998 Association between labor and neonatal respiratory morbidity in pregnancies complicated by gestational diabetes mellitus Hope Y. Yu, Phinnara Has, Melissa A. Clark, Lindsay Kanno, Dwight Rouse, Erika Werner Brown University, Women & Infants Hospital, Providence, RI OBJECTIVE: Neonates born to women with gestational diabetes mellitus (GDM) have historically been thought to be at increased risk of respiratory morbidity but it is uncertain whether or how exposure to labor affects this association. Therefore, we sought to assess the risk of respiratory complications in neonates who were delivered after labor compared to those who did not experience labor among a cohort of neonates born to women with GDM. STUDY DESIGN: A prospective cohort study of singleton pregnancies complicated by GDM in which the infants were live born at 34 weeks. The primary outcome for this analysis was respiratory distress syndrome (RDS) or transient tachypnea of the newborn (TTN) which resulted in neonatal intensive care unit (NICU) admission. Neonates who experienced labor (either spontaneous or induced) were compared to those who delivered by cesarean without labor. Covariates, including gestational age, were investigated and adjusted for if the p-value was <0.05. RESULTS: Of the 292 neonates meeting study inclusion criteria, 25 (8.6%) developed RDS or TTN and required NICU admission (median length of stay ¼ 7 days). Seventy four delivered without labor (25 by primary cesarean delivery [CD] and 49 by repeat CD) and 218 delivered after labor (46 by CD and 172 vaginally). Those who delivered without labor and those who labored delivered at similar gestational ages (38.4 v. 38.6 weeks), but differed with regard to age and race. Not laboring, compared to laboring, was associated with a significantly higher risk of the primary outcome, 13.5% v. 4.1% (OR 3.6; 95% CI 1.4-9.3). Even after adjusting for age and race, those who did not labor had a higher risk of the primary outcome (aOR 4.2; 95% CI 1.2-14.9). CONCLUSION: Exposure to labor is associated with a decreased risk of respiratory morbidity in neonates born in the late preterm or term period to mothers with GDM. 999 The Impact of Gestational Diabetes in Twin Pregnancies on Maternal and Neonatal Outcomes Kavita Vani 1 , Christina Scifres 2 , Katherine Himes 3 1 Magee-Womens Research Institute, University of Pittsburgh School of Medicine, Dept of OB/GYN/RS, Pittsburgh, PA, 2 University of Oklahoma Health Sciences Center, Oklahoma City, OK, 3 Magee-Womens Research Institute, University of Pittsburgh School of Medicine, Dept of OB/GYN/RS, Division of MFM, Pittsburgh, PA OBJECTIVE: Studies of gestational diabetes (GDM) in twin pregnan- cies are limited by small sample size and unknown chorionicity. We sought to determine how rates of GDM vary by chorionicity and determine the impact of GDM on maternal and neonatal outcomes in a large cohort of well-characterized twin pregnancies. STUDY DESIGN: This is a retrospective cohort study of all twin de- liveries from Magee-Womens Hospital, University of Pittsburgh between 1998 and 2013. Monochorionic monoamniotic twins were excluded. Maternal and neonatal data were collected using a peri- natal database, chart abstraction, and placental review. Large and small for gestational age (LGA/SGA) were defined as less than 10% using the NICHD growth study. Given the limited number of monochorionic diamniotic (mono/di) twins with GDM, univariate and multivariable logistic regression were used to interrogate the association between GDM and perinatal outcomes among dichorionic twins. Given GDM is a maternal exposure, analysis was done at the maternal level and outcomes reported are for either baby. Prepregnancy body mass index, race, parity, maternal age and type of assisted reproductive technology were included as potential confounders. RESULTS: 2927 twin pregnancies were included in the cohort: 79.4% (n¼2323) were dichorionic and 20.6% (n¼604) were mono/di. The median gestational age at delivery was 36 weeks (IQR: 33.4,37.5). 23.6% of women were obese, 468 (16%) conceived with in vitro fertilization and the average age was 30 years (+/-5.8y). The overall rate of GDM was 6.4% (n¼188) and there was no difference in rate of GDM by chorionicity with 154 (6.6%) cases in di/di twins and 34 (5.6%) cases in mono/di twins (p¼0.47). Among dichorionic twins, GDM was associated with an increased risk of preeclampsia with an OR ¼ 2.0 (CI: 1.4, 2.8; p<0.001). After adjusting for confounders, this remained significant with an aOR ¼ 1.6 (CI:1.1, 2.3, p¼0.02). Among dichorionic twins, GDM was not associated with an increased risk of cesarean delivery, LGA, SGA, or admission to the NICU for either twin (Table 1). CONCLUSION: The rate of GDM in this cohort was similar to rates in singleton pregnancies. Dichorionic twin pregnancies affected by GDM have an increased risk of preeclampsia compared to twin pregnancies without GDM. Neonatal outcomes were not different. Future studies to determine if glycemic control modifies these as- sociations are needed. Association between gestational diabetes and perinatal outcomes among dichorionic twins Gestational Dibetes (n=154) No Gestational Diabetes (n=2169) OR (95% CI) aOR (95% CI)* Preeclampsia 45 (29.2) 376 (17.3) 2.0 (1.4, 2.8) 1.5 (1.1, 2.3) Cesarean Delivery 100 (64.9) 1156 (53.2) 1.6 (1.2, 2.2) 1.3 (0.9, 1.9) Admission of either twin to NICU 61 (39.6) 759 (35.5) 1.2 (0.9, 1.7) NA LGA infant 19 (12.3) 212 (10.0) 1.2 (0.8, 2.1) NA SGA infant 17 (11.0) 232 (10.7) 1.0 (0.6, 1.7) NA Data are presented as n (%). OR = Odds ratio. CI = Confidence interval. NA = not applicable. *Adjusted for pre-pregnancy body mass index, race, parity, maternal age and type of assisted reproductive technology 1000 Fibroblast growth factor 21 and metabolic dysfunction in women with a prior glucose intolerant pregnancy Celeste P. Durnwald for the Eunice Kennedy Shriver NICHD Maternal Fetal Medicine Units Network, Bethesda, MD OBJECTIVE: Fibroblast growth factor 21 (FGF21) is associated with Type 2 diabetes (DM) and microvascular complications of DM. Therefore, we sought to compare FGF21 levels in women with and without metabolic dysfunction 5-10 years following a prior preg- nancy affected by glucose intolerance. STUDY DESIGN: Secondary analysis of women enrolled during preg- nancy with and without mild gestational diabetes (GDM) in a multicenter trial. At 5-10 years after the index pregnancy, women underwent anthropometric and blood pressure measurements and fasting blood was obtained. Women were then categorized as having no metabolic dysfunction, metabolic syndrome, or Type 2 DM at follow-up. Metabolic syndrome was defined per the AHA/NHLBI Scientific Statement. DM was defined as fasting glucose 126mg/dL, 2-hour glucose 200mg/dL after a 75 gram load, or undergoing current treatment for DM. FGF21 levels, using the Quantikine Human FGF-21 Immunoassay, were compared between women who did and did not have a history of GDM using multivariable linear regression. RESULTS: Of 1889 women enrolled in the original trial, 982 (52%) participated in the follow up study. 796 women had maternal plasma Poster Session V ajog.org S590 American Journal of Obstetrics & Gynecology Supplement to JANUARY 2018