ODFS Pace and Pace XL functional electrical stimulation devices for treating drop foot

Study component

Description

Objectives/hypotheses

To assess the effects of single channel common peroneal nerve stimulation (FES using the ODFS) on objective aspects of gait compared with exercise therapy for people with secondary progressive multiple sclerosis.

Study design

Randomised controlled trial.

Setting

Patients recruited from the waiting list for FES treatment at the FES clinic where the trial was conducted after being referred by either their GP or a consultant.

Inclusion/exclusion criteria

All patients were aged 18 years or over, with a diagnosis of SPMS and a rating of 4.5 to 6 on the EDSS. Patients also had to meet several criteria, including: predominantly unilateral dropped foot, a passive range of ankle dorsiflexion, response to common peroneal nerve stimulation and no previous use of FES. Patients were excluded if they had cognitive or psychiatric problems that affected their ability to understand or keep to treatment, or any other condition that may affect mobility or response to treatment.

Primary outcomes

The primary outcome variable was walking speed over 10 metres. Secondary outcome measures were PCI and distance walked in 3 minutes.

Statistical methods

Included: summary descriptive 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. ANCOVA was done on outcomes at 18 weeks, with baseline measures as covariates, after assessment of data conformity to ANCOVA assumptions.

Patients included

A total of 64 patients were recruited. The first 11 were used to pilot the trial protocol. Data produced from these pilot study patients were excluded in the analysis. Fifty three patients were recruited for the main trial.

Number analysed: FES=20, exercise=24.

Results

The comparator group showed a statistically significant increase in 10 metre walking speed relative to the intervention group, who showed no significant change in walking performance without stimulation. There were no differences between the intervention group and comparator group in distance walked. At each stage of the trial, the intervention group performed to a statistically significantly higher level with FES than without for the same outcome measures.

Conclusions

Exercise may provide a greater training effect on walking speed and endurance than FES for people with SPMS. FES using the ODFS may provide an orthotic benefit when the outcome is measured using the same parameters.

Abbreviations: ANCOVA, analysis of covariance test; EDSS, Kurtzke Expanded Disability Status Scale; FES, functional electrical stimulation; PCI, physiological cost index; SPMS, secondary progressive multiple sclerosis.

Intervention group: ODFS without stimulation

Comparator group: exercise

Analysis

Randomised

n=26

n=27

Efficacy

n=20

n=24

Primary outcomes

Mean walking speed over 10 metres (metres/second)

0.74±0.026

0.82±0.024

Difference: 0.081 (95% CI 0.01 to 0.15; p=0.028)

Walking speed over 10 metres (metres/second; mean±SD)

Week 0: 0.79±0.31

Week 6: 0.83±0.35

Week 12: 0.82±0.33

Week 18: 0.80±0.38

Change: p=0.592

Week 0: 0.68±0.28

Week 6: 0.72±0.27

Week 12: 0.72±0.27

Week 18: 0.77±0.29

Change: p=0.001

NR

Selected secondary outcomes

PCI recorded over 10 metres

(beats per minute/metres per minute, ANCOVA adjusted mean±SE)

0.69±0.041

0.70±0.037

Difference: 0.01 (95% CI −0.01 to 0.15; p=0.81)

PCI recorded over 10 metres

(beats per minute/metres per minute, mean±SD)

Week 0: 0.7±81.18

Week 6: 0.75±1.15

Week 12: 0.68±0.96

Week 18:0.74±1.12

Change: p=0.48

Week 0: 0.68±0.52

Week 6: 0.60±0.47

Week 12: 0.61±0.49

Week 18: 0.66±0.54

Change: p=0.53

NR

Distance walked in 3 minutes (metres, ANCOVA adjusted mean±SE)

124±8.5

112±7.9

Difference: 11 (95% CI −0.01 to 0.13; p=0.334)

Distance walked in 3 minutes (metres, mean±SD)

Week 0: n/a

Week 6: 122±56

Week 12: 124±51

Week 18: 125±55

Change: p=0.34

Week 0: 97±44

Week 6: 106±46

Week 12: 111±43

Week 18: 113±46

Change: p=0.005

NR

Intervention group: ODFS with stimulation (n=20)

