|Home | About | Journals | Submit | Contact Us | Français|
Ischemic mitral regurgitation, a complication of myocardial infarction, is associated with a poor prognosis and can result in postinfarction congestive heart failure. The preferred treatment of its chronic form is a matter of debate. Herein, we report the early and midterm results in 44 patients with chronic ischemic mitral regurgitation in whom concomitant mitral ring annuloplasty and coronary revascularization were performed at our hospital.
We reviewed their medical records. The patients had grades 3/4 and 4/4 chronic ischemic mitral regurgitation, or grade 2/4 regurgitation with left ventricular dilation and low left ventricular ejection fraction. All received circular, flexible annuloplasty rings.
Four patients died during the early postoperative period due to low cardiac output (9.1%). At the last follow-up echocardiographic examinations, performed a mean 13.14 ± 4.66 months after the surgical procedures (range, 6–22 mo), the 40 surviving patients were found to have significantly reduced left ventricular end-diastolic (P = 0.029) and end-systolic (P < 0.05) diameters and improved New York Heart Association functional class (P = 0).
Despite a risk of residual regurgitation, mitral ring annuloplasty appears to be a good treatment alternative in selected patients who have chronic ischemic mitral regurgitation. We discuss the procedure's rate of hospital mortality, and its potentially positive impact on survival.
Ischemic mitral regurgitation (IMR) is a mechanical complication of myocardial infarction (MI) and a predictor of poor outcome. It is most frequently seen after inferior MI (in 38% of cases) and anteroseptal MI (in 10%).1 Ischemic mitral regurgitation is clinically divided into 2 forms: acute and chronic. The acute form of IMR develops after an acute post-MI rupture of papillary muscle. Conversely, a widely accepted clinical definition of the chronic form has not yet been agreed upon. Borger and colleagues proposed a definition of the condition in their review.2 Several mechanisms are responsible for the development of chronic IMR, including annular dilation; global left ventricular (LV) dilation together with a coaptation defect caused by the traction of the mitral valve leaflets due to the apical, posterior, and lateral displacement of both papillary muscles; and a local malfunction in the LV wall that adjoins a papillary muscle.3 Chronic IMR is found in 10% to 20% of patients with coronary artery disease,4 and it is a major cause of congestive heart failure after MI.5 Despite such prevalence, surgical intervention to correct IMR remains open to debate, because early hospital mortality rates are high and long-term survival rates are not satisfactory.
Substantial improvements have been attained by use of valvular intervention in patients who exhibit heart failure and low LV ejection fraction (LVEF).6–8 Our current approach is valvular intervention in patients with IMR grades 3/4 and 4/4, and in patients who have grade 2/4 IMR with LV dilation and a low LVEF. Previously, we performed mitral valve replacement (MVR), but we have come to prefer mitral ring annuloplasty (MRA). Here, we present and discuss the early and midterm results in 44 patients who underwent concomitant coronary artery bypass grafting (CABG) and MRA at our institution.
Forty-four patients who underwent MRA in our hospital from February 2001 through June 2006 for chronic IMR were included in this retrospective analysis. Mitral ring annuloplasty had been performed, concomitantly with CABG, in 25 patients who had grade 3/4 IMR, in 7 patients who had grade 4/4 IMR, and in 12 patients who had grade 2/4 IMR with LV dilation and a low LVEF. Other patients who had acute IMR due to papillary muscle rupture or with organic mitral regurgitation in addition to their coronary artery disease were excluded from the analysis.
Cardiopulmonary bypass was performed by instituting aortic arterial–bicaval venous cannulation and pulsatile flow with the patient cooled to 30°C (measured rectally). Myocardial protection was achieved via antegrade blood cardioplegic arrest, followed by continuous administration of retrograde blood cardioplegic solution for the duration of cross-clamping. After cross-clamp application, the initial distal saphenous anastomoses were performed. This step was followed consecutively by ring implantation, distal anastomosis of the left internal mammary artery, and the proximal anastomoses. Cross-clamps were removed after de-airing of the heart.
