Secondary Logo

Share this article on:

Cannabis and cannabinoids for the treatment of people with chronic noncancer pain conditions: a systematic review and meta-analysis of controlled and observational studies

Stockings, Emilya,*; Campbell, Gabriellea; Hall, Wayne D.b,c; Nielsen, Suzannea; Zagic, Dinoa; Rahman, Rakina; Murnion, Bridind,e; Farrell, Michaela; Weier, Megana; Degenhardt, Louisaa

doi: 10.1097/j.pain.0000000000001293
Systematic Reviews and Meta-Analyses
Editor's Choice
Global Year 2018
Video
C

This review examines evidence for the effectiveness of cannabinoids in chronic noncancer pain (CNCP) and addresses gaps in the literature by: considering differences in outcomes based on cannabinoid type and specific CNCP condition; including all study designs; and following IMMPACT guidelines. MEDLINE, Embase, PsycINFO, CENTRAL, and clinicaltrials.gov were searched in July 2017. Analyses were conducted using Revman 5.3 and Stata 15.0. A total of 91 publications containing 104 studies were eligible (n = 9958 participants), including 47 randomised controlled trials (RCTs) and 57 observational studies. Forty-eight studies examined neuropathic pain, 7 studies examined fibromyalgia, 1 rheumatoid arthritis, and 48 other CNCP (13 multiple sclerosis–related pain, 6 visceral pain, and 29 samples with mixed or undefined CNCP). Across RCTs, pooled event rates (PERs) for 30% reduction in pain were 29.0% (cannabinoids) vs 25.9% (placebo); significant effect for cannabinoids was found; number needed to treat to benefit was 24 (95% confidence interval [CI] 15-61); for 50% reduction in pain, PERs were 18.2% vs 14.4%; no significant difference was observed. Pooled change in pain intensity (standardised mean difference: −0.14, 95% CI −0.20 to −0.08) was equivalent to a 3 mm reduction on a 100 mm visual analogue scale greater than placebo groups. In RCTs, PERs for all-cause adverse events were 81.2% vs 66.2%; number needed to treat to harm: 6 (95% CI 5-8). There were no significant impacts on physical or emotional functioning, and low-quality evidence of improved sleep and patient global impression of change. Evidence for effectiveness of cannabinoids in CNCP is limited. Effects suggest that number needed to treat to benefit is high, and number needed to treat to harm is low, with limited impact on other domains. It seems unlikely that cannabinoids are highly effective medicines for CNCP.

aNational Drug and Alcohol Research Centre, Faculty of Medicine, UNSW Sydney, Sydney, New South Wales, Australia

bCentre for Youth Substance Abuse Research, University of Queensland, Brisbane, Queensland, Australia

cNational Addiction Centre, Institute of Psychiatry, Psychology & Neuroscience, Kings College London, London, United Kingdom

dDiscipline of Addiction Medicine, Faculty of Medicine, University of Sydney, Sydney, New South Wales, Australia

eDrug Health Services, Concord Repatriation General Hospital, Sydney Local Health District, NSW Health, Sydney, New South Wales, Australia

Corresponding author. Address: 22-32 King St, Randwick 2031, New South Wales, Australia. Tel.: +61(2) 9385 0162; fax: +61 (2) 9385 0222. E-mail address: e.stockings@unsw.edu.au (E. Stockings).

Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Web site (www.painjournalonline.com).

Received December 07, 2017

Received in revised form April 22, 2018

Accepted May 01, 2018

Back to Top | Article Outline

1. Introduction

There has been increasing attention to the use of cannabis and cannabinoids in the treatment of chronic noncancer pain (CNCP). Changes in legislation and use globally mean that it is likely that there will be an increase in the coming years in the availability and use of cannabis and cannabinoid products for CNCP. In the United States, these products are most commonly cited for use in CNCP.48 Chronic noncancer pain conditions are prevalent and rank among the most significant causes of disability globally.30

Recent reviews of cannabis and cannabinoids for medicinal purposes have increased our knowledge in the understanding of their effectiveness on pain,55,88,93 although they are limited in the case of CNCP management and conclusions have been conflicting, with some reviews reporting moderate to large effects,48,93 whereas others have reported minimal60 or no benefit.3 Existing reviews have been limited in their searching for CNCP studies (eg, with a focus on specific types of cannabinoids2 or study designs60), and no single review has considered the following: all types of evidence; different CNCP conditions individually; potential differential effects of different cannabinoids; and the safety of cannabis for patients with CNCP. Each of these limitations reduces our understanding of the evidence for the use of cannabinoids for CNCP.

Chronic noncancer pain conditions are varied, and many people with CNCP live with complex physical and mental health comorbidities.9,70 Pain is considered by leading clinicians and researchers to be only one of a range of core outcomes that must be considered evaluating interventions for CNCP.82 The current review addresses the limitations of previous reviews and is the first to examine the evidence for the effectiveness of cannabinoids for CNCP for all study designs, all CNCP types, all types of cannabis and cannabinoids, and using the outcomes specified in the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT).82

Back to Top | Article Outline

2. Methods

2.1. Search strategy and study eligibility

To ensure full coverage of the literature, we conducted a multiphase search, comprising an initial review of reviews for cannabis and cannabinoids to treat CNCP, followed by 4 condition-specific systematic reviews.

A systematic review of reviews in October 2016 in the electronic databases MEDLINE, Embase, PsycINFO, and the Cochrane Database of Systematic Reviews to identify all reviews (and empirical studies contained within) that evaluated the evidence base for the administration of cannabis and cannabinoids to treat CNCP (PROSPERO registration CRD42016049475).

This search was supplemented by 4 systematic searches of empirical studies in July 2017 in the electronic databases MEDLINE, Embase, PsycINFO, the Cochrane Database of Systematic Reviews and clinicaltrials.gov to identify any trial that evaluated cannabis or cannabinoids in treating the specific pain conditions: neuropathic pain (PROSPERO registration: CRD42017065248), fibromyalgia (PROSPERO registration: CRD42017067057), arthritis (PROSPERO registration: CRD42017067059), and other or mixed groups of CNCP (Supplementary Material, page 6). Date of publication was restricted to between 1980 and July 2017. No restrictions were placed on language or publication type. Medline search strategies are shown in Appendix A of the supplementary appendix (available online at http://links.lww.com/PAIN/A592). Corresponding subject headings were used in each database where specialised thesauri existed.

Individual studies that were identified (N = 107) in the systematic review of reviews of cannabinoids for the treatment of pain were screened for eligibility in full by 2 independent reviewers. For reviews of empirical studies for neuropathic pain, fibromyalgia, arthritis, and CNCP, 2 reviewers independently examined titles and abstracts using the web-based systematic review program Covidence.84 All articles identified as potentially relevant (including review articles) were obtained in full and screened by 2 independent reviewers. Study screening was conducted in duplicate by 2 independent reviewers (any of G.C., E.S., M.W., D.Z., S.N., and R.R.). Interrater disagreement was resolved via consultation with an independent third reviewer (any of L.D., G.C., E.S., M.W., D.Z., and R.R.).

Back to Top | Article Outline

2.2. Types of pain conditions

We included studies that examined impacts of cannabis and cannabinoids on any CNCP condition. We followed Cochrane protocols determining studies for inclusion and extracting data; at least 80% of the patient population was required to be experiencing one of the included pain conditions (neuropathic pain, CNCP, arthritis, or fibromyalgia). If less than 80% of the sample had one of the target pain conditions but results were presented separately for the subsample experiencing one of these pain conditions, we included the study and extracted data for the target subgroup. Studies were required to examine cannabis and cannabinoids as a primary or secondary indication for pain and to measure at least 1 of our 3 primary pain outcomes: pain intensity and 30% or 50% reduction in pain.

Back to Top | Article Outline

2.3. Types of interventions

We considered studies examining tetrahydrocannabinol; cannabidiol; combination of tetrahydrocannabinol + cannabidiol; plant-based cannabis (eg, Cannabis sativa); and other cannabinoids, eg, tetrahydrocannabinolic acid (THCA), cannabidiolic acid, cannabidivarin, and the synthetic delta-9-tetrahydrocannabinol formulations nabilone and dronabinol.

Back to Top | Article Outline

2.4. Types of studies

We included randomised controlled trials (RCTs), nonrandomised controlled trials, quasi-experimental, before and after studies, prospective and retrospective cohort studies, case–control studies, analytical cross-sectional studies, observational studies, self-report, and N-of-1 studies. For studies with a comparison group, we considered any type of comparator, including placebo groups, waitlist controls, and other interventions.

