Reproductive Toxicology 22 (2006) 227–233

Review

Inherited and acquired thrombophilia: Pregnancy outcome and treatment

Marco De Santis a,∗, A.F. Cavaliere a, G. Straface a, E. Di Gianantonio b, A. Caruso a

a Telefono Rosso—Teratology Information Service, Department of Obstetrics and Gynecology,Catholic University of the Sacred Heart, Largo A. Gemelli, 8, 00168 Rome, Italy

b Servizio di Informazione Teratologica (SIT), Genetica Clinica ed Epidemiologica, University of Padua, Italy

Received 1 March 2006; received in revised form 3 May 2006; accepted 3 May 2006Available online 23 June 2006

Abstract

Maternal thrombophilias increases the risk of an adverse pregnancy outcome.An extensive literature review highlights the role of inherited and acquired thombophilic disorders in spontaneous abortion, both early and late,

recurrent or isolate, in intrauterine growth retardation, in placenta abruption, in pre-eclampsia and in venous thromboembolism. We have particularlyfocused attention on the following factors: antithrombin III (ATIII), proteins C (PC) and S (PS) deficiencies, genetic mutations particularly factorV Leiden (FVL), prothrombin gene G20210A (PTM) and the thermolabile variant of the methylene tetrahydrofolate reductase C677T (MTHFR)g

©

Kt

C

1

oa

0d

ene, lupus anticoagulant (LAC) and anticardiolipin antibodies, VIIIc factor, hyperhomocysteinemia and acquired activated protein C resistance.Appropriate treatment can improve pregnancy outcome without teratogenic effects.2006 Elsevier Inc. All rights reserved.

eywords: Inherited thrombophilia; Acquired thrombophilia; Pregnancy outcome; Spontaneous abortion; Pre-eclampsia; IUGR; Placenta abruption; Anticoagulantherapy

ontents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2271.1. Pregnancy loss . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2281.2. Pre-eclampsia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2291.3. Intrauterine growth retardation (IUGR) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2301.4. Placental abruption . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230

2. Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2302.1. Unfractionated heparin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2302.2. Low molecular weight heparin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2312.3. Oral anticoagulants . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2312.4. Acetylsalicylic acid (ASA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231

3. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232

. Introduction

Complications which occur during pregnancy are the causef morbility and mortality and necessitate medical interventionimed at improving the fetal and maternal outcome. Coagulation

anomalies, which play an important role in adverse pregnancyoutcome, are the subject of this review.

Thrombophilia are hemostatic disorders, classified as inher-ited and acquired, which affect about 15% of the Cau-casian population predisposed to thrombotic phenomenona[1]. Hereditary disorders include deficiencies of antithrom-bin III (ATIII), proteins C (PC) and S (PS) deficiencies,

∗ Corresponding author. Tel.: +39 06 30156525; fax: +39 06 30156572.E-mail address: marcodesantis@rm.unicatt.it (M. De Santis).

genetic mutations such as factor V Leiden (FVL), prothrom-bin gene G20210A (PTM) and the thermolabile variant of

890-6238/$ – see front matter © 2006 Elsevier Inc. All rights reserved.oi:10.1016/j.reprotox.2006.05.002

228 M. De Santis et al. / Reproductive Toxicology 22 (2006) 227–233

the methylene tetrahydrofolate reductase C677T (MTHFR)gene.

The most common acquired thrombophilias are due toantiphospholipid antibodies, which include lupus anticoagulant(LAC) and anticardiolipin antibodies. Thrombophilias derivedfrom a combination of hereditary and acquired components,such as the VIIIc factor, hyperhomocysteinemia and acquiredactivated protein C resistance, are identified [1]. In addition,pregnancy itself leads to a thrombophilic state as a result ofhemostatic and fibrinolitic changes. During pregnancy, proco-agulant factors (such as VIII, XII, VII, V) and the von Willebrandfactor and fibrinogen are increased, protein S and the activatedprotein C are reduced and fibrinolitic activity is diminished.All of these modifications, together with an enlarged plasmaticvolume, prepare the mother to face the hemostatic state duringdelivery.

The risk of venous thrombosis in pregnancy is increasedseven- to eight-fold, and even more after delivery, as a resultof all these modifications or as the first feature of an hereditarythrombophilia [2].

Thrombophilias have been investigated in relation to thefollowing obstetric complications: recurrent and non-recurrentmiscarriages in early or late pregnancy, intrauterine death,intrauterine growth retardation (IUGR), placental abruption,hypertensive disorders of pregnancy and maternal or neonatalthrombosis. In recent years, the importance of the use of antico-aae

1

(p

tafiaer

1awwc

wp5iwnr

In a total of over 3000 patients from two different studies,there were no statistical differences in late abortions regardingthe FV Leiden [8,9].

No association was observed between recurrent abortions andfactor V Leiden in a study conducted on 1111 Caucasian patients(904 with early recurrent miscarriages and 207 late miscarriages)compared with 150 controls. However, the same study noticed acorrelation between protein C resistance and recurrent abortions[10].

A greater prevalence of the FV Leiden and prothrombin (FII)G20210A mutations in 150 patients with adverse pregnancy out-comes (99 with a positive history for recurrent abortion between13 and 20 weeks of pregnancy and 51 with fetal loss after 20weeks) was recently found when compared with 115 controls[11]. By focusing our attention on the most important studies car-ried out in the last 2 years, we have found a correlation betweendifferent thrombophilic factors.

