Expert transvaginal ultrasound is determinant for diagnosing pelvic conditions after recurrent implantation failure in IVF

AS: ultrasonographic scan, design and conceptualized study and interpreted data; CR: ultrasonographic scan and revised the manuscript for intellectual content; DF: revised the statistical analysis; GM, EN, SV and FI: revised the manuscript for intellectual content; CE: ultrasonographic scan, revised the manuscript for intellectual content, conceptualized study and interpreted data. All the authors discussed the results and commented on the manuscript.

Aikaterini Selntigia: Writing – original draft, Validation, Investigation, Formal analysis, Conceptualization. Consuelo Russo: Supervision, Methodology, Investigation. Daniele Farsetti: Project administration, Methodology, Investigation, Formal analysis. Giulia Monaco: Visualization, Data curation. Elvira Nocita: Visualization, Data curation. Sara Valeriani: Visualization, Data curation. Federica Iacobini: Visualization, Data curation. Caterina Exacoustos: Writing – review & editing, Writing – original draft, Supervision, Software, Resources, Methodology, Investigation, Conceptualization.

This research did not receive any specific grant for funding agencies in the public, commercial, or not-for-profit sectors.

A total of 280 infertile patients with a history of RIF were assessed at our outpatient clinic between June 2020 and September 2024; among them, 152 fulfilled the predefined inclusion criteria ( Fig. 1 ). Table 1 summarizes the general characteristics of the study population and the medical data concerning the type of IVF treatment and the number of implantation failures. The average age of the included patients was 40.4 ± 5.3 years (range 31–45), with a mean BMI of 22.3 ± 3.4 kg/m². None of the included patients had a live birth pregnancy (parity=0), 9 women (5.9 %) in our study population had a previous spontaneous pregnancy with an early pregnancy loss, six (3.9 %) of which had a voluntary termination of pregnancy. Furhermore, 49 (32 %) of the included population had a previous pelvic surgery. Of the included patients, 98/152 (64.0 %) performed PGT-A with own oocyte, 26/152 (17.0 %) received oocyte donations. 28 patients (18.0 %) underwent both procedures, firstly experienced IF with euploid blastocyst following ovarian stimulation with their own oocytes and subsequent oocyte donation. The average number of implantation failures was 3.0 ± 1.3 per patient. Fig. 1 Figure illustrates the study time frame, inclusion and exclusion criteria. A total of 152 patients with Recurrent Implantation Failure (RIF) were included in the analysis. TVS= transvaginal ultrasound. IVF: In vitro fertilization. Fig. 1 Table 1 Demographic and infertility history data of the total population study. Table 1 DEMOGRAPHIC AND INFERTILITY CHARACTERISTICS OF THE TOTAL POPULATION (N = 152) Age (years) ( mean ± SD ) 40.4 ± 5.3 BMI ( mean ± SD ) 22.3 ± 3.4 Menarche ( mean ± SD ) 12.3 ± 2.1 Parity n (%) 0 (0.0) Primary infertility n (%) 143 (94.1) Secondary infertility n (%) 9 (5.9) Previous pelvic surgery n (%) 49 (32.0) IVF embryo’s characteristics PGT-A n (%) 98 (64.0) Oocyte donation n (%) 26 (17.0) PGT-A of their own oocyte followed then by oocyte donation n (%) 28 (18.0) Implantation Failures (IF) ( mean ± SD) 3.0 ± 1.3 2 IF n (%) 69 (45.0) 3 IF n (%) 51 (34.0) 4 IF n (%) 14 (9.0) 