Right ventricular function is a crucial determinant of long-term outcomes of children with heart disease. Quantification of right ventricular systolic and diastolic performance by echocardiography is of paramount importance, given the prevalence of children with heart disease, particularly those with involvement of the right heart, such as single or systemic right ventricles, tetralogy of Fallot, and pulmonary arterial hypertension. Identification of poor right ventricular performance can provide an opportunity to intervene.
In order to properly assess right ventricular function, it is important to highlight some of the features that differentiate the normal right ventricle from the left ventricle. The very anterior position of the right ventricle in the chest limits its visualisation by echocardiography. Its complex geometric shape, in contrast to the ellipsoid shape of the left ventricle, precludes the easy assessment of function by traditional echocardiography.Reference Ho and Nihoyannopoulos1 The right ventricle has prominent trabeculations, making border definition challenging when clear endocardial borders are required for some calculations, including fractional area change, as detailed later in the text. Finally, the muscle fibres of the right ventricle are arranged mainly longitudinally, such that most of the contraction occurs in that plane, as opposed to the oblique, longitudinal, and circular myofibres of the left ventricle.Reference Ho and Nihoyannopoulos1 Therefore, the contraction pattern of the right ventricle is characterised by a bellow-like motion of the free wall towards the septum, a longitudinal motion of the base towards the apex, and bulging of the ventricular septum into the right ventricular cavity.Reference Haber, Metaxas, Geva and Axel2–Reference Rushmer, Crystal and Wagner4 In disease states, however, the right ventricle shifts from the mainly longitudinal contraction pattern even in conditions in which the right ventricle is not the most affected chamber.Reference Okada, Rahmouni, Herrmann, Bavaria, Forfia and Han5 By way of example, patients with tetralogy of Fallot show a change in the right ventricular fibres with presence of a circular middle layer that resembles that of the left ventricle.Reference Sanchez-Quintana, Anderson and Ho6 In systemic right ventricles, circumferential over longitudinal free-wall shortening is seen, similarly to the left ventricle.Reference Pettersen, Helle-Valle and Edvardsen7 Owing to these factors, the assessment of right ventricular function by echocardiography is challenging and no single echocardiographic measure of function is devoid of limitations. Therefore, proper assessment should include various measures, taking into consideration different disease states. Despite these difficulties, there are guidelines for the functional assessment of the right ventricle by echocardiography in children.Reference Lopez, Colan and Frommelt8 In this review, we will go through the different systolic and diastolic indices, as well as their applications in practice. We will not review measures that assess global right ventricular function.
Systolic function
Qualitative assessment
The most commonly used method for the assessment of right ventricular systolic function is qualitative assessment or the “eyeball” method. Usually, abnormal function is reported as mildly, moderately, or severely diminished; however, the subjective estimation of right ventricular systolic function is highly dependent on the radiologist interpreting the study and can result in marked variability, and therefore poor inter-observer agreement.Reference Srinivasan, Sachdeva, Morrow, Greenberg and Vyas9 Moreover, qualitative assessment lacks sensitivity, as it does not distinguish well between mild-to-moderate and between moderate-to-severe systolic dysfunction when compared with cardiac magnetic resonance imaging studies.Reference Puchalski, Williams, Askovich, Minich, Mart and Tani10 Qualitative assessment should, therefore, be avoided as a single method or it should be used in conjunction with other quantitative measures and include several views to thoroughly examine right ventricular contractility.
