Worldwide Epidemiology of Pulmonary Hypertension Associated With Left Heart Disease

Abstract

Globally, around 1 in 2 patients with left heart disease (LHD) has pulmonary hypertension (PH). The prevalence of PH associated with heart failure, mitral regurgitation, aortic valve disease, hypertrophic cardiomyopathy, and cardiac amyloidosis has mainly been investigated in high-income countries, especially the United States and Europe, by transthoracic echocardiography without right heart catheterization confirmation. It is estimated that around 80% of patients with hypertrophic cardiomyopathy or cardiac amyloidosis, around 65% of patients with heart failure or mitral regurgitation, and around 40% of patients with aortic valve disease have PH. In contrast, data about PH associated with mitral stenosis and rheumatic heart disease are primarily from low-income countries. The prevalence is around 80% in mitral stenosis—as determined by right heart catheterization in candidates for surgery or balloon valvuloplasty—and around 45% in rheumatic heart disease based on transthoracic echocardiography screening. PH associated with LHD (PH-LHD) clinical trials have focused on PH secondary to heart failure and have most often been performed in the United States and Europe (sites from these countries in 53.3% and 43.3% of the trials, respectively). The heterogeneous epidemiology of PH-LHD may reflect actual differences in the frequency of the underlying types of LHD, but it may also be the consequence of detection and reporting biases. Regional health authority choices, diagnostic capacity, and socio-economic factors likely influence the selection of the sites of PH-LHD trials. Efforts are needed to better characterize the worldwide burden of PH-LHD and to conduct truly global clinical trials.

1. Introduction

Pulmonary hypertension (PH) is a common complication of left heart disease (LHD), ensuing from the backward transmission of elevated left-heart filling pressures to the pulmonary circulation.^[1] A subset of patients develops functional and structural alterations of pulmonary arterioles and venules, which heighten pulmonary vascular resistance (PVR) and compound PH associated with LHD (PH-LHD) [Figure 1].^[2] The pathogenesis of pulmonary vascular remodeling in PH-LHD includes defective nitric oxide signaling and enhanced endothelin-1 expression and activity, but it is only partially understood.^[3,4]

Figure 1.

Pathophysiology of PH associated with left heart disease. The insets on both sides contain representative transthoracic echocardiography images in heart failure with preserved ejection fraction (A), hypertrophic cardiomyopathy (B), cardiac amyloidosis (C), heart failure with reduced ejection fraction (D), mitral regurgitation (E), mitral stenosis (F), aortic stenosis (G), and aortic regurgitation (H). Cpc-PH: Combined post- and pre-capillary pulmonary hypertension; Ipc-PH: Isolated post-capillary pulmonary hypertension; LA: Left atrial; LV: Left ventricular; PAP: Pulmonary artery pressure; PAWP: Pulmonary artery wedge pressure; PH: Pulmonary hypertension; PVR: Pulmonary vascular resistance; RV: Right ventricular.

The hemodynamic diagnosis of PH-LHD requires that the mean pulmonary arterial pressure (mPAP) is > 20 mmHg and pulmonary artery wedge pressure is > 15 mmHg, as measured by right heart catheterization (RHC) at rest.^[5] The mPAP threshold was recently lowered from the previous one of 25 mmHg, since the current cut-off coincides with the upper limit of the mPAP range in healthy individuals.^[6] Moreover, it is in agreement with the evidence that a mPAP > 20 mmHg increases the risk of mortality.^[7,8]

PH-LHD is further categorized into 2 hemodynamic phenotypes: (1) isolated post-capillary PH (Ipc-PH) when PVR is ≤ 2 Wood units (WU) and (2) combined post- and pre-capillary PH (Cpc-PH) in the presence of PVR > 2 WU.^[5] Ipc-PH corresponds to stages of disease in which retrograde propagation of high left atrial pressure mostly accounts for the increase in pulmonary arterial pressure (PAP), whereas Cpc-PH reflects the occurrence of pulmonary vascular remodeling.^[9] The PVR threshold for the subclassification of PH-LHD in Ipc-PH and Cpc-PH was also recently reduced from 3 to 2 WU because large retrospective analysis highlighted that prognosis worsens starting from PVR beyond 2 WU.^[8,10]

The transpulmonary gradient (TPG; calculated as: mPAP − PAWP) and the diastolic pressure gradient (DPG; calculated as: diastolic PAP − PAWP) are other RHC-derived measures that have been proposed to distinguish between Ipc-PH and Cpc-PH. TPG > 12 mmHg and DPG > 7 mmHg are considered proxies of pulmonary vascular remodeling.^[11] However, TPG is influenced by stroke volume and PAWP. DPG is not affected by stroke volume, and studies based on both histology and RHC suggest that it is more accurate than TPG to detect the remodeling of pulmonary arterioles.^[12] Nonetheless, the prognostic value of DPG has been questioned.^[13,14]

Irrespective of the hemodynamic definition, patients with Cpc-PH have more pronounced ventilatory responses to exercise, lower pulmonary arterial compliance, impaired right ventricular function, and shorter life expectancy than those with Ipc-PH.^[15] The worse survival is mainly due to the development of right ventricular failure.^[16]

In this article, we review the literature about worldwide epidemiology of PH-LHD. Unlike the global burden of chronic respiratory diseases, this topic has not systematically been addressed in recent years.^[17] We also cover the prognostic impact of PH-LHD since it is commonly assessed together with PH-LHD prevalence.

2. Methods

The PubMed database was searched by entering the following keywords: “pulmonary hypertension” “group 2 pulmonary hypertension” “left heart disease” “heart failure” “mitral regurgitation” “mitral stenosis” “aortic regurgitation” “aortic stenosis” “rheumatic heart disease” “hypertrophic cardiomyopathy”, and “cardiac amyloidosis”. The relevant abbreviations (eg, PH) were also used.

Since the underlying etiology is not normally specified in studies on PH associated with heart failure (PH-HF), we considered this entity separately from PH associated with specific types of LHD (ie, valvular heart disease (VHD), hypertrophic cardiomyopathy (HCM), and cardiac amyloidosis). To provide aggregate measures for multiple studies, we calculated the average of the mean, median, or percent values in each study. Information about PH-LHD clinical trials was retrieved in the ClinicalTrials.gov and EudraCT databases.

3. PH-HF

Most studies on the prevalence of PH-HF have followed the classification of HF into HF with reduced ejection fraction (HFrEF), HF with mid-range or mildly reduced ejection fraction (HFmrEF), and HF with preserved ejection fraction (HFpEF). The main findings are shown in Figure 2.^{[16,18,19,20,21,22,23,24,25,26,27,28,29]}

Figure 2.

Prevalence and prognostic impact of PH-HF.^{[16,18,19,20,21,22,23,24,25,26,27,28,29]} The graph on the left side synthesizes the results of the main studies assessing PH prevalence in HF, with distinction between HFrEF and HFmrEF vs. HFpEF. For reference [22], the subgroups stratified by HFpEF and HFrEF/HFmrEF were further divided into 2 distinct cohorts, each represented by a separate bar. The graph on the right side depicts the hazard ratio along with the 95% confidence interval for all-cause mortality, with the exception of the studies by Shah et al^[27,29], which assessed the combined outcome of all-cause mortality or HF hospitalization. HF: Heart failure; HFmrEF: Heart failure with mildly reduced/mid-range ejection fraction; HFpEF: Heart failure with preserved ejection fraction; HFrEF: Heart failure with reduced ejection fraction; PH: Pulmonary hypertension; PH-HF: Pulmonary hypertension associated with heart failure; RHC: Right heart catheterization.

