Peroneal stimulator: Evaluation for the correction of spastic drop foot in hemiplegia

19

Peroneal Stimulator: Evaluation for the Correction of Spastic Drop Foot in Hemiplegia Malcolm H. Granat, PhD, Douglas J. Maxwell, BSc, Amanda C.B. Ferguson, PhD, Kennedy R. Lees, MB ChB, Joseph C. Barbenel, PhD ABSTRACT. Granat MH, Maxwell DJ, Ferguson ACB, Lees KR, Barbenel JC. Peroneal stimulator: evaluation for the correction of spastic drop foot in hemiplegia. Arch Phys Med Rehabil 1996; 77:19-24.

Objective: The objective of this study was to assess the orthotic and therapeutic value of the peroneal stimulator (PS) for adult hemiplegic patients. Design: This was a 2-period crossover study lasting 11 weeks. After recruitment each patient had a 4-week control period followed by a 4-week treatment period. The patients were assessed before the control period, after the control period, and after the treatment period. Setting: Nineteen patients were recruited from physiotherapy departments in the Glasgow area; 2 subsequently dropped out before the treatment period. Patients: All patients had hemiplegia as a result of a cerebrovascular accident (CVA) and were greater than 3 months but less than 36 months post-CVA. Average time since stroke was 7 months. Main Outcome Measures: The patients' gait was assessed over smooth linoleum, carpet, and uneven ground. Gait was evaluated using a switch-based portable system. Outcome measures were the temporal gait parameters of speed, symmetry, heel strike, and foot inversion during stance. The gait evaluation was repeated on 5 separate days at each assessment. The Barthel Index was applied at each session. Results: There was a significant orthotic improvement in inversion on all surfaces and for symmetry on linoleum (ANOVA, p = .05). There was no significant improvement in patients' gait when not using the PS. There was a significant improvement in the Barthel Index over the treatment period (Wilcoxon, p = .05). Conclusion: Use of the PS as an orthotic device late in the rehabilitation program would be appropriate for a selected subpopulation of patients.

© 1996 by the American Congress of Rehabilitation Medicine and the American Academy of Physical Medicine and Rehabilitation 'EMIPARESIS due to stroke often results in spastic drop.foot, ie, the loss of ability to dorsiflex the foot on the affected side. One approach to the management of spastic dropfoot is the prescription of an ankle foot orthosis ( A F t ) , which

H

From the BioengineeringUnit, WolfsonCentre,Universityof Strathclyde(Drs. Granat,Ferguson,Barbenel,and Mr. Maxwell)and the AcuteStroke Unit,University of Glasgow Department of Medicine and Therapeutics, Gardiner Institute, Western Infirmary(Dr. Lees), GlasgowScotland. Submittedfor publicationJanuary 17, 1995. Accepted in revisedform July 10, 1995. Supported by the Scottish Home and Health Department(K/CRED/4/C164). No commercial party having a direct financial interest in the results of the research supporting this article has or will confer a benefitupon the authors or upon any organizationwith which the authors are associated. Reprint requests to Malcolm H. Granat, PhD, BioengineeringUnit, Wolfson Centre, Universityof Strathclyde, 106 Rottenrow, GlasgowG4 ONW, Scotland. © 1996by the AmericanCongressof RehabilitationMedicineand the American Academy of Physical Medicineand Rehabilitation 0003-9993/96/7701-337853.00/0

