Is It Time to Move Toward Freeze-All Strategy? - A Retrospective Study Comparing Live Birth Rates between Fresh and First Frozen Blastocyst Transfer.

BACKGROUND: Cryopreservation of all embryos followed by transfer in subsequent cycles has emerged as an effective alternative to fresh embryo transfer (ET) in order to overcome the negative effect of superovulation on endometrial receptivity. AIMS: The aim of this study was to compare the reproductive outcomes between fresh ET and first frozen ET (FET) from "freeze-all" group of embryos.
SETTING: This study was conducted at a private in vitro fertilization center.
DESIGN: This was a retrospective study. PATIENTS AND METHODS: A total of 503 patients fulfilling the inclusion criteria between 2012 and 2017 were included. Of 503, 386 patients underwent fresh ET and 117 patients underwent FET following cryopreservation of all embryos. The results of only first FET were considered to eliminate the confounding factor of poor-quality embryos in subsequent transfer.
RESULTS: FET resulted in statistically significant higher live birth rate (44.44% vs. 33.41%), implantation rate (45.08% vs. 30.22%), and clinical pregnancy rate (57.26% vs. 38.6%) compared to fresh ETs. No difference was observed in the abortion rate between the two groups.
CONCLUSION: Reproductive outcomes were significantly better in the freeze-all group compared to fresh ET suggesting that the altered hormone levels during controlled ovarian stimulation could mediate an asynchrony between the endometrium and the transferred embryos, leading to implantation failure. Copyright:

INTRODUCTION

Implantation is the most critical step which determines success in Assisted reproductive technology (ART). Good-quality embryo and receptive endometrium are the two most important factors influencing implantation. Fresh embryo transfer (ET) has been the standard practice of care for many years. However, the negative effect of supraphysiological hormonal levels on the endometrium due to controlled ovarian stimulation (COS) in fresh cycles has always been an area of concern. High levels of estradiol (E2) and progesterone (P4) are known to cause asynchrony between endometrium and embryo, thereby affecting the reproductive outcomes.[1] With the use of better cryopreservation techniques, universal freezing of all embryos followed by transfer in subsequent cycles is now emerging as an effective alternative to improve ART results. The potential advantage of this method is that it provides a more physiologic environment in which ET can occur; this could lead to better pregnancy rates and is also known to decrease maternal and perinatal morbidity.[2] The aim of this study was to compare the reproductive outcomes (live birth rate [LBR]) between patients undergoing fresh ET and first frozen ET (FET).

PATIENTS AND METHODS

In a retrospective study conducted at a private in vitro fertilization (IVF) center, data were collected from the medical records maintained at the center. Any additional information was collected from the patients via telephonic conversation. All patients fulfilling the inclusion criteria between January 2012 and November 2017 were included in the study.

Patient selection

Fresh embryo transfer

Patients with following the inclusion criteria were selected for the fresh ET group – age between 25 and 35 years, normal ovarian reserve (day 2 follicle-stimulating hormone (FSH) <10 IU/L, anti-Mullerian hormone (AMH) >1.0 ng/ml, antral follicle count (AFC) between 5 and 15, good-quality blastocyst[3] available for fresh transfer (grade >3 with at least one A component, i.e. AA, AB, or BA), and optimal endometrium on the day of transfer (thickness of more than 7mm, triple-line pattern). Exclusion criteria were advanced maternal age >35 years, body mass index (BMI) >30 kg/m2, history of recurrent implantation failure, occult ovarian failure (AMH <1 ng/ml or AFC <5), severe male factor (oligospermia <1 million/ml), and any uterine anomaly affecting implantation (polyp, fibroid, and Asherman syndrome).

Frozen embryo transfer

Freeze-all strategy is adopted only for specific indications at our center. The indications are risk of threatened ovarian hyperstimulation syndrome (OHSS), number of optimal follicles >15 and E2 >3500 pg/ml on the day of trigger (both criteria need to be fulfilled to freeze-all embryos), P4 >1.5 ng/ml on the day of trigger, on patient's request due to personal reasons, unexpected and unrelated medical illness after pickup, unforeseen difficult ET, and suboptimal endometrium in stimulated cycles (thickness <7 mm, echogenic, fluid in cavity). Only patients who had at least one good-quality blastocyst cryopreserved and transferred in subsequent cycle (first FET) were included. The exclusion criteria were the same as for fresh ET patients.

