Antiphospholipid antibodies and pregnancy

Best Practice & Research Clinical Haematology Vol. 16, No. 2, pp. 211 –225, 2003 doi:10.1053/ybeha.2003.247

6 Antiphospholipid antibodies and pregnancy Monica Galli*

MD, PhD

Tiziano Barbui

MD

Head of Department Department of Haematology, Ospedali Riuniti, L.go Barozzi 1, Bergamo 24128, Italy

Antiphospholipid antibodies, i.e. lupus anticoagulants and anticardiolipin antibodies, are associated with obstetric complications. Fetal death and recurrent spontaneous abortions represent the obstetric criteria of the antiphospholipid syndrome. They occur with similar frequences and have an overall prevalence of 15 – 20%. Lupus anticoagulants carry a risk 3.0 to 4.8 times, and anticardiolipin antibodies 0.86 to 20 times higher than controls. The mechanism(s) by which antiphospholipid antibodies cause these events still has to be defined: thrombosis in the placental vessels, and impairment of embryonic implantation have been proposed. Unfractionated or low-molecular-weight heparin, alone or in combination with low-dose aspirin, represent the current standard treatment of pregnant antiphospholipid-positive women for preventing recurrent obstetric complications. Upon treatment, the live birth rate increases from 0– 40% to 70 – 80%. However, there is still an excessive frequency of maternal and/or fetal complications, indicating the necessity of a better calibration of the dosage, duration and timing of administration of heparin treatment. Key words: antiphospholipid syndrome; recurrent abortions; pregnancy loss; antiphospholipid antibodies.

Arterial and venous thrombosis are the most common and clinically relevant events of the so-called antiphospholipid syndrome, as they are reported in approximately one third of patients with antiphospholipid antibodies.1 The characteristics of thrombosis in the antiphospholipid syndrome are as follows: deep vein thrombosis (lower limbs) with or without pulmonary embolism; arterial thrombosis (cerebral ischaemia, transient ischaemic attacks, myocardial infarction); overall prevalence: approximately 30%; annual rate of first event: about 1%; annual rate of recurrence in patients not receiving oral anticoagulation: 10 –29%. The antiphospholipid syndrome is characterized also by obstetric complications2, other clinical manifestations have been variably associated with the syndrome * Corresponding author. Tel.: þ39-035-269-492; Fax: þ39-035-266-667. E-mail address: [email protected] (M. Galli). 1521-6926/03/$ - see front matter Q 2003 Elsevier Science Ltd. All rights reserved.

212 M. Galli and T. Barbui

(reviewed in Ref. 3). Clinical associations with antiphospholipid antibodies not included in the antiphospholipid syndrome are listed below: Central nervous system: Chorea, migraine, psychosis, epilepsy, sensorineural hearing loss, transverse myelopathy, cognitive impairment, pseudotumor cerebri, multiple sclerosis-like syndrome, hypoperfusion on SPECT scanning Gastrointestinal: Hepatic necrosis, acalcolous cholecystitis Vascular disease: Atherosclerosis, cardiac valvular disease, failed angioplasty, diastolic dysfunction, cardiomiopathy, Buerger’s disease Skin: Livedo reticularis, cutaneous ulcers, Dego’s disease, splinter haemorrhages Bone: Avascular necrosis, bone marrow necrosis Renal: Renal insufficiency, renal artery stenosis Pulmonary: Pulmonary hypertension, ARDS Endocrine: Adrenal failure, hypopituitarism Haematological: Thrombocytopenia, autoimmune haemolytic anaemia Antiphospholipid syndrome may occur either isolated (‘primary’ syndrome)4, or in the setting of an underlying disease, mainly systemic lupus erythematosus (‘secondary’ syndrome).5 An international consensus on classification criteria for definite antiphospholipid syndrome was published in 1999.6 Antiphospholipid antibodies are a rather wide and heterogeneous family of IgG, and/or IgM, less frequently, also IgA, immunoglobulins, long considered to react with negatively-charged phospholipids. Lupus anticoagulants and anticardiolipin antibodies are the first two such antibodies to be described. Lupus anticoagulants behave as acquired inhibitors of coagulation, which prolong the phospholipid-dependent coagulation reactions7, whereas anticardiolipin antibodies react with anionic phospholipids in solid-phase immunoassays.8

IMMUNOLOGICAL AND FUNCTIONAL PROPERTIES OF ANTIPHOSPHOLIPID ANTIBODIES In the 1990s, work from different laboratories made it clear that antiphospholipid antibodies do not recognize anionic phospholipids but rather plasma proteins bound to suitable anionic (not necessarily, phospholipid) surfaces. Among them, b2-glycoprotein I, and (human) prothrombin are the most common and investigated antigenic targets. Antigenic targets of the antiphospholipid antibodies are listed below: b2-glycoprotein I Prothrombin Protein C Protein S Annexin V Low- and high-molecular-weight kininogens Factor XII Tissue-type plasminogen activator b2-glycoprotein I is required by the great majority of anticardiolipin antibodies to react with cardiolipin in immunoassays.7 – 11 Specific subgroups of

Antiphospholipid antibodies and pregnancy 213

anti-b2-glycoprotein I12 and antiprothrombin13 antibodies are responsible for the lupus anticoagulant activity in phospholipid-dependent coagulation tests. b2-glycoprotein I is a single-chain glycoprotein with a molecular weight of 50 KDa (unreduced). It has a plasma concentration of approximately 4 mM (about 0.2 mg/ml), and its site of synthesis is probably the liver. Despite its anticoagulant behaviuor ‘in vitro’, its physiological role is unclear. b2-glycoprotein I is rich in proline and cysteine and contains 11 disulfide bridges. It consists largely of repeated units of approximately 60 amino acids each, the so-called ‘sushi domains’.14 The fifth domain is essential for binding to anionic phospholipids15, whereas the first domain contains significant epitopes for the immune recognition by anticardiolipin antibodies.16 For this reason, anticardiolipin antibodies would better be re-named anti-b2-glycoprotein I antibodies. The prevalence of anti-b2-glycoprotein I antibodies in antiphospholipid-positive patients approximates 50%.17 Antiprothrombin antibodies are of the non-neutralizing type because they do not inhibit the conversion of prothrombin to thrombin but rather cause a variable degree of hypoprothrombinaemia.18 They are present in approximately 50% of antiphospholipid-positive patients.19 The epitopes recognized by antiprothrombin antibodies have not yet been fully defined. Binding to both prethrombin I (the carboxy-terminal segment of prothrombin)18,20 and fragment 1 (the amino-terminal segment of prothrombin)20 has been reported, whereas no reactivity with thrombin has been shown.20,21 This suggests that most antiprothrombin antibodies are of polyor oligo-clonal nature. Soubgroups of anti-b2-glycoprotein I and antiprothrombin antibodies display lupus anticoagulant activity and affect the same coagulation tests through different mechanisms.22,23 In most patients, the anticoagulant activity is probably caused by a combination of both inhibitors, as shown by Horbach and coworkers.24 In their study of 28 lupus anticoagulant-positive plasma samples, the anticoagulant activity was totally dependent on antiprothrombin or anti-b2-glycoprotein I antibodies in four and seven cases, respectively. In the other 17 plasmas both antibodies contributed to lupus anticoagulant activity.

