Periovulatory 17β-estradiol pattern in conceptional and nonconceptional cycles during menotropin treatment of anovulatory infertility

FERTILITY AND STERILITY Copyright © 1987 The American Fertility Society

Vol. 47. No.2. February 1987 Printed in U.8A.

Periovulatory 17~-estradiol pattern in conceptional and nonconceptional cycles during menotropin treatment of anovulatory infertility

Daniel Navot, M.D.* Ehud J. Margalioth, M.D. Neri Laufer, M.D. Amnon Brzezinski, M.D. Arie Birkenfeld, M.D. Joseph G. Schenker, M.D. Department of Obstetrics and Gynecology, Hadassah University Hospital, Jerusalem, Israel

The value of multiple parameters in the prediction of fertile cycles was prospectively evaluated in 52 menotropincinduced cycles. The periovulatory pattern of estradiol (E;d was found to correlate with conceptional cycles. E2 levels> 500 pg/ml on the day of human chorionic gonadotropin administration (day 0) with a further increase on day + 1 (high ascending pattern-AJ) were found to have a 51% predictive value for fertile cycles. Twelve of the 17 fertile cycles had an AI> type of response (71%), whereas the overall incidence of an AJ pattern was 42% (22 of 52). No pregnancies have occurred with preovulatory follicles:%: 14 mm in diameter. The number of preovulatory follicles, E2 level on day 0, and midluteal progesterone had no predictive value for fertile cycles. Fertil Steril 47:234, 1987

In the natural ovulatory cycle the spontaneous luteinizing hormone (LH) surge is followed by a precipitous fall in serum 17J3-estradiol (E 2 ) levels. In contrast, during menotropin (human menopausal gonadotropin, hMG) induction of ovulation the ovulatory dose of human chorionic gonadotropin (hCG) may be followed by either a rise or fall in serum E2 levels. l In patients with mechanical infertility and spontaneous ovulation who are induced with hMG-hCG to superovulate for in vitro fertilization (IVF), a specific correlation was found between periovulatory E2 pattern and subsequent conception rates. 2- 4 In these cyReceived July 15, 1986; revised and accepted October 31, 1986. *Reprint requests: Daniel Navot; M.D., The Howard and Georgeanna Jones Institute for Reproductive Medicine, Hofheimer Hall, 825 Fairfax Avenue, 6th Floor, Norfolk, Virginia 23507. 234

Navot et al. E2 in conceptional cycles

cles it was shown that preovulatory serum E2 levels, which increase after hCG administration, were associated with the highest pregnancy rates. 2 -4 However, there is no available data that correlates peri ovulatory serum E2 secretory patterns with eventual conceptions in anovulatory infertile patients undergoing hMG-hCG ovulation induction. The primary aim of this work was to assess prospectively the periovulatory E2 patterns and their value in predicting conceptions after hMGhCG induction of ovulation in anovulatory infertility. MATERIALS AND METHODS

Forty anovulatory infertile patients were enrolled in the study. All the patients had anovulatory cycles with evidence of endogenous estrogen Fertility and Sterility

production and normal follicle-stimulating hormone (FSH) levels in accordance with group II of the World Health Organization definition. 5 Before hMG-hCG induction of ovulation, other caus~ es of infertility were excluded. Mean age of the patients was 32 years (range, 21 to 44 years), and mean duration of infertility 5.6 years (range, 2 to 19 years). Ovulation was induced with Pergonal (Ikapharm, Ramat Gan, Israel, 75 IU FSH and 75 IU LH), according to an individually adjusted treatment schedule. 6 Monitoring system relied, in addition to clinical parameters, on serial serum E2 measurements (Direct Radioimmunoassay [RIA], Isodan, Jerusalem, Israel) and ultrasonographic estimation of follicular size and number (Aloka Sector Scanner, Mure, Nitaki-shi, Tokyo, Japan). Target values for hCG administration were (1) serum E2 plateauing within an "E2 window" of 450 to 1400 pg/ml, and (2) one or two preovulatory follicles ~ 18 mm in largest diameter. Ideally both criteria should have been met; however, during actual clinical decision making no single criterion was rigidly held. The patients were divided according to their response to hMG-hCG administration into two groups of E2 patterns: Group A was defined by an increase (or plateau) of E2 on the day after hCG administration (day + 1). Any E2 value on day + 1 that was similar to (within 10% deviation) or higher than the E2 level on the previous day (day 0) was considered within that group. Group B was defined by a decrease of E2 levels of more than 20% (greater than twice the coefficient of variation of the RIA of E 2) by day + 1. Each of these groups was further subdivided into high responders (E 2 > 500 pg/ml on day 0) designated Al and B 1, respectively, and low responders (E 2 < 500 pg/ml on day 0) designated A2 and B 2, respectively. Ovulation was confirmed by a single midluteal serum progesterone (P) level higher than 10 ng/ml. (Solid phase 1251 RIA, Diagnostic Prod-

