| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Department of Clinical Science (B.-M.L., A.C., G.C.), Division of Obstetrics and Gynaecology, Karolinska Institutet, Huddinge University Hospital, SE-14186 Stockholm, Sweden; and Prince Henry’s Institute of Medical Research (H.G.B., L.B., D.M.R.), Clayton, Victoria, 3168, Australia
Address all correspondence and requests for reprints to: B.-M. Landgren, M.D., Ph.D., Department of Clinical Science, Division of Obstetrics and Gynaecology, Karolinska Institutet, Huddinge University Hospital, SE-14186 Stockholm, Sweden.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
These data indicate that there is a time-related change in the character of menstrual cycles as menopause approaches, with an increasing proportion of cycles observed with prolonged follicular phases that may either be delayed ovulatory cycles or anovulatory cycles. The increase in the proportion of B cycles with elevated early follicular phase FSH levels and low inhibin/FSH ratios toward menopause provides a basis for the apparent early increase in serum FSH and decrease in serum inhibins observed previously in studies of the menopause transition based on sampling confined to the follicular phase only. The data amplify and clarify current concepts of the endocrine basis of the menopause transition.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The variations in circulating FSH levels with increasing age are most probably due to changes in ovarian physiology affecting the secretory pattern of the gonadotrope, and the ovary becomes increasingly resistant to stimulation by gonadotropins, probably due to the decreased number of follicles, which leads to a decline in the production of both estrogens and inhibins. A number of studies (13, 14, 15, 16, 17, 18) have shown low circulating concentrations of inhibin B in older reproductive-age women. Whether the rise in FSH is in part the result of a primary neuroendocrine change or is explicable entirely by a response to falling feedback signals from the aging ovary is controversial (19). The present study aimed to make a detailed longitudinal investigation of the hormonal characteristics of women during transition to menopause by studying the same individuals annually with multiple samples through a single menstrual cycle. The study also aimed to try to identify early indicators of ovarian insufficiency.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Twenty-four women, presenting to the clinic for routine check-up or employed at the hospital, and apparently healthy and normally menstruating, volunteered to participate in the study. The selection criteria required that the volunteers were to be aged between 45 and 47 yr, with regular menstrual cycles and intermenstrual intervals of 26–35 d, and had seen no change in menstrual cycle pattern from previous years. Twenty-four of 30 women who originally volunteered met the selection criteria. The volunteers were 45–47 yr old at commencement of the study, and none had been pregnant, lactating, or using steroid hormones during the last 6 months preceding the study. When enrolled, none was taking any medication. The women had regular menstrual cycles, with cycle lengths between 26 and 35 d. Before entering the study, a general medical examination, including blood pressure, heart rate, hemoglobin, kidney, thyroid, and liver function tests was performed as well as a gynecologic examination including Papanicolaou smear. Clinical characteristics are shown in Table 1
.
|
All participants were seen once yearly until they reached menopause, defined as the spontaneous absence of menses for at least 12 months. Once yearly, blood was collected three times weekly for 4 wk starting with the onset of menses, between 0800 and 1100 h. If no bleeding was observed after 30–35 d, the blood collection was stopped because it was felt that a longer sampling period could not be justified ethically. If no bleeding was observed for at least 10 months, a standard 4-wk blood collection protocol was used. The blood was centrifuged within 30 min, and the serum was stored at –20 C until analyzed. All samples from an individual were analyzed in the same batch. The women kept bleeding diary cards throughout the whole study and brought them to the clinic at each yearly visit when a gynecological and general medical examination were performed.
Endocrine classification of menstrual cycles
Ovarian function was classified according to modifications of the classification system by Landgren and Diczfalusy (20), which is based on the peripheral estradiol (E2) and progesterone (P) levels characterizing the normal follicular and luteal phases of an ovulatory cycle (20). The luteal phase is characterized by a rise in P levels of at least 16 nmol/liter for at least 4 d. Four patterns of ovarian activity are described by this classification during the 4-wk blood collection period:
Type A: absence of ovarian activity; with low levels of E2 and P.
Type B: prolonged follicular phase with no luteal activity during the 4-wk collection period from the beginning of the last menses with E2 greater than 100 pmol/liter and P less than 5 nmol/liter (arbitrary limits adapted here). It is recognized that such cycles may result in a late ovulatory response if the cycle length is much prolonged beyond 4 wk. This outcome could not be assessed in the present study.
Type C: normal follicular activity, followed by insufficient luteal function (P levels > 5 and < 16 nmol/liter).
