13–17 September 2009, Hamburg, Germany Oral communication abstracts Rose &McCallum, Schild (2 formulae), Scott, Shepard, Siemer, Thurnau, Weiner (2 fomulae) and Warsof). To assess the accuracy of the formulae the differences between FBW and EFW (expressed as grams, as percentage of FBW and as absolute percentage (|FBW − EFW|/FBW) were calculated. To assess whether the accuracy changes with increasing FBW a linear regression analysis was performed. Results: 193 pregnancies fulfilled the inclusion criteria with a mean FBW of 1001g. The mean differences between FBW and EFW were within 10% with the formulae from Hadlock I and III, Scott, Hadlock IV, Hadlock II and Masal (absolute percentage of FBW: 9.2%, 9.2%, 9.3%, 9.4%, 9.9% and 9.9%). Of those, EFW and FBW was within 10% of the FBW in more than 60% of the cases with the formulae from Hadlock I and III and with the Scott formula (64.2%, 61.1%, 60.6%). Linear regression showed that with the Hadlock I and III formulae the accuracy did not change with increasing fetal weight (r=0.936 and p=0.938, p<0,0001, intercept and slope not significantly different from 0 and 1). With these weight estimation formulae, the 95% limits of agreement between FBW and EFW were −19% to 19% (Hadlock I) and −17.3% to 20.1% (Hadlock III) of the FBW. Conclusion: Of the 23 commonly available weight estimation formulae, the Hadlock I (BPD-HC-AC-FL) and III (BPD-AC-FL) formulae were most accurate in estimating fetal weight in fetuses with a birth weight of 1500g or less. OC26.02 Fetal weight estimation in pregnancy complicated by obesity: comparison between conventional and a new 3D ultrasound method B. Rovetto, N. Palai, S. Zatti, A. Lojacono, F. Prefumo, T. Frusca Maternal-Fetal Medicine Unit, University of Brescia, Brescia, Italy Objectives: To determine whether increasing body max index (BMI) decreases the accuracy of ultrasound estimation for fetal weight, and to compare the accuracy of birth weight estimation with the three- dimensional (3D) fractional thigh volume (Tvol) method against conventional bi-dimensional (2D) method. Methods: Obese pregnant women (BMI > 30 kg/m 2 ) were referred at 34–36 +6 weeks of gestation to obtain estimated fetal weight (EFW). In all cases the birth weight was estimated using the ‘‘Gestation adjusted projection method’’ as previously reported by Best and it was compared with neonatal weight at delivery. A conventional 2D ultrasound scan was performed to obtain EFW using Hadlock formula (BPD, HC, AC, FL). 3DTvol was acquired and used for EFW with the method previously described by Lee et al Results: 25 cases were studied. The median gestational age at delivery was 39 w (IQR 38 +3 –39 +2 ) and the median birth weight was 3500 g (IQR 3250–3800). Newborns with weighing > 4000 g were five, of which all were correctly indentified with Tvol (100%) but only 3 (60%) with Hadlock formula. The median difference between the predicted and the actual birth weight using Hadlock formula was 7,8% (IQR 5,5%–12,5%), using Tvol method was 4,3% (IQR 2,3–8) (p=0.002). The percentage of cases with an error > 10% with Hadlock formula was 40%, with Tvol method was 12%. Conclusions: In obese pregnant women, where fetal macrosomia is very difficult to detect with conventional 2D measurements, inclusion of fractional Tvol improves the accuracy of birth weight prediction. OC26.03 Derivation of new fetal weight estimation models using fractional limb volume W. Lee 1,2 , M. Balasubramaniam 4 , R. L. Deter 5 , L. Yeo 2,3 , S. Hassan 2,3 , F. Gotsch 2 , J. P. Kusanovic 2,3 , L. Goncalves 2,3 , R. Romero 2,3 1 Division of Fetal Imaging, William Beaumont Hospital, Royal Oak, MI, USA; 2 Perinatology Research Branch, NICHD/NIH/DHHS, Bethesda, MD, USA; 3 Department of OB/Gyn, Wayne State University, Detroit, MI, USA; 4 Division of Biostatistics, William Beaumont Hospital Research Institute, Royal Oak, MI, USA; 5 Department of OB/Gyn, Baylor College of Medicine, Houston, TX, USA Objective: Develop new fetal weight estimation models that combine fractional limb volume with conventional 2D biometry. Methods: A cross-sectional study of 271 fetuses used 3DUS to acquire biparietal diameter (BPD), abdominal circumference (AC), femoral diaphysis length (FDL), fractional arm volume (AVol) and fractional thigh volume (TVol) within 4 days of delivery. Weighted multiple linear regression was used to develop ‘modified Hadlock’ models for comparison to new models using natural logarithmic (Ln) transformed parameters for birth weight (BW). Predicted and observed weights were compared using mean % difference between predicted and actual BW (systematic error) and SD of the % differences (random error). The proportion of newborns with EFW within 5 or 10% of BW, were compared using McNemar’s test. Results: Most infants were delivered from uncomplicated pregnan- cies. BW ranged from 235 to 5,790 g. Six models were very accurate with mean systematic errors that were not significantly different from zero (t-test). Model 3 (0.18 ± 6.6%, R 2 = 0.99) and Model 6 (0.12 ± 6.6%, R 2 = 0.99) provided the best results when compared to modified Hadlock models (random error = 8.5%) (Pitman test). Model 6 classified an additional 7.3% and 4.1% of infants within 5% and 10% of BW, respectively (p<0.0001). Model 3 Ln BW = 0.5046 + 1.9665 (Ln BPD) − 0.3040 (Ln BPD) 2 + 0.9675 (Ln AC) + 0.3557 (Ln AVol) Model 6 Ln BW =−0.8297 + 4.0344 (Ln BPD) − 0.7820 (Ln BPD) 2 + 0.7853 (Ln AC) + 0.0528 (Ln TVol) 2 Conclusions: More precise fetal weight estimations are possible by combining fractional limb volume with 2D biometry. Substitution of FDL by fractional limb volume, use of Ln transformations with weighted regression analysis, and selective application of squared transformed parameter terms reduced the random error to 6.6%. New fetal weight estimation models, that incorporate soft tissue parameters, may offer new insight regarding the body composition of malnourished fetuses. OC26.04 Clinical utility of customized IUGR standard in a low risk population J. Zhang 1 , R. Mikolajczyk 1,2 , W. Sun 1 , J. Grewal 1 , M. Klebanoff 1 , M. Merialdi 3 1 Epidemiology Branch, Eunice Shriver Kennedy National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD, USA; 2 Department of Public Health Medicine, University of Bielefeld, Bielefeld, Germany; 3 Reproductive Health and Research, World Health Organization, Geneva, Switzerland Objective: Defining intrauterine growth restriction (IUGR) has been a long-standing challenge. Earlier studies suggested that a customized standard for IUGR (adjusting for maternal and Ultrasound in Obstetrics & Gynecology 2009; 34 (Suppl. 1): 1–61 51