In the studies published in 2005, an OR = 3.28 (95% CI1.34–8.04; P = 0.01) was identified in relation to factor VIII andmiscarriage in 92 cases compared to 380 controls and a clearassociation in Caucasian patients, but not in non-Caucasians(among whom the mutations were rare), was found when acase–control study of 3496 patients examined for the FVL andprothrombin G20210A mutations was carried out [12,13]. Bothmutations were independent risk factors only in those whoshowed clinical features from the 10th week of pregnancy. Inatfi1bt[

ictfnte

bMtcimwpbpfo

a

gulant therapy during pregnancy in preventing an adverse fetalnd maternal outcome due to thrombophilic disorders has beenmphasized.

.1. Pregnancy loss

Early and late abortion are defined on the basis of trimesterfirst or second) in which they occur; stillbirth is defined whenregnancy loss occurs after 24 weeks of gestation.

Since 1989, when Triplett and Harris identified a correla-ion between recurrent miscarriages and the antiphospholipidntibodies syndrome and 1996, when Sanson et al. correlatedetal loss with PC, PS and AT deficiencies, many studies havenvestigated the link between thrombophilias and spontaneousbortions [3,4]. Despite the substantial amount of data gath-red, there is still a certain amount of controversy in theesults.

In a retrospective cohort study, 228 FVL mutation carriers and21 controls evaluating fetal loss before and after 20 weeks werenalysed. The fetal loss was 31.6% versus 22.3% in the controls,ith particular occurrence <20 weeks (29.4% versus 17.4%)ith greater risk in FVL homozygous mutation [5], which was

onfirmed in further studies [6].However, the evaluation of FVL mutation in 52 Japanese

omen with recurrent spontaneous abortions and 41 of theirartners found no underlying differences when compared with5 controls [7]. Evaluating for FVL and MTHFR genotypesn 584 patients in Ireland found a greater FVL in the patientsho subsequently had a late spontaneous abortions but foundo significant differences between case and control groups withegard to IUGR and pre-eclampsia [8].

study, which examined the association between activated pro-ein C mutations and pregnancy complications such as abortion,etal loss, IUGR, placental abruption and pre-eclampsia [14],t was confirmed that a greater thrombotic risk occurs after the0th week. There was a similar association, in another study,etween recurrent abortion, intrauterine death, placental abrup-ion, pre-eclampsia and hereditary or acquired thrombophilia15].

Behjati et al., in a study published on Iranian patients (36 withdiopathic infertility and 65 with recurrent spontaneous abortion,ompared with 62 controls), found a greater frequency of muta-ions for the FVL in both infertility and abortion and a greaterrequency of MTHRF mutations in patients with abortions butot idiomatic infertility. There was, however, no correlation withhe factor II (FII) G20210A mutation and a reduction of the lev-ls of APCR [16].

In the same period, no differences of allelic frequencyetween the cases and the controls with regard to FVL, PTG andTHRF were reported, nor for multiple thrombophilic muta-

ions analysing 357 couples with early recurrent abortion and 68ontrols, all of Caucasian race [17]. However, among couplesn which one or other partner had more than one thrombophilic

utation, the relative risk of abortion in the following pregnancyithout therapeutic intervention was equal to 1.9 when com-ared to those couples who were not mutation carriers. It haseen established that hereditary trombophilia associated withositive histories for fetal loss, recurrent abortion, treatmentsor infertility and IUGR are not independent factors for the riskf perinatal mortality [18].

In a systematic review, also published in 2006, Robertson etl. evaluated the existing correlation between abortion and trom-

M. De Santis et al. / Reproductive Toxicology 22 (2006) 227–233 229

bophilia: 35 studies were selected comprised of 7167 patientswith early abortions and 15 studies with 4038 late abortionpatients. Both adverse outcomes revealed the existence of astatistically significant link with some hereditary and acquiredtrombophilias. In fact, the analysis shows:

1. Factor V Leiden: a significant association with early abor-tion, both in homozygosity and heterozygosity, with an OR,respectively, of 2.71 (95% CI 1.32–5.58) and of 1.68 (95%CI 1.09–2.58), and with late abortion in heterozygosity withan OR equal to 2.06 (95% CI 1.10–3.86).

2. Prothrombin heterozygosity: association with early abortionwith an OR 2.49 (95% CI 1.24–5.00), and with late abortionwith an OR of 2.66 (95% CI 1.28–5.53).

3. Anticardiolipin antibodies: significant association with earlyabortion shown from an OR 3.40 (95% CI 1.33–8.68) andwith late abortion with an OR 3.30 (95% CI 1.62–6.70).

It is interesting to observe that the comparison between recur-rent early abortion and single fetal loss in the II trimester showsthat the FV Leiden (homozygous and heterozygous) and theprothrombin heterozygosity are a statistically greater risk fornon-recurrent fetal loss of the II trimester when compared withearly recurrent abortion; protein activated C resistance wouldseem to be linked to a greater risk of early recurrent abortion.

Early abortion is associated with hyperhomocysteinemia (OR6ahtt3

1

pat

eopll

aw[

Ccp6f1v

sus 7.5% for anticardiolipin antibodies (P = 0.16). The preva-lence of these anomalies were 1.5–2 times greater in the patientswho gave birth prematurely (<28 sett), compared to those whodelivered after the 28th week [30].

Two years later, the same authors focused attention on theexisting relationship between the prothrombin 20210GA muta-tion and the FVL with pre-eclampsia and, at around 3 monthspostpartum, found no significant difference in 114 patients witha history of pre-eclampsia or HELLP (hemolysis, elevated liverenzymes, low platelets) syndrome [31].

