9–12 September 2012, Copenhagen, Denmark Oral communication abstracts OC17.05 Predictive value of fetal brain texture analysis by automatic quantitative MRI analysis in small-for-gestational-age fetuses for an abnormal neonatal neurobehavior M. Sanz-Cortes 1 , G. A. Ratta 1 , F. Figueras 1 , G. Ega ˜ na-Ugrinovic 1 , N. Bargall ´ o 2 , E. Gratac ´ os 1 1 Maternal-Fetal Medicine Department, ICGON, Hospital Clinic, Universitat de Barcelona; Fetal and Perinatal Medicine Research Group, Institut d’Investigacions Biomediques August Pi i Sunyer (IDIBAPS), Barcelona, Spain; 2 Department of Radiology, Hospital Clinic, Universitat de Barcelona, Barcelona, Spain Objectives: We tested the hypothesis whether a semiautomated method for automatic quantification of fetal magnetic resonance imaging (MRI) images based on texture analysis (TA) could identify patterns associated with an abnormal neurobehavior in small for gestational age (SGA) neonates. Methods: Ultrasound and MRI were performed on 91 SGA fetuses at 37 weeks of GA. Delineation of five brain regions from MRI scans was performed: bilaterally in two levels of the frontal lobe, basal ganglia and mesencephalon and in one single area in cerebellum. All neonates underwent a neonatal behavioral assessment scale (NBAS) test at 42 ± 1 weeks. Scores from NBAS test were classified as abnormal (cases) if ≥ 1 areas were < 5 th centile and normal (control) if all areas were > 5 th centile. Textural features were obtained from all the delineated areas using AQUA software. A classification algorithm based on feature selection of individual regions and discriminant analysis was used. The ability of the textural features from the different ROIs to predict neonatal neurobehavioral performance was analyzed using machine learning methods. Results: Based on NBAS scores, our sample was classified into 42 cases and 49 controls. Neither maternal clinical characteristics nor perinatal outcomes or age at NBAS test differed significantly between the two groups. The estimated accuracies to predict an abnormal neurobehavioral performance based on TA were 95.12% for the frontal lobe, 95.56% for basal ganglia, 93.18% for Mesencephalon and 83.33% for cerebellum. Conclusions: Fetal MRI brain TA is a potentially useful tool to predict those SGA fetuses that are at risk for an abnormal neurodevelopment. OC17.06 Increased fetal brain perfusion and neonatal neurobehavioral performance in the general population R. Mula 1 , A. Arranz 1 , S. Savchev 1 , F. Botet 3 , C. Costas 2 , E. Gratac ´ os 1 , F. Figueras 1 1 Maternal-Fetal Medicine Department, ICGON, Fetal and Perinatal Medicine Research Group, IDIBAPS, Hospital Clinic, Universitat de Barcelona, Barcelona, Spain; 2 Psychology Department, Universitat Aut ` onoma de Barcelona, Barcelona, Spain; 3 Neonatology Department, ICGON, Fetal and Perinatal Medicine Research Group, IDIBAPS, Hospital Clinic, Universitat de Barcelona, Barcelona, Spain Objectives: To explore in normal fetuses the association between fetal brain perfusion and middle cerebral artery (MCA) Doppler at third trimester with neonatal neurobehavioral performance. Methods: Frontal brain perfusion that was measured by fractional moving blood volume (FMBV) and middle cerebral artery (MCA) Doppler pulsatility index (PI) were assessed in 260 consecutive healthy fetuses at routine third trimester scan (30–34 weeks). Neonates were evaluated with the Neonatal-Behavioral-Assessment- Scale (NBAS), which evaluates 35 items clustered in 5 domains: social, motor, state regulation, state organization and attention. The association between Doppler parameters and neurobehavior were analyzed by MANCOVA and logistic regression, with adjustment for socio-economic class, ethnicity, parity and smoking. Results: Fetuses with increased FMBV (in the upper quartile) had lower neurobehavioral scores in all areas, reaching significance in social (6 vs. 6.4; p 0.025), motor (5.6 vs. 5.8; p 0.041) and attention (5.3 vs. 5.9; p 0.005). Fetuses with increased FMBV had higher risk of abnormal (< 10 th centile) social (OR 2.9 [95CI% 1.33–6.5]), motor (OR 3.3 [95% CI 1.36–8.1]) and attention (OR 2.5 [IC95% 1.1–5.8]) scores. Fetuses with lower MCA PI (in the lower quartile) did not differ in their neurobehavioral scores from those with normal values. Conclusions: Fetuses with increased frontal brain perfusion have poorer neurobehavioral competences, suggesting a disrupted neurological maturation. OC17.07 Placental specific microRNAs expression profile in placental insufficiency: correlation with severity of the disease and Doppler-determined umbilical artery PI I. Hromadnikova 1 , K. Kotlabova 1 , V. Novotna 2 , A. Kestlerova 2 , L. Krofta 2 1 Department of Molecular Biology and Cell Pathology, Charles University, Third Faculty of Medicine, Prague 10, Czech Republic; 2 Institute for the Care of the Mother and Child, Prague, Czech Republic Objectives: To assess expression profile of placental specific microR- NAs in placental insufficiency related complications. Placental spe- cific microRNA gene expression was correlated with severity of the disease with respect to the degree of clinical signs and requirements for the delivery (before and after 34 week) and Doppler ultrasound examination of a. umbilicalis pulsatility index. Methods: Relative quantification of placental specific microRNAs (miR-512-5p, miR-515-5p, miR-516-5p, miR-517 * , miR-518b, miR-518f * , miR-519a, miR-519d, miR-519e, miR-520a * , miR- 520 h, miR-524-5p, miR-525, miR-526a and miR-526b) was determined in placentas derived from 25 normal pregnancies and 37 complicated pregnancies (14 pre-eclampsia, 8 pre-eclampsia with IUGR, 7 IUGR, 6 SGA and 2 gestational hypertension) using real- time PCR. The cohort involved 17/12 patients with early/late onset of the disease, 6/16 patients with mild/severe pre-eclampsia and 10 patients with abnormal Doppler. Results: MiR-517 * , miR-519a, miR-520a * and miR-525 gene expression was significantly decreased in placental insufficiency when compared with control cohort. MiR-517 * represented a spe- cific marker for both early and late onset of the disease, mild and severe pre-eclampsia and correlated with a umbilicalis PI in placental insufficiency (ρ =−0.719, P = 0.003). MiR-519a was identified as a specific marker for early onset of the disease. MiR-520a * gene expression was significantly decreased in placental insufficiency with the onset before and after 34 weeks, mild and severe pre-eclampsia and SGA. MiR-525 seemed to be a marker for late placental insuffi- ciency, mild pre-eclampsia and SGA. MiR-516-5p was shown as an unique marker for SGA. Conclusions: This is the first study identifying gene expression profile of placental specific microRNAs in placental insufficiency related complications. Further studies will focus on the evaluation of diagnostical potential of extracellular miR-517 * , miR-520a * and miR-525 markers. Supported by GACR P304/12/1352. Ultrasound in Obstetrics & Gynecology 2012; 40 (Suppl. 1): 1–54 35