Impact of Intrauterine Infusion of Autologous Platelet‐Rich Plasma on Assisted Reproductive Outcomes in Patients With Recurrent Implantation Failure: A Meta‐Analysis of Randomized Controlled Trials

ABSTRACT

Background

This meta‐analysis evaluates the efficacy and safety of intrauterine autologous platelet‐rich plasma (PRP) infusion in patients with recurrent implantation failure (RIF).

Methods

Following PRISMA guidelines, we systematically searched major databases and included 10 randomized controlled trials (RCTs, n = 1 188). Study quality was assessed using the Cochrane RoB 2.0 tool.

Results

PRP infusion significantly improved clinical pregnancy rate (RR = 2.23, 95% CI 1.84–2.69) and live birth rate (RR = 3.74, 95% CI 2.68–5.22) without increasing miscarriage risk. Subgroup analysis indicated that a 0.5 mL PRP volume was as effective as 0.5–1.0 mL, and a notably greater treatment effect was observed in blastocyst transfer cycles compared to cleavage‐stage embryo transfers, suggesting potential enhanced benefit in this subgroup.

Conclusion

Intrauterine PRP infusion is an effective and safe treatment for RIF. The findings on dose equivalence and differential efficacy by embryo stage offer valuable insights for optimizing clinical application.

1. Introduction

Over the past four decades, assisted reproductive technology (ART) has achieved remarkable progress. However, approximately 15% of couples still experience RIF, a challenging condition that imposes significant psychological and economic burdens and remains a pressing clinical issue for reproductive specialists [1]. Currently, there is no unified definition of RIF. Studies have shown that approximately 18.7% of fertility centers diagnose RIF after two failed embryo transfers [2]. The widely accepted definition proposed by Coughlan in 2014 states that RIF refers to the failure to achieve clinical pregnancy in patients under 40 years old who have undergone In vitro fertilization‐embryo transfer (IVF‐ET) with at least four high‐quality embryos transferred across at least three fresh or frozen cycles [3]. The 2023 Practice Recommendations of the European Society of Human Reproduction and Embryology (ESHRE) on RIF define it as a condition in which specific patients fail to yield positive pregnancy test results after transfer of a sufficient number of viable embryos [4].

The etiology of RIF is complex, involving infections, anatomical abnormalities of the reproductive system, poor endometrial receptivity, immune disorders, prethrombotic states, and endocrine imbalances. Existing interventions range from hysteroscopy, endometrial scratching, preimplantation genetic testing, to immunomodulation (fat emulsion, glucocorticoids, low‐molecular‐weight heparin) and lifestyle optimization, yet none of these strategies have shown consistent benefits.

Among endometrial interventions, intrauterine infusion of PRP has attracted increasing attention in recent years. PRP is a plasma product obtained by centrifugation, containing high concentrations of platelets, leukocytes, and fibrin. Studies have reported that intrauterine PRP infusion can safely and effectively improve the clinical pregnancy rate (CPR), implantation rate (IR), and live birth rate (LBR) in RIF patients with non‐thin endometrium [5]. However, another study suggested that although PRP increases endometrial thickness, it does not significantly improve reproductive outcomes [6]. Recently, several RCTs have investigated PRP infusion prior to embryo transfer in RIF populations, but the results remain contradictory [7, 8]. Moreover, three additional meta‐analyses related to RIF patients with non‐thin endometrium were published in 2024 [7, 8, 9]. Therefore, it is necessary to conduct an updated meta‐analysis based on the latest high‐quality RCTs to provide more robust and detailed evidence regarding the efficacy and safety of PRP.

Compared with previous studies, this research offers several innovations: (1) strict restriction to RCTs to provide higher‐level evidence; (2) inclusion of newly published studies to ensure comprehensive and timely analysis; (3) further exploration of the potential impact of factors such as PRP volume and embryo transfer stage on therapeutic efficacy through subgroup analyses on the basis of evaluating overall effectiveness; (4) verification of the robustness of the main conclusions through rigorous sensitivity analysis, thereby providing a more solid evidence‐based medical foundation for the standardized application of PRP in RIF patients.

2. Materials and Methods

This study was reported based on the Preferred Reporting Items for Systematic Reviews and Meta‐analyses (PRISMA) 2020 Checklist (David et al.) (http://www.prisma‐statement.org/). The protocol of the meta‐analysis was registered in the International Prospective Register of Systematic Reviews (PROSPERO, https://www.crd.york.ac.uk/PROSPERO/) with the registration number of PROSPERO CRD420251179580. Chemical pregnancy was defined as a positive beta human chorionic gonadotropin (β‐hCG) 14 days after the ET, the threshold for positive β‐hCG varied across studies (typically 5–10 mIU/mL), as per each study's protocol. Clinical pregnancy was defined as one or more gestational sac (s) seen in the uterine cavity by transvaginal ultrasound at 4–6 weeks after ET, regardless of whether fetal heart activity was observed. The implantation rate was defined as the number of gestational sac on transvaginal ultrasound by the number of the transferred embryos. Ongoing pregnancy is defined as a pregnancy beyond 12 weeks of gestation. Live birth was defined as the delivery of one or more living infant (s), and a miscarriage was defined as the loss of a clinical pregnancy (i.e., disappearance of the gestational sac or lack of fetal heart development) before 20 weeks of gestation.

