doi: 10.1161/CIRCULATIONAHA.104.525188
(Circulation. 2005;112:I-402 – I-408.)
© 2005 American Heart Association, Inc.
Surgery for Valvular Heart Disease |
Mitral Valve Repair for Functional Mitral Regurgitation in End-Stage Dilated Cardiomyopathy
Role of the "Edge-to-Edge" Technique
From the Cardiac Surgery Department, San Raffaele University Hospital, Milan, Italy
Correspondence to Dr Michele De Bonis, Cardiac Surgery Department, San Raffaele University Hospital, Via Olgettina 60, 20132 Milan, Italy. E-mail michele.debonis{at}hsr.it
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background— The aim of this study was to assess the results of mitral valve (MV) repair in functional mitral regurgitation because of end-stage dilated cardiomyopathy (DCM).
Methods and Results— Seventy-seven patients with end-stage idiopathic (26 patients) or ischemic (51 patients) DCM underwent MV repair for functional mitral regurgitation (3 to 4+/4+). Fifty-eight patients (75.3%) were in New York Heart Association class III, and 19 (24.6%) were in IV. In 23 patients (29.8%) with a coaptation depth <1 cm, an isolated undersized annuloplasty was used. In the remaining 54 (70.1%), with a coaptation depth 
1 cm, the "edge-to-edge" technique was associated with the annuloplasty. In most of the cases (88.3%), a complete rigid/semirigid ring was used. Concomitant coronary artery bypass graft was performed in 39 patients (50.6%). Hospital mortality was 3.8% (3 of 77). Actuarial survival was 90.7±3.64%, and freedom from cardiac events was 81.8±7.96% at 2.7 years. At a mean follow-up of 18.4±9.8 months (range, 1 month to 5 years) New York Heart Association class improved from 3.4±0.4 to 1.4±0.6 (P<0.0001). Mitral repair failure (recurrence of MR 3+/4+) was documented in 7 patients (9%): 2 in the edge-to-edge (2 of 54, 3.7%) and 5 in the isolated annuloplasty group (5 of 23, 21.7%) (P=0.03). Freedom from repair failure at 1.5 years was 95.0±3.4% and 77±12.1%, respectively (P=0.04). The absence of the edge-to-edge was the only predictor of repair failure (P=0.03). When residual MR was absent or mild, a reverse left ventricular remodeling was clearly documented.
Conclusions— In patients with end-stage DCM, MV repair is feasible with low hospital mortality and important symptomatic improvement. The association of the edge-to-edge technique to the undersized annuloplasty can significantly improve the durability of the repair.
Key Words: mitral regurgitation • edge-to-edge technique • dilated cardiomyopathy
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
In ischemic or idiopathic dilated cardiomyopathy (DCM), left ventricular (LV) dysfunction is frequently complicated by functional mitral regurgitation (FMR). Undersized ring annuloplasty has been used for the treatment of FMR, although mitral insufficiency can recur in a significant number of cases.1–3 The durability of such a procedure seems to be particularly unreliable in patients with mild-to-moderate annular dilatation, severe tethering, and complex jets.1–2 In these cases, chordal-sparing valve replacement or, alternatively, other reconstructive procedures, including the "edge-to-edge" technique, have been proposed.2,4 Previous studies with the edge-to-edge repair for FMR have alternately reported encouraging,5 satisfactory,6 or disappointing results.7 Most of the unfavorable outcomes, however, have been described when the edge-to-edge technique has been used without any concomitant annuloplasty5 or in association with only a posterior flexible band,5,7 which could not prevent the progression of annular dilatation in the setting of DCM. In this study, we hypothesized that, in presence of FMR and important leaflet tethering, the association of the "edge-to-edge" technique to the undersized annuloplasty could improve the durability of the repair, provided that a complete ring is used and that the surgical correction is guided by specific echocardiographic information.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Study Population
The study population consisted of 77 patients with end-stage DCM and severe (4+/4+) or moderately severe (3+/4+) mitral regurgitation (MR), refractory to medical therapy, who underwent mitral repair from 1998 to March 2004. The etiology of DCM was ischemic in 51 cases (67.2%) and idiopathic in 26 (33.8%). All of the patients had severe LV dysfunction [ejection fraction (EF) <35%] and had been hospitalized 1 to 5 times for congestive heart failure (CHF) in the previous 6 months despite maximal medical therapy. The baseline characteristics are presented in Table 1. Patients with intrinsic mitral valve (MV) disease, unstable angina or recent myocardial infarction (<6 months), papillary muscle rupture, severe right ventricular dysfunction, multiple organ failure, EF >35%, concomitant LV reconstruction, or aortic valve procedures were excluded from the study.