3.1. HFrEF and HFmrEF

Left ventricular systolic dysfunction is the hallmark of HFrEF and is also found in HFmrEF. HFmrEF has other similarities with HFrEF, particularly the predominance of ischemic heart disease as etiology and the response to neurohormonal inhibitors and angiotensin receptor-neprilysin inhibitors.^[30] Thus, we discuss these 2 HF phenotypes together.

Key studies of the prevalence of PH in HFrEF and HFmrEF were conducted in high-income countries and are listed in Table 1.^{[16,18,19,20,21,22,23,24,25,26,27,28,29]} The mean age of the included patients was 63.5 years, and the majority of patients were male (58.9%). Overall, the proportion of subjects with PH was similar when determined by transthoracic echocardiography (TTE) or RHC (60.4% vs. 60.6% respectively). Except in 1 case, the investigators adopted the previous mPAP threshold of 25 mmHg for the diagnosis of PH, with potential exclusion of subjects with mPAP between 20 and 24 mmHg. There was a direct correlation between elevated PAP and all-cause mortality [Figure 2].^{[16,18,19,20,21,22,23,24,25,26,27,28,29]}

Table 1.

Studies of PH in HFrEF, HFmrEF, and HFpEF.

First author	n	Age (year)§	Female (%)	Countries and regions	Method of PH detection	PH definition	PH prevalence	Prognostic implications
Ghio[16]	377	51 ± 10	14	Italy	RHC	mPAP > 20 mmHg	62%	↑ mPAP: ↑ mortality
Miller[18]*	463	Passive PH: 57 ± 13 Mixed PH: 60 ± 15	Passive PH: 21 Mixed PH: 28	USA	RHC	Passive PH: mPAP ≥ 25 mmHg, PAWP > 15 mmHg, PVR < 3 WU Mixed PH: mPAP ≥ 25 mmHg, PAWP > 15 mmHg, PVR ≥ 3 WU	Passive PH: 45% of patients with PH, 33% of total Mixed PH: 55% of patients with PH, 40% of total	Passive PH: ↑ mortality Mixed PH: ↑ mortality
Vanhercke[19]	401	83 ± 5	50	Belgium	TTE	RVSP ≥ 30 mmHg	Any HF: 69% HFrEF: 67% HFpEF: 73%	↑ TR gradient: ↑ mortality
Szwejkowski[20]	1,612	75.2	42.6	Scotland	TTE	RVSP quintiles (< 33 mmHg, 33–38 mmHg, 39–44 mmHg, 45–52 mmHg, > 52 mmHg)	< 33 mmHg: does not have PH and not included 33–38 mmHg: 18.6% 39–44 mmHg: 17.2% 45–52 mmHg: 23.0% > 52 mmHg: 24.4%	6% relative increase in all-cause mortality every 5-mmHg rise of RVSP
Damy[21]	LVEF ≤ 45%: 1,026 LVEF > 45%: 354	LVEF ≤ 45%: 70 ± 10 LVEF > 45%: 76 ± 9	LVEF ≤ 45%: 27 LVEF > 45%: 51	UK	TTE	RVSP ≥ 35 mmHg	LVEF ≤ 45%: 29% LVEF > 45%: 18%	PH: ↑ mortality
Gerges[22]†	Cohort 1: 664 Cohort 2: 172	Cohort 1: 60.9 Cohort 2: 66.1	Cohort 1: 23 Cohort 2: 25.6	Austria	RHC	mPAP > 25 mmHg, PAWP > 15 mmHg Ipc-PH: DPG < 7 mmHg, Cpc-PH: DPG > 7 mmHg	Cohort 1: 67.9% (Ipc-PH: 55.9%, Cpc-PH: 12.0%) Cohort 2: 79.7% (Ipc-PH: 65.7%, Cpc-PH: 14.0%)	DPG > 7 mmHg: ↑ mortality
Gerges[22]‡	Cohort 1: 399 Cohort 2: 219	Cohort 1: 66.5 Cohort 2: 72.2	Cohort 1: 50.1 Cohort 2: 52.1	Austria	RHC	mPAP > 25 mmHg, PAWP > 15 mmHg Ipc-PH: DPG < 7 mmHg, Cpc-PH: DPG > 7 mmHg	Cohort 1: 54.4% (Ipc-PH: 42.1%, Cpc-PH: 12.3%) Cohort 2: 63.0% (Ipc-PH: 51.1%, Cpc-PH: 11.9%)	DPG > 7 mmHg: ↑ mortality
Adir[23]	LVEF < 50%: 168 LVEF ≥ 50%: 86	72.2	39	Italy, Israel	TTE/RHC	sPAP > 40 mmHg Cpc-PH: mPAP ≥ 25 mmHg, PAWP > 15 mmHg, DPG ≥ 7 mmHg	LVEF < 50%: Cpc-PH 17% LVEF ≥ 50%: Cpc-PH 38%	Low PCa, but not DPG: ↑ mortality
Vanderpool[24]	19,262	62	43.1	USA	RHC	mPAP ≥ 25 mmHg, PAWP > 15 mmHg	HFrEF: 18.1%, HFpEF: 25.8%	PH: ↑ hospitalizations and mortality, PVR ≥ 3 WU: ↑ mortality
Lam[25]	244	78	56	USA	TTE	sPAP ≥ 35 mmHg	83%	sPAP > 48 mmHg: ↑ mortality
Sera[26]	219	80.8	54	Japan	RHC	mPAP > 25 mmHg, PAWP > 15 mmHg Ipc-PH: DPG < 7 mmHg and PVR < 3 WU, Cpc-PH: DPG ≥ 7 mmHg and/or PVR > 3 WU	Ipc-PH: 20%, Cpc-PH: 4%	Cpc-PH: ↑ HF hospitalization and mortality within 1 year after discharge
Shah[27]	450	69.9 ± 9.7	49	USA, Argentina, Brazil, Canada, Georgia, Russian Federation	TTE	TRV > 2.9 m/s	36%	Elevated TRV: ↑ HF hospitalization and mortality
Leung[28]	455	67.8 ± 11.2	50.2	USA	RHC	mPAP > 25 mmHg	52.5%	–
Shah[29]	1,097	73.7 ±8	53	43 countries participating in PARAGON-HF	TTE	TRV > 2.9 m/s	31%	High TRV and TAPSE/sPAP ratio: ↑ HF hospitalization or cardiovascular death

–: not avaliable. ^*This study included patients with left ventricular systolic dysfunction (LVEF < 40%). ^†Patients with HFrEF/HFmrEF. ^‡Patients with HFpEF. ^§Age is presented as mean or mean ± standard deviation. Cpc-PH: Combined post -and pre-capillary pulmonary hypertension; DPG: Diastolic pulmonary gradient; HFmrEF: Heart failure with mildly reduced/mid-range ejection fraction; HFpEF: Heart failure with preserved ejection pressure; HFrEF: Heart failure with reduced ejection fraction; Ipc-PH: Isolated post-capillary pulmonary hypertension; LVEF: Left ventricular ejection fraction; mPAP: Mean pulmonary artery pressure; MR: Mitral regurgitation; PARAGON-HF: Prospective Comparison of ARNI With ARB Global Outcomes in HF With Preserved Ejection Fraction; PAWP: Pulmonary artery wedge pressure; PCa: Pulmonary artery compliance; PH: Pulmonary hypertension; PVR: Pulmonary vascular resistance; RHC: Right heart catheterization; RVSP: Right ventricular systolic pressure; sPAP: Systolic pulmonary artery pressure; TAPSE: Tricuspid annular plane systolic excursion; TR: Tricuspid regurgitation; TRV: Tricuspid regurgitation velocity; TTE: Transthoracic echocardiography; WU: Wood units.

