Peak serum estradiol and neonatal birth weight following in vitro fertilization

OBJECTIVE: To determine if peak serum estradiol (E2) during ovarian hyperstimulation for in vitro fertilization (IVF) is associated with birth weight. STUDY DESIGN: Retrospective cohort study of women aged 18–42 with a singleton live birth after fresh, autologous IVF between January 1, 2003, and December 31, 2014, at an academic institution. RESULTS: A total of 183 women were included in the study. Univariate analysis demonstrated a reduction in mean birth weight of 45.28 g (95% CI-84.14 to-6.39) for each 500 pg/mL increase in peak E2. After adjusting for potential confounders, mean birth weight decreased by 33.81 g for every 500 pg/mL increase in peak E2 (95% CI −64.31 to −3.30). The optimal peak E2 cut-point to predict low birth weight (LBW) based on the Youden index was 3,991 pg/mL, yielding 90% specificity, 33% sensitivity, 94% negative predictive value (NPV), and 23% positive predictive value (PPV). This dichotomized peak E2 together with age, parity, body mass index, ovulation trigger type, gestational age, and infant sex gave an area under the curve of 89% for LBW. CONCLUSION: Peak E2 during stimulation for IVF is negatively associated with neonatal birth weight. A peak E2 cut-point of 3,991 pg/mL has a high NPV but low PPV for LBW.

Duke Scholars

Author Yi-Ju Li Biostatistics & Bioinformatics, Division of Integrative Geno ...

Author Tracy Truong