OTHER ANTIGENIC TARGETS AND PROPOSED MECHANISMS OF THROMBOSIS Other proteins recognized by antiphospholipid antibodies have also been listed above.25 – 29 As most of them are involved in the regulation of the coagulation processes, it is conceivable that antibodies that reduce their plasma concentration and/ or hamper their function may produce an imbalance between the pro- and anticoagulant systems. This might represent the pathophysiological background underlying the increased thrombotic risk of antiphospholipid-positive patients. Impairment of the anticoagulant activity of the protein C system may be caused by antibodies with antigenic specificities other than protein C or protein S. Antibodies to phosphatidylethanolamine (PE) inhibit the anticoagulant activity of activated protein C in the plasma of antiphospholipid-positive patients.30 Marciniak and Romond31 elegantly demonstrated that IgGs with lupus anticoagulant activity hamper the inactivation of activated factor V by endogenous protein C. In plasma, this effect is caused by anti-b2glycoprotein I but not by antiprothrombin antibodies.32 Anti-b2-glycoprotein I antibodies have been shown to decrease the phospholipid-dependent inhibitory activity of tissue factor pathway inhibitor on factor X activation, leading to increased

214 M. Galli and T. Barbui

extrinsic generation of factor Xa.33 Yet another possible mechanism of thrombosis was suggested by Dobado-Berrios et al34 who found increased levels of tissue factor messenger RNA in mononuclear blood cells of patients with primary antiphospholipid syndrome. The highest values occurred in patients with a history of thrombosis. Finally, because thrombosis is a multifactorial event, factors other than antiphospholipid antibodies may contribute to its occurrence. The observation that the prevalence of the G1691A mutation of the factor V gene is higher in anticoagulant-positive patients with than without venous thrombosis sustains this concept.35

LABORATORY DIAGNOSIS OF ANTIPHOSPHOLIPID ANTIBODIES Detection by immunoassays Anticardiolipin antibodies have been detected by different techniques: in the first place, the Wassermann reaction, and subsequently radio- or enzyme-linked immunoassays (ELISAs).36 ELISAs with cardiolipin or other anionic phospholipids (or mixtures of phospholipids) in solid phase are the most commonly used methods as they allow the quantitative or semi-quantitative measurement and the identification of the antibody isotype(s). A number of kits are nowadays commercially available. However, despite much effort and several international standardization workshops37 – 40, a considerable degree of interlaboratory variation still exists. This is partially due to the high degree of sensitivity and a still rather weak specificity of the ELISAs for the detection of anticardiolipin antibodies. Specificity may be improved by increasing the cut-off points because clinically relevant associations are reported for persistently positive anticardiolipin antibodies only at medium to high levels. Increased titres of these antibodies may be found in a number of conditions unrelated to the antiphospholipid syndrome, mostly bacterial and viral infections.41 The use of ELISA plates coated with other antigens may help to overcome this problem. Because b2-glycoprotein I and prothrombin are presently reckoned to be the antigenic targets of most antiphospholipid antibodies, ELISAs with these proteins in solid phase are becoming increasingly used. Their major advantage is that they do not detect ‘infection-driven’ antiphospholipid antibodies, thus increasing the specificity of the assays towards the clinical manifestations of the syndrome. However, care must be exercised when carrying out these assays: the choice of the ELISA plate (such as g-irradiated polystyrene or high-binding polyvinylchloride plates) and the mode of presentation of the antigen (alone or bound to anionic phospholipids) are crucial to the immune recognition, owing to the low-affinity nature of anti-b2-glycoprotein I and antiprothrombin antibodies. Moreover, for both types of antibodies assay standardization is far from optimal, and their clinical association with thrombosis or pregnancy loss remains weak. Before replacing ELISAs for anticardiolipin detection with these newer assays, robust laboratory and clinical validation is required. Detection by coagulation tests Lupus anticoagulants are acquired inhibitors of coagulation that prolong the coagulation times of phospholipid-dependent coagulation tests. This prolongation is not corrected by mixing the patient’s plasma with normal control plasma, whereas a variable degree of correction is obtained upon increasing the concentration of

Antiphospholipid antibodies and pregnancy 215

anionic phospholipids in the test system. These are the diagnostic criteria first proposed by the Scientific Subcommittee for the Standardization of Lupus Anticoagulants/ Phospholipid-Dependent Antibodies in 198242 and subsequently updated.43 Despite the large number of screening and confirmatory tests available, no single assay is 100% sensitive and specific, owing to the heterogeneity of lupus anticoagulants. Thus, more than one assay must be carried out before excluding the presence of these inhibitors. Another important point is the pre-analytical handling of the plasma sample, which may largely affect the correct interpretation of the test result. At present, the recommended laboratory workout of patients suspected of having the antiphospholipid syndrome requires the performance of screening tests (among them, a sensitive activated partial thromboplastin time, the kaolin clotting time, and the dilute Russell’s viper venom time), with a mixing test if the assay is prolonged and a confirmation step with a platelet or a phospholipid neutralization procedure, and the detection of IgG (and, possibly, IgM) anticardiolipin antibodies by ELISA. The inclusion of anti-b2-glycoprotein I and antiprothrombin antibodies measurement is still under clinical validation.44

