Abstract
OBJECTIVE: This study compares early and long-term outcomes following mitral valve (MV) repair and replacement in patients with mitral regurgitation (MR) and reduced left ventricular ejection fraction (LVEF). METHODS: Patients with primary or secondary MR and LVEF <50% who underwent MV replacement or repair (with/without atrial septal defect closure and/or atrial fibrillation ablation) between 2005 and 2017 at our center were retrospectively analyzed using unadjusted and propensity score matching techniques (42 pairs). RESULTS: A total of 356 patients with either primary (n = 162 [45.5%]) or secondary MR (n = 194 [54.5%]) and LVEF <50% underwent MV repair (n = 293 [82.3%]) or replacement (n = 63 [17.7%]) during the study period. In-hospital mortality was 0.3% (repair) and 1.6% (replacement) in the unmatched cohort (P = .32); there were no in-hospital deaths after matching. Estimated survival was 72.8% (repair) versus 50.1% (replacement) at 8 years in the unmatched (P < .001), and 64.3% (repair) versus 50.7% (replacement) in the matched groups (P = .028). Eight-year cumulative incidence of reoperation was 7.0% and 11.6% in unmatched (P = .28), and 9.9% and 12.7% in matched (P = .69) repair and replacement groups, respectively. Markedly reduced LVEF (<40%) was among the independent predictors of long-term mortality (hazard ratio, 1.7; 95% CI, 1.2-2.4; P = .002). In secondary MR, MV repair showed an 8-year survival benefit over replacement (65.1% vs 44.6%; P = .002), with no difference in reoperation rate (11.6% [repair] vs 17.0% [replacement]; P = .11). CONCLUSIONS: MV repair performed in primary or secondary MR and reduced LVEF provides superior long-term results compared with replacement. Severe LV dysfunction is a significant predictor of reduced survival following MV surgery.