Paediatric Ketone Measurement and Intervention in Non-Diabetic Patients in Australasian Emergency Departments: A PREDICT Clinician Survey.

OBJECTIVE

To (i) assess variation in self-reported screening, monitoring and treatment of elevated ketone levels in non-diabetic children and (ii) establish future research priorities.

METHODS

Voluntary electronic survey of staff at Paediatric Research in Emergency Departments International Collaborative (PREDICT) network-affiliated hospitals. Respondents provided information on self-reported practice, beliefs about screening, monitoring, and treating elevated blood ketone levels, and identified future research priorities.

RESULTS

Four hundred twenty-three nurses and 397 doctors from 28 sites across Australia and NZ contributed responses. Nurses were more likely than doctors to screen (90% vs. 75%) and monitor (66% vs. 37%) ketone levels. Greater than 24 h of vomiting or poor oral intake triggered respondents to check ketone levels (83% and 75% respectively). The definition of 'elevated' ketones varied considerably, with nurses' median 1 (IQR 0.7-2) mmol/L and doctors' median 2 (IQR 1-2) mmol/L. Similarly, 'severely elevated' ketones varied considerably, with nurses' median 3 (IQR 2-4) mmol/L and doctors' median 4 (IQR 3-5) mmol/L. 25% of respondents changed route of fluid administration in response to a severely elevated level, though only half of respondents provided a higher-glucose fluid. Sixty-seven percent of respondents supported further prospective observational research on the role of ketone screening and targeted therapies in children aged 6 months-6 years who fail an oral fluid trial.

CONCLUSIONS

Nurses and doctors in Australia and NZ regularly screen for high ketone levels in non-diabetic children in ED, with significant differences in interpretation and management of 'elevated' levels.