Influence of superovulation on endometrial and embryonic development

FERTILITY AND STERILITY

Vol. 53, No.5, May 1990

Copyright" 1990 The American Fertility Society

Printed on acid·free paper in U.S.A.

Influence of superovulation on endometrial and embryonic development

Vinay Sharma, M.R.C.O.G.*t Malcolm Whitehead, F.R.C.O.G.t:!:§ Bridgett Mason, M.B.B.S. * John Pryse-Davies, F.R.C.Path.ll

Timothy Ryder, Ph.D.1f Mitch Dowsett, Ph.D.# Stuart Campbell, F.R.C.O.G.t William Collins, D.Sc.t

King's College School of Medicine and Dentistry, The Bourn Hallam Medical Centre, .Hammersmith Hospital, Queen Charlotte's and Chelsea Hospital, and Royal Marsdon Hospita~ London, United Kingdom

The authors have studied the temporal relationship between follicular rupture and endometrial development in 13 women during a natural ovarian cycle (length 25 to 35 days), and subsequently after standard treatment with clomiphene citrate, human menopausal gonadotropin and human chorionic gonadotropin (hCG) to induce multiple folliculogenesis for oocyte recovery, in vitro fertilization, and embryo freezing (cycle length 23 to 27 days). An endometrial biopsy was taken during both cycles 1.5 to 2.0 days after the oocytes had been released or removed. The samples were examined by light and transmission electron microscopy. Samples of peripheral blood were taken at defined times for hormone analysis. After treatment 11 subjects (85%) had advanced morphological development of the endometrium (8 women by 3 to 4 days, 3 women by 1 to 2 days). The concentrations of plasma estradiol (E2 ) and progesterone (P) on the days of follicular rupture and endometrial biopsy were significantly raised in the treatment cycles. The concentration of total urinary estrogens on the day of hCG administration and the mean change in the concentration of plasma E 2 (treatment/control) on the days of endometrial biopsy were positively correlated with the extent of endometrial advancement. In addition, the mean change in the concentration of plasma P (treatment/control) was markedly increased on the days of follicular rupture and endometrial biopsy in those subjects with an advanced endometrium. Embryonic development was not so obviously related to the extent of superovulation. Asynchronous endometrial and embryonic development may therefore contribute to the low pregnancy rate in these patients. Fertil Steril53:822, 1990

The study of endometrial biopsies from healthy, potentially fertile women has shown that cellular events in the glandular epithelium are regulated Received July 24, 1989; revised and accepted January 4, 1990.

* The Bourn Hallam Medical Centre.

t Department of Obstetrics and Gynaecology, King's College School of Medicine and Dentistry. t Reprint requests: Malcolm Whitehead, F.R.C.O.G., Academic Department of Obstetrics and Gynaecology, King's College Hospital, Denmark Hill, London, SE5 8RX, United Kingdom. § Supported by the Imperial Cancer Research Fund. II Department of Histopathology, Hammersmith Hospital, London, United Kingdom. 822

very precisely between the time of the luteinizing hormone (LH) peak and the implantation of a fertilized ovum. 1•2 There is, however, a possibility that treatment regimens to induce multiple folliculogenesis in subfertile patients may adversely affect the rate of endometrial development with respect to the time of ovulation or oocyte recovery for in vitro fertilization (IVF). In particular, the results

11 Electron Microscopy Unit, Queen Charlotte's and Chelsea Hospital, London, United Kingdom. #Department of Biochemical Endocrinology, Royal Marsdon Hospital, London, United Kingdom.

Sharma et al. Superovulation and endometrial advancement

Fertility and Sterility

of uncontrolled studies have shown that clomiphene citrate (CC), 3 and human menopausal gonadotropin (hMG) and human chorionic gonadotropin (hCG), 4 may increase the maturation rate of the endometrial epithelium during the proliferative and secretory phases, respectively. We have examined this phenomenon in more detail by undertaking a prospective study of endometrial development in young women during a natural ovarian cycle and after standard treatment with CC, hMG, and hCG before oocyte recovery and IVF. An endometrial biopsy was taken 2 days after follicular rupture during the natural cycle and again 1.5 to 2.0 days after oocyte recovery during a treatment cycle. Apparently healthy embryos from each subject were cultured and frozen for replacement during subsequent cycles. 5 We have determined the extent to which endometrial and embryonic development are related to the concentrations of plasma estradiol (E 2 ) and urinary total estrogens on the day when hCG was administered, and with the concentrations of plasma E 2 and progesterone (P) on the days when the follicles ruptured, and when the biopsies were taken. MATERIALS AND METHODS

