Imaging assessment of the right atrium: anatomy and function

Abstract

The right atrium (RA) is the cardiac chamber that has been least well studied. Due to recent advances in interventional cardiology, the need for greater understanding of the RA anatomy and physiology has garnered significant attention. In this article, we review how a comprehensive assessment of RA dimensions and function using either echocardiography, cardiac computed tomography, and magnetic resonance imaging may be used as a first step towards a better understanding of RA pathophysiology. The recently published normative data on RA size and function will likely shed light on RA atrial remodelling in atrial fibrillation (AF), which is a complex phenomenon that occurs in both atria but has only been studied in depth in the left atrium. Changes in RA structure and function have prognostic implications in pulmonary hypertension (PH), where the increased right ventricular (RV) afterload first induces RV remodelling, predominantly characterized by hypertrophy. As PH progresses, RV dysfunction and dilatation may begin and eventually lead to RV failure. Thereafter, RV overload and increased RV stiffness may lead to a proportional increase in RA pressure. This manuscript provides an in-depth review of RA anatomy, function, and haemodynamics with particular emphasis on the changes in structure and function that occur in AF, tricuspid regurgitation, and PH.

Due to recent advances in interventional cardiology, the need for a greater understanding of right atrial (RA) anatomy has garnered significant attention. In fact, the RA and the interatrial septum serve not just as direct targets of interventions, such as transcatheter ablation of the cavo-tricuspid isthmus (CVTI) but also as the entry gate to access left heart chambers for a wide array of complex electrophysiology and structural heart interventions. Thus, optimal outcomes of intracardiac procedures mandate an excellent knowledge of RA anatomy, including recognition of common variants (Table 1).

The normal RA shows an ellipsoid shape, it is located superior to the right ventricle (RV), and anterior and lateral to the left atrium. It consists of a venous component, an appendage, and a vestibule.¹ The RA receives venous blood from the superior (SVC) and inferior (IVC) cava veins, the coronary sinus, and the Thebesian veins that drain the myocardium. During cardiac catheterization, the SVC is often the preferred site of entry into the RA. The SVC inserts upon the RA superior wall, with a mean mediolateral diameter of 20 ± 3 mm and a anteroposterior diameter of 19 ± 3 mm.²

Anatomically, the RA (Figures 1 and 2) includes five key structures: (i) crista terminalis (CT), (ii) RA appendage (RAA), (iii) CVTI, (iv) Eustachian valve (EV), and (v) orifice of coronary sinus and Thebesian valve.^1,2,4 The CT (Figures 1 and 2) is a muscular band which arises from the RA anteromedial wall, extending vertically along the SVC and IVC, with a mean length and thickness of 51 ± 9 mm and 5.5 mm, respectively.⁵ On occasion, the CT can be quite prominent, thereby raising clinical concern for an RA mass (pseudo-mass, tumour, thrombus, or vegetation). Due to the non-uniformity of the myocardial fibre arrangements within and outside the CT, this region is the electroanatomic focus of up to two-thirds of focal RA tachycardia cases^2,6 (Table 1).

The RAA (Figures 1 and 2) is derived from the ‘embryological’ RA, generally exhibiting a triangular shape, and forming most of the anterior and lateral RA walls. Typically, an extensive array of pectinate muscles spreads perpendicularly or obliquely from the CT, lining the internal surface of the RAA, resulting in a characteristic ‘corrugated’ surface appearance. Between pectinate muscles, the atrial wall may be very thin, comprising of only a few layers of myocytes between the epicardium and endocardium. One prominent pectinate muscle, the sagittal bundle, crosses the RAA wall transversely, dividing the RAA into the anteromedial and posterolateral portion. This large pectinate muscle (12 mm × 0.4 mm) is solitary in 55–65%, numerous (>1) in 20–25%, and absent in 15–20% of human hearts.⁷

The presence of a highly trabeculated appendage can herald challenges in transcatheter ablation cases. Anteriorly, the RAA ends with a pouch-like structure, which provides a stable position for implanting leads, reducing the risk of dislodgment. The junction between RAA and the RA is larger than the left atrial appendage orifice, which facilitates a more favourable ‘washing out’ of blood flow. This anatomical consideration likely explains the observation that formation of thrombi is less common in the RAA when compared with the left counterpart (Table 1).

