Nuchal translucency measurement and pregnancy outcome after assisted conception versus spontaneously conceived twins

R. Maymon1,3, E. Jauniaux2, A. Holmes2, Y.M. Wiener1, E. Dreazen1 and A. Herman1

1 Department of Obstetrics and Gynecology, Assaf Harofe Medical Center, Zerifin 70300 (affiliated with Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv), Israel, and 2 Academic Department of Obstetrics and Gynaecology, Royal Free and University College, UCL campus, London, UK


    Abstract
 Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: Nuchal translucency (NT) measurement for Down's syndrome screening or detecting various fetal anomalies is a reliable sonographic marker. This study evaluates the contribution of NT screening in spontaneously conceived and assisted conception twin pregnancies. METHODS AND RESULTS: Maternal age at measurement, chorionicity, ultrasound features, karyotype results and pregnancy outcome were recorded prospectively and compared in 83 assisted reproduction treatment and 91 spontaneously conceived twins. Pregnancy outcome was evaluated according to maternal age, method of conception, NT data and chorionicity. NT measurements (>=95 centiles of the normal range) were considered screen-positive and mid-pregnancy fetal karyotyping was advised. Complicated pregnancy outcome, which could be signalled by increased NT, was defined as either chromosomal abnormalities, severe structural defects or fetal demise. Based on NT measurements, 16 fetuses (4.6%) were found to be screen-positive. Five of them had chromosomal aneuploidy and selective termination was performed. The parents also opted for this procedure in another five fetuses because of major structural abnormality diagnosed during NT assessment. No other chromosomal or major fetal abnormality were found post-natally. Although no difference was found in NT, crown–rump length and maternal age between spontaneous and assisted reproduction technology twin pregnancies, the former group had a significantly higher rate of screen-positive results (7 versus 2%, P = 0.047), amniocentesis uptake (33 versus 22%, P = 0.014), monochorionic twining (32 versus 4%, P = 0.001) and complicated pregnancy outcome (11 versus 5%, P = 0.02). CONCLUSION: The present study confirms that first trimester target scanning can improve outcome by early detection and management of cases with an anomalous co-twin. It also identifies some differences between spontaneously and artificially conceived twin pregnancies in relation to this area of testing.

Key words: twins/nuchal translucency/ultrasound/assisted conception/fetal outcome


    Introduction
 Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Acknowledgements
 References
 
Over the past decade, both maternal age of primiparas and the widespread increase in the application of assisted reproduction technology and ovulation-induction drugs have risen dramatically. Currently, more than 25% of pregnancies obtained by these procedures are twins (Olivenness et al., 1996). Furthermore, the twining rate is related to maternal age, and women >35 years of age are three times more likely to spontaneously conceive twins than are younger women (Spencer, 2000Go). This trend has a great impact on perinatal care since twin pregnancies pose numerous management problems from conception to delivery. For example, they are at higher risk both for chromosomal abnormality (Botting et al., 1987Go; Cuckle, 1998Go) and structural defect for each individual co-twin (Doyle et al., 1990Go). Additionally, Down's syndrome serum screening is of limited value in twin pregnancies (Wald et al., 1991Go) because of the unaffected co-twin's masking of the abnormal maternal serum distribution seen in singleton analyses (Cuckle, 1998Go). It has also been recently reported that the human chorionic gonadotrophin (HCG) serum levels are usually higher among twin pregnancies following assisted reproduction technology treatment, thus resulting in higher false positive rates for Down's syndrome screening (Räty et al., 2000Go).

