ASPINAL CORD INJURY is an insult to the spinal cord

Transcription

1 ORIGINAL ARTICLE A Randomized, Double-Blinded, Crossover Pilot Study Assessing the Effect of Nabilone on Spasticity in Persons With Spinal Cord Injury Sepideh Pooyania, FRCPC, Karen Ethans, FRCPC, Tony Szturm, PhD, Alan Casey, FRCPC, Daryl Perry, FRCPC 703 ABSTRACT. Pooyania S, Ethans K, Szturm T, Casey A, Perry D. A randomized, double-blinded, crossover pilot study assessing the effect of nabilone on spasticity in persons with spinal cord injury. Arch Phys Med Rehabil 2010;91: Objectives: To determine whether nabilone, a synthetic cannabinoid, alleviates spasticity in people with spinal cord injury (SCI). Design: A double-blind, placebo-controlled crossover study. Setting: Outpatient rehabilitation clinics. Participants: We recruited volunteers (N 12) with SCI and spasticity. One subject, a paraplegic man, dropped out of the study because of an unrelated cause. Eleven subjects completed the study; all subjects were men with an average age of years; 6 of them were persons with tetraplegia, and 5 were persons with paraplegia. Interventions: The subjects received either nabilone or placebo during the first 4-week period (0.5mg once a day with option to increase to 0.5mg twice a day), and then outcome measures were assessed. After a 2-week washout, subjects were crossed over to the opposite arm. Main Outcome Measures: The primary outcome was the Ashworth Scale for spasticity in the most involved muscle group, in either the upper or lower extremities, chosen by the subject and clinician. The secondary outcomes included the sum of the Ashworth Scale in 8 muscle groups of each side of the body measured by the clinician; Spasm Frequency Scale and visual analog scale, reported by the subject; Wartenberg Pendulum Test, in order to quantify severity of spasticity; and the Clinician s and Subject s Global Impression of Change. Results: One subject dropped out during the placebo arm because of an unrelated urinary stricture, and 11 subjects completed the study. There was a significant decrease on active treatment for the Ashworth in the most involved muscle (mean difference SD, ; P.003), as well as the total Ashworth score (P.001). There was no significant difference in other measures. Side effects were mild and tolerable. Conclusions: Nabilone may be beneficial to reduce spasticity in people with SCI. We recommend a larger trial with a more prolonged treatment period and an option to slowly increase the dosage further. From the Section of Physical Medicine and Rehabilitation, Departments of Medicine (Pooyania, Ethans, Casey, Perry) and Physical Therapy (Szturm), University of Manitoba, Winnipeg, MB, Canada. Supported by grants from the Valeant Company and Scholarship to Advance Research in Spasticity Program. No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefit on the authors or on any organization with which the authors are associated. Reprint requests to Sepideh Pooyania, MD, Physical Medicine and Rehabilitation Dept, 800 Sherbrook St, Winnipeg, MB, Canada R3A 1M4, sepidehali@yahoo.com /10/ $36.00/0 doi: /j.apmr Key Words: Cannabinoids; Muscle spasticity; Nabilone [name of substance]; Rehabilitation; Spinal cord injuries by the American Congress of Rehabilitation Medicine ASPINAL CORD INJURY is an insult to the spinal cord resulting in alterations of normal motor, sensory, and autonomic function. 1 The estimated incidence rate of SCI in Canada is approximately 35 per million population, or 1050 new injuries each year, with the primary cause being vehicular accidents, followed by falls, violence, and sports injuries. 2 Spasticity is an abnormal, velocity-dependent increase in muscle tone (tonic stretch reflexes) with exaggerated tendon jerks as a result of the hyperexcitability of the muscle stretch reflex. 3 The epidemiology of spasticity after traumatic SCI affirms the significance of this medical problem. 