Current Rheumatology Reports

, 16:403

Pregnancy Morbidity in Antiphospholipid Syndrome: What Is the Impact of Treatment?

Authors

    • Department of Obstetrics, Faculdade de Ciências MédicasUniversidade do Estado do Rio de Janeiro
  • Gustavo Rodrigues
    • Department of Obstetrics, Faculdade de Ciências MédicasUniversidade do Estado do Rio de Janeiro
  • Nilson R. de Jesús
    • Department of Obstetrics, Faculdade de Ciências MédicasUniversidade do Estado do Rio de Janeiro
  • Roger A. Levy
    • Department of Rheumatology, Faculdade de Ciências MédicasUniversidade do Estado do Rio de Janeiro
    • Federico Foundation
ANTIPHOSPHOLIPID SYNDROME (D ERKAN, SECTION EDITOR)

DOI: 10.1007/s11926-013-0403-6

Cite this article as:
de Jesús, G.R., Rodrigues, G., de Jesús, N.R. et al. Curr Rheumatol Rep (2014) 16: 403. doi:10.1007/s11926-013-0403-6
Part of the following topical collections:
  1. Topical Collection on Antiphospholipid Syndrome

Abstract

Women with persistently circulating antiphospholipid antibodies (aPL) have a higher incidence of recurrent abortions, fetal losses, pre-eclampsia, and placental insufficiency. Current treatment of patients with antiphospholipid syndrome (APS) during pregnancy with heparin and aspirin can act by preventing clot formation and improving live birth rates, but other obstetric morbidities remain high, especially in patients with a history of thrombotic events. In addition to the classical thrombotic placental events, other factors involving inflammation and complement activation seem to play a role in certain complications. In this article, we will review how medications interfere in the pathogenic mechanisms of APS, discuss the impact of current recommended treatment on pregnancy morbidity, and analyze new promising therapies.

Keywords

Antiphospholipid syndromeAPSObstetric APSObstetric antiphospholipid syndromePre-eclampsiaIntrauterine growth restrictionPrematurityTreatmentAbortionPregnancy lossFetal deathHeparinAspirinComplement activationThrombosisHydroxychloroquineEculizumab

Introduction

Pregnancy morbidity is a well-recognized presentation of antiphospholipid syndrome (APS), which is part of the classification criteria [1]. Women with persistent antiphospholipid antibodies (aPL) have a higher incidence of recurrent abortions, fetal losses, pre-eclampsia, and placental insufficiency than the general population [2••]. Lately, there have been reports of the pregnancy complications differing between patients with thrombotic or obstetric APS. After appropriate diagnosis, treatment of patients with APS shows remarkable improvement in live birth rates, but the risk of other obstetric morbidities, such as pre-eclampsia, intrauterine growth restriction (IUGR), and prematurity, remains high.

In this article, we will review how medications interfere in the pathogenic mechanisms of APS, discuss the impact of current recommended treatment on pregnancy morbidity and analyze new promising therapies.

Pathogenic Mechanisms of aPL-Mediated Pregnancy Morbidity

Thrombus formation is the key event of the vascular manifestations of APS, and many pathogenic mechanisms have been proposed to explain this thrombotic predisposition. However, the obstetric events related to the syndrome cannot be justified simply by thrombosis of the uteroplacental vasculature [3]. Initial studies reported placental infarction and thrombosis in patients with APS, and in vitro studies showed that aPL might produce a pro-coagulant state at the placental level through several mechanisms. These observations were not confirmed by subsequent studies, and hystopathologic analysis of most of miscarried placentas and fetuses from women with APS did not show signs of thrombosis [4••]. Some authors define obstetric APS as an inflammatory disorder [5], although the evidence of inflammation is not an obligatory finding in all animal models or in immunohistological analysis of abortive material or full-term placentae from patients with APS [4••].

