Outcome of Mitral Valve Repair for Severe Mitral Regurgitation

Abstract & Commentary

By Michael H. Crawford, MD, Editor

Source: David TE, et al. Late outcomes of mitral valve repair for mitral regurgitation due to degenerative disease. Circulation 2013;127:1485-1492.

Guideline-driven earlier mitral valve repair for patients with mitral regurgitation (MR) due to degenerative disease has increased in the last 25 years. These investigators from Canada examined the outcomes in 840 such patients followed prospectively from 1985 to 2004 approximately every other year for a median follow-up of 10 years. Clinical, hemodynamic, and pathological data were analyzed for predictors of outcomes. The Society of Thoracic Surgeons risk score ranged from 0.3 to 5.5% (mean 1.5%). The surgery was elective in 90% of cases. New York Heart Association (NYHA) classification was: I 15%, II 37%, III 37%, and IV 11%. In 95%, MR was graded severe. Mitral valve prolapse was present in 99%. Death within 30 days occurred in four patients and late death due to cardiac causes in 81. Multivariate predictors of mortality included age, lower ejection fraction (EF), and higher NYHA class. Reoperation was required in 38 patients and the valve was replaced in 30 patients. The probability of reoperation at 20 years was 6%. Only three patients were discharged from the hospital with moderate MR; all the rest had less severe or none. During follow-up, 37 patients developed severe MR and 61 developed moderate MR. At 20 years, freedom from recurrent severe MR was 91% and from moderate MR was 69%. Isolated anterior leaflet prolapse was the strongest predictor of recurrent moderate-to-severe MR. At the last follow-up, 69% of the 627 survivors were NYHA class I; 22% class II, and 9% class III. The authors concluded that in the absence of class IV symptoms and reduced EF, mitral valve repair for severe MR restored lifespan to normal, and recurrent moderate-to-severe MR or repeat surgery was unusual.


This single-center, single-surgeon, quarter century experience with mitral valve repair for MR due to mitral valve prolapse is of interest because randomized, controlled trials are unlikely to be done on the issues explored. In the population studied, almost all had severe MR and about half were NYHA class I-II. The excellent results presented support the guidelines recommendation that asymptomatic patients with severe MR should be repaired if feasible and appropriate based on mitral valve anatomy and comorbidities. Waiting for more marked symptoms was not supported by their experience as higher NYHA class was associated with higher mortality. The only caveat is the patient with isolated anterior leaflet prolapse, since this was the strongest predictor of recurrent MR after repair. Guidelines also recommend repair if LVEF falls below 60%, but waiting for this to happen in asymptomatic patients seems unwise based on their experience since EF was a strong predictor of mortality.

Recurrent significant MR occurred in about one-third of the survivors over 20 years. Surgery does not cure the degenerative process, and late recurrence of MR seemed to be related to further degeneration, whereas early recurrence seemed to be related to technical factors such as the inability to place a mitral annular ring. Also of interest was their experience with thromboembolism. There was a small but constant threat of thromboembolism throughout the observation period that was only associated independently with older age. Thromboembolism seemed to be unrelated to atrial fibrillation since all these patients were treated with anticoagulation. Freedom from thromboembolism was 86% at 20 years. They anticoagulated their patients for 3 months after surgery if there were no other indications for anticoagulation. After considering the data they posed the question of whether indefinite anticoagulation for all is justified. However, the risk of hemorrhage needs to be considered as there was an overall 22% serious hemorrhage rate among those on warfarin, resulting in eight deaths.