- Email: [email protected]
Hypoplastic left heart syndrome
CARDIOVASCULAR MEDICINE
Hypoplastic left heart syndrome
These staged operations are the definitive way of managing HLHS and often referred to as the ‘Norwood’ strategy after the surgeon who first described them. The heart can never be ‘normal’ as there is functionally only one developed ventricle, so these operations are often referred to as being ‘palliative’ in the sense that they do not correct the heart condition back to a normal (biventricular) circulation. The term ‘palliative’ can be misleading in this context and should not be confused with the commoner use of the word in the setting of palliative care. Primary cardiac transplantation is, in theory, an alternative treatment strategy but neonatal heart donors are essentially unknown in the UK and very rare around the world. Neonatal heart transplantation is therefore not a practical option in most countries, although a handful of centres have pursued this approach e accepting that many patients will die while awaiting a suitable organ. Given the success of the Norwood strategy, most units opt for this ‘staged palliation’ approach and only consider transplantation as a secondary option in children or adults who run into heart failure later in life. The ‘Norwood strategy’ is undertaken through three stages: (i) the Norwood operation on neonates; (ii) superior cavopulmonary anastomosis at 6e8 months of age; and (iii) total cavopulmonary connection between 18 months and 5 years of age (most commonly around 4 years). HLHS continues to be one of the highest risk lesions in children with congenital heart disease and, although surgical outcomes continue to improve, survival is currently around 65% at 5 years of age and 55% at 10 years of age. Despite these encouraging outcomes, many affected children may have developmental issues and a need for lifelong medical attention for ongoing health problems, both of which can place considerable strain on families. Counselling is essential for parents to make an informed decision and comfort care should be considered as an alternative pathway.
Shafi Mussa David J Barron
Abstract Hypoplastic left heart syndrome is a rare congenital heart defect in which the left-sided heart structures are underdeveloped, such that the left ventricle is unable to support the systemic circulation. It is almost always lethal without surgical treatment, and if left untreated would account for around 25e40% of neonatal cardiac deaths in the United Kingdom. Complex neonatal surgery to enable the right ventricle to support the circulation (the Norwood Procedure) has transformed management of the condition, but requires three staged procedures during early childhood. Management of these patients into adulthood will offer new challenges. Transplantation for this condition remains rare, and is reserved for those in which conventional treatment has failed. In this review article, we discuss epidemiology, genetics, morphology, pathophysiology, prenatal diagnosis, modes of clinical presentation, and management strategies in this challenging condition.
Keywords hypoplastic left heart syndrome; norwood; univentricular heart
Introduction Hypoplastic left heart syndrome (HLHS) is a complex congenital heart defect in which one or more of the left-sided cardiac structures are underdeveloped such that the heart is unable to support the systemic circulation. It is a fatal condition without intervention, which usually needs to be undertaken in the first few days after birth. Prior to the 1980s there was no treatment for the condition and it was viewed as being universally lethal until 1982. Surgical treatment was first undertaken in the UK in the early 1990s and now several congenital heart units in the UK have established programmes to treat affected children. Surgical management focusses on utilizing the right ventricle to support the systemic circulation. A series of staged operations then gradually redirect the systemic venous drainage directly into the lungs to create a circulation that is driven by a single (right) ventricle. This extraordinary series of procedures essentially creates a circulation in series rather than the normal situation of a circulation in parallel, driven by two ventricles.
Epidemiology HLHS is rare, occurring in about one in every 5000 live births, comprising approximately 2e3% of all children with congenital heart defects, and equating to approximately 150 affected children born in the UK per year. The condition is found worldwide with no particular ethnic or geographical preponderance, nor is there any described association with maternal age or parity. It is slightly more common in males. Despite the low incidence, the condition is of considerable significance because of its serious natural history e without treatment it would be responsible for 25e40% of all neonatal cardiac deaths, being fatal within a few weeks from birth in 95% of cases.
Pathophysiology Genetics As yet, no causative gene for HLHS has been found and most children with HLHS do not have chromosomal anomalies. There are, however, associations with Turner’s syndrome (monosomy X), Edwards syndrome (trisomy 18) and Patau syndrome (trisomy 13). The chromosomal abnormality with the most consistent incidence of HLHS is the terminal deletion of the long arm of chromosome 11 (Jacobsen syndrome), comprising between 5 and 10% of affected individuals. A candidate gene in this region, ETS-1, has been implicated in cardiac development in non-
Shafi Mussa MA MD FRCS (CTh) is a Registrar in Congenital Cardiac Surgery, at Birmingham Children’s Hospital NHS Foundation Trust, Birmingham, UK. Conflict of interest: none declared. David J. Barron MD FRCP FRCS (CTh) is a Consultant Congenital Cardiac Surgeon, at Birmingham Children’s Hospital NHS Foundation Trust, Birmingham, UK. Conflict of interest: none declared.
PAEDIATRICS AND CHILD HEALTH --:-
1
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
mammals and in mice. Other genes that have been associated with HLHS include GJA1 (chromosome 6), NKX2 on chromosome 5, and somatic mutations in the HAND1 gene, also on chromosome 5. A small, but well-recognised, risk of recurrence in future pregnancies exists. This is around 2e4%, although in families with two affected children the recurrence risk is considerably greater, estimated at 25%. This finding alone suggests that HLHS has a genetic component, although it is likely to be more complex and multifactorial in nature.
development and function. No single component can be considered in isolation as all may be involved in the condition. More than one and frequently all components are underdeveloped to varying degrees. Establishing the morphological subtype has prognostic significance, particularly in aortic atresia and mitral stenosis, in which there is inflow to but no outflow from a small, hypertensive left ventricular cavity. This subtype has been shown to have poorer outcomes than others, perhaps related to associated coronary anomalies. Other cardiac abnormalities can occur together with HLHS in about 7$5% of all cases. Examples are transposition of the great arteries, atrial isomerism, and total anomalous pulmonary venous drainage. These additional abnormalities may require additional surgical procedures and might affect outcome, but HLHS remains the dominant lesion and thus governs the management of these more complex patients.
Physiology In HLHS, the left side of the heart cannot support the systemic circulation. Survival is only possible where the systemic circulation is supported by the right ventricle via right-to-left flow through the ductus arteriosus (Figure 1). Without intervention the duct closes during the first few days after birth, exposing the left heart insufficiency, with subsequent failure of the systemic circulation. In addition, pulmonary venous return can only reach the systemic circulation by crossing the atrial septum (through a patent foramen ovale) to reach the right side of the heart. Thus the circulation relies on two shunts, with resultant obligate mixing of pulmonary venous and systemic venous return, creating a cyanotic condition. Without ductal flow the situation is incompatible with life: a duct-dependent systemic circulation.
Presentation and diagnosis The clinical presentation of HLHS varies depending on the flow through the obligate shunts. Often, babies are born in good condition by virtue of a patent ductus and the systemic circulation being supported by the right heart. At one end of the spectrum, if the patent foramen ovale and the ductus are widely open, then cyanosis may not be clinically obvious and abnormal findings on examination would suggest a large patent ductus (continuous murmur and wide pulse pressure) and little else. However, these patients have uncontrolled pulmonary blood flow and a large volume load on the circulation due to the shunts. As the pulmonary vascular resistance falls pulmonary blood flow increases and babies typically develop signs of congestive heart failure with increasing tachypnoea and hepatomegaly, and cardiomegaly and pulmonary congestion on the chest radiograph. This situation may progress into respiratory distress, increasing acidosis, and circulatory collapse. At the other extreme, babies may have significant pulmonary venous congestion and are cyanosed and tachypnoeic from birth secondary to a restrictive patent foramen ovale or even intact atrial septum. Severe restriction at the atrial septum leads to profound cyanosis, rapid decompensation, and may be incompatible with life. Survival depends on the degree of restriction and some patency of the mitral and aortic valves to allow at least some blood to leave the left heart. Within the spectrum, babies may have a moderate sized ductus and aortic coarctation with weak or absent femoral pulses and congestive heart failure because of the combination of high pulmonary blood flow and high systemic afterload. These signs may only become apparent a few days after birth as the ductus begins to close. The degree of cyanosis is variable and arterial saturations can be maintained in the 90s, with mixing of the circulation masked to some extent by high pulmonary blood flow. A plain chest radiograph can reveal cardiomegaly with pulmonary plethora and oedema, but is not diagnostic. An electrocardiogram (ECG) is generally non-specific, usually demonstrates a normal rhythm, and can show evidence of right ventricular hypertrophy with tall R-waves in the anterior chest leads. Echocardiography is the key to confirming the diagnosis and
Morphology The cardiac morphology of HLHS is heterogeneous. The condition cannot simply be called left ventricular hypoplasia as the left heart consists of components (i.e. the mitral valve, left ventricular cavity, left ventricular outflow tract, aortic valve, ascending aorta, and aortic arch) that are all inter-related in their
Figure 1 Hypoplastic left heart syndrome.
