The incremental value of left atrial (LA) deformation analysis by speckle tracking echocardiography compared with LA volume or LA ejection fraction as a cardiovascular risk marker has not been evaluated prospectively. We sought to compare LA function by speckle tracking echocardiography to other conventional LA parameters for prediction of adverse cardiovascular outcomes. This prospective study included 312 adults (mean age 71 ± 6 years, 56% men) in sinus rhythm who were followed for development of first atrial fibrillation, congestive heart failure, stroke, transient ischemic attack, myocardial infarction, coronary revascularization, and cardiovascular death. Global peak atrial longitudinal strain (PALS) by speckle tracking echocardiography was measured in all subjects by averaging all atrial segments. Left atrium was assessed with biplane LA volume, LA ejection fraction, 4-chamber LA area, and M-mode dimension. Of 312 subjects at baseline, 43 had 61 new events during a mean follow-up of 3.1 ± 1.4 years. All LA parameters, traditional parameters, and parameters derived by speckle tracking echocardiography were independently predictive of combined outcomes (p <0.0001 for all comparisons). Overall performance for prediction of cardiovascular events was greatest for global PALS (area under receiver operator characteristic curve: global PALS 0.83, indexed LA volume 0.71, LA ejection fraction 0.69, LA area 0.64, LA diameter 0.59). A graded association between degree of LA enlargement and risk of cardiovascular events was evident only for global PALS and indexed LA volume. In conclusion, global PALS is a strong and independent predictor of cardiovascular events and appears to be superior to conventional parameters of LA analysis.
Recent population-based studies have demonstrated the prognostic value of left atrial (LA) analysis for long-term outcome. Atrial longitudinal strain, deriving from application of the analysis of myocardial deformation using speckle tracking echocardiography at the LA chamber, is the first parameter useful for functional analysis of the left atrium and, as shown by recent studies, presents considerable feasibility and reproducibility. In this prospective study we compared for the first time the utility of LA longitudinal strain, LA volume, ejection fraction, area, and dimension for prediction of age-related cardiovascular outcomes.
Methods
Patients >50 years old referred for a general medical consultation were invited to participate. Exclusion criteria included a history or evidence of atrial arrhythmia, history of congenital heart disease, treatment with pacemaker implantation, valvular surgery, cardiac transplantation, or poor echocardiographic window. All subjects gave their written informed consent for participation in the study. All work was in compliance with the Declaration of Helsinki.
Baseline clinical data were obtained from a comprehensive interview of each patient at enrollment. All patients were prospectively followed for development of new outcome events, which included atrial fibrillation (AF), stroke, transient ischemic attack, myocardial infarction (MI), coronary revascularization, congestive heart failure, and cardiovascular death. Criteria defining outcomes of interest were specified a priori. Type and date of detection of each outcome were abstracted. AF was defined as the presence of clinically documented and electrocardiographically confirmed irregular rhythm with disorganized atrial activity and without discrete P waves. Stroke in this study included development of any type of stroke as defined by clinical documentation of the diagnosis with confirmatory findings on imaging studies. Transient ischemic attack was defined by clinical documentation of the diagnosis in the absence of other conditions that could be responsible for the neurologic symptoms. Presence of MI was considered present if 2 of 3 diagnostic criteria were fulfilled: compatible clinical presentation, diagnostic cardiac enzyme levels, and electrocardiographic changes consistent with MI. History of coronary artery disease was defined as typical symptoms or previous abnormal functional test result, history of MI, or previous coronary angiography demonstrating at least moderate coronary disease (50% stenosis in ≥1 vessel).
Coronary revascularization referred to coronary artery bypass grafting or percutaneous coronary intervention. Congestive heart failure was defined according to Framingham criteria requiring the presence of 2 major or 1 major and 2 minor Framingham criteria. Hypercholesterolemia was defined as use of cholesterol-lowering medication or, in the absence of cholesterol-lowering medication use, as having a total serum cholesterol >240 mg/dl.
Echocardiographic studies were performed using a high-quality echocardiograph (Vivid 7, GE, Milwaukee, Wisconsin) equipped with a 2.5-MHz transducer. Subjects were studied in the left lateral recumbent position. Measurements of left ventricular (LV) and LA dimensions, LV ejection fraction, LA ejection fraction, and diastolic LV filling velocities were made in accordance with current recommendations of the American Society of Echocardiography. LV ejection fraction measured using the Simpson method was used as a standard index of global LV systolic function. LV diastolic function was classified by Doppler assessment of mitral inflow, pulmonary venous flow, and tissue Doppler imaging of septal and lateral mitral annular motions. Grading of diastolic dysfunction was as follows: grade 1, impaired relaxation; grade 2, pseudonormal pattern; and grade 3/4, reversible/irreversible restrictive pattern. LA volumes were measured using the area–length method from apical 4- and 2-chamber views. LA volumes were subsequently indexed by body surface area.
