Metabolic score as a novel approach to assessing preeclampsia risk

S130 SMFM Abstracts 409 METABOLIC SCORE AS A NOVEL APPROACH TO ASSESSING PREECLAMPSIA RISK SINDHU SRINIVAS1, REBECCA MAZAR1, MICHAL ELOVITZ2, 1University of Pennsylvania, Obstetrics and Gynecology, Philadelphia, Pennsylvania,, 2 University of Pennsylvania, Philadelphia, Pennsylvania OBJECTIVE: Clinical trials have confirmed the presence of metabolic syndrome (MS) in women and its contribution to cardiovascular disease (CVD). Preeclampsia (PEC) and CVD have been noted to occur in similar populations with possible similar biologic mechanisms. We sought to determine if MS was associated with the development of PEC. STUDY DESIGN: This study was part of a large case control study, Preeclampsia: Mechanisms and Consequences. Cases are patients prospectively identified with PEC based on maternal criteria. Controls were patients presenting for delivery at term. Obstetric history, demographic, prenatal, and neonatal information were collected similarly for cases and controls. A pregnancy based metabolic score (0, 1, or 2C) was created using early pregnancy BMI (O = 30), chronic hypertension, and diabetes. Stratified analysis and logistic regression were used to evaluate the association of metabolic score with PEC and disease severity. RESULTS: 156 cases and 104 controls were evaluated. Among the cases, 45.5%, 38.5% and 16% women had a score of 0, 1, and 2C respectively. Among the controls, 63.5%, 27.9%, and 8.6% had a score of 0, 1, and 2C. Controlling for race and maternal age, women with PEC were more likely to have a MS of 1 (OR = 2.65, [1.39-5.02]) and 2C (OR = 3.18, [1.25-8.06]) compared to controls. A significant interaction was noted between MS and age. For women O = 30, a MS of 2C resulted in a 5.43 higher odds (CI [1.1625.5]) for PEC and a 9.05 higher odds (CI [1.76-46.6]) for severe PEC when compared to controls. Within cases, women O = 30 with a score 2C had a 10.88 higher likelihood of having severe PEC over mild PEC (p = 0.058). CONCLUSION: MS appears to be significantly associated with the development of PEC, particularly severe disease. The presence of components of the MS and advanced maternal age pose a significant risk for the development of PEC. More research is warranted to understand the contribution of the MS to the pathogenesis of PEC and to determine if the presence of MS and development of PEC results in long-term CVD in these women. 0002-9378/$ - see front matter doi:10.1016/j.ajog.2006.10.447

411 MATERNAL PLASMA AND AMNIOTIC FLUID SFLT-1 AND PLGF LEVELS AT THE TIME OF MID-TRIMESTER AMNIOCENTESIS IN WOMEN WHO SUBSEQUENTLY DEVELOP PREECLAMPSIA SHIN-YOUNG KIM1, HYUN-MEE RYU2, JAE-HYUG YANG2, MOONYOUNG KIM2, JUNG-YEOL HAN2, JOO-OH KIM2, JIN-HOON CHUNG2, SOYEON PARK1, MOON-HEE LEE1, DO-JIN KIM1, 1Cheil General Hospital and Woman’s Heathlcare Center, Laboratory of Medical Genetics, Seoul, South Korea, 2Cheil General Hospital and Women’s Healthcare Center, Sungkyunkwan University School of Medicine, Obstetrics and Gynecology, Seoul, South Korea OBJECTIVE: To determine whether the levels of soluble fms-like tyrosine kinase 1 (sFlt-1) and placental growth factor (PlGF) in maternal plasma (MP) and amniotic fluid (AF) are altered in women who subsequently develop preeclampsia. STUDY DESIGN: We performed a case-control study to compare the levels of sFlt-1 and PlGF in MP and AF of women who subsequently developed preeclampsia (n = 46) and normal pregnant women (n = 100) at the time of mid-trimester amniocentesis. The levels of sFlt-1 and PlGF were measured by enzyme-linked immunoassay. RESULTS: sFlt-1 levels in MP were significantly higher in preeclampsia than normal pregnancies, but PlGF levels were significantly lower (both for P ! .001). In AF, sFlt-1 levels were significantly decreased in preeclampsia compared with normal pregnancies, but PlGF levels were significantly increased (both for P ! .001). MP sFlt-1 levels of normal pregnancies were positively correlated with MP PlGF levels (r = 0.27, P = .008), whereas those of preeclampsia were negatively correlated with MP PlGF levels (r = ÿ0.423, P = .005). A significant correlation between AF sFlt-1 and PlGF levels was observed in normal pregnancies (r = 0.33, P = .001), but not observed in preeclampsia (r = 0.166, P = .277). Furthermore, the ratios of MP log[sFlt-1/ PlGF] were significantly higher, whereas those of AF log[sFlt-1/PlGF] were significantly lower in preeclampsia than normal pregnancies (both for P ! .001). There was significant correlation between MP and AF log[sFlt-1/PlGF] ratio in the study population (r = ÿ0.399, P ! .001). CONCLUSION: Preeclampsia is highly associated with the levels of sFlt1 and PlGF in MP and AF at the time of mid-trimester amniocentesis, which may provide an early prediction for the subsequent development of this disease based on a sFlt-1/PlGF ratio. 0002-9378/$ - see front matter doi:10.1016/j.ajog.2006.10.449

