Characteristics of autoantibodies associated with recurrent pregnancy loss

Pregnancy loss is one of the most important issues encountered by obstetricians and comprising 15–20% of pregnancy outcomes [1]. The most important cases of reproductive loss are disorders involving immune mechanisms. One example of such autoimmune disorders is the antiphospholipid syndrome (APS) that is associated with the persistent presence of antiphospholipid antibodies and the thrombophilic complications [2, 3]. The incidence of APS among patients with recurrent pregnancy loss reaches 27–42% [4].

Organ-specific autoantibodies can bind to negatively charged phospholipids on the cell membranes of platelets and endothelial cells [5], which may lead to thrombophilic complications resulting in preterm birth and miscarriage [6]. Of the diversity of existing autoantibodies, only three types are pathognomonic for APS: lupus anticoagulant, anticardiolipin and anti-b2-glycoprotein I [7]. However, there is also a fairly large range of antiphospholipid antibodies but are not included in the diagnostic criteria. These are antibodies to phosphatidylserine, phosphatidic acid, phosphatidylinositol, protein C, annexin 5, prothrombin, and others. Their role detection rates are controversial [8].

The pathogenesis of APS involves antibody-mediated thrombophilia. [9]. Activation of autoantibodies may be triggered by various factors including infections, malignancies, and medicinal substances. The autoantibodies bind to annexin 5, prothrombin, proteins C and S, interfering with the coagulation cascade [10], and potentiate platelet aggregation. Triggering the activation of endothelial cells may result in the development of endothelial dysfunction, which is one of the main pathogenic factors for the development of APS [11, 12]. Mediators secreted by damaged endothelial cells can serve as additional factors that aggravate the severity of the disease.

This study aimed to investigate the spectrum of APS-associated peripheral blood autoantibodies in women with recurrent pregnancy loss.

Material and methods

A total of 561 women of reproductive age was examined at the D.O. Ott Research Institute of Obstetrics, Gynecology, and Reproductology. The women were divided into two groups: group 1 (n = 76) included somatically healthy women with normal pregnancies, and group 2 (n = 485) comprised women, who were diagnosed with recurrent pregnancy loss. The age of women ranged from 25 to 47 (mean 32 ± 5.5) years. The inclusion criteria for the group I were the first trimester of pregnancy and the absence of pregnancy complications. The criteria for inclusion in group II were the first trimester of pregnancy and a history of threatened miscarriage or pregnancy loss. All women underwent clinical and laboratory examination, including complete blood count, urinalysis, biochemical, coagulation tests, and bacteriological studies. Exclusion criteria in both groups were diabetes mellitus, endocrine disorders, genetic factors of recurrent miscarriage, and uterine malformations.

Laboratory testing included measurement of serum levels of autoantibodies (IgG/IgM) to cardiolipin, phosphatidylserine, phosphatidylinositol, phosphatidic acid, beta2-glycoprotein-I, annexin 5, and prothrombin using enzyme-linked immunosorbent assay (ELISA) kits by Orgentec Diagnostika GmbH (Germany). The content of endothelial dysfunction markers (soluble intercellular adhesion molecule-1 (sICAM-I), endothelin-1, thrombomodulin, and von Willebrand factor) were determined by ELISA using sICAM-I BD (USA), ENDOTHELIN (1–21 ) (Biomedika, Austria), Thrombomodulin (USCN Life Science, USA) and Technozym vWF:Ag ELISA (Technoclone GmbH, Austria) kits, respectively. During testing, standard samples were used to build calibration curves and control samples. Testing results were registered instrumentally by measuring the optical density on an ELx808 microplate ELISA analyzer manufactured by BioTek Instruments Inc. (USA) at a wavelength of 450 nm. Testing results were evaluated by the manufacturer‘s recommendations. The lupus anticoagulant was determined on an ACL Elit PRO coagulometer (Instrumentation Laboratory, Spain) using reagents from Siemens (Germany).

Statistical analysis was performed with STATISTICA Windows 7.0 software using nonparametric statistics (Mann-Whitney test). Differences between the groups were considered significant at p <0.01 and p <0.001. The relation between variables was assessed with a nonparametric Spearman correlation coefficient, and the strength of the correlation was defined as follows: over 0.7 - stronger, from 0.3 to 0.7 - moderate, less than 0.3 - weak.

Results

Autoantibodies were found in 31.7% of patients with recurrent pregnancy loss (n = 485), while no autoantibodies were detected in the women with normal pregnancies (n = 76). Among patients with recurrent pregnancy loss, antibodies to cardiolipin were found in 14.3% of patients (n = 154), while an isolated increase in antibodies to cardiolipin was observed in 27.3% of cases; in 72.7% this was combined with other types of autoantibodies. Antibodies to phosphatidylserine, phosphatidylinositol and phosphatidic acid were detected in 7.1% of patients, there was no isolated increase in these autoantibodies, and in all cases, there was a combination with other types of autoantibodies. Antibodies to beta-2 glycoprotein-I were detected in 37.1% of patients both in isolation (56.1%) and combined (43.9%). Antibodies to annexin 5 were detected in 29.9% of patients both in isolation (71.7%) and combined with other antibodies (28.3%). Lupus anticoagulant was detected in 42.8% of patients, including 75.7% isolated and 24.3% combined (Fig. 1).

