Two randomised controlled trials have shown that maintenance of blood glucose levels below 110 mg/dl with intensive insulin therapy reduces mortality and morbidity of surgical and medical critically ill patients. An absolute reduction in the risk of death of 3-4 % is expected in intention-to-treat analysis, but the survival benefit increases when treatment is continued for at least a few days. Future studies set up to confirm the survival benefit and assign it as statistically significant in an intention-to-treat medical patient population should be adequately powered with inclusion of at least 5000 patients. For the observed benefits of intensive insulin therapy strict maintenance of normoglycaemia is primordial, whereas glycaemia- independent actions of insulin have minor, organ-specific impact. Pathophysiological mechanisms underlying the clinical effects are currently being unravelled further and might help to find new strategies for further improving outcome. Implementation of a strict glycemic control protocol in the intensive care unit is feasible and cost-effective, but asks for careful consideration of some practical aspects, such as prevention of hypoglycaemia, training of nurses and selection of accurate blood glucose measurement tools. Continuous blood glucose monitoring devices and closed-loop systems are under development and might be of great benefit to overcome these issues.