PAEDIATRICS AND CHILD HEALTH --:-
2
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
showing all the relevant morphological features to guide management. Further imaging is rarely indicated and cardiac catheterisation usually has no role. Some cases present after death at home. Historical studies have indicated that 15% of registered cases of HLHS were diagnosed at post-mortem examination, and that 78% of cases with HLHS were discharged from hospital before diagnosis. Such numbers are hopefully being reduced through the introduction of routine newborn oxygen saturation testing, together with better screening echocardiography.
intact atrial septum in utero, but this has been attempted in even fewer patients.
Postnatal management Medical care The main aim of initial management is to stabilise the baby to enable the diagnosis to be confirmed and a treatment plan formulated. The standard procedure is to ensure ductal patency with prostaglandin E2 infusion. Congestive cardiac failure is treated initially with diuretics but inotropic support (often dobutamine 5e10 mcg/kg/min) may be needed to support the volume loaded right ventricle. Intubation and ventilation alleviates the work of breathing and allows for haemodynamic stabilisation. Positive-pressure ventilation reduces pulmonary oedema, and permissive hypercapnia can raise pulmonary vascular resistance and so reduce pulmonary overcirculation, although this is rarely necessary. Nitrogen may be added to the ventilator circuit to reduce the fraction of inspired oxygen (FiO2) and further increase pulmonary vascular resistance, but is rarely used in current practice. These measures stabilise most patients, enabling surgery to be planned over the next few days. Patients with a restrictive or even intact atrial septum are more precarious, depending on the severity of the obstruction and resultant lung injury from pulmonary venous congestion. The most severe cases do not survive, but those who remain profoundly cyanosed despite ventilatory support need urgent decompression of the left atrium with either balloon atrial septostomy or early surgery. On confirmation of the diagnosis, all treatment options need to be clearly explained to families to enable them to make an informed decision. The option of no intervention (comfort care) should also be discussed. Most centres would agree that comfort care is a valid option although this has become a less frequent choice as the surgical outcomes have improved.
Prenatal diagnosis and management Currently in the UK, 35% of all congenital heart diseases are detected prenatally. HLHS is one easier lesions to identify and up to 75% of cases diagnosed prenatally. Typically ultrasound reveals a normally sited right ventricle with a hypoplastic left ventricle, often associated with hypoplasia of the ascending aorta. However, the four-chamber view might seem normal at 20 weeks and it is only at subsequent prenatal echocardiography (often in the third trimester) that subnormal left ventricular growth is identified. If HLHS is detected, a detailed scan to exclude extracardiac anomalies should be undertaken, and parents should be offered rapid fetal karyotyping. Prenatal diagnosis has two important roles. Firstly, it enables parents to be counselled in a timely and rational manner, provide an indication of prognosis and possible outcomes, and present the option of elective termination of pregnancy. Termination rates following a prenatal diagnosis of HLHS are decreasing. This is likely to be multifactorial but certainly indicates that counselling has changed with improving surgical outcomes. Secondly, prenatal diagnosis enables planning of postnatal management i.e. for the birth to be near or at a specialist cardiac surgical centre and with access to intensive care facilities for resuscitation and ongoing management of the newborn if required. However, only one small study has shown a survival advantage, although evidence exists that prenatal diagnosis improves pre-operative condition of patients and reduces co-morbidity. A recent systematic review has shown that prenatal diagnosis had no significant impact on either pre- or post-operative mortality, despite better haemodynamic stability of those neonates with a prenatal diagnosis. Fetal intervention for critical aortic stenosis in HLHS has been shown to be both feasible and moderately successful. The principle is to balloon dilate the aortic valve in utero to improve forward flow through the fetal left ventricle and encourage growth of the left heart structures, thereby avoiding the need for subsequent single ventricle palliation. Recently published voluntary registry data demonstrates that fetal cardiac intervention is being undertaken in at least 40 centres in 15 countries worldwide, but total numbers remain very small. Overall rates of fetal loss following intervention are at least 11%, with similar rates for fetal aortic valvuloplasty. Technical success was achieved in 81%, but only 53% surviving to hospital discharge, and 29% with a biventricular circulation at time of discharge. Consideration of fetal intervention outcomes alongside current outcomes with conventional surgical therapies therefore provokes serious ethical and clinical debate. Other fetal intervention in HLHS includes atrial septoplasty to alleviate restrictive or
PAEDIATRICS AND CHILD HEALTH --:-
Surgical care Stage 1 The Norwood procedure The aim of surgery is to modify the circulation such that the right ventricle ejects blood into the systemic circulation through an unobstructed outflow, while also securing adequate blood flow to the pulmonary circulation. To realise this: the atrial septum is removed to create a common atrial chamber which receives both the systemic and pulmonary venous return the main pulmonary artery is disconnected at its bifurcation and attached to a reconstructed aortic arch so that the right ventricle ejects directly into the systemic circulation. Any narrowing of the aortic arch is simultaneously repaired as part of the aortic arch reconstruction. a small synthetic tube (usually Goretex) is placed between the systemic and pulmonary circulations, either as a modified Blalock-Taussig shunt (classical procedure) or as a right ventricle to pulmonary artery conduit (modified procedure). This complex procedure was first described in 1980 by Bill Norwood, a surgeon at Children’s Hospital of Philadelphia, and has carried his name ever since. The surgery results in a
3
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
the neonatal myocardium. Mixed venous saturations are also closely monitored to optimize oxygen delivery, and some investigators have advocated the use of continuous in-line measurement of mixed venous saturations. The chest is commonly left open after these procedures to avoid any compressive effect of the closed chest on the oedematous myocardium. It is generally closed after 24e48 hours in the intensive care unit, following a period of maintained haemodynamic stability and resolution of tissue and myocardial oedema. The risk of shunt thrombosis is reduced by instituting a low dose heparin infusion in the intensive care unit and then substituting for low-dose aspirin (5e15 mg/kg) once oral feeding is established.
functionally univentricular circulation based on the right ventricle (Figure 2), in which the systemic and pulmonary circulations exist in parallel rather than in series (the normal situation). Ideally the same volume of blood will flow through both the systemic and pulmonary components with each cardiac cycle e a balanced circulation that aims to achieve adequate oxygen delivery without placing an unnecessary volume load on the circulation. Success of the procedure requires precise anaesthetic and cardiopulmonary bypass management, meticulous surgery, and careful post-operative care in the intensive care and ward settings. Initially only a few centres could emulate outcomes achieved in a handful of North American centres, but the procedure has gradually been adopted worldwide with steadily improving results. Early mortality reported in the early to mid-1990s was 30 e35% but has improved such that many large centres now report early survival (i.e. 30 day) of 85e90%. Nationally collected statistics from the UK National Institute of Cardiovascular Outcomes Research (NICOR) show an early survival of 89% for all cases (total 350) across the UK between 2012 and 2015. Postoperative management is one of the most challenging areas of neonatal cardiac intensive care hinging on balancing systemic and pulmonary circulations by aiming for systemic saturations of about 80%. High inspired oxygen or hypocarbia (low PaCO2) both lower pulmonary vascular resistance, volume loading the circulation, and lead to cardiac failure. Vasodilators such as milrinone are used to lower systemic vascular resistance, encouraging better systemic perfusion and reducing afterload on
RV-PA conduit and the Single Ventricle Reconstruction trial RV-PA conduit better initial approach in most, but not all cases of HLHS The Norwood procedure utilizes the main pulmonary artery as systemic outflow, and an alternative source of pulmonary blood flow is required. The classical Norwood procedure achieved this by means of a Blalock-Taussig (BT) shunt. A major concern is that flow through the shunt occurs throughout the cardiac cycle. This creates diastolic run-off from the aorta, which can result in retrograde diastolic flow in the aorta and risk of coronary steal. There have been incidences of sudden death in which no clear cause can be found, even at post-mortem examination, which have been attributed to low diastolic pressure and the coronary steal phenomenon. One of the most important surgical modifications to the Norwood procedure has been the use of a right ventricle-pulmonary artery conduit instead of the BT shunt as the source of pulmonary blood flow. A 5 or 6 mm diameter Goretex tube is connected between the right ventricle and the pulmonary arteries in the modified Norwood procedure, and a BT shunt is not required (Figure 3). Most flow occurs in systole and because there is no run-off during diastole, the diastolic blood pressure in maintained. However, the technique requires an incision to be made in the right ventricle and there are concerns that this might impair ventricular function of the single ventricle in the short or long term. The choice between the classical and modified techniques has been debated over the years. In order to answer this important question the Paediatric Heart Network investigators conducted the Single Ventricle Reconstruction (SVR) trial between 2005 and 2008. 555 neonates undergoing the Norwood procedure in 15 North American centres were prospectively randomised to either the classical Norwood (BT shunt) or modified Norwood (right ventricle to pulmonary artery (RV-PA) conduit). The trial itself represents a remarkable and outstanding achievement, especially in the context of paediatric cardiac surgery in which numbers of cases are often too small for meaningful statistical analysis. The data have yielded a huge amount of important information and continue to do so e there have been 30 major articles published on the primary and secondary outcomes of this trial to date. The primary endpoint was survival free of transplantation and the initial findings from the trial were significant survival advantage to the use of the RV-PA conduit over the BT shunt at 12 months (74% versus 64%). There was however more need for unplanned catheter-based intervention in the RV-PA conduit group (41% versus 26%), mainly for conduit stenoses. This
Figure 2 The ‘Classical’ Norwood Procedure. Pulmonary blood flow is secured through a Blalock-Taussing shunt between the innominate artery and the pulmonary artery.