|
Echocardiography
All of the patients underwent transthoracic (TTE) followed by transesophageal echocardiography (TEE). The severity of MR was graded as follows: mild, 1+ (jet area/left atrial area <10%); moderate, 2+ (jet area/left atrial area 10% to 20%); moderately severe, 3+ (jet area/left atrial area 20% to 45%); and severe, 4+ (jet area/left atrial area >45%). The vena contracta width, at the narrowest portion of the regurgitant jet, the coaptation depth, the tenting area of the MV, and the site of origin of the regurgitant jet were also measured and analyzed. The severity and precise mechanism of MR were confirmed by TEE. Dobutamine stress echocardiography was performed in all of the ischemic patients but 8 (15.6%) in whom atrial fibrillation, sinus tachycardia, or inducible ventricular arrhythmias were present. Dobutamine stress echocardiography was performed as described previously 8 to assess the viability and contractile reserve and to distinguish patients who could benefit from concomitant coronary artery bypass graft (presence of viability) from those who required only MV surgery (absence of viability).
Immediately after surgery, TEE was repeated to assess residual MR, transmitral diastolic gradient (from continuous-wave Doppler), and MV area (by direct planimetry). Serial TTEs were then performed at hospital discharge, within 3 months (average, 2.7±0.3) and every 6 months after surgery.
"Echo-Guided" Approach and Surgical Technique
In MR secondary to DCM, the MV is structurally normal, and the intraoperative inspection does not provide additional information useful to the choice of the surgical technique. The preoperative echocardiographic study, therefore, has literally to be used as a guide for the surgical correction providing detailed information about the degree of annular dilatation (intercommissural/septolateral dimensions), the severity of leaflet tethering (coaptation depth/tenting area), and the number and localization of the regurgitant jets.
The strategy adopted in our institution can be simplified as follows: in the presence of annular dilatation and moderate leaflet tethering (coaptation depth <1 cm), an undersized annuloplasty alone, with a complete ring, was used. On the other hand, when the degree of tethering was more pronounced (coaptation depth 1 cm), an association of the edge-to-edge technique to the undersized annuloplasty was preferred, with the aim to prevent the recurrence of MR and improve the durability of the repair. The edge-to-edge repair was always performed in correspondence with the site of origin of the regurgitant jet: centrally (in case of central jet) or posteromedially (when the regurgitant jet was in correspondence of the posterior commissure) (Figure 1). When more than 1 jet was identified, the edge-to-edge was applied on the largest one, relying on the undersized ring for the resolution of the others. All of the patients, with or without edge-to-edge, received an undersized annuloplasty with a complete ring, rigid or semirigid in most of the cases. An intraaortic balloon pump (IABP) was prophylactically inserted in 60 patients (77.9%) before skin incision. The distribution of those patients with IABP insertion was not different between the 2 groups (P=0.7). Although initially we were using IABP on an "as-needed basis" after the operation, from December 2000, we decided to adopt a prophylactic IABP in all of the patients with severe LV dysfunction. We used this approach also in nonischemic patients in whom the decreased afterload provided by the IABP can be helpful considering that those failing hearts, immediately after the operation, have to face the "afterload mismatch" produced by the newly competent repaired MV.
|
Follow-Up
All of the patients were examined in the heart failure clinic with physical examination, ECG, and TTE. The events included cardiac death, reoperation, being listed for heart transplantation, endocarditis, and thromboembolisms. Follow-up was 100% complete (mean time, 18±9.8 months; range, 1 month to 5 years; median, 14 months).
