The Congenital Heart Disease Brain: Prenatal Considerations for Perioperative Neurocritical Care

Abstract

Altered brain development has been highlighted as an important contributor to adverse neurodevelopmental outcomes in children with congenital heart disease. Abnormalities begin prenatally and include micro- and macrostructural disturbances that contribute to an altered trajectory of brain growth throughout gestation. Recent progress in fetal imaging has improved understanding of the neurobiological mechanisms and risk factors for impaired fetal brain development. The impact of the prenatal environment on postnatal neurological care has also gained increased focus. This review summarizes current data on the timing and pattern of altered brain development in congenital heart disease, the potential mechanisms of these abnormalities, and the association with perioperative neurological complications.

Why Focus on the Prenatal Period?

Over the past several decades, advancements in surgical and cardiac intensive care have improved survival for infants with congenital heart disease (CHD). As a result, there are now a greater number of adults than children living with this diagnosis.¹ This has shifted clinical and research efforts to reducing the associated morbidities of CHD. Neurodevelopmental impairment is increasingly recognized as the most common morbidity and spans deficits in motor function, cognition, language, executive function, visual-spatial skills, and behavioral and psychosocial outcomes.^2-4 By adolescence, 65% of children with CHD will receive remedial services and 50% will receive psychotherapy or counseling.² Impairments in cognitive and psychosocial functioning continue in adulthood and affect educational achievement, quality of life, employment, and insurance acquisition.^5-7 The growing population of CHD survivors and the significant long-term neurodevelopmental consequences of this diagnosis pose an urgent need to develop neuroprotective strategies. Many of the initial neurologically focused studies centered on perioperative mechanisms of neurodevelopmental impairment as potential targets. However, postnatal medical, surgical, and socioeconomic factors only explain 30% of the variance in developmental outcome.⁸ Thus, the prenatal environment likely plays an important role.

A growing number of brain imaging studies have contributed to our understanding of the underlying neurological abnormalities that lead to neurodevelopmental impairments in CHD. While brain injury is common, altered brain development has been highlighted as a key component of the long-term neurodevelopmental sequelae these children face.^9-11 Importantly, the presence of abnormal preoperative brain development suggests in utero disturbances.^12-16 The seminal work of Limperopoulos and colleagues applied fetal magnetic resonance imaging (MRI) methods to the CHD population and confirmed the existence of intrauterine abnormalities.¹⁷ Increasing focus on the prenatal environment has laid the foundation for current understanding of the pattern and underlying mechanisms of altered brain development in CHD and its implications for perioperative neurocritical care. These data are summarized throughout this review and are schematically displayed in Figure 1.

Figure 1.

Prenatal Brain Development in Congenital Heart Disease. This schematic represents the pattern of abnormal brain development in the CHD fetus by highlighting the primary macrostructural and microstructural abnormalities that have been identified on fetal brain imaging. The left panel summaries potential mechanisms that have been proposed for disrupted prenatal brain development. The right panel summarizes perioperative neurological abnormalities for which altered prenatal brain development increases risk (top) and prenatal factors that appear to be neuroprotective for the developing CHD brain (bottom).

Prenatal Disruption of Brain Development in CHD

Timing and Trajectory of Altered Brain Growth

The exact timing by which typical brain development is disrupted in CHD fetuses has not been fully elucidated, but ultrasonography and MRI studies have begun to provide insights. Brain growth has been investigated in multiple CHD cohorts using ultrasound measures of biparietal diameter and head circumference via absolute values, z-scores, or in relation to weight. While not all studies support a CHD-related vulnerability, the vast majority show second and third trimester reductions in fetal brain size.^18-25 A recent population-based study of over 57,000 pregnancies, including 295 with fetal isolated CHD, identified that abnormal brain growth begins in mid-gestation. Biparietal diameter z-scores were similar between groups at 12 weeks gestation, but head circumference z-scores were reduced in CHD fetuses at 20 weeks (z-score difference of −0.13). Even greater reductions were present at birth (z-score difference of −0.22).²⁵ Analyses of growth rates have shown slower second to third trimester biparietal diameter and head circumference growth in the CHD population.²⁴

