Bosentan Improves Exercise Capacity in Adolescents and Adults After Fontan OperationCLINICAL PERSPECTIVE
The TEMPO (Treatment With Endothelin Receptor Antagonist in Fontan Patients, a Randomized, Placebo-Controlled, Double-Blind Study Measuring Peak Oxygen Consumption) Study
Background—The Fontan procedure has improved survival in children with functionally univentricular hearts. With time, however, complications such as reduced exercise capacity are seen more frequently. Exercise intolerance is multifactorial, but pulmonary vascular resistance probably plays a crucial role. Elevated pulmonary vascular resistance has been associated with raised levels of endothelin-1, which are common both before and after Fontan operations. Treatment with endothelin-1 receptor antagonists could theoretically improve cardiopulmonary hemodynamics and exercise capacity. The aim of this study was therefore to examine the efficacy and safety of bosentan in Fontan patients.
Methods and Results—Seventy-five adolescents and adults were randomized 1:1 to 14 weeks of treatment with bosentan or placebo. Cardiopulmonary exercise test, functional class, blood samples, and quality-of-life questionnaires were evaluated at baseline and at the end of treatment. Sixty-nine patients (92%) completed the study. Peak oxygen consumption increased 2.0 mL·kg−1·min−1 (from 28.7 to 30.7 mL·kg−1·min−1) in the bosentan group compared with 0.6 mL·kg−1·min−1 (from 28.4 to 29.0 mL·kg−1·min−1) in the placebo group (P=0.02). Cardiopulmonary exercise test time increased by 0.48 minute (from 6.79 to 7.27 minutes) versus 0.08 minute (from 6.94 to 7.02 minutes; P=0.04). Nine bosentan-treated patients improved 1 functional class, whereas none improved in the placebo group (P=0.0085). Side effects were mild and occurred equally in both groups. No serious adverse effects were seen, and no patients had liver enzyme levels above the 3-fold upper limit.
Conclusions—Bosentan improves exercise capacity, exercise time, and functional class in Fontan patients without serious adverse events or hepatotoxicity.
Introduction of the Fontan operation in 19711 dramatically improved the survival of patients with functionally univentricular hearts, which was otherwise dismal.2 Although the surgical techniques have evolved,3,4 the procedure remains palliative with current 10-year survival rates reaching 90%.5,6 The prevalence of Fontan patients has consequently increased significantly over the past decades,7 and clinicians are now frequently caring for patients who suffer from a number of late comorbidities, including arrhythmias and exercise intolerance. These problems will continue to rise markedly.8,9
Editorial see p 1999
Clinical Perspective on p 2030
In the Fontan circulation, pulmonary blood flow is driven passively without a subpulmonary ventricle. Even a mild increase in pulmonary vascular resistance (PVR) will reduce pulmonary blood flow and thus the systemic ventricular preload, cardiac output, and exercise tolerance.10,11 Keeping PVR as low as possible is therefore mandatory, not only after Fontan completion but also preoperatively when raised pulmonary flow may increase PVR, particularly in susceptible patients. Other important factors, however, are nonmodifiable such as the loss of pulsatile pulmonary blood flow after the Fontan procedure, which has been suggested to increase PVR through vascular remodeling.12 Interestingly, levels of the potent pulmonary vasoconstrictor endothelin-1 (ET-1) are frequently elevated before surgical palliation,13 and raised levels of ET-1 have been shown to correlate with failure of the Fontan circulation.14,15 These observations suggest that pulmonary vasodilators and especially endothelin receptor antagonists (ERAs) may play a role in the treatment of these patients. Few studies have investigated the safety and efficacy of pulmonary vasodilators after the Fontan operation, and only 2 have tested ERAs.16,17 Randomized, placebo-controlled studies with ERAs have not been performed.
In this study, we report the results of the TEMPO (Treatment With Endothelin Receptor Antagonist in Fontan Patients, A Randomized, Placebo-Controlled, Double-Blind Study Measuring Peak Oxygen Consumption) study. The primary objective was to investigate the medium-term effect of treatment with the ERA bosentan on exercise parameters and New York Heart Association (NYHA) functional class. Furthermore, the safety profile of bosentan when administered for 14 weeks was assessed.
