First-trimester screening for trisomy 21 in singleton pregnancies achieved by assisted reproduction

A.W. Liao1, V. Heath1, N. Kametas1, K. Spencer2 and K.H. Nicolaides1,3

1 Harris Birthright Research Centre for Fetal Medicine, King's College Hospital Medical School, London and 2 Endocrine Unit, Clinical Biochemistry Department, Harold Wood Hospital, Romford, Essex, UK


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: The possible effect of assisted reproduction on first-trimester screening for trisomy 21 was examined by fetal nuchal translucency thickness (NT), maternal serum free ß-human chorionic gonadotrophin (HCG) and pregnancy-associated plasma protein-A (PAPP-A). METHODS: Parameters were measured at 11–14 weeks in 411 singleton pregnancies achieved by controlled ovarian stimulation, including 220 that had undergone IVF. Results were compared with 1233 singleton pregnancies conceived spontaneously. RESULTS: In the IVF pregnancies, the median fetal NT was not significantly different from that in controls, whilst the median free ß-HCG was significantly increased, and PAPP-A was significantly decreased. In the intracytoplasmic sperm injection group, fetal NT and free ß-HCG values were not significantly different from those in controls, but PAPP-A was significantly decreased. In those pregnancies achieved by ovarian stimulation, neither fetal NT, free ß-HCG nor PAPP-A were significantly different from the control group. CONCLUSIONS: In IVF pregnancies, screening for trisomy 21 by fetal NT, maternal serum free ß-HCG and PAPP-A levels may be associated with a 1.2% higher false-positive rate than in natural conception.

Key words: free ß-HCG/nuchal translucency/PAPP-A/screening/trisomy 21


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Screening for chromosomal abnormalities in pregnancy can be achieved effectively at 11–14 weeks gestation by a combination of maternal age, and monitoring fetal nuchal translucency thickness (NT) and concentrations of maternal serum free ß-human chorionic gonadotrophin (ß-HCG) and pregnancy-associated plasma protein-A (PAPP-A). This method, which can be carried out in One Stop Clinics for Assessment of Risk (OSCAR), has been shown to identify about 90% of cases of trisomies 21, 18 or 13, sex chromosome aneuploidies and triploidy at a 5–6% screen positive rate (Nicolaides et al., 1992Go; Snijders et al., 1998Go; Spencer et al., 1999Go, 2000aGo, bGo, cGo, dGo; Tul et al., 1999Go; Krantz et al., 2000Go).

In pregnancies achieved with assisted reproduction techniques, the prevalence of fetal abnormalities is not higher than in natural conceptions, except perhaps for sex chromosome aneuploidies (Koulischer et al., 1997Go). However, there is some evidence that second-trimester biochemical screening for trisomy 21 in assisted reproduction pregnancies is associated with higher false-positive results, than in naturally conceived pregnancies, because of an unexplained increase in maternal serum HCG concentration and a decrease in unconjugated oestriol (Maymon et al., 1999aGo). Although one study of 138 pregnancies achieved by IVF reported no significant difference in total HCG and alpha-fetoprotein (AFP) from naturally conceived pregnancies (Muller et al., 1993Go), another two studies of 41 and 67 IVF pregnancies reported that the median serum HCG concentration was 1.52 multiples of the median (MOM) and 1.29 MOM respectively (Heinonen et al., 1996Go; Ribbert et al., 1996Go). Similarly, a study of 151 IVF pregnancies reported that concentrations of unconjugated oestriol were 6% lower, total HCG 14% higher and free ß-HCG 9% higher than in normal pregnancies (Wald et al., 1999Go). Another study of 1632 women who had ovulation induction, and 298 who had IVF, found that total HCG was significantly increased (1.09 MOM) and unconjugated oestriol was decreased (0.92 MOM) (Barkai et al., 1996Go). Similarly, others (Frishman et al., 1997Go) examined 69 women who had IVF reported that total HCG was significantly increased (1.22 MOM) and unconjugated oestriol was decreased (0.90 MOM).

In this study, the possible effect of assisted reproduction on first-trimester screening for trisomy 21 was examined by monitoring fetal NT and concentrations of maternal serum free ß-HCG and PAPP-A.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The Fetal Medicine Centre offers screening for chromosomal abnormalities by a combination of maternal age, fetal NT and concentrations of maternal serum free ß-HCG and PAPP-A at 11–14 weeks of gestation (Spencer et al., 1999Go). Fetal NT was measured by transabdominal ultrasound examination as the maximum vertical distance between the skin and subcutaneous tissues at the back of the neck in a sagittal section of the fetus lying in the neutral position, by sonographers who received The Fetal Medicine Foundation Certificate of Competence in first-trimester scanning (Nicolaides et al., 1992Go; Snijders et al., 1998Go). Concentrations of free ß-HCG and PAPP-A were measured using a KRYPTOR analyser (CIS UK, High Wycombe, Bucks, UK). This is a random continuous access immunoassay analyser, using time-resolved amplified cryptate emission technology (TRACE), which provided results within 20 min of sampling (Spencer et al. 1999Go).

