Quantitative echocardiography of the mitral complex in dilated cardiomyopathy: the mechanism of functional mitral regurgitation.

We sought to elucidate the mechanism of mitral regurgitation (MR) in dilated cardiomyopathy (DCM). Quantitative two-dimensional echocardiographic examinations were performed in 27 patients, 18 with DCM (nine with MR on physical examination, nine without MR) and nine without underlying heart disease. The MR and "no MR" patients were clinically comparable. Spatial reconstructions from multiple apical cross sections were used to estimate the mitral leaflet area needed to occlude the orifice for a given midsystolic coaptation configuration (LEAF), as well as mitral annular area index, left ventricular volume, and left atrial volume. Similarly, reconstructions from parasternal short-axis views were used to estimate central chordae tendinae length and angulation. From selective parasternal views papillary muscle (PM) length and contraction and the tethering length from the PM base to the annular plane were measured. The MR group was characterized by markedly enlarged occlusional leaflet area (LEAF 19.8 + 3.1 in MR vs 13.8 + 2.8 in no MR group vs 6.3 + 0.9 cm2 in normal group; p < .01), striking mitral annular dilatation (midsystolic annular area index 7.5 + 0.8 in MR vs 4.6 + 0.9 in no MR group vs 2.9 + 0.4 cm2/m2 in normal group; p < .01), and left atrial enlargement (end-systolic left atrial volume 129 + 39 in MR vs 73 14 in no MR group vs 29 + 5 ml in normal group; p < .01). Chordal length and angulation, PM length, contraction, and tethering length, and left ventricular volume were not significantly different in the MR vs the no MR group. Noncoaptation of the mitral leaflets at their free margins was not observed in any MR patient. With the use of stepwise linear regression LEAF was determined chiefly by annular size (R2 .868) with left ventricular size having little additional influence (R2 increment .071). Thus, DCM is associated with enlargement of the mitral anulus, which is more pronounced in those patients with MR. Based on the quantitative estimates of occlusional leaflet area, we postulate that mitral leaflet tissue can stretch somewhat to accommodate dilatation of the mitral complex, but as the requirement for occlusional leaflet area increases less tissue is available for coaptation. Thus, although coaptation continues to occur, the valvular seal becomes ineffective once a critical LEAF is reached. The chief determinant of LEAF is the mitral annular size, while left ventricular size is a less important factor. Circulation 68, No. 3, 498-508, 1983. THE PROPER FUNCTION of the mitral valve requires coordinated motion of its six components: the posterior left atrial wall, the mitral anulus and leaflets, the chordae tendinae, papillary muscles, and free left From the Department of Medicine, University of Califomia School of Medicine and Wadsworth VA Medical Center, Los Angeles. Supported in part by the Arthur Dodd Fuller Foundation for Cardiovascular Research, Los Angeles. and VA Medical Center Research Funds. Address for correspondence: Pravin M. Shah, M.D., Cardiology (691/11lE), Wadsworth VA Medical Center, Wilshire and Sawtelle Blvds., Los Angeles, CA 90073. Received Jan. 20, 1983; revision accepted May 5, 1983. Dr. Boltwood was supported as a 1981-1982 Merck Fellow of the American College of Cardiology, and as an American Heart Association Fellow. Presented in part at the American Heart Association 55th Annual Scientific Sessions, Nov. 17, 1982, Dallas. 498 ventricular wall.' Mitral insufficiency commonly occurs in the presence of left ventricular dilatation, without any primary valvular disease.2 Such functional mitral regurgitation (MR) has been attributed to dilatation of the mitral anulus and to retraction of the leaflets by chordae and papillary muscles as the left ventricle dilates.2 However, functional mitral insufficiency has not been found to correlate well with mitral annular dilatation,3-5 and data obtained at autopsy imply that the mitral leaflets have reserve surface with a ratio of valve-to-orifice area ranging from 1.5 to 2.2.6 By exclusion, therefore, MR resulting from left ventricular enlargement has been attributed to the oblique direction of papillary muscle-chordal tension that presumably accompanies ventricular dilatation.' 7NevertheCIRCULATION by gest on N ovem er 6, 2017 http://ciajournals.org/ D ow nladed from PATHOPHYSIOLOGY AND NATIJRAL HISTORY-MYOCARDIAL DISEASE less, the mitral anulus normally shortens along its posterolateral segment during atrial and early ventricular systole, thereby reducing the valvular orifice.8' 9 Annular dilatation and hypokinesis therefore may contribute to mitral insufficiency, particularly in the presence of leaflet tethering abnormalities. Indeed, the benefit of mitral annuloplasty in various forms of MR has been reported.'0 Our laboratory group has reported on an echocardiographic method of reconstructing and measuring the mitral anulus in man. " After initial success with annular measurement, attention was turned to the mechanism of MR in dilated cardiomyopathy as an appropriate area of investigation. We hypothesized that certain clear anatomic differences might exist between patients with dilated cardiomyopathy with and without MR and that discerning these differences might yield insights into the pathophysiologic characteristics of the disease. Methods of quantitating the other features of the mitral apparatus were developed, and a comprehensive echocardiographic study was performed. The results of this study suggest a more important role for mitral annular dilatation in functional MR than has been appreciated heretofore. Materials and methods Patient selection. Patients with dilated cardiomyopathy who showed no clinical or echocardiographic evidence of