This was a randomized, double-blind, placebo-controlled, clinical trial. Subjects were randomized 1:1 via a computerized model to parallel groups of bosentan or placebo. Randomization block size was 40. Participants received oral placebo or bosentan 62.5 mg twice daily for 2 weeks, followed by 125 mg twice daily for 12 weeks when the end of the study was reached. The treatment regimen was therefore similar in terms of dosage to what has been recommended for the treatment of pulmonary arterial hypertension.18 Dosing was halved in patients with a body weight <35 kg. The placebo tablets were identical in appearance to the bosentan tablets. The trial was approved by the Danish Ethics Committee (protocol No. H-3-2010-045) and was in accordance with the Helsinki declaration and International Conference on Harmonization Good Clinical Practice guidelines.
Patients were recruited from a population database of patients with univentricular hearts >12 years of age who had undergone Fontan operation. Clinical stability (no clinical worsening, change in medication, or hospitalization 3 months before randomization) and the ability to perform a bicycle exercise test were required. Patients with other severe cardiac or extracardiac comorbidity, for example, critical heart failure (NYHA class IV), systemic hypotension (<80% of the reference for age), cyanosis (oxygen saturation <85%), neurological sequelae, and liver or renal impairment, were not included. Patients on pulmonary vasodilators or drugs contraindicated during bosentan treatment were excluded. All patients from the 2 Danish centers, Rigshospitalet in Copenhagen and Aarhus University Hospital, who fulfilled the criteria were invited to participate in the study. Additional participants were invited from 2 Swedish centers, the Karolinska University Hospital in Stockholm and Lund University Hospital. Before enrollment, written informed consent was obtained from all participants or from legal guardians for minors (<18 years of age).
The primary end point was peak oxygen consumption (o2) indexed by body weight. End points are listed in Table 1. Cardiopulmonary exercise testing (CPET) was performed, and blood samples, NYHA class, and Short Form-36 (SF-36) quality-of-life questionnaires were obtained at baseline and at the end of the study, all conducted and evaluated by the principal investigator.
Cardiopulmonary Exercise Testing
CPET was performed on a bicycle ergometer (Monark Ergomedic 839E, Monark Exercise AB, Sweden) at the Institute of Sports Medicine, Copenhagen. An appropriate incremental load protocol was chosen for each participant on the basis of expected fitness, with the purpose of reaching maximal effort within 6 to 12 minutes. All started with 1 to 2 minutes warm-up, after which the test protocol started at either 20 W and increased stepwise by 20 W/min or at 25 W and increased by 25 W/min. The same protocol was used for baseline and end-of-study CPET for each patient. ECG and transcutaneous oxygen saturation were measured continuously. The participants wore a mask with a TripleV flowmeter (Erich Jaeger GmbH, Germany) to measure respiratory volumes and a tube for breath-by-breath sampling to measure O2 and CO2 concentrations via Masterscreen CPX (CareFusion, San Diego, CA). All subjects were encouraged to exercise until exhaustion. A respiratory exchange ratio >1.1 was considered an indication that maximal effort was achieved. Measurements were given as the mean over 15-second intervals subtracting the highest and lowest value.
Peak o2 was defined as the 15-second interval with the highest o2. The anaerobic threshold was defined as the first 15-second interval at which the respiratory exchange ratio exceeded 1.0 without subsequently falling below 1.0.19 Maximal workload was the highest watt load achieved for at least 15 seconds.
Adverse Events and Compliance
Each patient had 5 visits: 1 visit at baseline, 1 visit for dose adjustment, 2 intermediate visits, and 1 visit at the end of the study. All tests at baseline and at the end of the study were performed in Copenhagen. Adverse events were recorded, and blood samples, including aminotransferases, hemoglobin, platelets, and packed cell volume, were analyzed at all visits. Drug compliance was evaluated by counting remaining tablets at the end of the study.
To secure acute tolerability, blood pressure, heart rate, and transcutaneous oxygen saturation were measured before and twice (at 1 and 2 hours) after the first drug administration at baseline and at the dose-adjustment visit. A drop in systolic blood pressure of >20% or symptoms such as palpitations or vertigo were considered to reflect significant hypotension and led to exclusion.
Female participants were required to use safe contraception, that is, 2 contraceptive methods simultaneously, including barrier methods, an intrauterine device, or hormonal methods. The use of hormonal contraception was accepted as long as it was dosed to account for the interaction with bosentan. For participants <15 years of age, sexual abstinence was accepted as safe. Serum human chorionic gonadotropin was measured before enrollment and at each subsequent visit to exclude pregnancy.
