Review of evidence for bosentan therapy for treatment of Eisenmenger syndrome

Benjamin J. Hartwig, DNP, AGACNP-BC (Doctor of Nursing Practice Candidate)1 & Benjamin Schultze, PhD, ARNP-BC (Associate Professor)

Journal of the American Association of Nurse Practitioners


Background and purpose: Eisenmenger syndrome (ES) is a rare condition caused by a right-to-left cyanotic shunt. To date, only heart–lung transplant has been shown to be curative. Bosentan is the only medication studied with a double-blind placebo-controlled randomized trial for management of this condition. The intent of this article is to explore the literature surrounding bosentan in ES and assess its efficacy.

Methods: A literature review was conducted with no limitation on date. Titles were scanned for applicability, and abstracts of those articles found to be pertinent were reviewed. Those articles considered relevant based on the abstract were read in entirety.

Conclusions: Eisenmenger syndrome remains incurable except through heart–lung transplant. Although no specific medical treatment or algorithm exists, three pharmacological classes show promise in disease management: endothelin receptor antagonists, phosphodiesterase inhibitors, and prostacyclins. Combined therapy with these agents may improve cardiopulmonary function. Bosentan has not been proven as a monotherapy for ES and is not appropriate in all patients as side effects are commonly reported.

Implications for practice: Further study is required to assess efficacy of combination therapy and utilization as a bridge to transplant or surgical correction of the underlying defect.

First described by Dr. Viktor Eisenmenger in 1897, Eisenmenger syndrome (ES) is a rare cardiac condition characterized by the reversal of blood flow through a defect in the cardiac chambers and/or vasculature leading to right heart failure. Reversal occurs because of the left heart forcing oxygenated blood at systemic pressures into the much lower pulmonary pressure oxygen-depleted right side of the heart, increasing pulmonary pressure and flow. This increased flow causes increased pulmonary vascular resistance, and pulmonary artery hypertension(PAH) ensues. To compensate, the right heart hypertrophies, eventually overpowering the pressure in the left heart. This forces venous blood through the communication into the arterial circulation, creating a cyanotic defect. The most common communications are atrial septal defects (ASDs), ventricular septal defects (VSDs), and patent ductus arteriosus (Connolly, 2017). Eisenmenger syndrome is the final progression of PAH in patients with congenital heart disease (CHD) (Lopes & Alnajashi, 2014). Eisenmenger syndrome's cardiopulmonary physiology changes are analogous to other types of PAH, although ES also affects other organ systems (Huang, Liang, & Zhou, 2012).

Epidemiology and prognosis

Congenital heart disease is the most frequent of the major congenital anomalies seen at birth with an incidence as high as 8 per 1,000 (Fine, Dias, Shoemaker, & Mehta, 2009). Of patients with CHD, approximately 10% develop PAH which, if left unmanaged, can ultimately lead to ES (Dimopoulos et al., 2010 Trojnarska & Plaskota, 2009) in approximately 1% of patients CHD (Schuuring, Vis, Duffels, Bouma, & Mulder, 2010). Although ES typically develops in early childhood or puberty (Kaneshiro & Chen, 2012), more patients are seen living into their 30s and 40s owing to improved medical and surgical techniques (Galie et al., 2006). These patients have a relatively stable clinical status when compared with other forms of PAH (Kula & Atasayan, 2015). Expected mortality in this population is within the first 4 decades (Huang et al., 2012), with only 55% of these patients surviving until 55 years of age (Schuuring et al., 2010). The most common causes of death are arrhythmias; noncardiac surgery, due to hemodynamic complications; heart failure (Trojnarska & Plaskota, 2009); and sequelae of multiorgan involvement such as hemoptysis, which is responsible for up to 30% of fatalities (Dimopoulos, Giannakoulas, Wort, & Gatzoulis, 2008 Trojnarska & Plaskota, 2009 Wood, 1958). The literature is clear that impaired exercise tolerance is an independent predictor of mortality (Galie et al., 2008 Macchia et al., 2007 Simonneau et al., 2014).