EPIDEMIOLOGY OF THE OBSTETRIC COMPLICATIONS Clinical obstetric criteria for the antiphospholipid syndrome6 are given in Table 1. Although not included among them, intrauterine growth retardation, pre-eclampsia, and prematurity are other common obstetric events.45 Approximately 15 – 20% of women with antiphospholipid antibodies have obstetric complications46, and up to 50 –75% of their pregnancies have a poor outcome.47

Table 1. Preliminary obstetric criteria for the classification of the antiphospholipid syndrome. Definite syndrome is present when at least one of the clinical and one of the laboratory criteria are met. Clinical criteria 1 One or more unexplained fetal deaths of a morphologically normal fetus at or beyond the 10th week of gestation, with normal fetal morphology documented by ultrasound or by direct examination, or 2 One or more premature births of a morphologically normal neonate at or beyond the 34th week of gestation because of severe pre-eclampsia or eclampsia, or severe placental insufficiency, or 3 Three or more unexplained consecutive spontaneous abortions before the 10th week of gestation, with maternal, anatomical, or hormonal abnormalities and paternal or maternal chromosomal causes excluded Laboratory criteria 1 Anticardiolipin antibody of IgG and/or IgM isotype in blood, present in medium or high titre, on two or more occasions, at least 6 weeks apart, measured by standard ELISA for b2-glycoprotein I-dependent anticardiolipin antibodies 2 Lupus anticoagulants present in plasma on two or more occasions, at least 6 weeks apart, detected according to the guidelines of the SSC for the Standardization of Lupus Anticoagulants/Phospholipid-dependent Antibodies

216 M. Galli and T. Barbui

From a developmental biology viewpoint, pregnancy losses may be divided into: 1. pre-embryonic, which occur within the 5th week from the first day of the last menstrual period; 2. embryonic, which occur from the 5th to the 9th week; 3. fetal, which occur from the 10th week. Because the majority of pre-embryonic and embryonic losses are commonly diagnosed between the 10th and 12th week—when the clinical symptoms develop— the only proofs of fetal death are the demonstration of the fetal cardiac activity before miscarriage, the passage of a conceptus measuring at least 3 cm, or the ultrasonographic measurement of a dead conceptus of a size consistent with a fetus. In the general population, the great majority of pregnancy losses occur during the preembryonic and embryonic periods.48 Women with documented heritable thrombophilia or essential thrombocythaemia have an increased prevalence of pregnancy loss49 – 51 which occurs mostly within the 10th week.50,51 On the contrary, antiphospholipidpositive women are more prone to fetal death.52 A summary of the prevalence and timing of pregnancy losses in these clinical conditions is reported in Table 2. The strength of the association between obstetric complications and antiphospholipid antibodies has been estimated by cross-sectional and case – control studies53 – 56: the presence of lupus anticoagulants carries an odds ratio ranging from 3.0 to 4.8, and that of anticardiolipin antibodies goes from 0.86 to 20.0. These figures vary, according to the primary or secondary nature of the antiphospholipid syndrome, the mother’s clinical history, and the titres, isotypes and/or types of antiphospholipid antibody investigated. The highest risk of pregnancy loss has been reported in patients with systemic lupus erythematosus.55 The maternal history of pregnancy loss represents a strong risk factor for further adverse obstetric events, whereas the presence of lupus anticoagulants or anticardiolipin antibodies does not seem to negatively affect the outcome of the first pregnancy.54 Whether a relationship between the antibody titer and the risk of obstetric complications exists still has to be clarified.46,57 The association of antiphospholipid antibodies different from anticardiolipin antibodies or lupus anticoagulants with fetal loss has been investigated by the NOHA study56: lupus anticoagulants, IgG anti-b2-glycoprotein I, IgG anti-annexin V and IgM antiphosphtidylethanolamine antibodies were found to be independent risk factors for unexplained primary early fetal loss. The same markers were prospectively found to be associated with a significant risk of fetal loss during subsequent pregnancies.56

Table 2. Epidemiology and timing of pregnancy loss in different clinical conditions. Pregnancy loss (%)

General population Antiphospholipid syndrome Essential thrombocythaemia Heritable thrombophiliaa a

Overall prevalence (%)

,10th week

.10th week

10– 20 50– 75 43 22– 29

87 50 83 86

13 50 17 14

Reference 48 47,52 51 49,50

Quantitative and/or qualitative defects of antithrombin, protein C, and protein S, the G1691A mutation of factor V gene, the G20210A mutation of prothrombin gene.

Antiphospholipid antibodies and pregnancy 217

Such a role, however, has been disputed by other authors.58,59 At present, the investigation of antiphospholipid antibodies other than lupus anticoagulants and anticardiolipin antibodies is not suggested in the routine workout of women with obstetric complications. PATHOGENESIS OF THE OBSTETRIC COMPLICATIONS Following the first description of intrauterine fetal death associated with placental infarction in a woman with lupus anticoagulants60, the hypothesis that thrombotic events occurring at the placental vasculature eventually lead to early abortion or fetal death was raised. Histopathological analysis of aborted placentae has been performed in relatively few women with antiphospholipid antibodies. Extensive infarctual areas, decrease in vasculosyncytial membranes, increase in fibrosis and in hypovascular villi have been reported61,62, which are consistent with placental hypoxia. Experimental models of antiphospholipid syndrome following mice immunization with human polyclonal and monoclonal anticardiolipin antibodies62,63 were characterized by a low fecundity rate and increased fetal resorption rate (equivalent to abortions in humans). These findings support the notion that—at least some—antiphospholipid antibodies may play a role in the pathogenesis of the obstetric complications of the antiphospholipid syndrome. The mechanism by which antiphospholipid antibodies may cause miscarriages still has to be defined. Placental thrombosis cannot be taken as the only and unifying explanation, because the described histopathological abnormalities are also found in the placentae of women without antiphospholipid antibodies. Alternative mechanisms have been proposed; these take into account the possibility that antiphospholipid antibodies may have a direct adverse effect on the process of embryonic implantation. Di Simone and coworkers64 – 66 have shown that IgG immunoglobulins from patients with antiphospholipid syndrome bind to throphoblast cells and reduce their in vitro invasiveness, differentiation and gonadotropin secretion. Two human monoclonal antiphospholipid antibodies, a b2-GPI-independent IgG and a b2-GPI-dependent IgM, reproduced the same effects, which were reversed by interleukin-3, and low-molecular weight heparin, but not by aspirin.65,66

