Impaired gait stability is a common feature of Multiple Sclerosis and contributes to fall risk and disease burden. Step-by-step foot placement control—adjusting foot placement based on the body’s kinematic state during swing—is a key mechanism for maintaining lateral and fore–aft stability during walking. The present study investigated how this stability-related control is realized in people with early clinically evident Multiple Sclerosis compared with neurologically healthy controls.
The investigators recorded kinematic data during steady-state treadmill walking and analyzed step-by-step relationships between the center-of-mass (CoM) kinematic state and subsequent foot placement using a linear feedback model. The approach focused on how variations in CoM position and velocity during the preceding swing phase predicted the next foot placement in both the mediolateral and anteroposterior directions.
Ten people with early-stage Multiple Sclerosis and 21 neurologically healthy control participants were included. All participants walked on a treadmill for one minute of steady-state walking at comparable average gait speeds. Group sample sizes, matching on average speed, and the 1-minute steady-state treadmill protocol are reported in the source.
Foot placement and CoM kinematic state data were analyzed using a linear feedback model that correlates each foot placement to the CoM position and velocity measured during the preceding swing phase. This model quantifies both the precision of foot placement control and the strength of the foot placement responses attributable to individual components of the CoM kinematic state (for example, mediolateral CoM velocity).
Both groups—people with early Multiple Sclerosis and controls—demonstrated step-by-step foot placement control in the mediolateral and anteroposterior directions as characterized by the feedback model.
There were no group differences in foot placement precision; precision of foot placement relative to the predicted control model was comparable between participants with Multiple Sclerosis and controls.
Despite similar precision, foot placement responses to variations in CoM velocity were stronger in the Multiple Sclerosis group than in controls. In other words, changes in CoM velocity during swing had larger influence on the subsequent foot placement in people with early Multiple Sclerosis.
Specifically, the contribution of mediolateral center-of-mass velocity feedback to the overall foot placement control mechanism was higher in the Multiple Sclerosis group compared to neurologically healthy controls.
The results indicate that step-by-step foot placement control is largely retained in early clinically evident stages of Multiple Sclerosis, as evidenced by preserved precision and the presence of predictive relationships between CoM state and foot placement. However, the control strategy appears to be implemented with different weighting of sensory feedback: people with early Multiple Sclerosis rely more heavily on mediolateral CoM velocity information to govern foot placement.
This altered weighting may reflect compensatory adaptations in sensorimotor integration or changes in how sensory cues are used to stabilize gait early in the disease. Clinically, the finding suggests that early Multiple Sclerosis does not necessarily abolish core stability mechanisms, but that interventions aimed at gait and balance could consider targeting sensory weighting and lateral stability control. The increased sensitivity to mediolateral CoM velocity could be relevant when designing balance training, gait rehabilitation, or fall-prevention strategies.
The ethical board of the medical faculty of Kiel University approved the study (project number D438/18). All participants provided written informed consent. The study is registered as Clinical Trial DRKS00022998. The authors declared no competing interests. The dataset used in the study is available upon reasonable request and after signing a data sharing agreement; a DOI was provided for data access in the source.
The source article reports the principal methods, sample sizes, modeling approach, and main outcomes described above. Specific details that were not reported in the source text provided here include full participant demographic breakdowns (for example, age, sex distribution), clinical characteristics of the Multiple Sclerosis group (such as disease duration, subtype, or disability scores), detailed model parameter estimates and statistical values, and effect size measures. These finer-grained data and statistical outputs are not included in the provided source text and would need to be consulted directly in the full preprint for comprehensive appraisal.