holds the foot in a neutral position to prevent it from dragging during the swing phase of gait] An alternative approach is active stimulation of the dorsiflexors. Electrical stimulation for the correction of spastic drop foot in hemiplegia was first applied by Liberson and coworkers in 1961. 2 Surface electrodes were applied over the peroneal nerve at the head of the fibula, and a stimulator, worn around the waist, was controlled by a switch in the heel of the shoe of the affected limb. When the patient lifted the heel to take a step, the stimulator was activated. Stimulation was stopped when the heel contacted the ground. This system, the peroneal stimulator (PS), produces dorsifiexion and eversion of the foot during the swing phase of gait. The PS has an advantage over an A F t in that it provides active gait correction and its action can be tailored to the individual patient's requirements. The PS has been used as a rehabilitation aid and has been further developed by Yugoslavian and American researchers. 3-5 It can be difficult to obtain an electrode placement that adequately controls both eversion and inversion using a single channel system. Some workers have therefore investigated the use of more complex multichannel systems to gain better control of gait patterns. 6'7 These multichannel systems, however, result in increased complexity of function, burdening the patient with a system that is unsightly, difficult to fine tune, and time-consuming to apply. There have been limited clinical evaluations of the PS with large patient populations. 8-~°Much of this work has been based on clinical reports from doctors and therapists involved in the patients' rehabilitation] ° The purpose of this study was to perform an objective assessment of the orthotic value (benefit gained when using the PS) and therapeutic value (benefit gained as a result of having used the PS) of the PS to adult hemiplegic patients. A control and treatment period was incorporated into the experimental design to determine the therapeutic benefits.

METHODS This was a 2-period crossover study lasting 11 weeks. After being recruited, each patient had a 1-month control period followed by a 1-month treatment period. The patients were assessed before and after the control period, and after the treatment period. The experimental protocol was approved by the West Ethical Committee of Greater Glasgow Health Board and all patients gave written informed consent to participation.

Selection Criteria Patients were referred to the investigators from physiotherapy departments in the Glasgow area. These patients were assessed for their suitability for the study against the following criteria: 1. Hemiplegia as a result of a cerebrovascular accident (CVA). 2. Spastic drop foot. 3. Sufficient communication skills to learn the use of the PS. 4. Ability to walk independently. 5. Ability to operate the PS (with help of carer where necessary).

Arch Phys Med Rehabi! Vol 77, January 1996

PERONEAL STIMULATOR IN HEMIPLEGIC GAIT, Granat

20

Table 1: Patient Details At Start of Program Subject

Age

Sex

Hemiparesis (Dominant Side)

A B C D F G H I J K L M N O P Q R S T

51 65 57 43 53 55 65 47 54 53 65 44 48 66 65 56 62 75 74

M F M M M M M M M M M M M M M M M F F

R(R) L(R) R(R) R(R) R(R) L(R) R(R) R(R) R(R) R(R) L(L) L(R) R(R) L(R) L(R) R(R) L(R) R(R) R(R)

Orthosis (Dependent pc/N])

Walking Aid (Dependent [Y/N])

AFO (N) 0 AFO (N) Air Splint (N) AFO (N) AFO (Y) Heel Spring (N) Knee cage, AFO (N) AFO (N) AFO (N) AFO (Y) AFO(N) 0 0 0 AFO (Y) 0 0 O

0 Stick Stick 0 Stick 0 0 g Stick Stick Stick 0 0 0 0 0 Stick Stick Stick

(Y) (Y) (Y)

(Y) (Y) (Y)

(N) (Y) (Y)

Months Post-CVA

Barthel Score

Classification

6 9 5 4 24 3 6 19 2 3 8 6 5 6 14 3 3 14 10

94 86 43 68 72 78 62 69 58 51 68 78 70 100 100 80 69 57 78

PACI Hemorrhage LACI PACI PACI LACI PACI Hemorrhage LACI LACI LACI Hemorrhage LACI LACI PACI PACI PACI LACI LACI

A value of 0 under Orthosis or Walking Aid indicates that the patient did not use one. The Barthel score was the initial value when the patient entered the program. Abbreviations: PACI, partial anterior circulation infarct; LACI, lacunar infarct.

6. Greater than three months but less than 36 months postCVA (to avoid the period of greatest rate of spontaneous recovery and to have a study population who were either undergoing rehabilitation or were not far removed in time from the rehabilitation period). Nineteen patients were recruited to the study; 2 subsequently dropped out before the treatment period. Another patient could not be included in the statistical analysis because he was unable to walk without wearing either an AFO or the PS. The average age of the remaining 16 patients was 56 years; they were on average 7 months post-CVA. Subjects were classified according to: affected side, dependency on an orthosis, dependency on a walking aid, disability as measured by the Modified Barthel Index, 1~ and the site of the lesion j2 (table 1).

