|
North American Congress on Biomechanics Canadian Society for Biomechanics - American Society of Biomechanics University of Waterloo Waterloo, Ontario, Canada August 14-18, 1998 |
|
Hemiparesis is linked to a high risk of falling (Gresham et al., 1995). A potentially important factor, which has not been examined, is the effect of the cortical lesions and associated impairments on the ability to execute the rapid, reflex-like com- pensatory limb movements that play a critical role in maintaining upright stance. We examined the influence of chronic hemiparesis on the control of these compensatory stepping and grasping reactions in stroke patients, using unpredictable antero- posterior platform motion to elicit the reactions.
Previous studies have revealed that stepping or grasping movements of the limbs are prevalent stabilizing reactions in maintaining upright stance. In healthy young and older adults, such compen- satory reactions feature rapid onset latencies and fast speed of movement (Maki et al., 1997).
Patients with stroke-induced hemiparesis often exhibit impaired ability to control the affected limbs (Shumway-Cook et al., 1995), and hence may present different compensatory strategies. For example, the impaired ability to control the paretic leg could lead to a preference to step with the un- affected limb in order to achieve the rapid and accurate foot movement required during compensa- tory stepping reactions. However, the benefits of stepping with the non-paretic leg could be offset by the delays that may be encountered in unloading this leg, since hemiparetic patients will typically stand with greater weight-bearing on the unaffected leg. Furthermore, this asymmetry in weight-bearing could exacerbate impairments in the ability to shift the weight over the stance leg, prior to swing phase, and hence could lead to heightened instabil- ity during the swing phase. Impaired control of compensatory stepping could lead to greater dependence on arm movements to increase external support; however, ability to grasp with the paretic arm is also quite likely to be severely impaired.
The primary objectives of this initial study were to determine how chronic hemiparesis, due to stroke, affects the ability to control rapid compensatory stepping and grasping movements of the limbs, and to elucidate the specific control strategies that are adopted by these patients. The compensatory stepping results are presented in this abstract. Analysis of results from compensatory grasping experiments is currently in progress and the findings will be presented at the conference.
This study involved eight patients with chronic hemiparesis (>6 months post-stroke), excluding those with brain stem or cerebellar lesions, spatial neglect or vision deficits. Patients were also screened for medication use and peripheral or central neurological problems (other than the stroke) that may influence their balance. Patients were included if they were clinically stable and ambulatory (without the use of an aid). Eight healthy control subjects were also tested. These subjects were asked to stand asymmetrically (70% of weight on the dominant leg) to mimic the asymmetric weight-bearing of the stroke patients.
Stepping and grasping reactions were evoked by unpredictable platform translation. Handrails (at the platform perimeter) and a harness were used as safety precautions. Testing in the first four stroke subjects and first four controls focussed on compensatory stepping. Testing in the remaining four stroke patients and controls focussed on arm reactions. Forward and backward platform translations were administered in random order. Light-cued voluntary grasping reactions were also tested in the arm-movement session. In the first set of 10 'unconstrained' trials, in which compensatory stepping reactions were elicited, subjects were not given any specific instructions on how to react. In the grasping trials, subjects were instructed to grasp a horizontal handrail (mounted transversely in front of them, at waist height) as quickly as possible with either one arm or both arms following either the light cue or the platform perturbation.
Force plates, mounted on the platform, were used to determine step timing; the measured shear forces were integrated to determine center-of-mass (COM) motion. Foot and arm movements were recorded using a video system. EMG's were also recorded.
Both stroke patients and control subjects relied on compensatory stepping to maintain stability in re- sponse to the imposed perturbations in 100% of the 'unconstrained' trials. However, unlike healthy subjects, the stroke patients most commonly stepped with the loaded (non-paretic) leg (68% of trials). This is striking because the patients stood, prior to perturbation, with greater weight-bearing on the unaffected leg, whereas a similar amount of pre- perturbation asymmetry in weight-bearing in healthy subjects led to an overwhelming tendency to step with the unloaded limb (91% of trials). In 41% of trials, stroke patients countered, to some extent, the tendency for lateral COM motion (toward the unsupported side) arising from lifting the loaded limb by generating a medial-lateral (ML) anticipatory adjustment (APA) prior to unloading the limb; however, this led to delayed step execution and increased antero-posterior (AP) COM displacement. When the non-paretic leg was lifted without a preceding APA (27% of trials), these problems were avoided; however, the COM fell laterally during step execution (see Fig. 1). Steps with the paretic leg (32% of trials) tended to be smaller and slower than steps executed with the non-paretic leg (see Fig. 2).
Fig.1. Stepping with non-paretic leg: effect of APA on COM at foot-off (FO) and foot-contact (FC). Schematic drawing shows relation to foot position.
Fig.2. Effect of stepping with paretic leg(*p<0.10, **p<0.05).
In contrast to healthy young adults, who typically recovered stability with a single step (84% of trials) and who never grasped the safety handrails, stroke patients had a much greater tendency to use multi- ple steps and/or grasping reactions in order to reco- ver balance, regardless of the stepping strategy adopted. For steps taken with the non-paretic leg, multiple-stepping and/or grasping occurred in 92% of the responses that included an APA and 75% of the responses that had no APA. For steps taken with the paretic leg, multiple-stepping and/or grasping occurred in 60% of the responses.
These preliminary results demonstrate that stroke patients may adopt one of three strategies for controlling compensatory stepping, yet each of these strategies has negative consequences that could jeopardize stability. Stepping with the paretic (unloaded) limb led to foot movements that tended to be small and slow. Stepping with the non- paretic (loaded) leg allowed for more robust stepping movements, but created difficulties in the control of the COM. If these steps were executed without an anticipatory adjustment (ML APA), the COM tended to fall laterally during the swing phase. Alternatively, if the ML APA was included, lateral stability was improved; however, the additional time required to execute the anticipatory phase led to substantial delays in step initiation and execution. The prevalence of multiple-step and grasping reactions in the stroke subjects suggests that the problems observed in controlling compensatory stepping do, in fact, compromise functional stability. The inability to execute an effective stepping reaction emerged even though patients demonstrated a relatively high degree of functional recovery when assessed clinically. These findings suggest that control of compensatory stepping should be considered in the assessment and rehabilitation of the stroke patient.
Gresham GE et al. Post-stroke Rehabilitation. Clinical Practice Guideline, No. 16. AHCPR, 1995.
Maki BE et al. Phys Ther, 77, 488-507, 1997.
Shumway-Cook A et al. Motor Control: Theory & Practical Applications. Williams & Wilkins, 1995.
This study was supported by grants from MRC to BEM and SEB, and a grant from NSERC to WEM.
