The level of a person’s mean blood glucose when admitted to a hospital for an acute myocardial infarction is known to be a predictor of mortality. However, mean blood glucose does not capture the fluctuations in glucose levels during an inpatient stay.

These researchers analyzed data on 18,563 patients hospitalized with acute myocardial infarction at 61 hospitals from 2000 to 2008 to see what blood glucose metrics predicted mortality during the first 48 hours of hospitalization.

Greater glucose variability was associated with higher mortality for five metrics but was no longer significant after multivariable adjustment. This contrasts with other study findings that critically ill patients with glycemic variations in various hospital units are at higher risk for death. The authors explain that it is possible that critically ill patients (other than heart attack patients) are more susceptible to the effects of fluctuating glucose levels. They point out that patients in medical/surgical intensive care units are more intensely monitored for blood glucose—and insulin treatment protocols—than patients in cardiac care units.

They conclude that calculating glucose variability from blood glucose levels in order to stratify risk, and intervening to lower glucose levels, may not be warranted in patients with acute myocardial infarction.