RESUMO
INTRODUCTION: The accurate interpretation of the cosyntropin (adrenocorticotropic hormone [ACTH]) stimulation test requires method- and assay-specific cutoffs of the level of cortisol. Compared with a historical cutoff (18 µg/dL) for polyclonal antibody-based immunoassays, lower thresholds were proposed for the Roche Elecsys II assay, which uses a monoclonal antibody. However, cutoffs for other commonly adopted, monoclonal antibody-based cortisol assays were not yet available. Here, we established the thresholds for the level of cortisol specific to the Abbott Architect immunoassay by comparing the measurements of the level of cortisol using 3 immunoassays. METHODS: The ACTH stimulation test was performed in patients with suspected adrenal insufficiency (n = 50). The serum cortisol level was measured using the Abbott Architect, Roche Elecsys II, and Siemens Centaur assays. The results of the Abbott assay were also compared with those of liquid chromatography-tandem mass spectrometry. The receiver operating characteristic analysis was performed to derive new diagnostic thresholds for the Abbott assay using the polyclonal antibody-based Siemens assay as the reference method. RESULTS: The concentrations of cortisol measured using the Abbott assay were similar to those measured using liquid chromatography-tandem mass spectrometry and the Roche Elecsys II assay but significantly lower than those measured using the Siemens assay. The optimized threshold for cortisol using the Abbott assay was 14.6 µg/dL at 60 minutes after stimulation (sensitivity, 92%; specificity, 96%) and 13.2 µg/dL at 30 minutes after stimulation (sensitivity, 100%; specificity, 89%). CONCLUSION: We recommend a threshold of 14.6 µg/dL for the level of cortisol at 60 minutes after ACTH stimulation for the Abbott assay. In comparison with the historical threshold of 18 µg/dL, the application of the new cutoff may significantly decrease false-positive results due to ACTH stimulation testing. The use of assay-specific cutoffs will be essential for reducing misclassification and overtreatment in patients with suspected adrenal insufficiency.