Sarcoidosis affects the lungs and thoracic lymph nodes in more than 90% of patients [31]. In the majority of affected people, acute granulomatous inflammation is self-limiting and resolves spontaneously with time. In about one-third of the cases, the inflammation becomes chronic. Subsequently, about 20% of patients with persistent inflammation develop lung fibrosis [32]. Based on chest X-ray appearance, five stages of intrathoracic sarcoidosis are distinguished, as presented in Table 1 [33].

Pulmonary involvement increases the risk of PH almost five times compared to non-pulmonary sarcoidosis. In the population from the Nationwide Inpatient Sample 2016–2018, the prevalence of PH was 19% in cases with pulmonary sarcoidosis and 3.5% with non-pulmonary sarcoidosis; however, there was no information on pulmonary sarcoidosis staging [34]. In published research, 65–80% of sarcoidosis patients with pre-capillary PH confirmed by RHC had stage 4 disease, as assessed using either chest X-ray or HRCT [4,5,6,7,8]. In about half of these patients, PH was mild, i.e., mPAP ≤ 35 mmHg [4,5]. Surprisingly, no correlation could be found between mPAP or pulmonary vascular resistance (PVR) and spirometric or plethysmographic parameters, and patients with similar radiologic and pulmonary function characteristics behaved differently in terms of PH development [4,5,8,24,35].

The ablation, compression, and distortion of pulmonary vessels, together with hypoxic pulmonary vasoconstriction (HPV), are believed to be the first elements of the pre-capillary PH development pathway in the parenchymal phenotype of SAPH by analogy to other chronic lung diseases [36,37]. HPV is a compensatory mechanism to maintain ventilation/perfusion homeostasis by diverting blood flow from more seriously to less seriously affected lung segments. Depending on the magnitude of the pulmonary vascular bed loss and the magnitude of HPV, pulmonary artery pressure may increase, albeit usually mildly. In some yet not identified circumstances, the increased shear stress from high blood flow induces a response from pulmonary vessels that leads to their remodelling and irreversible structural changes, resulting in significant elevations of PVR and PAP. This interaction between increasing shear stress and pulmonary vasculature remodelling may continue in a vicious cycle [38,39,40,41,42]. Remodelling involves all levels of the pulmonary vasculature and all layers of the vessel walls [40,43,44,45]. ¹⁸F-fluorodeoxyglucose (¹⁸F-FDG) uptake was found on positron emission tomography (PET) in the pulmonary artery wall in 19% of 175 sarcoidosis patients, and these patients had higher PAP and PVR compared to patients without the uptake [46]. The meaning of this phenomenon is not clear, but one of the theories is that it reflects the inflammation caused by the shear stress and could be helpful in the early detection of developing PH [47].

The distinctive features of sarcoid inflammation and fibrosis, including a specific lymphangitic fashion of distribution, can be responsible for differences observed between SAPH and PH due to other interstitial lung diseases (ILDs). For example, the same degree of fibrosis is associated with higher PAP in sarcoidosis than in idiopathic pulmonary fibrosis (IPF) and other ILDs [48,49]. The predilection to bronchovascular bundles, septa, and subpleural regions may be associated with a larger direct impact on nearby pulmonary vessels [50].

A clear definition of a parenchymal SAPH phenotype is lacking. Mathijssen et al. adopted the recommendations regarding PH associated with chronic lung diseases from the latest World Symposium on PH to define the criteria of a parenchymal SAPH. These criteria include pre-capillary PH and moderate to severe pulmonary parenchymal disease; the latter was defined as stage 3 or 4 pulmonary sarcoidosis or severe obstructive (forced expiratory volume in 1 s (FEV₁) ≤ 60%pred) or restrictive disease (forced vital capacity (FVC) ≤ 70%pred). They found the parenchymal phenotype in 73% of the patients from their SAPH cohort. Interestingly, not all stage 4 patients with pre-capillary PH were classified as the parenchymal phenotype despite the used definition; 17% of them were assigned to a compression phenotype, and one was diagnosed with chronic thromboembolic PH. Patients with stage 2 pulmonary sarcoidosis also fulfilled the criteria of the parenchymal phenotype. Additionally, one patient with stage 2 disease was categorised as having a parenchymal phenotype despite not fulfilling the criteria, but it was decided that this category was the best fit [7].

