Prediction for hemorrhagic transformation risk after intravenous thrombolysis in acute ischemic stroke patients in different therapeutic windows: comparison of 5 scoring systems / 第三军医大学学报

ABSTRACT

Objective To compare the predictive value of 5 scoring systems for hemorrhagic transformation risk after intravenous thrombolysis in patients with acute ischemic stroke (AIS) in different therapeutic windows.Methods A single-center and retrospective study was performed for 243 AIS patients who underwent intravenous thrombolysis using recombinant tissue plasminogen activator (rt-PA) in different therapeutic windows in our department during January 2014 and December 2016.Five scoring systems,including HAT model (hemorrhage after thrombolysis),MSS model (multicenter stoker survey),GRASPS model (glucose at presentation,race,age,sex,systolic blood pressure at presentation,severity of stroke at presentation),SEDAN model (baseline blood sugar,early infarct signs,hyperdense cerebral artery sign on admission CT,age,NIHSS on admission),and SITS model (safe implementation of thrombolysis in strokemonitoring study) were used to evaluate the risks for hemorrhagic transformation.The relationships between the 5 scoring systems and incidence rate of hemorrhagic transformation were analyzed among the patients in different therapeutic windows.The predictive values of the 5 scoring systems were compared using the areas (AUC) under the receiver operating characteristic (ROC) curve.Results When the AIS patients were treated by intravenous thrombolysis within 3 h,the AUC of GRASPS and HAT models were 0.698 and 0.619,respectively,higher than those of the other 3 systems.When the therapeutic window was between 3 to 4.5 h,HAT model and SEDAN model had highest AUC (0.719,0.744) than the other 3 systems (P ＜0.05).When the windows were ＞4.5 ～6 h,the HAT model had the highest AUC (0.676).Conclusion The 5 scoring systems show better predictive value for hemorrhagic transformation after intravenous thrombolysis.For the therapeutic window within 4.5 h,HAT model presents best predictive value than the other 4 scoring systems.