Effectiveness of robot-assisted gait training in children with cerebral palsy: a bicenter, pragmatic, randomized, cross-over trial (PeLoGAIT).

BACKGROUND: Walking ability is a priority for many children with cerebral palsy (CP) and their parents when considering domains of importance regarding treatment interventions. Partial body-weight supported treadmill training has become an established therapeutic treatment approach to address this demand. Further, new robotic rehabilitation technologies have increasingly been implemented in the clinical setting to allow for longer training sessions with increased step repetitions while maintaining a consistent movement pattern. But the current evidence about its clinical effectiveness in pediatric rehabilitation is weak. The aim of this research project is therefore to investigate the effectiveness of robot-assisted gait training on improvements of functional gait parameters in children with cerebral palsy. METHODS/DESIGN: Children aged 6 to 18 years with bilateral spastic cerebral palsy who are able to walk at least 14 m with or without walking aids will be recruited in two pediatric therapy centers in Switzerland. Within a pragmatic cross-over design with randomized treatment sequences, they perform 5 weeks of robot-assisted gait training (three times per week with a maximum of 45 min walking time each) or a 5-week period of standard treatment, which is individually customized to the needs of the child and usually consists of 1-2 sessions of physiotherapy per week and additional hippotherapy, circuit training as well as occupational therapy as necessary. Both interventions take place in an outpatient setting. The percentage score of the dimension E of the Gross Motor Function Measure-88 (GMFM-88) as primary outcome as well as the dimension D of the GMFM-88, 6-minute and 10-meter walking tests as secondary outcomes are assessed before and at the end of each intervention period. Additionally, a 5-week follow-up assessment is scheduled for the children who are assigned to the standard treatment first. Treatment effects, period effects as well as follow-up effects are analyzed with paired analyses and independent test statistics are used to assess carry-over effects. DISCUSSION: Although robot-assisted gait training has become an established treatment option to address gait impairments, evidence for its effectiveness is vague. This pragmatic trial will provide important information on its effects under clinical outpatient conditions. TRIAL REGISTRATION: ClinicalTrials.gov: NCT00887848 . Registered 23 April 2009.

International standards for neurological classification of spinal cord injury: classification skills of clinicians versus computational algorithms.

STUDY DESIGN: This is a retrospective analysis. OBJECTIVES: The objective of this study was to describe and quantify the discrepancy in the classification of the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI) by clinicians versus a validated computational algorithm. SETTINGS: European Multicenter Study on Human Spinal Cord Injury (EMSCI). METHODS: Fully documented ISNCSCI data sets from EMSCI's first years (2003-2005) classified by clinicians (mostly spinal cord medicine residents, who received in-house ISNCSCI training by senior SCI physicians) were computationally reclassified. Any differences in the scoring of sensory and motor levels, American Spinal Injury Association Impairment Scale (AIS) or the zone of partial preservation (ZPP) were quantified. RESULTS: Four hundred and twenty ISNCSCI data sets were evaluated. The lowest agreement was found in motor levels (right: 62.1%, P=0.002; left: 61.8%, P=0.003), followed by motor ZPP (right: 81.6%, P=0.74; left 80.0%, P=0.27) and then AIS (83.4%, P=0.001). Sensory levels and sensory ZPP showed the best concordance (right sensory level: 90.8%, P=0.66; left sensory level: 90.0%, P=0.30; right sensory ZPP: 91.0%, P=0.18; left sensory ZPP: 92.2%, P=0.03). AIS B was most often misinterpreted as AIS C and vice versa (AIS B as C: 29.4% and AIS C as B: 38.6%). CONCLUSION: Most difficult classification tasks were the correct determination of motor levels and the differentiation between AIS B and AIS C/D. These issues should be addressed in upcoming ISNCSCI revisions. Training is strongly recommended to improve classification skills for clinical practice, as well as for clinical investigators conducting spinal cord studies. SPONSORSHIP: This study is partially funded by the International Foundation for Research in Paraplegia, Zurich, Switzerland.

Extent of spontaneous motor recovery after traumatic cervical sensorimotor complete spinal cord injury.

