<b>Background:</b> In a previous study, we determined reference values for basal and thyrotropin-releasing hormone (TRH)- stimulated plasma concentrations of prolactin (PRL). The aim of the present study was to determine the clinical usefulness of the PRL response to TRH in the work-up of patients with hyperprolactinaemia.
<b>Methods:</b> We studied 92 consecutive patients referred for evaluation of hyperprolactinaemia. Patients with confirmed hyperprolactinaemia were divided into three groups: group A (pharmacological hyperprolactinaemia; n=2), group B (pathological hyperprolactinaemia; n=6) and group C (all other patients). Patients in group C underwent MRI of the pituitary and were subdivided into C1 (normal pituitary on MRI; n=6), C2 (slightly abnormal MRI; n=21), and C3 (evident microadenoma or macroadenoma on MRI; n=25 and 12, respectively). The MRI was technically insufficient in four patients. Basal PRL as determined by fluoroimmunometric assay and the PRL response to 400 microg TRH were determined in all patients.
<b>Results:</b> Hyperprolactinaemia was confirmed in 83% of the referred patients. Non-response, defined as a <2.5-fold PRL increase after TRH, occurred in one patient (50%) in group A, in 66% of patients in group B and in 99% of patients in group C. Within group C, basal PRL was not different between group C1 and C2, but higher (p=0.06) in group C3. The absolute PRL increase after TRH did not differ between the three subgroups. The relative PRL increase was smaller (p=0.03) in group C3 but overlapped considerably with groups C1 and C2. All patients except one in group C were so-called non-responders. Basal PRL and absolute PRL increases after TRH correlated with the adenoma diameter on MRI (r=0.66, p=0.0002 and r=0.49, p=0.008, respectively).
<b>Conclusion:</b> In patients referred for elevated serum PRL, hyperprolactinaemia should be confirmed under standardised conditions. The absolute or relative PRL increase after 400 microgram TRH does not help to differentiate between patients with prolactinoma or idiopathic hyperprolactinaemia. Therefore, the TRH stimulation test is not useful in the work-up of hyperprolactinaemia.