TREATMENT OF THE OBSTETRIC COMPLICATIONS The thrombosis of the placental vasculature and the defective embryonic implantation represent the biological rationale for the use of unfractionated and low-molecularweight heparins in the treatment of the obstetric complications of women with antiphospholipid antibodies. The live birth rate of women with antiphospholipid syndrome has been prospectively reported to be as low as 10% (0% in those with lupus anticoagulants, and 40% in those with anticardiolipin antibodies alone).67 The combination of heparin plus low-dose aspirin has been demonstrated by randomized, controlled clinical trials to significantly increase such a rate up to 71 –80% and to be superior to low-dose aspirin alone (live birth rate about 40%).68,69 Substitution of unfractionated heparin with low-molecular-weight heparin led to similar results.70 Heparin administration is well tolerated and, in general, it does not decrease bone density.71 Despite the high live birth rate, heparin-treated pregnancies of antiphospholipidpositive women are still characterized by an excessive frequency of maternal and/or

218 M. Galli and T. Barbui

fetal complications. Two studies on approximately 200 women reported problems in about 42– 50% of the women and in 28– 35% of the newborns.70,72 They were represented mainly by premature delivery (before the 37th week), gestational hypertension, antepartum haemorrhage, growth retardation and oligohydramnios. Approximately 50% of the pregnancies underwent (in most cases, elective) caesarean section. The outcome of newborns has been investigated by Ruffatti and coworkers73, who evaluated 55 infants born to 53 antiphospholipid-positive women. They were delivered from the 25th and 40th week of gestation; their birth weight ranged from 800 to 4000 g and had a mean Apgar score of 9.6 at 5 minutes. Twelve of them needed intensive care treatment due to prematurity. None of the newborns had thrombotic complications or malformations. Finally, no antiphospholipid-related manifestation developed during a follow-up ranging from 1.3 to 5.6 years. Steroids represented the first treatment option for the prevention of recurrent miscarriage in women with antiphospholipid antibodies.74 Their use progressively came into disfavour, because several clinical studies demonstrated that steroids are not as efficacious as heparin and aspirin and are associated with a high rate of feto/maternal complications (i.e. prematurity, hypertension and diabetes mellitus).75 – 78 Recently, Empson and coworkers79 performed a systematic review of randomized therapeutic trials on antiphospholipid-positive women with recurrent pregnancy loss published until 1999. Ten trials on 627 women were evaluated: three trials on aspirin alone showed no significant reduction in pregnancy loss; heparin plus aspirin significantly reduced pregnancy loss compared to aspirin alone; prednisone plus aspirin resulted in a significant increase in prematurity, and no significant reduction in pregnancy loss. High-dose immunoglubulin therapy has been reported to be effective in some cases, even though its precise mechanism(s) still has to be elucidated. This treatment has been evaluated by a multicentre, placebo-controlled pilot study carried out in 16 antiphospholipid-positive women with or without a history of recurrent miscarriages, and/or thromboembolic events.80 All of them received heparin and low-dose aspirin and were randomized either to monthly courses of immunoglobulins (1 g/kg body weight for 2 days) or placebo until the 36th week of gestation. The two groups were similar with respect to age, gravidity, number of previous pregnancy losses, and gestational age at the initiation of the treatment. All women delivered live-birth babies after the 32th week of gestation. No differences were observed with respect to the rate of pre-eclampsia, and placental insufficiency. There were fewer cases of fetal growth restriction, and neonatal intensive care unit admission among neonates of the high-dose immunoglobulin treatment group; these differences, however, were not statistically significant. A recent meta-analysis failed to demonstrate that intravenous immunoglobulins are of benefit in women with unexplained recurrent miscarriages.81 Therefore, at the present time intravenous immunoglobulins may be considered only for the treatment of women who cannot receive heparin and low-dose aspirin, or of those with proven inefficacy of this therapy. The sum of these findings indicates that heparin plus low-dose aspirin, despite being the standard treatment of a pregnant antiphospholipid-positive woman with a poor obstetric history, still requires to be better calibrated in terms of dosage, duration and timing of administration (Table 3). In the randomized clinical trials so far conducted, aspirin was started as soon as the pregnancy test was positive and heparin was introduced when fetal heart activity was noted by ultrasound.69,70 Therefore, the question whether this treatment should be commenced even before conception, in order to reduce the rate of pre-embryonic and embryonic loss, remains unanswered. Indeed, few and conflicting experimental data exist as to the effect of heparin and/or

Antiphospholipid antibodies and pregnancy 219

Table 3. Treatment options of antiphospholipid-positive women during pregnancy. Clinical scenarios

Therapy

Open questions

Asymptomatic

None

Aspirin?

, 3 abortions; no fetal death No previous thrombosis

Aspirin, 75 mg/day

No therapy? Prophylactic heparin instead of aspirin? Duration? With aspirin? Duration?

With previous thrombosis

3 abortions, or 1 fetal death No previous thrombosis

With previous thrombosis

a

LMW or UF heparina, at therapeutic dosage; warfarin during puerperium, PT-INR 2.0–3.0 LMW or HF heparin, at prophylactic dosage LMW or UF heparina, at therapeutic dosage;warfarin during puerperium, PT-INR 2.0–3.0

Fixed or adjusted dosage? Duration? Extended to puerperium? With aspirin? With aspirin? Duration?

LMW, low molecular weight; UF, unfractionated.

aspirin to facilitate the process of embryonic implantation. In these trials therapy was given up to the 34th week of gestation because of the risk of osteopenia. However, it is now recommended that administration of heparin and low-dose aspirin be continued until delivery in order to reduce the rate of late complications.82 Again, the best timing and dosage of heparin during labour, in relation to the mother’s haemorrhagic risk, has not yet been established. Treatment should be continued during the puerperium only in antiphospholipid-positive women with known risk factors for thrombosis. The observation that the use of dose-adjusted or a fixed dose of heparin led to similar rates of live births suggests that heparin possibly acts during pregnancy not only through its antithrombotic activity. Pregnant antiphospholipid-positive women should undergo regular platelet counts, at least in the first weeks of heparin administration, owing to the risk of heparininduced thrombocytopenia. The periodical control of antiphospholipid antibody titres during pregnancy does not seem to be helpful, because several factors (i.e. increased plasma levels of some coagulation factors, expansion of plasma volume) greatly influence the outcome of their determination. Therefore, the decrease or disappearance of antiphospholipid antibodies during pregnancy should not modify the therapeutic policy.