STUDY DESIGN: Retrospective, longitudinal analysis of motor recovery data from individuals with cervical (C4-C7) sensorimotor complete spinal cord injury (SCI) according to the International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI). OBJECTIVES: To analyze the extent and patterns of spontaneous motor recovery over the first year after traumatic cervical sensorimotor complete SCI. METHODS: Datasets from the European multicenter study about SCI (EMSCI) and the Sygen randomized clinical trial were examined for conversion of American Spinal Injury Association (ASIA) Impairment Scale (AIS) grade, change in upper extremity motor score (UEMS) or motor level, as well as relationships between these measures. RESULTS: There were no overall differences between the EMSCI and Sygen datasets in motor recovery patterns. After 1 year, up to 70% of subjects spontaneously recovered at least one motor level, but only 30% recovered two or more motor levels, with lesser values at intermediate time points. AIS grade conversion did not significantly influence motor level changes. At 1 year, the average spontaneous improvement in bilateral UEMS was 10-11 motor points. There was only moderate relationship between a change in UEMS and a change in cervical motor level (r(2)=0.30, P<0.05). Regardless of initial cervical motor level, most individuals recover a similar number of motor points or motor levels. CONCLUSION: Careful tracking of cervical motor recovery outcomes may provide the necessary sensitivity and accuracy to reliably detect a subtle, but meaningful treatment effect after sensorimotor complete cervical SCI. The distribution of the UEMS change may be more important functionally than the total UEMS recovered.

Development of quantitative and sensitive assessments of physiological and functional outcome during recovery from spinal cord injury: a clinical initiative.

The ability to detect physiological changes associated with treatments to effect axonal regeneration, or novel rehabilitation strategies, for spinal cord injury will be challenging using the widely employed American Spinal Injuries Association (ASIA) impairment scales (AIS) for sensory and motor function. Despite many revisions to the AIS standard neurological assessment, there remains a perceived need for more sensitive, quantitative and objective outcome measures. The purpose of Stage 1 of the Clinical Initiative was to develop these tools and then, in Stage 2 to test them for reliability against natural recovery and treatments expected to produce functional improvements in those with complete or incomplete spinal cord injury (SCI). Here we review aspects of the progress made by four teams involved in Stage 2. The strategies employed by the individual teams were (1) application of repetitive transcranial magnetic stimulation (rTMS) to the motor cortex in stable (chronic) SCI with intent to induce functional improvement of upper limb function, (2) a tele-rehabilitation approach using functional electrical stimulation to provide hand opening and grip allowing incomplete SCI subjects to deploy an instrumented manipulandum for hand and arm exercises and to play computer games, (3) weight-assisted treadmill walking therapy (WAT) comparing outcomes in acute and chronic groups of incomplete SCI patients receiving robotic assisted treadmill therapy, and (4) longitudinal monitoring of the natural progress of recovery in incomplete SCI subjects using motor tests for the lower extremity to investigate strength and coordination.

Obstacle stepping in children: task acquisition and performance.

The aim of this study was to investigate the locomotor capacity of children during the performance of different lower extremity tasks with increasing difficulty. Two subject groups of children (aged 6-8 and 9-12 years) and adult controls performed several motor tasks from the Zürich Neuromotor Assessment (ZNA) test, as well as a unilateral and bilateral obstacle stepping test during treadmill walking. Performance of ZNA items, changes in foot clearance, and obstacle hits were assessed. Correlations between children's age, ZNA and obstacle measures were calculated. Performance of all motor tasks improved with increasing age. All three groups improved foot clearance during unilateral obstacle stepping, while the younger children achieved a poorer performance level. During bilateral obstacle stepping, only the adult controls and the 9-12 years old children's group further improved foot clearance, while no further improvement occurred in the 6-8 years old children's group. A relationship between items of ZNA and bilateral obstacle stepping was found. It is concluded that children in the mid-childhood range are able to significantly improve performance of a high-precision locomotor task. However, children below 9 years of age have a poorer motor performance compared to older children and adults that becomes more pronounced with increasing complexity of the task. Finally, ZNA tests can improve the prediction of complex adaptive locomotor behaviour compared to calendar age alone.

