Multimodality imaging in valvular heart disease: how to use state-of-the-art technology in daily practice

Abstract

Our understanding of the complexities of valvular heart disease (VHD) has evolved in recent years, primarily because of the increased use of multimodality imaging (MMI). Whilst echocardiography remains the primary imaging technique, the contemporary evaluation of patients with VHD requires comprehensive analysis of the mechanism of valvular dysfunction, accurate quantification of severity, and active exclusion extravalvular consequences. Furthermore, advances in surgical and percutaneous therapies have driven the need for meticulous multimodality imaging to aid in patient and procedural selection. Fundamental decision-making regarding whom, when, and how to treat patients with VHD has become more complex. There has been rapid technological advancement in MMI; many techniques are now available in routine clinical practice, and their integration into has the potential to truly individualize management strategies. This review provides an overview of the current evidence for the use of MMI in VHD, and how various techniques within each modality can be used practically to answer clinical conundrums.

Open in new tabDownload slide

Introduction

Together with improved understanding on the mechanisms underlying valvular heart disease (VHD), technological advances in imaging and intervention during the last two decades have yielded significant innovation in the treatment of patients with VHD, but this has also increased the complexity of their management. Heart valve teams aim address these difficult decisions, by providing a standardized, high-quality, holistic evaluation of patients with VHD based on their clinical parameters (both specific to the VHD e.g. symptoms and comorbidities), biomarkers, invasive angiography, and haemodynamics, and, crucially, imaging. Echocardiography, including advanced techniques such as three-dimensional imaging and speckle tracking, remains the cornerstone of VHD assessment. However, the availability of other modalities such as computed tomography (CT), magnetic resonance imaging, and positron emission tomography (PET) have been recommended as minimum requirements for a comprehensive Heart Valve Centre. As such, the collaboration of cardiology and radiology within the Heart Team is crucial.

Multimodality imaging (MMI) plays a larger role in the management of VHD than simply providing a conclusive and comprehensive diagnosis where one imaging technique was inconclusive (Take home figure). It has the potential to offer insight into the pathophysiological drivers of disease and identify potential therapeutic targets for pharmacotherapy. Our understanding of the anatomy and geometry of dysfunctional valves and the severity of dysfunction is now more refined. Pathological changes to myocardial architecture and subclinical impairment predate overt ventricular remodelling and performance and negatively affect prognosis, challenging the guideline standard of waiting for reduction in left ventricular ejection fraction (LVEF) or ventricular remodelling to intervene in the absence of symptoms. Meticulous surgical and percutaneous interventional planning using imaging is improving patient selection, individualizing risk stratification, and improving procedural outcomes.

Native disease, of course, represents only one aspect of the vast subject of VHD; to discuss the role of MMI in the other important topics such as prosthetic valve disease and infective endocarditis would be beyond the scope of this paper. As such, this review focuses its aim in providing an overview of its additive value of multiparametric, MMI in the contemporary management of native VHD.

Aortic stenosis

Does my patient have severe aortic stenosis?

Discrepant echocardiographic grading of aortic stenosis (AS) severity is common, difficult to interpret and requires further assessment.¹ In the context of low-flow, low-gradient (LFLG) AS with reduced LVEF (<50%), low-dose dobutamine stress echocardiography (LD-DSE) is helpful not only to confirm AS severity but also to assess left ventricular (LV) recruitment and change in flow reserve, defined as an increase in stroke volume ≥20%.² Whilst the absence of LV flow reserve does not necessarily imply lack of improvement in LV function or clinical status following aortic valve replacement (AVR),³ it does connote increased surgical risk. Surgical AVR (SAVR) therefore comes with a weaker recommendation (IIb, level of evidence C)⁴ in such patients. However, prognosis in such patients is dismal with medical management alone, and as such transcatheter aortic valve implantation (TAVI) may be a more favourable option. Distinction between mild-to-moderate AS and paradoxical LFLG AS (with LVEF >50%) in the symptomatic patient can be challenging, but essential to guide management. Once measurement error has been excluded, and hypertension treated to reduced valvulo-arterial impedance, exercise or LD-DSE may also be helpful, to confirm a small valve area, and, in the case of exercise, provide an objective measure of symptomatic burden.⁵ Exercise stress echocardiography may be more often inconclusive owing to restrictive LV physiology.²

