Intrauterine insemination as treatment for antisperm antibodies in the female

Transcription

1 FERTILITY AND STERILITY Copyright<> 1988 The American Fertility Society Printed in U.S.A. Intrauterine insemination as treatment for antisperm antibodies in the female Ehud J. Margalloth, M.D. Elizabeth Sauter, M.D. Richard A. Bronson, M.D.* David L. Rosenfeld, M.D. Gerald M. Scholl, M.D. George W. Cooper, Ph.D. Department of Obstetrics and Gynecology, North Shore University Hospital, Cornell University Medical College, Manhasset, New York Ninty-one women with long-standing infertility in the presence of humoral antisperm antibodies (ASA) underwent 4 73 cycles of intrauterine insemination of washed sperm (lui), resulting in 26 pregnancies. Nine pregnancies were achieved in 67 women who underwent 285 lui during unstimulated cycles (13% pregnancy rate; 3% pregnancy/cycles of treatment). Twenty women underwent 86 lui after clomiphene citrate (CC) stimulation, resulting in 6 pregnancies (30%; 7% per cycle), while 28 underwent 102 cycles of lui after human menopausal gonadotropins (hmg) stimulation, resulting in 11 pregnancies (39%; 11% per cycle). Twenty-one of the 26 pregnancies (81%) were achieved in the first 2 lui cycles. During unstimulated and CC lui cycles, all pregnancies occurred in the first two cycles of treatment; while with hmg stimulation pregnancies also occurred in the third, fourth, and fifth lui cycles. Nine of these 91 women subsequently conceived spontaneously, and three others conceived through in vitro fertilization. Only 6 of the 38 pregnancies resulted in spontaneous abortion (16%). Thus, pregnancies achieved in women with ASA have no increased risk of abortion. Fertil Steril50:441, 1988 Cervical factor infertility is responsible for failure of conception in roughly 10% of infertile couples. An immunity to spermatozoa can be demonstrated in either partner in approximately one third of these cases. 1 Mechanisms by which antisperm antibodies (ASA) in women might interfere in vivo with reproduction include: (1) complement mediated sperm cytotoxicity, 2 (2) restricted motion of sp~rmatozoa within cervical mucus, 3 (3) interference with capacitation or the acrosome reaction,4 (4) inhibition of sperm attachment or penetration of egg membranes, 5 6 (5) interference with embryo implantation, 7 and (6) enhancement of phagoctosis of sperm in the genital tract by macrophages.8 The treatment of infertility caused by immunity Received February 2, 1988; revised and accepted May 10, * Reprint requests: Richard A. Bronson, M.D., Department of Obstetrics and Gynecology, Division of Human Reproduction, North Shore University Hospital, 300 Community Drive, Manhasset, New York to sperm remains controversial. While the use of condoms to prevent exposure to sperm-associated antigens was initialy described, 9 10 some found it ineffectual.11 The use of corticosteroids has been advocated, but results are variable and side effects are not uncommon. 10 ' 12 ' 13 In vitro fertilization has been successful, but this approach is both physically and emotionally intense and expensive. We have recommended the use of intrauterine insemination of washed sperm (lui) under gonadotropin stimulation in hormonally and sonographically monitored cycles. 1 The aim of this study is to present the results of 91 women who had undergone intrauterine inseminations as a method of treatment for immunity to sperm. A comparison was made of conception rates and the outcome of pregnancy after lui in unstimulated cycles versus those after ovarian stimulation with either clomiphene citrate (CC) or human menopausal gonadotropins (hmg). MATERIALS AND METHODS Couples presenting to the Division of Human Reproduction, North Shore University Hospital, Margalloth et al. lui for female antisperm antibodies 441