Hepatotoxicity (aminotransferases >3 times the upper reference limit), bone marrow depression (hemoglobin <6.5 mmol/L or a 40% decrease in platelets), positive human chorionic gonadotropin, desaturation (>10% decrease from baseline value), or hypotension led to immediate discontinuation of medication and study exclusion.
In adult patients after the Fontan operation, peak o2 has been reported to be ≈20 mL·kg−1·min−1, with a standard deviation of 4.3 mL·kg−1·min−1.9,20 Assuming a treatment effect of 4 mL·kg−1·min−1 and a required power of 90% at a 2-sided significance level of 5%, 24 patients in each group were needed. To account for exclusions for liver toxicity, medical noncompliance, and dropouts, the aim was to include 39 patients in each group.
SAS statistical software 9.3 (SAS Institute Inc, Cary, NC) was used for all analyses. The statistical analyses were conducted by the principal investigator and were reviewed by statistical consultants (ParamStat Ltd, Canterbury, Kent, UK).
Populations and Approaches for Handling Missing Data
The primary analysis was performed including data from all randomized subjects who completed the study; hence, this was an observed case analysis. Additionally, 3 sensitivity analyses were conducted to assess the impact of protocol deviations and the handling of missing data. A per-protocol population was defined as a subset of the subjects in the observed case analysis, excluding subjects with medical noncompliance (defined as intake of <75% of study medication) or respiratory exchange ratio <1.1 for baseline or end-of-study CPET. Two analyses were conducted including all randomized patients (intention-to-treat [ITT] population) with different approaches for handling the missing data. The first used worst-case imputations (ITT-WC), and the second used multiple imputations (ITT-MI) derived from the observed baseline and postbaseline data. For the WC analysis, missing data were replaced by the worst change from baseline within the appropriate treatment group. For the MI analysis, 100 data sets were created with different imputations for the missing data. These imputations came from a model including baseline peak o2 and 5 other variables that demonstrated the largest absolute correlation with the change in peak o2. These were also the only variables with an absolute correlation >0.2 with the change in peak o2. The MI analysis was a post hoc analysis suggested by peer reviewers.
Parametric and Nonparametric Analyses
When appropriate, changes from baseline were analyzed by ANCOVA, including terms for treatment group and the baseline variable for the parameter being analyzed. The difference between treatment groups was estimated and presented, along with its 2-sided 95% confidence interval and 2-tailed P value. This parametric test was used as long as the assumption of normality was not violated. The Shapiro-Wilk test and inspection of histograms and normality plots identified a few variables of concern, but only pro-brain natriuretic peptide (pro-BNP) was of sufficient concern to be analyzed with the nonparametric van Elteren test and with the Hodges-Lehmann estimator to determine the treatment effect. No adjustment was made for multiplicity because secondary end points are considered supportive to the primary end point.
Analysis of NYHA Class
Because there was only 1 subject in NYHA class III at baseline, a binary response was generated by combining classes II and III and leaving class I as a category on its own. Improvements and deteriorations from baseline were analyzed in a proportional-odds logistic regression model including baseline NYHA class as a covariate. The odds ratio and corresponding 95% confidence interval were presented for bosentan compared with placebo.
A wide variety of baseline variables were tested for their interaction with treatment effect to determine whether the treatment effect was greater for particular groups of subjects. For each variable with a significant interaction with treatment and for a few predefined variables (baseline values of peak o2, pro-BNP, and ET-1), the participants were divided into 2 groups, those with values lower than and those with values higher than median. The treatment responses in the 2 groups were then estimated. Because of the number of tests, these analyses should be regarded as exploratory and not conclusive.
The number of subjects reporting adverse events was entered into a browser-based database with terms for the expected adverse events and effect, as well as a field for any unexpected events and effects. These were compared between the treatment and placebo groups by use of the Fisher exact test. Systolic blood pressure changes from baseline were analyzed with ANCOVA.
