Freeze-All vs Fresh Embryo Transfer: What 4,064 IVF Patients Show

A meta-analysis of seven randomized trials involving 4,064 women found no statistically significant difference in cumulative live-birth rates between freeze-all and fresh embryo-transfer strategies. Freeze-all was associated with a 54% lower risk of ovarian hyperstimulation syndrome (OHSS). The findings support choosing a strategy for the individual patient rather than applying one approach to everyone.

In IVF, some patients increasingly favor frozen embryo transfer exclusively. After eggs are collected, fertilized and cultured, is it better to transfer a fresh embryo immediately, or freeze all embryos and transfer a thawed embryo in a later cycle?

Freeze-all has become widely used in fertility clinics. Hormone levels rise markedly during ovarian stimulation, so allowing a cycle to pass and the body to settle before transferring a frozen embryo may seem more favorable for implantation.

But does freezing first actually produce better results for every IVF patient?

A meta-analysis presented at the 2026 European Society of Human Reproduction and Embryology (ESHRE) meeting and published in Human Reproduction offers a qualified answer. In an unselected IVF population, it found no significant difference in cumulative live-birth rates between the two strategies.

The researchers selected randomized controlled trials that directly compared freeze-all with fresh embryo transfer. Their initial search retrieved 1,556 studies, but seven trials met the eligibility criteria. Together, they included 4,064 women: 2,030 in freeze-all groups and 2,034 in fresh-transfer groups, where transfer took place in the egg-retrieval cycle.

Freezing every embryo did not increase cumulative live birth

The principal outcome was cumulative live birth. This asks more than whether a baby was born after the first transfer. It considers whether a patient ultimately had a live birth using the embryos obtained from one egg retrieval, including a fresh transfer and subsequent frozen transfers where applicable.

Pooling the six trials that reported this outcome showed no statistically significant difference. The risk ratio was 0.98, with a 95% confidence interval of 0.87–1.11 and P=0.70.

In practical terms, the analysis did not establish that choosing freeze-all increased the overall likelihood of having a baby compared with a fresh-transfer strategy. It does not support a general claim that freezing all embryos and transferring later improves the chance of birth.

OHSS risk was lower with freeze-all

Freeze-all nevertheless had an important advantage: a lower risk of OHSS.

OHSS is a complication in which the ovaries respond excessively to stimulation medication, potentially causing ovarian enlargement and fluid accumulation in the abdomen. Severe cases may require hospitalization.

Across the four randomized trials reporting OHSS, the risk ratio for freeze-all versus fresh transfer was 0.46, with a 95% confidence interval of 0.27–0.77. This statistically significant finding corresponds to an approximately 54% relative reduction in risk.

The timing of transfer helps explain why. Fresh transfer takes place during the same stimulated cycle as egg retrieval, while the body remains under the influence of the hormones used to grow multiple follicles. Freeze-all postpones transfer to a later cycle.

For women who respond strongly to stimulation and develop many follicles, pregnancy-related human chorionic gonadotropin (hCG) can aggravate OHSS. Delaying transfer can therefore be important for safety in this group.

Does this mean fresh transfer is better?

No. The main finding is not that fresh transfer is superior. It is that routinely using freeze-all in unselected IVF patients did not produce a higher cumulative live-birth rate.

The meta-analysis excluded trials that enrolled only high responders or women with polycystic ovary syndrome (PCOS). Its results should not be interpreted as showing that patients at high risk of OHSS have no reason to freeze their embryos.

Earlier research points in a similar direction. In a multicenter randomized trial involving 460 women with regular menstrual cycles, live-birth rates were 27.4% with freeze-all and 28.7% with fresh transfer, without a significant difference. Time to pregnancy was longer in the freeze-all group.

The useful question is therefore whether a particular patient should have a fresh transfer now or freeze all embryos and wait for a later cycle. The choice depends on ovarian response, OHSS risk, the condition of the endometrium, hormone levels, the number and development of embryos, and the broader treatment plan.

The authors concluded that routine freeze-all in unselected IVF populations may not improve cumulative live birth, supporting a more individualized approach to embryo transfer.

The evidence is a conference abstract, not a full research paper

This meta-analysis was published as a 2026 ESHRE conference abstract, so the available research detail is more limited than in a full paper. Only seven randomized trials were included, and follow-up periods for defining cumulative live birth differed between studies.

The findings cannot, by themselves, determine the best transfer strategy for an individual patient.

This article is based on the meta-analysis by researchers at Harapan Kita Mother and Child Hospital and IPB University in Indonesia, published on July 8, 2026, in the volume 41 supplement of Human Reproduction. Its conference-abstract status should be considered when interpreting the results.

Research references

The impact of freeze-all versus fresh embryo transfer on cumulative live birth rates and ovarian hyperstimulation syndrome in unselected IVF patients: A meta-analysis of RCTs. Human Reproduction 41, Supplement 1, July 8, 2026. ESHRE conference abstract L26/P-139. DOI: 10.1093/humrep/deag083.475.

Freeze-all versus fresh blastocyst transfer strategy during IVF in women with regular menstrual cycles: multicentre randomised controlled trial. BMJ, 2020.