A PILOT STUDY TO
INVESTIGATE THE COMBINED USE OF BOTULINUM NEUROTOXIN A (BoNTA) AND FUNCTIONAL
ELECTRICAL STIMULATION (FES), WITH
PHYSIOTHERAPY, IN THE TREATMENT OF SPASTIC DROPPED FOOT IN
SUBACUTE STROKE.
C.A.
Johnson*, A.M. Tromans**, D.E. Wood*, D. O’Keefe*, D. Buhrs*, I.D. Swain*, J.H. Burridge***,
Department of Medical Physics and Biomedical Engineering* and Duke
of Cornwall Spinal Injuries Unit**, Salisbury District Hospital; University of Southampton***.
SUMMARY
The purpose of the study was to investigate the feasibility of
combining BoNTA and Functional Electrical Stimulation (FES) using a single
channel drop-foot stimulator, with physiotherapy, in the treatment of spastic
dropped foot in sub-acute stroke. Results of the study will be used to design a
randomised, controlled trial (RCT). Subjects were randomised into two groups;
both groups received physiotherapy, the treatment group also received BoNTA and
FES.
Patients recruited were within one year following stroke, and a baseline period
(weeks -4 to 0) was incorporated to allow change in rate of recovery to be
measured. BoNTA injections were given and FES treatment commenced at week 0, and follow-up assessments were made
at weeks 2, 4, 8 and 12. The primary outcome measures were walking speed and
the Physiological Cost Index (PCI) of gait. Measurements were taken of
stimulated and non-stimulated walking. Analysis of the results of the primary
outcome measures suggest that physiotherapy alone and the combined use of BoNTA
and FES with physiotherapy has a beneficial effect on walking speed and PCI.
The study has provided evidence of an additional effect of BoNTA and FES, and with sufficient
information on effect variability to enable sample size calculations for a
follow up RCT, to estimate the effect size with a useful degree of precision.
STATE OF THE ART
There
are approximately 100,000 hospital admissions for stroke in the United Kingdom annually; 20% are
fatal, 20% make a full recovery and 60% make a partial recovery but have a
restricted lifestyle /1/. Physiotherapists working with these patients aim to maximize the recovery of movement and
function through the re-education of postural control and normal activity.
There are no statistics on how many patients might benefit from the combined use
of FES and BoNTA, but 20% /2/ would be a conservative estimate of the
proportion of patients suffering from a spastic drop-foot. This group of
patients experience difficulty when walking because they are unable to
effectively dorsiflex their ankle during the swing phase of walking. The
problem arises partly through an inability to activate the anterior tibial muscles, and partly through restraint from the calf muscles.
In some, but not all cases, the restraint is due mainly to inappropriate
(spastic) calf activity; in others the mechanical resistance offered by the
calf is the predominant factor. Knuttson and Richards /3/ described premature calf activity in one third
of the study group of patients with spastic hemiparesis. EMG activity in the
triceps surae began and peaked earlier in the gait cycle, compared with normal;
peak activity was lower in triceps surae, and in tibialis anterior.
FES applied to the common peroneal nerve, and timed to the swing phase of
walking has been shown to be effective in patients with poor active ankle
dorsiflexion and calf spasticity /4/5/. FES is thought to have an inhibitory effect on antagonist activity, and
thus reduce calf spasticity; it may also improve the mechanical component of
calf restraint, through stretching of the triceps surae muscle group. BoNTA has
also been shown to reduce inappropriate calf activity /6/ particularly in
patients with premature calf activation during walking. The presence of spastic
dropped-foot frequently causes stroke patients to adopt an abnormal gait
pattern that may exacerbate extensor spasticity. Once established, such
abnormal patterns are difficult to correct. By applying these combined
treatments during the recovery phase, the long-term outcome of the mobility of
stroke patients may be improved.
