Peak Serum Estradiol Correlates With Low Neonatal Birth Weight Following IVF

Among women undergoing in vitro fertilization (IVF), every increase of 500 pg/mL in peak serum estradiol levels corresponds with a decrease of 29.92 g in neonatal birth weight, according to findings from a new study. This is the first study to quantify the dose-response relationship between peak estradiol level and neonatal birth weight in the IVF setting.

Tia Jackson-Bey, MD, a fourth-year resident of the Division of Reproductive Endocrinology and Infertility at Duke University Medical Center, presented results from the retrospective cohort study at the Scientific Congress & Expo of the American Society for Reproductive Medicine, held October 15 to 19, 2016, in Salt Lake City, UT.

Among infants conceived with IVF, elevated peak estradiol levels are known to correspond with low birth weight (LBW), defined as weight less than 2500 g. However, the recommended thresholds for optimal peak estradiol level in the existing IVF literature were arbitrarily selected and without evidence of a clear dose-response relationship.

In the present study, Jackson-Bey and colleagues sought to examine the association between peak estradiol level and neonatal birth weight following IVF. The analysis included 183 women aged 18 to 42 years with a singleton live birth following IVF with fresh embryos derived from autologous oocytes at the Duke Fertility Center from January 1, 2003, to December 31, 2014.

In total, 15 infants (8.2%) met the criteria for LBW. Among LBW infants, the mean birth weight was 2204.2 g and the mean gestational age was 35.5 weeks. By comparison, the mean birth weight was 3447.8 g and the mean gestational age was 39.3 weeks among non-LBW infants.

Mean peak estradiol levels were 2847.6 pg/mL among women with LBW infants and 2395.0 pg/mL among women with non-LBW infants. After adjusting for maternal age, body mass index (BMI), gestational age, infant sex, trigger type, diminished ovarian reserve, and parity, the dose-response relationship between birth weight and peak estradiol level remained intact. The mean birth weight decreased by 29.92 g for every 500-pg/mL increase in peak estradiol (P = .04; Table).

Table. Multivariate Analysis of Risk Factors for LBW

Variable Absolute Birth Weight LBW
Estimated Effect, g (95% CI)

P Value

Odds Ratio (95% CI)

P Value

Peak estradiol, every 500-pg/mL change –29.92 (–59.02 to –0.82) .04 1.25 (0.87 to 1.80) .22
Patient age 5.35 (–11.01 to 21.51) .525 1.12 (0.90 to 1.40) .31
Body mass index 14.06 (3.93 to 24.19) .01 1.00 (0.87 to 1.14) .96
Male infant sex 149.61 (28.29 to 270.93) .02 0.92 (0.20 to 4.31) .92
Gestational age 220.67 (186.73 to 254.62) < .01 0.35 (0.22 to 0.57) < .01
Trigger hCG –203.07 (–378.12 to –28.03) .02 1.17 (0.09 to 14.61) .90
Diminished ovarian reserve 201.69 (47.60 to 355.78) .01
Parity 2.13 (0.99 to 4.59) .05

CI = confidence interval, hCG = human chorionic gonadotropin, LBW = low birth weight.

Additional risk factors that correlated with absolute birth weight included BMI, male infant sex, gestational age, trigger type, and diminished ovarian reserve (see Table). However, gestational age was the only risk factor that significantly predicted LBW.

The secondary objective of the study was to identify an optimal peak estradiol threshold for predicting LBW. The research team determined that a peak estradiol level of 3991 pg/mL predicted LBW with 90% specificity and 33% sensitivity and had negative and positive predictive values of 94% and 23%, respectively.

A model that included dichotomized peak estradiol (higher or lower than 3991 pg/mL), in combination with maternal age, BMI, infant sex, gestational age, and parity, resulted in an area under the curve of 89% for predicting LBW. These findings provide the first evidence-based framework for predicting LBW among neonates conceived with IVF.

Source: Jackson-Bey T, Keyhan S, Li Y, Truong T, Eaton JL. Peak serum estradiol and neonatal birth weight following in vitro fertilization. Presented at: American Society for Reproductive Medicine 2016 Scientific Congress & Expo; October 15-19, 2016; Salt Lake City, UT. Abstract P-198.