Marked variation in antiphospholipid antibodies during pregnancy: relationships to pregnancy outcome

J. Topping1,5, S. Quenby2, R. Farquharson1, R. Malia3 and M. Greaves4

1 Liverpool Women's Hospital, Crown Street, Liverpool L8 7SS, 2 Department of Obstetrics and Gynaecology, University of Liverpool, Liverpool, UK, 3 Department of Haematology, University of Sheffield, Sheffield, UK and 4 Department of Medicine and Therapeutics, University of Aberdeen, Aberdeen, UK


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Variations in blood concentrations of antiphospholipid antibodies (APA) were investigated through the course of pregnancy in women who had a history of recurrent pregnancy loss, and were related to changes in pregnancy outcome. Serial measurements of APA were made in 16 women with antiphospholipid syndrome (APS) and 16 with negative APA tests pre-pregnancy. There was considerable intraindividual variation in test results through pregnancy. There was a significantly higher ratio of dilute Russell's viper venom time and IgG ACA titre in the first trimester compared with results pre-pregnancy in women with APS. Furthermore, transiently positive APA results were noted in the control group during pregnancy and some women with antiphospholipid syndrome tested negative for APA in mid- and late pregnancy. We have demonstrated clinically important variations in the results of tests for APA during pregnancy in women with APS.

Key words: antibodies/antiphospholipid/pregnancy


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Positive results in tests for antiphospholipid antibodies (APA) are associated with adverse pregnancy outcome, especially recurrent pregnancy loss and intrauterine growth retardation, as well as with maternal complications including thrombosis, thrombocytopenia and pregnancy induced hypertension (Lockshin et al., 1987Go; Triplett and Harris, 1989Go; Branch et al., 1992Go; Kwak et al., 1992Go; Balasch et al., 1996Go). Despite this, some women with positive tests for APA have uncomplicated pregnancies without treatment (Lockwood et al., 1989Go; Harris and Spinnato, 1991Go). Antiphospholipid syndrome (APS) has been defined as a condition where a history of recurrent pregnancy loss and/or thrombosis is accompanied by persistently positive tests for APA (usually a raised titre of antibody to cardiolipin and/or abnormalities of coagulation based assays consistent with the presence of a lupus anticoagulant) (Hughes and Khamashata, 1994). Because an immune mediated thrombotic pathogenesis is suspected, antithrombotics and corticosteroids have been used in an attempt to improve pregnancy outcome in women with APS.

There are diagnostic difficulties in APS which relate principally to the indirect nature of the tests used and problems in their performance and interpretation (the Lupus Anticoagulant Working Party, 1990Go; Jennings et al., 1998Go). In relation to anticardiolipin, there is controversy over the clinical significance of low titre positivity, and it has been suggested that intervention using potentially toxic drugs such as heparin and corticosteroids should be restricted to those women with strongly positive tests (Kutteh, 1996Go; Silver et al., 1996Go). In contrast, women with low titre antibodies have apparently benefited from treatment with heparin (Rai et al., 1997Go). There are also problems associated with the assays for lupus anticoagulant (LA), as they are essentially non-quantitative. Furthermore, the physiological haemostatic changes which occur during healthy pregnancy result in a shortening of clotting times (Wright et al., 1988Go). The complexity of the situation is compounded by the large intra-patient variation in test results over time, one-third to one-half of APA-positive subjects testing negative when reassessed (Creagh et al., 1991aGo; Rai et al., 1995Go). Because of these observations, it has been recommended that APS be diagnosed only when tests for APA are positive on two occasions at least 8 weeks apart in a subject with a consistent clinical history. Whether the recent introduction of tests for antibodies to the phospholipid binding proteins which carry the target antigens for APA will improve specificity, remains to be determined.

