Hypoplastic left heart syndrome: Management and outcome

INTRODUCTION — Hypoplastic left heart syndrome (HLHS) is characterized by a diminutive left ventricle and small left-sided structures incapable of supporting the systemic circulation (figure 1). If left untreated, HLHS is universally fatal. Surgical and medical interventions have improved outcomes, but mortality and morbidity remain high.

The management and outcome of HLHS will be reviewed here. The anatomy, physiology, clinical features, and diagnosis of HLHS and management of patients following the Fontan procedure are discussed separately. (See "Hypoplastic left heart syndrome: Anatomy, clinical features, and diagnosis" and "Management of complications in patients with Fontan circulation".)

CHOICE OF MANAGEMENT — Once the diagnosis of HLHS has been confirmed, treatment options should be clearly explained to parents/caregivers to allow them to make informed decisions in line with their values and preferences. HLHS remains one of the most challenging congenital heart defects to manage because of its high mortality and morbidity. However, innovations and improvements in medical and surgical care have improved outcomes considerably, and, increasingly, families are choosing intervention over comfort care measures [1,2].

Most pediatric cardiologists and surgeons in North America consider staged palliative surgery the preferred management approach for HLHS rather than primary cardiac transplantation or "comfort measures only" (CMO) [3]. Within the pediatric cardiology community, there remains a debate as to whether CMO should be offered as an option to parents/caregivers [4]. Some experts feel that CMO should not be offered as an option in light of the improving outcome of the staged surgical repair, while others argue parents/caregivers should be given the option of CMO because of the high burdens associated with surgical palliation, including multiple cardiac operations, unknown long-term morbidity, and considerable mortality [2,5]. Primary cardiac transplant for HLHS is generally not offered as an option given long waiting times for donor organs.

At the author's institution, the management choice is dependent on parent/caregiver preference after ensuring that parents/caregivers understand the different management options.

●Prenatal counseling – During prenatal counseling, discussion with the parents includes the natural history of HLHS and prenatal management decisions including termination of the pregnancy, pursuit of additional prenatal testing, choice of delivery setting, and, where clinically applicable, possibility of fetal intervention. The approach to prenatal counseling may differ in settings where options for pregnancy termination are limited.

Postnatal management choices are also discussed at prenatal visits, including staged palliative surgery, cardiac transplantation, and CMO. Longitudinal follow-up after prenatal diagnosis of HLHS suggests that approximately 40 percent of pregnancies end in termination or fetal demise [6].

●Postnatal counseling – Postnatal counseling includes discussion of staged surgical palliation, including a careful presentation of outcomes. Most parents/caregivers choose staged surgical palliation. Some families choose CMO, particularly if there are other comorbidities such as extreme prematurity, life-limiting genetic conditions, or other major malformations [5].

FETAL INTERVENTION — With the ability to identify prenatal anatomic and physiologic characteristics predictive of the development of HLHS, fetal intervention has become a potential therapeutic management option at select centers with expertise in fetal transcatheter intervention [6]. In highly selected cases of critical aortic valve stenosis and preserved left ventricular volume in the second trimester, prenatal intervention to perform aortic valvuloplasty can prevent the progression to neonatal HLHS [7-9]. This ultrasound-guided transcatheter procedure poses significant risks to the fetus and some risk to the mother as well. These risks have been felt to be justifiable given the uniform lethality of the untreated disease and the significant mortality associated with postnatal management [9,10].

INITIAL STABILIZATION — Initial management is focused on ensuring sufficient mixing of oxygenated and deoxygenated blood and optimizing ventricular function:

●Once the diagnosis of HLHS is made or suspected, continuous intravenous prostaglandin E1 (alprostadil 0.01 to 0.05 mcg/kg per minute) is initiated to maintain a patent ductus arteriosus. The patent ductus arteriosus provides vital flow from the right ventricle (RV) to the systemic circulation. (See "Diagnosis and initial management of cyanotic heart disease (CHD) in the newborn", section on 'Prostaglandin E1'.)

●Patients with a restrictive or intact atrial septum may require transcatheter balloon atrial septostomy or surgical atrial septoplasty to create a large-enough opening to decompress the left atrium and provide sufficient pulmonary venous return across the atrial septum [11]. Newborns with a restrictive or intact atrial septum, particularly infants with mitral atresia, can have profound cyanosis and may require emergency balloon atrial septostomy.

●Infants with ventricular dysfunction may require additional preoperative medical therapy, including diuretics, inotropic agents, and, possibly, mechanical ventilation. (See "Heart failure in children: Management".)

SURGICAL MANAGEMENT — Staged palliative repair is the preferred surgical approach over primary cardiac transplantation because of the scarcity of infant donors and the improving short-term success of the staged palliative approach [3].

Surgical palliation of HLHS consists of three staged procedures, typically performed at the following ages:

●Stage I procedure (eg, Norwood procedure) is performed in neonates during the first week of life (see 'Stage I procedures' below)

●Stage II procedure (bidirectional Glenn procedure) is typically performed at four to six months of age (see 'Stage II: Cavopulmonary shunt' below)

●Stage III procedure (Fontan procedure) is typically performed between two and five years of age (see 'Stage III: Fontan procedure' below)

These interventions are referred to as palliative procedures because they do not restore the patient to a normal biventricular circulation.