4-6 Spasticity by itself is not a disease state, and treatment should be made based on the functional limitations it can create. Treatment of spasticity includes nonpharmacologic management such as positioning, stretching, and physical modalities. Pharmacologic management of spasticity includes baclofen, benzodiazepines, dantrolene, clonidine, tizanidine, and gabapentin. Most of these medications share common side effects of sedation and asthenia. Nerve block, chemical denervation, and surgical interventions have also been used to treat spasticity. Marijuana is the common name for cannabis. Cannabis is composed of the leaves and flowering tops of the Cannabis sativa plant, which is known to contain more than 60 oxygencontaining aromatic hydrocarbon compounds known as cannabinoids. One of these cannabinoids, THC, is responsible for most of the plant s psychoactive properties. 7 Although not approved for clinical use, cannabis has shown benefits for patients with SCI-related spasticity. 8,9 Similar survey reports of the subjective benefits on spasticity with cannabis use can be found in the MS literature. 10 The illegality of smoking cannabis along with issues of varying potency makes smoked marijuana a difficult substance to analyze for efficacy in research studies. Clinical evidence supporting the use of oral THC in the SCI population has been described in different studies. 11,12 Clinical trials of THC in patients with MS have shown that the oral cannabinoids dronabilone (dronabinol) and nabilone, can produce objective and/or subjective relief from spasticity, pain, and tremor. 7 ASIA MS SCI THC VAS List of Abbreviations American Spinal Injury Association multiple sclerosis spinal cord injury delta-9-tetrahydrocannabinol visual analog scale

2 704 NABILONE AND SPASTICITY IN PERSONS WITH SPINAL CORD INJURY, Pooyania Another study found no change in overall spasticity scores on the Ashworth Scale. Secondary outcome measures in this study included the Rivermead Mobility Index, a timed 10-m walk, and 4 self-completion questionnaires: the United Kingdom Neurological Disability Score, the Barthel Index, the General Health Questionnaire, and a series of 9 category-rating scales. 13 Nabilone has a greater bioavailability and a longer duration of action than dronabilone, making nabilone an ideal drug to investigate in the area of spasticity management. There are no published studies regarding the effects of nabilone on spasticity in SCI. A similar pilot project was conducted in 2005 at our center. There was a problem with recruitment of volunteers, because the primary outcome for that study required that subjects have a certain level of severe spasticity at the knee extensors, which many screened patients did not have. In addition, at that time, the 0.5-mg formulation was not available. Thus, the starting dose was 1mg, and 50% of subjects who were enrolled had to drop out of the study because of the side effects they experienced while taking nabilone. Therefore, in this study, we started with a nabilone dosage of 0.5mg/d. The aim of this study was to determine whether nabilone has a significant beneficial effect on spasticity in people with SCI. A secondary objective of the study was to find a tolerable dosage of medication that would permit the design of larger trials. In this exploratory pilot study with a small number of subjects, our goal was to determine whether 0.5 to 1.0mg/d is a tolerable dosage of nabilone to control spasticity in persons with SCI. METHODS This randomized, double-blind, placebo-controlled, crossover, parallel-group study was carried out in an outpatient clinic (Rehabilitation Hospital). The study was approved by the Ethics Committee of the Faculty of Medicine of the University of Manitoba and by the Therapeutic Product Directorate of Health Canada. Participants Subjects with SCI were eligible for the study if they were aged 18 to 65 years, the level of injury was at C5 (ASIA grade A D) or below, and the injury occurred more than 1 year previously. They had to have a stable neurologic level (ie, no change in ASIA neurologic level in the last 6 months), with moderate spasticity (Ashworth 3). Spasticity medications had to be unchanged for at least 30 days before inclusion, and no botulinum toxin injections for more than 4 months. They were excluded if they had heart disease (because cannabinoids can reduce