RESTORING movements for walking in paraplegia by. Muscle Selection and Walking Performance of Multichannel FES Systems for Ambulation in Paraplegia

Transcription

1 IEEE TRANSACTIONS ON REHABILITATION ENGINEERING, VOL. 5, NO. 1, MARCH Muscle Selection and Walking Performance of Multichannel FES Systems for Ambulation in Paraplegia Rudi Kobetic, Member, IEEE, Ronald J. Triolo, Member, IEEE, and E. Byron Marsolais, Member, IEEE Abstract A minimal set of muscles (8 to 16) were identified as candidates for implantation in a clinical system to provide walking function to individuals with complete paraplegia using functional electrical stimulation (FES). Three subjects with complete motor and sensory paraplegia had percutaneous intramuscular electrodes implanted in all major muscles controlling the trunk, hips, knees, and ankles. Stimulation patterns for walking with FES were generated for different sets of eight and 16 muscles. The quality and repeatability of the resulting gait produced by walking patterns consisting of various combinations of muscles were determined. Most eight-channel stimulation patterns resulted in scissoring or insufficient hip flexion, preventing forward progression. One eight-channel system allowed a maximum speed of 0.1 m/s with a cadence of 22 steps/min and a stride length less than 0.3 m. Improved walking performance was observed with 16 channels of stimulation. This ranged from slow stepto gait at 0.1 m/s to smooth reciprocal gait at 0.5 m/s. In all three subjects, the favored combination of 16 channels included erector spinae for trunk extension; gluteus maximus, posterior portion of adductor magnus and hamstrings for hip extension; tensor fasciae latae and either sartorius or iliopsoas for hip flexion; vastus lateralis/intermedius for knee extension; and tibialis anterior/peroneous longus for ankle dorsiflexion. In one subject the 16-channel FES system provided repeatable day-to-day gait averaging 0.4 m/s, 58 steps/min and a stride length at 0.8 m. A maximum repeatable walking distance with 16 channels was 34 m. Multiple 34-m trials were possible with minimal rests between walks. Fatigue of both the hip extensors and upper body was a limiting factor. The selection of target muscles for implantation is critical to the performance of FES systems. This study provides guidelines to muscle selection for walking with FES based on objective measures of gait performance. The findings indicate that a 16-channel FES system for total implantation is feasible for repeatable short distance, independent, walker-support walking in paraplegia. Index Terms Functional electrical stimulation, paraplegia, walking. I. INTRODUCTION RESTORING movements for walking in paraplegia by means of functional electrical stimulation (FES) has been a research goal for the past 30 years. FES systems coordinate the actions of the paralyzed muscles by delivering sequences Manuscript received August 12, 1996; revised October 31, This work was supported by the Department of Veterans Affairs Rehabilitation Research and Development Service. The authors are with the Veterans Affairs Medical Center, Motion Study Laboratory, Cleveland, OH USA. Publisher Item Identifier S (97) of electrical pulses through a network of electrodes either implanted or attached to the surface of the body. This causes the appropriate muscles to contract and produce the stepping movements for the control of walking. Early attempts at FES for ambulation were made with four channels of surface stimulation [1]. Commonly two channels were assigned bilaterally for stimulation of the quadriceps to cause knee extension and two channels for stimulation of the peroneal nerve to induce withdrawal reflex causing simultaneous activation of hip and knee flexors, and ankle dorsiflexors [11]. Less often a pair of electrodes was added for activation of gluteal group for hip extension [4], [10]. A version of this walking system has been approved by the FDA for clinical application and has been commercially produced. A similar system using implanted electrodes and an implanted stimulator was also implemented [3]. Due to the hip and trunk instability of these systems, others have combined electrical stimulation with braces to improve stability and reduce metabolic energy requirements for walking [15]. There are many shortcomings of the surface stimulation systems. First, donning of the system is difficult, especially, to