Introduction

Currently many reproductive centres perform frozen embryo transfer (FET) of blastocysts developed on D6, however, it is unclear on which day of progesterone administration these blastocysts should be transferred in order to achieve the highest live birth rate, as only a limited number of studies explored this issue. Extending embryo development to day 6 (D6) allows the use of slower developing embryos and offers advantages in cases requiring pre-implantation genetic testing (PGT) or for working schedule purposes1,2.

Previous studies, observed that D6 blastocysts transferred in a FET cycle had a significantly higher chance of implantation compared to morphologically equivalent blastocysts transferred in a fresh transfer cycle3,4,5,6. One study suggested that even elective D6 transfer in fresh cycles resulted in lower implantation rates than D5 transfer5, though another study has reported comparable outcomes when the embryo had reached the blastocyst stage already on Day 57. As a result, current practice favors cryopreservation of D6 embryos and subsequent transfer in FET cycles.

The most suitable timing of D6 vitrified blastocyst transfers in FET cycles has yet to be determined.

While some reproductive centres, transfer D6 blastocysts on the 6th day of progesterone administration, others transfer both D5 and D6 blastocysts on the 5th day of progesterone8,9,10.

Until now, only a few studies reported better reproductive outcomes if vitrified warmed D6 blastocysts were transferred on the 5th day of progesterone compared to the 6th day of progesterone1,10.

The aim of our study is to examine whether D6 blastocysts should be transferred on the 5th or 6th day of progesterone administration during a hormonal replacement therapy (HRT) FET cycle.

Methods

This cohort study included a total of 746 vitrified D6 single blastocyst HRT-FET cycles performed between 2013 and 2022 at Fertility Clinic Skive Regional Hospital, Denmark. A total of 576 blastocysts were transferred on the 6th day of progesterone (2013–2021), and a total of 170 blastocysts were transferred on the 5th day of progesterone (20.09.2021–31.03.2023).

All methods were carried out in accordance with relevant guidelines and regulations.

The data employed in this analysis were derived from our clinical database. Given the retrospective nature of the study, ethical approval was waived by the Research Ethics Committees for Central Denmark Region.

Inclusion criteria were:

  • Hormone replacement therapy (HRT) frozen embryo transfer (FET) cycles,

  • Transfer of a Day 6 blastocyst,

  • Day 5 or Day 6 vitrified-warmed blastocyst transfer,

  • Available data on progesterone administration day and pregnancy outcome.

Women with PCOS or recurrent miscarriage were not excluded, as these conditions are representative of the clinical IVF population.

Exclusion criteria were:

  • Cycles using Day 5 or Day 7 blastocysts,

  • Transfers in natural or modified natural cycles,

  • Missing or incomplete data regarding embryo stage or progesterone exposure,

  • Donor oocytes.

Ovarian stimulation and ovulation trigger

Ovarian stimulation (OS) either involved gonadotropin-releasing hormone (GnRH) antagonist or GnRH agonist and follicle stimulating hormone (FSH), with human chorionic gonadotrophin (hCG) or GnRH agonist used as a trigger.

Vitrification and warming procedure

Blastocysts were cryopreserved using the Cryotop® vitrification protocol, as previously described by Kuwayama et al. with slight modifications. In brief blastocysts were equilibrated in a two-step vitrification solution containing ethylene glycol, dimethyl sulphoxide and sucrose at room temperature. Blastocysts were loaded on the Cryotop® strip in a volume of < 0.1 µl. vitrification solution and plunged immediately in liquid nitrogen. During the warming process, the Cryotop® strip was rapidly immersed from − 196 °C into a thawing solution at 37 °C containing sucrose to mitigate osmotic shock. The blastocysts were sequentially incubated in diluent and washing solution before transferring to culture medium and incubated to complete recovery. Blastocyst survival was evaluated after warming based on morphology and the ability of the blastocoel to re-expand before embryo transfer. Blastocyst transfer was performed 2–4 h after warming. The blastocysts transferred during the study were unscreened and did not undergo genetic testing.

