439: Do ruptured membranes remote from term affect neonatal mortality?

Fetus, Prematurity

www.AJOG.org 436 Correlation between liver location and congenital heart malformation in prenatally-diagnosed fetal omphalocele Freddy J. Montero1, Russell S. Miller1, Lynn L. Simpson1 1

Center for Prenatal Pediatrics, Division of Maternal Fetal Medicine, Department of Obstetrics and Gynecology, Columbia University Medical Center, New York, New York

OBJECTIVE: Omphalocele is an abdominal wall defect associated with an increased risk for other congenital malformations. Studies suggest omphalocele characteristics may relate to types of coexistent anomalies. This study assessed the correlation between presence or absence of liver herniation into the omphalocele sac and congenital heart defect in prenatally-diagnosed cases with normal karyotype. STUDY DESIGN: This was a retrospective chart review of consecutive pregnancies complicated by fetal omphalocele that underwent antenatal evaluation at Columbia University Medical Center between 2003 and 2009. Study inclusion required a comprehensive fetal anatomical sonogram, fetal echocardiogram, and known fetal or neonatal karyotype determination. Fetuses with an abnormal or unknown karyotype were excluded from consideration. RESULTS: Thirty-one cases of prenatally-diagnosed omphalocele underwent inclusion imaging. 3 fetuses with abnormal karyotype and 3 fetuses with unknown karyotype were excluded, leaving 25 subjects for analysis. 20 (80%) fetuses possessed an extracorporeal liver and 5 (20%) had intracorporeal livers. Overall, 7 (28%) fetuses were diagnosed with congenital heart defects. Euploid fetuses with an extracoporeal liver were less likely to have an associated cardiac malformation compared to those with an intra-abdominal liver (15% vs 80%, p⫽0.01). A greater percentage of major congenital heart defects was noted within the intracorporeal liver group, but this difference did not reach statistical significance (40% vs 5%, p⫽0.09). CONCLUSION: In chromosomally-normal fetuses with an intracorporeal liver, there is a higher likelihood of congenital heart malformation. CHD in euploid fetuses with omphalocele

Any cardiac defect

Intra- corporeal liver (nⴝ5)

Extra- corporeal liver (nⴝ20)

4 (80%)

3 (15%)*

..........................................................................................................................................................................................

Minor defect 2 (40%, both VSD) 2 (10%; ASD, VSD) .......................................................................................................................................................................................... Major defect 2 (40%; TGA, AVCD) 1 (5%; TOF) ..........................................................................................................................................................................................

Poster Session III

tubated was decreased for the more than 7 days elapsed time group (HR 9.46: 95% CI 1.34-66.4), as determined by Cox proportional hazard model. CONCLUSION: Elapsed time from steroid dosing to delivery has a significant effect on RDS, mortality, NEC and length of intubation. It had no effect on length of NICU admission or IVH. 0002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.603

438 Association of midgestation paraoxonase 1 activity and pregnancies complicated by preterm birth Arthur Baker1, Sina Haeri1, Richard Klein2, Kim Boggess1 1

University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, Medical University of South Carolina, Charleston, South Carolina

2

OBJECTIVE: Preterm birth (PTB) is a major cause of perinatal morbid-

ity and mortality. Women delivering preterm are at increased risk for atherosclerosis later in life. The HDL-associated antioxidant enzyme paraoxonase 1 (PON 1) is important in the pathophysiology of atherosclerosis, and recent genetic studies have suggested that it may be a possible risk factor for PTB. Our goal was to compare maternal serum PON 1 activity in early pregnancy between women with preterm and term birth. STUDY DESIGN: A case-control study of 30 women with spontaneous PTB (ⱖ23 0/7 and ⱕ33 6/7 weeks) and 90 women with uncomplicated term birth (ⱖ37 weeks) was conducted. Banked maternal serum collected at 15-18 weeks for routine genetic multiple marker screening was used to measure lipid concentrations (total cholesterol, HDL, LDL, triglycerides) and PON 1 activity using two substrates: paraoxon and phenylacetate (arylesterase activity). Mean PON 1 activity was compared between groups using student’s t-test. RESULTS: Maternal demographics were similar between the two groups. PON 1 activity (paraoxon) was significantly lower in women who delivered preterm compared to those who delivered at term (12.9 ⫾ 6.1 vs. 16.6 ⫾ 7.7 dA/min, p⫽0.02). Arylesterase activity and serum lipid concentrations were similar between women with preterm and term birth. CONCLUSION: Midgestation PON 1 activity is lower in women that later experience spontaneous PTB. Antioxidant capacity as measured by PON 1 activity may be important in PTB prediction. Our findings also support the theory that women delivering preterm are at risk for atherosclerosis later in life. 0002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.604

