According to the current guidelines for the diagnosis and treatment of pulmonary hypertension (PH) [1, 2], the Dana Point diagnostic classification separates pre-capillary PH, i.e. PH due to pulmonary vascular disease mainly affecting the pre-capillary arteriolar compartment (i.e. Dana Point groups 1, 3, 4 and 5), from post-capillary disease that originates distal to the capillaries and involves morphological changes in the pre-capillary compartment only occur after a significant pressure increase in the venous compartment (i.e. Dana Point group 2). Currently, a single haemodynamic parameter, i.e. wedge pressure [3], that is commonly flawed by methodological errors [4] is employed to distinguish between pre- and post-capillary disease. This distinction is vital because treatment is completely different for the two classes of disease [2], with a virtual “therapeutic gap” between the two disorders. For example, epoprostenol, the life-saving treatment for pulmonary arterial hypertension (PAH) [5], was shown to be detrimental in congestive heart failure [6].
PH with left-sided heart disease is classified as a non-PAH form that includes all conditions associated with increased left ventricular filling pressure. A transpulmonary gradient (mean pulmonary arterial pressure (Ppa)−pulmonary capillary wedge pressure (Ppcw)) >12 mmHg has been arbitrarily declared as the haemodynamic threshold indicating significant pre-capillary pulmonary vascular disease in left-sided heart disease, and has been classified as “reactive” or “out-of-proportion” PH. By contrast, transpulmonary gradients ≤12 mmHg have been classified as passive PH, i.e. PH due to hydrostatic pressure transmitted across the capillary bed of the lung, implying a lack of significant anatomical changes of the pre-capillary vessels. Naturally, these pressures can be directly changed by relieving left-sided pressures, for example by diuretic treatment, mitral valvulotomy in the case of mitral stenosis, or surgical mitral valve replacement. Whereas mitral stenosis was the most frequent cause decades ago, the most common causes of PH with left-sided heart disease today are systemic hypertension alone or in combination with metabolic syndrome and ischaemic heart disease, conditions in which PH develops as a consequence of compromised left ventricular relaxation and distensibility. An alternative approach, that has been proposed in the past to differentiate between isolated passive transmission of the left-sided filling pressures and concurrent remodelling of the pulmonary venules, may be the incorporation of the diastolic Ppa−mean Ppcw gradient into the diagnostic work-up of affected patients [7].
Although PH with left-sided heart disease is a common entity [8] associated with an increasing prevalence in the ageing population [9], and long-term follow-up trials have provided strong evidence that PH in left heart disease carries a worse outcome [10], available data on incidence, pathophysiology and therapy are sparse, and evidence providing firm grounds for haemodynamic classification is lacking. Therefore, basic and clinical work focusing on this important disease entity is needed.
The backward haemodynamic consequences of left-sided heart diseases are thought to progress from venous leakage to lung capillary injury involving impaired Ca2+ signalling and cytoskeletal reorganisation [11], plexiform lesions, arteriolar changes comprising medial hypertrophy, and intimal fibrosis with eventual right ventricular overload and failure. In the current issue of the European Respiratory Journal, Hoffmann et al. [12] have employed supracoronary aortic banding in Sprague-Dawley rats as an experimental model for PH owing to left heart disease and monocrotaline (MCT) injection to create PAH and provide unique comparative insights into both pathological mechanisms. Whole rat genome microarray analyses of lungs from banded and control rats deliver novel gene expression information related to these two experimental conditions. Unfortunately, the venous and arterial compartments have not been separately investigated in these studies. Still, molecular analysis in combination with genetically modified Ws/Ws rats, which are deficient in mast cells, and a pharmacological approach by treatment of aortic banded rats with the mast cell stabiliser ketotifen have provided the simple concept that mast cells play a role in the pathogenesis of both disorders.
