Arora S, Henderson SO, Long T et al.
University of Southern California, Los Angeles, CA, USA.
Diabetes Care 2011;34:852–4.
Editor’s note: Urine testing for glycosuria used to be a cornerstone in the monitoring of glycemic control in people with diabetes, before it was superseded by home blood glucose monitoring and measurement of blood glycated hemoglobin. Urine testing for ketonuria in patients with suspected diabetic ketoacidosis (DKA) has continued in mainstream practice, but here too the tide is turning. Clinicians have long-known that assessment of ketonuria is not fool-proof – false-positives are common, particularly in people who are fasting, and these can cause unnecessary anxiety and lead to a cascade of further blood tests and investigations, which in hindsight are not required (Ann Emerg Med 1999;34:342–6). Near-point estimation of blood β-hydroxybutyrate levels, using a capillary sample of blood, has become common practice in many hospitals in patients with suspected DKA. This provides a quantitative estimation of real-time blood ketone levels rather than the qualitative estimation yielded by urine testing, which reflects the ketone status in the preceding hours.