Multivariate analysis identifies the estradiol level at ovulation triggering as an independent predictor of the first trimester pregnancy-associated plasma protein-A level in IVF/ICSI pregnancies

STUDY QUESTION: Can independent predictors of pregnancy-associated plasma protein-A (PAPP-A) levels be identified in a group of women who conceived following IVF/ICSI? SUMMARY ANSWER: The significantly decreased PAPP-A level in IVF and ICSI pregnancies compared with non-IVF/ICSI pregnancies was correlated strongly with the serum estradiol (E2) level at ovulation triggering. WHAT IS KNOWN ALREADY: The first trimester prenatal combined screening test for fetal aneuploidies in pregnancies conceived following assisted reproduction techniques (ART) is complicated by an alteration of the maternal biomarkers free β-hCG and PAPP-A, causing a higher false-positive rate compared with pregnancies which are conceived naturally. The use of controlled ovarian stimulation prior to IVF/ICSI is suggested to be the principle reason for these alterations of biomarkers in ART pregnancies. STUDY DESIGN, SIZE, DURATION: Between January 2010 and December 2011, 1474 women who conceived naturally and 374 women who conceived following IVF (n = 89), ICSI (n = 204) or intrauterine insemination (IUI, n = 81) were included in this retrospective study. Only singleton pregnancies were eligible for this study. For all women, serum analysis was performed in the same clinical laboratory. Measurement of nuchal translucency (NT) thickness was performed by four physicians belonging to the same infertility centre. PARTICIPANTS/MATERIALS, SETTING, METHODS: First-trimester combined screening test of aneuploidy parameters (maternal age, PAPP-A and free β-hCG, NT thickness) were compared between non-ART and ART (IVF, ICSI and IUI) singleton pregnancies. Next, a minimal threshold E2 level at ovulation triggering was suggested for IVF/ICSI pregnancies above which the PAPP-A levels were significantly decreased compared with non-ART pregnancies. Finally, a multivariate analysis was performed to reveal independent predictors of PAPP-A level in IVF/ICSI pregnancies. MAIN RESULTS AND THE ROLE OF CHANCE: We showed a decrease of the multiple of the median (MoM) PAPP-A level in IVF and ICSI singleton pregnancies compared with non-ART singleton pregnancies (P < 0.001), with MoM values of 0.74 (0.16-3.16) and 0.81 (0.12-4.61) versus 0.98 (0.14-5.76), respectively. Analysis of variance of the overall model was highly significant (Fisher test 3.76, P = 0.01), indicating that the model explains a significant portion of the variation in the data. No difference in PAPP-A level was found between non-ART and IUI pregnancies. The free β-hCG level and NT thickness did not differ between ART and non-ART pregnancies. PAPP-A levels in IVF and ICSI pregnancies were strongly correlated with the E2 level at ovulation triggering. We showed by multivariate analysis that an E2 cut-off level of 1300 pg/ml at the time of ovulation could predict a significantly lower PAPP-A level at first trimester combined screening (β -0.239 ± 0.088, P < 0.005). LIMITATION, REASONS FOR CAUTION: The measures of biochemical markers can differ between laboratories and with the used equipment; therefore, extrapolation of the E2 cut-off level to other infertility centres should be undertaken with caution. WIDER IMPLICATIONS OF THE FINDINGS: One should be careful when using correction factors for ART patients undergoing the first trimester combined screening test. The proposed E2 cut-off level may help to identify a subgroup of women within the population of ART patients for whom use of a correction factor is justified. STUDY FUNDING/COMPETING INTEREST(S): None.