THROMBOSIS DURING PREGNANCY A formal estimation of the thrombotic risk during pregnancy has not yet been given, particularly in women without a history of thromboembolic events. Pregnancy and puerperium are known to increase the risk of deep venous thrombosis. It is therefore

220 M. Galli and T. Barbui

recommended that unfractionated or low-molecular-weight heparin be given throughout pregnancy to antiphospholipid-positive women with an ascertained history of thrombosis in order to reduce the risk of recurrence. Oral anticoagulation may substitute heparin after delivery and continue during puerperium. Therapeutic dosage must be given to antiphospholipid-positive women who were already on oral anticoagulation before pregnancy. Whether heparin or aspirin prophylaxis should be administered during pregnancy in order to prevent the occurrence of the first thrombosis has not been established.

IN VITRO FERTILIZATION It is not known whether antiphospholipid antibodies are involved in the pathogenesis of infertility. However, because approximately 10 – 15% of all couples at the reproductive age suffer from infertility, regardless of their antiphospholipid status, it is likely that a certain number of antiphospholipid-positive women may be candidates for ovulation induction, with or without in vitro fertilization and embryo transfer. Owing to the high levels of ovarian oestrogen (mainly oestradiol) reached by gonadotropin stimulation, concern may be raised as to whether antiphospholipidpositive women undergoing induction of ovulation are exposed to an increased thromboembolic risk. So far, no case of thrombosis has been reported, suggesting that such a risk is minor. A question still to be answered deals with the proper management of antithrombotic treatment during one or, most commonly, several procedures of ovulation induction and ovum retrieval. In fact, heparin administered because of the patient’s thrombotic history, may increase the risk of ovarian haemorrhage. It has been suggested that heparin be discontinued 12 –24 hours before the procedure and restarted 6 – 8 hours after ovum retrieval.83 The routine evaluation of antiphospholipid antibodies in women undergoing in vitro fertilization is not warranted, as pointed out by a meta-analysis84: seven studies on more than 2000 women, 703 of whom with at least one abnormal antiphospholipid antibody test, failed to report an association between antiphospholipid antibodies and reduced success of in vitro fertilization.

HORMONE CONTRACEPTION AND REPLACEMENT THERAPY It is common knowledge that hormone contraception increases a woman’s thrombotic risk, in particular when other congenital or acquired risk factors for thrombosis are present. Data about the arterial and venous thrombotic risk of antiphospholipidpositive women taking oral contraceptives are still limited and conflicting, even though the presence of lupus anticoagulants or anticardiolipin antibodies does not seem to increase such a risk dramatically.85 The patient’s clinical history (with or without thromboembolic events) should guide the decision about oral contraceptives. Experience with hormone replacement therapy is very limited, as it has been used only in few women with systemic lupus erythematosus: apparently, no increase in lupus flares was observed.86 No information is available regarding the thrombotic risk of women with the antiphospholipid syndrome.

Antiphospholipid antibodies and pregnancy 221

Practice points † lupus anticoagulants and, possibly, anticardiolipin antibodies are risk factors of fetal death, and other pregnancy complications † the routine laboratory workout of patients with pregnancy loss should include only the measurement of lupus anticoagulants and anticardiolipin antibodies. At present, the detection of other antiphospholipid antibodies is not recommended † heparins in combination with aspirin are the only treatment of proven efficacy in patients with the antiphospholipid syndrome during pregnancy for preventing the recurrence of both obstetric complications and thrombosis † the detection of antiphospholipid antibodies before in vitro fertilization is not recommended † limited information is available in relation to the thrombotic risk of antiphospholipid-positive women taking hormone contraception or replacement therapy

Research agenda † interaction between antiphospholipid antibodies and anionic phospholipids is mediated by plasma proteins with different immunological and functional properties. These differences probably, confer different risks of complications related to the antiphospholipid syndrome. This necessitates further study of the behaviour of the various antiphospholipid antibodies to establish which of them are pathogenetically linked with recurrent abortions and fetal death † the excessive frequency of maternal and/or fetal complications, despite antithrombotic treatment, indicates the need for a better calibration of the dosage, duration and timing of administration of heparin treatment

SUMMARY Experimental work has shown that interaction between antiphospholipid antibodies and anionic phospholipids is mediated by plasma proteins and that different antibodies may be recognized on the basis of their immunological and functional properties. Also emerging is the concept that these differences are likely to confer different risks of complications related to the antiphospholipid syndrome. This necessitates further study of the behaviour of the various antiphospholipid antibodies to establish which of them are pathogenetically linked with the clinical manifestations, particularly recurrent abortions and fetal death, of the antiphospholipid syndrome. Fetal death and recurrent spontaneous abortions represent the obstetric criteria of the antiphospholipid syndrome, with an overall prevalence of 15– 20%. In particular, fetal death represent 50% of the obstetric complications. The presence of lupus anticoagulants carries a risk ranging from 3.0 to 4.8, and that of anticardiolipin antibodies goes from 0.86 to 20. The mechanism(s) by which antiphospholipid antibodies cause these events still has to be defined: thrombosis in the placental vessels, and impairment of embryonic implantation have been proposed. Unfractionated or low-molecular-weight heparin in combination

222 M. Galli and T. Barbui

with low-dose aspirin represents the current standard treatment of pregnant antiphospholipid-positive women for preventing recurrent obstetric complications. Upon treatment, the live birth rate increases from 0– 40% to 70 –80%. However, there is still an excessively high frequency of maternal and/or fetal complications, indicating the need for a better calibration of the dosage, duration and timing of administration of heparin treatment.

