Abstract
PURPOSE: The present study evaluates the usefulness of an ACTH suppression test shortly after surgery, and to determine optimal cut-off values of included laboratory analyses, in predicting short- and long-term remission after surgery of Cushing's disease. METHODS: A 48 h suppression test with betamethasone 2 mg/day applied after 45 transphenoidal adenomectomies in 28 patients was evaluated. Receiver operating characteristic (ROC)-curves were created for the included assays: plasma cortisol, plasma adrenocorticotropic hormone (ACTH) and urinary free cortisol (UFC). Plasma levels of cortisol and ACTH were measured both at 24 and 48 h. Youden's index was used to determine cut-off with the highest sensitivity and specificity in predicting short- (3 months) and long-term (5 years or longer) remission. The area under curve (AUC) illustrated the clinical accuracy of the different assays. RESULTS: Plasma cortisol after 24 h with betamethasone was most accurate in predicting both short- and long-term remission. 3 months remission with cut-off 107 nmol/L: sensitivity 0.85, specificity 0.94, positive predictive value (PPV) 0.96 and AUC 0.92 (95% CI 0.85-1). 5 years remission with cut-off 49 nmol/L: sensitivity: 0.94, specificity 0.93, PPV 0.88, AUC 0.98 (95% CI 0.95-1). Analyses of ACTH or UFC did not improve diagnostic accuracy. CONCLUSIONS: A 48 h, 2 mg/day betamethasone suppression test after transphenoidal surgery of Cushing's disease could predict short- and long-term remission with a high accuracy. Suppression of plasma cortisol after 24 h with betamethasone to values excluding Cushings disease in the diagnostic setting yielded the highest accuracy in predicting long-term remission.