Association between insulin dose and hyperglycemia in hospitalized adults with ischemic stroke receiving continuous enteral nutrition: A retrospective cohort study.

Abstract

Background: Hyperglycemia following an acute ischemic stroke has been linked to increased morbidity and mortality. Because of changes in a hospital-wide sliding-scale insulin protocol to incorporate half-doses at midnight due to hypoglycemia risk, we aimed to evaluate the safety and efficacy of half-dose sliding-scale insulin at midnight compared with full doses in patients with acute ischemic stroke receiving enteral nutrition.

Methods: This single-center, retrospective cohort study involved 151 patients with acute ischemic stroke and receiving enteral nutrition in a neurocritical care unit between January 1, 2014, and December 31, 2021. The exposure of interest was half-dose sliding-scale insulin at midnight compared with full-dose sliding-scale insulin at midnight. The primary outcome was the incidence of hyperglycemia for a 48-h period after stability while receiving enteral nutrition. Secondary outcomes were new infections, incidence of hypoglycemia, and incidence of delirium.

Results: In the full-dose group, 52 patients experienced hyperglycemia compared with 60 patients in the half-dose group; however, after propensity matching for carbohydrate content in enteral nutrition and hemoglobin A1c, the results are not noninferior (risk difference, 3.9%; 95% CI, -21.2% to 13.3%; P = 0.1041). Delirium was significantly higher in the full-dose group, whereas the half-dose group had a higher rate of suspected bacterial infections.

Conclusion: The study indicates that administering half-dose sliding-scale insulin at midnight in patients with acute ischemic stroke receiving enteral nutrition is not noninferior to full doses in controlling hyperglycemia. However, differences in delirium and infection rates suggest that glycemic changes may influence other outcomes.