The natural history of the dilated cardiomyopathy

Transcription

1 ORIGINAL ARTICLES: CARDIOVASCULAR Mitral Valve Procedure in Dilated Cardiomyopathy: Repair or Replacement? Antonio M. Calafiore, MD, Sabina Gallina, MD, Michele Di Mauro, MD, Filoteo Gaeta, MD, Angela L. Iacò, MD, Stefano D Alessandro, MD, Valerio Mazzei, MD, and Gabriele Di Giammarco, MD Department of Cardiology and Cardiac Surgery, University G. D Annunzio of Chieti, Chieti, Italy Background. Mitral valve (MV) procedure for dilated cardiomyopathy is becoming popular. We analyzed the indications to MV repair or replacement according to our 10-year experience. Methods. From January 1990 to May 2000, 49 patients with dilated cardiomyopathy (12 idiopathic and 37 ischemic) underwent MV operation, 29 repair and 20 replacement. Preoperative evaluation included measurement of MV coaptation depth (CD) as a mirror of the abnormalities of MV apparatus leading to functional mitral regurgitation. Results. Thirty-day mortality was 4.2% (2 patients). One-, 3-, 5-, and 10-year actuarial survival was, respectively, 90%, 87%, 78%, and 73%. The possibility of survival with at least one New York Heart Association functional class improvement was 88%, 76%, 71%, and 65%. Return of functional mitral regurgitation after MV repair was nearly inevitable; however, using a scale from 0 to 4, mean postoperative functional mitral regurgitation was when preoperative MVCD was 10 mm or less and when preoperative MVCD was 11 mm or higher (p < 0.05). Globally, functional results were not influenced by the strategy of treatment (MV repair or replacement). Conclusions. Mitral valve operation can give satisfying survival and good palliation of dilated cardiomyopathy. The MVCD can be helpful in the choice of the surgical strategy on the MV. (Ann Thorac Surg 2001;71: ) 2001 by The Society of Thoracic Surgeons The natural history of the dilated cardiomyopathy (DCM) is often complicated by the appearance of functional mitral regurgitation (FMR). This is not an homogeneous entity, it is related both to changes in the geometry of the left ventricle and of the subvalvular mitral apparatus and to modifications in transmitral pressure and mitral annulus area. The complexity of the interactions among its different components are at the basis of the biphasic pattern of regurgitation (higher during protosystole, lower during midsystole, and again higher during telesystole), typical and exclusive of the FMR. The appearance of FMR has a negative impact on survival of patients with DCM [1 3], with a mortality rate from 40% to 70% after 12 months from the diagnosis of FMR. It is noteworthy that patients with mild FMR also showed a lower survival, about 40% after 12 months [2]. Recently [4, 5] mitral valve (MV) annuloplasty was proposed in selected patients to reduce left ventricle overload and, consequently, end-diastolic pressure. The competence of the MV has a direct effect on the stroke volume, which, independently from the ejection fraction, becomes antegrade, with a favorable effect on the cardiac output. As these concepts are becoming more Accepted for publication Oct 20, Address reprint requests to Dr Calafiore, G. D Annunzio University, Division of Cardiac Surgery, San Camillo de Lellis Hospital, Via C. Forlanini 50, Chieti, Italy; calafiore@unich.it. popular, more patients undergo conventional operation for DCM [6 8]. We reviewed our experience with MV operation in DCM, both idiopathic and ischemic, to evaluate from the analysis of our early and late results if the MV can be always preserved or if there are indications for MV replacement and, in case, what they are. Material and Methods Definition For the purpose of this presentation, DCM was present if (1) diastolic left ventricular size was 110 ml/m 2 or greater, (2) ejection fraction was 35% or lower, and (3) functional mitral regurgitation was present. Clinical Experience From January 1990 to May 2000, 49 patients with DCM (12 idiopathic and 37 ischemic) underwent isolated MV operation. None of them had organic MV disease. Age ranged from 42 to 79 years (mean, years) and 17 patients (34.7%) were 70 years or older. Seventeen patients (34.7%) were younger than 60 years and 5 (10.2%) were previously scheduled for cardiac transplant. Five This article has been selected for the open discussion forum on the STS Web site: by The Society of Thoracic Surgeons /01/$20.00 Published by Elsevier Science Inc PII S (00)