Stimulation protocols

In all patients, ovarian stimulation was started with a flexible starting dose of recombinant/highly purified FSH or highly purified human menopausal gonadotropin (HMG) ranging from 150 IU to 300 IU, depending on age, BMI, history of previous cycle response, and results of ovarian reserve tests. The FSH or HMG dose was then adjusted according to follicular growth monitored by transvaginal ultrasound every 2–3 days. After at least one follicle reaches 14 mm diameter, antagonist (cetrorelix – 0.25 mg) was added along with FSH or HMG. When at least two follicles reached 18 mm in diameter, ovulation was triggered using injection human chorionic gonadotropin (hCG) in patients planned for fresh transfer and with triptorelin (Decapeptyl) in patients where decision to freeze-all embryos was taken. E2 and P4 levels were done for all patients on the day of trigger. Oocyte retrieval was performed under transvaginal ultrasound guidance 34–36 h after administration of triggering agent. All embryos were cultured till day 5 under standard laboratory environment.

For patients undergoing fresh transfer, one or two good-quality blastocysts were transferred on day 5. Standard luteal phase support was given with vaginal micronized progesterone in a dose of 400 mg twice daily. In the freeze-all group, all blastocysts were cryopreserved by means of vitrification. On day 2 of next menstrual cycle, standard hormone replacement therapy was started with oral estradiol valerate in a dose of 8 mg/day. After 10–12 days of E2 replacement, an ultrasound scan was done for endometrium. If the endometrium was > 7 mm, triple-line, progesterone replacement was started with vaginal micronized progesterone in the dose of 400 mg twice daily. After 5 days of starting progesterone, one or two best-quality available blastocyst (s) were thawed and transferred. We only included the results of first FET in our study in order to eliminate the confounding factor of poor-quality embryos in subsequent transfer.

In both the groups, serum beta-hCG was measured 11 days postblastocyst transfer. If beta-hCG was positive (>10 IU/ml), the patient was called for scan a week later to look for gestational sac. Once intrauterine gestational sac was confirmed, repeat scan was done after 1 week to confirm fetal viability. In the fresh ET group, progesterone was continued till 10 weeks of gestation, while in patients undergoing FET, both estradiol valerate and progesterone were continued till 10 weeks of gestation. In both the groups, clinical pregnancies were then followed till the final outcome, i.e. delivery of at least one live-born baby.

Outcome measures

The primary outcome was LBR defined as rate of deliveries that resulted in at least one live-born baby per transfer. The secondary outcomes measured were implantation rate (IR), clinical pregnancy rate (CPR), and abortion rate (AR). IR was defined as the ratio of number of intrauterine gestational sacs detected on ultrasound to the number of transferred embryos. CPR was defined as pregnancy confirmed by ultrasound visualization of the gestational sac with fetal cardiac activity between the 6th and 7th weeks of gestation. AR was defined as miscarriage occurring before 22 weeks of gestation after confirmation of intrauterine gestational sac on early ultrasound.

Statistical analysis

Descriptive analysis was carried out by mean and standard deviation for quantitative variables and frequency and proportion for categorical variables. Outcome measures and the associated clinical variables were analyzed using the Chi-square and Mann–Whitney statistical tests as appropriate. SPSS software (version 22, SPSS Inc., Chicago, IL, USA) was used for statistical analysis. P < 0.05 was considered to be statistically significant.

RESULTS

A total of 503 patients fulfilled the inclusion criteria during the study period of which 386 patients underwent fresh ET and 117 patients had all their embryos cryopreserved followed by FET in subsequent cycle. The causes of infertility were comparable in both the groups [Table 1]. Among baseline characteristics, all parameters were comparable except the serum value of E2 and total number of oocytes retrieved [Table 2]. Both were found to be higher in the freeze-all group. However, there was no difference in the number of mature oocytes, total number of cleaved embryos, and blastocysts formed. Endometrial thickness on the day of pickup was also comparable in both the groups.

Table 1

Causes of infertility in both the groups

Infertility diagnosis	ET cycles		P
	Fresh ET, n (%)	Frozen ET, n (%)
Anovulatory	52 (13.4)	17 (14.5)	0.696
Endometriosis	10 (2.6)	4 (3.4)	0.633
Idiopathic	77 (19.9)	27 (23.1)	0.464
Male factor	81 (21)	24 (20.5)	0.919
Multiple (female)	64 (16.5)	22 (18.8)	0.364
Multiple (male and female)	59 (15.3)	12 (10.3)	0.171
Others	4 (1)	1 (0.9)	0.863
Tubal	39 (10.1)	10 (8.5)	0.287
Total	386	117