Intervention group: ODFS without stimulation (n=20)

Analysis

Walking speed over 10 metres (metres/second, mean±SD)

Week 0: 0.79±0.31

Week 6: 0.83±0.35

Week 12: 0.82±0.33

Week 18: 0.80±0.35

Change: p=0.592

Week 0: 0.79±0.35

Week 6: 0.78±0.35

Week 12: 0.77±0.34

Week 18: 0.73±0.35

Change: p=0.155

Difference:

Week 0: p>0.50

Week 6: p=0.001

Week 12: p=0.001

Week 18: p=0.001

PCI over 10 metres (beats per minute/metres per minute, mean±SD)

Week 0: 0.78±1.18

Week 6: 0.75±1.15

Week 12: 0.68±0.96

Week 18: 0.74±1.12

Change: p=0.48

Week 0: 0.68±0.77

Week 6: 0.92±1.66

Week 12: 0.73±0.97

Week 18: 0.82±1.17

Change: p=0.35

Difference:

Week 0: p=0.35

Week 6: p=0.17

Week 12: p=0.08

Week 18: p=0.38

Distance walked in 3 minutes (metres, mean±SD)

Week 0: n/a

Week 6: 122±56

Week 12: 124±51

Week 18: 125±55

Change: p=0.34

Week 0: 99±42

Week 6: 112±51

Week 12: 111±53

Week 18: 112±50

Change: p=0.24

Difference:

Week 0: p=n/a

Week 6: p=0.010

Week 12: p=0.003

Week 18: p=0.004

Abbreviations: ANCOVA, analysis of covariance test; CI, confidence interval; FES, functional electrical stimulation; n/a, not applicable; NR, not reported; PCI, physiological cost index; SD, standard deviation; SE, standard error.

Study component

Description

Objectives/hypotheses

To determine if the Odstock Dropped Foot Stimulator improved activities of daily living for people with multiple sclerosis compared with physiotherapy exercises.

Study design

Randomised controlled trial.

Setting

Patients referred to the National Clinical FES Centre, UK.

Inclusion/exclusion criteria

All people in the study had secondary progressive MS, with dropped foot that impaired mobility. Other inclusion criteria included: no previous use of FES, rating of 4–6.5 on the Kurtzke Expanded Disability Status Scale, and an effective response to common peroneal nerve stimulation.

Primary outcomes

Outcome measures were COPM scores and number of falls.

Statistical methods

The Wilcoxon Signed Rank test and Mann–Whitney U‑test used to test for intragroup and intergroup statistically significant differences, respectively (indicated by a p‑value of less than 0.05).

Patients included

64 people with unilateral dropped foot due to secondary progressive multiple sclerosis.

Results of 53 research volunteers are reported. Intervention group: n=26, comparator group: n=27.

Results

Improvements in performance and satisfaction scores were greater in the intervention group than the comparator group (p<0.05). The median number of falls over the study period was 5 in the intervention group and 18 in the comparator group (p=0.036).

Conclusions

The study shows that people with multiple sclerosis using the ODFS increased their COPM performance and satisfaction scores of their identified problems with activities of daily living more than those in a comparator group, who had physiotherapy exercises. The ODFS users also had fewer falls.

Abbreviations: COPM, Canadian Occupational Performance Measure; FES, functional electrical stimulation; MS, multiple sclerosis; ODFS, Odstock Dropped Foot Stimulator.

Intervention group: FES

Comparator group: Exercise

Analysis

Randomised

n=32

n=32

Efficacy

n=26

n=27

Primary outcomes

COPM Performance score

Median (IQR)

Month 0: 3.5 (2.7 to 4.0)

Month 3: 4.8 (4.2 to 5.6)

Change: 1.1 (0.1 to 2.0); p=0.0002

Month 0: 3.4 (3.0 to 4.2)

Month 3: 3.8 (3.2 to 5.0)

Change: 0.0 (0.0 to 0.9); p=0.0553

Between‑group comparison

Month 0: p=0.574

Month 3: p=0.089

Change: p=0.038

COPM Satisfaction score

Median (IQR)