Statistical Analysis
All of the data were prospectively entered in a dedicated database and analyzed. Calculations were performed using SPSS version 11.5 (SPSS Inc.) for the Windows (Microsoft Corp.) software package. Continuous data were compared using the Student t test for paired and unpaired data when appropriate. A comparison of categorical variables was performed using 
2 and Fisher’s exact test. Repeated measurements were compared by repeated-measures ANOVA. The Bonferroni test was used to adjust the observed significance level for the fact that multiple comparisons were made. Survival and freedom from events were analyzed with Kaplan-Meier actuarial methods. The comparison among groups was performed according to the log rank method. Univariate analysis of risk factors was performed with Cox proportional hazards regression. All of the data are presented as mean±SD (for actuarial estimates, SE is reported instead).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Of 77 patients, 23 had a coaptation depth <1 cm and underwent isolated undersized annuloplasty (ring-only group). In the remaining 54 patients, the coaptation depth was
1 cm, and an edge-to-edge was associated with the ring annuloplasty (edge-to-edge group). Preoperative characteristics were comparable in the 2 groups, although the mean tenting area and coaptation depth were obviously higher in the edge-to-edge one. All of the patients were receiving heart failure medication at the maximal tolerated dose without significant differences between the 2 groups (Table 2). A single central regurgitant jet was documented in 19 cases (82.7%) in the "ring-only group" and in 37 cases (68.6%) in the edge-to-edge one (P=0.2). The remaining patients in both groups presented a "complex jet" (an eccentric jet originating from the medial commissure or a central jet associated with a subcommissural or a commissural one). The prevalence of complex jets was comparable in the 2 groups. The edge-to-edge repair was performed centrally in 47 cases (87%) and posteromedially in 7 (13%). All of the patients, with and without edge-to-edge, received an undersized complete ring. The annuloplasty rings were semirigid (Seguin, St. Jude Medical) or rigid (Carpentier–Edwards, Classic) in 88.3% of the cases (68 of 77) and flexible (Duran, Medtronic Inc.) in 9 patients (11.7%) (Figure 2). The flexible rings were used exclusively at the beginning of our experience and then abandoned according to the new emerging evidence that more rigid devices should be preferred in the setting of DCM.9 The dimensions of the mitral annulus and the type of ring did not significantly differ between the 2 groups. Undersizing was initially achieved by using a ring that was 2 sizes smaller than the annular size measured intraoperatively. Intraoperative measurement was thereafter abandoned, and we simply used the smallest ring we could implant without inducing mitral stenosis, this being a number 26 in patients receiving an isolated undersized annuloplasty and a number 28 in those submitted to concomitant edge-to-edge repair. Thus, in patients receiving the edge-to-edge, the mean ring size was significantly higher (28.2±2.3; range, 26 to 31) than in patients undergoing undersized annuloplasty alone (27.3±1.2; range, 25 to 30; P=0.01) (Figure 3). Concomitant CABG was performed in 39 of 51 patients (76.4%) with ischemic DCM (mean 2.0±0.8 grafts per patient). The target coronary artery was the left anterior descending in 42.5% of the cases, obtuse marginal in 28.7%, posterior descending artery in 20%, and diagonal branches in 8.5%. In the remaining 12 ischemic patients (23.6%), myocardial revascularization was not performed because of unsuitable coronary anatomy (3 patients) or the presence of scar in the target region (9 patients). The distribution of those 12 patients was not significantly different between the 2 groups (P=0.4). Other associated procedures were tricuspid valve repair in 8 patients (10.3%), bipolar radiofrequency ablation of permanent atrial fibrillation in 9 (11.6%), and cardiac support device (CorCap, ACORN Cardiovascular) implant in 2 (2.5%).
|
|
|
Clinical Outcome
Postoperative complications are reported in Table 3. Three patients died within 30 days (3 of 77, 3.8%). Hospital mortality was 4.3% (1 of 23) in the ring-only group and 3.7% (2 of 54) in the edge-to-edge one (P value was not significant; low cardiac output in 2 patients, ventricular arrhythmia in 1). Three more patients (3.8%) died during the follow-up: 1 of pneumonia 6 months postoperatively and 2 of sudden death 3 and 7 months after surgery. The overall actuarial survival was 90.7±3.64% at 2.7 years (89.2±7.2% in the ring-only group and 91.4±4.12% in the edge-to-edge group; P=0.9). One patient was reoperated 12 months after an undersized annuloplasty with a 27-mm flexible ring for recurrence of severe MR. One more patient, with 2+ recurrent MR at 1 year, was listed for heart transplantation because of persistence of CHF symptoms. Freedom from cardiac events (cardiac death, reoperation, being listed for heart transplantation, endocarditis, and thromboembolisms) was 81.8%±7.96%. New York Heart Association (NYHA) class improved from a preoperative mean of 3.4±0.4 to 1.4±0.6 (P<0.0001) with all of the surviving patients but 3 being in class I or II and no significant difference between the 2 groups.
