The Fontan circulation is a palliative procedure performed in patients born with a functionally single-ventricular heart. The principle of the Fontan operation is based on redirecting systemic venous return directly through the pulmonary circulation. This means that there is no ventricular chamber that is directly involved in sustaining the pulmonary circulation.Reference Gewillig 1 It has been well documented that in the long term there is a continuing risk of failure of the Fontan circulation;Reference Driscoll, Offord, Feldt, Schaff, Puga and Danielson 2 , Reference Gentles, Gauvreau and Mayer 3 30% of all Fontan patients have late failure of the Fontan circulation at ~20 years of follow-up.Reference d’Udekem, Iyengar and Galati 4
Problems related to the Fontan operation are impaired ventricular function, increased pulmonary vascular resistance, cardiac arrhythmias, thrombosis, ascites, peripheral oedema, cyanosis, lymphatic dysfunction with protein-losing enteropathy, and plastic bronchitis.Reference Deal and Jacobs 5 For the purposes of this review, the term failing Fontan is used in case of reduced pulmonary blood flow, ventricular dysfunction, protein-losing enteropathy, or plastic bronchitis.
Failing of the Fontan circulation has been associated with a progressive decline in exercise capacity.Reference Giardini, Hager, Pace Napoleone and Picchio 6 The inability to increase pre-load and decrease after-load during exercise combined with abnormal ventricular function may lead to reduced cardiac output.Reference Stickland, Welsh and Petersen 7 , Reference Goldberg, Avitabile, McBride and Paridon 8
Fontan patients are also at risk for developing plastic bronchitis and protein-losing enteropathy. Plastic bronchitis leads to the formation of fibrinomucoid casts in the airways, caused by high central venous pressure and pulmonary resistance, and low cardiac output.Reference Caruthers, Kempa and Loo 9 Protein-losing enteropathy is characterised by leakage of proteins into the gastrointestinal tract.Reference Feldt, Driscoll and Offord 10 The mortality is high in Fontan patients with these complications.Reference Gewillig 1 , Reference Feldt, Driscoll and Offord 10
Despite these well-known problems, data to guide the use of medication to prevent Fontan circulation failure are few.Reference Ghanayem, Berger and Tweddell 11 A recent survey in 546 Fontan patients of seven centres reported a wide variation in the use of medication across the participating centres.Reference Anderson, Breitbart and McCrindle 12 This has been confirmed by other studies as well. In addition, the study showed that drug therapy has been used for several clinical indications.
The aim of this systematic review was to provide a systematic overview of studies that have been carried out on drug therapy aimed at prevention of failure of the Fontan circulation.
Methods
This systematic review was carried out according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) statement.Reference Moher, Liberati, Tetzlaff and Altman 13 The Embase database was used to search for relevant articles published up to December, 2014; date of search was 15 January, 2015. We used the following search terms: (“Fontan procedure”/de OR (fontan):ab,ti) AND (“drug therapy”/exp OR “cardiovascular agent”/exp OR (drug* OR pharmac* OR agent* OR medication* OR administration* OR sildenafil* OR bosentan* OR ((“dipeptidyl carboxypeptidase” OR ace OR “angiotensin converting enzyme”) NEAR/3 inhibitor*) OR diuretic* OR Enalapril*):ab,ti) AND (exercise/exp OR “exercise test”/exp OR “heart output”/exp OR (exercise OR ((heart OR cardiac*) NEAR/3 output)):ab,ti) AND [english]/lim NOT ([Conference Abstract]/lim OR [Letter]/lim OR [Note]/lim OR [Conference Paper]/lim OR [Editorial]/lim). An English language restriction was used. We also screened the references lists of review articles and relevant studies to identify other useful articles, but no relevant articles were found.
