THE COMPUSTIM 10B IN STROKE: CONTROL ALGORITHMS
AND
PATIENT SELECTION CRITERIA
Wright PA1,
Burridge JH1, Ewins DJ2, Mann GE1, McLellan DL3,
Swain ID1, Taylor PN1 and Wood DE1
1Address
for correspondence: Department of Medical Physics and Biomedical Engineering,
2Biomedical
Engineering Group,
3Rehabilitation
Research Unit, University of Southampton, Southampton General Hospital,
Tremona Rd, Southampton, SO16 6YD, UK
ABSTRACT
The aim of this study was:
·
To identify patients most likely to benefit from two
channel lower limb stimulation with the microcontroller based Compustim 10B
stimulator
·
To investigate the orthotic and re-education effects
of using a second channel of stimulation
·
To develop a database of control algorithms for
walking
Fourteen subjects (mean age
63.2, SD 6.7 yrs, 9 Male 5 Female) were recruited who had been treated with the
single channel Odstock Dropped Foot Stimulator (ODFS) for a minimum of 6
months. All subjects followed a CONTROL
— TREATMENT — CONTROL study design where the TREATMENT was two channel
stimulation with the Compustim 10B and CONTROL was single channel common peroneal
stimulation with the Compustim 10B. Walking speed and Physiological Cost Index
were measured at approximately four-weekly intervals throughout the study. Gait
analysis was performed at weeks 0, 12, 24 and 36.
A comparison of clinical
observations and the results of gait analysis will be presented.
KEYWORDS:
INTRODUCTION
Dropped foot following
stroke may be corrected by electrical stimulation of the common peroneal nerve
during the swing phase of the impaired leg. Significant improvements in walking
speed and effort of walking have been demonstrated in a randomised controlled
trial of the single channel ODFS [1, 2].
However, some patients still
experience problems through weakness or inadequate control of further muscle
groups in the leg. Preliminary work showed that stimulation of an additional
lower limb muscle group would improve gait; in particular stimulation of the
hamstrings muscle to improve knee flexion during the swing phase of gait and of
calf muscle during push-off.
STIMULATOR
A microcontroller based two
channel stimulator has been developed [3]. Footswitches underneath the heel and
first metatarsal head enable the microcontroller to identify specific stages of
the gait cycle. Stimulator algorithms, which included initiation and
termination conditions, frequency, and pulse width, were programmed via a user
friendly LabVIEW interface [4]. The interface features a real time display of
footswitch condition and stimulator output level, which is useful in setting up
subjects.
STUDY DESIGN
The study design for all
subjects was CONTROL (12 weeks, single channel Compustim) — TREATMENT (12
weeks, two channel Compustim) — CONTROL (12 weeks, single channel Compustim).
The second muscle group for
stimulation was selected by clinical observation of each subject’s gait.
Subjects observed to have poor 'push off' were selected for calf stimulation.
Three different algorithms were used for hamstring stimulation:
·
on initial floor contact to control knee
hyperextension
·
during the terminal third of stance phase to limit
knee hyperextension at this point and, in addition, to assist forward movement
of the knee over the forefoot and thus allow push-off
·
during swing phase to enhance knee flexion (probably
through quadriceps inhibition)
Two patients were observed to have been suitable for either calf or
hamstrings stimulation.
Walking speed and
Physiological Cost Index were measured at approximately four-weekly intervals
throughout the study. Gait analysis was performed at weeks 0, 12, 24 and 36.
This data will be used to investigate the orthotic and re-education effects of
using the second channel of stimulation.
Subjects were given a
questionnaire to complete on their use of the single channel ODFS before the
start of the trial. Subjects were asked to complete a second questionnaire at
the conclusion of the trial on their use of the Compustim 10B.
