Physiology

Left atrial strain is a feasible and rapid speckle-tracking tool for assessing atrial function, which has been shown to have diagnostic and prognostic value. Each of the phases of atrial function can be measured with strain imaging, and there are normal ranges (Appendix Table 1) for reservoir function (which accommodates pulmonary venous return during ventricular systole), conduit function (which accommodates pulmonary venous return during early diastole), and contractile function (which facilitates ventricular filling during atrial systole) [5, 6] (Fig. 1). Nonetheless, it is important to appreciate that none of these parameters purely assess atrial function, because of the interplay between atrial and ventricular activity, preload, and afterload. Thus, reservoir function is determined by atrial compliance during ventricular systole, but is also governed by LV end-systolic volume and LV base descent during systole. Likewise, the LA and LV freely communicate during diastole, so conduit function is dependent on LV relaxation and stiffness (chamber compliance) and as well as on reservoir function and atrial compliance. The atrial booster pump (contractile strain) is mainly related to the effectiveness of atrial contractility but the quality of pump function depends on LV systolic reserve, atrial preload (degree of venous return), and atrial afterload (LV end-diastolic pressures) [5, 6].

Fig. 1

Examples of LA strain curves. A Healthy subject; LA volume (53 mls) is normal, as are reservoir (S_R), conduit (S_CD), and contractile strain (S_CT). B Heart failure; in the presence of indeterminate left atrial pressure (normal LA volume, but borderline E/e’, and no tricuspid regurgitation jet), LA reservoir strain > 18% is reassuring that LA pressure is normal. More reassuring still would have been an LA contractile strain > 14% [15••]. C Risk of atrial fibrillation. LA volume is borderline. The reservoir strain (18%) is associated with a > sixfold increased hazard of developing AF, compared with people with a reservoir strain > 23%. In absolute risk terms, this would be a rate of > 4% per year vs. < 1% per year [50]

Techniques for Assessing LA Function

A number of imaging tools may be used for the assessment of LA size and function, but echocardiography remains the most accessible and least expensive [5]. While the assessment of LA size with either 2D or 3D measurement of LA volume indexed to body surface area (LAVI) is ubiquitous, volumetric analysis of the LA at multiple phases of the cardiac cycle is difficult. Measurement of LA volumes at their largest (at LV end-systole), smallest (at LV end-diastole), and immediately before atrial systole (at the onset of the P wave on the electrocardiogram) estimate the reservoir, conduit, and booster pump functions. Calculation of total, passive, and active emptying fractions, which pertain to the reservoir, conduit, and booster pump functions, is possible based on the given volumes [5, 6]. However, the volumetric approach is less feasible than STE for assessing LA function. Nonetheless, technical considerations are important for STE, including good image quality and an appropriate frame-rate (preferably 50–70 frames per second) (Fig. 1A). Potential technical challenges include limited endocardial resolution in the far field of apical views, thin atrial walls, and the existence of adjacent structures such as pulmonary veins or LA appendage.

Other imaging modalities for LA assessment include cardiac magnetic resonance (CMR) and cardiac computed tomography (CCT). In addition to LA anatomy and function, CMR is also able to assess LA fibrosis from late gadolinium enhancement (LGE) [7, 8]. CCT also permits the evaluation of pulmonary venous anatomy [9], estimation of risk of AF recurrence, and coronary artery disease [8].

LA Strain in the Assessment of LV Diastolic Function

Echocardiography remains the primary imaging technique for the assessment of LV diastolic dysfunction (DD), by determination of LV relaxation and compliance, and the estimation of LV filling pressure [10]. LA strain is significantly and negatively associated with the severity of LVDD [11,12,13], and conversely, the severity of DD is linked to the prevalence of abnormal LA strain values [12, 13]. LA strain will become in important adjunct in the estimation of LV filling pressure. In patients with a mitral E/A ratio from 0.8 to 2.0, LV filling pressure is likely raised when two of the three predictive features (average E/e’ > 14, peak TR velocity > 2.8 m/s, and LAVi > 34 ml/m2) are present. If one of the three criteria is absent and the remaining two provide conflicting results, LA reservoir strain can serve as the third parameter. The effectiveness of this approach was demonstrated in a recent multicenter study, which utilized invasive LV filling pressure as a reference point [14, 15••]; LA reservoir strain < 18% corresponded to an increased LV filling pressure. However, LA strain is more difficult to use in the assessment of LV filling pressure in AF.

LA contractile strain may be an even stronger correlate of LV filling pressure than LA reservoir strain. The associations of both LA reservoir and booster strains with LV filling pressure are modest in patients with normal LV ejection fraction. In such circumstances, high values of LA booster strain accurately identify non-elevated LV filling pressure [14]. However, the disadvantage of using LA contractile strain as a marker of high LV filling pressure is that low values may also reflect atrial stunning post-atrial arrhythmias. Lower velocities of septal and lateral a’ (which show mitral annular velocity during LA contraction), and reduced transmitral A wave (determined by atrial contraction and LV compliance), may be confirmatory of impaired LA function.