*pⴝ0.01

0002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.602

Yair Blumenfeld1, Henry Lee2, Jeffrey Gould3, Elizabeth Langen1, Anahita Jafari1, Yasser El-Sayed1

437 Does the time from corticosteroid dosing to delivery affect neonatal outcome?

1

Frank Schubert1, John Gianopoulos1 1

439 Do ruptured membranes remote from term affect neonatal mortality? Stanford University, Obstetrics & Gynecology, Stanford, California, University of California, San Francisco; CPQCC, Neonatology, San Francisco, California, 3Stanford University; CPQCC, Neonatology, Stanford, California 2

Loyola University Chicago, Maywood, Illinois

OBJECTIVE: To determine if the timing of corticosteroids affects neo-

natal morbidity and mortality. STUDY DESIGN: A retrospective cohort study. IRB approval was obtained. Women with codes associated with preterm birth from 2003 2007 were identified. Maternal and neonatal charts were reviewed for data extraction. Odds ratios (OR) were determined for respiratory distress syndrome (RDS), intraventricular hemorrhage (IVH), necrotizing enterocolitis (NEC), and mortality using multivariate logistic regression. RESULTS: The sample included 159 neonates born to 128 women divided by elapsed time from steroids to delivery. Power analysis was performed with power set at 0.9. The odds ratios for RDS, IVH, NEC, and mortality were calculated using multivariate logistic regression. Compared to 1-7 days, the OR for RDS was 19.8 (95% CI 1.82-216) for 0-24 hours and 3.77 (95% CI 1.23-11.6) for more than 7 days. There was also an increased risk of mortality (OR 15.4: 95% CI 1.09-217) and decreased risk of NEC (OR 0.18: 95% CI 0.04-0.79) when steroid dosing was more than 7 days before delivery. The length of time in-

OBJECTIVE: Preterm rupture of membranes (ROM) is the underlying

etiology in approximately 1/3 of all preterm deliveries. The objective of our study was to assess the effect of ROM on neonatal mortality. STUDY DESIGN: A retrospective review from a large state perinatal database (CPQCC – California Perinatal Quality Care Collaborative) was performed. Prenatal data including the presence of ROM, antenatal steroid administration, maternal age, race, hypertension, and prenatal care were recorded. Mortality rates were compared for neonates born in the setting of ROM to those that were not. Neonates with fetal anomalies were excluded. RESULTS: 17,501 infants born between 24 and 34 weeks gestation from 126 California NICUs from 2005 to 2007 were included. ROM rates decreased from 45% at 24-26 weeks gestation to 27% at 32-34 weeks. There was no difference in neonatal mortality between neonates born without ROM to those born with ROM ⬍ 18 hours prior to delivery. The presence of prolonged ROM ⬎ 18 hours was associated with decreased neonatal mortality at 24-26 weeks gestation when com-

Supplement to DECEMBER 2009 American Journal of Obstetrics & Gynecology

S167

Poster Session III

Fetus, Prematurity

pared with either the no ROM or ROM ⬍ 18 hours group (18% vs. 29% vs. 31%, p⬍0.0001). This protective effect was not found in other gestational ages and was still present when adjusting for possible confounding factors. Odds ratios for neonatal mortality at 24-26 weeks were 1.67 (CI 1.18-2.29) and 1.78 (CI 1.25-2.55) when comparing no ROM and ROM ⬍ 18 hours with ROM ⬎ 18 hours respectively. CONCLUSION: Overall, ROM rates decreased with increasing gestational ages. ROM ⬎ 18 hours was associated with decreased neonatal mortality at 24-26 weeks gestation but not at any other gestational ages. This finding warrants confirmation in other settings.

0002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.605

440 The effect of ruptured membranes on early neonatal sepsis Yair Blumenfeld1, Henry Lee2, Jeffrey Gould3, Mara Greenberg1, Joyce Sung1, Yasser El-Sayed1 1

Stanford University, Obstetrics & Gynecology, Stanford, California, University of California, San Francisco; CPQCC, Neonatology, San Francisco, California, 3Stanford University; CPQCC, Neonatology, Stanford, California

2

OBJECTIVE: Preterm rupture of membranes (ROM) is the underlying etiology in approximately 1/3 of all preterm deliveries. The objective of our study was to assess the effect of ROM on early neonatal sepsis. STUDY DESIGN: A retrospective review of neonatal birth data from a large state database (CPQCC –California Perinatal Quality Care Collaborative) was performed. Prenatal data including ROM, antenatal steroid administration, maternal age, race, hypertension, and prenatal care were recorded. Neonatal outcome data were analyzed. We compared early sepsis rates in neonates born with prolonged ROM ⬎ 18 hours to those with either no ROM or ROM ⬍ 18 hours. Early neonatal sepsis was defined as a positive CSF or blood culture in the first 3 days of life. RESULTS: 17,501 infants born between 24 and 34 weeks gestation from 126 California NICUs from 2005 to 2007 were included. Early sepsis rates were significantly higher in the prolonged ROM group vs. the no ROM group at gestational ages 26-34 weeks even when controlling for potential confounding factors. There were no significant differences in sepsis rates between neonates born with ROM less than or greater than 18 hours. CONCLUSION: Early neonatal sepsis was higher in the prolonged ROM group vs. the no ROM group. Latency greater than 18 hours was not found to have an adverse affect on neonatal sepsis rates. These findings warrant confirmation in other settings.