RESULTS
Group
receiving Calf Stimulation at Week 24 (n=6)
|
||
|
No stimulation |
1 Channel stimulation |
2 Channel stimulation |
|
+9% (0.03) |
+16% (0.00) |
+25% (0.00) |
Group
receiving Hamstrings Stimulation at Week 24 (n=6)
|
||
|
No stimulation |
1 Channel stimulation |
2 Channel stimulation |
|
+12% (0.13) |
+15% (0.08) |
+20% (0.03) |
Table 1. Percentage increase in walking speed at end of TREATMENT period (Week 24) compared to walking speed measured with no stimulation at beginning of CONTROL period (Week 0). P-values of paired t-tests are shown in brackets.
Results are not yet complete, as not all patients have finished the study at the time of writing. LabVIEW routines are being set up to analyse gait and present data.
Measurements of walking
speed and physiological cost index for walking with no stimulation at the
beginning of the first CONTROL period have been compared with those for no
stimulation, common peroneal stimulation, and two channel stimulation at the
end of the TREATMENT period. There were no significant changes in the effort of
walking both with and without stimulation for both the group receiving
hamstrings and for the group receiving calf stimulation (p>0.10, two tailed
t-test). There were significant increases in speed observed in both groups
(table 1). This confirms our preliminary findings that walking with two channel
stimulation is more efficient than walking without stimulation or with just common
peroneal stimulation. Graphs are shown of the percentage average increase in
walking speed (compared to walking speed with no stimulation at week 0)
throughout the study for the group with hamstrings stimulation and for the
group with calf stimulation (figures 1 & 2).
Nine patients have completed
end of trial questionnaires to date. All respondents reported that they were
able to walk further, faster and with less effort with the Compustim 10B.
Quality of walking, and confidence and independence in walking, were also
improved in all cases. All patients chose to continue using the Compustim 10B
at the end of the trial. The main reason for their choice was that the
Compustim 10B was more effective in improving walking. The stimulator was
considered too large to wear comfortably. Development in minimising the
stimulator to a more comfortable size is now underway in response to this
observation. Initial problems relating to footswitch reliability were addressed
by improving construction techniques and detection algorithms.
Figure 1. Percentage average increase
in walking speed throughout study for patients who received a second channel of
hamstrings stimulation (n=6). 95% confidence intervals for each data point are
between 5 and 15% but are not displayed for clarity.
Figure 2. Percentage average increase in walking speed throughout study for patients who received a second channel of calf stimulation (n=6). 95% confidence intervals for each data point are between 5 and 15% but are not displayed for clarity.
DISCUSSION
The group with hamstrings
stimulation showed a different pattern of improvement from the group who
received calf stimulation. There was evidence of a re-education effect of two
channel stimulation in the group receiving hamstrings stimulation but this was
less obvious in the group receiving calf stimulation. Both groups demonstrated
that walking was improved by single channel stimulation and further improved by
either second channel of stimulation. Both groups also showed an orthotic
benefit from the second channel of stimulation which increased with the length
of time they had used the second channel.
Gait analysis comprised
measurement of joint positions and velocities, force plate measurement of
ground reaction forces, and EMG studies. Indices are being calculated from this
data, e.g. coactivation indices from calf and Tibialis Anterior EMGs, muscle
contraction envelopes and timings, and peak forces of Ground Reaction Vectors.
These enable investigation of both the orthotic and possible re-education
effects of using the second channel of stimulation. Results of gait analysis
will be used to consider the importance of gait analysis in the context of
selecting and setting up a second channel of stimulation. A list of indicators
for selecting patients and setting up patients will then be developed for
future use of the Compustim 10B in clinical and research environments.
ACKNOWLEDGEMENT
This work is funded by
Action Research. We would like to thank Malcolm Burnett for technical support
during gait analysis sessions.
REFERENCES
1. Burridge JH, Taylor PN,
Hagan SA,
2. Burridge JH,
Taylor PN, Hagan SA,
3. Michael P. Developments in Surface Electrical Orthoses
for the Re-education of Hemiplegic Gait. PhD
Thesis,
4.
Ewins DJ, Wright PA, Burridge JH, Mann GE, Swain ID, Taylor PN and