Reply to the comment by Dr. Tayek

Excerpt

We appreciate the comments of Dr. Tayek on our study [1]. We would like to emphasize that the goal of the study was to evaluate the relationship between blood glucose level and mortality in a mixed ICU population, and of course not to confirm the reliability of the APACHE II score. In essence, Dr. Tayek asks us to re-evaluate our data, distinguishing known diabetics from nondiabetic. Since the prevalence of hyperglycemia in critically ill patients, using stringent criteria, approaches 100%, it appears to be somewhat artificial to divide the study population into patients with and those without a history of diabetes. We are not sure about the relationship between diabetes and stress-induced hyperglycemia. The pathophysiological mechanisms may share some components, but there are also major differences. Furthermore, it is known that one-half of patients with type 2 diabetes remain undiagnosed [2]. On the other side, hyperglycemic diabetic patients indeed seem to have a better prognosis than “new” hyperglycemic ICU patients [3, 4]. One could speculate about the reasons for this unexpected luck of having diabetes. It may be that insulin is started earlier in known diabetic ICU patients than in other critically ill patients. Diabetic patients may need more units of insulin, because they are more insulin resistant, which may be beneficial since it has not been yet confirmed whether it is the tight glucose regulation per se or the administration of insulin that causes the effects on morbidity and mortality [5]. However, being a diabetic also has drawbacks: the prevalence of known diabetes in our ICU population (and in the recent study by Van den Berghe et al. [6]) is approx. 17% while in the age-matched general population the prevalence is only 3.6% [2]. Apparently diabetics are more prone to ICU admission. …