Fetal MRI studies also demonstrate altered brain growth. Reduced total brain volume (TBV) is reported by 25 weeks gestation.²⁶ Third trimester deficits are consistently demonstrated across studies and display progressive worsening in growth over pregnancy.^17,26-28 A study of longitudinal prenatal and postnatal MRI showed a slower rate of growth that led to reductions by early preoperative imaging, supporting progressive third trimester impairments.²⁹ A small cohort comparing fetuses with transposition of the great arteries (TGA) to controls who underwent 3-5 fetal MRIs identified TGA fetuses to have a smaller brain/body ratio at 24 weeks and reduced TBV at 26 weeks that persisted throughout pregnancy.³⁰ Thus, ultrasound and MRI data demonstrate impaired brain growth by the second trimester that progresses at a slower rate of growth throughout gestation.

Tissues and Structures Affected

Early preoperative, neonatal MRI reflects the overall effects of the prenatal environment on brain development for CHD infants. Delayed myelination patterns, impaired white matter microstructural development, abnormal brain metabolism, reductions in regional brain size and volume, and abnormal cerebral cortical folding and development are all present preoperatively.^12-16 These findings suggest global white and gray matter disturbances are occurring in the CHD fetus. With advances in fetal MRI techniques, similar detailed assessments as those performed on neonatal MRI are now becoming feasible prenatally.

Specific to the developing white matter, CHD fetuses display macrostructural (via volumetric analysis) and microstructural (via diffusion tensor imaging) abnormalities.^26,31,32 In a cohort of HLHS fetuses, reductions in white matter volume began at 30 weeks gestation.³¹ Subcortical volume, which includes the developing white matter, is reduced at 25 weeks for Tetralogy of Fallot (TOF).²⁶ A cohort of heterogeneous cardiac diagnoses who underwent diffusion tensor imaging at 26-30 weeks demonstrated decreased fractional anisotropy in the corpus callosum.³² This represents atypical microstructural development of this major white matter tract.

Gray matter development is also disrupted prenatally in CHD fetuses. For HLHS, the white matter deficits described above occur alongside volumetric reductions in the cerebral cortex and subcortical gray matter, although subcortical structures are less affected.³¹ Despite this, a case series of three fetuses with CHD who underwent diffusion tensor imaging at 32-35 weeks showed altered microstructure in the thalamus.³³ Reductions in cerebellar volume have also been identified.³⁴ Cortical involvement is present by 25 weeks gestation and includes abnormalities in cortical folding and sulcation.^31,35 In HLHS fetuses, there is a three-week delay in the emergence of the cingulate sulcus, superior frontal sulcus, and anterior ascendant ramus.³¹ Left hemispheric sulcal pattern differences are present in the Sylvian Fissure and early emerging sulci in CHD fetuses imaged at 21-30 weeks gestation.³⁵ These alterations are primarily due to differences in the position of corresponding sulcal basins, which reflects the positional relationship between the deepest parts of neighboring cerebral sulci.³⁵

Studies investigating the trajectory of tissue-specific alterations and their contributions to overall brain growth are limited. Fetuses with TOF have a declining ratio of gray matter to subcortical volume over gestation.²⁶ This supports a proportionately slower rate of cortical growth. Two other cohorts have identified the cerebral cortex, subcortical gray matter, and cerebellum as primary contributors to the altered growth trajectory seen in CHD fetuses.^28,29 The constellation of findings on fetal MRI studies support global white and gray matter disturbances at the micro- and macro-structural level in the late second and early third trimester that progress along an altered trajectory of brain growth. Deficits in gray matter structures may provide the greatest contributions to this altered trajectory over time.