Thirteen patients (aged 26 to 33 years) attending the Bourn Hallam Medical Centre agreed to participater after the purpose of the investigation had been explained. The study was approved by the Ethics Committee of King's College Hospital. All volunteers had blocked oviducts, but apparently normal ovarian and uterine morphology as assessed by pelvic ultrasonography, and function as determined from their medical histories. All male partners had at least two normal spermiograms (i.e., sperm count> 20 X 106 mL seminal fluid, motility > 60%, abnormal morphology < 40%, and a negative mixed antibody reaction test. All patients were monitored for two cycles (natural and treatment). The cycles were consecutive in 11 women. Two subjects had either one or two intervening menstrual cycles. During the natural cycle both ovaries were examined daily by ultrasonography starting on the 8th day after the 1st day of menses (day 1). The mean maximum diameter of the leading follicle was recorded The day of follicular rupture (i.e., the time of presumptive ovulation) was determined from the time one or more of the following events occurred: (1) the follicle apparently disappeared, (2) the follicular wall collapsed, and (3) fluid was seen in the pouch

Vol. 53, No.5, May 1990

of Douglas. An endometrial biopsy was taken 1.5 to 2.0 days later (day of endometrial biopsy) from the anterior wall of the uterus using a Vabra curette (Cory Brothers, London, UK) and a Rocket aspiration pump (Rocket of London, Watford, Herts, UK). Portions of the biopsy were preserved in formol acetic (10% formalin and 5% acetic acid) for histological processing, and in 3% cacodylate buffered glutaraldehyde for transmission electron microscopy (TEM). Samples of peripheral venous blood (10 mL) were taken on alternate days during the follicular phase of the cycle, and on the days when the follicles ruptured and the endometrial biopsies were taken. An initial ultrasound scan was performed on the 2nd day to exclude the presence of an ovarian abnormality. During the treatment cycle all patients were given CC (100 mg/d, orally) from the 2nd to the 6th day after day 1. In addition, 150 to 225 IU of hMG (Serono Laboratories, Welwyn Garden City, Herts, United Kjngdom) were administered daily, intramuscularly, from the 4th day of the menstrual cycle. The subjects were scanned from day 8 to monitor follicular development. All follicles were observed in the transverse and longitudinal planes and the means of the horizontal and vertical diameters were recorded. The concentration of total urinary estrogens per 24 hours was also measured. The criteria for the administration of hCG (5,000 IU Profasi; Serono Laboratories) were as follows: (1) there were at least three follicles with a mean maximum diameter> 14 mm, (2) the leading follicle had a mean maximum diameter > 17 mm, and (3) the excretion oftotal urinary estrogen was >350 nmol/24 h. The oocytes were recovered under ultrasound guidance 35 hours later. The oocytes were inseminated, and the extent and rate of embryo development recorded. 6 Endometrial biopsies were taken 1.5 to 2.0 days later. Some of the cleaved embryos were cryopreserved on the following day, and others were allowed to develop into blastocysts. The blastocysts were also frozen. All embryos were replaced in a subsequent natural cycle. Blood samples were taken on alternate days during the follicular phase, on the day of hCG administration, and on the days of follicular rupture and endometrial biopsy. All blood samples were centrifuged immediately and the plasma stored at -20oC for a retrospective analysis. Light Microscopy

Histological features that have been compiled to describe an ideal 28-day cycle, in which ovulation

Sharma et al.

Superovulation and endometrial advancement

823

(or in our study follicular rupture) is assumed to occur on day 14, were used to date the endometrial biopsies. Daily changes in a number of criteria reflecting cellular growth, proliferation, and differentiation, together with glandular function, have been described in detail. 7 Briefly, the first distinct histological change occurs about 2 days after follicular rupture, when subnuclear vacuoles become apparent in the glandular epithelium and displace the nuclei upwards to create a pattern of pseudostratification. By day 3 there is an orderly row of nuclei with homogeneous cytoplasm above and vacuoles below. By day 4 small vacuoles have passed the nuclei and approached the lumen of the gland. Around day 5 the vacuoles are discharging their contents into the lumen and the nuclei are located at the base of the cells. The peak of acidophylic secretion in the lumen occurs by day 6. A best estimate for the time of endometrial sampling relative to follicular rupture (the presumed time of ovulation) was reported for each case by one investigator (J.P.-D.). Although this form of dating is a subjective process, the precision of the estimate is greatest during the early luteal phase of the ovarian cycle. 8 Transmission Electron Microscopy