Demographics and Baseline Characteristics
Of the 107 eligible Danish subjects, 66 (62%) were included. Patients who declined participation and those who were included had similar characteristics in terms of sex, age, and peak o2 at the last clinical visit. Including 9 Swedish patients, a total of 75 patients were enrolled. Baseline observations and measurements are shown in Table 2. The mean±SD age was 20±7.4 years, and the median age at Fontan completion was 3.9 years (quartiles 1–3, 2.8–6.9 years). Sixty percent were male. Two thirds had left or mixed systemic ventricular morphology. Peak o2 was 28.3±6.9 mL·kg−1·min−1. There were no differences in baseline characteristics between the 2 groups.
Efficacy Primary End Point
Sixty-nine patients (92%) completed the study (Figure 1). The overall treatment duration was 93.9±18.4 days, and the compliance was 94.4±9.2%. Two patients in the placebo group had compliance <75%. Treatment duration and compliance were similar in the 2 groups.
When the observed case data set was used, peak o2 increased from (mean±SD) 28.7±8.2 to 30.7±8.7 mL·kg−1·min−1 in the bosentan group and from 28.4±6.0 to 29.1±6.4 mL·kg−1·min−1 in the placebo group, a net treatment effect of 1.39 mL·kg−1·min−1 (95% confidence interval, 0.18–2.59; P=0.0245; Figure 2). A similar treatment effect was found in an analysis of the per-protocol and the ITT-MI data sets, whereas only a trend favoring bosentan was seen with the ITT-WC data set. In the ITT-WC data set, 4 patients in the bosentan group had missing data and were assigned worst change in the group, whereas this was the case for only 2 patients in the placebo group (Table 3).
A cumulative graph of the percentage of subjects in the 2 groups achieving specific changes from baseline is shown in Figure 3.
Secondary End Points
Patients in the bosentan group exhibited significant improvements in CPET duration and a significant reduction in pro-BNP levels. Changes in other secondary end points were not significant but generally in favor of bosentan (Table 4).
At baseline, 52 patients were in NYHA class I (22 in the bosentan group and 30 in the placebo group), 22 were in class II (13 in the bosentan group and 9 in the placebo group), and 1 patient (bosentan group) was in class III. For the entire population, the mean peak o2 was higher (29.2; n=52) with class I than class II (26.2; n=22). Similarly, the mean SF-36 total score was higher (130.4; n=51) with class I than class II (118.9; n=22). Both the mean SF-36 mental and physical scores were higher with class I, and consistency was observed with each parameter within treatment groups.
In the bosentan group, 9 patients improved 1 functional class (all from class II to I), and none deteriorated. In contrast, no subjects in the placebo group improved, and a single patient dropped 1 class (from class I to II; Figure 4). The odds ratio for improvement in the bosentan group compared with the placebo group was 0.069 (95% confidence interval, 0.00–0.41; P=0.0085). In addition to the planned observed case analysis shown above, sensitivity analyses were made as a result of the skewed baseline distribution. A WC-in-group data set was made with imputations for missing data, that is, no change for the 4 patients with missing data in the bosentan group and no change for the 2 patients in the placebo group. In this analysis, the odds ratio for improvement was 0.054 (95% confidence interval, 0.00–0.31; P=0.0028). Furthermore, an ITT-WC data set was made by imputing missing data in the bosentan group as deterioration and the placebo group as no change. The ITT-WC odds ratio for improvement was 0.5 (95% confidence interval, 0.10–2.16; P=0.47).
A minor but statistically significant decrease in hemoglobin was observed in bosentan-treated subjects from enrollment to the end of the study (Table 5). Patients in the bosentan group exhibited a significantly greater increase in peak oxygen pulse than the placebo group (10.4 to 11.3 versus 10.5 to 10.6 mL/kg; P=0.034). Desaturation during exercise and o2 pr. workload (amount of oxygen consumed per Watt workload during exercise) were unchanged from baseline to the end of the study.
A significantly greater treatment effect was observed in patients with higher than median heart rate at unloaded exercise and in those with higher than median ventilatory equivalents for O2 and CO2 at anaerobic threshold and peak exercise. Although not statistically significant, treatment responses tended to be greater in male subjects, in those with a low body mass index, in young patients, and in those with high plasma ET-1, whereas no difference was seen between patients with high and low baseline values of peak o2 and pro-BNP (Table 6).
Six of the 75 patients (8%) did not complete the trial. Three withdrew because of assumed adverse effects: 2 patients (1 from each group) had recurrence of their protein-losing enteropathy, and 1 patient (bosentan group) experienced chest pain. Three dropped out before their final visit for nondrug reasons: 1 patient (bosentan group) while hospitalized for influenza and 2 patients (1 from each group) who revoked their consent before the dose-adjustment visit without giving a reason.