5 IF n (%) 10 (6.5) > 5 IF n (%) 8 (5.0) BMI: Body Mass Index; IF: implantation failure; IVF: in vitro fertilization; PGT-A: Preimplantation genetic testing for aneuploidies. Figure illustrates the study time frame, inclusion and exclusion criteria. A total of 152 patients with Recurrent Implantation Failure (RIF) were included in the analysis. TVS= transvaginal ultrasound. IVF: In vitro fertilization. Demographic and infertility history data of the total population study. BMI: Body Mass Index; IF: implantation failure; IVF: in vitro fertilization; PGT-A: Preimplantation genetic testing for aneuploidies. The diagnosis of RIF was established by the IVF center, which referred 76 % of these patients to our facility. The remaining cases were referred either by general practitioners or through self-referral, and the doctors who performed the TVS in our unit were aware before the scan of the all the history of the patient. The ultrasound findings performed in the IVF center were compared to the Expert TVS findings in our unit. We divided the study population into two groups: patients with a previously identified pelvic pathology detected via TVS at the IVF center (45/152, 29.6 %) and those with normal findings (107/152, 70.4 %). Among the 45 patients with a known pelvic pathology on TVS performed at the IVF center, 26 (57.8 %) had their initial diagnosis confirmed during our ultrasound examination, while 19 (42.2 %) presented additional pathological findings. In contrast, among the 107 patients initially classified as normal, our Expert TVS scan detected pelvic pathology in 84 (78.5 %) cases, with only 23 (21.5 %) maintaining a normal diagnosis. ( Fig. 1 ) Among the 152 patients, Expert TVS detected pelvic pathology in 129 cases, compared to 45 identified in the initial IVF center TVS (p < 0.0001). A significant difference was also observed in the proportion of normal findings before and after Expert TVS (70.4 % vs 21.5 %, p < 0.0001). The comparison between pelvic pathologies identified by Expert TVS and those detected in IVF centres, including OA, Cohen’s K and PPA between the two TVS examinations, is summarized in Table 2 . Significant differences were observed for all pathological ultrasound findings between the initial IVF center scan and the ES TVS, except for isthmocele. Table 2 Ultrasonographic findings between IVF center TVS and Expert TVS. Table 2 PELVIC SITES TVS scan at IVF centres n (%) Expert TVS scan after RIF n (%) Overall agreement (95 % Cl) Cohen’s Kappa (95 % Cl) (95 % Cl) Positive Percent Agreement (95 % Cl) (95 % Cl) p value UTERUS 32 (21.1) 108 (71.1) 0.500 (0.418 – 0.582) 0.196 0.120–0.272) 29.6 % (21.2 – 39.2 %) < 0.01 Adenomyosis 13 (8.6) 83 (54.6) 0.539 0.457 – 0.620) 0.144 15.7 % < 0.01 Myomas 27 (17.8) 34 (22.4) 0.954 0.857 0.754–0.959) 79.4 % (62.1 – 91.3) 0.02 Submucous 7 (4.6) 8 (5.3) 0.974 (0.934–0.993) 0.655 (0.341–0.968) 50.0 % (15.7 – 84.3) 0.13 Intramural 25 (16.4) 27 (17.8) 0.974 (0.934–0.993) 0.904 (0.812–0.996) 85.2 % (66.3 – 95.8) 0.13 Subserous 14 (9.2) 16 (10.5) 0.98 (0.943 – 0.996) 0.886 (0.759–1.000) 81.2 % (54.3 – 95.9) 0.25 Uterine malformation 1 (0.7) 24 (15.8) 0.849 (0.782 – 0.902) 0.068 (-0.059–0.196) 4.2 % (0.1–21.1) < 0.01 Dysmorphic 1 (0.7) 18 (11.8) 0.888 (0.827 – 0.933) 0.094 (-0.078–0.266) 5.6 % (0.1– 27.3) < 0.01 Partial septum 0 (0.0) 6 (3.9) 0.960 (0.916 – 0.985) 0.000 (0.000 – 