Diagnoses at Highest Risk

In the recent population-based study from Lauridsen and colleagues, fetuses with single ventricle physiology and TOF had the greatest reductions in head circumference prenatally.²⁵ Whereas infants with TGA have the smallest head size relative to total body size at birth.³⁶ Reduced TBV has been shown to be most severe for fetuses with single ventricle physiology, specifically HLHS in one cohort of patients.^17,28 However, biventricular CHD, and particularly TOF, had the greatest reductions in another study.²⁷ The heterogeneity in cardiac diagnoses and large variations in anatomy and physiology, even within specific lesions, make it difficult to determine exact diagnosis-specific risk. However, moderate-severe lesions such as HLHS, TGA, and TOF are consistently affected.

CHD-Related Hypoxia as a Mechanism of Altered Prenatal Brain Development

Fetal Physiologic Response to Hypoxia

Fetuses rely on delivery of oxygen and nutrient-rich blood from the placenta. The oxygenated blood is preferentially shunted through the foramen ovale to the left side of the heart, out the aorta, and to the fetal brain (Figure 2A).^37,38 Many forms of CHD alter cardiac anatomy in such a way that disrupts this normal physiology. For example, in TGA, the aorta and pulmonary artery are transposed and blood supply to the fetal brain comes from the deoxygenated right side of the heart (Figure 2B).^37,38 In fetuses with HLHS, hypoplasia of left-sided structures facilitate mixing of oxygenated and deoxygenated blood in the right atrium. In the most severe cases, the relatively deoxygenated right-sided blood is supplied to the developing brain retrograde through the aortic isthmus (Figure 2C).^37,38 These examples highlight two ways in which distinct cardiac physiologies may contribute to fetal cerebral hypoxia. Although, other diagnoses also disrupt normal fetal physiology.

Figure 2.

Fetal Cardiac Physiology. Normal fetal circulation and changes with congenital heart disease course of blood flow in a late gestation fetus with normal heart anatomy (A), d-transposition of the great arteries (B) and hypoplastic left heart syndrome due to aortic atresia (C). Deoxygenated blood (blue-purple) flows to the placenta through the umbilical artery (UA) where gas exchange takes place. Blood with higher oxygen content (red) returns through the umbilical vein (UV) and ductus venosus (DV) to the inferior vena cava (IVC). The more highly saturated blood forms a stream in the IVC, which is preferentially directed across the foramen ovale into the left ventricle in the normal fetus and with d-TGA. Estimated hemoglobin oxygen saturation in percent is shown for each ventricle. Blood flow to the fetal lungs is limited by elevated pulmonary vascular resistance. In the fetus with d-TGA, the aorta arises from the right ventricle such that the brain receives less oxygenated blood, while the higher saturated blood is directed to the descending aorta through the ductus arteriosus. In HLHS, reduced or absent left ventricular ejection results in elevated left atrial pressure, limiting or reversing flow at the foramen and resulting in complete mixing of desaturated and well saturated blood in the right atrium and ventricle. Blood flow to the head and neck may occur in a retrograde fashion from the ductus arteriosus across the aortic isthmus. (Reprinted from Prog Pediatr Cardiol; 29(2), McQuillen et al, Effects of congenital heart disease on brain development, 79-85, 2010, with permission from Elsevier)

The fetal response to hypoxia appears to vary depending on gestational age and whether the exposure is acute or chronic. In acute hypoxia, cerebral vasodilation permits increased oxygen delivery to the developing brain, known as the “brain sparing effect.”³⁹ However, this response is insufficient for oxygen consumption early in gestation.⁴⁰ Furthermore, with chronic fetal hypoxia, as occurs with CHD, middle cerebral artery vasodilation is attenuated and decreased cerebral oxygen consumption occurs.³⁹ Initial studies investigating CHD-related hypoxia measured cerebrovascular resistance from ultrasonography as a surrogate marker of compensatory cerebral vasodilation. These data identified alterations in a variety of diagnoses, but differing effects were present depending on the study.^23,41-43 The most consistent alteration was decreased resistance for HLHS fetuses.^41-43 These cohort variations may reflect differences in severity and longevity of hypoxia, depending on gestational age at measurement and diagnoses included. Factors beyond oxygenation may also play a role, since it is known in adults that pCO2 is a potent stimulator of cerebral blood flow.⁴⁴

