Volume 14 Supplement 1

30th International Symposium on Intensive Care and Emergency Medicine

Open Access

Validation of HbA1c as a diagnostic marker for diabetes in the critically ill

  • I Gornik1,
  • A Vujaklija-Brajković1 and
  • V Gašparović1
Critical Care201014(Suppl 1):P581

https://doi.org/10.1186/cc8813

Published: 1 March 2010

Introduction

Diabetes mellitus (DM) is a highly prevalent disease: about one-quarter of patients admitted to hospital is diagnosed with DM, but there are also a large number of patients who are undiagnosed. An acutely ill patient with hyperglycaemia and no history of diabetes may have stress hyperglycaemia or a first manifestation of diabetes, but it is virtually impossible to establish diagnosis based on measurements of glucose. HbA1c is widely used as a marker of glycaemic control and has been proposed and used, but not yet officially accepted, as a diagnostic marker for diabetes. We have tested its usefulness for diagnosing diabetes in the intensive care setting.

Methods

We have included a cohort of adult patients admitted to a medical ICU. HbA1c was measured in all patients. Patients without hyperglycaemia during their illness and negative history of DM were used as negative controls; patients with known diagnosis of DM, impaired fasting glucose (IFG) or impaired glucose tolerance (IGT) were positive controls. All patients with negative history for DM and hyperglycaemia in ICU were called for measurement of fasting glucose and OGTT within 1 month after discharge.

Results

There were 618 patients in the cohort: 138 with no hyperglycaemia, 79 with a history of IFG or IGT, 108 with a history of diabetes; 293 patients had hyperglycaemia during the ICU stay and no information about DM in the history. When analysing only positive and negative controls, a cut-off value of 6% showed 95% sensitivity and 93% specificity for pre-diabetes (IFG and IGT), cut-off of 7% had 91% sensitivity and 96% specificity for DM. Similar results were obtained for the group of patients with hyperglycaemia and no history of impaired glucose metabolism: cut-off of 6% had 97% sensitivity and 93% specificity for pre-diabetes, 7% cut-off had 95% sensitivity and 98% specificity for DM.

Conclusions

HbA1c can be used for diagnosing diabetes and pre-diabetes in patients with critical care hyperglycaemia with high sensitivity and specificity. Since diabetes brings a burden of complications, impaired immune response, susceptibility to infections, and overall higher mortality, differentiating diabetics from patients without diabetes can be very valuable.

Authors’ Affiliations

(1)
University Hospital Centre Zagreb

Copyright

© BioMed Central Ltd. 2010

Advertisement