The impact of the blood glucose levels of non-diabetic critically ill patients on their clinical outcome.

Abstract

Stress hyperglycaemia is thought to result from a hormonal response (release of catecholamines, glucocorticoids, glucagon, etc.) following stress, sepsis or trauma. Although stress hyperglycaemia is a very common finding in critically ill populations, there are many non-diabetic critically ill patients who do not develop a hyperglycaemic stress response to trauma or acute illness. We suggest that the lack of a hyperglycaemic stress response during the acute phase of a critical illness may correlate significantly with the clinical outcome of these critically ill non-diabetic patients. This was a retrospective study of 700 non-diabetic critically ill patients admitted to the general intensive care unit (ICU) at Soroka Medical Center, Beer Sheva, Israel. We analyzed the clinical impact of the blood glucose levels of these patients measured during their first week of ICU hospitalization on their clinical outcome. Age, male gender, and the Acute Physiology and Chronic Health Evaluation (APACHE) score were found to be independent risk factors for new episodes of infection during the patients' stay in the ICU. Age and the APACHE and Sequential Organ Failure Assessment scores were found to be independent risk factors for intra-ICU mortality. In contrast, blood glucose analysis performed during the patients' stay in the ICU was not found to be an independent predictor for new infectious events or for mortality during the ICU stay. Our study did not demonstrate an association between blood glucose levels and clinical outcomes in non-diabetic critically ill patients.