496 Vol. 25 No. 6 June 2003Letters

Cannabis Reduces Opioid Dosein the Treatment of ChronicNon-Cancer Pain

To the Editor:Cannabinoids block pain responses in virtu-

ally every laboratory pain model tested. Inmodels of acute or physiological pain, canna-binoids are highly effective against thermal,mechanical, and chemical pain, and are com-parable to opioids in potency and efficacy.1 Inmodels of chronic pain, cannabinoids exhibitefficacy in the modulation of both inflamma-tory2 and neuropathic pain.3 Recent reviews de-scribe an endogenous cannabinoid systeminvolved in pain modulation that produces anal-gesia through the same brainstem circuitryinvolved in opioid analgesia.1,4,5,6 Althoughco-administration of ∆-9-tetrahydrocannabinol(THC) with µ opioid agonists can potentiatethe antinociceptive effects of each agent, anopioid is not required for cannabinoid analge-sia.5,6 Co-administration of a cannabinoid maylead to a lower opioid requirement. In an N-of-1 trial, oral THC reduced the pain of fam-ilial Mediterranean fever such that the use ofbreakthrough opioid for pain relief decreasedsignificantly.7

Recently, in Canada, the Medical MarijuanaAccess Program allows patients to apply toHealth Canada for access to dried cannabis formedicinal purposes. Although smoked canna-bis is not an ideal delivery system, it is efficientand results in plasma concentration curvesparallel to those seen after intravenous admin-istration.8 We present three patients who usedsmall doses of smoked marijuana in combina-tion with an opioid.

Case 1A 47-year-old woman with a ten-year history

of chronic progressive multiple sclerosis (MS)had headache, multisite joint pain, bladderspasm, and leg spasticity. Ambulation was signif-icantly compromised by the joint pain and legspasticity. She was wheelchair dependent, andalso suffered from severe insomnia and fatigue,which she attributed to the combination ofpain, bladder spasm, and leg spasticity. Physicalexamination revealed paraparesis, weakness

in the left upper extremity, tremor involvingboth hands, intranuclear ophthalmoplegiaand L’Hermitte’s sign. Previous treatment in-cluded steroids, physiotherapy, acupuncture,interdisciplinary pain management, intramus-cular injections of botulinum toxin, amitripty-line, fluoxetine, amantadine, acetaminophenwith codeine, oxycodone, nonsteroidal anti-inflammatory drugs (NSAIDs), and baclofen.The patient’s medications prior to access tosmoked marijuana consisted of long-acting mor-phine 75 mg per day, tizanidine 24 mg per day,and sertraline 150 mg at bedtime. In spite ofthese treatments, the patient did not obtainadequate control of her pain, spasticity, orsleep.

The patient received permission for access tosmoked marijuana and began to use a dose of2–4 puffs of smoked marijuana at bedtime ona regular basis. Over the next six months, themorphine was reduced to 45 mg per day, tizani-dine to 6 mg once per day, and sertraline tobetween 100 mg and 150 mg at bedtime. Thepatient reported improvement in pain, spas-ticity, bladder spasm, and sleep. The patientdenied any adverse side effects, other than shefelt somewhat “high” if she smoked more than4 puffs per dose. She was able to adjust the doseso that this did not occur. The patient receivedlegal access in the autumn of 2000 and contin-ues to use marijuana.

Case 2A 35-year-old HIV-positive man was initially

assessed in 1992 for four-month history ofHIV-related painful peripheral neuropathy in-volving the lower limbs and hands. His pain hadinitially involved the plantar aspect of his feetand had spread proximally to involve the lowerleg and thighs, and later his hands. The painwas described as severe “stinging, numb,tingling, and throbbing” pain that was unpre-dictable and poorly controlled. On physical ex-amination, positive findings included patchesof hypesthesia and hyperalgesia to pinprick test-ing in a stocking-glove distribution. Electromy-ography studies were reported as normal. Priortreatments had included trials of physiotherapy,acupuncture, psychological therapies, NSAIDs,tricyclic antidepressants, anticonvulsants, intra-venous and oral local anesthetics, and opioids.His medication regimen consisted of long-acting morphine 360 mg per day with morphine

Vol. 25 No. 6 June 2003 497Letters

sulfate 75 mg 4 times daily for breakthroughpain, and gabapentin 2,400 mg per day. Thepatient continued to report moderate to severelevels of pain.

