http://informahealthcare.com/jmf ISSN: 1476-7058 (print), 1476-4954 (electronic) J Matern Fetal Neonatal Med, 2015; 28(4): 443–447 ! 2014 Informa UK Ltd. DOI: 10.3109/14767058.2014.920811
ORIGINAL ARTICLE
Diagnostic value of screening tests in subgroups of women with recurrent pregnancy loss Ali Irfan Guzel1, Selc¸uk Erkılınc¸1, Irfan O¨zer1, Yusuf Celik2, Nafiye Yılmaz1, and Melike Dog˘anay1 1
Department of Obstetrics and Gynecology, Dr. Zekai Tahir Burak Women’s Health Research and Education Hospital, Ankara, Turkey and Department of Biostatistics, Dicle University, Diyarbakır, Turkey
2
Abstract
Keywords
Objective: To evaluate the diagnostic value of screening laboratory tests in women who had recurrent pregnancy loss (RPL). Methods: A total of 252 women with RPL managed in our tertiary referral research and education hospital were included in the study. Risk factors recorded involved age, gravidity, parity, number of prior live births, number of pregnancy losses, and thrombophlia tests. The cases were divided into three different groups and each group was analyzed separately. Results: There was no statistically significant difference between the first and second groups in terms of clinical and laboratory parameters (p40.05). In the third group, there was a statistically significant difference among cases in terms of parity, gravidity, number of pregnancy losses, serum AT III levels, APCR, and age of the women. According to the logistic regression model, odds ratios (95% CI) were 6.116 (3.797–9.852), 5.665 (2.657–12.079), 4.763 (3.099–7.321), 4.729 (3.080–7.260), 2.820 (1.836–4.333), and 1.911 (1.232–2.965), respectively. Conclusions: We do not recommend the screening of all women with RPL, but in women with high parity and those who had prior live birth pregnancies, increased AT III, and APCR may be diagnostic markers for subsequent pregnancy loss.
Diagnostic value, recurrent pregnancy loss, screening tests, subgroup
Introduction Recurrent pregnancy loss (RPL) is defined as three or more consecutive losses of clinically recognized intrauterine pregnancies before 20 weeks of gestation [1]. The incidence of women experiencing one clinically detected pregnancy loss is 15%. Of this, 2% have two losses and 0.4–1% have more than two [2]. In the recent years of obstetrics practice, the etiology of RPL has been a challenging topic that has been widely investigated. The Practice Committee of the American Society for Reproductive Medicine listed the etiological factors as the following: genetics, age of the woman, antiphospolipid anticor syndrome, abnormalities of the uterus, thrombophilia, metabolic and hormonal disorders, and infection [3]. Of these factors, the association between thrombophilia and RPL had been studied by previous reports [4–6]. Thrombophilia is defined as a hypercoagulable state that may create a tendency for blood clotting to occur.
Address for correspondence: Dr Ali Irfan Guzel, Division of Infertility and Gynecological Endocrinology, Department of Obstetrics and Gynecology, Dr. Zekai Tahir Burak Women’s Health Research and Education Hospital, Ankara, Turkey. Tel: + 90 532 293 71 31. Fax: + 90 312 306 59 17. E-mail: [email protected]
History Received 11 November 2013 Revised 14 January 2014 Accepted 30 April 2014 Published online 29 May 2014
Thrombophilia may be classified as acquired (smoking, obesity, fractures, surgery, and immobility) or inherited. These hypercoagulable states have been reported to include pregnancy, puerperium, exposure to hormonal contraception, and hormonal replacement therapy. Some conditions like RPL, stillbirth, placental abruption, intrauterine growth restriction, and severe pre-eclampsia are quasithrombotic events related to pregnancy [7]. Screening for these quasithrombotic events includes APCR, factor V leiden and II mutation, elevated homocysteine level, elevated Factor VIII levels and lupus anticoagulants with high priority, protein C and protein S activities, decreased antithrombin activity and increased anticardiolipin antibodies with intermediate priority and dysfibrinogenemia, elevated fibrinogen level, increased activity of factors IX and XI, and 5,10-methylenetetrahydrofolate reductase activity with low priority [8]. In Turkey, as a common trend, clinicians tend to evaluate the thrombophilia panel (prothrombin time (PT), activated partial thromboplastin time (APTT), international normalized ratio (INR), protein C and protein S deficiencies, fibrinogen levels, antithrombin III, anticardiolipin IgG and IgM antibody levels, and activated protein C resistance (APCR)) even if a patient has one or two pregnancy losses. We believe that this is an inaccurate approach. From this point in the current research, we conducted a study to evaluate the diagnostic
444
A. I. Guzel et al.
value of screening tests performed in our hospital in women with RPL and to assess whether these tests should be evaluated in every woman who had a pregnancy loss.
Methods This retrospective study was performed between January 2008 and December 2012 at Dr. Zekai Tahir Burak Women’s Health Research and Education Hospital, Ankara, Turkey, Department of Obstetrics and Gynecology Division of Infertility and Gynecological Endocrinology, a referral medical centre in the middle region of Turkey. During the study period, 252 women who had two or more RPL were included in the study. Patients who had anatomic uterine anomalies and abortions due to toxoplasmosis, rubella, cytomegalovirus, and herpes (TORCH) simplex virus infections were excluded from the study. All cases were Turkish women with no history of alcohol or drug use; eight women reported tobacco use throughout pregnancy. The data were collected from hospital records and patient files. Subject characteristics and demographics were analysed descriptively. The study was approved by the Ethics Committee of Dr. Zekai Tahir Burak Women’s Health Research and Education Hospital. Demographic characteristics recorded included age, gravidity, parity, number of pregnancies, and number of prior live births of the patients. After taking a complete history, including an obstetrical and gynecological history, we performed laboratory tests, such as prothrombin time (PT), activated partial thromboplastin time (APTT), international normalized ratio (INR), protein C and protein S deficiencies, fibrinogen levels, antithrombin III, anticardiolipin IgG and IgM antibody levels, and activated protein C resistance (APCR), in order to detect a thrombophilic tendency in our patients. We performed multiple analyses by dividing the cases into different groups to investigate the role of the mentioned parameters above as diagnostic screening tests on the etiology of RPL. The study population was divided into three groups according to the number of pregnancy loss, the age, and the number of prior live births. Means and standard deviations (SD) were calculated for continuous variables. Subject characteristics and demographics were analysed descriptively. The normal distribution of the variables was analysed using the Kolmogorov–Smirnov test. The Chi-square ( 2) test with Yates’ correction and Student’s t test were used to evaluate the associations between the categorical and continuous variables. The logistic regression method was used to find the risk variables by including all variables in the model and to calculate the odds ratios. Receiver operating characteristic curve (ROC curve) analysis was performed to calculate the cut-off valves of the risk factors. All variables were included in the backward stepwise procedure. Two-sided p values were considered statistically significant at p50.05. Statistical analyses were carried out using the statistical package SPSS 15.0 for Windows (SPSS Inc., Chicago, IL).