One of described mechanisms for aPL-mediated pregnancy failure is complement activation. In a normal pregnancy, components of complement cascade are observed in the stroma of the villi around fetal vessels, trophoblast membranes, and decidual spiral arteries. However, a local expression of complement-regulating proteins, such as CD59, decay accelerating factor (DAF), and MCP, inhibit complement activation at the syncitiotrophoblast level. APL-treated mice have complement activation and extensive C3 deposition in the decidua, while C3-deficient mice do not show fetal resorption after aPL infusion, as do mice treated with C3 inhibitor [6]. C5a induces neutrophil activation and release of TNF and tissue factor, with subsequent infiltration of placental tissues leading to trophoblastic injury, placental dysfunction, and damage to placenta and embryo [5].

Heparin can prevent clot formation and inhibit complement activation during pregnancy. Girardi et al. showed that heparin could reduce the risk of pregnancy loss in pregnant mice injected with human IgG aPL even in the absence of a detectable anticoagulant effect, while this protection was not achieved with other anticoagulants (fondaparinux and hirudin). The authors also found that aPL-IgG-initiated C3 cleavage products were lower in plasma and had fewer fragments bound to trophoblasts of mice treated with unfractioned heparin (UFH) or low-molecular weight heparin (LMWH) compared to untreated mice [7].

The invading throphoblast and maternal decidua can also be targeted by aPL, resulting in defective placentation. In vitro studies reported that aPL, especially beta 2-glycoprotein 1 (β2GP1)-dependent antibodies, induce trophoblast injury and apoptosis, inhibit syncitia proliferation and formation, decrease production of human chorionic gonadotrophin, and impair trophoblast invasion and adequate secretion of growth factors [4••]. Polyclonal and monoclonal β2GPI-dependent aPL can bind stromal decidua cell monolayers and induce a pro-inflammatory phenotype, characterized by increased ICAM-1 expression and TNF secretion, associated with an impaired endometrial differentiation [8].

Trophoblast invasion can be improved by interleukin-3 (IL-3) formation, a cytokine associated with placental formation and trophoblast invasion that is up-regulated by low-dose aspirin (LDA). Aspirin can also improve pregnancy results by decreasing thromboxane A2 and prostaglandin I2 formation, two molecules that are related to gestational hypertension and pre-eclampsia [9]. However, in vitro studies have demonstrated that LMWH and LDA augment basal trophoblast migration, but they are not capable of reversing aPL-mediated reduction of trophoblast migratory capacity [10••].

Antiphospholipid antibodies can also modify the production of angiogenic factors in human first-trimester trophoblast cells, increasing the secretion of vascular endothelial growth factor (VEGF), placental growth factor (PlGF), and soluble endoglin (sEng), with a reduction of soluble FMS-like tyrosine kinase-1 (sFlt-1). These changes were not reversed by LMWH or LDA in one study [10••], but other authors have described the restoration of the VEGF secretion on endometrial endothelial cells, previously blockaded by aPL, with the use of enoxaparin and tinzaparin [11]. Patients with pre-eclampsia without APS have elevated circulating levels of sEng and sFlt-1 with very low levels of VEGF and PlGF [12, 13], and it still unknown whether this different angiogenic pattern induced by aPL in vitro is the same in APS patients.

Interestingly, another in vitro finding was that LMWH, applied to the first trimester trophoblast without aPL, induced an anti-angiogenic state similar to pre-eclampsia by increasing sFlt-1 and PlGF and reducing VEGF, while LDA had no effect on the production of these factors [10••]. sFlt-1 is a soluble receptor of VEGF and mediates the signs and symptoms of pre-eclampsia, while its elevated circulating levels are associated with clinical pre-eclampsia [14••].

Han et al. suggested that these changes in the angiogenic milieu may explain the inability of heparin and LDA to prevent adverse outcomes in late gestation, such as pre-eclampsia or IUGR. LMWH may prevent early loss by altering the inflammatory environment in APS patients, while setting the foundation for impaired placentation by worsening angiogenesis [10••].

