Evolution of the Fontan Procedure
The Fontan operation is the treatment of choice for patients with a single anatomic or functional ventricular chamber. From the late 1940s, survivors with a functionally univentricular heart could be palliated with a systemic-to-pulmonary artery or Glenn shunt, but by the early 1970s, only 50% of those with tricuspid atresia—the most favorable form of functionally univentricular heart—survived 15 years ( Fig. 73.1 ). In that era, complications of cyanotic heart disease including stroke and cerebral abscess, and progressive ventricular failure and atrioventricular valve regurgitation due to chronic volume loading were the commonest causes of death.
The understanding that blood could flow through the lungs without a subpulmonary ventricle led to development of the Fontan operation, first described by Fontan and Baudet as “corrective surgery” for tricuspid atresia in 1971. The original procedure consisted of division of the right pulmonary artery from the main pulmonary artery and anastomosis of the superior vena cava to the right pulmonary artery. The left pulmonary artery was then anastomosed via a homograft valve to the right atrium, and the main pulmonary artery was disconnected from the hypoplastic ventricle. The valve was used to promote the “ventriculization” of the right atrium so that it might generate sufficient pressure to augment pulmonary flow ( Fig. 73.2 ). The atrial septal defect was closed and a valve placed at the junction of the inferior vena cava and the right atrium to prevent retrograde flow during atrial contraction. Six months later, Dr. Guillermo Kreutzer and team carried out their first atriopulmonary Fontan anastomosis. This procedure included a 6-mm fenestration in the interatrial septum and as such was the first fenestrated Fontan. No valve was positioned in the venous circulation because it was thought that this would result in a degree of obstruction—a hypothesis that was proven correct because it soon became apparent that valves within the Fontan circulation were associated with a high risk of stenosis. This technique evolved to generate an atriopulmonary anastomosis as wide as possible without the use of a valved conduit ( Fig. 73.3 ).
The atriopulmonary connection became the standard Fontan modification through to the late 1980s. However, over the long term, this circulation was associated with progressive dilatation of the systemic venous atrium, atrial thrombus, and intractable atrial arrhythmia. In a series of elegant hydrodynamic experiments, de Leval demonstrated the energy loss associated with the atriopulmonary anastomosis and potential for greater circulatory efficiency if much of the right atrium was excluded from the systemic atrial pathway by using an interatrial patch. This technique—termed the total cavopulmonary connection or lateral tunnel Fontan—reduced the degree of turbulence and energy loss and improved overall hemodynamics. Shortly after, the extracardiac conduit was introduced by Marcelletti et al. by interposing a prosthetic valveless conduit to connect the inferior vena cava to the pulmonary artery. This aimed to avoid progressive atrial dilation, late tachyarrhythmia, and sinus node dysfunction by reducing the number of suture lines and the pressure load within the right atrium ( Fig. 73.4 ).
Currently, both the lateral tunnel and extracardiac conduit are widely used, some preferring the former technique in younger patients and those with anomalous drainage of their hepatic veins. Studies demonstrate comparable hemodynamics in both circulations. Nevertheless the extracardiac conduit is the preferred technique in many centers because of the perception that it will be associated with a reduced late arrhythmia burden, although to date this has not been reliably demonstrated.
Late Outcome With a Fontan Circulation
As the fifth decade of Fontan surgery approaches, the burden of late morbidity and mortality has become apparent, with the risk of complications and death increasing the longer the duration of the Fontan circulation. Late outcome studies report a survival rate of 60% to 80% 20 years post-Fontan surgery. Variable case selection and duration of follow-up likely account for this range in outcome. At 25 years after Fontan surgery, almost half the cohort is predicted to face Fontan failure, defined as circulatory dysfunction with limited functional capacity (New York Heart Association [NYHA] class III or IV), Fontan takedown or conversion, the development of debilitating complications including protein-losing enteropathy (PLE) and plastic bronchitis, the need for cardiac transplantation, or death ( Fig. 73.5 ).
Risk Factors for Late Mortality
Identification of risk factors for long-term outcome and the development of models for risk stratification have the potential to target treatments to the highest risk patients and to guide the development of new treatment strategies. To date, risk stratification has been hampered by a relatively small number of patients and by institutional differences in patient selection and treatment regimens. However, with the increasing numbers of patients and the trend toward multiinstitutional networks and registries, it is likely that risk stratification will become an essential tool to guide the development of effective surveillance regimens and targeted interventions.
Preoperative risk factors for late death include male gender and the diagnosis of hypoplastic left heart syndrome. A higher pre-Fontan mean pulmonary arterial pressure is an important predictor of morbidity and mortality in both the early perioperative and late stages, with a threshold of 15 to 17 mm Hg or less being associated with a better outcome. A higher pulmonary artery pressure is also associated with prolonged pleural effusions in the early postoperative period, as well as the development of PLE in the late stage, both of which independently predict late mortality. Having a common atrioventricular valve (CAVV) is also a predictor of late death, with almost 50% of CAVVs having failed 20 years after Fontan surgery. Moreover, a CAVV is frequently associated with heterotaxy syndrome and anomalies of pulmonary and systemic venous drainage, both of which are also risk factors for late failure ( Table 73.1 ).
|Preexisting (pre-Fontan) factors||Male gender |
Hypoplastic left heart syndrome
Common atrioventricular valve
Higher mean pulmonary artery pressure (>16–18 mm Hg)
|Perioperative factors||Type of Fontan (atriopulmonary worse) |
Older age at Fontan operation (>7 years)
Operative complexity (e.g., aortic cross clamp time, bypass time, concomitant atrioventricular valve replacement)
|Early postoperative factors||Elevated Fontan circulation pressure (>20 mm Hg) |
Elevated ventricular filling pressure (>13 mm Hg)
Prolonged pleural drainage (>3 weeks)
|Late postoperative factors||Protein-losing enteropathy |
Reduced exercise capacity (peak VO 2 )
Those with an atriopulmonary Fontan are at greater risk of late death when compared with the more recent variations ( ). However, a survival advantage of the extracardiac conduit over the lateral tunnel has not been demonstrated. When Fontan and colleagues reviewed 160 Fontan surgeries from 1968 to 1988, they found older age at Fontan surgery was predictive of late death. A more recent experience similarly demonstrated a poorer late survival when the Fontan operation was undertaken after 7 years of age. Surrogate markers for surgical complexity including longer aortic cross-clamp time, bypass time, and concurrent atrioventricular valve replacement also impact on late survival.
The main factors in the postoperative course that influence late mortality relate to the presence of elevated pulmonary arterial or Fontan pathway pressure. A postoperative left atrial pressure greater than 13 mm Hg or Fontan pressure greater than 20 mm Hg is associated with a twofold increase in risk of late death. Prolonged pleural effusions, usually described as chest tube drainage for more than 3 weeks after surgery, is one of the strongest predictors of late death. Besides being a marker for elevated pulmonary arterial pressures, it may also be influenced by other factors, including longer cardiopulmonary bypass time, the presence of aortopulmonary collateral vessels, and the absence of a fenestration.
Beyond the perioperative period, the identification of risk factors becomes more challenging due to the insidious nature of disease progression. The development of late complications, including PLE and arrhythmia, and the requirement for ventricular pacing are markers for late failure and are described in detail later in this chapter.
Cardiopulmonary exercise stress testing is an important prognostic tool in the Fontan population. Of all the measured exercise variables, peak VO 2 is the most robust in predicting late morbidity and mortality. Those with a peak VO 2 of less than 16.6 mL/kg/min have a mortality risk seven times of those with a higher peak VO 2 ( Fig. 73.6 ). A lower peak heart rate or reduced heart rate reserve, defined as the difference between peak exercise and resting heart rates, has also been identified as a useful marker of function and prognosis. However, it is important to recognize that confounding factors such as antiarrhythmic therapy and pacemaker-dependence may influence exercise capacity and reduce its prognostic power.
Consequent to improved survival the Fontan population is becoming older, with the average age predicted to increase from 18 years in 2014 to 23 in 2025 and 31 years in 2045. The effective management of these patients will depend on the identification of those at greatest risk of decline, as well as potentially modifying the current approach to staged reconstruction on an individual basis. Well-defined patient surveillance strategies will allow physicians to deliver timely targeted interventions with the aim of increasing longevity and quality of life (QOL).
