ECE2010 Poster Presentations Adrenal (66 abstracts)
1Unit of Endocrinology and Diabetology, Department of Medical Sciences, Università degli Studi di Milano, Fondazione IRCCS Cà Granda - Ospedale Maggiore Policlinico, Milan, Italy; 2Unit of Endocrinology Casa Sollievo della Sofferenza, IRCCS, San Giovanni Rotondo, Italy; 3Unit of Radiology Casa Sollievo della Sofferenza, IRCCS, San Giovanni Rotondo, Italy; 4Unit of Endocrinology, Ospedale San Giuseppe, Università degli Studi di Milano, Milan, Italy; 5Unit of Endocrinology, Department of Medical and Surgical Sciences, Università degli Studi di Milano, IRCCS Policlinico San Donato Institute, San Donato Milanese, Italy.
Objective: Subclinical hypercortisolism (SH) has been associated with increased prevalence of hypertension, type 2 diabetes mellitus, dyslipidemia, central obesity, osteoporosis and vertebral fractures. We aimed to investigate the accuracy of different SH diagnostic criteria in predicting the concomitant presence of these complications.
Patiets: In this retrospective study we evaluated data of 231 patients (120 females and 111 males) affected with adrenal incidentalomas (AI).
Methods: We studied the accuracy of different SH diagnostic criteria (cortisol after 1 mg overnight dexamethasone suppression test 1 mg-DST at different cut-off such as 50, 83, 138 nmol/l, elevated urinary free cortisol, reduced ACTH levels alone or various combination of these parameters) in predicting the concomitant presence of the following three complications: hypertension, type 2 diabetes and vertebral fractures.
Results: The criterion characterized by the presence of two out of 1 mg-DST >83 nmol/l, elevated UFC, reduced ACTH, struck the best balance between sensitivity and specificity reaching a good accuracy in predicting the cluster of complications (61.9; 77.1 and 75.8%, respectively). The presence of the cluster of complications was associated with this criterion (OR 4.75, 95% CI 1.812.7, P=0.002) regardless for gonadal status, BMI and age. Indeed, the SH diagnostic criterion defined as cortisol after 1 mg-DST >138 nmol/l, showed the higher specificity (93.3%) but a poor sensitivity (lower than 23.8%), whereas the criterion by cortisol after 1 mg-DST >50 nmol/l showed the higher sensitivity (71.4%) but a low specificity and accuracy (49.5 and 51.5 respectively).
Conclusions: The SH criterion characterized by the presence of two out of 1 mg-DST >83 nmol/l, elevated UFC, reduced ACTH, seems the best in predicting the presence of chronic manifestations of subtle cortisol excess.