Spasticity reduction using electrical stimulation in the lower limb of spinal cord injury patients

Arjan van de Salm, MsC (Researcher)
Peter Veltink, PhD (Project leader)
Maarten IJzerman, PhD (Project leader)
Anand Nene, MD, PhD
Hermie Hermens, PhD

On October 21, 2005, Arjan van der Salm, defended his dissertation entitled: Spasticity reduction using electrical stimulation in the lower limb of spinal cord injury patients

Summary

In The Netherlands approximately 9.000 people suffer from a Spinal Cord Injury (SCI) and 19 percent of them have the ability to walk. Especially, patients with walking abilities are anxious to improve their gait performance. The gait of SCI patients is often impaired due to a decreased muscle activity, paralysis, or an increased muscle activity, spasticity. It has been found that important impairments in these patients are an exceeded plantar flexion and a decreased knee flexion, both occurring during swing. These impairments may decrease the swing limb advancement, which may affect the step length.

One of the treatment modalities, which is thought to reduce spasticity is electrical stimulation. Several studies describe that electrical stimulation has a carry-over effect, which can be used to reduce spasticity up to approximately 24 hours. In addition, it has also been found that electrical stimulation has an instant (or direct) effect. Both these effects may be useful to improve the gait performance in SCI patients. The goal of the thesis was to study the effect of electrical stimulation in SCI patients to reduce spasticity in muscles of the leg and to improve gait performance, using both the carryover and the instant effect. To achieve this goal it was necessary to have appropriate spasticity measures for the carry-over effect and the instant effect, thus during rest and gait. Using these measures, we studied the carryover and the instant effect of electrical stimulation on spasticity.

In chapter two the development of a new method for the measurement of spasticity is described. The method is based on the (modified) Ashworth scale and Tardieu scale, which are used frequently by clinicians. For the assessment, 30 to 45 stretches of the triceps surae over the whole range of motion were applied at varying velocities, measuring from 30 to 150 o/s. In normal gait the maximal angular velocities of ankle dorsal flexion are approximately 90 o/s. Other stretch velocities during daily life in spastic patients may be even higher, for example sudden foot-ground-contact during transfers. In addition, during activities of daily life the whole range of joint motion is used, therefore we used this range during the measurement. The electromyography (EMG) responses were measured to assess the reflex excitability, and the torque over the ankle joint was determined. The angle at which the reflex was initiated was also measured. The assessment method was used in 9 complete SCI patients. The results showed a significant increase in the EMG response at increasing velocities. It was concluded that the assessment method of spasticity using full range passive movements provides objective outcomes and distinguishes between reflexive and non-reflexive components of muscle stiffness.

In chapter three the assessment method for spasticity was studied for its criterion validity and reliability. A cross-sectional test-retest design over 3 to 4 separate days was performed in 8 complete SCI patients. To study the criterion validity the assessment outcomes were compared to the modified Ashworth scale, clonus score and H/M-ratio. The Ashworth scale and the clonus score are clinically used scales to determine spasticity. The H/M-ratio is an objective, neurophysiological measure, which provides an outcome for the spinal reflex excitability. The reliability was determined with the Intra-Class correlation (ICC) coefficient and the responsiveness was calculated. It was found that the EMG responses at stretches with a velocity of 75 or 100 °/s correlated significantly with the H/M-ratio (Spearman's rho ≥ 0.68). In addition, the results indicated that the clonus score was related to the EMG responses. No correlations were found with the modified Ashworth scale, but this might be expected because only measures for reflex excitability were studied, whereas, the torque outcomes were not included in this study, because the data of only 4 patients were available. The Modified Ashworth scale might be correlated to the torque outcomes.

The reliability was good for the EMG responses at stretch velocities of 75 °/s and 100 °/s (ICC ≥ 0.78), and for the angle at which the reflex was initiated (ICC 0.71). On the other hand, the calculated responsiveness was relatively low, being approximately 0.30 for the EMG responses and 0.54 for the reflex-initiating angle. This is mainly due to the relatively high variance caused by the variability in spasticity between the days, compared to the relatively low intervention effect of electrical stimulation we used.

