To determine the potential role of emergency surgical revascularization as treatment of acute myocardial infarction (AMI), results in 79 patients undergoing operation for myocardial revascularization during AMI from January 1986 to January 1991 were reviewed. Clinical characteristics for inclusion in the study were: 1) emergency operation; 2) persistent angina not controlled by medical therapy; 3) fixed ST segment elevation until surgical procedure, independently from magnitude of enzymatic levels. The 79 patients were divided in 2 groups: 27 with AMI or evolving AMI (Group 1); 52 with AMI due to complications during PTCA (Group 2). Twenty-eight patients had extremely severe clinical conditions. Mean interval between the beginning of AMI and operation was 4.2 +/- 6.7 hours, with a statistically significant difference between Group 1 (8.7 +/- 10.0) and Group 2 (1.9 +/- 1.0). One hundred ninety-two grafts were performed (2.4 +/- 1.1 grafts/patient). Overall hospital mortality was 10.1% (CL 6.7-13.3) (8 deaths) with a difference between Group 1 [18.5% (CL 10.7-25.3)] and Group 2 [5.8% (CL 4.7-6.6)] (p = 0.074). The incidence of perioperative myocardial infarction was 30.4% (CL 24.9-35.1) for that one in the area of ischemic muscle and 2.6% (CL 0.8-4.1) for infarction in remote muscle. Multivariate analysis for the entire series (79 patients) identified as independent predictors of increased in-hospital mortality: preoperative cardiogenic shock (p = 1.000E-4) and hyperlipidemia (p = 0.008). In Group 1 multivariate analysis identified as independent predictors of increased in-hospital mortality: the attempt of revascularization by PTCA and hyperlipidemia; in Group 2: preoperative need of mechanical ventilatory support.(ABSTRACT TRUNCATED AT 250 WORDS)