AN INTRODUCTION TO FUNCTIONAL ELECTRICAL    STIMULATION

2 channel open-loop gait assist electrical orthosis
(c) David Ewins 1990

These pages are an edited version of those created by Miss Amanda Lamb as part of her MSc project in Biomedical Engineering at the University of Surrey 1994-95. Although they are meant to provide an introduction to stimulation in general they do concentrate on functional electrical stimulation through the use of surface (transcutaneous) electrodes, as this is the area of group's expertise. Any suggestions for modifications should be addressed to David Ewins

Electrical stimulation is simply the application of electrical pulses to the body, be it for function or therapy. The classical example is that of the cardiac pacemaker. The range of clinical uses of electrical stimulation is wide and includes:

• pain relief (often known as TENS - Transcutaneous Electrical Nerve Stimulation)
• maintaining or increasing range of movement
• muscle strengthening
• facilitation of voluntary motor function
• orthotic training or substitution

Functional Electrical Stimulation (FES) is a subset of electrical stimulation. The term FES is applied to systems which attempt to restore lost or impaired neuromuscular function, such as standing and walking in cases of paraplegia, by the application of electrical pulses to neural pathways or, but less often, directly to muscles. FES is also sometimes known as Functional Neuromuscular Stimulation or FNS.

A portable stand-sit closed-loop electrical orthosis in use
(c) David Ewins 1990

It is important to realise that there are very few FES orthoses that could be considered to be clinically widely available, and that even those that are used on a daily basis require a good deal of input from clinicians, therapists and bioengineers, together with a considerable amount of commitment from the end user.

FES involves depolarising nerve/muscle fibers via externally applied electric current. Once depolarised these fibers conduct action potentials as occurs in healthy tissue.

The electrical pulses applied may be:

Monophasic

The monophasic waveforms tend to be asymmetric biphasic as the net charge built up in the body by the waveform discharges:

Charge Balanced

When charge balanced pulse types are used no net charge is introduced to the body. This is particularly important for some types of denervated muscle stimulation and for implanted electrodes, where electolysis at the electrode tissue interface could take place if there was a non-zero net charge.

Application of Faradic Stimulation

Faradic stimulation may be divided into four main classes:

• Transcutaneous
  
• Percutaneous
  
• Subcutaneous - via peripheral root nerves
  
• Subcutaneous - via spinal root nerves

These represent the depth that the electrode is placed at. Generally as the electrode gets closer to the nerve/muscle, then so the intensity required to initiate nerve depolarisation is reduced, thus less energy is required for the action. Equally important is the fact that that it is possible to be much more specific about which nerves, and therefore to some extent which motor units, are stimulated as the electrodes get closer to the nerve/muscle.

Transcutaneous Stimulation

Transcutaneous stimulation takes place on the skin via surface electrodes. The intensity needed varies greatly with the muscle, its condition, the stimulation pulsewidth used and the contraction required, but in some cases, eg when stimulating the quadriceps to effect standing, then amplitudes up to around 150 mA are not uncommon. The best results are obtained by removing hairs from under the electrodes (through clipping NOT shaving) and using a conducting jelly between skin and electrode.

Typical Electrode Positions for Quadricep Stimulation (-ve over motor point)
(c) David Ewins 1990

Typical Electrode Positions for Peroneal Stimulation -withdrawal reflex
(c) David Ewins 1990

Apart from the high energy required to maintain stimulation, the other major drawback with surface stimulation is that fine movements (as required to produce function in the hand) can be difficult to produce and reproduce. Monophasic or charge balanced pulse types may be used.

Percutaneous Stimulation

The electrical stimulation is transmitted through the skin via a wire or needle into the muscle belly. Flexible wires are most often used as the electrodes as they enable the subject to move around more. This type of stimulation is generally used for research only as it is not considered ethical in a clinical situation. MSc students, of course are expendable :-) and therefore suitable experimental subjects.

Subcutaneous Stimulation

Via peripheral root nerves
Electrode cuffs around nerve bundles are sutured to the muscle epimyseum.

Via spinal root nerves
Electrodes are implanted around spinal root nerves specific to the stimulation needed, eg sacral roots for bladder stimulation, and possibly lumbar roots for standing/walking.

