Basal 17β-estradiol did not correlate with ovarian response and in vitro fertilization treatment outcome

FERTILITY AND STERILITY威 VOL. 74, NO. 6, DECEMBER 2000 Copyright ©2000 American Society for Reproductive Medicine Published by Elsevier Science Inc. Printed on acid-free paper in U.S.A.

Basal 17␤-estradiol did not correlate with ovarian response and in vitro fertilization treatment outcome Phakphum Phophong, M.D., Domenico Massimo Ranieri, M.D., Iffat Khadum, M.D., Francesco Meo, M.D., and Paul Serhal, M.D. Assisted Conception Unit, University College London Hospital, London, United Kingdom

Objective: To verify the correlation of basal 17␤-E2 with ovarian response to stimulation and outcome of in vitro fertilization (IVF). Design: Retrospective observational study. Setting: The Assisted Conception Unit, University College London Hospitals. Patient(s): Three hundred five women undergoing IVF and IVF with intracytoplasmic sperm injection. Intervention(s): Basal follicle-stimulating hormone (FSH) and 17␤-E2 were assessed. The cutoff level for day 2 E2 established was 250 pmol/L. Each patient was noted for below (group A) or above (group B) the cutoff point according to her basal E2 level. Main Outcome Measure(s): Basal E2, age, duration of infertility, cycle day 2 FSH, number of ampules of gonadotropin used, number of days of stimulation, number of retrieved oocytes, fertilization rate, number of embryos transferred, number of cycles with embryo freezing, cancellation rate, clinical pregnancy rate, and implantation rate were compared between the two groups. Result(s): No differences were found between group A and group B in the number of oocytes retrieved (8.8 ⫾ 4.2 vs. 9.3 ⫾ 4.8), embryos transferred (2.5 ⫾ 0.8 vs. 2.7 ⫾ 0.7), cancellation (9.1% vs. 6.9%), pregnancy (24.8% vs. 30%), and implantation rate (12.3% vs. 15.6%). Correlation coefficient and coefficient of determination showed no significant correlation between basal E2 and the number of oocytes retrieved, age, and basal FSH. Conclusion(s): In our study population, basal E2 was not a sensitive predictor of ovarian response to stimulation and did not correlate with IVF outcome. (Fertil Steril威 2000;74:1133– 6. ©2000 by American Society for Reproductive Medicine.) Key Words: Basal E2, ovarian stimulation, IVF outcome

Received December 14, 1999; accepted June 23, 2000. Reprint requests: Domenico Massimo Ranieri, M.D., Assisted Conception Unit, University College London Hospital, 25 Grafton Way, Rosenheim Building, London WC1E 6DB, United Kingdom (FAX: 44-0171380-9957). 0015-0282/00/$20.00 PII S0015-0282(00)01621-6

The recruitment of a sufficient number of follicles during ovarian stimulation is a crucial factor in the success of assisted reproductive techniques. Failure of the ovaries to respond to gonadotropin stimulation is a negative prognostic factor in infertility treatment because it indicates reduced ovarian reserve. Although this is generally an age-related problem, its onset is highly variable and difficult to detect (1, 2). Patients with reduced ovarian reserve may have a shorter follicular phase with well-advanced follicular recruitment and a higher basal E2 level. Therefore, Licciardi et al. proposed basal E2 as an accurate predictor of ovarian reserve and IVF outcome. It was also suggested that in patients undergoing in vitro fertilization (IVF), basal E2 may be more ac-

curate than FSH in predicting ovarian response to stimulation and pregnancy outcome because elevated E2 levels can be associated, through negative feedback mechanism, with lower circulating FSH levels (3–5). Other investigators have suggested that normal baseline sex hormones are not able to confirm the ovulation performance in every patient (6, 7). A study by Lee et al. showed no difference in the E2 levels in different age groups of women with a regular menstrual cycle (8). Moreover, the different cutoff levels of basal E2 used in previous studies make the correlation of the data more difficult. This study was undertaken to verify the correlation of basal E2 to ovarian response to 1133

stimulation and outcome in the general population of patients undergoing IVF, rather than on specific subcategories of patients, such as older patients, who are known to have a worse prognosis.