The orifice of the IVC (Figures 1 and 2) is rounded and located in the RA posteroinferior region; its mean diameters are 24 ± 6 mm and mean area = 4.8 ± 2 cm². The EV, a remnant of the embryonic right valve of the sinus venosus, arises from the anterior border of the IVC orifice. In foetal life, the EV serves a critical role in channelling umbilical oxygenated blood from the IVC across the fossa ovalis to the systemic circulation. Following birth, the EV does not appear to have any physiological role, and with adulthood frequently regresses into a thin flap. Failure of reabsorption of the foetal EV may result in a redundant, mobile membrane that partly covers the IVC orifice, with the potential for complications during interventional procedures. If an substantial residual EV is present, the use of SVC access is suggested.⁸ In contrast, an incomplete, spot-like reabsorption may result in a mesh of thin filaments known as a Chiari network, which is present in ∼4.6% of the population, and is often found located close to the IVC orifice and the coronary sinus orifice. This structure is easily detectable by transthoracic echocardiography as a highly mobile and reflective structure.⁹ Notably, this structure may prevent a catheter from advancing freely into the atrium and limiting the range of the catheter manipulation.

The Eustachian ridge (26 ± 4 mm × 2 ± 2 mm) is an extension of the EV, which continues towards the central fibrous body, above the coronary sinus ostium. Excessive thickening of the Eustachian ridge is present in 47.9% of the hearts and may cause complete conduction block during atrial flutter, requiring total ridge ablation¹⁰ (Figure 2).

The coronary sinus (9–15 mm) drains into the posteromedial aspect of the RA, between the IVC and the right atrioventricular ostium (Figure 2). The coronary sinus (Figure 1) is used as a passage to the left heart epicardium during cardiac resynchronization therapy, catheter ablation of cardiac arrhythmias, defibrillation, and mitral valve annuloplasty.¹¹ An abnormally large diameter of the coronary sinus is an established risk factor for AV nodal re-entrant tachycardia. In nearly 80% of cases, its orifice is adjacent to the Thebesian valve. The shape of this valve varies from a clear crescentic flap (with or without fenestrations) to a thin strand-like, nearly invisible structure. Occasionally, an extensive imperforate flap over the ostium of the coronary sinus may make its cannulation challenging¹² (Table 1). If an obstructive Thebesian valve is detected by cardiac imaging prior to an intervention, access strategy to the coronary sinus should be modified. Radiofrequency energy has been used to perforate the Thebesian valve.

The CVTI is a quadrilateral-shaped, concave region demarcated anteriorly by the septal tricuspid leaflet attachment, and posteriorly by the EV. The CVTI is the prime target for ablation in an effort to interrupt the macro-re-entrant circuit of atrial flutter¹³ (Table 1).

The triangle of Koch, situated in the vestibule of the RA, is the key anatomic landmark in localizing the AV node. While the individual size of this landmark may be highly variable between individuals, general knowledge of the triangle of Koch dimensions is crucial when performing radio-frequency catheter ablation to avoid unintentional ablation of the atrioventricular node and subsequent complete heart block (Table 1).¹⁴

The interatrial septum and the fossa ovalis are two overlapping RA anatomic components (Figure 3). The interatrial septum is defined as the fibro-muscular area interposed between both atria. Two common phenotypic variants are the presence of a ‘lipomatous’ septum or a large septal aneurysm, both of which can be easily visualized by 2D echocardiography. The fossa ovalis is an oval/round concave region of 14 ± 4 mm × 12 ± 4 mm located in the infero-posterior aspect of the interatrial septum composed mainly of thin fibrous tissue. The fossa area (143 ± 65 mm²) typically enlarges with age. There are four variations of fossa ovalis anatomy as visualized from the RA perspective (Table 1).

