Abstract
[Purpose] To provide practical suggestions for the clinical application of robotic exoskeletal orthoses, we conducted a retrospective survey of patients with spinal cord lesions. [Participants and Methods] We studied the records of patients with spinal cord lesions who were trained using the ReWalk system (ReWalk Robotics). These records included the neurological level of injury, American Spinal Injury Association Impairment Scale, time since injury, total training time, adverse events, level of assistance required, dropouts, and reasons for dropout. [Results] Twenty-eight patients were trained with the ReWalk system at least once. Three of them underwent a modified training program owing to their severe paralysis. Eight patients dropped out from the regular training program or required a moderate or greater level of assistance. Twelve patients regained ambulation with the ReWalk system in indoor environments on a flat surface. The remaining five achieved ambulation with the ReWalk system in outdoor environments, including on slopes. Thirty-four adverse events were recorded. These events resolved within a few days with rest, adhesive dressing bandages, or other means. [Conclusion] The feasibility and safety of the training program and the outcomes achieved were comparable to those in previous studies. The device fit may need more careful attention.
Key words: Adverse events, Device fit, Gait training
INTRODUCTION
Spinal cord lesions cause severe paralysis, affecting all activities of daily living. In the case of bilateral paralysis, the motor function of both lower limbs is impaired. This is especially the case for people with spinal cord lesions of American Spinal Injury Association Impairment Scale (AIS) grades A to C; such complete or severely impaired but incomplete lesions result in the loss of motor function of both lower limbs, disabling ambulatory activity.
Several robotic exoskeletal orthoses have been developed to compensate for ambulatory disabilities. Some of these have been applied to patients with spinal cord lesions in clinical practice, with reports of their safety and efficacy in this population1,2,3).
The ReWalk system (ReWalk Robotics, Yokneam, Israel) has an exoskeletal structure and actuators on both hip and knee joints. It helps the wearer to maintain an upright standing posture and perform ambulatory motions. Using the ReWalk system and crutches on both sides, a person with complete paraplegia can learn to walk with or without further assistance4, 5). It has been approved by the US Food and Drug Administration for institutional use by persons with T4–L5 paraplegia and personal use by persons with paraplegia at T7–L56). The clinical applications and safety of this device have been reported5, 7). It allows patients to walk with a degree of independence6) and it reproduces motion similar to that of normal ambulation4).
However, the results for the ReWalk system have only been reported in the US4,5,6), Israel7), and Europe2, 8, 9). Herein, we report on a series of patients with spinal cord lesions who were trained with a robotic exoskeletal orthosis, aiming to provide practical suggestions.
PARTICIPANTS AND METHODS
Twenty-eight patients, including four females, were trained with the ReWalk system at least once from January 2015 to June 2021 in the Kanagawa Rehabilitation Center (Table 1).
Table 1. Characteristics of participants.
| Characteristics | Total | Trained with a body-weight- supporting walker | Dropped out or needed moderate or greater assistance | Achieved indoor walking | Achieved outdoor walking | |
| n | 28 | 3 | 8 | 12 | 5 | |
| Age, years | 34.3 ± 11.4 (31, 13–54) | 13–15 (n=1), 31–45 (n=2) | 37.8 ± 10.7 (36.5, 20–54) | 36.4 ± 10.5 (35.5, 20–53) | 23.8 ± 3.9 (21, 21–31) | |
| Severity (n) | ||||||
| Cervical-A | 1 | 1 | 0 | 0 | 0 | |
| Cervical-B | 1 | 1 | 0 | 0 | 0 | |
| Cervical-C | 2 | 1 | 0 | 1 | 0 | |
| Upper thoracic-A | 4 | 0 | 1 | 3 | 0 | |
| Upper thoracic-B | 0 | 0 | 0 | 0 | 0 | |
| Upper thoracic-C | 0 | 0 | 0 | 0 | 0 | |
| Lower thoracic/lumbar-A | 17 | 0 | 6 | 7 | 4 | |
| Lower thoracic/lumbar-B | 2 | 0 | 0 | 1 | 1 | |
| Lower thoracic/lumbar-C | 1 | 0 | 1 | 0 | 0 | |
| TSI (days) | 1,709.0 ± 2,080.3 (645, 51–7,240) | 1,769.0 ± 1,166.4 (1,783, 334–3,191) | 2,270.0 ± 2,480.0 (1,352, 160–7,240) | 1,841.1 ± 2,157.6 (685.5, 101–7,063) | 458.0 ± 541.6 (163, 51–1,499) | |
| Restricted joints (n) | 13 | 1 | 1 | 9 | 2 | |
| Spasticity (n) | 12 | 0 | 2 | 9 | 1 | |
Age is indicated either as the mean ± standard deviation (median, range) or as the number of individuals in a certain age range, following recommended standard of reporting by the Executive Committee for the Development of the International Spinal Cord Injury data sets10).
