495 FUTURE DEATH AND BOWEL COMPLICATIONS IN UNDELIVERED FETUSES WITHGASTROSCHISIS BY WEEK OF GESTATION DENA TOWNER1, 1University of Califor-nia, Davis, Obstetrics & Gynecology, Sacramento, California

OBJECTIVE: To determine the chance of future death and adverse boweloutcomes by gestational week in undelivered pregnancies with gastroschisis.

STUDY DESIGN: Using linked data from maternal and infant hospitaldischarge records and from birth and death certificates from acute care civilianhospitals in California during the years 1992-1997, we identified infants thatsurvived to gastroschisis repair. In the population of undelivered fetuses, thechance of future neonatal death, bowel atresia and need for major bowelsurgery (resections and ostomies) were calculated at 24 weeks and weekly from34 through 41 weeks. The statistic utilized was the chi square for linear trend.

RESULTS: There were 577 infants at risk for atresia and major bowelsurgery and 387 at risk for death as data on death was only available through1995. There was a significant trend for decreased chance of future bowelatresia (p=0.002) and bowel surgery (p=0.023) with increasing gestationalage but no trend found for death (Table1).

CONCLUSION: Fetuses with bowel atresia and those with major bowelsurgery tend to be delivered earlier in gestation, but there remains a constantrisk for death in the undelivered population of fetuses with gastroschisis.

Percent of undelivered infants with these future complications by completed

gestational week

Gestational week Bowel atresia Major bowel surgery Death

24 7.8% 14.6% 6.7%34 7.8% 13.8% 5.8%35 7.2% 13.2% 5.1%36 6.1% 12.5% 5.0%37 5.4% 11.8% 5.7%38 5.5% 11.9% 5.4%39 3.4% 10.3% 4.4%40 1.5% 9.2% 5.9%41 0% 3.6% 10.0%

496 MODERATE OR THICK MECONIUM-STAINED AMNIOTIC FLUID IS ASSOCIATEDWITH HIGHER RATES OF NEONATAL MORBIDITY SUSAN H. TRAN (F)1,J. T. PARER1, THOMAS J. MUSCI2, AARON B. CAUGHEY1, 1University of California,San Francisco, Obstetrics, Gynecology and Reproductive Sciences, San Fran-cisco, California, 2California Pacific Medical Center (CPMC), Obstetrics andGynecology, San Francisco, California

OBJECTIVE: The purpose of this study was to determine the relationshipbetween moderate or thick meconium-stained amniotic fluid (MSAF) andshort-term neonatal outcomes.

STUDY DESIGN: We designed a retrospective cohort study of all deliveriesbeyond 37 weeks gestational age affected by MSAF from 1981 to 2001 at asingle university hospital. Data was collected on quality of MSAF; umbilicalarterial pH (ua pH) and base excess (ua BE); rates of neonatal sepsis,respiratory distress syndrome (RDS), pneumothorax, jaundice, and chorio-amnionitis; and analyzed with bivariate and multivariate analyses.

RESULTS: There were 27,437 women who met the inclusion criteria. Theneonates with moderate/thick MSAF had higher rates of neonatal morbiditywhen compared to those with no meconium (Table). Moderate/thick MSAFwas associated with chorioamnionitis in 10.24% of pregnancies, as comparedto 4.26% of pregnancies with clear or thin MSAF (p!0.001). Controlling forethnicity, maternal age, gestational age, birthweight, and parity, moderate-thick MSAF was still associated with increased risks of the above neonataloutcomes.

CONCLUSION: We found that moderate/thick MSAF is associated withadverse neonatal outcomes, even when controlling for potential confoundingvariables. Whether MSAF acts as an indicator for high risk pregnancies orthere are interventions that may improve the outcomes in these patientsrequires further investigation.

Moderate/thick MSAF Clear amniotic fluid p-value

ua pH ! 7.0 1.09% 0.52% p!0.001ua BE ! �12 3.09% 1.66% p!0.001Sepsis 0.62% 0.20% p!0.001RDS 1.91% 0.73% p!0.001Pneumothorax 0.60% 0.24% p!0.001

497 LONG-TERM NEURODEVELOPMENTAL AND NEUROPSYCHOLOGICAL OUTCOMESAFTER TREATMENT FOR IMMUNE HYDROPS FETALIS BY INTRAVASCULARTRANSFUSION (IVT) DENNIS HARPER1, HANES SWINGLE1, JOHN WIDNESS1,DANIEL BONTHIUS1, GLEN AYLWARD2, CARL P. WEINER3, 1University of Iowa, Pe-diatrics, Iowa City, Iowa, 2Southern Illinois University, Pediatrics, Springfield,Illinois, 3University of Maryland at Baltimore, Physiology and Obstetrics,Gynecology and Reproductive Sciences, Baltimore, Maryland

OBJECTIVE: We tested the hypothesis that long-term neurodevelopmental /neuropsychological outcomes of IVT treated immune hydrops (IH) fetuses aresimilar to unaffected siblings using of a specific IVT protocol that addresses theunderlying pathophysiology of IH.