Little is known of variations in APA during pregnancy. In our recurrent miscarriage cases, ~20% of women have persistent APA pre-conceptually (Quenby and Farquharson, 1993Go), when we employ comprehensive laboratory testing. In some cases we have noted a rise in the anticardiolipin titre in early pregnancy. In view of this, and the recent potentially important suggestion that rising titres of anticardiolipin herald pregnancy loss (Kwak et al., 1994Go), we have monitored APA serially through pregnancy in women with APS and in a control group of women with recurrent miscarriage but negative results of tests for APA pre-conceptually.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Consecutive non-pregnant women (n = 145) with a history of at least two consecutive miscarriages were tested for APA. Of these, 30 (21%) had a raised titre of anticardiolipin and/or lupus anticoagulant detected on two occasions at least 8 weeks apart. Other causes of miscarriage were excluded: parental balanced translocations, endocrine causes identified by the presence of oligomenorrhoea (Quenby and Farquharson, 1993Go), hypothyroidism, uterine abnormalities as defined by hysteroscopy in women with previous second trimester miscarriage and cervical incompetence. Sixteen of the 30 gave informed consent for participation in the study. Sixteen control women from the same cohort, who had negative tests for APA and who were matched for age and number of miscarriages with the women with APS, also consented to the study (Table IGo). Women with APS received low-molecular weight heparin (Fragmin, Kabi) 5000 units daily by s.c. injection and aspirin 75 mg daily by mouth from the time the pregnancy was confirmed. All women received reassurance ultrasonography at fortnightly intervals until 14 weeks gestation. The cases (APS women) and controls (APA-negative recurrent miscarriage women) were comparable in terms of age and numbers of miscarriages (Table IGo).


View this table:
[in this window]
[in a new window]
 
Table I. Characteristics of the patient groups
 
For each patient, 20 ml of venous blood was sampled at intervals from 6–7 weeks gestation until the end of the pregnancy. When testing for lupus anticoagulant, 10 ml of blood was placed in citrate containing tubes, and immediately prepared by double centrifugation, at 2700 g for 10 min, in accordance with published guidelines (Machin et al., 1991Go). The supernatant (citrated plasma) was then stored at –70°C, transported in dry ice and tested within 6 weeks of sampling. Blood samples (10 ml) were prepared for measurement of anticardiolipin titre by single centrifugation, at 2700 g for 10 min, removal of the supernatant (serum), which was stored and transported under the same conditions.

IgG and IgM anticardiolipin were quantified in serum using a previously validated ELISA calibrated against an international standard (Hill et al., 1995Go). The upper limit of the normal range was determined from the log transformed mean plus 2 SD of results in 50 healthy adults [9 IgG phospholipid units (GPU)/ml IgG and 4 IgM phospholipid units (MPU)/ml IgM anticardiolipin]. Screening for lupus anticoagulant was by the Kaolin Cephalin Clotting Time (KCCT) utilizing sensitive reagents and by the Dilute Russell's Viper Venom Time (DRVVT) with a neutralization procedure using frozen–thawed platelets. Based on results in 50 healthy non-pregnant subjects a positive result was considered to be a DRVVT ratio to normal of >=1.1 with >=20% correction in the platelet neutralization step.

The data were analysed by trimester, where the first trimester was defined as 0–12 weeks, second trimester 12–24 weeks and third trimester after 24 weeks. The mean of the two results for each patient in each trimester was used for analysis. Statistical analysis was by the Wilcoxon signed ranks tests and was performed using the Arcus software package for personal computers.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Of the 16 women with APS, six had both lupus anticoagulant and a raised titre of anticardiolipin (IgG in six and IgM in two), four lupus anticoagulant only, and six raised anticardiolipin titre only (IgG in six and IgM in three). All 10 subjects with lupus anticoagulant had a positive DRVVT test.

