“Better Is the Enemy of Good”: The Deleterious Effects of Supra-Normal Left Ventricular Ejection Fraction

Abstract

Although not recognized as a distinct clinical entity in the current guidelines [1], a new phenotype of heart failure (HF), characterized by supra-normal left ventricular ejection fraction (LVEF) > 65% (snLVEF), draws the attention of the scientific community in the recent years. In a large paper, Wehner et al. reported a U-shaped relationship between LVEF and all-cause mortality [2], irrespective of age and other comorbidities, highlighting the fact that patients with snLVEF have an increased risk of death, which might be similar to that of patients with reduced LVEF [2]. Left atrial (LA) deformation assessed by speckle-tracking echocardiography (STE) is a sensitive marker of diastolic dysfunction [3] and current consensus recommendations suggest that LA reservoir strain (LARS) should be used as an additional parameter for evaluating LV filling pressures in patients with preserved LVEF [4]. Moreover, LARS was recently found to be an independent predictor of mortality, stroke, and HF in patients with normal LVEF [5].

Based on this previous knowledge, in the current issue of Echocardiography, Liu and colleagues [6] investigated LA and LV deformation patterns in hypertensive patients with snLVEF. Their study retrospectively enrolled 101 patients with essential arterial hypertension and preserved LVEF ≥50%, who were divided into low-normal LVEF (lnLVEF; 50%–59%), mid-normal LVEF (mnLVEF; 60%–69%), and supra-normal LVEF (≥70%). Their findings showed that hypertensive patients with snLVEF had impaired LA reservoir and conduit functions with preserved pump function, while patients with lnLVEF exhibited impairment of all three LA phasic functions. The authors also found an inverted U-shaped relationship between LARS and LVEF, proving that snLVEF has a deleterious effect on LA remodeling and mechanics, potentially leading to an adverse outcome.

During the past years, researchers have become increasingly aware of the potential detrimental effects of snLVEF. In a population-based cohort of 486 754 individuals, a LVEF≥70% was associated with decreased survival and underdiagnosed HF [7], while in women already diagnosed with HF, a snLVEF was associated with a higher risk of all-cause death, both in the acute [8] and chronic setting [9]. A recent study enrolling patients with transcatheter aortic valve replacement (TAVR) found that patients with LVEF>65% had worse outcomes after TAVR than patients with LVEF between 50% and 65% [10]. Moreover, HF with snLVEF seems to differ from HF with preserved LVEF (HFpEF) not only in terms of survival, but also of response to treatment. For example, the EMPEROR-Preserved trial proved the beneficial effects of empaglifozin in HFpEF, but these effects were not consistent in the subgroup of patients with snLVEF [11].

In an interesting magnetic resonance study on normal adults with LVEF > 57%, LVEF in the highest quartile was associated with a higher risk of adverse events, particularly in those with lower stroke volume [12]. Diminished stroke volume can be a consequence of increased afterload, which is usually encountered in hypertensive patients. Furthermore, arterial hypertension can determine LV hypertrophy (LVH), which reduces the LV cavity and can increase LVEF, thus explaining the common association between snLVEF and LVH [7]. Consequently, the presence of a snLVEF in hypertensives does not reflect increased myocardial contractility, but rather the adverse remodeling of the LV.

In a hemodynamic characterization of patients with HFpEF, a group of researchers proved that patients with LVEF>60% have a hypercontractile state with excessive LV afterload, increased myocardial stiffness, and diminished preload reserve, when compared with patients with LVEF between 50% and 60%, suggesting that these ranges of LVEF manifest as different hemodynamic phenotypes [13]. These distinct hemodynamic profiles have distinct effects on LA remodeling, and LA strain is a parameter that reflects early stages of atrial myopathy [14], while also being a sensitive marker of LV diastolic dysfunction [3].

These are consistent with the findings of Liu and colleagues [6], who reported significantly higher relative wall thickness and lower LV volumes in patients with snLVEF than in mnLVEF. It would have been interesting to evaluate the effect of LVH on LA strain reduction in their cohort, since previous research already stated that the presence or absence of LVH modulates the LA phasic function in hypertensives [15-17]. As the authors stated, proving the impairment of LA mechanics in hypertensive patients with snLVEF is a strong point of their research, since this is the first study to investigate all LA phasic functions in this clinical setting. The authors also showed that LARS and LVEF had an inverted U-shaped relationship, speculating that snLVEF might be an intermediate state between mnLVEF and lnLVEF throughout the course of hypertensive heart disease. Since this was a cross-sectional study, it was impossible to evaluate the effect of time on myocardial remodeling, but the authors’ speculation is plausible from a pathophysiological standpoint, as snLVEF might represent an initial step in the LV (mal)adaptation induced by hypertension.

The authors used the correlation between LARS and LV longitudinal strain as a surrogate for left atrio-ventricular coupling. The close relationship between these two parameters is well established [17, 18]. During the last years, a novel left atrioventricular coupling index emerged, defined as the ratio of LA to LV end-diastolic volumes assessed either by 2D echocardiography [19], 3D echocardiography [20], or magnetic resonance [21], and this index showed enhanced prognostic power in various cardiovascular diseases. It would be an interesting direction for future research to evaluate this coupling index in hypertensive patients with snLVEF.

Finally, another novelty of the study by Liu and colleagues [6] is the stratification of preserved LVEF into three categories (not two, as in other previous studies): lnLVEF, mnLVEF, snLVEF, which appear to be distinct phenotypes, potentially evolving into one another in the course of hypertensive heart disease. One question that arises is concerning the optimal LVEF cutoff to define snLVEF. Rather than regarding it as a continuous variable, probably the best approach would be to interpret LVEF while taking into consideration the concomitant LV remodeling, alteration of LA geometry, abnormal LV, and LA deformation. Artificial intelligence would likely be of great value in assembling all this information, in order to determine the true prognostic value of altered myocardial mechanics in patients with snLVEF. Until then, several questions regarding snLVEF arise. Is a separate clinical phenotype or just a transition state in the course of LV adaptation in cardiovascular disease? Which are the major determinants of prognosis in patients with snLVEF? Which is the true extent of left atrio-ventricular decoupling in this entity? Scientific research on this matter is probably just beginning!