Consequences of the Fontan Circulation
The Fontan circulation is characterized by elevated central venous pressure (CVP) and a low or low-normal cardiac output with a limited capacity to increase cardiac output with exercise. Arrhythmias are common and may be caused by atrial distension, especially in the case of the atriopulmonary Fontan, or by scarring subsequent to surgical interventions. Elevated CVP and reduced cardiac output adversely affect the function of a number of organs, including the hematologic, renal, liver, and lymphatic systems. Many of the resulting problems have an insidious onset, but, as the time passes, they contribute very significantly to morbidity, mortality, and QOL late after the Fontan procedure.
Impaired Exercise Capacity (see also Chapter 23 )
Performance of the Fontan circulation is limited at rest and with exercise, even with optimal anatomic and circulatory conditions. This remains an issue regardless of the type of Fontan procedure and suggests that the problem relates in a large part to the inherent limitations of the circulation itself. In the normal circulation, the subpulmonary ventricle has an important role to play in augmenting cardiac output with exercise. Its absence is central to the limited exercise capacity observed in the Fontan population ( Fig. 73.7 ). The magnitude of the reduction in exercise capacity is best demonstrated by cardiopulmonary exercise testing. Maximal exercise capacity is determined by the highest uptake and utilization of oxygen by the body during maximal exercise (VO 2 max) based on achieving a plateau of VO 2 despite an increase in workload. The highest achieved VO 2 value (VO 2 peak) is used as a substitute when this plateau is not achieved; a common occurrence in the Fontan population ( Fig. 73.8 ). In a structurally normal heart, the major factor limiting VO 2 max is cardiac output, which accounts for 70% to 85% of variance, with the remainder being derived by other factors, including pulmonary and skeletal muscle function and cellular metabolism. Multiple studies have demonstrated reduced VO 2 peak or VO 2 max in Fontan patients. Importantly, a lower VO 2 peak is associated with an increased risk of morbidity and mortality. There is also reduced workload at maximal effort, a variable reduction of VO 2 at ventilatory anaerobic threshold, a reduced peak O 2 pulse, and chronotropic incompetence with a blunted peak heart rate response.
There are several aspects of the Fontan circulation that contribute to impaired exercise capacity ( Fig. 73.9 ). Under normal conditions, cardiac output is augmented by increases in preload, heart rate, and myocardial contractility and a reduction in afterload. Stroke work is increased substantially more in the subpulmonary (right) ventricle than the systemic (left) ventricle. In the absence of a subpulmonary ventricle many of these adaptive responses are absent or compromised.
In the Fontan circulation, preload is chronically depleted, and this effect is magnified under exercise conditions. In the absence of a subpulmonary ventricle, systemic ventricular filling is dependent on diastolic function and low pulmonary vascular resistance to pull blood through the pulmonary circulation. These factors are the primary drivers of exercise capacity in the Fontan circulation. Following volume unloading during staged surgical reconstruction, the functionally univentricular heart reduces in size by 25% to 70%. Although remodeling could compensate for this change by reducing myofiber length, diastolic dysfunction predominates from early on in the majority of Fontan patients, suggesting that remodeling is inadequate. Diastolic dysfunction transmits increased filling pressures to the pulmonary veins. This has a progressive negative impact on the pulmonary vascular bed and systemic venous return, leading to further restriction of exercise performance. Increased systemic venous stiffness and reduced capacitance augment systemic return in the Fontan circulation at rest, but these adaptive mechanisms are less effective with exercise.
Pulmonary vascular resistance limits systemic ventricle preload and cardiac output in the Fontan circulation because it sets the level of energy required to deliver blood from the systemic veins to the systemic ventricle. A number of factors have the potential to adversely impact on pulmonary vascular resistance. For example, vascular compliance may be compromised by the lack of pulsatile flow, whereas reduced wall shear stress may lead to maladaptive changes in lung vasculature and increased pulmonary vascular resistance. Abnormal microscopic pulmonary vascular changes have been described in the Fontan circulation, as has impaired pulmonary artery growth. Others have described abnormal pulmonary vascular reactivity with exercise.
Although the mechanism of abnormal pulmonary vascular behavior is unclear, even a small increment of pulmonary vascular resistance has the capacity to reduce the ability to augment cardiac output with exercise. This has generated interest in the use of pulmonary vasodilator therapy in the Fontan population. However, the mixed results of these medications on Fontan exercise performance suggest that the factors responsible for exercise restriction are complex.
Chronotropic Incompetence (See also Chapter 22 )
The inability of the heart to increase rate commensurate with demand is a recognized predictor of future cardiovascular events and overall mortality in other settings of cardiovascular disease, including heart failure. Chronotropic incompetence is a marker of dysautonomic function and reduced sinus node reserve. In the Fontan population an impaired heart rate response to exercise is a common finding. There is some debate as to its relative contribution to impaired exercise capacity, with some even suggesting that it may form a useful adaptive response when there is diastolic dysfunction. Sinus node dysfunction may be related to damage to the sinus node and its arterial supply during cardiac surgery. More modern cavopulmonary connections have resulted in better preserved chronotropy, but sinus node dysfunction is not eliminated completely because contributors may include prior bidirectional Glenn or hemi-Fontan surgeries. The prognostic implication of chronotropic incompetence within the Fontan population is unclear.
Restrictive Lung Disease
In the absence of a subpulmonary ventricle, the Fontan circulation relies on efficient lung mechanics, with changes in intrathoracic pressure during respiration acting as a suction pump to draw blood through the lungs. It is increasingly recognized that optimized lung parenchyma is an important positive contributor to the Fontan circulation both at rest and under exercise conditions. Restrictive pulmonary function, as demonstrated by reduced forced vital capacity and forced expiratory volume in 1 second is well described in patients with congenital heart disease and is particularly apparent in those with a Fontan circulation. This pattern of abnormal respiratory function is multifactorial with contributions from thoracic surgeries, pleural stiffness, intrinsic lung development abnormalities, and Fontan pulmonary vascular flow dynamics. Side effects from the use of medications, especially amiodarone, may also play a role. Under exercise conditions, restrictive lung function can manifest with reduced O 2 pulse, higher peak minute ventilation (V E ), and reduced ventilation efficiency with higher V E /VCO 2 slope. In addition, several studies have found a significant correlation between impaired resting pulmonary function and reduced peak VO 2 max on exercise in the Fontan population. A small interventional study focused on reduced inspiratory muscle strength in Fontan patients by instituting inspiratory respiratory muscle training. Following training, resting inspiratory muscle strength, cardiac output, and ejection fraction increased, whereas during exercise there was an improvement in ventilatory efficiency.
The relative contribution of the pulmonary circulation and cardiovascular mechanics to exercise restriction in Fontan patients is yet to be fully understood. Further research will improve the understanding of these limitations and lead to new ways of “empowering” the Fontan circulation under exercise conditions.
Impaired Somatic Growth
Somatic growth is impaired during and following staged reconstruction for children with a functionally univentricular heart. Weight and height parameters are most often within the normal range at birth in the absence of prematurity or genetic abnormalities. Following the initial surgical procedure, there is a significant decrease in both height and weight z score. Weight follows a trajectory of relative recovery after the bidirectional Glenn or hemi-Fontan operation, and following Fontan completion in most children. However, height does not demonstrate the same trend. Heart failure, PLE, the presence of venovenous collaterals, and significant atrioventricular valve regurgitation may have a negative impact on weight and height trajectory, although these associations have not been demonstrated in a consistent way. By adulthood, males with a Fontan are shorter than normal population. This relationship is less striking in females. Mechanisms for the reduction in height potential may include lower bone density and reduced muscle mass. Of interest, but as yet unconfirmed, lower exercise participation in Fontan patients may impact on bone growth and subsequent bone/muscle development. Another potential contributor to reduced bone growth is prolonged hypoxemia prior to Fontan completion, although a fenestration after the Fontan operation (a marker for hypoxemia) does not appear to influence height recovery. The impact of the Fontan circulation on insulin-like growth factors (IGFs) and growth hormone and their interaction with somatic growth is yet to be established. In a cross-sectional study, lower IGF was found in Fontan patients with a higher brain natriuretic peptide (a marker for heart failure) and lower cardiac output. However, there was no relationship between IGF and somatic growth. The authors concluded that longitudinal studies were required to determine if these relationships contributed to impaired somatic growth in this population.
There is concern that obesity trends seen in the general population will be similarly seen in the Fontan population. Although obesity is less prevalent in Fontan cohorts compared with the general population, as well as other forms of congenital heart disease, the rate is not insignificant and ranges from 8% to 30%. Moreover, there is a tendency toward increasing weight and body mass index further out from Fontan surgery. Given the reliance of the Fontan circulation on optimal ventricular and vascular function, additional acquired cardiovascular risk factors can contribute only to a worse outcome.
