Multiple Sclerosis

0 downloads 17 Views 146KB Size Report
Mar 12, 2009 - electrical stimulation and therapeutic exercise on walking performance for people with .... tated and impairment–based approaches to physio-.

Multiple Sclerosis

A randomized trial to investigate the effects of functional electrical stimulation and therapeutic exercise on walking performance for people with multiple sclerosis CL Barrett, GE Mann, PN Taylor and P Strike Mult Scler 2009 15: 493 originally published online 12 March 2009 DOI: 10.1177/1352458508101320 The online version of this article can be found at:

Published by:

Additional services and information for Multiple Sclerosis can be found at: Email Alerts: Subscriptions: Reprints: Permissions: Citations:

Downloaded from at Bournemouth University on November 8, 2010


Multiple Sclerosis 2009; 15: 493–504

A randomized trial to investigate the effects of functional electrical stimulation and therapeutic exercise on walking performance for people with multiple sclerosis CL Barrett, GE Mann, PN Taylor and P Strike Background Functional electrical stimulation (FES), is a means of producing a contraction in a paralyzed or weak muscle to enable function through electrical excitation of the innervating nerve. Objective This two-group randomized trial assessed the effects of single channel common peroneal nerve stimulation on objective aspects of gait relative to exercise therapy for people with secondary progressive multiple sclerosis (SPMS). Methods Forty-four people with a diagnosis of SPMS and unilateral dropped foot completed the trial. Twenty patients were randomly allocated to a group receiving FES and the remaining 24 to a group receiving a physiotherapy home exercise program for a period of 18 weeks. Results The exercise group showed a statistically significant increase in 10 m walking speed and distance walked in 3 min, relative to the FES group who showed no significant change in walking performance without stimulation. At each stage of the trial, the FES group performed to a significantly higher level with FES than without for the same outcome measures. Conclusion Exercise may provide a greater training effect on walking speed and endurance than FES for people with SPMS. FES may provide an orthotic benefit when outcome is measured using the same parameters. More research is required to investigate the combined therapeutic effects of FES and exercise for this patient group. Multiple Sclerosis 2009; 15: 493–504. Key words: dropped foot; exercise; functional electrical stimulation; gait; multiple sclerosis; physiotherapy

Introduction Multiple sclerosis (MS) is the most common adult neurological disease in Western Europe and affects 85,000 people in the UK [1]. Approximately 75% of people with MS have difficulty with locomotion and are dependent upon a walking aid or wheelchair [2,3]. One clinical presentation that contributes to these problems with mobility is dropped foot. This can be described as an inability to dorsiflex the foot during the swing phase of gait, either due to weakness of the prime movers, increased tone in the plantarflexors, or disordered neural control causing co-contraction of agonist and antagonist muscles. This inappropriate neuromuscular

control around the ankle typically leads to abnormal patterns of gait, limiting aspects of walking such as speed, endurance, energy expenditure, and balance [4]. These deficits restrict participation in activities of daily living and negatively affect the quality of life. Functional electrical stimulation (FES), first used by Liberson [5] in 1961 to improve walking for people with dropped foot caused by stroke, is a means of producing a contraction in a paralyzed or weak muscle to enable function through electrical excitation of the innervating nerve. The researchers described the use of surface stimulation of the common peroneal nerve to produce ankle dorsiflexion and eversion, timed by a pressure-sensitive foot switch placed

Preliminary data from this study was presented as a poster at the International Functional Electrical Stimulation Society annual conference, Bournemouth, September 2004. The National Clinical FES Centre, Department of Clinical Sciences and Biomedical Engineering, Salisbury District Hospital, Salisbury, Wiltshire SP2 8BJ, UK Correspondence to: Paul N Taylor, Department of Clinical Sciences and Biomedical Engineering, Salisbury District Hospital, Salisbury, Wiltshire SP2 8BJ UK. Email: [email protected] Received 26 November 2007; accepted 23 November 2008

© SAGE Publications 2009


Downloaded from at Bournemouth University on November 8, 2010


CL Barrett et al.

inside the shoe and controlled by a battery-operated portable stimulator carried in the pocket. Stimulation began when the heel was lifted from the ground, continuing until weight returned to the foot switch. The authors reported positive responses with a therapeutic improvement in unassisted walking, as well as an immediate orthotic benefit when wearing the stimulator. The benefits of dropped foot stimulation for people with stroke were later confirmed by a randomized controlled trial [6], which led to FES being recommended by the National Clinical Guidelines for Stroke in 2000 [7]. Research supporting the efficacy of dropped foot stimulation for People With MS (PWMS), however, is limited to retrospective surveys and audits. Carnstam, et al. [8] were the first to report the use of common peroneal stimulation during walking for five people with MS. In a series of experiments involving one of these subjects using electromyographic gait recording, they reported an increase in dorsiflexor strength and a reduction in calf tone after 10– 15 min of stimulation. Karsnia, et al. [9] conducted a survey of 43 PWMS subjects, most of whom reported that they could walk further and for longer with stimulation, and the device was well accepted. More recently, Taylor, et al. [10] showed that after 18 weeks of using dropped foot stimulation, 21 PWMS demonstrated a mean increase of 16% in walking speed and a reduction in energy expenditure of 24% measured using the physiological cost index (PCI) when walking with stimulation compared to without stimulation. In a second audit study by the same group of researchers, it was reported that one third of a group of 78 PWMS showed an improvement in unassisted walking speed of more than 10%, and a questionnaire survey indicated that the most common reasons for using dropped foot stimulation were that it reduced the effort of walking, reduced tripping, and improved confidence during walking [11]. Interventions such as aerobic exercise and targeted physiotherapy are more established than FES as effective means of improving walking performance for PWMS. It has been demonstrated that PWMS benefit from general aerobic and resistance training, showing improvements in areas such as general strength [12], aerobic capacity, selfreported mobility levels [13], and 10 m walking speed [14]. Blocks of inpatient and outpatient physiotherapy have both been shown to improve general mobility and quality of life for PWMS [15–17], while Lord, et al. [18], showed that both task orientated and impairment–based approaches to physiotherapy produced significant improvements in walking quality. The specific types, dosages and elements of physiotherapy, and exercise programs that are most effective for PWMS, however, have yet to be confirmed [27].