A meta-analysis of 19 studies, which included 2742 womenaffected with hypertension and 2403 controls, compared tothe FVL, obtained a statistically significant result showing anincreased risk of hypertensive problems which was 2.25 timesgreater [32]. The mutation increases the risk of such disorders(OR 2.25; 95% CI 1.50–3.38) but an enormous heterogeneityamong the studies must be underlined. In fact, in the threelarge studies included in the meta-analysis, the association isnot referred to.

In a review carried out very recently which included 25studies with a total of 11,183 patients, the risk of develop-ing pre-eclampsia (both moderate and severe) in patients withtrombophilia was found to be present [19]. The association wasevident with factor V Leiden (heterozygosity) (OR 2.19; 95%CI 1.46–3.27) but with notable heterogeneity, prothrombin het-erozygosity (OR 2.54; 95% CI 1.52–4.23), homozygosity forM(bmaif

po((idccpipa

epiidTOrIs

.25; 95% CI 1.37–28.42), LAC (OR 2.97; 95% CI 1.03–8.56)nd acquired APCR (OR 4.04; 95% CI 1.67–9.76) with notableeterogeneity in the meta-analytic evaluation particularly forhe LAC. A statistical significance has been found between pro-ein S deficiency and late abortion with an OR = 20.09 (95% CI.70–109.15) [19].

.2. Pre-eclampsia

The FV Leiden mutations, the MTHRF 677 C/Ts polymor-hism and other hereditary thrombotic risk factors can moder-tely increase the risk of pre-eclampsia. However, the role ofrombophilias in this pathogenesis is still controversial.

The first studies were conducted in the 1990s. Dekker’s sci-ntific group, in 1995 and in 1999, identified a greater frequencyf thrombophilic alterations in cases with pre-eclampsia and inatients who had delivered before the 28th week [20,21]. Theink between FVL and pre-eclampsia was highlighted in the fol-owing analyses [22–25].

Other trombophilic disorders have also been investigated andcorrelation with APCR levels, with the mutation MHTRF andith elevated plasmatic levels of homocysteina has been found

25–29].Van Pampus et al. analysed the presence of activated proteinresistance, FVL mutation, hyperhomocysteinemia and anti-

ardiolipin antibodies in 345 patients with histories of severere-eclampsia, at least 10 weeks after the delivery, compared to7 controls. In the study group the presence of anomalies wereound in about 40% of the cases, and in particular 11.3% versus.5% of the controls for protein C resistance (P = 0.025), 12.1%ersus 4.5% for hyperhomocysteinemia (P = 0.115), 20.9% ver-

THFR (OR 1.37; 95% CI 1.07–1.76), hyperhomocysteinemiaOR 3.49; 95% CI 1.21–10.11) and with anticardiolipin anti-odies (OR 2.73; 95% CI 1.65–4.51). When the presence ofoderate and severe forms of pre-eclampsia are analysed sep-

rately, the existence of a heterogeneity in the moderate forms shown which is not found in the studies related to the severeorm.

A meta-analysis published in 2004 evaluated if polymor-hism C677T in the gene for MTHFR played a role in the devel-pment of pre-eclampsia. It revealed a moderate risk increaseOR 1.3; 95% CI: 1.0–1.4) in the carriers of the T mutationMTHFR 667 CT and TT) in comparison with the homozygos-ty for C mutation (MTHFR 667 CCs) [33]. Nevertheless, theata included in this analysis was contradictory and no studyarried out between 2001 and 2003 showed a significant asso-iation. When the plasmatic levels of homocysteina, in 1874atients were evaluated, no differences were observed in thenitial phases of the II half of pregnancy between cases withre-eclampsia and IUGR compared to controls, while there wasdifference at term [34].

Both the hereditary and acquired forms of trombophilias andvaluated factor V Leiden, factor II G20210A, MTHFR C677T,rotein S, protein C, ATIII deficiency, anticardiolipin antibod-es, lupus anticoagulant, hyperhomocysteinemia were examinedn a case–control study carried out on 808 cases which hadeveloped pre-eclampsia and an equal number of controls [35].he results showed that 406 cases of severe pre-eclampsia hadR = 4.9 (95% CI 3.5–6.9) and that they were thrombophilia car-

iers, excluding proteins S and C and antithrombin deficiency.n the remaining 402, affected by mild pre-eclampsia, the onlyignificant association was with the mutated prothrombin and

230 M. De Santis et al. / Reproductive Toxicology 22 (2006) 227–233

MTHFR homozygous mutation. The patients affected by severepre-eclampsia and found positive in the screening for throm-bophilia, showed a more elevated risk in comparison to thosewho were not thrombophilic or experiencing severe maternalcomplications such as acute renal insufficiency, disseminatedintravascular coagulation, placenta abruption and perinatal mor-tality [36].

1.3. Intrauterine growth retardation (IUGR)

A possible association may exist between IUGR and trom-bophilia but again the data is not homogeneous.

A correlation is shown in the studies carried out in the lastfew years concerning activated protein C resistance (APCR) andthe mutations for MTHFR homozygosity and PTM [37–39].

