fia0

aJC1

B

OgStDcijopRstCac0gcCnw0

ctJ1

OaOitScaC�sbaRc(nnGDCtc(s

Cmfhac0

tpER1

CCOaucwdSsptmLNCcbwpRrspNCtt0

6cKJ1

SSNaO(scSl(wtcM

www.AJOG.org Fetus, Prematurity Poster Session III

ndings suggest that the detection of PAMG in vaginal fluid is indic-tive of micro-perforations of the intact membranes.002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.694

529 The relationship between previous cervical surgerynd gestational age at delivery in twin gestationsennifer Jolley1, Robert Ehsanipoor1, Michael Haydon2,ecilia Lyons Gaffaney3, David Lagrew4

University of California, Irvine, Orange, California, 2Longeach Memorial Medical Center, Long Beach, California,

3Obstetrix Medical Group, San Jose, San Jose, California,4Saddleback Memorial Medical Center, Laguna Hills, CaliforniaBJECTIVE: To describe the impact of previous cervical surgery onestational age at delivery in twin gestations.TUDY DESIGN: A retrospective review of twin pregnancies delivered be-ween January 1998-December 2005 at two institutions was performed.ata collected included maternal demographics, history of procedure for

ervical dysplasia, and gestational age at delivery. Subjects were excludedf they had cerclage placement, twin-to-twin transfusion syndrome, ma-or anomalies, an indicated preterm delivery prior to 34 weeks gestation,r had undergone multifetal pregnancy reduction. Statistical analysis waserformed using JMP 7.0 software.ESULTS: 875 patients met inclusion criteria. 110 women had previousurgical procedures for cervical dysplasia, including cold knife coniza-ion, loop electrosurgical excision procedure (LEEP), cryotherapy, andO2 laser treatment. There was no significant difference in gestationalge at delivery between women who had undergone cervical surgeryompared to women who had not (mean 36.1 weeks vs. 36.2 weeks, p �.76). There was no difference in the rate of delivery before 34 weeksestation between women who had previous cervical surgery (11.0%)ompared to those who had not (10.9%) [RR 0.99 (95% CI: 0.56-1.76)].ONCLUSION: Previous surgical procedures for cervical dysplasia wereot associated with gestational age of delivery or delivery prior to 34eeks in twin gestations.002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.695

530 Association between post-randomization sonographicervical length and birth gestational age in a multicenterrial of ultrasound-indicated cerclageohn Owen1, Jeff Szychowski2

for the Vaginal Ultrasound Cerclage Trial Consortium, Department ofbstetrics & Gynecology, Birmingham, Alabama, 2University of Alabama

t Birmingham, Department of Biostatistics, Birmingham, AlabamaBJECTIVE: To estimate the predictive value of a 1-week post-random-

zation sonographic cervical length (CL) measurement on birth ges-ational age (GA).TUDY DESIGN: Planned secondary analysis of the NICHD-sponsorederclage trial. Women with prior spontaneous preterm birth (SPTB)t 17-336/7 wks and singletons underwent serial vaginal sonographicL assessment between 16 and 226/7 wks. Consenting women with CL25 mm were randomized to McDonald cerclage or to no cerclage. A

onogram was scheduled the following week, and clinicians werelinded to the results. The difference between CL at randomizationnd post-randomization CL, DELTA, was computed.ESULTS: Of the 301 women in the randomized trial, 271 (90%; 144erclage, 127 no cerclage) returned for their 1-week evaluation, meanSD) interval, 8.1 (3.6) days. As anticipated, mean DELTAs were sig-ificantly (p�0.003) different between cerclage (8.5 mm longer) ando cerclage groups (2.4 mm longer). In a regression model with birthA as the outcome and with randomization CL, cerclage group andELTA as predictors, both DELTA (p�0.0002) and randomizationL (p�0.0001) were significant; we also observed an interaction be-

ween cerclage group and DELTA (p�0.06). Stratifying analyses byerclage group, the effect of DELTA in the cerclage group was lowp�0.07), but in the no-cerclage group, DELTA remained a highly

ignificant predictor of birth GA (p�0.0002). a

Supplemen

ONCLUSION: Ultrasound-indicated McDonald cerclage for CL � 25m is associated with a mean 8.5 mm lengthening in CL at 1-week

ollow-up, but this value is a poor predictor of birth GA. Women atigh risk for recurrent PTB after the finding of CL � 25 mm, may havemodest degree of cervical lengthening even in the absence of cer-

lage, but continued shortening is predictive of earlier birth GA.002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.696