Back to Top | Article Outline

2.5. Outcomes

Guided by the IMMPACT core outcome domains for clinical trials in CNCP,82 we grouped the outcomes of interest into 6 categories: pain intensity, physical functioning, emotional functioning, global impression of change, adverse events (AEs), and withdrawals. We assessed the clinical significance of the changes by extracting data for a 30% reduction in pain (a “moderate” effect) and a 50% reduction in pain (a “substantial” effect).24

Back to Top | Article Outline

2.6. Assessment of risk of study bias

We used the Cochrane Collaboration risk of bias tool for RCTs.36 Randomised controlled trials were judged to have an overall “low risk” of bias if they had 6 to 8 risk domains rated as having a low risk of bias, “unclear risk” if 4 or more domains were judged as being unclear, and “high risk” if 3 or more domains were judged as being high risk. We additionally examined risk of bias because of sample size, where studies comprising at least 100 participants per treatment arm were classified as “low risk,” studies comprising 30 to 100 per arm were classified as “unclear risk,” and studies comprising <30 participants per arm were classified as “high risk.” Observational studies or case study reports were evaluated using an adapted version of the Cochrane Collaboration risk of bias in nonrandomised studies of interventions (ROBINS-I) assessment tool.76 Overall, risk of bias was determined by the most serious risk of bias allocated to that study across the tool.

Back to Top | Article Outline

2.7. Grading of evidence

As the review included RCTs and observational trials, we used an adapted version of the standard Grades of Recommendation, Assessment, Development and Evaluation (GRADE) tool to grade the overall study methodology.63 Randomised controlled trials began with a high rating that was downgraded if important limitations were identified in the study methodology. Observational trials began with a low rating and were upgraded if important strengths were identified. We additionally conducted a GRADE assessment using GRADEPro (https://gradepro.org/) for each reported pooled estimate that evaluated the risk of bias, inconsistency, indirectness, imprecision, and publication bias (through visual inspection of funnel plots).

Back to Top | Article Outline

2.8. Data extraction

We extracted details on the participants, interventions, comparisons, outcomes, and study design (PICOS) of each study, including: sample N, age, sex, medical and pain condition(s), length and type of treatment (including route of administration, place in therapeutic hierarchy, dose, and cointerventions), comparator type, study country, year, and design. Outcomes were extracted following IMMPACT recommendations. When data were not reported in full, we contacted authors for additional information. When studies reported multiple measures of a single domain (eg, pain intensity), we applied a hierarchy of evidence. When authors reported multiple analyses (eg, intention to treat [ITT], available case, or per protocol), we extracted the more conservative with a preference for ITT analyses. We reported AEs according to high-level Medical Dictionary for Regulatory Activities (MedDRA; https://www.meddra.org/) categories and report the 18 most common single AEs.

Data extraction, risk of bias, and GRADE assessments were conducted in duplicate by 2 independent reviewers (any of G.C., E.S., M.W., D.Z., S.N., and R.R.). Interrater disagreement was resolved via consultation with an independent third reviewer (any of L.D., G.C., E.S., M.W., D.Z., and R.R.).

Back to Top | Article Outline

2.9. Data analysis

We extracted data from all reported time points in each trial. Our primary analysis included data from the primary endpoint (or longest follow-up) in each trial. If multiple assessments were made on participants on the same day, we analysed the data taken from the longest follow-up.

Data were analysed separately for RCTs and observational study designs. All analyses were conducted using Review Manager (RevMan) version 5.379 and Stata 15.0.75 Continuous outcomes were pooled using fixed-effect generic inverse variance meta-analysis and expressed as standardised mean differences (SMDs) with 95% confidence intervals (CIs). To aid clinical interpretation of the continuous outcome of change in pain intensity, we additionally reexpressed the SMD for overall change in pain intensity as a mean difference on a 100 mm visual analogue scale (VAS) by multiplying the pooled SMD by a typical baseline among-person SD on a 100 mm VAS, obtained from the included studies.36,38 Dichotomous outcomes were summarised as odds ratios (ORs) using the Mantel–Haenszel fixed-effect model.22 For observational studies, we pooled event rates using the Stata metaprop command.57 Heterogeneity was assessed using the I2 statistic and described as low (≤25%), moderate (>25% and ≤50%), or high (≥75%).35 When data permitted, we assessed publication bias in the pooled estimates using the Stata15.0 metabias command to detect small study effects.33 If the test of small study effects was significant, we used the Stata15.0 metatrim command to conduct Duval and Tweedie's23 nonparametric trim and fill procedure and provide an adjusted treatment effect. We conducted sensitivity analyses using the inverse variance random effects model where I2 values exceeded 50%. For the primary pain intensity outcomes (30% reduction in pain, 50% reduction in pain, and change in pain intensity), we conducted subgroup analyses to assess for differences in RCT-pooled estimates based on overall study risk of bias (low, unclear, or high), study risk of bias due to sample size (low [100+ participants per treatment arm], unclear [30-100 per arm], and high [<30 per arm]), intervention length (1-day studies, very short term [<4 weeks], short term [4-12 weeks], intermediate term [13-26 weeks], or long term [>26 weeks]), and imputation method (none/ITT, completer-only, or last observation carried forward). We followed Cochrane Collaboration methods to overcome unit-of-analysis errors for multiarm studies.35 When raw data were not reported, we used the Generic Inverse Variance fixed-effect model to pool effect estimates and their standard errors.35

For dichotomous outcomes with at least a moderate GRADE rating, we calculated numbers needed to treat to benefit (NNTBs) and numbers needed to treat to harm (NNTHs) and their 95% CIs. We used pooled estimates of relative effect measures (ORs) to take into account the event rate in control groups.11 Number needed to treat to benefit was calculated for the outcomes 30% reduction in pain, 50% reduction in pain, and change in patient global impression of change (PGIC). Number needed to treat to harm was calculated for all-cause AEs and study withdrawals due to AEs. Panel G1 in Appendix G summarises the core statistics and metrics used in this article (available online at http://links.lww.com/PAIN/A592).

Back to Top | Article Outline

3. Results

The combined searches resulted in 2525 results. In total, 91 publications were eligible and included in the review, which reported on 104 distinct studies (Fig. 1, Figure B1 Appendix B, available online at http://links.lww.com/PAIN/A592). Table 1 (RCTs) and Table B1 in Appendix B (available online at http://links.lww.com/PAIN/A592) (observational studies) contain the list of included studies. The search additionally identified 17 ongoing studies for which results are yet to be reported (Appendix Table B2, available online at http://links.lww.com/PAIN/A592). Excluded studies are listed in Appendix Table B3 (appendices available online at http://links.lww.com/PAIN/A592).

Figure 1

Figure 1

Table 1-a

Table 1-a

Table 1-b

Table 1-b

Table 1-c

Table 1-c

Table 1-d

Table 1-d

Table 1-e

Table 1-e

Table 1

Table 1

Table 1

Table 1

Table 1

Table 1

Table 1

Table 1

Back to Top | Article Outline

3.1. Study characteristics

Characteristics of included studies, including sample characteristics, pain classification, cannabinoid classification, treatment length, dose, study outcomes, risk of bias rating, and imputation method are provided in Table 1 (RCTs) and Appendix Table B1 (observational studies, available online at http://links.lww.com/PAIN/A592). The 104 studies comprised 47 RCTs (24 parallel RCTs and 23 cross-over RCTs), and 57 observational studies, comprising a total of 9958 participants (n = 4271 RCTs; 5687 observational studies). We contacted 9 authors for additional information; 6 responded and 2 provided data that were used in analyses. Most studies were conducted in Western Europe (n = 47) or the United States (n = 34, Table 2).

Table 2

Table 2

When possible, we have examined CNCP categories separately. Overall, we found 48 studies of neuropathic pain (of which 16 were multiple sclerosis [MS]-related and 32 were non–MS-related), 7 studies for fibromyalgia, 1 for arthritis (specifically rheumatoid arthritis), and 48 studies for other CNCP (of which 13 were MS-related pain, 6 were visceral pain, and 29 were studies of samples with mixed or undefined non–MS-related CNCP, and Table 3).

Table 3

Table 3

Back to Top | Article Outline

3.2. Characteristics of participants

Detailed characteristics of participants in the studies are provided in Table 1 (RCTs) and Appendix Table B1 (available online at http://links.lww.com/PAIN/A592) (observational studies). Details of ongoing studies with no data available at time of current review are detailed in Appendix Table B2 (available online at http://links.lww.com/PAIN/A592). Details of studies excluded at the full-text review stage are presented in Appendix Table B3 (available online at http://links.lww.com/PAIN/A592). The number of participants ranged from 1 to 649, with a median of 42 (mean 136.8). All studies were conducted in adult samples, except for 2 case series of 2 adolescents (aged 14 and 15 years)67 and an open-label trial in young girls with adverse drug effects after vaccination.59 Where reported, mean age of adult participants ranged from 2843 to 6710 years (median 49.2, mean 50.5), and percentage of males ranged from 0% to 100% (median 46.7%; mean 45.1%). Mean baseline pain intensity scores were 59.6 (SD = 14.6; range: 30.1-87.5) on a 100 mm VAS, suggesting that patients had moderate to severe pain intensity at study intake.34

Pain was the primary indication in 76 studies and a secondary indication in 28 studies. Of the 104 included studies, 412,64,69,77 (n = 47 participants) examined cannabinoids as a first-line therapy, and 87 examined cannabinoids as a second-line therapy in addition to existing medication regimens. In 13 studies, the place of cannabinoids in the therapeutic hierarchy was not reported or unclear. The most common other adjunct medications were opioids, nonsteroidal anti-inflammatory drugs, and antispasticity medications. In nearly all RCT studies, patients were required to be on a stable dose of current medication before commencement of the trial.

