Placenta (2003), 24, S99–S103 doi:10.1016/S0143-4004(03)00178-4

Prenatal Diagnosis after ART Success: The Role of Early Combined Screening Tests in Counselling Pregnant Patients L. Ghisoni a, E. Ferrazzi a,c, C. Castagna a, P.E. Levi Setti b, A.C. Masini a and A. Pigni a a b

Department of Obstetrics and Gynecology, DSC L. Sacco Universita` degli Studi di Milano, 20157 Milan, Italy; Reproductive Medicine Unit, Humanitas Clinical Institute, Milan, Italy

Paper accepted 30 June 2003

First-trimester Down syndrome screening may cause a higher false positive rate in pregnant patients who have undergone ART (assisted reproductive technologies). The aim of this paper is to contribute to this analysis with the second largest series of combined biophysical and biochemical tests in the first trimester of pregnancy after ART. One hundred and forty-two singleton successful ART pregnancies were selected for this study: 50 pregnancies induced by using in-vitro fertilization (IVF), and 92 using intracytoplasmic sperm injection (ICSI). Each patient was matched with three naturally conceived pregnancies based on maternal age and gestational age. Free beta-HCG and PAPP-A were measured on dried blood spots and converted to MoMs. Nuchal translucency (NT) was measured by certified operators. Mean maternal age was 334. NT, free beta-HCG and PAPP-A values of the control cases were not significantly different from local standards evaluated on 3043 cases. NT between ART pregnancies and matched controls was not significantly different. PAPP-A was reduced but not significantly lower in ART pregnancies. Free beta-HCG was the only analyte that resulted in significantly higher values in ART pregnancies (1.12 MoM) versus controls (0.99 MoM). No significant differences were found for biochemical values observed between ICSI and IVF patients. The screen positive rates observed in ART and control pregnancies were 5.5 per cent and 4.6 per cent respectively NT measurements were not affected by ART pregnancies. Our results (non-significant lower values of PAPP-A and significantly higher free beta-HCG values) were consistent with other reported series. The increase in the screen positive rate determined by these biological variations was not greater than 0.9 per cent. This higher false positive rate has a negligible impact on counselling ART patients. The algorithm used to calculate the relative risk after the combined tests should not be changed until the detection rate of trisomies in ART pregnancies is not fully disclosed by larger series. Placenta (2003), 24, S99–S103  2003 Elsevier Ltd. All rights reserved.

INTRODUCTION There is a general consensus that the prevalence of aneuploidies is not significantly higher after successful assisted reproductive techniques (ART) (Koulischer et al., 1997). The debate is still open as far as more subtle genetic problems are concerned, especially after introcytoplasmatic sperm injection (Van Steirteghem, 2001). However, the clinical issue that prenatal counselling must address with these patients is mainly determined by the advanced maternal age of the patient at the time of the ART procedures. In addition to this, the patient should also be informed that even younger women can benefit by the high detection rate of screening tests for the risk of the most frequent trisomies, namely trisomy 21. Mid-trimester biochemical screening proved to cause a higher false positive rate in ART pregnancies (Heinonen et al., 1996; Ribbert et al., 1996). Similar findings have been reported c To whom correspondence should be addressed. Tel.: +390239042818; Fax: +39023565061; E-mail: enrico.ferrazzi@ unimi.it

0143-4004/03/$–see front matter

more recently by studies that investigated the results obtained with the analytes used in first trimester screening protocols, such as free-beta-HCG and PAPP-A (Liao et al., 2001; Wojdemann et al., 2001; Maymon and Shulman, 2002; Orlandi et al., 2002). It is very likely that for some of these papers variations of the results can be explained due to the limited number of cases studied (Wojdemann et al., 2001; Maymon and Shulman, 2002; Orlandi et al., 2002). However, even in the largest series published so far variations were still found between the results observed in pregnancies after in vitro fertilization and after intracytoplasmatic sperm injection (Liao et al., 2001). In this latter paper, Liao and co-workers found a significant increase in free beta-HCG, and a significant decrease of PAPP-A in IVF pregnancies, whereas in ICSI pregnancies a significant difference was not observed in both of these analytes. Biological explanations for these discrepancies are not currently available. Simple ovulation induction versus the more aggressive multiovulation protocols for IVF may play an important role in the observed variations of the analytes as hypothesized by Wojdemann. (Wojdemann et al., 2001).  2003 Elsevier Ltd. All rights reserved.