Stage I procedures — The stage I procedure is performed in neonates during the first week of life. The fundamental objectives of initial surgical management with the stage I procedure are to:

●Provide unobstructed blood flow from the right ventricle (RV) to the systemic circulation and the coronary arteries

●Establish a controlled source of pulmonary blood flow

●Ensure an unobstructed pathway between the pulmonary venous return and the systemic RV

The stage I procedure has three major modifications: the classic stage I procedure with a modified Blalock-Thomas-Taussig shunt (also commonly called the modified Blalock-Taussig shunt [mBTS]), the stage I procedure with an RV to pulmonary artery (RV-PA) conduit, and the stage I hybrid procedure (figure 2A).

Stage I with mBTS (classic Norwood) — The classic stage I procedure (referred to as a classic Norwood procedure) consists of the following (figure 2A):

●Creation of a neo-aorta by dividing the main PA, closing the defect in the distal vessel, and using the proximal main PA and homograft material to reconstruct the aortic arch, which incorporates the native ascending aorta.

●Establishment of a source of pulmonary blood flow by placing a right mBTS, typically using a 3.5 mm Gore-Tex tube, between the innominate artery and the proximal right PA.

●Resection of the atrial septum.

Stage I with RV-PA conduit (Sano modification) — Given the difficulty in managing the balance of pulmonary and systemic circulations in the early postoperative Norwood stage I patient, alternative palliations have been explored. Of these, using an RV-PA conduit to supply pulmonary blood flow (Sano modification) has become an acceptable alternative [12].

In the Sano modification, an RV-PA conduit (typically a 5 to 6 mm expanded polytetrafluoroethylene tube) serves as the source of pulmonary blood flow instead of the mBTS [13]. One disadvantage to the traditional Sano procedure is the need for a right ventriculotomy, with potential for negative effects on long-term function of the systemic RV. Modifications of the original Sano approach have included a limited RV incision without ventriculotomy and using a reinforced RV-PA conduit [14]. In some centers, a composite Sano conduit is created by attaching a systemic vein homograft to the distal end of the Sano to prevent conduit regurgitation and reduce volume overload [15].

Stage I hybrid procedure — The "hybrid" approach uses a combination of surgical and transcatheter approaches to accomplish the goals of a stage I palliative repair without neonatal cardiopulmonary bypass and circulatory arrest, thus avoiding the circulatory stressors imposed by the postbypass inflammatory state, myocardial reperfusion effects, and recovery from extensive dissection through a median sternotomy [16,17]. Avoiding cardiopulmonary bypass and circulatory arrest is a particular advantage in patients with underlying risk factors (eg, prematurity, very low birth weight, or severe shock or organ dysfunction) who may be deemed poor candidates for a more traditional stage I approach [18].

The stage I hybrid procedure entails the following:

●Pulmonary blood flow is controlled by surgically placing external constricting bands on the branch PAs.

●Systemic outflow is secured by using catheter techniques to place an intravascular stent in the ductus arteriosus, thus allowing the RV to continue to support the systemic circulation without surgically anastomosing the main PA to the aorta.

●The atrial septum is treated by catheter techniques with balloon atrial septostomy or placement of a stent in the interatrial septum.

The optimal use of the hybrid procedure has not been defined, and there is considerable variability among centers in patient selection, procedural details, and subsequent management. The hybrid procedure is performed in the minority of cases, and patients selected for this procedure frequently have underlying risk factors. A report from the Society of Thoracic Surgeons Congenital Heart Surgery Database included data on 1728 patients with HLHS managed at 100 centers in North America from 2010 through 2012 [18]. The hybrid procedure was the initial palliation in only 13 percent of patients, of whom 52 percent had at least one preoperative risk factor (including prematurity; low birth weight; chromosomal abnormalities; and organ dysfunctions such as shock, necrotizing enterocolitis, and renal dysfunction) compared with 37 percent of patients who underwent the traditional stage I procedure.

Comparison of approaches — The choice between the different stage I surgical approaches is based on the experience and preference of the surgeon, the institutional familiarity with the procedure, and certain patient factors (eg, anatomy and comorbidities such as prematurity or other noncardiac diseases). The available evidence suggests that clinical outcomes may depend more on patient and center factors (ie, comorbidities, surgical volume) than on the surgical technique itself.

●RV-PA conduit versus mBTS – The stage I procedure with mBTS (classic Norwood procedure) and the stage I procedure with RV-PA conduit (Sano modification) both result in a functionally univentricular circulation maintained by the RV. In both procedures, the goal is to achieve a distribution of blood flow to the pulmonary and systemic circulation that provides adequate oxygen delivery to the systemic organs and tissues without excessive pressure or volume load on the RV, as well as avoiding excessive systemic cyanosis. The main disadvantage of the mBTS is that blood flow through the shunt into the pulmonary circulation results in diastolic diversion of flow from the systemic circulation including, most importantly, the coronary circulation. The main disadvantage of the RV-PA approach is the need for a right ventriculotomy, with potential for negative effects on long-term function of the systemic RV.