The most commonly studied cannabinoid was nabiximols, followed by C. sativa. See Table B4 (available online at http://links.lww.com/PAIN/A592) for more information on the cannabinoids used in the included trials, including route of administration, duration, and dose.

Back to Top | Article Outline

3.3. Risk of bias ratings

Most parallel and cross-over RCTs were rated as unclear risk of bias across all domains because information was not fully reported or could not be obtained from the authors (see Appendix C for ratings of risk of bias, available online at http://links.lww.com/PAIN/A592). Several were rated as at high risk of bias because of selective reporting or other biases, such as omission of data and CIs, changes in selection of the primary endpoint, or a failure to take account of within-subject effects in cross-over studies (Appendix C, Figures C1, C2, available online at http://links.lww.com/PAIN/A592). Observational studies were judged to be at serious or critical risk of bias for key domains because of confounding, intervention measurement, high dropout, and selection of the reported result (Figure C3, available online at http://links.lww.com/PAIN/A592).

Back to Top | Article Outline

3.4. Outcomes

Tables D1 and D2 in Appendix D (available online at http://links.lww.com/PAIN/A592) describe IMMPACT outcomes collected in RCTs and observational studies, respectively. The most commonly studied outcomes were pain intensity (n = 100), AEs (n = 81), and withdrawals (n = 71). Fewer studies reported on physical functioning (n = 52), emotional functioning (n = 43), and patient's global impression of change (n = 24). Only 2 studies in which pain was the primary indication reported on all 6 outcomes.40,80

Back to Top | Article Outline

3.5. Pain

3.5.1. Thirty percent reduction in pain

3.5.1.1. Randomised controlled trial evidence

Of the 47 included RCTs, 13 assessed 30% reduction in pain (Table D1 in Appendix D, available online at http://links.lww.com/PAIN/A592), of which 8 RCTs (based on 9 data points) reported sufficient data and were used in the meta-analysis. Across all cannabinoids and CNCP conditions, cannabinoids were more likely than placebo groups to produce a 30% reduction in pain1,37,39,40,56,71,72,99 (n = 1734, OR 1.46, 95% CI 1.16-1.84, Table 4 and Table E1 and Figure E1 in Appendix E, available online at http://links.lww.com/PAIN/A592). A summary of key outcomes, including NNTB is shown in Table 6. No evidence of small study effects was detected (P = 0.08). We found significant effects for plant-based cannabis, THC:CBD extract, and ajulemic acid, but these were each based on a single study and our GRADE ratings for these estimates was moderate to very low. Among the specific pain conditions, we found effects for neuropathic pain and MS-related CNCP (Table 4 and Figure E1, available online at http://links.lww.com/PAIN/A592). Of the remaining 5 studies that assessed 30% reduction in pain but for which data were not reported or obtained from study authors, 3 reported a significant positive effect and 2 reported no benefit. When examined by overall study risk of bias rating and risk of bias due to sample size, the effect estimate remained significant for studies classified as having low risk and for studies with more than 100 participants per treatment arm, but was not significant for studies at unclear risk of bias, or for studies with less than 100 participants per arm, with notably larger but nonsignificant effects for the smallest studies (<30 participants per arm; Figures E1.1 and E1.1a, available online at http://links.lww.com/PAIN/A592). No significant differences in effect sizes were identified between studies with interventions of very short term (<4 weeks), short term (4-12 weeks), and intermediate term (13-26 weeks, Figure E1.2, available online at http://links.lww.com/PAIN/A592). All studies assessed outcomes using ITT analyses without imputation.

Table 4-a

Table 4-a

Table 4-b

Table 4-b

Back to Top | Article Outline
3.5.1.2. Observational evidence

In observational studies with a comparison group, one small open-label study with a randomised withdrawal phase (n = 2680) found that nabilone was significantly more likely to produce a 30% reduction in pain relative to placebo (Table 4). In observational studies with no comparison group, the pooled prevalence of receiving cannabinoids reported achieving a 30% reduction in pain was 72% (95% CI 66%-78%) (Figure E5 and Appendix F, available online at http://links.lww.com/PAIN/A592).

Back to Top | Article Outline

3.5.2. Fifty percent reduction in pain

3.5.2.1. Randomised controlled trial evidence

Five of the 47 included RCTs assessed 50% reduction in pain, all of which provided sufficient data for meta-analysis. We found no significant evidence that cannabinoids reduced pain by 50% compared with placebo groups (OR 1.43, 95% CI 0.97-2.11, Table 4 and Table E1 and Figure E2 in Appendix E, available online at http://links.lww.com/PAIN/A592). We found no effect for any of the specific cannabinoids; however, among pain conditions, a significant effect was found for non–MS-related neuropathic pain (Table 4). No evidence of small study effects was detected (P = 0.12). No subgroup analysis was able to be conducted for overall study risk of bias, as all studies were classified as low risk. When examined by risk of bias due to sample size, effects were larger and had substantial uncertainty for studies of <100 participants per treatment arm compared with studies with 100+ participants, but all estimates fell within overlapping bounds of uncertainty and were nonsignificant (Figure E2.1.a, available online at http://links.lww.com/PAIN/A592). No differences were detected between studies with interventions of very short term (<4 weeks), short term (4-12 weeks), and intermediate term (13-26 weeks, Figures E2.1 and E2.2, available online at http://links.lww.com/PAIN/A592). All studies assessed outcomes using ITT analyses without imputation.

Back to Top | Article Outline
3.5.2.2. Observational evidence

Two observational studies with a comparison group found evidence of a significant effect for 50% reduction in pain; however, the GRADE rating for this outcome was very low (Table 4 and Table E1 in Appendix E, available online at http://links.lww.com/PAIN/A592). Outcomes for observational studies with no comparison group were equivocal and are summarised narratively in Appendix F (available online at http://links.lww.com/PAIN/A592).

Back to Top | Article Outline

3.5.3. Change in pain intensity

3.5.3.1. Randomised controlled trial evidence

Of the 47 RCTs included in the review, 45 reported data on pain intensity, of which 30 (comprising 34 data points) reported sufficient data and were used in the meta-analysis for change in pain intensity. We found that cannabinoids overall produced a larger reduction in pain intensity than placebo groups (SMD −0.14, 95% CI −0.20 to −0.08, Table 4 and Table E1 and Figure E3 in Appendix E, available online at http://links.lww.com/PAIN/A592). We calculated this to be roughly equivalent to a reduction of 2.9 mm on a 100 mm VAS (95% CI −4.61 to −1.46) greater than placebo groups. Among the cannabinoids, there were significant effects for nabiximols and THC extract, both with a moderate GRADE rating (Table E1, available online at http://links.lww.com/PAIN/A592). We found an effect for neuropathic pain (MS and non–MS-related) and rheumatoid arthritis, but the latter was based on 1 small study and had a very low-grade rating (Table 4). No evidence of small study effects was detected (P = 0.49). Of the remaining 15 studies that assessed pain intensity but for which data were not reported or obtained from study authors, 12 reported a significant positive effect and 3 reported no benefit. When examined by overall risk of bias rating, the effect estimate remained significant for studies classified as low risk but was not significant for studies at unclear or high risk of bias (Figure E3.1, available online at http://links.lww.com/PAIN/A592), and effect sizes were larger for studies with smaller sample sizes (Figure E3.1a, available online at http://links.lww.com/PAIN/A592). When examined by study, intervention length effects seemed to dissipate with increasing study length: 1-day and very short term (<4 weeks) studies remained significant; however, studies conducted in the short (4-12 weeks), intermediate (13-26 weeks), or long term (>26 weeks) did not, with decreasing effect sizes as study length increased (Figure E3.2, available online at http://links.lww.com/PAIN/A592). The effect remained significant for studies using ITT analyses, however, was smaller and not significant for studies using last observation carried forward imputation methods, or where the handling of missing data was not reported (Figure E3.3, available online at http://links.lww.com/PAIN/A592).

