Gait Enhancing Mobile Shoe for Stroke Rehabilitation

Hemiparesis and other impairments are a frequent and disabling consequence of stroke and can
lead to asymmetric and inefficient walking patterns. Training on a split-belt treadmill,
which has two separate treads driving each leg at a different speed, can correct gait
asymmetries post-stroke (Reisman et al. 2005, Reisman et al. 2007). However, the effects of
split-belt treadmill training only partially transfer to everyday walking over ground and
extended training sessions are required to achieve long-lasting effects (Reisman et al.
2009). Our previous studies suggest that the iStride device that has been developed in our
laboratory can be used as an alternative gait training device for people with stroke
(Handzic, Reed, and colleagues 2011, 2011, 2012, 2013). The iStride device mimics the actions
of the split-belt treadmill, but can be used during over-ground walking and in one's own
home, thus enabling long-term training. The iStride device does not require any external
power and is completely passive; all necessary forces are redirected from the natural forces
present during walking since it utilizes the wearer's weight to generate its movements. While
the movements of the iStride device are similar to the split belt treadmill, and the iStride
device generates a similar aftereffect, the efficacy of this shoe in modifying the gait of an
individual with stroke is not yet verified. This research aims to continue to test the
iStride device on individuals with stroke to determine if the related effects in a clinic
setting, are also beneficial and safe in a home environment.

In this study, we will test the efficacy of the iStride device on individuals with stroke in
the clinic and within their own home. Efficacy will be evaluated based on the change in gait
coordination and also based on subjects' self-reported comfort on the device. We predict that
the iStride device will result in changes to interlimb coordination of gait and prove to be a
safe and effective device to be used in ones' home.

The asymmetric nature of hemiparetic gait can have a large impact on functional walking
ability. For example, swing phase asymmetry is a significant predictor of hemiparetic walking
performance because it strongly correlates with stages of motor recovery, walking speed, and
falls (Brandstater et al. 1983; Titianova and Tarkka 1995). Another measure of temporal
asymmetry - double support duration - is similarly correlated with walking speed (Olney et
al. 1994). In addition, spatial (e.g. step length) asymmetry is associated with decreased
propulsive force on the paretic leg (Balasubramanian et al. 2007; Bowden et al. 2006), which
limits forward motion of the body and reduces gait efficiency (Brouwer et al. 2009; Kahn and
Hornby 2009). The importance of gait efficiency should not be understated - the elevated
energy demands of hemiparetic gait combined with physical deconditioning post-stroke can
greatly limit performance of activities of daily living, contributing to poor cardiovascular
fitness and metabolic syndrome. In turn, this can increase the risks of a second stroke or
cardiovascular event and is associated with increased morbidity and mortality rates (Ivey et
al., 2005). Therefore, improving gait symmetry should be an important goal for therapy, to
not only improve functional mobility and reduce injury, but also to enhance general health
and well-being post-stroke.

Practicing walking on a split-belt treadmill can correct abnormal interlimb coordination of
gait in individuals with hemiparesis following stroke or other central nervous system lesions
(Reisman et al. 2005, Reisman et al. 2007). Asymmetric gait can manifest as a spatial
asymmetry, in which steps taken on one side are longer than those on the other. It can also
manifest as a temporal asymmetry, where the timing is uneven on the paretic and non-paretic
sides. Temporal asymmetries are often measured as differences in the duration of double
support periods, which are the amount of time both feet are simultaneously contacting the
ground and are measured separately for the paretic and non-paretic sides. The iStride device
is designed to cause changes in both spatial and temporal gait symmetry. We predict that the
iStride device would cause the steps on the side with the iStride device to be larger since
individuals would compensate for the backward rolling motion by placing their foot farther
forward in stance, thus increasing the distance between the two feet. Similarly, since the
stride is longer, it may also shorten the duration of stance relative to the other side. With
shortened stance duration, the amount of time spent in double-support at the end of stance
would likely decrease as well.

Although the original idea of the iStride device is derived from the motion of the
split-belt-treadmill, there are distinct differences between walking on the device and
walking on a split-belt treadmill with asymmetric belt velocities (Handzic and Reed, 2013).
While the body's velocity relative to ground is zero on a split-belt treadmill, the relative
velocity of the iStride device is non-zero and forward. The device forces the wearer's foot
forward or backward whereas the treadmill moves both feet backward, but at different speeds.
For both the split-belt treadmill and the iStride device, the relative velocity between both
feet is similar and the backward-moving iStride device takes the place of the faster tread.

