Combined Gait Training and Pharmacological Interventions

Drugs such as clonidine (a noradrenergic agonist), cyproheptadine (a serotonergic antagonist), baclofen (GABA agonist), GM-1 ganglioside, L-Dopa and 4-aminopyridine have been used in association with attempts to improve ambulation in individuals with SCI. The results from animal studies indicate that some of these drugs may act on the receptors in the spinal cord which facilitate interaction with a locomotor central pattern generator (spinal circuits which produce coordinated locomotor movement) (Chau et al. 1998; Rossignol et al. 1996; Barbeau and Rossignol 1990). Although not conclusive, there is some evidence that similar “central pattern generator” circuits exist in humans (Bussel et al. 1996; Illis 1995; Calancie et al. 1994; Bussel et al. 1989; Bussel et al. 1988) and provide the rationale for clinical use of these drugs.

Author Year; Country
Research Design
Sample Size
Methods Outcomes

DeForge et al. 2004; Canada
N SCI=11

Population: 11 participants with SCI; Age 22-70 yrs; all participants with diagnosis of AIS D; C3-T12 lesion level; 1-20 post-injury.
Treatment: Double-blind, placebo-controlled, crossover design; 4-Aminopyridine (4-AP): up-titration to 10 mg 4x/day stable dosing of 4-AP (n=15) versus Placebo (n=14), 2 weeks each condition
Outcome measures: Isometric muscle force, gait analysis.
  1. Some positive effects for both placebo and 4-AP treatment when compared to baseline, but no changes between groups were significant.
Effect Sizes: Forest plot of standardized mean differences (SMD ± 95%C.I.) as calculated from pre- and post-intervention data

Duffell et al. 2015; USA
Level 1
N= 48
Population: 26 individuals in locomotor treadmill training group (LTT)- 19 males and 7 females; mean age= 46.6 ± 12.6y; years post injury= 9.3 ± 8.9y; 22 individuals in combined LTT and tizanidine group- 15 males and 7 females; mean age= 46.5 ± 11.9y; years post injury= 10.2 ± 10.4y; Level of injury 30 C and 15 T
Treatment: Participants were randomly assigned into one of two intervention groups; LTT alone (LTT; n = 26) or combined LTT and Tizanidine (TizLTT; n = 22). Participants assigned to the TizLTT group, were initially provided with Tizanidine alone for a period of 4 weeks, and results for that period have been presented elsewhere, together with the LTT group clinical outcomes. Outcomes were measured at 0, 1, 2 and 4 weeks from the start of LTT for both groups
Outcome Measures: Timed up and go (TUG), 10MWT, 6MWT, MAS, Maximum voluntary isometric contractions (MVIC), active range of motion (AROM), Peak isokinetic velocity (Vp)
  1. Both LTT and TizLTT improved in walking speed and dorsiflexion maximum strength, with no significant differences between them, using group-averaging analysis at 0, 1, 2 and 4 weeks from the start of LTT for both groups.
  2. A higher proportion of participants in the TizLTT group achieved the MID for walking speed (40%) compared with LTT alone (13%)
  3. Those that achieved the MID for walking speed were significantly higher functioning at baseline than those that did not in the TizLTT group, and the change in walking speed was associated with the change in dorsiflexion peak velocity.

van der Bruggen 2001; The Netherlands
Level 1

Population: Age 25-70 yrs; all participants had an incomplete SCI; C2-L3 lesion level; 3-56 yrs post-injury.
Treatment: Double-blind, placebo-controlled, crossover design: up-titration to maximum of 15-45 mg, immediate-release 4-Aminopyridine capsules or Placebo, 4 weeks each condition.  2 week washout between conditions.
Outcome measure: comfortable and maximum walking speed.
  1. No statistically significant functional benefits were found.