In an Italian study, the association between IUGR (linkedto gestational hypertension or idiopatic) and FV Leiden, pro-thrombin mutation A20210, MTHFR TT 677 was examined in61 women with a history of fetal growth retardation and 93 con-trols of the same race. In all of the patients antiphospholipidanticoagulants, antithrombins, protein C and total and free pro-tein S antigen were also examined. The results underlined anassociation between growth retardation and the FV mutation(OR 6.9), with the prothrombin mutation (OR 5.9) and with TTMTHFR homozygosity (OR 1.5). The FVL and prothrombinmutations are shown to be independent factors for the parame-t

4tsp

ot4saGe

maO1to

yI

1

cpM

seem to increase the risk as shown in the studies of Wiener-Megnagi et al. [45–47]. On the other hand, in the study byProchazka et al., no statistically significant differences relatedto the analysis of the FVL were found [48].

The association between placenta abruption and throm-bophilias was underlined in the analysis very recently carriedout by Robertson et al. on 922 cases taken from seven studies,particularly with FV Leiden heterozygosity (OR 4.70; 95% CI1.13–19.59) and with the prothrombin heterozygosity (OR 7.71;95% CI 3.01–19.76) [19].

2. Treatment

The identification of a link between thrombophilic factors andadverse pregnancy outcome opens up treatment possibilities.

Anticoagulant therapy in pregnancy is suitable for the pre-vention and treatment of venous thromboembolism, of systemicembolism in women with mechanical valvular prosthesis and,often in combination with aspirin, to prevent pregnancy compli-cations in women with the antiphospholipid syndrome or otherthrombophilias and a history of previous pregnancy complica-tions. Currently the available antithrombotics include unfrac-tionated heparin, low molecular weight heparin, heparinoids,oral anticoagulants and aspirin. Direct inhibitors of thrombinasuch as the irudina are not used because they cross the placentaand their safety in pregnancy has not been established.

Cub

2

ppntsawrt

atddhtfcac[

i

ers analysed [40].In a case–control study on 493 babies born with IUGR and

72 controls, no correlation with maternal thrombophilic muta-ions was found. In a case–control study in 2003, however, theame author unexpectedly highlighted an association among thelasmatic levels of homocysteina [41,42].

In 2005, the same author carried out a case–control studyn 493 cases of IUGR and 472 controls in which the parame-ers analysed were the plasminogen activator inhibitor-1 (PAI-1)G/5G and the factor XIII Val134Leu variant, adding a familytudy on 250 cases. In all cases he evaluated the interactionsmong the polymorphisms, and also analysed the FV Leiden1691A and the Prothrombin G20210A. No strong association

merged with the IUGR [43].An association between IUGR and FVL or prothrombin gene

utations was found in a systematic review of the literature andmeta-analysis. The analysis of 10 case–control studies showedR equal to 2.7 (95% CI 1.3–5.5) for the FVL and 2.5 (95% CI.3–5.0) for the PTM. However, limitations must be attributedo the validity of the results because of the studies themselvesn which the figures were based [44].

In the 2006 review by Robertson et al., 195 cases were anal-sed, taken from five studies, in which the thrombophilic role inUGR was observed [19].

.4. Placental abruption

The studies aimed at evaluating a correlation between pla-ental abruption and thrombophilias seem to support such aossibility. Hyperhomocysteinemia, APCR, protein C mutation,THRF polymorphism and the combination of such factors

The Seventh ACCP (American College of Chest Physicians)onference on Antithrombotic and Thrombolytic Therapy drewp the guidelines for the prophylaxis of venous thromboem-olism in pregnancy [49].

.1. Unfractionated heparin

Unfractionated heparin is the drug of choice for the pro-hylaxis and treatment of thromboembolic manifestations inregnancy. Because of its molecular weight (MW), this doesot cross the fetus–placental barrier and therefore cannot haveeratogen effects and/or cause bleeding in the fetus [50]. Theide effects most frequently described which concern the motherre: osteoporosis, thrombocytopenia, allergy and hemorrhages,ith a 3% general incidence [50]. The percentage of hemor-

hages during therapy with unfractionated heparin is similar tohat recorded in non-pregnant patients (around 2%) [50].

Around 3% of the non-pregnant patients undergoing ther-py with unfractionated heparin develop an immune-mediatedhrombocytopenia. This form of thrombocytopenia should beifferentiated from the early, benign and transitory form whichevelops at the beginning of the therapy with unfractionatedeparin. In pregnant women who develop thrombocytopenia,he use of heparinoid danaparoid is recommended, which apartrom being an effective antithrombotic, does not cross the pla-enta and has less cross-reactivity than unfractionated heparinnd therefore less capacity of inducing a recurrent thrombo-ytopenia when compared to low molecular weight heparin49].

Long term therapy with heparin causes osteoporosis bothn animals and in humans [49]. Studies carried out in order to

M. De Santis et al. / Reproductive Toxicology 22 (2006) 227–233 231

appraise the risk of osteoporosis after 1 month of heparin ther-apy have shown that symptomatic vertebral fractures occurred inabout 2–3% of patients and a significant reduction of the bonedensity was recorded in 30% of the women undergoing longterm unfractionated heparin therapy [49,51].

2.2. Low molecular weight heparin

Low molecular weight heparin is shown to be comparable tounfractionated heparin with regard to effectiveness and safetyin the prophylaxis and in the treatment of venous thromboem-bolism [50]. It also does not cross the placenta and thereforecan be considered safe during pregnancy. When compared tounfractionated heparin it demonstrates a better biodisponibility,a longer half-life and a predictable anticoagulant activity, allow-ing, therefore, the use of fixed dosing based on individual bodyweight without the need to monitor the aPTT [50].

Some studies of pregnant women have shown that the lowmolecular weight heparins have a smaller incidence of throm-bocytopenia and a lower risk of osteoporosis [50].