gain the control of hip and trunk, thus making the use of such system impractical for every day use. Second, stimulation of hip flexors is difficult with surface stimulation, therefore, the system relies on attaining hip flexion by activation of withdrawal reflex [11]. The hip flexion using this method is often variable, slow in response, and diminishes with repetitions. Our approach to walking in paraplegia has been to control muscles individually by means of implanted percutaneous intramuscular electrodes [12]. This approach allowed the control of all major joints of lower extremities and trunk and provided the capability to synthesize movements approaching a more normal gait. Increasing the number of muscles under stimulation control to 48 channels resulted in improved speed and stability of gait and the ability to do side stepping [6] and stair climbing [5]. A major factor limiting practical use of the percutaneous system is the difficulty of implanting and maintaining percutaneous electrode system [13]. Implantable systems available for clinical application are generally limited to eight channels of stimulation [14]. Due to the stimulator size at most two implants per person are feasible for implantation at this time. This study was motivated to explore the walking generated /97$ IEEE

2 24 IEEE TRANSACTIONS ON REHABILITATION ENGINEERING, VOL. 5, NO. 1, MARCH 1997 TABLE I SUBJECT INFORMATION TABLE II MUSCLES AVAILABLE IN TEST SUBJECTS by a reduced set of eight or 16 muscles from a 48-channel percutaneous FES system. The purpose of this investigation was to define the minimal set of muscles to target for implantation in a clinical system that would produce adequate walking performance. Further, a case study was performed to measure the day-to-day repeatability of gait in a paraplegic subject to predict an outcome of a 16-channel implanted FES system. II. METHODS AND MATERIALS Three paraplegic subjects (Table I) who have had up to 48 channels of percutaneous and surface muscle stimulation available to control trunk, hips, knees, and ankles (Table II) for walking were included in this study. Muscles of the trunk and posterior thigh were stimulated with surface electrodes when unavailable with percutaneous electrodes. While multiple electrodes were implanted in some muscles, only one electrode per muscle was used in this study. Once conditioned with FES, these muscles produced from 20% to 60% of normal joint moments [8]. The study subjects were proficient in using their FES system with walker support and logged many hours of walking practice. The multi-channel system of each subject was scaled down to 16 or fewer channels of various muscle combinations which were programmed for walking and tested for objective outcome measures of gait. The basic criteria used for inclusion of muscles in the scaled down patterns were as follows. First, at least one quadriceps, excluding rectus femoris, was included for each leg to brace the knees and to allow the subject to stand. Next, at least one muscle controlling hip flexion was included to bring the leg forward in a reciprocal gait. To provide subjects with some capability to move the body forward at least one hip extensor was included. To allow them to stand erect without the need to assume a C posture, trunk extensors were included in the stimulation pattern. Therefore, the basic set of muscles required a minimum of eight channels. This system required bracing of ankles with a fixed ankle foot orthosis (AFO). An alternative eight-channel system option was to control the ankle dorsiflexion and leave trunk control to the upper extremities as is usually done in surface stimulation systems. In this manner, six different eight-channel patterns were configured (Table III) and tested. Additional channels were added to the basic set of muscles to help improve stability, bring the swing leg forward, or to add power for progression. Thus, eleven 16-channel muscle combinations were generated and tested (Table IV). Each pattern was tested in at least two subjects. TABLE III MUSCLE COMBINATIONS FOR EIGHT-CHANNEL FES SYSTEMS The stimulation patterns were programmed into a microprocessor controlled stimulator [2]. The typical stimulation pattern (Fig. 1) was divided into tics representing time units for which stimulation pulse width and interpulse interval were defined. The pulse widths were varied up to 150 s at an amplitude of 20 ma for intramuscular stimulation and up to 250 satan amplitude of 100 ma for surface stimulation. The interpulse interval was varied between 30 and 80 ms. Break points were inserted at various points of the stimulation pattern at which

