Disorders of the fetal circulation and the fetal brain

Automated 3D Fetal Brain Segmentation Using an Optimized Deep Learning Approach

BACKGROUND AND PURPOSE

MR imaging provides critical information about fetal brain growth and development. Currently, morphologic analysis primarily relies on manual segmentation, which is time-intensive and has limited repeatability. This work aimed to develop a deep learning-based automatic fetal brain segmentation method that provides improved accuracy and robustness compared with atlas-based methods.

MATERIALS AND METHODS

A total of 106 fetal MR imaging studies were acquired prospectively from fetuses between 23 and 39 weeks of gestation. We trained a deep learning model on the MR imaging scans of 65 healthy fetuses and compared its performance with a 4D atlas-based segmentation method using the Wilcoxon signed-rank test. The trained model was also evaluated on data from 41 fetuses diagnosed with congenital heart disease.

RESULTS

The proposed method showed high consistency with the manual segmentation, with an average Dice score of 0.897. It also demonstrated significantly improved performance (P < .001) based on the Dice score and 95% Hausdorff distance in all brain regions compared with the atlas-based method. The performance of the proposed method was consistent across gestational ages. The segmentations of the brains of fetuses with high-risk congenital heart disease were also highly consistent with the manual segmentation, though the Dice score was 7% lower than that of healthy fetuses.

CONCLUSIONS

The proposed deep learning method provides an efficient and reliable approach for fetal brain segmentation, which outperformed segmentation based on a 4D atlas and has been used in clinical and research settings.

De novo damaging variants associated with congenital heart diseases contribute to the connectome

Congenital heart disease (CHD) survivors are at risk for neurodevelopmental disability (NDD), and recent studies identify genes associated with both disorders, suggesting that NDD in CHD survivors may be of genetic origin. Genes contributing to neurogenesis, dendritic development and synaptogenesis organize neural elements into networks known as the connectome. We hypothesized that NDD in CHD may be attributable to genes altering both neural connectivity and cardiac patterning. To assess the contribution of de novo variants (DNVs) in connectome genes, we annotated 229 published NDD genes for connectome status and analyzed data from 3,684 CHD subjects and 1,789 controls for connectome gene mutations. CHD cases had more protein truncating and deleterious missense DNVs among connectome genes compared to controls (OR = 5.08, 95%CI:2.81-9.20, Fisher's exact test P = 6.30E-11). When removing three known syndromic CHD genes, the findings remained significant (OR = 3.69, 95%CI:2.02-6.73, Fisher's exact test P = 1.06E-06). In CHD subjects, the top 12 NDD genes with damaging DNVs that met statistical significance after Bonferroni correction (PTPN11, CHD7, CHD4, KMT2A, NOTCH1, ADNP, SMAD2, KDM5B, NSD2, FOXP1, MED13L, DYRK1A; one-tailed binomial test P ≤ 4.08E-05) contributed to the connectome. These data suggest that NDD in CHD patients may be attributable to genes that alter both cardiac patterning and the connectome.

Abnormal fetal cerebral and vascular development in hypoplastic left heart syndrome

Objective

To assess the cerebral and vascular development in fetuses with hypoplastic left heart syndrome (HLHS).

Methods

Pregnant women carrying fetuses diagnosed with HLHS who decided to interrupt their pregnancies were included in our study. Aortic size and blood flow were assessed based from fetal echocardiography. Immunohistochemical staining was performed in brain sections obtained from pathology in fetuses with HLHS and control fetuses without heart disease.

Results

Twenty‐seven midgestation fetal HLHS were included (gestational age, 23.3 ± 3.4 weeks). Head circumference z scores were lower in HLHS fetuses. Middle cerebral artery pulsatility index, a measure of cerebrovascular resistance, was inversely correlated with the ascending aortic z score (P < 0.05). Fetuses with HLHS had lower capillary density in the germinal matrix and their capillaries were larger compared with control fetuses with (P < 0.05). The expression of neuronal differentiation marker, FGFR1, and oligodendrocyte precursor, O4, were lower in HLHS brains compared with controls (P < 0.05).

Conclusion

Our study identified abnormalities of vascular flow and structural brain abnormalities in fetal HLHS associated with impaired neuronal and oligodendrocyte differentiation, as well as cerebral growth impairment, early in gestation. These findings may be related in part to early vascular abnormalities.