Results A total of 252 women with RPL were managed during the study period in our clinic and were included in the current
J Matern Fetal Neonatal Med, 2015; 28(4): 443–447
study. Table 1 shows the main characteristics and previous obstetric histories of the study participants. The mean age of the cases was 28.28 ± 5.72 years. The mean numbers of gravidity, parity, number of aborts, and number of prior live births were 3.72 ± 1.83, 0.54 ± 0.76, 2.99 ± 0.92, and 0.46 ± 0.73, respectively. The mean levels of the anticardiolipine antibody of IgG and IgM were 4.47 ± 0.31 GPLU/ml and 9.73 ± 3.27 MPLU/ml. APTT levels were 42.49 ± 12.19 s, PTT 11.74 ± 0.74 s, and INR 0.99 ± 0.04. Mean antitrombine III activity was 104.78 ± 18.17% while fibrinogen levels were at 334.11 ± 75.49 mg/dl. Protein C and S deficiencies were at 88.06 ± 22.15 and 105.92 ± 70.17%, respectively. Approximately 74.20% of our cases were negative for APCR. Table 2 depicts the characteristics of the groups classified according to the number of pregnancy losses. The result of cases with 52 pregnancy loss (n ¼ 72) and more than two (n ¼ 180) was as follows: anticardiolipine IgG 7.62 ± 2.45 versus 10.01 ± 4.16 GPLU/ml and IgM 4.76 ± 0.69 versus 4.22 ± 0.29 MPLU/ml, APTT 29.98 ± 3.21 versus 46.37 ± 211.38 s, PTT 11.83 ± 0.90 versus 11.66 ± 1.17 s, INR 0.9 ± 0.09 versus 0.99 ± 0.71, AT III 111.91 ± 16.50 versus 105.66 ± 15.60, serum fibrinogen levels 323.45 ± 50.56 versus 333.94 ± 75.66 mg/dl, Pro S deficiency 84.18 ± 11.69 versus 89.02 ± 22.47%, Pro C deficiency Table 1. The main characteristics and previous obstetric history of the study participants.
Age (years) Gravidy* Parity* Number of pregnancy loss* Number of living prior live birth* Anticardiolipin antibody of IgG Anticardiolipin antibody of IgM APTT PTT INR Antitrombine III activity Fibrinogen levels Protein S deficiency Protein C deficiency APCR (negative %)
n ¼ 252
SD
Range
28.28 3 (2–8) 0 (0–3) 3 (2–6) 0 (0–3) 4.47 9.73 42.49 11.74 0.99 104.78 334.11 88.06 105.92 74.2
5.72
27.53–29.01
0.31 3.27 12.19 0.74 0.04 18.17 75.49 22.15 70.17
3.86–5.08 3.27–16.18 25–38 10–12 0.82–1.11 75–125 140–350 59–118 70–140
*Data were calculated as median values (min–max values). Table 2. The characteristics of the groups classified according to the number of pregnancy losses.
Anticardiolipin antibody of IgG Anticardiolipin antibody of IgM APTT PTT INR Antitrombine III activity Fibrinogen levels Protein S deficiency Protein C deficiency APCR
No. pregnancy loss 2 (n ¼ 72)
No. pregnancy loss 3 (n ¼ 180)
p
7.62 ± 2.45
10.01 ± 4.16
0.648
4.76 ± 0.69
4.22 ± 0.29
0.082
29.98 ± 3.21 11.83 ± 0.90 0.9 ± 0.09 111.91 ± 16.50 323.45 ± 50.56 84.18 ± 11.69 90.91 ± 23.35 0.211 ± 0.41
46.37 ± 211.38 11.66 ± 1.17 0.99 ± 0.71 105.66 ± 15.60 333.94 ± 75.66 89.02 ± 22.47 106.57 ± 68.79 0.29 ± 0.45
0.512 0.292 0.913 0.373 0.146 0.414 0.922 0.337
Screening tests in women with recurrent pregnancy loss
DOI: 10.3109/14767058.2014.920811
90.91 ± 23.35 versus 106.57 ± 68.79%, and APCR was 0.211 ± 0.41 versus 0.29 ± 0.45. There was no statistically significant difference between the groups according to the number of pregnancy losses (p40.05). Cases were also divided into two groups according to age. Table 3 shows the parameters of these groups. The first group included cases 35 years while the other group included cases 435 years. The mean values of the parameters were as follows: anticardiolipine IgG 9.51 ± 0.31 versus 8.51 ± 2.55 GPLU/ml and IgM 4.38 ± 0.30 versus 4.11 ± 0.60 MPLU/ml, APTT 42.53 ± 2.35 versus 35.42 ± 2.03 s, PT 11.73 ± 0.07 versus 11.51 ± 0.17 s, INR 1.00 ± 0.51 versus 0.95 ± 0.10, mean AT III activity 105.97 ± 10.80 versus 104.72 ± 13.18%, serum fibrinogen levels 331.69 ± 15.04 versus 344.17 ± 13.18 mg/dl, Pro S deficiency 88.18 ± 1.45 versus 92.56 ± 4.39%, Pro C deficiency 105.93 ± 4.94 versus 105.01 ± 3.57%, and APCR was 0.29 ± 0.03 versus 0.16 ± 0.07. There was no statistically significant difference between the groups according to the age of the patients (p40.05). Table 4 summarizes the characteristics of the cases classified into two groups according to the number of prior live births. The women in these groups had at least one prior live birth and had two or more pregnancy losses after. The mean values of the parameters were as follows: anticardiolipine IgG 10.24 ± 4.32 versus 7.39 ± 2.37 GPLU/ml and IgM 4.44 ± 0.38 versus 4.23 ± 0.39 MPLU/ml, APTT Table 3. The characteristics of the groups classified according to the age of the cases.
