PROLACTIN-SECRETING PITUITARY MICROADENOMA: DETECTION AND EVALUATION*t

FERTILITY AND STERILITY Copyright 1978 The American Fertility Society Vol. 29, No.3, March 1978 PrinlRd in U.s.A. PROLACTIN-SECRETING PITUITARY MICROADENOMA: DETECTION AND EVALUATION*t R. HERBERT WIEBE, M,D,:j: CHARLES B, HAMMOND, M.D. STUART HANDWERGER, M.D. Endocrine Division, Department of Obstetrics and Gynecology, Duke University Medical Center, Durham, North Carolina 27710 Eleven women with secondary amenorrhea associated with hyperprolactinemia were studied. Base line hormone evaluations, visual field determinations, and routine sella turcica x-rays were normal. Prolactin-secreting pituitary microadenomas were found in all of the patients only after further diagnostic studies were carried out. These studies included poly tomography of the sella turcica; dynamic pituitary testing of growth hormone reserve, adrenocorticotropic hormone reserve, and gonadotropin reserve; and prolactin suppression with L-dopa. The early diagnosis of a small prolactin-secreting adenoma may be possible if several diagnostic criteria are utilized. The most sensitive diagnostic indices available are (1) poly tomography, (2) the magnitude of the plasma prolactin elevation, and (3) the failure to suppress prolactin secretion with L-dopa. Our findings emphasize the importance of an extensive evaluation of all women with amenorrhea associated with hyperprolactinemia. Hyperprolactinemia, with or without galactorrhea, is frequently associated with amenorrhea and represents a major disturbance of the hypothalamic-pituitary system. 1 The etiology of hyper - prolactinemia in patients with a radiographically normal sella turcica (as delineated by standard anteroposterior and lateral sella x-rays) is frequentlyobscure. However, with the recent advent of hypocycloidal polytomography2 and newer laboratory methods for assessment of hypothalamicpituitary function, it is now possible to identify in some of these patients a pituitary adenoma less than 1 cm in diameter (microadenoma). In a preliminary study we reported the efficacy of polytomography and the laboratory assessment Received September 6, 1977; revised November 9, 1977; accepted November 9, 1977. *Supported by Grant RR-30 from the General Clinical Research Centers Program, Division of Research Resources, National Institutes of Health. tpresented at the Ninth World Congress on Fertility and Sterility and the Thirty-Third Annual Meeting of The American Fertility Society, April 12 to 16, 1977, Miami Beach, Fla. :j:to whom reprint requests should be addressed. Recipient of Research Career Development Award 1-1000065-01 from the United States Public Health Service. 282 of pituitary function in the early diagnosis of pituitary adenomas in four women.3 We have subsequently studied an additional seven women in whom a pituitary microadenoma was diagnosed. These additional seven patients, along with the original four patients, are reported here and reemphasize the importance of the utilization of several diagnostic criteria in the early diagnosis of these tumors. Patients MATERIALS AND METHODS Eleven women (ages 19 to 37) with secondary amenorrhea of 1 to 7 years' duration were studied. Galactorrhea was present in eight of the women. The onset of amenorrhea, with or without galactorrhea, was spontaneous in five of the patients, postpartum in one patient, and coincident with the discontinuation of oral contraceptives in the remaining five patients. Complete physical examinations and visual field determinations were normal in all of the patients, except one who was congenitally blind. Base line hormone

Vol. 29, No.3 PROLACTIN-SECRETING PITUITARY MICROADENOMA 283 evaluations were also normal, with the exception of hyperprolactimemia in all of the patients. A transsphenoidal hypophysectomy with selective removal of a pituitary microadenoma (histologically confirmed) was performed in all of the patients. The relevant clinical aspects and base line laboratory data are summarized in Table 1. Radiologic Evaluation The sella turcica was examined by standard plain x-ray in all eleven patients and by routine tomography in seven of the patients. Poly tomograms of the sella were obtained in nine of the patients. The polytomograms were obtained at 2- mm intervals in the anteroposterior and lateral projections with the use of hypocycloidal movement. This newer radiologic technique allows much better clarity than was previously possible with routine tomography and allows the detection of subtle changes in the sella that may not be evident on routine tomographic and plain sella films.2 Radioimmunoassay Serum prolactin, 4 follicle-stimulating hormone (FSH),5 luteinizing hormone (LH),5 thyroidstimulating hormone (TSH),5 and growth hormone (GH)6 concentrations were determined by radioimmunoassay. Serum thyroxine was assayed by competitive protein-binding radioassay.7 Plasma cortisol levels were measured by competitive protein binding. 8 Laboratory ranges for women are as follows: follicular phase FSH, 4.6 to 22 miu/ml; follicular phase LH, 6.5 to 24.2 miulml; prolactin, 0 to 20 ng/ml; TSH, 0 to 10 miulml; GH, less than 2 /Lg/ml; T4, 4.6 to 10.7 /Lg/100 ml; RT 3 uptake, 25% to 35%; and cortisol, 5 to 20 /Lg/lOO ml. Studies of Pituitary Function Gonadotropic Hormone Reserve. LH-releasing factor (LRF; AY-24031) (100 /Lg) was given subcutaneously, and serum LH was assayed at -30, 0, 15, 30, 60, 120, and 240 minutes. Growth Hormone Reserve. After an overnight fast and 90 minutes following the placement of an indwelling venous catheter, regular insulin was injected intravenously at a dose of 0.1 unitlkg of body weight; serum samples were then obtained for growth hormone and glucose determinations at 15-minute intervals for 90 minutes. Hypothalamic-Pituitary-Adrenal Axis. To assess the hypothalamic-pituitary-adrenal axis, plasma cortisol concentrations were determined at 0, 60, and 90 minutes following the administration of insulin in nine of the eleven patients in whom growth hormone secretion was studied. Prolactin Study. Ninety minutes following the placement of an indwelling venous catheter, L dopa in an oral dose of 10 mg/kg of body weight was administered. Plasma prolactin was assayed at -30, 0, 30, 60, 90, 120, 180, and 240 minutes. Radiology RESULTS Plain sella turcica x-rays were reported normal in all eleven patients. Minimal asymmetry of the sella floor was evident on plain films in five of the patients. Routine tomography, performed in seven of the patients, was normal in five and abnormal in the remaining two patients. In one of these patients, initial routine tomography was normal; however, subsequent routine tomography 8 months later was abnormal. Polytomography, performed in nine of the patients, revealed definitive localized erosion of the sella floor with TABLE 1. Pituitary Microadenomas: Base Line Laboratory Data Patient Age Duration of amenorrhea Galactorrhea Prolactin FSH LH T. Cortisol (A.M.IP.M.l yr nglml mlulml mlu/ml 1'81100 ml 1'81100 ml L.S. 27 2 + 970 8.5 10.5 7.1 18.5/- J.G. 19 1 490 6.4 7.8 8.1 1414 C. P. 37 6 + 259 11.6 10.9 8.8 B.H. 19 Primary 253 9.5 7.8 5.0 15/- L.D. 27 7 + 229 4.0 7.0 5.0 2118 D.H. 21 3 + 225 9.7 13.4 7.9 21195 S. L. 27 51h + 221 14.0 15.0 10.4 1215.5 M.H. 27 2 150 14.1 26.4 6.2 19.6/- J. S. 31 3 + 105 12.4 14.2 8.7 15/5 N.T. 33 5 + 102 14.1 12.3 6.3 12/- B.M. 29 2 + 90 10.4 9.6 7.6 16.6/4 Normal laboratory values 0-20 4.6-22.0 6.5-24 4.6-10.7 5-20

284 WIEBE ET AL. March 1978 180 100 160 E "- 140 120 E 100 :J: -' Cf) 80 60 40 20 Time (minufes) 300 FIG. 1. Effect of LRF on serum LH levels in 11 women with prolactin-secreting pituitary microadenomas. e--e, Changes in serum LH levels from base line following the subcutaneous administration of 100 f.lg of LRF. The shaded area represents the range of response in 10 normal women studied in the early follicular phase. or without blister formation in eight patients, while asymmetry with questionable thinning of the sella floor was noted in one of the patients. Base Line Hormone Evaluation As is shown in Table 1, RT3 uptake, TSH levels, and cortisol levels were normal in all of the patients. Base line serum LH levels were normal in all of the patients, while base line serum FSH levels were normal in ten patients and depressed in one patient. Base line prolactin levels varied considerably in individual women, with wide fluctuations. The mean of the multiple prolactin levels obtained in individual patients c 0 <5 ct '" c 3i 0 (D '0 50 a LS DH SL JG CP NT BH MH LDBMJS FIG. 2. Effect of oral L-dopa (10 mg/kg of body weight) on plasma prolactin levels in 11 women with prolactin-secreting pituitary microadenomas. The responses are expressed as percentage change from the mean base line values obtained immediately prior to testing and are the nadirs of a progressive decline in plasma prolactin levels determined over a 4-hour period. varied between 90 and 970 ng/ml. A reasonable correlation between the size of the tumor and the mean base line level of prolactin was evident. Studies of Pituitary Function Hypothalamic-Pituitary-Adrenal Axis. A normal plasma cortisol response to insulin-induced hypoglycemia was noted in all nine women tested. Growth Hormone Reserve. In ten of the patients, growth hormone secretion and the response to insulin-induced hypoglycemia was normal according to the criteria of Eddy and associates. 