Surgical repair of the mitral valve is being increasingly performed to treat severe mitral regurgitation. Transesophageal echocardiography is an essential tool for assessing valvular function and guiding surgical decision making during the perioperative period. A careful and systematic transesophageal echocardiographic examination is necessary to ensure that appropriate information is obtained and that the correct diagnoses are obtained before and after repair. The purpose of this article is to provide a practical guide for perioperative echocardiographers caring for patients undergoing surgical repair of mitral regurgitation. A guide to performing a systematic transesophageal echocardiographic examination of the mitral valve is provided, along with an approach to prerepair and postrepair assessment. Additionally, the anatomy and function of normal and regurgitant mitral valves are reviewed.
Attention ASE Members:
ASE has gone green! Visit www.aseuniversity.org to earn free CME through an online activity related to this article. Certificates are available for immediate access upon successful completion of the activity. Non-members will need to join ASE to access this great member benefit!
Normal Anatomy and Function of the Mitral Valve
The MV is best conceptualized as a valve complex, comprising an annulus, leaflets, chordae, papillary muscles, and left ventricular muscle. Normal functioning of the MV requires the coordinated activity of all components of the valve complex.
Annulus
The mitral annulus is a fibrofatty ring that approximates a hyperbolic paraboloid, a geometric shape similar to a riding saddle. The annulus has two axes, a shorter and “higher” (more basal) anteroposterior (AP) axis and a longer and “lower” (more apical) commissural axis ( Figure 1 ). The anterior pole of the AP axis corresponds to the “riding horn” of the saddle and the commissural axis to the “seat” of the saddle. Anteriorly, the mitral annulus is thickened and fixed to the aortic annulus, a region termed the intervalvular fibrosa. The saddle shape of the mitral annulus acts to reduce tension on the leaflets, particularly the middle scallop of the posterior leaflet. Annular height is normalized to annular size by the ratio of height to commissural length at end-systole and is normally about 15%. Leaflet tension increases dramatically when this ratio falls below 10% (i.e., when the annulus becomes planar [decreased height] or dilated [increased commissural length]).
Ventricular contraction results in important conformational changes in the mitral annulus ( Figure 1 ). In early systole, left ventricular contraction causes a sphincter-like decrease in posterior annular area. The annulus shortens along the AP axis, and overall annular area is reduced by approximately 25%. Ventricular contraction also causes systolic folding of the anterior annulus, leading to a deepening of the saddle. The AP diameter returns to normal in midsystole, but increased annular height is maintained throughout systole. Annular folding and sphincteric contraction reduce leaflet tension and aid leaflet apposition, particularly in early systole when ventricular pressure is low.
Leaflets
The MV has an anterior and a posterior leaflet ( Figure 2 ). The anterior leaflet is oriented slightly medially (rightward) and the posterior leaflet slightly laterally (leftward). The leaflet edges meet at two commissures, termed anterolateral and posteromedial. The anterior leaflet is thicker, has a shorter annular attachment, and has a longer base-to-tip length than the posterior leaflet. In most people, the posterior leaflet is composed of three distinct scallops, which are not present on the anterior leaflet. Pleating of the scallops aids closure of the C-shaped posterior leaflet. By contrast, the anterior leaflet does not alter its circumferential length during systole, and therefore no pleating mechanism is required. Leaflet segments are usually named using the system popularized by Carpentier ( Figure 2 ). This nomenclature is useful for defining the location of leaflet pathology and for describing the relationship of the annulus to adjacent cardiac structures. The edges of the leaflets meet at a curved coaptation line that runs roughly along the commissural axis. There is normally approximately 10 mm of leaflet overlap (coaptation height) at end-systole.