2 Ann Thorac Surg CALAFIORE ET AL 2001;71: SURGICAL TREATMENT OF DILATED CARDIOMYOPATHY 1147 Fig 1. Correlation between mitral valve coaptation depth (MVCD) and (A) left ventricular (LV) sphericity index, (B) mitral annulus size (mm/m 2 ), and (C) ejection fraction. Dotted lines represent 95% confidence interval. patients (10.2%) were women. Twenty-five patients (51.0%) were in New York Heart Association functional class IV and 24 (49.0%) were in class III (mean, ). Mean duration of symptomatic heart failure was years. Preoperative ejection fraction ranged from 15% to 35% (mean, 27% 7%) and was 25% or less in 14 patients (28.6%). Patients with postischemic DCM had concomitant coronary disease, but none of them had angina or ischemic symptoms. Moreover, the dobutamine stress test did not show any evidence of segmental ischemia in 34 patients. Hemodynamic findings showed a mean pulmonary artery pressure of mm Hg and a mean cardiac index of L min 1 m 2. Echocardiographic Evaluation of Functional Mitral Regurgitation The degree of FMR was not the only key point to indicate MV operation, being changeable from time to time. The anatomy of the MV and of its subvalvular apparatus was carefully studied using transesophageal echocardiography. The distance between the mitral annular plane and the coaptation point of the mitral leaflets (MV coaptation depth, MVCD) was mainly taken into consideration. The MVCD is an early consequence of all the mechanisms that can influence the function of the MV apparatus. It is directly related to the left ventricular sphericity (transversal length/longitudinal length), one of the main determinants of FMR both in the clinical [9] and the experimental settings [10], to the annular size and to the ejection fraction (Fig 1). A single value is the mirror of the perturbations of the geometry of the heart and, consequently, of the MV. In 20 healthy controls, its mean value was mm and did not exceed 6 mm (Fig 2). Table 1 shows the preoperative echocardiographic data. Surgical Indications The patients included in this studies fulfilled the criteria previously given. Furthermore, they had normal or only moderate impaired right ventricular function with a mean pulmonary function of less than 40 mm Hg. Renal or liver disease, if present, was not severe. A severely depressed right ventricular function with severe pulmonary hypertension (mean pressure higher than 40 mm Hg) in presence of renal or liver failure is clear contraindication to operation. In this latter group of patients, results are uniformly poor [6]. If FMR is moderate to severe (3/4) or severe (4/4), operation is clearly indicated. In our opinion, in presence

3 1148 CALAFIORE ET AL Ann Thorac Surg SURGICAL TREATMENT OF DILATED CARDIOMYOPATHY 2001;71: Surgical Technique Intermittent antegrade warm blood cardioplegia was always used for myocardial protection [11]. If necessary, coronary artery bypass grafting was performed first. The MV was exposed through the interatrial septum; if the visual field was not satisfactory, the superior approach [12] was used. For this purpose, the superior vena cava was always cannulated directly. Mitral valve annuloplasty (n 29) was obtained by two different techniques. A gluteraldehyde-treated homologous pericardium, 52 mm long, was inserted between the two trigons in 10 patients; as the length was equivalent to a 26-ring size, the annulus was overreduced. In the remaining 19 patients, a De Vega-like annuloplasty with a single 2/0 TICRON (TI-CRON, Sherwood Medical, St. Louis, MO) was performed, going twice from the posteromedial to the anterolateral trigon and tying the suture at this level on a pledget. A 26 sizer was applied inside the valve to evaluate if the mitral annulus was reduced to the proper size. Mitral valve replacement was done on 20 patients limiting the excision of the valve only to a triangle of the anterior leaflet, which leaves in place all the chordae and, consequently, the papillary muscles. The interrupted pledgeted U sutures are brought to the limit of the leaflets, attaching the papillary muscles to the annulus. In this way the subvalvular apparatus and its function are preserved. When a tricuspid annuloplasty was needed (17 patients), it was performed after the MV operation. A De Vega-like procedure with a 2/0 Ticron was used. Coronary bypass grafting was performed in 33 patients, even if no ischemia was detected. Seventy-four coronary vessels were grafted (mean, distal anastomoses per patient): 27, left anterior descending artery; 4, diagonal; 17, obtuse marginal; and 19, right coronary artery or one of its branches. Before unclamping the aorta, all the patients received electively 5 g kg 1 min 1 of dobutamine. If necessary, stronger inotropic support with adrenaline and intraaortic balloon pump was used. If