Differential response to adrenocorticotropin hormone stimulation in polycystic ovarian disease with high and low dehydroepiandrosterone sulfate levels

Size: px

Start display at page:

Download "Differential response to adrenocorticotropin hormone stimulation in polycystic ovarian disease with high and low dehydroepiandrosterone sulfate levels"

Cynthia Gilbert
5 years ago
Views:

1 FERTILITY AND STERILITY Copyright C> 1982 The American Fertility Society PrintmJ. in U.SA. Differential response to adrenocorticotropin hormone stimulation in polycystic ovarian disease with high and low dehydroepiandrosterone sulfate levels Jonathan W. T. Ayers, M.D. Department of Obstetrics and Gynecology, University of Michigan Medical Center, Ann Arbor, Michigan The relative contributions of the ovary and the adrenal gland to androgen overproduction in polycystic ovarian disease (PCOD) remain controversial. In this investigation, patients with proven PCOD were divided into two groups, (1) those with low dehydroepiandrosterone sulfate (DHEAS) levels and (2) those with high DHEAS levels, and compared with controls for their response to adrenocorticotropin hormone (ACTH) stimulation. Significant differences in weight, degree of menstrual disturbance, and basal progesterone levels distinguished the two groups with PCOD. Although no discrete enzyme block was unmasked by ACTH, marked differences in steroid production ratios were apparent between the low and high DHEAS PCOD groups. These results suggest that in PCOD with high DHEAS (1) substantial differences in adrenal steroidogenesis pathways occur, (2) increased progesterone as well as inappropriate estrogen feedback may contribute to chronic anovulation, and (3) serum DHEAS levels may be a helpful screen in discerning those patients who have a significant adrenal component to their hyperandrogenism and may benefit from adrenal suppression alone or in combination. Fertil Steril37:645, 1982 In the 45 years since the original description of Stein-Leventhal ovarian syndrome 1 the clinical spectrum of polycystic ovarian disease (PCOD) has considerably broadened. Ingerso1l 2 and Gallagher et a1., 3 in the late 1950s, postulated an adrenal component to this hyperandrogenic state. Since then, numerous investigators have subclassified PCOD on the basis of serum luteinizing hormone (LH) levels, 4 menstrual disturbances, 5 ovarian size, 6 serum androgen levels, 7 and dynamic tests of suppression and stimulation. 8, 9 Although the relative contributions of adrenal and ovarian androgens in PCOD remain an un- Received September 8, 1981; revised and accepted January 5,1982. Reprint requests: Jonathan W. T. Ayers, M.D., Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio answered question, most investigators agree that perturbations in adrenal steroidogenesis play a role in chronic anovulation. While these frequently subtle abnormalities in adrenal function may not be detected by random measurements, they may be enhanced by maximal stimulation of the adrenal gland. 10 This study was designed to investigate the adrenal responsiveness to physiologic adrenocorticotropin hormone (ACTH) stimulation in patients with PCOD. Of particular interest was a possible differential response in patients with PCOD with and without evidence of preexisting hyperadrenalism,11, 12 as assessed by serum dehydroepiandrosterone sulfate (DHEAS)-the single most accurate measurement of C-19 adrenal androgen production. 13 Five key steroid hormones-progesterone (P), 17-hydroxyprogesterone (17-0H-P), cortisol (F), testosterone (T), and DHEAS were measured before and after ACTH administration. Ayers Adrenal DHEAS in PCOD 645

