Nikos Kouris a, *, Ignatios Ikonomidis b, Dimitra Kontogianni a, Peter Smith b, Petros Nihoyannopoulos b

Eur J Echocardiography (2005) 6, 435e442 Mitral valve repair versus replacement for isolated non-ischemic mitral regurgitation in patients with preoperative left ventricular dysfunction. A long-term follow-up echocardiography study Nikos Kouris a, *, Ignatios Ikonomidis b, Dimitra Kontogianni a, Peter Smith b, Petros Nihoyannopoulos b a Western Attica General Hospital, Athens, Greece b Hammersmith Hospital, Imperial College, London, UK Received 12 October 2004; received in revised form 15 January 2005; accepted 21 January 2005 Available online 2 April 2005 KEYWORDS Mitral regurgitation; Depressed LV function; Mitral valve repair Abstract The aim of this study was to evaluate LV function, by means of echocardiography, after mitral valve repair () or mitral valve replacement () in patients (pts) with chronic degenerative mitral regurgitation (MR) and depressed LV systolic function during a 6-years follow-up (FU) period. Patients and methods: Forty-five pts with moderately severe or severe MR and preoperative EF % 50% were divided into 2 groups: group (27 pts, 19 mene8 women, aged 62 G 10 years) and group (18 pts, 8 mene10 women, aged 60 G 12 years). The cause of MR was myxomatous mitral valve disease (/: 16/8), endocarditis (0/4) and degenerative mitral valves with ruptured chordae tendineae (11/6). All pts underwent transthoracic echocardiography preoperatively, postoperatively and annually during the FU period (6 G 3 years). Results: In group, 5 pts died, 5 were lost to FU and 2 pts underwent due to failure. In group, 6 pts died, 3 were lost to FU and 1 was re-operated due to prosthetic valve endocarditis. Atrial fibrillation was similar between the 2 groups. pts demonstrated significant LVEDD decrease postoperatively which was persistent during FU (p! 0.05). LVESD also decreased (p! 0.05), VTI improved (p! 0.05), while FS and EF showed a trend to improve. In pts, LVEDD was decreased * Corresponding author. 4, Agiou Georgiou Street, 15236, N. Penteli, Athens, Greece. Tel.: C30 210 8041601; fax: C30 210 5698465. E-mail address: nikoskou@otenet.gr (N. Kouris). 1525-2167/$30 ª 2005 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.euje.2005.01.003

436 N. Kouris et al. (p! 0.05) but increased during FU (p! 0.05) and LVESD remained high, resulting in a decrease of FS and EF (p! 0.05). VTI remained unchanged (p Z NS). Conclusion: in pts with non-ischemic MR and preoperative LV dysfunction achieves better preservation of LV systolic indices than, probably due to preservation of the subvalvular apparatus and LV geometry. ª 2005 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved. Mitral valve repair (), when feasible, has become the procedure of choice for surgical treatment of chronic mitral regurgitation (MR) because it provides better postoperative clinical outcome than mitral valve replacement (). 1e5 Poor preoperative left ventricular (LV) function has been associated with an adverse clinical outcome 6 in patients with MR, while preoperative LV ejection fraction (EF) may be a powerful predictor of postoperative heart failure and death, 5 regardless of the type of surgery used for correction of chronic MR. This is generally attributed to the postoperative increase in LV afterload and abolition of the pop-off valve effect of the regurgitant leak, further decreasing the EF and cardiac output. 6 The better clinical outcome after is directly related to the preservation of the subvalvular apparatus resulting in preservation of LV geometry. Patients with a significant degree of LV systolic dysfunction (EF! 50%) should undergo surgical correction of MR because: (a) surgical correction, compared to medical treatment, improves prognosis and (b) reduces the incidence of congestive heart failure. 7,8 The purpose of this study was to evaluate the recovery of the preoperative LV function after mitral valve repair or replacement in patients with impaired systolic LV function due to chronic degenerative MR. We also tested the hypothesis that these patients may present a greater improvement of LV function after than after. Patients and methods Study population The study group consisted of 45 patients with isolated moderately severe or severe MR and impaired LV function, without other significant associated valvular or coronary artery disease, documented by preoperative echocardiographic and cardiac catheterization studies. All patients had preoperative ejection fraction!50% (46 G 5%, range 31e50%), determined by echocardiography. The selection of LVEF % 50% as the cut-off point between normal and depressed LV function was based on Veterans Administration Co-operative Study on valvular heart disease. 