Importance of Assessing Right Ventricular Dimensions and Function

The measurement of RV dimensions, morphology, and function is important in several situations, such as congenital heart disease, LV heart failure, pulmonary hypertension, pulmonary embolism, valvular heart disease, lung disease, and arrhythmogenic RV cardiomyopathy.

RV failure may result from conditions that lead to impaired RV contractility, such as RV infarction, right-sided cardiomyopathies, valvular heart disease or severe sepsis; to RV pressure overload, including pulmonic stenosis, pulmonary primary hypertension, and pulmonary hypertension secondary to left heart disease, lung disease, or thromboembolic disease; and to RV volume overload, for instance, tricuspid or pulmonary regurgitation. Many disorders, such as corrected and uncorrected adult congenital heart disease and intracardiac shunts, may result in right ventricle failure through a combination of pathophysiologic mechanisms. Also, decompensated right ventricle (both acute and acute-on-chronic) is increasing as the prevalence of predisposing conditions grows.

The prognostic value of RV function has been shown in several conditions such as LV heart failure, valvular heart disease, pulmonary hypertension, congenital heart disease, or myocardial infarction (MI). Thus the early detection of RV dysfunction can have an impact on therapeutic decision making and on prognosis. Finally, improved understanding of the RV response to pressure and volume overload might lead to more optimal surgical and medical treatments.

Techniques for Assessing Right Ventricular Dimensions and Function

Global RV function has been traditionally difficult to assess, given its irregular shape, which cannot be assumed to any geometrical model. Several imaging techniques have been used in the past to assess RV dimensions and function. The chest radiograph is a simple method to assess RV size with the cardiothoracic ratio, but this may be misleading because an enlarged RV may compress the LV, resulting in a normal cardiothoracic ratio. Angiography used to be the gold standard for assessment of RV volumes and regional and global function. But this technique is invasive, involves ionizing radiation and the use of potentially nephrotoxic contrast, and is not as accurate as CMR. Echocardiography and radionuclide ventriculography have been used for the assessment of RV dimensions and function. More recently, “nongeometric” techniques such as three-dimensional (3D) echocardiography, CMR, and multidetector-row computed tomography (CT) permit accurate assessment of RV volumes, function, and mass.

Echocardiography

Echocardiography is the most frequently used technique for assessing the right ventricle; it is cheap, widely available, and can be used bedside in very ill patients. It provides information about RV morphology, dimensions, septum convexity, function, tricuspid regurgitation, and estimates of pulmonary arterial pressure and RV pressure. But the assessment of the right ventricle with echocardiography has several limitations. First, the location of the right ventricle behind the sternum restricts the window that can be accessed by the ultrasound beam. Second, the complex shape and thin walls of the right ventricle make it necessary to image the right ventricle from several projections, although the short-axis view is usually the most helpful. Third, the thick trabeculations in the chamber may be confused with a thrombus, a tumor, or hypertrophic cardiomyopathy. Finally, there is a lack of accurate mathematical models to quantify RV mass and volumes with M-mode or two-dimensional echocardiography (2DE) because quantitative measurements are based on geometric assumptions that do not apply to the right ventricle. A qualitative approach for RV volume assessment is usually applied, with the RV size being described as either normal or mildly, moderately, or severely enlarged. If the RV is the same size as the LV, it is usually characterized as moderately enlarged, and if larger than the LV, it is severely enlarged. Also, qualitative evaluation of RV function is usually applied with RV characterized as normal or mildly, moderately, or severely dysfunctional. An approximation to quantification has been introduced with M-mode measurement of the tricuspid annular plane systolic excursion (TAPSE). This parameter provides a rough estimate of RV function but it is not angle independent: it only takes into account the longitudinal function of the RV, which is the predominant, but not unique, component of RV systolic function, and, because it is measured at the basal segment, the presence of regional wall motion abnormalities will affect its accuracy. The fractional area change in the four-chamber view has also been used but, again, this takes into account only the lateral free wall and the presence of regional wall motion abnormalities or RV dilatation might affect its accuracy ( Fig. 39.2 ). Other indicators of RV function are flow Doppler-derived indices such as the myocardial performance index, tissue Doppler measurements of myocardial tricuspid annular myocardial velocities and time intervals, and strain and strain rate measurements of contractility whether with tissue Doppler or speckle tracking imaging (STI). Tissue Doppler is an angle-dependent technique, and incorrect alignment with the ultrasound beam, poor signal-to-noise ratio (SNR) and variability because of placement of the region of interest may affect the results. Speckle tracking imaging is a promising technique that has been applied to the right ventricle to measure longitudinal strain in the six segments in which the right ventricle is classically divided. This technique is useful for the diagnosis of various right heart diseases, but it may have problems in the case of thin wall or poor image quality in the RV lateral wall, and there is variability across vendors. Doppler transesophageal echocardiography is another echocardiographic method of RV assessment, but it is semiinvasive, is not well suited for evaluation of anteriorly positioned right ventricles, and requires special skills. Three-dimensional echocardiography (3DE) has emerged as a more accurate and reproducible approach to ventricular quantitation, mainly by avoiding the use of geometric assumptions of the ventricular shape. Real-time 3DE is an online acquisition of a 3D dataset of the heart without the need for electrocardiographic and respiratory gating. It calculates right ventricular ejection fraction (RVEF) using the volumetric semiautomated border detection method, which needs to be manually adjusted, and after acquisition and display of end-diastolic and end-systolic frame, long-axis, planes, and volumetric data of the right ventricle are analyzed offline. Eventually, curves of regional and global RVEF are produced and analyzed. However, there are practical problems, such as full cardiac visualization, good-quality endocardial border recognition for manual endocardial tracing, and time consumption. Also, these methods need a stable cardiac rhythm and constant cardiac function during image acquisition, although new single-beat 3DE techniques allow a fast acquisition in one short breath-hold. 3DE has been compared with CMR for the evaluation of RV function, and improved results in comparison with 2D echocardiography have been obtained, both with the usual technique and with the new single-beat 3DE, with consistent underestimation of RV volumes and ejection fraction with 3DE. Although 3DE has been used mainly for the left ventricle, assessment of the right ventricle with 3DE is feasible during routine standard echocardiography.