Women who conceived after assisted reproduction technology treatment are on average older and, in many cases, their current pregnancy (which may be their last) was achieved after long-standing infertility. For this reason, they would be extremely wary of an invasive fetal karyotyping procedure in which the miscarriage rate following amniocentesis is doubled in twins compared with that in matched singletons (Kidd et al.1997Go). It follows that every effort should be made to provide women who are pregnant with twins with the most accurate Down's syndrome screening tests (Maymon et al.1999aGo). The accumulation of fluid behind the fetal neck, defined as nuchal translucency (NT) thickness, was found to be a promising marker alerting for complicated pregnancy outcome. Those abnormalities include chromosomal aneuploidies, severe structural defects, a wide range of genetic syndromes and miscarriages (Nicolaides et al.1999Go). The application of this sonographic screening method in twin pregnancies is supported by the observation that NT measurements are similarly accurate in trisomic twin pregnancies and affected singletons (Pandya et al., 1995aGo; Sebire et al., 1996aGo; Nicolaides et al., 1999Go). Furthermore, NT assessment was found to be useful in helping to choose and perform the appropriate invasive diagnostic cytogenetic procedure in twin pregnancies (Sebire et al., 1996bGo). Finally, in monochorionic twins, a large NT thickness may alert for early onset of feto-fetal transfusion syndrome (Sebire et al. 1997aGo, 2000Go).

We present herein the results of a twin screening study in which two centres participated. Specifically, twins conceived with assisted reproduction technology were compared with spontaneous twin pregnancies for NT screening results and throughout their gestation. This subject is not satisfactorily addressed in the literature.


    Materials and Methods
 Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Acknowledgements
 References
 
The current study was carried out between June 1998 and November 1999 in two fetal medicine units: Assaf Harofe Medical Center, Zrifin, Israel and University College London Hospital, London, UK. All cases were recruited from the same clinics over the same period of time, and the same diagnostic protocol was used. Each institution has a tertiary unit known as the `twin clinic' to which pregnant women with multifetal pregnancies are referred for target scanning. Data from twins after fetal reduction of high order multiplicity were not included in the current study. The gestational ages were calculated from the last menstrual period and thereafter confirmed by the crown–rump length (CRL) (Robinson and Flemming, 1975). Specifically, the gestational age was calculated on the basis of the largest CRL length (Sebire et al., 1997aGo) and only fetuses with a CRL of 38–84 mm were included. The twin pregnancy group was further subdivided into twins conceived after assisted reproduction technology treatment (study group) and spontaneously conceived twins (control group).

A transabdominal and/or transvaginal ultrasound examination (Aloka 1600, Tokyo, Japan and HDI 3000; Advanced Technology Laboratories, Seattle, WA, USA) with a variable focus 3.5–5 MHz curvelinear transducer and a transvaginal 5 or 7 MHz transducer was performed in all cases. There was no set time limit for generating a satisfactory image. The ultrasound examination determined the viability, size and position of each fetus, as well as the location of the placenta, placental attachment to the inter-twin membrane and the relation between the amniotic sac and the cervix. Chorionicity was determined based on previous sonographic criteria (Sepulveda et al. 1996Go, 1997Go) in which monochorionic twin pregnancies were classified if there was a single placental mass and the absence of the {lambda}-sign at the inter-twin membrane–placental junction.

Following the description by Pandya et al. (1995b), precise measurements of the maximum thickness of the translucent area between the skin and the soft tissue overlying the cervical spine were obtained in the sagittal section of the fetus (Pandya et al.1995bGo). Cases with NT >=95 centiles of the normal range for various CRLs in singletons [2 mm for CRLs <44 mm, to 2.85 mm for CRLs of 83–84 mm (Herman et al., 2000Go)], in which the calipers were placed `on–on' the nuchal lines, were considered screen-positive and the parents were offered the option of invasive prenatal karyotype. Based on our published model (Herman et al., 1998aGo), each sonograph NT image was subject to regular internal audit to check quality control, standardization and performance. The intra-observer repeatability coefficients of our team (R.M. and A.H.) were 0.28 mm and 0.34 mm. Intra-observer repeatability of image measurement showed a significant difference (P < 0.01), but the mean difference (± SD) of 0.10 ± 0.18 mm was negligible (Herman et al., 1998bGo). This performance corresponds well with other groups (Pandya et al., 1995cGo). Prior to examination, the pregnant women were given an explanation on the nature and implications of the test. Following the test, they were counselled on the potential risks and specific recommendations were made. Genetic counselling, fetal karyotyping, mid-gestation detailed anomaly and fetal echocardiography scans were offered in all cases of NT >=95 centiles. Following the sonographic evaluation, the patients were referred to their obstetrician for further prenatal care.

