Mid-term echocardiographic results with different rings following restrictive mitral annuloplasty for ischaemic cardiomiopathy

BACKGROUND: Despite restrictive mitral annuloplasty (RMA) being considered effective for chronic ischaemic mitral regurgitation (CIMR), few data exist on mid-term echocardiographic results with different prosthetic rings. Therefore, comparative echocardiographic analysis has been performed. METHODS: Sixty-four consecutive coronary artery bypass graft surgery (CABG) + RMA (downsizing by two-ring sizes; median size: 26 mm) for CIMR with a follow-up of at least 6 months were prospectively followed-up with serial echocardiograms (preoperative, discharge, 6 months, follow-up ending). Hospital mortality, follow-up clinical and echocardiographic results were analysed and compared between three groups (group A: semi-rigid band, 17 patients; group B: complete symmetric semi-rigid, 22 patients; group C: complete asymmetric semi-rigid, 25 patients). RESULTS: Hospital mortality was 6.3%; 22.8 +/- 14.7 standard deviation (SD) months (range: 6-55) survival was 96.5 +/- 2.5%; freedom from re-intervention was 94.2 +/- 4.2%, from re-revascularisation 87.5 +/- 11.7%, from > or = grade-2 mitral regurgitation 58.2 +/- 9.8% and from heart failure (CHF) 71.6 +/- 10.5%. Recurrent (> or = grade-2) CIMR resulted in lower freedom-from-CHF (p = 0.0001), worsened New York Heart Association (NYHA) classification (p = 0.0001) and absence of reverse remodelling of the left ventricular end-diastolic diameter (LVEDD; p = 0.004), systolic diameter (LVESD; p = 0.014), indexed mass (LVMi; p = 0.005) and coaptation depth (p = 0.0001). Group A showed significant worse freedom from CHF (group A: 42.8 +/- 19.5% vs group B: 88.9 +/- 10.5% vs group C: 92.3 +/- 7.5%; p = 0.049) and from recurrent CIMR (17.4 +/- 13.8% vs 82.1 +/- 11.7% vs 94.1 +/- 5.7%, respectively; p = 0.0001). Complete rings decreased the hazard of recurrent CIMR (Physio = 0.141; Adams = 0.089). Higher NYHA during follow-up was found in group A (p = 0.002 for group B and p = 0.001 for group C) with a progressive reduction of trans-mitral mean gradient (p = 0.001), and a lower degree of reverse remodelling of LVEDD (p = 0.009 and p = 0.010) and coaptation depth (p = 0.040 and p = 0.002). CONCLUSIONS: Recurrent CIMR correlates with absent ventricular reverse remodelling. Despite a higher trans-mitral gradient, complete rings achieve better results in the treatment of CIMR.