For speckle tracking analysis, apical 4- and 2-chamber views were obtained using conventional 2-dimensional gray-scale echocardiography, during breath hold, and with a stable electrocardiographic recording. Three consecutive heart cycles were recorded and averaged. Frame rate was set to 60 to 80 frames/s.
Analysis of files recorded was performed offline by a single experienced and independent echocardiographer who was not directly involved in image acquisition and had no knowledge of other clinical and echocardiographic parameters representing LV, LA, and valvular structure and function using a commercially available semiautomated 2-dimensional strain software (EchoPac, GE). As previously described and as stated in the current American Society of Echocardiography/European Association of Echocardiography consensus, the LA endocardial border was manually traced in the 4- and 2-chamber views, thus delineating a region of interest composed by 6 segments. Then, after segmental tracking quality analysis and eventual manual adjustment of the region of interest, longitudinal strain curves are generated by the software for each atrial segment. As shown in Figure 1 , peak atrial longitudinal strain (PALS) measured at the end of the reservoir phase was calculated by averaging values observed in all LA segments (global PALS) and by separately averaging values observed in 4- and 2-chamber views (4- and 2-chamber average PALSs, respectively). Time to peak longitudinal strain was also measured as the average of all 12 segments (global time to peak longitudinal strain) and by separately averaging values observed in the 2 apical views (4- and 2-chamber average times to peak longitudinal strain). To assess the reproducibility of global PALS measurements, 30 patients were randomly selected; Bland–Altman analysis was performed to evaluate intra- and interobserver agreements.
Data are shown as mean ± SD. A p value <0.05 was considered statistically significant. Differences between group means were evaluated with t tests (continuous variables) or chi-square analyses (categorical variables), as appropriate. Receiver operator characteristic curves were generated to assess the overall performance of the various LA parameters for prediction of cardiovascular events. Differences in rates of cardiovascular events by categories of LA parameters were examined with log-rank tests. Categories of global PALS were obtained through measurements of percentiles. Cox proportional hazards modeling was used to determine the association of global PALS, LA volume, LA ejection fraction, LA area, and LA dimension to future cardiovascular events after adjusting for age, gender, and other significant covariates. Analyses were performed using SPSS 12.0 (SPSS, Inc., Chicago, Illinois).
Results
Of 357 patients screened, 312 patients (mean age 71 ± 8 years, 56% men) met eligibility criteria. Admitting indications for medical consultation included chest pain or discomfort (27%), dyspnea (25%); palpitations, presyncope, and/or syncope (19%); and other reasons (29%). Twenty-one patients were excluded for nonsinus rhythm or history of atrial arrhythmias, 8 for history of pacemaker implantation, 7 for history of valvular surgery, 2 for cardiac transplantation, and 7 for poor echocardiographic window. Tables 1 and 2 presents clinical and standard echocardiographic characteristics of the study population stratified for the presence of cardiovascular events. According to American Society of Echocardiography guidelines and to percentile subdivision of global PALS, Table 3 presents the distribution of degree of abnormality of LA size and function parameters in the study population.
| Variable | Cardiovascular Events | |
|---|---|---|
| No | Yes | |
| (n = 269) | (n = 43) | |
| Clinical data | ||
| Age (years) | 70.9 ± 8.1 | 66.7 ± 8.6 |
| Women | 43.0% | 42.5% |
| Body mass index (kg/m 2 ) | 25.9 ± 4.2 | 25.7 ± 5.2 |
| Heart rate (beat/min) | 71.1 ± 8.8 | 73.9 ± 9.1 |
| Systolic blood pressure (mm Hg) | 138.6 ± 8.2 | 141.2 ± 9.1 |
