1 Genesis Fertility Centre, 555 West 12th Avenue, Suite 550, Vancouver, British ColumbiaV5Z 3X7 and 2 Reproductive Biology Unit, Mount Sinai Hospital, Division of Reproductive Sciences, Department of Obstetrics and Gynecology, University of Toronto, 3rd Floor, 700 University Avenue, Toronto, Ontario M5G 1Z5, Canada
3 To whom correspondence should be addressed. Email: sfisher{at}genesis-fertility.com
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: coasting/estradiol/falling estradiol/IVF outcome
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
On the other hand, women at risk of ovarian hyperstimulation maintained excellent pregnancy rates in spite of experiencing significant drops in serum estradiol prior to hCG as a result of deliberate reductions or omissions of gonadotrophin doses. The occurrence of pregnancy as well as the development of ovarian hyperstimulation syndrome (OHSS) were reported to be unrelated to trends in estradiol concentrations in a group of women coasted prior to hCG (Egbase et al., 2000). The objective of this study was to further characterize the effect of falling estradiol levels prior to hCG on IVF outcomes.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Records of all IVF cycles from the period 19932002 were reviewed to identify women whose E2 levels fell prior to the day of hCG administration. For patients undergoing more than one treatment cycle during the above time-period, only data from the first treatment cycle in which estradiol fell were included in the analysis. All stimulation protocols during this period involved down-regulation with GnRH agonists in either a standard long or flare protocol. Appropriate controls whose E2 levels continued to rise until the day of hCG were selected from the same time-period and matched to cases for age and year of treatment.
Demographic information including age and diagnosis were collected for each patient. Stimulation parameters including protocol, total FSH dose, protocol modification (step-down or coasting), peak estradiol levels, and day of hCG were collected for both groups. Outcome data collected included number of oocytes retrieved, fertilization rate, number of 4872 h embryos, number of embryos transferred, implantation rate and clinical pregnancy rate. Given that the first pregnancy ultrasound performed in our programme is at 8 weeks gestational age, the implantation rate was defined as the number of gestational sacs identified on ultrasound at 8 weeks gestational age per number of embryos transferred, and clinical pregnancies were defined as those with fetal heart activity documented on ultrasound at 8 weeks gestational age.
Women whose E2 levels fell spontaneously and as a result of protocol modification (step-down or coasting) were analysed compared to controls to elicit the significance of this on IVF outcome.
Statistical analysis was carried out using SPSS (Version 11.0.1, SPSS Inc.). Normally distributed continuous variables were analysed using one-way analysis of variance (ANOVA), whereas non-normally distributed continuous variables were analysed using the KruskalWallis one-way ANOVA. Post hoc analysis was performed using Dunnett's t-test to compare the spontaneously falling E2 and protocol modification groups to the control group for statistical significance. Categorical variables were analysed using 2-test. P<0.05 was considered statistically significant.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Patients in all three groups were similar in age and choice of treatment protocol, whereas women in the protocol modification group were more likely to have a diagnosis of anovulation compared to women in the spontaneously falling E2 or control groups (Table I). The percentage decrease in E2 was significantly greater in the protocol modification group than in those whose E2 levels fell spontaneously [median 22.5% (range 195) versus 10% (range 195); P<0.0001, KruskalWallis test].
|
The number of oocytes retrieved was greatest in the protocol modification group (median 11, range 134, P=0.005 compared to controls, Dunnett's t-test) as might be expected given that a greater percentage of these women had polycystic ovarian syndrome. The spontaneously falling E2 group, however, had fewer oocytes retrieved than the control group (median 5, range 127 versus median 8, range 134; P=0.002, Dunnett's t-test). In addition, the spontaneously falling E2 group experienced higher rates of failed fertilization (P=0.024, 2-test) resulting in 24% of cycles with no embryo transfer in this group (P=0.07,
2-test) (Figure 1). The overall fertilization rate was not significantly different between the groups (control 62%, protocol modification 66%, spontaneously falling E2 60%; P=0.862, KruskalWallis test).
|
Similar mean (SD) numbers of embryos were transferred in each group [control 2.6 (0.8), protocol modification 2.6 (0.7) and spontaneously falling E2 2.4 (0.7), P=0.28, one-way ANOVA]. The vast majority of embryo transfers were performed at 4872 h, reflecting the standard practice during most of the time period of study. Two of 99 (2%) control cycles, three of 29 (10%) protocol modification cycles, and one of 60 (1.6%) spontaneously falling E2 cycles were performed at the blastocyst stage (P=0.052, 2-test).
