Chapter 13. Evaluation of Right Heart Function

As the role of the right ventricle (RV) in overall cardiac function has been more fully appreciated, greater efforts have been made to describe and quantify RV function using TEE in ways useful to clinicians. These efforts have been frustrated by the non-geometric, asymmetric shape of the chamber, its sequential contraction pattern, and the obscuring effect of epicardial fat on RV wall motion and thickness. In addition, the RV is exquisitely sensitive to loading conditions, overall pulmonary function, and the interdependence of the two ventricles.

The RV is anatomically divided into its inflow and outflow portions reflecting its dual embryonic origin. The inflow portion begins at the tricuspid valve and extends towards the apex to include the trabeculated, posteroinferior segments, while the outflow portion is usually free of trabeculations and includes the infundibulum (anterosuperior segments) and pulmonic valve. A series of muscular bands divide the two portions, the most important to the echocardiographer being the moderator band. This structure extends from the base of the anterior papillary muscle to the ventricular septum and should not be mistaken for a thrombus or intracavitary mass (see Chapter 3).1,2

Ventricular systolic ejection of the RV has a different pattern than that of the left ventricle. The ejection phase begins earlier and lasts longer, while the velocity profile is characterized by a lower and delayed peak.3 Right ventricular ejection is largely due to a bellows-like motion of the free wall and longitudinal shortening of the ventricle (apex to annulus) rather than the twisting and rotational motions that predominate on the left side.2,4 Finally, contraction of the RV is sequential, beginning with the free wall and moving towards the infundibulum.5

Although the vast majority of information about the RV can be obtained from a small number of images, the chamber's complexity defies standardized description or definition. The RV is a paradoxical structure, appearing triangular in one view but of an elongated, crescent shape in another. Likewise, while it usually appears smaller than the left ventricle (LV), its end-diastolic volume is actually greater.2

As the RV cannot be completely seen in any single image, multiple scan planes are required to adequately assess RV structure and function. The RV has approximately one-sixth the mass of the LV, and performs about one-quarter of its partner's stroke work.6 It consists of a free wall, an inferior or diaphragmatic wall, a septal wall, and an outflow tract (RVOT) region, although there is no formal segmental scheme for classifying wall motion as exists for the LV.

Despite this, a series of guidelines have been developed that attempt to establish standards for measurement of global RV size and function. Most important of these are the Recommendations for Chamber Quantification, developed jointly by the American Society of Echocardiography and the European Association of Echocardiography.7 These standards were developed by a combination of methods, including statistical calculation of standard deviation, expert opinion, and assessment of associated outcome. It is recommended that the same values be used to assess RV size for both transesophageal (TEE) and transthoracic echocardiography (TTE), even though the actual images obtained may be markedly different.7 Quantification with TEE is frequently very challenging due to the increased difficulty in achieving standard image planes and views.

Midesophageal Views

Probably the most useful view for RV assessment is the midesophageal four-chamber (ME-4C) view. To image the right ventricle, the standard ME-4C view is obtained and the probe is then turned slightly to the right to bring the tricuspid valve into the middle of the screen (Figure 13–1). The basilar RV free wall will be to the left of the screen while the apical portion of the free wall will be in the far field or slightly to the right, depending on the orientation of the heart. It may be useful to increase the multiplane angle to 10° to 20° to optimize the view of the RV cavity. A normal RV will be no more than two-thirds the longitudinal extent of the LV with the LV comprising the apex of the heart. From this imaging plane, the mid-cavity and longitudinal diameters of the RV may be measured and systolic motion qualitatively assessed (Figure 13–2). In addition, color-flow (CFD) and spectral Doppler analysis of trans-tricuspid flow as well as tissue Doppler examination of the tricuspid annulus may be performed from this image to evaluate possible valvular pathology and diastolic dysfunction. The ME-4C view is also an excellent view in which to quantify the right atrium (RA) by measuring its major and minor axes and comparing it to its partner to the left. As opposed to the left atrium (LA), normal ranges for RA size are not different for men and women. Normal reference limits for RV and RA size as well as partition values for mild, moderate, and severe enlargement are listed in Table 13–1.7

Rotation of the TEE multiplane angle to approximately 30° to 60° and slight rotation of the probe to the right will bring the inflow-outflow or wrap-around view onto the screen (Figure 13–3). In this view, the RA, tricuspid valve (TV), RV, pulmonic valve (PV), and proximal pulmonary artery (PA) appear in order from the left side of the screen in an arc sweeping counter-clockwise in the far field around the aortic valve to the right. The RV inferior or diaphragmatic free wall is visible in the far field and the infundibular portion of the RV is particularly well seen closer to the right side of the screen. This is also an excellent view in which to use CFD to interrogate the TV and PV and to measure the dimensions of the PV annulus as well as the RV outflow tract (RVOT) or the immediate subpulmonary regions (Figure 13–4; see Table 13–1).

