Straining to predict mortality in low flow low gradient aortic stenosis patients undergoing TAVR: can GLS and myocardial work parameters improve outcomes?

Abstract

Listen

Translate article

Abstract Background Classic low flow low gradient aortic stenosis (LFLG-AS) represents 5-10% of the Aortic Stenosis (AS) population and is associated with poorer clinical outcomes when compared to other AS subtypes. The survival rate of these patients is low (40–60% at 2-yrs) with conservative management. The TOPAS study found a transfemoral Transcatheter Aortic Valve Replacement (TAVR) adjusted hazard ratio (HR) of 0.23 and Surgical Aortic Valve Repair adjusted HR of 0.36. However, mortality within 2-yrs of TAVR can still be as high as 30%-50%. Currently there are no pre-TAVR echocardiographic markers that reliably predict post-TAVR outcomes. Global longitudinal strain (GLS), a load-dependent measurement, in conjunction with the recently emerging load-independent measurements, global myocardial work index (GWI) and global myocardial work efficiency (GWE), have the potential to improve peri-procedural risk stratification. Purpose This study sought to establish the sensitivity and specificity of using GLS, GWI, GWE, Contractile Reserve, and Left Ventricle Ejection Fraction (LVEF) in predicting mortality for LFLG AS patients undergoing TAVR. Methods We measured baseline GLS, GWI, GWE, Contractile Reserve, and LVEF, using GE Echo Pac AFI software (ver. 203) in patients with LFLG AS (n=29, age 79 ± 8 yrs, 69% males) that underwent low dose Dobutamine Stress Echocardiography (DSE) pre-TAVR (2017 to 2023). Contractile flow reserve (defined as >20% increase in LV stroke volume (SV) during DSE was recorded. LVEF was re-measured by a single operator. Receiver operating curves (ROC) were created to predict mortality using SPSS. Results Baseline LVEF was 37.6 ± 9.3% with mean AV gradient of 30 ± 10 mmHg prior to TAVR. 59% (17) had contractile reserve (25 ± 19% increase in SV). There was -0.3 ± 15.7% change in LVEF and 24% (10) patients had ≥10% improvement in LVEF at 1-yr Total mortality was 21% (6) and heart failure admission was 34% (10) at 1-yr Post TAVR, GLS (AUC 0.82, p=0.017), GWI (AUC 0.688, p=0.209) and GWE (AUC 0.82, p=0.001) can predict 1-yr mortality compared to baseline LVEF (AUC 0.536, p=0.83), and contractile reserve (AUC 0.551, p=0.132) (Figure 1). Baseline GLS of ≥ -7.5% (83% sensitivity, 83% specificity), GWI ≤ 817 mmHg% (83% sensitivity, 67% specificity) and GWE ≤74% (83% sensitivity, 67% specificity) can predict 1-yr mortality. Conclusion This study demonstrates that GLS, GWI, and GWE possess superior predictive capabilities for predicting mortality within 1-yr post-TAVR compared to baseline LVEF and contractile reserve. A baseline GLS of ≥ -7.5%, GWI ≤ 817 mmHg%, and GWE ≤ 74% were identified as thresholds with high sensitivity and specificity. This study has notable limitations, being a single-center retrospective study with a small sample size, however, the findings justify replication of this study in a larger, multicenter cohort, in an effort to improve risk stratification in the patients with LFLG-AS being considered for TAVR.