RA morphology should be assessed prior to any interventional procedure involving a transseptal approach. Both cardiac computed tomography (CCT) and cardiac magnetic resonance (CMR) may adequately demonstrate the atrial septal anatomy prior to procedures; however, for both pre- and intra-procedural assistance, transoesophageal echocardiography (TOE) is the modality of choice (Figure 3). Specifically, 3D TOE may be used to generate an ‘en-face’ view of the interatrial septum, in real-time during interventions as guidance to optimize the site of transseptal puncture, and monitor thereafter. Given the feasibility, low cost, and real-time imaging characteristic, TOE remains the imaging modality of preference for interatrial septal assessment in most clinical settings.

RA size and function measured using echocardiography, CMR, and CT

A comprehensive assessment of RA dimensions and function using either echocardiography, CCT, or CMR acts as a first step towards a better understanding of its pathophysiology. The function of the RA is complex and can be divided into three phases: (i) a reservoir phase, during which it acts as reservoir for the vena caval blood returning during ventricular systole (atrial filling); (ii) a conduit phase, during passive RA emptying into the RV; and (iii) booster pump, during atrial contraction, that augments RV filling during late ventricular diastole¹⁵ (Figures 4 and 5B).

Sex, age, and ethnic differences in RA size and function

Previous studies have reported sex differences in RA size, with larger RA volumes noted in men when using both 2D^16–20 and 3D^16,20 echocardiography even after body surface area (BSA) indexing. Conversely, RA function parameters (3D emptying fractions, strain) have been shown to be higher in women.^16,20

With ageing, end-systolic RA volumes seem to remain stable by both 2D^18,20 and 3D echocardiography,¹⁶ or even decrease slightly.²⁰ In contrast, both 3D end-diastolic volumes and 3D pre-A RA volumes enlarge with age.^16,20 Also, 3D RA total emptying fraction as well as reservoir strain, passive emptying fraction, and conduit strain magnitude decrease with age, whereas the magnitude of RA contractile strain increases.^16,20 Inter-vendor variability is likely responsible for some of the observed variability in RA strain normal values.²¹ Other factors known to impact RA size and function include strenuous athletic activity²² and body mass index.²³

Until recently, most datasets informing normative RA size ranges, including those as published in the current guidelines²⁴ were primarily derived from White North American and European populations. This was somewhat problematic, as there was evidence to demonstrate that normal values for cardiac chambers differ by country and/or ethnicity, and accordingly, recent studies of large international cohorts were conducted to elucidate these differences.^25–27 Contemporary data from the World Alliance Societies of Echocardiography (WASE) study noted geographic differences in RA size.²⁰ Interestingly, Asians were found to have significantly smaller RA dimensions (both longitudinal and transverse) and RA volumes (both 2D and 3D end-systolic RA volumes), compared to non-Asians, even when adjusted for BSA. These results underscore the need for established normal values for RA size and volumes according to ethnicity.

Echocardiographic measurements of RA size and function

Results from several key echocardiographic studies examining techniques for RA size and function in populations of healthy individuals are listed in Table 2. Using 2D echocardiography, RA linear dimensions, areas, and volumes should be assessed from the apical four-chamber view²⁴ (Figure 5B). The minor-axis dimension should be measured from a plane perpendicular to the long axis of the RA, extending from the lateral border of the RA to the interatrial septum. RA volume is determined from this single-view using the area–length method and/or method of disks summation.^{16,28,32–34} Studies have demonstrated that RA volumes obtained using the area–length method are slightly larger than those obtained using the method of disks summation.^9,17,18,35

To avoid foreshortening of the RA chamber, a dedicated RA-focused view should be utilized for measurement of RA areas, volumes, and strain.³⁶ The recent development of specific 3D echocardiography software dedicated to quantifying the atria has improved the accuracy and feasibility of these measurements.³⁷

