| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Department of Obstetrics and Gynecology I and II; Department of Obstetrics and Department of Clinical Immunology, Hôpitaux Universitaires de Strasbourg; Department of Epidemiology, Faculty of Medicine, Strasbourg; Department of Obstetrics and Gynecology, Pasteur Hospital, Colmar; Department of Obstetrics and Gynecology I and II, Hasenrein Hospital, Mulhouse; Department of Obstetrics and Gynecology, Civil Hospital, Haguenau; and the Department of Blood Transfusion, Coagulation Unit, Strasbourg, France.
Address reprint requests to: Michel Dreyfus, MD Department of Obstetrics and Gynecology Hôpitaux Universitaires de Caen Avenue Clemenceau 14033 Caen, Cedex France E-mail: dreyfus-m{at}chu-caen.fr
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: We conducted a prospective case-control study of 180 pregnant women with their first incidents of preeclampsia and no histories of thrombosis or systemic autoimmune diseases. Preeclampsia (n = 180) was defined as blood pressure (BP) at least 140/90 mmHg after 20 weeks’ gestation and proteinuria at least 0.3 g per 24 hours. Two control subjects were matched to each case (n = 360). They were pregnant women without hypertension or proteinuria and without histories of thrombosis or systemic autoimmune disease. Lupus anticoagulant (activated partial thromboplastin time, diluted thromboplastin time, platelet neutralization procedure) and anticardiolipin antibodies (immunoenzymatic assays) were assessed in both groups, and the coagulation state (levels of thrombin-antithrombin III complexes, fragments 1 + 2 of prothrombin) was also evaluated. The analysis design was a sequential plan with 5% type I error and 95% power.
Results: There was no association between antiphospholipid antibodies and preeclampsia. The odds ratio for the association was 0.95 (95% confidence interval 0.45, 2.61). Antiphospholipid antibodies were detected in eight of 180 preeclamptic women and in 19 of 360 controls. In contrast, there was a clear, confirmed activation of coagulation during preeclampsia.
Conclusion: Despite evidence of a prothrombotic state during preeclampsia, it is unlikely that antiphospholipid antibodies (lupus anticoagulant and anticardiolipin antibodies) represent risk factors for preeclampsia among women with no previous preeclampsia and no histories of thrombosis or systemic autoimmune disease
Antiphospholipid antibodies (lupus anticoagulant and anticardiolipin antibodies) are heterogeneous antibodies directed against proteins that bind anionic phospholipids. Although they can be found in normal individuals, a large body of evidence established clinical associations between antiphospholipid antibodies and venous and arterial thrombosis, thrombocytopenia, recurrent fetal loss, and many other conditions, with or without systemic lupus erythematosus.1–4 During pregnancy, thrombosis and placental infarction have been implicated in some complications, such as recurrent fetal wastage, idiopathic fetal growth restriction (FGR), and preeclampsia. Considering the thrombotic predisposition that is apparently linked to antiphospholipid antibodies in women,5,6 it was important to test for those antibodies during complicated pregnancies, in particular during preeclampsia, to determine possible predictive or therapeutic effects. The search for such an association between antiphospholipid antibodies and preeclampsia has resulted in conflicting results, probably because of case series with small samples or heterogeneity of selected women.7–14 Despite contrasting results, some authors suggested thromboprophylaxis during antiphospholipid antibody–associated pregnancies.15
The main goal of the present study was to determine whether women with their first occurrence of preeclampsia and no histories of thrombosis or autoimmune disease were more likely to have lupus anticoagulant or anticardiolipin antibodies than women who had normal courses during the same period of pregnancy. The second objective was to look for unusual subclinical hemostasis activation in positive antiphospholipid antibodies of preeclamptic women and controls.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Two control subjects were matched prospectively to each case (n = 360). Women who reported histories of preeclampsia were not eligible to be controls. Controls were matched to the cases’ terms within 2 weeks, to the cases’ age categorized as under 20 years, 20–35 years, and over 35 years, and to the cases’ parity (nulliparas versus multiparas). Controls were recruited among pregnant women who attended the same participating obstetric departments for routine examinations and blood analysis. Blood was collected only once at inclusion, depending on matching with cases, for testing for antiphospholipid antibodies and prethrombotic markers, and usual pregnancy follow-up. Eight eligible controls declined to participate