Further rotation of the multiplane angle to 90° to 120° while keeping the RA in view will bring the bicaval view onto the screen with the left atrium in the near field, the superior vena cava to the right, and the inferior vena cava (IVC) to the left (Figure 13–5). Frequently, the eustachian valve is seen emanating from the junction of the RA with the IVC at the lower end of the crista terminalis. This fetal remnant may have a mobile, filamentous structure associated with it that is known as a Chiari network, and is considered a normal variant (see Chapter 3). The interatrial septum is visible in the center of the screen and may be easily examined for a patent foramen ovale or other abnormalities. The tricuspid valve may be brought into this image at the far left field by further multiplane transducer angle rotation. This modified bicaval view is ideal for spectral Doppler analysis of a tricuspid regurgitation jet as the direction of the jet is usually closely aligned with the direction of the ultrasound beam (Figure 13–6). The coronary sinus is often seen just underneath the IVC to the left side of the screen and may prove useful in placement of a coronary sinus catheter during cardiac surgery (see Chapter 5).

Transgastric Images

Transgastric images of the RV should be carefully examined as it is not uncommon that some of the best images of the RV are obtained from this position. A transgastric short-axis (TG-SAX) view of the RV is obtained by advancing the TEE probe into the stomach (35 to 45 cm from the incisors), flexing the probe, identifying the LV, and turning the probe to the patient's right. The RV will appear crescent shaped and wrapped around the LV. The tricuspid leaflets are often seen in short axis with extreme flexion (Figure 13–7). A long-axis or inflow RV view may be obtained one of two ways. From the LV short-axis view, the probe is turned to the right to bring the RV short-axis view into the center of the screen and the multiplane angle is rotated to 90° to 140°, yielding a view with the RV to the left of the screen, the TV in the middle, and the RA to the right (Figure 13–8A). The inferior free wall will appear in the near field and the anterior free wall in the far field. Alternatively, from the LV short-axis view, the transducer angle is advanced to 100° to 120°, identifying the LV long-axis view, and the probe is then turned to the right to reveal the RV inflow or long-axis view. Spectral Doppler analysis is not optimal from this position given the divergence between the angle of the ultrasound beam and the direction of blood flow (Figure 13–8B).

An RV outflow view may occasionally be developed from the RV short-axis view by rotating the multiplane angle slowly towards 100° to 130°, but also turning the probe slightly towards the patient's left (Figure 13–9A). The RVOT and PV will appear in the mid–far field, parallel to the beam, often in optimal position for color and spectral Doppler analysis, should that be desired (Figure 13–9B).

The deep transgastric image of right-sided structures is sometimes optimal for assessment of tricuspid annular motion with tissue Doppler methods.8 This image is obtained by advancing the probe to the deep transgastric position, flexing the tip to identify the deep transgastric images of the LV, and turning the probe slightly towards the patient's right side (Figure 13–10). The result often resembles an apical four-chamber image obtained with transthoracic echocardiography.

The transgastric position is also the appropriate position from which to examine hepatic vein flow (Figure 13–11). From the transgastric position, the probe is turned to the patient's right side until the liver is seen. An appropriately sized and positioned vein is chosen and the multiplane angle is rotated such that the vein is as linear and as parallel with the ultrasound beam as possible. Spectral Doppler analysis of flow is then performed to examine flow patterns, in a manner analogous to that of pulmonary venous flow.

Exposure to chronic conditions of excessively high preload or afterload will eventually result in fundamental changes to the structure of the RV. However, the mechanisms by which the RV copes with pathologic changes depends on whether it is facing increases in pressure work or increases in volume work.

