Mitral valve repair versus replacement in patients with ischaemic mitral regurgitation and depressed ejection fraction: risk factors for early and mid-term mortality†
Antonio Lio,Antonio Miceli,Egidio Varone,Daniele Canarutto,Gioia Di Stefano,Francesca Della Pina,Daniyar Gilmanov,Michele Murzi,Marco Solinas,Mattia Glauber
DOI: https://doi.org/10.1093/icvts/ivu066
Abstract:Objectives: Mitral valve (MV) surgery for ischaemic mitral regurgitation (IMR) in patients with depressed left ventricular ejection fraction (LVEF) is associated with poor outcomes. The optimal surgical strategy for IMR in these patients remains controversial. The objective of this study was to compare the early mortality and mid-term survival of MV repair versus MV replacement in patients with IMR and depressed LVEF undergoing coronary artery bypass grafting (CABG). Methods: A retrospective, observational, cohort study was undertaken of prospectively collected data on 126 consecutive CABG patients with IMR and LVEF <40% undergoing either MV repair (n = 98, 78%) or MV replacement (n = 28, 22%) between July 2002 and February 2011. Results: The overall mortality rate was 7.9% (n = 10). MV replacement was associated with a 4-fold increase in the risk of death compared with MV repair [17.9%, n = 5 vs 5.1%, n = 5; odds ratio (OR) 4.04, 95% confidence interval (CI) 1.08-15.1, P = 0.04]. However, after adjusting for preoperative risk factors, the type of surgical procedure was not an independent risk factor for early mortality (OR 0.1, 95% CI 0.01-31, P = 0.7). Multivariable analysis showed that preoperative LVEF (OR 0.8, 95% CI 0.6-0.9, P = 0.018), preoperative B-type natriuretic peptide (BNP) levels (OR 1.01, 95% CI 1-1.02, P = 0.025), preoperative left ventricle end-systolic diameter (OR 0.8, 95% CI 0.7-1.0, P = 0.05) and preoperative left atrial diameter (OR 1.3, 95% CI 1.0-1.6, P = 0.015) were independent risk factors of early mortality. At the median follow-up of 45 months (interquartile range 20-68 months), the mid-term survival rate was 74% in the MV repair group and 70% in the MV replacement group (P = 0.08). At follow-up, predictors of worse survival were BNP levels [hazard ratio (HR) 1.0, 95% CI 1.0-1.01, P = 0.047], preoperative renal failure (HR 4.6, 95% CI 1.1-20.3, P = 0.039) and preoperative atrial fibrillation (HR 3.3, 95% CI 1.1-10, P = 0.032). Conclusions: MV repair in CABG patients with IMR and depressed LVEF is not superior to MV replacement with regard to operative early mortality and mid-term survival.