Pulmonary vasculature abnormalities

Vascular remodeling is a defining feature of pulmonary hypertension (PH), characterized by alterations in the vascular and functional systems that primarily affect the distal pulmonary vasculature. In PAH, these changes are most prominent in the small pulmonary arterioles. When remodeling becomes widespread, PVR increases, pulmonary arterial pressures rise, and the RV initially develops adaptive responses that become maladaptive. A central element of this remodeling process is the progressive accumulation of vascular and inflammatory cells within the vessel wall and the surrounding perivascular space. A complex interplay of pathogenic signaling pathways drives this accumulation. These reflect dysregulated endothelial production of vasoconstrictors and vasodilators, impaired proliferative control of pulmonary concomitant artery smooth muscle cells and fibroblasts, and the influence of prothrombotic and inflammatory mediators [13].

Additionally, perturbations in signaling networks, particularly within the TGF-β/BMP axis, play a critical role in promoting and sustaining the remodeling phenotype. PAH disrupts the low-pressure, high-compliance pulmonary circulation [13, 14]. Owing to its distensible and recruitable capillary bed, the pulmonary vasculature can normally accommodate increased blood flow with minimal changes in pressure [10, 15]. However, in PAH, vessels become stiff, resistant, and noncompliant. Complex vascular lesions and disorganized angiogenic structures, possibly arising from bronchial or vasa vasorum origins, are hallmark features of advanced PAH. This “pseudo-neoplastic” transformation involves upregulation of oncogenes, downregulation of Kv channels, and suppression of mitochondrial and apoptotic signaling pathways [13].

Histologically, PAH is characterized by complex vascular lesions, a hallmark of the disease, and advanced vascular remodeling. Additional features include medial hypertrophy, adventitial fibrosis, endothelial proliferation, thrombosis, and complex vascular lesions, which reduce lumen diameter and increase vascular rigidity [13, 14]. There are four types of vascular lesions: Type 1, arising from supernumerary arteries and connecting to the vasa vasorum; Type 2 involves linking the pulmonary arteries to the peribronchial or bronchial circulation. While this can partially relieve suprasystemic pressure, it can also lead to desaturation; Type 3, occurring at abrupt distal arterial terminations with runoff into dilated pulmonary venules; and Type 4, characterized by obstructed arteries undergoing recanalization. In contrast, PVOD shows predominant venous and microvascular remodeling with patchy capillary hemangiomatosis (Fig. 1). According to Poiseuille’s law, resistance increases in proportion to the fourth power of the vessel radius [16], and Laplace’s law states that, for a given pressure, a smaller radius, without wall thickening, increases the wall tension. Initially, adaptive medial thickening may therefore become maladaptive, consequently leading to progressive vascular stiffening and remodeling [14].

Fig. 1

Spectrum of pulmonary vascular diseases in PH. Reproduced from Guignabert C et al. Pathology and pathobiology of pulmonary hypertension: current insights and future directions. Eur Respir J. 2024; 64:2401095. doi:https://doi.org/10.1183/13993003.01095-2024. © The authors 2024. Distributed under the terms of the Creative Commons Attribution Non-Commercial 4.0 License (CC BY-NC 4.0)

