Human Reproduction, Vol. 15, No. 9, 1909-1912,
September 2000
© 2000 European Society of Human Reproduction and Embryology
Expected contribution to serum oestradiol from individual ovarian follicles in unstimulated cycles
1 University of Bristol Centre for Reproductive Medicine, St Michael's Hospital, Southwell Street and 2 University of Bristol Division of Child Health, Institute of Child Health, Royal Hospital for Sick Children, Bristol BS2 BEG, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Data relating serum oestradiol concentration to follicle size in unstimulated cycles are lacking. We provide precise data on serum concentrations expected for any follicle diameter (FD) in the mid- to late follicular phase. Infertile women (n = 35) with apparently normal ovulatory cycles were studied in detail in 128 unstimulated monofollicular cycles leading to IVF. Using mathematical modelling to account for repeated cycles in the same woman, the relationship between serum oestradiol and FD was explored and reference ranges for serum oestradiol at individual FD were calculated. Serum oestradiol concentrations [number of patients, geometric mean, 95% confidence interval (CI)] at the onset of the LH surge were higher in `fertilized' cycles (73, 1279, 1180–1378 pmol/l) compared with `unfertilized' cycles (31, 1055, 929–1197 pmol/l, P = 0.008) and `no oocyte' cycles (24, 1064, 922–1227 pmol/l, P = 0.03) respectively. In `fertilized' cycles, oestradiol concentrations rose exponentially with FD and for each size of follicle the oestradiol distribution was skewed. Functional oocyte competence varied in apparently normal ovulatory cycles and was correlated with pre-ovulatory serum oestradiol but not FD. Serum oestradiol varies within wide limits for maturing follicles of any given diameter prior to the onset of the LH surge.
Key words: competent oocytes/follicle diameter/monofollicular unstimulated cycles/serum oestradiol
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Serum oestradiol concentrations are commonly used as an indicator of follicle development in ovulation induction and assisted conception programmes. They are also of use, in conjunction with LH testing, for timing of natural intercourse in monitored cycles or for donor insemination. Surprisingly, definitive studies to relate oestradiol concentrations to follicle diameter (FD) in unstimulated cycles are few and small (Hackeloer et al., 1979
; Muasher et al., 1990; Paulson et al., 1990). As discussed in more detail later, an early study using abdominal scanning showed good correlation between oestradiol concentration and overall follicular growth but lacked specific measurements for particular follicle sizes (Hackeloer et al., 1979). Other reports suggested arbitrary reference values, not based on observations (Muasher et al., 1990; Paulson et al., 1990). More recent detailed use of transvaginal scanning has examined the role of FSH and follicle recruitment, though follicle output of oestradiol was not examined (van Santbrink et al., 1995). The present study aims to define reliable reference values based on a large number of cycles of proven ovulatory and oocyte competence, using transvaginal ultrasound scanning to measure follicular diameter. |
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The patients studied attended the Reproductive Medicine Clinic at St. Michael's Hospital, Bristol, UK, because of infertility and were taking part in a research programme investigating follicular and oocyte function in unstimulated cycles, including IVF of the pre-ovulatory oocyte. Informed consent was obtained after approval of the programme by the local ethics research committee. All the women had previously normal ovulatory cycles (25–35 days, mid-luteal serum progesterone
30 nmol/l) and the men had normal sperm counts and tests of function. Each woman had blood collected at 08:00 h daily for serum oestradiol measurement from day 9 of the menstrual cycle onwards, so as to ensure that a dominant follicle would be present (van Santbrink et al., 1995). One author (D.J.C.) measured follicle diameters (FD) daily between 09:00 and 11:00 h using a Sonoline SI-250 (Siemens AG, Erlangen, Germany) and 5 MHz vaginal probe. The internal diameters of developing follicles were measured in three perpendicular planes and the mean of these three values was taken as the FD. For this study, FD and serum oestradiol concentrations were measured up to and including the day of onset of the LH surge. To detect the LH surge accurately, blood samples were collected initially twice daily (08:00 h and 20:00) and when FD reached 14 mm, 4 hourly from 08:00 h (except at 04:00 h). The LH surge was deemed to have begun when the concentration rose 180% above the baseline, and continued to rise thereafter (Testart et al., 1981).
Evidence of follicular and oocyte competence was taken as normal two-pronucleate fertilization and progressive embryo cleavage (using laboratory methods as previously described) (Hull et al., 1985). As an end-point, implantation might seem to be the obvious choice to determine follicle competence; however, only a minority of embryos implant. Cycles with an unfertilized oocyte or in which an oocyte could not be collected were initially included to look for possible differences in oestradiol and FD from `fertilized' cycles, analysed on the day of onset of the LH surge. These differences were investigated using a two-tailed Student's t-test on the log-transformed data. If no oocyte was collected, this could have been due to technical difficulties encountered in the single follicle aspiration or because the `follicle' was in fact a follicular cyst. If an oocyte was collected but no embryo resulted, this was unlikely to be related to sperm dysfunction as we had controlled for that. It could have been because endogenous FSH or endogenous LH inadequately stimulated the follicle or because the oocyte recovery procedure was inappropriately timed. Therefore, for the definitive study of the normal relationship between oestradiol and FD, only data from the `fertilized' cycles were used because significant differences were found between the `fertilized`, `non-fertilized' and `no oocyte' groups.
