Unblinded Quasi-Randomized Pilot Study Exploring the Benefits of the ZeroG TRiP System to Improve Patient Balance Following an Acute Stroke
Unblinded Quasi-Randomized Pilot Study Exploring the Benefits of the ZeroG TRiP System to Improve Patient Balance Following an Acute Stroke
Strokes are neurological events that can lead to devastating physical and cognitive deficits, such as the inability to ambulate, impaired balance regulation, and loss of coordination. Due to the physical and cognitive deficits experienced following a stroke, many require admission to an inpatient rehabilitation facility to maximize their independence before returning to the home setting. The ability to walk, stand, climb stairs, and other mobility-related functional tasks, are critical components of achieving this functional independence. However, it is often difficult for post-stroke patients with balance impairments to safely practice balance and gait training without putting both therapists and patients at risk for injury. Incorporating robotic technologies to neurological rehabilitation can play a critical role in delivering safe and effective gait and balance therapy.
Body-weight support systems (BWSSs) unload paretic lower limbs, patients with gait impairments can practice a higher repetition of steps in a safe, controlled manner. As participants perform gait training, these systems support the participant's body-weight, permitting those with excessive weakness and poor coordination, to ambulate and perform more intensive therapy sessions sooner in their recovery, with minimal risk injurious fall. In addition to BWSSs, balance perturbation systems, which purposefully unbalance participants so to rehabilitate their postural control, have been used to improve gait and balance-control after stroke, or other age and disease related balance impairments.
The goal of this study was to evaluate the efficacy of a recently developed, not yet reported, balance perturbation module for the ZeroG BWSS. This new balance perturbation training module is directly integrated into the ZeroG BWSS and allows for the direct induction of safe lateral, anterior, and/or posterior perturbations via a Wi-Fi-enabled handheld device. During both stationary and ambulatory activities, this system was used unbalance participants in order to train their balance-control and balance-reactions. The purpose of this pilot study was to determine if this newly developed BWSS balance perturbation system more effectively rehabilitates participant gait and balance after stroke than the standard BWSS protocol without perturbations.
Body-Weight Support System Equipment and Interventions:
For this study, the BWSS used was the FDA listed ZeroG Gait and Balance System (Aretech, LLC, Ashburn, VA). ZeroG was first introduced to the investigators institution in September 2019. Unlike some BWSSs, this device is mounted on an overhead track that follows users during ambulation. Like other BWSSs, this system is designed to unload the users of up-to 200 pounds of their body weight while simultaneously protecting patients from falling. For this study, 10 pounds of participants' body-weight, the system minimum required to engage the BWSS, was continuously displaced. If a participant were to fall, the system would detect the change, decelerate, and stop the descent after a set distance; the fall distance was set between 8 to 12 inches for the purpose of this study. Unlike other BWSSs, a newly developed balance perturbation module known as the Training Responses in Postural rehabilitation or TRiP, is directly integrated to the ZeroG BWSS. This perturbation module is different than other systems as the balance perturbations are elicited directly through the BWSS and do not require a treadmill, tilt-table/shaking platform, or manual exertion by a. Further, they can be induced during normal gait and balance exercises during therapy. The BWSS control group interventions consisted of various balance activities, including: marching, side-stepping, retro-ambulation, step-taps, and step-ups. The BWSS control group also practiced various gait tasks, including: ambulation over the ground, going up and down stairs, and performing sit-to-stand transitions. The BWSS with perturbation (BWSS-P) intervention group performed the same activities as the control group, with the addition of lateral, anterior, and posterior perturbations. Assistive devices and equipment were used during intervention sessions as recommended by the participant's primary therapist, including: canes, rolling walkers, hemi-walkers, and ankle-foot-orthoses (AFO), ankle support braces, and upper extremity slings.
Investigators administered perturbations using a Wi-Fi-enabled handheld device linked to the BWSS and these consisted of a sudden and brief assistive or resistive force in the desired direction. Lateral perturbations were issued while participants were in a static stance, while anterior and posterior perturbations were issued during ambulation; eight perturbations, two in each direction, were completed each session. All participants started at perturbation level "one" and progressed up to a maximum perturbation level of "ten" through the course of the study. The amount of force exerted at each perturbation level is pre-set by the manufacturer. The perturbation level (i.e. intensity or force) used each session was based on the participant's progress and observational analysis made by the investigator of the participants' response to the perturbation level. If the participant was able to tolerate the initial perturbation level without exhibiting a balance reaction, the perturbation level was incrementally increased until an appropriate balance reaction was exhibited. If a participant was unable to recover and elicited a fall response in the system, the perturbation level was decreased by one level to ensure patient safety, and the exercise repeated to reinforce the exercise mechanics and participant confidence.
Participants in both study groups received a total of eight treatment sessions over two weeks. As necessary, participants received up to two sessions in one day to ensure they completed the required eight sessions before discharge. These sessions were incorporated into the participants' regular care. Treatment sessions were broken into 30 minute blocks as it standard at the host facility. This time includes participant transportation, equipment set-up, and in the case of this study, donning the BWSS harness. On average, participants received 20 minutes of active time in the BWSS for each 30 minute treatment block. All sessions were analyzed equally despite the length of time in the BWSS.
Missing data:
When outcome measurements are missing due to early participant withdrawal, if the participant completed 50% or less of the study sessions, that participant's data was removed. If the participant completed greater than 50%, then that participant's data was analyzed as normal.
Data Analysis:
Data was analyzed using GraphPad Prism version 9.0.0 (GraphPad Software, San Diego, CA). To compare the observed proportion of males and females in the BWSS groups, a Binomial Test and Fisher's exact test were used. The 95% confidence intervals reported the proportion of males and females in the BWSS-P group were calculated using the Wilson-Brown Method. Berg Balance Scale (BBS) and Activities specific Balance Confidence measurement changes between the pre- and post-intervention were compared directly, as well as between groups. The degree of change made by each individual was shown by calculating the percent change: [((Post assessment) - (Pre assessment)) / (Pre assessment)] ×100% .
BBS data of stroke rehabilitation patients from fiscal year 2018 served as a historical standard of care (SOC) baseline control. The SOC data was sorted to consist of patients with initial BBS scores of 21 or greater and who were admitted and discharged before the launch of the institution's BWSS in September 2018. This resulted in the inclusion of retrospective BBS data from 30 patients. Shapiro-Wilk testing was first used to test for normality; if one or more of the data-sets in the group failed (p<0.05), nonparametric tests were used going forward. For hypothesis testing between two groups, unpaired or paired two-tailed Student's t-test were conducted as appropriate. When indicated by an F-test for variance (p<0.05), Welch's correction was applied for unequal standard deviations between groups. When comparing three or more groups, if one or more groups were abnormally distributed, non-parametric Kruskal-Wallis analysis of variation (ANOVA) test and Dunn's multiple comparison test for statistical hypothesis testing were used. When normally distributed, an Ordinary one-way ANOVA with a Tukey's multiple comparisons test for statistical hypothesis testing was used. If Brown-Forsythe's test for variance indicated the variance of the groups were significantly different (p<0.05), a Brown-Forsythe correction was applied and Dunnett's T3 multiple comparisons test for statistical hypothesis testing was used instead.
Inclusion Criteria:
Exclusion Criteria: