Endocrine Abstracts

Introduction: Transsphenoidal surgery (TSS) is the first-choice treatment in Cushing’s disease (CD). When performed by experienced neurosurgeon, post-operative remission can be achieved in 70–80% of cases; however, recurrences during follow-up are quite frequent, involving around 20–30% of patients initially in remission. Early detection of relapses is crucial to avoid cortisol-related comorbidities. The aim of the study was to find reliable predictors of recurrence after neurosurgery in patients with CD.

Materials and methods: Forty CD patients (32 F/8 M, median age at diagnosis 39 years, range 13–67) in remission after TSS and with postoperative follow-up of at least 24 months (median 121.5, range 27–212 months) were included in the study. Remission was defined in the presence of at least two of the following criteria: i) post-surgical hyposurrenalism; ii) normal urinary free cortisol (UFC) and late-night salivary cortisol (LNSC) for at least 12 months; iii) serum cortisol < 50 nmol/l after 1 mg dexamethasone suppression test. All patients were submitted to at least one desmopressin (DDAVP) test during follow-up (6–12 months after surgery) and in 37/40 cases even in the diagnostic phase.

Results: Nine patients (22.5%, 8 F/1 M) experienced disease recurrence after a mean time of 32.2 months (range 18–62). Patients with macroadenomas were more prone to recur postoperatively P = 0.003). Early post-operative serum cortisol was lower in patients in prolonged remission compared to that of relapsed patients (55.5, IQR 33.3–54 nmol/l vs 191.5, IQR 57.5–417.5 nmol/l, P = 0.025), even though there was some degree of overlap between groups. A threshold of 76 nmol/l for serum cortisol was able to identify patient at high risk of relapse with sensitivity (SE) and specificity (SP) of 75%. Patients with recurrence displayed a greater ACTH and cortisol response (P = 0.0001) to DDAVP test compared to those in prolonged remission. An absolute increase in ACTH > 7.6 ng/l was identified through the ROC curve analysis (AUC = 0.8796; 95% CI :75–100) as the best predictor of recurrence, with SE of 88.9% and SP of 83.3%. After TSS, no differences in UFC and LNSC values were observed between groups.

Conclusions: The presence of corticotroph macroadenoma is a risk factor for CD relapse. DDAVP test is more accurate than the evaluation of serum cortisol in the immediate post-surgical period in predicting late recurrence. The re-appearance of a positive response to this test is an early marker of future relapse, that precedes the increase in LNSC and UFC by several months thus patients displaying such alteration should be closely monitored.