MATERIALS AND METHODS A total of 305 patients undergoing ovarian stimulation for IVF or IVF and intracytoplasmic sperm injection (ICSI) were enrolled in this study between June 1997 and June 1999. The causes of infertility were male factor in 145 cases (47.5%), tubal factor in 61 cases (20.0%), unexplained in 63 cases (20.7%), endometriosis in 26 cases (8.5%), and male and tubal factors in 10 cases (3.3%). On cycle day 2, all patients underwent an ultrasound scan to rule out the presence of ovarian cysts, and a blood sample was taken to assess basal FSH and E2 levels. Long downregulation was commenced in the midluteal phase of the same cycle with buserelin nasal spray (Suprefact; Hoechst, Hounslow, Middlesex, UK) 100 ␮g into each nostril six hourly (800 ␮g/d). A basal ultrasound scan was performed on day 3 of the following menstruation, and ovarian stimulation started with a daily dose (150 –300 IU) of either hMG (Menogon; Ferring GmbH, Kiel, Germany) or recombinant FSH (Gonal-F; Serono, Welling Garden City, Hertfordshire, UK). The initial starting dose was individualized according to the patient’s age (⬎37 or ⬍37), but the day 3 E2 did not influence the stimulation protocol. Follicular growth was monitored by ultrasound scans from day 7 of stimulation. The dose of gonadotropin was adjusted according to the ovarian response. When at least two leading follicles had reached 18 mm in diameter, hCG (10,000 IU; Choragon; Ferring GmbH, Kiel, Germany) was administered. Thirty-six hours later, oocyte retrieval was performed transvaginally under ultrasound control. If fewer than three follicles of 15-mm diameter were recruited on the day of proposed hCG administration, the cycle was canceled. In our study population, there were 169 IVF cycles (60.8%) and 109 IVF⫹ICSI cycles (39.2%). Fertilization was confirmed by the presence of two pronuclei 18 –20 h after insemination. Embryo transfer was performed two-three days after oocyte retrieval. The luteal phase was supported with either progesterone vaginal pessaries (Cyclogest; 400 mg twice daily; LD Collins & Co Ltd, Mill Hill, London, UK), or progesterone intramuscular injection (Gestone; 100 mg intramuscularly once daily; Ferring Pharmaceuticals Ltd, Langley, Berkshire, UK). A quantitative serum ␤-hCG was performed 16 d after embryo transfer. Patients with a positive result underwent transvaginal sonography 2 weeks later. Only clinical pregnancies (positive result of quantitative ␤-hCG with intrauterine gestation sac, fetal pole, and heartbeat seen on scan at 6 weeks of gestation) were considered for calculation of pregnancy rate. Estradiol levels were measured by RIA. Duplicate ali1134 Phophong et al.

quots (0.2 mL) of plasma were mixed with 0.1 mL of 1.5 mol/L sodium carbonate solution, pH 10.5, and extracted with 10 volumes of diethyl ether. The mixture was frozen, the ether was decanted and evaporated, and the residue was dissolved in phosphate-buffered saline and incubated with tritium-labeled E2 (Amersham International, Little Chalfont, Buckinghamshire, UK) and sheep anti-E2 antiserum (Bioclin Services, Cardiff, Wales, UK). Bound and free E2 were separated with dextran-coated charcoal. The interassay coefficients of variation for two control plasmas were 9.5% and 6.6% for E2 concentrations of 174 and 409 pmol/L, respectively. The cutoff level for day 2 E2 of 250 pmol/L was established, and each patient was noted as below (group A) or above (group B) the cutoff point according to the basal E2 level. Patient age, duration of infertility, cycle day 2 FSH, number of ampules of gonadotrophin used, number of days of stimulation, number of retrieved oocytes, fertilization rate, number of embryos transferred, number of cycles with freezing embryo, cancellation rate, clinical pregnancy rate, and implantation rate were compared in the two groups of each cutoff level. Statistical analysis was conducted using the Epi-Info program (version 6.0). Student’s t-test, ␹2, and Fisher’s exact test were used when appropriate. Correlation coefficient and coefficient of determination were calculated to measure the strength of association between basal E2 and ovarian response to stimulation. Power analysis was also applied.

