OBJECTIVES:To evaluate the performance of a machine-learning (ML) model compared with traditional logistic regression models for predicting a large-for-gestational-age (LGA) neonate at term and associated adverse perinatal outcomes, using sonographic biometric and Doppler parameters routinely assessed at a late third-trimester scan in combination with maternal demographic characteristics. METHODS:This cohort study involved a retrospective data analysis of singleton pregnancies that underwent a routine third-trimester ultrasound examination at 35 + 0 to 37 + 6 weeks' gestation and were delivered in a single tertiary referral center between December 2019 and December 2025. Pregnancies with major fetal structural anomalies, known genetic abnormalities or missing outcome data were excluded. We collected maternal demographic and obstetric characteristics, first-trimester biochemical screening results, third-trimester fetal biometry and Doppler parameters, and birth and neonatal outcome data. Both traditional statistical and contemporary ML models were developed to predict a LGA neonate (birth weight > 90th centile), as well as LGA complicated by one or more predefined adverse perinatal outcomes. Pregnancies with LGA complicated by adverse maternal outcome and those with adverse neonatal outcome were also analyzed separately to evaluate outcome-specific predictors and improve clinical interpretability. Logistic regression was used to identify predictors of each outcome with model discrimination assessed by the area under the receiver-operating-characteristics curve (AUC). In parallel, ML models (random forest, eXtreme Gradient Boosting (XGBoost), elastic net and support vector machine) were applied to capture potential non-linear relationships and interactions, providing complementary modeling approaches across different data structures. RESULTS:The study included 21 743 singleton pregnancies, with a median gestational age of 36 + 3 (interquartile range (IQR), 36 + 2 to 36 + 5) weeks at third-trimester ultrasound examination and 39 + 4 (IQR, 39 + 0 to 40 + 4) weeks at delivery, and a median birth-weight centile of 58.8 (IQR, 32.2-80.9). There were 3007 (13.8%) LGA neonates and 689 (3.2%) cases of LGA neonate with adverse outcome. The best-performing logistic regression and ML models demonstrated high discriminatory performance for predicting a LGA neonate (AUC, 0.888 (95% CI, 0.874-0.901) for both logistic regression and ML models) as well as LGA neonate with adverse perinatal outcome (AUC, 0.859 (95% CI, 0.831-0.884) for logistic regression and AUC, 0.859 (95% CI, 0.831-0.883) for ML model). Across both traditional logistic regression and ML models, the estimated fetal weight (EFW) centile was the most robust single predictor across all outcomes, with AUCs that ranged from 0.809 to 0.878. Model-specific 95% CIs are provided. Notably, ML models did not improve predictive performance compared to logistic regression. SHapley Additive exPlanations (SHAP) analysis revealed that EFW and abdominal circumference centiles were the principal determinants of risk prediction in our cohort. CONCLUSION:Routinely collected third-trimester ultrasound data can identify pregnancies at risk of LGA and associated adverse outcomes in an unselected population. EFW was the dominant predictor in both traditional logistic regression and ML models, with minimal incremental value obtained from additional maternal or ultrasound variables. ML did not improve predictive performance over traditional logistic regression, indicating that fetal size alone captured most clinically relevant information for risk stratification. These findings support a simpler, more clinically actionable approach to risk stratification, shifting focus from identifying large fetuses to predicting clinically meaningful complications. © 2026 International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVE:This systematic review aimed to evaluate the long-term outcomes of infants who had undergone fetal endoscopic tracheal occlusion (FETO) for congenital diaphragmatic hernia (CDH). METHODS:PubMed, MEDLINE, EMBASE and the Cochrane Central Register of Controlled Trials (CENTRAL) were searched from inception to October 2025 for studies reporting infant outcome ≥ 1 year post FETO in the following domains: growth and nutritional, neurodevelopmental, audiological, cardiac, respiratory, gastrointestinal and musculoskeletal. Quality was assessed using the revised Cochrane risk-of-bias tool for randomized trials and the Newcastle-Ottawa scale for cohort studies. Meta-analysis was not performed due to outcome heterogeneity and a narrative synthesis was conducted instead. RESULTS:Fourteen studies (one randomized controlled trial and 13 cohort studies), including a total of 395 infants, were analyzed. The duration of follow-up ranged from 7 to 87 months. Among domains of morbidity, respiratory complications were the most prevalent, although most improved over time. Growth failure affected a significant proportion of infants at 1 year, but considerable catch-up growth was demonstrated by 2 years. Gastrointestinal morbidity was common, with gastroesophageal reflux disease reported in 25-50% of cases and hernia recurrence in 21-43%. Neurodevelopmental outcome was generally reassuring. Multiple studies found no significant differences in long-term outcome between infants that underwent FETO and those that did not after adjusting for disease severity. CONCLUSION:FETO-treated CDH survivors experience substantial but improving multisystem morbidity that may be driven by factors independent of the intervention itself, emphasizing the need for long-term multidisciplinary follow-up. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVES:To evaluate the performance of the umbilical artery (UA) Doppler-based classification system, published by Gratacós et al., in stratifying selective fetal growth restriction (sFGR) by severity and predicting adverse outcomes in affected monochorionic diamniotic (MCDA) pregnancies. METHODS:This was a retrospective cohort study of MCDA pregnancies complicated by sFGR, seen at a single tertiary center (St George's Hospital, London, UK) from January 2000 to July 2024. Included cases were diagnosed with sFGR using the criteria of the recent Delphi consensus. Fetal growth was evaluated using twin-specific charts. A comparison of perinatal outcomes across the different groups of sFGR severity according to the UA Doppler-based classification was performed. Additionally, the discriminatory performance of this classification system and other ultrasound parameters for adverse outcomes was investigated using receiver-operating-characteristics (ROC)-curve analysis. RESULTS:Overall, 107 pregnancies complicated by sFGR were included in the analysis. The majority of cases were classified as Type-I sFGR (83/107 (77.6%)) as per the UA Doppler-based classification, while 15.0% (16/107) of cases were Type II and 7.5% (8/107) were Type III. Superimposed twin-twin transfusion syndrome occurred in 6.0% of Type-I sFGR, 18.8% of Type-II sFGR and 37.5% of Type-III sFGR cases (P = 0.010). The rate of intact survival of both twins without perinatal complications was found to be significantly higher in Type-I cases (54.2%) than in Type-II cases (18.8%) (P = 0.013). No significant differences were found in the rates of double intrauterine fetal demise (IUFD), IUFD of the smaller twin or composite adverse perinatal outcome (CAPO) of either twin. The discriminatory performance of the UA Doppler-based sFGR classification at diagnosis for predicting the occurrence of CAPO in the smaller twin was low (area under the ROC curve (AUC), 0.615 (95% CI, 0.506-0.723)). The combination of severity classification using the UA Doppler-based system with other variables (early diagnosis of sFGR, fetal weight discordance (%) and estimated fetal weight of one twin < 3rd centile) in different models adjusted for gestational age at delivery, laser photocoagulation and classification stage progression showed good discriminatory performance for predicting CAPO in the smaller twin (AUC > 0.9 for all). CONCLUSIONS:The UA Doppler-based sFGR classification has limited value in predicting adverse outcomes in pregnancies complicated by sFGR. The combination of severity classification using the UA Doppler-based system with other variables could improve the discriminatory performance and therefore its value in clinical practice. Larger studies are needed to investigate the optimal predictive tool to assess severity in monochorionic pregnancies complicated by sFGR. © 2026 International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVE:To evaluate whether fetuses exposed to threatened preterm labor (PTL) without intra-amniotic infection or inflammation (IAI), who were born ≥ 34 weeks' gestation, show signs of cardiac remodeling and dysfunction at admission and 4 weeks after the episode of threatened PTL. METHODS:This was a prospective cohort study conducted at two tertiary referral centers including pregnant patients admitted between September 2018 and April 2021. The study group comprised women with singleton pregnancies and intact membranes who were admitted with a diagnosis of PTL between 23 and 34 weeks' gestation. Pregnancies were eligible for inclusion if amniocentesis was performed at admission and ruled out the occurrence of IAI. Exclusion criteria included latency to delivery shorter than 4 weeks and delivery < 34 weeks. Fetal echocardiography was performed within 72 h after admission and was repeated at follow-up 4 weeks later. A control group of singleton pregnancies without PTL or preterm prelabor rupture of membranes (PPROM), matched with the study group at a ratio of 1:1 by gestational age at ultrasound, was also evaluated. Fetal cardiac morphology and function were assessed using a standardized echocardiographic protocol, and compared between the two groups. Amniotic fluid troponin I and N-terminal pro-brain natriuretic peptide (NT-proBNP) concentrations were measured in the study group at the time at which amniocentesis was performed, and were compared with 20 amniotic fluid samples from the Clinic-IDIBAPS Biobank that had been collected for indications other than PTL/PPROM or cardiac pathology. RESULTS:A total of 32 fetuses exposed to threatened PTL (of which 84.4% were born at term) and 43 controls were compared. At admission, fetuses exposed to threatened PTL already showed signs of cardiac remodeling compared with controls, including a more globular cardiac shape (median global sphericity index, 1.17 vs 1.23; P = 0.012), an increased right atrial area-to-heart area ratio (median, 0.18 vs 0.16; P = 0.006) and early signs of diastolic dysfunction (median lower tricuspid E/A ratio, 0.65 vs 0.72; P = 0.027) and increased systolic function (median tricuspid annular plane systolic excursion, 6.5 vs 6.0 mm; P = 0.036). At admission, the threatened-PTL group also showed significantly higher amniotic fluid troponin-I