Abstract

The purpose of this study was to compare the effect on paraplegic gait speed and stability of three hybrid systems utilizing a trunk-hip-knee-ankle-foot (THKAFO) orthosis and multi-channel functional electrical stimulation with implanted electrodes. H1 was a modified isocentric reciprocal gait orthosis (IRGO) with knees unlocked, ankles fixed and hips reciprocator coupled. H2 had a THKAFO instrumented with hip and knee wrap spring clutches that provided free extension and solenoid controlled flexion. The third prototype (H3) had knee joints instrumented with a multi-disk brake engaged via a wrap spring clutch. The ankles were fixed and hips were reciprocator coupled. A microprocessor controlled muscle stimulation and solenoid activation. Two volunteers tested the systems. H1 and H3 with a reciprocator provided erect posture but slow walking speed due to reciprocator limited step length. H2 achieved a faster walking speed; but with excessive trunk tilt. A hybrid system is needed that allows joint motion without excessive trunk tilt.

1. Introduction

Standing and short distance walking with multi-channel functional electrical stimulation (FES) in paraplegia has been proven feasible [7,5]. However, the FES standing time and walking distance have been limited because of instability resulting from poor muscle control or inadequate joint moments produced by electrical stimulation as muscles fatigue requiring considerable effort by the user and high metabolic energy consumption [6]. An approach to improve stability during standing and walking has been the use of hybrid orthoses – a combination of FES and bracing.

Several groups have designed hybrid systems to improve joint control [3], add supportive moments during stance [4] and reduce effects of muscle fatigue [1]. One such system used a reciprocating gait orthosis (RGO) with locked knees and reciprocally coupled hips in a 1:1 ratio so that hip flexion was powered by contra-lateral hip extension [9]. FES of the hip extensors (hamstrings) and flexors (rectus femoris) augmented the volitional effort of the user. Because the knees were locked in this system, the walking required considerable side to side movements so the foot could clear the floor during swing. In addition, the coupling between the hips restricted hip flexion resulting in short step length. By removing some of the joint motion constraints, significant improvements in gait were attained. Increasing the coupling ratio between hip flexion and extension to 2:1 and unlocking the knee joint during swing were the most effective strategies for increasing walking speed and reducing energy cost [8]. The addition of an electrical knee lock system can provide free movement of the leg during swing and stability during stance [2] without requiring stimulation [4], thus reducing knee extensor muscle fatigue. A hybrid system using controllable friction brakes at the knee and hip joints and four channels of surface FES significantly reduced the duty cycle of quadriceps stimulation. However, the system was too bulky and required too much power to be practical outside the laboratory [1]. The use of surface stimulation also limited the number of channels that could be used in a practical system and excluded deep muscles such as hip flexors that are important in walking.

This study was designed to determine how different components of a hybrid system affect the speed and stability of the gait. We implemented a hybrid system that selectively freed motion of the braced joints and provided power for walking with FES. The implanted FES of all major muscles controlling trunk, hips, knees and ankles provided power to stand up and to move the body forward during walking. Postural stability was provided by the trunk-hip-knee-ankle-foot-orthosis (THKAFO) with programmable joint locks that could be fixed or free to move during different phases of gait.

2. Methods

Two male subjects, 55 and 40 years old, with T-9 and T-7 complete thoracic spinal cord injury respectively, volunteered to participate in hybrid orthosis evaluation. Subject S1 and S2 had 28 and 15-channel FES systems, respectively. They had used their FES-only systems for exercise and walking with stand-by assistance for over 16 years.

H1 included 8 or 16 channels of FES and a modified commercial isocentric reciprocal gait orthosis (IRGO THKAFO. The ankle joints were locked, the knee joints were unlocked (free to move under FES control), and the hip reciprocator could be either engaged or disengaged.

