Published: June 16th, 2016
Training a person with paralysis to ambulate using a powered exoskeleton may present challenges. The goals are to present the candidate selection criteria and the training procedures for exoskeletal-assisted walking and other mobility skills that can be progressed as the participant's skill level improves.
Powered exoskeletons have become available for overground ambulation in persons with paralyses due to spinal cord injury (SCI) who have intact upper extremity function and are able to maintain upright balance using forearm crutches. To ambulate in an exoskeleton, the user must acquire the ability to maintain balance while standing, sitting and appropriate weight shifting with each step. This can be a challenging task for those with deficits in sensation and proprioception in their lower extremities. This manuscript describes screening criteria and a training program developed at the James J. Peters VA Medical Center, Bronx, NY to teach users the skills needed to utilize these devices in institutional, home or community environments. Before training can begin, potential users are screened for appropriate range of motion of the hip, knee and ankle joints. Persons with SCI are at an increased risk of sustaining lower extremity fractures, even with minimal strain or trauma, therefore a bone mineral density assessment is performed to reduce the risk of fracture. Also, as part of screening, a physical examination is performed in order to identify additional health-related contraindications.
Once the person has successfully passed all screening requirements, they are cleared to begin the training program. The device is properly adjusted to fit the user. A series of static and dynamic balance tasks are taught and performed by the user before learning to walk. The person is taught to ambulate in various environments ranging from indoor level surfaces to outdoors over uneven or changing surfaces. Once skilled enough to be a candidate for home use with the exoskeleton, the user is then required to designate a companion-walker who will train alongside them. Together, the pair must demonstrate the ability to perform various advanced tasks in order to be permitted to use the exoskeleton in their home/community environment.
Many persons with spinal cord injury (SCI) are unable to stand and ambulate with or without the use of an assistive device or physical assistance. For centuries, the only mobility option for those with severe SCI has been the wheelchair1. During the past few decades, persons with SCI have had the option to supplement their mobility by using passive orthotic devices such as a variety of reciprocating gait orthosis (RGO)2-7. These devices, however, have not become more widely used due to the physical effort required by the user to ambulate using these devices. The RGOs also have limitations in the ability to climb stairs, stand up, and sit down3,7. Efforts have been made to enhance the efficiency of these devices by incorporating Functional Electrical Stimulation (FES) to power the movement and help facilitate the forward swinging of the limb; however, these efforts have not progressed beyond concepts or prototypes8-12. In the 1970s, motors were incorporated with an orthosis to power the movement of the hip and knee joints and was successful in allowing a person with SCI to take steps13. However, inadequate battery and computer technology of the time limited the range of the device, and further development was abandoned10,13.
With recent technological advancements, several powered exoskeletons have been developed to enable persons with various pathologies to ambulate overground. These powered exoskeleton devices have been studied in persons with stroke14,15, persons with complete and incomplete SCI16-24, and other persons with disabilities causing reduced control of their lower extremities25-27. Although the devices differ, each one requires training and practice by the user for safe performance. Three of the referenced devices require the use of crutches to ambulate and maintain balance. The fourth one maintains balance and stability because of its large footplate and mass which enlarges the base of support and lowers the center of gravity20. The three devices that require crutching utilize the same principles even though there are some variations with the mechanics and methods of controlling the desired actions due to differences in the design of the devices.
A training program was developed at the James J. Peters VA Medical Center (JJPVAMC), Bronx, NY by a group of researchers consisting of a biomedical engineer, physiologist, physiatrist, exercise physiologist, neurologist and physical therapists. The training program was developed with one specific powered exoskeleton previously described 17,18 but it incorporates sets of skills that are applicable to other powered exoskeletons which require a set of crutches to maintain balance. All potential participants were screened prior to participating in the progressive training program. The importance of screening in persons with SCI is to ensure absence of contraindicated medical complications that may inhibit the safe use of these devices. One area of concern is low bone mineral density (BMD). Persons with SCI suffer dramatic bone loss immediately after injury28,29 which may continue throughout their lives30. This loss of BMD results in a high risk of long bone fractures. Currently, there is no effective treatment to mitigate bone loss for those with complete motor SCI. In addition, an established fracture threshold for person with SCI does not exist, but efforts have been made to identify criteria which may be used as a guide31-33 along with clinical judgement and fracture history. Other common contraindications may be treated and resolved, such as limited range of motion (ROM)34 and pressure ulcers35. Each of the different powered exoskeleton may require different conditions for eligibility, such as ROM criteria, to be a candidate to use the device, most of which have been described17-19,21,22,36.