With regard to the risk of osteoporosis, a group of womentreated with 5000 IU dalteparin was compared to another grouptreated with unfractionated heparin at a dose of 10,000 IU andpercentages of vertebral fractures of 3 and 15% were found,respectively [52]. In another randomised trial, the bone densityof a group of patients treated with dalteparin up to 3 years afterd[isdaa

db

aiwn[

va2tacgs

wf(mP

apy, blood loss during delivery was significantly higher (473 mlversus 365 ml, P = 0.02) [57].

Various studies have shown, however, that hemorrhagic com-plications while undergoing low molecular weight heparin ther-apy are very rare [50,51,56].

The FDA has recently reported cases of local haematomafollowing spinal or epidurale anesthesia in women treated withlow molecular weight heparin and it recommends effecting theanesthesia 10–12 h after the last heparin administration [58,59].

2.3. Oral anticoagulants

Oral anticoagulants cross the placental barrier and cause aspecific embryopathy characterized by nasal hypoplasia, tele-canthus, displasia of the epiphyses of the long bones and thecervical and lumbar vertebrae. This embryopathy has an inci-dence of 5–10% [60]. Some authors believe that the risk isparticularly high for exposures between the 6th and the 9th week[50]. A greater incidence of spontaneous abortions has also beenindicated in patients treated with coumarin derivatives whencompared with the women treated with unfractionated heparine.v. (34.6% versus 9.5%) [50].

In conclusion, malformations of the central nervous systemand of the eyes after exposure to the coumarin derivatives havealso been described independently of the gestational period [50].Some recent studies have hypothesized that the teratogenouseic

cdt

2

caaa

tatc

vepwdenrtp

elivery did not differ from that of an untreated control group53]. On the contrary, the bone density was significantly lowern women treated with unfractionated heparin both in compari-on to the untreated control group and the women treated withalteparin. In conclusion, cohort studies have not shown anyssociation between the use of low molecular weight heparin,nd osteoporosis [54].

In the absence of much data concerning the appropriate doseuring pregnancy, the generally administered dose is based onody weight.

The effectiveness of low molecule weight heparin in thentithrombotic prophylaxis of pregnancy has been demonstratedn various studies. Boda used nadroparin in seven pregnanciesith various types of hereditary thrombophilia. All of the preg-ancies had favourable results without maternal complications55].

Hunt used dalteparin in 32 patients with positive histories forenous thromboembolism and the only complication found waspostpartum vertebral collapse with a calculated incidence of

.2% [56]. Sanson, in a meta-analysis of all the studies regardinghe use of low molecular weight heparin in pregnancy, reportedn incidence of adverse fetus–neonatal events in 13.4% of thehildren with associated maternal pathologies and of 3.1% in theroup without associated pathologies, this last percentage beingimilar to that of the general population [54].

Considering vaginal delivery in women undergoing treatmentith low molecular weight heparin, Lindqvist and Dahlback

ound an increased risk of profuse hemorrhages (>600 ml)26.5% versus 7.4%, P = 0.001, OR 4.5) and of postpartum ane-ia (Hb <10 g/dl to 2 days from the birth) (29.4% versus 8.2%,< 0.001; OR 5.8). Furthermore, in the women undergoing ther-

ffect of the coumarin derivative is dose-dependent. In particulart would seem that a daily middle dose of 5 mg or less can beonsidered less dangerous for the fetus [50].

Moreover, because of the anticoagulant effect on the fetus, theoumarin derivatives can cause serious neonatal hemorrhagesuring the delivery when administered in the last few weeks ofhe pregnancy [50].

.4. Acetylsalicylic acid (ASA)

The studies on the use of acetylsalicylic acid in pregnancyoncern above all the prophylaxis of gestational hypertensionnd of the consequent intrauterine growth retardation, as wells the prevention of obstetric complications in women with thentiphospholipid syndrome.

All of the independently conducted studies have highlightedhat acetylsalicylic acid is a safe drug during pregnancy whendministered at low doses (50–150 mg). The risk of cardiopa-hy hypothesized by some studies has never been definitivelyonfirmed [50].

The effectiveness of low doses of ASA in the primary pre-ention of pre-eclampsia and fetal growth retardation has beenvaluated in many clinically controlled studies [50]. The pro-hylaxis with ASA (75 mg) in healthy pregnant women (12–32eeks) when compared to the placebo showed no significantifference in the incidence of pre-eclampsia, of preterm deliv-ry, of birth weight <1500 g or of intrauterine fetal death andeonatal death. No study has indicated an increased hemor-hagic risk for either the mother or the fetus. Low doses of ASAherapy are therefore not effective in the primary prevention ofre-eclampsia or its complications.

232 M. De Santis et al. / Reproductive Toxicology 22 (2006) 227–233

3. Conclusions

Although laboratory tests to identify the hereditary andacquired factors for thrombophilia are currently available in areasonably uniform way in developed countries, screening hasnot been recommended for all pregnant women. The patientssubmitted to such a study must be selected from those who havea positive history of thrombophilic and adverse obstetric out-comes in which confusing factors have been excluded.

The relationship between abortion and thrombophilia hasbeen demonstrated in the majority of studies concerning FVL,PTM and hyperhomocysteinemia but not so clearly in thoseexamining protein C, AT deficiency and MTHFR mutations.

The factor V Leiden mutation is a hereditary factor of throm-bophilia and has been identified in 5–6% of the Caucasianpopulation [61]. The frequency in the general population of thefactor V Leiden has been calculated as 3.6–6.0% in heterozy-gosity and 0.1–0.02% in homozygosity.