Mitral ring annuloplasty was performed in all patients in whom the mitral leaflets could be coapted. A circular Sovering™ flexible ring (Sorin Biomedica Cardio S.p.A.; Saluggia, Italy) was used in all instances. The appropriate ring size was determined from measurement of the anterior leaflet. Ring sizes 28 to 32 were used (no. 28 in 6 patients, no. 30 in 12 patients, and no. 32 in 26 patients). In each patient, 9 to 12 TiCron™ 2-0 sutures (Covidien Syneture; Mansfield, Mass) were placed. Additional procedures were performed in 2 patients each: chordae tendineae plication, chordae transfer, and attachment of the P2 scallop to the P3 scallop. After cardiopulmonary bypass was stopped, intraoperative transesophageal echocardiography was performed in order to rule out substantial valvular insufficiency. In our study, no patient required mitral reintervention.
Data were collected from the medical records of patients. During the study period, our routine institutional follow-up schedule consisted of echocardiographic evaluations at the 1st, 6th, and 12th postoperative months, and annually thereafter. We analyzed the results of each patient's last evaluable echocardiographic examination. The mean time between operation and examination was 13.14 ± 4.66 months (range, 6–22 mo). Changes in LVEF, LV end-diastolic diameter, LV end-systolic diameter, and degree of mitral regurgitation were analyzed. In addition, New York Heart Association (NYHA) functional capacity was analyzed.
The Statistical Package for Social Sciences version 15.0 was used for statistical analysis (SPSS Inc.; Chicago, Ill). Besides descriptive statistics (mean ± SD), comparisons between postoperative and follow-up data were made by using the Student t test for paired samples when the distribution was normal. For data without normal distribution, changes from baseline were analyzed by use of the Wilcoxon signed-rank test. Categorical data were compared by using the χ2 and Fisher exact tests. A P value <0.05 was considered statistically significant.
Of the 44 patients (mean age, 64.2 ± 7.6 yr), 20 were women (45.5%). Table I shows the preoperative characteristics and data of the patients. Most patients had 3-vessel disease, and approximately half had experienced an inferior MI. Half of the interventions were performed from 22 to 90 days after a patient's MI. Twenty-five of the patients had grade 3/4 mitral regurgitation (56.8%). The mean degree of mitral regurgitation was 2.8/4.
The left internal mammary artery was used for grafting in 36 patients (86.4%). The mean number of grafts was 3.2 ± 0.7. The mean durations of cross-clamping and total perfusion were 104.6 ± 46.5 min and 126.1 ± 54.5 min, respectively. The patients remained in the intensive care unit for a mean duration of 3.4 ± 1.9 days, and they were hospitalized for a mean duration of 8.2 ± 3.6 days.
Four patients died during the early postoperative period, all due to low cardiac output (hospital mortality rate, 9.1%). No relation was found between death and age, body mass index, LVEF, or the duration of cross-clamping or total perfusion.
Analysis of the follow-up echocardiographic examinations in the 40 surviving patients revealed significant reductions in LV end-diastolic and end-systolic diameters, in comparison with preoperative values (P = 0.029 and P < 0.05, respectively). In addition, statistically significant improvement occurred in the functional capacities of patients in accordance with NYHA classification. Although a tendency toward improved LVEF values was observed, the change did not reach statistical significance. The postoperative hemodynamic and functional changes are summarized in Table II. Upon the last follow-up, mean residual mitral regurgitation was 1.1/4, compared with the preoperative mean value of 2.8/4 (P < 0). Table III shows the distribution of patients by degree of mitral regurgitation preoperatively and at the last follow-up examination.
