Effect of antiphospholipid antibodies in women undergoing in-vitro fertilization: role of heparin and aspirin

Transcription

1 Human Reproduction vol.12 no.6 pp , 1997 Effect of antiphospholipid antibodies in women undergoing in-vitro fertilization: role of heparin and aspirin William H.Kutteh 1,4, Deborah L.Yetman 1, recurrent pregnancy loss or thrombosis with positive APA or Samuel J.Chantilis 2 and Jack Crain 3 lupus anticoagulant on two occasions (Harris, 1990). Although several treatments have been advocated, heparin and aspirin The University of Tennessee, Memphis Health Science Center, treatment is emerging as the treatment of choice for the APA Memphis, Tennessee, 2 The University of Texas Southwestern Medical Center, Dallas, Texas, and 3 The Nalle Clinic Center for syndrome associated with recurrent pregnancy loss (Lockshin, Women, Charlotte, North Carolina, USA 1992; Cowchock et al., 1992; Kutteh, 1996; Raziel et al., ). However, the significance of APA in a woman without To whom correspondence should be addressed at: Division of Reproductive Endocrinology, Department of Obstetrics and a prior pregnancy or in the absence of prior thromboembolic Gynecology, 956 Court Avenue, Room D324, Memphis, Tennessee phenomena is unclear (Bronson, 1995) , USA Previously, the reproductive autoimmune failure syndrome was described in women with increased autoantibodies, endo- To describe the prevalence of antiphospholipid antibodies metriosis and infertility, leading to a recommendation for in women undergoing in-vitro fertilization (IVF) and to immunological testing of women with infertility and endodetermine if heparin and aspirin affect implantation rates, metriosis (Gleicher et al., 1989). Recently, several investigators 191 women with a history of infertility undergoing IVF have advocated testing women undergoing in-vitro fertilization were prospectively tested for antiphospholipid antibodies. (IVF) for APA (Gleicher et al., 1994; Birdsall et al., 1996). This was a two-centre, non-randomized comparison of The theoretical rationale for the role of APA and the potential women with positive antiphospholipid antibodies receiving benefits of anticoagulation therapy for women undergoing IVF heparin and aspirin versus standard treatment. An enzyme- is based on several observations. Firstly, phospholipids function linked immunosorbent assay, with referenced standards as adhesion molecules during the formation of syncytiotrophoand known positive and negative sera on each plate, was blasts (Sessions and Horowitz, 1982). Secondly, the attachment utilized to measure antibodies to cardiolipin, phosphatidyl- of APA to surface phospholipids on trophoblasts may result inositol, phosphatidylglycerol, phosphatidylserine and in direct cellular injury. Moreover, inhibition of syncytiotrophophosphatidylethanolamine. Statistical analyses of results blast formation may cause indirect damage via intravascular included analysis of variance and Fisher s two-tailed exact thrombosis (Rote et al., 1992). More recently, some investigators test. Antiphospholipid antibodies were detected in 18.8% have recommended treatment of all APA-positive of patients undergoing IVF compared with only 5.5% women with heparin and aspirin (Sher et al., 1994; Kowalik in the 200 normal controls, 26% in 200 women with et al., 1996). Exogenous heparin has been shown to inhibit recurrent pregnancy loss, and 32% in 200 women with binding of APA with phospholipids (Ermel et al., 1995), systemic lupus erythematosus. In conclusion, antiphosphofunction and endogenous heparin manufactured by trophoblasts should lipid antibodies were found more frequently in women in the same fashion. The antithromboxane effects of undergoing IVF than in the normal control population. aspirin on inhibition of platelet aggregation are thought to work Although implantation rates appeared higher in the group in concert with heparin to promote and enhance implantation of women treated with heparin and aspirin, no statistically (Patrono, 1994; Hauth, 1995). significant differences were detected in implantation, pregprevalence of APA in women undergoing IVF has been The purpose of this study was 2-fold. First, we believe the nancy and ongoing pregnancy rates between those who received standard therapy and those treated with heparin exaggerated in recent investigations; therefore, we determined the prevalence of