Should frozen embryo transfer happen in the first cycle after egg retrieval, or after waiting another cycle? A Danish trial of 494 women did not establish that immediate transfer was non-inferior in modified natural cycles. Its findings do not show that waiting improves live birth for every patient or protocol.
- 494 women randomized across 7 Danish fertility clinics: Live birth rate was 32.4% for immediate vs. 36.0% for postponed transfer
- While the absolute difference was modest, the trial failed to prove that immediate transfer is non-inferior
- Researchers conclude: “immediate mNC-FET should not be offered as standard treatment”
“Now that egg retrieval is over, can I transfer the embryo during my very next period? Or is it better to let the uterus rest for about a month?”
This is one of the most common questions asked by patients undergoing In Vitro Fertilization (IVF). After undergoing days of ovarian stimulation injections and completing the retrieval, patients with cryopreserved embryos face two main pathways: preparing for a frozen embryo transfer (FET) as soon as the next menstrual period starts, or skipping one cycle before proceeding.
Among fertility patients, a widespread belief holds that “because the body underwent hyperstimulation, resting the uterus and body for a month enhances implantation.” Conversely, many patients eager to shorten the timeline ask whether it is safe to proceed immediately if they feel physically well.
Which approach actually leads to better pregnancy and childbirth outcomes?
A multicenter randomized controlled trial (RCT) presented at the 2026 Annual Meeting of the European Society of Human Reproduction and Embryology (ESHRE) directly addressed this question. The trial could not definitively conclude that “taking a month off is universally superior,” but it also failed to prove that immediate transfer was non-inferior within the prespecified margin.
How did the trial compare immediate and postponed FET?
The research team, encompassing seven fertility centers including Copenhagen University Hospital (Rigshospitalet) in Denmark, recruited 494 women between April 2021 and February 2025. Eligible participants were aged 18 to 40 with regular menstrual cycles who had either not conceived following a fresh embryo transfer or had undergone a “freeze-all” cycle and were preparing for their first FET.
The participants were randomized into two distinct arms:
- Immediate FET: Underwent frozen embryo transfer in the first menstrual cycle immediately following oocyte retrieval.
- Postponed FET: Postponed embryo transfer by at least one full menstrual cycle.
The protocol utilized in this trial was a modified natural cycle frozen embryo transfer (mNC-FET). In this protocol, natural follicular development is monitored via ultrasound until the leading follicle measures ≥17 mm, at which point an ovulation trigger injection (hCG) is administered, followed by blastocyst transfer 6 to 7 days later. This approach differs physiologically from artificial/programmed HRT cycles, where estrogen and progesterone are administered for endometrial preparation.
What did the live birth results show?
The primary clinical endpoint evaluated by the investigators was the live birth rate (LBR)—measuring delivered babies rather than biochemical or clinical pregnancy markers alone.
In the intention-to-treat (ITT) analysis of all 494 randomized women:
- Immediate FET Group: Live birth rate was 32.4%
- Postponed FET Group: Live birth rate was 36.0%
While the postponed group had a 3.6 percentage point numerical advantage, the difference did not reach statistical significance (P = 0.393).
Looking solely at the raw numbers (32% vs. 36%), one might assume the difference is negligible and conclude that transferring immediately is completely fine.
However, the statistical architecture of this trial prevents that interpretation.
This study was designed not as a conventional “superiority trial” testing whether waiting is better, but as a non-inferiority trial designed to prove that immediate transfer is not meaningfully worse than waiting. The researchers prespecified an absolute margin of 10 percentage points: the lower one-sided 95% confidence limit for the difference in live birth rates (immediate minus postponed) had to remain above −10 percentage points.
The statistical analysis showed that the lower bound crossed this prespecified margin. In other words, the trial could not rule out the possibility that immediate transfer results in an unacceptably lower birth rate.
Consequently, the study failed to statistically confirm that immediate FET is non-inferior to postponed FET.
What changed in the per-protocol analysis?
In the per-protocol (PP) analysis of 472 women, the gap widened:
- Live Birth Rate: 32.7% (Immediate) vs. 39.2% (Postponed) — an author-reported 6.6 percentage point difference (still failing to meet non-inferiority criteria).
- Pregnancy Rate: 44.9% (Immediate) vs. 57.3% (Postponed) — a statistically significant 12.4 percentage point difference (P = 0.007).
When evaluating only the cycles where an embryo transfer actually took place, the pregnancy rate remained significantly higher in the postponed arm (60.8% vs. 50.7%).
