There is growing interest in using cannabinoids for chronic pain. We performed a systematic review and meta-analysis of randomized controlled trials to evaluate the analgesic efficacy and adverse effects of cannabinoids for chronic non-cancer pain. PubMed, EMBASE, Web of Science, Cochrane CENTRAL and clinicaltrials.gov were searched up to December 2018. Information on the type, dosage, route of administration, pain conditions, pain scores, and adverse events were extracted for qualitative analysis. Meta-analysis of analgesic efficacy was performed. Meta-regression was performed to compare the analgesic efficacy for different pain conditions (neuropathic versus non-neuropathic pain). Risk of bias was assessed by The Cochrane Risk of Bias tool, and the strength of the evidence was assessed using the Grade of Recommendations Assessment, Development, and Evaluation (GRADE) approach. Forty-three randomized controlled trials were included. Meta-analysis was performed for 33 studies that compared cannabinoids to placebo, and showed a mean pain score (scale 0–10) reduction of −0.70 (p < 0.001, random effect). Meta-regression showed that analgesic efficacy was similar for neuropathic and non-neuropathic pain (Difference = −0.14, p = 0.262). Inhaled, oral, and oromucosal administration all provided statistically significant, but small reduction in mean pain score (−0.97, −0.85, −0.45, all p < 0.001). Incidence of serious adverse events was rare, and non-serious adverse events were usually mild to moderate. Heterogeneity was moderate. The GRADE level of evidence was low to moderate. Pain intensity of chronic non-cancer patients was reduced by cannabinoids consumption, but effect sizes were small. Efficacy for neuropathic and non-neuropathic pain was similar.
