September/October 2016

Neuroprosthetic Implant Improves Walking Ability in Stroke Patients
By Nathaniel Makowksi, PhD; Rudi Kobetic, MS; Lisa Lombardo, PT; Gilles Pinault, MD; Stephen Selkirk, MD, PhD; Kevin Foglyano, BSE; and Ronald Triolo, PhD
Today's Geriatric Medicine
Vol. 9 No. 5 P. 28

Stroke is a leading cause of long-term disability in the United States, with about 6.6 million people currently living with stroke and 800,000 new cases each year.1 Although these individuals benefit from therapy, about one-third require assistance to walk after completing their rehabilitation program.1 In addition to reducing functional independence, stroke also can result in reduced mobility, which can have negative effects on the cardiovascular system and be socially isolating.

Neuromuscular electrical stimulation, an approach that uses small electrical currents to stimulate nerves and activate muscles to produce useful movements, has been applied to stroke patients since the 1960s to produce ankle dorsiflexion via peroneal nerve stimulation,2 raising the patient's foot while he or she takes a step. Drop foot stimulators can improve toe clearance during walking, but many patients have more severe impairments that are not met by this approach. While this approach has shown some benefits,3 it does not address impairments at the hip and knee, or at the ankle during push off.

Drop foot stimulators commonly use electrodes that are placed on top of the skin to stimulate the nerve. However, surface stimulation is not well suited as a long-term approach when additional joints are affected. Some muscle functions, such as hip flexion, can be difficult to selectively stimulate with surface electrodes, and placement of several external electrodes each day is not practical for users with limited manual dexterity. In addition, stimulation of sensory fibers between the electrode and the target nerve can also produce uncomfortable sensations. Thus, the development of implantable multichannel stimulation systems provides an opportunity to control multiple joints, which could have a greater impact on walking ability. Initial feasibility of these multichannel implanted systems has primarily been demonstrated in patients who have sustained a spinal cord injury,4 but implanted neuroprostheses could also benefit patients who have had a stroke.

Demonstrated Efficacy
A recent case study demonstrated the initial feasibility of an implanted multijoint neuroprosthesis to improve walking after stroke.5 The participant was a household ambulator two years poststroke with weakness at the hip, knee, and ankle on the affected side. During surgery an eight-channel stimulator was implanted beneath the skin and adipose tissue in the abdominal area on the affected side. Electrodes were implanted near the nerves of the target muscles, which were chosen based on the participant's deficits. Electrode leads were tunneled under the skin to connect the electrodes to the stimulator. After allowing time to heal after the surgery, the participant was given a set of exercise stimulation patterns for home use to strengthen his muscles and improve fatigue resistance.

Following conditioning, a stimulation pattern to coordinate stimulation with his walking was developed. A heel switch in his shoe controlled the stimulation pattern. Taking weight off the heel initiated stimulation to the hip and knee flexors and ankle dorsiflexors for swing. After a delay, the hip and knee flexors were ramped down, and the knee extensors were activated to extend the knee for heel strike to complete the step. At heel strike, the hip extensors were activated for hip stability while the ankle dorsiflexors were ramped down. During stance, the knee extensors were ramped down, followed by the hip extensors after some delay. Using this stimulation pattern, the participant underwent supervised gait training with stimulation.

The participant's walking was assessed during three conditions: volitional walking at baseline (no device), volitional walking after training (device off), and walking with implanted stimulation after training (device on). Comparing differences between these conditions provides insight into the therapeutic and neuroprosthetic effects of multichannel stimulation. Therapeutic effects are improvements that result from using the device and remain present after the device is turned off. Neuroprosthetic effects are improvements that are present only when stimulation is turned on, as compared with when it is turned off. Assessments compared walking speed with the 10-meter walk test and six-minute timed walk test.

In addition, quantitative motion analysis was used to evaluate spatial-temporal characteristics. Outcomes showed improvements in walking speed corresponding to changes in spatial-temporal characteristics of gait. The participant had modest therapeutic improvements in walking speed (0.1 m/s) but had a significant neuroprosthetic improvement (0.4 m/s). Gait changes from the neuroprosthetic effect were more pronounced, resulting in more symmetric and dynamic walking.

Though walking speed is important and is used as an indicator of community ambulation, patients using these types of systems in the future may perceive additional benefit beyond improved walking. Being able to walk faster and longer may enable patients to exercise more, benefiting their cardiovascular health and increasing social interaction and societal participation.

Future Prospects
While there was a significant improvement in walking speed for this participant, additional work is needed to identify the type of patients who would benefit from this type of intervention. There are also opportunities to improve the technology. Techniques developed by Foglyano et al have shown that accelerometers can effectively predict gait events to coordinate stimulation with the user's own movements.6 Similar approaches with sensors may enable the controller to better coordinate with users' own efforts to create more natural movements. The current approach focuses on movements in the plane of progression and relies on the user to respond to external perturbations encountered in daily use. The addition of sensor-based feedback control of stimulation may improve balance and help the user react to disturbances.

The implanted stimulation should be tailored to add or complement an individual's own volitional effort during walking. For instance, stimulation could benefit patients who have difficulty flexing their knees during swing and are at increased likelihood of catching their toes and falling. Similarly, it may be beneficial for those who are unable to fully extend the knee at heel strike or have difficulty flexing the hip, resulting in shorter steps and slow walking. Presently, patients whose impairment results from weakness or paresis would be the most appropriate candidates. Stimulation may be able to overcome mild to moderate hypertonia, but in those with significant hypertonia, stimulation may provide limited benefits.