Chapter four describes the carry-over effect of electrical stimulation, to reduce spasticity in the triceps surae, for three stimulation methods. The used design was a placebo-controlled study with repeated measurements after the interventions. Ten complete SCI patients were included in the study. These patients were measured on 4 separate days. The intervention consisted of a 45 minutes cyclic electrical stimulation on the antagonist, agonist or dermatome. Alternatively, a placebo approach was applied. Each day another intervention was investigated. The outcomes measures were the modified Ashworth scale, clonus score, H/M-ratio, stretch response over the whole range of motion and the reflex-initiating angle.

It was found that stimulation of the agonist provided a significant reduction in the modified Ashworth scale of 46%. The antagonist stimulation reduced the reflex-initiating angle significantly (7%). The outcomes of the reflex excitability showed no significant changes in any of the stimulation methods. The effects found, in both the modified Ashworth scale and the reflex-initiating angle, were associated with changes in the visco-elasticity of the surrounding tissue. It was concluded that the carry-over effect of electrical stimulation provides no changes in the spinal reflex excitability.

The reflex excitability and its variability during gait are described in chapter five. To determine the spinal reflex excitability of the vastus lateralis, we measured the H-reflex during the mid stance and mid swing phases. This measurement was done in 10 healthy and 3 spastic incomplete SCI subjects. The H/M-ratios were determined during both phases of gait and their variability was determined using a modulation index. Results pointed out that the H/M-ratios of the spastic patients were approximately 3 times higher than the outcomes of the healthy subjects (P<0.05). The average modulation index in the healthy subjects group was 42%, whereas the modulation index in the patients group was 14%. Due to the large variation in the outcomes, this difference was not significant.

In chapter six the effect of hamstrings and L3/4 dermatome stimulation during the swing phase of gait was investigated. Both interventions were studied on separate days, and the intervention effect was compared with a baseline measurement. Gait performance outcomes, i.e. step length, maximum hip flexion and maximum knee flexion during swing, were determined in 5 spastic SCI patients. It was found that the hip flexion decreased during swing due to the electrical stimulation of the hamstrings, rather than an increase of knee flexion. Concerning the bi-articular position of the hamstrings on the dorsal side, this was explicable. No other significant changes in gait performance were found. Additionally, we investigated the effect of the electrical stimulation on the spinal reflex excitability during the swing phase of gait. For this the H/M-ratios of the vastus lateralis muscles of 3 patients were measured. In one patient the H/M-ratio was increased while the L3/4 dermatome was stimulated. Another patient showed a decrease of the H/M-ratio during hamstrings stimulation. One patient showed no relevant change at any of the interventions. This indicates that spastic muscles can be inhibited using the instant effect of electrical stimulation.

Publications from this thesis

Development of a new method for objective assessment of spasticity using full range passive movements. Van der Salm A, Veltink, PH, Hermens HJ, IJzerman MJ, Nene AV. Arch Phys Med Rehab, 86(10): 1991-7, 2005.
Gait impairments in a group of patients with incomplete spinal cord injuries and their relevance regarding therapeutic approaches using FES. Van der Salm AJ , NeneAV, Maxwell DJ, Veltink PH, Hermens HJ, IJzerman MJ. Artificial Organs, 29(1): 8-14, 2005.
Criterion validity and reliability of a method for objective assessment of spastic hypertonia using full range passive movements. Van der Salm A, Veltink PH, Hermens HJ, IJzerman MJ, Nene AV. Arch Phys Med Rehab. Submitted.
Comparison of electrical stimulation methods for reduction of triceps surae spasticity in SCI-patients. Van der Salm A, Veltink PH, IJzerman MJ, Groothuis-Oudshoorn KCG, Nene AV, Hermens HJ. Arch Phys Med Rehab, 87(2): 222-8, 2006.
Modulation of the vastus lateralis H-reflex during gait in healthy subjects and patients with spinal cord injury. Van der Salm A, Veltink PH, Hermens HJ, Nene AV, IJzerman MJ. Gait & Posture. Submitted.
Effect of electrical stimulation of hamstrings and L3/4 dermatome on H/M ratio and performance of gait in spastic SCI-patients. Van der Salm A, Veltink PH, Hermens HJ, Nene AV, IJzerman MJ. Neuromodulation. Submitted.