If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Cardiology, Peter Munk Cardiac Center, University Health Network, Toronto General Hospital, University of Toronto, Toronto, Ontario, Canada
Patients with a Fontan circulation for single-ventricle physiology are at increased risk of developing thromboembolic events. Thromboembolic events can lead to failure of the Fontan circulation, chronic sequelae in case of stroke, and early mortality. Controversies exist regarding the substrates, risk factors, and optimal detection methods for thromboembolic events. Despite the major clinical implications, there is currently no consensus regarding the optimal antithrombotic therapy to prevent or treat thromboembolic events after the Fontan procedure. In this review we aimed to untangle the available literature regarding antithrombotic prophylaxis and treatment for pediatric and adult Fontan patients. A decision-tree algorithm for thromboprophylaxis in Fontan patients is proposed. Additionally, the current state of knowledge is reviewed with respect to the epidemiology, pathophysiology, and detection of thromboembolic events in Fontan patients, and important evidence gaps are highlighted.
Résumé
Les patients ayant une circulation de type Fontan avec une physiologie à ventricule unique ont un risque accru de développer des complications thromboemboliques. Les événements thromboemboliques peuvent entraîner une défaillance de la circulation de Fontan, des séquelles chroniques en cas d'accident vasculaire cérébral et une mortalité précoce. Des controverses existent au sujet des substrats, des facteurs de risque et des méthodes optimales pour la détection des événements thromboemboliques. Malgré les implications cli-niques majeures, il n'existe actuellement aucun consensus concernant un traitement antithrombotique optimal pour prévenir ou traiter les événements thromboemboliques après une procédure de Fontan. Dans cette revue, nous avons cherché à démêler la littérature disponible concernant la prophylaxie et le traitement antithrombotique des patients Fontan, enfants et adultes. Un algorithme d'arbre de décision est proposé dans le cadre de la thromboprophylaxie chez les patients Fontan. En outre, l'état actuel des connaissances est examiné en ce qui concerne l'épidémiologie, la physiopathologie et la détection des événements thromboemboliques chez les patients Fontan, et les lacunes importantes en matière de données probantes sont exposées.
Since it was first reported in 1971, the Fontan procedure has enabled numerous children with single-ventricle physiology to survive into adulthood. Different variants of the original Fontan operation have evolved, all aiming to lead the systemic venous blood flow directly to the lungs, and reserving the functional single ventricle to support the systemic circulation. Although long-term survival after the Fontan procedure has drastically improved, up to 82% at 20 years, the complex hemodynamic challenges lead to an inevitable functional decline over time.
It remains an ongoing challenge to postpone complications, morbidity, and mortality for as long as possible.
One of the major threatening factors in Fontan patients is the occurrence of thromboembolic events (TEs), that are associated with an increased mortality and morbidity. A great deal of uncertainty exists regarding the incidence, risk factors and detection of TEs. Moreover, there is ongoing debate on the best thromboprophylaxis strategy to prevent TEs in patients with a Fontan circulation. In this review, we aim to untangle the issue of antithrombotic prophylaxis and treatment by reviewing the most pertinent literature available, and we propose a decision-tree algorithm for thromboprophylaxis in Fontan patients.
Epidemiology of Thromboembolic Events
The Fontan circulation is known to be associated with an elevated risk of TEs, but the exact extent of the problem remains unclear. Reports on incidence of TEs are complicated by differences in definition of TEs and in methods of detection, as well as by the heterogeneity of the cohorts and limited follow-up time. Several authors have attempted to report time-adjusted risk of TEs after the Fontan operation, although the results are quite divergent. Reported TE rates varied between 0.74% and 5.2% per patient-year, and cumulative TE incidence at 10 years ranged from 2% to 25% (Table 1, Fig. 1).
Assessment of late thromboembolic complications post-Fontan procedure in relation to different antithrombotic regimens: 30-years’ follow-up experience.
Assessment of late thromboembolic complications post-Fontan procedure in relation to different antithrombotic regimens: 30-years’ follow-up experience.
TEs were defined as the composite of left- and right-sided TE, unless stated otherwise. Imaging was either performed routinely, or on clinical suspicion. Cumulative incidence was estimated using Kaplan-Meier analysis.
∗ TE = pulmonary/systemic embolism or stroke.
† TE = Fontan-connection thrombosis, stroke, or systemic embolism.
TEs are subdivided according to location in the circulation. Right-sided TEs are defined as central venous thrombosis, right-sided intracardiac thrombi, or pulmonary emboli. Left-sided TEs can occur as strokes, systemic emboli, or left-sided intracardiac thrombi. Right-sided TEs are more frequent in Fontan patients, and represent 81% of the TEs in a meta-analysis of 1200 patients.
TEs do not always lead to a clear clinical presentation and can therefore be missed, depending on the surveillance protocol in place. The prevalence of clinically silent TEs is reported to be between 14% and 16%.
A multicenter, randomized trial comparing heparin/warfarin and acetylsalicylic acid as primary thromboprophylaxis for 2 years after the Fontan procedure in children.
