Background
Published reference values for echocardiographic measurements of right-heart dimensions and function do not stratify for gender and body size. The aim of this study was therefore to assess the impact of gender and biometric characteristics on right-heart dimensions and function.
Methods
From the echocardiography database at a tertiary care center, 1,625 subjects (mean age, 44 ± 14 years; 47% men) with normal echocardiographic findings between 2000 and 2009 were identified. Gender differences and association with body surface area were assessed retrospectively for right atrial long-axis and short-axis dimensions, right ventricular short-axis dimension, end-diastolic and end-systolic right ventricular area, right ventricular fractional area change, and tricuspid annular plane systolic excursion. The impact of normal values stratified for gender and body surface area was tested in 24 patients with moderate-sized to large atrial septal defects.
Results
All dimensional right-heart measurements were significantly lower in women. Differences became smaller when measurements were indexed for body surface area, but significant differences persisted, particularly for right ventricular end-diastolic area (7.9 ± 1.6 vs 8.7 ± 1.8 cm 2 /m 2 , P < .001) and right ventricular end-systolic area (4.0 ± 1.2 vs 4.7 ± 1.4 cm 2 /m 2 , P < .001). Fractional area change and tricuspid annular plane systolic excursion indexed to body surface area were significantly higher in women (50 ± 7% vs 46 ± 9% and 14 ± 3 vs 12 ± 2 mm/m 2 , respectively, P < .001 for both comparisons). The use of upper reference ranges for end-diastolic right ventricular area stratified for gender and body surface area improved the detection of enlarged right ventricles in patients with moderate-sized to large atrial septal defects (92% vs 54%, P < .007).
Conclusions
Gender and body surface area are important determinants of right ventricular dimensions and systolic function as measured on two-dimensional echocardiography. The investigators thus propose the use of measurements indexed to body surface area, with upper and lower reference ranges stratified for gender.
Enlarged right ventricular dimensions are important clues for the diagnosis of some common forms of congenital heart lesions, such as atrial septal defects or partial anomalous pulmonary venous drainage, and are among the major diagnostic criteria for arrhythmogenic right ventricular cardiomyopathy. Recently published European guidelines for adults with congenital heart disease underscore the importance of right ventricular enlargement on echocardiography as the key factor for intervention in patients with atrial septal defects.
Simple measures of echocardiographic right ventricular systolic function, such as fractional area change (FAC) of the right ventricle and longitudinal shortening of the right ventricle, have proven to bear important prognostic information in a variety of acquired cardiac lesions, such as right-heart failure in the setting of pulmonary hypertension, or in patients with heart failure or coronary artery disease.
Guidelines for the echocardiographic assessment of the right heart have recently been published. These guidelines provide a set of normal values for commonly used dimensional and functional right ventricular measurements. Normal values are derived from pooled data from different studies and do not stratify for gender, age, and biometric data.
The aim of this study was to determine the impact of age, gender, and biometric data such as height, weight, and body surface area on commonly used measures of right-heart dimensions and function.
Methods
Study Population
Patients were identified from the echocardiography database of a tertiary care center. In this database, for each patient, the indication for the study, measured height and weight, and a predefined set of echocardiographic measurements are recorded. For patients with structurally and functionally normal hearts, the study results are coded as “normal” by a staff cardiologist. The criteria for classification of echocardiographic results as normal are sinus rhythm, absence of structural abnormalities, presence of no or minimal valvar regurgitation, no evidence of pulmonary hypertension, normal systolic and diastolic function (except for the presence of an abnormal left ventricular relaxation pattern on myocardial inflow in patients aged > 60 years), and the absence of pericardial effusion.
For the purposes of this study, all adult patients (age ≥ 18 years) with study results coded as normal between January 1, 2000, and December 31, 2009, were identified from the echocardiography database. The indications for all studies were reassessed, and patients with preexisting conditions that might influence right-heart dimensions or function were excluded. These exclusion criteria included disorders with potential lung involvement (chronic obstructive lung disease, connective tissue disorders, history of pulmonary hypertension), disorders with potential involvement of the myocardium (inherited myopathies, coronary artery disease), and disorders potentially leading to acute or chronic changes of hemodynamics (systemic infections, pregnancy, disorders typically associated with anemia or changes in cardiac output). Analysis of the data adhered to our institutional ethics board guidelines.
Echocardiographic Measurements
All patients underwent standard transthoracic echocardiographic examinations with commercially available equipment. For the purposes of this study, the following echocardiographic dimensional measurements were assessed retrospectively: right ventricular end-diastolic midventricular short-axis dimension in the apical four-chamber view, right ventricular end-diastolic and end-systolic area in the apical four-chamber view, and end-systolic right atrial long-axis and short-axis dimensions in the apical four-chamber view. Measures of right ventricular systolic function included right ventricular longitudinal shortening assessed by M-mode measurement of the lateral tricuspid valve annulus (tricuspid annular plane systolic excursion [TAPSE] in millimeters) and right ventricular FAC, calculated [(end-diastolic right ventricular area in cm 2 − end-systolic right ventricular area in cm 2 )/end-diastolic right ventricular area in cm 2 ] × 100. Measurements are illustrated in Figure 1 . Height and weight were measured at the time of echocardiography, and body surface area was calculated using the Mosteller formula.
