Reoperative coronary artery bypass grafting: analysis of early and late outcomes.

BACKGROUND The purpose of this study was to evaluate early and late results of reoperative coronary artery bypass grafting compared with those of first coronary artery bypass grafting. METHODS From November 21, 1994, to December 31, 2001, 4,381 patients underwent isolated coronary revascularization: among these patients, 274 (6.3%) underwent a redo. Applying the propensity score, 239 redo patients (group R) were matched with 239 who underwent the first revascularization (group F). RESULTS Early mortality was 2.1% (group F) and 4.2% (group R), not significantly different. Group R showed significantly higher creatine kinase myocardial band release, length of intensive care unit stay, and incidence of incomplete myocardial revascularization than group F. In group R, off-pump patients showed higher incidence of incomplete revascularization. Redo was a risk factor for abnormal (>19 IU/L) creatine kinase myocardial band release (odds ratio, 1.7; p = 0.0066) and incomplete myocardial revascularization (odds ratio, 2.4; p = 0.0060). Five-year clinical outcome was significantly worse in group R, except for freedom from redo or percutaneous transluminal coronary angioplasty. Redo was an independent variable for lower freedom from death of any cause, cardiac death, acute myocardial infarction, cardiac events, and any event. Patients with higher creatine kinase myocardial band release or incomplete myocardial revascularization showed lower freedom from cardiac-related events. Incidence of incomplete myocardial revascularization and creatine kinase myocardial band release were significantly higher in group R by both univariate and multivariate analysis. This could explain the worse late outcome of redo patients. CONCLUSIONS Complete revascularization without damaging the heart, whichever technique is used, is the target of redo surgery, to achieve the same quality of results obtained in the first operation.