2D echocardiography

In the current echocardiographic guidelines issued by the American Society of Echocardiography (ASE) and European Association of Cardiovascular imaging (EACVI),²⁴ the normal values (mean ± standard deviation, or SD) of the indexed 2D RA major axis dimensions are 24 ± 3 mm/m² and 25 ± 3 mm/m² for men and women, respectively. For BSA-indexed 2D RA minor axis, the mean value is 19 ± 3 cm/m² for both sexes.^24,30 The reported normal values of the indexed 2D end-systolic RA volume (by area–length method) is 25 ± 7 mL/m² and 21 ± 6 mL/m² for men and women, respectively.^16,18,24 Studies using the single-plane method of disks have reported lower values for indexed 2D end-systolic RA volume. For example, in a large European multicentre study,¹⁸ 22.5 ± 6.5 mL/m² and 19.0 ± 6.2 mL/m² were reported for men and women, respectively. Similarly, in a large worldwide study,²⁰ 20.6 ± 6.4 mL/m² and 18.0 ± 5.3 mL/m² were reported for men and women, respectively.

3D echocardiography

No partition values of 3D RA volumes are reported in the current ASE/EACVI guidelines.²⁴ However, recently, 3D RA volumes were found to be larger than their corresponding 2D RA volumes.^16,20 In 200 healthy European subjects, the mean/SD indexed 3D end-systolic RA volume was reported to be 31 ± 8 mL/m² for men and 27 ± 8 mL/m² for women.¹⁶ In a large multicentre international cohort, the indexed 3D end-systolic RA volume was 23.2 ± 7.4 mL/m² for men and 20.5 ± 6.4 mL/m² for women.²⁰ No definitive normal values for RA function using either 2D or 3D are available in the current ASE/EACVI guidelines.²⁴ In single-centre studies, RA reservoir strain was reported to be 42 ± 9% and 45 ± 10% for men and women, respectively,⁵ while another study reported similar values of 44.6 ± 12.5% and 47.0 ± 13.4%, respectively.²⁰ With 3D echocardiography, total RA EmF has been reported in two studies to be 61 ± 8% and 52.5 ± 6.5% for men and 65 ± 8% and 53.3 ± 6.9% for women.²⁰

CMR and CT measurements of RA size and function

The technical strengths and limitations of non-invasive imaging modalities currently used for RA assessment are summarized in Table 3. CMR is the accepted gold standard for cardiac chamber volumetric assessment, providing both structural and functional information without associated ionizing radiation, and greater temporal resolution when compared with CCT. Yet, CMR has lower spatial resolution and requires longer acquisition times than CCT. A steady-state free precession is the most used technique in CMR for the acquisition of cine-images of the RA. RA area, longitudinal, and transverse diameters are typically obtained from the four-chamber and the RV two-chamber views, which are routinely acquired during CMR exams. RA volumes can be measured in these views by using the biplane area–length method³⁸ (Figure 6). The modified Simpson’s method and 3D-volumetric modelling can be performed by tracing the endocardial border in all contiguous slices on a stack of RA short-axis cine images using a slice thickness between 5 and 8 mm but comes at the expense of prolonged post-processing times. More precise 3D volume estimation by CMR can also be obtained with dedicated software that is capable of auto-detecting endocardial borders.³⁹ IVC and SVC contours should be excluded from the tracing, as well as the appendage.^38–41 Normal ranges for RA size vary significantly between methods. The area–length method is faster, but less reproducible compared to other volumetric methods. Volumetric methods are more accurate and provide larger values than the area–length method.

Intra-observer variability has been reported at 3.5% for RA volumes, 3.6% and 1.8% for areas in the two- and four-chamber views, and ∼4% for longitudinal and transverse diameters in the four- and two-chamber views, respectively. Inter-observer variability has been reported to be 3.9% for RA volumes, 5.2% and 5% for areas in the two- and four-chamber views, respectively, and 5.5% for longitudinal and transverse diameters in the four- and two-chamber views, respectively. RA reference values indexed to BSA are similar between men and women. There is a minimal decrease in RA size with age in healthy subjects, with some studies indicating that ethnicity should be interpreted in context as well.^39,40 There is moderate correlation and wide limits of agreement between MRI and CCT, ranging from ±13 to ±29 mL for RA volumes.⁴² Normal reference values for RA size by CMR are listed in Table 4. Age-related changes in blood flow and geometry of the caval veins and RA have been recently described using 4D flow CMR. These age-related changes have a significant impact on the haemodynamics of the RA inflow tract. In young subjects, blood flow of the RA showed a clockwise rotating helix without signs of turbulence. In contrast, this rotation was absent, and turbulences were more frequent in older subjects.⁴³

