31 Pulmonary hypertension

Pulmonary hypertension (PH) is defined haemodynamically as an increase in mean pulmonary arterial pressure (mPAP) of at least 25 mmHg at rest, as assessed by right heart catheterization (RHC). At the 2008 World Symposium on Pulmonary Hypertension in Dana Point, mPAP of 20–25 mmHg was considered abnormal, and the term ‘pre-PH’ was proposed to signify this.¹ Previously, the definition of PH included exercise-induced PH (mPAP ≥30 mmHg on exercise).² However, the data for exercise-induced PH is limited, and the recent reclassification of PH does not include exercise-induced PH in its definition.

PH is a severe, progressive disorder, which may be idiopathic, or related to a spectrum of medical disorders. The clinical classification of PH was revised at the Dana Point symposium (Table 31.1).¹ PH is classified into five groups with specific pathological, pathophysiological, and therapeutic characteristics. PH associated with left heart disease was identified as a distinct subgroup, with subcategories including PH related to systolic and diastolic dysfunction and valvular heart disease

PH may be considered pre- or postcapillary (Table 31.2).³ Precapillary PH is defined as mPAP 25 mmHg or more, with pulmonary capillary wedge pressure (PCWP) 15 mmHg or less and a normal or reduced cardiac output (CO). PH associated with left heart disease is considered postcapillary, and defined on haemodynamic terms as having mPAP 25 mmHg or more, PCWP greater than 15 mmHg, and a normal or reduced CO. Furthermore, postcapillary PH can be considered in two subgroups: passive PH (in which the transpulmonary gradient, TPG ≤12 mmHg) and reactive or disproportionate PH (in which the TPG 〉12 mmHg).

The transpulmonary gradient is the drop from mPAP to left atrial (or pulmonary capillary wedge) pressure, and is measured at right heart catheterization. A low TPG in the face of PH implies that the pulmonary circulation is likely to be structurally normal and that the PH is purely secondary to the back pressure effect of raised left heart filling pressure; whereas a high TPG implies that secondary changes have occurred to the pulmonary vasculature.

The prevalence of PH in left heart disease increases with the severity of functional impairment. Forty to seventy per cent of patients with isolated diastolic dysfunction, and up to 60% of those with left ventricular systolic dysfunction (LVSD) have PH at presentation.^3,4 In patients with valvular heart disease, the prevalence of PH increases with the severity of the valvular defect, and is reported in up to 100% of cases with severe left heart valvular disease.⁵

In general when PH develops in association with left heart disease, patients have a worse prognosis. In a chronic heart failure (HF) study, the mortality rate was higher in patients with PH (28%) than in those without PH (17%) at 28 months.⁶ PH (pulmonary vascular resistance 〉6–8 Wood’s units; TPG 〉16 mmHg) is also associated with an increased postoperative risk of right HF after cardiac transplantation.⁷

The histopathology of PH associated with left heart disease is characterized by pulmonary venous distension and thickening and pulmonary capillary and lymphatic dilatation, as well as interstitial oedema and alveolar haemorrhage. Over time, pulmonary vascular remodelling may lead to the classic changes of the distal pulmonary vasculature seen in pulmonary arterial hypertension (PAH) including medial hypertrophy, intimal proliferation and fibrosis, and adventitial changes (Fig. 31.1).^8,9

The pathophysiology of PH related to left heart disease is complex, and may be best considered as a combination of passive and active mechanisms. Passive pulmonary venous stretch resulting from backward transmission of elevated left heart filling pressures is important (Table 31.2). Purely passive PH is usually reversible with acute vasodilator testing, and may be more responsive to chronic HF treatment.¹⁰

However, in some cases there is an additional active component, leading to PH ‘out of proportion’ to the underlying left heart disease. In such patients, mPAP is disproportionately elevated above left heart filling pressures, leading to an elevation in TPG and a consequent increase in pulmonary vascular resistance (PVR). This elevation in PVR is attributed to an increase in the vasomotor tone of the distal pulmonary vasculature from pulmonary vasoconstriction and pulmonary vascular remodelling.¹¹

Pulmonary vascular remodelling results from increase in cell number in all components of the vessel wall. This leads to a narrowed lumen and increase in vascular resistance proportional to radius⁴. Although the formation of plexiform lesions, characteristic of severe PAH, are not generally seen in PH associated with left heart disease, true remodelling of resistance vessels has been reported. These fixed, obstructive lesions lead to PH that is generally unresponsive to acute vasodilator testing.

