Cumulative pregnancy and live birth rates in women with antiphospholipid antibodies undergoing assisted reproduction

T. Eldar-Geva1,2,3, C. Wood1, N. Lolatgis1, L. Rombauts1, G. Kovacs1, J. Fuscaldo1 and A.O. Trounson2

1 Monash IVF, Epworth Hospital, Richmond 3121, 2 Centre for Early Human Development, Monash University, Clayton 3168, Victoria, Australia


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The aims of this study were to investigate the influence of antiphospholipid antibodies (APA) on cumulative pregnancy and live-birth rates in patients undergoing assisted reproductive treatment. Serum samples from 173 patients were collected prior to initiation treatment cycle and tested by enzyme-linked immunosorbent assay (ELISA) for the presence of immunoglobulin (Ig)G, IgM and IgA against cardiolipin, phosphoserine, phosphoethanolamine, phosphoinositol, phosphatidic acid, and phosphoglycerol. Fifty-six samples from patients who had at least two failed cycles by assisted reproductive treatment were also tested by a bioassay for the presence of lupus anticoagulants. Both cumulative pregnancy and live birth rates were not affected by the presence of any specific or any number of seropositive APA. There was no association between multiple assisted reproductive treatment failures and APA seropositivity. Neither the serum concentration of any of the 18 APA, nor the number of positive APA was correlated with the number of assisted reproductive treatment failed cycles or affected the probability of pregnancy. No patient was found to be positive for lupus anticoagulants. Using life table analyses, which has been recognized as the most appropriate method available to analyse assisted reproductive treatment results, we conclude that there is no relationship between circulating APA and assisted reproductive treatment outcome. APA do not affect the early process of implantation or maintenance of pregnancy among assisted reproductive treatment patients.

Key words: antiphospholipid antibodies/assisted reproductive treatment/cumulative live birth rate/cumulative pregnancy rate/implantation failure


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Identifying the cause, and recommendation for treatment for infertile patients who have had multiple assisted reproduction treatment failures is a challenge. Low implantation rates have been attributed to a number of factors including the quality of gametes or embryos, altered endocrine milieu, the presence of uterine or endometrial pathology and autoimmunity. Antiphospholipid antibodies (APA) are a group of organ non-specific autoantibodies that bind to negatively charged phospholipids. Their presence has been associated with reproductive failure, the most consistently reported phenomenon is the association between recurrent spontaneous abortion and the presence of immunoglobulin (Ig) G anticardiolipin and lupus anticoagulant (LAC) (Cowchock et al., 1986Go, Lockwood et al., 1986Go; Gleicher et al., 1989Go). Although there is no proven mechanism for fetal loss, treatment of APA-positive pregnant women with heparin and aspirin improves pregnancy outcome in women with recurrent pregnancy loss (Branch et al., 1992Go; Cowchock et al., 1992Go; Kutteh, 1996Go; Rai and Reagen, 1997).

Women who present for in-vitro fertilization (IVF) have been shown to have a high prevalence of APA, especially those with pelvic pathology such as endometriosis and pelvic adhesions (El-Roeiy et al., 1987Go; Fisch et al., 1991Go; Sher et al., 1994Go; Dmowski et al., 1995Go; Nip et al., 1995Go; Schenk et al., 1996Go). However, the relevance of the presence of APA to IVF outcome is still a matter of debate. A study by Sher et al. generated much interest because of its claim that IVF failure is positively correlated with APA seropositivity (Sher et al., 1994Go). Further, Sher et al. (1994) claimed that high fecundity rates could be achieved in seropositive women treated with aspirin and heparin. Some studies have supported these findings (Birkenfeld et al., 1994Go; Dmowski et al., 1995Go; Geva et al., 1995Go; Kaider et al., 1996Go) but others have found no significant association between APA seropositivity and IVF failure (Gleicher et al., 1994Go; Nip et al., 1995Go; Birdsall et al., 1996Go; Kowalik et al., 1997Go; Kutteh et al., 1997Go). The conclusions arrived at in these studies have been viewed with some scepticism because of the use of non-randomized small patient numbers, non-comparable controls, determining small number of APA epitopes or the inappropriate definition of positive results. More recently, Denis et al. published results of a large study where 793 patients have been tested for the presence of 21 APA epitopes (Denis et al., 1997Go). Using various statistical methods, they demonstrated that elevated APA concentrations, at any cut-off point, were not associated with a change in IVF cycle outcome.