Fractional area change
The fractional area change is a two-dimensional measure of right ventricular global systolic function. It is obtained from the apical four-chamber view, and is calculated as the difference in end-diastolic area and end-systolic area divided by the end-diastolic area (Fig 1).Reference Solomon, Skali and Anavekar11 The image must be optimised with focus on maximising the right ventricular area and clearly delineating the border of the endocardium in the setting of trabeculations, particularly the free wall, to ensure accurate tracing of the right ventricular cavity.Reference Mertens and Friedberg12 A percentage of fractional area change >35% is considered normal in adults.Reference Lopez-Candales, Dohi, Rajagopalan, Edelman, Gulyasy and Bazaz13 The fractional area change has been found to correlate with magnetic resonance-derived right ventricular ejection fraction, as well as to predict outcome in adult patients with myocardial infarction and pulmonary hypertension.Reference Anavekar, Gerson, Skali, Kwong, Yucel and Solomon14, Reference Anavekar, Skali and Bourgoun15 Fractional area change is preserved after pericardiotomy, as opposed to other measures of right ventricular function, such as tricuspid annular plane systolic excursion, which is discussed below, and therefore is a preferable method with which to assess right ventricular function post-operatively in the setting of an altered right ventricular contractile pattern.Reference Okada, Rahmouni, Herrmann, Bavaria, Forfia and Han5 Fractional area change has not been extensively used in congenital heart disease, but some studies have reported on its use. In patients with surgically repaired tetralogy of Fallot, for example, the right ventricular outflow tract is patched and often dysfunctional. Given that fractional area change does not include the contribution of the right ventricular outflow tract to ejection, it is thought that the fractional area change may overestimate global right ventricular function in this subset of patients.Reference Mertens and Friedberg12 On the other hand, in adult patients with repaired tetralogy of Fallot, fractional area change demonstrates good correlation with ejection fraction measured by cardiac magnetic resonance imaging, whereas correlations with other echocardiographic measures such as tricuspid annular plane systolic excursion, tricuspid S’ velocity, and the myocardial performance index are poor. Finally, fractional area change can predict impaired right ventricular function before and after pulmonary valve replacement in this patient population.Reference Selly, Iriart and Roubertie16 Fractional area change is reproducible in adult studies.Reference Shiran, Zamanian and McConnell17 Although normative values are lacking in the paediatric population, the fractional area change should be incorporated in paediatric studies, with attention to acquiring an adequate four-chamber view of the right ventricle for proper tracing of the endocardial borders.Reference Lopez, Colan and Frommelt8
Figure 1 The fractional area change is obtained by tracing the right ventricular endocardial border at end diastole and end systole. The difference in the area at end diastole and end systole is divided by the area at end diastole. This is a reproducible measure of function that is not affected by pericardiotomy. Normal fractional area change is above 35%. In this example, the fractional area change is diminished (26%), indicating impaired right ventricular function.
Tricuspid Annular Plane Systolic Excursion (TAPSE)
Tricuspid annular plane systolic excursion is another two-dimensional measure with which one can assess systolic right ventricular function. It is obtained by placing the M-mode cursor through the lateral portion of the tricuspid valve annulus in the apical four-chamber view. The excursion of the tricuspid valve from the base of the heart towards the apex is measured as the distance from the annulus to the apex at end diastole minus that distance at end systole. This distance can also be simply measured using two-dimensional imaging techniques, with equally reproducible results (Fig 2).Reference Forfia, Fisher and Mathai18, Reference Mercer-Rosa, Parnell, Forfia, Yang, Goldmuntz and Kawut19 Limitations include load and angle dependence, as well as the potential influence of the functional status of the left ventricle. Moreover, this measure does not take into account the contribution of the ventricular septum and/or the right ventricular outflow tract to right ventricular performance.Reference Jurcut, Giusca, La Gerche, Vasile, Ginghina and Voigt20, Reference López-Candales, Rajagopalan, Saxena, Gulyasy, Edelman and Bazaz21 Tricuspid annular plane systolic excursion has proven utility in the adult population with pulmonary hypertension and heart failure, with good correlation with other indices of systolic function, such as radionucleotide angiography.Reference Saxena, Rajagopalan, Edelman and Lopez-Candales22–Reference Dini, Conti and Fontanive24 Given the change in contractile pattern of the right ventricle in disease states, even after pericardiotomy for aortic valve replacement, this measure has poor correlation with ejection fraction measured by cardiac magnetic resonance and should be used with caution in paediatric conditions that result in right ventricular volume load and in single ventricles. Preferably, it should be reserved for the individual longitudinal follow-up in such cases.Reference Mercer-Rosa, Parnell, Forfia, Yang, Goldmuntz and Kawut19, Reference Avitabile, Whitehead, Fogel and Mercer-Rosa25–Reference Koestenberger, Nagel and Ravekes27 Similar to adults, this measure is especially useful in children with pulmonary arterial hypertension, where tricuspid annular plane systolic excursion values <15 mm are associated with a threefold event rate compared with those with normal values.Reference Forfia, Fisher and Mathai18, Reference Ghio, Klersy and Magrini28, Reference Sato, Tsujino and Ohira29 Normative values for age are available for the premature, neonatal, and paediatric populations, with normal values ranging from 0.9 to 2.5 mm.Reference Koestenberger, Ravekes and Everett30, Reference Koestenberger, Nagel and Ravekes31
Figure 2 Tricuspid annular plane systolic excursion can be obtained using two-dimensional imaging (a and b) or by placing the M-mode cursor at the level of the tricuspid valve annulus (c). The electrocardiogram tracing can be used to determine the optimal frame to be measured. This measure reflects the longitudinal shortening of the right ventricle. In adults, normal values are greater than 1.6 centimetres. Normative data for age are available for children. Tricuspid annular plane systolic excursion is helpful for the assessment of right ventriuclar function in patients with pulmonary hypertension and those with unoperated hearts. Although reproducible, this measure correlates poorly with ejection fraction by magnetic resonance and shodul be used with caution in congenital heart defects involving the right ventricle, such as single and systemic right ventricles, and tetralogy of Fallot.
