INITIAL RESULTS FROM TWO TRIALS OF AN IMPLANTABLE TWO CHANNEL DROP FOOT STIMULATOR

Laurence Kenney¹, Gerrit Bultstra³, Rik Buschman^1,2, Paul Taylor⁴, Geraldine Mann⁴,Hermie Hermens¹, Jan Holsheimer³, Anand Nene¹, Martin Tenniglo¹, Hans van der Aa ², John Hobby ⁴.

1. Roessingh Research and Development bv, Enschede, The Netherlands; 2. Twente Institute for Neuromodulation, Enschede, The Netherlands; 3. University of Twente, Enschede, The Netherlands; 4. Salisbury District Hospital, Salisbury, UK.

SUMMARY

This paper reports preliminary results of pilot studies of a new implantable two channel drop foot stimulator. The two subjects use the stimulator on a daily basis and have shown increases in walking speed between 10% and 44% when compared to their baseline measurements. Isometric tests have demonstrated that the stimulator allows repeatable and selective stimulation of ankle joint muscles.

STATE OF THE ART

There is a growing body of evidence, including one RCT, supporting the orthotic benefits of using single channel, surface-mounted drop foot stimulator for Cardio Vascular Accident (CVA) patients and others [1],[2]. The clinical benefits are now being seen by growing numbers of patients in the UK and mainland Europe, not just at the centres with a direct research interest in the technology but also elsewhere. This move away from the research centres into more widespread clinical usage can be seen as a coming of age for the technology. However, this has been a very slow and patchy process and even in the UK where clinical usage is now relatively widespread, the total number of patients being treated remains relatively small [2]. While less than desired funding is undoubtedly a factor, technical limitations have also caused problems, chiefly those inherent with the use of surface electrodes

In the past, surface stimulators have suffered from a number of practical problems, particularly associated with the foot switch and leads. Despite considerable effort on the part of engineers to replace the footswitch with an alternative, for now it remains the sensor of choice in clinical drop foot systems. The traditional problem with the footswitch and leads was a poor robustness, with fatigue failures commonplace. In recent years, this low-tech, practical but important problem has been tackled and footswitches and leads are now available that typically last in excess of 6 months daily use.

Nevertheless, the problems inherent to stimulating the Common Peroneal nerve using surface electrodes remain. These include a lack of selectivity over the muscles and nerves recruited, sensitivity of muscle recruitment to electrode placement and pain and tissue irritation associated with passage of current through the skin. Taylor [3] identified problems with locating the electrodes as the most common non-physiological reason for discontinuing use of the surface stimulator. These issues have long been recognised and various attempts have been made to implant a single channel drop foot stimulator system on the Common Peroneal nerve [4],[5]. However, this approach failed to solve the selectivity problem as it was not possible to control the relative proportions of dorsiflexion and eversion. This project attempts to resolve this problem by stimulating the two branches of the Common Peroneal nerve separately. The Deep Peroneal nerve innervates muscles that primarily dorsiflex and invert the foot, whilst the Superficial innervates everting muscles.

MATERIAL AND METHODS

Stimulator development

The stimulator that is based on transcutaneous RF-coupling was developed over several years at the University of Twente and Roessingh Research and Development [6]. It is based on a very simple receiver design, using basic passive components, encapsulated in silicon rubber. The novel aspect of the design lies in the type and location of the electrodes. The electrodes are sub-epineural type, developed for this application, but similar in design concept to certain electrodes used in pain relief applications. The location of the receiver distal to the knee avoids the need for the cabling to cross a joint, a common cause of failure in similar applications. The transmitter is located over the site of the receiver, and is triggered in the same manner as the conventional surface stimulator, using a foot switch.

Clinical protocol

The pilot study was intended to investigate the following questions: does the stimulator function as predicted, is it safe for use in humans and, are there any side-effects. The predicted functions were that its use would result in an improvement in gait, that stimulation response would be relatively insensitive to minor (1-2cm) changes in transmitter positioning and that selective stimulation of the two branches of the Common Peroneal nerve could be achieved. Ethics and regulatory approval for both trials were granted from the appropriate authorities.

The first implant took place in The Netherlands in July 2000 [ 7]. Since then, a further 3 implants have taken place, one in The Netherlands and two in the UK. The subjects are all CVA patients with a stable neurology, at least 3 years post stroke and aged between 31 and 48 years. Baseline data on walking speed and endurance were gathered on at least 3 separate occasions, both without and, with their normal walking aid (if any). The UK group also measured Physiological Cost Index (PCI) data. Prior to the implant operation nerve conduction measures were taken to check the integrity of the Deep (DPN) and Superficial Peroneal (SPN) nerves [8]. Following implantation all measurements were repeated. Furthermore, isometric torque measures were taken following a period of recovery and at regular time intervals at follow up. Isometric measures of ankle moment were taken using custom-built devices described elsewhere [9],[10].