There are no evidence-based recommendations for the treatment of SAPH with predominant parenchymal causes. As for now, the treatment of underlying diseases and the symptomatic treatment of respiratory and/or heart failure, if needed, is a standard of care [1,10]. If SAPH is accompanied by active parenchymal inflammation, treatment with glucocorticoids and/or other anti-inflammatory medications is warranted. However, there is no solid proof of their effectiveness [10,31,51,52]. The use of immunosuppressive therapy in a SAPH setting is very common, it was applied in 26%-77% of SAPH patients from registries [8,19] and in up to 80% of retrospectively analysed cohorts [7], including SAPH with fibrotic pulmonary disease [53]. Reports on treatment outcomes are very limited and divergent [4,8]. Of note, haemodynamic improvement has been observed in some SAPH patients with stage 4 pulmonary disease, and reporting authors speculated the effect depended on a component of active inflammation [8]. There is clinical, histopathological, and metabolic evidence that the majority of patients with fibrotic pulmonary sarcoidosis have concomitant active granulomatous inflammation [54,55,56,57,58,59]. Up to 85% of patients with stage 4 pulmonary sarcoidosis showed active inflammation in lung parenchyma on ¹⁸F-FDG-PET [57,58,59]. Volumetric quantitative ¹⁸F-FDG-PET measurements showed a similar burden of active pulmonary inflammation in stage 4 as in stages 2–3, with the extent of the inflammation area corresponding to about 45% of lung volume [57]. Distinguishing the burnt-out disease from fibrotic pulmonary sarcoidosis with smouldering inflammation can be challenging when based only on conventional radiologic studies [60,61]. The usefulness of serum biomarkers in predicting the activity of sarcoidosis has been researched, but none appeared to be accurate enough [57,58,59,62,63,64,65].

The patency of pulmonary vessels, both arteries and veins, may be compromised by extrinsic compression caused by active sarcoidosis-based inflammation, fibrous lung parenchyma, lymph nodes, or fibrosing mediastinitis. The prevalence of mediastinal lymphadenopathy in sarcoidosis exceeds 90% [33]. The prevalence of fibrosing mediastinitis is unknown, but sarcoidosis may account for about 11% of fibrosing mediastinitis [95] and 42% of fibrosing mediastinitis causing PH [96]. The phenomenon of extrinsic vessel compression and its significance in SAPH development is poorly explored and understood [4,7,8,10,97,98,99]. The frequent compression of proximal pulmonary arteries, occasionally with luminal obliteration, has been revealed in the postmortem examination of the lungs of sarcoidosis patients [100]. Mathijseen et al. defined compression phenotype as a precapillary PH with the compression of central or segmental pulmonary arteries in imaging studies. They found this phenotype in 15% of their patients, with compression caused by either lung fibrosis, fibrosing mediastinitis, calcified lymph nodes, or active inflammatory masses. Among these six patients, five had stage 4 pulmonary sarcoidosis, and one had stage 1 disease. In two patients, including one with stage 1 pulmonary sarcoidosis, compressing structures were metabolically active on ¹⁸F-FDG-PET [7]. In a French cohort of 126 patients with severe pre-capillary SAPH, 4% had extrinsic compression of pulmonary arteries due to lymphadenopathy (two cases) or fibrosing mediastinitis (three cases). An increased uptake of ¹⁸F-FDG on PET was present in three patients: two with compressive lymph nodes and one with fibrosing mediastinitis [8]. In another cohort of 22 pre-capillary SAPH patients, the extrinsic compression of large pulmonary arteries by lymph nodes was recognised in 14% of the patients, all stage 4 disease [4]. In a meta-analysis performed on 17 SAPH patients who underwent pulmonary angioplasty for focal stenosis or external compression of the pulmonary vessels, the most frequently affected were the right pulmonary artery and its branches, and the least frequently affected were the pulmonary veins. Usually, multiple bilateral stenosis was present [97]. Extrinsic pulmonary vessel compression may appear in any stage of pulmonary sarcoidosis [98] and may be detected on contrast-enhanced CT, HRCT, or pulmonary angiography [4,7,8,97,98].

The assumption that the compressive phenotype of SAPH is confined to pre-capillary PH seems controversial and may be responsible for the underestimation of the prevalence of this phenotype. In a cohort of 27 patients with fibrosing mediastinitis and PH, 18% had post-capillary PH without concomitant cardiac disease. Sarcoidosis patients accounted for 48% of the whole group. In 52% of the patients, severe compression of pulmonary veins was present, the most probable cause of elevated pulmonary arterial wedge pressure (PAWP) [96].

