Left ventricular growth, but not the whole picture

Some of the most anatomically complex hearts present the simplest decision-making in congenital heart disease. Despite the anatomical ‘simplicity’, the borderline left heart remains one of the biggest challenges in paediatric cardiology. Those with significant left heart hypoplasia will proceed down the single ventricle pathway through hybrid or Norwood palliation [1, 2] and those with discrete coarctation or arch hypoplasia with otherwise normal aortic and mitral valves will usually do well with primary repair. Sometimes the ‘borderline’ group is obvious from birth—critical aortic stenosis, Shone complex, but other patients may present with well-functioning, but small aortic and mitral valves with a slender left ventricle. Surgery often needs to be performed in the first 2 weeks of life, and patients may have a significant atrial communication, making true assessment of the left heart challenging.

Contemporary outcomes for those palliated through to Fontan have improved significantly, and although morbidity accrues over time, most children will now transition to adult care having had 3 or 4 planned surgical procedures, with some diagnostic and/or interventional cardiac catheterizations [3].

The concept of rehabilitation of the left side of the heart [4–6] to achieve a biventricular circulation and potentially eliminate some of the comorbidities associated with the Fontan circulation is appealing. The application of repair techniques and more recently the replacement of the mitral valve at lower weights as well as increasing experience with Ross in infancy has enabled the more consistent optimization of the inflow and outflow of the left heart.

The question, however, has always been ‘at what cost?’. Rehabilitation of the left heart usually requires multiple surgical procedures [7, 8]; not infrequently necessitating eventual replacement of the aortic and mitral valves often with a Konno modification and its associated risks. Patients are potentially therefore being set up for a life of reintervention: at a minimum of valve dilatation or re-replacement and pacemaker generator and lead changes.

An additional concern of ‘forcing’ a heart down a biventricular pathway has been the effect on the pulmonary artery pressures. The pulmonary vascular bed will often see high left atrial pressures from mitral valve abnormalities compounded by elevated left ventricular end-diastolic pressure from endocardial fibroelastosis, chronic obstruction, hypoplasia and diastolic dysfunction [9]. Although there may be an increase in left ventricular volumes with recruitment, most borderline left ventricles display impaired relaxation despite an apparent ‘normal’ appearance in terms of size. Damage to the pulmonary vascular bed and pulmonary hypertension can not only cause clinical issues for the patient but also make conversion to a single ventricle pathway impossible, and may preclude the already limited candidacy for a transplant.

The ventricular–ventricular interaction is an important concept in normal physiology. Loss of a normal second ventricle can have an impact on the remaining systemic ventricle but the presence of a dysfunctional ventricle, particularly if it is dilated or hypertrophied, also represents an additional risk. Hence, whilst allowing growth of the left ventricle may assist the ventricular–ventricular interaction, if that ventricle is dysfunctional, then this could further hamper a systemic right ventricle if biventricular repair is then not possible.

Given the challenges of the Fontan pathway, it is always refreshing to see attempts to try and move patients to a biventricular pathway. The Super Glenn described by Marathe et al. [10] describes a targeted approach to increase pulmonary blood flow (with a fenestrated atrial communication) and therefore flow through the left heart with the aim of ‘growing’ the left heart to then be able to support the systemic circulation. This has been successful in converting some patients from a single to biventricular pathway, although the follow-up data are relatively short, with limited information on right heart pressures and, partially related to the age of the patients, no quality of life or objective measures such as exercise capacity.

The population included was mixed, with a range of hypoplastic left heart complex, critical aortic stenosis and unbalanced atrioventricular septal defects and if this were to be applied wider, selection criteria would need to be defined, and more work on the individual subgroups. The overall freedom from death and transplantation is superior to the majority of centres’ single ventricle programmes at 1 year (100%, but bearing in mind that these are all initial palliation survivors rather than all-comers), but by 5 years, this is similar (79%), with patients having more interventions than an average single ventricle patient.

Advances in strategies to improve the morbidity and mortality of patients with borderline left heart structures are welcome, but we must always be cautious when considering strategies where long-term outcomes are still being established. Although a 2-ventricle circulation should produce more stable and ‘normal’ physiology, the outcome is dependent on unobstructed inflow, outflow and adequate ventricular systolic and diastolic function. The adage of ‘a good Fontan is better than a bad biventricular repair’ still holds true, although we may be making progress towards improving the alternative.

REFERENCES