What does the research say?
In a study of 118 patients with various types of nerve pain (widespread nerve damage, pain after shingles, nerve injury) who had not responded to at least one treatment, pure CBD, pure THC, a mix of the two and placebo were compared, all taken by mouth. After eight weeks, no group did better than placebo; on one scale, the pain score in the CBD group was slightly worse than with placebo.
In a study in Denmark of 134 people with nerve pain from MS or spinal cord injury, pain in the group taking up to 45 mg of CBD a day was also no different from placebo. By contrast, in a study in Australia of 40 people with nerve pain from spinal cord injury, the CBD dose was raised to 800 mg a day over two weeks; during the 6-week treatment period, the pain score was slightly lower than with placebo (about half a point on a 0–10 scale). This difference is small and comes from a single small study; side effects were also reported more often during the CBD period (68% vs. 53%). In an older and very small study (24 people with neurological diseases such as MS and spinal cord injury, each product for 2 weeks), pain had decreased more with a CBD-rich plant extract than with placebo; but these patients were not selected for nerve pain and it is not known whether the extract contained THC, so the evidence is very weak.
In a small study of 29 people with nerve damage in the feet and legs, CBD oil applied to the skin was reported to reduce pain, cold sensations and itching; however, the study’s methods were weak, so this result should be interpreted with caution. In a crossover pilot study of 40 people with chemotherapy-related nerve damage, using pure CBD cream in 2-week periods, symptoms were similar during the CBD and placebo periods. In a study of 100 people with nerve pain from diabetes, a cannabis oil that was mainly THC and also contained small amounts of CBD and CBN was applied to the skin for 12 weeks, and pain symptoms decreased much more than with placebo. An effect this large was not seen in the other studies in the table; this single-center result needs to be confirmed by other studies, and some numbers in the publication don’t match between the tables and the text.
The most studied product is a prescription mouth spray containing roughly equal amounts of THC and CBD (nabiximols). In two 4–5-week studies (66 people with central pain from MS, 125 people with nerve pain that worsens with touch), pain decreased slightly more than with placebo. In two larger 14–15-week studies (339 people with MS, 246 people with touch-evoked nerve pain), there was no clear difference in average pain; in one of them, slightly more people had their pain reduced by at least 30% with the spray. Two studies by the drug’s maker, in 297 people with nerve pain from diabetes and in 116 people with spinal cord injury, have results only in the trial registry (no journal publication could be found as of October 2026); neither showed a difference in pain from placebo. The 2018 version of the independent Cochrane review found that 21% of people had at least half their pain relieved with cannabis products, compared with 17% with placebo. In the 2026 update (21 studies, 2,187 people), no product group showed a clear effect on this measure (very low certainty of evidence); with the product containing equal amounts of THC and CBD, slightly more people may have had at least 30% relief (low certainty), but the authors did not consider this difference clinically meaningful. Nervous system side effects (such as dizziness and drowsiness) were more common with products containing THC.
Synthetic THC (dronabinol) was tested in two studies of central pain from MS: in a small 3-week study of 24 people, pain eased a little, while in a larger 16-week study of 240 people there was no difference from placebo. Studies of cannabis inhaled as smoke or vapor (nerve damage from HIV, diabetic neuropathy, spinal cord injury, nerve pain after trauma) were small (16–50 people) and lasted from a single day to 5 days; in these, pain decreased more than with placebo in the short term, but they tell us nothing about long-term use. In an 8-week study of nerve-related low back pain, with results available only in the trial registry, inhaled cannabis was not clearly different from placebo; in an exploratory comparison in the same study, synthetic THC reduced pain slightly more than placebo. Of three studies (15–96 people) with nabilone, a THC-like drug, one found it less effective than a weak opioid; a study that selected people who had responded to the drug at the start and a study in which it was added to gabapentin reported a benefit.
In two small studies (12 and 46 people) of chemotherapy-related nerve damage using capsules containing CBD and a small amount of THC, pain did not decrease more than with placebo; in one, neuropathy symptoms got worse in the cannabis group. In HIV-related nerve pain, cannabidivarin (CBDV), a cannabinoid similar in structure to CBD, was also no better than placebo. In a study of 30 people with brachial plexus injury (injury to the network of nerves in the arm), a cannabis oil containing roughly equal amounts of THC and CBD, dropped under the tongue, reduced pain 1 point more than placebo; this difference was below the clinical importance threshold the researchers had set in advance.
There is also real-world data, but it doesn’t prove effectiveness. In Denmark’s national registries, 1,817 neuropathic pain patients who received a prescription cannabis product used less gabapentin and spent fewer days in the hospital than similar controls; those using THC, however, used more opioids. In a German pain registry, people using the spray containing THC and CBD reported more improvement than those using synthetic THC; there was no placebo group, the analysis was funded by a drug company, and all of the authors had financial ties to that company. In a 38-week follow-up with no comparison group, which continued two of the spray’s studies, average pain fell from 6.9 to 4.2, but more than a third of the patients (38%) did not complete the follow-up.
Complex regional pain syndrome is a severe, long-lasting pain that usually starts in an arm or leg after an injury or surgery, with burning, swelling, and changes in skin color and temperature. As of October 2026, there is no controlled cannabinoid study that enrolled only patients with this syndrome. Three short studies of inhaled cannabis in mixed groups of people with nerve pain also included patients with this syndrome: 22 of 38 people in one (most of the patients), 6 of 39 in another (plus 2 people with the type caused by nerve injury), and 6 of 27 in the third. In all three, cannabis reduced pain more than placebo in the short term (in one, only at the high dose), but no separate result was reported for this syndrome; one says the main result did not change noticeably when these 6 patients were excluded (numbers were not given). In the UK medical cannabis registry, 64 patients started on a prescription cannabis product for this syndrome (oil and dried flower; at month 6, the daily amount of THC in prescriptions was about six times that of CBD); their pain intensity fell over 6 months from an average of 6.7 to 6.1 out of 10, and a clinically important improvement (at least 1 point) was seen in 28% of patients. This study had no comparison group, more than half of the patients were already using cannabis before starting, most of the authors work for the clinic chain that treated the patients, and the most commonly prescribed products belong to the same group of companies; so it is not known how much of the improvement came from the product, and this data does not prove effectiveness. In the US, a 52-person study comparing a CBD-rich cannabis extract with very little THC against placebo is enrolling patients, according to its registration (the record has not been updated since June 2024; completion is expected in 2029). In Germany, a study aiming to prevent lasting pain with low-dose THC in newly started syndrome began in 2006; its record has not been updated since 2008 and its results have not been published.
In burning mouth syndrome (burning pain inside the mouth that persists without a visible cause), there are only two small studies with cannabis. In a single-center study in 26 people comparing full-spectrum CBD oil with an inactive product (placebo), pain decreased more in the CBD group; however, in this study only one side did not know which product was given (single-blind), and how much THC the oil contained was not reported. In a 17-person study without a control group, symptoms were also reported to decrease over time in people using an oil made from cannabis flowers; because there was no comparison group, it is not known how much of this was due to the oil. The evidence is very weak; these results need to be confirmed by larger, well-designed studies.
Strength of evidence: ●●○○ Weak evidence. Finding: Mixed results. What does this mean?
Ongoing research on this topic
Trials testing CBD for this topic on ClinicalTrials.gov: 6 under way (6 recruiting); results are awaited from 2 that ended in the last two years. A trial being under way does not show that CBD works.
- NCT04816994: Phase 3, recruiting, target 200 participants
- NCT05351801: Phase 2, recruiting, target 320 participants
- NCT06731894: Phase 2, recruiting, target 120 participants
Registries in other countries also list 2 recruiting trials on this topic: other registries.
Possible side effects and drug interactions
In studies of pure CBD taken by mouth, side effects were generally close to placebo (most often dry mouth); at a very high dose (800 mg a day), tiredness, drowsiness, nausea and diarrhea were more common. According to Cochrane 2026, serious side effects and stopping because of side effects with CBD-dominant products were not clearly different from placebo (low certainty). With products containing THC, nervous system side effects such as dizziness and drowsiness were more common; psychiatric side effects were found in 17% vs. 5% in the 2018 version of the Cochrane review, and in the 2026 update this difference became uncertain (very low certainty of evidence); in one study, stopping treatment because of side effects was 18% vs. 3%.
CBD can slow the breakdown of some medicines in the body. Taken together with gabapentin and pregabalin, which are commonly used for nerve pain, as well as with opioids and sleep medicines, it can increase drowsiness; this raises the risk of falls and of problems when driving. It may also interact with blood thinners such as warfarin and clopidogrel.
Interactions with medicines used for this condition
Interactions reported for these medicines in peer-reviewed publications and official prescribing information. A medicine that is not listed can still interact with CBD; tell your doctor about every medicine you take.
| Medicine | Importance | Warning | Source |
|---|---|---|---|
| Other opioids (morphine, oxycodone, fentanyl, tramadol, etc.), Methadone | Caution | Methadone and other opioids: In a child, a high methadone level was found and drowsiness developed while CBD was being used; two weeks after CBD was stopped, the level fell to less than half and the symptoms resolved. In small controlled studies with morphine and buprenorphine, no clear change in drug levels was seen; with a single intravenous dose of fentanyl, there was no respiratory depression. Drowsiness may increase together with opioids. Tell the doctor who manages your pain or addiction treatment. | PMID 32444381, PMID 41622700, PMID 41467610, PMID 25748562, FDA Epidiolex label §7.4 (05/2026) |
| Gabapentin (Gabateva, Neruda), Pregabalin (Lyrica) | Caution | Benzodiazepines, sleep medicines and sedatives: Drowsiness and sedation may increase together with CBD. Talk to your doctor before using them together. | DailyMed Epidiolex label 7.4, EMA Epidyolex SmPC 4.4 |
| Acetaminophen (paracetamol) | Caution | Acetaminophen and the liver: In a knee osteoarthritis study, liver enzyme elevations were more common in people taking 600 mg of CBD together with 3 g a day of acetaminophen (paracetamol) than with placebo (15 vs. 5 patients). Enzyme elevations were also seen in healthy adults at a lower CBD dose (5 mg/kg a day). If you take acetaminophen regularly, talk with your doctor about liver tests. | PMID 38033459, PMID 40622698 |
| Tizanidine | Caution | Theophylline, tizanidine: CBD can raise the blood levels of these medicines; the prescribing information for approved CBD recommends considering a dose reduction. The tizanidine prescribing information itself asks that it not be combined with weak CYP1A2 inhibitors such as CBD; low blood pressure, a slowed heart rate or excessive drowsiness may occur. Tell your doctor. | FDA Epidiolex label §7.2 (05/2026), EMA Epidyolex SmPC 4.5, PMID 33951339, Tizanidine prescribing information §7.2, dailymed.nlm.nih.gov |
| NSAIDs (ibuprofen, naproxen, celecoxib, etc.) | Information | NSAID pain relievers: There is no documented meaningful interaction between CBD and these pain relievers; there are no direct studies either. The bleeding and kidney risks come from the NSAID itself. | FDA Epidiolex label §12.3 (05/2026), Celecoxib prescribing information §7, dailymed.nlm.nih.gov |
To check interactions with the medicines you take: CBD drug interaction checker.
How was CBD used in studies?
In studies of low-dose pure CBD by mouth, the dose was adjusted from 5 mg to 50 mg a day in one study and up to a maximum of 45 mg in another (study dose; not a personal recommendation). In the study that reported a small benefit, people started at 200 mg a day and increased every 4 days, reaching 800 mg a day in two weeks (study dose; not a personal recommendation); the medicine was taken with food, divided into three doses a day. This amount is far above the doses in CBD oils on the market. In one of the studies of CBD applied to the skin, a bottle of oil containing 250 mg of CBD (study dose; not a personal recommendation) was used, and how much was applied was not reported; in the other, pure CBD cream was applied twice a day, and the amount of CBD in the cream was not reported. In studies of the THC-containing spray, patients adjusted the dose themselves; the daily upper limit ranged from 12 to 48 sprays depending on the study (study dose; not a personal recommendation). Studies of synthetic THC (dronabinol) used up to 10–15 mg a day, and studies of the THC-like drug nabilone used 1–4 mg a day (study dose; not a personal recommendation); most studies of inhaled cannabis were single-session.
