Mitral valve surgery for chronic ischemic mitral regurgitation

Antonio M. Calafiore, MD, Michele Di Mauro, MD, Sabina Gallina, MD, Gabriele Di Giammarco, MD, Angela L. Iaco`, MD, Giovanni Teodori, MD, and Isabella Tavarozzi, MD Division of Cardiac Surgery, University Hospital, Torino, and Department of Cardiology and Cardiac Surgery, “G D’Annunzio” University, Chieti, Italy

Background. Early and midterm clinical and echocardiographic results after mitral valve (MV) surgery for chronic ischemic mitral regurgitation were investigated to evaluate the validity of the criteria for repair or replacement applied by us. Methods. From 1988 to 2002, 102 patients with ischemic mitral regurgitation underwent MV surgery (82 repairs and 20 replacements). End-systolic distance between the coaptation point of mitral leaflets and the plane of mitral annulus was the key factor that allowed either repair (<10 mm) or replacement (>10 mm). Patients who had MV replacement showed higher New York Heart Association class (3.2 ⴞ 0.5 versus 3.4 ⴞ 0.5; p ⴝ 0.016), lower preoperative ejection fraction (0.33 ⴞ 0.9 versus 0.38 ⴞ 0.12; p ⴝ 0.034), and higher end-diastolic volume (161 ⴞ 69 mL versus 109 ⴞ 35 mL; p < 0.001) compared with repair. Mitral regurgitation was 3.2 ⴞ 0.7 in both groups. Results. Thirty-day mortality was 3.9% (2.4% MV repair versus 10.0% MV replacement; not significant). During the follow-up 26 patients died. Of the 72 survivors, 55

(76.3%) were in New York Heart Association classes I and II. Five-year survival was 75.6% ⴞ 4.7% in MV repair and 66.0% ⴞ 10.5% in MV replacement (not significant). Survival in New York Heart Association classes I and II was 58.9% ⴞ 5.4% in MV repair and 40.0% ⴞ 11.0% in MV replacement (not significant). Cox analysis identified preoperative New York Heart Association class, ejection fraction, end-diastolic volume, end-systolic volume, and congestive heart failure as risk factors common to both events. In 46 patients, late echocardiograms showed no volume or ejection fraction modifications. In patients who underwent MV repair, 50% had no or mild mitral regurgitation. Conclusions. Correction of chronic ischemic mitral regurgitation through either repair or replacement provides a good 5-year survival rate, with more than 75% of the survivors in New York Heart Association classes I and II. (Ann Thorac Surg 2004;77:1989 –97) © 2004 by The Society of Thoracic Surgeons

S

In most cases both PMs were normal (functional MR), and MR was caused by failure in coaptation of mitral leaflets because of their restricted motion. This could be caused by either global left ventricular (LV) dilation with posterolateral and apical displacement of both PMs or local malfunction of the LV wall adjacent to the posterior PM. The regurgitant jet was often central. However, when only the posterior PM was involved, an asymmetric pattern of MV deformation from medial to lateral side of the MV could be present, showing funnel-shaped deformity of the medial side and prolapselike deformity on the lateral side. This might develop as a result of preserved or excessive motion of the nontethered lateral side of the anterior leaflet [9]. As a consequence, in some cases, the regurgitant jet was double, one central and one lateral (Fig 1).

urgery for mitral regurgitation of ischemic origin (IMR) is not yet standardized. Whereas there is general agreement that the presence of IMR has a negative influence on survival after completed myocardial infarction (MI) [1–3], the necessity of correction of IMR is questioned [4, 5], and the benefit of mitral valve (MV) repair versus replacement is debated [6 – 8]. We evaluated our experience in patients with IMR to asses surgical indications, criteria for MV repair or replacement, and long-term results of surgery.

Material and Methods Definitions Mitral regurgitation (MR) was defined as being ischemic in origin as evidenced by clinical data and echocardiographic findings. Mitral leaflets and chordae were normal, and regurgitation was the result of completed MI, which is always present in the history of each patient. The posterior papillary muscle (PM) was rarely infarcted. Accepted for publication Nov 7, 2003. Address reprint requests to Dr Calafiore, Division of Cardiac Surgery, “S Giovanni Battista” Hospital, c.so Bramante 86, Torino, Italy; e-mail: [email protected].

© 2004 by The Society of Thoracic Surgeons Published by Elsevier Inc

Patient Population From June 1988 to December 2002, 102 patients with IMR underwent MV surgery; 82 of them had repair and 20 had replacement. Patients with organic MV disease with intermittent ischemia or with PM rupture were not included in this study. During the same time frame 5,768 patients had isolated myocardial revascularization and 205 underwent MV repair for nonischemic MR. 0003-4975/04/$30.00 doi:10.1016/j.athoracsur.2003.11.017

CARDIOVASCULAR

Mitral Valve Surgery for Chronic Ischemic Mitral Regurgitation

1990

CALAFIORE ET AL MITRAL VALVE SURGERY

Ann Thorac Surg 2004;77:1989 –97

CARDIOVASCULAR

(31.3%). In the remaining 80 cases (69.7%), only the posterior PM was malfunctioning, and in nine of them it was surely infarcted. The presence and entity of MR were evaluated by using colored areas of jet regurgitation and jet-to-left atrium area ratios [11]. Severity was graded as mild (1/4, if regurgitant area was up to 4 cm2), moderate (2/4, from 4 to 6 cm2), moderate to severe (3/4, from 6 to 8 cm2), and severe (4/4, more than 8 cm2). In 72 cases the left atrium area was calculated, and the percentage of regurgitant area was considered 1/4 if the regurgitant area was less than 25% of the left atrium area, 2/4 if it was from 25% to 40%, 3/4 if it was from 40% to 50%, and 4/4 if it was higher than 50%. Table 2 shows the echocardiographic findings.

