- Email: [email protected]
The association of antiphospholipid antibodies with intrauterine fetal death: A case–control study
Thrombosis Research 130 (2012) 32–37
Contents lists available at SciVerse ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
The association of antiphospholipid antibodies with intrauterine fetal death: A case–control study Linda Björk Helgadottir a, b, c,⁎, Finn Egil Skjeldestad a, d, Anne Flem Jacobsen b, Per Morten Sandset a, c, Eva-Marie Jacobsen a, c a
Department of Haematology, Oslo University Hospital Ullevål, Oslo, Norway Department of Obstetrics and Gynaecology, Oslo University Hospital Ullevål, Oslo, Norway Institute of Clinical Medicine, University of Oslo, Oslo, Norway d Department of Gynaecology and Obstetrics, Department of Clinical Medicine, University of Tromsø, Tromsø, Norway b c
a r t i c l e
i n f o
Article history: Received 22 August 2011 Received in revised form 4 October 2011 Accepted 16 November 2011 Available online 16 December 2011 Keywords: antiphospholipid antibodies fetal death pregnancy complications stillbirth thrombophilia
a b s t r a c t Introduction: Over the past few decades it has been recognized that antiphospholipid antibodies are associated with pregnancy loss. Other placenta-mediated pregnancy complications have also been associated with the presence of antiphospholipid antibodies. Most studies have measured antiphospholipid antibodies near the time of the event investigated. Objectives: To investigate the association of antiphospholipid antibodies and a history of intrauterine fetal death (IUFD) in a case–control design. Materials and methods: A case–control study of 105 women with a history of IUFD after 22 gestational weeks and 262 controls with live births. The prevalence of lupus anticoagulant, anticardiolipin- and anti-β2glycoprotein 1 antibodies were measured 3–18 years after the event of IUFD. Results: Total 9.5% of women with a history of IUFD and 5.0% of controls had at least one positive test for antiphospholipid antibodies (OR 2.0; 95% confidence interval (CI) 0.9-4.8). Women with a history of IUFD were significantly more often positive for lupus anticoagulant compared to controls (OR 4.3; 95% CI 1.0-18.4). The association of lupus anticoagulant with a history of IUFD was confined to women positive for other antiphospholipid antibodies in addition to lupus anticoagulant. Being positive for anti-β2-glycoprotein 1 or anticardiolipin antibodies alone was not significantly associated with a history of IUFD. Conclusions: Women with a history of IUFD after 22 gestational weeks were more often lupus anticoagulant positive. The association was confined to women with multiple positivity for antiphospholipid antibodies, although firm conclusions on the importance of multiple positivity cannot be made from this study. © 2011 Elsevier Ltd. All rights reserved.
Introduction Antiphospholipid antibodies (APAs) are heterogeneous autoantibodies that may be associated with increased risk of thrombotic and vascular complications [1]. Over the past few decades it has
Abbreviations: APAs, antiphospholipid antibodies; PMPC, placenta mediated pregnancy complications; aCL, anticardiolipin antibodies; IgG, immunoglobulin G isotype; LA, lupus anticoagulant; IUFD, intrauterine fetal death; VIP, the Venous Thromboembolism In Pregnancy study; anti-β2GP1, anti-β2-glycoprotein 1 antibodies; LR, lupus ratio; APTT, activated partial thromboplastin time; RVVT, Russell viper venom time; IgM, immunoglobulin M isotype; ELISA, enzyme-linked immunosorbent assay; OR, odds ratio; CI, confidence interval; ACCP, American College of Chest Physicians. ⁎ Corresponding author at: Oslo University Hospital Ullevål, Department of Obstetrics and Gynaecology, P.O.B. 4956 Nydalen, 0424 Oslo, Norway. Tel.: + 47 93429980; fax: + 47 23016211. E-mail address: [email protected] (L.B. Helgadottir). 0049-3848/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.thromres.2011.11.029
been recognized that APAs may be associated with pregnancy loss and other placenta mediated pregnancy complications (PMPC), such as intrauterine growth restriction, preeclampsia, placental abruption, as well as pregnancy loss in all trimesters [2–4]. PMPC affect more than one in six pregnancies, are major causes of maternal and fetal morbidity and mortality, and have been attributed to uteroplacental vascular insufficiency [5,6]. The pathophysiology of PMPC related to APAs probably does not only involve thrombosis of placental vessels, but also impaired conceptus implantation by damage to decidual or chorionic vessels or reduction of trophoblast invasiveness, which can lead to early pregnancy loss, or later in pregnancy, placental insufficiency and PMPC [7]. Previous studies have demonstrated an association of both inherited and acquired thrombophilia with PMPC, and it has been reported that up to 65% of women with PMPC have some form of acquired or inherited thrombophilia [8,9]. Both case–control and prospective cohort studies have found increased risk of PMPC in APA positive
L.B. Helgadottir et al. / Thrombosis Research 130 (2012) 32–37
women, and a systematic review reported a positive association between fetal death and the presence of both anticardiolipin (aCL) antibodies of immunoglobulin G (IgG) isotype and lupus anticoagulant (LA) [8–12]. Studies investigating the association between APAs and intrauterine fetal death (IUFD) have often been of small sample size [13], and they have differed in selection criteria for cases and controls [14,15]. In addition, laboratory detection of APAs is complicated by the heterogeneity of both the antibodies and the assays used for detection. All these issues can explain the inconsistent findings of the association of APAs with IUFD. Investigators reporting on the association between APAs and IUFD have usually analyzed the prevalence of APAs in blood samples collected within months after suffering IUFD. To our knowledge the prevalence of APAs several years after the incident, among women with a history of IUFD, has not been reported. Our hypothesis was that women with a history of IUFD were more often APA positive, compared to women with live births only. The aim of our study was to investigate this association between APAs and IUFD.
Materials and methods Ethics statement The Norwegian South-Eastern Regional Committee for Medical Research Ethics approved the study. Authorization for the use of information from medical records for research purposes was obtained from the Norwegian Ministry of Health and Social Affairs. The Norwegian Data Inspectorate approved the use of data comprising sensitive personal health information, and the merging of clinical and registerdata by using the unique 11-digit personal identification number given to all Norwegian citizens either by birth or immigration. A written informed consent was obtained from all study participants. The present study was a part of a larger hospital-based case–control study; the Venous Thromboembolism In Pregnancy (VIP) study, and was registered as a clinical observational study at www. clinicaltrials.gov, with registration number NCT00856076. Data on clinical risk factors for IUFD and clinical and biochemical risk factors
33
for venous thrombosis related to pregnancy have been published earlier [16–19]. Selection of cases and controls Women with a diagnosis of IUFD were identified retrospectively by a search for selected codes of the WHO International Classification of Diseases versions 9 or 10 that were registered in the patient administrative system of two Norwegian hospitals, that is, Oslo University Hospital Ullevål, Oslo, and Akershus University Hospital, Nordbyhagen, from January 1990 throughout December 2003. We identified 436 possible cases of IUFD, defined by fetal death after 22 completed gestational weeks or birth weight >500 g. We excluded 49 cases wrongly diagnosed and 8 with non-retrievable records leaving 379 women identified as cases with a verified diagnosis of IUFD in singleton or duplex pregnancies (Fig. 1). In 2006, the controls received an invitation to participate in the thrombosis part of the VIPstudy, in which 353/1229 (28.7%) agreed to participate. These 353 controls signed a consent to participate and agreed to receive a new questionnaire at a later time regarding the present study on IUFD [17,18]. The medical records for cases and controls were reviewed for validation of the diagnosis of IUFD, and information on demographics, general health, obstetrical history, details of the index pregnancy, labor, and delivery. The women's unique personal identification numbers were then merged with census data (Statistics Norway, Oslo, Norway). Women, who had emigrated, died, or had an invalid or foreign address, were excluded. Nine controls with a history of IUFD were also excluded. This left us with 346 cases and 326 controls eligible for study participation (Fig. 1). The participants were approached during 2006–8 by a letter outlining the purpose of the study. Those interested in participating contacted us by e-mail or telephone to schedule an appointment to donate a blood sample and to answer a questionnaire regarding socio-demographic factors, obstetrical history, general and psychological health, and quality of life. One of the cases did not donate a blood sample and was therefore excluded from the study. After two reminders the final study population comprised 105 cases and 262 controls (Fig. 1).
