Early Surgery in Patients With Mitral Regurgitation Due to Flail Leaflets
A Long-term Outcome Study
Background The optimal timing for surgery in patients with mitral regurgitation is disputed. Because of the frequency of left ventricular dysfunction, which is difficult to predict, early surgery has been recommended, but its potential benefits have not been demonstrated.
Methods and Results The outcomes of 221 patients (mean age, 65±13 years; 71% males) with flail leaflets diagnosed with two-dimensional echocardiography between 1980 and 1989 who were eligible for operation were analyzed. Group I comprised 63 patients who had early mitral valve surgery (within 1 month after diagnosis). Group II comprised 158 patients initially treated conservatively (80 of whom were operated on later). Group I patients were younger (P=.009), had more symptoms (P<.0001), and were more frequently in atrial fibrillation (P=.023) than group II patients. There was no difference in ejection fraction between the groups. The early surgery strategy was followed by an improved overall survival rate (P=.028) and a lower incidence of cardiovascular deaths (P=.025), congestive heart failure (P=.046), and new chronic atrial fibrillation (P=.032), as confirmed by multivariate analysis (adjusted risk ratios of 0.31, 0.18, 0.38, and 0.05, respectively; all P<.02).
Conclusions In patients with mitral regurgitation due to flail leaflets, the strategy of early surgery versus conservative management is associated with an improved long-term survival rate, decreased cardiac mortality, and decreased morbidity after diagnosis. This outcome advantage suggests that early surgery is a reasonable treatment option to be considered in low-risk candidates with repairable valves and severe mitral regurgitation.
Surgical correction of mitral regurgitation (MR) is followed by marked improvement in cardiac symptoms and has traditionally been used for patients who develop severe symptoms during the course of medical follow-up.1 However, this approach is associated with a disappointingly high incidence of postoperative left ventricular dysfunction and late mortality.2 3 4 This observation in combination with the excellent results of valvular reconstruction5 has led recently to the suggestion of early surgical correction of MR.6 7 However, evidence that a strategy of early surgery improves the long-term outcome of MR is not yet available.6 This would require comparing the outcome after early surgery with that after conservative treatment (ie, medical therapy with delayed surgery performed whenever judged indispensable) in patients with severe MR of uniform cause and mechanism.
Patients with organic MR due to flail leaflets are ideal candidates for such a comparison because it is invariably associated with a marked degree of regurgitation,8 9 it represents the commonest cause of surgical MR in developed countries,10 and it is diagnosed reliably with two-dimensional echocardiography.11 Although MR due to flail leaflets is associated with excess mortality and high morbidity,12 there is no agreement on the best method of treatment. On the basis of limited series, recommendations have been made for both early expeditious surgery13 and conservative management with possible deferred surgery.14 15 16 Therefore, it is uncertain whether early surgery or conservative management should be the preferred approach in these patients irrespective of symptoms.
Accordingly, we examined the long-term outcome of patients who were surgical candidates and had MR due to flail leaflets diagnosed with echocardiography and hypothesized that compared with conservative management, early surgical correction of MR is associated with improved long-term clinical outcome by (1) decreasing overall mortality and, specifically, cardiovascular mortality and (2) reducing cardiovascular morbidity.
Inclusion criteria were patients with flail leaflets demonstrated with two-dimensional echocardiography and a diagnosis that was originally made between January 1, 1980, and December 31, 1989, at the Mayo Clinic. Exclusion criteria were papillary muscle rupture; previous valvular surgery; associated mitral stenosis, aortic valve, and congenital heart disease; and patients who were not candidates for surgery (on the basis of comorbidity or age), as judged by the primary physician. Patients with incidental coronary artery disease or bypass grafting were not excluded.
The patients were separated into two groups: group I included those operated on within 1 month after diagnosis (“early surgery”), and group II included those treated conservatively (ie, those with initial medical follow-up). Cardiac surgery at our institution is performed the day after the consultation unless explicitly decided otherwise, and therefore delay was always a voluntary decision and in no case was due to an inability to accommodate the patient in the surgical schedule.
