Temporal trends and hospital variation in the management of severe hyperglycemia among patients with acute myocardial infarction in the United States

Background Elevated blood glucose is associated with higher mortality in patients with acute myocardial infarction (AMI). Although clinical guidelines recommend targeted glucose control in this group, clinical trials have yielded inconclusive results. Our objective was to understand how this lack of evidence impacts the management of severe hyperglycemia in routine practice. Methods We examined insulin use among 4,297 AMI admissions with a mean hospitalization blood glucose of ≥200 mg/dL across 55 US hospitals from 2000 to 2008. Temporal trends and interhospital variation in 2 measures of insulin use during hospitalization - any (subcutaneous, intravenous [IV], short acting, long acting) and IV insulin - were examined using hierarchical Poisson regression models. Results Of the 4,297 admissions, 2,618 (61%) received any insulin and 538 (13%) received IV insulin. After multivariable adjustment, a slight increase in insulin use was observed per admission year (relative risk [RR] 1.06, 95% CI 1.01-1.11). There was a modest (albeit nonsignificant) increase in IV insulin use seen before May 2004 (RR 1.18, 95% CI 0.96-1.47), with no significant change thereafter (RR 0.99, 95% CI 0.92-1.09). Marked variability in insulin use was observed across hospitals (median rate ratio 1.5 [any insulin] and 1.8 [IV insulin]), which did not change over time. Conclusions Insulin use among patients with AMI and severe hyperglycemia has remained low over the past decade, with substantial and persistent interhospital variation. These observations reflect marked clinical uncertainty with regard to glucose management in AMI, underscoring the imperative for a definitive clinical trial in this field.