9 The underlying cause of MR was myxomatous mitral valve disease (floppy mitral valve) in 24 patients, endocarditis in 4 and degenerative valves with ruptured chordae tendineae in 17. Twenty-seven patients (19 mene8 women, aged 62 G 10 years) ( group) underwent (according to the Carpentier type reconstruction 10,11 ) and a rigid Carpentier annular ring was inserted in 16 of them (Table 2). Eighteen patients (8 mene10 women, aged 60 G 12 years) ( group) underwent a conventional without preservation of papillary muscles and a low profile mechanical disc valve was implanted in all of them (Bjork-Shiley in 15 patients and St. Jude in 3). Patients in the 2 groups had similar age and degree of preoperative MR assessed by echocardiography and cardiac catheterization (Table 1). The decision to replace or to repair the valve was made on the table by the surgeon based on the valve anatomy. All patients were operated at the same center (Hammersmith Hospital, London, UK) by the same group of surgeons. Table 1 Preoperative patient characteristics Group A () Group B () No 27 18 Age (years) 62 G 10 60 G 12 Sex 19 mene8 women 8 mene10 women MR severity(median) 3/4 3/4 Moderate 2/4 8 4 Mod. severe 3/4 8 9 Severe 4/4 11 5 Underlying disease Floppy MV 16 8 Flail MV 11 6 Endocarditis 0 4 Leaflets involved Posterior 21 9 Anterior 2 3 Both 4 6 SR/AF/paced 16/10/1 11/7/0, mitral valve repair;, mitral valve replacement; MR, mitral regurgitation; MV, mitral valve; SR, sinus rhythm; AF, atrial fibrillation.

vs in depressed LV function 437 Table 2 Group A patients characteristics regarding ring insertion Age (years) Leaflets involved Degree of MR AF Ring (N Z 16) 61 G 8 Posterior: 13 Moderate 2/4: 5 6 Anterior: 1 Mod. severe 3/4: 6 Both: 2 Severe 4/4:5 No ring (N Z 11) 63 G 12 Posterior: 8 Moderate 2/4: 3 10 Anterior: 1 Mod. severe 3/4: 2 Both: 2 Severe 4/4: 6 Follow-up The mean follow-up period was 6 G 3 years in all patients and was divided in 3 consecutive time periods: 6e18 months after surgery (FU1), 18e36 months (FU2) and O36 months (FU3). All patients were evaluated at the outpatient clinic yearly. Echocardiography All patients were evaluated with transthoracic and transoesophageal echocardiography, within 5 months before surgery. Transthoracic echocardiography was repeated within 3 months after surgery and on an annual base during the follow-up (FU) period. Examinations were performed in the left lateral decubitus position. LV systolic function was estimated from LV end-diastolic and end-systolic diameters (LVEDD, LVESD), fractional shortening (FS), ejection fraction (EF) and velocity time integral of LV outflow track (VTI) with Pulse wave Doppler. EF was calculated using the modified Simpson s method. The size of the left atrium (LA) was also measured and pulmonary artery systolic pressure (PASP) was estimated from tricuspid regurgitation. Mitral regurgitation was evaluated semi-quantitatively by color Doppler imaging at transthoracic and transoesophageal echocardiography and was graded as follows: 1 Z mild, 2 Z moderate, 3 Z moderately severe, 4 Z severe. Table 3 Postoperative patients characteristics p FU time (years) 6 G 3 6G 4 NS Deaths 5 6 NS Re-operation 2 1 NS MR increase a 2 1 NS New AF 3 4 NS a MR increase is defined as an increase of MR degree from 1/4 to 2/4. Statistics Statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS) software. 12 Results were expressed as mean values G SD. Normal distribution of the echocardiographic indices was checked with Kolmogorove Smirnov test. Due to the fact that some measured data were skewed, analysis of the overall differences between subgroups was performed using the non-parametric ManneWhitney test or Kruskalle Wallis test for multiple comparisons. Paired analysis of each variable within the 2 groups was performed using the non-parametric Wilcoxon test or Friedman test for multiple comparisons. A p value!0.05 was considered significant. Results Clinical characteristics of the study population Patients clinical characteristics are presented in Table 1. No patient had any residual mitral regurgitation of any clinical significance at the first postoperative study. In the group, 5 patients were lost within the first year of FU and they were not included in the analysis of the data during FU period. In the group, 2 patients died 2 and 3 months, respectively, following surgery and 3 other were lost within the first year of FU and so the data of all 5 patients in this group were not included in the analysis of the data during FU period. Thus, of the 45 patients studied pre- and immediately postoperatively, 35 patients completed the FU3 period (O36 months) (22 in the group and 13 in the group). In-group analysis Overall changes of the echocardiographic indices postoperatively and during FU are presented in Table 4.