The aims of our study were to compare cumulative pregnancy and live birth rates for APA seronegative and seropositive patients and to determine whether there is a correlation between multiple IVF failure and the presence of any or a specific APA. To accomplish this, we tested sera from a population of IVF patients for the presence of IgG, IgM and IgA against six specific phospholipids: anticardiolipin, antiphosphoserine, antiphosphoethanolamine, antiphosphoinositol, antiphosphatidic acid, and antiphosphoglycerol. To avoid biases resulting from the many different methods available to determine APA status, the serological determinations were carried out by a laboratory specialized in reproductive immunology (Matzner et al., 1994Go). This group of investigators have published previously their APA results in a large group of infertile patients (Sher et al., 1994Go). In addition to the above immunoassays, we used a bioassay to determine serum concentrations of LAC.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Study design
Serum samples for this study were obtained routinely from patients attending assisted reproductive treatment treatment in Monash IVF between June 1996 and March 1997. Blood was collected prior to cycle initiation (treatment cycle number: median: 2; range: 1 to 12) and serum samples were kept at –20°C. Samples from 173 patients, aged 21–40 years, who had transfer of at least two embryos or eggs, were tested anonymously for the presence of 18 different APA. Monash IVF computerized database was searched for the patients' previous cycles. Cumulative pregnancy and live birth rates were calculated in various patient groups according to their APA status. We further analysed the results in patients with multiple IVF failure, compared to those who had successful treatment. The study group (n = 96) included patients who had at least two failed IVF-embryo transfer or gamete intra-Fallopian transfer (GIFT) cycles with at least two embryos or eggs transferred each time. GIFT cycles were included only if the fertilization rate of the extra eggs (>=2 eggs) was >=30%. The patients were further subdivided into two groups: group I (n = 67) included women with two to four failed assisted reproductive treatment cycles, and group II (n = 29) included women with five or more failed assisted reproductive treatment cycles. The control group (n = 45) included patients who became pregnant within their first two IVF cycles. APA results for 32 patients did not fulfil the selected criteria required to be included in any of the groups (such as patients who have had one failed cycle or those who conceived in their fourth treatment cycle). Their APA results were included only in the life table analysis. Fifty-six blood samples were tested within 2 h after collecting for the presence of LAC.

The study was approved by the research and ethics committee of Epworth Hospital, Richmond, Melbourne. Informed consent was obtained from all participants.

APA assay
Serum samples were tested for the presence of IgG, IgM and IgA against cardiolipin, phosphoserine, phosphoethanolamine, phosphoinositol, phosphatidic acid and phosphoglycerol, using an enzyme-linked immunosorbent assay (ELISA), as previously described (Matzner et al., 1994Go). For each APA epitope separately, seronegativity was defined as <2 SD above the mean and seropositivity was defined as >3 SD above the mean. All other results were defined as borderline.

Lupus anticoagulant was measured using dilute Russell's viper venom test (DRVVT) (LA-SCREEN and LA-CONFIRM kits, Gradipore, North Ride, NSW, Australia). Negative results were defined as 100–120% of the internal control.

Treatment protocols
Ovarian stimulation protocols used in our institute have been previously described (MacLachlan et al., 1989Go). Briefly, patients were down-regulated using a gonadotrophin-releasing hormone agonist Nafarelin 0.5 mg/day (Synarel; Searle, Sydney, NSW, Australia), starting in the luteal phase (long protocol) or on day 2 of the follicular phase (flare or boost protocol). Multiple follicular development was initiated using individually adjusted doses of purified urinary follicle-stimulating hormone (Metrodin; Serono, Melbourne, Victoria, Australia). The ovarian response was monitored with serum oestradiol concentrations and transvaginal ultrasonography. Patients were given 5000 IU human chorionic gonadotrophin (HCG) i.m. (Profasi; Serono) when at least three follicles of >17 mm were present and serum oestradiol was appropriately rising. Oocytes were retrieved transvaginally 36 h later under general anaesthesia. In GIFT cycles, one to four oocytes were transferred laparoscopically to one tube together with 1.5x105 motile spermatozoa purified on a Percoll density gradient. Routine IVF or intracytoplasmic sperm injection (ICSI) and embryo culture were used as indicated and one to four embryos were transferred 48–72 h after egg retrieval.

