Powered Gait Orthosis and Exoskeletons in SCI

Author Year; Country
Research Design
Sample Size
Methods Outcomes

Shin et al. 2014; Seoul
Prospective, Randomized Clinical Trial
Level 2
N= 53

Population: 53 individuals- 34 males and 19 females; 31 with cervical injuries and 22 with thoracic & lumbar injuries; 36 with traumatic SCI and 16 with non-traumatic SCI; mean age= 48.15 ± 11.14y; months post injury= 3.33 ± 2.02 months.
Treatment: Sixty patients with motor incomplete spinal cord injury (SCI) were included in a prospective, randomized clinical trial by comparing Robot-Assisted Gait Training (RAGT) to regular physiotherapy. The RAGT group received RAGT three sessions per week at duration of 40 minutes with regular physiotherapy in 4 weeks. The conventional group underwent regular physiotherapy twice a day, 5 times a week.
Outcome Measures: ASIA lower extremity motor score subscale (LEMS), ambulatory motor index (AMI), spinal cord independence measurement III mobility section (SCIM3-M), walking index for spinal cord injury version II (WISCI-II).
  1. At the end of rehabilitation, both groups showed significant improvement in LEMS, AMI, SCIM3-M, and WISCI-II.
  2. RAGT group members improved their WISCI-II scores.

Tanabe et al. 2013; Japan
Prospective controlled trial
Level 2

Population: 4 participants with complete paraplegia (3M 1F); 30-59 yrs old; 4-20 yrs post-injury.
Treatment: Participants performed ground-level walking test with both the conventional orthosis (PrimeWalk) and the WPAL orthosis.
Outcome Measures: Mean duration and distance of consecutive walking; Functional Ambulation Categories scale; PCI; modified Borg CR10 scale; EMG of upper extremities.
  1. Activation patterns of the EMG during gait indicate that WPAL needed only intermittent contraction while PrimeWalk demanded persistent contraction.
  2. The duration and distance of consecutive walking is higher for the WPAL than the conventional orthosis: for PrimeWalk: 5-12 min, 20-44 m consecutive walking; for WPAL: 7.8-40 min, 40-580m consecutive walking.
  3. The PCI, perceived exertion and EMG of upper extremities was lower for the WPAL than the conventional orthosis.

Tanabe et al. 2013b; Japan
Prospective controlled trial
Level 2

Population: 7 participants with motor-complete SCI (6M 1F); 6 AIS A, 1 AIS B; 32-61 yrs old; 6-20 years after injury.
Treatment: Participants performed ground-level walking test with both the conventional orthosis (PrimeWalk) and the WPAL orthosis.
Outcome Measures: Mean duration and distance of consecutive walking; Functional Ambulation Categories scale.
  1. With the WPAL, all users achieved independent gait on a level floor (Functional Ambulation Categories score of 4).
  2. Mean duration and distance of consecutive walking were 14.1(11.4) minutes and 165.6(202.6) m with the WPAL.
  3. With the orthosis, duration of walking ranged from 5-8 minutes and distance walked ranged from 20-107m. With the WPAL, duration of consecutive walking ranged from 4.5-40 minutes and distance walked ranged from 30-640m.

Arazpour et al. 2013c; Iran
Prospective controlled trial
Level 2

Population: 4 participants with thoracic level SCI (2M 2F); 22-29 yrs old; 9-51 months since injury; 3 incomplete 1 complete SCI.
Treatment: Patients performed orthotic gait training with a Powered Gait Orthosis (PGO) for a min of 6 wks, 1 hr/day for 5 days/wk prior to walking trials. Walking trials with an Isocentric Reciprocal Gait Orthosis (IRGO) and with both separate and synchronized movements with actuated orthotic hip and knee joints in a PGO were conducted.
Outcome Measures: Kinematics and temporal-spatial parameters of walking.
  1. Using separate and synchronized actuated movement of the hip and knee joints in the PGO increased gait speed and step length, and reduced lateral and vertical compensatory motions when compared to the IRGO, but there were no significant differences in these parameters.
  2. Using the new PGO improved knee and hip joint kinematics:
    1. Hip flexion (o): IRGO= 9.25(0.95); new PGO=18.75(2.36)
    2. Hip extension (o): IRGO=5.5(0.57); new PGO=7.75 (0.95)
    3. Knee flexion (o): IRGO = 6.75(0.95); new PGO = 37(1.82)

Kozlowski et al. 2015; USA
Level 2
N= 7

Population: 7 males; 2 with tetraplegia and 5 with motor-complete SCI; 3 AIS A, 1 AIS B, and 3 AIS C; median age= 36y; years post injury= 0.5y.
Treatment:  A convenience sample was enrolled to learn to use the first-generation Ekso powered exoskeleton to walk. Participants were given up to 24 weekly sessions of instruction. Data were collected on assistance level, walking distance and speed, heart rate, perceived exertion, and adverse events. Time and effort was quantified by the number of sessions required for participants to stand up, walk for 30 minutes, and sit down, initially with minimal and subsequently with contact guard assistance.
Outcome Measures: Primary outcomes: the number of sessions needed to achieve a rating of “minimal assistance”, number of sessions required until the rating became “contact guard only” for standing/sitting and for walking; Secondary outcomes: measures of walking tolerance and physical exertion.
  1. Walk times ranged from 28 to 94 minutes with average speeds ranging from 0.11 to 0.21 m/s.
  2. For all participants, heart rate changes and reported perceived exertion were consistent with light to moderate exercise.

