Types of pregnancy loss in recurrent miscarriage: implications for research and clinical practice

L. Bricker1,3 and R.G. Farquharson2

1 Department of Obstetrics and Gynaecology and 2 Miscarriage Clinic, University of Liverpool, Liverpool Women's Hospital, Crown St, Liverpool L8 7SS, UK


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: In recurrent pregnancy loss, there is much debate about cause and association, as exact pathophysiological mechanisms have not been elucidated. The aim of this study was to assess whether recurrent pregnancy loss (RPL) patterns differ according to causal/associated conditions, suggesting differing disease processes. METHODS: Following investigation, 427 women with RPL were classified into the following `diagnostic' groups: idiopathic, oligomenorrhoea, antiphospholipid syndrome (APS) and `possible' APS. A total of 323 subsequently conceived, and underwent serial ultrasonography in early pregnancy; of these, 87 (26.9%) miscarried, and the types of pregnancy loss for the four diagnostic groups were allocated to either embryo loss (fetal heart never seen) and fetal loss (fetal heart seen prior to pregnancy loss). RESULTS: Overall, there were 75 embryonic losses and 12 fetal losses. The fetal loss rates in each group were similar: idiopathic 5.1%, oligomenorrhoea 3.4%, `possible' APS 4.9% and APS 4.8%. CONCLUSIONS: Serial ultrasound helps to discriminate type of pregnancy loss and demonstrates that embryo loss is more common than fetal loss. More importantly, pregnancy loss patterns do not seem to differ between diagnostic groups in a treated population. The fetal loss rate in each of the diagnostic groups is similar to that in other reported populations.

Key words: antiphospholipid syndrome/embryonic loss/fetal loss/oligomenorrhoea/recurrent pregnancy loss


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Recurrent pregnancy loss (RPL) can be defined as more than two consecutive pregnancy losses at <24 weeks gestation and occurs in 1–2% of fertile women. When clinical pregnancy is established, the risk of spontaneous pregnancy loss is ~12–14% (Miller et al., 1980Go; Edmonds et al., 1982Go), and therefore the incidence of RPL by chance alone would be in the order of 0.35%. This discrepancy implies that in some cases there is an underlying pathological explanation. Following investigation of RPL, ~50% of cases have no cause or association found and are classified as idiopathic (Stirrat 1990Go; Quenby and Farquharson, 1993Go). It is generally accepted that within the idiopathic group there is considerable heterogeneity and it is unlikely that one single pathological mechanism can be attributed to their RPL. Known causes or associations of RPL fall into four categories: endocrine, immunological, anatomical and genetic. However, there is much debate about cause and association, as the exact pathophysiological mechanisms have not been elucidated. Current research is directed at theories related to implantation, trophoblast invasion and placentation, as well as factors which may be embryopathic.

Historically clinicians have grouped all pregnancy losses that occur at gestations prior to theoretical viability under the umbrella of `abortion'. More recently, among researchers in the field of RPL, it has been recognized that the classification of pregnancy loss is more complex, as the developing pregnancy undergoes various important stages, and differing pathology at the time of pregnancy loss is exhibited at these different stages. There has therefore been a plea to classify pregnancy losses according to the gestation at which they occur and detail of the event, for example intrauterine fetal death. In this way, possible pathophysiological mechanisms may be postulated and studied.

The objective of this study was to determine whether the pregnancy loss patterns in women with RPL differ according to causal or associated conditions, thus suggesting differing disease processes. If different conditions associated with RPL have specific, if as yet unexplained, mechanisms, one would expect the patterns of pregnancy loss to be different. This information is useful for the ongoing primary scientific research into RPL as laboratory findings are often correlated with subsequent pregnancy outcome. In addition, women who present themselves for investigation and management of RPL are generally highly motivated, and seek an understandable explanation. In fact, the finding of no association often causes much disappointment despite the fact that a subsequent pregnancy is more likely to be successful. Developing a knowledge base of gestational milestones for these women is very useful as an integral part of the supportive care provided by clinicians in recurrent miscarriage clinics.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The setting for this study was the Recurrent Miscarriage Clinic at Liverpool Women's Hospital, a teaching hospital of the University of Liverpool. Between 1988 and 1998, 850 women with a history of two or more pregnancy losses were referred to the clinic. Although the textbook definition of RPL is three or more pregnancy losses, we do accept referrals of women who have had two previous pregnancy losses, particularly if one or both have been second trimester losses or intrauterine deaths. Women are seen initially before a subsequent pregnancy, and at this preconceptual visit a thorough clinical history and investigation protocol is undertaken to exclude known associations with RPL, such as antiphospholipid syndrome (APS) and oligomenorrhoea. The full investigation protocol has been previously described (Drakeley et al., 1998Go). At the second clinic appointment, results and a plan of management for the next pregnancy are discussed. At this time, a repeat APS screen is undertaken. All women are advised to take preconceptual folic acid supplements.

