Tailored reference limits for urine corticoid:creatinine ratio in dogs to answer distinct clinical questions

To establish reference intervals for the urinary corticoid:creatinine ratio (UCCR) determined by chemiluminometric immunoassay, UCCR was measured by this method in 50 healthy dogs. To assess the diagnostic performance of different cut-off levels, the UCCR of 66 dogs with hyperadrenocorticism and 87 dogs with diseases mimicking hyperadrenocorticism were used to construct a receiver operating characteristic (ROC) curve. The upper reference limit derived from morning samples in healthy dogs was 30.81 x 10(-6). The area under the ROC curve was 0.94. The diagnostic cut-off with the highest negative likelihood ratio was 26.5 x 10(-6) (sensitivity 1, specificity 0.54), whereas the cut-off with the highest positive likelihood ratio was 161.2 x 10(-6)(specificity 0.988, sensitivity 0.515). The application of these two different diagnostic cut-offs eliminated the necessity to perform additional tests in 53 per cent of the patient population.