Congenital heart disease (CHD) remains a significant burden for thousands of infants born each year. Even with marked improvement in inpatient and surgical care, optimizing the transition to the outpatient medical home continues to be challenging. A keen understanding of risk stratification on the basis of cardiac anatomy and physiology, socioeconomic determinants of health, gestational age at birth, comorbidities and genetic syndromes, along with the need for potential technological assistance underlie the challenges in discharging and managing infants with all types of CHD. Outpatient feeding and nutrition outcomes remain important in the long-term success of these infants and continue to be the focus of future quality initiatives. This is based on the evolving evidence that morbidity in this arena continues through early childhood and interventions focused on infants may have drastic effects long-term. Engaging outpatient primary care providers, specialists, allied health providers, and parents in the discharge process is paramount in ensuring a smooth transition and reducing use of tertiary care resources. Furthermore, comprehensive home monitoring programs are essential in the management of certain high-risk infants with CHD and are continually being expanded to support all infants who may need ongoing assistance. Guidelines for immunoprophylaxis and vaccine administration continue to change on the basis of emerging evidence and represent potentially modifiable variables to reduce the burden of infection. Finally, exploring how ongoing use of tertiary care resources, readmissions, and emergency department encounters will continue to shed light on the importance of care coordination to maintain infant’s health while at home. About 40 000 infants with CHD (CHD) are born each year in the United States.1 Although survival has improved, issues remain when going home from the hospital after complex surgery or even while awaiting future procedures. Risk factors for mortality and morbidity postdischarge from the hospital from heart surgery are well-reported in infants with hypoplastic left heart syndrome (HLHS);2 yet, studies examining out-of-hospital outcomes for other defects are limited. Fatal adverse events and “near misses” in the first year after discharge from intervention occur at a 7% rate, emphasizing the importance of outpatient care for patients with CHD.3 Furthermore, although certain anatomic subtypes, such as the single ventricle, carry the highest risk, the presence of neurodevelopmental disability, genetic syndromes, ongoing respiratory support, and inability to orally feed can have profound effects on postdischarge.3–6
Consequently, we must educate and provide support for parents and outpatient caregivers when transitioning a neonate with CHD from the hospital to the home. The education should start quickly after the diagnosis is made, and recommendations should focus on attributes specific to the lesion and the patient. It remains critical to identify a primary physician for the infant, with transfer of pertinent information to the provider before the infant leaves the hospital as part of comprehensive home-care plan.7 Careful assessment of risk-stratification on the basis of congenital heart lesion, coupled with factors specific to the patient, can optimize communication with the outpatient team and set an expected framework for the infant’s posthospital care.
This manuscript is part of a larger series of articles simultaneously published as a Supplement in Pediatrics by the Neonatal Cardiac Care Collaborative (NeoC3). Please refer to the Executive Committee introductory paper for discussion on Class of Recommendations and Level of Evidence (LOE), writing committee organization, and document review and approval.
Risk Stratification of CHD by Lesion
Low-Risk Lesions
Isolated left-to-right shunt lesions including atrial septal defects, ventricular septal defects, complete atrioventricular septal defects, and patent ductus arteriosus are the most common forms of CHD.1 Flow through the defect is dependent on the size of the communication, as well as the pulmonary vascular resistance. Because of elevated pulmonary vascular resistance in neonates, newborns with isolated left-to-right shunts rarely develop symptoms at birth. Education about symptoms as the pulmonary vascular resistance falls is essential, because this can affect growth and development and susceptibility to respiratory infections. Families should be counseled on symptoms of pulmonary overcirculation, including tachypnea, diaphoresis, poor feeding, and failure to thrive. The timing of the development of these symptoms can be variable in the first few months of life; thus, open and frequent communication between the cardiologist and the primary care provider remains essential.