Normal RV free-wall thickness is less than half that of the LV due to low pressures characteristic of the right-side system. However, elevated pressures in the right ventricular cavity, be they a result of pulmonary hypertension, pulmonic valvular pathology, or intrinsic myocardial disease, will cause patterns of ventricular hypertrophy that are indistinguishable from those of the left. Furthermore, the diagnosis of RV hypertrophy may be confounded by the appearance of trabeculations, which often become extremely dense and prominent in cases of pulmonary hypertension, especially near the apex. Right ventricular wall thickness is best measured using the inferior or lateral wall rather than the anterior wall, and measures 5 mm or less in a normal heart but may exceed 10 mm in cases of cor pulmonale.9,10 Increased RV wall thickness may also be caused by primary myocardial infiltrative diseases such as amyloidosis or endocardial fibroelastosis.11

Under normal conditions, the center of mass in the heart is in the LV, and the ventricular septum maintains its concave curvature towards the LV throughout the cardiac cycle. As such, the LV maintains its circular shape in the transgastric midpapillary view. However, as the RV hypertrophies, the center of mass shifts rightward, towards the septum itself (Figure 13–12). The septum begins to flatten, and the LV thus begins to assume a D shape. Furthermore, this septal deformation results in the loss of the normal triangular or crescent shape of the RV in the transgastric view and the RV assumes a more spherical shape. Indeed, from the transgastric, midpapillary view, the two ventricles may begin to appear as two similar cavities with a flat wall dividing them. However, because the deformity in septal configuration is due to elevations in pressure within the ventricular cavities, this flattening is greatest when the intracavitary pressures are greatest, that is, at end-systole.12–14

In cases of myocardial failure, the RV will begin to dilate as a compensatory mechanism to maintain its stroke volume regardless of whether the failure is primary, secondary to volume overload, or following a long period of hypertrophy. This is detectable in a variety of echocardiographic views but the midesophageal four-chamber view adjusted to maximize the tricuspid annulus is the most useful. As noted, the LV should constitute the apex of the heart and the RV should extend no farther than two-thirds the length of the LV. A moderately dilated RV will reach to more than two-thirds the longitudinal extent of the LV, and a severely dilated RV will actually form the apex of the heart, displacing the LV completely (Figure 13–13).7 In addition, the RV will transform from a triangular to a circular shape as it dilates.9 More quantitative measurements have been established and are obtained from the midesophageal four-chamber as well as from the inflow-outflow view (see Table 13–1).

Septal motion is also abnormal in conditions of RV dilatation when the exaggerated motion of the septum occurs without any real shift in the center of mass, as is seen with RV hypertrophy. The septum appears to flatten, resulting in a D-shaped LV, but this occurs during the period of maximal filling, that is, during end-diastole.12,14 Then, as the ventricles empty, the septum is free to assume a more normal curvature. It is important, though, not to confuse this motion with the septal “bounce” that is commonly seen with ventricular pacing using epicardial leads following cardiac surgery and attributed to an abnormal pattern of contraction.15

Any assessment of RV function is complicated by the combined interplay of factors that distinguish it from the LV. As discussed above, its geometry is complex and difficult to image due to its distance from the probe as well as its structure. Because the wall is thinner and less muscular, systolic inward excursion is less than that of the LV, and it is affected by LV function by way of sharing a common wall. Furthermore, its size varies significantly during the respiratory cycle in opposing directions, depending on whether the patient is breathing spontaneously or is being mechanically ventilated.

Due in part to these difficulties, the ASE-SCA guidelines for intraoperative TEE examination concluded that mild degrees of RV hypocontractility are hard to define accurately, and have recommended that RV dysfunction would only be diagnosed when a wall segment is either akinetic or dyskinetic.16 In the ME-4C view (see Figure 13–1) one can evaluate the function of the free anterolateral RV wall. The midesophageal inflow-outflow view (see Figure 13–3) allows assessment of the RV diaphragmatic wall (in the far field and to the left) and of the infundibular portion of the RV (at the right side of the view). Last, the transgastric RV inflow view (see Figure 13–8) also allows evaluation of the RV diaphragmatic wall.

Similar to the LV, regional RV dysfunction can coexist with preserved global function, for example, when one of the acute marginal branches of the right coronary artery is occluded. A small portion of the anterior RV free wall is supplied by the left anterior descending artery through its conal branch, and this can lead to regional RV dysfunction with left coronary artery disease.17 During cardiac surgery, cardioplegia delivery may be inadequate for RV protection, especially with retrograde techniques.18 Also, the right coronary artery arises from the anterior aspect of the ascending aorta, serving as a common site for coronary air embolism during separation from cardiopulmonary bypass.

Hypercontractility of the infundibular RV outflow tract leading to obstruction of blood flow has been described in approximately 5% of patients after cardiac surgery.19 The end-systolic obliteration of RV outflow can be associated with significant pressure gradient across the RV outflow tract (>25 mm Hg). Several risk factors for this condition have been described, including RV hypertrophy, hypovolemia, and increased adrenergic tone. Clinically, it is associated with tricuspid regurgitation and hemodynamic instability.