2 with a complaint of delay in conception, in association with abnormal sperm penetration of or survival within cervical mucus were screened for ASA. One hundred ten women in whom ASA were documented underwent lui. Nineteen couples who were found to have major causes of infertility such as periadnexal adhesions, tubal occlusion, moderate to severe endometriosis, and oligospermia were eliminated from this study. The remaining 91 women constitute the study group of this report. All women had a complete history and physical examination. Tubal patency was confirmed by hysterosalpingography or laparoscopy. Ovulation and luteal phase adequacy were documented in all cases by basal body temperature charting and endometrial biopsy. Postcoital tests were performed as previously described,2 serially beginning 2 to 3 days before the anticipated basal body temperature (BBT) thermal shift, 8 to 12 hours after coitus. Of the 91 women in the group, 83 had less than five motile sperm per high-power field in normal preovulatory mucus, while 8 had many vibratory and shaking sperm in the mucus. Ovulation Monitoring and Insemination Insemination was performed as close to the time of ovulation as possible. In unstimulated cycles, patients were monitored serially with a home urinary luteinizing hormon (LH) monitoring kit (Ovustick, Monoclonal Antibodies, Inc., Sunnyvale, CA; or QTest, Becton Dickinson, Franklin Lakes, NJ), beginning 3 days before the anticipated LH surge. During CC or hmg treatment, patients were monitored both sonographically and by serum estradiol (E2) levels. CC 50 mg was administered on cycle days 5 to 9. Those who did not conceive in unstimulated or in CC cycles were offered stimulation with. hmg. Two ampules ofhmg were initiated on cycle day 2 to 4 and administered daily. E2 and sonography were performed serially beginning after 4 to 5 days of hmg administration. Human chorionic gonadotropin (hcg) 10,000 units was administered intramuscularly when lead follicle diameters were 16 to 20 mm and E2 > 400 pg/ml. Sperm were prepared by a swim-up technique, as described below. With the patient in the lithotomy position, a speculum was placed in the vagina, and the cervix was cleansed with a dry cotton swab. Insemination was performed with either a Weissman lui catheter (A.J. Medical Supplies, New Hyde Park, NY) or a Shepard lui catheter (Cook OB/ GYN, Spencer, IN) for high fundal placement of 0.3 to 0.5 ml of washed sperm. After insemination the patient rested for 1 to 5 minutes and resumed normal activity. Antisperm Antibody Testing ASA were detected by direct and indirect immunobead binding method as previously described Spermatozoa from the husband known to be free of surface-bound immunoglobulins by direct immunobead binding were selected by a swim-up, washed by centrifugation (600 X g) and resuspended in Dulbecco's phosphate-buffered saline (PBS) (Gibco, Grand Island, NY) and 5 mg/ml bovine serum albumin (BSA) (Pentex, Fraction V, Miles Scientific, Naperville, IL). One million of these sperm cells were added to 1 ml of diluted sera (1:10 in PBS/BSA) for each women in the study and incubated for 60 minutes at 37 C. Spermatozoa were washed three times, and the final pellet was resuspended in 0.1 ml PBS/BSA. In each case, the region of the sperm surface to which antisperm antibodies had bound, the proportion of sperm that were antibody bound, and the isotypes were determined by immunobead binding. Sperm Swim-Up Semen speciments were collected by masturbation into a sterile plastic cup. After liquefaction, 0.5 ml aliquots were placed in sterile capped plastic test tubes under 2 ml of Ham's F-10 medium (Gibco, Grand Island, NY) prepared with 2.1 mg/ ml sodium bicarbonate (S-8875, Sigma, St. Louis, MO), mg/ml streptomycin sulfate (S-6501, Sigma), and mg/ml penicillin-g (PEN-NA, Sigma). The tubes were incubated for 60 to 90 minutes at 37oC in an atmosphere of 5% C02 in air. The top 1.8 ml of all tubes were collected, pooled, and centrifuged at 600 X g for 6 minutes. The pellet was washed in 2 ml of the same medium, resuspended in 0.3 to 0.5 ml of medium, and delivered to the examination room. The yield of total motile sperm was usually between 10% and 30% of the original semen samples. RESULTS Ninety-one women underwent 4 73 cycles of lui, which resulted in 26 pregnancies in 24 women. In sixty-two women with primary infertility, 18 conceived after lui treatment (29% ).' Of the 29 who had secondary infertility, six conceived (21% ). Nine of these 91 women subsequently conceived 442 Margalloth et al. lui for female antisperm antibodies Fertility and Sterility