Reference values of RA size obtained using CCT are limited and not universal. End-systolic RA volume using 64-slice CCT in 103 healthy subjects were found to be 111.9 ± 29 mL with a reference range of 54.9–168.9 mL by Lin et al.⁴⁴ In the study by Takahashi et al.⁴⁵ with 320-slice CCT and semi-automated 3D segmentation technique, maximum RA volume was 82.1 ± 44.1 mL. Standard four-chamber and right ventricular two-chamber views are used for linear measurements. Linear measurements and volume calculations are performed in a similar way to CMR, including longitudinal, transverse and sagittal diameters, direct volume measurements (Simpson’s method), area–length method, and ellipsoid method for volume quantification.⁴⁶ Artefacts due to peculiarities of contrast use and excess cardiac motion may lead to errors in volume calculations. Therefore, tailored injection protocols are advised for an adequate visualization of the right atrium (RA). On the other hand, the excellent spatial resolution of CCT enables detailed anatomic evaluation of the RA, especially when performed with electrocardiographic gating (Figure 7). Disadvantages of CCT include exposure to ionizing radiation and iodinated contrast agents. CCT contrast protocols are mostly limited to diastole, while optimal assessment of RA size, remodelling, and systolic function requires full-cycle retrospectively gated acquisition. Arrhythmias can compromise the accuracy and reproducibility of the quantification and increase radiation exposure. It is also important to emphasize that currently, there is no clear consensus regarding which of the imaging modalities is superior in the assessment of RA dimensions.

The RA in atrial fibrillation

In the realm of atrial fibrillation (AF), most of the echocardiographic literature has focused on the phenomenon of left atrial remodelling^47–50 and incident stroke prediction.^51–54 This may be somewhat monolithic, as it is evidence that AF involves both atria, but the associated pathophysiology of the RA in AF has yet to be thoroughly investigated. The purpose of this section is to describe the different purported mechanisms regarding the role of RA in the pathogenesis of AF.

AF and RA remodelling

Atrial remodelling in AF is a complex phenomenon. Structural, electrical, and metabolic RA remodelling have all been described as responsible for the link between RA and AF. In a population free of clinical cardiovascular disease, CMR-derived RA volumes were independently associated with incident AF, even after adjusting for conventional cardiovascular risk factors and left atrial parameters.⁵⁵ Interestingly, following the restoration of sinus rhythm in AF, anatomical, and/or functional reverse remodelling has been noted in both the left and right atria.^56–59 In a population of endurance male athletes, the presence of higher RA volumes, and lower reservoir strain were associated with increased risk of paroxysmal AF.⁶⁰ In animal models of AF where remodelling was initiated by endurance exercise⁶¹ and/or pulmonary hypertension (PH)⁶² both left atrial and RA fibrosis were present and associated with increased AF susceptibility. Interestingly, investigators have described a subtype of AF originating from the RA,⁶³ which appears to be accompanied by specific RA structural remodelling patterns including a large RA appendage, less remodelled left atrium, and lower burden of epicardial adipose tissue surrounding the atria.⁶⁴

Besides RA structural remodelling, RA electrical remodelling has also been described. In a murine model of PH with AF, studies demonstrated that RA conduction slowing was coupled with greater AF inducibility when compared with controls.⁶² Additional studies described paroxysmal AF initiated by RA ectopy and maintained by a re-entrant circuit localized in the RA (so-called RA fibrillation).⁶³ In patients with chronic obstructive pulmonary disease (COPD) and concurrent AF, electrophysiological studies demonstrated that the slowing of RA conduction and a higher prevalence of non-pulmonary vein foci were localized in the RA.^65–67 In patients with persistent AF and left ventricular (LV) systolic dysfunction who underwent successful catheter ablation, long-term RA electrical remodelling analysis showed a recovery in the electrical atrial changes.⁵⁸