In a subsequent pregnancy, women are seen within a week of missing a period, and scanned twice weekly thereafter, until 12–14 weeks gestation. Thereafter follow-up occurs in the Pregnancy Support Antenatal Clinic. Transabdominal and/or transvaginal ultrasonography (using 3.5 and 7.5–5 MHz transducers respectively) are employed depending on the individual situation, to ensure optimal visualization of the pregnancy and the pelvis. Management options includes supportive care if idiopathic RPL, aspirin or a combination of heparin and aspirin if APS, HCG supplementation if oligomenorrhoeic, and transcervical or transabdominal cerclage if indicated for cervical weakness. There is a dedicated Miscarriage Nurse who is available for advice during working hours, and after-hours women are referred or have direct access to the gynaecological emergency room.

Data were entered onto a live database in an ongoing fashion, and included demographic details, relevant past history (medical, surgical, obstetric and gynaecological), results of investigations, details and outcome of the subsequent pregnancy. In the case of a subsequent miscarriage, the gestation at miscarriage and details of whether fetal cardiac activity was ever identified were recorded.

If miscarriage occurred before fetal cardiac activity was identified, the pregnancy loss was defined as embryonic loss. If pregnancy loss occurred after fetal cardiac activity was identified, the pregnancy loss was defined as fetal loss.

For the purposes of this study, women were included if they had a history of three or more previous pregnancy losses in the first trimester, and they had idiopathic RPL, APS or oligomenorrhoea. Women who had tested positive for APS on one occasion only were also analysed, but in a separate group (`possible' APS), as we wanted to assess whether clinical outcome differed from the other groups. Women with a history of one or more mid-trimester losses, or who had a combination of diagnoses (e.g. APS and oligomenorrhoea) or who had a confirmed genetic abnormality were excluded from the study. Pregnancies which were terminated or ectopic gestations were excluded from the study sample.

To calculate the embryonic loss rate in each group, the number of embryonic losses (i.e. cardiac activity never identified ultrasonographically) was divided by the total number of pregnancies. To calculate the fetal loss rate in each group, the number of fetal losses (i.e. loss after cardiac activity identified ultrasonographically) was divided by the total number of pregnancies minus the embryo losses.

The {chi}2-test was used to assess if there were significant differences in the pregnancy loss patterns of different diagnostic groups. If the {chi}2-test was invalid due to small numbers, Fisher's exact test was used.

Maternal age has been shown to be an independent factor in predicting pregnancy outcome. Subgroup analysis was therefore undertaken for women aged <35 and >=35 years. This age was chosen as it is the age at which the risk of aneuploidy in pregnancy rises significantly.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Over the 10 year period studied, 850 women with a history of RPL were seen, investigated and entered onto the database. In 357 (42%) women, no cause or association could be found and they were classified idiopathic RPL. Seventy-three (8.6%) had oligomenorrhoea (defined as a cycle >=35 days), 86 (10.1%) true APS and 93 (10.9%) tested positive for APS on one occasion (`possible' APS). A total of 241 women were classified as mid-trimester loss (14.6%), multiple pathology (12.9%) and genetic abnormality (0.8%), and were excluded from the study (Table IGo). A total of 427 women fulfilled the inclusion criteria for the study (248 idiopathic RPL, 54 oligomenorrhoea, 57 APS and 68 possible APS).


View this table:
[in this window]
[in a new window]
 
Table I. Diagnostic group classification
 
Of the 427 women, the median age was 32 years (range 18–45) and the median number of previous pregnancy losses was 3 (range 3–14). The number of primary losers (i.e. no previous pregnancy survival >24 weeks gestation) was 219 (53%). Between the diagnostic groups these demographic details showed similarity (Table IIGo).


