Gestational age determination in pregnancies conceived via assisted reproductive technology

To compare gestational age (GA) dating models for pregnancies conceived via assisted reproductive technology (ART), assess whether 14 or 15 days more accurately reflects the median follicular-phase duration in ART pregnancies and evaluate whether the gestational timelines of ART pregnancies are comparable with that of spontaneously conceived pregnancies.

We employed population data from the Medical Birth Registry of Norway (2015-2021), including 163 544 children conceived spontaneously, 2067 conceived via fresh embryo transfer (ET) and 2080 conceived via frozen ET. Among ART pregnancies, we compared two GA dating methods: a population-based ultrasound model (GA_US) and an ART-based formula based on the known oocyte fertilization date. Statistical agreement was evaluated by calculating individual pairwise differences in GA estimates. Bias and precision in the estimated date of delivery (EDD) were assessed by constructing cumulative birth distribution (Kaplan-Meier) curves for fresh and frozen ET pregnancies separately, employing time-to-event analysis to account for nonspontaneous births.

The absolute individual GA differences between ultrasound and ART-based dating, considering a 14-day median follicular-phase duration, were ≤ 1 day for 969/2067 (46.9%) fresh ET pregnancies and for 874/2080 (42.0%) frozen ET pregnancies. An approximate systematic difference of 1 day was observed, which we corrected by employing a 15-day median follicular-phase duration in the ART-based formula (GA_ART,15). There was a median pairwise difference between GA_US and GA_ART,15 of -0.4 (95% CI, -0.5 to -0.3) days in fresh ET pregnancies and -0.2 (95% CI, -0.4 to -0.1) days in frozen ET pregnancies. Both models demonstrated equivalent precision in the EDD. Measured using the ultrasound model, frozen ET pregnancies had a median pregnancy duration of 286.1 (95% CI, 285.5-286.7) days, which was 3.2 (95% CI, 2.4-3.9) days longer than that in fresh ET pregnancies, and 2.6 (95% CI, 2.0-3.2) days longer than that in spontaneously conceived pregnancies.

We found close statistical agreement between ultrasound and ART-based dating models for both GA and EDD. However, 15 (not 14) days of follicular-phase duration should be utilized in the ART-based formula to avoid systematic bias in ART-based GA estimates. Fresh and frozen ET pregnancies had different birth distributions and median pregnancy durations, underscoring that they are not comparable with spontaneously conceived pregnancies and they should be evaluated separately. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.