Although most single-institution case series and a multicenter prospective study suggested improved early survival with the RV-PA procedure compared with the mBTS procedure [19-22], a large multicenter clinical trial comparing the two techniques demonstrated that by six years after the stage I procedure, transplant-free survival was similar in both groups [23,24].

In the Single Ventricle Reconstruction (SVR) trial, 555 infants with single-ventricle physiology cared for at 15 North American centers were randomized to undergo a stage I procedure using either RV-PA or mBTS [23-25]. Survival among children assigned to RV-PA was better before the second stage of palliation, but by six years after randomization, differences in transplant-free survival were no longer statistically significant (64 percent in the RV-PA group versus 59 percent in the mBTS group) [24].

Additional findings of the SVR trial included [23-25]:

•There was an interaction of shunt type with annual stage I volume, with the highest volume and most senior surgeons having better outcomes using the mBTS. These data support the importance of tailoring the choice of shunt to the experience level of the surgeon and also highlight the case-volume relationship with stage I procedure outcomes.

•Children in the RV-PA group required more catheter interventions before undergoing the Fontan procedure; however, catheterization rates were similar after the Fontan procedure.

•Ventricular ejection fraction, clinical events, other morbidities, and New York Heart Association functional class were similar among survivors in both groups.

•After completion of stage II palliation, patients who initially underwent stage I with an RV-PA conduit had larger RV volumes and reduced strain patterns on echocardiography and cardiac magnetic resonance compared with those who underwent mBTS, suggesting that long-term ventricular function may be better preserved in those receiving mBTS initially [26,27].

Participants in the SVR trial are undergoing continued surveillance to characterize their longer-term outcomes. It is possible that the full impact of stage I strategy on RV function or PA architecture may only become apparent later in life.

●Hybrid versus traditional approaches – The main advantage of the hybrid procedure is that it accomplishes the goals of the stage I palliative repair without neonatal cardiopulmonary bypass and circulatory arrest. An important disadvantage is that patients who undergo the stage I hybrid procedure require a more extensive stage II operation.

There are limited data comparing outcomes with the stage I hybrid and traditional stage I procedures. In a retrospective study from a single center of infants with single-ventricle physiology who underwent initial stage I repair using a hybrid approach (n = 32) or mBTS (ie, classic Norwood; n = 43) and who subsequently completed stage II palliation, transplant-free survival at three years was similar in both groups (86 percent in the hybrid group versus 80 percent in the mBTS group) [28]. Patients in the hybrid group had higher rates of reintervention for PA obstruction (31 versus 11 percent). A subsequent report from the same institution (which included many of the same patients) evaluated outcomes in 138 infants with HLHS who underwent initial stage I palliation with a hybrid approach (n = 54), mBTS (n = 73), or RV-PA conduit (n = 11) [29]. Stage II palliation was achieved in 67 percent of patients. Interstage transplant-free survival, ventricular dysfunction, and atrioventricular valve regurgitation were similar among all three palliation strategies. Patients with preserved ventricular function had a higher rate of transplant-free survival regardless of palliation strategy. In another observational report of 40 selected patients who underwent hybrid stage I, the survival rate to hospital discharge was 90 percent, with an average length of stay that was approximately one-half of that of traditional surgical approaches [16].

Post-stage I management

Medical management — Following stage I palliation, infants typically are discharged from the hospital three to four weeks postoperatively with oxygen saturation levels in the range of 75 to 80 percent. Medical management of these patients may include the following:

●Aspirin – We recommend aspirin to prevent shunt thrombosis in all infants who have undergone stage I palliation, regardless of the type of procedure performed. The usual dose is 1 to 5 mg/kg per dose given orally once daily. However, the optimal dose is uncertain. Retrospective studies suggest that standard weight-based aspirin dosing may not achieve sufficient platelet inhibition in postsurgical pediatric patients [30-32]. In addition, there is a growing experience that aspirin resistance can occur in infants. Among infants with infants with shunt-dependent congenital heart disease, the prevalence of aspirin resistance may be as high as 10 percent [33]. Some centers routinely assess for aspirin resistance using commercial assays. (See "Nonresponse and resistance to aspirin", section on 'Laboratory testing'.)

The use of aspirin following stage I palliation is supported by observational data suggesting that aspirin greatly reduces the risk of shunt thrombosis. In a study from the National Pediatric Cardiology Quality Improvement Collaborative (NPC-QIC) registry that included 932 patients who underwent a stage I procedure at one of 52 sites in North America from 2008 to 2013, nearly all patients (94 percent) were discharged home on aspirin and only three patients (0.2 percent) presented with interstage shunt thrombosis [34]. This rate is considerably lower than reported rates of shunt thrombosis in earlier studies published before use of aspirin in this setting was standard practice (some reporting thrombosis rates as high as 40 percent) [35].

