In recent publications (1 3), we have reported the changes in

Transcription

1 ORIGINAL Endocrine ARTICLE Care Interrelationships between Ovarian and Pituitary Hormones in Ovulatory Menstrual Cycles across Reproductive Age David M. Robertson, Georgina E. Hale, Damien Jolley, Ian S. Fraser, Claude L. Hughes, and Henry G. Burger Prince Henry s Institute of Medical Research (D.M.R., H.G.B.), Clayton, Victoria 368, Australia; Department of Obstetrics and Gynaecology (G.E.H., I.S.F.), Queen Elizabeth II Research Institute for Mothers and Infants (DO2), University of Sydney, New South Wales 26, Australia; Monash Institute of Health Services Research (D.J.), Monash University, Clayton, Victoria 368, Australia; and Quintiles Inc. (C.L.H.), Research Triangle Park, North Carolina 2779 Context: Ovarian hormones regulate pituitary gonadotropin secretion across the menstrual cycle via negative and positive feedback mechanisms. The contribution of individual hormones is complex and is a continuing area of research. Objective: The aim of the study was to identify relationships between LH/FSH and estradiol, progesterone, inhibin A, inhibin B, and anti-mullerian hormone () in ovulatory menstrual cycles across reproductive age. Design: Serum ovarian and pituitary hormones were studied in a group of young ( 3 yr; n 2) and older ( 4 yr; n ) women. The slopes of the regression lines relating the ovarian and pituitary hormones were determined by multiple linear regression analysis and expressed with 9% confidence intervals for each ovarian hormone, with FSH and LH as independent variables. Both simultaneous and delayed (time lagged) relationships were examined. Results: Clear associations were evident for the lagged prediction of FSH, with significant negative associations being evident with inhibin B and in the follicular phase and with estradiol, inhibin B, progesterone, and in the luteal phase. For the lagged prediction of LH, significant positive and negative associations were observed with estradiol and inhibin B, respectively, in the follicular phase and a negative association with progesterone and inhibin B in the luteal phase. Conclusions: It is concluded that in the follicular phase, inhibin B is a major feedback regulator of FSH and may also be a negative feedback regulator of LH. may be indirectly involved in FSH regulation. (J Clin Endocrinol Metab 94: 38 44, 29) In recent publications ( 3), we have reported the changes in serum FSH, LH, estradiol, progesterone, inhibin A, inhibin B, and anti-mullerian hormone () levels throughout the menstrual cycle in women in the middle (age, 2 3; n 2) and later (age, 4 ; n 6) reproductive years. It was concluded that changes in hormonal patterns with age are a consequence of the age-related decline in ovarian follicle reserve, causing a decrease in ovarian factors (e.g. inhibin B) that are critical in the regulation of ovary:pituitary feedback and a secondary decline in luteal function. In some ovulatory cycles in the menopause transition, ISSN Print 2-972X ISSN Online Printed in U.S.A. Copyright 29 by The Endocrine Society doi:.2/jc Received August, 28. Accepted October 7, 28. First Published Online October 4, 28 a rise instead of a fall in estradiol during the mid and late luteal phases was observed (2). In light of human ovarian follicle data from Baerwald et al. (4), it was postulated that this atypical estradiol secretion pattern reflects the recruitment of a new dominant follicle during the mid-luteal phase of a cycle in which ovulation has already occurred. The large database available (77 cycles, seven hormones, time points) permits a comprehensive analysis with the basic tenet being that age-related hormone changes in ovarian and pituitary hormone levels are primarily attributable to a decrease Abbreviations:, Anti-Mullerian hormone; CI, confidence interval. 38 jcem.endojournals.org J Clin Endocrinol Metab. January 29, 94():38 44 on 3 December 27