Month 0: 2.2 (1.4 to 3)

Month 3:4.0 (2.9 to 5.5)

Change: 1.7 (0.3 to 2.7); p=0.0001

Month 0: 2.6 (1.8 to 3.0)

Month 3: 2.4 (1.6 to 4.0)

Change: 0.0 (0.0 to 1.0); p=0.0437

Between‑group comparison

Month 0: p=0.515

Month 3: p=0.027

Change: p=0.007

Selected secondary outcomes

Falls (median number over study period)

5

18

Between‑group comparison

p=0.036

COPM results for tripping

Median improvement (IQR)

Performance:

3.5^a (1.25 to 3.5)

Satisfaction:

4.5^ab (3.25 to 5.75)

Performance:

0 (0.0 to 0.0)

Satisfaction

0.0 (0.0 to 0.0)

Between‑groups (difference in improvement score for satisfaction)

4.5; p<0.05

COPM results for walking a distance

Median improvement (IQR)

Performance: 1.0 (0.75 to 3.0)

Satisfaction: 5.5^ab (4.0 to 7.0)

Performance: 0.5 (0.0 to 1.75)

Satisfaction: 3.0^a (0.0 to 2.75)

Between groups (difference in improvement score for satisfaction)

2.5, p<0.05

Abbreviations: COPM, Canadian Occupational Performance Measure; IQR, interquartile range.

^aStatistically significant within‑group difference (Wilcoxon signed rank test p<0.05).

^bStatistically significant between‑group difference (Mann–Whitney U‑test p<0.05).

Study component

Description

Objectives/hypotheses

To compare the motor relearning effect of a surface PNS (the ODFS) compared with usual care on lower limb motor impairment, activity limitation, and quality of life among chronic stroke survivors.

Study design

Single‑blinded randomised controlled trial.

Setting

Stroke rehabilitation outpatient program within a multihospital academic medical centre.

Inclusion/exclusion criteria

Patients were recruited from a stroke rehabilitation outpatient programme, with inclusion criteria of age at least 18 years, at least 12 weeks post‑stroke with unilateral hemiparesis, and ankle dorsiflexion strength of 4/5 or less on the Medical Research Council scale. A large number of exclusion criteria were also applied, including lower extremity oedema, skin breakdown, serious cardiac arrhythmias, pacemakers or other implanted electronic systems, pregnancy, or uncontrolled seizure disorder.

Primary outcomes

Lower extremity portion of the FM Assessment (motor impairment), the mEFAP measured without a device (functional ambulation), and the SSQOL scale.

Statistical methods

Intention‑to‑treat analysis done. Baseline characteristics evaluated using Wilcoxon rank sum test or Fisher's exact test. All outcome measures modelled using a linear mixed effects approach. Differences between treatment groups tested at each discrete time point using the nonparametric Wilcoxon rank sum test with Bonferroni's correction.

Patients included

110 chronic stroke survivors (at least 12 weeks post‑stroke) with unilateral hemiparesis and dorsiflexion strength of 4/5 or less on the Medical Research Council scale. After dropout and exclusions: PNS=54, control=56.

Results

There was no significant treatment group main effect or treatment group by time interaction effect on FM, mEFAP, or SSQOL raw scores (p>0.05). When comparing average change scores from baseline (T1) to end of treatment (T2, 12 weeks), and at 12 weeks (T3) and 24 weeks (T4) after the end of treatment, significant differences were noted only for the mEFAP and SSQOL scores. The change in the average scores for both mEFAP and SSQOL happened between T1 and T2, followed by relative stability afterwards.

Conclusions

No evidence of a motor relearning effect in either the PNS or comparator groups, but both PNS and comparator groups showed significant improvements in functional mobility and quality of life during the treatment period, with the effect maintained at 6‑month follow‑up.

There were no significant treatment group differences at any time point.

Abbreviations: FM, Fugl‑Meyer; mEFAP, modified Emory functional ambulation profile; ODFS, Odstock Dropped Foot Stimulator; PNS, peroneal nerve stimulation; SSQOL, stroke specific quality of life.

