Academic Editor: Buddhadeb Dawn
Left atrial (LA) enlargement and dysfunction increase the risk of atrial fibrillation (AF). Traditional echocardiographic evaluation of the left atrium has been limited to dimensional and semi-quantification measurement of the atrial component of ventricular filling, with routine measurement of LA function not yet implemented. However, functional parameters, such as LA emptying fraction (LAEF), may be more sensitive markers for detecting AF-related changes than LA enlargement. Speckle-tracking echocardiography has proven to be a feasible and reproducible technology for the direct evaluation of LA function. The clinical application, advantages, and limitations of LA strain and strain rate need to be fully understood. Furthermore, the prognostic value and utility of this technique in making therapeutic decisions for patients with AF need further elucidation. Deep learning neural networks have been successfully adapted to specific tasks in echocardiographic image analysis, and fully automated measurements based on artificial intelligence could facilitate the clinical diagnostic use of LA speckle-tracking images for classification of AF ablation outcome. This review describes the fundamental concepts and a brief overview of the prognostic utility of LA size, LAEF, LA strain and strain rate analyses, and the clinical implications of the use of these measures.
Atrial fibrillation (AF) is the most prevalent symptomatic cardiac arrhythmia in clinical practice worldwide. AF increases the risk of ischemic stroke, heart failure, cardiovascular events, and mortality [1, 2, 3, 4]. Atrial fibrosis has been increasing recognized as a contributing abnormality in the development of AF [5, 6, 7]. Atrial fibrosis increases local conduction heterogeneity in the atria and provides a AF-sustaining re-entry substrate [7, 8], which can be identified by delay-enhancement cardiac MRI and intracardiac electroanatomic mapping [9, 10]. However, their time-consuming and invasive nature limit the routine application of these tools in daily practice. Echocardiography provides a real-time and noninvasive method to assess cardiac anatomy and function. Because of its widespread availability and feasibility, echocardiography has been the imaging technique of choice for evaluating the left atrium. Several echocardiographic parameters of left atrial (LA) anatomy, function, and deformation have been used to evaluate atrial fibrosis and the risk of AF [1, 11, 12].
Catheter ablation (CA) is a common treatment strategy in symptomatic AF patients resistant to antiarrhythmic medications, but the long-term success rate is only around 50–80% [13, 14, 15]. LA remodeling is among the most important factors related to the recurrence of AF post CA. Previous studies have investigated the clinical predictors of AF recurrence after CA [16, 17, 18]. P-wave duration can serve as a low cost and widely available predictor of long-term outcome in AF patients undergoing CA [19, 20, 21, 22]; nevertheless, the predictive power of P-wave duration is weaker than that of LA emptying fraction (LAEF) . Echocardiography has the advantages of availability, efficacy, and providing real-time high temporal and spatial resolution images, and thus is best suited for evaluating the possibility of AF recurrence [23, 24, 25, 26].
In this review, we provide a comprehensive overview of the LA echocardiographic parameters associated with new-onset AF and AF recurrence after CA.
Left atrial dimension (LAD) in M-mode measurement is the traditional method used
to assess LA size. LA dilatation reflects the cumulative effects of left
ventricular (LV) filling pressure over time and the severity of diastolic
dysfunction, and can be used as a quantifiable surrogate of the arrhythmogenic
substrate in the development of AF. Previous studies have shown that dilated LAD
is a predictor of AF occurrence in general  and in elderly populations ,
and that the risk of developing AF is proportionate to the extent of LA
dilatation . The Cardiovascular Health Study revealed that patients with LAD
In normal subjects, LA function can be divided into three phases: reservoir,
conduit, and booster pump, which account for around 40%, 35%, and 25% of the
entire LV filling, respectively . To assess LA function, LA volumes are
measured at the mitral valve opening (LAV
Clinically LAEF is a significant echocardiographic parameter for predicting AF
occurrence. Cauwenberghs et al.  demonstrated that LAEF is a
significant predictor of cardiac events and of new-onset AF. The area under the
curve (AUC) of the receiver operating characteristic curve was 0.80 (95%
confidence interval [CI] 0.73–0.88) for new-onset AF at 8 years of follow-up. A
55.5% cutoff value of LAEF had a sensitivity of 0.77 and specificity of 0.72 for
predicting new-onset AF. The Copenhagen City Study also reported that not only
Several reports have highlighted the role of atrial fibrosis in AF pathogenesis
[7, 8, 42]. The development of fibrosis results in atrial myopathy , which is
associated with atrial dysfunction and conduction disturbance . Sung
et al.  reported that LAV
LA size may be a predictor of AF recurrence after CA. A meta-analysis of 22
studies revealed that dilated LAD increases the risk of AF recurrence after CA
regardless of the follow-up duration . Moreover, McCready et al.