Tissue Doppler-derived right ventricular systolic excursion velocity S’
Tissue Doppler systolic velocity of the tricuspid annulus is another measure of longitudinal right ventricular systolic performance, similar to tricuspid annular plane systolic excursion. This is a reproducible and easily obtainable measure of right ventricular systolic function with normal reference values available for the paediatric population (Fig 3).Reference Pavlicek, Wahl and Rutz32, Reference Koestenberger, Nagel and Ravekes33 A value <11.5 cm per second is associated with global right ventricular dysfunction with ejection fraction <45%.Reference Saxena, Rajagopalan, Edelman and Lopez-Candales22, Reference Pavlicek, Wahl and Rutz32, Reference Meluzin, Spinarova and Bakala34 The S’ velocity is diminished and inversely associated with end-diastolic volume in patients with repaired tetralogy of Fallot.Reference Ghio, Klersy and Magrini28, Reference Koestenberger, Nagel and Ravekes33, Reference Harada, Toyono and Yamamoto35, Reference Koestenberger and Ravekes36 Limitations of this technique are similar to tricuspid annular plane systolic excursion and include angle and load dependency.Reference Mertens and Friedberg12 An additional intrinsic limitation to tissue velocity imaging is a phenomenon called “tethering” by which the passive motion of the normal myocardium surrounding the diseased myocardium can result in falsely normal tissue velocities of the diseased segment in question.Reference Mertens and Friedberg12
Figure 3 Pulsed tissue Doppler can be used to calculate the tissue Doppler velocities of the tricuspid annulus. The S’ wave represents the systolic velocity, which assesses the right ventricular longitudinal function. The e’ wave represents the tricuspid annular early diastolic velocity, whereas the a’ wave represents the annular velocity with atrial contraction. E’ and a’ velocities are used in the assessment of right ventricular diastolic function.
Speckle tracking – strain and strain rate
Strain and strain rate together have recently emerged as a novel technique to quantitate myocardial contractility. Strain is defined as the degree of myocardial deformation compared with its original length, and is expressed as a percentage – where lengthening and thickening result in a positive value and shortening and thinning result in a negative value. Strain rate measures the rate of deformation, and is expressed as seconds−1. Its measurements can be obtained using Doppler and non-Doppler two-dimensional imaging, via speckle tracking or the tracking of natural myocardial acoustic markers (Fig 4). In addition, two-dimensional strain has the advantage of being angle independent. Strain also has the advantage of overcoming “tethering” or the movement of the diseased myocardium via pulling by the healthy myocardium surrounding it.Reference Mertens and Friedberg12 Secondary to their dependence on both extrinsic loading and intrinsic contractile force, strain and strain rate are load dependent, demonstrating increasing values in the setting of increased preload and decreasing values with increased ventricular size and afterload.Reference Sutherland, Di Salvo, Claus, D’Hooge and Bijnens37
Figure 4 Speckle tracking technology allows for the assessment of strain and strain rate as measures of the longitudinal function of the right ventricle. Image optimisation is required to demonstrate the entire right ventricular cavity in the imaging sector. Strain is a negative value, with more negative values indicating better function. Decreased strain appears to precede overt declines in right ventricular ejection fraction. Normal value in adults is −26±4%.