Combinations of autoantibodies were assessed depending on their titers. Women with recurrent pregnancy loss (n = 63) were subdivided into three groups with low (10–20 U/ml), medium (20–40 U/ml) and high titer (more than 40 U/ml) of autoantibodies to cardiolipin and beta-2 glycoprotein-I. Low, medium and high titers of autoantibodies were detected in 66.6%, 20.7%, and 12.7% of patients, respectively. It was noted that 30.1% of women with low autoantibody titers, along with antibodies to cardiolipin and beta-2 glycoprotein-I, also had antibodies to annexin 5, prothrombin, phosphatidylserine, phosphatidylinositol, phosphatidyl acid, and lupus anticoagulant; among patients with medium and high titers of autoantibodies these figures were 38.5% and 75.0%, respectively.

An assessment of endothelial dysfunction markers showed that blood levels of thrombomodulin and von Willebrand factor in women with recurrent pregnancy loss were higher than in women with normal pregnancies. Among women with recurrent pregnancy loss, levels of these markers were higher in those who had compared with those who had not had autoantibodies (Fig. 2). However, the Spearman correlation between antibodies to beta-2 glycoprotein-I and thrombomodulin was weak (r = 0.47, p < 0.05).

The peripheral blood level of soluble intercellular adhesion molecule-1 (sICAM-I) and serum annexin 5 in women with recurrent pregnancy loss was higher than in women with normal pregnancies. No significant differences were found between groups of women with recurrent pregnancy loss with and without autoantibodies in terms of sICAM-I and serum annexin 5 levels (Fig. 3). The blood levels of endothelin-1 did not differ significantly between groups of women with recurrent pregnancy loss and women with normal pregnancies.

Discussion

APS, as a condition with a different spectrum of clinical manifestations, has been studied for over 35 years. But despite the diversity of accumulated evidence on the alleged role of antiphospholipid antibodies obtained in experimental models [13] and clinical studies [14], there is still no convincing answer to the question of the relationship between them and miscarriage [15].

Our study showed that no autoantibodies were detected in women with normal pregnancies. At the same time, they were found in 31.7% of women with recurrent pregnancy loss, which shows their obvious involvement in the condition leading to pregnancy loss. Among the identified autoantibodies, the most prevalent were lupus anticoagulant and antibodies to beta-2 glycoprotein-I. In low and medium titers, antiphospholipid antibodies were determined mainly in isolation, whereas their high titers were found in combination with other types of autoantibodies in 75% of cases, which may indicate the activity of the autoimmune process as a whole.

Therefore, patients with recurrent pregnancy loss and other forms of autoimmune abnormalities require an in-depth examination. According to the literature, in women with a history of recurrent pregnancy loss and high positive antibody titers, 80% of next pregnancies culminate in a fetal demise [16, 17]. Patients with medium antiphospholipid antibody titers and planning pregnancy require prophylactic antithrombotic therapy. The clinical significance of low antibody titers is dubious and considered only as a risk factor in thrombophilic conditions [18], probably associated with transient instability of cell membranes.

Vascular endothelial cell dysfunction mediated by antiphospholipid antibodies plays a key role in APS pathogenesis. [19]. Activation of endothelial cells by antiphospholipid antibodies increases the expression of adhesion molecules (E-selectin, ICAM-1, VCAM-1) and enhance the synthesis and secretion of pro-inflammatory cytokines and chemokines, such as interleukin -1β and interleukin-6 [11]. There is an increase in the tissue factor expression on endothelial cells, which is one of the main triggers of the coagulation cascade, starting the external coagulation pathway. Activated endothelial cells also exhibit large amounts of von Willebrand factor (which is also a marker of endothelial cell damage). Besides, endothelial cells undergo desquamation of thrombomodulin from their surface [20, 21]. These changes lead to an increase in blood coagulation potential, increasing the risk of thrombosis.

Our study showed that concentrations of endothelial damage markers, such as thrombomodulin, von Willebrand factor, soluble intercellular adhesion molecule-1 (sICAM-I), and serum annexin 5 in women with recurrent pregnancy loss are higher than in women with normal pregnancies. It is also shown that the level of thrombomodulin and von Willebrand factor is higher in women with identified antiphospholipid antibodies. A positive correlation was found between antibodies to beta-2 glycoprotein-I and thrombomodulin. This fact confirms that antiphospholipid antibodies bind to membrane phospholipids of endothelial cells and initiate a cascade of reactions resulting in endothelial damage, procoagulant activity, and an increased risk of thrombosis, which ultimately leads to reproductive losses.

Conclusion

The study findings suggest that autoantibodies play a key role in the abnormalities of the coagulation process associated with recurrent pregnancy loss. They imply the need for new diagnostic tools to identify autoimmune abnormalities and pharmacological agents aimed at preventing the pathological effects of autoantibodies leading to premature termination of pregnancy. Comprehensive management of these disorders may involve intravenous immunoglobulins, coagulation modifiers and anticoagulants (antiplatelet agents, low molecular weight heparins), and therapeutic agents targeting the microcirculation.