Abstract 18060: Risk Stratification Following Mitral Valve Surgery for Chronic Ischemic Mitral Regurgitation: A Very Long-term Study

Background: Very few long-term data are available on patients undergoing mitral valve surgery for chronic ischemic mitral regurgitation (CIMR). Objective:to identify determinants of survival and adverse cardiovascular events, at very long-term outcome. Methods and Results: We reviewed complete left and right ventricular echocardiographic data, six-minute walking test (6-MWT) and BNP levels at pre, peri and follow-up, on 137 consecutive patients who underwent restrictive mitral annuloplasty (RMA) or mitral valve replacement (MVR) and CABG, for CIMR. Combined adverse cardiovascular events were defined as composite of death, heart failure, angina, myocardial infarction and re-hospitalization. Mean age was 67±0.7. Preoperative EF was 35±06%. Among 137 patients, 46% underwent RMA and 54% had MVR. Median follow- up was 7 years (range: 0.3-15.4). Early mortality was 7% (p=NS). In the RMA 42% of patients experienced MR recurrence. Overall survival at 5, 10 and 15 years were 84, 76 and 62% in RMA, and 87, 62 and 54% in MVR (p=0.65). At univariate analysis, preop B-blocker, 6-MWT, mean transmitral gradients and RV size were predictors of adverse events.The Cox Hazard multivariate analysis identified preoperative atrial fibrillation (p=0.005), preop BNP (p=0.025) as independent predictors of long-term mortality.Freedom from cardiovascular events at 5, 10 and 15 years were 90, 75 and 48% in RMA, and 90, 62 and 45% in MVR (p=0.57). . The Cox Hazard multivariate analysis identified preop B-blocker therapy (p=0.001), atrial fibrillation (p=0.01), postoperative mean transmitral gradients (p=0.047 ) and indexed effective orifice area (p=0.02) as independent predictors of adverse cardiovascular events. Conclusions: Our study confirms an high rate of true MR recurrence, at very long-term follow-up. Among collected variables, preoperative BNP and atrial fibrillation were independent predictors of survival, whereas B-blocker therapy, atrial fibrillation, postoperative mean transmitral gradients and indexed effective orifice area were independent predictors of adverse cardiovascular events. Type of surgery did not affect the very long-term outcome.