levels (median, 1222.8 pg/mL vs 841.3 pg/mL; P = 0.026). Amniotic fluid NT-proBNP was detectable in 25% of these cases, compared with none in the control group. At 4 weeks after the first ultrasound evaluation, in fetuses from mothers with threatened PTL, we observed that systolic function had normalized but cardiac remodeling persisted. These fetuses still exhibited a more globular cardiac shape (median global sphericity index, 1.16 vs 1.25; P = 0.036), and they now had a larger right ventricle (RV) (median RV area-to-heart area ratio, 0.26 vs 0.24; P = 0.037) without cardiomegaly. Diastolic function remained altered, with prolonged tricuspid inflow duration (median inflow time fraction, 0.39 vs 0.35; P = 0.001) and longer atrial contraction (median atrial time fraction, 0.23 vs 0.20; P = 0.039). CONCLUSIONS:Fetuses exposed to threatened PTL without IAI, who were born ≥ 34 weeks' gestation, exhibited signs of cardiac remodeling and subclinical diastolic dysfunction of the RV at admission that persisted for at least 4 weeks. Early identification of threatened PTL may provide opportunities to apply preventive strategies aimed at improving long-term cardiovascular health. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVE:To determine the retrospective yield of prenatal exome sequencing (PES) by establishing the proportion of children with a postnatal monogenic diagnosis that could have been diagnosed prenatally if PES had been available. METHODS:The study cohort comprised a sample of children in Northern Ireland, born between January 2010 and January 2018 (predating routine availability of PES), who received a monogenic diagnosis postnatally via next generation sequencing as part of either of two UK-wide studies (the 100 000 Genomes Project (2015-2018) or the Deciphering Developmental Disorders study (2011-2015)). Clinical data were collected retrospectively and correlated with the current UK National Health Service PES protocol, including the phenotypic eligibility criteria for PES and the associated fetal anomalies gene panel. Cases were considered retrospective diagnoses if the fetal phenotype would have been eligible for PES and the diagnostic gene was included on the test panel, meaning prenatal diagnosis in this current era could have been feasible. RESULTS:Of 101 children, 17.8% (95% CI, 10.3-25.3%) had both an eligible fetal structural anomaly (FSA) (i.e. high-risk FSA) and a diagnostic gene on the associated test panel, meaning that they could have been diagnosed prenatally in the current clinical landscape. The median length of the diagnostic odyssey for this subgroup of children was 3.7 years (1354 (range, 822-2450) days). Moreover, 58.4% (n = 59) of cases had no anomalies detected prenatally and 19.8% (n = 20) had a FSA that would not meet the eligibility criteria for PES (low-risk FSA). Although these cases would have been ineligible for PES under the current clinical pathway, 89.9% (n = 71/79) were affected by severe or profound syndromes. Postnatally, the most common functional anomalies were neurodevelopmental delay/intellectual disability and/or behavioral abnormality, which were observed in 80.2% (n = 81) of the included children. However, 80.2% (n = 65/81) of these affected children did not present with fetal anomalies eligible for PES. CONCLUSIONS:Almost one-fifth of children with a monogenic condition included in this study could have received a diagnosis via modern PES, avoiding a diagnostic odyssey lasting almost 4 years. However, despite having a monogenic condition, over half of the children did not present with any structural anomalies in utero. This demonstrates the degree to which fetal imaging is limited in its ability to reassure parents of the absence of a fetal genetic syndrome. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVE:To evaluate the diagnostic performance of transvaginal ultrasound (TVS) for adenomyosis using consensus-based direct and indirect ultrasound signs, with histopathology as the reference standard. METHODS:This prospective diagnostic accuracy study included women undergoing TVS examination at a tertiary referral center (Hospital San Juan de Dios, Santiago, Chile) between 1 December 2022 and 1 December 2023. Women aged ≥ 18 years who underwent hysterectomy within 6 months after TVS examination were enrolled consecutively. Cases with malignant histopathological findings were excluded. Ultrasound signs of adenomyosis were assessed according to the revised Morphological Uterus Sonographic Assessment consensus. Histopathology served as the reference standard. Diagnostic performance measures for individual ultrasound features and multivariable models were calculated. RESULTS:Among 2967 women evaluated using TVS, 345 (11.6%) underwent hysterectomy within 6 weeks and constituted the study cohort. Histopathology confirmed adenomyosis in 155 (44.9%) women. Individual ultrasound features showed high specificity but limited sensitivity. Among direct signs, echogenic subendometrial lines showed the best overall balance (sensitivity, 53.5% (95% CI, 45.4-61.6%); specificity, 94.7% (95% CI, 90.5-97.4%)). Individual indirect signs showed specificity > 97% but low sensitivity. A model combining direct ultrasound signs showed high specificity (93.7% (95% CI, 89.2-96.7%)) but limited sensitivity (60.0% (95% CI, 51.8-67.8%)); overall diagnostic accuracy was 78.6% (95% CI, 73.8-82.8%). The addition of indirect ultrasound signs into the model did not provide meaningful incremental