3 KOBETIC et al.: MULTICHANNEL FES SYSTEMS FOR AMBULATION IN PARAPLEGIA 25 TABLE IV MUSCLES COMBINATIONS TESTED FOR 16-CHANNEL FES SYSTEMS Fig. 1. Pulse width stimulation pattern for walking tested for repeatability in subject P3. Shaded areas indicate bursts of pulses at variable pulse widths. At break point number 0 (BP0), corresponding to the right heel strike (RHS), and at break point number 2 (BP2), corresponding to left heel strike (LHS), the subject presses the hand switch to initiate the left and right step, respectively. The break point 1 (BP1) and break point 3 (BP3) correspond approximately to the left toe off (LTO) and right toe off (RTO), respectively. delay times or switch functions were defined. These points corresponded approximately to gait events such as heel strike or toe-off. The processor progressed through the stimulation pattern and stopped at the break point for a pre-specified time or until the user pressed a ring-mounted switch to initiate the next step. The subject entered the stimulation pattern for walking (Fig. 1) through a gait initiation pattern usually consisting of a right step. At the break point 0 (BP0), the point of right heel strike (RHS), stimulation was provided to trunk, hip, and knee extensors bilaterally. For control of knee extension a quadriceps electrode that recruited vasti only at low pulse widths was used to eliminate the hip flexion component of rectus femoris during stance phase. The processor remained at BP0 until the subject pressed the hand switch indicating readiness for the next step. The period between BP0 and

4 26 IEEE TRANSACTIONS ON REHABILITATION ENGINEERING, VOL. 5, NO. 1, MARCH 1997 TABLE V GAIT PARAMETERS ACHIEVED WITH DIFFERENT EIGHT- AND 16-CHANNEL STIMULATION PATTERNS BP1 was where the weight shift to the right leg occurred and momentum for progression was renewed by increased activation of the right hip extensors by increasing the pulse width and decreasing the interpulse interval. At BP1 the left hip and knee extensors were turned off and hip flexors and ankle dorsiflexors were activated. The left toe off (LTO) occurred shortly thereafter. Tensor fasciae latae was used together with sartorius to balance the external rotation of the thigh [8]. Just before BP2, hip flexors were turned off and at the same time knee extensors were activated in preparation for the left heel strike (LHS). Hamstrings were also activated at BP2 to reduce the extensor moment on the knee. The gluteus maximus was activated together with erector spinae to provide stability of the hip and trunk during double support at BP2 until the subject initiated the right step with the hand switch. The stimulation to the posterior portion of adductor magnus was delayed or activated at low level since the time of the heel strike could not be predicted exactly and early activation often caused scissoring of the leg in terminal swing. Bilateral stimulation of the paraspinal muscles also provided a stabilizing effect on pelvis and was found to enhance the extension moment at the hip produced by stimulation of the hamstrings [7]. A tuning method consisting of varying the pulse width to control the amount of muscle activation and varying the stimulus burst initiation and duration of different channels was used to specify the stimulation patterns. This tuning of the stimulation patterns by trial and error [8] was done by observing video recording of the walking after each change was made. The subject was walking with a rolling walker for support over computer floor with cm grid. A video timer with one hundredth of a second resolution was superimposed on the video. The stimulation pattern producing the most normal gait as judged by the observer and the most comfortable to the subject was used for measurements of basic gait parameters of speed, cadence, stride length, and minimum and maximum trunk tilt. The trunk tilt was measured with respect to vertical and estimated from the video during the stride when the subjects was directly in front of the camera to minimize error due to parallax. These gait parameters provided an indication of the subject s ability to move forward, stability, and