Molecular analyses in AF have shown the expression of specific metabolic markers in the RA. In rats with PH and AF, activation of hypertrophic, proinflammatory, and profibrotic pathways were noted in the RA, akin to animal models of RA remodelling caused by LV dysfunction due to myocardial infarction.⁶² In a different study using the same rat model of PH and AF, the inflammation-resolution promoting molecule (RvD1) showed suppression of AF promotion, and attenuation of atrial fibrosis.⁶⁸ Other studies have shown that sustained adenosine-induced AF was driven by localized re-entry circuits in areas of the RA with the highest expression of adenosine A1 receptors.⁶⁹ Of note, some metabolic general disorders described in patients with AF were derived from analysis primarily focused upon the RA or RA appendage.^70–73

Recently, Hiram et al.,⁷⁴ used a monocrotaline-treated rat model demonstrating that the induced right heart disease produces a substrate for AF maintenance due to RA re-entrant activity, with an underlying substrate prominently involving RA fibrosis and conduction abnormalities. The authors emphasize some of the interesting differences compared with the AF substrate in LV dysfunction post-myocardial infarction. These differences provide new mechanistic insights and point to new approaches and tools for therapeutic interventions.⁶²

Aetiologies of RA remodelling leading to AF

As described in the pathophysiology of the RV,⁷⁵ RA remodelling occurs across a spectrum of clinical settings (i.e. pressure overload, volume overload, and intrinsic cardiomyopathy) leading to an underlying arrhythmogenic substrate and AF.

The pressure overloaded RA and AF

Several pulmonary diseases can induce RA pressure overload which may lead to RA structural, electrical, or metabolic remodelling, and subsequent AF⁷⁶ (Figure 8).

Atrial arrhythmias are frequently encountered in PH and appear to connote a worse clinical outcome. In clinical studies, the prevalence of AF in PH was estimated to be as high as 31%.⁷⁷ Moreover, elevated RA pressure and enlarged RA size have been described as established risk factors for adverse outcomes or death in PH.^78,79 In patients with PH and AF, New York Heart Association (NYHA)/World Health Organization (WHO) functional class, 6-min walking distance, NT-pro-B-type natriuretic peptide concentration, and renal function were significantly compromised compared to patients with PH and sinus rhythm.⁷⁷ Persistent AF has been associated with the risk of death from RV failure in PH patients.⁸⁰

COPD is another disease state well-associated with an increased risk of AF. In a large, retrospective cohort of patients on home oxygen admitted for COPD exacerbation, the prevalence of AF was 18.2% and served as a risk predictor for in-hospital death.⁸¹ In a multicentre cohort study, COPD was associated with AF and death.⁸² Reductions in forced expiratory volume in 1 s and obstructive respiratory disease were associated with higher incidence of AF after adjustment for measured confounders.⁸³ Of note, patients with AF and concomitant COPD have a more severely impaired RA function compared to AF patients without COPD.⁸⁴

Additional diseases that induce right heart disease are correlated with AF including obstructive sleep apnoea with conduction abnormalities related to connexin dysregulation and fibrosis^85,86; asthma and lack of asthma control.⁸⁷ Finally, recent data also underscored the link between AF and the risk of pulmonary embolism and deep vein thrombosis during the first months after AF diagnosis.⁸⁸ Gaynor et al., used a dog model of RA and RV pressure and volume overload and recorded findings after 3 months of progressive pulmonary artery banding. The authors demonstrated that RV systolic pressure was usually preserved, but diastolic function became impaired. To compensate, RA contractility increased, and the RA become more distensible to maintain filling of the stiffened ventricle. This compensatory response to the RA likely plays an important role in preventing clinical failure in chronic PH.⁸⁹

The volume-overloaded RA and AF

The presence of functional tricuspid regurgitation (FTR) is a key element in RA volume overload states resulting in RA structural, electrical, or metabolic remodelling, which may lead to AF (Figure 8).