438 N. Kouris et al. Table 4 Changes of the echocardiographic indices postoperatively and during follow-up Pre-op Post-op FU1 FU2 FU3 Pre-op Post-op FU1 FU2 FU3 LVEDD (mm) 61 G 6 53G 6* 52 G 6 50G 7 51G 9 59G 7 55G 8* 55 G 8 57G 8 59G 9x LVESD (mm) 43 G 4 39G 5* 38 G 7 35G 6 37G 8 42G 5 41G 8 41G 8 43G 9 44G 10 FS (%) 28 G 4 27G 5 29G 6 30G 5 29G 8 28G 4 25G 5x 25 G 6 25G 6 24G 9 EF (%) 48 G 6 46G 7 48G 10 50 G 9 48G 13 47 G 6 42G 9{ 43 G 11 42 G 11 40 G 14 VTI (cm) 13 G 3 15G 2* 15 G 2 15G 2 16G 2 13G 3 14G 2 14G 3 13G 4 14G 3 LA (mm) 56 G 10 50 G 9* 49 G 10 51 G 9 52G 9 53G 13 52 G 12 51 G 13 51 G 11 53 G 13 PASP (mmhg) 26 G 5 26G 6 28G 7 28G 6 31G 10 28 G 7 30G 7 32G 9 35G 9U 40 G 10U Values are expressed as mean G SD. FU1: follow-up 6e18 months, FU2: 18e36 months, FU3: O36 months. LVEDD, left ventricular end-diastolic diameter; LVESD, left ventricular end-systolic diameter; FS, fractional shortening; EF, ejection fraction; VTI, velocity time integral at left ventricular outflow tract; LA, left atrium; PASP, pulmonary artery systolic pressure. *p! 0.05 between pre- and postoperative values. xp! 0.02 between pre- and postoperative values and between post- and FU3 values. {p! 0.03 between pre- and postoperative values. Up! 0.05 between post- and FU2 and FU3 values. Values not indicated were NS (see text). Patients in group demonstrated a decrease of LVEDD from 61 G 6 mm preoperatively to 53 G 6 mm early postoperatively (p! 0.001), approaching normal values. The improvement of LVEDD remained unchanged throughout FU (Fig. 1). LVESD was also reduced from 43 G 4mm preoperatively to 39 G 5 mm early postoperatively (p Z 0.002), which was preserved during FU showing a trend for a further decrease (Fig. 2). FS and EF showed a mild deterioration from pre- to postoperative value (28 G 4% / 27 G 5%, p Z 0.2; 48 G 6% / 46 G 7%, p Z 0.25, respectively), but during FU, both indices showed a trend to improve (Fig. 3). Conversely, VTI showed an improvement LVEDD (mm) 65 60 55 50 45 pre-op Left Ventricular End-Diastolic Diameter post-op FU 1 Time FU 2 FU 3 OPER Figure 1 Changes in left ventricular end-diastolic diameter from preoperative value through follow-up time. p 1 : difference between pre- and postoperative value in each group (in-group analysis), p 2 : difference between postoperative and FU3 value in group B (ingroup analysis), p 3 : difference in FU2 and FU3 values between the 2 groups (between-groups analysis). from 13 G 3 cm preoperatively to 15 G 2 cm early postoperatively, (p! 0.05), which was preserved during FU (Fig. 4). The LA size was reduced after surgery (56 G 10 mm / 50 G 9mm,p Z 0.001) and remained unchanged during FU. Finally, PASP was similar preoperatively, postoperatively and during FU (26 G 5mmHg/ 26 G 6mmHg/ 31 G 10 mmhg, p Z NS). In the subgroup of patients there were no differences in LV diameters and EF during FU between patients with or without an annular ring after surgical repair of the mitral valve. However, a borderline increase in FS and EF (p Z 0.066 and p Z 0.05, respectively) was noted at FU1, in patients without an annular ring, compared to those LVESD (mm) 50 45 40 35 30 pre-op Left Ventricular End-Systolic Diameter post-op FU 1 Time FU 2 FU 3 OPER Figure 2 Changes in left ventricular end-systolic diameter from preoperative value through follow-up time. p 1, p 2 : differences between pre- and postoperative value in and group, respectively (in-group analysis) p 3 : difference in FU2 and FU3 values between the 2 groups (between-groups analysis).

