Echocardiographic evaluation of prosthetic heart
valves
Delineation of normal prosthetic valve function is usually possible with
transthoracic echocardiography. Despite the prominent acoustic shadowing that
accompanies mechanical prostheses, a Doppler signal that demonstrates normal
transprosthetic flow velocity and flow duration is usually sufficient to
exclude a stenotic valve. Exclusion of an incompetent valve is often more
difficult, especially for mechanical prostheses in the mitral position.
Transthoracic Doppler echocardiography provides important hemodynamic
information with regard to prosthetic valve pressure gradients and is
extremely useful in long-term follow-up. The "normal gradient"
across a prosthetic valve depends upon the type, size, and position of the
prosthesis as well as the cardiac output; guidelines are available for the
acceptable range of Doppler gradients encountered in properly functioning
valves [
1,2].
Acoustic shadowing caused by the prosthetic material may limit transthoracic
visualization of prosthetic leaflets, vegetations, and thrombi. In addition,
while prosthetic aortic valve regurgitation is usually well visualized on
surface color Doppler imaging, prosthetic mitral regurgitation is frequently
undetectable [
3].
As a result, transesophageal echocardiography (TEE) is the imaging method of
choice when the transthoracic echocardiogram is technically inadequate or when
there are borderline findings on the transthoracic echocardiogram in a patient
in whom there is a strong clinical suspicion of malfunction [
4].
Transthoracic echocardiography images anteriorly, and thus is usually
sufficient for evaluation of aortic prostheses. However, since the atria are
situated in the far field where resolution is low, they are shadowed by a
mechanical mitral prosthesis. Since TEE images are taken from a position
posterior to the heart, the atrial side of the mitral and tricuspid valves can
be seen. On the other hand, transthoracic echocardiography demonstrates the
ventricular side of the valve that lies in its near field, whereas TEE does
not. Both methods may have difficulty with the evaluation of aortic
prostheses, and the presence of mitral and aortic devices in the same patient
may make the task of the echocardiographer more difficult. In most cases,
however, both echocardiographic methods are complementary.
The physician evaluating a prosthetic device with TEE should be aware of the
range of abnormalities that are possible in these devices and should match
those possibilities to the patient's presentation. Complications of prosthetic
valves detected by TEE include:
• Perivalvular leak
• Endocarditis
• Extrinsic interference of function (pannus,
thrombus, vegetation) resulting in obstruction or regurgitation
• Ball variance
• Strut fracture and component escape
• Leaflet tears of bioprosthesis
• Leaflet calcification/stenosis of bioprosthesis
The role of the TEE for evaluating the abnormalities associated with
prosthetic valves will be reviewed here. The major complications encountered
with these valves and the role of echocardiography in the evaluation of
infective endocarditis are discussed fully elsewhere. (
See
"Complications of prosthetic heart valves" and
see
"Role of echocardiography in infective endocarditis-I").
PERIVALVULAR REGURGITATION ！ Perivalvular
regurgitation can arise from broken or dehisced sutures, from a poorly seated
ring, or from endocarditis (dehiscence). Hemolysis is a common, but generally
underappreciated, complication of these leaks, especially when they occur with
a mitral valve prosthesis [
5,6].
A TEE is required to detect perivalvular leakage and should routinely be
performed when a patient with a prosthetic valve presents with severe anemia.
Severe hemolysis can also occur after mitral valve repair, when regurgitation
develops around the annuloplasty ring. (
See
"Extrinsic nonimmune hemolytic anemia due to mechanical damage:
Fragmentation hemolysis and hypersplenism", section on Hemodynamic
turbulence).
To recognize a perivalvular leak, TEE must be performed with a high color
frame rate and middle range Nyquist limits (35 to 50 cm/sec) in several views
from several angles outside the sewing ring [
7].
There should be a careful search for periprosthetic leaks around as much of
the valve circumference as possible and an attempt to define the extent of the
regurgitation once it is identified. The origin of a periprosthetic leak may
appear deceivingly narrow when it is related to disruption of a limited number
of sutures. In some cases, these jets are seen inadvertently when one examines
other structures, such as the interatrial septum.
The most severe form of perivalvular regurgitation is seen when there is
dehiscence of a substantial portion of the sewing ring [
8].
In this setting, there is severe regurgitation, which may be so torrential
that the regurgitant flow is almost laminar. As in angiography, the sewing
ring on echocardiography is seen to rock with each cardiac cycle; mobile echo
densities representing the liberated suture material often can be visualized.
Both the laminar flow and the rocking motion of the ring may elude the
inexperienced echocardiographer, delaying recognition of this life threatening
condition.
PROSTHETIC VALVE OBSTRUCTION ！ When, in a
newly symptomatic patient, there is an unexpected rise in transprosthetic
gradient from a baseline determination or from established normal values for
valves of that type and size, the possibility of progressive obstruction
should be considered. Causes of obstruction include pannus ingrowth, thrombus,
and vegetation. Clinical clues to this possibility include the age of the
valve and the adequacy of anticoagulation. In a bioprosthesis or heterograft,
the leaflets themselves may become calcified and immobile.
Once there is a high suspicion of obstruction, TEE should be strongly
considered for both mechanical and bioprosthetic valves.
