This draft report is available in electronic format only (Draft Report, [PDF, 880.3 MB]; Draft Appendixes [PDF, 1.2 MB]; Draft Appendix Study Details Evidence Tables [XLSX, 93.9 KB]; Draft Appendix Cohort Studies Evidence Tables [XLSX, 13.8 KB]; Draft Appendix Rate of Bias Evidence Tables [XLSX, 15.3 KB]). For additional assistance, please contact us.
In patients with chronic (mainly neuropathic) pain with short-term treatment (4 weeks to <6 months):
- Studies of cannabis-related products were grouped based on their tetrahydrocannabinol (THC) to cannabidiol (CBD) ratio using the following categories: high-THC to CBD, comparable THC to CBD, and low-THC to CBD.
- Comparable THC to CBD ratio oral spray is probably associated with small improvements in pain severity and may be associated with small improvements in function. There was no effect in pain interference or serious adverse events. There may be a large increased risk of dizziness and sedation and a moderate increased risk of nausea.
- Synthetic THC (high-THC to CBD) may be associated with moderate improvement in pain severity and increased risk of sedation, and large increased risk of nausea. Synthetic THC is probably associated with a large increased risk of dizziness.
- Extracted whole-plant high-THC to CBD ratio products may be associated with large increases in risk of withdrawal due to adverse events and dizziness.
- Evidence on whole-plant cannabis, low-THC to CBD ratio products (topical CBD), other cannabinoids (cannabidivarin), and comparisons with other active interventions was insufficient to draw conclusions.
- Other key adverse event outcomes (psychosis, cannabis use disorder, cognitive deficits) and outcomes on the impact on opioid use were not reported.
- No evidence on other plant based compounds such as kratom met criteria for this review.
Objectives. To evaluate the evidence on benefits and harms of cannabinoids and similar plant-based compounds (PBCs) to treat chronic pain.
Data sources. Ovid® MEDLINE®, PsycINFO®, Embase®, the Cochrane Library, and SCOPUS® databases, reference lists of included studies, submissions received after Federal Register request, were searched to March, 2021.
Review methods. Using dual review, we screened search results for randomized controlled trials (RCTs) and observational studies of patients with chronic pain evaluating cannabis, kratom, and similar compounds with any comparison group and at least 1 month of treatment or followup. Dual review was used to abstract study data, assess study-level risk of bias, and rate the strength of evidence. Prioritized outcomes included pain, function, and adverse events. We grouped studies that assessed tetrahydrocannabinol (THC) and/or cannabidiol (CBD) based on their THC to CBD ratio and categorized them as high-THC to CBD ratio, comparable THC to CBD ratio, and low-THC to CBD ratio. We also grouped studies by whether the product was a whole-plant product (cannabis), or cannabinoids extracted or purified from a whole plant, or synthetic. We conducted meta-analyses using the profile likelihood random effects model and assessed between-study heterogeneity using Cochran's Q statistic chi square and the I2 test for inconsistency. Magnitude of benefit was categorized into no effect, or small, moderate, and large effects.
Results. From 2,671 abstracts, 20 RCTs (N=1,776) and 3 observational studies (N=653) assessing different cannabinoids were included; none of kratom. Studies were short-term (except one), and 75 percent enrolled patients with a variety of neuropathic pain. Comparators were primarily placebo or usual care. The strength of the evidence (SOE) was low, unless otherwise noted. Compared with placebo, comparable THC to CBD ratio oral spray was associated with a small benefit in change in pain severity (7 RCTs, N=702, 0 to10 scale, mean difference [MD] −0.54, 95% confidence interval [CI] −0.95 to −0.19, I2=30%; SOE: Moderate) and function (2 RCTs, N=183, statistically significant changes using different scales). There was no effect on pain interference or withdrawals due to adverse events. There was a large increased risk of dizziness and sedation and a moderate increased risk of nausea (dizziness: 6 RCTs, N=866, 30% vs. 8%, relative risk [RR] 3.57, 95% CI 2.42 to 5.60, I2=0%; sedation: 6 RCTs, N=866, 22% vs. 16%, RR 5.04, 95% CI 2.10 to 11.89, I2=0%; and nausea: 6 RCTs, N=866, 13% vs. 7.5%, RR 1.79, 95% CI 1.20 to 2.78, I2=0%). Synthetic high-THC to CBD was associated with a moderate improvement in pain severity, a moderate increase in sedation, and a large increase in nausea (pain: 5 RCTs, 0 to 10 scale, MD −1.08, 95% CI −1.96 to −0.43, I2=42%; sedation: 3 RCTs, 19% vs. 10%, RR 1.73, 95% CI 1.03 to 4.63, I2=0%, 2 RCTs, 12% vs. 6%, RR 2.19, 95% CI 0.77 to 5.39; I2=0%). We found moderate SOE for a large increased risk of dizziness (2 RCTs, 32% vs. 11%, RR 2.74, 95% CI 1.47 to 6.86, I2=0%). Extracted whole-plant high-THC to CBD (oral) was associated with a large increased risk of withdrawal due to adverse events (1 RCT, 13.9% vs. 5.7%, RR 3.12, 95% CI 1.54 to 6.33) and dizziness (1 RCT, 62.2% vs. 7.5%, RR 8.3, 95% CI 4.53 to 15.34). We observed a moderate improvement in pain severity when combining all studies of high-THC to CBD (7 RCTs MD −1.26, 95% CI −2.17 to −0.65, I2=59%; SOE: moderate). Evidence on whole-plant cannabis, topical, low-THC to CBD, other cannabinoids, comparisons with active products, and impact on use of opioids was insufficient to draw conclusions. Other important harms (psychosis, cannabis use disorder, and cognitive effects) were not reported.
Conclusions. Low to moderate strength evidence suggests small to moderate improvements in pain, and moderate to large increases in common adverse events (dizziness, sedation, nausea) and study withdrawal due to adverse events with cannabis products in short-term treatment (1 to 6 months). Evidence for other comparisons, outcomes, and PBCs were unavailable or insufficient to draw conclusions. Small sample sizes, lack of evidence for moderate and long-term use and key outcomes, such as other adverse events and impact on use of opioids during treatment indicate that more research is needed.