2 J Clin Endocrinol Metab, January 29, 94():38 44 jcem.endojournals.org 39 in ovarian feedback on the pituitary. Use of data from cycles in young and older reproductive-aged women was predicted to reveal relationships as ovarian feedback declined because of decreasing ovarian reserve, analogous to the analysis of TSH/T 4 relationships in primary hypothyroidism. Using multiple linear regression analysis, our objective was to explore: ) independent relationships between serum levels of ovarian hormones (estradiol, progesterone, inhibins A and B, ) and both pituitary hormones (FSH and LH) as assessed in simultaneously measured samples; 2) independent relationships between serum levels of ovarian hormones (estradiol, progesterone, inhibins A and B, ) and both pituitary hormones (FSH and LH) after the application of a 3-d time-lag; and 3) which of the ovarian hormones best predicts changes in FSH and LH as a basis for assessing their role in the feedback regulation of the two gonadotropins. We have concluded that inhibin B is most likely the primary ovarian factor regulating FSH and possibly LH, whereas steroids exert their action in a positive and negative fashion primarily on LH. is negatively associated with FSH; however, it is not believed to contribute directly to the feedback regulation of FSH. approach, the effects of pituitary gonadotropins on ovarian hormones 3 d later were explored. Statistical analyses The lack of symmetry in cross-sectional distributions of hormone levels was accommodated using a logarithmic transformation for each hormone measurement. When both variables in a regression are logtransformed, the regression slope coefficient as a power coefficient for the underlying raw variables can be interpreted as: log (y) a b log (x) 7 y Ax b, where log (A) a. The slope coefficient, b, from log-log regression represents the magnitude of relative change in y associated with a relative change in x. The slope coefficient, b, does not depend on the units of measurement of either predictor x or outcome y; hence, comparisons of magnitude between slope coefficients from different predictor variables are valid. Random-effects repeated-measures multivariable linear regression models were used to estimate the strength of relationships between hormone levels and time and between cross-sectional hormone values. To model the temporal evidence for cause-effect relationships, we lagged, by the least unit of time (3 d), the dependent variables, relative to the independent, in some regression models. Stata Release (StataCorp, College Station, TX) was used to perform all regression analyses and graphical displays. Results Subjects and Methods The subjects, methods, and study design have been presented in detail in our previous publications ( 3). In brief, 2 midreproductive-age control women (aged 2 3 yr) with regular menstrual cycles and 6 women (aged 4 yr) with variable cycle characteristics (late reproductive age with regular cycles and early and late menopausal transition, as defined using the STRAW classification (), were recruited by community advertisements in the area of the University of Sydney (Sydney, Australia). Women with amenorrhea for more than 3 months and smokers (within the last 2 months) were excluded. Blood was collected three times weekly throughout one entire cycle and the initial stages of the succeeding cycle. Serum LH, FSH, estradiol, progesterone, inhibin A, inhibin B, and were measured, and the results were presented as means within 3-d windows that were centered on the midcycle LH surge. Simultaneous analysis Multiple linear regression analyses were undertaken simultaneously: ) within 3-d windows in the follicular and luteal phases; and 2) across the whole follicular and luteal phases. Time-lagged analyses It was anticipated that the pituitary response to a change in ovarian activity is likely to be delayed and that a time-lagged analysis may be more informative than an analysis based solely on simultaneously obtained samples. Several studies (6, 7) have shown that increases in serum FSH and LH after steroid (and inhibin) withdrawal at ovariectomy in women are prolonged, taking days to weeks to reach a maximum response, with the earliest response between 2 and 24 h for FSH and 3 d for LH (6, 7). To explore the concept of time-lagged associations further, the linear regression analyses were performed using a 3-d lag interval within the follicular and luteal phases of the cycle. The main objective was to identify which ovarian factors correlated with a 3-d delayed response in FSH and LH. In other words, would low levels of inhibin B in the early