Ventricular contribution and the impact on outcome of the Fontan operation: less is more or more is less?

The Fontan procedure has become a standard cardiac surgical procedure for thousands of patients with functional single ventricle type of congenital heart disease since the first successful procedures that were reported in the early 1970s by Dr Fontan and colleagues in France and Dr Kreutzer and colleagues in Argentina.1,2 Initially, the procedure was used in children and adolescents with tricuspid atresia, a group of patients with single ventricle who were able to live extended years with a spectrum of aortic to pulmonary artery shunts and/or Glenn operations. Though these patients survived with these palliations, it was often at the expense of cyanosis, exercise intolerance and heart failure. By incorporating all systemic venous blood flow into the pulmonary arteries, the cyanosis was reversed with the potential for longer-term survival and better quality of life. Use of the Fontan procedure to treat other forms of single ventricle congenital heart disease expanded over the ensuing decades as new surgical strategies allowed children with more complex forms of single ventricle (including severe aortic outflow obstruction) to survive and reach the age when a total cavopulmonary anastomosis was feasible. Indications for the Fontan operation eventually extended to children with various forms of biventricular heart disease where the additional ventricle was deemed too small to provide adequate output or in cases where septation of the heart into 2 circulations was not possible. Over the recent decades, large cohort studies have been accomplished for patients who have functional single ventricle due to relatively low surgical mortality after total cavopulmonary anastomosis is achieved. This has enabled investigators to examine the impact of ventricular morphology (dominant right or left ventricle) and additional ventricle on outcome. Specifically, the question has remained, is a ‘two ventricle’ Fontan better or worse than a ‘single ventricle’ Fontan? Not all additional ventricles are alike. One could argue that an additional ventricle may be associated with better outcome because more ventricular mass potentially augments cardiac output. Yet, there are other additional ventricles that may cause an undue burden because of hypertrophy and poor contractility; this is seen particularly in those with severe outflow obstruction/atresia but a patent inflow such as in hypoplastic left heart syndrome of the aortic atresia/mitral stenosis subtype or pulmonary atresia with intact ventricular septum. The NHLBI-funded Paediatric Heart Network has been following a cohort of Fontan patients for almost 2 decades. In the most recent investigation of this cohort (Fontan 3),3 Atz et al. found that of the 373 patients included (mean age 21 years), survival was independent of ventricular morphology. However, in paired testing, there was a small but significant decrease in ejection fraction when compared to the original study. In a large single-centre cohort of 773 patients over 17 years with a large percentage having a morphologic right ventricle (50%), survival based on ventricular morphology was not different between those with a right ventricle versus a left ventricle but some divergence of the Kaplan–Meier curve was seen in later adolescence suggesting that over the long term, the morphologic left ventricle may indeed result in improved survival.3 From the same cohort, another study demonstrated that Fontan morbidity was no different based on ventricular morphology.4 Other large cohorts studies have had similarly concluded that ventricular morphology is not a risk factor for morbidity and mortality.5 Others have addressed the variable of an additional ventricle and its impact on outcome in single ventricle. In hypoplastic left heart syndrome, left ventricular size can be variable and several studies have shown minimal impact on survival at least early in life.6 Most recently, Marathe et al.7 examined the very issue of whether a ‘two ventricle’ Fontan is better than a ‘single ventricle’ Fontan. The group examined 1377 patients in the Australia-New Zealand Fontan (ANZFR) registry with a median follow-up of 11.5 years since Fontan. Only a very small proportion fit into the category of ‘two ventricle’ Fontan (79 of the 1377). They found that even with propensity-score matching for confounding factors, freedom from Fontan failure was no different between the groups. In the current issue of the European Journal of Cardiovascular Surgery, Rossi et al.8 have reported on a cohort of patients who have undergone the Fontan operation examining the question of whether an additional ventricular chamber impacts outcome. This retrospective study reviewed all patients who underwent the Fontan operation over a remarkably long period of 40 years. In contrast to the ANZFR study, 22% of the 210 patients met the criteria of biventricular Fontan. There was also good distribution of dominant ventricle type with 78 patients having a dominant right ventricle, 115 patients having a dominant left ventricle and 17 patients with an undetermined ventricle. The authors found no difference in early complications after Fontan or early death for those with biventricular Fontan in comparison to single ventricle Fontan. There was a significantly longer length of stay in the intensive care unit for the biventricular Fontan group and longer time of chest tube drainage that may be linked together. Of the cohort, Fontan revision after atriopulmonary Fontan was much more common in the univentricular Fontan group (one could surmise that this is because the older patients were more likely to have tricuspid atresia), which may be why that group was more likely to have rhythm disturbances. The arrhythmia burden and incidence of protein-losing enteropathy was surprisingly low considering that they are frequent complications of the Fontan procedure and cause significant morbidity.9 The authors showed that predictors of late mortality were heart failure (NYHA classification), liver disease (ultrasound and transient elastography) and pulmonary hypertension (patient on pulmonary vasodilator therapy). These risk factors are well known to be associated with mortality in this population from other large cohort studies.3–5,7,10 The authors hypothesized that an additional ventricle would be associated with worse outcome because of ventriculo-ventricular interactions and the potential for abnormal flow dynamics. They also suggested that diastolic function may be worse with a biventricular Fontan because of competing blood flow characteristics. They surmised that the longer length of stay and chest tube drainage in the biventricular Fontan group may be the result of these flow characteristics. Unfortunately, accurate measurement of diastolic function by non-invasive measures in the single ventricle is fraught with problems. Moreover, a recent study of a large cohort of Fontan patients, which specifically measured indices of diastolic function found no association with ventricular morphology.9 There are many reasons for prolonged chest tube drainage, which may have little to do with ventricular morphology including the importance of the lymphatic system. Interestingly, the authors did not report on systolic ventricular performance or atrioventricular valve regurgitation in this present publication. Both of these variables are known to be risk factors for survival for the functional single ventricle population. They also did not report on the type of additional ventricle seen and whether it was contracting or associated with a severely stenotic or atretic semilunar valve. Despite the potential concerns about diastolic function and the additional ventricle raised by the authors, no significant differences were found between the groups. Rossi et al. should be commended for this study addressing ventricular morphology in patients undergoing the Fontan procedure. They concluded that there was no significant difference in morbidity or mortality related to having an additional ventricle. From this and other studies, it is relatively safe to conclude that the ventricle itself, whether right or left, whether 2 or 1, does not have a major impact on short- or medium-term outcome for individuals with a Fontan circulation. Many other factors such as arrhythmia burden, pacemaker placement, protein-losing enteropathy, plastic bronchitis, liver fibrosis, thromboembolic events, ventricular failure and atrioventricular valve regurgitation have a significant impact on outcome in this population. Though we now have 40 years of experience with the Fontan procedure, following these cohorts over their lifetime remains essential to our continual education, so that we are able to properly counsel families and patients about what may lie ahead. It appears that the Fontan circulation, the strategy applied to treat the single ventricle, is itself the source of complication and failure, independent of the nature of the ventricles. Despite these known concerns, many Fontan patients are living relatively healthy lives with this physiology. For children born with functional single ventricle, the Fontan operation has been an extraordinarily successful palliation but it is also an unsuccessful cure for congenital heart disease.