Search tips
Search criteria 


Logo of bmjLink to Publisher's site
BMJ. 2008 January 26; 336(7637): 167–168.
Published online 2008 January 8. doi:  10.1136/bmj.39434.444583.80
PMCID: PMC2213791

Cannabinoids for chronic pain

Steven P Cohen, associate professor of anaesthesiology

Is effective but research is needed to decide which subgroups of people benefit most

Cannabis has been cultivated as an elixir for pain since as far back as 2000 years BCE. Amid increasing reports of psychosis, addiction, and other adverse effects the therapeutic use of smoked cannabis in the United States waned in the late 1930s, as a result of the Marihuana Tax Act and subsequent legislative measures, which posed nearly insurmountable obstacles for doctors. The past decade has, however, seen a resurgence in the interest in cannabinoids for alleviating pain, with the identification of at least two subtypes of cannabinoid receptors, and myriad clinical studies examining the effectiveness of tetrahydrocannabinol derivatives for acute pain, pain caused by chronic non-malignant disease, and pain from cancer. Evidence of the effectiveness of cannabinoids is strong for treating cancer, central pain, and spasticity related pain; the evidence is mixed for treating acute pain and weak for treating peripheral neuropathic pain.

In the accompanying paper, Frank and colleagues report the results of a randomised crossover controlled trial comparing the effectiveness of dihydrocodeine with the synthetic cannabinoid nabilone.1 The trial studied 96 patients with diverse neuropathic pain conditions.

The authors found that the mean visual analogue score at six weeks was 6.0 mm greater in people taking nabilone than in those taking dihydrocodeine (95% confidence interval 10.5 to 1.4) for the available case analysis and 5.6 mm (10.3 to 0.8) greater for the per protocol analysis. Although outcomes were not tabulated by diagnosis, only a few patients had central or spasticity related pain, the two pain conditions most responsive to treatment with cannabinoids. Almost half of the participants had post-traumatic pain or postsurgical pain, which often have multiple pain generators, a high prevalence of underlying psychopathology, and are difficult to treat.

These results are consistent with the literature on opioids and cannabinoids. Neuropathic pain was once believed to be refractory to opioids, but there is now indisputable evidence that it is responsive to opioids, albeit at higher doses than those used to treat nociceptive pain.2 In contrast, the evidence supporting cannabinoids for peripheral neuropathic pain and other chronic non-malignant pain disorders, while generally positive, is less strong and mostly confined to other synthetic analogues.3,4,5 Most people in Frank and colleagues’ study had peripheral neuropathic pain.

Before embarking on large scale trials designed to establish who will benefit most from an intervention, investigators should first ensure that trials demonstrate internal validity because failure to do so threatens to undermine the very concept of any new treatment. Even if a subgroup of patients might have benefited from nabilone, the current study design was not powered to identify them.

So how should we interpret these results and put them into practice? Given the poor analgesic effect and the high incidence of adverse effects (more than 80% of participants reported tiredness, more than 50% reported sleeplessness and sickness, and more than 25% reported tingling or feeling strange) seen in this and other studies, cannabinoids should not be used as first line agents in patients who present with uncategorised pain. Yet previous trials have shown that a small subset of patients with neuropathic pain and other chronic pain disorders may benefit from treatment with cannabinoids.3,4,5,6 The most compelling evidence for the effectiveness of cannabinoids is in the relief of pain associated with multiple sclerosis and, to a lesser extent, other central pain syndromes.7 8 So why is there a discrepancy?

One key difference between the current study and studies with positive findings5,6,7,8 is that Frank and colleagues used allodynia (which signifies evoked pain) and sympathetic dysfunction as inclusion criteria. Collectively, these signs are experienced by only a small subset of patients with chronic pain. In the current era, in which treatment of pain is ideally based on mechanism(s) of pain, pharmacotherapy should preferentially target specific pain generators.9 In multiple sclerosis, the condition with the strongest evidence for the effectiveness of cannabinoids, the most common manifestations of pain are continuous or spontaneous dysaesthesias, which are mechanistically different from allodynia.10 In addition, the evidence to support a sympathetic component in multiple sclerosis is scant.