After receiving approval to use marijuana in2000, the patient began using smoked mari-juana in a dose of 3–4 puffs 3–4 times per day.Over four months, the patient’s dose of mor-phine decreased to 180 mg per day and by 9months he had discontinued the morphine. BySeptember 2001, he discontinued gabapentin.The patient was able to manage without the useof an opioid or gabapentin until February 2002,at which time he developed an episode ofherpes zoster involving thoracic dermatomesT7 to T9 on the left side. Morphine was tempo-rarily re-introduced for 8 weeks during theacute episode in a dose of 15 mg three timesdaily. The patient reported that the smokedmarijuana was helpful for both the HIV neurop-athy and the herpes zoster pain, but he requiredthe additional analgesia provided by theopioid during the acute phase of the herpeszoster infection. He has since discontinued themorphine and remains opioid-free as of August2002. The patient denied side effects related tothe use of smoked marijuana.

Case 3A 44-year-old man presented with a 6-year

history of low back and left leg pain, which hadresulted from a work-related fall from a ten-foot high landing onto the edge of a metal tank,striking his lumbar spine. Physical examinationrevealed a decrease in lumbar lordosis, wastingof paravertebral musculature, decreased exten-sion at the lumbar spine, decreased straight legraising from the supine position to 60� on theright and to 45� on the left, tenderness to palpa-tion over the L3-4 to L5-1 facet joints, normaldeep tendon reflexes bilaterally and hypesthe-sia to pinprick testing over the lateral aspect ofthe left calf. Previous treatments had includedphysiotherapy, transcutaneous electrical nervestimulation, acupuncture, lumbar facet joint in-jections, radiofrequency facet neurolysis, andtrials of NSAIDs, tricyclic antidepressant analge-sics, muscle relaxants, tramadol, and long-acting morphine. The patient’s medicationsprior to the introduction of smoked marijuanaconsisted of long-acting morphine 150 mg perday and cyclobenzaprine 10 mg three timesper day.

The patient reported poor pain control withthis regimen and initiated a trial of smokedmarijuana. He reported using several puffs toone joint 4–5 times per day. After using thesmoked marijuana on a regular basis for 2weeks, the patient was able to decrease his mor-phine to 90 mg per day; after two more weeks,he decreased the morphine to 60 mg per dayand his cyclobenzaprine to 10 mg once daily. Hereported good pain control and was able tocontinue in his wage earning work. The patientcontinues to report improved pain controlusing the cannabis along with the lower doseof morphine as of August 2002.

CommentThese cases are consistent with preclinical

work demonstrating that cannabinoids exhibitanalgesic effects and may potentiate the antino-ciceptive effects of opioids. These patients wereable to decrease the dose of opioid by 60–100% as compared to before the regular useof smoked marijuana. With the introduction ofsmoked marijuana, each patient reportedbetter pain control. Unfortunately, the sourceof smoked marijuana used by patients, andthe percentage of THC in it, is unknown. Allpatients reported previous exposure to canna-bis at some time in their lives before the onset oftheir pain, and the relevance of this experiencealso is unknown. Standardized measures of painwere not used, and the information presentedwas based on the patients’ verbal report whenthey presented for follow-up appointments atthe Pain Management Unit. Nonetheless, thesecases suggest that further research regardingthe role of cannabinoids as analgesics and thecombination of cannabinoids with opioids inthe control of pain is needed.

Mary E. Lynch, MD, FRCPCAlexander J. Clark, MD, FRCPCQueen Elizabeth II Health Sciences CenterDalhousie UniversityHalifax, Nova Scotia, Canada

doi:10.1016/S0885-3924(03)00142-8

References1. Walker MJ, Strangman NM, Huang SM. Canna-binoids and pain. Pain Res Manage 2001;6:74–79.

2. Tsou K, Lowitz KA, Hohmann AG, et al. Suppres-sion of noxious stimulus-evoked expression of

498 Vol. 25 No. 6 June 2003Letters

Foslike immunoreactivity in rat spinal cord by aselective cannabinoid agonist. Neuroscience 1995;70:791–798.

3. Herzberg U, Eliav E, Bennett GJ, et al. The anal-gesic effects of R(+)-WIN 55,212-2 mesylate, a highaffinity cannabinoid agonist, in a rat model of neuro-pathic pain. Neurosci Lett 1997;221:157–160.

4. Martin WJ. Basic mechanisms of cannabinoid-induced analgesia. IASP Newsletter, Summer, 1999:3–6.

5. Richardson JD. Cannabinoids modulate pain bymultiple mechanisms of action. J Pain 2000;1:2–14.

6. Meng ID, Manning BH, Martin WJ, et al. An anal-gesia circuit activated by cannabinoids. Nature 1998;395:381–383.

7. Holdcroft A, Smith M, Jacklin A, et al. Pain reliefwith oral cannabinoids in familial Mediterraneanfever. Anesthesia 1997;52:483–488.

8. Ohlsson A, Lindgren JE, Whalen A, et al. Plas-ma delta-9-tetrahydrocannabinol concentrations andclinical effects after oral and intravenous administra-tion and smoking. Clin Pharmacol Ther 1980;28:409–416.