2.1. Search Strategy and Selection Criteria

A systematic literature search was performed in the Medline, Web of Science, Embase, Cochrane Library, and Scopus databases to identify RCTs investigating intrauterine PRP infusion for the treatment of RIF patients. The search period covered from the establishment of each database to November 2025, with language restricted to English. We employed various combinations of the following search terms: “Platelet‐Rich Plasma”, “Plasma, Platelet‐Rich”, “Platelet Rich Plasma”, “PRP”, “Recurrent Implantation Failure”, “Repeated Implantation Failure, RIF”.

Inclusion criteria: (1) Study design: RCTs; (2) Participants: infertile women who have experienced at least 2 implantation failures; (3) Intervention: intrauterine PRP infusion prior to embryo transfer; (4) Outcomes: evaluation of at least one of the following indicators: CPR, LBR, chemical pregnancy rate, IR, MR, ongoing pregnancy rate (OPR).

Exclusion criteria: (1) Single‐arm studies or pre‐post design studies; (2) Case reports; (3) Literature types including reviews or conference abstracts; (4) Duplicate publications; (5) Studies with unavailable detailed information on research methods or results, and data cannot be obtained even after contacting the original authors; (6) Original trials with inadequate study design, unclear description of sample data or outcome indicators.

2.2. Data Extraction

Two reviewers (Xinhong Liang and Xi Li) independently screened the titles and abstracts of the retrieved literature based on the inclusion and exclusion criteria, and cross‐checked the results. If a study was considered potentially relevant by the reviewers, the full text was retrieved for further evaluation. Data extracted from each included study were as follows: first author, publication year, region, study design, characteristics of study participants, intervention measures, detailed information on outcome indicators, and elements for risk of bias assessment. The methodological quality of the included RCTs was evaluated in accordance with the recommendations of the Cochrane Handbook.

2.3. Data Analysis

Meta‐analysis was performed using Review Manager 5.4 software. Since all included studies were prospective RCTs, the DerSimonian & Laird random‐effects model was employed to pool effect sizes for accounting for heterogeneity between trials. For dichotomous variables, the risk ratio (RR) and 95% confidence interval (CI) were calculated, with a p < 0.05 considered statistically significant. Heterogeneity among included studies was assessed using the I ² statistic. An I ² value > 50% indicated significant heterogeneity, and subgroup analyses were conducted to explore the sources of heterogeneity and eliminate the impact of significant heterogeneity. To evaluate the influence of individual studies on the pooled effect, a predefined sensitivity analysis was performed in this study: studies with the highest weight (> 30%) or extreme results were sequentially excluded, and the remaining studies were re‐pooled to observe whether the pooled effect exhibited directional changes or statistically significant alterations. If the 95% CI of the RR value crossed 1.0 and > 0.05 after exclusion, the original conclusion was considered robust. Publication bias was assessed using funnel plots and Egger's regression only when the number of included RCTs was ≥ 10; otherwise, the test power was deemed insufficient.

3. Results

3.1. Study Selection

A total of 577 studies were retrieved from the aforementioned databases. After removing duplicates, 197 studies were initially screened, among which 47 were assessed for eligibility. Finally, 10 RCTs involving 1188 women were included in the meta‐analysis (Figure 1). The reasons for exclusion were as follows: ongoing studies (n = 10), missing data (n = 4), studies involving thin endometrium/granulocyte colony‐stimulating factor (G‐CSF) (n = 23), and unavailable full text (n = 10). The flow chart of literature search and study selection is shown in Figure 1.

FIGURE 1.

PRISMA flow diagram of study selection.

3.2. Characteristics of the Included Studies

Table 1 provides an overview of the characteristics of the 10 included RCTs [7, 8, 9, 10, 11, 12, 13, 14, 15, 16]. All studies were published between 2020 and 2024, with sample sizes ranging from 33 to 196 participants. Autologous PRP was infused 48 h before embryo transfer in all studies, with the volume ranging from 0.5 mL to 1.5 mL. The majority of the studies adopted hormone replacement therapy (HRT) protocols for endometrial preparation, among which six studies transferred blastocyst‐stage embryos and four studies transferred cleavage‐stage embryos.

TABLE 1.

