Impact of prolonged embryo storage on reproductive and neonatal outcomes: a systematic review and meta-analysis

Abstract

Purpose

To analyze the influence of embryo storage on reproductive and neonatal results in patients undergoing in vitro fertilization (IVF) treatment.

Methods

PubMed, Embase, Cochrane Central Register of Controlled Trials, and BioMed Central databases were searched from inception up to June 2024 for studies comparing reproductive and neonatal outcomes in patients undergoing frozen embryo transfer within 12 months from embryo storage versus more than 12 months after embryo storage. Data were pooled by meta-analysis using a random effects model.

Results

A total of 16 studies, involving 233,315 embryos were included. Patients undergoing frozen embryo within 12 months from embryo storage were associated with higher rates of live birth (OR 1.17, 95% CI 1.09–1.26, I² = 78%) biochemical pregnancy (OR 1.26, 95% CI 1.08–1.47, I² = 77.8%) clinical pregnancy (OR 1.24, 95% CI 1.12–1.38, I² = 86.3%) and multiple pregnancy rate (OR 1.26, 95% CI 1.03–1.55, I² = 69%). No significant differences between groups were shown in terms of survival rate (OR 1.52, 95% CI 0.65–3.58, I² = 98.5%), miscarriage (OR 1.08, 95% CI 0.91–1.27, I² = 77%), implantation rate (OR 1.17, 95% CI 0.90–1.52, I² = 91.7%) and ectopic pregnancy (OR 0.98, 95% CI 0.80–1.20, I² = 0%). In addition, prolonged embryo storage was not associated with higher rates of preterm delivery (OR 0.86, 95% CI 0.67–1.09, I² = 8.3%), low weight at birth (OR 1.10, 95% CI 0.88–1.38, I² = 24.3%) or congenital malformations (OR 0.90, 95% CI 0.65–1.25, I² = 0.8%).

Conclusion

Prolonged embryo storage over 12 months is associated with lower rates of live birth, biochemical pregnancy, clinical pregnancy, and multiple pregnancy. However, important covariates such as reasons for delay of transfer, embryo quality, and improper handling of embryos could not be ruled out as causes of this reduction. Given these limitations, these conclusions should be viewed with caution.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10815-024-03283-7.

Introduction

Infertility treatments have undergone a radical change since the introduction of embryo vitrification. Vitrification allows for, among other things, the maximization of a single stimulation cycle, fertility preservation, increased safety in cases of ovarian hyperstimulation risk [1, 2], the implementation of preimplantation genetic diagnosis protocols, and the postponement of embryo transfer in cases of implantation window displacement [3]. Thawed embryo has provided the opportunity to perform selective single embryo transfer (SET), reducing maternal and neonatal morbidity and mortality associated with multiple pregnancies [4, 5]. No differences in reproductive outcomes in terms of cumulative live birth rate (LBR) or ongoing pregnancy rate were found comparing fresh versus frozen embryo transfer [6, 7]. However, frozen embryo transfer has shown higher rates of pregnancy hypertensive disorders, large for gestational age, and macrosomia. These findings were seemingly related to the absence of the corpus luteum, which is typical of hormone replacement endometrial preparation [8, 9]. Nowadays, oocyte, semen, and embryo vitrification are procedures performed in reproduction centers worldwide. Due to social or medical reasons, oocytes and embryos can be kept vitrified for extended periods of time. In this sense, it is pivotal to identify a safe vitrification cut-off period regarding reproductive and neonatal outcomes. Therefore, the aim of this systematic review and meta-analysis was to investigate if extended cryo-storage could impact on pregnancy and neonatal outcomes, thus providing and updated appraisal of evidence on this crucial topic.

Methods

Data sources and search strategy

A systematic review was performed according to the preferred reporting items for systematic reviews and meta-analyses 2009 guidelines [10]. Two reviewers (IZ, AMM) independently identified the relevant studies by an electronic search of the MEDLINE and Embase databases (from inception to June 2024). No language, publication date, study design, or publication status restrictions were imposed. This study is registered with PROSPERO (CRD42024553869).

Study selection

Two reviewers (IZ, AMM) independently assessed trial eligibility on the basis of titles, abstracts, and full-text reports. Discrepancies in study selection were discussed and resolved with another investigator (JMRR). Eligible studies had to satisfy the following prespecified criteria: (1) studies including patients undergoing IVF treatment; (2) investigations comparing transfer of thawed embryo stored for less or more than 12 months; (3) availability of pregnancy outcome data. Exclusion criteria were as follows: (1) studies using fresh embryos or slow freeze embryos; (2) lack of any pregnancy outcome data.

