PILOT APPLICATION OF A CLOSED-LOOP FES SYSTEM FOR THE STANDING UP TRAINING OF PARAPLEGIC PATIENTS

M. Ferrarin°, R. Spadone°, R. Cardini*

° Centro di Bioingegneria, Fond. Don Gnocchi IRCCS, Politecnico di Milano, Milano, Italy

* Servizio di Neuroriabilitazione, Fond. Don Gnocchi IRCCS, Milano, Italy

SUMMARY

In this paper the development and the application of an innovative training procedure for paraplegic patients based on the recovery of sit to stand to sit movements induced by FES and with the assistance of a mechanical device for partial weight relief is presented. A closed loop system based on knee goniometers and a PID regulator is used to control quadriceps muscle stimulation. First application on a paraplegic patient showed a relevant increase in muscle strength and endurance. It is argued that the recovery of standing posture at early stage of rehabilitation program can provide positive effects on musculo-skeletal, cardiovascular, respiratory and vestibular systems.

STATE OF THE ART

One of the most important phases before the application of FES to restore walking function in paraplegic patients is the increasing of muscle strength and fatigue resistance by means of specific training /1/. The standard training technique consists in the application of electrical stimulation on both quadriceps with the sitting patient that causes alternate oscillations of lower legs /1/. In general the stimulation is delivered with a pre-defined fixed pattern. Only in few studies closed loop controlled system has been used both for knee extension weight exercise /2/ and for leg cycling exercise /3/ with the advantage to adjust continually the stimulation current on the basis of muscle’s response and fatigue onset.

In our approach a closed loop controlled FES of quadriceps muscles combined with a mechanical device (Weight Relief System - WRS) is used to induce standing up-standing-sitting down movements /4/. A future step will be to consider the stimulation of gluteus muscles too.

MATERIAL AND METHODS

Fig.1 - Schematic drawing of Weight Relief System and closed loop controlled stimulator.

In fig.1 a scheme of the whole system is represented, both the mechanical device (WRS) and the stimulator controlling system. The WRS is composed by a see-saw construction : the patient is sitting on one side while on the other there are counterweights that relieve patient weight and facilitate the sit to stand movement. The amount of counterweight is decreased during training progression depending on the increase of muscle force, in order to reach the maximal load on lower limbs at the end of training.

The control system uses flexible electrogoniometers applied on the knee joint of each leg and surface electrodes for quadriceps stimulation. The measured knee angles are sent to a PC that adjusts in real time stimulation pulse width (PW) by means of a digital controller.

A multichannels PC-driven stimulator delivers rectangular monophasic current pulses with the following parameters :

- pulse frequency: 1 - 100 Hz, 25 Hz in our application;

- pulse amplitude: 0 - 150 mA, 110 mA in our application;

- pulse width: 0 - 500 ms, software controlled.

The control system is showed in fig. 2:

Fig.2 - Block scheme of closed loop controlled FES system.

A digital PID controller is used for each leg of the patient /5/. The knee angular reference trajectory was derived through an optimisation procedure whose target function was the minimisation of knee torque needed to stand up. In order to solve this problem an inverse dynamic model of the entire system (patient plus WRS), able to compute time course of torque necessary to provide a given standing up trajectory, was developed. The result was a modified hyperbolic tangent, opportunely repeated several times to obtain the reference trajectory used for patient exercise (see fig. 3).

Fig.3 - Reference knee joint trajectory.

Clinical case

This training procedure was applied on one paraplegic patient (male, 28 yrs), with a complete T5-T6 spinal cord lesion since 2 years and without any previous FES experience. The patient was recruited for a rehabilitation program for walking restoration with ARGO and FES. The results here reported are referred to the first four months of training (4 sessions/week and 1 hour/session). In the first weeks the patient was sitting on a table and the controlled stimulation was used to impose a sinusoidal trajectory (T=2.5 sec) at both knees alternatively with 2 Kg weight at each ankle (pre-training). The standing up training was started as soon as quadriceps force was enough for the application of the procedure described before.

Evaluations

Before, during and after the training some evaluations were performed to quantify the change in muscle strength, endurance and spasticity level. Maximal isometric knee torque induced by electrical stimulation (f=25 Hz, PW= 500 ms) was measured by means of an isokinetic device (Cybex) with 60° knee flexion. With the same device the fatigue index (FI%) was evaluated, using the following formulas:

FI% = M_max - M_30” / Mmax

where:

M_max = maximal isometric knee torque at the beginning of the test;

M_30” = isometric knee torque measured after 30 s of continuos stimulation.

