Vol. 22 • Issue 15
• Page 24
Bodyweight-supported treadmill training has become an essential intervention widely used with the stroke and spinal cord injury populations during rehabilitation. The inability to walk can be a devastating disability that affects individuals following severe stroke and in an even larger percentage of those who have spinal cord injury (SCI).
Research during the past two decades has ranged from studying animal models to examining the neural mechanisms within the central nervous system that demonstrate potential for locomotor recovery, to examining various clinical applications of locomotor training strategies and functional outcomes. This research has provided support for improved ambulation outcomes in both the motor-incomplete SCI population and stroke population.
Additionally, the research has contributed to a shift in the rehabilitation mindset of using compensatory training strategies-learning to adapt movement strategies and behaviors to achieve a goal-to one that focuses on activity-based training or task-specific training.1Bodyweight-supported gait training is a primary example of using task-specific training strategies to achieve a desired outcome; in this case, locomotion.
Role of Robotics
Bodyweight-supported treadmill training (BWSTT) can be applied in a variety of methods. One can provide manually assisted BWSTT over-ground, over a treadmill, or with the support of robotics such as the Lokomat, a robotic driven gait orthosis (DGO) developed to automate locomotor training.2The bodyweight support provides individuals with compromised nervous system function with the ability to initiate gait training despite having little to no ambulatory function.3,4The robotic device provides a simplified setting for postural control and balance, and minimizes the need to generate all the necessary forces for walking and coordinating stepping, which allows individuals to develop more effective and efficient movement strategies.1,3,5This intervention can be applied early during the rehab process and can prevent common compensatory movements-for example, altered gait patterns or use of extensive lower-extremity bracing.
In a study that examined functional outcomes following bodyweight support combined with treadmill training in subjects following stroke, the researchers found that "retraining gait in severely impaired stroke subjects with a percentage of their bodyweight supported resulted in better walking and postural abilities than did gait training in patients bearing their full weight."6
No matter which strategy of bodyweight-supported gait training is chosen, there are main principles and parameters that need to be applied in order to facilitate locomotion. As previously stated, locomotion training is a task-specific intervention. If one's goal is to achieve the ability to walk, the therapist must train walking. This intervention must be carefully directed and performed in a repetitive manner. Extensive research has highlighted key parameters that promote improved ambulatory outcomes while performing this activity-specific intervention.1,7,8There are required sensory inputs that contribute to the activation pattern of the lower extremities. These include achieving hip extension during late-stance phase, in order to provide proprioceptive input into the hip flexors to initiate the swing phase of gait. In addition, maximizing loading in the lower extremities while maintaining proper gait kinematics is key in increasing extensor muscle activity and facilitating locomotion. The input from the ankle extensor muscle group is especially necessary to regulate the stance-to-swing transition.
Research indicates that stepping is facilitated by limiting upper-extremity weight-bearing and demonstrating a reciprocal arm swing in a natural and coordinated fashion. Increased upper-extremity weight-bearing can even inhibit stepping in the lower extremities.
Research also demonstrates that by mimicking normal walking speeds, patients show improved independence with stepping, increased LE muscle activation and improved kinematics. By incorporating these principles into BWSTT, patients will benefit from optimal locomotor outcomes.
With the development of various BWSTT systems, both with and without the use of robotics, clinicians may have difficulty determining which approach to locomotor training will prove to be most beneficial for each patient. With manually assisted bodyweight-supported treadmill training, one to three therapists may be required to assist/advance the lower extremities as well as provide postural support or stability. Due to the repetitive demands and the need to maintain proper gait kinematics, this approach can be physically taxing on trainers.
As a result of the physical burden on the trainers, maintaining optimal stepping patterns and necessary alignment required for facilitating locomotion may become difficult. With the development of robotic gait training devices, the automated exoskeleton facilitates a normal gait cycle by guiding the individual's lower extremities in a coordinated, reciprocal pattern according to established joint kinematics. This device can allow for increased duration of training and improve overall gait quality, while maintaining the proper gait kinematics required throughout the entire training duration.
Due to these factors alone, robotic gait training may appear to be more effective, both in terms of outcomes and cost-effectiveness, and allow for increased duration of training, easier repeatability, and decreased therapist/trainer demands and injuries. More recently, research has examined the effects of robotic training versus conventional physical therapy, as well as robotic gait training versus BWSTT, to further examine functional outcomes, gait quality, postural control and balance.
Locomotor Training Reseach
In a locomotor training study that examined the effects of using the robotic device versus conventional physical therapy with 16 acute stroke survivors, robotic gait training was found to improve ambulation on functional and impairment scales, which included gait speed, endurance, muscle strength and tone, as compared with conventional PT.2In another recent study using a robotic device for gait training with 30 acute stroke survivors, individuals who participated in the training for four weeks demonstrated a significantly longer stance phase on their involved lower extremity as compared with those who received conventional PT alone for four weeks.3This supported the theory of "forced-use" effect on the involved lower extremity. However, there was no significant difference in functional score, or in walking speed or cadence.