We hypothesize that training over ground will lead to a change in the interlimb coordination
in individuals with asymmetric gait and allow individuals to develop a more persistent
symmetric gait. There are several differences between training on ground and a treadmill,
such as visual flow and vestibular information signaling forward movement that likely limit
the expression of learning in the over-ground context when trained on a treadmill. Visual
cues appear to be particularly important for context awareness (Keamey, 2003). Visual cues,
coupled with prior experience, are so powerful that predictive postural responses cause an
individual to stumble when stepping onto an escalator that is not moving (Reynolds and
Bronstein 2003; Bronstein et al. 2009). The body has learned an internal model that expects
an acceleration when stepping onto an escalator, but when that acceleration does not occur,
the person stumbles. A study of split-belt walking showed that transfer to over-ground
walking is enhanced when subjects are blindfolded during training on the treadmill and tested
over ground (Torres-Oviedo and Bastian, 2010). Since blindfolding eliminates visual cues
about the environment, this also suggests that vision is a key factor in determining the
context-dependence of learning. Since it is not realistic to blindfold stroke patients during
gait training, we designed the iStride device so that training could occur during over-ground
walking, thus visual cues during training and later walking over ground would be the same.

Data from control subjects using an earlier version of the iStride device (referred to as
GEMS at the time) has been published (Handzic et al. 2011) and a video of this previous
version can be found at http://reedlab.eng.usf.edu/publications/handzic2011GEMS.mp4. In this
study, we found that this earlier version of the device was capable of changing step length
as predicted, however the previous design was too heavy and too tall to be considered
practical for testing in stroke populations. The current version of the iStride device
produces similar motion to the previous version, but it weighs less (under 1 kg) and is
shorter (~4.4 cm). We will test the efficacy of wearing the current iStride device on gait
coordination during walking over ground on individuals with stroke in the clinic and within
their own home. All walking will be performed while subjects are closely guarded by an
experienced physical therapist to prevent falling. The effects of the iStride device on gait
coordination will be compared to those induced on a split-belt treadmill.

The proposed project represents one of the first attempts to build a device that corrects
walking symmetry while walking over ground. Not only would this allow people to experience
gait corrections while performing normal movements, but the simplicity and relative low cost
of these devices would also open up potential opportunities to train at home (for
high-functioning individuals with supervision) and in clinics where a split-belt treadmill is
not available. The studies outlined here will establish whether the iStride device is capable
of changing interlimb coordination of gait, and whether individuals with stroke can use these
devices for rehabilitation purposes. This work will thus build the foundation for future
training studies examining the effectiveness of long-term use of the iStride device for
improving symmetric walking patterns.

The question that this study targets is the modification of human walking patterns for use in
stroke rehabilitation. It is our ultimate objective to show that the iStride device can
change a person's temporal and spatial gait asymmetry into a symmetric gait. We have
demonstrated this in a clinic setting, and now we will test the device within the
participants own home. Our points of reference are results obtained by previous studies with
split-belt treadmills. We are also interested in how altering the interface between a foot
and the ground influences the adaptation to new walking patterns.

Inclusion Criteria:

- age 21-80

- one or more cerebral strokes, but all strokes on same side

- a stroke at least 6 months prior to enrollment

- Gait asymmetry, but able to walk independently with or without a cane

- Not currently receiving physical therapy

- no evidence of severe cognitive impairment that would interfere with understanding the
instructions

- no evidence of one-sided neglect, affecting ambulation

- At least 25 feet of walking space in home (does not need to be a straight line)

- Weight does not exceed 250 lbs

Exclusion Criteria:

- uncontrolled seizures

- metal implants (stents, clips, pacemaker)

- pregnancy

- History of a neurological disorder other than stroke ( Parkinson's, MS)

- Chronic Obstructive Pulmonary Disease

- Uncontrolled blood pressure

- Head injury in the past 90 days

- A myocardial infarction within the last 180 days

- Cannot rely on a rolling walker for ambulation