FERTILITY AND STERILITY Copyright
~
1996 American Society for Reproductive Medicine
Vol. 65, No.1, January 1996
Printed on acid-free paper in U. S. A.
The efficacy of immunotherapy in patients who underwent superovulation with intrauterine insemination
Chung Roon Kim, M.D.* Yoon Kyung Cho, M.D. Jung Eun Mok, M.D. Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, College of Medicine, University of Ulsan, Asan Medical Center, Seoul, Korea
Objective: To determine if the immunotherapy with corticosteroids would improve pregnancy rate in infertile patients with ovulatory factor and patients with unexplained infertility who undergo superovulation with lUI. Design: Prospective, controlled study. Setting: Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, University of Ulsan, Seoul, Korea. Patients: Ninety-one infertile patients with ovulatory factors and 78 patients with unexplained infertility who underwent superovulation with lUI. Results: The prevalence of autoantibodies (antinuclear antibody, lupus anticoagulant, anticardiolipin antibody, antidouble-stranded DNA antibody) was elevated significantly in patients with unexplained infertility compared with patients with ovulatory infertility (20.5% versus 3.3%). Forty-five patients with ovulatory factor underwent 72 cycles of superovulation with lUI, receiving corticosteroids. Thirty-eight patients with unexplained infertility underwent 75 cycles of superovulation with lUI, receiving corticosteroids. Forty-six patients with ovulatory factor who underwent 66 cycles of superovulation with lUI and 40 patients with unexplained infertility who underwent 75 cycles of superovulation with lUI served as controls, not receiving corticosteroids. In patients with ovulatory factor, there were no significant differences between the corticosteroid treatment group and control group in clinical pregnancy rate (38.9% versus 33.3%) or in spontaneous abortion rate (14.3% versus 13.6%). In patients with unexplained infertility, there was a significantly higher clinical pregnancy rate per cycle in the corticosteroid treatment group, with 45.3% (34/75) compared with 29.3% (22175) in the control group but no difference between the corticosteroid treatment and control groups in spontaneous abortion rate (17.6% versus 13.6%). There were no side effects due to the use of corticosteroids. Conclusion: These results demonstrate that immunotherapy with corticosteroids could improve pregnancy rate in patients with unexplained infertility who undergo superovulation with IUI. Fertil SteriI1996;65:133-8 Key Words: Immunotherapy, ovulatory factor, unexplained infertility, superovulation, lUI, autoantibodies
The peri-implantation period appears to be a time of significant pregnancy wastage. Several factors present at the preimplantation stage playa cooperative role in the outcome of pregnancy. Embryo-asso-
Received February 27, 1995; revised and accepted July 26, 1995. * Reprint requests: Chung Hoon Kim, M.D., Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, University of Ulsan, Asan Medical Center, 3881, Poongnap-Dong, Songpa-Gu, Seoul, 138-040, Korea
ciated immunosuppressor factor (1-3), platelet-activating factor (4, 5), 24K protein (6), D9B1 (7), and placenta protein 14 (PP14) also known as 2-pregnancy-associated endometrial globulin (8-10) have been proposed to be factors related to implantation. Among factors related to implantation, embryo-associated immunosuppressor factor and PP14 appears to play an immunomodulating role in pregnancy (13, 9, 10). These factors could facilitate implantation by modulating the maternal response to the fetal allograft. The concept that immune responsiveness
Vol. 65, No.1, January 1996
Kim et al. Immunotherapy in superovulation with lUI
133
in pregnancy generally is decreased arose from early studies that appear to show that lymphocyte receptivity, graft rejection, and the antibody response to certain antigens were reduced during pregnancy (11). Recently, it was elucidated that the fetal-placental semiallograft is afforded protection by local immunomodulating factors. Moreover, it has been demonstrated that methylprednisolone supports implantation of embryos (12) (Polak de Fried E, Blanco L, Lancuba S, Asch RH, abstract). We expect that immunotherapy with corticosteroids can improve pregnancy rate in infertile patients who undergo superovulation with lUI and may be more effective in infertile patients with etiologic factors related with abnormal immune function. The present study was designed to determine if the immunotherapy with corticosteroids would improve pregnancy rate in patients with ovulatory infertility or unexplained infertility who undergo superovulation with lUI.
