- Email: [email protected]
Low serum estradiol concentrations after five days of controlled ovarian hyperstimulation for in vitro fertilization are associated with poor outcome
FERTILITY AND STERILITY威 VOL. 74, NO. 1, JULY 2000 Copyright ©2000 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.
OVULATION INDUCTION
Low serum estradiol concentrations after five days of controlled ovarian hyperstimulation for in vitro fertilization are associated with poor outcome Yacoub Khalaf, M.D., M.R.C.O.G, Alison Taylor, M.D., and Peter Braude, Ph.D. Guy’s and St. Thomas’ Assisted Conception Unit, St. Thomas’ Hospital, London, United Kingdom
Objective: To evaluate the prognostic significance of low serum E2 concentrations in controlled ovarian hyperstimulation (COH) cycles for IVF. Design: Retrospective study. Setting: Assisted conception unit of a university hospital. Patient(s): One thousand four hundred and forty patients undergoing COH for IVF. Intervention(s): COH, serum E2 measurement, ultrasonographic scanning of ovarian follicles, oocyte retrieval, and ET. Main Outcome Measure(s): Cancellation and pregnancy rates. Result(s): Patients were classified into four groups according to serum E2 levels on the sixth day of COH: group A (E2 level ⬍ 50 pg/mL [114 cycles]), group B (E2 level 51–100 pg/mL [189 cycles]), group C (E2 level 101–200 pg/mL [320 cycles]), and group D (E2 level ⬎200 pg/mL [817 cycles]). Group A experienced the highest cancellation rates (65.1%) and lowest pregnancy rates (7.8%) despite requiring significantly more hMG ampules (47.8 ⫾ 1.7). The cancellation rate was higher (75.1%) and no pregnancy occurred in a subset of group A in whom COH was initiated with ⱖ3 ampules (225 IU) of gonadotropins. Conclusion(s): In COH cycles using luteal phase buserelin, low initial serum E2 concentrations are associated with poor outcome. (Fertil Steril威 2000;74:63– 6. ©2000 by American Society for Reproductive Medicine.) Key Words: Estradiol, controlled ovarian stimulation, IVF outcome
Received August 12, 1999; accepted January 27, 2000. Presented in part at the British Fertility Society meeting, Dundee, United Kingdom, December 15– 16, 1996. Reprint requests: Yacoub Khalaf, M.D., M.R.C.O.G, Guy’s and St. Thomas’ Assisted Conception Unit, St. Thomas’ Hospitals, London SE1 7EH, United Kingdom (FAX: 44 171 928 4639; E-mail: y.khalaf@umds .ac.uk). 0015-0282/00/$20.00 PII S0015-0282(00)00569-0
Controlled ovarian hyperstimulation (COH) is a standard practice in assisted conception programs. Monitoring of COH in IVF aims to evaluate whether the response to exogenous gonadotropins is sufficient to obtain an adequate number of mature oocytes. The second aim is to determine the optimal time for induction of final oocyte maturation by hCG administration. Currently, COH is monitored by repeated pelvic ultrasonography or serum E2 measurement. It is believed that ultrasonographic findings reflect growth, whereas serum E2 levels primarily detect functional activity of follicles.
significance of low serum E2 levels in the early follicular phase of GnRHa/hMG-stimulated IVF cycles.
Several investigators have evaluated various hormonal and ultrasonographic variables during stimulation to determine the prognosis for pregnancy in a given cycle (1– 4). The objective of this study was to examine the prognostic
Controlled ovarian hyperstimulation was performed by using hMG (Pergonal and Metrodin; Serono Laboratories Ltd., Welwyn Garden City, UK). The initial dose of hMG was individualized on the basis of age, basal day 3
MATERIALS AND METHODS Using a computerized database, we reviewed the records of all COH cycles for IVF-ET that were performed between January 1992 and March 1996. The search included all patients at St. Thomas’ Hospital who underwent IVF-ET using buserelin (Suprefact; Hoechst UK Ltd., Hounslow, UK) in the midluteal long protocol.