Once a person has successfully passed all of the screening criteria, fitting the device to the user and training may proceed. Proper fitting of the device is important to avoid inappropriate contact of the lower extremities with the exoskeleton because poor fitting may lead to bruising and/or skin abrasions16. Users may have limited or no lower extremity sensation and proprioception; this lack of sensory and tactile feedback from the feet can contribute to an overall lack of awareness of their center of balance, slowing the user's ability to master the device. This lack of awareness of the center of balance may also lead to challenges with appropriate weight shifting such as difficulty in gauging the extent of the forward and lateral shift necessary during the gait cycle and inappropriately timed weight shifting, resulting in excess use of weight bearing on the arms and crutches for balance maintenance. Once the basics mechanisms of standing balance and weight shifting are acquired, the user is taught to walk in the device. Multiple sessions are needed to improve walking and other mobility skills. Initially, surfaces that are flat and smooth within the medical center are used for training. However, with improved skill level, the user is challenged with incrementally more difficult tasks by introducing different walking surfaces such as carpet, asphalt, concrete, grass, and unleveled surfaces with different degrees of slopes.
The purpose of this manuscript is to report the screening criteria, proper fitting and training procedures for use a powered exoskeleton for overground walking. This program was developed for one device specifically, which is described by others16-18, but it addresses aspects and challenges that are common for staff trainers and persons with SCI who participate in exoskeletal-assisted walking programs which may use another powered exoskeleton. Certain aspects of this protocol are specific to the device used at the JJPVAMC. Additionally, some of the components of the training program were developed by the manufacture which includes orientation of the device components, basic guidelines for a proper fit and basic standing and sitting skill instructions. The researchers at the JJPVAMC developed all training activities performed once the user is standing up. These include enhancement of the standing and sitting training instructions, standing balance skills, indoor walking progression skills, outdoor walking progression skills, and other mobility tasks for reaching, stopping, turning, and various types of door/threshold navigation.
Note: The training protocol described within this manuscript was developed during a pilot project entitled: "The ReWalk Exoskeletal Walking System for Persons with Paraplegia" registered with ClinicalTrials.gov identifier NCT01454570. Developing a training program was not the objective of this pilot project however; the training program evolved during the course of conducting this study. The study protocol and informed consent form were reviewed and approved by the JJPVAMC's Institutional Review Board (IRB). The entire study and the procedures were explained to each study participant. The potential participant was given the opportunity to ask questions and was encouraged to take as much time as needed prior to consenting.
1. Participant Recruitment
Note: The fitting procedures were developed by the manufacture of the device. The methodology of fitting a person to the device will also vary among the different exoskeletons. Clinicians should refer to each of the specific manufacture's procedures.
Note: The donning procedures were developed by the manufacture of the device. The methodology of donning a person to the powered exoskeleton may vary among the different devices and clinicians should refer to the manufacture's procedures.
Note: The procedure to stand-up was developed by the manufacture of the device and may vary among the different exoskeletons. Clinicians should refer to the manufacture's procedures.
5. Standing Balance
Note: The standing balance procedures were developed by the researchers at the JJPVAMC. There may be some procedures that are specific to the device used but most of the procedures should translate to other powered exoskeletons.
Note: The walking procedures are a mixture of procedure developed by the staff at the JJPVAMC and the manufacture of the device. The mechanism of walking inbuilt into the powered exoskeleton and the dual crutch pattern used in the device was developed by the manufacture; however the approach of teaching the participant how to properly execute the walking, mechanism of providing assistance and the outcome measure used to record the level of assistance was the efforts of the researchers at the JPVAMC. Although, some procedures are specific to the powered exoskeleton used, most of the procedures are translatable to other powered exoskeletons that use crutches to maintain balance.
7. Progressive Goals of Mobility Training
Note: The goals of the mobility training were developed at the JJPVAMC and incorporated within the criteria for evaluating proficiency to use the powered exoskeleton in the home environment by the manufacture.