Anticardiolipin antibody of IgG Anticardiolipin antibody of IgM APTT PTT INR Antitrombine III activity Fibrinogen levels Protein S deficiency Protein C deficiency APCR
Cases 35 years old (n ¼ 227)
Cases 435 years old (n ¼ 25)
p
9.51 ± 0.31
8.51 ± 2.55
0.562
4.38 ± 0.30
4.11 ± 0.60
0.472
42.53 ± 2.35 11.73 ± 0.07 1.00 ± 0.51 105.97 ± 10.80 331.69 ± 15.04 105.93 ± 4.94 105.93 ± 4.94 0.29 ± 0.03
35.42 ± 2.03 11.51 ± 0.17 0.95 ± 0.10 104.72 ± 13.18 344.17 ± 13.18 105.01 ± 3.57 105.01 ± 3.57 0.16 ± 0.07
0.786 0.354 0.354 0.748 0.557 0.94 0.113 0.707
Table 4. The characteristics of the groups classified according to prior live birth pregnancies.
Anticardiolipin antibody of IgG Anticardiolipin antibody of IgM APTT PTT INR Antitrombine III activity Fibrinogen levels Protein S deficiency Protein C deficiency APCR
Cases with prior live birth (n ¼ 170)
Cases with no prior live birth (n ¼ 82)
10.24 ± 4.32
7.39 ± 2.37
0.439
4.44 ± 0.38
4.23 ± 0.39
0.472
p
30.11 ± 0.26 versus 66.93 ± 317.44 s, PT 11.79 ± 0.68 versus 11.57 ± 0.16 s, INR 0.95 ± 0.06 versus 1.07 ± 0.13, mean AT III activity 102.79 ± 1.26 versus 112.54 ± 1.26%, serum fibrinogen levels 330.62 ± 5.69 versus 339.47 ± 8.41 mg/dl, Pro S deficiency 85.89 ± 1.63 versus 90.17 ± 2.59%, Pro C deficiency 108.64 ± 6.17 versus 100.10 ± 2.51%, and APCR was 0.29 ± 0.03 versus 0.30 ± 0.04. In this group, there was a statistically significant difference between groups in terms of gravidity, parity, number of pregnancy loss, APCR, AT III, and age (p50.05). According to the logistic regression model, the odds ratios (95% CI) for these parameters were 6.116 (3.797–9.852), 5.665 (2.657–12.079), 4.763 (3.099– 7.321), 4.729 (3.080–7.260), 2.820 (1.836–4.333), and 1.911 (1.232–2.965) (Table 5). To find out if these parameters may be diagnostic markers in the third group, we performed ROC analysis. The ROC areas under the curves (AUCs) of these variables are shown in Figure 1. The AUCs (95% CI) for parity, gravidity, number of pregnancy losses, APCR, AT III, and age were 0.966 (0.96–0.987), 0.828 (0775–0.881), 0.740 (0.675–0.805), 0.747 (0.678–0.816), 0.681 (0.613–0.809), and 0.740 (0.671–0.809), respectively.
Discussion We conducted an analysis to evaluate the diagnostic value of screening laboratory tests among women with RPL. We also analysed the cases in three different groups according to the following: number of pregnancy losses, age, and number of prior live birth pregnancies. Two hundred fifty-two women with RPL were evaluated for demographic features and laboratory findings for thrombophilias. When we classified the cases according to the number of pregnancy losses (cases with 2 pregnancy loss and 3) and age (cases 35 years old and 435 years old), we did not find any statistically significant difference among groups in terms of laboratory parameters (Tables 2 and 3). The group classified according to having previous live births had different values in terms of the number of gravidity, parity, number of pregnancy losses, APCR, AT III, and age. In addition, the ROC curve AUCs revealed that these parameters may be diagnostic in this group. According to this study, women who had at least one previous live birth pregnancy and subsequent pregnancy losses should be screened by APCR and AT III. It is known that too many diagnostic tests are performed in women with RPL and there is a strong debate whether diagnostic tests should be performed in all women with this
Table 5. Odds ratios of the risk of the cases in group three by logistic regression method.
30.11 ± 0.26 11.79 ± 0.68 0.95 ± 0.06 102.79 ± 1.26 330.62 ± 5.69 85.89 ± 1.63 108.64 ± 6.17 0.29 ± 0.03
66.93 ± 317.44 11.57 ± 0.16 1.07 ± 0.13 112.54 ± 1.26 339.47 ± 8.41 90.17 ± 2.59 100.10 ± 2.51 0.30 ± 0.04
0.117 0.394 0.259 50.001 0.382 0.427 0.35 50.001
445
Gravidity Parity No. of pregnancy loss APCR AT III Age (years)
S.E.
Wald
p
Odds ratios
95 % CI
1811 243 1734 386 1561 219
55 437 50.001 20 156 50.001 50 651 50.001
6116 5665 4763
3797–9852 2657–12 079 3099–7321
1554 219 1037 219 647 224
50 433 50.001 22 404 50.001 8347 0.004
4729 2820 1911
3080–7260 1836–4333 1232–2965
446
A. I. Guzel et al.
J Matern Fetal Neonatal Med, 2015; 28(4): 443–447
Figure 1. ROC curve for detecting risk variables in cases of RPL in groups classified as having prior live birth pregnancies.
situation. Franssen et al. [9] designed a survey study and reported that too many diagnostic tests and ineffective therapeutic interventions were performed in women with recurrent miscarriage. They argued that it was important to test only the subgroups of RPL with high risk for underlying abnormality. Similar to the results of Franssen et al., clinicians in our country perform too many unnecessary tests on these women. Referring to the results of our study, we also recommend these tests to the subgroups of women who had RPL. Jaslow et al. [10] reported that investigative diagnostic tests did not differ among women with different numbers of pregnancy losses (women with two, three, or 4 pregnancy losses). We have also found no difference among women with two or more pregnancy losses. Marquard et al. [11] designed a study on women aged 435 years old with 3 first trimester miscarriages and showed that there was no association between thrombophilia workup (including anticardiolipin antibodies, lupus anticoagulant, proteins C and S, prothrombin gene, antithrombin III, factor V Leiden, and methylenetetrahydrofolate reductase) and risk of miscarriage. Our study was similar to this study and we also did not find any association between thrombophilia panel in women who were divided into groups according to the age and the number of pregnancy losses.
