Determinants of early follicular phase gonadotrophin and estradiol concentrations in women of late reproductive age

D.W. Cramer1 ,3 , R.L. Barbieri2 , A.R. Fraer1 and B.L. Harlow1

1 Obstetrics and Gynecology Epidemiology Center and 2 Department of Obstetrics and Gynecology, Brigham and Women's Hospital, Boston, MA 02115, USA


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: FSH and estradiol measured during the menstrual (basal) phase of cycles predict the success of infertility treatment; but the role of these hormones as markers for ovarian reserve in normal populations needs further study. METHODS AND RESULTS: From a cohort study of depressed and non-depressed women, a subset of 406 non-depressed women between the ages of 36 and 45 years with spontaneous periods were selected and their concentrations and determinants of basal hormones measured at study entry, 6 and 12 months later were described. FSH and LH increased significantly over the 12 months of observation (P <= 0.001), but considerable variation was noted in FSH and estradiol in some women monitored over the three cycles. Concentrations often varied between a pattern of low FSH, high estradiol in one cycle and high FSH, low estradiol in another. In multivariate models focusing on the maximum observed hormone concentration, significant predictors included: increasing age (P < 0.0001), smoking (P = 0.04), and shorter cycle length (P < 0.0001) during adolescence (P < 0.0001) associated with higher FSH; increasing age (P < 0.0266) and lower body mass index (P < 0.0289) associated with higher LH; and a greater number of estimated ovulatory cycles associated with higher estradiol (P < 0.0425). CONCLUSIONS: Early reproductive landmarks, smoking, body weight, and factors that determine number of ovulatory cycles impact on ovarian/pituitary physiology in late reproductive life.

Key words: body mass index/estradiol/gonadotrophins/menstrual cycle/ovulation


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The process of ovarian failure is generally a gradual one that coincides with a progressive loss of follicles as a woman ages. From an endowment of ~106 at birth, ~104 will remain by age 40 years (Faddy et al., 1992Go). By the time of menopause, essentially no follicles remain that are capable of developing into more advanced follicles with functioning granulosa–thecal components necessary for steroid production and feedback inhibition of the pituitary. The latter circumstance leads to the profile of high FSH and low estradiol typical for postmenopausal women. A precise description of endocrinological changes during the decades-long process of ovarian failure is complicated by constraints of physiology and feasibility. Marked changes in the concentrations of FSH, LH and estradiol during an individual menstrual cycle and the pulsatile nature of LH secretion would appear to mandate approaches that involve frequent sampling at multiple time points during the cycle; but such studies have been limited to small numbers of subjects (van Santbrink et al., 1995Go).

That measurement of reproductive hormones at a single point during the cycle might be informative was suggested by data indicating that basal hormone concentrations measured during the early follicular (i.e. menstrual) phase of the cycle were correlated with success after IVF. A high basal FSH (and, curiously, a high basal estradiol) predicted fewer oocytes recovered and lower chance of pregnancy (Muasher et al., 1988Go; Scott et al., 1989Go; Pearlstone et al., 1992Go; Licciardi et al., 1995Go; Smotrich et al., 1995Go; Witt, et al., 1995Go; Buyalos et al., 1997Go). Thus, measurement of basal FSH and estradiol, as potential markers for ovarian reserve, offers an approach to the study of the ovarian life cycle that is easier to use in large populations and yields simpler descriptive statistics. However, because recent studies (Burger, 1994Go; Santoro et al., 1996Go) suggest that there may be considerable variation in hormone concentrations from cycle to cycle, it is important to monitor hormones over more than one cycle. In this study, basal hormone concentrations were measured at entry, 6 and 12 months later in women aged 36–45 years. Description of reproductive hormones in this group may contribute to a greater understanding of the ovarian life cycle during a critical age period when fertility declines markedly.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
This study is based on subjects selected as part of the Harvard Study of Moods and Cycles—a prospective assessment of epidemiological and hormonal associations and consequences of major depression in women. The methods for this study have been described in detail elsewhere (Harlow et al., 2002Go). Briefly, all women between the ages of 36 and 44 years from seven communities in greater Boston were identified using Massachusetts Town Books (annual publications that list residents by name, age, and address according to voter precincts). Self-administered screening questionnaires were mailed to 6228 women with a verified address and telephone number. The questionnaire assessed reproductive landmarks, menstrual history, and past diagnosis or treatment of depression and included the Center for Epidemiologic Studies Depression Scale (CES-D). After two mailings and telephone follow-up of non-respondents, the screening component achieved a 73.5% response rate.