In short, despite several mechanisms discussed above, the pathogenesis of obstetric morbidity in aPL-positive patients has to be studied further in order to determine the impact of current and future medications.

Pregnancy Morbidity: What Is the Impact of the Treatment?

Pregnancy Loss

Current treatment for pregnancy loss (recurrent abortions or fetal loss) in patients with APS is based on randomized controlled trials developed in the late 1990s and in the early years of this century. Different inclusion criteria, especially concerning aPL profile, are probably the reason for the conflicting results that have been presented [2••]: LDA alone was found to be inferior to LDA plus heparin in two studies [15, 16], while another study reported similar results when those treatment groups were compared [17, 18]. Due to this discrepancy, there is no consensus on a treatment recommendation for recurrent abortion before 10 weeks of pregnancy, when LDA alone or LDA plus heparin can be used. Most authors recommend LDA plus full-dose heparin in patients with fetal loss (after 10 weeks of pregnancy) throughout pregnancy until 6 weeks post-partum [19••]. In both situations, there is a significant increase in live birth rates from as low as 4 % [20••] before the diagnosis and treatment of APS to up to 85 % after treatment [20••, 21].

In all aPL-positive patients with pregnancy losses, it is imperative that other causes of embryonic or fetal deaths are excluded, such as genetic syndromes, maternal anatomic alterations and infections, or fetal malformations. This investigation remains necessary even in patients with well-established diagnosis of APS. The finding of aneuploidy in a cytogenetic analysis, for example, classifies this death as non-APS related and the patient can receive the same treatment in the next pregnancy. On the other hand, if no cause for the pregnancy loss is found, there is no consensus as to what should be done. Prednisone, intravenous immunoglobulin (IVIG), hydroxychloroquine, and even anti-TNF drugs have been proposed to be used in this group, but there is no evidence to support the use of any of them [22]. Actually, a few drugs, like prednisone, have been associated with more adverse pregnancy results [23] and should be used with caution.

A recent prospective multicenter study evaluating risk factors for pregnancy failure in APS patients treated with conventional therapy has been published. The authors compared 57 pregnancy failures in patients with APS despite treatment to 57 successful pregnancies, matched for age and therapy. At multivariate analysis, SLE or other autoimmune diseases (OR 6.0; 95 % CI 1.7, 20.8; p = 0.01), history of both thrombosis and pregnancy morbidity (OR 12.1; 95 % CI 1.3, 115.3; p = 0.03) and triple aPL positivity (OR 4.1; 95 % CI 1.0, 16.7; p = 0.05) were associated with pregnancy failure. Patients with previous pregnancy morbidity alone and/or a single aPL positivity, in contrast, had successful outcomes when treated with conventional therapy. The type of previous pregnancy morbidity did not apparently affect the outcome of the new pregnancy [24••]. These findings could be used in the future to separate patients with high risk of pregnancy failure, which could benefit from additional therapies, from those with better prognosis and with higher likelihood of live birth with the current treatment.

In summary, treatment of APS patients with LDA and heparin can significantly improve live birth rates, but in a few cases pregnancy failure occurs despite recommended medications. In this scenario, there is no study to support the use of any different drug and proposed treatments rely on expert opinion. New pathogenic mechanisms of aPL-mediated pregnancy loss have been elucidated in recent years and may be helpful in finding specific targets for future treatment proposals for refractory cases of this syndrome. Innovative therapies will be discussed in the last section of this article.

Pre-eclampsia

The real frequency of aPL in patients with pre-eclampsia is still unknown. Older studies yielded conflicting results, but must of them reported a positive association with severe, early-onset (prior to 34 weeks gestation) pre-eclampsia. Once again, different inclusion criteria may be responsible for these controversial results [25]. A meta-analysis showed a pooled odds ratio for association of anticardiolipin antibodies (aCL) with pre-eclampsia of 2.86 [95 % confidence interval (CI), 1.37–5.98), and a pooled odds ratio with severe pre-eclampsia of 11.15 (95 % CI 2.66–46.75). Despite this significant association, the authors concluded that there is insufficient evidence to use aCL as predictors of pre-eclampsia [26].