symmetry of gait. The same parameters were measured in one subject (P3) using the best combination of 16 muscles to establish day-today repeatability. This subject has used a multichannel FES system daily for either walking or exercise for the past 12 years. During each test he completed two consecutive walking trials along the 8 m walkway, each separated by a 2 min rest interval. After a 4 min rest he walked uninterrupted for 34 m in a hallway. Our fatigue data shows that no significant fatigue occurs during the short 8 m walks in hip flexors [9] and extensors [8], while during long 34 m walks a reduction of hip extension moment by approximately 30% can be expected [8]. This test was repeated on 21 days over a seven-week period. On six of those days, the testing was repeated twice (one session at noon and the second session about three hours later) to determine within-day repeatability. The long walks were timed automatically by the stimulator to calculate average speed. Cadence, speed, and left and right stride lengths were estimated from the video for short walks only. One way ANOVA was used to test the hypothesis that parameter means of different trials and sessions are equal. III. RESULTS The results of muscle selection on the ability to provide walking in paraplegia are summarized in Table V. The values represent the best performance of the three subjects. In general, the same pattern produced similar gait in at least two subjects. The subject s ability to move forward and stability were highly dependent on the choice of muscles included in the stimulation pattern. The eight-channel systems provided poor stability with a large trunk tilt. The speed of walking was below 0.1 m/s. Steps were inconsistent and strides less than 0.3 m long. The most common problems were scissoring and lack of hip flexion. The two common channels of stimulation were the vasti part of the quadriceps bilaterally. The gluteus maximus stimulation provided the most hip extension without scissoring. Activation of posterior portion of adductor magnus in terminal stance in

5 KOBETIC et al.: MULTICHANNEL FES SYSTEMS FOR AMBULATION IN PARAPLEGIA 27 preparation for heel strike to provide stability during loading often caused scissoring because the timing of the heel strike was difficult to predict with preprogrammed stimulation. If this muscle was activated strongly during mid stance without sufficient regulation of abduction by the upper extremities on a walker, it caused reduction in swing clearance and bumping of the stance leg with the swinging limb. Two channels of hip flexors such as sartorius or iliopsoas often provided inadequate clearance past the stance leg during swing or resulted in scissoring. The activation of sartorius for hip flexion with concurrent external rotation of the thigh and flexion of the knee caused internal rotation of the leg and made it difficult to swing through without bumping into the stance leg in early swing. Stimulation of iliopsoas alone also caused bumping during early swing and crossing over in the terminal swing. Using the remaining two channels for activation of ankle dorsiflexors produced better results than using them for trunk extensors and bracing the ankles with a fixed AFO. The stimulation of the peroneal nerve for dorsiflexion was often accompanied by some flexion reflex which had a balancing effect or enhanced the response of direct stimulation of the hip flexors. The activation of tensor fasciae latae alone did not produce enough hip flexion to advance the leg. The activation of gracilis alone produced too much adduction, thus preventing the leg from swinging through with sufficient clearance. The best walking with eight-channel FES system were obtained with muscle combination 8/1 (Table III) including: gluteus maximus for hip extension, sartorius for hip flexion, vastus lateralis/intermedius for knee extension, and tibialis anterior/peroneus longus for ankle dorsiflexion. The addition of eight channels of stimulation to the basic set of muscles consisting of quadriceps, gluteus maximus, tibialis anterior and either sartorius or iliopsoas provided significant improvements in gait. These included: increased speed of walking (up to 0.5 m/s), eliminated scissoring, and reduced anterior trunk tilt (Table IV). The tensor fasciae latae was added to improve hip flexion and to provide clearance during the swing. The terminal swing