It is not clear why pulmonary vascular remodelling and disproportionate PH develop in only some chronic HF patients. It is possible that pulmonary vascular remodelling develops in response to chronic activation of stretch receptors located in the left atrium and pulmonary veins. Patients with congestive cardiac failure also have endothelial dysfunction, and raised serum endothelin-1 (ET-1) levels, a pathological mediator known to be important in vascular remodelling in PAH.^{12  13}

Other potential pathophysiological mechanisms include hypoxic pulmonary vasoconstriction and in situ thrombosis.

Recognition of PH may be delayed, as it is often masked by the clinical picture of the underlying left heart disease. Symptoms including dyspnoea and fatigue are common to both PH and left heart disease. Physical signs reflective of PH (such as a loud pulmonary component of the second heart sound) are often difficult to hear. Signs of right HF (including peripheral oedema) are generally late findings.

It may be difficult to distinguish between PH associated with diastolic dysfunction and true PAH. Left ventricular diastolic dysfunction should be suspected (rather than PAH) in the presence of one or more of a number of clinical or echocardiographic risk factors (Box 31.1).^3,14 However, a definite diagnosis of PH associated with left heart disease is only possible following RHC.

RHC remains the gold standard for the diagnosis of PH. It allows the measurement of mPAP, PCWP, and CO, and subsequent calculation of TPG and PVR. These data are important for the diagnosis of PH associated with left heart disease, and its subdivision into passive and active categories (Table 31.2).

The role of acute vasodilator testing is not clear in PH due to left heart disease. In general, patients with passive PH will have acute vasodilator reversibility, whereas patients with disproportionate PH will have fixed disease, unresponsive to vasodilator testing.¹⁰ At present, vasoreactivity testing is recommended for patients referred for cardiac transplantation.¹⁵ In these patients, lack of responseto vasodilator testing is associated with increased risk of post-transplantation right ventricular failure and early mortality.¹⁶

A variety of agents are used for vasoreactivity testing, including nitrous oxide, phosphodiesterase-5 (PDE-5) inhbitors, prostanoids, and other vasodilators.

In the diagnosis of left ventricular diastolic dysfunction, the role of exercise testing, or volume challenge, remains unclear (Fig. 31.2). However, in some patients with apparently ‘normal’ left ventricular filling pressures (often following diuretic therapy and fluid restriction prior to RHC), an increase in pressure following exercise or fluid challenge may unveil occult left ventricular dysfunction.¹⁴

RHC is a moderately invasive procedure, and so a number of noninvasive investigations are often used in the assessment for PH.

Continuous Doppler flow echocardiography is the best noninvasive tool for the assessment of PH associated with left heart disease. Echocardiography allows estimation of the systolic PAP from the maximal velocity of the tricuspid regurgitation jet (in the presence of tricuspid regurgitation). However, echocardiography also complements RHC as it provides other, structural and functional cardiac information.

Left ventricular systolic function is assessed by the left ventricular ejection fraction (LVEF). In contrast, left ventricular diastolic dysfunction is more difficult to assess, although there are several echocardiographic findings that should arouse suspicion for diastolic dysfunction (Table 31.3).^3,14 Left ventricular filling pressures may be estimated by the ratio of mitral valve flow velocity (E) divided by early diastolic lengthening velocities (E′). However, accurate evaluation of left ventricular filling pressures requires RHC.

B-type natriuretic peptide (BNP) is released in response to atrial and ventricular wall stretch. Plasma BNP concentrations are elevated in PAH, and are associated with poorer outcomes.¹⁷ However, in BNP concentrations are also elevated in left heart disease. Therefore, it is not clear whether increased BNP levels are useful in identification or prognostic evaluation of PH in left heart disease.

