Abstract
Introduction and objectives
Left ventricular reverse remodeling (LVRR) after transcatheter aortic valve replacement (TAVR) is associated with outcomes, but the relevance of aortic-root geometry remains uncertain. We examined whether the annular eccentricity index (EI) and sinus of Valsalva height on computed tomography angiography (CTA) were associated with 1-year LVRR.
Methods
This retrospective secondary analysis included 115 patients with severe aortic stenosis from an 11-center cohort who underwent successful TAVR and survived to undergo evaluable preprocedural CTA and 1-year echocardiography. EI was calculated as the difference between maximum and minimum annular diameters divided by the maximum diameter. Sinus height was the mean perpendicular distance from the annular plane to the sinus tip across the three sinuses. LVRR was defined as an increase in left ventricular ejection fraction of at least 5 percentage points or a decrease in indexed left ventricular end-diastolic volume of at least 15%. A fixed multivariable logistic model adjusted for baseline ventricular function and clinical covariates; Firth bias-reduced logistic regression was used for primary inference because the smaller outcome group contained 37 patients, with conventional maximum-likelihood estimation as a sensitivity analysis.
Results
LVRR occurred in 78 patients (67.8%). In the primary Firth analysis, each 1-standard deviation (SD) decrease in EI was associated with higher odds of LVRR (adjusted odds ratio, 2.40; 95% confidence interval, 1.52–3.92;
P
P
= 0.003). Conventional maximum-likelihood estimates were directionally consistent. Model performance was lower after bootstrap optimism correction (area under the curve, 0.77 vs. 0.81 apparent).
Conclusions
Among patients who survived to undergo evaluable 1-year echocardiography, a lower EI and greater sinus height were associated with LVRR after TAVR. These hypothesis-generating associations require external validation before clinical application.