Van den Boogaard et al. [12] also designed a retrospective study meant to evaluate the relationship between the number and sequence of preceding miscarriages and antiphospholipid syndrome. They found no difference in the women in terms of obstetric history and age of the women. In our study, in the third group classified according to having previous live birth pregnancies, women having previous live birth pregnancies had statistically different numbers of gravidity, parity, number of pregnancy losses, APCR, AT III, and age. These results were different from the findings of Van den Boogaard et al. [12]. These difference and interesting data are the additive point of our study to the literature. In another study, Metwally et al. [13] analyzed the effect of body mass index and the risk of miscarriage in women with RPL. They found that obesity increased the risk of RPL with a smaller but statistically significant difference. According to this study, the most important risk factor was the age of the women. In our study, age was not a single risk factor for RPL in women who had a previous live birth. Yildiz et al. [14] evaluated the role of Factor V Leiden (G1691A), prothrombin G20210A or methylenetetrahydrofolate reductase (MTHFR) C677T gene mutations in women with RPL and found that thrombophilia screening was not recommended in Turkish women with low risk of VTE.
DOI: 10.3109/14767058.2014.920811
Marquard et al. [11] determined the etiological factors of RPL in women aged 435 years old in a retrospective cohort study. They found that the most common etiological factor was fetal chromosomal abnormalities. About 4 out of 43 patients had antiphospholipid antibody (APA) syndrome and one patient had protein C deficiency. Our limitation in the current study is the absence of foetal chromosomal abnormalities, factor V Leiden, and methylenetetrahydrofolate reductase (MTHFR) gene mutations. Our study supports the findings of Marquard et al., as we also found no statistically significant difference between the groups formed according to the number of pregnancy losses and on the age of the women in terms of thrombophilia workup. There is a debate about the routine screening of all patients with RPL for inherited thrombophilias. The American Society for Reproductive Medicine (ASRM) recommends the screening of all women with RPL for lupus anticoagulant, anticardiolipin antibodies, and anti-b2 glycoprotein I. Stephenson [2] also recommends screening the same parameters. Laurino et al. [15] also recommend screening tests in women with RPL for antiphospholipid syndrome. McNamee et al. [16] reported that screening should be performed with limited parameters in women with high risks of thrombophilia. Blickstein et al. [8] recommended screening the high-risk cases for thrombophilias. We also found that screening is not necessary for all women with RPL but that subgroups, especially women with prior live birth pregnancies should undergo routine screening in terms of AT III and APCR. Our ROC curve results support this finding. In a case–control study, Hossain et al. [17] designed a study similar to ours and found no association between inherited thrombophilias and RPL in Pakistani women. Our results support this study, only it differs slightly in terms of the evaluation of subgroups. In conclusion, there is a strong association between antiphospholipid syndrome and RPL and screening may be useful in patients who are at high risk of VTE. We also do not recommend the screening of all cases. In women with high parity and those who had prior live birth pregnancies, although increased AT III and APCR may be a diagnostic marker for subsequent pregnancy losses.
Declaration of interest The authors declare that they have no conflict of interest.
Screening tests in women with recurrent pregnancy loss
447
References 1. Practice Committee of the American Society for Reproductive Medicine. Definitions of infertility and recurrent pregnancy loss. Fertil Steril 2008;89:1603. 2. Stephenson M, Kutteh W. Evaluation and management of recurrent early pregnancy loss. Clin Obstet Gynecol 2007;50: 132–45. 3. Practice Committee of the American Society for Reproductive Medicine. Evaluation and treatment of recurrent pregnancy loss: a committee opinion. Fertil Steril 2012;98:1103–11. 4. 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. 5. Preston FE, Rosendaal FR, Walker ID, et al. Increased fetal loss in women with heritable thrombophilia. Lancet 1996;348: 913–16. 6. Gris JC, Que´re´ I, Monpeyroux F, et al. Case-control study of the frequency of thrombophilic disorders in couples with late foetal loss and no thrombotic antecedent – the Nıˆmes Obstetricians and Haematologists Study5 (NOHA5). Thromb Haemost 1999;81: 891–9. 7. Blickstein I. Thrombophilia and women’s health: an overview. Obstet Gynecol Clin North Am 2006;33:347–56. 8. Blickstein D. Screening for thrombophilia. Obstet Gynecol Clin North Am 2006;33:389–95. 9. Franssen MT, Korevaar JC, van der Veen F, et al. Management of recurrent miscarriage: evaluating the impact of a guideline. Hum Reprod 2007;22:1298–303. 10. Jaslow CR, Carney JL, Kutteh WH. Diagnostic factors identified in 1020 women with two versus three or more recurrent pregnancy losses. Fertil Steril 2010;93:1234–43. 11. Marquard K, Westphal LM, Milki AA, Lathi RB. Etiology of recurrent pregnancy loss in women over the age of 35 years. Fertil Steril 2010;94:1473–7. 12. van den Boogaard E, Cohn DM, Korevaar JC, et al. Number and sequence of preceding miscarriages and maternal age for the prediction of antiphospholipid syndrome in women with recurrent miscarriage. Fertil Steril 2013;99:188–92. 13. Metwally M, Saravelos SH, Ledger WL, Li TC. Body mass index and risk of miscarriage in women with recurrent miscarriage. Fertil Steril 2010;94:290–5. 14. Yildiz G, Yavuzcan A, Yildiz P, et al. Inherited thrombophilia with recurrent pregnancy loss in Turkish women – a real phenomenon? Ginekol Pol 2012;83:598–603. 15. Laurino MY, Bennett RL, Saraiya DS, et al. Genetic evaluation and counseling of couples with recurrent miscarriage: recommendations of the National Society of Genetic Counselors. J Genet Couns 2005;14:165–81. 16. McNamee K, Dawood F, Farquharson RG. Thrombophilia and early pregnancy loss. Best Pract Res Clin Obstet Gynaecol 2012;26: 91–102. 17. Hossain N, Shamsi T, Khan N, Naz A. Thrombofilia investigation in Pakistani women with recurrent pregnancy loss. J Obstet Gynaecol Res 2013;39:121–5.