MATERIALS AND METHODS
19
patients were recruited from the stroke services of 4 NHS Trusts in England. All participants fulfilled the following selection criteria: 1) single stroke of vascular origin with
hemiplegia during the previous 12 months; 2) inability to achieve a heel
strike, correctable by FES; 3) between 3 and 6 inclusive on the Hauser
Ambulation Index; 4) an increased calf stretch response on examination; 5)
premature calf activation during gait identified by EMG. Patients were excluded
who had: 1) any additional medical condition that might influence response to
treatment; 2) prescribed anti-spastic medication; 3) prescribed medication that
may have influenced heart rate measurements, either at the time of recruitment
or during the 4 weeks prior to recruitment; 4) severe psychological problems;
5) any patient unable to give informed consent. Participants
were randomised into 2 groups; each selected a sealed envelope containing an
explanatory letter allocating them into one of the groups. The control group
(CG) received physiotherapy, the treatment group (TG) received physiotherapy
and the combined use of FES and BoNTA. Both groups continued with their physiotherapy programme
throughout the study period. In-patients received a minimum of 3 sessions per
week, out-patients 2 sessions per week.
Subjects in the TG received BoNTA
injections at week 0. In most cases the medial and lateral heads of
gastrocnemius were each injected with 200 units of Dysport, and tibialis
posterior with 400 units. This dose was modified for less spastic muscles or
smaller patients. Injections were given under electromyographical (EMG)
guidance. FES was used by the treatment group patients immediately following
BoNTA injections. An Odstock Dropped Foot Stimulator mark III (ODFS III) was
set up by the Research Physiotherapist, following the standard protocol, with a
full explanation and a copy of the user instruction manual. The device
delivered a train of electrical impulses, at a frequency of 40 Hz, via surface
electrodes. The common peroneal nerve was stimulated, and timed to the person’s
walking speed by a footswitch. Patients were asked to use the system on a daily
basis to assist their walking, and for most of each day. Cyclic exercise
stimulation was used in patients with higher levels of calf spasticity, and
where mechanical constraint of the triceps surae was present, to maximize
patient’s response to the use of the ODFSIII.
Assessments for
the study were carried out by the Research Physiotherapist in the patient’s
local Physiotherapy Department. The primary outcome measures were
non-stimulated (CG and TG) and stimulated (TG) walking speed and PCI of gait.
Secondary outcome measures included: a neurological examination, consisting of
tendon and cutaneous reflexes in the lower limb, the Modified Ashworth Scale to
measure spasticity in the quadriceps, hamstrings, dorsiflexors and
plantarflexors, and scoring of clonus; the Rivermead Motor Assessment scale;
the SF 36 Health Survey; a semi-structured interview; kinetic EMG for tibialis
anterior and the lateral head of gastrocnemius during walking.
RESULTS
Descriptive statistics based on the
demographic data of the study participants are tabulated below.
|
|
Control
|
Treatment
|
|
Number of subjects
|
9
|
10
|
|
Age mean (SD)
|
59.33 (12.46)
|
58.2 (12.72)
|
|
Age range
|
44 – 78
|
41 – 78
|
|
Sex
|
4 female, 5 male
|
2 female, 8 male
|
|
Side of hemiplegia
|
5 right, 4 left
|
7 right, 3 left
|
|
Time since stroke 0 - 6 months
|
3
|
6
|
|
Time since stroke 6 – 12 months
|
6
|
4
|
Results of the primary outcome measures
of walking speed and PCI are presented. Statistical methods included graphical
plots of individual patient response curves, non-parametric test statistics
(Mann-Whitney and Wilcoxon paired-sample tests), and a summary measures
approach to assessing between group differences using regressions of median
walking speed and PCI, and median change in walking speed and PCI on time.