Intervention group: PNS

Comparator group: UC

Analysis

Randomised

n=54

n=56

Efficacy

n=39

n=45

Primary outcome:

FM Assessment

NR (shown in plots)

NR (shown in plots)

No significant treatment group main effect (p=0.797) or treatment group by time interaction effect (p=0.321) on raw scores.

Time effect was significant (p=0.007) but no significant changes were seen from baseline to each time point (p>0.05).

Model parameter estimates for time effect during treatment: Difference 0.525 (95% CI −0.345 to 1.396; p=0.238).

Selected secondary outcomes

mEFAP

NR (shown in plots)

NR (shown in plots)

No significant treatment group main effect (p=0.968) or treatment group by time interaction effect (p>0.999) on raw scores.

Time effect was significant (p<0.001). Model parameter estimates of time effect at T2, T3 and T4 were all significantly lower than at baseline.

Model parameter estimates for time effect during treatment: Difference −13.864 (−21.256 to −6.473; p=<0.001).

SSQOL

NR (shown in plots)

NR (shown in plots)

No significant treatment group main effect (p=0.360) or treatment group by time interaction effect (p=0.627) on raw scores.

Time effect was significant (p<0.001).

Model parameter estimates of time effect at T2, T3 and T4 were all significantly higher than at baseline. Model parameter estimates for time effect during treatment: Difference 9.910 (3.724 to 16.096; p=0.002).

Abbreviations: CI, confidence interval; FM, Fugl‑Meyer; mEFAP, modified Emory functional ambulation profile; NR, not reported; PNS, peroneal nerve stimulation; SSQOL, stroke specific quality of life; t1, timepoint 1 (baseline); t2, timepoint 2 (end of device usage period); t3, timepoint 3 (12 weeks post‑treatment); t4, timepoint 4 (24 weeks post‑treatment); UC, usual care.

Study component

Description

Objectives/hypotheses

The objective of this study was to evaluate possible mechanisms for functional improvement and compare ambulation training with surface peroneal nerve stimulation (using the ODFS) compared with usual care via quantitative gait analysis.

Study design

Randomised controlled trial.

Setting

Patients were recruited from an academic medical centre.

Inclusion/exclusion criteria

Inclusion criteria age at least 18 years, at least 12 weeks post‑stroke with unilateral hemiparesis, ankle dorsiflexion strength of 4/5 or less on the Medical Research Council scale, and an effective response to common peroneal nerve stimulation. A large number of exclusion criteria were also applied, including lower extremity oedema, skin breakdown, serious cardiac arrhythmias, pacemakers or other implanted electronic systems; pregnancy, or uncontrolled seizure disorder.

Primary outcomes

Spatiotemporal (cadence, double support duration, stride length, walking speed), kinematic (peak hip flex swing, peak knee flex swing, peak ankle DF swing, ankle DF at IC, peak ankle abduction swing, peak ankle exterior rotation swing) and kinetic parameters (AP GRF, peak hip power pre‑swing, peak ankle power at push off) of gait. People were assessed while not wearing the ODFS.

Statistical methods

Intention‑to‑treat analysis done. A linear mixed‑effects model was used to evaluate the mean change in outcome measure within the treatment group. Bonferroni correction was used to control the family‑wise error rate and report the authors' adjusted p values.

Patients included

110 chronic stroke survivors (at least 12 weeks post‑stroke) with unilateral hemiparesis (PNS=54, UC=56).

Results

Cadence, stride length, walking speed, anterior‑posterior ground reaction force, peak hip power in pre‑swing, and peak ankle power at push‑off all significantly improved with respect to time. However peak ankle dorsiflexion in swing worsened. In general, the greatest change for all parameters happened during the treatment period. There were no statistically significant effects of the treatment on any of the spatiotemporal, kinematic, or kinetic parameters.

Conclusions

Gait training with PNS and UC was associated with improvements in peak hip power in pre‑swing and peak ankle power at push‑off, which may have resulted in improved cadence, stride length, and walking speed; but there were no differences between groups. Both groups also had a decrease in peak ankle dorsiflexion in swing, although the clinical implications of this finding are unclear.