Fig. 1. Flowchart – selection of study population.
34
L.B. Helgadottir et al. / Thrombosis Research 130 (2012) 32–37
Blood sampling and analyses Blood was collected in 5 mL Vacutainer tubes (Becton-Dickinson, Meylan-Cedex, France) containing 0.5 mL buffered citrate (0.129 mol/L). The tubes were centrifuged at 2000 g for 15 minutes within 1 hour, and plasma aliquots were frozen and kept at −70 °C until assayed. The blood was analysed for LA, aCL, and anti-β2-glycoprotein 1 (anti-β2GP1) antibodies. The assays were performed at the Hematologic Research Laboratory, Department of Hematology, Oslo University Hospital Ullevål (formerly Ullevål University Hospital) as described earlier [18]. In short, the presence of LA was identified using validated in-house lupus ratio (LR) tests, which are automated, quantitative, integrated tests for LA [20,21]. Two LR tests were performed, one based on the activated partial thromboplastin time (LR-APTT) and the other based on the Russell viper venom time (LR-RVVT). The LR tests were performed in 1:1 mixture of patient plasma and pooled normal plasma. For each of the LR tests two coagulation times were measured, one with a reagent with low and the other with a high phospholipid concentration. The ratio between the two coagulation times (low phospholipid/high phospholipid concentration) was divided by the corresponding ratio obtained with pooled normal plasma. The final ratio is defined as the LR of that patient's plasma [21]. The reagents were made from different concentrations of natural phospholipids (crude cephalin, generously provided by Dr. Tore Janson, AxisShield PoC AS, Oslo, Norway). In the APTT-based assay, a constant concentration of ellagic acid (Sigma-Aldrich, St. Louis, Missouri, USA) was used as activator. In the RVVT-based test, RVV (SigmaAldrich) activates factor X directly. The 99th percentile of the LR of the control group was chosen as the upper reference limit, and was 1.22 for the APTT-based LR test and 1.19 for the RVVT-based LR test. aCL IgG and IgM isotypes were analyzed with an in-house enzyme-linked immunosorbent assay (ELISA) essentially as described by Gharavi et al. [22]. We used serial dilutions of an in-house control drawn from a strongly aCL positive patient, which were standardized against Harris’ commercial standards (American Diagnostica Inc., Stamford, CT, USA). Values for IgG and IgM isotypes of aCL were reported in GPL units and MPL units, respectively. The cut-off values for a positive test were defined by the 99th percentile of the values of the control group, and were 10.7 for aCL IgG and 23.7 for aCL IgM. Anti-ß2GP1 IgG and IgM isotypes were assayed with commercial ELISA kits (QUANTA Lite TM β2 GP1 IgG/IgM, INOVA Diagnostics Inc., San Diego, USA) for semi-quantitative determination. Results were expressed in standard IgG and IgM anti-ß2GP1 units, that is, SGU and SMU, respectively. The 99th percentiles of the control group were used as upper reference limits, and were 6.5 for anti-ß2GP1 IgG and 30.3 for anti-ß2GP1 IgM. Statistical analyses Data were analyzed by Chi-square tests or Fisher's exact tests. Results were presented as percentages and odds ratios (OR) with 95% confidence intervals (CI). In the case of missing values for sociodemographic or clinical variables women were denoted the reference group in that particular analysis. Significance level was set at p b 0.05. All data was analyzed using the Statistical Package for Social Science version 16.0 (SPSS Inc, Chicago, Il, USA). Results Information from medical records of the eligible participants gave us the opportunity to compare participating women with the nonparticipating. There were no significant differences in the demographic and clinical data between the participating and nonparticipating women (data not shown). The maternal characteristics of cases and controls are displayed in Table 1. At the time of the
index pregnancy women with IUFD were younger, smoked more often at first visit, had more often placental abruption, their fetuses were more often small for gestational age and they had more often inherited thrombophilia. Prevalences of demographic and clinical risk factors were not significantly different between APA-positive and APA-negative women at the time of the index pregnancy (Table 2). At the time of blood sampling there was not a significant difference in the rate of miscarriages ≤22 gestational weeks between cases and controls, but the cases had more frequently had placental abruption (Table 3). The APA-positive women had more often a history of recurrent miscarriages compared to APA-negative women (Table 3). Mean time difference between index pregnancy and blood sampling were 8.7 years and 8.6 years for APA positive and APA negative women, respectively. The prevalence of APAs and the association of APAs with IUFD are shown in Table 4. Twenty-three women, 10 (9.5%) with a history of IUFD and 13 (5.0%) controls, were positive for at least one APA test (OR 2.0; 95% CI 0.9-4.8). The OR for multiple positivity for APAs among women with IUFD was 7.9 (95% CI 0.8-76.5) as compared to controls, using APA-negativity as a reference, were as the OR for having a single positive test was 1.5 (95% CI 0.6-4.0). Women with more than one positive APA test were all positive for LA. LA was significantly more often positive among women with a history of IUFD, compared to women with live births only (OR 4.3; 95% CI 1.0-18.4). Exploring the LA positive women by single/multiple positivity for APA we found that the risk related to LA was confined to women positive for LA in combination with other APAs. Three cases but no controls were positive for other APA tests in addition to LA. Anti-ß2GP1, both IgG and IgM, and aCL IgG and IgM were not significantly associated with a history of IUFD. Nineteen women had one APA test positive, that is 7 (6.7%) cases and 12 (4.6%) controls. One case and one control were positive for two tests, and one case each was positive for three and four tests.
Table 1 Maternal characteristics at index pregnancy. Variable
Age (at index pregnancy) b35 years ≥35 years Parity 0 1 ≥2 Multiple pregnancy Hypertensive disorders (HD) Preeclampsia Hypertension Small for gestational age (SGA) HD and/or SGA No HD or SGA HD, no SGA HD with SGA SGA, no HD Diabetes Placental abruption Placenta previa Smoking (at first visit) Inherited thrombophilia*
Cases
Controls
N = 105**
N = 262
p-value
%
%
80.0 20.0
68.3 31.7
n.s. b0.03
52.4 34.3 13.3 3.8
50.4 39.3 10.3 2.7
n.s. n.s.
5.7 5.7 34.3
7.3 4.6 1.5
59.0 6.7 4.8 29.5 1.0 8.6 2.9 27.6 18.4
87.8 10.7 1.1 0.4 0.4 0.8 0.8 8.4 11.8
n.s. n.s. b0.001 Reference n.s b0.02 b0.001 n.s. b0.001 n.s. b0.001 n.s.
p –value - represents results from univariate analysis. n.s.: not significant. *Inherited thrombophilia: factor V Leiden, prothrombin G20210A polymorphism, antithrombin, protein C- or protein S deficiency. **Two women using warfarin were excluded from analyses involving protein C and protein S.
L.B. Helgadottir et al. / Thrombosis Research 130 (2012) 32–37 Table 2 Demographic and clinical risk factors among antiphospholipid antibody (APA) positive and APA negative women. Variable
Cases Controls Age (at index pregnancy) b 35 years ≥35 years Parity 0 1 ≥2 Multiple pregnancy Hypertensive disorders (HD) Preeclampsia Hypertension Small for gestational age (SGA) HD and/or SGA No HD or SGA HD, no SGA HD with SGA SGA, no HD Diabetes Placental abruption Placenta previa Smoking (at first visit) Inherited thrombophilia*
APA positive
APA negative
N = 23**
N = 344**
%
%
p-value
43.5 56.5
27.6 72.4
n.s. n.s.
78.3 21.7
71.2 28.8
n.s. n.s.
43.5 34.8 21.7 0
51.5 38.1 10.5 3.2
n.s. n.s.
4.3 8.7 17.4
7.0 4.7 10.5
n.s. n.s. n.s.
73.9 8.7 4.3 13.0 0 0 0 21.7 4.5
79.9 9.6 2.0 8.4 0.6 3.2 1.5 13.4 14.3
Reference n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
p –value - represents results from univariate analysis. n.s.: not significant. *Inherited thrombophilia: factor V Leiden, prothrombin G20210A polymorphism, antithrombin-, protein C- or protein S deficiency. **Two women, one in each group, using warfarin were excluded from analyses involving protein C and protein S.