Experimental Protocol Each patient was on the program for 11 weeks. After selection the patients were tested during week 1 (session 1 tests). During the control period (weeks 2 to 5) the patients received their normal physiotherapy. In week 6 the patients were tested for a second time (session 2 tests). During this week they were instructed in the use of the PS. During the treatment period (weeks 7 to 10) the patients used the PS as their orthosis. They were encouraged to put on the PS in the morning and use it throughout the day. During this period weekly visits were made to the patients' homes to ensure correct application and attend to any problems. During week 11 the patients were tested for a third time (session 3 tests), and their gait was assessed both with and without the PS. They were tested first without the PS to prevent any carry over effects from influencing the results. The stimulator used for the project was a single channel drop foot stimulator? The stimulation output had rectangular wave monophasic pulses of 300/~s duration, adjustable frequency between 25 to 50Hz and user-controlled pulse amplitude setting between 0V and 60V. Stimulation was applied to the common peroneal nerve in the area between the popliteal fossa and the head of the fibula such that an effective dorsiflexing movement was produced at the ankle. The PS used by the patients had the capacity to record the length of time for which stimulation was delivered. Additionally, as a safety feature, the stimulator was active for 4 seconds only, after which it was retriggered by the heel switch. Consequently, it was not possible for the PS to inadvertently continue to stimulate. This also ensured that the

Arch Phys Med Rehabil Vol 77, January 1996

time recorded by the stimulator was a true representation of the level of activity of the user. After the 11-week testing period the patients were allowed to retain the PS for the remainder of the project. The following assessment tests were conducted before and after the control period and after the treatment period.

Gait Assessment The patient was required to walk along a straight path over 3 different surfaces: smooth linoleum (linoleum); thick pile carpet (carpet); and uneven ground simulated by lightweight carpet with battens, 1.0cm high and 2cm wide, placed at irregular intervals across the direction of progression (uneven ground). For each surface the patient walked along a defined straight path, with each walk designated as one run. On each day there were 5 recorded runs for linoleum and 2 for the other surfaces. The length of the recorded walk path was either 6 or 10 meters, according the patient's ability, and a 1.5-meter lead-in and nanout of the test walk path was used to eliminate variability arising Table 2: Definition of Gait Parameters Heel strike: This quantified heel contact at the start of the stance phase. This was calculated as the average time for which the heel switch only was ON at beginning of each stance period. This was expressed as a percentage of total foot contact time. Inversion: This quantified the medio-lateral stability of the foot during the stance phase. It was calculated by taking the difference of ON times between the 5th metatarsal and the 1st metatarsal switches, This was expressed as a percentage of continuous metatarsal-floor contact time. A value of +100% indicated that there was no contact under the 1st metatarsal, a value of 0% indicated equal contact times, and a value of -100% indicated that there was no contact under the 5th metatarsal. Symmetry: This quantified the swing symmetry of the gait pattern. This was calculated as the ratio of the swing time of affected leg to swing time of unaffected leg. Swing time was defined as the time between last floor contact of the foot to the first floor contact of the same foot. A normal value would be 100%, a value greater than 100% is indicative of a longer swing time for the affected leg. Speed: Speed was defined as the walkpath length divided by the time taken to traverse the walkpath.

21

PERONEAL S T I M U L A T O R IN HEMIPLEGiC GAIT, Granat

Table 3: Overall Results for All Parameters and Surfaces

Speed (m/min) Linoleum Session 1 Session 2 With PS W i t h o u t PS Uneven g r o u n d Session 1 Session 2 With PS W i t h o u t PS Carpet Session 1 Session 2 With PS W i t h o u t PS

Inversion (% difference between the ON and OFF times of the 5th and 1st metatarsal switches)

Symmetry (% swing time of unaffected leg to affected leg)

Heel strike (% of stance)