●Digoxin – Many centers, including ours, use digoxin throughout the interstage period. This practice is based on observational data from the NPC-QIC registry and post-hoc analysis of the SVR trial dataset, in which digoxin use appeared to be associated with improved interstage survival [36-39]. In one report of 544 patients in the NPC-QIC registry (excluding patients with a history of arrhythmia or who were discharged on antiarrhythmic medications after their stage I operation), interstage mortality was considerably higher among infants not discharged home on digoxin compared with on digoxin (9.9 versus 1.7 percent, respectively; odds ratio 8.6; 95% CI 1.9-38.2) [36]. Similarly, a post hoc analysis of the SVR trial dataset showed interstage mortality was higher among patients not prescribed digoxin (12.3 versus 2.9 percent; adjusted hazard ratio 3.5; 95% CI 1.1-11.7) [37].

●Diuretics – Most patients are discharged on diuretics [38]. Diuretics are typically weaned in the outpatient setting if the clinical status is stable.

●Angiotensin-converting enzyme (ACE) inhibitors – ACE inhibitors are commonly prescribed for infants with HLHS who have ventricular dysfunction following stage I palliation [38]. Empiric use of ACE inhibitors in the absence of ventricular dysfunction or systemic hypertension should be cautioned, as there are potential nephrotoxic effects, and the incidence of renal disease increases as these patients grow older. (See "Heart failure in children: Management", section on 'ACE inhibitors'.)

Data supporting the efficacy of ACE inhibitors in this setting are limited. In a clinical trial of 230 infants (<45 days old) with single-ventricle physiology randomized to receive enalapril or placebo, transplant-free survival at 14 months was similar in both groups (87 versus 86 percent, respectively) [40]. Somatic growth, neurodevelopment, ventricular function, heart failure severity, and adverse events were also similar in both groups. The trial had several limitations, including a high drop-out rate (only 185 of the 230 randomized infants completed the study), high rate of drug discontinuation, inability to achieve target doses in many patients, and limited follow-up. In addition, most of the patients enrolled in the trial had normal ventricular function at baseline; thus, the results may not be applicable to patients with ventricular dysfunction. Because of these limitations and because there is abundant indirect evidence from clinical trials in adult patients with heart failure demonstrating that ACE inhibitors improve survival in that setting, many providers continue to use ACE inhibitors in patients with single-ventricle physiology who have ventricular dysfunction.

Risk assessment — Following the stage I hospitalization, survivors remain at risk for interstage mortality prior to further surgical intervention. Reported interstage mortality rates range from 5 to 15 percent [19,41-45]. Causes of death include arrhythmias, residual anatomic lesions (aortic arch obstruction, restriction at the level of the atrial septum), shunt stenosis or thrombosis, imbalance of pulmonary and systemic blood flow, volume overload of the single ventricle, and diastolic run-off with coronary ischemia.

A risk score dubbed "NEONATE" was developed to identify neonates at highest risk for interstage mortality and was validated in 2128 interstage infants from the NPC-QIC registry [44]. The score ranges from 0 to 61, with points assigned for the following variables:

●Norwood (stage I) type:

•RV-PA conduit (Sano) – 0 points

•mBTS (classic Norwood) – 3 points

•Hybrid – 6 points

●ECMO postoperatively – 6 points

●Opiates at discharge – 6 points

●Not on digoxin at discharge – 9 points

●Arch obstruction on predischarge echocardiogram – 6 points

●Tricuspid regurgitation, moderate or greater without oxygen requirement – 12 points

●Extra oxygen (ie, moderate or greater tricuspid regurgitation plus oxygen requirement at discharge) – 28 points

A score of <17 identified lower-risk patients in whom the risk of interstage death or transplantation was 5 to 6 percent. A score of ≥17 identified higher-risk patients in whom the risk of interstage death or transplantation was 15 to 20 percent.

Monitoring and caregiver education — Strategies aimed at reducing interstage mortality and morbidity between stages I and II include the following [45,46]:

●Providing appropriate caregiver preparation during the stage I hospitalization

●Providing caregivers with a red flag action plan

●Optimizing feeding and weight gain

●Performing home monitoring of oxygen saturation and weight gain

●Establishing collaboration between the family, primary care provider, cardiologist, and other team members

●Standardizing assessments and action plans at clinic visits

Patients who are discharged home after stage I palliation typically are managed by an intensive home monitoring program that includes daily measurements of weight and oxygen saturation. Home care providers are given parameters that should prompt notification of the medical team for further evaluation and changes in management. This practice is based upon data from a single-center study that showed reduced mortality following initiation of a home surveillance program [47]. A study from the NPC-QIC demonstrated that incorporating many of the elements listed above into routine practice was associated with a decline in average interstage mortality from 9.5 percent in 2010 to 5.3 percent after 2014 [45].

Stage II: Cavopulmonary shunt — The second stage of surgical palliation for HLHS is the superior cavopulmonary shunt, also known as the bidirectional Glenn operation. This stage is typically performed at age four to six months of age, when the patient begins to outgrow the original shunt and becomes progressively more cyanotic. At stage II, the original shunt is removed and the superior vena cava is anastomosed end-to-side to the (typically right) PA (figure 2B). Thus, systemic venous return from the superior vena cava enters the PAs directly.