In PAH, pulmonary veins (PV) shift from pulsatile inflow to steady capillary perfusion. The stroke volume/pulse pressure ratio (which is microvasculature-dependent) and PVR are inversely related. Together, they form a nearly constant resistance-compliance product, given by PVR × compliance. This equation describes the decline of PA pressure during diastole [17, 18]. Proximal PAs, acting as Windkessel reservoirs, lose their energy-buffering function due to increased pulse wave velocity and early wave reflection [19]. This mechanism induces late-systolic RV overload by altering Pascal’s law, thereby resulting in labile, pulsatile afterload. The resistance-compliance time decreases, reflecting poor buffering and inefficient RV ejection [18]. Decreased ventricular–vascular coupling indicates significant uncoupling and impaired RV-PA interaction [12]. RV–PA coupling can be estimated noninvasively using the TAPSE (tricuspid annular plane systolic excursion) / PASP (PA systolic pressure) ratio, which reflects the balance between RV contractile function and afterload and mirrors the invasive Ees/Ea ratio. A preserved ratio indicates adequate coupling, whereas a declining ratio signals loss of contractile reserve and progressive uncoupling. As a noninvasive surrogate, a TAPSE/PASP ratio < 0.31 mm/mmHg has been associated with this uncoupling and is prognostic of RV failure and adverse outcomes [20]. Progressive RV-PA uncoupling leads to compromised forward flow, as reflected in a reduced RV outflow tract velocity integral. Uncoupling is suggested in right heart catheterization by elevated RA pressure, reduced stroke volume, and disproportionate RV failure relative to afterload. Other echocardiographic parameters include shortened pulmonary acceleration time (< 90 ms), mid-systolic notching, and septal flattening (systolic pressure overload and diastolic volume overload), which together produce a D-shaped LV. CMR reveals reduced myocardial perfusion reserve, particularly in the RV, resulting in coronary flow dysregulation and causing RV ischemia, dysfunction, and failure [21].

Additionally, pulmonary vascular remodeling in PAH is driven by progressive perivascular cellular accumulation and immune dysregulation within the arterial wall. Endothelial cells, smooth muscle cells, fibroblasts, and inflammatory cells interact in the context of disrupted TGF-β/BMP signaling, promoting endothelial-to-mesenchymal transition, cellular hyperproliferation, and extracellular matrix deposition. Vascular rarefaction and the formation of complex vascular lesions further increase PVR, progressively increase RV afterload, and contribute to functional decline. This evolving molecular environment perpetuates disease progression even when hemodynamic parameters initially appear compensated [14].

Impaired RV function and cardiac remodeling

As PVR increases and the proximal PAs stiffen, reflected pressure waves return earlier. This increases the late-systolic RV load, according to Pascal’s law. This phenomenon is evident in invasive hemodynamics as a late-systolic pressure rise. It correlates with echocardiographic signs, such as shortened PA acceleration time, RV outflow tract notching, and impaired strain or TAPSE [22, 23]. Once the contractile reserve is exhausted, RV-PA uncoupling occurs [20]. The RV, a thin-walled, compliant chamber suited for high-volume, low-pressure flow, is poorly equipped to handle the elevated afterload resulting from a stiffened pulmonary vasculature. Initially, the RV compensates through concentric hypertrophy and increased contractility, referring to the intrinsic myocardium force-generating capacity and systolic elastance, facilitated by autoregulation and the Frank-Starling mechanism. As the PVR increases 5 to 10-fold, the RV is subjected to higher afterload, and remodeling becomes maladaptive. The following characteristics define this phase: reduced RVEF, dilation, wall thinning, fibrosis, elevated RV end-diastolic pressure despite reduced RV filling and effective preload, impaired RV-PA coupling, and tricuspid regurgitation [24, 25]. CMR reveals this through reduced RV volumes, reduced RVEF, and late gadolinium enhancement at septal hinge points and in shear-stress-related fibrotic regions, all of which correlate with RV dysfunction and outcomes [21]. Functionally, RV strain, especially in the free wall, declines early, before systolic dysfunction [26]. However, persistent overload lowers the RVEF below 35%, inducing systemic hypoperfusion.

