490: Two year infant outcomes for children exposed to supplemental intravaginal progesterone gel in utero: secondary analysis of a multicenter, randomized, double-blind, placebo-controlled trial

Abstract

supplemental intravaginal progesterone gel in utero: secondary analysis of a multicenter, randomized, double-blind, placebo-controlled trial John M O’Brien, Jean J. Steichen, James A. Phillips, George W. Creasy University of Kentucky, Ob-Gyn, Lexington, KY, University of Cincinnati, Perinatal and Neonatal Medicine, Cincinnati, OH, Sage Statistical Solutions, Inc., Statistics, Efland, NC, Columbia Laboratories, Inc, Clinical Research and Development, Livingston, NJ OBJECTIVE: To describe select infant follow up for children born to women enrolled in the largest randomized controlled trial evaluating supplemental intravaginal progesterone for prevention of preterm birth. STUDY DESIGN: Infants born to women enrolled in the Progesterone Vaginal Gel Trial were followed at 6, 12, and 24 months to evaluate: 1) Biometry parameters (weight, length, and head circumference), 2) Developmental outcomes including cognition assessed by the Denver II Developmental Screening Test, 3) Identification of chronic morbid conditions, and 4) Congenital abnormalities not detected at birth. The treatment was 90 mg of intravaginal progesterone (or placebo) daily from 18-22 weeks until term, PROM, or preterm birth. RESULTS: Of the 637 patients treated, 445 were evaluated at 6 months (69.9%), 389 at 12 months (61.1%), and 293 at 24 months (46.0%) for at least one outcome variable. No meaningful differences were identified for infant biometry between groups or in their developmental screening results, congenital abnormalities, or chronic morbid conditions during follow-up. Biometry parameters are summarized in the table. Sub-analysis of these parameters of infants born either preterm or term also did not identify differences between groups. Suspected Denver II debilities were observed at any time during the 24 month follow-up in 33 infants in each group (10.3% progesterone and 10.4% placebo; Relative Risk (RR) 0.98; 95% confidence interval (CI) 0.62 1.55). Congenital abnormalities were identified in infants of 12 (3.8%) placebo subjects and 11 (3.4%) progesterone subjects (RR 0.90; 95% CI 0.40 2.01). Chronic morbid conditions were diagnosed at any time during follow up in 17 (5.4%) of placebo subjects infants and 18 (5.6%) progesterone subjects infants (RR 1.04; 95% CI 0.55 1.99). CONCLUSION: These results demonstrate similar follow up comparing infants exposed to supplemental progesterone in utero to placebo for a 90 mg daily dose. No detrimental effect related to supplemental progesterone administration was identified for up to 24 months follow up. 491 Bedside clinician IGFBP-1 immunoassay for ROM is feasible and accurate Pat Newcomb, Afreen Tariq-Fazili, Jorge Carrillo, Christopher LaRussa, Toni Golen Rochester General Hospital; Rochester Institute of Technology; University of Rochester, Rochester General Hospital OB GYN, Rochester, NY, Rochester General Hospital OB GYN Residency, Rochester General Hospital OB GYN, Rochester, NY, St Joseph Hospital Health Center, Department of Obstetrics and Gynecology at St Joseph Hospital Health Center, Syracuse, NY, Beth Israel Deaconess Medical Center, Dept. OB GYN Beth Israel Deaconess Medical Center, Boston, MA OBJECTIVE: To determine the accuracy of an insulin-like growth factor binding protein-1 (IGFBP-1)laboratory immunoassay as a bedside clinician test. STUDY DESIGN: Prospective ’gold-standard’ comparison in sequential consenting patients at three sites. Each local IRB approved the protocol: Women between 15-40 weeks gestation with signs and/or symptoms of PROM were enrolled. Presence of any 2 of 3 clinical signs: pooling of clear fluid in the vagina, an alkaline pH of the vaginal fluid, and microscopic ferning of the dried vaginal fluid was considered positive ROM (gold standard). Results of the study test were not used in clinical decision making. Age, race, EGA, presenting complaint and clinical details were recorded. At the time of evaluation for PROM a sample of vaginal secretions was obtained with a sterile swab according to the PRO-M Kit(Pro-Lab, Ontario, Canada) instructions. The vaginal swab was placed in extraction buffer immediately after sampling and the extracted sample was tested within six hours by an individual who was blinded to the clinical diagnosis. All sampling and testing was done by clinical staff including nurses, midwives, doctors, and residents. Whether or not delivery occurred within 48 hours was also documented using a clinical correlator unaware of the outcome of any test or exam. The PRO-M test results were compared to the gold standard. Sensitivity, Specificity, Positive % agreement, Negative % Agreement, and relative agreement were calculated with 95% confidence intervals. Detection of equivalance was calculated to be reached at 295 subjects with a rupture prevalence of 25%. RESULTS: N 272. Groups were the same in age (27.3 / 6.4 SD) and EGA (37.1 / 4.2). ROM by gold standard differed by site[p 0.001]. (BI 28.1%, RGH 57.7%, St. J 41.9%; mean 41.9%). As ROM in study was greater than estimated, the study halted at 272 subjects. The table shows that sensitivity was 91.2% and specificity 84.8%, with a relative agreement of 87.5%. CONCLUSION: Our results are similar to the published accuracy of the laboratory test. Bedside clinician ROM testing is feasible and accurate. www.AJOG.org Doppler Assessment, Fetus, Prematurity Poster Session III