0.000) 0.0 % (0.0–45.9) 0.04 Isthmocele 0 (0.0) 1 (0.7) 0.993 (0.964 – 1.000) 0.000 (0.000 – 0.000) 0.0 % (0.0 – 0.9) 1 ADNEXA 16 (10.5) 43 (28.3) 0.822 (0.752 – 0.880) 0.459 (0.304–0.615) 37.2 % (23.0 – 53.3) < 0.01 Endometriomas 16 (10.5) 36 (23.7) 0.868 (0.804 – 0.918) 0.550 (0.386–0.714) 44.4 % (27.9 – 61.9) < 0.01 Hydrosalpinx 0 (0.0) 13 (8.6) 0.914 (0.858 – 0.954) 0.000 (0.000 – 0.000) 0.0 % (0.0 – 24.7) < 0.01 DIE 10 (6.6) 59 (38.8) 0.678 (0.597 – 0.751) 0.200 (0.089–0.310) 16.9 % (8.4 – 29.0) < 0.01 Retrocervical (#ENZIAN A) 4 (2.6) 24 (15.8) 0.868 (0.804 – 0.918) 0.252 (0.049 – 0.458) 16.7 % (4.7 – 37.4) < 0.01 USL (#ENZIAN B) 4 (2.6) 50 (32.9) 0.697 (0.618 – 0.769) 0.104 (0.008 – 0.201) 8.0 % (2.2–19.2) < 0.01 Rectum (#ENZIAN C) 3 (2.0) 28 (18.4) 0.835 (0.767 – 0.891) 0.163 (-0.001 – 0.328) 10.7 % (2.3–28.2) < 0.01 Cohen’s K value is used to evaluate the agreement rate between classifications: K < 0 no agreement; K between 0 and 0.4: poor agreement; K between 0.4 and 0.6: fair agreement; K between 0.6 and 0.8: good agreement; K between 0.8 and 1: excellent agreement. DIE: deep infiltrating endometriosis; CI =confidence interval; IVF: in vitro fertilization; RIF: recurrent implantation failure; TVS: transvaginal; USL: utero-sacral ligaments. A p value < 0.05 was considered statistically significant. Ultrasonographic findings between IVF center TVS and Expert TVS. Cohen’s K value is used to evaluate the agreement rate between classifications: K < 0 no agreement; K between 0 and 0.4: poor agreement; K between 0.4 and 0.6: fair agreement; K between 0.6 and 0.8: good agreement; K between 0.8 and 1: excellent agreement. DIE: deep infiltrating endometriosis; CI =confidence interval; IVF: in vitro fertilization; RIF: recurrent implantation failure; TVS: transvaginal; USL: utero-sacral ligaments. A p value < 0.05 was considered statistically significant. OA between the two scans was low (OA: Uterus 0.5; Adnexa 0.822; DIE 0.678). Notably, adenomyosis had the lowest OA (0.539) ( Fig. 2 ), whereas myomas showed the highest agreement (OA: 0.954). Cohen’s K and PPA also indicated poor concordance rate for uterine pathology (K=0.196; PPA=29.6 %), adnexal pathology (K=0.459, PPA=37.2 %) and DIE (K=0.2, PPA=16.9 %). After Expert TVS, the detection of uterine pathologies increased significantly from 21.1 % to 71.1 % (p < 0.01). Similarly, adnexal pathologies, including ovarian and fallopian tube conditions, rose from 10.5 % to 28.3 % (p < 0.01), with a new diagnosis of hydrosalpinx in 13 patients (8.6 %). DIE detection also increased from 6.6 % to 38.8 % (p < 0.01), highlighting a significant underdiagnosis in initial IVF center scans. Specifically, Expert TVS identified 28 rectal nodules (#ENZIAN C), 24 cases of retrocervical/vaginal nodules (#ENZIAN A) and 50 cases of endometriosis of utero-sacral ligaments (#ENZIAN B), with the lowest agreement observed for #ENZIAN C (K=0.163, PPA=10.7 %). ( Fig. 3 , Fig. 4 ) Fig. 2 Two-dimensional (a,c) and three-dimensional (b,d) coronal section of a uterus with direct signs of adenomyosis and Y-shaped cavity (dysmorphic uterus, class U1c for the ESHRE/ESGE classification). Two-dimensional (a) and three-dimensional (b) longitudinal section of a uterus with hyperechoic myometrial islands (white arrows) on the posterior fundal wall, identified at Expert transvaginal ultrasound (TVS). Regarding Y-shaped