2026 Southeastern Residency Conference

Authors: Mikayla Texido, Jill Dunning, Caitlin Thomas
Purpose/Background: Hyperglycemia is commonly seen and associated with adverse outcomes among hospitalized patients, especially in the intensive care unit (ICU) because these patients are more vulnerable to poor outcomes. Previous trials demonstrate that improved glycemic management can reduce hospital complications, infections, and costs. In the critical care setting, guidelines recommend continuous intravenous (IV) insulin infusion as the preferred method for achieving specific glycemic goals while avoiding hypoglycemia. Despite these recommendations, there are limited studies that directly compare the safety and efficacy of IV insulin to subcutaneous (SubQ) insulin for glycemic management in the critical care setting. This study aims to evaluate patients who were eligible for IV insulin administration through an electronic glucose management system (eGMS) and compare those who were managed with IV insulin via eGMS versus those managed with SubQ insulin.
Methodology: This was a single center, retrospective chart review conducted at AdventHealth Orlando that evaluated patients in the ICU who received glycemic management with either IV or SubQ insulin from June 18, 2025, to August 18, 2025. Adult patients with either two consecutive blood glucose values ≥180 mg/dL or one blood glucose ≥300 while admitted to the cardiac, neuro, medical, surgical, or multi-system ICU were included. Patients were excluded if they were managed on IV insulin utilizing eGMS for less than 6 hours, were managed with IV insulin without eGMS, had cardiovascular surgery within 30 days prior to inclusion, or were being treated with insulin for diabetic ketoacidosis, hyperosmolar hyperglycemic state, or hyperkalemia. Patients managed with IV insulin through eGMS utilization were compared to those managed with SubQ insulin. The primary outcome was the percentage of blood glucose values within a goal range of 70-180 mg/dL. A total of 774 blood glucose values would provide 80% power to detect a 10% difference. Secondary efficacy endpoints included the time until blood glucose was in range, median blood glucose, and ICU length of stay in both groups, with time on eGMS and percent of successful transitions from IV to SubQ additionally evaluated in the IV group. Safety endpoints included the incidence of hypoglycemia (BG <70 mg/dL) and severe hypoglycemia (BG <40 mg/dL).
Results: Of the 431 patients screened, 149 patients met inclusion criteria and no exclusion criteria with 74 in the IV group and 75 in the SubQ group. The IV group had a total of 3,997 individual blood glucose values while the SubQ group had 1,599. Of the patients included, 82 (55.0%) were male, the median A1c was 6.7%, and the median blood glucose on ICU admission was 198 mg/dL. Blood glucose was within the goal range of 70-180 mg/dL in 79.9% of the measurements in the IV group versus 61.2% in the SubQ group (p<0.001). The median blood glucose in the IV group was 146 mg/dL versus 173 mg/dL in the SubQ group (p<0.001). Of patients who were able to maintain their blood glucose within goal range for at least 6 consecutive hours, the median time to blood glucose within goal range was 6.3 hours in the IV group (n=72) versus 14.0 hours in the SubQ group (n=60; U=1011; p<0.001). The patients in the IV group had a longer median ICU length of stay (7.5 days vs. 4.0 days; U=1877.5; p<0.001) and there were no incidences of severe hypoglycemia in either group.
Conclusion: Hyperglycemic patients admitted to the ICU had a higher percentage of blood glucose values within the goal range of 70-180 mg/dL with intravenous insulin administered via eGMS compared to patients managed with subcutaneous insulin.