Although the evidence for exercise and physiotherapy is increasing, there are currently no randomized trials that investigate the therapeutic effects of FES on the walking performance of PWMS. There is a clear need for such research to support clinical reports of its effectiveness. There is also a need to establish the effects of FES relative to more established therapeutic interventions for PWMS such as exercise.

Methods This study was a two-group randomized trial to assess the effects of single channel common peroneal nerve stimulation on objective aspects of gait relative to exercise therapy, each applied over an 18 week period. The primary outcome variable was walking speed over 10 m after 18 weeks of intervention. Secondary outcome measures were PCI and distance walked in 3 min. The design included an initial pilot element to test procedures and methods. Ethical approval for the study was obtained from the Salisbury and South Wiltshire Research Ethics Committee.

Sample size The between-subject variation in 10 m walking speed recorded after 18 weeks of Odstock Dropped Foot Stimulator (ODFS) use was estimated using retrospective audit data (n = 24 and SD = 0.3 m/s) [10]. The same audit data revealed a high degree of within-subject correlation in walking speeds recorded during 18 weeks of ODFS use (Pearson correlation coefficient circa 0.9), a feature that can be exploited by use of analysis of covariance (ANCOVA) with baseline walking speed as covariate. Using ANCOVA, a sample size of 25 subjects in each treatment group would be required to detect a minimum between-group difference in mean walking speed after 18 weeks of ODFS use of 0.1 m/s, assuming a two-sided test significance level of 5% with a test power of 80%. The possible need for a nonparametric ANCOVA was cautiously assumed in the sample size calculation, as was a possible ‘drop out’ rate up to 15%. Sample size calculation was performed using STATATM v.9 and checked using Bootstrap resampling of the audit sample data.

Subject selection Individuals were recruited to the study from the waiting list for FES treatment at the FES clinic where the trial was conducted. Participants lived

Multiple Sclerosis 2009; 15: 493–504

Downloaded from at Bournemouth University on November 8, 2010

MS FES and Exercise for dropped foot within a 100 mile radius of the clinic, and had been referred by either their General Practitioner (GP) or consultant specifically for treatment of their dropped foot with FES. Subjects were accepted for participation in the trial if they were at least 18 years of age with no upper age limit, and had a diagnosis of secondary progressive MS made by their neurologist and a rating in the range of 4–6.5 on the Kurtzke Expanded Disability Status Scale (EDSS) [19]. Subjects also needed to have a predominantly unilateral dropped foot impairing mobility (identified by both the referring doctor and the researchers), passive range of ankle dorsiflexion to at least plantigrade, a good response to stimulation of the common peroneal nerve when assessed for the trial by the researchers and no previous use of FES. Subjects with cognitive or psychiatric problems that affected their ability to understand or comply with treatment, and any other neurological or orthopedic problem that may have affected mobility or response to treatment were not accepted for participation in the trial.

Randomization In view of the relatively small number of subjects to be enrolled in the study, randomized allocation of eligible patients to one or the other treatment group was undertaken using computer-generated random


permuted blocks of size 4, in order to eliminate chance imbalance in group size. The allocation codes were prepared and concealed in sealed envelopes by a statistician independent of the study team, and revealed sequentially with each new patient enrolment. The envelopes were opened by participants in the presence of the recruiting researcher, after subject selection, consent to participate in the trial, and the first unassisted 3 min walk/first set of unassisted 3 × 10 m walks had been completed.

Interventions Subjects were allocated to one of two groups, receiving either FES or a programme of simple physiotherapy home exercises. All interventions were administered by the researchers conducting the trial. The protocol was designed to reflect current FES clinical practice (Figure 1) [11,20]. The FES group received common peroneal nerve stimulation to correct dropped foot using the ODFS. The device has provision for adjusting the rate of increase of the stimulation at the beginning of each output, rising ramp, to prevent a rapid dorsiflexion, which may cause a stretch reflex in the calf muscle. Additionally, the length of time that the stimulation continues after heel strike can be adjusted, providing an eccentric contraction of the anterior tibialis, lowering the forefoot to the ground at heel strike. These parameters, together with pulse

Trial Protocol FES Group

Week 0 Appointment 1

Week 0 Appointment 2

Week 6,12,18

Exercise Group

3 minute walk ↓ Falls diary explained ↓ 3 x 10m walk (no FES) ↓ Randomisation ↓ FES set up

3 minute walk ↓ Falls diary explained ↓ 3x 10m walk ↓ Randomisation ↓ Exercises taught

FES checked

Exercises checked

3x 10m walk (with FES)

3x 10m walk

3 minute walk ↓ Exercise check/ progression ↓ 6 x 10m walks ↓ 3 x 10m walks (With FES) 10 minute rest ↓ ↓ 3 minute walk 10 minute rest ↓ 3 minute walk (With FES) FES set up if volunteer wished 3 minute walk, (no FES) ↓ FES check/ adjustment ↓ 3x 10m walks (No FES)

Figure 1 Trial Protocol.

