Why are congenital heart defects being missed?

Automatic segmentation of 15 critical anatomical labels and measurements of cardiac axis and cardiothoracic ratio in fetal four chambers using nnU-NetV2

BACKGROUND

Accurate segmentation of critical anatomical structures in fetal four-chamber view images is essential for the early detection of congenital heart defects. Current prenatal screening methods rely on manual measurements, which are time-consuming and prone to inter-observer variability. This study develops an AI-based model using the state-of-the-art nnU-NetV2 architecture for automatic segmentation and measurement of key anatomical structures in fetal four-chamber view images.

METHODS

A dataset, consisting of 1,083 high-quality fetal four-chamber view images, was annotated with 15 critical anatomical labels and divided into training/validation (867 images) and test (216 images) sets. An AI-based model using the nnU-NetV2 architecture was trained on the annotated images and evaluated using the mean Dice coefficient (mDice) and mean intersection over union (mIoU) metrics. The model's performance in automatically computing the cardiac axis (CAx) and cardiothoracic ratio (CTR) was compared with measurements from sonographers with varying levels of experience.

RESULTS

The AI-based model achieved a mDice coefficient of 87.11% and an mIoU of 77.68% for the segmentation of critical anatomical structures. The model's automated CAx and CTR measurements showed strong agreement with those of experienced sonographers, with respective intraclass correlation coefficients (ICCs) of 0.83 and 0.81. Bland-Altman analysis further confirmed the high agreement between the model and experienced sonographers.

CONCLUSION

We developed an AI-based model using the nnU-NetV2 architecture for accurate segmentation and automated measurement of critical anatomical structures in fetal four-chamber view images. Our model demonstrated high segmentation accuracy and strong agreement with experienced sonographers in computing clinically relevant parameters. This approach has the potential to improve the efficiency and reliability of prenatal cardiac screening, ultimately contributing to the early detection of congenital heart defects.

Initiating a Fetal Cardiac Program from Scratch in Low- and Middle-Income Countries: Structure, Challenges, and Hopes for Solutions

Although fetal cardiac programs are well established in developed countries, establishing an efficient program in low- and middle-income countries (LMICs) is still considered a significant challenge. Substantial obstacles usually face the initiation of fetal cardiac service from scratch in LMICs. The primary structural frame of a successful fetal cardiac program is described in detail, emphasizing the required team members. The potential challenges for starting fetal cardiac services in LMICs include financial, awareness-related, prenatal obstetric screening, sociocultural, psychosocial, and social support factors. These challenges could be solved by addressing these barriers, such as collecting funds for financial support, raising awareness among families and health care providers, telemedicine, building international health partnerships, modifying training protocols for fetal cardiologists and sonographers, and initiating support groups and social services for families with confirmed fetal cardiac disease. Initiating a successful fetal cardiac program requires multi-aspect structural planning. The challenges for program initiation require diverse efforts, from modified training and promoting awareness of care providers and the community to governmental and nonprofit organizations' collaborations for proper building and utilization of program resources.

A deep learning framework for identifying and segmenting three vessels in fetal heart ultrasound images

BACKGROUND

Congenital heart disease (CHD) is one of the most common birth defects in the world. It is the leading cause of infant mortality, necessitating an early diagnosis for timely intervention. Prenatal screening using ultrasound is the primary method for CHD detection. However, its effectiveness is heavily reliant on the expertise of physicians, leading to subjective interpretations and potential underdiagnosis. Therefore, a method for automatic analysis of fetal cardiac ultrasound images is highly desired to assist an objective and effective CHD diagnosis.

METHOD

In this study, we propose a deep learning-based framework for the identification and segmentation of the three vessels-the pulmonary artery, aorta, and superior vena cava-in the ultrasound three vessel view (3VV) of the fetal heart. In the first stage of the framework, the object detection model Yolov5 is employed to identify the three vessels and localize the Region of Interest (ROI) within the original full-sized ultrasound images. Subsequently, a modified Deeplabv3 equipped with our novel AMFF (Attentional Multi-scale Feature Fusion) module is applied in the second stage to segment the three vessels within the cropped ROI images.

RESULTS

We evaluated our method with a dataset consisting of 511 fetal heart 3VV images. Compared to existing models, our framework exhibits superior performance in the segmentation of all the three vessels, demonstrating the Dice coefficients of 85.55%, 89.12%, and 77.54% for PA, Ao and SVC respectively.

CONCLUSIONS

Our experimental results show that our proposed framework can automatically and accurately detect and segment the three vessels in fetal heart 3VV images. This method has the potential to assist sonographers in enhancing the precision of vessel assessment during fetal heart examinations.

Prenatal detection rates for congenital heart disease using abnormal obstetrical screening ultrasound alone as indication for fetal echocardiography

OBJECTIVE

To determine the live born prenatal detection rate of significant congenital heart disease (CHD) in a large, integrated, multi-center community-based health system using a strategy of referral only of patients with significant cardiac abnormalities on obstetrical screening ultrasound for fetal echocardiography. Detection rates were assessed for screening in both radiology and maternal fetal medicine (MFM). The impact on fetal echocardiography utilization was also assessed.

METHODS

This was a retrospective cohort study using an electronic health record, outside claims databases and chart review to determine all live births between 2016 and 2020 with postnatally confirmed sCHD that were prenatally detectable and resulted in cardiac surgery, intervention, or death within 1 year.

RESULTS

There were 214,486 pregnancies resulting in live births. Prenatally detectable significant CHD was confirmed in 294 infants. Of those 183 were detected for an overall live-born detection rate of 62%. Detection rates in MFM were 75% and in radiology were 52%. The number of fetal echocardiograms needed to detect (NND) sCHD was 7.

CONCLUSIONS

A focus on quality and standardization of obstetrical screening ultrasound with referral to fetal echocardiography for cardiac abnormalities alone achieves benchmark targets for live-born detection of significant CHD requiring fewer fetal echocardiograms.

AI supported fetal echocardiography with quality assessment

This study aimed to develop a deep learning model to assess the quality of fetal echocardiography and to perform prospective clinical validation. The model was trained on data from the 18-22-week anomaly scan conducted in seven hospitals from 2008 to 2018. Prospective validation involved 100 patients from two hospitals. A total of 5363 images from 2551 pregnancies were used for training and validation. The model's segmentation accuracy depended on image quality measured by a quality score (QS). It achieved an overall average accuracy of 0.91 (SD 0.09) across the test set, with images having above-average QS scoring 0.97 (SD 0.03). During prospective validation of 192 images, clinicians rated 44.8% (SD 9.8) of images as equal in quality, 18.69% (SD 5.7) favoring auto-captured images and 36.51% (SD 9.0) preferring manually captured ones. Images with above average QS showed better agreement on segmentations (p < 0.001) and QS (p < 0.001) with fetal medicine experts. Auto-capture saved additional planes beyond protocol requirements, resulting in more comprehensive echocardiographies. Low QS had adverse effect on both model performance and clinician's agreement with model feedback. The findings highlight the importance of developing and evaluating AI models based on 'noisy' real-life data rather than pursuing the highest accuracy possible with retrospective academic-grade data.

Deep-learning model for prenatal congenital heart disease screening generalizes to community setting and outperforms clinical detection

OBJECTIVES

Despite nearly universal prenatal ultrasound screening programs, congenital heart defects (CHD) are still missed, which may result in severe morbidity or even death. Deep machine learning (DL) can automate image recognition from ultrasound. The main aim of this study was to assess the performance of a previously developed DL model, trained on images from a tertiary center, using fetal ultrasound images obtained during the second-trimester standard anomaly scan in a low-risk population. A secondary aim was to compare initial screening diagnosis, which made use of live imaging at the point-of-care, with diagnosis by clinicians evaluating only stored images.