The endometrial biopsies were postfixed in 1% osmium tetroxide in cacodylate buffer, dehydrated through a graded series of alcohols, and embedded in araldite. Ultrathin sections were cut on glass knives, stained with ceranyl acetate and lead citrate, and examined at 50 k V in an Hitachi HU 12A transmission electron microscope (Hitachi Scientific Instruments, Reading, United Kingdom). Endometrial dating was based on the most advanced fine structural features in the specimen. Detailed changes in the accumulation of glycogen, the development of giant mitochondria, and the appearance of a nucleolar channel system have been described and illustrated. 7 Deposits of glycogen and dense, supranuclear Golgi complexes are observed about 2 days after follicular rupture. The characteristic nucleolar channel system becomes progressively more evident over days 3 to 6, and the mitochondria continue to enlarge and elongate. Around days 5 to 6 large projections appear from the surface of epithelial cells and extensive amounts of secretion appear in the lumen. The number of nuclear cross-sections containing the unique nucleolar channel system increases 824

Sharma et al.

from 1 to 2 per 1,000 nuclei on day 3 to >40 per 1,000 nuclei on day 5. All samples were assessed and dated by one investigator (T.R.). Hormone Analysis

The concentration of plasma E 2 was measured by an extraction radioimmunoassay (RIA) that involved the use of a tritiated antigen and polyclonal antibodies to E 2 -6-carboxymethyl oxime-bovine serum albumen (BSA). The sensitivity was 60 pmoljL, and the mean within and between assay coefficients of variation (CVs) were 7.8% and 11.8%, respectively. The concentration of plasma P was measured by a nonextraction RIA that involved the use of an iodinated tracer and polyclonal antibodies to P-3-carboxymethyl oxime-BSA. The sensitivity was 2 nmol/L, and the mean within and between assay CVs were 6.7% and 12.3%, respectively. Total urinary estrogens were measured after acid hydrolysis by a traditional fluorometric assay. Statistical Analyses

Nonparametric methods were used. The effect of treatment on the rate of endometrial development relative to the time of follicular rupture was assessed by the application of Wilcoxon's matchedpairs signed-ranks test. The relationship between defined variables was assessed by the use of the Spearman rank correlation coefficient, or the Kruskal-Wallis one-way analysis of variance, x2 value.

RESULTS

Some characteristics of the natural and treatment cycles for all subjects are shown in Table 1. The length of the luteal phase (day of follicular rupture to day 1 of next menses - 1) for the natural cycles varied from 12 to 16 days, median 13 days. The corresponding values for the treatment cycles were 10 to 15 days, median 12 days. The concentration of total urinary estrogens on the day of hCG administration ranged from 400 to 2,815 nmolf 24h. Cell Morphology

The most important variables between the natural and treatment cycles were the amount of basal glycogen and the proportion of nuclei containing the nucleolar channel system. The analysis of an

Superovulation and endometrial advancement

Fertility and Sterility

Table 1 Characteristics of the Natural and Treatment Cycles for all Subjects Natural cycle

Subject no.

Day of follicular rupture

1 2 3 4 5 6 7 8 9 10 11 12 13

20 20 16 16 14 16 19 19 12 15 14 14 14

Median

16

28

Treatment cycle Day of follicular rupture

Length

Length

Urinary estrogens"

d

nmol/24h

31 35 27 28 26 29 31 31 25 26 26 26

618 1,460 967 2,602 >1,000 1,559 >1,000 2,815 2,630 400 502 >1,500 >1,000

14 13 13 14 13 13 13 13 14 14 12 13 12

23 25 27 25 26 27 23 24 25 24 23 23 23

>1,000

13

24

d

" Upper limit dependent on urine volume.

index based on a scoring system for all variables, excluding the nucleolar channel system, showed that there was no significant difference between the natural and treatment cycles. There was, however, a highly significant increase in the nucleolar channel system count during the treatment cycle (P < 0.001, Wilcoxon's test). Histological and Ultrastructural Dating

Table 2 Cytological Dating of Endometrial Biopsies Relative to Day of Follicular Rupture