In H2 the reciprocator was removed and the THKAFO augmented with electrically locking hip and knee joints. Commercially available mechanically actuated wrap spring clutches (Warner Electronic, Pitman NJ) were selected at the hips (PSI-5, rated at 28.25 Nm) and knees (PSI-2, rated at 2.8 Nm, or PSI-4, rated at 13.6 Nm). An eccentric knee joint took advantage of the ground reaction to minimize the clutch torque requirement.

The clutches were engaged by a solenoid that rotated a control collar to unwrap the spring from the hub, allowing free movement of the joint in flexion and extension. With the solenoid inactive, the joint was free to move in extension but was locked against flexion.

The H2 hybrid system used 26 channels of FES and required hip extensor muscles to bring hip into extension to maintain erect posture. Solenoid control was provided by a microprocessor controller and synchronized with the stimulation pattern for walking.

In H3 the knee joints were instrumented using our design for a locking joint that incorporated a multi-disk brake engaged via a wrap spring clutch. The hips were coupled with a reciprocating mechanism where flexion in one hip caused contra-lateral extension. The H3 hybrid system used 15 channels of FES with ankles fixed at neutral with a molded AFO.

Hybrids H1 and H2 were evaluated by subject S1 and hybrid H3 by subject S2, plus their FES-only systems. Subjects used either quad canes or a rolling walker with two or four wheels for balance and support. Basic gait parameters of speed, cadence, stride length and minimum and maximum trunk tilt were estimated from video.

3. Results

The hybrid systems with reciprocating mechanism engaged (H1 and H3) provided minimal trunk tilt, but slow speeds. On the other hand, H2 allowed much faster speed but had excessive trunk tilt. All prototypes provided effective sit-to-stand maneuvers and an erect posture while standing. In H2 and H3, the clutches were engaged to prevent collapse of the hips and knees, once standing was achieved with FES.

Only the H1 hybrid with engaged reciprocating mechanism provided enough stability and balance for walking with either quad canes or the rolling walker with four wheels. Walking with quad canes and eight channels of stimulation produced a stable erect gait with speed of 0.23m/s. Adding channels of stimulation increased the speed of walking but often caused jerky movements and reduced stability. The most stable arrangement involved using a rolling walker with two wheels. This system allowed walking with FES-only or with any of the hybrid systems. Using this walker and the H1 hybrid, subject S1 was able to walk with erect posture at speed of 0.32m/s. With the reciprocating mechanism disengaged or with FES-only system his posture was no longer erect and his trunk tilt was between 23 and 40 degrees. The speed was slightly faster with the reciprocator disengaged using the H1 hybrid configuration (0.49m/s) compared to FES-only walking (0.44m/s). This could be accounted for by the medio-lateral stability provided by the braces.

The H2 hybrid system with mechanically controlled joints allowed walking at a speed of 0.44m/s. This speed was slightly slower than the comparable FES-only walking. The subject reported that the H2 hybrid system provided good medio-lateral stability. The video recording showed slight reduction in trunk tilt from FES-only walking. However, the trunk tilt was considerably greater than that observed in the H1 and H3 hybrid systems with the reciprocating mechanism engaged.

Subject S2 evaluated the H3 hybrid configuration. The reciprocator was engaged upon standing and provided stable erect posture with less than 8 percent of body weight taken up by the arms. Walking with the H3 hybrid prototype was slow (0.33m/s) due to the step length limitation imposed by the reciprocating mechanism.

The FES-only gait provided the fastest speed of walking when compared to hybrid systems. However, the posture and stability were significantly compromised, as indicated by the large trunk tilt, and the need for a two-wheeled walker which provided more support then any other walking aid.

4. Summary and Conclusions

Of the three hybrid configurations tested, the reciprocator-based systems H1 and H3 provided the most erect posture and stable gait. While the walking was slow due to the limited step length imposed by the reciprocating mechanism, the gait was sufficiently stable to allow walking with quad canes or with a four-wheeled walker. The hybrid configuration H2 and FES-only walking posed no limitation on the step length and provided much faster walking. However, the user was unable to maintain an erect posture and stability was compromised, requiring the use of a stable two-wheeled walker.