Hormone replacement therapy

Prior to FET, all the patients underwent a standard HRT protocol, including oral oestradiol (6 mg/24hours), followed by vaginal micronized progesterone, 400 mg twice daily. Single day 6 blastocyst transfer was performed on either the 5th or 6th day of progesterone administration.

Outcome parameters

The main outcome parameter was live birth rate (LBR). Secondary outcomes were: biochemical pregnancy rate (PR), clinical pregnancy rate (CPR) and early pregnancy loss rate (EPLR).

LBR was defined as the ratio between number of patients with live-born babies after gestational week 22 and number of embryo transfers performed.

PR was defined as the number of pregnancies with a serum β-hCG level > 10 IU/L, measured 9–11 days after embryo transfer.

CPR was defined as the number of clinical pregnancies observed at vaginal ultrasound by visualization of a gestational sac and fetal heartbeat in the 7th gestational week.

EPLR was defined as an intrauterine pregnancy loss before 13 weeks of gestation, including both biochemical pregnancy losses and clinical losses.

Blastocyst scoring

Morphological grading of blastocysts was performed, according to the Gardner scoring system11. In brief based on expansion of the blastocoel cavity (1–6), development of inner cell mass (ICM) (A-C) and trophectoderm (TE) (A-C). Blastocysts were grouped as top quality ≥ 4AA/AB/BA, good quality 3AA/AB/BA, ≥3BB and poor quality 1 or 2, ≥ 3 with ICM C or TE C included.

Statistical analyses

SPSS software package 27.0 (IBM Corp., Armonk, NY, USA) was used for all statistical analyses. Categorical variables were compared using the Pearson’s chi-square test and for continuous variables, comparisons were made using the independent t-test, while the Mann-Whitney U test was utilized for non-normally distributed variables. Two-sided P-values were reported, with a significance threshold set at P < 0.05.

To account for potential confounding factors and repeated FET-cycle for some patients, a logistic regression model using Generalized Estimating Equations (GEE), since some patients underwent more than one embryo transfer. Models for LBR, PR, CPR, and EPLR were adjusted for maternal age, BMI, endometrial thickness (≥ 7 mm vs. <7 mm), and blastocyst quality (top vs. lower-quality), with adjusted odds ratios (aORs) and 95% confidence intervals (CIs) reported.

A p-value of < 0.05 was considered statistically significant.

No power calculation was performed for this study since it is a retrospective cohort analysis.

Results

Patient characteristics

In this cohort of a total of 746 HRT-FET cycles, D6 blastocysts transferred on either the 5th or 6th day of progesterone administration the mean age of the patients at embryo transfer was 33.6 ± 4.4 years vs. 32.7 ± 4.8 years, respectively, p = 0.06 (Table 1).

Table 1 Main patient characteristics.
Full size table

The mean BMI of patients on the 5th and 6th day of progesterone administration was 24.7 ± 3.5 vs. 24.5 ± 4.0, respectively (p = 0.25, Table 1). The endometrium was significantly thicker in the patients who had D6 blastocysts transferred on day 5, compared to those in the day 6 group (9.3 ± 2.0 vs. 8.8 ± 2.0, p = 0.03 (Table 1). A similar result was observed when endometrium thickness was analyzed categorically, using 7 mm as the cut-off (χ2 = 4.2, p = 0.04, Table 1). There was no significant difference between the groups regarding primary diagnosis, fertilization method (IVF vs. ICSI), or the number of blastocysts (p = 0.30, p = 0.10, p = 0.11, respectively, Table 1).

Blastocyst quality

There was no significant difference between the quality of blastocysts within the two groups:

Top quality 63.5% (108/170) vs. 62.8% (362/576), good quality 35.3% (60/170) vs. 36.6% (211/576) and poor quality 1.2% (2/170) vs. 0.5% (3/576), p = 0.63 for D6 transferred on day 5 versus day 6 transferred on day 6 of progesterone, respectively (Table 2).