METHODS

All pregnancies with isolated severe CHD in the Northwestern region of The Netherlands between 2015 and 2016 with available stored images were evaluated, as well as a sample of normal fetuses' examinations from the same region and time period. We compared the accuracy of the initial clinical diagnosis (made in real time with access to live imaging) with that of the model (which had only stored imaging available) and with the performance of three blinded human experts who had access only to the stored images (like the model). We analyzed performance according to ultrasound study characteristics, such as duration and quality (scored independently by investigators), number of stored images and availability of screening views.

RESULTS

A total of 42 normal fetuses and 66 cases of isolated CHD at birth were analyzed. Of the abnormal cases, 31 were missed and 35 were detected at the time of the clinical anatomy scan (sensitivity, 53%). Model sensitivity and specificity were 91% and 78%, respectively. Blinded human experts (n = 3) achieved mean ± SD sensitivity and specificity of 55 ± 10% (range, 47-67%) and 71 ± 13% (range, 57-83%), respectively. There was a statistically significant difference in model correctness according to expert-graded image quality (P = 0.03). The abnormal cases included 19 lesions that the model had not encountered during its training; the model's performance in these cases (16/19 correct) was not statistically significantly different from that for previously encountered lesions (P = 0.41).

CONCLUSIONS

A previously trained DL algorithm had higher sensitivity than initial clinical assessment in detecting CHD in a cohort in which over 50% of CHD cases were initially missed clinically. Notably, the DL algorithm performed well on community-acquired images in a low-risk population, including lesions to which it had not been exposed previously. Furthermore, when both the model and blinded human experts had access to only stored images and not the full range of images available to a clinician during a live scan, the model outperformed the human experts. Together, these findings support the proposition that use of DL models can improve prenatal detection of CHD. © 2023 International Society of Ultrasound in Obstetrics and Gynecology.

The comprehensive study on the role of POSTN in fetal congenital heart disease and clinical applications

BACKGROUND

Congenital heart defect (CHD) is the most common congenital abnormality, and it has long been a clinical and public health concern. Our previous findings have found Periostin (POSTN) and Pappalysin-1 (PAPPA) as potential biomarkers for fetal CHD. We aim to further elucidate POSTN's role in fetal heart development and explore the clinical applicability of POSTN and PAPPA as diagnostic marker for fetal CHD. This study is poised to establish a theoretical framework for mitigating the incidence of CHD and advance a novel approach for prenatal screening of fetal CHD.

METHODS

We verified differential expression of POSTN and PAPPA in gravida serum and fetal amniotic fluid based on our previous research. We established the Postn knockout mouse by CRISPR/Cas9 to investigate whether Postn deletion leads to cardiac abnormalities in mice. Besides, we explored the mechanism of POSTN on heart development through Postn knockout mouse model and cell experiments. Finally, we established the logistic regression model and decision curve analysis to evaluate the clinical utility of POSTN and PAPPA in fetal CHD.

RESULTS

We observed a significant decrease in POSTN and increase in PAPPA in the CHD group. Atrial septal defects occurred in Postn-/- and Postn± C57BL/6 fetal heart, while ventricular septal defects with aortic saddle were observed in Postn± C57BL/6 fetal heart. Disruption of the extracellular matrix (ECM) in cardiomyocytes and multiple abnormalities in cellular sub-organelles were observed in Postn knockout mice. POSTN may positively regulate cell behaviors and unsettle ECM via the TGFβ-Smad2/3 signaling pathway. The combination of serum biomarkers POSTN and PAPPA with Echocardiogram can enhance the diagnostic accuracy of CHD. Furthermore, the comprehensive model including POSTN, PAPPA, and two clinical indicators (NT and age) exhibits significantly higher predictive ability than the diagnosis group without the use of serum biomarkers or clinical indicators.

CONCLUSIONS

It is the first evidence that Postn deletion leads to cardiac developmental abnormalities in fetal mice. This may involve the regulation of the TGFβ signaling pathway. Importantly, POSTN and PAPPA possess clinical utility as noninvasive prenatal promising screening indicators of CHD.

The Lifelong Impact of Artificial Intelligence and Clinical Prediction Models on Patients With Tetralogy of Fallot

Medical advancements in the diagnosis, surgical techniques, perioperative care, and continued care throughout childhood have transformed the outlook for individuals with tetralogy of Fallot (TOF), improving survival and shifting the perspective towards lifelong care. However, with a growing population of survivors, longstanding challenges have been accentuated, and new challenges have surfaced, necessitating a re-evaluation of TOF care. Availability of prenatal diagnostics, insufficient information from traditional imaging techniques, previously unforeseen medical complications, and debates surrounding optimal timing and indications for reintervention are among the emerging issues. To address these challenges, the integration of artificial intelligence and machine learning holds great promise as they have the potential to revolutionize patient management and positively impact lifelong outcomes for individuals with TOF. Innovative applications of artificial intelligence and machine learning have spanned across multiple domains of TOF care, including screening and diagnosis, automated image processing and interpretation, clinical risk stratification, and planning and performing cardiac interventions. By embracing these advancements and incorporating them into routine clinical practice, personalized medicine could be delivered, leading to the best possible outcomes for patients. In this review, we provide an overview of these evolving applications and emphasize the challenges, limitations, and future potential for integrating them into clinical care.

Enhancing Fetal Anomaly Detection in Ultrasonography Images: A Review of Machine Learning-Based Approaches

Fetal development is a critical phase in prenatal care, demanding the timely identification of anomalies in ultrasound images to safeguard the well-being of both the unborn child and the mother. Medical imaging has played a pivotal role in detecting fetal abnormalities and malformations. However, despite significant advances in ultrasound technology, the accurate identification of irregularities in prenatal images continues to pose considerable challenges, often necessitating substantial time and expertise from medical professionals. In this review, we go through recent developments in machine learning (ML) methods applied to fetal ultrasound images. Specifically, we focus on a range of ML algorithms employed in the context of fetal ultrasound, encompassing tasks such as image classification, object recognition, and segmentation. We highlight how these innovative approaches can enhance ultrasound-based fetal anomaly detection and provide insights for future research and clinical implementations. Furthermore, we emphasize the need for further research in this domain where future investigations can contribute to more effective ultrasound-based fetal anomaly detection.

Prevalence and detection rate of major congenital heart disease in twin pregnancies in Denmark

OBJECTIVE

To investigate the national prevalence and prenatal detection rate (DR) of major congenital heart disease (mCHD) in twin pregnancies without twin-to-twin transfusion syndrome (TTTS)-associated CHD in a Danish population following a standardized prenatal screening program.

METHODS

This was a national registry-based study of data collected prospectively over a 10-year period. In Denmark, all women with a twin pregnancy are offered standardized screening and surveillance programs in addition to first- and second-trimester screening for aneuploidies and malformation, respectively: monochorionic (MC) twins every 2 weeks from gestational week 15 and dichorionic (DC) twins every 4 weeks from week 18. The data were retrieved from the Danish Fetal Medicine Database and included all twin pregnancies from 2009-2018, in which at least one fetus had a pre- and/or postnatal mCHD diagnosis. mCHD was defined as CHD requiring surgery within the first year of life, excluding ventricular septal defects. All pregnancy data were pre- and postnatally validated in the local patient files at the four tertiary centers covering the entire country.