However, whether whole genome approaches that are based on the magnitude of the effect truly reflect biologically important mechanisms is uncertain. It is well-known that minor gene expression differences may implicate large biological effects. For example, a three base pair deletion within the cystic fibrosis transmembrane conductance regulator protein causes major disease [13]. Most prominently upregulated gene ontology clusters may reflect the presence of distinct cell populations, albeit without proving pathophysiological evidence for causality to a certain disorder. The authors were successful in providing evidence beyond the expression analysis by utilising Ws/Ws rats deficient in mast cells. Still, the proof of causality is not complete as mast cell activation is complex and could be secondary to another principle that is operative in both pre- and post-capillary PH.
Mast cells release biogenic amines, including serotonin, which play a key role in pulmonary arterial vasoconstriction and smooth muscle cell proliferation [14], and histamine, which acts as a vasoconstrictor in pulmonary veins. In the circulation, thrombosis is driving mast cell accumulation [15]. However, Hoffmann et al. [12] observed that vascular thrombosis was not common in lungs of MCT-treated rats or in those from rats with aortic banding. Mast cells were most prominently found in the vicinity of pulmonary arteries and arterioles, with the constraint that remodelled pulmonary veins may easily be misinterpreted as arteries. Capillary haemorrhage could be seen in some areas of lungs with PH, yet did not co-localise with mast cell accumulations. However, visualisation of thrombosis in pathologic vascular specimens may be difficult because fresh thrombus is frequently fragile and disappears during tissue harvest due to unstable fibrin crosslinking [16]. Thus, mast cells may associate with fresh thrombus that escaped detection in the histological specimens, and the authors’ finding may indirectly indicate the pathophysiological importance of small vessel thrombosis in both conditions.
Mast cells may be involved in a variety of adaptive or pathological responses associated with chronic inflammation and it has been demonstrated that mast cells contribute to bronchial or cardiac remodelling. Growth and function, i.e. mediator production and secretion of mast cells, are regulated by mast cell growth factor [17], also called stem cell factor or kit ligand [18]. This cytokine represents the ligand of the c-kit tyrosine kinase receptor clustered as CD117, a marker for bone marrow (BM)-derived haematopoietic stem cells and mast cells. Mast cells are increased in the lungs and remodelled vessels of experimental PH and human idiopathic PAH [19–21]. In idiopathic PAH, the accumulation of c-kit+ cells was associated with an increased expression of c-kit mRNA in microdissected pulmonary arteries and with an increase of soluble c-kit plasma levels compared to controls [22]. The involvement of lung mast cells has recently also been reported in animal models of hypoxic PH [23]. The implication of c-kit+ cells in the pathophysiology of PH may have clinical implications in the development of innovative therapy for this devastating disease. Tyrosine kinase inhibitors, such as imatinib, or combined tyrosine and multiple kinase inhibitors, such as sorafenib, improve experimental PH in animal models [24–26] and imatinib appeared favourable in human PAH [27]. A main effect of tyrosine kinase inhibitors appears to be related to the inhibition of the platelet-derived growth factor receptor that attenuates migration and proliferation, and enhanced apoptosis of pulmonary arterial smooth muscle cells. By reducing perivascular accumulation of c-kit+ cells, mainly BM-derived progenitor cells but also mast cells in pulmonary arteries of mice exposed to chronic hypoxia, imatinib improved pulmonary vascular remodelling [28, 29]. These data indicate that the beneficial effect of tyrosine kinase inhibitors in experimental or human PAH may be partly due to the inhibition of c-kit+ cells, which include mast cells.
Hoffmann et al [12] augment this information, suggesting that mast cells are implicated in pulmonary vascular remodelling in experimental models of both PAH and PH due to left heart diseases. Thus, mast cells may provide the missing link between pre- and post-capillary PH. Future experimental studies are needed to improve the understanding of mechanisms by which mast cells promote pulmonary vascular remodelling. Mast cell targeted therapy, including inhibition of the tyrosine kinase c-kit, may represent an innovative treatment in PH, including the common variant of PH due to left heart disease.
Footnotes
Statement of Interest
Statements of interest for D. Montani and I.M. Lang can be found at www.erj.ersjournals.com/site/misc/statements.xhtml
- ©ERS 2011