Copyright of Journal of Maternal-Fetal & Neonatal Medicine is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL ARTICLE
Diagnostic value of screening tests in subgroups of women with recurrent pregnancy loss Ali Irfan Guzel1, Selc¸uk Erkılınc¸1, Irfan O¨zer1, Yusuf Celik2, Nafiye Yılmaz1, and Melike Dog˘anay1 1
Department of Obstetrics and Gynecology, Dr. Zekai Tahir Burak Women’s Health Research and Education Hospital, Ankara, Turkey and Department of Biostatistics, Dicle University, Diyarbakır, Turkey
2
Abstract
Keywords
Objective: To evaluate the diagnostic value of screening laboratory tests in women who had recurrent pregnancy loss (RPL). Methods: A total of 252 women with RPL managed in our tertiary referral research and education hospital were included in the study. Risk factors recorded involved age, gravidity, parity, number of prior live births, number of pregnancy losses, and thrombophlia tests. The cases were divided into three different groups and each group was analyzed separately. Results: There was no statistically significant difference between the first and second groups in terms of clinical and laboratory parameters (p40.05). In the third group, there was a statistically significant difference among cases in terms of parity, gravidity, number of pregnancy losses, serum AT III levels, APCR, and age of the women. According to the logistic regression model, odds ratios (95% CI) were 6.116 (3.797–9.852), 5.665 (2.657–12.079), 4.763 (3.099–7.321), 4.729 (3.080–7.260), 2.820 (1.836–4.333), and 1.911 (1.232–2.965), respectively. Conclusions: We do not recommend the screening of all women with RPL, but in women with high parity and those who had prior live birth pregnancies, increased AT III, and APCR may be diagnostic markers for subsequent pregnancy loss.
Diagnostic value, recurrent pregnancy loss, screening tests, subgroup
Introduction Recurrent pregnancy loss (RPL) is defined as three or more consecutive losses of clinically recognized intrauterine pregnancies before 20 weeks of gestation [1]. The incidence of women experiencing one clinically detected pregnancy loss is 15%. Of this, 2% have two losses and 0.4–1% have more than two [2]. In the recent years of obstetrics practice, the etiology of RPL has been a challenging topic that has been widely investigated. The Practice Committee of the American Society for Reproductive Medicine listed the etiological factors as the following: genetics, age of the woman, antiphospolipid anticor syndrome, abnormalities of the uterus, thrombophilia, metabolic and hormonal disorders, and infection [3]. Of these factors, the association between thrombophilia and RPL had been studied by previous reports [4–6]. Thrombophilia is defined as a hypercoagulable state that may create a tendency for blood clotting to occur.
Address for correspondence: Dr Ali Irfan Guzel, Division of Infertility and Gynecological Endocrinology, Department of Obstetrics and Gynecology, Dr. Zekai Tahir Burak Women’s Health Research and Education Hospital, Ankara, Turkey. Tel: + 90 532 293 71 31. Fax: + 90 312 306 59 17. E-mail: [email protected]
History Received 11 November 2013 Revised 14 January 2014 Accepted 30 April 2014 Published online 29 May 2014
Thrombophilia may be classified as acquired (smoking, obesity, fractures, surgery, and immobility) or inherited. These hypercoagulable states have been reported to include pregnancy, puerperium, exposure to hormonal contraception, and hormonal replacement therapy. Some conditions like RPL, stillbirth, placental abruption, intrauterine growth restriction, and severe pre-eclampsia are quasithrombotic events related to pregnancy [7]. Screening for these quasithrombotic events includes APCR, factor V leiden and II mutation, elevated homocysteine level, elevated Factor VIII levels and lupus anticoagulants with high priority, protein C and protein S activities, decreased antithrombin activity and increased anticardiolipin antibodies with intermediate priority and dysfibrinogenemia, elevated fibrinogen level, increased activity of factors IX and XI, and 5,10-methylenetetrahydrofolate reductase activity with low priority [8]. In Turkey, as a common trend, clinicians tend to evaluate the thrombophilia panel (prothrombin time (PT), activated partial thromboplastin time (APTT), international normalized ratio (INR), protein C and protein S deficiencies, fibrinogen levels, antithrombin III, anticardiolipin IgG and IgM antibody levels, and activated protein C resistance (APCR)) even if a patient has one or two pregnancy losses. We believe that this is an inaccurate approach. From this point in the current research, we conducted a study to evaluate the diagnostic
444
A. I. Guzel et al.
value of screening tests performed in our hospital in women with RPL and to assess whether these tests should be evaluated in every woman who had a pregnancy loss.
Methods This retrospective study was performed between January 2008 and December 2012 at Dr. Zekai Tahir Burak Women’s Health Research and Education Hospital, Ankara, Turkey, Department of Obstetrics and Gynecology Division of Infertility and Gynecological Endocrinology, a referral medical centre in the middle region of Turkey. During the study period, 252 women who had two or more RPL were included in the study. Patients who had anatomic uterine anomalies and abortions due to toxoplasmosis, rubella, cytomegalovirus, and herpes (TORCH) simplex virus infections were excluded from the study. All cases were Turkish women with no history of alcohol or drug use; eight women reported tobacco use throughout pregnancy. The data were collected from hospital records and patient files. Subject characteristics and demographics were analysed descriptively. The study was approved by the Ethics Committee of Dr. Zekai Tahir Burak Women’s Health Research and Education Hospital. Demographic characteristics recorded included age, gravidity, parity, number of pregnancies, and number of prior live births of the patients. After taking a complete history, including an obstetrical and gynecological history, we performed laboratory tests, such as prothrombin time (PT), activated partial thromboplastin time (APTT), international normalized ratio (INR), protein C and protein S deficiencies, fibrinogen levels, antithrombin III, anticardiolipin IgG and IgM antibody levels, and activated protein C resistance (APCR), in order to detect a thrombophilic tendency in our patients. We performed multiple analyses by dividing the cases into different groups to investigate the role of the mentioned parameters above as diagnostic screening tests on the etiology of RPL. The study population was divided into three groups according to the number of pregnancy loss, the age, and the number of prior live births. Means and standard deviations (SD) were calculated for continuous variables. Subject characteristics and demographics were analysed descriptively. The normal distribution of the variables was analysed using the Kolmogorov–Smirnov test. The Chi-square ( 2) test with Yates’ correction and Student’s t test were used to evaluate the associations between the categorical and continuous variables. The logistic regression method was used to find the risk variables by including all variables in the model and to calculate the odds ratios. Receiver operating characteristic curve (ROC curve) analysis was performed to calculate the cut-off valves of the risk factors. All variables were included in the backward stepwise procedure. Two-sided p values were considered statistically significant at p50.05. Statistical analyses were carried out using the statistical package SPSS 15.0 for Windows (SPSS Inc., Chicago, IL).