crossover was also prevented with abduction by tensor fasciae latae. In this system both ankle dorsiflexors and trunk extensors were activated. Thus, trunk tilt was reduced and foot clearance was provided during swing. The remaining channels were used for stabilization of the hip and to provide power for progression. In normal walking the power is generated mostly at the hip by extensors during loading and by the hip flexors in preswing and at the ankle during terminal stance. Either the posterior portion of the adductor magnus or the hamstrings could be added to increase hip extension. If only one was added, the posterior adductor magnus was found to be a better choice because it provided medio-lateral stability of the hip. This left one channel per side unused. If progression was poor, the hamstrings were the best choice. If hip flexion was marginal, then gracilis or one of the remaining hip flexors, either sartorius or iliopsoas, produced the best results. Increased hip flexion also provided some power for progression through ballistic swing. The stimulation of plantar flexors generally introduced instability and tended to push the subject out of balance. Their early activation often resulted in upward movement rather than in forward progression. The best combination of speed, cadence, and erect posture was obtained with muscle combination 16/5 (Table IV). This muscle combination included: erector spinae for trunk extension, gluteus maximus, hamstrings and the posterior portion of adductor magnus for hip extension, iliopsoas and tensor fasciae latae for hip flexion, vastus lateralis/ intermedius for knee extension, and tibialis anterior/peroneous longus for ankle dorsiflexion. This combination of muscles was reported by the patient as the most comfortable. Therefore, it was used for testing the day to day repeatability of gait parameters. Increasing the number of muscles up to 24 had a pronounced effect on function. In this system all hip extensors could be included which maximized the power input from hip. The medio-lateral stability was improved with addition of gluteus medius. By including plantar flexors, additional power for progression was generated. An extra channel for hip flexion further increased power input for walking and provided for increased balance during swing. The repeatability test in subject P3 using muscle combination 16/5 in a stimulation pattern (Fig. 1) showed that FES gait was symmetric with strides varying from day-to-day. The subject walked with an average cadence of 58 steps/min. There was a significant increase in speed from first ( m/s) to second ( m/s) trial. During the long walk the speed increased to m/s. The anterior trunk tilt did not change significantly between trials and varied between a minimum of to a maximum of from vertical. The stride lengths were statistically the same at m and m for left and right leg, respectively. The left stride length ( m, ) and the right stride length ( m, ) in the second trial were significantly longer than the left ( m, ) and right ( m, ) stride lengths in the first trial. However, there was no difference in stride lengths of the second trials in different sessions on the same day. Between sessions on different days, the stride length varied. One test of six long walks with 2 min rests was done to explore the limits of his FES system. These data showed no significant reduction in speed with each walk. IV. DISCUSSION The results of pattern 8/1, an eight-channel system (Table III), are in agreement with those published for four-channel surface stimulation systems [4]. These systems provide stepping at slow speeds, with short steps, and large trunk tilt. Using the withdrawal reflex, surface systems provide activation of at least eight muscles including knee extensors, hip and knee flexors and ankle dorsiflexors. The advantage of the withdrawal reflex is that it provides a balanced hip flexion that is difficult to attain with stimulation of individual hip flexors. As shown in Table V, most eight-channel patterns tested had problems with scissoring or insufficient hip flexion. Scissoring during swing was due to the adduction component of the iliopsoas electrode. When recruited by stimulation at the L2-L3 root, adductor longus and sartorius are usually recruited in addition to iliopsoas. The adductor longus causing adduction also acts as a hip extensor at larger hip flexion angles [9].