They are usually noted on transesophageal echocardiography (TEE), advanced imaging, or at autopsy. This was illustrated in a trial on thromboprophylaxis in 111 Fontan patients, in which 72% of the TEs were asymptomatic and detected during protocol assessment with TEE.
A multicenter, randomized trial comparing heparin/warfarin and acetylsalicylic acid as primary thromboprophylaxis for 2 years after the Fontan procedure in children.
In 2 cohorts of adult Fontan patients, the presence of occult pulmonary emboli was detected with a ventilation-perfusion (V/Q) scan in 17% of 30 cases, and evidence of silent stroke on cerebral magnetic resonance imaging (MRI) was detected in 33% of 84 cases.
Different studies have shown a time-related risk of TEs with an initial peak in TE incidence up to 16% in the first 6-12 months after surgery, followed by a plateau phase of low risk (Fig. 2).
Factors associated with thrombotic complications after the Fontan procedure: a secondary analysis of a multicenter, randomized trial of primary thromboprophylaxis for 2 years after the Fontan procedure.
Figure 2Short- and long-term risk of thromboembolic events (TEs) in Fontan patients. (A) Short-term TE risk. Hazard plot of TEs in children post-Fontan (n = 111). (B) Long-term TE risk. Kaplan-Meyer estimates of TE incidence in Fontan patients with atriopulmonary connection (APC), lateral tunnel (LT), or extracardiac conduit (ECC) (not significant). (A) Reproduced from McCrindle et al.
Factors associated with thrombotic complications after the Fontan procedure: a secondary analysis of a multicenter, randomized trial of primary thromboprophylaxis for 2 years after the Fontan procedure.
The effects of TEs in Fontan are often devastating. Intracardiac or conduit thrombosis can lead to local obstruction or embolize into the pulmonary circulation, with possible failure of the Fontan circulation as a consequence.
Multiple factors contribute to the risk of TEs after Fontan, and range from anatomical variations, to intrinsic coagulation abnormalities, all of which are related to Virchow’s triad of thrombogenesis: disturbed blood flow, endothelial injury, and coagulopathy (Fig. 3).
Figure 3Virchow’s triad of thrombogenesis in Fontan. CBP, cardiopulmonary bypass; CO, cardiac output; CVP, central venous pressure.
A key characteristic of the Fontan circulation is the abnormal flow state inherent to the surgical design. The loss of pulsatility and subsequent elevation in venous pressure leads to low blood flow and predisposes to thrombus formation. Stasis of blood occurs especially in intracardiac cul-de-sacs, such as a pulmonary artery stump, and in enlarged atria of those with an atriopulmonary connection (APC), atrial arrhythmias, or atrioventricular valvular regurgitation. This was reflected by an elevated TE risk for those with an APC (hazard ratio [HR], 2.0; P = 0.02), with atrial arrhythmias (HR, 2.3; P < 0.01), or with pulmonary atresia with an intact interventricular septum (HR, 3.6; P = 0.04), reported in 2 studies with 387 and 111 Fontan patients, respectively.
Factors associated with thrombotic complications after the Fontan procedure: a secondary analysis of a multicenter, randomized trial of primary thromboprophylaxis for 2 years after the Fontan procedure.
Further, in Fontan patients, cardiac output is often impaired because of the altered preload and pulmonary vascular resistance. An elevated TE risk was reported in those with a cardiac output below approximately 2.5L/min/m2.
The endothelial function in patients with a Fontan circulation is thought to be impaired, because of chronic hypoxia, surgical manipulation, the presence of artificial intravascular material, and elevated venous pressure. This has been reflected in increased levels of circulating Von Willebrand factor, a relatively specific indicator of endothelial injury, in 23 Fontan patients, compared with healthy control participants (P < 0.01).
The extracardiac Fontan conduits are at risk of development of a neointima inside the conduit, which disrupts the endothelial balance and incites platelet reactivity and thrombus formation.
The presence of minor conduit depositions was reported in 21 of 30 Fontan patients on computed tomography (CT) angiography within 2 years post-surgery in a trial on thromboprophylaxis.
Comparative trial of the use of antiplatelet and oral anticoagulant in thrombosis prophylaxis in patients undergoing total cavopulmonary operation with extracardiac conduit: echocardiographic, tomographic, scintigraphic, clinical and laboratory analysis.
Bovine jugular vein valved xenograft for extracardiac total cavo-pulmonary connection: the risk of thrombosis and the potential liver protection effect.
Increased biomarkers of platelet activation, such as thromboxane B2, soluble P-selectin, and sCD40L have been reported, and might predispose to TE in Fontan patients.
Deranged levels of coagulation proteins, found in Fontan patients, are presumed to be related to the chronic hepatic congestion with accompanying liver dysfunction, chronic consumption, and loss of clotting factors through mechanisms such as protein-losing enteropathy.
Fontan patients have decreased levels of vitamin K-dependent clotting factors II, VII, IX, and X as well as increased antithrombin and tissue factor pathway inhibitor levels, compared with healthy control participants.