heart rate and blood pressure), a history of psychotic disorders, schizophrenia, or any active psychologic disorder (because cannabinoids may cause psychologic disorders), a previously documented sensitivity to marijuana or other cannabinoid agents, severe liver dysfunction, cognitive impairment, a major illness in another body area, if they were pregnant or a nursing mother, had a history of drug dependency, smoked cannabis less than 30 days before the onset of the study or were unwilling not to smoke during the study, or if they had fixed tendon contractures. All subjects had blood tested for liver function, because cannabinoids may cause liver dysfunction. Subjects were recruited from Spasticity and Spinal Cord Injury Clinics and from the Spinal Cord Injury Research Database, and were included in the study in accordance with the eligibility and exclusion criteria. Twelve subjects were enrolled into the study. After satisfying the inclusion and exclusion criteria and signing an informed consent, basic outcome measures were conducted. The subjects blood pressure, pulse rate, and weight were documented. Outcome Measures The primary outcome measure for assessing spasticity in the most involved muscle group of the body, as chosen by the subject and clinician, was the Ashworth Scale. The Ashworth Scale is the most common clinical approach to the routine measurement of levels of spasticity. First proposed by Ashworth 14 in 1964, the scale grades resistance to rapid passive movement across a relaxed joint on an ordinal scale of 0 to 4 (appendix 1). A Modified Ashworth Scale has since been proposed by Smith and Bohannon, 15 adding a 1 point. This version of the scale is used clinically; however, it is not routinely used in research because the distances between 1 and 1 are not equal, making the data categorical. 16 The specific muscle group presenting with the greatest functional problems from spasticity was identified by both the clinician and subject at the first visit and subsequently followed up closely. For the Ashworth assessment, the subject was supine and resting for at least 15 minutes before testing. The secondary outcome measures included (1) the sum of the Ashworth Scale in the 8 muscle groups bilaterally (although Ashworth in the most involved muscle group was already considered our primary outcome measure, the sum of the Ashworth Scale in both upper and lower extremities was needed to reflect the general effect of antispasticity medication on the entire musculoskeletal system); (2) the Spasm Frequency Scale; (3) the VAS for spasticity; (4) the Wartenberg Pendulum Test; (5) the Subject s Global Impression of Change; and (6) the Clinician s Global Impression of Change. The sum of the Ashworth Scale was assessed in 8 muscle groups bilaterally including the elbow flexors and extensors, the wrist and finger flexors, the hip adductors, the knee flexors and extensors, and the foot plantar flexors. The Spasm Frequency Scale 17 testing was conducted by asking subjects to complete the Spasm Frequency Index throughout the period in question (appendix 2). Self-reporting and clinical examination scales used to assess spasticity have been shown to represent different dimensions of the clinical problem. 18 Therefore, a comprehensive assessment of spasticity that includes a subjective spasm frequency scale is appropriate. The VAS is valid and reliable in rating pain intensity. The VAS has also been used to evaluate the effect of clonidine on spasticity in people with SCI. 19 Skold 20 recommends that patients rate their spasticity on a 100-mm, pencil-and-paper VAS scale (fig 1) before and during interventions aimed at reducing spasticity. All subjects were requested to complete the 100-mm, pencil-and-paper VAS scale during each clinic visit. The Wartenberg Pendulum Test, 21 first described in 1951, can be used to assess spasticity. This test is performed with the subject either lying supine or seated. The examiner extends the knee and then drops the foot, allowing the leg to swing. In a subject without spasticity, the leg will oscillate about 6 times before stopping. In spasticity, Fig 1. VAS for spasticity.

