3.6 Congenital heart disease

The number of patients with congenital heart disease (CHD) who reach adulthood is continuously increasing. It can be stated that 90% of children with CHD survive into adulthood.¹ Moreover, in the last few years, the number of adults with CHD has exceeded the number of children with a congenital heart defect. As a consequence, today, increasing care is needed for adult congenital heart disease (ACHD). Although the majority of ACHD patients have a stable health, many will develop complications later in life ( Figure 3.6.1), of which arrhythmia and heart failure (HF) are the two most common. The literature reports an overall prevalence of complications in one-quarter of patients after a median follow-up of 35 years.^2,3 However, the prevalence might even rise to 60% according to the underlying ACHD diagnosis.⁴ These complications are associated with premature death and increased hospitalization, and negatively impact the quality of life. There is a tendency to extrapolate the treatment strategy used in patients with acquired heart disease to ACHD patients with HF. In some cases, this is acceptable and could have beneficial effects on clinical outcome, but in other cases, this standard approach to treat HF might have adverse effects, with progressive deterioration of the clinical status of the patient. The reason for this phenomenon is that ACHD patients are a highly heterogenous population, with failure of not only a systemic left ventricle (LV), but also of a systemic subpulmonary right ventricle (RV) or a single ventricle, with or without persistent left-to-right or right-to-left shunting. It is obvious that standard drug therapy could cause adverse effects. Today all therapeutic suggestions are based on expert opinion (level of evidence C).⁵ Even use of resynchronization therapy, the criteria for, and timing of, ventricular assist device (VAD) implantation, and timing of (combined) transplantation are unexplored in larger ACHD series. The relatively low number of ACHD patients per expert centre and the high heterogeneity of the ACHD HF population make it almost impossible to set up randomized controlled trials (RCTs) with a view to ameliorate the level of evidence from C to A. This chapter provides an overview of the complexity of ACHD HF patients and proposes some treatment recommendations. These recommendations must be appreciated on top of existing guidelines. In addition, emphasis is put on the clear need for further studies, preferably RCTs.

In general cardiology, HF is in the majority of cases related to a failing systemic LV. Failing might occur in systole and/or in diastole, known as HF with reduced ejection fraction (EF) and HF with preserved EF, respectively. There are many ACHD patients who might fit in these HF categories, in relation to impaired systemic LV function, such as those with aortic coarctation or aortic stenosis. A failing subpulmonary RV is also not that uncommon in general cardiology and is mostly related to pressure load on the RV secondary to moderate to severe pulmonary hypertension or volume load caused by severe tricuspid valve regurgitation. The subpulmonary RV in ACHD patients might also be volume-loaded due to a significant supratricuspid left-to-right shunt or moderate to severe long-standing pulmonary valve regurgitation (e.g. tetralogy of Fallot). These causes are rather rare in acquired heart disease. Moreover, not all subpulmonary ventricles in ACHD are morphological RVs, but LVs (after atrial switch repair for transposition of the great arteries (dextro (d)-TGA) and in congenitally corrected transposition of the great arteries (ccTGA)). The haemodynamic effects of pressure and/or volume overload in these cases differ from those on a subpulmonary RV and might need an altered therapeutic approach. As a consequence, many ACHD patients have a systemic RV instead of an LV (TGA with atrial switch or ccTGA). Which structural and functional parameters are considered to be normal for a systemic RV are unknown, and this knowledge gap makes it difficult to define a failing systemic RV and determine when to diagnose HF.⁶ Therefore, it still remains unclear when to start HF therapy and, if commenced, what kind of therapy should be given. General cardiology is dealing with a typical biventricular circulation: two pumps—one for the systemic circulation and one for the pulmonary circulation. In more complex ACHD, a univentricular circulation or a one and a half circulation might be present (e.g. tricuspid atresia). These ventricles might also become dysfunctional and lead to clinical signs of HF. However, these signs could be misleading, and be related to circulatory failure (which differs from HF), and not to ventricular dysfunction, but rather be caused by circulatory obstructions in the circuit. A typical example is an obstruction to the systemic venous return in a Fontan circuit. Clinical signs of HF might be present but should be considered as due to circulatory failure.⁷ Obviously, HF in ACHD might also occur due to tachyarrhythmias or even ischaemic heart disease.⁸ The incidence of acquired heart disease is rising because of the ageing ACHD population. Moreover, it is hypothesized that ACHD patients might have a genetic predisposition for an abnormal haemodynamic reaction, which might explain the faster progression to HF in some specific patients.

