PORTER, T. FLINT MD, MPH
Pregnancy loss is a common problem found in women with bonafide antiphospholipid syndrome (APS). In recent years, there also has been considerable interest in extending the relation between reproductive failure and antiphospholipid (APL) antibodies and reproductive failure to infertility and failure of in vitro fertilization–embryo transfer (IVF–ET). This issue has become one that is hotly debated among authorities and patients alike. Some authorities have recommended that all infertile women, particularly those undergoing IVF–ET, be tested for APL antibodies, usually in a “panel” assay that typically includes 5 to 7 APL antibodies. 1,2 Making the assumptions that a positive result has known prognostic significance and that treatment is safe and efficacious, women are often treated with immunologic therapy that traditionally has been reserved for patients with APS. In fact, the scientific evidence linking APL antibodies to infertility is, at best, conflicting, and several recent studies have found no clinically meaningful relation. This chapter summarizes the available data concerning APL antibodies, infertility, and IVF–ET.
Antiphospholipid Antibodies and Infertility
The same APL-associated immunologic dysfunction that causes midtrimester and recurrent first trimester pregnancy loss has been proposed as a factor in even earlier pregnancy loss and heretofore unexplained infertility. 3 This immunologic dysfunction is postulated to inhibit fertilization, impair embryonic development, and sabotage implantation or postimplantation/placental development. To this end, an unknown stimulus is proposed to cause a general polyclonal B-cell activation, resulting in the circulation of a variety of autoantibodies that cause failure of reproduction at any of the aforementioned stages. 4 One proposed stimulus for APL antibody formation is pelvic tissue damage caused by various clinical entities such as endometriosis, pelvic inflammatory disease, and even IVF–ET itself. 5–7 The latter association has since been refuted. 8,9
The initial evidence linking APL antibodies and infertility comes largely from retrospective serologic studies of women with unexplained infertility. One of the earliest reported that APL antibodies found in the serum of women undergoing IVF–ET could be transferred into follicular fluid and subsequently could result in failed implantation. 10 In a later study, the same group performed 33 separate assays (including IgM, IgA, and IgG for antibodies against two phospholipids in addition to five histone and four polynucleotide antigens) on blood samples from women with unexplained infertility and from patients with unexplained pregnancy wastage. 4 A statistically greater proportion of those with unexplained infertility (23 of 26) were positive for at least one of the 33 antibodies. Whether the proportion of infertile women positive for APL antibodies (lupus anticoagulant [LAC], anticardiolipin [aCL], and antiphosphotidylserine [aPS]) reached statistical significance was unclear. Moreover, the control group was inappropriate because it included women with other reproductive problems as well as men.
At approximately the same time, Taylor et al 11 reported that a significant proportion of 41 women with unexplained infertility had serologic evidence of either the LAC or aCL antibodies. This study has been criticized because the control group included pregnant women. A third study confirmed an association between APL antibodies and “reproductive failure,” but it also was plagued by several problems. 12 First, women with unexplained infertility were grouped together with women with known causes of infertility (ie, endometriosis, recurrent miscarriage, polycystic ovaries, and premature ovarian failure) to attain statistical significance. Second, the control group consisted of only 15 women with unknown reproductive histories. Third, the 15 women in the control group were never actually compared with those in the infertility group. Finally, the patients were tested for a myriad of autoantibodies and “embryotoxic factors,” so that meaningful interpretation relating to APL antibodies was impossible.
More recent studies supporting the relation between APL antibodies and infertility have relied on so-called APL antibody panel assays. These include assays for IgG, IgM, and IgA isotypes for as many as eight different APL antibodies. With the exception of aCL assays, assays for other APL antibodies are difficult to interpret because none have been standardized or has an established positive threshold. Aoki et al 13 tested 65 women with unexplained infertility and 64 women with endometriosis (many of whom had recurrent pregnancy loss) for multiple APL antibodies and compared them with 97 women with infertility of known cause and without a history of recurrent pregnancy loss. They reported that more women with unexplained infertility and endometriosis were positive for two or more of either aCL, aPS, or antiphosphatidyinositol (aPI) than were those in the control group (5% vs 0%). The findings were difficult to apply clinically, however, because the sera were selected from a bank enriched with patients who had previously demonstrated positive autoimmune test results. Further, the control group consisted of other infertile women with known causes rather than women with demonstrated fertility. Curiously, the proportion of infertile patients with positive APL antibody test results differed markedly from that presented in a previous report by the same senior author. 4
Recognizing the problems of interassay variation associated with nonstandardized assays, Hatasaka et al 14 studied the seroprevalence of APL antibodies in women with unexplained infertility. Their approach minimized bias by blinding investigators to clinical outcomes and by including an appropriate control group comprised of women with documented fertility. In addition, they used a normalized assay system standardized in a manner similar to that of the accepted aCL assay. Using these methods, their results differed markedly from those of previous investigators: they reported no difference between infertile women and fertile women in mean normalized APL antibody optical densities or in the proportions of women with positive APL antibody tests.
