Endocrine Abstracts

Background: Near-patient capillary beta-hydroxybutyrate (BOHB) meters have been available for several years but evidence as to their clinical utility and reliability in children is still growing. Anecdotal evidence suggests that patients may have a significantly high level of capillary ketones but look clinically well, little evidence is available as to the kind of levels of blood ketones that paediatric patients can clinically compensate for. There is a move towards home blood ketone testing but there is uncertainty about the impact on services.

Methods: A retrospective case note review was conducted to investigate near-patient BOHB testing in paediatric patients who presented at two district general services with established or new diagnosis of type 1 diabetes over 2 years. Data on clinical impression at presentation, paediatric early warning score (PEWS), blood parameters and treatment were collected and statistically analysed.

Results: 14/47 patients presenting as first episode of type 1 diabetes were assessed as clinically well (PEWS=0 or 1) and found to have a significant blood ketone level (>3 mmol/l). Eight of these were non-acidotic as defined by capillary blood gas.

When compared with the patients in established diabetes who were assessed as clinically well, with blood ketones >3 mmol/l where none had normal capillary gases, a significant difference was found (P=0.01684).

These ketotic patients, with or without acidosis, were treated with either i.v. or s.c. insulin. No significant difference in time to resolution of BOHB in these two groups was found (12.1 vs 9.3 h, P=0.402).

Conclusion: This evidence suggests that diabetic ketosis is common at the initial presentation of diabetes in clinically well children who may not be acidotic. Capillary BOHB is a better predictor of biochemical severity in established diabetes. This outcome supports the provision of home-ketone equipment in established diabetes but also provides insight into first presentations of type 1 diabetes in children.