Overall, there was no correlation between percentage decrease in estradiol level and clinical pregnancy rate with IVF. There were a total of 34 pregnancies in 112 cycles started (30%) in the control group (29 viable intrauterine pregnancies, three spontaneous abortions, one ectopic and one chemical pregnancy). In the protocol modification group, nine women conceived in 34 cycles started (26%), of which one pregnancy was a chemical pregnancy. Of a total of 13 pregnancies in the 78 cycles started in the falling E2 group (17%), there were two spontaneous abortions, one ectopic pregnancy and one chemical pregnancy. Spontaneously falling E2 was associated with a much lower clinical pregnancy rate (P=0.039, 2-test) and implantation rate (P=0.244, one-way ANOVA) than both the protocol modification and control groups (Figure 1). Of the clinical pregnancies identified, the multiple pregnancy rate was 31% (9/29) for the control group (eight twins, one triplet), 38% (3/8) for the protocol modification group (two twins, one triplet), and 33% (3/9) for the spontaneously falling E2 group (three twins, zero triplets) (P=0.768,
2-test). There was no statistically significant difference between the groups with respect to pregnancy outcome (P=0.703,
2-test) or multiple pregnancy rate (P=0.768,
2-test).
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In this study, there was no correlation between the percentage decrease in E2 level and IVF outcome. The aetiology of the falling E2 was a more important predictor of outcome than the magnitude of the drop in E2. Of the 112 patients who exhibited falling E2 levels in this study, 70% fell spontaneously, whereas the remainder fell as a result of protocol modification. Although there was no difference between the groups with respect to other important prognostic indicators such as age and choice of protocol (flare versus long protocol), women in the spontaneously falling E2 exhibited many features of poor responders. Specifically, this group had lower peak E2 levels, lower E2 levels on the day of hCG and fewer oocytes retrieved despite receiving significantly higher total doses of FSH. This observation underscores the importance of continued estradiol measurement as an adjunct to ultrasound monitoring in IVF as the pattern of E2 rise prior to hCG may provide useful prognostic information for the current and potentially future cycles.
It would seem that spontaneously falling E2 is a reflection of the underlying health of the cohort of follicles recruited. In early studies prior to the advent of GnRH analogue use in down-regulated cycles, falling E2 prior to hCG was presumed to be related to follicular atresia and/or premature luteinization (Jones et al., 1983; Ben-Rafael et al., 1986
). This resulted in retrieval of greater numbers of atretic oocytes, low fertilization and cleavage rates, high rates of polyspermy and ultimately low pregnancy rates. In such cycles, it is not uncommon to witness a minor LH surge prior to follicular maturity which is sufficient to cause premature luteinization, and falling estradiol levels as androstenedione production is reduced in favour of progesterone synthesis, thus limiting the substrate for aromatization to estradiol (Erickson et al., 1985
).
It is clear from both animal and human studies that the fate of the developing follicle is closely related to its ability to make estradiol. In vitro studies have shown that estradiol has important autocrine and paracrine roles within the developing follicle. In granulosa cells, estradiol enhances FSH-stimulated cell division, induces aromatase activity, and stimulates the expression of E2, LH and FSH receptors as well as the production of inhibin (Goldenberg et al., 1973; Dorrington et al., 1975
; Louvet and Vaitukaitis, 1976
; Erickson et al., 1979
; Adashi and Hsueh, 1982
; Kessel et al., 1985
; Hillier et al., 1989
). Androgen production is inhibited by estradiol in surrounding theca cells (Magoffin and Erickson, 1982
). Follicular fluid estradiol production has been shown to directly reflect aromatase activity in vitro (Hillier et al., 1981
), which in turn correlates with granulosa cell number and follicular maturity (size) (McNatty et al., 1979
). It is possible that women with spontaneously falling E2 exhibit this pattern either due to a reduction in the absolute number of granulosa cells, or to a relative reduction in aromatase activity within the developing follicle. Both of these may be the result of a relative insensitivity to FSH which is characteristic of poor responder patients.