benefit. CONCLUSIONS:Our findings highlight the importance of distinguishing between direct and indirect ultrasound features, with direct signs providing the most consistent diagnostic value, while indirect signs should be interpreted as supportive rather than definitive. © 2026 International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVES:To describe, characterize and determine the frequency of interhemispheric fissure (IHF) distortion on ultrasound in fetuses with open spina bifida (OSB), evaluate its association with other intracranial findings and biometric parameters and assess its evolution after prenatal closure of the spinal defect. METHODS:This retrospective cohort study included fetuses with suspected OSB referred to a single tertiary fetal center (Ontario Fetal Centre, Toronto, ON, Canada) between June 2017 and May 2025. Inclusion required prenatal diagnosis of OSB and at least one ultrasound examination with adequate axial transventricular and transthalamic planes of the fetal brain for evaluation of the IHF. IHF distortion was defined as deviation from the expected linear echogenic midline appearance. Biometric measurements were converted to gestational age-adjusted Z-scores. Associations between IHF distortion and clinical and imaging variables were assessed using univariable and multivariable logistic regression. Postoperative imaging was reviewed in fetuses that underwent prenatal spinal defect closure to evaluate changes in IHF appearance. RESULTS:Among 133 fetuses with confirmed OSB, IHF distortion was identified in 54 (40.6%). Interobserver agreement was excellent (Cohen's kappa coefficient = 0.83). Fetuses with IHF distortion had smaller transcerebellar diameter (TCD) Z-scores compared with those without distortion (-4.80 ± 2.11 vs -3.91 ± 2.54; P = 0.047). On multivariable logistic regression, later gestational age at OSB diagnosis (adjusted odds ratio (aOR), 1.43 (95% CI, 1.08-1.94); P = 0.01) and smaller TCD Z-score (aOR, 0.89 (95% CI, 0.76-0.99); P = 0.04) were independently associated with IHF distortion. Among fetuses that underwent prenatal closure and had postoperative imaging available, resolution of IHF distortion paralleled improvement in hindbrain herniation, whereas persistence of IHF distortion occurred only in cases with minimal or no improvement in hindbrain herniation. CONCLUSIONS:IHF distortion is a frequent and reproducible prenatal ultrasound finding in fetuses with OSB. Its association with a smaller TCD and its apparent resolution following prenatal closure of OSB in cases that demonstrated improvement in hindbrain herniation suggest that IHF distortion may represent a secondary mechanical manifestation within the Chiari II spectrum. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVE:To describe the imaging features of adnexal lesions during pregnancy and assess the performance of current models for the risk assessment of such lesions. METHODS:This was a PROSPERO-registered systematic review including studies published from inception to 1 August 2025 and was performed as per Preferred Reporting Systems for Systematic Reviews and Meta-Analysis guidance using Covidence software. Studies were included if they utilized an objective or standardized assessment of imaging features (based on the International Ovarian Tumor Analysis (IOTA) terms and definitions, or predefined categorical variables), expert description or quantitative data (such as lesion measurements and number of papillary projections) to assess the risk of adnexal lesions during pregnancy. Standardized extraction of imaging features was performed and classification of outcomes was based on histology, resolution or expert impression. Diagnostic model performance was also extracted. Meta-analysis of model performance was not performed owing to study heterogeneity. RESULTS:Overall, 3349 studies were screened, of which 40 underwent full-text review and 16 were included in the systematic review, comprising 1674 patients and 1660 adnexal lesions. Most included studies were retrospective and ultrasound-based. Excluding six studies that exclusively reported benign lesions, 93.1% (1284/1379) of patients had benign lesions, 3.4% (47/1379) had borderline ovarian tumors (BOTs) and 3.5% (48/1379) had malignant lesions. Excluding studies reporting exclusively on endometrioma, the most common benign adnexal lesions were simple cysts (39.0% (393/1007)), mature cystic teratomas (24.2% (244/1007)) and endometriomas (14.9% (150/1007)). Malignant lesions were most often epithelial ovarian cancer or metastases. Both BOTs and malignant lesions retained imaging characteristics comparable with those observed in non-pregnant women. Decidualization was present in 11.8-39.5% of reported endometriomas. While decidualized endometriomas (DEs) and BOTs typically share characteristics, experts noted distinguishing morphological features and behaviors. BOTs were larger (median, 88.0 mm vs 48.1 mm), and although both were predominantly unilocular-solid lesions, BOTs were relatively more often multilocular-solid (30.0% vs 22.4%) compared with DEs. Both showed high vascularity, with most cases having a color score of 3. Most DEs (83.9%) had one to three papillary projections, whereas most BOTs (66.7%) had more than three papillary projections. Diagnostic models, such as the IOTA two-step strategy and the Assessment of Different NEoplasias in the adneXa model alone, showed high specificity (70.0-90.2%) and negative predictive value (90.0-98.3%), but performance was limited by misclassification of DEs as malignant. CONCLUSIONS:Ultrasound can reliably assess adnexal lesions in pregnancy. DEs and BOTs can appear similar owing to their morphological overlap. Both BOTs and malignant lesions in pregnancy show morphology comparable with that in non-pregnant women, and IOTA models maintain a high negative predictive value for excluding malignancy in such cases. Future research should refine diagnostic tools to improve the differentiation of benign from malignant cysts with solid components. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVES:To evaluate the variability and reproducibility of umbilical artery (UA), fetal middle cerebral artery (MCA) and uterine artery (UtA) Doppler ultrasound measurements in pregnancy. METHODS:A systematic search of MEDLINE, EMBASE and the Cochrane Library (CENTRAL) was conducted from inception to 17 June 2024. Studies were included if they reported intra- or interobserver variability of UA, MCA or UtA Doppler ultrasound measurements in singleton or multiple pregnancies, using the intraclass correlation coefficient (ICC), concordance correlation coefficient, Cohen's kappa or limits of agreement. Data extraction was performed using standardized data-extraction forms. A meta-analysis was conducted using a random-effects model to account for heterogeneity, and variability estimates were pooled using Fisher's Z-transformation. Reproducibility was categorized according to the True Reproducibility of Ultrasound Techniques (TRUST) criteria. RESULTS:In total, 2426 records were screened. Of these, 26 studies including 2457 patients met the inclusion criteria; eight studies described UA Doppler, 11 described MCA Doppler and 11 described UtA Doppler results; two of the studies described cerebroplacental ratio results. The pooled intraobserver ICC for UA pulsatility index (PI) was 0.88 (95% CI, 0.77-0.94) and the pooled interobserver ICC was 0.68 (95% CI, 0.55-0.79). For MCA-PI, these values were 0.85 (95% CI, 0.69-0.93) and 0.80 (95% CI, 0.62-0.90), respectively, and for mean UtA-PI they were 0.90 (95% CI, 0.85-0.93) and 0.84 (95% CI, 0.78-0.89), respectively. Findings across studies suggested generally poor-to-moderate reproducibility according to the strict TRUST criteria. CONCLUSIONS:Our results emphasize that there is room for improvement regarding the reproducibility of Doppler ultrasound in routine obstetric practice. By standardizing ultrasound measurement protocols, advancing technologies such as artificial-intelligence-supported measurement applications and training programs, as well as obtaining multiple measurements per session and performing quality audits when outliers are detected, the field could move towards more accurate and reliable fetal assessments. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVE:To perform a head-to-head comparison of the diagnostic performance of ultrasonography (US) vs magnetic resonance imaging (MRI) for the preoperative detection of endometriosis using the #Enzian classification system. METHODS:This was a prospective observational study of postmenarchal premenopausal women who underwent laparoscopy between 1 April 2021 and 31 May 2024 in a tertiary referral center for clinical suspicion of endometriosis. All women underwent both transvaginal and transabdominal US imaging, according to the International Deep Endometriosis Analysis (IDEA) protocol, as well as MRI evaluation. The sonologist and the radiologist were blinded to each other's results. The diagnostic performance of US and MRI were determined for the following #Enzian compartments: O (ovarian endometriosis), T (adhesions at the level of the tubo-ovarian unit), A (deep endometriosis (DE) in the vagina, rectovaginal septum and torus uterinus), B (DE in the uterosacral ligament, cardinal ligament, parametrium and pelvic sidewall), C (DE in the rectosigmoid bowel), FI (lesions in other intestinal locations), FB (urinary bladder involvement), FU (ureteric involvement) and FO (lesions in other extragenital locations). The sensitivity and specificity were compared between imaging modalities using McNemar's test. Concordance between preoperative imaging and surgical assessment of #Enzian severity grade for endometriotic lesions/adhesions in #Enzian compartments O, T, A, B and C was assessed. RESULTS:In total, 463 women who underwent laposcopy for diagnosis and excisional treatment of endometriosis were enrolled in the study. For the detection of endometriotic lesions/adhesions in #Enzian compartments Oleft, Oright, Tleft, Tright, A, Bleft, Bright, C, FI, FU, FB and FO, US had a sensitivity of 97.52%, 97.22%, 85.15%, 85.26%, 92.66%, 88.81%, 86.79%, 95.79%, 75.00%, 76.19%, 83.72% and 60.00%, respectively, and a specificity of 96.77%, 97.73%, 96.63%, 94.79%, 91.02%, 96.66%, 96.92%, 95.60%, 98.77%, 97.96%, 99.05% and 99.56%, respectively. For MRI, the corresponding values for sensitivity were 95.04%, 94.44%, 77.23%, 74.74%, 90.37%, 88.06%, 90.57%, 94.21%, 83.93%, 71.43%, 74.42% and 90.00%, and those for specificity were 95.89%, 95.75%, 92.70%, 89.86%, 93.47%, 96.05%, 94.96%, 94.14%, 97.05%, 99.10%, 98.57% and 98.90%. US had significantly higher specificity in #Enzian compartment T on both the left (P = 0.0336) and right (P = 0.0153) sides compared with MRI, but no other comparisons of sensitivity or specificity between US and MRI showed statistically significant differences. US showed higher exact concordance with surgical assessment in severity grading for lesions/adhesions in the adnexa than did MRI (Oleft, 96.3% vs 94.8%; Oright, 97.2% vs 94.6%; Tleft, 93.9% vs 89.0%; Tright, 92.6% vs 86.7%). For DE nodules, US showed a 3.1 percentage point higher exact concordance with laparoscopy compared with MRI in #Enzian compartment C (94.0% vs 90.9%). CONCLUSIONS:US evaluation following the IDEA protocol has significantly higher specificity compared with that of MRI for pelvic adhesions in #Enzian compartment T. However, US and MRI have similarly high sensitivity and specificity for diagnosing DE in all relevant #Enzian compartments. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVE:To investigate, using transvaginal ultrasound imaging, the overlap between polycystic ovarian morphology (PCOM) and endometriosis, and to explore demographic and clinical factors associated with their coexistence across endometriosis phenotypes. METHODS:This was a retrospective cohort study of consecutive patients undergoing advanced transvaginal ultrasound examination for suspected or previously diagnosed endometriosis at a tertiary gynecological ultrasound clinic between February 2023 and June 2023. Ultrasound was performed following the International Deep Endometriosis Analysis consensus and the International Ovarian Tumor Analysis framework guided lesion characterization, with PCOM defined as per the 2018 and the modified 2023 International Polycystic Ovary Syndrome Guideline (≥ 20 follicles measuring 2-9 mm in diameter and/or ovarian volume > 10 mL in the absence of a dominant follicle, cyst or corpus luteum). Endometriosis phenotypes were classified sonographically as superficial (SE), ovarian (OE) or deep (DE) endometriosis. Demographic and clinical variables were compared between participants with endometriosis alone and those with concurrent endometriosis and PCOM. Logistic regression analysis was performed within the endometriosis-positive subgroup using PCOM as the dependent variable in a univariable model including age and a multivariable model including age and body mass index (BMI). An age-restricted (25-35 years) sensitivity analysis compared the frequencies of any endometriosis and of individual endometriosis phenotypes between participants with and those without PCOM. A secondary analysis within this subgroup compared clinical and demographic characteristics between endometriosis participants with and those without concurrent PCOM. RESULTS:Among 165 included patients, 62.4% (n = 103) were diagnosed with endometriosis, 37.0% (n = 61) demonstrated PCOM and 35.0% (36/103) of those with endometriosis had concurrent PCOM. In the whole cohort, the frequency of any endometriosis did not differ significantly between participants with and those without PCOM. DE was less frequent among those with PCOM, whereas differences in the frequency of OE and SE between participants with and those without PCOM were not statistically significant. Compared to individuals with endometriosis alone, those with concurrent PCOM were significantly younger and had lower body weight. The proportions of nulligravidae and of nulliparae were also higher. In age-restricted analyses limited to participants aged 25-35 years, no significant differences were observed between those with and those without PCOM for the presence of any endometriosis or for any individual endometriosis phenotype. However, among endometriosis-positive participants, increasing age was associated with lower odds of concurrent PCOM in both the univariable model including age (odds ratio (OR), 0.83 (95% CI, 0.77-0.90) per 1-year increase; P < 0.001) and the multivariable model including age and BMI (OR, 0.84 (95% CI, 0.77-0.92) per 1-year increase; P < 0.001), whereas BMI was not independently associated with PCOM (OR, 0.99 (95% CI, 0.91-1.09) per 1 kg/m2 increase; P = 0.874). In the sensitivity analysis restricted to participants aged 25-35 years, the differences observed previously in age, weight, gravidity and parity were no longer statistically significant. CONCLUSIONS:Endometriosis and PCOM can coexist on ultrasound in a tertiary referral cohort. However, the observed between-group differences were strongly influenced by age, and our findings should be interpreted as exploratory and hypothesis-generating rather than confirmatory. Prospective studies incorporating endocrine characterization, standardized cycle-phase assessment and side-specific ovarian assessment are needed to clarify the clinical and biological significance of this coexistence. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVES:The first aim was to describe longitudinal changes in levator hiatus area from 21 weeks' gestation in the first pregnancy to 1 and 8 years after first delivery, according to subsequent vaginal parity in women with vaginal delivery or Cesarean section (CS) at first delivery. The second aim was to assess the effect of subsequent vaginal delivery on levator hiatus area after first vaginal delivery. METHODS:In this prospective cohort study with an 8-year recall extension, 183 initially nulliparous women underwent three- and four-dimensional transperineal ultrasound examination at 21 weeks' gestation and 1 and 8 years after first delivery. The levator hiatus area at rest, during maximum pelvic floor muscle contraction and during maximum Valsalva maneuver was assessed at each timepoint. Participants with multiple vaginal deliveries (vaginal multiparae), only one vaginal delivery (vaginal primiparae), CS for all deliveries and