What's already known about this topic?

Structural brain abnormalities and abnormalities in neurodevelopment have been reported in HLHS.

What does this study add?

Our study describes abnormal vascular development that may account for abnormal neuronal and white matter differentiation in the developing fetus with HLHS.

A novel multimodal computational system using near-infrared spectroscopy predicts the need for ECMO initiation in neonates with congenital diaphragmatic hernia

BACKGROUND/PURPOSE

The purpose of this study was to develop a computational algorithm that would predict the need for ECMO in neonates with congenital diaphragmatic hernia (CDH).

METHODS

CDH patients from August 2010 to 2016 were enrolled in a study to continuously measure cerebral tissue oxygen saturation (cStO2) of left and right cerebral hemispheres. NIRS devices utilized were FORE-SIGHT, CASMED and INVOS 5100, Somanetics. Using MATLAB©, a data randomization function was used to deidentify and blindly group patient's data files as follows: 12 for the computational model development phase (6 ECMO and 6 non-ECMO) and the remaining patients for the validation phase.

RESULTS

Of the 56 CDH patients enrolled, 22 (39%) required ECMO. During development of the algorithm, a difference between right and left hemispheric cerebral oxygenation via NIRS (ΔHCO) was noted in CDH patients that required ECMO. Using ROC analysis, a ΔHCO cutoff >10% was predictive of needing ECMO (AUC: 0.92; sensitivity: 85%; and specificity: 100%). The algorithm predicted need for ECMO within the first 12h of life and at least 6h prior to the clinical decision for ECMO with 88% sensitivity and 100% specificity.

CONCLUSION

This computational algorithm of cerebral NIRS predicts the need for ECMO in neonates with CDH.

LEVEL OF EVIDENCE

II.

The Miracle Baby Grows Up: Hypoplastic Left Heart Syndrome in the Adult

PURPOSE OF REVIEW

Hypoplastic left heart syndrome (HLHS) is characterized by underdevelopment of the mitral valve, left ventricle, and aorta and is ultimately palliated with a single-ventricle repair. Universally fatal in infancy prior to the advent of modern surgical techniques, the majority of HLHS patients will now reach adulthood. However, despite improvements in early survival, the HLHS population continues to face significant morbidity and early mortality. This review delineates common sources of patient morbidity and highlights areas in need of additional research for this growing segment of the adult congenital heart disease population.

RECENT FINDINGS

It has become increasingly clear that palliated adult single ventricle patients, like those with HLHS, face significant life-long morbidity from elevated systemic venous pressures as a consequence of the Fontan procedure. Downstream organ dysfunction secondary to elevated Fontan pressures has the potential to significantly impact long-term management decisions, including strategies of organ allocation. Because of the presence of a morphologic systemic right ventricle, HLHS patients may be at even higher risk than other adult patients with a Fontan. Because the adult HLHS population continues to grow, recognition of common sources of patient morbidity and mortality is becoming increasingly important. A coordinated effort between patients and providers is necessary to address the many remaining areas of clinical uncertainty to help ensure continued improvement in patient prognosis and quality of life.

Head growth in fetuses with isolated congenital heart defects: lack of influence of aortic arch flow and ascending aorta oxygen saturation

OBJECTIVES

Congenital heart defects (CHDs) are reported to be associated with a smaller fetal head circumference (HC) and neurodevelopmental delay. Recent studies suggest that altered intrauterine brain hemodynamics may explain these findings. Our objectives were to evaluate the pattern of head growth in a large cohort of fetuses with various types of CHD, analyze these patterns according to the type of CHD and estimate the effect of cerebral hemodynamics with advancing gestation in the second and third trimesters.

METHODS

Singleton fetuses with an isolated CHD were selected from three fetal medicine units (n = 436). Cases with placental insufficiency or genetic syndromes were excluded. CHD types were clustered according to the flow and oxygen saturation in the aorta. Z-scores of biometric data were constructed using growth charts of a normal population. HC at different gestational ages was evaluated and univariate and multivariate mixed regression analyses were performed to examine the patterns of prenatal HC growth.