3.2. HFpEF

HFpEF is characterized by left ventricular ejection fraction (LVEF) ≥50% and predominant left ventricular diastolic dysfunction. This condition may be secondary to distinct forms of LHD (sometimes referred to as HFpEF mimickers)—for example, aortic stenosis—or may be self-standing, most frequently in obese individuals or elderly women with cardiovascular and non-cardiovascular comorbidities, such as arterial hypertension, atrial fibrillation (AF), and diabetes mellitus.^[1] It is anticipated that the number of patients with HFpEF will almost double by 2030, mainly because of increasing incidence of self-standing HFpEF.^[31]

In the main studies of PH in HFpEF, patients had or were assumed to have HFpEF without any underlying specific LHD, although it cannot be excluded that some instead had HFpEF mimickers. The included subjects were from Europe and the United States, but also from Japan and international cohorts [Table 1].^{[16,18,19,20,21,22,23,24,25,26,27,28,29]} The mean age was 63.6 years, and the mean body mass index was 29.1 kg/m². The female to male ratio was around 1. Like in patients with HFrEF/HFmrEF, it is possible that PH was underdiagnosed when RHC was employed since the mPAP cut-off was 25 mmHg. The mean patient age (62.6 years vs. 72.6 years) and the proportion of female patients (43.6% vs. 51%) were lower in high-income vs. low- or middle-income countries. In half of the investigations in high-income countries, PH was assessed by RHC. Combining TTE and RHC data, the prevalence was 47.5% [Figure 3]. Conversely, the prevalence of PH in cohorts from low- or middle-income countries was only determined by TTE and was 33.5%. Hospitalization and mortality were more likely in patients with higher systolic PAP (sPAP) and tricuspid regurgitation velocity.^[25] Furthermore, the presence of Cpc-PH, as defined by PVR >3 WU, portended an increased risk of mortality.^[24]

Figure 3.

Prevalence of PH among different types of LHD. The bar graphs in the upper part of the figure show the mean prevalence of PH in different types of LHD in the studies included in this article, after excluding those considering only patients with severe PH. The mean distribution of males and females and the mean age in the same studies are presented in the lower part of the figure. AR: Aortic regurgitation; AS: Aortic stenosis; CA: Cardiac amyloidosis; HCM: Hypertrophic cardiomyopathy; HFmrEF: Heart failure with mildly reduced ejection fraction; HFpEF: Heart failure with preserved ejection fraction; HFrEF: Heart failure with reduced ejection fraction; LHD: Left heart disease; MR: Mitral regurgitation; MS: Mitral stenosis; PH: Pulmonary hypertension; RHC: Right heart catheterization; RHD: Rheumatic heart disease; TTE: Transthoracic echocardiography.

4. PH associated with VHD

4.1. Mitral stenosis

Mitral stenosis represents a model of PH-LHD pathophysiology. The reduction in the mitral valvular area directly causes a rise of left atrial pressure and, thereby, pulmonary pressure. In addition, fibrosis of the chronically stretched wall reduces the compliance of the left atrium and magnifies the increase in filling pressure, which propagates to the pulmonary circulation. AF develops over time, further impairing left atrial function. As a result, virtually any patient presents with PH during the course of mitral stenosis.^[32]

There were 6 studies that determined the prevalence of PH associated with mitral stenosis in cohorts from India, the Middle East, and the United States, always by means of RHC [Table 2].^{[33,34,35,36,37,38,39,40,41,42,43,44,45,46,47]} The age of subjects from low- or moderate-income countries was younger than from high-income countries (39.8 years vs. 56.7 years). Of the patient group, 24.7% were male. LVEF was reported only in 1 investigation and was 52%.^[35] The main cause of mitral stenosis was rheumatic heart disease (RHD). Overall, more than 70% of patients with mitral stenosis had PH, with the caveat that they were symptomatic in most investigations and, thus, had the highest likelihood to have advanced VHD with PH. Patients in the United States had PH more often than patients from other countries (81.5% vs. 51.5%), suggesting an even stricter selection [Figure 4A].^{[33,34,35,36,37,38]} Even modest increases in PAP were associated with adverse surgical outcomes, although the degree of PH making surgery prohibitive is debated.^[48]

Table 2.

Studies of PH in mitral valve disease.

First author	n	Treatment	Age (year)*	Female (%)	Nation (s)	Method of PH detection	PH definition (mmHg)	PH prevalence (%)	Prognostic implications
Mitral stenosis
Fawzy[33]	559	MBV	NA	NA	Saudi Arabia	RHC	sPAP ≥ 50	38	Regression of PH 6–12 months after successful MBV
Fawzy[34]	531	MBV	NA	NA	Saudi Arabia	RHC	sPAP > 60	15	PH: ↓ event-free survival at 10 and 15 years
Pourafkari[35]	558	MBV	45 ± 13	76	Iran	RHC	mPAP ≥ 25	81	NA
Yang[36]	317	Surgery	61	77	USA	RHC	mPAP > 25	90	Moderate to severe PH: ↑ mortality after surgery
Hart[37]	295	MBV	52 ± 13	85	USA	RHC	mPAP > 25	73	Sustained benefit for up to 36 months for MBV in patients with normal TPG
Krishnamoorthy[38]	244	MBV	28	60	India	RHC	mPAP > 30	72	Severe PH: ↑ restenosis rate after MBV
Mitral regurgitation
Ratwatte[39]	9,638	NA	76 ± 12	62.6	Australia	TTE	sPAP ≥ 40	59.6	Higher sPAP: ↑ mortality
Al-Bawardy[40]	4,071	M-TEER	81	46	USA	TTE/RHC	sPAP ≥ 35	38	PH: ↑ mortality
Mentias[41]	1,318	MVR	62 ± 13	34	USA	TTE	sPAP ≥ 35	47	Higher baseline sPAP: ↑ mortality
Ghoreishi[42]	1,037	M-TEER	74 ± 12	44	USA	TTE/RHC	ICD-9 code	32.6	PH did not increase mortality after TMVR
Ahmed[43]	873	MVR	59 ± 14	41	USA	TTE/RHC	sPAP ≥ 40	53	Higher sPAP: ↑ mortality
Tigges[44]	643	NA	75	38	Germany, TRAMI registry	TTE	sPAP > 37	68.9	Higher sPAP: ↑ MACCE and mortality
Ben-Yehuda[45]	528	NA	72 ± 11	37	USA, Canada (COAPT trial)	TTE	sPAP > 50	34.8	Early reduction in sPAP after M-TEER: ↓ HF hospitalization or mortality
Barbieri[46]	437	MVR	67 ± 11	34	USA, France, Italy	TTE	sPAP > 50	23	sPAP > 50 mmHg: ↑ HF, CVD, and all-cause mortality
Tourneau[47]	256	NA	63 ± 12	34	France	TTE	sPAP > 50	32	Higher sPAP: ↑ mortality

^*Age is presented as mean or mean ± standard deviation. COAPT: Cardiovascular Outcomes Assessment of the MitraClip Percutaneous Therapy for Heart Failure Patients With Functional Mitral Regurgitation; CVD: Cardiovascular death; ICD-9: International Classification of Diseases-9; MACCE: Major adverse cardiac and cerebrovascular events; MBV: Mitral balloon valvuloplasty; mPAP: Mean pulmonary artery pressure; M-TEER: Mitral transcatheter edge-to-edge repair; MVR: Mitral valve repair; NA: Not available; PH: Pulmonary hypertension; RHC: Right heart catheterization; sPAP: Systolic pulmonary artery pressure; TMVR: Transcatheter mitral valve repair; TPG: Transpulmonary gradient; TRAMI: Transcatheter Mitral Valve Interventions; TTE: Transthoracic echocardiography.

