Patients with macroprolactinaemia: clinical and radiological features

Abstract

BACKGROUND: Macroprolactinaemia may represent a relevant cause of misdiagnosis, unnecessary investigation and inappropriate treatment. The aim of this study was to investigate the clinical and neuroradiological characteristics of patients with and without macroprolactinaemia and to evaluate the impact of macroprolactin determination on the diagnostic work-up of hyperprolactinaemic patients. MATERIALS AND METHODS: Retrospective analysis in 135 consecutive hyperprolactinaemic patients (111 women and 24 men; mean age 37 +/- 11.6 years) whose archived sera were subsequently tested for macroprolactin. Recoveries </= 40% after polyethylene glycol precipitation were indicative of macroprolactinaemia. RESULTS: Macroprolactin, entirely explaining biochemical hyperprolactinaemia, was found in 42.2% of patients, a third of whom presented with signs and symptoms of hyperprolactinaemia. Determination of macroprolactin changed the initial diagnosis in a consistent proportion of patients. In particular, idiopathic hyperprolactinaemia, initially diagnosed in 41 patients, was then excluded in 28 of them. Diagnosis of prolactin-secreting pituitary microadenoma shifted to non-secreting pituitary microadenoma in 10 of 49 patients, while in all patients with prolactin-secreting pituitary macroadenoma or hyperprolactinaemia due to stalk deafferentation the presence of macroprolactin was excluded and the initial diagnosis confirmed. Finally, macroprolactin was present in the majority of patients with magnetic resonance imaging (MRI) scans suggestive for primary empty sella (4 of 5 women) or pituitary hyperplasia (12 of 17 women, 3 of 3 men). Collectively, about half of subjects with macroprolactinaemia showed variable MRI abnormalities. CONCLUSIONS: The presence of macroprolactin was a relevant cause of misdiagnosis in patients with hyperprolactinaemia. However, due to the unexpected high frequency of pituitary abnormalities observed in the present series, we suggest that the diagnostic algorithm of hyperprolactinaemic states should include both polyethylene glycol precipitation test and MRI imaging.