From those who completed the initial survey, cohorts of women with and without current or past depression were invited to participate in a prospective study that included structured clinical psychiatric interviews, baseline and semi-annual medical questionnaires, and menstrually timed collection of blood specimens. The cohort of non-depressed women, who were the focus in this study, were those who scored <16 on the CES-D and did not meet criteria for a lifetime history of major depression based upon a Structured Clinical Interview for the Diagnostic and Statistical Manual (SCID-IV). Approximately 30% of the initially screened non-depressed women were eventually enrolled in the Harvard Study of Moods and Cycles. Among the women not enrolled, about half were unwilling to make the sizeable commitment required for participation in the Study of Moods and Cycles which entailed bi-annual menstrually timed blood donations and questionnaires. The remaining were ineligible for enrolment due to pregnancy, menopausal status, or loss to follow-up after the initial screening assessment.

The depressed and non-depressed cohorts were followed every 6 months over a 36 month follow-up period and were asked to provide early follicular phase blood specimens and to participate in medical and psychiatric follow-up interviews. The medical questionnaires assessed menstrual history including age at menarche, average cycle length currently and during the first 5 years after menarche, pregnancy and breastfeeding history, and oral contraceptive (OC) or hormone use. A variable called estimated ovulatory cycles was calculated, which has previously been found to be a predictor of risk for menopause (Cramer and Xu, 1996Go). This variable was calculated (for a premenopausal woman) by subtracting the woman's age at menarche from her current age which estimates her maximum `ovulatory years.' This was reduced by potential `anovulatory years' approximated by years of OC use, number of liveborns (roughly equating a term pregnancy with a year of anovulation), and total years of breastfeeding. This `adjusted ovulatory years' is multiplied by 365 and divided by the women's average cycle length (during the first 5 years after menarche) to yield the estimated number of ovulatory cycles. The cycle length that women recalled having during their first five years of menstrual life was used rather than the current cycle length, which on average would have been around age 40 years. It was felt that cycle length in early rather than late reproductive life was more relevant to the number of ovulations experienced during the majority of reproductive life.

For the prospective assessment of basal hormone concentrations, blood was drawn on any one day of the first 5 days of the menstrual cycle with day 1 defined as the first day of menstrual bleeding. The specimen was drawn during the first period after study enrolment and was scheduled every 6 months over the 3 year period of the study. Reminder calls were placed to women around the estimated time of their 6 monthly anniversary periods at which time the regularity of the menstrual cycle and the occurrence of events such as pregnancy, breastfeeding, and use of oral contraceptives or other hormones was reassessed. Because the blood draw had to accommodate holidays and interviewer's and subject's schedules, it was impossible to do the blood test on precisely the same menstrual day for each woman.

Although 643 non-depressed women were enrolled in the parent study, 221 (34.4%) who did not complete three consecutive draws, 11 (1.7%) who went on hormones during the year of observation and five who had unpredictable periods throughout their reproductive life were excluded. This left 406 women who had basal hormones measured during natural cycles at study entry, at 6 and 12 months after entry. Cycle lengths at study entry for these women were between 16 and 38 days and did not vary by more than 10 days from cycle to cycle in 384 (94.6%) subjects. No subject reported >=3 months between periods—the usual definition for oligo-amenorrhoea.

Measurements of serum gonadotrophins were performed using the Coat-A-Count immunoradiometric assays (Diagnostic Products Corp., Los Angeles, CA, USA). The World Health Organization (WHO) First International Reference Preparation (68/40) of human pituitary LH was used as the standard for LH and the WHO Second International Reference Preparation (78/549) of human pituitary FSH was used as the standard for FSH. Estradiol was measured in duplicate and averaged using the Coat-A-Count radioimmunoassay. Intra-assay coefficients of variation (CV) were <4% for the gonadotrophins and <7% for estradiol. Inter-assay CV were <10% for the gonadotrophins and <15% for estradiol.