General characteristics of the included studies.

study	year	Study type	region	RIF definition (n)	Age	Intervention case control		Time of PRP infusion	Sample size (n)	dose	concentration	Embryonic stage	outcomes
Baybordi	2022	RCT	Iran	≥ 3	18–45	PRP	Sham surgery method	48 h before embryo transfer	94 (48/46)	0.5–1 mL	No mentioned	Blastocyst	Chemical pregnancy rate, CPR, OPR, MR
Eftekhar	2024	RCT	Iran	≥ 2	18–42	PRP	HRT	48 h before embryo transfer	66 (33/33)	0.5–1 ml	Platelets 4–5 times more than peripheral blood	Cleavage	Chemical pregnancy, CPR,OPR,IR
Ershadi	2022	RCT	Iran	2–3	< 40	PRP	HRT	48 h before embryo transfer	85 (40/45)	0.5 ml	Platelets 4–5 times more than peripheral blood	Cleavage	Chemical pregnancy rate, CPR, IR, MR
Fazaeli	2024	RCT	Iran	≥ 2	< 40	PRP	Standard treatment	48 h before embryo transfer	64 (32/32)	0.5–1 ml	Platelets 4–5 times more than peripheral blood	Cleavage	Chemical pregnancy, CPR
Nazari	2020	RCT	Iran	≥ 3	< 40	PRP	HRT	48 h before embryo transfer	97 (49/48)	0.5 ml	Platelets 4–5 times more than peripheral blood	Blastocyst	Chemical pregnancy rate, CPR
Nazari	2022	RCT	Iran	≥ 3	18–38	PRP	HRT	48 h before embryo transfer	393 (196/197)	0.5 ml	Platelets 4–5 times more than peripheral blood	Blastocyst	Chemical pregnancy rate, CPR, MPR, LBR
Safdarian	2022	RCT	Iran	≥ 3	20–40	PRP	HRT	48 h before embryo transfer	120 (60/60)	0.5 ml	Platelets 4–5 times more than peripheral blood	Blastocyst	Chemical pregnancy rate, CPR, OPR, LBR, MPR, MPR
Yahyaei	2024	RCT	Iran	≥ 3	< 40	PRP	Standard treatment	48 h before embryo transfer	69 (34/35)	0.8‐1 ml	Platelets 4–6 times more than peripheral blood	Blastocyst	CPR, IR, LBR, MR
Zamaniyan	2021	RCT	Iran	≥ 3	20–40	PRP	HRT	48 h before embryo transfer	98 (55/43)	0.5 ml	Platelets 4–7 times more than peripheral blood	Blastocyst	Chemical pregnancy rate, CPR, IR, MR
Zargar	2021	RCT	Iran	≥ 2	< 41	PRP	Standard treatment	48 h before embryo transfer	80 (40/40)	1.5 ml	No mentioned	Cleavage	CPR, MR, LBR

Abbreviations: Clinical Pregnancy Rate (CPR); Implantation Rate (IR); Miscarriage Rate (MR); Live Birth Rate (LBR); Ongoing Pregnancy Rate (OPR); Multiple Pregnancy Rate (MPR).

The specific information about the preparation protocol of PRP, including centrifugation parameters, platelet concentration, activation method and perfusion volume, was summarized in Table S1. It is worth noting that the completeness of the reports on the characteristics of PRP preparation varied greatly among the included studies. Only a few studies described the centrifugation speed and time in detail, and no studies reported the absolute platelet count or white blood cell content.

3.3. Risk of Bias Assessment

Figure 2 presents the results of the risk of bias assessment for the included studies. Most studies exhibited a low risk of bias in terms of random sequence generation and allocation concealment. However, due to the inability to blind participants, the risk of performance bias and detection bias was high. The overall quality of evidence was rated as moderate, downgraded by one level each for risk of bias and imprecision.

FIGURE 2.

Assessments about risk of bias of included studies. (A) risk of bias graph and (B) risk of bias summary.

3.4. Primary Findings

The meta‐analysis included 10 eligible studies comparing intrauterine PRP infusion with control groups. Primary outcomes were CPR and LBR; secondary outcomes included BPR, IR, MR, and OPR. Figures 3, 4, 5, 6, 7, 8, 9, 10, 11 depict the measured values of each outcome in the forest plots.

FIGURE 3.

Forest plot of RR, 95% CI, and heterogeneity in studies that evaluated the risk of clinical pregnancy in interventions versus controls.

FIGURE 4.

Forest plot of RR, 95% CI, and heterogeneity in studies that evaluated the risk of positive serum β‐HCG rates 14 days after ET in interventions versus control.

FIGURE 5.

Forest plot of RR, 95% CI, and heterogeneity in studies that evaluated the risk of IRs in interventions versus control.

FIGURE 6.

Forest plot of RR, 95% CI, and heterogeneity in studies that evaluated the risk of OPRs in interventions versus control.

FIGURE 7.

Forest plot of RR, 95% CI, and heterogeneity in studies that evaluated the risk of LBRs in interventions versus control.

FIGURE 8.

Forest plot of RR, 95% CI, and heterogeneity in studies that evaluated the risk of MRs in interventions versus control.

FIGURE 9.

Forest plot of RR, 95% CI, and heterogeneity in subgroup analysis that evaluated the risk of CPRs with different PRP doses in interventions versus control.