M-mode and two dimensional echocardiographic methods for right ventricular (RV) function assessment. (A and B) RV planimetry in diastole (A) and systole (B) from which the fractional area change can be derived. (C and D) M-mode measurement of the tricuspid annular plane systolic excursion (TAPSE). Both parameters provide a rough estimate of RV function with significant accuracy and reproducibility problems. TAPSE only takes into account the longitudinal function of the right ventricle, which is the predominant, but not unique, component of RV systolic function, and does not account for regional wall motion abnormalities. The fractional area change is measured in only one plane, so the presence of regional wall motion abnormalities or RV dilatation greatly affect its accuracy.

Multislice Computed Tomography

Multislice CT (MSCT), although not a first-line technique for RV function quantification, is emerging as an alternative tool, especially for patients with implantable devices (a contraindication for CMR). However, MSCT uses ionizing radiation and potentially nephrotoxic contrast, and requires a low and stable heart rate during acquisition. The quantification of ventricular volumes throughout the cardiac cycle requires retrospective electrocardiogram (ECG)-gated helical acquisition, which increases the radiation significantly. Postprocessing is with either a volume threshold approach or, similar to CMR, myocardial analysis is automated but most times requires manual adjustment ( Fig. 39.3 ). In a study that compared accuracy and reproducibility of MSCT with CMR in the quantification of RV function, spiral MSCT with reconstruction at 5% intervals had good agreement with CMR for RVEF and volumes, with good accuracy and reproducibility, but at the expense of an average radiation dose of 12 to 18 mSv. Another study that compared volumetric quantification of RV-shaped phantoms with CMR, MSCT, and real-time 3DE showed that CMR images yielded the most accurate measurements, whereas MSCT measurements showed slight but consistent overestimation; and 3DE showed small underestimation.

Example of ventricular function analysis with computed tomography. (A and B) Analogous to cardiovascular magnetic resonance, they show myocardial analysis for left ventricular function quantification both with ejection fraction and with wall thickening. (C and D) The ventricular threshold method for measurement of both LV (C) and right ventricular (D) volumes.

Imaging Strategies for Cardiovascular Magnetic Resonance of the Right Ventricle

Before the study begins, it is essential to obtain an accurate ECG gating with minimal ectopy and to instruct the patient in breath-holding. Sometimes oxygen may be applied to improve breath-hold length. Ventricular ectopy can be a problem, mainly in patients with congenital heart disease or with suspicion of arrhythmogenic RV cardiomyopathy (ARVC). If this condition is present, pretreatment with an antiarrhythmic agent should be considered or, if this is unavailable or unsuccessful, prospective triggering with acquisition window limited to systole can diminish artifacts.

