Importance of Preserving the Mitral Subvalvular Apparatus in Mitral Valve Replacement

Transcription

1 Importance of Preserving the Mitral Subvalvular Apparatus in Mitral Valve Replacement Masafumi Natsuaki, MD, Tsuyoshi Itoh, MD, Shinji Tomita, MD, Koujirou Furukawa, MD, Masaru Yoshikai, MD, Hisao Suda, MD, and Hitoshi Ohteki, MD Department of Thoracic Surgery, Saga Medical School, Saga, Japan Background. This clinical study sought to determine whether mitral valve replacement (MVR) with the preservation of both anterior and posterior chordae tendineae (MVR group II) would be more effective on the improvement of left ventricular regional wall motion than MVR with the preservation of posterior chordae tendineae alone (MVR group I). Methods. Postoperative left ventricular wall motion was analyzed by a centerline method in three groups of MVRmgroup I (n = 13), group II (n = 15), and repair group (n = 15)--for mitral regurgitation. Shortening fraction of chordal length was determined in 100 chords, and these chords were divided into five regions. Results. The comparison of postoperative versus preoperative shortening fraction among the three groups revealed that postoperative wall motion improved more strikingly at apical and diaphragmatic regions in the MVR group II and repair group in comparison to the MVR group I. The postoperative shortening fraction at the apical region in the MVR group I! was significantly increased in comparison to preoperative shortening fraction (preoperative, 3.68% %; postoperative, 5.38% %; p < 0.05). However, postoperative shortening fraction in cardiac base was decreased in the MVR group II as well as other two groups. Conclusions. The MVR with the preservation of both anterior and posterior chordae tendineae contributed to the improvement of left ventricular regional wall motion in the apical and diaphragmatic regions. (Ann Thorac Surg 1996;61:585-90) C ardiac function after mitral valve replacement (MVR) for chronic mitral regurgitation has been reported to be impaired owing to postoperative elevation of left ventricular (LV) afterload [1], and postoperative management of severe cases using conventional MVR is sometimes troublesome. Some investigators have described a modified technique of MVR with preservation of the posterior submitral complex for mitral stenosis or regurgitation [2, 3]. However, it is questionable whether or not MVR with preservation of the posterior chordae tendineae alone can achieve satisfactory improvement of LV wall motion for severe mitral regurgitation with depressed cardiac function. An MVR procedure with preservation of all the chordae tendineae was later introduced by David and colleagues [4], who showed that the postoperative ejection fraction (EF) increased with exercise and that LV performance improved after operation. However, LV regional wall motion was not evaluated accurately by left ventriculography. We have used the technique of MVR sparing all the chordae tendineae with continuity of all the chordae and papillary muscles to examine whether this technique is more effective on LV regional cardiac function than MVR sparing the posterior chordae tendineae alone. The purpose of this clinical study was to examine the LV regional wall motion after mitral valve operation, and to clarify the contribution of Accepted for publication Sep 18, Address reprint requests to Dr Natsuaki, Department of Thoracic Surgery, Saga Medical School, Nabeshirna 5-1-1, Saga 849, Japan. the preserved chordae tendineae to the wall motion using the centerline method of contrast left ventriculography. Patients and Methods Mitral valve replacement or mitral valve repair was performed in 43 patients with severe chronic mitral regurgitation of Sellers degree III or IV evident by contrast left ventriculography [5]. The causes of mitral regurgitation were degenerative lesion in 26 and rheumatic lesion in 17. If the patients also had infective endocarditis, they were excluded from this study. The 43 patients were retrospectively divided into three groups according to the surgical method used between January 1992 and January 1995, and the preoperative and postoperative LV wall motions were compared among the groups. The type of surgical technique used to treat mitral regurgitation was selected according to the anatomic severity of prolapse or the thickness of the anterior leaflet. A summary of the patient data is shown in Table 1. The MVR was performed when the anterior leaflet was widely prolapsed from the mid-portion to the commissural side. For MVR, we used preservation of both the anterior and posterior chordae tendineae (MVR group II) when the anterior leaflet was smooth or slightly thickened, or preservation of posterior chordae tendineae and resection of the anterior chordae tendineae (MVR group I) when the anterior leaflet was calcified and thickened. The mean size of the prosthetic valve did not differ 1996 by The Society of Thoracic Surgeons /96/$15.00 Published by Elsevier Science lnc SSDI (95)