RESULTS

A total of 60 cases from 59 twin pregnancies were included. The prevalence of mCHD was 4.6 (95% CI, 3.5-6.0) per 1000 twin pregnancies (1.9 (95% CI, 1.3-2.5) per 1000 live births). The prevalences for DC and MC were 3.6 (95% CI, 2.6-5.0) and 9.2 (95% CI, 5.8-13.7) per 1000 twin pregnancies, respectively. The national prenatal DR of mCHD in twin pregnancies for the entire period was 68.3%. The highest DRs were in cases with univentricular hearts (100%) and the lowest with aortopulmonary window, total anomalous pulmonary venous return, Ebstein's anomaly, aortic valve stenosis and coarctation of the aorta (0-25%). Mothers of children with prenatally undetected mCHD had a significantly higher body mass index (BMI) compared to mothers of children with a prenatally detected mCHD (median, 27 kg/m2 and 23 kg/m2 , respectively; P = 0.02).

CONCLUSIONS

The prevalence of mCHD in twins was 4.6 per 1000 pregnancies and was higher in MC than DC pregnancies. The prenatal DR of mCHD in twin pregnancies was 68.3%. Maternal BMI was higher in cases of prenatally undetected mCHD. © 2023 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Classification of normal and abnormal fetal heart ultrasound images and identification of ventricular septal defects based on deep learning

OBJECTIVES

Congenital heart defects (CHDs) are the most common birth defects. Recently, artificial intelligence (AI) was used to assist in CHD diagnosis. No comparison has been made among the various types of algorithms that can assist in the prenatal diagnosis.

METHODS

Normal and abnormal fetal ultrasound heart images, including five standard views, were collected according to the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) Practice guidelines. You Only Look Once version 5 (YOLOv5) models were trained and tested. An excellent model was screened out after comparing YOLOv5 with other classic detection methods.

RESULTS

On the training set, YOLOv5n performed slightly better than the others. On the validation set, YOLOv5n attained the highest overall accuracy (90.67 %). On the CHD test set, YOLOv5n, which only needed 0.007 s to recognize each image, had the highest overall accuracy (82.93 %), and YOLOv5l achieved the best accuracy on the abnormal dataset (71.93 %). On the VSD test set, YOLOv5l had the best performance, with a 92.79 % overall accuracy rate and 92.59 % accuracy on the abnormal dataset. The YOLOv5 models achieved better performance than the Fast region-based convolutional neural network (RCNN) & ResNet50 model and the Fast RCNN & MobileNetv2 model on the CHD test set (p<0.05) and VSD test set (p<0.01).

CONCLUSIONS

YOLOv5 models are able to accurately distinguish normal and abnormal fetal heart ultrasound images, especially with respect to the identification of VSD, which have the potential to assist ultrasound in prenatal diagnosis.

Semi-Automatic Measurement of Fetal Cardiac Axis in Fetuses with Congenital Heart Disease (CHD) with Fetal Intelligent Navigation Echocardiography (FINE)

Congenital heart disease (CHD) is one of the most common organ-specific birth defects and a major cause of infant morbidity and mortality. Despite ultrasound screening guidelines, the detection rate of CHD is limited. Fetal intelligent navigation echocardiography (FINE) has been introduced to extract reference planes and cardiac axis from cardiac spatiotemporal image correlation (STIC) volume datasets. This study analyses the cardiac axis in fetuses affected by CHD/thoracic masses (n = 545) compared to healthy fetuses (n = 1543) generated by FINE. After marking seven anatomical structures, the FINE software generated semi-automatically nine echocardiography standard planes and calculated the cardiac axis. Our study reveals that depending on the type of CHD, the cardiac axis varies. In approximately 86% (471 of 542 volumes) of our pathological cases, an abnormal cardiac axis (normal median = 40-45°) was detectable. Significant differences between the fetal axis of the normal heart versus CHD were detected in HLHS, pulmonary atresia, TOF (p-value < 0.0001), RAA, situs ambiguus (p-value = 0.0001-0.001) and absent pulmonary valve syndrome, DORV, thoracic masses (p-value = 0.001-0.01). This analysis confirms that in fetuses with CHD, the cardiac axis can significantly deviate from the normal range. FINE appears to be a valuable tool to identify cardiac defects.

Evaluation of an artificial intelligent algorithm (Heartassist™) to automatically assess the quality of second trimester cardiac views: a prospective study

OBJECTIVES

The aim of this study was to evaluate the agreement between visual and automatic methods in assessing the adequacy of fetal cardiac views obtained during second trimester ultrasonographic examination.

METHODS

In a prospective observational study frames of the four-chamber view left and right outflow tracts, and three-vessel trachea view were obtained from 120 consecutive singleton low-risk women undergoing second trimester ultrasound at 19-23 weeks of gestation. For each frame, the quality assessment was performed by an expert sonographer and by an artificial intelligence software (Heartassist™). The Cohen's κ coefficient was used to evaluate the agreement rates between both techniques.

RESULTS

The number and percentage of images considered adequate visually by the expert or with Heartassist™ were similar with a percentage >87 % for all the cardiac views considered. The Cohen's κ coefficient values were for the four-chamber view 0.827 (95 % CI 0.662-0.992), 0.814 (95 % CI 0.638-0.990) for left ventricle outflow tract, 0.838 (95 % CI 0.683-0.992) and three vessel trachea view 0.866 (95 % CI 0.717-0.999), indicating a good agreement between the two techniques.

CONCLUSIONS

Heartassist™ allows to obtain the automatic evaluation of fetal cardiac views, reached the same accuracy of expert visual assessment and has the potential to be applied in the evaluation of fetal heart during second trimester ultrasonographic screening of fetal anomalies.

The Left Outflow Tract in Fetal Cardiac Screening Examination: Introduction of Quality Criteria Is Not Always Associated With an Improvement of Practice When Supervised by Humans

OBJECTIVES

Since 2016, the French CNEOF included the left ventricular outflow tract (LVOT) in the second and third trimester of pregnancy in addition to the four-chamber view and the parasagittal view of the right outflow tract. The objective of this study was to define quality criteria for fetal LVOT assessment and to perform a human audit of past and current practices, before and after the implementation of those quality criteria at a large scale.

METHODS

Seven quality criteria were investigated and rated from 0 to 1 during three periods of interest. Files were randomly selected from three centers, and average total and specific scores were calculated.

RESULTS

LVOT pictures were present in more than 94.3% of reports. The average quality score was 5.49/7 (95% confidence interval [CI]: 5.36-5.62), 5.91/7 (95% CI: 5.80-6.03), and 5.70/7 (95% CI: 5.58-5.82) for the three centers in the three periods of interest. There was no significant difference following the introduction of the quality criteria, 2017 versus 2020, P = .054.

CONCLUSION

Fetal LVOT images were present in most of ultrasound reports but the introduction of the proposed quality criteria under human supervision seems not associated with a significant change in practice.

First-Trimester Ultrasound Imaging for Prenatal Assessment of the Extended Cardiovascular System Using the Cardiovascular System Sonographic Evaluation Algorithm (CASSEAL)

INTRODUCTION AND OBJECTIVES

To compare fetal images obtained at the first- and second-trimester ultrasound scan when applying the Cardiovascular System Sonographic Evaluation Algorithm (CASSEAL).

METHODS

Using the CASSEAL protocol, nine sequential axial views were acquired in B-mode and color Doppler at the first- and second-trimester ultrasound scans, visualizing the main components of the extended fetal cardiovascular system. Images were compared qualitatively between both trimesters.

RESULTS

We obtained comparable images for all the nine axial views described in the CASSEAL protocol, with few differences and limitations.

CONCLUSIONS

The CASSEAL protocol is reproducible in the first trimester, and could help in the early detection of fetal cardiovascular abnormalities. It represents a promising additional tool in order to increase the CHD detection rate.

Controlled Prospective Study on the Use of Systematic Simulator-Based Training with a Virtual, Moving Fetus for Learning Second-Trimester Scan: FESIM III

OBJECTIVES

To analyze the feasibility of structured ultrasound simulation training (SIM-UT) in teaching second-trimester ultrasound screening using a high-end simulator with a randomly moving fetus.