cavity, note the tubular cavity (c,d): Rule of 10 of 8.4 mm [8] , lateral bulging of 8.5 mm on the left [4] and 6.1 mm on the right [3] , and fundal indentation of 7.9 mm [6] . Fig. 2 Fig. 3 Transvaginal ultrasound appearance of different localization of endometriosis identified at Expert ultrasound evaluation (TVS). A) Longitudinal section of a uterus showing adenomyosis in the posterior wall (white asterisk), with the right ovary (yellow dotted line) adherent posteriorly to the torus and the utero-sacral-ligament (USL) due to endometriotic fibrosis (white arrows). B) Longitudinal section of the right ovary with a small endometrioma of 21 mm [1] and a follicle of 14 mm. C) Longitudinal section of the right ovary adherent to the deep infiltrating endometriosis (DIE) of the right USL (yellow dotted lines). D) DIE of the rectal wall (yellow dotted lines) and of the rectovaginal septum (RVS) (white arrows). Fig. 3 Fig. 4 Transvaginal ultrasound appearance of adenomyosis and associated deep infiltrating endometriosis (DIE) identified at Expert evaluation. A-B) Longitudinal section of a uterus showing adenomyosis in the posterior wall (white arrows) with associated DIE of the rectal wall (yellow dotted lines). Fig. 4 Two-dimensional (a,c) and three-dimensional (b,d) coronal section of a uterus with direct signs of adenomyosis and Y-shaped cavity (dysmorphic uterus, class U1c for the ESHRE/ESGE classification). Two-dimensional (a) and three-dimensional (b) longitudinal section of a uterus with hyperechoic myometrial islands (white arrows) on the posterior fundal wall, identified at Expert transvaginal ultrasound (TVS). Regarding Y-shaped cavity, note the tubular cavity (c,d): Rule of 10 of 8.4 mm [8] , lateral bulging of 8.5 mm on the left [4] and 6.1 mm on the right [3] , and fundal indentation of 7.9 mm [6] . Transvaginal ultrasound appearance of different localization of endometriosis identified at Expert ultrasound evaluation (TVS). A) Longitudinal section of a uterus showing adenomyosis in the posterior wall (white asterisk), with the right ovary (yellow dotted line) adherent posteriorly to the torus and the utero-sacral-ligament (USL) due to endometriotic fibrosis (white arrows). B) Longitudinal section of the right ovary with a small endometrioma of 21 mm [1] and a follicle of 14 mm. C) Longitudinal section of the right ovary adherent to the deep infiltrating endometriosis (DIE) of the right USL (yellow dotted lines). D) DIE of the rectal wall (yellow dotted lines) and of the rectovaginal septum (RVS) (white arrows). Transvaginal ultrasound appearance of adenomyosis and associated deep infiltrating endometriosis (DIE) identified at Expert evaluation. A-B) Longitudinal section of a uterus showing adenomyosis in the posterior wall (white arrows) with associated DIE of the rectal wall (yellow dotted lines). A statistically significant misdiagnosis of adenomyosis was noted between the IVF center and ES TVS evaluations (8.6 % vs. 53.9 %, p < 0.0001). A significant underdiagnosis was also observed for uterine malformations (0.7 % vs 15.8 %, K=0.068, PPA=4.2 %, p < 0.01). Dysmorphic uteri-particularly T-shaped (12 cases) and Y-shaped (6 cases) were frequently overlooked in IVF center scans (0.7 % vs. 11.8 %, K=0.094, PPA=5.6 %, p < 0.01) ( Fig. 2 ). Similarly, partial septate uterus was underdiagnosed (0 % vs. 3.9 %, K=0, PPA=0 %, p = 0.04).

