Association between serum estradiol level decline in the days preceding ovulatory trigger and assisted reproductive technology outcomes.

OBJECTIVE

To study whether a decline in serum estradiol level in the days leading up to the ovulatory trigger in gonadotropin-releasing hormone antagonist ovarian stimulation cycles impacts cycle outcomes.

DESIGN

Retrospective cohort study.

SUBJECTS: Patients undergoing autologous retrieval cycles using a gonadotropin-releasing hormone antagonist protocol between 2010 and 2021 at a multicenter infertility practice network in the United States. Patients were excluded if an estradiol level was not measured on the 4 days before the ovulatory trigger or if they had conventional insemination.

EXPOSURE: Patients with a documented decrease of at least 1 pg/mL in serum estradiol on any of the 4 days before ovulatory trigger. The referent group experienced a consistent rise in serum estradiol each day before the trigger.

MAIN OUTCOME MEASURES: The primary outcome was the number of usable blastocysts, defined as the sum of blastocysts used for fresh transfer (if any) and the remaining cryopreserved blastocysts. Secondary outcomes included oocyte maturity, fertilization, ≥1 usable blastocyst, positive human chorionic gonadotropin, biochemical pregnancy, spontaneous abortion, clinical pregnancy, and live birth.

RESULTS

A total of 6,945 cycles met the inclusion criteria. Among this cohort, 5,653 demonstrated a consistent rise in serum estradiol, 1,103 cycles experienced one drop, and 189 cycles experienced 2-4 drops in estradiol during the 4 days preceding the ovulatory trigger. After adjusting for antimüllerian hormone and female age, there was no difference in the average number of usable blastocysts between the referent group (4.9 ± 4.0) and cycles with a single drop in estradiol (5.2 ± 4.8) (mean ratio 1.04, 95% confidence interval [CI] 0.97-1.10) and those with 2-4 drops in estradiol (mean ratio 0.98, 95% CI 0.85-1.14). There were no statistical differences in any of the secondary outcomes, including live birth. Live birth was observed at 41.1% for the consistent rise group, 36.8% for the one-drop group (risk ratio 0.91, 95% CI 0.81-1.02), and 42.1% for the 2-4 drops group (risk ratio 0.98, 95% CI 0.74-1.28).

CONCLUSION

The findings underscore that a decline in serum estradiol in the 4 days before the ovulatory trigger does not compromise oocyte quality. Patients with an estradiol decline preceding oocyte retrieval had comparable outcomes to those with a continued rise in estradiol.