OBJECTIVES: The number of coronary artery disease reoperations isincreasing. The aim of this paper is to identify risk factors and evaluatethe results of REDO coronary artery bypass grafting (CABG). MATERIAL:Between January 1984 and October 1994, 594 patients underwent REDO-CABG and3157 underwent primary-CABG. The mean age was 62 years with 84% men.Hypertension, hyperlipidemia, insulin dependent diabetes, smoking and renalinsufficiency were all more frequent in the REDO- group. A significantlyhigher number of patients undergoing REDO-CABG were in the CanadianCardiovascular Society (CCS) angina class 3 and 4, had instable angina, hadleft main stem stenosis of greater than 70% and 3-vessel disease comparedto those undergoing primary-CABG. The mean preoperative left ventricularfunction (LVEF) was 49.8 (REDO) vs. 58.2%, with a P value of less than0.001. RESULTS: The overall postoperative mortality rate forREDO-operations was 9.6 (57/594) vs. 2.8% for primary-CABG. Patients with areoperative interval of more than 1 year had an 8.9% mortality rate,compared to those reoperated less than 1 year after the initial CABG, wherethe mortality was 21% with a P value of less than 0.05. Postoperative lowcardiac output syndrome, intraaortic balloon pump support, prolongedventilatory support (> 24 h), hemorrhage and gastrointestinalcomplications were prominent features of the REDO-group (all P < 0.01).Urgent operation, CCS class 3 and 4, LVEF of less than 40%, generalizedarteriosclerotic disease and advanced age (> 80 years) were independentrisk factors for postoperative death in both groups. Preoperative renalinsufficiency, diabetes and short interval from primary-CABG were addedrisk factors in the REDO-group. The 5-years survival rate after REDO-CABGwas 89%, while the cardiac event-free survival rate was 79% and at 7 years84 and 62%, respectively. CONCLUSIONS: Reoperative CABG is effective, buthas an increased operative mortality and morbidity. The long-term resultsare encouraging. Unstable angina, poor preoperative left ventricularfunction, renal insufficiency, insulin dependant diabetes and an intervalshorter than 1 year of the initial operation were independent riskfactorsfor mortality.