435 Incidence of non-cardiac structural anomalies in twin-twin transfusion syndrome Shivani Patel 1 , Linda Randolph 2 , Kurt Benirschke 3 , Arlyn Llanes 1 , Larisa Yedigarova 1 , Ramen Chmait 1 1 University of Southern California, Los Angeles, CA, 2 Childrens Hospital Los Angeles, Los Angeles, CA, 3 University of California, San Diego, San Diego, CA OBJECTIVE: Monochorionic twins have increased perinatal morbidity and mortality due to a variety of factors, including twin-twin transfu- sion syndrome (TTTS) and increased risk of congenital anomalies. The objective of this study was to describe the non-cardiac anomalies in monochorionic twins afflicted with TTTS. STUDY DESIGN: This was a retrospective study of 221 consecutive cases of TTTS treated with laser surgery between March 2006 and May 2010. Major versus minor non-cardiac congenital anomalies were classified according to the Western Australian Birth Defects Registry system (BMJ 1997;315:1260-1265). Because of the secondary cardiac derangements that arise in association with TTTS, cases with cardiac anomalies were excluded from the analysis. Anomalies were diag- nosed by antepartum ultrasound and/or review of neonatal medical records. Statistical analyses were performed using Fisher’s exact and chi square tests. A p-value  0.05 was considered significant. RESULTS: Non-cardiac congenital anomalies were identified in a total of 24 individual twins (5.4%). There was a significantly increased rate of non-cardiac anomalies in the donor twin (8.1%, 18/221) versus the recipient twin (2.7%, 6/221, p=0.019). Comparing cases with and without non-cardiac anomalies, there were no differences in fre- quency of Quintero Stage III/IV cases (68% vs. 67%, p=1.0), gesta- tional week of laser surgery (20.2  1.9 vs. 20.7  2.4, p=0.39), ges- tational week at delivery (32.0  4.6 vs. 32.4  4.3, p=0.75), 30-day survival of the donor (68% vs. 74%, p=0.84), and 30-day survival of the recipient (73% vs. 83%, p=0.36), respectively. Of the 18 donor fetuses with anomalies, 10 (55.6%) were considered major. Of the 6 recipient fetuses with anomalies, 3 (50%) were major. CONCLUSIONS: An increased incidence of non-cardiac structural anomalies was found in donor twins versus recipient twins in cases of TTTS. The etiology of this difference is unknown, but it may in part be related to early blood flow disturbances seen in twin-twin transfusion syndrome. 436 Identification of growth restricted newborns at risk for neonatal death Suneet Chauhan 1 , Todd Lutton 2 , Everett Magann 3 , Eugene Chang 4 , John Morrison 5 , Alfred Abuhamad 1 1 Eastern Virginia Medical School, Norfolk, VA, 2 Sabre Systems Inc., Warimister, PA, 3 University of Arkansas for Medical Sciences, Little Rock, AK, 4 Medical University of South Carolina, Charleston, SC, 5 University Clinic Associates, Jackson, MS OBJECTIVE: Fetal growth restriction (FGR) is commonly defined as birth weight (BW) below 10% for gestational age (GA; Alexander et al Obstet Gynecol 1996) but most of these newborns are not at increased risk of neonatal death (ND). In 2005, Boulet et al (AJOG) published GA based BW thresholds for FGR that were linked with 2-, 2.5-, and 3-fold risks of ND (compared to newborns at 45-55%). This is a sec- ondary analysis of our earlier report (AJOG 2006) on the detection of FGR among women undergoing antepartum testing (ANT). The pur- pose of this study was to determine the likelihood of identifying BW  10% vs those at increased risk of ND. STUDY DESIGN: The inclusion criteria were: non-anomalous single- tons, ultrasound exam  22 weeks, medical or obstetric complica- tions requiring ANT, sonographic estimated fetal weight (SEFW) within 4 weeks of birth, GA & SEFW within the range provided by Boulet. SEFW was derived from measurements of 4 biometric param- eters and regression equation proposed by Hadlock et al. Non-over- lapping 95% confidence intervals (CI) were considered significant and a likelihood ratio of 10 or more was considered a useful diagnostic threshold for SEFW. RESULTS: During 6 years, 1,933 women met the inclusion criteria. The time interval between the SEFW and delivery was 6.6 6.5 days. Using Alexander’s thresholds of BW  10% for GA, 27% (n = 523) of newborns were FGR; using Boulet’s