Finally, similar to HF in acquired heart disease, HF in ACHD has to be considered as a multi-organ disease. HF in ACHD might be related to impaired renal function, and cardiac liver cirrhosis is not uncommon in a failing Fontan circuit or in ACHD pathology with elevated systemic atrial pressures. Protein-losing enteropathy might occur in a failing Fontan circuit. Chronic elevated systemic venous filling pressures are considered to trigger the development of protein-losing enteropathy (PLE) and plastic bronchitis. The literature reports that, in some cases, approximately 30–50% of ACHD patients show significantly impaired renal function.⁹ Haematological disorders occur often in cyanotic patients with elevated haematocrit levels, triggering symptoms and complications related to hyperviscosity.

The diagnostic process of symptomatic and clinical HF in ACHD basically is similar to diagnosing HF in acquired heart disease. Progressive impairment of functional capacity and typical signs of systemic and/or subpulmonary HF make the diagnosis quite clear. However, exercise capacity can also be lower in ACHD patients without obvious HF,¹⁰ which sometimes makes the diagnosis of HF very difficult. Moreover, clinical signs in ACHD might be misleading. Patients who underwent a bidirectional Glenn procedure have—even in the absence of HF—an increased jugular venous pressure because of the connection of the superior vena cava (SVC) to the pulmonary circulation. Also, in Fontan patients in whom both the SVC and the inferior vena cava are connected to the pulmonary circulation, more congestion occurs—in many cases, without signs of circulatory failure. Electrocardiography, echocardiography, and, if requested, cardiac magnetic resonance imaging or computed tomography might help to confirm a clinical diagnosis of HF and to identify the triggering aetiology. Measuring biomarker levels (B-type natriuretic peptide and N-terminal pro-B-type natriuretic peptide) in these patients is of utmost importance by confirming the diagnosis of HF and has added value in determining the prognosis and outcome. However, biomarker levels might be moderately increased in more complex ACHD without signs or symptoms of HF. The latter can also complicate the diagnosis of HF.

The diagnosis of HF in asymptomatic patients is less evident, especially when clinical signs of HF are absent. Technical examinations might help in assessing for signs of HF, but in many cases, only ventricular performance is impaired. In acquired heart disease, when the systemic LV function is reduced to an EF <50%, it can be considered as preclinical HF, in which case, treatment is generally initiated. However, whereas in acquired heart disease, myocardial dysfunction is progressing, this is not usually the case in typical ACHD patients with a systemic LV; for example, a patient who underwent a ventricular septal defect closure with a large patch might have an EF <50% and remain stable for many years or even lifelong, without the need for drug treatment. Thus, EF as an only parameter is not sufficiently reliable to confirm a diagnosis of HF. Here, changes in biomarker levels and/or changes in exercise capacity (measured on cardiopulmonary exercise testing) might help to identify the transition from still normal ventricular function to pathological ventricular dysfunction with output failure. A position paper from the Working Group of Grown-Up Congenital Heart Disease and the Heart Failure Association of the European Society of Cardiology (ESC) has proposed an algorithm that could be applied in daily practice.⁸ A more recent review by Leusveld et al. and the new ESC guidelines on ACHD⁵ recommend to shift the use of biomarkers earlier in the diagnostic algorithm for HF in ACHD.¹¹

The same problem occurs and is even more prominent in patients with a systemic RV. The systemic RV in ACHD patients has rarely an EF ≥50% and this is often the reason why patients with a systemic RV are too quickly and unjustifiably considered to be in HF. Also, the EF measurements obtained from patients with a single ventricle or with a one and a half circulation may not really help in identifying a deterioration of myocardial function. In both cases, changes in biomarker levels and/or changes in exercise capacity might help to identify the first stage in the evolution to symptomatic and clinical HF.⁸ An example of a diagnostic algorithm that could be applied is presented in Figure 3.6.2.

