Definitions and classification

The recent European Society of Cardiology/European Respiratory Society (ESC/ERS) guidelines have defined pulmonary hypertension (PH) as the elevation of mean pulmonary artery pressure (mPAP) ≥ 25 mmHg at rest, as assessed by right heart catheterization (RHC) . Normal pulmonary artery wedge pressure (PAWP) is ≤15 mmHg. Postcapillary PH is thus defined by mPAP ≥ 25 mmHg at rest and PAWP > 15 mmHg. On the other hand, precapillary PH is defined by mPAP ≥ 25 mmHg, PAWP ≤ 15 mmHg . In some conditions, chronic elevation of the left-sided filling pressure may cause excess vasoconstriction, with or without vascular remodelling, thus leading to elevated pulmonary vascular resistance (PVR) > 3 WU. This condition has been described as “reactive”, “out-of-proportion” or “mixed” PH, leading to a “disproportionate” increase in pulmonary artery pressure (PAP) . For a long time, a “transpulmonary pressure gradient” > 12 mmHg (i.e. the difference between mPAP and PAWP) has been used to describe this feature, but this gradient may be influenced by volume load and cardiac function, and does not prognosticate outcome in PH . The recent ESC/ERS guidelines favour measuring the diastolic pressure gradient (i.e. the difference between diastolic PAP and PAWP), which may be less dependent upon stroke volume and loading conditions . In healthy subjects, this gradient is <5 mmHg . Postcapillary PH is thus further classified as isolated postcapillary PH if the diastolic pulmonary gradient is <7 mmHg and/or the PVR is ≤3 WU, or as combined post- and precapillary PH, if the diastolic pulmonary gradient is ≥7 mmHg and/or the PVR is >3 WU ( Table 1 ).

The updated classification categorizes four types of PH associated with left heart disease (LHD), according to their origin: PH due to heart failure with reduced ejection fraction (HFrEF); PH due to heart failure with preserved ejection fraction (HFpEF); PH due to left-sided valvular heart disease; and PH due to congenital/acquired left heart inflow/outflow tract obstruction and congenital cardiomyopathies ( Fig. 1 ).

Causes of pulmonary hypertension due to left heart disease (PH-LHD). The fourth common etiology of PH is PH due to congenital/acquired left heart inflow/outflow tract obstruction and congenital cardiomyopathies (not shown). HFpEF: heart failure with preserved ejection fraction; HFrEF: heart failure with reduced ejection fraction.

In the setting of mPAP ≥ 25 mmHg at rest, measuring the precise value of PAWP is of major importance to discriminate precapillary PH from group 2 PH. Differentiating group 1 from group 2 patients may be difficult, and an exercise or a saline loading test may be used to unmask venous PH. A recent study has shown that exercise testing is more sensitive than saline loading to detect haemodynamic changes indicative of HFpEF . Combining mPAP > 30 mmHg and PVR > 3 mmHg*min per L is superior to mPAP > 30 mmHg alone for defining a pathological haemodynamic response of the pulmonary circulation during exercise . However, in exercising patients, there are no reliable data that define which level of exercise-induced changes in mPAP has prognostic implications; thus, a disease entity “PH on exercise” currently cannot be defined and should not be used .

Epidemiology and prognosis

Knowing the prevalence of PH in the population of patients with heart failure is not easy . The progression of heart failure is frequently associated with PH and right ventricular (RV) dysfunction, which is associated with a poor prognosis .

The prevalence of PH and RV failure in LHD varies depending on the population studied, the method used to diagnose PH (echocardiography or RHC) and the haemodynamic criteria used to define PH. Moreover, day-to-day variation in PAP may be observed in group 2 patients, depending on volume load.

In ambulatory outpatients with HFrEF, PH was found in 73% of patients referred for RHC . In patients with acute decompensated heart failure, PH was diagnosed in 25–75% of patients . PH seems to occur even more frequently in HFpEF. In three studies of patients with HFpEF, PH was present in 36%, 52% and 83% . While PH was most commonly described with mitral stenosis in the past, cardiologists are now increasingly facing the discovery of PH, given the increasing prevalence of HFpEF related to population ageing and co-morbidities. In parallel, cardiologists’ awareness of the measurement of pulmonary pressure has grown, and screening campaigns in pharmacovigilance of medicinal products (e.g. benfluorex or some cancer drugs) have been developed.