Control of Muscle Contraction

Surface electrical stimulation typically consists of a train of regular monophasic or biphasic pulses which may be described by the following parameters: frequency; amplitude; duration of individual pulses; duration of the pulse train; and rise time for the individual pulses.

Some of these are defined in the FES pulse train shown in below.

An FES Pulse Train

The amplitude and pulse width of the stimulation must be sufficient to meet the threshold of excitability of the stimulated tissue, changes under this level will have no effect as shown in the graphs below.

As the amplitude or pulse width rise the nerve fibres nearest the electrodes and largest in diameter are triggered to threshold and fire.

This continues until all fibres are firing at which point no more increase in force can be obtained, the muscle is said to be saturated.

This increase in recruitment is almost linear between threshold and saturation as shown below:

Effect of Amplitude on Force - Data taken from quadriceps femoris, surface electrodes, pulse duration 300 microseconds, frequency 35 Hz.
Adapted from: Functional Electrical Stimulation: A Practical Clinical Guide (2nd Edition). Benton, Baker et al., 1981 (see bibliography).

Effect of pulse duration on force - Data taken from surface stimulation of the dorsiflexors at a stimulation voltage of 50V and a frequency 30 Hz.
Adapted from: Functional Electrical Stimulation: Standing and Walking after Spinal Cord Injury. Kralj and Bajd, 1989 (see bibliography).

The rate of rise of the pulse can also be important. Too slow a rise time results in changes in the tissue membrane known as accommodation, which gradually elevates the threshold required for the nerve to fire. The pulse used in electrical stimulation do not, in general, allow this effect to occur.

The rate at which the nerve fibres fire is dependent on the frequency of pulse repetition. A single pulse produces a short lived muscle twitch of not more than 250ms. If pulses are repeated more frequently than this the muscle does not have time to relax in-between stimuli and eventually tetanic (continuous) contraction occurs.

Although these look similar to contractions evoked by voluntary stimuli, as voluntary motoneurons are innervated asyncronously, tetany is achieved at much lower rates - 5-25 Hz.

The Effect of Muscle Fatigue

The high rate of synchronous activity in electrical stimulation can cause decreased neural transmitter release. However, the biggest problem is that of muscle fatigue itself. This is because stimulation tends to elicit recruitment of the larger diameter motorneurons (they have a lower threshold), which recruit the faster and more powerful muscle fibres. These fibres, termed type 2 or white, fatigue quicker than the slower, but less powerful type 1 or red muscle fibres. Additonally, with long term paralysis, there can be a transformation of slow fibres to fast ones, which will exacerbate the fatigue problem

This type of fibre recruitment is often a reversal of the normal patterns which, in addition, involve an asynchronous firing which will allow fibres to "rest". Overall, therefore, stimulated muscle will generally fatigue sooner than the same response initiated voluntary.

NOTE: There is a good deal of work being undertaken in muscle condtioning, eg the use of skeletal muscle for cardaic assist (caridac myoplasty), and material on this will be added to these introductory pages at a later date.

As shown in below, the higher the stimulation frequency, the faster the muscle fatigues:

Effect of frequency on fatigue - This data is from an implanted electrode on the peroneal nerve.
From: Functional Electrical Stimulation: A Practical Clinical Guide (2nd Edition). Benton, Baker et al., 1981 (see bibliography).

Simulations

You can examine a simplification of the effect of amplitude, pulsewidth and frequency on muscle torque and fatigue in these Microsoft Excel based simulations (requires Excel Version 5 for Windows (not tried with the MAC yet), on the client platform)

Discussion of Simulations

When controlling the movement of limbs, a compromise frequency is generally sought for each muscle (generally around 20-40 pulses per second) which enables a fairly smooth response be achieved for a reasonable period.

In practical FES systems frequency is often held at a sufficient level for tetanus. Amplitude and pulse width are varied to control the contraction for the patient's needs. Closed loop systems, in which the stimulation is moderated automatically according to such parmeters as joint angles and foot pressure distribution, are now being developed, but few of these are in routine clinical practice.