Considering patients with definite APS, there are no randomized controlled trials that included patients with exclusively severe pre-eclampsia or eclampsia and premature birth before the 34th week, as stated by the classification criteria [1]. Therefore, recommended treatment for this group of patients (LDA plus heparin) is the same as for the patients with pregnancy loss; however, its real efficacy is unknown. The main question is whether the treatment can prevent the development of pre-eclampsia and intrauterine growth restriction in aPL-positive patients.

The frequency of pre-eclampsia in treated APS patients with or without previous pre-eclampsia can be as high as 50 %, with half the cases being severe [25]. This frequency is considerably higher than seen in the general population, estimated up to 7 % [27]. Serrano et al. reported 19.4 % of hypertensive disorders in 67 pregnancies (51 patients) with primary APS, including four cases of HELLP syndrome [21]. A multicenter European database of 590 women described lower frequency of pre-eclampsia and eclampsia (13.9 %) and 2 % of abruption placentae [28].

Bramham et al. also found a high incidence of pre-eclampsia in patients with APS, but those with previous thrombotic events tended to have higher rates of pre-eclampsia than women with purely obstetric features of the syndrome (24.4 vs. 9.5 %, respectively) [29••]. A group from India described 30.9 % of pre-eclampsia in APS patients with mainly obstetric criteria (40 patients with obstetric APS, 2 patients with history of thrombotic events), including 7.1 % of severe cases [20••]. Pregnant APS patients with history of cerebral ischemic events are at increased risk for : (1) pre-eclampsia (34.8 %), especially those positive for multiple aPL test; and (2) new cerebral ischemia if pre-eclampsia develops (OR 7.0, p = 0.15) [30•].

High prevalence of pre-eclampsia in APS patients despite treatment may represent a different pathway of non-aPL-related hypertensive disorder of pregnancy, as discussed above. For example, LDA, used in all recommended protocols for APS, reduced the occurrence of pre-eclampsia in 53 % of non-APS high risk patients when started before 16 weeks of pregnancy, with a reduction of almost 80 % of the severe cases [31, 32]. While studies in APS patients are still lacking, we do not seem to be closer to those numbers with the current treatment.

In summary, pre-eclampsia is still common in patients with APS receiving LDA and heparin, resulting in significant maternal and fetal morbidity in these pregnancies. Clinicians should be aware of obstetric APS-related events even in patients with history of only thrombotic manifestations of the syndrome.

Intrauterine Growth Restriction (IUGR)

An accurate diagnosis of IUGR is still lacking in the obstetric field. The usual definition is based on birth weight below the 10th percentile, including in the same group, small, but appropriately grown neonates who usually do well, and others for whom small size is a manifestation of IUGR [33]. The lower threshold (i.e. birth weight below the 5th percentile) will reduce the number of false positive cases with the cost of missing true pathological newborns. Some authors recommend that fetuses should be considered growth-restricted when there is a recognizable maternal pathology or abnormal umbilical or middle cerebral artery Doppler studies [34].

The small number of studies evaluating the prevalence of aPL in patients with IUGR is due to the previously mentioned controversies. Different inclusion criteria and controversial results have been reported [35, 36], with some authors describing similar frequency of aPL compared to controls [3739]. In addition, all studies had only a small number of patients enrolled, with a maximum of 50 patients with IUGR. Thus, it is difficult to estimate the real frequency of aPL in mothers with isolated growth-restricted neonates without other APS features.

As well as pre-eclampsia, IUGR is commonly found in pregnant women with APS before and after treatment. Both diseases represent different clinical manifestations of abnormal placental function and can be found in the same patient. Isolated fetal impairment, however, is possible and may develop with absent or minimal maternal signs, like small uterine height. For this reason, it is recommended that a monthly ultrasound evaluation should be performed and the Doppler-velocimetry studies started after 26 weeks of pregnancy [19••]. The more severe cases of IUGR show abnormal waveforms in umbilical artery, diagnosed by Doppler-velocimetry, or altered biophysical profile. Those findings represent chronic fetal distress and an increased risk for intrauterine death, eventually leading to premature interruption of pregnancy [33].