From March 1994 to December 1994, 91 infertile patients with ovulatory factor only and 78 patients with unexplained infertility who had begun superovulation with lUI in our unit were allocated randomly to the corticosteroid treatment group and the control group by a permuted block design. The patients with ovulatory infertility were resistant to clomiphene citrate (CC) (n = 15) or previously had failed to conceive despite three ovulatory cycles using CC (n = 76). Of patients with ovulatory infertility, patients with polycystic ovary syndrome (PCOS) or hyperandrogenism were excluded in this study. Polycystic ovary syndrome was defined as the typical ultrasound findings in the ovary of ~ 10 follicles < 9 mm and a hyperechogenic central stroma accompanied by oligo amenorrhea and/or hirsutism. There was no evidence of other factors in infertility in any patients with ovulatory infertility. Unexplained infertility was defined as primary or secondary infertility of ~2 years' duration, when all of the following criteria were met: [1] normal endocrine profile (PRL, TSH, LH, FSH, T, DHEAS, and midluteal P); [2] regular menstrual cycle and biphasic BBT with ~12 days duration of the luteal phase; [3] normal hysterosalpingography and laparoscopy within the past 2 years; [4] positive postcoital test; [5] normal findings on semen analysis according to World Health Organization (WHO) guidelines (13) and on sperm morphology evaluation using strict criteria by the authors (14). Ultrashort protocol of gonadotropin-releasing hormone agonist (GnRH-a) was used for superovulation in all patients. Blood samples were drawn for sero-
logic tests with respect to antinuclear antibody (ANA), lupus anticoagulant, anticardiolipin antibody, antidouble-stranded DNA (anti-dsDNA) antibody, and antiribonucleoprotein (anti-RNP) antibody in all patients on the day of commencement of superovulation. Patients were administered the GnRH-a, 0.1 mg/d triptorelin (D-Trp-6-LHRH, Decapeptyl; Ferring, Malmo, Sweden), days 2, 3, and 4 of the menstrual cycle. Human menopausal gonadotropin (Humegon; Organon, Oss, The Netherlands) and/or human FSH (Metrodin; Serono, Geneva, Switzerland) were started on the 3rd day of the menstrual cycle. Serial scans were performed using a Combison 320 US machine (Kretztechnik, Zipf, Austria) for monitoring of follicular growth and exogenous gonadotropin dosage was adjusted from day 6 of the menstrual cycle. Human chorionic gonadotropin (10,000 IU 1M, Pregnyl; Organon) was given to induce follicular maturation when a consistent rise in plasma E2 was identified and associated with one or more follicles sized ~ 17 mm in diameter. A single lUI procedure was performed approximately 36 to 40 hours after hCG administration. Husbands delivered a fresh semen sample by masturbation after approximately 3 days of sexual abstinence. The samples, collected into sterile plastic containers, were liquefied at room temperature for 30 minutes. Three-step Percoll gradient method was used for sperm preparation. A Percoll density gradient column was prepared using 90%, 70%, and 50% Percoll (Sigma Chemical Company, St. Louis, MO) solution plus insemination medium, Ham's F-10 (GIBCO, Staten Island, NY), supplemented with 3% human serum albumin, penicillin, and streptomycin. The neat ejaculate was layered carefully on top of Percoll gradient and then centrifuged at 600 X g for 15 minutes. After centrifugation, the Percoll layer was removed and the resultant sperm pellet was resuspended to 0.4 mL insemination medium. • The sperm suspension was drawn into an insemination catheter (Makler cannula; Sefi Medical, Haifa, Israel). Using a vaginal speculum, the catheter was passed through the uterine cervix and then the sample was deposited into the uterine cavity. The patient remained supine for approximately 20 minutes. Immunotherapy in the treatment group was performed through the administration of prednisolone (Delta-Cortef; Upjohn Co., Seoul, Korea). Prednisolone was administered in daily oral dosage of 10 mg from the 2nd day of menstrual cycle till the day before lUI and the daily dosage was increased to 60 mg (30 mg twice daily) from the evening oflUI and for 4 days afterward (Fig. 1). Luteal support was provided by administering 50 mg of P in oil (Progest, Samil Pharmacy, Seoul, Korea) daily from the day after lUI. Pregnancies were
Kim et aI. Immunotherapy in superovulation with lUI
Fertility and Sterility
MATERIALS AND METHODS
134
Day of menstrual cycle 2
3
4
I
I
I
+t t
+ +
IGnRH-al
I
hCG
lUI
FSH I hMG
0101000100001000000 00000
00000 000 000 000
o
Prednisolone 10mg
Figure 1 Schematic representation of the immunotherapy protocol in superovulation with lUI cycle. Ultrashort protocol of GnRH-a was used for superovulation.