2 MATERIALS AND METHODS Thirty-eight patients presenting with hirsutism and/or menstrual disturbances were included, with their consent, in the study protocol. PCOD was diagnosed by (1) monophasic basal body temperatures, (2) sclerocystic ovaries present at laparoscopy, (3) normal estrogen and prolactin (PRL) levels, and (4) absence of any other endocrinopathy. All patients had a positive response to medroxyprogesterone withdrawal. Twenty patients had basal serum levels of DHEAS over 4000 ng/ml (normal, ng/ ml}-"high DHEAS" PCOD; 18 patients had levels within the normal range-"low DHEAS" PCOD. Five normal, regularly cycling, healthy women served as a control population. All subjects were nulligravid and free from medication for at least 6 months. Cushing's disease was retrospectively excluded by normal 24-hour urinary F levels, and low-dose dexamethasone suppression of F and DHEA8. Subjects with normal or oligomenorrheic cycles were studied 5 to 8 days after the onset of bleeding. Nonluteal phase evaluation was assured by basal temperatures and serum P levels. Subjects were studied fasting between 8:00 A.M. and 10:00 A.M. while recumbant. Thirty minutes after placing a heparin lock, we obtained two samples 15 minutes apart and pooled them for bas line values. ACTH (Cortrosyn 250 IJ.g) (Organon, West Orange, NJ) was injected, and we obtained pooled stimulated samples at 30 and 45 minutes to avoid single sample error. Steroid, gonadotropin, and PRL determinations were performed by radioimmunoassay by the use of the double-antibody technique and commer- cially available reagents. For statistical analysis the Student's t-test, with analysis of variance, was used. BASAL STUDIES RESULTS Clinical presentations of high and low DHEAS patients differed markedly as to weight and degree of menstrual disturbances (Table 1). PCOD patienti'! with low DHEAS tended to be heavier and weighed more at menarche, although menarchal age did not differ. Seventy-eight percent of low DHEAS patients were amenorrheic, but only 30% of high DHEAS patients had completely stopped menstruating. Hirsutism was the presenting complaint in only half as many of low versus high DHEAS patients. Basal serum levels of follicle-stimulating hormone '(FSH), LH, and estradiol (E 2 ) were not significantly different between the three groups. PRL values tended to be higher (P < 0.1) in the high DHEAS group, although certainly within normallirnits (4 to 22 ng/ml). ACTH STIMULATION The basal levels and stimulated responses to ACTH of the five key steroids are depicted in Figure 1. DHEAS Significant basal and stimulated value differences were noted in all three groups. Paradoxically, the high DHEAS patients failed to show a significant response to ACTH. Table 1. Clinical Data Number patients Weight (kg) % Ideal body weight Age at menarche (yr) Weight at menarche (kg) Menses Regular Oligomenorrhea Amenorrhea Hirsutism Laboratory Data Luteinizing hormone (IU/mI) Follicle~stimulating hormone (IU/mI) Estradiol (pglmi) Prolactin (nglmi) "Significant at P = bp = Ayers Adrenal DHEAS in PCOD Control /5 (100%) o 2.1 ± ± ± ± 2.1 LowDHEAS HighDHEAS " " /20 (10%) 4/18 (22%) 12/20 (60%) (78%) 6/20 (30%) 5/18 (28%) 12/20 (60%) 4.0 ± ± ± ± ± ± ± ± 4.1 b Fertility and Sterility

