LEG POWERED
PARAPLEGIC CYCLING SYSTEM USING SURFACE FUNCTIONAL
ELECTRICAL STIMULATION
T.A.
Perkins*, N. de N. Donaldson*, R. Fitzwater**, G.F.Phillips**, D.E. Wood***
*UCL Department of Medical Physics, London,
UK
**Royal National Orthopaedic Hospital,
Stanmore, UK
***Salisbury District Hospital, Salisbury,
UK
SUMMARY
We
have established a simple system for Spinal Cord Injured (SCI) patients to
achieve leg propelled cycling using surface Functional Electrical Stimulation
(FES). We initially intended our partial lesion T11/12 male paraplegic subject
to achieve outdoor leg powered cycle rides that would be useful for both
exercise and mobility. We hoped this added function would encourage the patient
to greater effort, particularly as more intense exercise /1/ has reportedly
given some recovery in bone density as well as the other health benefits
expected from more moderate exercise levels /2/. Our first surface stimulation
volunteer has repeatedly cycled about 12km at a time (both indoors and
outside). After 20 months of FES exercise (at almost 150 minutes per week), bone
density in his more paralysed left leg had increased in several sites, that in
the tibial tuberosity by 44+/-2%. Other health benefits, including increased
muscle bulk and voluntary function /3/ were also noted.
We
are currently extending this English pilot study to complete-lesion
paraplegics. So far, two out of our three complete paraplegic subjects have
already attained kilometre plus leg powered cycling capability.
STATE OF THE ART
Leg
propelled cycling was first demonstrated in the 1980s, both with surface
stimulation /4/ and with implanted electrodes /5/. Since then, most FES cycling
exercise for SCI patients has been on static ergometers, often in a clinic or
laboratory for nominally three 30 minute sessions per week /6/, /7/ and /8/.
With compliance around 70%, this amounts to less than 10 minutes per day.
Following the demonstration of kilometre plus capability in two complete lesion
paraplegics in 1998, /9/ and /10/, we felt it was time FES cycling exercise
moved out of the laboratory and into patients’ everyday lives, where we
expected more frequent and prolonged exercise would be possible. A reliable
system for home use was therefore required.
MATERIAL AND METHODS
Equipment
Commercially
available recumbent tricycles (Trice and Windcheater) were chosen because of
their inherent stability and low seating pressure. Switches were attached to
the handlebars for starting and stopping the cycling program. A 7-bit shaft
encoder was driven synchronously by the cranks. Purpose-built foot-plates were
attached to the pedals to help secure the feet as shown in Fig 1. For the
paralysed limbs, Ankle Foot Orthoses (AFOs) were attached to the foot-plates to
hold the knees close to the sagittal plane. To control pedalling speed, a
throttle potentiometer adjusted stimulation strength (linearly interpolated
between threshold and maximal). A resistance trainer stand was provided so that
the tricycle was turned into a static ergometer for indoor cycle exercise at
home. For outdoor use, the cycle was removed from the trainer, keeping the
vital mechanical set-up for the patient exactly the same. An 8 channel surface
stimulator, giving pulses of up to 500ms
and 150mA, was used.
B E D A F
G G C
Key to Fig 1:-
A: Trainer. B: Electrode shorts. C: Tights to retain cables &
electrodes. D: Stimulator. E: Throttle & switches. F: Shaft encoder. G: foot-plates.
Fig
1. Partial lesion subject’s Windcheater tricycle at home.
Methods
Seat to pedal distance was adjusted for
minimum passive resistance to crank rotation without excessive knee extension.
Two of our complete lesion patients required special cranks for this to be
possible. For our partial lesion subject, quadriceps, hamstring and gluteal
muscles both sides were activated, with gastrocnemius and tibialis anterior on
the more paralysed left side. The complete paraplegics used gastrocnemius both
sides, leaving out tibialis anterior. In each case, stimulation frequency was
20Hz.
Initially,
static pedal force readings were taken with a spring balance for all likely
crank angles, as shown in Fig 2, in order to determine favourable crank angles
for stimulating each muscle group.
Fig
2. Maximal stimulation pedal forces.
Crank angle from Left hip
most
flexed Top Dead Centre (LTDC).
For
all our surface stimulated patients, control was derived from the crank angle
shaft encoder, having set start and stop angles for each muscle group as shown
in Table 1. A time advance compensated for muscle response delay. This time
advance was initially varied in 50ms increments and the resulting pedalling
speed against fixed resistance measured with a stopwatch. We found 150ms was
broadly optimal for these subjects.