Type D: cycles with normal ovarian activity.
Because each woman’s menstrual cycle history leading to menopause is known for the duration of the study, the specific cycles in which the serum hormone profiles were determined were designated by a cycle number relative to menopause in which cycle 1 is the last cycle before menopause. Thus, the cycles can be grouped in 10-month calendar periods before menopause, e.g. 1–10, 11–20, etc. This grouping of 10 months was chosen to enable the separation of cycles (e.g. cycles 1 and 12) collected within a 12-month period from the same individual.
The serum hormone data within each D cycle were grouped in 3-d stages centered around the estimated day of ovulation, which was identified from the composite of several hormonal indicators: elevated serum LH levels at midcycle, the initiation of a rise in serum P, and the inflection in serum P levels in the early luteal phase.
These stages were as follows:
Stage –4 (d 11–13 before estimated day of ovulation).
Stage –3 (d 8–10 before estimated day of ovulation).
Stage –2 (d 5–7 before estimated day of ovulation).
Stage –1 (d 2–4 before estimated day of ovulation).
Stage 0 (d (–1, 0, +1 around the estimated day of ovulation).
Stage +1 (d 2–4 after the estimated day of ovulation).
Stage +2 (d 5–7 after the estimated day of ovulation).
Stage +3 (d 8–10 after the estimated day of ovulation).
Group +4 (d 11–13 after the estimated day of ovulation).
As a separate analysis to compare hormone profiles between D and B cycles, the data from both B and D cycles were grouped according to the first day of the cycle (day of menses) and were averaged in 5-d groupings (1–5, 6–10, etc.).
Hormone assays
Serum levels of FSH and LH were measured by RIA using reagents from the Farmos Group (Oulu, Finland) at the Central Laboratory for Clinical Chemistry, Karolinska Hospital. Standards were the World Health Organization (WHO) Second International Reference preparation for FSH for Bioassay, no. 78/549, and the WHO Second IRP-HMG for LH. The sensitivity of the FSH assay was 0.7 IU/liter, and the LH assay was 0.2 IU/liter. The intraassay and interassay coefficients of variation were 2.7 and 6.3%, respectively, and 4.9 and 5.8%, respectively. E2 and P levels were determined by RIA using reagents from Diagnostic Products Corp. (Los Angeles, CA). The intraassay and interassay coefficients of variation were 7.2 and 6.8 and 4.9 and 5.8%, respectively. Prolactin was measured by RIA with reagents obtained from Hybritech (San Diego, CA). The intraassay and interassay variations were 5.4 and 4.9%.
Serum total inhibin was determined by RIA (9). Serum inhibin A was measured using 
-ßA subunit dimeric ELISA (19) with the WHO inhibin A preparation (91/624) used as standard. The assay sensitivity was 4 ng/liter. The within- and between-assay variation were 9 and 17% (n = 8 assays), respectively. Serum inhibin B concentrations were measured using the -ßB dimeric ELISA (21) using a human inhibin B preparation isolated from human follicular fluid provided by N. Groome (Oxford Brookes University, Oxford, UK) as standard. The assay sensitivity was 4 ng/liter. The within- and between-assay variations were 11.7 and 15.6%, respectively.
Ethical aspects
The study was approved by the local Ethics Committee of Karolinska sjukhuset, Stockholm, Sweden.
Statistics
The hormone data were log transformed before analysis. Comparisons within and between cycles/patient were assessed by one-way ANOVA with repeated measures with corrections for missing entries (SigmaStat, SPSS, Gorinchem, The Netherlands). In some cases a two-way ANOVA was used. Where normality criteria were not satisfied, the Mann-Whitney rank test was used.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
and Subjects and Methods. Although 24 women entered the study, 11 of the 24 women were not included for the final hormonal evaluations. Reasons for exclusion were: two dropped out for social reasons; one moved from Stockholm; four started hormonal replacement treatment after 1–2 yr in the study while still menstruating due to vasomotor symptoms; two presented with metrorrhagia, one of whom had a hysterectomy and one a levonorgestrel releasing intrauterine device inserted; one woman was treated for breast cancer; and one died from gastric cancer. The 11 women who withdrew from the study did not differ from the remaining 13 with respect to menarche, pregnancies, parity, menstrual duration, and menstrual interval (data not shown).
|
|
Menstrual cycle categories were assigned on the basis of E2 and P concentrations. Only two cycles in the study showed luteal phase insufficiency with maximum luteal phase P levels of 5.1 and 5.3 nmol/liter. The possibility that some B cycles may have shown luteal insufficiency cannot be excluded because of the 4-wk sampling protocol used. There were no A cycles recorded.
