Skip to content
View commentaries on this research

This is a plain English summary of an original research article

In people with some types of severe, drug-resistant epilepsy, adding cannabidiol to their treatment may reduce seizure frequency and improve quality of life compared with a placebo. The likelihood of being free from seizures for more than a year was still low, about 8%. However, an additional 12% of people had serious adverse effects with cannabidiol.

These findings come from a systematic review, which included six trials in 555 patients. Most were children and adolescents with rare forms of epilepsy, and findings may not apply to other forms of the condition. The included trials were poorly reported and show some bias. Further, more reliable research is needed and may produce different results.

Cannabidiol does not currently have approval for medicinal use in the UK for this purpose and will require more independent research before it is used with any confidence.

Why was this study needed?

Between 70 and 80% of people with epilepsy can control their seizures using antiepileptic drugs. However, the remaining 20 to 30% of people do not respond completely to these drugs and require an alternative.

There has been widespread interest from the public and the media in the medical use of cannabis and its active components (called cannabinoids). The medicinal grade cannabinoid studied in the main trials of this review do not have hallucinogenic effects. Laboratory and animal studies have suggested that cannabinoids might reduce epileptic seizures, and they have shown promise in some studies in people with severe epilepsy.

However, there has been concern about the quality of these studies, and the applicability of their findings to less severe forms of epilepsy. This review aimed to bring together the evidence on the safety and efficacy of cannabinoids as an add-on treatment for drug-resistant epilepsy.

What did this study do?

This systematic review looked at cannabinoids as add-on (adjunctive) treatments for drug-resistant epilepsy. It included six randomised controlled trials with 555 participants and 30 observational studies with 2,865 participants.

All of the trials compared a medicinal quality cannabinoid, cannabidiol, versus placebo. The most recent, larger randomised controlled trials used a dose of 10 or 20mg/kg/day over 14 weeks; earlier trials used 100mg two to three times a day over four to 26 weeks. The average age of trial participants was 16.3 years, and most had severe forms of epilepsy that start in childhood (Lennox-Gastaut or Dravet syndromes).

Only one trial was at low risk of bias (a phase III trial in Lennox-Gastaut syndrome). The remainder were at high risk (one trial), or risk of bias could not be judged due to very poor reporting. The observational studies were mostly at serious or critical risk of bias, and those results have not been included here. The trials are mainly funded by one manufacturer. The use of this drug, based on this evidence, should be very cautious.

What did it find?

  • Cannabidiol increased the likelihood of becoming seizure free (7.8% with cannabidiol vs 0.7% with placebo; relative risk [RR] 6.17, 95% confidence interval [CI] 1.50 to 25.32; three trials, 306 participants).
  • The likelihood of reducing seizure frequency by 50% or more was moderately increased (43.5% with cannabidiol vs 25.0% with placebo; RR 1.74, 95% CI 1.24 to 2.43; two trials, 291 participants).
  • Parents or carers reported an improvement in their child’s overall quality of life (59.8% with cannabidiol vs 34.5% with placebo; RR 1.73, 95% CI 1.33 to 2.26, two trials, 274 participants).
  • Any adverse events (88.4% with cannabidiol vs 69.7% with placebo; RR 1.24, 95% CI 1.13 to 1.36; 5 trials, 531 participants), serious adverse events (18.9% with cannabidiol vs 6.8% with placebo; RR 2.55, 95% CI 1.48 to 4.38; 4 trials, 516 participants), and treatment-related serious adverse events (6.8% with cannabidiol vs 0.6% with placebo; RR 5.93, 95% CI 1.38 to 25.46; 3 trials, 396 participants). The adverse events which increased included drowsiness, fatigue, diarrhoea, and elevated liver enzyme levels.

What does current guidance say on this issue?

Cannabidiol is not licensed for the treatment of epilepsy in the UK, and it is not covered by UK guidelines.

There are relatively limited options currently available for people whose epilepsy has not responded to existing drug treatments. According to NICE guidance from 2018, for children and young people, these options include a specific diet low in carbohydrates and high in fat (a ketogenic diet), as well as vagal nerve stimulation. The latter involves implantation of a device to provide ongoing cycles of electrical stimulation of the vagal nerve in the neck, to suppress abnormal electrical brain activity associated with seizures. Deep brain stimulation may also be an option in special circumstances.

What are the implications?

This review suggests some benefits of cannabidiol in a selected population with drug-resistant epilepsy, though bias might explain some or all of these results. As yet, however, cannabidiol is not licensed for this use in the UK.

The drug has been given orphan designation by the European Medicines Agency for the treatment of several severe and rare forms of epilepsy that have not responded to other treatments. This means that incentives are in place to encourage its development for these indications, and a pharmaceutical company has applied to do this.

Citation and Funding

Stockings E, Zagic D, Campbell G, et al. Evidence for cannabis and cannabinoids for epilepsy: a systematic review of controlled and observational evidence. J Neurol Neurosurg Psychiatry. 2018;89(7):741-53.

The review was funded by the Commonwealth Department of Health, the New South Wales Government Centre for Medicinal Cannabis Research and Innovation, the Victorian Department of Health and Human Services and the Queensland Department of Health in Australia.


EMA. Orphan designation. London: European Medicines Agency. 2018.

NICE. Deep brain stimulation for refractory epilepsy. IPG416. London: National Institute for Health and Care Excellence; 2012.

NICE. Epilepsies: diagnosis and management. CG137. London: National Institute for Health and Care Excellence; 2018.

Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre

  • Share via:
  • Print article


Drug-resistant epilepsy. Where adequate trials of two tolerated and appropriately chosen single drug or combination antiepileptic drug regimens have failed to stop seizures completely.

Seizure freedom. Defined in the review as having no seizures of any type for either 12 months or three times the longest seizure-free interval before the trial, whichever was the longest.


Back to top