Mobile digital gait analysis captures effects of botulinum toxin in hereditary spastic paraplegia

Alzhraa A. Ibrahim,Malte Ollenschläger,Stephan Klebe,Rebecca Schüle,Nicole Jeschonneck,Melanie Kellner,Evelyn Loris,Teresa Greinwalder,Bjoern M. Eskofier,Jürgen Winkler,Heiko Gaßner,Martin Regensburger
DOI: https://doi.org/10.1111/ene.16367
2024-06-12
European Journal of Neurology
Abstract:Background and purpose Hereditary spastic paraplegias (HSPs) comprise a group of inherited neurodegenerative disorders characterized by progressive spasticity and weakness. Botulinum toxin has been approved for lower limb spasticity following stroke and cerebral palsy, but its effects in HSPs remain underexplored. We aimed to characterize the effects of botulinum toxin on clinical, gait, and patient‐reported outcomes in HSP patients and explore the potential of mobile digital gait analysis to monitor treatment effects and predict treatment response. Methods We conducted a prospective, observational, multicenter study involving ambulatory HSP patients treated with botulinum toxin tailored to individual goals. Comparing data at baseline, after 1 month, and after 3 months, treatment response was assessed using clinical parameters, goal attainment scaling, and mobile digital gait analysis. Machine learning algorithms were used for predicting individual goal attainment based on baseline parameters. Results A total of 56 patients were enrolled. Despite the heterogeneity of treatment goals and targeted muscles, botulinum toxin led to a significant improvement in specific clinical parameters and an improvement in specific gait characteristics, peaking at the 1‐month and declining by the 3‐month follow‐up. Significant correlations were identified between gait parameters and clinical scores. With a mean balanced accuracy of 66%, machine learning algorithms identified important denominators to predict treatment response. Conclusions Our study provides evidence supporting the beneficial effects of botulinum toxin in HSP when applied according to individual treatment goals. The use of mobile digital gait analysis and machine learning represents a novel approach for monitoring treatment effects and predicting treatment response.
neurosciences,clinical neurology
What problem does this paper attempt to address?