Three pathways are crucial to the pathogenesis of ES: endothelin-1 (ET1); nitric oxide (NO); and prostacyclin (PGI2), all of which are targets of advanced therapy (AT) intended to mitigate ES progression (Dimopoulos et al., 2008). Prostacyclin-2 and NO elicit vasodilation in patients with increased pressure in the pulmonary artery and right ventricle. Eventually, despite PGI2 and NO, because of the inability to fully accommodate the increased flow, endothelial damage occurs. As this endothelial damage progresses, a mismatch between vasoconstriction and vasodilation, favoring a net state of vasoconstriction, develops (Trojnarska & Plaskota, 2009). Endothelin-1, a powerful vasoconstrictor, is noted to be elevated in patients with ES (Galie et al., 2006 Schuuring et al., 2010) and is linked to endothelial proliferation, fibrosis, and inflammation, as well as myocyte hypertrophy, resulting in a decreased luminal diameter of the pulmonary arteries with concomitant right ventricular hypertrophy (D'Alto et al., 2007). Endothelin-1, manufactured in endothelial cells, acts in the region of production on the ETA and ETB receptor subtypes, each of which has counteracting effects. ETA receptors mediate vasoconstriction and potentially cellular proliferation, whereas ETB receptors are involved in clearance of ET1 and promote vasodilatory action of NO and PGI2 (D'Alto et al., 2007 Galie et al., 2006). Bosentan, an oral dual ETA and ETB receptor antagonist proven effective for the treatment of idiopathic PAH and ES (Galie et al., 2006), has also shown to potentially reverse pulmonary vascular hypertrophy alleviating PAH by reverse remodeling (D'Alto et al., 2007). As ETB activation seems to be beneficial in PAH and ES, bosentan binds nonselectively to both ETA and ETB receptors with an affinity of 20:1 (Schuuring et al., 2010). Additional sequelae of ES include increased red cell production to compensate for systemic cyanosis. This in turn leads to increased blood viscosity and increases in hemoglobin and hematocrit. This increased viscosity and platelet dysfunction is associated with stroke and increased bleeding risks (Trojnarska & Plaskota, 2009).


Cardiac catheterization is considered the gold standard (Schuuring et al., 2010) to diagnose ES. Diagnostic criteria include flow reversal with hemodynamic parameters of mean pulmonary trunk pressures >25 mm Hg at rest or >30 mm Hg during exercise (Trojnarska & Plaskota, 2009) and a mean pulmonary capillary wedge pressure <15 mm Hg (Macchia et al., 2007). The 6-minute walk distance (6MWD) test is used in studies as a noninvasive measure of treatment effect and predictor of mortality (Galie et al., 2006). The validity of the 6MWD may be less reliable in those patients with mental disability (Schuuring et al., 2010) such as Down syndrome, many of whom also suffer from ES (Crepaz, Romeo, Montanaro, & Santis, 2013). The World Health Organization functional class (WHO FC) for PAH assesses the degree of functional decline based on characteristics of dyspnea, fatigue, chest pain, syncope at rest, or onset of exertion and assigns an I–IV value (IV being most advanced) based on that presentation (Table 1) (Simonneau et al., 2014). An additional tool commonly referenced is the subjective Borg dyspnea scale rating shortness of breath on an ascending scale of severity from 0 to 10 (Galie et al., 2006).

Current state of medical management

Historically, the only treatment options for ES were as follows: a heart, lung, or combination transplant; lung transplant with cardiac repair; or provision of palliative care with dramatic lifestyle modifications (Dimopoulos et al., 2010 Fukushima, 2015 Trojnarska & Plaskota, 2009). Until bosentan, no medical treatment had been shown effective for ES (Yao, 2015). As palliative measures, patients have been treated with medications such as, diuretics, digoxin, antiarrhythmics, anticoagulants, oxygen, and supplemental iron to manage the sequelae of this condition (Dimopoulos et al., 2010 Fukushima, 2015). These therapies have not been shown to increase the duration of life and require careful monitoring as therapeutic and harmful effects are separated by only a minor degree (Huang et al., 2012).

To date, only one placebo-controlled, double-blind, multicenter, randomized trial of any medication has been performed and shown to have definite benefit in medical management of ES (Dimopoulos et al., 2008). Bosentan has been shown to improve symptoms and delay the need for transplant. In addition, whether transplantation extends life is not fully known as posttransplant outcomes in this population are poor. In fact, the natural lifespan of patients with ES may exceed that of posttransplant patients with proper management (Adriaenssens, Delcroix, Deyk, & Budts, 2006). Furthermore, AT with bosentan may allow for surgical repair of the underlying defect once significant reverse proliferation and correction of pulmonary pressures has occurred (Huang et al., 2012).