3 ~ 300 '" c:: % 24% 130% 95%* 52% * * p<0.5 CD Normal CD Law DHEAS C.HillhDHEAS 158% 98% 200 j * 20 jjj I 100 jjj 1~i-'~,s I I 5 1** CorllOGI 17 0ltProQ8S1erono Figure 1 Baseline and stimulated values of the five steroids to ACTH stimulation. The data represent the mean values of the three groups. Statistical significance is based upon standard deviations of each sampled group. Normal values (darkened bars) are laboratory reference standards. Testosterone The high and low DHEAS groups had equally elevated basal values over normal controls. No significant stimulated change separated the two groups with PCOD. Progesterone The most consistent and striking variable between the three groups was the P response. Control and low DHEAS groups had basal P levels well within the normal follicular phase range and doubled their values upon ACTH stimulation. High DHEAS patients consistently had basal P levels twice as high as the other two groups, and similar to their stimulated DHEAS values, showed an attenuated response to ACTH. Cortisol No significant difference could be demonstrated between the three groups in either basal or stimulated F values. A normal twofold rise in serum F levels was noted with ACTH infusion in all subjects. 17 -Hydroxy progesterone Elevated basal values of 17-0H-P were seen in both groups with PCOD and failed to distinguish between high and low DHEAS groups. These values were at the upper limits of normal (35 to 78 ng/md for the follicular phase but were significantly increased over controls. High DHEAS subjects had a much higher response to ACTH than low DHEAS subjects. The ratio of incremental change between basal and stimulated levels of paired steroids, a mea- sure of adrenal responsiveness to ACTH, are shown in Figure 2. In general, patients with PCOD of both types exhibited higher androgen production by ACTH stimulation than normal women (Fig. 2d). Production of F from androstenedione (.!l4) precursors was similar in both groups of PCOD and in normal controls (Fig. 2c). However, comparisons of.!l4 and androstenediol (.!l5) steroids (Fig. 2a and b) revealed uniformly significant differences in incremental ratios between high and low DHEAS PCOD patients. While.!l4 to.!l5 pathway interconversion ratios were similar in normal women and low DHEAS patients, the high DHEAS PCOD patients produced much higher ratios of.!l4 steroids (P and F) per increment change in.!l5 steroid (DHEAS). Comparison of these incremental ratios shows a marked difference in steroidogenic responses between high and low DHEAS PCOD. DISCUSSION The validity of the ACTH stimulation test in the evaluation of adrenal function in PCOD has been documented. 14 Other investigators have found such dynamic testing to uncover subtle abnormalities in adrenal steroidogenesis that basal levels alone fail to reveal It * * I t o t.p/t.dheas t.f/t.dheas t.f/t.p(xiol A B c * Jj" Normal o Low DHEAS.Hlgh DHEAS *p<.05 t.tit.f Figure 2 Ratios of increments of selected steroid pairs. Values are the means of each group calculated by: (stimulated value1hbasal valuel)l(stimulated value2)-(basal value2)' Significant difference in 11 5 _11 4 interconversion between high and low DHEAS PCOD (see text). Ayers Adrenal DHEAS in peod 647 D

