Background: Symptomatic intracranial hemorrhage (sICH) is common after intravenous thrombolysis in acute ischemic strokes (AISs). Available predictive scoring systems were derived mostly in the Western countries.
Methods: Retrospective data in 1 provincial and 4 regional hospitals in the northern part of Thailand were reviewed. Patients with AIS, to whom recombinant tissue plasminogen activator (rt-PA) had been prescribed, were classified into 3 groups: no intracranial hemorrhage (no ICH), asymptomatic intracranial hemorrhage (asICH) and sICH. Coefficients under the multilevel ordinal logistic model were transformed into item scores and sum scores. Measures of discrimination, calibration, and internal validation were analyzed.
Results: Among 1172 patients, there were 78.8% with no ICH (n = 923), 13.1% with asICH (n = 154), and 8.1% with sICH (n = 95). The final model was named "SICH score" and included 6 variables: valvular heart diseases, use of aspirin, systolic blood pressure prior to thrombolysis that is 140 mmHg or higher, National Institutes of Health Stroke Scale scores higher than 10 and 20, a platelet count lower than 250,000 cell/mm3, and use of intravenous antihypertensive drugs during thrombolysis, with an Area under Receiver Operating Characteristic of .75 (95% confidence interval, .71-.80).
Conclusion: The SICH score could be an assisting tool to predict an individual risk of sICH after intravenous thrombolysis for AIS in Thai patients.
Keywords: Stroke; clinical prediction rule; intracranial hemorrhage; prognosis; thrombolytic therapy; tissue plasminogen activator.
Copyright © 2017 National Stroke Association. Published by Elsevier Inc. All rights reserved.