Echocardiographic planimetry, particularly using three-dimensional echocardiography (3DE), provides direct measure of AS severity. Computed tomography also offers the possibility of direct planimetry of the anatomical orifice area in AS, generating on average 0.2 cm² greater area as compared with echocardiography, but does not outperform echocardiography in area calculation or mortality prediction.⁶ Non-contrast CT AV calcium scoring (CT-AVC) is a highly reproducible, rapid, and low-risk alternative imaging modality used to adjudicate severity based on gender-specific cut-off values [>1300 Agatston units (AU) in women and >2000 AU in men].⁷ Computed tomography aortic valve calcium scoring also provides incremental prognostic information over echocardiography, and has been shown to be a powerful independent predictor of haemodynamic progression in AS, which may further individualize optimal timing of follow-up and/or intervention.⁸

One important caveat to this is bicuspidy. Shen et al.⁹ have recently reported that younger patients (<51 years) with bicuspid aortic valve (BAV) display less calcification, and the strong correlation between AV calcification and stenosis severity typical of older patients, or those with tricuspid aortic valve, is not seen. The authors postulate that this may be related to larger fibrotic, rather than calcific burden. This may, however, reflect a function of age, with a younger cohort being generally less susceptible to calcification in general. Novel imaging may provide insight into these inconsistencies. Hybrid imaging platforms such as PET-CT offer novel insights into the complex pathophysiological processes driving AS, including inflammation, calcium deposition, and ossification.¹⁰ Radiolabelled sodium fluoride (¹⁸F-NaF) has a powerful affinity for microcalcification, and is used to measure calcific activity, even before discernible calcification on standard CT imaging. Radiolabelled sodium fluoride may also localize to areas of valve degeneration and areas pre-disposed to progressive degeneration (Figure 1). Not only has ¹⁸F-NaF activity been shown to be a powerful predictor of disease progression and adverse events in AS and may also have the potential to identify therapeutic targets for novel pharmacotherapy.¹¹

My patient is asymptomatic—how can I optimally risk stratify?

Individualized risk stratification and timing of intervention can be a challenge. Current guidelines suggest that timing of intervention is dependent largely on the development of symptoms or reduction in LVEF.^{4  ,  12  ,  13} Symptomatic status is notoriously subjective, particularly in older or comorbid patients in whom traditional symptoms may be more difficult to ascertain. In the context of LV hypertrophy, reduction in LVEF is often a late phenomenon, occurring in the context of established tissue damage (fibrosis formation). Furthermore, SAVR in patients with reduced LVEF is associated with suboptimal perioperative course and postoperative improvements in LV function, symptoms and quality of life.¹⁴ Evidence from cardiovascular magnetic resonance (CMR) and myocardial deformation studies suggests that significant abnormalities in LV geometry, and myocardial architecture and function often predate reduction in LVEF.^15–18 Consequently, there is considerable interest in identifying robust and reproducible imaging biomarkers which may be used to assess myocardial health over time, and signify early or subclinical LV decompensation, to facilitate timely aortic valve intervention.

Speckle-tracking echocardiography (STE) is the gold standard for myocardial deformation or strain assessment. Feature- or tissue-tracking technology (using similar principals to STE) has been identified as being a potential alternative for use with CMR or CT, correlating well with STE,¹⁹ however, the latter remains superior owing to its higher temporal resolution, and lack of image interpolation. Global longitudinal strain (GLS) has potential as a powerful method of assessing subclinical LV dysfunction in multiple cardiac pathologies²⁰ (Figure 2). In asymptomatic severe AS with preserved LVEF, reduced GLS increases the risk of symptom development and therefore progression onto valve intervention compared with those with more preserved GLS.²¹ Ng et al.²² recently identified impaired GLS as an independent predictor of mortality in AS, irrespective of AS severity and LVEF. In a subgroup analysis of patients with asymptomatic severe AS, no survival difference was identified between those with impaired LVEF and those with preserved LVEF but reduced GLS. Indeed, recent meta-analysis evaluating the prognostic significance of reduced GLS in asymptomatic severe AS suggested >2.5-fold increase in mortality risk in patients with impaired GLS, even in those with LVEF >60%.¹⁵ It is important to recognize, however, that despite this compelling data, the widespread clinical use, and indeed its adoption into guidelines as a formal indication for treatment is limited by lack of standardization between vendors and overlap in values amongst those with health and disease.

