Abstract
Cannabinoid drugs are registered for postoperative nausea and emesis, Tourette syndrome and tumor-related anorexia, but are also used for spasticity and pain relief, among other conditions. Clinical studies for spasmolysis have been equivocal and even conclusions from meta-analyses were not consistent. This may be due to uncertainty in diagnostic criteria as well as a lack of direct spasmolytic activity (direct causality). In this review we used the Hill criteria to investigate whether a temporal association is causal or spurious. METHODS: A systematic literature search was performed to identify all clinical trials of cannabinoids for spasticity. Studies were evaluated for dose dependency and time association; all studies together were analyzed for reproducibility, coherence, analogy and mechanistic consistency. A Funnel plot was done for all studies to identify selection or publication bias. RESULTS: Twenty-seven studies were included in this meta-analysis. The spasmolytic activity (effect strength) was weak, with a nonsignificant small effect in most studies and a large effect only in a few studies ("enriched" studies, low patient numbers). No dose dependency was seen and plotting effect size vs. daily dose resulted in a slope of 0.004. Most studies titrated the cannabinoid to the optimum dose, e.g., 20 mg/d THC. The effect decreased with longer treatment duration (3-4 months). The spasmolytic effect is consistent for different European countries but not always within a country, nor is the effect specific for an etiology (multiple sclerosis, spinal cord injury, others). For other criteria like plausibility, coherence or analogous effects, no data exist to support or refute them. In most studies, adverse effects were frequently reported indicating a therapeutic effect only at high doses with relevant side effects. CONCLUSIONS: Current data do not support a specific spasmolytic effect; a general decrease in CNS activity analogous to benzodiazepines appears more likely. Whether individual patients or specific subgroups benefit from cannabinoids is unclear. Further studies should compare cannabinoids with other, nonspecific spasmolytic drugs like benzodiazepines.