Recent advances in fetal MRI now permit more direct measurement of tissue oxygenation to delineate the effects of CHD-related hypoxia on the developing brain. Application of these methods to CHD demonstrate reduced cerebral oxygenation in the third trimester.^27,45 The rate of change in oxygenation between 32 and 37 weeks gestation does not differ for CHD and control fetuses despite reductions at both time points.⁴⁵ MRI hemodynamic parameters at 36 weeks gestation have identified reduced fetal oxygen delivery, ascending aortic saturation, and fetal and cerebral oxygen consumption. The reduction in cerebral oxygen consumption is associated with lower brain weight by a full standard deviation.²⁷ In this same cohort, similar superior vena cava flow was present in CHD and control fetuses, which could suggest a more chronic hypoxic state where brain-sparing physiology is no longer present but hypoxia still contributes to altered brain development.²⁷

Hypoxia-Related Cellular Mechanisms of Altered Brain Development

Based on preterm data, the primary cellular mechanism for hypoxia-related disruption of brain development is thought to be injury to premyelinating oligodendrocytes (Pre-OLs).^46,47 Direct effects on developing white matter axons, subplate neurons, and migratory γ-aminobutyric acid (GABA)-ergic neurons can also occur.^46,47 Pre-OLs originate from radial glial cells in the subventricular zone and differentiate to form mature oligodendrocytes, which are responsible for myelination of developing white matter axons.⁴⁷ Pre-OLs are highly prevalent in the developing brain at 23-32 weeks gestation and are particularly vulnerable to hypoxia-ischemia.^48,49 Pre-OL injury occurs via excitotoxicity, free radical production, and microglial activation. This most commonly leads to a loss of cell processes as opposed to apoptosis.⁴⁷ Precursor cells are then upregulated but an arrested maturation occurs that prevents differentiation to mature myelin-producing cells and results in hypomyelination.^50-52 Axonal/neuronal involvement contributes to abnormalities in the cerebral cortex and subcortical gray matter structures via impaired thalamocortical connectivity (from hypomyelination) or via direct hypoxic effects.^53-56

Neuropathology data in fetuses and infants with CHD demonstrate evidence of white matter, cortical gray matter, and subcortical gray matter injury as early as 19-22 weeks gestation.^57,58 A recent model of intrauterine chronic hypoxia, similar to the severity of hypoxia in TGA, identified reduced neuronal density in the cerebral white matter, impaired myelination, reduced cortical and cerebellar weight, and cortical folding abnormalities.⁵⁹ Given the existence of cortical abnormalities, the subventricular zone, which is the primary postnatal contributor to cortical neurons,⁶⁰ has become a focus in CHD. A neonatal porcine model of chronic hypoxia identified reduced cortical gray matter volume and gyrification to coincide with fewer neuroprogenitor cells and decreased neurogenesis in the subventricular zone.⁶¹ Brain specimens from human infants with CHD demonstrated similar subventricular and cortical involvement.⁶¹ These data all suggest a complex interplay of white matter and axonal/neuronal disturbances that result in the global abnormalities seen on fetal MRI in CHD.

Decreased glucose supply has been proposed as the mechanism behind hypoxic effects on the developing brain.³⁸ The fetal brain accounts for one-third of total body glucose consumption during the second trimester of pregnancy.⁶² Glucose is the energy substrate for the brain and, in the setting of hypoxia, is metabolized through the tricarboxylic acid cycle to pyruvate and ultimately lactate.³⁸ Interestingly, altered brain metabolism, including elevated lactate, has been reported in CHD fetuses.¹⁷ Exposing fetal sheep to intrauterine chronic hypoxia while providing similar glucose and nutrient delivery as non-hypoxic animals does still render the white matter and cortical deficits described above.⁵⁹ However, chronic hypoxia could still decrease brain glucose supply via conversion to anaerobic metabolism even with adequate nutrient delivery from the placenta. Recently, it has been hypothesized that the fetal brain relies on aerobic glycolysis for typical brain development.⁶² This may provide further support for the important role of glucose. Abnormal placental functioning may also affect nutrient and oxygen delivery in the CHD fetus.