Cardiovascular magnetic resonance has increased our understanding of the variable LV response in AS in terms of cavity remodelling, degree of hypertrophy, and amount of fibrosis.^{17  ,  18  ,  23} Late gadolinium enhancement (LGE) correlates with the degree of interstitial fibrosis on endomyocardial biopsy,²⁴ and according to recent large meta-analyses and multicentre studies, is present in half of patients with AS^25–27 (Figure 3A and B). Treibel et al.²⁸ have recently demonstrated that myocardial fibrosis in AS is heterogeneous and complex, comprising both diffuse reactive interstitial (reversible) and more focal replacement (irreversible) forms, often with a subendocardial-to-epicardial gradient. The latter suggests an ischaemic process, which may be related to supply-demand mismatch; indeed, reduced myocardial perfusion reserve identified with CMR has been well described but is as yet to find its place as a clinical decision-aid.^{29  ,  30} Whilst earlier data suggested that mid-wall (non-infarct type) LGE was associated with worst outcomes [eight-fold mortality increase compared with the six-fold increase seen in subendocardial (infarct type) LGE],¹⁷ more recent data suggest that simply the presence of LGE is an independent predictor of outcome, being associated with higher all-cause and cardiovascular mortality, irrespective of the type of intervention. Indeed, a 1% increase in LGE burden has been associated with an 11% increase in mortality hazard.³¹

T1 and extracellular volume (ECV) mapping better assess the more diffuse interstitial processes occurring within the myocardium (Figure 3C and D). Although data are less well established as compared with LGE, native T1 mapping holds promise owing to its speed, ease, reproducibility, and independence of the need for gadolinium-based contrast agents. Increases in native T1 correlate with histologically assessed diffuse fibrosis, and ventricular remodelling on CMR,^{24  ,  32} whereas CMR-derived ECV correlates with replacement fibrosis.³³ Tissue characterization using CMR has also enables the identification of precise drivers of LV impairment, which may impact on decision to intervene. The recognition of such concomitant diagnoses can only assist in risk stratification, and better identification as to whom is likely to derive benefit from intervention, thereby better informing management choices. Of particular note is the recently identified association between AS and cardiac amyloidosis, which should be considered particularly in LFLG AS, excessive hypertrophy, low electrocardiographic voltages, or relatively higher levels of biomarkers. Typical LGE patters, and elevated T1 values and ECV may certainly elevate the index of suspicion,³⁴ however, the use of bone scintigraphy is supported by recent expert consensus recommendations to non-invasively confirm the diagnosis.³⁵ Encouragingly, TAVI has been shown to improve outcomes in this dual pathology with no increased periprocedural complications and mortality compared with AS alone.³⁶ ‘Staging’ AS in order to guide management has been proposed, using the parameters described above, as well as more severe markers of disease such as left atrial or right heart involvement.^{37  ,  38} Tastet et al.³⁹ have described incremental value of such an approach over traditional clinical parameters, being associated with a 30% increase in risk of mortality per stage of disease. Guidelines, however, are yet to adopt this approach.

Current guidelines do not recommend AVR in patients with moderate AS unless undergoing cardiac surgery for other reasons,⁴ despite subclinical LV impairment being identifiable by sensitive techniques such as strain and tissue Doppler imaging. Impaired GLS has recently been shown to an independent predictor of both prognosis and the eventual progression to AVR, further highlighting the multifaceted nature of AS.⁴⁰ There is controversy regarding the impact of moderate AS on outcomes in patients with reduced LVEF^{41  ,  42}; recent data propose reduced GLS, rather than LVEF may be a greater determinant of survival in such patients.⁴³ The TAVR UNLOAD (Transcatheter Aortic Valve Replacement to Unload the Left Ventricle in Patients With Advanced Heart Failure) study aims to address such clinical conundrums.⁴⁴

My patient requires intervention—transcatheter aortic valve implantation or surgery?