Treatment outcome
Failed cycles were categorized as blood ßHCG < 25 mIU/ml 16 days after oocyte recovery. Pregnancy was categorized as blood ßHCG > 25 mIU/ml 16 days after oocyte recovery with at least one increased value at 5 days later. Biochemical pregnancy was defined as ßHCG <1000 mIU/ml with no gestational sac visible on vaginal ultrasound. Early spontaneous abortion was defined as fetal loss during the first trimester, and late abortion as fetal loss during the second trimester.

Statistical analysis
Data analysis was carried out using several different methods. Cumulative pregnancy and live-birth rates in seropositive and seronegative patients were estimated by the Kaplan–Meier procedure. Differences between groups were evaluated using {chi}2 test and Mann–Whitney U test were appropriate. Cox's regression procedure was used to estimate the `risk ratio' of pregnancy with the number of cycles as a dependent variable by each APA epitope individually and by the number of positive epitopes. Chi-square tables and logistic regression models were used to calculate odds ratios of pregnancy for each APA epitope and for the number of positive epitopes. The same methods were used to analyse the association between APA status and probability of biochemical pregnancy, early and late abortions and live birth.

Analysis of the results were performed using the SAS version 6.2 computer analysis program. P < 0.05 was considered as statistically significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
One hundred and seventy three serum samples were tested for the presence of 18 APA–IgG, IgM and IgA against cardiolipin, phosphoserine, phosphoethanolamine, phosphoinositol, phosphatidic acid, and phosphoglycerol. Eighty-one patients (46.8%) had seronegative results for all 18 epitopes tested, namely, all APA concentrations were <2 SD above the mean of the normal controls. Sixty-four patients (37.0%) had at least one positive result (>3 SD above the mean). Twenty-eight patients (16.2%) had at least one borderline result (between 2 SD and 3 SD above the mean) with no positive results.

The cumulative pregnancy and live birth rates per transfer for APA seropositive and seronegative patients are presented in Tables I and IIGoGo respectively. No statistically significant difference could be found between the two groups. We further analysed the cumulative pregnancy and live birth rates for patients with different numbers of positive APA. For patients with >=2 positive epitopes the cumulative pregnancy rate per transfer was 18.5, 35.5, 41.4, 48, 48, and 48% for cycles 1–6 respectively. For patients with >=3 positive epitopes the cumulative pregnancy rate per transfer was 21.4, 29.3, 43.3, 57.5, and 57.5% for cycles 1–6 respectively. The number of positive or positive and borderline results (maximum 7 positive APA epitopes) was not associated with decreased cumulative pregnancy or live birth rates.


View this table:
[in this window]
[in a new window]
 
Table I. Cumulative pregnancy rate per transfera for APA seronegative, borderline, and seropositive patients
 

View this table:
[in this window]
[in a new window]
 
Table II. Cumulative live birth rates per transfera for APA seronegative, borderline, and seropositive patients
 
Pregnancy outcomes for patients with APA seronegative (all APA epitopes <2 SD above the means), APA seropositive (at least one epitope >3 SD above the mean) and borderline results (at least one epitope >2 SD above the mean) are presented in Table IIIGo. There was no statistically significant difference between the groups. Patients with borderline or positive APA were more likely to experience a first-trimester abortion compared to seronegative patients (45, 36.6 and 27% of the pregnancies respectively); however, this difference was not statistically significant. In the seronegative group, four patients had two miscarriages each and one patient had three consecutive second-trimester miscarriages. In the seropositive group four patients had two miscarriages each.


View this table:
[in this window]
[in a new window]
 
Table III. Pregnancy outcome for seronegative and seropositive patients receiving assisted reproductive treatment
 
No lupus anticoagulant (LAC) was found in the blood of 54 of the 56 tested samples. Two patients had borderline results (120% of the internal control): one patient had normal pregnancy and delivery after IVF embryo transfer in 1992 and two failed cycles in 1996–7. This patient had five positive IgG APA–anticardiolipin, antiphosphoserine, antiphosphoethanolamine, antiphosphoinositol and antiphosphatidic acid. The second patient had 2 failed cycles. She had 2 positive APA–IgG antiphosphoserine and IgA antiphosphatidic acid. Of the 54 negative LAC, 24 were APA seropositive and 32 were APA seronegative.

