Epilepsy and cannabidiol

Epilepsy is one of the few conditions where the support for a specific cannabinoid, not “cannabis” in general, has reached the level of regulatory recognition in New Zealand. This guide explains what that support is, how it fits into the legal prescribing pathway, and what it does not cover.

What the research actually shows

The strongest clinical backing for medicinal cannabis in epilepsy concerns pharmaceutical-grade cannabidiol (CBD), not THC-dominant preparations. In 2018, the New England Journal of Medicine published results from three randomised controlled trials of a purified CBD oral solution (Epidyolex) in people with Dravet syndrome and Lennox-Gastaut syndrome, two severe childhood-onset epilepsies. The trials found that adding CBD to existing anti-seizure medication reduced monthly seizure frequency by roughly 38-44% compared with placebo, depending on dose. The authors reported the medicine as generally well tolerated. The most common side effects were diarrhoea, decreased appetite, and drowsiness (Devinsky et al., 2018; Thiele et al., 2018).

Epidyolex (pharmaceutical CBD) is now registered with Medsafe for specific epilepsy syndromes and is the only cannabis-derived medicine with formal approval for epilepsy in New Zealand. The NZ Formulary lists it as an adjunctive treatment for seizures associated with Lennox-Gastaut syndrome or Dravet syndrome in people aged 2 years and older. That is a narrow, evidence-based indication. It is not a licence to treat all seizure types or all forms of epilepsy, and it does not mean that any CBD oil sold online is equivalent or safe.

How the pivotal trials were run

The strongest body of findings for CBD in epilepsy comes from three randomised, placebo-controlled trials that supported the approval of Epidyolex by the US FDA in 2018 and the European Medicines Agency in 2019. Those trials enrolled people with Lennox-Gastaut syndrome or Dravet syndrome who had not responded to other anti-seizure medicines. In the Lennox-Gastaut trials, adding pharmaceutical CBD at 10 mg/kg/day or 20 mg/kg/day reduced drop-seizure frequency by about 44% at the higher dose, compared with 27% for placebo (Devinsky et al., 2017; Thiele et al., 2018). In Dravet syndrome, CBD reduced convulsive seizure frequency by 39% versus 13% for placebo (Devinsky et al., 2017). The FDA approved Epidiolex in 2018 for Lennox-Gastaut syndrome and Dravet syndrome, and the European Medicines Agency followed in 2019. In New Zealand, Epidyolex is the registered medicine containing CBD for these syndromes. These are the only indications where the findings for CBD in epilepsy are considered robust.

The 2018 NEJM trial for Dravet syndrome was a randomised, double-blind, placebo-controlled phase 3 trial of Epidiolex. It showed that add-on CBD reduced monthly convulsive seizure frequency by 39%, from a median of 12.4 to 5.9, compared with a 13% reduction with placebo. The Lennox-Gastaut trial tested add-on therapy at two doses. The 20 mg/kg/day dose reduced monthly drop seizures by 41.9% versus 18.5% for placebo; the 10 mg/kg/day dose reduced them by 37.2%. A 41% reduction in drop-seizure frequency for Lennox-Gastaut syndrome was reported, compared with a 14% reduction for placebo. These trials were sponsored by GW Pharmaceuticals, the manufacturer of Epidyolex. They were not designed to test CBD as a monotherapy; participants continued their existing antiepileptic drugs.

Adverse effects in the 2018 NEJM trials included diarrhoea, decreased appetite, somnolence, and elevated liver enzymes. Elevated transaminases were more common in people taking valproate with higher CBD doses.

What the reviews and meta-analyses add

A 2018 systematic review of CBD in epilepsy (Lattanzi et al., published in Epilepsy & Behavior) found that, while CBD reduced seizure frequency in some treatment-resistant epilepsies, the studies were mostly small, short-term, or open-label. The only large trials used Epidiolex/Epidyolex. The same review noted that most participants in trials were already taking other anti-seizure medicines, so the added benefit of CBD is what was measured, not CBD alone.

A 2018 systematic review and meta-analysis by Stockings et al. examined cannabinoids for epilepsy. The findings for CBD were judged low certainty, with a 25% or greater reduction in seizures reported in roughly 50% of participants. The review also noted a higher rate of adverse events (diarrhoea, sedation, appetite changes) compared with placebo. For the pharmaceutical CBD product, the number needed to treat for a 50% seizure reduction was about 8, while the number needed to harm for any adverse event was about 6.