follicular phase predict raised levels of FSH in the midfollicular phase as assessed with a 3-d time delay? The lagged data analysis was performed across the follicular phase (menstruation 3 early follicular phase (fp); early fp 3 mid fp; mid fp 3 late fp) and the luteal phase (lp) (early lp 3 mid lp; mid lp 3 late lp). In addition, using the same Independent relationships between ovarian hormones and serum FSH and LH Simultaneous analyses As a representative assessment, the relationship between FSH and all the other hormones in the early follicular phase is shown in Fig. and the correlation coefficients of key relationships are presented in Table. The slopes with 9% confidence intervals (CI) of the regression lines as determined by multiple linear regression analyses for each ovarian hormone (log FSH and log LH as the independent variables) are shown: ) across the various phases of the menstrual cycle (Fig. 2); and 2) combined within follicular and luteal phases in Fig. 3, A and C. In the follicular phase (Fig. 3A), the simultaneous prediction of FSH resulted in negative associations with estradiol [slope 9% CI,.44 ( 6 to.32)], [.22 (.27 to.7)] and weakly, inhibin B [.9 (.8 to.)], and a positive association with progesterone [.2 (.8 to.3)]. For LH, a positive association was observed with progesterone [.3 (.7 to.46)], a negative association with inhibin B [.3 (.4 to.8)], and a weak association with [.7 (. to.2)]. Little evidence of an association between LH and estradiol or inhibin A was observed (Figs. 2 and 3). In the luteal phase (Fig. 3C), a negative association was observed between FSH and estradiol [.33 ( 3 to.3)], [ (. to.2)], and progesterone [ 4 (.68 to.39)]. In this phase, a significant (P.) and marked negative association was observed between LH and progesterone [.82 (. to.63)]. The slopes of regression lines for FSH vs. estradiol or inhibin B were not significantly different when the above analyses were undertaken in the less than 4 yr and more than 4 yr age groups (Fig. 4), despite the elevated levels of FSH in the older age group. on 3 December 27

3 4 Robertson et al. Ovarian:Pituitary Feedback Regulation J Clin Endocrinol Metab, January 29, 94():38 44 TABLE. Correlation coefficients between ovarian and pituitary hormones as assessed in simultaneously obtained samples in different stages of the menstrual cycle (n 42 4) FSH (IU/L) pmol/l nmol/l 2 2 µg/l µg/l LH -2 IU/L µg/l log (hormone) concentration FIG.. Scatterplots between FSH and LH/ovarian hormones in the early follicular phase of ovulatory menstrual cycles from women in mid and late reproductive ages. See Table for more details. Time-lagged analyses In contrast to the simultaneously observed associations described above, clear associations were evident for the lagged prediction of FSH in the follicular phase (Fig. 3B). Significant (P.) negative associations were evident for inhibin B [slope 9% CI,.3 (.4 to )] and [.6 (.2 to.)], but not for estradiol, progesterone, or inhibin A. Thus, lower inhibin B and in one 3-d window were associated with higher FSH in the succeeding window. For the lagged prediction of LH, significant positive and negative associations were observed for estradiol [.7 (.42 to.)] and inhibin B [.33 ( 6 to.)], respectively. In the luteal phase, in the lagged prediction of FSH (Fig. 3D), negative associations were evident for estradiol [.27 (.43 to.)], inhibin A [.2 (.3 to.)], inhibin B [.7 (.3 to.)], progesterone [.8 (.28 to.)], and [. ( to.)]. For the corresponding lagged prediction of LH, a clear negative association with progesterone [.39 ( to.2)] was evident, but there was little or no evidence of association for, inhibin B, and estradiol (Fig. 3D). Correlation coefficients r P Follicular phase FSH vs. estradiol.24 ns.76 P. FSH vs. progesterone.8 ns.3 ns FSH vs. LH.7 P..68 P. FSH vs. inhibin A.7 ns.4 ns FSH vs. inhibin B 6 P..49 P. FSH vs. 2 P..4 P. vs..6 P..39 P. vs. inhibin A.86 P..49 P. Luteal phase FSH vs. LH.78 P. FSH vs..34 P FSH vs. progesterone.37 P. LH vs. estradiol.36 P. LH vs. ns LH vs. progesterone P. vs. progesterone.42 P. vs. inhibin B.4 P., Early follicular phase;, mid follicular phase;, late follicular phase;, early luteal phase;, mid luteal phase;, late luteal phase; ns, not significant. The associations for the