Moderate-Risk Lesions
Some forms of CHD can progress in the first few months, requiring close surveillance by both parents and providers. Cyanotic CHD such as tetralogy of Fallot may have progressive cyanosis with age, with the potential for hypercyanotic spells. Discussion of signs and symptoms with parents is essential, and potential home treatments should be discussed. Infants with left ventricular outflow tract obstruction, such as aortic stenosis or aortic coarctation, can develop ventricular dysfunction related to progression of stenosis and thus heart failure symptoms. Tachypnea and difficulty feeding should prompt an urgent evaluation, with assessment of perfusion and pulses. Some forms of CHD can have postoperative changes, such as the development of pulmonary venous obstruction after repair of total anomalous pulmonary venous return. Respiratory distress, sometimes manifesting as recurrent wheezing or “bronchiolitis,” needs to be urgently evaluated in any child with CHD.8
High-Risk Lesions
Perhaps the most well-reported high-risk lesions being managed at home are HLHS and other single ventricle variants. Mortality for infants with HLHS, whereas shunt-dependent, historically was 12%4 with the need for supplemental tube feeding, growth failure, earlier gestational age, specific anatomic subtype, and history of seizures found to be risk factors for decompensation at home.5,9 Although significant center variation exists, ∼75% of patients with HLHS are diagnosed prenatally10 with initial discussion focused on delivery plans and timing of surgery. Nevertheless, it is critical that the challenges of the interstage period (time between Stage 1 and Stage 2 palliative procedures) be discussed and that a comprehensive home monitoring program be considered for these families (Class I, Level B-NR). Other high-risk lesions that can compromise cardiac output have risk for at home decompensation and include Ebstein anomaly with significant tricuspid valve regurgitation, cardiomyopathy with reduced function, and substantial residual left ventricular outflow tract obstruction.
Risk Stratification of Infants With CHD by Gestational Age and Birth Weight
Gestational age (GA) is the most significant predictor of mortality and significant morbidities in extremely preterm infants without congenital anomalies.11 It is well known that survivors of preterm birth often suffer from severe neonatal morbidities, including chronic lung disease, retinopathy of prematurity, sequelae of necrotizing enterocolitis, intraventricular hemorrhage, or periventricular leukomalacia.11–13
This becomes even more important when prematurity is associated with a need for cardiac surgery because the risk of these specific complications is increased. There is a negative relationship between critical CHD morbidity and prematurity with lower GA leading to a higher risk of significant morbidities. In infants under 29 weeks, at least 1 severe neonatal morbidity was present in > 60% of infants with critical CHD compared to ∼35% of infants without CHD. This is in comparison to term infants in which only 3% develop significant morbidity with critical CHD compared to 0.1% without CHD.14–17 Although not as prominent, this effect is also demonstrated in late preterm infants, born between 34 completed weeks and 36 weeks and 6 days of gestation.14 One year, all-cause mortality outcomes mirror this finding with only 60% survival in infants with critical CHD born < 29 weeks of gestation. Furthermore, >80% of neonates with CHD born <29 weeks have suffered either a mortality event or continue to have significant morbidity,14,16,18 The culmination of these findings is a higher proportion of infants that will require more significant outpatient coordination and observation than their term counterparts.
In addition to GA, birth weight, especially as it relates to gestational age, has a significant effect on both operative and postoperative outcomes. Just being born <2.5 kg portends an elevated risk for overall survival.19,20 Furthermore, being small for gestational age (SGA) in addition to being low birth weight is an important predictor of mortality in infants with CHD undergoing cardiothoracic surgery <2 months of age.20
Hence, for infants with CHD, risk stratification is required depending upon the GA and birth weight.
Recommendations:
Neonates with CHD born prematurely, at low birth weight, and/or SGA require significant care coordination and assessment of their risk factors during follow-up to minimize the long-term impact on their survival and quality of life (Class 1, LOE B-NR)
Focused attention on outpatient care coordination in neonates with CHD born late preterm or early term can be beneficial secondary to the higher rates of morbidity associated with this specific population (Class IIa, LOE B-NR)
Follow-up strategies designed to stratify infants by birth weight can be effective, because SGA infants with CHD have a higher risk of mortality and morbidity than infants with GA appropriate birth weight (Class IIa, LOE B-NR)
Risk Stratification of Infants With CHD on the Basis of Socioeconomic Status and Race
Socioeconomic determinants of health play an important, yet challenging role in risk stratifying infants at the time of discharge. A lower socioeconomic status along with high rates of local area-based poverty has frequently been associated with worse postdischarge survival, medium-term survival, rates of readmissions, and higher morbidity scores.21–25 Conversely, higher socioeconomic status along with greater parental education has been linked to improved short-term and long-term outcomes including mortality across all types of CHD.23,24 Crowded living conditions and smoking increase the risk of acquired infections within the first year in infants with CHD.26 Black and Hispanic patients with CHD demonstrate worse infant survival and higher rates of readmissions through the first year of life.24,27 Consequently, while arranging follow-up for infants soon after cardiac surgery, social determinants of health of the entire family, especially as it impacts transportation, housing and food security, education of the parents, family members smoking status, and overall living arrangements, clearly should be taken into consideration.