8. Assessments of Walking
Note: The walking assessments used are standard clinical tests that have been established by others.
Note: The procedures to sit down were developed by the manufacture of the device and may vary among the different exoskeletons. Clinicians should refer to the manufacture's procedures.
Note: The doffing procedures were developed by the manufacture of the device. The methodology of doffing the powered exoskeleton may vary among the different devices. Clinicians should refer to the manufacture's procedures.
The following measurements are obtained throughout the training. Two handed and one handed crutch balance skills are each assessed for 1 min as "able" or "not able" to maintain balance (Figure 2). Walking assessments for time and distance are obtained throughout the training sessions using the 6MWT, 10MWT and the TUG. Exoskeletal-assisted walking on commonly encountered surfaces are tested indoors (Figures 3 and 4) and outdoors (Figures 5-6). Other mobility skills such as navigating doors (Figures 7 and 8), reaching over head into a cabinet (Figure 9) and sitting outside on a park bench (Figure 10) are assessed as "able" to perform or "not able" to perform.
Average walking speeds during the 10MWT in 10 session intervals for the first 60 sessions are depicted (Figure 11). This graph shows participants have varying initial ability to use the powered exoskeleton and varying rates of improvement among users. The average ± standard deviation of the best fit line's slope is 0.0048 ± 0.004 m/sec and values ranged from 0.00026 to 0.015 m/sec. This indicates that although each participant improved at variable rates they walked an average of 0.0048 m/sec faster each session. The average ± standard deviation of the best fit intercept is 0.16 ± 1.8 m/sec and the values ranged from -0.026 to 0.50 m/sec. This indicates that on average participants have an average initial speed of 0.16 m/sec; with some participants having almost no ability to ambulate and others have a very good ability in the early stages of training.
Trainer assistance affects performance; those who need a greater level of assistance walk slower than those who are more proficient and independent in using the system18. The three walking test measurements, although similar, provide different proficiency information. The 10MWT provides an indication of the best effort for speed (m/sec) that the user is able to ambulate in the device. The 6MWT distance, when converted to speed in m/sec, provides an average walking speed and is an indication of the consistency of walking in the exoskeleton. Since the timer continues when the user accidently stops walking, the speed from a 6MWT that is closer to the best effort 10MWT indicates that the person had consistent walking and fewer stops. The TUG requires many skills to be performed in consecutive combination. The TUG is a measure of the person's overall ability to incorporate standing up, walking, turning, stopping, and sitting down in the powered exoskeleton. An overview of the 6MWT, 10MWT and the TUG measurements have been described previously by Yang18 and are presented in Table 1 along with the patient demographic information of the participants.
Figure 1. Two handed crutch balance. This figure demonstrates a person standing still and balancing with both crutches. Please click here to view a larger version of this figure.
Figure 2. One handed crutch balance. This figure demonstrates a person standing still and balancing with only 1 crutch. Please click here to view a larger version of this figure.
Figure 3. Walking indoors on a smooth surface. This figure demonstrates a person walking indoors on a flat surface. Please click here to view a larger version of this figure.
Figure 4. Walking on carpet. This figure demonstrates a person walking indoors on a carpeted surface. Please click here to view a larger version of this figure.
Figure 5. Walking outdoors on grass. This figure demonstrates a person walking outdoors on grass. Please click here to view a larger version of this figure.
Figure 6. Walking on slopes. This figure demonstrates a person walking outdoors down a curb cutout. Please click here to view a larger version of this figure.
Figure 7. Navigating an elevator. This figure demonstrates a person walking out of a timed door setting such as an elevator door. Please click here to view a larger version of this figure.
Figure 8. Walking out of a revolving door. This figure demonstrates a person walking out of a revolving door. Please click here to view a larger version of this figure.
Figure 9. Overhead cabinet and countertop reaching. This figure demonstrates a person taking items out of an overhead cabinet. Please click here to view a larger version of this figure.
Figure 10. Sitting outside on a park bench. This figure demonstrates a person sitting outside on a park bench. Please click here to view a larger version of this figure.