Papillary Muscles, Chordae, and Left Ventricle
Two papillary muscles, the anterolateral and the posteromedial, support the mitral leaflets. The papillary muscles run parallel with the long axis of the left ventricle, aligned with the commissures. Systolic contraction of the papillary muscles offsets the base-to-apical shortening of the left ventricle, which would otherwise cause leaflet prolapse. The larger anterolateral muscle typically arises from the mid anterolateral wall of the left ventricle and supports the ipsilateral half of both leaflets: A1/P1 and the anterolateral part of A2/P2. The smaller posteromedial muscle typically arises from the mid inferior wall of the left ventricle and supports the ipsilateral part of both leaflets: A3/P3 and the posteromedial part of A2/P2. Branches of the left anterior descending and circumflex coronary arteries supply the anterolateral muscle, whereas the posteromedial muscle is supplied entirely by branches of the right coronary artery and is therefore more vulnerable to rupture after myocardial infarction.
The papillary muscles attach to the leaflets via chordae tendineae. Primary chords attach to the free edges of the leaflets, and secondary chords attach to the undersurface of the leaflets. Primary chords support the free edges of the leaflets during systole. Rupture of primary chords causes acute MR. Secondary chords help maintain left ventricular geometry, particularly the two thicker strut chords, which attach to the undersurface of the anterior leaflet.
Normal Anatomy and Function of the Mitral Valve
The MV is best conceptualized as a valve complex, comprising an annulus, leaflets, chordae, papillary muscles, and left ventricular muscle. Normal functioning of the MV requires the coordinated activity of all components of the valve complex.
Annulus
The mitral annulus is a fibrofatty ring that approximates a hyperbolic paraboloid, a geometric shape similar to a riding saddle. The annulus has two axes, a shorter and “higher” (more basal) anteroposterior (AP) axis and a longer and “lower” (more apical) commissural axis ( Figure 1 ). The anterior pole of the AP axis corresponds to the “riding horn” of the saddle and the commissural axis to the “seat” of the saddle. Anteriorly, the mitral annulus is thickened and fixed to the aortic annulus, a region termed the intervalvular fibrosa. The saddle shape of the mitral annulus acts to reduce tension on the leaflets, particularly the middle scallop of the posterior leaflet. Annular height is normalized to annular size by the ratio of height to commissural length at end-systole and is normally about 15%. Leaflet tension increases dramatically when this ratio falls below 10% (i.e., when the annulus becomes planar [decreased height] or dilated [increased commissural length]).
Ventricular contraction results in important conformational changes in the mitral annulus ( Figure 1 ). In early systole, left ventricular contraction causes a sphincter-like decrease in posterior annular area. The annulus shortens along the AP axis, and overall annular area is reduced by approximately 25%. Ventricular contraction also causes systolic folding of the anterior annulus, leading to a deepening of the saddle. The AP diameter returns to normal in midsystole, but increased annular height is maintained throughout systole. Annular folding and sphincteric contraction reduce leaflet tension and aid leaflet apposition, particularly in early systole when ventricular pressure is low.
Leaflets
The MV has an anterior and a posterior leaflet ( Figure 2 ). The anterior leaflet is oriented slightly medially (rightward) and the posterior leaflet slightly laterally (leftward). The leaflet edges meet at two commissures, termed anterolateral and posteromedial. The anterior leaflet is thicker, has a shorter annular attachment, and has a longer base-to-tip length than the posterior leaflet. In most people, the posterior leaflet is composed of three distinct scallops, which are not present on the anterior leaflet. Pleating of the scallops aids closure of the C-shaped posterior leaflet. By contrast, the anterior leaflet does not alter its circumferential length during systole, and therefore no pleating mechanism is required. Leaflet segments are usually named using the system popularized by Carpentier ( Figure 2 ). This nomenclature is useful for defining the location of leaflet pathology and for describing the relationship of the annulus to adjacent cardiac structures. The edges of the leaflets meet at a curved coaptation line that runs roughly along the commissural axis. There is normally approximately 10 mm of leaflet overlap (coaptation height) at end-systole.
Papillary Muscles, Chordae, and Left Ventricle
Two papillary muscles, the anterolateral and the posteromedial, support the mitral leaflets. The papillary muscles run parallel with the long axis of the left ventricle, aligned with the commissures. Systolic contraction of the papillary muscles offsets the base-to-apical shortening of the left ventricle, which would otherwise cause leaflet prolapse. The larger anterolateral muscle typically arises from the mid anterolateral wall of the left ventricle and supports the ipsilateral half of both leaflets: A1/P1 and the anterolateral part of A2/P2. The smaller posteromedial muscle typically arises from the mid inferior wall of the left ventricle and supports the ipsilateral part of both leaflets: A3/P3 and the posteromedial part of A2/P2. Branches of the left anterior descending and circumflex coronary arteries supply the anterolateral muscle, whereas the posteromedial muscle is supplied entirely by branches of the right coronary artery and is therefore more vulnerable to rupture after myocardial infarction.