There was considerable intra-individual variation in test results throughout pregnancy. Amongst the controls none was positive for APA on pre-pregnancy testing, by definition. However, five control women had positive tests in the first trimester (three were positive for lupus anticoagulant and two had a raised titre of IgG anticardiolipin) (Table IIGo). These increases in assay results for women negative for APS pre-pregnancy, were surprisingly large (for the DRVVT ratio: from 0.9, 1.0 and 1.09 to 1.36, 1.15, and 1.19 respectively, and for the IgG ACA: from 2 and 4.4 to 41.7 and 15.8 respectively). Among the women in the APS group, highest ACA and DRVVT results were also found in the first trimester (Figures 1 and 2GoGo). The highest titres of IgG anticardiolipin were seen in the first trimester in 24 out of the total 32 patients studied longitudinally. The highest ratios in the DRVVT were also seen at this time, in 23 out of 32 patients. In the entire study cohort of 32 women a statistically significant difference in both the DRVVT and IgG anticardiolipin titres was demonstrated between pre-pregnancy and the first trimester samples (Tables III and IVGoGo). In the control group, the change did not persist later in pregnancy; APA positivity was noted in the second and third trimesters in one subject only (Table IIGo). The APS group had a progressive decline in the level of positivity as pregnancy progressed (Figures 1 and 2GoGo).


View this table:
[in this window]
[in a new window]
 
Table II. APS assay positivity in the longitudinal study, for those women with APS pre-conceptually and controls (APS negative pre-conceptually)
 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 1. Variation in the dilute Russell's viper venom test (DRVVT) ratio for the antiphospholipid syndrome (APS) group during pregnancy; mean ± 95% confidence intervals.

 


View larger version (13K):
[in this window]
[in a new window]
 
Figure 2. Variation in the IgG anticardiolipin antibodies in APS group during pregnancy; mean ± 95% confidence intervals.

 

View this table:
[in this window]
[in a new window]
 
Table III. The statistical difference in the dilute Russell's viper venom test assay between pre-pregnancy results and pregnancy value by trimester, using the Wilcoxon signed ranks test results (n = 32)
 

View this table:
[in this window]
[in a new window]
 
Table IV. The statistical difference in the IgG anticardiolipin antibodies assay between pre-pregnancy results and pregnancy value by trimester, using the Wilcoxon signed ranks test results (n = 32)
 
Among the 16 APS pregnancies there were 12 full-term live births (none had growth restriction or pregnancy induced hypertension) and four miscarriages. In the four women who miscarried, APA remained detectable throughout the pregnancy, with no apparent fall in the degree of positivity. Of the 12 women with APS and uncomplicated pregnancies, the APA tests had become negative in eight women by the second trimester, and by the third trimester nine APS women had test results within the normal range (Table IIGo). All 16 control women had uncomplicated live births, reflecting the high success rate in women with recurrent pregnancy loss of unknown aetiology (Clifford et al., 1997Go).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
In this study, the proportion of women with recurrent miscarriage found to have persistent APA is comparable to that reported by others (Parke et al., 1991Go). These longitudinal data suggest that the results of tests for APA are influenced by gestation in women with APS. Both the DRVVT ratio and IgG anticardiolipin titre have higher values during the first trimester. The pathophysiological basis of these changes is obscure. As part of the physiological preparation for the challenge of parturition there is an up-regulation of the coagulation and haemostatic mechanisms, with a shortening of clotting times. Factor VIII and fibrinogen are particularly affected by stage of gestation, but a change in their plasma concentration should not influence the DRVVT assay (Derksen et al., 1992Go). In addition, the main rise occurs later in pregnancy and these adaptive responses are not apparent during the first trimester (Wright et al., 1988Go). We have previously found no difference in the normal range for the DRVVT in 30 normal pregnant women in the second trimester (14 and 18 weeks gestation) compared with that measured when they were not pregnant (Creagh et al., 1991bGo) in contrast to the current results, in women with APS. It is important to note, however, that all tests for lupus anticoagulant are indirect and influenced by changes in coagulation factor concentration and are not truly quantitative. This may account for the observation that it has not been possible to demonstrate a relationship between the degree of prolongation of clotting times due to lupus anticoagulant and the likelihood of miscarriage or thrombosis, although a relationship between the titre of anticardiolipin antibodies and chance of miscarriage or thrombosis has been found to exist (Silver et al., 1996Go).