63
TABLE 1 Comparison of IVF cycles according to E2 concentration after 5 days of hMG (ⱖ150 IU/d) stimulation. Variable
Group A (⬍50 pg/mL)
Group B (51–100 pg/mL)
Group C (101–200 pg/mL)
Group D (⬎200 pg/mL)
No. of cycles Day 9 E2 concentration (pg/mL) No. of hMG ampules per cycle Cycle cancellation (%) No. of oocytes per retrieval Fertilization (2PN) (%) Pregnancy per cycle (%)
114 160.2 ⫾ 29.7 47.8 ⫾ 1.7 65.1 7.6 ⫾ 0.5 60.4 ⫾ 5.0 7.8
189 305.2 ⫾ 13.1 46.0 ⫾ 1.4 25.9 7.0 ⫾ 0.4 59.0 ⫾ 3.0 13.2
320 574.2 ⫾ 29.7 40.2 ⫾ 0.9 16.6 7.4 ⫾ 0.2 60.9 ⫾ 2.0 22.1
817 1346.0 ⫾ 35.2a 32.6 ⫾ 1.0b 9.9a 9.5 ⫾ 0.2 63.1 ⫾ 3.0 26.4b
a b
P⬍.001. P⬍.01.
Khalaf. Low serum E2 concentrations. Fertil Steril 2000.
FSH level, and response to COH in previous cycle(s). Generally, patients ⱕ30 years of age were started on 2 ampules, patients 31–35 years of age were started on 3 ampules, and patients ⬎35 years of age were started on 4 ampules. Patients who previously showed suboptimal response to these standard doses or have a day 3 FSH level ⱖ10 U/L were started on 6 ampules. Subsequent adjustments were made on the basis of individual responses. In patients whose day 6 E2 concentration was ⬍100 pg/mL, the initial daily dose was doubled, increased to 6 ampules if the initial dose was ⬍6 ampules, and was unchanged if the initial dose was the maximum daily dose (6 ampules).
9.7, 5.1, and 4.6% at concentrations of 57, 567, and 1564 pg/mL, respectively. We used ultrasonography and serum E2 measurement to monitor ovarian response to stimulation. Patients were categorized into four groups on the basis of day 6 serum E2 concentrations: group A (⬍50 pg/mL [114 cycles]), group B (51–100 pg/mL [189 cycles]), group C (101–200 pg/mL [320 cycles]), and group D (⬎200 pg/mL [817 cycles]) (conversion factor to SI unit, 3.671). Results are expressed as means ⫾ SE. Statistical comparisons were made by using the 2, Student t-test, or analysis of variance where appropriate; Statview software was used for statistical analysis (Abacus, Berkeley, CA). Statistical significance was set at P⬍.05. Because of the study design, consultation with our Institutional Ethics Committee was not required.
Human chorionic gonadotropin 10,000 IU (Profasi; Serono Laboratories Ltd.), was administered when at least three follicles were ⱖ18 mm. Transvaginal follicular aspiration was performed 34 –36 hours later. Serum samples were analyzed for E2 by using the DELFIA time-resolved fluorescence assay of Wallace (EG&G Ltd., Milton Keynes, UK). The reported intraassay imprecision for E2 was 10%, 5.0%, and 5.7% at concentrations of 38, 294, 2969 pg/mL, respectively. The total between-assay imprecision for E2 was
RESULTS The mean patient age was similar in all groups. Groups A and B had significantly lower E2 concentrations on day 9 of stimulation. The total numbers of hMG ampules required for
TABLE 2 Comparison of IVF cycles according to E2 concentration after 5 days of hMG (ⱖ225 IU/d) stimulation. Variable
Group A (⬍50 pg/mL)
Group B (51–100 pg/mL)
Group C (101–200 pg/mL)
Group D (⬎200 pg/mL)
No. of cycles Day 9 E2 concentration (pg/mL) No. of hMG ampules per cycle Cycle cancellation (%) No. of oocytes per retrieval Fertilization (2PN) (%) Pregnancy per cycle (%)
50 160.2 ⫾ 66.0 57.9 ⫾ 2.7 74.1 5.5 ⫾ 0.9 60.4 ⫾ 9 0.0
106 262.6 ⫾ 16.1 58.9 ⫾ 1.8 27.4 6.1 ⫾ 0.4 56.0 ⫾ 3 6.5
165 561.1 ⫾ 20.2 51.3 ⫾ 1.3 21.8 7.0 ⫾ 0.3 57.0 ⫾ 2 20.9
452 1323.7 ⫾ 46.4a 41.6 ⫾ 1.2a 9.7a 8.5 ⫾ 0.2 62.0 ⫾ 2 24.2b
a b
P⬍.001. P⬍.01.