The cytological dating of endometrial biopsies by light and electron microscopy relative to the day of follicular rupture for both natural and treatment cycles is shown in Table 2. Two subjects (10 and 11) had similar dates for both cycles (i.e., no endometrial advancement). Three subjects (1, 2, and 6) had advanced endometrial development of 1 or 2 days. Eight subjects had advanced endometrial development of 3 to 4 days by at least one method. The median increased rate of development was 3 days by light microscopy and 2 days by transmission electron microscopy. The advancement observed by both methods was significant (P < 0.001, Wilcoxon's test). Hormone Analyses

The concentrations of plasma E 2 and P on the day of follicular rupture and the day of endometrial biopsy are summarized in Table 3. The median concentration of plasma E 2 in the treatment cycle Vol. 53, No.5, May 1990

on the day offollicular rupture was 9.2 times higher than the corresponding day during the natural cycle. Similarly, the level was 6.3 times higher on the day of endometrial biopsy in treatment cycles as compared with the natural cycles. The median concentration of P was also higher in treatment cycles when compared with the natural cycles. The value was 2.6 times and 6.2 times higher on the days of follicular rupture and endometrial biopsy, respectively. In addition, whereas the ratio of plasma P on the days of endometrial biopsy/follicular rupture was 3.4 in natural cycles, it was 8.1 in treatment cycles. The factorial increases in the concentration of plasma E 2 (treatment/natural cycle for all subjects) were positively correlated with the extent of endometrial advancement on the day of endometrial biopsy (P < 0.05, Spearman's test), but not on the day of follicular rupture. The corresponding increases in the concentration of plasma P. on the days of follicular rupture and endometrial biopsy were markedly higher in the 11 subjects who had advanced endometrial development. The increases, however, were not related to the extent of the morphological changes and hence to the cytological dating. The concentrations of total urinary estrogens and plasma E 2 on the day of hCG administration were positively correlated with advanced endometrial development (P < 0.001 for both variables, Kruskal-Wallis test).

Cytological dating

Natural cycle Subject no.

LM"

1 2 3 4 5 6 7 8 9 10 11 12 13

2 to 3 1 to 2 1 2 to 3 0 2 to 3 0 to 1 Oto 1 Oto 1 Oto 1 3 to4 0 1 to 2

Treatment cycle

TEMb

LM

2 to 3 1 to 2 2

1 to 2

3 to 4 2 to 3 4to5 5 to 6 4 to 5 4 to 5 4 to 5 4to5 5 Oto 1 2 to 3 4to5 4to 5

3 to 7 5 5 3 5 1 to 2 2 to 3 3 5

1.5

4.5

3.5

d

Median

1.0

Endometrial advancement

TEM

LM

3to 4 2 to 3 3 to 4

1 1 3 3 4 2 4 4 4 0 -1 4 3

d

d

0 3 1 to 2 1 0 to 1 1 to 2 2to 3

TEM

3

1 1 2 3 2 4 2 4 0 0 3 2

" Light microscopy. b Transmission electron microscopy.

Sharma et al.

Superovulation and endometrial advancement

825

Table 3 Concentration of Plasma E 2 and P on the Day of Follicular Rupture• and the Day of Endometrial Biopsy during Natural and Treatment Cycles Natural cycle Plasma hormone

Treatment cycle

Day follicular rupture

Day endometrial biopsy

Day follicular rupture

Day endometrial biopsy

360 140to 1,010

400 230to 1,800

3,300 2,100 to 5,300

2,500 1,100 to 5,600

6.1 3.1 to 13.4

21.0 7.2to52.3

16.0 2.9to 42.1

130.0 12.6 to 370.2

E 2 (pmol/L) Median Range P (nmol/L) Median Range

• Natural cycle; day of oocyte collection treatment cycle.

Oogenesis and Embryogenesis

that developed into blastocysts from subjects grouped according to the extent of endometrial advancement is shown in Table 5. There was no apparent relationship between the two processes.

