[III-JCKAP7-2-1] Risk Prediction Models for Significant Adverse Event in Pediatric Cardiac Catheterization Laboratory: CRISP Registry in the United States
キーワード：adverse event, risk prediction, cardiac catheterization laboratory
Objective: Pre-procedural risk prediction models for significant adverse event (SAE) in pediatric cardiac catheterization laboratory (PCCL) has been developed by the multi-centered CRISP registry in the United States. The risk models for children and adults are CRISP and CRISA scores, which has been widely utilized internationally. In this presentation, we present these risk prediction scores and our updated CRISP registry. Methods: The CRISP registry was launched in 2008 in the United States. CRISP and CRISA scores were published in 2016 and 2019. We present the risk score development methodology along with selection of final predictive variables and the performance/application of risk prediction models. Results: After identifying potential independent predictors for SAE in the univariable analysis, multivariable analysis reached the final risk prediction models in the datasets of CRISP registry. The performance of CRISP was later validated in the subsequent cohort in 2018. CRISP score has been referenced/utilized in many research projects related to PCCL. The online CRISP score calculator is available at the CCISC website (new.ccisc.net). Since 2023, the upgraded CRISP registry has been started with the new sets of data variables. Conclusion: Risk prediction models (CRISP and CRISA) have been applied to the clinical practice internationally and become a valuable tool for pre-procedural preparation, risk awareness and patient counseling. In the new CRISP registry, we aim to improve these risk prediction models by incorporating additional unique pre-procedural variables.