6 28 IEEE TRANSACTIONS ON REHABILITATION ENGINEERING, VOL. 5, NO. 1, MARCH 1997 During slow walking the extension component is usually not a problem. The results of the 16-channel system tests showed that hip flexion was an important component in providing gait at speeds greater than 0.2 m/s. Only sartorius could be used by itself and still provide enough clearance of the stance leg during swing. Iliopsoas by itself produced too much adduction and caused scissoring or would not clear the stance leg. Best results were attained when either sartorius or iliopsoas was used together with tensor fasciae latae to balance the leg during swing. With 16 channels of stimulation available, at least two were used for hip extension to provide assistance in progression. When there was a choice between posterior adductor and hamstring, the posterior adductor, in general, produced better results as shown by comparing patterns 16/4 and 16/9. When both posterior adductor and hamstrings were used together with gluteus maximus, stance leg clearing was marginal. In the same combination, gluteus medius instead of hamstrings produced faster walking but at increased medio-lateral sway as seen by comparison of patterns 16/7 and 16/8. The advantage of including tensor fasciae latae and the posterior portion of adductor magnus in the 16-channel walking system lies in the added ability to perform a side step. These muscles provide an important function during side stepping. The tensor fasciae latae provides abduction to move the leg to the side and after the weight transfer to the leading leg; the posterior adductors of both legs bring the feet together, thus completing the step. Including plantar flexors in the pattern increased the speed of walking by providing power input for progression at ankles. The highest speed for the 16-channel system was achieved with this combination as shown for patterns 16/10 and 16/11. Here again the combination using the posterior adductor instead of the hamstrings produced faster gait. Using plantar flexors in the patterns produced large variability in step length partly due to difficulty in timing their activation. The FES produced a symmetric gait with repeatable speed from day-to-day in a paraplegic individual who used FES daily for either walking or exercise. The results suggest a warmup or loosening effect with repeated walks as reflected in increased stride length and speed in the second trial. The subject s arm fatigue was the limiting element when FES system was used for repeated long walks. The 16-channel FES system provided repeatable day-to-day gait without the need to adjust stimulation parameters. During short walks no significant muscle fatigue occurs in well-conditioned subjects, thus adjustments in stimulation parameters were not required. Therefore, it can be concluded that an implantable 16-channel FES system with currently available technology can provide repeatable short distance walker-supported walking in paraplegia. The basic set of muscles to target should include: erector spinae for trunk extension; hamstrings, gluteus maximus and the posterior portion of adductor magnus for hip extension; tensor fasciae latae and either sartorius or iliopsoas for hip flexion; vastus lateralis/intermedius for knee extension; and tibialis anterior recruited together with peroneous longus for dorsiflexion. While in most combinations the posterior portion of adductor magnus was a better choice than hamstrings, the preferred system 16/5 includes both muscles. This provides the media-lateral stability of the hip by adductor magnus and extension of the hip by both muscles. In addition, hamstrings in terminal swing provide protection to the knee from hyperextension when coactivated with vasti. On the other hand, the rectus femoris which is a very active muscle during normal gait is excluded from stimulation pattern because its hip flexion component is difficult to control with FES during standing. The sartorius, another biarticular muscle, produces knee flexion in addition to its primary function as a hip flexor, thus reducing the need for a separate knee flexion channel. As hardware is developed that offers additional channel of stimulation, it should be incorporated in a clinical system to improve stability of the body and control of the joints to allow other functions including stair climbing. In addition, further development of nerve electrodes, which allow selective stimulation of fascicles in peripheral nerves, can reduce the number of channels needed and increase the amount of muscle recruited by each channel, thus providing greater joint muscle moments than currently available. REFERENCES [1] T. Bajd, A. Kralj, R. Turk, H. Benko, and