 demonstrated that a LAD cutoff value of 43 mm predicted long-term success
following CA for those with persistent AF, with a sensitivity of 92% and a
specificity of 52%. Similar results have been reported for LAV. In the
meta-analysis by Njoku et al.  large LAV and LAV index (LAVI)
increased the odds (odds ratio [OR] 1.032, 95% CI 1.012–1.052) and were
independent predictors of AF recurrence post CA. Shin et al. 
reported that LAV was the only predictor of AF recurrence after CA in
multivariate analysis, and a LAVI cutoff value of 34 mL/m
Several studies have demonstrated that LAEF is useful in predicting the
maintenance of sinus rhythm in patients with AF post CA [52, 53, 54]. Our group
demonstrated that LAEF, but not LAD or LAV, provides optimal prognostic
information for risk stratification in 483 AF patients undergoing CA, which
implies that LA dysfunction is an earlier indicator of atrial remodeling than LA
dilatation . Oka et al.  demonstrated the superiority of
pre-ablation baseline LAEF over LAVI in predicting AF recurrence after CA in 292
patients with PAF undergoing single or multiple procedures. Charitakis et
al.  investigated the association of the risk of AF recurrence with
echocardiographic parameters (LAV
The LV diastolic phase could be divided into early rapid filling, diastasis, and
atrial systole. There is a close interaction between the left atrium and LV
diastolic function. Increased LV filling pressure reduces passive emptying volume
from the left atrium to the left ventricle, triggering a compensatory mechanism
that increases the active emptying volume by enhancing active LA contraction in
the late diastole period . Therefore, structural and functional LA remodeling
is often the consequence of LV diastolic dysfunction . Several
echocardiographic parameters have been suggested as useful in evaluating LV
diastolic function, such as LAVI, transmitral E/A ratio, isovolumic relaxation
time, decelerating time of mitral early velocity, e’ on tissue Doppler imaging,
and E/e’ and tricuspid regurgitation velocity [57, 58]. Although both LAV
LV diastolic dysfunction adversely affects LA structural, functional, and
electrical remodeling ; therefore, patients with the diagnosis of LV
diastolic dysfunction have an increased risk of AF . Tsang et al.
