New Point-of-Care test identifies diabetic ketoacidosis in triage

Abstract

Study objectives: Diabetic ketoacidosis (DKA) is a common, life-threatening complication of diabetes. Immediate treatment is necessary to avoid rapid clinical deterioration and the need for higher levels of care. Patients presenting with elevated blood glucose levels have become so common in the emergency department (ED) that clinical suspicion of DKA may not be sufficient to discriminate the true metabolic emergency from less serious hyperglycemic conditions. Treatment delays are common because of nonclassic presentation and the need for laboratory test results to make the diagnosis of DKA. The diagnosis of DKA relies on signs and symptoms plus laboratory findings of blood glucose level greater than 250 mg/dL, an anion gap greater than 15 mmol/L, and carbon dioxide (CO 2 ) level less than 20 mmol/L, excluding other, less common causes of anion gap acidosis. We compared the results of a point-of-care test for the ketone β-hydroxybutyrate (β-OHB) with standard measures for accuracy in predicting the diagnosis of DKA. Methods: After providing informed consent, 101 patients who presented with blood glucose greater than 250 mg/dL or clinical signs of DKA underwent testing for β-OHB at Barnes-Jewish Hospital ED triage using the Precision Xtra meter (Abbott Laboratories, North Chicago, IL). Additional diagnostic evaluation followed clinical guidelines but included a measured glucose and electrolyte panel in all cases. The diagnosis of DKA was made by experienced clinicians using standard clinical criteria. Logistic regression was used to identify clinical indicators of DKA, and χ 2 analysis was used to evaluate binary decision rules. Results: Demographics of the study group include mean age 46.4±16.3 years (range 20 to 83 years); 45.1% female patients; 83.3% black patients, and 16.6% white patients (including 1 Asian and 1 Hispanic patient). The mean blood glucose level was 511±265 mg/dL (range 76 to 1,340 mg/dL); creatinine level 1.76±1.56 mg/dL (range 0.5 to 9.8 mg/dL); and anion gap level 15.7±8.7 mmol/L. A clinical diagnosis of DKA was made in 37 patients. Values of β-OHB ranged between 0 and 6 mmol/L (mean 1.86±2.12 mmol/L). The β-OHB values correlated with anion gap (0.614 mmol/L, P 2 (–0.603 mmol/L, P P 1.5 mmol/L, patient at high risk for DKA; β-OHB from 0.6 to 1.5 mmol/L with blood glucose >250 mg/dL, patient at intermediate risk of DKA). This decision rule was significantly related to DKA status (χ 2 =38.176, P 2 ), implied a complex decision rule with sensitivity of .949 and specificity of .919. Additional logistic regression analysis, using only information available at triage (β-OHB and BG), showed that blood glucose level added little to classification of patients. Therefore, an alternate decision rule using only β-OHB was tested, in which β-OHB greater than 1.1 mmol/L was used as an indicator of DKA status. This decision rule left diagnostic sensitivity at .949, reduced specificity to .806, and provided an overall correct classification rate of .861 (χ 2 =54.66, P Conclusion: The point-of-care test for β-OHB was as sensitive as more established clinical indicators of DKA and offers immediate preliminary diagnosis of patients so that timely, critical interventions can be initiated.