Evidence of progression over time

In the first reported case in 1897, Eisenmenger's patient was found on autopsy to have succumbed to hemoptysis secondary to PAH in the setting of a VSD. In 1958, Paul Wood established the pathophysiological mechanisms underlying ES. At that time, attempts at surgical repair were responsible for more deaths in this patient population than any other cause with the exception of hemoptysis. Advocating venesections for therapeutic phlebotomy (a practice still used today), Wood noted that PAH treatments were ineffective, leaving only permanent anticoagulation as a method to prevent thromboembolism (Wood, 1958). Unfortunately, these venesections are only a temporizing measure as therapeutic phlebotomy is associated with iron deficiency and microcytosis, leading to increased risk of thrombosis (Trojnarska & Plaskota, 2009). The first, and only, placebo-controlled randomized trial of WHO FC III ES patients occurred in 2006.

The Bosentan Randomized Trial of Endothelin Antagonist Therapy-5 (BREATHE-5) study showed functional improvement with bosentan after 16 weeks of therapy when compared with the placebo control group. Fifty-four patients were randomized 2:1 to treatment (n = 37) or placebo (n = 17), baseline characteristics with regard to age, sex, time from diagnosis, defect type, 6MWD, and hemodynamics were equivalent between the two groups (Galie et al., 2006). In 2010, a survival benefit was shown for AT in a retrospective analysis, indicating that appropriate therapy with a combination of epoprostenol, sildenafil, or bosentan can offer a viable alternative to transplant (Dimopoulos et al., 2010). Historically, lifestyle modifications have been the only nonpharmacological modalities available to patients with ES who cannot receive a transplant. Such modifications include avoidance of the following: physical activity, dehydration (Trojnarska & Plaskota, 2009), high altitude, and pregnancy in female patients (Schuuring et al., 2010). These life style changes may palliate symptoms but also likely decreases quality of life (D'Alto et al., 2007).

Argument in favor of bosentan

Bosentan has shown promise not only as a temporizing measure but also as a potential means to reverse the physiologic effects of ES, resulting in improved quality of life and functionality (Adriaenssens et al., 2006 Fukushima, 2015 D'Alto et al., 2007). The BREATHE-5 trial showed that bosentanis as effective as, or superior to, no treatment in regards to arterial oxygenation, as well as establishing the safety profile of this medication in ES, demonstrating improvement in 6MWD and hemodynamic profiles over a 40-week period (Galie et al., 2006). This population has historically documented poor outcomes of transplant secondary to multiorgan involvement (Dimopoulos et al., 2008) and has been shown to have the highest perioperative mortality rate among heart–lung transplants (Trojnarska & Plaskota, 2009). Bosentan shows promise in delaying the need for transplant while increasing quality of life in the interim (Adriaenssens et al., 2006). Given this information, a provider may be inclined to prescribe bosentan for ES while considering earlier implementation of treatment in patients with inoperable CHD with the intent of delaying disease progression. The Endothelin Antagonist Trial in Mildly Symptomatic PAH Patients (EARLY) study documented efficacy in this area in WHO FC II patients who suffered CHD but had not yet developed ES physiology (Galie et al., 2008 Simonneau et al., 2014).

Argument against bosentan

Bosentan is metabolized through the cytochrome p (CYP) 450, CYP2C9, and CYP3A4 pathways (Schuuring et al., 2010) and should be avoided in patients with liver failure in general. One of the most commonly reported side effects of bosentan therapy is elevated liver enzymes ≥3 times the upper normal limit (Galie et al., 2006 Fine et al., 2009 D'Alto et al., 2007). Because of potentiation of right-sided heart failure, severe peripheral edema can occur causing some patients to discontinue the drug (Galie et al., 2006 D'Alto et al., 2007). Typically, liver enzymes returned to normal and edema resolved after cessation or dose reduction, although one case study showed that dose reduction was ineffective at improving liver status or maintaining the intended effects of bosentan (Yao, 2015).

