14–17 September 2014, Barcelona, Spain Electronic poster abstracts The presence of pain and dyspareunia was obtained by an electronic questionnaire, including questions from the International Consultation on Incontinence Modular Questionnaire (ICIQ) and a pain drawing. Pain-related questions in the ICIQ female sexual matters, vaginal and bowel symptoms module were used. No pain was coded, if the answer was ‘‘not at all’’, ‘‘never’’ or if the reported the pain was not bothersome at all. The pain drawing showed a female body from the front and the back, with the possibility to mark out the following areas to be pain full: pubic symphysis, inguinal area, iliac crest, sacroiliac joints, lower back, coccyx, buttocks and thighs. A cross on the drawing was coded as pain, the absence of cross as no pain. Each area was analysed separately. The ultrasound examiners were blinded to the women’s questionnaire data. Results: There were no differences in pelvic pain and dyspareunia assessed by the ICIQ (table 1) in women with and without major muscle defects. Furthermore, the frequencies of marked areas on the pain drawing did not differ in women with and without defects. Conclusions: Our data suggest that there are no differences in self-reported pelvic pain and dyspareunia in women with and without muscle defects 12 month post-partum. P23.02 Use of ultrasonography and magnetic resonance imaging in the diagnosis of placenta membranacea X. Chen Department of Ultrasonography Diagnosis, Wuhan, China Objectives: To analyze the contribution of ultrasonography and magnetic resonance imaging in the evaluation of placenta membranacea. Methods: This was a prospective study involving two fetuses sus- pected of having placenta membranacea on ultrasound examination. MRI was used to analyze the location of the placenta, and to distinguish the normal placenta and placenta membranacea in a twin pregnancy. All the results were compared with pathology results. Results: A 25-year-old woman and a 24-year-old woman were referred to our unit for abnormal placenta, at 24 weeks and 3 days and 25 weeks and 5 days, respectively. The latter was a twin pregnancy. No obvious abnormality were detected in all the fetuses, however, placenta abnormalities were detected in the first fetus and one of the latter fetuses. Displayed by ultrasonography, the abnormal placenta nearly occupied the whole uterine cavity, and showed diffuse low-level internal echoes inside the placenta. Displayed by MRI, the abnormal placentas showed hyperintense T2-weighted signal and flowing void effect consistent with vascular branches. In the first case, a few normal placenta were seen. In the latter case, normal placenta were not seen in the fetus with placenta membranacea. Postpartum histologic examination revealed chorionic villi directly attached to the fetal membranes in these two cases, consistent with the diagnosis of placenta membranacea. Conclusions: Ultrasonography could be used to display the 2D image and color Doppler image of the placenta membranacea in real time, and the hemodynamics change are very important in the prognosis analysis. MRI are good complementary tool to ultrasonography for identifying the outline of placenta, especially the spatial relationship between normal placenta and placenta membranacea in twin pregnancy. P23.03 Eye tracking as a tool to assess sonographer behaviour when quality-scoring fetal ultrasound images M. Ahmed 2 , C.L. Knight 1,2 , A.T. Papageorghiou 1 , J.A. Noble 2 1 Nuffield Department of Obstetrics & Gynecology, University of Oxford, Oxford, United Kingdom; 2 Institute of Biomedical Engineering, University of Oxford, Oxford, United Kingdom Objectives: Quality-scoring fetal ultrasound images is an important process; there are no objective measures to assess sonographers’ behaviour. Analysing eye movements may provide a quantitative insights into efficient scoring. This study aims to establish if assessors fixate on similar image regions, in a similar sequence, during scoring. Methods: 3 ultrasound experts (one fellow, two 3rd year engineering PhDs) and 3 novices (1st year engineering PhDs) scored 150 fetal abdominal ultrasound images (22-36 weeks’ gestation) using Salomon’s objective criteria. Eye movements were recorded and analysed. Results: Fixations falling on the stomach, umbilical vein, and spine were 14%, 14%, 9% for experts and 12%, 12%, 11% for novices. The mean scoring time was 8.3sec/image for experts and 4.8sec for novices. The most common fixation sequences were stomach-vein-stomach for experts (24%) and vein-stomach-vein for novices (24%). Expert fixation sequences had 16 steps on average, whereas novice fixation sequences had 10. Expert scores were on average 1 point higher than novices. Conclusions: Experts and novices fixate on similar regions. How- ever, experts spend more time assessing images and cross-reference more between structures than novices, thus employing a more thor- ough strategy. The spine plays a significant role in quality scoring, despite not being one of Salomon’s criteria. It is likely that the spine is used sub-consciously for orientation. This first insight into sonographers’ visual search behaviour may prove useful for training and assessment. Supporting information can be found in the online version of this abstract P23.04 Ultrasound measurement of subcutaneous fat thickness as an independent predictor for adverse pregnancy outcomes N. Kennedy, A.E. Quinton, A. Martin, M.J. Peek, R. Nanan Sydney Medical School Nepean, University of Sydney, Sydney, NSW, Australia Objectives: Observational studies suggest body mass index (BMI) is a good predictor for adverse pregnancy outcomes. Experimental studies suggest central adiposity measured via subcutaneous fat thickness (SFT) is a more significant measure of obesity related outcomes. The aim was to measure SFT via ultrasound at 10-14weeks (SFT1) and 18-22weeks (SFT2) to assess as a predictor of adverse pregnancy outcomes. Methods: Prospectively women (n = 1462) were recruited at two time points. SFT was measured at the cervical placenta image. Pregnancies outcomes were collated. Logistic regression modelling assessed SFT performance. Results: SFT1 n = 1462 and SFT2 n = 1363 with outcomes available for n = 1385. Mean (range) SFT was 21.2mm (6.9-73.87) and 20.3mm (7.47-67.97) respectively. 27% of women were overweight and 27% obese. Correlation between BMI and SFT1 was R2 = 0.56 and BMI and SFT2 was R2 = 0.55. Ultrasound in Obstetrics & Gynecology 2014; 44 (Suppl. 1): 181–369. 321