Arrhythmia (see also Chapter 22 )
Arrhythmia is a common problem in the Fontan population, has an increasing prevalence in older patients, and is often associated with Fontan failure. The two most frequent arrhythmias are bradycardia due to sinus node dysfunction, and atrial flutter. The latter is more correctly termed intraatrial reentrant tachycardia (IART). Both of these arrhythmias become more prevalent with time but are not necessarily linked to each other ( Fig. 73.10 ). In a population-based report, bradyarrhythmias are present in 7% at 10 years and 15% at 20 years after the Fontan procedure, and tachyarrhythmia in 9% and 31%, respectively. Tachyarrhythmia is commoner in those with functional limitations, isomerism, and an atriopulmonary Fontan connection when compared with the extracardiac Fontan. A contemporary series suggests atrial tachyarrhythmia is present in most if not all patients 25 years after the atriopulmonary Fontan procedure. The extracardiac Fontan may result in less IART than the lateral tunnel, although the evidence for this is less conclusive. IART is also more common when atrioventricular valve repair or pulmonary vein surgery is required at the initial surgery.
Focal, atrial ectopic tachycardias occur in approximately 13% of patients over long-term follow-up, many in the same patients who have IART. Atrial fibrillation is becoming more frequent in older patients (19% in one series) with risk factors overlapping those of the aging population (such as overweight and hypertension).
The occurrence of bradyarrhythmia or tachyarrhythmia signals a 50% to 60% risk of Fontan failure over the next 10 years.
Ventricular tachycardia (VT) is relatively uncommon and usually asymptomatic, with Holter recordings suggesting a prevalence of approximately 6% 10 years after Fontan. However, symptomatic VT or ventricular fibrillation can occur in up to 3%. The presence of VT correlates with larger ventricular volumes, reduced ejection fraction, and magnetic resonance imaging (MRI) evidence of myocardial fibrosis.
Sudden cardiac death occurs at late follow-up in 5% to 9%. Risk factors include the presence of atrial tachyarrhythmia, atrioventricular valve replacement at the time of the Fontan surgery, and an immediate postoperative systemic venous pressure greater than 20 mm Hg. Preoperative sinus rhythm is protective.
Bradycardia and Pacing
Pacemakers may be used in up to 25% of cases at late follow-up, including those implanted for the management of atrial tachycardia. Pacing for bradyarrhythmia is required in approximately 7% to 15% of patients during long-term follow-up. In approximately two-thirds, the indication is sinus bradycardia and, in one-third, atrioventricular block. The latter is more common among patients with congenitally corrected transposition of the great arteries ( Fig. 73.11 ). Pacemakers are commonly placed when an atriopulmonary Fontan is converted to an extracardiac Fontan. This procedure usually includes antiarrhythmia surgery. Some centers will implant biatrial antitachycardia pacing devices prophylactically during the same procedure.
Bradyarrhythmia is more common after the atriopulmonary Fontan than the lateral tunnel or external cardiac conduit ( Fig. 73.12 ). Heart rate variability, a subtle marker of sinus node dysfunction, is reduced in lateral tunnel and external conduit Fontan in equal measure when compared with healthy controls. There is some suggestion that the external conduit may be associated with more sinus node dysfunction than the lateral tunnel, but this is not a consistent finding. Sinus node dysfunction may relate more to the nature of the prior superior vena cava pulmonary anastomosis (as well as native sinus node function) because that surgery is close to the sinus node and sinus node artery. Atrial pacing, which usually must be epicardial and may require extensive thoracic surgery to be achieved, is generally reserved for those with symptomatic chronotropic incompetence. Depending on the anatomy, it may be feasible to place transvenous atrial leads; however, it is not unusual to have to place leads in nonstandard positions because areas of viable myocardial tissue can be limited ( Fig. 73.13 ).
Ventricular pacing should be avoided or minimized as far as possible because of the risk of causing ventricular dyssynchrony and pacemaker-induced cardiomyopathy, cardiac failure, and atrioventricular valve regurgitation. Fontan patients with ventricular pacing have a fivefold risk of transplant or death compared with matched nonpaced controls. The value of cardiac resynchronization therapy is being explored; results are generally disappointing, but there may be a place in specific cases such as postpacing cardiomyopathy or when there is a systemic left ventricle and left bundle branch block ( Fig. 73.14 ).
Sudden death after the Fontan procedure is not that uncommon, usually occurring in the context of end stage of the Fontan circulation failure. It may be related to events such as pulmonary and cerebral embolism or poorly controlled atrial tachycardias. Implantable cardioverter-defibrillators are a class IB indication for secondary prevention following resuscitated cardiac arrest due to sustained VT or ventricular fibrillation. However, implantation carries a significant risk in those with Fontan failure given that it entails thoracic surgery. If pacing is not required, a subcutaneous implantable cardioverter-defibrillator may be an option in some patients. Careful consideration should be made regarding defibrillation threshold testing at the time of implantation because this process can be lethal in those with severe ventricular failure.
Intraatrial Reentrant Tachycardias
The association of atrial tachycardias with poor outcome is at least partly correlated to the underlying substrate of the arrhythmia, rather than the arrhythmia per se. In the atriopulmonary Fontan, there is an electromechanical correlation between the occurrence of arrhythmia and the degree of atrial dilation and thickening. Risk factors also include previous pulmonary artery banding, isomerism, and a systemic right ventricle.
The observation that atrial dilatation was associated with IART led in part to the adoption of the lateral tunnel procedure. The prevalence of IART has proven to be lower with this type of Fontan connection. The external cardiac conduit approach reduces intracardiac surgery and avoids progressive dilation of the atrial wall, but it is not evident that this operative strategy is associated with a reduced prevalence of IART compared with the lateral tunnel.
Invasive electrophysiologic studies demonstrate that the mechanism of IART commonly involves surgical scars created during suturing of the lateral tunnel. These studies reveal large areas of low-voltage diseased atrial myocardium, with fractionated signals demonstrating delayed and nonhomogeneous electrical conduction. This substrate is ideal for the development of intraatrial reentry ( Fig. 73.15 ).
Atrial reentry circuit depends on areas of slow conduction in diseased atrial myocardium with electrically silent tissue on each side. This results in a slow conducting bridge, or isthmus. An electrical signal enters the isthmus, and by the time the electrical signal is released from this isthmus, the healthy myocardium is able to conduct again; the signal propagates around the atrium and back to the entry point of the isthmus. These areas of slow conduction usually develop around areas of scar either surgical or due to progressive atrial fibrosis ( Fig. 73.16 ). After all forms of Fontan, the commonest position for such an isthmus is between the bottom end of a right atriotomy scar and the inferior vena cava (“pericaval origin”). This is different to other postoperative congenital heart groups and the structurally normal heart with atrial flutter, where the isthmus commonly runs across the anatomic cavotricuspid junction. It has been proposed that an additional surgical line should be made at the time of the lateral tunnel surgery, to prevent such an isthmus developing. Unfortunately, given the time lag in the development of IART after the Fontan procedure, it will be decades before we know if this has been successful. As mentioned before, some modifiable surgical techniques may help to prevent IART. Certainly, the move away from atriopulmonary connection has been beneficial, as has reduced age at the time of the Fontan operation.
Other congenital arrhythmia substrates such as accessory pathway–mediated tachycardias and atrioventricular nodal reentrant tachycardias account for up to 30% of the tachycardias in Fontan patients treated in a tertiary electrophysiology laboratory. These tachycardias are more responsive to medical and ablative therapy than IART.
In the acute setting, medical management can be difficult, and the patient may have decompensated cardiac failure, as both a cause and effect of the IART. Direct current cardioversion can fail in a quarter of patients, with increased success rate if type I or III antiarrhythmic medications are started prior. Medication for rate control can be difficult to manage because of the commonly associated sinus node dysfunction, and although amiodarone can be effective, side effects can be harmful when this medication is used long term. Thus, in the adult patient with an atriopulmonary connection, medical management is frequently unsuccessful. Interventional strategies involve a choice of (or combination of) a catheter ablation strategy versus a surgical takedown to a lateral tunnel or extracardiac conduit with concomitant surgical ablation techniques, usually a maze procedure.
Catheter ablation for IART in the atriopulmonary Fontan can be successful in the short term, but there is a high recurrence rate. This is not surprising given the fact that the underlying substrate—the atrial dilation and wall thickening with large areas of scarred and electrically inhomogeneous tissue—is not altered. The grossly dilated atrium is also a nidus for thrombus formation and is hemodynamically inefficient. The early Fontan conversion experience was one of considerable mortality outside of several high-volume centers. The results of this surgery are improving, and this improvement relates at least in part to a better appreciation of the indications for operation. Many centers have published favorable results, with an early mortality rate of approximately 5%, improved NYHA functional class, and reduction in arrhythmia incidence over 10 years (see later, “ Surgical Management of Fontan Failure ”).