Patients who had a stage I hybrid procedure require a more extensive stage II operation, also known as a "comprehensive stage II procedure." In these patients, the ductal stent is removed and the surgical aortopulmonary reconstruction is completed. The distal main PA is repaired, and the PA bands are removed, usually with branch PA repairs as well.

The optimal timing for undertaking stage II palliation is uncertain. Based on a post hoc analysis of the SVR trial dataset, an interval of three to six months following stage I appears to be associated with lowest risk of mortality [48]. Of 399 patients who survived to stage II, 349 (87 percent) survived to a three-year follow-up, and overall transplant-free survival was 68 percent. Transplant-free survival was highest among patients who underwent stage II at an interval of three to six months following stage I. Optimal timing of the stage II palliation did not differ by shunt type. Another analysis from the NPC-QIC dataset of 789 patients from 31 centers showed that centers who generally performed stage II operations later than 5.1 months of age had higher interstage mortality [49]. This suggests a benefit for earlier stage II procedures. However, infants who required a stage II operation before three months of age had increased mortality, likely secondary to other high risk clinical factors [50].

Estimated surgical mortality following a traditional bidirectional Glenn procedure is low (between 1 and 2 percent) since the repair resolves the issue of balancing pulmonary and systemic flow through a shunt [51,52]. Data from the NPC-QIC showed postoperative complications occurred in one-third of patients and included unplanned cardiac catheterization (9 percent), reoperation (6 percent), cardiac arrest (3 percent), new neurologic deficit including seizures (2 percent), and feeding/gastrointestinal difficulties resulting in interventions such as fundoplication and/or feeding tube placement (15 percent) [52]. In a multivariate analysis, risk factors for postoperative complications included failure to thrive, female gender, and presence of aortic atresia and mitral stenosis.

The post-Glenn circulation is characterized by a passive source of pulmonary blood flow, a volume-unloaded ventricle, and persistent cyanosis due to the obligate flow of inferior vena caval return to the RV, which supplies the systemic arterial output. Infants are typically hospitalized for one week postoperatively and have oxygen levels in the low 80s at discharge. Following stage II, there is a low mortality rate and most children are able to tolerate this level of systemic oxygenation. As children become more active with crawling, walking, and, subsequently, running, the degree of cyanosis becomes more severe, especially during periods of exercise and exertion. At this point, stage III (Fontan) palliation should be considered. In addition, a small subset of patients will have ventricular dysfunction, which is associated with increased mortality [53].

Stage III: Fontan procedure — The third stage, the Fontan procedure, is typically performed between the age of two to five years. The Fontan procedure creates a venous pathway (total cavopulmonary connection) that directs the inferior vena caval flow into the PAs, resulting in the entire systemic venous return flowing passively into the PAs (figure 2C). This creates a system with a single ventricle pumping blood into separate, in-series systemic and pulmonary circulations, thereby relieving cyanosis.

There have been many modifications to the Fontan procedure. The modern versions of the Fontan are the lateral tunnel Fontan and the extracardiac Fontan (figure 2C). The choice between these largely depends on center and surgeon preference. Although this remains controversial, there are no definitive data showing superiority of either modern approach.

A surgically created fenestration between the conduit and the atrium can be made at the time of the Fontan operation. There is considerable practice variation in the use of the fenestration. Some centers perform the fenestration as a routine; others have moved toward using the fenestration only in "high-risk" patients (eg, those with elevated pulmonary vascular resistance [PVR]). The optimal approach is uncertain. The fenestration allows for a small right-to-left shunt and serves as a "pop-off" when PVR increases. This "pop-off" is particularly advantageous in the immediate postoperative period in patients with ventricular dysfunction and/or high PVR. Because there is a small amount of right-to-left shunting, patients with a fenestrated Fontan typically have oxygen saturations in the low 80s initially, rising to the high 80s to low 90s with time. The degree of cyanosis depends on the amount of shunting. The fenestration may close spontaneously many months after surgery or it can also be closed using a transcatheter device in the cardiac catheterization laboratory if necessary. One of the long-term risks of keeping a fenestration patent is that a thrombus in the Fontan baffle may go across the fenestration, which can result in stroke.

The Fontan procedure has a low mortality rate (3 percent), similar to the stage II procedure [54]. Although this total cavopulmonary connection is sometimes presented as the final stage of palliation for HLHS, follow-up after the Fontan procedure has demonstrated that a significant proportion of patients may require additional intervention due to cyanosis, elevated central venous pressures, persistent anatomic obstructions, arrhythmia, valve dysfunction, or ventricular dysfunction. Failure of the Fontan circulation may present as rare complications, such as protein-losing enteropathy, which occurs in 6 percent of cases [54]. Similarly, plastic bronchitis is rare but may require takedown of the Fontan or heart transplant to relieve symptoms.