The technical difficulties that hamper accurate assessment of regional RV function also confound assessment of global function, leading to a diverse array of quantitative indices (Table 13–2).

Ejection phase indices are especially affected by the non-geometric shape of the RV. In addition, tricuspid regurgitation is very common and the decreased afterload might compensate for decreased contractility. The unidimensional index shortening fraction can be calculated along either the short or long axis of the RV from the ME-4C view or the TG-SAX (with the probe rotated rightward to show the RV), but is rarely adequately reflective of RV contraction.20 A variant of shortening fraction that has been found more useful is tricuspid annular plane systolic excursion (TAPSE). It has been shown that the main contribution to RV systole is the contraction of longitudinal muscle fibers that cause the tricuspid annulus to move toward the RV apex. The majority of this movement occurs in the lateral aspect of the ventricular free wall, as the medial septal attachment of the tricuspid annulus is relatively fixed. Hence the tricuspid annulus moves like a hinge during systole, resulting in a measurable long-axis excursion of the lateral aspect of the tricuspid annulus. In TTE, TAPSE is usually measured in the apical four-chamber view. However, for TEE imaging, the transgastric RV inflow view (see Figure 13–8A) more accurately aligns the axis of annular movement with the direction of the ultrasound beam, allowing more accurate measurement of annular motion using M-mode (Figure 13–14).21 In one study, TAPSE had a better correlation with radionuclide-measured RV ejection fraction compared to RV fractional area change.22 Other studies have shown that decreased TAPSE is associated with a worse prognosis in patients with acute myocardial infarction and in patients after mitral valve repair surgery.23,24 It should be mentioned, though, that TAPSE can also be affected by LV function independent of RV function.25

A two-dimensional ejection phase index is the fractional area change (FAC), calculated as follows:

Similar to shortening fraction, it can be measured by planimetry in either the ME-4C or the TG-SAX views. The degree of RV dysfunction can be classified as mild (FAC 25% to 31%), moderate (FAC 18% to 24%), or severe dysfunction (FAC <18%).7 In patients undergoing coronary artery bypass surgery, RV FAC values less than 35% were associated with a worse early and late postoperative outcome.26

A three-dimensional ejection fraction can be calculated using a summation of smaller geometric volumes (ellipsoid, prism, or pyramid) to model the RV.27,28 Similar to the LV, a modified Simpson's method can also be used to calculate RV ejection fraction.29 In the future, three-dimensional echocardiography might be employed to obtain more accurate measurements of RV size (see Chapter 24).30

Several non-geometric indices of RV function have been suggested. In patients with tricuspid regurgitation, the acceleration of the regurgitant jet into the right atrium (dP/dt) reflects the rate of increase in RV pressure during systole, and is determined by RV contractility. To calculate dP/dt, the tricuspid regurgitation jet is identified at the transesophageal RV inflow-outflow view with color-flow Doppler. The omniplane angle is adjusted to align the tricuspid regurgitation jet with the ultrasound beam, and the continous-wave spectral Doppler trace of the regurgitation jet is examined (Figure 13–15). dt is defined as the time interval between jet velocities of 1 m/s (corresponding to a ventriculo-atrial pressure gradient of 4 mm Hg) and 2 m/s (corresponding to a pressure gradient of 16 mm Hg). The dP/dt is then calculated by dividing the pressure difference (16 – 4 = 12 mm Hg) by the time interval (in seconds). Calculation of RV dP/dt using echocardiography has a good correlation with values derived from invasive cardiac catheterization.31 Compared to ejection phase indices, dP/dt is more sensitive to preload changes but less sensitive to afterload.32

Another ventricular function index free of any geometrical assumptions is the myocardial performance index (MPI), originally defined for the left ventricle as the sum of the two isovolumic intervals (isovolumic contraction and isovolumic relaxation) divided by ejection time (Figure 13–16).33 For the RV, a spectral pulsed-wave Doppler trace of pulmonary artery flow provides the ejection time and is recorded either from the upper esophageal aortic arch short-axis or the transgastric RV outflow view. Total systolic time can be measured as the duration of a tricuspid regurgitation jet if it exists, or as the interval between the end of the A and the start of the E wave on a tricuspid inflow spectral Doppler trace (Figure 13–17). The isovolumic period is calculated by subtracting the ejection time from the total systolic time. MPI is affected by both systolic and diastolic function, and is relatively independent of heart rate, afterload, and preload.33 MPI value above 0.5 was found to predict hemodynamic instability and mortality after cardiac valvular surgery.34