Atrial fibrillation commonly occurs earlier in the Fontan population than in other patients with postoperative congenital heart disease and is generally poorly tolerated. Onset often occurs in the third decade, usually as an intermittent arrhythmia that commonly coexists or alternates with other atrial tachycardias. Progression to sustained atrial fibrillation is common within 5 years of the first episode. The inclusion of left atrial (Cox) maze with right atrial maze at the time of Fontan conversion may prove effective in reducing the recurrence rate of this arrhythmia, especially in older patients and those who already have atrial fibrillation. However, it is not known what proportion of atrial fibrillation has a left atrial/pulmonary vein origin in the Fontan circulation, even though this is the commonest mechanism in the structurally normal heart. There is anecdotal evidence, and it makes intuitive sense, that some atrial fibrillation in these patients has a right atrial origin.
Role of Catheter Ablation
Although a surgical approach may be most appropriate for those with atrial tachycardia with an atriopulmonary Fontan, catheter ablation has a role in other cases. An ablation can be a useful palliation where conversion is contraindicated, atrial dilation is not excessive, or the patient has declined surgery. Focal atrial tachycardias can be relatively straightforward to ablate, along with congenital arrhythmia such as accessory pathways, atrioventricular node reentrant tachycardia, and rare cases with twin atrioventricular nodes. With the extracardiac conduit or lateral tunnel Fontan, the critical isthmus is usually on the cardiac side of the baffle, so that access for ablation catheters is difficult. However, there has been increasing confidence in the use of transbaffle puncture technique in these cases because there is commonly a safe puncture point at the lower end of the baffle at the junction with the inferior vena cava/atrial border.
The dominant arrhythmia post Fontan is atrial tachycardia, with complex atrial reentry circuits. Their appearance is commonly coincident with hemodynamic deterioration and the arrhythmia typically contributes further to lower cardiac output, forming a vicious cycle that may be lethal. Atrial fibrillation commonly alternates with other atrial tachycardias. Current management is tending toward early conversion of the atriopulmonary Fontan to an extracardiac conduit, with concurrent atrial arrhythmia prevention surgery.
Following all types of Fontan surgeries, medical management is usually not sustainable for more than 2 to 3 years and, although amiodarone is the most effective medication, side effects are common. Invasive electrophysiology studies and catheter ablation strategies can be very helpful and are recommended early in the absence of gross atrial dilation and to diagnose and treat concurrent congenital arrhythmia substrates including accessory pathways.
Hematologic and Immunologic Complications
It is well recognized that the Fontan circulation presents a hypercoagulable state, with the incidence of thromboembolic events reported to vary between 8% and 20% of the population. This is likely an underestimation, in view of the occurrence of silent thromboembolism in this group. A prospective multicenter randomized controlled trial assessing several anticoagulant regimens reported a total thrombosis rate of 23% over 2 years. Only one-third of these events (8%) were symptomatic, with the remainder being detected during intensive surveillance as part of the study design. Studies have described a peak thrombotic risk in the first year following Fontan completion, which plateaus over the next 3 to 4 years, before a second peak after 10 years. Moreover, the incidence of thromboembolic complications is higher in adults compared with children, suggesting an increase in risk with time that might relate to a gradual deterioration in vascular and liver function, exacerbated by a tendency to a more sedentary lifestyle in older and more debilitated patients.
The etiology of this prothrombotic state is multifactorial and involves all three factors of the Virchow triad, namely abnormal hemodynamics, a hypercoagulable state, and endothelial dysfunction ( Fig. 73.17 ). Potential factors include the low-velocity flow in the systemic veins, cavopulmonary connection and pulmonary arteries, atrial arrhythmias, persistent cyanosis related to right-to-left shunts, and an imbalance of intrinsic procoagulant and anticoagulant factors.
Risk Factors for Thromboembolism
Older age at the time of the Fontan operation is a risk factor for silent thromboembolism. Surprisingly, there appears to be a similar risk of thromboembolism among the different variants of the Fontan ( Fig. 73.18 ; ). Although the presence of a right-to-left shunt is known to increase the risk of cerebral vascular embolization, the presence of a fenestration has not been associated with increased thromboembolic or stroke risk. This suggests that intrinsic hematologic abnormalities may be the most significant factors in the prothrombotic state in the Fontan circulation. Compared with healthy controls, Fontan patients have reduced levels of procoagulant factors, including factors II, V, VII, and X, and coagulant inhibitors, such as protein C, protein S, plasminogen, and antithrombin III. An elevated level of factor VIII is a strong risk factor for venous thromboembolism in the normal adult population, with a predicted incidence of recurrent thrombosis of more than 10% per year in those with increased serum levels. Longitudinal studies monitoring serum factor VIII levels in patients with a functionally univentricular heart have demonstrated a conversion from low serum levels early in the course of staged reconstruction, to significantly raised levels after Fontan completion. Increased factor VIII activity correlates with higher superior vena cava pressure in the Fontan circulation. As such, it is hypothesized that increased pressure transmitted to the liver sinusoidal endothelium leads to the upregulation of factor VIII synthesis. Thrombocytopenia may also contribute, particularly if related to heparin treatment, in which it may be associated with a high risk of thrombosis, or when associated with portal hypertension and failure of the Fontan circulation (see Fontan Failure). Lastly, progressive endothelial dysfunction develops with prolonged exposure to the Fontan circulation. Even well-functioning adult patients may have underlying endothelial dysfunction, indicated by increased plasma concentrations of endothelin-1 and abnormal digital pulse amplitude tonometry. This multitude of factors leads to the high incidence of thromboembolic events that contribute to significant morbidity and mortality early and late after Fontan surgery.
Many children with a functionally univentricular heart have immunologic anomalies on routine laboratory investigations, the most common being lymphopenia that predominantly involves CD4 T cells. Absolute lymphocyte counts decrease with time after the Fontan operation. Patients who are more than 10 years post-Fontan surgery have been found to be four times as likely to have significant lymphopenia as compared with patients in the first decade post Fontan. However, the clinical significance of these findings is unclear because there does not appear to be an increase in opportunistic infections even in the setting of significant lymphopenia. Increased lymphatic recirculation may be a compensatory mechanism, allowing for preservation of normal tissue-level T-cell function even in the setting of low cell counts. Nevertheless, there is an abnormally high incidence of atopy (approximately 60%), suggesting abnormal skewing of the distribution of residual T cells.
The most significant deficiencies were noted in patients with PLE (see later), although lymphopenia occurs even in the absence of PLE. All patients with PLE are lymphopenic, with preferential loss of T cells (CD4 more than CD8) but preservation of normal levels of B and natural killer cells. Hypogammaglobulinemia is common, mainly affecting immunoglobulin G (IgG) and IgA levels.
Patients with PLE have higher rates of nonresponsiveness to vaccination, particularly to hepatitis B and measles, mumps, and rubella, and may require repeated vaccinations and avoidance of live vaccines. Some advocate for antibiotic prophylaxis against opportunistic infections such as Pneumocystis jirovecii and Mycobacterium avium , although supporting clinical evidence is lacking. In assessing the immunocompromised state of patients with PLE, confounding factors such as malnutrition and the side effects of immunosuppressive therapies should also be considered.
Late survivors of the Fontan surgery invariably experience multiorgan sequelae including progressive liver dysfunction and PLE; however, the long-term progression of their renal function is poorly understood. The early occurrence of acute kidney injury is currently well recognized following complex surgical reconstruction in the neonate, with increasing evidence of late renal dysfunction in these patients (see Chapter 78 ). In addition, in the Fontan circulation, there is reduced renal perfusion as the chronic elevation of CVP increases efferent arteriolar pressure. Glomerular filtration pressure is also increased, leading to a high incidence of microalbuminuria. In a small retrospective cohort study of 21 patients at mean 11 years post-Fontan completion, almost half the group had an increased urine microalbumin/creatinine ratio. A strong correlation between urine microalbumin/ creatinine ratio and superior vena cava mean pressure was also demonstrated.
