| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Unipath Ltd. (F.M., J.C., P.W.P., J.E.E.), Bedford, United Kingdom MK44 3UP; Unilever Research Colworth (L.J.A.), Bedford, United Kingdom MK44 1LQ; and Parkwood Clinic (S.W.P.), Bournemouth, United Kingdom BH7 7DW
Address all correspondence and requests for reprints to: Dr. F. Miro, Unipath Ltd., Priory Business Park, Bedford, United Kingdom MK44 3UP. E-mail: fernando.miro{at}unipath.com.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Due to its complexity, there are substantial difficulties in investigating the endocrinology of the menopausal transition. A few studies, however, have revealed some important features. Remarkably, the first half of these cycles is characterized by intervals with a substantial rise in circulating levels of FSH, with minimal estradiol production or inhibin (11, 12). Also, during the luteal phase, progesterone levels are generally lower, and often there is no significant rise.
Studies based on measurements of urinary hormones have shown results similar to those for serum. These studies have established the usefulness of analysis of urinary hormones in the study of the menstrual cycle and have focused attention on other abnormalities in these elongated cycles, with potentially significant health consequences, such as the existence of an increased estrogen to pregnanediol ratio during the luteal phase (13, 14, 15, 16).
Although the work performed to date provides important clues for understanding the endocrinology of the menopausal transition, it is limited to the description of hormonal patterns in a handful of cycles. To determine general trends of the origin, consequences, and progression of the menopausal transition, relatively large quantitative studies are required.
Thus, we undertook this study with the objective of determining 1) the relevance of the timing in ovarian response to the elongation of the menstrual cycle during the menopausal transition, 2) whether the changes in ovarian response are due to insufficient stimulus from the pituitary or temporary impeded responsiveness from the ovary, and 3) the consequences of the elongation of the cycle to steroid profiles during the luteal phase. To achieve this, we examined profiles of four major reproductive urinary hormones, FSH, LH, estrone 3-glucuronide (E1G), and pregnanediol 3-glucuronide (PdG), over several menstrual cycles in 34 women with perimenopausal menstrual irregularity. The main parameters investigated were the onset of ovarian response during the follicular phase, the variation in FSH production before the onset of ovarian response, and midluteal levels of PdG, E1G, and LH.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The data presented here are part of the FREEDOM study, which included a population of 112 healthy Caucasian women recruited in response to advertisements. Volunteers were selected by interview and were assigned to one of four preliminary clinical groups (fertile, premenopausal, perimenopausal, and postmenopausal) on the basis of menstrual history, vasomotor symptoms, and other symptoms, such as premenstrual changes. Exclusion criteria were the use of hormonal contraception or hormone replacement therapy during the trial or in the 6 months before study entry, pregnancy, breastfeeding, hysterectomy, or a past history of diagnosed pituitary disorder. Also excluded were women taking any medication known to interfere with the secretion and action of reproductive hormones. Each volunteer provided urine samples for a period of 6–18 months.
The present study is based on data from 34 women with perimenopausal menstrual irregularity, Stages of Reproductive Aging Workshop stages –2 to –1 (17). Ages ranged from 40–53 yr, with a median body mass index of 25.
The study was performed in accordance with current guidelines on good practice in clinical research and the Declaration of Helsinki. Ethical approval for the study was obtained from the Unilever Research Laboratory Colworth and the east Dorset local medical research ethical committees. Informed, written consent was obtained from all volunteers.
Hormonal analyses
Each volunteer provided daily urine samples (first morning void), which were collected in universal containers with sodium azide as preservative (0.1%). Volunteers were asked to keep the samples at 4 C until delivery to the laboratory (on a weekly basis). At arrival, the specimens were aliquoted and stored at 4 C until analyzed. All samples were analyzed within 3 months of arrival at the laboratory. Stability studies carried out in our laboratory show no significant loss of immunoactivity in the samples for any of the hormones measured for at least 1 yr.
Each sample was analyzed for FSH, LH, estrone 3-glucuronide (E1G), and pregnanediol 3-glucuronide (PdG). The assays were carried out by immunoassay, using AutoDelfia (Perkin-Elmer Life Sciences, Cambridge, UK) with highly specific in-house developed antibodies and following in-house established and validated protocols (18).
