Publication: Accuracy of clinical risk scores in predicting post-rtPA intracerebral hemorrhage in a Thai cohort
Issued Date
2019-05-01
Resource Type
ISSN
01252208
Other identifier(s)
2-s2.0-85068802771
Rights
Mahidol University
Rights Holder(s)
SCOPUS
Bibliographic Citation
Journal of the Medical Association of Thailand. Vol.102, No.5 (2019), 540-546
Suggested Citation
A. Suengtaworn, G. Saposnik, C. P. Hurst, N. Poungvarin, Y. Nilanont Accuracy of clinical risk scores in predicting post-rtPA intracerebral hemorrhage in a Thai cohort. Journal of the Medical Association of Thailand. Vol.102, No.5 (2019), 540-546. Retrieved from: https://repository.li.mahidol.ac.th/handle/20.500.14594/51689
Research Projects
Organizational Units
Authors
Journal Issue
Thesis
Title
Accuracy of clinical risk scores in predicting post-rtPA intracerebral hemorrhage in a Thai cohort
Abstract
© JOURNAL OF THE MEDICAL ASSOCIATION OF THAILAND | 2019. Objective: Several prediction scores for post-rtPA symptomatic intracerebral hemorrhage (SICH) have been developed based on western datasets. The authors compared the accuracy of eight clinical risk scores to predict post-rtPA SICH in the Thai population. Materials and Methods: The authors applied eight risk scores to a retrospective cohort of acute ischemic stroke patients who received IV rtPA between 2005 and 2015 in a tertiary care center (Siriraj Hospital, Mahidol University, Thailand). The main outcomes were SICH defined according to the ECASSII and the NINDS definitions. All risk scores were then compared using ROC curve, sensitivity, specificity, NPV and PPV, and the LR+ and LR–. Results: Four hundred five patients were included. The rates of SICH-ECASSII and SICH-NINDS definition were 7.1% and 11.1%, respectively. Among the eight risk scores, the DRAGON score, the HAT score, and the GRASPS score were the three best scores for predicting by SICH-ECASSII definition. The DRAGON score achieved 66% sensitivity and 58% specificity (AUC 0.60, PPV 11%, NPV 96%, LR+ 1.56, LR– 0.59), the HAT score had 72% sensitivity, 50% specificity (AUC 0.65), while GRASPS score reached 79% sensitivity, but only 40% specificity (AUC 0.63). Conclusion: The present study demonstrates that the existing SICH risk scores did not perform well in the studied population.