Several important hemodynamic values can be derived from spectral Doppler interrogation of right-side flows. Cardiac output can be calculated by placing the pulsed-wave Doppler sample volume just proximal to the pulmonary valve either in the transgastric RV outflow view or the upper esophageal aortic arch view to measure the velocity-time integral (Figure 13–18A and B). The RVOT diameter is measured in the same view or in the midesophageal RV inflow-outflow view (see Figure 13–4). Stroke volume is then calculated as follows:

SV = RVOTArea × RVOTVTI

In patients without significant tricuspid valve regurgitation, this measurement correlates well with cardiac output measured by thermodilution.35

In the presence of tricuspid regurgitation and in the absence of RV outflow obstruction, continuous-wave Doppler can be used to estimate pulmonary artery (PA) systolic pressure because the peak velocity of the tricuspid regurgitant jet reflects the pressure gradient between the RV and RA (Figure 13–19A and B). Most patients with pulmonary hypertension have some degree of tricuspid regurgitation, making quantitative estimation possible. It is important to align the ultrasound beam to within 20° of parallel to the regurgitant jet to avoid significant underestimation of the pressure gradient.

Estimation of RA pressure is based on the absolute measurement and respiratory variation of IVC size using the TG-SAX view with the probe rotated to the right (Figure 13–20). A maximal diameter less than 20 mm quite accurately differentiates patients with RA pressures less than 10 mm Hg from those with greater than 10 mm Hg.36 While the decrease in diameter in response to a forced sniffing maneuver (normal decrease is 35% to 55%) can add some information about RA pressure within 5 mm Hg gradations, it is significantly less reliable.36 In spontaneously breathing patients, IVC diameter decreases during inspiration, while in mechanically ventilated patients the opposite is usually true depending on intravascular volume status. An increase of more than 12% during inspiration predicts an increase in cardiac output in response to volume loading.37

Right ventricular diastolic function can be evaluated in a manner analogous to the left ventricle, although it has been less studied and its clinical significance is still largely unknown. Compared to the mitral annulus, the tricuspid annulus diameter is larger, resulting in lower inflow velocities.38 Also, tricuspid inflow is more subject to respiratory variations, with inflow velocities increasing during spontaneous inspiration.39 When evaluating diastolic function, it is important to ensure that the respiratory variation measured is not an artifact resulting from respiratory-induced translocation and rotation movements of the heart itself, which might lead to situations in which the interrogated flow is not parallel to the ultrasound beam.

Right ventricular filling is evaluated by measuring tricuspid inflow velocities, and can usually be measured from the midesophageal, modified bicaval view with the probe turned to the right and the transducer angle rotated to somewhere between 90° and 120°. The ideal position of the transducer angle is determined by attempting to align the direction of blood flow with the direction of the ultrasound beam, and is highly dependent on how the heart is positioned in the chest with respect to the esophagus. Normal tricuspid inflow consists of an early diastolic E wave corresponding to right atrial emptying upon opening of the tricuspid valve, and a late diastolic A wave corresponding to atrial contraction (Figure 13–21). Right atrial filling is usually evaluated using TEE by interrogating hepatic vein flow or IVC flow. From the transgastric position, the probe is rotated to the right, and adjusted such that a hepatic vein branch is as parallel as possible with the ultrasound beam. The hepatic vein inflow usually includes two antegrade flows—S during systole and D during early diastole—and a retrograde end-diastolic A wave corresponding to atrial contraction (see Figure 13–11). A second retrograde end-systolic V wave is variably present. Normal values for these flow velocities are summarized in Table 13–3.

A third modality used to assess RV diastolic function is tricuspid annular tissue Doppler analysis. The normal spectral curve includes one positive systolic wave (S′) and two negative diastolic waves—an early (E′) wave and a late (A′) wave. Normally, the E′ wave is higher than the A′ wave. In a pattern similar to the LV, the first stage of diastolic dysfunction is related to impaired ventricular relaxation, and is denoted by decreased E- and E′-wave velocities, E/A and E′/A′ ratios of less than 1, and increased deceleration times on trans-tricuspid flow assessment. More severe diastolic dysfunction resulting from decreased ventricular compliance and restricted filling is characterized by a high E/A ratio (usually >1.5), normal or short trans-tricuspid deceleration times, and an S-wave peak velocity that is lower than the D wave on the hepatic vein flow. In contrast to the E/A ratio on the tricuspid inflow trace, the E′/A′ ratio on the tricuspid annulus tissue Doppler tracing remains less than 1 and continues to diminish with worsening function.