0002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.606

S168

www.AJOG.org 441 17-hydroxyprogesterone for triplet pregnancy does not reduce prematurity or neonatal morbidity, may increase midtrimester loss CA Combs1, T. Garite1, M. Porto2, K. Maurel1, for the Obstetrix Collaborative Research Network1 1

Obstetrix Medical Group, Various Sites, California, University of California, Irvine, Orange, California

2

OBJECTIVE: To test the hypothesis that prophylactic 17-alpha-hy-

droxyxprogesterone caproate (17P) given to mothers with triplet pregnancy will reduce composite neonatal morbidity by decreasing the rate of preterm delivery. STUDY DESIGN: Placebo-controlled, double-blind, multicenter, randomized clinical trial. Mothers with trichorionic triplets were randomized to 17P 250 mg or placebo at 16-23 wks’ gestational age (GA), IM injections repeated weekly until 34 wks’ GA. Randomization was 2-to-1 17P vs placebo. Sample size 81 mothers (243 babies) gave 80% power to detect reduction of composite neonatal morbidity from 60% with placebo to 40% with 17P. RESULTS: 56 mothers were randomized to 17P, 25 to placebo at mean GA 20 wks. Baseline characteristics were similar between the groups. There was no significant difference in composite neonatal morbidity (38% with 17P vs 41% with placebo), or in mean GA at delivery (31.9 wks vs 31.8 wks), delivery ⬍28 wks (16% vs 8%), ⬍32 wks (34% vs 52%, P ⫽ 0.13), ⬍ 35 wks (77% vs 84%). The 17P group had 4 midtrimester pregnancy losses ⬍ 24 wks and 1 fetal death at 23 wks, a total loss of 13 of 168 fetuses before viability (8%) versus none with placebo (P ⬍ 0.02). CONCLUSION: Prophylactic 17P is not indicated for triplets because there is no evidence that it reduces preterm delivery or neonatal morbidity. An association of 17P with midtrimester loss was suggested in our results and in the Meis-MFMU trial. Previous publications evaluating 17P in twin or triplet pregnancy have not reported midtrimester loss rates. We encourage publication of pregnancy loss statistics with the results of future trials evaluating 17P for various indications. 0002-9378/$ – see front matter • doi:10.1016/j.ajog.2009.10.607

442 Intraamniotic infection not reliably predicted by clinical signs and symptoms CA Combs1, D. Hickok2, T. Garite1, R. Porreco1, D. Hollemon2, L. Harden1, for the Obstetrix Collaborative Research Network1 1

Obstetrix Medical Group, Various Sites, California, ProteoGenix, Inc, Costa Mesa, California

2

OBJECTIVE: In evaluation of preterm labor, the decision to perform

amniocentesis is often based on whether there are signs or symptoms of intraamniotic infection (IAI). This study was performed to assess the screening effectiveness of traditional clinical indicators of IAI. STUDY DESIGN: 111 women with preterm labor, singleton pregnancy, intact membranes, ⬍37 weeks gestational age (GA) underwent amniocentesis as part of an evaluation for IAI. Amniotic fluid cultures were performed for eubacteria, Mycoplasma sp. and Ureaplasma sp. A research-use PCR assay that detects highly-conserved 16S rRNA gene sequences was used to detect microbes across a broad taxonomic range. IAI was defined as either a positive culture or 16S rDNA PCR. RESULTS: Prevalence of IAI was 14% overall (16/111), 12 cases positive for both culture & 16S rDNA and 4 positive for 16S rDNA alone. Prevalence of IAI varied by GA: 21% before 28 wks, 19% at 28-32 wks, 0% after 32 wks. Table shows low sensitivity of clinical signs as predictors of IAI. CONCLUSION: IAI is common in preterm labor, especially before 32 wks. Clinical signs and symptoms have low sensitivity for prediction of IAI in the setting of preterm labor, but higher specificity. Relying on these characteristics alone for a decision to perform amniocentesis would fail to detect IAI in the majority of cases. Improved noninvasive screening or diagnostic tests would be of great clinical utility because prediction of IAI by clinical characteristics is of limited value.

American Journal of Obstetrics & Gynecology Supplement to DECEMBER 2009