coronary or primary valvular disease were enrolled in the study. The nine subjects in the "no MR" group had no detectable systolic murmur when examined separately by two cardiologists, while the nine in the MR group had easily detectable typical high-pitched apical holosystolic murmurs radiating to the axilla (grade 2/6 or greater) the presence of which were confirmed with phonocardiographic examination. Several patients with faint apical systolic murmurs not easily recorded on a phonocardiogram were excluded from the study. No patient was studied during an acute exacerbation of heart failure, and only one patient (in the MR group) had clinical and contrast echocardiographic evidence of associated tricuspid regurgitation. Nine age-comparable normal control subjects were enlisted from the hospital wards or from the medical staff. Subjects with technically unsatisfactory echocardiograms were excluded from the study. All subjects gave written informed consent. Echocardiographic measurements. A two-dimensional echocardiographic system (Varian 3000) was used and recordings were made on half-inch videotape (Sony). The images could be redisplayed in real time, slow motion, or as single frames. Annular and central chordal measurements were taken directly off the screen with a large protractor and/or ruler, while other measurements required tracings on transparencies. Except when specified otherwise, measurements were averaged over 4 sinus or 8 atrial fibrillation beats. The screen height relative to the observer was adjusted to minimize parallax. Vertical and horizontal screen calibrations were repeatedly measured by imaging a 30 mm in diameter wire ring in a water-filled chamber. Mitral anulus. This method has been described in detail in a prior communication from this laboratory."I Briefly, subjects were studied in the left lateral decubitus position. With an inclinometer'2 six separate apical recordings were obtained at 30 degree rotational intervals around the annular circumference. Each apical view was selected to maximize the annular diameter at a given rotational orientation and was recorded during held expiration. By convention, a hinge point was defined as the center of the triangular tissue supporting the base of a leaflet, as identified in cross section. At a given point in the cardiac cycle diameters between the two hinge points measured at 30 degree rotational intervals were assumed to bisect each other, and a smooth circumference was drawn to connect the outer edges of these diameters. In the original method" the assumption of diameters bisecting other diameters was made at only one point in the cardiac cycle, namely end-isovolumetric relaxation, and a floating reference frame was used at other points in the cardiac cycle. In this study this assumption was extended to reconstruction at points of maximum and minimum annular size, i.e., use of a fixed reference frame throughout the cardiac cycle was assumed to be adequate. Also, only two points in the cardiac cycle were measured, namely maximum annular size (at the peak of the P wave, or peak of R wave in atrial fibrillation) and minimum annular size (at ventricular midsystole)."I The area and circumference of each annular outline was obtained by computerized planimetry. This method yielded the midsystolic and middiastolic annular areas indexed to body surface area (Almin and Almax), the fractional contraction in this area from maximum to minimum (FAC), and the midsystolic and middiastolic annular circumferences (Cmin and Cmax). Leaflet coaptation. By convention, the apical views obtained with the inclinometer mentioned above were labeled 8', 9', 10', 1', 12', and 1'. Each view subtends 30 degrees of arc on two sides of the rotational center so that six views complete a 360 degree rotation. In general, the 12' view corresponds to a conventional four-chamber view, and the 9' view corresponds to a conventional two-chamber apical view. With the use of tomographic planes relatively perpendicular to the mitral closure line, the apical recordings obtained above at 10', 11', and 12' were used to study leaflet coaptation. In each recording the leaflet coaptation angle A and perpendicular distance from the annular plane D were measured at midsystole and averaged (figure 1). A and D were further averaged over the 10', 11 ', and 12' views. Occlusional leaflet area (LEAF). The leaflet contour in a given apical view obtained as described above may be approximately described by the four heights hl,...,h4 and the three radii rl,...,r3 (figure 2). Assuming that this contour is swept through 30 degrees of rotation about the origin of r,, the area produced can be calculated as the sum of fractions of cone or frustum areas according to A = 7-f[(r, + r9) (h12 + (rI r2)2)/2 + r2(r22 + h22)')'2 + r3(r32 + h32)'1 + (rI + r3) (h42 + (rI r3)2)'/j] The approximate LEAF would then be the sum of the fractional areas over the six apical rotational views. For this method only one representative midsystolic frame per view was analyzed, without multiframe averaging, and a programmable calculator was used to expedite the analysis. Note that this measurement does not account for leaflet infolding or compression of leaflet tissue at the line of coaptation; hence the term occlusional leaflet area. Presumably this LEAF is the minimum area needed to occlude the mitral orifice for a given midsystolic leaflet configuration and is less than the total leaflet tissue area anatomically present. Left ventricular and atrial volumes. With the 9' and 12' views obtained as described above, left ventricular and left atrial endocardial outlines were obtained. The ventricular outlines were obtained at ventricular end-systole and end-diastole, while the 499 Vol. 68, No. 3, September 1983 by gest on N ovem er 6, 2017 http://ciajournals.org/ D ow nladed from