Article Text

Download PDFPDF

Tetrahydrocannabinol (THC) for cramps in amyotrophic lateral sclerosis: a randomised, double-blind crossover trial
  1. M Weber1,2,
  2. B Goldman1,
  3. S Truniger2
  1. 1Neuromuscular Diseases Unit/ALS Clinic, Kantonsspital St Gallen, St Gallen, Switzerland
  2. 2Department of Neurology, University Hospital Basel, Basel, Switzerland
  1. Correspondence to PD Dr Markus Weber, Neuromuscular Diseases Unit, Kantonsspital St Gallen, 9007 St Gallen, Switzerland;{at}


Background Many patients with amyotrophic lateral sclerosis (ALS) experience cramps during the course of the disease but so far, none of the medications used has been of proven benefit. The objective was to determine the effect of orally administered tetrahydrocannabinol (THC) on cramps in ALS patients.

Methods The authors conducted a randomised, double-blind, placebo-controlled crossover trial in 27 ALS patients suffering from moderate to severe (visual analogue scale (VAS); VAS≥4) daily cramps. There were 7 women and 20 men with a mean age of 57 years and a mean functional ALS score (ALSFRS-R) of 38.4. Patients were randomly assigned to receive 5 mg THC twice daily followed by placebo or vice versa. Each treatment period lasted for 2 weeks and was preceded by a 2-week drug-free observation period (run-in, wash-out period respectively). The primary outcome measure was change in cramp intensity as assessed by a VAS. Secondary outcome measures included the number of cramps per day, number of cramps during daytime and bedtime, intensity of fasciculations (VAS) as well as validated measures of quality of life (ALSAQ-40), quality of sleep (SDQ), appetite (FAACT) and depression (HADS).

Results Complete data were available from 22 patients. THC was well tolerated. There was no evidence for a treatment effect on cramp intensity, number of cramps, fasciculation intensity or any of the other secondary outcome measures.

Conclusions This interventional study with orally administered THC 5 mg twice daily did not demonstrate subjective improvement of cramp intensity in ALS patients.

  • Amyotrophic lateral sclerosis
  • cannabinoids
  • randomised trial
  • cramps
  • fasciculations
  • ALS
View Full Text

Statistics from


Cramps in amyotrophic lateral sclerosis (ALS) patients may affect any muscle or muscle group (eg, jaw, abdominal, chest, back, neck, or extremities). Severity varies from mild, without affecting daily activities and sleep, to disabling, where almost any voluntary muscle activity induces longstanding, severely painful cramping. In a survey of symptom management in 84 ALS patients cramps were present in 62%, requiring treatment in 56%.1 Various medications such as quinine sulfate, magnesium, lioresal, dantrolene, clonazepam, diphenylhydantoin and gabapentin are used for the treatment of cramps in ALS.2 However, treatment of cramps in ALS has been a primary endpoint only in a single open-label, non-randomised study.3 Moreover, quinine sulfate, the most widely prescribed medication for cramps in the US2 has recently been banned by the FDA.4 To date only five publications of randomised, placebo-controlled clinical trials in ALS mention the effect of the study medication on cramps (Vitamin E, gabapentin, riluzole, dextrometorphan/quinidine).5–9 None of them, including gabapentin seemed to alleviate cramps. This implies that to date there are no medications with proven benefit for the treatment of cramps in ALS.

ALS patients who smoke herbal cannabis or drink hemp tea report a lessening of cramps and fasciculations. Up to 10% of ALS patients use cannabis (marijuana) for symptom control (eg, cramps, spasticity, drooling).10 A recent, randomised double-blind, crossover study investigating the safety and tolerability of Δ9 –tetrahydrocannabinol (THC, dronabinol) in 19 ALS patients revealed a non-significant reduction in cramps and fasciculations.11 In addition, symptomatic effects in insomnia and appetite were found. Oral doses from 2.5 to 10 mg were well tolerated.