REFERENCES 1. Tanner D, Levine R & Kittner SJ. Epidemiology of antiphospholipid antibodies and vascular disease. In Levine SR & Brey RL (eds) Clinical Approach to Antiphospholipid Antibodies. Boston: Butterworth Heinemann, 2000, pp 18. 2. Dizon-Townson D & Branch DW. Obstetric vascular disease in antiphospholipid syndrome. In Levine SR & Brey RL (eds) Clinical Approach to Antiphospholipid, Antibodies. Boston: Butterworth Heinemann, 2000, pp 81–94. 3. Godfrey T & D’Cruz D. Antiphospholipid syndrome: general features. In Khamashta MA (ed.) Hughes’ Syndrome. London: Springer, 2000, pp 8–19. 4. Asherson RA. A ‘primary’ antiphospholipid syndrome? Journal of Rheumatology 1988; 15: 1742–1746. 5. Alarcon-Segovia D, Deleze M, Oria CV et al. Antiphospholipid antibodies and the antiphospholipid syndrome in systemic lupus erythematosus. Medicine 1989; 68: 353–365. * 6. Wilson WA, Gharavi AE, Koike T et al. International consensus on preliminary classification criteria for definite antiphospholipid syndrome. Arthritis Rheumatism 1999; 42: 1309– 1311. 7. Mackie IJ, Donohoe S & Machin SJ. Lupus anticoagulant measurement. In Khamashta MA (ed.) Hughes’ Syndrome. London: Springer, 2000, pp 214–224. 8. Loizou S, McCrea JD, Rudge AC et al. Measurement of anticardiolipin antibodies by an enzyme-linked immunosorbent assay: standardization and quantitation of results. Clinical and Experimental Immunology 1985; 62: 739–744. 9. Galli M, Comfurius P, Maassen C et al. Anticardiolipin antibodies (ACA) directed not to cardiolipin but to a plasma protein cofactor. Lancet 1990; 335: 1544–1547. 10. McNeil HP, Simpson RJ, Chesterman CN & Krilis SA. Anti-phospholipid antibodies are directed against a complex antigen that includes a lipid-binding inhibitor of coagulation: b2-glycoprotein I (apolipoprotein H). Proceedings of the National Academy of Sciences of the USA 1990; 87: 4120–4124. 11. Matsuura E, Igarashi Y, Fujimoto M et al. Anticardiolipin cofactor(s) and differential diagnosis of autoimmune disease. Lancet 1990; 335: 177 –178. 12. Galli M, Comfurius P, Barbui Tet al. Anticoagulant activity of b2-glycoprotein I is potentiated by a distinct subgroup of anticardiolipin antibodies. Thrombosis and Haemostasis 1992; 68: 297– 300. 13. Galli M, Beretta G, Daldossi M et al. Different anticoagulant and immunological properties of antiprothrombin antibodies in patients with antiphospholipid antibodies. Thrombosis and Haemostasis 1997; 77: 486–491. 14. Polz E & Kostner GM. The binding of b2-glycoprotein I to human serum lipoproteins. FEBS Letters 1979; 102: 183 –186. 15. Sheng Y, Sali A, Herzog H et al. Site-directed mutagenesis of recombinant human b2-glycoprotein I identifies a cluster of lysine residues that are critical for phospholipid binding and anticardiolipin antibody activity. Journal of Immunology 1996; 157: 3744–3751. 16. Reddel SW, Wang YX, Sheng YH & Krilis SA. Epitope studies with anti-b2-glycoprotein I antibodies from autoantibody and immunized sources. Journal of Autoimmunity 2000; 15: 91– 96. 17. Day HM, Thiagarajan P, Ahn C et al. Autoantibodies to b2-glycoprotein I in systemic lupus erythematosus and primary antiphospholipid syndrome: clinical correlations in comparison with other antiphospholipid antibody tests. Journal of Rheumatology 1998; 25: 667–674. 18. Bajaj SP, Rapaport SI, Fierer DS et al. A mechanism for the hypoprothrombinemia of the acquired hypoprothrombinemia-lupus anticoagulant syndrome. Blood 1983; 61: 684–692. 19. Arvieux J, Darnige L, Reber G et al. Development of an ELISA for autoantibodies to prothrombin showing their prevalence in patients with lupus anticoagulants. Thrombosis and Haemostasis 1995; 74: 1120–1125. 20. Rao LVM, Hoang AD & Rapaport SI. Mechanism and effects of the binding of lupus anticoagulant IgG and prothrombin to surface phospholipids. Blood 1996; 88: 4173–4182. 21. Malia RG, Brooksfield C, Bulman I, et al. Prothrombin fragment 1-2: the epitope for antiphospholipid antibody expression. Presented at the XVIth Congress of the International Society of Thrombosis and Haemostasis. Florence, Italy, June 6–12, 1997 (abst. 689).

