Individuals with chronic stroke have long-term walking problems that limit community engagement and quality of life, lead to secondary disabilities, and increase healthcare costs and burden. These walking issues often persist despite rehabilitation. One novel target for stroke gait rehabilitation is interlimb coordination-the phase-dependent cyclical relation of the legs. Interlimb coordination is altered during walking after stroke, compromising walking stability, phase transitions, and responses to perturbation and contributing to motor compensation. It is unclear what neural pathways contribute to impaired interlimb coordination after stroke and what impact this has on walking-related outcomes.

This proposal consists of two aims to address these issues, with the long-term goal of developing therapeutic interventions to improve interlimb coordination and walking after stroke.

Aim 1 will identify which neural sources contribute to impaired interlimb coordination after stroke. During bilateral, cyclical recumbent stepping (analogue of walking), interlimb coordination will be assessed as relative leg phasing. During the task, transcranial magnetic stimulation and peripheral nerve stimulation will be applied to assess supraspinal, interhemispheric, spinal interneuronal, and sensory pathways. The relation of interlimb coordination with these outcomes will be assessed to determine potential contributors.

Aim 2 will test the association between interlimb coordination and walking after stroke. Interlimb coordination will be quantified during split-belt treadmill walking, and associations with walking speed, endurance, mobility, independence, daily activity, quality of life, and community engagement will be tested.

An additional exploratory aim will determine the effect of targeted neuromodulation on lower limb interlimb coordination. Electrical stimulation will be applied to three locations in a cross-over study: the primary motor cortex (supraspinal/interhemispheric), thoracolumbar spine (spinal interneuronal), and peripheral nerves (sensory).

Walking problems are common, disruptive, and persistent after stroke. Of 9.4 million people in the U.S. with chronic stroke, most have long-term walking problems such as decreased walking speed and endurance, and ~36% (~3.4 million) cannot walk independently. These walking issues have secondary impacts, including limited community integration, impaired quality of life, decreased physical activity, increased risk of other diseases and disabilities, and ~$56 billion in annual costs. Hence, a major goal of stroke rehabilitation research is to improve walking. Most walking rehabilitation approaches have focused on restoring neurotypical patterns in the more affected limb without regard for the less affected limb or focused on improving walking function (e.g., maximizing walking speed) without regard for how improvements are achieved. Unfortunately, walking limitations persist even after intensive rehabilitation, perhaps because these current approaches do not adequately address the complex neural control required for walking.

Interlimb coordination is essential for walking, but impaired after stroke. One construct that reflects the complex control of walking is interlimb coordination, the phase-dependent cyclical relation of the legs [and arms]. Bipedal locomotion cannot occur without interlimb coordination, and precise control is essential for stability, phase transitions, and responses to perturbations. After stroke, interlimb coordination becomes impaired. The investigators and others have shown that during bilateral, cyclical leg movements (e.g., walking and pedaling), the phase relation between the legs is abnormal and highly variable after stroke. Lower limb muscle activity has abnormal timing and amplitude that is exacerbated during bilateral movements, and ankle motor control is worse during bilateral than unilateral movement.

Impaired interlimb coordination may contribute to worsened walking-related outcomes. Impaired interlimb coordination after stroke increases metabolic cost of walking, compromises responses to perturbation, and contributes to detrimental behaviors like motor compensation. However, it is unclear how impaired interlimb coordination impacts walking. Some studies have shown that impaired interlimb coordination is related to reduced walking speed (or pedaling velocity), but others have found no association. Although the relation of walking with interlimb coordination is unclear, related constructs (spatiotemporal asymmetry and variability) are associated with walking. Spatiotemporal asymmetry is associated with motor impairment and slower walking, and spatiotemporal variability is related to fall risk and reduced community ambulation.

Interlimb coordination is a distinct construct that is understudied. One reason why few studies have investigated interlimb coordination after stroke is that much prior work has used walking symmetry as a measure of interlimb coordination. However, less than 50% of the variance in interlimb coordination is explained by measures of symmetry, symmetry does not reflect the phase-dependent relation between limbs, and asymmetrical movements may be appropriate and well-coordinated. These data suggest that symmetry and interlimb coordination are distinct features of gait. Also, symmetry may offer limited insight into walking; improvements in symmetry and in walking are not associated, and targeting symmetry does not benefit walking more than other approaches. In contrast, measures that fully reflect interlimb coordination patterns during walking (e.g., muscle activation, kinetic, or kinematic patterns) may better inform walking rehabilitation.