Overall, the net effect seems to point toward a more prothrombotic phenotype, but might vary between individuals.
Detection of Thromboembolic Events
The clinical presentation of TEs in Fontan is variable and ranges from acute decompensation of the Fontan circulation, sudden cardiovascular collapse, stroke, respiratory distress, or localized edema, to incidental findings associated with asymptomatic thrombi. The challenge is to detect TEs early and to prevent further complications and sequelae. An example of the multimodality imaging of thrombosis in the Fontan conduit is shown in Figure 4.
Figure 4Multimodality imaging of a thrombus in a Fontan patient. Right atrial thrombus (asterisks) in an adult with tricuspid atresia, hypoplastic right ventricle, and ventricular septal defect, palliated with an atriopulmonary connection type Fontan. (A) Transthoracic echocardiography. (B) Cardiac magnetic resonance imaging with late gadolinium enhancement. (C) Cardiac computed tomography angiography. RA, right atrium; PA, pulmonary artery.
Clinical implication of these findings is complicated by the difference with the generally accepted cutoff level of 0.5-1.0 μg/mL to exclude pulmonary embolism.
Transthoracic echocardiography (TTE) is the most accessible imaging tool for routine patient evaluation. Yet clear imaging of the Fontan circuit is difficult and thrombi can be missed. The superiority of TEE over TTE has been shown in multiple studies, with 20%-85% of the TEE-detected thrombi missed using TTE.
However, TEE is a much more invasive examination, warranting sedation in most of the patients, which limits its application during routine follow-up.
MRI
MRI has become an important tool in the evaluation of patients with congenital heart disease (CHD). Cardiac MRI with late gadolinium enhancement has a high accuracy for the identification of left ventricular thrombi (sensitivity 88%; specificity 99%; gold standard pathological diagnosis) in acquired heart disease.
Yet, costs are high, there is a limited availability, and in children it often requires sedation.
CT angiography
CT angiography offers an attractive alternative, with a high anatomic resolution of the heart, as well as the ability to assess the pulmonary vasculature, provided image quality is adequate. An adjusted technique with dual injection and dual phase acquisition is needed to optimize contrast distribution, which is impaired because of slow flow and aberrant connections in Fontan patients.
However, misinterpretation due to aberrant blood flow can occur, comparable with CT angiography, and experience within the Fontan population is scarce.
To conclude, the American Heart Association recommended annual evaluation with TTE and evaluation with MRI every 2-3 years for detection of Fontan-related complications, or earlier if any clinical signs occur including atrial arrhythmias.
Because of the substantial prevalence and effect of TEs in the Fontan population, one of the main pillars of Fontan care is the prevention of TEs. The consensus is that thromboprophylaxis is indicated, yet the optimal strategy remains a subject of ongoing debate.
Thromboprophylaxis strategies
No thromboprophylaxis
In the earlier days of the Fontan procedure, the experience with several patients who presented large right atrial thrombi in the postoperative phase, incited the appreciation of thromboprophylaxis post surgery.
Patients with all types of Fontan were included, age at index surgery was between 3.4 and 8.5 years, and the average follow-up was 7.1 years. Overall, a TE occurred in 18.6% of patients without thromboprophylaxis, in 8.6% of patients who used acetylsalicylic acid (ASA), and in 9% who used vitamin K antagonists (VKAs). This was translated into a reduction in TE risk with an odds ratio of 0.36 (P < 0.01) for those who were using any thromboprophylaxis compared with none. Comparable findings were reproduced in multiple observational reports.
Antiplatelet therapy
Antiplatelet therapy generally consists of ASA in Fontan care, in a dosage of 1-5 mg/kg or 75-81 mg once daily. ASA is an inhibitor of platelet cyclooxygenase-1, which results in suppression of platelet aggregation, by decreasing thromboxane A2 levels. In Fontan patients ASA can reduce platelet reactivity and the deposition of platelet-fibrin aggregates in the conduit’s neointima.
Total cavopulmonary connection is superior to atriopulmonary connection Fontan in preventing thrombus formation: computer simulation of flow-related blood coagulation.
Resistance to ASA due to increased platelet turnover and platelet activation has been described by several authors. It was shown in 52% of 44 Fontan patients in one study, indicated by a prolonged elevation of thromboxane B2 levels during platelet function testing.
VKAs produce their anticoagulant effect by inhibiting the activation of clotting factors II, VII, IX, and X, and protein C and S. The net effect is that growth of thrombi in situ is prevented, and new clot formation within the circulation is impaired. Dosing is monitored according to the international normalized ratio (INR) with a target of 2-3 for Fontan, and is known to be affected by numerous drug and food interactions.
In a risk analysis of TEs, 15 of 54 children (28%) with Fontan who were receiving VKAs were in a subtherapeutic range for > 70% of the time, associated with an increased risk of TEs (HR, 3.5; P = 0.01), compared with those receiving either ASA or VKAs within the therapeutic range.