View this table:
[in this window]
[in a new window]
 
Table II. Demographic characteristics of each diagnostic group
 
In the study group of 427 women, 323 became pregnant, of whom 87 had a further miscarriage (26.9%). The remaining 104 women had not conceived again, verified by direct patient contact or GP contact. None of the women were lost to follow-up. Table IIIGo shows the pregnancies, pregnancy losses and types of pregnancy losses overall and for each diagnostic group. Pregnancies, pregnancy losses and types of pregnancy losses overall and for each diagnostic group are shown for the subgroups women aged <35 and >=35 years in Tables IV and VGoGo respectively.


View this table:
[in this window]
[in a new window]
 
Table III. Pregnancies, pregnancy outcomes, and types of pregnancy loss overall and for each diagnostic group
 

View this table:
[in this window]
[in a new window]
 
Table IV. Pregnancies, pregnancy outcomes, and types of pregnancy loss overall and for each diagnostic group in women aged >=35 years
 

View this table:
[in this window]
[in a new window]
 
Table V. Pregnancies, pregnancy outcomes, and types of pregnancy loss overall and for each diagnostic group in women aged <35 years
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
In normal pregnancy, the presence of fetal cardiac activity on ultrasound scan is generally reassuring, as the risk of spontaneous loss is lower following this gestational milestone (Brigham et al., 1999Go). In recent years, with the advancing ultrasonographic technology including the advent of transvaginal probes, fetal cardiac activity can be confirmed at earlier gestations (Goldstein et al., 1988Go). In very early pregnancy, the fetal heart is established and beating at 22 days post-conception (Moore 1982Go). Theoretically on transvaginal scan, a fetal pole and yolk sac should be seen if the gestation sac diameter is >20 mm and fetal cardiac activity should be visualized if the crown–rump length is >5 mm, i.e. after 6 weeks gestation (Royal College of Radiologists and the Royal College of Obstetricians and Gynaecologists Standing Joint Committee, 1995Go). On transabdominal scan these features may only be seen a week later. At present there is no clear evidence that ultrasound examination during pregnancy is harmful (Neilson, 2001Go).

This study demonstrates the value of serial ultrasound examination in discriminating type of pregnancy loss. To our knowledge, this is the first study to report both embryonic and fetal loss rates in a recurrent miscarriage population, presented according to diagnostic groups. Several studies have been published which evaluate pregnancy loss rates after ultrasonographic documentation of fetal cardiac activity (fetal loss rates) (see Table VIGo). However, marked differences in study design make comparisons difficult. These differences include the type of population studied, indication for ultrasound, type of ultrasound scan performed, whether ultrasound was single or serial and gestation at which first ultrasound was performed. In some studies, details of the aforementioned factors are unclear, further compounding the difficulty in comparisons. In addition, ultrasound is a relatively new technology with rapid advances which constantly expand clinical application, making earlier studies less relevant in the present clinical setting. This is particularly important when considering first trimester ultrasound assessment, because, as mentioned before, the introduction of transvaginal ultrasonography in the late 1980s has greatly improved early pregnancy assessment.


View this table:
[in this window]
[in a new window]
 
Table VI. Studies reporting pregnancy loss rates (published from 1990 onward)
 
We have demonstrated that the overall pregnancy loss rate is 26.9% and that for each diagnostic group, the rates are similar (17–29%, P = 0.39, not significant) (Table IIIGo). Embryo loss is far more common than fetal loss (embryo:fetal loss ratio = 6:1), and the observed number of fetal losses is small. The overall embryonic loss rate is 23.2% and is similar between diagnostic groups (12.8–25.7%, P = 0.30, not significant). The overall fetal loss rate is 4.8% and is similar between diagnostic groups (3.4–5.1%, P = 0.99, not significant). The fetal loss rate is similar to that described in other populations.

APS is associated with pregnancy complications and pregnancy loss in all three trimesters of pregnancy (Branch et al., 1992Go). By definition, pregnancy loss is one of the diagnostic criteria. To make the diagnosis of APS requires: (i) a clinical history of RPL; or (ii) pregnancy complications due to placental insufficiency; or (iii) thrombosis; or (iv) thrombocytopenia; and (v) positive tests for antiphospholipid antibodies (anticardiolipin antibodies and/or lupus anticoagulant) on at least two occasions 6 weeks apart (Wilson et al., 1999Go). APS activity is most likely to be detected in the first trimester of pregnancy (Topping et al., 1999Go) and despite repeat testing some women test positive on only one occasion, thus defying the diagnostic criteria. We have also shown that these women (`possible' APS) have a fetal loss rate (4.9%) similar to those women who have never tested positive. The significance of one positive test is not known and further evaluation of this group of women is required.