| Diastolic blood pressure (mm Hg) | 77.9 ± 6.1 | 78.3 ± 6.4 |
| Hypertension | 164 (61%) | 31 (72%) |
| Diabetes mellitus | 65 (24%) | 12 (29%) |
| Hypercholesterolemia | 172 (64%) | 32 (75%) |
| Current smoker | 32 (12%) | 6 (15%) |
| Myocardial infarction | 32 (12%) | 12 (29%) |
| History of coronary artery disease | 92 (25%) | 20 (46%) |
| Previous stroke | 13 (5%) | 3 (6%) |
| Previous transient ischemic attack | 19 (7%) | 3 (8%) |
| Previous congestive heart failure | 29 (11%) | 11 (25%) |
| Medical therapy | ||
| Angiotensin-converting enzyme inhibitors or angiotensin receptor blocker | 121 (44.9%) | 20 (46.5%) |
| β Blockers | 63 (23.4%) | 11 (25.5%) |
| Calcium antagonists | 64 (23.8%) | 12 (27.9%) |
| Spironolactone | 41 (15.2%) | 9 (20.9%) |
| Loop diuretics | 86 (31.9%) | 16 (37.2%) |
| Statins | 159 (59.1%) | 27 (62.7%) |
| Platelet aggregation inhibitors | 110 (40.9%) | 25 (58.1%) |
| Variable | Cardiovascular Events | |
|---|---|---|
| No | Yes | |
| (n = 269) | (n = 43) | |
| Echocardiographic data | ||
| Left atrial diameter (mm) | 35.6 ± 7.8 | 41.5 ± 8.2 |
| Left atrial area (cm 2 ) | 17.9 ± 5.9 | 20.6 ± 6.6 |
| Left atrial volume indexed (ml/m 2 ) | 34.2 ± 9.1 | 42.1 ± 10.1 |
| Left atrial ejection fraction (%) | 65.6 ± 7.1 | 55.6 ± 7.0 |
| Left ventricular end-diastolic diameter (mm) | 46.3 ± 8.3 | 50.3 ± 8.0 |
| Left ventricular mass index (g/m 2 ) | 106.6 ± 21.3 | 116.3 ± 23.5 |
| Left ventricular ejection fraction (%) | 55.7 ± 7.5 | 54.1 ± 8.3 |
| Mitral E (cm/s) | 65.2 ± 30 | 68.6 ± 31 |
| Mitral E/A ratio | 0.95 ± 0.2 | 1.07 ± 0.3 |
| Any diastolic dysfunction | 251 (93.3%) | 40 (93.0%) |
| Grade 2–4 diastolic dysfunction | 78 (29.0%) | 15 (34.9%) |
| E/E′ (cm/s) | 7.6 ± 3.4 | 8.9 ± 4.2 |
| Global peak atrial longitudinal strain (%) | 35.4 ± 9.5 | 20.7 ± 9.8 |
| 4-Chamber peak atrial longitudinal strain (%) | 33.2 ± 7.1 | 19.8 ± 9.6 |
| 2-Chamber peak atrial longitudinal strain (%) | 37.6 ± 10.2 | 22.1 ± 9.9 |
| Global time to peak atrial longitudinal strain (ms) | 395 ± 75 | 409 ± 85 |
| LA M-Mode Diameter | LA Area | LA Indexed Volume | LA Ejection Fraction | Global PALS | |
|---|---|---|---|---|---|
| Normal | 64% | 36% | 18% | 20% | 17% |
| Mildly abnormal | 21% | 51% | 31% | 33% | 28% |
| Moderately abnormal | 10% | 9% | 20% | 22% | 26% |
| Severely abnormal | 5% | 4% | 31% | 25% | 29% |
Over a mean follow-up period of 3.1 ± 1.4 years, 43 patients had 61 new events (16 new AF, 4 strokes, 3 transient ischemic attacks, 9 MIs, 11 developed congestive heart failure, 15 coronary revascularizations, and 3 cardiovascular deaths). Of the 4 strokes recorded, 2 occurred in patients in sinus rhythm and the other 2 occurred after onset of AF (74 ± 34 days after onset of AF). Patients who developed cardiovascular events had decreased global PALS, decreased LA ejection fraction, and larger left atrium at baseline irrespective of method of quantitation. To further investigate the value of these conventional and innovative echocardiographic indexes to predict cardiovascular events, we performed receiving operating characteristics curve analyses. Among all LA parameters analyzed, global PALS showed the highest diagnostic accuracy (area under curve 0.83) to predict cardiovascular events using a cut-off value <−19% ( Figure 2 and Table 4 ). This trend was evident even if coronary revascularization was excluded as an end point (area under curve for global PALS 0.86 vs 0.83).
| Cutoff Value | Sensitivity (%), (95% CI) | Specificity (%), (95% CI) | Area Under Curve | |
|---|---|---|---|---|
| Left atrial diameter (mm) | >44.6 | 59.2 (53.0–87.4) | 61.1 (26.3–76.5) | 0.61 |
| Left atrial area (cm 2 ) | >26.3 | 62.5 (53.6–88.1) | 66.2 (41.3–82.6) | 0.66 |
| Left atrial ejection fraction (%) | <37.6 | 65.6 (52.3–81.6) | 78.4 (68.2–88.1) | 0.71 |
| Left atrial indexed volume (ml) | >48.2 | 69.1 (55.6–84.2) | 80.9 (75.1–91.6) | 0.72 |
| Global peak atrial longitudinal strain (%) | <18.8 | 78.2 (69.6–92.6) | 85.2 (77.9–97.9) | 0.83 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