Survival Rates. Ischemic mitral regurgitation worsens the prognosis and substantially increases operative death in patients who have coronary artery disease. In the study by Ellis and associates9 among such patients who received percutaneous intervention, the 3-year survival rate in the absence of IMR was 92.3%. On the other hand, mild, moderate, and severe IMR was associated with survival rates of 84.5%, 74.6%, and 68.6%, respectively (all P < 0.001). Rates were even worse for patients with a LVEF ≤0.40: absence of IMR, 75.7% survival; mild, 66.9%; moderate, 53.7%; and severe, 46.5% (all P < 0.001). Mallidi and coworkers10 found a higher prevalence of heart failure and shorter cardiac-event-free survival in CABG-only patients who had mild-to-moderate mitral regurgitation, in comparison with patients who had no regurgitation.
General Indications for Surgery. Although IMR is of interest to surgeons because of its unfavorable outcomes, a consensus has not been reached regarding its surgical indications. Isolated CABG performed early in the post-MI period improves survival and reduces the LV remodeling that results in mitral regurgitation.11 Therefore, intervention to the mitral valve is not suggested during the acute period if there is no papillary muscle rupture. According to previous studies, mitral valve surgical intervention in patients with grades 1/4 and 2/4 chronic IMR is of debatable benefit; in contrast, the chief issue in the presence of grades 3/4 and 4/4 chronic IMR is to choose which surgical procedure will be performed.
Because the long-term outcomes of concomitant CABG and mitral intervention due to chronic IMR have led to no surgical consensus, early and midterm results have influenced the course of action in these patients.
In-Hospital Death. The hospital mortality rate of CABG-only patients who have no mitral regurgitation ranges from 0 to 6.9%,6,10,12–14 and the corresponding rate for CABG-only patients with moderate mitral regurgitation ranges from 1.8% to 12%.6,7,10,12–16 These rates seem similar, but when intervention to the mitral valve is performed in the latter group of patients, the mortality rate increases (range, 10%–21%) (Table IV). Regarding the long-term results in such patients, several investigators have proposed that IMR may progress after CABG-only procedures and worsen the survival rates. For example, Mallidi and associates10 reported a regression to grade 3/4 or 4/4 mitral regurgitation in 30% of CABG-only patients who had mild-to-moderate IMR. Aklog and co-authors17 reported the persistence of moderateor severe regurgitation in 77% of their CABG-only patients. In contrast, when Wong and colleagues16 reported the long-term results in their patients who had moderate mitral regurgitation, there was no difference between the groups who underwent CABG only or CABG plus mitral intervention.
According to the Society of Thoracic Surgeons' database, concomitant mitral valve surgery doubles the perioperative risks of CABG.10 Therefore, isolated CABG may be preferable in patients who have mild-to-moderate mitral regurgitation plus multiple comorbidities or a life expectancy of less than 5 years. In contrast, concomitant mitral valve surgery may be the treatment of choice in low-risk patients, considering the acceptable 5% mortality rate in this group.2
Indications for Surgery in Grade 2/4 Mitral Regurgitation. Calafiore and co-investigators8 performed surgery on the mitral valve in the presence of chronic IMR of grade 2/4 or higher when there was also LV dilation and low LVEF, because those symptoms are poor short- and long-term prognostic signs. In such patients, the mitral intervention improved long-term survival without any significant increase in early hospital death (30-day mortality rate, 3.9%; 5-year survival rate, 73.5%). Harris and colleagues7 determined a 5-year survival rate of only 10% for IMR patients who underwent CABG-only operations—in contrast with a 57% survival rate for patients who underwent CABG plus mitral valve surgery. In that study, all patients had congestive heart failure preoperatively. During the follow-up period of our study, patients who underwent both procedures experienced significant reductions in LV end-diastolic and end-systolic diameters, and improvement in functional capacity; there was a trend toward improvement in LVEF, although not to statistical significance. These findings suggest benefits for survival.