3 NABILONE AND SPASTICITY IN PERSONS WITH SPINAL CORD INJURY, Pooyania 705 the number of swings is reduced in accordance with the velocity dependence. 18 Fowler et al 22 studied the Wartenberg Pendulum Test in patients with cerebral palsy and found that the first swing excursion was the best predictor of spasticity and not the duration of swings. The Wartenberg Pendulum Test was conducted both with the patient lying supine and while seated with knees flexed. The examiner first placed a small accelerometer (approximately 1 1 1cm) on the patient s right leg. The examiner extended this knee passively and released the foot, allowing the leg to swing. This was then repeated for the other side, which resulted in data from 22 separate legs rather than 11 patients, because each subject was measured twice. The accelerometer was interfaced with a computer and a customized software package to register data for analysis. Two variables, damping ratio and natural frequency, were assessed by pendulum tests. The damping ratio, a measure of how quickly an oscillation diminishes, was assessed in the rotational component of the leg drop. A smaller number here indicates lower resistance of motion and less spasticity. The other pendulum test variable measured was natural frequency, which is the frequency at which a pendulum oscillates as it comes to rest. It was measured in the rotational component of the leg drop. A lower natural frequency indicates more spasticity. The Subject s Global Impression of Change and the Clinician s Global Impression of Change were scored on a 7-point scale from markedly worse to markedly improved. The Subject s Global Impression of Change was assessed using a questionnaire asking subjects to rate their impression of the effects of the study medication using 7 scores from terrible to delighted. These 2 outcome measures had no baseline value. They were conducted after each 4-week period of treatment and placebo in order to assess the overall satisfaction of the subject and clinician, respectively. The pharmacy, using a computerized randomization system, randomly assigned participants to receive either nabilone 0.5mg or a placebo orally once a day for the first 2 weeks of the study. Subjects were assessed for the safety and efficacy of their prescription based on the primary outcome measures at the end of this initial 2-week period. This was done using 30-minute phone interviews, during which the subjects were questioned about any side effects. Pending tolerance of any side effects associated with the drug/placebo, study subjects were given the option to increase their treatment dosage to 0.5mg twice a day for the remaining 2 weeks of that 4-week period. They could return to the 0.5-mg orally once-a-day dose at any time. At the end of week 4 of treatment, the subjects were examined in the clinic. They reported whether they had increased the dosage to 2 tablets per day. They were examined and their vital signs were charted, any possible side effects were reviewed, and assessments of tolerance and outcome measures were conducted. All subjects had a 2-week washout period. Because nabilone has a half-life of 2 hours, 2 weeks was a reasonable washout period to ensure the medication had been excreted, as well as proving convenient for scheduling clinic visits. After the washout period, subjects were crossed over to the opposite arm in which the same patterns of outcome assessment and dose titration were maintained. The study concluded at 10 weeks. At the end of 10 weeks, our subjects were interviewed about any side effects, vital signs were examined, adverse reactions or events were monitored, and outcome measures were documented. Throughout the study, the subjects were able to drop out anytime with instructions to inform us of their decision and about any adverse effects. Overall, each subject had 3 clinical visits after enrollment in the study. All 3 visits, including examination of outcome measures and Ashworth Scales, were conducted by the same subject. Statistical Analysis All analyses were performed with the SAS version 9.1. a The comparison between the 2 arms of the study, the group that initially received nabilone with the one that received the placebo, was done with chi-square analysis. A 2-sample t test was used to compare baseline weights between the 2 groups. The Wilcoxon rank-sum test was used to calculate whether the mean difference of all the outcome measures between treatment and placebo periods, with the exception of the pendulum results, is significant. For the pendulum test results, as parametric variables, a Student t test was used for the same purpose. The correlation between body weight and response to treatment was assessed by Spearman