FIGURE 10.

Forest plot of RR, 95% CI, and heterogeneity in subgroup analysis that evaluated the risk of pregnancy rates with different embryonic stages in interventions versus control.

FIGURE 11.

Forest plot of RR, 95% CI, and heterogeneity in subgroup analysis that evaluated the risk of sensitivity analysis on MRs in interventions versus control.

3.4.1. Clinical Pregnancy Rates

All 10 studies compared the clinical pregnancy rate between the PRP group and the control group, involving a total of 1188 participants (592 in the PRP group and 596 in the control group). The results of the meta‐analysis using the random‐effects model indicated that PRP significantly increased the clinical pregnancy rate (CPR) (252/592 vs. 114/596; RR = 2.23, 95% CI: 1.84–2.69, p < 0.00001), with no significant heterogeneity observed among the studies (I ² = 0%, χ² = 7.88, df = 9, p = 0.55) (Figure 3).

3.5. Sensitivity Analysis for RIF Definition

In order to evaluate the impact of different definitions of RIF on the stability of primary outcomes, this study excluded 4 studies that included patients with less than 3 previous implant failures and performed sensitivity analysis. After exclusion, the results were still robust and statistically significant (RR = 2.34, 95% CI: 1.90–2.87, < 0.00001), and there was no heterogeneity (I ² = 0%). The above results confirm that the difference in the definition of RIF does not affect the robustness of the main conclusions.

3.5.1. Chemical Pregnancy Rates

All 10 studies compared the chemical pregnancy rate between the two groups. The PRP group demonstrated a significantly higher chemical pregnancy rate than the control group (237/518 vs. 125/521; RR = 1.91, 95% CI: 1.60–2.28, p < 0.00001). No heterogeneity was detected (I ² = 0%, χ² = 2.56, df = 7, p = 0.92), indicating consistency in the treatment effect across the included studies (Figure 4).

3.5.2. Implantation Rates

Three studies compared the IR between the PRP group and the control group. The results showed that PRP improved the IR (RR = 1.93, 95% CI: 1.41–2.64, p < 0.0001), with significant heterogeneity observed among the studies (I ² = 68%, χ² = 6.33, df = 2, p = 0.04) (Figure 5).

3.5.3. Ongoing Pregnancy Rates

Four studies compared the OPR between the two groups. PRP significantly increased the OPR (RR = 1.74, 95% CI: 1.15–2.63, p = 0.008), with moderate heterogeneity detected (I ² = 45%, χ² = 5.46, df = 3, p = 0.14) (Figure 6).

3.5.4. Live Birth Rates

Five studies compared the LBR between the PRP group and the control group. PRP significantly improved the LBR (RR = 3.74, 95% CI: 2.68–5.22, p < 0.00001), with significant heterogeneity observed (I ² = 79%, χ² = 19.39, df = 4, p = 0.0007) (Figure 7).

3.5.5. Sensitivity Analysis for Live Birth Rate

Given the substantial heterogeneity (I ² = 79%) among the five included studies reporting live birth rate (LBR), we performed a leave‐one‐out sensitivity analysis to evaluate the robustness of the combined effect. Exclusion of any single study did not alter the statistical significance of the result. All the combined RRs values were > 1.0, and their 95% CIs did not cross 1.0. It is worth noting that the study of Nazari et al. (2022) [13] was identified as the main source of heterogeneity. After excluding this study, the pooled RR decreased from 3.74 to 2.32 (95% CI: 1.55–3.49, p < 0.00001), and the I ² value decreased from 79% to 60%. These findings indicate that the positive effect of PRP on LBR is robust, although the magnitude of the effect estimate is sensitive to the inclusion of this large study. Due to the limited number of studies, formal subgroup analyses could not be performed, but the sensitivity analysis suggests that between‐study variability, particularly driven by the Nazari et al. study, contributes to the high initial heterogeneity.

3.5.6. Miscarriage Rates

Six studies compared the MR between the two groups. There was no significant difference in the MR between the PRP group and the control group (RR = 0.66, 95% CI: 0.34–1.28, p = 0.22), indicating a favorable safety profile of PRP (Figure 8).

3.5.7. Subgroup Analysis Results

Subgroup analyses were performed for exploratory purposes, as the tests for subgroup differences did not reach statistical significance.

For PRP volume (Figure 9): both 0.5 mL (RR = 2.33, 95% CI: 1.87–2.90) and 0.5–1.0 mL (RR = 1.88, 95% CI: 1.29–2.76) significantly improved clinical pregnancy rate compared with control, with no statistically significant difference between the two dose subgroups (P for interaction = 0.34, I ² = 0%).

For embryo transfer stage (Figure 10): PRP infusion appeared to be associated with a numerically larger effect in cycles with blastocyst transfer (RR = 2.37, 95% CI: 1.93–2.91) than in those with cleavage‐stage embryo transfer (RR = 1.67, 95% CI: 1.03–2.71). However, the subgroup difference test did not reach statistical significance (interaction = 0.16, I 2 = 50%), suggesting that the observed differences may be caused by accidental factors. Therefore, these results should be regarded as the basis for making assumptions, rather than conclusive conclusions.

