Year of Publication


Degree Name

Doctor of Nursing Practice

Committee Chair

Dr. Melanie Hardin-Pierce

Clinical Mentor

Dr. Ellen Ratcliffe

Committee Member

Dr. Sheila Melander


BACKGROUND: Hypoglycemia can be a complication of diabetic ketoacidosis (DKA) protocols. This prolongs time to DKA resolution, increasing hospital stay and mortality risk. Does a revised DKA protocol reduce the incidence of hypoglycemia and reopening of the anion gap due to inappropriate transition to subcutaneous insulin? There is a lack of published data in the U.S. on factors affecting time to resolution of DKA and LOS in the ICU. This review was focused on evaluation of safety outcomes and protocol effectiveness by comparing hypoglycemia and hypokalemia events and instances of anion gap reopening for the duration of DKA treatment.

PURPOSE: To evaluate the effectiveness and safety of a diabetes ketoacidosis (DKA) protocol at UK Healthcare in patients with diabetes type 1: whether time to AG resolution, hypoglycemia, hypokalemia and anion gap reopening incidences different following transition from old protocol to a revised protocol.

METHODS: Retrospective chart review of patients managed with a DKA protocol before and after protocol revision. Protocol efficacy was evaluated by assessing time to resolution of AG, length of stay (LOS) in the hospital. Protocol safety evaluated by assessing the number of incidences of hypoglycemic, hypokalemic events and events of anion gap reopening.

ANALYSIS: Comparison was done using descriptive statistics as well as parametric and nonparametric tests to determine incidences of hypoglycemia, hypokalemia, anion gap that reopened, time to anion gap resolution difference of 2 protocols. Comparisons of demographic and clinical data of cohorts: t-test for continuous variables and the Mann-Whitney U tests.

RESULTS: 67 patients met biochemical inclusion criteria for DKA: median ages 32 and 31 years, 55% were males and 45% were females on average. The revised protocol (group 2, n=42) did not show to be safer than old algorithm (group 1, n=28) in hypoglycemia events for duration of the treatment with 44% (n=11) in the first group and 53% (n=22) for the second group. It did not show to be safer in terms of hypokalemia, 40 % (n =10) of hypokalemia incidences in the first group and 50% (n=21) in the second group (p=0.458). But it showed time in to anion gap resolution was 3 hours faster when no hypoglycemia happened in the second cohort compared to the first cohort and that the length of stay in ICU decreased by 1 day when no incidences of anion gap reopening happened in the second group as compared to the first cohort. The protocol showed to be safer for patients in terms of faster DKA resolution and shorter ICU LOS but not in terms of incidences of adverse events. Individual factors associate with slower resolution of DKA were lower admission pH (p=.029) in the first group but no correlation found in the second group (p=0.735).

IMPLICATION/CONCLUSION: This project showed no difference in safety outcomes such as hypoglycemia or hypokalemia but improved effectiveness outcomes such as faster AG resolution between two groups. But it showed that increased safety (avoidance of hypoglycemia and AG reopening) of the protocol leads increase in effectiveness and shorter ICU LOS in the second group. Future studies should focus on the staff and providers compliance with following protocol and timely transition from IV insulin infusion to SC insulin.