The risk of mortality or significant morbidity is high among critically ill patients who are treated in the intensive care unit (ICU) for > 5 days. These patients are susceptible to sepsis, excessive inflammation, critical illness polyneuropathy, and multiple organ failure, the latter often being the cause of death. Most intensive care patients, even those who did not previously suffer from diabetes, are hyperglycemic, which is presumed to reflect an adaptive development of insulin resistance. In the K.U. Leuven study it was hypothesized that hyperglycemia is not a beneficial adaptation to severe illness but rather predisposes patients to many of the typical intensive care complications--prolonged intensive care dependence and death. The effects of intensive insulin therapy to maintain normoglycemia during critical illness were studied in a large group (N = 1548) of ventilated, surgical ICU patients. An algorithm was proposed for implementing this procedure. The randomly assigned intensive insulin therapy group received insulin infusion tailored to control blood glucose (BG) levels in the range 80-110 mg/dL, whereas the conventional treatment group received insulin only when glucose levels exceeded 200 mg/dL, and in that event were maintained in a target range of 180-200 mg/dL. Intensive insulin therapy induced a 43% reduction of intensive care mortality risk (P = 0.036 after correction for interim analyses) and a 34% reduction of hospital mortality (P = 0.005). A reduced risk of severe infections by 46% (P = 0.003) was associated with a 35% reduction in prolonged (> 10 d) requirement for antibiotic therapy (P < 0.001). In addition, excessive inflammation was prevented. Logistic regression analysis indicated that control of BG levels, rather than insulin administration itself, likely explains the observed clinical benefits. Use of insulin infusion to maintain normoglycemia using a titration algorithm, at least in populations similar to those in the Leuven study, improves outcome. Further data are needed to establish the applicability of this strategy to other patient groups, such as those in a medical ICU and in general hospital care.