PAEDIATRICS AND CHILD HEALTH --:-
4
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
residual aortic-arch obstruction, restrictive atrial septal defects, imbalance of pulmonary and systemic blood flow, diastolic runoff with coronary ischaemia, shunt stenosis or thrombosis, and chronic volume overload of the single ventricle. Babies in the interstage period are particularly vulnerable to shunt thrombosis if they become significantly dehydrated and great care needs to be taken in managing even ‘minor’ viral infections which result in poor feeding or diarrhoea. Home monitoring of oxygen saturations and daily weights, alongside parent education and strict criteria for notification to healthcare professionals, has achieved a sustained reduction in interstage mortality to 2% over a 10-year period. This additional vigilance is likely to have identified at-risk infants and prompted earlier investigation. The most common problems are residual coarctation (in 5e10% of patients after the Norwood procedure), tricuspid regurgitation, and left pulmonary artery stenosis or hypoplasia. Stage 2 Cavopulmonary shunt The second stage procedure is undertaken at 4e6 months of age when the pulmonary vascular resistance has fallen. A highpressure supply of blood to the lungs is no longer needed and passive flow of blood through the pulmonary vascular bed can suffice. The exact timing varies but most centres assess the circulation with cardiac catheterisation or MRI at 3e4 months of age and plan stage II surgery after this. Decisions are also guided by saturation levels that tend to fall as the child outgrows the shunt placed at the Norwood procedure. Surgery aims to remove the original shunt, disconnect the superior vena cava from the heart and join it directly into the pulmonary arteries. The venous return from the superior vena cava is therefore directed into the lungs, removing the volume load from the heart and improving the mechanical efficiency of the right ventricle. The azygous vein is ligated as part of the procedure to prevent venous run-off to the lower body and ensures that all superior vena cava flow is directed into the pulmonary circulation. This so-called cavopulmonary shunt or bidirectional Glenn shunt (Figure 4) is still a major procedure needing cardiopulmonary bypass, but is a lower risk procedure than stage I, with a survival of 96e99%. Completion of stage II is an important landmark with patients being generally less precarious after this surgery than before. The anastomosis grows with the patient and saturations are maintained around 80e85%. Most children remain well for several years and it is only with increased activity and progression from crawling to walking and running that the degree of desaturation becomes gradually more severe, particularly during exercise. There is little mortality between second and third stages, and the timing of the third (and final) stage procedure partly depends on institutional preference with some centres following a policy of routinely progressing to stage III surgery at a fixed age (sometimes as young as 18 months, but most commonly around the age of 4 years) but most are guided by deteriorating symptoms and degree of desaturation. Stage 3 Total cavopulmonary connection The final stage of palliation for HLHS is to redirect the remaining venous return (i.e. inferior vena caval blood) to the pulmonary vasculature. All systemic venous blood now runs passively into the lungs, and for the first time, the child is no longer cyanosed. The systemic and pulmonary circulations are
Figure 3 The ‘modified’ or ‘Sano’ Norwood Procedure utilising an RVPA Conduit to provide pulmonary blood flow.
represented unequivocal class I evidence for the use of the modified Norwood strategy with the use of an RV-PA conduit. In order to predict outcomes following Norwood surgery considerable efforts have been made to find risk factors. Patients with co-existing genetic syndromes e.g. CHARGE association or Turner’s syndrome pose a greater risk, as do patients with smaller diameter of the ascending aorta, those with obstruction to pulmonary venous return (i.e. restrictive or intact atrial septum), those of lower gestational age, and patients with lower socioeconomic scores. Term babies with aortic atresia (51% of the cohort) had significantly lower 12-month mortality with a RV-PA conduit (about one third that of similar patients with a BT shunt), fitting with the physiological explanation of superior diastolic coronary flow and lack of coronary steal in RV-PA conduit patients. Conversely, in preterm babies with a patent aortic valve (4% of the cohort), survival following a classical Norwood was significantly better e perhaps a ventriculotomy was less well tolerated in preterm babies. Term babies with aortic stenosis showed similar survival regardless of surgical strategy. The interstage period This refers to the few months following the hospital discharge after the Norwood procedure up to the second stage of palliation for HLHS. Careful post-operative monitoring and treatment are vital elements of any successful management strategy for this complex condition. Mortality during this period is high, with 2 e20% of those discharged from hospital dying before the second stage procedure. Potential causes of interstage mortality include
PAEDIATRICS AND CHILD HEALTH --:-
5
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
Figure 5 The Total Cavo-Pulmonary Shunt (TCPC) or Stage III Procedure utilising ans extra-cardiac conduit to divert the IVC flow into the Pulmonary arteries. Also called the ‘Fontan circulation’.
Figure 4 The cavo-pulmnary shunt, or stage II procedure.
Fontan circulation and the common atrium that allows some blood return directly to the systemic circulation and therefore bypasses the lungs; this improves systemic blood flow at the expense of a small degree of desaturation and is used by most centres. Overall outcome following the Fontan procedure depends mostly on factors that are independent of the underlying cardiac diagnosis i.e. preoperative pulmonary artery pressures and ventricular function. Data from the Australia and New Zealand Fontan registry has shown that 13-year survival following extracardiac TCPC is 97%, although the proportion of patients with HLHS in this group was only 15%. However, HLHS patients were almost four times more likely to develop failure of the Fontan circulation than those patients with more favourable morphologies.