The papillary muscles attach to the leaflets via chordae tendineae. Primary chords attach to the free edges of the leaflets, and secondary chords attach to the undersurface of the leaflets. Primary chords support the free edges of the leaflets during systole. Rupture of primary chords causes acute MR. Secondary chords help maintain left ventricular geometry, particularly the two thicker strut chords, which attach to the undersurface of the anterior leaflet.
Etiology of Mitral Regurgitation
In developed countries, degenerative disease and FMR are the two most common indications for surgical treatment of MR, accounting for approximately 70% and 20% of cases, respectively. Rheumatic heart disease is relatively uncommon in developed nations but remains the most frequent cause of valvular heart disease in developing countries. Other important causes of MR include endocarditis, clefts, and papillary muscle rupture.
Degenerative MR
Degenerative MV disease encompasses a range of pathology, including chordal stretching or rupture, leaflet thickening and redundancy, annular dilatation, and calcification of the leaflets and chordae. Leaflet and chordal thickening is due to proliferation of cellular and connective tissue elements, particularly the accumulation of glycosaminoglycans in the extracellular matrix, a process termed myxomatous change.
Two forms of degenerative disease are recognized: fibroelastic deficiency (FED) and Barlow disease. With FED, there is chordal elongation or rupture resulting in prolapse or flail of an isolated segment, most commonly P2. The affected segment may be morphologically normal or demonstrate myxomatous change. Annular dimensions are only mildly increased. Patients with FED are typically older (aged >60 years) and have short clinical histories consistent with the abrupt onset of MR due to chordal rupture. Barlow disease is characterized by widespread myxomatous change involving multiple leaflet segments and the subvalvular apparatus. Patients are often younger (aged <60 years) and have long-standing MR. Although FED and Barlow disease are separate clinical entities, they represent two ends of a disease spectrum. Barlow disease is associated with mitral annular disjunction, in which there is wide separation (5–15 mm) between the left ventricular wall and the atrial wall–MV junction posteriorly, resulting in hypermobility of the posterior annulus. Degenerative MV disease usually occurs in isolation but is also associated with systemic connective tissues disorders such as Marfan and Ehlers-Danlos syndromes.
Quantitative three-dimensional (3D) echocardiography demonstrates important abnormalities in valvular dimensions and motion in patients with degenerative MV disease. Values for annular area, leaflet area, AP diameter, commissural diameter, and prolapse height are all increased compared with normal, being greater for Barlow disease than FED. The ratio of commissural diameter to AP diameter is reduced with Barlow disease compared with FED, reflecting the more circular shape of the annulus in Barlow disease. Annular dynamics are also abnormal. During systole, shortening along the AP axis occurs normally, but there is marked pathologic expansion along the commissural axis in late systole, leading to a diminished reduction in annular area, which exacerbates MR. Annular folding is also reduced, resulting in a more planar annulus, which also contributes to MR.
Successful repair of degenerative MV disease is possible in the majority of patients, particularly when disease is limited to the posterior leaflet. Anterior and bileaflet repairs are more challenging and are associated with a lower rate of success and a higher need for reoperation. However, success for all types of repairs is increasing. In recent series, success rates approaching 100% have been reported for isolated posterior leaflet repairs. For complex anterior and bileaflet repairs, success rates of >90% have been reported at high-volume centers. These figures are unlikely to be achieved in nonspecialist units.