Spin echo (black-blood) sequences (including turbo spin echo, half-Fourier acquisition single-shot turbo spin echo, or spin echo-echo planar imaging) are used mainly for anatomic assessment ( Fig. 39.4 ), because the detection of fatty replacement/infiltration in the RV free wall is subject to high interobserver variability and, at the moment, is not considered a diagnostic criteria for ARVC to improve their specificity.

Short-axis and transaxial spin echo images of the heart in a healthy subject. The right ventricular free wall is perfectly depicted between the black-blood pool and the black pericardium. Note that there is epicardial fat deposition around the right coronary artery (anterior atrioventricular groove), left anterior descending artery (anterior interventricular groove), and left circumflex (posterior atrioventricular groove). It is important to know the normal patterns of fat distribution to prevent false-positive reading of scans in assessing patients for possible fat infiltration in arrhythmogenic right ventricular cardiomyopathy.

Functional evaluation of the RV is performed by using steady-state free precession (SSFP) cines with segmented fast imaging, allowing the whole stack of images to be acquired in 5 to 8 minutes, thereby reducing breathing and movement artifact ( Fig. 39.5 ). Moreover, real-time SSFP imaging can acquire the whole stack in just one breath-hold with acceptable accuracy and image quality. In patients who are unable to hold their breath consistently, solutions using the same sequence with more signal averages or combined with navigator echo are successful with free breathing, with a slight increase in the scanning time.

Steady-state free precession cine images in eight short-axis planes from cines acquired from base to apex. The series of short-axis slices begins in the atrioventricular groove in diastole and goes through the ventricles down to the apex, thus covering the whole left ventricle (LV) and right ventricle (RV). In the most basal slice (upper left), the pulmonary artery will appear in systole because of atrioventricular ring descent. The high contrast between blood and myocardium allows the drawing of the endocardial and epicardial borders for volume and mass determinations. For the images to be accurate and reproducible, it is important to always use the same slice thickness and the same gap between slices. Also, retrospective gating should be used to acquire the whole cardiac cycle.

RV function can be estimated by CMR, in a way analogous to echocardiography, with the fractional area change method ( Fig. 39.6 ). This is a fast method but lacks accuracy, owing to the need for geometric assumptions. A better method of measuring volumes, and thereby function, is by using Simpson’s rule. This method requires a stack of contiguous tomographic slices to be acquired that encompass both ventricles entirely.

Measurement of uniplane fractional area change (red shaded area) of the right ventricle through planimetry of the right ventricle in diastole (left) and systole (right) in a four-chamber view. This is a fast approximation to right ventricular function but, as for two-dimensional echocardiography, the use of a single plane makes it insensitive to regional wall motion abnormalities or geometric distortion, affecting its accuracy and reproducibility.

The right ventricle is well shown in the transaxial plane from the tricuspid valve to the apex. Consequently, a stack of transverse images was primarily used for quantification, with good agreement with pulmonary flow and LV stroke volume ( Fig. 39.7 ), as well as better interobserver and intraobserver reproducibility than the short-axis plane. However, it can be problematic because it is subject to considerable partial volume effects, especially in the anterior and inferior walls. The question has been raised as to whether the RV volumes should always be measured in the axial orientation. Yet the interstudy reproducibility of RV measurements in the short-axis orientation is good, with differences not clinically significant and this orientation allows the LV and RV dimensions to be measured simultaneously, thus saving time ( Fig. 39.8 ). A recent study has shown comparable reproducibility for the two approaches when measuring volumes of the systemic right ventricle in 22 patients with congenital heart disease. Although ejection fraction and stroke volume were measured more consistently in the axial orientation, ventricular mass was measured more consistently in the short-axis orientation. Nowadays, nearly all CMR units have adopted the short-axis stack approach, leaving the axial orientation for specific congenital heart disease patients.

(A) To plan the transversal slices, a cine sequence through the right ventricular (RV) outflow tract aligned in an oblique sagittal plane should be acquired. This view is useful for examining the function of the right ventricular free wall from the apex to the pulmonary valve (PV) and visualizing pulmonary regurgitation. (B) Multiple transaxial steady-state free precession cine sequences in contiguous planes should be acquired from the pulmonary valve level down to the inferior right ventricular wall, using the right ventricular outflow tract view to pilot them. These planes are invaluable for examining regional right ventricular wall motion. Again, the slice thickness and gap between slices should be consistent among studies. Two features of the normal right ventricle are seen in this transversal stack: the muscular outflow tract and the more apical insertion of the tricuspid valve in the septum compared with the septal leaflet of the mitral valve . Ao, Aorta; LA, left atrium; LV, left ventricle; PA, pulmonary artery; RA, right atrium.

