Sensorimotor impairments following stroke often involve complex pathological changes across multiple joints and multiple degrees of freedom of the arm and hand, thereby rendering them difficult to diagnose and treat. The objective of this study is to evaluate multi-joint neuromechanical impairments in the arm and hand, then conduct impairment-specific treatment, and determine the effects of arm versus hand training and the effects of passive stretching before active movement training.

Sensorimotor impairments following stroke can lead to substantial disability involving the upper extremity. These impairments often involve complex pathological changes across multiple joints and multiple degrees-of-freedom of the arm and hand, thereby rendering them difficult to diagnose and treat. Many potential mechanisms, such as weakness, motoneuronal hyperexcitability, and elevated passive impedance, can contribute and it is currently unclear where to focus treatment. The objectives of this study are to address allocation of therapy resources between the arm and hand and to examine the benefits of combining passive stretching with active movement training.

Aim 1. To compare the efficacy of training the arm versus the hand in promoting upper extremity rehabilitation.

Hypothesis 1: Treating the proximal larger joints in the arm alone will lead to greater improvement than treating the distal hand alone.

Aim 2. To examine the efficacy of combining passive stretching with active (assistive or resistive) training for the shoulder, elbow, wrist, and hand.

Hypothesis 2: Multi-joint intelligent stretching followed by active (assistive or resistive) movement facilitated by use of the IntelliArm arm rehabilitation robot and a Hand rehabilitation robot will improve motor control of the upper extremity more than standard movement therapy alone.

Subjects will be assigned randomly with equal chance to one of four groups. Groups are split into 2 conditions based on stretching and 2 conditions based on target of intervention (arm or hand). Half of all the subjects will be assigned to the stretching groups and the other half to the passive movement groups. Half of the subjects will be assigned to the arm-training and the remaining half to hand-training groups. Arm-training groups will use the IntelliArm, hand-training groups will use the hand robot. For those assigned to the stretching groups, subjects will complete up to 30 minutes of passive stretching with the IntelliArm or the hand robot. For those assigned to the passive movement condition, subjects will do the robot according to their group assignment and wear it for up to 30 minutes with little to no stretching preceding the active therapy session. For each group, the initial about 30 minutes of stretching or relaxing will be followed by 45-60 minutes of active therapy with the IntelliArm or hand robot (depending on group assignment), for a total session time of 75-90 minutes.

The 4 groups of subjects will be compared against each other.

Inclusion Criteria:

• First focal unilateral lesion, ischemic or hemorrhagic
• Had a stroke 1-12 months prior to enrollment
• Rated between stages 2-4 on the Chedoke McMaster Stroke Assessment Impairment Inventory: Stage of Recovery of the Arm and Hand

Exclusion Criteria:

• Apraxia
• Score of less than 22 on the Mini Mental Status Exam
• Severe pain in the shoulder by a self-rating of 7 out of 10 or greater
• Severe contracture in the upper extremity
• Unable to sit in a chair for 3 consecutive hours
• Unrelated musculoskeletal injuries
• Poor fit into equipment used in study
• Botox injection in upper extremity within 4 months
• Concurrent participation in gait or upper extremity intervention studies