This study was planned to assess the relationship of perioperative glycemic control to the subsequent risk of infectious complications and to compare early clinical outcomes of coronary artery bypass surgery in diabetics with non-diabetics in a single center. A total of 1090 adults who underwent coronary artery surgery in a five year period were included in a retrospective cohort study based on available chart review. Of 1090 patients, 400 had type II diabetes mellitus. Intraoperative and postoperative blood glucose levels in diabetic group were manipulated by means of a continuous insulin infusion. Data of pre- and postoperative blood glucose levels were evaluated with respect to postoperative infection risk for diabetics. Risks of early mortality, cerebrovascular accident, and postoperative infection in diabetic patients were compared with the nondiabetic group. High preoperative mean glucose levels were the main risk factor for the development of postoperative infection (p = 0.012 and p = 0.028 for the mean glucose levels 1 and 2 days before operation, respectively). For diabetic group, of 400 patients 20 (5%) were diagnosed to have postoperative infection (superficial sternal wound in 3 (0.75%), donor site infection in 4 (1%), mediastinitis in 5 (1.25%), urinary tract infection in 6 (1.5%), and lung infection in 2 (0.5%) patients). The diabetic group had significantly higher prevalence of mediastinitis, donor site infection, urinary tract infection and total infection (p values were 0.048, 0.013, 0.009, and 0.044, respectively). Early mortality was higher among diabetics than in nondiabetics (1.73% vs 3%, p = 0.048) but the risk of cerebrovascular accident in diabetics was not greater than in nondiabetics in early period. In patients with diabetes who undergo coronary artery bypass surgery, preoperative hyperglycemia is an independent predictor of short-term infectious complications and total length of stay in hospital.