vs in depressed LV function 439 EF (%) with an annular ring implanted, but this difference was not evident in FU2 and FU3. In the group, LVEDD showed an early postoperative decrease, compared to its preoperative value (59 G 7 mm versus 55 G 8mm, p! 0.05). Conversely, LVEDD increased at 36 months of FU (post-op versus FU3 value: 55 G 8 mm versus 59 G 9mm,p Z 0.016, Fig. 1). LVESD remained high early postoperatively (Fig. 2), resulting in a significant decrease of FS and EF (pre-op / post-op value: 28 G 4% / 25 G 5%, p! 0.02, 47 G 6% / 42 G 9%, p! 0.03, respectively). The reduction in VTI (cm) 55 50 45 40 35 pre-op 17 16 15 14 13 12 11 10 pre-op post-op Velocity Time Integral post-op Ejection Fraction FU 1 Time FU 1 Time FU 2 FU 2 FU 3 FU 3 OPER OPER Figure 4 Changes in velocity time integral from preoperative value through follow-up time. p: difference between pre- and postoperative value in group (in-group analysis). All other differences were NS. Figure 3 Changes in left ventricular ejection fraction from preoperative value through follow-up time. p 1, p 2 : differences between pre- and postoperative value in and group, respectively (in-group analysis) p 3 : difference in FU2 and FU3 values between the 2 groups (between-groups analysis). FS and EF further increased at 36 months of FU (Fig. 3). VTI and LA size remained practically unchanged. PASP did not change significantly after surgery (28 G 7 mmhg / 30 G 7 mmhg, p Z NS), but at FU3 it increased from the postoperative value (30 G 7 mmhg / 40 G 10 mmhg, p Z 0.005). Finally, a modest correlation was found in both groups, between preoperative and postoperative LVEF (r Z 0.40). Between-groups analysis Patients in the 2 groups did not show any difference in any echocardiographic parameter preoperatively, as well as postoperatively. During FU1, no difference was found in any parameter, but at FU2 patients in group compared to patients in group demonstrated lower LVEDD (50 G 7 mm versus 59 G 8 mm, p! 0.05), lower LVESD (35 G 6 mm versus 43 G 9mm, p! 0.05), higher FS (30 G 5 mm versus 25 G 6 mm, p Z 0.07) and EF (50 G 9 mm versus 42 G 11 mm, p Z 0.07), smaller LA (51 G 9 mm versus 61 G 11 mm, p! 0.05), lower PASP (28 G 6 mmhg versus 35 G 9 mmhg, p! 0.05) and similar VTI (15 G 2 mm versus 13 G 4 mm, p Z 0.3). Finally, similar differences were observed between patients in the and patients in the group during FU3 (Figs. 1e4). Follow-up In group, five patients died at 5 G 1 years after surgery. All patients who died had completed the FU3 period, so their data were included in the analysis. Two other patients underwent due to failure (1 and 2 years after surgery, respectively). Two more patients demonstrated a progressive increase of mitral regurgitation during the FU period, not requiring surgical intervention. In the remaining 18 patients with, the mild residual MR observed early postoperatively, remained unchanged throughout FU period. Three patients developed atrial fibrillation (AF) during FU. None of the patients developed clinical or echocardiographic signs suggestive of coronary artery disease (akinesis or scar). Also, no perioperative myocardial infarction was found in the group. In the group, 2 patients died during early postoperative period (!3 months) and their data were not included in the analysis. Four other patients died at 7 G 4 years having completed the FU3 period, so their data were analyzed. Additionally, 1 patient was re-operated due to prosthetic valve endocarditis and four patients

440 N. Kouris et al. developed AF (2 permanent, 2 paroxysmal) (Group A versus Group B p Z NS, Table 3). Occurrence of AF did not differ between the 2 groups. Furthermore, in the group the mitral prosthesis remained well functioning with the pressure half time remaining unchanged throughout the FU period (PHT post-op: 94 G 35 ms, FU1: 105 G 30 ms, FU2: 88 G 32 ms, FU3: 82 G 27 ms, p Z NS). Finally, only 1 patient demonstrated increase of MR during FU period, not requiring surgical intervention, while in the rest of patients the mild residual MR remained unchanged. Thus, there was no significant difference between the 2 groups in the number of patients with deterioration of MR during FU (Table 3). Discussion Degenerative mitral valve disease is currently the most common cause of mitral regurgitation in Europe. 3 Mitral regurgitation leads to chronic volume overload of the left ventricle, progressively leading to LV dysfunction. 