9 One patient had no response. Gonadotropin Reserve. As is shown in Figure 1, nine patients had normal responses to LRF, while two patients had responses which were borderline low. Prolactin Suppression. As is shown in Figure 2, plasma prolactin concentrations decreased greater than 50% from base line in seven pa- Patient TABLE 2. Prolactin-Secreting Microadenomas: Summary of Pituitary Function Tests Prolactin Plain sella Routine Poly tomography GH reserve ACTH reserve LRF response L-Dopa films tomography suppression ng/ml L. S. 970 Normala Abnormal Normal Borderline None J.G. 490 Normala Abnormal No response Normal Normal Borderline C.P. 259 Normal" Abnormal Normal Normal Normal Normal B.H. 253 Normal Normal Abnormal Normal Normal Normal Normal L.D. 229 Normal Normal Abnormal Normal Normal Borderline Normal D.H. 225 Normal a Asymmetry Asymmetry Normal Normal Normal None S. L. 221 Normala Abnormal Normal Normal Normal Minimal M.H. 150 Normal Normal Abnormal Normal Normal Normal Normal J. S. 105 Normal Normal Abnormal Normal Normal Normal N.T. 102 Normal Abnormal Normal Normal Normal Normal B.M. 90 Normal Abnormal Normal Normal Normal Normal aminimal asymmetry.

Vol. 29, No.3 PROLACTIN-SECRETING PITUITARY MICROADENOMA 285 tients receiving L-dopa. However, L-dopa failed to suppress prolactin secretion in two of the patients, only minimal suppression (83%) was noted in one patient, and the magnitude of suppression was borderline (62%) in the remaining patient. A summary ofthe diagnostic studies is shown in Table 2. DISCUSSION Secondary amenorrhea, with or without galactorrhea, was the only clinical abnormality in our patients. Base line evaluations were normal with the exception of hyperprolactinemia. The finding of hyperprolactinemia in three of the patients without galactorrhea emphasizes the importance of measuring plasma prolactin in all women with unexplained amenorrhea. The demonstration of pituitary microadenomas in these eleven women, despite normal sella turcica x-rays, raises the important question of the frequency of this disorder. These pituitary tumors were found only after further diagnostic studies were conducted. Our findings emphasize the importance of an extensive evaluation of all women with amenorrhea associated with hyperprolactinemia. That all of the women with abnormal polytomograms had normal plain x-rays of the sella attests to the utility of polytomography in the diagnosis of these tumors. Furthermore, the subtle changes noted on polytomography were not detected on routine tomography in five of seven women. It is possible that, had poly tomography been available, the radiologic diagnosis of a pituitary adenoma may have been made earlier in the two women with positive routine tomography. Our results would indicate that polytomography is of value in the early diagnosis of pituitary microadenoma and may be positive prior to any associated trophic loss. Although polytomography is probably the most sensitive diagnostic technique available, caution should be exercised in interpretation, as variations in the normal anatomy of the sella turcica may be misinterpreted as pathology. Definitive bony erosion should be demonstrated in order to make the radiologic diagnosis. Iftomograms are suspicious and not definitive, then repeat studies, with proper ocular shielding, should be performed at 6- to 12-month intervals to determine whether there have been any progressive changes. Furthermore, to avoid diagnostic errors, polytomography should not be utilized alone, but in conjunction with other diagnostic tests. Although absent or decreased reserves of growth hormone and adrenocorticotropic hormone and a decreased LH response to LRF in patients with hyperprolactinemia may indicate significant pituitary compromise and aid in the diagnosis of pituitary tumors, our results would indicate that these diagnostic studies are oflimited value in the early diagnosis of pituitary microadenomas. Since L-dopa increases dopamine levels in the hypothalamus, increasing the prolactin-inhibiting factor and thus decreasing prolactin secretion, it has been utilized as a clinical test for evaluation of prolactin secretion. 