The noncardiac comorbidities were summated as a score.17 Follow-up information was obtained by consultation or by use of the institutional medical survey center to send questionnaires or to make telephone calls to patients, their relatives, or physicians. The modes and causes of death and complications were noted as previously recommended.18
On the index echocardiograms, the classic criteria for diagnosis of flail leaflets were used.11 The degree of MR was assessed semiquantitatively.19 Left atrial diameter, left ventricular diameter, ejection fraction,20 21 and wall thickness were measured and used unaltered by electronic transfer from the echocardiogram database to the study file.2 3
Group data were expressed as mean±SD or percentages. Group comparisons were performed with the t test or χ2 test as appropriate. End points analyzed were overall survival, cardiac death, congestive heart failure (CHF), new-onset chronic atrial fibrillation, thromboembolism, major bleeding, and infective endocarditis. All analyses were performed in accordance with intention to treat. The rates of survival and other end points were estimated by use of the Kaplan-Meier method. Follow-up time was defined in two alternative ways: first, from the time of echocardiographic diagnosis in both groups, and second, from the time of surgery in group I patients and from the mean time to surgery (of group I patients) for group II. Overall and cardiac mortality were analyzed both ways, whereas nonfatal complications (CHF, chronic atrial fibrillation, thromboembolism, major bleeding, and endocarditis) were analyzed using the second definition. Patients with complications occurring between the diagnosis and the starting time of the analysis were excluded from the specific analysis. Direct comparison of outcomes between groups I and II used the log-rank test.
Multivariate analysis was performed with the use of Cox proportional hazards analysis. For each end point, the baseline independent determinants of outcome were included in the model, and a group variable (ie, early surgery versus conservative treatment) was also forced into the model. This model was also repeated with the comorbidity index to adjust for comorbidity. To determine the effect on outcome of early surgery, if defined beyond 1 month after diagnosis, the analysis was repeated using a definition of 3 and 6 months after the index echocardiography. A value of P<.05 was considered significant.
Of 229 patients with flail leaflets diagnosed during the study period, 8 (mean age, 79±8 years) were considered by their primary physician to be unsuitable candidates for operation because of advanced physiological age (4 patients) or comorbid diseases (4 patients) and were excluded from further analysis. Partial data on the natural history with medical treatment of these patients have been presented previously.12 The remaining 221 patients (mean age, 65±13 years; 71% males) form the subject of the present study. Of these, 63 patients formed group I (early surgery) and 158 patients formed group II (conservative management). The mean time to surgery was 9±7 days (range, 0 to 31 days) in group I. Of the group II patients, 80 were subsequently operated on between 1 month and 10 years after diagnosis (mean, 41±33 months) and 78 did not have surgery. Among the 158 patients in group II, surgery was deferred in 138 (87.3%) because they had no or minimal symptoms with medical therapy, in 10 (6.3%) to allow improvement of heart failure by medical treatment (8 of whom eventually had surgery), and in 10 (6.3%) because surgery was declined. Follow-up was 100% complete to February 1995.
As noted in Table 1⇓, group I patients were slightly younger, more severely symptomatic, more often in atrial fibrillation at presentation, and had a slightly lower comorbidity index than group II patients. There were no differences between the groups in left ventricular function, left atrial size, or cause of the flail leaflets (P=.62). During follow-up, group II patients took diuretics more frequently (58% versus 32%; P=.001) and warfarin less frequently (31% versus 52%; P=.011) than group I patients. The maintenance doses of warfarin were similar in group I (4.2±1.8 mg) and group II (4.4±2.1 mg) (P=.61), as was the use of ACE inhibitors (P=.14), β-blockers (P=.66), digoxin (P=.21), hydralazine (P=.90), class I antiarrhythmic agents (P=.22), and antiplatelets (P=.56).