Study details (for health professionals)
Low-dose oral pure CBD was not superior to placebo in two RCTs: 5–50 mg/day in various peripheral nerve pains (8 weeks; on one scale the result was worse than placebo) and up to 45 mg/day in nerve pain from MS or spinal cord injury (6 weeks). In a crossover RCT (40 people) in neuropathic pain from spinal cord injury with titration to 800 mg/day, the pain score was 0.54 points (0–10 scale) lower than with placebo; the effect is small and rests on a single small study. There are two small RCTs of CBD applied to the skin: a methodologically weak study (29 people) in lower-extremity peripheral neuropathy reported a positive result; a crossover pilot study of pure CBD cream in chemotherapy-induced neuropathy (40 people, 2 + 2 weeks) found no difference from placebo. In HIV-associated neuropathy, CBDV, a cannabinoid closely related to CBD (400 mg/day, 32 people), was also not superior to placebo. In a series of single-patient crossover trials in 24 people with persistent symptoms due to neurological diseases such as MS and spinal cord injury (Wade 2003; each product for 2 weeks), pain was significantly better with a CBD-rich plant extract than with placebo; the patients were not selected for neuropathic pain, the THC content of the extract was not reported and the abstract gives no effect size, so the finding counts as very weak evidence. There are 28 RCTs of products containing THC: THC:CBD oromucosal spray (nabiximols) showed a small benefit in two 4–5-week studies; in two large 14–15-week studies the difference in mean pain was not significant (in one, the ≥30% responder rate was significantly higher with the spray); the manufacturer’s negative results in diabetic neuropathy (297 people) and spinal cord injury (116 people) are available only on ClinicalTrials.gov. Small studies of inhaled cannabis lasting from a single day to 5 days showed short-term pain reduction (individual patient data meta-analysis of 5 RCTs: NNT 5.6). Dronabinol was beneficial in a small 3-week study in MS and ineffective in a 16-week, 240-person study. A single-center, 100-person, 12-week RCT of a THC-dominant oil (THC:CBD:CBN) applied to the skin in diabetic neuropathy reported a very large benefit; it has not been independently confirmed. Of 3 RCTs with nabilone (15–96 people), one found it worse than dihydrocodeine, and two (one with an enriched design enrolling only initial responders) found it beneficial. Cochrane 2026 (21 RCTs, 2,187 people; only double-blind studies lasting at least 2 weeks): no clear effect on ≥50% relief for any of THC-dominant, balanced THC:CBD or CBD-dominant products (very low certainty); with the balanced product, ≥30% relief and the patient’s global impression may increase slightly (RD 0.07; low certainty); nervous system side effects increased (RD 0.25–0.39; low to very low certainty); two-thirds of the studies were funded by drug companies. Reviews reach different conclusions: a meta-analysis of 8 RCTs reported a moderate-certainty benefit for pain and sleep. Guidelines and expert bodies do not recommend cannabinoids as a first choice for nerve pain; where they differ is on whether they can be tried when other treatments have not been enough. The neuropathic pain special interest group of the International Association for the Study of Pain (NeuPSIG) issued a weak recommendation against using cannabinoids in 2015; cannabinoids are also absent from the first-, second- and third-line treatments recommended in the group’s 2025 update. The German Neurological Society’s 2019 guideline does not recommend cannabinoids because their effect is small and their side effects common; it states that, when other treatments have failed, they may be tried off-label only in individual patients as part of a multimodal treatment plan. A 2022 Canadian primary care guideline (PEER) finds the benefit of cannabinoids in nerve pain uncertain and lists them among options that can be discussed with the patient after treatments with clear benefit have been considered. There are also more favorable views: the Canadian Pain Society recommended cannabinoids as third-line treatment in 2014 (and wrote that they may be considered first-line for pain from spinal cord injury); the European Pain Federation’s 2018 position paper states that, when first- and second-line drugs do not work well enough or are not tolerated, experienced physicians may consider cannabis-based products as part of multimodal treatment, preferably as add-on drugs; and the Brazilian Academy of Neurology in 2020 placed products containing THC and CBD at the third line for central nerve pain, not as monotherapy but as add-on drugs in selected patients who have not responded to other drugs. The same academy’s 2021 position paper, however, states that for peripheral nerve pain the evidence points more toward their being ineffective, and that in some situations without an established evidence-based treatment (nerve pain from spinal cord injury, central post-stroke pain, chemotherapy or trauma) they may be considered as add-on treatment within an individual plan (one of its 49 authors is an executive at a CBD company); in chemotherapy-induced nerve damage, three small controlled studies published after this paper (2024–2025) found that pain or neuropathy symptoms did not improve more than with placebo. October 2026 search (PubMed, Cochrane, ClinicalTrials.gov): 35 controlled studies of cannabinoids in patients with neuropathic pain were found (all randomized; 3 with results only in the registry); all of them are in the table below, together with 6 topic-specific systematic reviews (including the 2018 and 2026 versions of the Cochrane review) and 4 real-world studies (one specific to complex regional pain syndrome). Complex regional pain syndrome was searched separately on October 28, 2026 (PubMed, ClinicalTrials.gov): no controlled study specific to this condition was found; three short studies of inhaled cannabis in mixed nerve pain groups also included patients with this syndrome (in one, most of the patients), but no separate result was given for them; a study comparing a CBD-rich extract with placebo is ongoing according to its registration. There is no separate page for burning mouth syndrome (burning pain inside the mouth that persists without a visible cause); it was searched separately on October 29, 2026 (PubMed, ClinicalTrials.gov): in a single-center, single-blind study comparing full-spectrum CBD oil (THC content not reported) with placebo in 26 patients (Parise 2026), pain decreased more than in the placebo group from the third assessment onward, and oral health–related quality of life also improved in the CBD group; in a 4-week study without a control group in 17 patients using an oil made from cannabis flowers (Gambino 2021), symptoms were reported to decrease over time. Both are very small and the evidence is very weak; the placebo-controlled study is in the table (not added to the neuropathic pain counts above), and the uncontrolled study is mentioned only here and in the research section. Studies in groups not selected for neuropathic pain (except Wade 2003, which also tested a CBD-rich extract: that study is in the table to show every piece of data on the CBD side, but it was not added to the count of controlled studies above), experimental synthetic drugs that target the body’s own cannabinoid system, and three small registry records whose comparative pain results could not be interpreted were not included in the table; completed studies with unpublished results are listed under unverifiable items.
Evidence assessment: Weak; conflicting (oral 45–50 mg/day negative in two RCTs, 800 mg/day small benefit in a single small RCT; Cochrane 2026: no clear effect with CBD-dominant products, very low certainty)
CBD alone
Low-dose pure CBD by mouth (45–50 mg/day) was no better than placebo in two RCTs (in Zubcevic 2023, +1.14 NRS worse on one scale); at a very high dose (800 mg/day), a single small RCT in spinal cord injury showed a small benefit (−0.54/10). In a 40-person crossover pilot RCT (2 + 2 weeks) of pure CBD cream applied to the skin in chemotherapy-related neuropathy, there was no difference. In Cochrane 2026, CBD-dominant products (5 studies, 208 people) showed no clear effect on ≥50% relief (RD −0.08; 95% CI −0.20 to 0.05; very low certainty); the high-dose study was published after this review’s search. In an old 24-person crossover study in people with persistent symptoms due to neurological diseases (Wade 2003), pain decreased significantly more with a CBD-rich plant extract than with placebo; but the patients were not selected for neuropathic pain, each product was tried for only 2 weeks and the THC content of the extract was not reported (very weak evidence).
Products containing THC
There are 28 RCTs of products containing THC (THC:CBD spray, dronabinol, sublingual THC, sublingual THC:CBD oil, inhaled cannabis, CBD+THC capsules, THC oil applied to the skin; 3 have results only in the registry, and 2 are the THC arms of four-arm CBD studies). In most studies lasting from a single day to 5 weeks, pain decreased slightly more than with placebo; in most studies lasting 6 weeks or longer, there was no clear difference (a single 100-person study with an oil applied to the skin reported a large but unconfirmed benefit). Cochrane 2026: no clear effect on ≥50% relief (THC-dominant RD 0.14, 95% CI −0.07 to 0.37; balanced THC:CBD RD 0.04, 0.00–0.08; very low certainty); with the balanced product, ≥30% relief may increase slightly (RD 0.07; low certainty). Nervous system side effects increased (THC-dominant RD 0.25, low certainty; balanced RD 0.39, very low certainty). There are also 3 RCTs with nabilone (15–96 people).
Evidence: Low to very low certainty (Cochrane 2026): benefit uncertain or small, nervous system side effects more common
Dose range in studies
Oral pure CBD 5–50 mg/day (6–8 weeks) and up to 800 mg/day (6 weeks); topical 250 mg CBD / 3 fl oz oil; pure CBD cream twice a day (amount not reported). For products containing THC: mouth spray up to 12–48 sprays a day (2.5–2.7 mg THC per spray), dronabinol up to 10–15 mg a day, nabilone 1–4 mg/day, inhaled cannabis 1–9.4% THC (mostly single sessions; in one study 0.5–1 mg THC with a metered-dose device), CBD+THC capsules 125–300 mg CBD + 6–15 mg THC a day, THC-dominant oil on the skin 1–6 drops depending on pain intensity (3.2 mg THC per drop)
Starting dose and titration
Flexible dosing in Zubcevic 2023; 3-week titration in Hansen 2023; in Robertson 2026, increased by 200 mg every 4 days to 800 mg/day on day 13; in the nabiximols study, patients self-titrated
These are doses researchers gave under medical supervision; they are not a recommended dose for you. "mg/kg/day" means milligrams per kilogram of body weight per day.
| Study | Participants | Product | Dose and titration | Duration | Result | Source |
|---|---|---|---|---|---|---|
| Zubcevic 2023 Double-blind RCT, 4 arms (CBD / THC / CBD+THC / placebo) | 118 randomized (115 ITT) Polyneuropathy, postherpetic neuralgia, peripheral nerve injury; not responding to at least 1 treatment | Oral capsules: pure CBD; pure THC; CBD/THC 2:1 | CBD 5 mg at start → 50 mg (in the combination 5/2.5 mg → 50/25 mg; THC 25 mg) Increased up to 10 capsules/day (CBD 50 mg) over the first 4 weeks, then fixed for 4 weeks; median 10 capsules in the CBD arm at week 8 | 8 weeks | No group was superior to placebo; in the CBD group, pain at week 8 was 1.14 points worse than placebo (0–10 scale; 95% CI 0.11–2.19). The difference was 0.38 (−0.65 to 1.4) in the THC group and −0.12 (−1.13 to 0.89) in the CBD+THC group; neither differed from placebo Side effects: Dry mouth slightly more common in all active arms; discontinuation due to side effects 2/27 (7.4%) with CBD, 0 with placebo | PMID 36571471, doi:10.1002/ejp.2072, EudraCT 2017-005198-38 |
| Robertson 2026 Double-blind, placebo-controlled crossover RCT (4-week washout) | 40 randomized (38 analyzed) Neuropathic pain from spinal cord injury lasting at least 3 months (Australia) | Oral pure CBD capsules (200 mg, 99% purity); no THC detected in the blood | Up to 800 mg/day, with food and divided into 3 doses a day (largest dose 400 mg in the evening) 200 mg/day (days 1–4) → 400 (days 5–8) → 600 (days 9–12) → 800 mg/day (days 13–42) | 2 × 6 weeks | During the treatment period, pain was 0.54 points lower with CBD than with placebo (0–10 scale; 95% CI −0.88 to −0.21); the effect is small Side effects: Side effects reported in 68.4% with CBD, 52.6% with placebo; most often tiredness/drowsiness (15), nausea (10), diarrhea (7); 1 person stopped because of nausea | PMID 42256679, doi:10.1016/j.eclinm.2026.103986, ANZCTR ACTRN12622000634774 |
| Hansen 2023 Multicenter, double-blind, placebo-controlled, 4-arm parallel RCT (THC / CBD / THC+CBD / placebo) | 134 (MS 119, spinal cord injury 15; CBD 31, THC 32, THC+CBD 31, placebo 40) Neuropathic pain and/or spasticity from MS or spinal cord injury (Denmark) | Oral capsules: synthetic pure CBD; THC; both together | CBD up to 45 mg/day, THC up to 22.5 mg/day (divided into 3 doses a day, up to 9 capsules) Standard escalation over 3 weeks; a lower dose was kept if there was enough effect or side effects | 6 weeks + 1 week tapering | In the pure CBD group, the improvement in neuropathic pain did not differ from placebo (difference 0.45 points; 95% CI −0.47 to 1.38; 0–10 scale, a positive value means less reduction than placebo); there was also no difference in the THC (0.42; −0.54 to 1.38) and THC+CBD (0.16; −0.75 to 1.08) groups; for spasticity too, no group differed from placebo. Only 134 of the target 448 people could be enrolled Side effects: Side effects most common in the THC arms; dry mouth 29% with CBD, 27.5% with placebo; 1 hospitalization in the CBD arm (gallstone attack) | PMID 37630995, doi:10.3390/ph16081079, EudraCT 2018-002315-98, protocol: Hansen 2021, PMID 34573231, pharmacokinetic substudy in 23 MS patients: Hansen 2024, PMID 38054364 |