This retrospective study included patients after RIF during IVF process. All included patients were admitted to the Gynecological Ultrasound Unit of the University of Rome 'Tor Vergata' between 2020 and 2024 for a second-opinion ultrasound. The study was conducted in accordance with the following inclusion and exclusion criteria. Inclusion Criteria: • Age 30–45 years • Diagnosis of primary or secondary infertility • History of ≥ 2 implantation failures after transfer of blastocyst • Embryo Transfer of either: o Euploid blastocysts confirmed by PGT-A, and/or o Blastocysts from donor oocytes • TVS was performed by experienced specialists (Expert TVS) using 2D, 3D, and power Doppler (PD) to accurately assess the pelvic organs • Complete IVF center reports: o Embryo transfer modalities o Ultrasound scans o Infertility screening (genetic, endocrinological, metabolic, trombhilic, infection) • Signed informed consent Age 30–45 years Diagnosis of primary or secondary infertility History of ≥ 2 implantation failures after transfer of blastocyst Embryo Transfer of either: o Euploid blastocysts confirmed by PGT-A, and/or o Blastocysts from donor oocytes Euploid blastocysts confirmed by PGT-A, and/or Blastocysts from donor oocytes TVS was performed by experienced specialists (Expert TVS) using 2D, 3D, and power Doppler (PD) to accurately assess the pelvic organs Complete IVF center reports: o Embryo transfer modalities o Ultrasound scans o Infertility screening (genetic, endocrinological, metabolic, trombhilic, infection) Embryo transfer modalities Ultrasound scans Infertility screening (genetic, endocrinological, metabolic, trombhilic, infection) Signed informed consent Exclusion Criteria: • Lack of documentation regarding previous IVF procedures • Ongoing pregnancy at the time of evaluation • Incomplete or inconsistent clinical or anamnestic data • Untreated metabolic, endocrine, or hematological conditions • Inaccurate or insufficient TVS report from the referring IVF center • Embryo transfers involving: o Cleavage-stage embryos from own oocytes o Cleavage-stage embryos from donor oocytes o Unscreened embryos from own oocytes • Absence of signed informed consent Lack of documentation regarding previous IVF procedures Ongoing pregnancy at the time of evaluation Incomplete or inconsistent clinical or anamnestic data Untreated metabolic, endocrine, or hematological conditions Inaccurate or insufficient TVS report from the referring IVF center Embryo transfers involving: o Cleavage-stage embryos from own oocytes o Cleavage-stage embryos from donor oocytes o Unscreened embryos from own oocytes Cleavage-stage embryos from own oocytes Cleavage-stage embryos from donor oocytes Unscreened embryos from own oocytes Absence of signed informed consent Approval was granted by the Ethics Committee of University of Rome “Tor Vergata” (Date 28/11/2024, No 264.24 CET2). Informed consent, allowing the collection and use of anonymized personal data and TVS findings for research purposes, was obtained from all individual participants included in the study. Patient information was recorded according to a pre-established format using the File maker pro® software Version 9.0. Infertility was defined as the absence of pregnancy after 12 months of targeted intercourse [6] . The infertility evaluation reports generated during the IVF process were collected and thoroughly analyzed. The data related to the IVF process included the following: years of infertility, genetic analyses (karyotype, cystic fibrosis, hereditary thrombophilia), acquired thrombophilia, pathological hemoglobin chains, endocrinological hormone test (TSH, PRL, AMH, FSH, glycemic and insulinemic curve), vitamin D, infection and serological test and eventual therapy of the above-mentioned pathologies were recorded. Furthermore, all data relating to embryo quality were recorded, in terms of blastocysts quality and results of PGT-A [7] . Regarding embryo quality, the study excluded embryos at a premature developmental stage (cleavage stage) derived from donated oocytes, as well as those from the patient’s own oocytes that had not undergone PGT-A analysis. Before Expert TVS all previous ultrasound reports from the IVF center were reviewed, and any diagnosed morphological pathology of pelvic organs was recorded. All ultrasounds at the IVF center were performed by gynecologists specialized in assisted reproductive technologies. All patients underwent 2D, 3D and PD TVS examination in order to evaluate all possible pelvic pathologies. The ultrasound examination was performed by two experienced sonographers in gynecological field (C.E., A.S.) using a Voluson E6 or E8 device (GE Healthcare, Zipf, Austria) with a transvaginal probe and if necessary additional transabdominal probe. The scan was first performed with a conventional 2D ultrasound for the assessment of the pelvis. The uterus, endometrium and adnexa were evaluated. The 2D examination was followed by a 3D acquisition of volumes using the 3D volume mode. The 3D examination was performed to acquire the volume of the uterus in order to obtain a coronal section and evaluate cavity morphology and junction zone and myometrial morphology [8] . The TVS examination was performed at any phase of the menstrual cycle. Patients with suspected congenital malformations who were also reassessed during the secretory phase. Patients with newly