Preconceptual hypersecretion of LH associated with polycystic ovarian disease (PCOD) was thought to be associated with an increased risk of pregnancy loss (Regan et al., 1990Go). However, studies aimed at suppressing LH failed to show an improvement in pregnancy outcome (Clifford et al., 1996Go). By contrast, oligomenorrhoea has been shown to be an independent adverse risk factor for pregnancy loss in a RPL population and this group of women have lower luteal phase estradiol levels suggesting poor endometrial receptivity (Quenby and Farquharson, 1993Go).

In recurrent miscarriage associated with oligomenorrhoea, the presence of a live fetus is encouraging, with the risk of fetal loss at 3.4%.

Table VIGo summarizes studies published from 1990 onwards which have reported pregnancy loss rates. In unselected populations with no history of infertility or RPL, fetal loss rates are 3.3–6.1%. The reported fetal loss rates in infertile populations vary widely, ranging from 7 to 15%. Three studies evaluating pregnancy loss after ultrasonographic documentation of fetal cardiac activity in women with a history of RPL (Opsahl and Pettit, 1993Go; Van Leeuwen et al., 1993Go; Laufer et al., 1994Go) report fetal loss rates of 17–22.7%. The first two studies do not report the diagnostic work-up of these women and therefore may represent an unknown and complex case mix, a feature that we dealt with by removal of mid-trimester loss and dual pathology groups.

It is difficult to explain why the fetal loss rate in our population of women with a history of RPL is relatively lower than other reported recurrent miscarriage populations. It may be attributable to our inclusion criteria of women with only previous first trimester losses, as it is known for example that APS is strongly associated with later pregnancy loss (Branch et al., 1992Go; Drakeley et al., 1998Go). Alternatively, the effect of the various treatments we offer, and perhaps more importantly the intensive supportive care given throughout pregnancy, may result in more favourable outcomes. There is evidence in the literature to support the effectiveness of `tender loving care' on successful pregnancy outcome (Stray-Pederson and Stray-Pederson 1984Go; Liddell et al., 1991Go). Of course, the fact that the women included in our study received various treatment regimens may introduce bias, but, if anything, one would expect higher pregnancy loss rates and perhaps differences in the types of pregnancy loss between diagnostic groups had the pregnancies been untreated. It would have been unethical to withhold treatment for the purposes of the study. Therefore, the results of this study should be applied to women undergoing treatment, and, as most women attending recurrent miscarriage services receive some form of treatment, especially if they have antiphospholipid antibodies, our data are pragmatic.

The pregnancy loss pattern according to diagnostic group shows some variation in women aged >=35 years (Table IVGo), and both the pregnancy loss rates and fetal loss rates are higher than for women aged <35 years (Table VGo); a similar finding in an infertile population has been demonstrated (Smith and Buyalos, 1996Go).

A significant shortcoming in this paper is the absence of karyotypic analysis of pregnancy loss tissue. It is generally accepted that ~50% of early pregnancy losses are associated with abnormal karyotype. A retrospective analysis of products of conception from 224 pregnancies (Stern et al., 1996Go) reported no difference in the frequency of abnormal karyotype in pregnancy losses from women with RPL compared with women with no history of RPL (57% abnormal karyotypes in both groups). In a more robust, large, prospective study of 1309 pregnancies (Ogasawara et al., 2000Go) it was shown that an overall loss rate of 35% can be expected in RPL groups, irrespective of associated pathology. In their study, of 458 pregnancy losses, 234 (51%) were karyotyped of which 120 (51%) had an abnormal karyotype. By contrast, of 114 sporadic losses in a general population, 77% had an abnormal karyotype, and this difference was significant. Of interest was the finding that the frequency of normal karyotypes significantly increased with the number of previous pregnancy losses and this further confirms the likelihood of underlying pathophysiology in RPL apart from abnormal karyotype. Unfortunately, for pragmatic and financial reasons, over the 10 year period of our study, we have not routinely undertaken cytogenetic analysis of the products of conception in our population.