Blood (4.5 mL) was collected in 0.109 M (ie, 3.2%) trisodium citrate anticoagulant (0.5 mL). Plasma was collected after 5 minute centrifugation at 10,000 g. The supernatant was drawn off and tested immediately for activated partial thromboplastin time and lupus anticoagulant detection, or frozen at -78C for later analysis. Serum was collected after 4 hours clotting at 37C and 10-minute centrifugation at 3000 g. Supernatant was frozen at -25C until analysis. Four biologic tests were done for subjects and controls. Activated partial thromboplastin time reagent used was PTT-A (Diagnostica Stago, Asnières, France) according to the directives of the manufacturer. Normal values of test plasma to normal plasma ratio were 0.7 to 1.2. Lupus anticoagulant was detected according to methods of Brand et al16; screening test was diluted thromboplastin time, a prothrombin time realized with high dilutions of thromboplastin in calcium chloride 0.25M. The reagent used was Thromboplastin IS (Dade Behring, Rueil Malmaison, France). Two dilutions of thromboplastin were used for each plasma sample. Lupus anticoagulant was considered present when the patient plasma:control plasma ratio was higher than 1.3 with 1/50 dilution and 1.5 with 1/500 dilution. In the presence of prolonged activated partial thromboplastin time, mixing studies were done with a subject:control plasma ratio of 1:1. Correction was defined as an activated partial thromboplastin time of the mixture less than or equal to the activated partial thromboplastin time of the normal plasma plus 5 seconds, as described.16 The confirmation test was a platelet neutralization procedure. The reagent used was Staclot PNP (Diagnostica Stago). The platelet neutralization procedure depended on shortening of the activated partial thromboplastin time after patients’ plasma was mixed with aliquots of platelet membranes. A shortening of 8 seconds was considered significant according to the manufacturer’s instructions. Anticardiolipin antibodies were detected with an enzyme-linked immunosorbent assay (ELISA) according to Harris.17 The second antibodies coupled with peroxidase were anti-immunoglobulin (Ig)G (Nordic Immunological Laboratories, Tilburg, The Netherlands), anti-IgA (Sigma Aldrich Corporation, St Louis, MO), and anti-IgM (Dako A/S, Glostrup, Denmark). Fragments 1 + 2 of prothrombin (nM/L) and thrombin-antithrombin III complexes (µg/L) were determined with a commercial ELISA kit (Enzygnost fragments 1 + 2 of prothrombin micro and Enzygnost TAT micro) (Dade Behring) according to the directives of the manufacturer. Normal values were established in 30 normal women from the third to the ninth months of pregnancy.
A comparison of the means of thrombin-antithrombin III complexes and fragments 1 + 2 of prothrombin, according to antiphospholipid antibodies status and status as a case or control, was made using Brown-Forsythe analysis of variance18 because the variances were not equal (Levene tests19 were significant). The risk of preeclampsia associated with the presence of antiphospholipid antibodies was assessed using a sequential matched case-control study based on a conditional logistic regression.20 This type of analysis is represented in Figure 1
by the triangle. The null hypothesis is accepted when the path crosses the lower side of the triangle. The alternative hypothesis is accepted when the path crosses the upper side. As long as the path stays within the triangle, the recruitment continues. The sequential plan was a one-sided triangular test21 with a 5% type I error and a 95% power to detect a tripling of the risk in order to maintain clinical relevance of our results. An analysis was made every 25 triplets. The other possibility would have been a traditional fixed sample analysis with a single test at the end of the study. The gain in time for the same type I and II error is important.
|
The combined effects of antiphospholipid antibodies, thrombin-antithrombin III complexes, and fragments 1 + 2 of prothrombin were estimated by a conditional logistic regression. The following three approaches were investigated: all three variables were dichotomized into positive and negative, thrombin-antithrombin III complexes and fragments 1 + 2 of prothrombin are used as continuous variables, and the logarithm of thrombin-antithrombin III complexes and fragments 1 + 2 of prothrombin were used. The model presented is the one that gave the best fit (logarithm of thrombin-antithrombin III complexes and fragments 1 + 2 of prothrombin). The subgroups of severe preeclampsia and early onset (under 34 weeks’ gestation) were analyzed separately using the same strategy as the one used for the whole sample.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
. Among the cases, one had lupus anticoagulant and seven had anticardiolipin antibodies. Among the matched controls, five had lupus anticoagulant and 14 had anticardiolipin antibodies. The OR for the association between antiphospholipid antibodies (lupus anticoagulant and anticardiolipin antibodies) and preeclampsia was 0.95 (95% CI 0.45, 2.61; P = .50). The OR for the association between anticardiolipin antibodies (IgG, IgM, or IgA) and preeclampsia was 0.96 (95% CI 0.35, 2.38; P = .58). The path of the sequential analysis concerning detection of antiphospholipid antibodies (Figure 1
) crossed the lower boundary, and additional information from the triplets that arrived after the last analysis strengthened the conclusion. Considering the immunoglobulin class or the levels of anticardiolipin antibodies (GPL or MPL units), we did not find a statistical difference between cases and controls. For instance, the levels of the IgG anticardiolipin antibodies were 22, 32, 34, and 77 GPL units in the cases and 12, 15, 17, 20, 27, and 58 GPL units in the controls (stratified Mann-Whitney test P = .09). Only one pregnant woman (a control) had both lupus anticoagulant and anticardiolipin antibodies (IgG class).