Coronary angiography was performed in 34 patients in group I (54%) and in 69 in group II (44%) and showed ≥70% diameter stenosis of coronary arteries in 10 patients (16%) and 20 patients (13%), respectively (P=.44).
Surgery was performed in all 63 group I patients and eventually in 80 of the group II patients. There was no difference between these two groups of patients who had surgery in the percentage of those who were in New York Heart Association (NYHA) class III or IV preoperatively (63% versus 69%; P=.10), those with valve repair (67% versus 66%; P=.67), or those with coronary bypass surgery performed (16% versus 24%; P=.33), nor with regard to anesthesia time (P=.32), bypass time (P=.17), or cross-clamp time (P=.24).
Comparison of Outcome Between the Two Management Strategies
Nine deaths occurred among the patients in group I and 47 among those in group II (Table 2⇓). The overall survival rate at 5 and 10 years, respectively, was 89±4% and 79±8% for group I patients and 78±3% and 65±5% for group II patients (P=.028) (Fig 1⇓). The operative mortality rate was 1.6% (1 of 63 patients) in group I and 6.3% (5 of 80) for the patients in group II who had surgery (P=.17).
The analysis was repeated after the definition of early surgery was modified to that performed within 3 months and within 6 months after diagnosis. When both definitions were used, early surgery continued to be associated with lower mortality rates (P=.013 and P=.0038, respectively). These analyses were repeated using the second definition of follow-up starting time as defined in “Methods.” This had the effect of eliminating three early deaths that occurred in group II. The results were essentially unchanged, with probability values of .047, .024, and .0078 for the 1-, 3-, and 6-month definitions, respectively.
Multivariate analysis adjusting for the predictors of outcome at baseline demonstrates that early surgery is independently associated with better survival (Table 3⇓). Fig 2⇓ displays the model-based adjusted survival curves constructed separately in NYHA class I or II and class III or IV, showing the higher mortality rate and greater benefit of early surgery in class III or IV. Forcing the comorbidity index in the multivariate model did not alter the significant reduction of mortality associated with early surgery (adjusted risk ratio [RR] [95% CI]=0.30 [0.12 to 0.71]; P=.0076). Similar benefits of early surgery were maintained when the second definition of follow-up starting time was used (adjusted RR [95% CI]=0.37 [0.15 to 0.89]; P=.026).
Although cardiovascular deaths represented a high percentage of all deaths (67% and 75% in groups I and II, respectively), there were only 6 cardiovascular deaths in group I and 35 in group II (Table 2⇑). The 5- and 10-year cardiac death rates were 8±4% for both time periods in group I and 16±3% and 29±5%, respectively, in group II (P=.025) (Fig 3⇓). In multivariate analysis, early surgery was associated with decreased cardiovascular mortality (adjusted RR, 0.18; P=.002) (Table 3⇑).
CHF occurred in 13 patients in group I and 61 in group II during follow-up. In group I, this complication was related to recurrent valvular regurgitation in 5 patients who had failed valve repair and to myocardial dysfunction in 8. For group II patients, the first episode of CHF was related to valvular regurgitation in 53 patients (1 case due to prosthetic failure), to myocardial dysfunction in 7 after valve surgery (2 had prostheses implanted), and to both abnormalities in 1 patient. The cumulative likelihood of CHF in group I at 5 and 10 years, respectively, was 18±5% and 27±7% compared with 26±4% and 59±7% in group II (P=.046) (Fig 4⇓). The adjusted RR of CHF in the early surgery group was 0.38 (P=.015) (Table 3⇑).
Of the 159 patients initially in NYHA class I or II, 2 in group I (9.1%) and 54 in group II (39.4%) had progressed to NYHA class III or IV or experienced CHF during follow-up. The incidence of this event at 5 and 10 years, respectively, was 5±5% and 10±7% in group I and 24±4% and 58±7% in group II (P=.022).
