Relation of admission glucose levels, short- and long-term (20-year) mortality after acute myocardial infarction.

We examined temporal trends in mortality after myocardial infarction from 1985 to 2008 depending on admission glucose levels. We included 11,324 consecutive patients admitted to our intensive coronary care unit for myocardial infarction from 1985 to 2008. Patients were categorized into normal, mild, and severe hyperglycemia groups (admission glucose levels <140, 140 to 200, and ≥200 mg/dl, respectively). Temporal trends were determined using 3 groups: 1985 to 1990, 1990 to 2000, and 2000 to 2008. The prevalence of hyperglycemia increased from 26% in the 1980s to 49% in the 2000s. The prevalence of hyperglycemia primarily increased in patients without diabetes. Kaplan-Meier mortality was 4%, 8%, and 17% at 30 days and 64%, 71%, and 82% at 20 years in patients with normal, mild, and severe hyperglycemia, respectively. Compared with normal admission glucose level, adjusted 30-day mortality was 3.6-fold greater (95% confidence interval 2.9 to 4.3) in patients with severe hyperglycemia. This association was not dependent on diabetic status (p for interaction = 0.43) but was dependent on the decade of hospitalization with a stronger association from 2000 to 2008 (adjusted odds ratio 7.7, 95% confidence interval 5.4 to 11, p for interaction <0.001). Compared with diabetes, hyperglycemia was a better discriminator for 30-day mortality. Mortality at 30 days decreased from 1985 to 2008, however, it decreased less in patients with hyperglycemia compared with those with normoglycemia. In conclusion, elevated admission glucose levels are common in patients with myocardial infarction and are strongly associated with increased mortality. Mortality decreased less from 1985 to 2008 in patients with hyperglycemia compared with those with normoglycemia. Efforts that establish optimal treatment for these patients remain warranted.