anticardiolipin antibodies (ACA) and APA and aspirin. in women undergoing IVF. Secondly, in order to determine Key words: anticardiolipin antibodies/antiphospholipid antiwhether the proposed enhancement of IVF success in women bodies/heparin/immunotherapy/in-vitro fertilization with APA being treated with heparin and aspirin was secondary to effects on implantation and/or early pregnancy loss, we evaluated the potential effect of heparin and aspirin therapy Introduction on pregnancy, ongoing pregnancy and implantation rates in Antiphospholipid antibodies (APA), a group of autoantibodies women with positive APA undergoing IVF. that bind to negatively charged phospholipids (Harris and Hughes, 1988), have clinical significance because of their Materials and methods association with thromboembolic events and adverse pregnancy Patient population outcomes (Cowchock, 1991). The antiphospholipid antibody We evaluated 191 different women undergoing IVF. The work was syndrome has been described in women with a history of carried out between 1 June 1995 and 30 August 1996 at two centres: European Society for Human Reproduction and Embryology 1171

2 W.H.Kutteh et al. centre 1, Southwestern Fertility Associates at the University of Texas standardization as described previously (Kutteh et al., 1994). Briefly, Southwestern Medical Center in Dallas, Texas, USA and centre 2, phospholipid units for IgG and IgM were calculated for each serum the Nalle Clinic Center for Women in Charlotte, North Carolina, sample, and the median value was determined from the non-gaussian USA. For this study, only women undergoing their first cycle of IVF distribution. The cut-off value in phospholipid units for each APA at age 42 years at the time of retrieval with follicle stimulating was then determined by using the 99th percentile of the normal hormone (FSH) of 20 miu/ml, oestradiol 50 pg/ml on cycle day population, approximately 3 times the median value. All values 3 and a normal uterine cavity were included. All women had a reported as positive were the mean of triplicate determinations with complete evaluation that included a history and physical examination, background absorbence obtained from wells prepared without the hysterosalpingogram or hysteroscopy and mid-luteal progesterone or coating phospholipid subtracted. Any values with SE 10% were late-luteal phase endometrial biopsy. Patients were excluded from discarded and samples re-assayed. Interassay and intra-assay variation this study if they had two or more prior clinical pregnancy losses were 8 and 6% respectively. (not including ectopic pregnancies), one or more prior stillbirths, thrombocytopenia, allergies to aspirin, a prior history of thrombo- Stimulation protocol embolic disorder, a history of osteopenia, osteoporosis, or any other Both centres utilized similar protocols for IVF. Briefly, women were bone disorder. None of the women included in this study suffered down-regulated with Lupron (Tap Pharmaceuticals, Deerfield, IL, any serious adverse effects from heparin therapy which include USA) starting on cycle day 21 until oestradiol was 35 pg/ml and bleeding, bleeding/infection at the injection site, thrombocytopenia, no ovarian cysts 20 mm diameter were present on transvaginal and hypersensitivity. The control populations consisted of 200 non- ultrasound. Gonadotrophins were initiated using a Metrodin (Serono, pregnant, reproductive-aged, parous women who had no history of Randolph, MA, USA) weighted stimulation protocol based on con- reproductive problems, 200 non-pregnant women with recurrent centrations of FSH and oestradiol measured on cycle day 3, patient pregnancy loss (three or more consecutive losses), and 200 nonachieved a mean diameter of 18 mm with appropriate oestradiol weight and prior stimulation cycles, if any. When at least two follicles pregnant women with systemic lupus erythematosus. concentrations, human chorionic gonadotrophin (HCG, Profasi; Laboratory evaluation Serono) was administered, and transvaginal oocyte aspiration was scheduled in 34 h. Oocytes were recovered, cumulus was stripped, All women in this study had cycle day 3 serum samples evaluated and insemination of oocytes occurred after 3 h with motile for the presence of APA using the enzyme-linked immunosorbent assay spermatozoa. Gametes were cultured in human tubular fluid (HTF; (ELISA) method as described by Harris (1990). Briefly, individual Irvine Scientific, Santa Ana, CA, USA) with 0.5% human serum 96-well microtitre plates (Immulon-2; Dynatech Labs, Chantilly, VA, albumin (HAS; Irvine Scientific). After 18 h, fertilization was USA) were coated with 30 µl of cardiolipin, phosphatidylinositol, determined, and embryos were cultured in HTF with 0.3% albumin. phosphatidylglycerol, phosphatidylserine or phosphatidylethanol- Transfer occurred at h post-aspiration. Generally, three to four amine (Sigma Chemical Co., St Louis, MO, USA) at a concentration embryos at the six- to 12-cell stage were transferred based on embryo of 45 µg/ml (cardiolipin) in ethanol and 50 µg/ml (all other phosphoquality, endometrial thickness and pattern, patient age and desires of lipids) in methanol. The plates were air-dried overnight at 4 C, the couple. Luteal phase support was maintained with progesterone blocked with 200 µl of 10% fetal calf serum (GIBCO, Long Island, in oil, 50 mg i.m. daily. Pregnancy tests were performed in 12 days. NY, USA) in 1 phosphate buffered saline (GIBCO), washed, and Pregnancy rate was defined as number of women with at least two incubated at 37 C for 2 h with 50 µl of patients sera diluted 1:50 in appropriately rising quantitative βhcg values. Ongoing pregnancy 10% fetal calf serum in phosphate buffered saline. Each unknown rate was defined as a pregnancy documented with fetal heart motion sample was run in triplicate. The plates were then washed to remove by transvaginal ultrasonography detected on at least one occasion. unbound antibody and proteins, and a secondary antibody, alkaline Implantation rate was determined based on the number of ongoing phosphatase conjugated antihuman immunoglobulin G (IgG) (Caltag pregnancies (with fetal heart motion) divided by the number of Labs, San Francisco, CA, USA), IgM (Biosource, Tago Immunoembryos transferred. logicals, Camarillo, CA, USA), or IgA (Calbiochem, La Jolla, CA, USA), was added to the plate. After incubation and washing, p- Heparin and aspirin treatment nitrophenyl phosphate substrate (Sigma 104) was added and used to Baseline activated partial thromboplastin time (aptt) and platelet measure indirectly the level of APA in a patient s serum. The optical counts were obtained prior to cycle initiation. Patients with prolonged density of the samples, caused by the cleavage of the substrate by aptt or thrombocytopenia ( platelets/ml) were excluded the enzyme, was determined at 405 nm by a BioRad Microplate from the study. At centre 2, women with positive APA initiated Reader Model 450 (BioRad Laboratories, Richmond, CA, USA) and aspirin 81 mg/day at the time of gonadotrophin start and s.c. heparin was used to quantify the amount of APA in the sera. Every assay 5000 units twice daily beginning on the night of aspiration. Midplate also included a known high positive anticardiolipin sample interval aptt and platelets were evaluated at 1 and 2 weeks after [ 100 GPL (see below)] run in triplicate. Plates were incubated until heparin initiation. If platelet counts decreased to /ml or if the high positive wells achieved an optical density of 1.0; typically, the patient s aptt values were outside the normal range, heparin this required an incubation of min. Referenced standard sets dosage was reduced by 1000 units per dose, and laboratory tests were for cardiolipin (Louisville APL Diagnostics, Louisville, KY, USA) re-evaluated in 1 week. Treatment was continued for 13 weeks as a and known negative sera were used on every plate. All results were part of this study. Thereafter, treatment was continued or terminated defined in phospholipid units for IgG (GPL) and IgM (MPL) as based on a consensus between the patient, her obstetrician and her follows: 10 units, negative; units, borderline; units, reproductive endocrinologist. Patients who did not receive heparin positive; and 80 units, high positive. Results for IgA were defined and aspirin were treated according to our standard IVF protocol in phospholipid units for IgA (APL) as follows: 30 units, negative; without any additions (standard treatment). 30 units, positive. Phosphatidylinositol, phosphatidylglycerol, and phosphatidylserine values were interpreted based on the anticardiolipin Statistical analysis standards of Harris (1990). For phosphatidylethanol- Statistical analyses were performed using analysis of variance amine, the multiples of the median method was used for (ANOVA) and two-tailed Fisher s exact test. 1172