Impact of Indexed Reference Values on the Detection of Enlarged Right Ventricles in a Group of Adults with Moderate-Sized to Large Atrial Septal Defects
To demonstrate the clinical impact of normalized versus nonnormalized right-heart measurements, we analyzed a group of patients with moderate-sized to large atrial septal defects. For this purpose, we identified 24 patients who underwent interventional closure of secundum-type atrial septal defects with large Amplatzer septal occluder devices (≥20 mm; St. Jude Medical, St. Paul, MN). Within this patient group, we tested whether upper reference ranges stratified for body surface area and gender would improve the detection of enlarged right ventricles compared with nonindexed reference ranges as proposed in the guidelines (end-diastolic right ventricular area in the four-chamber view > 25 cm 2 ).
Statistical Analysis
Statistical analysis was performed using SPSS version 19.0 (SPSS, Inc., Chicago, IL). Descriptive data for continuous variables are presented as mean ± SD and dichotomous variables as percentages. Histograms were used to confirm normal distributions of continuous variables. For comparisons between groups, Student’s t tests were used. Pearson’s correlation analysis was used to assess associations between right-heart dimensional and functional measurements, age, and biometric data. Lower and upper limits of normal were defined as mean ± 2 SDs. Two-sided P values < .05 were considered to be significant.
Results
Study Population
From a total of 46,367 echocardiographic studies in adult patients performed between 2000 and 2009, in 2,743 (6%), the results were coded as normal. Of these subjects, 1,118 were excluded from the present analysis because of potentially confounding preexisting disease. Reasons for exclusion were coronary artery disease in 168, lung disease in 97, connective tissue disorders in 211, inherited myopathies in 16, human immunodeficiency virus infection in 20, studies performed in patients in the intensive care unit in 201, acute leukemia or lymphoma in 278, known anemia in 66, and 61 studies in women who were pregnant.
The remaining 1,625 subjects (47% men; mean age, 44 ± 14 years; range, 18–92 years) were analyzed for the purposes of this study. The mean height was 171 ± 10 cm, the mean weight was 72 ± 15 kg, and the mean body surface area was 1.84 ± 0.22 m 2 . Indications for echocardiography in these subjects are shown in Table 1 . Not every measurement was available in every patient. Table 2 illustrates the number of available measurements for each of the analyzed echocardiographic parameters, ranging from 70% (right-heart end-diastolic area) to 92% (right atrial long-axis dimension), and the corresponding averages of age and biometric parameters for these groups.
Indication | Number of subjects (%) |
---|---|
Suspected heart disease | 873 (54) |
Cardiac murmur | 136 (8) |
Rule out structural heart disease | 720 (44) |
Nonspecific chest discomfort or dyspnea | 228 (14) |
Arrhythmia | 353 (22) |
Systemic hypertension | 139 (9) |
Abnormal cardiac appearance on chest radiography | 17 (1) |
Healthy individuals | 246 (15) |
Living organ donor | 141 (9) |
“Checkup” | 28 (2) |
Screening for familial diseases | 77 (5) |
Rule out cardiac source of embolism | 210 (13) |
Assessment before potentially cardiotoxic drug treatment | 209 (13) |
Others | 87 (5) |
Parameter | Number of subjects | Men (%) | Average age (y) | Average height (cm) | Average weight (kg) | Average body surface area (m 2 ) |
---|---|---|---|---|---|---|
Right atrial long axis | 1,496 | 47 | 43 ± 14 | 171 ± 10 | 72 ± 15 | 1.84 ± 0.22 |
Right atrial short axis | 1,473 | 47 | 43 ± 14 | 171 ± 10 | 72 ± 15 | 1.84 ± 0.22 |
Right ventricular end-diastolic short axis | 1,235 | 47 | 43 ± 14 | 171 ± 10 | 71 ± 15 | 1.84 ± 0.22 |
Right ventricular end-diastolic area | 1,145 | 47 | 43 ± 14 | 171 ± 10 | 71 ± 15 | 1.83 ± 0.22 |
Right ventricular end-systolic area | 1,133 | 47 | 43 ± 14 | 171 ± 10 | 71 ± 15 | 1.83 ± 0.21 |
Right ventricular FAC | 1,133 | 47 | 43 ± 14 | 171 ± 10 | 71 ± 15 | 1.83 ± 0.21 |
TAPSE | 1,144 | 47 | 45 ± 15 | 171 ± 9 | 73 ± 15 | 1.86 ± 0.22 |
Correlation of Right-Heart Measurements with Age
There were no statistically significant correlations between age and right ventricular systolic function (FAC and TAPSE), right atrial short-axis dimension, right ventricular short-axis dimension, and right ventricular systolic area ( r = −0.05 to 0.01, P > .10 for all). There was a weak, although statistically significant, positive correlation between age and right atrial long-axis dimension and a weak negative correlation between age and right ventricular end-diastolic area ( r = 0.09 and −0.09 respectively, P = .001 and .004, respectively; Figure 2 ). These weak correlations are not of clinical importance.