Demographic data, ultrasound findings and the results of biochemical testing were entered into a fetal database at the time of assessment, and from this information patient-specific risks were produced. Outcome follow-up data were obtained from the patients or their doctors. In this retrospective study a search was made of the database to identify all singleton pregnancies (one live fetus and one gestational sac) achieved by assisted reproduction, which had first-trimester screening. A series of spontaneously conceived control pregnancies (three per case) matched for maternal age, ethnic origin and smoking status were similarly identified.

Statistical analysis
All free ß-HCG, PAPP-A and NT measurements were converted to MOM, derived from previous studies (Snijders et al., 1998Go; Spencer et al., 1999Go) of unaffected pregnancies and gestational age calculated from fetal crown–rump length. MOM values for biochemical markers were calculated after correction for maternal weight using the reciprocal weight procedure (Neveux et al., 1996Go) as described previously (Spencer et al., 2000eGo). The non-parametric Mann–Whitney test was used to determine the significance of differences between the groups, because even with logarithmic transformation the data of the smaller subgroups were not normally distributed. Statistical analysis of data was performed with Analyse-It (Smart Software, Leeds, UK) a statistical software add-in for Microsoft Excel 7.

The performance of first-trimester screening for trisomy 21 using fetal NT and maternal serum biochemical markers in spontaneously conceived and IVF pregnancies was assessed using standard statistical modelling techniques (Royston and Thompson, 1997Go). Population parameters for unaffected pregnancies and those affected by trisomy 21 from an earlier study (Spencer et al., 1999Go) were used. Using these parameters, a series of 15 000 random MOM values were selected for each marker from within the distribution of trisomy 21 and the unaffected pregnancies. To simulate IVF pregnancies, the MOM values for PAPP-A and free ß-HCG were adjusted by the percentage changes observed in this study. These values (trisomy 21 and unaffected MOMs; trisomy 21 and IVF adjusted unaffected MOMs) were then used to calculate likelihood ratios which were then used together with the age-related risk for trisomy 21 in the first trimester (Snijders et al., 1999Go), to calculate the expected detection and false-positive rate at a fixed cut-off, in a population with the maternal age distribution of pregnancies in England and Wales (Office of National Statistics, 1997–1999Go).


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
First-trimester assessment was carried out in 411 singleton pregnancies achieved after controlled ovarian stimulation, including 220 that had undergone IVF and 30 that had undergone ICSI. In the naturally conceived pregnancies the percentage of nulliparous women was lower than in those achieved by assisted reproduction (P < 0.0001); there were no other significant differences in demographic characteristics between the two groups (Table IGo).


View this table:
[in this window]
[in a new window]
 
Table I. Demographic characteristics of the study populationsa
 
In the assisted reproduction pregnancies, free ß-HCG, PAPP-A and NT all fitted a Gaussian distribution after log10 transformation (Kolmogorov–Smirnov test, P < 0.001) as was confirmed for unaffected pregnancies (Spencer et al., 1999Go). Compared with the controls, in the IVF pregnancies the median maternal serum free ß-HCG concentration was significantly increased (P = 0.001) and that of PAPP-A was decreased (P = 0.025), whilst in the ICSI-conceived pregnancies median PAPP-A concentration was decreased (P = 0.037), and in the ovarian stimulation group there were no significant differences in either free ß-HCG or PAP-A (Table IIGo).


View this table:
[in this window]
[in a new window]
 
Table II. Comparison of pregnancies achieved by assisted reproduction with those conceived naturally in terms of fetal nuchal translucency thickness (NT) and maternal serum free ß-HCG and PAPP-A at 11–14 weeks gestation
 
In the control group, the false-positive rate at a 1 in 300 first-trimester risk cut-off was 7.0% in this age-biased population. By comparison, in the IVF group the false-positive rate was 15.9% for a similar age distribution. When the differences in the IVF group were used to model the impact of assisted conception on the false-positive rate when screening for trisomy 21 in the general population, the increased maternal serum free ß-HCG and reduced PAPP-A was found to increase the false-positive rate at a 1 in 300 risk cut-off by 1.2%.

In the assisted reproduction group, there were six cases with chromosomal defects (Table IIIGo), 401 cases with normal fetal karyotype or the delivery of phenotypically normal babies, and four cases of miscarriages with unknown fetal karyotype.


