Discussion
The main result of this study is that in CHF outpatients, the LA reservoir, a 2D-STE measure of atrial function, is independently associated with worsening heart failure. Moreover, the integration of the LA reservoir with the ratio between mitral early diastolic peak velocity at pulsed Doppler and TDI provides a measure reflecting the interaction between atrial and ventricular function which is even more accurate to stratifying prognosis.
LA function is often considered only a passive bystander of LV dysfunction. However, several studies have shown that LA contractile function is determined by intrinsic LA contractility, LV compliance, LV filling pressures (LA afterload), and pulmonary vascular capacitance [1]. Atrial function improves ventricular filling by storing blood during ventricular systole while the atrioventricular valves are closed (reservoir function), as a passive conduit and emptying during early ventricular filling (conduit function), and providing additional ventricular filling with active contraction before atrioventricular valve closure (booster pump function). The interplay between LA and LV functions throughout the cardiac cycle (LA–LV coupling) is crucial in several pathophysiological conditions [3]. In this pathophysiological setting 2D-STE could represent a useful tool not only to evaluate the LV and RV functions [11-13] but also to provide measures reflecting the atrial function [14]. In particular, it allows the study of the complexity of the atrial deformation by analysing three sequential phases, i.e. the reservoir, the conduit, and the contraction.
The reservoir phase begins at the closure of the atrio-ventricular valves and ends with their opening at the end of isovolumic ventricular relaxation. In this phase, the atrial pressure quickly rises while receiving blood from the veins, and the atria stretch to their maximum capacity. The greater the atrial stretch, the greater the value of the reservoir. The other two phases of atrial function are reciprocal to the reservoir. The first is the conduit phase, which is related to the first half of ventricular diastole, after the opening of the atrioventricular valves. During this phase, the atria quickly unload in the ventricles the blood accumulated in the reservoir phase (passive atrial emptying), and the strain analysis will register a negative deflection of the curve, followed by a plateau which corresponds to the diastasis, when the blood directly flows from the veins to the ventricles. This phase is influenced by the compliance of the ventricles and the diastolic function [14]. The last phase is the contraction phase, corresponding to the atrial systole. This phase reflects the contractile capacity of the atria, along with the ventricular stiffness and the ventricular end diastole filling pressures [14].
Previous studies have demonstrated the role of peak atrial strain in predicting heart failure prognosis [5, 6]. However, different from these studies, we attempted to evaluate the prognostic role of all the phases of the atrial function. Interestingly, all the measures were associated with heart failure worsening, but only LASr remained independently associated with the occurrence of events during multivariate regression analyses. This evidence could be related to the ability of the reservoir to reflect a global atrial dysfunction. Consequently, although the conduit and contraction phases could help to define more clearly the mechanisms responsible for atrial dysfunction, the reservoir, being the expression of both the other phases, is associated with more important prognostic information.
This hypothesis is supported by the studies demonstrating the close relationship between a reduced LASr and the pathophysiological mechanisms underlying atrial dysfunction [14–16], i.e. the increased LV filling pressure, the LA remodelling, and the loss of contractile capacity. For this reason, the impairment of LASr could precede anatomical alterations [15], thus allowing a more accurate identification of diastolic dysfunction than conventional echo variables, such as LA volume index or E/e’ [17-20]. Moreover, the therapeutic strategies able to improve LV function can be associated with the improvement of atrial function [21].
The relevance of our study is also related to the demonstration that the combination of the strain measure of LA function, LASr, and that of LV filling pressure based on pulsed Doppler and TDI, E/e’, is characterised by even greater prognostic accuracy in predicting heart failure worsening. In previous studies, the ratio between LASr and E/e’ showed greater accuracy in detecting high filling pressure [7, 22], but this measure should also be considered as a parameter that can better reflect the relationship between LA and LV functions. In fact, LV function worsening could cause an increase in LV end-diastolic pressures, which can then cause LA wall stiffening, thereby resulting in decreased LA relaxation and decreased LA strain peak during LV systole. Conversely, the decreased LA function could further favour the increase in LV filling pressure, thus leading to a vicious circle responsible for heart failure progression. Consequently, lower ratio between LASr and E/e’ corresponds to worse interaction between LA and LV functions, i.e. greater LV filling pressure and worse LA function, thus allowing to provide a significant parameter of prognostic stratification.
In conclusion, 2D-STE is an echocardiographic technique able to provide measures of atrial function independently associated with heart failure worsening. Among these measures, the reservoir shows greater accuracy in predicting the events. Moreover, this accuracy is even greater when the strain measure of the reservoir is combined with the Doppler measures related to LV filling pressure. These data further strengthen the possible usefulness of 2D-STE in providing information about both ventricular and atrial functions.