Of the 171 patients (77%) initially in sinus rhythm, chronic atrial fibrillation developed in 1 patient in group I and in 18 in group II. At 5 and 10 years, 0% and 4±4% of group I patients, respectively, had new chronic atrial fibrillation, whereas 8±3% and 26±7% of group II patients had developed this arrhythmia (P=.032) (Fig 5⇓). In multivariate analysis, permanent atrial fibrillation was less frequent in group I (RR, 0.05; P=.011) (Table 3⇑).
A total of 15 thromboembolic events occurred in group I and 33 in group II. The cumulative incidence of thromboembolic events at 5 and 10 years, respectively, was 19±5% and 27±6% in group I and 18±3% and 37±8% in group II (P=.66).
Endocarditis occurred in only 11 group II patients, of whom 2 contracted it before late operation, 2 contracted it after valve surgery, and 7 ultimately did not have surgery. The likelihood of endocarditis was 5±2% at 5 years and 8±3% at 10 years in group II (P=.038). Because of the small number of events, no further analysis of this complication was performed.
Major Bleeding Episodes
Bleeding complications occurred in five patients in group I and six in group II. The probability of bleeding at 5 and 10 years was 6±3% and 12±5%, respectively, in group I and remained unchanged at 5±2% in group II (P=.25).
The present study shows that in patients with MR due to flail leaflets, the two treatment strategies (early surgery or conservative management followed by surgery whenever deemed necessary) are associated with significantly different outcomes. After adjustment for baseline differences in symptoms, age, comorbidity, and ejection fraction, early surgery is followed by a lower long-term overall mortality rate, which is explained by decreased cardiovascular mortality. Early surgery is also followed by a decreased incidence of atrial fibrillation and heart failure, without a significant difference in embolic or hemorrhagic complications.
Rationale for Early Surgery
Surgical correction of MR undisputedly results in an improvement of symptoms when they are present preoperatively.1 The evidence that it may improve survival is limited22 ; therefore, the most common management strategy has been to wait for frank symptoms to develop before performing surgery. Two types of arguments have served as a rationale for early surgery.
First, in series in which surgical correction of MR was indicated mostly for severe symptoms, left ventricular dysfunction was a frequent postoperative complication3 4 and was associated with poor postoperative survival.2 The objective measures of left ventricular function are useful for detecting left ventricular dysfunction2 3 23 24 25 but are imperfect tools for detecting early contractile abnormality.23 26 This lack of sensitivity may lead to the occurrence of “unexpected” left ventricular dysfunction.3 Also, prolonged duration of MR is associated with the increased occurrence of postoperative left ventricular dysfunction.3 Waiting for severe symptoms to indicate surgical correction of MR is also associated with excess mortality,27 excess incidence of postoperative left ventricular dysfunction,3 and heart failure.6 All these arguments tend to suggest that delaying an operation that is almost unavoidable12 may be a source of poor results.
Second, the spectacular results of valve repair28 29 30 represent a major argument for early surgery. Repair can now be accomplished in a high percentage of patients with severe MR5 28 30 and at a low risk in experienced medical centers.5 30 Repair is associated with improved survival and less morbidity than valve replacement.5 31 32
However, although the success and low risk of valve repair make early surgery an acceptable concept, they do not prove that it should be the preferred management strategy. An early surgical approach may necessitate unexpected valve replacement, and reservations regarding early surgery persist because a beneficial effect on outcome has never been substantiated in controlled studies.30
In MR due to flail leaflets, the degree of regurgitation is uniformly large,8 9 33 which eliminates an important source of variation in management. Despite this, various opinions on management strategy have been expressed. Several reports have insisted on the severity of cardiac decompensation,9 13 34 35 36 whereas others have found it indistinguishable clinically and hemodynamically from others with chronic MR,8 37 often with minimal symptoms at diagnosis37 that are easily controlled with medical therapy.14 15 Therefore, recommendations as diverse as favoring immediate operative therapy13 34 and providing conservative treatment15 have been made. This uncertainty is disturbing because flail leaflets represent a major cause of surgical MR in developed countries.10 30
The present study is, to the best of our knowledge, the first to analyze the outcome of MR due to flail leaflets in two subgroups defined by the two possible contemporary management strategies (ie, early surgery and conservative treatment). The results of the present study significantly favor early surgery, in particular because of improved survival even after adjustment for all other prognostic indicators and comorbidities.
