5

5

5

5

www.AJOG.org SMFM Abstracts

57 USE OF ETHNIC-SPECIFIC MEDIANS FOR HISPANIC PATIENTS REDUCES ETHNICDISPARITIES IN MULTIPLE MARKER SCREENING LUISA WETTA1, JOSEPH BIGGIO1,JOHN OWEN1, 1University of Alabama at Birmingham, Birmingham, Alabama

OBJECTIVE: Race affects analyte concentrations used for maternal multiplemarker serum screening, so laboratories derive race-specific medians for black andnon-black patients. Since Hispanic ethnicity is included in the non-black group, weinvestigated whether Quad screen analyte concentrations differ between Caucasianand Hispanic women and whether using ethnic-specific medians affects screeningperformance.

STUDY DESIGN: With IRB approval, we identified all patients with a Quadscreen (15-22 weeks) performed in our laboratory (11/04 – 6/08). We identifiednon-black patients and categorized self-reported ethnicity as Caucasian or His-panic, limiting our analysis to non-diabetics with singleton gestations. Medianswere derived using commercially available software. AFP, estriol, hCG, and inhibinA medians were determined for the Composite group and individually for Cauca-sians and Hispanics. Using the Composite medians, the mean MoM for each ana-lyte was compared between groups. Using ethnic-specific medians, new MoMswere calculated and utilized with age in a multiple marker risk estimation algo-rithm. The screen positive rates (SPR) for trisomy 21 (T21) and neural tube defects(NTD) were calculated.

RESULTS: 5,478 Caucasian and 2,246 Hispanic pregnancies were evaluated.Intergroup mean MoMs were significantly different for all analytes: AFP, hCG, andinhibin were lower in Hispanics, while estriol was higher (p�0.0001). Using theComposite medians, the SPR for T21 was 5.39% in Caucasians and 3.29% in His-panics. Utilization of ethnic-specific medians reduced the ethnic disparity--the SPRwas 4.76% in Caucasians and 4.05% in Hispanics. The SPR for NTD using com-posite medians was 1.44% for Caucasians, and 0.89% for Hispanics, whereas withethnic-specific medians the SPR was 1.42% for Caucasians and 1.07% for Hispan-ics.

CONCLUSION: Serum analyte concentrations differ between Caucasian andHispanic gravidas. Use of ethnic-specific medians reduces the ethnic disparity inthe screen-positive rates for T21 and NTD and should improve screening perfor-mance.

0002-9378/$ - see front matterdoi:10.1016/j.ajog.2008.09.587

58 THE EFFECTS OF SSRI EXPOSURE ON FETAL, MATERNAL AND PLACENTAL VASCULARPARAMETERS KENNETH LIM1, DANNY RURAK1, TIM OBERLANDER2, ARI SANDERS3,WAYNE RIGGS2, 1University of British Columbia, Vancouver, British Columbia, Can-ada, 2University of British Columbia,, British Columbia, Canada, 3Univeristy ofBritish Columbia,, British Columbia, Canada

OBJECTIVE: The objective of the study was to examine the effects of SSRI ex-posure on fetal, maternal and placental arterial vascular parameters.

STUDY DESIGN: Prospective observational cohort. The PI, RI and Flow weremeasured in fetal (middle cerebral, right pulmonary artery) and maternal (rightand left uterine artery) using standard ultrasound doppler techniques on an AlokaProsound 5500. Umbilical artery PI and RI was also determined. At least 5 mea-surements were done for each parameter in patients exposed (N�14) and not-exposed (N�26) to SSRI at both trough (AM) and 4.5 hours post morning dose(PM).

RESULTS: MCA PI was signficantly lower in the exposed fetuses (1.81[SD�0.38] vs 1.55[SD�0.28], P�0.01) vs non exposed fetuses in the AM but not inthe PM. This was due to a significant decrease in PI and increase in flow in thecontrol fetuses (AM vs PM) that did not occur in exposed fetuses. A trend towardssignificance (P�0.05) in MCA flow was found in the PM (204.6 [SD�12] vs 129.5[SD�41.5], in exposed vs non exposed fetuses but not in the AM. In the non-exposed fetuses, a significant change in UA PI between AM and PM was seen in thenon-exposed fetuses (0.94 [SD�0.15] vs 0.88 [SD�0.16], P�0.03), but not in theexposed fetuses.

CONCLUSION: SSRI exposure may alter various vascular parameters in the fetaland placental compartments.