Microalbuminuria may be a more sensitive indicator of early renal disease than the estimated glomerular filtration rate (eGFR). In a review of 68 patients a decade after Fontan completion, 90% had a normal eGFR (eGFR >90 mL/min per 1.73 m 2 ) but more than 40% had microalbuminuria. Furthermore, serum creatinine may not be a reliable indicator of kidney function in this population. Fontan patients often have a lower muscle mass and are relatively malnourished, as demonstrated by their lower serum creatinine concentration compared with age-matched controls. Nevertheless, elevated serum creatinine is a strong predictor of death or cardiac transplantation, whether as an independent variable or as part of the Model for End-stage Liver Disease Excluding INR (MELD-XI) score. The decline in renal function is insidious and protracted but is likely to play an important role in the prognostication of late survivors.
Fontan-Associated Liver Disease
As survival rates following staged reconstruction have improved, many patients are currently living into their third and fourth decade. The long-term consequences of elevated CVP and low cardiac output have become more apparent. Although the Fontan circulation affects many organ systems outside the heart, its impact on the liver may be the most prevalent complication. Despite this, understanding of the evolution of liver injury is limited, and the role of various screening tests is only now evolving as new information becomes available.
For children with single-ventricle physiology, the insult to the liver probably begins well before the creation of the Fontan circulation ( Fig. 73.19 ). Shortly after birth, infants with functionally univentricular heart disease are subject to one or more surgeries and associated alterations in hemodynamics and oxygen saturations. These derangements may have a profound impact on the architecture of the liver as hepatocytes are subjected to impaired perfusion and hypoxemia. In a series of children who did not survive beyond the Fontan circulation, autopsy demonstrated the consistent finding of fibrosis, confirming the notion that liver injury begins prior to the Fontan.
Although liver injury may not start with the Fontan operation, it is clear that additional changes to the hepatic environment are relatively immediate following Fontan completion. In a study in which an abdominal ultrasound was performed just prior to the Fontan and then repeated 3 to 6 months following total cavopulmonary connection, the liver span was increased and velocities within the hepatic arteries were decreased after the Fontan procedure. This finding is consistent with hepatic congestion and, when coupled with a mild elevation in liver enzymes, suggests that congestion is immediate and likely begins a process of chronic low-level liver injury.
Hepatic congestion and the resultant fibrosis are not a static process but rather one that progresses slowly over time. In a study of adolescents with Fontan physiology, the only confirmed risk factor for the degree of fibrosis was the amount of time that had passed since the initial Fontan operation. Although the sample size was relatively small, ventricular morphology, atrioventricular valve regurgitation, and ventricular function were not associated with the degree of liver fibrosis. Interestingly, systemic venous pressure was not associated with the degree of fibrosis. This may have been related to the relatively narrow range of Fontan pressures of the patients included in the analysis. In another recent report, almost all patients had some evidence of liver abnormality 20 years following the Fontan procedure, with a third having regenerative nodules and 6% established cirrhosis. As reported elsewhere, there was little correlation between liver abnormalities and clinical status. Other factors may also contribute to liver disease in the Fontan population, including hypoxemic damage occurring during episodes of low cardiac output. Hepatitis C infection should be considered in older patients who had cardiac surgery prior to screening for the virus, as should alcohol-related liver damage.
The clinical characteristics of Fontan-associated liver disease remain to be clearly defined, but, as more patients survive into adulthood, it is becoming apparent that the disease is similar to other forms of liver disease with variable degrees of fibrosis and cirrhosis. Mild abnormalities of liver enzymes, especially γ-glutamyl transferase, are common, as is a minor elevation in the indirect bilirubin. Excepting for a mild elevation in the prothrombin time, indices of synthetic function such as albumin are usually normal unless there is advanced cirrhosis (or PLE).
Ultrasound of the liver frequently demonstrates heterogeneous echotexture and arterialized nodules. These nodules are striking but appear to be benign and may be an attempt to increase blood flow to the liver by increasing arterial supply. MRI or ultrasound elastography of the liver demonstrates increased liver stiffness. Although this finding is likely a combined result of congestion and fibrosis, one small study has shown that liver and splenic stiffness on elastography was strongly correlated with the degree of biopsy-proven fibrosis.
Perhaps the most feared consequence of advanced liver disease in the Fontan circulation is the development of hepatocellular carcinoma. What once seemed isolated to rare case reports has become more common as more patients survive well into adulthood. Unfortunately, the detection of hepatocellular carcinoma may be challenging, particularly on a background of abnormal hepatic parenchyma. Screening tests such as serum α-fetoprotein may be helpful but are not sensitive or specific enough to reliably diagnose each case. Regular abdominal ultrasound to detect nodular growth may be helpful. Contrast computed tomography (CT) or MRI, both the gold standard for diagnosis of hepatocellular carcinoma in other settings, may be less reliable in the Fontan circulation. The tumor is supplied from the hepatic arterial circulation, whereas the normal liver receives most of its blood supply from the portal vein. These tests detect early contrast enhancement in the tumor and later enhancement in the surrounding tissue. Elevated systemic venous pressure in the Fontan circulation may interfere with this relationship.
In addition to the development of hepatocellular carcinoma, progression of cirrhosis and its complications may herald the failure of the Fontan circulation. Portal hypertension is not uncommon, with splenomegaly seen in 20% of patients at a median of 10 years after the Fontan operation. Venous collaterals from the liver or esophageal varices are seen in more than half of those with functional limitation. Abdominal ascites may be cardiac in origin but can also be associated with liver cirrhosis. In this setting, the development of ascites is a poor prognostic sign. As intraabdominal fluid increases, a vicious cycle ensues with increased abdominal pressure leading to increased venous hypertension and an even more pronounced decrease in cardiac output. Ascites is often relatively resistant to diuretic treatment and, while not reported in the literature, some patients have been managed with repeated peritoneal taps to drain abdominal fluid in the hope of maintaining a functional Fontan circulation either as a palliative procedure or while awaiting heart or heart and liver transplantation.
Liver disease is a frequent, serious and progressive entity following the Fontan operation, but screening algorithms and management are less well defined. Serial monitoring with liver biopsy is not practical and may be impacted by the heterogeneity of the hepatic manifestations. For now, regular surveillance with abdominal ultrasound and elastography, as well as serial measurements of serum α-fetoprotein, may be the best option, particularly in those patients more than 10 years out from the Fontan procedure. Multiple societies, interest groups, and individuals are working on follow-up protocols for all organ systems, particularly the liver. Continued work toward the development of medical and surgical strategies to lower venous pressure and improve cardiac output may slow the progression of liver disease. The Fontan operation has helped to save or prolong many lives, but more work is needed to help manage the complications that result from this unique circulation.
The physiology created by the Fontan operation results in both obligate central venous hypertension and persistent low cardiac output. Although these physiologic abnormalities may be well tolerated, at least for a few decades, there is a subset of patients in whom severe complications may occur much earlier. Plastic bronchitis and PLE are both feared complications of the Fontan circulation, and both may lead to significant morbidity and mortality.
In recent years, a great deal has been learned about the role of the lymphatic system in the pathophysiology of plastic bronchitis and PLE. The lymphatic system, the scavenger of the circulatory system, is responsible for retrieving interstitial fluid and returning it to the central circulation via the connection of the thoracic duct to the innominate vein. For patients with Fontan physiology, there is obligate lymphatic hypertension that results from the transmission of the elevation in CVP ( Fig. 73.20 ). The lymphatic system is further inundated by the increase in lymphatic fluid production that results from increased intravascular and intrahepatic hydrostatic pressure associated with heart failure. Although MRI imaging demonstrates universal dilation of the lymphatics in patients with Fontan physiology, there appear to be some patients in whom the lymphatic hypertension and dilation lead to lymphatic insufficiency, often with severe consequences.
Plastic bronchitis is characterized by the development of abnormal lymphatic vessels in the peribronchial region. These abnormal vessels form tiny fistulous connections to the airways, allowing for a slow but insidious leakage of lymphatic fluid. The fluid itself dissipates over time with respiration, leaving behind a proteinaceous material composed of fibrin and inflammatory cells that ultimately coalesce into “plastic casts” of the airway ( Fig. 73.21 ). These casts cause obstruction of the airways, leading to cough, to ventilation-perfusion mismatch, and, in severe cases, to asphyxia ( Table 73.2 ). Exacerbations are more common in the winter months and may be precipitated by respiratory infections. The onset often occurs within a few years of the Fontan procedure, and the condition is more common in those who had chylothorax at the time of surgery.
|Plastic bronchitis||Elevated intralymphatic pressure||Cough||Pulmonary vasodilation|
|Increased lymphatic production||Low oxygen saturations||Inhaled tPA|
|Abnormal lymphatic connections to the airways||Asphyxia||Lymphatic intervention|
|Protein-losing enteropathy||Elevated intralymphatic pressure||Diarrhea||Diuretics|
|Increased lymphatic production||Ascites||Pulmonary vasodilation|
|Abnormal lymphatic connections to the intestines||Peripheral edema||Controlled-release budesonide|
|Muscle wasting||Lymphatic intervention|
Treatment for plastic bronchitis has evolved rapidly over the past decade as the mechanism of the disease has become clear. Medical management strategies for plastic bronchitis include treatment with bronchodilators, inhaled steroids, and pulmonary vasodilators. For those with chronic cast production, inhaled tissue plasminogen activator can be added to the medical regimen to dissolve the fibrin within the casts. This treatment can be quite successful at controlling the symptoms and severity of the disease. However, although medical therapies may be effective at controlling symptoms, they do not change the underlying abnormalities of the lymphatic vessels nor do they eliminate the connections between the lymphatic vessels and the airways.