Mathijssen et al. defined a vasculopathy phenotype as a pre-capillary PH with PVR > 3 Wu, no or mild pulmonary disease, and excluded other causes of PH. Based on that, they classified only 1 patient of the group of 40 as a suspected vasculopathy phenotype [7]. Yet, the lung pathological evaluation proves that intrinsic vascular involvement at all levels of pulmonary vasculature is very common in sarcoidosis. It appears in two forms: sarcoidosis-specific granulomatous angiitis and non-specific microangiopathy [4,100,101,102]. Granulomatous pulmonary angiitis means the presence of granulomas within walls of blood vessels with the destruction of lamina elastica and was found in 100% of sarcoidosis cases in specimens from autopsy series [100], in 69% of specimens from open lung biopsies [102], and in 41% of transbronchial biopsies [101]. It was also found in 80% of explanted SAPH lungs [4]. Both granulomas and healed lesions of various stages were observed, often coexisting. Granulomatous vascular involvement correlated positively with the extent of granulomas in lung parenchyma, but there was no correlation between the occurrence of granulomatous angiitis and radiographic pulmonary sarcoidosis stages [100,101]. Interestingly, venous involvement predominated and accounted for up to 90% of cases [4,100,101,102]. Granulomas were found in vessels of all calibres, but 75% of the involved veins and 54% of arteries were less than 100 µm in diameter [101]. Microangiopathy refers to the alterations of endothelial and basement membrane of precapillary, capillary, and postcapillary pulmonary vessels and was found in 35% of transbronchial biopsy specimens from patients with sarcoidosis [101] and 100% of explanted lungs of five patients with SAPH [4]. Both angiitis and microangiopathy caused vessel occlusion and destruction [4,45,50,100,101,102,103,104]. Plexiform lesions were not typically seen in sarcoidosis-associated vasculopathy [4,100,101]; however, venous involvement with vessel obliteration and subsequent capillary congestion may resemble pulmonary veno-occlusive disease (PVOD) [4,105,106]. The assumption that pulmonary vasculopathy is to be expected only in cases with mild pulmonary disease and severe PH, in the absence of other possible causes, is debatable in light of the facts presented above. Most probably, it contributes significantly to the development of SAPH believed to be of predominantly different phenotypes, and it may be responsible for severe or out of proportion to perceptible factors PH. Also, given the vast venous involvement, a post-capillary PH could be expected. A negative correlation between mPAP or PVR and TLco was found in SAPH, and septal lines and ground glass opacifications appeared more often on the lung HRCT of SAPH patients compared to sarcoidosis patients without PH [4,7]. These clinical and radiologic features are known to occur commonly in PH with overt involvement of pulmonary veins [1]. Pulmonary vasculopathy is potentially the most underappreciated mechanism of SAPH, as it is difficult to establish the diagnosis and to assess the extent of vascular involvement antemortem.

Immunosuppressive agents are suggested as the first-line treatment when sarcoidosis vasculopathy is suspected, with the option of the pulmonary vasodilators used in some patients [10]. In patients with venous involvement resembling PVOD, pulmonary vasodilators should be used carefully in order to avoid worsening pulmonary congestion [1,10].

It is believed that post-capillary SAPH results from cardiac involvement in the course of sarcoidosis that causes LV function impairment. This hypothetical cause-and-effect relationship has not been validated practically, though no studies addressing this topic can be found. Cardiac involvement has been observed in up to 27% of Caucasian and African American sarcoidosis patients and even up to 80% of Japanese patients; in the majority of them, it was clinically silent [107,108,109]. Post-capillary PH accounted for 7–28% of SAPH cases in reported groups [5,7,19]. In the group reported by Mathijssen et al., none of the patients with post-capillary SAPH had cardiac sarcoidosis, and all had preserved systolic LV function [7]. In another study, 65% of post-capillary SAPH patients had normal LV ejection fraction, data on cardiac involvement were not reported. In that study, patients with post-capillary SAPH and impaired LV function were clinically indistinguishable from patients without LV disease but had better clinical outcomes [24]. It is also interesting that no pulmonary hypertension was observed in a French cohort of cardiac sarcoidosis patients despite systolic or diastolic LV dysfunction in many of them [109]. These data may undermine the theory of such a straightforward cause-and-effect relationship between post-capillary SAPH and cardiac sarcoidosis. An alternative explanation could include the known vast intrinsic and extrinsic involvement of pulmonary veins of all calibre in sarcoidosis [4,97,100,101,102]. In the cohort of sarcoidosis patients listed for LTx, the PAWP was significantly higher in the group with PH compared to patients with no PH [35]. Unfortunately, patients with post-capillary PH are often excluded from studies on SAPH [5,6,8], and reports on cardiac sarcoidosis do notexplore this intriguing plot.

In post-capillary SAPH due to cardiac sarcoidosis, glucocorticoids and other immunosuppressive agents are recommended. In addition, the standard treatment for cardiac failure should be applied as appropriate [1,10,51]. This SAPH phenotype requires more attention in future studies.

The liver involvement can be found in 50–75% of sarcoidosis patients, but symptomatic disease occurs in only 5–15% [110,111]. Portal hypertension (PoH) develops in less than 1% of sarcoidosis patients [111] and in up to 33% of patients with proven hepatic sarcoidosis [110]. PoH can be caused by biliary cirrhosis or the compression of the portal vein by involved lymph nodes or parenchymal granulomas, as well as sarcoid hepatic vasculitis [110,111]. It is believed that 2–6% of patients with portal hypertension of various aetiology develop pulmonary arterial hypertension, also called portopulmonary hypertension (PoPH) [1]. The overall occurrence of PoPH in sarcoidosis seems very rare, but it should not be forgotten in a differential workup of SAPH.

Glucocorticoids and other immunosuppressive agents are recommended for PH caused by sarcoid liver disease [10]. Pulmonary vasodilators should also be considered [1].