Although often referred to as “the forgotten chamber,” the right ventricle plays a critical role in the clinical expression and prognosis of patients with congenital heart disease, left ventricular (LV) dysfunction, and pulmonary hypertension in addition to primary disorders of the right heart. One reason for this oversight historically is that echocardiographic imaging of the right ventricle is demanding. Specifically, the thin walls and coarse trabeculations of the right ventricle handicap endocardial border detection, the anterior position in the chest may challenge the limited near-field resolution of echocardiography, and the chamber’s complex three-dimensional (3D) structure (crescentic in the coronal plane and triangular in the sagittal plane) with distinct inflow and outflow tracts complicates geometry-dependent calculation of right ventricular (RV) volumes and mass. In addition, partly because of differences in embryology, workload, and the impedance of their respective downstream vascular beds, the right ventricle is physiologically distinct from the left ventricle; this is manifested by a predominance of longitudinally arranged myofibers, “peristaltic-like” contraction propelling blood from the inflow to the outflow tract, less torsional contribution to contraction, and negligible isovolumic intervals, all of which may influence estimates of RV structure and function.

RV pressure and volume overload and intrinsic myocardial dysfunction often present with RV enlargement, and therefore accurate measurement of RV volume is essential in the management of a wide variety of cardiovascular diseases. Cardiac magnetic resonance (CMR) is considered the reference standard for quantifying RV volumes, but it is relatively expensive, is not widely available, and requires optimization of image acquisition and consistency in postprocessing. Echocardiography is therefore the imaging modality of choice for screening and serially following patients with disease involving the right ventricle. Quantitative two-dimensional (2D) echocardiographic measurements increase accuracy (compared with CMR) and interreader agreement compared with qualitative (i.e., biretinal algorithm) estimates. However, because of the right ventricle’s complex 3D geometry, standardized 2D echocardiographic measurements of RV dimensions correlate poorly with CMR RV volume. In contrast, 3D echocardiographic volumes measured using either the method of disks or semiautomated border detection correlate with (but underestimate) CMR RV volume, which is often indexed to body surface area (BSA). Real-time 3D echocardiography will likely become the method of choice for calculating RV volume and ejection fraction, and accordingly, it is pertinent to ask whether there is a need to develop normal reference values for indexed RV dimensions and volume when measured using echocardiography.

There is considerable evidence that cardiovascular structure and function scale with body size, and body size may account for up to half of the variability of LV dimensions in adults. In addition, the influence of body size on cardiovascular structure and function contributes to the overlap of chamber measurements between disease states and normal variants. This ambiguity is particularly noteworthy in athletes, in whom physiologic hypertrophy may mimic disease. Indexing may resolve these diagnostic uncertainties and thereby facilitate clinical decision making. Although the American Society of Echocardiography recommends that echocardiographic LV dimensions and mass and left atrial volume be indexed to BSA, indexed RV dimensions are conspicuously absent in the most recently published guidelines of normal values for the right heart. In a study of 102 endurance athletes, RV dimensions exceeded these guideline normal values in up to 69% of the subjects. In that study, 28% of the absolute RV outflow tract dimensions fulfilled major criteria for arrhythmogenic RV cardiomyopathy, which when indexed to BSA was reduced to 6%. Finally, consider that unlike their adult professional counterparts, pediatricians routinely index RV measurements to body size, because the variability owing to body growth supports this approach in children. However, the variability in body size in adults is considerable and itself is sufficient to justify scaling these measurements.

Left-heart measurements may also vary by age, gender, and race or ethnicity. Although men have larger LV volumes and mass and lower ejection fractions than women, the effect of age is inconsistent. The influences of age and gender on right-heart measurements have also been demonstrated in several small studies, although the results (lower RV mass with age and in women) are less reliable. Most recently, Kawut et al. measured RV volumes and mass in 4,204 normal, nonobese participants of the Multi-Ethnic Study of Atherosclerosis (MESA) using CMR and found body size–independent associations of RV volume, mass, and ejection fraction with age, sex, and race. Similar normative data in the echocardiographic literature have been lacking. Three studies in the current issue of JASE attempt to fill this void.

In the first, D’Oronzio et al. retrospectively studied the influence of age, gender, and body size on measurements of RV size (RV end-diastolic and end-systolic area, RV short-axis dimension, right atrial long and short axes) and function (tricuspid annular plane systolic excursion and fractional area change) in 1,625 patients (predominantly Caucasian) with normal results on echocardiography. Gender and BSA, but not age, were significant determinants of RV size and function. Thus, absolute RV dimensions were less and fractional shortening was greater in women; gender differences were smaller when RV variables were indexed to BSA, although differences remained, with average differences ranging between 6% and 18%. The potential clinical impact of indexing was cleverly demonstrated by comparing absolute and indexed RV end-diastolic areas in a group of 24 patients with percutaneously closed moderate-sized to large atrial septal defects (a group with prima facie RV enlargement). Using upper reference ranges indexed to BSA and gender, enlarged right ventricles were identified in 92% of patients, whereas using the guideline-recommended nonindexed reference range, only 54% of patients were correctly identified. As the authors note, the sample population was homogenous, measurement variability was not determined, and it is likely that individuals with RV dimensions that were considered abnormal by staff cardiologists were not included in the sample, resulting in a bias toward smaller upper limits of normal. Nevertheless, D’Oronzio et al. provide a rich resource of gender-stratified, indexed right-heart parameters of structure and function.