METHODS

This was a prospective, controlled trial. A trial group of 11 medical students with minimal obstetric ultrasound experience underwent 12 hours of structured SIM-UT in individual hands-on sessions within 6 weeks. Learning progress was assessed with standardized tests. Performance after 2, 4, and 6 weeks of SIM-UT was compared with two reference groups ((A) Ob/Gyn residents and consultants, and (B) highly skilled DEGUM experts). Participants were asked to acquire 23 2nd trimester planes according to ISUOG guidelines in a realistic simulation B-mode with a randomly moving fetus as quickly as possible within a 30-minute time frame. All tests were analyzed regarding the rate of appropriately obtained images and the total time to completion (TTC).

RESULTS

During the study, novices were able to improve their ultrasound skills significantly, reaching the physician level of the reference group (A) after 8 hours of training. After 12 hours of SIM-UT, the trial group performed significantly faster than the physician group (TTC: 621±189 vs. 1036±389 sec., p=0.011). Novices obtained 20 out of 23 2nd trimester standard planes without a significant time difference when compared to experts. TTC of the DEGUM reference group remained significantly faster (p<0.001) though.

CONCLUSION

SIM-UT on a simulator with a virtual, randomly moving fetus is highly effective. Novices can obtain standard plane acquisition skills close to expert level within 12 hours of self-training.

Use of artificial intelligence and deep learning in fetal ultrasound imaging

Deep learning is considered the leading artificial intelligence tool in image analysis in general. Deep-learning algorithms excel at image recognition, which makes them valuable in medical imaging. Obstetric ultrasound has become the gold standard imaging modality for detection and diagnosis of fetal malformations. However, ultrasound relies heavily on the operator's experience, making it unreliable in inexperienced hands. Several studies have proposed the use of deep-learning models as a tool to support sonographers, in an attempt to overcome these problems inherent to ultrasound. Deep learning has many clinical applications in the field of fetal imaging, including identification of normal and abnormal fetal anatomy and measurement of fetal biometry. In this Review, we provide a comprehensive explanation of the fundamentals of deep learning in fetal imaging, with particular focus on its clinical applicability. © 2022 International Society of Ultrasound in Obstetrics and Gynecology.

Shone Complex: A Case Report of Congenital Heart Disease Detected Using Point-of-care Ultrasound

INTRODUCTION

Undiagnosed congenital heart disease and management of pediatric cardiogenic shock presents a diagnostic challenge for the emergency clinician. These diagnoses are rare and require a high index of suspicion given the overlap with more common pediatric pathology. Point-of-care ultrasound can assist in differentiating these presentations. We present a case of neonatal cardiogenic shock secondary to a previously undiagnosed congenital heart disease, specifically Shone complex, detected using point-of-care ultrasound.

CASE REPORT

A six-week-old female presented with severe respiratory distress and was found to be in cardiogenic shock secondary to a previously undiagnosed congenital heart defect.

CONCLUSION

Initial diagnosis of congenital heart disease is uncommon in the emergency department, but it should be recognized by clinicians given the high associated morbidity and mortality. Point-of-care ultrasound is a powerful tool to assist in evaluating for cardiac abnormalities as an etiology for undifferentiated shock in the pediatric population.

Controlled prospective study on ultrasound simulation training in fetal echocardiography: FESIM II

PURPOSE

To analyze the learning curves of ultrasound novices in fetal echocardiography during structured simulation-based ultrasound training (SIM-UT) including a virtual, randomly moving fetus.

METHODS

11 medical students with minimal (< 10 h) prior obstetric ultrasound experience underwent 12 h of structured fetal echocardiography SIM-UT in individual hands-on sessions during a 6-week training program. Their learning progress was assessed with standardized tests after 2, 4, and 6 weeks of SIM-UT. Participants were asked to obtain 11 fetal echocardiography standard planes (in accordance with ISUOG and AHA guidelines) as quickly as possible. All tests were carried out under real life, examination-like conditions on a healthy, randomly moving fetus. Subsequently, we analyzed the rate of correctly obtained images and the total time to completion (TTC). As reference groups, 10 Ob/Gyn physicians (median of 750 previously performed Ob/Gyn scans) and 10 fetal echocardiography experts (median of 15,000 previously performed Ob/Gyn scans) were examined with the same standardized tests.

RESULTS

The students showed a consistent and steady improvement of their ultrasound performance during the training program. After 2 weeks, they were able to obtain > 95% of the standard planes correctly. After 6 weeks, they were significantly faster than the physician group (p < 0.001) and no longer significantly slower than the expert group (p = 0.944).

CONCLUSION

SIM-UT is highly effective to learn fetal echocardiography. Regarding the acquisition of the AHA/ISUOG fetal echocardiography standard planes, the students were able to reach the same skill level as the expert group within 6 weeks.

Evaluation fetal heart in the first and second trimester: Results and limitations

Background

Cardiac heart defects affect nearly 6-12 per 1000 live births in the general population and are more frequent than common trisomies.

Aim

To assess the efficacy and technical limitations of first-trimester fetal heart evaluation in the 11-14^th-weeks' scan and comparison with the second-trimester anatomical exam by ultrasound.

Material and Method

Between April 2015 and July 2020, medical records and ultrasound data of 3295 pregnancies who underwent first-trimester fetal anatomy exams by ultrasound were reviewed retrospectively. All ultrasound exams were performed by the same two operators (TUKD, OP) with transabdominal transducers. Fetal situs, four-chamber view, outflow tracts, and three-vessel trachea view are the cornerstones of first-trimester fetal heart examination. Conventional grayscale mode and high-definition power Doppler mode were utilized. The same operators re-examined all cases between the 18 and 23 weeks of gestation by ISUOG guidelines.

Results

We performed a combined transvaginal and transabdominal approach for only 101 cases (3.06%). The mean maternal age was 31.28 ± 4.43, the median gestational age at the first-trimester ultrasound exam was 12.4 weeks, and the median CRL was 61.87 mm (range was 45.1-84 mm). Even combined approach situs, cardiac axis, and four-chamber view could not be visualized optimally in 28 cases (0.7%). Outflow tracts were visualized separately in 80% (2636 in 3295) cases. Three vessel-trachea views were obtained in 85.4% (2814 in 3295) cases by high-definition Doppler mode. There were 47 fetuses with cardiac defects in 3295 pregnancies with the known pregnancy outcome. Ten cases had abnormal karyotype results. Thirty-two fetuses with cardiac anomalies (9.7 in 1000 pregnancies) were detected in the first-trimester examination, and the remaining 15 (4.55 in 1000 pregnancies) cases were diagnosed in the second-trimester examination. The prevalence of congenital cardiac anomalies was 14.25 in 1000 pregnancies. Fifteen cases were missed in the first-trimester exam. Also, ten fetuses which had abnormal cardiac findings in the first-trimester exam were not confirmed in the second-trimester exam. Sensitivity, specificity, positive, and negative predictive values were calculated as 65.3%, 99.7%, 66.8%, and 99.67%, respectively.

Conclusion

Late first-trimester examination of the fetus is feasible and allows earlier detection of many structural abnormalities of the fetus, including congenital heart defects. Suspicious and isolated cardiac abnormal findings should be re-examined and confirmed in the second-trimester exam. Previous abdominal surgery, high BMI, and subtle cardiac defects can cause missed cardiac abnormalities.

