The role of ECV-CT in left ventricular remodeling after TAVR

Introduction Left ventricular (LV) reverse remodeling (RR) can occur after transcatheter aortic valve replacement (TAVR), but it remains incompletely understood. Pre-TAVR CT-derived extracellular volume (ECV-CT), a marker of extracellular matrix expansion (including both focal replacement fibrosis and diffuse interstitial fibrosis), has been correlated with worse outcomes, but its association with LV RR is ill defined. We aimed to assess the association between pre-TAVR ECV-CT and LV RR at 1-year post-TAVR. Methods Single-center retrospective study including consecutive patients with severe symptomatic AS undergoing TAVR between 2022-2024. Only patients with echocardiographic (TTE) data both before and 1-year after TAVR were included. All patients underwent a pre-TAVR-planning CT on a 192-slice dual-source scanner, using an additional post-contrast low-radiation-dose prospective acquisition. ECV-CT was calculated according to the equation and example in Fig1. Patients were grouped by mean ECV-CT, and TTE measures were compared before and after TAVR within each group using paired analyses. Results A total of 81 patients were included (median age 82 (IQR 78-87) years; 47% male). Pre-TAVR TTE showed a median transaortic gradient of 47mmHg (IQR 41–56), a mean aortic valve area of 0.75±0.19 cm², a median indexed LV mass (LVMi) of 138 g/m² (IQR 106–161) and a median indexed LV end-diastolic volume (LVEDVi) of 51mL/m² (IQR 41–68). The median LV ejection fraction (LVEF) was 56% (IQR 51–63) and mean global longitudinal strain (GLS) was -15 ± 4%. No patient had a clinical diagnosis of cardiac amyloidosis. The mean ECV-CT was 33.9 ± 8.0%. Patients with ECV-CT ≥33.9% had numerically higher LVMi (143 g/m² [IQR 107–161] vs 130 g/m² [IQR 106–161], p=0.481). At 1-year post-TAVR, TTE demonstrated a significant reduction in LVMi to 121 g/m² (IQR 98–148; p=0.018). No significant changes were observed in other parameters [LVEDVi 51 mL/m² (IQR 38–66), LVEF 56% (IQR 52–61) and GLS -16 ± 4%); p>0.05 for all]. After stratification according to mean ECV-CT, patients with ECV-CT <33.9% (n=42) showed a non-statistically regression in LVMi (130 g/m² [IQR 106–161] vs 121 g/m² [IQR 101–149]; p=0.333), whereas those with ECV-CT ≥33.9% (n=33) demonstrated a significant reduction in LVMi (143 g/m² [IQR 107–161] vs 122 g/m² [IQR 91–142]; p=0.023). Both groups achieved similar LVMi at 1-year after TAVR (p=0.717) (Figure 1). Conclusion In this preliminary study, patients with elevated ECV-CT showed a post-TAVR reduction in LVMi that brought them to levels similar to those with normal ECV. This pattern suggests that elevated ECV-CT, despite associated with poorer prognosis in previous studies, is not necessarily a marker of irreversible adverse remodeling in aortic stenosis.