(298) Bolus Versus No Bolus Insulin Algorithm in the Treatment of Diabetic Ketoacidosis

Introduction: Diabetic ketoacidosis (DKA) is a severe complication of diabetes defined as a combination of hyperglycemia, metabolic acidosis, and ketonemia. The mainstay of treatment is intravenous insulin to reverse the underlying ketogenesis while simultaneously treating the hyperglycemia. The exact insulin dosing regimen for patients with DKA, however, has been debated given a lack of literature. The 2009 American Diabetes Association guidelines recommend the use of either an initial insulin bolus (0.1 units/kg) followed by a continuous infusion of 0.1 units/kg/hr or a no bolus approach with a continuous infusion of 0.14 units/kg/hr. The objective of this study was to compare the efficacy and safety of an initial bolus to a no bolus insulin algorithm for the treatment of patients with diabetic ketoacidosis after an institutional guideline change from the bolus to no bolus approach.

Methods: This was a retrospective chart review of patients admitted to a large academic medical center from August 2022 to April 2024 that were diagnosed with DKA. Patients ages 18 to 89 were included if a chart diagnosis of DKA was present and received either the initial bolus or no bolus strategy. Patients were excluded if diagnosed with euglycemic DKA or hyperosmolar hyperglycemic syndrome. The primary outcome was a composite of hypokalemia ( < 3.3 mEq/L) or hypoglycemia ( < 70 mg/dL) within 24 hours of insulin initiation for the treatment of DKA. Secondary endpoints included time to DKA resolution, re-opening of anion gap, cumulative insulin dose to DKA resolution, length of hospital stay, and 28 day mortality.

Results: In total, 273 patients were screened, 90 patients were included (50 in the bolus group and 40 in the no bolus group). The mean age was 43 years and 54% were female. The primary composite outcome of hypokalemia or hypoglycemia occurred in 19 (38%) patients in the bolus group and 11 (28%) patients in the no bolus group (P=0.294). There were no differences in secondary outcomes except hospital length of stay which was reduced in the no bolus group (6.9 days vs 3.75 days; P=0.032).

Conclusions: After transition to a no bolus insulin algorithm for the management of DKA, the no bolus strategy was found to have similar efficacy and safety outcomes as compared with an initial bolus strategy.