Table 2 Blastocyst quality.
Full size table

Reproductive outcome

The LBR in the group of patients transferred on the 5th day of progesterone was 23.5% (40/170), while in the group of patients transferred on the 6th day of progesterone administration it was 25.7% (148/576) (p = 0.57 (Table 3)).

Table 3 Main reproductive outcomes.
Full size table

The PR in the group of patients transferred on the 5th day of progesterone was 38.8% (66/170), while in the group of patients transferred on the 6th day of progesterone administration it was 41.2% (237/576), (p = 0.60 (Table 3)).

The CPR in the group of patients transferred on the 5th day of progesterone was 25.9% (44/170), while in the group of patients transferred on the 6th day of progesterone administration it was 29.5% (170/576), (p = 0.39 (Table 3)).

The EPLR in the group of patients transferred on the 5th day of progesterone was 33.3% (22/66), while in the group of patients transferred on the 6th day of progesterone administration it was 28.3% (67/237), (p = 0.45 (Table 3)).

Reproductive outcomes – Multivariable GEE analysis

Multivariable logistic regression using Generalized Estimating Equations (GEE) showed that transfer on Day 5 versus Day 6 of progesterone was not a statistically significant predictor of any reproductive outcome. The adjusted odds ratio (aOR) for LBR was 1.33 (95% CI: 0.83–2.14, p = 0.230); for CPR, 1.39 (95% CI: 0.82–2.34, p = 0.158); for PR, 1.30 (95% CI: 0.79–2.12, p = 0.305); and for EPLR, 0.67 (95% CI: 0.34–1.32, p = 0.239) (Table 4).

Table 4 Generalized estimating equations (GEE) analysis for predictors of live birth, clinical pregnancy, positive pregnancy test, and early pregnancy loss (adjusted odds ratios).
Full size table

Among covariates, maternal age was a significant negative predictor of LBR (aOR = 0.95, 95% CI: 0.91–0.99, p = 0.014). Good-quality blastocyst was the strongest positive predictor across most outcomes: LBR (aOR = 2.94, 95% CI: 1.63–5.31, p < 0.001), CPR (aOR = 3.08, 95% CI: 1.67–5.66, p < 0.001), and PR (aOR = 2.34, 95% CI: 1.31–4.19, p = 0.003). For EPLR, blastocyst quality showed a trend toward lower risk (aOR = 0.49, 95% CI: 0.23–1.10, p = 0.065), though not statistically significant. BMI and endometrial thickness were not significant predictors for any outcome (Table 4).

Discussion

This retrospective cohort study evaluated whether the timing of D6 blastocyst transfer - on the 5th versus 6th day of progesterone administration in HRT-FET cycles, affects reproductive outcomes.

We found no significant differences in live birth rate (LBR), clinical pregnancy rate (CPR), biochemical pregnancy rate (PR), or early pregnancy loss rate (EPLR) between the two groups. These findings remained consistent in multivariable models, adjusted for potential confounders, including maternal age, BMI, endometrial thickness and blastocyst quality. Blastocyst quality emerged as the most significant predictor of LBR, CPR and PR, while maternal age was a significant negative predictor for LBR. Neither BMI nor endometrial thickness significantly influenced the outcomes.

Our findings differ from two previous retrospective cohort studies. Bilgory et al. reported higher CPR and LBR when D6 blastocysts were transfered on the 5th day of progesterone exposure compared to the 6th day (21 vs. 21 transfers)1. Similarly, Xu et al. found significantly higher CPR and implantation rate (IR) with day 5 vs. day 6 transfers (103 vs. 415 cycles), while the early miscarriage rate and the multiple pregnancy rates were similar10. These discrepancies may reflect differences in study design, inclusion criteria and embryo quality. Additionally, small sample sizes and retrospective nature of those studies may contribute to different conclusions.

