RESUMEN
The aim of this randomised controlled study was to compare continuous subcutaneous insulin infusion using an insulin pump with the traditional continuous intravenous infusion method for tight glycaemic control. Sixty patients admitted to our University Hospital medical intensive care unit with an initial blood glucose level over 6.1 mmol/l, were enrolled and randomised into two treatment groups: the subcutaneous insulin group received continuous subcutaneous insulin infusion and the intravenous group received insulin by traditional intravenous infusion with infusers. Three patients died in the first 24 hours and were excluded from the final analysis. Insulin therapy was administered to both groups according to the previously designed and used protocol in the department. The target glucose level was 4.4 to 6.1 mmol/l. There was no significant difference in mortality between the groups. However mean blood glucose level was found to be lower (6.56+/-0.82 mmol/l vs. 7.85+/-1.6 mmol/l, P=0.00055) in the subcutaneous insulin group. According to Vogelzang's hyperglycaemic index, better glycaemic control was achieved in the subcutaneous insulin group while there was no significant difference in terms of hypoglycaemic events. Daily insulin bolus and infusion requirements were also significantly lower in the subcutaneous insulin group. Despite the small number of patients involved in this study in a medical intensive care unit, strict blood glucose control using a subcutaneous insulin pump was achieved more efficiently than the traditional intravenous infusion method without increasing hypoglycaemic events.