Discussion In the present study, we found that women with a history of IUFD, as compared with women with live births only, were significantly more often positive for LA 3–18 years after the index pregnancy. This was attributed to women positive for LA in combination with other APAs, although this study does not allow firm conclusions on the importance of multi-positivity. Our finding of increased prevalence of LA among women with a history of IUFD is in agreement with other studies. However, most investigators, as summarized in a systematic review [9], have found aCL to be an even stronger predictor of IUFD than LA [13–15,23]. Gonen et al. studied the prevalence of both inherited and acquired thrombophilia among women with unexplained IUFD after 26 gestational weeks and found LA and aCL to be significantly more prevalent among women with a history of IUFD compared to controls, with OR 6.1 (95% CI 1.4-36.2) and 8.5 (95% CI 1.6-42.3), respectively [13]. The NOHA-study found, in univariate analysis, LA as well as aCL IgG and anti-ß2GP1 IgG to be associated with IUFD after 22 weeks of pregnancy. However, in the multivariate analysis, adjusting for other APAs and inherited thrombophilia, LA was protective of IUFD, but both aCL IgG and anti-β2GP1 IgG were still positive predictors for IUFD [15]. The cause was found to be that no cases, but two controls, had LA as
35
Table 4 Prevalence of APA and OR with 95% CI for APA among women with IUFD compared to controls. Variable
APA positive – ≥ 1 tests APA positive – 1 test APA positive – >1 tests LA positive – ≥ 1 test LA positive + anti-ß2GP1/aCL negative LA positive + anti-ß2GP1/aCL positive LR-APTT positive LR- RVVT positive Anti- ß2GP1 positive Unique anti- ß2GP1 positivity Anti- ß2GP1 + ≥1 LA/aCL positive test aCL positive Unique aCL positivity aCL + ≥1 LA/ anti- ß2GP1 positive test
Cases (N = 105)
Controls (N = 262)
%
(n)
%
(n)
9.5 6.7 2.9 4.8 1.9 2.9 4.8 1.9 4.8 2.9 1.9 3.8 1.9 1.9
(10) (7) (3) (5) (2) (3) (5) (2) (5) (3) (2) (4) (2) (2)
5.0 4.6 0.4 1.1 1.1 0 0.8 0.8 1.5 1.5 0 2.3 2.3 0
(13) (12) (1) (3) (3) 0 (2) (2) (4) (4) 0 (6) (6) 0
OR
95% CI
2.0 1.5 7.9 4.3 1.7 6.5 2.5 3.2 1.9 1.7 0.8 -
0.9-4.8 0.6-4.0 0.8-76.5 1.0-18.4 0.3-10.5 1.2-34.0 0.4-18.2 0.8-12.3 0.4-8.8 0.5-6.1 0.2-4.3 -
APA: antiphospholipid antibodies, OR: odds ratio, CI: confidence interval, IUFD: intrauterine fetal death, LA: lupus anticoagulant, LR: lupus ratio, APTT: activated partial thromboplastin time-based, RVVT: Russell viper venom test-based, aCL: anticardiolipin antibodies, anti-ß2GP1: anti-β2 glycoprotein 1.
a unique APA marker. This is in line with our findings that the risk of IUFD related to LA is mainly attributed to women positive for other APAs as well. In studies on thrombosis, a more consistent association with both arterial and venous thrombosis has been found for LA than for aCL [24]. Results of studies are largely influenced by differences in study design; for instance eligibility criteria, laboratory methods, and cut-off values. In accordance with the recent update on the guidelines for LA detection, we used two LR tests for LA, one based on the APTT, the other based on RVVT [25]. The LR tests are automated tests; integrating screening, mixing and confirmatory procedures. The results are calculated as LA ratios (screen/confirm) and normalized against values obtained with a pooled normal plasma, as recommended by the guidelines [25]. The LR tests have proven high reproducibility and low interlaboratory variation in an international multilaboratory study and where hence chosen for this study [21]. The majority of studies on APA and IUFD have applied various exclusion criteria in the attempt to investigate women with unexplained IUFD only. Thus they have excluded women with known causes of IUFD. In most studies, IUFD explained by congenital malformations, abnormal karyotypes, uterine malformations, fetal hydrops or infections have been excluded. In some studies, women with maternal risk factors for IUFD, like diabetes, preeclampsia, placental abruption or intrauterine fetal growth restriction have also been excluded, as well as women with a history of miscarriages and known thrombophilia. Such differences in study-design make comparison of results difficult. Our cases were unselected related to cause and we did not exclude women with known thrombophilia, miscarriages or known risk factors for IUFD. By this our findings are more applicable to the general obstetric population. In a recent review article, Kist et al. demonstrated that the relationship between adverse pregnancy outcome and thrombophilia was influenced by
Table 3 Obstetrical history, until time of blood sampling. Variable
Cases (N = 105) %
Controls (N = 262) %
p-value
APA-positive (N = 23) %
APA-negative (N = 344) %
p-value
Miscarriage ≤ 22 weeks Recurrent miscarriage (≥3) ≤22 weeks Abruption in any pregnancy
39.0 5.7 11.4
30.9 2.3 1.1
n.s n.s. b0.001
47.8 13.0 0
32.3 2.6 4.4
n.s. 0.03 n.s
p –value - represents results from univariate analysis. n.s.: not significant. APA = antiphospholipid antibodies.
36
L.B. Helgadottir et al. / Thrombosis Research 130 (2012) 32–37
confounding factors like ethnicity, methods of testing, and severity of disease [26]. About 5-7% of healthy pregnant women have been reported to have positive tests for APAs [27]. The prevalence of APAs depends on the definition of a positive test. It is recommended by international consensus to define the cut-off value by the 99th percentile of a control group. Since by definition 1% of the controls will have a positive result for each test, the prevalence of APAs in the healthy population will be influenced by the number of tests for APAs examined. Thus, if six different tests are used, up to 6% of the control/normal population will be APA positive. Although there is some increased relative risk of IUFD associated with APAs, the absolute risk for APA positive women without previous clinical events is low, and the probability of a successful pregnancy outcome is high. Thus, the screening for APAs in an unselected population of pregnant women is not recommended. The 8th Guidelines on Antithrombotic Therapy of the American College of Chest Physicians (ACCP) from 2008 recommend screening for APAs among women with a history of PMPC [28]. Low molecular weight heparin and low dose acetylsalicylic acid are currently recommended for the prevention of recurrent pregnancy loss in women with APAs [28–31]. Current recommendations on treatment of the APA syndrome may have long-term implications. For example, indefinite treatment is recommended after a first venous thrombosis if the patient has tested positive for APA twice with three months interval. This is a more rigorous secondary thromboprophylactic therapy than usually advised, even though little or no data support such recommendations [32]. This study implies that multiple positivity is probably important although the nature of the study does not allow firm conclusion. This is in agreement with other recent studies that have demonstrated that multiple positivity for APAs is more frequently associated with pregnancy complications than single positivity [33,34]. Collection of blood samples only at a single time point and a long time after the index pregnancy are limitations of our study. However, one of the rationales for repeated testing is to avoid false positive tests due to transiently elevated APAs, which is not a concern in the present study, when the samples were collected 3–18 years after the index pregnancy. Moreover, we can not exclude the possibility that some women have either turned negative or turned positive for APAs after the index pregnancy, but there is no reason to believe that this would differ between cases and controls. Remarkably little is known about the sustainability of APAs over time. In one study, Erkan et al. found sustained positivity over time in approximately 75% of tests initially positive, but with a mean follow-up time of only 2.4, 3.5 and 1.0 years for LA, aCL, and anti-β2 GP1, respectively [35]. Another limitation of our study is that only 29% of the identified cases participated, which probably reflects the burden of this serious complication. In spite of this our sample size is relatively large compared to other studies in this field. We had information from medical records from all the eligible participants and we did not find significant differences in sociodemographic and clinical factors between participating and non-participating women. We therefore do not believe that the low participation applies a selection bias, but it does affect the power of the study. We conclude that women with a history of IUFD after 22 gestational weeks were more often LA positive, 3–18 years after the incident, but the risk of IUFD related to LA was confined to women positive for other APAs in addition to LA. However, there is still great uncertainty related to the association of APAs and IUFD and the clinical importance is not easily predicated.
Conflict of Interest No conflicts of interest.