3 Antiphospholipid antibodies and in-vitro fertilization Table I. Positive anticardiolipin (ACA) and antiphospholipid antibodies (APA) a Table II. Demographics of women with positive antiphospholipid antibodies (APA) undergoing in-vitro fertilization (IVF) a Group n Positive % positive Positive % positive Diagnosis n Positive APA % positive ACA APA Endometriosis Controls b b 5.5 Pelvic/tubal factor IVF c d 18.8 Male factor RPL Unexplained SLE Anovulation Total a APA included antibodies to cardiolipin (ACA) in addition to phosphatidylserine, phosphatidyl-glycerol, phosphatidylinositol, and a Differences between each group were not significant by ANOVA phosphatidylethanolamine. (P 0.08). b Differences between the normal controls and all other groups were significant, P c ACA differences in IVF group versus controls P 0.014, RPL P 0.039, and SLE P Women with positive APA received standard treatment at d APA differences in IVF group versus controls P 0.001, RPL P 0.115, centre 1, while centre 2 administered heparin and aspirin and SLE P to women with positive APA. Table III demonstrates the RPL recurrent pregnancy loss; SLE systemic lupus erythematosus. similarities between the two centres. There was no difference Results in the average age of women, in the frequency of positive APA, the number of embryos transferred, the implantation Antiphospholipid antibodies in women undergoing IVF rates, pregnancy or ongoing pregnancy rates. Blood samples from 191 women undergoing IVF and 200 parous women with no history of reproductive problems were Pregnancy rate in women undergoing IVF tested by ELISA for APA. Overall, seven out of 200 (3.5%) different women from the control group were positive for As demonstrated in Table IV, 52.6% women who had positive ACA, while 19 out of 191 (9.9%) different women with a APA and received heparin and aspirin therapy became pregnant. history of infertility were positive for IgG and/or IgM ACA In comparison, women with positive APA and standard treat- (P 0.014), as demonstrated in Table I. In comparison, 35 ment had a pregnancy rate of 47.0%. In addition, women out of 200 (17.5%) women with recurrent pregnancy loss and testing negative for APA and receiving standard treatment had 50 out of 200 (25%) women with systemic lupus erythematosus a pregnancy rate of 50.3%. Four women with borderline levels had positive ACA. Significant differences were found in of APA (classified as negative) desired treatment and received ACA positivity in women undergoing IVF when compared to heparin and aspirin. Three became pregnant (75.0%). The women with recurrent pregnancy loss (P 0.039) and SLE differences in pregnancy rates between groups were not signi- (P 0.001). ficant. To determine if any women with negative ACA were positive for one of the other antiphospholipid antibodies, results Ongoing pregnancy rate (OPR) in women undergoing IVF for IgM, IgG and IgA antibodies against phosphatidylinositol, To determine if APA might play a role in early pregnancy phosphatidylglycerol, phosphatidylserine and phosphatidyl- loss, thus affecting the number of ongoing pregnancies, we ethanolamine were evaluated. As expected, when women were evaluated the ongoing pregnancy rates (Table IV). Eight of 19 tested for the full panel of APA in addition to anticardiolipin, (42.1%) women undergoing IVF who had positive APA and more positive women were identified in each group included received heparin and aspirin therapy had an ongoing pregnancy. in this study. Out of 191 women undergoing IVF who were Similarly, women with positive APA and standard treatment negative for IgM, IgG and IgA ACA, 17 (8.9%) tested positive had an ongoing pregnancy rate of 35.3%. Additionally, women for another APA. In all, 36 out of 191 (18.8%) women with no APA and standard treatment had an ongoing pregnancy undergoing IVF were positive for APA compared with 11 of rate of 39.7%, while women with no APA (borderline) receiving 200 (5.5%) controls (P 0.001), considering patients with heparin and aspirin had an ongoing pregnancy rate of 75.0%. more than one positive immunoglobulin only once (Table I). The differences in ongoing pregnancy rate between each group Of these women with positive APA, the isotype distribution were not significant. was 25 of 36 (69.4%) IgG, 11 of 36 (30.6%) IgM, and 0 