Demographic and clinical data for the control group, group I and group II are presented in Table IVGo. No statistically significant difference between the groups was found with respect to patient's age and the number of eggs/embryos transferred per cycle. The diagnosis of tubal infertility was more frequent in group I compared to the control group (P < 0.05). The proportions of patients who had the flare versus the long down-regulation protocol for ovarian stimulation were equally distributed in all groups.


View this table:
[in this window]
[in a new window]
 
Table IV. Demographic and clinical data of the control group (pregnant within their first 2 cycles), group I (2–4 failed assisted reproductive cycles), and group II (>=5 failed assisted reproductive cycles)
 
Borderline or seropositive results (>2 SD above the mean) for at least one of the 18 APA epitopes were found in 24 patients (53.3%) in the control group, 37 patients (55.2%) in group I and 17 patients (58.6%) in group II (statistically not significant). No significant difference was found between the groups with respect to the results of any of the 18 APA epitopes (Table VGo), the total number of positive epitopes (Table VIGo), or the total number of borderline and positive epitopes (data not shown). A comparison between the whole study group (groups I and II together) and the control groups showed no difference with respect to any of the above mentioned parameters.


View this table:
[in this window]
[in a new window]
 
Table V. Seropositive results for each APA epitope in the various groups
 

View this table:
[in this window]
[in a new window]
 
Table VI. Antiphospholipid antibodies (APA) serology status in patients who conceived within their first two assisted reproductive treatment attempts (control group), patients with 2–4 failed assisted reproductive treatment cycles (group I) and patients with >=5 failed assisted reproductive treatment cycles (group II)
 
We used Cox's regression procedure to estimate the odds ratio of pregnancy with the number of cycles as a dependent variable by each APA epitope separately, by the number of positive and borderline and positive epitopes and for none versus one and two versus >=3 positive and borderline and positive epitopes. None of the above significantly affected the probability of pregnancy, biochemical pregnancy, early or late abortions and live birth. Odds ratios were between 0.4–2.1.

Association between the presence of APA and of pelvic pathology (endometriosis or tubal disease) was assessed. Of the 79 patients with pelvic pathology 28 (34.4%) were seronegative and 37 (46.8%) were seropositive. Of the 91 patients without pelvic pathology 53 (58.2%) were seronegative and 27 (29.7%) were seropositive (P < 0.01, {chi}2 test).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The role of antiphospholipid antibodies (APA) in implantation failure has been the focus of many recent research efforts and is still a mater of debate. Our results demonstrate that cumulative pregnancy and live birth rates are not affected by the presence of any specific or any number of positive APA. There is no correlation between multiple assisted reproductive treatment failures and APA seropositivity. Neither the serum concentration of any of 18 different APA, nor the number of positive APA was correlated with the number of assisted reproductive treatment failed cycles or affected the probability of pregnancy. In addition, by using a bioassay to determine the concentrations of lupus anticoagulant (LAC), we showed that the occurrence of LAC is a rare phenomenon in patients with assisted reproductive treatment failure without recurrent miscarriage, regardless of their APA status.

All of the APA studies to date have compared pregnancy and implantation rates in a cohort of assisted reproductive treatment patients with various APA results. However, live table analyses, namely cumulative pregnancy and life birth rates, have been recognized as the most appropriate method available to present success rates for assisted reproductive treatment (Guzick and Bross, 1992Go; Hull, 1992Go). This is the first study to use life table analyses to evaluate a possible correlation between APA status and assisted reproductive treatment outcome.

In the study by Sher et al. all serum samples in our study were tested by the Reproductive Immunology Associates Laboratory for the presence of 18 APA and the same normal controls were used (Sher et al., 1994Go). In contrast with previous findings (Sher et al., 1994Go), we found no correlation between the presence of these APA and assisted reproductive treatment outcome. Our results support recent works (Denis et al., 1997Go; Kowalik et al., 1997Go; Kutteh et al., 1997Go). By using life table analyses we add strength to the conclusion that there is no relationship between circulating APA and the early process of implantation or maintenance of pregnancy among assisted reproductive treatment patients.