preoperative creatinine was higher than 1.4 mg/l when the pump was off, a continuous infusion of dopamine (3 g kg 1 min 1, renal range) and furosemide (20 mg/h) was always started. Fig 2. Mitral valve coaptation depth in the healthy subject. of moderate FMR (2/4), MV operation is always indicated. This FMR degree, with a normally contracting heart and normal volumes, represents a small portion of the cardiac output. In a dilated heart with a low ejection fraction (50% of the stroke volume), there is a need for correction. Choice Between Mitral Valve Repair and Replacement At the beginning of our experience (1990 to 1993, 9 patients) the MV was always replaced. Later on, we tried to repair the MV at each occasion. However, as we considered MV replacement in these patients, our strategy changed. The anatomy of the MV is the key point of the choice of technique. If the MVCD is between 7 and 10 mm, MV annuloplasty must be performed. However, if the MVCD exceeds 10 mm, the alterations of the MV subvalvular apparatus are such that FMR is expected to return in a short period of time (Fig 3). In this situation MV replacement must be performed. Postoperative Course After the operation, the patients were admitted to the intensive care unit and then to the surgical and cardiologic wards. Follow-up data were obtained 3, 6, 9, and 12 months after operation and then every 6 months. Echocardiographic examination was performed every time patients were seen at our outpatients clinic; if the patient could not come to the clinic, the referring cardiologist sent us clinical information. Follow-up was 100% complete. Table 1. Preoperative Echocardiographic Data Data EDv (ml/m 2 ) ESv (ml/m 2 ) SV (ml/m 2 ) EF (%) 27 8 Sphericity index Mitral annulus (mm/m 2 ) MVCD (mm) Mitral regurgitation (1 to 4) grade 4 17 grade 3 22 grade 2 9 EDv end-diastolic volume; ESv end-systolic volume; EF ejection fraction; MVCD mitral valve coaptation depth; SV stroke volume.

4 Ann Thorac Surg CALAFIORE ET AL 2001;71: SURGICAL TREATMENT OF DILATED CARDIOMYOPATHY 1149 Fig 3. Perioperative transesophageal echocardiography. (A) Mitral annulus was dilated (25 mm/m 2 ) and (B) severe functional mitral regurgitation was present. Mitral valve coaptation depth was 13 mm. This patient underwent mitral valve repair. (C) Despite reduction annuloplasty (16 mm/m 2 ), moderate functional mitral regurgitation persisted. The patient had a successful mitral valve replacement. Statistical Analysis Results are expressed as mean value standard deviation unless otherwise indicated. Statistical analysis comparing two groups was performed with unpaired twotailed t tests for the means or 2 test for categoric variables. Survival and event-free survival curves were obtained with the Kaplan-Meier method (SPSS Software, Chicago, IL). The statistical significance was calculated with the log-rank. p value less than 0.05 was considered significant. Table 2. Operative Data Operation Data Mitral valve repair 29 Isolated 2 CABG 18 Tricuspid repair 5 CABG tricuspid repair 4 Mitral valve replacement 20 Isolated 4 CABG 8 Tricuspid repair 5 CABG tricuspid repair 3 CPB time (min) Ao cross-clamping time (min) Ao aortic; CABG coronary artery bypass grafting; CPB cardiopulmonary bypass. Results Early Mortality and Morbidity Tables 2 and 3 show, respectively, the operative and the postoperative data. Two patients (4.1%) died in the first 30 days after operation. The causes of death were multiorgan failure on the fifth postoperative day and supradiaphragmatic aortic rupture due to an undiagnosed perforating aortic ulcer. The death happened in the seventh postoperative day with the patient in perfect hemodynamic conditions. Intraaortic balloon pump was not used in this patient at anytime. All the patients had elective inotropic support from 11 to 141 hours (mean, hours). Three patients needed intraaortic balloon pump assistance in the operating room and 1 patient needed it during the intensive care unit stay. Mean intensive care unit stay was hours. Five patients needed a second readmission in the intensive care unit and were all redischarged after a Table 3. Postoperative Data Data Deaths 2 (4.1%) AMI... CVA 1 (2.0%) IABP 4 (8.2%) OR 3 (6.1%) ICU 1 (2.0%) Acute renal failure 1 (2.0%) Acute respiratory failure 2 (4.1%) Bleeding (ml/12 h) Redo for bleeding 4 (8.2%) Transfused patients 23 (47.0%) CK-MB peak ICU stay (h) Readmission in the ICU 5 (10.6%) Awaking time (h) Ventilation time (h) In-hospital stay (d) AMI acute myocardial infarction; CVA cerebrovascular accident; IABP intraaortic balloon pump; ICU intensive care unit; OR operating room.