rank correlation coefficient. Differences between placebo and treatments were adjusted by baseline measures (where available and a 1-sample sign rank tested that the mean difference equated to 0). Any difference was analyzed as a change from the baseline, except for the Clinician s and Subject s Global Impression of change. The primary outcome measure was the Ashworth Test, with a P value less than.05 considered statistically significant. The present statistical analysis of secondary measures encompasses a number of formal comparisons, and thus a group-wise type I error could occur. One common solution to minimize the occurrence of a group-wise type I error is to apply the Bonferroni correction; divide the alpha probability level by the number of multiple comparisons. 23,24 In the present study, 6 secondary measures were examined, and thus the corrected level will be set at.008 (.05 divided by 6). RESULTS We recruited 12 volunteers with SCI and spasticity. One subject dropped out of the study after receiving a diagnosis of urinary stricture. He was a paraplegic patient who received 4 doses of placebo before leaving the study. Eleven subjects completed the study. All were men with an average age of years. Six had tetraplegia and 5 paraplegia. Five of the 11 subjects were randomly assigned to the treatment arm first (group 1), and 6 to the placebo arm first (group 2). Using chi-square analysis, we compared baseline data between the 2 groups. Group 1 had a higher baseline sum of Ashworth than group 2 (35.6 vs 25.1; P.005). This occurred by coincidence, as the subjects were randomly assigned by the pharmacist in order to keep both subject and clinician blind of the randomization. Other baseline data including weight did not differ significantly between the groups. Spasticity improved significantly in the most spastic muscle group while subjects were receiving active treatment as compared with placebo, as measured by the primary outcome measure, the Ashworth Scale (table 1). The statistical test was the Wilcoxon rank-sum test (P.003). After the Bonferroni correction, there was a significant difference in one of the secondary outcome measures, the total Ashworth Scale, as a result of active treatment with nabilone (P.001) (see table 1), but the trend toward improvement in the Spasm Frequency Scale and VAS did not reach statistical significance after considering the Bonferroni correction as a result of active treatment when compared with the baseline (see table 1). The pendulum tests were done both in sitting and supine positions (see table 1). As mentioned earlier, a smaller damping ratio indicates lower resistance of motion and less spasticity. With the use of the Student t test, there was no significant decrease from baseline with active treatment (see table 1). With

4 706 NABILONE AND SPASTICITY IN PERSONS WITH SPINAL CORD INJURY, Pooyania Table 1: Comparison of Outcome Measures From Baseline Level for the Periods on Placebo and Nabilone, Using Bonferroni Correction and P<.008 Outcome Measure n Mean at the Baseline Mean After Placebo Period Mean After Treatment Period Mean Difference Between Treatment and Placebo Period in Each Subject, Adjusted to the Baseline Ashworth in most involved muscle group (significant) Ashworth in 8 muscle groups (significant) VAS (trend toward significant) Spasm Frequency Scale Subject Global Impression 11 NA *.312 Clinician Global Impression 11 NA *.789 Rotational damping ratio, sitting, pendulum variable Rotational natural frequency, sitting, pendulum variable (trend toward significant) *These measures have no baseline; thus, they are a comparison of difference between placebo and active arms. Values are mean difference SD. P the rotational natural frequency in the sitting position, the subjects had a trend toward significant improvement with active treatment (P.018) (see table 1). In the lying position, some of the subjects had such a severe tone that the natural frequency and damping ratio were not measurable, resulting in missing data for these subjects, and accordingly, these data could not be accurately analyzed and reported. Clinician s and Subject s Global Impression scores did not show a significant difference after the course of active treatment, compared with the course of placebo (see table 1). There was no significant correlation between each subject s weight and the response to treatment. In general, nabilone with a dosage ranging from 0.5mg at night to 0.5mg twice a day was tolerated by all the subjects. No serious adverse events emerged during the trial to concern any of the subjects and cause them to drop out of the study. All the subjects receiving placebo increased the dosage to 1 tablet twice a day, and no side