Efficacy and adverse event profile of cannabidiol and medicinal cannabis for treatment-resistant epilepsy: Systematic review and meta-analysis

This paper aimed to systematically examine the efficacy and adverse event (AE) profile of cannabidiol and medicinal cannabis by analyzing qualitative and meta-analytic data. We used the terms ("Cannabidiol" OR "Cannabis") AND "Epilepsy" AND ("Treatment" OR "Therapeutics") as keywords to retrieve studies indexed on PubMed, ScienceDirect, and CENTRAL databases. The inclusion criteria were as follows: clinical studies with a longitudinal observational design and intervention using cannabinoid derivatives, especially cannabidiol and medicinal cannabis, whereby some results involved the frequency of epileptic seizures. We used Cochrane Collaboration's Review Manager software (RevMan 5.1.6) for the meta-analysis and dichotomized the articles to a confidence interval of 95%. From 236 articles, we selected 16 for descriptive analysis; we selected only 4 for the meta-analysis. According to the results, a statistically meaningful effect of cannabidiol compared with placebo was observed (p < 0.00001). When comparing treatment with cannabidiol or medicinal cannabis, significance was not found for the AE profile (p = 0.74). As AEs for cannabidiol were more common under short-term than under long-term treatment (p < 0.00001), this approach was favorable in the long term. Furthermore, cannabidiol is more effective than placebo, regardless of the etiology of epileptic syndromes and dosage. Overall, the AE profile did not differ across treatments with cannabidiol or medicinal cannabis, though it did differ favorably for long-term than for short-term treatment.