|
Echocardiography
At hospital discharge, TTE showed mild MR in 29 (37.6%) patients and no MR in 48 (62.3%). On average, the patients had grade 0.6±0.3 MR, a mean MV area of 2.8±0.6 cm2 (range, 1.9 to 3.4 cm2) and a mean transmitral diastolic gradient of 3.8±1.2 mm Hg. No significant differences were found between the 2 groups. The first follow-up TTE (<3 months after surgery) showed similar results. At the last echocardiogram (mean, 1.5 years after surgery), MV area (mean, 2.8±0.4 cm2) and transmitral diastolic gradient (mean, 3.3±2.1 mm Hg) had remained unchanged, whereas MR grade had increased to 1.2±1.0 (P=0.0001 compared with predischarge). In particular, 79.2% of the patients had no or mild MR, and 11.6% showed moderate (2+) mitral insufficiency (Table 4). Recurrence of 3 to 4+ MR occurred in 7 patients (9%): 5 in the ring-only group (5 of 23, 21.7%) because of reappearance of bileaflet tethering and 2 in the edge-to-edge group (2 of 54, 3.7%; P=0.02). In those last 2 patients, a posteromedial edge-to-edge had been performed, and its failure was likely related to insufficient extension of the suture with the reappearance of a commissural regurgitant jet. All of those 7 patients developed MV repair failure within the first postoperative year. A reoperation was performed in 1 of them. The other 6 are currently closely followed-up, as they are still in functional class II. Univariate analysis identified the absence of the edge-to-edge as the only predictor of recurrence of MR of grade 3 to 4+ (hazard ratio, 4.7; P=0.03). No other risk factor for MV repair failure reached statistical significance or even a probability value 
0.10, and, therefore, a multivariable analysis was not performed (Table 5). The ring size was reduced over the duration of the trial, but this had no impact on the mitral repair failure rates. The freedom from recurrence of 3 or 4+ MR at 1.5 years was 95±3.3% in the edge-to-edge group and 77±12.1% in the isolated undersized annuloplasty group (P=0.04) (Figure 4). Echocardiographic changes in LV dimensions and function were assessed in patients who had 2 TTE follow-ups. Patients submitted to radiofrequency ablation of atrial fibrillation or cardiac support device implantation and those with recurrence of MR 2+ were excluded from the analysis, because all of these conditions would have been confounding factors in the LV remodeling evaluation. Using the criteria described above, 28 patients (20 ischemic and 8 idiopathic) were available for the purpose of this analysis. Four of them were in the ring-only and 24 were in the edge-to-edge group. The mean LV end-diastolic diameter, end-diastolic volume, and end-systolic volume significantly decreased from baseline to early follow-up, remaining stable from early to late postoperative stage (Figure 5). A significant improvement of the EF was documented only in the ischemic DCM patients who received a mean of 1.5±0.3 grafts. In the 8 patients with idiopathic DCM, on the other hand, we observed a significant decrease in the LV dimensions without changes of the EF (from 22±6.5% to 25±9.8%; P=0.22). The same analysis performed in 5 of the 7 patients with 3 to 4+ recurrent MR (1 excluded for CorCap implantation and 1 for having had only 1 TTE follow-up), showed that LV dimensions and function had remained comparable with the preoperative values. Insufficient data were available for the patients with moderate (2+) recurrent MR. Finally, differences in the remodeling pattern between ring-only and edge-to-edge patients could not be assessed for the small number of suitable cases in the isolated undersized annuloplasty group.