Before medical treatment is started, structural optimization always needs to be considered and performed when possible. As in acquired heart disease, pressure and/or volume unloading of a ventricle by carrying out a surgical or interventional procedure might lead to positive remodelling of the systemic or subpulmonary ventricle and prevent the development of clinically evident HF. In addition, arrhythmias need to be treated when and where possible. Although the ESC HF guidelines¹² recommended for patients with acquired heart disease are frequently also applied to ACHD patients, the level of evidence for doing so is still lacking. Therefore, extrapolation of recommendations may not be appropriate and an individualized approach is warranted for ACHD patients.

In ACHD with output failure from a systemic LV, neurohormonal and cardiac autonomic activity is increased. With the diagnostic criteria in mind, initiation of standard HF therapy can be considered, more specifically renin–angiotensin–aldosterone system (RAAS) inhibitors, beta-blockers, mineralocorticoid receptor antagonists, and diuretics. Although use of digoxin remains controversial, it may be beneficial in patients with diminished ventricular function and atrial fibrillation with increased ventricular response. The decision to start treatment is straightforward when symptoms and clinical signs are present. However, the benefit of medical treatment in ACHD patients with impaired systemic LV function with no increase in biomarker levels or decrease in exercise capacity has never been documented.

Although not supported by any scientific evidence, standard HF therapy is initiated as part of routine management of most patients with a failing systemic RV. Beneficial effects are expected in theory, based on increased neurohormonal and autonomic activity. However, it remains unclear whether patients with a systemic RV and reduced EF, with or without an increase in biomarker levels and/or impaired exercise capacity, would benefit from standard HF therapy. Many centres do not initiate treatment if there is no change in functional status and no increase in biomarker levels. It remains doubtful of whether to initiate standard HF treatment even when patients remain completely asymptomatic, with increasing biomarker levels. RAAS inhibitors have not been proven to show clear clinical benefits, except in some symptomatic patients,^13,14,15 whereas beta-blockers might have a positive effect on functional capacity, atrioventricular valve regurgitation, and RV remodelling.^16,17,18 Use of diuretics achieves the most rapid subjective positive response, and when there is evidence of fluid retention, diuretics should be prescribed.

The subpulmonary RV in ACHD is more frequently exposed to volume and/or pressure overload than in acquired heart disease. When timely structural optimization is not possible, RV failure can occur, leading to systemic venous congestion. As long as patients remain asymptomatic, with no clinical signs of HF, no specific medical treatment is needed. When symptoms and clinical signs of HF are present, loop diuretics are the preferred choice. In cases of therapy-resistant right-sided HF, thiazides may be added. RAAS inhibitors and beta-blockers have never been investigated in this setting. In RV failure due to severe pulmonary hypertension, use of selective pulmonary vasodilators is preferred. This is discussed in detail in the ESC guidelines on treatment of pulmonary hypertension.¹⁹

Failure of a single ventricle usually occurs in patients who underwent Fontan palliation. The dominant ventricle can be a morphological RV or LV. When the single ventricle becomes dysfunctional, pressures in the Fontan circuit increase, the circulation fails, and the patient becomes symptomatic. Loop diuretics are mostly used to lower the filling pressures and achieve symptomatic relief. However, overuse of diuretics can reduce the preload and cause a cardiorenal imbalance. Mineralocorticoid receptor antagonists can have a positive impact on PLE. There are no data to suggest a beneficial effect of RAAS inhibitors in Fontan patients. Only carvedilol has proven effects on symptoms and signs of HF. If the Fontan circuit fails due to progressive increase in pulmonary vascular resistance, selective pulmonary vasodilators, such as endothelin receptor antagonists or phosphodiesterase-5 inhibitors, can be considered.