Acknowledgments Financial support was received from the South-Eastern Norway Regional Health Authority Trust, Hamar, Norway, the Oslo University Hospital Ulleval Scientific Trust, Oslo, Norway, and the Research Council of Norway. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
References [1] Miyakis S, Lockshin MD, Atsumi T, Branch DW, Brey RL, Cervera R, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006;4:295–306. [2] Lynch A, Marlar R, Murphy J, Davila G, Santos M, Rutledge J, et al. Antiphospholipid antibodies in predicting adverse pregnancy outcome. A prospective study. Ann Intern Med 1994;120:470–5. [3] Kujovich JL. Thrombophilia and pregnancy complications. Am J Obstet Gynecol 2004;191:412–4. [4] Triplett DA, Harris EN. Antiphospholipid antibodies and reproduction. Am J Reprod Immunol 1989;21:123–31. [5] Rodger MA, Paidas M, Claire M, Middeldorp S, Kahn S, Martinelli I, et al. Inherited thrombophilia and pregnancy complications revisited. Obstet Gynecol 2008;112: 320–4. [6] Berg CJ, Atrash HK, Koonin LM, Tucker M. Pregnancy-related mortality in the United States, 1987–1990. Obstet Gynecol 1996;88:161–7. [7] Rand JH, Wu XX, Guller S, Scher J, Andree HA, Lockwood CJ. Antiphospholipid immunoglobulin G antibodies reduce annexin-V levels on syncytiotrophoblast apical membranes and in culture media of placental villi. Am J Obstet Gynecol 1997;177: 918–23. [8] Kupferminc MJ, Eldor A, Steinman N, Many A, Bar-Am A, Jaffa A, et al. Increased frequency of genetic thrombophilia in women with complications of pregnancy. N Engl J Med 1999;340:9–13. [9] Robertson L, Wu O, Langhorne P, Twaddle S, Clark P, Lowe GD, et al. Thrombophilia in pregnancy: a systematic review. Br J Haematol 2006;132:171–96. [10] 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. [11] Lockwood CJ, Romero R, Feinberg RF, Clyne LP, Coster B, Hobbins JC. The prevalence and biologic significance of lupus anticoagulant and anticardiolipin antibodies in a general obstetric population. Am J Obstet Gynecol 1989;161:369–73. [12] Katano K, Aoki A, Sasa H, Ogasawara M, Matsuura E, Yagami Y. beta 2Glycoprotein I-dependent anticardiolipin antibodies as a predictor of adverse pregnancy outcomes in healthy pregnant women. Hum Reprod 1996;11:509–12. [13] Gonen R, Lavi N, Attias D, Schliamser L, Borochowitz Z, Toubi E, et al. Absence of association of inherited thrombophilia with unexplained third-trimester intrauterine fetal death. Am J Obstet Gynecol 2005;192:742–6. [14] Bocciolone L, Meroni P, Parazzini F, Tincani A, Radici E, Tarantini M, et al. Antiphospholipid antibodies and risk of intrauterine late fetal death. Acta Obstet Gynecol Scand 1994;73:389–92. [15] Gris JC, Quere I, Monpeyroux F, Mercier E, Ripart-Neveu S, Tailland ML, et al. Case– control study of the frequency of thrombophilic disorders in couples with late foetal loss and no thrombotic antecedent–the Nimes Obstetricians and Haematologists Study5 (NOHA5). Thromb Haemost 1999;81:891–9. [16] Helgadottir LB, Skjeldestad FE, Jacobsen AF, Sandset PM, Jacobsen EM. Incidence and risk factors of fetal death in Norway: a case–control study. Acta Obstet Gynecol Scand 2011;90:390–7. [17] Jacobsen AF, Skjeldestad FE, Sandset PM. Ante- and postnatal risk factors of venous thrombosis: a hospital-based case–control study. J Thromb Haemost 2008;6:905–12. [18] Bergrem A, Jacobsen EM, Skjeldestad FE, Jacobsen AF, Skogstad M, Sandset PM. The association of antiphospholipid antibodies with pregnancy-related first time venous thrombosis - a population-based case–control study. Thromb Res 2010;125:e222–7. [19] Jacobsen AF, Dahm A, Bergrem A, Jacobsen EM, Sandset PM. Risk of venous thrombosis in pregnancy among carriers of the factor V Leiden and the prothrombin gene G20210A polymorphisms. J Thromb Haemost 2010;8:2443–9. [20] Schjetlein R, Sletnes KE, Wisloff F. A quantitative, semi-automated and computerassisted test for lupus anticoagulant. Thromb Res 1993;69:239–50. [21] Jacobsen EM, Barna-Cler L, Taylor JM, Triplett DA, Wisloff F. The Lupus Ratio test– an interlaboratory study on the detection of Lupus anticoagulants by an APTTbased, integrated, and semi-quantitative test. Fifth International Survey of Lupus Anticoagulants–ISLA 5. Thromb Haemost 2000;83:704–8. [22] Gharavi AE, Harris EN, Asherson RA, Hughes GR. Anticardiolipin antibodies: isotype distribution and phospholipid specificity. Ann Rheum Dis 1987;46:1–6. [23] Alfirevic Z, Roberts D, Martlew V. How strong is the association between maternal thrombophilia and adverse pregnancy outcome? A systematic review. Eur J Obstet Gynecol Reprod Biol 2002;101:6–14. [24] Galli M, Luciani D, Bertolini G, Barbui T. Lupus anticoagulants are stronger risk factors for thrombosis than anticardiolipin antibodies in the antiphospholipid syndrome: a systematic review of the literature. Blood 2003;101: 1827–32.
L.B. Helgadottir et al. / Thrombosis Research 130 (2012) 32–37 [25] Pengo V, Tripodi A, Reber G, Rand JH, Ortel TL, Galli M, et al. Subcommittee on Lupus Anticoagulant/Antiphospholipid Antibody of the Scientific and Standardisation Committee of the International Society on Thrombosis and Haemostasis. Update of the guidelines for lupus anticoagulant detection. J Thromb Haemost 2009;7:1737–40. [26] Kist WJ, Janssen NG, Kalk JJ, Hague WM, Dekker GA, de Vries JI. Thrombophilias and adverse pregnancy outcome - A confounded problem! Thromb Haemost 2008;99:77–85. [27] Parke AL, Wilson D, Maier D. The prevalence of antiphospholipid antibodies in women with recurrent spontaneous abortion, women with successful pregnancies, and women who have never been pregnant. Arthritis Rheum 1991;34:1231–5. [28] Bates SM, Greer IA, Pabinger I, Sofaer S, Hirsh J, American College of Chest Physicians. Venous thromboembolism, thrombophilia, antithrombotic therapy, and pregnancy: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133(Suppl.-886S). [29] Rai R, Cohen H, Dave M, Regan L. Randomised controlled trial of aspirin and aspirin plus heparin in pregnant women with recurrent miscarriage associated with phospholipid antibodies (or antiphospholipid antibodies). BMJ 1997;314:253–7.
37
[30] Brenner B, Bar J, Ellis M, Yarom I, Yohai D, Samueloff A. Live-Enox Investigators. Effects of enoxaparin on late pregnancy complications and neonatal outcome in women with recurrent pregnancy loss and thrombophilia: results from the Live-Enox study. Fertil Steril 2005;84:770–3. [31] Greer IA, Nelson-Piercy C. Low-molecular-weight heparins for thromboprophylaxis and treatment of venous thromboembolism in pregnancy: a systematic review of safety and efficacy. Blood 2005;106:401–7. [32] Derksen RH, de Groot PG. Towards evidence-based treatment of thrombotic antiphospholipid syndrome. Lupus 2010;19:470–4. [33] Ruffatti A, Tonello M, Cavazzana A, Bagatella P, Pengo V. Laboratory classification categories and pregnancy outcome in patients with primary antiphospholipid syndrome prescribed antithrombotic therapy. Thromb Res 2009;123:482–7. [34] Tripodi A, de Groot PG, Pengo V. Antiphospholipid syndrome: laboratory detection, mechanisms of action and treatment. J Intern Med 2011;270:110–22. [35] Erkan D, Derksen WJ, Kaplan V, Sammaritano L, Pierangeli SS, Roubey R, et al. Real world experience with antiphospholipid antibody tests: how stable are results over time? Ann Rheum Dis 2005;64:1321–5.