Oestradiol was measured as a daily routine (not in batches) using a fluoroimmunoassay (DELFIA; Wallac UK Ltd., Milton Keynes, UK) as previously described (Harlow et al., 1995). The interand intra-assay coefficients of variation were 13.8 and 7.1% (at 603 pmol/l). The coefficient of variation for ultrasound FD measurement was 1.7% at 11 mm and 1.2% at 20 mm. As is usual in clinical practice, estimations of follicle size were rounded up or down to the nearest whole integer. For example, an FD of 16 mm included all measurements from 15.5 to 16.4 mm.
For statistical analysis, oestradiol values were transformed to log base 10 and the relationship to FD was investigated by a `random intercept' model which allowed for variation between subjects, between repeated cycles in the same subject (a maximum of four in this instance), and repeated measurements within cycles (maximum six). In this way, a formula was proposed for the calculation of serum oestradiol from FD as:
 |
In this model, u, v and e were assumed to be normally distributed variables, with means equal to 0 and variances that were estimated as part of the modelling process.
There was no significant variation in the slope (b), either across women or across cycles per woman (data not shown).
Model fitting was undertaken using SAS v6.12, PROC MIXED, (SAS Institute, Cary, NC, USA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
One hundred and sixty-nine cycles were monitored in 35 women. In 14 cycles (8%), a dominant follicle failed to develop within 3 weeks of the onset of menses and these were excluded from the study. Follicular development up to and including an LH surge was present in the remaining 155 cycles. In these, ovulation occurred in 21 before oocyte collection could be undertaken. In six cycles, there was a second large follicle, which yielded an additional oocyte, and these cycles were excluded; in only two of these cycles did either oocyte fertilize. In the remaining 128 monofollicular cycles, an oocyte was not collected in 24 (19%), an oocyte was collected but failed to fertilize in 31 (24%) and an oocyte was collected, fertilized in-vitro and embryo transfer took place in 73 (57%).
FD and oestradiol concentration were measured on the day of the onset of the LH surge in all 128 monofollicular cycles and the results are summarized in Table I. The distribution of oestradiol was skewed but normalization of the distribution by log-transformation indicated a coefficient of variation of only 5%, compared with 10–13% for FD. Mean FD appeared to be smaller in the `no oocyte' group by 0.8 mm compared with the other groups but the difference was not significant. However, mean oestradiol was significantly lower in the `no oocyte' and `unfertilized' cycles compared with the `fertilized' cycles (P = 0.03 and P = 0.008 respectively). Therefore further analysis was restricted to the `fertilized' group.
|
During the late follicular phase in the 73 monofollicular `fertilized' cycles, there were 289 paired oestradiol and FD measurements (73 cycles from 33 women, with up to four cycles per woman and up to six results per cycle). Selected results from individual cycles are illustrated in Figure 1
to demonstrate a consistent exponential rise in oestradiol with increasing FD. In Figure 2, the relationship between serum oestradiol and FD is shown (r2 = 0.494).
|
|
Table II
gives the calculated range of serum oestradiol values about each FD, from 10 to 24 mm. The distribution was skewed about a geometric mean. Using restricted maximum likelihood (REML) and the statistical model formula presented above, the following relationships were found:
 |
 |
). In our data set, 14/289 measurements (4.8%) lay outside these limits.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The present study offers information on the relationship between serum oestradiol concentrations and ovarian follicle diameter (FD) during the mid- to late follicular phase. These data are relevant to current ultrasound practice, large enough to be reliable, and also biologically specific to follicles which were demonstrably competent from fertilization of the oocyte they contained. Oocytes were collected at a specific and consistent time after the onset of the mid-cycle LH surge.
Failure to collect an oocyte or failure of an oocyte to fertilize may occur for several reasons, as discussed previously. Data from such cycles were excluded, as they were likely to represent sub-optimal follicular development. This is supported by the significantly lower serum oestradiol concentrations that were measured in these cycles.