Serrano et al. described 15.5 % (9/58) cases of IUGR [21], while Dadhwal et al. had the same finding in 9 (21.4 %) of 42 pregnancies [20••]. The group from St Thomas’ Hospital, London, had almost 40 % of small for gestational age (SGA) neonates (below the 10th percentile) in patients with thrombotic APS, compared to 16 % of patients with obstetric APS [29••]. Fischer-Betz et al. reported 28.6 % of SGA in pregnancies after cerebral ischemic events related to aPL [30•].

A publication evaluating the efficacy of prophylaxis using LDA and UFH in the prevention of IUGR in patients with APS provided negative results. Of the study group, 32.3 % had low birth weight newborns (below the 10th percentile) compared to 2.5 % of the control group. The mean birth weight of the newborn babies of the study group was also smaller than controls (2,798 vs. 3,124 g, respectively) [40].

Once again, treatment for this presentation of APS is disappointing, with a higher frequency of IUGR in APS patients. Intrauterine growth restriction is commonly associated with pre-eclampsia, but isolated IUGR can be found even in patients with no previous aPL-related obstetric events. Routine evaluation of fetal growth is recommended for early diagnosis [19••].

Prematurity

Premature birth is the most frequent neonatal complication of APS during pregnancy (20–25 % of patients; range 11–66 % based on different studies) [20••, 21, 28, 41••], mainly caused by medical intervention due to pre-eclampsia and IUGR. Preterm delivery is required in approximately one-third of APS patients with pre-eclampsia [25]; spontaneous premature birth is less frequent. Bramham et al. described higher rates of preterm delivery in patients with thrombotic APS when compared to women with history of recurrent abortion [29••].

Since the frequencies of pre-eclampsia and IUGR remain high in women with APS, it is difficult to determine the current estimates of premature births on those patients. Specific treatment to prevent premature birth is not recommended, as it is a common consequence of other obstetric complications related to APS.

Pregnancy Morbidity: Our Experience

At Hospital Universitário Pedro Ernesto, we conducted a retrospective review (2001–2012) of 43 pregnancies in 32 APS patients (unpublished data). Table 1 demonstrates the characteristics of our patients and Table 2 shows the gestational outcomes of those patients, according to the clinical presentation of APS (obstetric APS vs. thrombotic APS).
Table 1

Characteristics of patients with APS followed in outpatient clinic for rheumatic diseases, thrombophilic disorders, and fetal loss at Hospital Universitário Pedro Ernesto

Patients (n)

32

Pregnancies (n)

43

Age (mean)

30.5 years

Primary APS (n/%)

26/81 %

Secondary APS (n/%)

6/19 %

Live birth rate

 Before treatment (%)

22.2 %

 After treatment (%)

83.8 %

Fetal death after treatment (n)

6

Late abortion (n)

1

APS Antiphospholipid syndrome

Table 2

Gestational results of 43 pregnancies in 32 patients with antiphospholipid syndrome (APS), according to the presentation of the disease (obstetric APS vs. thrombotic APS)

 

Total (43 pregnancies)

Obstetric APS (18 pregnancies)

Thrombotic APS ( 25 pregnancies)

p value comparing Obstetric APS vs. Thrombotic APS (CI = 95 %)

Pre-eclampsia (n/%)

11/25.5 %

4/22.2 %

7/28 %

0.47a

Intrauterine growth restriction (n/%)

10/23.2 %

2/11.1 %

8/32 %

0.10a

Mean weight of live births

2,615 g

3,067 g

2,345 g

<0.001b

Altered dopplevelocimetry study (n/%)

4/9.3 %

0/0 %

4/16 %

N/A

Premature birth (n/%)