In support of the apparent change in clotting times however, the IgG anticardiolipin titre was also highest in the first trimester in women with APS, with a subsequent fall in some women, often into the normal range. This apparent increase in anticardiolipin titre in the first trimester is of interest. The serum concentration of total IgG falls during pregnancy (El-Roiey et al., 1990Go). Although levels of autoantibodies appear to increase in healthy pregnancy they remain within the range of normal, including anticardiolipin (Patton et al., 1987Go). It is therefore possible to speculate that maternal exposure to antigenic stimuli may result in an increase in the autoimmune response in early pregnancy in women with APS. Alternatively, deficiency in the cellular adhesion mechanism of the maternal–conceptus interface caused by antiphospholipid antibodies, may lead to increased exposure of fetal cells to the maternal circulation (Lim et al., 1996Go). Further evidence for such autoimmune aetiologies of recurrent pregnancy loss has been discussed in detail by Christiansen (1996).

It is unlikely that the treatment given was responsible for the changes in APA noted through pregnancy. Aspirin is not known to have any immunomodulatory effect and does not influence the fluid phase of coagulation. Heparin can result in prolongation of the KCCT and DRVVT, but does not do so when administered in prophylactic doses. Also, low molecular weight heparin has limited antithrombin activity and does not influence the clotting time in the doses used. Heparin has been shown to reduce APA binding to cardiolipin in vitro (Ermel et al., 1995Go). However, in this study, the low molecular weight heparin treatment was initiated prior to the first trimester test which was generally higher than the pre-pregnancy test (Figures 1 and 2GoGo), and therefore the change in DRVVT was independent of the administered low molecular weight heparin.

It is noteworthy that a decrease in the anticardiolipin titre was a feature in all of those pregnancies with successful outcomes including the control patients, who although tested negative preconceptually had the highest levels in the first trimester that decreased progressively until delivery. Because we had no late pregnancy complications, we could not demonstrate whether a persistently high level would have predicted an adverse event, e.g. pre-eclampsia, abruption or sudden intrauterine death. Our findings are consistent with the observations of Kwak et al. (1994), who suggested that a down-regulation of maternal antiphospholipid antibodies is a good prognostic feature, although no attempt was made to assay for lupus anticoagulant in pregnancy in that study. Some data from studies in experimental animals suggest a possible causal role for anticardiolipin in pregnancy loss (Branch et al., 1990Go), although this is disputed and no such evidence exists in humans. Despite this, serial measurement of anticardiolipin titre in pregnancy may serve as a surrogate marker for the pathogenic process which leads to miscarriage and permit better discrimination of those at particular risk. As evidence on the efficacy of various therapeutic regimens accrues (Rai et al., 1997Go) such a facility could be of benefit, especially as treatment with heparin is not without risk.

The findings in the control group were unexpected, but consistent with the known occurrence of transient antiphospholipid antibodies. These have frequently been associated with intercurrent and chronic infections (Al-Saeed et al., 1994Go), although there is no reason to suspect an infectious pathogenesis in these women with recurrent miscarriage. Persistently positive tests for APA were found in ~3% of 501 unselected women in the first trimester of apparently healthy pregnancies and a similar proportion had transiently positive tests (Creagh et al., 1991aGo). There is, therefore, a low prevalence of often transient APA positivity in apparently healthy pregnancies and the data presented here indicates that this may be increased in pregnancies in women with a history of recurrent miscarriage. The transience of the test abnormality and the uniformly good outcome of the pregnancies distinguishes this situation from that in women with APS, where persistence of APA is found pre-conceptually and maintained through pregnancy, especially in those who miscarry. The occurrence of transient APA in the control group may reflect the rise in titres seen in the APS group. It is unknown whether the development of APA during pregnancy has pathological significance.