Khalaf. Low serum E2 concentrations. Fertil Steril 2000.
64
Khalaf et al.
Significance of low day 6 estradiol in IVF
Vol. 74, No. 1, July 2000
ovarian stimulation were 48.9 ⫾ 2.7 and 46.0 ⫾ 1.8 in groups A and B, respectively, compared with 40.2 ⫾ 9 and 32.6 ⫾ 1.0 ampules in groups C and D, respectively (P⬍ .001). A significantly higher cancellation rate (55%) was observed in group A even though the hMG dose was increased during the course of stimulation in over 60% of this group. In group B, although the cancellation rate (27.5%) was significantly lower than that in group A, it was significantly higher (P⬍.001) than the cancellation rate in groups C and D (16.9 and 9.8%, respectively). The mean numbers of oocytes obtained were similar among groups A, B, and C (7.6 ⫾ 0.6, 7.0 ⫾ 0.3, and 7.5 ⫾ 0.2, respectively); but significantly more oocytes were obtained in group D (9.2 ⫾ 0.2). However, the differences in fertilization rates were not statistically significant among the groups. Pregnancy rates were significantly lower (P⬍.001) in groups A and B (7.8% and 13.2%, respectively) than in groups C and D (22.1% and 26.4%, respectively) (Table 1). Cycles in which ⱖ225 IU of hMG were used to initiate COH (n ⫽ 727) were analyzed. The differences among groups were more remarkable. Although groups A and B required a significantly higher mean total number of hMG ampules for stimulation (P⬍.001), cancellation rates were significantly higher (74.1% and 27.4%, respectively; P⬍.001) and significantly fewer oocytes were produced (5.5 ⫾ 0.9 and 6.1 ⫾ 0.4, respectively. No pregnancy was achieved in group A, and only 6.5% of patients conceived in group B. In contrast pregnancy rates of 20.9% and 24.2% were achieved in groups C and D, respectively (P⬍.001) (Table 2).
DISCUSSION Early follicular phase E2 level predicts the outcome of COH cycles using buserelin acetate. Cycles in which E2 concentrations are ⬍100 pg/mL have a poor outcome compared with cycles in which serum E2 concentrations are higher. Very low levels of E2 (⬍50 pg/mL) are associated with worse outcome. In our study, groups A and B maintained low E2 levels even though the dose of hMG was increased during the course of stimulation cycle. Significantly more ampules of hMG were used for stimulation in these two groups. Increasing or doubling the dose of hMG administered during the course of stimulation does not enhance poor initial response to ovarian stimulation (5). Furthermore, in groups A and B, significantly more cycles had to be cancelled because of very low E2 concentrations (⬍300 pg/mL) or too few follicles (⬍3 follicles 18 mm in diameter) on the day of hCG administration. The lower the level of E2 on day 6 of stimulation, the higher the likelihood of cycle cancellation, especially in patients in whom COH was initiated with ⱖ225 IU of hMG. In all groups, patients in whom cycles were not cancelled produced similar numbers of oocytes retrieved. Fertilization rates were similar FERTILITY & STERILITY威
among the four groups, whereas pregnancy rates were significantly lower in groups A and B. Hershlag et al. (8) examined 18 IVF cycles in which the E2 level after 5 days of gonadotropin stimulation was ⬍50 pg/mL. They reported a very poor outcome of these cycles despite increasing the dose of hMG during the course of stimulation (8). However, for obvious pharmacologic and physiologic reasons, the follicular phase variables used for cycle cancellation in hMG-stimulated IVF cycles cannot be extrapolated to GnRHa/hMG cycles (6). However, in a recent study, some investigators have demonstrated that low day 4 serum E2 levels in COH cycles using luteal phase leuprolide acetate predict higher cancellation rates because of poor response (4). Some investigators have suggested that pregnancy rate is independent of E2 level on the day of hCG administration (6). In contrast, other investigators have demonstrated that the pattern of E2 production in the terminal phase of follicular growth can predict IVF outcome (1). This latter assertion is supported by a recent study (7) that examined assisted reproduction outcome in five patient groups with different ratios of serum E2 levels on the day of hCG administration to oocytes. Highly significant differences in pregnancy rates were observed among patient groups. In patients who had a very low E2/oocyte ratio (⬍70 pg/mL per oocyte), the pregnancy rate was just 5.4% compared with 31.3% per oocyte retrieval when the E2/oocyte