Replacing the hip reciprocator with controllable joints at the hips and knees (hybrid configuration H2), greatly improved joint motion range to levels similar to FES-only walking. However, the resulting improvement in step length was accompanied by less erect posture during walking. This may have been due to two factors. First, any decrease in hip extensor force with muscle fatigue failed to bring the hip into full extension and compromised the erect posture. Second, the clutches required a few degrees of flexion to engage the spring and further rotation. Therefore, hip hyperextension was required to compensate for backlash of the clutch to keep the subject erect. Optimal hip and trunk strength and timely activation of hip extensors were necessary to reduce muscle fatigue and to bring the hip into extension so that the brace could take over and stabilize the hip. In the reciprocator-based hybrid configuration, the activation timing and muscle strength were not as critical because the stimulation of hip flexors at the opposite leg helped in bringing the stance hip into extension.

A selectively engaging reciprocating mechanism may be able to provide both erect posture and unencumbered motion. A hybrid system with this mechanism would be self- righting, with both the extensors and the contra-lateral hip flexors contributing support forces to keep the body erect. Unlike current reciprocal gait orthoses, however, this configuration would not limit step length during walking or step height during stair climbing. The reciprocator would be engaged through the initial part of contra-lateral swing phase to provide additional force in bringing and keeping the stance hip in extension. During the mid-swing, the stance hip would be locked in extension and the reciprocating mechanism would be disengaged, allowing the swing leg to extend without the constraint of the reciprocator. While the reciprocator is disengaged, the stance hip would be kept in extension with a clutch-controlled joint lock. At heel strike, the reciprocating mechanism would be re-engaged to prepare for the next step. This next-generation hybrid system is currently under development.

Acknowledgment

This work was supported by the NIH grant R01 NS33287-05 and the Rehabilitation Research and Development Service of the Department of Veterans Affairs.

References

[1] M. Goldfarb, W. K. Durfee. “Design of a Controlled-Brake Orthosis for FES-Aided Gait.” IEEE Trans Rehab Eng. 1996;4(1):13-24.

[2] S. E. Irby, K. R. Kaufman, J. W. Mathewson, D. H. Sutherland. “Automatic control design for a dynamic knee-brace system.” IEEE Trans Rehab Eng. 1999;7(2):135-139.

[3] S. E. Irby, K. R. Kaufman, R. W. Witra, D. H. Sutherland. “Optimization and application of a wrap-spring clutch to a dynamic knee-ankle-foot orthosis.” IEEE Trans Rehab Eng. 1999; (7)2:130-133.

[4] H. Kagaya, Y. Shimada, K. Sato, M. Sato, K. Iizuka, G. Obinata. “An electrical knee lock system for functional electrical stimulation.” Arch Phys Med Rehabil. (77):870-873, 1996.

[5] A Kralj, T. Bajd, Functional Electrical Stimulation: Standing and Walking After Spinal Cord Injury. Boca Raton, FL: CRC Press; 1989.

[6] E. B. Marsolais, B.G. Edwards. Energy costs of walking and standing with functional neuromuscular stimulation and long leg braces. Arch Phys Med Rehabil. 1988;69:243-249.

[7] E. B. Marsolais, R Kobetic. “Functional Electrical Stimulation for walking in paraplegia,” J Bone Joint Surg (Am). June 1987;69-A(5):728-733.

[8] E. B. Marsolais, R. Kobetic, G. Polando, K. Ferguson, S. Tashman, R. Gaudio, S Nandurkar, H. R. Lehneis. “The Case Western Reserve University hybrid gait orthosis.” J Spinal Cord Med. 2000;23(2):100-108.

[9] M. Salomonow, R. Baratta, S. Hirokawa, N. Rightor, W. Walker, P. Beaudette, H. Shoji, R. D’Ambrosia. “The ROG generation II: Muscle stimulation powered orthosis as a practical walking system for thoracic paraplegics.” Orthopedics. 1989;12(10):1309-1315.