vaginal birth after CS were compared. The differences between study groups in levator hiatus area over time were assessed using a linear mixed model. RESULTS:Overall, vaginal multiparae showed a statistically significant non-linear increase in levator hiatus area from 21 weeks' gestation to 8 years after first delivery (P < 0.001). In contrast, participants who delivered exclusively by CS showed a decrease in levator hiatus area from 21 weeks' gestation to the 8-year follow-up. Subsequent vaginal delivery after an initial vaginal delivery did not affect the levator hiatus area during Valsalva maneuver. At the 1-year and 8-year follow-up after the first vaginal delivery, levator hiatus area at rest remained unchanged in vaginal multiparae but decreased in vaginal primiparae. At the 8-year follow-up, vaginal multiparae had a larger hiatus area during pelvic floor muscle contraction than did vaginal primiparae. CONCLUSIONS:In this study, the levator hiatus area was influenced by the mode of the first delivery and by subsequent vaginal deliveries. Overall, the levator hiatus area increased from 21 weeks' gestation to 8 years in participants with at least one vaginal delivery and decreased in those who delivered exclusively by CS. Subsequent vaginal delivery after a first did not seem to affect the levator hiatus area measured during Valsalva maneuver. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.
ABSTRACT Objective To review systematically and assess quantitatively the incidence and risk of placenta accreta spectrum (PAS) disorders and hysterectomy in Cesarean scar pregnancy (CSP) continuing into mid‐to‐late gestation. Methods A comprehensive literature search was performed in PubMed, EMBASE, Cochrane Library, ClinicalTrials.gov , China National Knowledge Infrastructure (CNKI) and Yiigle databases from inception to 12 February 2025. Studies reporting the rates of PAS and/or hysterectomy in women with a singleton CSP that continued ≥ 14 weeks' gestation were eligible for inclusion. Data extracted included information on medical history, CSP classification and clinical manifestation, gestational age at delivery and pregnancy outcome. Data were synthesized as pooled incidences and odds ratios with 95% CIs, and random‐effects models were used owing to high heterogeneity. Subgroup analyses were performed according to CSP subtype (Type I vs Type II vs Type III; and endogenous vs exogenous) and residual myometrial thickness (RMT) (≤ 3 mm vs > 3 mm). Results Among 61 studies included in the systematic review, 20 studies (432 cases) were eligible for inclusion in the meta‐analysis. Qualitative synthesis captured severe clinical phenotypes, including spontaneous uterine rupture and massive hemorrhage during Cesarean section. The pooled incidences of PAS and hysterectomy were 0.90 (95% CI, 0.81–0.96) and 0.39 (95% CI, 0.22–0.57), respectively. Marked heterogeneity was observed across CSP subtypes: Type‐III and exogenous CSPs were associated with significantly higher odds of invasive PAS, hysterectomy and preterm delivery compared with Types‐I/II and endogenous subtypes, respectively. Reduced RMT was associated significantly with increased odds of overall PAS and invasive PAS, and was predictive of hysterectomy on receiver‐operating‐characteristics‐curve analysis. The pooled incidence of reaching ≥ 28 weeks’ gestation was 0.85 (95% CI, 0.73–0.94), whereas the pooled incidence of reaching term (≥ 37 weeks) was 0.19 (95% CI, 0.10–0.30). Among studies that reported uterine rupture, the pooled incidence of this outcome was 0.05 (95% CI, 0.02–0.10). Conclusions Type‐III and exogenous CSPs that progress into mid‐to‐late gestation present substantial risks for PAS and hysterectomy, with a low likelihood of term delivery. Type‐I, Type‐II and endogenous CSPs are associated with a comparatively lower risk of adverse outcome. CSP subtype and RMT are key determinants for risk stratification and individualized counseling. © 2026 International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVE:Gestational diabetes mellitus (GDM) is associated with future maternal cardiovascular disease, independent of the development of Type-2 diabetes mellitus. As a major cardiovascular risk factor, we examined the incidence and predictors of hypertension at 5 months postpartum in women with previous GDM. METHODS:Between September 2023 and January 2025, we conducted a single-center observational prospective cohort study of all women who received routine prenatal care at 12 weeks' gestation at King's College Hospital, London, UK, and developed GDM during the index pregnancy. Those with chronic hypertension were excluded from the analysis. Women were invited to a 5-month postpartum visit, at which assessment was undertaken for hypertension (systolic blood pressure (BP) ≥ 130 mmHg or diastolic BP ≥ 80 mmHg, or receiving antihypertensive treatment), obesity, adiposity, dysglycemia, dyslipidemia and renal dysfunction. Logistic regression was performed to assess predictors of postnatal hypertension. RESULTS:Among the 678 included women with previous GDM in the index pregnancy, 179 (26.4%) developed hypertension at a median of 5.1 (interquartile range (IQR), 4.4-6.7) months postpartum. Multivariable analysis showed that postpartum hypertension was associated with: higher maternal age, Black or mixed ethnicity, higher median weight in early pregnancy, higher systolic and diastolic blood pressure in early pregnancy, and development of pre-eclampsia (PE) or