21st World Congress on Ultrasound in Obstetrics and Gynecology Oral communication abstracts and yolk sac were recorded in a database. MSD and YSD were cal- culated as the average of the three measurements. Outcome at time of the nuchal scan and final pregnancy outcomes were subsequently recorded. FLDA was performed on a restricted GA range in order to have measurements from viable pregnancies and miscarriages in the upper range to study growth of CRL, MSD and YSD in pregnancies that were viable and in the ones that miscarried by 12 weeks GA. Results: 90 pregnancies resulting in a live birth which had CRL (n = 305), MSD (n = 285) and YSD (n = 234) measurements and 24 pregnancies that resulted in a miscarriage with measurements of CRL (n = 31), MSD (n = 40) and YSD (n = 26) were included. Restricted GA ranges were: from 32 to 91 days for CRL and from 32 to 84 days for MSD and YSD. FLDA of CRL and YSD discriminated viable pregnancies from miscarriages from 45 days GA onwards. Although some miscarriages showed a normal MSD growth, the discrimination between a viable pregnancy and a miscarriage was clear with FLDA in the whole GA range studied. Conclusions: The ability to discriminate viable pregnancies from miscarriages with FLDA of CRL, MSD and YSD was shown in a population with known conception dates. OC04.03 Expectant management of delayed miscarriage – are we setting ourselves up to fail? S. Merritt 1 , M. McVeigh 1 , P. Seed 2 , J. Hamilton 1 1 Emergency Gynaecology Unit, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom; 2 Department of Women’s Health, King’s College London, London, United Kingdom Objectives: To examine: the uptake of and drop out rate from expectant management of delayed miscarriage; which ultrasound parameters may predict success; when follow up should occur to give the best chance of complete miscarriage. Methods: A retrospective analysis of computerised databases in a tertiary referral Early Pregnancy Unit from January 2007 – December 2009. Parity/gravidity and ultrasound parameters (mean sac diameter, presence of yolk sac, presence of embryo and crown – rump length) were examined. Results: 316 women began expectant management, 65 were excluded with incomplete miscarriage or lost to follow up. Of the remaining 251, 90 (35.9%) opted out of expectant management. Median time to opting out was 8 (range 1 – 26) days. 55/130 (42.3%) who continued completed their miscarriage by 14 days, a further 68/130 (52.3%) did so by 28 days. The median time to completed expectant management was 16 (range 4–71) days. 31 women (22 who did not miscarry by 4 weeks and 9 who had complications) had surgery. The median time to clinically indicated surgery was 8 (range 2 – 15) days. The median number of repeat scans was 1 (range 0–5). Presence of yolk sac, fetal pole, CRL ≤ 15 mm and nulliparity were not predictive of successful management (Fisher’s exact test). Nulligravidity showed a 96.6% sensitivity (95% CI 82 – 99.9%) and specificity of 21.1% (95% CI 14–29%) for not having an ERPC. The mean gestational sac diameter is most predictive of needing an ERPC [area under curve 0.6 (95% CI 0.4–0.7) ROC analysis]. Conclusions: Many women who start expectant management opt out early. This study suggests women with delayed miscarriage should be counselled about the likely length of time to completion. Four weeks rather than the more commonly used cut off of two weeks increases the number who could achieve a successful outcome without increasing the complication rate. Mean GSD should be used to give a more specific likelihood of successful outcome. OC04.04 # The appearance of an embryo with or without visible heartbeat and yolk sac in small gestation sacs in accurately dated fertility treatment patients A. Pexsters 1 , A. Daemen 1 , C. Bottomley 2 , Y. Abdallah 3 , O. Naji 3 , T. D’Hooghe 1 , T. Bourne 1,3 , D. Timmerman 1 1 UZ Gasthuisberg, K.U. Leuven, Leuven, Belgium; 2 Chelsea and Westminster Hospital, London, United Kingdom; 3 Imperial College London, Hammersmith Campus, London, United Kingdom Objectives: The optimal ultrasound criteria to diagnose miscarriage are not clear. We aimed to assess the timing of appearance and ability to measure structures: embryo with or without heartbeat and yolk sac in small gestational sacs when the date of conception is accurately known. Methods: Prospectively collected patients who were pregnant after in vitro fertilisation, intra-uterine insemination and ovulation induction underwent two-weekly transvaginal ultrasound scans from 5 and 6 weeks gestational age (GA). Measurements were performed by a single operator and included crown–rump length (CRL), three measurements of gestation sac to calculate mean sac diameter (MSD) and three measurements of yolk sac (YS) at every visit. The presence of an embryonic heartbeat was stated. Pregnancy outcome was subsequently recorded. The timing of the appearance of embryo, heartbeat and YS and the relationship between their appearance and measurements for CRL and MSD was investigated at 5 and 6 weeks GA using t-test. Results: 90 singleton pregnancies resulting in a live birth were included. Above 42 days GA, CRL and YS were always measurable and a heartbeat was always seen. In a gestation sac with an MSD of 13.5 mm, both heartbeat and YS could always be identified. A heartbeat could always be seen when the embryo had a CRL of at least 1.9 mm. When the CRL of the embryo was 5 mm at 5 or 6 weeks GA, a YS was always visible. Conclusions: Cut-off values for the timing of appearance of structures inside small sacs and their relation to CRL and MSD size were established in patients with exactly known GA. These results can be used to define the optimal timing of an ultrasound for ascertaining viability in patients after fertility treatment. Care must be taken not to extrapolate these data to the general population with uncertain dates, but the data do provide evidence from which rational definitions for miscarriage might be derived. OC04.05 The limitations of using mean gestation sac diameter (MSD) measurements to define miscarriage T. Bourne 1,3 , Y. Abdallah 1 , A. Daemen 2 , O. Naji 1 , A. Pexsters 3 , C. Bottomley 1 , C. Stalder 1 , D. Timmerman 3 1 Department of Obstetrics & Gynecology, Imperial College London, Hammersmith Campus, London, United Kingdom; 2 Department of Electrical Engineering (ESAT), Katholieke Universiteit Leuven, Leuven, Belgium; 3 Department of Obstetrics & Gynecology, Katholieke Universiteit Leuven, University Hospitals, Leuven, Belgium Objectives: There is a lack of clarity regarding cut-off values for MSD to define miscarriage. Values suggested in the literature range from 13 to 25 mm. We aimed to define the mean gestation sac diameter (MSD) with or without a yolk sac (YS) that can definitively diagnose a miscarriage. Methods: 252 consecutive pregnancies of unknown viability (PUV) with no YS or embryo and 53 with a YS but no embryo were recruited (total n = 305). PUV was defined as an empty gestation sac or sac with a YS but no embryo seen with MSD < 30 mm. The sensitivity, specificity, PPV and NPV were calculated for potential # This presentation is eligible for the Young Investigator award (to be presented at the closing ceremony). 8 Ultrasound in Obstetrics & Gynecology 2011; 38 (Suppl. 1): 1–55