| Xu 2020 Placebo-controlled RCT (followed by crossover) | 29 Lower-extremity peripheral neuropathy | Topical CBD oil (250 mg CBD / 3 fl oz); THC not reported | Not reported (in the abstract) | 4 weeks | Significant reductions in intense pain, sharp pain, cold sensations and itching were reported (methodologically weak) Side effects: No adverse events reported | PMID 31793418, doi:10.2174/1389201020666191202111534 |
| D'Andre 2024 Double-blind, placebo-controlled, crossover pilot RCT | 40 randomized (20 + 20); 36 completed both periods Chemotherapy-induced peripheral neuropathy with pain or tingling of at least 4/10, in people who had finished chemotherapy and had no evidence of cancer (USA, Mayo Clinic) | Pure CBD (isolate) cream applied to the skin; placebo cream | Twice a day to the affected areas (the amount of CBD in the cream is not in the abstract or the registry) | 2 × 2 weeks | Neuropathy symptoms (QLQ-CIPN20 sensory subscale) were similar during the CBD and placebo periods; according to the researchers, the results do not support CBD cream reducing this pain. Small, short pilot study Side effects: Side-effect scores were similar in both periods; generally well tolerated; no serious side effects (registry) | PMID 39016024, doi:10.1089/can.2023.0253, NCT05388058 |
| Wade 2003 Double-blind, placebo-controlled, randomized series of single-patient crossover trials (each patient tried four products for 2 weeks each) | 24 (18 MS, 4 spinal cord injury, 1 brachial plexus injury, 1 limb loss due to neurofibromatosis); 20 people per product in the table of a 2026 review Patients with neurological symptoms (pain, spasticity, muscle spasm, bladder problems) not responding to standard treatment (UK); patients were not selected for neuropathic pain | Sublingual spray: CBD-rich, THC-rich and 1:1 THC:CBD plant extracts, and placebo (the THC content of the CBD-rich extract is not in the sources read) | 2.5–120 mg per 24 hours (based on symptoms and unwanted effects) Self-titrated by the patient | 2-week periods for each product (four periods) | Pain relief was significantly better than placebo with both the CBD-rich and the THC-rich extract (no effect size or numbers in the abstract). Limitations: very small; each product was tried for only 2 weeks; patients were not selected because their pain was neuropathic, and there is no separate result by type of pain; because it was not reported whether the CBD-rich extract contained THC, it does not count as a pure CBD result; the full text could not be read. In later, larger studies testing low-dose oral pure CBD, pain was no different from placebo (Zubcevic 2023, Hansen 2023). The evidence is very weak for both extracts Side effects: With rapid dose increases of the THC-containing extract, temporary drops in blood pressure and intoxication in 3 patients; unwanted effects were generally well tolerated (abstract) | PMID 12617376, doi:10.1191/0269215503cr581oa |
| Parise 2026 (burning mouth syndrome) Single-center, single-blind, placebo-controlled RCT | 26 randomized (13 CBD, 13 placebo); 35 people screened Patients with burning mouth syndrome (96% women, mean age 61.9; Brazil, orofacial pain outpatient clinic) | Full-spectrum CBD oil (THC content not in the abstract) | Not reported (in the abstract) | Not reported (in the abstract) | Pain intensity and impact (VAS) decreased over time in the CBD group and did not change in the placebo group; the difference between groups was significant from the third assessment on (p < 0.001). Oral health-related quality of life (OHIP-14) improved in the CBD group (p = 0.001; no between-group comparison in the abstract). Very small, single-center and single-blind; duration, dose and the product’s THC content are not in the abstract, and the full text could not be read. Burning mouth syndrome is not a group selected by this page’s criterion for neuropathic pain. The authors reported no conflicts of interest. The evidence is very weak Side effects: No serious side effects were seen (abstract) | PMID 42951760, doi:10.1016/j.oooo.2026.07.010, PMID 33123730, 17 patients, cannabis flower extract, 4 weeks) |
| Nurmikko 2007 Double-blind, placebo-controlled RCT | 125 Peripheral neuropathic pain with allodynia | Sativex (THC:CBD roughly 1:1, oromucosal) | Patients adjusted the dose themselves; the daily upper limit is contradictory in the registry: up to 24 sprays a day (THC 65 mg + CBD 60 mg) in the intervention description, up to 48 sprays a day (THC 130 mg + CBD 120 mg) in all group descriptions in the results section; the full text of the publication could not be read Patients self-titrated | 5 weeks | Pain on a 0–10 scale decreased by 1.48 points with the active drug and 0.52 with placebo (difference −0.96; 95% CI −1.59 to −0.32; p = 0.004); allodynia, sleep and the patient’s global impression were also better. 5-week study; sponsor is the drug’s manufacturer Side effects: Sedative and gastrointestinal side effects more common; discontinuation 18% vs. 3% | PMID 17997224, doi:10.1016/j.pain.2007.08.028, NCT00711880, a 52-week open-label extension is mentioned in the abstract |
| Eibach 2021 Double-blind, placebo-controlled, crossover RCT (3-week washout) | 32 HIV-associated neuropathic pain (Germany, Berlin) | Oral cannabidivarin (CBDV; a cannabinoid found in cannabis that is similar in structure to CBD but is not CBD) | 400 mg/day | 2 × 4 weeks | Pain was 0.62 points higher during the CBDV period than with placebo (0–10 scale; 95% CI −0.27 to 1.51; p = 0.16), meaning there was no clear difference; use of additional pain relievers, pain quality and quality of life did not change either Side effects: The frequency of side effects was similar in both periods; there were no unexpected serious side effects | PMID 32770831, doi:10.1002/cpt.2016 |
| Berman 2004 Double-blind, placebo-controlled, three-period crossover RCT | 48 Treatment-resistant central neuropathic pain from brachial plexus root avulsion (tearing of the nerve roots of the network of nerves in the arm) (United Kingdom) | Oromucosal spray: plant extract containing roughly equal amounts of THC and CBD (GW-1000-02, Sativex) and plant extract containing mainly THC (GW-2000-02); placebo spray | 2.5 mg THC per spray (plus 2.5 mg CBD in the combined spray); up to 48 sprays a day (registry) | 3 × 2 weeks (preceded by a 2-week baseline period) | The expected 2-point reduction in the primary outcome was not seen. Pain decreased slightly more than with placebo: by 0.58 points with the THC+CBD spray (95% CI 0.18–0.98) and 0.64 points with the THC spray (0.24–1.03) (11-box, 0–10 scale; registry results); sleep measures also improved. Each period lasted 2 weeks; the study was conducted with the drug’s manufacturer Side effects: Side effects reported: THC+CBD 34/46, THC 37/47, placebo 20/48 (registry); mostly mild to moderate, including feeling intoxicated, and resolved on their own | PMID 15561385, doi:10.1016/j.pain.2004.09.013, NCT01606189, sponsor is the drug’s manufacturer) |
| Rog 2005 Single-center, double-blind, placebo-controlled, parallel RCT | 66 randomized (THC+CBD 34, placebo 32); 64 completed Central pain from MS (burning-tingling pain in 59 people, painful muscle spasms in 7) (United Kingdom, Liverpool) | Oromucosal spray: THC:CBD roughly 1:1 plant extract (Sativex), added to current pain treatment | 2.7 mg THC + 2.5 mg CBD per spray; up to 48 sprays a day; mean 9.6 sprays/day at week 4 (19.1 in the placebo group) Patients gradually adjusted the dose themselves | 1-week baseline period + 4 weeks of treatment | Pain intensity decreased by 2.7 points with the active drug and 1.4 with placebo (0–10 scale; between-group p = 0.005; difference in the registry −1.25, 95% CI −2.11 to −0.39); sleep disturbance also decreased more (p = 0.003). Small, 4-week study Side effects: Dizziness, dry mouth and drowsiness more common with the active drug; cognitive side effects limited to long-term memory; side effects reported 30/34 vs. 22/32 (registry) | PMID 16186518, doi:10.1212/01.wnl.0000176753.45410.8b, NCT01604265, sponsor is the drug’s manufacturer). Open-label 2-year extension: Rog 2007, PMID 18035205, 28 completed 2 years, at least one drug-related side effect in 58) |
| Selvarajah 2010 Double-blind, placebo-controlled, parallel RCT | 30 randomized; 29 analyzed Painful diabetic peripheral neuropathy (United Kingdom, Sheffield) | Oromucosal spray: THC 27 mg/ml + CBD 25 mg/ml (Sativex), added to current treatment | Under the tongue, in divided doses up to 4 times a day; increased to the highest tolerated dose (amount not reported as a number) To the highest tolerated dose over 2 weeks | 2 weeks of dose adjustment + 10 weeks fixed dose | Pain decreased in both groups, but the difference between groups was not significant (total pain score p = 0.40; 95% CI −11.3 to 27.8; 0–100 scale); no difference in quality of life either. People with depression had higher baseline pain and improved regardless of treatment Side effects: 6 of 30 people stopped because of side effects (distribution by group not given) | PMID 19808912, doi:10.2337/dc09-1029, NCT00238550, the publication gives no registration number) |
| NCT00710424 (manufacturer’s diabetic neuropathy study, registry results) Double-blind, placebo-controlled, parallel phase 3 RCT; results only in the registry (no peer-reviewed publication) | 297 randomized (149 active, 148 placebo); primary analysis 146 and 148 Painful diabetic neuropathy lasting at least 6 months and not relieved by current treatment (type 1 or type 2 diabetes) | Oromucosal spray: THC 27 mg/ml + CBD 25 mg/ml (Sativex), added to current treatment | Up to 24 sprays a day (THC 65 mg + CBD 60 mg), no more than 8 in 3 hours | 14 weeks (98 days) | Pain decreased similarly in both groups: by 1.67 points with the active drug and 1.55 with placebo (0–10 scale; difference −0.12, 95% CI −0.60 to 0.36; p = 0.63). At least 30% relief 54/149 vs. 59/148 (p = 0.52); no difference in sleep, quality of life or use of additional pain relievers either. No peer-reviewed publication Side effects: Side effects reported 120/149 vs. 101/148; possibly drug-related side effects 96 vs. 52; serious side effects 14 vs. 12; intoxication score rose with the active drug; did not complete the study 44/149 vs. 23/148 (registry) | NCT00710424, results in the registry, July 31, 2012, sponsor is the drug’s manufacturer, and the registry states that publications must be submitted for company review, no publication found on PubMed in October 2026) |
| NCT01606202 (manufacturer’s spinal cord injury study, registry results) Double-blind, placebo-controlled, parallel phase 3 RCT; results only in the registry (no peer-reviewed publication) | 116 randomized (56 active, 60 placebo); primary analysis 55 and 59 Central neuropathic pain from spinal cord injury not relieved by current treatment (pain at least 4/10) | Oromucosal spray: THC 27 mg/ml + CBD 25 mg/ml (GW-1000-02, Sativex), added to current treatment | 2.7 mg THC + 2.5 mg CBD per spray; no more than 8 sprays in 3 hours, up to 48 sprays a day | Up to 51 days (about 7 weeks) | No difference in the primary outcome: pain decreased by 0.74 points with the active drug and 0.69 with placebo (0–10 scale; difference −0.08, 95% CI −0.51 to 0.35; p = 0.71). Among secondary outcomes, those saying their overall condition had improved at least slightly were 30/56 vs. 12/60 (p < 0.001), and the Brief Pain Inventory score was slightly better (p = 0.032); no difference in spasms, sleep or quality of life. No peer-reviewed publication Side effects: Side effects reported 46/56 vs. 29/60; serious side effects 3 vs. 2 | NCT01606202, results in the registry, September 28, 2012, sponsor is the drug’s manufacturer, in October 2026, mentioned on PubMed only in a scoping review, PMID 33465022 |
| Langford 2013 Multicenter, double-blind, placebo-controlled, parallel phase 3 RCT (phase A); followed by 14 weeks open-label + 4 weeks double-blind randomized withdrawal (phase B) | 339 randomized (167 active, 172 placebo); 58 in phase B Central neuropathic pain from MS without adequate relief from current medicines | Oromucosal spray: THC:CBD (Sativex), added to current treatment | Up to 12 sprays a day (THC 32.5 mg + CBD 30 mg; registry) | 14 weeks (A) + 18 weeks (B) | The primary outcome was not met: at week 14, at least a 30% reduction in pain was 50% with the active drug and 45% with placebo (p = 0.234); difference in mean pain −0.17 points (95% CI −0.62 to 0.29; registry). In the interim analysis at week 10 there was a difference (p = 0.046). In the randomized withdrawal phase (58 people), treatment failure was 57% in those switched to placebo and 24% in those who continued the drug (p = 0.04). The authors found the findings inconsistent between the two phases Side effects: Side effects reported 99/167 vs. 65/172; serious side effects 3 vs. 2 (registry) | PMID 23180178, doi:10.1007/s00415-012-6739-4, NCT00391079, sponsor is the drug’s manufacturer) |