effects were reported. In the active treatment arm, 7 of the 11 subjects increased the dosage to 0.5mg twice a day. Two of the subjects who increased the dosage experienced side effects and dropped back to the initial dosage of 0.5mg at night. One subject reported headaches, and another reported mild vertigo, ataxia, and lack of motivation. Four subjects stayed on the dosage of 0.5mg at night, largely because they were happy with the improvement and did not want to risk any side effects with the higher dosage. While receiving nabilone, 8 of 11 subjects reported side effects. Three subjects (27.2%) experienced drowsiness. Dry mouth and asthenia were each reported by 2 subjects (18.1%). Two subjects reported mild vertigo (18.1%), one at the dosage of 0.5mg at night and the other at the dosage of 0.5mg twice a day. Mild ataxia, headache, and lack of motivation were reported in 1 subject (9.09%) while taking nabilone at the dosage of 0.5mg twice a day. None of the subjects reported any remarkable interference with activities of daily living as a result of reported side effects. No clinically relevant changes were observed in the physical examination, including pulse rate and blood pressure. Larger dosages may result in further benefits with a longer treatment period, as some of the subjects continued to have severe spasticity even while taking nabilone. This randomized, double-blind, placebo-controlled, crossover pilot study of an orally administered cannabinoid, nabilone, in subjects with spasticity resulting from SCI has found nabilone, 0.5 to 1.0mg/d, to be a safe and well-tolerated dosage. DISCUSSION In this study, we found a significant reduction in spasticity as measured by the Ashworth Scales in the most spastic muscle group of persons with SCI after treatment with nabilone compared with a placebo. We were able to identify a relatively tolerable dosage of nabilone. The same effect of another oral THC (dronabilone) on reducing spasticity was previously observed in 2 other studies. 25,26 Study Limitations This study has some limitations. The design of this project as a pilot study and the low number of subjects made this study a preliminary one to explore a tolerable dosage and reasonable outcome measures. It was recognized that the numbers of subjects would be too small to reach firm conclusions, with a high likelihood of type II errors. A further limitation arose because the subjects were allowed to continue their antispasticity medication and did not go through a washout period from the previous medication. This could introduce validity issues because of possible medication interactions. The study was also hampered by a lack of dose-effect assessment, since our subjects were given the option of either continuing or increasing their dosage. The improvement we saw in the Ashworth of almost 20% (0.9/5) could be considered clinically significant. The rating scales, however, differ between the Ashworth and Spasm Frequency Scales. The Spasm Frequency Scale, which was used as one of the secondary outcome measures (see appendix 2), has only 5 grades, with all 10 or more spasms per day considered as grade 5, and is therefore unlikely to be sensitive to changes in the most severe category (eg, despite a decrease in spasms from 50 to 15 in 1 day, subjects could not describe themselves as improving 1 grade). In the Subject s and Clinician s Global Impression of Change, the small number of subjects and possible type II error could explain why no statistically significant difference was reported while patients were taking medication. The pendulum variables that were addressed were rotational natural frequency and damping ratio in both sitting and lying positions. It is not clear which measurement is more clinically significant and why rotational natural frequency showed a trend towards improvement with nabilone while the damping ratio did not. It may be due to the small sample size and short period of study. Some subjects had relatively normal baseline pendu-

5 NABILONE AND SPASTICITY IN PERSONS WITH SPINAL CORD INJURY, Pooyania 707 lum testing; their spasticity was more severe in the muscle groups other than the knee extensors, and the pendulum test only examines knee extensor spasticity. Another factor was that some of the subjects had such severe spasticity that the variables were not detectable by the software, resulting in some missing data in the pendulum test (18 lower extremities instead of 22) and causing difficulty in the statistical analysis. This problem was more dominant in the supine position, which made it difficult to accurately analyze those results, and we thus only reported the results from the sitting position. The secondary outcome measures were