The method of conception, maternal age at measurement, chorionicity, karyotype results and pregnancy outcome were recorded prospectively on a computer database. For the current study, complicated pregnancy outcome was defined as aneuploidy, severe structural defects or miscarriages, since those abnormalities have been reported to be associated with increased NT (Nicolaides et al.1999Go). The fetal outcome was categorized into miscarriage, intrauterine fetal death, neonatal death or alive and well. All disorders or abnormalities were recorded in detail. Karyotype results and pregnancy outcome of all the twins and their medical history were obtained from the parents by a telephone interview and, when necessary, their medical records were requested from the respective departments.

Statistics
Standardized kurtosis showed that the data derived from a normal distribution of the twin population, and thus the variables, are expressed as mean and SD. Frequencies are expressed as a percentage. Prediction of a complicated pregnancy outcome was classified as aneuploidy, severe structural defects which justified selective termination, or fetal demise. Those parameters were assessed based on mode of conception, maternal age, NT value and CRL. A stepwise logistic regression model was applied to the data. The cases were further subdivided according to method of conception (assisted reproduction technology versus spontaneous) and chorionicity. Differences between the groups in chorionicity, amniocentesis rate and cases with complicated versus uncomplicated pregnancy outcome were compared using {chi}2 or Fisher's exact test, where applicable. The two-sample t-test was used to assess the difference between CRL, NT and maternal age in those groups. After excluding all abnormal cases, a linear regression model for NT and CRL was fitted to the data. P-values < 0.05 were considered statistically significant. Statistical analysis was carried out by the statistical department of the Tel Aviv University, using SPSS software.


    Results
 Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Acknowledgements
 References
 
This study comprised 91 twins conceived spontaneously and 83 twin pregnancies which followed assisted reproduction technology treatment, in which complete follow-up was available. A total of 107 pregnancies were from Assaf Harofe of which 60 were part of a previous twin study (Maymon et al.1999bGo). Sixty-seven pregnancies were from University College London Hospital. Based on first trimester NT screening results, 16 (4.6%) fetuses out of 348 co-twins which had been assessed were found to be screen-positive and the women were offered prenatal karyotyping (Table 1Go). However, the karyotype was available in another 80 fetuses because of other indications for testing, including second trimester abnormal sonographic or biochemistry results and the parents' request. Thus, the overall prenatal karyotyping uptake was 28% of the twins studied. There were five cases with aneuploidy, including three cases of Turner's syndrome and two with Down's syndrome. The observed prevalence of aneuploidy among fetuses with abnormal NT (>=95 centiles) was 5/16 fetuses (31%), and for the complete group it was 5/348 (1.4%). Post-natally, no traits were observed that would have warranted chromosomal analysis in those neonates in which prenatal karyotyping was not performed. Overall, 10 fetuses (3%) underwent selective termination at the request of the parents (nine of them in the second trimester and one during the third trimester). This was because of chromosomal aneuploidy in five cases, and major structural anomalies in the other five (four cases because of neural tube defect and one bichorionic euploid co-twin because of hydrops). In one case, the entire pregnancy was lost following the procedure. A co-twin with increased NT and a large thoracic cyst which resolved spontaneously and uneventful outcome was reported elsewhere (Jauniaux et al.2000Go). The overall spontaneous miscarriage rate was 5% of the twins in the current study. There were three cases of monochorionic twins which developed feto-fetal transfusion syndrome, among which one co-twin had intrauterine death at 30 weeks and the other an early neonatal death.


View this table:
[in this window]
[in a new window]
 
Table I . The distribution and comparison of the 182 spontaneous and 166 assisted reproduction technology conceived co-twins pregnancies by maternal age, sonographic measurements, amniocentesis rate and complicated pregnancy outcome*
 
There were no differences between maternal age and method of conception between the two groups (complicated/uncomplicated outcome). Therefore, only CRL and NT (which were found to be significant after applying the logistic regression analysis) were included in the model. The probability of a complicated pregnancy outcome (dependent variable) and the CRL and NT (independent variable) were thus calculated using a logistic regression model.