View this table:
[in this window]
[in a new window]
 
Table III. Fetal nuchal translucency thickness (NT) and maternal serum free ß-HCG and PAPP-A concentrations in the chromosomally abnormal group from pregnancies achieved by assisted reproduction
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The findings of this study suggest that in pregnancies achieved by IVF and ICSI, the performance of screening for chromosomal defects by a combination of maternal age, fetal NT and maternal serum free ß-HCG and PAPP-A at 11–14 weeks gestation is likely to lead to a higher false-positive rate than in natural conceptions and in those where assisted reproduction is confined to controlled ovarian stimulation. Consequently, in IVF and ICSI pregnancies the invasive diagnostic testing rate—and therefore the risk of miscarriage—would be increased. The demographic characteristics of our assisted reproduction pregnancies and controls were similar, but in controls the incidence of nulliparity was lower. Nevertheless, we have shown previously that parity does not have a significant effect on maternal serum free ß-HCG or PAPP-A (Spencer et al., 2000fGo).

Concentrations of maternal serum free ß-HCG in IVF pregnancies were about 15% higher than in naturally conceived pregnancies. This increase is similar to that reported in the second trimester both for free ß-HCG [9% by Wald et al. (1999)] and for total HCG [29% by Ribbert et al. (1996); 22% by Frishman et al. (1997); 14% by Wald et al. (1999)]. In ICSI pregnancies, maternal serum HCG concentrations in the second trimester are not significantly different from normal, but AFP concentrations are reduced (Lam et al., 1999Go). Similarly, we found that in the first trimester there was no significant change in concentrations of free ß-HCG, but those of PAPP-A were reduced.

Accurate risk estimates in IVF pregnancies undergoing first-trimester screening for chromosomal defects requires correction of the measured concentrations to reduce the overestimate of risk. It was suggested previously that in IVF pregnancies undergoing second-trimester biochemical screening for chromosomal defects the measured MOM should be divided by the median MOM obtained in IVF pregnancies (Wald et al., 1999Go). This procedure was shown to bring the false-positive rate back to the rate observed in naturally conceived pregnancies. Correcting in a similar manner in the current first-trimester study would have also reduced the false-positive rate to that in the control group. However, before such adjustment of risk calculation is considered, it would be necessary to study a large number of trisomy 21 pregnancies achieved by assisted reproduction to determine the effectiveness of screening with adjusted means.

Fetal NT was not affected by assisted reproduction, and this is in agreement with a previous study of 75 singleton pregnancies achieved by assisted reproduction (Maymon et al., 1999bGo). In the current study, NT screening alone would have identified all six chromosomally abnormal fetuses. However, in a multicentre study of 96127 pregnancies, including 326 with trisomy 21, the sensitivity of screening by nuchal translucency was 77% for a false- positive rate of 5% (Snijders et al., 1998Go). Screening by a combination of nuchal translucency and first-trimester biochemistry improves the detection of trisomy 21 to ~90% (Spencer et al., 1999Go, 2000dGo; Krantz et al., 2000Go). It would therefore be reasonable to offer women with pregnancies achieved by assisted reproduction the option of combined screening, despite the small increase (of ~1%) in the false-positive rate. Another option is that in a woman with an IVF pregnancy the measured MOM of free ß-HCG is divided by the median MOM of 1.2, found in this study.


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
This study was supported by a grant from The Fetal Medicine Foundation (Registered charity no: 1037116)


    Notes
 
3 To whom correspondence should be addressed at: Harris Birthright Research Centre for Fetal Medicine, King's College Hospital Medical School Denmark Hill, London SE5 9RS, UK. E-mail: kypros{at}fetalmedicine.com Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Barkai, G., Goldman, B., Ries, L. et al. (1996) Down's syndrome screening marker levels following assisted reproduction. Prenat. Diagn., 16, 1111–1114.[ISI][Medline]

Frishman, G., Canick, J., Hogan, J. et al. (1997) Serum triple marker screening in in vitro fertilization and naturally conceived pregnancies. Obstet. Gynecol., 90, 98–101.[Abstract/Free Full Text]

Heinonen, S., Ryynanen, M., Kirkinen, P. et al. (1996) Effect of in vitro fertilization on human chorionic gonadotropin serum concentrations and Down's syndrome screening. Fertil. Steril., 66, 398–403.[ISI][Medline]

Koulischer, L., Verloes, A. and Lesenfants, S. et al. (1997) Genetic risk in natural and medically assisted procreation. Early Pregnancy, 3, 164–171.[Medline]

Krantz, D.A., Hallahan, T.W., Orlandi, F. et al. (2000) First trimester Down syndrome screening using dried blood biochemistry and nuchal translucency. Obstet. Gynecol., 96, 207–213.[Abstract/Free Full Text]