lagged prediction of ovarian hormones by FSH and LH were also investigated. Significant positive lagged associations were observed in the follicular phase between inhibin A and LH [.38 (.9 to.67)], and estradiol and LH [.3 (.9 to.6)], whereas a significant negative lagged association was noted between and FSH [.32 (.62 to.2)] in the follicular phase, but not LH. and B showed no significant relationships (data not shown). FSH (but not or inhibin B) in the midfollicular phase predicts estradiol positively [.4 (.2:.7)] and progesterone negatively [.3 (.6 to.92)] in the midluteal phase. Both these associations were significant in the more than 4 yr age group [estradiol,.4 (.2 to.7); progesterone,.4 ( 3 to.7)], but not in the less than 4 yr age group. Discussion A number of significant relationships were identified between circulating ovarian hormones and pituitary gonadotropins in this study in particular, the inverse relationships between FSH and both inhibin B and, LH and inhibin B, and steroids with both gonadotropins. However, it is unclear to what extent these ovarian hormones are bona fide feedback regulators of gonadotropin secretion. A feedback regulator in this context is an ovarian factor that responds to pituitary gonadotropin stimulation and inhibits gonadotropin secretion when its circulating on 3 December 27

4 J Clin Endocrinol Metab, January 29, 94():38 44 jcem.endojournals.org 4 A Multivariable FSH (with LH simultaneously) B Multivariable LH (with FSH simultaneously) Slope estimates & 9% confidence intervals Slope estimates & 9% confidence intervals FIG. 2. The independent relationships between ovarian hormones and FSH (A) and LH (B) during the follicular and luteal phases of the menstrual cycle. Data are presented as the slopes with 9% CI of the regression lines as determined by multiple linear regression analysis for each ovarian hormone, with both FSH and LH simultaneously. By including LH in models for FSH (and vice versa), we remove any possible confounding effects between the two outcome variables; the direct influence of LH itself on FSH has been nullified. Positive or negative y-axis values reflect a positive or negative slope, and the significance of that slope can be assessed from the overlap of the confidence limits with the y-axis where the y-axis value is zero. Vertical dashed line separates the follicular phase of the second cycle (b, b, b) from first cycle., Menstruation;,, and, early, mid, and late follicular phase;, ovulation;,, and, early, mid, and late luteal phase. levels are sufficiently elevated. For example, inhibin B secretion is stimulated by FSH and when the circulating levels of inhibin B are sufficiently elevated, it feeds back to the pituitary to decrease the secretion of FSH (8, 9). Nonetheless, it is recognized that there may be other mechanisms which exhibit either stimulatory or inhibitory actions, without necessarily being part of a feedback regulatory system. To characterize these relationships, simultaneous and 3-d time lagged multiple regression analyses between ovarian and pituitary hormones were performed within the follicular and luteal phases of the cycle according to age categories. The former analysis highlights relationships between hormones that are attributed to age-related changes, including changes associated with the menopause transition. The second approach was chosen on the premise that a regulatory role of a hormone produced by one organ (the ovary) will take time to exert its effect on the hormone production of another organ (the hypothalamo-pituitary unit). Because the samples in this study were pooled into 3-d intervals across the follicular and luteal phases, a 3-d interval was chosen for the time-lag analyses between ovarian hormone levels and pituitary hormone levels. Relationships between ovarian and pituitary hormones Analyses of data from simultaneously obtained samples within each phase of the menstrual cycle revealed a strong inverse relationship (P.) between FSH as the independent variable and both estradiol and in the follicular phase of ovulatory cycles (Figs. 2 and 3). This relationship was not altered with age (Fig. 4). showed a significant negative relationship with FSH (P.8) in the midfollicular stage of the cycle (Fig. 2), although at lower