Trajectories of Neurodevelopment and Opportunities for Intervention Across the Lifespan in Congenital Heart Disease

Abstract

Children with congenital heart disease (CHD) are at increased risk for neurodevelopmental challenges across the lifespan. These are associated with neurological changes and potential acquired brain injury which occur across a developmental trajectory, and which are influenced by an array of medical, socio-demographic, environmental and personal factors. These alterations to brain development lead to an array of adverse neurodevelopmental outcomes, which impact a characteristic set of skills over the course of development. The current paper reviews existing knowledge of aberrant brain development and brain injury alongside associated neurodevelopmental challenges across the lifespan. These provide a framework for discussion of emerging and potential interventions to improve neurodevelopmental outcomes at each developmental stage.

Introduction

Advances in prenatal diagnosis, neonatal and perioperative care, and medical, educational and psychosocial care have led to increased survival rates and improved outcomes for individuals with Congenital Heart Disease (CHD). Within the context of these remarkable gains, there are clear patterns of changes in brain development, brain injury risk, and associated neurodevelopmental challenges observed across the lifespan, particularly for those children requiring surgery within the first year of life (Marino et al., 2012).

These neurological differences have been referred to as a unique ‘encephalopathy of CHD,’ with intertwined influences of alterations in typical brain development and acquired brain injury (Volpe, 2014). That is, changes to the trajectory of brain development begin in the fetal period and are further compounded by neonatal and perioperative brain injury. There are a multitude of medical, environmental, personal, and sociodemographic risk factors that impact these trajectories. This is reflected in the resulting spectrum of neurodevelopmental outcomes associated with CHD, which are heterogenous but impact a characteristic set of skills (Cassidy et al., 2017; Sanz et al., 2021).

Most notably, neurological changes and associated neurodevelopmental challenges in CHD occur over the course of development, likely in a developmental cascade, with earlier delays or impairments impacting the continued course of both skill development and brain development (Sanz et al., 2021). This paper will review our existing knowledge of brain development, injury, and recovery over the course of the lifespan in individuals with CHD, along with associated neurodevelopmental challenges. These are used as a frame to explore potential key timepoints for interventions that might interrupt the cascade of changes described, and to review those care strategies and interventions that are currently in use or under investigation, and to reveal understudied opportunities for change.

Prenatal Brain Development in CHD

Human brain development is a complex process with morphologic events occurring across developmental stages, and a prolonged period of refinement of connections that occurs in the third trimester and extends into the early postnatal period. Technical advancements in brain magnetic resonance imaging (MRI) have revealed an increased frequency of structural and developmental abnormalities of the fetal and neonatal brain in the setting of CHD (Peyvandi et al., 2019). It was long thought that developmental abnormalities in children with CHD were secondary to surgical and peri-operative factors, such as the need for cardio-pulmonary bypass in infancy. However, in one of the first neonatal brain MRI studies performed, Miller et al. observed that neonates with complex forms of CHD (d-transposition of the great arteries and single ventricle physiology) have evidence of delayed brain development even before going to the operating room (S. P. Miller et al., 2007). Subsequently, Limperopoulos et al. observed these same patterns in the 3^rd trimester fetus with complex CHD, demonstrating that delayed brain development begins in late gestation (Limperopoulos et al., 2010). Importantly, this divergence from typical patterns of brain development begins in the 3^rd trimester of fetal life—a time of rapid brain growth and development that presumably requires a significant increase in oxygen and substrate delivery. Overall, late gestation fetuses and newborns with significant CHD have smaller brains (Limperopoulos et al., 2010) with simplified cortical gyrification (Clouchoux et al., 2013), less organized white matter tracts (Miller et al., 2007), and immature biochemistry (Limperopoulos et al., 2010). With advancements in fetal MRI, including technology to account for fetal and maternal motion, there is new evidence that fetuses with CHD have decreased cerebral oxygenation (Peyvandi et al., 2021). Importantly, perinatal impairments in brain growth appear to affect subsequent brain growth trajectories (Ortinau et al., 2018) and neurodevelopmental outcomes in infancy (Sadhwani et al., 2022). The etiology of delayed brain development in the late gestation fetus with CHD is likely multi-factorial, with contributions from genetic abnormalities, cardiovascular physiology, and environmental factors.

Cardiovascular physiology

The third trimester of fetal life is marked by a period of rapid brain growth and refinement of connections. Consequently, blood flow to the fetal brain increases and is estimated to be approximately 25% of the combined ventricular output in the third trimester (Rudolph, 2018). In the normal fetus, cerebral blood flow is supplied by highly oxygenated blood from the ductus venosus preferentially streaming across the foramen ovale to the left atrium and ventricle. Depending on the sub-type of CHD, cerebral blood flow and thus oxygen delivery can be impaired. For example, in D-transposition of the great arteries, the aorta and pulmonary artery are transposed, and thus the higher oxygenated blood reaches the pulmonary vasculature as opposed to the aorta. In hypoplastic left heart syndrome, inadequate left heart structures lead to reversal of blood flow in the foramen ovale, with mixing of oxygenated and deoxygenated blood in the right ventricle and in the cases of aortic atresia, retrograde flow in the ascending aorta.