Back to Top | Article Outline
3.5.3.2. Observational evidence

In the observational studies with a comparison group, we found no significant evidence of effect for cannabinoids in reducing pain intensity (Table 4). A significant reduction in pain intensity was identified in within-person pre–post assessments of pain in observational studies with no comparison group (Appendix F, available online at http://links.lww.com/PAIN/A592). Five RCTs examined reductions in analgesic use. People taking nabiximols had a greater reduction in the frequency and quantity of use of rescue analgesics compared with placebo groups (SMD −0.13, 95% CI −0.26 to −0.01, I2 = 48%); this had a moderate GRADE rating.

Back to Top | Article Outline

3.6. Physical functioning

No significant effect of cannabinoids on overall physical functioning in 18 RCTs, Table E2 and Figure E6 (available online at http://links.lww.com/PAIN/A592) or quality of life (n = 11 RCTs) compared with placebo groups was found (Table E2 and Figure E8, available online at http://links.lww.com/PAIN/A592). There was a significant effect of cannabinoids in reducing sleep problems when compared with placebo groups (SMD −0.29, 95% CI −0.40 to −0.19), but the GRADE assessment for this was low (Table E2 and Figure E7, available online at http://links.lww.com/PAIN/A592). We found a reduction in sleep problems when compared with placebo groups for nabiximols with a moderate GRADE rating (SMD −0.32, 95% CI −0.44 to −0.20, Table E3 in Appendix E, available online at http://links.lww.com/PAIN/A592). No small study effects were detected for any of these outcomes (P's range from 0.14 to 0.84).

Back to Top | Article Outline

3.7. Emotional functioning

Patients receiving any cannabinoids did not report any difference compared with comparator groups in overall emotional functioning, or in depressive or anxiety symptoms specifically (Table E2 and Figures E9–E11, available online at http://links.lww.com/PAIN/A592). No evidence of small study effects was identified for overall emotional functioning (P = 0.10) or anxiety symptoms (P = 0.06); however, a significant effect was detected for depression (P = 0.01). The trim and fill procedure to account for small study effects revealed that the adjusted estimate did not differ significantly from the original estimate (SMD 0.04, 95% CI −0.14 to 0.22, Table E2, availableonlineathttp://links.lww.com/PAIN/A592). A significant improvement in emotional functioning was identified for dronabinol compared with placebo based on a single study; we had low confidence in this effect (Table E3 in Appendix E, available online at http://links.lww.com/PAIN/A592).

Back to Top | Article Outline

3.8. Patient global impression of change

In the 4 RCTs which reported PGIC as a continuous outcome on the 7-item PGIC scale, there were significant increases among patients receiving any cannabinoid compared with placebo (Table E2 and Figure E12, available online at http://links.lww.com/PAIN/A592), with no evidence of small study effects (P = 0.28). Nine RCTs reported PGIC scores as a dichotomous outcome (much or very much improved vs slightly improved, no change, or worse), with significant improvement among patients receiving any cannabinoid compared with placebo (Table 4 and Figure E13, available online at http://links.lww.com/PAIN/A592), and no evidence of small study effects (P = 0.3). Confidence in these outcomes was low to very low. Most of the evidence was for nabiximols, with some evidence for nabilone, C. sativa, and THC extract.

Back to Top | Article Outline

3.9. Study withdrawals

Patients with CNCP who received a cannabinoid had 2 times the odds of withdrawing from a trial for any reason than patients who received placebo (Table E4 in Appendix E, available online at http://links.lww.com/PAIN/A592). They had 3.47 times the odds of withdrawing because of AEs (Table 5); no evidence of small study effects was found (P = 0.44). Patients with CNCP who received placebo were slightly more likely to withdraw from trials because of a lack of efficacy than those receiving cannabinoids. There was some variation between cannabinoids in reasons for withdrawal (Table E4 in Appendix E, available online at http://links.lww.com/PAIN/A592).

Table 5

Table 5

Back to Top | Article Outline

3.10. Adverse events

Patients with CNCP receiving cannabinoids had 2.33 times the odds of experiencing an AE compared with placebo groups (Table 5 and Table E4 in Appendix E, available online at http://links.lww.com/PAIN/A592). Significant evidence of small study effects was detected (P = 0.01); however, the adjusted estimate did not differ significantly from the original (OR = 2.22, 95% CI 1.60-3.01). Serious AEs were reported in a smaller number of studies (Table 5), and patients receiving cannabinoids had higher rates of serious AEs, but this did not reach statistical significance. No small study effects were detected (P = 0.52). Compared with placebo groups, patients receiving cannabinoids were more likely to report individual AEs such as dizziness (OR 5.52, 95% CI 4.47-6.83), cognitive attention or disturbance (OR 5.67, 95% CI 2.72-11.79), and confusion and disorientation (OR 5.35, 95% CI 2.31-12.39, Table 5).

Back to Top | Article Outline

3.11. Summary statistics

Table 6 summarises the pooled ORs, pooled event rates for cannabinoids vs placebo groups, and NNTB or NNTH for dichotomous outcomes with a moderate or higher GRADE rating in RCTs. Note: because we only had continuous measures of sleep outcomes, cannabinoids' impacts on improving sleep cannot be included in these summary statistics.

Table 6

Table 6

For cannabinoids' impact on pain outcomes, pooled event rates for 30% reduction in pain intensity were 29.0% vs 25.9%, respectively. The NNTB was 24 (95% CI 15-61, Table 6). For a 50% reduction in pain, the pooled event rate for cannabinoids was 18.2%, compared with 14.4% for placebo groups (Table 6). The NNTB for 50% reduction in pain was unable to be calculated, as the estimate crossed the line of no effect.

For studies where outcomes were presented dichotomously, participants receiving cannabinoids had slightly increased odds of reporting global improvements (PGIC) than patients who received placebo (Table 6). In participants receiving cannabinoids, the pooled percentage reporting “much” or “very much” global improvement was 18.9% compared with 11.8%; the NNT was 38 (95% CI 27-62).

Pooled statistics for AEs and study withdrawals are also presented in Table 6. The estimated pooled rate of all-cause AEs was 81.2% among people receiving cannabinoids, compared with 66.2% of those receiving placebo; the NNTH was 6 (95% CI 5-8). The pooled event rate for study withdrawals due to AEs was 15.8% in those receiving cannabinoids compared with 4.6% of those receiving placebo, and the NNTH was 40 (35-49).

Back to Top | Article Outline

4. Discussion

To the best of our knowledge, this is the first systematic review of the evidence for the effectiveness and safety of cannabinoids for CNCP that included all cannabinoids, all study designs, and considered all outcomes recommended by the IMMPACT group. We also assessed the clinical relevance of these findings using event rates, NNTB, and NNTH.

We found moderate evidence for a reduction in pain for cannabinoids when compared with placebo groups. Pooled analyses suggested that 30% reduction in pain was reported by 29.0% in cannabinoids, compared with 25.9% in placebo groups. A 50% reduction in pain was reported by 18.2% in cannabinoid groups and 14.4% in placebo groups; however, this did not reach statistical significance. The NNTB to achieve a 30% reduction in pain for 1 person using cannabis or cannabinoids (compared with placebo groups) was estimated at 24 (95% CI 15-61), and the NNTH for 1 person to experience any AE was 6 (95% CI 5-8). Although caution needs to be used in comparing NNTs across studies involving different groups and timeframes,44 these NNTBs are much higher than those for other analgesics: previous studies in neuropathic pain suggested NNTs for strong opioids of 4.3 (95% CI 3.4-5.8), pregabalin (7.7, 95% CI 6.5-9.4), and tricyclic antidepressants (3.6, 95% CI 3.0-4.4).28 The NNTH in our review was similar to that for opioids for CNCP, with a recent Cochrane review indicating that the NNTH for 1 person using opioids to experience any AE (compared with placebo) was 5 (95% CI 4-9).27 When reexpressed as a mean change on the commonly used 100 mm VAS, the pooled SMD for the continuous outcome of change in pain intensity was equivalent to a 3 mm greater reduction on this scale compared with placebo, which is well below the 30 mm reduction regarded to represent a clinically important difference in pain intensity.41,58 In contrast to more optimistic conclusions from earlier reviews,2,48 our findings are largely consistent with a recent Cochrane review examining cannabinoids for neuropathic pain, indicating that these medicines are unlikely to be effective in the treatment of pain.46 In their review, Mücke et al.46 report an NNTB of 20 for 50% or greater reduction in pain, and NNTHs of 3 and 6 for AEs relating to nervous system and psychiatric disorders, respectively, suggesting a similar efficacy and safety profile of cannabinoids for pain as reported in our review.