Recent advances in lymphatic imaging and intervention have allowed for a more selective approach to the treatment of plastic bronchitis. Using T2 MRI imaging or dye injection into the lymph nodes, the lymphatic system can be visualized and abnormal lymphatic networks can be mapped. Once mapped, new interventional techniques for accessing the lymphatic system can be used for the delivery of agents to embolize the network of abnormal lymphatic vessels, thereby reducing the leakage of lymphatic fluid into the airways and eliminating the formation of casts. The early experience with this technique is encouraging. Its use is expanding, and it has the potential to become a definitive therapy for those who develop this complication ( Fig. 73.22 ).
Like plastic bronchitis, PLE is a consequence of lymphatic insufficiency. In the case of PLE, the abnormal lymphatic connections form between the lymphatics originating in the liver and the small intestine. These abnormal connections allow protein-rich lymphatic fluid to drain from the high-pressure lymphatic system into the low-pressure gastrointestinal tract. Over time, this results in a profound loss of proteins and a phenotype characterized by diarrhea, extreme muscle wasting, particularly of the extremities, and often a combination of peripheral edema and chronic ascites. Although 5-year survival after PLE diagnosis has improved dramatically over the past 2 decades (from 50% to 88%), the disease remains severely debilitating, with a significant impact on quality and duration of life. In addition to the phenotypic abnormalities, patients with PLE have abnormalities in bone structure and in the characteristics of the immune system. The diseases present some years after the Fontan operation. Although there is an association with complex heart disease, including hypoplastic left heart syndrome, and with postoperative chylothorax, many affected individuals have no obvious risk factors. The onset is often insidious, but once PLE is established, patients with this condition are among the most chronically debilitated of all those with congenital heart disease.
Although it is well understood that abnormal lymphatic connections to the intestine are responsible for the loss of lymphatic fluid that occurs with PLE, there are likely other contributing factors to this disease. Endoscopy often demonstrates an inflammatory pattern similar to that observed in children with inflammatory bowel disease. Like inflammatory bowel disease, there are cases of PLE that can be controlled through the use of corticosteroids. Oral controlled-release budesonide is a steroid formulated to target its antiinflammatory properties to the distal small intestine. Budesonide has the added advantage of first pass hepatic metabolism, theoretically leading to a lower degree of systemic absorption than with other corticosteroids.
Unfortunately, some cases of PLE are steroid resistant. In older adolescents and young adults, controlled release budesonide has not been as effective a treatment. Although the response to steroids in younger children suggests an inflammatory component of the disease, the absence of a response in adolescents and young adults suggests that inflammation may play less of a role in older patients. Although the underlying abnormal connections between the lymphatic vessels and the intestine are present in both age groups, the differential response to treatment suggests that the breakdown of the integrity of the intestinal mucosa may have more of a relationship to inflammation in younger patients.
In any age group, the treatment strategy for PLE begins with optimizing the overall Fontan physiology. Diuretics may be helpful by decompressing the lymphatic system and may also treat some of the symptoms associated with PLE. Pulmonary vasodilators may be useful adjuncts because the reduction in pulmonary vascular resistance leads to a drop in CVP and therefore a reduction in intralymphatic pressure. Other potential therapies have been reported and include the administration of heparin and even treatment with low-dose dopamine infusion. Unfortunately, although medical treatment has led to improved outcomes, the burden of morbidity and mortality remains substantial.
As in the case of plastic bronchitis, the emergence of lymphatic imaging and intervention has the potential to alter the strategy for the treatment of those with PLE. By accessing and occluding the abnormal lymphatic connections from the hepatic lymphatics to the intestines, one could substantially alter the rate of loss of lymphatic fluid. However, although promising, lymphatic intervention for PLE is not as successful as it has been for plastic bronchitis. Nevertheless, recent reports of lymphatic intervention suggest that with further technical refinement, there is the potential to fundamentally alter the trajectory of those who develop this feared complication.
Peripheral Vascular Disease
Increased arterial stiffness and small arterial lumen diameter have been reported in Fontan patients, the latter presumably a response to a chronic reduction in cardiac output. These arterial abnormalities, in combination with those of lymphatic drainage described earlier and chronic venous insufficiency, manifest in the lower extremities as varicose veins, edema, and, in serious cases, chronic ulceration. These lower extremity problems occur in approximately 20% of Fontan patients older than 18 years and are more common in those who have had multiple cardiac catheterization procedures or central venous catheters via the femoral veins. They are also associated with deep venous thrombosis and as such may predispose to pulmonary and systemic embolism.
Neurocognitive and Psychosocial Function
The diagnosis of a life-threatening illness during a child’s formative years can have far-reaching effects that ripple through the family and across a lifetime. Children following the Fontan pathway experience profound physical, emotional, behavioral, neurodevelopmental, and social challenges in the early years of life, and these challenges have the potential for lifelong consequences, particularly in relation to future health, well-being, and QOL.
Quality of Life
The World Health Organization defines quality of life as a dynamic, multidimensional concept, unique to each individual’s perception of his or her position in life and his or her physical health, psychological well-being, level of independence, relationships, personal beliefs and values, and environmental context. In the context of functionally univentricular congenital heart disease, stressors associated with diagnosis and treatment, the uncertainty of progressive functional limitations, and the possibility of heart failure, heart transplantation, arrhythmias, and sudden death may compromise QOL. Overall, the majority of published studies report lower QOL for children and adults with a Fontan circulation compared with normative data, age-matched controls, or healthy siblings ; however, a small number of studies report QOL similar to the general population. Although some studies have demonstrated an association between lower QOL and greater complexity of cardiac abnormality, daily medication use, greater length of hospital stay, and greater number of medical interventions, most research has found that social and psychological constructs, such as greater psychological stress, fewer social supports, and lower family socioeconomic status, play a more influential role in determining QOL. Consensus on the role of these risk factors in the Fontan population has not been established.
From a neurocognitive perspective, it is well established that children with complex congenital heart disease are at increased risk of neurodevelopmental impairment, particularly those with a functionally univentricular heart. Although studies report a lower mean intelligence quotient for groups of children and young people with a Fontan circulation compared with healthy peers, the majority of patients have intellectual function within the normal range. However, there is a higher prevalence of impairments in executive functioning, visual construction and perception, fine and gross motor skills, language, attention, and academic performance in childhood compared with population norms. Risk and severity of neurodevelopmental impairment are associated with individual factors (e.g., presence of a genetic syndrome, hypoplastic left heart syndrome, structural brain abnormalities, cyanosis, genetic factors) and environmental factors (e.g., prolonged deep hypothermic circulatory arrest during cardiac surgery, postoperative seizures, longer length of hospital stay, lower socioeconomic status, greater psychological stress). Although many pediatric cardiac centers currently include neurodevelopmental clinics, the same attention has not been paid to neurocognitive health in adult CHD care. Without effective intervention and support, hardships encountered during childhood can endure for years after cardiac diagnosis and treatment. It is also possible for difficulties to emerge for the first time in adolescence or adulthood, with heart failure, atrial fibrillation, cardiac surgery, and recurrent strokes increasing vulnerability to neurocognitive impairment later in life.
Illness-related stressors can challenge children and young people’s emerging coping skills during the peak years of onset for mental health disorders. Children and adolescents with a Fontan circulation (aged 10 to 19 years) have been found to have higher rates of lifetime psychiatric diagnosis (65%) compared with healthy referents (22%), particularly anxiety disorders (Fontan: 35%, referent: 7%) and disruptive behavior disorders such as attention-deficit/hyperactivity disorder (Fontan: 34%, referent: 6%). Health-related fears, separation anxiety, body image concerns, and sleeping difficulties are also common. Overall, patient-specific demographic, perinatal, medical, and psychosocial factors tend to be better predictors of later psychological outcomes than intraoperative factors. Several mechanisms for psychological morbidity in complex CHD are important to consider. Exposure to early physiological risk, such as in utero brain immaturity, perioperative hemodynamic alterations, and systemic inflammation, may adversely affect neurobiologic development and consequently alter long-term responses to stress, increasing the risk of psychological morbidity. In addition, studies of individuals exposed to high levels of stress early in life consistently show that the experience of early adversity is associated with disrupted child-parent attachment and alterations in the developmental trajectories of networks in the brain associated with emotion and cognition.