2 586 NATSUAKI ET AL Ann Thorac Surg PRESERVING MITRAL SUBVALVULAR APPARATUS 1996;61: Table 1. Summary of Patient Data Characteristic Demographics Age (y) Range Men Women NYHA IV ]II Group I Group II Group III (n = 13) (n - 15) (n - 15) {38-69) {38-72) (37-70) II Preoperative hemodynamics AF EF ± 0.70 AF atrial fibrillation; EF = ejection fraction; NYHA New York Heart Association. between MVR groups I and II. For preservation of the anterior chordae tendineae in group II, the anterior leaflet was divided into two parts at its center, and the two parts were separately shifted to both commissures. This method was originally reported by Miki and colleagues in 1988 [6]. In our method, all the anterior chordae tendineae were preserved to maintain an adequate tension of the papillary muscle, and the posterior chordae tendineae were completely preserved. A St. Jude medical valve with a low profile was selected as prosthetic valve whose bileaflet movement was not disturbed by the preserved submitral complex. Mitral valve repair was performed (repair group, group IIl) when the anterior leaflet was not so widely prolapsed or when only the posterior leaflet was prolapsed. The procedure of mitral repair involved quadrangular resection and plication for posterior leaflet prolapse, or reconstruction using artificial polytetrafluoroethylene chordae for anterior leaflet prolapse. Artificial mitral ring annuloplasty was used in 6 patients. All 43 patients were operated on by the same surgeon. Preoperative and postoperative LV volume and regional wall motion were compared among the three groups. Contrast left ventriculography was performed in the 30-degree right anterior oblique projection and 60- degree left anterior oblique projection using 35-mm cineangiography. The LV end-systolic and end-diastolic volumes were measured by the area-length method. The preoperative end-diastolic volume index ranged from 130 to 260 ml/m 2. The LV regional wall motion was analyzed by the modified centerline method (CAA100; Nishimoto Inc, Osaka, Japan) in 13 patients in group I, 15 in group II, and 15 in the repair group. A centerline was constructed by computer midway between the end-diastolic and end-systolic contours, and 100 chords (numbered 1 to 100) were drawn perpendicular to the centerline (Fig 1). The LV regional wall motion was expressed as the mean shortening fraction (SF) in each group of 20 chords in the anterobasal (chords 1 to 20), anterolateral (21 to 40), apical (41 to 60), diaphragmatic (61 to 80), and posterobasal (81 to 100) regions. Shortening fraction was calculated as SF = chord-length/end-diastolic perimeter 100 (%). Motion at each chord was normalized for heart size by dividing by the perimeter of the end-diastolic contour [7]. Normal control mean SF was 4.1% in the anterobasal, 2.5% in the anterolateral, 3.3% in the apical, 3.4% in the diaphragmatic, and 3.2% in the posterobasal region (n = 10). Postoperative SF in these five regions were compared with the preoperative value in each group, and the ratio (percentage) of postoperative versus preoperative SF was compared among the three groups. In patients with atrial fibrillation, LV volume and regional wall motion were measured as the mean values in three continuous heart beats. Postoperative left ventriculography was performed about 1 month after operation in MVR or mitral valve repair. Postoperative global and regional cardiac functions were compared statistically with preoperative function by Student's paired t test, and the ratio (percentage) of the postoperative versus preoperative value was corn- (WALL MOTION) SHORTENING (%) i 0 2' '0 160 Fig 1. A centerline constructed by computer in the center between the end-diastolic and end-systolic contours. One hundred chords are drawn perpendicular to &e centerline. Each chord length reflects the regional wall motion of the corresponding point on the centerline (upper). The shortening fraction at each chord is calculated with the correction of the end-diastolic perimeter at 100 points. Dotted lines show the value of the shortening in normal control. Standard deviation is shown above and below the mean value. Solid line shows the shortening in a patient. The shortening in this case is decreased at apical (chords 50 to 70) and posterobasal (chords 90 to 100) regions (lower). (Chords 1 to 20: anterobasal, 21 to 40: anterolateral, 41 to 60: apical, 61 to 80: diaphragmatig and 81 to 100: posterobasal.)