Complicated pregnancy outcome was defined as aneuploidy, major organ anomalies diagnosed prenatally which justified selective termination and spontaneous miscarriages. The odds ratio of complicated pregnancy outcome associated with increasing NT measurement with every 1 mm was 4.36 [95 confidence interval (CI) 2.40–7.93, P < 0.001] whereas an increase in the CRL was related to a better pregnancy outcome with an odds ratio of 0.91 (CI 0.87–0.96, P < 0.001). The odds ratios were calculated using continuous variables of NT and CRL.

When an arbitrary cut-off point of P = 0.5 was used, a complicated twin pregnancy outcome could be anticipated with sensitivity, specificity, and positive and negative predictive values of 33, 100, 100 and 95% respectively.

In the uncomplicated twin pregnancies, a linear regression model (R2 = 0.1227, P < 0.001) between an individual CRL and NT was used and showed an increased NT value as gestation progressed (Figure 1Go).



View larger version (18K):
[in this window]
[in a new window]
 
Figure 1 . Linear regression between nuchal translucency (NT, in mm) and crown-rump length (CRL, in mm) with 95% confidence limits for unaffected individuals ({square} = assisted reproduction technology; {blacktriangleup} = spontaneous conception).

 
Table IGo presents and compares the data of the assisted reproduction technology treatment and spontaneous conception twin pregnancies. The values of the mean maternal age, CRL and NT thickness were comparable. The spontaneous conception pregnancies had a significantly higher rate of screen-positive results, monochorionic twins, amniocentesis uptake and complicated pregnancy outcome (Table IGo). The pregnancies were further subdivided according to their chorionicity (Table IIGo). No significant difference between the monochorionic and bichorionic twins was found for either the screen-positive rate or a complicated pregnancy outcome (Table IIGo).


View this table:
[in this window]
[in a new window]
 
Table II . The distribution of the twins pregnancies, their screen-positive rate (NT >95th centile), chorionicity and complicated pregnancy outcome*
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Acknowledgements
 References
 
Preventing the birth of a twin with chromosomal or structural anomalies whilst minimizing the risk to the normal co-twin continues to pose an important prenatal challenge. This requires early recognition of twin gestation, determination of zygosity and a target scanning to detect any markers and signs of malformations.

The obstetric outcomes of twins are usually analysed by the method of conception and chorionicity. However, the NT data from the first trimester emerge as being highly informative for determining other aspects of pregnancy outcome, such as chromosomal and structural abnormalities (Nicolaides et al., 1999Go). Studies comparing the obstetrical outcome in assisted and spontaneous conception twin pregnancies have yielded conflicting results. Moise et al. (1998) found that IVF twins were at a higher risk of neonatal morbidity and mortality than spontaneously conceived twins, whereas Olivennes et al. (1996) and Koudstaal et al. (2000) reported no significant difference in perinatal outcome, the latter in a matched control study (Olivennes et al., 1996Go; Moise et al.1998Go; Koudstaal et al.2000Go). We found no significant difference in maternal age between assisted and spontaneous conception twins. However, the former group had a significantly higher amniocentesis uptake and complicated pregnancy outcome rate. Various theories were proposed to explain this difference: Olivennes et al. (1996) found fewer obstetric visits among parents with spontaneous twin pregnancies and argue that this trend may explain the different outcome (Olivennes et al., 1996Go). Our findings did not confirm theirs. A more reasonable explanation might lie in the higher monozygosity rate found in the spontaneously-conceived twins. Although it was recently reported in a series of 303 monochorionic twin pregnancies that there were 16 (3%) pregnancies in which at least one fetus was found to be structurally or chromosomally abnormal (Sebire et al., 2000Go). Nevertheless, and in agreement with one of the above studies (Koudstaal et al.2000Go), we argue that this phenomenon implies that the risk for a complicated outcome in the spontaneous conception twins group is higher than in the assisted reproduction technology fetuses because of the difference in the zygosity.

