Intensive insulin therapy versus glycemic control in critically ill patients: a prospective controlled trial

Abstract

Intensive insulin therapy significantly reduced morbi-mortality in a population of critically ill surgical patients [1]. The results in medical ICU patients were less clear [2]. The objective of this study was to determine whether intensive insulin therapy is more safe and efficient than a glycemic control strategy without the use of high doses of insulin in a heterogeneous population of critically ill adult patients. Methods Included in the study were all adult patients admitted from 1 July 2004 to 31 December 2006 to a 20-bed multidisciplinary ICU of a general hospital and to an 11-bed trauma center ICU that had at least two blood glucose levels above 150 mg/dl from three measurements obtained in the first 12 hours after admission. Patients were randomly assigned to strict normalization of blood glucose levels (80120 mg/dl) with the use of insulin infusion or to glycemic control through glucose-free venous hydration, hypoglycidic nutritional formula and subcutaneous insulin if the blood glucose level was higher than 180 mg/dl in the measurements taken every 6 hours. Results Three hundred and thirty-seven patients were enrolled in the study. At admission the two groups were comparable regarding age, sex, APACHE III score, prevalence of diabetes mellitus and nosologies. Patients in group 1 (n = 168) received 52 (3574.5) units regular insulin per day, while group 2 (n = 169) received 2 (06.5) units insulin/day (P < 0.001). The median glucose level during treatment was 133.6 (119.7153.3) mg/dl in group 1 and was 144 (123174.2) mg/dl in group 2 (P = 0.003). ICU mortality was 22.6% in group 1 and 25.0% in group 2 (P = 0.6). There was no difference between the two groups regarding length of ICU stay, infectious complications and organ dysfunctions. Hypoglycemia occurred in 27 patients (16%) in group 1 and six patients (3.5%) in group 2 (P < 0.001). No permanent cognitive defects were recorded in patients with hypoglycemia. When a subgroup of patients who stayed in the ICU for more than 5 days was analyzed, although a small trend toward mortality reduction was noted (25.5% in group 1 and 30.3% in group 2; relative reduction of 16%), this difference did not reach statistical significance. Conclusion This study demonstrates the need to use protocols to control hyperglycemia that allows a less strict blood glucose control. With this approach it is possible to limit the hazards of hypoglycemia and, at the same time, to maintain the benefits of glycemic control. With such an approach it would be possible to extend the benefits of blood glucose control to ward patients.