Results A total of 252 women with RPL were managed during the study period in our clinic and were included in the current
J Matern Fetal Neonatal Med, 2015; 28(4): 443–447
study. Table 1 shows the main characteristics and previous obstetric histories of the study participants. The mean age of the cases was 28.28 ± 5.72 years. The mean numbers of gravidity, parity, number of aborts, and number of prior live births were 3.72 ± 1.83, 0.54 ± 0.76, 2.99 ± 0.92, and 0.46 ± 0.73, respectively. The mean levels of the anticardiolipine antibody of IgG and IgM were 4.47 ± 0.31 GPLU/ml and 9.73 ± 3.27 MPLU/ml. APTT levels were 42.49 ± 12.19 s, PTT 11.74 ± 0.74 s, and INR 0.99 ± 0.04. Mean antitrombine III activity was 104.78 ± 18.17% while fibrinogen levels were at 334.11 ± 75.49 mg/dl. Protein C and S deficiencies were at 88.06 ± 22.15 and 105.92 ± 70.17%, respectively. Approximately 74.20% of our cases were negative for APCR. Table 2 depicts the characteristics of the groups classified according to the number of pregnancy losses. The result of cases with 52 pregnancy loss (n ¼ 72) and more than two (n ¼ 180) was as follows: anticardiolipine IgG 7.62 ± 2.45 versus 10.01 ± 4.16 GPLU/ml and IgM 4.76 ± 0.69 versus 4.22 ± 0.29 MPLU/ml, APTT 29.98 ± 3.21 versus 46.37 ± 211.38 s, PTT 11.83 ± 0.90 versus 11.66 ± 1.17 s, INR 0.9 ± 0.09 versus 0.99 ± 0.71, AT III 111.91 ± 16.50 versus 105.66 ± 15.60, serum fibrinogen levels 323.45 ± 50.56 versus 333.94 ± 75.66 mg/dl, Pro S deficiency 84.18 ± 11.69 versus 89.02 ± 22.47%, Pro C deficiency Table 1. The main characteristics and previous obstetric history of the study participants.
Age (years) Gravidy* Parity* Number of pregnancy loss* Number of living prior live birth* Anticardiolipin antibody of IgG Anticardiolipin antibody of IgM APTT PTT INR Antitrombine III activity Fibrinogen levels Protein S deficiency Protein C deficiency APCR (negative %)
n ¼ 252
SD
Range
28.28 3 (2–8) 0 (0–3) 3 (2–6) 0 (0–3) 4.47 9.73 42.49 11.74 0.99 104.78 334.11 88.06 105.92 74.2
5.72
27.53–29.01
0.31 3.27 12.19 0.74 0.04 18.17 75.49 22.15 70.17
3.86–5.08 3.27–16.18 25–38 10–12 0.82–1.11 75–125 140–350 59–118 70–140
*Data were calculated as median values (min–max values). Table 2. The characteristics of the groups classified according to the number of pregnancy losses.
Anticardiolipin antibody of IgG Anticardiolipin antibody of IgM APTT PTT INR Antitrombine III activity Fibrinogen levels Protein S deficiency Protein C deficiency APCR
No. pregnancy loss 2 (n ¼ 72)
No. pregnancy loss 3 (n ¼ 180)
p
7.62 ± 2.45
10.01 ± 4.16
0.648
4.76 ± 0.69
4.22 ± 0.29
0.082
29.98 ± 3.21 11.83 ± 0.90 0.9 ± 0.09 111.91 ± 16.50 323.45 ± 50.56 84.18 ± 11.69 90.91 ± 23.35 0.211 ± 0.41
46.37 ± 211.38 11.66 ± 1.17 0.99 ± 0.71 105.66 ± 15.60 333.94 ± 75.66 89.02 ± 22.47 106.57 ± 68.79 0.29 ± 0.45
0.512 0.292 0.913 0.373 0.146 0.414 0.922 0.337
Screening tests in women with recurrent pregnancy loss
DOI: 10.3109/14767058.2014.920811
90.91 ± 23.35 versus 106.57 ± 68.79%, and APCR was 0.211 ± 0.41 versus 0.29 ± 0.45. There was no statistically significant difference between the groups according to the number of pregnancy losses (p40.05). Cases were also divided into two groups according to age. Table 3 shows the parameters of these groups. The first group included cases 35 years while the other group included cases 435 years. The mean values of the parameters were as follows: anticardiolipine IgG 9.51 ± 0.31 versus 8.51 ± 2.55 GPLU/ml and IgM 4.38 ± 0.30 versus 4.11 ± 0.60 MPLU/ml, APTT 42.53 ± 2.35 versus 35.42 ± 2.03 s, PT 11.73 ± 0.07 versus 11.51 ± 0.17 s, INR 1.00 ± 0.51 versus 0.95 ± 0.10, mean AT III activity 105.97 ± 10.80 versus 104.72 ± 13.18%, serum fibrinogen levels 331.69 ± 15.04 versus 344.17 ± 13.18 mg/dl, Pro S deficiency 88.18 ± 1.45 versus 92.56 ± 4.39%, Pro C deficiency 105.93 ± 4.94 versus 105.01 ± 3.57%, and APCR was 0.29 ± 0.03 versus 0.16 ± 0.07. There was no statistically significant difference between the groups according to the age of the patients (p40.05). Table 4 summarizes the characteristics of the cases classified into two groups according to the number of prior live births. The women in these groups had at least one prior live birth and had two or more pregnancy losses after. The mean values of the parameters were as follows: anticardiolipine IgG 10.24 ± 4.32 versus 7.39 ± 2.37 GPLU/ml and IgM 4.44 ± 0.38 versus 4.23 ± 0.39 MPLU/ml, APTT Table 3. The characteristics of the groups classified according to the age of the cases.