52.39 56.24 55.78 56.96

± 33.50 _+ 37.60 ± 35.12 ~ 34.91

207.30 199.52 172.09 159.78

_+ 93.70 ± 80.93 ± 55.65 ± 44.48

-42,60 -36.34 -41.88 -17.33

± 56.18 ± 45.86 ± 39.67 _+ 31.90

6.98 6.27 5.43 6.36

_+ 7.90 ± 9.41 ± 7.07 _+ 13.96

49.27 56.43 53.08 55.80

± ± £ +

34.62 37.76 34.46 34.11

222.28 201.63 179.67 208.04

± ± ± ±

105.93 87.35 64.67 114.73

-40.52 -30.29 -36.68 -11,22

± 57.61 ± 49.60 ± 41.27 _+ 42.87

6.23 7.92 6,91 7.63

± ± ± ±

59.33 66.25 58.93 63.03

± 34.69 ± 37.63 _+ 36.24 ± 35.22

201.65 176.69 167.39 154.60

± ± ± ±

128,35 71.76 48.58 53.55

-40.39 -28.91 -46.32 -9,60

± 55.15 _+ 40.37 ± 36.01 ± 33.67

5.59 6.82 4.95 5.07

_+ 7.26 _+ 10.08 ± 6.07 _+ 16.00

5.56 11.80 8.18 14.67

Values are m e a n ± SD.

from gait changes during the initial and final stages of the walking assessment. This assessment was repeated on 5 separate days in each testing session giving sufficient data to characterize changes in the gait parameters.J2 The temporal parameters of gait were measured using instrumented insoles and a portable, body-worn data logging system.13 The instrumented insoles consisted of 4 switches: one under the heel, another under the big toe and 2 under the heads of the first and fifth metatarsals. The insoles were located inside the shoe with the subject's sock limiting any mechanical abrasion of the foot by the insole. No measurements were taken when the subject was wearing an AFO. The activity of the switches was recorded at a sampling frequency of 50Hz, and processing of this data after the test generated the gait parameters. Gait parameters calculated were restricted to those that could be affected, directly or indirectly, by the action of PS. The calculated parameters affected indirectly were speed and symmetry of gait. Those affected directly were associated with the foot-floor contact pattern and were a direct result of the action of the spastic drop foot, ie, heel contact and inversion of the foot. The gait parameters are defined in table 2. The Barthel Index ~ was applied once during each testing session. For session 3 the patients assumed the use of the PS in their replies. Although upper limb function heavily influences the Barthel Index, it was applied to establish the level of disability of the study population and to detect any substantial improvement or deterioration in that level. It proved to be more difficult than we had anticipated to find patients who met our selection criteria. Just over 2% of screened patients from the acute stroke unit met the selection criteria. In some patients established gait problems masked the correction

of the PS on the patient's gait. Other problems encountered were some subjects' poor communication skills and an inability to operate the PS effectively. Eight patients could apply the PS independently; however, in some cases its effectiveness was compromised by poor electrode placement. There was a day-today variation of the electrode site at which the most appropriate movement was obtained; this was possibly due to changes in electrode and skin condition.

Data Analysis Analysis of Variance (ANOVA) was used to determine the overall orthotic benefit to the group and the individual orthotic benefit for all gait parameters. To evaluate the therapeutic benefit for the whole group, a paired t test was performed, comparing the change over the period between session 1 and session 2 and the change over the period between session 2 and session 3, on all gait parameters. The nonparametric Wilcoxon Test was used to compare the Barthel Index scores (related samples with a non-normal distribution). A significance level of .05 was used in all the tests. RESULTS

Orthotic Benefit The analysis was performed for the group as a whole and for each individual patient. The orthotic benefit to the group was calculated by comparing the results from session 3 when using or not using the PS (table 3). There was a significant improvement in inversion on all surfaces and for symmetry on linoleum. The magnitude of change in the inversion parameter was more B

1.5

pul "~ O j::

Fig 1. Speed (m/sec) using the PS plotted against speed ( m / sec) w i t h o u t using the PS for the three different surfaces: (A) linoleum, {B) uneven ground, and (C) carpet. The dashed line indicates no change of speed using the PS. Points above this line indicate an increase in speed when using the PS.

