Background
The 2015 American Society of Echocardiography/European Association of Cardiovascular Imaging recommendations for chamber quantification suggest new abnormality threshold and severity partition values for left atrial (LA) volume that are equally valid for the biplane method of disk (MOD) summation and the area-length method (ALM). However, they have never been clinically validated. Thus, we compared the clinical and prognostic impact of LA volume assessed by MOD and ALM by using both the 2015 and 2005 abnormality thresholds.
Methods
In a retrospective study of 467 patients with sinus rhythm and various cardiac conditions (median age 61 years, 68% men), maximal LA volumes were measured with MOD and ALM. Patients were followed for 3.7 ± 1.1 years to record both all-cause mortality and cardiac death.
Results
Applying the 2015 cutoff values, 21% of patients with dilated LA according to the 2005 recommendations were reclassified as normal. Severity of LA dilatation was reclassified in 48% (222/467) patients. ALM provided significantly larger LA volumes than MOD (41 [32; 58] mL/m 2 vs 39 [30; 55] mL/m 2 ; P = .0150), reclassifying 18% (84/467) of patients. Patients who died had larger LA volumes measured with both MOD (57 [38; 77] mL/m 2 vs 37 [30; 51] mL/m 2 ; P < .0001) and ALM (58 [40; 82] mL/m 2 vs 40 [32; 54] mL/m 2 ; P < .0001). Regardless of the method used, LA volume was a significant factor associated with mortality, with both the 2015 and 2005 cutoff values providing similar prognostic power.
Conclusions
The use of 2015 partition values and different methods of LA volume measurement leads to significant changes in patients’ clinical profiles. LA enlargement is an important prognostic indicator independent of cutoff values and methods used. Care should be taken to ensure consistent measurements and interpretation of two-dimensional echocardiography LA volume during patient follow-up.
Highlights
- •
The 2015 recommendations proposed a higher abnormality threshold for left atrial volume.
- •
This led to reclassification of 48% of patients according to the left atrial size.
- •
The new and previous abnormality thresholds maintained similar prognostic power.
- •
The area-length method provided larger left atrial volumes than method of disks.
- •
It reclassified 18% of patients but did not influence prognostic significance.
Left atrial (LA) enlargement is an important risk factor and a powerful independent predictor of adverse cardiovascular events in various clinical scenarios, such as acute and chronic coronary artery disease, dilated cardiomyopathy, heart failure, diabetes mellitus, cohorts of unselected patients, and healthy volunteers.
Using the results of large normative studies, the recently updated recommendations for cardiac chamber quantification issued jointly by the American Society of Echocardiography (ASE) and the European Association of Cardiovascular Imaging (EACVI) proposed new abnormality threshold and severity partition values for LA maximum volume measured with two-dimensional echocardiography (2DE), which were significantly larger than those published in 2005. Applying the higher abnormality threshold for LA volume index (34 vs 28 mL/m 2 ) is expected to lead to the reclassification of patients previously diagnosed with mildly or even moderately dilated LA as having normal LA. However, the actual extent of this change as well as its clinical and prognostic impact remain unknown.
Moreover, the coexistence of two different algorithms for the routine calculation of biplane LA maximum volume, such as the method of disks (MOD; Simpson disks summation method) and the area-length method (ALM), which may yield slightly different measurements in the same patient, adds to the complexity of the LA quantification by 2DE. Despite the fact that some studies reported slightly larger LA volumes when measured by ALM in comparison with MOD, there is a lack of method-specific reference values and evidence that these methods can actually be used interchangeably.
Accordingly, the present study was designed to (1) determine the clinical impact of different reference values and severity partition values (recommended by ASE/European Association of Echocardiography [EAE] in 2005 and ASE/EACVI in 2015) and methods of LA volume measurement (MOD and ALM) on diagnosis and reclassification of severity of LA enlargement in a cohort of patients with structural heart diseases undergoing clinically indicated routine 2DE; and (2) compare the prognostic value of LA enlargement assessed by MOD versus ALM and using 2015 versus 2005 abnormality thresholds and severity cutoff values.
Methods
Study Design and Population
Between October 2010 and December 2012, patients with sinus rhythm and various cardiac conditions who underwent clinically indicated transthoracic echocardiography were enrolled into a single-center retrospective observational cross-sectional study aimed to assess the reference values and the prognostic power of three-dimensional echocardiography LA volumes. For the present retrospective analysis, the following inclusion criteria were used: (1) recordings of both 2DE apical four- and two-chamber views allowing for LA quantification, (2) sufficient image quality to trace LA endocardium; (3) availability of follow-up data. Patients with more than 5 mm difference between LA lengths in the two apical views were excluded ( Figure 1 ). Demographic and clinical data (age, weight, height, body surface area, body mass index, cardiovascular risk factors, and comorbidities) were retrieved from the electronic clinical records of the hospital database. The study was approved by the local Ethics Committee.
Echocardiographic Analysis
Digitally stored data sets in raw-data format were analyzed offline using commercially available software (EchoPAC BT12, GE Vingmed, Horten, Norway). All measurements were performed according to the ASE/EACVI guidelines.