RESULTS The mean age (⫾ SD) of the 305 patients was 34.2 ⫾ 4.5 y, and mean duration of infertility was 4.0 ⫾ 2.1 y. The mean basal FSH level was 8.3 ⫾ 2.7 IU/L. Cycle day 2 E2 levels were 193.8 ⫾ 81.1 pmol/L. hMG was used for ovarian stimulation in 193 patients (63.3%) and recombinant FSH in 112 patients (36.7%). The treatment cycle was canceled for 27 patients (8.9%): 26 because of poor response and one because of risk of hyperstimulation. Of 305 cycles, 74 (24.3%) clinical pregnancies were established. When the cutoff level of day 2 E2 was applied to all 305 patients, no statistically significant differences in patient age or causes and duration of infertility were found. There was no statistically significant difference between cycle day 2 FSH level in group A vs. group B. Of 262 patients in group A, 24 (9.1%) had their cycle canceled. Three patients of 43 had their cycle abandoned in group B (6.9%). The mean E2 level in group A was 170 ⫾ 37.1, and in group B, it was 337 ⫾ 20.7. The E2 cutoff level of 250 pmol/L was then applied to the 278 patients who completed the cycle (Table 1). There were no significant differences in age (34 ⫾ 4.4 y vs. 34.0 ⫾ 4.9 y) and duration of infertility (4.0 ⫾ 2.2 y vs. 4.2 ⫾ 2.1 y). Cycle day 2 FSH levels in group A were 8.1 ⫾ 2.7 IU/L, and

Basal estradiol: ovarian response and IVF outcome

Vol. 74, No. 6, December 2000

TABLE 1 Comparison of ovarian response and IVF outcome at the cutoff level of basal estradiol of 250 pmol/L. Day 2 estradiol (pmol/L) ⬍250 pmol/L (n ⫽ 238)

Variable Age (y) Duration (y) Basal FSH (IU/L) No. of ampules of gonadotropin No. of d of stimulation No. of oocytes retrieved Fertilization rate (%) No. of embryos transferred No. of cycles with embryo freezing No. of pregnancies/total no. of cycles (%) No. of embryos that implanted/total no. of embryos transferred (%)

ⱖ250 pmol/L (n ⫽ 40)

34.0 ⫾ 4.4 4.0 ⫾ 2.2 8.1 ⫾ 2.7 47.1 ⫾ 20.0 11.0 ⫾ 1.2 8.8 ⫾ 4.2 63.8 ⫾ 26.2 2.5 ⫾ 0.8 19/238 (7.9%)

34.0 ⫾ 4.9 4.2 ⫾ 2.1 8.5 ⫾ 2.4 40.8 ⫾ 17.1 11.0 ⫾ 1.0 9.3 ⫾ 4.8 65.3 ⫾ 24.6 2.7 ⫾ 0.7 4/40 (10.0%)

59/238 (24.8%)

12/40 (30.0%)

75/612 (12.3%)

17/109 (15.6%)

Note: Values are means ⫾ SD unless otherwise indicated. All P values were not statistically significant. Phophong. Basal estradiol: ovarian response and IVF outcome. Fertil Steril 2000.