3 Ann Thorac Surg NATSUAKI ET AL ;61: PRESERVING MITRAL SUBVALVULAR APPARATUS Table 2. Preoperative and Postoperative Left Ventricular Volume and Ejection Fraction '~ Variable Group I Group II Group III LV EDVI (ml/m 2) Preop Postop ~ ~ 100 _+ 16 Paired t test p < p < 0.01 p < Ratio (%) _ 15 LV ESVI (ml/m 2) Preop 63 ± _ ± 20 Postop ~ ~ 35 _+ 12 Paired t test p NS p < 0.01 p < 0.01 Ratio (%) _+ 19 b 71 _+ 27 b kv EF (%) Preop _ _- 7 Postop b b Paired t test p < 0.01 p NS p = NS Ratio (%) 82 ± b 100 _+ 10 b Data are mean -- standard deviation. Statistical examination was performed by analysis of variance and group comparisons were performed using Fisher's method, l, p < 0.05 versus group I; ~ p < 0.05 versus group III. Ratio postop value/preop value 100(%). EDVI - end-diastolic volume index; EF ejection fraction; ESVI end-systolic volume index; LV - left ventricular; Postop - postoperative; Preop preoperative. pared among the three groups. Results were expressed as the mean _+ standard deviation. Comparison between three groups was carried out by one-way analysis of variance. When statistical significance was obtained, group comparisons were performed using Fisher's method. A p value of less than 0.05 was judged to indicate a statistically significant difference. Results Preoperative LV cineangiography showed no significant differences among the three groups in LV end-systolic volume index, end-diastolic volume index, and EF. Postoperative end-diastolic volume index was significantly decreased in all three groups, and postoperative endsystolic volume index was significantly decreased in MVR group II or the repair group compared with the preoperative value. Postoperative end-systolic volume index was not significantly decreased in MVR group I. The degree of postoperative decrease in end-systolic volume index was judged to be greater in MVR group II or repair group from the ratio of postoperative versus preoperative end-systolic volume index compared with that of MVR group I (Table 2}. Postoperative EF was not changed in MVR group II or the repair group, whereas the EF of MVR group I was significantly impaired compared with preoperative value. The ratio of postoperative versus preoperative EF indicated that the postoperative EF in MVR group II or repair group more significantly improved than that in MVR group I (Table 2). In the analysis of LV wall motion, postoperative SF in I pre-op SF 8 MVR-Group I 1.] post-op (%) r *** *: P<0.05 **: p<0.01 AB AL AP DP PB Fig 2. Comparison between preoperative and postoperative left ventricular regional wall motion with the centerline method. Postoperative left ventricular shortening fraction (SF) deteriorated in the anterobasal, apical' diaphragmatic, and posterobasal ref~ions in MVR group I (n - 13) compared with preoperative value (paired t test). (AB = anterobasal; AL = anterolateral; AP apicah DP = diaphragmatit, PB posterobasal.) MVR group I was decreased in all five regions (Fig 2), whereas the SF in MVR group II was increased in the anterolateral, apical, and diaphragmatic regions compared with the preoperative value (Fig 3). The postoperative SF in the repair group was slightly increased at the SF (%) MVR-Group li pre-op D post-op *: p<0.05 **: p<0.0! AB AL AP DP PB Fig 3. Comparison between preoperative and postoperative left ventricular regional wall motion with the centerline method. Postoperative left ventricular shortening fraction (SF) improved significantly in the anterolateral and apical regions in MVR group 1I (n = 15) compared with the preoperative value (paired t test). (AB anterobasal; AL anterolateral; AP - apical; DP diaphragmatic; PB posterobasal.)