A 2018 meta-analysis of four randomised trials reported a reduction in convulsive seizures that was modest but statistically significant, with a number needed to treat of 8 for a 50% or greater reduction in seizures.

A 2022 network meta-analysis (Lattanzi et al., published in Epilepsia) compared add-on antiseizure medications in Dravet and Lennox-Gastaut syndromes. It found pharmaceutical CBD reduced seizure frequency for both conditions, but the effect size was moderate and adverse effects (diarrhoea, decreased appetite, somnolence) were common.

A 2024 meta-analysis of 14 RCTs in treatment-resistant epilepsy found a significant reduction in seizure frequency (standardised mean difference 0.30, 95% CI 0.11-0.48), but no significant change in quality of life. It noted that most findings came from Dravet and Lennox-Gastaut syndromes.

A 2024 Cochrane review of pharmaceutical CBD for epilepsy found that CBD (at 10-20 mg/kg/day) reduced seizures by 50% or more in roughly 20% more people than placebo for Dravet and Lennox-Gastaut syndromes. It did not clearly improve quality of life and increased risks of adverse events (diarrhoea, decreased appetite, somnolence, and transaminase elevation, particularly with clobazam).

A 2020 systematic review found that adding CBD to conventional anti-seizure medications in treatment-resistant epilepsy produced modest seizure reduction but significant adverse events, including diarrhoea, decreased appetite, and sedation.

No trial has yet demonstrated that unregulated CBD oil preparations reduce seizures. The body of findings is built on pharmaceutical-grade CBD (Epidyolex), not on botanical oils or isolates sold as supplements.

THC and epilepsy: what the research says

The findings for THC alone or THC-dominant preparations in epilepsy are weak and inconsistent. Most trials use CBD-dominant or CBD-only preparations. A 2018 systematic review in Frontiers in Pharmacology found that CBD reduced seizure frequency in some treatment-resistant epilepsies, but the review noted that THC-dominant preparations had insufficient backing and a higher risk of psychoactive adverse effects.

The New Zealand regulatory pathway

The NZ Formulary currently lists cannabis preparations for chronic pain, not for epilepsy. It does not list a cannabis product for epilepsy as a funded or routine indication. Epidiolex (pharmaceutical CBD) is approved by Medsafe for Dravet and Lennox-Gastaut syndromes, but it is not funded and is accessed via special authority or named-patient supply.

The Misuse of Drugs (Medicinal Cannabis) Amendment Act 2018 legalised the prescription pathway. The Medicinal Cannabis Scheme came into force on 1 April 2020. Under the scheme, all preparations must meet GMP standards and be approved by the Ministry of Health before they can be prescribed. The NZ Formulary notes: “The current evidence base for medicinal cannabis products is still developing. The place of cannabis-based products in modern therapeutics remains to be more fully defined.”

Key points to check before a prescribing conversation

• Epidiolex (pharmaceutical CBD) is the only cannabis-derived medicine with high-quality trial findings for epilepsy. In New Zealand it is approved by Medsafe for seizures associated with Lennox-Gastaut syndrome, Dravet syndrome, or tuberous sclerosis complex in people aged two years and older. This is a regulatory statement of the product’s indication, not a claim about general cannabis.
• Its body of findings is distinct from that of other CBD preparations, because the trials were done with a specific, purified oral CBD formulation (pharmaceutical-grade, ~99% CBD, with known excipients) and under standardised dosing.
• Cannabidiol is not a general “anti-seizure” agent for all epilepsy types. The strongest backing is for three rare childhood-onset syndromes: Dravet syndrome, Lennox-Gastaut syndrome, and tuberous sclerosis complex.
• The key caveat is that the pivotal findings are industry-funded, so the magnitude of benefit may be overstated in marketing. Later independent systematic reviews have confirmed the effect but noted the limitation.

What this guide covers

This guide answers one question: what do the findings say about cannabis-based medicines for epilepsy, and how does that fit into New Zealand’s regulated prescribing pathway? It focuses on pharmaceutical-grade CBD and the specific, licensed preparations that have backing in epilepsy. It does not cover recreational use, unregulated preparations, or “CBD wellness” claims.