Grijalva et al. 2010; Mexico
Level 1

Population: 10 males, 4 females; mean age 29; average YPI 6.2; 8 cervical SCI, 6 thoracic
Phase 1: protocol found in Grijalva et al., 2003
Phase 2 (3 months): Administration of 10mg/day of 4-AP the first week, 20 mg/day the second week, 30 mg/day the third week, and 10mg increase per day every 2-3 months if patients were not experiencing any adverse reactions.
Outcome Measures: AIS motor and sensory scale, SCIM, SEPs
  1. There were no significant differences between placebo and 4-AP in phase one.
  2. There were no significant differences comparing pretreatment and highest 4-AP doses evaluations in phase 2.
  3. Two patients experienced adverse reactions severe enough to cause drop out; one of them had epileptic seizures and the other persistent anxiety and insomnia.
Effect Sizes: Forest plot of standardized mean differences (SMD ± 95%C.I.) as calculated from pre- and post-intervention data

Maric et al. 2008; Switzerland
RCT with crossover
Level 1
N = 12

Population: 12 participants with incomplete SCI, 4 female, 8 male, ages 31-75.
Treatment: Double-blind, placebo-controlled crossover study design: Participants were randomly divided into two groups. Group 1 first received 6 weeks of treatment of 200mg L-Dopa and 50mg dopa decarboxylase inhibitor with physiotherapy for 30-45min 1-4hrs after L-dopa intake. Group 1 then received then 6 weeks of treatment with placebo. Group 2 first received 6 weeks of placebo and then 6 weeks of treatment with 200mg L-Dopa and 50mg dopa decarboxylase inhibitor with physiotherapy.
Outcome Measures: Asia motor score (AMS); WISCI II; SCIM II.
  1. There was no effect of L-Dopa and physiotherapy on the outcomes.
  2. The treatment group had greater improvement than control in AMS (+7.8 in treatment, vs. +6.6 in control) and SCIM (+16.6 vs. +11.7), but the difference was not significant (p=0.49 for AMS, p=0.31 for SCIM).
  3. The control showed greater average improvement than treatment group in WISCI II score (+2.9 in treatment, vs. +3.4 in control) but the difference was not significant.

Walker & Harris 1993; USA
Level 1

Population: Age 21-44 yrs; all participants incomplete SCI; C5-L1 lesion level; 1-13 yrs post-injury
Treatment: Double-blind, placebo-controlled crossover study design: Intravenous GM-1 ganglioside (Sygen ®) or placebo + 2 hr PT (gait training) 6x/wk for 2 months, followed by switch of drug administration (total 4 months). All participants given 6 months of PT before trial.
Outcome measures: Motor score, walking distance, and velocity.
  1. GM-1 + PT resulted in increase in motor scores, walking distance, and walking velocity.
    *Note: Klose and Calancie (1994) wrote a critique of this article, re-calculated data, and came up with different results.

Stewart et al. 1991; Canada
Crossover Design
Level 1
N = 9

Population: 6 participants with paraplegia, 3 participants were paretic; age 19-57 yrs; AIS A-D; C7-T10 lesion level; 1-10 yrs post-injury.
Treatment: Double-blind, placebo-controlled, crossover design: Two periods of 4 weeks of medication (Clonidine (up to 0.1-0.5 mg daily) or Placebo, randomly assigned) separated by a 2 week washout period.
Outcome measures: Kinematic measures during body weight support gait, spasticity, adverse effects.
  1. No group differences were presented or analyzed in this study.
  2. Clonidine did not elicit locomotor activity in the paraplegic patients, but there were reductions in stretch reactions and clonus during assisted locomotion.
  3. Side effects were experienced by 8 of 9 patients during dosage increases, although for most patients these symptoms were transient. Dryness of the eyes and mouth were the most common, though other participants experienced lethargy, mild hypotension and constipation.

