Adrenalectomy Improves Blood Pressure and Metabolic Control in Patients With Possible Autonomous Cortisol Secretion: Results of a RCT

OBJECTIVE: The best approach to patients with adrenal incidentaloma (AI) and possible autonomous cortisol secretion (PACS) is debated. The aim of this study was to assess the metabolic effect of adrenalectomy in AI patients with PACS in relation to cortisol secretion parameters, peripheral activation, and glucocorticoid sensitivity. DESIGN: This is a multicenter randomized study (NCT number: NCT04860180). METHODS: Sixty-two AI outpatients (40–75 years) with AI >1 cm and cortisol after overnight dexamethasone suppression test (F-1mgDST) between 50 and 138 nmol/L were randomized to adrenalectomy (Arm A) or a conservative approach (Arm B). Fifty-five patients completed the 6-month follow-up, 25 patients in Arm A (17 female patients, aged 62.5 ± 10.4 years) and 30 patients in Arm B (24 female patients, 66.1 ± 9.1 years). Plasma adrenocorticotroph hormone (ACTH), 24-h urinary free cortisol, 24-h urinary free cortisone, F-1mgDST, glucose, lipids, glycated hemoglobin (HbA1c) levels, blood pressure (BP), body weight, and treatment variations were assessed. The 24-h urinary free cortisol/cortisone ratio (an 11-beta hydroxysteroid dehydrogenase type 2 activity marker), BclI, and the N363S variants of glucocorticoid receptor (GR) polymorphisms were also evaluated. RESULTS: BP control improved in 68% and 13% of the subjects in Arm A and Arm B, respectively (p = 0.001), and the glycometabolic control improved in 28% and 3.3% of the subjects in Arm A and Arm B patients, respectively (p = 0.02). Arm A subjects more rarely showed the BP and/or glycometabolic control worsening than Arm B patients (12% and 40%, respectively, p = 0.03). The surgical approach was independently associated with BP amelioration (OR 3.0, 95% CI 3.8–108.3, p < 0.001) but not with age, F-1mgDST levels, BMI, and hypertension and diabetes mellitus presence at baseline. The 24-h urinary free cortisol/cortisone ratio and the presence of sensitizing GR polymorphisms were not associated with the surgical outcome. The receiver operating characteristic (ROC) curve analysis showed that the BP control amelioration was associated with F-1mgDST [area under the curve (AUC), 0.82 ± 0.09 p = 0.012]. The F-1mgDST cutoff with the best compromise in predicting the BP amelioration was set at 75 nmol/L (sensitivity 77%, specificity 75%). CONCLUSIONS: AI patients with PACS benefit from surgery in terms of BP and glycometabolic control.