ratio was 70 –140 pg/mL. These results support our finding of significantly lower pregnancy rates in IVF cycles with low E2 concentrations. A starting dose of 225 IU of gonadotropins may be superior to a dose of 150 IU in producing adequate ovarian response with significantly lower cancellation rates (9). Failure of a daily dose of 225 IU of hMG to result in E2 concentrations ⬎50 pg/mL on the sixth day of a COH cycle was associated with poor outcome. No pregnancy was achieved in these patients in our study. The poor outcome of cycles with low initial E2 levels could be explained by the direct effect of E2 on modulating oocyte function (10). It may also result from lack of an adequate estrogenic intrafollicular milieu, which is important for the oocyte’s acquisition of developmental competence and cytoplasmic maturation, including activation, synthesis of the male pronucleus growth factor, and preimplantation development (11). Additionally, some investigators have suggested that intrafollicular E2 level is highly correlated with and may be a good predictor of oocyte function (12). We conclude that in down-regulated cycles, low E2 concentration after 5 days of gonadotropin stimulation predicts a high likelihood of cycle cancellation because of poor response even if the dose of hMG is increased during the course of the same treatment cycle. Our study also suggests that IVF cycles with low initial E2 concentration have lower fecundability even when similar numbers of oocytes and fertilization rates have been obtained. It is not always pos65
sible to predict poor responders before COH; therefore, it is important to identify hyporesponders as early as possible. Such patients can then make an informed decision about whether to proceed with COH or stop ovarian stimulation, thereby reducing the emotional and financial burden of therapy. Should they opt to continue with COH, they would at least have a more realistic expectation of its effectiveness. References
Khalaf et al.
7.
8.
1. Jones HW Jr, Acosta A, Andrews MC, Garcia JE, Jones GS, Mantzavinos T, et al. The importance of the follicular phase to success and failure in in vitro fertilization. Fertil Steril 1983;40:317–21. 2. Vargyas J, Marrs R, Kletsky D, Mishell D. Correlation of ultrasound measurement of ovarian follicle size and serum estradiol levels in ovulatory patients following clomiphene citrate for in vitro fertilization. Am J Obstet Gynecol 1982;144:569 –73. 3. Mettler L, Tavmergen EN. Significance of oestradiol values in IVF-ET under a combined GnRH analogue— desensitization and simultaneous gonadotrophin stimulation for the outcome of pregnancies. Hum Reprod 1989;4(8 Suppl):59 – 64. 4. Phelps JY, Levine AS, Hickman TN, Zacur HA, Wallach EE, Hinton EL. Day 4 estradiol levels predict pregnancy success in women undergoing controlled ovarian hyperstimulation for IVF. Fertil Steril 1998; 69:1015–9. 5. van Hoof MH, Alberda AT, Huisman GJ, Zeilmaker GH, Leerentveld
66
6.
Significance of low day 6 estradiol in IVF
9.
10. 11. 12.
RA. Doubling the human menopausal gonadotrophin dose in the course of an in-vitro fertilization treatment cycle in low responders: a randomized study. Hum Reprod 1993;8:369 –73. Forman RG, Robinson J, Egan D, Ross C, Gosden B, Barlow DH. Follicular monitoring and outcome of in vitro fertilization in gonadotropin-releasing hormone agonist-treated cycles. Fertil Steril 1991;55: 567–73. Loumaye E, Engrand P, Howles CM, O’Dea L. Assessment of the role of serum luteinizing hormone and estradiol response to follicle-stimulating hormone on in vitro fertilization treatment outcome. Fertil Steril 1997;67:889 –99. Hershlag A, Asis MC, Diamond MP, DeCherney AH, Lavy G. The predictive value and the management of cycles with low initial estradiol levels. Fertil Steril 1990;53:1064 –7. Abu-Heija AT, Yates RW, Barrett T, Jamieson ME, Fleming R, Coutts JR. A comparison of two starting doses of human menopausal gonadotrophin for follicle stimulation in unselected patients for in-vitro fertilization. Hum Reprod 1995;10:801–3. Danforth DR. Endocrine and paracrine control of oocyte development. Am J Obstet Gynecol 1995;172(2 Pt 2):747–52. Armstrong DT, Zhang X, Vanderhyden BC, Khamsi F. Hormonal actions during maturation influence fertilization and early embryonic development. Ann N Y Acad Sci 1991;626:137–58. McNatty KP, Smith DM, Makris A, Osathanondh R, Ryan KJ. 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 1979;49:851– 60.
