Usefulness of atrial function for risk stratification in asymptomatic severe aortic stenosis

We aimed to evaluate the usefulness of left atrial (LA) mechanics and stiffness over global left ventricular (LV) longitudinal strain (GLS) for risk stratification in severe aortic stenosis (AS).

From a cohort of 89 prospective asymptomatic patients with severe AS and normal LV ejection fraction, 82 (32 men, mean age 73±10 years) truly asymptomatic patients, scheduled after a negative exercise echocardiogram, were enrolled. Forty age- and gender-matched prospective, asymptomatic subjects served as controls. Predefined end points were the occurrence of symptoms (dyspnea, angina, syncope), and death during follow-up.

At study entry, patients had: impaired LV GLS (p=0.001), reduced LA reservoir (p<0.001), high LA stiffness (p<0.001), and increased valvulo-arterial impedance (p<0.001) compared to controls. During follow-up [16±14.9 months (ranging from 1 month to 4.2 years)], 53 patients (64.6%) reached one of the endpoints. Patients with events showed lower LV GLS (p>0.001), lower LA reservoir (p<0.001), and greater LA stiffness (p<0.001) than those asymptomatic. On univariate Cox regression analysis, LV GLS (p<0.001), LA reservoir (p<0.001), and LA stiffness (p=0.004) were strong predictors of adverse events. Kaplan-Meier curves showed that event-free survival was significantly higher in patients with a LV GLS ≥16.8% [p<0.001; area under the curve (AUC)=0.922; sensitivity=86%, specificity=80%], a LA reservoir ≥19.8% (p=0.001; AUC=0.860, sensitivity=71%, specificity=84%), and a LA stiffness <0.78 (p<0.001; AUC 0.819, sensitivity 70%, specificity 89%). On multivariate analysis, only LV GLS remained significantly associated with patients' prognosis (hazard ratio=1.49, 95% CI=1.11-2.01, p=0.008).

In asymptomatic patients with severe AS, an efficient cardiovascular system is based on an effective atrial-ventricular interplay. LA function assessment is useful for early identification of risk in these patients. LV GLS however was confirmed to be the best predictor of patients' outcome.