Multiple Sclerosis 2009; 15: 493–504

Downloaded from at Bournemouth University on November 8, 2010


CL Barrett et al.

width (0–365 μs) and current amplitude (20– 100 mA), were adjusted to generate sufficient muscle contraction to produce dorsiflexion and eversion. A fixed frequency of 40 Hz was used. The device was adjusted when required at six weekly reviews throughout the study period. When initially set up with FES at week 0, (appointment 1), subjects were given 3 days to practice with the device before a follow-up check was made (appointment 2). Use of the ODFS was initially restricted to short walks increasing daily over 2 weeks, after which unrestricted daily use was permitted. Subjects were encouraged to wear the device most of the day and to switch it on and off as required, which all were able to do either independently or with the assistance of a care taker. Participants in the exercise group were taught a series of simple home exercises that aimed to improve trunk and pelvis stability, lower limb muscle length and strength, and balance and control of movement in a range of positions from lying through to standing. Specific exercises appropriate to each individual were chosen from the list in Appendix 1. Exercise programmes were reviewed and progressed where appropriate by choosing more challenging ones from the list at six weekly intervals throughout the study. They were instructed to perform the exercises every day for approximately half an hour, one to two times daily at home. All exercises were prescribed by the research physiotherapists involved in conducting the trial. To mirror the FES intervention, after being taught the exercises at week 0, (appointment 1), subjects were given 3 days to practice at home before returning to have the exercises checked (appointment 2). At the end of the study, subjects randomized to the exercise group were given the opportunity to try electrical stimulation for 6 months.

Outcome measures Self-selected walking speed over 10 m Self-selected walking speed has been shown to be a valid and reliable method of measuring walking performance for PWMS [27]. Walking speed was timed over a 10 m distance, with 1 m allowed at either end for acceleration and deceleration. The total distance walked was therefore 12 m. For the FES group, three walks with stimulation and three without were completed at each assessment stage. At week 0, the first three walks without stimulation were completed at appointment 1, and the second three walks with stimulation at appointment 2, (see Figure 1). This allowed the subject a

period of time to become accustomed to the stimulation. At weeks 6, 12, and 18, the order of walks with and without stimulation was randomized to avoid the effects of fatigue. To mirror the FES group, the exercise group completed two sets of three walks at the same stages of each appointment that the FES group completed their walks with and without stimulation. Both groups were allowed to use their walking aid of choice, as long as this remained constant throughout the study.

PCI over 10 m This is an indication of the energy efficiency of walking based on the increase in heart rate during walking compared to the resting heart rate [21]. Resting and working heart rates were measured using a heart rate monitor and watch, and PCI was calculated using the following calculation: change in heart rate (beats per minute)/walking speed (ms−1) × 60. Resting heart rate was defined as the lowest reading recorded while the patient was sitting at rest for 2 min, and working heart rate as the highest recording immediately after the subject had finished the 10 m walk.

Walking distance in 3 min The distance walked in 3 min was recorded. Endurance walking over 6 min has been shown to be a feasible, reproducible and reliable method of testing endurance for people with MS [24]. Unfortunately, the 6 min walk proved to be too demanding for the initial 11 pilot study patients and the walking time was reduced to 3 min. The implications of this will be examined in the limitations section of the discussion. Both groups completed two 3 min walks at each assessment stage except the first. In the FES group, one walk was made without and one with stimulation. At week 0, only one walk was completed without stimulation because it was considered that the subjects would have built up insufficient fatigue resistance to complete endurance tests with FES at initial set-up. If a subject failed to complete the 3 min walk, the distance completed at the point where they stopped and the time taken to complete this distance was recorded.

Data collection and analysis Data were collected and recorded by five researchers who also provided the treatment interventions.

Multiple Sclerosis 2009; 15: 493–504

Downloaded from at Bournemouth University on November 8, 2010

MS FES and Exercise for dropped foot Three of these researchers were also involved in the design and write up of the study. The nature of the treatment allocations was concealed in subsequent data analysis by the independent statistician. Methods of analysis included descriptive summary measures, Anderson-Darling test of Normal distribution assumption and basic tests for equality of means or mean changes using standard independent sample or paired sample tests, parametric or nonparametric as appropriate. ANCOVA was performed on outcomes at 18 weeks, with baseline measures as covariate [25], after assessment of data conformity to ANCOVA assumptions. All analyses were repeated using nonparametric ANCOVA [25], as a check on the integrity of conclusions drawn. All analysis was undertaken using the software package STATATM v.9.