Factors associated with thrombotic complications after the Fontan procedure: a secondary analysis of a multicenter, randomized trial of primary thromboprophylaxis for 2 years after the Fontan procedure.
Bleeding is the most important adverse event (see the section on ASA vs VKAs). Other considerations are the teratogenicity during pregnancy and low bone mineral density associated with long-term usage.
Management of pregnancy in patients with complex congenital heart disease: a scientific statement for healthcare professionals from the American Heart Association.
The non-VKA oral anticoagulants (NOACs), with a significantly broader therapeutic window and fewer interactions, are an interesting alternative to VKAs. However, experience with NOACs in CHD patients is still limited, and to date they are not licensed for Fontan patients. NOACs inhibit a single coagulation protease, either factor Xa or thrombin, through reversible, competitive binding. This results in rapid restoration of coagulation when the drug level decreases and so compliance is highly important. Of note, the effect of the NOAC can be overcome as soon as the amount of natural substrate increases, such as in prothrombotic conditions.
This concept of diminishing anticoagulant effects in increasing procoagulant circumstances raises a concern in Fontan patients, considered to be in a chronic hypercoagulable state. Moreover, drug metabolism of NOACs might be impaired in the case of hepatic dysfunction in Fontan patients. Experience with NOACs in Fontan patients is discussed below in the section on VKAs vs NOACs. As with VKAs, the main adverse event of NOACs is the associated bleeding (see VKAs vs NOACs). Other side effects are gastrointestinal disorders and in rare cases, severe liver injury.
Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials.
Teratogenicity of NOACs is currently unknown and experience in children is limited.
Heparin
The main anticoagulant action of heparin is mediated by the inactivation of thrombin and factor Xa, thereby preventing fibrin formation and inhibiting thrombin-induced activation of platelets and coagulation factors.
When anticoagulation is indicated in Fontan patients, heparin can be preferred over VKAs or NOACs in case of pregnancy, an increased bleeding risk, or in some cases for children to avoid burdensome venous sampling.
Management of pregnancy in patients with complex congenital heart disease: a scientific statement for healthcare professionals from the American Heart Association.
A multicenter, randomized trial comparing heparin/warfarin and acetylsalicylic acid as primary thromboprophylaxis for 2 years after the Fontan procedure in children.
Heparin is available as unfractionated heparin (UFH) and low molecular-weight heparin (LMWH), the latter being more stable and suitable for ambulatory usage.
The most serious complication is heparin-induced thrombocytopenia, an acute immune-mediated hypercoagulable state with arterial and venous thrombosis. It occurs in approximately 2% of adults, and in < 1% of children treated with UFH.
Many studies have attempted to provide an answer to the clinical issue of thromboprophylaxis in Fontan patients. Superiority of either anticoagulation or antiplatelet therapy has nonetheless not been proven, and remains one of the most debated subjects within the field. An overview of comparative studies reported in this review is shown in Supplemental Table S1.
ASA vs VKAs
To date, 2 randomized controlled trials (RCTs) are published that compared VKAs with ASA, both in children. The first, published in 2011, included 111 children who underwent Fontan surgery with a total cavopulmonary connection at a mean age of 4.8 years, who were randomized to receive either VKAs or ASA post surgery.
A multicenter, randomized trial comparing heparin/warfarin and acetylsalicylic acid as primary thromboprophylaxis for 2 years after the Fontan procedure in children.
The TE incidence at 2 years was 24% for VKAs and 14% for ASA, but this difference was not significant (P = 0.45). Bleeding was more common in those taking VKAs compared with ASA, with 33% and 14% minor bleeding, respectively (P = 0.03), and 1 major gastrointestinal bleed associated with an INR of 12.
The second RCT included 30 patients who underwent conversion to Fontan with an extracardiac conduit, and reported 7 TEs in the group receiving ASA (47%) during 2 years of follow-up, compared with 3 in the VKA group (20%), which did not reach statistical significance (P = 0.12).
Comparative trial of the use of antiplatelet and oral anticoagulant in thrombosis prophylaxis in patients undergoing total cavopulmonary operation with extracardiac conduit: echocardiographic, tomographic, scintigraphic, clinical and laboratory analysis.
Age at Fontan conversion varied from 3.4 to 8.5 years and the follow-up period was 7 years on average. The period prevalence of TE was similar for patients receiving ASA (8.6%), compared with those receiving VKAs (9%). Bleeding was not reported in most of the included studies. Because of the retrospective nature of most included studies, important selection bias complicated the extrapolation of the results to clinical practice. Some of the reasons that VKAs were prescribed included arrhythmias, fenestrations, a history of TE, suboptimal hemodynamics, or a prosthetic valve.
As for adult Fontan patients, many have acquired an indication for formal anticoagulation, such as atrial arrhythmia, which eliminates the option of ASA. For the remainder, observational studies report a tendency toward superiority of anticoagulation over antiplatelets for all adult Fontan patients, but the results remain inconclusive.