Resulting publications were assessed independently by two reviewers (N.O. and A.M.). Of the articles found in Embase, titles and abstracts were screened and selected based on the following pre-defined criteria: the study population had to consist of patients with a Fontan circulation, there had to be any kind of drug intervention, efficacy or safety of the medication had to be measured, the drug therapy had to be aimed at decreasing pulmonary resistance, improving exercise performance or ventricular function, or at reducing protein-losing enteropathy, and the article had to be published between 1997 and 2014. As patients who develop plastic bronchitis are generally considered to have developed a failing Fontan circulation and as the primary focus of this article was the prevention of failure of the Fontan circulation, we did not include studies on treatment of plastic bronchitis. If titles and abstracts met the inclusion criteria, the full-text articles were studied. Full-text articles of the selected studies were included if the article still matched the first inclusion criterion, was not a case report, and if the full text was available. In case of disagreement, an agreement was negotiated. The risk of bias within each study was assessed using the Cochrane risk of bias tool.Reference Higgins and Green 14 If the risk of all types of bias was high – selection bias, detection bias, performance bias, attrition bias, and reporting bias – quality was considered insufficient and the article was not included in this review.
The following data were extracted: publication year, study design, medication and dosage, number of included patients, type of Fontan procedure, age at inclusion, and follow-up duration. Outcome measures of interest were as follows: ventricular performance – myocardial performance index, velocity time integral, and heart rate – exercise capacity – respiratory rate, minute ventilation, ratio of ventilator effort over carbon dioxide production, peak oxygen uptake (VO2), pulmonary blood flow index, cardiac index, systemic oxygen saturation, and heart rate – exercise haemodynamics – cardiac index, stroke volume indexed to body surface area, ejection fraction, end-systolic volume indexed to body surface area, systemic vascular resistance index, total pulmonary resistance index, end-systolic volume indexed to body surface area, VO2, and O2 saturation.
Results
The search method as described above produced 162 articles in Embase. Of these articles, 26 met the inclusion criteria, and after screening of the full text a total of 10 articles were included in this review. In these articles, nine study populations were described; one of the populations had been described in two articles, using different outcome measures.Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16 The flowchart of the literature search is given in Figure 1. The study by Tunks et alReference Tunks, Barker and Benjamin 17 was not included in this review (figure: quality of article was not sufficient), due to lack of randomisation (selection bias), knowledge of the allocated interventions by outcome assessors, patients, and personnel (detection and performance bias), and there was incomplete outcome data and selective outcome reporting (attrition and reporting bias). The study characteristics and populations are described in Table 1. Of the included studies, four were cross-over, randomised controlled trials,Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16 , Reference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 – Reference Hebert, Mikkelsen and Thilen 20 two open-label randomised controlled trials,Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 , Reference Schuuring, Vis and van Dijk 22 and three were uncontrolled open-label trials.Reference Hager, Weber, Muller and Hess 23 – Reference Van De Bruaene, La Gerche and Claessen 25 Moreover, two articles by Goldberg et alReference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16 on the use of sildenafil were on the same single trial; one of the articles focussed on the exercise test findingsReference Goldberg, French and McBride 15 and a separate paper evaluated echo findings on ventricular performance.Reference Goldberg, French and Szwast 16 Ventricular performance was a secondary end point of the original trial.Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16
Figure 1 Flow chart of article selection process.
Table 1 Overview of included studies.
APC=right atrium-to-pulmonary artery connection; AVC=right atrium-to-right ventricular connection; BNP=brain natriuretic peptide; CI=cardiac index; CO=cardiac output; CPET=cardiopulmonary exercise test; ECC=extracardiac conduit; EDVi=end-diastolic volume index; EF=ejection fraction; ESVi=end-systolic volume index; HR=heart rate; ICC=intracardiac conduit; IQR=interquartile range; LT=lateral tunnel; MPI=myocardial performance index; PBFi=pulmonary blood flow index; PVR=pulmonary vascular resistance; RCT= randomised controlled trials; SaO2=systemic oxygen saturation; SBP=systolic blood pressure; SVi=stroke volume index; SVRi=systemic vascular resistance index; TCPC=total cavopulmonary connection; TPRi=total pulmonary resistance index; VE/VCO2=ventilatory equivalents for carbon dioxide; VTI=velocity time integral (of left ventricular outflow tract); VTI×HR=a surrogate of cardiac output
* Dropout of three patients
** Measured by MPI
*** Patients received placebo/sildenafil for 6 weeks and were switched after a washout period of another 6 weeks
**** Including control group of nine patients
***** Tested by cardiopulmonary exercise test (CPET/CPX)
****** Exercise tests only
******* Dropout of six patients
******** Dropout of 10 patients due to lack of motivation
********* All patients received 6 months of bosentan, the control group was untreated for the first 3 months
********** Dropout of three patients
*********** Tested by exercise cardiac magnetic resonance
************ Dropout of one patient
Most articles used exercise capacity as the primary outcome measure (n=8).Reference Goldberg, French and McBride 15 , Reference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 – Reference Hager, Weber, Muller and Hess 23 , Reference Van De Bruaene, La Gerche and Claessen 25 The other articles reported ventricular performanceReference Goldberg, French and Szwast 16 or clinical state/O2 saturationReference Ovaert, Thijs and Dewolf 24 as the primary outcome measure.