In the left ventricle, strain is a useful tool to assess regional myocardial function in ischaemic heart disease and in cardiomyopathies.Reference Sutherland, Di Salvo, Claus, D’Hooge and Bijnens37–Reference Weidemann, Wacker and Rauch41 It also appears to be a prognostic marker in adults with significant cardiovascular disease.Reference Kalam, Otahal and Marwick42 Strain has been used to assess right ventricular function in conditions such as pulmonary arterial hypertension, arrhythmogenic right ventricular cardiomyopathy, and tetralogy of Fallot.Reference Jurcut, Giusca, La Gerche, Vasile, Ginghina and Voigt20, Reference Teske, Cox, De Boeck, Doevendans, Hauer and Cramer43 It correlates well with ejection fraction measured by cardiac magnetic resonance, and may serve as a better predictor of outcome and functional capacity than ejection fraction.Reference Leong, Grover and Molaee44, Reference Park, Negishi, Kwon, Popovic, Grimm and Marwick45 In adults and children with pulmonary hypertension, lower strain, especially in the apical and free wall regions, has a strong correlation with mean pulmonary artery pressures, right ventricular size and function, exercise capacity, and mortality.Reference Dambrauskaite, Delcroix and Claus46–Reference Okumura, Humpl, Dragulescu, Mertens and Friedberg49 Global longitudinal strain is decreased in children and adults with repaired tetralogy of Fallot, and declines in strain appear to precede overt declines in right ventricular ejection fraction.Reference Weidemann, Mertens, Gewillig and Sutherland40, Reference Li, Xie, Wang, Lu, Zhang and Ren50–Reference Scherptong, Mollema and Blom53 Although strain appears to be reproducible, is angle independent, and easy to obtain, normative values are lacking, and therefore its utility at present lies in its use for individual patient follow-up.Reference Levy, Sanchez Mejia, Machefsky, Fowler, Holland and Singh54
Myocardial acceleration during isovolumic contraction
Myocardial acceleration during isovolumic contraction is a load-independent measure of ventricular contractile function. This Doppler-derived index is measured as the ratio of systolic velocity to the time to peak systolic velocity, with normal values in the right ventricular free wall of >1.1 m/s2 (Fig 5).Reference Jurcut, Giusca, La Gerche, Vasile, Ginghina and Voigt20 Although the isovolumic acceleration is heart rate dependent, and therefore useful to assess myocardial contractile reserve during exercise stress testing, normative data are available in children with adjustment for heart rate. Mean values in children are 2.3 m/s2 (range 1.1–3.6). Isovolumic acceleration is strongly associated with tricuspid annular plane systolic excursion and tricuspid myocardial systolic velocity (S’). Isovolumic acceleration is diminished in conditions such as transposition of the great arteries with systemic right ventricles, tetralogy of Fallot, and acute pulmonary embolism.Reference Cetiner, Sayin and Yildirim55–Reference Vogel58
Figure 5 Colour tissue Doppler interrogation of the tricuspid valve annulus for measurement of the isovolumic acceleration. Isovolumic acceleration is calculated by dividing the peak velocity at isovolumic contraction by the time from the onset of the wave to its peak velocity, also measured by tissue Doppler interrogation at the tricuspid valve annulus. Normal values are >1.1 m/s2.
Three-dimensional echocardiography
In brief, three-dimensional calculation of right ventricular volumes is possible with the use of specific softwares.Reference Khoo, Young, Occleshaw, Cowan, Zeng and Gentles59, Reference van der Zwaan, Helbing and McGhie60 This method typically involves manual tracing of the RV endocardial border in different planes, therefore adequate two-dimensional views of the right ventricle are required.Reference Leary, Kurtz, Hough, Waiss, Ralph and Sheehan61 Multiple studies comparing right ventricular volumes and ejection fraction with values obtained by cardiac magnetic resonance imaging show consistent underestimation of volumes measured by three-dimensional echocardiography. Importantly, this underestimation appears to be more pronounced as the right ventricular volume increases (Fig 6).Reference Khoo, Young, Occleshaw, Cowan, Zeng and Gentles59, Reference van der Zwaan, Helbing and McGhie60, Reference Grewal, Majdalany, Syed, Pellikka and Warnes62, Reference Niemann, Pinho and Balbach63 Nevertheless, due to its availability and cost-effectiveness, three-dimensional echocardiography is an attractive alternative to magnetic resonance imaging to assess right ventricular volumes in function. Further studies are required in order to establish normative data and improve accuracy in the paediatric population
Figure 6 Three-dimensional calculation of right ventricular volumes and ejection fraction. Three-dimensional image acquisition is performed from the apical four-chamber view in order to maximise visualisation of the right ventricle. Data sets are analysed offline using a dedicated software. Manual tracing of the endocardial border in three planes – sagittal, frontal, and coronal – is required for this calculation. The figure on the right depicts normal ejection fraction (61%).
