Editorial " ~ ~ Does Pituitary Stalk Compression Cause Hyperprolactinemia ? Accepted thinking has been that all adenohy- releasing factors that could be responsible, at pophyseal hormones are under stimulatory least in part, for the hyperprolactinemia control of hypothalamic-releasing factors ex- found in this group of patients. Thyrotropin- cept for prolactin, which is believed to be releasing hormone (TRH) causes prolactin controlled by an inhibitory factor, dopamine. release, but its physiological role is unclear: If the hypothalamic factors are denied access Prolactin release is not impaired in hyperthy- through interruption of the portal circulation, roidism in which TRH is probably suppressed, which occurs when the stalk is damaged, but the hyperprolactinemia found in associa- prolactin secretion increases and that of the tion with primary hypothyroidism is most other pituitary hormones falls (7,14). The probably related to TRH excess, although hyperprotactinemia found in association with hypothalamic dopamine depletion may also pituitary tumors that do not secrete prolactin play a roleo Vasoactive intestinal polypeptide or with other space-occupying lesions such as (VIP) is a 28-amino acid peptide that is craniopharyngiomas, metastases, and the like synthesized in the paraventricular region of is believed to be due to pressure on the the hypothalamus and that has a marked pituitary stalk, which then deprives the prolactin-releasing effect. It is secreted as a normal prolactin-secreting cells of adequate prohormone, which also contains a 27-amino inhibition (3,6,9,11). How tenable is this acid peptide, called peptide histidine methio- supposition? nine (PHM 27) (13). PHM, like VIP, is a First of all, do tumors pressing on the member of the secretin-glucagon group of stalk interrupt the portal circulation? As the peptides and has a wide distribution in adenohypophysis relies for its blood supply various tissues including the hypothalamus on the portal circulation (8), experimental and pituitary stalk, paralleling that of VIP. It stalk section leads to extensive infarction (2). too has potent prolactin-releasing activity. Slowly growing tumors may be associated The physiological role of these two closely with neovascular channels, which might related peptides is still unclear. In animal explain the rarity of infarction. The "stalk- experiments, destruction of the VIP-secreting related" hyperprolactinemia is not necessarily cells causes decreased protactin cell respon- paralleled by hyposecretion of other pituitary siveness to various stimuli such as 5- hormones nor is it associated with evidence of hydroxytryptophan, which is a serotonin diabetes insipidus. Furthermore, it is very precursor (12). The prolactin rise in response likely that reverse flow of prolactin along the to stress, related to adrenergic effects, is also portal vessels is at least partially responsible blocked. VIP (and probably PHM 27) is also for its suppressive effect on gonadotropin- secreted by the pituitary. Addition of block- releasing hormone. ing antibodies inhibits prolactin release in It has been suggested that raised intrasel- tissue culture (12). Its function is probably a lar pressure due to pituitary tumors or other paracrine one, modulating prolactin release. space-occupying lesions might compress the Beta-endorphin too has a potent prolac- long portal vessels and cause interruption of tin-releasing effect. It is present in pituitary blood flow (8). Although there may be a two- corticotrophs and may exert a paracrine to threefold rise in intrasellar pressure, there stimulatory effect on lactotrophs to account is no correlation with serum prolactin levels; for the hyperprolactinemia frequently found nevertheless, two-thirds of the patients stud- in association with Cushing's disease (4), ied showed mild hyperprolactinemia. although it may also cause hypothalamic Recently, compelling evidence has been dopamine depletion. Posterior pituitary beta- presented for the presence of prolactin- endorphin rises after stalk section and appears 66 Endocrine Pathology Volume 1 Number 2 June 1990 to be inhibited by dopamine (5). Its role in hyperprolactinemic states not caused by this prolactin secretion is unclear, especially as the tumor type. prolactin-releasing activity of posterior pitu- The precise role of the various protactin- itary extracts following stalk section dimin- modulating factors still needs to be elucidated ishes by up to 90% (5). Another small peptide and is likely to clarify the mechanism of the with potent prolactin-releasing properties frequent prolactin elevations not associated was found in the posterior pituitary, but its with radiological lesions as well as those structure