Impact of transcatheter aortic valve replacement on aortic root remodeling: a pre- and post-procedure CTA comparative analysis

Figures

• 

  Fig. 1. Anatomical Measurement Protocol and Representative CTA Images. A Pre-procedural CTA imaging protocol demonstrating the standardized measurement planes. B Post-procedural CTA at 1-year follow-up showing the implanted transcatheter aortic valve prosthesis within the native aortic root. C Annulus angle measurement methodology, illustrating the angle between the aortic annular plane and the left ventricular outflow tract (LVOT) central axis. The typical pre-TAVR angle of approximately 50° becomes more acute post-TAVR (approximately 46°)

• 

  Fig. 2. Aortic Root Dimensional Changes at 1-Year Follow-up. A Mean dimensional changes showing statistically significant expansion of the sinus of Valsalva (SoV, +2.6 ± 2.4 mm, +6.8%), sinotubular junction (STJ, +1.6 ± 1.8 mm, +5.0%), and ascending aorta (AAo, +0.6 ± 1.4 mm, +1.7%). Error bars represent standard deviation. B Individual patient trajectories demonstrating heterogeneity in remodeling response. Color coding indicates magnitude of change: gray (< 2 mm), light orange (2-3 mm), and dark orange (> 3 mm). The thick black line represents the cohort mean. C Distribution of patients relative to the predefined dimensional threshold (3 mm for SoV, indicated by dashed line). Only 10.3% of patients (n = 12) exceeded this threshold, with the majority showing changes within the expected measurement variability

• 

  Fig. 3. Annulus Angle Changes and Clinical Significance. A Distribution of annulus angle changes at 1 year. The histogram shows the observed changes (gray bars) overlaid with a normal distribution curve (blue solid line) and theoretical normal distribution (gray dashed line). The mean change was -3.1 ± 3.8°. Red dashed line indicates the threshold for clinically significant change (> 5° decrease), which occurred in 24.8% of patients (n = 29). Blue dashed line shows extreme increases (> 5°), observed in 4.3% of patients (n = 5). B Proportion of patients exceeding predefined dimensional thresholds across different parameters. Error bars represent 95% confidence intervals. Angular changes were more prevalent than dimensional changes, with 32.5% of patients meeting at least one threshold-exceeding criterion

• 

  Fig. 4. Correlation Structure and Multivariate Predictors of Hemodynamic Outcomes. A Correlation matrix of remodeling parameters and mean pressure gradient. Heat map colors indicate correlation strength (dark red: positive correlation; dark blue: negative correlation). Weak interdependence among dimensional parameters (r = -0.03 to 0.18) justified their inclusion as independent variables in multivariate modeling. B Forest plot showing standardized β coefficients from multivariate linear regression for 1-year mean pressure gradient. Significant predictors (red bars, p < 0.05) were annulus angle change (β = 0.24, p = 0.012) and baseline calcium volume (β = 0.19, p = 0.042). Dimensional expansion parameters (SoV, STJ) showed no significant association. Dashed vertical line represents the null hypothesis (β = 0)

• 

  Fig. 5. Association Between Anatomical Parameters and Hemodynamic Performance. A Scatter plot showing the relationship between annulus angle change and 1-year mean pressure gradient (β = 0.24, p = 0.012). Greater angular change was associated with higher gradients. Gray shaded area represents 95% confidence interval. Horizontal dashed line indicates the adverse outcome threshold (MPG ≥ 20 mmHg). B Relationship between baseline annular calcium volume and mean pressure gradient (β = 0.19, p = 0.042; slope = 0.029 mmHg/mm³). Data points are color-coded by risk category: green (low risk, 0 points), yellow (intermediate risk, 1 point), and red (high risk, 2 points). Vertical dashed line indicates the high-risk calcium threshold (400 mm³). C Box plots comparing mean pressure gradient across risk groups defined by combined anatomical features. The high-risk group (n = 19) demonstrated significantly higher gradients compared to low-risk patients (p = 0.008). Boxes represent interquartile ranges; horizontal lines indicate medians. D Scatter plot demonstrating the absence of association between SoV diameter change and mean pressure gradient (β = 0.08, p = 0.384, R² < 0.01). Vertical dashed line…

• 

  Fig. 6. Exploratory Risk Associations for Adverse Hemodynamic Outcomes. A Adverse outcome rates by risk category. Patients with both high-risk features (annulus angle change > 5° and baseline calcium > 400 mm³) had significantly elevated adverse hemodynamic outcome rates (26.3% vs 4.3% in low-risk group, p = 0.008). Error bars represent 95% confidence intervals. Numbers in parentheses indicate patients with adverse outcomes out of total in each category. B Receiver operating characteristic (ROC) curve demonstrating model discrimination. The C-statistic was 0.72 (95% CI: 0.58-0.86), with bootstrap-corrected C-statistic of 0.69 (95% CI: 0.54-0.84). The red dot indicates the optimal threshold for identifying high-risk patients (2 points). C Calibration plot comparing predicted versus observed probabilities of adverse outcomes. Points represent the three risk categories with error bars showing 95% confidence intervals. The black diagonal line represents perfect calibration. Hosmer-Lemeshow test indicated adequate calibration (p = 0.64)