Other Anti-Spasmodics

Other potential anti-spasmodics which have been tested in the SCI population include Traditional Chinese Medicine (TCM), Riluzole, L-threonine, and orphenidrine citrate.

TCM is believed to have medical benefits and scientific documentation of these benefits, specifically with respect to spasticity, pain and sleep, have begun.

Riluzole is an antiglutamatergic agent approved to slow the progression of Amyotrophic Lateral Sclerosis. Due to its spinal locus of activity where it acts more strongly on polysynaptic reflex pathways but not on direct motoneuron excitabilty, riluzole was investigated and shown to decrease spastic flexion reflexes in response to cutaneous stimuli in spinal cord injured rats (Kitzman 2009).

L-threonine is an α-amino acid with a putative mechanism of anti-spastic action through increasing spinal glycine levels (Paisley et al. 2002).

Orphenadrine citrate is a non-competitive NMDA-type (N-Methyl-D-aspartate) glutamate antagonist which acts centrally as an anticholinergic and non-opioid analgesic (Clark 2002).

Although diazepam and dantrolene continue to be used in SCI spasticity, no new evidence is available to support continued recommendation for use in the presence of current first line treatments

Other drugs that have been assessed against spasticity treatment in SCI include the opiod antagonist, Naloxone and the anti-epileptic, levetiracetam.

Author Year

Research Design

Total Sample Size

Methods Outcome
Traditional Chinese Medicine
Liu et al. 2014





Population: Age range: 18-65yr; Level of Injury: C5; Level of severity: AIS A-D.

Intervention: Individuals were randomized into perfume-water administered placebo (n=80) and traditional Chinese medicine (TCM; n=80) treated groups (TCM formula was composed of six kinds of herbs). All individuals received 1 yr treatment. Bath barrels (500L) were filled with either perfume water or TCM in which individuals bathed for 30 min every 3 days at a constant temperature for 1-yr.

Outcome Measures: Ashworth Scale (AS), Spasm Frequency Scale (SFS), Visual Analog Scale (VAS), Subject Global Impression of Change (SGIC), Clinician’s Global Impression of Change (CGIC), Pittsburgh Sleep Quality Index (PSQI), Regular Sleep-Wake Rhythm, Skin Allergies. Outcomes were assessed post-intervention.

1.     Post-intervention AS in the most involved muscle group of the body (as chosen by the subject and clinician) significantly improved (p<0.05) in the TCM group (7.10±0.79) when compared to the placebo group (7.56±0.89). 2.     Post-intervention sum of the AS in both upper and lower extremities (including the elbow flexors and extensors, the wrist and finger flexors, the hip adductors, the knee flexors and extensors, and the foot plantar flexors) in the TCM group (25.38±3.11) was below (p>0.05) that of the placebo group (28.34±3.35).

3.     Post-intervention SFS was significantly improved (p<0.05) in the TCM group (2.86±0.12) when compared to the placebo group (3.55±0.23).

4.     Post-intervention VAS significantly improved in the TCM group when compared to the placebo group (p<0.01).

5.     Post-intervention VAS significantly improved (p<0.01) in the TCM group (40.34±2.89) when compared to the placebo group (55.89±3.67).

6.     Post-intervention SGIC significantly improved (p<0.01) in the TCM group (2.99±0.17) when compared to the placebo group (3.75±0.35).

7.     Post-intervention CGIC significantly improved (p<0.01) in the TCM group (3.21±0.16) compared to the placebo group (3.88±0.24).

8.     No skin allergies were found.

9.     Post-intervention PSQI significantly improved (p<0.01) in the TCM group (5.8±0.97) when compared to the placebo group (7.7±1.1).

10.   There was a significant difference (p<0.01) between the 18.75% (15/80) of individuals in the placebo group and the 43.75% (35/80) of individuals in the TCM group who reported better sleep quality and a more regular sleep-wake rhythm.

Effect Sizes: Forest plot of standardized mean differences (SMD±95%C.I.) as calculated from pre- and post-intervention data.

Casale et al. 1995


Prospective Controlled Study


Population: Mean age:31.2 yr; Gender: males=9, females=2; Injury etiology: SCI=11; Level of injury: paraplegia=11; Time since injury range: 9-11 yr.

Intervention: 60 mg of intravenous orphenadrine citrate.

Outcome Measures: Threshold of flexion reflex (mAmp); Ashworth Scale (AS); at baseline, initial injection, 10, 20, 30 and 60 min post injection.


1.     A significant difference was observed when comparing the use of orphenadrine and placebo (p<0.0001).

2.     Orphenadrine increased the flexion reflex threshold within 30 min in ten individuals and within 60min in one patient.