A Diagnostic Dilemma: Transposition of the Great Arteries

Transposition of the great arteries (TGA) remains one of the most common and severe underdiagnosed congenital cardiac anomalies in the prenatal period. Unfortunately, despite advances in prenatal ultrasound screening, the detection rate of major congenital heart defects (CHDs) remains low. We present the case of a preterm male infant delivered limp with generalized cyanosis and in respiratory distress at 36 weeks gestation with postnatal echocardiography (ECHO) depicting dextro-TGA (d-TGA). Maternal prenatal targeted fetal anomaly ultrasonography at 18 weeks gestation showed abnormal right ventricle and right ventricular outflow tract. Subsequent two-time repeat fetal ECHO showed ventricular septal defect. This case represents how challenging and unrecognized critical CHDs can be. Furthermore, it highlights the need for clinicians to have a high index of suspicion when newborns present with clinical manifestations of critical CHDs and manage it accordingly to avoid severe complications.

Fetal arrhythmias case series: Experiences from a fetal screening center in Taiwan

OBJECTIVE

Fetal arrhythmias are common and in rare cases can be associated with severe mortality and morbidity. Most existing articles are focused on classification of fetal arrhythmias in referral centers. Our main objective was to analyze types, clinical characteristics, and outcomes for arrhythmia cases in general practice.

CASE REPORT

We retrospectively reviewed a case series of fetal arrhythmias in a fetal medicine clinic between September 2017 and August 2021.

FETAL ARRHYTHMIAS IN OUR SAMPLE PRESENTED BY

Ectopies (86%, n = 57), bradyarrhythmias (11%, n = 7), and tachyarrhythmias (3%, n = 2). One tachyarrhythmia case was associated with Ebstein's anomaly. Two cases of second-degree AV block received transplacental fluorinated steroid therapy with recovery of fetal cardiac rhythm in later gestation. One case of complete AV block developed hydrops fetalis.

CONCLUSION

Detection and careful stratification of fetal arrhythmias in obstetric screening is crucial. While most arrhythmias are benign and self-limited, some require prompt referral and timely intervention.

Trends in Delayed Diagnosis of Critical Congenital Heart Defects in an Era of Enhanced Screening, 2004-2018

OBJECTIVE

To describe trends in delayed diagnosis of critical congenital heart defects (CCHDs) with prenatal and postnatal screening advances.

STUDY DESIGN

We evaluated a retrospective cohort of live births with CCHD delivered between 2004 and 2018 from a statewide, population-based birth defects surveillance system in Massachusetts. Demographic information were obtained from vital records. We estimated timely (prenatal or birth/transfer hospital) and delayed diagnosis (after discharge) proportions by year and time periods coinciding with the transition to mandatory pulse oximetry in 2015.

RESULTS

We identified 1524 eligible CCHD cases among 1 087 027 live births. By 2018, 92% of cases received a timely diagnosis, most prenatally. From 2004 to 2018, prenatal diagnosis increased from 46% to 76% of cases, while hospital diagnosis decreased from 38% to 17%, and delayed diagnosis declined from 16% to 7%. These trends were consistent across all characteristics evaluated. Among cases without a prenatal diagnosis, the proportion with delayed diagnosis did not change over time, even after implementation of mandatory pulse oximetry screening. Prenatal detection increased the most among severe cases (treated or died in first month of life). Well-appearing newborns without prenatal diagnosis made up 79% of delayed diagnosis cases by 2015-2018. Delayed diagnosis was most common for coarctation.

CONCLUSIONS

While prenatal diagnosis of CCHD increased dramatically, there was no reduction in delayed diagnosis among postnatally diagnosed infants, even after pulse oximetry screening became mandatory. Pulse oximetry may not reduce delayed diagnosis in settings with high prenatal detection, and other strategies are needed to ensure timely diagnosis of well-appearing newborns.

Are there foetal extracellular vesicles in maternal blood? Prospects for diagnostic biomarker discovery

Prenatal diagnosis of congenital disease improves clinical outcomes; however, as many as 50% of congenital heart disease cases are missed by current ultrasound screening methods. This indicates a need for improved screening technology. Extracellular vesicles (EVs) have attracted enormous interest in recent years for their potential in diagnostics. EVs mediate endocrine signalling in health and disease and are known to regulate aspects of embryonic development. Here, we critically evaluate recent evidence suggesting that EVs released from the foetus are able to cross the placenta and enter the maternal circulation. Furthermore, EVs from the mother appear to be transported in the reverse direction, whilst the placenta itself acts as a source of EVs. Experimental work utilising rodent models employing either transgenically encoded reporters or application of fluorescent tracking dyes provide convincing evidence of foetal-maternal crosstalk. This is supported by clinical data demonstrating expression of placental-origin EVs in maternal blood, as well as limited evidence for the presence of foetal-origin EVs. Together, this work raises the possibility that foetal EVs present in maternal blood could be used for the diagnosis of congenital disease. We discuss the challenges faced by researchers in translating these basic science findings into a clinical non-invasive prenatal test.

Improvement of diagnostic efficiency in fetal congenital heart disease using fetal intelligent navigation echocardiography by less-experienced operators

OBJECTIVE

This study investigated the feasibility and accuracy of fetal intelligent navigation echocardiography (FINE) for the prenatal diagnosis of congenital heart disease (CHD) by inexperienced and experienced operators.

METHOD

In this prospective study, all volume data sets from 120 fetuses with a broad spectrum of CHD were acquired using spatiotemporal image correlation technology. The prenatal diagnostic procedures were performed by two operators with different experience (beginner: 1 year and expert: 15 years) using FINE and traditional fetal echocardiography. Data were analyzed on the time of examination and acquisition of results.

RESULTS

Diagnoses made by FINE and traditional echocardiography were completely consistent with the final diagnosis of CHD in 98 (81.66%) versus 20 (16.66%) (P < 0.001) beginners and 87.50% (n = 105) versus 101 (84.16%) experts, respectively. On the contrary, there was significant difference using traditional echocardiography (16.66% versus 84.16%, P < 0.001) by two examiners. Furthermore, the examination time decreased when using FINE compared with using traditional echocardiography (beginner operators: 4.54 ± 1.03 min versus 20.58 ± 3.36 min, P < 0.001; expert operators: 3.89 ± 0.96 min versus 12.73 ± 1.62 min, P < 0.001).

CONCLUSION

Based on our results, a prenatal diagnosis of CHD can be made with high feasibility and accuracy using FINE compared with traditional fetal echocardiography for beginner operators.

Prenatal detection of aortic coarctation in a well-organized screening setting: Are we there yet?

OBJECTIVE

We aimed to assess current prenatal detection rate (DR) of aortic coarctation (CoA) and its impact on neonatal outcome in the Netherlands to evaluate the efficacy of the Dutch screening protocol in which the cardiac four-chamber view, outflow tracts and three-vessel view are compulsory.

METHODS

All prenatally and postnatally diagnosed CoA cases between 2012 and 2021 were extracted from our PRECOR-registry. Annual DRs were calculated with a focus on the trend over time and attributing factors for detection. Postnatal outcome was compared between prenatally detected and undetected cases.

RESULTS

49/116 cases (42.2%) were detected prenatally. A higher chance of detection was found for cases with extracardiac malformations (71.4%; p = 0.001) and the more severe cases with an aortic arch hypoplasia and/or ventricular septal defect (63.2%; p = 0.001). Time-trend analysis showed no improvement in DR over time (p = 0.33). Undetected cases presented with acute circulatory shock in 20.9% and were more likely to have severe lactic acidosis (p = 0.02) and impaired cardiac function (p < 0.001) before surgery.

CONCLUSION

Even in a well-organized screening program, the DR of CoA still requires improvement, especially in isolated cases. The increased risk of severe lactic acidosis in undetected cases stresses the need for urgent additions to the current screening program, such as implementation of the three-vessel trachea view and measurement of outflow tracts.

Prenatally diagnosed congenital heart disease: the cost of maternal care

OBJECTIVE

Little is known regarding the effects of a prenatal diagnosis of congenital heart disease (CHD) on the cost of antenatal and delivery care. We sought to compare the maternal costs of care in pregnancies where the fetus or child was diagnosed prenatally vs. postnatally.