ucts Corporation, Los Angeles, CA.) Statistical analysis was performed with the BMDP statistical software (BMDP, Los Angeles, CA). For prediction of chances for conception in each cycle, stepwise logistic regression was performed on multiple parameters: the four patterns of E2 response, serum E2 level on day 0, number and size of preovulatory follicles on day 0, and midluteal serum P estimations.

RESULTS In the 40 anovulatory patients 52 cycles were prospectively evaluated. Seventeen of the 52 cycles were conceptional (33%), whereas 35 of the cycles were infertile (67%). The relative number of cycles for each type ofE 2 response is illustrated in Table 1. An Al pattern was observed in 42% of cycles. A similar proportion had a Bl pattern (40%), 14% responded in a B2 pattern, and only 4% had an A2 type of response. The relative distribution of conceptional cycles, according to E2 pattern, is shown as well in Table 1. There is a significantly higher rate of fertile cycles (P < 0.05) in patients with an Al response, compared with either Bl or B2 (A 2 is too small to permit statistical analysis). Seventy-one percent of conceptional cycles had an Al type of response (12 of 17). For assessment of the value of ultrasonography in monitoring timing of hCG administration, preovulatory follicles were classified into three groups: follicles ~ 14 mm in largest diameter, follicles measuring 15 to 17 mm, and follicles that were ~ 18 mm. No pregnancies occurred in patients with follicles < 15 mm on day 0 (0/4). The ratio of fertile cycles in the group with intermediate-size follicles was 43% (6/14), whereas in cycles with follicles ~ 18 mm 32% were fertile (11134). The level of serum E2 was compared between conceptional and nonconceptional cycles on day 0;

Table 1. Distribution of Cycles According to Specific E2 Response Periovulatory E2 response

Proportion of specific E2 response (%)

Ai (high-ascending) BI (high--descending) A2 (low-ascending) B2 (low-descending)

22/52 21152 2/52 7/52

(42) (40) (4) (14)

Total

52/52 (100)

Proportion of conceptional cycles (%)

12/22 (55) 3/21 (14) 112 117 (14)

pc'value

< 0.05 Al vs BI Al vs BI + B2 + A2

< 0.05 < 0.05

17/52 (33)

aFischer's exact test. Vol. 47, No.2, February 1987

Navot et al. E2 in conceptional cycles

235

no statistically significant difference was found (Table 2). Midluteal serum P (day + 8) was similar in conceptional (22.9 ± 3 [standard error of the. mean] ng/ml) and non conceptional cycles (21.6 ± 4.65 ng/mD (Table 2). Stepwise logistic regression 7 was applied to the previously mentioned parameters to evaluate the value of each in the prediction of conception in a given cycle. The only parameter that could predict eventual conception was the grouping according to E2 response (Table 2). An A1 pattern predicted conception in 51% of cycles, whereas B1 and B2 patterns could predict conception in 15%. The size and number of preovulatory follicles, level of E2 on day 0, and midluteal P had no predictive value by themselves and did not improve the prediction given by the pattern of peri ovulatory E2 alone.

DISCUSSION

Conception rates after gonadotropin induction of ovulation in anovulatory patients, classified as group II according to the World Health Organization may be as low as 14%.8 Although these patients are closely monitored by serum E 2, usually in conjunction with ultrasonography, these two parameters in themselves proved to be of limited value in predicting conceptions. 9 This study has prospectively identified a pattern of E2 that is associated with the highest conception rate per

Table 2. Value of Various Parameters for Prediction of Conceptional and Nonconceptional Cyclesa Parameter

Fertile

Nonfertile

Predictive value %

Periovulatory E2 response Al 12 A2 1 BI 3 B2 1 E2 on day 0 741 ± 79.3 (pg/ml) Follicular size (mm) ;3 18 11 15-17 6 "" 14 P on day +8 22.9 ± 3 (ng/ml)