Cardiopulmonary exercise tests (CPET) may be useful to identify early, or exercise-induced, PH as a cause of exercise-limitation in patients with underlying left heart disease. However, patients are often unable to perform a CPET, and six-minute walk testing is used instead. In patients with PAH, lower six-minute walk test distance portends a poorer prognosis.^18,19

Cardiac MRI (CMR) provides the most accurate measurement of right ventricular mass and ejection fraction. Its role in the diagnosis of PH in this patient group is promising, but needs further study.

The main and segmental pulmonary artery size may be measured on CT scanning. There are few data to support the use of CT scanning in the diagnosis of PH in patients with left heart disease.

Management of PH related to left heart disease centres on the treatment of the underlying disorder with medications such as diuretics, angiotensin converting enzyme (ACE) inhibitors, β-adrenoreceptor blockers, or other interventions.¹⁰ In patients with valvular heart disease, corrective valve surgery is recommended, and is usually associated with clinical improvement, and resolution of the PH.^5,20 The improvement may take several weeks to months, and may be incomplete, due to the fixed obstructive changes of pulmonary vascular remodelling.

Supplemental oxygen is recommended to reverse resting hypoxaemia in appropriate patients. Full assessment and treatment of comorbidities (including pulmonary emboli and obstructive sleep apnoea) is important. Early referral for cardiac transplantation is essential for selected cases. Patients with endstage cardiac failure and PH may be candidates for cardiac transplantation (in milder, largely reversible forms of PH), or heart–lung transplantation when PH is severe and/or fixed.

Specific PH therapy is not routinely recommended for patients with PH secondary to heart disease, as there have been no successful placebo-controlled trials of disease-targeted PH therapies in this patient group.^3,21 There is a potential risk of increasing intrapulmonary shunting and ventilation–perfusion mismatch, particularly with pulmonary vasodilators. However, the risk may be ameliorated by treatment to reduce left ventricular filling pressures. A number of pharmceutical trials have been performed, and the results are summarized below.

Nitric oxide (NO) and prostacyclin are potent pulmonary vasodilators. Both NO and prostacyclin have been shown to improve pulmonary haemodynamics acutely, with decreased PVR, and pulmonary arterial pressures.^22,–24 In patients with left heart disease, inhaled NO leads to a decrease in PVR, with an increase in left ventricular filling pressures. However, an international trial using intravenous epoprostenol in patients with severe left ventricular failure (FIRST trial) was terminated prematurely because of a trend towards increased mortality in patients receiving epoprostenol.²⁵

The endothelin system is activated in chronic HF, and elevated plasma ET-1 (and its precursor big-ET1) levels are associated with increased morbidity and mortality in patients with left heart disease.^26,27 Plasma ET-1 levels correlate with the symptomatic and haemodynamic severity of the HF.^28,29 Intravenous infusions of ET-1 lead to increased systemic vascular resistance, and decreased cardiac index.³⁰ ET-1 acts via two receptor subtypes (ET_A and ET_B receptors). Both receptor subtypes lead to pulmonary vasoconstriction, and vascular smooth muscle proliferation. Activation of ET_B receptors also leads to vasodilation (via NO and prostacyclin), antiproliferative and antithrombotic effects, and clearance of ET-1. It is unclear whether selective ETA blockade is advantageous in chronic HF.