In more complex congenital heart defects, pulmonary blood flow might be decreased due to high pulmonary vascular resistance or mechanical outflow tract obstruction. Co-occurrence of an intra- or extracardiac shunt mediates right-to-left shunting, resulting in central cyanosis. Moreover, the systemic, subpulmonary, or single ventricle can also fail in these patients. In contrast to patients without shunting, reduced systemic afterload can further exacerbate right-to-left shunting and progressive systemic desaturation. In these patients, medical HF therapy always needs to be initiated with great caution.

The above paragraphs have mainly focused on therapeutic regimens for systolic dysfunction. However, HF with preserved EF can also occur in ACHD patients. LV outflow tract obstruction, including (sub)valvular aortic stenosis and coarctation, and genetic predisposition, chronic cyanosis, and chronic deprived ventricular filling, are potential causes of diastolic dysfunction of the ventricle. To date, no studies on medical treatment for diastolic HF have reported a beneficial effect on outcome. Again, data from general cardiology are extrapolated for application in the treatment of patients with ACHD, but without proven benefit. Diuretics might improve signs and symptoms, and use of beta-blockers is expected to prolong the ventricular filling time.⁸

Besides standard HF therapy, other pharmacological agents are also used in patients with ACHD. When beta-blockers prove inadequate in lowering the heart rate, ivabradine can be given, although the latter has only been tested in acquired heart disease.²⁰ When HF is resistant to RAAS inhibitors, a combination of valsartan and sacubitril could help maintain therapy-resistant patients compensated.²¹ Adding sodium–glucose cotransporter 2 inhibitors might also be beneficial. Clinical evidence for iron supplementation in ACHD patients with HF is lacking, although it sounds scientifically logical to extend this treatment approach for general cardiology patients with HF to ACHD patients with HF. Moreover, iron deficiency is not that uncommon in ACHD patients²² and needs to be treated to improve functional capacity.²³ Antiplatelet therapy or oral anticoagulation is only indicated in cases of coronary or peripheral atherosclerosis, or where there is a risk of local thrombus formation, in the presence of mechanical valves, or thrombogenic supraventricular arrhythmias, respectively.

A summary of medical treatment in ACHD patients with HF is provided in Table 3.6.1.

In patients who are unresponsive to medical treatment or in whom significant electromechanical dyssynchrony is present, cardiac resynchronization therapy (CRT) could play a beneficial role. CRT has been proven effective in the failing systemic LV where it can lead to functional improvement and a reduction in HF-associated morbidity and mortality. CRT is frequently associated with the implantation of an automatic defibrillator when the patient is at high risk of malignant arrhythmia and sudden cardiac death. In general, the proportion of non-responders are higher among ACHD patients, compared to patients with ischaemic heart disease and idiopathic dilated cardiomyopathy. Moreover, the long-term effects of CRT in ACHD patients are currently still unknown. However, there are some positive features that have been observed in ACHD patients treated with CRT. A systemic LV resynchronizes better than a systemic RV.²⁴ Also, when only the RV is continuously paced, the risk of dyssynchrony increases, in which case applying CRT could improve the resynchronization parameters.²⁵ CRT is also recommended for patients with double discordance and complete atrioventricular block, although anatomical issues may hamper device implantation. The pacing site for optimal results is of utmost importance, particularly in patients with single ventricle physiology.²⁵ In functional class IV patients, CRT can serve as a bridge to mechanical assist devices and/or heart transplantation. Remote monitoring, such as with use of implantable loop recorders, might help in early detection of malignant arrhythmia. More details on CRT and ACHD can be found in the respective guideline.²⁶