Another potential flaw in the current study is that it assessed outcomes after only six weeks, which is too short a time frame to detect the low but clinically significant risk of adverse psychiatric effects such as psychosis, depression, cognitive deficits, and addiction.4 Similarly, short-term reductions in pain scores do not necessarily translate into long-standing improvements in pain, functional capacity, and psychological wellbeing. Future trials evaluating treatment with cannabinoids might consider using risk assessment tools, like those used for opioids.11

Finally, there is the problem of the prohibition surrounding the medicinal use of cannabinoids. The therapeutic use of cannabinoids is about 50 years behind that of opioids, and the associated stigma poses a considerable barrier to research and development.12 Yet history tells us that the line dividing what is and isn’t socially acceptable behaviour is often arbitrary and inseparable from the cultural-political context. Not so long ago, blood-letting and the curative use of heavy metals were considered state of the art medical care. Although many such practices have been abandoned, others remain that do more harm than good.

Strong evidence supports the use of cannabinoids for chronic pain, but more research is needed to determine which diagnoses, pain characteristics, and clinical variables are most amenable to treatment; the long-term effectiveness of these drugs; optimal drug selection and dosages; the risk-benefit ratio of combining cannabinoids with other drugs; and how adverse effects can be minimised. This research can occur only in a political climate conducive to the continued use and investigation of cannabinoids to alleviate pain and suffering.


Competing interests: None declared.

Provenance and peer review: Commissioned; not externally peer reviewed.


1. Frank B, Serpell MG, Hughes J, Matthews JN, Kapur D. Comparison of analgesic effects and patient tolerability of nabilone and dihydrocodeine for chronic neuropathic pain: randomised, crossover, double blind study. BMJ 2008. 10.1136/bmj.39429.619653.80.
2. Benedetti F, Vighetti S, Amanzio M, Casadio C, Oliaro A, Bergamasco B, et al. Dose-response relationship of opioids in nociceptive and neuropathic postoperative pain. Pain 1998;74:205-11. [PubMed]
3. Campbell FA, Tramer MR, Carroll D, Reynolds JM, Moore A, McQuay HJ. Are cannabinoids an effective and safe treatment option in the management of pain? A qualitative systematic review. BMJ 2001;323:13-6. [PMC free article] [PubMed]
4. Rice AS, Lever I, Zarnegar R. Cannabinoids and analgesia, with special reference to neuropathic pain. In: McQuay HJ, Kalso E, Moore RA, eds. Systematic reviews in pain research: methodology refined. Seattle: IASP Press, 2007:233-46.
5. Abrams DI, Jay CA, Shade SB, Vizoso H, Reda H, Press S, et al. Cannabis in painful HIV-associated sensory neuropathy. A randomized placebo-controlled trial. Neurology 2007;68:515-21. [PubMed]
6. Notcutt W, Price M, Miller R, Newport S, Phillips C, Simmons S, Sansom C. Initial experiences with medicinal extracts of cannabis for chronic pain: results from 34 “N of 1” studies. Anaesthesia 2004;59:440-52. [PubMed]
7. Svendsen KB, Jensen TS, Bach FW. Does the cannabinoid dronabinol reduce central pain in multiple sclerosis? Randomised double blind placebo controlled trial. BMJ 2004;329:253. [PMC free article] [PubMed]
8. 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 study. Lancet 2003;362:1517-26. [PubMed]
9. Cohen SP, Verdolin MH, Chang AS, Kurihara C, Morlando BJ, Mao J. The IV ketamine test predicts subsequent response to an oral dextromethorphan treatment regimen in fibromyalgia patients. J Pain 2006;7:391-8. [PubMed]
10. O’Connor AB, Schwid SR, Herrmann DN, Markman JD, Dworkin RH. Pain associated with multiple sclerosis: systematic review and proposed classification. Pain 2007 Online first 8 October 2007.
11. Webster LR, Webster RM. Predicting aberrant behaviors in opioid-treated patients: preliminary validation of the opioid risk tool. Pain Med 2005;6:432-42. [PubMed]
12. Zimmerman B, Bayer R, Crumpacker N. Is marijuana the right medicine for you? A factual guide to the medical uses of marijuana. New Canaan, CT: Keats Publishing, 1998.

Articles from BMJ : British Medical Journal are provided here courtesy of BMJ Group