significance (P.68) when tested over the combined stages of the follicular phase (Fig. 3). When comparisons were made within cycle in the follicular phase between ovarian hormone levels and pituitary hormone levels obtained 3 d later (lagged samples), inhibin B (P.) and (P.), but not estradiol, progesterone, or inhibin A, showed a strong negative association with FSH (Fig. 3). This negative association with inhibin B is seen at both younger and older ages but becomes obvious as a result of the age-related decline in ovarian reserve and thus inhibin B secretion, leading to a concomitant increase in FSH. The inverse relationship between FSH and inhibin B is consistent with the recognized feedback role of inhibin B on FSH secretion (8 ). on 3 December 27

5 42 Robertson et al. Ovarian:Pituitary Feedback Regulation J Clin Endocrinol Metab, January 29, 94():38 44 A B FSH C D LH Follicular phase Luteal phase Follicular phase Luteal phase Slope estimates & 9% confidence intervals FIG. 3. The independent relationships between ovarian hormones and FSH/LH as independent variables during the follicular (A) and luteal (B) phases of the menstrual cycle using simultaneous (A and C) and lagged (B and D) linear regression analysis. For additional information, see Fig. 2 legend. Overall, however, estradiol (a regulator of overall FSH tone or setting ) and inhibin B (the major negative feedback factor) appear to be the main factors regulating FSH across age and within cycle, although further analysis will be needed to confirm whether these conclusions apply within discrete phases of the menstrual cycle. and FSH has no known role in the regulation of FSH, nor is known to inhibit FSH secretion or to be stimulated by gonadotropins ( 7). is primarily a product of granulosa cells from small but not primordial follicles (2). A number of studies [recently reviewed by Visser et al. (3)], suggest that plasma reflects the size of the resting pool of pre-fshdependent follicles (4) and thus is a good marker of ovarian reserve. These data and those in mice () suggest that plays a role in inhibiting the expansion of the primordial follicle FSH (IU/L) FSH (IU/L) 2 2, Age < 4 y, Age > 4 y pmol/l pmol/l, Age < 4 y, Age > 4 y ng/ml ng/ml FIG. 4. Scatterplots and regression analyses comparing FSH and estradiol and FSH and inhibin B in the early follicular phase with age ( 4 yr vs. 4 yr). Slope values comparing the below 4 yr vs. above 4 yr age groups: FSH vs. estradiol [ 4 ( to.6) vs..2 (.22 to 2.2), not significant]; FSH vs. inhibin B[.76 (.7 to.44) vs. (.4 to.89), not significant]. pool with no evidence of an endocrine role in regulating gonadotropin secretion. Nonetheless, in this study, a highly significant negative association between and FSH was noted over the menstrual cycle at all ages and within cycle (as assessed by lagged- phase regression analysis) even after accounting for the known FSH regulators (estradiol, inhibin A, inhibin B, and progesterone) in the analysis. These findings are supported by studies of normal ovulatory menstrual cycles of young women (6) where a significant negative correlation was observed between FSH and and in women undergoing in vitro fertilization treatment (, 7). In these latter studies in which the ovary was hyperstimulated with FSH, serum decreased by up to 2% whereas FSH increased (, 7). It was concluded that the decrease in could be due to a fall in the number of 2- to -mm follicles (occurring with FSH hyperstimulation) (7) or FSH inhibition of production as seen in vitro in granulosa cells obtained from PCOS patients (8). However, there is no evidence for a regulatory feedback mechanism between and FSH. It remains unknown whether FSH exerts any effect on secretion, either directly or indirectly, but it appears that is not a feedback regulator of FSH within the ovarianpituitary axis. however is a marker of declining ovarian reserve and may contribute to the elevations in FSH by unknown mechanisms. Based on the proposition that is a marker of follicle reserve, it would be interesting to assess whether antral follicle count rather than is an independent correlate of on 3 December 27