Antiphospholipid Antibodies and In Vitro Fertilization– Embryo Transfer
Though the question of APL antibody-mediated infertility is far from settled, this association also has been proffered as a cause of IVF–ET failure. This hypothesis holds that APL antibodies interfere with implantation after IVF–ET, resulting in early pregnancy loss. 3 In keeping with that notion, Geva et al 2 reported a greater incidence of several circulating autoimmune antibodies, including APL antibodies, antinuclear antibodies (ANA), and antidouble-stranded DNA antibodies, in patients undergoing IVF–ET who experienced biochemical pregnancy loss. IgG aCL antibodies were detected in 14.2% of the study group, whereas no patients with successful IVF–ET were positive for aCL antibodies. The same group later reported that serologic positivity for one or more of seven autoantibodies was more common in 50 women (22%) who did not become pregnant after three or more IVF–ET cycles compared with 40 women with successful IVF–ET (2.5%) and 40 healthy nulligravidas (7.5%). 15 The investigators concluded that all women with multiple unsuccessful IVF–ET attempts should be tested for these various autoantibodies. Such broad recommendations would seem unwarranted given the small sample sizes in both studies and because none of the women in the second study 15 tested positive for LAC antibodies, and only three of the women with IVF–ET failure were positive for aCL antibodies.
In a similar study, Birkenfeld et al 3 measured LAC and aCL antibodies in women with documented tubal disease undergoing IVF–ET and found that 18 (32%) of 56 patients with failed IVF–ET were positive for one or both APL antibodies compared with none of the 14 women with successful implantation. These findings are difficult to interpret because this study also reported that previous ovulation induction, ovum retrieval, and IVF–ET may increase the likelihood of a positive test. It was unclear from the data whether the two comparison groups differed in these areas.
Not all investigators have found an association between APL antibodies and IVF–ET failure. In the earliest study, El Roeiy et al 10 reported that sera from 10 of 26 women undergoing IVF were positive for at least one of 11 autoantibodies, including aCL and aPS. A positive result, however, had no bearing on subsequent IVF–ET outcome (ie, the number of IVF–ET attempts, the success of oocyte retrieval, the number of abnormal oocytes, fertilization rates, cleavage rates, or pregnancy rates). This group later reported that no combination of 15 autoantibody tests, including APL antibodies, predicted the frequency of chemical or clinical pregnancy loss or success. 16
Similar results were reported by Nip et al 17 who evaluated IVF–ET outcome in women who were positive for a large number of autoantibodies, including APL antibodies, in both sera and follicular fluid. There was no relation between autoantibodies of any kind, in either sera or follicular fluid, and IVF–ET outcome, whether the suspected cause of infertility was endometriosis, tubal factor, or unexplained. Others support this view. Birdsall et al, 9 testing sera from 240 women undergoing IVF–ET for aCL and aPS antibodies, found no correlation between these antibodies and IVF–ET outcome. Kowalik et al 18 reported similar findings in a study of midfollicular sera from more than 500 women undergoing IVF–ET. Of these, 6.9% and 11.2% were positive for either IgG or IgM isotypes of aCL and aPS, respectively. Like El Roeiy et al 10 had found 10 years earlier, however, a positive APL antibody test had no bearing on overall IVF–ET outcome.
Proponents of APL antibody-associated immunologic dysfunction and its role in infertility have suggested that APL antibodies other than the LAC and aCL antibodies have an important role in IVF–ET failure. 19 In keeping with that notion, a positive test for any of three isotypes of several APL antibodies would predict IVF–ET outcome and subsequently impact therapy. Accordingly, Kaider et al 1 retrospectively tested sera from 42 women with failed IVF–ET and 42 women with successful IVF–ET for IgG, IgM, and IgM isotypes of seven different APL antibodies. They found that more women in the IVF–ET failure group (26%) had increased levels of APL antibodies than women in the IVF–ET success group (5%). It should be noted, however, that the most common positive test was for antiphosphatidylcholine (aPC), an antibody for which there is no standardized assay. In addition, the majority of positive tests were for the IgM isotype, which has been shown to be of little clinical significance in studies of women with bonafide APS. 20 Indeed, only three patients were positive for aCL antibodies in the study by Kaider et al. 1 Similarly, Coulam et al 19 reported that although 22% of women were positive for at least one of three isotypes of antibodies to seven phospholipids (21 separate assays), only 4% of women with IVF–ET failure were positive for aCL antibodies. The implication was that patients with “reproductive autoimmune failure syndrome” may be missed if only aCL antibody assays were performed.