Contents lists available at SciVerse ScienceDirect
Thrombosis Research journal homepage: www.elsevier.com/locate/thromres
Regular Article
The association of antiphospholipid antibodies with intrauterine fetal death: A case–control study Linda Björk Helgadottir a, b, c,⁎, Finn Egil Skjeldestad a, d, Anne Flem Jacobsen b, Per Morten Sandset a, c, Eva-Marie Jacobsen a, c a
Department of Haematology, Oslo University Hospital Ullevål, Oslo, Norway Department of Obstetrics and Gynaecology, Oslo University Hospital Ullevål, Oslo, Norway Institute of Clinical Medicine, University of Oslo, Oslo, Norway d Department of Gynaecology and Obstetrics, Department of Clinical Medicine, University of Tromsø, Tromsø, Norway b c
a r t i c l e
i n f o
Article history: Received 22 August 2011 Received in revised form 4 October 2011 Accepted 16 November 2011 Available online 16 December 2011 Keywords: antiphospholipid antibodies fetal death pregnancy complications stillbirth thrombophilia
a b s t r a c t Introduction: Over the past few decades it has been recognized that antiphospholipid antibodies are associated with pregnancy loss. Other placenta-mediated pregnancy complications have also been associated with the presence of antiphospholipid antibodies. Most studies have measured antiphospholipid antibodies near the time of the event investigated. Objectives: To investigate the association of antiphospholipid antibodies and a history of intrauterine fetal death (IUFD) in a case–control design. Materials and methods: A case–control study of 105 women with a history of IUFD after 22 gestational weeks and 262 controls with live births. The prevalence of lupus anticoagulant, anticardiolipin- and anti-β2glycoprotein 1 antibodies were measured 3–18 years after the event of IUFD. Results: Total 9.5% of women with a history of IUFD and 5.0% of controls had at least one positive test for antiphospholipid antibodies (OR 2.0; 95% confidence interval (CI) 0.9-4.8). Women with a history of IUFD were significantly more often positive for lupus anticoagulant compared to controls (OR 4.3; 95% CI 1.0-18.4). The association of lupus anticoagulant with a history of IUFD was confined to women positive for other antiphospholipid antibodies in addition to lupus anticoagulant. Being positive for anti-β2-glycoprotein 1 or anticardiolipin antibodies alone was not significantly associated with a history of IUFD. Conclusions: Women with a history of IUFD after 22 gestational weeks were more often lupus anticoagulant positive. The association was confined to women with multiple positivity for antiphospholipid antibodies, although firm conclusions on the importance of multiple positivity cannot be made from this study. © 2011 Elsevier Ltd. All rights reserved.
Introduction Antiphospholipid antibodies (APAs) are heterogeneous autoantibodies that may be associated with increased risk of thrombotic and vascular complications [1]. Over the past few decades it has
Abbreviations: APAs, antiphospholipid antibodies; PMPC, placenta mediated pregnancy complications; aCL, anticardiolipin antibodies; IgG, immunoglobulin G isotype; LA, lupus anticoagulant; IUFD, intrauterine fetal death; VIP, the Venous Thromboembolism In Pregnancy study; anti-β2GP1, anti-β2-glycoprotein 1 antibodies; LR, lupus ratio; APTT, activated partial thromboplastin time; RVVT, Russell viper venom time; IgM, immunoglobulin M isotype; ELISA, enzyme-linked immunosorbent assay; OR, odds ratio; CI, confidence interval; ACCP, American College of Chest Physicians. ⁎ Corresponding author at: Oslo University Hospital Ullevål, Department of Obstetrics and Gynaecology, P.O.B. 4956 Nydalen, 0424 Oslo, Norway. Tel.: + 47 93429980; fax: + 47 23016211. E-mail address: [email protected] (L.B. Helgadottir). 0049-3848/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.thromres.2011.11.029
been recognized that APAs may be associated with pregnancy loss and other placenta mediated pregnancy complications (PMPC), such as intrauterine growth restriction, preeclampsia, placental abruption, as well as pregnancy loss in all trimesters [2–4]. PMPC affect more than one in six pregnancies, are major causes of maternal and fetal morbidity and mortality, and have been attributed to uteroplacental vascular insufficiency [5,6]. The pathophysiology of PMPC related to APAs probably does not only involve thrombosis of placental vessels, but also impaired conceptus implantation by damage to decidual or chorionic vessels or reduction of trophoblast invasiveness, which can lead to early pregnancy loss, or later in pregnancy, placental insufficiency and PMPC [7]. Previous studies have demonstrated an association of both inherited and acquired thrombophilia with PMPC, and it has been reported that up to 65% of women with PMPC have some form of acquired or inherited thrombophilia [8,9]. Both case–control and prospective cohort studies have found increased risk of PMPC in APA positive
L.B. Helgadottir et al. / Thrombosis Research 130 (2012) 32–37
women, and a systematic review reported a positive association between fetal death and the presence of both anticardiolipin (aCL) antibodies of immunoglobulin G (IgG) isotype and lupus anticoagulant (LA) [8–12]. Studies investigating the association between APAs and intrauterine fetal death (IUFD) have often been of small sample size [13], and they have differed in selection criteria for cases and controls [14,15]. In addition, laboratory detection of APAs is complicated by the heterogeneity of both the antibodies and the assays used for detection. All these issues can explain the inconsistent findings of the association of APAs with IUFD. Investigators reporting on the association between APAs and IUFD have usually analyzed the prevalence of APAs in blood samples collected within months after suffering IUFD. To our knowledge the prevalence of APAs several years after the incident, among women with a history of IUFD, has not been reported. Our hypothesis was that women with a history of IUFD were more often APA positive, compared to women with live births only. The aim of our study was to investigate this association between APAs and IUFD.
Materials and methods Ethics statement The Norwegian South-Eastern Regional Committee for Medical Research Ethics approved the study. Authorization for the use of information from medical records for research purposes was obtained from the Norwegian Ministry of Health and Social Affairs. The Norwegian Data Inspectorate approved the use of data comprising sensitive personal health information, and the merging of clinical and registerdata by using the unique 11-digit personal identification number given to all Norwegian citizens either by birth or immigration. A written informed consent was obtained from all study participants. The present study was a part of a larger hospital-based case–control study; the Venous Thromboembolism In Pregnancy (VIP) study, and was registered as a clinical observational study at www. clinicaltrials.gov, with registration number NCT00856076. Data on clinical risk factors for IUFD and clinical and biochemical risk factors
33
for venous thrombosis related to pregnancy have been published earlier [16–19]. Selection of cases and controls Women with a diagnosis of IUFD were identified retrospectively by a search for selected codes of the WHO International Classification of Diseases versions 9 or 10 that were registered in the patient administrative system of two Norwegian hospitals, that is, Oslo University Hospital Ullevål, Oslo, and Akershus University Hospital, Nordbyhagen, from January 1990 throughout December 2003. We identified 436 possible cases of IUFD, defined by fetal death after 22 completed gestational weeks or birth weight >500 g. We excluded 49 cases wrongly diagnosed and 8 with non-retrievable records leaving 379 women identified as cases with a verified diagnosis of IUFD in singleton or duplex pregnancies (Fig. 1). In 2006, the controls received an invitation to participate in the thrombosis part of the VIPstudy, in which 353/1229 (28.7%) agreed to participate. These 353 controls signed a consent to participate and agreed to receive a new questionnaire at a later time regarding the present study on IUFD [17,18]. The medical records for cases and controls were reviewed for validation of the diagnosis of IUFD, and information on demographics, general health, obstetrical history, details of the index pregnancy, labor, and delivery. The women's unique personal identification numbers were then merged with census data (Statistics Norway, Oslo, Norway). Women, who had emigrated, died, or had an invalid or foreign address, were excluded. Nine controls with a history of IUFD were also excluded. This left us with 346 cases and 326 controls eligible for study participation (Fig. 1). The participants were approached during 2006–8 by a letter outlining the purpose of the study. Those interested in participating contacted us by e-mail or telephone to schedule an appointment to donate a blood sample and to answer a questionnaire regarding socio-demographic factors, obstetrical history, general and psychological health, and quality of life. One of the cases did not donate a blood sample and was therefore excluded from the study. After two reminders the final study population comprised 105 cases and 262 controls (Fig. 1).
