鈥淚鈥橫 NOT bionic,鈥� insists Julie Hill as she stands up. 鈥淚鈥檓 just fitted with
some clever electronics, but it鈥檚 my muscles, my bones that are doing this.鈥� For
most people, standing up is no big deal. But for Hill it鈥檚 different. Six years
ago, a car crash severed her spinal cord, leaving her with no sensation or
muscular control below the waist. Last year she hit the headlines as the first
person to have 12 electrodes surgically implanted along her spine to let her
stand up at the touch of a button.
The pioneering operation performed on Hill is just one of several great
strides made over the past couple of years in FES, or functional electrical
stimulation of muscles. The current fed to her electrodes mimics the control
signals that usually pass from the brain to the legs. For Hill, FES has been a
real success, restoring at least some movement. But for bioengineers, standing
is just the beginning鈥攖heir ultimate goal is to get people out of their
wheelchairs and walking.
To help a paralysed person walk, researchers must synchronise the contraction
and relaxation of dozens of muscles, and at the same time build up stamina and
conquer fatigue. At present, FES falls far short of such sophistication. Indeed,
today鈥檚 technology allows Hill to stand for only 4 minutes. But the prospects
are far from bleak. FES can now help people to grip with paralysed hands and
promises a safer way of restoring bladder control. And while these advances are
important in themselves鈥攂ladder infection, for example, is a major cause
of death and disease among paralysed people鈥攖hey also point towards ways
that might help people take their first steps in future. 鈥淥ur ideal is to
combine the strengths from all these technologies,鈥� says biomedical engineer
Paul Taylor, who works with the team at Salisbury District Hospital where Hill鈥檚
operation was performed.
Advertisement
The nerves that control muscles in the arms, legs and abdomen run down the
hollow centre of the spine, before branching off to their final destinations.
Spinal injuries often crush or sever these nerves at one point, leaving the
nerves below undamaged. And while the muscles fed by these nerves may also be
fine, they are cut off from their chain of command. 鈥淲hat these people require
is a form of jump-lead that could be plugged into their nerves below the damaged
area of spine, to stimulate these nerves and restore the lost function,鈥� says
Nick Donaldson, a biomedical engineer from University College London. The
electronics he designed for Hill鈥檚 system do just that.
Hill鈥檚 electrodes stimulate nerve bundles where they leave her spine. The
electrodes, six on each side, fix firmly to the vertebrae and send between 12
and 15 electrical pulses a second down the nerves to the muscles they serve. The
impulses to Hill鈥檚 leg muscles now have a voltage of 10 to 100 microvolts, a
current of between 4 and 20 milliamps and a duration of about 200
microseconds.
Before Hill鈥檚 operation, FES systems simply fired pulses to muscles that
straighten the knee. With their legs straight, patients then had to haul
themselves upright. Hill鈥檚 system is much smarter. She wears a tiny computer,
which controls the delivery of pulses to her electrodes. Its instructions are
relayed across the skin to a receiver implanted over the ribs of her chest. When
she presses a button on the computer, it triggers a 鈥渇iring sequence鈥� that
changes the duration and frequency of pulses to each electrode by precise
amounts. The muscles work in concert, contracting by the right amount to raise
her to her feet. 鈥淲e are the first people to attempt to activate all the muscles
in the leg simultaneously, which gives us the possibility of controlling
balance,鈥� says Donaldson.
Overstimulation
For all its benefits, Hill鈥檚 system still has major disadvantages. For
example, each nerve bundle contains fibres, or axons, that go to several
different muscles. 鈥淓ach time we send a signal to one electrode, more than one
muscle will contract,鈥� says Donaldson. 鈥淔or example, the electrode that
stimulates the muscles that cause Hill鈥檚 knee to straighten also causes her body
to jack-knife.鈥� The Salisbury team is not yet certain which muscle is causing
the problem. To identify it, the researchers are now anaesthetising individual
muscles that run through Hill鈥檚 groin and then switching on her electrodes. 鈥淥ur
hope is to pinpoint the muscle,鈥� says Duncan Wood, another member of the team.