Multi-modal fetal imaging techniques have demonstrated how this altered circulation can lead to flow disturbances affecting in-utero growth and brain development. Cerebral Doppler ultrasound can assess cerebral vascular resistance in the middle cerebral artery (MCA). Several studies have observed altered cerebral blood flow patterns in the fetus with CHD (Donofrio et al., 2003; Kaltman et al., 2005). In particular, fetuses with hypoplastic left heart syndrome consistently have the most abnormal patterns, with decreased resistance in the MCA thought to be an autoregulatory phenomenon in response to decreased flow and oxygenation to the brain, similar to the growth restricted fetus. However, it is unclear if fetuses with CHD have normal autoregulatory capacities of the brain. A recent study measuring MCA vascular resistance at baseline and after brief administration of maternal hyperoxia demonstrated variable response across different sub-types of CHD (Hogan et al., 2021). This may reflect differences in cerebral autoregulatory capacities in the setting of complex CHD. More recently, novel fetal cardiac MRI techniques have been developed, which enables measurements of flow and oxygen saturation in fetal blood vessels. By combining fetal brain MRI and cardiovascular magnetic resonance, Sun et al. found a correlation between fetal cerebral oxygen consumption and brain size (estimated brain weight) among fetuses with complex CHD in late gestation (Sun et al., 2015). Specifically, there was a direct correlation between estimated brain weight and cerebral oxygen consumption and a modest association between cerebral oxygen delivery and brain size. Both ultrasound and MRI based studies support the hypothesis that aberrations in cardiovascular physiology can result in decreased perfusion and oxygen delivery to the brain affecting brain development and increasing susceptibility to brain injury (Peyvandi et al., 2019).

Environmental influences on the developing fetal brain and the role of the placenta

Natural experiment studies and other observational studies have demonstrated a link between prenatal maternal stress and developmental outcomes in offspring among children without congenital anomalies (Scheinost et al., 2016, 2017). The biological mechanism of these findings is hypothesized to be through epigenetic mechanisms at the level of the placenta acting as a mediator of maternal and environmental signals to the developing fetus (Bale, 2015; Nugent & Bale, 2015). Not surprisingly, maternal stress levels are noted to be significantly higher after the diagnosis of a fetal anomaly including fetal CHD (Rychik et al., 2013). In a study of pregnant women carrying a fetus with CHD, 65% experienced significant stress, 44% reported anxiety and 29% reported depression. Furthermore, maternal stress and anxiety was associated with smaller hippocampal and cerebellum volumes in the fetus with CHD (Wu et al., 2020).

Placental pathology is known to be abnormal in pregnancies affected by significant CHD, and there is a relationship between cardiac development and placental health which in combination can contribute to in utero brain development. Common abnormalities on pathologic exam have included thrombosis, infarction, chorangiosis, immature villi and abnormal placental perfusion (Rychik et al., 2018; You et al., 2020). In addition, recent studies demonstrate abnormalities in placental vasculature in the setting of CHD with high rates of vascular malperfusion lesions (Leon et al., 2022). It is unclear whether placental abnormalities precede the development of CHD or if the placental pathology develops secondary to abnormal cardiovascular physiology. There is likely a complex interplay between maternal health, placental function, and developmental programming in the fetus that can have long term effects on outcomes such as neurodevelopment, though future studies are necessary to identify these causal pathways.

Possible prenatal intervention strategies

Given the multi-factorial nature of delayed fetal brain development in utero, identifying an intervention to optimize outcomes is challenging. For certain complex forms of CHD, fetal cardiac interventions to change the natural history of disease have been in use for several years (Schidlow et al., 2017), though the impact of altering cardiovascular physiology on long term developmental outcomes remains unclear (Laraja et al., 2017). Trials and experiments are underway to assess the utility of maternal hyper-oxygenation therapy, which is thought to increase oxygen flow to the brain, though it is unclear if this therapy holds promise given the potential impact on the placenta as well as flow patterns in the fetus (Edwards et al., 2019; Li et al., 2022). Behavioral interventions in utero to manage maternal stress hold great promise in the general population (Li et al., 2022) and can be applied to mothers with a fetal anomaly. Finally, animal models are underway to understand complex and cumulative events in the developing cortex and white matter and for development of potential neuroprotective approach (Leonetti et al., 2019; Morton et al., 2015). Studies using a large animal model found that neural stem/progenitor cells contribute to perinatal corticogenesis, and suggest that restoration of neurogenic potential of the unique cell population is a candidate therapeutic target for improving cortical growth (Morton et al., 2017). In a rodent model of chronic hypoxia, treatment with tetrahydrobiopterin (BH4) mitigated the deleterious effects of chronic hypoxia on the developing white matter (Romanowicz et al., 2019). Because BH4 is already approved by the FDA and shown to be safe during pregnancy, there is translational potential for BH4 to become a new neuroprotective therapy for fetuses with CHD. In addition, genetic etiologies are now recognized as a very important contributor. Integrative approaches involving genetics, cell biology, and molecular biology to model brain development in CHD will play a key role in defining the underlying causes of brain dysmaturation and optimal windows for treatment to improve neurodevelopmental outcomes in CHD.

