OBJECTIVES

Although mitral valve (MV) repair is known to be superior to replacement in overall clinical outcomes, the appropriateness of valve repair for rheumatic MV disease remains controversial because of the risks of recurrent mitral dysfunction and the need for re-operation.

METHODS

From 1997 to 2007, 540 patients underwent either isolated MV repair (n=122) or replacement with a mechanical prosthesis (n=418) in treatment of rheumatic MV disease. Survival and morbidity were evaluated using Kaplan-Meier analysis and Cox regression, including propensity score analysis.

RESULTS

Follow-up was complete in 96.1% of patients (mean, 71.8+/-39.1 months). Patients undergoing repair were younger; more likely to have predominant mitral regurgitation; and less likely to show atrial fibrillation (AF), significant tricuspid regurgitation or pulmonary hypertension, than those undergoing replacement. The 10-year freedom from cardiac death rate was 92.0+/-4.2% following repair and 86.8+/-2.3% following replacement (P=0.042). After adjustment for baseline differences, repair and replacement were found to be similar in terms of cardiac survival (P=0.25), re-operation (P=0.68) and thrombo-embolic complication (P=0.20) rates. Replacement patients had more anticoagulation therapy-related complications (P=0.030). Independent factors positively associated with combined cardiac death and major morbidities included older patient age (P=0.010), uncorrected AF (P=0.015) and the presence of significant tricuspid regurgitation (P=0.012) or coronary disease (P=0.043). The influence of the type of MV surgery was statistically marginal (P=0.093).

CONCLUSIONS

When performed for selected patients, MV repair had excellent durability comparable to mechanical valve replacement in rheumatic disease. Both MV repair and replacement had comparable long-term clinical results; therefore, repair surgery seems to be more beneficial by avoiding troublesome life-long anticoagulation and risks of bleeding.