Evaluation of postoperative outcomes of valve reoperation: a retrospective study

Abstract

Abstract   OBJECTIVES The aim of this study was to compare the incidence of operative death and postoperative complications between primary and reoperation valve surgeries and to identify independent risk factors for these events among valve-reoperation patients. METHODS Between 2013 and 2015, 54 269 patients who underwent valve surgery were retrospectively analyzed using the Japan Cardiovascular Surgery Database. They were divided into the primary (group P; n = 49 833) and reoperation (group R; n = 4436) surgery groups. Among the reoperation patients, we conducted multivariable logistic regression analyses to identify risk factors for the incidences of operative mortality and postoperative complications. Then, we also conducted propensity score matched analyses to compare the incidences of these 2 outcomes for primary versus reoperation procedures separately for patients with and without infective endocarditis (IE). RESULTS Incidences of postoperative mortality (4.6% vs 9.1%; P < 0.001) and any complications (36.6% vs 41.4%; P < 0.001) were higher in the reoperation group. For patients undergoing reoperation, strong risk factors for operative mortality included urgency status, ejection fraction <30%, IE, dialysis, chronic kidney disease, New York Heart Association class 3/4, concomitant coronary artery bypass grafting and aorta procedure, tricuspid valve surgery only, multivalve surgery and age. In the propensity score matched cohort, the relative odds of operative mortality were 1.53 (95% confidence interval: 1.26–1.86, P < 0.001) among patients with IE and were 1.58 (95% confidence interval: 1.18–2.13, P < 0.002) among those without. CONCLUSIONS Outcomes for reoperation were significantly worse than those for primary surgery. At the primary operation, the risk of reoperation should be considered and when considering the indications for reoperation, the preoperative state, surgical timing and intervention method should be considered.