Cycle lengths of B and D cycles when averaged/subject (51.8 ± 25.8 vs. 35.4 ± 26.0 d, n = 9 and 13, respectively, mean ± SD) were not significantly different. However, cycle lengths for all B and D cycles were significantly different (52.1 ± 36.2 vs. 30.1 ± 14.5 d, n = 19 and 58, respectively, P < 0.005). As seen in Table 2, the length of D cycles between 30 and 90 cycles before menopause is relatively stable [27.1 (18.6–39.3) d; geometric mean ± 2 SD]; however, in the last 10 cycles, four of the five D cycles showed cycle length of 39 d or more in comparison with seven of eight B cycles with more than 39 d for the last 10-cycle period.
Hormonal patterns in menstrual cycles in women approaching menopause
The 11 women (of the total of 13) in whom final menses were accurately defined, exhibited D cycle characteristics more than 30 cycles from menopause (corresponding to 2–4 yr before menopause). The proportion of D cycles then decreased to 62, 20, and 25% with successive 10-cycle groupings approaching menopause. The decrease in proportion of D cycles was associated with an increase in B cycles beginning in the 21–30 cycle group before menopause. This data are presented in Fig. 1 and Table 2. The transition between D and B cycles between women can occur between 1 and 27 cycles before menopause but can, in some instances, be followed by the occurrence of additional D cycles. Two cycles (B and D) were not followed by menstruation, and these cycles were defined as xxx and not compared with those cycles before menopause.
|
The serum hormone levels in each D cycle throughout the study period were organized with reference to the LH midcycle peak [with groups designated as group LH –4, –3, –2, –1, LH peak (LH 0), LH +1, LH +2, LH +3, LH +4]. The hormonal data were then combined into groups according to their cycle number from menopause (1–10, 11–20, 21–30, etc., Fig. 2) and their geometric means (± 2 SD) calculated. A statistical comparison of hormone levels for each group of the menstrual cycle was undertaken by one-way-ANOVA with repeated measures across all cycle groups. Significant differences were observed for FSH but not LH, E2, P, prolactin, total inhibin, and inhibin A and B.
|
These were indeed ovulatory cycles established from the hormonal pattern observed during the initial 30–35 d of sampling in which clear-cut luteal phases with normal serum P levels were noted. However, the fall in E2 and P was not followed by menstrual bleeding, which occurred much later.
B cycles
The limited number of B type cycles identified in the study (n = 16–17) prevented a longitudinal assessment of changes in hormone levels within subject as shown in D cycles. Thus, all B cycles were assessed as a single group, and statistical comparisons were undertaken between all B and D cycles (n = 19) matching the same cycle number from FMP (between 1 and 30 cycles from FMP) (Fig. 3). When B and D cycles were organized in 5-d groups from the first day of menses in both B and D cycles, serum FSH and LH were significantly elevated at the beginning of the menstrual cycle in the B cycles, compared with the D cycles, and remained elevated throughout the 4-wk collection period. Serum P was low (mean values, 1.2–1.8 nmol/liter) throughout the B cycles, whereas a cyclical pattern of P values was observed in D cycles peaking at 19–21 nmol/liter between d 16 and 25 of the cycle (Fig. 3). Serum E2, total inhibin, inhibin A, and inhibin B were also significantly lower in B cycles but not uniformly across the cycle. Serum prolactin was not significantly different between B and D cycles (Fig. 3).
|
A comparison was undertaken between serum hormone levels and the ratio of FSH to inhibin B, inhibin A, and total inhibin for d 1–5 of the cycle for l B 1–27 and D cycles 1–30 cycles from menopause. Whereas serum inhibin A and B were not significantly different between B and D cycles, serum FSH, LH, and FSH/inhibin B and FSH/total inhibin ratios were significantly elevated in B cycles.
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
D cycles
It is apparent from the serum hormone profiles of cycles with evidence of normal P concentrations that they are regular in terms of cycle length, with exception of the final cycles. It was anticipated from earlier studies (17) in which serum was collected between cycle d 1–5 of normal cycling women at various ages (20–50 yr) of their reproductive life that serum FSH and possibly LH would increase toward menopause and inhibin B and then inhibin A would decrease, whereas serum E2 would be the last to change. However, it is clear from the present studies that no significant changes were seen in these hormones in any phase of D cycles when examined over 4–9 yr before FMP, with the exception of luteal-phase FSH levels in cycles 10–20 from menopause. In fact, visual inspection and confirmation by statistical analyses of the data (Fig. 3) showed a high degree of uniformity between cycles in women as they approach menopause. We conclude that D cycles remain essentially unchanged in their hormonal parameters over the last 36 cycles before menopause. This finding cannot be compared directly with previous studies (e.g. Ref. 4, 5, 6) in which FSH levels in older women were observed to be raised in comparison with those in younger women because our study covers only the last years before final menses.