Patients in AF without severe FTR have been described as having RA and tricuspid annulus remodelling, independent of the presence of left heart disease.⁹⁰ Accordingly, it has been hypothesized that FTR could be either considered as the cause or consequence of RA dilatation and AF development.⁹¹ In patients in AF with FTR, a significant correlation has been demonstrated between FTR severity and RA volume and tricuspid annulus area.⁹² Irrespective of cardiac rhythm and RV loading conditions, RA volume is a major determinant of tricuspid annulus area in patients with FTR, and RA enlargement is an important mechanism for tricuspid annulus dilatation.⁹³

Two distinct mechanisms should be noted for FTR in AF: (i) in ventricular FTR, a combination of remodelling and dysfunction of the RV, dilatation of the tricuspid annulus, and tethering of the tricuspid valve leaflets; and (ii) atrial FTR in patients with normal RV geometry and function but with dilated tricuspid annulus and RA (Figure 8).

Ventricular functional tricuspid regurgitation

Ventricular FTR may induce RA remodelling leading to AF (Table 5). In a healthy population, morphologic analysis of the normal RV using 3D echocardiography-derived curvature indices has been reported in strategies to describe differing RV morphologies with distinct implications in heart failure, PH, and left-to-right interatrial shunt.⁹⁴ In PH, RV elongation, and elliptical/spherical RV deformation induced valvular tethering leads to ventricular FTR (Figure 9).⁹⁶ Other studies have reported the role of tricuspid annulus dilatation and/or papillary muscles displacement in the development of ventricular FTR,⁹⁶ as well as an observation of a more circular tricuspid annulus shape in comparison with healthy subjects.⁹⁷

Atrial functional tricuspid regurgitation

Though more recently described, atrial FTR (Table 5) is probably the cornerstone of RA remodelling in AF (Figure 9). In cases of atrial FTR, 3D echocardiography has demonstrated a larger tricuspid annular area, weaker annular contraction, smaller tethering angle, and similar leaflet coaptation status when compared with patients with FTR and left-sided heart disease.⁹⁸ Interestingly, tricuspid annular area in mid-systole was independently associated with atrial FTR and the tricuspid annulus area in atrial FTR was more closely correlated with the RA volume than the RV volume.⁹⁸ Thus, in contrast to ventricular FTR, dilatation of the tricuspid annulus appears to be the primary cause of regurgitation in atrial FTR.⁹⁹

The cardiomyopathic RA and AF

AF is the primary example of autonomous disease affecting both atria and is responsible for the development of left and RA cardiomyopathy. Few data are available on the existence of an intrinsic RA cardiomyopathy in conjunction with other cardiovascular diseases. This compelling but preliminary hypothesis should be evaluated further investigations in different clinical settings (e.g. myocardial infarction, arrhythmogenic RV cardiomyopathy, amyloidosis, myocarditis, cardiotoxicity) due to different conditions, such as chemotherapy. RA remodelling in heart failure with preserved ejection fraction is associated with higher prevalence of AF, mild tricuspid regurgitation, severe PH, and impaired reservoir function.¹⁰⁰

The RA in PH

In the early phase of PH, the increased RV afterload induces RV remodelling which is mainly characterized by hypertrophy. As PH progresses, RV dysfunction and dilatation, which represent the main maladaptive features of this disease, may begin and eventually lead to RV failure.¹⁰¹ RV overload and increased RV stiffness then leads to a proportional increase in RA pressure. In addition, RV dilatation can cause FTR through annular dilatation and leaflet tethering, which in turn may further increase RA pressure.⁹⁹ As such, increased RA pressure determines RA remodelling, dysfunction, dilatation, and eventually failure with worsening of RV filling, pulmonary perfusion, and systemic congestion (Figure 5A).