The molecular basis for the pharmacological actions of cannabinoids are the cannabinoid receptor type 1 (CB1) which is expressed predominantly in the central and peripheral nervous system and cannabinoid receptor type 2 (CB2) with high expression levels in the spleen and thymus. CB1 is localised presynaptically in the brain, spinal cord and peripheral nerve.12 It is accepted that cannabinoid agonists such as THC inhibit glutamate release and enhance the effect of the inhibitory neurotransmitter GABA.12 The role of glutamate as the main excitatory neurotransmitter in the generation of cramps has not been addressed in the scientific literature, but it has been proposed that cramps in ALS may result from impairment of GABAergic function.13 The current understanding of the pathophysiology of cramps suggests that peripheral nerve hyperexcitability and motor neuron bistability (anterior horn cell level) may play a role.14–17 The distribution of CB1 receptors in the peripheral nerves including the neuromuscular junction and spinal cord may pose an advantage of THC over medications acting solely at the peripheral nerve/neuromuscular junction (quinine sulfate) or centrally (neurontin).18 19 Consequently, in view of the lack of randomised controlled trials, we sought to investigate the effectiveness of THC in the treatment of cramps by using a randomised, double-blind, placebo-controlled crossover design. Based on patients' self-reports and a single pilot trial11 it was hypothesised that severity of cramps will improve significantly in patients treated with THC compared with treatment with placebo. We also evaluated the potential role of THC in treating other aspects of ALS.


The study was approved by the local research ethics committee of the Kanton St Gallen, by the federal Department of Health (Bundesamt für Gesundheit, BAG) and Swiss Medicines Control Agency (Swissmedic). Written informed consent was obtained from all participants prior to enrolment. The study was registered on, clinical trial identifier number NCT0081285.


Adult patients diagnosed as having possible, probable laboratory supported, probable or definite ALS according to the revised El Escorial criteria20 and with an average daily cramp severity score of 4 and more (on a visual analogue scale (VAS)) were eligible. They were recruited from the ALS outpatient clinic at the Kantonsspital St Gallen, Switzerland. Further inclusion criteria were the ability to communicate and report adverse events by phone, no use of cannabis or cannabinoids for at least 1 month prior to study entry, agreement not to use them during the study and willingness to stop driving or operating dangerous machinery for the study period. Patients were allowed to take any other concomitant medications for the treatment of ALS or symptom control (eg, riluzole, antidepressants) but were asked not to change this medication during the study period. Patients were excluded if they were demented, had significant concomitant illness(-es), had a past history of a psychiatric disorder, explicitly of schizophrenia, were current drug or alcohol abusers, or had experienced significant adverse effects from or hypersensitivity to THC or any cannabinoid.

Study design

A randomised controlled double-blind crossover design consisting of four periods each of 2 weeks duration was applied (figure 1). Table 1 shows a power calculation for the primary outcome measure (VAS) with sample sizes of 18 and 24 subjects, assuming within subject standard deviations of σ=0.5, 1 and 1.5, and a mean reduction of Δ=1 (in VAS) at a significance level of α=0.05. A sample size of 24 subjects would have allowed a 20% drop out of patients (eg, due to possible side effects) without changing the power significantly.

Figure 1

Study sequences and visits. THC, tetrahydrocannabinol.

Table 1

Power calculation for the primary outcome measure (VAS) comparing placebo versus treatment

A computer-generated blockwise random sequence was provided by the study statistician. Each subject was given a subsequent number. Two copies of the complete randomisation list were prepared. One was used by the hospital pharmacy in conjunction with the patient's trial number to dispense either THC and then placebo or vice versa. Another was stored at a confidential location by the statistician until the time of unblinding. All investigators and patients were blinded to treatment allocation for the duration of the study.


During the run-in period and wash-out period, patients received no medication. Study visits were scheduled at study entry, and after each treatment period. At study entry, patients were instructed how to complete the cramp diary and take the study medication. The study medication was given out at the first and second study visit, and the unused medication recollected after each treatment period (study visits 2 and 3). Safety assessments consisted of monitoring and recording all adverse events and serious adverse events, the regular monitoring of haematology, blood chemistry and urine values, regular measurement of vital signs and body weight measurements at all study visits. Patients were provided with a phone number (7 h, 24 h) so that they could contact the investigator if adverse reactions were to occur.


The active study medication consisted of sesame oil containing Δ9-THC (dronabinol) 2.5 mg per three drops. Dronabinol was provided by THCpharm (Frankfurt am Main, Germany). The sesame oil was prepared by the local hospital pharmacy. The placebo medication contained only sesame oil and looked and tasted identical to the active medication. The study medication (six drops equivalent to 5 mg Dronabinol) was taken twice daily. All other medication was taken as usual. Patients were allowed to use clonazepam 1–2 mg up to five times daily as escape medication for severe, disabling cramps but had to document this in the cramp diary.