FMR
FMR is MR that occurs in the presence of structurally normal mitral leaflets. FMR may be ischemic or nonischemic, the latter due primarily to dilated cardiomyopathy. The main mechanism of FMR is leaflet tethering due to ventricular dilatation. Left ventricular remodeling causes lateral and/or apical displacement of the papillary muscles, resulting in leaflet tethering in systole. However, the relationship between ventricular dilatation and MR is complex. FMR is more common after inferior or posterior myocardial infarction than anterior infarction, despite greater ventricular dilatation with the latter ( Figure 3 ). Inferior or posterior infarction causes more displacement of the posteromedial papillary muscle than occurs to the anterolateral papillary muscle after anterior infarction. There are several reasons for the reduced impact of anterior infarction on mitral geometry : the annulus is better supported anteriorly by the intervalvular fibrosa, the ventricular septum helps prevent lateral displacement of the anterolateral papillary muscle, and anterior infarctions tend to be more apical, with relative sparing of the basal left ventricular wall.
Left ventricular systolic dysfunction and annular dilatation contribute to FMR but are not primary etiologic mechanisms. In clinical studies, there is an inconsistent relationship between left ventricular ejection fraction (LVEF) and the severity of FMR. Thus, it is not unusual for patients with severe left ventricular dysfunction to have minimal FMR and vice versa. Annular dilatation, particularly along the AP axis, is a consistent finding but is less marked than with degenerative MR. Annular height is variable, but in general, the annulus is more planar than normal. During systole, there is reduced contraction along the AP axis and a minimal increase in annular height.
Isolated annular dilatation, in the absence of ventricular remodeling, is an uncommon cause of FMR but can occur because of atrial dilatation secondary to atrial fibrillation.
The durability of MV repair for FMR is less than for degenerative disease, with recurrence rates for moderate or severe MR of 20% to 30% typical. Recurrence is more likely when annular dilatation is severe and there is marked leaflet tethering (see below). In a recently published randomized trial, no difference in survival was observed between repair or replacement for severe FMR, but recurrence of moderate or severe regurgitation was 32.6% for repair versus 2.3% for replacement at 12-month follow-up. However, the trial was not powered to detect a mortality difference, and given the randomized design, some patients at high risk for recurrence would have undergone valve repair.
Rheumatic Disease, Endocarditis, Clefts, and Papillary Muscle Rupture
Rheumatic MR is characterized by leaflet thickening and retraction, chordal shortening, and commissural fusion. Leaflet motion is restricted in both systole and diastole, and the leaflet tips have a characteristic rolled-edge appearance. Calcification may be present in the annulus, leaflets, and subvalvular apparatus. Valve repair for rheumatic MR is challenging and associated with a high failure rate. In most circumstances, valve replacement is the preferred treatment.
Endocarditis can occur on normal valves but is more common on diseased valves. MR arises because of leaflet perforation, destruction, or deformity. Leaflet perforation commonly occurs at the site of attachment of vegetations. Endocarditis can also cause aneurysm or abscess formation in the valve and surrounding tissues, which may perforate causing MR. If leaflet destruction is not severe, MV repair is feasible in most patients.
Mitral clefts are typically congenital in origin. Anterior clefts are more common than posterior clefts and usually occur in association with other congenital heart disease, particularly endocardial cushion defects such as inlet ventricular septal defect or primum atrial septal defect. Clefts of the posterior leaflet are very uncommon and are not associated with other congenital heart disease. Clefts that present in adulthood are strongly associated with degenerative MV disease, at least for the posterior leaflet. Degenerative change may reflect regurgitation-induced mechanical injury. The great majority of mitral clefts can be successfully repaired.
Most cases of papillary muscle rupture are due to myocardial infarction, but rupture occasionally occurs after chest trauma. Rupture of the posteromedial muscle is associated with inferior myocardial infarction and occurs approximately 10 times more commonly than rupture of the anterolateral muscle, reflecting the single-vessel blood supply to the former. MV replacement is usually required.
Transesophageal Echocardiographic Assessment of the Mitral Valve
Patients undergoing MV repair should undergo a systematic transesophageal echocardiographic examination according to published guidelines. In addition, the MV apparatus should be further examined using a combination of two-dimensional (2D) and 3D imaging ( Figure 4 ). Two-dimensional and 3D imaging are complementary modalities, each with its own strengths and limitations. In general, qualitative 3D imaging is more suitable to the operating room environment than quantitative analysis. Qualitative 3D imaging is more accurate than standard 2D imaging in localizing leaflet pathology, whereas 2D imaging is superior to qualitative 3D imaging for making measurements and for rapidly quantifying severity. In standard 2D and 3D views, it is important to adjust the depth or zoom function to focus on the mitral apparatus and to examine the valve with and without color Doppler.