/

/

/

/

/ /

//

1.0 /

p

/

/O/~

/

/

o ~/

90"

/

/

/

o

t

o /

9"

/

O"

¢° 0

I [

i

I

0.5

1.0

1.5

0

I

I

I

I

I

I

0.5

1.0

1.5

0.5

1.0

1.5

Without Orthosis

Arch Phys Med Rehabil Vol 77, J a n u a r y 1996

22

PERONEAL S T I M U L A T O R IN HEMIPLEGIC GAIT, G r a n a t

A 40O /

/

/

C

/ /1

/1

/

/I

200

0 I /

/

4•

qo,o

/

o

/

I

0

0

/ /

O

//~)

o/2 o

i

/

//

//

0 4...

/

/

//

'

200

|

I

400

0

/i /

o¢ 200

/

I

I

400

0

°

/

/

,, /

,

,

0

400

0

Without Orthosis

improved this significantly. Above the 10% value the changes in heel strike were variable with subject and surface. There were statistically significant improvements with 8 subjects on linoleum, 4 on carpet and 5 on uneven ground. There was also a statistically significant deterioration in 1 subject on linoleum. The most pronounced effect was on medio-lateral stability of the foot as measured by the inversion parameter (fig 4). There were statistically significant improvements with 9 subjects on linoleum, 6 on carpet and 6 on uneven ground. There were no statistically significant deteriorations. Many of the positive changes in this parameter were large. On uneven ground, however, the values for inversion (without the PS) were considerably lower than on linoleum and carpet. This could be explained possibly by subjects being more careful about their foot placement over the uneven ground surface. This was also reflected in speed, which was lowest on uneven ground (table 2). In 10 of the 17 patients there was a clear orthotic benefit, and 11 subjects used it regularly after the treatment period.

than 50% on all 3 surfaces. At the 0.1 level there were improvements in speed on carpet and symmetry on uneven ground. The individual results for all the patients are shown in figures 1 through 4. For any given parameter, the closer its value is to a normal value the smaller is the potential for improvement. Therefore, the greatest improvements could be expected in parameters with values well below the normal range. Some patients, for example, may have a high walking speed but have a poor gait pattern with low stability; there would be no expected improvement in walking speed but improvements in other parameters would indicate a safer gait pattern. There was a range of original walking speeds (approx. 0.1 to 0.6m/sec) that improved when subjects were using the PS. Walking speeds above 0.6m/sec offer little potential for detectable improvement. At speeds below 0.1m/sec mobility was so impaired that the alterations made to gait by the PS were minor. There were statistically significant improvements with 5 subjects walking on linoleum, 3 walking on carpet, and 5 walking on uneven ground. There were also a statistically significant deteriorations in 2 subjects walking on linoleum. Two patients showed improvements in symmetry, heel strike, and inversion without significant improvement in speed. Both had speeds in the region of .60m/sec, suggesting that speed alone is not adequate to describe hemiplegic gait. There was a trend towards reducing swing symmetry to 100% (fig 2), with only 2 subjects on each surface lying above the reference line. This reduction was larger for higher values of swing symmetry. There were statistically significant improvements with 8 subjects walking on linoleum, 4 walking on carpet and 2 walking on uneven ground. There were also a statistically significant deteriorations in 2 subjects on the linoleum, 1 subject on the carpet and 2 subjects walking on uneven ground. The improvements in heel strike (fig 3) were mainly in subjects whose heel strike was less than 10% of stance. In these subjects, heel strike without the PS was poor or totally absent, and the direct action of the PS, raising the foot during swing,

Therapeutic Benefit To determine if the gait parameters improved without using the PS after the treatment period the change in the mean value of the measured parameter over the treatment period was compared with the change over the control period. The rate of increase of speed over the control period was greater than over the treatment period (paired t test, p < .05). This indicated that during the treatment period the rate of improvement as measured by speed decreased. Eleven of the 17 patients started the program within 6 months of their CVA and would be expected to have some recovery during the control period. As their gait improved their potential for further improvement would reduce. This result does not exclude a therapeutic benefit of the PS but any such benefit is clearly limited. No other changes were found. At the start of the program 2 patients (K and P) were dependent on AFOs because of medio-lateral ankle instability. Both used the PS successfully for the entire treatment period and

