Evaluation of basal cortisol and the adrenocorticotropic hormone stimulation test in 982 dogs undergoing investigation for hypoadrenocorticism.

BACKGROUND: The adrenocorticotropic hormone (ACTH) stimulation test is the gold standard test for diagnosing hypoadrenocorticism. However, basal cortisol is recommended for screening but fails to consider the effect of analytical variation and previous glucocorticoid therapy on test performance and additional information that a stimulation test could provide. HYPOTHESIS/OBJECTIVES: To investigate the effect of varied cut-offs and previous glucocorticoid therapy on the diagnostic performance of basal cortisol and whether ACTH stimulation test results impart additional prognostic information. ANIMALS: Fifty-five and 927 hospitalized dogs with and without hypoadrenocorticism, respectively. METHODS: Retrospective case record analysis of dogs undergoing ACTH stimulation testing. Cortisol concentrations were compared between glucocorticoid-exposed (GS, n = 198) and non-exposed (NGS, n = 729) groups. RESULTS: Basal cortisol concentrations ≤ 2.0 μg/dL were significantly (P < .0001) more common in the GS (n = 91, 46.0%) than in the NGS group (n = 226, 31.0%). False-positive flat-line ACTH stimulation test results occurred in 11 (1.2%) dogs. Basal cortisol ≤ 2.0 μg/dL demonstrated imperfect diagnostic sensitivity (98.2%; 95% CI, 90.3%-100%) and poor specificity (65.8%; 95% CI, 62.7%-68.9%). Increasing the cut-off from 2.0 to 2.4, 2.7, or 3.1 μg/dL improved diagnostic sensitivity at the expense of specificity. In the NGS group, basal > 7.0 μg/dL, post-ACTH > 15.6 μg/dL, and delta < 5.3 μg/dL cortisol concentrations were associated with increased risk of not surviving to discharge. CONCLUSIONS AND CLINICAL IMPORTANCE: Higher basal cortisol concentration cut-offs (>2 μg/dL) maximize test sensitivity to the detriment of specificity. Adrenocorticotropic hormone stimulation test results rule out hypoadrenocorticism in the majority of dogs and can provide prognostic information.