Vol. 74, No. 1, July 2000
OVULATION INDUCTION
Low serum estradiol concentrations after five days of controlled ovarian hyperstimulation for in vitro fertilization are associated with poor outcome Yacoub Khalaf, M.D., M.R.C.O.G, Alison Taylor, M.D., and Peter Braude, Ph.D. Guy’s and St. Thomas’ Assisted Conception Unit, St. Thomas’ Hospital, London, United Kingdom
Objective: To evaluate the prognostic significance of low serum E2 concentrations in controlled ovarian hyperstimulation (COH) cycles for IVF. Design: Retrospective study. Setting: Assisted conception unit of a university hospital. Patient(s): One thousand four hundred and forty patients undergoing COH for IVF. Intervention(s): COH, serum E2 measurement, ultrasonographic scanning of ovarian follicles, oocyte retrieval, and ET. Main Outcome Measure(s): Cancellation and pregnancy rates. Result(s): Patients were classified into four groups according to serum E2 levels on the sixth day of COH: group A (E2 level ⬍ 50 pg/mL [114 cycles]), group B (E2 level 51–100 pg/mL [189 cycles]), group C (E2 level 101–200 pg/mL [320 cycles]), and group D (E2 level ⬎200 pg/mL [817 cycles]). Group A experienced the highest cancellation rates (65.1%) and lowest pregnancy rates (7.8%) despite requiring significantly more hMG ampules (47.8 ⫾ 1.7). The cancellation rate was higher (75.1%) and no pregnancy occurred in a subset of group A in whom COH was initiated with ⱖ3 ampules (225 IU) of gonadotropins. Conclusion(s): In COH cycles using luteal phase buserelin, low initial serum E2 concentrations are associated with poor outcome. (Fertil Steril威 2000;74:63– 6. ©2000 by American Society for Reproductive Medicine.) Key Words: Estradiol, controlled ovarian stimulation, IVF outcome
Received August 12, 1999; accepted January 27, 2000. Presented in part at the British Fertility Society meeting, Dundee, United Kingdom, December 15– 16, 1996. Reprint requests: Yacoub Khalaf, M.D., M.R.C.O.G, Guy’s and St. Thomas’ Assisted Conception Unit, St. Thomas’ Hospitals, London SE1 7EH, United Kingdom (FAX: 44 171 928 4639; E-mail: y.khalaf@umds .ac.uk). 0015-0282/00/$20.00 PII S0015-0282(00)00569-0
Controlled ovarian hyperstimulation (COH) is a standard practice in assisted conception programs. Monitoring of COH in IVF aims to evaluate whether the response to exogenous gonadotropins is sufficient to obtain an adequate number of mature oocytes. The second aim is to determine the optimal time for induction of final oocyte maturation by hCG administration. Currently, COH is monitored by repeated pelvic ultrasonography or serum E2 measurement. It is believed that ultrasonographic findings reflect growth, whereas serum E2 levels primarily detect functional activity of follicles.
significance of low serum E2 levels in the early follicular phase of GnRHa/hMG-stimulated IVF cycles.
Several investigators have evaluated various hormonal and ultrasonographic variables during stimulation to determine the prognosis for pregnancy in a given cycle (1– 4). The objective of this study was to examine the prognostic
Controlled ovarian hyperstimulation was performed by using hMG (Pergonal and Metrodin; Serono Laboratories Ltd., Welwyn Garden City, UK). The initial dose of hMG was individualized on the basis of age, basal day 3
MATERIALS AND METHODS Using a computerized database, we reviewed the records of all COH cycles for IVF-ET that were performed between January 1992 and March 1996. The search included all patients at St. Thomas’ Hospital who underwent IVF-ET using buserelin (Suprefact; Hoechst UK Ltd., Hounslow, UK) in the midluteal long protocol.