Abbreviations: AP GRF, anterior‑posterior ground reaction force; DF, dorsiflexion; IC, initial contact; PNS, peroneal nerve stimulation; UC, usual care.

Intervention group

Comparator group

Analysis

Randomised

54

56

Efficacy

39

45

Assumed same as Sheffler 2013 study see table 7.

Primary outcome: Cadence (steps/minute, mean±SD)

t1: 65.0±22.0

t2: 67.4±21.5

t3: 69.3±26.4

t4: 70.8±26.8

t1: 66.7±22.7

t2: 72.6±22.6

t3: 72.0±23.2

t4: 73.7±22.7

Treatment group main effect p<0.001

Treatment group × time effect p>0.999

Time effect at t4 p<0.0001

Selected secondary outcomes

Double support (seconds, mean±SD)

t1: 1.14±0.76

t2: 1.05±0.75

t3: 1.08±0.87

t4: 1.02±0.77

t1: 1.15±0.87

t2: 0.92±0.72

t3: 0.97±0.74

t4: 0.91±0.68

Treatment group main effect p>0.999

Treatment group × time effect p>0.999

Stride length (metres, mean±SD)

t1: 0.62±0.24

t2: 0.68±0.27

t3: 0.71±0.28

t4: 0.71±0.28

t1: 0.67±0.24

t2: 0.74±0.22

t3: 0.72±0.24

t4: 0.73±0.2)

Treatment group main effect p=0.998

Treatment group × time effect p>0.999

Time effect at t4 p=0.0003

Walking speed (metres/second, mean±SD)

t1: 0.35±0.20

t2: 0.40±0.25

t3: 0.44±0.28

t4: 0.44±0.28

t1: 0.40±0.24

t2: 0.47±0.24

t3: 0.46±0.25

t4: 0.47±0.24

Treatment group main effect p>0.999

Treatment group × time effect p>0.999

Time effect at t4 p<0.0001

Peak hip flex swing (degrees, mean±SD)

t1: 32.5±8.0

t2: 33.2±11.3

t3: 32.8±9.6

t4: 34.5±10.1

t1: 35.3±9.3

t2: 35.1±9.4

t3: 36.2±9.2

t4: 35.4±8.6

Treatment group main effect p=0.350

Treatment group × time effect p>0.999

Peak knee flex swing (degrees, mean±SD)

t1: 25.1±13.1

t2: 27.6±14.1

t3: 27.8±15.9

t4: 29.6±16.4

t1: 30.5±15.1

t2: 31.3±16.6

t3: 31.6±17.0

t4: 31.3±15.5

Treatment group main effect p=>0.999

Treatment group × time effect p=>0.999

Peak ankle flex swing (degrees, mean±SD)

t1: 2.1±7.7

t2: 1.0±8.5

t3: 1.8±8.5

t4: 1.1±8.5

t1: 2.6±6.7

t2: −0.3±7.9

t3: −1.2±10.2

t4: 0.1±8.8

Treatment group main effect p=0.293

Treatment group × time effect p>0.999

Time effect at t4 p= 0.058

Ankle DF at IC (degree mean±SD)

t1: −7.3±9.4

t2: −6.9±9.3

t3: −5.9±8.0

t4: −6.8±8.4

t1: −5.9±8.1

t2: −8.2±10.2

t3: −9.3±11.4

t4: −6.7±9.4

Treatment group main effect p>0.999

Treatment group × time effect p=0.181

Peak ankle abduction swing, (degrees, mean±SD)

t1: −4.1±6.2

t2: −3.8±4.5

t3: −3.8±4.5

t4: −4.2±5.3

t1: −7.5±17.6

t2: −6.1±7.7

t3: −6.0±6.0

t4: −4.7±4.9

Treatment group main effect p=0.464

Treatment group × time effect p=0.999

Peak ankle external rotation swing (degrees, mean±SD)

t1: 0.1±18.0

t2: 1.6±19.3

t3: −0.8±16.5

t4: 1.2±20.3

t1: 4.4±20.1

t2: 3.4±17.8

t3: 4.7±19.9

t4: 0.8±17.9

Treatment group main effect p>0.999

Treatment group × time effect p>0.999

AP GRF (Nm, mean±SD)