Studies with more than two arms for which a subset of interventions satisfied the inclusion criteria were kept in the analysis after having discarded the arms that did not satisfy the inclusion criteria.

Data extraction and quality assessment

Two investigators (IZ, AMM) independently extracted data (baseline characteristics, definition of outcomes, and number of events) using a standardized data abstraction form. The same investigators independently and systematically assessed the studies’ methodological quality using the risk of bias in non-randomized studies of interventions assessment tool from the Cochrane handbook (ROBINS-I) [11], assessing seven domains of bias for each outcome: (1) confounding, (2) selection of participants, (3) classification of interventions, (4) deviations from intended interventions, (5) missing data, (6) measurement of outcomes, and (7) selection of the reported result. Disagreements were resolved with another investigator (LSG).

Data synthesis and data analysis

Outcome measures

The primary endpoint was live birth. Secondary endpoints included cryo-survival, biochemical pregnancy, implantation rate, ectopic pregnancy, miscarriage, clinical pregnancy, multiple pregnancy, low birth weight, preterm delivery, and neonatal defects.

Statistical analysis

For dichotomous outcomes, the odds ratios (ORs) with 95% confidence intervals (CIs) were calculated from the available data, and trial-specific ORs were combined with the DerSimonian and Laird random effects model with the estimate of heterogeneity being taken from the Mantel–Haenszel model [12]. The presence of heterogeneity among studies was evaluated with the Cochran Q chi-square test with p ≤ 0.1 considered to be of statistical significance, estimating the between-studies variance tau-square, and using the I² test to evaluate inconsistency. The I² statistic is derived from the Q statistic (100% × (Q − df)/Q) and describes the percentage of total variation across studies that is due to heterogeneity; a value of 0% indicates no observed heterogeneity, and larger values show increasing heterogeneity. I² values of 25%, 50%, and 75% have been assigned adjectives of low, moderate, and high heterogeneity, respectively [13]. The presence of publication bias was assessed for the primary endpoint with Harbord tests, and by visual estimation with funnel plots. Pre-specified subgroup analyses for the primary endpoint were performed to assess the influence of embryo storage on treatment effect and by iteratively removing one study at a time to confirm that our findings were not driven by any single study. In order to account for different lengths for embryo transfer across studies, another sensitivity analysis was performed using the Poisson regression model with random intervention effects to calculate inverse-variance weighted averages of study-specific log stratified incidence rate ratios (IRRs). Results were displayed as IRR, which are exponential coefficients of the regression model. In addition, a random effects meta-regression was performed to assess the impact on treatment effect of maternal age, blastocyst-stage transfer, percentage of close system, cryogenic tank employed and good quality embryo transferred. The statistical level of significance was 2-tailed p < 0.05. Analyses were performed using Stata version 13.1 (Stata Corp., College Station, TX).

Results

Search results

Figure 1 displays the preferred reporting items for systematic reviews and meta-analyses flow diagram for study search and selection. Of the 458 citations screened, 432 were excluded because of a preclinical design (no embryo storage was considered), 16 because of lack of clinical data, slow-freezing protocol, different storage classification, or because they were case reports. Through snowballing sampling, we identified four studies that satisfied inclusion criteria. Therefore, a total of 16 studies involving 233,315 embryos were selected and included in this systematic review and meta-analysis.

Fig. 1.

The preferred reporting items for systematic reviews and meta-analyses flow diagram for study search and selection. Flow diagram of the search for studies included in the meta-analysis according to the preferred reporting items for systematic reviews and meta-analyses statement

Study characteristics

The main trial features and patient characteristics of the included studies are reported in Table 1 and Supplementary appendix Table 1. All studies have a retrospective cohort design [14–29] and included patients undergoing embryo transfer in blastocysts stage, seven of them also included cleavage stage [16, 18, 20, 24–27]. One study included embryos with preimplantation testing [14].

Table 1.