gestational hypertension (GH) in the index pregnancy. The area under the receiver-operating-characteristics curve for the prediction of postpartum hypertension was 0.72 (95% CI, 0.68-0.77) without inclusion of development of PE or GH, and 0.74 (95% CI, 0.70-0.78) with their inclusion. The corresponding detection rates for postpartum hypertension were 49.5% and 53.7%, respectively, at a 20% false-positive rate. Notably, women with GDM in the index pregnancy frequently had dysglycemia (351/678 (51.8%)) as well as body mass index ≥ 30 kg/m2 (232/678 (34.2%)), waist-to-height ratio > 0.5 (502/678 (74.0%)), dyslipidemia (186/678 (27.4%)) and/or renal dysfunction (78/678 (11.5%)). At the 5-month postpartum review, among women with previous GDM who developed postpartum hypertension compared to those who were normotensive at follow-up, median BMI was significantly higher (30.1 (IQR, 26.7-36.8) vs 27.0 (IQR, 23.5-30.6) kg/m2), a waist-to-height ratio > 0.5 was significantly more prevalent (87.7% vs 69.1%) and dyslipidemia was significantly more common (35.8% vs 24.4%). CONCLUSION:One in four women with recent GDM developed hypertension at 5 months postpartum. However, risk prediction from factors observed in early pregnancy was only modest, reinforcing the need for universal postpartum BP assessment. The high prevalence of hypertension and other cardio-renal-metabolic abnormalities following GDM emphasizes the associated enhanced cardiovascular risk, and the postnatal opportunities for intervention early in the life course of these women. © 2026 The Author(s). Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.
OBJECTIVES:To evaluate the outcome of monochorionic diamniotic (MCDA) pregnancies complicated by twin reversed arterial perfusion (TRAP) sequence and to evaluate potential prognostic ultrasound markers as predictors of pump-twin survival. METHODS:This was a retrospective cohort study of MCDA twin pregnancies diagnosed with TRAP sequence at two medical centers in China between January 2010 and December 2024. All cases were analyzed retrospectively for intrauterine course and outcome. Cases that underwent expectant management were assigned to the primary cohort (cases from the First Affiliated Hospital of Sun Yat-sen University, Guangzhou) or the external-validation cohort (cases from Dalian Women and Children's Medical Group, Dalian). Cases that underwent intrauterine intervention at the First Affiliated Hospital of Sun Yat-sen University were assigned to the intrauterine-intervention cohort. The following parameters were evaluated in the primary and external-validation cohorts for their predictive value regarding pump-twin survival: in the first trimester, the crown-rump length (CRL) and nuchal translucency thickness of the pump twin, the upper pole-rump length (URL) of the acardiac twin, the URL/CRL ratio, and the presence of reversed a-wave in the ductus venosus and the umbilical artery pulsatility index (UA-PI) in the acardiac twin; and in the second trimester, the URL of the acardiac twin, the estimated fetal weight (EFW) of each twin, the twin weight ratio (TWR) (EFWacardiac twin/EFWpump twin) and the UA-PI of each twin. Univariate Cox regression analysis was performed to identify the parameters associated with pump-twin survival. Performance was evaluated by the area under the receiver-operating-characteristics curves (AUCs) and calibration curves. Finally, Kaplan-Meier survival analysis was performed. RESULTS:A total of 117 MCDA pregnancies with TRAP sequence were included in the analysis: 55 in the primary cohort, 38 in the intrauterine-intervention cohort and 24 in the external-validation cohort. The overall pump-twin survival rate was 65.8% (77/117). Three variables (URL and URL/CRL ratio in the first trimester and TWR in the second trimester) were identified to be independent risk factors influencing pump-twin survival in both the primary and external-validation cohorts. The AUCs were indicative of good predictive discrimination in both cohorts: 0.811 (95% CI, 0.667-0.956) for URL, 0.780 (95% CI, 0.619-0.941) for the URL/CRL ratio and 0.859 (95% CI, 0.730-0.988) for TWR in the primary cohort; and 0.844 (95% CI, 0.662-1.000) for URL, 0.823 (0.633-1.000) for the URL/CRL ratio and 0.925 (95% CI, 0.766-1.000) for TWR in the external-validation cohort. The optimal cut-off values were 4.85 cm for URL, 0.54 for the URL/CRL ratio and 0.295 for TWR in the primary cohort. The TWR exhibited good calibration, but URL and the URL/CRL ratio exhibited poor calibration. Kaplan-Meier survival analysis revealed that pump twins with TWR ≥ 0.295 experienced a significantly shorter intrauterine survival duration in both the primary and external-validation cohorts, and pump twins with URL ≥ 4.85 cm or URL/CRL ratio ≥ 0.54 experienced a significantly shorter intrauterine survival duration in the primary cohort. However, the optimal cut-offs of URL and the URL/CRL ratio showed no statistical difference in the external-validation cohort. CONCLUSION:After 14 + 0 weeks' gestation, the threshold of TWR < 0.295 can reliably predict survival of the pump twin in MCDA pregnancies with TRAP sequence. However, the predictive value of URL and the URL/CRL ratio in the first trimester for pump-twin survival is poor. A well-designed prospective multicenter study is warranted to develop a first-trimester multivariable predictive model with enhanced predictive performance. © 2026 International Society of Ultrasound in Obstetrics and Gynecology.