Neonatal Brain Development and Injury in CHD

As outlined previously, differences in brain development begin in the fetal period in CHD, and may be related to alterations in brain perfusion and oxygenation (Petit et al., 2009; Sethi et al., 2013). Neonates with TGA, ToF, and HLHS have significantly smaller head circumference (HC) than children without CHD (Barbu et al., 2009; Manzar et al., 2005; Rosenthal, 1996; Shillingford et al., 2007), which may be related to these differences in fetal brain development. Newly acquired brain injury is common among neonates with critical CHD, and this population is known to be at high risk for neurologic and neurodevelopmental differences (Donofrio et al., 2011; Licht et al., 2009; S. P. Miller et al., 2007; Sarajuuri et al., 2012; Shillingford et al., 2007). Brain injury and neurologic outcomes are due to multiple, cumulative influences beginning in the fetal period and extending throughout a survivor’s lifetime (Donofrio & Massaro, 2010; Limperopoulos et al., 1999, 2000; Mulkey et al., 2013). .

Trouble Transitioning: Fetal Predictors of Brain Injury

Delayed brain development, in particular, has been shown to correlate with abnormal postnatal brain development and pre-and postoperative brain injury in neonates with CHD (Andropoulos et al., 2010; Beca et al., 2013; Brossard-Racine et al., 2016; Claessens, Khalili, et al., 2019; Dimitropoulos et al., 2013). Preoperative brain injury in CHD appears to be strongly related to microstructural and metabolic brain development (Dimitropoulos et al., 2013), and may be related to differences in the cerebral circulation (Ortinau et al., 2012). Specific markers of delayed brain development, including enlarged ventricular and extra-axial CSF spaces (Brossard-Racine et al., 2016; Claessens, Khalili, et al., 2019), brain maturation scores (Andropoulos et al., 2010; Beca et al., 2013), and abnormal microstructural and metabolic brain development (Dimitropoulos et al., 2013) are associated with more severe pre-and post-operative brain injury.

Cardiac physiology and timing of diagnosis have also been associated with the degree of brain injury (Peyvandi et al., 2016; Peyvandi, Kim, et al., 2018). Single-ventricle physiology has been most clearly associated with a higher risk for postoperative brain injury compared to newborns with TGA (Beca et al., 2013; Limperopoulos et al., 2000; Mcquillen et al., 2007; Mulkey et al., 2013; Peyvandi, Kim, et al., 2018). Interestingly, prenatal diagnosis of single-ventricle heart disease or transposition of the great arteries may be protective, as prenatally diagnosed neonates with CHD have a less severe postnatal brain injury than postnatally diagnosed neonates (Peyvandi et al., 2016).

Contributing Factors to Neonatal Brain Injury

There are multiple risk factors for pre-and postoperative brain injury in neonates with CHD (Table 1). Physiologic factors that have been most commonly associated with brain injury include hypoxemia and time to surgery (Lynch et al., 2014; Petit et al., 2009), hypotension (Dimitropoulos et al., 2013; Galli et al., 2004; Mcquillen et al., 2007), and decreased regional cerebral oxygen saturation (rSO2) (Dent et al., 2005; Mcquillen et al., 2007). Lower preoperative oxygen saturation has been associated with more severe preoperative brain injury in studies evaluating neonates with heterogeneous CHD diagnoses, and in TGA, specifically (Block et al., 2010; Dimitropoulos et al., 2013; Petit et al., 2009). Preoperative hypotension is predictive of preoperative brain injury score in CHD (Dimitropoulos et al., 2013). Autonomic function is immature in infants with CHD (Mulkey et al., 2019), and worse autonomic function correlates with preoperative brain injury scores in this population (Schlatterer et al., 2021).

Table 1:

Risk Factors and Proposed Interventions for Brain Injury in Complex Congenital Heart Disease