3 Table 1 Pregnancy Rate and Outcome in thij Study Group Number of women in the study 91 Number of women who conceived after lui 24 Number of pregnancies with lui 26 Number of women who conceived in untreated cycles 9 Pregnancies after IVF 3 Total number of pregnancies 38 Live births 22 Ongoing pregnancies (>3 months) 8 Stillbirths 1 Fetal malformation 1 Spontaneous abortion 6 spontaneously a few months to 2 years after the lui treatments. Three other women conceived through IVF. Clinical data on these women and the results of the 38 pregnancies that occurred are presented in Table 1. The spontaneous abortion rate was 16% (6/38). One case resulted in a stillborn (1/ 38 = 2.6%), and another resulted in the birth of a congenital hydrocephalic child. The mean age of all wives was 32.3 ± 3.6 years with a range of 21 to 41. The mean age of those pregnant was 31.7 ± 3 years, compared with 32.6 ± 3.9 years for those who did not achieve pregnancy. The mean duration of infertility of the entire group was 4 ± 2.4 years, with a range of 1 to 14 years. The mean duration of infertility for those who conceived was 3 ± 2 years versus 4.4 ± 2.5 years for those who did not (P < 0.01). In 39 cases, only a single isotype of ASA was found (14 lga, 11 lgg, 16 lgm), whereas two isotypes were detected in 38 sera, and 14 had all three. Although a decrease in pregnancy rate was seen when only lgg ASA was present (11% versus 36% for lga and 36% for lgm) and when more than one isotype was detected (28% one, 26% two, 21% three), these differences were of no statistical significance. Table 2 presents the pregnancy rate according to the different types of ASA, their regional specificity for the sperm surface, and percent sperm-binding immunobeads. The extent of immunity to spermatozoa, as reflected by the percent of husband's sperm showing bound antibody after exposure to wife's serum at 1:10 dilution, was shown to be predictive of conception for head-directed antibodies (P < 0.02). This was not the case for taildirected antibodies. After the swim-up procedure, sperm motility was above 90% in 86% of cases. In the rest; it was above 50% in at least one cycle. Motility of < 50% was noted in only 4% of cycles. The total number of motile sperm inseminated was <5 X 10 6 in only two cases, and both failed to conceive. In 59 other cases, the total motile sperm cells inseminated in 310 cy- cles was between 5 and 15 million cells and 13 (22%) became pregnant. Of 30 women who had 112 cycles of lui with > 15 million motile sperm (range of 16 to 75 million), 11 (37%) achieved conception. This difference, however, was of no statistical significance (P = 0.19). Sixty-seven women underwent lui during unstimulated cycles on the day after detection of the urinary LH surge. Nine pregnancies were achieved in 285 cycles (one pregnancy per 32 cycles of treatment, 3%). Twenty women underwent 86 lui after CC, resulting in 6 pregnancies (1:14 cycles, 7%). Twenty-eight women underwent 102 cycles of lui after hmg and hcg stimulation, resulting in 11 pregnancies (1:9 cycles, 11 %). HMG stimulation was started in 25 cases after failure of lui during unstimulated or CC cycles and in only three cycles as the primary treatment (Table 3). The conception rate per cycle according to the method of treatment is presented in Table 4. The pregnancy rate of 67 women who underwent 285 luis in unstimulated cycles was 13.4% per woman and 3% per cycle. Nine pregnancies were documented during the follow-up period, one of a woman who had been pregnant during an lui cycle, delivered, and a year later conceived spontaneously. The other eight conceived spontaneously after failing to conceive on lui. Of the 75 women followed during a cumulative period of 800 months, the spontaneous pregnancy rate was 12% per woman and 1.1% per cycle. The pregnancy rate per cycle for the lui treatment was significantly higher (3% vs 1.1 %; p < 0.05). Table 2 Pregnancy Rate According to the Different Types of ASA, Their Regional Specificity for the Sperm Surface, and Percent Sperm-Binding Immunobeads in the Study Group % Sperm All Pregnancy binding" women Pregnant (%) Head IgA <50% % bound >50% % IgG <50% % >50% % IgM <50% % >50% % All <50% %b >50% %b Tail IgA <50% % bound >50% % IgG <50% % >50% % IgM <50% % >50% % All <50% % >50% % " See Materials and Methods. b P<0.02. Margalloth et al. lui for female antisperm antibodies 443

