Echocardiographic Recognition and Quantitation of Prosthetic Valve Dysfunction

Patient-Prosthesis Mismatch

Patient-prosthesis mismatch (PPM) is not an intrinsic dysfunction of the prosthesis. This problem occurs when the effective orifice area (EOA) of a normally functioning prosthesis is too small in relation to the patient’s body size and cardiac output requirements, resulting in abnormally high postoperative gradients.

The most widely accepted and validated parameter for identifying a PPM is the indexed EOA , which is the EOA of the prosthesis divided by the patient’s body surface area. Table 31.1 shows the indexed EOA cutoff values used to identify a PPM and quantify its severity and its prevalence according to severity and prosthesis position. ^, Lower cutoff values of indexed EOAs should be used to define PPMs for obese patients (see Table 31.1 ). Transthoracic echocardiography (TTE) is essential to differentiate PPM from intrinsic prosthetic valve dysfunction.

Prosthetic Valve Thrombosis and Pannus

Obstruction of prosthetic valves may be caused by thrombus formation (see Fig. 31.1A ), pannus ingrowth (see Fig. 31.1B ), or a combination of both. Pannus ingrowth alone may be encountered in bioprostheses and mechanical valves. It can manifest as a slowly progressive obstruction caused by a subvalvular annulus, in which case it may be difficult to visualize and distinguish from progressive structural valve deterioration (SVD). Pannus is usually encountered in patients with a normal anticoagulation profile and with subacute or chronic symptoms. ^,

Valve thrombosis should be suspected in a patient with any type of prosthetic valve who has had a recent increase in dyspnea or fatigue. Suspicion should be stronger if there has been a period of interrupted or subtherapeutic anticoagulation in the recent past. In such cases, echocardiography should be done promptly and should include transesophageal (TEE) studies, particularly if the prosthesis is in the mitral position. Valve thrombosis is most often encountered in patients with mechanical valves and inadequate antithrombotic therapy. Thrombosis may also be seen in surgical or transcatheter bioprosthetic valves, in which it most often occurs in the early postoperative period (3–6 months) (see Fig. 31.1E ). ^,

Subclinical thrombosis (i.e., valve leaflet thickening not associated with valve dysfunction or clinical symptoms) can occur in 5% to 25% of patients within the first year after transcatheter aortic valve replacement (TAVR), and it appears to be more common than after surgical aortic valve replacement (SAVR). In the PARTNER 3 randomized trial, [CR] the rate of subclinical thrombosis was 13% for the TAVR group versus 5% for the SAVR group ( P = 0.03) at 30 days and 27% versus 20%, respectively, at 1 year ( P = 0.19). Moreover, 50% of the positive cases at 30 days had spontaneously (i.e., without anticoagulation) regressed by 1 year, whereas a large proportion of the positive cases at 1 year were negative at 30 days.

The clinical significance of subclinical thrombosis is unclear. Some studies reported no association with clinical events, whereas others reported an association with cerebrovascular events. A meta-analysis of observational studies reported that subclinical thrombosis detected by multidetector computed tomography (MDCT) was associated with a 3.38-fold increase in the risk of cerebrovascular events.

Structural Valve Deterioration

Mechanical prostheses have excellent durability, and SVD is rare with contemporary valves, although mechanical failure (e.g., strut fracture, leaflet escape, occluder dysfunction due to lipid adsorption) has occurred with some models in the past (see Fig. 31.1C ). Bioprosthetic SVD is expressed clinically by the development of a progressive stenosis due to leaflet calcification or by regurgitation due to leaflet tear (see Figs. 31.1D and F ). SVD is the major cause of bioprosthetic valve failure, and the rate of reoperation for SVD has been as high as 30% at 15 years. ^,

The traditional definition of SVD is based on the composite of valve reintervention or death related to structural valve failure. This definition, however, underestimates the true incidence of SVD because it captures only the most severe cases of SVD associated with heart failure symptoms. A substantial proportion of patients with severe SVD may not undergo valve reintervention because they are considered to be at high risk for poor outcomes with reoperation or a valve-in-valve procedure. Some deaths may not be classified as valve related even though SVD might have directly or indirectly contributed to the death.

More sensitive and granular definitions of SVD based on echocardiography have been proposed, ^, and they include four stages: stage 0, no SVD; stage 1, morphologic SVD (i.e., morphologic abnormalities consistent with SVD but with no deterioration in valve hemodynamic function); stage 2, moderate hemodynamic SVD (i.e., occurrence of moderate prosthetic valve stenosis or transvalvular regurgitation during follow-up); and stage 3, severe hemodynamic SVD. In one study, the rate of stage 2 or greater SVD was 41%, whereas the rate of reintervention was 3.5% at 10 years after SAVR.

Infective Endocarditis

Prosthetic valve endocarditis is the most severe form of infective endocarditis. It occurs in 1% to 6% of patients with valve prostheses and accounts for 10% to 30% of all cases of infective endocarditis. Prosthetic valve endocarditis (PVE) is an extremely serious condition with high mortality rates (30%–50%).