Patients with HLHS may have long-term complications following Fontan completion since it remains unknown how well the morphologic RV, which is responsible for sustaining both systemic and pulmonary flow, will perform into adulthood [55-57]. It is expected that the function of these ventricles will decline over time, based on observations of RV function in the systemic circulations of other cardiac disorders, such as following atrial switch operations for D-transposition of the great arteries. In a retrospective series of patients with HLHS who completed Fontan procedure, patients who underwent an mBTS or RV-PA conduit as stage I palliation had clinically significant ventricular dysfunction (17 and 31 percent, respectively) and atrioventricular valve regurgitation (40 and 16 percent, respectively) at a mean follow-up of 58 months following the Fontan procedure [56].

Complications following the Fontan procedure are discussed in greater detail in a separate topic review. (See "Management of complications in patients with Fontan circulation", section on 'Cardiovascular complications'.)

Heart Transplantation — Heart transplantation is an important option for patients with HLHS. The main benefit of heart transplant in this population is that it exchanges single-ventricle physiology for a biventricular circulation. However, heart transplantation carries with it the burdens and risks associated with immunosuppression and potential for organ rejection. Because of considerable supply versus demand mismatch for organs that often results in long recipient wait times, primary heart transplantation is generally not offered. Rather, heart transplantation is reserved for patients in whom the Fontan circulation is failing or has failed. Criteria for listing patients with HLHS for heart transplant are not standardized. Common indications for cardiac transplant in this population include:

●Inability to wean from mechanical support

●Intravenous inotrope dependence

●Clinically symptomatic ventricular dysfunction

●Other comorbidities unresponsive to medical or interventional therapy (eg, protein-losing enteropathy, plastic bronchitis, arrhythmias)

Approximately 3 to 5 percent of patients with HLHS require heart transplantation by age six years [58,59]. For patients who survive to Fontan completion, approximately 5 percent are listed for heart transplant by the time they reach adulthood, and the likelihood continues to rise thereafter [60].

Patients with a history of congenital heart disease, when compared with patients with cardiomyopathy, experience greater morbidity and mortality when undergoing heart transplantation, possibly related to immune, clinical, and anatomical risk factors [61,62]. In a report of 253 pediatric patients who had undergone prior Norwood procedure and were listed for heart transplantation between 1993 and 2013 at 35 centers participating in the Pediatric Heart Transplant Study, 188 patients (74 percent) received heart transplantation, 58 (23 percent) died before transplantation, and seven (3 percent) were delisted [62]. Most patients were listed for transplantation between the Glenn and Fontan stages (38 percent) or between the Norwood and Glenn stages (35 percent); 27 percent were listed post-Fontan. At the time of listing, 13 percent of patients were on mechanical circulatory support. Overall survival at 1, 5, and 10 years following transplantation was 82 percent, 68 percent, and 59 percent, respectively.

Cardiac transplantation in pediatric patients and in patients post-Fontan is discussed further separately. (See "Heart failure in children: Management", section on 'Heart transplantation' and "Overview of the management and prognosis of patients with Fontan circulation", section on 'Heart or multiorgan transplant'.)

OUTCOME

Mortality — The medical and surgical interventions discussed in the sections above have improved outcomes, but mortality and morbidity remain high. HLHS carries a higher risk of mortality compared with most other congenital heart disease (CHD) defects [63]. For infants who undergo stage I repair, estimated three- to six-year survival rates are approximately 60 to 70 percent [24,55,56,64-66]. Mortality during the stage I hospitalization and interstage period contribute substantially to overall mortality. For patients who survive to the age of 12 months, long-term survival is approximately 90 percent [64,66].

In a population-based study using vital statistics data over a 26-year period in the Atlanta area, the survival of infants with nonsyndromic HLHS was 0 percent for those born between 1979 and 1984 and increased to 43 percent for those born between 1999 and 2005 [66]. For infants who underwent surgery, survival was 58 percent through the first year. Among children who survived to the age of one year, 94 percent survived to age 18 years.

In a single-center review of 499 consecutive patients with single-ventricle physiology undergoing palliation from 1990 to 2008, survival rates for the subset of patients with HLHS (n = 109) at 1, 5, and 10 years were 75, 70, and 65 percent, respectively [65].

●Mortality by stage – Surgical mortality rates differ with the different stages of palliation. The highest risk of mortality occurs following stage I:

•Stage I – Surgical mortality associated with the stage I procedure is sufficiently high that this operation ranks in the highest risk category of the Risk Adjustment in Congenital Heart Surgery scoring system [67]. With increasing experience with this procedure, however, select tertiary centers have reported early survival rates that approach 90 percent [19,68].

In the previously described multicenter Single Ventricle Reconstruction (SVR) trial, overall survival at one year was approximately 70 percent [69]. Causes of death included cardiovascular disease (42 percent of deaths), unknown (24 percent), multisystem organ failure (7 percent), infection (7 percent), respiratory failure (7 percent), and surgical complications (5 percent). The median age of death was 1.6 months, and most deaths occurred during the stage I hospitalization.