The NT screen-positive rate of 5% is consistent with our previous twins study (Maymon et al., 1999bGo) and is lower than the 7% and 8.6% in dichorionic and uncomplicated monochorionic twins respectively, reported by others (Sebire et al., 1996aGo, 1997aGo). Additionally, the prevalence of chromosomal abnormalities on the basis of maternal age and fetal NT measurement in the present study was 31% in comparison with 23% reported by the same group (Sebire et al., 1996bGo). However, the median maternal age in their study population was 35 years and, considering that the maternal background age-related risk is part of the first trimester NT screen results, this may contribute to higher false positive rate.

The current finding of a 5% overall spontaneous fetal loss rate is in agreement with other reports (Sebire et al., 1997bGo; Nicolaides et al., 1999Go), and is more than double the 2% rate observed in an unselected population at 10–14 weeks gestation (Pajkrt et al., 1999Go). These results may extend our understanding of the epidemiological observation of a low Down's syndrome prevalence in twin pregnancies. On the basis of zygosity calculations, the prevalence per term pregnancy should be almost double that in singletons whereas, in fact, the prevalence in twins and singletons is similar (Cuckle, 1998Go). The overall fetal loss rate in singleton pregnancies with Down's syndrome is about 50% (Macintosh et al., 1995Go). If the relative risk of loss in apparently normal twins is the same as that observed in singletons, then it might also apply to Down's syndrome pregnancies, in which case the loss per term pregnancy would be much greater in affected than in unaffected pregnancies. This effect could account for the apparently similar occurrence in twins and singletons.

An 8.6 and 37.5% incidence of increased NT has been previously reported in uncomplicated and complicated monochorionic twins respectively. The latter were found to be at risk for developing feto-fetal transfusion syndrome (Sebire et al., 1997aGo, 2000Go). This finding in 32 sets of monochorionic twins has not yet been confirmed, and the increased NT rate being similar to that of our dizygotic twins (4.7 and 4.9% respectively) may be because of the small size of our series. The current study sheds more light on those aspects as well as providing comparative data on the effectiveness of NT screening in twin pregnancies. Thus, by using the logistic regression model presented herein, accurate calculations of the probability for an uncomplicated pregnancy outcome could be provided as early as 10–14 weeks gestation. The NT values for both the assisted reproduction technology twins and their controls were found to increase with advancing gestation (Figure 1Go), and an increase in NT was related to a positive significance odds ratio for complicated pregnancy outcome. In contrast, an increase in CRL was related to a better pregnancy outcome. Our data (Figure 1Go) validate the reference ranges of NT measurements that had originally been developed for singletons (Braithwaite et al., 1996Go).

In twins, NT measurements can be performed either alone or in combination with first trimester serum biochemistry (Spencer, 2000Go). By combining NT measurement and various serum markers (Cuckle and Shemi, 1999Go; Wald et al., 1999Go), an improvement of the detection of the trisomic co-twins can be expected, with the benefit of ultrasound being able to specifically locate the affected one (Spencer, 2000Go). Hopefully, those algorithms might counterbalance the increased midgestation HCG maternal serum levels reported in assisted reproduction technology twins (Räty et al.2000Go).

In conclusion, our data indicate that assisted reproduction technology-derived twin pregnancies are not associated with increased risk of various pregnancy complications. The findings of this study should be reassuring both for physicians and their patients who are engaged in assisted reproduction technology treatment. However, they should not be misguiding: overall, twin pregnancies still have increased adverse outcome when compared with singletons. Our experience and that of others (Nicolaides et al., 1999Go) shows that first trimester target NT scan can improve outcome by early detection and management of cases with an anomalous co-twin. Therefore, the implementation of NT assessment should be favorably considered in antenatal twin clinics.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Acknowledgements
 References
 
The authors are grateful to the statistical department at the Tel Aviv University for their assistance. Esther Eshkol is thanked for her editorial assistance.


