Re-repair of the failed mitral valve: insights into aetiology and surgical management†

OBJECTIVES: Mitral valve (MV) repair is the gold standard for treatment of degenerative mitral regurgitation. A variety of surgical techniques allow surgeons to achieve a high rate of MV repair even with MV diseases of other aetiologies. However, a certain number of repairs fail over time. The aim of this study was to review our single-centre experience of MV re-repair and analyse the mode of repair failure, re-repair safety and efficiency in relation to the initial aetiology. METHODS: Between 1997 and 2015, 91 patients underwent redo MV re-repair. The first MV repair was performed in our institution in 59% of cases. Follow-up information was available for 93% of our patients. The median follow-up was 56 months. RESULTS: The initial aetiology was degenerative disease in 40 (44%) patients, rheumatic disease in 25 (27.5%), endocarditis in 10 (11%), ischaemic in 6 (7%), severe mitral annulus calcification in 5 (5.5%), congenital disease in 4 (4%) and unknown in 1 (1%). The mean age was 58 ± 15 years. The median delay between the 1st and 2nd repair was 49 months with 6 early re-repairs. Re-repair was urgent or emergent in 19% of cases; indications for surgery were mitral regurgitation in 48%, stenosis in 19%, endocarditis in 19%, mitral disease in 11%, ring thrombosis in 2% and systolic anterior motion in 1%. The main mechanisms of failure included technical error (30%), progression of disease (35%), new disease (29%) and unknown (6%.) Re-repair was performed through a median sternotomy in 96% of cases, and 34% of patients had concomitant procedures. Eight (9%) postoperative deaths (4 of mitral annulus calcification, 2 of endocarditis, 1 of degenerative disease, 1 of ischaemia) and 5 (6%) early failures occurred (3 of rheumatic disease, 1 of degenerative disease, 1 of a congenital condition), requiring MV replacement in 4 and new repair in 1. Overall survival at 5 and 10 years was 76% and 57%, 83% and 49% in patients with degenerative diseases and 95% and 95% in patients with rheumatic disease. Overall freedom from reoperation at 5 and 10 years was 82% and 61%, 94% and 87% with degenerative disease and 60% and 45% with rheumatic disease. CONCLUSIONS: MV re-repair is feasible and has good mid-term results in patients with degenerative MV disease. Rheumatic MV disease is associated with a certain risk of failure over time; nevertheless, these patients show excellent survival after re-repair.