447 Quantification of ultrasound umbilical veinchanges in fetal diaphragmatic herniaDouglas Richards1, David Kays2

1Intermountain Medical Center and the University of Utah, Maternal FetalMedicine, Murray, UT, 2University of Florida, Pediatric Surgery, Gainesville, FLOBJECTIVE: Deviation of the intraabdominal umbilical vein has beendescribed as a marker of congenital diaphragmatic hernia and hasbeen shown to help identify liver prolapse into the fetal chest. Devia-tion has previously been described only as a subjective finding. Thepurpose of this study is to quantify the degree of deviation in affectedfetuses and to determine the accuracy of measurements for determin-ing liver prolapse.STUDY DESIGN: In patients undergoing prenatal sonography for CDHthe mid-hepatic portion of the umbilical vein was identified, and thedistance between the lateral edge of the vein and the inner rib marginwas measured. The same was done on the right, and the ratio of left toright measurement was termed the umbilical vein ratio (UVR). Forthis analysis, deviation of the UV, as quantified by the UVR, was usedto predict the presence and side of the hernia, as well as to diagnoseliver prolapse. The 5th and 95th confidence limits for the normal UVRwere determined from 80 unaffected fetuses. Of 160 cases of CDHseen at our center between 11/1995 and 9/2010, 56 fetuses were stud-ied prospectively, and 76 had adequate images available for retrospec-tive measurements. The first sonogram at our center was used foranalysis.RESULTS: The gestational age at the time of the ultrasound was 29.4�/� 6.2 weeks. Of the 132 patients, 14 had a right-sided and 118 hada left-side hernia. All but one of the cases with right-sided hernias hadan UVR above the normal range, and all had liver prolapse. Of thosewith a left-sided hernia, only 2 had a ratio within the normal range. Ofthose with a left-sided hernia, an UVR of � 0.4 was shown by ROCanalysis to be the best predictor of liver prolapse. This cut off had asensitivity of 0.89 for predicting prolapse, with a false positive rate of0.14.CONCLUSION: 93% of right-sided CDH and 98% of left-sided lesionshave an UVR outside of the normal range. This demonstrates thatdeviation may a useful indicator of CDH in screening ultrasound ex-aminations. A UVR less 0.4 is predictive of liver prolapse.

448 Predictors for acute fetal demise immediatelyafter fetoscopic laser coagulation intwin-twin transfusion syndromeRamesha Papanna1, Rodrigo Ruano2, Anthony Johnson2,Karen Moise2, Michael Belfort2, Kenneth Moise2

1Yale University, Ob/Gyn & Reprod Sci., New Haven, CT, 2BaylorCollege of Medicine, Obstetrics and Gynecology, Houston, TXOBJECTIVE: Fetal demise following fetoscopic laser surgery (FLS) fortwin twin transfusion syndrome (TTTS) has been well documented(Muratore et al 2009 and Skupski et al 2010). However, the perioper-

ative and operative risk factors for acute fetal demise (� 24 hrs) afterFLS remain unclear.STUDY DESIGN: Retrospective case-control study of prospectively col-lected data from 284 consecutive patients undergoing FLS (selectivesequential with Solomon technique) for TTTS between November2006 and June 2011. All patients were evaluated �24 hours after theprocedure. Exclusions: selective reduction secondary to failed lasersurgery (n�11 cases) and delivery �24 hours after the procedure (n�4 cases). The following pre and intra-operative variables were ana-lyzed: gestational age at procedure, maternal age, parity, Quinterostage, fetal growth discordance, preoperative Doppler studies in um-bilical arteries (UA), umbilical veins (UV), ductus venosus (DV),middle cerebral artery, placental location and umbilical cord inser-tions in the placenta, type of anesthesia, trocar time, total laser energy,number and type of vessels, before and after maximum verticalpocket, net amnioreduction, performance of Solomon technique. Therisk factors for acute recipient death (ARD) and acute donor death(ADD) were compared with dual survival fetuses. Univariate analysiswas performed using t-test, Mann-Whitney test and Chi-square test,as well as ROC curves followed by multiple logistic regression analysis.RESULTS: 269 patients were included in the analysis: 241(89.6%) dualsurvivors, 13 (4.8%) ARD and 15(5.6%) ADD in 27% (OR: 4.2;95%CI: 1.3-13.4), which was provided by the ROC analysis (area un-der the curve: 0.69, sensitivity: 67%, specificity: 61%; p�0.02). Noneof the other parameters studied were significantly associated with dif-ferences in the rate of ARD and ADD.CONCLUSION: Pre and intra-operative findings predict both ARD andADD in TTTS with FLS. Further studies are needed to reduce the acutefetal loss after the procedure, specifically, protective value of Solomontechnique in ARD and severe growth discordance in ADD.

449 Predictive value of combined cervicovaginal cytokinesand gestational age at assessment for intraamnioticinfection in preterm premature rupture of membranesAeli Ryu1, Kyo Hoon Park1, Sung Youn Lee1,Kyung Joon Oh1, Eun Ha Jeong1, Shi Nae Kim1

1Seoul National University College of Medicine, Seoul NationalUniversity Bundang Hospital, Department of Obstetricsand Gynecology, Seongnam-si, South KoreaOBJECTIVE: To determine whether or not cervicovaginal cytokine lev-els and gestational age at assessment alone or in combination canidentify microbial invasion of amniotic cavity in women with pretermpremature rupture of membranes (PPROM).STUDY DESIGN: Cervicovaginal secretions were collected immediatelybefore amniocentesis in 74 consecutive women admitted for PPROM(20.0-35.0 weeks) with singleton pregnancies. Amniotic fluid (AF)obtained by amniocentesis was cultured and the white blood cell(WBC) count was determined, and the following cytokines were mea-sured in cervicovaginal fluid: interleukin (IL)-1�, IL-6, and IL-8. The

Poster Session III Doppler Assessment, Fetus, Prematurity www.AJOG.org

S206 American Journal of Obstetrics & Gynecology Supplement to JANUARY 2012