Abstract
ABSTRACTThe absence of effective pharmacological interventions in acute traumatic spinal cord injury is a major problem in its management. A critical barrier in identifying such interventions lies in the vast heterogeneity of recovery profiles, which masks the potential efficacy of treatments in clinical trials. To determine the impact of temporal recovery profiles on long-term functional independence, we used EMSCI (European Multicenter Study about Spinal Cord Injury) data. Total motor scores from the International Standards for the Neurological Classification of Spinal Cord Injury (ISNCSCI) and the Spinal Cord Independence Measure (SCIM) were used to assess neurological and functional outcomes, respectively. We developed a classification method consisting of thresholding and unsupervised machine learning clustering and applied it to the total motor score profiles. Comparing SCIM scores between classes revealed that functional independence is significantly higher among patients displaying advanced neurological recovery profile. Our study suggests that the evaluation of temporal recovery profiles can provide novel insights in spinal cord injury clinical trials.
Publisher
Cold Spring Harbor Laboratory