Follow-up duration varied between 1 day and 6 months. Of the nine studies, only four measured short-term effects of medication before and during exercise tests, and the follow-up duration was 1–2 days.Reference Rhodes, Ubeda-Tikkanen and Clair 19 , Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 , Reference Hager, Weber, Muller and Hess 23 , Reference Van De Bruaene, La Gerche and Claessen 25 The rest of the studies reported a longer follow-up duration, varying between 6 weeks and 6 months.Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16 , Reference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 , Reference Hebert, Mikkelsen and Thilen 20 , Reference Schuuring, Vis and van Dijk 22 , Reference Ovaert, Thijs and Dewolf 24
In eight articles, children and young adults were studied,Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16 , Reference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 – Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 , Reference Ovaert, Thijs and Dewolf 24 , Reference Van De Bruaene, La Gerche and Claessen 25 and in two articles only adults were studied.Reference Schuuring, Vis and van Dijk 22 , Reference Hager, Weber, Muller and Hess 23 The age of patients in the studies ranged from 1.2 to 55 years. A total of 267 Fontan patients participated in the studies. The number of patients per study ranged from 10 to 75, with a median of 27. After correction for dropouts, 244 patients completed the studies. Most patients included in the studies reviewed had been operated according to current surgical strategies – that is, intralateral tunnel or extracardiac conduit type of Fontan operation (n=153).Reference Rhodes, Ubeda-Tikkanen and Clair 19 – Reference Van De Bruaene, La Gerche and Claessen 25 Of the studies included in this review, five also reported older techniques such as an atriopulmonary connection or atrioventricular connection.Reference Rhodes, Ubeda-Tikkanen and Clair 19 – Reference Hager, Weber, Muller and Hess 23 Only one study included patients with an atriopulmonary connection or atrioventricular connection,Reference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 and in another study the type of Fontan connection was not mentioned.Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16
All the articles reported dropout rates. In three studies, all of the included patients completed the study.Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 , Reference Hager, Weber, Muller and Hess 23 , Reference Van De Bruaene, La Gerche and Claessen 25 Overall, the dropout range was 0–10 patients with a median of two patients.
The results and conclusions of all nine studies (10 articles) are also shown in Table 1. Of these nine studies (10 articles), four studies evaluated the effect of sildenafil. The effect was assessed by changes in exercise capacity,Reference Goldberg, French and McBride 15 , Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 , Reference Hager, Weber, Muller and Hess 23 exercise haemodynamics,Reference Van De Bruaene, La Gerche and Claessen 25 or changes in ventricular function as measured by myocardial performance index.Reference Goldberg, French and Szwast 16 In three studies, exercise tests were performed 1 day after completion of drug use for the pre-defined amount of time, without further follow-up.Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 , Reference Hager, Weber, Muller and Hess 23 , Reference Van De Bruaene, La Gerche and Claessen 25 In the other study, a follow-up duration of 6 weeks was obtained.Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16
A significant improvement after the use of sildenafil was noted for exercise capacity in two studies, for exercise haemodynamics in one, and for ventricular function in another. The improvements were explained by an increase in cardiac index, maximal oxygen uptake, ejection fraction, and/or pulmonary blood flow.
Goldberg et al did not report an improvement in maximal oxygen uptake, the primary outcome measure of their study; however, there was an increased ventilatory efficiency during peak and submaximal exercise and a suggestion of improved O2 consumption in two subgroups during submaximal exercise, in those with single left ventricular or mixed ventricular morphology and in those with brain natriuretic peptide levels ⩾100 pg/ml.
