Assessment of Potential Predictors of Aortic Stenosis Severity Using ECG-Gated Multidetector CT in Patients with Bicuspid and Tricuspid Aortic Valves Prior to TAVI

Background/Objectives: The aim of this study was to evaluate the usefulness of selected predictive parameters obtainable from cardiac multidetector computed tomography for assessing the severity of aortic valve stenosis in patients scheduled for transcatheter aortic valve implantation (TAVI). Methods: A detailed retrospective analysis was performed on 105 patients with a bicuspid aortic valve (BAV), selected from a cohort of 1000 patients with BAV confirmed on ECG-gated CT, and on 105 patients with a tricuspid aortic valve (TAV) matched for sex and age. All patients included in both groups had significant aortic stenosis confirmed on transthoracic echocardiography. Results: Across the entire cohort, a trend toward higher aortic valve calcium scores was observed in patients with bicuspid compared to tricuspid aortic valves (4194.8 ± 2748.7 vs. 3335.0 ± 1618.8), although this difference did not reach statistical significance (p = 0.080). However, sex-stratified analysis showed higher calcium scores in males with BAV than with TAV (5596.8 ± 2936.6 vs. 4061.4 ± 1659.8, p = 0.002), with no significant difference observed among females (p > 0.05). Univariate regression analysis showed that the aortic valve calcium score was the strongest statistically significant predictor of aortic stenosis severity in both groups, with R² = 0.224 for BAV and R² = 0.479 for TAV. In the multiple regression model without interaction terms, the explanatory power increased to R² = 0.280 for BAV and R² = 0.495 for TAV. Conclusions: In patients scheduled for TAVI, linear regression models assess the severity of aortic stenosis more accurately than any individual predictive parameter obtainable from ECG-CT, with the aortic valve Agatston score emerging as the most reliable single CT-derived predictor of stenosis severity in both TAV and BAV subgroups.