The sensitivity of the FSH assay was 0.17 IU/liter. The intraassay coefficients of variation (CVs) were 5.1%, 2.4%, and 1.8% for concentrations of 1.8, 8.2, and 42.9 IU/liter, respectively, whereas the interassay CVs at the same concentrations were 4.0%, 3.4%, and 2.4%. The sensitivity of the LH assay was 0.09 IU/liter. The intraassay CVs were 6.4%, 3.6%, and 2.0% for concentrations of 2.3, 10.6, and 51.6 IU/liter, respectively, whereas the interassay CVs at the same concentrations were 10.0%, 4.8%, and 1.6%, respectively. The sensitivity of the E1G assay was 0.21 nM. The intraassay CVs were 5.1%, 2.5%, and 2.2% for concentrations of 3.4, 17.1, and 85.4 nM, whereas the interassay CVs were 4.2%, 2.6%, and 2.7%, respectively. The sensitivity of the PdG assay was 0.14 µM. The intraassay CVs were 15.6%, 7.8%, and 7.7% for concentrations of 1.6, 8.1, and 40 µM, whereas the interassay CVs were 6.6%, 1.6%, and 4.9%, respectively.
Data preparation and analyses
When all hormonal measurements were completed, we developed graphic profiles for all hormones, adjusted, and smoothed. The statistical approach used to smooth and adjust the profiles was based on the obtained PdG curve. Briefly, a smooth curve was fitted to ln(PdG), and the residuals were used to adjust the other hormones. This was achieved using the SAS/IML Splinec routine with a smoothing parameter of 10 (19). This retrospective approach was found to have equivalent effects to creatinine adjustment (19A ). FSH and LH results are expressed as international units per liter, E1G as nanograms per milliliter, and PdG as micrograms per milliliter.
Parameters investigated
Cycle length. Cycle length was considered the length of the interval between two consecutive menstrual bleeding episodes, with d 1 as the first day of menstrual bleeding. Spotting was identified from diary entries; it was sporadic and usually occurring on the days after menstrual bleeding. Midcycle bleeding was distinguished in retrospective from d 1 bleeding, because it was unrelated to a drop in PdG or E1G levels. There were 13 episodes of midcycle bleeding, six of them associated with high levels of FSH and low E1G and PdG levels, five in the presence of increasing E1G levels; there was an episode of bleeding 2 d after a real menstrual period; and finally, there was a case with 2 d of bleeding 3 d after an LH peak.
Day of E1G take-off (ETO). ETO was used to estimate the timing of the ovarian response. It indicates the time taken from d 1 of the cycle to the start of the first sustained rise in E1G and is estimated through the application of an algorithm, as described previously (18). Briefly, the algorithm was applied to smoothed-adjusted values of E1G throughout the menstrual cycle and measured the magnitude of the increase in E1G for each day in the cycle with a positive slope, excluding the last 10 d of the cycle.
Mean FSH levels before E1G take-off (FSHETO). This parameter was used to determine changes in FSH secretion in relation to the onset of the ovarian response. It is estimated as the mean value of FSH from the first day of the cycle to ETO.
Midluteal PdG. This was the maximum value of PdG during the cycle. Because the smoothing process produced symmetrical profiles of luteal PdG, the maximum PdG values typically occurred at the midluteal phase. This parameter was used to estimate the ability of the corpus luteum to produce progesterone.
Midluteal E1G. Midluteal E1G was used to estimate estrogen production during the luteal phase. It was determined as the value of E1G concomitant to maximum PdG.
Midluteal LH. Midluteal LH was used to estimate the levels of LH during the luteal phase. It was determined as the value of LH on the day of maximal PdG.
Statistical analyses
To determine the impact of delayed ETO on cycle elongation, we divided the cycle into two intervals: 1) from the beginning of the cycle to ETO, and 2) the rest of the cycle. For each individual, the SD for total cycle length, ETO, and total length-ETO were calculated. The values obtained were compared using the Sign test, contrasting the hypothesis that ETO is the most variable interval in the cycle, and thus: SD (ETO) > SD (length of cycle – ETO).