4 Table 3 Pregnancy Rate after lui for ASA in Unstimulated and Stimulated Cycles No. of women No. of lui cycles No. of Women Cycles per pregnancies pregnant pregnancy Unstimulated cc HMG+hCG % Twenty-six women underwent one or two cycles of lui, 17 women underwent three or four cycles, 28 women five or six, 17 women seven to 10 cycles, and 3 women underwent 11 cycles of lui. Twenty-one of the 26 pregnancies (81%) were achieved in the first two lui cycles. During unstimulated and CC lui cycles, all pregnancies occurred in the first two cycles of treatment, while no pregnancy occurred in the third to tenth cycles of treatment. Conception with hmg was significantly enhanced in that (1) pregnancy rate in the first three cycles was 14.3% compared with 5% in the unstimulated and 9. 7% in the CC cycles (P < 0.05) and (2) pregnancies occurred also in the third, fourth, and fifth cycles. Of the 28 patients who underwent hmg treatment, 25 previously had failed luis in unstimulated or CC cycles. Thus, the pregnancy rate of lui and hmg as the primary treatment is higher than the 39% observed. There was no difference in the mean preovulatory E 2 level in stimulated cycles when conceptive cycles were compared with the nonconceptive cycles (918 ± 661 pg/ml versus 859 ± 528 pg/ml). The lowest peak preovulatory E 2 level in a conception cycle was 526 pg/ml. Confounding infertility factors were found in 57 cases. Mild endometriosis was documented in 21 cases. Fourteen had minimal pelvic adhesions or a prior ectopic pregnancy. A borderline semen analysis (motility of 30% to 40% or sperm concentration of 10 to 20 X 10 6 cells/ml) was found in eight cases. Fourteen women had some uterine malformation such as a diethylstilbestrol (DES)-associated T shaped uterus (7), a septate uterus (6), or a unicornuate uterus. Table 5 summarizes the pregnancy rate of each group. In the 34 women whose only positive finding after an extensive evaluation was humoral ASA, the pregnancy rate was 35.3%, while in the other 57 cases with additional infertility factors, the pregnancy rate was 21% (P = 0.1; not significant [NS]). In more than 500 cycles of insemination no complications were noted. We did not have any case of infection, abdominal cramping pains, or abnormal vaginal bleeding. DISCUSSION The treatment of immunity to sperm in women is controversial. lui as a method of treatment in such cases has been advocated since 1978, 18 but very few cases using this treatment in women with ASA have been published. In a recent review of the literature/ 9 only 14 women with ASA were reported to have undergone lui and 5 pregnancies (36%) were achieved. In another current review, 10 a success rate of 23% was reported in 40 couples with ASA who underwent lui, but the author states that it was difficult to determine which partner had the antisperm antibodies in most instances, although it seemed that many more antibody-positive men were treated. Our overall success rate of 26% is comparable to that reported previously. The fact that most pregnancies occurred during the first few treatment cycles supports the argument of a cause-and-effect relationship. More important than the overall conception rate is the success rate in well-defined groups and the comparison of different treatment methods. While Table 4 Conception Rate per Cycle of lui Treatment in the Various Unstimulated and Stimulated Groups No. of cycle Unstimulated 4/67 5/62 0/47 0/38 0/24 pregnancy% 6% 8% CC pregnancy % 2/25 4/23 0/14 0/9 0/7 8% 17% HMG pregnancy % 5/28 1/20 3/15 1/11 1/8 18% 5% 20% 9% 12% /47 0/8 0/20 Table 5 Conception Rate in Females With ASA With and Without Confounding Infertility Factors Cases Pregnant Pregnancy % Female ASA OnliY Mild Endometriosis Uterine Malformation Borderline Semen Analysis Pelvic Adhesions or S/P Ectopic Pregnancy Margalloth et al. lui for female antisperm antibodies Fertility and Sterility