• In the case of suspected aortic pannus, the
distal end of the left ventricular outflow tract should be examined both with
imaging and with color flow. Pannus tends to lie close to the valve ring and
can be easily overlooked. There can be a prevalve jet that suggests pannus,
which, on further searching with a variety of frequencies, angles, and gain
settings, will be detected.
• In the mitral position, the same procedure
should be followed. Finding a high grade of spontaneous contrast in the left
atrium, with or without thrombi, or finding thrombus around the sewing ring in
the setting of adequate anticoagulation should heighten suspicion of pannus
formation. Thin fibrillar strands are also encountered on the mitral annulus
and the left ventricular outflow tract. These structures are brightly
reflective and highly mobile and may or may not be associated with a
pathologic process.
When a thrombus is present in the vicinity of the sewing ring and when
obstruction is clinically plausible and suggested by Doppler imaging,
consideration should be given to the use of thrombolysis or surgery [
9].
(
See
"Complications of prosthetic heart valves").
Distinction between thrombus and pannus ！
The most common etiology for prosthetic valve obstruction is thrombus
formation; pannus formation due to fibrous tissue ingrowth is less common. In
one surgical study of 112 obstructed mechanical valves, pannus formation was
the underlying cause in 11 percent of valves, pannus formation in combination
with thrombus was present in 12 percent, while thrombus alone was the etiology
in the remaining cases [
10].
Though associated with substantial morbidity, thrombolytic therapy is often an
effective alternative to surgery for treatment of thrombus. Therefore, it is
important to distinguish between these two causes. (
See
"Complications of prosthetic heart valves", section on Pannus
formation).
Echocardiographic differentiation of pannus and thrombus may be difficult. In
general:
• Thrombus tends to be mobile, somewhat less
echo-dense, and associated with spontaneous contrast
• Pannus is highly echogenic, consistent with its
fibrous composition, and is usually firmly fixed to the valve apparatus
Color flow aliasing with proximal acceleration of the flow jet in the vicinity
of the mass may aid in the identification of pannus. In one study of 23
patients who presented with 24 obstructed valves, clinical, transthoracic, and
TEE data were compared to pathology upon surgery in order to determine the
clinical and echocardiographic characteristics that differentiate thrombus
from pannus formation [
11].
Thrombus was established in 14 valves and pannus in 10. Pannus
formation was more common in the aortic position. From the standpoint of
echocardiography, thrombus was more likely to be associated with soft
ultrasound density (92 versus 29 percent).
PROSTHETIC VALVE REGURGITATION ！
Physiologic regurgitation, the so-called seating puff of angiography, is
universally encountered and dependent in degree on the type of prosthesis
used. However, severe regurgitation may result from bioprosthetic valve
degeneration , mechanical valve pannus, thrombus, or vegetation.
Physiologic regurgitation ！ All mechanical
valves exhibit some degree of obligatory regurgitation of up to 15 mL of blood
[
12].
Seating regurgitation or "closure backflow" appears only briefly and
is due to retrograde volume displacement as the valve leaflets close. This
type of regurgitation is detected by highly sensitive color flow Doppler
imaging on TEE. In addition, a certain amount of more prolonged "leakage
backflow" regurgitation occurs after the valve closes [
13].
In an evaluation of 136 mechanical prosthetic valves, TEE color Doppler
imaging revealed regurgitant jets in 95 percent of the mitral prostheses and
44 percent of the aortic prostheses [
14].
In contrast transthoracic echocardiography documented regurgitant jets in only
28 and 29 percent, respectively. The lower incidence of detected regurgitation
in the aortic position may be explained by the shadowing of the left
ventricular outflow tract in TEE imaging [
15].
Normally functioning mechanical valves, such as the bileaflet St. Jude,
usually have two to four centrally directed regurgitant jets. Features
associated with these jets include a low intensity and only minimal
penetration into the atrium, generally less than 3 cm [
13,16,17].
The monodisc Medtronic-Hall valve has two jets, one of which is prominent and
longer [
17,18].
Normally functioning heterografts are less likely to have these small
regurgitant signals; when mild regurgitation is present, there is usually one
central jet [
14,19].
Pathologic regurgitation ！ Pathologic jets
tend to be high velocity, central, intense, broad, and highly aliased; they
tend to be eccentric, hugging the wall of the atrium (Coanda effect),
penetrating to its roof, continuing around its perimeter and terminating where
they began. The offending mass or tissue can usually be seen, but the actual
locus of obstruction causing the regurgitation may not be visible.
In contrast to mechanical valves, bioprostheses with leaflet degeneration may
exhibit central pathologic regurgitation that is broad-based when severe. The
bioprosthetic leaflets can usually be seen and are often thickened and
calcified with possible perforation or disruption leading to a flail leaflet.
Forward flow velocities measured by spectral Doppler may be increased due to
the larger flow volume associated with regurgitation as well as concurrent
obstruction.
REGURGITATION DUE TO STRUT FRACTURE AND DISK
LIBERATIONS ！ There has been considerable concern about large
Bjork-Shiley monodisc valves because of their low, but significant, incidence
of strut fracture, leading to liberation of the disk and rapid hemodynamic
compromise. Fortunately, this is very uncommon. When it does occur, clinical
deterioration is often so rapid that diagnostic TEE cannot be performed and
the patient is frequently taken from the emergency room to surgery based on
clinical suspicion.