B

A 3O

C O

// //

/

/I

O

20

O¢/~

~,

O

/

O

O/ •/¢O

Arch Phys Med Rehabil V o l 77, J a n u a r y 1996

0

/

/

0

0

/ /

/

I

I

I

10

20

30

/

/

/6

/

0 /

P

I

i

I

10

20

30

Without Orthosis

/

•/

/

No 0

/

/

/

/ /

//

/

10

/

//O

//

(3

•

/

/

//

/

Fig 3. Heel strike (% stance) using the PS p l o t t e d against heel strike (% stance) without using t h e PS for the three different surfaces: (A) linoleum, (B) uneven ground, and (C) carpet. The dashed line indicates where heel strike with and without the PS are the same. Points above the line indicate an oncreased duration in heel strike when using the PS.

Fig 2. Swing symmetry (%} using the PS plotted against swing symmetry (%) without using the PS for the three different surfaces: (A) linoleum, (B) u n e v e n ground, and (C) carpet, The dashed line indicates where swing symmetry with and without the PS are the same. Points below the dashed line and above the x-axis indicate an improvement in swing symmetry w h e n using t h e PS.

0

/

i

¢0

/

l

I

I

10

20

30

PERONEAL STIMULATOR IN HEMIPLEGIC GAIT, Granat

A

B / O

Fig 4. Inversion (% midstance) using the PS plotted against inversion (% midstance) without using the PS for the three different surfaces: (A) linoleum, (B) uneven ground, and (C) carpet. The dashed line indicates where inversion with and without the PS are the same. Points below the dashed line and above the xaxis indicate an improvement in inversion when using the PS.

O e~

/

,

o

,

/

I

I

I

50

100

-100

-50

_

/

/

//

o,

/

•

100

e 0

0

0

0

0

0

75

3000•

m

-50

o 1500- 25

A

O

N

B

M

D

P

-

/

/

/

O o

/"

to 50

i 100

-so q~"

/

/

/

/

// /

/

I

I

50

100

O//v

/

O

t

• OO

-ld0

/

/

/

¢

/

/

,o

Without Orthosis

DISCUSSION Multichannel stimulation, which has been used with some success as a training aid in our clinic7'15 to improve a patient's gait, is not a practical 'home-therapy' or an orthotic option for this group of patients. Other studies have shown improvements when using the PS and in these studies the patients received significant therapy input and long-term training within a clinical or laboratory settingJ ° More recent reports have concentrated

•

O

O0

,,0 G

I

Stimulator Usage Stimulator usage was calculated as the average time the stimulator was active each day. The average daily walking distance was estimated using this information. Figure 5 shows the estimated average daily walking distances for each patient and their Barthel Index after the treatment period.

•

/

o,a; •"QID

both preferred it. Patient P, when assessed for inclusion into this program, could walk without an AFO, but before starting the initial tests he was prescribed an AFO. When patient P returned for the session 2 test he was unable to walk without the AFO, but after using the PS for 4 weeks he recovered sufficient voluntary control to walk with an unsupported ankle. There was a significant improvement in the Barthel Index over the treatment period but not over the control period. The patients who showed an improvement in the Barthel Index reported that with the PS they were better able to cope with stairs and uneven ground; it may have been that walking over uneven ground during the tests had a training effect. It appeared that patients whose Barthel Index was within a certain range improved when using the PS. Patients with a Barthel Index of 100 (patients A, N, and O) showed little improvement when using the PS, nor did patient C, who had a very low score. A Barthel Index of 100 is the maximum possible, though this does not mean that they were without residual disability.