63
TABLE 1 Comparison of IVF cycles according to E2 concentration after 5 days of hMG (ⱖ150 IU/d) stimulation. Variable
Group A (⬍50 pg/mL)
Group B (51–100 pg/mL)
Group C (101–200 pg/mL)
Group D (⬎200 pg/mL)
No. of cycles Day 9 E2 concentration (pg/mL) No. of hMG ampules per cycle Cycle cancellation (%) No. of oocytes per retrieval Fertilization (2PN) (%) Pregnancy per cycle (%)
114 160.2 ⫾ 29.7 47.8 ⫾ 1.7 65.1 7.6 ⫾ 0.5 60.4 ⫾ 5.0 7.8
189 305.2 ⫾ 13.1 46.0 ⫾ 1.4 25.9 7.0 ⫾ 0.4 59.0 ⫾ 3.0 13.2
320 574.2 ⫾ 29.7 40.2 ⫾ 0.9 16.6 7.4 ⫾ 0.2 60.9 ⫾ 2.0 22.1
817 1346.0 ⫾ 35.2a 32.6 ⫾ 1.0b 9.9a 9.5 ⫾ 0.2 63.1 ⫾ 3.0 26.4b
a b
P⬍.001. P⬍.01.
Khalaf. Low serum E2 concentrations. Fertil Steril 2000.
FSH level, and response to COH in previous cycle(s). Generally, patients ⱕ30 years of age were started on 2 ampules, patients 31–35 years of age were started on 3 ampules, and patients ⬎35 years of age were started on 4 ampules. Patients who previously showed suboptimal response to these standard doses or have a day 3 FSH level ⱖ10 U/L were started on 6 ampules. Subsequent adjustments were made on the basis of individual responses. In patients whose day 6 E2 concentration was ⬍100 pg/mL, the initial daily dose was doubled, increased to 6 ampules if the initial dose was ⬍6 ampules, and was unchanged if the initial dose was the maximum daily dose (6 ampules).
9.7, 5.1, and 4.6% at concentrations of 57, 567, and 1564 pg/mL, respectively. We used ultrasonography and serum E2 measurement to monitor ovarian response to stimulation. Patients were categorized into four groups on the basis of day 6 serum E2 concentrations: group A (⬍50 pg/mL [114 cycles]), group B (51–100 pg/mL [189 cycles]), group C (101–200 pg/mL [320 cycles]), and group D (⬎200 pg/mL [817 cycles]) (conversion factor to SI unit, 3.671). Results are expressed as means ⫾ SE. Statistical comparisons were made by using the 2, Student t-test, or analysis of variance where appropriate; Statview software was used for statistical analysis (Abacus, Berkeley, CA). Statistical significance was set at P⬍.05. Because of the study design, consultation with our Institutional Ethics Committee was not required.
Human chorionic gonadotropin 10,000 IU (Profasi; Serono Laboratories Ltd.), was administered when at least three follicles were ⱖ18 mm. Transvaginal follicular aspiration was performed 34 –36 hours later. Serum samples were analyzed for E2 by using the DELFIA time-resolved fluorescence assay of Wallace (EG&G Ltd., Milton Keynes, UK). The reported intraassay imprecision for E2 was 10%, 5.0%, and 5.7% at concentrations of 38, 294, 2969 pg/mL, respectively. The total between-assay imprecision for E2 was
RESULTS The mean patient age was similar in all groups. Groups A and B had significantly lower E2 concentrations on day 9 of stimulation. The total numbers of hMG ampules required for
TABLE 2 Comparison of IVF cycles according to E2 concentration after 5 days of hMG (ⱖ225 IU/d) stimulation. Variable
Group A (⬍50 pg/mL)
Group B (51–100 pg/mL)
Group C (101–200 pg/mL)
Group D (⬎200 pg/mL)
No. of cycles Day 9 E2 concentration (pg/mL) No. of hMG ampules per cycle Cycle cancellation (%) No. of oocytes per retrieval Fertilization (2PN) (%) Pregnancy per cycle (%)
50 160.2 ⫾ 66.0 57.9 ⫾ 2.7 74.1 5.5 ⫾ 0.9 60.4 ⫾ 9 0.0
106 262.6 ⫾ 16.1 58.9 ⫾ 1.8 27.4 6.1 ⫾ 0.4 56.0 ⫾ 3 6.5
165 561.1 ⫾ 20.2 51.3 ⫾ 1.3 21.8 7.0 ⫾ 0.3 57.0 ⫾ 2 20.9
452 1323.7 ⫾ 46.4a 41.6 ⫾ 1.2a 9.7a 8.5 ⫾ 0.2 62.0 ⫾ 2 24.2b
a b
P⬍.001. P⬍.01.
Khalaf. Low serum E2 concentrations. Fertil Steril 2000.
64
Khalaf et al.