In two of the four studies on sildenafil use, mild side-effects were reported. These included flushing and mild headache.Reference Goldberg, French and McBride 15 , Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 Sildenafil was well tolerated in most patients.
A single study looked at the effects of inhaled iloprost.Reference Rhodes, Ubeda-Tikkanen and Clair 19 This was a cross-over, randomised controlled trial with 18 patients. The primary outcome measure was exercise capacity, and follow-up was limited to the 1-month period between exercise tests. The results of this study showed that iloprost increased O2 pulse at peak exercise with 1.2 ml/beat, and that VO2 max was increased in all patients who had a baseline peak VO2 of <30 ml/kg/minute. Side-effects related to the inhalation of iloprost, including mild throat or chest discomfort, were reported by 10 of the 18 patients.
Bosentan use was evaluated in three studies. Outcome was measured by exercise capacityReference Hebert, Mikkelsen and Thilen 20 , Reference Schuuring, Vis and van Dijk 22 or clinical state.Reference Ovaert, Thijs and Dewolf 24 These three studies were randomised controlled trials, in which the treatment group received bosentan for 14 weeks to 6 months, respectively. In the study by Schuuring et al, the control group only received bosentan for 6 months after the first 3 months. All studies used a relatively long follow-up period. The largest number of patients was reported by Hebert et al; 69 of their patients completed the trial. In this treatment with endothelin receptor antagonist in fontan patients, a randomized, placebo-controlled, double-blind study measuring peak oxygen consumption (TEMPO) trial, a small increase in exercise capacity was noted.Reference Hebert, Mikkelsen and Thilen 20 Schuuring et al and Ovaert et al concluded that bosentan had no significant effect on either outcome measure. Side-effects reported in these three studies included severe headache, flushing, dyspnoea, chest pain, anaemia and leucocytopaenia, peripheral oedema, and mild transient fatigue.
The study by Kouatli et alReference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 was the only study that analysed the effects of an angiotensin-converting enzyme inhibitor – enalapril – on exercise capacity in Fontan patients; 10 weeks of enalapril treatment was compared with 10 weeks of treatment with placebo in this cross-over, randomised controlled trial. After 20 weeks, no differences in blood pressures, heart rate, respiratory rate, maximal oxygen uptake, or O2 saturations were detected. In contrast, a decrease in the cardiac index increase from rest to peak exercise was noted after enalapril therapy (125% after placebo versus 102% after enalapril). From this study, it was concluded that enalapril does not alter exercise capacity, diastolic function, or any of the baseline haemodynamics. There were no differences in reported side-effects between enalapril use and placebo.
The Cochrane risk of bias tool was used to assess the quality of the included studies (Appendix). There was a low risk of bias in two studies.Reference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 , Reference Hebert, Mikkelsen and Thilen 20 There was a substantial risk of performance and detection bias in five studies, due to lack of blinding.Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 – Reference Van De Bruaene, La Gerche and Claessen 25 In two studies, the method used to conceal the allocation was not described.Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16 , Reference Rhodes, Ubeda-Tikkanen and Clair 19
Discussion
This systematic review provides an overview of studies that have been carried out on drug therapies to prevent failure of the Fontan circulation. We found that sildenafil and iloprost improve exercise capacity, mainly in short-term studies.Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16 , Reference Rhodes, Ubeda-Tikkanen and Clair 19 , Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 , Reference Hager, Weber, Muller and Hess 23 , Reference Van De Bruaene, La Gerche and Claessen 25 In one study with 14 weeks of follow-up, bosentan showed a significant effect to improve exercise capacity.Reference Hebert, Mikkelsen and Thilen 20 This is in contrast with other bosentan studies and with enalapril that did not affect exercise capacity.Reference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 , Reference Schuuring, Vis and van Dijk 22 , Reference Ovaert, Thijs and Dewolf 24 Exercise capacity has been recognised as a key factor in survival of Fontan patients.Reference Kempny, Dimopoulos and Uebing 26 In a recent study in 321 Fontan patients, an association between exercise capacity and morbidity in Fontan patients has been reported.Reference Diller, Giardini and Dimopoulos 27 Patients with poor exercise capacity had an increased risk of cardiac-related hospitalisation. Therefore, it seems essential to prevent exercise intolerance in Fontan patients. The normal response to exercise in a biventricular circulation includes a decrease of up to 50% of pulmonary vascular resistance compared with the value at rest, an increase of ventricular end-diastolic volume, and a decrease of end-systolic volume, resulting in an increase of cardiac output.Reference Argiento, Chesler and Mule 28 Fontan patients are unable to decrease pulmonary vascular resistance and increase ventricular pre-load, which limits the exercise capacity.Reference Goldberg, Avitabile, McBride and Paridon 8 Furthermore, 47% of the Fontan patients have elevated pulmonary vascular resistance at ~9 years after the Fontan operation.Reference Khambadkone, Li, de Leval, Cullen, Deanfield and Redington 29 In Fontan patients, pharmacological vasodilators might decrease pulmonary vascular resistance and subsequently increase pulmonary blood flow to sustain cardiac output during exercise.
