Hyperglycemia is common in critically ill patients, with approximately 90% of patients treated in an ICU developing blood glucose concentrations > 110 mg/dL (6.1 mmol/L). Landmark trials in Leuven, Belgium, suggested that targeting normoglycemia (a blood glucose concentration of 80-110 mg/dL [4.4-6.1 mmol/L]) reduced mortality and morbidity, but other investigators have not been able to replicate these findings. Recently, the international multicenter Normoglycemia in Intensive Care Evaluation-Survival Using Glucose Algorithm Regulation (NICE-SUGAR) study reported increased mortality with this approach, and recent meta-analyses do not support intensive glucose control for critically ill patients. Although the initial trials in Leuven produced enthusiasm and recommendations for intensive blood glucose control, the results of the NICE-SUGAR study have resulted in the more moderate recommendation to target a blood glucose concentration between 144 mg/dL and 180 mg/dL (8-10 mmol/L). As critical care practitioners pay greater attention to glycemic control, it has become clear that currently used point-of-care measuring systems are not accurate enough to target tight glucose control. Unresolved issues include whether increased blood glucose variability is inherently harmful and whether even moderate hypoglycemia can be tolerated in the quest for tighter blood glucose control. Future research must first address whether intensive glucose control can be delivered safely, and whether computerized decision support systems and newer technologies that allow accurate and continuous or near-continuous measurement of blood glucose can make this possible. Until such time, clinicians would be well advised to abide by the age-old adage to "first, do no harm."