6 J Clin Endocrinol Metab, January 29, 94():38 44 jcem.endojournals.org 43 FSH. In such an analysis, would no longer be associated with FSH. Unfortunately, antral follicle count was not determined in this study. and LH Surprisingly, a significant negative association was seen in the follicular phase between inhibin B and LH, suggesting that inhibin B may also be a regulator of LH. This is an area of considerable controversy because it is generally recognized that inhibins exert a differential inhibitory effect on FSH (8), whereas LH is under the influence of ovarian steroids and possibly gonadal factors such as gonadotropin surge-attenuating factor (9). Administration of purified preparations of inhibin A either as a bolus or by infusion in monkeys (2 22), sheep (23), and rats (24, 2) results in a decline in circulating FSH but not LH. However, there is in vivo evidence in immature (but not mature) male rats (2) and in vitro in rat primary pituitary cell cultures that inhibin A suppresses LH secretion (but not LH synthesis), LH response to GnRH stimulation (26, 27), and GnRH receptor levels (26). These data suggest that the effect of inhibins on LH may be a combination of a decrease in LH stores and a decrease in the ability of the gonadotroph to respond to GnRH, of which the latter is affected by age in males (2) and cycle stage in females (27). Based on these data, we postulate that an inhibin B:LH feedback mechanism may exist in the human female, but that this mechanism is less pronounced than that seen with the inhibin B:FSH feedback mechanism, and that this inhibition is likely to be modified by the changing endocrinology of the menstrual cycle. and FSH/LH The inverse relationship between estradiol and FSH observed over the follicular and luteal phases of the menstrual cycle (Figs. 2 and 3) is consistent with previous studies (28 32), in support of an inhibitory effect of estradiol on FSH secretion, although by time-lagged analysis a significant difference was only detected in the luteal phase. This latter finding is consistent with findings of in vivo studies with the estradiol receptor antagonist, tamoxifen (33), in which estradiol was more effective in inhibiting FSH in the luteal compared with the follicular phase; however, results from this study did not support this distinction. Based on the failure to obtain a significant response in the lagged-phase hormone analyses in the follicular phase, these data suggest that estradiol (and probably progesterone) is involved in defining the tonic pituitary:ovarian hormonal settings within cycle, which alter with reproductive age. The positive association between estradiol and LH seen in the lagged analysis in the follicular phase is consistent with the recognized augmentation effect of estradiol on GnRH secretion before ovulation (28). and FSH/LH In the follicular phase, progesterone is positively correlated with FSH and LH (Figs. 2 and 3). This relationship is supported by findings (34) that progesterone has an augmentative effect on FSH/LH in the late follicular phase of the cycle and that this effect is attributed in part to augmentation of GnRH release of gonadotropins by pituitary gonadotropes (34). In the luteal phase, an inverse relationship was observed between progesterone and both FSH and LH, similar to studies in postmenopausal women after a sequential estradiol plus progesterone treatment (28, 3), and this suggests an inhibitory role, with primarily a hypothalamic site of steroid action. Although analyses in this study were primarily centered on the influence of ovarian hormones on FSH and LH, a reverse time-lagged analysis was undertaken correlating gonadotropin levels with ovarian hormone levels lagged 3 d later. In the follicular phase, there was no association between FSH and either lagged inhibin A or B. Significant positive associations between LH and inhibin A and estradiol in the follicular phase support the known trophic action of LH. Our previous studies ( 3) noted that the patterns of pituitary and ovarian hormones throughout the menstrual cycle change with the onset of irregular cycles, leading to decreases in serum inhibins,, and progesterone and elevations in gonadotropins and late luteal estradiol levels. The latter increase in estradiol was seen only during irregular menopause transition ovulatory cycles and was associated with lower luteal phase progesterone, higher early cycle FSH, and