Duffel et al. 2015; USA
Level 2
N= 83

Population: 29 individuals in control group- 9 males and 10 females; mean age= 47.8 ± 13.1y; years post injury= 8.1 ± 8.1y; 27  individuals in lokomat group- 19 males and 8 females; mean age= 46.6 ± 12.6y; years post injury= 9.3 ± 8.9y; 27 individuals in tizanidine group- 19 males and 8 females; mean age= 47.4 ± 11.6y years post injury= 10.9 ± 10.8y; motor incomplete SCI, AIS C or D
Treatment: Participants were assigned to 3 groups: no intervention, Lokomat, or tizanidine. For the Lok group, locomotor training was provided using a robot-assisted locomotor training device. This device provides bodyweight- supported gait assistance. For the Tiz group, 0.03 mg/kg of tizanidine was administered 4 times a day for 4 weeks.
Outcome Measures: Timed up and go (TUG), 10MWT, 6MWT, MAS
  1. There was no difference between interventions, though overall walking speed and endurance did improve.
  2. Only a small number of participants achieved the MID. Both MID and GMM-RCR analyses revealed that tizanidine improved endurance in high-functioning participants.
  3. GMMRCR classification also showed that speed and mobility improved after locomotor training.

Remy-Neris et al. 1999; France
Prospective study
Level 2
N = 11

Population: 9 males, 2 females; ages 25-66 yrs, mean age 40 years.
Treatment: Each patient received 3 doses of 15-90 µg clonidine and a placebo by lumbar puncture. Each injection separated by a minimum of 3 days.
Outcome measures: Spatiotemporal gait data, Ashworth scores, soleus H-reflex, and polysynaptic flexion reflexes recorded before and every hour for 4-6 hours after injection.
  1. No significance testing was conducted in this study.
  2. Three participants improved their gait velocity after clonidine administration; one (S6) increased his stride amplitude; the two others decreased their cycle durations.
  3. 3 of 8 ambulatory participants had greater maximum overground walking speed with clonidine. These participants were more severely impaired and had shorter times post-injury.

Wainberg et al. 1990; Canada
Prospective controlled trial
Level 2
N = 8

Population: 8 participants with spinal spastic paresis; 7 male, 1 female, 2 wheelchair-bound, 6 can walk with aids; age 23-56 years; C4-T11 lesion level; 1-15 yrs post-injury
Treatment: 6 of the 8 participants were included in double-blind, placebo controlled, crossover design: Two periods of 3 weeks of medication (2-8 mg 3x daily Cyproheptadine or Placebo, randomly assigned) separated by a 1-week washout period. Four participants continued in an open label, long term trial (>6 months)
Outcome measures: Temporal measures, EMG, joint angles, spasticity, comfortable walking speed. No statistical analysis.
  1. No significant differences between groups were reported in this study.
  2. Maximum comfortable walking speed increased in ambulatory participants, with a decrease in cycle duration and double support duration.
  3. Two patients that required body weight support during placebo could walk with full weight bearing during cyproheptadine therapy. Muscle coordination improved, and clonus was reduced.

Leech et al. 2014; USA
Randomized Cross Over Design
Level 2
N= 10

Population: 10 individuals; level of injury ranging from C2 to C7; months post injury= 12 – 301 months;
Treatment: Participants were involved in a double-blinded, randomized, cross-over design to assess the effects of acute pharmacological manipulation of 5HT transmission on various measures of locomotor performance. The agents used were overencapsulated, orally administered doses of a selective serotonin reuptake inhibitor (SSRI): 10 mg of escitalopramoxalate and a 5HT antagonist: 8 mg of cyproheptadin. Participants participated in 2 days of testing separated by at least 1week. Before initial locomotor testing, a licensed physical therapist assessed standardized measures of strength, spastic motor activity, and walking ability.
Outcome Measures: AIS Lower Extremity Motor Scores, Spinal Cord Assessment Tool for Spastic Reflexes, Modified Ashworth Scale, The Walking Index for Spinal Cord Injury II, EMG, O2 consumption
  1. Neither medication led to improvements in locomotion, with a significant decrease in peak overground gait speed observed after 5HT antagonists.
  2. Additionally, 5-HT medications had differential effects on EMG activity, with 5HT antagonists decreasing extensor activity and SSRIs increasing flexor activity