Figure 4.

Prevalence of PH associated with mitral valve disease.^{[33,34,35,36,37,38,39,40,41,42,43,44,45,46,47]} The graphs synthetize the results of the main studies assessing PH prevalence in mitral stenosis (A) and mitral regurgitation (B), grouped by the continents where they were carried out. PH: Pulmonary hypertension; RHC: Right heart catheterization.

In patients in the United States with mitral stenosis and sPAP 35–44 mmHg, cardiac surgery halted PH progression and improved long-term survival at the cost of acceptable 30-day mortality.^[36] In a small sample study from India, mortality rose from 5.5% to 28.5% when PAP was supra-systemic. Mubeen et al^[49] concluded that surgery for mitral stenosis is feasible as long as PAP is lower than systemic pressure. Interestingly, histological analysis of lung biopsy specimens did not reveal specific changes in pulmonary arterioles in the group with supra-systemic PAP.

The outcomes of patients with PH associated with mitral stenosis undergoing percutaneous balloon mitral commissurotomy were different depending on the study location. In the United States, the procedure was shown to be safe and effective to treat patients with mitral stenosis and severe PH, with immediate pulmonary hemodynamic improvement and sustained benefit up to 3 years after the procedure.^[37] Conversely, PVR remained elevated after percutaneous balloon mitral commissurotomy or valvuloplasty in patients from Saudi Arabia, possibly because the presence of Cpc-PH that was not modified by intervention.^[34]

4.2. Mitral regurgitation

Like mitral stenosis, mitral regurgitation increases left atrial pressure directly and indirectly, via left atrial dilation and stiffening [Figure 1]. The high left atrial pressure is transmitted to the pulmonary circulation in a retrograde manner, with a peak during the systolic backflow through the regurgitant mitral valve, which causes the appearance of large V-waves in the PAWP waveform at RHC.^[50]

Figure 4B^{[39,40,41,42,43,44,45,46,47]} and Table 2^{[33,34,35,36,37,38,39,40,41,42,43,44,45,46,47]} present key studies evaluating PH resulting from primary and secondary mitral regurgitation. All were from high-income countries in Europe, North America, and Australia, mainly because the focus was the effect of transcatheter mitral valve repair (TMVR).

The included patients had a mean age of 74.7 years, and 47.6% were male. Mean LVEF, as calculated based on the information in 7 investigations, was 56.7%. The prevalence of PH was highly variable due to differences in assessment methods, PH definition, and mitral regurgitation severity. It varied from 23% in degenerative mitral regurgitation with leaflet flail, when the cut-off of sPAP was 50 mmHg,^[46] to 69.9% in German patients with sPAP >37 mmHg enrolled in the Transcatheter Mitral Valve Interventions (TRAMI) registry.^[44] In the largest TTE-based investigation, 59.6% of 9,638 patients affected by moderate to severe mitral regurgitation had PH, and 30% had sPAP 30–39.9 mmHg.^[39]

Both TTE-estimated sPAP and RHC-measured PAP were reported to decrease early after TMVR or cardiac surgery, and a prompt reduction in PAP was a strong independent predictor of favorable long-term clinical outcomes.^[51,52,53] It is noteworthy that TMVR reduced sPAP to a greater extent than guideline-directed medical therapy alone.^[45] Even mild PH was an independent predictor of perioperative mortality in patients undergoing surgery for mitral regurgitation.^[42] Conversely, the presence of PH did not increase the risk of in-hospital mortality in patients undergoing TMVR, nor did it affect the rate of procedural success and in-hospital major adverse cardiovascular events.^[43] Higher sPAP correlated with higher post-discharge short- and long-term mortality after both mitral regurgitation surgery and TMVR.^[42,44]

4.3. Aortic stenosis

The main studies of the prevalence of PH in aortic stenosis included patients undergoing transcatheter aortic valve replacement (AVR) or surgical AVR in high-income countries in Europe, North America, and Australia. The mean age was 77.9 years, and 45.2% of subjects were male [Table 3].^{[54,55,56,57,58,59,60,61,62,63,64,65,66]} LVEF was always reported, and the mean value was 55.6%. PH was found in 6% of asymptomatic patients with aortic stenosis^[67] and in 47%–75% of those treated by surgical AVR or transcatheter AVR [Figure 5].^{[54,55,56,57,58,59,60,61]} Around 10% of subjects scheduled for transcatheter AVR and evaluated by RHC had Cpc-PH.^[57]

Table 3.

Studies of PH in aortic valve disease.

First author	n	Treatment/Grade	Age (year)*	Female (%)	Nation	Method of PH detection	PH definition	PH prevalence (%)	Prognostic implications
AS
Ratwatte[54]	14,980	AS	77 ± 8	57.4	Australia	TTE	sPAP ≥ 30 mmHg	86.3	eRVSP > 40 mmHg: ↑ mortality
Luçon[55]	2,435	TAVR	83 ± 7	51	France	TTE	sPAP ≥ 40 mmHg	65	PH: ↑ mortality; severity of PH does not affect procedural risk or 30-day mortality
Melby[56]	1,080	SAVR	NA	NA	USA	TTE/RHC	sPAP ≥ 35 mmHg	47	PH: ↑ mortality
Sultan[57]	561	TAVR	82 ± 8	49	USA	RHC	mPAP ≥ 25 mmHg, PAWP > 15 mmHg Ipc-PH: PVR ≤ 3 WU, Cpc-PH: PVR > 3 WU	PH: 64; Ipc-PH: 34 Cpc-PH: 20†	Cpc-PH: ↑ mortality despite successful TAVR
Ben-Dor[58]	509	Severe AS	> 80	25	USA	RHC	sPAP ≥ 40 mmHg	68.2	PH: ↑ mortality
Weber[59]	503	SAVR	74 ± 10	42	Switzerland	RHC	mPAP ≥ 25 mmHg, PAWP > 15 mmHg Ipc-PH: PVR ≤ 3 WU, Cpc-PH: PVR > 3 WU	PH: 48; Ipc-PH: 28.6,Cpc-PH: 12.7†	Cpc-PH: ↑ mortality
O’Sullivan[60]	433	TAVR	82 ± 5	55	Switzerland	RHC	mPAP > 25 mmHg, LVEDP > 15 mmHg Ipc-PH: DPG < 7 mmHg, Cpc-PH: DPG > 7 mmHg	PH: 75; Ipc-PH: 51, Cpc-PH: 11†	Cpc-PH: ↑ mortality
Maeder[61]	421	SAVR	75 ± 10	41	Canada	RHC	mPAP > 25 mmHg	49.3	PVR > 3 WU and/or sPAP > 60 mmHg: ↑ mortality
AR
Ratwatte[62]	8,392	At least moderate AR	74 ± 14	58.4	Australia	TTE	sPAP ≥ 40 mmHg	44.4	PH: ↑ mortality
Parker[63]	1,156	At least mild AR	72 ± 14	51	Israel	TTE	Moderate or greater PH (not specified)	26	PH: ↑ mortality
Magne[64]	802	NA	NA	NA	France	RHC	mPAP > 25 mmHg	36.7	NA
Khandhar[65]	506	severe AR	65	55	USA	TTE	Severe PH: sPAP ≥ 60 mmHg	16	Better survival in patients surgically treated than medically treated
Naidoo[66]	139	severe AR	NA	NA	South Africa	RHC	Severe PH: sPAP ≥ 60 mmHg	24	PH does not influence the early outcome after AVR

^*Age is presented as mean, mean ± standard deviation or range. ^†Pre-capillary PH is not listed. AR: Aortic regurgitation; AS: Aortic stenosis; AVR: Aortic valve replacement; Cpc-PH: Combined post-and pre-capillary pulmonary hypertension; DPG: Diastolic pressure gradient; eRVSP: Estimated right ventricular systolic pressure; Ipc-PH: Isolated post-capillary pulmonary hypertension; LVEDP: Left ventricular end-systolic pressure; mPAP: Mean pulmonary artery pressure; NA: Not available; PAWP: Pulmonary artery wedge pressure; PH: Pulmonary hypertension; PVR: Pulmonary vascular resistance; RHC: Right heart catheterization; SAVR: Surgical aortic valve replacement; sPAP: Systolic pulmonary artery pressure; TAVR: Transcatheter aortic valve replacement; TTE: Transthoracic echocardiography; WU: Wood units.