Reduced myocardial perfusion reserve in PAH may contribute to impaired myocardial oxygen supply-demand balance and ventricular mechanical dysfunction [27]. The RA enlarges as RV compliance declines; an RA volume index > 74 mL/m² predicts a worse prognosis [28]. Systemic conditions, such as systemic sclerosis, can cause early RV dysfunction even with a modest hemodynamic burden [29, 30]. As PA pressures rise, right coronary perfusion decreases, and RV ischemia becomes a major contributor. Additionally, microvascular rarefaction, metabolic shifts, reduced oxidative phosphorylation, increased glycolysis, and fatty acid oxidation further impair RV contractility [13]. This progressive transition from adaptive RV hypertrophy to maladaptive remodeling and eventual RV–PA uncoupling is paralleled by worsening hemodynamics, rising PVR, and declining CO. Ultimately, RV failure is a key determinant of clinical outcome in patients with PAH. Figure 2 shows the stepwise pathophysiological cascade integrating vascular, ventricular, and hemodynamic changes.

Fig. 2

Progression of PAH from adaptive RV remodeling to RV failure. Schematic representation of PAH progression integrating echocardiographic changes (upper panels), vascular remodeling (lower panels), and hemodynamic changes (central graph). The stages of development are illustrated across three stages: 1) a compensated phase (RVEF >54%) with adaptive RV hypertrophy and preserved CO; (2) a symptomatic/unstable phase (RVEF 37–54%) marked by progressive RV–pulmonary arterial uncoupling; and (3) an advanced right heart failure phase (RVEF <37%) characterized by RV dilation, severe tricuspid regurgitation, reduced forward flow, and LV underfilling. In the central hemodynamic graph, red solid curves represent PAP and VM, both progressively increasing; the purple dashed curve represents PVR, which rises steadily; and black dotted curves represent SvO₂, CO, and RVSV, which decline following RV–PA uncoupling. The horizontal dashed line indicates the clinical symptom threshold. Advanced stages demonstrate interventricular septal shift and reduced LV cavity size, reflecting ventricular interdependence and late biventricular involvement.

Abbreviations: CO: Cardiac output; LA: Left atrium; LV: Left ventricle; PA: Pulmonary artery; PAH: Pulmonary arterial hypertension; PAP: Pulmonary arterial pressure; PVR: Pulmonary vascular resistance; RV: Right ventricle; RVEF: Right ventricular ejection fraction; RVSV: Right ventricular stroke volume; SvO₂: Mixed venous oxygen saturation; VM: Ventricular mass.

The left heart and its coupling with the pulmonary circulation

Although normal filling pressures define PAH, previous reports suggest that LV mechanics may be altered [31]. As the PVR and RV function deteriorate, LV filling is limited by reduced preload and interventricular septal shift. This phenomenon further contributes to LV dysfunction [32]. Even mild PAH can reduce LV output during stress, and targeted RV volume reduction (e.g., via inferior vena cava occlusion) enhances LV preload, demonstrating the systemic ripple effect of RV strain [33,34,35]. Other contributors to cardiac workload, including decreased LV torsion, increased RV stress dysfunction, ventricular desynchrony, pericardial constraint, and heart rate, are also likely to contribute to PAH-associated LV changes [31, 36].

Left atrial dysfunction

Reduced pulmonary capillary blood return due to increased PVR and RV dysfunction in advanced PAH leads to decreased LA preload, resulting in diminished LA volume and reduced LV volume [37]. Reports suggest potential atrial remodeling secondary to hemodynamic changes, such as reduced pulmonary venous return; however, strong evidence on LA fibrosis remains limited. A cohort of 326 patients with PAH was analyzed to assess the relationship between RA remodeling and the risk of atrial fibrillation. The LA volume index remained within the RA limits, indirectly suggesting an absence of significant hemodynamic alterations involving the LA [38]. The current evidence suggests that chronic underfilling predominantly drives functional impairment [1, 36], while structural remodeling remains insufficiently characterized.