$PRODUCTION \ INAPPROPRIATE / \ DHEAS ~ ESTROGEN/ FEEDBACK (PERIPHERAL t ADRENAL AROMATIZATlON) ANDROGENS PROGESTERONE 1) Partial Enzyme Block? 2) Altered Adrenal Steroidogenic Dynamics?$

4 CHRONIC ANOVULATION ("POLYCYSTIC OVARIAN DISEASE") GONADOTROPHIN DESYNCHRONIZAIION (tlh tfsh) OVARIAN ("POL YCYSTIC / STROMA OVARIES") ~ STIMULATION HYPOTHALAMIC-PITUITARY I DYSFUNCTION t ANDROGEN PRODUCTION \ INAPPROPRIATE / \ DHEAS ~ ESTROGEN/ FEEDBACK (PERIPHERAL t ADRENAL AROMATIZATlON) ANDROGENS PROGESTERONE 1) Partial Enzyme Block? 2) Altered Adrenal Steroidogenic Dynamics? 3) CRF - ACTH Dependent? 4) Alterations 01 Prolactin Metobolism? Figure 3 Postulated mechanism of chronic anovulation in high DHEAS PCOD. (1) Increased adrenal androgens may be peripherally converted, generating inappropriate estrogen feedback. (2) Increased basal, follicular levels of progesterone may act directlyon the hypothalamic/pituitary axis to suppress the normal gonadotropin sequence. It is tempting to postulate a partial enzymatic defect as the cause of hyperandrogenism in PCOD. While there is tentative evidence of partial 21-, ll-hydroxylase l6 and 3~-hydroxysteroid dehydrogenase l8 and isomerase deficiency in some hyperandrogenic women, unequivocal proof of a classic enzyme block is lacking. Gibson et al. l9 have presented evidence suggesting frequent occurrence of deficiencies in several hydroxylase enzymes in hirsute women when challenged with ACTH. However, there was a considerable overlap in adrenal responsiveness between study and control subjects. \ Lachelin et al. l4 concluded from their PCOD ACTH responses that there was no evidence for a primary enzyme block. Rather, selected abnormal respollses in several androgens were due to functional changes in steroidogenesis induced by the hyperandrogenic milieu itself. However, they did note increases of DHEA, P, and 17-0H-P consistent with a mild 21- or ll-hydroxylase deficiency under sustained ACTH infusion. In the present study of high and low DHEAS PCOD, ACTH did not uncover an overt enzyme block. Serum F response to ACTH stimulation was appropriate in both groups ofpcod and similar to control subjects. Elevations in basal T and androgenic precursors, notably 17-0H-P, were present in both groups, as has been described.5, 14, 20 Whether this "androgen backup" is secondary to partial deficiencies in the 21-, ll-hydroxylase system or to functional disturbances in steroidogenesis cannot be ascertained 648 Ayers Adrenal DHEAS in PCOD from these data. However, both high and low DHEAS PCOD groups show abnormal responses in both stimulated and basal states, suggesting a physiologic aberration in steroid production. It seems that maintenance of normal F production in both types of PCOD is at the expense of increased levels of androgens. Further scrutiny of the data reveals discrete, uniform differences between the two types of PCOD. Clinically, the low DHEAS PCOD patients tended toward the classic Stein-Leventhal l presentation: obese, hyperandrogenic, and amenorrheic. High DHEAS PCOD patients in contrast were: (1) thinner, both at menarche and present; (2) oligomenorrheic rather than amenorrheic; (3) complaining more of hirsutism than menstrual disturbances; and (4) found to have higher serum PRL levels. Of note, neither gonadotropin nor E2 levels were statistically different in the two groups. Basal levels ofdheas and P were much higher in the high DHEAS than low DHEAS group and showed an attenuated response to ACTH stimulation. This finding may suggest that the adrenal steroidogenesis in the high DHEAS PCOD patient is under tonic hyperstimulation in order to maintain normal F production. Future investigation of serum ACTH and metapyrone blockade might be useful in exploring this attractive postulate. Mechanisms that might account for this "basal hyperandrogenism" in high DHEAS PCOD include (1) a more significant 21-, ll-hydroxylase deficiency requiring higher endogenous ACTH drive for normal F; (2) alterations in PRL metabolism with concomitant increases in DHEAS 2l, 22; or (3) a more pronounced functional change in adrenal steroidogenesis due to increased levels of androgens (DHEAS) at the cellular level. Elevated P levels, the most consistent discrepancy between high and low DHEAS groups, suggest a further differentiation of the two groups of PCOD. It has been shown that midfollicular P administration can ablate the LH surge and ovulation through central gonadotropin-releasing hormone (GnRH) suppression. Tonically elevated P, in high DHEAS patients, might enter the gonadotropin desynchronization cycle and produce anovulation through a different mechanism than Stein-Leventhal (low DHEAS) PCOD (Fig. 3). In summary, tw<;l groups ofpcod patients have been defined on the basis of their DHEAS levels. Clinical presentation, basal and ACTH-stimulated steroid values, and comparison of the ratio of Fertility and Sterility