In preliminary small studies using the endothelin-1 receptor antagonists (ETRAs) bosentan, darusentan, and BQ-123, there was an acute improvement in mPAP, right atrial pressure, PCWP, and CO.³¹–³³ Intravenous bosentan led to acute improvement in pulmonary haemodynamics in 24 patients with chronic HF.²⁹ Acute treatment with the ETA receptor antagonist sitaxentan led to significant decreases in pulmonary arterial pressures and PVR.³⁴ In a large placebo-controlled study of patients with acute decompensated HF, intravenous tezosentan led to an acute reduction in left ventricular filling pressures, and increased cardiac index.³⁵

Longer-term studies have not confirmed a benefit for the use of ETRAs in patients with PH associated with left heart disease. The Research on Endothelin Antagonists in Chronic Heart Failure (REACH-1) study, a placebo-controlled study using bosentan, was terminated early because of elevated hepatic transaminase levels.³⁶ However, there was a trend towards reduced HF mortality and morbidity with bosentan. The Endothelin Antagonist Bosentan or Lowering Cardiac Events in Heart Failure (ENABLE 1 and 2) studies did not show any benefit of bosentan over placebo with regard to morbidity or mortality.³⁷ In a further study with enrasentan (Enrasentan Cooperative Randomized Evaluation, ENCOR study), no benefit was shown.³⁸ The Endothelin A Receptor Antagonist Trial in Heart Failure (EARTH) study also showed no difference of left ventricular end-systolic volume on CMR imaging in patients treated with darusentan or placebo.³⁹ The results of the ETRA trials in cardiac failure patients are summarized in Table 31.3.

It is somewhat difficult to reconcile the negative results in the trials described above with the positive findings in acute studies. One explanation for the disparity is that patients who have PH disproportionate to their underlying left heart disease are more likely to benefit from specific PH therapy, and many of the trials discussed included all patients with congestive HF, not only patients specifically with PH who might be expected to benefit. Alternatively, it is possible that the acute haemodynamic response to ETRAs does not lead to a sustained clinical benefit, or that any benefit is masked by the use of other HF therapies.^40,41 Currently, the use of ETRAs is not recommended for the treatment of PH associated with left heart disease.

In PAH, NO mediates pulmonary vasodilation via cyclic guanosine monophosphate (cGMP), which is degraded by phosphodiesterases (particularly PDE-5). In PH, PDE-5 is up-regulated. PDE-5 inhibitors, such as sildenafil, cause pulmonary vasodilation, antiproliferative actions on pulmonary vascular smooth muscle cells, and protection from ischaemic reperfusion injury. In PAH, sildenfil is associated with increased six-minute walk test distance, and improved pulmonary haemodynamics at 12 weeks, with a sustained clinical benefit at 12 months.⁴²

In chronic HF, sildenafil acutely lowers PVR and pulmonary arterial pressures, and improves endothelium-dependent flow-mediated pulmonary vasodilation.^43,–45 Longer-term studies have shown that sildenafil is associated with improvement in exercise capacity, and CO and skeletal muscle blood flow during exercise.^46,47 A study of 34 patients with symptomatic HF and PH, showed an improvement in six-minute walk test distance, quality of life, exercise capacity, and fewer hospitalizations for HF, with sildenafil compared to placebo at 12 weeks. Despite these promising results, the routine use of sildenafil in patients with PH associated with left heart disease cannot be recommended. There is an urgent need for further longer-term trials, particularly in patients with PH disproportionate to the underlying heart disease.

To date, there have been no trials specifically addressing patients with left ventricular diastolic HF, in which PH is common. The PhosphodiesteRasE-5 Inhibition to Improve Quality of Life And EXercise Capacity in Diastolic Heart Failure (RELAX) study, a double-blind, placebo-controlled clinical trial using sildenafil, is currently under way. The primary endpoint of the study is the change in exercise capacity as assessed by peak VO₂ after 24 weeks of treatment with sildenafil or placebo.

PH is common in patients with left heart disease, and is associated with increased morbidity and mortality. The underlying pathophysiological mechanisms are complex, including both passive and active components. PH disproportionate to the underlying left heart disease may be attributable to reactive pulmonary vascular remodelling. Left ventricular diastolic dysfunction may be difficult to diagnose, but is best assessed with echocardiography and RHC. Management of PH is focused on treatment of the underlying left heart disease, and reversal of hypoxaemia. There is no supporting evidence for the routine use of specific PH therapies at present. However, there is some suggestion that PDE-5 inhibitors may be useful, and their safety and efficacy needs to be formally evaluated in controlled trials before further recommendations are made.