The number of oocytes recovered and fertilized from each subject are shown in Table 4 together with the number of embryos that underwent cleavage, the number that were subsequently frozen, and the number that were cultured an!;! formed blastocysts. Oocytes were not recovered from subject 10, and only 2 were obtained from subject 11 (i.e., from the subjects with no endometrial advancement). From 5 to 16 oocytes were recovered from each of the other subjects. The majority of oocytes were fertilized and underwent cleavage. There was no significant correlation between the concentration of total urinary estrogens and the number of oocytes recovered, the number that were viable, or the number that were fertilized. Whenever possible a maximum of four embryos were frozen, and the remainder were allowed to develop in culture media. The number of embryos

DISCUSSION

This is the first report of a prospective, controlled study to examine the rate of change in endometrial morphology relative to the time of follicular rupture in patients being treated with a combination of agents to induce multiple follicular development and oogenesis. The results clearly showed endometrial advancement of 1 to 4 days in 85% of subjects. The limitations associated with attempts to date the endometrium by histological and ultrastructural methods are well known, so it was particularly reassuring that a similar advancement was observed by two independent assessors

Characteristics of Oocytes Recovered From Treatment Cycles and Subsequent Embryogenesis

Table4

Embryos

Oocytes Subject no.

Blastocysts•

Recovered

Fertilized

Cleaved

Frozen/cultured

No.

Frozen

9 3 9 14 7 14 1 7 10 0 2 8 4

9 2 7 12 7 14 1 7 9 0 2 8 1

4/5 1/1 4/3 4/8 4/3 4/10 1/0 4/3 4/5

4 0 0 3 0 3

4

12 13

9 5 9 15 7 16 8 10 1:,;l 0 2 9 8

Median

9

7

7

1 2 3 4 5 6 7 8 9 10 11

2/0 4/4 0/0

3 3

0 4 4 0

4

• From cultured embryos.

826

Sharma et al.

Superovulation and endometrial advancement

Fertility and Sterility

Table5 Relationship Between Endometrial Dating and the Number ofBlastocysts Patients cultured embryos

No. with blastocysts

No. of embryos

Blastocysts•

+1 to2 >+3

3 6

2 3

16 26

7 (44) 11 (42)

Totals

9

5

42

18-

Endometrial dating d

• ValU!lS in parentheses are percents.

(one using light microscopy, and the other electron microscopy) working in different laboratories. The advancement was associated with higher concentrations of plasma E 2 and P on the day offollicular rupture and 1.5 to 2.0 days later at the time when the endometrial biopsy was taken. The extent of the advancement was also correlated with the concentrations of urinary total estrogens and plasma E 2 and P on the day of follicular rupture, and 1.5 to 2.0 days later at the time when the endometrial biopsy was taken. The extent of the advancement is correlated with the concentrations of urinary total estrogens and plasma E 2 on the day that hCG was administered, and with the increase in the plasma E 2 (treatment/natural cycle) on the day that the biopsy was taken. Those subjects, whose endometrium responded most to treatment tended to have higher concentrations of plasma E 2 during the natural cycle (data not shown). Our study was concerned solely with morphological changes in the glandular epithelium of the endometrium. Other research during the last decade has revealed that the endometrium secretes a cyclical profile of regulator peptides. In particular, it is now well established that the endometrium secretes prolactin (PRL), and that this function is not dependent upon the presence of a trophoblast. 9 It has been suggested that the principal function of decidual PRL in pregnancy involves osmoregulation, 10 although other activities may include embryonic growth regulation, immunoregulation, and fetal lung maturation. The production of endometrial prolactin starts before implantation although it is not certain whether the.. hormone is secreted into the uterine lumen or blood vessels, or both. It is therefore possible that the hormone may be able to influence the growth and development of the pre-embryos even in the preimplantation period. Although >60% of the secretory products of the decidualized endometrium comprises a large variety

Vol. 53, No.5, May 1990

of polypeptides, 11 their function is still unknown in terms of implantation and early embryonic development. Nevertheless, it can be expected that drug effects on endometrial maturation will also alter its function. Although our results showed that there were no apparent differences in the number and overall progression of embryos from those subjects whose endometrium was advanced 3 to 4 days, we cannot exclude the possibility that important time differences exist in their morphological and functional development. Furthermore, the extent to which human embryonic and endometrial development should be in synchrony is unknown. In the rhesus monkey, however, early pregnancy loss has been shown to· be higher when the embryos were in asynchrony with the endometrium. 12 The role of the human endometrium is difficult to assess, because embryos can implant at extragenital sites after in vivo as well a~ IVF. Moreover, a delayed rise in the concentration of hCG, implying delayed implantation in what is probably an asynchronous endometrium, has also been reported in humans. 13 Early results from ovum donation programs have suggested that the implantation window may extend up to the 19th or 20th day of the menstrual cycle although its duration remains uncertain. 14 A more recent study of different treatment regimens has shown that a very short exposure of the endometrium to physiological levels of E 2 or conversely a more prolonged stimulation at the same dose will allow normal endometrial development with the addition of P. Supraphysiological doses of P significantly enhanced the endometrial maturation process, 15 which is consistent with our findings. There has also been a suggestion that although the time available for implantation is short it may appear at different times of the luteal phase in different women. 16 This phenomenon may explain instances of delayed implantation of the human embryo. The results from various clinical studies have shown that overstimulation with estrogens can result in rapid and extensive vacuolization of the luminal epithelial cells followed by their complete exfoliation.17 Furthermore, patients with high levels of plasma E 2 may also exhibit a spontaneous surge in the level of circulatory LH when the follicles are small, or may even start spontaneous luteinization of the granulosa cells. 18 In addition, in the present study the antiestrogenic effect of CC may mimic the preovulatory decrease in plasma E 2 and the