4 588 NATSUAKI ET AL Ann Thorac Surg PRESERVING MITRAL SUBVALVULAR APPARATUS 1996;61: Table 3. Shortening Fraction of Left Ventricular Five Regions by Centerline Method ~ Variable Group I Group II Group III AB (%) Preop 4.10 _ _ _ Postop 3.46 _ _ _ Paired t test p < 0.05 NS NS AL (%) Preop 3.88 _ _ Postop 3.06 _ _ Paired t test NS p < 0.01 NS AP (%) Preop 4.84 _ _ Postop 2,88 +_ _ b 5.42 _ b Paired t test p < p < 0.05 NS DP (%) Preop 4.31 _ _ _ Postop 2.84 _ _ b 4.72 _ b Paired t test p < 0.01 NS NS PB (%) Preop 3.39 _ _ Postop 2.31 _ _ 0.94 Paired t test p < 0.01 NS p < 0.01 a Data are mean -+ standard deviation. Statistical examination was performed by analysis of variance, b p < 0.05 versus group I, Fisher's method. AB = anterobasal; AL = anterolateral; AP = apical; DP = diaphragmatic; PB posterobasal; Postop - postoperative; Preop - preoperative. anterolateral, apical, and diaphragmatic regions compared with the preoperative value (Table 3). Also the ratio of postoperative versus preoperative SF indicated that postoperative LV wall motion was relatively improved in the apical and diaphragmatic regions in MVR group II and the repair group compared with MVR group I (Table 4). Postoperative SF in the anterobasal and posterobasal regions was decreased in all three groups compared with the preoperative value. Table 4. Comparison of the Ratio (Percentage) of Postoperative Versus Preoperative Shortening Fraction in Left Ventricular Regions ~ Region Group I Group II Group IIl AB _ AL 97 +_ _ _ 94 AP b 139 _+ 75 b DP b 107 _+ 41 b PB _ a Statistical examination was performed by analysis of variance, b p < 0.05 versus group I Fisher's method. Ratio: Postop SF/preop SF x 100 (%). AB = anterobasal; AL = anterolateral; AP = apical; DP = diaphragmatic; BP - posterobasal; Postop = postoperative; Preop = preoperative; SF - shortening fraction. C o m m e n t Most investigators have noted a significant reduction of the EF in patients with mitral regurgitation after MVR without preservation of the chordae tendineae [8-10]. The decrease in the LV EF after conventional MVR was caused by several factors such as decreased preload, increased afterload, or impaired contractile function. After MVR for mitral regurgitation, preload is decreased by removal of the regurgitant volume and afterload is increased by disappearance of the low impedance ejection root into the left atrium. The main factor responsible for decreased ejection performance is thought to he interruption of ventricular valvular interaction with the change in loading condition. Several studies [4, 11] have demonstrated that ejection performance did not worsen after correction or replacement of the mitral valve using procedures that preserve the mitral valve apparatus. Our present findings suggest that preservation of all the chordae tendineae prevents the depression of LV function and allows LV end-systolic volume and shortening of the regional anterior wall to return to near normal. In MVR group II where all the chordae tendineae were preserved, the shortening fraction of postoperative anterior regional wall motion improved to the same level of the postoperative SF after mitral repair. In contrast, MVR group I in which the posterior chordae tendineae alone were preserved did not attenuate the end-systolic volume as much as MVR group II because of the increased afterload resulting from elevation of end-systolic wall stress. This elevation was probably caused by interruption of the continuity of papillary muscle and the anterior chordae tendineae and the impaired function of the papillary muscle. Because of the inadequate function of the papillary muscle, normal LV geometry and axis shortening could not be maintained in the MVR group I [12]. The continuity of all chordae tendineae and papillary muscle maintained functional LV geometry and improved LV shortening in the MVR group II. The important effects of connection of the papillary muscle to the anterior mitral leaflet through the chordae tendineae on LV performance have been demonstrated experimentally by several investigators [13-15]. If the concept that continuity between the chordae tendineae and the papillary muscle enhances contractile function through regional afterload reduction is correct, then it follows that the