RESULTS

Text Box:
Figure 1: Walking speed Text Box:
Figure 2: Distance in 6 minutes

The results of the first two implanted subjects in the Netherlands and the UK are presented. Figures 1 and 2 show the results of the walking speed and 6 minutes endurance measurements pre and post implant, respectively for the Dutch (NL) and English (UK) patient. The Dutch patient was an occasional walker with an AFO and therefore measurements with and without the orthosis were taken. When using the implanted system in both patients the walking speed and distance were increased by respectively 10% (UK) and 44% (NL) from mean baseline values. Figures 3 and 4 show typical graphs of the isometric response to stimulation at ‘optimal’ setting for the 2 subjects. The patients themselves defined the optimal setting. The stimulation times and ramping varied between the Dutch and UK transmitter, due to minor changes in the settings for the two patients. These graphs show that at the onset of stimulation the force produced increased rapidly and was maintained at a stable level. After termination of stimulation the force rapidly declined.

Text Box:
Figure 4: Isometric moments from stimulation of Deep and Superficial Peroneal nerves (UK) Text Box:
Figure 3: Isometric moments from stimulation of Deep and Superficial Peroneal nerves (NL)

Text Box:
Figure 5: Sensitivity of dorsiflexion response to transmitter placement The sensitivity of isometric response to transmitter movement was also investigated. In the case presented here, sensitivity was defined as change in dorsiflexion moment with proximal/distal displacement. Results from the UK and Dutch patients are shown in figure 5 (2^nd order polynomial curve fitted to data). This graph shows that a displacement of about 1cm would not significantly affect the moment produced, indicating that the implantable system is relatively insensitive to minor positioning errors.

DISCUSSION

As may be seen in figures 1 and 2, both patients gained orthotic benefit using the stimulator. Taylor [11] showed a mean change in walking speed amongst CVA patients of 12% (0.07 m/sec) at 6 weeks and 27% (0.16 m/sec) at 4 ½ months use of the surface stimulator. The results obtained in the present study are similar to those reported by Taylor [11]. Figures 3 and 4 show that stimulation on user-defined optimal settings resulted in a response characterised by smooth dorsiflexion with moderate eversion. These results were repeatable both within, and between test sessions [10].

The sensitivity to transmitter placement was determined to be relatively low. There are no reports in the literature quantifying typical sensitivity to surface stimulation of the CPN, but experience suggests that it is significantly higher. This ease of positioning may be of specific benefit when we consider that typical CVA patients also have less control over their upper extremities. This new system may therefore also help these subjects to gain more independence.

The implants in two other patients have shown failures after having functioned properly for periods of months. An investigation of one of the explanted systems has shown that the system failure was caused by a fault in the receiver manufacturing process. Prior to failure, both of these patients also showed orthotic benefit from the device and similar isometric results to those reported here. The manufacturing process has now been adapted and a new receiver version is currently in production. Subject to regulatory approval, the clinical trials will continue in the near future.

REFERENCES

1. Burridge JH, Taylor PN, Hagan SA, et al. The effects of common peroneal stimulation on the effort and speed of walking: a randomized controlled trial with chronic hemiplegic patients. Clin Rehabil 1997;11: 201-210

2. Burridge J. Does the drop-foot stimulator improve walking in hemiplegia? Neuromodulation 2001;4: 77-83

3. Taylor PN, Burridge JH, Dunkerley AL, et al. Patients' perceptions of the Odstock Dropped Foot Stimulator (ODFS). Clin Rehabil 1999;13: 439-446

4. Waters RL, McNeal DR, Faloon W, et al. Functional electrical stimulation of the peroneal nerve for hemiplegia. Long-term clinical follow-up. J Bone Joint Surg Am 1985;67: 792-793

5. Strojnik P, Acimovic R, Vavken E, et al. Treatment of drop foot using an implantable peroneal underknee stimulator. Scand J Rehabil Med 1987;19: 37-43

6. Holsheimer, J., Bultstra , G., Verloop, A. J., van de Aa, H. E., and Hermens, H. J. Implantable dual channel peroneal nerve stimulator. Proc Ljubljana FES Conf (ed. Jaeger, R. and Bajd, T.) 1993: 43 -44.

7. van der Aa HE, Bultstra G, Verloop AJ, Kenney L, Holsheimer J, Nene A, Hermens HJ, Zilvold G, Buschman HPJ. Application of a dual channel peroneal nerve stimulator in a patient with „central“ drop foot. Proceedings of the World Federation of Neurosurgical Societies. 2001 (in press).

8. Lee HJ et al. Peroneal nerve conduction to the proximal muscles – an alternative to conventional methods. Am J Phys Med Rehab 1997;76:197-199.

9. Wood DE, Donaldson NN, Perkins TA. Apparatus to measure simultaneously 14 isometric leg joint moments. Part 2: Multi-moment chair system. Med Biol Eng Comput 1999;37: 148-154

10. Buschman HPJ, Kenney LJ, Nene AV, Bultstra G, Tenniglo M, Hermens HJ, van der Aa HE. Development and performance of an implantable 2 channel peroneal nerve stimulator for dynamic equinovarus foot in stroke. Proceedings IEEE-student branch. Eindhoven, The Netherlands May 2001

11. Taylor PN, Burridge JH, Dunkerley AL, et al. Clinical use of the Odstock dropped foot stimulator: its effect on the speed and effort of walking. Arch Phys Med Rehabil 1999;80: 1577-1583

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the support of the EU (NEUROS TMR network) and the UK Medlink programme (IMPULSE project). Thanks also to Duncan Woods of Salisbury District Hospital.