Long-term remission and recurrence rates in Cushing's disease: predictive factors in a single-centre study.

To investigate the early and late outcomes of patients with Cushing's disease (CD) submitted to a neurosurgical procedure as first-line treatment. In this single-centre retrospective case notes study, 131 patients with CD with a minimum follow-up period of 6 years (124 operated by transsphenoidal surgery (TSS) and seven by the transcranial approach) were studied. Apparent immediate cure: post-operative 0900 h serum cortisol level <50 nmol/l; remission: cortisol insufficiency or restoration of 'normal' cortisol levels with resolution of clinical features; and recurrence: dexamethasone resistance and relapse of hypercortisolaemic features. In patients operated by TSS, remission of hypercortisolaemia was found in 72.8% of 103 microadenomas and 42.9% of 21 macroadenomas, with recurrence rates 22.7 and 33.3% respectively with a 15-year mean follow-up (range, 6-29 years). Of 27 patients with microadenomas operated after 1991, with positive imaging and pathology, 93% obtained remission with 12% recurrence. In multivariate analysis, the time needed to achieve recovery of hypothalamo-pituitaryadrenal axis was the only significant predictor of recurrence; all patients who recurred showed recovery within 3 years from surgery: 31.3% of patients had total hypophysectomy with no recurrence; 42% of patients with selective adenomectomy and 26.5% with hemi-hypophysectomy showed recurrence rates of 31 and 13% respectively (χ(2)=6.275, P=0.03). Strict remission criteria were not superior in terms of the probability of recurrence compared with post-operative normocortisolaemia. Lifelong follow-up for patients with CD appears essential, particularly for patients who have shown rapid recovery of their axis. The strict criteria previously used for 'apparent cure' do not appear to necessarily predict a lower recurrence rate.