The results of this study emphasize the need for a comprehensive clinical and laboratory approach to the diagnosis of APS. Future studies, in larger cohorts of women, should further address the prognostic value of the changes in APA reactivity which occur through pregnancy in women with APS; ideally, these studies should include women on no treatment.


    Notes
 
5 To whom correspondence should be addressed at: Arrowe Park Hospital, Arrowe Park Road, Upton, Wirral L49 5PE, UK Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Al-Saeed, A., Makris, M., Malia, R.G. et al. (1994) The development of antiphospholipid antibodies in haemophilia is linked to infection with hepatitis C. Brit. J. Haematol., 88, 845–848.[ISI][Medline]

Balasch, J., Creus, M., Fabregues, F. et al. (1996) Antiphospholipid antibodies and human reproductive failure. Hum. Reprod., 11, 2310–2315.[Abstract]

Branch D.W., Dudley D.J., Mitchell M.D. et al. (1990) Immunoglobulin G fractions from patients with antiphospholipid antibodies cause fetal death in BALB/c mice: a model for autoimmune loss. Am. J. Obstet. Gynaecol., 163, 210–216.[ISI][Medline]

Branch, D.W., Silver, R.M., Blackwell, J.L. et al. (1992) Outcome of treated pregnancies in women with antiphospholipid syndrome: an update of the Utah experience. Obstet. Gynecol., 80, 614–620.[Abstract]

Christiansen, O.B. (1996) A fresh look at the causes and treatments of recurrent miscarriage, especially its immunological aspects. Hum. Reprod. Update, 2, 271–279.[Abstract]

Clifford, K., Rai, R. and Regan, L. (1997) Future pregnancy outcome in unexplained recurrent first trimester miscarriage. Hum. Reprod., 12, 387–389.[Abstract]

Creagh, M.D., Duncan, S.L.B., McDonnell, J.M. et al. (1991a) Failure of the detection of antiphospholipid antibodies alone to predict poor pregnancy outcome. Br. J. Haematol., 77, 4–7.

Creagh, M.D., Malia, R.G., Cooper, S.M. et al. (1991b) Screening for the lupus anticoagulant and anticardiolipin antibodies in women with fetal loss. J. Clin. Pathol., 44, 45–47.[Abstract]

Derksen, R.H., Out, H.J., Blokzijl, L. and De-Groot, P.G. (1992) Detection of lupus anticoagulant in pregnancy (letter). Clin. Exp. Rheumatol., 10, 323–324.[ISI][Medline]

El-Roiey, A., Myers, S.A. and Gleicher, N. (1990) The prevalence of autoantibodies and lupus anticoagulant in healthy pregnant women. Obstet. Gynecol., 75, 390–396.[Abstract]

Ermel, L.D., Marshburn, P.B. and Kutteh, W.H. (1995) Interaction of heparin with antiphospholipid antibodies from the sera of women with recurrent pregnancy loss. Am. J. Reprod. Immunol., 33, 14–20.[ISI][Medline]

Harris, E.N. and Spinnato, J.A. (1991) Should anticardiolipin tests be performed in otherwise healthy pregnant women? Am. J. Obstet. Gynecol., 165, 1272–1277.[ISI][Medline]

Hill, M.B., Phipps, J.L., Malia, R.G. et al. (1995) Characterisation and specificity of anti-endothelial cell membrane antibodies and their relationship to thrombosis in primary antiphospholipid syndrome (APS). Clin. Exp. Immunol., 102, 368–372.[ISI][Medline]

Hughes, G.R.V. and Khamashta, M.A. (1994) The antiphospholipid syndrome. J. R.. Coll. Phys. Lond., 28, 301–304.[ISI][Medline]

Jennings, I., Woods, T.A.L., Kitchen, S. et al. (1998) Clinically important inaccuracies in testing for the lupus anticoagulant: an analysis of the results from the surveys of the UK national external quality assurance scheme (NEQAS) in blood coagulation. Thromb. Haemostas., 77, 934–937.[ISI]