Hartigan et al. 2015; USA
Level 4
N= 16

Population: 16 individuals- 13 males and 3 females; SCI ranging from C5 complete to L1 incomplete; age range= 18-51 years.
Treatment: To assess how quickly each participant could achieve proficiency in walking, each participant was trained in the system for 5 sessions, each session lasting approximately 1.5 hours. Following these 5 sessions, each participant performed a 10MWT and a 6MWT.
Outcome Measures: 10 MWT, 6MWT, donning and doffing times, ability to walk on various surfaces.
  1. At the end of 5 sessions (1.5 hours per session), average walking speed was 0.22 m/s for persons with C5-6 motor complete tetraplegia, 0.26 m/s for T1-8 motor complete paraplegia, and 0.45 m/s for T9-L1 paraplegia.
  2. Distances covered in 6 minutes averaged 64 meters for those with C5-6, 74 meters for T1-8, and 121 meters for T9-L1.

Additionally, all participants were able to walk on both indoor and outdoor surfaces

Yang et al. 2015; USA
Post Test
Level 4
N= 12

Population: 12 individuals- 10 males and 2 females; 9 AIS A, 2 AIS B and 1 AIS C; Level of injury between C8 to T11; age range= 31 to 75.
Treatment: Twelve individuals with SCI ≥1.5 years who were wheelchair users participated. They wore a powered exoskeleton (ReWalk) with crutches to complete 10-meter (10MWT) and 6-minute (6MWT) walk tests. LOA was defined as modified independence (MI), supervision (S), minimal assistance (Min), and moderate assistance (Mod). Best effort EAW velocity, LOA, and observational gait analysis were recorded.
Outcome Measures: 10 MWT, 6 MWT, level of assistance (LOA), degree of hip flexion, degree of knee flexion, step time.
  1. 7 of 12 participants ambulated ≥0.40 m/s. 5 participants walked with MI, 3 with S, 3 with Min, and 1 with Mod. Significant inverse relationships were noted between LOA and EAW velocity for both 6MWT and 10MWT.
  2. There were 13 episodes of mild skin abrasions. MI and S groups ambulated with 2-point alternating crutch pattern, whereas the Min and Mod groups favored 3-point crutch gait.

Fineberg et al. 2013; USA
(with Able-Bodied normative comparisons)
Level 4
N=6 participants with SCI (3 control)

Population: 6 participants with chronic, motor-complete thoracic SCI (5M 1F); 24-61 yrs old; 3 requiring minimal assistance and 3 requiring no assistance. 3 AB controls.
Treatment: Participants underwent training sessions consisting of 1-2 hours of combined standing and walking 3 times/week for 5-6 months on the ReWalk powered exoskeleton assisted walking system.
Outcome Measures: magnitude and pattern of mechanical loading via vertical ground reaction force (vGRF): peak stance average (PSA), peak vGRF for heel strike, mid-stance, and toe-off.
  1. Participants in the SCI minimal-assist group demonstrated the lowest vGRF compared with the no-assist and AB controls. The min-assist group had significantly lower area under the curve for gait cycle (p<0.05); no significant difference was found for SCI no-assist vs AB control group.
  2. SCI min-assist group had statistically slower mean walking velocity than the SCI no-assist group (P = 0.0148).
  3. Participants with SCI demonstrated mechanical loading magnitudes and patterns similar to able-bodied gait:
    1. SCI no-assist: avg vGRFHS=66(8)%, vGRFMS=91(12)%, vGRFTO=107(7)%
    2. SCI min-assist:  avg vGRFHS=36(15)%, vGRFMS=47(12)%, vGRFTO=62(21)%
    3. AB control:  avg vGRFHS=91(9)%, vGRFMS=70(9)%, vGRFTO=105(18)%.

Esquenazi et al. 2012; USA
Level 4

Population: 12 participants with chronic SCI (8M 4F); 18-55 yrs old; all motor-complete cervical and thoracic; >6 months post-injury.
Treatment: All participants had gait training using the ReWalk powered exoskeleton; participants were trained for up to 24 sessions of 60-90 min duration over approximately 8 weeks.
Outcome Measures: 6MWT; 10MWT; gait laboratory evaluation; dynamic electromyogram; survey containing questions about comfort and confidence using the ReWalk; assessment of spasticity and pain; physical examination; Short Form-36 v2 Health Survey Questionnaire.
  1. By completion of the trial, all participants had walked under their own control without human assistance while using the ReWalk for at least 50-100m continuously and for a period of at least 5-10 minutes.
  2. Excluding 2 participants with considerably reduced walking abilities, average distances and average walking speed significantly improved. Average walking speed was 0.25m/s (0.03-0.45m/s). (no significance testing done)
  3. 3 participants reported their overall spasticity improved after training.
  4. All participants had strong positive comments regarding the emotional/psychosocial benefits of the use of ReWalk.
  5. At the 12-month follow-up, general health status as measured by study clinicians did not change.


New technology has advanced passive bracing to exoskeletons which are wearable robotic devices that have powered joints and extensive software programming to enable synchronized, functional and safe movement. In addition, the weight of the device can be borne by the exoskeleton and not the patient. While the gait speeds are still relatively slow due to safety issues (to minimize loss of balance and potential falls), the major advance is the reduction of energy that is required to utilize these devices to walk. Patients with primarily thoracic injuries have utilized these devices. With the price continuing to drop for these technologies, this will provide opportunity to evaluate their long-term use as more people acquire them for home use. Furthermore, newer versions are accommodating the ability to sit or wheel a wheelchair while wearing the device, increasing utility in a clinical setting (assisting with rehabilitation goals).


Studies ranging from level 1b to level 4 evidence show that PGOs can enable safe walking and reduce energy expenditure compared to passive bracing in patients with thoracic injuries, or those with adequate triceps functioning.