separated from each other and, in contrast to normal circulation in which a pumping chamber separates the pulmonary and systemic circulations, the two circulations are now directly in series. The single ventricle must provide enough energy to the circulating blood to drive it through both systemic and pulmonary vascular beds before it returns to the heart. Francis Fontan (a surgeon in Bordeaux) originally conceived this innovative series circulation over 40 years ago as a palliative procedure for tricuspid atresia and other less complex forms of functionally univentricular circulation. Following various technical modifications over the years this so-called Fontan circulation is now referred to as the Total Cavo-Pulmonary Connection (TCPC) but remains the final common pathway for a whole range of congenital heart defects characterised by a functionally single ventricle circulation. The procedure is undertaken on cardiopulmonary bypass, tunneling the inferior vena cava to the underside of the right pulmonary artery (Figure 5). This is achieved either with a baffle sewn within the right atrium to direct the inferior vena cava blood along a lateral tunnel within the atrium, or by disconnecting the inferior vena cava from the right atrium and placing a large-bore Goretex tube between the inferior vena cava and the pulmonary arteries e the ‘Extracardiac TCPC’. Outcomes from Stage 3 surgery In-hospital mortality is generally 3e4%. Outcomes have been improved by creating a small fenestration (hole) between the
PAEDIATRICS AND CHILD HEALTH --:-
Hybrid approaches An innovative alternative to the stage I Norwood procedure has been developed in an effort to avoid the potentially damaging effects of cardiopulmonary bypass in these fragile neonates, alongside combining interventional cardiac catheterisation techniques with surgery e the so-called ‘hybrid’ approach. The procedure aims to replicate the physiological state of the Norwood procedure by placing pulmonary artery bands on the branch pulmonary arteries to limit pulmonary blood flow whilst placing a stent (a bare metal transcatheter stent similar to those used in coronary artery disease) in the arterial duct to ensure patency, thus securing the systemic blood flow (an alternative to
6
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
paediatric intensive care. The Norwood procedure is also the subject of the largest multi-centre prospective randomised trial in paediatric cardiac surgery. The trial has produced an immense amount of useful data, and all of those involved should be applauded for their important contribution to the literature. The Norwood procedure was first conceived as a “one-sizefits-all” operation for all variants of HLHS. Sufficient expertise and increasing amounts of data now inform practice to the extent that centres may able to tailor their approach to individual patients, depending on anatomical features or physiological state, such that some babies may be better served with a RV-PA conduit, some with a BT shunt, and some with a hybrid management strategy. Technical refinements such as the use of ring-reinforced Goretex tubes for the RV-PA conduit have reduced the need for catheter-based intervention to augment pulmonary blood flow in the interstage period, and have resulted in improvements in pulmonary artery growth, although have not improved transplant-free survival through to 12 months. This evolution of surgical technique is absolutely imperative in terms of achieving the best possible outcomes. This must be borne in mind when interpreting the outcomes from the literature that may have used out-moded or inferior techniques and management strategies. Transplantation may be an option for some patients who develop heart failure in later life, although organ availability and suitable donors are the major limiting factors. Currently, the staged Norwood-type procedures remain the prevalent surgical option and provide a good quality of life for the majority of patients, the oldest of whom are now reaching adulthood. Worldwide experience suggests less than 5% of HLHS patients have undergone heart transplantation, but this is partly because some patients who develop heart failure are not candidates for transplantation due to multi-organ failure or high antigen load precluding donor matching. The best outcomes can only be achieved for these children and their families if the centres responsible for their management have ongoing, regular exposure to an appropriate number and complexity of cases to develop and maintain expertise, not only with surgery, but all aspects of care from the prenatal period through to adulthood. Data, mainly from North American centres, have demonstrated better outcomes in those centres with greater numbers of cases than those centres managing relatively few patients with HLHS, although a recent systematic review of the data was inconclusive. There remains an argument to concentrate expertise within fewer, larger, specialist centres. Those centres which are prepared to offer surgery to the highest risk patients i.e. those with restrictive or intact atrial septum or co-existing genetic syndromes, will extend the boundaries of what may be achieved. Finally, the efforts being made to improve outcomes have been rewarded with increasing numbers of patients reaching adolescence and adulthood with a single ventricle circulation. This success has produced a population of patients with a unique and challenging set of problems, which also need to be addressed and overcome. A
Figure 6 The Hybrid Norwood Procedure. Placement of bilateral branch pulmonary artery bands and a stent in the arterial duct. A balloon atrial septostomy or even stent placement is also performed to ensure good inter-atrial communication.
stenting the arterial duct is to maintain patients on a long term infusion of prostaglandin E2). Finally, atrial mixing is ensured with a balloon septostomy with or without a stent in the atrial septum to maintain patency (Figure 6). The surgeon and interventional cardiologist work together in the operating theatre, undertaking the procedure through a standard sternotomy but without cardiopulmonary bypass, using image intensification to deploy the stent(s) and surgical application of bilateral pulmonary artery bands. It may be undertaken sequentially in the operating theatre and then in the angiography suite, but hybrid operating theatres containing imaging equipment, where surgery and intervention can be performed simultaneously, are becoming more commonplace. Preliminary results were encouraging with early mortality comparable with conventional approaches (about 15e20%). However, as the outcomes from the conventional Norwood procedure have improved, only enthusiasts pursue the hybrid approach as a primary strategy for the management of all HLHS. The hybrid approach has, in the best hands, outcomes that are comparable to those of conventional surgery. As much of the mortality and morbidity from the conventional approach occurs in the neonatal period, it is intuitive to utilise the hybrid approach in higher risk patients, the preferred strategy in many centres. While this may lower initial mortality, there is an imposed burden during the interstage period and at comprehensive stage II surgery, which needs to be addressed before the hybrid approach genuinely improves outcomes.
The future Surgery for hypoplastic left heart syndrome has transformed the management of this otherwise uniformly fatal condition, and in doing so has led to improvements in other aspects of paediatric cardiac surgery, interventional paediatric cardiology and
PAEDIATRICS AND CHILD HEALTH --:-
7
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
Yerebakan C, Valeske K, Elmontaser H, et al. Hybrid therapy for hypoplastic left heart syndrome: Myth, alternative, or standard? J Thorac Cardiovasc Surg 2016 Apr; 151: 1112e21. 1123.e1e5.
FURTHER READING GENETICS Benson DW, Martin LJ, Lo CW. Genetics of hypoplastic left heart syndrome. J Pediatr 2016 Jun; 173: 25e31.
TRANSPLANTATION Chinnock RE, Bailey LL. Heart transplantation for congenital heart disease in the first year of life. Curr Cardiol Rev 2011 May; 7: 72e84.
PRENATAL DIAGNOSIS Atz AM, Travison TG, Williams IA, et al. Prenatal diagnosis and risk factors for preoperative death in neonates with single right ventricle and systemic outflow obstruction: screening data from the Pediatric Heart Network Single Ventricle Reconstruction Trial (*). J Thorac Cardiovasc Surg 2010 Dec; 140: 1245e50. Thakur V, Munk N, Mertens L, Nield LE. Does prenatal diagnosis of hypoplastic left heart syndrome make a difference? e a systematic review. Prenat Diagn 2016 Aug; 36: 854e63.
VOLUME/OUTCOME RELATIONSHIP Hirsch JC, Gurney JG, Donohue JE, Gebremariam A, Bove EL, Ohye RG. Hospital mortality for Norwood and arterial switch operations as a function of institutional volume. Pediatr Cardiol 2008; 29: 713e7. Welke KF, O’Brien SM, Peterson ED, et al. The complex relationship between pediatric cardiac surgical case volumes and mortality rates in a national clinical database. J Thorac Cardiovasc Surg 2009 May; 137: 1133e40 [Welke JTCVS 2009].
FETAL INTERVENTION Moon-Grady AJ, Morris SA, Belfort M, et al. International fetal cardiac intervention registry: a worldwide collaborative description and preliminary outcomes. J Am Coll Cardiol 2015 Jul 28; 66: 388e99. SURGERY e ORIGINAL PAPERS Fontan F, Baudet E. Surgical repair of tricuspid atresia. Thorax 1971; 26: 240e8. Norwood WI, Lang P, Hansen DD. Physiologic repair of aortic atresiahypoplastic left heart syndrome. N Engl J Med 1983; 308: 23e6. Sano S, Ishino K, Kawada M, et al. Right ventricle-pulmonary artery shunt in first stage palliation of hypoplastic left heart syndrome. J Thorac Cardiovasc Surg 2003; 126: 504e9.
Clinical practice points C
C
UK OUTCOME DATA National Congenital Heart Disease Audit Report 2012e15 http://www. ucl.ac.uk/nicor/audits/congenital/documents/annual-reports/ NCHDA2012-15 (accessed Aug 2016). C
SVR TRIAL Ohye RG, Sleeper LA, Mahony L, et al. Comparison of shunt types in the Norwood procedure for single-ventricle lesions. N Engl J Med 2010 May 27; 362: 1980e92. C
HYBRID APPROACH Galantowicz M, Cheatham JP, Phillips A, et al. Hybrid approach for hypoplastic left heart syndrome: intermediate results after the learning curve. Ann Thorac Surg 2008; 85: 2063e70.
PAEDIATRICS AND CHILD HEALTH --:-
8
Hypoplastic left heart syndrome (HLHS) is a rare condition, occurring in 1 per 5000 live births, and found in 2e3% of all children with congenital heart defects. It is almost always fatal without treatment. The cardiac morphology is heterogeneous but one or more left sided heart structures are too small, and the left heart cannot support the systemic circulation. The condition is relatively straightforward to identify during prenatal scanning, and up to 75% of all cases are diagnosed prenatally. The Norwood procedure is the first of three surgical procedures along the univentricular palliation pathway for HLHS, usually undertaken in the first week of life. This is followed by a superior cavopulmonary anastomosis at 6e8 months of age, and finally total cavopulmonary connection at 18 months to 5 years of age. Babies are particularly vulnerable in the period between Stage I of the Norwood procedure and prior to superior cavopulmonary anastomosis. Paediatricians should very carefully assess the fluid status of these babies if they present with apparently minor illness.