11/25.5 %

2/11.1 %

9/40 %

0.06a

Obstetric APS only obstetric events related to antiphospholipid antibodies, without known vascular thrombotic event; Thrombotic APS history of vascular thrombotic events related to antiphospholipid antibodies according to APS criteria, with or without obstetric events

ap value of pre-eclampsia, premature birth, and intrauterine growth restriction according to Fisher’s exact probability test

bp value of mean weight of live births according to Student’s t test

Four pregnancies in patients with history of early recurrent abortion were treated with LDA only and 9 pregnancies were treated with LDA plus LMWH throughout the entire pregnancy. The remaining 28 pregnancies received LDA plus heparin from confirmation of pregnancy until 14 weeks, then this combination was changed to LDA plus warfarin to maintain INR 2–3. Warfarin was stopped and heparin was re-started at 34 weeks of pregnancy.

Warfarin was prescribed in our patients because of the high cost of heparin in our country, even UFH, and also due to our experience with women with metallic heart valves, in whom heparin is not efficient for the prevention of valve thrombosis [42]. We do not start vitamin K antagonist during the first trimester of pregnancy to prevent the development of warfarin embryopathy, which has been described only in this period [42], and due to probable benefits of heparin during trophoblast differentiation, as was described above. We re-start heparin near term for a better management of delivery. We had no cases of major bleeding in the mother or the fetus, and no observation of chromosomal or anatomic alterations in the patients that received warfarin during pregnancy.

In our group of patients, we found an improvement of live birth rates from 22.2 to 83.8 % (6 fetal deaths and 1 late abortion). Our results are similar to what has been reported by other studies when patients used only LDA with heparin throughout the entire pregnancy [20••, 21].

Of 43 pregnancies, 25,5 % were complicated by pre-eclampsia: 28 % (7/25) in thrombotic APS patients, and 22 % (4/18) in patients with only obstetric APS (CI 95 %, p = 0.47) (Table 2). One patient with previous deep venous thrombosis and stroke developed HELLP syndrome, resulting in fetal death at 25 weeks. Two (11 %) of the 18 pregnancies of patients with obstetric APS had IUGR, while patients with thrombotic APS had higher number of cases of fetal growth restriction, not reaching statistical significance (8/25, 32 %; CI 95 %, p = 0.10). The mean weight of the live births, though, was significantly lower in patients with thrombotic APS (2,345 compared to 3,067 g of the patients with obstetric APS; CI 95 %, p < 0.001). Four of our patients with thrombotic APS had altered umbilical artery waveform in Doppler ultrasound, while none of the patients with obstetric APS had this finding. These results suggest a more aggressive disease in patients with thrombotic APS.

Considering prematurity, 11 (30.5 %) of the 36 live births of our group of patients were born before term: 6 patients had pre-eclampsia and 2 fetuses had chronic fetal distress diagnosed by Doppler-velocimetry study without hypertensive disorder of the mother. Interestingly, only 2 patients with premature births had obstetric APS while 8 had thrombotic events, including 1 patient with previous stroke that had two births before 32 weeks due to severe pre-eclampsia while on full-dose anticoagulation in both pregnancies.

Future Possible Therapies for Obstetric APS

Hydroxychloroquine

The most promising drug for treatment of obstetric APS is hydroxychloroquine (HCQ), an old antimalarial used for systemic lupus that can possibly target specific steps of obstetric APS that apparently are not modified by LDA or heparin [43•]. It is safe to use during pregnancy [42], and acts inhibiting inflammatory cytokines such as TNF, IL-1, IL-2, and IL-6, inhibiting Toll-like receptor activation, interrupting antigen processing, and inhibiting TCR- and BCR-induced calcium signaling. HCQ also inhibits platelet aggregation and arachidonic acid release from stimulated platelets [44•]. Observational studies suggest that HCQ has an antithrombotic effect in aPL-positive patients, especially those with associated systemic lupus [45].