confirmed with rising serum ,B-hCG levels and transvaginal ultrasound examination confirming the presence of an intrauterine gestational sac. Biochemical pregnancies were excluded from analysis. Statistical analysis was performed using Student's t-test, X2 test with Yates' continuity correction, or Fisher's exact test as appropriate. Statistical significance was defined as P < 0.05. RESULTS
Forty-five patients with ovulatory factor underwent 72 cycles of superovulation with lUI, receiving corticosteroids. Thirty-eight patients with unexplained infertility underwent 75 cycles of superovulation with lUI, receiving corticosteroids. Forty-six patients with ovulatory factor who underwent 66 cycles of superovulation with lUI and 40 patients with unexplained infertility who underwent 75 cycles of superovulation with lUI served as controls, not receiving corticosteroids. Table 1 shows that the corticosteroid treatment group and control group in each group according to the etiologic factors were similar with respect to clinical, endocrine features, and male partner sperm parameters. Serologic findings with respect to ANA, lupus anticoagulant, anticardiolipin antibody, and anti-dsDNA antibody are shown in Table 1. The prevalence of these antibodies was elevated significantly in patients with unexplained infertility (16/78; 20.5%) compared with patients with ovulatory infertility (3/91; 3.3%) (P < 0.01 for total antibodies). There were no patients who were positive for two or more kinds of antibodies. None of the patients had anti-RNP antibodies. Vol. 65, No.1, January 1996
In the patient groups with ovulatory infertility, there were no significant differences between treatment group and control group in amount of gonadotropin(s) required, days of gonadotropin(s) administration, serum E2 levels on the day of hCG administration, the number of mature (2::14 mm) follicles, or endometrial thickness (Table 2). In the patient groups with unexplained infertility, there were also no significant differences between treatment group and control group in clinical response (Table 2). In patient groups with ovulatory infertility, it seemed that a clinical pregnancy rate per cycle was higher in the treatment group, with 38.9% (28/72) compared with 33.3% (22/66) in the control group, but this difference was not statistically significant. There were no significant differences between the treatment group and control group in spontaneous abortion rate (14.3% versus 13.6%) or in multiple pregnancy rate (17.9% versus 13.6%) (Table 3). In patient groups with unexplained infertility, there was a significantly higher clinical pregnancy rate per cycle in the treatment group, with 45.3% (34/75) compared with 29.3% (22/75) in the control group (P < 0.05), but no differences between the treatment group and the control group in spontaneous abortion rate (17.6% versus 13.6%) or in multiple pregnancy rate (17.6% versus 9.1%) (Table 3). The pregnancy outcome in patients with positive antibodies is shown in Table 4. In the unexplained infertility group with positive antibodies, clinical pregnancy rate per cycle was higher in the treatment group than that in the control group (46.2% versus 13.3%) but the difference was not statistically significant. Of 16 unexplained infertility patients who were positive for antibodies, 2 with anticardiolipin antibodies in the treatment group became negative in the next treatment cycles but none in the control group became negative. There were no side effects due to the use of corticosteroids. DISCUSSION