Quantitative analysis of 2DE-derived LA volume was performed by a single experienced echocardiographer (E.S.) unaware of either clinical information or outcomes. LA maximal volume was measured at the end of ventricular systole (the last frame before the mitral valve opening) in the apical four- and two-chamber views. While tracing the LA endocardium, the LA appendage and the ostia of pulmonary veins were excluded. When dropout was encountered, a straight line joining the closest visualized structures completed the atrial outline. Mitral annulus was considered as LA atrioventricular border. The length of LA was measured from the middle of the mitral annular plane to the posterior wall ( Figure 2 ). The software package was set in order to calculate the LA maximal volume by both MOD and ALM using the same endocardial tracings. LA volumes were indexed to body surface area. Indexed maximal LA volumes were classified as normal, mildly, moderately, or severely dilated according to the abnormality thresholds and severity partition values recommended by both ASE/EAE in 2005 and ASE/EACVI in 2015 ( Table 1 ).
π / 4 ( h ) ∑ ( D 1 ) ( D 2 ) ,
where h is the height of the disks and D 1 and D 2 are orthogonal minor and major transverse axes of each disk. (2) ALM, applying the following equation: <SPAN role=presentation tabIndex=0 id=MathJax-Element-2-Frame class=MathJax style="POSITION: relative" data-mathml='8/3π⋅[(A1⋅A2)/L],’>8/3π⋅[(A1⋅A2)/L],8/3π⋅[(A1⋅A2)/L],
8 / 3 π ⋅ [ ( A 1 ⋅ A 2 ) / L ] ,
where A 1 and A 2 are the LA areas measured in the apical two- and four-chamber views and L is the shortest of the two long axes. According to the 2015 ASE/EACVI recommendations, the LA volume index (body surface area = 1.58 m 2 ) should be classified as normal, when measured by MOD, and mildly dilated, when measured by ALM. Applying the previous cutoff values from 2005, LA is classified as moderately dilated. 2Ch , two-chamber view; 4Ch , four-chamber view; LAV , LA volume; LAVi , LA volume indexed to body surface area.
| Chamber quantification guidelines | LA volume index (mL/m 2 ) | |||
|---|---|---|---|---|
| Normal | Mildly abnormal | Moderately abnormal | Severely abnormal | |
| Lang et al , 2005 | 16-28 | 29-33 | 34-39 | ≥40 |
| Lang et al , 2015 | 16-34 | 35-41 | 42-48 | >48 |
In addition, left ventricular size and function parameters were obtained from the analysis of two-dimensional and three-dimensional echocardiography data sets of the left ventricle, as described elsewhere.
Reproducibility Analysis
Intraobserver variability was tested on 45 good-quality apical four- and two-chamber images by an experienced echocardiographer (E.S.) who reanalyzed the same recordings one more time, being blinded to the first measurements. For interobserver variability, the same images were analyzed independently by a different experienced researcher (J.B.).
Follow-up
All patients were followed for occurrence of new outcome events, which included all-cause mortality and cardiac death (defined as death resulting from an acute myocardial infarction, heart failure, cardiovascular procedures, and sudden cardiac death). Follow-up data were collected by an investigator (D.G.), who was not involved in the echocardiographic measurements, through analysis of clinical records and telephone contacts to patients, physicians, or the next of kin when the patient was not available.
Statistical Analysis
The Kolmogorov-Smirnov test was used to verify the normal distribution of variables. Continuous variables were reported as the mean and SD for normally distributed variables, while nonnormally distributed variables were reported as the median and interquartile range. Statistical significance was tested using a χ 2 test for categorical variables. T -test or Mann-Whitney test was used to test the differences in the continuous variables according to the data distribution. Receiver-operator characteristic curves were generated to assess the overall performance and predictive value of the different cutoffs and methods of LA volume assessment. Kaplan-Meier survival analysis was used to plot all-cause mortality and cardiac death. Differences between survival curves were assessed by the log-rank test. Statistical analysis was performed using R (ver. 3.3.1) and GraphPad Prism (ver. 7.0a). A P value of <.05 was considered statistically significant.
Results
A total of 467 patients were included into the study cohort. Demographic and clinical characteristics and echocardiographic data of the study patients are summarized in Table 2 .
| Variable | Overall ( N = 467) |
|---|---|
| Age, years | 61 (46; 72) |
| Gender, male | 316 (67.7) |
| Height, m | 1.70 (1.63; 1.76) |
| Weight, kg | 73 (63; 83) |
| Body surface area, m 2 | 1.84 (1.70; 1.98) |
| Body mass index, kg/m 2 | 24.9 (22.8; 27.6) |
| Heart rate, bmp | 68 (60; 76) |
| Systolic blood pressure, mm Hg | 120 (110; 140) |
| Diastolic blood pressure, mm Hg | 75 (70; 80) |
| Sinus rhythm | 467 (100) |
| Smoking | 170 (36.4) |
| Diabetes | 67 (14.3) |
| Dyslipidemia | 196 (42.0) |
| Family history of CAD | 111 (23.8) |
| Hypertension | 262 (56.1) |
| CAD | 163 (34.9) |
| LA volume, MOD biplane, mL | 71 (55; 100) ∗ |
| LA volume, ALM biplane, mL | 76 (58; 106) ∗ |
| LA volume index, MOD, mL/m 2 | 38.55 (30.11; 55.15) † |
| LA volume index, ALM, mL/m 2 | 40.77 (32.33; 57.66) † |
| 3DE LV end-diastolic volume index, mL/m 2 | 75 (62; 94) |
| 3DE LV end-systolic volume index, mL/m 2 | 33 (25; 51) |
| 3DE LV ejection fraction, % | 55 (43; 62) |
| 3DE LV mass, g | 180.5 (146; 210) |
| 2DE LV global longitudinal strain, % | −15.7 (−18.95; −11.8) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