identified suspected ovarian or endometrial lesions were reassessed during the subsequent early proliferative phase. Our TVS examination was performed, on average, three months after the initial ultrasound evaluation at the IVF center. In order to ensure diagnostic criteria that were precise, consistent, reproducible and general accepted we use standardized consensus protocols and both national and international guidelines. All possible myometrial pathologies were described according to the Morphological Uterus Sonographic Assessment (MUSA) guidelines [9] , [10] . Adenomyosis was diagnosed in the presence of at least one direct TVS sign. Uterine fibroid’s anatomical locations were classified according to the FIGO classification system [11] ; to describe endometrial pathology, the International endometrial tumor analysis (IETA) terminology was used [12] To assess ovarian pathologies, we used the International Ovarian Tumor analysis (IOTA) terminology [13] . Endometriosis was recorded following the International Deep Endometriosis Analysis (IDEA) [14] , previously published mapping systems [15] and the most recent #ENZIAN classification [16] . Furthermore, congenital uterine anomalies (CUAMs) were diagnosed using all the international classifications currently in use through the coronal scan obtained in 3D [17] , [18] . The presence and the US characteristics of isthmocele was described using the last published Delphi consensus [19] . All statistical analyses were carried out using Medcalc version 9.2.0.2 (Medcalc Software, Mariakerke, Belgium). The quantitative variables were expressed as mean ± standard deviation (SD). The qualitative or categorical variables were expressed as the number of cases (n) and relative rate (%). Each specific pelvic condition analyzed by ultrasound at the IVF center and by the Expert TVS was evaluated and compared. The result of the ultrasound examination, for each specific condition, can be either positive or negative, based on the final ultrasound experts findings, and is therefore considered a binary diagnostic test. In order to evaluate the agreement of normal and pathological findings at IVF TVS and Expert TVS overall agreement (OA), Cohen’s kappa (K), percent positive agreement (PPA) and McNemar’s Chi-squared test were calculated. Statistical significance was considered for p value < 0.05.

This study emphasizes the importance of Expert TVS evaluations in women undergoing IVF treatment, especially after RIF. The discrepancy in diagnosing pelvic pathologies between IVF centres and Expert TVS highlights the need for specialized evaluation. Given that many couples undergoing IVF have limited blastocysts, each transfer is vital. Future research should explore the impact of expert ultrasound on identifying treatable pathologies to enhance treatment strategies, improve IVF outcomes, and personalize care for better reproductive results.

This study highlights the importance of advanced 2D and 3D TVS in diagnosing pelvic pathologies in women undergoing IVF treatment, especially after RIF. The study underscores the need for Expert TVS to detect pelvic pathologies, which could play an important role in identifying potential causes of infertility and IF. Pelvic pathologies were more frequently detected in expert TVS, emphasizing the importance of thorough TVS evaluations before embryo transfer. One of the key challenges in managing RIF is the ability to tailor treatments for each patient. ESHRE group defines RIF as the repeated failure of viable embryos to implant, necessitating further investigation or intervention [4] . In this study, we choose for the RIF only failure of implant after at least two ETs of euploid blastocysts, to avoid as much as possible the embryo factor of IF. While the probability of implantation varies widely among infertile patients, individualizing treatment for RIF remains essential. Recent studies have shown that patients undergoing multiple embryo transfers with euploid blastocysts have progressively higher cumulative live birth rates (CLBR), reaching up to 95.5 % after three embryo transfers and 98.2 % after five. [2] , [3] These findings suggest that true RIF is exceedingly rare when blastocyst are euploid and uterine and tubal pathologies are excluded. Uterine abnormalities such as myomas, adenomyosis, and uterine malformations significantly impact fertility [20] . Typically, a comprehensive fertility workup is performed before IVF treatment to identify such pathologies. Transvaginal ultrasound, particularly 3D ultrasound, plays an integral role in diagnosing these conditions. Despite the uncertainty surrounding the utility of 3D ultrasound in improving IVF outcomes, this study found considerable diagnostic discrepancies between IVF centres and expert ultrasound centres, particularly regarding adenomyosis and uterine malformations ( Table 2 ). It is widely known that myomas causing uterine cavity distortion reduce clinical pregnancy rate (CPR), highlighting the importance of accurate ultrasound diagnosis [20] . Our study found the highest OA (0.95) and PPA (79.4 %) regarding the diagnosis of myomas; while significant diagnostic discrepancies with the lowest agreement and PPA for adenomyosis and uterine malformations, were revealed. This suggests that the detection of these pathologies with potential clinical implications for the infertility population, remains