The Role of the Placenta

The placenta is essential for oxygen and nutrient exchange in the fetus and for health and development over the lifespan.⁶³ Preeclampsia, fetal growth restriction, and preterm birth are associated with placental abnormalities.^64,65 These factors are also more common with CHD and are linked with longer hospital length of stay and decreased 3-year survival in the cardiac population.^66-69 Reduced placental weight occurs across multiple cardiac defects and correlates with lower birth head circumference.^36,70 A recent publication of 120 pregnancies with fetal CHD showed increased chorioangiosis, vascular hyperplasia in terminal chorionic villi, and villus hypomaturity of the placenta.⁷⁰ A small cohort of HLHS pregnancies has shown abnormal development of placental villi, alongside reduced vascular area and density and dysregulation of leptin, an angiogenic factor.⁷¹ Placental vascular malperfusion includes several patterns of vascular abnormalities that can result from obstruction in fetal blood flow or hypoxia and is also seen with CHD.^65,72,73 Placental MRI has identified reduced umbilical vein oxygenation and altered placental perfusion in CHD fetuses.^27,74 These data support altered placental functioning in addition to the disrupted vascular development that has been reported. Recently, placental abnormalities have been associated with more severe brain injury postnatally.⁷⁵ Thus, abnormal placental development and function may play an important role in CHD-associated neurological abnormalities.

Genetic and Epigenetic Considerations

It is well known that syndromic patients with CHD and those with extra-cardiac anomalies are at greater risk for adverse neurodevelopmental outcomes.^76-78 Rapid advances in sequencing technologies are now beginning to reveal the extent to which genetic and epigenetic factors contribute to neurodevelopment. For example, copy number variants appear to play an important pathogenic role for CHD and include variants in genes associated with neuronal development.⁷⁹ Furthermore, pathogenic copy number variants correlate with impaired motor development at 14 months of age.⁸⁰ The apolipoprotein E ε2 allele has also been associated with impaired motor development, as well as behavioral problems.^81,82 Additionally, de novo damaging mutations in genes highly expressed in the heart and brain have been identified in CHD patients with neurodevelopmental disability.⁸³ These data support a genetic contribution to altered brain development in CHD.

Disruption of epigenetic regulation may alter transcription factor binding and expression of genes critical for both cardiac and brain development.^84,85 Hypoxia-related altered gene expression in angiogenic pathways may also be important. Hypoxia inducible factors 1- and 2-alpha (HIF-1/2α) regulate adaptation to hypoxia by activating gene transcription.⁸⁶ Under hypoxic conditions, HIF suppresses maturation of oligodendroglia and increases vascular density via the WNT signaling pathway.⁸⁷ Impaired maturation of oligodendroglia results in hypomyelination, as described above. Intrauterine chronic hypoxia increases capillary density in the cerebral white matter and correlates with impaired myelination.⁵⁹ Thus, this is a plausible pathway by which brain development may be altered in CHD. Specific to CHD fetuses, HIF-2α expression has been shown to be increased in the basal ganglia.⁸⁸ Expression of soluble fms-like tyrosine kinase-1 and vascular endothelial growth factor-A, both angiogenic factors, are also altered.⁸⁸ Damaging angiogenic variants identified by whole exome sequencing have recently been associated with smaller birth head circumference and increased umbilical artery pulsatility index z-scores (suggesting placental insufficiency).⁸⁹ These findings support a common pathway by which angiogenic variants result from and/or cause abnormal heart, brain, and placental development.

Despite growing evidence linking genetic variants to neurodevelopmental outcome, minimal data have studied their association with the micro- and macrostructural brain abnormalities that occur on fetal MRI in CHD. Sulcal pattern analysis has identified alterations in the earliest emerging sulci,³⁵ which are thought to be under tighter genetic control than later emerging sulci.⁹⁰ Whole exome sequencing in adolescents with CHD have identified that damaging de novo variants were only present in patients with the most atypical sulcal patterns.⁹¹ Additional studies are needed to determine the effects of genetic and epigenetic factors on the timing, trajectory, and region/structure-specific alterations in brain development that occur in CHD.