5 1150 CALAFIORE ET AL Ann Thorac Surg SURGICAL TREATMENT OF DILATED CARDIOMYOPATHY 2001;71: Fig 4. Actuarial survival. mean of hours. Mean postoperative stay in the Division of Cardiac Surgery was days. From there, all the patients were sent in the cardiologic ward. Postoperative course was globally days. Chronic medical treatment included angiotensinconverting enzyme inhibitors, diuretics, and -blockers like carvedilol. Late Survival After a mean of months, 11 patients (23.4%) died; 5 of them had MV repair and 6 MV replacement. Causes of death were cardiac in 10 (2 sudden death and 8 heart failure) and noncardiac in 1 (malignancy). One-, 3-, 5-, and 10-year actuarial survival was, respectively, 90%, 87%, 78%, and 73% (Fig 4). Five-year survival in ischemic and idiopathic patients was similar (78% 10% versus 75% 14%, p not significant). Patients who had MV repair had better 5-year survival than patients who had MV replacement, although not statistically significant (83% 7% versus 70% 10%). Functional Results In the 36 survivors, after a mean follow-up of months New York Heart Association functional class decreased from to (p 0.001). The possibility of survival with at least one functional class improvement 1, 3, 5, and 10 years after operation was, respectively, 88%, 76%, 71%, and 65% (Fig 5) and was higher in patients who had repair rather than replacement, although the values were not statistically significant (76% 8% versus 65% 11% after 5 years). Echocardiographic Results Twenty-eight of the 34 survivors were followed up with serial echocardiographic evaluations (Table 4). The MV repair gave similar results to the MV replacement, although this latter group included the most dilated patients. Volumes, stroke volumes, and ejection fractions did not change, but the New York Heart Association functional class was lower than the preoperative one, independently from the surgical procedure on the MV. Residual FMR was present, in different degrees, in every patients who had MV repair except for 2 patients. However, if we group the echocardiographic results according to the MVCD (10 mm or less and 11 mm and more), patients who had MV repair showed some difference in the late outcome (Table 5). If MV repair was performed in the group of patients with MVCD 11 mm or more, postoperative New York Heart Association functional class remained similar to the preoperative one and residual FMR degree was higher than in the other group of patients ( versus , p 0.006). This was not attributable to annular redilatation, as its size was similar in both group of patients ( mm versus mm, p not significant). In 14 patients who had MV repair, followed every 3 months for 1 year, FMR reappeared during the first 6 months and then remained constant. This pattern was independent from MVCD, but the degree of FMR was mainly mild if MCVD was 10 mm or less, and mainly moderate if it was 11 mm or higher. The mitral annulus never redilated (Table 4). Comment Functional mitral regurgitation is a key point in the natural history of DCM. Its appearance increases the degree of heart failure and shortens life expectancy [1 3]. Mechanisms of FMR are several. Left ventricular enlargement causes displacement of both papillary muscles posterolaterally and apically. Geometric distortion of the mitral apparatus is its natural consequence. These changes increase the distance over which the mitral leaflets are tethered from the papillary muscles to the anterior annular ring and restrict their possibility to close at the annular level [13]. The leaflets take a tented geometry, the length of leaflet coaptation is reduced, the excursion angle of the posterior leaflet decreases, and the coaptation depth of the leaflets increases. When FMR starts, the mitral annular area increases together with the dilatation of the mitral annulus and a consequent increase of the base of the heart. With the same level of papillary muscle displacement, dilation of the annulus causes a higher degree of FMR. The combination of posterolateral and apical displacement and dilated annulus gives the highest degree of FMR. Mitral regurgitation that follows postischemic DCM can have, if a papillary muscle is involved in the ischemic event, a different start, as restricted motion of a leaflet Fig 5. Actuarial possibility to be alive and improved by at least one New York Heart Association functional class.