Gender Differences in Right-Heart Dimensions and Function
Right atrial and right ventricular dimensions were all significantly larger in men, whereas FAC was significantly higher in women and no statistically significant difference was found for TAPSE. Figure 3 illustrates gender differences for these measurements. The most striking differences were found for measurements of right ventricular end-diastolic and end-systolic areas, for which the absolute differences between means for genders were 3.4 and 2.3 cm 2 , respectively, accounting for 25% and 34%, respectively, of absolute values.
Correlation between Biometric Data and Right-Heart Dimensions and Systolic Function
Men were on average significantly taller (178 ± 8 vs 165 ± 7 cm, P < .0001) and heavier (79 ± 14 vs 66 ± 13 kg, P < .0001) than women and had corresponding higher body surface areas (1.97 ± 0.19 vs 1.73 ± 0.18 m 2 , P < .0001) and higher body mass indexes (25.2 ± 4.6 vs 24.2 ± 4.9 kg/m 2 , P < .0001).
Moderate correlation for all dimensional right-heart measurements with all biometric characteristics was found, with the strongest correlations for body surface area ( Table 3 , Figure 4 ). Correlations of TAPSE and FAC with biometric data were relatively weak ( Table 3 , Figure 4 ).
Echocardiographic parameter | Height | Weight | Body surface area | Body mass index |
---|---|---|---|---|
Right atrial long axis | 0.30 ∗ | 0.39 ∗ | 0.42 ∗ | 0.24 ∗ |
Right atrial short axis | 0.35 ∗ | 0.30 ∗ | 0.35 ∗ | 0.11 ∗ |
Right ventricular end-diastolic short axis | 0.32 ∗ | 0.32 ∗ | 0.35 ∗ | 0.16 ∗ |
Right ventricular end-diastolic area | 0.42 ∗ | 0.42 ∗ | 0.47 ∗ | 0.20 ∗ |
Right ventricular end-systolic area | 0.37 ∗ | 0.38 ∗ | 0.42 ∗ | 0.19 ∗ |
Right ventricular FAC | −0.13 ∗ | −0.16 ∗ | −0.17 ∗ | −0.09 ∗ |
TAPSE | 0.07 † | 0.09 ∗ | 0.09 ∗ | 0.06 |
To demonstrate the potential clinical significance of correlations of dimensional right-heart measurements with biometric data, Figure 5 demonstrates the differences for all dimensional right-heart measurements between subjects in the lowest and the highest quartiles of body surface area.
Gender Differences for Right-Heart Dimensions and Function When Normalized to Body Surface Area
When right-heart dimensions were normalized to body surface area, right atrial short-axis and right ventricular short-axis dimensions were no longer significantly different between genders. However, even when normalized for body surface area, a significant difference persisted for right atrial long-axis dimensions and particularly for measurements of end-diastolic and end-systolic right ventricular areas in the four-chamber view ( Table 4 ). Although the differences between genders became smaller when values were indexed for body surface area, average differences between genders were still 6% to 18%. Interestingly, indexed TAPSE became significantly higher in women, which fits with the observation that FAC, another parameter of systolic right ventricular function, was significantly higher in women.
Echocardiographic parameter indexed to body surface area | Men | Women | P | Difference between means (%) |
---|---|---|---|---|
Right atrial long axis (cm/m 2 ) | 2.4 ± 0.3 | 2.5 ± 0.3 | <.0001 | 6 |
Right atrial short axis (cm/m 2 ) | 1.9 ± 0.3 | 1.9 ± 0.3 | .90 | 0 |
Right ventricular end-diastolic short axis (cm/m 2 ) | 1.5 ± 0.3 | 1.5 ± 0.3 | .90 | 0 |
Right ventricular end-diastolic area (cm 2 /m 2 ) | 8.7 ± 1.8 | 7.9 ± 1.6 | <.0001 | 10 |
Right ventricular end-systolic area (cm 2 /m 2 ) | 4.7 ± 1.4 | 4.0 ± 1.2 | <.0001 | 18 |
TAPSE (mm/m 2 ) | 12.0 ± 2.3 | 13.5 ± 2.5 | <.0001 | 13 |