Statistics from Altmetric.com
Hypoplastic left heart syndrome (HLHS), morphologically subdivided in mitral and/or aortic atresia or stenosis, accounts for 2–3% of all congenital heart defects and is a fatal congenital heart defect if untreated.1 Early postnatal survival requires a patent ductus arteriosus (DA) to provide systemic blood flow from the RV to aorta and an atrial septal defect to allow mixing of oxygenated pulmonary venous and deoxygenated systemic venous blood flow. Neonates that present with HLHS are clinically stable as long as the parallel pulmonary and systemic circulations are balanced, although the postnatal decrease in pulmonary vascular resistance may result in clinical decompensation. In about 6% of HLHS patients, the atrial septum is intact at birth, and up to 22% have a severely restrictive atrial septum,2 which results in a critically ill neonate. In prenatally undetected HLHS, neonates are often admitted in cardiogenic shock because of severe obstruction of the DA.
Surgical options are based on a three-staged procedure or heart transplantation. The three-stage Norwood procedure includes: Stage I with separation of systemic and pulmonary arterial flow using either a Blalock–Taussig shunt or a right ventricular to pulmonary artery conduit; Stage II with systemic venous flow from the superior vena cava directed to the pulmonary artery (and the shunt or conduit taken down); Stage III with a complete Fontan procedure directing the inferior vena caval flow to the pulmonary artery. Newer Hybrid approaches use a combination of surgical and transcatheter approaches for correction of HLHS. Despite significant advances in the surgical and intensive care for newborns with HLHS, decision making remains challenging for parents and physicians with a choice of the classical surgical Norwood procedure; a Hybrid surgical-interventional approach; heart transplantation or compassionate therapy after intense and repetitive communication about the choices and prognosis.
Lloyd et al3 draw attention to the preoperative condition of HLHS and analysed the operative, interstage and 1-year mortalities for either a Norwood or Hybrid procedure at the Evelina Children's Hospital in London. Risk was adjusted for patient complexity using a modified comprehensive Aristotle Score. A Hybrid approach was performed in 27 of 138 newborns, all with a significant higher Aristotle score. Despite this risk difference, the overall unadjusted survival at 1 year was similar for the Norwood and Hybrid procedures at 58.6% and 51.9%, respectively. Overall, 44% of patients were able to complete all three stages of these complex repair procedures.
Similar results were reported in 2003 from the American Congenital Heart Surgery Society; in 710 neonates undergoing a classical Norwood procedure, only 28% survived to the final Fontan completion.4 In 2010, the Pediatric Heart network5 reported a prospective randomised comparison of the conventional Norwood (with a modified Blalock–Taussig shunt) to the Sano modification (with a right ventricular to pulmonary artery conduit) in 555 preselected patents from a total cohort of 921 patients. One-year transplant-free survival was 64% for the conventional Norwood versus 74% for the Sano cohort with non-fatal serious adverse events in an additional 46% and 37%, respectively. In this context, the paper by Lloyd et al is unsparing in pointing out the current, still unsatisfactory, surgical results for newborns with HLHS, in particular, in those with high-risk scores. Based on these findings, the author recommend using the Hybrid procedure as a bridge to an eventual Norwood or Sano palliation, converting an infant from a high-risk category to moderate or low-risk group. However, given that this means the surgical options for HLHS patients will be based on a four-step (instead of three-step) procedure; the author's conclusion perhaps might be additionally influenced by the growing body of literature that the brain of newborns with HLHS is underdeveloped, thereby being more vulnerable to injury from open-heart surgery.6
Based on our own hybrid experience in Giessen,7–9 and the study from Columbus, Ohio,10 ,11 the paper by Lloyd et al raises a couple of interesting questions. Why is the Hybrid approach per se associated with mortality? From the current available techniques and materials, at least in Europe, there now is almost no reason for death from the procedural point of view, compared to when the hybrid procedure was initially developed.8 ,12 In our centre in Giessen, the Hybrid approach is used as an initial palliation for babies with HLHS, and similar forms of single ventricle as well as for some defects with borderline left heart structures. In terms of the physiopathology, the Hybrid approach consists of (1) bilateral pulmonary banding (bPAB) for lung protection, (2) maintaining an open ductus by percutaneous stenting or long-term prostaglandin infusion to preserve adequate systemic perfusion, (3) an atrioseptostomy, including stent placement for unloading of the left atrium. This approach allows all subsequent options for treatment, even in patients with a high Aristotle score, as described in this issue.3 In cases of prostaglandin refractory duct obstruction with consequent metabolic acidosis, percutaneous duct stenting is the treatment of choice; in systemic low cardiac output due to pulmonary runoff, surgical bPAB is the most effective measure; for a restrictive or intact atrial septum, pulmonary venous decompression by catheter techniques is recommended at the immediate first step.