 demonstrated that the risk of incident AF was proportionate to the severity
of LV diastolic dysfunction, and LAVI was the strongest predictor of AF in the
840 elderly patients studied. Rosenberg et al.  used data from the
Cardiovascular Health Study to analyze the influence of echocardiographic
diastolic parameters on the risk of AF. They found that early mitral inflow
velocity (peak E velocity), late mitral inflow (A-wave) velocity-time integral,
and LAD were the predictors of incident AF. Vasan et al.  examined
the diastolic parameters in patients in the longitudinally followed Framingham
Heart Study and found that an E/e’ ratio greater than the median (1.23) increased
the rate of incident AF. Arai et al.  revealed that an E/e’
Heart failure with preserved ejection fraction (HFpEF) is characterized by
elevated LV filling pressures with clinical signs and symptoms of heart failure,
LV diastolic dysfunction and a LV ejection fraction
Parameters of LV diastolic dysfunction may serve as surrogate markers for AF recurrence post CA. Cha et al.  demonstrated an increased relative risk of AF recurrence of 1.8 (95% CI 1.1–3.1) in systolic dysfunction and 1.7 (95% CI 1.0–2.7) in isolated diastolic dysfunction compared with normal function at 1 year after CA. Kumar et al.  in a study of 124 patients undergoing CA for AF, found that high-grade LV diastolic dysfunction, defined by e’ on tissue doppler imaging and deceleration time, was an independent predictor of AF recurrence after adjustment for AF type and LAV (Hazard Ratio [HR] 2.6, p = 0.009). However, Kosiuk et al.  demonstrated that the E/A ratio and decelerating time could predict AF recurrence during the first week after CA, but that long-term results were not influenced by pre-procedural echocardiographic parameters that indicate LV diastolic dysfunction. A possible explanation for this finding is that LV diastolic function may deteriorate after AF ablation, mediated by longer ablation time, with a subsequent impact on LA and LV hemodynamics . In addition, Nedios et al.  revealed that LA asymmetry was associated with LA dilatation and LV diastolic dysfunction and correlated with reduced success after AF ablation, but the presence or the grade of LV diastolic dysfunction was not associated with procedural success. Further investigations are needed into the definition and precise cutoff values to identify LV diastolic dysfunction and into the influence of LV diastolic dysfunction on AF recurrence after CA.
Total atrial conduction time (TACT) is an atrial conduction parameter affected
by atrial conduction velocity and anatomy . The gold standard method of TACT
measurement is intracardiac measurement using invasive electrophysiologic study
. Alternatively, TACT can be estimated noninvasively by PA-TDI interval,
which is defined as the time interval between the onset of P wave on the surface
electrocardiogram and the peak of A’ wave on tissue Doppler imaging . Erdem
at al.  revealed that TACT measured by PA-TDI correlated with that
measured via invasive electrophysiologic study. Prolonged PA-TDI interval
reflecting atrial remodeling [80, 81] has been shown to increase the risk of AF
in various cohorts [82, 83, 84, 85]. Vos et al.  reported that a prolonged
PA-TDI interval is vulnerable to new-onset AF in patients with various
cardiovascular diseases with a HR of 1.375 per 10 ms increase in PA-TDI interval.
Muller et al.  revealed that patients with prolonged PA-TDI
intervals in the cryptogenic stroke cohort had higher incidences of AF detection.
The AUC of the receiver operating characteristic curve was 0.94 for occult AF
detection, and a PA-TDI interval cutoff value of 145 ms had a sensitivity of
93.8% and a specificity of 90.5% for identifying occult AF at 1-year follow-up.
Leung et al.  investigated the relation between echocardiographic
markers of LA fibrosis and AF progression in patients with new-onset AF (620
subjects) and controls (342 subjects). They found that PA-TDI interval and LA
reservoir strain were correlated negatively, and patients with persistent AF had
a longer PA-TDI interval and smaller LA reservoir strain than those with PAF. In
predicting AF recurrence after successful electrical cardioversion or CA, Mueller
et al.  demonstrated that PA-TDI interval at a cutoff value of 152
ms had a sensitivity of 87% and a specificity of 100% for predicting early AF
recurrence after successful cardioversion in patients with non-PAF; Uijl
et al.  demonstrated that PA-TDI interval had a better discriminative
performance than LAV
Speckle-tracking echocardiography (STE) is a novel, non-Doppler echocardiographic method to measure the magnitude and rate of atrial myocardial deformation by calculating the longitudinal strain and strain rate independent of cardiac rotational motion and the tethering effect [90, 91]. Strain is a dimensionless index that reflects total deformation of the myocardium relative to its initial length during the cardiac cycle , expressed as a positive value for lengthening or a negative value for shortening. STE tracks the natural acoustic markers within a region of interest (kernel) frame-by-frame, evaluating the geometric shift of each kernel throughout the cardiac cycle . Fig. 1A shows an example of LA strain via the apical 4-chamber view. LA strain reaches its maximal value just before the mitral valve opening, and LA strain during the reservoir phase (LASr) is measured as the strain value at the mitral valve opening minus that at the ventricular end-diastole (a positive wave occurring during the ventricular systole) . When the LA conduit phase begins, LA volume gradually decreases to a plateau until the 2nd late peak, just before the onset of the active atrial contractile phase. The strain value at the onset of atrial contraction minus that during the mitral valve opening is a surrogate of LA strain at the conduit phase (LAScd). The strain value at the ventricular end diastole minus that during the onset of atrial contraction is a surrogate of LA strain at the contraction phase (LASct). Strain rate is the rate by which the deformation occurs. Fig. 1B shows an example of LA strain rate. There is one positive peak during the reservoir phase (pLASRr) and two consecutive negative peaks during the LV diastolic phase. The first peak represents passive myocardium shortening (pLASRcd) and the second peak is the minimal value after the LA active pump phase (pLASRct). The assessment of LA strain and strain rate can use a 4-chamber view or both 4- and 2-chamber views to report the average values from 6 or 12 segments, respectively .