Sharma et al.

Superovulation and endometrial advancement

827

preprogrammed epithelial cells may start the secretory transformations earlier. These changes in the endometrium may influence embryo apposition and initiate a reduction in the negative charge on the cells of the blastocyst and the endometrium, which allows surface contact similar to that observed in the rat. 19 The blastocysts and the endometrium also cooperate in the lysis of the zona pellucida. Human blastocysts orientate themselves so that the part neighboring the inner cell mass comes in contact with the epithelial surface first. 20 This close contact may allow the epithelium to influence the metabolism of the embryonic plate. Superovulation beyond the optimum amount has been shown to result in high abortion rates in vivo 21 and low pregnancy rates after IVF and embryo transfer (ET). 6 The proportion of multiple pregnancies is reduced when more than nine oocytes are retrieved, or more than seven cleaved embryos are available, even though only four are transferred to the uterus. In nature a considerable number of preimplantation embryos are lost. 22 Similarly, after ET up to 53% of patients may have viable embryos although less than half of these result in a clinical pregnancy. 23 These results are remarkably similar to the high early pregnancy loss rate reported in the rhesus monkey when the embryo and the endometrium are not in synchrony. 12 The superovulation regimens have been shown to significantly and sharply reduce the proportion of follicles with normal steroid concentrations, the number of patients with a normal follicle, and the number of follicles that can be considered normal. 24 Even though the oocyte may be able to adapt to an atypical hormonal environment in the final stages of its maturation, its viability and the future development of the embryo may be compromised in extreme situations. Our results show that endometrial advancement frequently occurs after superovulation and may in part be responsible for low conception rates, although a cause and effect relationship has not been demonstrated. Nevertheless, due attention to this factor must be given while preparing patients for treatment by IVF and ET. For example, agonists of gonadotropin-releasing hormone may be used to control endogenous ovarian function and so improve uterine receptivity during treatment cycles for assisted conception. 25 Alternatively, tests should be developed to identify the most appropriate treatment regimen for a particular patient, and for monitoring the response to different drug dos-

828

Sharma et al.

ages. Finally, the replacement of frozen and thawed embryos during a natural or controlled cycle 15 might be the preferred methods oftreatment in the future.

Acknowledgments. The authors are grateful to Nigel Shannon, Ph.D., for undertaking the statistical analysis, and Margaret Mobberley, F.I.M.L.S. for assistance with the transmission electron microscopy.