in group B, they were 8.5 ⫾ 2.4 IU/L (P⬎0.05). There were no significant differences noted in the starting dose of ampules (75 IU each) of gonadotropin (3.9 ⫾ 1.3 vs. 3.7 ⫾ 1.2), the total number of ampules of gonadotropin used (47.1 ⫾ 20.0 vs. 40.8 ⫾ 17.1), or the number of ovarian stimulation days (11.0 ⫾ 1.2 vs. 11.0 ⫾ 1.0). The number of retrieved oocytes in group A was 8.8 ⫾ 4.2 compared with 8.5 ⫾ 4.2 in group B. Fertilization rates were not different (63.8% ⫾ 26.2% vs. 65.3% ⫾ 24.6%). Similarly, no significant differences were found in the number of embryos transferred (2.5 ⫾ 0.8 vs. 2.7 ⫾ 0.7), pregnancy rate (24.8% vs. 30.0%), and implantation rate (12.3% vs. 15.6%). Correlation coefficient and coefficient of determination did not show significant correlation between basal E2 and, respectively, number of oocytes retrieved (r⫽0.00; r 2 ⫽ 0.00), age (r⫽⫺0.01; r 2 ⫽ 0.00), and basal FSH (r⫽0.03; r 2 ⫽ 0.00).

DISCUSSION The recruitment of a sufficient number of oocytes in patients undergoing ovarian stimulation for IVF is a crucial factor to successful treatment because it provides a selection of embryos for transfer (9). Diminished ovarian reserve can be defined as reduction in the number and quality of oocytes and is associated with reduced female fecundity and poor IVF outcome (10, 11). Although this is usually an agerelated phenomenon, it can unexpectedly manifest in younger patients and may be evidenced by a surprisingly poor ovarian response to stimulation for IVF. An accurate test of ovarian reserve has always been an attractive proposition to clinicians because of the attendant FERTILITY & STERILITY威

advantages of optimizing drug regimens for individual patients and assessing the likelihood of a successful outcome before commencing IVF treatment. From the various different tests available, basal E2 was proposed, and is still used, as an accurate predictor of ovarian response and IVF outcome (3–5). The present study was conducted to verify the correlation between basal E2, ovarian response to stimulation, and outcome in IVF treatment. Licciardi et al. (3) showed a significantly lower pregnancy rate in patients with basal E2 levels between 31 and 75 pg/mL and no pregnancy if the E2 levels were ⬎75 pg/mL in patients aged 35–36 y. Smotrich et al. (4) showed a lower pregnancy rate when the E2 level was ⬎80 pg/mL in 18 patients aged 35.8 ⫾ 0.9 y, and no pregnancy was achieved in 8 patients aged 37.1 ⫾ 1.3 with E2 levels ⬎ 100 pg/mL. In a more recent paper by Evers et al., the cutoff level of E2, correlated with poor ovarian response, was 60 pg/mL in 16 patients aged 33 ⫾ 4 y (5). We applied the cutoff point of 250 pmol/L (conversion factor, 3.671 ⫽ 68 pg/mL), which is a good balance among the cutoff points previously introduced. Forty patients aged 34 ⫾ 4.9 were found to have their basal E2 over the cutoff level considered. When the selected E2 cutoff level was applied to the 305 patients, the cycle day 2 E2 was not predictive of the cancellation rate. Day 2 FSH was similar in the group of patients with higher cutoff level of E2 (⬎250 pmol/L), thus confirming the lack of correlation between basal E2 and basal FSH (4). The cutoff level was then applied to the 278 patients who completed the cycle (Table 1). No difference was noticed between the two groups for any of the parameters compared. The lack of correlation between basal E2, number of oocytes retrieved, age, and basal FSH was also confirmed by the correlation coefficient and coefficient of determination. This was also in accordance with our data published in 1998, in which no correlation was found between ovarian response and basal E2 in patients undergoing IVF (12). In the current study, basal E2 could not be confirmed as an accurate predictor of ovarian response to stimulation. Our data have also shown that in the same group of patients, there was no correlation with the outcome of the IVF treatment. However, to reach a power analysis of 80% with 95% confidence, 1,000 patients for oocytes retrieved and 4,900 patients for the pregnancy rate would have to have been included in the study. Unfortunately this is beyond the ability of our treatment center. In our opinion, further studies are necessary on a larger number of patients to confirm the validity of basal E2 as predictor of ovarian response to stimulation and outcome in patients undergoing IVF. Meticulous individualization of its application may reveal that the test can be of benefit to specific subcategories, such as older women and patients with higher levels of basal E2. 1135