Regardless of the aetiology of the low estradiol levels, the effect of low estradiol levels, and particularly the E2:androgen ratio in follicular fluid, has been well documented (Andersen, 1993; Fukuda et al., 1995
; Akaboshi et al., 1998
). It has been previously proposed that gonadotrophins and estrogens suppress the apoptotic DNA fragmentation associated with follicular atresia, whereas androgens induce follicular atresia (Billig et al., 1993
). Administration of diethylstilboestrol (DES) and estradiol to hypophysectomized rats resulted in prevention of granulosa cell apoptosis, whereas removal of DES and treatment with androgen were associated with increased DNA fragmentation (Billig et al., 1993
). Similarly, lowering androstenedione levels by active immunization has been shown to reduce follicular atresia and increase ovulation rates in sheep (Scaramuzzi et al., 1980
).
Clinical studies evaluating follicular fluid steroidogenesis have similarly correlated low E2:androgen ratios with IVF outcome. Follicular fluid samples yielding an oocyte known to result in clinical pregnancy after IVF were analysed compared to follicular fluid from non-conception cycles (Andersen, 1993). Pregnancy potential of the oocytes correlated with a high E2:androgen ratio in follicular fluid, suggesting that low E2:androgen ratios reflect early follicular atresia and reduced viability of the associated oocyte. Similar results have been reported by others (Fukuda et al., 1995
; Akaboshi et al., 1998
).
This study was retrospective and data involving follicular fluid E2:androgen ratios and serum androgen levels were not available. If one infers from the basic science and clinical literature that small spontaneous drops in estradiol prior to hCG reflect early follicular atresia and compromised oocyte quality, it is not surprising that this group experience high rates of failed fertilization (18%), low implantation rates (9%) and low clinical pregnancy rates per cycle started (12%). What remains to be seen is whether this is a repetitive phenomenon in subsequent IVF cycles.
In contrast to the spontaneously falling E2 group, women whose E2 fell as a result of intentional protocol modification aimed at reducing the risk of ovarian hyperstimulation had IVF outcomes similar to control patients. This group was more likely to have a diagnosis of anovulation than either the control or spontaneously falling E2 group, reflecting the increased risk of hyperstimulation in patients with anovulation due to polycystic ovarian syndrome. As might be expected, this group required lower doses of FSH, and achieved greater numbers of oocytes than the spontaneously falling E2 or control groups.
Despite greater numbers of oocytes retrieved, women in the protocol modification group had similar numbers of healthy embryos on day 3 as the control group. IVF outcomes (rate of failed fertilization, fertilization rate, implantation rate, and clinical pregnancy rate) were comparable between the protocol modification and control groups.
Although it seems that falling E2 as a result of protocol modification does not have a detrimental effect on IVF outcome, this comparison is obviously limited by lack of an appropriate control group.
Given that PCOS patients notoriously respond well to gonadotrophin stimulation, one would expect that this group might have an improved prognosis with IVF compared to control patients, many of whom have unexplained infertility. A more appropriate control group might include PCOS patients not requiring protocol modification due to concern regarding hyperstimulation.
In summary, it is clear that the aetiology of falling estradiol rather than the magnitude of the decline in estradiol is critical to IVF outcome. Patients experiencing sharp declines in estradiol level as a result of deliberate protocol modification may be reassured that this phenomenon does not compromise IVF outcomes. On the other hand, patients experiencing subtle but spontaneous declines in estradiol level should be counselled that this is associated with very poor IVF outcomes, and cycle cancellation may be warranted.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Akaboshi K, Oda T, Yoshida J, Kohriyama S, Miyazaki T and Yoshimura Y (1998) The effects of gonadotropin-releasing hormone agonist on androstenedione production and follicular development during controlled ovarian hyperstimulation. J Assist Reprod Genet 15, 478484.[CrossRef][ISI][Medline]
Andersen CY (1993) Characteristics of human follicular fluid associated with successful conception after in vitro fertilization. J Clin Endocrinol Metab 77, 12271234.[Abstract]
Ben-Rafael Z, Kopf GS, Blasco L, Flickinger GL, Tureck RW, Strauss JF and Mastroianni L, Jr (1986) Follicular maturation parameters associated with the failure of oocyte retrieval, fertilization, and cleavage in vitro. Fertil Steril 45, 5157.[ISI][Medline]
Billig H, Furuta I and Hsueh AJ (1993) Estrogens inhibit and androgens enhance ovarian granulosa cell apoptosis. Endocrinology 133, 22042212.[Abstract]
Dorrington JH, Moon YS and Armstrong DT (1975) Estradiol-17beta biosynthesis in cultured granulosa cells from hypophysectomized immature rats; stimulation by follicle-stimulating hormone. Endocrinology 97, 13281331.[Abstract]
Egbase PE, Al Sharhan M, Berlingieri P and Grudzinskas JG (2000) Serum oestradiol and progesterone concentrations during prolonged coasting in 15 women at risk of ovarian hyperstimulation syndrome following ovarian stimulation for assisted reproduction treatment. Hum Reprod 15, 20822086.