Our results are perhaps not surprising, given that the ability to achieve a pregnancy is not impaired in women with the full-blown antiphospholipid antibody syndrome (recurrent miscarriage, thrombosis and thrombocytopenia). If the presence of APA could prevent or damage the implantation process, we would expect the frequency of infertility to be higher in this group of patients than in the normal population. To the best of our knowledge there is no evidence in the literature supporting this association.

The association between recurrent miscarriage and APA is a consistently reported feature of primary antiphospholipid syndrome (Asherson et al., 1989Go; Rai et al., 1995Go). APA, especially IgG anticardiolipin and LAC, are found in 15% of women with recurrent miscarriage. Our study demonstrated that the rate of first trimester miscarriage was higher in patients having positive and/or borderline APA (45 and 36.6% respectively), compared with APA-negative patients (27%). This difference was not statistically significant. Similar results have been found previously (Kowalik et al., 1997Go). Other investigators found no relationship between APA concentrations and pregnancy loss rates in IVF patients (Birdsall et al., 1996Go; Denis et al., 1997Go; Balasch et al., 1998Go). In our study only one patient had recurrent (three consecutive) miscarriages and no patients were found to be positive for LAC. These findings suggest that although LAC may be important in recurrent fetal loss, it appears to have no significance in recurrent implantation failure.

Transitory positive APA tests are a common occurrence and are often due to viral or other infections present at the time of testing (Rai et al., 1995Go). In a study of 500 consecutive women with recurrent miscarriage, only 66% of those with an initial positive test for LAC, 37% of those who were IgG anticardiolipin positive and 36% who were IgM anticardiolipin positive had a repeat positive test (Rai et al., 1995Go). In our study and in all others referred to, serum APA concentrations were measured once only. In order to exclude those with only transitory positive tests, repeat testing at least 8 weeks after the initial test is necessary to confirm the presence of APA in patients with an initially positive result. Until such a comprehensive study is conducted we cannot know the real importance of APA in infertility.

In agreement with Sher et al. (1994), we found that the prevalence of APA was significantly higher in patients with organic pelvic disease than in patients without pelvic pathology (respectively 46.8 and 29.7% in our study and 53 and 14% in Sher's study) (Sher et al., 1994Go). Other studies found no difference in the prevalence of anticardiolipin and antiphosphatidylserine between patients with tubal factor infertility, male factor infertility and idiopathic infertility (Birdsall et al., 1996Go; Kowalik et al., 1997Go). Enlarging the tested APA panel may account for the difference between the studies. It is possible that the tissue damage in pelvic diseases such as endometriosis or pelvic inflammatory disease predisposes the occurrence of autoantibodies such as APA. However, their occurrence does not predispose assisted reproductive treatment failure.

Five groups have performed non-randomized treatment of APA-positive women undergoing assisted reproductive treatment (Birkenfeld et al., 1994Go; Sher et al., 1994Go; Schenk et al., 1996Go; Kutteh et al., 1997Go; Rubinstein et al., 1997Go; Sher, 1997Go). In a small group of patients no significant benefit of heparin and aspirin treatment in APA-positive women undergoing IVF was demonstrated (Kutteh et al., 1997Go). The four other groups of investigators reported improved outcome after treatment with aspirin, heparin or glucocorticosteroids. It is still possible that there is a benefit to this empiric treatment; however, this benefit presumably would be achieved through other mechanisms not related to APA status. Aspirin, a cyclo-oxygenase synthesis inhibitor, might influence other systems involving prostaglandins. For example, previous studies (Nye et al., 1997Go) found that aspirin could inhibit hypothalamic–pituitary–adrenal response to some specific stimulations. Taking into account the well known severe side-effects of long lasting anticoagulation treatment, we cannot support routine treatment with different combinations of these agents at this time. Randomized prospective controlled trials to investigate the efficiency of these agents, regardless of APA status, are certainly needed.


