690 Equivalence of erythromycin and azithromycin for treatment of PPROM Shari Gelber 1 , Elyssa Brent 1 , Aneesha Varrey 1 , Bella Fridman 2 , Katherine Sapra 3 , William Frayer 2 1 Weill Cornell Medical College, Obstetrics and Gynecology, New York, NY, 2 Weill Cornell Medical College, Pediatrics, New York, NY, 3 Columbia University, Pediatrics, New York, NY OBJECTIVE: A regimen of ampicillin and erythromycin (ERY) improves outcomes in patients with Preterm Premature Rupture of Membranes (PPROM). National shortages of ERY, patient sensitivity to ERY, and similar mechanism of action have encouraged providers to substitute Azithromycin (AZ). Some expert guidelines now recommend use of AZ, despite limited knowledge of efficacy in this setting. We sought to determine if treatment with AZ is equivalent to treatment with ERY for PPROM. STUDY DESIGN: Retrospective chart review of patients with PPROM admitted to labor and delivery at an academic center who received either ERY or AZ between January 2009 and March 2012. Inclusion criteria were gestational age 24-34 wk, planned expectant manage- ment and receipt of 1 dose of ERY or AZ. Exclusion criteria were delivery prior to receipt of antibiotics or receipt of both ERY and AZ. At our institution, decision to use AZ is determined by unavailability or patient intolerance of ERY and is not practitioner dependent.  2 and Fisher tests for categorical variables and Mann-Whitney test for continuous variables were done. Primary composite outcome was in- fant death, RDS, grade 3-4 IVH, NEC or sepsis 72 h of life. RESULTS: 103 patients met inclusion criteria; 7 were excluded for re- ceiving both ERY/AZ. Data from 96 patients were analyzed. 67 pa- tients received ERY and 29 patients received AZ. There were no sta- tistically significant differences in age, ethnicity, BMI, GA at PPROM (ERY: 30.42 vs AZ: 30.85 p=0.53), history of preterm deliveries, par- ity, smoking, GBS status, or receipt of ampicillin or steroids between groups. There were no statistically significant differences in maternal or fetal outcomes between the groups(Table 1). There was no differ- ence in latency (ERY: 65.6h vs AZ 61.7h). CONCLUSION: AZ provides the same benefit as ERY in the setting of PPROM. Although there is no significant difference in outcomes between groups, the trend toward improvement in several outcomes indicates studies are warranted to see if AZ is the preferred agent for PPROM. 691 The effect of maternal age on the risk of stillbirth if remaining undelivered Alicia Mandujano 1 , Thaddeus Waters 1 , Stephen Myers 1 1 Metrohealth Medical Center, Obstetrics and Gynecology, Maternal-Fetal Medicine, Cleveland, OH OBJECTIVE: To examine the effect of maternal age at delivery on the risk of fetal death for pregnancies that remain undelivered between 34-42 weeks gestational age. STUDY DESIGN: Linked birth and infant death data for the U.S. from the National Center for Health Statistics analyzed non-anomalous singleton pregnancies between 2003 and 2005. Pregnancies were clas- sified according to maternal age at delivery (18, 18-24, 25-29, 30-34, 35-39 and =40). Outcomes of 8,785,132 live births and 12,777 fetal deaths between 34 and 42 completed weeks gestation were examined. The risk of fetal death if remaining undelivered (FDRRe) was determined using the following equation: (# fetal deaths that occurred at a given GA + all fetal deaths that occurred after that GA/Total deliveries at a given GA + all remaining undelivered pregnancies). The FDRRe for each maternal age group was compared to the neonatal death rate (NDR) for each week of gestation. Additional subgroup comparisons were made for women with (HR) and without (LR) medical complications of pregnancy. RESULTS: At each week of gestation, the FDRRe was similar for all maternal age groups except for women =40 (Figure 1). When com- pared to the neonatal death rate, the number of fetal deaths that could be avoided by delivery exceeded the NDR between 35 and 36 weeks gestation for women =40 and between 37-38 weeks gestation for women 40. The FDRRe was greatest for women =40 with medical complications (Figure 2). CONCLUSION: These findings suggest that maternal age = 40 at de- livery is an independent additional risk factor for fetal death. Delivery beyond 36 weeks gestation for women =40 could result in an in- creased number of perinatal deaths. The observed FDRRe for women =40 is independent of preexisting medical conditions. 692 Oxytocin receptor polymorphisms, postpartum mood and breastfeeding duration Alison Stuebe 1 , Alison Wise 2 , Thutrang Nguyen 3 , Samantha Meltzer-Brody 4 , Karen Grewen 4 , Anna-Maria Siega-Riz 5 1 University of North Carolina School of Medicine, Obstetrics and Gynecology, Chapel Hill, NC, 2 UNC Gillings School of Global Public Health, Biostatistics, Chapel Hill, NC, 3 Harvard Medical School, Division of Genetics and Endocrinology, Boston, MA, 4 University of North Carolina School of Medicine, Psychiatry, Chapel Hill, NC, 5 UNC Gillings School of Global Public Health, Epidemiology, Chapel Hill, NC OBJECTIVE: Oxytocin plays a central role maternal behavior and lacta- tion. We sought to determine the extent to which oxytocin receptor (OXTR) polymorphisms are associated with perinatal mood disor- ders and breastfeeding duration. STUDY DESIGN: We genotyped 260 Caucasian women in the Preg- nancy, Infection and Nutrition Postpartum Study for three OXTR single-nucleotide polymorphisms (SNPs). In prior work, authors found that social support buffers stress-induced cortisol increases among carriers of the low risk rs53576 G allele, but not among ho- mozygotes for the A allele. The OXTR rs2254298 G and rs1042778 T alleles have been associated with reduced parental sensitivity and lower circulating oxytocin levels. We hypothesized carriage of these risk alleles would be associated with postpartum depression and reduced breastfeeding duration. Postpartum mood was measured using the Ed- inburgh Postnatal Depression Scale (EPDS). Breastfeeding duration was assessed by interview at 3, 12 and 36 months postpartum. We used Fish- er’s exact tests and Cox proportional hazards regression to measure the association between OXTR polymorphisms and postpartum phenotype. P values of 0.05 were considered statistically significant. RESULTS: Among the 260 women in our study, 11 women (4.2%) had symptoms of postpartum depression at three months postpartum. Among participants homozygous for the low risk OXTR rs53576 G allele, 0.9% (1/116) had depressive symptoms (EPDS13), compared Outcomes by administration of erythromycin vs azithromycin a. Fisher exact; b. Chi-square. www.AJOG.org Clinical Ob, Epidemiology, ID, Intrapartum Fetal, Operative Ob, Med-Surg-Diseases, Ob Quality & Safety, Public & Global Health Poster Session V Supplement to JANUARY 2013 American Journal of Obstetrics & Gynecology S291