Visual contribution to postural stability: Interaction between target fixation or tracking and static or dynamic large-field stimulus.

Stationary visual information has a stabilizing effect on posture, whereas moving visual information is destabilizing. We compared the influence of a stationary or moving fixation point to the influence of stationary or moving large-field stimulation, as well as the interaction between a fixation point and a large-field stimulus. We recorded body sway in 20 healthy subjects who were fixating a stationary or oscillating dot (vertical or horizontal motion, 1/3 Hz, +/-12 degrees amplitude, distance 96 cm). In addition, a large-field random dot pattern (extension: approximately 80 x 70 degrees) was stationary, moving or absent. Visual fixation of a stationary dot in darkness did not reduce antero-posterior (AP) sway compared to the situation in total darkness, but slightly reduced lateral sway at frequencies below 0.5 Hz. In contrast, fixating a stationary dot on a stationary large-field pattern reduced both AP and lateral body sway at all frequencies (0.1-2 Hz). Ocular tracking of the oscillating dot caused a peak in body sway at 1/3 Hz, i.e. the stimulus frequency, but there was no influence of large-field stimulus at this frequency. A stationary large-field pattern, however, reduced AP and lateral sway at frequencies between 0.1 and 2 Hz when subjects tracked a moving dot, compared to tracking in darkness. Our results demonstrate that a stationary large-field pattern has a stabilizing effect in all conditions, independent of whether the eyes are fixing on a stationary target or tracking a moving target.

Outcome measures in spinal cord injury: recent assessments and recommendations for future directions.

STUDY DESIGN: Review by the spinal cord outcomes partnership endeavor (SCOPE), which is a broad-based international consortium of scientists and clinical researchers representing academic institutions, industry, government agencies, not-for-profit organizations and foundations. OBJECTIVES: Assessment of current and evolving tools for evaluating human spinal cord injury (SCI) outcomes for both clinical diagnosis and clinical research studies. METHODS: a framework for the appraisal of evidence of metric properties was used to examine outcome tools or tests for accuracy, sensitivity, reliability and validity for human SCI. RESULTS: Imaging, neurological, functional, autonomic, sexual health, bladder/bowel, pain and psychosocial tools were evaluated. Several specific tools for human SCI studies have or are being developed to allow the more accurate determination for a clinically meaningful benefit (improvement in functional outcome or quality of life) being achieved as a result of a therapeutic intervention. CONCLUSION: Significant progress has been made, but further validation studies are required to identify the most appropriate tools for specific targets in a human SCI study or clinical trial.

Obstacle stepping involves spinal anticipatory activity associated with quadrupedal limb coordination.

Obstacle avoidance steps are associated with a facilitation of spinal reflexes in leg muscles. Here we have examined the involvement of both leg and arm muscles. Subjects walking with reduced vision on a treadmill were acoustically informed about an approaching obstacle and received feedback about task performance. Reflex responses evoked by tibial nerve stimulation were observed in all arm and leg muscles examined in this study. They were enhanced before the execution of obstacle avoidance compared with normal steps and showed an exponential adaptation in contralateral arm flexor muscles corresponding to the improvement of task performance. This enhancement was absent when the body was partially supported during the task. During the execution of obstacle steps, electromyographic activity in the arm muscles mimicked the preceding reflex behaviour with respect to enhancement and adaptation. Our results demonstrate an anticipatory quadrupedal limb coordination with an involvement of proximal arm muscles in the acquisition and performance of this precision locomotor task. This is presumably achieved by an up-regulated activity of coupled cervico-thoracal interneuronal circuits.

Vertical perturbations of human gait: organisation and adaptation of leg muscle responses.