Volumetric methods for assessment of right ventricular (RV) function (Simpson’s rule). (A) Simultaneous quantification of left ventricular and RV volumes and systolic function in the short-axis orientation. There is difficulty in delineating the most basal slice of the right ventricle, where the right atrium and the pulmonary trunk may be incorrectly included in the RV blood pool. (B) Quantification of only the right ventricle in the axial orientation; although the tricuspid valve and right atrium are easily depicted here, problems with partial volume effects in the inferior slices are significant.

As commented, Simpson’s rule is used for measuring volumes and function. A standardized method of combined ventricular functional analysis is described in Chapter 16 . When using this method, it is critical to ensure that the most basal part of the free wall of the right ventricle in diastole is included in the most basal cine slice because it can be easily truncated. Manual or semiautomatic planimetry of the endocardial borders of each ventricle at both end diastole and end systole and of the epicardial borders at end diastole is done. Care must be taken to exclude the right and left atria as they come into the basal imaging planes during systole. The ventricular volume is equal to the sum of the endocardial areas multiplied by the distance between the centers of each slice ( Fig. 39.9 ). The stroke volume is equal to the difference between the end-diastolic volume (EDV) and the end-systolic volume (ESV). The ejection fraction (EF) is calculated as the stroke volume divided by the EDV. The mass is calculated as the volume of tissue occupied by the free wall multiplied by an assumed density of 1.05 g/mL. The volumes obtained by this method are independent of geometric assumptions and dimensionally accurate. Mass measurements agree well with autopsy studies. RV parameters obtained with this technique are reproducible, with an interobserver variability of 6.3% for EDV, 8.6% for ESV, 7% for stroke volume, 4.4% for EF, and 7.8% for RV mass and an intraobserver variability of 3.6% for EDV, 6.5% for ESV, 5.9% for stroke volume, 4% for EF, and 5.7% for RV mass. Current analysis software allows for automated quantification of the RV, with automated endocardial border detection of the right ventricle in contiguous short-axis slices and tracking of the tricuspid valve throughout the cardiac cycle. A volume–time curve can then be produced from which a velocity–time curve can be derived for subsequent calculation of diastolic function parameters.

Summation of discs is used to quantify ventricular volumes and mass. The ventricular volume is equal to the sum of the endocardial areas multiplied by the distance between the centers of each slice, both for the end-diastolic volume (A) and the end-systolic volume (B). The ventricular mass is calculated as the sum of the epicardial minus the endocardial areas multiplied by the distance between the centers of each slice, as shown in A. LV, Left ventricle; RV, right ventricle.

If necessary, other CMR techniques could be used in the assessment of the right ventricle. Flow velocity maps allow for accurate assessment of pulmonary valve regurgitation (regurgitant fraction) and systemic-to-pulmonary flow ratio. Magnetic resonance angiography (MRA) must be done in case assessment of great vessel anatomy is advisable, such as in a number of congenital cardiac conditions. New free-breathing time resolved MRA techniques enable accurate diagnosis and vessel measures compared with conventional breath-hold MRA and can be performed in children and adults who find breath-holding difficult. Recently, four-dimensional (4D) flow has been shown to be able to detect mild impairment of RV function in a small group of patients with ischemic heart disease, and RV kinetic energy, a 4D flow-derived parameter, has been suggested as a more robust measure of RV efficiency than RV function. Diffusion tensor imaging has also been applied to the assessment of RV fibers’ geometry. Although promising, these techniques are still in development and more work is needed to fully understand all the information that can be derived from them, along with their potential clinical use ( Fig. 39.10 ). Myocardial tissue tagging, feature tracking, and myocardial phase contrast are briefly summarized later. Finally, CMR with LGE can detect myocardial fibrosis in a number of both ischemic and nonischemic cardiomyopathies.

Diffusion tensor imaging of the left ventricle/right ventricle (A) and four-dimensional flow assessment of the pulmonary artery and branches (B) of the right ventricle. Both techniques are promising for assessment of fiber geometry and flow dynamics, respectively.