Norman et al. 1998; Canada
Level 4
N (enrolled) = 12
N (completed) =7

Population: 12 males recruited, 7 completed evaluations; age 19-35 yrs; participants had diagnosis of AIS C-D; C4-T12 lesion level; 1.1-5.3 yrs post-injury
Treatment: 3 different oral tablets in order of convenience: Clonidine (≤0.25 mg/day) or Cyproheptadine (≤24 mg/day) or Baclofen (≤80mg/day): each drug trial had incremental increase to maximum dose and stable dosing over 3 weeks followed by incremental decrease from maximum dose and washout over 2 weeks.
Outcome measures: Surface EMG and kinematic gait analysis during treadmill walking.
*Note: No statistical analysis was done.
  1. No significance testing was completed in this study.
  2. 7/12 participants had evaluations of all 3 drugs; adverse effects for 4/5 participants prevented completion of all conditions. The greatest effects in more severely disabled participants.
  3. Cyprohyeptadine resulted in decreased need for assistance, an increase in maximum treadmill speed and decreased clonus.
  4. Clonidine resulted in an increase in maximal treadmill speed and a generally more upright posture.
  5. Baclofen resulted in minor changes in walking. Maximal treadmill speed increases and other changes were often retained following washout of drugs.

Segal and Brunnemann 1998; USA
Level 4

Population: 9 males; age 28-60 yrs; participants had diagnosis of AIS C-D; C2-L4 lesion level; 4-28 yrs post-injury
Treatment: 4-AP (single 10mg immediate-release capsule). Comparison of means at baseline and at intervals over 24-hour follow-up.
Outcome measures: Ambulation parameters. 
  1. Significant improvements in gait velocity (increased by 36% from 24.1(16.5) m/min to 32.7 (22.9) m/min (p£0.04) and in stride length (increased from 0.9 (0.3) meters to 1.0 (0.3) meters) (p£0.02).
  2. There was also no increase in cadence (p=0.06) and no decrease in gait cycle duration (p=0.1).
  3. Gait changes began 6 hours after drug administered and persisted after the 24-hour follow-up.

Azouvi et al. 1996; France
Level 4
N = 18


Population: 18 patients with severe and disabling spinal spasticity, 12 of 18 participants with thoracic or low cervical lesion (Frankel A-D); age 21-59; C4-T11 lesion level; 0.5-27 years post-injury.
Treatment: Implanted intrathecal baclofen pump. 17 patients had an electronically driven programmable pump filled with 18 cc of 500 or 2000 ug/cc baclofen delivered by continuous infusion or by intermittent bolus. One patient had a manually operated pump delivering a bolus of 50 ug. Follow up assessment was 6-72 months after implantation.
Outcome Measures: Ashworth scale, spasm frequency scores, FIM. No statistical analysis on FIM walking score.
  1. Average motor FIM score for the 18 patients was 39.9 _+ 18.1 before treatment, 58.5 _+ 28.7 at 6 months (Wilcoxon Z = -3.62, p < .001)
  2. In 5 patients, walking ability improved (average initial FIM walking score: 3.6 ± .87, 6-month score: 5.8 ± .2), and 2 of these patients acquired the ability to climb stairs.
  3. Severe side effects were observed in 2 patients and one patient’s treatment interrupted after 9 months due to severe side effects.