Our search resulted in the identification of four studies that observed the effect of sildenafil, a phosphodiesterase inhibitor, on exercise capacity, exercise haemodynamics, or ventricular performance.Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16 , Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 , Reference Hager, Weber, Muller and Hess 23 , Reference Van De Bruaene, La Gerche and Claessen 25 , Reference Nemoto, Sasaki and Ozawa 30 Of the three studies that measured exercise capacity, two showed an increased exercise capacity.Reference Goldberg, French and McBride 15 , Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 , Reference Hager, Weber, Muller and Hess 23 Van de Bruaene et alReference Van De Bruaene, La Gerche and Claessen 25 demonstrated an improvement in exercise haemodynamics, and Goldberg et alReference Goldberg, French and Szwast 16 noted improvement in ventricular function. Sildenafil decreases pulmonary vascular resistance, which might improve ventricular pre-load in the Fontan circulation. Furthermore, it may decrease systemic vascular resistance. This might explain the improvement of exercise capacity in the study by Hager et alReference Hager, Weber, Muller and Hess 23 and Giardini et al;Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 however, these assumptions have not been supported for all aspects by the results of the elegant invasive study by Van de Bruaene et al. They demonstrated a decrease in pre-load with exercise, which was unaffected by sildenafil, despite a decrease in pulmonary resistance.Reference Van De Bruaene, La Gerche and Claessen 25 At the same time, sildenafil resulted in a further decrease of end-systolic volume during exercise, resulting in a net increase of exercise stroke volume. This may relate to the decrease of systemic vascular resistance by sildenafil. In contrast to the results of Hager et alReference Hager, Weber, Muller and Hess 23 and Giardini et al,Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 Goldberg et alReference Goldberg, French and McBride 15 observed no effect on exercise capacity after sildenafil treatment. This difference in results could be explained by the differences in baseline conditions of the patients, the dose of sildenafil, and follow-up duration.Reference Hager, Weber, Muller and Hess 23 Hager et al hypothesised that patients with a poor baseline condition profit more from sildenafil. Hager et al and Giardini et al included patients with a poorer baseline condition than Goldberg et al. Furthermore, Goldberg et al used lower doses of sildenafil than Hager et al and Giardini et al. Sildenafil has a short half-life time, which could possibly result in a lower sildenafil serum concentration in the patients of Goldberg et al during the exercise tests. In addition, both the studies by Hager et al and Giardini et al were not blinded, in contrast to the study by Goldberg et al. Finally, Goldberg et al had a longer follow-up duration than Hager et al and Giardini et al.
Further, two smaller studies measured exercise capacity after intake of bosentan, an endothelin receptor antagonist, and found no significant improvement in exercise capacity.Reference Schuuring, Vis and van Dijk 22 , Reference Ovaert, Thijs and Dewolf 24 Ovaert et alReference Ovaert, Thijs and Dewolf 24 only showed an increase in exercise capacity in patients with limited exercise capacity. Therefore, the authors hypothesised that only patients with an elevated pulmonary vascular resistance would benefit from bosentan. This in contrast to the results of Schuuring et al,Reference Schuuring, Vis and van Dijk 22 who reported no increase in exercise capacity, although all patients had limited baseline exercise capacity. These authors suggested that endothelin-1 levels in Fontan patients are not elevated enough to demonstrate improved exercise capacity, in contrast to the situation in patients with pulmonary arterial hypertension, for whom bosentan was originally developed.Reference Duffels, van der Plas and Surie 31 In the largest study in the field so far, Hebert et al used a randomised controlled design to demonstrate the superiority of bosentan over placebo to improve VO2, exercise time, and reduce pro-brain natriuretic peptide levels, without serious adverse effects.
