Background Cerebral oedema is responsible for the majority of deaths secondary to diabetic ketoacidosis (DKA) in children and young people. The administration of large volumes of intravenous fluid in paediatric DKA has been associated with an increased risk of cerebral oedema. Despite the existence of national guidelines, intravenous fluid prescription is often noncompliant with recommendations, causing children to receive inappropriate volumes of intravenous fluid therapy.

Aims I aim to identify risk factors for excessive intravenous fluid prescription in paediatric DKA. I aim to test the null hypothesis that admission pH, heart rate centile and Glasgow Coma Scale (GCS) score are not significantly associated with the prescription of more than 40 millilitres/kilogram of fluid in the first four hours of admission. I intend to use the results to make recommendations for guideline development, clinical education and future research.

Study Design This retrospective cohort study investigated children under sixteen years, with biochemical features of DKA, prior to retrieval to a Paediatric Intensive Care Unit. Between 2003 and 2013, twenty-three out of sixty-seven admissions were prescribed more than 40 millilitres/kilogram of intravenous fluid in the first four hours of admission. Logistic regression evaluated the relationship between pH, heart rate centile, GCS score and the outcome variable, at the p£0.05 level of significance. ChiSquared and Wilcoxon Rank-Sum tests were used for univariable analysis.

Results An initial GCS score of <9 was associated with the prescription of more than 40 millilitres/kilogram of fluid in the first four hours of admission, displaying a degree of significance (p=0.05; Odds Ratio: 4.25 [95% Confidence Interval: 0.99 to 18.36]). Admission pH and heart rate centile were not associated with the outcome variable.

Conclusion A GCS score of <9 on arrival to hospital may be associated with excess intravenous fluid prescription in paediatric DKA. Guidelines should explicitly state how to manage children presenting to hospital with DKA and a low GCS score with regards to fluid prescription. Education strategies should encourage prompt recognition and management of cerebral oedema. Further research is required to corroborate these findings and clarify the relationship between intravenous fluid and cerebral oedema in paediatric DKA.