t1: 0.51±0.28

t2: 0.60±0.29

t3: 0.64±0.38

t4: 0.74±0.56

t1: 0.55±0.30

t2: 0.69±0.33

t3: 0.64±0.27

t4: 0.67±0.31

Treatment group main effect p>0.999

Treatment group × time effect p>0.999

Time effect at t4 p=0.032

Peak hip power in pre‑swing (W/kg, mean±SD)

t1: 0.30±0.21

t2: 0.45±0.43

t3: 0.48±0.41

t4: 0.53±0.52

t1: 0.38±0.31

t2: 0.50±0.44

t3: 0.53±0.48

t4: 0.59±0.61

Treatment group main effect p=0.003

Treatment group × time effect p>0.999

Time effect at t4 p<0.0001

Peak ankle power at push‑off (W/kg mean±SD)

t1: 0.42±0.41

t2: 0.54±0.49

t3: 0.56±0.54

t4: 0.62±0.63

t1: 0.51±0.58

t2: 0.66±0.65

t3: 0.64±0.64

t4: 0.64±0.64

Treatment group main effect p>0.999

Treatment group × time effect p>0.999

Time effect at t4 p=0.003

Average time standing/day (minutes, mean±SD)

t1: 153.5±87.2

t2: 170.6±78.6

t3: 156.4±79.2

t4: 173.2±115.2

t1: 127.3±75.5

t2: 127.1±71.7

t3: 138.1±91.5

t4: 157.5±115.5

Time effect: F_3,151 1.05; p>0.999

Treatment group × time effect: F_3,151 0.57, p>0.999

Average time walking/day (minutes, mean±SD)

t1: 67.8±58.6

t2: 67.3±58.5

t3: 77.5±56.1

t4: 71.8±54.1

t1: 65.9±53.6

t2: 67.5±48.2

t3: 71.1±46.8

t4: 78.8±50.7

Time effect: F_3,148 0.54, p>0.999

Treatment group × time effect: F_3,148 1.13, p>0.999

Average number of steps/day (mean±SD)

t1: 3223±3134

t2: 3383±3470

t3: 3991±3397

t4: 3738±3211

t1: 3270±2947

t2: 3555±2951

t3: 3734±2820

t4: 4038±2848

Time effect: F_3,153 0.78, p>0.999

Treatment group × time effect: F_3,153 0.78, p>0.999

Abbreviations: AP GRF, anterior‑posterior ground reaction force; DF, dorsiflexion; F_{x, y}, degrees of freedom with _x representing the number of treatment arms minus 1 and _y representing the residual error; IC, initial contact; Nm, newton per kilogram; SD, standard deviation; t1, timepoint 1 (baseline); t2, timepoint 2 (end of device usage period); t3, timepoint 3 (12 weeks post‑treatment); t4, timepoint 4 (24 weeks post‑treatment); W, watt.

Study component

Description

Objectives/hypotheses

To measure the effect of the ODFS on the effort and speed of walking.

Intervention/comparator

Intervention: OFDS and physiotherapy

Comparator: Physiotherapy alone

Study design

Randomised controlled trial.

Setting

Specialist UK‑based FES centre

Inclusion/exclusion criteria

Inclusion criteria were: stroke causing a hemiplegia for at least 6 months; ability to stand from sitting without help; and walk a minimum of 50 metres independently before to stroke.

Exclusion criteria were: bilateral dropped foot, discomfort with stimulation; mental impairment; severe expressive and receptive dysphagia; unable to walk 10 metres; unable to elicit functional ankle dorsiflexion; unable to attend physiotherapy sessions; no improvement seen in walking with stimulation.

Primary outcomes

Changes in walking speed measured over 10 metres and effort of walking measured by PCI.

Statistical methods

Non‑parametric statistical tests were used. Wilcoxon signed ranks test was used to determine significance of differences within each group and Mann–Whitney test for differences between groups. Categorical variables were correlated using the chi‑square test and relationships between quantitative variables using Spearman's p.