531 N-acetylcysteine prevents preterm birth by attenuatinghe LPS-induced expression of contractile associatedroteins in an animal model for preterm birthugene Chang1, Jingmei Zhang2, Scott Sullivan1,oger Newman3, Inderjit Singh1

Medical University of South Carolina, Charleston, South Carolina,2Medical University of South Carolina, Obstetrics and Gynecology,

harleston, South Carolina, 3Medical University of Southarolina, Mount Pleasant, South CarolinaBJECTIVE: Intrauterine infection is associated with maternal immunectivation (MIA), which can lead to preterm birth through NF-mediatedpregulation of contractile associated proteins (CAPs) resulting in in-reased contractility. We hypothesized that N-acetylcysteine (NAC),ould decrease preterm birth through a decrease in NF activation and aecrease in CAP expression in a MIA model for preterm birth.TUDY DESIGN: Pregnant CD-1 mice were given intrauterine LPS oraline on gestation day 15/20. They received systemic NAC or salinerior to injection and were monitored until delivery. The rate of pre-erm birth in the control, LPS, and LPS � NAC animals was deter-

ined. In another experiment, animals were sacrificed 6 hours afterPS � NAC treatment and myometrium was collected and processed.F activation was assessed by immunoflourescence. Expression of theAP’s: cyclo-oxygenase-2 (COX-2), connexin 43, and oxytocin re-

eptor was assessed by RT-PCR. To compare NF and CAP expressionetween groups, non-parametric testing with the Kruskal-Wallis testas used, as normality of data was not assumed. Post-hoc testing waserformed when appropriate.ESULTS: As previously reported with this model, LPS administrationesulted in preterm birth in and this effect was attenuated by NAC. LPSignificantly increased COX-2, connexin 43, and oxytocin receptor ex-ression. NAC significantly decreased COX-2 expression. LPS increasedF activation; this was significantly attenuated by NAC.ONCLUSION: NAC may be beneficial in prevention of MIA-related pre-erm birth through attenuation of NF activation and COX-2 upregula-ion. It is a promising agent for study in the prevention of preterm birth.002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.697

532 Amniotic fluid matrix metalloproteinase-9, interleukin-, and serum C-reactive protein in predicting histologichorioamnionitis and intraamniotic infectionyo Hoon Park1, Joon-Seok Hong2, Shi Nae Kim1, HeeJung

ung1, Woong Sun Kang3, Dong Myoung Shin2

Seoul National University Bundang Hospital, Seongnam-si, Gyeonggi-do,outh Korea, 2Seoul National University Bundang Hospital, , Gyeonggi-do,outh Korea, 3Seoul National University College of Medicine, Seoulational University Bundang Hospital, Department of Obstetrics

nd Gynecology, Seongnam-si, Gyeonggi-do, South KoreaBJECTIVE: To compare the diagnostic performance of amniotic fluidAF) matrix metalloproteinase-9 (MMP-9), interleukin-6 (IL-6), anderum C-reactive protein (CRP) in the identification of histologichorioamnionitis and a positive AF culture.TUDY DESIGN: One hundred three consecutive patients with preterm

abor, premature rupture of membranes, or cervical incompetence35 weeks gstation) who delivered within 72 hours of amniocentesisere studied. Maternal blood was collected to measure CRP concen-

rations immediately after transabdominal amniocentesis. AF wasultured for aerobic and anaerobic bacteria and for Mycoplasmas, and

MP-9 and IL-6 were determined by enzyme linked immunosorbent

ssay. Placentas were examined histologically. Receiver operating

t to DECEMBER 2009 American Journal of Obstetrics & Gynecology S197