B cycles
B cycles were characterized by an absence of a luteal phase, elevated gonadotropins, and increased length. It should be noted that sampling was for a 4-wk period, although the cycle length was prolonged (average, 10 wk) in the majority of cases. The possibility that ovulation may have occurred subsequently to sampling in subjects with long B cycles cannot be excluded. These cycles are comparable with those that may be seen in women taking low-dose P alone for contraception (20). FSH in particular, is elevated early in the cycle. These findings suggest that ovarian events early in the cycle determine the occurrence of either an anovulatory cycle or a cycle with delayed ovulation and lead to an early elevation in serum FSH in particular. The occurrence of B cycles may be either the result of the failure of follicular recruitment or the recruitment of antral follicles that are defective in either the oocyte or the associated granulosa cells. The proposed defect may relate to a reduction in granulosa cell number or sensitivity of granulosa cells to gonadotropic stimulation, leading to lower serum inhibin and E2 levels and hence elevated FSH and LH levels. As a consequence, there will be a failure to ovulate and an absence of a luteal phase at least over the period of blood sampling. The increase in serum E2 toward the end of the B cycles, as opposed to the decrease in D cycles (Fig. 3), can be attributed to the elevated LH drive at this stage of the cycle stimulating the residual granulosa cells (5). It can be postulated that after a prolonged period of exposure to elevated gonadotropins, one or more follicles become responsive, and hence E2 levels rise. A recent report (22) has shown that in women approaching menopause, some individual cycles showed an early increase in gonadotropins with low estrogens. With time the estrogen levels rose, leading to ovulatory-type cycles, which may correspond to the B cycles examined in this study. Future studies are proposed to establish whether such cycles show abnormal follicular development as assessed by ultrasound.
It is interesting to note that D cycles also tended to become irregular in length in the last 10-cycle period before menopause, although the hormonal patterns appeared normal. Even though the number of observations is low, these findings suggest that cycle irregularity occurs in the final cycles before menopause, irrespective of whether apparent normal hormonal activity is evident. It can be concluded that as menopause approaches, menstruation may fail to occur despite a fall in the concentrations of E2 and P, as noted previously by Metcalf et al. (7).
Can the present findings showing changing patterns of D and B cycles approaching menopause provide an explanation for the changing patterns of serum inhibin A and B and serum gonadotropins in age studies of women approaching menopause (5, 17)? A comparison between the present study and the serum age study of Burger et al. (17) shows some support for this possibility. Changes in serum FSH and inhibin A are consistent with the changing patterns of B and D cycles approaching menopause. It should be noted that in the study by Lee et al. (5), women in the 46- to 50-yr age group showed significantly elevated follicular phase and midcycle serum FSH concentrations together with slightly raised E2 and unchanged luteal-phase P. The data were, however, presented as grouped data, with marked variability in hormone concentrations. The possibility cannot be excluded that some of the subjects in Lee’s study actually had defective or absent luteal-phase P concentrations and that their study may therefore have included a mixture of D and B cycles. Other studies (4, 5, 6) compared FSH levels in older women with those in younger controls, whereas in the present study, no rise in FSH was found in ovulatory cycles within 9 yr of final menses.
Can one readily identify a B cycle? Retrospectively, in women with regular/irregular menstrual cycles, serum FSH and LH levels are highly elevated, serum P values are suppressed at least for the first 30 d, and serum E2 levels increase with time. In addition, because serum inhibin B and total inhibins are lower in B cycles, whereas serum FSH and LH are elevated, a ratio of FSH/inhibin B or FSH/total/inhibin may improve the discriminatory power of the analysis. To assess whether a cycle is a B or D cycle prospectively, a combination of gonadotropin measurements with an ovarian ultrasound assessment for follicle size and count probably would be discriminatory, although at this stage, it is conjecture to presume that failure of follicle development is the cause of the abnormal hormonal patterns in B cycles. Clearly, additional studies are required on this issue.