The evidence on the effectiveness of cannabinoids for CNCP is limited for several reasons. First, sample size is an issue, with only 21 of the 104 included studies having at least 100 participants per treatment arm. Although we made multiple attempts to minimise risk of bias in the effect estimates due to small sample sizes, this risk cannot be fully mitigated. For some estimates, effect sizes were notably larger in studies with <30 participants per treatment arm compared with studies of 100+ per arm; however, these estimates fell within overlapping bounds of uncertainty. There is a growing body of evidence indicating that effect estimates tend to be larger in studies with small sample sizes,20 and as such, caution should be taken when interpreting outcomes based on studies with small sample sizes in this review. Well conducted, large RCTs comprising at least 100 participants per treatment arm should be considered a priority in this space. Second, most studies were of limited duration (median of 8 weeks): given that CNCP is a chronic condition, this sheds little light on the appropriateness of long-term use of cannabinoids in CNCP, in terms of both treatment efficacy and safety. Of the little evidence available, we found that reductions in pain intensity were largest for 1-day studies, and smaller and nonsignificant in studies of 13-week duration or longer, providing some initial suggestion that the effectiveness of cannabinoids for CNCP may diminish over time. Third, the issues of cannabinoid tolerance, risks of iatrogenic dependence, and of withdrawal symptoms if long-term cannabinoids are ceased, remain poorly understood. Short-term clinical trials such as those included in this review are often of insufficient power and duration to detect potential harms and AEs associated with long-term cannabis use, such as elevated risk of psychosis and substance dependence.32,45 It is crucial that these long-term outcomes identified in the epidemiological literature are considered alongside evidence of efficacy from clinical trials when determining overall suitability of cannabinoids as medicines for CNCP. Fourth, cannabinoid dose was often poorly recorded. Often, only a maximum recommended dose was reported and data on participants' actual cannabinoid consumption were seldom recorded, so it is difficult to make strong recommendations on doses that are maximally effective and safe. Fifth, by far, the greatest amount of high-quality evidence was for nabiximols, resulting in small numbers of studies (and in some cases, single study) in some analyses for other types and formulations of cannabinoids (eg, ajulemic acid), meaning that we are be less confident about their efficacy. Sixth, although almost all studies reported data on change in pain intensity, very few reported outcomes for 30% and 50% reduction in pain. Given that pain was a secondary outcome in many studies, it is possible that authors did not report these outcomes because they are drawn from the pain-specific IMMPACT guidelines; however, there is also the possibility that study authors chose not to report outcomes for 30% and 50% reduction in pain when the continuous pain intensity outcome indicated no benefit. Although we have made multiple attempts to account for publication bias throughout this review, there remains the possibility that the studies for which 30% and 50% reduction in pain were not reported did not find evidence of effect. If this is the case, NNTBs for these outcomes may be higher than those reported here; however, our overall conclusion that cannabinoids are unlikely to be effective medicines for CNCP will remain unchanged. Finally, to ensure that all the available evidence of cannabinoids as a treatment for CNCP was considered in this review, we included evidence from RCTs and less rigorous observational study designs. This approach allows researchers, clinicians, and policymakers to map current research activity and to identify knowledge gaps. Although observational studies provide some insight into the efficacy of cannabinoids for CNCP, ultimately only data from high-quality RCTs will be used to inform national treatment guidelines. We noted that most of the higher-quality RCT evidence was for neuropathic pain and MS-related pain. There is scant, low-quality evidence on cannabinoids used for fibromyalgia or visceral pain, and very few studies of cannabinoids' use in the most common and burdensome CNCP conditions, namely back/neck problems, migraines, and arthritides. Thus, the conclusions of this review primarily relate to neuropathic or MS-related pain. Several ongoing studies targeting these more common CNCP conditions were identified and will be analysed when results become available.

Most studies used a placebo comparator and added cannabinoids to stable doses of analgesics, nonsteroidal anti-inflammatory drugs, and antispasticity drugs, so the evidence for cannabinoid use in CNCP is largely around cannabinoids as adjuvant medicines. Often, multiple analgesics were used, which varied between groups, and the ways they were used were not consistently reported. Most studies held doses of other analgesic medications constant, although some studies documented changes in breakthrough medication or adjunctive analgesia.

Back to Top | Article Outline

4.1. Limitations of this review

The findings of this review need to be considered in light of several potential limitations. Some of these limitations have already been noted and include the high risk of bias in many studies because of small N and missing information on study design and rigour of controls; most studies also evaluated cannabinoids as adjunct to other analgesic medications. We attempted to assertively minimise these limitations. Many documents were reviewed by a small research team, which might have led to errors in assessing eligible studies. However, internal checks were conducted by members within this team and a process of double and triple checking existed; we also checked all identified reviews to ensure that no studies had been missed that had been reported in any other reviews of evidence. Third, errors may have been made in data interpretation. To reduce such errors, all sources and data extracted were double checked by at least 2 reviewers and conflicts were resolved by third reviewer when necessary.

Back to Top | Article Outline

5. Conclusions

It seems unlikely that cannabinoids are highly effective medicines for CNCP. There is moderate- to high-grade evidence supporting use of nabiximols to achieve modest reductions in pain as adjunctive therapy in MS-related pain. However, NNTBs were high and NNTHs low, with high rates of dropout for AEs, and long-term efficacy and safety is unknown. We also found minimal evidence that cannabinoids are effective in improving other important domains in people with CNCP such as emotional and physical functioning. Cannabinoids are unlikely to be a monotherapy for CNCP. People living with CNCP often have complex comorbidities,9,70 and multidisciplinary treatment that includes physical and psychological therapy rather than reliance on medicines alone is likely to be most effective.

Back to Top | Article Outline

Conflict of interest statement

G. Campbell, S. Nielsen, M. Farrell, and L. Degenhardt have all been investigators on untied investigator-driven educational grants funded by Reckitt Benckiser. M. Farrell and L. Degenhardt have received an untied educational grant from Mundipharma for post-marketing surveillance studies of a potentially tamper-resistant formulation of controlled-released oxycodone. S. Nielsen, M. Farrell, and L. Degenhardt have been investigators on untied investigator-driven educational grants funded by Indivior. M. Farrell and L. Degenhardt have been investigators on an untied investigator-driven educational grant funded by Seqirus. The remaining authors have no conflict of interest to declare.

Funding was received from the Commonwealth Department of Health, the NSW Government Centre for Medicinal Cannabis Research and Innovation, the Victorian Department of Health and Human Services, and the Queensland Department of Health. E. Stockings, G. Campbell, S. Nielsen, and L. Degenhardt are supported by NHMRC research fellowships (#1104600; #1119992; #1132433; and #1041472). The National Drug and Alcohol Research Centre at the University of NSW is supported by funding from the Australian Government under the Substance Misuse Prevention and Service Improvements Grant Fund.

Back to Top | Article Outline

Acknowledgements

The authors acknowledge Mary Kumvaj who assisted in the development of the search strategy.

Author contributions: L. Degenhardt and M. Farrell conceived the Review. E. Stockings, G. Campbell, S. Nielsen, D. Zagic, R. Rahman, and M. Weier did the systematic search, selected papers, and extracted data. E. Stockings conducted statistical analyses. G. Campbell, L. Degenhardt, and W.D. Hall drafted the manuscript with critical revisions from all authors. B. Murnion provided clinical important intellectual content. All authors reviewed the paper before submission.

Back to Top | Article Outline

Supplemental digital content

Supplemental digital content associated with this article can be found online at http://links.lww.com/PAIN/A592.

Back to Top | Article Outline

Supplemental video content

Video content associated with this article can be found online at http://links.lww.com/PAIN/A593.