410 RETHINKING IUGR IN PREECLAMPSIA: DEPENDENT OR INDEPENDENT OF PAMELA NEFF1, MATERNAL CONDITION? SINDHU SRINIVAS1, ANDREA GOLDBERG1, MARY SAMMEL, SCD2, MICHAL ELOVITZ3, 1University of Pennsylvania, Obstetrics and Gynecology, Philadelphia, Pennsylvania, 2 University of Pennsylvania, Biostatistics, Philadelphia, Pennsylvania, 3 University of Pennsylvania, Philadelphia, Pennsylvania OBJECTIVE: Traditionally, intrauterine growth restriction (IUGR, !10%) has been included in the diagnosis of severe preeclampsia. These studies were performed to determine the validity of this classification and to investigate the relationship between severity of preeclampsia and preexisting maternal factors with IUGR. STUDY DESIGN: This study was part of a large case control study, Preeclampsia: Mechanisms and Consequences (PMC). Cases are patients prospectively identified with preeclampsia based on maternal criteria. Controls were patients presenting for delivery at term (O37 wks). Obstetric history, demographic, prenatal and neonatal information were collected similarly for all cases and controls. IUGR, preeclampsia and severity of disease was evaluated using chi-square analyses. Significant confounders were controlled for using multivariable logistic regression. RESULTS: The rates of IUGR (!10%) in patients with mild preeclampsia (n=61), severe preeclampsia (n=117) and controls (n=125) were 28%, 34% and 12%. Controlling for chronic hypertension, tobacco, diabetes, maternal age, race, and obesity, preeclamptic women have a 3.06 (CI [1.58-5.95]) and 4.61 (CI [1.68-12.68]) times increased odds of having a fetus with IUGR at !10% and !5% compared to controls. Severe preeclamptic patients (n=117) have an increased likelihood of IUGR over controls (!10% OR=3.57 ([1.757.28]) (!5% OR= 5.29 [1.83-15.32]). Within the cases, severity was not associated with IUGR !10% (OR=1.88 [0.87-3.99]). A history of preeclampsia did not increase the risk for IUGR (OR = 0.53, [0.15-1.82]). Obesity (BMI O = 30) in early pregnancy was not associated with severity of disease within cases (p = 0.48). CONCLUSION: Controlling for traditional confounders, preeclampsia is independently associated with the development of IUGR. Maternal disease severity does not correlate with poor fetal growth. The pathogenesis of IUGR in the setting of preeclampsia may be disparate from the mechanisms that result in severe maternal disease. Classification of severity of preeclampsia by fetal growth should be reevaluated. 0002-9378/$ - see front matter doi:10.1016/j.ajog.2006.10.448

412 OBSTETRIC OUTCOMES IN AGE MATCHED COHORTS OF OVUM DONORS AND IVF PATIENTS SHANI DELANEY1, PETER KLATSKY1, AARON CAUGHEY2, NAM TRAN1, GLEN SCHATTMAN3, 1University of California, San Francisco, Obstetrics, Gynecology and Reproductive Sciences, San Francisco, California, 2 University of California, San Francisco, Department of Obstetrics, Gynecology and Reproductive Sciences, San Francisco, California, 3Cornell University, Reproductive Medicine and Infertility, New York, New York OBJECTIVE: To determine the differences in outcomes between IVF patients undergoing ovum donation as compared to those utilizing their own ova. STUDY DESIGN: We conducted a retrospective study of patients undergoing ovum donor recipient (ODR) cycles and compared their obstetric outcomes with an age matched cohort of IVF patients who used embryos derived from their own ova. We reviewed 89 pregnancies in two age matched cohorts of patients undergoing IVF (n = 44) and ODR (n = 45). Chi square, t-test and multivariate regression was performed to assess the impact of embryonic origin on obstetric outcomes, including preeclampsia (PreE), pregnancy induced hypertension (PIH), gestational age (GA) and low birth weight infants (LBR). RESULTS: Both PIH and PreE were diagnosed more often in patients undergoing ODR cycles than IVF, 31% vs. 9% (p = 0.01) and 10% vs. 3% (p = 0.04), respectively. ODR patients were also at risk for having more LBW and preterm deliveries, 22% vs. 7% (p = 0.02) and 39% vs. 14% (p!0.01), respectively. ODR cycles resulted in mean birthweights of 363g lower than IVF cycles (CI = 90, 636g). ODR patients also had significantly higher rates of twin gestations, 14% vs. 31% (p = .05). After controlling for maternal age and twin gestations, the increased incidence of PIH among ODR cycles persisted with an odds ratio of 3.6 (95% CI, 1.03-12.49); differences in birth weight were no longer statistically significant, however a trend toward lower birth weights persisted. CONCLUSION: Providers can counsel patients and providers that while ovum donation increases pregnancy rates, patients remain at higher risk for other obstetric complications. Our findings support an embryonic role in obstetric outcomes and disease, particularly in the etiology of preE and PIH. Further investigation into the role of antigenicity and immune response are needed to explain these findings and better understand the etiology of preeclampsia. 0002-9378/$ - see front matter doi:10.1016/j.ajog.2006.10.450