Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. Which factors influence glycemic control in the intensive care unit? Jean-Charles Preiser a , Philippe Devos b and Rene ´ Chiolero c Introduction Before 2001, the hyperglycemia found in most critically ill patients was considered a component of the stress response [1  ]. Current understanding was completely changed by the publication of the first Leuven study in 2001 [2]. This investigation compared an intensive insu- lin regimen targeting a blood glucose level within the 80–110 mg/dl range, as compared with a ‘conventional’ management cohort in which blood glucose was treated only when above 200 mg/dl. Van den Berghe et al. [2] demonstrated a 4% decrease in the absolute mortality of critically ill patients randomized to intensive insulin therapy (IIT). These unexpectedly impressive results triggered a huge wave of enthusiasm. Recommendations to implement tight glucose control in intensive care units were rapidly issued by several healthcare agencies (Joint Commission on Accreditation of Healthcare Organiza- tion, the Institute for Healthcare Improvement and the Volunteer Hospital Organization). Simultaneously, several different teams tried to reproduce the results and to examine the underlying mechanisms of the findings of the Leuven team. Overall, the results of the Leuven study have not been reproduced, including in the medical intensive care unit of the same hospital. Of note, the follow-up studies have given rise to several controversies and raised important but as yet unan- swered questions for the physicians taking care of critically ill patients. In addition to glycemic control itself, the heterogeneity of several other aspects represents a major difficulty for the interpretation of the data reported in these prospective randomized con- trolled trials. As a consequence, the clinician is now left with several unanswered issues, such as what is the optimal value of blood glucose, what are the risks asso- ciated with hypoglycemia and what categories of patient might benefit from tight glucose control via IIT? The updated recommendations of the American Diabetes Association, the American Association of Clinical Endo- crinologists and of the French-speaking societies of intensive care reflect these uncertainties. Pathophysiological considerations It has long been recognized that critically ill patients often are hyperglycemic. For many years, this was attrib- uted to ‘stress’ and was believed to be a part of the host response to critical illness [1  ]. Thus, hyperglycemia was believed to be a marker of the severity of illness rather than a mediator of cell injury. The Leuven studies started with the hypothesis that hyperglycemia was not just a biomarker. These investigators postulated elevations in serum glucose contributed to the pathophysiology of critical illness. This proposal boosted the current field of investigation. The initial question might be rephrased as ‘What is the optimal blood glucose concentration in the a Department of General Intensive Care, University Hospital Centre of Liege, b Department of Intensive Care, Saint-Joseph Clinic, Liege, Belgium and c Department of Intensive Care, University Hospital Lausanne, Lausanne, Switzerland Correspondence to Jean-Charles Preiser, MD, PhD, Department of General Intensive Care, University Hospital Centre of Liege, Domaine universitaire du Sart-Tilman, 4000 Liege, Belgium Tel: +32 4 366 74 95; fax: +32 4 366 88 98; e-mail: Jean-Charles.Preiser@chu.ulg.ac.be Current Opinion in Clinical Nutrition and Metabolic Care 2010, 13:205–210 Purpose of review Intensive insulin therapy titrated to restore and maintain blood glucose between 80 and 110 mg/dl (4.4 – 6.1 mmol/l) was found to improve survival of critically ill patients in one pioneering proof-of-concept study performed in a surgical intensive care unit. The external validity of these findings was investigated. Recent findings Six independent prospective randomized controlled trials, involving 9877 patients in total, were unable to confirm the survival benefit reported in the pioneering trial. Several hypotheses were proposed to explain this discrepancy, including the case-mix, the features of the usual care, the quality of glucose control and the risks associated with hypoglycemia. Summary Before a better understanding and delineation of the conditions associated with and improved outcome by tight glycemic control, the choice of an intermediate glycemic target appears as a safe and effective solution. Keywords intensive insulin therapy, meta-analysis, prospective randomized controlled trial, stress hyperglycemia, tight glucose control Curr Opin Clin Nutr Metab Care 13:205–210 ß 2010 Wolters Kluwer Health | Lippincott Williams & Wilkins 1363-1950 1363-1950 ß 2010 Wolters Kluwer Health | Lippincott Williams & Wilkins DOI:10.1097/MCO.0b013e328335720b