Symptomatic intracerebral hemorrhage after intravenous thrombolysis in Chinese patients: comparison of prediction models.

Background: To assess the performance of risk scores in predicting symptomatic intracranial hemorrhage (SICH) after intravenous thrombolysis (IVT).

Methods: A multicenter prospective study was performed in 811 patients who underwent IVT with standard-dose recombinant tissue plasminogen activator within 4.5 hours of acute ischemic stroke (AIS) onset in 67 stroke centers involved in the Thrombolysis Implementation and Monitor of acute ischemic Stroke in China program from May 2007 to April 2012. SEDAN (blood sugar, early infarct signs, [hyper]dense cerebral artery sign, age) score, Safe Implementation of Thrombolysis in Stroke (SITS)-SICH score, Glucose Race Age Sex Pressure Stroke Severity (GRASPS) score, Multicenter Stroke Survey (MSS) score, and Stroke Prognostication using Age and National Institutes of Health Stroke Scale (SPAN)-100 index were calculated in selected patients, and their predictive performance for SICH was compared according to the National Institute of Neurological Disorders and Stroke (NINDS), Safe Implementation of Thrombolysis in Stroke-Monitoring Study (SITS-MOST), and European Cooperative Acute Stroke Study (ECASS)-II criteria.

Results: For predicting the risk of SICH (NINDS definition) after IVT, the area under the receiver operating characteristic (ROC) curve of MSS score was the highest (.71, P < .0001). For predicting the risk of SICH (SITS-MOST definition) after IVT, the area under the ROC curve of GRASPS score was the highest (.73, P = .005). For predicting SICH (ECASS-II definition) after IVT, the area under the ROC curve of MSS score was the highest (.73, P < .0001).

Conclusions: SITS-SICH, GRASPS, and MSS scores predicted the risk of SICH after IVT in patients with AIS, but only the latter 2 were better in the Chinese population. MSS score had the best predictive performance for SICH using NINDS and ECASS-II definitions, whereas GRASPS score was the best for SICH using the SITS-MOST definition.