In our study, the only indication for D6 blastocyst vitrification was late blastulation. This minimized heterogeneity, as embryos were not vitrified due to poor day 3 quality, preimplantation genetic testing (PGT), or workload issues, as it was case in Bilgory et al.1. Xu et al. also used late blastulation as the only indication for D6 vitrification10,

Blastocyst quality in our cohort was comparable between groups and grading was performed using the Gardner system11, with top-quality blastocysts comprised 63.5% in the Day 5 group and 62.9% in the Day 6 group; good-quality blastocysts were 35.3% versus 36.6%, and poor-quality blastocysts accounted for less than were 1. 5% in both groups. These proportions were slightly higher than in previous studies. For example, Xu et al. classified 55–58% as “good” (> 3BB) and 42–45% as “qualified” for cryopreservation (> 3CC)10. Bilgory et al. reported only 30–38% as “excellent”1. A possible explanation is, that transfer timing becomes more critical when embryo quality is compromised. In such cases, optimizing endometrial synchrony may partially compensate for lower developmental potential. In contrast, when embryo quality is consistently high, as in our cohort, the relative impact of transfer timing may be diminished.

This hypothesis aligns with prior literature showing poorer outcomes of D6 compared to D5 blastocysts, including lower CPR and LBR and higher miscarriage rates3,4,5,7,9,12,13,14,15, potentially due to the intrinsic embryonic factor and a loss of embryonic-endometrial synchrony6. Slowly developing blastocysts have also been shown to have a higher aneuploidy rate, impacting reproductive outcomes16. In the meta-analysis by Li et al. euploid D5 blastocysts had a higher IR and LBR compared to euploid D6 embryos14. A possible explanation for this finding could be intracellular factors, resulting in a slower development of D6 embryo as previously suggested14,17. As our study did not involve PGT, we could not assess euploidy status, which remains a limitation.

Although embryo quality is not modifiable, transfer timing remains a modifiable factor in optimizing implantation. Personalizing timing through Endometrial Receptivity Analysis (ERA) has previously shown comparable outcomes for Day 5 and Day 6 transfers when transfer was synchronized with the receptive window18. Although ERA was not used in our cohort, our outcomes suggests that standard protocols may be sufficient when embryo quality is high.

Another strength of our study is its exclusive focus on HRT-FET cycles, avoiding the variability, introducing the natural cycle FET or modified natural cycle (MNC), which were present in earlier studies1,10. A previous meta-analysis found no significant difference in CPR, IR or LBR between natural cycle FET and HRT-FET19, but a significantly higher miscarriage rate was seen after HRT-FET20. In our cohort, endometrial thickness was slightly greater in the Day 5 group, however, all values remained within clinical recommendations and were unlikely to explain the outcome differences alone.

Importantly, no changes occurred in laboratory protocols, culture conditions, vitrification, or warming procedures during the study period. Although the two groups represented different time frames, consistency in laboratory conditions minimizes the risk of technical bias. The decision to make the protocol change from transfer on the 5th day instead of the 6th day was based on the previous studies1,10, but our results suggest that this change does not affect reproductive outcomes when embryo quality is high.

As this was a real-world cohort including patients with common infertility-related conditions such as PCOS and recurrent miscarriage, some heterogeneity is expected. Although these factors were not adjusted for individually, their inclusion reflects routine clinical practice, enhancing the generalizability of our findings.

The main limitations of our study include its retrospective design and the limited number of transfers, especially on the 5th day of progesterone. Nevertheless, this represents the largest single-center cohort to date addressing this question, and we used robust statistical methods to account for repeated measures and confounding factors.

Conclusion

In this retrospective cohort, D6 blastocyst transfer on either the 5th or 6th day of progesterone exposure in HRT-FET cycles resulted in similar reproductive outcomes. Our findings suggest that transfer timing may be less critical when embryo quality is high. These results contrast with earlier studies and underscore the need for large, prospective randomized controlled trials to determine the optimal timing for D6 blastocyst transfer in HRT-FET cycles.