Endogenous progesterone guided timing of natural cycle frozen embryo transfer and reproductive outcomes in a large multicenter cohort in the United States.

OBJECTIVE

To study whether an endogenous progesterone (P4) measurement-based strategy for timing natural cycle frozen embryo transfer (NC-FET), with embryo transfer performed two days after serum P4 reaches ≥5 ng/mL in the luteal phase, is associated with differences in live birth rates compared with true natural, modified natural, and programmed FET protocols.

DESIGN

Retrospective cohort study.

SUBJECTS: Autologous single-blastocyst FET cycles performed between January 2020 and December 2023.

EXPOSURE: Frozen embryo transfer using one of four endometrial preparation protocols: endogenous P4-based NC-FET, true NC-FET, modified NC-FET, or programmed FET.

MAIN OUTCOME MEASURES: Live birth rate per embryo transfer.

RESULTS

A total of 26,668 FET cycles met inclusion criteria, including P4-based NC-FET (n = 74), true NC-FET (n = 999), modified NC-FET (n = 680), and programmed FET (n = 24,915). The NC-FETs were more frequently later-order transfers, and programmed FETs were mostly first-order transfers. Therefore, a cycle order-standardized programmed cohort was used for comparison. Baseline characteristics were comparable across groups. In the P4-based NC-FET group, serum P4 reached ≥5 ng/mL on a median cycle day 16 (IQR 15-18), approximately 4 days after luteinizing hormone surge, with embryo transfer occurring predominantly on cycle days 19-21. Live birth rate was 50% in the P4-NC group and 47% in programmed FET cycles, with no statistically significant difference after adjustment (risk ratio [RR] 1.12, 95% confidence interval [CI] 0.88-1.43). Clinical pregnancy, biochemical pregnancy, and pregnancy loss rates did not differ significantly across protocols. True and modified NC-FET demonstrated higher live birth rates compared with programmed FET; however, no statistically significant differences were noted when compared with P4-NC FET. Analyses restricted to euploid embryo transfers yielded similar findings.

CONCLUSION

Endogenous progesterone-based timing of NC-FET, with embryo transfer performed two days after serum P4 reaches ≥5 ng/mL, was not associated with a statistically significant difference in live birth rates compared with programmed or other established natural cycle protocols. This strategy represents a feasible alternative for NC-FET timing and permits salvage of a natural cycle that might otherwise be cancelled because of missed luteinizing hormone surge detection.