thresholds, 19% (373) were at increased risk of ND (4% [69] at 2X risk; 1% [24] at 2.5X, and; 14% (279) at 3.0X). Predictive accuracy of identifying abnormal growth is provided in the table. The ND rate was 9/1000 births. CONCLUSIONS: SEFW is significantly more reliable in identifying growth  10% for GA than newborns at increased risk of ND. To decrease the mortality linked with FGR, we need to identify those at risk. Detection of abnormal birth weight (BW) BW < 10% for GA* (N  523) At Risk of ND ^ (N  373) ND / 1,000 births 14 9 .......................................................................................................................................................................................... Sensitivity 63% (59-68%) 44% (39-49%) .......................................................................................................................................................................................... Specificity 94% (93-95%) 88% (87-90%) .......................................................................................................................................................................................... Positive predictive value 80% (75-83%) 54% (49-60%) .......................................................................................................................................................................................... Negative predictive value 87% (86-89%) 83% (81-85%) .......................................................................................................................................................................................... Likelihood ratio 10.8 3.8 .......................................................................................................................................................................................... Bolded if non-overlapping CI. * Using nomograms published by Alexander; ^ Using nomograms published by Boulet. 437 Fetal pleuroamniotic shunting for macrocystic lung lesions Susanne Schrey 1 , Yoav Yinon 1 , Edmond Kelly 2 , Jacob Langer 3 , Rory Windrim 1 , Gareth Seaward 1 , Greg Ryan 4 1 Fetal Medicine Unit, Mount Sinai Hospital, Toronto, ON, 2 Paediatrics, Mount Sinai Hospital, Toronto, ON, 3 Division of General and Thoracic Surgery, Hospital for Sick Children, University of Toronto, Toronto, ON, 4 Fetal Medicine Unit, Mount Sinai Hospital, University of Toronto, Toronto, ON OBJECTIVE: To evaluate the role of fetal pleuro-amniotic shunting for large macrocystic congenital cystic adenomatoid malformations (CCAM) of the lungs. STUDY DESIGN: Retrospective review of perinatal outcome of 13 fe- tuses with isolated large macrocystic CCAM’s who had pleuro-amni- otic shunts inserted. RESULTS: Shunts were inserted at a mean of 26 wks (17-36) in 13 fetuses. A single shunt resulted in complete decompression of all mul- ticystic lesions. One shunt dislodged after 2 wks. 5 fetuses were hy- dropic, and 2 had polyhydramnios. Marked mediastinal shift was present in all; 6 had neither hydrops nor polyhydramnios, but all 13 had a poorly prognostic CCAM–volume ratio (1.6). Hydrops & hydramnios resolved after shunting in all cases. CCAM’s decreased significantly in size in all survivors. One severely hydropic 17 wk fetus died 1 day after shunting. None laboured prematurely or had PPROM soon after shunting, Delivery was at a mean of 36.3 wks (37.9 wks if 17 wk IUD is excluded). 9 delivered at term, one at 36 & one at 33 wks, one pregnancy is ongoing. 11 were delivered vaginally & one by cae- sarean section. There were no maternal complications. Postnatally, all had uneventful lobectomies within the first week and CCAM was confirmed by pathology in all. CONCLUSIONS: Fetuses with large CCAM’s are at risk of developing hydrops +/- hydramnios. If hydrops develops, mortality is almost 100%. Fetal pleuro-amniotic shunts have a relatively low complica- tion rate and result in excellent perinatal outcome. We suggest that fetal pleuroamniotic shunting should be considered for very large macrocystic CCAM’s, even in the absence of hydrops. Poster Session III Doppler Assessment, Fetus, Neonatology, Prematurity www.AJOG.org S176 American Journal of Obstetrics & Gynecology Supplement to JANUARY 2011