With growing clinical evidence highlighting the feasibility and efficacy of TAVR across all surgical risk groups, there is growing awareness of the importance of CT to help guide decision-making between SAVR and TAVR. Computed tomography helps identify specific patient-specific anatomical features that connote an increased risk from TAVR. Findings such as severe protruding sub-annular calcification, low coronary ostial heights, and short membranous septum are all easily identified on CT and can be found independent of surgical risk. As well, CT is very helpful in characterization BAV disease allowing the identification of specific anatomical findings that drive significant risk in TAVR such as severe leaflet and raphe calcification. Ultimately decisions rest with the local multidisciplinary heart teams but over time the role of CT in informing these discussions and decisions has grown significantly.

Computed tomography is the gold standard for the assessment of the aortic annulus, root, aorta, and peripheral vasculature for access. The importance of the pre-procedural, intraprocedural, and post-procedural use of transoesophageal or transthoracic echocardiography (TTE) in patients undergoing TAVI is highlighted by consensus documents and registry data.^{45  ,  46} Intraprocedurally, the echocardiographer may be crucial in highlighting hostile anatomy, guiding wire and device positioning, assessing in real time for complications, and assessing post-procedural success. Intraprocedural TTE can be performed; however, transoesophageal echocardiography (TEE) is more optimal for annular sizing, assessment of coronary obstruction, positioning, and paravalvular leak.⁴⁷

Aortic regurgitation

Does my patient have severe aortic regurgitation and how severe is it?

Transthoracic echocardiography and TEE remain the gold standard for the assessment aortic regurgitation (AR); semi-quantitative and quantitative measures are the foundation of severity quantification. Taken together, TEE, Doppler, and 3DE assessments may overcome typical challenges associated with two-dimensional echocardiography (2DE), such as suboptimal alignment with eccentric jets, an inability to accurately identify the jet origin as it relates to the valve plane, and the potential non-hemispheric shape of the flow convergence zone.⁴⁸ Indeed, not only can it provide exquisite anatomical detail, but 3DE is considered the ideal imaging modality for the assessment of the effective regurgitant orifice area, at the most optimal imaging plane and phase within the cardiac cycle. Additionally, the integration of LV volumes derived from 3DE may enable an accurate estimation of stroke, and therefore regurgitant volume and fraction.⁴⁹

Unlike in AS, where peak velocity may be underestimated, CMR is recommended by both the ESC and AHA/ACC as a valuable complementary modality to quantify AR where echocardiography proves inconclusive,^{4  ,  12} and shows better correlation with 3DE, compared with 2DE.⁴⁸ Cardiovascular magnetic resonance can readily quantify regurgitant volumes (RegV) and fractions (RegF) through phase-contrast velocity mapping planned at the sinotubular junction level.⁵⁰ It is important to stress, however, that no formal cut-off values to define haemodynamically significant AR using RegV or RegF have yet been published. Certainly, the CMR-derived RegF required to predict symptoms or other indication for surgery for intervention is lower than echocardiographic cut-offs (CMR >30%, echocardiography >50%, Table 1).⁵⁶ Phase contrast may also be used to quantify holodiastolic retrograde flow (HDR), which has been shown to correlate with echocardiographically derived severity of AR. Early data suggest that CMR-derived HDR may also be an independent predictor of event-free survival.⁵⁷ Direct planimetry of the anatomical orifice area is possible using 3DE, CMR, or CT.⁵⁸ Using either a 3D-derived area of vena contracta (VC), or direct planimetry of the coaptation defect area, >0.5 cm² correlates with angiographically assessed AR severity,⁵⁹ standard 2DE measures of severity,⁶⁰ and CMR-derived RegF >30%.⁶¹

My patient is asymptomatic—how can I optimally risk stratify?