Means and 95% confidence limits on the mean for the gonadotrophins and estradiol were calculated based upon log transformed values to accommodate skewness of the hormonal distributions. Hormone concentrations were distinguished by whether they were measured at 0, 6 or 12 months and by whether they were the maximum or minimum of the three values regardless of study period. Paired t-tests were used to determine whether subjects' 6 or 12 month values differed from their baseline values or whether maximum differed from minimum values. Generalized linear modelling was used to determine whether (log transformed) entry or maximum hormone concentrations differed by categories of demographic, reproductive, or menstrual history variables after adjusting for age at study entry or cycle day of study. Significant associations were distinguished by those with a two-sided P-value of <=5%. All analyses were performed using the SAS system (Statistical Analyses System Institute, SAS Campus, Cary, NC, USA).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Table IGo shows selected characteristics of the study group together with the means and 95% confidence intervals of basal gonadotrophins and estradiol concentrations measured at study enrolment. FSH increased significantly with increasing age at enrolment (P = 0.003). After adjustment for age and cycle day of blood draw, women with shorter cycle lengths during adolescence had statistically significantly higher FSH (P = 0.005) as did women who were current or former smokers (P = 0.02). As the number of estimated ovulatory cycles increased, FSH concentrations also increased significantly (P = 0.031). Variables associated with baseline LH included race and body mass index (BMI) with lower LH occurring in non-whites (P = 0.005) and those with a greater BMI (P = 0.038). There were no significant predictors of estradiol at study entry, although associations of borderline significance (0.5>P<0.10), included lower estradiol in those with later age at menarche and those who had breastfed for more than 1 year.


View this table:
[in this window]
[in a new window]
 
Table I. Basal reproductive hormones measured at study entry by selected characteristics of study population
 
Table IIGo compares mean FSH, LH and estradiol observed at entry, 6 and 12 months and the maximum or minimum hormone value for women regardless of the month in which it was measured. A paired t-test based on log transformed values indicated that both the 6 and 12 month values for FSH and LH, but not estradiol, were significantly higher than the baseline values. In addition, maximum hormone concentrations were significantly different compared with minimum hormone concentrations for each of the hormones measured. The difference between the maximum and minimum hormone concentrations was especially pronounced in some subjects. Fourteen women (3.4%) had an FSH clearly in the postmenopausal range (>20 IU/l) which returned to a premenopausal level in a subsequent cycle. A total of 187 (46%) of subjects had a >=50% difference between both their minimum and maximum basal FSH and their minimum and maximum basal estradiol concentrations; and 82 (20.2%) of subjects had a >=100% difference between both sets of hormones. In women with such differences, it was more likely (P < 0.001) for the maximum basal FSH to occur in a cycle different from the one in which the maximum basal estradiol was observed, compared to women without such extreme differences (Table IIIGo). Table IVGo shows basal FSH and estradiol values in selected women who best illustrated how low FSH and robust estradiol concentrations in one cycle may be followed by high FSH and low estradiol in another cycle.


View this table:
[in this window]
[in a new window]
 
Table II. Baseline, 6 month, 12 month, maximum and minimum hormone valuesa for the study group
 

View this table:
[in this window]
[in a new window]
 
Table III. Likelihood of a maximum FSH or maximum estradiol value occurring in the same or different cycle in women characterized by the degree of difference in their basal hormone concentrations
 

View this table:
[in this window]
[in a new window]
 
Table IV. Subjects having marked changes in basal FSH and estradiol concentrations
 
Figures 1 and 2GoGo display statistics related to the maximum and minimum values for basal FSH and estradiol for women in 2 year age groups. As illustrated by the medians for the maximum and minimum values for the basal hormones, there was a clear trend for FSH values to increase over the entire age range; but the trend for estradiol was less consistent. As illustrated by the upper 90th percentile value of the maximum and the lower 10th percentile value of the minimum hormone value, variability increased with age. The widening gap between the 10th and 90th percentile values for the basal hormones as women entered their forties reflects the hormonal oscillations described in Table IVGo. Trends in the concentration of, and range for, basal LH were less apparent (data not shown).



View larger version (19K):
[in this window]
[in a new window]
 
Figure 1. Maximum and minimum basal FSH concentrations for women in 2 year age groups from 36 to 45 years.

 


View larger version (21K):
[in this window]
[in a new window]
 
Figure 2. Maximum and minimum basal estradiol concentrations for women in 2 year age groups from 36 to 45 years.

