From January 1994 to December 2005, 40 consecutive patients with SMI (grade III-IV) and impaired left ventricular pump function (EF < 35% in echocardiography and angiography) underwent mitral valve surgery at our institution. Patients with previous coronary artery bypass grafting (CABG) were included in the study; patients with an additional left ventricular remodeling procedure were excluded.

The study population included 20 male and 20 female patients with a mean age of 64 ± 9 years. 26 patients suffered from dilative cardiomyopathy, 14 from ischemic cardiomyopathy. Mean preoperative ejection fraction was 28 ± 5%. 24 had chronic atrial fibrillation. 27 patients were in New York Heart Association (NYHA) class III and 13 patients in NYHA class IV despite optimized heart failure therapy which typically included digoxin, angiotensin – converting enzyme inhibitors, diuretics, beta – blockers and spironolactone. 16 patients had been initially referred to our institution for evaluation of heart transplantation. Patient characteristics and comorbidities are shown in Table 1.

All patients underwent transthoracic or transoesophageal echocardiography on hospital admission, transoesophageal echocardiography after induction of anesthesia prior to sternotomy and intraoperative to control the results after mitral valve surgery. Postoperative follow-up echocardiography was obtained prior to discharge and on follow-up. If no follow-up was available in our institution (two patients), the investigation was performed at a city hospital.

The perioperative measurement of left ventricular chamber size at end-diastole and end-systole as well as the assessment of mitral regurgitation were performed by transthoracic two-dimensional echocardiographic images in the parasternal long-axis view (including M-mode) and by apical four-chamber view. The intraoperative analysis was performed by employing short-axis and long-axis views using a transoesophageal approach. Left ventricular volumes and ejection fraction were calculated by modification of Simpson's rule method with two apical views. Stroke volume was calculated as the difference between the diastolic and systolic volumes, and ejection fraction was calculated as the ratio of stroke volume to end-diastolic volume. Colour Doppler flow mapping of regurgitant jets with visualisation of the vena contracta and proximal isovelocity surface area was used for quantification of valvular regurgitation. The severity was graded as mild (I), moderate (II), moderate-severe (III), and severe (IV) 319.

29 patients (72.5%) had grade 3, and 11 (27.5%), grade 4 mitral regurgitation. All patients had normal or moderate impaired right ventricular function without presence of liver failure, although in 10 patients the mean pulmonary pressure was higher than 40 mm Hg.

Preoperative echocardiography and hemodynamic data are listed in Table 2.

Surgical treatment in patients with SMI and considerably impaired LV function had been identified as an indicator of poor prognosis. This is due to the primary ventricular problem that causes MV dysfunction and, despite of the SMI correction, the disease further exists 21. MV surgery is associated with a high mortality which in the series of other investigators varies between 2.3% and 19.4% 31622. The patients in our study, therefore, represented a high predicted mortality among patients with dilated or ischemic cardiomyopathy. The surgical mortality in patients with ischemic cardiomyopathy with an age > 60 years have been reported from 10 to 48% 23. A high operative mortality of 21% has been reported among a group of 28 patients undergoing mitral valve replacement and additional CABG 24. Despite high operative risks the 30-days mortality in our group of patients was moderate with 7.5% (n = 3). All patients with early death had ischemic cardiomyopathy with previous CABG. In addition, in two cases it was a combination of MV replacement with CABG and in one a combination of MV repair, CABG and TV repair. Redo's and prolonged cross-clamping times are the mainly responsible for the high mortality in patients with -ischemic cardiomyopathy. There were no deaths in patients with dilated cardiomyopathy and our results confirm the reports of other authors about poor early outcomes in patients with ischemic cardiomyopathy.

MV surgery in patients with LV dysfunction is associated with higher postoperative complication rates. Two thirds of our patients had preoperative atrial fibrillation and/or compensated renal insufficiency and ten of them were older than 70 years. Postoperative morbidity rate exceeded 40% and was expected in this cohort of patients with serious comorbidities. Early complications such as redo for bleeding, acute respiratory failure, acute renal failure with dialysis and ventricular tachycardia developed in 11 (27.5%) of our patients. This is the reason for longer ventilation, ICU- and hospital stay and that coincides with reports of other investigators 325.

All patients showed improvement in exercise tolerance at the follow-up, and the NYHA class improved significantly. There were no differences between patients with ischemic or dilated cardiomyopathy. Quantitative echocardiographic analyses showed a markedly increase of ejection fraction, a significant reduction of pulmonary hypertension and of the left ventricular end-diastolic diameter at follow-up. Furthermore six from ten patients (60%) with severe pulmonary hypertension were still alive (mean survival 58 ± 29 months) at time of follow-up. These results do not confirm the opinion of other investigator about poor outcomes in this cohort of patients 26. The mitral regurgitation improved significantly from severe to moderate at follow-up in the repair/annuloplasty group. The mid-term survival in this study group was none significantly shorter compared to heart transplantation.

Following the guidelines for the management of patients with valvular heart disease asymptomatic patients with an ejection fraction below 60% and/or end-systolic dimension over 40 mm and symptomatic patients with EF above 30% and LVESD below 55 mm are candidates for mitral surgery 27. Still controversy exists regarding the operation (MVR versus HTX) in patients with severe reduced ejection fraction.

Our results could show that MV surgery in patients with considerably impaired LV function has a similar mid-term outcome compared to transplantation. We therefore conclude that in MV surgery is an acceptable alternative procedure which improved quality of life in the present time with increasing organ shortage.

Of importance is the preservation of the mitral valve apparatus. Previous clinical studies have compared the results of MV reconstruction against those following MV replacement and have concluded that preservation of the annular-chordal-papillary muscle continuity results in maintenance of LV function and geometry, leading to better patient outcome 282930. However, we could not observe a difference in outcome between MV repair and replacement. One reason could be the preservation of the mitral valve apparatus despite MV replacement, which was observed in our studies, too 31. But we think that chordal sparing mitral valve replacement is not a better way to treat SMI because of the need for anticoagulation for mechanical prosthesis in mitral position and inevitable degeneration of bioprosthesis.

Through the restoration of the mitral competency and ventricular geometry, MV surgery offers a new treatment strategy for the treatment of end-stage heart failure.

This study has several limitations. First, the population represents a relatively small number of patients. However, significant differences were detected. Second, the accuracy of the volume quantification is dependent to operator experience, and calculation of flows has inherent errors because of limitations in measurement accuracy. Third, the assessment for the functional status at the follow-up was subjective. We tried to objective our results using standardized exercise tests. Forth, the comparison of the mitral valve with the transplant patient group is not very accurate, because of the more depressed LV function and co-morbidities in the transplant group. However, we tried to overcome this obstacle by age, gender and time period matching.

High risk mitral valve surgery for secondary mitral regurgitation in patient with ischemic and dilated cardiomyopathies and considerably impaired LV function corrects effectively mitral regurgitation and represents an alternative procedure in a high-risk population with an acceptable perioperative mortality rate. Decrease in mitral regurgitation after surgical correction contributes to restoration of left ventricular geometry and may be an alternative to heart transplantation in selected patients.