Transthoracic echocardiography in the apical four-chamber view
showing left atrial (LA) longitudinal strain and strain rate. (A) LA strain in a
paroxysmal atrial fibrillation (PAF) patient 1-day post ablation. Taking the R
wave as the starting point, the first positive peak corresponds to the LA
reservoir phase (LASr) (red arrow), the second peak corresponds to the LA
contractile phase (LASct) (orange arrow), and the difference between the two
peaks corresponds to the conduit phase (LAScd) (blue arrow). The traces are
time–displacement displays, with the x-axis representing time and the y-axis
showing myocardial shortening as negative and lengthening as positive (%). The
depicted LA wall is divided into six segments marked by different colors. (B) The
LA longitudinal strain rate in the same patient as in panel (A). The traces are
time-velocity displays, with the x-axis representing time and the y-axis
representing velocity (s
STE can be used to assess atrial fibrosis [12, 93] and LV diastolic dysfunction , and serve as a surrogate marker of LA remodeling to detect early LA dysfunction even prior to structural changes of the left atrium [94, 95]. Kuppahally et al.  described an inverse relationship between the degree of atrial fibrosis detected by delay-enhancement cardiac MRI and the LA strain and strain rate as shown by STE. Eichenlaub et al.  reported that LASr, LAScd, and LASct were correlated with LVA, as measured by intracardiac voltage mapping, in patients with persistent AF undergoing CA; among the three strain parameters, LASr was the most powerful predictor of atrial fibrosis. Laish-Farkash et al.  demonstrated good correlation between LASr and LA LVA as assessed by invasive intracardiac electroanatomic mapping. The LASr cutoff value of 19.7% had a sensitivity of 85.2% and a specificity of 73.3% in predicting the presence of LVA. Therefore, a reduced LA deformation during the reservoir phase may be an early marker of the extent of LA fibrosis , which is associated with the incidence of AF . Park et al.  demonstrated that LASr was a significant predictor of new-onset AF in heart failure patients (397 of 4312 patients) regardless of the LA size.
Most cutoff values of LA strain are based on studies involving a small number of
subjects and depend on age, sex, ultrasound manufacturer, and post-processing
software package . To establish age- and sex-based normative values of LA
strain in the general population and to assess the prognostic yield of lower
limits of normal LA strain in relation to future AF, a substudy of the fifth
Copenhagen City Heart Study evaluated 1641 healthy participants and reported the
median values (and the corresponding limits of normality) for LASr, LAScd, and
LASct were 39.4% (23.0–67.6%), 23.7% (8.8–44.8%), and 15.5% (6.4–28.0%),
respectively . These values were similar to the results of the meta-analysis
by Pathan et al. , which showed cut-off values of 39% for LASr,
23% for LAScd, and 17% for LASct in healthy adults. To investigate whether LA
strain can be used to predict new-onset AF in the general population, Hauser
et al.  conducted a prospective longitudinal study including 3590
participants from the fifth Copenhagen City Heart Study. Compared to the
reference group (patients with LASr
In addition to enabling the identification of patients with a history of AF, STE
provides prognostic information for the risk stratification of AF patients
undergoing CA. Hammerstingl et al.  demonstrated that LASr was
significantly reduced in patients with recurrent AF compared to those without AF
recurrence. Motoki et al.  demonstrated that a low LASr at a cutoff
value of 23.2% could predict the status of sinus rhythm maintenance after CA
with a sensitivity of 76% and a specificity of 66%. Parwani et al.
