DEVELOPMENT AND VALIDATION OF B2AND2SV2ASC SCORE FOR PREDICTING RISK OF STROKE ONE YEAR AFTER TRANSCATHETER AORTIC VALVE REPLACEMENT

INTRODUCTION Stroke after transcatheter aortic valve replacement (TAVR) is a significant cause of morbidity and mortality. It increases the risk of mortality within one year of TAVR threefold. It is also associated with significant disability and increased health care costs. A validated risk prediction tool for stroke one year after TAVR may be useful in risk stratification to guide decisions at the point of clinical care and in research. OBJECTIVES 1. To develop and internally validate a new risk prediction score (B2AND2SV2ASc) for predicting stroke within one year of TAVR; 2. To assess the predictive performance of the R2CHA2DS2VASc, CHA2DS2VASc, R2CHADS2, and CHADS2 scores for predicting stroke within one year of TAVR. METHODS This was a retrospective cohort study using data from the University of Kansas Hospital TAVR program. Patients who underwent TAVR from 2012 to the first quarter of 2018 who had one-year follow-up data were included in the study. The association between candidate variables and stroke within one year of TAVR was estimated using the Firth multivariable logistic regression method. Integer points for variables that were included in the model were calculated based on the model β coefficients. Internal validation was performed by bootstrapping. The model discrimination and calibration were assessed using the area under the receiver operating characteristic curve (c-statistic) and the Hosmer-Lemeshow test. RESULTS Eight variables were included in the final model; bicuspid/unicuspid valve, diabetes mellitus and prior peripheral vascular disease were weighted with two points each, and aortic valve area of <0.5 cm2, ≥2 diseased coronary vessels, prior stroke/transient ischemic attack, body surface area ≤1.75 m2, and being female were weighted with one point each. The new score (B2AND2SV2ASc score) had a c-statistic of 0.709 (0.640 on internal validation) and a Hosmer-Lemeshow χ2 p-value of 0.146. The R2CHA2DS2VASc, CHA2DS2VASc, R2CHADS2, and CHADS2 scores had c-statistics of 0.591, 0.596, 0.607, and 0.622, respectively. The Hosmer-Lemeshow χ2 p-values were 0.762, 0.422, 0.463 and 0.146 respectively. CONCLUSION The B2AND2SV2ASc score had a good discriminative ability and was well calibrated for stroke prediction one year after TAVR.

Grantor dc:publisher

University of Kansas

Year dc:date.issued

2019

Author dc:creator

• NDUNDA, PAUL MUUTU

Advisor dc:contributor.advisor

• Kallail, K. James

• Medicine
• One Year
• Risk Prediction Model
• Risk Prediction Score
• Stroke
• Transcatheter Aortic Valve Replacement

Statement dc:rights

• Copyright held by the author.

Language dc:language.iso

en

Dc Identifier Other

http://dissertations.umi.com/ku:16783

OAI identifier oai:identifier

oai:kuscholarworks.ku.edu:1808/37236