Currently, prolonged controlled trials to monitor the efficacy of bosentan over time are lacking. The longest studies found for this review followed results for less than 4 years, at which point the efficacy of bosentan was waning. One study found that improvement in 6MWD peaks at 4 months and WHO FC improvement peaks at 1 year, after which improvements gradually return to baseline despite continued therapy (Van Loon et al., 2007). Similar results have been found in other studies and case reports. Apostolopoulou, Manginas, Cokkinos, and Rammos (2007) showed a return to baseline of 6MWD, exercise capacity as measured by metabolic equivalents and duration, and Borg dyspnea index by 2 years. Gatzoulis (2011) reported the case of a 41-year-old patient who did well on bosentan with an improved quality of life for 3.5 years, at which time her 6MWD decreased and her dyspnea worsened, requiring the addition of sildenafil to return to her improved state. Schuuring et al. (2010) reported similar findings in patients returning to baseline exercise parameters by 2 years.

Patients not able to receive transplants typically continue bosentan therapy lifelong to continue receiving the benefits. If the beneficial effects do in fact continue for that long, cost cannot be ignored. At the time of publication, one study (Macchia et al., 2007) estimated the annual cost of bosentan at $80,000, whereas the addition of other advanced therapies such as sildenafil or parenteral PGI can add an annual cost of $50,000–$100,000. Assuming that the therapeutic benefit only lasts 3.5 years, this treatment regimen costs over a quarter million dollars only to potentially return to baseline, requiring additional costly therapy to regain benefit or leaving patients without options for further medical recourse. This finding counters the results of the EARLY study and indicates that earlier initiation of AT would be detrimental to long-term outcomes and that AT should be withheld until the latest possible time to prolong quality and duration of life.


Bosentan has been demonstrated as effective by the BREATHE-5 (Galie et al., 2006) and EARLY (Galie et al., 2008) trials to improve hemodynamics, 6MWD, Borg dyspnea index, and in certain cases WHO functional class. There has not however, been a proven benefit in reducing mortality with bosentan over time (Macchia et al., 2007 Trojnarska & Plaskota, 2009). Van Loon et al. (2007)showed that the persistence of bosentan's effect at one year was 68% and at 2 years dropped to 43%.

BREATHE-5 is the controlled study of greatest duration in the ES patient population which only assessed outcomes for up to 40 weeks. The initial 16 weeks were the randomized arm, whereas the last 24 weeks were an open-label continuation study in which the placebo group was also given bosentan. This showed improvement in WHO functional class in most patients; maintenance of improvement in 6MWD in patients who had initially been randomized to the bosentan group; and significant improvement in 6MWD in those patients beginning bosentan at the open-label extension point (Galie et al., 2006). One of the shortfalls with the BREATHE-5 study, aside from its short duration and low power, is the focus on atrial septal defect and VSD to the exclusion of patent ductus arteriosus and complex congenital abnormalities (Diaz-Caraballo et al., 2009), which may call into question the utility of this drug for such patients with ES. With the BREATHE-5 trial lasting only 40 weeks and a mean age in the literature reviewed for this paper of 34 years, controlled studies of greater duration are needed though unfortunately not available.

The EARLY study is a longer placebo-controlled randomized controlled trial for patients in WHO class II, advocating the potential benefit of earlier implementation of bosentan for CHD management to prevent progression to ES. Unfortunately, it did not include the ES demographic as patients had not yet progressed to that point (Galie et al., 2008). Despite this, results revealed mixed promise with a 5-year open-label continuation study showing 158 WHO FC I or II patients at baseline with 122 remaining in class I or II and 35 deteriorating to class III or IV. Maximum follow-up period within the study was 37.9 months (Galie et al., 2008). Overall, 18.4% of patients improved, whereas 22.8% of patients deteriorated (Simonneau et al., 2014), indicating that this treatment is not valid for long-term reversal but may be useful for maintaining the status quo and preventing further functional decline. This theory is supported by studies showing bosentanand AT as a bridge or alternative to transplant (Adriaenssens et al., 2006), especially if started early (Galie et al., 2008).

ES is the endpoint of PAH in patients with CHD who have a 55% life expectancy at age 55 (Crepaz et al., 2013 Diaz-Caraballo et al., 2009). Although there is no cure for ES, bosentan offers hope of improved quality of life and at least temporary cessation of functional decline. Dimopoulos et al. (2010) demonstrated that treatment with AT, including bosentan, was superior to the alternative of no treatment or conservative management. After a 4-year follow-up of 229 patients, 68 of whom were on AT including bosentan, sildenafil, and epoprostenol, 52 patients had died, only two of whom were on AT. Following the information from this study and others similar to it, the conclusion may be reached that bosentan alone may not be the answer for patients with ES but that combination therapy with phosphodiesterase inhibitor (PDEi) and PGI2 in combination with bosentan and traditional therapies may provide added benefit.