Erickson GF, Wang C and Hsueh AJ (1979) FSH induction of functional LH receptors in granulosa cells cultured in a chemically defined medium. Nature 279, 336338.[ISI][Medline]
Erickson GF, Magoffin DA, Dyer CA and Hofeditz C (1985) The ovarian androgen producing cells: a review of structure/function relationships. Endocr Rev 6, 371399.[ISI][Medline]
Fukuda M, Fukuda K, Yding Andersen C and Byskov AG (1995) Healthy and atretic follicles: vaginosonographic detection and follicular fluid hormone profiles. Hum Reprod 10, 16331637.[Abstract]
Goldenberg RL, Reiter EO and Ross GT (1973) Follicle response to exogenous gonadotropins: an estrogen-mediated phenomenon. Fertil Steril 24, 121125.[ISI][Medline]
Hillier SG, Reichert LE, Jr and Van Hall EV (1981) Control of preovulatory follicular estrogen biosynthesis in the human ovary. J Clin Endocrinol Metab 52, 847856.[ISI][Medline]
Hillier SG, Wickings EJ, Saunders PT, Dixson AF, Shimasaki S, Swanston IA, Reichert LE, Jr and McNeilly AS (1989) Control of inhibin production by primate granulosa cells. J Endocrinol 123, 6573.[Abstract]
Jones HW Jr, Acosta A, Andrews MC, Garcia JE, Jones GS, Mantzavinos T, McDowell J, Sandow B, Veeck L, Whibley T et al. (1983) The importance of the follicular phase to success and failure in in vitro fertilization. Fertil Steril 40, 317321.[ISI][Medline]
Kessel B, Liu YX, Jia XC and Hsueh AJ (1985) Autocrine role of estrogens in the augmentation of luteinizing hormone receptor formation in cultured rat granulosa cells. Biol Reprod 32, 10381050.[Abstract]
Licciardi FL, Liu HC and Rosenwaks Z (1995) Day 3 estradiol serum concentrations as prognosticators of ovarian stimulation response and pregnancy outcome in patients undergoing in vitro fertilization. Fertil Steril 64, 991994.[ISI][Medline]
Louvet JP and Vaitukaitis JL (1976) Induction of follicle-stimulating hormone (FSH) receptors in rat ovaries by estrogen priming. Endocrinology 99, 758764.[Abstract]
Magoffin DA and Erickson GF (1982) Direct inhibitory effect of estrogen on LH-stimulated androgen synthesis by ovarian cells cultured in defined medium. Mol Cell Endocrinol 28, 8189.[CrossRef][ISI][Medline]
McNatty KP, Smith DM, Makris A, Osathanondh R and Ryan KJ (1979) The microenvironment of the human antral follicle: interrelationships among the steroid levels in antral fluid, the population of granulosa cells, and the status of the oocyte in vivo and in vitro. J Clin Endocrinol Metab 49, 851860.[ISI][Medline]
Phelps JY, Levine AS, Hickman TN, Zacur HA, Wallach EE and Hinton EL (1998) Day 4 estradiol levels predict pregnancy success in women undergoing controlled ovarian hyperstimulation for IVF. Fertil Steril 69, 10151019.[CrossRef][ISI][Medline]
Scaramuzzi RJ, Baird DT, Clarke IJ, Martensz ND and Van Look PF (1980) Ovarian morphology and the concentration of steroids during the oestrus cycle of sheep actively immunized against androstenedione. J Reprod Fertil 58, 2735.[ISI][Medline]
Sharma V, Riddle A, Mason BA, Pampiglione J and Campbell S (1988) An analysis of factors influencing the establishment of a clinical pregnancy in an ultrasound-based ambulatory in vitro fertilization program. Fertil Steril 49, 468478.[ISI][Medline]
Smotrich DB, Widra EA, Gindoff PR, Levy MJ, Hall JL and Stillman RJ (1995) Prognostic value of day 3 estradiol on in vitro fertilization outcome. Fertil Steril 64, 11361140.[ISI][Medline]
Submitted on June 23, 2004; accepted on September 10, 2004.
|