S100

Placenta (2003), Vol. 24

Table 1. Crown rump length, nuchal translucency, free Beta-HCG and PAPP-A are reported and compared for successful ART pregnancies, matched controls and local standards. See text for statistical comparison

C.R.L. (mm) N.T. (mm) free B-HCG (MoM) PAPP-A (MoM)

Local standards

P

Matched controls no. 435

P

ART no. 145

mean s.d. mean s.d. mean

60.3 8.9 1.5 0.4 1.00

0.9

60.1 8.9 1.5 0.4 0.99

0.6

59.7 9.8 1.4 0.3 1.12

s.d. mean s.d.

0.53 1.03 0.62

0.9 0.2

0.53 1.02 0.60

0.9

0.3* 0.02 0.5

0.67 0.98 0.61

* t-test performed after logarithmic transformation of the observed values.

Whereas until now the different biological background between IVF and ICSI is of difficult interpretation. In addition to this, different laboratory methodologies can introduce other variables (Liao et al., 2001; Orlandi et al., 2002). As a consequence of these discrepancies, counselling these patients becomes more difficult because of a supposed increase of the false positive rate. Nuchal translucency measurements (NT) provided more stable data which were not significantly different from controls. According to these findings nuchal translucency measurements may provide a more reliable risk assessment. However, since the combination of NT and biochemical measurements could provide a higher detection rate (Spencer et al., 1999; Krantz et al. 2000), the clarification of the relative and combined role of biochemical analytes remains of utmost importance. The aim of this paper is to contribute to this discussion with the second largest series, reported so far, of combined biophysical and biochemical tests in the first trimester of pregnancy.

control group consisted of 429 uneventful, spontaneous pregnancies. Three thousand and forty-three cases collected in the same period, in which NT measurements were performed by certified operators were used as local reference values. Methods Each maternal blood sample was collected and dried as spots on filter paper and analysed for free beta-hCG and PAPP-A by enzyme-linked immunosorbent assays (ELISA) (Schleicher and Shull, Keane, NH, USA). All biochemical values were converted to the multiple of the median (MoM) for these biochemical markers according to the medians calculated on 10 000 unaffected pregnancies (Krantz et al., 2000). All nuchal translucency (NT) measurements were performed by certified operators according to the protocol of the Fetal Medicine Foundation. Statistical analysis

PATIENTS AND METHODS Patients As a part of our ongoing prospective study (January 2000 to January 2003) of first trimester aneuploidies screening in ART successful pregnancies, 142 singleton consecutive pregnancies were selected for this study. Fifty pregnancies by in vitro fertilization (IVF), and 92 by intracytoplasmic sperm injection (ICSI) were identified. The whole series was collected from the ART program of the Reproductive Medicine Unit, Humanitas Clinical Institute, and from the Prenatal Diagnostic Centre at the Department of Obstetrics and Gynaecology of the Sacco University Hospital of Milan. In this series only one fetus was affected by chromosomal aneuploidy, i.e. trisomy 21. This data was not included in the comparison of the ART series with matched controls and local standards. Each patient was matched with three naturally conceived pregnancies based on maternal age and gestational age. This

Biochemical MoM values and biophysical measurements of control cases were tested for their values versus the whole population of 3500 screened in the same laboratory during the same time period. Biochemical and biophysical measurements were then compared between cases (ART pregnancies) and controls (matched spontaneous pregnancies). A sub analysis was then performed for IVF cases and ICSI cases. The screen positive rate in cases (ART pregnancies) was compared to the screen positive rate in controls and in the whole population.

RESULTS Mean maternal age, both in ART and controls was (334). NT, free Beta-HCG and PAPP-A were not significantly different at the level of 0.05 between the overall local population and the matched controls selected for this study (Table 1, first vs second column). Figure 1 shows the NT values observed

Ghisoni et al.: Prenatal Diagnosis after ART Success

S101

Figure 1. Nuchal translucency measurements vs crown rump length in ART pregnancies (full circle), matched controls (empty circles), local standards (dots). Linear regressions for ART cases (heavy black line), controls and local standards (black lines).

between ART pregnancies and controls. The two regressions were compared and were not significantly different (P=0.09). Similarly NT findings compared for their mean values after semi-logarithmic transformation were not significantly different for both tests (P=0.07). Similar findings were observed for PAPP-A values between ART and controls (P=0.5). The only significant difference between ART and controls was found for the free beta-HCG values which were significantly higher than controls (P