Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Cannabinoids and codeine have similar effects on pain relief, but cannabinoids commonly cause adverse psychotropic effects


ACP J Club. 2002 Jan-Feb;136:18. doi:10.7326/ACPJC-2002-136-1-018

Related Content in this Issue
• Companion Abstract and Commentary: Review: Cannabinoids control chemotherapy-induced nausea and vomiting but increase the risk for side effects

Source Citation

Campbell FA, Tramèr MR, Carroll D, et al. Are cannabinoids an effective and safe treatment option in the management of pain? A qualitative systematic review. BMJ. 2001;323:13-6. [PubMed ID: 11440935] (All 2002 articles were reviewed for relevance, and abstracts were last revised in 2008.)



What are the effectiveness and safety of cannabinoids for pain management?

Data sources

Full publications of studies were identified by searching MEDLINE (1966 to 1999), EMBASE/Excerpta Medica (1974 to 1999), the Oxford Pain Database (1950 to 1994), and the Cochrane Library (1999, Issue 3) with the terms marijuana, marihuana, mariuana, cannabis, cannabinoids, tetrahydrocannabinol (THC), delta-9-THC, nabilone, pain, analgesia, and random. Bibliographies of relevant studies were scanned.

Study selection

Studies were selected if they were randomized controlled trials (RCTs) that compared cannabis or cannabinoids with an analgesic agent or placebo for pain management. Studies of experimental pain were excluded.

Data extraction

Data were extracted on study quality, interventions, and outcomes.

Main results

20 RCTs were identified, 9 of which met the selection criteria (222 patients). The types of cannabinoids used were oral delta-9-THC, 5 to 10 mg; an oral synthetic nitrogen analog of THC, 4 mg; oral benzopyranoperidine, 2 to 4 mg; and intramuscular levonantradol, 1.5 to 3 mg. No study assessed cannabis or other inhaled or smoked cannabinoids. The other active treatment drugs were oral codeine, 50 to 120 mg, and oral secobarbital, 50 mg. Studies varied in interventions and regimens, clinical settings, duration of follow-up, and outcomes. 5 studies were on cancer pain; 2 were on chronic, nonmalignant pain; and 2 were on postoperative pain. In 7 studies, follow-up was 6 to 7 hours; in 1 study, follow-up was 6 weeks; and in 1 study, follow-up was 5 months. In patients with cancer pain, oral delta-9-THC and an oral synthetic nitrogen analog of THC were as effective as codeine, but oral benzopyranoperidine was less effective than codeine. In patients with postoperative pain, intramuscular levonantradol was more effective than placebo, and adverse effects with levonantradol were common but mild. In all studies, adverse effects (often psychotropic) were common.


Cannabinoids and codeine have similar effects on pain relief, but cannabinoids commonly cause adverse psychotropic effects.

Sources of funding: Swiss National Research Foundation and Royal College of Nursing Institute.

For correspondence: Ms. F. A. Campbell, Pain Management Centre, Undercroft, South Block, Queen’s Medical Centre, Nottingham NG7 2UH, England, UK. FAX 44-115-970-9196.


Studies done in the past several years have suggested that cannabinoids may be effective in certain populations for alleviating nausea and symptoms of anorexia, although their clinical indications have been unclear because of the high incidence of psychotropic side effects and the lack of sound trials comparing cannabinoids with standard therapy (1, 2). Despite the lack of clear evidence on safety and efficacy, a survey done in 1994 showed that most physicians in the United Kingdom wished to have cannabinoids available by prescription (3). In this setting, Campbell and colleagues and Tramèr and colleagues completed their respective systematic reviews, which examined the efficacy of cannabinoids in the treatment of pain and chemotherapy-induced nausea and vomiting.

Cannabinoids in pain management provide analgesic effects similar to those of codeine. Cannabinoids commonly cause psychotropic side effects, but whether these effects are more common than with codeine is uncertain. At present, insufficient evidence exists to support the widespread use of cannabinoids over currently available analgesic therapies for pain management. However, the effects of inhaled cannabinoids were not assessed in this analysis.

The findings regarding chemotherapy-induced nausea and vomiting were somewhat more intriguing: Cannabinoids were shown to be slightly superior to such conventional antiemetics as prochlorperazine or metoclopramide. Although psychotropic side effects remained common, some of these effects, including euphoria and sedation, may have been perceived as beneficial. Many patients reported having a preference for cannabinoids, but whether this preference was because of beneficial side effects was not evaluated.

What is to be gained from these 2 reviews? The review by Campbell and colleagues assessing the efficacy of cannabinoids in the treatment of pain puts to rest the question of whether these oral or intramuscular agents are useful in that setting. Clearly, the review of published trials shows that unless more potent and less toxic cannabinoid derivatives are developed and studied, these agents are unlikely to have any important role as therapeutic options in pain management.

The results of the review by Tramèr and colleagues assessing the efficacy of cannabinoids in the treatment of nausea and vomiting are less clear. One can certainly argue that any relevance of the “slightly superior” outcomes of treatment with cannabinoids over conventional antiemetics may be offset by the concerns about the psychotropic side effects that routinely accompany the use of these agents. However, as pointed out in Belknap’s commentary (4) on the Voth study (1), no clinical trials were reported comparing cannabinoids to 5-hydroxytryptamine-3 receptor antagonists, which have become the standard of care in the prevention of chemotherapy-induced nausea and vomiting. Until such trials are done, the use of cannabinoids for this indication remains uncertain.

Of greater importance is the observation that patients often expressed a preference for the cannabinoids over other antiemetics for future chemotherapy despite the related psychotropic effects. This observation cannot be ignored. One can only speculate whether patient preference was a result of the beneficial side effects, such as euphoria, or another unknown effect. Irrespective of the reason, patient preference should be included as a valid outcome in the design of future studies that evaluate the efficacy of cannabinoids in the treatment of chemotherapy-induced nausea and vomiting.

Barry M. Kinzbrunner, MD
Vitas Healthcare Corporation
Miami, Florida, USA

Barry M. Kinzbrunner, MD
Vitas Healthcare Corporation
Miami, Florida, USA


1. Voth EA, Schwartz RH. Medicinal applications of delta-9-tetrahydrocannibinol and marijuana. Ann Intern Med. 1997;126:791-8. [PubMed ID: 9148653]

2. Nelson KA, Walsh D, Sheehan FA. The cancer-anorexia cachexia syndrome. J Clin Oncol. 1994;12:213-25. [PubMed ID: 8270980]

3. Meek C. Doctors want cannabis prescriptions allowed. BMA News Review. 1994;February:1.

4. Belknap S. Commentary on “Review: Delta-9-tetrahydrocannabinol is effective in alleviating cancer chemotherapy induced nausea.” ACP J Club. 1997 Nov-Dec;127:65. Comment on: Volta EA, Schwartz RH. Medicinal applications of delta-9-tetrahydrocannabinol and marijuana. Ann Intern Med. 1997;126:791-8.