Orientation of the MV can be confusing; the three commonly used orientations are shown in Figure 2 .
Echocardiographic-Anatomic Correlations
Various “roadmaps” have been published describing the anatomic-echocardiographic relationships for the standard midesophageal views. However, differences exist regarding which leaflet segments are visualized in each view, particularly with respect to the four-chamber view, which has variously described as visualizing A2/P2, A2/A1/P2, and A2/A3/P2/P3. Furthermore, in a recent study by Mahmood et al ., in which experienced echocardiographers were shown a video sequence of various midesophageal views, the correct mitral segments were identified in only 30.4% of cases. Correct identification of A2 and P2 occurred 69.4% of the time in the long-axis view and 50% of the time in the four-chamber view. A1 and P1 were correctly identified 13.89% of the time in the long-axis view and 47.2% of the time in the four-chamber view, while correct identification of A3 and P3 occurred in only 5.6% of cases in both views. In the commissural view, the correct segments were identified 92% of the time. The reasons for the low success rate include individual variability in cardiac shape and position and the lack of anatomic references in the four-chamber and long-axis views ( Figure 5 ). By contrast, qualitative 3D imaging in the en face view ( Figure 6 ) allows all mitral segments to be identified accurately.
Midesophageal Views
In the four-chamber view ( Figure 4 A) the anterior leaflet (usually A2) is on the left, and the posterior leaflet (usually P2) is on the right of the image. The scan plane typically cuts the coaptation line slightly obliquely. In a high or mid probe position, the anterior leaflet appears relatively longer, whereas when the probe is deep, the posterior leaflet appears relatively longer. Oblique imaging is confirmed by visualizing the posteromedial papillary muscle on the right of the image. Withdrawing or anteflexing the probe sweeps the image plane toward the anterolateral commissure (A1/P1) and displays the left ventricular outflow tract (LVOT). Advancing or retroflexing the probe sweeps the image plane toward the posteromedial commissure (A3/P3).
In the commissural view ( Figure 4 B), the MV appears trileaflet, with P1 on the right, A2 in the center, and P3 on the left. The anterolateral papillary muscle may be visible on the right and the posteromedial papillary on the left of the image. The image plane runs parallel to the axis of the curved coaptation line, which can result in a broad or double jet of MR when the regurgitant orifice is elliptical ( Figure 7 ). The commissural view passes through the “low,” “long” axis of the MV, yielding higher values for annular dimension and prolapse height compared with the long-axis view. A flail or prolapsing P2 segment may appear to rise above A2 in the center of the valve during systole (the cobra sign; Figure 8 ). In the two-chamber view ( Figure 4 C), the three segments of the anterior leaflet (A3/A2/A1) are to the right, and P3 is to the left of the image. The coaptation line is cut at A3/P3. The left atrial appendage is on the right, and the coronary sinus is on the left.
In a true long-axis view ( Figure 4 D), the mitral and aortic valves are both visualized, but neither papillary muscle is seen. When correctly obtained, the long-axis view cuts the coaptation line perpendicularly through A2/P2, along the “high,” “short” axis of the valve. Turning the probe to the left (counterclockwise) sweeps the scan plane toward the anterolateral commissure (A1/P1) and, eventually, the left atrial appendage. Turning the probe to the right (clockwise) sweeps the scan plane toward the posteromedial commissure (A3/P3) and, eventually, the right atrium.
Transgastric Views
The basal short-axis view ( Figure 4 E) provides an en face view of the MV, potentially displaying all mitral segments. With color Doppler imaging, it is sometimes possible to identify the segments (e.g., A1/P1), but not necessarily the leaflet, involved. However, to visualize the regurgitant jet, the image plane must be sufficiently basal (i.e., at or above the annular plane in systole), which is frequently not possible. The two-chamber view ( Figure 4 F) is useful for visualizing the subvalvular apparatus.