	Risk Factors	Supporting Studies	Interventions under Investigation
Prenatal	Immature Brain Development	Miller, S. P. et al., 2007 Limperopulos, C et al, 2010 Andropoulos, D. B. et al. 2010 Beca, J. et.al. 2013 Dimitropoulos, A. et al. 2013 Brossard-Racine, M. et al. 2016 Claessens, N. H. P. et al. 2019	Maternal Hyperoxygenation tetrahydrobiopterin (BH4) Maternal Progesterone
	Cardiac Physiology (Single vs Two Ventricle)	Limperopoulos, C. et al. 2000 McQuillen, P. et.al. 2007 Mulkey, S. B. et al. 2013 Beca, J. et.al. 2013 Peyvandi, S. et al. 2018
	Timing of Diagnosis of Cardiac Disease *	Peyvandi, S. et al. 2016	Decreased time to surgery/Delivery planning
	Placental Pathology	Rychik et al., 2018 You et al., 2020 Leon et al., 2022
	Maternal Stress	Rychik et al., 2013 Wu et al., 2020	Psychological Intervention for Maternal Stress
Preoperative	Lower O2 Saturation	Petit, C. J. et al. 2009 Block, A. J. et al. 2010 Dimitropoulos, A. et al. 2013	Allopurinol Precision monitoring/therapy to optimize blood pressure, oxygenation, and cerebral perfusion
	Hypotension	Dimitropoulos, A. et al. 2013
	Immature Autonomic Function	Schlatterer, S.D. et al. 2021
	Length of Time to Surgery	Petit, C. J. et al. 2009 Lynch, J. M. et al. 2014	Decreased time to surgery
	Balloon Atrial Septostomy	Mcquillen, P. S. et al. 2006 Block, A. J. et al. 2010 Dimitropoulos, A. et al. 2013 Kelly, C.J. et al. 2019
Intraoperative	Prolonged Cardiopulmonary Bypass	Beca, J. et.al. 2013	Non-neurotoxic/neuroprotective anesthesia Mesenchymal Stromal Cells Precision monitoring
	Prolonged Circulatory Arrest	Beca, J. et.al. 2013
	Anesthesia	Andropoulos et al., 2014
Postoperative	Lower O2 saturation	Galli, K. K. et al. 2004	Precision monitoring
	Lower Cerebral rSO2 **	Dent, C. L. et al.2005 Mcquillen, P. S. et al. 2007
	Hypotension	Galli, K. K. et al. 2004 Mcquillen, P. S. et al. 2007 Dimitropoulos, A. et al. 2013
Outpatient	Developmental Cascades Increasing Environmental Demands/Expectations	Sanz, J. et al., 2021	Early Interventions (OT, Speech, PT) Parent-oriented Psychoeducation Neurocognitive Intervention Psychopharmacology Psychosocial Interventions
	Comorbid Medical/Genetic Conditions	Marino, B. et al., 2012 Latal, B. et al., 2016 Wernovksy, G. et al. 2006
	Vascular Processes During Aging Cardiovascular Events	Keir, M. et al., 2019 Melazzini, L. et al., 2019	Psychosocial Interventions Transition Readiness Interventions Expansion of Neurodevelopmental Care
	Socio-Economic and Demographic Factors	Bucholz et al., 2021	Improving access/utilization of services Promoting a more stimulating home environment

^*Peyvandi, S. et al. 2016 found infants prenatally diagnosed with CHD had less brain injury.(Peyvandi et al., 2016)

^**Claessens, N.H.P et al. 2019 found lower cerebral rSO2 was NOT associated with brain injury.(Claessens, Jansen, et al., 2019)

Aberrant electrocortical activity, as measured by EEG and amplitude-integrated EEG (aEEG), is common among neonates with CHD, and is associated with abnormal brain development and brain injury (Mebius et al., 2018; Mulkey et al., 2015). Abnormal background activity on aEEG is highly associated with pre-operative brain injury (60% in one study)(3,4) as well as brain atrophy (low brain volume) prior to surgery (Mulkey et al., 2015). Immature structural and microstructural brain development also correlates with abnormal brain electrical activity (Birca et al., 2016). Neonates with immature structural brain development and pre-operative brain injury had increased high-frequency connectivity on EEG, and neonates with delayed microstructural brain development had weaker low-frequency connectivity on EEG (Birca et al., 2016). Moreover, failure to recover continuous background activity on aEEG by 48 hours post-operatively is associated with mortality and worse neurodevelopmental outcomes on BSID-III at age two years (Gunn et al., 2012). Thus, early differences in aEEG and EEG are highly associated with developmental outcomes and mortality, indicating long-term impact (Birca et al., 2016; Gunn et al., 2012; Mebius et al., 2018; Mulkey et al., 2015).

Postoperatively, brain injury is associated with lower postoperative oxygen saturation (Galli et al., 2004), hypotension in the first 24–48h following surgery (Dimitropoulos et al., 2013; Galli et al., 2004; Mcquillen et al., 2007), and lower postoperative regional cerebral rSO2 (Galli et al., 2004; Mcquillen et al., 2007). However, one recent study did not find an association between lower postoperative cerebral rSO2 and postoperative brain injury (Claessens, Jansen, et al., 2019).

Surgical considerations associated with neonatal brain injury in CHD include time to surgery (Lynch et al., 2014; Petit et al., 2009), need for balloon atrial septostomy (BAS) (Block et al., 2010; Dimitropoulos et al., 2013; Kelly et al., 2019; Mcquillen et al., 2006), and prolonged bypass and circulatory arrest times (Beca et al., 2013). Both pre- and postoperative white matter injury are associated with a longer time between birth and surgery (Lynch et al., 2014; Petit et al., 2009). This association may be explained by a progressive decrease in cerebral tissue oxygenation between birth and surgery in TGA or HLHS (Lynch et al., 2018). BAS is associated with stroke in several studies (Block et al., 2010; Dimitropoulos et al., 2013; Kelly et al., 2019; Mcquillen et al., 2006), and prolonged cardiopulmonary bypass and circulatory arrest times correlate with new postoperative white matter injury (Beca et al., 2013). Overall, infants with CHD undergo a complex clinical course during which they are likely to experience multiple physiologic changes that may contribute to development of brain injury.