Left ventricular dilatation is considered essential to define severe AR.⁶² Cardiovascular magnetic resonance is the gold standard for LV volumetric assessment, and as such may be a more reproducible tool in monitoring LV dilatation. Mentias et al.⁶³ have recently reported increased mortality in patients undergoing surgery for severe AR in the context of LV end-systolic dimension of >2.0 cm², smaller than current guideline cut-off values for surgery. Where resource limitation precludes the liberal use of CMR, a trend towards such dimensions, or if comprehensive echocardiographic LV analysis (including strain) suggests a trend towards adverse remodelling, a more thorough assessment is warranted.

Subclinical LV dysfunction may again be assessed using GLS, which provides incremental prognostic data in asymptomatic individuals not meeting standard LV size or functional indications for surgery,^{64  ,  65} and, according to ESC guidelines, may be useful in guiding decision-making.⁴ Stress echocardiography may unmask latent symptoms and can be used to assess contractile reserve and RV function during exercise, which may aid decision-making in borderline cases.⁶⁶ Compared with AS, studies examining the role of imaging biomarkers of myocardial fibrosis in AR are smaller and less numerous. Nevertheless, there is a clear signal that elevated T1 time constants, increased ECV, and the presence of LGE convey adverse prognosis.^67–69 Their utility in patient selection for intervention, however, is yet to be determined.

Mitral valve disease

Why does my patient have mitral regurgitation?

Two-dimensional echocardiography and 3DE are the cornerstone of mitral valve assessment; for distinguishing aetiology, assessing severity, and determining suitability for valve repair, be it percutaneous or surgical (Figure 4). Volumetric and anatomical assessment using CMR supplements echocardiographic data (Table 1) and provides important information regarding myocardial viability and perfusion of ischaemic MR, where revascularization may improve LV performance and regurgitation. Similarly, CT may also provide an assessment of prior infarction and LV remodelling, with the spatial resolution required to provide accurate geometrical information to ascertain the mechanism of MR, without limitation in acoustic windows. Like TAVI, planning for transcatheter mitral valve replacement using CT improves risk stratification and procedural outcomes (Figure 5).

Calcific degenerative MV disease is a growing problem in our aging population, with degenerative mitral stenosis (MS) representing 12.5% of patients with MS.⁷⁰ It is incompletely understood in terms of pathophysiology, assessment of severity, and treatment indications. The underlying lesion is mitral annular calcification (MAC), which typically begins posteriorly, and progresses to involve the base of the mitral valve leaflets,⁷¹ limiting mobility and inducing distortion into the mitral valve geometry. Insights from ¹⁸F-fluoride and ¹⁸F-FDG PET suggest that, like AS, the process is far from being mere ossification. In fact, MAC comprises a vicious cycle of injury, inflammation, and calcification. Calcification appears to beget further calcification, evidenced by progression of disease more likely in those with higher baseline mitral annular calcium scores and ¹⁸F-fluoride activity.⁷²

Both MS and mitral regurgitation (MR) may occur. Direct planimetry using 2D TTE in degenerative MS may be challenging; 3D TEE is reported to be more accurate, reproducible, and feasible.⁷³ Mitral valve area (MVA) calculated by the continuity equation is the gold standard in calcific, degenerative MS,⁷⁴ but this may be limited by concomitant MR. Degenerative MS patients are older, with concomitant cardiac diseases such as, hypertension, renal dysfunction, and coronary artery disease, all of which decrease LV and atrial compliance and impair relaxation, which renders MVA by pressure half time invalid. Whilst degree of calcification correlates with mean mitral valve gradient, severe MAC does not necessarily translate to severe stenosis; a recent study of patients requiring haemodialysis suggested that measurable MV gradients were identifiable in 58% of patient with moderate to severe MAC. Extension of valve calcification may limit bileaflet motion and opening angles; moderately or severely reduced anterior mitral leaflet mobility induced a mean mitral valve gradient of 7 ± 3 mmHg compared with 3 ± 1  mmHg in those with normal leaflet mobility.⁷⁵