RESULTS

Fetuses with severe and less severe types of CHD demonstrated statistically significant HC growth restriction with increasing gestational age (slope of -0.017/day); however, there was no statistically significant effect of fetal hemodynamics on HC growth. Fetuses with CHD but normal brain oxygenation and normal aortic flow showed a significant decrease in HC growth (slope of -0.024/day). Only fetuses with isolated tetralogy of Fallot demonstrated a smaller HC z-score at 20 weeks of gestation (-0.67 (95% CI, -1.16 to -0.18)).

CONCLUSIONS

Despite the decline in head growth in fetuses with a prenatally detected isolated CHD, HC values were within the normal range, raising the question of its clinical significance. Furthermore, in contrast to other studies, this large cohort did not establish a significant correlation between aortic flow or oxygen saturation and HC growth. Factors other than altered fetal cerebral hemodynamics may contribute to HC growth restriction with increasing gestational age, such as (epi)genetic or placental factors. Copyright © 2016 ISUOG. Published by John Wiley & Sons Ltd.

Fetal MRI detects early alterations of brain development in Tetralogy of Fallot

OBJECTIVE

Prenatal imaging has identified alterations of brain growth in fetuses with congenital heart disease. However, little is known about the timing of altered brain development and its occurrence in specific congenital heart disease subgroups. This magnetic resonance imaging study aimed to identify early (median, 25 gestational weeks [GW]) changes in fetal total brain (TBV), gray matter (GMV), and subcortical brain (SBV) volumes in Tetralogy of Fallot (TOF) cases in utero.

STUDY DESIGN

Fetal magnetic resonance imaging (1.5 Tesla) was performed in 24 fetuses who were diagnosed with TOF and 24 normal age-matched control fetuses (20-34 GW). TBV, GMV, SBV, intracranial cavity, cerebellar, ventricular, and external cerebrospinal fluid volumes were quantified by manual segmentation based on coronal T2-weighted sequences. Mixed model analyses of variance and t-tests were conducted to compare cases and control fetuses.

RESULTS

TBV was significantly lower (P < .001) in early (<25 GW) and late TOF cases. Both GMV (P = .003) and SBV (P = .001) were affected. The GMV-to-SBV ratio declined in fetuses with TOF (P = .026). Compared with normal fetuses, ventricular volume was increased (P = .0048). External cerebrospinal fluid was enlarged in relation to head size (P < .001). Intracranial cavity volume (P = .314) and cerebellar volume (P = .074) were not significantly reduced in fetuses with TOF.

CONCLUSION

TOF is associated with smaller volumes of gray and white matter and enlarged cerebrospinal fluid spaces. These changes are present at ≤25 GW and indicate altered fetal brain growth in this pathophysiologic entity during early stages of human brain development.

3-D volumetric MRI evaluation of the placenta in fetuses with complex congenital heart disease

INTRODUCTION

Placental insufficiency remains a common cause of perinatal mortality and neurodevelopmental morbidity. Congenital heart disease (CHD) in the fetus and its relationship to placental function is unknown. This study explores placental health and its relationship to neonatal outcomes by comparing placental volumes in healthy pregnancies and pregnancies complicated by CHD using in vivo three-dimensional MRI studies.

METHODS

In a prospective observational study, pregnant women greater than 18 weeks gestation with normal pregnancies or pregnancies complicated by CHD were recruited and underwent fetal MR imaging. The placenta was manually outlined and the volume was calculated in cm(3). Brain volume was also calculated and clinical data were also collected. Relationships, including interactive effects, between placental and fetal growth, including brain growth, were evaluated using longitudinal multiple linear regression analysis.

RESULTS

135 women underwent fetal MRI between 18 and 39 weeks gestation (mean 31.6 ± 4.4). Placental volume increased exponentially with gestational age (p = 0.041). Placental volume was positively associated with birth weight (p < 0.001) and increased more steeply with birth weight in CHD-affected fetuses (p = 0.046). Total brain and cerebral volumes were smaller in the CHD group (p < 0.001), but brainstem volume (p < 0.001) was larger. Placental volumes were not associated with brain volumes.

DISCUSSION

Impaired placental growth in CHD is associated with gestational age and birth weight at delivery. Abnormalities in placental development may contribute to the significant morbidity in this high-risk population. Assessment of placental volume by MRI allows for in vivo assessments of placental development.