0002-9378/$ - see front matterdoi:10.1016/j.ajog.2008.09.588

Supplemen

59 PRENATAL ARRAY COMPARATIVE GENOMIC HYBRIDIZATION ANALYSIS AMONGSTWOMEN CHOOSING INVASIVE PRENATAL DIAGNOSIS: A DECISION ANALYSIS SUSANTRAN1, BRIAN SHAFFER1, MARY NORTON2, KATHERINE MALABED1, SHERRI PENA1, AARONCAUGHEY1, 1University of California, San Francisco, Obstetrics, Gynecology, andReproductive Sciences, San Francisco, California, 2Kaiser Permanente, Obstetricsand Gynecology, San Francisco, California

OBJECTIVE: To examine the decision of whether to undergo prenatal arraycomparative genomic hybridization (aCGH) analysis for the detection of chromo-somal imbalances amongst women already electing to undergo invasive prenataldiagnosis (PNDx).

STUDY DESIGN: A decision-analytic model was designed for women who havealready elected to undergo PNDx. The model compared: (1) undergoing aCGH todetermine if the fetus is affected by a microdeletion/duplication syndrome, a find-ing of unknown clinical significance (FUS), or a normal outcome to (2) not under-going aCGH. Baseline assumptions included a live-birth incidence of known ge-netic syndromes and subtelomeric rearrangements of 1:1000 and a 1% rate (range:0.66 - 2%) of detecting a FUS after parental specimens have been analyzed. Out-comes were assigned utilities and quality adjusted life years (QALYs) were calcu-lated. Inputs were varied over a wide theoretical range in sensitivity analyses.

RESULTS: In the baseline analysis, obtaining an aCGH analysis is the preferredoption leading to greater diagnoses of affected fetuses with an improvement of0.00132 QALYs. However, this decision was extremely sensitive to both the risk ofgenetic syndromes and the risk of FUS (figure). With a rate of FUS of 2% or aboveat a rate of genetic syndromes and subtelomeric rearrangements of 1/1000, indi-viduals would do better not to obtain this information.

CONCLUSION: While aCGH can provide valuable prenatal diagnostic informa-tion, the decision whether to provide it to all women is dependent upon both therisk of genetic syndromes that would be potentially diagnosed or missed and therate of FUS. Further research in this area would be well-served to characterize theseFUS and to delineate the rate of microdeletion/duplication syndromes.

0002-9378/$ - see front matterdoi:10.1016/j.ajog.2008.09.589

60 IDENTIFICATION OF RNA-SNP MARKERS FOR NONINVASIVE PRENATAL DIAGNOSIS(NIPD) OF T18 AND T13 MENGIA TANG1, MIN LEE1, FIROUZ MOHSENIAN1, TAO SHI1,BETTY DRAGON1, JIAN-HUA DING1, YANFENG YANG1, 1SEQUENOM, Inc., San Diego,California

OBJECTIVE: Trisomy 18 and 13 are the only two live born trisomies apart fromDown syndrome. Less than 50% of these trisomy disorders survive to full termbirth. and rarely do affected infants survive one year of life. In this study, we use asystematic screening approach to identify chromosome 18 and 13 (C18 and C13)RNA-SNP markers. These markers enable further test development of NIPD testsfor T18 and T13.

STUDY DESIGN: An exon array was used to compare gene expression profilesand identify SNPs using matched placenta and maternal PBMC RNA samples. AllSNP candidates were then screened using 100 human genomic DNA samples tomeasure the heterozygote rate (HR) for each SNP. SNPs with an HR of 4% wereretested using placenta RNA samples.

RESULTS: Five C13 genes and six C18 genes with differential expression wereidentified using exon arrays. 21 C13 and 44 C18 SNP candidates from these geneswere screened against the DNA samples. 16 C13 and 37 C18 SNPs showed a HR of�4%. Five SNPs from two C13 genes and 13 SNPs from six C18 genes were selectedfor further evaluations and assay development based on positive placenta RNAresults. Optimum assay conditions have been achieved by Design of Experiment forsample preparation, reverse transcription PCR and detection.

CONCLUSION: We have identified C18 and C13 RNA-SNP markers that will beused to develop an NIPD for T13 and T18.

0002-9378/$ - see front matterdoi:10.1016/j.ajog.2008.09.590

t to DECEMBER 2008 American Journal of Obstetrics & Gynecology S163