Back to Top | Article Outline

References

[1]. Abrams DI, Jay C, Shade S, Vizoso H, Reda H, Press S, Kelly M, Rowbotham M, Petersen K. Cannabis in painful HIV-associated sensory neuropathy: a randomized placebo-controlled trial. Neurology 2007;68:515–21.
[2]. Andreae MH, Carter GM, Shaparin N, Suslov K, Ellis RJ, Ware MA, Abrams DI, Prasad H, Wilsey B, Indyk D, Johnson M, Sacks HS. Inhaled cannabis for chronic neuropathic pain: a meta-analysis of individual patient data. J Pain 2015;16:1221–32.
[3]. Aviram J, Samuelly-Leichtag G. Efficacy of cannabis-based medicines for pain management: a systematic review and meta-analysis of randomized controlled trials. Pain Physician 2017;20:E755–96.
[4]. Ball S, Vickery J, Hobart J, Wright D, Green C, Shearer J, Nunn A, Cano MG, MacManus D, Miller D, Mallik S, Zajicek J. The Cannabinoid Use in Progressive Inflammatory brain Disease (CUPID) trial: a randomised double-blind placebo-controlled parallel-group multicentre trial and economic evaluation of cannabinoids to slow progression in multiple sclerosis. Health Technol Assess 2015;19:vii–viii, xxv–xxxi, 1–187.
    [5]. Berman JS. A study to evaluate the effects of cannabis based medicine in patients with pain of neurological origin. 2012. NCT01606176, Vol. 2017, Clinicaltrials.gov.
      [6]. Berman JS, Symonds C, Birch R. Efficacy of two cannabis based medicinal extracts for relief of central neuropathic pain from brachial plexus avulsion: results of a randomised controlled trial. PAIN 2004;112:299–306.
        [7]. Bestard JA, Toth CC. An open-label comparison of nabilone and gabapentin as adjuvant therapy or monotherapy in the management of neuropathic pain in patients with peripheral neuropathy. Pain Pract 2011;11:353–68.
          [8]. Blake DR, Robson P, Ho M, Jubb RW, McCabe CS. Preliminary assessment of the efficacy, tolerability and safety of a cannabis-based medicine (Sativex) in the treatment of pain caused by rheumatoid arthritis. Rheumatology (Oxford) 2006;45:50–2.
            [9]. Campbell G, Nielsen S, Bruno R, Lintzeris N, Cohen M, Hall W, Larance B, Mattick RP, Degenhardt L. The pain and opioids IN Treatment study: characteristics of a cohort using opioids to manage chronic non-cancer pain. PAIN 2015;156:231–42.
            [10]. Carroll CB, Bain PG, Teare L, Liu X, Joint C, Wroath C, Parkin SG, Fox P, Wright D, Hobart J, Zajicek JP. Cannabis for dyskinesia in Parkinson disease: a randomized double-blind crossover study. Neurology 2004;63:1245–50.
            [11]. Cates CJ. Simpson's paradox and calculation of number needed to treat from meta-analysis. BMC Med Res Methodol 2002;2:1.
            [12]. Chung SA, Hossain NK, Blackman AS, Shapiro CM. Can the cannabinoid nabilone help with pain and sleep in fibromyalgia patients? Sleep 2009;32:A325–6.
            [13]. Deleted in proof.
              [14]. Collin C, Ehler E, Waberzinek G, Alsindi Z, Davies P, Powell K, Notcutt W, O'leary C, Ratcliffe S, Nováková I. A double-blind, randomized, placebo-controlled, parallel-group study of Sativex, in subjects with symptoms of spasticity due to multiple sclerosis. Neurol Res 2010;32:451–9.
                [15]. Corey-Bloom J, Wolfson T, Gamst A, Jin S, Marcotte TD, Bentley H, Gouaux B. Smoked cannabis for spasticity in multiple sclerosis: a randomized, placebo-controlled trial. Can Med Assoc J 2012;184:1143–50.
                  [16]. Curtin F, Elbourne D, Altman DG. Meta-analysis combining parallel and cross-over clinical trials. II: binary outcomes. Stat Med 2002;21:2145–59.
                    [17]. Curtin F, Elbourne D, Altman DG. Meta-analysis combining parallel and cross-over clinical trials. III: the issue of carry-over. Stat Med 2002;21:2161–73.
                      [18]. de Vries M, Van Rijckevorsel DC, Vissers KC, Wilder-Smith OH, Van Goor H. Single dose delta-9-tetrahydrocannabinol in chronic pancreatitis patients: analgesic efficacy, pharmacokinetics and tolerability. Br J Clin Pharmacol 2016;81:525–37.
                        [19]. de Vries M, van Rijckevorsel DCM, Vissers KCP, Wilder-Smith OHG, van Goor H. Tetrahydrocannabinol Does not reduce pain in patients with chronic abdominal pain in a phase 2 placebo-controlled study. Clin Gastroenterol Hepatol 2017;15:1079–86.e4.
                          [20]. Dechartres A, Trinquart L, Boutron I, Ravaud P. Influence of trial sample size on treatment effect estimates: meta-epidemiological study. BMJ 2013;346:f2304.
                          [21]. Degenhardt L, Lintzeris N, Campbell G, Bruno R, Cohen M, Farrell M, Hall WD. Experience of adjunctive cannabis use for chronic non-cancer pain: findings from the Pain and Opioids IN Treatment (POINT) study. Drug Alcohol Depend 2015;147:144–50.
                            [22]. Deeks J, Higgins J, Altman D, Cochrane Statistical Methods Group. Chapter 9: analysing data and undertaking meta-analysis. In: Higgins J, Green S, editors. Cochrane handbook for systematic reviews of interventions 510 (updated March 2011). London, United Kingdom: The Cochrane Collaboration, 2011.
                            [23]. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis. Biometrics 2000;56:455–63.
                            [24]. Dworkin RH, Turk DC, Wyrwich KW, Beaton D, Cleeland CS, Farrar JT, Haythornthwaite JA, Jensen MP, Kerns RD, Ader DN, Brandenburg N, Burke LB, Cella D, Chandler J, Cowan P, Dimitrova R, Dionne R, Hertz S, Jadad AR, Katz NP, Kehlet H, Kramer LD, Manning DC, McCormick C, McDermott MP, McQuay HJ, Patel S, Porter L, Quessy S, Rappaport BA, Rauschkolb C, Revicki DA, Rothman M, Schmader KE, Stacey BR, Stauffer JW, von Stein T, White RE, Witter J, Zavisic S. Interpreting the clinical importance of treatment outcomes in chronic pain clinical trials: IMMPACT recommendations. J Pain 2008;9:105–21.
                            [25]. Elbourne DR, Altman DG, Higgins JP, Curtin F, Worthington HV, Vail A. Meta-analyses involving cross-over trials: methodological issues. Int J Epidemiol 2002;31:140–9.
                              [26]. Ellis RJ, Toperoff W, Vaida F, Van Den Brande G, Gonzales J, Gouaux B, Bentley H, Atkinson JH. Smoked medicinal cannabis for neuropathic pain in HIV: a randomized, crossover clinical trial. Neuropsychopharmacology 2009;34:672–80.
                                [27]. Els C, Jackson TD, Kunyk D, Lappi VG, Sonnenberg B, Hagtvedt R, Sharma S, Kolahdooz F, Straube S. Adverse events associated with medium- and long-term use of opioids for chronic non-cancer pain: an overview of Cochrane Reviews. Cochrane Database Syst Rev 2017;10:CD012509.
                                [28]. Finnerup NB, Attal N, Haroutounian S, McNicol E, Baron R, Dworkin RH, Gilron I, Haanpaa M, Hansson P, Jensen TS, Kamerman PR, Lund K, Moore A, Raja SN, Rice ASC, Rowbotham M, Sena E, Siddall P, Smith BH, Wallace M. Pharmacotherapy for neuropathic pain in adults: systematic review, meta-analysis and updated NeuPSIG recommendations. Lancet Neurol 2015;14:162–73.
                                [29]. Frank B, Serpell MG, Hughes J, Matthews JNS, Kapur D. Comparison of analgesic effects and patient tolerability of nabilone and dihydrocodeine for chronic neuropathic pain: randomised, crossover, double blind study. BMJ 2008;336:199–201.
                                  [30]. GBD 2016 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 2017;390:1211–59.
                                  [31]. Hagenbach U, Luz S, Ghafoor N, Berger J, Grotenhermen F, Brenneisen R, Mäder M. The treatment of spasticity with [Delta] 9-tetrahydrocannabinol in persons with spinal cord injury. Spinal Cord 2007;45:551.
                                    [32]. Hall W, Degenhardt L. Adverse health effects of non-medical cannabis use. Lancet 2009;374:1383–91.
                                    [33]. Harbord R, Harris R, Sterne J. Updated tests for small-study effects in meta-analyses. Stata J 2009;9:197–210.
                                    [34]. Hawker GA, Mian S, Kendzerska T, French M. Measures of adult pain: Visual Analog Scale for Pain (VAS pain), Numeric Rating Scale for Pain (NRS pain), McGill Pain Questionnaire (MPQ), Short-Form McGill Pain Questionnaire (SF-MPQ), Chronic Pain Grade Scale (CPGS), Short Form-36 Bodily Pain Scale (SF-36 BPS), and measure of intermittent and constant osteoarthritis pain (ICOAP). Arthritis Care Res (Hoboken) 2011;63(suppl 11):S240–52.
                                    [35]. Higgins J, Green S. Cochrane handbook for systematic reviews of interventions version 5.1.0. London, United Kingdom: 2011.
                                    [36]. Higgins JP, Green S. Cochrane handbook for systematic reviews of interventions. Vol. 4. London, United Kingdom: John Wiley & Sons, 2011.
                                    [37]. Hoggart B, Ratcliffe S, Ehler E, Simpson KH, Hovorka J, Lejcko J, Taylor L, Lauder H, Serpell M. A multicentre, open-label, follow-on study to assess the long-term maintenance of effect, tolerance and safety of THC/CBD oromucosal spray in the management of neuropathic pain. J Neurol 2015;262:27–40.