| Serpell 2014 Multicenter, double-blind, placebo-controlled, parallel RCT | 246 randomized (128 active, 118 placebo); 303 people screened Treatment-resistant peripheral neuropathic pain with pain triggered by touch (allodynia) | Oromucosal spray: THC:CBD (Sativex), added to current treatment | Up to 24 sprays a day (THC 65 mg + CBD 60 mg), no more than 8 in 3 hours (registry) | 15 weeks | Of the two co-primary outcomes, at least a 30% reduction in pain was significantly more frequent with the active drug (p = 0.034 in the full analysis set; odds ratio CI 1.05–3.70); the reduction in mean pain was numerically greater with the active drug but not significant (difference in the registry −0.34; 95% CI −0.79 to 0.11). Sleep quality (p = 0.007) and the patient’s global impression (p = 0.023) were better Side effects: Side effects reported 109/128 vs. 83/118; serious side effects 10 vs. 6; did not complete the study 49/128 vs. 24/118 (registry); according to the authors, no new safety concerns | PMID 24420962, doi:10.1002/j.1532-2149.2013.00445.x, NCT00710554, sponsor is the drug’s manufacturer) |
| Lynch 2014 Double-blind, placebo-controlled, crossover pilot RCT; followed by an open-label extension | 16 Established chemotherapy-induced neuropathic pain (Canada, Halifax) | Oromucosal spray: THC:CBD (Sativex) | Starting with 1 spray in the evening, up to a maximum of 12 sprays a day (registry) Increased by 1–2 sprays a day until pain relief without side effects (registry) | Not reported (in the abstract) | In the whole group, there was no significant difference in pain from placebo. In 5 of 16 people, pain decreased by at least 2 points (mean 2.6 points; 0.6 with placebo); this responder analysis also did not reach significance (number needed to treat 5). Pilot study Side effects: Not reported (in the abstract) | PMID 23742737, doi:10.1016/j.jpainsymman.2013.02.018, NCT00872144 |
| Svendsen 2004 Double-blind, placebo-controlled, crossover RCT (3-week washout) | 24 Central pain from MS, ages 23–55 (Denmark, Aarhus) | Oral dronabinol (synthetic THC) | Up to 10 mg a day | 2 × 3 weeks (15–21 days) | In the last week, median spontaneous pain was 4.0 with dronabinol and 5.0 with placebo (0–10 scale; p = 0.02); number needed to treat for 50% relief 3.5 (95% CI 1.9–24.8). The authors considered the effect modest; small, short study Side effects: More people had side effects with dronabinol, especially in the first week; most often dizziness. Functional status did not change | PMID 15258006, doi:10.1136/bmj.38149.566979.AE, Second publication in Danish: Svendsen 2005, PMID 16014264 |
| Rintala 2010 Double-blind, active-controlled (diphenhydramine), crossover pilot RCT | 7; 5 people completed both drugs Spinal cord injury and neuropathic pain below the level of injury (USA, Houston) | Oral dronabinol (synthetic THC); comparator drug diphenhydramine (a sedating allergy medicine, used as an active placebo) | Not reported (in the abstract) | Not reported (in the abstract) | The change in pain did not differ between the two drugs: +0.20 with dronabinol, −1.80 points with diphenhydramine (0–10 scale; p = 0.10); dronabinol was no better than the comparator. Very small pilot study Side effects: With both drugs, most often dry mouth, constipation, tiredness and drowsiness; 2 people left the study while taking dronabinol, their first drug | PMID 20855984, doi:10.1097/PHM.0b013e3181f1c4ec |
| Schimrigk 2017 Double-blind, placebo-controlled, parallel phase 3 RCT; followed by a 32-week open-label period | 240; 100 people continued treatment for up to 119 weeks Central neuropathic pain from MS (Germany) | Oral dronabinol (synthetic THC) solution | Up to 15 mg a day (registry) | 16 weeks double-blind + 32 weeks open-label (some up to 119 weeks in total) | At 16 weeks, pain decreased by 1.92 points with dronabinol and 1.81 with placebo (0–10 scale); there was no difference (p = 0.68). The study was run by the drug’s manufacturer; 4 of the 6 authors are company employees. The authors emphasize that the drug could be used safely over the long term Side effects: Drug-related side effects in 50% with dronabinol, 25.9% with placebo; 26% with long-term use; no signs of abuse, 1 possible dependence | PMID 29073592, doi:10.1159/000481089, NCT00959218, no results in the registry) |
| Weizman 2018 Double-blind, placebo-controlled, within-subject two-session single-dose RCT; brain MRI | 15 Chronic neuropathic pain radiating from the lower back to the leg (radicular) (Israel, Tel Aviv) | Sublingual THC oil; placebo oil | Not reported (in the abstract); in the same team’s 2024 publication with the same design, 0.2 mg/kg (mean 15 mg) | Single dose (two separate sessions) | THC reduced pain significantly more than placebo (numbers not in the abstract); the reduction in pain was associated with reduced connectivity between emotional-cognitive control regions and sensory regions of the brain. Single-dose, very small study Side effects: Not reported (in the abstract) | PMID 30185448, doi:10.1212/WNL.0000000000006293, The same team’s 12-male-patient publication with the same design: Weizman 2024, PMID 38597988, THC 0.2 mg/kg, pain on a 0–100 scale from 53 to 35 with THC and from 52 to 44 with placebo, p < 0.05 vs. placebo). Whether the participants of the two publications are the same is not stated |
| Abrams 2007 Randomized, placebo-controlled, parallel study (inpatient) | 50 completed (number randomized not in the abstract) HIV-associated painful sensory neuropathy (USA, San Francisco) | Smoked cannabis (3.56% THC); placebo cigarettes with the cannabinoids extracted | 3 cigarettes a day for 5 days | 5 days | Daily pain decreased by a median of 34% with cannabis and 17% with placebo (p = 0.03); people with more than 30% pain reduction 52% vs. 24% (p = 0.04). Experimentally induced hyperalgesia also decreased. The study lasted 5 days Side effects: No serious side effects reported | PMID 17296917, doi:10.1212/01.wnl.0000253187.66183.9c, NCT00046722 |
| Wilsey 2008 Double-blind, placebo-controlled, crossover RCT (single-session laboratory trials) | 38 Central or peripheral neuropathic pain (of 38 people, 22 had complex regional pain syndrome type I, 10 central pain from spinal cord injury or MS, and 6 diabetic neuropathy or nerve damage; all had used cannabis before) (USA, California) | Smoked cannabis: high dose 7% THC, low dose 3.5% THC; placebo | Per session using a standardized smoking procedure (amount not in the abstract) | 3 separate sessions (each a single dose) | Cannabis reduced pain more than placebo (mixed model; numbers not in the abstract); there was no effect on pain evoked by touch or heat. Single-dose, short laboratory study Side effects: Few and tolerable intoxicating effects; at the high dose, short-term cognitive effects, especially on memory | PMID 18403272, doi:10.1016/j.jpain.2007.12.010, NCT00254761 |
| Ellis 2009 Double-blind, placebo-controlled, crossover phase 2 RCT (2-week washout) | 34 randomized; 28 completed both periods HIV-associated distal sensory polyneuropathy pain that had not responded to at least two classes of pain relievers (USA, San Diego) | Smoked cannabis (1–8% THC); placebo | 4 times a day for 5 days; THC strength adjusted per person between 1% and 8% Dose escalation and adjustment on the first day of each treatment week | 2 × 5 days of treatment | Among completers, pain decreased more with cannabis than with placebo (median difference 3.3 points; effect size 0.60; p = 0.016); people with at least 30% pain reduction 46% vs. 18%. Mood and daily functioning improved similarly in both periods Side effects: Mostly mild and temporary; treatment-limiting side effects in 2 people; difficulty concentrating, tiredness, drowsiness and dry mouth more common with cannabis | PMID 18688212, doi:10.1038/npp.2008.120, NCT00255580, PMID 22133305, during the cannabis week, plasma ghrelin and leptin rose, PYY fell, and insulin fell similarly to placebo) |
| Ware 2010 Double-blind, placebo-controlled, 4-period crossover RCT | 23 randomized; 21 completed Chronic neuropathic pain after trauma or surgery (Canada, Montreal) | Cannabis smoked in a pipe: 0% (placebo), 2.5%, 6% and 9.4% THC | 25 mg of cannabis at a time, 3 times a day, for 5 days | 4 × 14 days (5 days of treatment + 9 days of washout) | In the prespecified comparison, mean daily pain was 5.4 with 9.4% THC and 6.1 with placebo (0–10 scale; difference 0.7; 95% CI 0.02–1.4); the intermediate strengths made no significant difference. Falling asleep and sleep quality improved; no difference in mood or quality of life Side effects: During the 9.4% THC period, most often headache, dry eyes, burning in the painful area, dizziness, numbness and cough | PMID 20805210, doi:10.1503/cmaj.091414, ISRCTN68314063 |
| Wilsey 2013 Double-blind, placebo-controlled, crossover RCT (single-day laboratory trials) | 39 (59 people consented, 20 did not take the drug) Central or peripheral neuropathic pain persisting despite current treatment (of 39 people, 6 had complex regional pain syndrome type I and 2 the type caused by nerve injury; when the type I patients were excluded, the main result did not change noticeably); all had used cannabis before (USA, California) | Vaporized inhaled cannabis: medium dose 3.53% THC, low dose 1.29% THC; placebo | Two doses per session using a standardized inhalation procedure (cumulative) | 3 separate sessions | With both doses, pain decreased more than with placebo; there was no difference between the two doses (p > 0.7). Number needed to treat for a 30% pain reduction: 3.2 at the low dose, 2.9 at the medium dose. Single-day laboratory study Side effects: Few intoxicating effects; cognitive effects resolved within 1–2 hours | PMID 23237736, doi:10.1016/j.jpain.2012.10.009, NCT01037088 |
| Wallace 2015 Double-blind, placebo-controlled, 4-period crossover RCT (single-session trials, 2-week washout) | 16 Treatment-resistant painful diabetic peripheral neuropathy (USA, San Diego) | Vaporized inhaled cannabis: 1%, 4% and 7% THC; placebo | A single dose per session (400 mg of cannabis or placebo, with a vaporizer; registry) | 4 separate sessions (each a single dose, 4 hours of follow-up) | Spontaneous pain decreased more than with placebo at all three doses (low p = 0.031, medium p = 0.04, high p < 0.001), and the effect was dose-related; the high dose also reduced evoked pain. Single-dose, very small study Side effects: The high dose impaired performance on 2 of 3 attention tests; euphoria 56% with placebo, 100% with the high dose; drowsiness 38% vs. 73% | PMID 25843054, doi:10.1016/j.jpain.2015.03.008, NCT00781001, secondary analysis: Wallace 2020, PMID 32565122 |
| Wilsey 2016 Double-blind, placebo-controlled, crossover RCT (8-hour laboratory sessions) | 42 Neuropathic pain from spinal cord injury or disease (USA, California) | Vaporized inhaled cannabis: 2.9% and 6.7% THC; placebo | 4 puffs; 3 hours later, 4–8 more puffs at the person’s choice For the second dose, the person chose the number of puffs (4–8) | 3 separate sessions | Cannabis reduced pain significantly more than placebo; the two doses did not differ from each other. People with at least 30% pain reduction: 45% with placebo, 70% with the low dose, 88% with the high dose (number needed to treat 4 and 3). Single-day laboratory study Side effects: Intoxication and subjective effects were dose-related; the authors interpreted the low dose as the best balance of benefit and harm | PMID 27286745, doi:10.1016/j.jpain.2016.05.010, NCT01555983, blood level analysis of the same trial: Wilsey 2016, PMID 27621666 |
| Almog 2020 Double-blind, placebo-controlled, three-arm crossover RCT (single-inhalation dose) | 27 (21 neuropathic pain, 6 complex regional pain syndrome) Chronic neuropathic pain or complex regional pain syndrome (Israel, Haifa) | Cannabis flower vaporized with a metered-dose inhaler (22% THC, less than 0.1% CBD); placebo | 0.5 mg or 1 mg of THC in a single inhalation | 3 separate sessions, 150 minutes of follow-up | With both doses, pain decreased from baseline and stayed stable for 150 minutes; a significant difference from placebo was seen only at the 1 mg dose. Single-dose study; supported by the device manufacturer, and 7 of the 8 authors have ties to the company Side effects: Mostly mild and resolved on their own; no consistent impairment of cognitive function was seen | PMID 32445190, doi:10.1002/ejp.1605 |
| NCT02460692 (dronabinol and vaporized cannabis, registry results) Randomized, triple-blind, placebo-controlled, three-arm parallel phase 2 study (everyone received vapor and capsules); results only in the registry (no peer-reviewed publication) | 131 randomized (placebo 44, dronabinol 43, cannabis 44); 91 completed Chronic low back pain described as related to nerve damage (neuropathic); the registry’s enrollment criteria do not separately require a neuropathic diagnosis (USA, San Diego) | Vaporized inhaled cannabis (3.7% THC + 5.6% CBD) or oral dronabinol (synthetic THC); placebo | Not reported (in the registry) | 8 weeks | At week 8, pain fell from 6.03 to 5.22 with placebo, from 6.11 to 4.19 with dronabinol and from 6.24 to 4.45 with cannabis (0–10 scale). There was no difference in the main comparison, cannabis vs. dronabinol (p = 0.78); cannabis vs. placebo p = 0.070; exploratory dronabinol vs. placebo comparison p = 0.044. No peer-reviewed publication Side effects: No serious side effects; other side effects reported with placebo 30/44, dronabinol 31/43, cannabis 30/44; 15, 17 and 8 people left the study | NCT02460692, results in the registry, September 28, 2023, no publication found on PubMed in October 2026) |