It must be emphasised that the outcome of newborns with HLHS is strongly dependent on thoughtful decision making with the goal of offering an effective, but ‘gentle medicine’ in terms of minimal invasiveness. The bPAB approach still requires an open chest but does not require cardiopulmonary bypass. For this approach to be favoured, a flow-oriented anaesthesia is necessary together with a smooth, bloodless surgical banding of the left and right pulmonary arteries and transpulmonary stent placement; immediate postoperative extubation should be self-evident.
The authors from London adopted in their Hybrid approach from the Columbus method with bPAB and transpulmonary duct stenting, but combined in selected patients surgical atrioseptostomy utilising cardiopulmonary bypass; for me, this is a huge change, far away from a minimally invasive procedure. Several groups have reported on other modifications of the Hybrid approach. The Toronto group expanded the ‘Hybrid’ stage I to an extensive surgical procedure by introducing a reversed BT-shunt, but with the expenditure of the ratio of oxygen consumption to delivery more pronounced than with the classical Norwood procedure.13 The decision to expand the Hybrid approach was based on an institutional experience with classical Norwood/Sano procedure; nowadays the extended Hybrid approach represents a novel option for some patients at the high-risk extreme of the variable cohort of HLHS newborns. Another Hybrid approach consisting of bPAB and long-term prostaglandin infusion was first published from Japan14; this much less invasive strategy has its own benefits and pitfalls, particular as a long-term therapy. Considering the multiple variants of the so-called ‘Hybrid approach’ in HLHS and variation in the technical aspects, differences in outcomes with these approaches requires further assessment and validation.
Another question arises in terms of why a Hybrid programme mostly started with high-risk, but not low-risk HLHS patients? Was this novel approach only possible in patients considered at high risk of an adverse outcome, similar to our experience 15 years ago in Giessen7? Additionally, is it possible that lack of adequate collaboration between the surgeon and interventionalist affects the decision for a surgical versus Hybrid approach? There also are increasing concerns about the effects of retrograde flow on cerebral perfusion, the risk of retrograde coarctation, distortion of pulmonary artery anatomy induced by the bPAB, the effects of pulmonary valve dilation, and so on; as well as the potential to neglect all the problems which may occur when performing a ‘classical’ postnatal Norwood/Sano procedure with cardiac and/or circulatory arrest. Could some arguments also be related on economic aspects? What will happen when the Hybrid stage I can be performed on a sleeping, spontaneous breathing, newborn in the catheterisation suite, and the surgeon focuses ‘only’ on a comprehensive stage II procedure with a subsequent transcatheter Fontan completion? Will we have a comparable debate, as occurred with the advent of transcatheter atrial septal defect closure years ago? According to a report by the Centers for Disease Control and Prevention in 2007,15 analysing 35 common birth defects for their impact on hospital charges resources and mortality, neonates with HLHS undergoing traditional surgical management accounted for the most expensive care of all birth defects, with the fourth longest hospital stay, despite a 34% in-hospital mortality rate. The healthcare system in Germany currently financially rewards prolonged intensive care, intubation and ventilation. By contrast, novel low-invasive procedures, such as a Hybrid stage I or strategies to avoid heart transplantation are not supported under such financial aspects.
An additional reason that the Hybrid approach is currently used in most centres only as a rescue treatment in high-risk newborns might be biased by the definition of the Hybrid procedure itself. The success of a Hybrid approach requires integration of surgical and interventional procedure, and also multiple people with different types of medical and technical expertise. Bacha and Hijazi16 mentioned years ago that becoming a ‘learning leader’ is sine qua non to achieve satisfactory results. Just as starting a Norwood programme, the hybrid approach requires experience; it makes little sense to start with inexperienced teams, for example, with duct stenting intraoperatively, or by percutaneous transcatheter technique, in the hope of good results from the beginning of the programme. In the Columbus and Giessen centre experiences, the surgeons and paediatric cardiologists made the decision to change their classical programme to favouring the Hybrid stage I approach in most, but not all, neonates with HLHS or univentricular variants. Parallel to the improvements of the classical Norwood/Sano procedure, the evolution of the Hybrid-HLHS approach was achieved by sharing the international experiences and innovative modifications to improve the procedure for their patients. In the near future, it will be interesting to hear from the ‘Hybrid group’ in London, if their new four-step approach will improve the outcome of neonates with HLHS, including long-term quality of life.
Competing interests None.
Provenance and peer review Commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.