More recently, a pilot study was performed that compared two forms of bodyweight-supported gait training: manually assisted BWSTT versus robotic gait training with 16 individuals with chronic hemiplegia.4The study examined the effect of each approach on self-selected over-ground walking speed and paretic step-length ratio (comparison of step length of paretic lower extremity to the step length of the non-involved lower extremity), as well as fast over-ground walking speed, 6-minute walk test, and clinical measures for balance, motor scores and a battery of physical performance tests.
Based on the outcomes of this study, no significant difference was noted between the groups in terms of walking speed or paretic step-length ratio. However, within the robotic device group, there was a significant change pre- to post-intervention with self-selected walking speed, paretic step-length ratio, and four out of six other clinical measures versus the manually assisted BWSTT group, who improved in balance scores only.
Another factor that deserves more consideration during the rehabilitation process following both stroke and SCI is the cardiovascular benefits of BWSTT. Evidence shows that stroke patients who have participated in robotic gait training have seen a significant increase in muscle tissue and a significant loss of fat mass.3Likewise, individuals who participated in conventional PT demonstrated increased weight and fat mass.
This can only suggest that robotic gait training is associated with an increase in aerobic metabolism similar to cardiovascular training. The cardiovascular benefits of this training also may play a major role in the prevention of secondary complications that can occur following a stroke or SCI.
Bodyweight-supported treadmill training can be applied to a variety of populations, ranging from mild to severe disability, and can greatly impact whether one is a wheelchair user, a short-distance household ambulator, or one who is fully ambulatory.
References are available at www.advanceweb.com/pt or by request.
Kathleen Doyle is therapy practice leader for the spinal cord injury program at Spaulding Rehabilitation Hospital in Boston, MA.
Measuring Gait Changes After Intervention-Applying Technology in the Clinic
By Dobrivoje S. Stokic, MD, DSc, Terry S. Horn, PhD, John M. Ramshur, BS, and John W. Chow, PhD
The methods for identifying gait impairments after stroke include clinical assessment scales, observational gait analysis, and instrumented measurement techniques of various complexities. Ideally, the applied technique should be easy to administer and have sound psychometric properties. There is an abundance of previous research evaluating validity and reliability of various clinical scales for assessing mobility and gait, either exclusively or as a part of the global outcome measure. Determining an agreement between measures derived from different instrumented techniques is required for proper interpretation of research results and making valid clinical comparisons.
Electronic walkways have been increasingly used for the assessment of gait after stroke, whereas 3D video-based gait analysis is typically used in more advanced settings. However, a formal comparison of temporospatial parameters derived from electronic walkway and video-based gait analysis has not been made in people with stroke. This is of interest because gait changes after stroke, such as foot landing flat or in plantar flexion and variable pressures over different foot areas.may potentially introduce discrepancies in temporospatial gait parameters reported by these two systems among people who sustained a stroke.
The goal of method comparison studies is to determine the agreement between the methods or instruments designed to measure the same function. The obtained results should be appraised within the proper context to determine their implications. To capture the true change in walking performance over time or as a result of intervention, the variability inherent to measurement techniques or instruments should be far smaller than the observed difference in outcome measure. This is essential when the expected change is modest.
All gait parameters were significantly different between healthy and stroke subjects. As expected, stroke subjects walked slower than healthy, with shorter strides and steps, and spent proportionally less time in single support and more time in total support. Stroke subjects who used a walking aid, compared with those without a walking aid, walked significantly slower, with bilaterally shorter strides and steps, and spent proportionally less time in single support (P_0.05) and more time in total support. Asymmetries in walking were apparent in stroke subjects (e.g., longer step length and smaller percent single support on the affected side), but the mean values did not differ significantly.
Our results may help clinicians choose the appropriate instrument for assessing gait function in healthy subjects and in people with stroke. The findings suggest that the two gait analysis methods may be used interchangeably across the range of walking velocities. For example, a motion analysis system could be used at baseline and study endpoints to acquire comprehensive temporospatial, kinematic, and kinetic data, whereas temporospatial data may suffice for interim or follow-up assessments conveniently performed using an electronic walkway. Furthermore, published reports using either instrument in healthy and stroke populations may be compared with confidence in meta-analyses.
From Stokic, D., et al. (2009). Agreement between temporospatial gait parameters of an electronic walkway and a motion capture system in healthy and chronic stroke populations. American Journal of Physical and Medical Rehabilitation, 88(6).