Exploring approaches to implement high intensity gait training for people with stroke during inpatient rehabilitation

Research Objectives High intensity gait training (HIGT) is established as best practice for patients following stroke. The optimal mode to implement HIGT is unknown. This study compares the use of usual care approaches and overground robotic exoskeletons (ORE) to achieve HIGT in people with stroke during inpatient rehabilitation. Design Prospective Observational Study. Setting Large urban inpatient rehabilitation facility (IRF). Participants Patients admitted to IRF within 6 months post-stroke with an appropriate walking goal. Interventions Patients received physical therapy (45-90 minutes) 5 days per week focused on HIGT including: conventional overground gait training, body weight supported treadmill training (BWSTT), LiteGait, and ORE. Heart rate (HR) was measured using the Polar A370 Fitness Watch. ORE was provided using Ekso Bionics exoskeleton. Main Outcome Measures Intensity achieved during HIGT sessions was assessed by HR, time spent walking per session, and number of steps. Results Patients (n=46; age 64 ± 13 years; 52% female) were admitted for inpatient rehabilitation at 16.55 ± 22.22 days since onset of stroke for an average length of stay of 22.72 ± 12.19 days. Conventional overground gait training (51%) was the most common mode of all HIGT sessions, followed by BSWTT (28%), and ORE (11%). ORE sessions (63%) attained the greatest amount of time in moderate-to-vigorous intensity, followed by BSWTT (46%), Litegait (36%), and conventional overground (30%). Time spent walking across sessions was greatest during ORE (24.6±-7.1 minutes), followed by conventional overground gait training (19.3±10.5), BSWTT (20.9±6.4), and Litegait (19.6±6.0). The average number of steps during ORE sessions was 505.59±208.10. Conclusions No mode of gait training attained HIGT cardiovascular intensity recommendations for patients with stroke. However, ORE elicited the greatest cardiovascular intensity and amount of time spent walking during IRF physical therapy sessions. Author(s) Disclosures The authors do not have any conflicts of interest.