Re-repair vs. Replacement for Failed Mitral Valve Repair: A Systemic Review and Meta-Analysis

Objective The objective of this study was to compare outcomes of re-repair with those of mitral valve replacement (MVR) for failed initial mitral valve repair (MVr). Methods We searched the Pubmed, Embase, and Cochrane Library databases for studies that compared mitral valve re-repair with MVR for the treatment of failed initial MVr. Data were extracted by two independent investigators and subjected to a meta-analysis. Odds ratio (OR), risk ratio (RR), hazard ratio (HR), ratio difference (RD), mean difference (MD), and 95% confidence interval (CI) were calculated with the Mantel-Haenszel and inverse-variance methods for mode of repair failure, perioperative outcomes, and follow-up outcomes. Results Eight retrospective cohort studies were included, with a total of 938 patients, and mean/median follow-up ranged from 1.8 to 8.9 years. Pooled incidence of technical failure was 41% (RD: 0.41; 95% CI: 0.32 to 0.5; P = 0.00; I 2 = 86%; 6 studies, 846 patients). Pooled mitral valve re-repair rate was 36% (RD: 0.36; 95% CI: 0.26–0.46; P = 0; I 2 = 91%; 8 studies, 938 patients). Pooled data showed significantly lower perioperative mortality (RR: 0.22; 95% CI: 07 to 0.66; I 2 = 0%; P = 0.008; 6 studies, 824 patients) and significantly lower long-term mortality (HR:0.42; 95% CI: 0.3 to 0.58; I 2 = 0%; P = 0; 7 studies, 903 patients) in the re-repair group compared with MVR. Conclusions Mitral valve re-repair was associated with better immediate and sustained outcomes for failed MVr and should be recommended if technically feasible.