Other investigators using similarly large APL panel assays have failed to find any relation between APL antibodies and IVF–ET failure. Denis et al 21 measured IgG, IgM, and IgA isotypes to seven phospholipid antigens in 793 women undergoing assisted reproduction for a variety of reasons. Pregnancy rates were similar (>60%) regardless of whether the women were positive for any number of the seven APL antibodies. Further, no definable threshold of positivity could be identified that predicted IVF–ET failure.
Proponents of APL-antibody reproductive dysfunction criticized Denis et al 21 because of laboratory methods and patient selection. Notwithstanding, reports from two other groups support their findings. 22,23 Porter et al 22 performed blinded assays for seven APL antibodies in 167 women undergoing IVF–ET, 144 fertile control subjects, and 51 women with APS. Only a positive test for antiphosphatidylethanolamine (aPE) was associated with infertility. When controlling for the contribution of other APL antibodies with logistic regression, no positive test for any of the seven APL antibodies predicted IVF–ET failure. Interestingly, aPE was the only APL antibody for which APS women were not positive. Chilcott et al 23 tested sera of 380 women for APL antibodies including the LAC, aCL, and aPS and anti-β2 glycoprotein I before IVF–ET. Although 23% of all women tested positive for at least one of the APL antibodies, there was no ad-verse effect on pregnancy rates, live birth rates, or birth weights. A meta-analysis of seven published studies came to similar conclusions. 24
Treatment of Antiphospholipid Antibody-Related Infertility
Several groups have suggested treatment of infertile women positive for APL antibody with therapies reserved typically for patients with known autoimmune disease (ie, APS). Based on nonrandomized trials, heparin, corticosteroids, low-dose aspirin, and even intravenous immune globulin (IVIG) currently are used in some centers, both before and after IVF–ET.
Birkenfeld et al 3 treated 15 women with a history of failed IVF–ET who were positive for APL antibodies with 10 mg of prednisone and low-dose aspirin daily starting 2 weeks before cycle initiation. The prednisone dose was doubled after pregnancy was achieved. Seven women were able to sustain pregnancies, but no control group was used for comparison. In a similar but larger study, Hasagewa et al 25 treated 307 women who were positive for antinuclear antibodies and/or APL antibodies with 10 mg of prednisone and low-dose aspirin. Pregnancy rates were 40% in treated women compared with 20% in the untreated group. From the data presented, however, it appeared there was no difference in pregnancy rates if women were positive for APL antibodies alone.
The studies by Sher et al 6,26 have received considerable attention, and APL antibody-related IVF–ET treatment regimens often are based on their findings. In the first report, Sher et al 6 treated women undergoing IVF–ET who were APL-antibody positive and who had documented pelvic pathology with heparin and aspirin starting the day of cycle initiation. Clinical pregnancies were achieved in 49% of treated patients compared with 16% of untreated women (P < 0.05). Whereas these findings are impressive, closer inspection of the data and study design reveal flaws that render them, at best, inconclusive. The most obvious problem is that the study was not randomized, and the investigators were not blinded to which patients were receiving treatment. Indeed, the method of selecting patients for treatment or nontreatment was not explained. Although the differences in pregnancy rates are statistically significant, the 49% pregnancy rate among the treated patients positive for APL is comparable with rates among women undergoing IVF–ET for pelvic pathology in other experienced centers. Also, the 16% pregnancy rate among nontreated women with pelvic pathology is lower than would be expected among such patients if APL antibody status were unknown. Finally, when the separate disease categories of the patients were taken into account (ie, endometriosis, pelvic inflammatory disease, or pelvic adhesions), no statically significant differences were found.