Surgical Indication

Fig 1. The regurgitant jet is generally central (A), but can be double (B) owing to asymmetric deformity of the mitral valve.

Preoperative clinical data are shown in Table 1. Four patients with mild aortic stenosis were included in the study because they underwent prophylactic surgery.

Echocardiograms All the patients had a preoperative transthoracic echocardiogram. The mitral annulus was identified as the leaflet hinge point, and its size was measured in the apical long axis, four- and two-chamber views at the end of systole; the mean value was considered. The distance between the point where the MV leaflets coapt and the mitral annulus plane (coaptation depth, CD) was measured at end-systole in the four-chamber apical long axis view. Left ventricular volumes (end-diastolic and endsystolic) were measured according to the biapical Simpson disk method [10] indexed to body size, and ejection fraction (EF) was calculated. Mitral valve anatomy was assessed. In every case mitral leaflets and chordae were normal. The PMs were normal, but in 32 cases they were both malpositioned

Mitral regurgitation, if 3/4 or 4/4, was always corrected. When MR was 2/4, surgery was indicated in the presence of a dilated LV (end-diastolic volume ⬎ 110 mL/m2) or low EF (ⱕ0.35), as in the case of dilated cardiomyopathy [12]. All the patients had a standard monitoring, including a Swan Ganz catheter. Transesophageal echocardiography was obtained in 92 patients. Care was taken on the anatomy of the MV, being the degree of MR often downgraded [13] owing to the peculiar hemodynamic situation in the operating room. The ascending aorta was cannulated for arterial inflow, whereas the inferior vena cava was cannulated as usual, and the superior vena cava was directly cannulated. Moderately hypothermic perfusion and antegrade cold crystalloid cardioplegia were used in 6 patients. In the remaining 96, normothermic perfusion and intermittent antegrade warm blood cardioplegia were used. Surgical approach was always transseptal. In case of incomplete vision, the incision was continued to reach the roof of the left atrium. Mitral annuloplasty always involved the posterior annulus and both commissures, and it was obtained by means of a suture annuloplasty or by a 40-mm-long autologous pericardial strip, as shown by most recent experiences [14]. A specially designed 40-mm-long ring (Sovering Miniband, Sorin Biomedica, Saluggia, Italy) was used in 5 patients. When the MV was replaced, the valve remained intact in all its components, and only a small triangle of the anterior leaflet was excised (Fig 2) to preserve the integrity of the subvalvular apparatus. At the end of the procedure, all patients electively received 5 ␮g · kg⫺1 · min⫺1 of dobutamine and either nitroglycerin or sodium nitroprusside according to arterial resistance. Other inotropic agents, as well as an intraaortic balloon pump, were used when necessary.

Indication for Repair or Replacement Ischemic mitral regurgitation, as the LV is involved in the mechanisms that generate it, behaves as a functional MR, and its degree of regurgitation is not constant. For this reason, in our opinion, the anatomy must be carefully evaluated because it represents a stable element. Transthoracic echocardiogram or transesophageal echocardio-

Ann Thorac Surg 2004;77:1989 –97

CALAFIORE ET AL MITRAL VALVE SURGERY

1991

Variable Age (y) ⬍50 50 –59 60 – 69 ⱖ70 Sex Male Female Preoperative NYHA class II III IV CHF Angina Previous AMI Anterior (or septal) Lateral Inferior (or posterior) Time from AMI to surgery (mo) AMI ⬍ 1 month Previous cardiac operation Preoperative IABP Dilated cardiomyopathy Diabetes CRF (Cr ⱖ 2.0 mg/dL) Hypertension ECV Chronic atrial fibrillation

All (n ⫽ 102)

MV Repair (n ⫽ 82)

MV Replacement (n ⫽ 20)

p Value

66.5 ⫾ 8.5 5 (4.9%) 15 (14.7%) 39 (38.2%) 43 (41.2%)

66.6 ⫾ 8.3 3 (3.7%) 12 (14.6%) 33 (40.2%) 34 (41.6%)

66.2 ⫾ 9.7 2 (10.0%) 3 (15.0%) 6 (30.0%) 9 (45.0%)

NS NS NS NS NS

79 (77.5%) 23 (22.5%) 3.3 ⫾ 0.5 3 (2.9%) 69 (67.6%) 30 (29.5%) 59 (57.8%) 61 (59.8%) 111 36 (32.4%) 7 (6.3%) 68 (61.3%) 72 ⫾ 82

62 (75.6%) 20 (24.4%) 3.2 ⫾ 0.5 3 (3.7%) 57 (69.5%) 22 (26.8%) 42 (51.2%) 49 (59.8%) 89 27 (30.3%) 5 (5.6%) 57 (64.0%) 62 ⫾ 76

17 (85.0%) 3 (15.0%) 3.4 ⫾ 0.5 0 12 (60.0%) 8 (40.0%) 17 (85.0%) 12 (60.0%) 22 9 (45.0%) 2 (10.0%) 11 (55.0%) 107 ⫾ 88

NS NS 0.016 NS NS NS 0.006 NS NS NS NS NS NS

3 (2.9%) 10 (9.8%)

2 (2.4%) 8 (9.7%)

1 (5.0%) 2 (10.0%)

NS NS

3 (2.9%) 69 (67.4%) 29 (28.4%) 7 (6.8%) 53 (52.0%) 15 (14.7%) 22 (21.6%)