鈥淲e could then block its activity using a long-lasting anaesthetic, surgically
cut the nerve that supplies it, or possibly relocate the muscle tendons so that
it performs some other function.鈥�
Until this type of problem is solved, fine control of movement will be
impossible. In future, it may be feasible to divide each nerve bundle into many
different branches and attach an electrode to each. But so far this idea has
been discounted because it would make the surgery and the electronic implant too
complicated.
While this approach may be years away, surgeons are already relocating
tendons. At the MetroHealth Medical Center in Cleveland, Ohio, surgeon Michael
Keith and bioengineer Hunter Peckham use the technique to restore hand and arm
movement to people with high spinal injuries. These people cannot move their
hands, but have limited control of their arms. Keith first detaches the bottom
of the deltoid muscle, which runs over the shoulder, and with the help of a
length of artificial tendon, joins it to the lower end of the triceps. This
allows people to extend their elbow. Next, he relocates the brachioradialis in
the forearm, allowing patients to extend their wrists.
In a second operation, Keith and Peckham also implant electrodes to restore
movement to the fingers. They sew seven electrodes onto the muscles of the
forearm which, when fired, cause the fingers to clench. By varying the pattern
of impulses to the electrodes, the patients can choose which type of grip they
want. So they can, for example, hold a pen or even pull a cork from a bottle.
They change the grip with a shoulder switch (see
Diagram). More than 30 people
have now undergone this procedure, called the NeuroControl Freehand System, and
last year Roger Fenn, a telecommunications engineer from Southampton, became the
first man in Britain to be wired up when Keith visited the Salisbury team.
For now, this type of FES holds no promise for people like Hill. The idea of
sewing electrodes to individual leg muscles has already been tried and rejected.
Donaldson found that the risk of infection from as many as 30 electrodes was too
great.
Complex mimicry
Where possible, bioengineers aim to copy the body鈥檚 own muscular controls.
But the complexity of the nervous system often makes this a difficult task.
Every nerve bundle contains a number of different types of fibres. At one
extreme are large diameter axons, which stimulate muscle fibres that contract
rapidly. These muscles have a poor blood supply and generate energy
anaerobically by burning the carbohydrate glycogen鈥攁 process that releases
lactic acid and causes fatigue. At the other extreme are small axons, which
stimulate slowly contracting muscle fibres. These have a good blood supply,
consume glucose aerobically, and are less prone to fatigue. Each axon-muscle
fibre package is called a motor unit. And the body tends to use slow motor units
for everyday tasks such as holding a limb in a fixed position, and fast units
when sudden short bursts of activity are needed.
A typical muscle has about 300 motor units, although no more than 5 per cent
usually contract at once. When more muscle power needed, the brain makes every
unit work harder by increasing the frequency of nerve impulses. It also
stimulates more units, maintaining smooth contractions by cycling the activity
between them, giving the nonactive ones a chance to recover.
That鈥檚 how the body does it. But FES distorts this system in vital ways, says
Stanley Salmons, professor of medical cell biology at the University of
Liverpool. 鈥淔irst there is no orderly recruitment of axons, so it is difficult
to gradually increase the number of motor units being activated and obtain a
graded response,鈥� he says. 鈥淪econdly, the physics of ion movement through nerves
dictates that large axons are stimulated in favour of small ones.鈥�
These problems are reflected in Hill鈥檚 case. Her electrodes tend to stimulate
fast motor units, and lock them in their contracted position. And although Hill
cannot feel it, her muscles soon become exhausted. They give way after a few
minutes and she sinks into her chair.
Bioengineers at the University of Glasgow are tackling these problems with a
device that stimulates specific types of nerve fibre. Their focus is not the
arms or legs, however, but the bladder. 鈥淎fter spinal injury, people often
become incontinent because they cannot empty their bladder,鈥� says physiologist
Ronald Blaxendale from Glasgow. 鈥淭hey have no voluntary control over the muscle
that squeezes the bladder, the detrusor muscle, nor can they chose to relax the
sphincter muscle that acts as the valve.鈥�
Existing systems for artificially emptying the bladder exploit the difference
in response times between fast and slow motor units. The nerve bundle leading to
the bladder contains a mixture of large axons that supply the sphincter, and
small axons that innervate the detrusor muscle. When this bundle is stimulated
the sphincter contracts, and relaxes, rapidly. The slow motor units of the
detrusor muscle take longer to contract and squeeze the bladder as the sphincter
relaxes, allowing urine to pass. The problem is that it creates a back pressure
that risks damaging the kidneys.