Visualization of the regurgitant jet, VC, or assessment of proximal isovelocity surface area may be challenging using 2DE due to acoustic shadowing, particularly in the presence of severe MAC. 3D colour Doppler can identify the number, location, and severity of regurgitant jets.⁷³ Cardiovascular magnetic resonance may help with volumetric assessment of MR severity and is more reproducible. Similar to AR, however, the RegV and RegF associated with adverse outcomes are lower than those derived by echocardiography,^{52  ,  76} with a recent consensus document providing formalized CMR-based grading of severity⁵³ (Table 1). Both non-contrast and contrast enhanced CT are used to assess the extent and location of MAC, and involvement of subvalvular apparatus, valve leaflets and myocardium, and also facilitates direct planimetry of MVA.⁷⁷

Intervention in degenerative MS is indicated for symptoms but may be challenging. Percutaneous balloon mitral valvuloplasty not suitable; surgery is often high risk due to comorbidity and surgical factors relating to the degree of calcification. TMVR, or ‘valve-in-MAC’, is a new treatment option for those whose risk profile precludes surgical intervention.⁷⁸ Computed tomography is fundamental for assessment prior to surgery to assess the risk of surgical debridement to surrounding structures. Prior to transcatheter device implantation, CT is the gold standard for quantifying the degree of circumferential MAC (e.g. >270° is required for anchoring for certain devices), assessing the hostility of calcification (e.g. the presence of protruding barb of calcium), and virtual valve implantation to prediction of left ventricular outflow tract obstruction post-device implantation, the latter being the most important and independent predictor of 30-day and 1-year mortality.⁷⁸

My patient is asymptomatic—how can I optimally risk stratify?

Key imaging biomarkers of risk have been identified in primary MR: reduced LVEF, impaired strain, and LV fibrosis. Current guidelines suggest LVEF <60% as a trigger for surgical intervention in asymptomatic primary MR.⁴ However, well-established registry data suggest that even minimal relative reduction in LVEF to 50–59% results in 80% increased mortality risk over a 10-year follow-up period compared with those whose LVEF remained above 60%.⁷⁹ More contemporary data support this; the Mitral Regurgitation International Database registry examined a large cohort of patients with flail leaflet: LVEF <45% conveyed a four-fold increase in all-cause mortality over a mean follow-up period of 9.3 ± 5.5 years, with increased mortality risk with decreasing LVEF [adjusted HR, 1.12 (1.02–1.20) per 1% LVEF decrement; P = 0.005].⁸⁰

Twenty per cent of patients with resting LVEF >60% develop LV dysfunction (<50%) postoperatively.⁸¹ Global longitudinal strain is reduced in asymptomatic patients with preserved LVEF,⁸² predicting subsequent reduction of functional capacity,⁸³ and postoperative LV impairment.⁸⁴ Indeed, Kim et al.⁸⁵ identified that a cut-off value for preoperative GLS of >18.1% predicted worse LV reverse remodelling postoperatively, and was independently and incrementally more predictive of cardiovascular and all-cause mortality, prompting the authors to propose surgical timing on a strain basis. It is important to note, however, that GLS assessment is load dependent,⁸⁶ and therefore cautious interpretation of individual patient results in the volume loaded state of severe MR is required.

Exercise echocardiography, as opposed to pharmacological stress, can unmask functional limitation in ‘asymptomatic’ patients (New York Heart Association class I–II) in 20–30% of cases, a finding which conveys increased risk of mortality and morbidity.⁸³ Furthermore, increasing severity of MR, impaired contractile reserve (either derived from LVEF⁸⁷ or GLS⁸⁸) and inducible increases in pulmonary arterial pressure to >60 mmHg⁸⁹ or RV dysfunction⁹⁰ all predict symptom development or surgical indication. Exercise may also expose increased ectopic activity or induce sustained/non-sustained malignant arrhythmia; these findings may indicate increased risk of sudden cardiac death in mitral valve prolapse when found on Holter monitoring.