Parents and siblings also experience higher levels of psychological distress compared with population norms, yet these groups may be vulnerable to falling between the cracks in terms of clinical assessment and access to evidence-based emotional health care. Parents with high distress report poorer physical health, greater parenting burden, higher health service use, and more suicidal ideation compared with parents of children with complex CHD with lower distress. Many who suffer from these difficulties never receive psychological treatment.
Without effective intervention, psychological effects can be enduring and can influence a patient’s capacity to successfully transition from pediatric to adult health services, with potentially life-threatening consequences. Evidence-based, theoretically grounded interventions may mitigate the development of mental health difficulties in response to serious illness across the family system. Regular screening and assessment for psychological morbidity in people of all ages with a Fontan circulation and their families are clearly indicated. In general, integrating psychosocial assessment within a clinical setting with which patients are already engaged is a key factor distinguishing successful and unsuccessful early mental health interventions, improving treatment uptake and mental health outcomes. However, at this time of writing there are no published data on the efficacy of psychological interventions developed specifically for children or adults with a Fontan circulation.
Although it is clear we need to address psychological vulnerabilities across the family system, the best framework through which to do so remains to be determined. Early theories conceptualized the illness experience as a trauma for the patient, leading to an emphasis on trauma-focused psychological therapies. Despite the inclusion of life-threatening illnesses as meeting criteria for a traumatic event within the Diagnostic and Statistical Manual of Mental Disorders, researchers have more recently critiqued this diagnostic conceptualization due to the lack of clarity regarding one triggering traumatic event. Going forward, we need to better understand the mechanisms underlying psychological morbidity for children and adults with a Fontan circulation and their families and to continue efforts to foster integrated models of psychological and neurocognitive care. These represent some of the next frontiers of research and clinical practice in the field.
Management of the “Well” Fontan
Routine Surveillance and Testing: the Role of Practice Guidelines
The Fontan population is heterogenous with a wide spectrum of functional performance. The morbidity profile is highly variable depending on a number of factors including underlying anatomy and ventricular morphology, time from the Fontan procedure, and the era in which Fontan surgery was performed. Hence practice guidelines must be adaptable to these features and must also take into account the life stage of the patient. For example, the pediatric population has care priorities that differ to those of adults. Moreover, the transitions between life stages, lifestyles, and physiologic and psychological states are important milestones that need to be anticipated, and discussed and planned in advance. There is a paucity of published practice guidelines that adequately traverse the lifetime care of a Fontan patient. This is largely due to the lack of a demonstrable advantage of any particular treatment options or management strategies. Consequently, there is significant practice variation. There are few practice guidelines for pediatric care, especially for the “well” Fontan, in part due to a general perception that these patients have a low resource requirement. Moreover, there is no evidence base to assess the impact of a more structured approach to ongoing surveillance on outcome in this group. The majority of published guidelines are for adult patients, often with a focus on the failing Fontan. Nevertheless, a structured approach to surveillance in childhood is likely to lead to lower resource use and a better understanding as to which investigations are productive at a given time. In addition, specific timing of discussion and counseling in relation to exercise participation, teenage risk taking, contraception, and transition to adult care is likely to improve outcomes in these areas. The latter is especially important given the mortality and morbidity risk associated with drop out from cardiology surveillance at the time of transition.
The Fontan patient requires regular surveillance over his or her lifetime ( Fig. 73.23 ). The core requirement is a regular clinical review with a pediatric cardiologist during childhood and an adult congenital cardiologist thereafter. The frequency of review in pediatric practice is debatable; however, American and European guidelines for adult congenital heart disease recommend annual review. Echocardiography forms the mainstay of imaging given its relative ease of access and noninvasive properties. Useful information primarily relates to the function of the ventricle and the assessment of valve regurgitation and the outflow tracts, although additional information including the quantification of the gradient between the Fontan pathway and the pulmonary venous atrium through Doppler interrogation of a patent fenestration or the diagnosis of thrombosis can be useful. The primary limitations of echocardiography include a reliance on geometric indices to assess ventricle size and function, which is often problematic given the heterogeneous ventricular morphology encountered in the Fontan population and the increasingly challenging acoustic properties in older patients. Nevertheless, it continues to have a place in surveillance for older patients who are MRI incompatible. Its usefulness would increase if nongeometric indices of diastolic and systolic performance were able to predict outcome. Small studies have shown inconsistent performance to date in this regard, although deformation assessment appears to hold the most promise. Cardiac MRI is superior to echocardiography in the assessment of Fontan flow dynamics and the size and systolic function of the systemic ventricle, especially when it is a right ventricle. It is a useful adjunct to surveillance particularly in the adult with a Fontan circulation, although current guidelines leave its inclusion in regular surveillance to individual assessment. The ability to perform exercise MRI may lead to its inclusion in future surveillance algorithms. The utility of cardiac MRI is limited in the pediatric population given the frequent need for general anesthesia. Cardiac CT and cardiac catheterization are useful where there are specific questions not answered by echocardiography or MRI.
With half of all patients with late Fontan failure having preserved systolic ventricular function, the utility of this type of routine surveillance may be limited. More sensitive markers of cardiac function would be useful. In particular, an increased end-diastolic volume (EDV) index, most reliably measured with cardiac MRI, may be a better marker for cardiac status and appears to have better prognostic value in the Fontan population. EDV reflects the preload of the systemic ventricle and is influenced by a combination of factors including venous capacitance and degree of ventricular dilatation. It is elevated immediately after Fontan completion but reduces to normal levels after the first year. The etiology behind the progressive increase in EDV in the later years is not understood. It may be the result of chronic volume overload, secondary to fluid retention, aortopulmonary collaterals, atrioventricular valve regurgitation, and ventricular dyssynchrony secondary to arrhythmia.
The electrocardiogram provides insight into loss of sinus node function, heart block, and other arrhythmias, which are particularly prevalent in those with original atriopulmonary type Fontan connections. Holter monitoring and event and implantable loop recorder monitoring add layers of additional surveillance where arrhythmia is suspected.
Symptoms may be underreported in the Fontan population due to a lifetime adjustment to a different functional normality, and self-assessment of functional status is poorly correlated with quantitative assessment. Hence serial assessment with cardiopulmonary exercise testing is useful for ongoing surveillance. Exercise capacity declines over time, and the rate of decline may better predict future adverse events than the absolute exercise capacity at a particular point in time. The frequency and age at which to start is not clearly established, especially in those in New York Heart Association class 1 or 2.
The liver and kidney are detrimentally impacted by elevated systemic venous pressure and restricted cardiac output. Consequently, abnormalities of these organs are often apparent in the Fontan population and can be found even in those with little in the way of functional limitation. Current recommendations are for intermittent screening with liver and renal function serum testing and ultrasound scans without clearly specified intervals. The sensitivity and specificity of these tests as early screening tools in this population are not conclusive. New biomarkers and imaging modalities are emerging, but, as with existing renal and liver function testing, their predictive value needs to be established prior to their being incorporated into surveillance programs. Recent reports of hepatocellular carcinoma in older Fontan patients with cirrhosis highlight the importance of screening for both of these conditions.
Routine surveillance and testing in the Fontan population are an evolving process. As information from larger population-based studies emerges, more robust guidelines can be developed for the lifetime care of the Fontan patient.
Given the propensity for thrombosis in the Fontan circulation, the need for antithrombotic prophylaxis is generally accepted, with the highest rates of thrombosis described in retrospective studies where prophylaxis was not universal. The two most widely used prophylactic medications are aspirin and warfarin. Even with these agents, there remains a substantial rate of thrombosis (7% to 19% of cases). The mortality risk associated with clinically evident thrombosis is significant, ranging from 12% to 28%. Moreover, the recurrence risk is substantial, with further thromboses occurring in more than a quarter of patients. Two time periods of greatest risk for thrombosis have been identified, within the first year of Fontan operation and late (≥10 years post Fontan).