3 (3.7%) 52 (63.4%) 26 (31.7%) 6 (7.3%) 47 (57.3%) 14 (17.1%) 19 (23.2%)

0 17 (85.0%) 3 (15.0%) 1 (5.0%) 6 (30.0%) 1 (5.0%) 3 (15.0%)

NS NS NS NS 0.028 NS NS

AMI ⫽ acute myocardial infarction; CHF ⫽ chronic heart failure; Cr ⫽ creatinine; CRF ⫽ chronic renal failure; ECV ⫽ extracardiac vasculopathy; IABP ⫽ intraaortic balloon pump; MV ⫽ mitral valve; NS ⫽ not significant; NYHA ⫽ New York Heart Association.

gram gives us important information, such as annular diameter, tenting area, restricted motion of one or both mitral leaflets, and CD. We consider this latter measure a mirror of the abnormalities of the subvalvular apparatus. Inasmuch as, with time, PMs move posteriorly, laterally, and apically, the forces that tether the mitral leaflets increase, limiting their movements toward the annulus. Coaptation depth is proportional to the tethering forces, giving us an idea of the relation between PMs and the annular insertion of MV leaflets. The higher the CD, the more important the PMs displacement and the more dilated the midportion of the LV. From a retrospective analysis [15], we found that when CD is 10 mm or less, MV repair has good midterm results, but when CD is 11 mm or more, abnormalities in the subvalvular apparatus prevent proper coaptation of mitral leaflets, causing residual MR. In this case, MV replacement is indicated.

Postoperative Course Patients remained in the intensive care unit until the inotropic support was represented only by dobutamine, then they were transferred to the surgical ward. In most

cases oral treatment with ␤-adrenergic blockers and angiotensin-converting enzyme inhibitors was started between the second and fourth postoperative days. From the surgical ward the patients were discharged to the cardiology ward and then to cardiac rehabilitation. All the patients were followed up at our outpatient clinic 3, 6, and 12 months postoperatively, and then yearly. At that moment, when possible, a transthoracic echocardiogram was performed by our cardiologists. Follow-up was 100% complete.

Statistical Method Results are expressed as mean ⫾ standard deviation unless otherwise indicated. Statistical analysis comparing two independent groups was performed with unpaired two-tailed Student’s t test for the means or ␹2 test for categorical variables. Preoperative and postoperative data were compared with paired two-tailed Student’s t test. Preoperative, echocardiographic, and operative variables considered in the study are listed in the appendix. Logistic regression was used to identify risk factors for early mortality. Actuarial curves were obtained with the Kaplan-Meier method. The statistical significance was

CARDIOVASCULAR

Table 1. Preoperative Clinical Data

1992

CALAFIORE ET AL MITRAL VALVE SURGERY

Ann Thorac Surg 2004;77:1989 –97

Table 2. Echocardiographic Data Variable CARDIOVASCULAR

EF ⱕ0.35 0.36 – 0.50 ⬎0.50 EDv (mL/m2) ESv (mL/m2) SV (mL/m2) MR 2⫹ 3⫹ 4⫹ Regurgitant jet Single Double Annulus (mm) MVCD (mm) Wall abnormalities Akinesia/dyskinesia Anterior Septal Lateral Inferior posterior hypokinesia anterior septal lateral inferior posterior

All (n ⫽ 102)

MV Repair (n ⫽ 82)

MV Replacement (n ⫽ 20)

p Value

0.37 ⫾ 0.12 60 (58.8%) 28 (27.4%) 14 (13.7%) 116 ⫾ 44 76 ⫾ 37 40 ⫾ 14 3.2 ⫾ 0.7 18 (17.6%) 41 (40.2%) 43 (42.2%)

0.38 ⫾ 0.12 44 (53.7%) 25 (30.5%) 13 (15.9%) 109 ⫾ 35 70 ⫾ 29 39 ⫾ 18 3.1 ⫾ 0.8 13 (15.9%) 33 (40.2%) 36 (43.9%)

0.33 ⫾ 0.09 16 (80.0%) 3 (15.0%) 1 (5.0%) 161 ⫾ 69 119 ⫾ 53 41 ⫾ 16 3.3 ⫾ 0.7 5 (25.0%) 8 (40.0%) 7 (35.0%)

0.034 0.032 NS NS ⬍0.001 ⬍0.001 NS NS NS NS NS

91 (89.2%) 11 (10.8%) 38.2 ⫾ 7.0 9.2 ⫾ 2.0

72 (87.8%) 10 (12.8%) 38.4 ⫾ 7.3 9.1 ⫾ 2.2

19 (95.0%) 1 (5.0%) 36.5 ⫾ 5.4 11.0 ⫾ 1.8

NS NS NS ⬍0.001

144 25 (17.4%) 23 (16.0%) 6 (4.2%) 66 (45.7%) 24 (16.7%) 194 45 (23.2%) 44 (22.7%) 40 (20.6%) 22 (11.3%) 43 (22.2%)

111 18 (16.3%) 17 (15.3%) 4 (3.6%) 55 (49.5%) 17 (15.3%) 162 38 (23.4%) 36 (22.3%) 34 (21.0%) 17 (10.5%) 37 (22.8%)

33 7 (21.2%) 6 (18.2%) 2 (6.1%) 11 (33.3%) 7 (21.2%) 32 7 (21.9%) 8 (25.0%) 6 (18.8%) 5 (15.5%) 6 (18.8%)

EDv ⫽ end-diastolic volume; EF ⫽ ejection fraction; ESv ⫽ end-systolic volume; MVCD ⫽ mitral valve coaptation depth; NS ⫽ not significant; SV ⫽ stroke volume.

calculated with the log-rank test. Cox analysis was used to evaluate the independent risk factors for reduced late events. In the Cox analysis model independent variables were expressed as hazard ratio with the 95% confidence limit; the related probability value was also reported. The SPSS software (SPSS, Inc, Chicago, IL) was used. Probability values less than 0.05 were considered significant.