Echocardiography, particularly TEE, plays a key role in the diagnostic and prognostic assessment of PVE because the diagnosis relies predominantly on the combination of positive blood cultures and echocardiographic evidence of prosthetic infection, such as vegetations, paraprosthetic abscesses, or a new paravalvular regurgitation. The diagnosis and management of PVE remains difficult, and a multidisciplinary approach is increasingly considered to be best practice. The team includes a consultant in cardiology and echocardiography with specialist competencies in valve disease. The incidence of PVE after TAVR is 1.1% per patient-year, which is similar to that after SAVR.

Timing of Echocardiographic Follow-Up

An initial TTE examination performed 6 weeks to 3 months after prosthetic valve implantation is recommended to assess the results of surgery and serve as a baseline for comparison if complications or deterioration occurs later. ^, Repeat TTE along with TEE is recommended for patients with prosthetic valves if there is a change in clinical symptoms or signs suggesting valve dysfunction. ^,

The American College of Cardiology/American Heart Association (ACC/AHA) guidelines suggest performing a TTE at 5 and 10 years and then annually thereafter, even in the absence of a change in clinical status. However, about 25% to 35% of patients with a bioprosthesis implanted for less than 10 years in the aortic position have some degree of valve degeneration or dysfunction at the Doppler echocardiographic examination. ^{, ,} The data support more frequent echocardiographic follow-up studies 5 years after implantation, as recommended in the 2009 American Society of Echocardiography (ASE) guidelines. For patients with mechanical valves, routine annual echocardiography is not indicated in the absence of a change in clinical status. Routine follow-up TTE after TAVR should include a baseline study (ideally within 30 days) at 1 year, and annually thereafter.

Clinical Data

The reason for the echocardiographic study and the patient’s symptoms should be clearly documented. Because the interpretation of Doppler echocardiographic findings depends on the type and size of the prosthesis, this information and the date of implantation should be incorporated in the report. The exact type and model of prosthesis should be recorded because the design and hemodynamic performance may differ substantially from one model generation to another. The approach used for surgical (e.g., standard sternotomy, ministernotomy, right anterior thoracotomy) or transcatheter (e.g., transfemoral, transaxillary, transcarotid, transapical, transaortic, transcaval) valve replacement should also be collected.

Blood pressure and heart rate should be measured at the time of echocardiogram. The patient’s height, weight, body surface area, and body mass index should be recorded to assess for the presence and severity of PPM and to index LV dimensions.

Recognition and Quantitation of Prosthetic Valve Stenosis

The appearance of a new murmur with new congestive heart failure symptoms in a patient with prosthetic aortic valve should prompt an urgent TTE study and, if indicated, a TEE study. However, the initial suspicion of prosthetic valve stenosis may be the incidental finding of abnormally high flow velocities and gradients detected during a routine examination. Tables 31.2 and 31.3 show the Doppler echocardiographic criteria for the recognition and quantitation of prosthetic valve stenosis. ^{, ,}

Leaflet Morphology and Mobility

Echocardiographic imaging should identify the sewing ring, the valve occluder, and the surrounding area. Imaging of the ball or disc of mechanical valves is often difficult to obtain because of reverberations and shadowing caused by the valve components. Left parasternal short-axis views and especially off-axis views are useful for assessing mitral prosthetic valve leaflet mobility. The leaflets of bioprosthetic valves normally appear to be thin, with unrestricted motion and no evidence of prolapse. Stentless homograft or autograft valves may be indistinguishable from native valves.

Prosthetic valve stenosis generally is associated with abnormal valve morphology and/or mobility ( Figs. 31.2–31.4 ; see Tables 31.2 and 31.3 ). In the case of mechanical valves, the mobility of the occluder is usually reduced or absent (see Fig. 31.2A ). Thrombus, pannus, or vegetations are often visualized at the level of the prosthesis ring or hinge mechanism ( Fig. 31.3D ). In the case of bioprosthetic valves, prosthetic valve stenosis is most often associated with thickening, calcification, and reduced mobility of the leaflets (see Fig. 31.2B ). Valve leaflet thickening is considered significant when the thickness of at least one leaflet is greater than 2 mm. Obstruction of bioprostheses may also be caused by thrombus, pannus, or vegetation (see Fig. 31.1 ).

Two- (2D) and three-dimensional (3D) TEE can improve assessment of leaflet mobility and detection of cusp calcification and thickening, valvular vegetations due to endocarditis, thrombus or pannus, and reduced leaflet, disc, or ball mobility (see Fig. 31.3 ). In the case of mechanical prostheses, evaluation of occluder mobility can be attempted with some degree of success, but in our experience, valve cinefluoroscopy is definitely the best, most economical, and least invasive technique that can be used for this purpose (see Fig. 31.4A ). MDCT may also be used to evaluate leaflet mobility of mechanical and bioprosthetic valves and to assess for pannus or thrombus in all types of valves and for valve leaflet thickening and calcification in bioprosthetic valves ( Fig. 31.4B –F).