|
confirm that thrombin-antithrombin III complexes and fragments 1 + 2 of prothrombin were higher in preeclamptic women compared with controls. The results of the multivariate analysis from logarithmic expression of the data are given in Table 2. However, when we compared women with and without detectable antiphospholipid antibodies (Table 3), we did not find any statistically significant difference in the levels of the markers of activated coagulation in cases or controls. Table 3 indicates the levels of thrombin-antithrombin III complexes and fragments 1 + 2 of prothrombin in cases and controls that had positive antiphospholipid antibodies and in cases and controls with negative antiphospholipid antibodies. By analysis of variance those differences were not significant. Levels of fragments 1 + 2 of prothrombin were not different in pregnant women with positive versus negative antiphospholipid antibodies (P = .87). The results for levels of thrombin-antithrombin III complexes for the same comparison were not different (P = .51). These results indicate that antiphospholipid antibodies are unlikely to induce increased coagulation activation markers in pregnant women.
|
|
shows OR, CI, and P by multivariate analysis for antiphospholipid antibodies, thrombin-antithrombin III complexes, and fragments 1 + 2 of prothrombin. Only thrombin-antithrombin III complexes values were statistically different between severely preeclamptic women and their matched controls. We also separately analyzed women who had early-onset (before 34 weeks’ gestation) preeclampsia (n = 79), and we did not find a difference in the antiphospholipid antibody status with controls (n = 158). Four women with early-onset preeclampsia and seven controls were positive for antiphospholipid antibodies (OR 1.33, CI 0.38, 4.72; P = .66). By analysis of variance, those differences were not significant. Levels of fragments 1 + 2 of prothrombin were not different in pregnant women with positive versus negative antiphospholipid antibodies (P = .11). The results for levels of thrombin-antithrombin III complexes for the same comparison were not different (P = .30).
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Our study was intended to include consecutive cases of preeclamptic women in the participating centers who met inclusia criteria. We do not think that the 16 preeclamptic women who did not participate in the study would have changed the conclusions because they were excluded for different reasons. We decided to test directly lupus anticoagulant and anticardiolipin antibodies instead of using a stepwise screening of antiphospholipid antibodies, as proposed for patients with fetal losses,25 because those antibodies could represent distinct members of antiphospholipid antibodies that only partially overlap.26 The frequency of antiphospholipid antibodies detected in our population of pregnant women (5%) was not different from those of previous series ranging from 1.8% to 7%.27 We estimated that a traditional analysis would have lasted twice as long (9 years).
By the increased levels of thrombin-antithrombin III complexes and fragments 1 + 2 of prothrombin, we confirmed that preeclampsia is probably associated with hemostasis activation in vivo although the latter marker was not significantly increased in the severe preeclampsia subgroup. The conversion of prothrombin to thrombin by the prothrombinase results in the generation of peptides, active thrombin, and fragments 1 + 2 of prothrombin. When reacting with its inhibitor, antithrombin III, thrombin generates the thrombin-antithrombin III complex. Thus, those two markers are generally, but not definitively, considered to quantify the activation state of the coagulation system.28,29 During preeclampsia, increased hemostatic activation is probably linked to the known fibrin deposition in glomerular capillary walls and placental spiral arteries. However, although antiphospholipid antibodies were associated with a prothrombotic state in lupus patients,5 the antiphospholipid antibodies detected in preeclamptic women and in controls was not associated with increased thrombin-antithrombin III complexes and fragments 1 + 2 of prothrombin. One possible drawback of this study could have been low statistical power of the tests in Table 3 because of the small number of patients with positive antiphospholipid antibodies (eight cases and 18 controls). We calculated a posteriori the power of those tests, based on the observed standard errors, to detect a difference of the same range to the one we observed between cases and controls (two for fragments 1 + 2 of prothrombin and 20 for thrombin-antithrombin III complexes). The numbers required in each group (positive and negative antiphospholipid antibodies) to reach a power of 80% for a 5% type I error are 12 for fragments 1 + 2 of prothrombin and 28 for the thrombin-antithrombin III complexes. These calculations assume that their repartition is similar in the two groups. The fact that the number of negative antiphospholipid antibodies (514 women) was far greater ensures that the power is over 80% and even over 90% for fragments 1 + 2 of prothrombin. In the subgroup of early-onset preeclampsia (before 34 weeks’ gestation), the power was 65%.