Baseline trial and patient characteristics included in the meta-analysis

Reference	Design	Embryos (total no)	Embryo stage	Preimplantation genetic testing	Maximum storage duration (months)	Maternal age (mean, years)	Cryo-storage device	no embryo transferred
Cimadomo (2022) [14]	Retrospective	2688	Blastocyst	Yes	84	38	Cryo-storage liquid nitrogen tanks	SET
Cobo (2024) [15]	Retrospective	58,001	Blastocyst	No	132	39.9	Vapor tanks and liquid nitrogen tanks	SET/DET
Cui (2021) [22]	Retrospective	9806	Blastocyst	No	> 60	34.4	Cryo-storage liquid nitrogen tanks	SET
Lee (2021) [23]	Retrospective	1632	Blastocyst	No	> 25	35.8	Cryo-storage liquid nitrogen tanks	SET/DET
Li J. (2020) [24]	Retrospective	24,698	Blastocyst and cleavage	No	24	31	Cryo-storage liquid nitrogen tanks	SET/DET
Li W. (2017) [26]	Retrospective	1739	Cleavage	No	60	31.2	Cryo-storage liquid nitrogen open device	SET/DET/TET
Li X. (2023) [25]	Retrospective	1037	Blastocyst and cleavage	No	84	32.9	Cryo-storage liquid nitrogen tanks	SET/DET
Liang (2024) [27]	Retrospective	10,167	Blastocyst and cleavage	No	> 12	33.1	Cryo-storage liquid nitrogen tanks	SET/DET/TET
Lin (2021) [28]	Retrospective	7579	Blastocyst	No	126	31.2	Cryo-storage liquid nitrogen open device	SET
Ma (2023) [29]	Retrospective	2938	Blastocyst	No	98	31.2	Cryo-storage liquid nitrogen tanks	SET
Mao (2022) [16]	Retrospective	31,143	Blastocyst and cleavage	No	109	34.3	Cryo-storage liquid nitrogen tanks	SET/DET
Ueno (2018) [17]	Retrospective	8736	Blastocyst	No	97	38.2	Cryo-storage liquid nitrogen open device	SET
Wang (2024) [18]	Retrospective	47,006	Blastocyst and cleavage	No	> 72	33	Cryo-storage liquid nitrogen open device	SET/DET
Wirleitner (2013) [19]	Retrospective	1992	Blastocyst	No	72	36	Cryo-storage liquid nitrogen tanks	SET/DET
Zhang (2021) [20]	Retrospective	17,826	Blastocyst and cleavage	No	> 13	31.4	NA	SET/DET
Zheng (2023) [21]	Retrospective	6327	Blastocyst	No	120	31.7	Cryo-storage liquid nitrogen open device	SET

DET double embryo transfer, NA not available, SET single embryo transfer, TET triple embryo transfer

Clinical outcomes

Patients undergoing frozen embryo transfer within 12 months from embryo storage were associated with higher rates of live birth (OR 1.17, 95% CI 1.09–1.26, I² = 78%) biochemical pregnancy (OR 1.26, 95% CI 1.08–1.47, I² = 77.8%) clinical pregnancy (OR 1.24, 95% CI 1.12–1.38, I² = 86.3%) and multiple pregnancy rate (OR 1.26, 95% CI 1.03–1.55, I² = 69%) (Fig. 2). No significant differences between groups were shown in terms of survival rate (OR 1.52, 95% CI 0.65–3.58, I² = 98.5%), miscarriage (OR 1.08, 95% CI 0.91–1.27, I² = 77%), implantation rate (OR 1.17, 95% CI 0.90–1.52, I² = 91.7%) and ectopic pregnancy (OR 0.98, 95% CI 0.80–1.20, I² = 0%) (Fig. 3). In addition, prolonged embryo storage was not associated with higher rates of preterm delivery (OR 0.86, 95% CI 0.67–1.09, I² = 8.3%), low weight at birth (OR 1.10, 95% CI 0.88–1.38, I² = 24.3%) or congenital malformations (OR 0.90, 95% CI 0.65–1.25, I² = 0.8%) (Fig. 4).

Fig. 2.

Pooled analysis of studies comparing frozen embryo transfer more than 12 months from embryo storage versus within 12 months after embryo storage for a live birth rate, b biochemical pregnancy, c clinical pregnancy, and d multiple pregnancy. Forest plot reporting study-specific and summary odds ratios (ORs) with 95% confidence intervals (CIs)

Fig. 3.

Pooled analysis of studies comparing frozen embryo transfer more than 12 months from embryo storage versus within 12 months after embryo storage for a cryo-survival, b miscarriage, c implantation rate, and d ectopic pregnancy. Forest plot reporting study-specific and summary odds ratios (ORs) with 95% confidence intervals (CIs)

Fig. 4.

Pooled analysis of studies comparing frozen embryo transfer more than 12 months from embryo storage versus within 12 months after embryo storage for a preterm delivery, b low term at birth, and c congenital malformation. Forest plot reporting study-specific and summary odds ratios (ORs) with 95% confidence intervals (CIs)

Risk of bias assessment

Supplementary appendix Table 2 summarizes the results of the risk of bias assessment with the ROBINS-I tool. Four studies were considered at low risk for overall risk of bias [17, 19, 28, 29], and 12 studies presented some concerns [14–16, 18, 21–27, 30].