10

51

1 18 6 793 ± 52.5

15 15 None

34 14 4 21.6 ± 4.65

aStepwise logistic regression. 236

Navot et al. E2 in conceptional cycles

None None None None

cycle (55%). In addition, about 70% of all conceptions occurring in this series show that same pattern. It was previously shown 2-4 in patients undergoing IVF that the highest conception rates after embryo transfer were obtained in those having an A 1 type of E2 response. The concept of menotropin treatment in ovulatory patients undergoing IVF and anovulatory patients is different: in IVF patients the objective is to mature numerous follicles that will yield a high number offertilizable oocytes. In anovulatory patients the clinician strives for a close reproduction of the natural cycle, taking precautions to avoid ovarian hyperstimulation and high multiple pregnancies. In spite of these differences in the strategies to ovarian stimulation, the pattern of E2 response that is conducive to a conception is common to both groups of patients. In the natural cycle the maturation of the ovarian follicular elements is synchronous with the development of an endometrium favorable for embryo implantation. In contrast, it seems that in cycles in which ovulation is pharmacologically induced, the development of multiple follicles is associated with significantly higher levels of P in the early luteal phase 10, 11 and enhanced maturation of the endometrium. 11 These events may lead in turn to the arrival of embryos into a uterus whose endometrium is too advanced for optimal implantation. It may therefore be that an A1 pattern with high and sustained E2 may directly overcome this potential asynchrony between the developmental stage of the embryo and the actual maturity of the endometrium by counteracting this high postovulatory progestational milieu of stimulated cycles. Alternatively it was suggested that the LH content of hMG may induce premature luteinization or atresia in menotropin-stimulated cycles. 12 It may be therefore that successful cycles (A 1 pattern) are those in which the ovulatory dose of hCG is administered relatively early, at a stage in which the theca cells respond to hCG by increased androgen production while the granulosa is still capable of enhanced aromatization to estrogens. An A1 pattern may therefore be an indirect index of optimal timing of hCG administration. Our observation that no conception occurs with follicles < 15 mm corroborates the report of Seibel et aI.,13 who have had no conceptions with follicles < 16 mm. The rate of fertile cycles with follicles intermediate in size (15 to 17 mm) and large follicles (;::: 18 mm) did not differ significantly. Fertility and Sterility

REFERENCES

I

jl

1. Bergquist L, Nillius SJ, Wide L: Human gonadotropin therapy. II. Serum estradiol and progesterone patterns during nonconceptual cycles. Fertil Steril 39:766, 1983. 2. Laufer N, DeCherney AH, Tarlatzis BC, Naftolin F: The association between preovulatory serum 17J3-estradiol pattern and conception in human menopausal gonadotropin-human chorionic gonadotropin stimulation. Fertil Steril 46:73,1986 3. Jones HW Jr, Acosta A, Andrews MC, Garcia JE, Jones GS, Mantzavinos T, McDowell J, Sandow B, Veeck L, Whibley T, Wiles C, Wright G: The importance of the follicular phase to success and failure in in vitro fertilization. Fertil Steril40:317, 1983 4. Lopata A: Concepts in human in vitro fertilization and embryo transfer. Fertil Steril 40:289, 1983 5. WHO Scientific Group: Agents stimulating gonadal function in the human. World Health Organization Technical Report Series No. 514, Geneva, 1973, p 7 6. Brown JB: Gonadotropins. In Infertility: Male and Female, Edited by V Insler, B Lunenfeld. Singapore, Churchill Livingstone, 1986, p 371

Vol. 47, No.2, February 1987

7. Engleman L: Stepwise Logistic Regression in BMDP Statistical Software, Edited by WJ Dixon. Berkeley, University of California Press, 1983, p 330 8. Lunenfeld B, Insler V: Gonadotropins. In Diagnosis and Treatment of Functional Infertility, Edited by B Lunenfeld, V Insler. Berlin, Grosse Verlag, 1978, p 85 9. Brown JB: Gonadotropins. In Infertility: Male and Female, Edited by V Insler, B Lunenfeld. Singapore, Churchill Livingstone, 1986, p 379 10. Laufer N, Navot D, Schenker JG: The pattern of luteal phase plasma progesterone and estradiol in fertile cycles. Am J Obstet Gynecol 143:808, 1983 11. Bergquist C, Nillius SJ, Wide L: Human gonadotropin therapy. I. Serum estradiol and progesterone patterns during conceptual cycles. Fertil Steril 39:761, 1983 12. Jones GS: Update on in vitro fertilization. Endocr Rev 5:62,1984 13. Seibel MM, McArdle CR, Thompson EE, Berger MJ, Taymor ML: The role of ultrasound in ovulation induction: a critical appraisal. Fertil Steril 36:573, 1981

Navot et al. E2 in conceptional cycles

237