Resting motor threshold as a predictor of rehabilitation outcomes in multiple sclerosis

To investigate resting motor threshold (RMT) as a predictor of clinically meaningful gait improvement after multidisciplinary rehabilitation (MDR) + rTMS in people with Multiple Sclerosis (pwMS).

We retrospectively analysed 122 pwMS undergoing inpatient MDR combined with 12 sessions of excitatory rTMS. Gait was assessed pre- and post-treatment using the Timed 25-Foot Walk Test (T25FW) and the 6-Minute Walk Test (6MWT). Group- and individual-level longitudinal RMT changes were modelled using linear mixed-effects models. RMT slope was defined as the subject-specific slope from a linear mixed model with session number as continuous time, and random slopes and intercepts for each participant. Baseline RMT and the slope across 6 (50 % of total), 9 (75 % of total) and 12 (100 % of total) rTMS sessions were then used to predict improvement, adjusting for demographic and clinical covariates.

Clinically meaningful improvement occurred in 24.6 % (T25FW) and 48.4 % (6MWT) of pwMS. RMT slope significantly predicted improvement at the T25FW, even after accounting for baseline motor performance, baseline RMT, and clinical and sociodemographic covariates. No association was found between RMT and 6MWT improvement.

Longitudinal RMT trajectory, rather than baseline excitability, predicts gait speed improvement after MDR + rTMS in pwMS. This may reflect residual corticospinal plasticity and excitability potential, which may contribute to determining the efficacy of MDR in pwMS for gait speed improvement.

RMT slope may be a feasible rule-out biomarker to identify likely non-responders and support precision rehabilitation strategies.