With regard to grade 2/4 IMR patients, we do not intervene except when there is also LV dilation and low LVEF, because long-term results of mitral intervention are questionable in patients with grade 2/4 IMR but no LV dysfunction, and operative death is increased. In our study, 1 of 12 patients who had grade 2/4 IMR died (8.3%). Age, congestive heart failure, severe LV dysfunction, preoperative intra-aortic balloon pump use, a high number of diseased coronary vessels, and the presence of comorbid conditions constitute risk factors for hospital death in IMR cases.18 Accordingly, when patients are selected for IMR correction on the basis of having fewer risk factors, a mortality rate of 8.3% seems to justify mitral valve intervention in patients with grade 2/4 mitral regurgitation, LV dilation, and low LVEF.
Mitral Ring Annuloplasty versus Valve Replacement. Currently, the most common interventions for chronic IMR are MRA and MVR. The aim of MRA is to decrease the size of the mitral annulus in order to enable the coaptation of leaflets, and its short-term results include encouragingly low perioperative mortality rates.19 Grossi and associates20 reported 30-day mortality rates of 10% in patients undergoing MRA and 20% in MVR, and Gillinov and co-authors21 reported corresponding figures of 6% and 19%. If the integrity of the valve is not preserved during MVR, severe postoperative LV dysfunction may develop, which entails mortality rates of up to 40%. Conversely, when chordal protection techniques are used during MVR, LV dysfunction is not much more common than in MRA operations22; however, protection of both leaflets may not always be possible. There are several other disadvantages of MVR, including valvular thrombosis, warfarin overdose, and extended duration of cardioplegic arrest. Although the superiority of MRA over MVR with resection of the native valve is well recognized, recurrent and residual regurgitation are not seen after MVR, and—despite the high reported mortality rates—LV functions are preserved after MVR when the complete valvular apparatus has been preserved. Because the early and long-term results of MRA are satisfactory, we prefer performing MRA in IMR patients.
Better 5-year survival rates have been achieved with use of MRA than with MVR. In the study by Grossi and associates,20 the survival rate was 71% in MRA patients versus 34% in MVR patients; the corresponding figures were 58% and 36% for low-risk patients in the Gillinov study.21 However, in more than 30% of MRA patients, persistent or recurrent regurgitation occurs as LV remodeling continues. Although recurrence tends to happen during the early postoperative period, stability usually results after 6 months. In contrast, the risk of postoperative recurrence is eliminated with MVR, yet the procedure is associated with decreased survival.23
Annuloplasty Rings. Several types of rings are used for annuloplasty. They may be rigid or flexible and circular or partial (covering the posterior annulus only). McGee and coworkers23 compared results from the use of flexible, rigid, and bovine pericardial rings. After 6 months, the development risk of grade 3/4 or 4/4 mitral regurgitation was 25% for flexible and rigid rings, and 66% for bovine pericardial rings; however, there was no association between the type of annuloplasty and survival. Although there is no consensus on the type of ring to use in annuloplasty, we use a flexible ring, because that style is most readily available.
Future Studies. Very large series involving the surgical treatment of IMR have not yet been reported. Studies with large numbers of patients and long durations of follow-up would help to resolve the debate regarding optimal treatment. Furthermore, results obtained from series of the subgroups of IMR might help to standardize IMR surgery. Finally, including a control group that comprises MVR patients in whom the complete valvular apparatus is preserved would render more meaningful conclusions. Although a limitation of our study is its small sample size, we believe that our results contribute to the currently available evidence.
In summary, MRA in chronic IMR is associated with varying degrees of recurrent mitral regurgitation, yet we consider MRA to be a preferable treatment in selected patients: MRA is associated with a low mortality rate, a substantial improvement in midterm functional capacity, and potentially improved survival.
Address for reprints: Hayrettin Tekumit, MD, Ibrahim Aga Mah. Esatpasa Sok., Girgin Sitesi A4 Blok Daire:9, Acibadem, Kadikoy, 34718 Istanbul, Turkey