Abstract 3180: The Relationship Between Spontaneous and Iatrogenic Hypoglycemia and Mortality in Patients Hospitalized with Acute Myocardial Infarction
Background: While blood glucose (BG) control is recommended by professional societies in hyperglycemic patients (pts) with acute myocardial infarction (AMI), enthusiasm for BG-lowering interventions is tempered by concerns of inducing hypoglycemia. Hypoglycemia has been associated with higher mortality in prior studies, but whether this risk is similar in pts who develop hypoglycemia spontaneously and those whose hypoglycemia follows insulin therapy, is unknown.
Methods: We evaluated 7,924 AMI pts hospitalized from 01/00 –12/05 at 40 US hospitals, who were hyperglycemic on admission (initial BG ≥ 140 mg/dL). Pts were stratified based on whether they developed a hypoglycemic event (random BG <60 mg/dL) during subsequent hospitalization. Logistic regression models were used to evaluate the association between hypoglycemia and in-hospital mortality within subgroups of pts who did and did not receive insulin therapy.
Results: Overall, 482 pts (6%) developed hypoglycemia, of which 346 (72%) had hypoglycemia after insulin treatment, and 136 had spontaneous hypoglycemia. In both insulin-treated and non-treated groups, pts with hypoglycemia were older and had more comorbidity. Hypoglycemia severity was similar in the insulin-treated and non-treated groups (nadir BG 46 vs. 47 mg/dL). Hypoglycemia was associated with increased mortality in the non-treated group (18% mortality (25/136) vs. 9% (425/4214), p=0.0003), but not in the insulin-treated group (10.4% mortality (36/346) vs. 10.5% (294/2803), p=0.96). After multivariable adjustment, there was a strong interaction between hypoglycemia and insulin therapy (p for interaction 0.008). Hypoglycemia was a predictor of higher mortality in pts that did not receive insulin (OR 2.40, 95% CI 1.35– 4.26), but not in pts treated with insulin (OR 0.89, 95% CI 0.57–1.41).
Conclusions: While hypoglycemia is associated with increased mortality in pts hospitalized with AMI, this risk appears to be confined to pts who develop hypoglycemia spontaneously. In contrast, iatrogenic hypoglycemia after insulin therapy is not associated with higher mortality risk. These data provide some reassurance to clinicians in their efforts to control BG after AMI, which frequently requires insulin administration.