Rand and colleagues showed that HCQ can dissociate aPL IgG–β2GP1 complexes and reduce the amount of aPL IgG that binds to phospholipid bilayers trapping annexin A5, a potent natural anticoagulant with high affinity for anionic phospholipids. Annexin A5 is required for the maintenance of placental integrity in mice and plays a thrombomodulatory role at the maternal–fetal interface within the placentary blood circulation by shielding apical membrane phospholipids of placental villous syncytiotrophoblasts [46].

The reversion of aPL IgG–β2GP1 complexes mediated by HCQ permits more annexin A5 to crystallize and may also ameliorate some of the other proposed thrombogenic effects. They also found that HCQ promotes the formation of second layers of annexin A5 crystal “patches” over areas where the mentioned immune complexes had disrupted crystallization, thereby further reducing the exposure of thrombogenic phospholipids [46].

Human syncytiotrophoblasts, obtained from placentas of women undergoing elective cesarean sections at term, had markedly reduced expression of annexin A5 when they were exposed to monoclonal and polyclonal aPL IgGs, with significant binding of the antibodies to the cells. As was expected from the results of previous studies, all these aPL-mediated effects were completely reversed with the addition of HCQ [43•].

In summary, in vitro studies indicate that HCQ may be beneficial to APS patients, but currently there are no human data confirming these findings. Considering its well-known safety profile and absence of adverse fetal effects [45], HCQ could be used in patients with APS, especially those with pregnancy morbidity despite adequate treatment. Prospective randomized trials should be designed to evaluate its efficacy in patients with APS, initially as additional therapy of current treatment with LDA and aspirin.

Complement Activation

Complement activation shown in both thrombosis and adverse obstetric events related to aPL is also found in patients with pre-eclampsia and HELLP syndrome without APS, inducing dysregulation of angiogenic factors [47••]. Inhibition of complement activation, therefore, is an interesting target for new therapies. Eculizumab is a humanized monoclonal antibody directed against complement protein C5, which inhibits its cleavage to C5a and C5b and prevents membrane attack complex. One report described benefits of eculizumab in a patient with catastrophic APS who was refractory to conventional therapy [44•]. Considering pregnancy morbidity, a recent article described the use of eculizumab in a previously healthy patient with HELLP syndrome at 26 weeks gestation, prolonging pregnancy for 17 days and probably reducing neonatal morbidity [47••]. Similar studies depicting benefits in pregnant APS patients are lacking.

Eculizumab is classified as category C by FDA as there are no well-controlled studies in humans, but potential benefits may warrant use of the drug in pregnant women with severe pregnancy morbidity despite potential risks. An important drawback of this medication is the high cost, which will limit its use unless the efficacy and safety in APS patients are demonstrated with larger studies.

Extracorporeal Apheresis

The possibility to treat pre-eclampsia changes a paradigm that was held until a few years ago, when delivery was the only known therapy [48]. A pilot study used extracorporeal apheresis in five patients with very preterm severe pre-eclampsia to adsorb circulating soluble fms-like tyrosine kinase 1 (sFlt-1). Dextran sulfate apheresis lowered circulating sFlt-1, reduced proteinuria, stabilized blood pressure, and prolonged pregnancy for at least 15 days without apparent adverse events to mother or fetus, with evidence of fetal growth during this period.

There is no such study in APS patients, but the results are encouraging, considering how often premature severe pre-eclampsia is diagnosed in this group of patients [14••]. Larger trials are needed to confirm the reported benefits and apparent safety before extracorporeal apheresis can be recommended for patients with pre-eclampsia, with or without APS.

Double Anti-Aggregant Therapy

Double anti-aggregant therapy has been recommended for some thrombotic events not related to pregnancy, such as ischemic heart disease, atrial fibrillation, and for secondary prevention of stroke [49], and the most used combinations are LDA plus dipyridamole and aspirin plus clopidogrel. Dipyridamole is an adenosine transporter inhibitor which increases extracellular levels of adenosine [50] and clopidogrel is an irreversible blocker of the platelet ADP receptor that is crucial in their activation and aggregation [51]. In vitro, dipyridamole was able to regulate VEGF and sFlt-1 secretion in the hypoxic placenta and could, therefore, control the balance of these competing angiogenic factors in diseases characterized by placental ischemia [50]. The anti-inflammatory action of clopidogrel [52] could contribute, in an auxiliary way, reducing APS-induced obstetric complications.

