Objective Babies may use alternative cerebral fuels including ketones when blood glucose concentrations are low, but laboratory ketone measurements are slow and expensive. Point-of-care measurement of ketone concentrations, if sufficiently accurate, may provide useful information for clinical care.

Patients and design Eligible babies were 35–42 weeks’ gestation, ≤10 days old and admitted to the newborn intensive care unit. At the time of clinically indicated blood tests, additional samples were taken to measure beta-hydroxybutyrate using a point-of-care analyser and the laboratory method.

Results One-hundred and fifty babies had 142 paired samples. Overall point-of-care accuracy was excellent (mean difference 0.00 mmol/L) and precision was moderate (SD 0.18 mmol/L). A point-of-care measurement ≥0.4 mmol/L was highly predictive of a laboratory measurement ≥0.4 mmol/L (area under the curve 0.98).

Conclusion Point-of-care measurement of blood beta-hydroxybutyrate concentrations is sufficiently accurate in newborns to be potentially useful in clinical care.

Clinical trial registration number Registered with the Australian and New Zealand Clinical Trials Registry ACTRN: 12616000784415. The study was registered before recruitment commenced.