TTE lacks sensitivity to detect subclinical thrombosis of bioprosthetic valves. Contrast MDCT is superior to TTE for this purpose and is useful for identifying markers of leaflet thrombosis, such as hypoattenuated leaflet thickening (HALT) and reduced leaflet motion (RLM) (see Fig. 31.4C –E). Given the dynamic nature of subclinical leaflet thrombosis and its absence of or weak association with clinical events, routine screening with contrast MDCT is not recommended. Noncontrast MDCT may be useful to identify bioprosthetic valve leaflet calcification, which is a marker of SVD (see Fig. 31.4F ). ^,

Transprosthetic Velocity and Gradient

The principles of interrogation and recording of flow velocity through prosthetic valves are similar to those used in evaluating native valve stenosis. This includes pulsed-wave (PW) and continuous-wave (CW) Doppler and color Doppler, using several windows for optimal recording and minimizing angulation between the Doppler beam and flow direction. CW Doppler evaluation of aortic prostheses must be performed from multiple transducer positions, including apical, right parasternal, right supraclavicular, and suprasternal notch. Measurements of prosthetic valve velocity and gradients are made from the transducer position, yielding the highest velocities. ^,

The fluid dynamics of mechanical valves may differ substantially from those of the native valve ( Fig. 31.5 ). The flow is eccentric in the monoleaflet valves and composed of three separate jets in the bileaflet valves. Because the direction of the jets across prosthetic valves may be eccentric, multiwindow CW interrogation is essential to obtain the highest transprosthetic velocity signal. Occasionally, an abnormally high jet gradient corresponding to a localized high velocity may be recorded by CW Doppler interrogation through the smaller central orifice of bileaflet mechanical prostheses in the aortic or mitral position (see Fig. 31.5 ). ^, This phenomenon may lead to overestimation of gradient, underestimation of the EOA, and therefore a false suspicion of prosthesis dysfunction.

Pressure gradient is calculated with the use of the simplified Bernoulli equation as follows: ΔP = 4 × V _Pr ² , where V _Pr is the velocity of the transprosthetic flow jet (in m/s). In patients with aortic prostheses and a high flow rate or narrow LV outflow tract (LVOT), the velocity proximal to the prosthesis may be increased and therefore not negligible. If proximal velocity (V _LVOT ) is greater than 1.5 m/s, estimation of the pressure gradient is more accurately determined by including V _LVOT in the following equation:

$\Delta \text{P}={4\text{×}(\text{V}}_{\text{PrAv}}^2-{\text{V}}_{\text{LVOT}}^2)$

Because of the pressure recovery phenomenon, the gradients measured by Doppler echocardiography may be higher and EOAs smaller compared with the values obtained by cardiac catheterization (see Fig. 31.5 ). The extent of pressure recovery is more pronounced in patients with a small aorta. It is important to distinguish between the pressure recovery phenomenon that may occur with any type of native or prosthetic aortic valve and the localized high gradient phenomenon that occurs in bileaflet mechanical valves (see Fig. 31.5 ).

Prosthetic valve stenosis is usually associated with increased transprosthetic peak flow velocity or mean gradient (≥3 m/s or ≥20 mmHg for aortic prostheses and ≥1.9 m/s or ≥6 mmHg for mitral prostheses) (see Tables 31.2 and 31.3 ). ^{, ,} However, a high velocity or gradient alone is not proof of intrinsic prosthetic valve obstruction and may be caused by PPM, high-flow conditions, prosthetic valve regurgitation, or localized high central jet velocity in bileaflet mechanical valves (data interpretation and differential diagnosis are discussed later). Significant prosthetic valve regurgitation may cause the flow rate across the prosthesis to increase, resulting in increased velocities and gradients.

Transprosthetic Jet Contour and Flow Ejection Dynamics

The contour of the velocity through the prosthesis is a qualitative index that may be used to assess prosthetic aortic valve function in conjunction with the other quantitative indices. In a normal valve, the contour of the CW flow velocity usually has triangular shape, with early peaking of the velocity and a short acceleration time (AT), which is the time from the onset of flow to maximal velocity (AT < 80 ms) ( Fig. 31.6 ). With prosthetic valve obstruction, a more rounded velocity contour is seen, with the velocity peaking almost in mid-ejection and prolonged AT (>100 ms) (see Fig. 31.6 and Tables 31.2 and 31.3 ). ^,

The main limitation of the AT is that it highly depends on chronotropy. To overcome this limitation, AT should be indexed to the LV ejection time (LVET); a ratio greater than 0.37 suggests prosthetic aortic valve stenosis. The advantage of these indices is that they are independent of Doppler beam angulation in relation to flow direction. They are, however, influenced by LV systolic function. For example, a patient with normally functioning aortic prosthesis and concomitant depressed myocardial contractility may nonetheless exhibit a rounded velocity contour with late velocity peaking, as has been witnessed in patients with severe PPM.