Use of VADs is also applicable in ACHD patients with a failing systemic LV. They can be used as destination therapy or as a bridge to transplantation.²⁷ Use of VAD therapy has been described in some patients with a failing systolic RV, but success rates can be compromised by anatomical complexity, trabecularization of the RV, and associated comorbidity,^28,29 and even more so in patients with a failing Fontan circulation. The function of a single ventricle, both morphological LV and RV, can be supported by VAD therapy. However, liver cirrhosis, associated haematological and coagulations disorders, and anatomical/technical issues can undermine feasibility and beneficial clinical outcome.^30,31 Finally, VAD therapy can also be applied in cases of a failing subpulmonary ventricle, but again here no hard clinical outcome data are available to support a definite level of evidence.²⁷

The number of ACHD patients with end-stage HF is continuously increasing. In general, when the systemic LV or RV starts to fail in biventricular circulation, pulmonary vascular resistance is meticulously monitored to avoid a combined heart–lung transplantation. It has been suggested that the short-term outcome of transplantation in ACHD patients is worse, compared to that in patients with non-ACHD pathology (20–30% 30-day mortality in ACHD patients). However, long-term outcome seems to be similar to, or even better than, that in patients with acquired heart disease. Following the first year post-transplantation, ACHD patients showed improved survival: 5-year survival ranged from 69% to 80%, compared to approximately 72% in non-ACHD patients, whereas 10-year survival was even further improved, ranging from 52.8% to 57.4%, compared with 50.9% to 53.6% in non-ACHD transplant recipients.³² The transplant risk is higher in ACHD patients with single-ventricle lesions, compared to those with a biventricular circulation.³³ Moreover, human leucocyte antigen sensitization is detrimental to transplant outcome and is a specific concern in patients with ACHD who already underwent several surgical procedures or blood transfusion, and/or had pregnancies in the past.³⁴ Assist devices might be used as a bridge to transplantation or as destination therapy. The results of heart transplantation for both failing LV and RV are good. In some patients, only a heart–lung transplantation is an option, especially in patients with Eisenmenger syndrome or those in whom the pulmonary vascular circulation remains immature. The high number of bronchial collateral arteries, and consequently associated high intraoperative bleeding risk, could reduce the success rate of a heart–lung transplantation down to only 50%. Moreover, there is the constant ethical balance between the number of donors and the number of patients on the waiting list. A heart–lung transplantation uses three organs that could be implanted in three other patients. Many centres have abandoned combined heart–lung transplantation because of the high operative risk and the shortage of donors. Some specific cases are chosen for double lung transplantation and repair of (simple) underlying congenital heart defects. Most of these results are similar to, or even worse than, standard heart–lung transplantation. Long bypass time, as well as longer ischaemic time, compromises, in most of these cases, this type of approach. An emerging issue is progressive failure of the Fontan circulation. A single heart transplantation should be the most effective approach, but a failing Fontan circuit is frequently associated with liver cirrhosis. Many transplant centres opt for combined heart–liver transplantation, but this intervention has to be seen as high risk and relevant experience is very limited. A failing Fontan circulation is characterized by multi-organ dysfunction, which, in turn, influences the clinical outcome. Finally, the anatomical complexity and the number of previous interventions have to be taken into account for risk estimation. Timely counselling offered by expert transplant centres is needed in this complex decision-making process.

The beneficial effect of cardiopulmonary rehabilitation has been proven in HF patients with acquired heart disease. A minimum of physical activity or recreational sports is also recommended in stable HF patients. Although there are no specific studies that investigated the impact of exercise in ACHD patients with HF, the position paper of the ESC Working Group on Adult Congenital Heart Disease and the Section of Sports Cardiology promote at least minimal physical activity for all ACHD patients, including those with stable HF.³⁵ If the type and intensity of exercise are aligned with the haemodynamic and electrophysiological status of the patient, it is assumed that exercising can be done in a safe manner. A group from Rotterdam followed up a cohort of ACHD patients for 10 years and concluded that a required minimum of exercise is not related to impaired outcome.³⁶ Moreover, the risk of sudden death during exercise is low for the overall ACHD population. Sudden cardiac events are very rare, occurring in only 10% of patients, during exercise.³⁷ Obviously, when new symptoms occur, patients need to be re-evaluated, and the individualized exercise prescription adapted accordingly.³⁵

Approximately 7% of all hospital admissions of patients with ACHD are related to HF.³⁸ With an ageing population, it is expected that this number will increase further over time.