3.5.8. Subgroup Analysis by RIF Definition (Number of Prior Implantation Failures)

To explore the potential impact of varying RIF definitions across studies, we performed a subgroup analysis stratifying studies by the threshold of prior embryo transfer failures: < 3 failures (4 studies, n = 295) versus ≥ 3 failures (6 studies, n = 893). As shown in Figure S1, PRP infusion significantly increased the clinical pregnancy rate in both subgroups. In the subgroup that defined RIF as the number of previous failures < 3, the combined RR value was 1.56 (95% CI: 0.98–2.48, p = 0.06), and there was no heterogeneity (I ² = 0%). In the subgroup that defined RIF as the number of previous failures ≥ 3, the combined RR value was 2.34 (95% CI: 1.90–2.87, p < 0.00001), and there was no heterogeneity (I ² = 0%). The test for subgroup differences was not statistically significant (χ² = 3.52, df = 1, p = 0.06, I ² = 71.6%), suggesting that while the effect size appeared larger in the ≥ 3 failures group, the difference did not reach formal significance. Notably, both subgroups demonstrated consistent benefits of PRP without heterogeneity.

3.5.9. Sensitivity Analysis Results

Given that the study by Yahyaei et al. (2024) [8] accounted for 35.6% of the weight in the initial pooling of miscarriage rates, we excluded this study and re‐pooled the remaining 5 studies. The results showed that the pooled RR of the miscarriage rate changed from 0.66 (95% CI: 0.34–1.28) to 0.97 (95% CI: 0.44–2.11), and the p value increased from 0.22 to 0.93. The effect tended to be neutral, and the heterogeneity decreased from 37% to 9% (Figure 11). These findings suggest that the conclusion that “PRP has no effect on the MR” is robust and not driven by a single study.

3.5.10. Publication Bias

A funnel plot was applied to qualitatively evaluate the publication bias. The funnel plot presented in Figure 12 is symmetrical, indicating that the likelihood of publication bias is relatively low.

FIGURE 12.

Funnel plot of publication bias analysis.

4. Discussion

This study shows that for patients with RIF, intrauterine infusion of PRP may significantly increase the clinical pregnancy rate (CPR), live birth rate (LBR), biochemical pregnancy rate, implantation rate (IR), and ongoing pregnancy rate (OPR), and does not increase the risk of abortion. However, in view of the significant heterogeneity of the main outcome indicator LBR (I ² = 79%), the findings related to LBR should be interpreted with caution. Exploratory subgroup analysis suggested that PRP infusion may bring clinical benefits regardless of PRP dose or embryo transfer stage. Among them, low‐dose PRP (0.5 mL) seems to be more effective, and the benefit is more obvious in the blastocyst transfer group. However, these potential mechanisms are mainly based on previous basic research, which has not been directly verified in this meta‐analysis. The potential improvement of clinical outcomes in RIF patients may be associated with the following mechanisms, as illustrated in Figure S2.

4.1. PRP Promotes Endometrial Angiogenesis and Improves Endometrial Blood Flow

PRP is an autologous plasma product obtained by centrifugation that contains platelet concentrations far exceeding physiological levels. It is rich in growth factors, including platelet‐derived growth factor (PDGF), transforming growth factor‐β (TGF‐β), vascular endothelial growth factor (VEGF), insulin‐like growth factor (IGF), as well as various cytokines such as interleukin (IL)‐4, IL‐8, IL‐13, IL‐17, tumor necrosis factor‐α (TNF‐α), and interferon‐α. These growth factors and cytokines can promote endometrial angiogenesis and stromal cell proliferation, thereby increasing endometrial thickness, improving endometrial blood supply, and supporting embryo implantation and subsequent pregnancy maintenance. In vitro experiments have found that PRP significantly promotes the linear growth of endometrial glands and blood vessels [17]. An RCT involving patients with thin endometrium found that intrauterine PRP infusion not only increased endometrial thickness but also significantly improved the CPR [18]. Similarly, a retrospective cohort study reported that PRP improved endometrial thickness and uterine hemodynamics in patients with RIF and thin endometrium, leading to a significant improvement in pregnancy outcomes [19]. In contrast, in RIF patients with normal endometrial thickness, PRP does not appear to further increase endometrial thickness. A retrospective study observed that in patients with non‐thin endometrium, the CPR, embryo IR, and LBR in the PRP group were significantly higher than those in the non‐PRP group (p < 0.05). However, there was no difference in endometrial thickness between the two groups, suggesting that PRP may improve clinical outcomes by enhancing endometrial secretory function during the secretory phase [5]. We speculate that in addition to promoting endometrial growth, improvements in endometrial blood flow are also contributing to the beneficial effects of PRP on pregnancy outcomes. Although many studies have reported that PRP increases endometrial thickness and improves clinical outcomes in patients with thin endometrium, its effectiveness across different severities of thin endometrium remains unclear. First, the impact of endometrial thickness severity on clinical outcomes is still debated. Studies have found that the LBR is significantly reduced in patients with thin endometrium, with a more pronounced decline when endometrial thickness is < 5 mm [20]. Second, the therapeutic response to PRP may vary with the severity of thin endometrium. A self‐controlled case study across menstrual cycles found that although PRP treatment significantly increased endometrial thickness in most patients with thin endometrium, some still required cycle cancellation due to insufficient endometrial response [21]. It is speculated that patients with more severe thin endometrium, intrauterine adhesions, or other complications may not benefit sufficiently from PRP. Therefore, it is necessary to conduct high‐quality RCTs to explore the role of PRP in patients with varying degrees of thin endometrium, so as to identify the suitable population and implement stratified management.