TSI is indicated as the mean ± standard deviation (median, range).
TSI: time since injury.
The inclusion criteria for the ReWalk training program were as follows: paraplegia or tetraplegia caused by spinal cord lesions, a height of 160–190 cm, body weight less than 100 kg, currently living or able to live in a home or community, being motivated to train regularly with the ReWalk system, and being able to safely perform exercises while standing or walking with orthoses or by other means. Criteria for exclusion from the training program were a poor fit of the ReWalk system; osteoporosis (T-score < −0.25); pregnancy or potential pregnancy; severe concurrent medical diseases such as infections of the circulatory system, including the heart or lungs; pressure sores; severe spasticity (Modified Ashworth Scale score ≥3) or uncontrolled clonus; unstable spine or unhealed limb or pelvic fractures; heterotopic ossification that restricts the functional range of motion; substantial contractures; and psychiatric or cognitive comorbidities that might have interfered with the training.
The mean ± standard deviation (median, range) age was 34.3 ± 11.4 (31, 13–54) years, height was 170.3 ± 6.2 cm, and body weight was 62.9 ± 8.4 kg. The causes of spinal cord lesions were trauma (n=22), ischemia (n=3), hemorrhage (n=1), a tumor (n=1), and an arteriovenous fistula (n=1). The time since injury (TSI) ranged from 51 days to 7,240 days, with a median of 645 days.
The training program with the ReWalk system was the same as that in previous reports5, 9, 11). Briefly, the training program was developed by ReWalk Robotics to restore independent ambulation on a flat surface indoors (a basic course) or on an outdoor course that includes slopes (an advanced course). The participants learn skills such as standing with two crutches, weight-shifting, standing-to-sitting, sitting-to-standing, and walking. Two physical therapists who were certified as ReWalk trainers trained or supervised each participant unless they were skilled enough to walk safely with a single assistant. Immediately before and after each training session, the participant’s blood pressure, heart rate, and any symptomatic episodes were recorded, and the condition of the participant’s skin was observed.
The patients were divided into four subgroups. Patients in subgroup 1 trained with modified training program for various reasons, such as the severity of paralysis. Patients in subgroup 2 dropped out of regular training to walk independently or required a moderate or higher level of assistance by June 2021. Patients in subgroup 3 had their ambulation restored with the ReWalk system and required minimal assistance or close monitoring when walking indoors on a flat floor. Patients in subgroup 4 had their ambulation restored with the ReWalk system and required minimal assistance or close monitoring when walking outdoors, including on a slope. We collected the records of patients trained with the ReWalk system at least once from January 2015 to June 2021 in the Kanagawa Rehabilitation Center. The following items were included: training progress logs, including training period, number of sessions, total training time, and adverse events; and training outcomes, including the level of assistance, 10-m walk time (10MWT), 6-minute walk distance (6MWD), dropouts, and the reasons for dropout. These items were compared among subgroups with descriptive statistics.
This study was approved by the Institutional Review Board of the Kanagawa Rehabilitation Hospital (krh-2014-2). Informed consent was obtained in written form from all individual participants who were included in the study.
RESULTS
The patients’ training progress is displayed in Fig. 1, and the training logs are summarized in Table 2. Three patients out of the 28 underwent a modified training program owing to their severe paralysis. All three had tetraplegia and limited motor function of the trunk and arms. They were trained with a body-weight-supporting walker and achieved ambulation with moderate assistance, one of them with the added support of two crutches. No adverse events were recorded in these three cases.
Fig. 1.
Training progress.
Twenty-eight persons with spinal cord lesion (SCL) participated in the gait training program with the ReWalk. Of these, three underwent training with the ReWalk system and a body-weight-supporting walker as part of the modified training program. Eight attempted to restore independent walk with the ReWalk but dropped out from the program or stayed in need for moderate or greater physical assistance. Twelve participants achieved gait on a flat surface with minimal or less assistance through the program. Five successfully achieved gait including on a slope.