has not yet been elucidated (5). associated with non-prolactin-secreting tu- Arginine vasopressin, oxytocin, and neuroten- mors. It is unlikely that stalk pressure alone is sin cause protactin release provided only that the mechanism explaining the serum prolac- the hypothalamus-pituitary flow is intact. tin elevations found in association with Recently, galanin has been found to inhibit non-prolactin-secreting pituitary tumors. hypothalamic dopamine, causing prolactin William Singer, M.B. release, but it can also act as a direct prolactin Associate Professor of Medicine and releaser after intraventricular injection, as Clinical Biochemistry demonstrated in rats (15). In this species, University of Toronto estrogens may stimulate galanin gene expres- Toronto, Ontario sion in both hyperplastic lactotrophs and Canada prolactin-secreting adenomas and may there- fore have a role in the induction and maintenance of the "stalk pressure effect." Growth factors may also play an impor- References tant part in prolactin secretion. Epidermal 1. Birman P, Michard M, Li JY. Epidermal growth factor (EGF) increases prolactin gene growth factor binding sites present in normal expression, reduces growth hormone (GH) human and rat pituitaries are absent in human synthesis in GH 3 and GH4 cells, and is pituitary adenomas. J Clin Endocrinol Metab produced by some pituitary tumors (15). 65:275-281, 1987. 2, Daniel PM, Prichard ML. Studies of the However EGF receptors are absent in prolac- hypothalamus and the pituitary gland. Acta tin- and GH-secreting and nonsecreting Endocrinol [Suppl] 80(201):1-21, 1975. pituitary tumors (1). 3. Grossman A, Besser GM. Prolactinomas. Br Lastly, suckling induces a prompt sus- MedJ 290:182-184, 1985. tained prolactin release that appears to be 4. Horvath E, Kovacs K. The pituitary gland. under posterior pituitary control, presumably Pathol Res Pract 183:12%142, t988. by a fraction of the propressorphysin prohor- 5. Hyde JF, Murai I, Ben-Jonathan N. Differen- mone (12). tial effects of pituitary stalk section on From the clinical point of view, the posterior pituitary and hypothalamic content correlation between size and secretory activity of prolactin-releasing factor, oxytocin, dopa- mine and beta-endorphin. Neuroendocrinol- is a useful guideline (10). Prolactin values ogy 48:314-319, 1988. greater than approximately 150-200 ~g/L are 6. KapscalaLP, Molitch ME, Post KD, Biller BJ, almost always associated with prolactin- Prager RJ, Jackson IMD, Reichlin S. Galactor- secreting tumors. The larger the tumor and rhea, oligomenorrhea and hyperprolactinemia the more invasive, the higher is the serum in patients ,with craniopharyngiomas. J Clin prolactin level (2,3). Some of the immunore- Endocrinot Metab 51:798-800, 1980. active prolactin secreted, however, may be 7. Kikuchi K, Fujisawa I, Momo T, Yamanaka C, biologically inert, and so the clinical picture Kagi M, Nakano Y, Konishi J, Mikawa H, may be that ofa nonfunctioning tumor. Sudo M. Hypothalamic-pituitary function in What of those patients having serum growth hormone--deficient patients with pi- tuitary stalk transections. J Clin Endocrinol prolactin levels below 150-200 txg/L? Those Metab 67:817-823, 1988. with large tumors probably do not have a 8, Lees PD, Pickard JD. Hyperprolactinemia, prolactin-secreting adenoma; however, intrasellar pituitary tissue pressure and the prompt tumor shrinkage in response to a pituitary stalk compression syndrome. J Neu- dopamine agonist would make the diagnosis rosurg 67:192-196, 1987. likely. No provocative tests thus far have 9. Lundberg PO, Osterman PO, Wide L. Serum distinguished prolactin-secreting tumors from prolactin in patients with hypothalamus and Pituitary Stalk Compression and Hyperprolactinemia 67 pituitary disorders. J Neurosurg 55:194--199, O, Hanew tC Yumita S, Andoh N, Sasano N, 1981. Yoshinago IC Distribution, plasma concentra- 10. McComb DJ, Kovacs K, Horvath E, Singer W, tion and in vivo prolactin-releasing activity of Kitlinger DW, Smyth HS, Ezrin C, Weiss peptide histidine methionine in humans. J MH. Correlative ultrastructural morphometry Clin Endocrinot Metab 65:683-688, 1987. of human prolactin-producing adenomas, Acta 14. Vaughan L, Carmel PW, Dyrenfurth I, Frantz Neurochir (Wien) 53:217-225, 1980. AG, Antunes JL, Ferin M. Section of the 11. Nabarro JDN. Pituitary prolactinomas. Clin pituitary stalk in rhesus monkey. Neuroendo- Endocrinol (Ox~3 17:129-155, 1982. crinology 30:70-75, 1980. 12. Reichlin S. Neuroendocrinology of the pitu- 15. Webster J, Ham J, Bevan JS, et al. Growth itary gland. Toxicol Pathol 17:250-255, 1989. factors and pituitary tumors. Trends Endo- 13. Sasaki A, Sam S, Go M, Shimizv Y, Murakami crinol Metab 1: 95-98, 1989. .