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
Hypoplastic left heart syndrome
These staged operations are the definitive way of managing HLHS and often referred to as the ‘Norwood’ strategy after the surgeon who first described them. The heart can never be ‘normal’ as there is functionally only one developed ventricle, so these operations are often referred to as being ‘palliative’ in the sense that they do not correct the heart condition back to a normal (biventricular) circulation. The term ‘palliative’ can be misleading in this context and should not be confused with the commoner use of the word in the setting of palliative care. Primary cardiac transplantation is, in theory, an alternative treatment strategy but neonatal heart donors are essentially unknown in the UK and very rare around the world. Neonatal heart transplantation is therefore not a practical option in most countries, although a handful of centres have pursued this approach e accepting that many patients will die while awaiting a suitable organ. Given the success of the Norwood strategy, most units opt for this ‘staged palliation’ approach and only consider transplantation as a secondary option in children or adults who run into heart failure later in life. The ‘Norwood strategy’ is undertaken through three stages: (i) the Norwood operation on neonates; (ii) superior cavopulmonary anastomosis at 6e8 months of age; and (iii) total cavopulmonary connection between 18 months and 5 years of age (most commonly around 4 years). HLHS continues to be one of the highest risk lesions in children with congenital heart disease and, although surgical outcomes continue to improve, survival is currently around 65% at 5 years of age and 55% at 10 years of age. Despite these encouraging outcomes, many affected children may have developmental issues and a need for lifelong medical attention for ongoing health problems, both of which can place considerable strain on families. Counselling is essential for parents to make an informed decision and comfort care should be considered as an alternative pathway.
Shafi Mussa David J Barron
Abstract Hypoplastic left heart syndrome is a rare congenital heart defect in which the left-sided heart structures are underdeveloped, such that the left ventricle is unable to support the systemic circulation. It is almost always lethal without surgical treatment, and if left untreated would account for around 25e40% of neonatal cardiac deaths in the United Kingdom. Complex neonatal surgery to enable the right ventricle to support the circulation (the Norwood Procedure) has transformed management of the condition, but requires three staged procedures during early childhood. Management of these patients into adulthood will offer new challenges. Transplantation for this condition remains rare, and is reserved for those in which conventional treatment has failed. In this review article, we discuss epidemiology, genetics, morphology, pathophysiology, prenatal diagnosis, modes of clinical presentation, and management strategies in this challenging condition.
Keywords hypoplastic left heart syndrome; norwood; univentricular heart
Introduction Hypoplastic left heart syndrome (HLHS) is a complex congenital heart defect in which one or more of the left-sided cardiac structures are underdeveloped such that the heart is unable to support the systemic circulation. It is a fatal condition without intervention, which usually needs to be undertaken in the first few days after birth. Prior to the 1980s there was no treatment for the condition and it was viewed as being universally lethal until 1982. Surgical treatment was first undertaken in the UK in the early 1990s and now several congenital heart units in the UK have established programmes to treat affected children. Surgical management focusses on utilizing the right ventricle to support the systemic circulation. A series of staged operations then gradually redirect the systemic venous drainage directly into the lungs to create a circulation that is driven by a single (right) ventricle. This extraordinary series of procedures essentially creates a circulation in series rather than the normal situation of a circulation in parallel, driven by two ventricles.
Epidemiology HLHS is rare, occurring in about one in every 5000 live births, comprising approximately 2e3% of all children with congenital heart defects, and equating to approximately 150 affected children born in the UK per year. The condition is found worldwide with no particular ethnic or geographical preponderance, nor is there any described association with maternal age or parity. It is slightly more common in males. Despite the low incidence, the condition is of considerable significance because of its serious natural history e without treatment it would be responsible for 25e40% of all neonatal cardiac deaths, being fatal within a few weeks from birth in 95% of cases.
Pathophysiology Genetics As yet, no causative gene for HLHS has been found and most children with HLHS do not have chromosomal anomalies. There are, however, associations with Turner’s syndrome (monosomy X), Edwards syndrome (trisomy 18) and Patau syndrome (trisomy 13). The chromosomal abnormality with the most consistent incidence of HLHS is the terminal deletion of the long arm of chromosome 11 (Jacobsen syndrome), comprising between 5 and 10% of affected individuals. A candidate gene in this region, ETS-1, has been implicated in cardiac development in non-
Shafi Mussa MA MD FRCS (CTh) is a Registrar in Congenital Cardiac Surgery, at Birmingham Children’s Hospital NHS Foundation Trust, Birmingham, UK. Conflict of interest: none declared. David J. Barron MD FRCP FRCS (CTh) is a Consultant Congenital Cardiac Surgeon, at Birmingham Children’s Hospital NHS Foundation Trust, Birmingham, UK. Conflict of interest: none declared.
PAEDIATRICS AND CHILD HEALTH --:-
1
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
mammals and in mice. Other genes that have been associated with HLHS include GJA1 (chromosome 6), NKX2 on chromosome 5, and somatic mutations in the HAND1 gene, also on chromosome 5. A small, but well-recognised, risk of recurrence in future pregnancies exists. This is around 2e4%, although in families with two affected children the recurrence risk is considerably greater, estimated at 25%. This finding alone suggests that HLHS has a genetic component, although it is likely to be more complex and multifactorial in nature.
development and function. No single component can be considered in isolation as all may be involved in the condition. More than one and frequently all components are underdeveloped to varying degrees. Establishing the morphological subtype has prognostic significance, particularly in aortic atresia and mitral stenosis, in which there is inflow to but no outflow from a small, hypertensive left ventricular cavity. This subtype has been shown to have poorer outcomes than others, perhaps related to associated coronary anomalies. Other cardiac abnormalities can occur together with HLHS in about 7$5% of all cases. Examples are transposition of the great arteries, atrial isomerism, and total anomalous pulmonary venous drainage. These additional abnormalities may require additional surgical procedures and might affect outcome, but HLHS remains the dominant lesion and thus governs the management of these more complex patients.
Physiology In HLHS, the left side of the heart cannot support the systemic circulation. Survival is only possible where the systemic circulation is supported by the right ventricle via right-to-left flow through the ductus arteriosus (Figure 1). Without intervention the duct closes during the first few days after birth, exposing the left heart insufficiency, with subsequent failure of the systemic circulation. In addition, pulmonary venous return can only reach the systemic circulation by crossing the atrial septum (through a patent foramen ovale) to reach the right side of the heart. Thus the circulation relies on two shunts, with resultant obligate mixing of pulmonary venous and systemic venous return, creating a cyanotic condition. Without ductal flow the situation is incompatible with life: a duct-dependent systemic circulation.
Presentation and diagnosis The clinical presentation of HLHS varies depending on the flow through the obligate shunts. Often, babies are born in good condition by virtue of a patent ductus and the systemic circulation being supported by the right heart. At one end of the spectrum, if the patent foramen ovale and the ductus are widely open, then cyanosis may not be clinically obvious and abnormal findings on examination would suggest a large patent ductus (continuous murmur and wide pulse pressure) and little else. However, these patients have uncontrolled pulmonary blood flow and a large volume load on the circulation due to the shunts. As the pulmonary vascular resistance falls pulmonary blood flow increases and babies typically develop signs of congestive heart failure with increasing tachypnoea and hepatomegaly, and cardiomegaly and pulmonary congestion on the chest radiograph. This situation may progress into respiratory distress, increasing acidosis, and circulatory collapse. At the other extreme, babies may have significant pulmonary venous congestion and are cyanosed and tachypnoeic from birth secondary to a restrictive patent foramen ovale or even intact atrial septum. Severe restriction at the atrial septum leads to profound cyanosis, rapid decompensation, and may be incompatible with life. Survival depends on the degree of restriction and some patency of the mitral and aortic valves to allow at least some blood to leave the left heart. Within the spectrum, babies may have a moderate sized ductus and aortic coarctation with weak or absent femoral pulses and congestive heart failure because of the combination of high pulmonary blood flow and high systemic afterload. These signs may only become apparent a few days after birth as the ductus begins to close. The degree of cyanosis is variable and arterial saturations can be maintained in the 90s, with mixing of the circulation masked to some extent by high pulmonary blood flow. A plain chest radiograph can reveal cardiomegaly with pulmonary plethora and oedema, but is not diagnostic. An electrocardiogram (ECG) is generally non-specific, usually demonstrates a normal rhythm, and can show evidence of right ventricular hypertrophy with tall R-waves in the anterior chest leads. Echocardiography is the key to confirming the diagnosis and
Morphology The cardiac morphology of HLHS is heterogeneous. The condition cannot simply be called left ventricular hypoplasia as the left heart consists of components (i.e. the mitral valve, left ventricular cavity, left ventricular outflow tract, aortic valve, ascending aorta, and aortic arch) that are all inter-related in their
Figure 1 Hypoplastic left heart syndrome.
