Early assessment of outcome of pituitary surgery for Cushing's disease

Abstract

In recent years selective pituitary adenomectomy by trans­ sphenoidal microsurgery has becom e the treatment o f choice for Cushing’s disease. However, it is not successful in all patients. Recently, the European Cushing’s Disease Survey Study Group has reported on 668 patients with Cushing’s disease from 25 institutions throughout Europe treated with transsphenoidal surgery between 1975 and 1990 (Bochicchio et al., 1995). clinical and biochem ical remission o f hypercortisolism after surgery occurred in 76-3% o f patients. Unfortunately, not all o f the patients in remission after surgery were really cured: the disease recurred at a mean time o f 39-3 months in 12-7% o f 5 1 0 patients who were in remission after the operation. The distribution o f the recurrences did not show any plateau or cluster throughout the follow-up period, so it may be expected that with longer follow-up more relapses will be observed. Accurate early assessment o f outcome o f pituitary surgery for Cushing’s disease is important to expedite further treatment in patients with persistent hypercortisolism. Patients in whom the morning serum cortisol level and/or 24-hour urinary cortisol excretion are above the normal range in the early postoperative period (5--14 days o f surgery) are considered surgical failures. They have to be referred for further treatment, e.g. early repeat surgery (Ram et al ., 1994) or pituitary irradiation (Estrada et ill., 1997). On the contrary, the finding o f a morning serum cortisol level and 24-hour urinary cortisol excretion clearly below normal, especially when accompanied by symptoms o f secondary adrenocortical insufficiency, will almost invariably a clinical remission of Cushing’s disease. However, a small number of patients in whom pituitary surgery induces a remission o f the disease are not hypocortisolaemic immediately after surgery: serum cortisol may decline only gradually to a level below normal in patients with adrenal macronodules and a period o f hypocortisolaem ia after surgery may even be lacking com pletely in patients with periodic hypersecretion o f cortisol or in those receiving medical treatment to lower their cortisol production before surgery (Friedman & Chrousos, 1993). Therefore, when in