Objective: The study objective was to identify characteristics differentiating patients undergoing valve replacement versus valve repair for degenerative mitral valve disease and to use this information to compare survival and reoperation after each procedure.
Methods: From 1985 to 2005, 3286 patients underwent isolated primary operation for degenerative mitral valve disease. Valve repair was performed in 3051 patients (93%), and valve replacement was performed in 235 patients (7.2%). A propensity model and score developed for fair comparison of outcomes yielded 195 matched pairs.
Results: Patients undergoing replacement were older (70 +/- 12 years vs 57 +/- 13 years) and had more complex valvar pathology, symptoms, and left ventricular dysfunction. Thus, the characteristics of the propensity-matched patients undergoing repair more resembled those of the patients undergoing replacement (older, complex valvar pathology) than patients undergoing typical repair. Eight patients died in the hospital (0.26%) after repair and 5 patients (2.1%) died after replacement (P = .001). Unadjusted survival at 5, 10, and 15 years was 95%, 87%, and 68% after repair and 80%, 60%, and 44% after replacement, respectively (P < .0001); however, among propensity-matched patients, survival was similar (P = .8): 86% versus 83% at 5 years, 63% versus 62% at 10 years, and 43% versus 48% at 15 years. Freedom from reoperation among propensity-matched patients was 94% at 5 and 10 years after repair and 95% and 92% at 5 and 10 years after replacement, respectively (P = .6).
Conclusion: It is reasonable to perform valve repair in elderly patients with complex degenerative mitral valve pathology because it can eliminate the need for anticoagulation and risk of prosthesis-related complications. However, when valve pathology is so complex that repair is infeasible, this study demonstrates that valve replacement does not diminish long-term outcomes.