These data are valuable in both the research and clinical settings. In the research setting, clinical outcome is often correlated with in-vitro laboratory experiments on tissue samples from women with RPL (for example endometrial and serum studies), which are undertaken in an attempt to identify the pathophysiology of early pregnancy failure. More detailed information about the type of pregnancy loss is useful in interpreting the findings. While numerous studies have reported fetal loss rates in different populations on the basis of ultrasound viability, none have correlated these findings with karyotypic characterization of the pregnancies. As pregnancy losses accumulate, abnormal karyotypes remain constant and loss of normal karyotypes increases (Ogasawara et al., 2000Go). We suspect that embryonic losses are more likely to be karyotypically abnormal and fetal losses more likely to be karyotypically normal. Future research is needed to evaluate types of pregnancy losses in an RPL population divided into clinical subgroups, and to correlate these findings with cytogenetic analysis of products of conception to elucidate whether different types of pregnancy losses are characterized by chromosomally normal or abnormal conceptuses.

In the clinical setting, women with a history of RPL experience intense stress and anxiety following conception in a subsequent pregnancy. Emotional support and regular pregnancy assessment with ultrasound forms the mainstay of the Miscarriage Clinic service. Providing accurate and realistic information about the chances of success in an ongoing pregnancy is essential in this process and it is clear that success prediction will depend more on maternal age than on the diagnostic group the women fits into. Once fetal cardiac activity has been confirmed ultrasonographically, reassurance that there is a 95% chance of pregnancy survival can be given. In addition, these data are useful when assessing the clinical outcome of treatment intervention trials before implementation of evidence-based practice.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors would like to thank Kelly Wood and Jo Teare for their maintenance of the Miscarriage Clinic database, Anne Marie Hughes the Miscarriage Clinic nurse and the women of Liverpool and beyond who attend the Miscarriage Clinic.


    Notes
 
3 To whom correspondence should be addressed. E-mail: lbricker{at}liv.ac.uk Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Branch, D.W., Silver, R.M., Blackwell, J.L., Reading, J.C. and Scott, J.R. (1992) Outcome of treated pregnancies in women with antiphospholipid syndrome: an update of the Utah experience. Obstet. Gynecol., 80, 614–620.[Abstract]

Brigham, S.A., Conlon, C. and Farquharson, R.G. (1999) A longitudinal study of pregnancy outcome following idiopathic recurrent miscarriage. Hum. Reprod., 14, 2868–2871.[Abstract/Free Full Text]

Clifford, K., Rai, R.,Watson, H., Franks, S. and Regan, L. (1996) Does suppressing luteinising hormone secretion reduce the miscarriage rate? Results of a randomised trial. Br. Med. J., 312, 1508–1511.[Abstract/Free Full Text]

Drakeley, A.J., Quenby, S. and Farquharson, R.G. (1998) Mid-trimester loss|appraisal of a screening protocol. Hum. Reprod., 13, 1975–1980.[Abstract]

Edmonds, D.K., Lindsay, K.S., Miller, J.F., Williamson, E. and Woods, P.J. (1982) Early embryonic mortality in women. Fertil. Steril., 38, 447–453.[ISI][Medline]

Goldstein, S.R. (1994) Embryonic death in early pregnancy: a new look at the first trimester. Obstet. Gynecol., 84, 294–297.[Abstract]

Goldstein, S.R., Snyder, J.R., Watson, C. and Danon, M. (1988) Very early pregnancy detection with endovaginal ultrasound. Obstet. Gynecol., 72, 200–204.[Abstract]

Hill, L.M., Guzick, D., Fries, J. and Hixson, J. (1991) Fetal loss rate after ultrasonically documented cardiac activity between 6 and 14 weeks, menstrual age. J. Clin. Ultrasound, 19, 221–223.[ISI][Medline]

Keenan, J.A., Rizvi, S. and Caudle, M.R. (1998) Fetal loss after early detection of heart motion in infertility patients: prognostic factors. J. Reprod. Med. Obstet. Gynecol., 43 (Suppl.), 199–202.

