Since 1987, the Portland Diabetic Project has enrolled >5000 patients with diabetes mellitus (DM)—including patients with noninsulin-dependent DM and patients with insulin-dependent DM—in a prospective nonrandomized study of the effects of perioperative insulin administration on morbidity and mortality after open-heart surgical procedures. In the early stages of the study, patients received SC insulin. In 1991, we instituted a continuous intravenous insulin (CII) protocol, which has been refined several times since its introduction. Our principal measure of glycemic control is the mean blood glucose value obtained by measuring glucose every 0.5 to 2 hours during the initial 3-day postoperative period (3-BG). There is a strong relationship between postoperative mortality and 3-BG value, even after adjusting for other clinical variables that influence mortality. The postoperative mortality rate was significantly lower among patients who received CII rather than SC insulin after coronary artery bypass grafting (CABG) or cardiac surgery of any type. We have also found that 3-BG is strongly associated with the risk of postoperative deep sternal wound infection (DSWI), and that treatment with CII significantly reduces the risk of infection. Our CII glucose protocol is slightly more expensive to administer than conventional SC insulin. However, when cost savings associated with a reduced rate of serious infection and reduced length of hospital stay are considered, CII results in total savings of >$4500 per patient. Widespread use of our CII protocol to manage perioperative hyperglycemia in patients with diabetes who undergo open-heart surgical procedures would result in substantial reductions in morbidity, mortality, and overall costs to the health care system.