Background: Diabetic ketoacidosis (DKA) is a common, life-threatening complication of diabetes. The diagnosis of DKA relies on signs and symptoms, plus laboratory findings of blood glucose (BG) of >250 mg/dL, an anion gap (AG) of ≥15 mmol/L, and carbon dioxide (CO2) of ≤18 mmol/L when other causes of acidosis are excluded. Objectives: To compare the results of a point-of-care test for the ketone beta-hydroxybutyrate (β-OHB) with standard measures for accuracy in predicting DKA. Methods: After providing informed consent, 160 patients who presented with BG of >250 mg/dL underwent testing for β-OHB with the Precision Xtra meter (Abbott Laboratories, North Chicago, IL) at triage in a large urban hospital emergency department. The diagnosis of DKA was made by clinicians by using standard clinical criteria without knowledge of the β-OHB test. Results: A diagnosis of DKA was made in 57 of 160 subjects. The β-OHB values correlated strongly with AG (r = 0.66, p < 0.001) and with CO2 (r = -0.69, p < 0.001), as well as with glucose (r = 0.31, p < 0.001). Cross-classification of DKA vs. β-OHB yielded sensitivity of 98% (95% CI = 91% to 100%), specificity of 85% (95% CI = 78% to 91%), with a positive likelihood ratio of 6.7 (95% CI = 4.22 to 10.78), and negative likelihood ratio of 0.021 (95% CI = 0.003 to 0.144) at the manufacturer-suggested β-OHB level of 1.5. Conclusions: The point-of-care test for β-OHB was as sensitive as more established indicators of DKA. It is more useful than glucose alone for the diagnosis of DKA and offers immediate diagnosis of patients at triage.
- anion gap