Bleeding complications are generally more significant in those taking VKAs compared with ASA. A study of 431 Fontan patients with a median age of 2.4 years reported a rate of clinically relevant bleeding for those taking VKAs of 0.82% per patient-year, compared with 0.47% per patient-year for those taking ASA (P < 0.05).
Assessment of late thromboembolic complications post-Fontan procedure in relation to different antithrombotic regimens: 30-years’ follow-up experience.
Rivaroxaban dosage was adjusted using a physiologically based pharmacokinetic model, to obtain a drug exposure equivalent to 10 mg once daily in adults, and was available as an oral suspension.
Dosing regimen prediction and confirmation with rivaroxaban for thromboprophylaxis in children after the Fontan procedure: insights from the phase III UNIVERSE study.
During a follow-up time of 12 months, which did not include routine imaging, 1 TE was reported in the rivaroxaban group, as opposed to 3 TEs in the ASA group (2% vs 9%; P = 0.095), yet the study was not powered for efficacy. Clinically relevant bleeding incidence was similar in both groups (8% vs 9%) with 1 major bleeding (epistaxis) occurring in a patient taking rivaroxaban. Hence, the safety profile of rivaroxaban seemed comparable with ASA. Further studies are needed to confirm these findings.
VKAs vs NOACs
Seventy-four adult Fontan patients (mean age, 32 years) using NOACs were studied in a multinational registry, with a TE rate of 2.9% per patient-year and a similar rate for major bleeding, which seemed comparable with event rates for those taking VKAs from historical data (Fig. 5).
A retrospective study of adult Fontan patients (mean age, 27 years) reported no TEs and an event rate of 0.6% per patient-year for major bleeding in 36 patients receiving NOACs, compared with 1.8% for TE and 1.9% for major bleeding in 42 patients taking VKAs (P = not significant).
Figure 5Event-free survival of adult Fontan patients receiving non-vitamin K antagonist oral anticoagulants. Kaplan-Meier estimates of event-free survival for thromboembolic events, major bleeding, and minor bleeding in adult Fontan patients.
However, good-quality data for the efficacy and safety of NOACs in patients with Fontan are limited, and larger prospective studies with longer follow-up are warranted to support definitive recommendations.
Phased approach in thromboprophylaxis
There are reasons to consider a phased approach in thromboprophylaxis strategy, by adjusting it during the lifespan of a Fontan patient, in correspondence to the varying risk over time.
The first days to months directly after Fontan conversion are considered to be the most vulnerable period in terms of TEs because of the composite of endothelial damage, change in hemodynamics, and loss of coagulant proteins during and directly after the surgery. The importance of early institution of thromboprophylaxis was emphasized in a study focused on the first 30 days post Fontan in 192 children (median age, 3 years), which reported an odds risk of 45 if thromboprophylaxis was not used (P < 0.01). No difference in TE risk was reported for either ASA or VKAs, and bleeding was not reported.
However, in a meta-analysis of 1075 Fontan patients, most (80%) started VKA treatment during the first postoperative period of 3-12 months, which mainly reflects clinical practice.
Assessment of late thromboembolic complications post-Fontan procedure in relation to different antithrombotic regimens: 30-years’ follow-up experience.
Factors associated with thrombotic complications after the Fontan procedure: a secondary analysis of a multicenter, randomized trial of primary thromboprophylaxis for 2 years after the Fontan procedure.
A retrospective study on late TEs, beyond 1 year after Fontan surgery, showed no difference in efficacy of either VKAs or ASA in 475 Fontan patients during a median follow-up of 7 years. They did report a higher bleeding risk for VKAs (3.7% vs 0.6%), suggesting a preference for ASA in this period.
In adulthood the TE risk increases again. As was mentioned previously, a potential advantage of anticoagulation over antiplatelet therapy in adults is suggested in various reports, and larger comparative studies are needed to compare VKAs with NOACs.
Consideration of specific risk factors
Atrial arrhythmias
More than half of Fontan patients will develop atrial arrhythmias during their lifetime, and this is recognized to be an important risk factor for TEs.
The general consensus is that anticoagulation should be initiated for primary prevention in atrial arrythmias, for pediatric as well as for adult Fontan patients.
2018 AHA/ACC guideline for the management of adults with congenital heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.
Clinicians generally maintain a lower threshold to anticoagulate Fontan patients with an APC, a Y-graft extracardiac conduit, or a Kawashima Fontan conduit for azygos continuation, as well as those with dilated lateral tunnels or a pulmonary artery stump, all of which seem to be associated with a higher likelihood of TEs.
Fenestrations, applied to the Fontan circulation for hemodynamic advantages, have the potential of paradoxical embolism through right-to-left shunting. One study of 431 patients reported an odds ratio of 9.7 (P < 0.01) for stroke in case of fenestrations, although there were few events.
Assessment of late thromboembolic complications post-Fontan procedure in relation to different antithrombotic regimens: 30-years’ follow-up experience.
Choice of thromboprophylaxis in fenestrated Fontan patients varies among centres. In case of significant right-to-left shunting, either through fenestrations or through other shunts, there might be a reason to start anticoagulation.