Conversely, other secondary reproductive outcomes, including miscarriage rates, showed no significant disparities. Cycle cancellation rates were 10.1% in the immediate group and 6.2% in the postponed group, which did not reach statistical significance.
Why might clinics postpone transfer?
Why has delaying FET by one cycle become common practice?
During controlled ovarian hyperstimulation, high doses of gonadotropins stimulate multiple follicles simultaneously, creating a supraphysiological hormonal environment unlike that of a natural cycle. Clinicians have long hypothesized that residual endocrine effects could persist into the immediately following cycle, potentially impairing endometrial receptivity or disrupting normal follicular recruitment.
Additionally, multiple resolving corpora lutea frequently persist in the ovaries immediately post-retrieval, making it difficult to distinguish an emerging dominant follicle on ultrasound. For these physiological and diagnostic reasons, many fertility clinics routinely postpone FET.
However, waiting has its own downsides. Deferring transfer extends the overall time to pregnancy (TTP), and for couples who have already navigated prolonged infertility, waiting another month can add emotional strain.
In a prior sub-study of this trial, women in the immediate arm reported more physical symptoms. More women in the postponed arm reported emotional effects of waiting, but that difference was not significant after adjustment for multiple testing. Total infertility-related stress and severe-depression symptoms did not differ significantly between groups.
How do these results compare with a programmed-cycle trial?
A previous randomized trial reported results favoring immediate FET in a different treatment setting.
A randomized trial of 724 women published in 2021 reported per-protocol live birth rates of 47.2% (163/345) with immediate FET and 37.7% (129/342) with delayed FET. Live birth was a secondary outcome; the trial was powered for ongoing pregnancy.
One difference between the trials is the endometrial preparation protocol:
- The 2021 study evaluated artificial/programmed cycles (HRT-FET), where exogenous estrogen and progesterone prepare the endometrium.
- The 2026 Danish trial evaluated modified natural cycles (mNC-FET), which rely on endogenous follicular growth and ovulation.
These differing findings should be interpreted in the context of each trial’s protocol and population. Comparing separate trials does not establish that the preparation protocol caused the difference or prove a timing-by-protocol interaction.
What can patients conclude about transfer timing?
Interpreting these findings as definitive proof that “waiting a month always increases birth rates” would be inaccurate.
The difference in live birth rates between the two groups did not achieve clear statistical superiority. The primary finding of this trial is that immediate transfer failed to demonstrate non-inferiority.
The investigators highlighted that a single trial of roughly 500 women cannot establish definitive clinical policy and that larger prospective studies are warranted. Furthermore, because these findings are specific to modified natural cycles, they cannot be directly extrapolated to programmed HRT cycles or true unmedicated natural cycles.
For now, evidence remains insufficient to assert that every patient must wait a month after retrieval, but neither is it justified to claim that immediate transfer offers identical success rates in natural cycle transfers.
The study authors concluded that in modified natural cycles, routinely offering immediate FET in the first cycle post-retrieval as standard care cannot yet be recommended.
Ultimately, the takeaway for patients is not simply whether to wait or proceed. Transfer timing should be decided collaboratively with the clinical team, taking into account ovarian recovery, cycle regularity, endometrial readiness, and the specific protocol selected for embryo transfer.
For the separate question of how to prepare the uterine lining, read our report on natural versus hormone-replacement cycles for frozen embryo transfer. That comparison concerns preparation protocols, rather than whether to wait after retrieval.
※ This article was written based on the multicenter randomized controlled trial “Immediate versus postponed frozen embryo transfer in a modified natural cycle – A multicentre randomised controlled trial” conducted by researchers from Copenhagen University Hospital (Rigshospitalet) and six other fertility centers, published in Human Reproduction, Volume 41, Supplement 1, on July 8, 2026. The study enrolled 494 women aged 18–40 (Presentation: L26/P-902, Abstract: deag083.1226, DOI: 10.1093/humrep/deag083.1226). It does not provide individualized clinical diagnoses or treatment recommendations, and decisions regarding transfer timing and endometrial preparation should always be made in consultation with an attending fertility specialist.
Additional primary sources: Danish trial protocol (BMJ Open, 2021); Li et al., programmed-cycle timing trial (Human Reproduction, 2021); psychosocial substudy (Human Reproduction, 2025).
※ The images associated with this article were generated using generative AI (ChatGPT, OpenAI) as illustrative visual references and do not depict real individuals.
※ Copyright © 2026 THE FERTILITY NOTES. All rights reserved. Unauthorized reproduction, duplication, distribution, scraping, AI training, or data utilization of all content, including articles, photos, and images, is strictly prohibited and may result in legal liability.