Moreover, the test used to measure the E2 level can give different results depending on the methodology used by the laboratory. Clinicians should establish independently their cutoff levels according to methods, threshold values, and clinically defined end points. Other tests, such as simultaneous evaluation of basal FSH and the response pattern of E2 to GnRH-a administration or inhibin, should also be investigated as more accurate predictors of response to stimulation in the general population (12, 13). References 1. Toner JP, Philput CB, Jones GS, Muasher SJ. Basal follicle-stimulating hormone level is a better predictor of in vitro fertilization performance than age. Fertil Steril 1991;55:784 –91. 2. Navot D, Rosenwaks Z, Margalioth EJ. Prognostic assessment of female fecundity. Lancet 1987;2:645–7. 3. Licciardi FL, Liu HC, Rosenwaks Z. Day 3 estradiol serum concentrations as prognosticators of ovarian stimulation response and pregnancy outcome in patients undergoing in vitro fertilization. Fertil Steril 1995; 64:991– 4. 4. Smotrich DB, Widra EA, Gindoff PR, Levy MJ, Hall JL, Stillman RJ. Prognostic value of day 3 estradiol on in vitro fertilization outcome. Fertil Steril 1995;64:1136 – 40.

1136 Phophong et al.

5. Evers JLH, Slaats P, Land JA, Dumoulin JCM, Dunselman GAJ. Elevated levels of basal 17␤ estradiol predict poor response in patients with normal basal levels of follicle-stimulating hormone undergoing in vitro fertilization. Fertil Steril 1998;69:1010 – 4. 6. Wallach EE. Pitfalls in evaluating ovarian reserve. Fertil Steril 1995; 63:12– 4. 7. Scott RT, Hofmann GE. Prognostic assessment of ovarian reserve. Fertil Steril 1995;63:1–11. 8. Lee SJ, Lenton EA, Sexton L, Cooke ID. The effect of age on the cyclical patterns of plasma LH, FSH, oestradiol and progesterone in women with regular menstrual cycles. Hum Reprod 1988;7:851–5. 9. Roest J, van Heusden AM, Mous H, Zeilmaker GH, Verhoeff A. The ovarian response as a predictor for successful in vitro fertilization treatment after the age of 40 years. Fertil Steril 1996;66:969 –73. 10. Meldrum DR. Female reproductive aging-ovarian and uterine factors. Fertil Steril 1993;59:1–5. 11. Navot D, Bergh PA, Williams MA, Garrisa GJ, Guzman I, Sandler B, et al. Poor oocyte quality rather than implantation failure as a cause of age-related decline in female infertility. Lancet 1991;337: 1375–7. 12. Ranieri DM, Quinn F, Makhlouf A, Khadum I, Ghutmi W, McGarrigle H, et al. Simultaneous evaluation of basal follicle-stimulating hormone and 17␤-estradiol response to gonadotropin-releasing hormone analogue stimulation: an improved predictor of ovarian reserve. Fertil Steril 1998;70:227–33. 13. Seifer DB, Gardiner AC, Lambert-Messerlian G. Day 3 serum inhibin-B is predictive of assisted reproductive technologies outcome. Fertil Steril 1997;67:110 – 4.

Basal estradiol: ovarian response and IVF outcome

Vol. 74, No. 6, December 2000