The interactions of these pharmacological interventions are complex and appear to affect walking ability and spasticity to varying effects. The studies on clonidine (oral or intrathecal), cyproheptadine and baclofen demonstrate improvements in various aspects of gait (i.e. walking speed, posture, spasticity), but no improvements led to significant functional changes in walking. Norman et al. (1998) found the greatest improvements in more severely disabled subjects and in many cases, the effects were retained following washout of clonidine. Bradycardia and hypotension, common side-effects of oral clonidine can be ameliorated with intrathecal injection of clonidine (150-450µg) (Filos et al. 1994). The combined effect of different drugs has not been well explored. One very small study (not tabled due to its small sample size, Fung et al. 1990) showed that a combination of clonidine, cyproheptadine and treadmill training improved SCI locomotion in its 2 participants.

Conflicting evidence exists on the use of GM-1 ganglioside for neurologic recovery for walking in SCI. A large scale multicenter RCT (n=760) (Geisler et al. 2001) suggested that although GM-1 treatment may have accelerated initial SCI recovery (at 8 weeks), it did not improve the final extent of recovery (26 weeks). However, walking ability was not assessed.

Immediate release, 4-AP capsules have been shown to have no benefit to ambulation as indicated by 2 RCTs (van der Bruggen et al. (2001), n=20; DeForge et al. (2004), n=15). However, the study of van der Bruggen et al. (2001) was not directed solely at exploring the effects on walking and therefore the heterogenous nature of the subject groups may have confounded the ambulation results. Furthermore, differences in intervention (i.e. up-titration to 15-45mg/day over 4 weeks in van der Bruggen et al. (2001) vs. up to 10mg 4X/day for 2 weeks in DeForge et al. (2004)) and the lack of consistent clinically relevant outcome measures complicates the interpretation of the available evidence.

Two of the studies noted above used a combination of pharmacological and physical therapy gait training interventions. One small RCT (Walker & Harris 1993) (N=9) provided some evidence that a combination of physical therapy (including gait training) and GM-1 ganglioside improved motor scores, walking distance, and walking speed in chronic SCI participants compared to physical therapy plus placebo (though Klose and Calancie (1994) posted a critique and re-calculated their results with different conclusions). Other results from the pre-test/post-test study conducted by Fung et al. (1990) provide level 5 evidence that clonidine and cyproheptadine in conjunction with BWSTT may be effective in enabling nonambulatory incomplete SCI patients to achieve overground ambulation with assistive devices.

A more recent study examined the effects of combining L-Dopa (dopamine precursor) with gait retraining in a group of individuals with acute/sub-acute incomplete SCI (Maric et al. 2008). Unlike the promising effects of L-Dopa on motor recovery following stroke (Scheidtmann et al. 2001), there was no added benefit in this SCI group. Although spinal neural circuits can certainly undergo plastic changes, the results of this study suggest that dopaminergic neurons may not have been sufficiently stimulated by the dosage used here, or that they may not contribute to motor recovery associated with gait retraining.

There is limited evidence that oral Baclofen may improve walking after SCI from two Level 5 pre-post studies that examined the effects of Baclofen on gait (total N = 21, 0.5-27 years post-injury (Azouvi et al. 1996; Norman et al. 1998). Avouzi et al. (1996) showed increases in the Functional Independence Measure (FIM™) walking scores in 5 of 18 patients, and 2 people acquired the ability to climb stairs following Baclofen administration. Participants in the Norman et al. (1998) study only showed minor changes in walking when using this drug.


There are nine Level 1 and 2 studies (DeForge et al. 2004; Duffell et al. 2015 (2); Leech et al. 2014; van der Bruggen 2001; Grijalva 2010; Maric 2008 ; Stewart et al. 1991; Wainberg et al. 1990) that found no significant differences of the effects of medication on walking ability.

There is level 1 evidence (Walker & Harris 1993), limited by a small sample size, that GM-1 ganglioside combined with physical therapy improves walking ability in chronic incomplete SCI patients (though Klose and Calancie (1994) posted a critique and re-calculated their results with different conclusions).

There are two level 4 studies (Azouvi et al. 1996; Segal & Brunnemann 1998) that found significant differences in aspects of gait after baclofen administration.