External validation of two different cardiac damage staging systems for aortic stenosis in a surgical cohort of patients who underwent surgical aortic valve replacement

European Heart Journal - Cardiovascular Imaging

29 January 2025
Organised by: Logo
ESC Journals

Abstract

AbstractPurpose

In recent years, several cardiac damage staging systems for severe aortic stenosis (AS) have been proposed. These systems may help to stratify patients’ risk and select the adequate timing for intervention. Our aim was to perform an external validation of two staging systems (1,2) in patients who underwent surgical aortic valve replacement (SAVR).

Methods

Retrospective analysis of a prospective cohort of patients with AS who underwent SAVR in a tertiary care center from 2017 to 2022. Echocardiographic parameters measured prior to surgery were analyzed, and patients were classified into the different stages of two published staging systems.

The area under the ROC curve (AUC ROC) was used to measure the discriminatory capacity of the two staging models to predict the primary endpoint of 1-year mortality after SAVR.

Results

244 patients were analyzed. Mean age was 69 (9.5) years, and 37.7%f were female. Mean EuroSCORE II was 2.6 (3.3). The primary endpoint occurred in 11 (4.5%) patients.

The distribution of patients in the different stages of cardiac damage is displayed in Figure 1. According to Staging system 1, the majority of patients (50%) were intervened on in a Stage 1 of cardiac damage (left ventricular subclinical damage, considered when global longitudinal strain [GLS] was ≥ −17%). In the case of Staging system 2, 29% of patients were classified as left ventricular damage and 47.9% as left atrium or mitral damage.

The area under the ROC curve of Staging system 1, which incorporated cardiac mechanics, was 0.688 (95% CI: 0.510-0.865), and showed better predictive performance than Staging system 2 [0.593 (95% CI 0.432–0.736)] (Figure 2).

Conclusions

In a surgical cohort of patients with AS, a cardiac staging system that include GLS and right ventricular-arterial coupling showed better discriminatory yield for one-year mortality. Therefore, when available, assessing LV mechanics and the interrelation between right ventricular function and afterload could help in better risk stratification and prognostic assessment in this context.

Cardiac damage staging systems

AUC curve for 1-y mortality

Contributors

R Roman
R Roman

Author

C Olmos
C Olmos

Author

San Carlos Clinical Hospital Madrid , Spain

C Gil
C Gil

Author

M Luque
M Luque

Author

R Bassa
R Bassa

Author

E Pozo
E Pozo

Author

P Mahia
P Mahia

Author

F Islas
F Islas

Author