challenging. Recent advancements in 2D and 3D TVS imaging have led to a diagnosis with high sensitivity and specificity, ranging from 7 % to 88 % and 67–93 %, respectively [21] , [22] , [23] . Adenomyosis, commonly linked to infertility, is associated with higher miscarriage rates, especially with the junctional zone involvement [22] , [24] , [25] , [26] . In our study, a diagnostic discordance was noted for detecting dysmorphic and partial septate uteri. Evidence on the reproductive benefits of septal and dysmorphic metroplasty is inconclusive, with studies linking partial septate uteri to poor reproductive outcomes but not demonstrating universal improvements with surgical intervention [27] , [28] . T-shaped uterus diagnosis remains inconsistent without an universally accepted consensus [29] , [30] , [31] consecutively, however, hysteroscopic metroplasty showed potential benefits, future prospective randomized studies are needed to achieve strong evidence [32] , [33] . The findings of this study underscore the pivotal role of Expert TVS in diagnosing pelvic pathologies, which could significantly impact the management and outcomes of IVF. While pelvic pathologies detected via TVS after RIF are not always determinant in reproductive outcomes, as spontaneous pregnancies can occur even in the presence of DIE or a dysmorphic uterus, such abnormalities should still be considered, particularly prior to IVF and especially after RIF. Most cases of implantation failure in IVF patients can be effectively managed with personalized clinical/surgical interventions. Advanced imaging techniques, such as 3D transvaginal ultrasound, could be considered as part of the standard “fertility workup” in IVF. Pelvic pathologies, when detected early, could significantly influence treatment strategies and improve overall reproductive outcomes. The primary limitation of this study is its retrospective nature, which may result in missing data. Additionally, we included IVF procedures of several European centres with a wide range of heterogeneity. Furthermore, all ultrasound evaluations were performed by expert sonographers who were not blinded to IVF history and performed TVS with attention to investigate possible findings that can be associated to IF. Although this high level of expertise may limit reproducibility, it highlights the importance of referring patients to diagnostic imaging specialists before initiating IVF. Despite these limitations, this study’s strengths include experts sonographers, which is an important point as ultrasound is an operative-dependent examination. All known causes of RIF were excluded to limit the bias of other concomitant diseases. Last but not least, an important strength of this study is the inclusion of only euploid blastocyst (own oocytes) or blastocyst from oocyte donation.

Recurrent implantation failure (RIF) is defined as the absence of implantation after multiple embryo transfers, affecting approximately 5–10 % of couples undergoing in vitro fertilization (IVF) and good quality embryo transfer (ET) [1] , [2] , [3] . However, RIF is frequently observed in couples who have undergone IVF without embryo genetic analysis, particularly when the maternal age exceeds 35 years, leading to significant frustration and emotional distress. Maternal age significantly influences prognosis, while RIF is associated with multifactorial etiologies, including immunological, thrombophilic, and endometrial receptivity disorders, microbiome imbalances, anatomical abnormalities, pelvic diseases, male factors, and embryo aneuploidy [4] . Despite extensive research, there is no universally accepted definition or standardized diagnostic and therapeutic protocol for RIF. The European Society of Human Reproduction and Embryology (ESHRE) defines RIF as repeated implantation failure following the transfer of viable embryos, warranting further investigation and intervention [4] . Conversely, the 2022 Lugano Consensus recommends diagnosing RIF only after at least three failed transfers of euploid blastocysts or an equivalent number of unscreened embryo transfers, adjusted for maternal age and euploidy rates [5] . Recent literature reflects a shift toward an embryo-centric perspective, defining RIF as a failure of sustained implantation after transferring blastocysts into a morphologically normal uterus, as confirmed by ultrasound, saline infusion sonography, or hysteroscopy, excluding significant uterine pathologies. [2] , [3] . A thorough fertility evaluation should include clinical, endocrinological, genetic, and pelvic morphological assessments. High-resolution two-dimensional (2D), three-dimensional (3D), and power Doppler transvaginal ultrasound (TVS) are critical for detecting pelvic pathologies contributing to infertility and implantation failure. The aim of this study was to evaluate pelvic pathology in patients with RIF, following at least two embryo transfers of good-quality blastocyst, either obtained from egg donation or euploid blastocyst from their own oocytes. This assessment was conducted using a dedicated transvaginal ultrasound performed by experienced specialists.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. The author is an Editorial Board Member/Editor-in-Chief/Associate Editor/Guest Editor for this journal and was not involved in the editorial review or the decision to publish this article.