The relationship between abortion and the factor V Leidenhas been the subject of numerous studies and the results are notalways homogeneous, even in different studies carried out in thesame geographical area. In the heterozygous form there is a four-to eight-fold increased thrombotic risk which is 80-fold in thehomozygous form [61]. Studies related to the association withan adverse pregnancy outcome show a connection with abortion,but also with placental abruption and IUGR and highlight thesa

wtiw(tatt

iRafw1

ra

ppilwvs[

In conclusion, the strict collaboration between obstetricianand hematologist in cases with thrombophilic etiology can bethe key to the success of the pregnancy.

References

[1] Walker ID, Greaves M, Preston FE. On behalf of the haemostasis andthrombosis task force British Committee for standards in haematology.Investigation and management of heritable thrombophilia. Br J Haema-tology 2001;114:512–28.

[2] Martinelli I, De Stefano V, Taioli E. Inherited thrombophilia and firstvenous thromboembolism during pregnancy and puerperium. ThrombHaemost 2002;87:791–5.

[3] Triplett DA, Harris EN. Antiphospholipid antibodies and reproduction.Am J Reprod Immunol 1989;21:123–31.

[4] Sanson BJ, Friederich PW, Simioni P. The risk of abortion and stillbirthin antithrombin-, protein C-, and protein S-deficient women. ThrombHaemost 1996;75:387–8.

[5] Meinardi JR, Middeldorp S, de Kam MM, van Pampus EC, HamulyacK, Prins MH, et al. Increased risk for fetal loss in carrierrs of the factorV Leiden mutation. Ann Intern Med 1999;130:736–9.

[6] Wramsby ML, Sten-Linder M, Bremme K. Primary habitual abortionsare associated with high frequency of factor V Leiden mutation. FertilSteril 2000;74(5):987–91.

[7] Hashimoto K, Shizusawa Y. The factor V Leiden mutation in Japanesecouples with recurrent spontaneous abortion. Hum Reprod Embr1999;14:1872–4.

[8] Murphy RP, Donoghue C, Nallen RJ, D’Mello M. Prospective evaluationof the risk conferred by factor V Leiden and thermolabile methylen-

[

[

[

[

[

[

[

[

[

[

[

[

trong suspicion of a correlation between placental pathologiesnd fetal carriership of a hereditary defect [61–65].

The maternal mutation for the FV Leiden is not associatedith a greater risk of abortion, pre-eclampsia, placental abrup-

ion or SGA but nevertheless the fetal mutation FVL carriers associated with a higher incidence of pre-eclampsia amongomen of ethnic groups in comparison to those of white race

P = 0.04). Therefore, in patients with negative histories forhromboembolism the heterozygous FVL mutation is associ-ted with a lower risk of thromboembolism in pregnancy andherefore neither the screening of all pregnant women nor thereatment of the low risk carriers is recommended.

Regarding the prothrombin gene mutation, an Italian studydentified the association with abortion and the meta-analysis ofey et al., which has already been quoted, calculated an associ-tion between PTM and with both the recurrent abortion shownrom an OR calculated equal to 14.7 (95% CI 0.99–218) andith non-recurrent late abortion from an OR of 7.4 (95% CI.2–42.6) [62,63].

The association between hyperhomocysteinemia and recur-ent abortion was sustained in the meta-analysis of Nelen et al.nd from a subsequent case–control study by Gris et al. [64,65].

Rey’s meta-analysis did not underline a role for protein Colymorphisms in the abortion, AT deficiency and MTHFRolymorphisms [63]. Lindqvist et al., who studied 2480 womenn Sweden in the early stages of pregnancy, observed a preva-ence of activated resistance protein C equal to 11% [9]. Thereere no statistically significant differences between the acti-ated protein C resistant subgroup and the controls but pre-ented an eight-fold increased risk of venous thromboembolia63].

netetrahydrofolate reductase polymorphism in pregnancy. ArteriosclerThromb Vasc Biol 2000;20:266–70.

[9] Lindqvist PG, Svensson PJ, Marsaal K, Luterkort M, Dahlback B. Acti-vated protein C resistance (FV: Q506) and pregnancy. Thromb Haemost1999;81:532–7.

10] Rai R, Shlebak A, Cohen A, Backos M. Factor V Leiden and acquiredactivated protein C resistance among women with recurrent miscarriage.Hum Reprod 2001;16:961–5.

11] Santoro R, Iannaccaro P, Sottilotta G. Prothrombotic gene mutations inwomen with recurrent abortions and intrauterine fetal death. MinervaGinecol 2005;57:447–50.

12] Gluck CJ, Pranikoff J, Aregawi D, Haque M, Zhu B, Yracy T, et al. Thefactor V Leiden mutation, high factor VIII, and high plasminogen acti-vator inhibitor activity: etiologies for sporadic miscarriages. Metabolism2005;54:1345–9.

13] Lissalde-Lavigne G, Cochery-Nouvellon E. The association betweenhereditary thrombophilias and pregnancy loss. Haematologica2005;90:1223–30.

14] Sugiura M. Pregnancy and delivery in protein C-deficiency. Curr DrugTargets 2005;6:577–83.

15] Hossain N, Shamsi T, Soomro N. Frequency of thrombophilia in patientswith adverse pregnancy outcome. J Pak Med Assoc 2005;55:245–7.