of 36 IgA. Only values that were 20 phospholipid units of Implantation rates in women undergoing IVF IgG or IgM were considered positive. The average positive value for women undergoing IVF was GPL and As seen in Table V, women with positive APA receiving MPL. As seen in Table II, there were no differences heparin and aspirin had an implantation rate of 25.0%. In in APA positive rates based on the presumptive diagnosis comparison, the implantation rate of women with positive made prior to IVF. APA receiving standard treatment was 19.4%. In addition, the implantation rates of women with negative APA receiving Comparison of centre 1 and centre 2 standard treatment compared to heparin and aspirin were 19.4 Stimulation protocols, criteria for administration of HCG, and and 26.7% respectively. The differences in implantation rate laboratory culturing methods were similar in both centres. between these groups were not significant. 1173

4 W.H.Kutteh et al. Table III. Comparison of results for all women undergoing in-vitro fertilization (IVF), grouped according to treatment centre: in centre 1, women with positive antiphospholipid antibodies (APA) received standard treatment, in centre 2, women with positive APA received heparin and aspirin Centre 1 Centre 2 a Significance b (n 79) (n 112) Age (years) NS Positive APA (%) 13/79 (16.4) 23/112 (20.5) NS No. embryos transferred SE NS Implantation rate (%) 50/247 (20.2%) 66/324 (20.3%) NS Pregnancy rate (%) 39/79 (49.3%) 58/112 (51.7%) NS Ongoing rate (%) 30/79 (37.9%) 47/112 (41.9%)NS APA antiphospholipid antibody. a Four women with positive APA refused heparin and aspirin treatment. b NS not significant. pregnancy or implantation rates based on the presence or Table IV. Pregnancy rates and ongoing pregnancy rates of women who absence of APA and women receiving standard treatment or received heparin/aspirin versus standard treatment a heparin and aspirin. It is important to emphasize that this study APA treatment No. patients No. pregnant No. ongoing was non-randomized; therefore, the strength of the conclusions pregnancies is limited. However, the two centres had similar protocols and Positive Hep/Asp (52.6) b 8 (42.1) overall success rates. Our data relate to other reports in the Positive standard 17 8 (47.0) 6 (35.3) literature, with the exception of Sher et al. (1994). Briefly, Negative c Hep/Asp 4 3 (75.0) 3 (75.0) Gleicher et al. (1994) found 27.6% of 105 women undergoing Negative standard (50.3) 60 (39.7) IVF had positive APA but reported no differences in pregnancy, Values in parentheses are percentages. biochemical pregnancy or early loss rates. Nip et al. (1995) Asp aspirin; Hep heparin. reported no differences in biochemical pregnancy, clinical Differences between each group were not significant by ANOVA (P 0.79 and P 0.53). pregnancy or live birth rates in 100 women undergoing IVF Pregnancy rate: the number of women with at least two appropriately with unexplained infertility, endometriosis and tubal factor rising quantitative βhcg values, and ongoing pregnancy rate: a pregnancy infertility. Birdsall et al. (1996) reported 15.0% of 240 women documented with fetal heart motion by transvaginal ultrasonography detected on at least one occasion. undergoing IVF had positive ACA and antiphosphatidylserine Borderline values for APA. antibodies and reported an association with intrauterine growth retardation. However, they found no difference in the implantation rates between controls and women undergoing IVF. A Table V. Implantation rates of women who received heparin/aspirin versus recent retrospective study reported an 18.1% positivity for standard treatment a ACA and antiphosphatidylserine antibodies in women under- No. embryos No. embryos Implantation rate going IVF (Kowalik et al., 1996). The authors concluded that implanted transferred (%) b the presence of APA did not affect pregnancy rates, but there Positive Hep/Asp was a trend towards increased APA in women with early Positive standard miscarriage. Schenk and coworkers (1996) found that 58% of Negative c Hep/Asp women undergoing IVF had positive APA. They utilized Negative standard a positive value definition of 4 SD from the mean, a method Asp aspirin; Hep heparin. that may not be appropriate given the non-gaussian distribution Differences were not significant by ANOVA (P 0.66). of APA values. Nonetheless, they concluded that