Context: Adrenalectomy is a radical treatment for hypercortisolism in Cushing's disease. However, it may lead to Nelson's syndrome, originally defined by the association of a pituitary macroadenoma and high plasma ACTH concentrations, a much feared complication.
Objective: The objective of the study was to reconsider Nelson's syndrome by investigating corticotroph tumor progression based on pituitary magnetic resonance imaging scan and search for predictive factors.
Design: This was a retrospective cohort study.
Setting: The complete medical records of Cushing's disease patients at Cochin Hospital were studied.
Patients: Patients included 53 Cushing's disease patients treated by adrenalectomy between 1991 and 2002, without previous pituitary irradiation.
Measurements: Clinical data, pituitary magnetic resonance imaging data, and plasma ACTH concentrations for all patients and pituitary gland pathology data for 25 patients were recorded. Corticotroph tumor progression-free survival was studied by Kaplan-Meier, and the influence of recorded parameters was studied by Cox regression.
Intervention: There was no intervention.
Results: Corticotroph tumor progression ultimately occurred in half the patients, generally within 3 yr after adrenalectomy. A shorter duration of Cushing's disease (adjusted hazard ratio: 0.884/yr), and a high plasma ACTH concentration in the year after adrenalectomy [adjusted hazard ratio per 100 pg/ml (22 pmol/liter): 1.069] were predictive of corticotroph tumor progression. In one case, corticotroph tumor progression was complicated by transitory oculomotor nerve palsy. During follow-up, corticotroph tumor progression was associated with the increase of corresponding ACTH concentrations (odds ratio per 100 pg/ml of ACTH variation: 1.055).
Conclusion: After adrenalectomy in Cushing's disease, one should no longer wait for the occurrence of Nelson's syndrome: modern imaging allows early detection and management of corticotroph tumor progression.