LAVi is a reliable parameter for assessing the long-term effect of increased LV filling pressure on the LA. However, it has been demonstrated that LAVi combined with LA strain significantly improves the detection rate of LV diastolic dysfunction and increased LV filling pressure compared to using only LAVi. This finding is relevant for patients with preserved ejection fraction, among whom elevated LV filling pressure is indicated by reductions in LA reservoir strain and LA pump strain [16]. Moreover, LA reservoir strain has a stronger association with invasive LV filling pressure than does LAVi [17, 18]. Importantly, the measurement of LA reservoir strain using STE has shown very high feasibility, with a rate of ~ 95% [14].

Prognostic Role of LA Strain in Heart Failure

Diastolic dysfunction and raised LV filling pressures are markers of adverse outcome in people with and without symptomatic heart failure. Left atrial strain delivers a practical and reliable diagnostic tool to identify elevated LA pressure both during rest and exercise in patients with HF or suspected HF (Fig. 1B). Moreover, LA strain demonstrates the highest prognostic value in patients with HF among the other non-invasive indices of filling pressure [19]. LA strain represents a robust approach for evaluating pulmonary artery wedge pressure (PAWP) at rest. Furthermore, it demonstrates consistent capability in detecting pathological increases in PAWP during exertion, even when resting pressures appear normal [19].

The progressive worsening of exercise capacity from stage A through stage B to stage C in HF is accompanied by a gradual impairment of LA reservoir and contractile strain and strain reserve. Individuals without HF with reduced resting LA reservoir strain and without the ability to increase LA reservoir strain after passive leg raise have been found to have a shorter 6-min walk distance than those with preserved atrial strain and strain reserve [20]. Peak VO2 is directly associated with LA compliance. LA structure and function are associated with non-specific HF symptoms in stages A and B, even after adjustment for comorbidities, risk factors, echocardiographic parameters of cardiac structure, and diastolic function.

In patients with heart failure with preserved ejection fraction (HFpEF), LA strain seems superior to LAVi for predicting adverse outcomes. Among patients with HFpEF, LA strain is more strongly associated with adverse outcomes than longitudinal LV and RV strain measurements [21]. Similarly, in patients with heart failure with reduced ejection fraction (HFrEF), reduced peak atrial longitudinal strain (PALS) was significantly associated with more advanced HF and greater impairment in both LV systolic and diastolic function indices. Impaired PALS strongly predicts adverse outcomes, independent of other clinical and echocardiographic factors used to assess prognosis. Moreover, PALS provides additional prognostic information concerning LAVi, LV filling pressures, and LV global longitudinal strain (GLS) [22].

While an increase in LAVi can indicate the chronic impact of elevated LV or LA filling pressures and be a predictor of adverse outcomes in patients with HFrEF [23], LA enlargement can also occur in patients with normal filling pressures, i.e., healthy athletes or individuals with lone arrhythmias. Therefore, measurements of LA dimensions may not always offer a dependable estimation of LA pressure or function [22].

Detection of Atriopathy and Prediction of AF with LA Strain

Cardiac output can be reduced by around 15–20% by loss of atrial contraction when atrial fibrillation occurs in the context of HF. Conversely, increase of the LA booster pump function may compensate for decreased early filling in patients with impaired LV relaxation. In this setting, the likely reason for increased LA contractility is increased LA volume (Frank-Starling’s law). However, prolonged exposure to increased LV filling pressure leads to extreme dilatation and exhaustion of the Frank-Starling response, with both impaired LA function and the risk of AF [24]. A significant evidence base has been gathered regarding the prediction of AF with LA strain (Appendix Table 2).

In HFpEF, patients who develop AF are characterized by reduced LA strain (Fig. 1C)—independent of older age, higher BNP, and creatinine levels, increased LAVi, LV mass, impaired diastolic function, as well as reduced exercise capacity. Peak atrial contractile strain (PACS), peak atrial longitudinal strain (PALS), and LAVi demonstrate the highest predictive value for AF (PACS and PALS were independent of clinical data, LAVi, and E/e’ ratio). These three parameters (PACS, PALS, and LAVi) identify three key predictors of atrial fibrillation, and their combination is able to effectively distinguish high and low AF risk in HFpEF. This high-risk subset demonstrated a 33-fold increase in hazard [25].

The recurrence of AF after ablation is an important problem. In patients in sinus rhythm, LA reservoir and LA conduit strain have been strongly associated with atrial fibrillation recurrence, independent of LAVi, BMI, and LA pressure. However, some evidence points toward left atrial appendage velocity (LAAV), obtained at transesophageal echocardiography, as the strongest independent predictor [26]. In patients with AF, LAVi has been proposed as the only independently associated risk factor for AF recurrence after ablation, and LA strain appears to be less useful.

LA strain may help identify a group of patients with a high risk of developing AF. The appropriate clinical response to this predictor needs further study. It seems likely that pre-emptive lifestyle intervention, including stopping alcohol intake, may reduce the risk of progression. Whether there is sufficient justification to initiate empirical anticoagulation in this setting also requires further study [27].