Early post-operative ACTH and cortisol as predictors of remission in Cushing’s disease

Abstract

To study the value of early (24 h) post-operative ACTH and serum cortisol as predictors of remission after transsphenoidal surgery in Cushing's disease. We prospectively studied 44 patients who underwent transsphenoidal surgery for Cushing's disease between 1997 and 2005. The mean follow-up period of patients after surgery was 49 months (19-102 months). The predictive value of clinical characteristics, pre-operative hormonal studies, radiological, surgical and histological findings, and post-operative hormonal studies were analysed. For the post-operative hormonal study plasma ACTH and serum cortisol were determined at 8.00 a.m. the day after surgery. After surgery, Cushing's disease remitted in 39 patients (89%) and persisted in 5 patients (11%). Three patients relapsed during the follow-up period. Only three study variables were predictive of persistence of Cushing's disease after surgery: the non identification of the adenoma in histology (an adenoma was found in 87% of the patients in remission, and in 20% of treatment failures, p = 0.01), the early post-operative plasma ACTH (patients in remission: 2 pmol/L (1.1-10.8 pmol/L), treatment failures: 8.2 pmol/L (1.1-12 pmol/L), p = 0.019), and the early post-operative serum cortisol (patients in remission: 128.4 nmol/L (27.6-4644 nmol/L), treatment failures: 797 nmol/L (606-1037 nmol/L), p = 0.003). ROC curves indicated that plasma ACTH < or = 7.55 pmol/L distinguished patients in remission from treatment failures with 80% sensitivity and 97.4% specificity, and serum cortisol < or = 585 nmol/L with 100% sensitivity and 90% specificity. Twenty-four hours after transsesphenoidal surgery for Cushing's disease, and without glucocorticoids replacement, patients with serum cortisol concentrations higher than 585 nmol/L, and/or plasma ACTH higher than 7.55 pmol/L, and/or those in which an adenoma is not identified in the histological study, have a high risk of treatment failure.