Systemic ventricular dysfunction in Fontan patients might lead to, or is a sign of circulatory decline with associated complications such as arrythmias, hypoxia, heart failure, and protein-losing enteropathy, which can, among others, result in TEs.
Protein-losing enteropathy itself is not identified as an independent risk factor but is often part of the multifactorial dysfunction.
Fontan-related bleeding risk
The derangement of coagulant proteins in the Fontan circulation, can, in certain patients, lead to a more anticoagulant hemostatic profile, correlated with bleeding risk.
Additionally, hemoptysis is a serious complication in Fontan, hypothesized to be secondary to pulmonary arteriovenous malformations and aortopulmonary collateral vessels, arising among others from the bronchial arteries.
The statements of the American Heart Association and the European Society of Cardiology regarding thromboprophylaxis in Fontan recommended VKAs or LMWH for 3-12 months after the Fontan procedure. Long-term prophylaxis during childhood may consist of antiplatelets, with possible intensification to anticoagulation upon adulthood, although no firm recommendation is made. Of note, anticoagulation with VKAs is recommended in those with a mechanical heart valve, known thromboembolism, or atrial arrhythmia.
2018 AHA/ACC guideline for the management of adults with congenital heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.
Other, less established risk factors might justify anticoagulation, such as the presence of a blind ending pulmonary artery, specific types of Fontan such as the APC, Y-graft, or Kawashima conduit, systemic ventricular dysfunction, and significant right-to-left shunting.
Figure 6 shows a suggested decision tree algorithm on the basis of the previously mentioned recommendations.
Figure 6Suggested decision-tree algorithm for thromboprophylaxis post-Fontan repair. ∗According to the guidelines mentioned in this review: in adults with definite risk factors and in all patients with mechanical valves anticoagulation is recommended (level I). In children with any risk factor other than mechanical valve or in adults with only potential risk factors anticoagulation should be considered (level IIa). APC, atriopulmonary connection; ASA, acetyl salicylic acid; LMWH, low molecular-weight heparin; NOAC, non-vitamin K antagonist oral anticoagulant; PA, pulmonary artery; R-L, right-to-left; TE, thromboembolic event; VKA, vitamin K antagonist.
Different management options exist for confirmed TE in Fontan patients, including surgical or percutaneous embolectomy, thrombolytic therapy, anticoagulation, and, in case of failing Fontan, surgical modification, take-down of the Fontan circulation, or heart transplantation.
The choice of a certain therapy depends on the clinical presentation, the type of Fontan, the location and size of the thrombus, and the underlying potentiating factors. The decision should be made in a multidisciplinary setting, and will be different for each patient.
Anticoagulation
Acute treatment
Treatment of TEs often consists of anticoagulation, in the form of heparin (UFH or LMWH), or VKAs either with or without heparin bridging, and necessitates close monitoring for thrombus regression. In a retrospective cohort of 235 adult Fontan patients, anticoagulation with heparin or VKAs was the most frequently chosen treatment for hemodynamically stable patients with an intracardiac thrombus, and clot resolution was achieved in 9/19 (47%).
In another cohort of adult Fontan patients, all 98 patients with TE were treated with VKAs, with additional heparin bridging in 30%, leading to proven thrombus resolution in only 16%.
A recurrent TE occurred in 24%, and 7% needed surgical Fontan revision. As for children, in the previously mentioned RCT of thromboprophylaxis, all 25 patients with TEs were treated with VKAs and 64% received additional heparin initially.
A multicenter, randomized trial comparing heparin/warfarin and acetylsalicylic acid as primary thromboprophylaxis for 2 years after the Fontan procedure in children.
Clot resolution was not reported, but TEs recurred in 28%, leading to 1 death.
Whether a differentiation should be made in terms of treatment between laminated conduit thrombosis, which are generally silent, and clinically overt TEs, remains unclear. Even silent thrombi are thought to be associated with worse prognosis and recurrent TEs in the future, so anticoagulation of these patients seems to be the safest option.
Chronic treatment
As soon as a Fontan patient has experienced a TE, the risk of recurrent TE is reported to be as high as 24-28%.
A multicenter, randomized trial comparing heparin/warfarin and acetylsalicylic acid as primary thromboprophylaxis for 2 years after the Fontan procedure in children.
Investigations should be performed to determine possible predisposing factors, such as atrial arrhythmias, atrioventricular regurgitation, elevated venous or pulmonary pressures, and dilated cardiac compartments. These factors, if present, are rarely transient and so the recurrence risk will remain elevated during the lifetime. The guidelines for adults with CHD therefore recommend to continue the anticoagulation after resolution of the thrombus, for secondary prevention, in general with VKAs.
2018 AHA/ACC guideline for the management of adults with congenital heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.