Sensitivity analyses

At leave-one-out sensitivity analysis, results remained consistent with the main analysis when iterative removing one study at a time (Supplementary appendix Table 3). Similarly, in a sensitivity analysis with the use of estimated IRR to account for different lengths of embryo transfer, findings were unchanged (IRR 0.14, 95% CI 0.09–0.23).

Meta-regression analysis did not show any significant interaction between maternal age (p = 0.105), blastocyst-stage transfer (p = 0.427), percentage of close system (p = 0.685), cryogenic tank employed (p = 0.924), good quality embryo transferred (p = 0.785) and treatment effects.

Funnel-plot distributions of the pre-specified outcomes as well as the Harbord tests indicated the absence of publication bias and small study effect for all the outcomes (Supplementary appendix Figs. 1–11).

Discussion

In this study, we evaluated the impact of embryo storage duration on reproduction and neonatal outcomes in patients undergoing ET. The main findings of our investigation can be summarized as follows:

1. Patients undergoing prolonged embryo storage were associated with a lower rate of live birth, biochemical pregnancy, clinical pregnancy, and multiple pregnancy as compared to patients undergoing frozen embryo transfer within 12 months from embryo storage.

2. No impact on treatment effect was found for maternal age, blastocyst-stage transfer, percentage of close system, cryogenic tank employed, and good quality embryo transferred.

In the early days of in vitro fertilization (IVF), slow-freezing was the predominant method used to preserve embryos [31]. Parmigiani et al. and Machtinger et al. investigated the effect of this freezing method on oocytes and embryo competence. They found prolonged preservation did not affect embryo and oocyte survival or reproductive outcomes [32, 33]. However, this technique often led to the formation of ice crystals, which could damage the embryos and reduce their viability [31]. The concept of vitrification, a rapid-freezing process that avoids ice crystal formation, was introduced in the 1980s and is nowadays widely used in fertility clinics worldwide [34].

The ability to store embryos for extended periods without compromising their quality has pivotal implications offering numerous benefits, including increased flexibility in treatment timing, better synchronization with the recipient’s menstrual cycle, and the possibility to preserve surplus embryos for future use for oncological patients, women undergoing medical treatments that may impair fertility, and those who wish to delay childbearing for personal, professional or others reasons [35].

Vitrification process is not without its challenges, with complex biological mechanisms at play [36]. Vitrification involves a multitude of molecules, proteins, and cells that can be influenced by the vitrification technique [37]. The formation of ice crystals in the intra- and extracellular space is avoided through the use of high concentrations of cryoprotectants combined with a very rapid temperature drop to reach the vitreous state [38]. During the vitrification process, cells are dehydrated just before the cooling process begins by exposure to concentrations of cryoprotective agents (CPAs) high enough to reach the vitreous state in the intra- and extracellular space. This transformation is achieved with a combination of high CPA concentrations and extremely high cooling rates [39]. The main risk of the vitrification procedure is the potential cytotoxicity posed by the high CPA concentration required. A cryopreserved cell must return to its normal physiological temperature, surviving thawing and the accompanying ice warming [40]. During thawing/warming, two processes can drastically reduce the survival of cryopreserved cells: de novo crystal formation, recrystallization, and the formation of a strong osmotic gradient due to the drastic melting of intracellular ice in the context of rapid thawing [41–43]. Not least the pivotal influence of temperature fluctuation due to tanks opening might have an impact on clinical outcomes. Storage tanks with failed vacuum have a much higher evaporation rate than those with intact vacuum [44].

Some studies have investigated the effects of extended embryo storage on embryo structural integrity and viability, yielding contradictory results [19, 21, 32]. Therefore, there was an urgent need for an updated systematic appraisal of treatment effects and quality of evidence regarding the impact of embryo storage duration on reproduction and neonatal outcomes in patients undergoing embryo transfer.

A previously published study from Yan et al. has postulated that embryo storage, exceeding 6 years, might have a detrimental effect on embryo competence bringing to a decrease on clinical pregnancy, biochemical pregnancy, and LBR. Nevertheless, no impact was found on ectopic pregnancy, miscarriage, and neonatal outcomes [45]. Of note, a significantly time-dependent decrease on cryo-survival rates of the vitrified blastocysts with prolonged embryo storage was shown [45]. In a retrospective cohort study of 2688 euploid embryos, Cimadomo et al. analyzed the effect of prolonged vitrification storage across seven groups, with storage durations up to a maximum of 68 months. Only SET was considered. A lower LBR was observed in the univariate analysis [14], consistent with the findings of Li et al. and Zhang et al. in their investigations included in our systematic review meta-analysis [20, 24]. Those results were attributed by the authors to embryo quality, with the best embryo grades being selected for the first transfers. However, Cimadomo et al. included only euploid embryos, analyzed by Next-Generation Sequencing minimizing the effect of this bias. Against this background, in our systematic review and meta-analysis, we performed a meta-regression considering good embryo quality with no impact on treatment effect, by employing the best quality embryo for transfer. In a large retrospective clinical study, Cobo et al. found a statistical difference for LBR across certain, but not all, storage time subgroups [15].