 
The next part of the study was to determine which key variables influenced the maximum basal hormone concentrations observed during the three cycles (Table VGo). The focus was on the maximum hormone concentration because it was felt that this may have greater biological significance than the minimum concentration; and a `step-down' procedure was used in a multivariate analysis beginning with all variables in Table IGo and dropping those of least significance. Cycle day of the blood draw and age (except where noted) were included as potential confounding factors in the models. Variables found to independently predict a greater maximum basal FSH concentration were advancing age (P < 0.0001), shorter cycle length during adolescence (P = 0.001), and former or current smoking (P = 0.04). When the single variable `estimated ovulatory cycles' was substituted for age and cycle length (which are components of that variable), increasing ovulatory cycles were strongly associated with increasing FSH, while the significance of smoking was retained. Variables found to independently predict a greater maximum basal LH concentration were advancing age (P = 0.0266) and BMI (P = 0.0289). Race did not persist as a significant predictor in the multivariate model; nor did ovulatory cycles significantly influence maximum LH concentrations. Variables found to predict maximum estradiol were parity (increasing estradiol with greater parity) and age at first livebirth (lower estradiol with later age at first livebirth). When the variable, ovulatory cycles, was substituted for age and the pregnancy-related variables, increasing ovulatory cycles were associated with increasing estradiol concentrations.


View this table:
[in this window]
[in a new window]
 
Table V. Key characteristics predicting maximum basal gonadotrophin and estradiol concentrations during 1 year observation period
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The decade beginning around age 36 years is an important one during a woman's reproductive life. Models based on counts of primordial follicles in ovaries removed at autopsy or surgery suggest that an accelerated loss of follicles begins in the late thirties (Faddy et al., 1992Go). Beginning around the same time, marked declines in fertility occur in women being treated with artificial insemination (Schwartz and Mayaux, 1982Go) or IVF (Engmann et al., 1999Go). It is also a decade during which rates for breast, endometrial, and ovarian cancer rise sharply. Thus, description of reproductive hormones in this age group may be informative. In this study, a large number of women was selected from the general population and their hormone concentrations were measured during the early follicular phase in three cycles, 6 months apart. The focus of the study was on both the initial concentrations observed as well as the maximum during the three cycles. The findings appear to complement and expand current knowledge of reproductive physiology.

The results confirm that FSH concentrations increase as a woman ages (Sherman et al., 1976Go; Reyes et al., 1977Go; Metcalf and Livesy, 1985Go; Lenton et al., 1988Go; Cramer et al., 1994Go). Even between their baseline and 6 month period of observation, women had a small but significant increase in basal FSH and LH. The physiological basis for the increase in FSH, which begins well before the menopausal transition, is likely to be due to a declining germ cell pool with a consequent decrease in concentrations of the ovarian hormone, inhibin B, which regulates FSH secretion (Buckler et al., 1991Go; Welt et al., 1999Go). Unfortunately, due to the high cost and lack of availability of the inhibin assay when this study began, inhibin concentrations were not measured in this study.

The calculated variable `estimated ovulatory cycles' was also strongly related to FSH. Components of this variable include: years at risk for ovulation determined by current age and age at menarche; events associated with anovulation including pregnancies, breastfeeding, and OC use; and frequency of ovulation, as determined by cycle length. Cycle length in early reproductive life rather than late was chosen for this variable, although the two were correlated in this study (r = 0.24, P = 0.0001). Interestingly, cycle length in early reproductive life, by itself, was a strong predictor of basal FSH. It can be speculated that short cycle length in early reproductive life may signal a smaller germ cell endowment to begin with since it is known that cycle length decreases as women age and ovarian reserve decreases (Treloar et al., 1967Go; Lenton et al., 1984Go; Munster et al., 1992Go). Follicle counts performed at autopsy by Block ranged between 85 000 and 684 000 in women aged <20 years, suggesting that varying germ cell endowments are likely to occur (Block, 1952Go).

An additional variable, which might also mediate its effect on FSH through germ cell depletion, is smoking. This study confirmed prior observations that current and former smokers have higher FSH concentrations (Velasco et al., 1990Go; Cramer et al., 1994Go; Cooper et al., 1995Go). Smokers undergoing IVF have poorer oocyte recovery and lower fertilization rates (Van Voorhis et al., 1992Go). A logical extension of the search for factors which increase basal FSH concentrations would be that these same variables hasten age at menopause. Indeed, it is well established that smoking decreases the age at menopause (Cramer et al., 1995Go) and it has also reported that the `estimated ovulatory cycles' variable predicted an earlier than normal menopause (Cramer and Xu, 1996Go).