To determine the variation in FSHETO with increasing ETO, the distribution of ETO was divided into four intervals according to quartiles, and the adjusted mean values of FSHETO for each interval were compared. To take into account subject differences, two-way ANOVA was used. The same approach was used to determine PdG and E1G variations in relation to ETO (as well as FSHETO), and LH in relation to PdG.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
20 ng/ml) with considerably reduced FSH levels. Because of the existing negative feedback between estrogens and FSH, we excluded these cycles from the analysis of FSH. Four other cycles were excluded because there was an abundant number of missing samples. Overall, the median number of cycles analyzed per volunteer was nine, and the median and mean length of the cycle were, respectively, 28 and 34.8 d, with a range of 167 d. Elongated cycles
All volunteers showed at least one elongated cycle in their profiles. These cycles were chiefly characterized by an increased duration in the period before ETO, and presented other particular features too. The difference between this type of cycle and the normal type is shown in Fig. 1
. Cycle 1 is a case of typical duration (28 d), whereas cycle 2 is an elongated cycle (47 d). ETO occurs considerably later in cycle 2 (d 20 vs. d 7), and FSH reaches much higher levels (FSHETO is 14.7 IU/liter in cycle 1 and 44.31 IU/liter in cycle 2). During the luteal phase, maximum values for PdG are lower in cycle 2 (7.7 vs. 18.7 µg/ml), whereas E1G levels are higher (56.3 vs. 24.4 ng/ml). The mean (±SD) cycle length values after ETO were 20.95 ± 4.34 for typical cycles and 27.95 ± 11 for elongated cycles.
|
Application of the Sign test revealed a much higher magnitude of SD for the interval for ETO than for the rest of the cycle (P < 0.0001), and indeed, there was a very strong association between the SD for ETO and the total length of the cycle (Fig. 2).
|
ETO distribution was divided into quartiles, and the adjusted mean FSHETO values for the intervals were compared. A simultaneous increase in both variables was found [F(3,252) = 47.56; P < 0.0001; Fig. 3].
|
The CV values for PdG and E1G were 79 and 49, respectively. To investigate the relationship between delayed ETO and midluteal levels of PdG and E1G, we applied the same approach used previously for ETO and FSHETO. In this case, a gradual decrease in PdG levels in parallel with increasing ETO was found [F(3, 252) = 7.05; P < 0.0001; Fig. 4A]. By contrast, an increase in E1G was found in association with ETO [F(3, 252) = 11.85; P < 0.0001; Fig. 4B].
|
To investigate the variation in midluteal PdG and E1G in relation to increasing FSHETO levels, the same approach as that for ETO was applied. The results were comparable to those found for ETO, with an inverse relationship between FSHETO and PdG levels [F(3, 252) = 8.24; P < 0.0001] and a direct one with E1G [F(3, 252) = 16.67; P < 0.0001; Fig. 5].
|
To determine the nature of the PdG deficiency observed in relation to the increased delay in ovarian response, we compared the variations in luteal PdG and LH. When the distribution of midluteal PdG values was divided into four intervals according to its quartiles, and the adjusted mean values of LH for each interval were compared, a marked inverse correlation between both variables was found [F(3, 252) = 18.76; P < 0.0001; Fig. 6].
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A prominent feature of the perimenopausal elongated cycle is the existence of a lag phase in ovarian response during the follicular phase. This feature has been described as a prolonged hypoestrogenic phase at the beginning of the cycle (13) or as the inactive phase of the cycle (20). To determine the significance of this lag phase in the elongation of the menstrual cycle, we defined the variable ETO to measure the onset of ovarian response. The marked parallelism found between the variability in ETO and total cycle length indicates that elongation of the menstrual cycle is fundamentally the result of the delay in ovarian response.
A delayed ovarian response has two major causes: inadequate FSH stimulus from the pituitary, and ovarian refractoriness to FSH. Due to the nature of the pituitary-ovary interaction, the absence of an ovarian response to FSH results in a progressive increase in FSH; indeed, after ovariectomy, FSH levels rise gradually over several days until reaching a maximal value (21). Our results indicate an association between longer ETO and higher FSH levels, which is consistent with a temporary lack of responsiveness to FSH. Elevated FSH during the follicular phase of the elongated cycles has been widely reported (11, 12, 13, 16, 22, 23, 24, 25).
The origin of this temporary refractoriness of the ovary is as yet unknown. Possible causes include an increase in the rate of follicular atresia or a slower rate of follicular growth. Several investigators have concluded that the human ovary undergoes a process of accelerated follicular depletion toward the end of the fourth decade of life (26, 27, 28). Equally, in vitro fertilization results show a steep decline in the rate of success at this age (29), and histological studies have described increased atresia of primordial and primary follicles in the ovaries of women approaching the menopause (30). Although this increase in follicular demise apparently involves early stages, it might disrupt the sequence of follicular development, resulting in the delay in recruitment observed in the elongated cycles.