Diastolic function
Assessment of right ventricular diastolic function should be included in routine echocardiograms, particularly in those performed for conditions in which diastolic dysfunction can precede declines in systolic function, such as in pulmonary arterial hypertension.Reference Shiina, Funabashi and Lee64 In tetralogy of Fallot, measures of diastolic dysfunction appear to correlate with risk factors for future re-interventions, such as greater degree of right ventricular dilatation.Reference Maskatia, Morris, Spinner, Krishnamurthy and Altman65
The assessment of right ventricular diastolic function is obtained by Doppler interrogation of the tricuspid inflow, tissue Doppler interrogation of the lateral tricuspid valve annulus, Doppler interrogation of the hepatic veins, assessment of the right atrium, as well as assessment of the size and collapsibility of the inferior caval vein. Owing to the different phases of diastole, the evaluation of diastolic function should include multiple parameters. Doppler signals should be obtained at end-expiration during quiet breathing – a task that can be quite challenging in children. Alternatively, five to seven beats should be acquired to account for the effect of respiration on the inflow velocities.Reference Zoghbi, Habib and Quinones66
Right ventricular diastolic dysfunction can result in increased right atrial pressures. As such, the collapsibility of the inferior caval vein might be decreased or absent. This is easily identified by two-dimensional imaging or by interrogating the caval vein using M-mode (Fig 7). Impaired right ventricular compliance and increased end-diastolic pressures can result in reversal of flow with atrial contraction in the hepatic veins and/or the inferior caval vein.
Figure 7 Inferior caval vein collapse can be identified by simple two-dimensional observation or using M-mode. The figure shows M-mode interrogation of normal inferior caval vein collapse with respiration, suggesting normal right atrial pressure.
Tricuspid inflow and myocardial velocities
The early, rapid filling phase of diastole is represented by the E-wave. The E-wave deceleration time reflects right ventricular relaxation. Atrial contraction occurs in late diastole, and is represented by the A-wave. The early diastolic tissue Doppler velocity, commonly denoted as e’, represents right ventricular relaxation. Increased E/e’ ratios represent increased right ventricular filling pressures. Normal values in adults have recently been published. E/e’ ratios <15 are considered normal.Reference Caballero, Kou and Dulgheru67 With normal diastolic function, the early filling velocity is higher than the atrial contraction velocity; therefore, reversal of the E/A ratio (<0.8) with increased deceleration time represents impaired ventricular relaxation. An accentuated relationship of the rapid filling and the atrial contraction velocities (E/A>2.1) with decreased deceleration time (<120 ms) represents restrictive physiology – a late phase of diastolic dysfunction. Short deceleration time helps discern between normal diastolic function and “pseudo-normalisation” – the intermediate phase of diastolic dysfunction characterised by preserved E/A relationship (E/A ratio between 0.8 and 2.1 with E/e’ ratio >6) (Fig 3). Increased deceleration time is directly associated with tau in subjects with pulmonary arterial hypertension.Reference Okumura, Humpl, Dragulescu, Mertens and Friedberg49
In summary, the echocardiographic assessment of right ventricular function is challenging due to limitations inherent to the right ventricle and due to paucity of normative data in children. Nevertheless, quantification of right ventricular function is possible and should be routinely performed using a combination of different measures, taking into account each disease state. Quantification is extremely useful for individual patient follow-up. Laboratories should continue to strive to optimise reproducibility through quality improvement and quality assurance efforts in addition to investing in technology and training for new, promising techniques, such as three-dimensional echocardiography.
Acknowledgements
We thank Yan Wang for selected images.
Conflicts of Interest
None.
Ethical Standards
The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national guidelines and with the Helsinki Declaration of 1975, as revised in 2008.