During the last several years, evidence has arisen that the neuronal control of human locomotion depends on feedback from load receptors. The aim of the present study was to determine the effects and the course of sudden and unexpected changes in body load (vertical perturbations) on leg muscle activity patterns during walking on a treadmill. Twenty-two healthy subjects walking with 25% body weight support (BWS) were repetitively and randomly loaded to 5% or unloaded to 45% BWS during left mid-stance. At the new level of BWS, the subjects performed 3-11 steps before returning to 25% BWS (base level). EMG activity of upper and lower leg muscles was recorded from both sides. The bilateral leg muscle activity pattern changed following perturbations in the lower leg muscles and the net effect of the vertical perturbations showed onset latencies with a range of 90-105 ms. Body loading enhanced while unloading diminished the magnitude of ipsilateral extensor EMG amplitude, compared to walking at base level. Contralateral leg flexor burst activity was shortened following loading and prolonged following unloading perturbation while flexor EMG amplitude was unchanged. A general decrease in EMG amplitudes occurred during the course of the experiment. This is assumed to be due to adaptation. Only the muscles directly activated by the perturbations did not significantly change EMG amplitude. This is assumed to be due to the required compensation of the perturbations by polysynaptic spinal reflexes released following the perturbations. The findings underline the importance of load receptor input for the control of locomotion.

Standardized assessment of walking capacity after spinal cord injury: the European network approach.

OBJECTIVES: After a spinal cord injury (SCI), walking function is an important outcome measure for rehabilitation and new treatment interventions. The current status of four walking capacity tests that are applied to SCI subjects is presented: the revised walking index for spinal cord injury (WISCI II), the 6 minute walk test (6MinWT), 10 meter walk test (10MWT) and the timed up and go (TUG) test. Then, we investigated which categories of the WISCI II apply to SCI subjects who participated in the European Multicenter Study of Human Spinal Cord Injury (EM-SCI), and the relationship between the 10MWT and the TUG. METHODS: In the EM-SCI, the walking tests were applied 2 weeks and 1, 3, 6 and 12 months after SCI. We identified the WISCI II categories that applied to the EM-SCI subjects at each time point and quantified the relationship between the 10MWT and the TUG using Spearman's correlation coefficients (rho) and linear regression. RESULTS: Five WISCI II categories applied to 71% of the EM-SCI subjects with walking ability, while 11 items applied to 11% of the subjects. The 10MWT correlated excellently with the TUG at each time point (rho>0.80). However, this relationship changed over time. One year after SCI, the time needed to accomplish the TUG was 1.25 times greater than the 10MWT time. DISCUSSION: Some categories of the WISCI II appear to be redundant, while some discriminate to an insufficient degree. In addition, there appear to be ceiling effects, which limit its usefulness. The relationship between the 10MWT and TUG is high, but changes over time. We suggest that, at present, the 10MWT appears to be the best tool to assess walking capacity in SCI subjects. Additional valuable information is provided by assessing the needs for walking aids or personal assistance. To ensure comparability of study results, proposals for standardized instructions are presented.

Facilitation of spinal reflexes assists performing but not learning an obstacle-avoidance locomotor task.

The aim of this study was to investigate spinal reflex (SR) modulation during the performance and learning of a precision locomotor task. Healthy subjects had to minimize foot clearance when repeatedly stepping on a treadmill over a randomly approaching obstacle. The subjects walked with reduced vision and were informed about the approaching obstacle and task performance by acoustic warning and feedback signals, respectively. SRs were randomly evoked by tibial nerve stimulation (with non-nociceptive and nociceptive stimulus intensity) during the mid-stance phase in both normal and pre-obstacle stepping. Foot clearance and electromyographic activity of the tibialis anterior and biceps femoris muscles of the right leg were analysed. Only if a delay was introduced between warning signal and nerve stimulation, was the SR amplitude in both muscles enhanced prior to obstacle steps compared with normal steps for both stimulus intensities. Thus, the reflex enhancement depended on the subject's awareness of the approaching obstacle. Improved performance was reflected in a decreased foot clearance, but did not correlate with the course of SR amplitude. It is concluded that obstacle stepping is associated with a facilitation of SR pathways, probably by supraspinal drive. This facilitation might provide assistance in safe obstacle stepping, e.g. to compensate quickly if resistance is encountered.

Modulation of leg muscle activity and gait kinematics by walking speed and bodyweight unloading.