Significance of low day 6 estradiol in IVF
Vol. 74, No. 1, July 2000
ovarian stimulation were 48.9 ⫾ 2.7 and 46.0 ⫾ 1.8 in groups A and B, respectively, compared with 40.2 ⫾ 9 and 32.6 ⫾ 1.0 ampules in groups C and D, respectively (P⬍ .001). A significantly higher cancellation rate (55%) was observed in group A even though the hMG dose was increased during the course of stimulation in over 60% of this group. In group B, although the cancellation rate (27.5%) was significantly lower than that in group A, it was significantly higher (P⬍.001) than the cancellation rate in groups C and D (16.9 and 9.8%, respectively). The mean numbers of oocytes obtained were similar among groups A, B, and C (7.6 ⫾ 0.6, 7.0 ⫾ 0.3, and 7.5 ⫾ 0.2, respectively); but significantly more oocytes were obtained in group D (9.2 ⫾ 0.2). However, the differences in fertilization rates were not statistically significant among the groups. Pregnancy rates were significantly lower (P⬍.001) in groups A and B (7.8% and 13.2%, respectively) than in groups C and D (22.1% and 26.4%, respectively) (Table 1). Cycles in which ⱖ225 IU of hMG were used to initiate COH (n ⫽ 727) were analyzed. The differences among groups were more remarkable. Although groups A and B required a significantly higher mean total number of hMG ampules for stimulation (P⬍.001), cancellation rates were significantly higher (74.1% and 27.4%, respectively; P⬍.001) and significantly fewer oocytes were produced (5.5 ⫾ 0.9 and 6.1 ⫾ 0.4, respectively. No pregnancy was achieved in group A, and only 6.5% of patients conceived in group B. In contrast pregnancy rates of 20.9% and 24.2% were achieved in groups C and D, respectively (P⬍.001) (Table 2).
DISCUSSION Early follicular phase E2 level predicts the outcome of COH cycles using buserelin acetate. Cycles in which E2 concentrations are ⬍100 pg/mL have a poor outcome compared with cycles in which serum E2 concentrations are higher. Very low levels of E2 (⬍50 pg/mL) are associated with worse outcome. In our study, groups A and B maintained low E2 levels even though the dose of hMG was increased during the course of stimulation cycle. Significantly more ampules of hMG were used for stimulation in these two groups. Increasing or doubling the dose of hMG administered during the course of stimulation does not enhance poor initial response to ovarian stimulation (5). Furthermore, in groups A and B, significantly more cycles had to be cancelled because of very low E2 concentrations (⬍300 pg/mL) or too few follicles (⬍3 follicles 18 mm in diameter) on the day of hCG administration. The lower the level of E2 on day 6 of stimulation, the higher the likelihood of cycle cancellation, especially in patients in whom COH was initiated with ⱖ225 IU of hMG. In all groups, patients in whom cycles were not cancelled produced similar numbers of oocytes retrieved. Fertilization rates were similar FERTILITY & STERILITY威
among the four groups, whereas pregnancy rates were significantly lower in groups A and B. Hershlag et al. (8) examined 18 IVF cycles in which the E2 level after 5 days of gonadotropin stimulation was ⬍50 pg/mL. They reported a very poor outcome of these cycles despite increasing the dose of hMG during the course of stimulation (8). However, for obvious pharmacologic and physiologic reasons, the follicular phase variables used for cycle cancellation in hMG-stimulated IVF cycles cannot be extrapolated to GnRHa/hMG cycles (6). However, in a recent study, some investigators have demonstrated that low day 4 serum E2 levels in COH cycles using luteal phase leuprolide acetate predict higher cancellation rates because of poor response (4). Some investigators have suggested that pregnancy rate is independent of E2 level on the day of hCG administration (6). In contrast, other investigators have demonstrated that the pattern of E2 production in the terminal phase of follicular growth can predict IVF outcome (1). This latter assertion is supported by a recent study (7) that examined assisted reproduction outcome in five patient groups with different ratios of serum E2 levels on the day of hCG administration to oocytes. Highly significant differences in pregnancy rates were observed among patient groups. In patients who had a very low E2/oocyte ratio (⬍70 pg/mL per oocyte), the pregnancy rate was just 5.4% compared with 31.3% per oocyte retrieval when the E2/oocyte ratio was 70 –140 pg/mL. These results support