Key points are not available for this paper at this time.
PURPOSE: To test the hypothesis that partial body weight-supported treadmill training (PBWSTT) provides more effective gait training than an equally supportive but less physiologic aggressive bracing assisted walking (ABAW) program. METHODS: Following informed consent, patients participating in an inpatient rehabilitation program with significant leg weakness and need for at least moderate assistance for walking, without orthostatic hypotension, symptomatic dyspnea, or angina pectoris were randomized to receive PBWSTT vs. ABAW. PBWSTT was provided by a commercially available, overhead motorized hoist attached to a parachute-type body harness, which provided partial support of the patient's weight over a treadmill. Therapists assisted with weight shifting, leg advancement, and foot placement as needed. ABAW included aggressive early therapist-assisted ambulation using knee-ankle combination bracing and hemi-bar if needed. Treatment sessions of up to 45 minutes per day, five days per week were given as tolerated for the duration of the inpatient stay or until patients could walk over-ground unassisted. All patients had an additional 45-minute session of functionally oriented physical therapy each day with or without bracing as judged appropriate by the patient's individual therapist. RESULTS: Fifty-six patients a mean age of 71 +/- 1 SEM were enrolled 40 +/- 3 days post stroke. Although the outcome of the two groups as a whole did not differ, a subgroup with major hemispheric stroke defined by the presence of hemiparesis, hemianopic visual deficit, and hemihypesthesia who received more than 12 treatment sessions showed significantly better over-ground endurance (90 +/- 34 vs. 44 +/- 10 meters) and speed scores (12 +/- 4 vs. 8 +/- 2 meters/minute) for PBWSTT vs. ABAW, respectively. CONCLUSIONS: PBWSTT and ABAW are equally effective gait training techniques except for a subset of patients with major hemispheric stroke who are difficult to mobilize using ABAW alone.
Kosak et al. (Wed,) studied this question.