Blaxendale and his colleagues hope to overcome the muscular stalemate with
three electrodes built into a single cuff that encircles the nerve bundle
leading to the bladder (see
Diagram). When impulses travel down a nerve, the
axon鈥檚 voltage suddenly become positive relative to the surrounding tissue. The
first electrode, an anode, blocks any such signals that might still be coming
from the spine. The second, a cathode, stimulates the nerve on command, while
the third, the anode, acts like a filter, allowing the impulses to travel only
along certain fibres. 鈥淚f you juggle with the timing and current of pulses at
the anodes it is possible to block the transmission of the big fibres but allow
action potentials in the small fibres to leak past,鈥� says Blaxendale.
When applied to the bladder, this 鈥渁nodal block鈥� turns off all nerve traffic
to the sphincter, allowing it to relax. But impulses still pass along the small
fibres to the detrusor muscle, making it contract. 鈥淚n experiments using
animals, the system now works at least four times out of five,鈥� says
Blaxendale.
Cuff combination
Another possible refinement would be to use five or more of these cuffs. This
would allow different populations of muscle cells to contract and relax in turn,
and share out the work within the detrusor muscle. Such a system would allow
different populations of muscle cells to contract and relax in turn. Such a
system would be closer to the body鈥檚 own controls, says Blaxendale, but not that
close. 鈥淚t would still be far from the 300 or so separately controlled motor
units that healthy bodies have to play with in each muscle,鈥� he says. Blaxendale
reckons that such a system could help someone like Hill by preferentially
stimulating slow motor units in the legs, reducing the problems caused by
fatigue.
In order to increase the length of time she spends standing, Hill is already
fighting fatigue with the help of a discovery made by Salmons in the 1970s. He
found that stimulating fast-reacting muscle fibres for a while every day
increased the number of capillaries running through them and converted them into
slow, fatigue-resistant fibres. To exploit this dicovery, Hill spends an hour a
day lying down, her feet dangling off the bed and 2-kilogram weights strapped to
her ankles. Her control unit has an 鈥渁utomatic鈥� setting that repeatedly lifts
and lowers her legs.
One of the critical requirements of any system that aims to get people
walking again is a feedback system. The Salisbury team is experimenting with a
pressure switch in the shoe that has already proved useful for stroke patients
who have lost some control of their legs. When you take a step forward, your
toes lift very slightly to stop you tripping up. Stroke patients often lose this
involuntary mechanism. By linking the pressure sensor to an electrode that
sticks on over the muscle that pulls the toes up, the Salisbury researchers have
restored this action. They have now joined forces with engineers at the
University of Surrey to study how to stimulate a second muscle above the knee,
to bend the knee and lift the foot even higher off the ground.
All these developments in FES offer a glimpse of the kinds of techniques that
would be needed to help paralysed people to walk again. But this goal is still a
long way off. A feedback system for walking, for example, would probably have to
include sensors that could monitor such things as the angle of joints and the
load carried by different muscles.
For the present, though, Hill is happy that she can stand. Besides the added
freedom of movement it gives her, standing also has a therapeutic value. Muscles
that receive no stimulation rapidly thin. Bones press against the weakened
tissue and can cut off the blood supply causing pressure sores. 鈥淚f a person鈥檚
buttocks are allowed to waste away, then the risk of pressure sores increases
dramatically,鈥� says Donaldson. Infected pressure sores are another major threat
to the lives of paralysed people.
Standing up a few times a day, Donaldson says, both reduces the pressure on
the buttocks and helps to maintain muscle bulk. This last factor is particularly
important to Hill. 鈥淎s far as my legs are concerned they are in better shape
than they have been for a long time,鈥� she says. 鈥淭hey had wasted away after the
accident, but now look very presentable.鈥