anterior mitral leaflet and its chordal connection must have a greater impact on regional loading conditions than the posterior mitral leaflet. Greater tension is generated by the anterior leaflet with a large number of chordae tendineae at a given LV pressure [13]. Postoperative cineangiography, performed in the early period, revealed that the regional SF improved in the anterolateral, apical, and diaphragmatic regions and that the LV SF of both axes improved in MVR group II. In contrast, the postoperative regional SF deteriorated at the same regions, and LV postoperative SF in both axes was decreased in MVR group I. Thus, preservation of both the anterior leaflet with its connection and the posterior

5 Ann Thorac Surg NATSUAKI ET AL ;61: PRESERVING MITRAL SUBVALVULAR APPARATUS leaflet contributed to the improvement of regional anterior, apical, and diaphragmatic wall motion. We expected that diaphragmatic and posterobasal regional wall motions would improve in MVR group I. However, we found that diaphragmatic regional wall motion deteriorated although the posterior submitral complex was preserved. These findings suggested that the adequate cooperative tension of the anterior chordae tendineae together with the posterior chordae played an important role in tethering to the papillary muscles. Previous studies have shown that LV cardiac functions improved by MVR with the preservation of the posterior chordae alone [16-18]. In contrast, our clinical data for MVR group I revealed no improvement in global or regional cardiac function. Thus, it was suggested that the posterior chordae alone did not have a tethering effect on the papillary muscle in MVR for chronic mitral regurgitation. The improved ventricular performance after MVR preserving all the chordae tendineae or rnitral valve repair is thought to be caused by the buttressing effect of the papillary muscle on the region as it contracts. This buttressing effect may reduce the afterload stress on the papillary muscle region [19, 20]. A recent study has demonstrated improvement of postoperative LV regional wall motion by analysis using the radial method in MVR with all the chordae tendineae preserved [20]. However, the radial method cannot show LV regional wall motion precisely in enlarged elliptical LV geometry. The method depends on the change of one ventricular perimeter point toward one reference point on the long axis. It is the assumption on radial method that the point on the perimeter located in diastole is approximately the same point measured in systole. This may be an invalid assumption and sometimes it is clearly not the case. And the radial line in the high cardiac base in diastole may be seen in systole to intersect the aortic valve plane. In contrast, the centerline method includes a long perimeter of the ventricular contour for each segment. The method has anatomically coinciding systolic and diastolic regions. The method can be applied to various types of LV contour because its method measures the wall motion along locally determined vectors, and requires no geometric reference figure [21]. Therefore, we used the centerline method in our analysis of wall motion, and detected a significant postoperative improvement in the anterolateral, apical, and diaphragmatic regions and slight impairment in the anterobasal and posterobasal regions after MVR preserving all the chordae tendineae or mitral valve repair. This finding suggests that a satisfactory buttressing effect of both papillary muscles was occurring in the anterolateral, apical, and diaphragmatic regions, and the buttressing effect was incomplete in the anterobasal and posterobasal regions, even when all the chordae tendineae were preserved. An experimental study demonstrated that impaired LV global function and regional anterior wall motion in experimentally created mitral regurgitation were almost reversed to control values by surgical correction preserving all the chordae tendineae [22, 23]. Another study has shown that systolic shortening of the LV major axis was decreased by cutting both the chordae tendineae in MVR [24]. This experimental finding reflects our clinical observation that postoperative apical wall motion was improved by MVR with preservation of both the chordae tendineae. In contrast, the