    Notes
 
3 To whom correspondence should be addressed. Email: intposgr{at}post.tau.ac.il Back


    References
 Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Acknowledgements
 References
 
Botting, B.J., Macdonald Davies, J. and Macfarlane A.J. (1987) Recent trends in the incidence of multiple birth and associated mortality. Arch. Dis. Child., 62, 941–950.[Abstract]

Braithwaite, J.M., Morris, R.W. and Economides, D.L. (1996) Nuchal translucency measurements: frequency distribution and changes with gestation in a general population. Br. J. Obstet. Gynaecol., 103, 1201–1204.[ISI][Medline]

Cuckle, H. (1998) Down's syndrome screening in twins. J. Med. Screening, 5, 3–4.[ISI][Medline]

Cuckle, H. and Shemi, I. (1999) Calculating correct Down's syndrome risks. Br. J. Obstet. Gynaecol., 106, 371–372.[ISI][Medline]

Doyle, P.E., Beral, V., Botting, B.J. et al. (1990) Congenital malformations in twins in England and Wales. J. Epidemiol. Community Health, 45, 43–48.[Abstract]

Herman, A., Maymon, R., Dreazen, E. et al. (1998a) Nuchal translucency audit: a novel image screening method. Ultrasound Obstet. Gynecol., 12, 398–403.[ISI][Medline]

Herman, A., Maymon, R., Dreazen, E. et al. (1998b) Image magnification doesn't contribute to the repeatability of nuchal translucency caliper placing. Ultrasound Obstet. Gynecol., 11, 1–5.[ISI][Medline]

Herman, A., Dreazen, E., Samandarov, A. et al. (2000) On-to-on versus on-to-out nuchal translucency measurements. Ultrasound Obstet. Gynecol., 15, 126–130.[ISI][Medline]

Jauniaux, E., Hertzkovitz, R. and Hall, J.M. (2000) First trimester prenatal diagnosis of a thoracic cyst lesion associated with fetal skin edema. Ultrasound Obstet. Gynecol., 15, 74–77.[ISI][Medline]

Kidd, S.A., Lancaster, P.A., Anderson, J.C. et al. (1997) A cohort study of pregnancy outcome after amniocentesis in twin pregnancy. Paediatr. Perinat. Epidemiol., 11, 200–213.[ISI][Medline]

Koudstaal, J., Bruinse, H.W., Helmerhorst, F.M. et al. (2000) Obstetric outcome of twin pregnancies after in-vitro fertilization: a matched control study in four Dutch University hospitals. Hum. Reprod., 15, 935–940.[Abstract/Free Full Text]

Macintosh, M.C., Wald, N.J., Chard, T. et al. (1995) Selective miscarriage of Down's syndrome fetuses in women age 35 years and older. Br. J. Obstet. Gynaecol., 102, 798–801.[ISI][Medline]

Maymon, R., Dreazen, E., Weinraub, Z. et al. (1999a) Antenatal screening for Down's syndrome in assisted reproductive pregnancies. Hum. Reprod., Update, 5, 530–534.[Abstract/Free Full Text]

Maymon, R., Dreazen, E., Rozinsky, S. et al. (1999b) Comparison of nuchal translucency measurement and second-trimester triple serum screening in twin versus singleton pregnancies. Prenat. Diagn., 19, 727–731.[ISI][Medline]

Moise, J., Laor, A., Arnon, Y. et al. (1998) The outcome of twin pregnancies after IVF. Hum. Reprod., 13, 1702–1705.[Abstract]

Nicolaides, K.H., Sebire, N.J. and Snijders, R.J.M. (1999) The 11–14 week scan: The diagnosis of fetal abnormalities. Parthenon Pub., NY.