PAEDIATRICS AND CHILD HEALTH --:-
2
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
showing all the relevant morphological features to guide management. Further imaging is rarely indicated and cardiac catheterisation usually has no role. Some cases present after death at home. Historical studies have indicated that 15% of registered cases of HLHS were diagnosed at post-mortem examination, and that 78% of cases with HLHS were discharged from hospital before diagnosis. Such numbers are hopefully being reduced through the introduction of routine newborn oxygen saturation testing, together with better screening echocardiography.
intact atrial septum in utero, but this has been attempted in even fewer patients.
Postnatal management Medical care The main aim of initial management is to stabilise the baby to enable the diagnosis to be confirmed and a treatment plan formulated. The standard procedure is to ensure ductal patency with prostaglandin E2 infusion. Congestive cardiac failure is treated initially with diuretics but inotropic support (often dobutamine 5e10 mcg/kg/min) may be needed to support the volume loaded right ventricle. Intubation and ventilation alleviates the work of breathing and allows for haemodynamic stabilisation. Positive-pressure ventilation reduces pulmonary oedema, and permissive hypercapnia can raise pulmonary vascular resistance and so reduce pulmonary overcirculation, although this is rarely necessary. Nitrogen may be added to the ventilator circuit to reduce the fraction of inspired oxygen (FiO2) and further increase pulmonary vascular resistance, but is rarely used in current practice. These measures stabilise most patients, enabling surgery to be planned over the next few days. Patients with a restrictive or even intact atrial septum are more precarious, depending on the severity of the obstruction and resultant lung injury from pulmonary venous congestion. The most severe cases do not survive, but those who remain profoundly cyanosed despite ventilatory support need urgent decompression of the left atrium with either balloon atrial septostomy or early surgery. On confirmation of the diagnosis, all treatment options need to be clearly explained to families to enable them to make an informed decision. The option of no intervention (comfort care) should also be discussed. Most centres would agree that comfort care is a valid option although this has become a less frequent choice as the surgical outcomes have improved.
Prenatal diagnosis and management Currently in the UK, 35% of all congenital heart diseases are detected prenatally. HLHS is one easier lesions to identify and up to 75% of cases diagnosed prenatally. Typically ultrasound reveals a normally sited right ventricle with a hypoplastic left ventricle, often associated with hypoplasia of the ascending aorta. However, the four-chamber view might seem normal at 20 weeks and it is only at subsequent prenatal echocardiography (often in the third trimester) that subnormal left ventricular growth is identified. If HLHS is detected, a detailed scan to exclude extracardiac anomalies should be undertaken, and parents should be offered rapid fetal karyotyping. Prenatal diagnosis has two important roles. Firstly, it enables parents to be counselled in a timely and rational manner, provide an indication of prognosis and possible outcomes, and present the option of elective termination of pregnancy. Termination rates following a prenatal diagnosis of HLHS are decreasing. This is likely to be multifactorial but certainly indicates that counselling has changed with improving surgical outcomes. Secondly, prenatal diagnosis enables planning of postnatal management i.e. for the birth to be near or at a specialist cardiac surgical centre and with access to intensive care facilities for resuscitation and ongoing management of the newborn if required. However, only one small study has shown a survival advantage, although evidence exists that prenatal diagnosis improves pre-operative condition of patients and reduces co-morbidity. A recent systematic review has shown that prenatal diagnosis had no significant impact on either pre- or post-operative mortality, despite better haemodynamic stability of those neonates with a prenatal diagnosis. Fetal intervention for critical aortic stenosis in HLHS has been shown to be both feasible and moderately successful. The principle is to balloon dilate the aortic valve in utero to improve forward flow through the fetal left ventricle and encourage growth of the left heart structures, thereby avoiding the need for subsequent single ventricle palliation. Recently published voluntary registry data demonstrates that fetal cardiac intervention is being undertaken in at least 40 centres in 15 countries worldwide, but total numbers remain very small. Overall rates of fetal loss following intervention are at least 11%, with similar rates for fetal aortic valvuloplasty. Technical success was achieved in 81%, but only 53% surviving to hospital discharge, and 29% with a biventricular circulation at time of discharge. Consideration of fetal intervention outcomes alongside current outcomes with conventional surgical therapies therefore provokes serious ethical and clinical debate. Other fetal intervention in HLHS includes atrial septoplasty to alleviate restrictive or
PAEDIATRICS AND CHILD HEALTH --:-
Surgical care Stage 1 The Norwood procedure The aim of surgery is to modify the circulation such that the right ventricle ejects blood into the systemic circulation through an unobstructed outflow, while also securing adequate blood flow to the pulmonary circulation. To realise this: the atrial septum is removed to create a common atrial chamber which receives both the systemic and pulmonary venous return the main pulmonary artery is disconnected at its bifurcation and attached to a reconstructed aortic arch so that the right ventricle ejects directly into the systemic circulation. Any narrowing of the aortic arch is simultaneously repaired as part of the aortic arch reconstruction. a small synthetic tube (usually Goretex) is placed between the systemic and pulmonary circulations, either as a modified Blalock-Taussig shunt (classical procedure) or as a right ventricle to pulmonary artery conduit (modified procedure). This complex procedure was first described in 1980 by Bill Norwood, a surgeon at Children’s Hospital of Philadelphia, and has carried his name ever since. The surgery results in a
3
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
the neonatal myocardium. Mixed venous saturations are also closely monitored to optimize oxygen delivery, and some investigators have advocated the use of continuous in-line measurement of mixed venous saturations. The chest is commonly left open after these procedures to avoid any compressive effect of the closed chest on the oedematous myocardium. It is generally closed after 24e48 hours in the intensive care unit, following a period of maintained haemodynamic stability and resolution of tissue and myocardial oedema. The risk of shunt thrombosis is reduced by instituting a low dose heparin infusion in the intensive care unit and then substituting for low-dose aspirin (5e15 mg/kg) once oral feeding is established.
functionally univentricular circulation based on the right ventricle (Figure 2), in which the systemic and pulmonary circulations exist in parallel rather than in series (the normal situation). Ideally the same volume of blood will flow through both the systemic and pulmonary components with each cardiac cycle e a balanced circulation that aims to achieve adequate oxygen delivery without placing an unnecessary volume load on the circulation. Success of the procedure requires precise anaesthetic and cardiopulmonary bypass management, meticulous surgery, and careful post-operative care in the intensive care and ward settings. Initially only a few centres could emulate outcomes achieved in a handful of North American centres, but the procedure has gradually been adopted worldwide with steadily improving results. Early mortality reported in the early to mid-1990s was 30 e35% but has improved such that many large centres now report early survival (i.e. 30 day) of 85e90%. Nationally collected statistics from the UK National Institute of Cardiovascular Outcomes Research (NICOR) show an early survival of 89% for all cases (total 350) across the UK between 2012 and 2015. Postoperative management is one of the most challenging areas of neonatal cardiac intensive care hinging on balancing systemic and pulmonary circulations by aiming for systemic saturations of about 80%. High inspired oxygen or hypocarbia (low PaCO2) both lower pulmonary vascular resistance, volume loading the circulation, and lead to cardiac failure. Vasodilators such as milrinone are used to lower systemic vascular resistance, encouraging better systemic perfusion and reducing afterload on
RV-PA conduit and the Single Ventricle Reconstruction trial RV-PA conduit better initial approach in most, but not all cases of HLHS The Norwood procedure utilizes the main pulmonary artery as systemic outflow, and an alternative source of pulmonary blood flow is required. The classical Norwood procedure achieved this by means of a Blalock-Taussig (BT) shunt. A major concern is that flow through the shunt occurs throughout the cardiac cycle. This creates diastolic run-off from the aorta, which can result in retrograde diastolic flow in the aorta and risk of coronary steal. There have been incidences of sudden death in which no clear cause can be found, even at post-mortem examination, which have been attributed to low diastolic pressure and the coronary steal phenomenon. One of the most important surgical modifications to the Norwood procedure has been the use of a right ventricle-pulmonary artery conduit instead of the BT shunt as the source of pulmonary blood flow. A 5 or 6 mm diameter Goretex tube is connected between the right ventricle and the pulmonary arteries in the modified Norwood procedure, and a BT shunt is not required (Figure 3). Most flow occurs in systole and because there is no run-off during diastole, the diastolic blood pressure in maintained. However, the technique requires an incision to be made in the right ventricle and there are concerns that this might impair ventricular function of the single ventricle in the short or long term. The choice between the classical and modified techniques has been debated over the years. In order to answer this important question the Paediatric Heart Network investigators conducted the Single Ventricle Reconstruction (SVR) trial between 2005 and 2008. 555 neonates undergoing the Norwood procedure in 15 North American centres were prospectively randomised to either the classical Norwood (BT shunt) or modified Norwood (right ventricle to pulmonary artery (RV-PA) conduit). The trial itself represents a remarkable and outstanding achievement, especially in the context of paediatric cardiac surgery in which numbers of cases are often too small for meaningful statistical analysis. The data have yielded a huge amount of important information and continue to do so e there have been 30 major articles published on the primary and secondary outcomes of this trial to date. The primary endpoint was survival free of transplantation and the initial findings from the trial were significant survival advantage to the use of the RV-PA conduit over the BT shunt at 12 months (74% versus 64%). There was however more need for unplanned catheter-based intervention in the RV-PA conduit group (41% versus 26%), mainly for conduit stenoses. This
Figure 2 The ‘Classical’ Norwood Procedure. Pulmonary blood flow is secured through a Blalock-Taussing shunt between the innominate artery and the pulmonary artery.