Results A total of 64 subjects were recruited. The first 11 were used to pilot the trial protocol and patient compliance with/acceptance of those procedures. Data generated by these pilot study patients have not been included in the analysis. Subsequently, 53 subjects were recruited, 26 to the FES group and 27 to the Exercise group. Seven subjects dropped out at a very early stage of trial, one failed to start and one breached protocol. Participant flow and the reasons that participants dropped out are shown using the Consort Diagram in Figure 2. The participant that dropped out because of deteriorating MS, allocated to the FES group, was subsequently the subject of a sensitivity analysis because they could not be characterized as ‘missing at random.’ Data are therefore available for 44 subjects, 20 in the FES group and 24 in the Exercise group. We did not adopt an intention-to-treat protocol, as the study was essentially exploratory, that is, to assess whether FES has benefit for this patient group. The nine “drop outs”/excluded subjects are therefore not included in analysis. The drop outs all occurred so early in the trial that for the majority we have no data, or baseline data only, and minimal or no information would have been gained as a result of completing an intention to treat analysis. No obvious errors were found when study data was initially checked using appropriate range filters, for example, flag up age 110. No significant departures from normal distribution assumptions were observed in treatment responses when distributional properties and presence of outliers were assessed for using graphical plots, examination of summary measures (mean, median, standard deviation, range), and Anderson-Darling Normality tests. Consequently, all analyses were undertaken using parametric tests that included paired samples


t-test, and parametric ANCOVA. All conclusions were checked using nonparametric test equivalents, including ANCOVA. No discrepancies were found between parametric tests and their nonparametic equivalents relating to any of the conclusions drawn. Basic demographic statistics are presented in Table 1. There are small but noticeable disparities between the FES and Exercise group means for both age and time since the diagnosis. As a precautionary measure, both variables were used as covariates in subsequent ANCOVA analysis to assess their possible impact on treatment response at 18 weeks. EDSS scores for both groups were clearly very similar and ranged from a score of 4, (able to walk 300 m without a walking aid) to 6.5 (able to walk 20 m with a walking aid). Although a score of 6.5 implies that they would perform poorly on the endurance walk, subjects at this level were included in the study on the basis that they were able to complete the primary outcome assessment, (10 m walk). Guidelines for the 6 min endurance walk test for people with MS clearly state that rest periods are allowed during the test at any stage [24]. Many subjects, not just those with an EDSS score of 6.5, took advantage of this allowance. Tables 2–4 present the means (standard deviations in brackets) for walking speed over 10 m, PCI, and distance walked in 3 min for both treatment groups at baseline for 6, 12, and 18 weeks, respectively. There are two columns for the FES group, one displaying data recorded with stimulation and the other displaying data recorded without stimulation. The column for the exercise group contains the mean walking speeds calculated from data generated by both sets of walks at each appointment.

FES vs exercise between group comparisons Initially three variables, baseline response, age, and time since diagnosis, were entered as covariates in the ANCOVA calculations. The ANCOVA results revealed that age and mean time since diagnosis had no significant impact on any of the response variables at 18 weeks (ANCOVA P > 0.25 for both). Conversely, baseline measures were, as expected, clearly associated with outcome at 18 weeks, (P < 0.001 for all three response variables). The ANCOVA baseline adjusted mean values for the FES and the Exercise treatment groups along with corresponding ANCOVA statistics are presented in Table 5. The only significant difference between groups identified after 18 weeks of treatment was observed for 10 m walking speed: the exercise group walked, on average, +0.08 m/s faster (95% confidence interval +0.01 to +0.15) than the Multiple Sclerosis 2009; 15: 493–504

Downloaded from at Bournemouth University on November 8, 2010


CL Barrett et al. Consort Flowchart Assessed for eligibility (n= 87)

Excluded (n= 34) Enrolment


Allocated to FES Group (n= 26) Received allocated intervention (n= 25) Did not receive allocated intervention (n= 1) Reason; Subject felt light headed when FES applied- decided to discontinue


Lost to follow-up (n= 1) Reason; Patient moved abroad

Reasons Failure to meet inclusion criteria (n = 23) Pilot subjects (n = 11)

Allocated to Exercise Group (n= 27) Received allocated intervention (n= 27)

Lost to follow-up (n= 0) Discontinued intervention (n= 2)

Discontinued intervention (n= 4) Reasons


Reasons 1. Fall unrelated to intervention 2. Development of lower back unrelated to intervention

1. Patient fell and fractured femur 2. Swollen legs made FES ineffective 3. Lifestyle factors made FES inconvenient 4. Deteriorating MS and unable to attend appointments

Analysed (n= 20 )


Analysed (n= 24)

Excluded from analysis (n= 0 )

Analysis Excluded from analysis (n= 1) Reason; Patient breached protocol

Figure 2 Consort Flowchart.

Table 1

Demographic data Final cohort

Age (mean) Sex Time since diagnosis (mean) Kurtze score (mean)

Pilot subjects

Drop outs

FES (n = 20)

Exercise (n = 24)

FES (n = 6)

Exercise (n = 5)

FES (n = 7)

Exercise (n = 2)

52.1 years SD 6.7 40–63 years 15 Female 5 Male 13.6 years SD 8.3 5–32 years 5.9 SD 0.76 4–6.5

56.6 years SD 9.0 39–73 years 16 Female 8 Male 17.7 years SD 8.3 6–42 years 5.8 SD 0.78 4–6.5

57.3 years SD 11.1 42–71 3 Female 3 Male 9.2 years SD 6.7 3–22 years 6 SD 0.55 5–6.5

60.4 years SD 3.8 55–64 4 Female 1 Male 19.0 years SD 9.8 7–34 years 5.9 SD 0.55 5–6.5

49.1 years SD 9.1 31–57 4 Female 3 Male 12.0 years SD 6.2 6–20 6 SD 0.0 6–6

56.5 years SD 6.4 52–61 1 Female 1 Male 15.5 years SD 0.7 15–16 6 SD 0.0 6–6

Multiple Sclerosis 2009; 15: 493–504

Downloaded from at Bournemouth University on November 8, 2010

MS FES and Exercise for dropped foot Table 2

Mean (SD) walking speed over 10 m/s


Mean walking speed

0 6 12 18 Paired t-test Week 0 vs week 18

Table 3


Paired t-test Mean walking Speed FES group without Exercise ms−1 (n = 24) stimulation vs FES group with stimulation