Non-invasive cardiac imaging and prognostic role of right atrial size and function

Current guidelines recommend the evaluation of RA size and pericardial effusion as the only imaging markers for risk stratification in PH.¹⁰² A dilated RA has been associated with poor prognosis in patients with PH.¹⁰³ Indeed, in a small cohort of 25 consecutive patients with PAH, RA area, and presence of FTR were independently related to mortality or eventual lung transplantation during a mean follow-up of 29 months.¹⁰⁴ Furthermore, in a cohort of 66 consecutive PH patients, RA volume was positively associated with RV afterload assessed by echocardiography (i.e. pulmonary artery systolic pressure).¹⁰⁵ Therefore, RA area quantified on a four-chamber view with either echocardiography or CMR, is suggested in the current guidelines as one of the criteria for risk stratification in patients with PH. The proposed cut-off values of 18 and 26 cm² for RA area should be used to identify patients with PH at intermediate- (5–10%) or high risk (> 10%) of 1-year mortality, respectively. It is prudent to emphasize that these recommendations are based on relatively small studies. In addition, the proposed cut-off values are predominantly derived from expert consensus opinion, and they differ notably from the cardiac chamber quantification guidelines which recommend an indexed RA volume with the following sex-specific cut-off values for RA dilatation of 39 and 33 mL/m² for males and females, respectively.²⁴ The PH guidelines also suggest assessment of RA area for follow-up risk stratification, evaluation of the effect of specific therapies and to eventually intensify PH-specific treatments. More advanced measures of RA size with 3D echocardiography, such as the 3D RA minimum volume combined with the 3D assessment of the left atrial size, showed to be useful to distinguish between pre-capillary and post-capillary PH,¹⁰⁶ therefore, having potential implications on choice of PH treatment.¹⁰²

In addition to size, RA function appears to play a pivotal role in the prediction of the symptom development as well as prognosis in patients with PH.¹⁰⁷ In 65 patients with PH from the IMPRES (Imatinib in Pulmonary Arterial Hypertension) trial, RA reservoir and conduit function assessed with longitudinal strain were impaired in patients with PH compared to healthy controls. In the same study, RA reservoir function was associated with RV enlargement and dysfunction as well as N-terminal pro-brain natriuretic peptide (NT-proBNP), independently of RA size and pressure.¹⁰⁸ This suggests an incremental value of RA function over RA size parameters in assessing decompensation in PAH.¹⁰⁸ Regarding prognostic value, prior work demonstrated an association of all RA speckle tracking echocardiography strain-derived phasic functions with hospitalizations or death during a median follow-up of 44 months in a prospective cohort of 63 consecutive patients with PAH.¹⁰⁹ Similarly, in a cohort of 80 PAH patients, there was an association observed between CMR feature tracking-derived RA function parameters and haemodynamic status, with RA conduit function demonstrating the highest predictive value for adverse events.¹¹⁰ Furthermore, a recent study showed an increase of RA reservoir and conduit function assessed with CMR after balloon pulmonary angioplasty in a cohort of 29 patients with chronic thromboembolic PH, representing RA functional recovery. These findings correlated well with a decrease in pulmonary vasculature resistance and NT-proBNP levels and was associated with a significant improvement in functional capacity, thus confirming the role of RA in the pathophysiology of PH-related symptoms.¹¹¹

RA peak longitudinal strain has additive prognostic usefulness to other clinical measures, including RV strain, RA area, and RA pressure, in patients with PH. RA mechanical function by strain imaging has the potential for clinical applications in adult and paediatric patients with PH.^112–114

Future perspectives

The importance of RA size and function has so far been predominantly investigated in patients with pre-capillary PH using echocardiography. Evaluation of the RA in different patient cohorts, such as patients with combined pre- and post-capillary PH or the evaluation of PH-specific treatments on RA morphology and function using additional imaging modalities, such as CT represent important areas for future investigations.

Clinical implications of a comprehensive assessment of the RA

The RA is the cardiac chamber that has been least well studied, even though recent data have shown that changes in RA structure and function have prognostic implications in specific cardiac diseases such as heart failure and PH. In addition, the recently published normative on RA size and function will likely shed light on the role of RA volume in tricuspid annular dilation, AF, and FTR.

Conflict of interest: Three of the authors are on speakers’ bureaus of three different companies: RML and FFF for Philips Healthcare, and LES for General Electric. In addition, M.C. has served as a consultant for Novo Nordisk and Astra Zeneca.

References