not the prime objective of the study. Thus, for this small exploratory study, we have presented the secondary measures individually, but recognize that solid conclusions cannot be made about any of these secondary measures. CONCLUSIONS This randomized, double-blind, placebo-controlled study suggests that an orally administered cannabinoid, nabilone, may be beneficial to improve spasticity. The numbers in this study are small, and we recommend a larger trial with a more prolonged treatment period and an option to continue to slowly increase dosages. This will allow further assessment of the efficacy and safety of nabilone in spasticity in patients with SCI. Grade APPENDIX 1: ASHWORTH SCALE Description 0 No increase in tone 1 Slight increase in tone giving a catch when the limb is moved in flexion or extension 2 More marked increase in tone, but limb easily flexed 3 Considerable increase in tone, passive movement difficult 4 Limb rigid in flexion or extension APPENDIX 2: PENN SPASM FREQUENCY SCALE Grade Description 1 No spasms 2 1 spasm or fewer per day 3 Between 1 and 5 spasms per day 4 Between 5 and 9 spasms per day 5 10 or more spasms per day References 1. DeLisa JA, Gans BM, Bockenek WL. Rehabilitation medicine: principles and practice. 4th ed. Philadelphia: Lippincott-Raven; Pertwee RG. Cannabinoids and multiple sclerosis. Pharmacol Ther 2002;95: Young RR. Spasticity a review. Neurology 1994;44(11 Suppl 9)S Maynard FM, Karunas RS, Waring WP. Epidemiology of spasticity following traumatic spinal cord injury. Arch Phys Med Rehabil 1990;71: Walter JS, Sacks J, Othman R, et al. A database of self-reported secondary medical problems among VA spinal cord injury patients: its role in clinical care and management. J Rehabil Res Dev 2002;39: Johnson RL, Gerhart KA, McCray J, Menconi JC, Whiteneck GG. Secondary conditions following spinal cord injury in a population based sample. Spinal Cord 1998;36: Pertwee RG. Cannabis and cannabinoids: pharmacology and rationale for clinical use. Forsch Komplementarmed 1999;6(Suppl 3): Dunn M, Davis R. The perceived effects of marijuana on spinal cord injured males. Paraplegia 1974;12: Malec J, Harvey RF, Cayner JJ. Cannabis effect on spasticity in spinal cord injury. Arch Phys Med Rehabil 1982;63: Cansroe P, Musty R, Rein J, Tillery W, Pertwee R. The perceived effects of smoked cannabis on patients with multiple sclerosis. Eur Neurol 1997;38: Maurer M, Henn V, Dittrich A, Hofmann A. Delta-9-tetrahydrocannabinol shows antispastic and analgesics effects in a single case double-blind trial. Eur Arch Psychiatry Clin Neurosci 1990; 240: Brenneisen R, Egli A, Elsohly MA, Henn V, Spiess Y. The effect of orally and rectally administrated delta 9-tetrahydrocannabinole on spasticity, a pilot study with 2 patients. Int J Clin Pharmacol Ther 1996;34: Zajicek J, Fox P, Sanders H, Wright D, et al. UK MS research group. Cannabinoids for treatment of spasticity and other symptoms related to multiple sclerosis (CAMS study): multicenter randomized placebo-controlled trial. Lancet 2003;362: Ashworth B. Preliminary trial of carisolrodol in multiple sclerosis. Practitioner 1964;192: Bohannon RW, Smith MB. Inter-rater reliability of a modified Ashworth Scale of muscle spasticity. Phys Ther 1987;67: Johnson GR. Outcome measures of spasticity. Eur J Neurol 2002;(9 Suppl 1): Penn RD. Intrathecal baclofen for severe spasticity. Ann N Y Acad Sci 1988;531: Priebe MM, Sherwood AM, Thornby JI, Kharas NF, Markowski J. Clinical assessment of spasticity in spinal cord injury: a multidimensional problem. Arch Phys Med Rehabil 1996;77: Stewart JE, Barbeau H, Gauthier S. Modulation of locomotor patterns and spasticity with clonidine in spinal cord injured patients. Can J Neurol Sci 1991;18: Skold C. Spasticity in spinal cord injury: self- and clinically rated intrinsic fluctuations and intervention-induced changes. Arch Phys Med Rehabil 2000;81: Wartenberg R. Pendulousness of the legs as a diagnostic test. Neurology 1951;1: Fowler EG, Nwigwe AI, Ho TW. Sensitivity of the pendulum test for assessing spasticity in persons with cerebral palsy. Dev Med Child Neurol 2000;42: Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B 1995;57: Strassburger K, Bretz F. Compatible simultaneous lower confidence bounds for the Holm procedure and other Bonferroni-based closed tests. Stat Med 2008;27: Hagenbach U, Luz S, Ghafoor N, et al. The treatment of spasticity with 9 -tetrahydrocannabinol in persons with spinal cord injury. Spinal Cord 2007;45: Kogel RW, Johnson PB, Chintam R, Robinson CJ, Nemchausky BA. Treatment of spasticity in spinal cord injury with dronabinol, a tetrahydrocannabinol derivative. Am J Ther 1995;2: Supplier a. SAS Institute Inc, 100 SAS Campus Dr, Cary, NC