Although the combination of anti-aggregant agents seems to be more protective against recurrent thrombotic events, they are most closely associated with hemorrhagic complications, especially the combination of aspirin plus clopidogrel [53].

In summary, the use of combined therapy with anti-aggregants may be an alternative in selected cases, such as refractory obstetric APS, but the potential risk of bleeding during pregnancy should be better evaluated [49].

Tumor Necrosis Factor-Alpha (TNF-α)

Tumor necrosis factor-alpha (TNF-α) is a pleiotropic cytokine that plays a crucial role in the implantation and development of pregnancy and in the pathophysiology of rheumatic and autoimmune diseases. It influences the blastocyst implantation, vascular permeability of the endometrium, and uterine deciduation [54]. High levels of TNF-α have been found in pregnancy complications such as infection, IUGR, and early unexplained spontaneous abortions [55, 56]. Berman et al., in a murine model of APS, found that TNF-alpha deficiency and TNF blockade promote a protective effect in C5-deficient mice treated with aPL. Thus, they concluded that TNF blockade is a potential therapy for pregnancy complications of APS [57].

Anti-TNF drugs are classified by the US Food and Drug Administration (FDA) as pregnancy risk category B, with infliximab, etanercept, adalimumab, and certolizumab being the most frequently used medications [22]. Although some authors suggest that anti-TNF drugs could be used for refractory cases of APS during pregnancy [22], there are no studies to support this recommendation. Therefore, the use of these medications in APS patients should be reserved for research purposes.

Statins

Statins are cholesterol-lowering agents that also have anti-inflammatory properties with a direct effect on the endothelial expression of adhesion molecules, plaque formation, and thromboxane synthesis [44•]. APL induce increased expression, function, and transcription of tissue factor on endothelial cells that could be reversed by treatment with statins in in vivo experimental models [44•]. While miscarriages have been reduced by using pravastatin in aPL-treated mice [58], other authors have reported that the same drug did not mitigate the adverse in vitro effect of aPL on trophoblast migration and had no impact on the secretion of pro-inflammatory cytokines and angiogenic factors by primary human first trimester trophoblast cells exposed to aPL [59].

Concerns about the safety of statins during pregnancy remain, as they are related to teratogenicity, disruption of gonadal stem cell development in fetuses, and theoretical long-term fetal neurological damage [60], being classified as Category X for pregnancy (contraindicated) by FDA. Trials of statin during human pregnancy should only be conducted if there is a clear indication of human safety, including primate trials, and IRB approvals under strictest conditions, including long-term follow-up of the newborns [60]. Thus, statins should not be used prevent pregnancy loss in patients with APS due to known adverse fetal effects and conflicting in vitro results reported.

Conclusions

Treatment of patients with APS during pregnancy can significantly improve live birth rates, but other obstetric complications such as pre-eclampsia, fetal growth restriction, and prematurity remain high despite treatment. Patients with history of thrombotic APS develop more pregnancy morbidity compared to exclusively obstetric APS patients. Novel therapies for obstetric APS may further increase the live birth rates and could be an option for refractory cases, but ideally they should also be able to treat other forms of pregnancy morbidity in APS patients.

Compliance with Ethics Guidelines

Conflict of Interest

Guilherme R. de Jesús and Roger A. Levy have had travel/accommodations expenses covered/reimbursed by APS Action.

Gustavo Rodrigues and Nilson R. de Jesús declare that they have no conflict of interest.

Human and Animal Rights and Informed Consent

This article does not contain any studies with human or animal subjects performed by any of the authors.

Copyright information

© Springer Science+Business Media New York 2013