Article Text

Download PDFPDF

Pulmonary arterial hypertension in adult congenital heart disease
  1. Margarita Brida1,2,3,4,
  2. Michael A Gatzoulis1,2
  1. 1 Adult Congenital Heart Centre and National Centre for Pulmonary Hypertension, Royal Brompton Hospital, London, UK
  2. 2 National Heart and Lung Institute, Imperial College, London, UK
  3. 3 Division of Adult Congenital and Valvular Heart Disease, Department of Cardiovascular Medicine, University Hospital Muenster, Muenster, Germany
  4. 4 Division for Adult Congenital Heart Disease, Department of Cardiovascular Medicine, University Hospital Centre Zagreb, Zagreb, Croatia
  1. Correspondence to Dr Margarita Brida, Royal Brompton Hospital, Sydney Street, London SW3 6NP, UK; margarita.brida{at}


Pulmonary arterial hypertension (PAH) is commonly associated with congenital heart disease (CHD) and relates to type of the underlying cardiac defects and repair history. Large systemic to pulmonary shunts may develop PAH if untreated or repaired late. PAH, when present, markedly increases morbidity and mortality in patients with CHD. Significant progress has been made for patients with Eisenmenger syndrome in pathophysiology, prognostication and disease-targeting therapy (DTT), which needs to be applied to routine patient care. Patients with PAH–CHD and systemic to pulmonary shunting may benefit from late defect closure if pulmonary vascular resistance (PVR) is still normal or near normal. Patients with PAH and coincidental defects, or previous repair of CHD should be managed as those with idiopathic PAH. Patients with a Fontan circulation, despite not strictly fulfilling criteria for PAH, may have elevated PVR; recent evidence suggests that they may also benefit from DTT, but more data are required before general recommendations can be made. CHD–PAH is a lifelong, progressive disease; patients should receive tertiary care and benefit from a proactive DTT approach. Novel biomarkers and genetic advances may identify patients with CHD who should be referred for late defect closure and/or patients at high risk of developing PAH despite early closure in childhood. Ongoing vigilance for PAH and further controlled studies are clearly warranted in CHD.

  • congenital heart disease
  • pulmonary vascular disease
  • secondary pulmonary hypertension

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Pulmonary hypertension (PH), commonly encountered in congenital heart disease (CHD), represents a wide spectrum of disease entities defined by a mean pulmonary arterial pressure (mPAP) ≥25 mm Hg at rest. The clinical classification of PH, first introduced in 1998, has since undergone numerous modifications. Currently, there are five main groups: pulmonary arterial hypertension (PAH, group I), PH due to left heart disease (II), PH due to lung disease (III), chronic thromboembolic PH (IV) and unclear multifactorial PH (group IV).1 Most patients with CHD are classified as group I PAH haemodynamically characterised by precapillary PAH (pulmonary vascular resistance (PVR) >3 Wood units (WU) and pulmonary artery wedge pressure ≤15 mm Hg).2 PAH in association with CHD (PAH–CHD) usually develops due to intracardiac or extracardiac shunts with unrestricted pressure and/or volume overload of the pulmonary circulation. This, in turn, and with time, induces shear stress, arterial endothelial damage and adverse pulmonary vascular remodelling. Although the exact prevalence of PAH–CHD remains unknown, it is estimated that approximately 10% of adults with CHD have PAH, which impacts on their quality of life and outcome.3 Conversely, in the setting of unexplained PAH congenital cardiac anomalies (eg, patent ductus arteriosus, sinus venosus atrial septal defect, partial anomalous pulmonary venous return, aortopulmonary window etc) should be excluded, as if present, they have a direct bearing on management and outcome.

Treatment of PAH–CHD has evolved in recent years with options for either late repair in some patients (surgical or catheter), or PAH disease-targeting therapy (DTT) in others.3 4 Our review focuses on clinical aspects and current management strategies in the four main clinical subgroups of PAH–CHD, namely Eisenmenger syndrome (ES), PAH associated with persistent systemic to pulmonary shunts, PAH with small/coincidental shunts and PAH after defect correction.1 5 Moreover, we included patients with a Fontan circulation in our discussion who do not strictly fulfil PAH criteria, but may nevertheless have elevated PVR and may potentially benefit from PAH–DTT (figure 1).

