Average glucose levels at hospital admission for acute myocardial infarction predict 30-day mortality better than does diabetes history, even though the latter is routinely used as an assessment tool, according to an analysis of two large trials.
“Patients with no diabetes history with elevated in-hospital glucose levels have the same high risk for short-term death after AMI [acute myocardial infarction] as patients with diabetes history,” the investigators wrote.
Dr. Abhinav Goya of the Emory Schools of Public Health and Medicine in Atlanta, along with an international team of investigators, conducted a post hoc analysis of two randomized, controlled trials of AMI with ST-segment elevation: the Clinical Trial of Reviparin and Metabolic Modulation in Acute Myocardial Infarction Treatment and Evaluation-Estudios Clinicos Latino America (CREATE-ECLA) and the Organization for the Assessment of Strategies for Ischemic Syndromes-6 (OASIS-6). For the analysis, the investigators looked at glucose measurements taken at hospital admission, at 6 hours post admission, and at 24 hours post admission, and took as “average in-hospital glucose” the mean of the three. They also assessed diabetes history, and then estimated the ability of these to forecast death at 30 days. In cases where there were fewer than all three measurements available, the average glucose was taken to be the mean of two measures, or the one measure by itself.
A total of 30,536 patients were analyzed. Of them, 13,100 (43%) had an average glucose of at least 144 mg/dL, the cut-off point that was used to predict risk of 30-day mortality. Of these 13,100 patients, 8,388 (64%) did not have a history of diabetes. At 30 days, a total of 2,808 deaths were documented. Average glucose predicted mortality with a highly significant odds ratio of 1.10, and this remained undiminished after adjustment for diabetes history, according to Dr. Goya, also of the Population Health Research Institute, Hamilton, Ont. Diabetes history alone also predicted death at 30 days, with a highly significant odds ratio of 1.63; however, after adjustment for average glucose, the odds ratio fell to a nonsignificant 0.98.
Additionally, nondiabetic patients with glucose levels above 144 mg/dL had an average rate of death that nearly matched that of diabetic patients with similarly high glucose—13.2% versus 13.7%, respectively (Am. Heart J. 2009 Feb. 23 [doi:10.1016/j.ahj.2008.12.007]).
“These hyperglycemic patients with no diabetes history would have been overlooked as high risk if diabetic status alone were used for risk assessment,” the authors concluded.
The CREATE-ECLA trial had no external funding; OASIS-6 was funded by Sanofi Aventis, Organon, and GlaxoSmithKline.