Comment on “Mean glucose level is not an independent risk factor for mortality in mixed ICU patients” by Ligtenberg et al.

Excerpt

The recent study by Ligtenberg et al. [1] examined at diabetic and nondiabetic ICU patients with various blood glucose measurements and found that the Acute Physiology and Chronic Health Evaluation (APACHE) II score was the most important factor in predicting mortality. Recent data suggest that mortality in patients with known diabetes is less than that in patients with injury-induced hyperglycemia. In the Van den Berghe et al. study [2] all medical ICU patients received intravenous insulin administration or conventional therapy regardless of admission glucose. Interestingly, patients given IVIA without known diabetes had a trend to lower mortality (40.9% vs. 36.8%). Intravenous insulin administration to medical and surgical ICU patients with two measurements of blood glucose above 199 mg/dl reduced overall (20.9% vs. 14.8%) and septic shock mortality (60% vs. 33%) [3]. It appears that patients with injury-induced hyperglycemia have increased mortality [4, 5, 6] despite the fact that their injury scores were similar to those of nonhyperglycemic patients [6], and serum albumin levels (an excellent marker of severity of injury) were similar between groups (personal Communication) [5]. Many of these patients with injury-induced hyperglycemia may perhaps be patients with prediabetes who have recently undergone a major metabolic injury (see Table 1). …