10 Recent investigations in laboratory animals have demonstrated a direct effect of dopamine on the pituitary lactotroph, suggesting that dopamine, in fact, may be the prolactin-inhibiting factory Therefore, a failure to respond to L-dopa should indicate autonomous prolactin secretion. Following L-dopa administration, eight of our patients showed significant suppression of prolactin, thereby suggesting that some of these tumors are not autonomous and retain some degree of responsiveness. Therefore, significant suppression of prolactin following L-dopa administration does not rule out a pituitary adenoma. However, a failure to respond, as was demonstrated in three of our patients, would suggest autonomous production of prolactin and therefore may be indicative of a pituitary adenoma. This diagnostic test was of particular value in one of our patients, as there was no definitive radiologic evidence of a microadenoma in this patient. The magnitude of the prolactin elevation in this patient (Le., over 200 ng/ml), combined with asymmetry of the sella floor on poly tomography and a failure to suppress prolactin with L-dopa, made the diagnosis of pituitary adenoma quite likely. Plasma prolactin levels may vary markedly in individual patients, and therefore it is difficult to assign a specific plasma level of prolactin that is diagnostic of pituitary tumors unless multiple levels are obtained. To the present time, all patients whom we have evaluated with prolactin levels over 300 ng/ml have had demonstrable pituitary tumors. Obviously, a very large series of patients will be required before we can definitively assign a plasma prolactin level that is diagnostic of a pituitary adenoma. In addition, continued follow-up of patients in whom prolactin levels are below 300 ng/ml may allow us to lower the level that is diagnostic of a pituitary adenoma. As three of our patients had prolactin levels in the range of 100 ng/ml, we presently view any

286 WIEBE ETAL. March 1978 patient with a prolactin level of over 50 ng/ml as possibly having a tumor. In summary, the early diagnosis of a prolactinsecreting micro adenoma may be possible if several diagnostic criteria are utilized in conjunction. It would appear that the most sensitive diagnostic indices available are (1) poly tomography, which frequently is positive before any evidence of associated hormone loss; (2) the magnitude of the plasma prolactin elevation; and (3) the failure to suppress prolactin secretion with L-dopa. REFERENCES 1. Bohnet HG, Dahlen HG, Wutke W, Schneider HPG: Hyperprolactinemic anovulatory syndrome. J Clin Endocrinol Metab 42:132, 1976 2. Vezina JL, Sutton TJ: Prolactin-secreting pituitary microadenoma. Am J Roentgenol 120:46, 1974 3. Wiebe RH, Hammond CB, Borchert LG: Diagnosis of prolactin-secreting microadenomas. Am J Obstet Gynecol 126:993, 1976 4. Sinha ZN, Selby FW, Lewis VJ, Vanderlaan WP: A homologous radioimmunoassay for human prolactin. J Clin Endocrinol Metab 36:509, 1973 5. Odell WD, Rayford PL, Ross GT: Simplified, partially automated method for radioimmunoassay of human thyroid-stimulating, growth, luteinizing, and folliclestimulating hormones. J Lab Clin Med 70:973, 1967 6. Frohman LA, Horton ES, Lebovitz HE: Growth hormone releasing action of a Pseudomonas endotoxin (piromen*). Metabolism 16:57, 1967 7. Murphy BEP, Pattee CJ, Gold A: Clinical evaluation of a new method for the determination of serum thyroxine. J Clin Endocrinol Metab 26:247, 1966 8. Murphy BEP: Some studies of the protein-binding steroids and their application to the routine micro and ultra micro measurement of various steroids in body fluid by competitive protein binding radioassay. J Clin Endocrinol Metab 27:973, 1967 9. Eddy RL, Gilliland PF, Ibarra JD, McMurray JF, Thompson JQ: Human growth hormone release. Comparison of provocative test procedures. Am J Med 56:179, 1974 10. Friesen H, Guyda H, Hwang JE, Tyson JE, Barbeau A: Functional evaluation of prolactin secretion: a guide to therapy. J Clin Invest 51:706, 1972 11. MacLeod RM, Lehmeyer JE: Studies on the dopaminemediated inhibition of prolactin secretion. Endocrinology 94:1077, 1974