STUDY DESIGN: Eighteen consecutive IH fetuses were treated by onephysician from July 1985 to October 1995; 16 (89%) survived. Because IHreflects cardiac dysfunction secondary to inadequate oxygen carrying capacitycharacterized by increased UV pressure (UVP), the IVT protocol utilized a 2-step HCT deficit correction (final target HCT 50%) over a 2-4d intervalcombined with UVP measurement to avoid over transfusion and bicarbonateadministration to assure a post transfusion UVpHO7.30. All survivorsparticipated in the follow-up at a mean age of 10y. Unaffected older siblingsserved as controls. Physical and neurological exams and comprehensiveneuropsychological evaluations were performed. The neuropsychological testswere a balance of tasks to identify general and subtle neurocognitive deficits.They included a comprehensive neuropsychological battery measuring verbaland nonverbal language, spatial tasks, multiple measures of memory, visualmotor coordination and perception, and a computerized measure of attention.

RESULTS: Two (12%) had major neurological sequelae associated witheither procedural (post IVT bradycardia after umbilical artery transfusion) orneonatal (kernicterus) complications. Each occurred early in the learningcurve. IH children had physical, neurological, and neuropsychological,including cognitive, outcomes statistically similar to their siblings, except forlower scores on a computerized measure of attention. Fourteen IH childrenhad a normal head CT scan and two minor abnormalities (NS from unrelatedcontrols). All 16 children had normal brain volumes.

CONCLUSION: This, the most comprehensive and longest follow-up of IH,is good news for fetal medicine. IVT for IH is associated with a high likelihoodof survival and favorable neurological and neuropsychological long-termoutcomes.

498 DOES A MYELOMENINGOCELE SAC COMPARED TO NO SAC INCREASE MORBIDITY?ROBERT WILSON1, M. P. JOHNSON2, M. W. BEBBINGTON2, A. W. FLAKE2, H. L.HEDRICK2, L. N. SUTTON3, N. S. ADZICK2, 1University of Pennsylvania, Surgery/OBGYN, Philadelphia, Pennsylvania, 2Children’s Hospital of Philadelphia,Surgery, Philadelphia, Pennsylvania, 3Children’s Hospital of Philadelphia,Neurosurgery, Philadelphia, Pennsylvania

OBJECTIVE: A recent fetus with a large sac S1 myelomeningocele (MMC)but bilateral talipes prompted the question’ does the presence or size of aMMC sac affect postnatal leg function’?

STUDY DESIGN: A MMC database with prenatal, birth and a minimum ofone year follow-up evaluations was reviewed (IRB 2002-2354) . All fetuses hadundergone in-utero MMC repair (prior to NICHD MOMS trial) at 20C0 to25C6 weeks at a single institution . Fifty-four fetuses had prenatal evaluationwith 48 children completing a birth and one year evaluation of leg neuro-function. Sac presence and volume were calculated from presurgical ultra-sound imaging. Pearson statistical analysis was used to compare functionaloutcome differences between MMC sac and MMC no sac.

RESULTS: A MMC sac was present in 38/54 (70%) of fetuses evaluated in-utero and had been present in 35/48 (73%) of children evaluated at one year ofage. Sac volume was !5cc, 5-10cc (5.3,6.9,7.0,7.4), and O10cc (10.6,11.1,11.3,12.6,26.3,55.5) in 28, 4, and 6 fetuses respectively. Leg functional levels were 2or more levels higher (worse) than the bony level when a sac was present in23% and 17% compared to no sac in 0% and 0% at birth and one yearevaluation respectively (p = 0.0951) . At one year, MMC sac fetuses had legneurofunctional exams that were better (17%), equal to (66%) or worse (17%)than predicted by bony lesion levels while fetuses without sacs were better(30%) and equal to (70%) predicted level . Spinal level was a signficant factorfor function but it was presence or absence of a MMC sac not the size that wasthe added factor to affect function.

CONCLUSION: The presence of a MMC sac may cause spinal nerve damageat a higher level than the bony level due to traction on the cord or nerve roots .Sac size/volume or spinal level did not appear to contribute to this observationbut this may be due to small patient population.

S144 SMFM Abstracts