A deficiency in luteal progesterone (or PdG) in perimenopausal elongated cycles has been noticed in a large number of studies (12, 13, 14, 16, 31, 32). We found a gradual decline in PdG levels with increasing ETO. Although we detected an overall inverse relationship between FSH and PdG, there was no clear gradual pattern allowing the inference of a causal effect for elevated FSH in the reduction of PdG. Future work is required to establish the origin of the reduction in PdG.
The reduction in PdG resembles luteal deficiency. Clinically, there are two recognized origins for this condition: hypothalamic and ovarian (33, 34, 35, 36). We observed an inverse relationship between PdG and LH consequent with an ovarian origin for the luteal defect (33, 34). The occurrence of high luteal LH levels during the perimenopause has been described previously (14). It remains to be determined whether the cause of the reduced PdG is reduced cell number, deficient cell differentiation, or reduced LH responsiveness.
In contrast to PdG, midluteal E1G levels increase as ETO gets longer and FSH higher. This was expected, because abundant research has described increased circulating (37) and excreted (13, 16, 38, 39) luteal estrogen levels in perimenopausal elongated cycles. Because of the disparity with PdG and the lack of an inverse relationship between luteal LH and E1G, the excess estrogens seem to have a source different from the corpus luteum. We found a very robust relationship between increasing FSH and E1G. The elevated FSH might exert a hyperstimulating effect, disturbing the normal process of follicular development. Multicystic follicular growth with hyperestrogenism has been related to exogenous ovarian hyperstimulation (40) and FSH-secreting adenomas (41). Moreover, multicystic follicular growth with high concentration of estrogens (37) and persistence of follicles during the luteal phase (25) have been described in perimenopausal women.
Additional research is required to better characterize as well as determine the precise causes of the hormonal abnormalities in these elongated cycles. Studies designed to partially suppress FSH levels before the onset of ETO might be useful in determining the contribution of this gonadotropin. Equally, detailed ultrasound follow-up should provide first-hand information on follicular growth dynamics for comparison with the hormonal findings. Finally, measurement of other hormonal markers of follicular recruitment and development, in particular inhibin B (42, 43), should bring more accurate information on the onset of the ovarian response.
The possible relationship between perimenopausal hyperestrogenism (further aggravated by reduced progesterone) and morbidity is a matter of concern among some researchers (16, 7, 39, 44, 45). Naturally elevated estrogen levels have been associated with thickening (46) and even cystic glandular hyperplasia of the endometrium (37, 38). Given the characteristics of the elongated cycles, it is paramount to determine the real impact on health of this hormonal imbalance, particularly because nearly 47% of women experience at least 5 yr of menopausal transition (10).
In summary, the menstrual irregularity occurring during the menopausal transition is characterized by the occurrence of elongated cycles. Our study indicates that this elongation is mostly the result of a delay in the onset of ovarian response and suggests that this delay is due to a temporary lack of ovarian responsiveness to FSH. Concomitant with the increase in the delay in ovarian response is a clear tendency toward reduced luteal levels of PdG and increased estrogen levels. A link between increased and elevated FSH appears likely; however, the origin of the reduced PdG is less clear.
|
Footnotes |
|---|
Abbreviations: CV, Coefficient of variation; E1G, estrone 3-glucuronide; ETO, E1G take-off; FSHETO, FSH levels before ETO; PdG, pregnanediol 3-glucuronide.
Received October 2, 2003.
Accepted July 3, 2004.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  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]
|
|
|
|
 |
|
 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]
|
|
|
|
|
|
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]
|
|
|
|
 |
|
 P. M. Meyer, S. L. Zeger, S. D. Harlow, M. Sowers, S. Crawford, J. L. Luborsky, I. Janssen, D. S. McConnell, J. F. Randolph, and G. Weiss Characterizing Daily Urinary Hormone Profiles for Women at Midlife Using Functional Data Analysis Am. J. Epidemiol., April 15, 2007; 165(8): 936 - 945. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. A. MacGregor, A. Frith, J. Ellis, L. Aspinall, and A. Hackshaw Prevention of menstrual attacks of migraine: A double-blind placebo-controlled crossover study Neurology, December 26, 2006; 67(12): 2159 - 2163. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.A. O'Connor, E. Brindle, R.C. Miller, J.B. Shofer, R.J. Ferrell, N.A. Klein, M.R. Soules, D.J. Holman, P.K. Mansfield, and J.W. Wood Ovulation detection methods for urinary hormones: precision, daily and intermittent sampling and a combined hierarchical method Hum. Reprod., June 1, 2006; 21(6): 1442 - 1452. [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 |