Figure 5.

Prevalence of PH associated with aortic stenosis.^{[54,55,56,57,58,59,60,61]} The graph synthesizes the results of the main studies assessing PH prevalence in aortic stenosis, grouped by the continents where they were carried out. PH: Pulmonary hypertension; RHC: Right heart catheterization.

PH was associated with worse prognosis in patients with aortic stenosis, with the risk of all-cause death rising from sPAP values > 35 mmHg.^[54] The severity of PH did not influence the risk of transcatheter AVR-related complications or 30-day mortality. However, sPAP > 60 mmHg was an independent predictor of 1-year mortality.^[55] RHC data indicated that 1-year mortality is higher in subjects undergoing transcatheter AVR who have Cpc-PH but not in those who have Ipc-PH, as compared to patients without PH. Moreover, improvement of sPAP and right ventricular function at 1-year follow-up is less common in the former than in the latter.^[57,60]

PH before surgical AVR was associated with prolonged ventilation, longer stay, and higher in-hospital mortality. Long-term survival was also negatively influenced by pre-operative PH, with 5-year survival being 62%–68% in patients with no or mild PH at the time of surgical AVR and 52%–57% in those with moderate or severe PH.^[56] Post-procedural, residual PH was associated with worse prognosis than pre-procedural PH.^[56]

4.4. Aortic regurgitation

We found a small number of studies on the prevalence of PH in patients with aortic regurgitation [Table 3]^{[54,55,56,57,58,59,60,61,62,63,64,65,66]} [Figure 3]^{[62,63,64,65,66]}. All were conducted in high-income countries. The mean age was 70 years, and females outnumbered males (55% vs. 45%). LVEF was reported in 2 studies, and the mean was 58.5%. In the largest investigation, based on TTE, PH was found in almost 45% of 8,392 patients with isolated aortic regurgitation and preserved LVEF.^[62] In another study relying on RHC, PH was demonstrated in 36.7% of 805 subjects with aortic regurgitation.^[64] The prevalence of PH dropped to between 16% and 24% when only severe PH was considered.^[65,66]

Patients with aortic regurgitation with PH had greater left ventricular end-diastolic and end-systolic dimensions, lower LVEF, and higher grades of mitral regurgitation than those without PH.^[65] Moreover, they faced higher all-cause mortality.^[62] AVR was shown to lower sPAP from 66 mmHg to 31 mmHg in a small surgical sample and operative mortality was acceptable (3%) even in patients with severe PH, supporting the choice to perform AVR also in this challenging subgroup.^[63]

4.5. RHD

RHD is still highly prevalent in Africa, the southeast and middle-east regions of Asia, the Caribbean islands, and some Latin American countries because the likelihood of the disease is increased by factors relating to overcrowding, poor hygienic conditions, low education, and limited access to care.^[68] RHD typically affects the left cardiac valves, especially the mitral one. Annulus dilatation, chordal elongation, and leaflet prolapse cause mitral regurgitation, and commissural fusion leads to mitral stenosis. Isolated involvement of the aortic valve is rare, whereas bi-valvular disease is common.^[69]

PH is an expected complication of RHD, along with HF and AF, and it has been estimated that 3.75 million individuals in the world suffer from RHD-related PH, with this etiology being the cause of 10% of PH cases in Africa and 6% of PH cases in Asia.^[70] Real-world data suggest that RHD accounts for even higher proportions of patients with PH-LHD.^[71,72,73] Furthermore, RHD is the most common cause of PH in African children, together with unoperated congenital heart disease.^[74] Table 4^{[75,76,77,78,79,80,81,82,83]} lists the 9 studies reporting the prevalence of PH in subjects with RHD, without details of the affected valves. The investigations were mainly carried out in Africa and South Asia,^{[75,76,77,78,79,80,81,82,83]} and no investigations were done in high-income countries. The mean age was 31.6 years, and 34% of the patient population was male. The mean LVEF was 58.4% (based on 3 studies).

Table 4.

Studies of PH in rheumatic heart disease.

First author	n	Type of rheumatic heart disease	Age (year)*	Female (%)	Nation(s)	Method of PH detection	PH definition	PH prevalence (%)
Zühlke[75]	3,343	MS: 72.9% MR: 60.4% PS: 59.5% TS: 54.2% AS: 61.9% AR: 55.2%	28	67	Africa, India, Yemen	TTE	Not specified	28.8
Kingué[76]	1,385	VALVAFRIC study	30	60	Africa	TTE	Not specified	28.7
Ahmed Abdi[77]	433	Isolated mitral valve: 51.3% Combined mitral and aortic valve: 44%	46 ± 20	66	Somalia	TTE	sPAP > 30 mmHg	23.8
Sriharibabu[78]	403	MR: 43.4% MS: 42.4% AS: 4.0% AR: 3.5% PS: 0.2% Mitral and aortic valve lesions: 6.5%	33 ± 4	70	India	TTE	Not specified	52
Sliwa[79]	344	NA	43	68	South Africa	TTE	sPAP > 35 mmHg	18
Okello[80]	309	NA	30	63.4	Uganda	TTE	sPAP > 35 mmHg	32.7
Tchoumi[81]	270	MR: 61.5% AR: 38.5% MS: 6.0% Mitral disease: 8.0%	12 ± 4	63	Cameroon	TTE	Not specified	87
Laudari[82]	235	MS: 11.06% MR: 25.96% AS: 3.83% AR: 11.06%	39 ± 4	70	Nepal	TTE	sPAP > 30 mmHg	51.6
Sani[83]	129	MR: 38.0% Mixed mitral valve disease: 27.9% Combined mitral and aortic disease: 19.5%	24 ± 12	64	Nigeria	TTE	Not specified	72.1

^*Age is presented as mean or mean ± standard deviation. AR: Aortic regurgitation; AS: Aortic stenosis; MR: Mitral regurgitation; MS: Mitral stenosis; NA: Not available; PH: Pulmonary hypertension; PS: Pulmonary stenosis; sPAP: Systolic pulmonary artery pressure; TS: Tricuspid stenosis; TTE: Transthoracic echocardiography; VALVAFRIC: Registry of Rheumatic Heart Disease in Western and Central Africa.

On average, 44% of patients had PH, but with great variability [Figure 3]. In the Global Rheumatic Heart Disease Registry (REMEDY) study, up to 28% of patients presented with elevated PAP.^[75] In patients admitted to hospitals in Africa, India, and Yemen and enrolled in the Registry of Rheumatic Heart Disease in Western and Central Africa (VALVAFRIC) or REMEDY, the frequency of PH was almost 30%.^[75,76] The prevalence of PH was higher in studies with smaller samples, peaking at 87% in children referred for cardiac surgery.^[81] In a pooled sample of 3,750 patients with RHD from 2 studies, the prevalence of PH was 12.9%.^[84] The variability of these metrics may be explained by differences in patient populations and RHD phenotypes.^[85]

5. PH associated with HCM

We found 10 studies that evaluated the prevalence of PH in HCM [Table 5].^{[86,87,88,89,90,91,92,93,94,95]} All were performed in high-income areas—the United States, Europe, China, and Japan—and were mostly in tertiary centers.