METHODS

Costs of maternal care were determined for pregnancies in which the fetus or child was diagnosed with CHD between 1997 and 2012 in the state of Utah. Cases of CHD were identified via a statewide birth defect surveillance program which included data on the timing of diagnosis, maternal demographic and clinical data, and linked to statewide inpatient maternal hospital discharge records. Antenatal testing costs were determined using Medicaid fee estimates and total facility costs were determined for all hospitalizations including delivery. The association of timing of diagnosis of CHD with costs was analyzed using univariable and multivariable models.

RESULTS

Of 2128 pregnancies included in the study, 36% had a fetus prenatally diagnosed with CHD. The prenatal diagnosis group was more likely to have a termination or stillbirth and were younger at delivery (gestational age 37.3 vs 38.0 weeks, p < .001). Labor induction and cesarean delivery rates were similar between groups. Antenatal testing and delivery hospitalization costs were higher in the prenatal diagnosis group: $5819 vs $4041 (p < .001) and $10,509 vs $7802 (p < .001), respectively. Patients in the prenatal diagnosis group had longer lengths of hospital stays (3.5 vs 2.4 d, p > .001). After controlling for significant differences between the groups, including lesion severity, the prenatal diagnosis remained directly associated with antenatal testing costs (+$1472), maternal hospitalization costs (+$2713), and maternal hospital length of stay (+1.0 d).

CONCLUSION

A prenatal diagnosis of fetal CHD was associated with increased prenatal costs, hospitalization costs, and hospital length of stay for affected pregnant patients.

Artificial intelligence in the pediatric echocardiography laboratory: Automation, physiology, and outcomes

Artificial intelligence (AI) is frequently used in non-medical fields to assist with automation and decision-making. The potential for AI in pediatric cardiology, especially in the echocardiography laboratory, is very high. There are multiple tasks AI is designed to do that could improve the quality, interpretation, and clinical application of echocardiographic data at the level of the sonographer, echocardiographer, and clinician. In this state-of-the-art review, we highlight the pertinent literature on machine learning in echocardiography and discuss its applications in the pediatric echocardiography lab with a focus on automation of the pediatric echocardiogram and the use of echo data to better understand physiology and outcomes in pediatric cardiology. We also discuss next steps in utilizing AI in pediatric echocardiography.

Impact of Socioeconomic Status and Residence Distance on Infant Heart Disease Outcomes in Canada

Background Socioeconomic status (SES) impacts clinical outcomes associated with severe congenital heart disease (sCHD). We examined the impact of SES and remoteness of residence (RoR) on congenital heart disease (CHD) outcomes in Canada, a jurisdiction with universal health insurance. Methods and Results All infants born in Canada (excluding Quebec) from 2008 to 2018 and hospitalized with CHD requiring intervention in the first year were identified. Neighborhood level SES income quintiles were calculated, and RoR was categorized as residing <100 km, 100 to 299 km, or >300 km from the closest of 7 cardiac surgical programs. In-hospital mortality at <1 year was the primary outcome, adjusted for preterm birth, low birth weight, and extracardiac pathology. Among 7711 infants, 4485 (58.2%) had moderate CHD (mCHD) and 3226 (41.8%) had sCHD. Overall mortality rate was 10.5%, with higher rates in sCHD than mCHD (13.3% versus 8.5%, respectively). More CHD infants were in the lowest compared with the highest SES category (27.1% versus 15.0%, respectively). The distribution of CHD across RoR categories was 52.3%, 21.3%, and 26.4% for <100 km, 100 to 299 km, and >300 km, respectively. Although SES and RoR had no impact on sCHD mortality, infants with mCHD living >300 km had a higher risk of mortality relative to those living <100 km (adjusted odds ratio [aOR], 1.43 [95% CI, 1.11-1.84]). Infants with mCHD within the lowest SES quintile and living farthest away had the highest risk for mortality (aOR, 1.74 [95% CI, 1.08-2.81]). Conclusions In Canada, neither RoR nor SES had an impact on outcomes of infants with sCHD. Greater RoR, however, may contribute to higher risk of mortality among infants with mCHD.

Circulating microRNA: Myocardium-derived prenatal biomarker of ventricular septal defects

Background: Recently, circulating microRNAs (miRNAs) from maternal blood and amniotic fluid have been used as biomarkers for ventricular septal defect (VSD) diagnosis. However, whether circulating miRNAs are associated with fetal myocardium remains unknown.

Methods: Dimethadione (DMO) induced a VSD rat model. The miRNA expression profiles of the myocardium, amniotic fluid and maternal serum were analyzed. Differentially expressed microRNAs (DE-microRNAs) were verified by qRT–PCR. The target gene of miR-1-3p was confirmed by dual luciferase reporter assays. Expression of amniotic fluid-derived DE-microRNAs was verified in clinical samples.

Results: MiRNAs were differentially expressed in VSD fetal rats and might be involved in cardiomyocyte differentiation and apoptosis. MiR-1-3p, miR-1b and miR-293-5p were downregulated in the myocardium and upregulated in amniotic fluid/maternal serum. The expression of amniotic fluid-derived DE-microRNAs (miR-1-3p, miR-206 and miR-184) was verified in clinical samples. Dual luciferase reporter assays confirmed that miR-1-3p directly targeted SLC8A1/NCX1.

Conclusion: MiR-1-3p, miR-1b and miR-293-5p are downregulated in VSD myocardium and upregulated in circulation and may be released into circulation by cardiomyocytes. MiR-1-3p targets SLC8A1/NCX1 and participates in myocardial apoptosis. MiR-1-3p upregulation in circulation is a direct and powerful indicator of fetal VSD and is expected to serve as a prenatal VSD diagnostic marker.

Artificial intelligence in perinatal diagnosis and management of congenital heart disease

Prenatal diagnosis and management of congenital heart disease (CHD) has progressed substantially in the past few decades. Fetal echocardiography can accurately detect and diagnose approximately 85% of cardiac anomalies. The prenatal diagnosis of CHD results in improved care, with improved risk stratification, perioperative status and survival. However, there is much work to be done. A minority of CHD is actually identified prenatally. This seemingly incongruous gap is due, in part, to diminished recognition of an anomaly even when present in the images and the need for increased training to obtain specialized cardiac views. Artificial intelligence (AI) is a field within computer science that focuses on the development of algorithms that "learn, reason, and self-correct" in a human-like fashion. When applied to fetal echocardiography, AI has the potential to improve image acquisition, image optimization, automated measurements, identification of outliers, classification of diagnoses, and prediction of outcomes. Adoption of AI in the field has been thus far limited by a paucity of data, limited resources to implement new technologies, and legal and ethical concerns. Despite these barriers, recognition of the potential benefits will push us to a future in which AI will become a routine part of clinical practice.

Disparities in surgical outcomes of neonates with congenital heart disease across regions, centers, and populations

OBJECTIVE

To summarize existing literature on neonatal disparities in congenital heart disease surgical outcomes and identify potential policies to address these disparities.

FINDING

Disparities in outcomes for neonatal congenital heart surgery were largely published under four domains: race/ethnicity, insurance type, neighborhood/socioeconomic status, and cardiac center characteristics. While existing research identifies associations between these domains and mortality, more nuanced analyses are emerging to understand the mediators between these domains and other non-mortality outcomes, as well as potential interventions and policies to reduce disparities. A broader look into social determinants of health (SDOH), prenatal diagnosis, proximity of birth to a cardiac surgical center, and post-surgical outpatient and neurodevelopmental follow-up may accelerate interventions to mitigate disparities in outcomes.

CONCLUSION

Understanding the mechanisms of how SDOH relate to neonatal surgical outcomes is paramount, as disparities research in neonatal congenital heart surgery continues to shift from identification and description, to intervention and policy.