4.2. PRP Improves the Endometrial Immune Microenvironment in RIF Patients

Immune cells within the female reproductive tract, including T cells, B cells, natural killer (NK) cells, and macrophages (MQ), play a critical role in defending against pathogens and maintaining tissue homeostasis. Proper immune regulation is essential for establishing endometrial receptivity, as these cells enhance epithelial adhesion, regulate decidual cell differentiation, remodel uterine blood vessels, control and resolve inflammation, and inhibit destructive immune responses against paternal allogeneic antigens. Dysregulation may trigger inflammation, thereby impairing fertility. One of the most common immune disorders observed in RIF is the alteration of the Th1/Th2 ratio, as well as changes in the number or function of NK cells and MQ [22, 23]. A study found that women with RIF exhibited a significantly increased Th17/Treg ratio, a reduced absolute number of endometrial Treg cells, and a higher RORγt/FOXP3 ratio compared with fertile women. Meanwhile, the levels of pro‐inflammatory factors such as TNF‐α, IL‐6, IL‐8, and CCL2 were also significantly higher than those in healthy fertile women [24]. After intrauterine PRP injection, decreased cell proliferation and attenuated inflammatory immune responses were observed during endometrial decidualization, a finding confirmed by other studies. For instance, PRP treatment significantly decreased levels of endometrial NK cells, CD8+ T cells, and Th1 cells, while endometrial receptivity‐related gene expression patterns became more similar to those of fertile controls, indicating improved endometrial receptivity [25]. PRP treatment has shown promising results in improving the endometrial immune environment in RIF patients. The uterine immune microenvironment differs from systemic immunity and has unique characteristics. Intrauterine PRP infusion only exerts local intervention in the uterus, which may avoid the risks associated with systemic immunomodulation, representing a clinical advantage. However, evaluating the endometrial immune status before and after PRP treatment remains challenging due to the need for endometrial biopsy and inconsistent detection standards, which bring difficulties to its clinical application. If the immunomodulatory effect of PRP can be indirectly reflected through more convenient peripheral blood immune indicators or intrauterine fluid cytokine detection, invasive operations on patients can be reduced and the convenience of clinical diagnosis and treatment can be improved.

4.3. PRP Enhances Endometrial Receptivity

Studies have confirmed that PRP can significantly upregulate the expression of key endometrial receptivity markers, such as integrin αvβ3 and leukemia inhibitory factor (LIF). A self‐controlled case series study found that the expression of endometrial receptivity markers HOXA‐10, Ki67, and αvβ3 was increased after intrauterine PRP infusion [21]. Another study showed that PRP treatment upregulated the expression of LIF, COX2, and p53 genes, as well as estrogen and progesterone receptors in endometrial cells of RIF patients, enhancing the endometrial autocrine/paracrine mechanism and improving endometrial receptivity [26]. A study showed that in patients with persistent chronic endometritis (CE), intrauterine PRP infusion significantly increased the expression of multiple receptivity‐related genes, including CD36, DUOX1, DUOX2, CLDN1, CLDN3, HPSE, KLF5, MET, TLR4, ARG2, LGALS9C, and DEFB1 [27]. PRP improves endometrial receptivity, thereby enhancing the endometrium's capacity to support embryo implantation. Our meta‐analysis further confirmed that PRP significantly improved clinical outcomes in RIF patients. In addition to enhancing endometrial receptivity, an intriguing question arises: could PRP also modulate the implantation window in patients with potentially shifted or unsynchronized windows? Currently, most relevant studies are observational or small‐scale, and the direct causal link between receptivity markers and live birth outcomes has not been fully confirmed. In the future, endometrial organoid models or single‐cell sequencing technology can be adopted to dynamically observe the precise regulation of PRP on the expression of receptivity‐related genes in different endometrial subsets at a more bionic and high‐resolution level.

