Performance Of Neurophysiological, Neuroimaging, Biomechanical And Clinical Predictors And Prediction Models Of Upper Extremity Motor Recovery Following Stroke-A Systematic Review Protocol

Background: Stroke has been identified as a major cause of disability often due to the upper-extremity impairments, recovery from which, takes place over a period of time. Prediction models or predictors have been formulated for predicting upper-extremity recovery following stroke. However, we still lack knowledge about the hierarchy of available prediction models for recovery prognosis and hence a systematic review for the same is warranted. Objectives: To systematically review and assess the level of evidence underpinning the performance of prediction models or predictors (incorporating neurophysiological, neuroimaging, biomechanical and/or clinical measures) in predicting upper-extremity motor recovery post-stroke. Methods: Six online databases (PubMed, Web of Science, Scopus, OvidSP, Proquest and CINAHL) will be searched using database-specific terms. Two reviewers will independently screen relevant studies and extract data from eligible articles using a modified CHecklist for critical Appraisal and data extraction for systematic Reviews of prediction Modelling Studies (CHARMS). Risk of bias of included articles will be assessed using the Prediction model Risk Of Bias ASsessment Tool (PROBAST) and overall level of evidence will be determined using the Grading of Recommendations Assessment Development and Evaluation (GRADE). Discussion: Extant systematic reviews have analogized data from recovery prediction models based on therapist-rated and patient-reported outcome measures. Even so, the Stroke Recovery and Rehabilitation Roundtable Task Force have recommended quantifying movement quality using instrument-based outcome measures to determine recovery. Therefore, this protocol elucidates the steps of a systematic review assessing the evidence on prediction models or predictors comprising certain instrument-based outcomes that predict upper-extremity recovery post-stroke.