Implantation of embryos in the wall of the uterus is the result of a complex series of interactive steps beginning with fixation of the blastocyst in the uterus and ending with formation of a definite placenta. Several factors present at the preimplantation stage playa cooperative role in the outcome of pregnancy. Embryo-associated immunosuppressor factor has been shown to be secreted by pre-embryos, and correlation between successful pregnancy and embryo-associated immunosuppressor factor activity has been shown in embryo growth medium (1). Bose (2) reported that embryo-associated immunosuppressor factor affects T cell and B cell function Kim et aI. Immunotherapy in superovulation with lUI
135
i
Table 1 Patient Characteristics Ovulatory infertility
No. of patients Age of patients (y) Age of husbands (y) Infertility duration (y) Patients with primary infertility (%) Semen Volume (mL) Concentration (106/mL) Motility (%) Sperm morphology evaluation using strict criteria Normal morphology (%) Morphology index (%) Endocrine profile Basal FSH (mID/mL) Basal LH (mID/mL) T (ng/mL) PRL (ng/mL) No. of patients with positive autoantibodies Antinuclear antibody Lupus anticoagulant Anticardiolipin antibody:\: Anti-dsDNA antibody Total
Unexplained infertility
Treatment
Control
Treatment
Control
45 31.0 ± 2.9* 34.2 ± 2.8 2.4 ± 1.1 63.6
46 31.5 ± 2.7 33.9 ± 2.9 2.6 ± 1.3 56.5
38 32.1 ± 3.0 34.3 ± 3.1 2.9 ± 1.3 57.6
40 32.8 ± 3.3 35.0 ± 3.6 2.8 ± 1.2 60.0
2.6 ± 0.4 89.2 ± 31.3 59.2 ± 13.2
2.8 ± 0.4 80.3 ± 27.1 57.3 ± 12.5
2.8 ± 0.4 78.1 ± 35.7 56.1 ± 14.2
2.6 ± 0.3 85.1 ± 34.6 58.4 ± 13.8
6.8 ± 15.0 ±
2.5 6.5
7.4 ± 2.7 15.3 ± 6.6
6.9 ± 2.8 14.7 ± 6.7
7.3 ± 2.9 15.5 ± 7.2
7.4 6.7 0.4 12.2
1.9 1.8 0.1 3.5
7.1 6.6 0.3 13.1
± ± ± ±
o
± ± ± ±
2(4.3)t
o
o
o
1(2.2)
o o
1(2.2)
2(4.3)
* Values are means ± SD. t Values in parentheses are percentages.
1.6 1.5 0.1 3.9
7.1 6.9 0.3 12.7
± ± ± ±
2.0 1.9 0.1 5.0
3(7.9) 1(2.6) 4(10.5) 0 8(21.0)
6.9 6.7 0.4 11.9
± ± ± ±
1.6 1.8 0.1 4.1
3(7.5) 1(2.5) 3(7.5) 1(2.5) 8(20.0)
:\: All immunoglobulin G.
directly and this active process may be one of several responses by which the maternal immune response against the fetus is prevented. Placental protein 14 isolated from the human placenta was shown to be predominantly synthesized and secreted by secretory and decidualized endometrium (15). A hypothesis that PP14 plays an immunomodulating role in pregnancy, has been proposed (9, 10), on the basis ofthe observation that PP14 suppresses the proliferation of lymphocytes in the mixed lymphocyte reaction. The concept that immune responsiveness in pregnancy generally is decreased arose from studies that appear to show that lymphocyte receptivity, graft rejection, and the antibody response to certain anti-
gens were reduced during pregnancy (11). Recently, it was elucidated that the fetal-placental semiallograft is afforded protection by local immunomodulating factors. Moreover, it has been demonstrated by Cohen et al. (12) that methylprednisolone supports implantation of embryos with small holes in their zona. A prospective study by Polak de Fried et al. (Polak de Fried E, Blanco L, Lancuba S, Asch RH, abstract) suggested that immunosuppressive doses of corticosteroids administered for a short period of time could significantly improve implantation rate, pregnancy rate, and delivery rate of patients undergoing IVF-ET. /. There are several mechanisms that can explain the efficacy of immunotherapy. Cortisol can stimu-
Table 2 Comparison of Clinical Response Between Treatment and Control Groups Ovulatory infertility Treatment No. of patients No. of cycles 75 IU ampules of gonadotropins Duration of gonadotropin administration (d) Response on the day of hCG administration E2 (pg/mL)t No. of follicles ;",14 mm Endometrium (mm)
45 72 23.7 ± 7.2 ±
5.6* 1.6
1,318.1 ± 414.3 8.0 ± 3.6 10.7 ± 3.5
* Values are means ± SD. 136
Kim et al. Immunotherapy in superovulation with lUI
Unexplained infertility
Control 46 66 24.4 ± 7.4 ±
Treatment
5.4 1.8
1,398.3 ± 395.2 8.3 ± 3.7 3.8 11.3 ±
38 75 24.3 ± 7.0 ±
5.5 1.5
1,488.1 ± 499.3 8.4 ± 3.7 4.0 11.5 ±
Control 40 75 25.0 ± 7.3 ±
5.8 1.7
1,378.3 ± 425.2 8.1 ± 3.6 3.7 11.1 ±
t Conversion factor to SI unit, 3.671.
