567: Comparing the sensitivity of the Hadlock and Williams growth curves in predicting perinatal morbidity and mortality

Abstract

growth curves in predicting perinatal morbidity and mortality Jonathan M. Snowden, Amy E. Doss, Aaron B. Caughey, Yvonne W. Cheng Oregon Health and Science University, Department of Obstetrics & Gynecology, Portland, OR, Oregon Health & Science University, Department of Obstertics and Gynecology, Portland, OR, University of California, San Francisco, Obstetrics & Gynecology, San Francisco, CA OBJECTIVE: To compare two commonly applied fetal growth curves, Hadlock and Williams, with regards to their sensitivity in identifying small for gestational age (SGA) births that are at elevated risk of adverse perinatal outcomes. STUDY DESIGN: This was a population-based retrospective cohort study of births in the US in 2005, as recorded in the National Center for Health Statistics natality database. We categorized all singleton births as SGA ( 10th percentile of birthweight at a given gestational age) or non-SGA according to two fetal growth curves: the Hadlock and the Williams curves. We fit multivariable logistic regressions to analyze the impact of SGA on several fetal and neonatal outcomes: intrauterine fetal demise (IUFD), infant death, and neonatal intensive care unit (NICU) admission, controlling for confounders including parity, advanced maternal age, and race. RESULTS: Of the 3,705,893 singleton deliveries (births and IUFDs), all 238,799 deliveries categorized as SGA by Williams were also categorized thusly by Hadlock (6.4%). An additional 160,181 (4.3%) were categorized as SGA by Hadlock but not by Williams. The regression analysis demonstrated that births considered SGA by both Hadlock and Williams had significantly elevated odds for all perinatal outcomes analyzed, with adjusted odds ratios (aORs) ranging from 3.66 (95% Confidence Interval: 3.55 3.77) for NICU admission to 8.35 (7.81 8.93) for IUFD. The models also demonstrated that the births identified as SGA by only Hadlock also had significantly increased odds of adverse perinatal outcomes, but with an attenuated magnitude. CONCLUSION: The SGA cutoffs from both the Hadlock and the Williams growth curves identified deliveries with increased odds of IUFD, infant death, and NICU admission. While the more restrictive Williams cutoffs identified births at highly elevated risk, the deliveries identified as SGA by only Hadlock were also at increased risk for these adverse outcomes. Clinicians should be aware that even within a binary category (e.g. SGA and non-), there exists a gradient of risk. 568 Risk factors for spontaneous preterm birth in African-American and Caucasian women receiving 17 -hydroxyprogesterone caproate Julia Timofeev, Jasbir Singh, Niki Istwan, Debbie Rhea, Rita Driggers Washington Hospital Center, Obstetrics and Gynecology, Washington, DC, Alere Health, Department of Clinical Research, Atlanta, GA OBJECTIVE: To determine risk factors for recurrent preterm birth in African-American and Caucasian women receiving 17 -hydroxyprogesterone (17P). STUDY DESIGN: Retrospective analysis of a cohort of women with prior spontaneous preterm delivery (SPTD) enrolled in an outpatient 17P administration program at 27 weeks’ gestation. Maternal characteristics, obstetric history and rates of recurrent SPTD were determined in women of African-American and Caucasian descent using 2 and multivariable logistic regression at two-tailed 0.05. Primary study outcome was the rate of recurrent SPTD 34 weeks. RESULTS: There were 1,913 African-American and 5,195 Caucasian women meeting the inclusion criteria. The overall rate of recurrent SPTD was 28.3%. Rates of recurrent SPTD 37 and 34 weeks were similar among women starting 17P at 16-20.9 vs. 21-26.9 weeks. Table 1 describes maternal characteristics and current pregnancy outcomes. Significant risk factors associated with SPTD 34 weeks in multivariable analysis are presented in Table 2. CONCLUSION: Cervical length 25mm before 27 weeks had the strongest association with SPTD 34 weeks within each maternal race. One or more prior term deliveries attenuated the risk for recurrent SPTD in African-American but not Caucasian women, while tobacco abuse increased the risk for recurrent SPTD in African-American but not Caucasian women. Delineation of race-specific risk factors associated with SPTD can facilitate counseling of women receiving 17P. PosterSessionIV Epidemiology, Infectious Disease, Intrapartum Fetal Assessment, Operative Obstetrics, Obstetric Quality & Safety, Public Health-Global Health www.AJOG.org