Evaluation of the urinary cortisol: creatinine ratio in the diagnosis of hyperadrenocorticism in dogs.

The diagnostic accuracy of the urinary cortisol:creatinine ratio (CCR), with the cortisol being measured by ELISA, was evaluated by subjecting data from 18 dogs with and 20 dogs without hyperadrenocorticism to receiver operating characteristic (ROC) curve analysis. The area under the ROC curve (W 0.93, SE(w) 0.044) was much higher than 0.5, indicating that the CCR did distinguish between dogs with and without hyperadrenocorticism. A cutoff value of about 60 x 10(-6) was associated with the highest sensitivity (1.0) and specificity (0.85). At the disease prevalence rate of the present study (0.47), the positive and negative predictive values were 0.87 and 1.0, respectively. These numbers indicate that canine hyperadrenocorticism may be safely excluded when the CCR is below 60 x 10(-6) but that a test of higher specificity (eg, the ACTH stimulation test) should be used to confirm the diagnosis of canine hyperadrenocorticism when the CCR is above 60 x 10(-6).