Neuroimaging predictors of later outcomes

Neonates with CHD may experience two separate but related processes that contribute to neurodevelopmental outcomes, namely, delayed brain development and brain injury. Structural brain immaturity during the neonatal period is associated with neurodevelopmental outcomes in survivors of CHD at 2 years of age (Beca et al., 2013), and another study revealed an association between smaller cortical gray matter and cerebellar volumes and lower fine motor scores at 9 and 18 months (Stegeman et al., 2022).

The relationship between brain injury and neurodevelopmental outcomes in CHD is an area of active investigation. Neonatal white matter injury, particularly involvement of the posterior limb of the internal capsule, is associated with lower IQ and more severe motor and attention deficits in school-age survivors of CHD (Claessens et al., 2018). One study showed that preoperative white matter injury correlates with lower Bayley Scales of Infant and Toddler Development III scores at age 12 months in CHD (Andropoulos et al., 2012). Another study found that moderate to severe WMI before or after surgery is associated with lower motor scores at 2.5 years, but not 12 months (Peyvandi, Chau, et al., 2018). Clinically silent, small neonatal strokes were not associated with adverse ND outcomes and even trended to better outcomes in TGA (Peyvandi, Chau, et al., 2018). However, there does appear to be an association between acute ischemic stroke in the corticospinal tract and abnormalities of muscle tone and gross motor delay, and severe ischemic brain injury is associated with a diagnosis of cerebral palsy in CHD (Stegeman et al., 2022). In addition, less is known regarding the association between early alterations in brain development or brain injury and longer-term neurodevelopmental outcomes in school age, adolescence, or adulthood.

Genetic Influences on Neurodevelopment

CHD has strong heritability, and a genetic abnormality can be identified in up to 50% of syndromic CHD and 10% of non-syndromic sporadic cases (Homsy et al., 2015). Genetic anomalies include chromosomal duplication or deletion (aneuploidy, 13%), copy number variation (10%), de novo single nucleotide variant (10%) (Zaidi & Brueckner, 2017). Aneuploidies were the earliest genetic anomalies to be linked to CHD and have well described association with adverse neurodevelopmental outcome (e.g., Down, Turner syndrome). Gaynor et al. were first to focus on the importance of patient specific factors, including genetic syndromes and ApoE genotype for neurodevelopmental outcome in early single center studies (Gaynor et al., 2007) subsequently confirmed in secondary analyses of prospective trials (Newburger et al., 2012). Review of published outcome studies over two decades, found the presence of genetic/extracardiac anomaly as a significant risk factor for worse neurodevelopmental outcome (Gaynor et al., 2015). With improvements in genomic sequencing technology, multicenter projects have embarked on large scale sequencing efforts (Hoang et al., 2018). De novo variants in genes with high heart and brain expression were identified in 20% of subjects with CHD and neurodevelopmental delay but only 2% with isolated CHD (Homsy et al., 2015). Rare de novo variants in isolated CHD overlap substantially with those identified in children with autism (Jin et al., 2017). These include genes involved in chromatin remodeling, cilia genes and signaling pathways involving Notch and Ras known to be important for both brain and heart development (Zaidi & Brueckner, 2017). Additionally, de novo damaging variants were associated with post-operative outcomes in CHD (Boskovski et al., 2020). Together, this suggests that genetic variants can affect neurodevelopmental outcomes independently, but also as potential modifiers of neurodevelopmental outcomes in CHD patients. This highlights the need for a systematic search and inclusion of genetic risk factors in neurodevelopmental outcome prediction models in CHD.

Neurobehavioral Outcomes in the Neonatal Period

Neurodevelopmental impairments are present even in the earliest stages of development. Neonatal evaluations have suggested a range of impairments, that are present even prior to surgical intervention. Compared with typically developing term neonates, neonates with CHD are known to have poorer automatic and motor regulation, lower attention and arousal scores, poorer self-regulation, and poor feeding (Butler et al., 2017, 2019; Desai et al., 2023; Hogan et al., 2018). Neonatal attentional and self-regulation scores have been of particular interest, since they may relate to improved feeding at discharge (Gakenheimer-Smith et al., 2019) and to later motor delays, but not later cognitive or language delays (Campbell et al., 2022). Poor feeding also predicts worse motor outcomes (Medoff-Cooper et al., 2016).

Interventions in the Neonatal and Perioperative Period

Identified risk factors for brain injury and altered development can suggest opportunities for neuroprotective interventions that may prevent injury and promote development, leading to improved neurodevelopmental outcomes.