Recommendations:
Racial and socioeconomic disparities exist in both delivery of care and outcomes for infants with CHD and should be considered to properly coordinate the infant’s outpatient care. (Class IIb, Level B-NR)
Risk Stratification of Infants With CHD on the Basis of Parental Health
Postpartum depression and its effects on neonatal outcomes after discharge has been well established in all types of neonates. Specifically, in children with CHD, 38% of parents report high rates of emotional distress28 and the presence of maternal distress has been strongly linked to infant social withdrawal after cardiac surgery lasting well after discharge.29
Recommendations:
Although direct links between postpartum depression and outcomes in CHD have not previously been established, extrapolation from other neonatal conditions should encourage providers to take this into context while transitioning the patient to the care of the parents at home (Class IIb, LOE C-EO)
Risk Stratification of Infants With CHD on the Basis of Location
Although there is no published evidence showing that living far away from a tertiary center significantly increases the risk of morbidity or mortality, taking care of infants with CHD, especially with critical CHD or single ventricle physiology, who live at a greater distance from a tertiary facility remains challenging.
Recommendations:
Poor access to emergency services, local hospitals with minimal experience managing infants with CHD, lack of broadband connectivity, and distance from tertiary care facilities are challenges that may need to be factored into the discharge process (Class IIb, LOE C-EO)
Need for Outpatient Nutrition, Feeding Support, and Allied Health Services
Neonates with CHD are at increased risk for neurodevelopmental morbidity in both cognitive and motor domains.30 Concomitant with neurodevelopmental morbidity are deficiencies in both the feeding and nutrition spheres.30 Frequently, infants discharged from the hospital in the postoperative period are discharged with supplemental feedings because of either growth failure or poor oral feeding skills.30,31 Importantly, infants orally fed at discharge from the hospital, and who continue to feed orally, have been shown to have higher mental development index and psychomotor development index scores at 6 months of age.30
Careful attention should be shown to infants with a high risk of needing an indwelling feeding tube.32 Infants considered at high risk for feeding difficulties that will benefit from more intensive outpatient therapy include those with an underlying genetic disease, those who had a prolonged postoperative course, high cardiac disease complexity, and low birth weight.32,33
Weight gain is a challenge associated with the feeding difficulties in neonates with CHD. Possible causative etiologies for growth failure include an ongoing catabolic state of the neonate, altered cardiac hemodynamics, as well as poor oral feeding mechanics.31 Importantly, few infants with CHD consistently use breast milk as an outpatient given perceived barriers to breastfeeding or human milk in this population.31,32 Possible perceived barriers include the false belief that breastfed infants suffer desaturations in higher number and expend more calories when compared to bottle-fed infants, a misconception that practitioners are unable to assess transfer from the breast, and growth failure.31,34
Growth failure, regardless of feeding route or type of milk, should prompt evaluation by the primary cardiologist in conjunction with a nutritional assessment.34 Expectations of dietitians’ consultation should include: measurement of anthropometric and percentiles, feeding regimen assessment, calculated nutrition (both volumetric and caloric), and review of supplemental services.35 Speech therapy services may also be warranted if there is concern that the infant does not possess adequate feeding skills.34 Growth failure in the absence of poor feeding skills or hemodynamic compromise may initially be addressed with either an increase in volume or caloric density.34 In addition, the majority of these infants would benefit from state-specific Early Intervention services. Please refer to the Nutritional article in this supplement for further discussion.