In the second study, Willis et al. established reference ranges of absolute and normalized RV dimensions (10 2D caliper and 2D areas) from 205 prospectively studied healthy volunteers stratified by gender, age, and ethnicity (European, Indian, Chinese, and Malay). Gender, BSA, and ethnicity were found to have significant impacts on RV dimensions, whereas the effects of age were minimal. Thus, RV dimensions and biometric measures were larger in men and Europeans; when indexed to BSA, most dimensions (notably not RV end-diastolic area) were greater in women than in men. However, the influence of ethnicity on absolute and especially indexed RV parameters was minimal and inconsistent and therefore not included in the gender-specific reference ranges. Although there are inconsistencies in the values for the linear measurements between studies, it is reassuring that the reference ranges of indexed RV end-diastolic area for men and women are similar to those reported by D’Oronzio et al. Intraobserver and interobserver differences and test-retest variability were dependent on the measurement site and, except for wall thickness, midventricular minor axis (referred to by Willis et al. as “RVD-2”), and RV end-systolic area, were good to excellent. Although one may quibble about the purity of the self-declared ethnicities, the precise sites of measurements, and the relatively small size of the sample population, the study confirms the need for gender-stratified indexed RV dimensions.

Although these two studies support the need to index, an unsettled and critical issue is how these measurements should be indexed. Most scaling methods are ratiometric; that is, they divide the scaled variable by a measure of body size (usually BSA), which results in a linear relationship of the form s = x / y , where s is the scaled cardiovascular parameter, x is the cardiovascular parameter, and y is the body size parameter. This approach is problematic for three reasons. First, the correlation between the cardiovascular parameter and body size may not be linear and may have unanticipated variation; second, difficulties may result when scaling a cardiovascular parameter to a body size parameter that has different dimensions (e.g., volume vs area); and finally, it has been shown that ratiometric scaling does not produce size-independent scaled cardiovascular variables (i.e., the ratiometrically scaled variable correlates with body size). In contrast, allometric scaling divides the cardiovascular variable by an exponent, b ; in this case, s = x / y ^b . This nonlinear approach does not make assumptions between the variation and correlation between body size and cardiovascular parameter, is dimensionally consistent, and, importantly, effectively eliminates the effect of body size on the parameter. For example, allometric scaling minimizes and in some instances abolishes gender differences in cardiac dimensions. In addition, whereas scaling RV dimensions ratiometrically to BSA did not produce size independence in the aforementioned study of endurance athletes, scaling allometrically did. In this context, a third article in the current issue of JASE by D’Andrea et al. is particularly pertinent.

D’Andrea et al. report reference 2D and real-time 3D echocardiographic values for the right ventricle in 440 elite athletes (220 endurance trained and 210 strength trained) and 250 age-matched and gender-matched controls without detectable cardiovascular disease or risk factors. All RV measurements were greater in the endurance athletes than either strength-trained athletes or controls, and importantly, the upper range of normal in the endurance-trained athletes exceeded considerably the upper limits of normal as reported in the guidelines. Three-dimensional RV end-diastolic volume was associated with age, male gender, duration and type of training, increased LV stroke volume, and pulmonary arterial systolic pressure; only the latter four were independent predictors of RV end-diastolic volume on multivariate analysis. RV end-diastolic volume was only modestly correlated with BSA ( r = 0.29), but indexing RV volumes to BSA did not produce size-independent parameters, as did allometric scaling. Importantly, RV end-diastolic volumes remained significantly greater in the endurance-trained versus strength-trained athletes and controls after body size–independent allometric scaling.

Taken together, these three studies support scaling of RV parameters to body size and suggest that allometric scaling is the best approach, particularly when dealing with special groups such as athletes and perhaps ethnic populations. Indexing to BSA is a practical compromise but is inaccurate insofar as body composition affects the relation between body mass or surface area and cardiovascular variables; this problem may be particularly troublesome when normalizing dimensions in obese populations. It should be recognized that the subjects in the three RV indexing studies were nonobese, with BSAs ranging from 1.77 to 1.89 m ² . It has been suggested that the most appropriate variable with which to index cardiovascular parameters is fat-free mass (i.e., body mass with high metabolic activity), unfortunately an approach that is both inconvenient and often impractical. It has been argued that for both empirical and theoretical reasons, allometric scaling to fat-free mass is ideal (height may be a reasonable alternative). In terms of practicality, developing normative values of right-heart dimensions by indexing to BSA is an important first step and should be encouraged. Further study is needed to determine to what and how indexing the right ventricle affects individuals with obesity, influences the frequency of chamber-size reclassification, and affects cardiovascular outcomes.