Potential roles of acute and chronic hypoxic ischemic and inflammatory mechanisms have led to trials of maternal progesterone in the fetal period (NCT02133573) and perioperative allopurinol (NCT04217421). An initial single center trial of allopurinol before, during and after surgery did not show significant benefit to infants with two ventricle physiology but may have decreased adverse cardiovascular events (Clancy et al., 2001). The consistent finding that increased time to surgery is associated with risk of WMI in both TGA and HLHS newborns (Lim Jessie Mei et al., 2019; Lynch et al., 2014; Petit et al., 2009) suggests that early surgery, when feasible, would be an effective non-pharmacologic intervention.

Lower postoperative blood pressure has been associated with increased risk of new postoperative white matter injury (Galli et al., 2004; Mcquillen et al., 2007), suggesting the simple intervention of targeting higher blood pressures with preload, inotropic support or vasopressors. While this is likely to be well tolerated and beneficial in newborns with TGA, concern arises for postoperative management of HLHS where intraoperative and postoperative afterload reduction have improved survival after the Norwood procedure by preventing episodes of elevated systemic vascular resistance and low cardiac output syndrome (Hoffman et al., 2021). To achieve the optimal balance between cerebral perfusion and overall cardiac output with single ventricle physiology will likely require individualized precision therapy guided by new methods for bedside monitoring of cerebral blood flow, oxygenation and autoregulation along with cardiac output.

A final area of concern and opportunity relates to exposure of the vulnerable neonatal brain to repeated anesthetics, narcotics and benzodiazepines that have been associated with neurotoxicity and programmed neuronal cell death (Ing et al., 2022; Sarić et al., 2022). These findings, while well established in animal models, are a topic of multiple clinical research studies to establish their significance in humans (Andropoulos et al., 2014; McCann et al., 2019). Nevertheless, neuroprotective clinical trials have been proposed using putative non-neurotoxic and possibly directly neuroprotective anesthesia and sedation with dexmeditomidine. Trials are also underway examining the delivery of mesenchymal stromal cells through cardiopulmonary bypass as a neuroprotective measure, as this has been shown to protect stem/progenitor cells in the subventricular zones, and to promote migration of neuroblasts in animal models (Maeda et al., 2020). A major challenge to neuroprotective trials in CHD newborns is the need for early surrogate outcome measures that reliably predict long-term childhood and adult neurodevelopmental outcomes.

Developmental care for Neonates and Infants

Recently, more focus has been placed on neurodevelopmental care in the CICU setting (Lisanti et al., 2016; Torowicz et al., 2012). Individualized developmental care in the ICU setting is known to improve neurobehavioral functioning and structural development in preterm infants (Als et al., 2004). Currently, developmental care is variable among CICUs within North America (Sood et al., 2016), in part because critically ill infants with CHD face unique challenges (Lisanti et al., 2019). Proposed models for developmental care in the CICU include the following elements: parental engagement/family-centered care; an individualized, cue-based plan; and an emphasis on a supportive hospital environment, with particular attention to circadian rhythms, noise/light levels, feeding, social interaction/play, kangaroo care, and developmentally considerate approaches to medical procedures (Desai et al., 2023; El-Farrash et al., 2020; Lisanti et al., 2019; Peterson & Evangelista, 2017). Interdisciplinary engagement, including parents, speech/language pathologists, physical/occupational therapy, child/family life therapists, psychology, social work, dieticians, nurses, and physicians is critical for implementation of any successful developmental care program in the CICU (Butler et al., 2017; Peterson & Evangelista, 2017), and a well-planned process of implementation with a strong commitment from leadership and staff is necessary for success (Sood et al., 2016). The inclusion of parents, families, and children with CHD in the design of supportive psychosocial interventions is also critically important to ongoing research (Sood et al., 2022). Many CICUs are beginning to implement their own developmental care programs based on the above ideas, and the community of caregivers and families of CHD patients await further studies to determine their long-term impact.

Neurodevelopmental Outcomes from Infancy to Adulthood

The broad range of influences on the brain, beginning in fetal life and continuing through the neonatal and perioperative periods, fundamentally alter the trajectory of brain development and associated neurodevelopmental skills. Just as CHD is heterogeneous, outcomes can be quite varied, and often involve a range of neurocognitive domains (Cassidy et al., 2017).

For infants and toddlers, early difficulties with feeding, motor development, language acquisition, and self-regulation are observed (Brosig et al., 2007; Clancy et al., 2020; Latal, 2016; Ware et al., 2020). As children enter school, which presents increased cognitive, academic, and social-emotional demands, additional challenges become evident. Specifically, weaknesses in attention, executive functioning, visual-spatial processing, and academic achievement emerge (Bellinger & Newburger, 2010; Cassidy et al., 2015, 2017; Griffin et al., 2003; Sanz et al., 2017), as well concerns for social cognition, mood, emotional and behavioral regulation, and/or anxiety (Bellinger et al., 2009; Kovacs et al., 2009; Wilson et al., 2015). There are also higher rates of neurodevelopmental disorder diagnoses (ADHD, autism spectrum disorder, learning disabilities) in children with CHD compared to the general population (Loblein et al., 2022; Razzaghi et al., 2015; Ryan et al., 2019; Tsao et al., 2017; Verrall et al., 2019), with increased risks and poorer outcomes for those with comorbid genetic conditions (Latal, 2016; Marino et al., 2012; Wernovsky, 2006). Further, deficits in these areas can hinder the development of daily living (adaptive) skills needed to successfully navigate the transition to adulthood and ultimately achieve independence (Cassidy et al., 2017; Ilardi et al., 2017).