Fig. 1. Flowchart – selection of study population.
34
L.B. Helgadottir et al. / Thrombosis Research 130 (2012) 32–37
Blood sampling and analyses Blood was collected in 5 mL Vacutainer tubes (Becton-Dickinson, Meylan-Cedex, France) containing 0.5 mL buffered citrate (0.129 mol/L). The tubes were centrifuged at 2000 g for 15 minutes within 1 hour, and plasma aliquots were frozen and kept at −70 °C until assayed. The blood was analysed for LA, aCL, and anti-β2-glycoprotein 1 (anti-β2GP1) antibodies. The assays were performed at the Hematologic Research Laboratory, Department of Hematology, Oslo University Hospital Ullevål (formerly Ullevål University Hospital) as described earlier [18]. In short, the presence of LA was identified using validated in-house lupus ratio (LR) tests, which are automated, quantitative, integrated tests for LA [20,21]. Two LR tests were performed, one based on the activated partial thromboplastin time (LR-APTT) and the other based on the Russell viper venom time (LR-RVVT). The LR tests were performed in 1:1 mixture of patient plasma and pooled normal plasma. For each of the LR tests two coagulation times were measured, one with a reagent with low and the other with a high phospholipid concentration. The ratio between the two coagulation times (low phospholipid/high phospholipid concentration) was divided by the corresponding ratio obtained with pooled normal plasma. The final ratio is defined as the LR of that patient's plasma [21]. The reagents were made from different concentrations of natural phospholipids (crude cephalin, generously provided by Dr. Tore Janson, AxisShield PoC AS, Oslo, Norway). In the APTT-based assay, a constant concentration of ellagic acid (Sigma-Aldrich, St. Louis, Missouri, USA) was used as activator. In the RVVT-based test, RVV (SigmaAldrich) activates factor X directly. The 99th percentile of the LR of the control group was chosen as the upper reference limit, and was 1.22 for the APTT-based LR test and 1.19 for the RVVT-based LR test. aCL IgG and IgM isotypes were analyzed with an in-house enzyme-linked immunosorbent assay (ELISA) essentially as described by Gharavi et al. [22]. We used serial dilutions of an in-house control drawn from a strongly aCL positive patient, which were standardized against Harris’ commercial standards (American Diagnostica Inc., Stamford, CT, USA). Values for IgG and IgM isotypes of aCL were reported in GPL units and MPL units, respectively. The cut-off values for a positive test were defined by the 99th percentile of the values of the control group, and were 10.7 for aCL IgG and 23.7 for aCL IgM. Anti-ß2GP1 IgG and IgM isotypes were assayed with commercial ELISA kits (QUANTA Lite TM β2 GP1 IgG/IgM, INOVA Diagnostics Inc., San Diego, USA) for semi-quantitative determination. Results were expressed in standard IgG and IgM anti-ß2GP1 units, that is, SGU and SMU, respectively. The 99th percentiles of the control group were used as upper reference limits, and were 6.5 for anti-ß2GP1 IgG and 30.3 for anti-ß2GP1 IgM. Statistical analyses Data were analyzed by Chi-square tests or Fisher's exact tests. Results were presented as percentages and odds ratios (OR) with 95% confidence intervals (CI). In the case of missing values for sociodemographic or clinical variables women were denoted the reference group in that particular analysis. Significance level was set at p b 0.05. All data was analyzed using the Statistical Package for Social Science version 16.0 (SPSS Inc, Chicago, Il, USA). Results Information from medical records of the eligible participants gave us the opportunity to compare participating women with the nonparticipating. There were no significant differences in the demographic and clinical data between the participating and nonparticipating women (data not shown). The maternal characteristics of cases and controls are displayed in Table 1. At the time of the
index pregnancy women with IUFD were younger, smoked more often at first visit, had more often placental abruption, their fetuses were more often small for gestational age and they had more often inherited thrombophilia. Prevalences of demographic and clinical risk factors were not significantly different between APA-positive and APA-negative women at the time of the index pregnancy (Table 2). At the time of blood sampling there was not a significant difference in the rate of miscarriages ≤22 gestational weeks between cases and controls, but the cases had more frequently had placental abruption (Table 3). The APA-positive women had more often a history of recurrent miscarriages compared to APA-negative women (Table 3). Mean time difference between index pregnancy and blood sampling were 8.7 years and 8.6 years for APA positive and APA negative women, respectively. The prevalence of APAs and the association of APAs with IUFD are shown in Table 4. Twenty-three women, 10 (9.5%) with a history of IUFD and 13 (5.0%) controls, were positive for at least one APA test (OR 2.0; 95% CI 0.9-4.8). The OR for multiple positivity for APAs among women with IUFD was 7.9 (95% CI 0.8-76.5) as compared to controls, using APA-negativity as a reference, were as the OR for having a single positive test was 1.5 (95% CI 0.6-4.0). Women with more than one positive APA test were all positive for LA. LA was significantly more often positive among women with a history of IUFD, compared to women with live births only (OR 4.3; 95% CI 1.0-18.4). Exploring the LA positive women by single/multiple positivity for APA we found that the risk related to LA was confined to women positive for LA in combination with other APAs. Three cases but no controls were positive for other APA tests in addition to LA. Anti-ß2GP1, both IgG and IgM, and aCL IgG and IgM were not significantly associated with a history of IUFD. Nineteen women had one APA test positive, that is 7 (6.7%) cases and 12 (4.6%) controls. One case and one control were positive for two tests, and one case each was positive for three and four tests.
Table 1 Maternal characteristics at index pregnancy. Variable
Age (at index pregnancy) b35 years ≥35 years Parity 0 1 ≥2 Multiple pregnancy Hypertensive disorders (HD) Preeclampsia Hypertension Small for gestational age (SGA) HD and/or SGA No HD or SGA HD, no SGA HD with SGA SGA, no HD Diabetes Placental abruption Placenta previa Smoking (at first visit) Inherited thrombophilia*
Cases
Controls
N = 105**
N = 262
p-value
%
%
80.0 20.0
68.3 31.7
n.s. b0.03
52.4 34.3 13.3 3.8
50.4 39.3 10.3 2.7
n.s. n.s.
5.7 5.7 34.3
7.3 4.6 1.5
59.0 6.7 4.8 29.5 1.0 8.6 2.9 27.6 18.4
87.8 10.7 1.1 0.4 0.4 0.8 0.8 8.4 11.8
n.s. n.s. b0.001 Reference n.s b0.02 b0.001 n.s. b0.001 n.s. b0.001 n.s.
p –value - represents results from univariate analysis. n.s.: not significant. *Inherited thrombophilia: factor V Leiden, prothrombin G20210A polymorphism, antithrombin, protein C- or protein S deficiency. **Two women using warfarin were excluded from analyses involving protein C and protein S.
L.B. Helgadottir et al. / Thrombosis Research 130 (2012) 32–37 Table 2 Demographic and clinical risk factors among antiphospholipid antibody (APA) positive and APA negative women. Variable
Cases Controls Age (at index pregnancy) b 35 years ≥35 years Parity 0 1 ≥2 Multiple pregnancy Hypertensive disorders (HD) Preeclampsia Hypertension Small for gestational age (SGA) HD and/or SGA No HD or SGA HD, no SGA HD with SGA SGA, no HD Diabetes Placental abruption Placenta previa Smoking (at first visit) Inherited thrombophilia*
APA positive
APA negative
N = 23**
N = 344**
%
%
p-value
43.5 56.5
27.6 72.4
n.s. n.s.
78.3 21.7
71.2 28.8
n.s. n.s.
43.5 34.8 21.7 0
51.5 38.1 10.5 3.2
n.s. n.s.
4.3 8.7 17.4
7.0 4.7 10.5
n.s. n.s. n.s.
73.9 8.7 4.3 13.0 0 0 0 21.7 4.5
79.9 9.6 2.0 8.4 0.6 3.2 1.5 13.4 14.3
Reference n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.
p –value - represents results from univariate analysis. n.s.: not significant. *Inherited thrombophilia: factor V Leiden, prothrombin G20210A polymorphism, antithrombin-, protein C- or protein S deficiency. **Two women, one in each group, using warfarin were excluded from analyses involving protein C and protein S.