Figure 1

Basic characteristics of four main congenital heart disease subgroups with pulmonary arterial hypertension and of the distinct Fontan cohort with pulmonary vascular disease. ASD, atrial septal defect; iPAH, idiopathic pulmonary arterial hypertension; PAH, pulmonary arterial hypertension; PVD, pulmonary vascular disease; PVR, pulmonary vascular resistance; VSD, ventricular septal defect. Blue colour indicates deoxygenated blood; pink colour indicates oxygenated blood; purple colour indicates mixed deoxygenated and oxygenated blood.

Eisenmenger syndrome (PAH with severe PVD)

ES, the most severe form of PAH–CHD, results from large systemic-to-pulmonary shunts (atrial, ventricular or arterial) triggering the development of PAH and pulmonary vascular disease (PVD), and leading with time to reversal of shunting. Chronic cyanosis is a distinct feature in ES linked to multisystem involvement, including the haematopoietic system with secondary erythrocytosis, thrombocytopenia, coagulation abnormalities and so on (figure 1A). Defect closure in ES is contraindicated as the defect functions as a ‘relief valve’ for the high pulmonary artery pressures, maintaining systemic cardiac output through right-to-left shunting at the expense of cyanosis. Closing the defect in ES may precipitate right heart failure and death. In contrast to previous misbelief that patients with ES have much better survival, whether independently or in comparison with other PH aetiologies, a recent study accounting for immortal time bias showed that untreated patients have a poor survival with 10-year mortality rates ranging between 30% and 40%, underscoring the need for proactive therapy.6 Patients with Down syndrome account for approximately one-third of patients with ES. These patients may have more complex underlying CHD (atrioventicular septal defect being the most common lesion) and seem to develop accelerated PVD compared with patients with similar underlying cardiac defects without Down syndrome (floppy upper airways may be contributory to this). Patients with Down syndrome present certain challenges during diagnostic and follow-up assessment processes, including regular blood sampling, and may have higher non-compliance rates. Furthermore, acquisition and reliability of certain periodic parameters such as the 6 min walk test may be challenging, although the latter did not seem to be problematic—with appropriate supervision—in a recent study7 assessing the effects and safety of macitentan in ES including patients with Down syndrome. Also, perhaps causal, data from multinational registry showed that the proportion of patients on DTT was the lowest among patients with Down syndrome (18.6% vs 45.5% in the remainder).8

Treatment of ES includes general measures, supportive therapy and PAH–DTT. Heart and lung transplantation, or lung transplantation with repair of the underlying cardiac defect are also therapeutic options when other treatment strategies have been exhausted (table 1). General measures and supportive therapies should include regular assessment by trained and experienced PAH–CHD physicians, avoidance of pregnancy with effective contraceptive methods, psychosocial support, maintaining active lifestyle, endocarditis prophylaxis in selected defects and regular immunisation against influenza and pneumococcal infections.9

There has been considerable progress regarding prognostication in ES in recent years. Baseline characteristics (age, level of severity of underlying CHD, degree of cyanosis),8 10 11 functional aspects (New York Heart Association functional class, 6 min walk distance (6MWD)),4 10 12 biomarkers (brain natriuretic peptide, C reactive protein),13 14 echocardiographic indices,8 15 severity of PVD (PVR, acute vasoreactivity testing (AVT) at baseline)16 and other markers have been shown to provide prognostic information in ES (figure 2). Similarly, iron deficiency anaemia (with transferrin saturation <20% being a good marker for its detection), commonly found in ES, is associated with worse outcome.17 18 Patients with ES should, therefore, be routinely screened for iron deficiency and, if present, treated with iron supplementation (oral or intravenous) until they are iron replete.5 19 This is associated with an increase in haematocrit with improved quality of life and 6MWD.20 In contrast, routine venesections have become a thing of the past, as there is little evidence of derived benefit, and in fact they may be harmful by inducing or enhancing iron deficiency, compromising secondary erythrocytosis and impairing oxygen transport capacity.