Outcome measures

Beginning with the run-in period and extending throughout the entire study, patients completed a diary each morning. The visual analogue rating scales (0–10; 0=no symptoms, 10=most severe symptoms) were explained to each patient at the beginning of the study and one sample of it completed in the outpatient department. The primary outcome measure was the average cramp intensity during the preceding 24 h rated every morning on the VAS. Cramp intensity as assessed by a VAS is the most common measure used in cramp trials.21

Secondary outcome measures were the number of cramps during the preceding 24 h, during daytime and bedtime (documented in the diary before going to bed and after awakening) and severity of fasciculations (VAS). Validated, patient-based measures of functional ability (ALSFRS-R22; ALS functional rating scale revised), quality of life (ALSAQ-4023; ALS assessment questionnaire), quality of sleep (SDQ24; sleep disorder questionnaire, items F003, F004, F023, F045, F055, F154, F155), appetite (FAACT25; functional assessment of anorexia/cachexia therapy) and depression (HADS26; hospital anxiety and depression scale) were applied during each study visit. All assessments were carried out by the same investigator (GB). In addition, cramp diaries where checked for completeness at the study visits.


Statistical analyses were conducted according to a pre-established analysis plan. For daily rated outcome measures, the mean of the last 3 days of each period was calculated. To compare the verum with the placebo values, a linear mixed effect model with fixed factors sequence, period and treatment was performed for all measures. All comparisons were adjusted for the corresponding values of the preceding period (run-in period, wash-out period respectively). A 95% CI of the difference of means was estimated for all parameters. A p value <0.05 was considered significant. All analyses were performed using the statistical software R version 2.7.1, a language and environment for statistical computing. Linear mixed-effect models were performed using the function lme in the package nlme.


Study patients

Recruitment took place between May 2005 and February 2008. A total of 27 patients were eligible (cramp severity ≥4, daily cramps) and randomised (figure 2, flow chart). Fourteen were randomised to sequence 1/0 (THC followed by placebo) and 13 to sequence 0/1 (placebo followed by THC). Two patients withdrew during the run-in period because cramps spontaneously disappeared. Two other patients who were admitted to hospital did not complete the diary. Another patient did not document the run-in period while on vacation, leaving 11 patients in each sequence who completed the entire study. Their data are included in the intention-to-treat analysis.

Figure 2

Flow of participants through the trial. THC, tetrahydrocannabinol.

Seven patients were women, and 20 patients were men, with a mean age of 57±12 years (range 34–78) and a mean ALSFRS-R of 38.4±6.0 (range 22–46). Four patients had bulbar, 23 limb onset. The mean disease duration was 20±18 months (4–63). All patients were on riluzole (Rilutek) twice daily. At study entry only a single patient used medications for the treatment of cramps (quinine sulfate 250 mg twice daily). The medication was kept stable during the entire study.

Primary outcome measure

The mean baseline ratings (entire run-in period) are shown in table 2. The mean of the last three daily ratings of each period was used in the statistical analysis to compare the values of the verum and placebo period. There was no significant treatment effect (table 2). The time course over the entire study period of medians and their variability was slightly different between sequence 1/0 and sequence 0/1 (figure 3). However, no significant period or sequence effect was present (p>0.1).

Table 2

Results of daily rated outcome measures (OM): baseline values (14-day run-in period)

Figure 3

Box plots of cramp intensity (VAS) of each period over entire study period. THC, tetrahydrocannabinol; x-axis, VAS, on a visual analogue scale; y-axis, periods.

Secondary outcome measures

Analyses were performed the same way as for the primary outcome measure. None of the secondary outcome measures (number of cramps per day, number of cramps during daytime and night-time, fasciculation intensity) showed a significant difference between the placebo and THC values (table 2). Measures of quality of life (ALSAQ-40), sleep (SDQ), appetite (FAACT) and depression (HADS) were not significantly different between the placebo and THC period (table 3).

Table 3

Results of outcome measures rated after each treatment period at each study visit

Adjusted analysis

An adjusted analysis was performed for cramp intensity, total number of cramps and intensity of fasciculations using the means of the 14-day run in period as baseline values. They were compared with the mean of the last three daily ratings of the verum and placebo period. There was no significant treatment effect.

As a potential carryover effect cannot be adequately estimated in a 2×2 crossover design, the time course of the primary outcome measure over the entire study period was modelled in each patient (supplemental data). In sequence 1/0, four out of 11 patients experienced lessening of cramp intensity during the THC period which persisted during the subsequent wash-out period and placebo period compatible with a significant carryover effect.

Adverse events

Two serious adverse events occurred. Both patients were admitted to hospital. One patient developed pneumonia during the wash-out period (after THC period) and later died; the other developed deep venous thrombosis before the THC period. These adverse events were felt not to be study-related. None of the remaining patients withdrew from the study. One patient experienced mild dizziness while on THC (sequence 0/1). The patient continued the study with half the dosage. Otherwise, none of the patients reported any side effects.