Recommendations:
Understanding that human milk is beneficial to the preterm neonates, it follows that it may also confer benefits to the neonate with CHD, though this is an extrapolation (Class IIb, LOE C-EO)
Poor oral feeding skills require more urgent supplemental tube feeding as growth failure has a significant association with poor outcomes, especially in infants with single ventricle physiology (Class I, LOE B-NR)36,37
In addition to speech and feeding challenges, neonates with CHD are also at risk for motor development delay. Despite the well-known association of certain types of CHD with motor delay, few trials examining the efficacy of ongoing physical therapy exist.38 Those studies that exist vary greatly in design, inclusion criteria, assessment of risk factors, attrition, and outcome measures.38 Postsurgery rehabilitation programs are present in adult cardiology, yet despite available data, they are infrequently used.38 Similar to risk factors for poor feeding, infants born: prematurely, with a genetic syndrome, higher disease complexity, or with single ventricle anatomy are at increased risk for requiring some sort of rehabilitation therapy post discharge.38 Importantly, delays in all domains are worse for infants who require feeding assistance technologies.30,38,39
Tracheostomy and CHD
Although nutrition and neurodevelopmental support are the 2 most prevalent allied health systems engaged, there are other technical aspects to discharge that must be addressed. Respiratory support, including tracheostomy, adds an additional element of risk when discharging the neonate after congenital heart surgery. Tracheostomy is rare in CHD, requiring placement between 0.8% and 2% of the time.40–42 Readmissions, however, are incredibly frequent, occurring on average 2.4 times per patient-year in infants in whom tracheostomy was placed before discharge.43 Owing to the significant outpatient burden and risk that a tracheostomy places on the patient in the medical home, mortality is high in the short-term period with 1- to 3-year survival only reaching 12% to 68%.40,42–45
Recommendations:
Careful planning and assessment of the home environment, socioeconomic factors, primary care provider comfort level, and local medical services can be beneficial when transitioning the patient from a tertiary care facility to the medical home (Class IIa, LEO C-EO)
Impact of Genetic Syndromes
Genetic syndromes, especially Trisomy 21, account for nearly 30% of extracardiac comorbidities (ECC).46 Suspicion of aneuploidy based on the type of cardiac condition, ECC, or physical exam findings should prompt swift evaluation, if not done previously. Trisomy 21, 22q11 deletion syndromes, Turner Syndrome, Williams Syndrome, Trisomy 13, Trisomy 18, and Noonan Syndrome comprise the majority of known genetic disorders highly associated with CHD affecting neonates, although many more single gene defect disorders that would affect outpatient management within the medical home should be considered.7 Furthermore, connecting parents to local, state, and national support organizations for various genetic syndromes can be extremely beneficial. Please refer to the fetal cardiology section of this supplement for a more complete list of genetic syndromes associated with congenital heart defects. (REF)
Recommendations:
Associated morbidities due to a patient’s genetic syndrome may be reasonably considered when coordinating the infant’s outpatient care plan (Class IIb, LOE C-EO)
Importance of Primary Care Providers and the Medical Home
Primary care providers (PCP) working in medical homes are tasked with providing and coordinating the multiple health care needs of infants with CHD. The American Academy of Pediatrics and the American College of Cardiology assembled a team of experts, including representation from the American Academy of Pediatrics Committee on Practice and Ambulatory Medicine, to develop a policy statement with the goals of improving patient outcomes and influencing the care of these neonates.7 This includes emphasis on ongoing and essential communication among caregivers, the heart center team, and the PCP (Fig 1). Providers in heart centers must make clear communication with the PCP a priority and should strive to create systems ensuring success.
Example checklist ensuring careful communication among all stakeholders: family, heart center team, and primary care team.
Example checklist ensuring careful communication among all stakeholders: family, heart center team, and primary care team.
CHD is often associated with the presence of chronic ECC. The presence of ECC has been associated with higher resource utilization during the operative period; however, the impact beyond hospital discharge is minimally understood. Still, an increasing number of ECCs coupled with the complexity of the index surgical procedure.46 This association suggests that a proactive approach to determine which children might benefit most from a cardiac complex care program that partners families and all providers may reduce excessive resource utilization, although data on this is lacking. The needs of infants with CHD and any established significant morbidities increase exponentially requiring long-term multidisciplinary input. Furthermore, in-hospital teaching along with parental engagement before discharge can affect discharge success profoundly and improve transition to outpatient care and the medical home.
