We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Severity of Illness Is Not a Major Determinant of Hyperglycemia in Critically Ill Patients.
- Authors
Falciglia, Mercedes; Freyberg, Ron W.; Almenoff, Peter L.; D'Alessio, David A.; Render, Marta L.
- Abstract
Hyperglycemia has been identified as a common abnormality in critically ill individuals and a risk factor for mortality. Recent studies have shown that inpatient hyperglycemia occurs frequently in individuals without a prior diagnosis of diabetes, and that these patients have an even greater risk of death from hyperglycemia than individuals with known diabetes. The term "stress hyperglycemia" has been used to describe a transient and reversible glycemic excursion in hospitalized patients related to the illness response, presumably cytokines and counter-regulatory hormones like cortisol and catecholamines. However, there is very little direct evidence that hyperglycemia among hospitalized patients not known to have diabetes is solely due to factors associated with acute stress. To address this question We evaluated the correlation between mean glucose and severity of illness in a large and heterogeneous cohort of individuals admitted to intensive care units (ICU) from 2002 through 2005. A severity of illness measure was calculated for each patient from a validated ICU logistic regression model that uses admission diagnosis, co-morbid disease, laboratory variables, and age. Mean glucose was determined for each patient using all ICU glucose measures except the admission value. The association between severity of illness score and mean glucose was determined using a linear regression model with mean glucose as the dependent variable. We included 181,073 first admissions to medical, cardiac and surgical ICUs at 118 geographically diverse Veterans Affairs hospitals. For the entire cohort, R-square was 0.007, indicating that the contribution of severity of illness to the variance in mean glucose was only 0.7%. R-square was 0.003 and 0.035 for patients with and without a diagnosis of diabetes respectively. This finding challenges the idea that hyperglycemia in hospitalized patients is explained solely by the stress related to illness. Alternatively, it is possible that. the hyperglycemia observed during hospitalization is a reflection of underlying abnormalities in glucose metabolism. Future studies are necessary to elucidate these potential mechanisms.
- Publication
Diabetes, 2007, Vol 56, pA177
- ISSN
0012-1797
- Publication type
Academic Journal