3.     One individual who had severe spastic hypertonia did not see an improvement in reflex threshold.

4.     There was a significant decrease in the AS for orphenadrine treatment (p<0.0001), as compared to no effect for the placebo.

Theiss et al. 2011


RCT Crossover



Population: Mean age: 43.7 yr; Gender: males=6, females=1; Level of injury: cervical=5, thoracic=2; Level of severity: AIS C=2, AIS D=4, AIS C-D=1; Mean time since injury: 14.4 yr.

Intervention: Riluzole (50 mg) and placebo (50 mg) were administered in a randomized order, separated by washout period of 7 days.

Outcome Measures: Stimulus threshold for flexion-withdrawal reflex, Peak amplitude of torque during reflex response, H-reflex and M-wave responses, Hmax/Mmax ratio, Torque of maximum voluntary contractions (MVC), Average sustained torque, Modified Ashworth Scale (MAS), Spinal Cord Assessment Tool for Spasticity (SCATS).

1.     The threshold stimulus intensity for the reflex response significantly increased after riluzole (p=0.03) but not after placebo.

2.     Compared with placebo, a significant decrease in the peak amplitude of torque following riluzole was found (p=0.0003).

3.     The torque of MVC during dorsiflexion decreased significantly after placebo (p=0.02) but not after riluzole.

4.     The percent change in average sustained torque increased significantly with riluzole administration compared to placebo (p=0.04).

5.     The Hmax/Mmax ratio, MAS, and the SCATS scores were not significantly different between the placebo and the riluzole conditions.

Effect Sizes: Forest plot of standardized mean differences (SMD±95%C.I.) as calculated from pre- and post-intervention data.

Lee & Patterson 1993

USA & Ireland




Population: Injury etiology: MS, traumatic/non-traumatic SCI.

Intervention: 6 g/day-500 mg L-Threonine or Placebo capsules taken 3x/day on empty stomach.

Outcome Measures: Ashworth Scale (AS) (bilat). Hip adductors-flexors-extensors and knee flexors-extensors)–6 highest summed for spasticity score, which was used throughout the study. Spasm frequency and severity score (spasm score was derived by multiplying two variables together over 2 wk period using specially designed chart), BI, Kurtzke Disability Status Scale, Individual & Caregiver subjective responses, Aes, and glycine/threonine plasma concentrations. Measurements pre-and post intervention. All measures conducted by a single investigator.

Modest but definite antispastic effect in favour of L-threonine versus placebo:

1.     Mean spasm score reduced for both interventions–weak correlation between spasm score and spasticity reduction.

2.     No change in BI or Kurtzke with either intervention.

3.     Dramatic rise in plasma threonine during active intervention but no change in plasma glycine.

4.     Weak correlation between plasma threonine and spasticity reduction.

5.     Patient-carer’s subjective report–6/2 threonine/placebo responders.

6.     Aes included minor side effects during intervention (n=2) and Placebo (n=1).

7.     Four dropouts–2 for medical and 2 for non-medical reasons.

Effect Sizes: Forest plot of standardized mean differences (SMD±95%C.I.) as calculated from pre- and post-intervention data.


Brackett et al. 2007

USA & Canada



Population: SCI (n=3): Age range: 30-42 yr; Level of Injury: T6, C4 & C4; Time since injury range: 5-26 yr; Able Bodied Controls (n=3): Age range: 22-37 yr.

Intervention: 30hr protocol implemented: physiologic saline was injection on day 1 as a control for the infusion of naxolone on day 2.

Outcome Measures: N/R.

1.    No spasticity was experienced in able bodied subjects during the study.

2.    SCI subjects had a large increase in the duration and occurrence of spasticity.

3.    This increase was only present when SCI subjects were injected with naxolone. Spastic events occurred within 30 min of injection and could easily be triggered, even when there was lack of an obvious stimulus.

Finnerup et al. 2009


RCT Crossover




Population: Mean age: 51.0 yr; Gender: males=21, females=3; Level of injury: cervical=10, thoracic=12, lumbar=2; Level of severity: AIS A=10, AIS C=2, AIS D=12.

Intervention: Individuals were randomized to receive either 5-wk treatment with levetiracetam (500 mg x2 in the first week to 1000 mg x2 in the second week and 1500 mg x2 in wk 3–5) or placebo. After a 1 wk washout period subjects were crossed to the other treatment arm.

Outcome Measures: Pain score, Penn Spasm Frequency Scale (PSFS), Modified Ashworth Scale (MAS).