                                    [38]. Johnston BC, Patrick DL, Thorlund K, Busse JW, da Costa BR, Schünemann HJ, Guyatt GH. Patient-reported outcomes in meta-analyses—Part 2: methods for improving interpretability for decision-makers. Health Qual Life Outcomes 2013;11:211.
                                    [39]. Karst M, Salim K, Burstein S, Conrad I, Hoy L, Schneider U. Analgesic effect of the synthetic cannabinoid CT-3 on chronic neuropathic pain: a randomized controlled trial. J Am Med Assoc 2003;290:1757–62.
                                    [40]. Langford R, Mares J, Novotna A, Vachova M, Novakova I, Notcutt W, Ratcliffe S. A double-blind, randomized, placebo-controlled, parallel-group study of THC/CBD oromucosal spray in combination with the existing treatment regimen, in the relief of central neuropathic pain in patients with multiple sclerosis. J Neurol 2013;260:984–97.
                                    [41]. Lee JS, Hobden E, Stiell IG, Wells GA. Clinically important change in the visual analog scale after adequate pain control. Acad Emerg Med 2003;10:1128–30.
                                    [42]. Lynch ME, Cesar-Rittenberg P, Hohmann AG. A double-blind, placebo-controlled, crossover pilot trial with extension using an oral mucosal cannabinoid extract for treatment of chemotherapy-induced neuropathic pain. J Pain Symptom Manage 2014;47:166–73.
                                      [43]. Maurer M, Henn V, Dittrich A, Hofmann A. Delta-9-tetrahydrocannabinol shows antispastic and analgesic effects in a single case double-blind trial. Eur Arch Psychiatry Clin Neurosci 1990;240:1–4.
                                      [44]. McAlister FA. The “number needed to treat” turns 20—and continues to be used and misused. CMAJ 2008;179:549–53.
                                      [45]. Moore THM, Zammit S, Lingford-Hughes A, Barnes TRE, Jones PB, Burke M, Lewis G. Cannabis use and risk of psychotic or affective mental health outcomes: a systematic review. Lancet 2007;370:319–28.
                                      [46]. Mücke M, Phillips T, Radbruch L, Petzke F, Hauser W. Cannabis-based medicines for chronic neuropathic pain in adults. Cochrane Database Syst Rev 2018;3:Cd012182.
                                      [47]. Narang S, Gibson D, Wasan AD, Ross EL, Michna E, Nedeljkovic SS, Jamison RN. Efficacy of dronabinol as an adjuvant treatment for chronic pain patients on opioid therapy. J Pain 2008;9:254–64.
                                        [48]. National Academies of Sciences Engineering and Medicine. The health effects of cannabis and cannabinoids: the current state of evidence and recommendations for research. Washington, DC: The National Academies Press, 2017.
                                        [49]. NCT00710424 (GW Pharmaceuticals). A study of Sativex® for pain relief due to diabetic neuropathy. 2008. NCT00710424, Vol. 2017, Clinicaltrials.gov.
                                          [50]. NCT01606176 (GW Pharmaceuticals). A study to evaluate the effects of cannabis based medicine in patients with pain of neurological origin. 2012. NCT01606176, Vol. 2017, Clinicaltrials.gov.
                                            [51]. NCT01606202 (GW Pharmaceuticals). A study of cannabis based medicine extracts and placebo in patients with pain due to spinal cord injury. 2012. NCT01606202, Vol. 2017, Clinicaltrials.gov.
                                              [52]. Notcutt W. A study of cannabis based medicine extracts and placebo in patients with pain due to spinal cord injury. 2012. NCT01606202, Vol. 2017, Clinicaltrials.gov.
                                                [53]. Notcutt W, Phillips C, Hughes J, Lacoux P, Vijayakulasingam V, Baldock L. A retrospective description of the use of nabilone in UK clinical practice—extension study (conference poster). Mult Scler 2014;468.
                                                  [54]. Novotna A, Mares J, Ratcliffe S, Novakova I, Vachova M, Zapletalova O, Gasperini C, Pozzilli C, Cefaro L, Comi G, Rossi P, Ambler Z, Stelmasiak Z, Erdmann A, Montalban X, Klimek A, Davies P. A randomized, double-blind, placebo-controlled, parallel-group, enriched-design study of nabiximols* (Sativex), as add-on therapy, in subjects with refractory spasticity caused by multiple sclerosis. Eur J Neurol 2011;18:1122–31.
                                                    [55]. Nugent SM, Morasco BJ, O'Neil ME, Freeman M, Low A, Kondo K, Elven C, Zakher B, Motu'apuaka M, Paynter R, Kanasagara D. The effects of cannabis among adults with chronic pain and an overview of general harms: a systematic review. Ann Intern Med 2017;167:319–31.
                                                    [56]. Nurmikko TJ, Serpell MG, Hoggart B, Toomey PJ, Morlion BJ, Haines D. Sativex successfully treats neuropathic pain characterised by allodynia: a randomised, double-blind, placebo-controlled clinical trial. PAIN 2007;133:210–20.
                                                    [57]. Nyaga VN, Arbyn M, Aerts M. Metaprop: a stata command to perform meta-analysis of binomial data. Arch Public Health 2014;72:39.
                                                    [58]. Ostelo RWJG, de Vet HCW. Clinically important outcomes in low back pain. Best Pract Res Clin Rheumatol 2005;19:593–607.
                                                    [59]. Palmieri B, Laurino C, Vadala M. Short-term efficacy of CBD-enriched hemp oil in girls with dysautonomic syndrome after human papillomavirus vaccination. Isr Med Assoc J 2017;19:79–84.
                                                    [60]. Petzke F, Enax-Krumova EK, Hauser W. Efficacy, tolerability and safety of cannabinoids for chronic neuropathic pain: a systematic review of randomized controlled studies (article in German). Schmerz 2016;30:62–88.
                                                    [61]. Pini LA, Guerzoni S, Cainazzo MM, Ferrari A, Sarchielli P, Tiraferri I, Ciccarese M, Zappaterra M. Nabilone for the treatment of medication overuse headache: results of a preliminary double-blind, active-controlled, randomized trial. J Headache Pain 2012;13:677–84.
                                                      [62]. Pinsger M, Schimetta W, Volc D, Hiermann E, Riederer F, Pölz W. Benefits of an add-on treatment with the synthetic cannabinomimetic nabilone on patients with chronic pain—a randomized controlled trial. Wiener Klinische Wochenschrift 2006;118:327–35.
                                                        [63]. Platt L, Reed J, Minozzi S, Vickerman P, Hagan H, French C, Jordan A, Degenhardt L, Hope V, Hutchinson S. Effectiveness of needle/syringe programmes and opiate substitution therapy in preventing HCV transmission among people who inject drugs. Cochrane Database of Systematic Reviews 2016;1:1–13. CD012021.
                                                        [64]. Rintala DH, Fiess RN, Tan G, Holmes SA, Bruel BM. Effect of dronabinol on central neuropathic pain after spinal cord injury: a pilot study. Am J Phys Med Rehabil 2010;89:840–8.
                                                        [65]. Riva N, Mora G, Soraru G, Lunetta C, Falzone Y, Marinou K, Maestri E, Fazio R, Comola M, Comi G. The canals study: a randomized, double-blind, placebo-controlled, multicentre study to assess the safety and efficacyon spasticity symptoms of a cannabis sativa extract in motor neuron disease patients. Amyotroph Lateral Scler Frontotemporal Degener 2016;17:44.
                                                          [66]. Rog DJ, Nurmikko TJ, Friede T, Young CA. Randomized, controlled trial of cannabis-based medicine in central pain in multiple sclerosis. Neurology 2005;65:812–19.
                                                            [67]. Rudich Z, Stinson J, Jeavons M, Brown SC. Treatment of chronic intractable neuropathic pain with dronabinol: case report of two adolescents. Pain Res Manag 2003;8:221–4.
                                                            [68]. Schimrigk S, Marziniak M, Neubauer C, Kugler EM, Werner G, Abramov-Sommariva D. Dronabinol is a safe long-term treatment option for neuropathic pain patients. Eur Neurol 2017;78:320–9.
                                                              [69]. Schley M, Legler A, Skopp G, Schmelz M, Konrad C, Rukwied R. Delta-9-THC based monotherapy in fibromyalgia patients on experimentally induced pain, axon reflex flare, and pain relief. Curr Med Res Opin 2006;22:1269–76.
                                                              [70]. Scott KM, Lim C, Al-Hamzawi A, Alonso J, Bruffaerts R, Caldas-de-Almeida JM, Florescu S, de Girolamo G, Hu C, de Jonge P, Kawakami N, Medina-Mora ME, Moskalewicz J, Navarro-Mateu F, O'Neill S, Piazza M, Posada-Villa J, Torres Y, Kessler RC. Association of mental disorders with subsequent chronic physical conditions: world mental health surveys from 17 countries. JAMA Psychiatry 2016;73:150–8.
                                                              [71]. Selvarajah D, Gandhi R, Emery CJ, Tesfaye S. Randomized placebo-controlled double-blind clinical trial of cannabis-based medicinal product (Sativex) in painful diabetic neuropathy: depression is a major confounding factor. Diabetes Care 2010;33:128–30.