reduction of contractility after replacement with a prosthetic valve or rigid ring in the mitral position was reported to occur through mechanical restriction of the basal portion of the heart [25]. The cause of postoperative impairment of the anterobasal or posterobasal wall motion with our MVR technique or mitral repair may have been mechanical restriction of contraction in the basal portion of the heart because of limited movement of mitral valve ring by prosthetic valve (or prosthetic ring) or narrowing of the mitral orifice by quadrangular resection in mitral repair. Technically, we have been using a method involving shifting of the anterior leaflet to both commissures for preservation of the anterior chordae tendineae in MVR [6]. This procedure is feasible for reattachment of separated segments of the thick anterior leaflet to the mitral ring of both commissures. In contrast, the preservation technique using the in situ position of the anterior leaflet and chordae is thought to be technically problematic for a thick anterior leaflet and chordae [26, 27]. Feikes and colleagues [28] described the another preservation technique involving rotation of the anterior leaflet and chordae to the posterior leaflet. In an experimental study, Moon and co-workers [15] demonstrated that the LV pressure-volume relationship did not differ between the two MVR methods with the preservation of all chordae tendineae--an anterior chordal-sparing technique (Khonsali technique [29]) and a posterior chordal-sparing technique (Feikes technique [28]). The posterior chordalsparing method was different from our MVR technique with the preservation of posterior chordae tendineae alone. The function of papillary muscle may be preserved in the Feikes technique or the Khonsali technique. However, anterior regional wall motion may not always improve in the Feikes technique, because the movement of posterior wall may be strengthened excessively by the preserved chordae tendineae. The anterior chordalsparing technique of Khonsali may maintain an adequate global and regional cardiac function after MVR. We believe it is a problem that LV outflow obstruction may occur when excessive retained leaflet tissue is folded up into the annulus in the Khonsali technique. In contrast, the Miki technique [6] can maintain nearly normal chordal tension on the anterior and posterior mitral ring owing to the more normal anatomic position of the anterior chordae tendineae. We tried to insure that the tension of anterior chordae tendineae was maintained from the commissural side in an anterior portion, and that LV outflow was not obstructed by the excessive leaflet tissue. In conclusion, MVR preserving all the chordae tendineae for mitral regurgitation is an excellent procedure for improving postoperative LV regional wall motion in the anterolateral, apical, and diaphragmatic regions and for decreasing postoperative LV end-systolic volume index. We found that preservation of both the anterior and

6 590 NATSUAKI ET AL Ann Thorac Surg PRESERVING MITRAL SUBVALVULAR APPARATUS 1996;61: posterior mitral subvalvular apparatus in MVR played an important role in preserving LV regional wall motion. In contrast, MVR preserving the posterior chordae tendineae alone was not enough to improve global LV function or regional wall motion. References 1. Ross J Jr. Afterload mismatch in aortic and mitral valve disease: implications for surgical therapy. J Am Coll Cardiol 1985;5: Lillehei CW, Levy MJ, Bonnabeau RC Jr. Mitral valve replacement with preservation of papillary muscles and chordae tendinae. J Thorac Cardiovasc Surg 1964;47: Hetzer R, Bougioukas G, Franz M, Borst HG. Mitral valve replacement with preservation of papillary muscles and chordae tendineae--revival of a seemingly forgotten concept. Cardiovasc Surg 1983;31: David TE, Burns RJ, Bacchus CM, Druck MN, Weisel RD. Mitral valve replacement for mitral regurgitation with and without preservation of chordae tendineae. J Thorac Cardiovasc Surg 1984;88: Sellers RD, Levy MJ, Amplatz K, Lillehei CW. Left retrograde cardioangiography in acquired cardiac disease. Am J Cardiol 1964;14: Miki S, Kusuhara K, Ueda Y, Komeda M, Okita Y, Tahata T. Mitral valve replacement with preservation of chordae tendineae and papillary muscles. Ann