16] Behjati R, Modarressi MH, Jeddi-Therani M, Dokoohaki P, GashemiJ, Zarnani AH, et al. Thrombophilic mutations in Iranian patientswith infertility and recurrent spontaneous abortion. Ann Hematol 2006;1:1–4.

17] Jivray S, Rai R, Underwood J, Regan L. Genetic thrombophilic muta-tions among couples with recurrent miscarriage. Hum Reprod 2006:23.

18] Weintraub AY, Sheiner E, Levy A, Yerushalmi R, Mazor M. Pregnancycomplications in women with inherited thrombophilia. Arch GynecolObstet; 2006.

19] Robertson L, Wu O, Langhorne P. Thrombophilia in pregnancy: aaystematic review. Br J Haematol 2006;132:171–96.

20] Dekker GA, de Vries JI, Doelitzsch PM. Underlying disorders asso-ciated with severe early onset preeclampsia. Am J Obstet Gnnecol1995;173:1042–8.

21] Dekker GA. Risk factor for preeclampsia. Clin Obstet Gynecol1999;42:422–35.

M. De Santis et al. / Reproductive Toxicology 22 (2006) 227–233 233

[22] Dizon-Townson DS, Nelson LM, Easton K, Ward K. The factor V Lei-den mutation may predispose women to severe preeclampsia. Am JObstet Gynecol 1996;175:902–5.

[23] Grandone E, Margaglione M, Colaizzo D. Factor V Leiden, C > TMTHFR polymorphism and genetic susceptibility to preeclampsia.Thromb Haemost 1997;77:1052–4.

[24] Nagy B, Toth B, Rigo J, Karadi I, Romics L, Papp Z. Detection offactor V Leiden mutation in severe preeclamtic Hungarian women. ClinGenet 1998;53:478–81.

[25] Kupferminc MJ, Fait G, Many A, Gordon D, Eldor A, Lessing JB.Severe preeclampsia and high frequency of genetic thrombophilic muta-tions. Obstet Gynecol 2000;96:45–9.

[26] Lindoff C, Ingemarsson I, Martinsson G, Segelmark M. Preeclampsia isassociated with a reduced response to activated protein C. Am J ObstetGynecol 1997;176:457–60.

[27] Sohda A, Arinami T, Hamada H. Methylenetetrahydrofolate reductasepolymorphism and pre-eclampsia. J Med Genet 1997;34:525–6.

[28] Rajkovic A, Catalano PM, Malinow MR. Elevated homocysteine levelswith preeclampsia. Thromb Haemost 1997;77:1052–4.

[29] El-Khairy L, Vollset SE, Refsum H, Ueland PM. Plasma total cysteine,pregnancy complications, and adverse pregnancy outcomes: the Hord-land Homocysteine Study. Am J Clin Nutr 2003;77:467–72.

[30] Van Pampus MG, Dekker GA, Wolf H, Huijgens PC. High prevalence ofhemostatic abnormalities in women with a history of severe preeclamp-sia. Am J Obstet Gynecol 2000;182:989–90.

[31] Van Pampus MG, Wolf H, Koopman MM, van der Ende A, Buller HR,Reitsma PH. Prothrombin 20210 G: a mutation and factor V Leidenmutation in women with a history of severe preeclampsia and (H)ELLPsyndrome. Hypertens Pregnancy 2001;20:291–8.

[32] Kosmas IP, Tatsioni A, Ioannidis JP. Association of Leiden mutationin factor V gene with hypertension in pregnancy and preeclampsia: a

[

[

[

[

[

[

[

[

[

[

[

[44] Howley HE, Walker M, Rodger MA. A systematic review of the associa-tion between factor V Leiden or prothrombin gene variant and intrauter-ine growth restriction. Am J Obstet Gynecol 2005;192:694–708.

[45] Wiener-Megnagi Z, Ben-Shlomo I, Goldberg Y, Shalev E. Resistance toactivated protein C and the Leiden mutation: high prevalence in patientswith abruptio placentae. Am J Obstet Gynecol 1998;179:1565–7.

[46] Van der Molen EF, Verbruggen B. Hyperhomocysteinemia and otherthrombotic risk factors in women with placental vasculopathy. BJOG2000;107:785–91.

[47] Facchinetti F, Marozio L, Grandone E, Pizzi C, Volpe A, Benedetto C.Thrombophilic mutations are a main risk factor for placental abruption.Haematologica 2003;88:785–8.

[48] Prochazka M, Happach C, Marsal K. Factor V Leiden in pregnanciescomplicated by placental abruption. BJOG 2003;110:462–6.

[49] Bates SM, Greer IA, Hirsh J, Giinsberg JS. Use of antithrombotic agentsduring pregnancy. The Seventh ACCP Conference on Antithromboticand Thrombolytic Therapy. Chest 2004;126:627S–44S.

[50] Grandone E, Guiotto G, Di Minno G. Linee guida sull’uso dei farmacianticoagulanti e antiaggreganti in ostetricia e ginecologia. Haematologica2002;87(S12):1–20.

[51] Greer IA, Nelson-Pierci C. Low-molecular-weight heparins for throm-boprophylaxis and treatment of venous thromboembolism in pregnancy:a systematic review of safety and efficacy. Blood 2005;106:401–7.

[52] Monreal M, Lafoz E, Olive A, Del Rio L, Vedia C. Comparison of sub-cutaneos unfractionated heparin with a low-molecular weight heparin(Fragmin) in patients with venous thromoembolism and contraindica-tions for coumarin. Thromb Haemost 1994;71:7–11.