Discussion In the present study, we found that women with a history of IUFD, as compared with women with live births only, were significantly more often positive for LA 3–18 years after the index pregnancy. This was attributed to women positive for LA in combination with other APAs, although this study does not allow firm conclusions on the importance of multi-positivity. Our finding of increased prevalence of LA among women with a history of IUFD is in agreement with other studies. However, most investigators, as summarized in a systematic review [9], have found aCL to be an even stronger predictor of IUFD than LA [13–15,23]. Gonen et al. studied the prevalence of both inherited and acquired thrombophilia among women with unexplained IUFD after 26 gestational weeks and found LA and aCL to be significantly more prevalent among women with a history of IUFD compared to controls, with OR 6.1 (95% CI 1.4-36.2) and 8.5 (95% CI 1.6-42.3), respectively [13]. The NOHA-study found, in univariate analysis, LA as well as aCL IgG and anti-ß2GP1 IgG to be associated with IUFD after 22 weeks of pregnancy. However, in the multivariate analysis, adjusting for other APAs and inherited thrombophilia, LA was protective of IUFD, but both aCL IgG and anti-β2GP1 IgG were still positive predictors for IUFD [15]. The cause was found to be that no cases, but two controls, had LA as
35
Table 4 Prevalence of APA and OR with 95% CI for APA among women with IUFD compared to controls. Variable
APA positive – ≥ 1 tests APA positive – 1 test APA positive – >1 tests LA positive – ≥ 1 test LA positive + anti-ß2GP1/aCL negative LA positive + anti-ß2GP1/aCL positive LR-APTT positive LR- RVVT positive Anti- ß2GP1 positive Unique anti- ß2GP1 positivity Anti- ß2GP1 + ≥1 LA/aCL positive test aCL positive Unique aCL positivity aCL + ≥1 LA/ anti- ß2GP1 positive test
Cases (N = 105)
Controls (N = 262)
%
(n)
%
(n)
9.5 6.7 2.9 4.8 1.9 2.9 4.8 1.9 4.8 2.9 1.9 3.8 1.9 1.9
(10) (7) (3) (5) (2) (3) (5) (2) (5) (3) (2) (4) (2) (2)
5.0 4.6 0.4 1.1 1.1 0 0.8 0.8 1.5 1.5 0 2.3 2.3 0
(13) (12) (1) (3) (3) 0 (2) (2) (4) (4) 0 (6) (6) 0
OR
95% CI
2.0 1.5 7.9 4.3 1.7 6.5 2.5 3.2 1.9 1.7 0.8 -
0.9-4.8 0.6-4.0 0.8-76.5 1.0-18.4 0.3-10.5 1.2-34.0 0.4-18.2 0.8-12.3 0.4-8.8 0.5-6.1 0.2-4.3 -
APA: antiphospholipid antibodies, OR: odds ratio, CI: confidence interval, IUFD: intrauterine fetal death, LA: lupus anticoagulant, LR: lupus ratio, APTT: activated partial thromboplastin time-based, RVVT: Russell viper venom test-based, aCL: anticardiolipin antibodies, anti-ß2GP1: anti-β2 glycoprotein 1.
a unique APA marker. This is in line with our findings that the risk of IUFD related to LA is mainly attributed to women positive for other APAs as well. In studies on thrombosis, a more consistent association with both arterial and venous thrombosis has been found for LA than for aCL [24]. Results of studies are largely influenced by differences in study design; for instance eligibility criteria, laboratory methods, and cut-off values. In accordance with the recent update on the guidelines for LA detection, we used two LR tests for LA, one based on the APTT, the other based on RVVT [25]. The LR tests are automated tests; integrating screening, mixing and confirmatory procedures. The results are calculated as LA ratios (screen/confirm) and normalized against values obtained with a pooled normal plasma, as recommended by the guidelines [25]. The LR tests have proven high reproducibility and low interlaboratory variation in an international multilaboratory study and where hence chosen for this study [21]. The majority of studies on APA and IUFD have applied various exclusion criteria in the attempt to investigate women with unexplained IUFD only. Thus they have excluded women with known causes of IUFD. In most studies, IUFD explained by congenital malformations, abnormal karyotypes, uterine malformations, fetal hydrops or infections have been excluded. In some studies, women with maternal risk factors for IUFD, like diabetes, preeclampsia, placental abruption or intrauterine fetal growth restriction have also been excluded, as well as women with a history of miscarriages and known thrombophilia. Such differences in study-design make comparison of results difficult. Our cases were unselected related to cause and we did not exclude women with known thrombophilia, miscarriages or known risk factors for IUFD. By this our findings are more applicable to the general obstetric population. In a recent review article, Kist et al. demonstrated that the relationship between adverse pregnancy outcome and thrombophilia was influenced by
Table 3 Obstetrical history, until time of blood sampling. Variable
Cases (N = 105) %
Controls (N = 262) %
p-value
APA-positive (N = 23) %
APA-negative (N = 344) %
p-value
Miscarriage ≤ 22 weeks Recurrent miscarriage (≥3) ≤22 weeks Abruption in any pregnancy
39.0 5.7 11.4
30.9 2.3 1.1
n.s n.s. b0.001
47.8 13.0 0
32.3 2.6 4.4
n.s. 0.03 n.s
p –value - represents results from univariate analysis. n.s.: not significant. APA = antiphospholipid antibodies.
36
L.B. Helgadottir et al. / Thrombosis Research 130 (2012) 32–37
confounding factors like ethnicity, methods of testing, and severity of disease [26]. About 5-7% of healthy pregnant women have been reported to have positive tests for APAs [27]. The prevalence of APAs depends on the definition of a positive test. It is recommended by international consensus to define the cut-off value by the 99th percentile of a control group. Since by definition 1% of the controls will have a positive result for each test, the prevalence of APAs in the healthy population will be influenced by the number of tests for APAs examined. Thus, if six different tests are used, up to 6% of the control/normal population will be APA positive. Although there is some increased relative risk of IUFD associated with APAs, the absolute risk for APA positive women without previous clinical events is low, and the probability of a successful pregnancy outcome is high. Thus, the screening for APAs in an unselected population of pregnant women is not recommended. The 8th Guidelines on Antithrombotic Therapy of the American College of Chest Physicians (ACCP) from 2008 recommend screening for APAs among women with a history of PMPC [28]. Low molecular weight heparin and low dose acetylsalicylic acid are currently recommended for the prevention of recurrent pregnancy loss in women with APAs [28–31]. Current recommendations on treatment of the APA syndrome may have long-term implications. For example, indefinite treatment is recommended after a first venous thrombosis if the patient has tested positive for APA twice with three months interval. This is a more rigorous secondary thromboprophylactic therapy than usually advised, even though little or no data support such recommendations [32]. This study implies that multiple positivity is probably important although the nature of the study does not allow firm conclusion. This is in agreement with other recent studies that have demonstrated that multiple positivity for APAs is more frequently associated with pregnancy complications than single positivity [33,34]. Collection of blood samples only at a single time point and a long time after the index pregnancy are limitations of our study. However, one of the rationales for repeated testing is to avoid false positive tests due to transiently elevated APAs, which is not a concern in the present study, when the samples were collected 3–18 years after the index pregnancy. Moreover, we can not exclude the possibility that some women have either turned negative or turned positive for APAs after the index pregnancy, but there is no reason to believe that this would differ between cases and controls. Remarkably little is known about the sustainability of APAs over time. In one study, Erkan et al. found sustained positivity over time in approximately 75% of tests initially positive, but with a mean follow-up time of only 2.4, 3.5 and 1.0 years for LA, aCL, and anti-β2 GP1, respectively [35]. Another limitation of our study is that only 29% of the identified cases participated, which probably reflects the burden of this serious complication. In spite of this our sample size is relatively large compared to other studies in this field. We had information from medical records from all the eligible participants and we did not find significant differences in sociodemographic and clinical factors between participating and non-participating women. We therefore do not believe that the low participation applies a selection bias, but it does affect the power of the study. We conclude that women with a history of IUFD after 22 gestational weeks were more often LA positive, 3–18 years after the incident, but the risk of IUFD related to LA was confined to women positive for other APAs in addition to LA. However, there is still great uncertainty related to the association of APAs and IUFD and the clinical importance is not easily predicated.
Conflict of Interest No conflicts of interest.
