This systematic review and meta-analysis were conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (17). The protocol of this meta-analysis was registered in PROSPERO (CRD42024563268).

We conducted a systematic review and meta-analysis compared the pregnancy outcomes after different embryo transfer strategies in patients with EMS and adenomyosis. In all cases, EMS and adenomyosis were diagnosed by biopsy through surgery or medical imaging evidence like transvaginal ultrasound (TVUS) or magnetic resonance imaging (18). In TVUS, ovarian endometrioma appeared as a persistent unilocular or multilocular cyst with homogeneous low-level echogenicity of the cyst fluid and absent or moderate cyst wall vascularization. Deeply infiltrating endometriosis in the TVUS appeared as thick tissue blocks, nodular formations, or irregularly shaped hypoechoic, commonly affecting the uterosacral ligament, pouch of Douglas, and/or vagina. In addition, adenomyosis showed myometrial cystic areas, hyperechoic islands, linear striations, and buds or irregular/infiltrated endometrial-myometrial junction zones (19–21).

First, we conducted a comparative analysis across the entire study population, including individuals with EMS or adenomyosis. Subsequently, we performed a subgroup analysis based on the included studies. One analysis comprised nine studies focusing on EMS, while the other included two studies on adenomyosis. We separately summarized the gestational outcomes for patients with EMS or adenomyosis.

A systematic search for relevant papers was carried out in PubMed, EMBASE, MEDLINE, Web of Science, Google Scholar, China National Knowledge Infrastructure, Wanfang Data Knowledge Service Platform, and China Biomedical Literature Database. English search keywords included “endometriosis”, “adenomyosis”, “in vitro fertilization/IVF-ET”, “Intracytoplasmic sperm injection/ICSI”, “freeze-all/frozen embryo transfer”, “Fresh Embryo Transfer”, and “pregnancy outcomes”, as showed in support information. No restrictions were placed on the language or publication date of the studies. Additionally, we examined the reference lists of the eligible studies and review articles to identify any additional relevant articles.

Studies were included in this meta-analysis if they matched the following inclusion criteria (1): subjects of study were women diagnosed with either EMS or adenomyosis. (2) study focused on the pregnancy outcomes of different embryo transfer types in IVF/ICSI cycles. The exposure group consisted of women undergoing FET, while the control group comprised those undergoing fresh ET. (3) the outcomes include at least one of the following: implantation rate, miscarriage rate, clinical pregnancy rate, ectopic pregnancy rate, or live births rate. (4) randomized controlled trials (RCTs) or cohort studies. exclusion criteria: (1) study involved with donor or recipient oocyte treatments. (2) Review articles, abstracts, letters, conference papers, and case reports were excluded. (3) Necessary data was not available. (4) studies included animal experiments.

All titles and abstracts were independently reviewed by two investigators (Yixian Han and Lukanxuan Wu). Studies that potentially met the inclusion criteria were further assessed through full-text review. Any discrepancies were resolved by consulting a third author(Chang Liu). The kappa statistic was used to evaluate inter-examiner agreement in study selection (22). Both researchers manually extracted data from the included studies using specially designed data collection forms. The following data were collected: lead author, publication year, country, study period, study design, intervention, age, body mass index, duration of infertility, diagnosis, EMS phenotype or stage, stimulation protocol, FET protocol, oocyte retrieved, and pregnancy outcomes.

The quality of the included studies was evaluated using the Newcastle Ottawa scale (NOS) for assessing observational studies based on selection, comparability, and outcome domains. The total score of this scale was 9 points (0–3: poor quality, 4–6: fair quality,7–9: good quality) (23).

The primary outcomes were clinical pregnancy and live birth rate, and the secondary outcomes were miscarriage, implantation rate, and ectopic pregnancy rate. The implantation rate was defined as the proportion of transferred embryos that were successfully implanted. Clinical pregnancy was confirmed by the presence of at least one gestational sac on ultrasound, including ectopic pregnancy. Ectopic pregnancy was defined as the detection of a gestational sac outside the uterine cavity using TVUS, or clinically diagnosed when no gestational sac is observed within the uterine cavity but serum hCG levels continue to rise. A miscarriage was defined as the loss of gestation before 28 weeks of pregnancy. Live birth was described as a live birth event after 28 weeks.

Statistical analysis was performed with Review Manager 5.4 software. All the outcomes in our analysis were binary. Dichotomous data was analyzed using the Mantel–Hansel odds ratio and the CIs between groups. The I² statistic and Cochran’s Q-statistic were used to assess methodologic and clinical heterogeneity across studies. High heterogeneity was defined as I² ≥ 50% or p < 0.10, in which case a random-effects model was applied; otherwise, a fixed-effects model was used. Sensitivity analysis and subgroup analysis were conducted when I² or Cochran’s Q-statistic detected significant heterogeneity. Funnel plot analysis was used to assess the potential publication bias. All statistical tests were two-sided, and p values < 0.05 were considered statistically significant. The GRADE approach was employed to evaluate the quality of each outcome (24). This assessment considered factors such as risk of bias, inconsistency, indirectness, imprecision, and publication bias. Based on the GRADE criteria, the evidence was categorized into four levels: high, moderate, low, and very low quality.