Only age and BMI significantly affected the maximum LH concentration. Similar to FSH, LH increases with age; but the effect of BMI on LH may not be a simple linear one. It appears that a lower LH may occur at both extremes of weight (Table IGo). This association between BMI has been previously described (Bohlke et al., 1998Go) and may be modulated through other hormones including insulin or leptin. Fasting insulin concentrations are positively correlated with BMI or body fat and negatively correlated with LH (Dale et al., 1992Go) which could account for depressed concentrations of LH in obese women. In thin women, the association might involve the adipose hormone, leptin, which is known to be strongly and positively correlated with BMI and which also appears to have a number of hypothalamic and ovarian effects (Mantzoros et al., 1998Go).

One variable which was found to significantly influence the maximum estradiol concentration was ovulatory cycles. Although the effect could be an indirect one mediated by the effect of past ovulations on FSH concentrations, it is reasonable also to consider a direct ovarian effect. A greater number of ovulatory cycles would imply that more follicles with developed granulosa–thecal compartments have been formed. This could impact on the number of residual cells, which have previously been luteinized and may be capable of steroid production, that remain in ovaries which have undergone a greater number of ovulations. Though highly speculative, this theory would have considerable relevance to the postulated effects of a longer reproductive life and greater number of ovulations on breast and ovarian cancer risk. Clearly, estradiol concentrations eventually decrease in all women regardless of their past reproductive history, so it will be important to describe precisely how the transition from fluctuating to low estradiol concentrations occurs.

We believe that an important observation of the current study is that there is considerable short-term variation in reproductive hormone concentrations. Instances in which wide fluctuations in reproductive hormones occurred in the same woman have been described, but mostly in women in their late forties (Burger, 1994Go; Santoro et al., 1996Go), while the current study suggests the fluctuations begin earlier in reproductive life. The `oscillation' of basal hormones pointed out in Table IVGo raises the question of what would be the physiological basis for `menopausal' concentrations of FSH and estradiol to be measured in one sampling period and low FSH and high estradiol in another. Based on the observations from Table IIIGo, it is unlikely that this shift occurs simply at different time points during the early follicular phase of the same cycle. More likely, extremes in one cycle set up the dynamics for an opposite extreme to be observed in the next cycle. This could explain the observation from IVF clinics that both an elevated basal FSH and elevated basal estradiol are correlated with poor outcomes (Licciardi et al., 1995Go; Smotrich et al., 1995Go; Witt et al., 1995Go). However, rather than being a distinct physiological event, an elevated basal estradiol may simply reflect the response to the high FSH from the previous cycle. Studies that monitor basal hormones in two or more consecutive cycles should be informative and clarify whether patterns tend to be repeated by the same ovary (since it is postulated that ovulation alternates between ovaries).

Limitations of the current study include the facts that the sample consisted mostly of Caucasians, excluded women with major depression, and involved a selected group of women willing to participate in a longitudinal study. These factors may limit the generalizability of the results. In addition, the key exposures which were assessed were based on recall and may be subject to error, although such biases generally obscure rather than create associations. Focusing on a single point during the cycle rather than multiple time points precluded investigation into whether all of the cycles studied were ovulatory and cycle dynamics. Logistical issues prevented collection of specimens on precisely the same day in all subjects. Finally, adrenal hormones such as dehydroepiandrosterone were not studied, which may also contribute to estrogen concentrations.

Despite these limitations, we believe this study has yielded some important conclusions. Basal FSH increases during late reproductive life and wide fluctuations in FSH and estradiol may occur, suggesting that basal hormone concentrations need to be characterized over more than one cycle. Higher basal FSH concentrations are seen in smokers and those with a shorter cycle length in early reproductive life. Estimated number of ovulatory cycles was also strongly related to FSH, providing epidemiological support for the physiological relationship between germ cell mass, inhibin, and FSH concentrations. Past ovulatory cycles also influenced maximum estradiol concentrations observed, inviting the speculation that the number of prior ovulations may also affect the steroid-producing capability of the ovary in later life. BMI may also influence hormone concentrations, especially LH, and mediate some of the known effects of BMI on fertility. Further studies involving a broader spectrum of reproductive hormones followed in consecutive cycles would advance understanding of the ovarian life cycle.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Supported by a grant (MH-50013) from the National Institutes of Mental Health. We are indebted to Rebecca Liberman, Evelyn Li, and Vanessa Pratomo for their assistance in data processing and laboratory assessments.