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The main finding of this study is that the association of the edge-to-edge repair to a complete-ring undersized annuloplasty significantly increases the durability of mitral repair in FMR because of end-stage DCM. This technique restores leaflet coaptation also in patients with excessive tethering and complex jets without inducing mitral stenosis. Patients with CHF and FMR (3 to 4+/4+) have a life expectancy of
70% at 1 year after the diagnosis.10 In this setting, MV repair with a standard undersized ring annuloplasty has been performed with low hospital mortality, symptomatic improvement, and good midterm survival.11 However, if substantial apical tenting is present or the pattern of the regurgitant jet is complex, then ring annuloplasty alone may not be sufficient.4 The edge-to-edge technique has been used in patients with FMR and DCM, particularly by the Cleveland Clinic group with a disappointing recurrence rate of moderately severe (3+) MR at 2 years of 24%.7 These results, however, were mainly explained by the fact that, in that series, the edge-to-edge repair was always used in association with a posterior flexible band, which may not prevent the recurrence of annular dilatation in the setting of advanced DCM. Indeed, the patients requiring reoperation almost invariably presented redilatation of the mitral annulus. In our institution, we decided to use the edge-to-edge technique exclusively in patients with important leaflet tethering with the aim to improve the durability of the undersized annuloplasty and avoid MV replacement. However, the edge-to-edge was always associated with an undersized annuloplasty performed with a complete ring (preferentially rigid or semirigid) and the site of leaflet approximation was literally guided by the TTE/TEE findings. This approach resulted in a 3.7% recurrence rate of 3 to 4+ MR at follow-up that was 6-fold lower compared with that registered with the undersized annuloplasty alone (21.7%) despite having, in the edge-to-edge patients, the more-advanced degree of leaflet tethering. Freedom from recurrence of severe MR at 1.5 years was 20% higher in the edge-to-edge group. Even considering all the cases of recurrence of MR 2+, the 14.8% rate of our edge-to-edge experience compared favorably with the 34% registered in the ring-only group of our series and with the 29% reported by Tahta et al,1 using a flexible downsized ring. In the current study, the use of the edge-to-edge technique was identified as the only significant predictor of durability of MV repair. By ensuring leaflet coaptation exactly where the tethering is more pronounced, the edge-to-edge approach could abolish the occurrence of the "loitering effect"12 and prevent the recurrence of MR. Moreover, by anchoring the leaflets together, it could exert a kind of "reins" effect on the LV chamber, counteracting the progression of the LV remodeling, which can lead to recurrence of MR.13 The serial echocardiographic follow-ups allowed the assessment of LV reverse remodeling after surgery showing that, when volume overload is abolished by a successful repair, a significant decrease of LV dimensions develops early after surgery and remains substantially stable in the following 15 months in both idiopathic and ischemic patients. Finally, for "very sick" patients, as those included in this series, the opportunity of performing a reconstructive procedure rather than a chordal-sparing valve replacement has been questioned.4,14 We have always been rather concerned by the impact of MV replacement (although performed preserving all the subvalvular apparatus) on LV systolic function over time. If a fairly durable MV repair can be performed, we still believe this would represent the best option for these high-risk patients.
Limitations
This study has several limitations. The number of patients is relatively small, together with the number of late events. The assignment of the patients to one or another technique was not performed in a randomized fashion. However, it should be considered that only patients with MR secondary to really end-stage DCM (EF <35%; NYHA III-IV) were prospectively included in this series and that the number of events, although relatively small, was enough to show a statistically significant difference between the 2 groups. As far as the assignment of the patients to one or another technique is concerned, we used criteria that have been reported to be associated with failure of the undersized annuloplasty (coaptation depth >1 cm; complex jets) to select the cases where adding the edge-to-edge technique could have been potentially beneficial. In particular, we used the coaptation depth, which represents the ultimate mechanism of functional MR independent of LV function and shape (sphericity index).15 Several factors (ischemic or idiopathic etiology of DCM, age, previous AMI, EF, LV dimensions, annuloplasty ring type and number, associate CABG, and type of regurgitant jet) might have influenced the results. Nevertheless, none of them were shown to be significantly correlated with the recurrence of MR. The potential role of concomitant myocardial revascularization on MV function has also to be considered. We cannot exclude that concomitant CABG might have improved the mitral apparatus function by some recruitment of the hibernating myocardium. However, there were no differences between the 2 groups in terms of presence and distribution of hibernating myocardium, number of patients undergoing CABG, and number of grafts and coronary arteries revascularized. Therefore, if any effect on mitral function has been exerted by concomitant CABG, this should have been comparable in the 2 groups. Finally, these data have to be considered as preliminary results, which need to be confirmed by a larger number of patients and a significantly longer follow-up.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Tahta SA, Oury JH, Maxwell JM, Hiro SP, Duran CM. Outcome after mitral valve repair for functional ischemic mitral regurgitation. J Heart Valve Dis. 2002; 11: 11–19.[Medline] [Order article via Infotrieve]
2. Calafiore AM, Gallina S, Di Mauro M, Gaeta F, Iaco AL, D’Allesandro S, Mazzei V, Di Giammarco G. Mitral valve procedure in dilated cardiomyopathy: repair or replacement. Ann Thorac Surg. 2001; 71: 1146–1152.