Thrombolysis has proven its efficacy in the general adult population for hemodynamically unstable pulmonary embolism. For Fontan patients, pulmonary embolism and intracardiac or central venous thrombi can cause obstruction of blood flow with consequent hemodynamic derangement. The experience with thrombolytics in Fontan patients is limited. In a retrospective study of 28 adult Fontan patients with intracardiac thrombi, thrombolytics were administered in 1 hemodynamically unstable patient, in 1 stable patient as initial treatment, and in 2 stable patients because of clot progression with VKA treatment. This led to survival and successful clot resolution in only 1 patient (25%).
In a systematic review, 18 of 67 cases with a TE were treated with thrombolysis, leading to survival and clot resolution in 7 patients (37%). The clinical status of these patients was not reported.
Right-sided thrombi could disseminate into the pulmonary vasculature, and lead to chronically elevated pulmonary vascular resistance in Fontan patients.
The indication for systemic thrombolysis for a TE in Fontan patients is generally limited to those with hemodynamic instability, and warrants a multidisciplinary discussion for each individual case.
A more recently developed technique is catheter-directed thrombolysis, consisting of in situ administration of low-dose thrombolytics in combination with mechanical fragmentation using ultrasound. There is anecdotal experience with this technique in Fontan patients directed toward right atrial thrombi and pulmonary embolism.
Surgical embolectomy for TEs in Fontan patients is considered for those with hemodynamic instability or if the thrombus persists despite anticoagulation. In the previously mentioned retrospective study of adult Fontan patients with TEs, surgical embolectomy was performed in 7 patients, of which 4 patients (57%) survived.
Embolectomy can also be performed percutaneously with a mechanical suction device known as AngioVac (AngioDynamics, Latham, NY), as was shown in an adult Fontan patient with a saddle pulmonary embolism.
As the number of adult patients with a Fontan circulation is steadily growing, pregnancy in these patients is a highly relevant topic. The current European Society of Cardiology guideline classifies all pregnancies in patients with a Fontan circulation as high risk (modified World Health Organization classification 3) or as very high risk (modified World Health Organization classification 4) for those with previous cardiovascular complications.
Management of pregnancy in patients with complex congenital heart disease: a scientific statement for healthcare professionals from the American Heart Association.
Thromboembolism is an important risk in pregnant Fontan patients, because the hypercoagulable profile specific to the Fontan circulation is aggravated during pregnancy, which is itself a prothrombotic state.
However, bleeding risk should be minimized, so advice on the type of thromboprophylaxis during pregnancy, delivery, and the early postpartum period is challenging. A retrospective study on 84 pregnant Fontan patients reported bleeding complications in 35% and TEs in 3%.
Management of pregnancy in patients with complex congenital heart disease: a scientific statement for healthcare professionals from the American Heart Association.
Women with an indication for anticoagulation before pregnancy can be managed with LMWH, which does not cross the placenta, in contrast to VKAs, thereby avoiding the risk of teratogenicity, fetopathy, and fetal bleeding.
Management of pregnancy in patients with complex congenital heart disease: a scientific statement for healthcare professionals from the American Heart Association.
Monitoring of peak anti-Xa levels is recommended in those with mechanical heart valves, and might be considered in Fontan patients without as well. With regard to NOACs, at present time they are contraindicated in pregnancy.
Contraceptives
All Fontan patients of reproductive age should be counselled about the effect of pregnancy, and appropriate contraception. Estrogen-containing contraceptives increase the risk of thrombosis and should be avoided. The desogestrel progestogen-only pill has efficacy similar to the combined oral contraceptive pill and is widely and safely used by women with a Fontan circulation in many parts of the world, including Europe, Australasia, and South America. Other safe contraceptive methods include intrauterine devices or progestogen-eluting subdermal implants.
Fontan patients treated with antithrombotics might undergo planned or unplanned invasive procedures, necessitating interruption of their antithrombotic therapy. In some patients receiving VKAs, bridging with heparin might be needed. In a retrospective study, a selection of 6 of 20 Fontan patients receiving VKAs were bridged with heparin to undergo a planned procedure, and no bleeding or TE was reported. The selection consisted of patients with a history of TEs or a mechanical prosthetic valve.
Surgical bleeding risk and patient-related thrombotic and bleeding risk factors should be considered when deciding on heparin bridging.
Children and effect of anticoagulation
The hemostatic physiology in children is different from adults, including lower concentrations of coagulation proteins, which results in greater fluctuations in INR with VKA treatment.
In combination with the burdensome venous sampling for INR monitoring, this leads to the choice for LMWH in some cases, with infrequent testing of anti-FXa levels. Further, for the growing child, the potential of reduced bone density due to VKAs, and osteoporosis due to heparin, raises a concern.
Finally, patients treated with anticoagulation are recommended against contact sports to prevent bleeding, which might impair quality of life in children and adults.
Gaps in Evidence
A great deal of uncertainty exists regarding TEs after Fontan. The exact incidence of TEs remains unclear because of differing follow-up intervals and variable definitions and screening methods of TEs. Risk factors are difficult to identify because of small numbers of events and heterogeneous cohorts. Furthermore, follow-up studies of Fontan patients reflect results of an earlier surgical era and do not necessarily represent the expectation for Fontan patients who undergo surgery in the current era.