Acknowledgments Financial support was received from the South-Eastern Norway Regional Health Authority Trust, Hamar, Norway, the Oslo University Hospital Ulleval Scientific Trust, Oslo, Norway, and the Research Council of Norway. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
References [1] Miyakis S, Lockshin MD, Atsumi T, Branch DW, Brey RL, Cervera R, et al. International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS). J Thromb Haemost 2006;4:295–306. [2] Lynch A, Marlar R, Murphy J, Davila G, Santos M, Rutledge J, et al. Antiphospholipid antibodies in predicting adverse pregnancy outcome. A prospective study. Ann Intern Med 1994;120:470–5. [3] Kujovich JL. Thrombophilia and pregnancy complications. Am J Obstet Gynecol 2004;191:412–4. [4] Triplett DA, Harris EN. Antiphospholipid antibodies and reproduction. Am J Reprod Immunol 1989;21:123–31. [5] Rodger MA, Paidas M, Claire M, Middeldorp S, Kahn S, Martinelli I, et al. Inherited thrombophilia and pregnancy complications revisited. Obstet Gynecol 2008;112: 320–4. [6] Berg CJ, Atrash HK, Koonin LM, Tucker M. Pregnancy-related mortality in the United States, 1987–1990. Obstet Gynecol 1996;88:161–7. [7] Rand JH, Wu XX, Guller S, Scher J, Andree HA, Lockwood CJ. Antiphospholipid immunoglobulin G antibodies reduce annexin-V levels on syncytiotrophoblast apical membranes and in culture media of placental villi. Am J Obstet Gynecol 1997;177: 918–23. [8] Kupferminc MJ, Eldor A, Steinman N, Many A, Bar-Am A, Jaffa A, et al. Increased frequency of genetic thrombophilia in women with complications of pregnancy. N Engl J Med 1999;340:9–13. [9] Robertson L, Wu O, Langhorne P, Twaddle S, Clark P, Lowe GD, et al. Thrombophilia in pregnancy: a systematic review. Br J Haematol 2006;132:171–96. [10] 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. [11] Lockwood CJ, Romero R, Feinberg RF, Clyne LP, Coster B, Hobbins JC. The prevalence and biologic significance of lupus anticoagulant and anticardiolipin antibodies in a general obstetric population. Am J Obstet Gynecol 1989;161:369–73. [12] Katano K, Aoki A, Sasa H, Ogasawara M, Matsuura E, Yagami Y. beta 2Glycoprotein I-dependent anticardiolipin antibodies as a predictor of adverse pregnancy outcomes in healthy pregnant women. Hum Reprod 1996;11:509–12. [13] Gonen R, Lavi N, Attias D, Schliamser L, Borochowitz Z, Toubi E, et al. Absence of association of inherited thrombophilia with unexplained third-trimester intrauterine fetal death. Am J Obstet Gynecol 2005;192:742–6. [14] Bocciolone L, Meroni P, Parazzini F, Tincani A, Radici E, Tarantini M, et al. Antiphospholipid antibodies and risk of intrauterine late fetal death. Acta Obstet Gynecol Scand 1994;73:389–92. [15] Gris JC, Quere I, Monpeyroux F, Mercier E, Ripart-Neveu S, Tailland ML, et al. Case– control study of the frequency of thrombophilic disorders in couples with late foetal loss and no thrombotic antecedent–the Nimes Obstetricians and Haematologists Study5 (NOHA5). Thromb Haemost 1999;81:891–9. [16] Helgadottir LB, Skjeldestad FE, Jacobsen AF, Sandset PM, Jacobsen EM. Incidence and risk factors of fetal death in Norway: a case–control study. Acta Obstet Gynecol Scand 2011;90:390–7. [17] Jacobsen AF, Skjeldestad FE, Sandset PM. Ante- and postnatal risk factors of venous thrombosis: a hospital-based case–control study. J Thromb Haemost 2008;6:905–12. [18] Bergrem A, Jacobsen EM, Skjeldestad FE, Jacobsen AF, Skogstad M, Sandset PM. The association of antiphospholipid antibodies with pregnancy-related first time venous thrombosis - a population-based case–control study. Thromb Res 2010;125:e222–7. [19] Jacobsen AF, Dahm A, Bergrem A, Jacobsen EM, Sandset PM. Risk of venous thrombosis in pregnancy among carriers of the factor V Leiden and the prothrombin gene G20210A polymorphisms. J Thromb Haemost 2010;8:2443–9. [20] Schjetlein R, Sletnes KE, Wisloff F. A quantitative, semi-automated and computerassisted test for lupus anticoagulant. Thromb Res 1993;69:239–50. [21] Jacobsen EM, Barna-Cler L, Taylor JM, Triplett DA, Wisloff F. The Lupus Ratio test– an interlaboratory study on the detection of Lupus anticoagulants by an APTTbased, integrated, and semi-quantitative test. Fifth International Survey of Lupus Anticoagulants–ISLA 5. Thromb Haemost 2000;83:704–8. [22] Gharavi AE, Harris EN, Asherson RA, Hughes GR. Anticardiolipin antibodies: isotype distribution and phospholipid specificity. Ann Rheum Dis 1987;46:1–6. [23] Alfirevic Z, Roberts D, Martlew V. How strong is the association between maternal thrombophilia and adverse pregnancy outcome? A systematic review. Eur J Obstet Gynecol Reprod Biol 2002;101:6–14. [24] Galli M, Luciani D, Bertolini G, Barbui T. Lupus anticoagulants are stronger risk factors for thrombosis than anticardiolipin antibodies in the antiphospholipid syndrome: a systematic review of the literature. Blood 2003;101: 1827–32.
L.B. Helgadottir et al. / Thrombosis Research 130 (2012) 32–37 [25] Pengo V, Tripodi A, Reber G, Rand JH, Ortel TL, Galli M, et al. Subcommittee on Lupus Anticoagulant/Antiphospholipid Antibody of the Scientific and Standardisation Committee of the International Society on Thrombosis and Haemostasis. Update of the guidelines for lupus anticoagulant detection. J Thromb Haemost 2009;7:1737–40. [26] Kist WJ, Janssen NG, Kalk JJ, Hague WM, Dekker GA, de Vries JI. Thrombophilias and adverse pregnancy outcome - A confounded problem! Thromb Haemost 2008;99:77–85. [27] Parke AL, Wilson D, Maier D. The prevalence of antiphospholipid antibodies in women with recurrent spontaneous abortion, women with successful pregnancies, and women who have never been pregnant. Arthritis Rheum 1991;34:1231–5. [28] Bates SM, Greer IA, Pabinger I, Sofaer S, Hirsh J, American College of Chest Physicians. Venous thromboembolism, thrombophilia, antithrombotic therapy, and pregnancy: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008;133(Suppl.-886S). [29] Rai R, Cohen H, Dave M, Regan L. Randomised controlled trial of aspirin and aspirin plus heparin in pregnant women with recurrent miscarriage associated with phospholipid antibodies (or antiphospholipid antibodies). BMJ 1997;314:253–7.
37
[30] Brenner B, Bar J, Ellis M, Yarom I, Yohai D, Samueloff A. Live-Enox Investigators. Effects of enoxaparin on late pregnancy complications and neonatal outcome in women with recurrent pregnancy loss and thrombophilia: results from the Live-Enox study. Fertil Steril 2005;84:770–3. [31] Greer IA, Nelson-Piercy C. Low-molecular-weight heparins for thromboprophylaxis and treatment of venous thromboembolism in pregnancy: a systematic review of safety and efficacy. Blood 2005;106:401–7. [32] Derksen RH, de Groot PG. Towards evidence-based treatment of thrombotic antiphospholipid syndrome. Lupus 2010;19:470–4. [33] Ruffatti A, Tonello M, Cavazzana A, Bagatella P, Pengo V. Laboratory classification categories and pregnancy outcome in patients with primary antiphospholipid syndrome prescribed antithrombotic therapy. Thromb Res 2009;123:482–7. [34] Tripodi A, de Groot PG, Pengo V. Antiphospholipid syndrome: laboratory detection, mechanisms of action and treatment. J Intern Med 2011;270:110–22. [35] Erkan D, Derksen WJ, Kaplan V, Sammaritano L, Pierangeli SS, Roubey R, et al. Real world experience with antiphospholipid antibody tests: how stable are results over time? Ann Rheum Dis 2005;64:1321–5.