Figure 11. Averaged Ten-session 10MWT Speeds. The data demonstrates the 10MWT speeds for the first 60 sessions of training averaged by ten session intervals. The x-axis describes the sessions and the y-axis describes the average speed (m/sec) calculated from the 10MWT result obtained during the participants training session. A linear best fit line was overlaid on each participant's results. Please click here to view a larger version of this figure.
|Walk Tests (WT) and Levels of Assistance (LOA)
|10 m WT
Table 1. Characteristics of the Participants and Walk Test Results. SID=subject identification number; y=years; cm=centimeters; kg=kilograms; DOI=duration of injury; LOI=level of injury; AIS=American Spinal Injury Association Impairment Scale; LOA=level of assistance; s=seconds; m=meters; NP=Not-Performed and NT=non-traumatic SCI. LOA was adapted from the FIM as one of the following: moderate assistance (Mod) - participant performs 50% to 74% of the task; minimal assistance (Min) - the user performs 75% or more of the task; supervision (S) - the trainer is not touching the participant but is close enough to reach in to provide support for balance or guidance as needed; and modified independence (MI) - the trainer does not provide any assistance, and the participant is fully independent while walking in device. Re-print with permission, from Yang A, Asselin P, Knezevic S, Kornfeld S, Spungen A. Assessment of in-hospital walking velocity and level of assistance in a powered exoskeleton in persons with spinal cord injury. Top Spinal Cord Inj Rehabil. 2015; 21(2):100-109. Copyright (C) 2015 Thomas Land Publishers, Inc.
During the past five years, our group has developed a successful screening and training program for participants to use the type of powered exoskeleton that requires crutches. We have trained individuals with motor complete paralysis as well as those with incomplete paralysis. This training program has the potential to be modified and built upon with additional devices that require the use of crutches, or newer versions of the existing devices.
Standardization of a training program is important to ensure participant safety, successful use of the device, identify staff resources, and to acquire consistent results. Key points in a good training program include appropriate candidate selection, proper fitting of the device, appropriate skill progression, and providing assistance on the shoulders or on an area with intact sensation to enable the user to recognize the required force and movement, promoting adaption of their movements during the subsequent stepping actions. It is important to practice this strategic dance between the trainer and the user in order to minimize trainer support, thus helping the user gain expertise and independence in the device. Trainers should avoid assisting below the participant's level of sensation since this action results in difficulty in becoming independent in the exoskeleton. Another key point to improve walking skill is to challenge the participant with walking on various surfaces and in different environments. Participants perceive walking indoors and on flat/smooth surfaces in the medical center to be easier than ambulating on a carpeted floor. Walking on carpeted floors, in turn, is reported to be easier than walking outdoors on uneven surfaces such as concrete or asphalt. Walking up and down different slope gradients force the participant to adapt their walking strategy because the method of weight shifting becomes more challenging due to the altered center of balance presented by the slope. All of these challenging environments are commonly encountered within the community and therefore, are very important to practice in a controlled setting to properly prepare the participant.
There have been several reports in persons with SCI who have learned to use a powered exoskeleton to safely ambulate overground16-19,21,36. Many of the participants in these reports had little to no residual function or sensation in their lower extremities. No serious adverse events were reported from these studies and the devices were deemed safe to use with the proper training. The adverse events reported included skin abrasions, bruising or redness of the skin, and fatigue of the upper extremities, especially during the initial training sessions16,19,36. It was noted that with continued training, participants noticed a reduction of upper extremity fatigue and skin abrasions resolved quickly with better fitting of the device. Future bruising and redness were avoided with adjustment of the straps and strategic placement of additional padding surrounding the affected area.