The studies analysed in this review suggest that bosentan, sildenafil, and iloprost are vasodilators that improve exercise capacity in Fontan patients.
With regard to angiotensin-converting enzyme inhibitors, commonly used in heart-failure treatment in acquired heart disease, Kouatli et alReference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 reported no effect of enalapril in exercise capacity, exercise haemodynamics, and diastolic function. It is possible that the study failed to show an effect because the renin–angiotensin system was not activated in the Fontan patients and may not be the optimal target in Fontan patients.Reference Roche and Redington 32 Further explanations were the use of a small number of patients, a low dose of enalapril, and a short follow-up period. In addition, study patients were in relatively good condition at baseline. Further research is clearly needed, particularly as at present a considerable percentage of Fontan patients are on angiotensin-converting enzyme inhibitors, without real evidence supporting their use in these patients.Reference Anderson, Breitbart and McCrindle 12
There are several limitations to this review. Despite the selected time period, well after the introduction of the total cavopulmonary connection, there were six studies that included patients with an atrioventricular or atriopulmonary connection type of Fontan operation,Reference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 – Reference Hager, Weber, Muller and Hess 23 as well as one study that did not mention the Fontan type.Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16 This heterogeneity in the included patients resulted in less comparability of the studies in this review.
As mentioned earlier, the term failing Fontan in this review has been used in case of reduced pulmonary blood flow, ventricular dysfunction, protein-losing enteropathy, or plastic bronchitis. Considering the focus on prevention of Fontan failure, we have excluded one article about plastic bronchitis. We did not include articles about protein-losing enteropathy, as we only found case reports on this patient group. Further randomised controlled studies are needed to evaluate plastic bronchitis and protein-losing enteropathy treatment. In addition, we could identify only one article that studied the effect of angiotensin-converting enzyme inhibitors in Fontan patients.Reference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 Considering the limitations of that study, further research is needed.
Considering the generally small number of patients, heterogeneity in surgical techniques, and drug types used, we did not perform meta-analysis.
Remarkably, several of the included studies did not use a control group; three studies were uncontrolled.Reference Hager, Weber, Muller and Hess 23 – Reference Van De Bruaene, La Gerche and Claessen 25 Many of the studies that assessed exercise capacity, exercise haemodynamics, or ventricular performance had a short follow-up period.Reference Goldberg, French and McBride 15 , Reference Goldberg, French and Szwast 16 , Reference Kouatli, Garcia, Zellers, Weinstein and Mahony 18 , Reference Rhodes, Ubeda-Tikkanen and Clair 19 , Reference Giardini, Balducci, Specchia, Gargiulo, Bonvicini and Picchio 21 , Reference Hager, Weber, Muller and Hess 23 , Reference Van De Bruaene, La Gerche and Claessen 25 Consequently, the long-term effects of sildenafil, iloprost, and enalapril in Fontan patients are still unknown. Furthermore, some studies did not evaluate safety of the drug under study.Reference Hager, Weber, Muller and Hess 23
We conclude that, based on the studies analysed in this review, bosentan, sildenafil, and iloprost may improve exercise capacity at the short term in Fontan patients. Given the small number of studies, the small number of patients, the short follow-up duration, and the lack of control groups in several studies, the results need to be interpreted carefully. More, larger, and longer placebo-controlled studies are needed to evaluate efficacy and safety of drug therapies in the prevention and treatment of failure of the Fontan circulation.
Acknowledgement
None.
Financial Support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Appendix: Assessment of risk of bias of included studies according to the Cochrane risk of bias tool
A: Risk of bias graph for the studies included in this systematic review
B: Risk of bias summary
1=random sequence generation; 2=allocation concealment; 3=blinding of participants and personnel; 4=blinding of outcome assessment; 5=incomplete outcome data; 6=selective reporting; “+”=low risk of bias; “?”=unclear risk of bias; “−”=high risk of bias.