In a cross-sectional CMR study by Leong et al., the LA volume index and peak LA strain distinguished pre- and post-capillary PH, with patients with idiopathic PAH having normal LA volumes. Nevertheless, they found a reduced LA strain relative to healthy controls, suggesting that functional LA impairment may occur independently of volumetric changes, possibly secondary to reduced preload due to RV failure [38]. Left atrial reservoir strain (LARS), assessed by speckle-tracking echocardiography, has emerged as a sensitive marker of impaired LA compliance and reduced pulmonary venous return in PAH [18, 37]. The LA strain may be a more sensitive and earlier marker of LA dysfunction than volume alone, potentially aiding in risk stratification and early detection of left-sided heart involvement [35, 39]. Recent evidence from Guo et al. [1], demonstrated a modest inverse relationship between LA global strain and the transpulmonary gradient in pre-capillary PH. Another possible explanation is subclinical diastolic dysfunction without LA enlargement [1]. However, more research is needed to explore this observation.

Left ventricular underfilling

RV enlargement directly affects LV diastolic filling through ventricular interdependence, LV end-diastolic volume (LVEDV), and stroke volume. Joong-Gan et al. [39] reported echocardiographic findings that support this hypothesis, demonstrating: (1) early diastolic filling rate is closely linked to interventricular septum curvature at the time of maximal curvature, (2) this filling rate correlates with LVEDV, and (3) LA filling was standard, and with no relation to LVEDV, suggesting LV underfilling is predominantly driven by mechanical septal displacement rather being purely a consequence of LA volume depletion. Advanced PAH compromises LV filling through two parallel mechanisms: serial interaction and direct ventricular interaction. While chronically increased PVR and RV dysfunction reduce pulmonary capillary blood return, thereby reducing baseline LA preload and volume (serial interaction), the dynamic underfilling of the LV during early diastole is largely dictated by direct ventricular interaction. Specifically, right-to-left interventricular septal displacement mechanically restricts the LV cavity in early diastole, an impairment that operates independently of changes in LA volume (Fig. 3)

Fig. 3

Temporal Evolution of Left Heart Involvement in PAH. Schematic representation of the temporal evolution of left-heart involvement during PAH. The upper panels illustrate progression from compensated RV pressure overload to RV–PA uncoupling and ultimately chronic LV underload remodeling. Early disease is characterized by preserved LV filling and output despite rising PVR. As RV–PA uncoupling develops, reduced transpulmonary forward flow, septal shift, and pericardial constraint impair pulmonary venous return, progressively lowering LV preload, LVOT-VTI, SV, and CO. Advanced disease is associated with a small-cavity LV phenotype, reduced contractile reserve, diminished LVEDV, and impaired LA/LV deformation reserve.The lower panel schematically depicts the temporal trajectories of these physiologic changes during PAH progression, demonstrating rising RV afterload/PVR alongside progressive reductions in pulmonary venous return, LV preload, LVOT-VTI, SV/CO, and LA/LV deformation reserve.

Abbreviations: CO: Cardiac output; LA: Left atrium; LV: Left ventricle; LVEDV: Left ventricular end-diastolic volume; LVOT-VTI: Left ventricular outflow tract velocity-time integral; PAH: Pulmonary arterial hypertension; PVR: Pulmonary vascular resistance; RV: Right ventricle; RV–PA: Right ventricular–pulmonary arterial; SV: Stroke volume.

A mechanistic approach using PV area and flow has been proposed to assess LV underfilling. A recent CMR imaging study demonstrated reduced LV size and end-diastolic volume compared with healthy controls [37]. Notably, PV area was not associated with LV global longitudinal strain (LV-GLS), suggesting that LV dysfunction in PAH is primarily driven by underfilling rather than by an intrinsic myocardial defect. Greater RV dysfunction, abnormal septal configuration, and reduced LV preload are other mechanisms of LV-GLS impairment [39]. These findings underscore the potential value of LV underfilling and PV area measurement as prognostic markers of LV dysfunction in patients with PAH. LV underfilling has been independently associated with adverse clinical outcomes and correlated with more advanced disease [3].

