20th World Congress on Ultrasound in Obstetrics and Gynecology Short oral presentation abstracts OP19.06 How accurate are we in determining the placental cord insertion site in multiple gestations? L. E. Romine 1 , M. Hwang 5 , N. Trivedi 2 , J. Wan 4 , Y. LaCoursiere 2 , P. Do 4 , K. Benirschke 3 , D. Schrimmer 2 , D. H. Pretorius 1 1 Radiology, University of California, San Diego, CA, USA; 2 Reproductive Medicine, University of California, San Diego, CA, USA; 3 Pathology, University of California, San Diego, CA, USA; 4 School of Medicine, University of California, San Diego, CA, USA; 5 School of Medicine, Vanderbilt University, Nashville, TN, USA Objectives: To determine accuracy of ultrasound (US) in classifying placental cord insertion (PCI) site in multiple gestations at a tertiary care center. Methods: A retrospective study of multiple gestations delivered from 2/2003 to 7/2009 was performed. Data included gestational age (GA) at PCI identification, use of in-vitro fertilization (IVF), and PCI characterization by both US and pathology. The distance from PCI to placental edge was measured. To correlate US with pathology data, PCI was classified as central/eccentric, marginal (≤ 1 cm from placental edge), or velamentous (into membranes). Statistical analysis was performed with Chi-square comparison amongst categorical variables. US images were reviewed in discordant cases to determine causality. Results: Twins and triplets were analyzed. 638 fetuses (302 pregnancies) met criteria. Mean GA at PCI evaluation was 20.9 wks, median 19.0. Using pathology results as gold standard, US correctly identified PCI as central/eccentric, marginal, or velamentous in 69.3% (P < 0.01). US was more accurate for diamniotic-dichorionic twins (73.8% of 382) than diamniotic-monochorionic (59.4% of 155, P = 0.002). US correctly classified 97.5% of central/eccentric, 8.5% of marginal, and 6.1% of velamentous PCIs (438, 118, and 82 cases). Conversely, pathology confirmed US diagnosis of central/eccentric in 71.4%, marginal in 32.3%, and velamentous in 11.1% (of 598, 31, and 9). Comparing normal (central/eccentric) to abnormal (marginal/velamentous), US was less accurate in identifying abnormal (7.5% of 200) than normal (97.5% of 438) PCIs (P < 0.01); no difference was seen for twins (71.9% of 549) vs. triplets (73.5% of 89, P > 0.05). No difference was found in occurrence of velamentous PCI in IVF vs. non-IVF cases (12.5% for both). Conclusions: US classification of PCI was often inconsistent with pathological results in multiple gestations. Discrepant cases were evaluated and suggest evolution/peripheral infarction of the placenta with apparent evolution of the PCI in many instances. OP19.07 Cervical length changes and the risk of preterm delivery in twin pregnancies T. Ghi, M. Kuleva , E. Maroni, A. M. Youssef, F. Guasina, F. Giorgetta, T. Arcangeli, F. Bellussi, E. Contro, G. Pilu, G. Pelusi U.C. Ostetricia e Ginecologia, University of Bologna, Bologna, Italy Objectives: To assess if longitudinal cervical changes may predict the risk of preterm delivery in twin pregnancies Methods: Cervical length was measured at rest in supine position at 20 and 24 weeks in a cohort of twin pregnancies. These two measurements and their difference were then compared between patients who had spontaneous onset of labor < 32 weeks (group 1) or ≥ 32 weeks of gestation (group 2). Cases where early preterm delivery was indicated due to a maternal or fetal indications were excluded. Results: Overall 84 patients were included in our study. The mean gestational age at delivery was 35.5 ± 1.7 weeks, with 6 (7%) patients delivered prior to 32 weeks. The mean cervical length was significantly lower in group 1 vs. group 2 at 20 (30.8 ± 9.5 mm vs. 37.1 ± 7.2 mm, P = 0.046) and at 24 weeks (24.3 ± 10.1 mm vs. 34.3 ± 7.1 mm, P = 0.002). In addition, cervical length shortening in group 1 was greater in comparison with group 2 in the 20–24 weeks interval (7.5 ± 7.5 mm vs. 2.8 ± 4.3 mm, P < 0.017). Conclusions: Cervical length shortening between 20–24 weeks seems to be considerably greater among twin pregnancies undergoing spontaneous onset of labor prior to 32 weeks. OP19.08 Effect of prophylactic progesterone on incidence of preterm labour in spontaneous twin pregnancy, randomized controlled study A. Z. Elsheikhah , S. Dahab, S. Negm, A. Ebrashy, M. Momtaz Cairo University, Cairo Fetal Medicine Unit, Cairo, Egypt Objectives: To study the role of prophylactic progesterone versus placebo in prevention of preterm labour in twin pregnancy. Methods: One hundred women pregnant in twins were enrolled for this study. They were divided in 2 groups. Group (1) 50 pregnant women received vaginal progesterone 200 mg daily for 10 wks (from the 24 th to the 34 th week of gestation). Group (2) 50 pregnant women received placebo for the same duration. Transvaginal ultrasound measurement of the cervical length was done at 18–24 weeks gestation. patients had a first trimester and second trimester detailed scan using Medison Accuvix XQ machine (Korea). Patients with anomalies were excluded. Results: Patients demographic details, age, parity, previous risk factors were not statistically different between the 2 groups. The mean cervical length was 34 mm in the progesterone and 36 mm in the placebo group. Mean gestational age at delivery was 35.7 weeks with a mean birth weight of 2160 grms in the progesterone group which is not significantly different from the placebo group mean gestational age 35.6 weeks with gestational weight of 2210grms. Foetal complications or neonatal ICU admissions were not different in the 2 groups. Conclusions: This study showed that the prophylactic vaginal progesterone administration has no significant value in the prevention of spontaneous preterm labour in twin pregnancy. OP19.09 Spontaneous version of fetal presentation in twin pregnancies during third trimester J. W. Park , S. Y. Kim, E. S. Im, C. W. Park, J. S. Park, J. K. Jun, H. C. Syn Obstetrics & Gynecology, Seoul National University Hospital, Seoul, Republic of Korea Objectives: Vaginal delivery of a non-vertex presenting twin is not recommended. Our objective is to evaluate changes in fetal presentation of twin pregnancies and to counsel mothers with evidence. Methods: We performed a retrospective analysis of the pattern of spontaneous version in twin pregnancies from early third trimester to birth. We identified a total of 310 eligible women of live-born twin pregnancies in Seoul National University Hospital during 1999–2009. Gestational age was divided in two intervals: early- 3 rd trimester (27–30 weeks); and mid-3 rd trimester (31–34 weeks). We analyzed change of fetal presentation between two intervals and compared the pattern of change of presentation during antepartum (divided in 4 groups; group [1]–[4]) and delivery. Results: The first twin’s presentations at delivery according to 4 groups are described below (Table 1). The chances of vertex presenting twin at delivery differ according to whether an individual presenting twin was vertex or not at early-3 rd trimester, even if 108 Ultrasound in Obstetrics & Gynecology 2010; 36 (Suppl. 1): 52–167