45007

/ //

/

oo,~

o oo,'o

C

o,

/

//

-soo°,"

-1~

:5

/

/

/

//

O

0

23

F

H

L

Q

J

E

I

G

K

C

Patient

Fig 5. Distance walked, as indicated by the stimulator timer, and final Barthel Index for all patients. The patients are ordered by decreasing Barthel Score.

on improvements in the technologyJ 6'~7 This study evaluated the PS as an orthotic device that would be prescribed for community use. The stimulator chosen was one that could now be used at home by this group of patients, and the amount of usage of the PS was a combination of patients' expectations and abilities. The average increase in walking speed was quite small, although other parameters, notably inversion and heel strike, showed greater improvements. The action of the PS is to dorsiflex and evert the foot during swing phase. This direct action improved inversion and heel strike during stance phase and enhanced the quality of the gait. These improvements may increase the safety of the gait and the patient's confidence irrespective of effects on walking speed. Foot-Floor Contact Without the PS, the sole of the foot was flat during stance in all patients. Those patients who had observable inversion during early stance (patients B, D, F, L, and P) or toe strike during early stance (patient I) achieved improvement in heel strike and/or inversion (improved foot pronation) using the PS. All these patients showed improvements in other parameters. In 4 of the remaining patients, where early inversion or toe strike was not observable, improvements were also made in heel strike and inversion (patients A, H, J, and Q). In 2 patients (H and J), there were also improvements in other gait parameters such as speed and symmetry. Lack of further improvement in patient Q may be explained by this patient's rapid fatigue and in A by an adequately fast gait. Improvement in the quality of foot-floor contact during stance, therefore, appeared to be the key to improving other gait parameters. Figure 5 shows the average distance walked using the PS and the Barthel Index for each patient. The patients with a daily walking distance of less than one kilometer must rarely venture outside. The Barthel Index is not a good indicator of the distance a patient may walk in a day. That distance is more likely to be affected by the lifestyle and expectations of the patient. Consequently, patients with similar Barthel Indices show very different levels of activity. The benefit gained from the use of the stimulator is in improved walking, but if the patient's lifestyle does not normally include extended periods of walking (eg, going outdoors), then the PS may not be the most appropriate orthosis for that individual. The action of the PS in lifting the fore foot during swing will be of most importance where increased ground clearance is required (eg, stairs and rough ground). The importance of this action of the PS was borne out in the improvements found in traversing rough round and the patient comments on improvements in climbing stairs. To apply the PS successfully the patient had to understand the use of the PS (to set the correct amplitude level and position

Arch Phys Med Rehabil Vol 77, January 1996

24

PERONEAL STIMULATOR IN HEMIPLEGIC GAIT, Granat

the electrodes appropriately to give the required foot movement) and apply the PS (with the help of a carer if necessary). The PS is an active orthosis and therefore requires far more active involvement from the patient than a passive orthosis such as an AFO. Also, the design of the PS makes it difficult to apply and to connect the electrodes with one hand and, therefore, many patients required daily help. Given the population from which these subjects were selected, it is our opinion that the PS, applied in the late stage of rehabilitation, would be applicable to a few patients (2%) who present with CVA. The PS would therefore have a role in the management of spastic drop foot in a selected patient population, in particular in patients with medio-lateral instability of the foot and reduced ground clearance in swing leading to forefoot contact. Further work would be required to determine the benefit of the PS early in the patients' rehabilitation program. This would allow the patient to become familiar with the PS while still an inpatient and be incorporated into his gait reeducation program before gait problems become established. At this early stage it may have much wider applicability as a training aid, although this project did not address this application. SUMMARY 1. For the patient group there was an overall orthotic benefit in the medio-lateral stability of foot but no significant change in the other measured parameters. There were demonstrable orthotic benefits for certain individual patients, however. 2. Patients who would benefit from the PS are those in w h o m the foot-floor contact pattern can be improved (particularly where early inversion and toe strike are obvious problems) and whose levels of independence and mobility are within certain limits that can be defined by scoring systems and objective measurement. 3. Most patients could cope with the daily use of the PS but some required assistance in applying the PS. Acknowledgments: The financial support of the Scottish Home and Health Department (K/CRED/4/C164) is gratefully acknowledged.