PAEDIATRICS AND CHILD HEALTH --:-
4
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
residual aortic-arch obstruction, restrictive atrial septal defects, imbalance of pulmonary and systemic blood flow, diastolic runoff with coronary ischaemia, shunt stenosis or thrombosis, and chronic volume overload of the single ventricle. Babies in the interstage period are particularly vulnerable to shunt thrombosis if they become significantly dehydrated and great care needs to be taken in managing even ‘minor’ viral infections which result in poor feeding or diarrhoea. Home monitoring of oxygen saturations and daily weights, alongside parent education and strict criteria for notification to healthcare professionals, has achieved a sustained reduction in interstage mortality to 2% over a 10-year period. This additional vigilance is likely to have identified at-risk infants and prompted earlier investigation. The most common problems are residual coarctation (in 5e10% of patients after the Norwood procedure), tricuspid regurgitation, and left pulmonary artery stenosis or hypoplasia. Stage 2 Cavopulmonary shunt The second stage procedure is undertaken at 4e6 months of age when the pulmonary vascular resistance has fallen. A highpressure supply of blood to the lungs is no longer needed and passive flow of blood through the pulmonary vascular bed can suffice. The exact timing varies but most centres assess the circulation with cardiac catheterisation or MRI at 3e4 months of age and plan stage II surgery after this. Decisions are also guided by saturation levels that tend to fall as the child outgrows the shunt placed at the Norwood procedure. Surgery aims to remove the original shunt, disconnect the superior vena cava from the heart and join it directly into the pulmonary arteries. The venous return from the superior vena cava is therefore directed into the lungs, removing the volume load from the heart and improving the mechanical efficiency of the right ventricle. The azygous vein is ligated as part of the procedure to prevent venous run-off to the lower body and ensures that all superior vena cava flow is directed into the pulmonary circulation. This so-called cavopulmonary shunt or bidirectional Glenn shunt (Figure 4) is still a major procedure needing cardiopulmonary bypass, but is a lower risk procedure than stage I, with a survival of 96e99%. Completion of stage II is an important landmark with patients being generally less precarious after this surgery than before. The anastomosis grows with the patient and saturations are maintained around 80e85%. Most children remain well for several years and it is only with increased activity and progression from crawling to walking and running that the degree of desaturation becomes gradually more severe, particularly during exercise. There is little mortality between second and third stages, and the timing of the third (and final) stage procedure partly depends on institutional preference with some centres following a policy of routinely progressing to stage III surgery at a fixed age (sometimes as young as 18 months, but most commonly around the age of 4 years) but most are guided by deteriorating symptoms and degree of desaturation. Stage 3 Total cavopulmonary connection The final stage of palliation for HLHS is to redirect the remaining venous return (i.e. inferior vena caval blood) to the pulmonary vasculature. All systemic venous blood now runs passively into the lungs, and for the first time, the child is no longer cyanosed. The systemic and pulmonary circulations are
Figure 3 The ‘modified’ or ‘Sano’ Norwood Procedure utilising an RVPA Conduit to provide pulmonary blood flow.
represented unequivocal class I evidence for the use of the modified Norwood strategy with the use of an RV-PA conduit. In order to predict outcomes following Norwood surgery considerable efforts have been made to find risk factors. Patients with co-existing genetic syndromes e.g. CHARGE association or Turner’s syndrome pose a greater risk, as do patients with smaller diameter of the ascending aorta, those with obstruction to pulmonary venous return (i.e. restrictive or intact atrial septum), those of lower gestational age, and patients with lower socioeconomic scores. Term babies with aortic atresia (51% of the cohort) had significantly lower 12-month mortality with a RV-PA conduit (about one third that of similar patients with a BT shunt), fitting with the physiological explanation of superior diastolic coronary flow and lack of coronary steal in RV-PA conduit patients. Conversely, in preterm babies with a patent aortic valve (4% of the cohort), survival following a classical Norwood was significantly better e perhaps a ventriculotomy was less well tolerated in preterm babies. Term babies with aortic stenosis showed similar survival regardless of surgical strategy. The interstage period This refers to the few months following the hospital discharge after the Norwood procedure up to the second stage of palliation for HLHS. Careful post-operative monitoring and treatment are vital elements of any successful management strategy for this complex condition. Mortality during this period is high, with 2 e20% of those discharged from hospital dying before the second stage procedure. Potential causes of interstage mortality include
PAEDIATRICS AND CHILD HEALTH --:-
5
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
Figure 5 The Total Cavo-Pulmonary Shunt (TCPC) or Stage III Procedure utilising ans extra-cardiac conduit to divert the IVC flow into the Pulmonary arteries. Also called the ‘Fontan circulation’.
Figure 4 The cavo-pulmnary shunt, or stage II procedure.
Fontan circulation and the common atrium that allows some blood return directly to the systemic circulation and therefore bypasses the lungs; this improves systemic blood flow at the expense of a small degree of desaturation and is used by most centres. Overall outcome following the Fontan procedure depends mostly on factors that are independent of the underlying cardiac diagnosis i.e. preoperative pulmonary artery pressures and ventricular function. Data from the Australia and New Zealand Fontan registry has shown that 13-year survival following extracardiac TCPC is 97%, although the proportion of patients with HLHS in this group was only 15%. However, HLHS patients were almost four times more likely to develop failure of the Fontan circulation than those patients with more favourable morphologies.