Thorac Surg 1988;45: Sheehan FH, Stewart DK, Dogde HT, Mitten S, Bolson EL, Brown BG. Variability in the measurement of regional left ventricular wall motion from contrast angiograms. Circulation 1983;68: Carabello BA, Williams H, Gash AK, et al. Hemodynamic predictors of outcome in patients undergoing valve replacement. Circulation 1986;74: Wong CYH, Spotnitz HM. Systolic and diastolic properties of the human left ventricle during valve replacement for chronic mitral regurgitation. Am J Cardiol 1981;47: Nakano K, Swindle MM, Spinale F, et al. Depressed contractile function due to canine mitral regurgitation improves after correction of the volume overload. J Clin Invest 1991; 87: Bonchek LI, Olinger GN, Siegel R, Tresch DD, Keelan MR Jr. Left ventricular performance after mitral reconstruction for mitral regurgitation. J Thorac Cardiovasc Surg 1984;88: Deutsch HJ, Curtius JM, Bongarth C, et al. Left ventricular geometry and function before and after mitral valve replacement. J Heart Valve Dis 1994;3: Hansen DE, Cahill PD, Derby GC, Miller DC. Relative contributions of the anterior and posterior mitral chordae tendinae to canine global left ventricular systolic function. J Thorac Cardiovasc Surg 1987;93: Yunn KL, Farm JI, Rayhill SC, et al. Importance of the mitral subvalvular apparatus for left ventricular segmental systolic mechanics. Circulation 1990;82(Suppl 4): Moon MR, DeAnda Abe, Daughtex GT, et al. Experimental evaluation of different chordal preservation methods during mitral valve replacement. Ann Thorac Surg 1994;58: Tasdemir O, Katircioglu F, Catav Z, et al. Clinical results of mitral valve replacement with and without preservation of the posterior mitral valve leaflet and subvalvular apparatus. J Cardiovasc Surg 1991;32: Goor DA, Mohr R, Lavee J, Serraf A, Smolinsky A. Preservation of the posterior leaflet during mechanical valve replacement for ischemic mitral regurgitation and complete myocardial revascularization. J Thorac Cardiovasc Surg 1988;96: Hennein HA, Swain JA, Mclntosh CL, Bonow PO, Stone CD, Clark RE. Comparative assessment of chordal preservation versus chordal resection during mitral valve replacement. J Thorac Cardiovasc Surg 1990;99: Rushmer RF. Initial phase of ventricular systole: asynchronous contraction. Am J Physiol 1956;184: Okita Y, Miki S, Kusuhara K, et al. Analysis of left ventricular wall motion after mitral valve replacement with a technique of preservation of all chordae tendineae. Comparison with conventional mitral valve replacement or mitral valve repair. J Thorac Cardiovasc Surg 1992;104: Sheehan FH, Bolson EL, Dodge HT, Mathey DG, Schofer J, Woo HW. Advantages and applications of the centerline method for characterizing regional ventricular function. Circulation 1986;74: Carabello BA, Nakano K, Corin W, Biederman R, Spann JF Jr. Left ventricular function in experimental volume overload hypertrophy. Am J Physiol 1989;256:H Ishihara K, Zile MR, Kanazawa S, et al. Left ventricular mechanics and myocyte function after correction of experimental chronic mitral regurgitation by combined mitral valve replacement and preservation of the native mitral valve apparatus. Circulation 1992;86(Suppl 2): Gains E, Schad H, Heimisch W, Hagl S, Mendler N, Sebening F. Preservation versus severance of the subvalvular apparatus in mitral valve replacement: an experimental study. Eur J Cardiothorac Surg 1990;4: Rijk Zwikker GL, Schipperheyn JJ, Huysmans HA, Bruschke AVG. Influence of mitral valve prosthesis or rigid mitral ring on left ventricular pump function--a study on exposed and isolated blood-perfused porcine hearts. Circulation 1989; 80(Suppl 1): David TE. Mitral valve replacement with preservation of chordae tendineae: rationale and technical considerations. Ann Thorac Surg 1986;41: Rose EA, Oz MC. Preservation of anterior leaflet chordae tendineae during mitral valve replacement. Ann Thorac Surg 1992;57: Feikes HL, Daugharthy JB, Perry, JE, Bell JH, Hieb RE, Johnson GH. Preservation of all chordae tendineae and papillary muscle during mitral valve replacement with a tilting disc valve. J Cardiac Surg 1990;5: Khonsali S, Sintek C. Mitral valve replacement with chordal preservation: clinical significance. In: Rivera R, Duran E, eds. Actualization cardiovascular 5. Madrid: Aran Ediciones, 1992: