Pre-intervention decompensation in severe aortic stenosis: the role of CT-derived extracellular volume

Introduction

Diffuse myocardial fibrosis represents an advanced stage of myocardial involvement in severe aortic stenosis (AS) and contributes to impaired myocardial reserve and clinical instability. Extracellular volume (ECV) quantification using dual-energy computed tomography (CT) can be performed during standard pre-transcatheter aortic valve implantation (TAVI) assessment, providing a non-invasive measure of myocardial fibrosis. The clinical implications of CT-derived ECV before valve intervention remain incompletely defined. This study aimed to evaluate the association between ECV and clinical decompensation during the pre-intervention period in patients with severe AS.

Methods

Retrospective cohort of patients with severe AS undergoing pre-TAVI CT (April 2024–April 2025). A delayed dual-energy acquisition was performed for ECV quantification. Global ECV was defined as the mean of basal and mid-ventricular short-axis values. Patients were included irrespective of subsequent management strategy. Outcomes of interest were pre-procedural unplanned cardiovascular hospitalization and urgent aortic valve replacement.

Results

253 patients were included (median age 81 years; 48.6% male). Median follow-up was 11 months [9-15]. 146 patients underwent valve implantation (136 TAVI, 10 SAVR), and 107 were managed conservatively. Median ECV was 30.4% [27.55 – 33.70%]. ECV did not differ between management strategies (p = 0.118).

During follow-up, 29 patients (11.5%) experienced unplanned cardiovascular hospitalization, and 28 of the 146 treated patients (19.2%) required urgent valve replacement. ECV was higher in patients with cardiovascular hospitalization compared with those without hospitalization (34.10% vs. 29.85%, p = 0.002). Similarly, patients requiring urgent valve intervention had higher ECV values than those treated electively (33.93% vs. 29.33%, p < 0.001).

ECV predicted unplanned cardiovascular hospitalization (OR 1.12, 95% CI 1.04–1.22; p = 0.005) and urgent aortic valve replacement (OR 1.24, 95% CI 1.11–1.38; p < 0.001). ECV increased progressively with worsening NYHA class, with median values of 26.55%, 28.75%, 32.25%, and 34.90% across NYHA classes I to IV, respectively (p < 0.001).

Conclusions

CT-derived ECV is associated with an increased risk of clinical decompensation in patients with severe aortic stenosis. Higher ECV identifies patients at greater risk of unplanned cardiovascular hospitalization and urgent valve replacement, and is associated with more advanced NYHA class. These findings support the potential role of ECV in routine pre-TAVI assessment to refine risk stratification and optimize timing of intervention