Accuracy of serum beta-hydroxybutyrate measurements for the diagnosis of diabetic ketoacidosis in 116 dogs

The aim of this study was to evaluate the accuracy of serum beta-hydroxybutyrate (beta-OHB) measurements for the diagnosis of diabetic ketoacidosis (DKA) in dogs. One hundred sixteen diabetic dogs were prospectively enrolled in the study: 18 insulin-treated (IT) diabetic dogs that had a positive urine ketone test and 88 untreated, newly diagnosed diabetic dogs. Venous blood gas tensions and pH, serum glucose and urea nitrogen (SUN), and electrolyte (Na+, Cl-, and K+) and urine acetoacetate (AA) concentrations were measured concurrently with serum beta-OHB concentrations. On the basis of laboratory findings, the patients were assigned to I of 3 groups: diabetic ketoacidosis (n = 43); diabetic ketosis (DK, n = 41); and nonketotic diabetes (NDK, n = 31). Serum beta-OHB concentrations differed significantly (P < .001) among the study groups. Although marked differences in beta-OHB concentrations were found, a considerable overlap exists between the distributions of dogs with DK and those with DKA. The overall accuracy of beta-OHB determination as a diagnostic test for DKA, determined by the area under the receiver operating characteristic (ROC) curve, was 0.92. In the 1.9- to 4.8-mmol/L range, serum beta-OHB determination sensitivity varied from 100 to 35.7%, whereas specificity varied from 39 to 100%. The cutoff value of 3.8 mmol/L showed the best equilibrium between specificity (95%), sensitivity (72%), and likelihood ratio (14.8). We concluded that the quantitative measurement of serum beta-OHB may be a potential tool for diagnosing and monitoring ketosis and ketoacidosis in diabetic dogs.