There is a diversity of opinion as to whether warfarin or aspirin should be used as primary antithrombotic prophylaxis in Fontan patients. This is reflected in marked practice variation as detailed in recent surveys. Dosage regimes and INR targets are poorly defined, and current guidelines are unable to provide conclusive evidence for their recommendations. The only prospective randomized control trial to date comparing aspirin and warfarin did not demonstrate a clear difference in thrombotic events between the two regimes over a 2-year period, despite intensive thrombosis surveillance. Nevertheless, subanalysis suggests that those receiving warfarin who have consistently subtherapeutic INR measurements are at higher risk of thrombotic events. This finding is supported by several retrospective analyses. Anticoagulation with warfarin carries a significant risk of serious hemorrhagic events. Of interest in the aforementioned study, this type of complication occurred in 1.75% in the warfarin group over 2 years compared with none in the aspirin group. Others have reported significant bleeding events on prophylactic anticoagulation with events primarily occurring in patients anticoagulated with warfarin, especially when taken for many years. In the adult Fontan population, the ideal antithrombotic regime is a conundrum because there may be an increased risk of hemorrhage related to gastric varices and other comorbidities, in addition to the risk of thrombosis. To further complicate matters, there is a high risk of a second thrombotic event if a patient commenced on warfarin following a thrombotic event has his or her warfarin discontinued because of a bleeding event.
In addition to the elevated risk profile, long-term warfarin therapy carries a higher financial cost and has greater impact on QOL compared with aspirin. The need for regular blood tests, the difficulty in maintaining a consistent therapeutic window, and the need to avoid at-risk activities has significant economic social and psychological cost.
Given the aforementioned, many centers limit warfarin prophylaxis to the high-risk early postoperative Fontan period, using aspirin after the first postoperative year in all but those at higher than normal risk of thrombotic episodes. There is little experience with the use of newer antithrombotic agents in the Fontan circulation, but this may change. Perhaps of greater interest is the potential role pharmacogenetics may have in individualizing the prophylaxis regimen. Genetic variants are known to influence warfarin dose requirements and the risk of bleeding events early after starting treatment. In addition, aspirin resistance is well recognized. Routine testing for genetic susceptibility to thrombosis, propensity for bleeding, and resistance to antithrombotic medication may form part of the assessment and treatment individualization in the future. However, with the possible exception of aspirin resistance, there is currently insufficient evidence to support a recommendation in this area.
Role of Angiotensin-Converting Enzyme Inhibition, Aldosterone Antagonists, and β-Blockade
In the well Fontan patient, the use of angiotensin-converting enzyme (ACE) inhibitors, known to be efficacious in adult patients with structurally normal hearts and congestive heart failure, has little evidence base. There have been two randomized controlled trials in children with a functionally univentricular heart: one prior to the Fontan and one smaller study after. Neither demonstrated benefit in terms of improved growth parameters, ventricular function, or exercise performance. ACE inhibitors are frequently used in Fontan patients with and without systolic dysfunction, on the assumption that the development and/or progression of ventricular dysfunction will be delayed, particularly in the setting of a systemic right ventricle. Despite their wide spread use, the efficacy of ACE inhibition in the Fontan circulation remains unproven, and the extrapolation of treatment effect from populations with different forms of heart failure should be made with caution.
Aldosterone antagonists reduce mortality when used in conjunction with ACE inhibition in adults with congestive heart failure and a structurally normal heart. The potential mechanisms for this effect include modification of adverse remodeling by reducing interstitial fibrosis, and a reduction in the risk of ventricular arrhythmia and sudden death due to an increase in serum potassium levels. The Fontan circulation is associated with increased activation of the renin-angiotensin-aldosterone (RAA) system. There is evidence in children with congenital heart disease and adults with Fontan failure that diastolic ventricular dysfunction is associated with increased RAA system activation and with high-risk RAA system genotypes. There is a high prevalence of diastolic dysfunction in Fontan patients, and a therapeutic role for these medications has been postulated. However, there is little evidence of therapeutic efficacy in the Fontan population to date.
As with ACE inhibition, β-blockade carries a mortality and morbidity benefit for adult patients with congestive heart failure and systolic dysfunction. The same effect has not been demonstrated in children, although studies have been underpowered. A single study in a small group of children and adults with Fontan failure and ventricular dysfunction showed an improvement in ejection fraction following treatment with carvedilol. β-Blockade can be useful for the treatment of atrial and ventricular arrhythmia in the Fontan circulation, but careful monitoring is required in the presence of heart block and sinus node dysfunction. There is no other indication for its use in the Fontan patient with normal ventricular function.
In summary, there are few studies assessing the efficacy of these medications and those that do exist are generally small and often statistically underpowered. Although there is no strong evidence in the Fontan setting to support the use of medications that have become the mainstay of heart failure treatment strategies in the adults with acquired heart disease, there is little to suggest harm when used in the setting of heart failure accompanied by systolic dysfunction in the Fontan population. Pediatric and congenital heart failure guidelines support their use, and they remain potential options to ameliorate the Fontan circulation where evidence of overt systolic dysfunction is found. There is currently no evidence to suggest these medications provide a protective effect in patients with normal ventricular function despite their not infrequent use in this context.
Reduced exercise participation and physical deconditioning are common in the Fontan population. The etiology is complex, in part related to the decreased exercise capacity but also to perceptions of physical ability and the psychosocial reaction to living with chronic disease. Inactivity is thought largely responsible for the reduced bone density and muscle mass reported in Fontan patients, although medication use, especially diuretics and antithrombotic treatment, may also contribute. These factors are increasingly recognized as having a detrimental impact on exercise capacity. Recent studies have demonstrated that not only is exercise safe for Fontan patients, especially at submaximal levels, but exercise may be even more important as a therapeutic entity because it compensates for the lack of a pulmonary pump and may improve chronotropic incompetence and respiratory reserve. Although exercise as a therapy is still in its infancy, it is generally acknowledged that encouraging regular exercise to promote cardiorespiratory fitness and muscle conditioning should be recommended as part of long-term Fontan care, with the aim of preserving or enhancing functional capacity.
Despite a high incidence of infertility in women with a Fontan circulation, there are an increasing number of reports of successful pregnancies. If pregnancy is considered, prepregnancy counseling and careful planning ensure that risks are understood and mitigated as far as is possible. Unintended pregnancy in women with a Fontan circulation carries a significant risk to the fetus and mother. For the fetus, the risk includes exposure to medications that are potential teratogens and placental dysfunction as a consequence of the Fontan circulation. The latter is manifest in high rates of miscarriage and intrauterine growth retardation. Risks to the mother include a limited ability to increase cardiac output, an increased risk of arrhythmia, progression of ventricular impairment, and a hypercoagulable state. Sexual health and discussion related to pregnancy risk should be undertaken at an early stage—ideally before sexual maturity is reached. There are multiple potential contraception methods for women with a Fontan circulation ( Table 73.3 ). Preparations containing estrogen are not recommended, due to their prothrombotic risk. There are multiple progesterone-only options for women, including tablet, implantation, and long-acting intramuscular injection; however, side effects and a limited window for missed doses with some oral forms, make these medications less appealing for some women. Intrauterine devices are not contraindicated, but recognition of the potential for severe vagal response during insertion needs to be taken into account, as does the risk of infection. The latter is highest at the time of insertion and then falls to the background rate of sexually transmissible infection. Simple barrier protection methods can also be used, but reliance on these as the only contraception method carries a higher risk of contraception failure even with reliable use.
|Type of Contraception||Benefits||Risks||Recommendation in Fontan Patients|
|Abstinence||No interactions||Relies on absolute compliance||Recommended|
|Combined estrogen/progesterone pill||High intrinsic efficacy |
Long window of cover for missed dose (12 h)
Interacts with warfarin
Dependent on daily compliance
|Progesterone-only pill||Not thrombogenic||Side effects, especially menstrual irregularity |
Some only 3 h window of cover for missed dose
Formulation variation in overall efficacy
Dependent on daily compliance
Reduced efficacy with Bosentan
|Progesterone implant||Not thrombogenic |
Long lasting (3 years)
High intrinsic efficacy
|Side effects, especially menstrual irregularity |
Reduced efficacy with Bosentan
|Progesterone intramuscular injection||Not thrombogenic |
Long-lasting contraception (12 wk)
High intrinsic efficacy
|Small risk of hematoma with warfarin |
Dependent on strict 12 weekly compliance
|Condoms/barrier methods||Not thrombogenic |
Additional protection against sexually transmitted infections
|User dependency |
General lower efficacy
|Recommended ideally in conjunction with other method|
|Intrauterine device||Not thrombogenic |
Long-lasting contraception (5 years)
|Potential for severe vagal reaction during insertion |
Copper coils higher risk of menorrhagia and dysmenorrhea
Infection risk, especially during insertion
|Second line recommendation|