A final area that merits further investigation is the potential for surgical repair of the underlying cardiac defect once the patient has been optimized on bosentan or combination AT. This approach has historically failed as the removal of the cyanotic shunt drastically overloads the pulmonary circulation, resulting in precipitous right heart failure necessitating shunt repair reversal. With further assessment of these treatments, it may become possible to reverse the pulmonary vasculature remodeling to the point that surgical correction of the shunt allows for a curative solution (Huang et al., 2012).


ES is a fatal disease involving a pulmonary to systemic cyanotic shunt, resulting in multiple organ involvement and comorbidities (Schuuring et al., 2010) with no current definitive nonsurgical cure or established treatment protocol (Diaz-Caraballo et al., 2009). Originally believed to only be treatable through anticoagulation, management of heart failure sequelae (Wood, 1958), palliative care (Fukushima, 2015), and transplant (Yao, 2015), we now have a better understanding of the pathophysiology involved. With that understanding, several pharmacological alternatives to traditional treatments have been created with promising results. Although bosentan use in ES was the focus of this article, it was discovered that concomitant treatment with PDEi and PGI2 are useful adjuncts. No survival benefit of bosentan or any other agent as a monotherapy has been conclusively demonstrated in the literature, and the evidence is not unequivocal regarding treatment options. Instead, individual cases require careful assessment and tailoring of implementation based on the progression of the disease and associated comorbidities. What is broadly agreed upon in the literature, and the conclusion of this review, is that further research is needed into the utility of bosentan and combined AT for ES.


Adriaenssens, T., Delcroix, M., Deyk, K. V., & Budts, W. (2006). Advanced therapy may delay the need for transplantation in patients with the Eisenmenger syndrome. European Heart Journal, 27, 1472–1477.

American Heart Association. (2017). Classes of heart failure. Retrieved from American Heart Association: http://www.heart. org/HEARTORG/Conditions/HeartFailure/AboutHeartFailure/ Classes-of-Heart-Failure_UCM_306328_Article.jsp#.Wr0t-dPwYsk.

Apostolopoulou, S., Manginas, A., Cokkinos, D., & Rammos, S. (2007). Long-term oral bosentan treatment in patients with pulmonary arterial hypertension related to congenital heart disease: A 2-year study. Heart, 93, 350–354.

Connolly, H. M. (2017). Evaluation and prognosis of Eisenmenger syndrome. Retrieved from UpToDate: contents/evaluation-and-prognosis-of-eisenmenger-syndrome? source=search_result&search=Management%20of%20Eisenmenger%20syndrome&selectedTitle=3;49#H6.

Crepaz, R., Romeo, C., Montanaro, D., & Santis, S. D. (2013). Long-term results of treatment with bosentan in adult Eisenmenger’s syndrome patients with Down’s syndrome related to congenital heart disease. BMC Cardiovascular Disorders, 13, 1–7.

D’Alto, M., Vizza, C., Romeo, E., Badagliacca, R., Santoro, G., Poscia, R.,… Calabro, R. (2007). Long term effects of bosentan treatment in adult patients with pulmonary arterial hypertension related to congenital heart disease (Eisenmenger physiology): Safety, tolerability, clinical, and haemodynamic effect. Heart, 93, 621–625.

Diaz-Caraballo, E., Gonzalez-Garcia, A. E., Renones, M., SanchezRecalde, A., Garcia-Rio, F., & Oliver-Ruiz, J. M. (2009). Long-term bosentan treatment of complex congenital heart disease and Eisenmenger’s syndrome. Revista Espanola de cardiologia, 62, 1046–1049.

Dimopoulos, K., Giannakoulas, G., Wort, S. J., & Gatzoulis, M. A. (2008). Pulmonary arterial hypertension in adults with congenital heart disease: Distinct differences from other causes of pulmonary arterial hypertension and management implications. Current Opinion in Cardiology, 23, 545–554.