lower early cycle inhibin B. The luteal phase rise in estradiol was attributed to the premature initiation of folliculogenesis during the luteal phase of an existing ovulatory cycle. This premature folliculogenesis was proposed to have been driven by markedly elevated follicular phase FSH and was referred to as a LOOP (luteal out of phase) event (2). In this study, a significant positive association between follicular phase FSH and luteal phase estradiol was observed in further support of this hypothesis. Conclusion These analyses extend our understanding of the ovarian regulation of pituitary secretion of FSH/LH. The key roles of inhibin B in regulating FSH and of estradiol/progesterone in the regulation of FSH/LH are also supported. The potential role of inhibin B in regulating LH, however, requires further investigation. Although it is concluded that is not an ovarian feedback regulator of pituitary FSH/LH, its marked inverse association with FSH is surprising and may yet indicate an unrecognized role. Acknowledgments The contributions of Dr. Pavel Sluka and Enid Pruysers in the analysis and preparation of this manuscript are gratefully acknowledged. Address all correspondence and requests for reprints to: David Robertson, Ph.D., Prince Henry s Institute of Medical Research, P.O. Box 2, Clayton, Victoria 368, Australia. david.robertson@ princehenrys.org. This work was supported by National Health and Medical Research Council of Australia Program Grant 24 and Research Fellowship 692 (to D.M.R.). Authors Disclosure Information: G.E.H., C.L.H., D.J., and I.S.F. have nothing to disclose. H.G.B. and D.M.R. are inventors on patents AU8/9 and AU86/97. on 3 December 27

7 44 Robertson et al. Ovarian:Pituitary Feedback Regulation J Clin Endocrinol Metab, January 29, 94():38 44 References. Hale GE, Zhao X, Hughes CL, Burger HG, Robertson DM, Fraser IS 27 Endocrinology of reproductive aging: the endocrine features of menstrual cycles in mid-reproductive age, late reproductive age and menopause transition subjects classified according to the STRAW staging system. J Clin Endocrinol Metab 92: Hale GE, Hughes CL, Burger HG, Robertson DM, Fraser IS, Atypical estradiol secretion and ovulation patterns caused by luteal out-of-phase ( LOOP ) events underlying irregular ovulatory menstrual cycles in the menopause transition. Menopause, in press 3. Robertson DM, Hale GE, Fraser IS, Hughes CL, Burger HG 28 A proposed classification system for menstrual cycles in the menopausal transition based on changes in serum hormone profiles. Menopause : Baerwald AR, Adams GP, Pierson RA 23 A new model for ovarian follicular development during the human menstrual cycle. Fertil Steril 8:6 22. Soules MR, Sherman S, Parrott E, Rebar R, Santoro N, Utian W, Woods N 2 Stages of reproductive aging workshop (STRAW). J Womens Health Gend Based Med : Yen SSC, Tsai CC 97 The effect of ovariectomy on gonadotropin release. J Clin Invest : McCartney CR, Blank SK, Marshall JC 26 acutely increases LH pulse amplitude but does not acutely influence nocturnal LH pulse frequency slowing during the late follicular phase in women. Am J Physiol Endocrinol Metab 292:E9 E96 8. Welt CK 24 Regulation and function of inhibins in the normal menstrual cycle. Semin Reprod Med 22: Burger HG, Dudley EC, Robertson DM, Dennerstein L 22 Hormonal changes in the menopause transition. Recent Prog Horm Res 7: Klein NA, Houmard BS, Hansen KR, Woodruff TK, Sluss PM, Bremner WJ, Soules MR 24 Age-related analysis of inhibin A, inhibin B, and activin a relative to the intercycle monotropic follicle-stimulating hormone rise in normal ovulatory women. J Clin Endocrinol Metab 89: van Rooij IA, Broekmans FJ, te Velde ER, Fauser BC, Bancsi, de Jong FH, Themmen AP 22 Serum anti-müllerian hormone levels: a novel measure of ovarian reserve. Hum Reprod 7: Stubbs SA, Hardy K, Da Silva-Buttkus P, Stark J, Webber LJ, Flanagan AM, Themmen AP, Visser JA, Groome NP, Franks S 2 Anti-müllerian hormone protein expression is reduced during the initial stages of follicle development in human polycystic ovaries. J Clin Endocrinol Metab 9: Visser JA, de Jong FH, Laven JS, Themmen AP 26 Anti-Müllerian hormone: a new marker for ovarian function. Reproduction 3: 9 4. McIlveen M, Skull JD, Ledger WL 27 Evaluation of the