    Acknowledgments
 
We thank Prof. Eng Gan and Erica Millan who tested the blood samples for the presence of lupus anticoagulant. We thank Dr Philip McCloud and Darshi Arachige from the Department of Mathematics, Monash University, for the statistical analysis of the data. We also thank all the clinical, nursing, embryology, and administration staff of Monash IVF for their invaluable help. The study was supported by a research grant to C.Wood by Monash IVF Ltd.


    Notes
 
3 To whom correspondence should be addressed Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Asherson, R.A., Khamashta, M.A., Ordi Ross, J. et al. (1989) The `primary' antiphospholipid syndrome: major clinical and serological features. Medicine Baltimore, 68, 366–374.[ISI][Medline]

Balasch, J., Creus, M., Fabregues, F. et al. (1998) Antiphospholipid antibodies and the outcome of pregnancy after the first in-vitro fertilization and embryo transfer cycle. Hum. Reprod., 13, 1180–1183.[Abstract]

Birdsall, M.A., Lockwood G.M., Ledger, W.L., Johnson, P.M. and Chamley L.W. (1996) Antiphospholipid antibodies in women having in-vitro fertilization. Hum. Reprod., 11, 1185–1189.[Abstract]

Birkenfeld, A., Mukaida, T., Minichiello, L., Jackson, M., Kase, N.G. and Yemini, M. (1994) Incidence of autoimmune antibodies in failed embryo transfer cycles. Am. J. Reprod. Immunol., 31, 65–68.[ISI][Medline]

Branch, D.W., Silver, R.M., Blackwell, J.L., Reading, J.C. and Scott, J.R. (1992) Outcome of treated pregnancies in women with antiphospholipid syndrome: an update of the Utah experience. Obstet. Gynecol., 4, 614–620.

Cowchock, S., Smith, J.B. and Gocial, S. (1986) Antibodies to phospholipids and nuclear antigens in patients with related abortions. Am. J. Obstet. Gynecol., 155, 1002–1010.[ISI][Medline]

Cowchock, S., Reece, E.A., Balaban, D., Branch, D.W. and Plouffe, L. (1992) Repeated fetal losses associated with antiphospholipid antibodies: a collaborative randomized trial comparing prednisone to low-dose heparin treatment. Am. J. Obstet. Gynecol., 166, 1318–1323.[ISI][Medline]

Denis, A.L., Guido, M., Adler, R.D., Bergh, P.A., Brenner, C. and Scott, R.T. (1997) Antiphospholipid antibodies and pregnancy rates and outcome in in vitro fertilization patients. Fertil. Steril., 67, 1084–1090.[ISI][Medline]

Dmowski, W.P., Rana, N, Michalowska, J., Friberg, J., Papierniak, C. and El-Roeiy A. (1995) The effect of endometriosis, its stage and activity, and of autoantibodies on in vitro fertilization and embryo transfer success rates. Fertil. Steril., 63, 555–562.[ISI][Medline]

El-Roeiy, A. Gleicher N., Friberg, J. et al. (1987) Correlation between peripheral blood and follicular fluid autoantibodies and impact on in vitro fertilization. Obstet. Gynecol., 70, 163–170.[Abstract]

Fisch, B., Rikover, Y., Shohat, L. et al. (1991) The relationship between in vitro fertilization and naturally occurring antibodies: evidence for increased production of antiphospholipid autoantibodies. Fertil. Steril., 56, 718–724.[ISI][Medline]

Geva, E., Amit, A., Lerner-Geva, L., Azem, F., Yovel, Y. and Lessing, J.B. (1995) Autoimmune disorders: another possible cause for in vitro fertilization and embryo transfer failure. Hum. Reprod., 10, 101–104.