References 1. Lehmann JF, Condon SM, Price R, deLateur BJ. Gait abnormalities in hemiplegia: their correction by ankle-foot othoses. Arch Phys Med Rehabil 1987;68:763-71. 2. Liberson WT, Holmquest HJ, Scott D, Dow M. Functional electrotherapy: stimulation of the peroneal nerve synchronized with the swing phase of the gait of hemiplegic patients. Arch Phys Med Rehabil 1961;42:101-5.

Arch Phys Med Rehabil Vol 77, January 1996

3. Kljajic M, Bajd T, Stanic U. Quantitive gait evaluation of hemiplegic patients using electrical stimulation orthoses. IEEE Trans Biomed Eng 1975;22:438-41. 4. Kralj A, Acimovic R, Stanic U. Enhancement of hemplegic patient rehabilitation by means of functional electrical stimulation. Prosthet Orthot Int 1993; 17:104-7. 5. McNeal DR. Peripheral nerve stimulation, superficial and implanted. In: Fields WS, Leavitt, editors. Neural organization and its relevance to prosthetics. New York: Intercontinental Medical Book Corporation, 1973:77-99. 6. Malezic M, Bogataj U, Gros N, Decman I, Vrtacnik P, Kljajic M, et al. Application of a programmable dual-channel adaptive electrical stimulation system for the control and analysis of gait. J Rehab Rcs Dev 1992;29:41-53. 7. Stanic U, Acimovic-Janezic R, Gros N, Trnkoczy A, Bajd T, Kljajic M. Multichannel electrical stimulation for correction of hemiplegic gait. Scand J Rehab Med 1978; 10:75-92. 8. Kljajic M, Malezic M, Acimovic R, Vavken E, Stanic U, Pangrsic B, et al. Gait evaluation in hemiparetic patients using subcutaneous peroneal electrical stimulation. Scand J Rehab Med 1992;24:1216. 9. Kljajic M, Bajd T, Stanic U. Quantitive gait evaluation of hemiplegic patients using electrical stimulation orthoses. IEEE Trans Biomed Eng 1975;22:438-41. 10. Merletti R, Andina A, Galante M, Furlan I. Clinical experience of electronic peroneal stimulators in 50 hemiparetic patients. Scand J Rehab Med 1979; 11:111-21. 11. Granger CV, Albrecht GL, Hamilton BB. Outcome of comprehensive medical rehabilitation: measurement by PULSES profile and the Barthel Index. Arch Phys Med Rehabil 1979;60:145-54. 12. Hill KD, Goldie PA, Baker PA, Greenwood KM. Retest reliability of the temporal and distance characteristics of hemiplegic gait using a footswitch system. Arch Phys Med Rehabil 1994;75:577-83. 13. Granat MH, Maxwell DM, Bosch CJ, Ferguson ACB, Lees KR, Barbenel JC. A body-worn gait analysis system for evaluating herniplegic gait. Med Eng Physics 1995; 17:390-4. 14. Bamford J, Sandercock P, Dennis M, Bum J, Warlow C. Classification and natural history of clinically identifiable subtypes of cerebral infarction, Lancet 1991;337:1521-6. 15. Malezic M, Kljajic M, AcimovicJanezic, Gros N, Krajnik J, Stanic U. Therapeutic effects of multisite electrics stimulation in motor disabled patients. Arch Phys Med Rehabil 1987;68:553-60. 16. Rozman J, AcimovicJanezic R, Tekavcic I, Kljajic M, Trlep M. Implantable stimulator for selective stimulation of the common peroneal nerve: A preliminary report. J Med Eng Tech 1994; 18:4753. 17. Wllemsen ATM, Bloemhof F, Boom HBK. Automatic stance-swing phase detection from the accelerometer data for peroneal nerve stimulation. IEEE Trans Biomed Eng 1990;37:1201-8.

Supplier a. FES Medical Electronics, Flat 12, 355 Argyle Street, Glasgow GZ 8LT, UK.