the presence Implantation rate was determined based on the number of ongoing pregnancies (with fetal heart motion) divided by the number of embryos of APA did not affect pregnancy rates, ongoing pregnancy transferred. rates, or implantation rates. In the report by Sher et al. (1994), 27% of 171 women who were negative for APA had an Discussion ongoing pregnancy compared with only 15% of 25 women In this study, only 3.5% of parous women with no history of who were positive for APA. However, in another group of 187 adverse obstetric outcomes had positive ACA, and 5.5% were women who were positive for APA and received heparin and positive for APA. These findings agree with previous reports aspirin therapy, 47% had an ongoing pregnancy. The authors that 1 4% of healthy women have positive APA (Lockwood suggested heparin and aspirin therapy overcome the adverse et al., 1989; Harris and Spinnato, 1991). However, our results effects of APA on implantation, and advocated that universal suggest that 36 of 191 (18.8%) of women undergoing IVF testing and treatment for APA would greatly enhance IVF had positive APA. These data relate to several different success. Their study was non-randomized, and it is unclear investigations which found 15 45% of women undergoing which women with positive APA were treated. The authors IVF had positive APA (Gleicher et al., 1994; Sher et al., 1994; Nip et al., 1995; Birdsall et al., 1996). In this report, we were unable to detect a difference in pregnancy, ongoing 1174 suggested that the difference in the pregnancy rates of normal women (27%) versus women with positive APA receiving heparin and aspirin (47%) may have been related to transient,

5 Antiphospholipid antibodies and in-vitro fertilization undetected APA produced during ovarian stimulation. In reality, Gleicher, N., Liu, H., Dudkiewicz, A. et al. (1994) Autoantibody profiles and immunoglobulin levels as predictors of IVF success. Am. J. Obstet. Gynecol., when it is considered that women were overdiagnosed with 170, APA in their study, the real question becomes does heparin Harris, E.N. (1990) Annotation: antiphospholipid antibodies. Br. J. Haematol., and aspirin improve the success rate in all women undergoing 74, 1-9. Harris, E.N. and Hughes, G.R. (1988) Thrombosis and miscarriages: the IVF? antiphospholipid antibody story. In Atlas of Science: Immunology. Institute Considerable attention has been directed to recent investi- of Scientific Information, Philadelphia, PA, pp gations which reported positive APA in as many as 60% of Harris, E.N. and Spinnato, J.A. (1991) Should anticardiolipin tests be performed women with a history of infertility and endometriosis who in otherwise healthy pregnant women? Am. J. Obstet. Gynecol., 165, were undergoing IVF (Sher et al., 1994). These very high Hauth, J.L. (1995) Low-dose aspirin: lack of association with an increase in positive rates of APA in apparently healthy women under- abruptio placentae or perinatal mortality. Obstet. Gynecol., 85, going IVF must be considered in the context of the reports Kowalik, A., Vichnin, M., Branch, W. and Berkeley, A. (1996) Mid-follicular that only 25 42% of patients diagnosed with systemic lupus anticardiolipin and antiphosphatidylserine antibody titres do not correlate with IVF outcome. American Society of Reproductive Medicine Meeting, erythematosus have ACA (Alarcon-Segovia et al., 1989; Love November, 1996, Boston, MA, USA. Abstract P-130. and Santoro, 1990; Qamar et al., 1990; Kutteh et al., 1993). Kutteh, W.H. (1996) Antiphospholipid antibody-associated recurrent Therefore, one must seek other explanations for the investilow-dose aspirin alone. Am. J. Obstet. Gynecol., 174, pregnancy loss: treatment with heparin and low-dose aspirin is superior to gations which report that approximately half of women under- Kutteh, W.H., Lyda, E.C., Abraham, S.M. and Wacholtz, M.C. (1993) going IVF have significant amounts of positive APA (Sher Association of anticardiolipin antibodies and pregnancy loss in women with et al., 1994; Schenk et al., 1996). In one controversial report, systemic lupus erythematosus. Fertil. Steril., 60, % of the women undergoing IVF had positive APA, and Kutteh, W.H., Webster, R. and Kutteh, C.C. (1994) Multiples of the median: an alternative method for reporting antiphospholipid antibodies in women the authors recommended universal testing of APA (Sher et