Table 2. Training records.
| Parameters | Trained with a body-weight-supporting walker | Dropped out or needed moderate or greater assistance | Achieved indoor walking | Achieved outdoor walking | |
| n | 3 | 8 | 12 | 5 | |
| Training period, days, median (range) | 321 (69–1,502) | 146 (7–439) | 77.5 (38–134) | 166 (53–233) | |
| Training sessions, median (range) | 18 (4–69) | 8 (3–28) | 19 (10–32) | 27 (19–60) | |
| Total training time, h, median (range) | 12.7 (2.7–51.0) | 8.0 (2–28.7) | 22.8 (20.0–49.0) | 40.5 (38.0–54.0) | |
| Adverse events | |||||
| Reddening | - | Bottom of foot: 3 | Bottom of foot: 4 Fibular head: 2 | Fibular head: 3 Posterior superior iliac spine: 1 | |
| Front of shank: 1 | |||||
| Fibular head: 1 | |||||
| Pain | - | Front of thigh: 1 | Hand: 1 | - | |
| Wrist: 1 | |||||
| Blistering | - | - | Bottom of foot: 1 | Front of thigh: 2 | |
| Subcutaneous bleeding | - | - | At fibular head: 1 | - | |
| Gluteal region: 1 | |||||
| Great trochanter: 1 | |||||
| Side of thigh: 1 | |||||
| Mid-upper arm: 1 | |||||
| Intramuscular bleeding | - | Hip adductor: 1 | Hip adductor: 1 | - | |
| Iliopsoas: 1 | |||||
| Muscle fatigue | - | - | Neck: 1 | - | |
| Orthostatic hypotension | - | - | Present: 1 | - | |
| Abrasion | - | - | - | Fibular head: 1 | |
| Sacrum region: 1 | |||||
| Epidermal split | - | - | - | Fibular head: 1 | |
Eight other patients initially attempted to restore the independent ambulation with the ReWalk but subsequently dropped out from the regular training program or required the same (moderate or greater) level of assistance by the end of the study period. All of them had paraplegia, and most had a relatively low neurological level of injury (NLI), indicating that they had trunk motor function. One of these patients was able to walk with two crutches with minimal to moderate assistance, and the others needed moderate or greater assistance by the end of the study period. The reasons for dropout were as follows: transferring to a more demanding ambulation training program without a robotic device (n=1, a patient with an AIS grade of C), logistical difficulties in regular participation in the training program due to social reasons such as having a job (n=5), and discontinuation owing to the coronavirus disease 2019 pandemic (n=2). Seven adverse events were reported for these eight patients, including reddening of the foot and the shank, pain in the thigh, and intramuscular bleeding of the leg.
Twelve patients (one with incomplete tetraplegia and 11 with paraplegia with or without trunk motor function) regained ambulation with the ReWalk system and two crutches in indoor environments on a flat surface with close monitoring. Their training took 1–4 months, with one or two sessions a week. At the end of the training program, their 10MWT was 42.8 ± 6.5 s and their 6MWD was 81.3 ± 22.2 m (one 6MWD record was missing). Eighteen adverse events were recorded for 7 of these 12 patients, including reddening of the foot and in the area of the fibular head, pain in the hand and wrist, blistering of the foot, subcutaneous bleeding of the leg and arm, intramuscular bleeding, muscle fatigue, and orthostatic hypotension.
The remaining five patients achieved ambulation with the ReWalk system and two crutches in outdoor environments, including on slopes, with close monitoring. These patients were, on average, younger and had a lower NLI and shorter TSI than the rest of the study participants. After 2–7 months’ training, with one or two sessions a week, their 10MWT was 21.8 ± 2.0 s and their 6MWD was 157.8 ± 24.8 m (one 6MWD record was also missing for this subgroup). Nine adverse events were recorded for four of these five patients, including abrasion, blistering, reddening, and epidermal splitting of the leg.
All adverse events resolved within a few days with rest, adhesive dressing bandages, or other means.
DISCUSSION
In the present study, patients with spinal cord lesions underwent gait training by using an exoskeletal robotic device. We noted the restoration of ambulation as well as adverse events that occurred.