In a follow-up study, Sher et al 26 reported the selective use of heparin and aspirin in combination with in IVIG in another observational, nonrandomized, nonblinded study. During a 4-year period, 678 infertile women had positive tests for APL antibodies: 603 received heparin and low-dose aspirin, whereas only 84 women received no treatment. Live births occurred in 46% and 22%, respectively (P < 0.001). A significant difference is not surprising, given the marked disparity in sample sizes. In the next phase of their study, they compared the prevalence of IgG, IgM, and IgA isotypes of six different APL antibodies in the sera of women for whom IVF–ET had failed after treatment and in the sera of women with successful IVF–ET. They inexplicably combined the women positive for aPS with the women positive for aPE and reported that they had a lower live birth rate than women who were negative for either one of the antibodies (5% vs 12%, P < 0.001). Although statistically significant, this difference did not control for the significantly greater number of multiple gestations in the aPS/aPE-negative group. Importantly, the authors were unable to find a difference in birth rates between women positive for any APL antibody individually, including aPS and aPE, compared with women who tested negative. In the final phase of their study, treated women who did not achieve pregnancy in the first phase of the study had IVIG added to their treatment regimen. Only women positive for aPS and/or aPE benefited from treatment with IVIG; however, live birth rates before and after the addition of IVIG treatment were used for comparison rather than a control group of women who were negative for the APL antibodies in question.
The lack of scientific rigor and questionable designs of all of the previously mentioned reports would make the use of immunologic therapy in the treatment of women undergoing IVF–ET seem unwarranted. Several groups support this view. Schenk et al 27 treated APL antibody-positive IVF–ET patients with heparin and low-dose aspirin starting the day of oocyte retrieval and found no difference in the embryo implantation rates between the treated and untreated women (13% vs 8%). Further, neither clinical nor ultimate ongoing pregnancy rates differed according to APL antibody status. Kutteh et al 28 also used heparin and low-dose aspirin to treat 19 women undergoing IVF–ET who tested positive for aCL or other APL antibodies. Neither pregnancy rates (53% vs 47%) nor implantation rates (26% vs 19%) were different, regardless of whether women positive for APL antibody women received treatment.
Antiphospholipid Antibody-Mediated Infertility and Clinical Practice
In summary, there is some evidence that APL antibodies may be more prevalent among women with infertility when compared with fertile controls. For these tests to be clinically useful, they must impact outcome and have a scientifically validated treatment available for patients who test positive. To date, there is not adequate evidence to support either impact or treatment. Instead, appropriately designed studies, using standardized assays for APL antibodies associated with bonafide autoimmune disease, have failed to confirm the findings of observational, retrospective studies using various laboratory methods. In the case of treatment for APS antibodies in women undergoing IVF–ET, the evidence is confined to what at best can be described as anecdotal reports.
Unfortunately, many clinicians have adopted these treatments and offered them to infertile couples in their care. Indeed, the current application of immunologic treatment (ie, IVIG and/or heparin and aspirin) has preceded any understanding of possible immune-mediated mechanism(s) of infertility. In other areas of medicine, it would be deemed unacceptable to routinely use potentially harmful therapy in women with a given illness when there is no proven biologic basis for the disease and no epidemiologic evidence of the treatment’s efficacy. In the case of APS antibody-mediated infertility, the drive to increase pregnancy rates should not run the risk of jeopardizing the best interests of infertile women by subjecting them to unnecessary health risks and by needlessly expending their time, money, and emotional well-being.
1. Kaider BD, Price DE, Roussev RG, et al. Antiphospholipid antibody prevalence in patients with IVF failure. Am J Reprod Immunol. 1996; 35: 388–393.
2. Geva E, Yaron Y, Lessing JB, et al. Circulating autoimmune antibodies may be responsible for implantation failure in in vitro fertilization. Fertil Steril. 1994; 62: 802–806.
3. Birkenfeld A, Mukaida T, Minichiello L, et al. Incidence of autoimmune antibodies in failed embryo transfer cycles. Am J Reprod Immunol. 1994; 31: 65–68.
4. Gleicher N, El-Roeiy A, Confino E, et al. Reproductive failure because of autoantibodies: Unexplained infertility and pregnancy wastage. Am J Obstet Gynecol. 1989; 160: 1367–1385.
5. Gleicher N, el-Roeiy A, Confino E, et al. Is endometriosis an autoimmune disease? Obstet Gynecol. 1987; 70: 115–122.
6. Sher G, Feinman M, Zouves C, et al. High fecundity rates following antiphospholipid antibody seropositive women treated with heparin and aspirin. Hum Reprod. 1994; 9: 2278–2283.
7. Fisch B, Rikover Y, Shohat L, et al. The relationship between in vitro fertilization and naturally occurring antibodies: Evidence for increased production of antiphospholipid autoantibodies. Fertil Steril. 1991; 56: 718–724.
8. Fisch B, Fried S, Manor Y, et al. Increased antiphospholipid antibody activity in in vitro fertilization patients is not treatment-dependent but rather an inherent characteristic of the infertile state. Am J Reprod Immunol. 1995; 34: 370–374.