 demonstrated that a LASr cutoff value of 10% predicted post-CA AF
recurrence with a sensitivity of 97.9% (95% CI 88.9–99.6%) and a specificity
of 78.2% (95% CI 65.6–87.1%). One meta-analysis study including 12 studies
and a total 1025 AF patients revealed that LASr was a significant predictor of
post-CA AF recurrence by multivariable pooled analysis (OR 1.16, 95% CI
1.09–1.24) . In addition to LASr, LASct is also reported to be associated
with the outcome of AF ablation. Wen et al.  demonstrated that
LASct is an independent risk factor for AF recurrence; the 5-year cumulative
recurrence probability was much higher in patients with LASct
Even if those with LAD
Even if STE provides a feasible and reproducible assessment of LA function, STE is dependent on the quality of echocardiographic images and frame rates, and requires time-consuming offline analysis. Therefore, it may not suitable for all clinical settings . In addition, intervendor discordance of LA strain assessed by STE remains a problem to be solved. For example, LA reservoir strains differ significantly by using different speckling tracking analysis systems (GE vs. Siemens) . The intervendor/intersoftware variability should be considered when discussing published LA strain values.
LA electrical and mechanical dysfunction coexist in the early phase before LA enlargement . Sarvari et al.  demonstrated that inhomogeneous contraction of the left atrium potentially predicted AF recurrence after ablation. Because STE is angle-independent and can assess regional myocardial function and timing accurately, the regional differences in 2-dimensional (2D) STE-derived LA strain and strain rate potentially could be used to measure heterogeneous LA fibrosis and dysfunction indirectly. LA mechanical dispersion is calculated as the standard deviation in time to peak strain of the LA segments . It is greater in AF patients than in healthy individuals, increases proportionately to the duration of AF , and provides prognostic information on the risk of AF recurrence in patients after ablation [116, 119]. In a case-control study, patients with new-onset AF had significantly worse LASr and LASct, and more pronounced LA mechanical dispersion, than those without AF . However, it is time-consuming to calculate the standard deviation for parameters of LA mechanical dispersion because sophisticated mathematics is needed for averaging the 2–3 instances of six segmental values per apical 4-chamber and 2-chamber views (3 peaks of LA strain rate curve in sinus rhythm). Alternatively, the curved M-mode color images of LA strain and strain rate provide detailed spatial and temporal information on LA deformation mechanics. These images provide a unidimensional view of LA strain and strain rate, illustrating the changes in length and in strain/sec of the depicted LA wall along the time axis, respectively. As shown in Fig. 1C–D, the spatial and temporal information of LA deformation can be displayed in these images, on which blue or red color, deep or light hue, and pattern of color distribution indicate the direction, strength, and homogeneity of LA deformation, respectively. However, it is challenging to use visual estimation to precisely differentiate these images. Recently our group demonstrated that a deep convolutional neural network (CNN) analysis can successfully incorporate spatial and temporal features from these STE images into an overall assessment of LA deformation mechanics; indeed, the STE image-based CNN model outperformed the logistic regression model using LAD, LAEF, LA strain, and strain rate in predicting AF recurrence after CA . This study demonstrated the potential advantages of supervised deep learning with CNNs to classify images to provide prognostic information for AF intervention. Note that this retrospective study included only 606 patients, and large prospective studies are needed to optimize CNN model performance. Recently, manufacturers have begun developing dedicated software packages for LA strain measurement after publication of the common standards to assess LA strain . Newly-available softwares, such as AutoStrain (TomTec) or LA Automated Function Imaging (Echo-Pac), allow for a quick assessment of LA strain. Future goals would be to achieve fully automatic generation and interpretation of LA STE images, provide fast and reproducible assessment of LA deformation properties, and validate and enhance the performance of CNN models in this domain.