5 the pregnancy rate of the unstimulated group was 13% and the pregnancy/cycle ratio was 1:32, in the hmg group, the pregnancy rate was 39% and the pregnancy/cycle ratio was 1:9. The exact mechanism by which hmg stimulation improves conception rates is unknown. It is assumed that superovulation improves the chances of pregnancy by increasing the number of oocytes available for fertilization. But in cases of female immunity to sperm, other important factors may play a role. The levels of antibodies in the cervical mucus and in oviductal fluid had been shown to be negatively correlated to estrogen levels and to decrease after hmg treatment.20 Elevated estrogen levels were also found to lower complement concentration in the reproductive tract. The increased conception rate with the use ofhmg may be attributed to a decrease in the levels of ASA and complement within the reproductive tract due to the elevated estrogens. The fact that in our study group no pregnancy occurred in stimulated cycles with E2 < 526 pg/ml supports this hypothesis. Additional benefit of hmg-hcg treatment over urinary LH monitoring could be attributed to the better timing of insemination relative to ovulation. Ovulation occurs 36 to 42 hours after hcg administration. 21 Because the initiation of the LH surge is not defined by single daily urinary LH monitoring, the estimated ovulation window is 12 to 24 hours. As sperm may have a shortened in vivo lifespan in the presence of ASA, the better timing of insemination relative to ovulation after hmg-hcg treatment may have contributed to the increased pregnancy rates in this group. Our study group was not large enough to address fully the optimal number of sperm to be placed in the uterine cavity. Because couples with male infertility were excluded in this study, we usually had more than 5 million motile sperm for insemination. However, there was an increased conception rate when > 15 million motile sperm were inseminated when compared to the group who were inseminated with 5 to 15 million motile cells (37% versus 22%; P = 0.19, NS). This gives some support to the theory that one of the advantages ofiui is the increase in number of sperm entering the uterus. The aggregate numbar of antigenic sites on inseminated sperm might then exceed the level of available ASA so that the antibody burden (amount of immunoglobulins per sperm) is diminished. The level ofimmunobead binding to the sperm tail did not influence the chance of pregnancy after lui. In contrast, the pregnancy rate when more than 50% of the sperm were antibody-bound over the head was one third (13%) of that when binding levels were <50% (38%). Head-directed antibodies have been shown to impair gamete interaction, whereas taildirected antibodies do not influence sperm capacitation, zona binding, and egg membrane penetration.1'5'6 Hence, although lui overcomes the block to penetration of cervical mucus caused by head- or tail-directed antibodies, it cannot circumvent the impaired fertilizing ability of sperm caused by exposure to head-directed antibodies in the uterine and tubal fluids. The incidence of early pregnancy loss in women with antisperm antibodies was initially reported to be approximately 50%. 7 Pregnancy wastage has been shown to occur in animals experimentally stimulated to produce ASA. Menge and associates22 reported that immunization with an extract of solubilized surface antigens of the rabbit's sperm plasma membrane led to impaired fertility due to inhibition of fertilization, while immunization with a nonionic detergant extract of sperm impaired fertilization at a postfertilization level. They postulated that antigens responsible for inducing postfertilization infertility originate from a subsurface site on spermatozoa and might be conserved on the preimplantation embryo or trophectoderm. In another study, Menge et al.23 reported that four of seven pregnant women with ASA aborted; all had a titer of circulating ASA of >1:16. Haas et al., 24 studying ASA in recurrently aborting women, reported that all women with ASA detected by radiolabeled antiglobulin assay (RAA) continued to abort subsequent pregnancies, despite leukocyte immunization. This was in contrast to a similar group of recurrently aborting women that had no ASA as detected by RAA, who carried their fetuses to term after leukocyte immunization. In contrast to these findings, we did not observe any increased risk of miscarriage in women with sperm-surface ASA who were not recurrent aborters. The 16% (6/ 38) abortion rate is well within that reported for women without ASA. Similar to our results, lngerslev and lngerslev25 were unable to demonstrate an increase miscarriage rate in ASA positive women. In conclusion, our study demonstrates that infertile women with immunity to sperm can benefit from lui. If ASA is the only factor found, the chance of conception with hmg stimulation is 40%, with a 70% to 80% chance that the pregnancy would occur in the first two lui cycles. Our data suggest that the treatment of women with immunity to sperm by lui should be performed with ovarian etimulation. We recommend CC stimulation for 2 months, followed by stimulation with Margalloth et al. lui for female antisperm antibodies 445