| Haney 2025 Double-blind, placebo-controlled, parallel pilot RCT | 12 (6 active, 6 placebo); all completed Women with breast cancer and painful peripheral neuropathy caused by taxane chemotherapy drugs (USA, New York) | Oral cannabis capsule: CBD 100 mg + THC 5 mg | Target 1 capsule 3 times a day; participants reduced the dose (mean 2.0 capsules/day in the active group, 2.4 with placebo) | 8 weeks | Pain, sleep and well-being improved over time in both groups; however, neuropathy scores in the cannabis group were worse than placebo every week (p < 0.035), and sleep and the impact of pain on daily life also did worse than placebo (p < 0.05). No sign of benefit was seen; very small feasibility study Side effects: Well tolerated according to the authors; requests to reduce the dose in both groups | PMID 40611810, doi:10.1089/can.2025.0028, NCT03782402, results in the registry) |
| Weiss 2025 Double-blind, placebo-controlled, parallel RCT; followed by a 12-week open-label follow-up | 46 randomized (target 100); 43 completed; primary analysis 43 (23 active, 20 placebo) Patients with non-metastatic breast, colorectal, uterine or ovarian cancer who had finished chemotherapy and had grade 2–3 chemotherapy-induced peripheral neuropathy (88% women) (USA, Pennsylvania) | Oral hemp extract gel capsules: CBD 13.9–15.1 mg + THC 0.67–1.20 mg per capsule | 3 × 3 capsules a day: CBD 125–136 mg + THC 6.0–10.8 mg/day Days 1–3: 3 × 1, days 4–6: 3 × 2, from day 7: 3 × 3 capsules | 12 weeks | In the primary analysis, there was no difference between groups in the sensory symptoms of neuropathy (QLQ-CIPN20) or in pain; the difference in the impact of pain on daily life (favoring placebo) disappeared after adjustment for the baseline difference. In a secondary analysis of 46 people who completed at least 8 weeks, sensory symptoms improved more in the active group (−10.4; 95% CI −20.5 to −0.3; p = 0.044). The target number of participants was not reached Side effects: No difference in side effects between groups; raised bilirubin in one person each, raised alkaline phosphatase in one person (did not exceed the withdrawal criterion) | PMID 41211445, doi:10.3389/fonc.2025.1590168, NCT04398446, the authors declared no conflicts of interest |
| Seevathee 2025 Double-blind, placebo-controlled, parallel phase 3 RCT (single center) | 100 randomized (50 + 50); all in the intention-to-treat analysis; 7 people did not complete follow-up Adults with type 2 diabetes and painful diabetic peripheral neuropathy of the lower extremities who were not taking neuropathic pain medicines (Thailand, Kalasin) | THC-dominant cannabis oil applied to the skin (per drop THC 3.2 mg, CBD 0.32 mg, CBN 0.65 mg); placebo MCT and coconut oil. The product was supplied by the Department of Thai Traditional and Alternative Medicine of the Thai Ministry of Public Health | 1–2, 3–4 or 5–6 drops depending on pain intensity (daily frequency of application not stated in the full text that was read) | 12 weeks | The neuropathic pain symptom score (NPSI-T total) fell from 25.6 to 5.6 with the cannabis oil and from 25.2 to 22.9 with placebo; the between-group difference at week 12 was −17.3 points (p < 0.001). The effect is much larger than in the other diabetic neuropathy studies in the table; this single-center result has not been confirmed by another study, and some numbers in the publication don’t agree (between-group difference at week 4: −5.50 in the table, −8.14 points in the text; at week 8: −12.56 in the table, −14.54 in the text; placebo group score at week 4: 22.22 in the table, 24.86 in the text). 3 of the 9 authors work at the ministry department that supplied the product; the authors declared no conflicts of interest Side effects: Side effects 10% in both groups, all mild (hives, redness, skin pain); resolved by the end of follow-up | PMID 39720705, doi:10.1159/000542511, Thai Clinical Trials Registry TCTR20211220008 |
| Kittithamvongs 2025 Triple-blind, placebo-controlled, two-period crossover RCT (14-day washout) | 30 randomized; 28 analyzed Moderate to severe pain lasting at least 6 months after traumatic injury of the network of nerves in the arm (brachial plexus), ages 20–60 (Thailand, Bangkok) | Sublingual cannabis oil, THC:CBD 1:1 (27 mg THC and 25 mg CBD per mL; made by the Government Pharmaceutical Organization of Thailand), added to current pain medicines; identical-looking placebo | Starting at 4 × 1 drop a day; up to 4 × 5 drops a day (under 30 mg of THC, according to the authors) Increase by 1 drop every 2 days if pain the previous day was 4/10 or higher | 2 × 10 days | Pain decreased 1 point more with the cannabis-based medicine than with placebo (0–10 scale; 99% CI −0.03 to 2.1; p = 0.01); this was below the prespecified clinical importance threshold of 2 points. The neuropathic pain screening test was 75% positive in both periods; sleep quality was 1.5 points better with cannabis (99% CI 0.7–2.4). The authors do not recommend adding this medicine Side effects: Mild dizziness in 4 people (during the cannabis period); all continued treatment | PMID 39115545, doi:10.1097/CORR.0000000000003221, Thai Clinical Trials Registry TCTR20191106004 (the registry page returned no content in October 2026), commentary: Hoffmeyer 2025, PMID 39330963 |
| Frank 2008 Double-blind, active-controlled (dihydrocodeine, a weak opioid), crossover RCT (2-week washout) | 96 randomized; 73 in the available-data analysis, 64 in the per-protocol analysis Chronic neuropathic pain, ages 23–84 (United Kingdom, 3 hospitals) | Oral nabilone (a THC-like synthetic cannabinoid); comparator drug dihydrocodeine | Nabilone up to 2 mg a day; dihydrocodeine up to 240 mg a day The dose was increased gradually in each 6-week period | 2 × 6 weeks (14 weeks in total) | Pain was worse with nabilone than with dihydrocodeine: mean pain 6.0 mm higher (0–100 scale; 95% CI 1.4–10.5); dihydrocodeine provided better pain relief. No placebo group Side effects: Side effects more common with nabilone; no major side effects with either drug | PMID 18182416, doi:10.1136/bmj.39429.619653.80, ISRCTN15330757 |
| Toth 2012 Double-blind, placebo-controlled, flexible-dose RCT with enriched enrollment and randomized withdrawal design (everyone first took single-blind nabilone for 4 weeks) | 37 started single-blind nabilone; 26 responders were randomized (13 nabilone, 13 placebo) Treatment-resistant painful diabetic peripheral neuropathy, pain at least 4/10 (Canada, Calgary) | Oral nabilone (a THC-like synthetic cannabinoid), added to current medicines | 1–4 mg a day (flexible); mean 2.9 mg/day at the end of the study | 4 weeks single-blind + 5 weeks double-blind | Only those who responded to nabilone in the first 4 weeks (at least 30% relief) were randomized; in these people, pain decreased 1.27 points more with nabilone than with placebo (0–10 scale; 95% CI 0.25–2.29; p = 0.02); anxiety, sleep and quality of life also improved. Because non-responders were excluded, the effect may be smaller in the general patient population; in both groups, 62% of participants may have noticed which drug they were taking Side effects: In the single-blind period, 2 of 37 people stopped because of drug-related confusion | PMID 22921260, doi:10.1016/j.pain.2012.06.024, A registration by the same institution with the same design, NCT01035281, 4 weeks single-blind nabilone, responders to the double-blind period, no results in the registry, the publication gives no registration number) |
| Turcotte 2015 Double-blind, placebo-controlled, parallel RCT | 15 Patients with relapsing-remitting MS and MS-related neuropathic pain without adequate relief despite at least 1,800 mg of gabapentin a day (Canada, Winnipeg) | Oral nabilone (a THC-like synthetic cannabinoid), added to gabapentin | 2 × 1 mg a day Increased by 0.5 mg a week over 4 weeks | 9 weeks (4 weeks titration + 5 weeks fixed dose) | In a model adjusted for age, sex, disability, MS duration and baseline pain, pain intensity and the impact of pain on daily life decreased faster with nabilone than with placebo (both p < 0.01). The size of the between-group difference is not in the abstract; very small study Side effects: Most often dizziness and drowsiness; dropout rates did not differ between groups | PMID 25288189, doi:10.1111/pme.12569, NCT00480181, no results in the registry, the drug’s marketer Bausch Health is a collaborator in the registry) |
| Ateş 2026 (Cochrane) Systematic review and meta-analysis (Cochrane; update of the 2018 review) | 21 RCTs, 2,187 people (6 new studies, 450 people) Chronic neuropathic pain in adults (5 studies central, 14 studies peripheral, 2 studies both types) | THC-dominant (7 studies), balanced THC and CBD (9) and CBD-dominant (5) products; herbal, plant-derived and synthetic | 2–26 weeks | No clear effect on at least 50% pain relief in any product group: THC-dominant RD 0.14 (95% CI −0.07 to 0.37; 7 studies, 534 people), balanced THC:CBD RD 0.04 (0.00–0.08; 8 studies, 746 people), CBD-dominant RD −0.08 (−0.20 to 0.05; 5 studies, 208 people); all very low certainty. With the balanced product, at least 30% relief (RD 0.07; 0.02–0.12) and those saying they were 'much or very much improved' (RD 0.07; 0.02–0.11) may increase slightly (low certainty; the authors did not consider this clinically meaningful). 14 of the studies (two-thirds) were funded by drug companies. The review included only double-blind studies lasting at least 2 weeks; single-session or few-day inhaled cannabis studies were therefore excluded Side effects: Nervous system side effects with THC-dominant RD 0.25 (0.14–0.37; low certainty), with the balanced product RD 0.39 (0.23–0.55; very low); discontinuation due to side effects with the balanced product RD 0.05 (0.02–0.09); no difference in nervous system or psychiatric side effects with CBD-dominant products (low certainty) | PMID 41548880, doi:10.1002/14651858.CD012182.pub3, Previous version: Mücke 2018 (Cochrane, in this table) | |
| Mücke 2018 (Cochrane) Systematic review and meta-analysis | 16 RCTs, 1,750 people Chronic neuropathic pain | Mainly THC:CBD spray; nabilone, dronabinol, inhaled cannabis | 2–26 weeks | ≥50% relief 21% vs. 17% (NNTB 20, low quality); 'the potential benefits may be outweighed by potential harms' Side effects: Nervous system adverse events 61% vs. 29% (NNTH 3); psychiatric adverse events 17% vs. 5% | PMID 29513392, doi:10.1002/14651858.CD012182.pub2, Update: Ateş 2026 (Cochrane, PMID 41548880, in this table) | |
| McParland 2023 (meta-analysis) Systematic review and meta-analysis | 8 RCTs (6 studies in the sleep analysis, 8 in the pain analysis) Chronic neuropathic pain | Synthetic and natural cannabinoids, compared with placebo | Not reported (in the abstract) | Cannabinoids improved sleep quality (SMD 0.40; 95% CI 0.19–0.61) and daily pain (SMD −0.55; −0.69 to −0.19) more than placebo; the authors rated the certainty as moderate. However, heterogeneity between studies was large (I² 82% for pain) and the prediction intervals include zero (pain −1.51 to 0.39; sleep −0.12 to 0.88). This review reaches a more favorable conclusion than Cochrane 2026 Side effects: Daytime sleepiness, nausea and dizziness more common | PMID 36598058, doi:10.1136/rapm-2021-103431, PROSPERO CRD42017074255 | |
| Andreae 2015 (individual patient data meta-analysis) Systematic review and individual patient data meta-analysis (Bayesian method) | 5 RCTs, 178 people (405 observations) Neuropathy with chronic pain | Inhaled cannabis (smoked or vaporized), compared with placebo | Days to a few weeks | In the short term, inhaled cannabis provided additional pain reduction in 1 of every 5–6 patients (number needed to treat 5.6; Bayesian 95% credible interval 3.4–14). The authors write that the result is limited because of the small number of studies and patients, short follow-up, shortcomings in allocation concealment and many dropouts Side effects: Discontinuation due to side effects was rare (full text) | PMID 26362106, doi:10.1016/j.jpain.2015.07.009, PROSPERO CRD42011001182, the 5 included studies: Abrams 2007, Ellis 2009, Ware 2010, Wilsey 2008 and Wilsey 2013 (all in this table), 4 of the 12 authors are investigators of these studies | |
| Sherman 2026 (systematic review) Systematic review of RCTs (no meta-analysis) | 4 RCTs (15,377 records screened) Painful peripheral neuropathy from diabetes | Medical cannabis, single cannabinoids and approved cannabinoid medicines (added to treatment) | Not reported (in the abstract) | In 3 of four small and differing RCTs, neuropathic pain decreased more than with placebo; in 1 there was no difference. The authors write that no firm conclusion can be drawn because of risk of bias, the small number of studies and the variety of products, and that the doses observed can only guide new studies Side effects: Dizziness and cognitive symptoms common, mostly mild to moderate; discontinuation due to side effects varied across studies | PMID 41714301, doi:10.1177/25785125261425444 | |