Anticardiolipin antibody of IgG Anticardiolipin antibody of IgM APTT PTT INR Antitrombine III activity Fibrinogen levels Protein S deficiency Protein C deficiency APCR
Cases 35 years old (n ¼ 227)
Cases 435 years old (n ¼ 25)
p
9.51 ± 0.31
8.51 ± 2.55
0.562
4.38 ± 0.30
4.11 ± 0.60
0.472
42.53 ± 2.35 11.73 ± 0.07 1.00 ± 0.51 105.97 ± 10.80 331.69 ± 15.04 105.93 ± 4.94 105.93 ± 4.94 0.29 ± 0.03
35.42 ± 2.03 11.51 ± 0.17 0.95 ± 0.10 104.72 ± 13.18 344.17 ± 13.18 105.01 ± 3.57 105.01 ± 3.57 0.16 ± 0.07
0.786 0.354 0.354 0.748 0.557 0.94 0.113 0.707
Table 4. The characteristics of the groups classified according to prior live birth pregnancies.
Anticardiolipin antibody of IgG Anticardiolipin antibody of IgM APTT PTT INR Antitrombine III activity Fibrinogen levels Protein S deficiency Protein C deficiency APCR
Cases with prior live birth (n ¼ 170)
Cases with no prior live birth (n ¼ 82)
10.24 ± 4.32
7.39 ± 2.37
0.439
4.44 ± 0.38
4.23 ± 0.39
0.472
p
30.11 ± 0.26 versus 66.93 ± 317.44 s, PT 11.79 ± 0.68 versus 11.57 ± 0.16 s, INR 0.95 ± 0.06 versus 1.07 ± 0.13, mean AT III activity 102.79 ± 1.26 versus 112.54 ± 1.26%, serum fibrinogen levels 330.62 ± 5.69 versus 339.47 ± 8.41 mg/dl, Pro S deficiency 85.89 ± 1.63 versus 90.17 ± 2.59%, Pro C deficiency 108.64 ± 6.17 versus 100.10 ± 2.51%, and APCR was 0.29 ± 0.03 versus 0.30 ± 0.04. In this group, there was a statistically significant difference between groups in terms of gravidity, parity, number of pregnancy loss, APCR, AT III, and age (p50.05). According to the logistic regression model, the odds ratios (95% CI) for these parameters were 6.116 (3.797–9.852), 5.665 (2.657–12.079), 4.763 (3.099– 7.321), 4.729 (3.080–7.260), 2.820 (1.836–4.333), and 1.911 (1.232–2.965) (Table 5). To find out if these parameters may be diagnostic markers in the third group, we performed ROC analysis. The ROC areas under the curves (AUCs) of these variables are shown in Figure 1. The AUCs (95% CI) for parity, gravidity, number of pregnancy losses, APCR, AT III, and age were 0.966 (0.96–0.987), 0.828 (0775–0.881), 0.740 (0.675–0.805), 0.747 (0.678–0.816), 0.681 (0.613–0.809), and 0.740 (0.671–0.809), respectively.
Discussion We conducted an analysis to evaluate the diagnostic value of screening laboratory tests among women with RPL. We also analysed the cases in three different groups according to the following: number of pregnancy losses, age, and number of prior live birth pregnancies. Two hundred fifty-two women with RPL were evaluated for demographic features and laboratory findings for thrombophilias. When we classified the cases according to the number of pregnancy losses (cases with 2 pregnancy loss and 3) and age (cases 35 years old and 435 years old), we did not find any statistically significant difference among groups in terms of laboratory parameters (Tables 2 and 3). The group classified according to having previous live births had different values in terms of the number of gravidity, parity, number of pregnancy losses, APCR, AT III, and age. In addition, the ROC curve AUCs revealed that these parameters may be diagnostic in this group. According to this study, women who had at least one previous live birth pregnancy and subsequent pregnancy losses should be screened by APCR and AT III. It is known that too many diagnostic tests are performed in women with RPL and there is a strong debate whether diagnostic tests should be performed in all women with this
Table 5. Odds ratios of the risk of the cases in group three by logistic regression method.
30.11 ± 0.26 11.79 ± 0.68 0.95 ± 0.06 102.79 ± 1.26 330.62 ± 5.69 85.89 ± 1.63 108.64 ± 6.17 0.29 ± 0.03
66.93 ± 317.44 11.57 ± 0.16 1.07 ± 0.13 112.54 ± 1.26 339.47 ± 8.41 90.17 ± 2.59 100.10 ± 2.51 0.30 ± 0.04
0.117 0.394 0.259 50.001 0.382 0.427 0.35 50.001
445
Gravidity Parity No. of pregnancy loss APCR AT III Age (years)
S.E.
Wald
p
Odds ratios
95 % CI
1811 243 1734 386 1561 219
55 437 50.001 20 156 50.001 50 651 50.001
6116 5665 4763
3797–9852 2657–12 079 3099–7321
1554 219 1037 219 647 224
50 433 50.001 22 404 50.001 8347 0.004
4729 2820 1911
3080–7260 1836–4333 1232–2965
446
A. I. Guzel et al.
J Matern Fetal Neonatal Med, 2015; 28(4): 443–447
Figure 1. ROC curve for detecting risk variables in cases of RPL in groups classified as having prior live birth pregnancies.
situation. Franssen et al. [9] designed a survey study and reported that too many diagnostic tests and ineffective therapeutic interventions were performed in women with recurrent miscarriage. They argued that it was important to test only the subgroups of RPL with high risk for underlying abnormality. Similar to the results of Franssen et al., clinicians in our country perform too many unnecessary tests on these women. Referring to the results of our study, we also recommend these tests to the subgroups of women who had RPL. Jaslow et al. [10] reported that investigative diagnostic tests did not differ among women with different numbers of pregnancy losses (women with two, three, or 4 pregnancy losses). We have also found no difference among women with two or more pregnancy losses. Marquard et al. [11] designed a study on women aged 435 years old with 3 first trimester miscarriages and showed that there was no association between thrombophilia workup (including anticardiolipin antibodies, lupus anticoagulant, proteins C and S, prothrombin gene, antithrombin III, factor V Leiden, and methylenetetrahydrofolate reductase) and risk of miscarriage. Our study was similar to this study and we also did not find any association between thrombophilia panel in women who were divided into groups according to the age and the number of pregnancy losses.