LA substrate modification in addition to pulmonary vein isolation improves AF ablation outcome [122, 123, 124]. Maintenance of sinus rhythm leads to histological reverse remodeling and functional recovery, shown by reduced LA size, improved LA function, and increased LA conduction velocity [55, 125]. However, LA ablation itself impairs LA function, a result related to the extent of scarring . As a result of the different degrees of myocardial damage associated with the different ablation strategies and AF populations, a discrepancy exists in the literature regarding LA functional reverse remodeling after successful AF ablation. Tops et al.  found that LA structural reverse remodeling was associated with a concomitant improvement in LA strain. Spethmann et al.  demonstrated that LASr and LASct normalized within 6 months after CA in PAF patients with no AF recurrence. Perea et al.  used cardiac MRI to reveal that extensive LA linear lesions reduced LA volume and preserved or even increased LAEF in most patients after successful CA. However, Lemola et al.  found that LA linear ablation restored sinus rhythm but compromised LA systolic function in patients with PAF. A meta-analysis by Jeevanantham et al.  revealed that successful CA significantly decreased LAD and LAV without significant influences on LAEF. To evaluate the influence of CA outcome on LA reverse remodeling in the same patients, Yang et al.  studied 38 patients undergoing a repeat CA for AF recurrence after a 1st circumferential pulmonary vein isolation. The absence of LA size reduction after a 1st unsuccessful CA and the presence of significant LA size reduction after a successful second CA in the same patients imply that procedural success was associated with LA structural reverse remodeling. However, LAEF, LA strain, and LA strain rate were not concomitantly improved. Another meta-analysis by Xiong et al.  (25 studies, 2040 patients) revealed that LAEF is significantly decreased in PAF but insignificantly changed in persistent AF after CA. It is likely that differences in the extent of scarring associated with different ablation strategies, preexisting LA fibrosis, and clinical outcome contribute to variable changes in LAEF after CA between PAF and persistent AF patients. Recently, we noted significant LA reverse remodeling, evidenced by reduced LA size and improved LAEF, in non-PAF patients undergoing a successful LVA-guided LA linear ablation . LA functional reverse remodeling was noted even in patients undergoing extensive LA linear ablation. Possibly, the LA linear ablation strategy targeting LVA to avoid damage to otherwise healthy LA myocardium could help preserve the effect of LA functional reverse remodeling.
Although the results are variable regarding LA functional change after
successful AF ablation, LA structural reverse remodeling has been consistently
observed after successful AF ablation and might be considered as a marker of
freedom from AF recurrence. By using different variables and definitions, Kagawa
et al.  demonstrated that a reduction of
Compared with cross-sectional observational studies, longitudinal studies can avoid time-invariant unobserved individual differences, detect changes in parameters beyond a single moment in time, and establish sequences of events to suggest cause-and-effect relationships. Our group conducted a three-year longitudinal study to evaluate the long-term prognostic influence of LAD remodeling on the outcome of AF ablation. We found that a longitudinal linear mixed model-based two stage model outperformed a logistic model using the baseline LAD in classifying outcome status after AF ablation . In addition, LAD was shortened over the first 3 months and remained stable up to 36 months after CA. Similarly, Reant et al.  also found a reduction in LAD during the first 3 months after CA, which then remained stable up to 12 months post CA. The degree of LAD reduction was significantly influenced by the baseline LAD . Interestingly, LAD was reduced in both the success and failure groups. Because the ablation lesions themselves also decrease LA size , the prognostic value of LAD reduction in predicting the outcomes of AF ablation remains a matter that needs clarification. In addition, further longitudinal studies of LA functional remodeling may unveil the long-term prognostic influence of the extent of reversibility of LA deformation parameters on AF ablation outcome.