FES group without stimulation ms−1 (n = 20)

FES group with stimulation ms−1 (n = 20)

0.79 (0.35) 0.78 (0.35) 0.77 (0.34) 0.73 (0.35) P = 0.155

0.79 (0.31) 0.83 (0.35) 0.82 (0.33) 0.80 (0.35) P = 0.592

P > 0.50 P = 0.001 P = 0.001 P = 0.001

0.68 (0.28) 0.72 (0.27) 0.72 (0.27) 0.77 (0.29) P = 0.001

Mean (SD) PCI over 10 m


FES group without stimulation Btm−1 (n = 20)

FES group with stimulation Btm−1 (n = 20)

Paired t-test No stimulation versus stimulation

Median PCI exercise group Btm−1 (n = 24)

0 6 12 18 Paired t-test Week 0 vs week 18

0.68 (0.77) 0.92 (1.66) 0.73 (0.97) 0.82 (1.17) P = 0.35

0.78 (1.18) 0.75 (1.15) 0.68 (0.96) 0.74 (1.12) P = 0.48

P = 0.35 P = 0.17 P = 0.08 P = 0.38

0.68 (0.52) 0.60 (0.47) 0.61 (0.49) 0.66 (0.54) P = 0.53

Table 4

Mean (SD) distance walked in 3 min (meters)


FES without stim (n = 20)

FES with stim (n = 20)

Paired t-test FES with stimulation vs FES without stimulation

Exercise (n = 24)

0 6 12 18 Paired t-test Week 0 vs week 18

99 (42) 112 (51) 111 (53) 112 (50) P = 0.24

n/a 122 (56) 124 (51) 125 (55) P = 0.34

n/a P = 0.010 P = 0.003 P = 0.004

97 (44) 106 (46) 111 (43) 113 (46) P = 0.005

Table 5 Baseline-adjusted mean standard error (SE) treatment outcomes at 18 weeks Outcome variable


ANCOVA-adjusted mean + (standard error)

Walking speed over 10 m

FES (n = 20) Exercise (n = 24) Difference 95% ci FES (n = 20) Exercise (n = 24) Difference 95% ci FES (n = 20) Exercise (n = 24) Difference 95% ci

0.74 (0.026) 0.82 (0.024) 0.081, P = 0.028 +0.01 to +0.15 0.69 (0.041) 0.70 (0.037) 0.01, P = 0.81 −0.01 to +0.15 124 (8.5) 112 (7.9) 11, P = 0.334 −0.01 to +0.13

PCI recorded over 10 m Distance walked in 3 min (in meters)

FES-assisted group, (P = 0.028). The nonparametric rank ANCOVA [23] analysis delivered the same overall conclusions. The impact of the nonrandom drop-out (no data collected) referred to above was explored through a

sensitivity analysis comprising three scenarios for the patient’s possible outcome, best, the average, and the worst of the results actually recorded in this study. ANCOVA analysis of each scenario delivered P values of 0.014, 0.020, and 0.065, respectively, and the estimated differences in 10 m walking speed being:   

Difference = 0.088, with a 95% confidence interval of 0.02 to 0.16 Difference = 0.081, with a 95% confidence interval of 0.01 to 0.15 Difference = 0.065, with a 95% confidence interval of −0.004 to 0.14

Within-group mean changes Mean changes between baseline and 18 week measures were nonsignificant for all three outcome measures in the FES group, both with and without stimulation. For the exercise group, there was a Multiple Sclerosis 2009; 15: 493–504

Downloaded from at Bournemouth University on November 8, 2010


CL Barrett et al.

highly significant increase in both walking speed over 10 m (P = 0.001) and distance walked in 3 min (P = 0.005) over the 18 week trial period. There was no significant change observed for PCI for either group.

FES within-group comparisons; with stimulation vs without stimulation For each of the outcome measures, (10 m walking speed, PCI, and distance walked in 3 min), the performance of the FES group with and without stimulation was compared at each of the time points (6, 12, and 18 weeks). There were significant increases in mean values for 10 m walking speed and distance walked in 3 min with stimulation compared to without stimulation at each time point, (Tables 2 and 4). No significant differences were recorded for mean PCI (Table 3). It was noted that for 10 m walking speed, (the primary outcome measure), at each individual time point, mean responses recorded with stimulation essentially did not change. There was a concurrent decrease in performance over time without stimulation. Although this did not reach significance, (P = 0.155), it requires further investigation.

Discussion This study was a two-group randomized trial designed to estimate and compare the effects of single channel common peroneal nerve stimulation for dropped foot with a simple exercise program on objective aspects of walking performance. To the authors’ knowledge, this is the first study to examine the effects of FES on walking speed for people with MS or to compare its effects with the more widely accepted intervention of exercise. The reader is reminded that the primary outcome measure on which sample size calculations were based, was unassisted walking speed over 10 m. All conclusions drawn in respect to other measures are therefore made cautiously.