In addition, with improving care, there is now a growing population of adults with CHD, and adults with CHD now outnumber children with CHD (Marelli et al., 2014). Despite this, we have an extremely limited understanding of the continued evolution of neurodevelopmental concerns over the lifespan, particularly during the aging process. Though mechanisms are poorly understood, underlying disease factors present increased risk for vascular problems and acquired cardiovascular events (Keir et al., 2019; Melazzini et al., 2019). These presumably contribute to substantially increased rates of dementia and neurocognitive decline in older adults with CHD, a topic which remains woefully understudied (Keir et al., 2019).

Though research up to this point has generally focused on specific skills sets or diagnoses in a cross-sectional fashion, researchers and clinicians have been encouraged to shift their conceptual framework away from static models that define a unitary, “neurodevelopmental signature,” towards a dynamic conceptual framework that includes a broader range of outcomes that interact and evolve over time, in a developmental cascade (Sanz et al., 2021). When viewed this way, interventions across the lifespan become important to interrupting these cascades and promoting improvement in neurodevelopmental trajectories. An emerging lifespan perspective also becomes critically important.

Interventions Across the Lifespan

Despite this, to date there has been notably less focus on the development and implementation of interventions for the CHD population, with only a small number of single-center investigations focused on interventions for children, adolescents, and young adults with CHD (Calderon & Bellinger, 2015; Cassidy et al., 2021). The efficacy of intervention programs developed and implemented for high-risk populations such as children born preterm and those diagnosed with chronic medical conditions and/or neurodevelopmental disorders is well documented in the literature (Case-Smith, 2013; Spittle et al., 2007). As there are many specific areas of concern that are shared between chronic medical conditions, including CHD, these interventions offer a reasonable starting point for developing intervention strategies.

Interventions in Infancy and Early Childhood

The importance and efficacy of direct early intervention services (e.g., physical therapy, occupational therapy, speech and language therapy) for children who are at increased risk for or diagnosed with neurodevelopmental disability is well established (Majnemer, 1998; Nores & Barnett, 2010). While these services are frequently utilized by young children with CHD, little is known about the longitudinal impact of these child-directed therapies for this specific population. Parent-oriented psychoeducational interventions for young children with CHD, investigated through a series of controlled trials conducted as part of the Congenital Heart Disease Intervention Project (CHIP), have demonstrated improved cognitive development, maternal adjustment, and family functioning (McCusker et al., 2010, 2012; van der Mheen et al., 2019).

Interventions for Attention/Executive Function

Across medical and neurodevelopmental disorder populations, there is a growing interest in neurocognitive interventions, particularly to address executive functioning deficits; however, to date, the effectiveness of these programs is mixed (Diamond & Ling, 2016; Melby-Lervåg & Hulme, 2013). A single-center study examining the effectiveness of a computerized neurocognitive intervention (Cogmed) for adolescents with CHD found no improvement in the primary outcome measure (working memory), but improvements in inhibitory control and parent-report of cognitive regulation were reported at 3-month follow-up (Calderon et al., 2020). Pharmacological treatment of symptoms of ADHD (inattention, hyperactivity, impulsivity) is often the most effective treatment for children and adolescents (Wolraich et al., 2019). The AHA released a scientific statement in 2008 (Vetter et al., 2008), with considerations for individuals with CHD recommending thorough evaluation and continued monitoring due to the cardiovascular effects of stimulant medication. However, to date there have not been any studies investigating the efficacy of these medications for individuals with CHD with or without diagnosis of ADHD. This is also the case for pharmaceutical treatment of mood and anxiety disorders in this population, which has not been directly studied.

Social-Emotional Functioning and Transition to Adulthood

Interventions to address social-emotional functioning are especially crucial during adolescence and adulthood given the bidirectional relationship between psychosocial and medical health (Kovacs & Bellinger, 2021), though few programs have been evaluated. Aerobic/physical activity programs have been shown to have positive effects on psychosocial outcomes, cognitive functioning, and quality of life based on self- and proxy-report (Dulfer et al., 2017). While some psychosocial intervention protocols have been described and determined feasible for adolescents and adults with CHD, there is limited evidence for the efficacy of these interventions (Tesson et al., 2019).