Figure 2

Prognostication in Eisenmenger syndrome: a goal-oriented proactive approach. Evolving markers for assessing disease severity, prognosis, disease progression and response to disease-targeting therapy. Adopted from Gatzoulis et al.5 BNP, brain natriuretic peptide; CHD, congenital heart disease; CRP, C reactive protein; CI, cardiac index; LA, left atrium; NYHA, New York Heart Association; PDA, patent ductus arteriosus; PVRi, pulmonary vascular resistance index; RA, right atrium; RAP, right atrium pressure; TAPSE, tricuspid annular plane systolic excursion; VSD, ventricular septal defect.

Oxygen supplementation for chronic cyanosis has not been shown to have a positive effect on exercise capacity nor on survival in ES.21 The use of oral anticoagulants remains somewhat controversial; there is a relatively high prevalence of pulmonary arterial thrombosis in situ,22 arrhythmia and stroke in ES, but also a concurrent risk of pulmonary haemorrhage and haemoptysis. Our own empirical approach towards the adult with ES has been to advocate oral anticoagulation with warfarin and engage haematology colleagues to assist with titration of international normalised ratio but with the specific remark not to interfere with high levels of haemoglobin/haematocrit. There may be a role for non-vitamin K antagonist oral anticoagulants in ES, but data are currently lacking.

Baseline assessment of haemodynamics with cardiac catheterisation, allowing for estimation of PVR and AVT with nitric oxide, should be offered to all patients with PAH–CHD as it confirms the diagnosis but also provides prognostic information for patients with ES. A positive AVT, however, is not an indication for therapy with calcium channel blockers (which is standard practice for patients with idiopathic PAH), as in ES they may cause significant peripheral vasodilation, increased right-to-left shunting (hypoxia), syncope and sudden cardiac death.23

There has been recent and important advance on PAH–DTT in ES. PAH–DTT, in general, targets three main pathophysiological pathways: the endothelin pathway using endothelin-1 receptor antagonists (ERAs), nitric oxide pathway using phosphodiesterase type-5 inhibitors (PDE-5i) and potentially guanylate cyclase inhibitors, and prostacyclin pathway using prostanoids.

ERAs act by blocking endothelin-1 receptors whose activation results in elevation of intracellular-free calcium leading to powerful vasoconstriction and contributes to pathogenesis of PAH by inducing proliferation, fibrosis and inflammation.24 Current ERAs include bosentan, macitentan (affecting both endothelin A and B receptors) and ambrisentan (selectively binding only to endothelin A receptors). BREATHE-5 was a pivotal study to show safety and efficacy of bosentan therapy on improving PVR index and 6MWD after 16 weeks of therapy versus placebo in ES.25 26 A recent randomised controlled study in patients with ES (MAESTRO) showed no safety concerns regarding macitentan, but was neutral in its primary endpoint of 6MWD (largely due to the improved 6MWD observed in the placebo arm).7 Bosentan therapy was recommended in the latest guidelines for patients with ES in functional class III (class IB).9 We submit that a new consensus is now due, given the progressive nature of ES, the improved prognostication and accumulated experience from PAH–DTT. Functional class II patients with ES should also be considered for PAH–DTT given that we know of their existence and that they also have considerable mortality, although half compared with mortality of class III patients with ES at baseline.12 Patients deteriorating on monotherapy should be offered sequential combination therapy, whereas patients with severe form of ES in functional class III/IV should be tried on upfront combination therapy, including intravenous prostanoids.27 28

Sildenafil and tadalafil represent selective PDE-5i that target the nitric oxide pathway through prevention of the breakdown of cyclic GMP in the endothelium of the pulmonary vasculature. This in turn decreases the availability of calcium for vascular smooth muscle cell contraction. PDE-5i have also shown beneficial effect in patients with ES in terms of symptoms, oxygen saturation and pulmonary haemodynamics,29 30 although the evidence is somewhat less robust compared with bosentan. PDE-5i are, nevertheless, commonly used in ES often as a second-line PAH–DTT.2 9 Riociguat, a soluble guanylate cyclase stimulator, also works on the nitric oxide pathway and enhances the production of cyclic GMP, but data in ES are currently lacking.

