19th World Congress on Ultrasound in Obstetrics and Gynecology Oral poster abstracts from the vertical position of the occipitofrontal diameter axis. The operators were blinded to their measurements; the 60 3D blocks were analysed twice by each operator. Mean difference and 95% limits of agreement between paired measurements of fetal NT by the same and by two different sonographers were determined. Results: In the mid-sagittal plane the median NT was 1.8. The median NT at 5 ◦ to 25 ◦ was significantly lower (−0.1, −0.3, −0.4, −0.5, −0.6 mm) than in the mid-sagittal plane. The 95% limits of agreement for NT in the mid-sagittal section were −0.1 to 0.3 mm and −0.1 to 0.5 mm for both operators. At 15 ◦ for delta NT, 95% of the cases were between −0.2 to 0.7 mm and −0.3 to 0.5 mm, respectively. Between operators it was −0.1 to 0.2 in the mid-sagittal section and −0.1 to 0.4 at a deviation of 15 ◦ . Conclusion: Fetal NT measurement depends on the position of the fetal head. The measurement is biggest in the mid-sagittal plane, which is clearly defined and smaller in lateral planes. OP02.04 A NTMoM and DVPIVMoM based model to predict outcome in fetuses at increased first trimester risk: a promising method to reduce false positive rates E. Timmerman , E. Pajkrt, K. Oude Rengerink, C. Bilardo Obstetrics & Gynecology, Academic Medical Centre, Amsterdam, Netherlands Objective: To evaluate whether a model based on NT and DVPIV can reduce the false positive rate (FPR) of first trimester screening. Methods: Review of prenatal database for all cases with increased risk at the combined test and known DVPIV. Logistic regression was used to build a maternal age-, NT- and DVPIV-based model to predict chromosomal anomalies. Splines were used to investigate whether the influence of the continuous variables in the model was linear. Results: Linearity was not proven for maternal age < 30 years. Therefore 369 cases with maternal age above 30 years were included, of which 193 (52,3%) had a DVPIV >P95. An increased DVPIV entailed a relative risk for chromosomal anomalies of 3,7 (51 vs 14%). The following model was built using logistic regression analysis: predicted probability (chromosomal anomaly) = 1/(1 + exp − (−7,26 + (0,44 × NTMoM) + (1,21 × DVPIVMoM) + (0,10 × age)). The area under the ROC curve of the model used for the prediction of chromosomal anomalies was 0,80. Furthermore the model could indentify 10% of fetuses without chromosomal anomalies and thereby reduce FPR. Conclusions: In a population of fetuses with increased risk for Down syndrome after first trimester screening the combination in a binary logistic regression model of NT and DVPIV improves the prediction of chromosomal anomalies and reduces FPR. Although measurement of DV is well established as second stage screening for chromosomal anomalies, this is the first model combining NT and DVPIV as continuous variables. Supporting information can be found in the online version of this abstract. OP02.05 Performance of first trimester combined screening test for trisomy 21 and trisomy 13/18 with the double test taken before or after 10 weeks in 44,537 singleton pregnancies O. B. Petersen 1 , I. Kirkegaard 3 , A. Holmskov 4 , N. Uldbjerg 1 , N. Tørring 2 1 Fetal Medicine Unit, Aarhus University Hospital Skejby, Aarhus N, Denmark; 2 Clin Biochemistry, Aarhus University Hospital Skejby, Aarhus N, Denmark; 3 Obstetrics and Gynecology, Horsens Hospital, Horsens, Denmark; 4 Obstetrics and Gynecology, Viborg Hospital, Viborg, Denmark Objective: To compare the performance of two-step first trimester combined screening for trisomy 21 and trisomy 13/18 with biochemistry taken early (week 8 + 0 to 9 + 6) or late (week 10 + 0 to 13 + 6) in a large cohort of singleton pregnancies. Methods: The study includes 44 537 singleton pregnancies with complete first trimester screening data from November 2003 to March 2009. All cases of prenatal or postnatal aneuploidy were identified through the Danish national cytogenetic register. The double test was taken between week 8 + 0 and 13 + 6, in the majority of cases by the GP, mailed to the laboratory and analyzed prior to the NT scan. Or taken at the day of the NT scan. All scans were performed, or supervised by FMF-certified sonographers. Risk calculations were performed using Astraia Fetal Database. Results: 110 out of 124 cases of trisomy 21 were diagnosed in the first trimester screening (detection rate 88,7%). When the double test was taken before week 10 + 0, the detection rate of trisomy 21 was 95,9% (70 out of 73), whereas 78,4% were detected (40 out of 51) with the double test taken week 10 + 0 or later (P = 0.0007). For trisomy 13 and 18, 65,0% (13 out of 20) were detected before 10 th gestational week, and 73,3% (11 out of 15) after (N.S). The screen positive rate and the maternal age were similar before and after week 10. Conclusion: Screening for fetal trisomy can be performed with good results with the blood sample taken as early as the 8 th week of gestation. When the blood sample was taken before week 10, the detection rate of trisomy 21 was significantly improved, compared to after week 10, with no difference in the detection rate of trisomy 13/18 with biochemistry taken early or late in 1 st trimester. OP02.06 Feasibility of a contingent test completed in the first trimester to screen for trisomy 21 A. Borrell 1 , M. Munoz 1 , M. Arigita 1 , A. Gonc´ e 1 , A. Seres 1 , G. Falguera 2 , D. Guix 3 , R. Clavera 3 , R. Espel 4 , E. Baldrich 4 , A. Rodriguez-Veret 5 , A. Torrent 6 , E. Quesada 7 , R. Garcia-Moreno 8 , A. Acera 9 , E. L. Lopez-Quesada 10 , M. Degollada 11 1 Obstetrics & Gynecology, Hospital Clinic Barcelona, Barcelona, Spain; 2 SRHCC, Central Region, Catalan Health Institute, Terrassa, Spain; 3 SRHCC, Granollers, Granollers, Spain; 4 SRHCC, Sabadell, Sabadell, Spain; 5 Obstetrics & Gynecology, Consorci Sanitari de Terrassa, Terrassa, Spain; 6 SRHCC, Mollet, Spain; 7 SRHCC, Bages-Bergueda-Solsones, Manresa, Spain; 8 SRHCC, Osona, Vic, Spain; 9 SRHCC, Cerdanyola, Cerdanyola, Spain; 10 Obstetrics & Gynecology, Hospital Mutua de Terrassa, Terrassa, Spain; 11 SRHCC, Anoia, Igualada, Spain Objective: To study the feasibility of a contingent test completed in the first trimester, in which the combined test was performed locally as a first step, and secondary ultrasound markers were assessed in a tertiary referral center as a second step. 66 Ultrasound in Obstetrics & Gynecology 2009; 34 (Suppl. 1): 62–176