Three-dimensional (3D) echocardiography is a novel approach providing a
non-invasive method to analyze cardiac anatomy and function. The measurement of
LAV by 2D echocardiography is based on geometric assumptions, which often results
in underestimation of LAV compared with that measured using cardiac MRI. 3D
echocardiography provides a more accurate measure of LAV because of automated
border detection, the acquisition of 3D data sets at different phases of the
cardiac cycle, and more accurate assessment of asymmetric remodeling of the left
atrium [139, 140, 141]. Badano et al.  revealed that LAD and area
measurements significantly underestimated actual LA size and misclassified the
grade of severity of LA dilatation in 43–70% of patients if 3D LAV was used as
the gold standard. In addition, 3D echocardiography provides unique measurement
of phasic changes of LAV during the cardiac cycle and detailed information of the
different LA functions . Marsan et al.  demonstrated that a
significant reduction of LAV
LA myocardial fibers are arranged not only in the longitudinal direction, and LA fibrosis may occur heterogeneously in patients with AF . Studies have shown that 3D-STE-derived circumferential, longitudinal, radial, as well as area strain are significantly reduced in patients with AF compared to matched controls [146, 147]. Because 2D-STE only provides longitudinal deformation information, some LA dysfunctions may be overlooked by 2D-STE.
3D-STE has the advantage of combining longitudinal and circumferential strain information , and a few studies have demonstrated the superiority of 3D-STE over 2D-STE in predicting AF occurrence or recurrence after CA [145, 147, 148]. Moreover, 3D-STE can be used to detect LA functional reverse remodeling by showing improvement of global strain and LA dyssynchrony . Theoretically, 3D-STE also has the advantage of overcoming the out-of-plane motion that may occur with 2D-STE, as the advent of 3D acquisition allows tracking of speckles in the myocardium in the 3D space . However, 3D echocardiography is limited by the slow temporal resolution and motion artifacts, and evaluation of the clinical utility of 3D-STE remains insufficient. Further studies are needed to clarify whether the diagnostic and prognostic value of 3D-STE is superior to that of 2D-STE .
Echocardiography is a safe and non-invasive technique providing quantitative
analyses of cardiac chamber size and function, but clinical measurement of the
left atrium has so far been limited to evaluation of LAD and LAV. Considerable
data support the use of LAEF to predict incident AF and AF recurrence after CA.
STE enables early detection of LA dysfunction before anatomical changes and also
helps identify patients with a severely dilated left atrium at risk for AF. The
studies discussed in this review support the contention that LAEF and LA strain
provide optimal diagnostic and prognostic information for assessing AF patients.
It is likely that future guidelines for patient evaluation and guidance of AF
ablation will include evaluation of not only LA chamber size but also LA function
parameters. Compared with cardiac MRI, echocardiography provides a real-time and
feasible method to assess LA function (LAVI
2D, 2-dimensional; 3D, 3-dimensional; AF, atrial fibrillation; AUC, area under the curve; CA, catheter ablation; CI, confidence interval; CNN, convolutional neural network; HFpEF, Heart failure with preserved ejection fraction; HFrEF, Heart failure with reduced ejection fraction; HR, hazard ratio; LA, left atrial; LAD, left atrial dimension; LAEF, left atrial emptying fraction; LAV, left atrial volume; LAVI, LAV index; LAScd, conduit left atrial strain; LASct, contractile left atrial strain; LASr, reservoir left atrial strain; LV, left ventricular; LVA, low voltage area; OR, odds ratio; PAF, paroxysmal atrial fibrillation; pLASRcd, peak conduit left atrial strain rate; pLASRct, peak contractile left atrial strain rate; pLASRr, peak reservoir left atrial strain rate; STE, speckle-tracking echocardiography; TACT, total atrial conduction time.
HTL and CCC designed the research study. HTL and HLL performed the research. All authors wrote the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.
Thanks to all the peer reviewers for their opinions and suggestions.
This research received no external funding.
The authors declare no conflict of interest.