Impact of Endometrial Receptivity Analysis on Pregnancy Outcomes In Patients Undergoing Embryo Transfer: A Systematic Review and Meta-Analysis

Abstract

To analyze the influence of endometrial receptivity analysis (ERA) on embryo transfer (ET) results in patients undergoing in vitro fertilization (IVF) treatment. PubMed, Embase, Cochrane Central Register of Controlled Trials, and BioMed Central databases were searched from inception up to December 2022 for studies comparing pregnancy outcomes in patients undergoing personalized embryo transfer (pET) by ERA versus standard ET. Data were pooled by meta-analysis using a random effects model. We identified twelve studies, including 14,224 patients. No differences were observed between patients undergoing ERA test and those not undergoing ERA test prior to ET in terms of live birth (OR 1.00, 95% CI 0.63–1.58, I² = 92.7%), clinical pregnancy (OR 1.20, 95% CI 0.90–1.61, I² = 86.5%), biochemical pregnancy (OR 0.83, 95% CI 0.46–1.49, I² = 87%), positive pregnancy test (OR 0.99, 95% CI 0.80–1.22, I² = 0%), miscarriage (OR 0.91, 95% CI 0.62–1.34, I² = 67.1%), and implantation rate (OR 1.18, 95% CI 0.44–3.14, I² = 93.2%). pET with ERA is not associated with any significant differences in pregnancy outcomes as compared to standard ET protocols. Therefore, the utility of ERA in patients undergoing IVF should be revisited.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10815-023-02791-2.

Introduction

Reaching a good-quality embryo is one of the main and most challenging goals in in vitro fertilization (IVF) treatment. Therefore, an unsuccessful embryo transfer (ET) having obtained a good-quality blastocyst is frustrating for both patients and doctors.

Despite many studies have been conducted with the aim to improve embryo quality [1–3], endometrial factor yet represents a challenge and is still one of the principal causes of ET failure, especially when biopsied euploid blastocysts are transferred [4].

The endometrial cycle is regulated by a complex hormone interaction with progesterone as the main ovulation component, leading to endometrial epithelium changes, induced by the luteinizing hormone peak, known as the window of implantation (WOI). In natural and hormone replaced cycles, the WOI can suffer a relevant shift, estimated in 2 or even more days [5].

In this context, many efforts have been made to achieve a personalized endometrial preparation in consequence of the powerlessness shown in the histological dating and morphological observations [6]. In 2011, Diaz-Gimeno et al. created an algorithm based on a panel of 238 genes: the endometrial receptivity array (ERA) test [7]. Using an endometrial biopsy by scratching the lining of the womb, the ERA test tries to identify the optimum timing of progesterone exposure and determine the exact moment for personalized embryo transfer (pET) by evaluating the gene expression profile with the array methodology. ERA provides two possible scenarios: a receptive or non-receptive endometrium and, additionally, itemizing a pre-receptive or post-receptive endometrium. Due to this data, the endometrium and programmed frozen ET can be prepared anticipating or delaying the WOI, if necessary. Some reports uphold that ERA results are reproducible within a timeframe of up to 40 months after the first biopsy is performed [5].

Since 2011, the use of ERA has increased in daily clinical practice even though the first randomized clinical trial aimed to evaluate its effectiveness was published in 2020. Published literature to date has shown conflicting results and individual studies might not provide an adequately powered analysis, raising a need for a systematic appraisal of treatment effects and quality of evidence. Therefore, the aim of this investigation was to provide an updated comprehensive and quantitative assessment of evidence concerning the usefulness of ERA test.

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 [8]. Two reviewers (IZ, JJHM) independently identified the relevant studies by an electronic search of the MEDLINE and Embase databases (from inception to December 2022). The search strategy is available in the Supplementary appendix. No language, publication date, study design, or publication status restrictions were imposed. This study is registered with PROSPERO (CRD42022332891).

Study selection

Two reviewers (IZ, JJHM) independently assessed trial eligibility on the basis of titles, abstracts, and full-text reports. Discrepancies in the study selection were discussed and resolved with another investigator (JMRR). Eligible studies had to satisfy the following pre-specified criteria: (1) studies including patients for IVF treatment; (2) investigations comparing the use of ERA test vs non-test; (3) the availability of pregnancy outcome data. Exclusion criteria were as follows: (1) studies using other endometrial receptivity test; (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.