Proficiency in the use of the device is determined by the ability to achieve faster ambulation speeds, reduced levels of assistance, and safe ambulation in varied environments. Prior reports of walking ability showed that those who were more independent would ambulate faster than those who needed assistance. A report by van Hedel et al. categorized walkers as "assisted walkers" if they could ambulate with a minimum speed of 0.44 ± 0.14 m/sec; a speed associated with those who chose to walk outdoors with assistance over using their wheelchair42. This walking speed is similar to the 0.40 m/sec speed of the limited community ambulators reported in persons with stroke.43 Although only a few studies have reported ambulation speed and level of assistance using robotic exoskeletons, these studies indicated that many participants were able to achieve the 0.40 m/sec walking speed mentioned in these previous reports. A report using a powered exoskeleton showed that 7 of 12 participants were able to ambulate faster than 0.40 m/sec18. Another investigation using a different powered exoskeleton was able to illustrate 6 of 16 participants successfully ambulating greater than 0.40 m/sec36. Although reports using a third powered exoskeleton have not demonstrated walking speeds of 0.40 m/sec22,44, future reports may show increased walking speeds with further training and/or adaptions in that device. Thus far, all studies using powered exoskeletons have reported those needing greater levels of assistance walked at slower speeds. One thought discussed in these reports was that although some of the participants did not ambulate above the 0.40 m/sec speed, they were able to ambulate at the level of "supervision" as defined in the FIM scale. These reports suggest that, with additional training or modifications to the devices, ambulation at these faster speeds may be achieved.
Energy expenditure measured by oxygen consumption has been demonstrated to be increased with exoskeletal-assisted walking, but not above the threshold that is unduly fatiguing. Eight participants who ambulated in the powered exoskeleton at an average pace of 0.22 ± 0.11 m/sec demonstrated walking oxygen consumption rates of 11.2 ± 1.7 ml/kg/min and heart rates of 118 ± 21 bmp (48% ± 16% heart rate reserve), both of which were a significant increase from sitting and standing17, but significantly below the maximum predicted values. Another report using a different powered exoskeleton, evaluated oxygen consumption in 5 participants during 2 bouts of walking and reported 9.5 ± 0.8 ml/kg/min when walking at 0.19 ± 0.01 m/sec and 11.5 ± 1.4 ml/kg/min when walking at 0.27 ± 0.05 m/sec21. Both of these studies demonstrated that participants ambulating at a moderate intensity were above the minimal training intensity threshold determined by the American College of Sports Medicine to be effective for cardiorespiratory benefits45. This suggests that these devices have the potential to be used for longer periods of time, providing a form of activity that if performed regularly may be expected to lead to improvements in the user's fitness, body composition and lipid profile.
The powered exoskeletons offer a form of modified independence (level six as defined by the FIM) for standing and overground ambulation for persons with upper extremity function. Future devices may be designed to ambulate at faster speeds or provide a greater ability to vary the desired ambulation speed. Future exoskeletons may also be designed for those with limited hand and arm function (such as those with tetraplegia) by maintaining the user's balance with additional trunk support and providing another mechanism than holding a crutch for maintaining balance. Advancements in brain control may one day be available to be incorporated to control the walking movement20. Within this emerging field, the basic training concepts presented may be applicable to the current and future powered exoskeletons, but should be tailored to the user and the exoskeleton being used.
Standardized training strategies are currently utilized for successful participant exoskeletal-assisted walking; future modifications of these devices may need adaptions to the training paradigm. Teaching qualified SCI health-care professionals to appropriately train persons with SCI to perform exoskeletal-assisted walking is needed for the continued use and prescription of these devices. The future is bright for these devices; the use of powered exoskeletons by persons with SCI would become more widespread with the establishment of training programs in medical and rehabilitation centers throughout the world. Additionally, future research may show that regular exoskeletal-assisted walking improves many of the secondary medical complications that are associated with immobility and paralysis from spinal cord injury.
The authors have nothing to disclose.
Support for this work was obtained by the VA Rehabilitation Research & Development National Center of Excellence for the Medical Consequences of Spinal Cord Injury (VA RR&D #B9212C). Two of the four powered exoskeleton devices were used on a loaner basis at no cost from ReWalk Robotics, Inc. (Marlborough, Massachusetts). Additionally a portion of participants obtained Orthopedic shoes which were donated by Aetrex Worldwide Inc. (Teaneck, New Jersey).
Assistance from Denis Doyle-Green was invaluable during the training program and we thank him for this. We would also like to thank the physical therapists from the Rehabilitation and Spinal Cord Injury Services at the James J. Peters VA Medical Center for their advisement and consultations.
|Powered Exoskeleton such as ReWalk™, Ekso™, REX®, and Indego®, etc.
|Loft strand Crutches
Copyright © 2024 MyJoVE Corporation. All rights reserved