(300) Effect of Insulin Bolus Omission on Time to Resolution of Diabetic Ketoacidosis

Introduction: Diabetic ketoacidosis (DKA) is an acute, life threatening complication of diabetes that is characterized by hyperglycemia, metabolic acidosis, and increased ketone production. Insulin therapy is required to correct the underlying metabolic process but recommendations on its dosing vary between guidelines. This study aims to evaluate if giving an insulin bolus prior to a fixed-rate intravenous insulin infusion (FRIII) influences the time to resolution of DKA.

Methods: This single-center, IRB-approved, retrospective review evaluated patients presenting to the emergency department with DKA from January 2021-December 2022. Patients were grouped based on if they received an insulin bolus or no bolus prior to beginning an FRII I. Patients were excluded if their initial blood glucose (BG) was ≤250mg/dl, they did not receive an FRIII or their FRIII was initiated at 0.14u/kg/hr, acidosis resolution did not occur prior to discharge, they were transferred from an outside facility, or they were pregnant or incarcerated. The primary objective was time to resolution of DKA defined as a blood BG < 200 mg/dl and two of the following criteria: a serum bicarbonate level ≥15 mEq/l, a venous pH >7.3, and a calculated anion gap ≤12 mEq/l. Secondary objectives included assessing time on insulin drip, time to BG < 200 mg/dl, and safety signals of hypoglycemia and hypokalemia between groups.

Results: Patients in the bolus group presented with significantly higher initial blood glucose (BG) and anion gap as well as lower serum bicarbonate and pH. The time to resolution of DKA from the start of insulin therapy did not differ between the bolus and no bolus group (14.4 vs 14.5 h, p=0.81). There was also no difference in time on insulin drip (20.0 vs. 18.2 h, p=0.41), time to BG < 200mg/dL (7.5 vs. 6.3 h, p-0.22), rates of hypoglycemia (8.2 vs. 5.8%, p=0.72), or rates of hypokalemia (14.8 vs 7.7%, p=0.38). The initial rate of blood glucose decrease between the first two basic metabolic panels (BMPs) was higher in the bolus group (60.2 vs. 42.7 mg/dL/hr, p=0.002).

Conclusions: In this retrospective chart review, administration of an insulin bolus prior to initiation of an FRIII did not reduce time to resolution of DKA when compared to patients that did not receive an insulin bolus.