A similar pattern of response for
walking speeds and PCI was seen under both non-stimulated and stimulated test
conditions. In the non-stimulated walking tests, a significant upward trend in
median walking speed for both the CG (P=0.020) and TG (P=0.004) was seen, the
trend lines being significantly different in location (P=0.040). Comparison of
stimulated walking tests show a significant upward trend in median walking
speed for both the CG (P=0.020) and the TG (P=0.042), the trend lines being
significantly different in location (P=0.009). In non-stimulated walking tests
a significant downward trend was demonstrated in median PCI for the treatment
group (P=0.007), but not for the control group (P=0.292), the trend lines being
significantly different in location (P=0.038). In stimulated walking tests a
significant downward trend in median PCI was evident for the treatment group
(P=0.020), but not for the control group (P=0.292), the trend lines being
significantly different in location (P=0.016).
Regression of median change in walking
speed was significant for the CG (P=0.043) and TG (P=0.025); the trend lines
being in a sufficiently different location (P=0.0194). The following graph
illustrates the changes from baseline Week 0, in median walking speed (m/s) for
both CG and TG (stimulated).
The following box and whisker plots
demonstrate the changes from baseline Week 0 in median PCI (heart beats/m) for
both CG and TG (stimulated). No trends emerged as significant.
DISCUSSION
This pilot study has provided evidence
of a real treatment effect. It has also given sufficient information on the
variability of the outcome measure to facilitate sample size calculations for a
subsequent study, to clarify the magnitude of the treatment effect with a
meaningful degree of precision. Given a baseline median walking speed of 0.2
m/s, with a pooled standard deviation for median change in walking speed of
0.106 m/s, a test significance level of 5% (assuming the use of a
non-parametric test statistic) and test power set at 80%, 470 subjects per
group would be required to detect a 10% difference (0.02 m/s) between the
treatment and control groups (as defined above) in respect of median change in
walking speed from baseline (at 12 weeks), 120 subjects per group to detect a
20% difference (0.04 m/s), 55 subjects per group to detect a 30% difference
(0.06 m/s) and 32 subjects per group to detect a 40% difference (0.08 m/s).
REFERENCES
/1/ Royal College of Physicians (1989) ‘Stroke: towards better
management. Summary & recommendations of a report of the
Royal College of Physicians.’ Journal
of the Royal College of
Physicians 24, 1, 15-17
/2/ Gracanin F. (1984) ‘Functional electrical
stimulation in external control of motor activity and movements of paralysed
extremities.’ International Rehabilitation Medicine 6, 25–30
/3/ Knutsson E., Richards C., (1979) ‘Different
types of disturbed motor control in gait of hemiparetic patients.’ Brain 102, 405-430
/4/ Burridge J.H., Taylor P.N., Hagan S.A., Wood D.E., Swain, I.D.,
(1997) ‘The effects of common peroneal stimulation on the effort and speed of
walking’, Clinical Rehabilitation 11, 201-210.
/5/ Burridge J.H., Taylor P.N., Wood D.E., McLellan D.E., (1998) ‘The
effect of different patterns of abnormal muscle activation during walking on
the response to common peroneal stimulation’, presented at the Human
Performance Meeting, University
of East London, July.
/6/ Hesse S., Ct at (1996) ‘Ankle
muscle activity before and after botulinum toxin therapy for lower limb
extensor spasticity in chronic hemiparetic patients.’ Stroke 27, 3.
ACKNOWLEDGEMENTS
This work was supported by a grant from
Ipsen Ltd, including the supply of BoNTA (Dysport), and Salisbury Area
Healthcare NHS Trust. The author would also like to thank the team at the
Department of Medical Physics and Biomedical Engineering, Salisbury District Hospital; the
research team, including consultants, pharmacists and physiotherapists in the
participating NHS Trusts; the Research and Development Support Unit at Salisbury District Hospital; and
especially the participants themselves and their families.
AUTHOR’S ADDRESS
CA Johnson, Clinical Research
Physiotherapist, Department of Medical Physics and Biomedical Engineering,
Salisbury District Hospital, Wiltshire, SP2 8BJ. Telephone: 01722 336262 x
4060, Fax 01722 425263, e-mail calj@mpbe-sdh.demon.co.uk