4.4. Anti‐Inflammatory Effect of PRP

Chronic endometritis (CE) is a chronic inflammatory disease of the endometrium, characterized by abnormal infiltration of stromal cells. The main causes of asymptomatic CE are common intrauterine bacteria (, Streptococcus, and Staphylococcus) and/or Mycoplasma/Ureaplasma infections. Oral antibiotics are the standard treatment and achieve a high cure rate [28]. CE prevalence among RIF patients ranges from 7.7% to 66.3% [29, 30, 31]. Changes in the distribution and function of immune cells during CE can lead to abnormal local immune responses, affect the endometrial immune microenvironment, and disturb maternal‐fetal immune balance, thereby interfering with endometrial receptivity, embryo implantation, and development [32]. A study reported that the CPR of RIF patients complicated with CE was significantly lower than that of RIF patients without CE (20.0% vs. 46.9%, p = 0.04) [33]. Effective CE treatment focuses on eradicating endometrial inflammation (mainly IL‐1, TNF‐α, etc.). Platelet degranulation promotes the release of growth factors such as PDGF and IGF, which facilitate tissue repair, angiogenesis, and cell differentiation. Additionally, platelets contain chemokines, mucins, and antifibrinolytic proteins, which are widely used in soft tissue repair and infected wound treatment, exerting direct antibacterial effects. Therefore, PRP has been applied clinically in RIF patients with persistent CE positivity. In one study, patients receiving oral antibiotics combined with intrauterine PRP infusion showed statistically significant increases in CPR, LBR, and hCG‐positive rate [34]. Another study reported favorable outcomes with intrauterine PRP infusion alone, administered 3–4 times during the endometrial proliferative phase in patients with persistent CE. The implantation rate and clinical pregnancy rate in the cured group were significantly higher than those in the non‐cured group. Moreover, after PRP treatment, the expression of genes related to endometrial receptivity and antibacterial genes was upregulated, while genes involved in immune response processes were downregulated in cured patients. Meanwhile, the proportions of endometrial CD8+ T cells, CD56+ NK cells, Foxp3+ Treg cells, and T‐bet+ Th1 cells were significantly reduced [27]. These findings suggest that PRP may mitigate the adverse inflammatory environment caused by CE through its potent anti‐inflammatory and tissue repair capabilities, providing a novel solution for these refractory patients. PRP has been explored in patients with persistent CE, but current evidence is limited to small observational studies. PRP is not yet a standard treatment for CE, and further RCTs are needed. Meanwhile, given the high incidence of CE in RIF patients and its adverse impact on clinical outcomes, we cautiously recommend that RIF patients preparing for another embryo transfer, especially those with a history of CE or clinical signs suggesting possible endometrial inflammation, undergo endometrial biopsy to confirm the diagnosis and provide more precise indications for PRP application.

Previous meta‐analyses have confirmed that intrauterine PRP infusion can improve clinical outcomes in patients with RIF, but the strength of their evidence has been limited by substantial heterogeneity, such as failure to exclude patients with thin endometrium and the inclusion of mixed study designs [35, 36]. A recent study that only included high‐quality RCTs and excluded patients with thin endometrium reached consistent conclusions with this study, namely that PRP can effectively improve CPR and LBR without increasing the risk of miscarriage, thereby strengthening the evidence base [37]. On this basis, the present study further expands the relevant evidence in several ways. First, by including the latest RCTs and performing predefined sensitivity analysis, we found that after excluding the study with the highest weight, the pooled effect size of the miscarriage rate approached 1 (RR ≈ 1.0), and the heterogeneity decreased significantly. This result not only reinforces the conclusion that PRP does not increase the risk of miscarriage but also more clearly supports its neutral safety profile, providing a more reliable basis for clinical decision‐making. Second, subgroup analyses were performed based on clinically relevant factors such as PRP dose and embryo transfer stage. In terms of dose, both 0.5 mL and 0.5–1.0 mL PRP can significantly improve the CPR, with a more favorable trend observed in the lower‐dose group. This may reflect the presence of an optimal concentration window for PRP, as excessively high concentrations of growth factors and proteins may cause excessive cell proliferation or improper cell signal transduction, thereby interfering with normal tissue repair [38, 39]. In terms of the embryo transfer stage, a higher effect size was observed in the blastocyst transfer group, although the interaction effect did not reach statistical significance. This finding suggests that PRP‐optimized endometrium may act synergistically with blastocysts of higher developmental potential, consistent with the theory of embryo–endometrium synchronization [40, 41, 42, 43].

In conclusion, through enhancing evidence quality, verifying safety, and exploring clinical application conditions, this study provides more detailed references for the precise application of PRP in RIF patients.