Torque does not usually decline linearly with time as was the case in the simulation above. The fatigue observed in the simulations is attributed to three factors:

• decreased transmitter release, leading to failure of the synaptic junction
• fatigue of the muscle fibres due to contraction-produced ischemia, which is not relieved as the same fibres keep contracting
• metabolic fatigue of the contractile mechanism (production of lactic acid)

In the fatigue simulation there is no fatigue until 18 Hz and no increase in fatigue above about 80 Hz. This is not the case in reality.

Practical FES Systems

• Requirements of a Practical FES System
• Dropped Foot Stimulation [under construction]
• The University of Surrey Dropped Foot Stimulator [under construction - page not the stimulator :-) ]

Requirements of a Practical FES System (electrical orthosis)

1. Should be Simple to Don and Doff
This is vital as it will to a large extent determine the amount of use the patient will get out of their system. This will limit the number of connections and leads that are external to the body.

2. Function must be Relevant to the User
Some functions seem to have 'obvious' relevance to anyone. But even a function like standing may be of very little use to paraplegic in adapted accommodation.

3. System must Consistently Provide the Desired Function
The electrical orthosis must provide the desired function under a range of working conditions both external, eg location, and internal, eg electrode positioning.

4. The System must include the User
The condition of the user's muscles, bones, ligaments and cardiovascular performance are of vital importance in ensuring that the required function can be attained safely and repeatably.

5. User must be Aware of the Limitations of the System
It must be ensured that the user has realistic expectations. With open-loop systems the user should understand the problems that may occur, eg with fatigue, and realise how these will affect the performance of the system.

6. User must Understand Committment Required to Maximise the Benefits
Generally a long term commitment to a training program (>3 months for "sit to stand") is required.

7. System should Ideally be Fail Safe
This is not at present always possible, eg how does one make an electrical stimulation only standing system fail safe ! With open-loop systems the user should possess the strength/control to cope in the event of a systems failure. The degree to which failure is dangerous depends on the system - but the possibility of failure needs to be carefully considered in programming the stimulator.

A practical electrical orthosis therefore consists of two components. Firstly, a reliable and adaptable system capable of responding to changing parameters. Secondly, a trained user with the necessary motivation. Both parts are of equal importance.

Electrical Stimulation Bibliography

In the following list are publications - books, papers etc., which are good sources of background information for electrical stimulation in general and functional electrical stimulation in particular.

For Surrey Students: Should you require further details, and if you have a copy of the EndNOTE Plus database installed on your PC, I would be happy to send you a copy of my reference database - I have copies of most of the entries in this.

• Benton L A, Baker L L, Bowman B R and Waters R L (1981). Functional Electrical Stimulation - A Practical Clinical Guide (2nd Edition). Rancho Rehabilitation Engineering Program. Rancho Los Amigos Medical Centre, 7601 East Imperial Highway, Downey, California 90242.

• Baker L L, McNeal D R, Benton L A, Bowman B R and Waters R L (1993). Neuromuscular Electrical Stimulation - A Practical Guide (3rd Edition). Rancho Rehabilitation Engineering Program. Rancho Los Amigos Medical Centre, 7601 East Imperial Highway, Downey, California 90242.

• British Standard, Medical Electrical Equipment. Part 2, Particular requirements for safety. Section 2.10, Specification for nerve and muscle stimulators (1988). British Standards Institution.

• Functional Electrical Stimulation: Applications in Neural Prostheses (1977). Edited by F Terry Hambrecht and James B Reswick. Published by Marcel Dekker, Inc.ISBN 0-8247-6632-6.

• Neural Prostheses: Fundamental Studies (1990). Edited by William F Agnew and Douglas B McCreery. Published by Prentice Hall. ISBN 0-13-615444-1.

• Kralj A and Bajd T (1989). Functional Electrical Stimulation: Standing and Walking after Spinal Cord Injury. CRC Press. ISBN 0-8493-4529-4.

• Stillwell G K (1967). Chapter: Clinical Electric Stimulation in Therapeutic Electricity and Ultraviolet Radiation. Edited by Sidney Licht. Published by Elizabeth Licht, New Haven, Connecticut. 2nd Edition, pg 105-155.

• Winter D A (1990). Biomechanics and Motor Control of Human Movement. John Wiley & Sons, Inc. ISBN 0-