Lam, Y.H., Yeung, W.S., Tang, M.H. et al. (1999) Maternal serum alpha-fetoprotein and human chorionic gonadotrophin in pregnancies conceived after intracytoplasmic sperm injection and conventional in-vitro fertilization. Hum. Reprod., 14, 2120–2123.[Abstract/Free Full Text]

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 mid-gestation serum screening in assisted reproduction versus naturally conceived singleton pregnancies. Prenat. Diagn., 19, 1007–1011.[ISI][Medline]

Muller, F., Aegerter, P. and Boue, A. (1993) Prospective maternal serum human chorionic gonadotropin screening for the risk of fetal chromosome anomalies and subsequent fetal and neonatal deaths. Prenat. Diagn., 13, 29–43.[ISI][Medline]

Neveux, L.M., Palomaki, G.E., Larivee, D.A. et al. (1996) Refinements in managing maternal weight adjustment for prenatal screening results. Prenat. Diagn., 16, 1115–1119.[ISI][Medline]

Nicolaides, K.H., Azar, G., Byrne, D. et al. (1992) Fetal nuchal translucency: ultrasound screening for chromosomal defects in first trimester of pregnancy. Br. Med. J., 304, 867–889.[ISI][Medline]

Office of National Statistics (1997–1999) Birth Statistics 1996–1998. Series FM1 nos. 25–27. The Stationery Office, London.

Ribbert, L.S.M., Kornman, L.H., De Wolf, B.T.H.M. et al. (1996) Maternal serum screening for fetal down syndrome in IVF pregnancies. Prenat. Diagn., 16, 35–38.[ISI][Medline]

Royston, P. and Thompson, S.G. (1997) Model based screening for risk with application to Down's syndrome. Stat. Med., 11, 257–268.

Snijders, R.J.M., Noble, P., Sebire, N. et al. (1998) UK multicentre project on assessment of risk of trisomy 21 by maternal age and fetal nuchal translucency thickness at 10–14 weeks of gestation. Lancet, 351, 343–346.

Snijders, R.J.M., Sundberg, K., Holzgreve, W. et al. (1999) Maternal age and gestation- specific risk for trisomy 21. Ultrasound Obstet. Gynecol., 13, 167–170.[ISI][Medline]

Spencer, K., Souter, V., Tul, N. et al. (1999) A rapid screening program for trisomy 21 at 10–14 weeks using fetal nuchal translucency, maternal serum free ß-hCG and PAPP-A. Ultrasound Obstet. Gynecol., 13, 231–237.[ISI][Medline]

Spencer, K., Ong, C., Skentou, H. et al. (2000a) Screening for trisomy 13 by fetal nuchal translucency and maternal serum free beta hCG and PAPP-A at 10–14 weeks of gestation. Prenat. Diagn., 20, 411–416.[ISI][Medline]

Spencer, K., Tul, N., Nicolaides, K.H. et al. (2000b) Maternal serum free beta hCG and PAPP-A in fetal sex chromosome defects in the first trimester. Prenat. Diagn., 20, 390–394.[ISI][Medline]

Spencer, K., Liao, A.W., Skentou, H. et al. (2000c) Screening for triploidy by fetal nuchal translucency and maternal serum free ß-hCG and PAPP-A at 10–14 weeks of gestation. Prenat. Diagn., 20, 495–499.[ISI][Medline]

Spencer, K., Spencer, C.E., Power, M. et al. (2000d) One Stop Clinic for Assessment of Risk for fetal anomalies: a report of the first year of prospective screening for chromosomal anomalies in the first trimester. Br. J. Obstet. Gynaecol., 107, 1271–1275.[ISI]

Spencer, K., Ong, C.Y.T., Liao, A.W. et al. (2000e) The influence of ethnic origin on first trimester biochemical markers of chromosomal abnormalities. Prenat. Diagn., 20, 491–494.[ISI][Medline]

Spencer, K., Ong, C.Y.T., Liao, A.W. et al. (2000f) The influence of parity and gravidity on first trimester markers of chromosomal abnormality. Prenat. Diagn., 20, 792–794.[ISI][Medline]

Tul, N., Spencer, K., Noble, P. et al. (1999) Screening for trisomy 18 by fetal nuchal translucency and maternal serum free beta hCG and PAPP-A at 10–14 weeks of gestation. Prenat. Diagn., 19, 1035–1042.[ISI][Medline]

Wald, N.J., White, N., Morris, J.K. et al. (1999) Serum markers for Down's syndrome in women who have had in vitro fertilisation: implications for antenatal screening. Br. J. Obstet. Gynaecol., 106, 1304–1306.[ISI][Medline]

Submitted on February 5, 2001; accepted on April 4, 2001.