•Stages II and III – As previously noted, the mortality rates associated with stages II and III are low (approximately 1 to 3 percent) [51,52,54,65,70]. In the SVR trial, of the 399 patients who underwent stage II, 87 percent survived to a three-year follow-up [48]. Survival was highest among patients who underwent stage II at an interval of three to six months following stage I.

Outcomes following Fontan in patients with HLHS are less favorable compared with other single-ventricle CHD defects. In a study from the Australia and New Zealand Fontan Registry, which included data on >1000 Fontan procedures, patients with HLHS were nearly three times more likely to experience early death or Fontan failure (defined as reoperation, transplantation, reoperation, or poor functional status) compared with other single-ventricle morphologies [71].

Survival from each stage to the next and overall survival for children with HLHS was described in a retrospective review of patients who underwent surgical palliation at a single center from 2002 to 2012 [64]. Of the 219 patients included in the study, 38 (17 percent) died during the stage I hospitalization and another 18 infants (8 percent) died during the interstage period. Of the 163 patients who underwent the stage II procedure, 22 (13 percent) subsequently died or required heart transplantation, 95 (58 percent) underwent the Fontan procedure, and 46 (28 percent) were alive awaiting Fontan palliation at last follow-up. The overall eight-year survival following the stage I operation was 66 percent.

Additional details on outcomes following Fontan palliation are provided separately. (See "Overview of the management and prognosis of patients with Fontan circulation", section on 'Prognosis after Fontan procedure'.)

●Risk factors – Risk factors for mortality, especially after the stage I operation, include [64,72-79]:

•Noncardiac congenital anomalies and/or genetic disorders, particularly chromosomal defects

•Prematurity or low birth weight

•Restrictive or intact atrial septum (even if adequately decompressed in the neonatal period)

•Certain anatomic variants (eg, mitral stenosis with aortic atresia, small ascending aorta)

•Significant right ventricular (RV) dysfunction and/or tricuspid regurgitation following stage I

•Longer duration of deep hypothermic circulatory arrest during stage I surgery

•Open sternum following stage I

•Need for extracorporeal membrane oxygenation

•Unplanned cardiac reoperation

•Living in a high-poverty neighborhood

Data from the Texas Birth Defects Registry showed that mortality increased as the distance from the birth center to cardiac surgical center increased [80]. This provides evidence for the utility of prenatal diagnosis in HLHS and may also support the prenatal transfer of affected fetuses to a tertiary center with the expertise to care for patients with HLHS. (See "Congenital heart disease: Prenatal screening, diagnosis, and management", section on 'Delivery planning'.)

Neurodevelopmental outcome — Patients with HLHS and other single-ventricle conditions who survive staged surgical palliation and/or heart transplantation are at risk for neurodevelopmental impairment (NDI) [81-85].

In the previously mentioned SVR trial, 321 of 373 patients who survived without cardiac transplantation underwent neurodevelopmental assessment at 14 months using the Bayley Scales of Infant Development-II [83]. The mean psychomotor developmental index (PDI) was 74±19 and mean mental developmental index (MDI) was 89±18, both significantly lower than normative means. Neither the PDI nor MDI scores were associated with the type of stage I palliative procedure. Multivariate analysis identified the following as independent risk factors for low PDI and/or MDI scores:

●Low birth weight

●Prolonged duration of postoperative mechanical ventilation

●Prolonged stage I hospitalization

●Greater number of complications postdischarge to age 12 months

●Presence of noncardiac anomalies or genetic syndrome

●Lower maternal education level

In a subsequent report from the SVR trial, children with single-ventricle lesions continued to have NDI at three years of age based upon lower scores on the Ages & Stages Questionnaire compared with normative values [86].

In a multicenter cross-sectional study, 47 school-age survivors of either Norwood palliation or infant heart transplant had neuropsychiatric testing at a mean age of 12.4±2.5 years. Mean full scale intelligence quotient (IQ) was 86±14, well below population normative values. Surgical strategy (Norwood versus transplantation) was not associated with any developmental test score in multivariable analysis [81].

These results demonstrate that patients with HLHS are at risk for NDI, but this appears to be independent of the type of surgical intervention. Patients who undergo fetal intervention appear to have a similar risk of NDI [87].

Thus, children with HLHS are at high risk for developmental delay and should be referred for developmental evaluation and early intervention or early childhood special education, as well as periodic reevaluation [88]. (See "Developmental-behavioral surveillance and screening in primary care".)

Although the etiology of neurodevelopmental disability in patients with HLHS is not fully understood, it is thought to be multifactorial and includes malnutrition as a risk factor. One study of 170 patients (60 percent of whom had HLHS) demonstrated that low height trajectory was associated with worse neurodevelopmental outcome at 14 months of age [89]. Clinical efforts toward ensuring delivery of adequate nutrition to these vulnerable infants are ongoing, and poor low-height trajectory in infancy may serve as an indication to monitor neurodevelopment more closely.