Van den Boogaard et al. [12] also designed a retrospective study meant to evaluate the relationship between the number and sequence of preceding miscarriages and antiphospholipid syndrome. They found no difference in the women in terms of obstetric history and age of the women. In our study, in the third group classified according to having previous live birth pregnancies, women having previous live birth pregnancies had statistically different numbers of gravidity, parity, number of pregnancy losses, APCR, AT III, and age. These results were different from the findings of Van den Boogaard et al. [12]. These difference and interesting data are the additive point of our study to the literature. In another study, Metwally et al. [13] analyzed the effect of body mass index and the risk of miscarriage in women with RPL. They found that obesity increased the risk of RPL with a smaller but statistically significant difference. According to this study, the most important risk factor was the age of the women. In our study, age was not a single risk factor for RPL in women who had a previous live birth. Yildiz et al. [14] evaluated the role of Factor V Leiden (G1691A), prothrombin G20210A or methylenetetrahydrofolate reductase (MTHFR) C677T gene mutations in women with RPL and found that thrombophilia screening was not recommended in Turkish women with low risk of VTE.
DOI: 10.3109/14767058.2014.920811
Marquard et al. [11] determined the etiological factors of RPL in women aged 435 years old in a retrospective cohort study. They found that the most common etiological factor was fetal chromosomal abnormalities. About 4 out of 43 patients had antiphospholipid antibody (APA) syndrome and one patient had protein C deficiency. Our limitation in the current study is the absence of foetal chromosomal abnormalities, factor V Leiden, and methylenetetrahydrofolate reductase (MTHFR) gene mutations. Our study supports the findings of Marquard et al., as we also found no statistically significant difference between the groups formed according to the number of pregnancy losses and on the age of the women in terms of thrombophilia workup. There is a debate about the routine screening of all patients with RPL for inherited thrombophilias. The American Society for Reproductive Medicine (ASRM) recommends the screening of all women with RPL for lupus anticoagulant, anticardiolipin antibodies, and anti-b2 glycoprotein I. Stephenson [2] also recommends screening the same parameters. Laurino et al. [15] also recommend screening tests in women with RPL for antiphospholipid syndrome. McNamee et al. [16] reported that screening should be performed with limited parameters in women with high risks of thrombophilia. Blickstein et al. [8] recommended screening the high-risk cases for thrombophilias. We also found that screening is not necessary for all women with RPL but that subgroups, especially women with prior live birth pregnancies should undergo routine screening in terms of AT III and APCR. Our ROC curve results support this finding. In a case–control study, Hossain et al. [17] designed a study similar to ours and found no association between inherited thrombophilias and RPL in Pakistani women. Our results support this study, only it differs slightly in terms of the evaluation of subgroups. In conclusion, there is a strong association between antiphospholipid syndrome and RPL and screening may be useful in patients who are at high risk of VTE. We also do not recommend the screening of all cases. In women with high parity and those who had prior live birth pregnancies, although increased AT III and APCR may be a diagnostic marker for subsequent pregnancy losses.
Declaration of interest The authors declare that they have no conflict of interest.
Screening tests in women with recurrent pregnancy loss
447
References 1. Practice Committee of the American Society for Reproductive Medicine. Definitions of infertility and recurrent pregnancy loss. Fertil Steril 2008;89:1603. 2. Stephenson M, Kutteh W. Evaluation and management of recurrent early pregnancy loss. Clin Obstet Gynecol 2007;50: 132–45. 3. Practice Committee of the American Society for Reproductive Medicine. Evaluation and treatment of recurrent pregnancy loss: a committee opinion. Fertil Steril 2012;98:1103–11. 4. 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. 5. Preston FE, Rosendaal FR, Walker ID, et al. Increased fetal loss in women with heritable thrombophilia. Lancet 1996;348: 913–16. 6. Gris JC, Que´re´ I, Monpeyroux F, et al. Case-control study of the frequency of thrombophilic disorders in couples with late foetal loss and no thrombotic antecedent – the Nıˆmes Obstetricians and Haematologists Study5 (NOHA5). Thromb Haemost 1999;81: 891–9. 7. Blickstein I. Thrombophilia and women’s health: an overview. Obstet Gynecol Clin North Am 2006;33:347–56. 8. Blickstein D. Screening for thrombophilia. Obstet Gynecol Clin North Am 2006;33:389–95. 9. Franssen MT, Korevaar JC, van der Veen F, et al. Management of recurrent miscarriage: evaluating the impact of a guideline. Hum Reprod 2007;22:1298–303. 10. Jaslow CR, Carney JL, Kutteh WH. Diagnostic factors identified in 1020 women with two versus three or more recurrent pregnancy losses. Fertil Steril 2010;93:1234–43. 11. Marquard K, Westphal LM, Milki AA, Lathi RB. Etiology of recurrent pregnancy loss in women over the age of 35 years. Fertil Steril 2010;94:1473–7. 12. van den Boogaard E, Cohn DM, Korevaar JC, et al. Number and sequence of preceding miscarriages and maternal age for the prediction of antiphospholipid syndrome in women with recurrent miscarriage. Fertil Steril 2013;99:188–92. 13. Metwally M, Saravelos SH, Ledger WL, Li TC. Body mass index and risk of miscarriage in women with recurrent miscarriage. Fertil Steril 2010;94:290–5. 14. Yildiz G, Yavuzcan A, Yildiz P, et al. Inherited thrombophilia with recurrent pregnancy loss in Turkish women – a real phenomenon? Ginekol Pol 2012;83:598–603. 15. Laurino MY, Bennett RL, Saraiya DS, et al. Genetic evaluation and counseling of couples with recurrent miscarriage: recommendations of the National Society of Genetic Counselors. J Genet Couns 2005;14:165–81. 16. McNamee K, Dawood F, Farquharson RG. Thrombophilia and early pregnancy loss. Best Pract Res Clin Obstet Gynaecol 2012;26: 91–102. 17. Hossain N, Shamsi T, Khan N, Naz A. Thrombofilia investigation in Pakistani women with recurrent pregnancy loss. J Obstet Gynaecol Res 2013;39:121–5.
Copyright of Journal of Maternal-Fetal & Neonatal Medicine is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