Effect of Early Surgery on Survival
Few studies have compared medical therapy with surgery for MR,22 38 39 40 41 and they have yielded conflicting results. These uncertainties probably result from small40 41 and old22 38 series with inadequate controls,38 39 41 poorly defined degrees of MR,38 and lack of data on comorbidity.22 38 40 41 The relevant question, however, is not whether surgery improves survival (because surgery is almost unavoidable eventually for relief of symptoms)1 12 but whether early surgery may improve long-term survival by avoiding the complications of MR that may occur during the phase of medical observation.12 39 41 To the best of our knowledge, the comparison of the two management strategies, early surgery and conservative management, has not been reported previously. In the present series, early surgery was associated both with lower long-term mortality because of a marked decrease in cardiac mortality and with very low incidences of sudden death and intractable heart failure. This result, found in univariate analysis, was confirmed in multivariate analysis after adjustment for the baseline predictors of survival. This result is confirmed when adjusted for comorbidity in the multivariate analysis. Also, comorbidity was low; only a few deaths were noncardiac, and this did not explain the difference in mortality rates between the two groups. Also important is that conservative management was motivated principally by cardiac status, particularly with regard to patients who were less symptomatic and less frequently in atrial fibrillation and who had a borderline smaller left ventricle. Finally, the lower mortality rate in the early surgery group was not related to more valve repairs performed in this group.5 Therefore, the improved survival and decreased cardiac mortality rate of the patients in group I appear to be intrinsically related to the early surgery, although definitive confirmation should be provided by a randomized study.
The beneficial effect on survival and left ventricular function of valve repair5 31 might be taken as an argument to delay surgery if it consistently restored normal left ventricular mechanics. However, valve repair does not eliminate the risk of left ventricular dysfunction.5 Therefore, the potential of preserving left ventricular ejection performance by means of valvular reconstruction or chordal preservation should not lull clinicians into delaying operation. However, early surgery is conceivable only with a low operative mortality rate; consequently, valve repair is a major incentive for early surgery, which should be performed by surgeons proficient with this technique. Another important determinant of the operative risk is age2 ; therefore, in elderly patients (≥75 years), conservative management may be more appropriate in the absence of frank symptoms.
Effect of Early Surgery on Morbidity
Most studies combining medical and surgical follow-up have provided little information about the impact of surgery on cardiovascular morbidity.22 38 39 40 41 Surgical correction of mitral regurgitation is performed to abolish symptoms and to prevent the development of CHF, but a notable incidence of CHF occurs late after surgery, due mostly to left ventricular dysfunction.6 In the present study, the incidence of new or recurrent CHF was lower in patients who had early surgery, consistent with a decreased morbidity rate after early surgery.
Atrial fibrillation tends to occur less in patients who have had surgery than in those receiving medical treatment.38 In the present study, early surgery was associated with a decreased late incidence of new chronic atrial fibrillation compared with conservative management. The beneficial effect of early surgery in sinus rhythm or of surgery early after the occurrence of atrial fibrillation42 suggests that the suppression of volume overload may interrupt atrial remodeling and prevent its consequences.
There was a significantly lower incidence of endocarditis after early surgery, consistent with the low rates of infection after valve repair,43 but a larger series is required to confirm it as a benefit of early surgery.
The present study is observational, with significant differences in baseline characteristics between the treatment groups and relatively small numbers of patients. Although multivariate analysis provides an opportunity to adjust for these differences, in particular, for age, symptoms, ejection fraction, and comorbidity, caution must be used in applying these results. However, these results are important, especially because no randomized trial comparing these two therapeutic strategies is available. In light of our results, such a randomized trial should be performed.