    Notes
 
3 To whom correspondence should be addressed at: Ob-Gyn Epidemiology Center, 221 Longwood Avenue, Boston, MA 02115, USA. E-mail: dcramer{at}partners.org Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Block, E. (1952) Quantitative morphological investigations of the follicular system in women. Variations at different ages. Acta Anat., 14, 108–123.[ISI]

Bohlke, K., Cramer, D.W., Barbieri, R.L. (1998) Relation of luteinizing hormone levels to body mass index in premenopausal women. Fertil. Steril., 69, 500–504.[ISI][Medline]

Buckler, H.M., Evans, C.A., Mantora, H. et al. (1991) Gonadotropin, steroid and inhibin levels in women with incipient ovarian failure during anovulatory and ovulatory rebound cycles. J. Clin. Endocrinol. Metab., 72, 116–124.[Abstract]

Burger, H.G. (1994) The menopause: when it is all over or is it. Aust. NZ J. Obstet. Gynaecol., 34, 293–295.[ISI][Medline]

Buyalos, R.P., Daneshmand, S. and Brzechffa, P.R. (1997) Basal estradiol and follicular-stimulating hormone predict fecundity in women of advanced reproductive age undergoing ovulation induction therapy. Fertil. Steril., 68, 272–277.[ISI][Medline]

Cooper, G.S., Baird, D.D., Hulka, B.S. et al. (1995) Follicle-stimulating hormone concentrations in relation to active and passive smoking. Obstet. Gynecol., 85, 407–411.[Abstract/Free Full Text]

Cramer, D.W. and Xu, H. (1996) Predicting age at menopause. Maturitas, 23, 319–326.[ISI][Medline]

Cramer, D.W., Barbieri, R.L., Xu, H. et al. (1994) Determinants of basal-follicle stimulating hormone levels in premenopausal women. J. Clin. Endocrinol. Metab., 79, 1105–1109.[Abstract]

Cramer, D.W., Harlow, B.L., Xu, H. et al. (1995) Cross sectional and case-controlled analyses of the association between smoking and early menopause. Maturitas, 22, 79–87.[ISI][Medline]

Dale, P.O., Tanbo, T., Vaaler, S. et al. (1992) Body weight, hyperinsulinemia and gonadotropin levels in the polycystic ovary syndrome; evidence of two distinct populations. Fertil. Steril., 58, 487–491.[ISI][Medline]

Engmann, L., Maconochie, N., Bekir, J.S. et al. (1999) Cumulative probability of clinical pregnancy and live birth after a multiple cycle IVF package: a more realistic assessment of overall and age-specific success rates? Br. J. Obstet. Gynaecol., 106, 165–170.[ISI][Medline]

Faddy, M.J., Gosden, R.G., Gougeon, A. et al. (1992) Accelerated disappearance of ovarian follicle in mid-life: implications for forecasting menopause. Hum. Reprod., 7, 1342–1346.[Abstract]

Harlow, B.L., Cohen, L.S., Otto, M. et al. (2002) Demographic, family, and occupational characteristics associated with major depression: The Harvard Study of Moods and Cycles. Acta Psychiatr. Scand., in press.

Lenton, E.A., Landgren, B.M., Sexton, L. et al. (1984) Normal variation in the length of the follicular phase of the menstrual cycle: effect of chronological age. Br. J. Obstet. Gynaecol., 91, 681–684.[ISI][Medline]

Lenton, E.A., Sexton, L., Lee, S. et al. (1988) Progressive changes in LH and FSH and LH:FSH ratio in women throughout reproductive life. Maturitas, 10, 35–43.[ISI][Medline]

Licciardi, F.L., Hung-Ching, L. and Rosenwaks, Z. (1995) Day 3 estradiol serum concentrations as prognosticators of ovarian stimulation response and pregnancy outcome in patients undergoing in vitro fertilization. Fertil. Steril., 64, 991–994.[ISI][Medline]