Synthetic prostacyclin epoprostenol and prostacyclin analogues iloprost, treprostinil, beraprost and selexipag activate prostanoid receptors on endothelial cells within the pulmonary vasculature causing vasodilatation. Beneficial effects on haemodynamics, functional class and exercise capacity have been reported in ES; however, their route of administration (intravenous epoprostenol, inhaled iloprost, subcutaneous treprostinil) makes them unattractive for first-line therapy in ES and has confined their use mostly to markedly limited patients who are already on combination oral DTT. Intravenous prostanoids should also be employed in pregnancy, where aggressive PAH–DTT approach, combined with early delivery of the fetus, seems to convey a better outcome (ERAs cannot be used because of potential teratogenicity).31 It should, however, be noted that pregnancy is contraindicated in ES and female patients and their families should be made aware of the major risk of pregnancy (mortality as high as 1:3 or 1:4, and considerable morbidity) (table 1). Beraprost and selexipag represent oral prostacyclin analogues, with positive results in the treatment of PAH,32–34 and thus merit independent assessment in PAH–CHD, including patients with ES.

PAH with persistent systemic to pulmonary shunts

(PAH with variable severity of PVD)

This subgroup includes adult patients with large intracardiac or extracardiac shunts who have PAH but have not developed ES (figure 1B). Severity of PAH and in particular of PVR determines whether shunt closure may be beneficial. As most large post-tricuspid shunts (ventricular septal defect, patent ductus arteriosus) develop pulmonary vascular disease if not closed early in childhood, this subgroup concerns mostly patients with atrial septal defects. It has to be emphasised that PH and PVD are not the same thing, as patients from this subgroup may have PH but not elevated, or just mildly elevated PVR, that is, mPAP >25 mm Hg, with normal or near-normal PVR (common scenario among older patients with atrial septal defect, compliant pulmonary vascular bed and large left to right shunts; typically, such patients would be suitable for and benefit from defect closure). It is, therefore, paramount that formal assessment of PVR during cardiac catheterisation is performed in these patients to guide management, in addition to prognostic information derived at the time from AVT as discussed in the ES section. Defect closure, crucially, should not be based on feasibility of the procedure but on long-term benefits. Current guidelines regarding defect closure in the setting of PAH are conservative2 9 35 based on the uncertain long-term benefits of late defect closure in the presence of PVD. Retrospective studies have suggested that defect closure in the presence of PAH carries the worst prognosis among all types of PAH–CHD, including ES.36 37 Closure of the defect is contraindicated when PVR index >8 WU m2, but permitted when PVR index is <4 WU m2.2 Patients with borderline PVR index between 4 and 8 WU m2 (in the ‘grey zone’) should be assessed in tertiary centres with expertise in PAH–CHD. Controversy remains on the merits of closing such defects from the ‘grey zone’ when patients respond positively to PAH–DTT. We could make the case for closing such defects in the presence of a persistent large left to right shunt, when PVR falls significantly during AVT and/or after a positive trial period of PAH–DTT, always in the absence of any cyanosis (at rest or during exercise).38–40 Fenestrated closure under these circumstances may also be considered, and patients must remain under tertiary care and should continue with their pre-closure PAH–DTT medication. Repeated cardiac catheterisation for reassessing PVR 6 to 12 months after closure is advisable, if not necessary, to establish a new baseline for the individual patient and guide the ongoing need for PAH–DTT (table 1). A national/international registry with longer-term outcome of patients from the ‘grey zone’ is warranted as we are currently lacking late outcome data of patients with PAH–CHD and large systemic to pulmonary shunts in the DTT era, whether they have been managed conservatively or with treat-and-repair approach. New biomarkers such as levels of circulating endothelial cells in the peripheral blood41 and future genetic advances may shed additional light on patients likely to benefit long term from late defect closure.

Table 1

Management of pulmonary arterial hypertension in adult congenital heart disease

PAH with small or coincidental defects

(PAH with PVD)

Small defects (in general, ventricular septal defect <1 cm or atrial septal defect <2 cm) are not considered to be primarily responsible for developing PAH and may even be beneficial in patients with PAH by allowing occasional right-to-left shunting, and thus decompressing the right ventricle and maintaining systemic cardiac output (figure 1C). Patients with PAH and small or coincidental defects have similar pathophysiology and clinical phenotype as patients with idiopathic PAH, and should therefore be treated with PAH–DTT, considered for anticoagulation, whereas the defect should not be closed (table 1).