LV underfilling and underload lead to remodeling over time. CMR studies by Simpson et al. [2] and Sjögren et al. [3] have shown that predicted LV mass values are lower in idiopathic and connective-tissue disease PAH subgroups [2], suggesting atrophy secondary to LV underfilling. Normal fluctuations in LV filling are reduced in a sheltered and unchallenged LV, thus eliminating a myocardial growth factor [36]. Chronic underutilization provides the most plausible explanation for LV atrophy. In other settings where the LV is underutilized, such as mitral stenosis, bed rest, zero-gravity conditions, or certain congenital heart diseases, a decrease in its mass has also been reported [36]. However, Guo et al. [1] found that in patients with PAH, the LV undergoes cavity shrinkage and elevated filling pressures. However, no reduction in myocardial mass was observed, resulting in LV underfilling without LV atrophy. Whether chronic LV underfilling leads to LV atrophy in PAH remains controversial. In this context, it is important to distinguish LV preload from LV filling pressure. Although these terms are often used interchangeably in clinical practice, preload more accurately refers to LV end-diastolic myocardial stretch or chamber filling volume. In contrast, filling pressure reflects the pressure measured within the ventricle at end diastole. In PAH, LV preload is typically reduced because of chronic LV underfilling, septal displacement, and impaired pulmonary venous return. However, LV filling pressures or LV end-diastolic pressure may be normal or even elevated due to altered ventricular compliance, pericardial constraint, and geometric distortion related to ventricular interdependence. Accordingly, elevated LV end-diastolic pressure does not necessarily indicate increased effective LV preload in PAH.

Left ventricular atrophy and dysfunction

Several complementary mechanisms have been proposed to explain LV dysfunction in PAH, including chronic LV underfilling secondary to ventricular interdependence, structural remodeling related to reduced LV workload and atrophy, and secondary intrinsic myocardial alterations involving sarcomeric and molecular remodeling. The following section discusses the available evidence supporting each of these proposed mechanisms. Chronic LV underfilling leads to LV preload deficiency and hypovolemia. Although LVEF may remain within normal limits, global intrinsic LV force-generating capacity may remain relatively preserved despite evidence of molecular, sarcomeric, and ultrastructural remodeling described in advanced PAH. An adaptive atrophic change opposite to RV hypertrophy could explain this phenomenon [36]. The combination of low flow and small left-sided chambers will reduce the need for high contractile force on the left side due to decreased myocardial stretch [4], resulting in an unchallenged LV that becomes deconditioned to higher workloads. Evidence has shown reduced LV contractility and strain, as well as lower LV mass, in patients with PAH compared with healthy controls [31, 40]. Additionally, another study shows that patients with PAH had significantly elevated myocardial fibrosis markers and increased LV T1, which may induce atrophy-related extracellular changes, such as fibrosis and collagen deposition [40]. Importantly, several experimental studies have failed to demonstrate significant LV fibrosis or major collagen remodeling in PAH models despite evidence of altered LV geometry and function. Ishikawa et al. reported no significant increase in LV collagen content in monocrotaline-treated rats. At the same time, Fowler et al. similarly found no increase in myocardial fibrosis despite evidence of ventricular dysfunction [41,42,43]. Nevertheless, some studies have demonstrated molecular, neurohumoral, and electrophysiologic alterations in the LV despite the absence of overt fibrosis or hypertrophy. Lourenço et al. demonstrated impaired LV contractility and relaxation associated with neurohumoral activation without structural remodeling. At the same time, Hardziyenka et al. identified electrophysiologic remodeling and connexin alterations in the atrophic LV of pressure-overloaded RV failure models [44, 45]. These observations suggest that secondary myocardial adaptation may occur in advanced disease, although the extent to which these changes represent intrinsic irreversible myocardial dysfunction remains uncertain. The LV-GLS and left atrioventricular plane displacement have also been shown to be markedly reduced in patients with PAH, further suggesting impaired LV contractility [4]. Molecular analysis of PAH myocardial biopsies by Manders et al. [