Antiphospholipid antibodies and antibodies to tissue factor pathway inhibitor in women with implantation failures or early and late pregnancy losses

Abstract

The association of antiphospholipid antibodies (aPL) with early recurrent spontaneous abortions and fetal deaths is well established [1,2]. A role for aPL as a possible cause of failure to achieve pregnancy after in vitro fertilization (IVF), however, is controversial. An association between the hypercoagulable state because of the presence of aPL and unsuccessful embryo implantation has been observed in some [3], but not in other studies [4–6]. Pregnancy complications constitute one of the two major clinical criteria of the antiphospholipid syndrome (APS) [7] and adverse pregnancy outcomes may result from poor placental perfusion because of localized thrombosis. Recently, we found that high titers of antibodies to tissue factor pathway inhibitor (TFPI) (anti-TFPI) in women with autoimmune aPL seem to increase the risk of aPL-related reproductive failures [8]. TFPI is aKunitz-type protease inhibitor that tightly regulates tissue factor-mediated coagulation. During normal pregnancy, some of the hemostatic changes occurring in the placenta are characterized by increased tissue factor expression and low expression of TFPI. Both tissue factor and TFPI seem to be essential for the maintenance of hemostasis in the placenta [9]. The objective of the present study was to evaluate the presence of aPL [lupus anticoagulant (LA) andmoderate or high titers of anticardiolipin antibodies (aCL)] and anti-TFPI in women with a history of pregnancy loss, as well as in women with recurrent IVF failures. The study included patients referred to our institution in Argentina for evaluation of aPL because of a history of two or more consecutive spontaneous abortions (early pregnancy loss before 10 weeks of gestation), at least one fetal death (late pregnancy loss at or beyond 10 weeks of gestation) or recurrent (two ormore) IVF failures despite good visual quality embryos. Patients were referred after excluding other common etiologies of pregnancy failures (infections and hormonal, metabolic, uterine anatomic or genetic abnormalities). There were 243 women (median age 32 years old, range 21–37) with early and/ or late pregnancy losses (98 with early, 116 with late, and 29 with both), and 48 (median age 33 years old, range 23–39) with IVF failures. Themedian number of pregnancy losses was three (range 2–7) in the early group and two (range 1–5) in the late group, and the median number of IVF procedures was three (range 2–6) in the IVF failure group. A group of 80 normal control women (median age 35 years old, range 25–42) was also evaluated. They had had only successful pregnancies. None of the patients or control women had a history of thrombosis or systemic lupus erythematosus. In all cases, blood collection took place at least 3 months after pregnancy complications and if aPL were positive a second blood sample was taken 3 or 6 months later. LA activity was identified through at least three different screening tests, mixing studies and confirmatory procedures according to the ISTHguidelines. aCL (IgG and IgM isotypes) were measured by using a standardized in-house ELISA and titers above 20 GPL or MPL units were considered to be a positive result. The in-house ELISAs for IgG and IgM antibodies to b2 glycoprotein I (antib2GPI) and to prothrombin (anti-PT) were performed as previously reported using electron beam(100 kGy) and c-irradiated microtiter plates, respectively. The cut-off values (10 arbitrary units for IgG or IgM) were previously determined as the 99th percentiles of 95 normal sera [10]. Antibodies to TFPI (IgG and IgM isotypes) were detected in patients sera as recently described using recombinant full-length TFPI coated on c-irradiated microtiter plates [8]. Results were expressed as U mL, referred to internal standards arbitrarily fixed at 100 U mL. The 99th percentiles in 70 healthy controls were previously chosen as the cut-off points (18 U mL and 15 U mL for IgG and IgM anti-TFPI, respectively). Of the 80 control women, positive aCL was found in only one subject (1.2%). Similarly, only two of 48 (4.2%) patients with IVF failures had aCL. None of them had either LA or anti-b2GPI and/or anti-PT. On the other hand, 43 of the 243 (17.7%) women with pregnancy loss had persistent classical aPL. LAwas positive in 32; aCL in 37 and 34 women had antib2GPI and/or anti-PT. All women with positive anti-b2GPI and/or anti-PT had LA and/or aCL. The prevalence of aPL Correspondence: Ricardo R. Forastiero, Hematologia, Universidad Favaloro, Solis 453, (C1078AAI) Buenos Aires, Argentina. Tel.: +54 11 4378 1145; fax: +54 11 4378 1311; e-mail: forastiero@favaloro.edu.ar