Previous studies relating serum oestradiol to ovarian FD are based on small numbers of observations using transabdominal ultrasound scanning. A close correlation between FD and serum oestradiol concentration was shown but this was based on only 15 cycles studied using transabdominal ultrasound (Hackeloer et al., 1979). Furthermore, oestradiol estimations did not relate to specific FD. Others set arbitrary oestradiol criteria for human chorionic gonadotrophin (HCG) administration in otherwise unstimulated IVF cycles and also administered HCG if follicles showed disproportionate oestradiol:FD ratios (Paulson et al., 1990). They proposed that higher oestradiol concentrations were required at small follicle diameters (1100 pmol/l at 15 mm) than at larger diameters (735 pmol/l at 20 mm) to justify giving HCG, prior to oocyte recovery but there are no scientific data to support this hypothesis.
In our study, also involving unstimulated IVF, the cycle was not perturbed by any hormonal treatment and oocyte collection was timed by the onset of the endogenous LH surge. The onset of the surge was precisely estimated by 4 hourly measurements. Only monofollicular cycles producing a fertilizable oocyte were used for our purpose of deriving optimal reference criteria for individual follicles. The present study is based on a large number of observations using transvaginal ultrasound scanning, the method of choice for ovarian imaging in most current gynaecological practice (Kossoff et al., 1991). The study was continued in each cycle until the day of onset of the LH surge, as after this oestradiol concentrations fall with the onset of luteinization (Muasher et al., 1990); also, any practical predictive value would have passed. Transvaginal scanning did not begin until day 9 of the cycle, primarily to ensure the presence of a dominant follicle at the onset of monitoring (van Santbrink et al., 1995; Fauser and Van Heusden, 1997).
In cycles in which an oocyte could not be collected or fertilized, the peak FD was not reduced. By contrast, peak serum oestradiol concentrations were significantly reduced, suggesting an association between impaired follicular secretion of oestradiol and defective oocyte maturation. The findings suggest that serum oestradiol may be a better index of follicular functional capacity. Serum oestradiol concentrations rose exponentially with follicular enlargement suggesting progressively enhanced activity of individual granulosa cells.
Other reported attempts to predict serum oestradiol concentrations have not been helpful. Based on 21 conception cycles achieved by ovulation induction (Fink et al., 1982), follicular maturity did not correspond with oestrogen production in one-third of cases, even though a wide range of peak urinary oestrogen output (100–600 µmol/24 h) was accepted as being normal. Ultrasound assessment of endometrial thickness or morphology has been advocated as an indirect index of oestrogen state for monitoring exogenous gonadotrophin-stimulated cycles (Zaidi et al., 1995; Remohi et al., 1997), but has not proved valuable. It is probably too insensitive for the high oestradiol concentrations expected in gonadotrophin-stimulated cycles.
In this study, we determined means and lower limits of serum oestradiol for follicles of every diameter from 10 to 24 mm and specifically on the day of the LH surge (see Tables I and II
). At the onset of the LH surge, the mean FD was 21 mm and lower limit 14.5 mm, and geometric mean oestradiol 1279 pmol/l and lower limit 594 pmol/l. Those oestradiol values are consistent with the findings in Table II for all follicles of 21 mm diameter, but the lower oestradiol limit was also consistent with the mean for follicles of 15–16 mm diameter. Thus there is wide functional variation between follicles of similar size. Nevertheless, our findings provide reliable minimum and mean values of oestradiol to be expected for a follicle of any given diameter that can produce a competent oocyte, and these criteria might reasonably be applied to follicles of mixed sizes within a stimulated cohort as in IVF or ovulation induction treatment. Whether ovulation induction or down-regulation have significant effects on these relationships has yet to be tested in future studies. From these data, we know that for any FD, a particular serum oestradiol concentration can be said to be within expected limits for a competent follicle, that is, one which contains a fertilizable oocyte. However, these data cannot predict whether a follicle of a particular size is likely to contain a competent fertilizable oocyte, and can only give an indication of what might be expected from a particular follicle.
Our data from apparently competent pre-ovulatory follicles show considerable variance for FD and oestradiol, FD more so than oestradiol. It appears from these data that precise assessment of follicle maturity can only be obtained through considering both size and endocrine function. However, reference to oestradiol concentrations as an index of demonstrable follicle maturity (Table I) shows that an FD of 21 mm on average or 15 mm minimum is required. The clinical value of these data is to provide definition of what normal endocrine output is from a follicle which provides an oocyte capable of fertilization. Adjustment will need to be made for inter-laboratory differences but the assay systems used are widely available commercially, together with correction factors from assay to assay.
|
Acknowledgments |
|---|
D.J.C. was supported by a grant from the Medical Research Council, London (grant number 9107952).
|
Notes |
|---|
3 To whom correspondence should be addressed at: University of Bristol Division of Obstetrics and Gynaecology, St Michael's Hospital, Southwell Street, Bristol BS2 BEG, UK. E-mail: d.j.cahill{at}bris.ac.uk
* After a short illness, Michael Hull, Professor of Reproductive Medicine and Surgery in the Division of Obstetrics and Gynaecology, University of Bristol, died on 22 November, 1999, aged 60 years. He went to Bristol from London in 1976 as Consultant Senior Lecturer in Obstetrics and Gynaecology with a developing reputation in clinical reproductive endocrinology from his time in London and elsewhere. His contributions to knowledge and practice of reproductive medicine, particularly in all aspects of male and female infertility, brought international renown to his unit. He was appointed to a Personal Chair in 1989. He brought strong ethical principles to his research and clinical practice and campaigned locally and nationally for public understanding and funding of infertility services.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Fauser, B.C. and Van Heusden A.M. (1997) Manipulation of human ovarian function: physiological concepts and clinical consequences. Endocr. Rev., 18, 71–106.