In addition to interventions that directly target psychosocial and cognitive health in adolescents and young adults, interventions to address loss to medical follow up have become more important. Namely, transition to continued specialized medical care is critical to reduce morbidity and mortality, and additional invasive interventions, which presumably would also improve psychological and neurocognitive outcomes over the remainder of the lifespan (Kollengode et al., 2018; Moceri et al., 2015; Nitta et al., 2021). There is also a more recent focus on developing interventions focused on preparing adolescents with CHD to enter adult care. One such nurse-led program has demonstrated efficacy in improving transition readiness (Mackie et al., 2018). Implementation of another CHD Transition Clinic intervention, including focused teaching, completing of self-assessment questionnaires, and tracking through a clinical registry, resulted in improved follow-up rates and self-ratings of transition readiness (Gaydos et al., 2020). Additional clinical trials geared towards improving independence and mental health over the course of the transition to adulthood are underway (Saarijärvi et al., 2021).

Despite the rapidly increasing population of older adults with CHD, the topic of monitoring and intervention in this group remains understudied and in its early phases (Keir et al., 2019). From a practical standpoint, few centers offer neurodevelopmental monitoring past adolescence and young adulthood; indeed, most centers still focus care on early childhood (T. A. Miller et al., 2020). While expansion of neurodevelopmental care into adulthood presents logistical challenges and would require collaboration across specialties, this should be viewed as a critical opportunity to improve long term care and outcomes.

Dilemmas and Future Directions

In individuals with CHD, alterations to the trajectory of brain development begin in utero, evolve over time, and are compounded by additional injury. These interact with a multitude of medical and demographic factors and lead to a heterogeneous array of adverse neurodevelopmental outcomes. Improved understanding of brain development in the context of these factors provides us with a conceptual framework for the development of targeted interventions that might improve neurodevelopmental trajectories and outcomes.

Despite the promising interventions described above, there are several gaps in our understanding of neurodevelopment in CHD that limit our progress. Firstly, the majority of research on neurodevelopment in CHD is cross-sectional, with fewer longitudinal studies in limited cohorts (for a review, see Sanz et al., 2021). As a result, we have a poor understanding of the relationship between early neurological and neurodevelopmental findings and later outcomes at school age, adolescence, and young adulthood. It is likely that many of the impacts of these early alterations to fetal and brain development manifest in later development, though these connections have not yet been clearly characterized. We also have limited data regarding the efficacy of earlier interventions on subsequent neurodevelopmental trajectories and outcomes. Additionally, the heterogeneity of this patient population presents a significant research challenge. These factors highlight the need for larger, collaborative clinical data registries that will allow for more complex analyses of factors that contribute to neurodevelopmental trajectories across the lifespan. Fortunately, these collaborative efforts are underway, with the Cardiac Neurodevelopmental Outcome Collaborative and Cardiac Networks United leading effort to establish longitudinal, multisite registries to answer these questions (Gaies et al., 2019; Marino et al., 2020).

Additionally, social determinants of health have consistently been a significant contributor to both medical and neurodevelopmental outcomes in CHD (Lopez et al., 2022). These social determinants of health impact even the earliest stages of care, with lower rates of prenatal diagnosis (Krishnan et al., 2021) and poorer surgical outcomes (Gallegos et al., 2022) in specific demographic groups. Emerging research also strongly suggests that socioeconomic factors are important, primary drivers of neurodevelopmental outcomes in CHD (Bucholz et al., 2020; Favilla et al., 2021), and that the impacts of socioeconomic factors not only persist, but become more pronounced over the course of a child’s development (Bucholz et al., 2021). Our understanding of racial and socioeconomic contributors to neurodevelopmental outcomes is inadequate and needs to become a focus of continued research. Important barriers to address include improved inclusion of diverse populations in research, development of improved culturally and linguistically appropriate measurement tools for neurodevelopmental and psychological outcomes in diverse populations, and improved representation of diverse populations in our clinical and research teams (Sanz et al., 2021). Though addressing these barriers will improve our understanding of the issue, we also need clear, actionable changes to our care pathways and specific interventions to address these inequities in neurodevelopmental and health outcomes. Examples of this may include interventions to improve access to neurodevelopmental evaluations, which may improve access to needed therapies or early intervention services that are currently underutilized in CHD (Mussatto et al., 2018; Brosig Soto et al., 2011). Other possible interventions include supporting parents in providing a more stimulating home environment to promote cognitive development (Bonthrone et al., 2021).

Along these lines, we suspect that addressing disparities in access to clinical care, specifically equitable participation in formal neurodevelopmental follow-up programs, is critical. This is needed to provide earlier identification and clear documentation of neurodevelopmental problems, which is often required in order to access needed therapies and services in the community. Despite this, there are likely financial barriers (e.g., insurance coverage), language barriers, and cultural barriers that reduce access to needed care and thus create disparities in neurodevelopmental trajectories. Breaking down these barriers to care across the lifespan will also be critical to improving outcomes for all individuals with CHD.

In sum, we are now developing a better understanding of neurological and neurodevelopmental trajectories in individuals with CHD. This understanding should now begin to serve the development and evaluation of specific intervention strategies to improve the trajectory, with targets for intervention across the lifespan, beginning in fetal life and continuing through adulthood. The development of large-scale registries and collaborations will be necessary to better understand the long-term consequences of these interventions. Finally, improved inclusion of diverse populations in our clinical and research efforts will be critical to begin to weave in an understanding of the impacts of social determinants of health.