Results Table 3 shows the surgical details of the patients included in our study. Eighty-two (82) patients (80.4%) underwent MV repair, 15 (18.3%) had a suture annuloplasty, and 67 (81.7%) had a posterior annuloplasty (from commissure to commissure), which was performed using either a pericardial strip in 62 (52 mm long in 29 patients and 40 mm long in 34) or a specially designed 40-mm-long ring (Sovering Miniband) in 5 patients. An edge-to-edge valvuloplasty was added in 5 patients because of possible anterior leaflet prolapse. A prosthesis (biologic in 11 [55%] and mechanical in the remaining 9 [45%]) was inserted into the mitral orifice of 20 patients (19.6%).

MR ⫽ mitral regurgitation;

NS NS NS NS NS NS NS NS NS NS MV ⫽ mitral valve;

Four patients (3.9%) died during the first 30 postoperative days: 1 died as a result of low output syndrome, and 3 were lost for noncardiac causes (rupture of abdominal aneurysm, tracheal bleeding, and multi-organ failure as a result of bleeding). Two of them had undergone MV repair (2 of 82, 2.4%) and 2 had undergone MV replacement (2 of 20, 10.0%; not significant. Logistic regression did not show any risk factors for early mortality. Table 4 shows the postoperative outcome of the patients in this study. After a mean of 24 ⫾ 33 months, 26 patients died, 23 of cardiac causes (heart failure in 13, sudden death in 7, and acute MI in 3) and 3 died of noncardiac causes (septicemia in 1, car accident in 1, acute respiratory failure in 1). Eighteen of them (21.9%) had undergone MV repair and 8 had undergone MV replacement (40.0%). The risk factors identified by Cox analysis were preoperative NYHA class, EF, echocardiographic end-diastolic and end-systolic volumes, and congestive heart failure (Table 5). Mean follow-up of the survivors was 39 ⫾ 35 months, with 14 patients surviving more than 5 years. Five-year survival was 73.5% ⫾ 4.4% (Fig 3), with no differences between repair or replacement.

CALAFIORE ET AL MITRAL VALVE SURGERY

1993

shows the data from the transthoracic echocardiogram performed in our institution in 46 patients after a mean of 24 ⫾ 21 months. Globally, volumes and EF did not change. In 40 patients who had MV repair (5 with a suture annuloplasty and 35 by means of a pericardial strip), 50% had no or mild MR (2 of 5 and 18 of 35, respectively). Patients with residual MR 2/4 or higher (n ⫽ 20) improved their NYHA class from 3.3 ⫾ 0.5 to 2.4 ⫾ 0.9 (p ⬍ 0.001), even if at a lower extent than the remaining 20 patients with no or 1/4 residual MR (from 3.3 ⫾ 0.6 to 1.9 ⫾ 0.5; p ⬍ 0.001). Patients who had replacement had larger preoperative and postoperative volumes and lower EF than patients who had MV repair.

Comment

Fig 2. Only a small triangle of the anterior leaflet is excised (A). The remainder of the anterior leaflet is fixed laterally using the prosthetic sutures (B).

Clinical results are shown in Table 6. Of the 72 survivors, 55 (76.3%) were in NYHA classes I and II; 79.1% of them had repair, and 60.0% had replacement (not significant). Possibility of being alive and in NYHA classes I and II was 56.9% ⫾ 4.9% (Fig 4), which was similar for both repair or replacement. Cox analysis showed that preoperative EF, echocardiographic end-diastolic and end-systolic volumes, and congestive heart failure were risk factors (Table 5). A postoperative transthoracic echocardiogram was obtained in 68 patients (86.1% of the survivors). Table 7

Mitral regurgitation that follows completed MI has an incidence of approximately 20% [3, 16], and it is higher in patients with previous inferior MI. Kumanohoso and associates [16] reported some degree of MR after a mean follow-up of more than 24 months, in 21.4% of 102 consecutive patients with previous MI (61 anteroseptal and 41 inferior). Mitral regurgitation incidence was higher after inferior (38%) as compared with anterior (10%) MI. On the other hand, anteroseptal MI causes extensive ischemia with minor involvement of the backpositioned MV apparatus, and inferior MI causes less extensive ischemia but with a direct involvement of important components of the MV apparatus, including the basal inferior LV wall adjacent to the posterior PM and MV annulus. In most cases, the mechanism of IMR is related to local LV remodeling, with PM displacement producing apical tethering or tenting of the leaflets (restricted systolic leaflet motion). When global LV dilation occurs, both PMs are displaced posteriorly, laterally, and apically. As a consequence, the tethering forces on both leaflets increase, reducing their movement. Patients with IMR have a worse natural history than patients without IMR. In patients with previous acute MI, Grigioni and colleagues [2] reported a 5-year survival of 61% ⫾ 6% in the absence of IMR, and of 38% ⫾ 5% in the

Table 3. Surgical Detailsa Variable

All (n ⫽ 102)

MV Repair (n ⫽ 82)

MV Replacement (n ⫽ 20)

CABG Anastamoses/patient Tricuspid valve repair Aortic valve replacement AoV repl. ⫹ AscAo repl AscAo replacement ASB pacing device AF ablation CPB time (min) Aortic XCI time (min)

93 (91.2%) 2.2 ⫾ 0.9 18 (17.6%) 3 (2.9%) 1 (1.0%) 1 (1.0%) 4 (3.9%) 1 (1.0%) 102 ⫾ 44 81 ⫾ 30

73 (89.0%) 2.2 ⫾ 1.1 15 (18.3%) 2 (2.4%) 1 (1.2%) 1 (1.2%) 4 (4.9%) 1 (1.2%) 102 ⫾ 44 82 ⫾ 32

20 (100%) 1.7 ⫾ 0.8 3 (15.0%) 1 (5.0%) 0 0 0 0 102 ⫾ 41 79 ⫾ 19

a

There were no significant differences in any variables studied.