Challenges and controversies in fetal diagnosis and treatment: hypoplastic left heart syndrome

Today, almost 70% of babies with hypoplastic left heart syndrome (HLHS) will survive into adulthood, although significant long-term morbidity and mortality still exists. Prenatal diagnosis of HLHS is increasingly common, allowing improved counseling, and the potential for fetal intervention if indicated. Exciting progress continues to be made in the area of fetal diagnosis and intervention, specifically catheter intervention for intact atrial septum or severe aortic stenosis. Pediatric cardiologists should be keenly aware of the flaws of staged palliation for the treatment of HLHS, and need to keep abreast of the emerging data regarding fetal diagnosis and intervention.

Brain abnormalities and neurodevelopmental delay in congenital heart disease: systematic review and meta-analysis

OBJECTIVES

Studies have demonstrated an association between congenital heart disease (CHD) and neurodevelopmental delay. Neuroimaging studies have also demonstrated a high incidence of preoperative brain abnormalities. The aim of this study was to perform a systematic review to quantify the non-surgical risk of brain abnormalities and of neurodevelopmental delay in infants with CHD.

METHODS

MEDLINE, EMBASE and The Cochrane Library were searched electronically without language restrictions, utilizing combinations of the terms congenital heart, cardiac, neurologic, neurodevelopment, magnetic resonance imaging, ultrasound, neuroimaging, autopsy, preoperative and outcome. Reference lists of relevant articles and reviews were hand-searched for additional reports. Cohort and case-control studies were included. Studies reporting neurodevelopmental outcomes and/or brain lesions on neuroimaging in infants with CHD before heart surgery were included. Cases of chromosomal or genetic abnormalities, case reports and editorials were excluded. Between-study heterogeneity was assessed using the I(2) test.

RESULTS

The search yielded 9129 citations. Full text was retrieved for 119 and the following were included in the review: 13 studies (n = 425 cases) reporting on brain abnormalities either preoperatively or in those who did not undergo congenital cardiac surgery and nine (n = 512 cases) reporting preoperative data on neurodevelopmental assessment. The prevalence of brain lesions on neuroimaging was 34% (95% CI, 24-46; I(2) = 0%) in transposition of the great arteries, 49% (95% CI, 25-72; I(2) = 65%) in left-sided heart lesions and 46% (95% CI, 40-52; I(2) =18.1%) in mixed/unspecified cardiac lesions, while the prevalence of neurodevelopmental delay was 42% (95% CI, 34-51; I(2) = 68.9).

CONCLUSIONS

In the absence of chromosomal or genetic abnormalities, infants with CHD are at increased risk of brain lesions as revealed by neuroimaging and of neurodevelopmental delay. These findings are independent of the surgical risk, but it is unclear whether the time of onset is fetal or postnatal.

Structural congenital brain disease in congenital heart disease: results from a fetal MRI program

OBJECTIVE

To identify the type and incidence of fetal brain pathology in fetuses with a prenatal diagnosis of congenital heart disease (CHD).

PATIENTS AND METHODS

67 pregnant women underwent a fetal MR-examinations between 20 and 38 gestational weeks. MR was done on a 1.5 T superconducting system. The type of cardiac malformation was defined by fetal echocardiography. Fetuses with a chromosomal abnormality or an extracardiac anomaly were excluded.

RESULTS

Fetal MRI scans in the final study cohort (53 fetuses) yielded normal results in 32 fetuses and a brain abnormality in 21 fetuses. Congenital brain disease (CBD) was found in 39% of the final study cohort of fetuses with CHD. MRI findings were classified into malformations, acquired lesions and widening of the ventricles and/or outer CSF spaces (malformations: 7 fetuses, acquired lesions: 5 fetuses, changes in CSF spaces: 9 fetuses). Asymmetry of the ventricles was the most common finding in the CSF group.

CONCLUSIONS

Our data suggest that fetal MRI can be used to characterize structural CBD in CHD. Advanced MRI techniques such as diffusion tensor imaging and proton spectroscopy are tools that, in the future, will certainly shed light on the spectrum of structural and functional CBDs that are associated with CHD.

Development and validation of a semiquantitative brain maturation score on fetal MR images: initial results

PURPOSE

To develop a valid, reliable, and simple-to-use semiquantitative visual scale of fetal brain maturation for use in clinical fetal MR imaging assessment and interpretation.