Regarding the haemodynamic characteristics of PH in HFrEF, a study of 320 patients found that PVR was normal (<1.5 WU) in 28%, mildly elevated (1.5–2.49 WU) in 36%, moderately elevated (2.5–3.49 WU) in 17% and severely elevated (>3.5 WU) in 19% . When comparing HFrEF and HFpEF, baseline stroke volume and cardiac output (CO) are higher in HFpEF than in HFrEF, while mPAP, PAWP and PVR are similar .

Prognostic significance of PH

Whereas the prevalence of PH varies greatly, depending on the study, all cohorts of patients show that PH is associated with increased mortality ( Fig. 2 ). In a recent retrospective analysis of 21,727 veterans undergoing RHC, a continuum of risk according to mPAP level was documented, and the 21–24 mmHg mPAP range was associated with increased mortality and hospitalization . In both HFrEF and HFpEF, PH seems to be a reliable prognostic marker . However, it is difficult to argue that PH is an independent marker of poor prognosis, because mPAP integrates several components (e.g. blood volume status, left ventricular [LV] compliance, mitral valve function), and is also influenced by remodelling and loss of compliance of the left atrium, pulmonary vascular remodelling and renal function – all factors known to affect prognosis . Finally, PH prognosis is more linked to RV function than to the level of PAP . In the late stages of heart failure, mPAP may decrease because of RV dysfunction, and carries an adverse short-term prognosis . RV dysfunction is common, is associated with evidence of more advanced heart failure and is predictive of poorer outcome .

Pulmonary hypertension (PH), right ventricular (RV) function and clinical outcome in acute decompensated heart failure.

Adapted from .

As far as the question of the prognostic value of the diastolic gradient is concerned, studies are not unanimous , and these points deserve further investigation.

Physiopathology

Pulmonary circulation is a low-pressure, low-resistance and high-compliance circulation. Left atrial (LA) pressure is estimated by PAWP. Under physiological conditions, mean PAP is defined as follows: mPAP = PAWP + PVR * CO; and thus PVR is defined as PVR = (mPAP − PAWP)/CO.

In healthy subjects, any increase in CO (effort, stress), is accompanied by an only moderate increase in pulmonary pressure, resulting from a decrease in PVR . This adaptation of PVR is partly caused by dilatation of the pulmonary vascular bed. Endothelial production of nitric oxide is involved in the flow-dependent vasodilation . In the upright position, another mechanism is the large capacity of recruiting blood vessels not perfused at rest . All these mechanisms contribute to optimize the ventilation/perfusion ratio on exercise.

The first event in the development of PH due to LHD (PH-LHD) is the increase in left heart filling pressures, leading to pulmonary venous hypertension . Functional mitral regurgitation, LV remodelling, increased LV stiffness and LV dysfunction will result in chronic elevation of LA pressure; this contributes to increase LA size, and reduce LA compliance and contractility. Systolic and diastolic LA functions are altered, and the left atrium cannot play its role as a buffer reservoir before the pulmonary circuit. The elevation of the left filling pressure is transmitted to PAP in a nearly 1:1 proportion, thus leading to increased pulmonary pressure, especially during exercise . Variation in congestive state also induces mPAP variation.

At the pulmonary capillary level, endothelial dysfunction is characterized by decreased production of nitric oxide, overproduction of endothelin-1, activation of the renin–angiotensin–aldosterone system and neurogenic activation; this leads to pulmonary artery vasoconstriction and PVR elevation. Next, elevated PAP leads to vascular damage, with pathological remodelling of the pulmonary arterioles, such as thickening of the alveolar-capillary membrane, medial hypertrophy and neointimal proliferation , whereas “plexiform lesions” that are pathognomonic of pulmonary arterial hypertension are not found .

Finally, RV is highly sensitive to slight increases in afterload, i.e. to increased PVR (steady afterload), and to PA stiffening (pulsatile load) ; this progressively leads to decreased RV stroke volume. The adaptation of RV to increased afterload varies over time and depending upon patients’ phenotypes. First, the remodelling is characterized by RV hypertrophy, to normalize RV wall stress (Laplace law, and thus RV myocardial oxygen demand). Then, dilatation, spherization and functional tricuspid regurgitation may be observed. These changes mark a pejorative milestone in the evolution of the disease. Chronic pressure elevation in the superior and inferior vena cava progressively leads to congestion, including impairment of renal function (cardiorenal syndrome) , and of hepatic, splanchnic and gut function.

It is noteworthy that some patients will develop severe PH and RV dysfunction whereas others will not. The factors underlying this susceptibility are unknown, but genetic and environmental contexts, co-morbidities and the period since the beginning of LHD are potential contributing factors . Studies are currently in progress to better understand this point.