Abstract Body: Background: Stroke is a major cause of chronic disability and often leads to balance and gait disorders. Evidence shows impaired voluntary step initiation and reactive stepping response are associated with increased risks of falling. Stepping training is a standard component of stroke rehabilitation to improve gait and balance. However, the clinicians primarily focuses on voluntary step initiation with less emphasis on reactive step training, which involves responding to unexpected postural perturbations. The purpose of this study is to investigate differences in the reactive step control among healthy young adults (HYA), healthy older adults (HOA), and persons with chronic stroke (PCS). Hypotheses: (1) compared to HYA and HOA, PCS will show impaired reactive step control, as evidenced by longer step reaction time (RT) and larger step error (Err), and (2) there will be correlations between walking speed (WS), single limb stance (SLS), and double limb stance (DLS) time and step RT.
Methods: This study used a novel step initiation paradigm, employing a 3D camera system to provide real-time visual feedback of foot trajectory and create a virtual target to guide a forward step. To examine a person's ability to modify the ongoing foot movements via reactive control (i.e. reactive step control), we unexpectedly shifted the visual target's location after they had begun stepping. We quantified step control by step ST, peak step velocity (Pv), time to peak velocity (tPv), and step Err. We quantified walking ability by WS and step length (SL), and dynamic standing balance by SLS and DLS times.
Results: Twenty HYA, 15 HOA, and 17 PCS participated in this study. PCS had slower step RTs than HOA and HYA (p<=.0002). Reactive step control declined with age, as indicated by increased step RT in HOA compared to HYA (p=0.03). Stepping error was higher in PCS compared to HYA (p< = 0.02). The paretic (P) leg had smaller Pv and longer tPV compared to both HOA and HYA (p<0.0001). Step RT on the NP leg was correlated with WS (r= -0.84, p =0.005), P SLS (r= -0.88, p =0.002), and P DLS times (r= 0.76, p =0.02).
Conclusions: PCS demonstrated slower reactive step times on both P and NP legs, which correlated with slower WS and reduced SLS time on the P leg. Preliminary findings suggest that reactive stepping paradigm is a promising predictor of WS and standing balance in PCS. Improving reactive step control could enhance post-stroke gait function and decrease the risk of falls.