AF ⫽ atrial fibrillation; AoV ⫽ aortic valve; ASB ⫽ atrial synchronous biventricular; AscAo ⫽ ascending aorta; CABG ⫽ coronary artery bypass grafting; CPB ⫽ cardiopulmonary bypass; MV ⫽ mitral valve; repl ⫽ replacement; XCl ⫽ cross-clamping.

CARDIOVASCULAR

Ann Thorac Surg 2004;77:1989 –97

1994

CALAFIORE ET AL MITRAL VALVE SURGERY

Ann Thorac Surg 2004;77:1989 –97

Table 4. Postoperative Data Variable CARDIOVASCULAR

Deaths AMI CVA LOS IABP Acute renal failure Acute respiratory failure Bleeding (mL/12 h) Transfused patients ICU stay (h) In-hospital stay (d)

All (n ⫽ 102)

MV Repair (n ⫽ 82)

MV Replacement (n ⫽ 20)

p Value

4 (3.9%) 5 (4.9%) 5 (4.9%) 9 (8.8%) 3 (3.9%) 19 (18.6%) 10 (9.8%) 847 ⫾ 784 74 (72.5%) 34.6 ⫾ 36.6 7.8 ⫾ 6.2

2 (2.4%) 4 (4.9%) 4 (4.9%) 6 (7.3%) 2 (2.4%) 14 (17.1%) 8 (9.8%) 890 ⫾ 850 59 (72.0%) 29.3 ⫾ 33.4 7.7 ⫾ 6.6

2 (10.0%) 1 (5.0%) 1 (5.0%) 3 (15.0%) 1 (5.0%) 5 (25.0%) 2 (10.0%) 700 ⫾ 476 15 (75.0%) 54.4 ⫾ 45.0 8.2 ⫾ 3.8

NS NS NS NS NS NS NS NS NS 0.028 NS

AMI ⫽ acute myocardial infarction; CVA ⫽ cerebrovascular accident; ⫽ low-output syndrome; MV ⫽ mitral valve; NS ⫽ not significant.

IABP ⫽ intraaortic balloon pump;

ICU ⫽ intensive care unit;

presence of IMR (p ⬍ 0.001). Survival was even worse (29% ⫾ 9%) if the actual regurgitant orifice was at least 20 mm2. Lamas and coworkers [3] found that after a mean of 3.5 years after MI, patients with IMR had a higher incidence of cardiovascular mortality (29% versus 12%, p ⬍ 0.01) and severe heart failure (24% versus 16%, p ⫽ 0.0153) compared with patients without IMR. In patients who had undergone percutaneous intervention, Ellis and associates [1] found that the 3-year actuarial survival in patients with no, mild, moderate, and severe IMR differed considerably (92.3%, 84.5%, 74.6%, and 68.6%; p ⬍ 0.001). This difference was even more notable in patients with EF of 0.40 or less (75.7%, 66.9%, 53.7%, and 46.5%; p ⬍ 0.001). In spite of these data, surgeons were never aggressive toward IMR. Some papers have questioned the benefit of MR correction in the natural history of surgical patients. In 1988 Arcidi and colleagues [4] reported 58 patients with 3/4 MR who had undergone coronary artery bypass grafting (CABG; however, 6 patients had LV aneurysmectomy or placation, 2 had septal myectomy, and 3 had a previous coronary operation) and compared them with an unmatched cohort of 20 patients who had undergone MV replacement and CABG. Five-year survival was 77% versus 31%, and the authors concluded that CABG alone was a good option in patients with coronary artery disease and 3/4 MR. However, mean EF was high (0.53),

and ischemia was the cause of MR in 74% of the patients in the first group and in only 30% of the patients in the control group. Multivariate analysis showed that congestive heart failure and low EF were independent predictors of lower survival, but congestive heart failure was present in only 10% of the patients in the first group and in 80% of the patients in the control group. Only 38% of the patients in the first group had LV akinesis. In 1999 Duarte and coworkers [17] compared a group of 58 patients with 3/4 MR with 58 case-matched patients without MR and the same EF. Fifteen-year survival after surgery was the same. However, again low EF (0.40 or less), congestive heart failure, and pulmonary wedge pressure more than 17 mm Hg were independent predictors of lower survival. As a consequence, both papers demonstrated that when EF is low and heart failure symptoms are present, IMR has to be corrected. The same concept was shown by Harris and colleagues [18], reporting on two groups of patients with the same degree of IMR (2/3) who had undergone either CABG alone (n ⫽ 142; EF, 0.387) or CABG and MV surgery (n ⫽ 34; EF, 0.380). While in-hospital mortality was 9% versus 21%, the 5-year survival was similar (52% versus 58%). Patients with preoperative heart failure had a 5-year survival of 10% with CABG alone and 57% with CABG and MV surgery. The authors also found that CABG alone did not avoid early progression of IMR to more severe degrees, whereas MV surgery stabilized the

LOS

Table 5. Cox Analysis Possibility of Being Alive and in NYHA Class I or II

Survival Variable Preoperative NYHA class EF (continuous) EDv/m2 (continuous) ESv/m2 (continuous) CHF CHF ⫽ chronic heart failure; ume; HR ⫽ hazard ratio;

HR 2.1 ⫺1.07 1.02 1.01 2.8

95% CL

p Value

HR

95% CL

p Value

(1.1– 4.3)

0.0490

...