Data extraction and quality assessment

Two investigators (IZ, JJHM) 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) [9], 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 (AMM).

Data synthesis and data analysis

Outcome measures

The primary endpoint was live birth. Secondary endpoints included positive pregnancy test, biochemical pregnancy, implantation rate, clinical pregnancy, and miscarriage.

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 [10]. 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 [11]. The presence of publication bias was assessed for the primary endpoint using funnel plots. The presence of publication bias was investigated 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 repeated implantation failure 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 addition, a random effects meta-regression was performed to assess the impact on treatment effect of maternal age, body mass index, percentage of PGT-A prior to ET, and percentage of donor cycles. 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 1492 citations screened, 7 were excluded as they were considered non-relevant, 1240 were excluded because of a preclinical design without ERA assessment, 3 because of lack of pre-specified clinical outcomes. Fourteen studies were not performed on humans, 19 included other diseases, and 123 for other reasons. Therefore, a total of 12 studies including 142,24 patients 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 patient characteristics of the included studies are reported in Table 1 and the main trial features in Tables 2 and 3. Two studies had a randomized design, three had a prospective cohort design, and seven studies had a retrospective cohort design. All studies included patients undergoing ET in blastocysts stage with a Gardner grade BB or higher except one including a cleavage stage [12]. In two studies, both fresh and frozen ETs were performed [12, 13]; in the remaining studies, only frozen embryos were transferred [14–23].

Table 1.

Baseline trial and patient characteristics included in the meta-analysis

Type	Bassil et al.		Bergin et al.		Cozzolino2020 et al.		Cozzolino2022 et al.
	ERA test	Non ERA test	ERA test	Non ERA test	ERA test	Non ERA test	ERA test	Non ERA test
Age (years) (mean)	36.30	35.60	36.25	34.84	38.50	38.01	39.19	38.0
Nº oocytes retrieved (mean)	13.20	12.80	17.01	16.58	14.26	14.80	NR	NR
Nº of previous embryo transferred (mean)	1	1.1	1.60	0.53	1.62	1.44	NR	NR
Endometrial thickness (mm) (mean)	8.95	9.1	NR	NR	8.3	8.9	NR	NR
Type	Doyle et al.		Doyle2022 et al.		Fodina et al.		Jia et al.
	ERA test	Non ERA test	ERA test	Non ERA test	ERA test	Non ERA test	ERA test	Non ERA test
Age (years) (mean)	36.7	36.7	34.7	34.5	35.0	34.0	32.01	31.87
Nº oocytes retrieved (mean)	NR	NR	NR	NR	11.5	13.0	NR	NR
Nº of previous embryo transferred (mean)	NR	NR	NR	NR	NR	NR	NR	NR
Endometrial thickness (mm) (mean)	NR	NR	10.1	10.0	NR	NR	9.7	9.5
Type	Neves et al.		Ohara et al		Riestenberg et al.		Simón et al.
	ERA test	Non ERA test	ERA test	Non ERA test	ERA test	Non ERA test	ERA test	Non ERA test
Age (years) (mean)	40.93	41.58	38.2	38.5	36.9	34.9	33	32.8
Nº oocytes retrieved (mean)	NR	NR	NR	NR	NR	NR	NR	NR
Nº of previous embryo transferred (mean)	3.01	2.65	5.64	5.83	NR	NR	NR	NR
Endometrial thickness (mm) (mean)	10.02	10.25	NR	NR	9.0	8.8	NR	NR

ERA, endometrial receptivity analysis; NR, not recorded

Table 2.

Study key features

Study	Year of publication	Design	N of patients			Multicentric	Primary outcome/s
			Overall	ERA test	Control
Bassil et al.	2018	Retrospective cohort study	556	53	503	No	Pregnancy rate
Bergin et al.	2021	Retrospective cohort study	486	133	353	No	Live birth rate
Cozzolino et al.	2020	Retrospective cohort study	2598	153	2445	Yes	Implantation and ongoing pregnancy rate
Cozzolino et al.	2022	Retrospective cohort study	5372	802	4570	Yes	Live birth rate
Doyle et al.	2022	Retrospective cohort study	2591	307	2284	No	Live birth rate
Doyle et al.	2022	Randomize controlled trial	767	381	386	Yes	Live birth rate
Fodina et al.	2021	Retrospective cohort study	253	94	159	No	Biochemical and clinical pregnancy
Jia et al.	2022	Prospective cohort study	281	140	141	No	Implantation and clinical pregnancy rate
Neves et al.	2019	Retrospective cohort study	333	56	277	No	Implantation and clinical pregnancy rate
Ohara et al.	2022	Retrospective cohort study	550	244	306	No	Clinical pregnancy rate and live birth rate
Riestenberg et al.	2021	Prospective cohort study	228	147	81	No	Live birth rate
Simón et al.	2020	Randomize controlled trial	302	148	154	Yes	Live birth rate

Table 3.

