Background and objective: The usefulness of dynamic tests of PRL release in determining underlying pathology in hyperprolactinaemic patients is not well recognized by endocrinologists, only 15% of whom routinely perform such tests. High resolution pituitary magnetic resonance imaging (MRI) has become more widely available during the past 5 years and is now generally regarded as the pituitary imaging method of choice. Since few prolactinoma patients are now submitted to surgery, it is important to ascertain the usefulness of these techniques in suggesting a pathological diagnosis.
Design: A 3 year retrospective audit of the information derived from measurement of PRL and TSH responses to the dopamine antagonist domperidone and from high resolution pituitary MRI in patients being investigated for hyperprolactinemia in regional endocrine unit.
Patients: Eighty-four patients (10 male, 74 female) whose investigation of hyperprolactinaemia included a domperidone test and high resolution pituitary MRI. Patients who had domperidone tests performed after pituitary surgery or who did not have pituitary MRI were excluded from the analysis.
Measurements: PRL and TSH were measured basally and at 30 and 60 minutes following domperidone (10 mg i.v.) and gadolinium-enhanced pituitary MRI was performed in all patients.
Results: 20 patients had a normal PRL response to domperidone (defined as PRL30/PRL0 > 3) and this group included five patients in whom hyperprolactinaemia was not sustained. Pituitary MRI showed evidence of a microadenoma in only two patients, imaging being unequivocally normal in the majority. Sixty-four patients had an abnormal PRL response to domperidone and 18 of these had a macrolesion (nine prolactinomas, nine other tumours). Pituitary MRI was performed in the remaining 46 patients with abnormal PRL response to domperidone and suggested microadenoma in 29 (63%), identified other structural abnormalities in six cases but was entirely normal in 11 cases. Neither the basal PRL level nor the TSH response could refine the diagnosis further because of overlap between the various subgroups.
Conclusions: The majority of patients with a normal dynamic response of PRL to domperidone had a normal or near normal pituitary MRI scan. In the two cases where an abnormality was detected it could have been an incidental microadenoma or cyst, thus suggesting that pituitary scanning could normally be omitted in patients whose PRL response to domperidone is normal (24% of our total). The group of patients with an abnormal dynamic response of PRL to domperidone was not generally amenable to further diagnostic refinement by considering the degree of hyperprolactinaemia or the TSH response to domperidone because of overlap of these parameters between the diagnostic subgroups. Therefore any degree of hyperprolactinaemia associated with a blunted PRL response to domperidone warrants pituitary imaging.