Recovery from muscle atrophy was followed up by measuring thigh circumference and with ecographic examination of Rectus Femoris and Vastus Medialis. Muscle spasticity was tested with Pendulum Tests.

RESULTS

Fig. 4 - Changes in maximal isometric knee torque during the training.

An increase in muscle force and endurance was found during training progress. After four months counter weights were reduced more than 30% respect the initial value, nevertheless the number of standing up movements in one session increased from 4 to 35 (5 series of 7 consecutive sit to stand movements separated by 15 min of rest).

Fig. 5 - Changes in fatigue index during the training.

In fig.4 the histogram of the maximal isometric knee torque is showed, where an increase from 20 Nm to 40 Nm can be found in both legs. The increase started with the specific standing-up training, allowing to suppose a better efficacy than the traditional oscillating training. This hypothesis can be confirmed only with a larger number of patients.

The change in muscle endurance is shown in fig. 5 where a decrease in the fatigue index, meaning an increase of fatigue resistance, is presented mainly in the left leg; an increase of the fatigue index in the pre-training period can be noted, probably due to the different loads applied in the two phases.

In fig. 6 the diameter of the thigh measured 15 cm over the top of the patellae with extended legs is reported. A progressive improvement of muscle conditions (an increase of about 10% of thigh diameter in both legs) since the first phase of the pre-training can be noted. This data were confirmed by the ecographic study where a growth of about 80% in Rectus Femoris and 20% in Vastus Lateralis of both lower limbs was found.

The Pendulum Test was showing a low level of spasticity in all knee extensors muscle that didn’t change significantly during the whole training period.

Fig.6 - Changes in thigh circumference during the training.

DISCUSSION

The positive results obtained up to now from the application of this training procedure allow to be optimistic about its efficacy. In particular the increase of muscle volume and maximal isometric knee torque and the improvement of muscle endurance demonstrate its effectiveness on muscle conditioning. From the rehabilitation point of view, the possibility to obtain a FES induced standing-up movements since the very initial phase of the training (when the lower limb muscles are not developing enough torque yet) without overloading the upper limbs and assuring safe experimental condition for the patient, makes this technique promising. It can be supposed that the early restoration of active standing-up/sitting-down movements and standing posture, besides the well known therapeutic benefits of FES training /6/, facilitates the functional use of walking system (FES and hybrid orthoses) by an indirect stimulation of postural control system. Patient’s impressions are in this direction.

Moreover the presented procedure can be used for the development of innovative FES control strategies (closed-loop techniques, adaptive control, neural networks, patient driven control, see the accompanying paper of Riener et al.) and their testing on patients in controlled and safe experimental conditions.

We have already started to apply this training procedure on other two paraplegic patients. Further development will be to consider also Gluteus Muscles in the motor scheme.

REFERENCES

/1/ Bajd T., Kralj A., Functional Electrical Stimulation : standing and walking after Spinal Cord Injury, CRC press Inc., Boca Raton, Florida, 1989.

/2/ Ezenwa B.N, Glaser R.M., Couch W., Figoni S.F., Rodgers M.M., Adapatative control of functional neuromuscolar stimulation-induced knee extension exercise, J Rehab Res & Dev, 28(4):1-8, 1991.

/3/ Petrofsky J.S., Phillips C.A., Heaton H.H., Glaser R.M., Bicycle ergometer for paralyzed muscles, J Clin Eng, 9 :13-19, 1984.

/4/ Ferrarin M., Frigo C., Spadone R., Pedotti A., Development of a closed loop FES system and application to standing up movement for paraplegics, Rehab R&D Progress Reports 1997, supp. J Rehab Res & Dev (submitted to).

/5/ Ferrarin M., D’Acquisto E., Mingrino A., Pedotti A., An experimental PID controller for knee movement restoration with closed loop FES system, Proc. 18th Annual International Conference IEEE - EMBS, paper n. 386, 31 october - 3 november 1996, Amsterdam, Olanda, 1996.

/6/ Glaser R.M., Functional Neuromuscolar Stimulation : exercise conditioning of Spinal Cord Injuried Patients, Int. J Sports Med., 15 : 142-148, 1994.

AUTHOR ADDRESS

Dr.Eng. Maurizio Ferrarin, PhD

Centro di Bioingegneria, Fond. Don Gnocchi IRCCS, Politecnico di Milano

Via Capecelatro, 66 ; I-20148 Milano, Italy

e-mail : ferramau@mail.cbi.polimi.it ; Tel: +39-2-40.308.305 ; Fax: +39-2-40.48.919