Laufer, M.R., Ecker, J.L. and Hill, J.A. (1994) Pregnancy outcome following ultrasound-detected fetal cardiac activity in women with a history of multiple spontaneous abortions. J. Soc. Gynecol. Invest., 1, 138–142.[ISI][Medline]

Liddell, H.S., Pattison, N.S. and Zanderigo A. (1991) Recurrent miscarriage: outcome after supportive care in early pregnancy. Aust. NZ J. Obstet. Gynaecol., 31, 320–322.[ISI][Medline]

Miller, J.F., Williamson, E., Glue, J., Gordon, Y.B., Grudzinskas, J.G. and Sykes, A. (1980) Fetal loss after implantation: a prospective study. Lancet, ii, 554–556.

Molo, M.W., Kelly, M., Balos, R., Mullaney, K. and Radwanska, E. (1993) Incidence of fetal loss in infertility patients after detection of fetal heart activity with early transvaginal scan. J. Reprod. Med., 38, 804–806.[ISI][Medline]

Moore, K.L. (1982) The Developing Human, 3rd edn. W.B.Saunders, Philadelphia, pp. 70–109.

Neilson, J.P. (2001) Ultrasound for fetal assessment in early pregnancy (Cochrane Review). In The Cochrane Library, Issue 2. Update Software, Oxford.

Ogasawara, M., Aoki, K., Okada, S. and Suzumori, K. (2000) Embryonic karyotype of abortuses in relation to the number of previous miscarriages. Fertil. Steril., 73, 300–304.[ISI][Medline]

Opsahl, M.S. and Pettit, D.C. (1993) First trimester sonographic characteristics of patients with recurrent spontaneous abortion. J. Ultrasound Med., 12, 507–510.[Abstract]

Qasim, S.M., Sachdev, R., Trias, A., Senkowski, K. and Kemmann, E. (1997) The predictive value of first-trimester embryonic heart rates in infertility patients. Obstet. Gynecol., 89, 934–936.[Abstract]

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

Royal College of Radiologists and the Royal College of Obstetricians and Gynaecologists Standing Joint Committee (1995) Guidance and Ultrasound Procedures in Early Pregnancy. HMSO, London.

Regan, L., Owen, E.J. and Jacobs, H.S. (1990) Hypersecretion of LH: infertility and miscarriage. Lancet, 336, 1141–1144.[ISI][Medline]

Rosen, G.F., Silva, P.D., Patrizio, P., Asch, R.H. and Yee, B. (1990) Predicting pregnancy outcome by the observation of a gestational sac or of early fetal cardiac motion with transvaginal ultrasonography. Fertil. Steril., 54, 260–264.[ISI][Medline]

Smith, K.E., and Buyalos, R.P. (1996) The profound impact of patient age on pregnancy outcome after early detection of fetal cardiac activity. Fertil. Steril., 65, 35–40.[ISI][Medline]

Stern, J.J., Dorfmann, A.D., Gutierrez-Najar, A.J., Cerrillo, M. and Coulam, C.B. (1996) Frequency of abnormal karyotypes among abortuses from women with and without a history of recurrent spontaneous abortion. Fertil. Steril., 65, 250–253.[ISI][Medline]

Stirrat, G.M. (1990) Recurrent miscarriage; definition and epidemiology. Lancet, 348, 1402–1406.

Stray-Pederson, B. and Stray-Pederson, S. (1984) Etiologic factors and subsequent reproductive performance in 195 couples with a prior history of habitual abortion. Am. J. Obstet. Gynecol., 148, 140–146.[ISI][Medline]

Topping, J., Quenby, S., Farquharson, R.G., Malia, R. and Greaves, M. (1999) Marked variation in antiphospholipid antibodies during pregnancy: relationships to pregnancy outcome. Hum. Reprod., 14, 224–228.[Abstract/Free Full Text]

Van Leeuwen, I., Ware Branch, D. and Scott, J.R. (1993) First-trimester ultrasonography findings in women with a history of recurrent pregnancy loss. Am. J. Obstet. Gynecol., 168, 111–114.[ISI][Medline]

Wilson, W.A., Gharavi, A.E., Kolke, T., Lockshin, M.D., Branch, D.W. and Plette, J.C. (1999) International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome: report of an internal workshop. Arthrit. Rheumat., 42, 1309–1311.[ISI]

Submitted on August 3, 2000; resubmitted on October 19, 2001; accepted on January 14, 2002.