Table 5.

Studies of PH in hypertrophic cardiomyopathy.

First author	n	HCM phenotype	Age (year)*	Female (%)	Nation	Method of PH detection	PH definition	PH prevalence (%)	Prognostic implications
Ong[86]	1,570	OHCM: 74.8%, non-OHCM: 25.2%	57 ± 15	58.8	USA	TTE	sPAP > 36 mmHg	38.2	PH: ↑ mortality SRT, mainly septal myectomy, eliminates the adverse prognostic impact of PH
Ahmed[87]	1,342	OHCM with septal myectomy	57.9	49.5	USA	TTE	sPAP ≥ 35 mmHg	47.8	Higher sPAP: ↑ mortality
Anand[88]	542	ApHCM	59.4 ± 16.6	44	USA	TTE	sPAP > 36 mmHg	34	PH: worse symptoms ↑ mortality
Kanbayashi[89]	401	LVOT-OHCM: 88, mid-ventricular OHCM: 41, ApHCM: 124	50.7	35	Japan	TTE	sPAP > 40 mmHg	17.6	sPAP: ↑ stroke and systemic embolic events
Musumeci[90]	361	OHCM: 165, non-OHCM: 126, end-stage: 20	51.2 ± 16.9	38.9	Italy	TTE	sPAP > 35 mmHg	18.3	PH: ↑ HCM-related morbidity
Finocchiaro[91]	324	NA	50 ± 16	41	USA	TTE	sPAP > 35 mmHg	24	PH: ↑ mortality or heart transplantation
Geske[92]	306	OHCM with septal myectomy	53	52	USA	TTE	sPAP > 35 mmHg	53	Surgical myectomy reduces PH
Wu[93]	301	OHCM	NA	NA	China	TTE	Not specified	12.3	NA
Covella[94]	162	OHCM + advanced HF, treated with SRT	59 ± 11	51	USA	RHC	mPAP > 25 mmHg, PAWP > 15 mmHg	PH: 51; Cpc-PH: 34	No prognostic impact of elevated PAP
							Cpc-PH: PVR > 3 WU
Klarich[95]	193	ApHCM	58 ± 17	38	USA	TTE	sPAP > 35 mmHg	62	NA

^*Age is presented as mean or mean ± standard deviation. ApHCM: Apical hypertrophic cardiomyopathy; Cpc-PH: Combined post-and pre-capillary pulmonary hypertension; HCM: Hypertrophic cardiomyopathy; HF: Heart failure; LVOT: Left ventricular outflow tract; mPAP: Mean pulmonary artery pressure; NA: Not available; OHCM: Obstructive hypertrophic cardiomyopathy; PAWP: Pulmonary artery wedge pressure; PH: Pulmonary hypertension; PVR: Pulmonary vascular resistance; RHC: Right heart catheterization; sPAP: Systolic pulmonary artery pressure; SRT: Septal reduction therapy; TTE: Transthoracic echocardiography; WU: Wood units.

One study relied on RHC performed before septal myectomy^[94] and the other ones on TTE^{[86,87,88,89,90,91,92,93,95]}. The mean age of the included patients was between 50 and 60 years, 44.8% were female, and mean LVEF was 67.5% (based on 8 studies).

PH was reported in 12%–62% of patients, depending on the definition [Figure 3]. The highest rates were found in HCM with an obstructive phenotype^[96] and in apical variants.^[95] Older age (especially in women), AF, HF symptoms, moderate-severe mitral regurgitation, and higher obstruction gradients were correlates of more severe PH in obstructive HCM.^[92,93,96] More than two-thirds of patients with end-stage HCM had PH, contributing to the poor prognosis of the disease at this stage.^[90]

PH portended an increased risk of morbidity and mortality^[86] and was associated with higher peri-operative risk in subjects undergoing cardiac surgery.^[97] However, it did not negatively influence the outcomes of septal alcohol ablation and surgical myectomy.^[94] In fact, patients with moderate-severe PH had maximum benefit from the elimination of the left ventricular tract outflow obstruction.^[92] The reduction in PAP may be progressive over time and long-lasting after treatment of left ventricular outflow tract obstruction.^[92]

6. PH associated with cardiac amyloidosis

The 5 studies investigating the prevalence of PH in patients affected by cardiac amyloidosis were performed in the United States or Europe [Table 6].^{[98,99,100,101,102]} All were based on RHC, taking advantage of the need of endomyocardial biopsy for the diagnosis of cardiac amyloidosis. The mean age was 70 years, men and women were equally represented, and LVEF was overall mildly reduced. All studies included subjects with light chain cardiac amyloidosis; 3 also included subjects with transthyretin cardiac amyloidosis.

Table 6.

Studies of PH in cardiac amyloidosis.

First author	n	Cardiac amyloidosis type	Age (year)*	Female (%)	Nation	Method of PH detection	PH definition	PH prevalence (%)	Prognostic implications
Martens[98]	469	ATTR-CA: 52%, AL-CA: 42%	73 ± 10	22	USA	RHC	mPAP > 25 mmHg	79	No impact of PH on mortality, transplant, and LVAD placement
Longinow[99]	132	AL-CA: 52%, ATTR-CA: 47%	67	47	USA	RHC	mPAP > 20 mmHg, PAWP > 15 mmHg	PH: 74.9; Ipc-PH: 50.7, Cpc-PH: 24.2	Higher mPAP: ↑ mortality
							Ipc-PH: PVR < 3 WU, Cpc-PH: PVR > 3 WU
Duca[100]	61	ATTR-CA: 57.4%, AL-CA: 42.6%	74	62	Austria	RHC	Not specified	> 75	Greater impact of PH in ATTR-CA than AL-CA
Holt[101]	57	ATTR-CA: 100%	74	50	Norway	RHC	mPAP > 20 mmHg, PAWP > 15 mmHg	PH: 77; Ipc-PH: 28, Cpc-PH: 28	NA
							Ipc-PH: PVR ≤ 2 WU, Cpc-PH: PVR > 2 WU
Slivnick[102]	52	ATTR-CA: 60%	69 ± 9	15	USA	RHC	mPAP > 25 mmHg, PAWP > 15 mmHg, PVR > 3 WU	PH: 90; Ipc-PH: 55, Cpc-PH: 29	Higher TPG: ↑ adverse outcomes

^*Age is presented as mean or mean ± standard deviation. AL-CA: Amyloid light chain cardiac amyloidosis; ATTR-CA: Transthyretin cardiac amyloidosis; Cpc-PH: Combined post- and pre-capillary pulmonary hypertension; Ipc-PH: Isolated postcapillary pulmonary hypertension; LVAD: Left ventricular assist device; mPAP: Mean pulmonary artery pressure; NA: Not available; PAWP: Pulmonary artery wedge pressure; PH: Pulmonary hypertension; PVR: Pulmonary vascular resistance; RHC: Right heart catheterization; TPG: Transpulmonary gradient; WU: Wood units.

