Cannabis-based medicines for chronic neuropathic pain in adults

Estimates of the population prevalence of chronic pain with neuropathic components range from 6% to 10%. Current pharmacological treatments for neuropathic pain help only a minority. New treatments are needed. Cannabis is increasingly promoted in the media as a treatment for chronic pain. This is an update of a review first published in 2018.

To assess the benefits and harms of cannabis-based medicines (herbal, plant-based, synthetic) compared to placebo or conventional drugs for chronic neuropathic pain conditions in adults.

We searched CENTRAL, MEDLINE, Embase, and three trial registries, together with reference checking. The latest search date was 29 January 2025.

We selected randomised, double-blind controlled trials of medical cannabis, plant-derived and synthetic cannabinoids, against placebo or any other active treatment for chronic neuropathic pain conditions in adults, with a treatment duration of at least two weeks. We excluded studies whose double-blind duration was less than two weeks and studies which did not explicitly state that the pain was of a neuropathic nature.

Critical outcomes were the number of participants reporting pain relief of at least 50%, a Patient Global Impression of Change (PGIC) rating of 'much' or 'very much' improved, serious adverse events, and withdrawals due to adverse events.

We assessed the risk of bias (RoB) for seven outcomes reported in three summary of findings tables using the Cochrane RoB 1 tool.

We synthesised results for each outcome using meta-analysis with a random-effects model by calculating absolute risk differences (RD) and standardised mean differences (SMD) with 95% confidence intervals (CI) for dichotomous outcomes and continuous outcomes, respectively. We used GRADE to assess the certainty of evidence for prespecified outcomes.

We included six new studies involving 450 participants, along with 15 studies involving 1737 participants from the 2018 review, for a total of 21 studies with 2187 participants. The studies ranged from two to 26 weeks in duration. Sample sizes ranged from 18 to 339 participants. Participants' mean age ranged from 34 to 61 years, and the proportion of women ranged from 0% to 90%. Five studies included participants with central neuropathic pain, 14 studies included participants with peripheral neuropathic pain, and two studies included both types. Seven studies administered tetrahydrocannabinol (THC)-dominant medicines; nine studies, balanced THC and cannabidiol (CBD) medicines; and five studies, CBD-dominant medicines. Twenty studies compared cannabis-based medicine to placebo, and one study's comparator was dihydrocodeine. We judged the overall risk of bias to be low in six studies, unclear in 10 studies, and high in five studies.

THC-dominant medicines versus placebo. There is no clear evidence for an effect on pain relief of at least 50% (RD 0.14, 95% CI -0.07 to 0.37; 7 studies, 534 participants), PGIC rating of 'much' or 'very much' improved (RD 0.17, 95% CI -0.24 to 0.58; 2 studies, 72 participants), withdrawals due to adverse events (RD 0.03, 95% CI -0.02 to 0.08; 6 studies, 511 participants), serious adverse events (RD 0.02, 95% CI -0.01 to 0.06; 7 studies, 537 participants), pain relief of at least 30% (RD 0.16, 95% CI -0.08 to 0.40; 7 studies, 566 participants), and psychiatric disorder-related adverse events (RD 0.01, 95% CI -0.01 to 0.03; 4 studies, 368 participants), all with very low-certainty evidence. They may increase nervous system adverse events (RD 0.25, 95% CI 0.14 to 0.37; 5 studies, 439 participants; low-certainty evidence). THC/CBD-balanced medicines versus placebo. There is no clear evidence for an effect on pain relief of at least 50% (RD 0.04, 95% CI 0.00 to 0.08; 8 studies, 746 participants) and serious adverse events (RD 0.01, 95% CI -0.02 to 0.03; 11 studies, 1449 participants), both with very low-certainty evidence. The evidence is very uncertain about the effect on nervous system-related (RD 0.39, 95% CI 0.23 to 0.55; 11 studies, 1445 participants) and psychiatric disorder-related adverse events (RD 0.08, 95% CI 0.03 to 0.13; 9 studies, 1375 participants), both very low-certainty evidence. They may increase PGIC ratings of 'much' or 'very much' improved (RD 0.07, 95% CI 0.02 to 0.11; 7 studies, 1145 participants), pain relief of at least 30% (RD 0.07, 95% CI 0.02 to 0.12; 10 studies, 1285 participants), and withdrawals due to adverse events (RD 0.05, 95% CI 0.02 to 0.09; 11 studies, 1449 participants), all with low-certainty evidence, though these effects were not clinically relevant. CBD-dominant medicines versus placebo. There is no clear evidence for an effect on pain relief of at least 50% (RD -0.08, 95% CI -0.20 to 0.05; 5 studies, 208 participants; very low-certainty evidence). They may increase or decrease PGIC ratings of 'much' or 'very much' improved (RD -0.03, 95% CI -0.22 to 0.16; 2 studies, 79 participants), withdrawals due to adverse events (RD 0.02, 95% CI -0.03 to 0.06; 5 studies, 213 participants), serious adverse events (RD 0.02, 95% CI -0.03 to 0.06; 5 studies, 213 participants), pain relief of at least 30% (RD -0.04, 95% CI -0.17 to 0.09; 5 studies, 218 participants), nervous system-related adverse events (RD -0.03, 95% CI -0.10 to 0.03; 5 studies, 208 participants), and psychiatric disorder-related adverse events (RD -0.01, 95% CI -0.06 to 0.04; 5 studies, 208 participants), all with low-certainty evidence.

There is no clear evidence for an effect of THC-dominant medicines on pain relief of 50% or greater, PGIC ratings of 'much' or 'very much' improved, withdrawals due to adverse events, and serious adverse events (very low-certainty evidence). There is no clear evidence for an effect of THC/CBD-balanced medicines on pain relief of 50% or greater and serious adverse events (very low-certainty evidence). They may increase PGIC ratings of 'much' or 'very much' improved, and withdrawals due to adverse events (low-certainty evidence). There is no clear evidence for an effect of CBD-dominant medicines on pain relief of 50% or greater (very low-certainty evidence). They may increase or decrease PGIC ratings of 'much' or 'very much' improved, serious adverse events, and withdrawals due to adverse events (low-certainty evidence).

No funding.

DOI 2018 review: 10.1002/14651858.CD012182.pub2.