The presence of LGE has been noted in 20% of patients with primary MR and is thought to reflect fibrotic remodelling caused by chronic volume overload. Location and type of LGE are variables, but conveys increased risk, either of worse postoperative outcome,⁹¹ sudden cardiac death (where LGE affecting the papillary muscle or inferolateral wall is most prevalent⁹²), or arrhythmia (associated with replacement fibrosis⁹³) ECV is also increased, correlating with other markers of LV dysfunction.⁹⁴ But like AS, these data are not yet incorporated into guidelines as indications to intervene but may facilitate aggressive watchful waiting, and facilitation of early surgery. Whilst early intervention in MVP may reduce arrhythmia burden, even in those with minimal MR, current data are of small cohorts,⁹⁵ and further study is required.

Determining whom will benefit from intervention in functional MR remains controversial, and of considerable interest. A comprehensive understanding of the interplay between the primary LV pathology and subsequent functional MR is essential, a concept described as the ‘proportionality’ of MR.⁹⁶ Accurate assessment of the regurgitant fraction is the most important determinant of severity in functional MR⁹⁷ and is reproducibly quantified by CMR, compared with echocardiography.

Tricuspid valve disease

When is multimodality imaging used in tricuspid regurgitation?

The negative clinical and prognostic implications of TR are well documented irrespective of aetiology, with surgical intervention often being high risk.^{98  ,  99} As such, there is increasing awareness of the need to address significant TR in a manner that comes with (i) an acceptable level of risk and (ii) a meaningful clinical outcome. Central to prescribing the correct management strategy (be it medical, surgical, or percutaneous) is a detailed understanding of tricuspid valve (TV) anatomy and function, RV size and function, and mechanism of disease.

2D TTE and TEE remain the principal methods of TR assessment; however, its use in anatomical delineation is limited by the non-planar nature of the annulus,¹⁰⁰ and the anatomical variation between individuals. All three leaflets are rarely seen in a single view; distinguishing between them can therefore be a challenge. 3D TTE and TEE provide insights into the complex and dynamic nature of the TV apparatus. The TV annulus is more asymmetrical than the MV and is highly dynamic.^{101  ,  102} Comparison with the normal annular dimensions reported by both the European Society of Cardiology and the American Society of Echocardiography is dependent on the view used, and subtle changes in angulation may significantly alter measurements. Two-dimensional echocardiography systematically underestimates annular dimensions by ∼4 mm compared with 3DE, CT, and CMR, highlighting the importance of 3DE for annuloplasty sizing.^103–105

Functional TR is related to annular dilatation (resulting in distortion of the typical configuration of the annulus) with leaflet remodelling and potential lack of coaptation, and distortion in subvalvular geometry, resulting in increased tenting angles and area.⁵¹ These features, as well as structural leaflet abnormalities are readily assessed using 3DE, and indeed may predict residual TR after surgical annuloplasty.¹⁰⁶ Leaflet impingement or perforation by implantable device leads can also be identified by 3DE, which may impact on future ability to deploy transcatheter devices.¹⁰⁷ The right heart is notoriously challenging to image using echocardiography on account of limited retrosternal acoustic windows. Three-dimensional right heart anatomy, function and morphology can be easily delineated using CT. Segmentation of the TV annulus can also be performed, and parameters such as tenting area and height can be measured.

Numerous transcatheter therapies have emerged as a potential alternative to medical therapy alone in those with significant TR. Comprehensive 3D imaging is paramount inpatient selection for such therapies. Spatial relationship of the TV apparatus to specific surrounding structures can be reproducibly described using CT in order to optimize outcome and minimize complications, e.g. right coronary artery course and position when considering annular reduction, vena cava characterization, orientation of the annular plane relative to the RV apex when considering spacer devices, and annular geometry and landing zone characteristics in the context of percutaneous transcatheter replacement.¹⁰⁸ Annular and RV segmentation is also feasible and reproducible using CT.

