Dexamethasone-suppressed corticotropin-releasing hormone stimulation test for diagnosis of mild hypercortisolism

Context: The definitive diagnosis of Cushing's syndrome (CS) in the setting of mild disease, as well as exclusion of CS in the setting of conditions that might mimic this clinical entity (pseudo-Cushing's syndrome), continues to present a significant challenge to the clinician. Objective: The aim of the study was to review characteristics of the combined dexamethasone-suppressed CRH stimulation test in patients evaluated at an academic center for the possibility of mild CS. Design, Patients, and Methods: We conducted a retrospective review of 66 patients. A total of 51 patients underwent final statistical analysis: 21 (41%) had Cushing's disease, and 30 were considered to have pseudo-CS based on the clinical scenario, comorbidities, and follow-up. Sensitivity, specificity, and diagnostic accuracy of cortisol and ACTH levels for the diagnosis of Cushing's disease were calculated at 1 min before, and 15, 30, 45, and 60 min after CRH administration. Diagnostic cutoffs for each parameter were determined by minimizing the absolute difference between sensitivity and specificity. Diagnostic accuracy was characterized by the area under the receiver operating characteristic curve, determined using the trapezoid rule. Results: The highest diagnostic accuracy was provided by the serum ACTH level at 15 min post-CRH, in which the area under the receiver operating characteristic curve was 99.7%, and a cutoff of more than 27 pg/ml (>5.9 pmol/liter) provided a sensitivity of 95% and specificity of 97% for the diagnosis of CS. A 15-min post-CRH cortisol greater than 2.5 μg/dl (70 nmol/liter) provided a sensitivity and specificity of 90 and 90%, respectively. Conclusions: Our results differ from previous studies because our data suggest that when using the combined dexamethasone-suppressed CRH stimulation test, a 15-min post-CRH ACTH value greater than 27 pg/ml (5.9 pmol/liter) had the highest diagnostic accuracy for the detection of CS. However, the sensitivity and specificity for this test were not statistically different from the sensitivity and specificity of other tests, such as those measuring post-CRH stimulated ACTH levels or post-CRH cortisol levels at other time points. Therefore, clinicians should be cautious about interpretation of suppression and stimulation tests in the diverse population of patients with hypercortisolism.