No serious adverse events occurred. Half of the bosentan-treated subjects reported adverse events compared with 59% in the placebo group (Table 7). The most frequently reported adverse effect in the bosentan group was flushing (6 of 36, 17%), which was seen in only 1 subject in the placebo group (3%; P=0.050). No patients reached the prespecified 3-fold upper limit for aminotransferases, but 5 patients (1 in the treatment group) had ≥1 measurements between 1 and 2 times the upper reference limit. A minor, albeit insignificant, reduction in systolic blood pressure was found in both groups (−6 versus −2 mm Hg; P=0.57).
The present study is the first to examine the effect of medium-term ERA treatment on exercise capacity in adolescents and adults after the Fontan operation in a placebo-controlled design. The study indicates that bosentan not only is safe but also improves exercise capacity and functional class. Corrected for placebo, peak o2 increased 1.4 mL·kg−1·min−1 (5%), which is a smaller change than anticipated. CPET time increased 23.4 seconds (6%). Importantly, this beneficial treatment response translated to a significant improvement in NYHA class.
In patients after the Fontan procedure, PVR is crucial for maintaining optimal hemodynamics at rest but perhaps more importantly during exercise. In this situation, PVR needs to be not only normal but preferably as low as possible to secure systemic ventricular filling and a relevant increase in cardiac output. There is evidence to suggest that PVR after the Fontan operation may increase slowly over time.11,21 This could be multifactorial, but alterations in pulmonary vascular or endothelial function are likely to be of importance,15 whether secondary to ventricular dysfunction or secondary to the loss of pulmonary pulsatile flow.12 Whatever the exact mechanisms, pulmonary vasodilators that predominantly modulate endothelial function could theoretically improve exercise-dependent hemodynamic requirements and thus exercise tolerance and well-being in these patients. This issue has been addressed scarcely, in small patient groups with different drugs and mostly in single-dose acute studies. So far, only 1 placebo-controlled trial with sustained oral treatment has been published.22
A recent open-label study on 42 Fontan patients failed to show effects of 6 months of bosentan treatment.17 However, the study was hampered by the facts that 24% of patients withdrew before the final visit and that the CPET was performed in multiple centers by different investigators. In another open-label study of 27 patients randomized to either a single oral dose of sildenafil (n=18) or no treatment (n=9), sildenafil increased pulmonary blood flow and the confidence interval at rest and at peak exercise, leading to a 9% increase in peak o2.20 These findings, however, could not be confirmed in a subsequent randomized, double-blind, placebo-controlled trial in 28 patients. Six weeks of sildenafil treatment (20 mg 3 times daily), improved respiratory efficiency during peak and submaximal exercise, whereas peak o2 did not change. However, borderline improvement at the anaerobic threshold was seen in those with single left or mixed ventricular morphology and in those with a high BNP level.22 In another placebo-controlled study, a 5% increase in peak o2 was found after a single dose of inhaled iloprost. The 9 patients with a baseline o2 <30 mL·kg−1·min−1 were all responders, whereas a positive treatment response was only seen in 3 of 6 patients with peak o2 >30 mL·kg−1·min−1.23 From these observations with pulmonary vasodilators, it was anticipated that subgroup analyses in our study could pinpoint measures such as baseline peak o2,23 pro-BNP,22 or plasma ET-1, which could serve to identify subjects most likely to be responders. Although patients with plasma ET-1 >67 pmol/L had marginally better treatment response, no marker has consistently proved an ability to predict a treatment effect in either the present or previous studies. The most promising variable to predict treatment effect in the current data was a higher than median ventilatory equivalent for O2/CO2 at peak exercise. Because of the present high cost of bosentan and the potential adverse effects, it would be desirable, if at all feasible, to distinguish responders from nonresponders before beginning life-long treatment. One solution could be to begin treatment in eligible subjects and then evaluate treatment response by, for example, CPET and symptoms after a test period of 3 to 6 months.