Mantzoros, C.S. and Moschos, S.J. (1998) Leptin: in search of role(s) in human physiology and pathophysiology. Clin. Endocrinol., 49, 551–567.[ISI][Medline]

Muasher, S.J., Oehninger, S., Simonetti, S. et al. (1988) The value of basal and/or stimulated serum gonadotropin levels in prediction of stimulation response and in vitro fertilization outcome. Fertil. Steril., 50, 298–307.[ISI][Medline]

Metcalf, M.G. and Livesy, J.H. (1985) Gonadotropin excretion in fertile women: effect of age and the onset of the menopausal transition. J. Endocrinol., 105, 357–362.[Abstract]

Munster, K., Schmidt, L. and Helm, P. (1992) Length and variation in the menstrual cycle|a cross-sectional study from a Danish county. Br. J. Obstet. Gynaecol., 99, 422–429.[ISI][Medline]

Pearlstone, A.C., Fournet, N., Gambone, J.C. et al. (1992) Ovulation induction in women age 40 and older: the importance of basal follicle-stimulating hormone level and chronological age. Fertil. Steril., 58, 674–679.[ISI][Medline]

Reyes, F.I., Winter, J.S.D. and Faiman, C. (1977) Pituitary–ovarian relationships preceding the menopause. I. A cross-sectional study of serum follicle-stimulating hormone, luteinizing hormone, prolactin, estradiol, and progesterone levels. Am. J. Obstet. Gynecol., 129, 557–564.[ISI][Medline]

Santoro, N., Rosenberg Brown, J., Adel, T. et al. (1996) Characterization of reproductive hormonal dynamics in the perimenopause. J. Clin. Endocrinol. Metab., 81, 1495–1501.[Abstract]

Schwartz, D. and Mayaux, M.J. (1982) Female fecundity as a function of age: results of artificial insemination in 2193 nulliparous women with azoospermic husbands. Federations CECOS. N. Engl. J. Med., 306, 404–406.[ISI][Medline]

Scott, R.T., Toner, J.P., Muasher, S.J. et al. (1989) Follicle-stimulating hormone levels on cycle day 3 are predictive of in vitro fertilization outcome. Fertil. Steril., 51, 651–654.[ISI][Medline]

Sherman, B.M., West, J.H. and Korenman, S.G. (1976) The menopausal transition: analysis of LH, FSH, estradiol, and progesterone concentrations during menstrual cycles of older women. J. Clin. Endocrinol. Metab., 42, 629–636.[Abstract]

Smotrich, D.B., Widra, E.A., Gindoff, P.R. et al. (1995) Prognostic value of day 3 estradiol on in vitro fertilization outcome. Fertil. Steril., 64, 1136–1140.[ISI][Medline]

Treloar, A.D., Boynton, R.E., Benn, B.G. et al. (1967) Variation of the human menstrual cycle through reproductive life. Int. J. Fertil., 12, 77–126.[ISI][Medline]

Van Santbrink, E.J.P., Hop, W.C., van Dessel, T.J.H.M. et al. (1995) Decremental follicle-stimulating hormone and dominant follicle development during the normal menstrual cycle. Fertil. Steril., 64, 37–43.[ISI][Medline]

Van Voorhis, B.J., Syrop, C.H., Hammitt, D.G. et al. (1992) Effects of smoking on ovulation induction for assisted reproductive techniques. Fertil. Steril., 51, 609–615.

Velasco, E., Malacara, J.M., Cervantes, F. et al. (1990) Gonadotropins and prolactin serum levels during the perimenopausal period: correlation with diverse factors. Fertil. Steril., 53, 56–60.[ISI][Medline]

Welt, C.K., McNicholl, D.J., Taylor, A.E. et al. (1999) Female reproductive aging is marked by decreased secretion of dimeric inhibin. J. Clin. Endocrinol. Metab., 84, 105–111.[Abstract/Free Full Text]

Witt, B.R., Barad, D.H., Barg, P. et al. (1995) Basal serum follicle stimulating hormone (FSH) and estradiol levels as predictors of pregnancy in unstimulated donor insemination cycles. J. Assist. Reprod. Genet., 12, 157–160.[ISI][Medline]

Submitted on February 2, 2001; accepted on September 18, 2001.