Likewise, Zheng et al. analyzed 6327 vitrified-warmed blastocyst transfer cycles, at the first attempt, and found a correlation between the chance of clinical pregnancy, ongoing pregnancy, and live birth before and after adjustment for possible confounding factors and propensity score matching. The authors described a substantial decrease in reproductive outcomes, particularly LBR 44.57% vs 36.55% vs 35.09% vs 28.77%, for < 3 months, 4–6 months, 7–12 months, 13–24 months, respectively [21]. Contrarily, Lee et al. found no significant differences between cryo-storage time groups among 1632 autologous vitrified-warmed blastocyst transfer cycles [23].

Of note, in none of the aforementioned investigations consequences on neonatal outcomes were found, apart from Wang et al., who described an increased risk of preterm birth in pregnancies obtained from embryos vitrified for more than 12 months in 47,006 cycles [18]. Our systematic review and meta-analysis included 16 studies and 233,315 vitrified-thawed embryos, making it the most extensive systematic assessment of the effects of cryo-storage on embryo competence to date. In our research, we have used the 12-month cut-off due to its clinical application, as this time interval can align with daily practice patient requirements, such as the time needed for reproductive surgery and its recovery, weight loss, correcting hormonal imbalances, the desire for a new child after a successful pregnancy from the first embryo transfer, or a personal decision due to work or psychological reasons [46]. We also evaluated the impact of maternal age, blastocyst-stage transfer, percentage of close versus open system, type of cryogenic tank used (vapor vs liquid nitrogen tanks), and good quality embryo transferred, with no impact on treatment effects. Our results are in line with a recent meta-analysis that evaluated open vs close vitrification systems and did not find significant differences between the two methods in embryo or pregnancy outcomes [43]. However, our results conflict with a previously published meta-analysis which found no differences in clinical outcomes among patients undergoing early (< 12 months) versus prolonged (> 12 months) embryo transfer. The aforementioned metanalysis only included five studies and 18,047 embryos and did not represent an updated appraisal of current evidence [47].

Our results may be crucial for decision-making by assisted reproduction professionals and patients, helping them to be aware of the possible outcomes of delaying embryo transfer. Notwithstanding, further investigations are needed to determine a specific cut-off that could alter the prognosis for these patients.

Limitations and strengths

The present study should be interpreted in light of some limitations. The main limitation of this review is its reliance on retrospective studies, but currently constitute the sole source of published evidence. In addition, this is a study-level meta-analysis providing average treatment effects. The lack of patient-level data prevents us from assessing the impact of baseline clinical characteristics (i.e., presence of uterine factor) and other changes in therapeutic strategies (i.e., endometrial preparation, transfer technique, and embryo quality) on treatment effects. Moreover, euploidy status of the embryos only was stated in one study. However, this systematic review and meta-analysis present the following strengths:—including 16 studies and 233,315 embryos, is the largest meta-analysis to the best of our knowledge assessing reproductive and neonatal outcomes after frozen embryo transfer comparing stored embryos for less than or more than 12 months—all stratified analyses are combined with meta-regression analyses to determine any potential impact of the tested variables on effect estimates (blastocyst or cleavage embryo, close or open system, vapor or liquid nitrogen tank), although—considering the number of studies included—these analyses should only be considered as hypothesis-generating. Lastly, different definitions have been used in the studies here included, limiting the reliability of effect estimates for such outcomes.

Conclusions

Prolonged embryo storage for more than 12 months is associated with a statistical reduction of live births, biochemical pregnancy rate, clinical pregnancy, and multiple pregnancy rate. However, important covariates such as reasons for delay of transfer, embryo quality, and improper handling of embryos could not be ruled out as causes of this reduction. Given these limitations, these conclusions should be viewed with caution.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CPAs

Cryoprotective agents

CI

Confidence interval

IVF

In vitro fertilization

LBR

Live birth rate

ORs

Odds ratios

SET

Single embryo transfer

Data Availability

Data will be provided under request to corresponding author.

Declarations

Conflict of interest

The authors declare no competing interests.