                                                              [72]. Serpell M, Ratcliffe S, Hovorka J, Schofield M, Taylor L, Lauder H, Ehler E. A double-blind, randomized, placebo-controlled, parallel group study of THC/CBD spray in peripheral neuropathic pain treatment. Eur J Pain 2014;18:999–1012.
                                                              [73]. Shah A, Craner J, Cunningham JL. Medical cannabis use among patients with chronic pain in an interdisciplinary pain rehabilitation program: characterization and treatment outcomes. J Subst Abuse Treat 2017;77:95–100.
                                                                [74]. Skrabek RQ, Galimova L, Ethans K, Perry D. Nabilone for the treatment of pain in fibromyalgia. J Pain 2008;9:164–73.
                                                                  [75]. StataCorp. Stata Statistical Software: Release 15, Vol. 15.0. College Station: StataCorp LLC, 2017.
                                                                  [76]. Sterne J, Higgins J, Reeves B. Extending the risk of bias tool to allow for assessment of non-randomised studies, cluster-randomised trials and cross-over trials: a Cochrane methods innovation fund project (Workshop). Proceedings of the Book Extending the Risk of Bias Tool to Allow for Assessment of Non-Randomised Studies, Cluster-Randomised Trials and Cross-Over Trials: A Cochrane Methods Innovation Fund Project (Workshop). In: Better Knowledge for Better Health | Un meilleur savoir pour une meilleure santé. Abstracts of the 21st Cochrane Colloquium; 2013 19-23 Sep; Québec City, Canada. John Wiley & Sons, 2013. p. 203–4.
                                                                  [77]. Svendsen KB, Jensen TS, Bach FW. Does the cannabinoid dronabinol reduce central pain in multiple sclerosis? Randomised double blind placebo controlled crossover trial. BMJ 2004;329:253.
                                                                  [78]. Tesfaye SA. Study of Sativex® for pain relief due to diabetic neuropathy. 2008. NCT00710424, Vol. 2017, Clinicaltrials.gov.
                                                                    [79]. The Nordic Cochrane Centre. Review Manager (RevMan). Copenhagen, Denmark: The Cochrane Collabration, 2014.
                                                                    [80]. Toth C, Mawani S, Brady S, Chan C, Liu C, Mehina E, Garven A, Bestard J, Korngut L. An enriched-enrolment, randomized withdrawal, flexible-dose, double-blind, placebo-controlled, parallel assignment efficacy study of nabilone as adjuvant in the treatment of diabetic peripheral neuropathic pain. PAIN 2012;153:2073–82.
                                                                    [81]. Turcotte D, Doupe M, Torabi M, Gomori A, Ethans K, Esfahani F, Galloway K, Namaka M. Nabilone as an adjunctive to gabapentin for multiple sclerosis-induced neuropathic pain: a randomized controlled trial. Pain Med 2015;16:149–59.
                                                                      [82]. Turk DC, Dworkin RH, Allen RR, Bellamy N, Brandenburg N, Carr DB, Cleeland C, Dionne R, Farrar JT, Galer BS. Core outcome domains for chronic pain clinical trials: IMMPACT recommendations. PAIN 2003;106:337–45.
                                                                      [83]. van Amerongen G, Kanhai K, Baakman AC, Heuberger J, Klaassen E, Beumer TL, Strijers RLM, Killestein J, van Gerven J, Cohen A, Groeneveld GJ. Effects on spasticity and neuropathic pain of an oral formulation of delta9-tetrahydrocannabinol in patients with progressive multiple sclerosis. Clin Ther 2017 Feb 9. pii: S0149-2918(17)30054-1. doi: [epub ahead of print].
                                                                        [84]. Veritas Health Innovation. Covidence systematic review software. Melbourne. Available at: http://www.covidence.org. Accessed 2 July 2018.
                                                                        [85]. Vermersch P, Trojano M. Tetrahydrocannabinol: cannabidiol oromucosal spray for multiple sclerosis-related resistant spasticity in daily practice. Eur Neurol 2016;76:216–26.
                                                                          [86]. Wade DT, Makela P, Robson P, House H, Bateman C. Do cannabis-based medicinal extracts have general or specific effects on symptoms in multiple sclerosis? A double-blind, randomized, placebo-controlled study on 160 patients. Mult Scler J 2004;10:434–41.
                                                                            [87]. Wade DT, Robson P, House H, Makela P, Aram J. A preliminary controlled study to determine whether whole-plant cannabis extracts can improve intractable neurogenic symptoms—cross over study. Clin Rehabil 2003;17:21–9.
                                                                              [88]. Walitt B, Klose P, Fitzcharles MA, Phillips T, Hauser W. Cannabinoids for fibromyalgia. Cochrane Database Syst Rev 2016;7:Cd011694.
                                                                              [89]. Wallace MS, Marcotte TD, Umlauf A, Gouaux B, Atkinson JH. Efficacy of inhaled cannabis on painful diabetic neuropathy. J Pain 2015;16:616–27.
                                                                                [90]. Ware MA, Fitzcharles M-A, Joseph L, Shir Y. The effects of nabilone on sleep in fibromyalgia: results of a randomized controlled trial. Anesth Analg 2010;110:604–10.
                                                                                  [91]. Ware MA, Wang T, Shapiro S, Collet JP; Team Cs. Cannabis for the management of pain: assessment of safety study (COMPASS). J Am Pain Soc 2015;16:1233–42.
                                                                                    [92]. Ware MA, Wang T, Shapiro S, Robinson A, Ducruet T, Huynh T, Gamsa A, Bennett GJ, Collet JP. Smoked cannabis for chronic neuropathic pain: a randomized controlled trial. Can Med Assoc J 2010;182:E694–701.
                                                                                      [93]. Whiting PF, Wolff RF, Deshpande S, Di Nisio M, Duffy S, Hernandez AV, Keurentjes JC, Lang S, Misso K, Ryder S, Schmidlkofer S, Westwood M, Kleijnen J. Cannabinoids for medical use: a systematic review and meta-analysis. JAMA 2015;313:2456–73.
                                                                                      [94]. Wilsey B, Marcotte T, Deutsch R, Gouaux B, Sakai S, Donaghe H. Low-dose vaporized cannabis significantly improves neuropathic pain. J Pain 2013;14:136–48.
                                                                                        [95]. Wilsey B, Marcotte T, Tsodikov A, Millman J, Bentley H, Gouaux B, Fishman S. A randomized, placebo-controlled, crossover trial of cannabis cigarettes in neuropathic pain. J Pain 2008;9:506–21.
                                                                                          [96]. Wilsey B, Marcotte TD, Deutsch R, Zhao H, Prasad H, Phan A. An exploratory human laboratory experiment evaluating vaporized cannabis in the treatment of neuropathic pain from spinal cord injury and disease. J Pain 2016;17:982–1000.
                                                                                            [97]. Wissel J, Haydn T, Müller J, Brenneis C, Berger T, Poewe W, Schelosky LD. Low dose treatment with the synthetic cannabinoid Nabilone significantly reduces spasticity-related pain. J Neurol 2006;253:1337–41.
                                                                                              [98]. Wong BS, Camilleri M, Busciglio I, Carlson P, Szarka LA, Burton D, Zinsmeister AR. Pharmacogenetic trial of a cannabinoid agonist shows reduced fasting colonic motility in patients with nonconstipated irritable bowel syndrome. Gastroenterology 2011;141:1638–47. e1637.
                                                                                                [99]. Zajicek J, Fox P, Sanders H, Wright D, Vickery J, Nunn A, Thompson A; UK MS Research Group. Cannabinoids for treatment of spasticity and other symptoms related to multiple sclerosis (CAMS study): multicentre randomised placebo-controlled trial. Lancet 2003;362:1517–26.
                                                                                                [100]. Zajicek JP, Hobart JC, Slade A, Barnes D, Mattison PG, Group MR. Multiple sclerosis and extract of cannabis: results of the MUSEC trial. J Neurol Neurosurg Psychiatry 2012;83:1125–32.
                                                                                                  Keywords:

                                                                                                  Cannabis; Chronic noncancer pain; Neuropathy; Systematic review; Meta-analysis; Number needed to treat

                                                                                                  Supplemental Digital Content

                                                                                                  Back to Top | Article Outline
                                                                                                  © 2018 International Association for the Study of Pain