Impact of Prenatal Brain Development on Perioperative Neurological Risk

Association of Prenatal Brain Development with Perioperative Brain Injury

Postnatal MRI studies have utilized a brain maturational scoring system in infants with CHD and have identified an association between preoperative brain immaturity and increased risk of white matter injury.^9,92 Impaired white matter microstructure and brain metabolism have also been associated with perioperative white matter injury and stroke.⁹³ Fetal MRI data are now beginning to further define these associations. Qualitative findings of altered fetal brain development, such as increased extra-axial spaces and ventricular dilatation, increase the risk of preoperative white matter injury.⁹⁴ A recent cohort of CHD fetuses who underwent fetal volumetric analysis discovered prenatal reductions in total brain, cortical gray matter, unmyelinated white matter, and cerebellar volume to correlate with neonatal ischemic brain injury. Increased cerebrospinal fluid volume had a similar effect.⁹⁵

Animal models have identified a plausible cellular mechanism by which abnormal brain development leads to brain injury. As described above, injury to Pre-OLs results in upregulation of progenitors that are unable to form mature oligodendrocytes, leading to a preponderance of Pre-OLs.^50-52 Persistent Pre-OLs in chronic white matter lesions demonstrate increased susceptibility to injury from repeated hypoxic-ischemic exposure.⁵² In the clinical setting, cardiac diagnoses that result in intrauterine hypoxia typically display postnatal hypoxia and require surgical interventions with cardiopulmonary bypass and deep hypothermic circulatory arrest. A preclinical model mimicking the ischemia-reperfusion/reoxygenation of cardiopulmonary bypass and circulatory arrest has shown a vulnerability of Pre-OLs.⁹⁶ An upregulation of precursors cells and downstream hypomyelination results.⁹⁶ These findings are similar to the cellular mechanisms proposed for prenatal hypoxic brain alterations. Furthermore, hypoxia prior to ischemia-reperfusion/reoxygenation leads to additional susceptibility of Pre-OLs.⁹⁷ Therefore, CHD-related fetal hypoxia may render the developing brain more vulnerable during cardiac surgery, particularly if altered brain development pre-exists. Of interest, mature oligodendrocytes are also susceptible during ischemia-reperfusion/reoxygenation when exposed to prior hypoxia.⁹⁷ However, deep hypothermic circulatory arrest mitigates this risk, suggesting this surgical strategy could be somewhat neuroprotective.⁹⁷

Prenatal Brain Development and Infant Electroencephalography

In addition to brain injury, CHD infants often display abnormal background pattern on electroencephalography (EEG) perioperatively, and 8-30% have evidence of electrographic seizures.^98-102 In a small cohort of heterogeneous cardiac diagnoses, nearly 60% of infants demonstrated an abnormal EEG on the first day of life.¹⁰³ Furthermore, infants with brain atrophy on preoperative MRI, which would be reflective of global disturbances in brain development, were more likely to have severely abnormal EEG background pattern and were unlikely to have sleep-wake cycling.¹⁰³ Subsequent data found no association of brain maturation or microstructural development with EEG background activity, although infants with brain injury were more likely to have a discontinuous pattern.¹⁰⁴ While the data are not conclusive, it is possible that EEG abnormalities in CHD infants result from altered prenatal brain development, postnatal brain injury secondary to impaired development, or a combination of both.