It is likely that those antibodies were not able to activate the coagulation cascade that already had been activated by another mechanism during preeclampsia. Therefore, we consider the antiphospholipid antibodies detected in our selected population to represent natural nonpathogenic autoantibodies. Although this type of antibody is mainly low affinity IgM, they can switch their heavy chain to IgG or IgA class. Currently, we have no precise idea about the natural history of normal and preeclamptic pregnant women who had antiphospholipid antibodies. In particular, how many, if any, will develop thrombosis and antiphospholipid syndrome? Genetic predisposition and environmental circumstances could create the conditions of an antigendriven affinity maturation of the antibodies which will become pathogenic. Among the heterogeneous group of antiphospholipid antibodies, lupus anticoagulant and anticardiolipin antibodies might not be as informative as other more recently described members of this group of autoantibodies. Although there is still no general consensus on the precise type of antiphospholipid antibodies that should be tested in different clinical conditions, antibodies directed against phospholipid-associated plasma protein cofactors (ß2 glycoprotein I, prothrombin, annexin V, protein C, and protein S) have been reported to be linked to an increased risk of thrombosis and even preeclampsia.30
Thus, it seems likely that lupus anticoagulant and anticardiolipin antibodies should not be tested in preeclamptic women who have no histories of thrombosis and no systemic autoimmune disease and should not be used to decide anticoagulation therapy in a selected population. Another study will have to determine the clinical interest of antiphospholipid antibodies in predicting first or recurrent preeclampsias in high-risk women.
|
Footnotes |
|---|
Received February 18, 2000. Received in revised form August 17, 2000. Accepted September 28, 2000.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. Feinstein DI. Lupus anticoagulant, anticardiolipin antibodies, fetal loss, and systemic lupus erythematosus. Blood 1992;80:859–62.
3. Hugues GRV. The antiphospholipid syndrome: Ten years on. Lancet 1993;342:341–3.[Medline]
4. Triplett DA. Antiphospholipid antibodies and recurrent pregnancy loss. Am J Reprod Immunol 1989;20:52–67.
5. Ginsberg JS, Demers C, Brill-Edwards P, Johnston M, Bona R, Burrows RF, et al. Increased thrombin generation and activity in patients with systemic lupus erythematosus and anticardiolipin antibodies: Evidence for a prothrombotic state. Blood 1993;81: 2958–63.
6. Santoro SA. Antiphospholipid antibodies and thrombotic predisposition: Underlying pathogenetic mechanisms. Blood 1994;83: 2389–91.
7. Branch DW, Andres R, Digre KB, Rote NS, Scott JR. The association of antiphospholipid antibodies with severe preeclampsia. Obstet Gynecol 1989;73:541–5.
8. Pattison NS, Chamley LW, McKay EJ, Liggins GC, Butler WS. Antiphospholipid antibodies in pregnancy: Prevalence and clinical associations. Br J Obstet Gynaecol 1993;100:909–13.[Medline]
9. Yasuda M, Takakuwa K, Tokunaga A, Tanaka K. Prospective studies of the association between anticardiolipin antibody and outcome of pregnancy. Obstet Gynecol 1995;86:555–9.[Abstract]
10. Allen JY, Tapia-Santiago C, Kutteh WH. Antiphospholipid antibodies in patients with preeclampsia. Am J Reprod Immunol 1996;36:81–5.
11. Yamamoto T, Takahashi Y, Geshi Y, Sasamori Y, Mori H. Antiphospholipid antibodies in preeclampsia and their binding ability for placental villous lipid fractions. J Obstet Gynaecol Res 1996;22:275–83.[Medline]
12. Milliez J, Lelong F, Bayani N, Jannet D, el Medjadi M, Latrous H, et al. The prevalence of antibodies during third-trimester pregnancy complicated by hypertension or idiopathic fetal growth retardation. Am J Obstet Gynecol 1991;165:51–6.[Medline]
13. Out HJ, Bruinse HW, Christiaens GCML, van Vliet M, de Groot PG, Nieuwenhuis HK, et al. A prospective, controlled multicenter study on the obstetric risks of pregnant women with antiphospholipid antibodies. Am J Obstet Gynecol 1992;167:26–32.[Medline]
14. D’Anna R, Scilipoti A, Leonardi J, Scuderi M, Jasonni VM, Leonardi R. Anticardiolipin antibodies in preeclampsia and intrauterine growth retardation. Clin Exp Obstet Gynecol 1997;24:135–7.[Medline]
15. Rai R, Regan L. Obstetric complications of antiphospholipid antibodies. Curr Opin Obstet Gynaecol 1997;9:387–90.[Medline]
16. Brand JT, Barna LF, Triplett DA. Laboratory identification of lupus anticoagulants: Results of the second international workshop for identification of lupus anticoagulants. On behalf of the subcommittee on lupus anticoagulants/antiphospholipid antibodies of the ISTH. Thromb Haemost 1995;74:1597–603.[Medline]