Recommendations:
Risk stratification and coordination among providers may be useful while arranging follow-up for infants with CHD and established comorbidities (Class IIb, LEO C-EO)
Clear and consistent communication between heart center providers and PCPs is essential in the outpatient management of infants with CHD (Class IIb, LEO C-EO)
Discussion of signs and symptoms with parents can be effective, and it is reasonable to discuss potential home treatments (eg, knee-chest position) (Class IIa, LEO C-EO)
Remote Monitoring Programs
Remote monitoring programs (RMP) such as those used in interstage home monitoring have become important for centers that manage high-risk infants with CHD. Born out of collaboration across centers, RMPs initially demonstrated success by significantly lowering the at-home, interstage, mortality rate for infants with single ventricle disease to < 5%.4,6,47–49 Through work led by the National Pediatric Cardiology Quality Improvement Collaborative, these programs have been expanded at many centers to include all infants with CHD that require more care-coordination and more extensive family engagement.50
The makeup of RMPs are very center-specific and frequently are led by an advanced practice provider (physician, nurse practitioner, or physician assistant) along with nurse coordinators, dietitians, speech therapists, data managers, and other allied health professionals. Data management and communication is frequently written and using a telephone as the primary source of communication. Yet, digital platforms with electronic medical record integration are becoming much more common and seem to demonstrate better outcomes, security, and expandability.51–54 Although the exact structure and frequency of data collection is perhaps not the driver of success, it is evident that open communication and better parent and PCP engagement plays a significant role.55
Recommendations:
The involvement of the PCP in coordination with the cardiologist as part of a robust RMP is essential in the management of infants with at least moderate-risk CHD (lass 1, LEO B-NR)
For infants with single ventricle physiology, it is recommended that the interstage period (time between Stage 1 and Stage 2 palliative procedures) be discussed and that comprehensive interstage home monitoring be available for these families (Class I, LEO B-NR)
Immunization and Palivizumab
Infants with CHD, both repaired and nonrepaired, are prone to infection from various types of pathogens. Coupled to the fact that many undergo heart surgery within the first year of life, there is the possibility of leaving a vulnerable group of infants undervaccinated during their first year. Significant center variation exists in how programs alter the recommended vaccination schedule56 for infants undergoing surgery.57 There does not seem to be a significant effect from cardiopulmonary bypass on antibody titers, suggesting that deviation away from the recommended schedule may be unnecessary before surgery.58 Evidence on optimal immunization timing immediately after cardiopulmonary bypass in infants who just underwent surgery is lacking; yet, normal vaccination of infants even with single ventricle CHD appears to be safe regardless of timing.59 Furthermore, use of the 23-valent polysaccharide pneumococcal vaccine guidelines in patients with heterotaxy syndromes has been extrapolated from guidelines for other groups of patients with questionable splenic function.60
Infections with respiratory syncytial virus (RSV) are prominent in this population, and risk for decompensation from infections remains high.26,61 Many infants with CHD will qualify for palivizumab administration, and various societal recommendations for the use of this antibody preparation are continually updated and should be followed.62 It is also essential to recognize that infants with simple left-to-right shunt lesions who demonstrate congestive heart failure do qualify for palivizumab administration. Furthermore, cardiopulmonary bypass does reduce these antibody levels, and thus repeat administration before discharge is recommended in infants receiving the series.63
Recommendations:
Delay or significant modification of the standard vaccination schedule is not recommended for infants with CHD (Class I, LOE B-NR)
For patients with questionable splenic function, administration of the 23-valent polysaccharide pneumococcal vaccine can be beneficial (Class IIa, LOE B-NR)
Neonates with potential immunocompromised states may benefit from consultation with immunology before vaccine administration (Class IIb, LOE B-EO)
Palivizumab immunoprophylaxis for RSV can be effective in lowering the risk of complications associated with infection in neonates with significant CHD (Class IIa, LOE B-NR)
Emergency Department Visits and Readmissions
Despite careful coordination, robust RMPs, and well-developed action plans involving all providers, emergency department (ED) visits and readmissions to tertiary care facilities remain a prevalent issue. Furthermore, as medicine has evolved and become more metric-based through a myriad of quality improvement initiatives, rates of ED visits and readmissions continue to be increasingly scrutinized in the hope of improving outcomes, quality of care, patient satisfaction and resource utilization.
