High blood glucose linked to increased mortality in intensive care unit

Hyperglycemia increased the risk of death from 1.6 to five times the expected rate for those admitted to ICU for myocardial infarction.

Issue: July 2006

Hyperglycemia has been found associated with increased mortality in patients in intensive care units, with the rates the highest in patients without diagnosed diabetes and in those admitted with cardiovacular disorders, such as myocardial infarction, unstable angina and stroke, according to a study presented at the American Diabetes Association 66th Annual Scientific Sessions in Washington.

“Overall, there was a significant relationship between mean hyperglycemia and adjusted mortality for every level of hyperglycemia,” said Mercedes Falciglia, MD, assistant professor of medicine, division of endocrinology at the University of Cincinnati College of Medicine, Ohio.

Researchers theorized that the magnitude of mortality risk varied with the disease for which patients were admitted. High glucose levels were not found linked to mortality in conditions such as chronic obstructive pulmonary disease or liver failure.

Veterans Affairs cohort

Researchers evaluated 252,000 severely ill patients. Falciglia’s presentation focused on data from 216,000 within that group.

Patients were admitted to 177 Veterans Affairs ICUs between September 2002 and March 2005. Two-thirds of the mostly male cohort were over 60 years old; the rest were between 40 to 60 years. Thirty percent had undergone surgery as reason for ICU admission. Admission diagnoses included cardiovascular, respiratory, gastrointestinal, renal and other disorders.

Patients were stratified by mean glucose groups. Normal blood glucose of 70 mg/dL to 110 mg/dL was the range to which subsequent hyperglycemic categories were compared. The other glucose groups ranged from 111 mg/dL to 145 mg/dL; 146 mg/dL to 199 mg/dL; 200 mg/dL to 300 mg/dL; and greater than 300 mg/dL.

Researchers used a validated mathematical model to calculate the risk of death for each patient based on the diagnoses that brought them into the hospital, presence of any other diseases, lab test results and age. A second mathematical model looked at the independent link between a patient’s average blood glucose during their ICU stay and their risk of mortality.

Graded risk

According to Falciglia, risk was graded and began with mild hyperglycemia at levels of only 111 mg/dL to 145 mg/dL. At that range, there was already a 30% increase in mortality.

In 154,000 patients without diagnosed diabetes, the significant association started with mild hyperglycemia and ranged from about a 40% increase in mortality risk to almost a fourfold risk, Falciglia said. In patients with diagnosed diabetes, the significant association did not start until one reached glucose levels of 145 mg/dL.

“Although there was a significant association in both groups, there were clearly striking differences in the magnitude of risk in patients without a diagnoss of diabetes when compared [with] those with a diagnosis of diabetes,” Falcigia said.

In all patients, hyperglycemia increased the risk of death from 1.6 to five times the expected rate for those admitted due to MI, 3.4 to 15.1 times for stroke and 1.7 to 6.2 times for unstable angina.

Researchers found a significant association between hyperglycemia and mortality in all ICU types, Falciglia said. However, in cardiac ICU, the magnitude of mortality was greater than in the medical and surgical ICUs.

Although previous randomized trials have demonstrated that treatment of hyperglycemia decreases morbidity and mortality, more studies are needed, Falciglia said.

“Future randomized trials in glycemic control of hospitalized patients might benefit from focusing on diseases where the risk of death from hyperglycemia seems to be the greatest.” – by Rachel Eskenazi

For more information:

Falciglia M. Hyperglycemia and mortality in 252,000 critically ill patients. Presented at: American Diabetes Association 66th Annual Scientific Sessions; June 13, 2006; Washington.