CT-derived extracellular volume provides added prognostic value over global longitudinal strain in severe aortic stenosis
A Lobo, M C Almeida, C Castro, J Cavadas, A Goncalves, I Rodrigues, P Braga, N Ferreira, F Sampaio, R Fontes-CarvalhoAbstract
Background
In aortic stenosis (AS), myocardial extracellular volume (ECV) can be non-invasively quantified by CT and reflects adverse myocardial remodeling and prognosis. Echocardiographic global longitudinal strain (GLS) is a sensitive marker of left ventricular systolic dysfunction and adverse outcomes in AS. However, the relationship between CT-derived ECV and GLS, as well as their relative prognostic value, remains incompletely defined.
Methods
Retrospective cohort analysis of patients with severe AS undergoing cardiac CT for TAVI planning between April 2024 and April 2025. All examinations included a standardized delayed-phase dual-energy acquisition enabling myocardial ECV quantification. Echocardiographic GLS was assessed using two-dimensional speckle-tracking echocardiography from standard apical views using vendor-specific software on General Electric and Philips systems. Associations between ECV and GLS were evaluated, and prognostic value for all-cause mortality was assessed using Kaplan–Meier analysis and Cox regression.
Results
A total of 253 patients with available CT-derived ECV data were included; GLS data were available for 149 patients. In this subgroup, median ECV was 30.25% [IQR 5.25] and median GLS was −13.10% [IQR 4.80]. Higher ECV was associated with worse GLS, reflected by less negative GLS values (Spearman ρ = 0.217, p = 0.008).
In univariable Cox regression analysis, ECV was significantly associated with mortality (HR 1.238, 95% CI 1.049–1.461; p = 0.012), whereas GLS showed a non-significant trend (HR 1.186, 95% CI 0.979–1.438; p = 0.081). In a multivariable model including both parameters, ECV remained independently associated with mortality (HR 1.198, 95% CI 1.012–1.418; p = 0.036), while GLS was not significantly associated (HR 1.119, 95% CI 0.921–1.359; p = 0.258).
Survival differed according to ECV stratified by the median (17.34 vs. 19.82 months, p = 0.013). Reduced GLS (≥ −16%) showed a non-significant trend towards lower survival (p = 0.071).
Conclusion
CT-derived extracellular volume independently predicts mortality in severe aortic stenosis, whereas GLS does not retain independent prognostic significance after adjustment for ECV. These findings support CT-derived ECV as a robust marker for risk stratification beyond echocardiographic strain assessment.