Blood Glucose Response to Rescue Dextrose in Hypoglycemic,Critically Ill Patients Receiving an Insulin Infusion

Background: There is inadequate guidance for clinicians on selection of the optimal dextrose 50% (D₅₀W) dose for hypoglycemia correction in critically ill patients. Objective: The purpose of this study was to determine the blood glucose (BG) response to D₅₀W in critically ill patients. Methods: A retrospective analysis was conducted of critically ill patients who received D₅₀W for hypoglycemia (BG < 70 mg/dL) while on an insulin infusion. The primary objective of this study was to determine the BG response to D₅₀W. The relationship between participant characteristics and the dose-adjusted change in BG following D₅₀W was analyzed using simple and multiple linear mixed-effects models. Results: There were 470 hypoglycemic events (BG < 70 mg/dL) corrected with D₅₀W. The overall median BG response was 4.0 (2.53, 6.08) mg/dL per gram of D₅₀W administered. Administration of D₅₀W per protocol resulted in 32 episodes of hyperglycemia (BG > 150 mg/dL), resulting in a 6.8% rate of overcorrection; 49% of hypoglycemic episodes (230/470) corrected to a BG >100 mg/dL. A multivariable GEE analysis showed a significantly higher BG response in participants with diabetes (0.002) but a lower response in those with recurrent hypoglycemia (P = 0.049). The response to D₅₀W increased with increasinginsulin infusion rate (P = 0.022). Burn patients experienced a significantly larger BG response compared with cardiac, medical, neurosurgical, or surgical patients. Conclusions: The observed median effect of D₅₀W on BG was approximately 4 mg/dL per gram of D₅₀W administered. Application of these data may aid in rescue protocol development that may reduce glucose variability associated with hypoglycemic episodes and the correction.