MATERIALS AND METHODS

This is a retrospective assessment of data from a previous study that was prospective, institutional review board approved, and HIPAA compliant. Forty-eight normal pregnancies with a gestational age (GA) of 25 to 35 weeks were studied. A fetal total maturation score (fTMS) was developed by utilizing six subscores that evaluated cortical sulcation, myelination, and the germinal matrix and provided a single combined numerical value to be evaluated as a marker of brain maturity. The fTMS was correlated with GA and segmented brain volume. A regression model that associated GA based on the visual fTMS scoring was determined. The model was validated with a leave-one-out cross validation procedure.

RESULTS

Mean GA was 29.3 weeks ± 2.9 (standard deviation) (range, 25.2-35.3 weeks) and mean fTMS was 8.6 ± 3.7 (range, 4-16). The intraclass correlation coefficient between the three readers (independent and blinded) was 0.948 (P < .001). The correlations were as follows: GA and brain volume, r = 0.964 (P < .001); fTMS and brain volume, r = 0.970 (P < .001); and GA and fTMS, r = 0.975 (P < .001). A regression model to calculate GA based on fTMS yielded the following equation: calculated GA (weeks) = 22.86 + 0.748 fTMS (P < .001; adjusted R(2) = 0.946). The standard error of the model for calculation of fetal GA from the visual fTMS scale was ± 4.8 days.

CONCLUSION

If validated further, the fTMS scale might be used to assess morphologic brain maturity of fetuses between 25 and 35 weeks GA on a single-case basis in a clinical setting.

Neurological injury and anesthetic neurotoxicity following neonatal cardiac surgery: does the head rule the heart or the heart rule the head?

The improvements in care of children with heart disease have resulted in a major decrease in mortality and increased attention to adverse events and quality of survival. There is important neurological morbidity in children with congenital heart disease. Some problems such as stroke or seizure may be immediately apparent, but others, such as learning disability and motor delay emerge over time. The etiology is multifactorial and includes genetic, procedural and social causes. Only some factors are modifiable. Over the last decade, evidence has been presented that anesthetic drugs may be a potential cause of CNS morbidity. Neonates and infants may be particularly vulnerable to this. The purpose of this article is to describe the multiple known causes of neurodevelopmental impairment in children with heart disease, including anesthetic agents, and to explore the relationship between congenital heart disease and its treatment in this regard.

Hypoplastic left heart syndrome: current considerations and expectations

In the recent era, no congenital heart defect has undergone a more dramatic change in diagnostic approach, management, and outcomes than hypoplastic left heart syndrome (HLHS). During this time, survival to the age of 5 years (including Fontan) has ranged from 50% to 69%, but current expectations are that 70% of newborns born today with HLHS may reach adulthood. Although the 3-stage treatment approach to HLHS is now well founded, there is significant variation among centers. In this white paper, we present the current state of the art in our understanding and treatment of HLHS during the stages of care: 1) pre-Stage I: fetal and neonatal assessment and management; 2) Stage I: perioperative care, interstage monitoring, and management strategies; 3) Stage II: surgeries; 4) Stage III: Fontan surgery; and 5) long-term follow-up. Issues surrounding the genetics of HLHS, developmental outcomes, and quality of life are addressed in addition to the many other considerations for caring for this group of complex patients.

Impact of congenital heart disease on fetal brain development and injury

PURPOSE OF REVIEW

Advances in cardiac surgical techniques and intensive care have led to improved survival in babies with congenital heart disease (CHD). Although it is true that the majority of children with CHD today survive, many have impaired neurodevelopmental outcome. Although continuing to improve short-term morbidity and mortality are important goals, recent research has focused on defining the impact of CHD on brain development and brain injury in utero.

RECENT FINDINGS

The impact of CHD on the developing brain of the fetus and infant will be discussed. Neurologic abnormalities detectable prior to surgery will be described and postnatal progression of abnormalities will be highlighted. Potential causes of these findings will be discussed, including altered cerebral blood flow in utero, and brain development and risk for in-utero and postnatal brain injury. Finally, neurologic and developmental outcome after surgical repair of CHD will be reviewed.

SUMMARY

Neurodevelopmental evaluation preoperatively and postoperatively in CHD patients should be standard practice, not only to identify those with impairments who would benefit from intervention services but also to identify risk factors and strategies to optimize outcome. Fetal management and intervention strategies for specific defects may ultimately play a role in improving in-utero hemodynamics and increasing cerebral oxygen delivery to enhance brain development.