ET=Embryo transfer

Table 2

Comparison of baseline characteristics in both the groups

Baseline characteristics	Mean±SD		P
	Fresh ET	FET
Age of female (years)	30.98±3.86	30.57±3.58	0.311
Total number of oocytes retrieved	13.75±5.84	15.82±6.9	0.001
Total number of mature oocytes	11.76±4.77	12.24±4.8	0.065
Number of fertilized embryos	10.13±4.65	10.74±5.75	0.244
Number of cleaved embryos	9.7±4.53	10.28±5.57	0.250
Number of blastocysts formed	5.57±3.65	5.02±3.80	0.384
Serum E2 levels on trigger day (pg/ml)	2140.28±695.45	3773.3±403.34	0.001
Serum P4 levels on trigger day (ng/ml)	0.79±0.32	1.27±0.86	<0.001
Endometrial thickness on the day of transfer (mm)	9.21±1.35	9.41±1.43	0.473
Number of embryos transferred	622	173
Single blastocyst	150	61	0.01
Double blastocyst	236	56
Average number of embryos transferred per patient	1.61	1.47	0.01

ET=Embryo transfer, FET=Frozen ET, SD=Standard deviation, E2=Estradiol, P4=Progesterone

On comparing various reproductive outcome parameters in both the groups, it was found that FET resulted in a statistically significant increase in the LBR (44.44% vs. 33.41%) compared with the rate observed with fresh transfer (P = 0.029). A similar pattern of results was observed on analyzing IR and CPR, which were higher in the women undergoing FET. However, AR was found to be comparable in both the groups [Figure 1]. Another observation worth mentioning is that although the average number of embryos transferred was more in patients undergoing fresh ET, the IR was higher in the FET group [Table 2]. Furthermore, the mean birth weight of babies at delivery was comparable in both the groups [Table 3].

Figure 1

Cluster bar graph for comparison of various outcome parameters between two groups

Table 3

Comparison of various outcome parameters between the two groups (n=503)

Parameter	ET cycles		P
	Fresh ET (n=386), n (%)	FET (n=117), n (%)
CPR	149/386 (38.6)	67/117 (57.26)	<0.001
IR	188/622 (30.22)	78/173 (45.08)	<0.001
Miscarriage rate	45/386 (11.65)	15/117 (12.82)	0.477
LBR	129/386 (33.41)	52/117 (44.44)	0.029
Birth weight at delivery (g), mean±SD	2812.88±458.12	3011.25±354.73	0.159

ET=Embryo transfer, FET=Frozen ET, SD=Standard deviation, LBR=Live birth rate, CPR=Clinical pregnancy rate, IR=Implantation rate

DISCUSSION

Embryo implantation failure is still one of the greatest challenges in the field of ART. In the past few years, much effort has been focused on embryo quality and selection. On the other hand, no effective noninvasive clinical tools are available to evaluate endometrial receptivity and select an appropriate time for ET. Freeze-all strategy is now being adopted as a preferred option by many centers to improve ART results. It has been proposed that with this strategy, the best embryos can be cryopreserved and transferred into a more receptive endometrium, compared with a fresh cycle, as COS may jeopardize endometrial receptivity.[4]

In this study, we compared reproductive outcomes between fresh ET and first FET of top-grade blastocyst (s) in patients with optimal endometrial thickness. Among baseline characteristics, the total numbers of oocytes retrieved, serum E2 and P4 levels were significantly higher in the freeze-all group compared to fresh. This difference could be attributed to the fact that a large percentage of patients in the freeze-all group had polycystic ovaries with risk of threatened OHSS during stimulation. The mean number of oocytes in the fresh ET group is relatively high because only patients with good ovarian reserve were included in this group to minimize the component of bias. Women with low ovarian reserve were excluded from the analysis. Furthermore, even though the cutoff for freeze-all is >15 follicles, we follow minimal dose stimulation protocol for polycystic ovary syndrome (PCOS) patients to get optimum number of oocytes without increasing the risk of OHSS. Hence, the difference in the number of oocytes retrieved (though statistically significant) is relatively less. It was also observed that the difference in E2 levels in both the groups did not correlate with the number of oocytes retrieved. This could be explained by the fact that majority of patients in the freeze-all group had polycystic ovaries which produce more estradiol hormone on the day of trigger due to the presence of many small- and medium-sized follicles. This accounts for the more discrepancy seen in E2 levels between the two groups even though the difference in the total number of oocytes retrieved was comparatively less. However, there was no difference in the number of mature oocytes, cleaved embryos, and blastocysts formed between the two groups. Another factor affecting implantation, i.e. endometrial thickness, was also comparable between the two groups.