| Datta 2026 (systematic review) Systematic review of RCTs (meta-analysis not possible) | 3 RCTs, 68 people (plus 3 ongoing studies) Chemotherapy-induced peripheral neuropathy (treatment or prevention) | THC:CBD mouth spray; oral capsule (the review classified it as 'oral CBD'; based on participant numbers, most likely a capsule that also contained a small amount of THC alongside CBD); CBD applied to the skin | Not reported (in the abstract) | In no study were pain or quality of life better than with placebo (p > 0.05). The authors write that current evidence is insufficient for routine use in chemotherapy-induced neuropathy. The study the review classified as 'oral CBD' is, based on participant numbers, most likely Haney 2025, whose capsule contained 5 mg of THC alongside 100 mg of CBD (the full text of the review could not be read). 2 of the 8 authors work for a clinic chain that prescribes medical cannabis (Curaleaf Clinic, London) Side effects: Mild to moderate; most often tiredness, dizziness and dry mouth | PMID 42085369, doi:10.1080/15360288.2026.2664668 | |
| Hoggart 2015 Multicenter, open-label follow-up study (extension of two RCTs); randomized withdrawal phase in a subgroup | 380 (234 completed, 62%) Peripheral neuropathic pain from diabetes or with pain triggered by touch (allodynia); patients who completed the two parent RCTs | Oromucosal spray: THC:CBD (Sativex), added to current treatment | Up to 24 sprays a day (THC 65 mg + CBD 60 mg; registry) | 38 weeks | No comparison group. Pain fell from 6.9 points at the start of the parent studies to 4.2 at the end of follow-up (0–10 scale); at every assessment, at least half of the patients reported 30% or greater relief; the dose was not increased over time. In the 19-person randomized withdrawal part in the registry, there was no difference in pain between those who continued the drug and those switched to placebo (p = 0.27). 38% did not complete follow-up; the results reflect those who continued. The patients come from two parent RCTs; the registry does not name them, but the patient group, period and dosing rule match the diabetic neuropathy (NCT00710424) and allodynic pain (Serpell 2014) studies in the table, so the patients may largely be from those two studies. Manufacturer-supported Side effects: Serious side effects in 40 of 380 people; non-serious side effects in 295 (registry); according to the authors, no new safety concerns | PMID 25270679, doi:10.1007/s00415-014-7502-9, NCT00713323, randomized withdrawal), results in the registry, sponsor is the drug’s manufacturer |
| Ueberall 2022 (German Pain e-Registry) Retrospective, propensity score–matched registry study (comparison of two drugs) | 674 (337 THC:CBD spray, 337 dronabinol) Outpatients with severe neuropathic pain who had not responded adequately to established treatments (Germany) | THC:CBD oromucosal spray (nabiximols) or oral dronabinol (synthetic THC), added to current treatment | Mean THC: 16.6 mg/day in the spray group, 17.2 mg/day in the dronabinol group | 24 weeks | A composite score of 9 pain- and health-related measures improved by 55.4% in the spray group and 40.5% in the dronabinol group (difference 14.0; 95% CI 12.6–15.4). No placebo group; retrospective registry data. The analysis was funded by a drug company (Almirall); all 4 of the 4 authors have financial ties to this company (one is a full-time employee) Side effects: Drug-related side effects 21.1% vs. 35%; discontinuation due to side effects 5.9% vs. 14.8% | PMID 35140513, doi:10.2147/JPR.S340968, Other analyses of the same registry: Ueberall 2019, PMID 31190969, 497 with neuropathic pain, no control group), Ueberall 2022, PMID 34480564, 655 + 655) |
| Hjorthøj 2022 (Danish national registry) Cohort using national registry data; comparison with propensity score–matched controls | 1,817 people receiving prescription cannabinoids for neuropathic pain (plus 924 for other pain) and 1 matched control for each Patients with neuropathic pain who received a prescription in Denmark’s medical cannabis pilot program, which started in 2018 | Prescription cannabinoid medicines and medical cannabis: CBD, THC or CBD + THC (separate analysis by class; distribution not in the abstract) | Not reported (in the abstract) | Not reported (in the abstract) | In neuropathic pain, those taking CBD, THC or both used less gabapentin and spent fewer days in the hospital than controls; those using THC, however, used more opioids than controls. Pain intensity was not measured; observational data, confounding cannot be ruled out. No similar benefit was seen in other types of pain Side effects: The authors found it generally safe; no details in the abstract | PMID 34624164, doi:10.1002/ejp.1874, PMID 34743399 |
| Evans 2025 (UK Medical Cannabis Registry, complex regional pain syndrome) Case series (from a prospective patient registry; no comparison group) | 64 Patients treated for complex regional pain syndrome who had entered the registry at least 6 months earlier; 56% were already using cannabis at baseline (United Kingdom, private clinic) | Prescription cannabis products chosen by the physician: extracts in MCT oil (CBD-dominant and THC-dominant) and THC-dominant dried flower; at baseline, 28 people used oil only, 6 flower only and 30 both | Median prescribed daily amount at baseline THC 12.5 mg and CBD 11 mg; at month 6, THC 117.9 mg and CBD 20 mg At the physician’s discretion | 6 months (measurements at baseline and at months 1, 3 and 6) | Pain intensity (BPI, 0–10) fell from 6.69 to 5.85 at month 1 and to 6.05 at month 6 (p < 0.05; missing data imputed with baseline values); a clinically important improvement at month 6 was seen in 28% of patients on the BPI and 41% on the SF-MPQ-2. Improvement from baseline was reported on anxiety, sleep and quality-of-life scales; the patient global impression of change (PGIC) did not change significantly. Because there was no comparison group, it cannot be known how much of the improvement came from the product; this is not evidence of effectiveness. 7 of the 11 authors work for the clinic chain (Curaleaf) that treated the registry patients; the most commonly prescribed oils and flower are also products of the same group of companies Side effects: 5 patients (7.8%) reported a total of 50 side effects (20 severe, none life-threatening); most often nausea and headache (4 times each) | PMID 40898690, doi:10.1002/brb3.70823, PMC12405601 |
Reported side effects
- Nabiximols: nervous system adverse events 61% vs. 29%, psychiatric 17% vs. 5% (Cochrane 2018)
- Discontinuation with Sativex 18% vs. 3% (Nurmikko 2007)
- High dose (800 mg/day): side effects reported in 68% with CBD, 53% with placebo; most often tiredness/drowsiness, nausea, diarrhea (Robertson 2026)
- Cochrane 2026: nervous system side effects with THC-dominant products RD 0.25 (95% CI 0.14–0.37; low certainty), with balanced THC:CBD RD 0.39 (0.23–0.55; very low); discontinuation due to side effects with the balanced product RD 0.05 (0.02–0.09; low); no difference in nervous system or psychiatric side effects with CBD-dominant products (low certainty)
- Unpublished manufacturer studies: in diabetic neuropathy, side effects 120/149 vs. 101/148, serious 14 vs. 12 (NCT00710424); in spinal cord injury 46/56 vs. 29/60 (NCT01606202)
Monitoring done in studies
- Epidiolex label: at 10–20 mg/kg/day, ALT >3×ULN in 13% (placebo 1%); risk rises markedly with valproate; monitor at baseline, at months 1, 3 and 6, and after dose changes
- In healthy adults, 5 mg/kg/day led to ALT/AST >3×ULN in 5.6% (placebo 0%) and eosinophilia (Florian 2025, PMID 40622698); 1,500 mg/day led to >5×ULN in 31% (Watkins 2021, PMID 33022751)
- Liver enzymes should be monitored with acetaminophen (paracetamol) and other medicines that strain the liver (Pramhas 2023, PMID 38033459)
Who should be careful
- CBD is not an approved medicine for any type of pain
- Use caution with CNS depressants because of the risk of sedation, falls and impaired driving
Drug interactions
- Opioids: no increase in side effects with fentanyl + 400–800 mg CBD; 'bad drug effect' increased with hydromorphone + 200 mg CBD; pain threshold decreased with methadone + 1,200 mg CBD. An opioid-sparing effect has not been proven; monitor for sedation
- CNS depressants (gabapentin, pregabalin, benzodiazepines, alcohol): additive sedation; somnolence with Epidiolex 32% vs. 11%. No direct clinical study of CBD + gabapentinoids has been reported
- NSAIDs: the label shows no significant change in tolbutamide, a CYP2C9 substrate; no direct study of CBD + NSAIDs has been reported; no documented significant interaction
- Acetaminophen (paracetamol): in knee OA, enzyme elevations were more frequent with acetaminophen 3 g + CBD 600 mg; monitor liver enzymes
- Warfarin: in a case report, INR rose and the warfarin dose was reduced; monitor INR when starting or stopping CBD
- Clopidogrel: antiplatelet effect may be reduced via CYP2C19 (label); consult cardiology
- DOACs (apixaban, rivaroxaban): no direct data reported; uncertain, should be disclosed and monitored
- Valproate (liver, high risk) and clobazam (sedation 46% vs. 16%)
- Tizanidine, theophylline (CYP1A2): the label recommends considering a dose reduction
- Tacrolimus, sirolimus, everolimus, digoxin: monitoring or dose reduction (label)
- Strong inducers such as rifampin: CBD levels fall by ~32%
- Note: label data come from high-dose pharmaceutical CBD; with nonprescription products the size of an interaction may differ, but its direction should be assumed to be the same
Abbreviations: RCT, randomized controlled trial (a study in which participants are assigned to groups by chance); placebo, an inactive substance; SMD, OR and RR, effect measures used in analyses that pool studies; CYP, liver enzymes that break down medicines.
If you are considering CBD
- You can talk with your doctor about the cause of your nerve pain and the treatment options that suit you.
- If you take gabapentin, pregabalin, an opioid or a sleep medicine, talk with your doctor and pharmacist about the fact that CBD can increase drowsiness.
- If you take a blood thinner, be sure to mention it.
- If a product is being considered, check its label and its independent certificate of analysis; find out whether it contains THC.
- Don’t stop your current medicines on your own.
Choose your age group, medicines and situation to prepare a short summary you can take to your doctor. Nothing you enter is sent anywhere.
Bottom line
It has not been shown that pure CBD taken by mouth at low doses works better than an inactive substance for nerve pain; the benefit seen in the single small study with a very high dose was small and needs to be confirmed. There are many studies of products containing THC; although short studies show a small benefit, this benefit becomes uncertain in long and large studies, and side effects are more common. The best course is to weigh the treatment options for your long-lasting pain together with your doctor.
Frequently asked questions
Does CBD oil work for neuropathic pain?
The evidence is weak and the results conflict. In two well-designed studies that tested low doses of pure CBD by mouth (45–50 mg a day; study dose, not a personal recommendation), CBD was no better than an inactive substance (placebo). In a small study of nerve pain from spinal cord injury in which the dose was raised to 800 mg a day, pain was slightly lower; this is a single, small result. In an older 24-person study in people with neurological disease, pain had also decreased more with a CBD-rich plant extract than with placebo; but the patients were not selected for nerve pain and the THC content of the extract is not known. There are two small studies of CBD applied to the skin: one with weak methods reported a positive result, while the other, using pure CBD cream in chemotherapy-related nerve damage, found no difference from an inactive cream (placebo).
Does CBD work for nerve pain from diabetes?
There is no published study of pure CBD by mouth in nerve pain from diabetes; in the study that looked at different kinds of nerve pain together, CBD by mouth was not found to be better than placebo. There are a few studies of products containing THC: a mouth spray containing THC and CBD was no better than placebo in a 30-person study and in a 297-person study whose results are only in the trial registry; in a 16-person study of inhaled cannabis using single-dose sessions, pain eased in the short term; and in a 26-person study of the THC-like drug nabilone that selected people who had responded to the drug at the start, a benefit was seen. In a 12-week study of 100 people using a THC-dominant cannabis oil applied to the skin, pain symptoms decreased much more than with placebo; this single-center result needs to be confirmed by other studies. A 2026 review writes that the studies on this topic are small and differ from one another, and that no firm conclusion can be drawn.
Are products containing THC more effective for nerve pain?
Many more studies have been done with products containing THC than with pure CBD, and in short studies pain decreased slightly more. But in most of the long and large studies there was no clear difference from an inactive substance. The Cochrane review updated in 2026 found no clear effect on at least half the pain being relieved, either with THC-dominant products or with products containing equal amounts of THC and CBD; with the equal-ratio product, slightly more people may have had at least 30% relief (low certainty of evidence). Nervous system side effects (such as dizziness and drowsiness) are more common with products containing THC.
Can CBD be used together with gabapentin or pregabalin?
No study has been reported that directly looked at using CBD together with these medicines. But because they can all cause drowsiness, using them together may increase the risk of sleepiness and falls. It’s important to talk to your doctor and pharmacist.
Can CBD oil be obtained with a prescription for neuropathic pain in Turkey?