therapy appears to deliver a more effective means of improving unassisted walking performance than FES for people with secondary progressive multiple sclerosis (SPMS). A note of caution is indicated in respect of the between group difference in walking speed at 18 weeks, the sensitivity analysis delivering a nonsignificant (P = 0.065) result for the most pessimistic of the three scenarios evaluated. These findings reflect previous clinical audit data for the effects of FES on the unassisted walking performance of PWMS, [10], which also showed no changes in unassisted 10 m walking speed over an 18 week intervention period. In terms of the exercise group, however, these findings were interesting. Previous research has reported a significant improvement in objective parameters of walking for PWMS using more intensive bursts of therapy of two to three times weekly sessions lasting 3– 8 weeks [15–17] and UK National Clinical Guidelines for MS recommend more comprehensive multidisciplinary input to achieve clinical benefit [24]. A possible explanation is that prior to the trial, many subjects received very little therapeutic input, were physically inactive and were consequently quite debilitated. It is therefore likely that any encouragement given by a health professional to increase activity would improve physical performance, fitness, and independence through reducing fear of exertion and increasing motivation to exercise. Furthermore, subjects in the exercise group may have been motivated by the promise of being given a walking stimulator at the end of the trial and were, therefore, more prepared to work on their mobility in preparation for this than the group who had already been given FES. The findings indicate that further investigation into the minimum dosage of therapeutic exercise required for PWMS to produce a positive effect on objective parameters of gait is needed. They also support recent suggestions that more work is required to distinguish the psychological and motivational aspects of therapy from the effects of the specific physical intervention [28,29].

Unassisted walking performance

FES-assisted walking performance compared to unassisted walking performance

Although the exercise group showed a significant improvement in unassisted walking speed (P = 0.001) and distance walked (P = 0.005) in 3 min over the 18 week intervention period, the FES group did not. At week 18, the exercise group walked significantly faster than the FES group when means of the two groups were compared using baseline-adjusted ANCOVA (P = 0.028). Therefore in this instance, a simple program of home exercise

At each assessment stage of the study, (6, 12, and 18 weeks), significant improvements were recorded for the FES group with stimulation compared to without stimulation for 10 m walking speed and distance walked in 3 min. These are secondary analyses and as such must be considered as exploratory, but the magnitude and consistent pattern of responses, the associated low P values and evidence of previous clinical observations [10,11,20], support

Multiple Sclerosis 2009; 15: 493–504

Downloaded from at Bournemouth University on November 8, 2010

MS FES and Exercise for dropped foot the suggestion that FES functions as an effective orthosis for PWMS. It was concerning, however, that for 10 m walking speed, this orthotic benefit was attributable to a notable downwards, but nonsignificant, trend in unassisted walking speed (0.79–0.73 ms−1), whereas the FES-assisted speed remained stable. It is possible that in this RCT the cohort’s performance was affected by fatigue due to the significant amount of walking required of them. The observed trend may have been partly due to the effects of electronic stimulation exacerbating this fatigue in a way that exercise did not. Alternatively, the FES group may have become reliant on the FES, so that as the trial progressed they found it increasingly difficult to walk without it. Conversely, a nonsignificant trend towards an improvement in the distance walked in 3 min without FES was observed over the 18 weeks, (99– 112 m). It is possible that with a larger cohort of subjects a statistically significant treatment effect may have been detected. The sample size for this study was generated using 10 m walk data because at the time of calculation there was no published data for endurance walking for PWMS. The effects of FES on distance walking for this population need further examination before results can be accepted with any certainty. Consequently, the effects of FES on the unassisted walking performance of PWMS remain uncertain. Clinically, at least two groups of PWMS have been noted, one of which demonstrates an increase in their unassisted walking as a result of using FES, and one of which does not [11]. It is possible that this may relate to the degree to which the individual is affected by muscle weakness as opposed to loss of nerve conduction. Those who experience a predominance of muscle weakness may benefit to a greater extent from the strengthening and reeducation effects of FES. More investigation into the clinical predictors for response to FES in PWMS is essential so that clinicians can make better informed decisions about its prescription.

Physiological cost index No significant changes in PCI were observed for either group over the trial period for any of the within group or between group analyses completed, except for a significant reduction in PCI at week 6 when assisted and unassisted walking was compared for the FES group. Therefore, no significant training or orthotic effects on energy expenditure were observed overall for either FES or exercise interventions. This does not reflect the limited published data that has examined the effects of dropped foot stim


ulation on energy expenditure during walking for PWMS. In a similar way to walking speed, previous clinical audit data has recorded a beneficial immediate orthotic effect, but no change in unassisted walking performance measured using PCI [10,11]. An explanation for this may be that PCI recordings in this study were significantly influenced by the large amounts of walking demanded of the subjects at each assessment, which led to significant increases in cardiovascular demand and fatigue levels, possibly affecting the reliability of heart rate recordings both at rest and during activity. Additionally, recently published research has indicated that although possibly the most feasible method of measuring energy expenditure clinically, PCI may not be the most reliable or valid measure. Recent comparisons with the “gold standard” measurement of oxygen consumption (VO2) in both healthy subjects [25] and stroke subjects [26] have called into question the reliability and validity of PCI in small samples due to poor test–retest reliability and weak correlations with VO2.