Fertility and Sterility
late the secretion of E2 and P by human granulosa cells (16). It can be postulated that, either directly or by means of P production, glucocorticoids may cause adequate secretion of immunosuppressive factor(s) from the endometrium, thus supporting implantation. In addition, corticosteroids would reverse an endometrial inflammatory process that can be induced by ET or lUI procedures. Therefore, immunotherapy with corticosteroids may improve the pregnancy rate in all infertile patients undergoing assisted reproductive technologies. In the present study, it seemed that the pregnancy rate in immunosuppressed patients with ovulatory infertility was higher than in patients included in control group, but this difference was not statistically significant. However, a significant difference may be represented if the sample size in immunotherapy and control groups is enlarged. Therefore, larger series will be required to determine how pregnancy rate in immunotherapy group compares with that in control group according to the etiologic factors of infertile patients. Patients with unexplained infertility can demonstrate significant abnormalities in immune function. Gleicher et al. (17) reported total immunoglobulin abnormalities and autoantibody abnormalities inclusive of phospholipid antibodies in patients with unexplained infertility. Similar results also were reported by Taylor et al. (18). These authors reported significantly elevated autoantibody levels to smooth muscle, phospholipids, as well as nuclear antigens. In our study, ANA, lupus anticoagulant, anticardiolipin antibody, anti-dsDNA antibody, and anti-RNP antibody were assayed in all patients. The prevalence of these antibodies was elevated significantly in patients with unexplained infertility (20.5%) compared with patients with ovulatory infertility (3.3%). These results suggest that a high proportion of paTable 3 Comparison of Pregnancy Outcome Between Treatment and Control Groups Ovulatory infertility Treatment No. of patients No. of cycles No. of clinical pregnancies* Abortiont Ectopict Ongoingt Multiple pregnanciest Twin Triplet
Unexplained infertility
Control
Treatment
45 72
46 66
38 75
40 75
28 (38.9) 4 (14.3) 0 24 (85.7)
22 (33.3) 3 (13.6) 1 (4.5) 18 (81.8)
34 (45.3)t 6 (17.6) 0 28 (82.4)
22 (29.3) 3 (13.6) 0 19 (86.4)
3 (13.6) 2 0
6 (17.6) 5 1
2 (9.1) 2 0
5 (17.9) 4 1
* Values in parentheses are percents per cycle.
t p < 0.05 versus control in unexplained infertility group. t Values in parentheses are percents per clinical pregnancy.
Vol. 65, No.1, January 1996
Control
Table 4 Comparison of Pregnancy Outcome Between Treatment and Control Groups in Patients With Positive Antibodies Ovulatory infertility Treatment No. of patients No. of cycles No. of clinical pregnancies* Abortiont Ectopic Ongoingt Multiple pregnanciest
Control
Unexplained infertility Treatment
Control
1 2
2 3
1 (50.0) 1 (100.0) 0 0
1 (33.3) 0 0 1 (100.0)
6 (46.2) 2 (33.3) 0 4 (66.7)
2 (13.3) 1 (50.0) 0 1 (50.0)
0
0
2 (33.3)
0
8 13
8 15
* Values in parentheses are percents per cycle. t Values in parentheses are percents per clinical pregnancy.
tients with unexplained infertility is associated with abnormalities in immune function. In the present study, pregnancy rate in immunosuppressed patients with unexplained infertility was significantly higher than in those patients who did not receive exogenous corticosteroids. In the unexplained infertility group with positive antibodies, clinical pregnancy rate per cycle was higher in the treatment group than in the control group and this difference in pregnancy rate between two groups was distinct (46.2% versus 13.3%). However, the difference was not statistically significant because sample size was small. Moreover, two patients with positive antibodies (25%) in the treatment group of unexplained infertility became negative in the following treatment cycles but none in the control group became negative. There still are other mechanisms that can explain our results. Corticosteroids can modulate the abnormal immune function such as antibodies, thereby allowing the embryo to develop normally. The present study demonstrated that immunotherapy could improve pregnancy rate in infertile patients who undergo superovulation with lUI, especially in patients with etiologic factors related with abnormal immune function. It is necessary to go further with largescale prospective clinical trials in infertile patients who undergo assisted reproductive technologies, to evaluate the efficacy of immunotherapy according to the etiologic factors. REFERENCES 1. Bose R, Mahadevan MM. Immunosuppressive activity in human embryo growth media is associated with successful pregnancy: effect of gonadotropin releasing hormone agonist (GnRHa) treatment of patients undergoing in vitro fertilization and embryo transfer (IVF-ET). J Clin Immunol 1990; 10:175-81.