...

...

(1.02–1.1) (1.01–1.03) (1.003–1.02) (1.1– 6.9)

0.0043 0.0189 0.0090 0.0273

⫺1.05 1.01 1.01 2.3

(1.01–1.08) (1.005–1.02) (1.004 –1.02) (1.1– 4.6)

0.0041 0.0031 0.0028 0.0238

CL ⫽ confidence limit; EDv ⫽ end-diastolic volume; NYHA ⫽ New York Heart Association.

EF ⫽ ejection fraction;

ESv ⫽ end-systolic vol-

CALAFIORE ET AL MITRAL VALVE SURGERY

1995

CARDIOVASCULAR

Ann Thorac Surg 2004;77:1989 –97

Fig 3. Survival according to mitral valve (MV) repair or replacement. (Dashed line ⫽ mitral valve repair; straight line ⫽ overall; straight line with black squares ⫽ mitral valve replacement.)

amount of residual MR. Also, Aklog and coworkers [13] found that CABG alone for moderate IMR leaves many patients with significant residual MR, and that this may not be the optimal therapy for most patients. On the contrary, in two groups of 89 patients with and without IMR and the same EF (0.42 versus 0.44) who had CABG alone, Ryden and associates [19] evidenced a similar 3-year survival. However, survival curves started to diverge after 3 years, and the difference, as stated also by the authors, began to be more evident with a longer follow-up. Recently, two important studies reported long-term results after MV surgery for IMR. Grossi and colleagues [7] reported a 30-day mortality rate of 13.4% in patients operated on in 1988 or after, with a 5-year survival of 71% for repair and 34% for replacement. Gillinov and associates [8] had a 30-day mortality of 13% and, in the lower-risk group, a 5-year survival of 58% after MV repair and of 36% after MV replacement; in the higherrisk group, survival after either repair or replacement was similarly poor. The authors concluded that even though most patients with IMR benefit from MV repair, in the most-complex, high-risk settings, survival after either repair or replacement is similar. Differently from our study, both series included patients who were operated on for PM rupture. These reports encompassed a

Fig 4. Possibility of being alive and in New York Heart Association classes I and II according to mitral valve (MV) repair or replacement. (Dashed line ⫽ mitral valve repair; straight line ⫽ overall; straight line with black squares ⫽ mitral valve replacement.)

long-lasting experience, with different surgeons using different techniques. Our experience has been developed in an epoch in which the mechanisms of IMR are better understood and the surgical techniques are more effective. Our results (30-day mortality of 3.9% and 5-year survival of 73.5% with 76.3% of the survivors in NYHA classes I and II) emphasize the benefits of an extensive correction of chronic IMR; as evidenced by the literature, MR impairs the outcome in patients after completed MI, and CABG alone is not able to control the postoperative natural history of untreated MR. Furthermore, as far as we know, to date no study shows that in comparable patients with and without IMR, surgery worsens the outcome in treated patients compared with similar untreated patients. In our opinion, 3/4 or 4/4 IMR has to be considered as nonischemic MR with the same degree of MR. We know that the natural history and survival of patients with 3/4 or 4/4 MR, even in the absence of symptoms and pharmacologically treated, is worse than the expected mortality rate in the general population [20]. Moreover, early surgery improves survival compared with a similar population with the same disease treated only pharmacologically, independent of symptoms [21]. Applying these

Table 6. Clinical Results MV Repair (n ⫽ 62) Variable NYHA class I II III IV Impaired Unchanged Improved Follow-up (mo) MV ⫽ mitral valve;

MV Replacement (n ⫽ 10)

Pre

Post

p Value

Pre

Post

p Value

3.2 ⫾ 0.5 0 3 (4.8%) 46 (74.2%) 13 (21.0%)

2.1 ⫾ 0.6 5 (8.1%) 44 (71.0%) 12 (19.4%) 1 (1.6%)

⬍0.001 NS ⬍0.001 ⬍0.001 0.002

3.5 ⫾ 0.5 0 0 7 (70%) 3 (30%)

2.5 ⫾ 0.7 0 6 (60%) 3 (30%) 1 (10%)

0.002

2 (3.2%) 12 (19.4%) 48 (77.4%) 38 ⫾ 35 NS ⫽ not significant;

NYHA ⫽ New York Heart Association;

0 4 (40%) 6 (60%) 46 ⫾ 36 Post ⫽ postoperative;

Pre ⫽ preoperative.

0.015 NS NS NS NS NS NS

1996

CALAFIORE ET AL MITRAL VALVE SURGERY

Ann Thorac Surg 2004;77:1989 –97

Table 7. Echocardiographic Results MV Repair (n ⫽ 40) Variable CARDIOVASCULAR

EF EDV (mL/m2) ESV (mL/m2) SV (mL/m2) MR (from 0 to 4) none 1⫹ 2⫹ 3⫹ 4⫹ Follow-up (mo)

MV Replacement (n ⫽ 6)

Pre

Post

p Value

0.37 ⫾ 0.12 110 ⫾ 39 69 ⫾ 35 40 ⫾ 15 3.2 ⫾ 0.8 ... ...

0.40 ⫾ 0.16 98 ⫾ 42 61 ⫾ 38 36 ⫾ 10 1.2 ⫾ 1.1 11 (27.5%) 9 (22.5%) 19 (47.5%) 1 (2.5%) ...