Study limitations and considerations This study was completed using people with SPMS referred to the FES clinic specifically for dropped foot. Consequently, the subjects represented a highly selected group of patients. This is emphasized by the fact that almost a third of those assessed did not fit the inclusion criteria, either because they did not gain a good response to FES or because their mobility did not place them at the correct point on the EDSS scale to complete the required outcome measures. Additionally, seven out of 27 volunteers dropped out during the trial for a variety of reasons that included incompatibility of the device with lifestyle and deteriorating health or mobility that rendered the device ineffective. Results should therefore be generalized with caution to the MS population as a whole. Additionally, power calculations were only completed using the primary outcome measure of 10 m walking speed. Results for the other outcome measures should therefore also be accepted cautiously. Several important limitations to this study relate to its design. Clinical assessors were not blinded to the interventions and also provided treatment, which could have contributed to measurement bias. However double or single blinding of treatments was impractical because subjects meeting the inclusion criteria demonstrated an immediate motor response to FES that would be visually obvious to any assessor. It would also be clearly felt by the volunteers, who needed to actively participate in training to achieve a consistent treatment benefit Multiple Sclerosis 2009; 15: 493–504

Downloaded from at Bournemouth University on November 8, 2010


CL Barrett et al.

from FES. The use of independent assessors would have been more practical, and should be considered for future studies. An important issue is the large amount of walking that subjects were asked to complete at each assessment. Subjects in general found this difficult, and consequently were extremely fatigued by the end of each session. It is possible that this increased cardiovascular demand influenced heart rate recordings giving inaccurate PCI readings. It is also likely that because of the increased fatigue levels, they did not perform as well on walking speed tests as those whose data was included in clinical audits [10,11]. A consequence of these issues with fatigue was that the 6 min walking test had to be reduced to 3 min after the initial 11 subjects had completed piloting of the methods. Unfortunately, the validity and reliability of the 3 min walk for PWMS remains unconfirmed to date. However, reducing the time allowed for the distance walk was the only option available in order to gain data for endurance walking performance during this study. Another consideration relating to the protocol is that there was no baseline period at the start of the trial to assess whether subjects’ walking performance was changing spontaneously. However, as already described, it was noted that when subjects had completed the walking tests once and knew what was expected of them, they were motivated to practice walking further and faster at home. It is therefore likely that a small improvement in walking performance would have been recorded even during a baseline period with no intervention. Finally, the motivation of the “promise” of being given a walking stimulator is an important factor to consider for the exercise group. The significance of the effect of motivation on compliance with exercise regimes cannot be underestimated and it is well documented that the psychological benefits of rehabilitation may be as important as the physical intervention [28,29]. It is likely that the exercise group had fewer “drop outs” and adhered closely to their exercise regime with positive results partly due to the knowledge that they would receive FES at the end of the trial.

Conclusions This study has explored the effects of FES relative to a basic program of home exercise therapy on objective parameters of gait for people with SPMS. Primarily, while a simple program of home exercise therapy appears to significantly increase walking speed and endurance over an 18 week intervention period, single channel common peroneal stimulation does not. However, it does appear to have a significant orthotic benefit, resulting in signifi-

cantly increased walking speed and endurance when performance without stimulation is compared to performance with stimulation.

Areas for further research Further research into the effects of FES on unassisted walking performance in relation to the clinical features of MS is indicated to allow clinicians to make accurate predictions about their patients’ response to the intervention. Additionally, further investigation into the effects of FES on endurance walking for PWMS is required using sample sizes calculated specifically for this aspect of walking performance. More information about the minimum dosage of physiotherapy home exercise that produces a positive change in ambulatory function and the implications of this for current best practice is required. Most importantly, there is a need to investigate how the effects of FES and therapeutic exercise given in combination differ from the effects when either intervention is given on its own.

Acknowledgement The authors gratefully acknowledge the support of the Multiple Sclerosis Trust UK which funded this work. Supplier: Stata is a product of StataCorp, 4905 Lakeway Drive, College Station, TX 77845, USA. The ODFS is CE marked and FDA approved and Manufactured by Odstock Medical Limited, Salisbury District Hospital, Salisbury, Wiltshire, SP2 8BJ, UK.

References 1. Compston, A, Coles, A. Multiple Sclerosis. Lancet 2002; 359: 1221–1331. 2. Lord, SE, Halligan, PW, Wade, DT. A comparison of two physiotherapy treatment approaches to improve walking in multiple sclerosis: a pilot randomized controlled study. Clin Rehabil 1998; 12: 477–486. 3. Jacobs, LD, Wende, KE, Brownscheidle, CM, Apatoff, B, Coyle, PK, Goodman, A, et al. A profile of multiple sclerosis: The New York State Multiple Sclerosis Consortium. Mult Scler 1999; 5: 369–376. 4. Kottink, AI, Hermens, HJ, Nene, AV, Tenniglo, MJ, van der Aa, H, Buschman, HP, et al. A randomized controlled trial of an implantable 2-channel peroneal nerve stimulator on walking speed and activity in poststroke hemiplegia. Arch Phys Med Rehabil 2007; 88: 971–978. 5. Liberson, WT, Holmquest, HJ, Scott, D, Dow, M. Functional electrotherapy; stimulation of the peroneal nerve synchronised with the swing phase of gait in hemiplegic patients. Arch Phys Med Rehabil 1961; 42: 101–105. 6. Burridge, JH, Taylor, PN, Hagan, SA, Wood, DE, Swain, ID. The effects of common peroneal stimulation on the effort and speed of walking; a randomised controlled trial

Multiple Sclerosis 2009; 15: 493–504

Downloaded from at Bournemouth University on November 8, 2010

MS FES and Exercise for dropped foot

7. 8. 9. 10.

11. 12.







19. 20.



