more adverse effects, including diarrhea, vomiting, fatigue, sleepiness, fever, and abnormal liver-function tests.
A separate 2018 randomized trial in Lennox–Gastaut syndrome tested two CBD doses in children and adults with frequent drop seizures. The median reduction in drop-seizure frequency was 41.9% with 20 milligrams of CBD per kilogram of body weight per day and 37.2% with 10 milligrams, compared with 17.2% with placebo. Diarrhea, reduced appetite, sleepiness, and elevated liver enzymes were among the reported risks.
These studies were not tests of cannabis broadly. They evaluated a defined CBD formulation, given at carefully measured doses, alongside other antiseizure drugs. The results therefore cannot automatically be applied to dispensary products, edible cannabis, high-THC preparations, or products labeled simply as “CBD oil.”
What about THC-containing cannabis oils?
A small Canadian open-label study of a CBD-dominant CBD/THC oil followed 20 children with Dravet syndrome for 20 weeks. The researchers reported fewer motor seizures and improvements in some electroencephalogram and quality-of-life measures. However, the study had no placebo group and was too small to determine whether the oil itself caused the improvements. Results from an uncontrolled study can be influenced by reporting effects, natural fluctuations in seizure frequency, changes in other treatments, and participants’ expectations.
For that reason, the findings should not be treated as equivalent to the randomized trials of purified CBD. The NHS England guidance has similarly distinguished between licensed, pharmaceutical-grade CBD and other cannabis-based products. It notes that robust evidence in childhood epilepsy relates to purified CBD, while evidence for THC-containing products remains limited and concerns persist about THC exposure during brain development.
Current clinical significance
The research has nevertheless produced a meaningful change in clinical practice. In the United States, the Food and Drug Administration has approved Epidiolex, a purified CBD medicine, for seizures associated with Dravet syndrome, Lennox–Gastaut syndrome, and tuberous sclerosis complex in patients aged one year and older. Approval applies to that specific prescription product and its studied uses; it is not an endorsement of cannabis products generally.
CBD is not a cure, and many patients continue to have seizures despite treatment. It can also interact with other medicines and may require monitoring of liver enzymes, particularly when used with certain antiseizure drugs. Treatment decisions should therefore be made with an epilepsy specialist rather than by substituting nonprescription cannabis products for established medication.
The overall conclusion is more nuanced than either broad enthusiasm or outright dismissal. Early evidence was insufficient because rigorous studies were lacking. Later controlled trials demonstrated that adjunctive pharmaceutical-grade CBD can reduce seizure frequency in particular severe epilepsy syndromes, while also causing adverse effects. The remaining research challenge is to determine which patients benefit, how long benefits persist, how CBD interacts with other medicines, and whether any THC-containing or whole-plant preparations offer benefits that outweigh their additional uncertainties and risks.