Incremental Detection of Severe Congenital Heart Disease by Fetal Echocardiography Following a Normal Second Trimester Ultrasound Scan in Québec, Canada

Background:

The benefit of fetal echocardiograms (FE) to detect severe congenital heart diseases (SCHD) in the setting of a normal second-trimester ultrasound is unclear. We aimed to assess whether the increase in SCHD detection rates when FE are performed for risk factors in the setting of a normal ultrasound was clinically significant to justify the resources needed.

Methods:

This is a multicenter, population-based, retrospective cohort study, including all singleton pregnancies and offspring in Quebec (Canada) between 2007 and 2015. Administrative health care data were linked with FE clinical data to gather information on prenatal diagnosis of CHD, indications for FE, outcomes of pregnancy and offspring, postnatal diagnosis of CHD, cardiac interventions, and causes of death. The difference between the sensitivity to detect SCHD with and without FE for risk factors was calculated using generalized estimating equations with a noninferiority margin of 5 percentage points.

Results:

A total of 688 247 singleton pregnancies were included, of which 30 263 had at least one FE. There were 1564 SCHD, including 1071 that were detected prenatally (68.5%). There were 12 210 FE performed for risk factors in the setting of a normal second-trimester ultrasound, which led to the detection of 49 additional cases of SCHD over 8 years. FE referrals for risk factors increased sensitivity by 3.1 percentage points (95% CI, 2.3–4.0; P<0.0001 for noninferiority).

Conclusions:

In the setting of a normal second-trimester ultrasound, adding a FE for risk factors offered low incremental value to the detection rate of SCHD in singleton pregnancies. The current ratio of clinical gains versus the FE resources needed to screen for SCHD in singleton pregnancies with isolated risk factors does not seem favorable. Further studies should evaluate whether these resources could be better allocated to increase SCHD sensitivity at the ultrasound level, and to help decrease heterogeneity between regions, institutions and operators.

Application of machine learning in screening for congenital heart diseases using fetal echocardiography

There is a growing body of literature supporting the utilization of machine learning (ML) to improve diagnosis and prognosis tools of cardiovascular disease. The current study was to investigate the impact that the ML framework may have on the sensitivity of predicting the presence or absence of congenital heart disease (CHD) using fetal echocardiography. A comprehensive fetal echocardiogram including 2D cardiac chamber quantification, valvar assessments, assessment of great vessel morphology, and Doppler-derived blood flow interrogation was recorded. The postnatal echocardiogram was used to ascertain the diagnosis of CHD. A random forest (RF) algorithm with a nested tenfold cross-validation was used to train models for assessing the presence of CHD. The study population was derived from a database of 3910 singleton fetuses with maternal age of 28.8 ± 5.2 years and gestational age at the time of fetal echocardiography of 22.0 weeks (IQR 21-24). The proportion of CHD was 14.1% for the studied cohort confirmed by post-natal echocardiograms. Our proposed RF-based framework provided a sensitivity of 0.85, a specificity of 0.88, a positive predictive value of 0.55 and a negative predictive value of 0.97 to detect the CHD with the mean of mean ROC curves of 0.94 and the mean of mean PR curves of 0.84. Additionally, six first features, including cardiac axis, peak velocity of blood flow across the pulmonic valve, cardiothoracic ratio, pulmonary valvar annulus diameter, right ventricular end-diastolic diameter, and aortic valvar annulus diameter, are essential features that play crucial roles in adding more predictive values to the model in detecting patients with CHD. ML using RF can provide increased sensitivity in prenatal CHD screening with very good performance. The incorporation of ML algorithms into fetal echocardiography may further standardize the assessment for CHD.

Ventilatory management of critically ill children in the emergency setting, during transport and retrieval

Critical illness in children is uncommon. The acute stabilization and resuscitation of critically ill children remains challenging to even the most experienced operator. Cardiorespiratory illness represents the largest subgroup of diseases causing critical illness and, thus adds a layer of complexity and additional challenge to the safe intubation and establishment of effective ventilation of this group of children. Children have unique physiological and anatomical differences to adults, and present the team involved in their resuscitation and stabilization with challenges exaggerated by critical illness. The consideration of pathophysiological implications of disease and the equipment available during transport and retrieval from the roadside or nonspecialist setting to pediatric intensive care allows the clinician involved in resuscitation, stabilization, and establishment of ventilation to employ targeted strategies to optimize ventilatory success. This review focuses on the types of ventilatory challenges that must be addressed when managing critically ill children in the local settings in which they present, and the resources available to optimize the outcome prior to and during transfer to a higher level of care.

First-trimester ultrasound detection of fetal heart anomalies: systematic review and meta-analysis

OBJECTIVES

To determine the diagnostic accuracy of ultrasound at 11-14 weeks' gestation in the detection of fetal cardiac abnormalities and to evaluate factors that impact the detection rate.

METHODS

This was a systematic review of studies evaluating the diagnostic accuracy of ultrasound in the detection of fetal cardiac anomalies at 11-14 weeks' gestation, performed by two independent reviewers. An electronic search of four databases (MEDLINE, EMBASE, Web of Science Core Collection and The Cochrane Library) was conducted for studies published between January 1998 and July 2020. Prospective and retrospective studies evaluating pregnancies at any prior level of risk and in any healthcare setting were eligible for inclusion. The reference standard used was the detection of a cardiac abnormality on postnatal or postmortem examination. Data were extracted from the included studies to populate 2 × 2 tables. Meta-analysis was performed using a random-effects model in order to determine the performance of first-trimester ultrasound in the detection of major cardiac abnormalities overall and of individual types of cardiac abnormality. Data were analyzed separately for high-risk and non-high-risk populations. Preplanned secondary analyses were conducted in order to assess factors that may impact screening performance, including the imaging protocol used for cardiac assessment (including the use of color-flow Doppler), ultrasound modality, year of publication and the index of sonographer suspicion at the time of the scan. Risk of bias and quality assessment were undertaken for all included studies using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool.

RESULTS

The electronic search yielded 4108 citations. Following review of titles and abstracts, 223 publications underwent full-text review, of which 63 studies, reporting on 328 262 fetuses, were selected for inclusion in the meta-analysis. In the non-high-risk population (45 studies, 306 872 fetuses), 1445 major cardiac anomalies were identified (prevalence, 0.41% (95% CI, 0.39-0.43%)). Of these, 767 were detected on first-trimester ultrasound examination of the heart and 678 were not detected. First-trimester ultrasound had a pooled sensitivity of 55.80% (95% CI, 45.87-65.50%), specificity of 99.98% (95% CI, 99.97-99.99%) and positive predictive value of 94.85% (95% CI, 91.63-97.32%) in the non-high-risk population. The cases diagnosed in the first trimester represented 63.67% (95% CI, 54.35-72.49%) of all antenatally diagnosed major cardiac abnormalities in the non-high-risk population. In the high-risk population (18 studies, 21 390 fetuses), 480 major cardiac anomalies were identified (prevalence, 1.36% (95% CI, 1.20-1.52%)). Of these, 338 were detected on first-trimester ultrasound examination and 142 were not detected. First-trimester ultrasound had a pooled sensitivity of 67.74% (95% CI, 55.25-79.06%), specificity of 99.75% (95% CI, 99.47-99.92%) and positive predictive value of 94.22% (95% CI, 90.22-97.22%) in the high-risk population. The cases diagnosed in the first trimester represented 79.86% (95% CI, 69.89-88.25%) of all antenatally diagnosed major cardiac abnormalities in the high-risk population. The imaging protocol used for examination was found to have an important impact on screening performance in both populations (P < 0.0001), with a significantly higher detection rate observed in studies using at least one outflow-tract view or color-flow Doppler imaging (both P < 0.0001). Different types of cardiac anomaly were not equally amenable to detection on first-trimester ultrasound.