Prenatal Diagnosis, Brain Injury, and Postnatal Brain Development

In the modern era of fetal cardiology, prenatal diagnosis is becoming more common and appears to have a neuroprotective effect. For example, Mahle and colleagues have shown that infants with HLHS who were prenatally diagnosed were less likely to have seizures or a comatose state than those diagnosed after birth.¹⁰⁵ Prenatally diagnosed single ventricle and TGA infants demonstrate less preoperative brain injury and less severe lesions if injury is present.¹⁰⁶ Postnatally diagnosed infants are more likely to have metabolic acidosis and higher lactate levels before cardiac surgery.^105,107 This hypoperfusion state has been proposed as a mechanism for increased neurological risk in infants who are not diagnosed until after birth. However, surgery often occurs at a later age in patients with a postnatal diagnosis.^105,106,108 Longer time to surgery has been associated with increased risk of postoperative white matter injury due to lower cerebral oxygen saturations.¹⁰⁹ Thus, the timing of cardiac surgery in postnatally diagnosed CHD infants may also impact risk of brain injury¹¹⁰ and could be a potentially modifiable factor.

Prenatal diagnosis has also been associated with more optimal brain development in the perioperative period. Specifically, prenatally diagnosed infants with single ventricle physiology and TGA display a faster rate of pre- to post-operative microstructural brain development in white and gray matter brain regions.¹⁰⁶ The neuroprotective effects of a prenatal diagnosis also extend beyond the perioperative period. Children with a prenatal diagnosis of TGA display better neurocognitive outcomes at 4-6 years of age than those postnatally diagnosed.¹¹¹

While prenatal diagnosis plays an important role in decreasing perioperative and long-term neurological risk (potentially via delivery planning and optimization of immediate postnatal critical care¹¹²), there may be some detrimental effects because of the parental psychological distress coincident with the diagnosis. Parental wellbeing is becoming an area of increased focus in the CHD literature. Parents show elevated depression, anxiety, and stress symptoms after receiving a prenatal diagnosis of CHD.^113,114 Prenatal stress exposure is just beginning to be studied for its effects on fetal brain development. Term infant data have shown an association of stress exposure with abnormal development of the amygdala and hippocampus.¹¹⁵ As knowledge of parental psychological wellbeing and its impact on brain development increases, it will be important to evaluate this fetal exposure in the CHD population.

Timing of Delivery is Important for the CHD Brain

Once a prenatal diagnosis is established in a fetus with CHD, timing of delivery becomes an essential component of clinical care. Gestational age at birth is a well-known predictor of neurodevelopmental impairments in cognitive, language, motor, executive functioning, behavioral and psychiatric domains.^116-118 This relationship is also seen with brain injury. Lower gestational age at birth has been associated with greater white matter injury volume¹¹⁹ and increased risk of intraventricular hemorrhage in CHD.¹²⁰ Preterm infants in a cardiac intensive care unit are 3.2 times more likely to display acute neurological events that include intracranial hemorrhage or white matter injury.¹²¹ Early term birth also increases risk of brain injury in CHD. Infants born at 37-38 weeks gestation are 2.5 times more likely to have cerebrovascular injury.¹²² This risk continues to increase as gestational age decreases.¹²² MRI data also support an association between preterm birth and greater disruption of brain development for CHD infants. Diffusion tensor imaging shows microstructural abnormalities across multiple white matter tracts that appear to be driven by white matter injury.¹²³ Compared to preterm infants without CHD and to critically ill term infants, preterm CHD infants display greater microstructural abnormalities in peripheral and deep white matter and in the thalamus.¹²⁴ To reduce morbidity and mortality, including postnatal neurological risks, recent guidelines recommend that elective delivery should not occur before 39 weeks for CHD fetuses.¹¹² Furthermore, an algorithm of delivery planning and resuscitation based on cardiac diagnosis/physiology has been provided to prevent postnatal hemodynamic compromise.¹¹²

Conclusions

Fetuses with CHD display alterations in brain development that span multiple tissues and regions within the developing brain. These deficits are present by the second trimester and progress along an altered trajectory of brain growth that predisposes the CHD infant to increased perioperative neurological risk. The underlying mechanisms of these alterations are likely multifactorial and include hypoxia, genetic and epigenetic variants, placental abnormalities, and prenatal stress exposure. Prenatal diagnosis and delivery at term gestation have neuroprotective effects for the developing CHD brain.