In conclusion, the present detailed menstrual cycle study of a small number of women as they approach menopause demonstrates that within 30 cycles of final menstruation, there is an increased frequency of apparently anovulatory cycles characterized by elevated follicular-phase FSH concentrations. In women who continued to have hormonal evidence of ovulatory cycles, follicular-phase FSH did not consistently increase. Thus, previous studies indicating a progressive increase in follicular-phase serum FSH with increased reproductive age reflect the fact that menstrual cycle characteristics may be heterogeneous and that the occurrence of an anovulatory cycle (or a cycle with delayed ovulation) is a determinant of elevated FSH concentrations. The data indicate that, in addition to the menstrual cycle irregularity characteristic of the menopause transition, there is hormonal heterogeneity.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Abbreviations: E2, Estradiol; FMP, final menstrual period; P, progesterone; WHO, World Health Organization.
Received May 12, 2003.
Accepted March 4, 2004.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  The ESHRE Capri Workshop Group Female contraception over 40 Hum. Reprod. Update, November 1, 2009; 15(6): 599 - 612. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A.L. Chand, C.A. Harrison, and A.N. Shelling Inhibin and premature ovarian failure Hum. Reprod. Update, September 14, 2009; (2009) dmp031v1. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. J. Broekmans, M. R. Soules, and B. C. Fauser Ovarian Aging: Mechanisms and Clinical Consequences Endocr. Rev., August 1, 2009; 30(5): 465 - 493. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. R. Sowers, H. Zheng, D. McConnell, B. Nan, S. Harlow, and J. F. Randolph Jr. Follicle Stimulating Hormone and Its Rate of Change in Defining Menopause Transition Stages J. Clin. Endocrinol. Metab., October 1, 2008; 93(10): 3958 - 3964. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Sowers, H. Zheng, D. McConnell, B. Nan, S. D. Harlow, and J. F. Randolph Jr. Estradiol Rates of Change in Relation to the Final Menstrual Period in a Population-Based Cohort of Women J. Clin. Endocrinol. Metab., October 1, 2008; 93(10): 3847 - 3852. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Santoro, S. L. Crawford, W. L. Lasley, J. L. Luborsky, K. A. Matthews, D. McConnell, J. F. Randolph Jr., E. B. Gold, G. A. Greendale, S. G. Korenman, et al. Factors Related to Declining Luteal Function in Women during the Menopausal Transition J. Clin. Endocrinol. Metab., May 1, 2008; 93(5): 1711 - 1721. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Dennerstein, P. Lehert, H. G. Burger, and J. R. Guthrie New findings from non-linear longitudinal modelling of menopausal hormone changes Hum. Reprod. Update, November 1, 2007; 13(6): 551 - 557. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H.G. Burger, G.E. Hale, D.M. Robertson, and L. Dennerstein A review of hormonal changes during the menopausal transition: focus on findings from the Melbourne Women's Midlife Health Project Hum. Reprod. Update, November 1, 2007; 13(6): 559 - 565. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
The ESHRE Capri Workshop Group Endometrial bleeding Hum. Reprod. Update, September 1, 2007; 13(5): 421 - 431. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. E. Hale, X. Zhao, C. L. Hughes, H. G. Burger, D. M. Robertson, and I. S. Fraser Endocrine Features of Menstrual Cycles in Middle and Late Reproductive Age and the Menopausal Transition Classified According to the Staging of Reproductive Aging Workshop (STRAW) Staging System J. Clin. Endocrinol. Metab., August 1, 2007; 92(8): 3060 - 3067. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. H. Eliassen, S. A. Missmer, S. S. Tworoger, D. Spiegelman, R. L. Barbieri, M. Dowsett, and S. E. Hankinson Endogenous steroid hormone concentrations and risk of breast cancer among premenopausal women. J Natl Cancer Inst, October 4, 2006; 98(19): 1406 - 1415. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Grewal, M. R. Sowers, J. F. Randolph Jr., S. D. Harlow, and X. Lin Low Bone Mineral Density in the Early Menopausal Transition: Role for Ovulatory Function J. Clin. Endocrinol. Metab., October 1, 2006; 91(10): 3780 - 3785. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. L. Jones, J. K. Findlay, P. G. Farnworth, D. M. Robertson, E. Wallace, and L. A. Salamonsen Activin A and Inhibin A Differentially Regulate Human Uterine Matrix Metalloproteinases: Potential Interactions during Decidualization and Trophoblast Invasion Endocrinology, February 1, 2006; 147(2): 724 - 732. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V Wiseman, L Conteh, and F Matovu Using diaries to collect data in resource-poor settings: questions on design and implementation Health Policy Plan., November 1, 2005; 20(6): 394 - 404. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