5 increments reveal two distinct types of PCOD: low DHEAS, the classic Stein-Leventhal PCOD', and high DHEAS, the hyperandrogenic. PCOD with tonically elevated ~4, ~5 steroids, P, and DHEAS. Distinction between these two groups is of great clinical importance in. choosing the most appropriate therapy for ovulation induction or androgen suppression. While interruption of inappropriate estrogen feedback with clomiphene may be the therapy of choice for low DHEAS PCOD; adrenal suppression alone, or in combination may be a more physiologic therapy for the tonic hyperadrenalism in the high DHEAS PCOD patient. REFERENCES 1. Stein IF, Leventhal ML: Amenorrhea' associated with polycystic ovaries. Am J Obstet Gynecol 29:181, Ingersoll FM: Differential diagnosis of Stein-Leventhal syndrome (polycystic ovarian syndrome). In Progress in Gynecology, Edited by EJ Meigs, SH Sturgis. New York, Gruen & Stratton, 1957, P Gallagher TF, Kappos A, Hellman L, Lipsett MB, Pearson OH, West CD: Adrenocortical hyperfunction in "idiopathic" hirsutism and the Stein-Leventhal syndrome.. J Clin Invest 37:794, Givens JR, Andersen RN, Vurstot ES, Wiser WL: Clinical findings and hormonal responses in patients with PCOD. with normal versus elevated LH levels. Obstet Gynecol 47:388, Abraham GE, Chakmakjau ZH, Buster JE, Marshall JR: Ovarian and adrenal contributions to peripheral androgens in hirsute women. Obstet Gynecol 46:169, Shailaja GR, Thompson IE, Berger MJ, Taymor ML: Clinical aspects of the polycystic ovary syndrome. Obstet Gynecol 49:552, Shailaja GR, Thompson IE, Berger MJ, Talert LM, Taymor ML: Diagnostic value of androgen measurements in polycystic ovary syndrome. Obstet GynecoI52:169, Lisse K, Schurenkamper P, Friedrich W, Rutkowsky J: Diural change of serum androstenedione and testosterone and response to hcg and dexamethasone in women with polycystic ovaries, adrenal hyperandrogenism, and unexplained hirsutism. Acta Endocrinol (Copenh) 93:216, Bardiu CW, Hembree WC, Lipsett MB: Suppression of testosterone and' androstenedione production rates with dexamethasone in women with idiopathic hirsutism and polycystic ovaries. J Clin Endocrinol Metab 28:1300, Newmark S, Dluhy RG; Williams GH, PochiP, Rose LI: Partial and 21-hydroxylase deficiencies in hirsute women. Am J Obstet Gynecol 127:594, Abraham GE, Maroulis GB, Buster JE, Chang J, Marshall JR: Effect of dexamethasone on serum cortisol and androgen levels in hirsute patients. Obstet Gynecol 47: 395, Abraham GE: Ovarian and adrenal.contribution to peripheral androgens during the menstrual cycle. J Clin Endocrinol Metab 39:340, Lobo RA; W1!llington LP, Goebelsmann U: DHEAS as an indication of adrenal androgen function. Obstet Gynecol 57:69, Lachelin GeL, Barwell M, Hopper BR, Bunk G, Yen SSC: Adrenal function in normal women and women with the polycystic ovary syndrome. J Clin Endocrinol Metab 49: 892, GivensJR, Anderson RN, Rogland JB, Wiser WL, Vinstot ES: Adrenal function in hirsutism: I. Diurnal change and response of plasma androstenedione, testosterone, 17 -hydroxyprogesterone, cortisol, LH, and FSH to dexamethasone and 1/2 unit of ACTH. J Clin Endocrinol Metab 40: ' 16. Newmark S, Dluhy RG, Williams GH, Pochi P, Rose LI: Partial 11- and 21-hydroxylase deficiencies in hirsute women. Am J Obstet Gynecol 127:59{, ~ Halperin G, Maschler I: Isolation of urinary c-20- and C-20-hydroxy-C-21 steroid metabolites in cases of congenital adrenal hyperplasia, postpubertal virilizing syndrome, and polycystic ovary syndrome. Steroids 33: ' 18. Axelrod CR, Goldzieher JW, Ross SD: Concurrent 3-hydroxysteroid dehydrogenase deficiency in adrenal and sclerocystic ovary. Acta Endocrinol (Copenh) 48:392, Gibson M, Lackritz R, Schiff I, Tulchinsky D: Abnormal adrenal responses to adrenocorticotropic hormone in hyperandrogenic women. Fertil Steril 33:43, Sojo-Arana I, Carrano-Lopez A, Cortes-Gallegos V: The polycystic ovary: hirsutism 'and 171i-hydroxyprogesterone. Fertil Steril 31:687, Vermeulen A, Suy E, Eubens R: Effect of prolactin on plasma DHEA(S) levels. J Clin Endocrinol Metab 44: 1222, C~r IN, Tyson JE, Warne GL, McNeilly AS, Faiman C,. Frieson HG: Adrenocortical function in hyperprolactinemic women. J Clin Endocrinol Metab 45:973, Schuiling GA, Pols-Valkhof N, Zurcher AF: Some observations concerning the role of progesterone in the control of ovulation. Acta Endocrinol (Copenh) 94:246, Zeilmaker GH: The.biphasic effect of progesterone on ovulation in the rat. Acta Endocrinol (Copenh) 51:461, Dierschke DJ, Yamaji T, Karsch FJ, Weick RF, Weiss G, KnobilE: Blockade by progesterone of estrogen-induced LH and FSH release in th1! rhesus monkey. Endocrinology 92:1496, 1973 Ayers Adrenal DHEAS in peod 649

ROLE OF HORMONAL ASSAY IN DIAGNOSING PCOD DR GAANA SREENIVAS (JSS,MYSURU)

ROLE OF HORMONAL ASSAY IN DIAGNOSING PCOD DR GAANA SREENIVAS (JSS,MYSURU) In 1935, Stein and Leventhal described 7 women with bilateral enlarged PCO, amenorrhea or irregular menses, infertility and masculinizing