CONCLUSIONS

First-trimester ultrasound examination of the fetal heart allows identification of over half of fetuses affected by major cardiac pathology. Future first-trimester screening programs should follow structured anatomical assessment protocols and consider the introduction of outflow-tract views and color-flow Doppler imaging, as this would improve detection rates of fetal cardiac pathology. © 2021 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Maternal serum-derived exosomal lactoferrin as a marker in detecting and predicting ventricular septal defect in fetuses

Among different types of congenital heart diseases, ventricular septal defect is the most frequently diagnosed type and is frequently missed in early prenatal screening programs. Herein, we explored the role of maternal serum-derived exosomes in detecting and predicting ventricular septal defect in fetuses in the early stage of pregnancy. A total of 104 pregnant women consisting of 52 ventricular septal defect cases and 52 healthy controls were recruited. TMT/iTRAQ proteomic analysis uncovered 15 maternal serum exosomal proteins, which showed differential expression between ventricular septal defect and control groups. Among these, four down-regulated proteins, lactoferrin, SBSN, DCD, and MBD3, were validated by Western blot. The protein lactoferrin was additionally verified by ELISA which was able to distinguish ventricular septal defects from controls with area under the ROC curve (AUC) 0.804 (p < 0.001). Our findings reveal that lactoferrin in maternal serum-derived exosomes may be a potential biomarker for non-invasive prenatal diagnosis of fetal ventricular septal defects.

Evaluation of Fetal Cardiac Size and Shape: A New Screening Tool to Identify Fetuses at Risk for Tetralogy of Fallot

OBJECTIVE

Prenatal detection rates for tetralogy of Fallot (TOF) vary between 23 and 85.7%, in part because of the absence of significant structural abnormalities of the 4-chamber view (4CV), as well as the relative difficulty in detection of abnormalities during the screening examination of the outflow tracts. The purpose of this study was to evaluate whether the 4CV and ventricles in fetuses with TOF may be characterized by abnormalities of size and shape of these structures.

METHODS

This study retrospectively evaluated 44 fetuses with the postnatal diagnosis of TOF. Measurements were made from the 4CV (end-diastolic length, width, area, global sphericity index, and cardiac axis) and the right (RV) and left (LV) ventricles (area, length, 24-segment transverse widths, sphericity index, and RV/LV ratios). Logistic regression analysis was performed to identify variables that might separate fetuses with TOF from normal controls.

RESULTS

The mean gestational age at the time of the last examination prior to delivery was 28 weeks 5 days (SD 4 weeks, 4 days). The mean z-scores were significantly lower in fetuses with TOF for the 4CV and RV and LV measurements of size and shape. Logistic regression analysis identified simple linear measurements of the 4CV, RV, and LV that had a sensitivity of 90.9 and specificity of 98.5% that outperformed the 4CV cardiac axis (sensitivity of 22.7%) as a screening tool for TOF.

CONCLUSIONS

Measurements of the 4CV, RV, and LV can be used as an adjunct to the outflow tract screening examination to identify fetuses with TOF.

The Use of Artificial Intelligence in Automation in the Fields of Gynaecology and Obstetrics - an Assessment of the State of Play

The long-awaited progress in digitalisation is generating huge amounts of medical data every day, and manual analysis and targeted, patient-oriented evaluation of this data is becoming increasingly difficult or even infeasible. This state of affairs and the associated, increasingly complex requirements for individualised precision medicine underline the need for modern software solutions and algorithms across the entire healthcare system. The utilisation of state-of-the-art equipment and techniques in almost all areas of medicine over the past few years has now indeed enabled automation processes to enter - at least in part - into routine clinical practice. Such systems utilise a wide variety of artificial intelligence (AI) techniques, the majority of which have been developed to optimise medical image reconstruction, noise reduction, quality assurance, triage, segmentation, computer-aided detection and classification and, as an emerging field of research, radiogenomics. Tasks handled by AI are completed significantly faster and more precisely, clearly demonstrated by now in the annual findings of the ImageNet Large-Scale Visual Recognition Challenge (ILSVCR), first conducted in 2015, with error rates well below those of humans. This review article will discuss the potential capabilities and currently available applications of AI in gynaecological-obstetric diagnostics. The article will focus, in particular, on automated techniques in prenatal sonographic diagnostics.

Exploring a new paradigm for the fetal anomaly ultrasound scan: Artificial intelligence in real time

OBJECTIVE

Advances in artificial intelligence (AI) have demonstrated potential to improve medical diagnosis. We piloted the end-to-end automation of the mid-trimester screening ultrasound scan using AI-enabled tools.

METHODS

A prospective method comparison study was conducted. Participants had both standard and AI-assisted US scans performed. The AI tools automated image acquisition, biometric measurement, and report production. A feedback survey captured the sonographers' perceptions of scanning.

RESULTS

Twenty-three subjects were studied. The average time saving per scan was 7.62 min (34.7%) with the AI-assisted method (p < 0.0001). There was no difference in reporting time. There were no clinically significant differences in biometric measurements between the two methods. The AI tools saved a satisfactory view in 93% of the cases (four core views only), and 73% for the full 13 views, compared to 98% for both using the manual scan. Survey responses suggest that the AI tools helped sonographers to concentrate on image interpretation by removing disruptive tasks.

CONCLUSION

Separating freehand scanning from image capture and measurement resulted in a faster scan and altered workflow. Removing repetitive tasks may allow more attention to be directed identifying fetal malformation. Further work is required to improve the image plane detection algorithm for use in real time.

Prenatal diagnosis and planned peripartum care improve perinatal outcome of fetuses with transposition of the great arteries and intact ventricular septum in low-resource settings

OBJECTIVE

To report on the feasibility of establishing a regional prenatal referral network for critical congenital heart defects (CHDs) and its impact on perinatal outcome of fetuses with transposition of the great arteries and intact ventricular septum (TGA-IVS) in low-resource settings.

METHODS

This was a retrospective study of consecutive fetuses with a diagnosis of TGA-IVS between January 2011 and December 2019 in Kochi, Kerala, India. A regional network for prenatal diagnosis and referral of patients with critical CHDs was initiated in 2011. Pregnancy and early neonatal outcomes were reported. The impact of the timing of diagnosis (prenatal or after birth) on age at surgery, perinatal mortality and postoperative recovery was evaluated.

RESULTS

A total of 82 fetuses with TGA-IVS were included. Diagnosis typically occurred later on in gestation, at a median of 25 (interquartile range (IQR), 21-32) weeks. The majority (78.0%) of affected pregnancies resulted in live birth, most (84.4%) of which occurred in a specialist pediatric cardiac centers. Delivery in a specialist center, compared with delivery in a local maternity center, was associated with a significantly higher rate of surgical correction (98.1% vs 70.0%; P = 0.01) and overall lower neonatal mortality (3.7% vs 50%; P = 0.001). The proportion of cases undergoing arterial switch operation after prenatal diagnosis of TGA-IVS increased significantly, along with the prenatal detection rate, over the study period (2011-2015, 11.1% vs 2016-2019, 29.4%; P = 0.001). Median age at surgery was significantly lower in the prenatally diagnosed group than that in the postnatally diagnosed group (4 days (IQR, 1-23 days) vs 10 days (IQR, 1-91 days); P < 0.001). There was no significant difference in postoperative mortality (2.0% vs 3.6%; P = 0.49) between the two groups.

CONCLUSIONS

This study demonstrates the feasibility of creating a network for prenatal diagnosis and referral of patients with critical CHDs, such as TGA, in low-resource settings, that enables planned peripartum care in specialist pediatric cardiac centers and improved neonatal survival. © 2020 International Society of Ultrasound in Obstetrics and Gynecology.