Gleicher, N., El-Roeiy, A., Confino, E. and Friberg, J. (1989) Reproductive failure because of autoantibodies: unexplained infertility and pregnancy wastage. Am. J. Obstet. Gynecol., 160, 1376–1385.[ISI][Medline]

Gleicher, N., Liu, H.C., Dudkiewicz, A. et al. (1994) Autoantibody profiles and immunoglobulin concentrations as predictors of in vitro fertilization success. Am. J. Obstet. Gynecol., 170, 1145–1149.[ISI][Medline]

Guzick, D.S. and Bross, D. (1992) Convenient numerical procedures for estimating cumulative pregnancy curves. Fertil. Steril., 57, 85–91.[ISI][Medline]

Hull, M.G.R. (1992) Infertility treatment: relative effectiveness of conventional and assisted conception methods. Hum. Reprod., 7, 785–796.[Abstract]

Kaider, B., Price, D.E., Rousev, R.G. and Coulan, C.B. (1996) Antiphospholipid antibody prevalence in patients with IVF failure. Am. J. Reprod. Immunol., 35, 388–393.[ISI][Medline]

Kowalik, A., Vichnin, M., Liu, H-C., Branch, W. and Berkeley, A. (1997) Mid-follicular anticardiolipin and antiphosphatidylserine antibody titers do not correlate with in vitro fertilization outcome. Fertil. Steril., 68, 298–304.[ISI][Medline]

Kutteh, W.H. (1996) Antiphospholipid antibody-associated recurrent pregnancy loss: treatment with heparin and low dose aspirin is superior to low dose aspirin. Am. J. Obstet. Gynecol., 174, 1584–1589.[ISI][Medline]

Kutteh, W.H., Yetman, D.L., Chantilis, S.J. and Crain, J. (1997) Effect of antiphospholipid antibodies in women undergoing in vitro fertilization: role of heparin and aspirin. Hum. Reprod., 12, 1171–1175.[ISI][Medline]

Lockwood, C.J., Reece, E.A., Romero, R. and Hobbins, J.C. (1986) Antiphospholipid antibody and pregnancy wastage. Lancet, ii, 742–743.

MacLachlan, V., Besanko, M., O'Shea, F. et al. (1989) A controlled study of luteinizing hormone releasing hormone agonist (buserelin) for the induction of folliculogenesis before in vitro fertilization. N. Engl. J. Med., 320, 1233–1237.[Abstract]

Matzner, W., Chong, P., Xu, G. and Ching, W. (1994) Characterisation of antiphospholipid antibodies in women with recurrent spontaneous abortions. J. Reprod. Med., 39, 27–30.[ISI][Medline]

Nip, M.M.C., Taylor, P.V., Rutherford, A.J. and Hanock, K.W. (1995) Autoantibodies and antisperm antibodies in sera and follicular fluids of infertile patients; relation to reproductive outcome after in-vitro fertilization. Hum. Reprod., 10, 2564–2569.[Abstract]

Nye, E.J., Hockings, G.I., Grice, J.E. et al. (1997) Aspirin inhibits vasopressin-induced hypothalamic–pituitary–adrenal activity in normal humans. J. Clin. Endocrinol. Metab., 82, 812–817.[Abstract/Free Full Text]

Rai, R.S., Regan, L., Clifford, K. et al. (1995) Antiphospholipid antibodies and ß-2 glycoprotein-I in 500 women with recurrent miscarriage: results of a comprehensive screening approach. Hum. Reprod., 10, 2001–2005.[Abstract]

Rai, R. and Regen, L. (1997) Antiphospholipid antibodies, infertility and recurrent miscarriage. Curr. Opin. Obstet. Gynecol., 9, 279–282.[ISI][Medline]

Rubinstein, M., Marazzi, A. and Polak de Fried, E. (1997) Does low dose aspirin treatment improve implantation and pregnancy rates in IVF patients? (abstract). 10th World Congress on IVF and Assisted Reproduction, Vancouver, Canada.

Schenk, L.M., Butler, L., Morris et al. (1996) Heparin and aspirin treatment yields higher implantation rates in IVF patients with antiphospholipid antibodies seropositivity compared to untreated seronegative patients (abstract). Proceedings of the Annual Meeting of the American Society of Reproductive Medicine, Boston, USA.

Sher, G. (1997) A rational basis for selective immunotherapy in IVF patients testing seropositive to autoantibodies (abstract). 10th World Congress on IVF and Assisted Reproduction, Vancouver, Canada.

Sher, G., Feinman, M., Zouves, C. et al. (1994) High fecundity rates following in-vitro fertilization and embryo transfer in antiphospholipid antibody seropositive women treated with heparin and aspirin. Hum. Reprod., 9, 2278–2283.[Abstract]

Submitted on October 30, 1998; accepted on February 10, 1999.