The use of bilateral internal mammary artery (BIMA) grafting for myocardial revascularization has been demonstrated to provide long-term benefits compared to revascularization using single left internal mammary artery (SIMA) and venous conduits. However, it is still controversial whether the use of BIMA is associated with a higher hospital mortality and morbidity. The present study retrospectively evaluated the possible advantages related to the use of BIMA at 3-year follow-up and whether the presence of operative risk factors in patients with BIMA could limit the application of the procedure in myocardial revascularization. We compared two groups of 100 patients matched for preoperative clinical characteristics, who underwent myocardial revascularization on the left coronary system with BIMA (93 males and 7 females, mean age 59 +/- 4 years) or with SIMA and venous conduits (86 males and 14 females, mean age 63 +/- 6 years). Hospital mortality rate was 2% in both groups, the use of BIMA being not a significant risk factor for hospital mortality and morbidity. The mean follow-up was 36 +/- 6 months for the BIMA group and 40 +/- 10 months for the SIMA group. At 3 years, there was no significant differences in the actuarial freedom from cardiac death (96 +/- 2% for BIMA vs 94 +/- 2% for SIMA patients), myocardial infarction (98 +/- 2 vs 97 +/- 2%), angina (93 +/- 2 vs 91 +/- 2%), symptomatic heart failure (92 +/- 3 vs 92 +/- 2%), coronary angioplasty/reoperation (96 +/- 2 vs 97 +/- 2% ), and total cardiac events (80 +/- 4 vs 76 +/- 4%). BIMA grafting was not an independent predictor of late cardiac events. In 66 patients who underwent a late angiographic or echo-Doppler study, the patency rate was 100% for the left mammary artery, 94% for the right mammary artery and 69% for venous conduits. In conclusion, myocardial revascularization with BIMA in situ is associated with low hospital mortality and morbidity, good clinical outcome and excellent patency rate at 3 years, with apparently no significant differences when compared to the use of SIMA and venous conduits. The low hospital mortality and morbidity and the satisfactory medium-term results in our opinion justify a more extensive use of BIMA in myocardial revascularization.