Changes in cadence from about 25 to 55 rpm could then be accommodated on load.
Mode 3, program 4 (commutator) Cycling program
for partial T11/12
Channel 1
2 3 4
5 6 7
8
Muscle group
RQuad RGlut
RHam LGlut LQuad LTibA LHam LGstrc
Pulse Current
110 60 60
60 80 40
80 40 mA
Pulse Period
50 50 50
50 50 50
50 50 ms
Threshold p.w.
40 40 40
40 100 40
40 40 ms
Maximal p.w.
304 300 300
300 296 300
302 300 ms
Start Angle 183
205 318 45
3 225 115
180
Stop Angle 312
318 68 139
129 315 219
270
LTDC offset°
315 150
Time Advance in ms
Shaft sense
255 +ve=0/-ve=255
Table
1: Parameter block (from stimulator print out). Start and stop angles in
degrees past LTDC.
RESULTS
Over the first 16 months, our partial
lesion patient averaged 21 minutes of daily FES training with a simpler system
/3/, /9/. He has exercised for a further 16 months with the system described
here. At 20 months, bone density in his more
paralysed limb segments had increased significantly, that in his left tibial
tuberosity by 44+/-2%. With this new system, our subject has repeatedly cycled
ca. 12km at a time either in a sports hall or on near level ground outdoors
(Fig 3).
Fig
3. Partial lesion subject cycling with
FES. Note AFO on left.
In
this English pilot study, we now have three complete lesion paraplegics (one
implant and two surface stimulated), all using Trice recumbent tricycles (with
moulded cushions to reduce seating pressure). So far, two out of these three
complete paraplegics have already cycled more than one kilometre at a time
while powered by their own FES activated leg muscles.
DISCUSSION
We
have clearly established a practicable system for FES leg powered cycling. This
may be a valuable function following spinal cord injury, giving considerable
health benefits, especially as our patients, training at home, are able to
maintain a high level of exercise. We hope to design a new stimulator with more
channels, so that tibialis anterior can be included for the complete
paraplegics. Readily replaceable battery packs and improved electrode suits are
also under consideration. The high proportion of our patients (3 out of 4) who
have already achieved a kilometre plus cycling capability is most encouraging.
/1/ Mohr T., Podenphant J., Biering-Sorensen F.,
Galbo H., Thamsborg G. and Kjaer M., Increased bone mineral density after
prolonged electrically induced cycle training of paralyzed limbs in spinal cord
injured man, Calcified Tissue International, 61, 1997, 22-25.
/2/ Janssen T.W.J., Glaser R.M. and Shuster D.B., Clinical efficacy of electrical stimulation exercise training: effects on health, fitness, and function, Topics in Spinal Cord Injury Rehabilitation, 3, 1998, 33-49.
/3/ Donaldson N., Perkins T.A., Fitzwater R.,
Wood D.E. & Middleton F., FES cycling may promote recovery of leg function
after incomplete spinal cord injury, Spinal Cord, 38, 2000, 680-682.
/4/ Petrofsky J.S., Heaton H.H. and
Phillips C.A., Outdoor bicycle for exercise in paraplegics and quadriplegics, J.Biomed.Eng.,
5, 1983, 292-296.
/9/ Carmen Bruck cycling, Tomorrow’s World,
BBC1 TV program, 4 November, 1998.
/10/ Perkins
T.A., Donaldson N.deN., Harper V.J., Tromans A.M., Wood D.E. and Rushton D.N.,
Provision of standing, stepping and cycling for a T9 paraplegic with a
lumbo-sacral anterior root stimulator implant (LARSI), in IFESS Conf. Proc.:
1996-1998 CDROM, Ed
P. Meadows, Pub IFESS, Internat.Funct.
Elec., 5030 N. Hill Street, La Canada Flintridge, CA 91011-2335, USA,
INSIFESS98 Proceedings, Lucerne, 20-26
August1998, 1999, \Full
Papers\Perkins.htm.
ACKNOWLEDGEMENTS
This work was sponsored by INSPIRE, the
Wellcome Trust and Royal National Orthopaedic Hospital (Stanmore). We also
thank Mr. Tromans & Professor Swain (Salisbury District Hospital), Drs.
Middleton & Mathew (Stanmore Spinal
Injury Unit) and Professor Hunt of Glasgow University for their support.
AUTHOR’S ADDRESS