separated from each other and, in contrast to normal circulation in which a pumping chamber separates the pulmonary and systemic circulations, the two circulations are now directly in series. The single ventricle must provide enough energy to the circulating blood to drive it through both systemic and pulmonary vascular beds before it returns to the heart. Francis Fontan (a surgeon in Bordeaux) originally conceived this innovative series circulation over 40 years ago as a palliative procedure for tricuspid atresia and other less complex forms of functionally univentricular circulation. Following various technical modifications over the years this so-called Fontan circulation is now referred to as the Total Cavo-Pulmonary Connection (TCPC) but remains the final common pathway for a whole range of congenital heart defects characterised by a functionally single ventricle circulation. The procedure is undertaken on cardiopulmonary bypass, tunneling the inferior vena cava to the underside of the right pulmonary artery (Figure 5). This is achieved either with a baffle sewn within the right atrium to direct the inferior vena cava blood along a lateral tunnel within the atrium, or by disconnecting the inferior vena cava from the right atrium and placing a large-bore Goretex tube between the inferior vena cava and the pulmonary arteries e the ‘Extracardiac TCPC’. Outcomes from Stage 3 surgery In-hospital mortality is generally 3e4%. Outcomes have been improved by creating a small fenestration (hole) between the
PAEDIATRICS AND CHILD HEALTH --:-
Hybrid approaches An innovative alternative to the stage I Norwood procedure has been developed in an effort to avoid the potentially damaging effects of cardiopulmonary bypass in these fragile neonates, alongside combining interventional cardiac catheterisation techniques with surgery e the so-called ‘hybrid’ approach. The procedure aims to replicate the physiological state of the Norwood procedure by placing pulmonary artery bands on the branch pulmonary arteries to limit pulmonary blood flow whilst placing a stent (a bare metal transcatheter stent similar to those used in coronary artery disease) in the arterial duct to ensure patency, thus securing the systemic blood flow (an alternative to
6
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
paediatric intensive care. The Norwood procedure is also the subject of the largest multi-centre prospective randomised trial in paediatric cardiac surgery. The trial has produced an immense amount of useful data, and all of those involved should be applauded for their important contribution to the literature. The Norwood procedure was first conceived as a “one-sizefits-all” operation for all variants of HLHS. Sufficient expertise and increasing amounts of data now inform practice to the extent that centres may able to tailor their approach to individual patients, depending on anatomical features or physiological state, such that some babies may be better served with a RV-PA conduit, some with a BT shunt, and some with a hybrid management strategy. Technical refinements such as the use of ring-reinforced Goretex tubes for the RV-PA conduit have reduced the need for catheter-based intervention to augment pulmonary blood flow in the interstage period, and have resulted in improvements in pulmonary artery growth, although have not improved transplant-free survival through to 12 months. This evolution of surgical technique is absolutely imperative in terms of achieving the best possible outcomes. This must be borne in mind when interpreting the outcomes from the literature that may have used out-moded or inferior techniques and management strategies. Transplantation may be an option for some patients who develop heart failure in later life, although organ availability and suitable donors are the major limiting factors. Currently, the staged Norwood-type procedures remain the prevalent surgical option and provide a good quality of life for the majority of patients, the oldest of whom are now reaching adulthood. Worldwide experience suggests less than 5% of HLHS patients have undergone heart transplantation, but this is partly because some patients who develop heart failure are not candidates for transplantation due to multi-organ failure or high antigen load precluding donor matching. The best outcomes can only be achieved for these children and their families if the centres responsible for their management have ongoing, regular exposure to an appropriate number and complexity of cases to develop and maintain expertise, not only with surgery, but all aspects of care from the prenatal period through to adulthood. Data, mainly from North American centres, have demonstrated better outcomes in those centres with greater numbers of cases than those centres managing relatively few patients with HLHS, although a recent systematic review of the data was inconclusive. There remains an argument to concentrate expertise within fewer, larger, specialist centres. Those centres which are prepared to offer surgery to the highest risk patients i.e. those with restrictive or intact atrial septum or co-existing genetic syndromes, will extend the boundaries of what may be achieved. Finally, the efforts being made to improve outcomes have been rewarded with increasing numbers of patients reaching adolescence and adulthood with a single ventricle circulation. This success has produced a population of patients with a unique and challenging set of problems, which also need to be addressed and overcome. A
Figure 6 The Hybrid Norwood Procedure. Placement of bilateral branch pulmonary artery bands and a stent in the arterial duct. A balloon atrial septostomy or even stent placement is also performed to ensure good inter-atrial communication.
stenting the arterial duct is to maintain patients on a long term infusion of prostaglandin E2). Finally, atrial mixing is ensured with a balloon septostomy with or without a stent in the atrial septum to maintain patency (Figure 6). The surgeon and interventional cardiologist work together in the operating theatre, undertaking the procedure through a standard sternotomy but without cardiopulmonary bypass, using image intensification to deploy the stent(s) and surgical application of bilateral pulmonary artery bands. It may be undertaken sequentially in the operating theatre and then in the angiography suite, but hybrid operating theatres containing imaging equipment, where surgery and intervention can be performed simultaneously, are becoming more commonplace. Preliminary results were encouraging with early mortality comparable with conventional approaches (about 15e20%). However, as the outcomes from the conventional Norwood procedure have improved, only enthusiasts pursue the hybrid approach as a primary strategy for the management of all HLHS. The hybrid approach has, in the best hands, outcomes that are comparable to those of conventional surgery. As much of the mortality and morbidity from the conventional approach occurs in the neonatal period, it is intuitive to utilise the hybrid approach in higher risk patients, the preferred strategy in many centres. While this may lower initial mortality, there is an imposed burden during the interstage period and at comprehensive stage II surgery, which needs to be addressed before the hybrid approach genuinely improves outcomes.
The future Surgery for hypoplastic left heart syndrome has transformed the management of this otherwise uniformly fatal condition, and in doing so has led to improvements in other aspects of paediatric cardiac surgery, interventional paediatric cardiology and
PAEDIATRICS AND CHILD HEALTH --:-
7
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002
CARDIOVASCULAR MEDICINE
Yerebakan C, Valeske K, Elmontaser H, et al. Hybrid therapy for hypoplastic left heart syndrome: Myth, alternative, or standard? J Thorac Cardiovasc Surg 2016 Apr; 151: 1112e21. 1123.e1e5.
FURTHER READING GENETICS Benson DW, Martin LJ, Lo CW. Genetics of hypoplastic left heart syndrome. J Pediatr 2016 Jun; 173: 25e31.
TRANSPLANTATION Chinnock RE, Bailey LL. Heart transplantation for congenital heart disease in the first year of life. Curr Cardiol Rev 2011 May; 7: 72e84.
PRENATAL DIAGNOSIS Atz AM, Travison TG, Williams IA, et al. Prenatal diagnosis and risk factors for preoperative death in neonates with single right ventricle and systemic outflow obstruction: screening data from the Pediatric Heart Network Single Ventricle Reconstruction Trial (*). J Thorac Cardiovasc Surg 2010 Dec; 140: 1245e50. Thakur V, Munk N, Mertens L, Nield LE. Does prenatal diagnosis of hypoplastic left heart syndrome make a difference? e a systematic review. Prenat Diagn 2016 Aug; 36: 854e63.
VOLUME/OUTCOME RELATIONSHIP Hirsch JC, Gurney JG, Donohue JE, Gebremariam A, Bove EL, Ohye RG. Hospital mortality for Norwood and arterial switch operations as a function of institutional volume. Pediatr Cardiol 2008; 29: 713e7. Welke KF, O’Brien SM, Peterson ED, et al. The complex relationship between pediatric cardiac surgical case volumes and mortality rates in a national clinical database. J Thorac Cardiovasc Surg 2009 May; 137: 1133e40 [Welke JTCVS 2009].
FETAL INTERVENTION Moon-Grady AJ, Morris SA, Belfort M, et al. International fetal cardiac intervention registry: a worldwide collaborative description and preliminary outcomes. J Am Coll Cardiol 2015 Jul 28; 66: 388e99. SURGERY e ORIGINAL PAPERS Fontan F, Baudet E. Surgical repair of tricuspid atresia. Thorax 1971; 26: 240e8. Norwood WI, Lang P, Hansen DD. Physiologic repair of aortic atresiahypoplastic left heart syndrome. N Engl J Med 1983; 308: 23e6. Sano S, Ishino K, Kawada M, et al. Right ventricle-pulmonary artery shunt in first stage palliation of hypoplastic left heart syndrome. J Thorac Cardiovasc Surg 2003; 126: 504e9.
Clinical practice points C
C
UK OUTCOME DATA National Congenital Heart Disease Audit Report 2012e15 http://www. ucl.ac.uk/nicor/audits/congenital/documents/annual-reports/ NCHDA2012-15 (accessed Aug 2016). C
SVR TRIAL Ohye RG, Sleeper LA, Mahony L, et al. Comparison of shunt types in the Norwood procedure for single-ventricle lesions. N Engl J Med 2010 May 27; 362: 1980e92. C
HYBRID APPROACH Galantowicz M, Cheatham JP, Phillips A, et al. Hybrid approach for hypoplastic left heart syndrome: intermediate results after the learning curve. Ann Thorac Surg 2008; 85: 2063e70.
PAEDIATRICS AND CHILD HEALTH --:-
8
Hypoplastic left heart syndrome (HLHS) is a rare condition, occurring in 1 per 5000 live births, and found in 2e3% of all children with congenital heart defects. It is almost always fatal without treatment. The cardiac morphology is heterogeneous but one or more left sided heart structures are too small, and the left heart cannot support the systemic circulation. The condition is relatively straightforward to identify during prenatal scanning, and up to 75% of all cases are diagnosed prenatally. The Norwood procedure is the first of three surgical procedures along the univentricular palliation pathway for HLHS, usually undertaken in the first week of life. This is followed by a superior cavopulmonary anastomosis at 6e8 months of age, and finally total cavopulmonary connection at 18 months to 5 years of age. Babies are particularly vulnerable in the period between Stage I of the Norwood procedure and prior to superior cavopulmonary anastomosis. Paediatricians should very carefully assess the fluid status of these babies if they present with apparently minor illness.
Crown Copyright Ó 2016 Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Mussa S, Barron DJ, Hypoplastic left heart syndrome, Paediatrics and Child Health (2016), http://dx.doi.org/ 10.1016/j.paed.2016.12.002