4.5. Heterogeneity, Limitations, and Future Directions

The main limitation of this meta‐analysis is the substantial heterogeneity in the primary outcome of live birth rate (I ² = 79%); sensitivity analysis identified the study by Nazari et al. (2022) [13] as a major contributor, with I ² decreasing to 60% after its exclusion. Potential sources of heterogeneity may include differences in PRP preparation protocols, RIF definitions, patient baseline characteristics, and embryo transfer procedures between studies. In view of these sources of heterogeneity, and although the combined effect in the sensitivity analysis is still statistically significant, its effect size still needs to be interpreted cautiously, so the results of this study should be viewed with caution. Specifically, significant heterogeneity represents the main challenge, which may originate from the following sources: differences in baseline patient characteristics, such as RIF definitions, age, body mass index (BMI), embryo quality, and the number of transferred embryos, which directly affect pregnancy outcomes but failed to be unified or stratified in most included studies. Substantial variations in PRP preparation and application protocols, which constitute the primary source of heterogeneity, as different centrifugation protocols result in different platelet concentrations and growth factor components; and diverse infusion doses, frequencies, and timing, which complicate the direct comparison of “PRP effects” across different studies, and it is precisely to address this core issue that the dose subgroup analysis in this study was conducted.

Our subgroup analysis was stratified according to the threshold of previous implantation failure (< 3 vs. ≥ 3). The results showed that PRP could significantly improve the clinical pregnancy rate in both subgroups, and the effect size of ≥ 3 times failure group was higher (RR = 3.65 vs. 1.92). Although the subgroup interaction was not statistically significant (p = 0.06), this finding suggests that patients with more severe RIF (≥ 3 failures) may have greater absolute benefits from PRP, which may be related to their more significant endometrial dysfunction. However, this observation should be interpreted as exploratory given the limited number of studies in the < 3 failures subgroup (n = 4) and the lack of standardized RIF definitions across trials.

This study has several limitations. First, the overall methodological quality of the included RCTs was moderate, and the geographic distribution of studies was limited to Iran, which restricts the generalizability of our findings to other populations with different ethnic, socioeconomic, or healthcare system backgrounds. Second, as summarized in Table S1, the PRP preparation protocols across the 10 included RCTs exhibited considerable heterogeneity. Although most studies reported a target platelet enrichment factor of 4 to 7 times that of the baseline, no absolute platelet count or white blood cell concentration was provided, and only two studies clearly indicated that PRP was not activated before perfusion. Centrifugal parameters (centrifugal force, time, number of spins) are also quite different. This lack of standardization limits the comparability of research results and also hinders the determination of the best PRP for the treatment of RIF. Third, the pooled live birth rate (LBR) showed substantial heterogeneity (I ² = 79%), which likely reflects variations in patient baseline characteristics (e.g., age, embryo quality, RIF definition) as well as differences in PRP preparation and infusion protocols across studies. This heterogeneity suggests that caution should be exercised when interpreting the results of LBR. Future studies should adopt a standardized definition of outcomes and report the results of stratified analysis to reduce heterogeneity.

Based on the above limitations, future research should focus on the following aspects: (1) conducting large‐scale, international multicenter high‐quality RCTs to obtain high‐level evidence; (2) establishing international consensus and standards for PRP preparation and clinical application; (3) identifying biomarkers that can predict the therapeutic efficacy of PRP using omics technologies to achieve precise classification and individualized treatment of RIF; (4) designing well‐conducted studies to explore the suitable population and optimal application methods of PRP.

5. Conclusion

Intrauterine PRP infusion may improve pregnancy outcomes in RIF patients potentially via multiple biological pathways, such as endometrial repair, modulation of the uterine immune microenvironment, improvement of endometrial receptivity, and anti‐inflammatory properties. Our study further suggests that low‐dose PRP may be sufficiently effective, and its use in blastocyst transfer cycles should be prioritized when conditions permit.

Funding

National Natural Science Foundation of China (NO. 82571880).

Ethics Statement

This review article included no patients, and thus, it did not require approval from Ethics Committee.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Figure S1: Subgroup analysis of clinical pregnancy rate according to the number of previous embryo transfer failures (< 3 times vs. ≥ 3 times) in women with recurrent implantation failure (RIF).

Forest plot of subgroup analyses comparing the clinical pregnancy rate between intrauterine PRP infusion and control groups, stratified by the number of prior embryo transfer failures (< 3 vs. ≥ 3). Relative risks (RRs) and 95% confidence intervals (CIs) were calculated using a random‐effects model. Between‐study heterogeneity was evaluated using the I ² statistic.

Figure S2: Proposed mechanisms by which intrauterine autologous platelet‐rich plasma (PRP) infusion improves endometrial function and reproductive outcomes in patients with recurrent implantation failure (RIF).

Schematic illustration summarizing four putative core pathways that potentially underlie the beneficial effects of PRP: promotion of endometrial angiogenesis and perfusion, improvement of endometrial receptivity, modulation of the endometrial immune microenvironment, and anti‐inflammatory and reparative effects against chronic endometritis‐related injury. Key growth factors (PDGF, VEGF, TGF‐β, IGF) secreted by PRP may mediate these downstream biological effects, potentially optimizing the endometrial microenvironment and enhancing embryo implantation potential.

Table S1: Summary of PRP Preparation and Administration Characteristics in Included Studies.

Data Availability Statement

The data that support the findings of this study are openly available within the paper.