6 Ann Thorac Surg CALAFIORE ET AL 2001;71: SURGICAL TREATMENT OF DILATED CARDIOMYOPATHY 1151 Table 4. Postoperative Echocardiographic and Clinical Evolution Total (n 28) MV Repair (n 16) MV Replacement (n 12) Pre Post p Value Pre Post p Value Pre Post p Value EDv (ml/m 2 ) NS NS NS ESv (ml/m 2 ) NS NS NS SV (ml/m 2 ) NS NS NS EF (%) NS NS NS Annulus (mm) FMR No residual FMR 2 12 NYHA class Follow-up (mo) EDv end-diastolic volume; EF ejection fraction; ESv end-systolic volume; FMR functional mitral regurgitation; NS not significant; NYHA New York Heart Association; SV stroke volume. (generally the posterior one) can be the basis of the process. However, when left ventricular dilation starts, the further mechanisms of FMR are similar to those described previously. Surgical possibilities, when heart failure is not controlled pharmacologically, are directed to the restoration of the competence of the MV, with a concomitant reduction of the base of the heart [4, 5]. This effect causes a reduction of the left ventricular volume, as the base of the cardiac cone becomes smaller. With time, a favorable remodeling of the left ventricle can follow, with further reduction of the volume; however, as demonstrated by our experience, this is not always the case. The purpose of operation is to maintain the same stroke volume as preoperatively, which is antegrade, with a concomitant increase in cardiac output. According to this viewpoint, as operation being only palliatory, the possibility of returning to the situation where symptoms could be more easily controlled by medical treatment, can be considered a success of the procedure. To decide when FMR has to be treated, moderate to severe or severe FMR is the obligatory surgical indication. However, in our opinion, moderate FMR, in any patient symptomatic of heart failure, is huge enough to justify its correction. An accurate study of the anatomy of the MV will provide the real potential for FMR that, at the moment of the evaluation, can change according different factors (relative hypovolemia, vasodilation, vasoconstriction due to anxiety, etc). The higher the ventricular volume, the lower the ejection fraction, the more aggressive the MV operation must be. To correct FMR, MV can be repaired or replaced. Mitral annuloplasty was considered the standard to correct mitral regurgitation [4, 5]. However, FMR can reappear in the follow-up of patients treated with ring annuloplasty [14]. The reduction of posterior annulus limits annular area, improving coaptation. However, ring insertion shifts the posterior annulus toward the fixed anterior annulus, increasing the distance between the papillary muscles and the posterior annulus and further reducing its ability to move anteriorly and to coapt. This can still cause FMR [13], even with an annular overreduction [15]. Table 5. Postoperative Echocardiographic and Clinical Evolution in Patients Who Had Mitral Valve Repair, Grouped According to the Coaptation Depth of Mitral Leaflets MVCD 10 mm (n 11) MVCD 10 mm (n 5) Pre Post p Value Pre Post p Value EDv (ml/m 2 ) NS NS ESv (ml/m 2 ) NS NS SV (ml) NS NS EF (%) NS NS Annulus (mm) FMR No residual FMR 2... NYHA class NS Follow up (mo) EDv end-diastolic volume; EF ejection fraction; ESv end-systolic volume; FMR functional mitral regurgitation; NS not significant; NYHA New York Heart Association; SV stroke volume.

7 1152 CALAFIORE ET AL Ann Thorac Surg SURGICAL TREATMENT OF DILATED CARDIOMYOPATHY 2001;71: In our opinion, in certain patients FMR is difficult to correct due to the deep geometric alterations among the components of the MV apparatus, and MV replacement has to be considered. We found that MVCD is a simple measurement that can give us an indication of the altered leaflet tethering geometry. If it is 11 mm or more, MV, preserving intact its subvalvular apparatus, must be replaced. Two goals are fulfilled: the mitral annulus is fixed to a certain size, reducing the base of the heart (a 25- or 27-mm prosthetic valve is generally inserted) and FMR will not be a problem in the future. In these patients, if MV repair is performed, return of FMR is very likely to happen, with a degree that can impair the functional result of the procedure. We observed that when MV repair was performed in patients with a MVCD 11 mm or more, return of FMR was the rule and all the patients had some degree of FMR (mean, 2.5). On the contrary, when MVCD was 10 mm or less, FMR was generally mild to moderate (mean, 1.2). This was not attributable to mitral annulus redilation, as its size remained constant with time. We believe that there are indications both for MV repair and for MV replacement and, in selected patients, this latter solution has to