Predictability of hypoadrenalism occurrence and duration after adrenalectomy for ACTH-independent hypercortisolism

To evaluate if the parameters of hypothalamic-pituitary-adrenal (HPA) axis activity could predict the occurrence and duration of post-surgical hypocortisolism (PSH) in patients with Cushing's syndrome (CS) and with adrenal incidentaloma (AI).

We studied 80 patients (54 females, age 53.3 ± 11 years), who underwent adrenalectomy for CS (17 patients) or for AI (53 patients). Before surgery, we measured adrenocorticotroph hormone (ACTH), urinary free cortisol (UFC) and serum cortisol after 1 mg dexamethasone suppression test (1 mg-DST) levels. After surgery, all patients were given a steroid replacement therapy, and PSH was searched after 2 months by a low-dose (1 µg, iv) corticotropin stimulation test, that was repeated every 6 months in PSH patients for at least 4 years.

The PSH occurred in 82.4 and 46% of CS and AI patients, respectively. In the whole sample and in AI patients separately considered, the PSH was independently predicted by the preoperative cortisol levels after 1 mg-DST, however, with a low (< 70%) accuracy. In AI patients the PSH occurrence was not ruled out even by the cortisol levels after 1 mg-DST lower than 1.8 μg/dL (50 nmol/L). In the 50% of CS patients and in 31% of AI patients the PSH lasted more than 18 months and in 35.7% of CS patients it persisted for more than 36 months. In AI patients, the PSH duration was not predictable by any parameter. However, a PSH duration of at least 12 months was significantly predicted before adrenalectomy (sensitivity 91.7%, specificity 41.2%, positive predictive value 52.4%, negative predictive value 87.5%, p = 0.05) by the presence of at least 2 out of low ACTH levels, increased UFC levels and cortisol levels after 1 mg-DST ≥ 3.0 µg/dL (83 nmol/L).

The PSH occurrence and its duration are hardly predictable before surgery. All patients undergoing unilateral adrenalectomy should receive a steroid substitutive therapy.