Fink, R.S., Bowes, L.P., Mackintosh, C.E. et al. (1982) The value of ultrasound for monitoring ovarian responses to gonadotrophin stimulant therapy. Br. J. Obstet. Gynaecol., 89, 856–861.[Web of Science][Medline]
Hackeloer, B.J., Fleming, R., Robinson, H.P. et al. (1979) Correlation of ultrasonic and endocrinologic assessment of human follicular development. Am. J. Obstet. Gynecol., 135, 122–128.[Web of Science][Medline]
Harlow, C.R., Cahill, D.J., Maile, L.A. et al. (1995) Time-dependent effects of transforming growth factor alpha on aromatase activity in human granulosa cells. Hum. Reprod., 10, 2554–2559.
Hull, M.G.R., Joyce, D.N., McLeod, F.N. et al. (1985) An economic and ethical way to introduce in-vitro fertilization to infertility practice, and findings related to post-coital sperm/mucus penetration in isolated tubal, `cervical' and unexplained infertility. Ann. N. Y. Acad. Sci., 442, 318–323.[Web of Science][Medline]
Kossoff, G., Griffiths, K.A. and Dixon, C.E. (1991) Is the quality of transvaginal images superior to transabdominal ones under matched conditions? Ultrasound Obstet. Gynaecol., 1, 29–35.[Web of Science][Medline]
Muasher, S.J., Kruithoff, C., Webster, S. et al. (1990) Natural cycle in-vitro fertilization: a simplified treatment method. Fertil. Steril., 53, 826–827.
Paulson, R.J., Sauer, M.V., Francis, M.M. et al. (1990) In vitro fertilization in unstimulated cycles: a clinical trial using hCG for timing of follicle aspiration. Obstet. Gynecol., 52, 785–791.
Remohi, J., Ardiles, G., Garcia-Velasco, J.A. et al. (1997) Endometrial thickness and serum oestradiol concentrations as predictors of outcome in oocyte donation. Hum. Reprod., 12, 2271–2276.
Testart, J., Frydman, R., Feinstein, M.C. et al. (1981) Interpretation of plasma luteinizing hormone assay for the collection of mature oocytes from women: definition of a luteinizing hormone surge initiating rise. Fertil. Steril., 36, 50–54.[Web of Science][Medline]
van Santbrink, E.J., Hop, W.C., van Dessel, T.J. et al. (1995) Decremental follicle-stimulating hormone and dominant follicle development during the normal menstrual cycle. Fertil. Steril., 64, 37–43.[Web of Science][Medline]
Zaidi, J., Campbell, S., Pittrof, R. and Tan, S.L. (1995) Endometrial thickness, morphology, vascular penetration and velocimetry in predicting implantation in an in vitro fertilization program. Ultrasound Obstet. Gynecol., 6, 191–198.[Web of Science][Medline]
Submitted on February 24, 2000; accepted on May 22, 2000.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Ziomkiewicz, P.T. Ellison, S.F. Lipson, I. Thune, and G. Jasienska Body fat, energy balance and estradiol levels: a study based on hormonal profiles from complete menstrual cycles Hum. Reprod., November 1, 2008; 23(11): 2555 - 2563. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. A. Azim, M. Costantini-Ferrando, and K. Oktay Safety of Fertility Preservation by Ovarian Stimulation With Letrozole and Gonadotropins in Patients With Breast Cancer: A Prospective Controlled Study J. Clin. Oncol., June 1, 2008; 26(16): 2630 - 2635. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. G. M. Beckers, N. S. Macklon, M. J. Eijkemans, M. Ludwig, R. E. Felberbaum, K. Diedrich, S. Bustion, E. Loumaye, and B. C. J. M. Fauser Nonsupplemented Luteal Phase Characteristics after the Administration of Recombinant Human Chorionic Gonadotropin, Recombinant Luteinizing Hormone, or Gonadotropin-Releasing Hormone (GnRH) Agonist to Induce Final Oocyte Maturation in in Vitro Fertilization Patients after Ovarian Stimulation with Recombinant Follicle-Stimulating Hormone and GnRH Antagonist Cotreatment J. Clin. Endocrinol. Metab., September 1, 2003; 88(9): 4186 - 4192. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||