PAH after defect closure (PAH with PVD)

This is a concerning subgroup of patients presenting with PAH often many years after defect closure (figure 1D); reasons for this are not clear and may include late diagnosis and closure, and/or possibly genetic, or other predisposition to PAH. The denominator for this subgroup is uncertain, as many of the patients who underwent defect closure in childhood have been lost to follow-up. In a recent report of 1103 shunt patients, the cumulative incidence of PAH ranged from 2.1% directly after closure to more than 15% 50 years later. Moreover, the 10-year survival was significantly worse in patients who developed PAH after closure (OR 3.5).42 Given the common association of PAH and CHD and its adverse impact of the former on quality of life and prognosis on affected patients, we submit that all patients with CHD and previous shunt closure should undergo at least one thorough assessment in a specialist centre during teenage or adulthood years. Apart from excluding PAH, other key information on overall prognosis and lifestyle issues, including family planning and exercise, can be provided at the same time. The clinical course for patients with PAH and closed shunts is very similar to idiopathic PAH, and these patients should be treated with DTT as dictated by severity and course of PAH (table 1).

Fontan operation (PVD possible without PAH)

The Fontan operation, performed in patients with complex CHD unsuitable for biventricular repair, separates the systemic from pulmonary circulation by directing systemic venous return to the pulmonary artery, without an interposed subpulmonary ventricle (figure 1E). Fontan is a radical but palliative procedure associated with many late complications and limited surgical alternatives.43 Pulmonary blood flow and cardiac output are dependent on low PVR particularly as there is no subpulmonary ventricle. In the absence of effective ventricle, even a slight increase in PVR could result in low cardiac output with extravascular fluid accumulation and hepatic congestion associated with increased mortality. Indeed, recent data indicate unique pattern of adverse pulmonary vascular remodelling in long-term Fontan circulation44 that may be contributory to these haemodynamic changes and adverse outcome. The limited evidence at present suggests that Fontan patients with elevated PVR (but no PAH, as conventionally defined) might benefit from PAH–DTT.45–50 The consensus on selection criteria from the Fontan cohort that might benefit from DTT has not yet been reached, thus recommendations for general PAH–DTT cannot be made before more data become available.


PAH is relatively common in the context of CHD adversely affecting quality of life and survival. Sufficient progress has been made at the extreme end of the spectrum, namely ES; these patients should be proactively treated with PAH–DTT in tertiary centres employing a goal-oriented approach. Patients with PAH and small defects, or previous defect closure should also receive PAH–DTT. More data are required for patients with PAH and systemic to pulmonary shunts, and for patients with a Fontan operation before general recommendations for PAH–DTT use can be made. Late defect closure in PAH–CHD is generally contraindicated, with the exception of selected patients with PAH–CHD with significant systemic to pulmonary shunts, following detailed assessment in tertiary centres, inclusive of PVR and AVT. Novel biomarkers and genetic advances may identify patients with PAH–CHD who may benefit from late defect closure and/or those at risk of developing PAH despite timely, childhood defect closure. Lifelong tertiary care for all patients with known PAH–CHD, and ongoing vigilance for late development of PAH for the remainder, with additional data derived from registries and controlled studies in PAH–CHD are all warranted.



  • Contributors Both authors contributed equally to the planning, conduct and reporting of the work described in the article; MAG assumes responsibility for the overall content as guarantor.

  • Funding MB is the recipient of the International Training and Research Fellowship EMAH Stiftung Karla VÖLLM, Krefeld, Germany. MAG has received support from the British Heart Foundation, London, UK, and Actelion, Allschwil, Switzerland.

  • Competing interests MAG is a steering committee member for Actelion Pharmaceuticals and has received unrestricted educational grants from Actelion Pharmaceuticals, Pfizer and GlaxoSmithKline.

  • Patient consent Not required.

  • Provenance and peer review Commissioned; externally peer reviewed.