Dexamethasone measurement during low-dose suppression test for suspected hypercortisolism: threshold development with and validation.

Abstract

Dexamethasone Suppression Test (DST), recommended for Cushing's Syndrome (CS) diagnosis, explores the pituitary feedback to glucocorticoids. Its diagnostic accuracy could be affected by dexamethasone bioavailability, and therefore, we have developed and validated a dexamethasone threshold after 1-mg DST. We studied 200 subjects: 125 patients were considered retrospectively and 75 were enrolled prospectively as the validation cohort. Serum dexamethasone, Late Night Salivary Cortisol (LNSC), and Urinary Free Cortisol (UFC) were measured with LC-MS/MS. Normal LNSC and UFC levels were used to exclude CS. The lower 2.5th percentile of dexamethasone distribution in non-CS patients with cortisol ≤ 50nmol/L after 1-mg DST was used as threshold. 16 patients were CS and 184 non-CS (108 adrenal incidentaloma and 76 excluded CS); 4.5nmol/L resulted the calculated threshold. Cortisol after 1-mg DST confirmed high sensitivity (100% at 50nmol/L cut-off) and moderate-low specificity (63%, increased to 91% at 138nmol/L) to diagnose CS in the whole cohort of patients. We could reduce the number of false-positive results (from 10 to 6 and from 7 to 4 in AI and excluded CS) considering adequate dexamethasone levels. Dexamethasone levels were not affected by hypercortisolism, age, gender, smoke, weight, and creatinine. 6% of non-CS patients did not achieve adequate dexamethasone levels (40% of tests with serum cortisol > 138nmol/L after 1-mg DST). We developed and validated the routine dexamethasone measurement during 1-mg DST: it is independent from patient's clinical presentation, and it should be used to increase the specificity of serum cortisol levels.