1.      There was no difference in the median pain intensity during levetiracetam and placebo treatment (p=0.46).

2.      Levetiracetam did not significantly change PSFS or MAS scores (p>0.05).


Traditional Chinese Medicine

Liu et al. (2014) reported significant improvements in pain, spasticity and sleep after a randomized, double blind, study comparing an herbal bath of six TCMs to a placebo of perfumed bath water. No treatment related side effects or allergies were observed in 160 participants during the year long study. Outcome assessments included the Ashworth Scale, the Visual Analog Scale for pain, subject and clinician global impression scales and the Pittsburg Sleep Quality Index. Liu et al. (2014) concluded that the treatment was effective and economical for long-term use. However, they acknowledge that even though the herbs used in this study are believed to have medical benefits after more than 1000 years of use, a detailed chemical analysis is needed to understand the mechanism of action.


Casale et al. (1995) were able to demonstrate a significant reduction of spasticity (p<0.0001) as measured by Ashworth in favour of intravenous administration orphenadrine citrate versus placebo. This anti-spasmodic effect was demonstrated by an increased flexion reflex threshold as early as 30 minutes post administration. The authors suggest that this drug, given its immediate action, could be used as a preparatory solution for physical therapy sessions in spastic patients. Given its known side effect profile, this treatment may be appropriate for short term application.


Theiss et al. (2011) demonstrated significantly reduced spasticity without a generalized reduction in strength, uniformly among seven patients with chronic, motor-complete SCI. Although the study was randomized, double-blinded and controlled, measurements were made only on a single dose and outcome assessments were not clinically feasible (e.g. stimulus thresholds and torque measurements). At least one confirmatory RCT would be beneficial before consideration that continued use of riluzole is beneficial for spasticity treatment in SCI. In the interim, riluzole may be appropriate to trial on individuals who do not respond to or tolerate other anti-spasmodics.


A randomized, controlled study of L-threonine (Lee & Patterson 1993) showed minimal effects on spasticity. Interestingly there was no correlation between the reduction in spasm score and tone reduction suggesting that these components of the disordered upper motor neurone syndrome may have different pathophysiological causes and therefore may require different pharmacological treatments. This is especially important to note since many patients do not experience satisfactory control of spasticity using first line treatments for reasons other than intolerance.


An inadvertent side effect discovered during an investigation of neuroendocrine function in SCI using naloxone was the profound increase in spasticity after naloxone treatment in all three SCI participants compared to no such activity in the three non-disabled study volunteers (Brackett et al. 2007). This interesting finding agrees with a previous suggestion that a relationship between opioid receptors and spasticity in patients exists where morphine is found to be effective when intrathecal baclofen tolerance is an issue.


Levetiracetam, an anti-epileptic (i.e., Keppra), was studied by Finnerup et al. (2009) for its effect on neuropathic pain, evoked pain or spasms and spasticity. Although well tolerated, no effects were recorded for spasm severity or pain.

Although diazepam and dantrolene continue to be used in the treatment of SCI related spasticity, no new evidence has been found to support their continued use given that Level 1 evidence supports treatments now routinely used for spasticity in SCI (Nance et al. 1989). See Table 20 for data regarding a study focusing on clonidine in SCI. However, an earlier crossover study (Corbett et al. 1972) showed that effects of diazepam and placebo were not different from pre-treatment and that Valium was more effective than amytal and placebo in reducing spasticity (p<0.02-0.05).


There is level 1b evidence (from one RCT: Liu et al. 2014) that traditional Chinese medicine is a safe, effective and economical long-term treatment for spasticity in SCI.

There is level 2 evidence (from one prospective controlled trial: Casale et al. 1995) for short-term use of intravenous orphenadrine citrate for treatment of spasticity secondary to SCI.

There is level 1b evidence (from one RCT: Theiss et al. 2011) that riluzole is effective in treating spasticity post SCI.

There is level 1b evidence (from one RCT: Lee & Patternson 1993) that L-threonine produces minimal anti-spasmodic effects post SCI.

There is level 4 evidence (from one pre-post study: Brackett et al. 2007) that naloxone causes a profound increase in spasticity in individuals with SCI.

There is level 1b evidence (from one RCT: Finnerup et al. 2009) that Levitiracetam is not effective for treating spasm severity in SCI.

Continued use of diazepam and dantrolene for reducing SCI-related spasticity is not supported with up to date evidence and would benefit from new controlled comparison studies. Whether effective or not, replicating results in well-designed trials is warranted before alternative recommendations for new or older treatments will be accepted into current practice.