Antiphospholipid antibodies and pregnancy 223 22. Galli M & Barbui T. Antiprothrombin antibodies: detection and clinical significance in the antiphospholipid syndrome. Blood 1999; 93: 2149–2157. 23. Willems GM, Janssen MP, Pelsers MMAL et al. Role of divalency in the high-affinity binding of anticardiolipin antibody/b2-glycoprotein I complexes to lipid membranes. Biochemistry 1996; 35: 13833–13842. 24. Horbach DA, van Oort E, Derksen RHWM et al. The contribution of anti-prothrombin antibodies to lupus anticoagulant activity: discrimination between functional and non-functional anti-prothrombin antibodies. Thrombosis and Haemostasis 1998; 79: 790–795. 25. Oosting JD, Derksen RHWM, Bobbink IWG et al. Antiphospholipid antibodies directed against a combination of phospholipids with prothrombin, protein C or protein S—an explanation for their pathogenic mechanisms. Blood 1993; 81: 2618–2625. 26. Matsuda J, Saitoh N, Gohchi K et al. Anti-annexin V antibody in systemic lupus erythematosus patients with lupus anticoagulant and/or anticardiolipin antibody. American Journal of Hematology 1994; 47: 56 –58. 27. Berard M, Sugi T, McIntyre JA et al. Prevalence and kininogen-dependence of antiphosphatidylethanolamine antibodies. Nouvelle Revue Francais d’ Hematologie 1995; 37(supplement II): S69–S72. 28. Jones DW, Mackie IJ, Gallimore MJ & Winter M. Antibodies to factor XII and recurrent fetal loss in patients with the anti-phospholipid syndrome. British Journal of Haematology 2001; 113: 550–552. 29. Cugno M, Dominguez M, Cabibbe M et al. Antibodies to tissue-type plasminogen activator in plasma from patients with primary antiphospholipid syndrome. British Journal of Haematology 2000; 108: 871–875. 30. Smirnov MD, Triplett DA, Comp PC et al. On the role of phosphatidylethanolamine in the inhibition of activated protein C activity by antiphospholipid antibodies. Journal of Clinical Investigation 1995; 95: 309–316. 31. Marciniak E & Romond EH. Impaired catalytic function of activated protein C: a new in vitro manifestation of lupus anticoagulant. Blood 1989; 74: 2426–2432. 32. Galli M, Ruggeri L & Barbui T. Differential effects of anti-b2-glycoprotein I antibodies on the anticoagulant activity of activated protein C. Blood 1998; 91: 1999–2004. 33. Salemink I, Blezer R, Willems GM et al. Antibodies to b2-glycoprotein I associated with antiphospholipid syndrome suppress the inhibitory activity of tissue factor pathway inhibitor. Thrombosis and Haemostasis 2000; 84: 653– 656. 34. Dobado-Berrios PM, Lopez-Pedrera C, Velasco E et al. Increased levels of tissue factors mRNA in mononuclear blood cells of patients with primary antiphospholipid syndrome. Thrombosis and Haemostasis 1999; 82: 1578–1582. 35. Galli M, Finazzi G, Duca F et al. The G1691A mutation of factor V, but not the G20210A mutation of factor II or the C677T mutation of methylenetetrahydrofolate reductase genes, is associated with venous thrombosis in patients with lupus anticoagulants. British Journal of Haematology 2000; 108: 865– 870. 36. Loizou S, McCrea JD, Rudge AC et al. Measurement of anticardiolipin antibodies by an enzyme-linked immunosorbent assay: standardization and quantitation of results. Clinical and Experimental Immunology 1985; 62: 739– 744. 37. Harris EN, Gharavi AE, Patel S & Hughes GRV. Evaluation of the anti-cardiolipin antibody test: report of an international workshop held April 4 1986. Clinical and Experimental Immunology 1987; 68: 215– 222. 38. Harris EN. The second international anticardiolipin standardization wet workshop/the Kingston Antiphospholipid Antibody study (KAPS) group. American Journal of Clinical Pathology 1990; 94: 476– 484. 39. Harris EN, Pierangeli S & Birch D. Anticardiolipin wet workshop report: Vth International Symposium on Antiphospholipid Antibodies. American Journal of Clinical Pathology 1994; 101: 616–624. 40. Pierangeli SS, Stewart M, Silva LK & Harris EN. Report on anticardiolipin wet workshop during the VIIth International Symposium on antiphospholipid antibodies. Journal of Rheumatology 1998; 25: 156–162. 41. Giordano P, Galli M, Del Vecchio G et al. Lupus anticoagulant, anticardiolipin antibodies and hepatitis C virus infection in thalassemia. British Journal of Haematology 1998; 102: 903–906. 42. Green D, Hougie C & Kazmier FJ. Report on the working party on acquired inhibitors of coagulation: studies on the lupus anticoagulant. Thrombosis and Haemostasis 1982; 49: 144– 147. * 43. Brandt JT, Triplett DA, Alving B & Scharrer I. Criteria for the diagnosis of lupus anticoagulant: an update. Thrombosis and Haemostasis 1995; 74: 1185– 1190. 44. Arnout J. Antiphospholipid syndrome: diagnostic aspects of lupus anticoagulants. Thrombosis and Haemostasis 2001; 86: 83–91. 45. Lockshin MD. Pregnancy loss and antiphopsholipid antibodies. Lupus 1998; 7: S86–S89. 46. Lynch A, Marlar R, Murphy J et al. Antiphospholipid antibodies in predicting adverse pregnancy outcome. A prospective study. Annals of Internal Medicine 1994; 120: 470 –475. 47. Lookwood CJ, Romero R, Feinberg RF et al. The prevalence and biologic significance of lupus anticoagulant and anticardiolipin antibodies in a general obstetric population. American Journal of Obstetrics and Gynecology 1989; 161: 369–373.

