Ona Wu1,2, Sren Christensen2, Niels Hjort2, Anders Neumann2, A Gregory Sorensen1, Gotz Thomalla3, Jens Fiehler4, Thomas Kucinski4, Joachim Rother3, Leif stergaard2
1Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Charlestown, MA, USA; 2Center for Functionally Integrative Neuroscience, Aarhus University Hospital, Aarhus, Denmark; 3Neurology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany; 4Neuroradiology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
Identification of patients most likely to benefit from reperfusion therapy is a key question in the management of acute stroke patients. Algorithms combining DWI/PWI for predicting infarction in stroke patients (< 6h) were trained on patients receiving standard non-thrombolytic treatment and then applied to patients who received thrombolysis. Predicted lesion volumes were significantly higher in patients without early reperfusion compared to patients who did reperfuse. Our results show that for patients given thrombolytic therapy, the likelihood of reperfusion depends on the degree and extent of the ischemic injury at the time of treatment, which we measured using multiparametric predictive algorithms.