The prevalence of PH was > 75% in all studies, even when the cut-off of mPAP was 25 mmHg instead of 20 mmHg [Figure 3].^{[98,99,100,101,102]} Ipc-PH was the predominant PH-LHD subtype, followed by Cpc-PH. Strikingly, purely pre-capillary PH was also described, possibly because of amyloid deposition in pulmonary capillaries and arterioles.^[103,104] No difference in PH prevalence was found between transthyretin cardiac amyloidosis and light chain cardiac amyloidosis, although pulmonary hemodynamics affected the prognosis of transthyretin cardiac amyloidosis more than the prognosis of light chain cardiac amyloidosis.^[100] Higher mPAP and PVR were predictors of increased mortality.^[99]

7. Clinical trials

We identified 59 clinical trials for patients with PH-LHD in the ClinicalTrials.gov and EudraCT databases [Table 7].^{[105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141]} The vast majority focused on PH-HF. It is possible that a few subjects with PH associated with VHD, PH associated with HCM, or PH with cardiac amyloidosis were unintentionally enrolled in these trials, but these types of LHD were exclusion criteria.

Table 7.

Clinical trials on PH-LHD.

NCT/Eudract number	Interventions	Design	Type of LHD	Participating countries
Concluded trials with published results
FIRST[105]	Epoprostenol vs. SoC	Multicenter, randomized, open-label	HF, LVEF < 25% or LVEF < 30% and recently treated with IV inotrope	UK, USA, Australia, Austria, Canada, Denmark, Finland, France, Germany, Ireland, Italy, Netherlands, Scotland, Spain, Sweden
NA[106]	Sildenafil vs. placebo	Single-center, randomized, open-label	HF, LVEF ≤ 35%	Italy
NCT00781508[107]	Sildenafil vs. placebo	Single-center, randomized, double-blind, placebo-C	HF, any LVEF	USA
NCT00309790[108]	Sildenafil vs. placebo	Single-center, randomized, double-blind, placebo-C	HF, LVEF < 40%	USA
NA[109]	Bosentan vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF <35%	USA
NCT01156636[110]	Sildenafil vs. placebo	Two-center, randomized, double-blind, placebo-C	HF, LVEF ≥ 50%	Italy
NCT02133352*	Ranolazine	Single-center, single-arm, open-label	HF, LVEF ≥ 50%	USA
NCT00407446[111]	Sildenafil vs. placebo	Single-center, randomized, double-blind, placebo-C	HF, LVEF < 45%	Italy
NCT01800292*	Sildenafil	Single-center, open-label, single-group	HF, LVEF ≥ 50%	USA
NCT00458276[112]	Tezosentan vs. placebo	Multicenter, randomized, double-blind, placebo-C	Cardiac surgery with CPB	USA, Austria, Canada, Czech Republic, France, Germany, India, Israel, Italy, Poland, Serbia, Slovakia, Sweden, UK
NCT01172756[113]	Riociguat vs. placebo	Multicenter, randomized, double-blind, placebo-C	LVEF < 50%	Austria, Czech Republic, Germany
NCT01646515[114]	Udenafil vs. placebo	Single-center, randomized, double-blind, placebo-C	HF, LVEF ≤ 40%	Korea
NCT01726049[115]	Sildenafil vs. placebo	Single-center, randomized, double-blind, placebo-C	HF, LVEF ≥ 45%	Netherlands
NCT01932606[116]	Sodium nitrite vs. placebo	Single-center, randomized, double-blind, placebo-C	HF, LVEF ≥ 50%	USA
NCT02262078[117]	Nebulized sodium nitrite vs. placebo	Single-center, randomized, double-blind, placebo-C	HF, LVEF ≥ 50%	USA
NCT00820352[118]	Bosentan vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF ≥ 50%	Austria
NCT01726049[119]	Sildenafil vs. placebo	Single-center, randomized, double-blind, placebo-C	HF, LVEF ≥ 45%	Netherlands
ChiCTR-INR-16009511[120]	Fasudil	Two-center, single-arm, open-label	HF, LVEF ≥ 50%	China
NCT02070991[121]	Macitentan vs. placebo	Multicenter, randomized, double-blind, placebo-C	LVEF ≥ 30%	USA, Austria, Belgium, Canada, Czech Republic, France, Germany, Israel, Italy, Spain, Switzerland
NCT00862043[122]	Sildenafil vs. placebo	Multicenter, randomized, double-blind, placebo-C	Surgery or percutaneous treatment of VHD	Spain
NCT02885636[123]	Albuterol vs. placebo	Single-center, randomized, double-blind, placebo-C	HF, LVEF ≥ 50%	USA
NCT03015402[124]	Sodium nitrite vs. placebo	Single-center, randomized, double-blind, placebo-C, crossover	HF, LVEF ≥ 40%	USA
NCT02220335[125]	PA denervation vs. sildenafil plus sham PA denervation	Multicenter, randomized, open-label, sham-C	HF, any LVEF	China
NIS 01201257849[126]	Sildenafil	Single-center, case-control	HF, LVEF > 50%	Russian Federation
NCT02554903[127]	Macitentan vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVAD implanted within 90 d prior randomization	USA
NA[128]	Beraprost plus sildenafil vs. beraprost	Single-center, randomized, double-blind, placebo-C	HFrEF and HFmrEF	China
NCT04157660[129]	Oxygen-enriched air vs. placebo	Single-center, randomized, single-blind, placebo-C, crossover	HF, LVEF > 50%	Switzerland
NCT03541603[130]	Levosimendan vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF > 40%	USA
NCT03153111[131]	Macitentan vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF ≥ 40%	USA, Argentina, Austria, Brazil, Bulgaria, Czech Republic, Denmark, France, Germany, Hungary, Israel, Poland, Romania, Russian Federation, Spain, Sweden, UK
NCT02744339[132]	Riociguat vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF ≥ 50%	Austria
NCT02775539[133]	Mirabegron vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, any LVEF	Spain
NCT01959828[134]	Inhaled nitric oxide	Multicenter, open-label	HF with LVAD implant; children with CHD requiring surgery	Japan
Prematurely terminated trials; trials without reported results
NCT00840463*	Ambrisentan vs. placebo	Multicenter, randomized, double-blind	HF, LVEF ≥ 40%	USA
NCT01270750*	Bosentan	Single-center, non-randomized, open-label	HF, any LVEF, non-operable rheumatic MS	Greece
NCT01065051[135]	Riociguat vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, any LVEF	USA
NCT01275339[136]	Tadalafil vs. placebo	Randomized, double-blind, placebo-C	Moderate-severe AS with LVEF ≥ 50%	USA
NCT01272388*	Tadalafil vs. placebo	Randomized, double-blind, placebo-C	Severe AS with LVEF ≥ 45%	USA
NCT01910389[137]	Tadalafil vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF < 40%	USA, Canada
NCT01960153*	Tadalafil vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF < 40%	USA
NCT02078557*	MK-8892	Single-group assignment, open-label	HFrEF	Belgium, Moldova, Romania
EudraCT 2008-008135-28[138]	Sildenafil vs. placebo while breathing air or 15% oxygen	Double-blind, crossover	HF, LVEF < 40% or NT-proBNP > 600 ng/L	UK
NCT02304705*	Sildenafil vs. placebo	Single-center, randomized, double-blind, placebo-C	HF, any LVEF	USA
NCT03037580*	Treprostinil vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF ≥ 45%	USA
EudraCT 2017–003688-37[139]	Tadalafil vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF ≥ 50%	Germany
NCT03043651[140]	Treprostinil	Multicenter, open-label extension of NCT03037580	HF, LVEF ≥ 45%	USA
NCT03714815[131]	Macitentan vs. placebo	Multicenter, long-term, open-label extension of NCT03153111	HF, LVEF ≥ 40%	USA, Argentina, Brazil, Austria, Bulgaria, Denmark, France, Germany, Hungary, Israel, Poland, Romania, Russian Federation, Sweden, UK
NCT03624010*	Levosimendan	Open-label, single-group	HF, LVEF ≥ 50%	USA
NCT01616381*	Sildenafil vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF ≤ 40%	Israel, Italy, Norway, UK
Ongoing trials
NCT01065454[141]	Riociguat vs. placebo	Multicenter, randomized, double-blind, placebo-C	HFrEF	USA, Australia, Austria, Belgium, Canada, China, Czech Republic, Denmark, France, Germany, Italy, Japan, Netherlands, Poland, Singapore, Spain, Switzerland, UK
NCT01913847*	Sildenafil vs. placebo	Single-center, randomized, double-blind, placebo-C	HF, LVEF ≤ 40%	Korea
NCT03611270*	PA denervation	Multicenter, non-randomized, open-label	HFpEF or HFrEF at chronic state	USA
NCT03629340*	Metformin vs. placebo	Single-center, randomized, double-blind, crossover	HF, LVEF ≥ 50%	USA
NCT04945460*	Sotatercept vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF ≥ 50%	USA, Canada, Belgium, Israel, Italy, France, Germany, Poland, Spain, Sweden, UK
EudraCT 2021-005032-30†	Nitrosooxypropanol vs. placebo	Multicenter, open-label	CABG, AVR, or MVR	Sweden
NCT04484675*	Inhaled milrinone vs. IV milrinone	Single-center, randomized, double-blind	Mitral valve surgery with CPB	Egypt
NCT05737940*	Relaxin mimetic AZD3427 vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, any LHD	USA, Austria, Canada, China, Czech Republic, Denmark, Germany, Italy, Japan, Netherlands, Poland, Spain, Sweden, UK
NCT05824923*	PA denervation vs. SoC	Multicenter, randomized, single-blind, controlled	HF, any LVEF	China
NCT05983250*	Levosimendan vs. placebo	Multicenter, randomized, double-blind, placebo-C	HF, LVEF > 40%	USA
NCT03356353*	Sildenafil	Multicenter, single-arm, open-label	HFrEF requiring LVAD	Canada

