IntroductionPatients admitted to the intensive care unit often develop hyperglycaemia, but the underlying mechanisms have not been fully described. The incretin effect is reduced in patients with type 2 diabetes. Type 2 diabetes and critical illness have phenotypical similarities, such as hyperglycaemia, insulin resistance and systemic inflammation. Previous studies have shown beneficial effects of exogenous glucagon-like peptide (GLP)-1 on glycaemia in critically ill patients, a phenomenon also seen in patients with type 2 diabetes. In this study, we hypothesised that the incretin effect, which is mediated by the incretin hormones GLP-1 and glucose-dependent insulinotropic peptide (GIP), is impaired in critically ill patients.MethodsThe incretin effect (i.e., the relative difference between the insulin response to oral and intravenous glucose administration) was investigated in a cross-sectional case–control study. Eight critically ill patients without diabetes admitted to a mixed intensive care unit and eight healthy control subjects without diabetes, matched at group level by age, sex and body mass index, were included in the study. All subjects underwent an oral glucose tolerance test (OGTT) followed by an intravenous glucose infusion (IVGI) on the next day to mimic the blood glucose profile from the OGTT. Blood glucose, serum insulin, serum C-peptide and plasma levels of GLP-1, GIP, glucagon and proinflammatory cytokines were measured intermittently. The incretin effect was calculated as the increase in insulin secretion during oral versus intravenous glucose administration in six patients. The groups were compared using either Student’s t test or a mixed model of repeated measurements.ResultsBlood glucose levels were matched between the OGTT and the IVGI in both groups. Compared with control subjects, proinflammatory cytokines, tumour necrosis factor α and interleukin 6, were higher in patients than in control subjects. The endogenous response of GIP and glucagon, but not GLP-1, to the OGTT was greater in patients. The insulin response to the OGTT did not differ between groups, whereas the insulin response to the IVGI was higher in patients. Consequently, the calculated incretin effect was lower in patients (23 vs. 57 %, p = 0.003).ConclusionsIn critically ill patients, the incretin effect was reduced. This resembles previous findings in patients with type 2 diabetes.Trial registrationClinicalTrials.gov identifier: NCT01347801. Registered on 2 May 2011.Electronic supplementary materialThe online version of this article (doi:10.1186/s13054-015-1118-z) contains supplementary material, which is available to authorized users.