Van De Bruaene et al. have proposed treatment algorithms for acute subaortic and acute subpulmonary ventricular failure.³⁴ Both algorithms distinguish between the presence and absence of shunt lesions. Further differentiation in shunt lesions is made between patients with high pulmonary vascular resistance (with a predominant right-to-left shunt) and those with low pulmonary vascular resistance (with a predominant left-to-right shunt).

Oxygen has to be given to all hypoxic HF patients, including those with ACHD. In patients with a right-to-left shunt, it might be that the systemic saturation does not increase much because of the extra-pulmonary shunt; this patient group is not influenced by oxygen therapy. Similar to acute HF in acquired heart disease, oxygen is not indicated in patients with a failing systemic ventricle. However, regardless of the systemic saturation, oxygen therapy has to be instituted in ACHD patients with a failing subpulmonary ventricle, as oxygen lowers the pulmonary vascular resistance and reduces afterload.

Use of diuretics—both loop diuretics and, in some cases, in combination with thiazides—is very important in ACHD patients with failing systemic and/or subpulmonary ventricles. However, this population also has to deal with the typical heart–kidney balance: the lower the circulating volume, the higher the risk of insufficient renal perfusion and renal function impairment. If requested, dialysis should be commenced. Focus on this heart–kidney balance is even more important in patients with a Fontan circulation in whom a minimum volume load is required as the driving force for circulation. Too high doses of diuretics might significantly decrease the single-ventricle cardiac output.³⁹

Similar to treatment of patients with acquired HF, vasodilators play a major role in treatment of a failing systemic ventricle. The benefit of using vasodilators in a failing subpulmonary ventricle or a single ventricle is not proven. As in chronic heart failure, vasodilators might aggravate a right-to-left shunt, resulting in further systemic desaturation.

Vasopressors, such as noradrenaline, are also indicated in some specific ACHD patients who present with acute HF. Severe hypotension can lead to impaired perfusion of the coronary arteries; vasopressors help to redistribute blood from the extremities to the vital organs. In patients with right-to-left shunting, vasopressors are used to increase the systemic vascular resistance, in general more so than the pulmonary vascular resistance. As a consequence, the right-to-left shunt improves, as well as the systemic saturation. Low to moderate doses of noradrenaline are most effective in controlling the balance between the systemic and pulmonary vascular resistance.

Inotropes are also beneficial. Besides the general use of milrinone in the post-operative intensive care setting, it increases the contractility of the volume-loaded subpulmonary RV and reduces the systemic and pulmonary vascular resistance in ACHD patients. Low-dose dobutamine decreases the systemic and pulmonary vascular resistance, in addition to improving myocardial dysfunction. High-dose dobutamine increases the pulmonary vascular resistance and can have an adverse effect on afterload of the subpulmonary ventricle. Finally, levosimendan, a calcium sensitizer, has both a positive inotropic effect and a vasodilatory effect. In patients with severe low-output HF, levosimendan seems to be more efficient than dobutamine.⁴⁰ ACHD patients with end-stage HF can be treated with levosimendan as outpatients on a regular basis.⁴¹

Selective pulmonary vasodilators are specifically used in patients with high pulmonary vascular resistance. Inhaled nitric oxide and inhaled iloprost, a prostacyclin analogue, are the most commonly used. When inhaled, they have no effect on the systemic vascular resistance, but only on the pulmonary vascular resistance. Intravenous pulmonary vasodilators, such as prostacyclin, can cause systemic hypotension and, if possible, should be avoided. Moreover, as a precaution, with respect to all intravenous therapy in ACHD patients with shunt lesions, paradoxical air and/or thrombus embolism should be avoided.