Although the recent case study was performed in the chronic phase after stroke, some patients may benefit from multijoint stimulation during the acute phase after stroke. Having the opportunity to learn to coordinate their movements with stimulation could help prevent some of the compensatory strategies people develop to compensate for their impairments. The neuroprosthesis could be incorporated into gait training during the rehabilitation period. This may allow individuals to begin walking earlier in the rehabilitation process and delay or prevent the need for bracing. Similar to current efforts with robotics, the neuroprosthesis could increase the number of movement repetitions patients could complete, as well as improving movement quality. Stimulation provides the added benefit that the patient's own muscles provide movement assistance. As the patient experiences improvements during the rehabilitation period, stimulation patterns could be adapted to meet their needs with the patient prepared to use the system during daily life following completion of the therapy.

Additional factors affect eligibility for this type of intervention. It would not be appropriate, for example, for patients with severely impaired cognition and communication, as they need to understand the risks and be able to communicate with therapists. Uncompensated hemineglect would suggest that a patient may be unable to respond to unexpected movements. Comorbidities such as poor cardiovascular health that increase the risk of surgery would also preclude patients from participating.

Patient populations who could benefit from implanted multijoint stimulation are not limited to those who have suffered a stroke. Patients who have sustained a spinal cord7,8 or traumatic brain injury or who have multiple sclerosis and have difficulty walking could also benefit from this type of approach with stimulation providing walking assistance.

— Nathaniel Makowksi, PhD, is an instructor in the department of physical medicine and rehabilitation at MetroHealth Medical Center and Case Western Reserve University's School of Medicine and an investigator in the Cleveland Functional Electrical Stimulation Center.

— Rudi Kobetic, MS, is an investigator at the Louis Stokes Cleveland VA Medical Center and a principal investigator in the Advanced Platform Technology Center.

— Lisa Lombardo, PT, is a physical therapist at the Louis Stokes Cleveland VA Medical Center.

— Gilles Pinault, MD, is a surgeon at the Louis Stokes Cleveland VA Medical Center and a principal investigator in the Advanced Platform Technology Center.

— Stephen Selkirk, MD, PhD, is a neurologist at the Louis Stokes Cleveland VA Medical Center, whose clinical focus centers on the care of patients with spinal cord injury, advanced multiple sclerosis, and amyotrophic lateral sclerosis (ALS). He conducts research on neurodegeneration and central nervous system repair as well as on clinical mechanisms to enhance survival and quality of life in ALS patients and to enhance and restore ambulation in patients with neurological disorders.

— Kevin Foglyano, BSE, is a biomedical engineer at the Louis Stokes Cleveland VA Medical Center, who specializes in implementing neuroprostheses for standing, walking, and trunk control; motion capture data analysis; algorithm development; and exoskeletal bracing.

— Ronald Triolo, PhD, is a professor of orthopaedics and biomedical engineering at Case Western Reserve University and a senior career scientist with the VA's Rehabilitation Research & Development Service. He is the executive director of the VA's Center for Advanced Platform Technology and directs the Motion Study Laboratory of the Louis Stokes Cleveland VA Medical Center, where he pursues research in the development and clinical application of neuroprostheses and restorative technologies, limb prosthetics and orthotics, dynamic exoskeletons, musculoskeletal biomechanics and the control of human movement, rehabilitation engineering, and the assessment of assistive technology.

References
1. Mozaffarian D, Benjamin EJ, Go AS, et al. Heart disease and stroke statistics — 2016 update: a report from the American Heart Association. Circulation. 2016;133(4):e38-e360.

2. Liberson WT, Holmquest HJ, Scot D, Dow M. Functional electrotherapy: stimulation of the peroneal nerve synchronized with the swing phase of the gait of hemiplegic patients. Arch Phys Med Rehabil. 1961;42:101-105.

3. Everaert DG, Stein RB, Abrams GM, et al. Effect of a foot-drop stimulator and ankle-foot orthosis on walking performance after stroke: a multicenter randomized controlled trial. Neurorehabil Neural Repair. 2013;27(7):579-591.

4. Ho CH, Triolo RJ, Elias AL, et al. Functional electrical stimulation and spinal cord injury. Phys Med Rehabil Clin N Am. 2014;25(3):631-654.

5. Makowski NS, Kobetic R, Lombardo LM, et al. Improving walking with an implanted neuroprosthesis for hip, knee, and ankle control after stroke [published online May 26, 2016]. Am J Phys Med Rehabil. doi: 10.1097/PHM.0000000000000533.

6. Foglyano KM, Schnellenberger JR, Kobetic R, et al. Accelerometer-based step initiation control for gait assist neuroprosthesis. J Rehabil Res Dev. In press.

7. Nataraj R, Audu ML, Triolo RJ. Simulating the restoration of standing balance at leaning postures with functional neuromuscular stimulation following spinal cord injury. Med Biol Eng Comput. 2016;54(1):163-176.

8. Bailey SN, Hardin EC, Kobetic R, Boggs LM, Pinault G, Triolo RJ. Neurotherapeutic and neuroprosthetic effects of implanted functional electrical stimulation for ambulation after incomplete spinal cord injury. J Rehabil Res Dev. 2010;47(1):7-16.