In Reply: Since the 2001 study by van den Berghe et al,1 practices in the intensive care unit have been altered toward avoiding persistent hyperglycemia. Since then, our study and others2 have provided additional evidence suggesting that normalization of blood glucose levels should not be a target, at least with tools available at the bedside. As pointed out by Dr Castellanos and colleagues, hypoglycemia may be the price of blood glucose control. In the COIITSS trial, a total of 62 patients experienced a total of 108 episodes of hypoglycemia, ranging from 1 to 10 per patient. Of these patients, 51 were hypoglycemic on a single day, 10 had hypoglycemic episodes on 2 distinct days, and 1 patient had this complication occurring over 4 days. These findings suggest that hypoglycemia may be predominantly a complication of the initiation of intensive insulin therapy. In the COIITSS trial, this complication was not associated with an increased risk of death.
Annane D, Lejeune J, Chevret S. Tight Glycemic Control in Critically Ill Patients—Reply. JAMA. 2010;303(17):1694-1695. doi:10.1001/jama.2010.516