Oliviennes, F., Kadhel, P., Rufat, P. et al. (1996) Perinatal outcome of twin pregnancies obtained after in vitro fertilization: comparison with twin pregnancies obtained spontaneously or after ovarian stimulation. Fertil. Steril., 66, 105–109.[ISI][Medline]

Pajkrt, E., Mol, B.W.J., Bleker, O.P. et al. (1999) Pregnancy outcome and nuchal translucency measurements in fetuses with normal karyotype. Prenat. Diagn., 19, 1104–1108.[ISI][Medline]

Pandya, P.P., Hilber, F., Snijders, R.J.M. et al. (1995a) Nuchal translucency thickness and crown-rump-length in twin pregnancies with chromosomally abnormal fetuses. J. Ultrasound Med., 14, 565–568.[Abstract]

Pandya, P.P., Snijders, R.J.M., Johnson, S.J. et al. (1995b) Screening for fetal trisomies by maternal age and fetal nuchal translucency thickness at 10–14 weeks of gestation. Br. J. Obstet. Gynaecol., 102, 957–962.[ISI][Medline]

Pandya, P.P., Altman, D.G., Brizot, M.I. et al. (1995c) Repeatability of measurement of fetal nuchal translucency thickness. Ultrasound Obstet. Gynecol., 5, 334–337.[ISI][Medline]

Räty, R., Virtanen, A., Koskinen, P. et al. (2000) Maternal midtrimester serum AFP and free b-hCG levels in in vitro fertilization twin pregnancies. Prenat. Diagn., 20, 221–223.[ISI][Medline]

Robinson, H.P. and Fleming, J.E.E. (1975) A critical evaluation of sonar crown-rump length measurements. Br. J. Obstet. Gynaecol., 82, 702–710.[ISI][Medline]

Sebire, N.J., Snijders, R.J.M., Huges, K. et al. (1996a) Screening for trisomy 21 in twin pregnancies by maternal age and fetal nuchal translucency thickness at 10–14 weeks of gestation. Br. J. Obstet Gynaecol., 103, 999–1003,[ISI][Medline]

Sebire, N.J., Noble, P.L., Psarra, A. et al. (1996b) Fetal karyotyping in twin pregnancies: selectional technique by measurement of fetal nuchal translucency. Br. J. Obstet. Gynaecol., 103, 887–890.[ISI][Medline]

Sebire, N.J., D'Ercole, C., Hughes, K. et al. (1997a) Increased nuchal translucency thickness at 10–14 weeks of gestation as a prediction of severe twin to twin transfusion syndrome. Ultrasound Obstet. Gynecol., 10, 86–89.[ISI][Medline]

Sebire, N.J., Thornton, S, Huges, K. et al. (1997b) The prevalence and consequence of missed abortion in twin pregnancies at 10–14 weeks of gestation. Br. J. Obstet Gynaecol., 104, 847–849.[ISI][Medline]

Sebire, N.J., Souka, A., Skenton, H. et al. (2000) Early prediction of severe twin-to-twin transfusion syndrome. Hum. Reprod., 15, 2008–2010.[Abstract/Free Full Text]

Sepulveda, W., Sebire, N.J., Hughes, K. et al. (1996) The lambda sign at 10–14 weeks gestation as a predictor of chorionicity in twin pregnancies. Ultrasound Obstet. Gynecol., 7, 421–423.[ISI][Medline]

Sepulveda, W., Sebire, N.J., Kalogeropoulos A. et al. (1997) Evolution of the lambda or twin/chorionic peak sign in dichorionic twin Obstet. Gynecol., 89, 439–441.

Spencer, K. (2000) Screening for trisomy 21 in twin pregnancies in the first trimester using free ß-hCG and PAPP-A combined with fetal nuchal translucency thickness. Prenat. Diagn., 20, 91–95.[ISI][Medline]

Wald, N.J., Cuckle, H., Wu, T.S. et al. (1991) Maternal serum unconjugated estriol and human chorionic gonadotropin levels in twin pregnancies: implications for screening for Down's syndrome. Br. J. Obstet. Gynaecol., 98, 905–908.[ISI][Medline]

Wald, N.J., Watt, H.C. and Hackshow, A. (1999) Integrated screening for Down's syndrome based on tests performed during the first and second trimesters. N. Engl. J. Med., 341, 461–467.[Abstract/Free Full Text]

Submitted on October 2, 2000; accepted on May 24, 2001.