Three-Dimensional Imaging
The recommended orientation for displaying the MV with 3D imaging is en face from the left atrial side in the surgical orientation ( Figure 6 ). This view facilitates unambiguous communication between the echocardiographer and surgeon and allows all mitral segments to be accurately identified. The aortic valve and left atrial appendage should be included in the data set, as this helps when orienting the image.
Three-dimensional data sets may be acquired in a single heartbeat (real time) or over several heartbeats. With multiple-beat acquisition, narrow volumes of data are acquired each heartbeat and electronically stitched together. Real-time acquisition is less affected by electrocautery and movement but has lower temporal and spatial resolution than multiple-beat acquisition. Disconnecting the patient from the ventilator and pausing diathermy can minimize movement and electrocautery artifacts. Single-beat acquisition is usually satisfactory for standard 3D imaging, but to achieve adequate resolution, multiple-beat acquisition, over four to six heartbeats, is necessary for color 3D imaging. Zoomed mode is preferred over full-volume mode, as the data set can be limited to the MV, further improving resolution.
Examination before Cardiopulmonary Bypass
The goals of the transesophageal echocardiographic examination before cardiopulmonary bypass (CPB) are to assess the mechanism, location, and etiology of the regurgitation; to grade severity; and to identify associated pathology such as ventricular dysfunction, pulmonary hypertension, and tricuspid regurgitation (TR). For minimally invasive MV surgery, TEE is also used for assessing the position of cannulae used for CPB.
Assessing Valvular Morphology
Annulus
Degenerative and FMR are associated with annular dilatation. The annulus should be measured at end-systole in the midesophageal long-axis view with the calipers placed at the base of the aortic valve and the hinge point of the posterior mitral leaflet. Using the base of the aortic valve, rather than the hinge point of the anterior leaflet, includes the intervalvular fibrosa and therefore overestimates the true annular dimension by approximately 5 mm. The aortic valve is chosen because the hinge point of the anterior leaflet is often difficult to assess in patients with MV disease. Using this dimension, the upper limit of normal for the mitral annulus is 35 mm, with values > 40 mm considered to indicate severe dilatation. If not diseased, the length of the anterior mitral leaflet provides a useful guide to the appropriate annuloplasty size. The anterior mitral leaflet length is measured from the base of the aorta to the leaflet tip in the midesophageal long-axis view in late diastole, when the leaflet is straight.
The finding of mitral annular calcification or mitral annular disjunction should be discussed with the surgeon, as these findings may complicate the surgery or, in the case of mitral annular disjunction, necessitate the reattachment of the annulus to the ventricular endocardium as part of the repair procedure.
Leaflets
MV pathology may be classified on the basis of leaflet motion ( Figure 9 ). Excessive leaflet motion (type 2) is typically due to degenerative disease and involves leaflet flail, prolapse, or both ( Figure 10 ). Prolapse occurs because of excessive leaflet tissue and/or chordal lengthening. During systole, the body of the leaflet domes above annular plane, but the leaflet tip is directed toward the left ventricle. Less severe prolapse, in which the leaflet tip remains in the ventricle at end-systole, is termed “billowing.” Prolapse height should be assessed in the midesophageal long-axis view (i.e., the “high” axis), to avoid overdiagnosis. Leaflet flail occurs because of chordal (or, occasionally, papillary muscle) rupture. The leaflet tip is directed toward the left atrium, and ruptured chordae may be seen flicking in the left atrium in late systole. Isolated prolapse and flail cause eccentric regurgitation, with the jet directed away from the affected side. Ruptured chordae arising from the anterolateral (P1/A1) and posteromedial (P3/A3) segments can flick into the central part of the valve (P2/A2) in late systole, which, with 2D imaging, can give the false impression of P2/A2 flail. However, with 3D imaging, the origin of the chordae is usually readily apparent. Color 3D imaging helps localize the origin of regurgitant jets, which may not directly match the leaflet pathology seen with 2D or 3D imaging ( Figure 11 ).
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