The present study is the largest one to report the outcome of surgical candidates with MR due to flail leaflets. However, the benefit of early surgery, which is significant overall even after adjustment for symptoms and ejection fraction, cannot be confirmed in individual subgroups at relatively low risk, such as patients with an ejection fraction ≥60% or those with no or minimal symptoms, because of the relatively limited sample size. Future studies, in particular randomized ones, should analyze these important subgroups.
The present series included a relatively early experience with mitral valve repair, and increasing feasibility and experience in repair procedures5 30 should amplify the advantages of early surgery. By contrast, the ability of pharmacological agents to modify the natural history of MR remains unproved.
A 1-month time frame, although empirical, was chosen to minimize both the inclusion period and the events that might occur during that period that might be difficult to interpret. Exclusion of the three deaths in the conservative management group that occurred before the median date of surgery of group I did not change the results. However, our results should not be taken as mandating surgery within 1 month after diagnosis, because our conclusions were not altered using a 3- or 6-month threshold, suggesting that decision making within that delay is appropriate.
The present series of patients with MR due to flail leaflets compares for the first time the outcome associated with the management strategies chosen at diagnosis. Compared with the conservative management strategy, the early surgery strategy is associated with improved long-term survival, decreased cardiac mortality, and decreased morbidity, even after adjustment for age, ejection fraction, symptoms, and comorbidity. This outcome advantage suggests that early surgery is a reasonable treatment option for low-risk candidates with repairable valves and should be considered in patients with the diagnosis of severe MR.
Dr Ling was supported by a Faculty Development Program scholarship from the National University Hospital, Singapore, and the Mayo Foundation. We gratefully acknowledge the statistical assistance of Christine M. Boos.
Reprint requests to Maurice Enriquez-Sarano, MD, Mayo Clinic, 200 First St SW, Rochester, MN 55905.
- Received November 11, 1996.
- Revision received April 23, 1997.
- Accepted April 28, 1997.
- Copyright © 1997 by American Heart Association
Gray R, Helfant R. Timing of surgery for valvular heart disease. Cardiovasc Clin. 1990;21:209-215.
Enriquez-Sarano M, Tajik AJ, Schaff HV, Orszulak TA, Bailey KR, Frye RL. Echocardiographic prediction of survival after surgical correction of organic mitral regurgitation. Circulation. 1994;90:830-837.
Crawford MH, Souchek J, Oprian CA, Miller DC, Rahimtoola S, Giacomini JC, Sethi G, Hammermeister KE, on behalf of Participants in the Department of Veterans Affairs Cooperative Study on Valvular Heart Disease. Determinants of survival and left ventricular performance after mitral valve replacement. Circulation. 1990;81:1173-1181.
Enriquez-Sarano M, Schaff HV, Orszulak TA, Tajik AJ, Bailey KR, Frye RL. Valve repair improves the outcome of surgery for mitral regurgitation: a multivariate analysis. Circulation. 1995;91:1022-1028.
Enriquez-Sarano M, Schaff HV, Orszulak TA, Bailey KR, Tajik AJ, Frye RL. Congestive heart failure after surgical correction of mitral regurgitation: a long-term study. Circulation. 1995;92:2496-2503.
Acar J, Michel PL, Luxereau P, Vahanian A, Cormier B. Indications for surgery in mitral regurgitation. Eur Heart J. 1991;12(B):52-54.
Baxley WA, Kennedy JW, Feild B, Dodge HT. Hemodynamics in ruptured chordae tendineae and chronic mitral regurgitation. Circulation. 1973;48:1288-1294.
Sanders CA, Austen WG, Harthorne JW, Dinsmore RE, Scannell JG. Diagnosis and surgical treatment of mitral regurgitation secondary to ruptured chordae tendineae. N Engl J Med. 1967;276:943-949.
Childress RH, Maroon JC, Genovese PD. Mitral insufficiency secondary to ruptured chordae tendineae. Ann Intern Med. 1966;65:232-244.
