Predicting recovery and outcomes for the lower limb after stroke

Abstract

Regaining the ability to walk after stroke is the most common rehabilitation goal for patients. Recovery from motor impairment plays a role in achieving independent walking and the symmetry of the walking pattern. The objectives of this thesis were to identify predictors of recovery from lower limb motor impairment and walking outcome at the subacute stage of stroke, and to investigate an intervention for improving gait symmetry at the chronic stage of stroke. Two prediction studies were conducted using clinical measures, transcranial magnetic stimulation (TMS), and magnetic resonance imaging (MRI), within the first week of stroke, with follow-up assessments at 6 and 12 weeks post-stroke. The first study (n = 32) found that patients recovered about 70% of their initial lower limb motor impairment, with baseline motor impairment (lower limb Fugl-Meyer score) the only predictor of recovery. This is the first report of proportional recovery from lower limb motor impairment. The second study (n = 41) identified variables that predicted whether a patient would walk independently by six or 12 weeks, or remain dependent at 12 weeks post-stroke. The study produced the Time to Walking Independently after STroke (TWIST) algorithm that made accurate predictions for 95% of patients, and is the first to predict when a patient will walk independently post-stroke. TMS and MRI measures were not significant predictors in either study, which may be due to the presence of alternate descending pathways to the lower limb. The final study investigated unilateral step training and conventional treadmill training in 20 patients with chronic stroke. The effects of training were dependent on the direction of gait asymmetry, as both types of training improved step-length asymmetry only for participants who took a shorter step with their paretic leg. This highlights the need to identify subsets of patients who are most likely to respond to a given intervention. In summary, this thesis presents novel findings of proportional recovery from lower limb impairment, and the TWIST algorithm for predicting when a patient will achieve independent walking. Unilateral step training in the chronic stage had modest effects, but may have benefit at the subacute stage.