Pulmonary vasodilators are used extensively in patients with idiopathic pulmonary arterial hypertension and Eisenmenger syndrome with consistent beneficial effects.24–27 In idiopathic pulmonary arterial hypertension and Eisenmenger syndrome, PVR is severely elevated, which is in contrast to the normal to mildly elevated PVR after the Fontan operation. In these patients, several other factors may contribute to the exercise limitations and explain the somewhat lower treatment effect compared with what is found in idiopathic pulmonary arterial hypertension and Eisenmenger syndrome. However, the significant improvement in NYHA class seen in this study suggests that the effect is of clinical relevance. Heart catheterization may provide information on pulmonary blood flow and PVR to support the findings of positive effects on peak o2 and functional class reported here.
Although treatment significantly improved functional class, the skewed distribution of NYHA classification at baseline between the bosentan and placebo groups was of concern. Two ITT-WC analyses were therefore performed for sensitivity. The ITT-WC-in-group analysis showed highly significant improvement with treatment, whereas the ITT-WC analysis showed no effect. The latter was, however, a very pessimistic analysis because deterioration was imputed for the 4 patients with missing data in the bosentan group, even though this was not seen in any patients with complete data. It is highly unlikely that all 4 patients who withdrew from the bosentan group did so because of symptomatic deterioration.
The effect on functional class may be underestimated as a result of underreporting of exercise limitation. At baseline, nearly 70% of the patients reported to be in NYHA class I despite a peak o2 of only 28 mL·kg−1·min−1, that is, close to half of what is seen in healthy, lean individuals. This may be explained by the fact that NYHA class was developed for patients with congestive heart failure,28 whereas Fontan patients have a life-long chronic disease and may have adapted daily exercise and expectations accordingly, thus regarding their own physical capacity as normal. The same may be the case for the SF-36 quality-of-life questionnaire, in which many patients, despite physical limitations, achieved close to maximum scores, leaving only small margins for improvement.29 Unlike NYHA class, no change was detected in SF-36 after treatment.
Half of the bosentan-treated subjects reported side effects versus 59% in the placebo group. Side effects were mild, and no serious adverse events occurred during 14 weeks of treatment. It is not surprising to see a small but significant decrease in hemoglobin during treatment because mild anemia is a common side effect of bosentan. Mild elevations in aminotransferase levels were seen in only 2 patients on bosentan and 3 patients in the placebo group, whereas no patients reached the predefined 3-fold upper limit, which, according to US Food and Drug Administration drug information,30 would be expected in 11% of bosentan-treated subjects. Because hepatic abnormalities, possibly related to chronic venous congestion and low cardiac output,31 have increasingly been described years after the Fontan operation, increased susceptibility to increases in liver function tests might have occurred. If liver abnormalities are induced predominantly by abnormal hemodynamics, bosentan may potentially improve or prevent rather than increase the risk for this otherwise significant problem. However, the long-term safety of bosentan in Fontan patients still needs to be demonstrated.
From our findings and the scarce literature available, it is premature to provide general recommendations for the use of bosentan or other pulmonary vasodilators in patients after the Fontan operation. The beneficial effects now observed in 2 randomized, controlled trials suggest that the drugs have small but significant beneficial effects on important cardiopulmonary measures. Timing of ERA treatment remains unknown. It could be argued that treatment should be started early to gain the maximum effect. In older Fontan patients, a number of other factors, including progressive diastolic ventricular function, may limit the response to lower PVR. Furthermore, ERAs, phosphodiesterase inhibitors, and prostanoids use different pathways for decreasing PVR, and combined treatment may, as previously observed in patients with idiopathic pulmonary arterial hypertension,32 theoretically have an additive effect in patients after the Fontan operation.
The sample sizes in our subgroup analyses were inadequate to differentiate responders from nonresponders on the basis of, for example, peak o2, plasma pro-BNP, and plasma ET-1. Furthermore, it was only demonstrated that bosentan is a safe and efficacious treatment for 14 weeks; long-term responses remain to be investigated. Finally, it should be noted that no invasive hemodynamic measurements were obtained in this study; therefore, the inferences concerning the effect of bosentan through lowering the PVR are somewhat speculative. Despite randomization, the distribution of NYHA class between the bosentan and placebo groups at baseline was skewed. This complicated interpretation of the observed changes in functional class.
This randomized, placebo-controlled, double-blind study demonstrates that 14 weeks of bosentan treatment improves oxygen consumption, functional class, and exercise time in clinically stable adolescents and adults after the Fontan operation with minimal side effects, no hepatic toxicity, and no serious adverse events. The long-term impact of bosentan is worthy of further investigation.