The training outcomes were comparable to those of previous reports of gait training with the ReWalk system4,5,6,7,8,9, 12). In those studies, 80 patients were trained, 57 of whom achieved ambulation with the ReWalk system with no physical assistance. In the present study, 17 out of 28 patients achieved ambulation with close monitoring. In the previous studies, the median (range) 10MWT and 6MWD were reported as 38 (12–325) s (n=53)4,5,6,7,8, 12) and 90.2 (10.8–255.9) m (n=47)5,6,7,8, 12), respectively. The characteristics of those participants were as follows: 17 out of 80 were female4,5,6,7,8,9,10,11,12), their mean ± standard deviation (median, range) age was 37.3 ± 10.9 (36, 18–49) years (n=49)4, 5, 7, 8, 12); height, 173.4 ± 9.3 cm (n=45)4,5,6, 12); and body weight, 71.0 ± 11.2 kg (n=45)4,5,6, 12); all similar to those of the present study. The TSI in previous studies was 4 (1.0–24.3) years, longer than that in the present study. Seven injuries were at the cervical level, 30 at the upper-thoracic level, 40 at the lower-thoracic level, and 1 at the lumbar level4,5,6,7,8,9,10,11,12). In the previous reports, 71 patients had an AIS grade of A or B, 8 had an AIS grade of C, and 1 had an AIS grade of D4,5,6,7,8,9,10,11,12). This indicates that the cases in the present study might have been slightly less severe than those in previous studies. Thus, the present study suggests that the equivalent outcomes could be achievable even in the clinical setting in Japan or Asian population. Noted that the participants’ racial or social backgrounds might vary in the previous studies. As the physical, financial, and logistical costs are high in training programs with exoskeletal robotic devices13), these backgrounds and study settings could affect the compliance and the outcomes of the programs.
Adverse events were more common in this study, at 34 events for the 28 patients, than that in previous reports, at 51 cases for 53 patients5, 6, 8, 9, 11). Most adverse events were skin-related problems in both the present and previous studies. Similar to what Yang et al.6) reported, we added padding to avoid mechanical stress from contact with the device, frequently examined participants’ skin condition, and stretched their leg muscles before each training session as preventive measures in the present study. Despite these measures, the rate of adverse events was higher than those in past studies in the US4,5,6), Israel7), and European countries2, 8, 9). Langton et al. observed that geographical origin is related to differences in the gross morphology and composition of the skin and suggested that this could affect the skin’s resistance to mechanical stress14). The geographical distribution of the participants in the present study differs from that of previous studies, which might partially account for the observed variations in adverse event rates. However, publication bias should also be considered. Zeilig et al. reported that several reports on ambulatory training with devices did not mention excessive skin pressure among the adverse events; certain research groups may see such events as ordinary and not worth reporting7). As with conventional orthoses, the comfort and fit of exoskeletal robotic devices are crucial points of application. The high rate of adverse events in this case series suggests the need for careful attention during training and to manufacture a more comfortable and safer device.
The number of patients who ceased the training program should not be neglected. Our results suggest that adequate consideration be given to the participants enrolled in training programs with exoskeletal robotic devices, as the physical, financial, and logistical costs are high13). The characteristics of the patients who could not complete the training program were similar to those who completed the program in terms of age and injury severity. Neurological or general motor function may be a minor reason for dropping out. The most frequent reason in these cases was the difficulty attending regular training sessions owing to jobs. Community-living individuals with chronic spinal cord injuries, especially those with jobs, may need support to complete their training while they are busy with their jobs and other aspects of daily life.
The outcomes of ambulatory training with an exoskeletal robotic device, such as the gait velocity and level of assistance, are related to physical functioning. In a previous study, the NLI influenced gait performance12). The participants who achieved ambulation in outdoor environments had a lower NLI, were younger, and had a shorter TSI than participants who achieved ambulation only on flat surfaces. Although patients with cervical lesions achieved limited outcomes in this study, even those with motor-complete tetraplegia conducted a fair amount of gait training. Another study also demonstrated that gait training with an exoskeletal robotic device for people with tetraplegia is feasible and safe15).