In Turkey, cannabis-derived products are governed by the Kenevirden Elde Edilen Ürünlere Dair Yönetmelik (Regulation on Products Derived from Hemp) dated January 31, 2026. These products, including CBD, can be sold only in pharmacies. Those classed as medicines and health products are dispensed with a prescription; for supplement products and cosmetics with a THC content of no more than 0.3%, the regulation does not state a prescription requirement. As of October 2026, there is no approved (licensed) CBD medicine in Turkey. If you are thinking of using it, talk with your doctor first.
References
Show 47 references
- Zubcevic 2023. Double-blind RCT, 4 arms (CBD / THC / CBD+THC / placebo). PMID 36571471, doi:10.1002/ejp.2072, EudraCT 2017-005198-38
- Robertson 2026. Double-blind, placebo-controlled crossover RCT (4-week washout). PMID 42256679, doi:10.1016/j.eclinm.2026.103986, ANZCTR ACTRN12622000634774
- Hansen 2023. Multicenter, double-blind, placebo-controlled, 4-arm parallel RCT (THC / CBD / THC+CBD / placebo). PMID 37630995, doi:10.3390/ph16081079, EudraCT 2018-002315-98, protocol: Hansen 2021, PMID 34573231, pharmacokinetic substudy in 23 MS patients: Hansen 2024, PMID 38054364
- Xu 2020. Placebo-controlled RCT (followed by crossover). PMID 31793418, doi:10.2174/1389201020666191202111534
- D'Andre 2024. Double-blind, placebo-controlled, crossover pilot RCT. PMID 39016024, doi:10.1089/can.2023.0253, NCT05388058
- Wade 2003. Double-blind, placebo-controlled, randomized series of single-patient crossover trials (each patient tried four products for 2 weeks each). PMID 12617376, doi:10.1191/0269215503cr581oa
- Parise 2026 (burning mouth syndrome). Single-center, single-blind, placebo-controlled RCT. PMID 42951760, doi:10.1016/j.oooo.2026.07.010, PMID 33123730, 17 patients, cannabis flower extract, 4 weeks)
- Nurmikko 2007. Double-blind, placebo-controlled RCT. PMID 17997224, doi:10.1016/j.pain.2007.08.028, NCT00711880, a 52-week open-label extension is mentioned in the abstract
- Eibach 2021. Double-blind, placebo-controlled, crossover RCT (3-week washout). PMID 32770831, doi:10.1002/cpt.2016
- Berman 2004. Double-blind, placebo-controlled, three-period crossover RCT. PMID 15561385, doi:10.1016/j.pain.2004.09.013, NCT01606189, sponsor is the drug’s manufacturer)
- Rog 2005. Single-center, double-blind, placebo-controlled, parallel RCT. PMID 16186518, doi:10.1212/01.wnl.0000176753.45410.8b, NCT01604265, sponsor is the drug’s manufacturer). Open-label 2-year extension: Rog 2007, PMID 18035205, 28 completed 2 years, at least one drug-related side effect in 58)
- Selvarajah 2010. Double-blind, placebo-controlled, parallel RCT. PMID 19808912, doi:10.2337/dc09-1029, NCT00238550, the publication gives no registration number)
- NCT00710424 (manufacturer’s diabetic neuropathy study, registry results). Double-blind, placebo-controlled, parallel phase 3 RCT; results only in the registry (no peer-reviewed publication). NCT00710424, results in the registry, July 31, 2012, sponsor is the drug’s manufacturer, and the registry states that publications must be submitted for company review, no publication found on PubMed in October 2026)
- NCT01606202 (manufacturer’s spinal cord injury study, registry results). Double-blind, placebo-controlled, parallel phase 3 RCT; results only in the registry (no peer-reviewed publication). NCT01606202, results in the registry, September 28, 2012, sponsor is the drug’s manufacturer, in October 2026, mentioned on PubMed only in a scoping review, PMID 33465022
- Langford 2013. Multicenter, double-blind, placebo-controlled, parallel phase 3 RCT (phase A); followed by 14 weeks open-label + 4 weeks double-blind randomized withdrawal (phase B). PMID 23180178, doi:10.1007/s00415-012-6739-4, NCT00391079, sponsor is the drug’s manufacturer)
- Serpell 2014. Multicenter, double-blind, placebo-controlled, parallel RCT. PMID 24420962, doi:10.1002/j.1532-2149.2013.00445.x, NCT00710554, sponsor is the drug’s manufacturer)
- Lynch 2014. Double-blind, placebo-controlled, crossover pilot RCT; followed by an open-label extension. PMID 23742737, doi:10.1016/j.jpainsymman.2013.02.018, NCT00872144
- Svendsen 2004. Double-blind, placebo-controlled, crossover RCT (3-week washout). PMID 15258006, doi:10.1136/bmj.38149.566979.AE, Second publication in Danish: Svendsen 2005, PMID 16014264
- Rintala 2010. Double-blind, active-controlled (diphenhydramine), crossover pilot RCT. PMID 20855984, doi:10.1097/PHM.0b013e3181f1c4ec
- Schimrigk 2017. Double-blind, placebo-controlled, parallel phase 3 RCT; followed by a 32-week open-label period. PMID 29073592, doi:10.1159/000481089, NCT00959218, no results in the registry)
- Weizman 2018. Double-blind, placebo-controlled, within-subject two-session single-dose RCT; brain MRI. PMID 30185448, doi:10.1212/WNL.0000000000006293, The same team’s 12-male-patient publication with the same design: Weizman 2024, PMID 38597988, THC 0.2 mg/kg, pain on a 0–100 scale from 53 to 35 with THC and from 52 to 44 with placebo, p < 0.05 vs. placebo). Whether the participants of the two publications are the same is not stated
- Abrams 2007. Randomized, placebo-controlled, parallel study (inpatient). PMID 17296917, doi:10.1212/01.wnl.0000253187.66183.9c, NCT00046722
- Wilsey 2008. Double-blind, placebo-controlled, crossover RCT (single-session laboratory trials). PMID 18403272, doi:10.1016/j.jpain.2007.12.010, NCT00254761
- Ellis 2009. Double-blind, placebo-controlled, crossover phase 2 RCT (2-week washout). PMID 18688212, doi:10.1038/npp.2008.120, NCT00255580, PMID 22133305, during the cannabis week, plasma ghrelin and leptin rose, PYY fell, and insulin fell similarly to placebo)
- Ware 2010. Double-blind, placebo-controlled, 4-period crossover RCT. PMID 20805210, doi:10.1503/cmaj.091414, ISRCTN68314063
- Wilsey 2013. Double-blind, placebo-controlled, crossover RCT (single-day laboratory trials). PMID 23237736, doi:10.1016/j.jpain.2012.10.009, NCT01037088
- Wallace 2015. Double-blind, placebo-controlled, 4-period crossover RCT (single-session trials, 2-week washout). PMID 25843054, doi:10.1016/j.jpain.2015.03.008, NCT00781001, secondary analysis: Wallace 2020, PMID 32565122
- Wilsey 2016. Double-blind, placebo-controlled, crossover RCT (8-hour laboratory sessions). PMID 27286745, doi:10.1016/j.jpain.2016.05.010, NCT01555983, blood level analysis of the same trial: Wilsey 2016, PMID 27621666
- Almog 2020. Double-blind, placebo-controlled, three-arm crossover RCT (single-inhalation dose). PMID 32445190, doi:10.1002/ejp.1605
- NCT02460692 (dronabinol and vaporized cannabis, registry results). Randomized, triple-blind, placebo-controlled, three-arm parallel phase 2 study (everyone received vapor and capsules); results only in the registry (no peer-reviewed publication). NCT02460692, results in the registry, September 28, 2023, no publication found on PubMed in October 2026)
- Haney 2025. Double-blind, placebo-controlled, parallel pilot RCT. PMID 40611810, doi:10.1089/can.2025.0028, NCT03782402, results in the registry)
- Weiss 2025. Double-blind, placebo-controlled, parallel RCT; followed by a 12-week open-label follow-up. PMID 41211445, doi:10.3389/fonc.2025.1590168, NCT04398446, the authors declared no conflicts of interest
- Seevathee 2025. Double-blind, placebo-controlled, parallel phase 3 RCT (single center). PMID 39720705, doi:10.1159/000542511, Thai Clinical Trials Registry TCTR20211220008
- Kittithamvongs 2025. Triple-blind, placebo-controlled, two-period crossover RCT (14-day washout). PMID 39115545, doi:10.1097/CORR.0000000000003221, Thai Clinical Trials Registry TCTR20191106004 (the registry page returned no content in October 2026), commentary: Hoffmeyer 2025, PMID 39330963
- Frank 2008. Double-blind, active-controlled (dihydrocodeine, a weak opioid), crossover RCT (2-week washout). PMID 18182416, doi:10.1136/bmj.39429.619653.80, ISRCTN15330757
- Toth 2012. Double-blind, placebo-controlled, flexible-dose RCT with enriched enrollment and randomized withdrawal design (everyone first took single-blind nabilone for 4 weeks). PMID 22921260, doi:10.1016/j.pain.2012.06.024, A registration by the same institution with the same design, NCT01035281, 4 weeks single-blind nabilone, responders to the double-blind period, no results in the registry, the publication gives no registration number)
- Turcotte 2015. Double-blind, placebo-controlled, parallel RCT. PMID 25288189, doi:10.1111/pme.12569, NCT00480181, no results in the registry, the drug’s marketer Bausch Health is a collaborator in the registry)
- Ateş 2026 (Cochrane). Systematic review and meta-analysis (Cochrane; update of the 2018 review). PMID 41548880, doi:10.1002/14651858.CD012182.pub3, Previous version: Mücke 2018 (Cochrane, in this table)
- Mücke 2018 (Cochrane). Systematic review and meta-analysis. PMID 29513392, doi:10.1002/14651858.CD012182.pub2, Update: Ateş 2026 (Cochrane, PMID 41548880, in this table)
- McParland 2023 (meta-analysis). Systematic review and meta-analysis. PMID 36598058, doi:10.1136/rapm-2021-103431, PROSPERO CRD42017074255
- Andreae 2015 (individual patient data meta-analysis). Systematic review and individual patient data meta-analysis (Bayesian method). PMID 26362106, doi:10.1016/j.jpain.2015.07.009, PROSPERO CRD42011001182, the 5 included studies: Abrams 2007, Ellis 2009, Ware 2010, Wilsey 2008 and Wilsey 2013 (all in this table), 4 of the 12 authors are investigators of these studies
- Sherman 2026 (systematic review). Systematic review of RCTs (no meta-analysis). PMID 41714301, doi:10.1177/25785125261425444
- Datta 2026 (systematic review). Systematic review of RCTs (meta-analysis not possible). PMID 42085369, doi:10.1080/15360288.2026.2664668
- Hoggart 2015. Multicenter, open-label follow-up study (extension of two RCTs); randomized withdrawal phase in a subgroup. PMID 25270679, doi:10.1007/s00415-014-7502-9, NCT00713323, randomized withdrawal), results in the registry, sponsor is the drug’s manufacturer
- Ueberall 2022 (German Pain e-Registry). Retrospective, propensity score–matched registry study (comparison of two drugs). PMID 35140513, doi:10.2147/JPR.S340968, Other analyses of the same registry: Ueberall 2019, PMID 31190969, 497 with neuropathic pain, no control group), Ueberall 2022, PMID 34480564, 655 + 655)
- Hjorthøj 2022 (Danish national registry). Cohort using national registry data; comparison with propensity score–matched controls. PMID 34624164, doi:10.1002/ejp.1874, PMID 34743399
- Evans 2025 (UK Medical Cannabis Registry, complex regional pain syndrome). Case series (from a prospective patient registry; no comparison group). PMID 40898690, doi:10.1002/brb3.70823, PMC12405601
Points that could not be verified from the full text
- Xu 2020 (amount applied, THC) was read only from the abstract; the maximum daily dose in Nurmikko 2007 is contradictory in the registry (NCT00711880: 24 sprays in the intervention description, 48 in the results section), and the full text of the publication could not be read
- The full text of the Cochrane 2026 update (which 21 studies, and in particular which 5 CBD-dominant studies, were included) could not be read because of the Cochrane Library’s bot verification; the numbers are from the review’s abstract
- The Thai registry page (TCTR20191106004) for the brachial plexus injury study (Kittithamvongs 2025) still returned no content on October 28, 2026; the registered plan could not be compared with the publication (product and dose information was read from the full text of the publication)
- There are 5 randomized studies that appear completed on ClinicalTrials.gov but whose results were found neither in the registry nor on PubMed as of October 2026: nabilone in phantom limb pain with 50 people planned (ended in 2011; actual number not in the registry) and in spinal cord injury with 12 people (2015), a metered-dose inhaler in diabetic neuropathy with 122 people (ended June 2026), nabilone and diet in spinal cord injury with 14 people (2024), and CBD/CBDA with 6 people (2026)
- Dose information is missing in two studies of products applied to the skin: in the pure CBD cream study (D'Andre 2024), the amount of CBD in the cream is not in the abstract or the registry, and the full text is not open access; in the THC-dominant oil study (Seevathee 2025), the daily frequency of application is not stated in the full text
- Of the guidelines in the evidence summary, the 2025 NeuPSIG update and the European Pain Federation’s 2018 position paper were read only from the abstract (the full texts required a subscription or the publisher’s page); for this reason, NeuPSIG 2025’s detailed decision on cannabinoids and the content of its published correction are not known
- The full text of Wade 2003 could not be read (paywalled): the THC content of the CBD-rich extract and the size of the effect on pain for each product are not known. The two studies in burning mouth syndrome (Gambino 2021, Parise 2026) were read from the abstract only; for Parise 2026, the THC content of the product, the length of treatment and who was blinded are not in the abstract
Trusted sources from around the world
Show 1 reference
- Cochrane: Cannabis-based medicines for chronic nerve painCochrane, English, January 2026
A plain-language summary of an independent systematic review: benefits and risks, and the certainty of the evidence.