[53] Pettila V, Leinonen P, Markkola A, Hiilesmaa V, Kaaja R. Post-partum bone mineral density in women treated for thromboprophy-laxis with unfractionated heparin or LMW heparin. Thromb Haemost2002;87:182–6.

[

[

[

[

[

[

[

[

[

[

[

[

meta-analysis. J Hypertens 2003;21:1221–8.33] Kosmas IP, Tatsioni A, Ioannidis JP. Association of C677T polymor-

phism in the methylenetetrahydrofolate reductase gene with hyper-tension in pregnancy and preeclampsia: a meta-analysis. J Hypertens2004;22:1655–62.

34] D’Anna R, Baviera G, Corrado F, Ientile R, Granese D, Stella NC.Plasma homocysteine in early and late pregnancies complicated withpreeclampsia and isolated intrauterine growth restriction. Acta ObstetGynecol Scand 2004;83:155–8.

35] Mello G, Parretti E, Marozio L, Pizzi C, Lojacono A, Frusca T, et al.Trombophilia is significantly associated with severe preeclampsia: resultsof a large-scale, case-controlled study. Hypertension 2005;46:1252–3.

36] Sibai BM. Thrombophilia and severe preeclampsia: time to screen andtreat in future pregnancies? Hypertension 2005;46:1252–5.

37] Clark P, Walker ID, Greer I. Acquired activated protein-C resistance inpregnancy and association with increased thrombin generation and fetalweight. Lancet 1999;353:292–3.

38] Kupferminc MJ, Eldor A, Steinman N. Increased frequency of genetictrombophilia in women with complications of pregnancy. N Engl J Med1999;340:9–130.

39] Kupferminc MJ, Peri H, Zwang E. High prevalence of the prothromingene mutation in women with intrauterine growth retardation, abrup-tio placentae and second trimester loss. Acta Obstet Gynecol Scand2000;79:963–7.

40] Martinelli P, Grandone E, Colaizzo D, Paladini D, Sciannama N, Mar-gaglione M, et al. Familial thrombophilia and the occurrence of fetalgrowth restriction. Haematologica 2001;86:428–31.

41] Infante-Rivard C, Rivard GE, Yotov WV. Absence of association ofthrombophilia polymorphism with intrauterine growth restriction. N EnglJ Med 2002;347:19–25.

42] Infante-Rivard C, Rivard GE, Gauthiter R, Theoret Y. Unexpected rela-tionship between plasma homocysteine and intrauterine growth restric-tion. Clin Chem 2003;49:1476–82.

43] Infante-Rivard C, Rivard GE, Guiguet M, Gauthier R. Throm-bophlic polymorphisms and intrauterine growth restriction. Throm-bophilic polymorphism and intrauterine growth restriction. Epidemiology2005;16:281–7.

54] Sanson BJ, Lensing AW, Prins MH, Ginsberg JS, Barkagan ZS, Lavenne-Pardonge E, et al. Safety of low-molecular-weight heparin in pregnancy:a systematic review. Thromb Haemost 1999;81(5):668–72.

55] Boda Z, Laszlo P, Rejto L, Tornai I, Pflieger G, Blasko G, et al.Low molecular weight heparin as thromboprophylaxis in familial throm-bophilia during the whole period of pregnancy. Thromb Haemost1996;76(1):128.

56] Hunt BJ, Doughty HA, Majumdar, Copplestone A, Kerslake S, BuchananN, et al. Thromboprophylaxis with low molecular weight heparin (Frag-min) in high risk pregnancies. Throm Haemost 1997;77(1):39–43.

57] Lindqvist PG, Dahlback B. Bleeding complications associated withlow molecular weight heparin prophylaxis during pregnancy. ThrombHaemost 2000;84(1):140–1.

58] Reports of epidural or spinal haematomas with the concurrent use oflow molecular weight heparin and spinal/epidural anesthesia or spinalpuncture. Food and Drug Administration, Washington; December 1997.

59] Sugiyama T, Itoh M, Ohtawa M, Natsuga T. Study on neutralizationof low molecular weight heparin (LHG) by protamine sulfate and itsneutralization characteristics. Thromb Res 1992;68(2):119–29.

60] Chan WS, Anand S, Ginsberg JS. Anticoagulation of pregnant womenwith mechanical heart valves: a systematic review of the literature. ArchIntern Med 2000;160(2):191–6.

61] Calderwood CJ, Geer IA. The role of factor V Leiden in maternal healthand the outcome of pregnancy. Curr Drug Targets 2005;6:567–76.

62] Martinelli I, Taioli E, Cetin I, Marinoni A, Gerosa S, Villa MV, et al.Mutations in coagulation factors in women with unexplained late fetalloss. N Engl J Med 2000;343(14):1015–8.

63] Rey E, Kahn SR, David M, Shrier I. Thrombophilic disorders and fetalloss: a meta-analysis. Lancet 2003;361(9361):901–8.

64] Nelen WL, Blom HJ, Steegers EA, den Heijer M, Eskes TK. Hyper-homocysteinemia and recurrent early pregnancy loss: a meta-analysis.Fertil Steril 2000;74(6):1196–9.

65] Gris JC, Perneger TV, Quere I, Mercier E, Fabbro-Peray P, Lavigne-Lissalde G, et al. Antiphospholipid/antiprotein antibodies, hemostasis-related autoantibodies, and plasma homocysteine as risk factors fora first early pregnancy loss: a matched case–control study. Blood2003;102(10):3504–13.