Acquired prolactin deficiency (APD) after treatment for Cushing's disease is a reliable marker of irreversible severe GHD but does not reflect disease status.

OBJECTIVE: We have previously reported that acquired prolactin deficiency (APD) is a marker for severe hypopituitarism and observed a high prevalence of APD in patients treated for Cushing's disease. Recovery of GH secretion is recognized to occur in a proportion of patients treated for Cushing's disease after the effects of glucocorticoid excess on GH secretion have subsided. The aim of this study was to investigate further the association between APD, treated Cushing's disease and, in particular, to determine whether recovery of GH secretion may occur in these patients.

METHODS: Fifty-seven patients (42 female), in remission after treatment for Cushing's disease, were studied. The cohort comprised 13 patients with, and 44 without APD. APD was defined as a serum prolactin persistently below the detection limit of the assay. Severe GH deficiency was defined as a peak GH response of less than 9 mU/l during a GH stimulation test. Age and gender did not significantly differ between subgroups.

RESULTS: Of the 13 patients with APD, a macroadenoma was present in one patient, a microadenoma was present in 10, no lesion was detected in one, and in one patient (treated with an yttrium implant) the size of the tumour was unknown. Of the 28 patients who did not have APD, who were treated with primary surgery a microadenoma was present in 23 and a macroadenoma was present in five. Detailed pituitary imaging was not available in 16 patients who did not have APD, who were treated with primary external XRT. Deficiencies of GH, TSH, LH/FSH (P < 0.0001) and ADH (P = 0.006) status, by conventional testing, were present more frequently in the APD subgroup. In contrast, the prevalence of ACTH deficiency after treatment was not different between the APD and non-APD groups. However, the requirement for additional therapy, targeting the pituitary or adrenal gland, after primary treatment, in those patients not rendered ACTH-deficient, was significantly greater in the APD compared with the non-APD groups (P = 0.003). After pituitary surgery, a significant correlation between peak GH response and interval since remission of Cushing's syndrome was found in the subgroup without APD (r = 0.4, P = 0.04). Four patients who did not have APD, who had documented severe GHD in the immediate postoperative period displayed normalization of GH secretion, when re-tested after a mean interval of 27.2 months. In contrast, no patient with APD after pituitary surgery demonstrated a detectable GH response after up to 132 months of follow-up. No patient with APD showed recovery of prolactin secretion by the time of the most recent measurement (mean 57 months). All 10 patients who developed APD immediately after pituitary surgery had undergone a radical procedure (either a subtotal or total hypophysectomy). In contrast, of 28 patients with Cushing's disease who did not develop APD, only four underwent radical surgery (P < 0.0001). Seven of 14 patients (50%) who underwent a radical operation and two of 20 treated by selective adenomectomy (10%) required additional treatment to achieve control of Cushing's syndrome (P = 0.04).

CONCLUSION: In the presence of APD, patients with Cushing's disease do not experience recovery of GH secretion after treatment, even after the effects of glucocorticoid excess subside. In the absence of APD, GH status may normalize after successful surgical treatment. Although a marker for severe hypopituitarism, APD does not indicate success of treatment of Cushing's disease and may be associated with detectable ACTH secretion from residual corticotroph adenoma activity. APD after pituitary surgery for Cushing's disease occurs only after a radical operation. When a selective adenomectomy is not possible, control of Cushing's disease by operation is less frequent and when achieved, is more often at the cost of hypopituitarism. The optimal management of such patients requires further study.