6 hmg for 5 to 6 months. lui treatment should be continued for six to eight cycles before proceeding to IVF. REFERENCES 1. Bronson RA, Cooper GW, Rosenfeld DL: Factors effecting the population of the female reproductive tract by spermatazoa: their diagnosis and treatment. Sem Reprod Endocrinol 4:371, Price RJ, Boettcher B: The presence of complement in human cervical mucus and its possible relevance to infertility in women with complement-dependent sperm-immobilizing antibodies. Fertil Steril 32:61, Jager S, Kremer J, de Wilde-Janssen IW: Are sperm immobilizing antibodies in cervical mucus an explanation for a poor postcoital test? Am J Reprod Immunol 5:56, Beer AE, Neaves WB: Antigenic status of semen from the viewpoints of the female and the male. Fertil Steril 29:3, Bronson RA, Cooper GW; Rosenfeld DL: Sperm-specific isoantibodies and autoantibodies inhibit the binding of human sperm to the human zona pellucida. Fertil Steril 38: 724, Tsukui S, Noda Y, Yano J, Fukuda A, Mori T: Inhibition of sperm penetration through human zona pellucida by antisperm antibodies. Fertil Steril 46:92, Jones WR: Immunological aspects of infertility. In Immunology of Human Reproduction, Edited by JS Scott, WR Jones. London, Academic Press, 1976, p London SN, Haney AF, Weinberg JB: Diverse humoral and cell-mediated effects of antisperm antibodies on reproduction. Fertil Steril41:907, Franklin RR, Dukes CD: Antispermatozoal antibody and unexplained infertility. Am J Obstet Gynecol89:6, Haas GG: Immunologic infertility. Obstet Gynecol North Am 14:1069, Isojima S, Li TS, Ashitaka Y: Immunologic analysis of sperm immobilizing factor found in sera of women with unexplained infertility. Am J Obstet Gynecol101:677, Alexander NJ, Sampson JH, Fulgham DL: Pregnancy rates in patients treated for antisperm antibodies with prednisone. Int J Fertil 28:63, Shulman S, Harin B, Davis P, Reyniak JV: Immune infertility and new approaches to treatment. Fertil Steril29:309, Clark GN, Lopata A, Johnston WIH: Effect of sperm antibodies in females on human in vitro fertilization. Fertil Steril 46:435, Yovich JL, Stanger JD, Kay D, Boettcher B: In vitro fertilization of oocytes from women with serum antisperm antibodies. Lancet 1:369, Bronson RA, Cooper GW, Rosenfeld DL: Membrane bound sperm specific antibodies: their role in infertility. In Bioregulators in Reproduction, Edited by H Vogel, G Jagiello. New York, Academic Press, 1981, p Bronson RA, Cooper GW, Rosenfeld DL: Sperm antibodies: their role in infertility. Fertil Steril42:171, Kremer J, Jager S, Kuiken J: Treatment of infertility caused by antisperm antibodies. Int J Fertil 23:270, Alexander NJ, Ackerman S: Therapeutic insemination. Obstet Gynecol North Am 14:905, Yang SL, Schumacher GFB, Broer KA, Holt JA: Specific antibodies and immunoglobulins in the oviductal fluid of the rhesus monkey. Fertil Steril 39:359, Testart J, Frydman R: Minimum time lapse between luteinezing hormone (LH) surge or human chorionic gonadotropin (hcg) administration and follicular rupture. Fertil Steril 37:50, Menge AC, Peegel H, Riolo ML: Sperm fractions responsible for immunologic induction of pre- and post-fertilization infertility in rabbits. Biol Reprod 20:931, Menge AC, Medley NE, Mangione CM, Dietrich JW: The incidence and influence of antisperm antibodies in infertile human couples on sperm-cervical mucus interactions and subsequent fertility. Fertil Steril38:439, Haas GG, Kubota K, Quebbeman JF, Jijon A, Menge AC, Beer AE: Circulating antisperm antibodies in recurrently aborting women. Fertil Steril45:209, Ingerslev HJ, Ingerslev M: Clinical findings in infertile women with circulating antibodies against spermatozoa. Fertil Steril33:514, 1980 I. 446 Margalloth et al. lui for female antisperm antibodies Fertility and Sterility