Timing of intervention remains controversial, but should be considered in the asymptomatic individuals with progressive RV dilatation and dysfunction.^{4  ,  13} Cardiovascular magnetic resonance is the gold standard for RV volumetric analysis, although in an experienced centre, 3DE volumetric analysis also performs well, despite a systematic relative underestimation of volumes of ∼20%.¹⁰⁹ Cardiovascular magnetic resonance-derived indexed RV end-diastolic volumes (RVEDV_i) predict RV dysfunction after TV surgery in a number of disease states, with an RVEDV_i of 150–170 mL/m² predictive of a lack of postoperative reverse remodelling.¹¹⁰

Pulmonary valve disease

When is multimodality imaging used in pulmonary valve disease?

Isolated pulmonary valve (PV) disease is rare. Pulmonary stenosis (PS) usually occurs in the context of congenital heart disease (CHD), whereas haemodynamically relevant regurgitation (PR) is typically acquired, and related to rheumatic heart disease, endocarditis, or carcinoid. The pulmonary valve is an anterior structure, and therefore may difficult to image echocardiographically, particularly with TEE. Detection of PR is principally made using colour Doppler. The pathological PR jet is typically broad and holodiastolic, and qualitative assessment of severity may be performed on the basis of the diameter of the jet at it origin, with increasing density of the CW Doppler envelope providing supporting evidence of worsening severity. It is important to note, however, that the colour jet area in severe PR may variable in duration and area.¹¹¹

Minimal amounts of PR are seen in the majority of healthy individuals; however, the finding of more than mild PR is rarely physiological, and a thorough assessment of predisposing conditions, such as pulmonary hypertension, or associated conditions or anomalies should be performed. Because of the association of PV disease and CHD, cross-sectional imaging is rapidly becoming indispensable, although much of the data available relates to patients following Tetralogy of Fallot (TOF) repair. The role of advanced cross-sectional imaging for PR is of increasing importance moving from diagnosis to treatment planning.

Cardiovascular magnetic resonance offers unlimited views of the heart with the ability to examine for concomitant CHD and quantify severity, function, shunting. As such, it is the gold standard for the comprehensive cardiac assessment of a patient with haemodynamically relevant PR.¹¹² Planning of through-plane phase-contrast velocity mapping imaging just distal to the valve tips provides rapid quantification of regurgitant fraction, which compares well to quantification using ventricular stroke volumes. A regurgitant fraction of 40% is considered severe using this method. Flow reversal in the branch pulmonary arteries is also easily demonstrated using CMR and is an indirect identifier of severe PR.¹¹³ In further, data are required to determine the clinical utility of 4D flow over standard CMR in this scenario.¹¹⁴ Similar to TR, surgery is indicated in patients with progressive RV dilatation, i.e. if RVEDVi is >150 mL/m².

Both CT and CMR offer multiplanar views of the right ventricular outflow tract (RVOT) and entire pulmonary trunk, which may be significantly abnormal in both PS and PR. Accurately defining the level and aetiology of stenosis may be challenging on TTE, and even more so on TEE, owing to the relative posterior location of the PV from the oesophagus. Computed tomography is indispensable for transcatheter pre-procedural planning to describe the landing zone characteristics, particularly in patients with calcified or stented pulmonary conduits or RVOT, in whom artefact may preclude meaningful assessment using CMR. Computed tomography is also superior for assessment of coronary artery anatomy and course, to delineate risk of compression during percutaneous valve implantation.¹¹⁵

Conclusion

The recognition of VHD as a complex syndrome highlights why MMI is crucial to optimized patient management. As we enter an era of emphasis on early disease identification, coupled with increasingly safer interventional options, the use of MMI in routine work-up is inevitable. Robust evidence that earlier intervention than is currently recommended with further drive the need for further study into MMI’s role in individualizing timing of intervention. Furthermore, MMI allows for safe, non-invasive bench-to-bedside transition of laboratory-based research to inform on previously unknown pathophysiological processes. Increased understanding of these processes may facilitate the development of novel treatments targeting newly discovered pathological pathways.

Conflict of interest: Dr Blanke and Leipsic provide institutional CT core lab services for Edwards Lifesciences, Abbott, Medtronic, Neovasc, Pi cardia. Dr Leipsic is a consultant to MVRX and Circle CVI. Dr Blanke is a consultant to CIrcle CVI, Abbott , MVRX and Edwards

References