Moreover, there is little agreement regarding the relative efficacy of various thromboprophylaxis strategies. The evidence consists of a few small RCTs and a multitude of inconclusive observational studies. Clinical decisions are therefore often guided by expert opinion. Further prospective trials are needed to identify the most appropriate thromboprophylaxis strategy, taking into account the identification of discriminative risk factors and increasing age of Fontan patients. However, the organization of such a trial is impeded by low patient numbers and difficulties in TE detection.
Future Perspectives
Recently, risk scores have been developed to risk-stratify Fontan on adulthood.
The correlation of this risk score with the risk of TE and possible guidance for the choice in thromboprophylaxis in adulthood is yet to be studied.
As for the role of NOACs in Fontan patients, the recently published RCT on rivaroxaban vs ASA in children showed a reassuring safety profile for rivaroxaban.
Further follow-up data from the multinational Non-vitamin K antagonist Oral anticoagulant for prevention of Thromboembolic Events registry on adult CHD patients receiving NOACs, among whom 92 Fontan patients, are expected.
However, additional high-quality data will be necessary to elucidate the role of NOACs vs VKAs in adults, and to date no definite recommendation can be made.
TEs are important complications after the Fontan procedure and remain a serious cause of morbidity and mortality in the early and late follow-up after surgery. Thrombogenesis in Fontan is promoted through various factors acting on each of the 3 aspects of Virchow’s triad—disturbed blood flow, endothelial injury, and coagulopathy. The effect of each of these factors differs per patient, and their exact clinical implication remains inconclusive. The detection of TEs in Fontan is done primarily via routine TTE, although this modality is known to miss a substantial proportion of asymptomatic TEs. Additionally, TEE, MRI, or CT angiography can be performed in case of suspicion of a TE, as well as at regular follow-up intervals.
As for the prevention of TEs, the superiority of either anticoagulation or antiplatelet therapy has not been proven, and residual risk of TEs remains. VKA therapy is the most studied anticoagulation regimen in Fontan patients, and data on NOACs are still scarce. Because of the higher incidence of TEs early postoperatively, as well as beyond the first decade, a phased approach could be used, starting with anticoagulation for 3-12 months post surgery, followed by antiplatelets during childhood. Patients can return to anticoagulation in adulthood or at appearance of clear indications such as atrial arrhythmias or a TE. Additional reasons to start anticoagulation consist of a multitude of theoretically predisposing factors. One should also take into account the effect on daily life, for instance the requirement of frequent INR monitoring for VKAs in children and the risk of bleeding in the case of a trauma. Other life events such as pregnancy and additional surgery might require changes to thromboprophylaxis strategies.
Different treatment options exist for a confirmed TE in Fontan, ranging from intensification of anticoagulation to thrombolysis or embolectomy. The decision should be made in a multidisciplinary setting, and will be different for each patient, depending on the location and size of the thrombus, and the underlying Fontan physiology.
There is a clear need for prospective studies with a focus on the comparison of different thromboprophylaxis strategies in Fontan patients, as well as on the potential of risk stratification to guide the therapeutic decision.
Acknowledgements
The patient whose images are depicted in Figure 4 deceased before the writing of this review and consent for the reproduction of anonymized data in this review could therefore not be obtained. In life, the patient provided written informed consent to be included in the Concor registry for adults with CHD with the explicit purpose to use patient data for scientific and clinical research.
Funding Sources
None.
Disclosures
The authors have no conflicts of interest to disclose.
Assessment of late thromboembolic complications post-Fontan procedure in relation to different antithrombotic regimens: 30-years’ follow-up experience.
A multicenter, randomized trial comparing heparin/warfarin and acetylsalicylic acid as primary thromboprophylaxis for 2 years after the Fontan procedure in children.
Factors associated with thrombotic complications after the Fontan procedure: a secondary analysis of a multicenter, randomized trial of primary thromboprophylaxis for 2 years after the Fontan procedure.
Comparative trial of the use of antiplatelet and oral anticoagulant in thrombosis prophylaxis in patients undergoing total cavopulmonary operation with extracardiac conduit: echocardiographic, tomographic, scintigraphic, clinical and laboratory analysis.
Bovine jugular vein valved xenograft for extracardiac total cavo-pulmonary connection: the risk of thrombosis and the potential liver protection effect.
Total cavopulmonary connection is superior to atriopulmonary connection Fontan in preventing thrombus formation: computer simulation of flow-related blood coagulation.
Management of pregnancy in patients with complex congenital heart disease: a scientific statement for healthcare professionals from the American Heart Association.
Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta-analysis of randomised trials.
Dosing regimen prediction and confirmation with rivaroxaban for thromboprophylaxis in children after the Fontan procedure: insights from the phase III UNIVERSE study.
2018 AHA/ACC guideline for the management of adults with congenital heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.
2018 AHA/ACC guideline for the management of adults with congenital heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.