Value of Computed Tomography-Based Computational Fluid Dynamics in the Prediction of Acute Coronary Syndrome.

This study investigated the ability of coronary computed tomography angiography (CCTA)-derived computational fluid dynamics (CFD) parameters to identify lesions associated with subsequent acute coronary syndrome (ACS). The study included 37 patients with well-documented ACS and available CCTA performed at least 1 week before the event. Lesions identified on CCTA were classified as culprit (n=37) or non-culprit (n=42). Information on clinical characteristics and anatomical features was collected. CFD analysis was performed to compute wall shear stress (WSS) and axial plaque stress (APS) at both the segment (seg) and arc levels, with minimum (min) and maximum (max) values recorded. Univariate and multivariate logistic regression analyses were used to identify predictors of ACS. Significant stenosis was more frequent in culprit than non-culprit lesions (P=0.033). Compared with non-culprit lesions, culprit lesions had lower min(APSseg) values but higher max(|APS|seg) and max(WSSseg), although the differences were not statistically significant. In multivariate analysis, vessel location, min(APSseg) (odds ratio [OR] 3.17, P=0.047), and max(WSSseg) (OR 6.99, P=0.020) were independently associated with the occurrence of ACS events. Incorporating CFD parameters max(|APS|seg), min(APSseg) and max(WSSseg) into a model containing clinical and anatomical variables significantly improved ACS prediction (area under the curve 0.862 vs. 0.781; P=0.044). CCTA-derived CFD parameters are independently associated with the development of ACS. Integrating multiple CFD metrics enhances the predictive performance beyond traditional clinical and anatomical characteristics, supporting their potential role in risk stratification.