作者: Erin Burke Quinlan , Lucy Dodakian , Jill See , Alison McKenzie , Vu Le
DOI: 10.1002/ANA.24309
关键词:
摘要: Objective This study was undertaken to better understand the high variability in response seen when treating human subjects with restorative therapies poststroke. Preclinical studies suggest that neural function, injury, and clinical status each influence treatment gains; therefore, current hypothesized a multivariate approach incorporating these 3 measures would have greatest predictive value. Methods Patients 6 months poststroke underwent battery of assessments before receiving weeks standardized upper extremity robotic therapy. Candidate predictors included brain injury (including gray white matter), function (cortical cortical connectivity), (demographics/medical history, cognitive/mood, impairment). Results Among all 29 patients, gains identified (smaller corticospinal tract [CST] injury), (greater ipsilesional motor cortex [M1] activation), connectivity interhemispheric M1–M1 connectivity). Multivariate modeling found best prediction achieved using both CST (r2 = 0.44, p = 0.002), result confirmed Lasso regression. A threshold defined whereby no subject >63% clinically significant gains. Results differed according stroke subtype; patients lacunar were predicted by measure intrahemispheric connectivity. Interpretation Response therapy after is model includes function. Neuroimaging may an ascendant role decision making for rehabilitation, which remains largely reliant on behavioral assessments. across subtypes, suggesting utility lesion-specific strategies. ANN NEUROL 2015;77:132–145