Kim et al. Immunotherapy in superovulation with lUI
137
2. Bose R. Properties of human pre- and post-implantation embryo-associated immunosuppressor factor(s). Immunol Lett 1991;27:119-25. 3. Bose R, Mahadevan MM. Embryo associated immunosuppressor factor(sl secreted by preembryo and serum estradiol levels are predictive of pregnancy outcome: effect of gonadotropin releasing hormone agonist (GnRHa) treatment of patients undergoing in vitro fertilization and embryo transfer (IVF-ET). Immunol Invest 1992;21:25-38. 4. Collier M, O'Neil C, Ammit AJ, Saunders DM. Biochemical and pharmacological characterization of human embryo-derived platelet activating factor. Hum Reprod 1988;3:993-8. 5. Harper MJK. Platelet-activating factor: a paracrine factor in preimplantation stages of reproduction? BioI Reprod 1989; 40:907-13. 6. Manners CV. Endometrial assessment in a group of infertile women on stimulated cycles for IVF: immunohistochemical findings. Hum Reprod 1990;5:128-32. 7. SeifM, Aplin JD, Buckley CH. Luteal phase defect: the possibility of an immunohistochemical diagnosis. Fertil Steril 1989;51:273-9. 8. Bohn H, Krans W, Winckler W. New soluble placental tissue proteins: their isolation, characterization, localization and quantification. Placenta Suppl 1982;4:67-81. 9. Bolton AE, Pockeley AG, Clough KJ, Mowles EA, Stocker RJ, Westwood OMR, et al. Identification of placental protein 14 as an immunosuppressive factor in human reproduction. Lancet 1987; 1:593-5. 10. Pockeley AG, Mowles EA, Stocker RJ, Westwood OMR, Chap-
138
Kim et a1. Immunotherapy in superovulation with lUI
11. 12.
13.
14.
15.
16.
17.
18.
man MG, Bolton AE. Suppression of in vitro lymphocyte reactivity of phytohaemagglutinin by placental protein 14. J Reprod Immunol1988;13:31-9. Gill TS III, Repetti CF. Immunological and genetic factors influencing reproduction. Am J Pathol 1979;95:463-570. Cohen J, Malter H, Elsner C, Kort H, Massey J, Mayer MP. Immunosuppression supports implantation of zona pellucida dissected human embryos. Fertil Steril 1990;53:662-5. World Health Organization. Laboratory manual for the examination of human semen and semen-cervical mucus interaction. 3rd ed. New York: Cambridge University Press, 1993:1-107. Kim CH, Moon SY, Chang YS. Comparison between sperm morphology evaluation using strict criteria and sperm penetration assay as prognostic factors in in vitro fertilization. Korean J Obstet Gynecol 1994;37:1772-91. Julkunen M, Koistinen R, Sjoberg J, Rutanen EM, Wahlstrom T, Seppala M. Secretory endometrium synthesizes placental protein 14. Endocrinology 1986; 118:1782-6. Ben-Rafael Z, Benadiva CA, Garcia CJ, Flickinger GL. Cortisol stimulation of estradiol and progesterone secretion by human granulosa cells is independent of follicle-stimulating hormone effects. Fertil Steril 1988;49:813-6. Gleicher N, EI-Roeiy A, Confino E, Friberg J. Reproductive failure because of autoantibodies: unexplained infertility and pregnancy wastage. Am J Obstet Gynecol 1989; 160: 137680. Taylor PV, Campbell JM, Scott JS. Presence of autoantibodies with unexplained infertility. Am J Obstet Gynecol 1989; 161:377 -9.
Fertility and Sterility