6 The regurgitant volume, ejected into the low-resistance left atrial chamber during systole, increases the total stroke volume and the EF. Since the LVEF, in the presence of severe MR, is often supranormal, the slightest reduction in EF is synonymous to significant LV dysfunction. Myocardial failure is also an important postoperative complication and it is responsible for the majority of postoperative cardiac related mortality and morbidity. 8 Many reports have shown that is superior to, as far as preservation of LV function is concerned. 1e4,13 To our knowledge, there have been no reports on the effects of mitral valve replacement versus repair on an already impaired LV function prior to valve surgery for isolated MR. In our study, LVEDD was significantly reduced in both groups postoperatively. This was an expected finding, since the cause of LV volume overload was eliminated. 14 During FU however, while in patients LVEDD remained within normal limits, in the group it increased again, 3 years after surgery, reaching the preoperative values, without any significant residual MR. This finding could be explained by the fact that in without chordal preservation, LV remodeling alters the shape of the left ventricle toward a spherical geometry (decreased eccentricity index), whether in the LV shape is not affected. Previous reports support this finding, although FU period was shorter. 5,15 It should be also noted that during FU, no perioperative acute myocardial infarction, which develops more frequently during, 16 was found in our group. Consequently, at 6 years FU, patients with demonstrated a lower FS and EF than patients with, compared to the early postoperative study. A possible explanation for these alterations is based upon the aforementioned changes in LV geometry. It has been proposed that, despite expected increase in LV afterload caused by the competent mitral valve, in as well as in with chordal preservation, LV afterload decreases after mitral surgery. 17 Preservation of the subvalvular apparatus in these techniques maintains LV geometry and allows a reduction of LV radius, leading to a reduction in wall stress (LV afterload), according to the Laplace equation. This LV afterload reduction is even more beneficial in patients with preoperatively reduced LV contractility. 6 Conversely, endsystolic wall stress is increased after without chordal preservation due to the loss of chordal support and LV end-systolic volume fails to decrease. 18 Ren and colleagues 15 suggest that preservation of LVEF in and in with chordal preservation is not caused by a decrease in endsystolic volume; the decrease in end-systolic volume is the result of a preserved EF (due to the maintenance of the functional components of the subvalvular apparatus), even though there is a decrease in end-diastolic volume. 15 Our results are in agreement with other reports that showed preservation (not improvement) of LVEF after 1,6,15 and deterioration after 1,13,15 and this preservation is maintained during a long period of time postoperatively after, while it further deteriorates after. The decrease in FS and EF that was noted in both groups at the early postoperative period, is consistent with the reports of Enriquez-Sarano et al., who reported this finding almost 10 years ago, in patients with preoperatively normal systolic function. 2 The weak correlation that was also found in both groups, between preoperative and postoperative LVEF (r Z 0.40), is consistent with previous reports. 19 The data of this cohort demonstrate that patients with MR, who have developed LV dysfunction because they were not operated early enough during the natural history of their valve disease, can benefit from valve reconstruction. However, neither with nor with a normalization of LVEF with mean values O55% could be obtained. Incidence of atrial fibrillation Incidence of AF did not differ between the 2 groups neither preoperatively, nor postoperatively or during FU. Because of the small number of patients, we could not demonstrate whether AF played any role in LV remodeling, although such