These sources sell no products. All trusted sources
Research history
Here we record what changed on this page, when and why, with dates. The newest change is at the top.
The last sentence of the page's short summary was rewritten in calmer, more guiding language: instead of "Talk to your doctor before you start.", it now says who you can talk with, and about what, if you are considering CBD. The result and the strength of evidence did not change.
Two studies were added to the table. A small 2003 crossover study in 24 people with persistent symptoms due to neurological diseases that also compared a CBD-rich plant extract with placebo (Wade 2003): pain had decreased more with this extract than with placebo; but because the patients were not selected for neuropathic pain, each product was tried for only 2 weeks and the THC content of the extract is not known, this finding does not count as evidence for pure CBD. No separate page was opened for burning mouth syndrome: a single-center, single-blind study comparing full-spectrum CBD oil with placebo in 26 patients with this condition (Parise 2026) was added to the table, and a 17-person study without a control group was added to the research section; both are very small and the evidence is very weak. The result and the strength of evidence did not change.
Related studies: Wade 2003, published February 2003 (PMID 12617376, doi:10.1191/0269215503cr581oa); Parise 2026 (burning mouth syndrome), published July 24, 2026 (online) (PMID 42951760, doi:10.1016/j.oooo.2026.07.010, PMID 33123730, 17 patients, cannabis flower extract, 4 weeks))
October 2026 additional review: sources found for this page during searches for other topics were read and added to the page. The full text of the brachial plexus injury study (Kittithamvongs 2025) was read: the product tested was a cannabis oil containing roughly equal amounts of THC and CBD, dropped under the tongue; the product type changed from 'Other' to 'Contains THC', dose information was added, and the number of randomized studies with products containing THC rose from 27 to 28. Added: the numerical inconsistencies in the publication to the study of an oil applied to the skin in nerve pain from diabetes (Seevathee 2025); the authors’ company ties, and the fact that the study the review counts as 'oral CBD' most likely also contained a small amount of THC, to the review of chemotherapy-induced neuropathy (Datta 2026); and the appetite hormone substudy of the same trial to the HIV-associated neuropathy study (Ellis 2009). A separate search was done for complex regional pain syndrome and a short section was added: there is no controlled study specific to this condition; three short studies in mixed nerve pain groups also included these patients, and their numbers were written in two rows; a 64-patient series with no comparison group from the UK medical cannabis registry was added to the table. The views of guidelines and expert bodies were added to the evidence summary: both those recommending against use and those writing that cannabinoids may be considered when other treatments have not been enough. An uncontrolled CBD study in chemotherapy-induced neuropathy (Vigano 2025) was not added to the table: patients had not been selected for nerve pain, and no pain result was reported for the period when CBD was given alone. The conclusion and the strength of evidence did not change.
Related studies: Evans 2025 (UK Medical Cannabis Registry, complex regional pain syndrome), published September 2025 (PMID 40898690, doi:10.1002/brb3.70823, PMC12405601)
Older changes (4)
In October 2026, PubMed, Cochrane and ClinicalTrials.gov were searched systematically; 38 studies were added: 24 randomized studies of products containing THC (3 with results only in the trial registry: the drug maker’s diabetic neuropathy and spinal cord injury studies, and a study of dronabinol and inhaled cannabis in nerve-related low back pain; one with a THC-dominant oil applied to the skin in diabetic neuropathy), 1 randomized study with pure CBD cream applied to the skin, 3 with the THC-like drug nabilone, 1 with CBDV, a cannabinoid closely related to CBD, and 1 with a cannabis medicine of unreported composition, 5 systematic reviews including the 2026 Cochrane update, and 3 real-world studies. The 2018 numbers from the Cochrane review were replaced with those from the 2026 update; the conclusion and strength-of-evidence text for products containing THC were rewritten. The phrase 'a single study' for CBD applied to the skin was corrected: there is also a second study with a negative result. For two existing four-arm studies, the results of the THC arms were added; it was noted that the maximum dose of the mouth spray study (Nurmikko 2007) is contradictory in the registry; the product type of the 2018 Cochrane review became 'Multiple product types' and that of the CBD oil applied to the skin became 'CBD-rich extract, THC content unclear'. The conclusion and the strength of evidence did not change.
Related studies: Eibach 2021, published April 2021 (PMID 32770831, doi:10.1002/cpt.2016); Berman 2004, published December 2004 (PMID 15561385, doi:10.1016/j.pain.2004.09.013, NCT01606189, sponsor is the drug’s manufacturer)); Rog 2005, published September 27, 2005 (PMID 16186518, doi:10.1212/01.wnl.0000176753.45410.8b, NCT01604265, sponsor is the drug’s manufacturer). Open-label 2-year extension: Rog 2007, PMID 18035205, 28 completed 2 years, at least one drug-related side effect in 58)); Selvarajah 2010, published January 2010 (PMID 19808912, doi:10.2337/dc09-1029, NCT00238550, the publication gives no registration number)); NCT00710424 (manufacturer’s diabetic neuropathy study, registry results), published July 31, 2012 (registry results) (NCT00710424, results in the registry, July 31, 2012, sponsor is the drug’s manufacturer, and the registry states that publications must be submitted for company review, no publication found on PubMed in October 2026)); NCT01606202 (manufacturer’s spinal cord injury study, registry results), published September 28, 2012 (registry results) (NCT01606202, results in the registry, September 28, 2012, sponsor is the drug’s manufacturer, in October 2026, mentioned on PubMed only in a scoping review, PMID 33465022); Langford 2013, published April 2013 (PMID 23180178, doi:10.1007/s00415-012-6739-4, NCT00391079, sponsor is the drug’s manufacturer)); Serpell 2014, published August 2014 (PMID 24420962, doi:10.1002/j.1532-2149.2013.00445.x, NCT00710554, sponsor is the drug’s manufacturer)); Lynch 2014, published January 2014 (PMID 23742737, doi:10.1016/j.jpainsymman.2013.02.018, NCT00872144); Svendsen 2004, published July 31, 2004 (PMID 15258006, doi:10.1136/bmj.38149.566979.AE, Second publication in Danish: Svendsen 2005, PMID 16014264); Rintala 2010, published October 2010 (PMID 20855984, doi:10.1097/PHM.0b013e3181f1c4ec); Schimrigk 2017, published October 2017 (PMID 29073592, doi:10.1159/000481089, NCT00959218, no results in the registry)); Weizman 2018, published October 2, 2018 (PMID 30185448, doi:10.1212/WNL.0000000000006293, The same team’s 12-male-patient publication with the same design: Weizman 2024, PMID 38597988, THC 0.2 mg/kg, pain on a 0–100 scale from 53 to 35 with THC and from 52 to 44 with placebo, p < 0.05 vs. placebo). Whether the participants of the two publications are the same is not stated); Abrams 2007, published February 13, 2007 (PMID 17296917, doi:10.1212/01.wnl.0000253187.66183.9c, NCT00046722); Wilsey 2008, published June 2008 (PMID 18403272, doi:10.1016/j.jpain.2007.12.010, NCT00254761); Ellis 2009, published February 2009 (PMID 18688212, doi:10.1038/npp.2008.120, NCT00255580, PMID 22133305, during the cannabis week, plasma ghrelin and leptin rose, PYY fell, and insulin fell similarly to placebo)); Ware 2010, published October 5, 2010 (PMID 20805210, doi:10.1503/cmaj.091414, ISRCTN68314063); Wilsey 2013, published February 2013 (PMID 23237736, doi:10.1016/j.jpain.2012.10.009, NCT01037088); Wallace 2015, published July 2015 (PMID 25843054, doi:10.1016/j.jpain.2015.03.008, NCT00781001, secondary analysis: Wallace 2020, PMID 32565122); Wilsey 2016, published September 2016 (PMID 27286745, doi:10.1016/j.jpain.2016.05.010, NCT01555983, blood level analysis of the same trial: Wilsey 2016, PMID 27621666); Almog 2020, published September 2020 (PMID 32445190, doi:10.1002/ejp.1605); NCT02460692 (dronabinol and vaporized cannabis, registry results), published September 28, 2023 (registry results) (NCT02460692, results in the registry, September 28, 2023, no publication found on PubMed in October 2026)); Haney 2025, published October 2025 (PMID 40611810, doi:10.1089/can.2025.0028, NCT03782402, results in the registry)); Weiss 2025, published October 2025 (PMID 41211445, doi:10.3389/fonc.2025.1590168, NCT04398446, the authors declared no conflicts of interest); Kittithamvongs 2025, published February 1, 2025 (PMID 39115545, doi:10.1097/CORR.0000000000003221, Thai Clinical Trials Registry TCTR20191106004 (the registry page returned no content in October 2026), commentary: Hoffmeyer 2025, PMID 39330963); Frank 2008, published January 26, 2008 (PMID 18182416, doi:10.1136/bmj.39429.619653.80, ISRCTN15330757); Toth 2012, published October 2012 (PMID 22921260, doi:10.1016/j.pain.2012.06.024, A registration by the same institution with the same design, NCT01035281, 4 weeks single-blind nabilone, responders to the double-blind period, no results in the registry, the publication gives no registration number)); Turcotte 2015, published January 2015 (PMID 25288189, doi:10.1111/pme.12569, NCT00480181, no results in the registry, the drug’s marketer Bausch Health is a collaborator in the registry)); Ateş 2026 (Cochrane), published January 19, 2026 (PMID 41548880, doi:10.1002/14651858.CD012182.pub3, Previous version: Mücke 2018 (Cochrane, in this table)); McParland 2023 (meta-analysis), published April 2023 (PMID 36598058, doi:10.1136/rapm-2021-103431, PROSPERO CRD42017074255); Andreae 2015 (individual patient data meta-analysis), published December 2015 (PMID 26362106, doi:10.1016/j.jpain.2015.07.009, PROSPERO CRD42011001182, the 5 included studies: Abrams 2007, Ellis 2009, Ware 2010, Wilsey 2008 and Wilsey 2013 (all in this table), 4 of the 12 authors are investigators of these studies); Sherman 2026 (systematic review), published April 2026 (PMID 41714301, doi:10.1177/25785125261425444); Datta 2026 (systematic review), published September 2026 (PMID 42085369, doi:10.1080/15360288.2026.2664668); Hoggart 2015, published January 2015 (PMID 25270679, doi:10.1007/s00415-014-7502-9, NCT00713323, randomized withdrawal), results in the registry, sponsor is the drug’s manufacturer); Ueberall 2022 (German Pain e-Registry), published February 2022 (PMID 35140513, doi:10.2147/JPR.S340968, Other analyses of the same registry: Ueberall 2019, PMID 31190969, 497 with neuropathic pain, no control group), Ueberall 2022, PMID 34480564, 655 + 655)); Hjorthøj 2022 (Danish national registry), published February 2022 (PMID 34624164, doi:10.1002/ejp.1874, PMID 34743399); D'Andre 2024, published December 2024 (PMID 39016024, doi:10.1089/can.2023.0253, NCT05388058); Seevathee 2025, published 2025 (PMID 39720705, doi:10.1159/000542511, Thai Clinical Trials Registry TCTR20211220008)
Newly published and previously missed clinical studies were searched on PubMed and ClinicalTrials.gov. 2 studies added: Robertson 2026 and Hansen 2023. In a small randomized study of pure CBD, pain eased slightly compared with placebo (Robertson 2026); in another randomized study, no difference was seen (Hansen 2023). With the new studies, the strength of evidence changed from "Very weak" to "Weak" and the conclusion from "No effect found" to "Mixed results."
Studies that led to the change: Robertson 2026, published June 2026 (PMID 42256679, doi:10.1016/j.eclinm.2026.103986, ANZCTR ACTRN12622000634774); Hansen 2023, published July 28, 2023 (PMID 37630995, doi:10.3390/ph16081079, EudraCT 2018-002315-98, protocol: Hansen 2021, PMID 34573231, pharmacokinetic substudy in 23 MS patients: Hansen 2024, PMID 38054364)
The full texts of the studies were reread and the information on the page was checked against its sources. Corrected: side-effect and dose-titration information in the study table. Points that could not be verified from the full text went down from 2 items to 1. The conclusion and the strength of evidence did not change.
Page published. This first review looked at 4 studies.