Dimopoulos, K., Inuzuka, R., Goletto, S., Giannakoulas, G., Swan, L., Wort, S. J., & Gatzoulis, M. A. (2010). Improved survival among patients with Eisenmenger syndrome receiving advanced therapy for pulmonary arterial hypertension. Circulation, 121, 20–25.

Eisenmenger, V. (1897). Die angeborenen Defecte der Kammerscheidewand des Herzens. Zeitschrift fur klinische Medicin, 32, 1–28.

Fine, N., Dias, B., Shoemaker, G., & Mehta, S. (2009). Endothelin receptor antagonist therapy in congenital heart disease with shuntassociated pulmonary artery hypertension: A qualitative systematic review. Canadian Journal of Cardiology, 25, e63–e68.

Fukushima, H. (2015). Update on medical treatment of patients with Eisenmenger syndrome. International Heart Journal, 56, S4–S7.

Galie, N., Beghetti, M., Gatzoulis, M., Granton, J., Berger, R. M., Lauer, A.,… Landzberg, M. (2006). Bosentan therapy in patients with eisenmenger syndrome: A multicenter, double-blind, randomized, placebo controlled study. Circulation, 114, 48–54.

Galie, N., Rubin, L., Hoeper, M., Jansa, P., Al-Hiti, H., Meyer, G.,… Simonneau, G. (2008). Treatment of patients with mildly symptomatic pulmonary arterial hypertension with bosentan (early study): A double-blind, randomised controlled trial. Lancet, 371, 2093–2100.

Gatzoulis, M. (2011). The management of Eisenmenger syndrome in the modern treatment era: A case report. European Respiratory Review, 20, 293–301.

Huang, J. B., Liang, J., & Zhou, L. Y. (2012). Eisenmenger Syndrome: Not always inoperable. Respiratory Care, 57, 1488–1495.

Humbert, M. (n.d.). The global alliance against chronic respiratory diseases. Retrieved March 29, 2018, from World Health Organization: %20hyperthension-Dr%20M.%20Hmpert.pdf.

Kaneshiro, N., & Chen, M. (2012). Eisenmenger syndrome. Retrieved February 14, 2017, from University of Maryland Medical Center:

Kula, S., & Atasayan, V. (2015). Surgical and transcatheter management alternatives in refractory pulmonary hypertension: Potts shunt. Anatolian Journal of Cardiology, 15, 843–847.

Lopes, A., & Alnajashi, K. (2014). Saudi guidelines on the diagnosis and treatment of pulmonary hypertension: Pulmonary arterial hypertension associated with congenital heart disease. Annals of Thoracic Medicine, 9, S21–S25.

Macchia, A., Marchioli, R., Marfisi, R., Scarano, M., Levantesi, G., Travazzi, L., Tognoni, G. (2007). A meta-analysis of trials of pulmonary hypertension: A clinical condition looking for drugs and research methodology. American Heart Journal, 153, 1037–1047.

Schuuring, M. J., Vis, J. C., Duffels, M. G., Bouma, B. J., & Mulder, B. J. (2010). Adult patients with pulmonary arterial hypertension due to congenital heart disesae: A review on advanced medical treatment with bosentan. Therapeutics and Clinical Risk Management, 6, 359–366.

Simonneau, G., Galie, N., Jansa, P., Meyer, G. M., Al-Hiti, H., Kusic-Pajic, A.,… Rubin, L. (2014). Long-term results from the EARLY study of bosnentan in WHO functional class II pulmonary arterial hypertension patients. International Journal of Cardiology, 172, 332–339.

Trojnarska, O., & Plaskota, K. (2009). Therapeutic methods used in patients with Eisenmenger syndrome. Cardiology Journal, 16, 500–506.

Van Loon, R. L., Hoendermis, E., Duffels, M., Vonk-Noordegraaf, A., Mulder, B., Hillege, H., Berger, R. M. (2007). Long-term effect of bosentan in adults versus children with pulmonary arterial hypertension associated with systemic-to-pulmonary shunt: Does the beneficial effect persist? American Heart Journal, 154, 776–782.

Wood, P. (1958). The Eisenmenger syndrome or pulmonary hypertension with reversed central shunt. British Medical Journal, 2, 755–762.

Yao, A. (2015). Medical treatment for an adult patient with Eisenmenger syndrome. International Heart Journal, 56, S8–S11.