vs in depressed LV function 441 a role seems inexplicit to influence the results of one surgical technique over the other. This is supported by Lim et al., 20 who suggested that AF did not affect early outcome of. Ring versus no ring Patients who underwent with concomitant annular ring insertion did not show any significant difference in postoperative LV function, compared to those without ring implantation. The decision whether a ring should be inserted was based upon the surgeon s choice for best anatomical correction of the responsible leaflet, since the 2 subgroups with and without ring were identical, concerning the affected leaflet and the degree of MR preoperatively. The use of annular ring during is controversial and not all surgeons do recommend its implantation, since it improves the results of but, on the other hand, it increases the frequency of systolic anterior motion (SAM). 21 Moreover, the effects of rigid or flexible rings on LV function are also controversial. David et al. 22 reported better LV function after insertion of flexible ring, while Castro et al. did not show any significant difference between the 2 rings. 23 Study limitations It is possible that due to the relatively small number of patients involved, some of our results might have been affected, particularly in the subgroup analysis. The patient s surgical procedure was not randomized preoperatively and was determined by the surgeon at the time of the operation. However, since a randomized trial will probably never be performed, the only acceptable method of comparison of the 2 surgical procedures is to adjust the baseline preoperative parameters. We reported a significant increase of LVEDD in group during 6 years FU. However, subclinical coronary artery disease may not be excluded, as coronary angiography was not performed in any patient postoperatively. The paradoxic septal motion associated with cardiac surgery might raise concern about the use of echocardiography for postoperative assessment of LVEF. This, however, has been attributed to a translational movement rather than a regional wall motion abnormality and it does not impair the ability to assess LV function by echocardiography. 24 Finally, we did not perform any formal quantitation of MR due to the absence of any standardized criteria at the time of the study. Conclusion In this study we report that during a 6-year followup, patients with chronic degenerative MR and LV dysfunction show further deterioration of LV function after. Conversely, appear to better preserve the overall LV function. Mitral valve repair should therefore be the surgical procedure of choice, not only for the lower mortality during the operation itself, but also for the better preservation of LV function in the long term and the level of ejection fraction should not be a reason for denying the benefits of repair to patients with MR. Nevertheless, although seems to preserve LV function, it rarely achieves complete normalization of LVEF and therefore early surgery should be performed before these patients develop LV dysfunction. References 1. Goldman ME, Mora F, Guarino T, Fuster V, Mindich BP. Mitral valvuloplasty is superior to valve replacement for preservation of left ventricular function: an intraoperative twodimensional echocardiographic study. J Am Coll Cardiol 1987;10:568e75. 2. Enriquez-Sarano M, Schaff HV, Orszulak TA, Tajik AJ, Bailey KR, Frye RL. Valve repair improves the outcome of surgery for mitral regurgitation. A multivariate analysis. Circulation 1995;91:1022e8. 3. Lee EM, Shapiro LM, Wells FC. Superiority of mitral valve repair in surgery for degenerative mitral regurgitation. Eur Heart J 1997;18:655e63. 4. Tischler MD, Cooper KA, Rowen M, LeWinter MM. Mitral valve replacement versus mitral valve repair. A Doppler and quantitative stress echocardiographic study. Circulation 1994;89:132e7. 5. Enriquez-Sarano M, Tajik AJ, Schaff HV, Orszulak TA, McGoon MD, Bailey KR, et al. Echocardiographic prediction of left ventricular function after correction of mitral regurgitation: results and clinical implications. J Am Coll Cardiol 1994;24:1536e43. 6. de Varennes B, Haichin R. Impact of preoperative left ventricular ejection fraction on postoperative left ventricular remodeling after mitral valve repair for degenerative disease. J Heart Valve Dis 2000;9:313e20. 7. Enriquez-Sarano M, Tajik AJ, Schaff HV, Orszulak TA, Bailey KR, Frye RL. Echocardiographic prediction of survival after surgical correction of organic mitral regurgitation. Circulation 1994;90:830e7. 8. Enriquez-Sarano M, Schaff HV, Orszulak TA, Tajik AJ, Bailey KR, Frye RL. Congestive heart failure after surgical correction of mitral regurgitation. A long term study. Circulation 1995;92:2496e503. 9. Crawford MH, Souchek J, Oprian CA, Miller DC, Rahimtoola S, Giacomini JC, et al. Determinants of survival and left ventricular performance after mitral valve replacement: Department of Veterans Affairs Cooperative Study on Valvular Heart Disease. Circulation 1990;81: 1173e81.