Study key features

Study	Study period	Embryo stage	Type of transfer	PGT-A
Bassil et al.	2016–2017	Blastocyst	Frozen embryo transfer	No
Bergin et al.	2014–2019	Blastocyst	Frozen embryo transfer	No
Cozzolino2020 et al.	2013–2018	Blastocyst	Frozen embryo transfer	Yes and no
Cozzolino2022 et al.	NR	Cleavage and blastocyst	Fresh and frozen embryo transfer	No
Doyle et al.	2018–2019	Blastocyst	Frozen embryo transfer	Yes
Doyle2022 et al.	2018–2020	Blastocyst	Frozen embryo transfer	Yes
Fodina et al.	2017–2020	Blastocyst	Frozen embryo transfer	Yes and no
Jia et al.	2019–2021	Blastocyst	Frozen embryo transfer	No
Neves et al.	2012–2018	Blastocyst	Frozen embryo transfer	Yes or donor
Ohara et al.	2019–2020	Blastocyst	Frozen embryo transfer	No
Riestenberg et al.	2018–2019	Blastocyst	Frozen embryo transfer	Yes
Simón et al.	2013–2017	Blastocyst	Fresh and frozen embryo transfer	No

NR, not reported

Clinical outcomes

No differences were observed between patients undergoing ERA test and those not undergoing ERA test prior to ET in terms of live birth (OR 1.00, 95% CI 0.63–1.58, I² = 92.7%), clinical pregnancy (OR 1.20, 95% CI 0.90–1.61, I² = 86.5%), biochemical pregnancy (OR 0.83, 95% CI 0.46–1.49, I² = 87%), positive pregnancy test (OR 0.99, 95% CI 0.80–1.22, I² = 0%), miscarriage (OR 0.91, 95% CI 0.62–1.34, I² = 67.1%), and implantation rate (OR 1.18, 95% CI 0.44–3.14, I² = 93.2%) (Figs. 2 and 3).

Fig. 2.

Pooled analysis of studies analyzing the impact of endometrial receptivity analysis on live birth (A), clinical pregnancy (B), and biochemical pregnancy (C). Forest plot reporting study-specific and summary odds ratios (ORs) with 95% confidence intervals (CIs)

Fig. 3.

Pooled analysis of studies analyzing the impact of endometrial receptivity analysis on positive pregnancy test (A), miscarriage (B), and implantation rate (C). Forest plot reporting study-specific and summary odds ratios (ORs) with 95% confidence intervals (CIs)

Risk of bias assessment

Supplementary appendix Tables 3 and 4 summarize the results of the risk of bias assessment with the ROBINS-I and RoB 2.0 tools. Two studies were considered at low risk for overall risk of bias [13, 18], three were considered at moderate overall risk of bias [20–22], five were considered at serious risk [12, 15–17, 19], and two at critical overall risk of bias [14, 23].

Sensitivity analyses

At leave-one-out sensitivity analysis after removing Ohara et al, patients undergoing ERA test were associated with a lower risk of live birth (OR 0.75, 95% CI 0.57–0.99) (Supplementary appendix Table 3). Results remained consistent with the main analysis when iterative removing the rest of included studies.

Meta-regression analysis did not show any significant interaction between maternal age (p = 0.588), body mass index (p = 0.229), percentage of PGT-A prior to ET (p = 0.709), and percentage of repeated implantation failure (RIF) (p = 0.219) and treatment effects.

Funnel plot distributions of the pre-specified outcomes as well as Harbord tests indicated absence of publication bias and small study effect for all the outcomes (Supplementary appendix Figures 1–6).

Discussion

In this study, we evaluated the impact of ERA test on pregnancy outcomes in patients undergoing ET. The main findings of our investigation can be summarized as follows:

1. The risk of live birth, positive pregnancy test, biochemical pregnancy, miscarriage, clinical pregnancy, and implantation rate did not differ between patients undergoing ERA test and those not undergoing ERA test prior to ET.