^*All these trials are ongoing, unfinished, or have posted results only on ClinicalTrials.gov. These trials are registered and available on the National Library of Medicine’s clinical trials database, https://clinicaltrials.gov/. ^†This trial is ongoing, so it does not have posted results. It is registered and available on the EU Clinical Trials Register, https://www.clinicaltrialsregister.eu/ctr-search/search. AS: Aortic stenosis; AVR: Aortic valve replacement; CABG: Coronary artery bypass graft surgery; CHD: Congenital heart disease; CPB: Cardiopulmonary bypass; FIRST: The Flolan International Randomized Survival Trial; HF: Heart failure; HFmrEF: HF with mildly reduced ejection fraction; HFpEF: HF with preserved ejection fraction; HFrEF: HF with reduced ejection fraction; IV: Intravenous; LHD: Left heart disease; LVAD: Left ventricular assist device; LVEF: Left ventricular ejection fraction; MS: Mitral stenosis; MVR: Mitral valve replacement; NA: The institutional research review board approved the study but no NCT number is available; NCT: National clinical trial; NT-proBNP: N-terminal pro-B-type natriuretic peptide; PA: Pulmonary artery; PH: Pulmonary hypertension; Placebo-C: Placebo-controlled; Sham-C: Sham-controlled; SoC: Standard of care; VHD: Valvular heart disease.

The most represented nations were the United States and those in Europe, accounting for 26 trials (43.3%) and 32 trials (53.3%), respectively. Within Europe, the trials in Germany, Italy, Austria, the United Kingdom, and Spain outnumbered those in other countries. Regions other than North America and Europe were much less frequently involved. There were 16 studies (26.7%) that recruited Asian patients, mainly from Israel and China (10% for each nation); only 1 to 3 trials included subjects from the Russian Federation, India, Korea, Singapore, or Japan. Only 1 or 2 trials enrolled Brazilian, Argentinian, Australian, or Egyptian patients.

The tested interventions were drugs, except for 2 investigations of pulmonary artery denervation, which were conducted in China.^[126] One trial in Japan focused on children with PH requiring surgery for congenital heart disease.^[134] The active trial in Egypt is evaluating the effects of milrinone in patients undergoing surgery for mitral valve disease, who also have sPAP ≥ 55 mmHg at TTE.

8. Discussion

The data about PH-LHD epidemiology largely vary across studies in different regions of the world. This discrepancy may have several causes. Investigators may tend to focus on the most relevant type of LHD in a certain area—for example, RHD is a major health issue in Africa—whereas aortic stenosis or cardiac amyloidosis are objects of intense research in industrialized countries because of the treatment options that have been developed in the last years. This leads to a reporting bias. Methodological heterogeneity should be also considered; local availability will dictate reliance on RHC vs. TTE, which in turn affects the accuracy of PH diagnosis. It is possible that patients come to medical attention at different phases of LHD. Since PH becomes more common with LHD progression, the timing of assessment may influence the frequency of PH detection. Supporting the hypothesis that the way studies are conducted influence the estimates of PH-LHD prevalence, we found a fluctuating proportion of patients with PH-RHD only in Africa.

Nonetheless, it cannot be excluded that the burden of PH-LHD is different across the globe due to ethnical and social factors, such as genetic susceptibility, influence of diet, and promptness and completeness of therapy. Intriguingly, there is also a clear imbalance in the geographical distribution of PH-LHD clinical trials. High-income countries normally have well-structured health care systems, ensuring diagnostic capacity, resources for patient recruitment and follow-up, and control of competing diseases, such as infections. These characteristics are preconditions for the successful completion of a clinical study, and it is expected that they are looked for by sponsors and clinical research organizations.

Studies assessing PH with the same approach in similarly selected cohorts worldwide are needed to obtain reliable information about PH-LHD epidemiology. Involvement of centers of excellence in low-income nations may facilitate the design of more inclusive trials. The creation of partnerships between hospitals and funding agencies in high- and low-income nations is another possible strategy to overcome the selection bias in PH-LHD trials.

9. Conclusions

The epidemiology of PH-LHD is characterized by apparent regional variability, with PH-LHD due to mitral stenosis and RHD prevailing in low- and middle-income countries, and PH due to other LHD types being predominant in high-income countries. A bias toward reporting only the most common forms of PH-LHD in a certain area likely enforces these differences. Enrollment in PH-LHD clinical trials has mostly been limited to patients in high-income countries, arguably because of the better context that these offer as compared to low-income nations. At present, there is no global consistency in the evidence informing PH-LHD knowledge and management.

Funding

Italo Porto was supported by the Italian Ministry of University and Research (PRIN 2022 PNRR funded by the European Union - Next Generation EU; project P20225BYWX, CUP D53D23020960001).

Author contributions

Suela Vani and Filippo Novarese contributed to literature search, data synthesis, and writing. Matteo Toma and Italo Porto helped with the design of the study and critically reviewed the article draft. Pietro Ameri conceived the article and contributed to literature search, data synthesis, and writing. All authors read and approved the final manuscript.

Conflicts of interest

Italo Porto received speaker and/or consultant fees from Medtronic, Abbott Vascular, Edwards, ABIOMED, GE, Siemens Healthcare, Sanofi, Amgen, Daiichi-Sankyo, Astra Zeneca, Bayer, PIAM, and Chiesi. Pietro Ameri received speaker and/or advisor fees from Astra Zeneca, Boehringer Ingelheim, Bayer, Daiichi Sankyo, MSD, Janssen, and Gossamer Bio. The Department of Internal Medicine of the University of Genova received fees from Bayer for scientific consultancy performed by Pietro Ameri.