NS NS NS NS ⬍0.001 ... ... ... ... ...

7 (17.5%) 16 (40.0%) 17 (42.5%)

Pre 0.25 ⫾ 0.07 162 ⫾ 58 120 ⫾ 49 43 ⫾ 11 3.2 ⫾ 0.8 ... ... ... ... ...

24 ⫾ 17

0.31 ⫾ 0.09 158 ⫾ 67 113 ⫾ 59 45 ⫾ 13 ... ... ... ... ... ...

p Value NS NS NS NS ... ... ... ... ... ...

23 ⫾ 30

EDv ⫽ end-diastolic volume; EF ⫽ ejection fraction; ESv ⫽ end-systolic volume; MR ⫽ mitral regurgitation; not significant; Post ⫽ postoperative; Pre ⫽ preoperative; SV ⫽ stroke volume.

concepts to all patients with MR, independent of its origin, MV surgery has to be performed in every patient with either moderate-to-severe or severe IMR, especially when EF is low, as in most IMR patients. The experience deriving from MV surgery in patients with dilated cardiomyopathy (mostly ischemic) demonstrated that a 30day surgical risk is low and midterm survival is satisfactory [15, 22–24]. The question is whether to perform surgery when MR is moderate (2/4). In our opinion, in the presence of low EF and dilated LV, moderate MR has to be corrected. Low EF is by itself a risk factor for long-term survival after CABG, and, in this subset of patients, when MR is present, survival can be still worse. Tolis and associates [5] reported a series of patients with EF of 0.30 or less and mean MR of 1.7 who had undergone CABG alone. Five-year survival was 50%, which was worse than the 5-year survival of 71% reported by Grossi and colleagues [7]. Although it is difficult to compare the two different series, the figures suggest a different outcome. Time of surgery is another important key point. Our experience shows that patients with preoperative chronic heart failure, enlarged hearts, and lower EF have lower survival and fewer possibilities to be alive and in NYHA classes I and II. Even in the absence of available data, it is likely that a more anticipated surgical indication will improve midterm results. Mitral valve repair is often possible, and overreductive annuloplasty is nearly always able to correct MR, even in the presence of double regurgitant jets. We prefer to use a 40-mm ring, which is either commercially available or made of autologous pericardium, because it totally eliminates the posterior leaflet from the mechanism of MV closure, thus acting only as a doorstop. However, the possibility of allowing both leaflets to coapt depends on the ability of the anterior leaflet to move toward the annulus and to reach the posterior one. If this movement is insufficient, the mitral leaflets never coapt no matter how much the posterior annulus is reduced. For this reason, for each patient, we evaluate the depth of the anterior leaflet during systole. According to our experi-

Post

MV ⫽ mitral valve;

NS ⫽

ence [15], this value is crucial for deciding whether to repair (if 10 mm or less) or to replace (if more than 10 mm) the MV. It is evident that the more the PMs are displaced, the higher the possibility that the anterior leaflet could be limited in its movement. Therefore, MV replacement with intact subvalvular apparatus is reserved for patients with the most-displaced PMs. These patients have generally the largest hearts and the lowest EFs. Nevertheless, the 5-year results appear to be similar to the results in patients undergoing MV repair, although it is likely that curves can diverge significantly with a longer follow-up and a greater number of patients. This finding focuses on preventing MR recurrence (or reducing it as much as possible), which is the main target of MV surgery for IMR. In conclusion, there is evidence that untreated IMR may worsen the outcome in patients who need myocardial revascularization. Moderate-to-severe and severe IMR need to be corrected, inasmuch as today we have appropriate surgical means for this purpose. In cases of moderate IMR, we think that surgery is indicated in the presence of low EF or large LV volumes. In other cases and in the absence of experimental evidence, it is very likely that correction of IMR will not change the natural history of patients. Mitral valve repair has to be performed looking at the possibility that the anterior leaflet may reach the annulus to meet the frozen posterior leaflet. When CD is more than 10 mm, MV replacement accompanied by prosthesis insertion inside the MV seems to be the best option. Early and midterm results seem to be satisfactory, even when most of these patients are in preoperative congestive heart failure.

References 1. Ellis SG, Whitlow PL, Raymond RE, Schneider JP. Impact of mitral regurgitation on long-term survival after percutaneous coronary intervention. Am J Cardiol 2002;89:315–8. 2. Grigioni F, Enriquez-Sarano M, Zehr KJ, Bailey KR, Tajik AJ. Ischemic mitral regurgitation long-term outcome and prognostic implications with quantitative Doppler assessment. Circulation 2001;103:1759 –64.