Kutteh, W.H. (1996) Antiphospholipid antibody-associated recurrent pregnancy loss: treatment with heparin and low-dose aspirin is superior to low-dose aspirin alone. Am. J. Obstet. Gynecol., 174, 1584–1589.[ISI][Medline]

Kwak, J.Y.H., Gilman Sachs, A., Beaman, K.D. et al. (1992) Autoantibodies in women with primary recurrent spontaneous abortion of unknown etiology. J. Reprod. Immunol., 22, 15–31.[ISI][Medline]

Kwak, J.Y.H., Barini, R., Gilman Sachs, A. et al. (1994) Down-regulation of maternal antiphospholipid antibodies during early pregnancy and pregnancy outcome. Am. J. Obstet. Gynecol., 171, 239–246.[ISI][Medline]

Lim, K.J.H., Odukoya, O.A., Li, T.C. and Cooke, I.D. (1996) Cytokines and immuno-endocrine factors in recurrent miscarriage. Hum. Reprod., Update, 2, 469–481[Abstract/Free Full Text]

Lockshin, M.D., Druzin, M.L. and Qamar, T. (1987) Antibody to cardiolipin, lupus anticoagulant and fetal death. J. Rheumatol., 14, 259–262.[ISI][Medline]

Lockwood, C.J., Romero, R., Feinberg, R.F. et al. (1989) The prevalence and biologic significance of lupus anticoagulant and anticardiolipin antibodies in a general obstetric population. Am. J. Obstet. Gynecol., 161, 369–373.[ISI][Medline]

Machin, S.J., Giddings, J.C., Greaves, M. et al. (1991) Guidelines on testing for the lupus anticoagulant. Prepared by the Working Party of the BCSH Blood Transfusion Task Force. J. Clin. Pathol., 44, 885–889.[ISI][Medline]

Parke, A.L., Wilson, D., Bayer-Maierd (1991) The prevalence of antiphospholipid antibodies in women with recurrent spontaneous abortion, women with successful pregnancies and women who have never been pregnant. Arthr. Rheum., 34, 1231–1235.[ISI][Medline]

Patton, P.E., Coulam, C.B. and Bergstralh, E. (1987) The prevalence of autoantibodies in pregnant and non-pregnant women. Am. J. Obstet. Gynecol., 157, 1345–1350.[ISI][Medline]

Quenby, S.M. and Farquharson, R.G. (1993) Predicting recurring miscarriage: what is important? Obstet. Gynecol., 82, 132–138.[Abstract]

Rai, R.S., Regan, L., Clifford, K. et al. (1995) Antiphospholipid antibodies and beta–2-glycoprotein-I in 500 women with recurrent miscarriage: results of a comprehensive screening approach. Hum. Reprod., 10, 2001–2005.[Abstract]

Rai, R., Cohen, H., Dave, M. et al. (1997) Randomised controlled trial of aspirin and aspirin plus heparin in pregnant women with recurrent miscarriage associated with phospholipid antibodies. Br. Med. J., 314, 253–257.[Abstract/Free Full Text]

Silver, R.M., Porter, T.F., Van Leeuween, I. et al. (1996) Anticardiolipin antibodies: clinical consequences of `low titers'. Obstet. Gynecol., 87, 494–500.[Abstract/Free Full Text]

The Lupus Anticoagulant Working Party (1990) Detection of the lupus-like anticoagulant: current laboratory practice in the UK. J. Clin. Pathol., 43, 73–75.[Abstract]

Triplett, D.A. and Harris, E.N. (1989) Antiphospholipid antibodies and reproduction. Am. J. Reprod. Immunol., 21, 3–4.

Wright, J.G., Cooper, P., Astedt, B. et al. (1988) Fibrinolysis during normal human pregnancy; complex inter-relationships between plasma levels of tissue plasminogen activator and inhibitors on the euglobulin clot lysis time. Br. J. Haematol., 69, 253–258.[ISI][Medline]

Submitted on April 23, 1998; accepted on October 20, 1998.