Abstract TP373: Classification of sICH in Stroke Center Hospitals: Reliability in Question

Abstract

Background: Measuring sICH is an important accountability of Stroke Centers. Since the NINDS rt-PA Study, the sICH definition has changed as knowledge of reperfusion-associated hemorrhagic transformation has grown. We aimed to determine what sICH definition was used by Stroke Centers and how this impacts sICH rates. Methods: Stroke Centers were invited to participate in a survey with the option to complete it via SurveyMonkey TM or by mail. Instructions to adhere to the sICH definition adopted in policies/procedures were provided, and to ask for clarification from Stroke Team members if needed. Data were assembled in SPSS, and analyzed using descriptive statistics and Student t-tests. Results: 229 responses were received representing 84% of U.S. states and the District of Columbia; 31% represented academic medical centers and 69% community hospitals. 64% of respondees were responsible for collecting the stroke quality data that supports certification. Overall tPA treatment rate for the sample was 8.7% + 6.4 (median 7%), with an overall reported sICH rate of 9.5% + 16.4 (median 5%). Official definitions supported sICH for 86% of responding hospitals, however the most common definition (48%) reported was, “any hemorrhage on non-contrast CT or MRI in combination with any clinical deterioration.” Only 17% identified the definition for sICH adopted by TJC for Comprehensive Stroke Center reporting. Among those that adhered to the TJC definition, sICH rates were significantly lower at 3%+2.3 (median 3%; t=4.7; mean difference = 7.7%; p<.0001, 95% CI 4.4-10.95), compared to 10.6%+17.5 (median 6%). Conclusions: Our study documents a significant need for education and inter-rater reliability monitoring of the use of sICH classification after intravenous tPA to ensure accuracy in local quality improvement processes, as well as the validity of data submitted to national stroke registries. Additionally, because sICH associated with reperfusion therapy is a new measure undergoing testing by TJC that could ultimately be tied to future pay-for-performance and public reporting, consensus on its definition as well as reliable sICH classification will be essential to future Stroke Center evaluation.