9. Birdsall MA, Lockwood GM, Ledger WL, et al. Antiphospholipid antibodies in women having in vitro fertilization. Human Reprod. 1996; 11: 1185–1189.
10. El-Roeiy A, Gleicher N, Friberg J, et al. Correlation between peripheral blood and follicular fluid autoantibodies and impact on in vitro fertilization. Obstet Gynecol. 1987; 70: 163–170.
11. Taylor PV, Campbell JM, Scott JS. Presence of autoantibodies in women with unexplained infertility. Am J Obstet Gynecol. 1989; 161: 377–379.
12. Roussev RG, Kaider BD, Price DE, et al. Laboratory evaluation of women experiencing reproductive failure. Am J Reprod Immunol. 1996; 35: 415–420.
13. Aoki K, Dudkiewicz AB, Matsuura E, et al. Clinical significance of B2-glycoprotein I-dependent anticardiolipin antibodies in the reproductive autoimmune failure syndrome: Correlation with conventional antiphospholipid antibody detection systems. J Obstet Gynecol. 1995; 172: 926–931.
14. Hatasaka H, Porter TF, Silver RM, et al. Antiphospholipid antibody levels are not elevated among women with tubal factor and unexplained infertility. Annual Proceedings of the American Society of Reproductive Medicine, 1997.
15. Geva E, Amit A, Lerner-Geva L, et al. Autoimmune disorders: Another possible cause for in vitro fertilization and embryo transfer failure. Hum Reprod. 1995; 10: 2260–63.
16. Gleicher N, Liu HC, Dudkiewicz A, et al. Autoantibody profiles and immunoglobulin levels as predictors of in vitro fertilization success. Am J Obstet Gynecol. 1994; 170: 1145–1149.
17. Nip MMC, Taylor PV, Rutherford AJ, et al. Autoantibodies and antisperm antibodies in sera and follicular fluids of infertile patients: Relation to reproductive outcome after in vitro fertilization. Hum Reprod. 1995; 10: 2564–2569.
18. Kowalik A, Vichnin M, Liu H-C, et al. Midfollicular anticardiolipin and antiphosphatidylserine antibody titers do not correlate with IVF outcome. Fertil Steril. 1997; 68: 298–304.
19. Coulam CB, Kaider BD, Kaider AS, et al. Antiphospholipid antibodies associated with implantation failure after IVF/ET. J Assist Reprod Genet. 1997; 14: 603–608.
20. Silver RM, Porter TF, van Leeuwen I, et al. Anticardiolipin antibodies: Clinical consequences of “low titers.” Obstet Gynecol. 1996; 87: 494–500.
21. Denis AL, Guido M, Adler RD, et al. Antiphospholipid antibodies and pregnancy rates and outcome in in vitro fertilization patients. Fertil Steril. 1997; 67: 1084–1090.
22. Porter TF, Branch DW, Silver RM, et al. Antiphospholipid antibodies are not related to the success of implantation in women undergoing in vitro fertilization–embryo transfer. Obstet Gynecol. 2000; In press.
23. Chilcott IT, Margara R, Cohen H, et al. Pregnancy outcome is not affected by antiphospholipid antibody status in women referred for in vitro fertilization. Fertil Steril. 2000; 73: 526–530.
24. Hornstein MD, Davis OK, Massey JB, et al. Antiphospholipid antibodies and in vitro fertilization success: A meta-analysis. Fertil Steril. 2000; 73: 330–333.
25. Hasegawa I, Yamanoto Y, Suzuki M, et al. Prednisolone plus low-dose aspirin improves the implantation rate in women with autoimmune conditions who are undergoing in vitro fertilization. Fertil Steril. 1998; 70: 1044–1048.
26. Sher G, Matzner W, Feinman M, et al. The selective use of heparin/aspirin therapy, alone or in combination with intravenous immunoglobulin G, in the management of antiphospholipid antibody-positive women undergoing in vitro fertilization. Am J Reprod Immunol. 1998; 40: 74–82.
27. Schenk LM, Butler L, Morris JP, et al. Heparin and aspirin treatment yields higher implantation rates in IVF patients with antiphospholipid antibody seropositivity compared to untreated seronegative patients (abstract). Proceedings of the Annual Meeting of the American Society of Reproductive Medicine, Boston, Massachusetts, 1996.
28. Kutteh WH, Yetman DL, Chantilis SJ, et al. Effect of antiphospholipid antibodies in women undergoing in vitro fertilization: Role of heparin and aspirin. Hum Reprod. 1997; 12: 1171–1175.
© 2001 Lippincott Williams & Wilkins, Inc.