This review describes the current status and the clinical data related to the effects of tight glucose control in critically ill patients. In contrast to decreases in mortality and morbidity reported in one study in which insulin rate was titrated to keep blood glucose between 80 and 110 mg/dL, the benefits were not confirmed in multicenter prospective studies. Retrospective data found an association between a mean blood glucose level below 140 to 150 mg/dL and improved outcome. Currently unanswered issues include the optimal target for blood glucose, effects of blood glucose variability, hazards of hypoglycemia, and potential influence of the underlying disorder on the effects of tight glucose control. Therefore, recommendations regarding practical aspects of tight glucose control by intensive insulin therapy cannot be supported by currently available data. An intermediate target level for blood glucose of 140 to 180 mg/dL seems associated with the lowest risk-to-benefit ratio.