The inability of dynamic tests of prolactin and TSH secretion to differentiate between tumorous and non-tumorous hyperprolactinemia.

Certain hyperprolactinemic patients have an obvious pituitary tumor while others with normal pituitary radiology may or may not harbor a pituitary microadenoma. A variety of biochemical tests have been proposed to distinguish between those with and those without pituitary tumors. The aims of this study were: firstly to examine these tests to assess their efficacy in differentiating between patients with radiologically-demonstrated pituitary tumors and normal controls; and secondly to establish if those hyperprolactinemic patients with normal radiology formed two distinct groups biochemically as might be expected if some did and some did not have tumors. The prolactin (PRL) and thyroid stimulating hormone (TSH) response to domperidone and the PRL response to TRH and insulin-induced hypoglycemia have thus been examined in hyperprolactinemic subjects with and without radiological evidence of an adenoma and in normal controls. The basal serum PRL was similar in patients with and without radiological evidence of a pituitary adenoma. The serum PRL response to all stimuli studied, expressed as a percentage of initial values, was blunted in patients with known pituitary tumors with total separation from values in control subjects. Results for patients with normal pituitary radiology were similar to those for patients with tumors with minimal overlap with controls. The peak TSH increment after domperidone was exaggerated in patients with known tumors, but overlap with control values was observed in 25%. In patients with normal radiology the peak TSH increment after domperidone was similarly increased but again overlap with control values occurred in 28%. Cluster analysis showed no evidence of two subgroups of response with in the hyperprolactinemic patients.(ABSTRACT TRUNCATED AT 250 WORDS)