Although CHD remains the most common congenital anomaly, there is minimal data available related to the frequency of visits to the ED related to their heart disease. Patients with CHD account for < 1% of all pediatric ED visits. Yet, they require inpatient admission from the ED tenfold more frequently, at a rate of 46%, compared to infants without CHD.8,64 Among shunt-dependent neonates, nearly 80% will require an ED visit within the first 6 months, with close to one-half requiring inpatient admission.65 Furthermore, ED physicians may feel uncomfortable managing infants with complex CHD, demonstrating a clear opportunity to provide education that would likely improve care.65,66
Overall readmission, from any location after an index operation, remains high in this population at 8.7%.67 This rate goes up significantly when infants are shunt-dependent, rising to between 10% and 73%.68–74 This wide range suggests significant center and regional variation that can be improved through standardization. The most common factors associated with readmission are Hispanic ethnicity, a longer postoperative length of stay, younger age, government insurance, single ventricle physiology, and high social risk factors, including regional poverty level, type of primary medical insurance, maternal age, primary parental language, and distance to outpatient clinics.21,67,69,70,72,73,75,76
Recommendations:
Evaluating the presence of certain social determinants of health, longer length of hospital stay, younger age, government insurance, and single ventricle physiology at the time of neonatal discharge can be a beneficial component of the outpatient plan, especially in coordination with the local care team (Class IIa, LOE B-NR)
Respiratory distress, sometimes manifesting as recurrent wheezing or “bronchiolitis,” may need to be urgently evaluated in any child with a CHD (Class IIa, LOE C-LD)
Preventing ED visits and readmissions can also have a direct societal financial benefit. Although penalties for readmissions may not be as robust in this population compared to adults with heart disease, there is some evidence that cost can be nearly doubled in patients who were readmitted to the hospital within 30 days of their initial discharge.76 This finding remains significant even after adjusting for case complexity and type of initial operation.
Unfortunately, the rates of ED utilization and readmissions are significant for those with CHD, especially the most vulnerable of the population. Many risk factors have been elucidated, and there has been considerable work done to alter modifiable factors and improve care coordination. Still, there is more to be done to close gaps in health disparities that may impact this patient population.
Conclusions and Future Directions
Despite significant advances surrounding inpatient care of infants with CHD, notable morbidity and potentially mortality exist after discharge. Coupling high-complexity medical conditions with a need for multiple types of care providers, assistance with required technology and underlying patient-specific and society-specific risk factors are challenging hurdles to completely overcome. Furthermore, short-term outcomes as infants have a direct effect on long-term success for children surviving with CHD.
It is for these reasons that careful coordination at the time of discharge, identification of modifiable and nonmodifiable patient-specific variables, open communication with all providers involved in the infant’s care, and ongoing monitoring of outcomes remains essential in the care of these types of patients. Parents also need to be well-versed in the knowledge of their child’s CHD and potential complications that can ensue. Focusing just on inpatient and surgical success is only a part of the lifelong care these neonates are facing.
In the future, equal consideration for outpatient management and transitions to the medical home will need to hold as much weight as short-term inpatient survival. Having a keen understanding of the variables surrounding development, feeding and nutrition, rehabilitation services, weaning from technology dependence, and lessening potential socioeconomic disparities will go a long way to advancing the care of infants with CHD.
Drs Vergales, Figueroa, Frommelt, Putschoegl, Singh, and Murray conceptualized the article, performed extensive literature review, and contributed to the writing of the manuscript; Drs Allen and Villafane along with Mr Wood conceptualized the article, developed the content outline, oversaw the intellectual content, and were the principal reviewers of the manuscript; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.
FUNDING: The Neonatal Heart Society (NHS) contributed an educational grant to the project, NeoC3. The NHS, on a regular basis, applies and receives several unrestrictive educational grants for several internal projects from the following organizations and companies: Abbott Formula, Mead Johnson, Cheisi, Mallinckrodt, Prolacta, and Medtronic. The grants received from industry partners were used solely to offset the cost of publishing this supplement in Pediatrics. The industry supporters did not suggest manuscript content, nor did they participate in any way to the writing or editing of the manuscript.
CONFLICT OF INTEREST DISCLOSURES: The authors have no conflicts of interest relevant to this article to disclose.
The guidelines/recommendations in this article are not American Academy of Pediatrics policy, and publication herein does not imply endorsement.
Comments