utility of multiple endocrine and ultrasound measures of ovarian reserve in the prediction of cycle cancellation in a high-risk IVF population. Hum Reprod 22: Durlinger AL, Visser JA, Themmen AP 22 Regulation of ovarian function: the role of anti-müllerian hormone. Reproduction 24: Wunder DM, Bersinger NA, Yared M, Kretschmer R, Birkhäuser MH 28 Statistically significant changes of antimüllerian hormone and inhibin levels during the physiologic menstrual cycle in reproductive age women. Fertil Steril 89: Catteau-Jonard S, Pigny P, Reyss AC, Decanter C, Poncelet E, Dewailly D 27 Changes in serum anti-mullerian hormone level during low-dose recombinant follicular stimulating hormone therapy for anovulation in polycystic ovary syndrome. J Clin Endocrinol Metab 92: Pellatt L, Hanna L, Brincat M, Galea R, Brain H, Whitehead S, Mason H 27 Granulosa cell production of anti-müllerian hormone is increased in polycystic ovaries. J Clin Endocrinol Metab 92: Fowler PA, Sorsa-Leslie T, Harris W, Mason HD 23 Ovarian gonadotrophin surge-attenuating factor (GnSAF): where are we after 2 years of research? Reproduction 26: Ramaswamy S, Pohl CR, McNeilly AS, Winters SJ, Plant TM 998 The time course of follicle-stimulating hormone suppression by recombinant human inhibin A in the adult male rhesus monkey (Macaca mulatta). Endocrinology 39: Stouffer RL, Dahl KD, Hess DL, Woodruff TK, Mather JP, Molskness TA 994 Systemic and intra luteal infusion of inhibin A or activin A in Rhesus monkeys during the luteal phase of the menstrual cycle. Biol Reprod : Molskness TA, Woodruff TK, Hess DL, Dahl KD, Stouffer RL 996 Recombinant human inhibin-a administered early in the menstrual cycle alters concurrent pituitary and follicular, plus subsequent luteal, function in rhesus monkeys. J Clin Endocrinol Metab 8: Tilbrook AJ, De Kretser DM, Clarke IJ 993 Human recombinant inhibin A suppresses plasma follicle-stimulated hormones to intact levels but has no effect on luteinizing hormone in castrate rats. Biol Reprod 49: Robertson DM, Prisk M, McMaster JW, Irby DC, Findlay JK, de Kretser DM 99 Serum FSH suppressing activity of human recombinant inhibin A in male and female rats. J Reprod Fertil 9: Winters SJ, Pohl CR, Adedoyin A, Marshall GR 996 Effect of continuous inhibin administration on gonadotropin secretion and subunit gene expression in immature and adult male rats. Biol Reprod : Farnworth PG, Robertson DM, de Kretser DM, Burger HG 988 Effects of 3 kda bovine inhibin on FSH and LH in rat pituitary cells in vitro: antagonism of gonadotrophin-releasing hormone agonists. J Endocrinol 9: Childs GV, Miller BT, Miller WL 997 Differential effects of inhibin on gonadotropin stores and gonadotropin-releasing hormone binding to pituitary cells from cycling female rats. Endocrinology 38: Lasley BL, Wang CF, Yen SS 97 The effects of estrogen and progesterone on the functional capacity of the gonadotrophs. J Clin Endocrinol Metab 4: Marshall JC, Case GD, Valk TW, Corley KP, Sauder SE, Kelch RP 983 Selective inhibition of follicle-stimulating hormone secretion by estradiol. Mechanism for modulation of gonadotropin responses to low dose pulses of gonadotropin-releasing hormone. J Clin Invest 7: de Ziegler D, Bergeron C, Cornel C, Medalie DA, Massai MR, Milgrom E, Frydman R, Bouchard P 992 Effects of luteal estradiol on the secretory transformation of human endometrium and plasma gonadotropins. J Clin Endocrinol Metab 74: Gill S, Lavoie HB, Bo-Abbas Y, Hall JE 22 Negative feedback effects of gonadal steroids are preserved with aging in postmenopausal women. J Clin Endocrinol Metab 87: Lahlou N, Chabbert-Buffet N, Christin-Maitre S, Le Nestour E, Roger M, Bouchard P 999 Main inhibitor of follicle stimulating hormone in the lutealfollicular transition: inhibin A, oestradiol, or inhibin B? Hum Reprod 4: Welt CK, Pagan YL, Smith PC, Rado KB, Hall JE 23 Control of folliclestimulating hormone by estradiol and the inhibins: critical role of estradiol at the hypothalamus during the luteal-follicular transition. J Clin Endocrinol Metab 88: McCartney CR, Blank SK, Marshall JC 26 acutely increases LH pulse amplitude but does not acutely influence nocturnal LH pulse frequency slowing during the late follicular phase in women. Am J Physiol Endocrinol Metab 292:E9 E96 on 3 December 27