Hypoglycaemia and predisposing factors among clinical subgroups treated with intensive insulin therapy.

Abstract

In previous studies, conflicting intensive insulin therapy (IIT) results have been observed, whereby IIT-related mortality seems to be lower in specific clinical subgroups. The aim of this study was to assess differences in glycaemic control, the risk of critical hypoglycaemia (≤ 2.2 mmol/l), the associated predisposing factors, and the in-hospital mortality in different clinical subgroups treated with IIT. Prospective, observational study in a university-affiliated intensive care unit (ICU) conducted from 2004 to 2005. All patients (n = 1667) belonging to one of the six most common surgical intervention groups (cardiac, neuro, abdominal, vascular, orthopaedic, and spinal surgeries) and medical patients were included. IIT was performed with a target blood glucose level of 4.4-7.8 mmol/l. Different indices were analysed to evaluate glucose control and glycaemic variability. The rate of critical hypoglycaemia was significantly different within the different clinical subgroups and varied from 0.8% to 4.5%. Similar results were obtained for hyperglycaemia. Multivariable analyses for the predisposing factors of critical hypoglycaemia showed a heterogeneous distribution pattern among the different clinical subgroups. Similar results were obtained for the risk factors of in-hospital mortality. The risk of critical hypoglycaemia and the associated predisposing factors depended on the clinical subgroup involved. Critical hypoglycaemia is a potential threat for our patients, and the high risk of critical hypoglycaemia in some clinical subgroups appeared to reverse the benefits of IIT. As a result, it is crucial that the different subgroups involved in a study are defined to further interpret the potential benefits of IIT and the risk of critical hypoglycaemia.