442 N. Kouris et al. 10. Carpentier A, Relland J, Deloche A, Fabiani JN, D Allaines C, Blondeau P, et al. Conservative management of the prolapsed mitral valve. Ann Thorac Surg 1978;26: 294e302. 11. Carpentier A. Cardiac valve surgery e the French Correction. J Thorac Cardiovasc Surg 1983;86:323e37. 12. SPSS. Statistical Package for the Social Sciences. Windows version, release 10.0. Chicago: SPSS Inc; 2000. 13. Huikuri H. Effect of mitral valve replacement on left ventricular function in mitral regurgitation. Br Heart J 1983;49:328. 14. Tischler MD, Cooper KA, Battle RW, Leavitt BJ. Effect of mitral valve repair for mitral valve prolapse on regression of left ventricular mass. Am J Cardiol 1992;70:1216e7. 15. Ren J, Aksut S, Lighty GW, Vigilante GJ, Sink JD, Segal BL, et al. Mitral valve repair is superior to valve replacement for the early preservation of cardiac function: relation of ventricular geometry to function. Am Heart J 1996;131: 974e81. 16. Obarski TP, Loop FD, Cosgrove DM, Lytle BW, Stewart WJ. Frequency of acute myocardial infarction in valve repairs versus valve replacement for pure mitral regurgitation. Am J Cardiol 1990;65:887e90. 17. Okita Y, Miki S, Veda Y, Tahata T, Sakai T, Matsuyama K. Comparative evaluation of left ventricular performance after mitral valve repair or valve replacement with or without chordal preservation. J Heart Valve Dis 1993;99: 228e37. 18. Rozich TD, Carabello BA, Usher BW, Kratz JM, Bell AE, Zile MR. Mitral valve replacement with or without chordal preservation in patients with chronic mitral regurgitation. Circulation 1992;86:1718e26. 19. Matsumura T, Ohtaki E, Tanaka K, Misu K, Tobaru T, Asano R, et al. Echocardiographic prediction of left ventricular dysfunction after mitral valve repair for mitral regurgitation as an indicator to decide the optimal timing of repair. JAm Coll Cardiol 2003;42:464e5. 20. Lim E, Barlow CW, Hosseinpour AR, Wisbey C, Wilson K, Pidgeon W, et al. Influence of atrial fibrillation on outcome following mitral valve repair. Circulation 2001;104 (12 Suppl.1):I59e63. 21. Bisbos AD, Spanos PK. Surgical correction of mitral valve regurgitation. Hell J Cardiol 2003;44:469e79. 22. David TE, Komeda M, Pollick C, Burns RJ. Mitral valve annuloplasty: the effect of the type on left ventricular function. Ann Thorac Surg 1989;47:524. 23. Castro LJ, Moon MR, Rayhill SC, Niczypornk MA, Ingels Jr NB, Daughters 3rd GT, et al. Annuloplasty with flexible or rigid ring does not alter left ventricular systolic performance, energetics or ventricularearterial coupling in conscious, closed-chest dogs. J Thorac Cardiovasc Surg 1993;105:643. 24. Lehmann KG, Lee FA, McKenzie WB, Barash PG, Prokop EK, Durkin MA, et al. Onset of altered interventricular septal motion during cardiac surgery: assessment by continuous intraoperative transesophageal echocardiography. Circulation 1990;82:1325e34.