During rehabilitation, many patient groups are being trained using bodyweight-supported treadmill training. However, little is known about modulation of time and distance parameters, joint movements and leg muscle EMG patterns by very low walking speeds or partial bodyweight unloading. We collected data from 20 healthy young volunteers who walked on a treadmill at walking speeds varying between 0.5 and 5.0 km h(-1) (0.14-1.39 ms(-1)) and with 0%, 25%, 50% and 75% bodyweight unloading. We found that cadence and stride length were largely influenced by walking speed, while bodyweight unloading influenced these measures only at 75%. However, the relative duration of the gait phases changed largely only at walking speeds less than 2.5 km h(-1), but was influenced by all different bodyweight unloading conditions. Joint trajectories of knee and ankle joint, as well as leg muscle EMG activity patterns changed largely at walking speeds slower than 2.5 km h(-1) and with 75% bodyweight unloading. We concluded that the parameters we investigated changed minimally at walking speeds faster than 2.5 km h(-1) and bodyweight unloading conditions less than 50%. Therefore, standards for EMG activity and joint angle trajectories should only be compared when the training is done with velocities higher than 2.5 km h(-1) and less than 50% body weight unloading.

Improving walking assessment in subjects with an incomplete spinal cord injury: responsiveness.

STUDY DESIGN: Prospective longitudinal study. OBJECTIVES: To investigate the responsiveness of the Walking Index for Spinal Cord Injury II (WISCI II), 6-Min Walk (6MWT) and 10-Meter Walk Tests (10MWT) for the assessment of walking capacity in incomplete spinal cord injury (iSCI) and to validate these tests with the lower extremity motor score (LEMS). SETTING: European Multicenter Study of Human Spinal Cord Injury. METHODS: The walking tests of 22 iSCI subjects who achieved functional ambulation and could stand or walk within 1 month after iSCI were analyzed at 3, 6 and 12 months after injury. Responsiveness was assessed by determining differences between the time intervals, and Spearman's correlation coefficient was calculated to quantify validity. RESULTS: All walking tests were able to assess the improvement of walking capacity within the first 3 months after injury. Between 3 and 6 months, only the 10MWT and 6MWT were responsive to the ongoing improvement in locomotor capacity. Overall, correlations between the tests were good within the first month, but became poorer over time. CONCLUSION: The 6MWT and 10MWT were more responsive in demonstrating an improvement in walking capacity compared to the WISCI II. The testing of functional outcome after iSCI as provided by the ordinal ASIA motor score can be improved by interval scaled measures. This allows increasing the responsiveness of functional outcome measures and should be advantageous in assessing therapeutical approaches in iSCI subjects. In iSCI subjects with walking ability, we recommend the additional use of timed tests to monitor improvement in locomotor capacity.

Limits of locomotor ability in subjects with a spinal cord injury.

STUDY DESIGN: Case-control study with pair-wise matching. OBJECTIVES: To investigate the level of locomotor performance in spinal cord injury (SCI) subjects. SETTING: Spinal Cord Injury Center of Balgrist University Hospital in Zurich, Switzerland. METHODS: Seven well-recovered SCI subjects and controls performed three tests: (1) 6-Min Walk Test (6MWT), (2) treadmill walking with full and restricted vision and (3) repetitive stepping over an obstacle. In the latter task, the subjects had to step over an obstacle as low as possible, without touching it. The subjects performed three blocks of 50 trials each. They received acoustic signals that indicated obstacle approach and foot clearance. During the third block, vision was restricted. The main outcome measures were: (1) distance walked, (2) leg muscle activity, joint kinematics and double-support duration, (3) foot clearance, number of obstacle hits, leg muscle activity, joint kinematics and swing phase duration. RESULTS: (1) No difference in the 6MWT between the groups. (2) Independent from vision, the double-support duration was increased in SCI subjects. (3) SCI subjects learned generally at a slower rate. They touched the obstacle more frequently when vision became restricted. Furthermore, we observed differences between the groups in rectus femoris and tibialis anterior EMG activity and swing phase duration. CONCLUSION: It appears that even in well-recovered SCI subjects, small deficiencies in functional gait can be determined, which remain undetected by clinical walking tests. In addition, SCI subjects learned a locomotor task at a slower rate. This indicates that rehabilitation programs should train (adaptive) locomotion in different (sensory) environments.