Patients included

32 hemiplegic patients who had a single stroke at least 6 months before the start of the trial.

Results

Mean increase in walking speed between the start and end of the trial was 20.5% in the FES group with stimulation, and 5.2% in the control group. There was a reduction of 24.9% in PCI in the FES group with stimulation, and 1% in the control group. There was no improvement in these parameters in the FES group when the stimulator was not used. When the FES group was compared with the control group, there was a statistically significant improvement in walking speed at week 12, but not at week 4, and no statistically significant improvements in PCI at any time point.

Conclusions

Walking was statistically significantly improved when the ODFS was worn but no 'carry‑over' (training effect) was seen. Physiotherapy alone did not improve walking.

Abbreviations: FES, functional electrical stimulation; m, metres; ODFS, Odstock Dropped Foot Stimulator; PCI, physiological cost index.

FES

Control

Analysis

Randomised

n=16

n=16

n=32

Efficacy

n=15

n=16

n=31 (1 patient from the FES group excluded from analysis)

Primary outcome: 10‑m walking speed (metres/second, mean±SD)

FES without stimulation

Week 0: 0.64±0.46

Week 4: 0.62±0.41

Week 12: 0.63±0.39

FES with stimulation:

Week 0: 0.68±0.49

Week 4: 0.75±0.51

Week 12: 0.77±0.43

Control

Week 0: 0.48±0.25

Week 4: 0.51±0.25

Week 12: 0.51±0.27

FES without stimulation vs control

Week 0: p=0.318 (95% CI −0.130 to 0.360)

Week 4: p=0.621 (95% CI −0.170 to 0.280)

Week 12: p=0.407 (95% CI −0.120

to 0.320)

FES with stimulation vs control

Week 0: p=0.228 (95% CI −0.380, to 0.100)

Week 4: p=0.221 (95% CI −0.400, to 0.900)

Week 12: p=0.044 (95% CI −0.460 to 0.001)^a

Changes in 10‑m walking speed (metres/second, mean±SD)

FES without stimulation

Week 4 vs Week 0: 0.06±0.21

Week 12 vs Week 0: 0.01±0.15

Control

Week 4 vs Week 0: 0.03±0.08

Week 12 vs Week 0: 0.03±0.10

FES without stimulation

Week 4 vs Week 0: % change 1.2, p=0.551 (95% CI −0.130 to 0.045)

Week 12 vs Week 0: % change 0.12, p=1.00 (95% CI −0.085 to 0.075)

Control

Week 4 vs Week 0: % change 5.21, p=0.255 (95% CI −0.0200 to 0.065)

Week 12 vs Week 0: % change 5.21, p=0.379 (95% CI −0.030 to 0.080)

FES group with stimulation week 0 vs FES group without stimulation week 0

Mean=0.04±0.11, % change 6.1, p=0.187 (95% CI −0.020 to 0.100)

FES group with stimulation week 4 vs FES group without stimulation week 0

Mean=0.11±0.24, % change 17.61, p=0.205 (95% CI −0.020 to 0.195)

FES group with stimulation week 12 vs FES group without stimulation week 0

Mean=0.13±0.13, % change 20.50, p=0.004 (95% CI 0.060 to 0.210)

Selected secondary outcomes

PCI (beats per minute/metre per minute, mean±SD)

FES without stimulation

Week 0: 0.80±0.74

Week 4: 0.71±0.71

Week 12: 0.76±0.64

FES with stimulation:

Week 0: 0.59±0.49

Week 4: 0.61±0.67

Week 12: 0.54±0.56

Control

Week 0: 1.03±0.67

Week 4: 0.98±0.74

Week 12: 1.00±0.69

FES without stimulation vs control

Week 0: p=0.220 (95% CI −0.141 to 0.052)

Week 4: p=0.327 (95% CI −0.179 to 0.620)

Week 12: p=0.127 (95% CI −0.080 to 0.770)

FES with stimulation vs control

Week 0: p=0.057 (95% CI −0.010 to 0.640)

Week 4: p=0.127 (95% CI −0.080 to 0.770)

