Internally Versus Externally Guided Body Weight-Supported Treadmill Training (BWSTT) for Locomotor Recovery Post-stroke

• Physiological and biomechanical effects of Body Weight Supported Treadmill Training [ Time Frame: 20 Weeks plus a 6 month follow-up. ] [ Designated as safety issue: No ]
  

The investigators hypothesize that key differences are present in both biomechanical (i.e., loading, kinetic energy at toe off, trunk energetic cost, muscle tendon lengthening velocity) and physiological (i.e., temporal patterning of intermuscular electromyogram [EMG], central reflex modulation) constituents of locomotion between internally and externally-driven modes of BWSTT. Externally-driven BWSTT (Lokomat) produces a more controlled, consistent and mechanically appropriate locomotor pattern promoting positive adaptation in the spinal locomotor circuitry and improved integration of descending motor drive which in combination promote improved gait dynamics. These therapeutically-induced differences will be manifest in the ability to generalize the effects of BWSTT to overground locomotion and will include: improved gait symmetry, increased knee flexion during swing phase, normalization of limb kinetic energy at the stance-to-swing transition, and the ability to scale gait speed effectively between self-selected and fast speeds. The investigators further hypothesize that externally-driven BWSTT will produce more persistent treatment-related effects.

In this pilot study, the investigators will conduct a series of twelve single-case, ABA or BAB, designs in which hemiparetic subjects will experience both internally and externally-driven BWSTT. Training parameters (i.e., body weight support, treadmill speed and support stiffness) will be held constant between modes of BWSTT and physiological and biomechanical responses will be compared between modes for individual subjects. Adaptations in overground gait parameters (i.e., limb kinetic energy at toe off, knee flexion, trunk mechanical energetic cost, gait symmetry, gait speed) will be compared using reference normal gait data obtained from non-disabled, age and gender-matched control subjects walking at matched speeds. The response of non-disabled control subjects to both forms of BWSTT will also be studied.