We thank the GCP unit of Copenhagen and the staff at the participating centers in Copenhagen, Aarhus, Stockholm, and Lund for their support of this study.
Sources of Funding
This study was supported by an investigator-initiated research grant from Actelion Pharmaceuticals, which also provided the study drug. Actelion Pharmaceuticals was not involved in the writing or editing of the report or analysis of the data.
Dr Hebert is the recipient of an investigator initiated grant from Actelion Pharmaceuticals. The other authors report no conflicts.
- Received December 26, 2013.
- Accepted September 18, 2014.
- © 2014 American Heart Association, Inc.
- Fontan F,
- Baudet E
- Gewillig M
- Hosein RBM,
- Clarke AJB,
- McGuirk SP,
- Griselli M,
- Stumper O,
- De Giovanni JV,
- Barron DJ,
- Brawn WJ
- Ono M,
- Boethig D,
- Goerler H,
- Lange M,
- Westhoff-Bleck M,
- Breymann T
- Diller GP,
- Dimopoulos K,
- Okonko D,
- Li W,
- Babu-Narayan SV,
- Broberg CS,
- Johansson B,
- Bouzas B,
- Mullen MJ,
- Poole-Wilson PA,
- Francis DP,
- Gatzoulis MA
- La Gerche A,
- Gewillig M
- Gewillig M,
- Brown SC,
- Eyskens B,
- Heying R,
- Ganame J,
- Budts W,
- La Gerche A,
- Gorenflo M
- McLaughlin VV,
- Archer SL,
- Badesch DB,
- Barst RJ,
- Farber HW,
- Lindner JR,
- Mathier MA,
- McGoon MD,
- Park MH,
- Rosenson RS,
- Rubin LJ,
- Tapson VF,
- Varga J
- Giardini A,
- Balducci A,
- Specchia S,
- Gargiulo G,
- Bonvicini M,
- Picchio FM
- Goldberg DJ,
- French B,
- McBride MG,
- Marino BS,
- Mirarchi N,
- Hanna BD,
- Wernovsky G,
- Paridon SM,
- Rychik J
- Galiè N,
- Manes A,
- Negro L,
- Palazzini M,
- Bacchi-Reggiani ML,
- Branzi A
- Kuwana M,
- Watanabe H,
- Matsuoka N,
- Sugiyama N
- Galiè N,
- Beghetti M,
- Gatzoulis MA,
- Granton J,
- Berger RM,
- Lauer A,
- Chiossi E,
- Landzberg M
- 28.↵Boston: Little, Brown. Diseases of the Heart and Blood Vessels: Nomenclature and Criteria for Diagnosis. Boston, MA: Criteria Committee, New York Heart Association; 1964:114.
- 30.↵Drugs.com. Tracleer Official FDA information, side effects and uses. http://www.drugs.com/pro/tracleer.html. Accessed December 10, 2013.
- Humbert M,
- Barst RJ,
- Robbins IM,
- Channick RN,
- Galiè N,
- Boonstra A,
- Rubin LJ,
- Horn EM,
- Manes A,
- Simonneau G
The natural history of children born with functional univentricular hearts was previously dismal. After the introduction of the Fontan operation and its subsequent modifications, survival into adulthood is now the norm, although long-term complications seem unavoidable and survival remains by no means normal. As a natural consequence of the single-ventricle physiology, Fontan patients have reduced exercise capacity, typically 60% to 70% of the expected for normal control subjects. The main limiting factor of circulation is the lack of a subpulmonary ventricle, which leaves the pulmonary blood flow passively driven without propulsion from a pumping chamber. The passive transpulmonary flow is greatly dependent on a low pulmonary resistance. Unfortunately, the pulmonary resistance often rises over time, leading to clinical worsening. The present article presents the first double-blind, randomized, placebo-controlled study of the pulmonary vasodilator bosentan. In 75 patients treated for 14 weeks with bosentan, improvement was seen in oxygen uptake, exercise time, and New York Heart Association functional class. Furthermore, bosentan was safe to use for 14 weeks with no serious adverse effects. The only side effect more frequent in the bosentan group than in the placebo group was flushing. Although it is still too early to recommend this treatment generally in patients with Fontan circulation, these findings may be of great importance in future research and management of patients with Fontan circulation.