al., with recurrent pregnancy loss. Obstet. Gynecol., 84, ). However, it is critical to realize that women were Lockshin, M.D. (1992) Antiphospholipid antibody syndrome. JAMA, 268, considered positive at any level of APA, and that 17% of Lockwood, C.J., Romero, R., Feinberg, R.F. et al. (1989) The prevalence and the control population in that study was composed of women biological significance of lupus anticoagulant and cardiolipin antibodies in with no obstetric problems who had positive APA. Based on a general obstetric population. Am. J. Obstet. Gynecol., 161, the very high proportion of positive APA found in their normal Love, P.R. and Santoro, S.A. (1990) Antiphospholipid antibodies: anticardiolipin and the lupus anticoagulant in systemic lupus erythematosus population (17%) and the notation that women undergoing (SLE) and in non-sle disorders. Ann. Int. Med., 112, IVF who had APA at any level were considered positive, Nip, M.M.C., Taylor, P.V., Rutherford, A.J. and Hancock, K.W. (1995) one can surmise that their APA assay was configured to Autoantibodies and antisperm antibodies in sera and follicular fluids of overdiagnose APA. Our study emphasizes the importance of infertile patients; relation to reproductive outcome after in-vitro fertilization. Hum. Reprod., 10, standardization of the ELISA, quality control and careful Patrono, C. (1994) Aspirin as an antiplatelet drug. N. Engl. J. Med., 330, interpretation of positive results. Future studies should include data on control populations, definition of positive values of Qamar, T., Levy, R., Sammaritano, L. et al. (1990) Characteristics of high- APA, standardization methods for the APA assay and the titer IgG antiphospholipid antibody in systemic lupus erythematosus patients with and without fetal death. Arthr. Rheum., 33, randomization of patients to treatment. To address these Raziel, A., Herman, A., Bukovsky, I. et al. (1996) Intravenous immunoglobulin important issues, we are currently involved in a multicentre, treatment of pregnant patients with unexplained recurrent abortions. Hum. prospective, randomized clinical trial to evaluate the role of Reprod., 11, Rote, N.S., Walter, A., and Lyden, T.W. (1992) Antiphospholipid antibodies: APA in women undergoing IVF and the effects of heparin and lobsters or red herrings? Am. J. Reprod. Immunol., 28, aspirin therapy on outcome. Schenk, L.M., Butler, L., Morris, J.P. et al. (1996) Heparin and aspirin treatment yields higher implantation rates in IVF patients with antiphospholipid antibody seropositivity compared to untreated seronegative References patients. American Society of Reproductive Medicine Meeting, November 1996, Boston, MA, USA, Abstract O-076. Alarcon-Segovia, D., Deleze, M., Oria, C.O. et al. (1989) Antiphospholipid Sessions, A. and Horowitz, A.F. (1982) Differentiation related difference in antibodies and antiphospholipid syndrome in systemic lupus erythematosus: the plasma membrane phospholipid symmetry of myogenic and fibrogenic a prospective analysis of 500 consecutive patients. Medicine (Baltimore), cells. Biochem. Biophys. Acta, 728, , Sher, G., Feinman, M., Zouves, C. et al. (1994) High fecundity rates following Birdsall, M.A., Lockwood, G.M., Ledger, W.L. et al. (1996) Antiphospholipid in-vitro fertilization and embryo transfer in anti-phospholipid antibody antibodies in women having in-vitro fertilization. Hum. Reprod., 11, seropositive women treated with heparin and aspirin. Hum. Reprod., 9, Bronson, R. (1995) Editorial: immunology and reproductive medicine. Hum. Reprod., 10, Received on December 17, 1996; accepted on April 8, 1997 Cowchock, F.S. (1991) The role of antiphospholipid antibodies in obstetric medicine. In Lee, R.V. (ed.), Current Obstetric Medicine. Mosby-Yearbook, St Louis, MO, pp Cowchock, F.S., Reece, E.A., Balaban, D. et al. (1992) High fetal losses associated with antiphospholipid antibodies: a collaborative randomized trial comparing prednisone with low-dose heparin treatment. Am. J. Obstet. Gynecol., 166, Ermel, L.D., Marshburn, P.B. and Kutteh, W.H. (1995) Interaction of heparin with antiphospholipid antibodies (APA) from sera of women with recurrent pregnancy loss (RPL). Am. J. Reprod. Immunol., 33, Gleicher, N., El-Roeiy, A., Carfino, E. and Friberg, J. (1989) Reproductive failure because of autoantibodies: unexplained infertility and pregnancy wastage. Am. J. Obstet. Gynecol., 160,