Meta-Analysis of Mitral Valve Repair Versus Replacement for Rheumatic Mitral Valve Disease

Background Rheumatic heart disease remains one of the leading causes of heart valve disease worldwide despite being a preventable condition. Mitral valve repair is superior to replacement in severe degenerative mitral valve disease, however its role in rheumatic valve disease remains controversial. This meta-analysis compared mitral valve repair and replacement in rheumatic heart disease. Methods Medline, EMBASE, Cochrane and Scopus were searched from January 1980 to June 2016 for original studies reporting outcomes of both mitral valve repair and replacement in rheumatic heart disease in adults, children or both. Two (2) authors independently assessed studies for inclusion, followed by data extraction and analysis. Results The search yielded 930 articles, with 98 full-texts reviewed after initial screening and 13 studies subsequently included for analysis, totalling 2,410 mitral valve repairs and 3,598 replacements. Pooled rates and odds ratio (95% confidence interval) for operative mortality of repair versus replacement was 3.2% versus 4.3%, 0.68 (0.50-0.92; p4.01). Pooled odds ratios (95% confidence interval) were for long-term mortality 0.41 (0.30-0.56; p<0.001); reoperation 3.02 (1.72-5.31; p<0.001); and bleeding 0.26 (0.11-0.63; p=0.003). There was a trend towards lower thrombo-embolism 0.42 (0.17-1.03; p=0.06), and no significant difference in endocarditis (p=0.76), during follow-up. Conclusion Mitral valve repair is associated with reduction in operative and long-term mortality and bleeding, so is recommended in rheumatic mitral valve disease where feasible, but it does entail a higher rate of reoperation during follow-up.

Automated summary

This study compared mitral valve repair and replacement in patients with rheumatic heart disease and found that mitral valve repair is associated with a reduction in operative and long-term mortality and bleeding, but it does entail a higher rate of reoperation during follow-up.