Other morbidities — With an expanding cohort of survivors of surgical palliation through Fontan completion, increasing information is being accumulated on long-term morbidities in these patients. Common issues complicating management of patients post-Fontan include the following, which are discussed in detail separately (see "Management of complications in patients with Fontan circulation" and "Overview of the management and prognosis of patients with Fontan circulation"):

●Arrhythmias (see "Management of complications in patients with Fontan circulation", section on 'Arrhythmias')

●Atrioventricular valve regurgitation (see "Management of complications in patients with Fontan circulation", section on 'Atrioventricular valve regurgitation')

●Heart failure and exercise intolerance (see "Management of complications in patients with Fontan circulation", section on 'Heart failure')

●Thromboembolic events (see "Management of complications in patients with Fontan circulation", section on 'Thrombosis')

●Protein-losing enteropathy (see "Management of complications in patients with Fontan circulation", section on 'Protein-losing enteropathy')

●Kidney disease (see "Management of complications in patients with Fontan circulation", section on 'Renal complications')

●Liver disease (see "Management of complications in patients with Fontan circulation", section on 'Liver disease')

●Lung disease (see "Management of complications in patients with Fontan circulation", section on 'Respiratory complications')

●Venous insufficiency (see "Management of complications in patients with Fontan circulation", section on 'Venous insufficiency')

●Growth restriction (see "Management of complications in patients with Fontan circulation", section on 'Somatic growth issues')

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Congenital heart disease in infants and children".)

SUMMARY AND RECOMMENDATIONS

●Prenatal and initial postnatal counseling – Fetuses and neonates diagnosed with hypoplastic left heart syndrome (HLHS) should be managed at centers with expertise in both medical and surgical treatment for HLHS. In these centers, prenatal counseling for parents/caregivers includes a review of the natural history of HLHS and discussion of prenatal management decisions (eg, terminating versus continuing the pregnancy, pursuing additional prenatal testing, choice of delivery setting, and, where clinically applicable, the possibility of fetal intervention). Postnatal counseling includes discussion of staged surgical palliation, including a careful presentation of outcomes. Most informed parents/caregivers choose staged surgical palliation; however, some choose "comfort measures only," particularly if there are other comorbidities such as extreme prematurity, life-limiting genetic conditions, or other major malformations. (See 'Choice of management' above.)

●Initial stabilization – Initial management focuses on providing sufficient mixing of oxygenated and deoxygenated blood and adequate systemic perfusion by maintaining patency of the ductus arteriosus with prostaglandin E1 (alprostadil) infusion. Patients with a restrictive or intact atrial septum may require transcatheter balloon atrial septostomy. (See 'Initial stabilization' above.)

●Surgical palliation – Surgical palliation of HLHS consists of three staged procedures (figure 2A-C) (see 'Surgical management' above):

•Stage I – Stage I palliation is performed during the neonatal period once the newborn is medically stable. It provides unobstructed blood from the right ventricle (RV) to the systemic circulation and unobstructed flow between the pulmonary venous return and the RV and also establishes a reliable source of pulmonary blood flow. Stage I procedures can be performed with a modified Blalock-Thomas-Taussig shunt (also referred to as the classic Norwood procedure), with an RV-to-pulmonary artery (PA) conduit (also referred to as the Sano modification), or with a combination of surgical and transcatheter techniques (referred to as the hybrid procedure) (figure 2A). (See 'Stage I procedures' above.)

•Interstage management – Following stage I palliation, infants typically are discharged from the hospital three to four weeks postoperatively, with oxygen saturation levels in the range of 75 to 80 percent. These infants remain at risk of interstage mortality prior to further surgical intervention. For all infants who have undergone stage I palliation, regardless of the type of procedure performed, we recommend aspirin therapy to prevent shunt thrombosis (Grade 1B). Other medications that may be prescribed, depending on the clinical circumstances, include digoxin, diuretics, and angiotensin-converting enzyme (ACE) inhibitors. (See 'Post-stage I management' above.)

•Stage II – Stage II palliation (the bidirectional Glenn operation) creates a superior cavopulmonary shunt (figure 2B). It is typically performed at four to six months of age. (See 'Stage II: Cavopulmonary shunt' above.)

•Stage III – Stage III palliation (the Fontan procedure) creates a venous conduit that directs blood flow from the inferior vena cava into the PA (figure 2C). It is typically performed between two and five years of age. (See 'Stage III: Fontan procedure' above.)

●Outcome – With modern medical and surgical interventions, HLHS has transformed from being a condition that was almost always fatal to having six-year survival rates approaching 60 to 70 percent. For children who survive to the age of 12 months, long-term survival is approximately 90 percent. However, there is considerable morbidity, including an increased risk of neurodevelopmental impairment, thrombotic complications, exercise intolerance, and other complications. (See 'Outcome' above and "Management of complications in patients with Fontan circulation" and "Overview of the management and prognosis of patients with Fontan circulation", section on 'Prognosis after Fontan procedure'.)

●Developmental surveillance and early intervention – Children with HLHS are at high risk for developmental delay and should be referred for developmental evaluation, early intervention, and, if indicated, early childhood special education. (See 'Neurodevelopmental outcome' above and "Developmental-behavioral surveillance and screening in primary care".)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Audrey Marshall, MD, who contributed to an earlier version of this topic review.