our finding of significantly lower pregnancy rates in IVF cycles with low E2 concentrations. A starting dose of 225 IU of gonadotropins may be superior to a dose of 150 IU in producing adequate ovarian response with significantly lower cancellation rates (9). Failure of a daily dose of 225 IU of hMG to result in E2 concentrations ⬎50 pg/mL on the sixth day of a COH cycle was associated with poor outcome. No pregnancy was achieved in these patients in our study. The poor outcome of cycles with low initial E2 levels could be explained by the direct effect of E2 on modulating oocyte function (10). It may also result from lack of an adequate estrogenic intrafollicular milieu, which is important for the oocyte’s acquisition of developmental competence and cytoplasmic maturation, including activation, synthesis of the male pronucleus growth factor, and preimplantation development (11). Additionally, some investigators have suggested that intrafollicular E2 level is highly correlated with and may be a good predictor of oocyte function (12). We conclude that in down-regulated cycles, low E2 concentration after 5 days of gonadotropin stimulation predicts a high likelihood of cycle cancellation because of poor response even if the dose of hMG is increased during the course of the same treatment cycle. Our study also suggests that IVF cycles with low initial E2 concentration have lower fecundability even when similar numbers of oocytes and fertilization rates have been obtained. It is not always pos65
sible to predict poor responders before COH; therefore, it is important to identify hyporesponders as early as possible. Such patients can then make an informed decision about whether to proceed with COH or stop ovarian stimulation, thereby reducing the emotional and financial burden of therapy. Should they opt to continue with COH, they would at least have a more realistic expectation of its effectiveness. References
Khalaf et al.
7.
8.
1. Jones HW Jr, Acosta A, Andrews MC, Garcia JE, Jones GS, Mantzavinos T, et al. The importance of the follicular phase to success and failure in in vitro fertilization. Fertil Steril 1983;40:317–21. 2. Vargyas J, Marrs R, Kletsky D, Mishell D. Correlation of ultrasound measurement of ovarian follicle size and serum estradiol levels in ovulatory patients following clomiphene citrate for in vitro fertilization. Am J Obstet Gynecol 1982;144:569 –73. 3. Mettler L, Tavmergen EN. Significance of oestradiol values in IVF-ET under a combined GnRH analogue— desensitization and simultaneous gonadotrophin stimulation for the outcome of pregnancies. Hum Reprod 1989;4(8 Suppl):59 – 64. 4. Phelps JY, Levine AS, Hickman TN, Zacur HA, Wallach EE, Hinton EL. Day 4 estradiol levels predict pregnancy success in women undergoing controlled ovarian hyperstimulation for IVF. Fertil Steril 1998; 69:1015–9. 5. van Hoof MH, Alberda AT, Huisman GJ, Zeilmaker GH, Leerentveld
66
6.
Significance of low day 6 estradiol in IVF
9.
10. 11. 12.
RA. Doubling the human menopausal gonadotrophin dose in the course of an in-vitro fertilization treatment cycle in low responders: a randomized study. Hum Reprod 1993;8:369 –73. Forman RG, Robinson J, Egan D, Ross C, Gosden B, Barlow DH. Follicular monitoring and outcome of in vitro fertilization in gonadotropin-releasing hormone agonist-treated cycles. Fertil Steril 1991;55: 567–73. Loumaye E, Engrand P, Howles CM, O’Dea L. Assessment of the role of serum luteinizing hormone and estradiol response to follicle-stimulating hormone on in vitro fertilization treatment outcome. Fertil Steril 1997;67:889 –99. Hershlag A, Asis MC, Diamond MP, DeCherney AH, Lavy G. The predictive value and the management of cycles with low initial estradiol levels. Fertil Steril 1990;53:1064 –7. Abu-Heija AT, Yates RW, Barrett T, Jamieson ME, Fleming R, Coutts JR. A comparison of two starting doses of human menopausal gonadotrophin for follicle stimulation in unselected patients for in-vitro fertilization. Hum Reprod 1995;10:801–3. Danforth DR. Endocrine and paracrine control of oocyte development. Am J Obstet Gynecol 1995;172(2 Pt 2):747–52. Armstrong DT, Zhang X, Vanderhyden BC, Khamsi F. Hormonal actions during maturation influence fertilization and early embryonic development. Ann N Y Acad Sci 1991;626:137–58. McNatty KP, Smith DM, Makris A, Osathanondh R, Ryan KJ. 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 1979;49:851– 60.
Vol. 74, No. 1, July 2000