The main result of this RCT is that orally administered THC (2×5 mg daily) does not seem to alleviate cramp intensity, cramp frequency and fasciculation intensity assessed by a VAS. There was also no effect on measures of quality of life (ALSA Q40), sleep (SDQ), appetite (FAACT) and depression (HADS). This contrasts with patients' self reports and a single pilot trial.11 The lack of a treatment effect found in this study could reflect failure to achieve sufficiently high systemic medication levels. This might be, in part, due to the limited treatment period of 2 weeks which is shorter than in the THC pilot trial (4-week treatment period) of Gelinas and colleagues.11 Only a single patient in this study experienced side effects, which suggests that plasma levels of THC, which is the main psychoactive cannabinoid, and its metabolites were low. In a study in MS patients with a mean daily dose of 25 mg, the greatest benefit was noted in patients with painful muscle spasms, but side effects occurred in 88.2%.27 We did not implement a dose titration phase,27–29 as this bears a high risk of unmasking the blinding due to side effects. This has been reported in up to 70% of the study population in cannabinoid trials.28 29 The dose chosen aimed to avoid psychoactive side effects and was based on a pilot trial in ALS in which single doses of 2.5–10 mg caused relatively few side effects.11

Baseline values, standard deviations and variability of the primary outcome measure over the course of the study were different between treatment groups. However, the statistical model accounted for this imbalance, making it unlikely that group variability may have unmasked a treatment effect. Nevertheless, a potential carryover effect cannot be adequately estimated in a crossover design. Modelling of individual data indeed suggested that in some patients, a carryover effect occurred (see online figure). A possible explanation is the metabolism of THC. THC is well absorbed in the gut and underlies a high first metabolism effect in the liver (80%) before it enters the blood and is stored in fatty tissues (eg, brain). The slow release of THC from these tissues produces low levels of THC concentration after a single dose.30 The mean urinary excretion half life of inactivated metabolites such as THCCOOH in healthy individuals varies between 1.3 days in infrequent and 4.1 days in frequent users.31 However, nothing is known about the pharmacokinetics of THC and its metabolites in ALS patients. This makes it difficult to determine the length of the wash-out period. The wash-out period of 2 weeks in this study was based on previously published THC trials in MS patients and Parkinson's disease.28 29 One of our patients tested positive for THC 3 weeks after stopping intake, suggesting that pharmacokinetics and excretion half life in ALS may be even more variable than in healthy individuals.

Another factor which may have confounded the study results is the natural course of cramps, which is unknown. Cramps develop and disappear during the course of the disease, but no formal studies exist analysing these dynamic changes. In two of our eligible patients, cramps spontaneously disappeared during the run-in period. Such spontaneous fluctuations may interfere with the treatment and placebo periods, and would also argue for a parallel group study design in future trials.

Despite the negative results for the symptomatic treatment of cramps, the finding that THC is tolerable to ALS patients may be important, because cannabinoids exert neuroprotective effects in a variety of in vivo32 and in vitro models.33 Cannabinoids delay disease progression and prolong survival when administered after symptom onset in mouse models.34–36 Moreover, the cannabinoid system seems to be involved in the pathogenesis of ALS. Microglial CB 2 receptor expression is upregulated in the spinal cord of ALS patients,37 and the sensitivity of the CB1 receptor is increased in SOD mice.38

In summary, a low dosage of 5 mg twice daily is well tolerated in ALS patients but does not seem to eleviate cramp frequency, cramp severity and fasciculation intensity. It cannot be ruled out that higher doses may have been beneficial for symptom control. This study has highlighted some of the difficulties inherent in crossover studies with cannabinoids. Due to the unknown pharmacokinetics of THC, variable half-lives, a possible carryover effect and the unknown natural course of cramps, future studies with THC in ALS patients should apply a parallel group study design rather than a crossover design.


We are thankful to R Brenneisen, Department of Clinical Research, University of Berne, for conducting THC stability measurements, to D Schilling, Department of Pharmacy, Kantonsspital St Gallen, for preparing the study medication, and to A Schoetzau, University of Basel, for the statistical analyses.


View Abstract


  • Funding This study received funding from the Amyotrophic Lateral Sclerosis Association (ALSA) clinical management research grant.

  • Competing interests None.

  • Ethics approval Ethics approval was provided by the Kantonale Ethikkomission des Kantons St Gallen, Kantonsspital St Gallen, 9007 St Gallen, Switzerland.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.