In our study, the main outcome measure, i.e. LBR, was significantly higher in patients who underwent FET compared to those who had fresh ET. Similarly, CPR and IR were also found to be significantly higher in the FET group indicating better synchronization between embryo and endometrium. AR was found to be comparable between the two groups.

Various studies have depicted similar results highlighting the detrimental effect of ovarian hyperstimulation on endometrial receptivity and decreased IR in fresh ET compared to FET.[56] In a meta-analysis by Roque et al.[1] to assess whether cryopreservation of all embryos followed by FET results in better outcomes than fresh transfer, it was found that FET resulted in significantly higher ongoing pregnancy rates and CPR. However, none of the randomized controlled trials (RCTs) included in this meta-analysis have commented on the LBR which is the primary goal of any ART treatment.

Chen et al.[7] conducted a multicentric RCT where 1508 infertile patients with PCOS undergoing their first IVF cycle were randomized to either fresh ET or embryo cryopreservation followed by FET of one or two cleavage stage embryos. FET resulted in a higher frequency of live birth than fresh ET (49.3% vs. 42.0%), lower frequency of pregnancy loss, and OHSS but a higher frequency of preeclampsia. There were no significant differences between the two groups in the rates of other pregnancy and neonatal complications. Another study by Shapiro et al.[8] showed that following implantation failure with fresh blastocysts, patients have a significantly greater chance of live birth in a subsequent cycle with freeze-all and FET than with fresh transfer. A meta-analysis of observational studies also showed that FET was associated with a lower risk of perinatal death than fresh ET (relative risk, 0.68; 95% confidence interval, 0.48–0.96).[9]

Studies have found that superovulation leads to lowering of expression of specific integrins which play a crucial role in implantation.[1011] There is evidence suggesting that superovulation may also alter the timing of the receptivity window, i.e. the time duration, for which the endometrium is receptive to implantation of embryo. This shift in the window of endometrial receptivity can have a detrimental effect on implantation; endometrial advancement of more than 3 days was found to be associated with implantation failure.[12] This shift in the window may also have a negative impact on embryo development postimplantation; studies on mouse have shown that embryos implanting beyond the normal receptivity window are more likely to have defective placental formation and fetal growth abnormalities with increased risk of low-birth-weight babies.[13] Recent data also suggest an important role of natural killer (NK) cells with respect to endometrial receptivity.[14] Decidual NK cells secrete multiple factors such as vascular endothelial growth factor, leukemia-inhibiting factor, and cytokines that may help in implantation. Superovulation has been shown to alter the immune environment of the endometrium. A study of endometrial biopsies taken from oocyte donors during stimulated and natural cycles has shown reduced levels of endometrial NK cells following superovulation in comparison to the natural cycles.[15] Thus, superovulation negatively affects the genetic and immunological profile of endometrium making it unfavorable not only for implantation but also for further embryo growth and development. All these detrimental effects of superovulation can be overcome by adopting the policy of freeze-all followed by transfer in subsequent cycle.

However, there are certain issues which mandate more research and clarification before universal freeze-all policy can become a norm in the future. Although vitrification is associated with extremely good thaw rates, there are some studies in mice suggesting that vitrification may result in methylation defect in early embryo.[1617] Moreover, there is some evidence for increased risk of preeclampsia in mothers and large for gestational age babies in FET which require further evaluation.[1819] Furthermore, the optimal hormonal regimen for FET needs to be explicated, particularly when we consider the hormonal milieu as the major culprit affecting reproductive outcomes in fresh ET.

Most of the studies published have included day 2, day 3, and day 5 ETs in their analysis which acts as an important confounding factor in predicting the reproductive outcomes. To the best of our knowledge, this is the first study from India which has compared LBRs between fresh ET and FET of only good-quality blastocyst (s). Furthermore, few RCTs have analyzed LBRs as their primary outcome. The present study is limited by its retrospective nature and policy of using freeze-all strategy for a selected group of patients. A prospective randomized study comparing fresh ET and universal freezing followed by FET will give a better assessment of reproductive outcomes.

CONCLUSION

Freeze-all strategy followed by transfer in subsequent cycles is associated with improvement in reproductive outcomes by overcoming the pernicious effect of superovulation in fresh cycles. However, further well-defined studies are needed using LBRs and perinatal outcomes as the primary end points before freeze-all policy can be accepted as the standard practice of care.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.