Interlimb coordination has multi-level neural control, but which neural pathways contribute to impaired interlimb coordination after stroke is unclear. During walking, interlimb coordination is controlled via supraspinal, spinal interneuronal, and sensory pathways. Supraspinal motor regions have bilateral effects via ipsilateral uncrossed (e.g., corticospinal, reticulospinal) and interhemispheric pathways. Spinal interneuronal pathways control bilateral locomotor rhythms, and interlimb sensory pathways provide essential input to supraspinal regions, spinal circuits, and motoneurons. Each pathway contributes to interlimb coordination, and there is a complex interplay between pathways. Supraspinal descending pathways modulate spinal interneuronal and sensory pathways; sensory pathways are modulated by spinal interneuronal circuits; and spinal interneuronal circuits depend on sensory inputs. Stroke causes imbalanced transcallosal pathways and altered strength of bilateral motor tracts and causes secondary changes in descending supraspinal modulation of spinal interneuronal and sensory pathways, leading to altered function of these pathways For example, crossed limb reflexes, H-reflexes, and cutaneous reflexes all have abnormal timing and amplitude during walking and pedaling after stroke, and sensory loss is common after stroke. These changes may contribute to impaired interlimb coordination and walking disability after stroke. Despite its essential role for walking, there has been minimal work that investigates interlimb coordination and seeks to identify the neural sources of impairment after stroke.

Targeting interlimb coordination may improve walking. Most walking rehabilitation approaches have focused on restoring neurotypical patterns in the more affected limb with little focus on interlimb interactions or improving walking function (e.g., maximizing walking speed) without regard for movement quality. These approaches may have achieved limited success because they ignore the essential role of interlimb coordination for walking. Yet, few interventions have targeted interlimb interactions, largely because interlimb coordination has been conflated with symmetry and can be more expensive to measure. Interventions to target interlimb coordination during walking include: 1) promoting an antiphasic interlimb pattern (e.g., using visual or audio cues); 2) stimulating coordinated bilateral responses (e.g., perturbing walking surface stiffness on a variable stiffness treadmill to evoke responses to maintain coordination and stability); and 3) augmenting interlimb phasing errors with a split-belt treadmill to elicit aftereffect improvements in coordination. Repeated interlimb coordination training may yield improvements in walking. Moreover, determining the neural sources of impaired interlimb coordination could reveal additional targets for intervention. For example, neuromodulatory stimulation of supraspinal, spinal interneuronal, and/or sensory pathways may serve as a powerful adjuvant to interlimb coordination training.

The aims of this proposal address several issues that must be resolved before interlimb coordination can be effectively targeted: 1) identifying the relative contributions of supraspinal, spinal interneuronal, and sensory pathways to interlimb coordination impairments after stroke, and 2) determining the association between interlimb coordination and walking.

Inclusion Criteria:

• Age: 25 - 90 years of age
• Monohemispheric stroke
• Chronic phase (> 6 months post stroke)
• Ability to walk for at least 6 minutes at a self-selected comfortable speed

Exclusion Criteria:

• Lesions affecting the brainstem or cerebellum
• Other neurological disorders
• Current botox treatments for the lower limb
• Significant cognitive or communication impairment

TMS exclusion criteria

• Previous adverse reaction to TMS
• Skull abnormalities or fractures
• Concussion within the prior 6 months
• Unexplained, recurring headaches
• Implanted cardiac pacemaker
• Metal implants in the head or face
• History of seizures or epilepsy
• Use of medications that could increase risk of seizure
• Current pregnancy

PNS & DCS exclusion criteria

• Skin hypersensitivity at any sites of stimulation, including the scalp, thoracolumbar spine, and peripheral limbs
• History of contact dermatitis at any of the sites of stimulation
• History of allodynia and/or hyperalgesia
• Active skin infection
• Skin lesions
• Deep vein thrombosis
• Any other skin or scalp condition that could be aggravated by stimulation
• Implanted electronic, metallic, or highly conductive devices near site of stimulation that cannot be removed without permission from a health professional

All underlying data for this study will be deidentified and then will be shared.

All shared data will be made available at latest by the time of associated publication or at the end of the performance period, whichever comes first. As permitted by the relevant repositories, data will be made available in perpetuity, but at minimum for 10 years.

All data described above will be shared on UIC INDIGO and UIC Research Data Glacier, university institutional and data repositories. All shared data will be accessible via persistent unique identifiers (Digital Object Identifier [DOI]). DOIs will be referenced in any related publications.