Barlow JB. Mitral regurgitation. In: Barlow JB, ed. Perspectives on the Mitral Valve. Philadelphia, Pa: FA Davis Co; 1987:113-150.
Helmcke F, Nanda NC, Hsiung MC, Soto B, Adey CK, Goyal RG, Gatewood RP Jr. Color Doppler assessment of mitral regurgitation with orthogonal planes. Circulation. 1987;75:175-183.
Quinones MA, Waggoner AD, Reduto LA, Nelson JG, Young JB, Winters WL Jr, Ribeiro LG, Miller RR. A new, simplified and accurate method for determining ejection fraction with two-dimensional echocardiography. Circulation. 1981;64:744-753.
Hammermeister KE, Fisher L, Kennedy W, Samuels S, Dodge HT. Prediction of late survival in patients with mitral valve disease from clinical, hemodynamic, and quantitative angiographic variables. Circulation. 1978;57:341-349.
Carabello BA, Nollan SP, McGuire LB. Assessment of preoperative left ventricular function in patients with mitral regurgitation: value of the end-systolic wall stress–end-systolic volume ratio. Circulation. 1981;64:1212-1217.
Wisenbaugh T. Does normal pump function belie muscle dysfunction in patients with chronic severe mitral regurgitation? Circulation. 1988;77:515-525.
Sarano ME, Tajik AJ, Schaff HV, Orszulak TA, Frye RL. Should mitral regurgitation surgery be performed in minimally symptomatic patients? Analysis of the impact of the class III-IV symptoms on postoperative outcome. J Am Coll Cardiol. 1996;27:73a. Abstract.
Loop FD, Cosgrove DM, Stewart WJ. Mitral valve repair for mitral insufficiency. Eur Heart J. 1991;12(suppl B):30-33.
Perier P, Deloche A, Chauvaud S, Fabiani JN, Rossant P, Bessou JP, Relland J, Bourezak H, Gomez F, Blondeau P, D’Allaines C, Carpentier A. Comparative evaluation of mitral valve repair and replacement with Starr, Björk, and porcine valve prostheses. Circulation. 1984;70(suppl 1):I-187-I-192.
Sanders CA, Scannell JG, Harthorne JW, Austen WG. Severe mitral regurgitation secondary to ruptured chordae tendineae. Circulation. 1965;31:506-516.
Roberts WC, Braunwald E, Morrow AG. Acute severe mitral regurgitation secondary to ruptured chordae tendineae: clinical, hemodynamic, and pathologic considerations. Circulation. 1966;33:58-70.
Delahaye JP, Gare JP, Viguier E, Delahaye F, De Gevigney G, Milon H. Natural history of severe mitral regurgitation. Eur Heart J. 1991;12(suppl B):5-9.
Hochreiter C, Niles N, Devereux RB, Kligfield P, Borer JS. Mitral regurgitation: relationship of noninvasive descriptors of right and left ventricular performance to clinical and hemodynamic findings and to prognosis in medically and surgically treated patients. Circulation. 1986;73:900-912.
Horstkotte D, Loogen F, Kleikamp G, Schulte HD, Trampisch HJ, Bircks W. Der Einfluss des prothetischen Herzklappenersatzes auf den natürlichen Verlauf von isolierten Mitral- und Aortenklappenfehlern sowie Mehrklappenerkrankungen: klinische Ergebnisse bei 783 Patienten bis zu 8 Jahren nach Implantation von Björk-Shiley-Kippscheibenprothesen. Z Kardiol. 1983;72:494-503.
Fernandez J, Joyce DH, Hirschfeld KJ, Chen C, Yang SS, Laub GW, Adkins MS, Anderson WA, Mackenzie JW, McGrath LB. Valve-related events and valve-related mortality in 340 mitral valve repairs: a late phase follow-up study. Eur J Cardiothorac Surg. 1993;7:263-270.