Implantable device for long-term electrical stimulation of denervated muscles in rabbits H. Lanmi]ller I Z. Ashley 2 E. Unger I H. Sutherland 2 M. Reichel I M. Russold 2 J. Jarvis 2 W. Mayr I S. Salmons 2 1Center for Biomedical Engineering & Physics, Medical University of Vienna, Austria 2Department of Human Anatomy & Cell Biology, University of Liverpool, Liverpool, UK Abstract--Although denervating injuries produce severe atrophic changes in mamma- lian skeletal muscle, a degree of functional restoration can be achieved through an intensive regime of electrical stimulation. An implantable stimulator was developed so that the long-term effects of different stimulation protocols could be compared in rabbits. The device, which is powered by two lithium thionyl chloride batteries, is small enough to be implanted in the peritoneal cavity. All stimulation parameters can be specified over a wide range, with a high degree of resolution; in addition, up to 16 periods of training (10-180 min) and rest (1-42 h) can be set in advance. The micro- controller-based device is programmed through a bidirectional radiofrequency link. Settings are entered via a user-friendly computer interface and annotated to create an individual study protocol for each animal The stimulator has been reliable and stable in use. Proven technology and rigorous quality control has enabled 55 units to be implanted to date, for periods of up to 36 weeks, with only two device failures (at 15 and 29 weeks). Changes in the excitability of denervated skeletal muscles could be followed within individual animals. Chronaxie increased from 3.24 +_ 0.54 ms to 15.57 +_ 0.85 ms (n - 55, p < 0.0001) per phase in the 2 weeks following denervation. Keywords--lmplantable stimulator, Skeletal muscle, Denervation, Rabbit, Excitability, Radiotelemetry Med. Biol. Eng. Comput., 2005, 43, 535-540 1 Introduction SOME SPINAL cord and peripheral injuries cause irrecoverable lower motor neurone damage. This results in flaccid paraplegia and, over time, extremely severe atrophy of the affected muscles. For many years, no effective treatment could be offered to people with this condition. Recently, however, it was shown that the denervated muscles can undergo a remarkable recovery of mass, excitability and force if they axe conditioned with a sufficiently intensive regime of electrical stimulation (KERN et al., 2002a, b; 2004). Treating the muscles in this way will not restore normal mobility, but it may, at some future point, pave the way to neuroprosthetic approaches of the type that have been applied to patients with upper motor neurone lesions (HORCH and DHILLON, 2003). There are, in any case, potential secondary benefits, including: improved cardiovascu- lax fitness; better skin condition and muscle cushioning, with a corresponding reduction in the risk of pressure sores; and improvements in cosmetic appearance and patient self-esteem. Correspondence should be addressedto ProfessorStanley Salmons; emaih s.salmons@liverpool.ac.uk Paper received 4 February 2005 and in final form 31 May 2005 MBEC online number: 20054030 © IFMBE: 2005 Medical & Biological Engineering & Computing 2005, Vol. 43 The present study forms part of a collaborative project designed to place this application of electrical stimulation on a secure scientific footing. In particular, the use of suitable animal models will enable us to devise protocols for clinical use that provide the maximum beneficial effects with the minimum intrusion into the patient's normal daily activities. For this purpose we needed a stimulator that would allow us to investigate the effects of long-term conditioning on the denervated tibialis anterior muscle of the rabbit. The cell membrane of denervated muscle fibres is much less excitable than that of nerve fibres and activating it calls for high currents and long pulse durations. At the whole muscle level, the absence of functional nerve branches means that adequate recruitment depends on conveying electrical excitation directly to all parts of the muscle, and the active area of the electrodes must be large so as to create the necessary volume and intensity of the current field while minimising local heating. For these reasons, the pulse charge required to activate a denervated muscle is 100-500 times greater than that needed to activate an innervated muscle. Our clinical studies in patients suffering from long-term flaccid paraplegia (KERN et al., 1999; 2002b; 2004) were carried out with surface electrodes connected to a stimulator designed for external use (HOFER et al., 2002). Surface electro- des have also been used in some chronic animal studies, but the need for restraint necessarily limits the daily duration of stimu- lation that can be applied (MOKRUSCH et al., 1990). 535