REFERENCES 1. Dockery P, Li TC, Rogers A W, Cooke ID, Lenton EA, Warren MA: An examination of the variation in timed endome. trial biopsies. Hum Reprod 3:715, 1988 2. Dockery P, Li TC, Rogers AW, Cooke ID, Lenton EA: The ultrastructure of the glandular epithelium in the timed endometrial biopsy. Hum Reprod 3:826, 1988 3. Birkenfeld A, Navot D, Levij IS, Laufer N, Beier-Hellwig K, Goecke C, Schenker JG, Beier HM: Advanced secretory changes in the proliferative human endometrial epithelium following clomiphene citrate treatment. Fertil Steril 45: 462, 1986 4. Garcia JE, Acosta AA, Hsiu JG, Jones HW: Advanced endometrial maturation after ovulation induction with human menopausal gonadotropin/human chorionic gonadotropin for in vitro fertilization. Fertil Steril41:31, 1984 5. Fehilly CB, Cohen JC, Simons RF, Fishel SB, Edwards RG: Cryopreservation of cleaving embryos and expanded blastocysts in the human: a comparative study. Fertil Steril 44: 638, 1985 6. Sharma V, Riddle A, Mason BA, Pampiglione J, Campbell S: An analysis of factors influencing the establishment of a clinical pregnancy in an ultrasound-based ambulatory in vitro fertilization program. Fertil Steril49:468, 1988 7. Wynn RM: The human endometrium: cyclic and gestational changes. In Biology of the Uterus, 2nd edition, Edited by RM V. ;nn and WP Jollie. New York, Plenum Press, 1989, p 289 8. Li T-C, Dockery P, Rogers AW, Coote ID: How precise is histologic dating of endometrium using the standard dating criteria? Fertil Steril51:759, 1989 9. Daly DC, Maslar IA, Rosenberg SM, Tohan N, Riddick DH: Prolactin production by luteal phase defect endometrium. Am J Obstet Gynecol140:587, 1981 10. Tyson JE: The evolutionary role of prolactin in mammalian osmoregulation: effects on fetoplacental hydromineral transport. Semin Perinatal 6:216, 1982 11. Bell SC: Comparative aspects of decidualization in rodents and human: cell types, secreted products and associated function. In Implantation of the Human Embryo, Edited by RG Edwards, JM Purdy, PC Steptoe. London, Academic Press, 1985, p 71 12. Marston JH, Penn R, Sirrelle PC: Successful auto transfer of tubal eggs in the rhesus monkey (Macaca mulatta). J Reprod Fertil49:175, 1977 13. Englert L Y, Roger M, Belaisch-Allart J, Jondet M, Frydman R, Testart J: Delayed appearance of plasmatic chori-

Superovulation and endometrial advancement

Fertility and Sterility

14. 15.

16.

17. 18.

19.

20.

onic gonadotropin in pregnancies after in vitro fertilization. Fertil Steril 42:835, 1984 Rosenwaks Z: Donor eggs: their application in modern reproductive technologies. Fertil Steril47:895, 1987 Navot D, Anderson TL, Droesch K, Scott RT, Kreiner D, Rosenwaks Z: Hormonal manipulation of endometrial maturation. J Clin Endocrinol Metab 68:801, 1989 Edwards RG: Normal and abnormal implantation in the human uterus. In Implantation of the Human Embryo, Edited by RG Edwards, JM Purdy, PC Steptoe. London, Academic Press, 1985, p 303 Blaustein A: Interpretation of Biopsy of Endometrium. New York, Raven Press, 1980, p 21 Trounson AO: Manipulation of endocrine requirements for in vitro fertilization. Proc Endocr Soc Australia 25(suppl. 1):1,1982 Hewitt K, Beer AE, Grinnel F: Disappearance of aniotic sites from the surface of the rat endometrial epithelium at the time of blastocyst implantation. Bioi Reprod 21:691, 1979 . Psychoyos A: Endocrine control of implantation. In Hand Book of Physiology: Endocrinology, Vol. 2. Edited by RO Greep. Baltimore, Williams and Wilkins, 1973, p 187

Vol. 53, No.5, May 1990

21. Schwartz M, Jewelwicz R, Dyrenfurth I, Tropper P, Van de Wiele RL: The use of human menopausal and chorionic gonadotrophins for induction of ovulation: sixteen years experience at the Sloane Hospital for Women. Am J Obstet Gynecol138:801, 1980 22. Steer C, Campbell S, Davies M, Mason B, Collins W: Spontaneous abortion rates following natural and assisted conception. Br Med J 299:1317, 1989 23. O'Neil C, Gidley Baird AA, Pike IL, Porter RN, Sinosich MJ, Saunders DM: Maternal blood platelet physiology and luteal phase endocrinology as means of monitoring pre and post implantation embryo viability following in vitro fertilization. J In Vitro Fert Embryo Transfer 2:87, 1985 24. Testart J, Frydman R, Castanier M, Lassalle B, Belaisch J: Influence of ovarian stimulation on follicular fluid steroid levels and fertilization of the oocyte. In Fertilization of the Human Egg In Vitro, Edited by HM Beier, HR Lindner. Berlin, Springer Verlag, 1983, p 73 25. Testart J, Forman R, Belaisch-Allert J, Volante M, Hazout A, Strubb N, Frydman R: Embryo quality and uterine receptivity in in vitro fertilization cycles with or without agonists of gonadotropin-releasing hormone. Hum Reprod 4: 198,1989

Sharma et al. Superovulation and endometrial advancement

829