Week 12: p 0.083 (95% CI −0.020 to 0.749)

Changes in PCI (beats per minute/metreper minute, mean±SD)

FES without stimulation

Week 4 vs Week 0: −0.09±0.05

Week 12 vs Week 0: −0.04±0.12

Control

Week 4 vs Week 0:

−0.05±0.27

Week 12 vs Week 0:

−0.03±0.20

FES without stimulation

Week 4 vs Week 0: % change −5.62, p=0.335 (95% CI −0.085 to 0.255)

Week 12 vs Week 0: % change −11.83, p=0.670 (95% CI −0.105 to 0.2245)

Control

Week 4 vs Week 0: % change −5.80, p=0.148 (95% CI −0.065 to 0.160)

Week 12 vs Week 0: % change −3.90, p=0.469 (95% CI −0.060 to 0.120)

FES with stimulation week 0 vs FES without stimulation week 0

Mean −0.20±0.32, % change −20.68, p=0.010 (95% CI 0.040 to 0.325)

FES with stimulation week 4 vs FES without stimulation week 0

Mean −0.18±0.44, % change −20.49, p=0.094 (95% CI −0.025 to 0.380)

FES with stimulation week 12 vs FES without stimulation week 0

Mean −0.26±0.37, % change −24.87, p=0.008 (95% CI 0.040 to 0.430)

Abbreviations: FES, functional electrical stimulation; m/s, metres per second; m, metre; min, minute; PCI, physiological cost index; vs, versus.

^aResults may be reported incorrectly because the p value is <0.05 but the 95% CI spans zero.

Study component

Description

Objectives/hypotheses

To determine the effectiveness of FES on drop foot in patients with MS, using data from standard clinical practice.

Study design

Case series with data collected between 2008 and 2013.

Setting

UK‑based specialist FES centre.

Inclusion/exclusion criteria

Study exclusion criteria were: inability to walk 10 metres with the assistance of a walking aid; poorly controlled epilepsy; or fixed skeletal deformities. Other precautions included recent injury, fracture, or surgery; major skin conditions; and cancerous tissue near the site of stimulation.

Primary outcomes

Clinically meaningful changes in functional walking category and walking speed over 10 metres.

Statistical methods

A repeated‑measures analysis of variance was used to analyse the walking speed over 10 metres. Planned comparisons were done using paired t tests.

Patients included

166 people with MS.

Results

An increase in walking speed was found to be significant both initially and after 20 weeks with a substantial clinically meaningful change. No significant training effect was found. Functional walking category lasted or improved in 95% of people responding to treatment.

Conclusions

FES is a well‑accepted intervention that aids clinically meaningful changes in walking speed, leading to a preserved or increased functional walking category.

Abbreviations: FES; functional electrical stimulation. MS; multiple sclerosis.

Unassisted walk

FES walk

Analysis

Efficacy

‑

‑

n=153

Primary outcome: walking speed over 10 metres

(m/s)

‑

‑

Main effect for stimulation vs no stimulation (F_1,152 91.88, p<0.001). Interaction effect between stimulation over time (F_1,152 9.79, p=0.002).

Selected secondary outcomes

Initial orthotic effect

(metres/second, mean±SD)

0.72±0.33

0.79±0.31

Mean difference: 0.07± 0.11^a, p=0.001 (95% CI 0.05 to 0.08)

Continuing orthotic effect

(metres/second, mean±SD)

0.72±0.35

0.82±0.34

Mean difference: 0.11±0.16^b, p=0.001 (95% CI 0.08 to 0.13)

Total orthotic effect

(metres/second, mean±SD)

0.72±0.33

0.82±0.34

Mean difference: 0.10±0.22^b, p=0.001 (95% CI 0.07 to 0.14)

Training effect (metres/second, mean±SD)

0.72±0.33

0.72±0.35

Mean difference: 0.00±0.26, p=0.53 (95% CI ‑0.04 to 0.03)

Abbreviations: CI, confidence interval; m/s, metres/second; SD, standard deviation; vs, versus.

^aMinimal meaningful change.

^bSubstantial meaningful change.