Is Point of Care (POC) Beta-Hydroxybutyrate (BOHB) [ketones] greater than or equal to 3.3 mmol/L?
Diabetic ketoacidosis (DKA) is a life-threatening complication of insulin dependent diabetes. Historically, the diagnosis is based on serum bicarbonate (CO2) level < 16 mmol/L or pH of < 7.30, the presence of ketonuria (acetoacetate) or ketonemia (beta-hydroxybutyrate), and serum glucose level >200 mg/dL.
At Children’s Mercy Hospital, a CO2 ≥ 16 mmol/L suggests that DKA is unlikely, whereas a CO2 < 16 mmol/L suggests DKA and the patient is then treated using the DKA Clinical Practice Guideline (CPG). As the name implies, the primary driver of the anion gap metabolic acidosis is the ketone, beta-hydroxybutyrate (βOHB). Only recently has the ability to measure this specific ketone at the bedside been available. The Nova Max Plus β-Ketone Monitoring System (Nova Biomedical) is used to monitor whole blood β-hydroxybutyrate (βOHB) at the bed-side in patients in the acute care and inpatient setting during DKA
Comparative values for point-of-care (POC) βOHB and serum bicarbonate (CO2) were evaluated at Children’s Mercy Hospital and the POC βOHB value corresponding to the CO2 value < 16 was established. Receiver Operating Characteristic (ROC) analysis indicated that a POC βOHB value of 3.3 mmol/L predicts DKA with 92.5% sensitivity and 76.2% specificity. The ROC-AUC (area under the curve) was 0.922 with an efficiency of 85%.
Figure 1 - ROC curve with optimal cutoff for DKA diagnosis by POC βHB
The Receiver Operating Characteristic curve (or ROC curve) is a plot of the true positive rate against the false positive rate for the different possible cut-points of a diagnostic test.
An ROC curve demonstrates several things:
It shows the tradeoff between sensitivity and specificity (see example below):
The closer the curve follows the left-hand border and then the top border of the ROC space, the more accurate the test.
The closer the curve comes to the 45-degree diagonal of the ROC space, the less accurate the test.
The area under the curve is a measure of test accuracy.
To assess the stability of the estimated cutoff of 3.3 mmol/L, the data was bootstrap sampled 1,000 times with the results supporting the POC βOHB serum value 3.3 mmol/L as achieving the highest efficiency when predicting DKA. Statistical “Bootstrapping” is often used to see how good a test result performs. Statistical bootstrapping is useful for:
Calculating standard errors
Forming confidence errors
Performing hypothesis tests