J. Sega, The use of a four channel electrical stimulator as an ambulatory aid for paraplegic patients, Phys. Therapy, vol. 63, no. 7, pp , [2] G. Borges, K. Ferguson, and R. Kobetic, Development and operation of portable and laboratory electrical stimulation system for walking in paraplegic subjects, IEEE Trans. Biomed. Eng., vol. 36, pp , July [3] J. Holle, M. Frey, H. Gruber, H. Kern, H. Stohr, and H. Thoma, Functional electrostimulation of paraplegics-experimental investigations and first clinical experience with an implantable stimulation device, Orthoped., vol. 7, p. 1145, [4] E. Isakov, J. Mizrahi, and T. Najenson, Biomechanical and physiological evaluation of FES-activated paraplegic patients, J. Reh. Res. Dev., vol. 23, no. 3, pp. 9 19, [5] R. Kobetic, S. G. Carroll, and E. B. Marsolais, Paraplegic stair climbing assisted by electrical stimulation, in Proc. 39th ACEMB Conf., Baltimore, MD, Sept. 1986, p [6] R. Kobetic, J. M. Pereira, and E. B. Marsolais, Electromyographic study of the side step for development of electrical stimulation patterns, in Proc. 9th RESNA Conf., Minneapolis, MN, 1986, pp [7] R. Kobetic, S. G. Carroll, and E. B. Marsolais, Functional electrical stimulation of hip extension and abduction affected by activity of the trunk, in Proc. 10th RESNA Conf., San Jose, CA, 1987, pp [8] R. Kobetic and E. B. Marsolais, Synthesis of paraplegic gait with multichannel functional neuromuscular stimulation, IEEE Trans. Rehab. Eng., vol. 2, pp , June [9] R. Kobetic, E. B. Marsolais, and P. C. Miller, Function and strength of electrically stimulated hip flexor muscles in paraplegia, IEEE Trans. Rehab. Eng., vol. 2, pp , Mar [10] A. Kralj, T. Bajd, R. Turk, J. Krajnik, and H. Benko, Gait restoration in paraplegic patients: A feasibility demonstration using multichannel surface electrodes FES, J. Rehab. Res. Dev., vol. 20, pp. 3 20, [11] K. H. Lee and R. Johnston, Electrically induced flexion reflex in gait training of hemiplegic patients: Induction of the reflex, Arch. Phys. Med. Rehab., vol. 57, p. 311, [12] E. B. Marsolais and R. Kobetic, Functional electrical stimulation for walking in paraplegia, J. Bone Joint Surgery, vol. 69-A, no. 5, pp , [13] A. Scheiner, G. Polando, and E. B. Marsolais, Design and clinical application of a double helix electrode for functional electrical stimulation, IEEE Trans. Biomed. Eng., vol. 41, pp , May [14] B. Smith, P. H. Peckham, M. W. Keith, and D. D. Roscoe, An externally powered, multichannel, implantable stimulator for versatile control of paralyzed muscle, IEEE Trans. Biomed. Eng., vol. BME-34, pp , July [15] M. Solomonow, R. Baratta, S. Hirokawa, N. Rightor, W. Walker, P. Beaudette, H. Shoji, and R. D Ambrosia, The RGO generation II: Muscle stimulation powered orthosis as a practical walking system for thoracic paraplegics, Orthoped., vol. 12, no. 10, pp , 1989.

7 KOBETIC et al.: MULTICHANNEL FES SYSTEMS FOR AMBULATION IN PARAPLEGIA 29 Rudi Kobetic (M 91) was born in Maribor, Slovenia. He received both the undergraduate and graduate degrees in biomedical engineering from Case Western Reserve University, Cleveland, OH. Since 1980, he has been with the Motion Analysis Laboratory at the Cleveland Veterans Affairs Medical Center developing FNS systems for restoring functional movements in paralyzed patients. Ronald J. Triolo (S 77 M 86), for a photograph and biography, see this issue, p. 22. E. Byron Marsolais (M 88) received the M.D. and Ph.D. degrees from the University of Iowa, Iowa City. He interned at St. Luke s Hospital, New York, NY, and did his orthopaedic residency at the University of Iowa. His Ph.D. studies were in the engineering fields of mechanics and hydraulics. In 1970, he joined Case Western Reserve University Medical School, Cleveland, OH, where he began his association with the Veterans Affairs Medical Center. He is Director of the Rehabilitation Division of the Department of Orthopaedic Surgery at Case Western Reserve University, a Senior Instructor in orthopaedic surgery, Associate Professor of Orthopaedic Medicine and Rehabilitation Service at the Veterans Affairs Medical Center since 1972, and Director of the Motion Study Laboratory since He led the University Hospitals Problem Low Back Team from 1975 to 1985 and the Veterans Affairs Medical Center Back Clinic from 1975 to present. In the past years, he has been increasingly involved, along with other investigators at Case Western Reserve University, in research on functional electrical stimulation of paralyzed muscle. He has headed a 25-member multidisciplinary team at the Cleveland Veterans Affairs Medical Center since 1983, working on restoration of mobility in paralyzed individuals.