This survey has some limitations. It was conducted at a single center. For generalizability, data from several centers are needed. Selection bias should also be considered, as the cases were sampled retrospectively. To participate in the training program, patients had to agree to weekly visits over several months. Thus, motivation, functional ability, social participation, and similar factors may affect the progress and outcomes of training in the general population. In order to establish a more precise picture of the feasibility of the training with the ReWalk, a multi-center study in clinical settings is required. In conclusion, the presented cases demonstrated comparable outcomes to those of previous studies and supported the evidence that patients with paraplegia may achieve walking with the device with minimum assistance. Remaining problems include training-related burdens such as physical or logistical costs and adverse events assumed to be due to mechanical stress between the device and the body.
Funding
The data used in this study were obtained during routine clinical practice. No additional funding was received.
Conflict of interest
The authors declare no competing interests.
REFERENCES
- 1.Fang CY, Tsai JL, Li GS, et al. : Effects of robot-assisted gait training in individuals with spinal cord injury: a meta-analysis. BioMed Res Int, 2020, 2020: 2102785. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.van Dijsseldonk RB, van Nes IJ, Geurts AC, et al. : Exoskeleton home and community use in people with complete spinal cord injury. Sci Rep, 2020, 10: 15600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Tamburella F, Lorusso M, Tramontano M, et al. : Overground robotic training effects on walking and secondary health conditions in individuals with spinal cord injury: systematic review. J Neuroeng Rehabil, 2022, 19: 27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Fineberg DB, Asselin P, Harel NY, et al. : Vertical ground reaction force-based analysis of powered exoskeleton-assisted walking in persons with motor-complete paraplegia. J Spinal Cord Med, 2013, 36: 313–321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Esquenazi A, Talaty M, Packel A, et al. : The ReWalk powered exoskeleton to restore ambulatory function to individuals with thoracic-level motor-complete spinal cord injury. Am J Phys Med Rehabil, 2012, 91: 911–921. [DOI] [PubMed] [Google Scholar]
- 6.Yang A, Asselin P, Knezevic S, et al. : 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: 100–109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zeilig G, Weingarden H, Zwecker M, et al. : Safety and tolerance of the ReWalk™ exoskeleton suit for ambulation by people with complete spinal cord injury: a pilot study. J Spinal Cord Med, 2012, 35: 96–101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Benson I, Hart K, Tussler D, et al. : Lower-limb exoskeletons for individuals with chronic spinal cord injury: findings from a feasibility study. Clin Rehabil, 2016, 30: 73–84. [DOI] [PubMed] [Google Scholar]
- 9.Platz T, Gillner A, Borgwaldt N, et al. : Device-training for individuals with thoracic and lumbar spinal cord injury using a powered exoskeleton for technically assisted mobility: achievements and user satisfaction. BioMed Res Int, 2016, 2016: 8459018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.DeVivo MJ, Biering-Sørensen F, New P, et al. International Spinal Cord Injury Data Set: Standardization of data analysis and reporting of results from the International Spinal Cord Injury Core Data Set. Spinal Cord, 2011, 49: 596–599. [DOI] [PubMed] [Google Scholar]
- 11.Khan AS, Livingstone DC, Hurd CL, et al. : Retraining walking over ground in a powered exoskeleton after spinal cord injury: a prospective cohort study to examine functional gains and neuroplasticity. J Neuroeng Rehabil, 2019, 16: 145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Charette C, Déry J, Blanchette AK, et al. : A systematic review of the determinants of implementation of a locomotor training program using a powered exoskeleton for individuals with a spinal cord injury. Clin Rehabil, 2023, 37: 1119–1138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Guanziroli E, Cazzaniga M, Colombo L, et al. : Assistive powered exoskeleton for complete spinal cord injury: correlations between walking ability and exoskeleton control. Eur J Phys Rehabil Med, 2019, 55: 209–216. [DOI] [PubMed] [Google Scholar]
- 14.Langton AK, Sherratt MJ, Sellers WI, et al. : Geographical ancestry is a key determinant of epidermal morphology and dermal composition. Br J Dermatol, 2014, 171: 274–282. [DOI] [PubMed] [Google Scholar]
- 15.Rodriguez Tapia G, Doumas I, Lejeune T, et al. : Wearable powered exoskeletons for gait training in tetraplegia: a systematic review on feasibility, safety and potential health benefits. Acta Neurol Belg, 2022, 122: 1149–1162. [DOI] [PubMed] [Google Scholar]