3. Lamas GA, Mitchell GF, Flaker GC, et al. Clinical significance of mitral regurgitation after acute myocardial infarction. Circulation 1997;96:827–33. 4. Arcidi JM Jr, Hebeler RF, Craver JM, et al. Treatment of moderate mitral regurgitation and coronary disease by coronary bypass alone. J Thorac Cardiovasc Surg 1988;95:951–9. 5. Tolis JA, Korkolis DP, Kopf GS, Elefteriades JA. Revascularization alone (without mitral valve repair) suffices in patients with advanced ischemic cardiomyopathy, and mild-tomoderate mitral regurgitation. Ann Thorac Surg 2002;74: 1476 –81. 6. Cohn LH, Rizzo RJ, Adams DH, et al. The effect of pathophysiology on the surgical treatment of ischemic mitral regurgitation: operative and late risks of repair versus replacement. Eur J Cardiothorac Surg 1995;9:568 –74. 7. Grossi EA, Goldberg JD, LaPietra A, et al. Ischemic mitral valve reconstruction and replacement: comparison of longterm survival and complications. J Thorac Cardiovasc Surg 2001;122:1107–24. 8. Gillinov AM, Wierup PN, Blackstone EH, et al. Is repair preferable to replacement for ischemic mitral regurgitation? J Thorac Cardiovasc Surg 2001;122:1125–41. 9. Kwan J, Shiota T, Agler DA, et al. Geometric differences of the mitral apparatus between ischemic and dilated cardiomyopathy with significant mitral regurgitation real-time three-dimensional echocardiography study. Circulation 2003;107:1135–40. 10. American Society of Echocardiography Committee on Standards. Recommendations for quantitation of the left ventricle by two dimensional echocardiography. J Am Soc Echocardiogr 1989;2:358 –67. 11. Helmecke F, Nanda NC, Hsiung MC, et al. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation 1987;75:175–83. 12. Calafiore AM, Gallina S, Contini M, et al. Surgical treatment of dilated cardiomyopathy with conventional technique. Eur J Cardiothorac Surg 1999;16(Suppl):S73–8. 13. Aklog L, Filsoufi F, Flores KQ, et al. Does coronary artery bypass grafting alone correct moderate ischemic mitral regurgitation? Circulation 2001;104(Suppl 1):I-68 –75. 14. Calafiore AM, Di Mauro M, Gallina S, Canosa C, Iaco` AL. Optimal length of pericardial strip for posterior mitral overreductive annuloplasty. Ann Thorac Surg 2003;75: 1982–4. 15. Calafiore AM, Gallina S, Di Mauro M, et al. Mitral valve procedure in dilated cardiomyopathy: repair or replacement? Ann Thorac Surg 2001;71:1146 –53. 16. Kumanohoso T, Otsuji Y, Yoshifuku S. Mechanism of higher incidence of ischemic mitral regurgitation in patients with inferior myocardial infarction: quantitative analysis of left ventricular and mitral valve geometry in 103 patients with prior myocardial infarction. J Thorac Cardiovasc Surg 2003; 125:135–43. 17. Duarte IG, Shen Y, MacDonald MJ, Jones EL, Craver JM, Guyton RA. Treatment of moderate mitral regurgitation and coronary disease by coronary bypass alone: late results. Ann Thorac Surg 1999;68:426 –30. 18. Harris KM, Sundt TM, Aeppli D, Sharma R, Barzilai B. Can late survival of patients with moderate ischemic mitral regurgitation be impacted by intervention on the valve? Ann Thorac Surg 2002;74:1468 –75. 19. Ryden T, Bech-Hanssen O, Brandrup-Wognsen G, Nilsson F, Svensson S, Jeppsson A. The importance of grade 2 ischemic mitral regurgitation in coronary artery bypass grafting. Eur J Cardiothorac Surg 2001;20:276 –81. 20. Ling LH, Enriquez-Sarano M, Seward JB, et al. Early surgery in patients with mitral regurgitation due to flail leaflets; a long-term outcome study. Circulation 1997;96:1819 –25. 21. Ling LH, Enriquez-Sarano M, Seward JB, et al. Clinical

CALAFIORE ET AL MITRAL VALVE SURGERY

1997

outcome of mitral regurgitation due to flail leaflet. N Engl J Med 1996;335:1417–23. 22. Bolling SF, Pagani FD, Deeb GM, Bach DS. Intermediateterm outcome of mitral reconstruction in cardiomyopathy. J Thorac Cardiovasc Surg 1998;115:381–8. 23. Radovanovic N, Mihajlovic B, Selestiansky J, et al. Reductive annuloplasty of double orifices in patients with primary dilated cardiomyopathy. Ann Thorac Surg 2002;73:751–5. 24. Bishay ES, McCarthy PM, Cosgrove PM, et al. Mitral valve surgery in patients with severe left ventricular dysfunction. Eur J Cardiothorac Surg 2000;17:213–21.

Appendix List and Definitions of Variables Preoperative Clinical Variables Age continuous (years) Sex dichotomous Left main disease stenosis 50% or more Number of vessels diseased from 1 to 3 Diabetes medical treatment for hyperglycemia at rest History of hypertension need of medical treatment (Ca2⫹ blockers, ␤-blockers, angiotensin-converting enzyme inhibitors) Chronic obstructive pulmonary disease forced expiratory volume in 1 second less than 75% of predicted value, air partial pressure of oxygen lower than 60 mm Hg or chronic medical treatment. Long-term use of bronchodilators or steroids for lung disease. Unstable angina presence of angina at rest, stable angina with worsening pattern, or de novo angina. Rest angina requiring intravenous nitrates until arrival in the anesthetic room. New York Heart Association class from I to IV Chronic heart failure heart failure in the history or at the present admission, without angina Chronic renal failure creatinine value higher than 2.0 mg/dL. Redo previous CABG operation Preoperative intraaortic balloon pump use of intraaortic balloon pump for cardiogenic shock or to stabilize an unstable angina Infarcted area anterior or lateral or inferior or septal or posterior

Preoperative Echocardiographic Variables Mitral regurgitation grade from 1 to 4 Ejection fraction continuous End-diastolic volume continuous (mL/m2) End-systolic volume continuous (mL/m2) Mitral valve coaptation depth continuous (mm) Mitral annulus size continuous (mm)

Operative Variables Cardiopulmonary bypass time continuous (minutes) Mitral valve replacement or repair dichotomous Tricuspid repair dichotomous

CARDIOVASCULAR

Ann Thorac Surg 2004;77:1989 –97