17. Harris EN. Antiphospholipid antibodies (Annotation). Br J Haematol 1990;74:1–9.
18. Brown MB, Forsythe AB. Robust tests for the equality of variances.J Am Stat Assoc 1974;69:364–7.
19. Levene H. Robust tests for the equality of variance. In: Olkin I, ed Contributions to probability and statistics. La Jolla, California: Standford University Press, 1960.
20. Breslow NE, Day NE. Statistical methods in cancer research: The analysis of case-controls studies. IARC scientific publications, 32.Vol 1. Lyon, France: IARC, 1980.
21. Hedelin G. Sequential tests for conditional logistic regression and combination of 2 x 2 tables in case-controls studies. Technical report, Laboratoire d’épidé miologie et de santé publique. Stras-bourg, France: Université Louis Pasteur, 1993.
22. Whitehead J. The design and analysis of sequential clinicals trials Chichester, United Kingdom: Ellis Horwood Limited, 1983.
23. de Boer K, ten Cate JW, Sturk A, Borm JJ, Treffers PE. Enhanced thrombin generation in normal and hypertensive pregnancy. Am J Obstet Gynecol 1989;160:95–100.[Medline]
24. Schjetlein R, Haugen G, Wisloff F. Markers of intravascular coagulation and fibrinolysis in preeclampsia: Association with intrauterine growth retardation. Acta Obstet Gynecol Scand 1997;76: 541–6.[Medline]
25. Infante-Rivard C, David M, Gauthier R, Rivard GE. Lupus anticoagulants anticardiolipin antibodies, and fetal loss. N Engl J Med 1991;325:1063–6.[Abstract]
26. Triplett DA. Antiphospholipid-protein antibodies: Laboratory detection and clinical relevance. Thromb Res 1995;78:1–31.[Medline]
27. Lockshin MD. Antiphospholipid antibody syndrome. Rheum Dis Clin North Am 1994;20:45–59.[Medline]
28. Teitel JM, Bauer KA, Lau HK, Rosenberg RD. Studies of the prothrombin activation pathway utilizing radioimmunoassay for the F2/F 1 + 2 fragment and thrombin-antithrombin complex. Blood 1982;59:1086–97.
29. Pelzer H, Stuber W. Markers of haemostatic activation: New perspectives using immunochemical methods for determination of prothrombin fragment F 1 + 2 and thrombin-antithrombin III complex. Thromb Haemostasis 1989;62:165–9.
30. Faden D, Tincani A, Tanzi P, Spatola L, Lojacono A, Tarantori M, et al. Anti-beta 2 glycoprotein I antibodies in general obstetric population: Preliminary results on the prevalence and correlation with pregnancy outcome. Anti-ß2 glycoprotein I antibodies are associated with some obstetrical complications, mainly preeclampsia-eclampsia. Eur J Obstet Gynecol Reprod Biol 1997;73:37–42.[Medline]
This article has been cited by other articles:
|
|
 |
|
  B. M. Sibai Thrombophilia and Severe Preeclampsia: Time to Screen and Treat in Future Pregnancies? Hypertension, December 1, 2005; 46(6): 1252 - 1253. [Full Text] [PDF]
|
|
|
|
 |
|
 K. Duckitt and D. Harrington Risk factors for pre-eclampsia at antenatal booking: systematic review of controlled studies BMJ, March 12, 2005; 330(7491): 565. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Lee, M. A. Brown, D. W. Branch, K. Ward, and R. M. Silver Anticardiolipin and Anti-{beta}2-Glycoprotein-I Antibodies in Preeclampsia Obstet. Gynecol., August 1, 2003; 102(2): 294 - 300. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. W. Branch and M. A. Khamashta Antiphospholipid Syndrome: Obstetric Diagnosis, Management, and Controversies Obstet. Gynecol., June 1, 2003; 101(6): 1333 - 1344. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Heilmann, G.-F. v. Tempelhoff, and K. Pollow Antiphospholipid Syndrome in Obstetrics Clinical and Applied Thrombosis/Hemostasis, April 1, 2003; 9(2): 143 - 150. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