3. Hung J, Handschumacher MD, Rudski L, Chow CM, Guerrero JL, Levine RA. Persistence of ischemic mitral regurgitation despite annular ring reduction: mechanistic insights from 3D echocardiography. Circulation. 1999; 100: I–73.
4. Miller DC. Ischemic mitral regurgitation redux –To repair or to replace? J Thorac Cardiovasc Surg. 2001; 122: 1059–1162.
5. Umana JP, Salehizadeh BS, DeRose JJ, Nahar T, Lotvin A, Homma S, Oz MC. "Bow-tie" mitral valve repair: an adjuvant technique for ischemic mitral regurgitation. Ann Thorac Surg. 1998; 66: 1640–1646.
6. Kinnaird TD, Munt BI, Ignaszewski AP, Abel JG, Thompson CR. Edge-to-edge repair for functional mitral regurgitation: an echocardiographic study of the hemodynamic consequences. J Heart Valve Dis. 2003; 12: 280–286.[Medline] [Order article via Infotrieve]
7. Bhudia SK, McCarthy PM, Smedira NG, Lam B, Rajeswaran J, Blackstone EH. Edge-to-edge (Alfieri) mitral repair: results in diverse clinical settings. Ann Thorac Surg. 2004; 77: 1598–1606.
8. La Canna G, Alfieri O, Giubbini R, Gargano M, Ferrari R, Visioli O. Echocardiography during infusion of dobutamine for identification of reversibly dysfunction in patients with chronic coronary artery disease. J Am Coll Cardiol. 1994; 23: 617–626.[Abstract]
9. Tibayan FA, Rodriguez F, Langer F, Zasio M, Bailey L, Liang D, Daughters GT, Ingels NB, Miller DC. Annular remodeling in chronic ischemic mitral regurgitation: ring selection implications. Ann Thorac Surg. 2003; 76: 1549–1555.
10. Trichon BH, Felker GM, Shaw LK, Cabell CH, O’Connor CM. Relation of frequency and severity of mitral regurgitation to survival among patients with left ventricular systolic dysfunction and heart failure. Am J Cardiol. 2003; 91: 538–543.[CrossRef][Medline] [Order article via Infotrieve]
11. Romano MA, Bolling SF. Update on mitral repair in dilated cardiomyopathy. J Card Surg. 2004; 19: 396–400.[CrossRef][Medline] [Order article via Infotrieve]
12. Glasson JR, Komeda M, Daughters GT, Bolger AF, Karlsson MO, Foppiano LE, Hayase M, Oesterle SN, Ingels NB Jr, Miller DG. Early systolic mitral leaflet "loitering" during acute ischemic mitral regurgitation. J Thorac Cardiovasc Surg. 1998; 116: 193–205.
13. Hung J, Papakostas L, Tahta SA, Hardy BG, Bollen BA, Duran CM, Levine RA. Mechanism of recurrent ischemic mitral regurgitation after annuloplasty. Continued LV remodeling as a moving target. Circulation. 2004; 110 (suppl II): II85–II89.
14. Gillinov AM, Wierup PN, Blackstone EH, Bishay ES, Cosgrove DM, White J, Lytle BW, McCarthy PM. Is repair preferable to replacement for ischemic mitral regurgitation? J Thorac Cardiovasc Surg. 2001; 122: 1125–1141.
15. Yiu SF, Enriquez-Sarano M, Tribouilloy C, Seward JB, Tajik AJ. Determinants of the degree of functional mitral regurgitation in patients with systolic left ventricular dysfunction: a quantitative clinical study. Circulation. 2000; 102: 1400–1406.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||