224 M. Galli and T. Barbui 48. Goldstein SR. Embryonic death in early pregnancy: a new look at the first trimester. Obstetrics and Gynecology 1994; 84: 294–297. 49. Sanson B-J, Friederich PW, Simioni P et al. The risk of abortions and stillbirth in antithrombin-, protein C-, and protein S-deficient women. Thrombosis and Haemostasis 1996; 75: 387 –388. 50. Preston FE, Rosendaal FR, Walker ID et al. Increased fetal loss in women with heritable thrombophilia. Lancet 1996; 348: 913 –916. 51. Griesshammer M, Heimpel H & Pearson T. Essential thrombocythaemia and pregnancy. Leukemia and Lymphoma 22 1996; 22(supplement 1): 57–63. 52. Branch DW & Silver RM. Criteria for antiphospholipid syndrome: early pregnancy loss, fetal loss, or recurrent pregnancy loss? Lupus 1996; 5: 409 –413. * 53. Barbui T, Cortelazzo S, Galli M et al. Antiphospholipid antibodies in early repeated abortions: a casecontrolled study. Fertility and Sterility 1988; 50: 589 –592. 54. Infante-Rivard C, David M, Gauthier R & Rivard G-E. Lupus anticoagulant, anticardiolipin antibodies, and fetal loss. New England Journal of Medicine 1991; 325: 1063–1066. 55. Ginsberg JS, Brill-Edwards P, Johnston M et al. Relationship of antiphospholipid antibodies to pregnancy loss in patients with systemic lupus erythematosus: a cross-sectional study. Blood 1992; 80: 975– 980. * 56. Gris JC, Que´re´ I, Sanmarco M et al. Antiphospholipid and antiprotein syndromes in non-thrombotic, nonautoimmune women with unexplained recurrent primary early foetal loss. The Nimes Obstetricians and Haematologists Study—NOHA. Thrombosis and Haemostasis 2000; 84: 228 –236. 57. Balash J, Reverter JC, Creus M et al. Human reproductive failure is not a clinical feature associated with b2-glycoprotein I antibodies in anticardiolipin and lupus anticoagulant seronegative patients (the antiphospholipid/cofactor syndrome). Human Reproduction 1999; 14: 1956–1959. 58. Franklin RD, Hollier N, Kutteh WH et al. b2-glycoprotein I as a marker of antiphospholipid syndrome in women with recurrent pregnancy loss. Fertility and Sterility 2000; 73: 531–535. 59. Nilsson IM, Astedt B, Hedner U & Berezin D. Intrauterine death and circulating antcoagulant (‘antithromboplastin’). Acta Medica Scandinavica 1975; 197: 153 –159. * 60. De Wolf F, Carreras LO, Moerman P et al. Decidual vasculopathy and extensive placental infarction in a patient with repeated thromboembolic accidents, recurrent fetal loss, and a lupus anticoagulant. American Journal of Obstetrics and Gynecology 1982; 142: 829 –834. 61. Out HJ, Kooijman CD, Bruinse HW & Derksen RH. Histopathological findings in placentae from patients with intra-uterine fetal death and anti-phospholipid antibodies. European Journal of Obstetrics & Gynecology and Reproductive Biology 1991; 41: 179 –186. 62. Blank M, Cohen J, Toder V & Shoenfeld Y. Induction of anti-phospholipid syndrome in naive mice with mouse lupus monoclonal and human polyclonal anti-cardiolipin antibodies. Proceedings of the National Academy of Sciences of the USA 1991; 88: 3069–3073. 63. Bakimer R, Fishman P, Blank M et al. Induction of primary antiphospholipid syndrome in mice by immunization with a human monoclonal anticardiolipin antibody (H-3). Journal of Clinical Investigation 1992; 89: 1558–1563. 64. Di Simone N, Meroni PL, Del Papa N et al. Antiphospholipid antibodies affect trophoblast gonadotropin secretion and invasiveness by binding directly and through adhered b2-glycoprotein I. Arthritis and Rheumatism 2000; 43: 140–150. * 65. Di Simone N, Caliandro D, Castellani R et al. Low-molecular weight heparin restores in-vitro trophoblast invasiveness and differentiation in presence of immunoglobulin G fractions obtained from patients with antiphospholipid syndrome. Human Reproduction 1999; 14: 489 –495. 66. Di Simone N, Caliandro D, Castellani R et al. Interleukin-3 and human trophoblast: in vitro explanations for the effect of interleukin in patients with antiphospholipid syndrome. Fertility and Sterility 2000; 73: 1194–2000. 67. Rai RS, Clifford K, Cohen H & Regan L. High prospective fetal loss rate in untreated pregnancies of women with recurrent miscarriage and antiphospholipid antibodies. Human Reproduction 1995; 10: 3301–3304. 68. Kutteh WH. Antiphospholipid antibody-associated recurrent pregnancy loss: treatment with heparin and low-dose aspirin is superior to low-dose aspirin alone. American Journal of Obstetrics and Gynecology 1996; 174: 1584–1589. 69. Rai R, Cohen H, Dave M & Regan L. Randomized controlled trial of aspirin and aspirin plus heparin in pregnant women with recurrent miscarriages associated with phospholipid antibodies (or antiphospholipid antibodies). British Medical Journal 1997; 314: 253–257. 70. Backos M, Rai R, Baxter N et al. Pregnancy complications in women with recurrent miscarriage associated with antiphospholipid antibodies treated with low dose heparin and aspirin. British Journal of Obstetrics and Gynaecology 1999; 106: 102–107. 71. Backos M, Rai R, Thomas E et al. Bone density changes in pregnant women treated with heparin: a prospective, longitudinal study. Human Reproduction 1999; 14: 2876–2880.

Antiphospholipid antibodies and pregnancy 225 72. Ruffatti A, Orsini A, Di Leonardo L et al. A prospective study of fifty-three consecutive calcium heparin treated pregnancies in patients with antiphospholipid antibody-related fetal loss. Clinical and Experimental Rheumatology 1997; 15: 499–504. 73. Ruffatti A, Dalla Barba B, Del Ross T et al. Outcome of fifty-five newborsn of antiphospholipid-positive mothers treated with calcium heparin during pregnancy. Clinical and Experimental Rheumatology 1998; 16: 605–610. 74. Lubbe WF, Butler WS, Palmer SJ & Liggins GC. Fetal survival after prednisone suppression of maternal lupus-anticoagulant. Lancet 1983; i: 1361–1363. 75. Cowchock FS, Reece EA, Balaban D et al. Repeated fetal losses associated with antiphospholipid antibodies: a collaborative randomized trial comparing prednisone with low-dose heparin treatment. American Journal of Obstetrics and Gynecology 1992; 166: 1318–1323. 76. Silver RK, McGregor SN, Sholl JS et al. Comparative trial of prednisone plus aspirin versus aspirin alone in the treatment of anticardiolipin-positive obstetric patients. American Journal of Obstetrics and Gynecology 1993; 169: 1411–1417. 77. Lockshin MD, Druzin ML & Qamar T. Prednisone does not prevent recurrent fetal death in women with antiphospholipid antibody. American Journal of Obstetrics and Gynecology 1989; 160: 439– 443. 78. Laskin CA, Bombardier C, Hannah ME et al. Prednisone and aspirin in women with autoantibodies and unexplained recurrent fetal loss. New England Journal of Medicine 1997; 337: 148 –153. 79. Empson M, Lassere M, Craig JC & Scott JR. Recurrent pregnancy loss with antiphospholipid antibody: a systematic reivew of therapeutical trials. Obstetrics and Gynecology 2002; 99: 135 –144. 80. Branch DW, Peaceman AM, Druzin M et al. A multicenter, placebo-controlled pilot study of intravenous immune globulin treatment of antiphospholipid syndrome during pregnancy. American Journal of Obstetrics and Gynecology 2000; 182: 122–127. 81. Daya S, Gunby J, Porter F et al. Critical analysis of intravenous immunoglobulin therapy for recurrent miscarriage. Human Reproduction Update 1999; 5: 475–482. * 82. Rai R. Obstetric management of antiphospholipid syndrome. Journal of Autoimmunity 2000; 15: 203– 207. 83. Udoff LC & Branch DW. Management of patients with antiphospholipid antibodies undergoing in vitro fertilization. Journal of Autoimmunity 2000; 15: 209– 211. 84. Homstein MD, Davis OK, Massey JB et al. Antiphospholipid antibodies and in vitro fertilization success: a meta-analysis. Fertility and Sterility 2000; 73: 330–333. 85. Petri M & Robinson C. Oral contraceptives and systemic lupus erythematosus. Arthritis and Rheumatism 1997; 40: 797– 803. 86. Arden NK, Lloyd ME, Spector TD & Hughes GRV. Safety of hormone replacment therapy (HRT) in systemic lupus erythematosus. Lupus 1994; 3: 11–13.