Diagnosing ketoacidosis

Disturbance of normal acid-base balance is a defining feature of diabetic ketoacidosis (DKA), the acute and potentially life-threatening complication of diabetes that results from severe insulin deficiency. The cause of the acidosis in such cases is abnormal accumulation in blood of the two ketoacids, acetoacetate and β-hydroxybutyrate. 

These acids are excreted in urine, and for many years semi-quantitative urine dipstick testing (ketostix) for the presence of acetoacetate has been used to screen for DKA at the patient’s bedside. An alternative approach, made practicably possible by development of a portable POCT analyzer, is to measure β-hydroxybutyrate in blood. 

This analyzer provides quantitation of β-hydroxybutyrate in just 30 sec using a finger-prick (5 µL) capillary sample. A recent French study sought to discover if the newer blood testing system is any more effective in screening for DKA than the traditional dipstick testing of urine. 

This retrospective study focused on 173 patients who, on admission to a Paris Emergency Department, were found to be hyperglycemic (defined as finger-prick blood glucose > 13.7 mmol/L). The finding of hyperglycemia triggered a protocol for investigation of possible DKA, which involved testing each of the study patients’ blood for β-hydroxybutyrate and urine for acetoacetate (ketosis). 

A final diagnosis of DKA was subsequently made in 10 of 173 (6 %) hyperglycemic patients. Retrospective analysis of test results demonstrated that urine and blood tests were equally effective in identifying those patients who were not suffering DKA, (negative predictive value for both tests 100 %). 

However, specificity of the two tests at this cutoff (93.9 % for blood test and 77.3 % for urine test) revealed that a falsely positive result is much more likely if the urine test is used to screen for DKA than if the blood test is used. Delay in obtaining urine samples – more than 2 hrs in 10 % of cases – is an additional disadvantage of urine testing, highlighted by this study. 

The authors are in accord with the American Diabetic Association in recommending that quantitation of blood β-hydroxybutyrate rather than urine dipstick testing be used to screen for DKA in the emergency room.