The influence of age on learning a locomotor task.

OBJECTIVE: Knowledge about locomotor task performance and learning in the elderly is important in optimizing rehabilitation strategies. The aim of this study was to evaluate differences between young and elderly subjects in the acquisition and performance of a precision locomotor task, with full and restricted vision. METHODS: The subjects walked on a treadmill and had to step as low as possible over an obstacle, without touching it. They received acoustic warning and feedback signals, indicating obstacle appearance and foot clearance, respectively. Full vision was provided during the first two runs and became restricted during the third run. The number of obstacle hits and adaptations in foot clearance, leg muscle activity, range of motion of leg joints and swing phase duration were assessed. RESULTS: With vision, the performance improved in both groups. Restricted vision reduced the task accuracy in both the young and the elderly. However, only the young subjects regained optimal foot clearance with practice. CONCLUSIONS: Elderly subjects rely more on visual control when acquiring and performing a precision locomotor task. We suggest that this is due to an impaired function of proprioceptive feedback mechanisms, which can replace visual information in young subjects. SIGNIFICANCE: In the elderly, therapeutical attention should be directed towards optimizing the use of the remaining proprioceptive inputs.

Obstacle avoidance during human walking: H-reflex modulation during motor learning.

The goal of this study was to investigate changes of H-reflex amplitudes during a motor learning task. Subjects with reduced vision were instructed to step over an obstacle on a treadmill as low as possible, while the soleus H-reflex was elicited. Acoustic warning and feedback signals about performance were provided. Performance improvement was associated with a decrease of muscle activity, needed to step over the obstacle (rectus femoris, biceps femoris, tibialis anterior and gastrocnemius medialis muscles), and of foot clearance, while joint angle trajectories from knee and ankle became more stable. The experiment consisted of five runs, three with normal treadmill walking and two with randomly stepping over the obstacle (100 times). H-reflexes were elicited at early and late stance phase before stepping over the obstacle. H/M ratio, latency and duration were determined. The values of these measures were calculated for the onset and end of a run and their course over time was evaluated using a correlation coefficient. The largest adaptations with a significant increase of reflex amplitude occurred during the first obstacle run. This increase lasted only briefly and the reflex amplitudes decreased to their previous values. During the later obstacle run, no H-reflex modulation occurred. It is concluded that a motor learning task causes adaptational effects not only on performance, but also on H-reflex responses. The results indicate that most of the modulation of H-reflexes is probably due to supraspinal influences on reflex transmission. The observations made are probably less specific for this motor task (stepping over the obstacle), but rather associated with the increased attention required by the motor learning task during the first obstacle run.

Obstacle avoidance during human walking: transfer of motor skill from one leg to the other.

The aim of this study was to evaluate whether a newly acquired locomotor skill can be transferred to the mirror condition. Subjects were trained to step over an obstacle on a treadmill, the appearance of which was signalled by an acoustic stimulus, while visual information was prevented. Feedback information about foot clearance was provided by acoustic signals. During two successive runs (each consisting of 100 steps over the obstacle) the same leg was leading (i.e. the leg crossing the obstacle first). In the following third run, the leading and trailing legs were changed. During each of the three successive runs the adaptational changes were analysed by recording leg muscle electromyographic (EMG) activity, joint angle trajectories and foot clearance over the obstacle. The training effect gained between the first and second runs and the transfer to the mirror condition (third run) were evaluated. Adaptational changes of all measures, except ankle joint trajectory, could to a significant extent be transferred to the mirror condition. No side-specific differences in the amount of transfer were found, neither from the right to the left side, nor vice versa. These observations are at variance with adaptational changes observed during split-belt walking or one-legged hopping on a treadmill, where no transfer to the mirror condition occurred. It is assumed that this might be due to the specific requirements of the tasks and the leg muscles involved. While in the split-belt and hopping experiments leg extensor muscles are mainly involved, leg flexors predominate in the performance of the present task. It is hypothesised that the learning effects observed in the present experiments are mediated at a higher level (e.g. brainstem) of locomotor control.