MANAGEMENT OF DIABETIC KETOACIDOSIS: A BEFORE-AND-AFTER IMPLEMENTATION STUDY OF THE TWO-BAG METHOD

Document Type

Conference Proceeding

Publication Date

1-1-2024

Publication Title

Crit Care Med

Abstract

INTRODUCTION: Conventional management of diabetic ketoacidosis (DKA) utilizes a variable rate insulin infusion requiring multiple changes to IV fluid contents to resolve hyperglycemia and anion-gap acidosis. The author's institution standardized DKA management by incorporating the two-bag method into a revised orderset in May 2022. The two-bag method has been associated with earlier anion gap closure, faster normalization of blood glucose, and less hypoglycemia. However, not all studies evaluating the twobag method have reached the same conclusion. This study compares conventional management of DKA to the two-bag method. METHODS: This single-center, pre-post study included adults started on an IV insulin infusion while admitted to the emergency department for DKA treatment from July 1st, 2021 to October 31st, 2022. Patients were divided into two-bag or conventional groups based on timing in relation to the two-bag method orderset go-live in May of 2022. The following patients were excluded: age < 18 years, pregnant, and those not meeting DKA diagnostic criteria at the beginning of treatment. The primary outcome was time to anion-gap resolution (≤12mEq/L). Secondary outcomes included ICU and hospital length of stay (LOS), insulin infusion duration, and time to blood glucose ≤250mg/ dL. Safety outcomes included rates of hypoglycemia and hypokalemia during infusion. RESULTS: Time to anion-gap resolution (conventional(n=43): 11 hours vs. two-bag(n=61): 13 hours; p=0.2), median ICU LOS (1.5 days vs. 1.8 days; p=0.3) and hospital LOS (3.7 days vs. 4.0 days; p=0.9) were all similar between groups. Median duration of insulin infusion (19 hours vs. 15 hours; p=0.02) was shorter in the two-bag group while median time to blood glucose ≤250mg/dL (6 hours vs. 8 hours; p=0.03) was longer in the two-bag group. Incidence of hypoglycemia (28% vs. 5%; p=0.001) and hypokalemia (53% vs. 26%; p=0.005) were lower in the twobag group. CONCLUSIONS: The two-bag method was associated with similar efficacy to the conventional method with a potential improvement in the incidence of hypoglycemia and hypokalemia. However, more studies are needed for full evaluation of two-bag method.

Volume

52

Issue

1

First Page

S650

Share

COinS