In the initial screening, 2,465 citations were identified across the databases. After removing duplicates, 1,513 articles were evaluated based on their titles and abstracts. Subsequently, 43 articles were retrieved for full-text assessment. After a thorough examination, 5 articles were excluded due to unsuitable study types, 24 for lack of relevance, and 3 for not reporting relevant outcomes. Ultimately, 11 articles that met our eligibility criteria were included in the final analysis. A high level of concordance was observed among the reviewers in terms of data screening and integration (kappa = 0.81). Of the included studies, 4 were published in Chinese and 7 in English. The flow chart of the study selection is demonstrated in Figure 1 .

A total of 5650 patients were enrolled, with 3127 undergoing FET and 2523 undergoing fresh ET in the IVF/ICSI cycles. Six studies originated from China (15, 25–29), two from France (11, 12). and the remaining three from Canada, England, and Turkey, respectively (13, 14, 30). Of the eleven studies, two primarily addressed adenomyosis (12, 15), while the remaining nine focused on various types of EMS (8, 11, 13, 14, 25, 27–30). Among the EMS studies, three exclusively included patients with ovarian endometriomas, rASRM stage I-II EMS, or rASRM stage III-IV EMS, respectively (25, 26, 30). In four studies, the diagnosis of EMS was made via laparoscopy or laparotomy, while in five studies, it was determined using either laparoscopy or imaging examinations such as TVUS and magnetic resonance imaging. Adenomyosis was diagnosed by TVUS. Among the included studies, all involved clinical pregnancy rate, five involved embryo implantation rate (25, 26, 28–30), five involved ectopic pregnancy rate (25, 26, 28–30), ten involved live birth rate (11–13, 15, 25–30), and ten involved miscarriage rate (11, 13–15, 25–30). However, the definitions of miscarriage rate varied, with pregnancy loss ranging from 6 to 28 weeks, potentially introducing confounding factors. Therefore, we adopted the most common definition of miscarriage rate, specifically as a pregnancy loss occurring before 28 weeks after clinical pregnancy confirmation, and excluded four studies with differing definitions of miscarriage rate (13, 14, 26, 30). Table 1 presents the characteristics of each study included and Table 2 outlines the main results from each study included in the analysis.

Funnel plots for each of the meta-analyses, as shown in Supplementary Figure S1 , appeared to be symmetrical, except for the ectopic pregnancy rate, which suggested a potential, though minimal, publication bias likely due to the non-publication of smaller studies.

The NOS used for assessing the quality of studies in this meta-analysis is presented in Table 3 . The included studies received scores ranging from 7 to 9, indicating they were classified as high-quality studies with a low risk of bias.

The quality of evidence for each outcome was evaluated using the GRADE approach as demonstrated in Supplementary Figure S4 . The findings revealed that the clinical pregnancy and live birth rates, as well as the ectopic pregnancy rate, following different embryo transfer strategies for patients with EMS and adenomyosis, or EMS alone, were assessed as having moderate certainty of evidence. In contrast, all other outcomes were rated as low or very low certainty of evidence.

This review has several limitations and strengths that may have influenced the findings. The strengths of our study include the following: it is a large-scale meta-analysis based on 11 studies with a novel focus on patients with EMS and adenomyosis. We conducted a systematic literature search adhering to strict inclusion and exclusion criteria, ensuring a rigorous methodology. In addition to reporting the pregnancy outcomes of clinical pregnancy and live birth, we also addressed key reproductive outcomes, such as ectopic pregnancy, which is particularly significant in the context of assisted reproductive treatments. Despite the precautions taken, our study is subject to certain limitations. The quality of each of the included studies varies. Since there are some significant differences in baseline characteristics concerning the age of patients, type of infertility, ovarian reserve function and ovarian protocols, among others. it could introduce confounding factors into the results. Additionally, the included studies differed in their ascertainment of EMS and adenomyosis, without explicitly distinguishing between the two conditions. This lack of distinction may have led to the analysis of gestational outcomes in a mixed group. Furthermore, our study did not examine pregnancy complications or obstetric outcomes, which warrant further investigation.

In clinical practice, factors such as uterine environment, endometrial receptivity, and the patient’s overall health should be carefully considered during IVF/ICSI cycles before embryo transfer. Based on current evidence, for women with EMS undergoing IVF/ICSI, clinicians might prioritize FET protocols over fresh embryo transfer. despite the potential for increased financial costs. Stimulation protocols should be adjusted to optimize oocyte retrieval and subsequent cryopreservation for FET cycles. For women with adenomyosis, pregnancy outcomes appear comparable between FET and fresh ET strategies. Clinicians should evaluate uterine abnormalities using imaging or hysteroscopy to guide decisions regarding endometrial preparation and the appropriate type of embryo transfer.

Heterogeneity exists across the studies due to substantial variations in study design, participant characteristics, and sample size. Although subgroup analyses for EMS and adenomyosis were performed to mitigate some of the heterogeneity, differences in the types of observational studies and patient characteristics, particularly variations in disease severity among EMS or adenomyosis patients, contribute to variability in baseline data, potentially impacting the outcomes. Consequently, randomized controlled trials focusing on different subtypes and severity levels of EMS and adenomyosis are warranted to better evaluate the efficacy and applicability of various embryo transfer strategies.