23. 24. 25.

with chronic hemiplegic patients. Clin Rehabil 1997; 11: 201–210. Intercollegiate working party for stroke, (2000) National clinical guidelines for stroke. London, Royal College of Physicians ISBN 1860 161 2000. Carnstam, B, Larsson, LE, Prevec, TS. Improvement in gait following electrical stimulation. Scand J Rehabil Med 1977; 9: 7–13. Karsnia, A, Dillner, S, Ebefors, I, Lundmark, P. Why patients use or reject a peroneal muscle stimulator. Adv Extern Control Hum Extrem 1990; 10: 251–260. Taylor, PN, Burridge, JH, Dunkerley, AL, Woof, DE, Norton, J, Singleton, C, et al. Clinical use of the ODFS: its effect on speed and effort of walking. Arch Phys Med Rehabil 1999; 80: 1577–1583. Taylor, P. The use of electrical stimulation for correction of dropped foot in subjects with upper motor neurone lesions. ANCR 2002; 2: 16–18. Debolt, LS, McCubbin, J. The effects of home–based resistance exercise on balance, power and mobility in adults with multiple sclerosis. Arch Phys Med Rehabil 2004; 85: 290–297. Mostert, S, Kesselring, J. Effects of a short-term exercise training program on aerobic fitness, fatigue, health perception and activity level of subjects with multiple sclerosis. Mult Scler 2002; 8: 161–168. Romberg, A, Virtanen, A, Ruutiainen, J, Aunola, S, Karppi, SL, Vaara, M, et al. Effects of a 6-month exercise programme on patients with multiple sclerosis, a randomized study. Neurology 2004; 63: 2034–2038. Wiles, CM, Newcombe, RG, Fuller, KJ, Shaw, S, FurnivalDoran, J, Pickersgill, TP, et al. Controlled randomized cross-over trial of the effects of physiotherapy on mobility in chronic multiple sclerosis. J Neurol Neurosurg Psychiatry 2001; 70: 174–179. Patti, F, Ciancio, MR, Reggio, E, Lopes, R, Palermo, F, Cacopardo, M, et al. The impact of outpatient rehabilitation on quality of life in multiple sclerosis. J Neurol 2002; 249: 1027–1033. Solari, A, Filippini, G, Gasco, P, Colla, L, Salmaggi, A, Mantia, L, et al. Physical Rehabilitation has a positive effect on disability in multiple sclerosis patients. Neurology 1999; 52: 57–62. Lord, SE, Halligan, PW, Wade, DT. A comparison of two physiotherapy treatment approaches to improve walking in multiple sclerosis: a pilot randomized controlled study. Clin Rehabil 1998; 12: 477–486. Kurtze, JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983; 33: 1444–1452. Swain, I, Taylor, P. The clinical use of functional electrical stimulation in neurological rehabilitation. Horizons in medicine, vol. 16. Royal College of Physicians, London; 2004. p. 315–322. Bailey, M, Ratcliffe, C. Reliability of physiological cost index measurements in walking normal subjects using steady-state, non-steady state and post-exercise heart rate recording. Physiotherapy 1995; 81: 618–623. National Institute of Clinical Excellence. Management of multiple sclerosis in primary and secondary care. Clinical Guideline, vol. 8; Royal College of Physicians, London; 2003. Senn, S. Testing for baseline balance in clinical trials. Stat Med 1994; 13: 1715–1726. Goldman, MD, Marrie, RA, Cohen, JA. Evaluation of the six- minute walk in multiple sclerosis subjects and healthy controls. Mult Scler 2008; 14: 383–390. Graham, R, Smith, N, White, C. The reliability and validity of the physiological cost index in healthy subjects

26. 27. 28. 29.


while walking on 2 different tracks. Arch Phys Med Rehabil 2005; 86: 2041–2046. Danielsson, A, Willen, C, Sunnerhagen, K. Measurement of energy cost by the physiological cost index in walking after stroke. Arch Phys Med Rehabil 2007; 88: 1298–1303. Vaney, C, Blaurock, H, Gattlen, B, Meisels, C. Assessing mobility in multiple sclerosis using the rivermead mobility index and gait speed. Clin Rehabil 1996; 10: 216–226. Wiles, CM. Physiotherapy and related activities in multiple sclerosis. Mult Scler 2008; 14: 863–871. Stack, E. Physiotherapy - the ultimate placebo. Physiother Res Int 2006; 11: 127–128.

Appendix 1

Simple home exercises taught to the exercise group Small pieces of equipment readily available at home were used to promote joint alignment, provide resistance or challenge balance (e.g., rolled towels, theraband, balls, light hand weights). Activation of the deep stabilizing muscles around the trunk and pelvis to promote proximal alignment and stability was promoted throughout. Stretches (in supine, prone, long sitting, sitting or standing as appropriate) Hamstrings Quadriceps Hip flexors Hip abductors Ankle planti-flexors Exercises in crook lying Bridging (two legs/single leg) Trunk rotation Pelvic tilt Unilateral hip abduction Bilateral hip abduction Hip and knee flexion/extension Exercises in side lying Unilateral hip abduction Unilateral hip lateral rotation Unilateral hip abduction/lateral rotation Unilateral knee flexion/extension Exercises in prone Unilateral hip extension Unilateral/bilateral knee flexion Bilateral isometric gluteal contraction Unilateral/bilateral hip rotation Exercises in unsupported sitting Anterior/posterior pelvic tilt Trunk rotation Forward trunk flexion Unilateral trunk extension (reach out of base of support) Unilateral knee extension/flexion Multiple Sclerosis 2009; 15: 493–504

Downloaded from at Bournemouth University on November 8, 2010


CL Barrett et al. Balancing on one leg (single-leg stance) Sideways stepping Backwards stepping Balancing in step-stance Lateral reaching out of base of support

Unilateral hip abduction Bilateral hip abduction Exercises in standing Squats (two legs/single leg) Step-ups onto low step.

Multiple Sclerosis 2009; 15: 493–504

Downloaded from at Bournemouth University on November 8, 2010