be preferred. The MVCD gives the indication of the geometry of the left ventricle and can identify the patients in whom recurrence of mitral regurgitation is more likely to occur. Midterm results in our patients and in the series of other investigators [4, 5, 7, 8, 16] show satisfactory survival rates and functional palliation. A small increase in ejection fraction and stroke volume was observed, but this was not significant. However, stroke volume becomes mainly antegrade, with a concomitant increase in cardiac output. The great benefit of the procedure is the restoration of MV competence or the reduction of FMR, as observed in patients who had MV replacement or MV repair with MVCD 10 mm or less. Modifications of left ventricle volume were not important and were not related to the clinical improvement. How long this situation will remain constant is difficult to say. As patients always have DCM and remain with a DCM after operation, our purpose is to obtain a palliation that will reduce symptoms and increase life expectancy, often related to the competence of the MV. Aggressive medical treatment, which has recently made gigantic improvements, will allow an improvement in the quality of life, reducing as much as possible the progression of the disease. References 1. Romeo F, Pelliccia F, Cianfrocca C, et al. Determinants of end-stage idiopathic dilated cardiomyopathy: a multivariate analysis of 104 patients. Clin Card 1989;12: Blondheim DS, Jacobs LE, Kotler MN, Costacurta GA, Parry WR. Dilated cardiomyopathy with mitral regurgitation: decreased survival despite a low frequency of left ventricular thrombus. Am Heart J 1991;122: Junker A, Thayssen P, Nielsen B, Andersen PE. The hemodynamic and prognostic significance of echo-dopplerproven mitral regurgitation in patients with dilated cardiomyopathy. Cardiology 1993;83: Bolling SF, Deeb GM, Brunsting LA, Bach DS. Early outcome of mitral valve reconstruction in patients with end-stage cardiomyopathy. J Thorac Cardiovasc Surg 1995;109: Bolling SF, Pagani FD, Deeb GM, Bach DS. Intermediateterm outcome of mitral reconstruction in cardiomyopathy. J Thorac Cardiovasc Surg 1998;115: Calafiore AM, Gallina S, Contini M, Iacó AL, Barsotti A, Gaeta F, Zimarino M. Surgical treatment of dilated cardiomyopathy with conventional technique. Eur J Cardiothorac Surg 1999;16:S Bishay ES, McCarthy PM, Cosgrove PM, et al. Mitral valve surgery in patients with severe left ventricular dysfunction. Eur J Cardiothorac Surg 2000;17: Buffolo E, Paula IA, Palma H, Branco JN. A new surgical approach for treating dilated cardiomyopathy with mitral regurgitation. Arq Bras Cardiol 2000;74: Kono T, Sabbah HN, Stein PD, Brymer JF, Khaja F. Left shape as a determinant of functional mitral regurgitation in patients with severe heart failure secondary to either coronary artery disease or idiopathic dilated cardiomyopathy. Am J Cardiol 1991;68: Kono T, Sabbah HN, Rosman H, Alam M, Jafri S, Goldstein S. Left ventricular shape is the primary determinant of functional regurgitation in heart failure. J Am Coll Card 1992;20: Calafiore AM, Teodori G, Mezzetti A, et al. Intermittent antegrade warm blood cardioplegia. Ann Thorac Surg 1995; 59: Guirondon GM, Ofiesh JG, Kaushik R. Extended vertical transatrial septal approach to the mitral valve. Ann Thorac Surg 1991;52: Otsuji Y, Handschumaker MD, Schwammenthal E, et al. Insights from three-dimensional echocardiography into the mechanism of functional mitral regurgitation. Direct in vivo demonstration of altered leaflet tethering geometry. Circulation 1997;96: Liehl Cohen N, Otsuji Y, Viahakes GJ, Akins CW, Levine RA. Functional ischemic regurgitation can persist despite ring annuloplasty: mechanistic insights [Abstract]. Circulation 1997;96(Suppl 1): Hung J, Handschumaker MD, Rudski L, Chow C-M, Guerrero JL, Levine RA. Persistence of ischemic mitral regurgitation despite annular ring reduction: mechanistic insights from 3D echocardiography [Abstract]. Circulation 1999; 100(Suppl 1): Bach DS, Bolling ST. Improvement following correction of secondary mitral regurgitation in end-stage cardiomyopathy with mitral annuloplasty. Am J Cardiol 1996;78: INVITED COMMENTARY Doctor Calafiore and associates should be congratulated for the results of their series summarized in this manuscript. In this series, they retrospectively analyzed the results of mitral valve surgery in 49 patients with dilated cardiomyopathy over a ten-year period. Their results were quite good with a 30-day mortality of 4% and a 5- and 10-year actuarial survival of 78% and 73%, respectively. The patients also had an improvement in their 2001 by The Society of Thoracic Surgeons /01/$20.00 Published by Elsevier Science Inc PII S (01)