Endocrine Abstracts

Background: Diabetes-related ketoacidosis (DKA) is a common and potentially life threatening complication in people with diabetes. Despite national and international guidelines, interhospital guideline variation and mismanagement during admission are important contributory factors to increased DKA duration and length of stay.

Aim: To establish a common DKA registry to identify gaps in management, assess outcomes and share best practises across centres.

Methods: Retrospective analysis of all DKA admissions between 1st January 2021 to 1st December 2021 across six hospitals in the United Kingdom was undertaken. People aged <18 years, admission pH >7.3 or self-discharged before treatment completion were excluded. Information on fluid and insulin prescriptions, glucose and ketone monitoring, DKA duration and length of hospitalisation was collected. Comparison between hospitals was performed using the Independent-Samples Kruskal-Wallis Test. Data was analysed using SPSS version 27.0 and presented in median interquartiles, frequencies and proportion as appropriate.

Results: Since the objective is to identify best practice and not to compare, hospital names are coded A to F to ensure anonymity. A total of 465 DKA episodes across the six hospitals were included. There were differences observed in the DKA duration (median in hours; A- 13.1, B-11, C-9.7, D-15.7, E-19.5, F-15.2; P value< 0.001) and length of hospitalisation (median in days;A-4.6, B-5.4, C-2, D-3.9, E-4.5, F-3.5; P value< 0.001) across hospitals. Similarly, variations were noticed in appropriateness of glucose monitoring(A- 110.9%, B- 86.3%, C- 95.9%, D- 89.1%, E- 92.6%, F- 117.8%; P value <0.001), appropriateness of ketone monitoring (A- 61.3%, B- 83.6%, C- 91.5%, D- 67.3%, E- 62.6%, F- 69.6%; P value<0.001) and fluid prescription (A- 83.6%, B- 80.0%, C- 102.8%, D- 100%, E- 100%, F- 133.3%; P value<0.001). No significant differences were noted in the appropriateness of fixed rate intravenous insulin infusion (A- 100%, B-100%, C- 1008%, D- 98.8%, E- 98%, F- 100%; P value: 0.156).

Conclusion: With the exception of fixed rate intravenous insulin infusion, significant inter-hospital variation in other individual parameters were observed. A centralised DKA registry can abet identifying gaps in DKA management and dissemination of best practises across centres to aid improved patient outcomes.