2. No impact on treatment effect was found for maternal age, BMI, percentage of PGT-A prior to ET, or percentage of donor cycles.

The interaction between endometrium and embryo, essential for the evolution of pregnancy, is subject to the control of multiple factors. Endometrial receptivity and its dysfunction are considered a frequent cause of infertility and many studies have been carried out to investigate endometrial receptivity by analyzing biomarkers from histological, immunochemical, proteomic, and genetic samples [24–27].

A previously published study has postulated that the mid-luteal phase decidualization might have two essential functions for the correct endometrial implantation: selective or receptive. If the receptive function prevails, we could face a non-viable embryo implantation with a higher risk of abortion, while a high selectivity without endometrial receptivity would lead to implantation failure.

Until 2011, the unique method used to study endometrial receptivity was based on the histological dating of the endometrium (Noyes test), from an endometrial biopsy. The Noyes test was, however, associated with poor clinical correlation and deficient correspondence between fertile and infertile populations [6]. The introduction of the ERA test in 2011 opened a new horizon in the personalization of the progesterone-endometrium interaction [7]. Since then, ERA has been widely used in daily clinical practice and many studies have been performed to assess ERA’s efficacy with conflicting results [28–30], postulating that different results in ERA test may result from a test error [31]. In its early days, ERA used the array technology to analyze the sequence of genes involved in endometrial receptivity. However, since January 2017 [32], this hybridization method was replaced by next generation sequencing (NGS), a more powerful molecular analysis based on DNA polymerase sequencing procedure [33, 34]. Seven studies included in the present meta-analysis were performed after 2017 [14, 17–20, 22, 23], four studies were performed prior to 2017 [13, 15, 16, 21], and one did not report the enrollment period [12]. Hence, further studies using only NGS technology will be needed to corroborate our findings.

The extensive use of the ERA test is not risk-free. It is crucial to consider that this test requires an invasive method to obtain the sample. Endometrial biopsy is associated, even if in rare cases, with infections, bleeding, and pain [35]. Moreover, the time required for the diagnostic test and its results will delay the entire IVF process which might cause psychological stress in some patients and increases the cost of the treatment.

A previous published meta-analysis with fewer included studies and smaller sample size did not lead to clarify if ERA test increases or decreases pregnancy rates in ET [36].

This systematic review and meta-analysis included thirteen studies and 14,224 patients, 2686 tested with ERA. In this extensive investigation, we analyzed three prospective clinical trials and two randomized controlled trials. Simon et al. showed a significant increase in cumulative live birth, implantation, and pregnancy rates in frozen ET personalized with ERA test versus fresh or frozen embryo with standard timing transfer [13]. In a recent multicentric trial, including 30 private centers, Doyle et al found no benefit from ERA test in terms of live births [18]. Just one more study was identified with a positive impact of ERA on pregnancy rates. Ohara et al. found a higher clinical pregnancy rate and live birth rate in pET as compared to standard ET in 1000 patients with RIF [22]. Those findings were confirmed when considering maternal age. Cozzolino et al., in two retrospective cohorts, analyzed the effect of ERA on a RIF population with a single previous ET. None of the aforementioned studies was an improvement in pregnancy or live birth rates secondary to the personalization of ET by ERA found. Contrariwise, a recently published study described worse pregnancy outcomes in patients undergoing pET [12, 16].

In our investigation, after pooling data from 12 studies and 14,224 patients, the risk of live birth, positive pregnancy test, biochemical pregnancy, miscarriage, clinical pregnancy, and implantation rate did not differ between patients undergoing ERA test and those not undergoing ERA test prior to ET. Therefore, the utility of ERA in patients undergoing IVF should be revisited.

Limitations

The present study should be interpreted in light of some limitations. First, 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 and other changes in therapeutic strategies on treatment effects. However, all stratified analyses are combined with meta-regression analyses to determine any potential impact of the tested variables on effect estimates, although—considering the low number of studies included—these analyses should only be considered hypothesis-generating. Second, different definitions have been used in the studies here included, limiting the reliability of effect estimates for such outcomes.

Conclusions

pET with ERA is not associated with any significant differences in pregnancy outcomes as compared to standard ET protocols. Therefore, the utility of ERA in patients undergoing IVF should be revisited.

Declarations

Competing interests

The authors declare no competing interests.