Depending on the severity of their symptoms, ACHD patients with HF face limitations in their daily life. These limitations not only involve reduced physical capacity, but also impact the patients’ mental health and social life. Such limitations and the need for hospitalizations can negatively affect the quality of life. In addition to systematic medical follow-up, psychological support is often required. Symptomatic relief and mitigating the need for regular hospitalizations could help stabilize patients’ mental health. Psychologists and nurse practitioners play an important role in achieving these goals. CHD patients are considered to have a chronic disease condition, particularly those with HF. It is advocated that these patients are offered advanced care management, initiated during the transition period from paediatric CHD to ACHD. Where at the beginning this advanced care management still may focus on reparative interventions, it is possible that, after a while, a palliative approach is needed and a dialogue about end-of-life care is initiated.^42,43 This is a rather difficult process that can be organized by a palliative care support team with sufficient expertise in this matter. Some of these approaches have been described for HF patients with acquired heart disease and could be applied to the ACHD population.^42,43 However, there are no clinical data that justify implementing this approach without the requested critical background.

The heterogeneity of the ACHD population means that only one definition of HF for all types of ACHD will not suffice. One does not fit all. The scientific community has to continue to search for definitions adapted to specific pathologies, so these disease-oriented definitions can be used for diagnosis, therapy, and evaluation of treatment effect.

In many cases, the original cause of HF is known. Early structural optimization, modified surgical approaches, and more appropriate interventional procedures may contribute to a lower risk of developing HF later in life. Continuing feedback from the adult programme to congenital surgeons, paediatric cardiologists, and all health-care professionals involved helps in the process for treatment optimalization.

As in many disease processes, there would be added value in being able to predict adverse cardiovascular events before obvious structural, functional, or biochemical abnormalities become apparent. A combination of minimal changes in variables could provide the opportunity to predict adverse cardiovascular events and consequently anticipate them in a timely manner.⁴⁴ Early therapeutic strategies based on data obtained by deep phenotyping is one of future possibilities, but the development is still in its infancy.

RCTs represent the only way to improve the level of evidence of ACHD therapy. While RCTs usually include relatively homogeneous patient groups, this is not always the case in daily clinical practice. Patient characteristics often differ from those of the original study patients. This is especially true for the ACHD patient group where there is pronounced heterogeneity. Despite the fact that RCTs represent the only way to achieve a higher level of evidence, it would be very difficult in practice to set up RCTs in the ACHD population. To recruit enough patients and achieve the necessary statistical power, multicentre collaboration is needed. The will is certainly not lacking among the ACHD community, but the road to its achievement is difficult (logistically and financially).

One option to address the problems associated with RCTs is to analyse large data sets with use of artificial intelligence. Hospitals, governments, and scientific organizations have large databases, longitudinal in time. A combination of these databases can simulate a patient’s digital life course via deep learning.⁴⁵ Using the principle of evidence-based medicine, a prediction model can then be created for the real patient, with particular attention to the therapeutic approach and outcome. This too is still in its infancy, but it will be an important step towards personalized medicine. Especially in a heterogeneous group of patients, this approach will have a clear added value.

HF is a common problem in ACHD patients, and with ageing of the population and the rise in the number of complex cases, the incidence will increase further. In addition to clinical signs and symptoms, significant impairment in exercise performance and a clear increase in circulating biomarker levels make a diagnosis of HF more likely, although the diagnosis remains challenging in this specific group. The heterogeneity of the ACHD patient population hinders the ‘blind’ application of HF in acquired heart disease guidelines in CHD patients. Moreover, even drug trials in more homogeneous groups of ACHD failed to show any positive effect on clinical outcome. Only surrogate endpoints were achieved. The more complex the CHD, the more the treating physician has to be aware of adverse effects of the interventions used. In-depth knowledge of the underlying disease is of utmost importance, and if lacking, a referral to, or at least advice from, an expert centre is required. In addition to drugs, device therapy and transplantation can both still offer alternative, albeit temporary, treatment solutions. Unfortunately, for some patients who end up on a palliative trajectory, end-of-life care management should be offered. Many aspects discussed in this chapter are based on small studies and expert opinion. Large multicentre trials are warranted to answer the remaining questions.