|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Rheumatology (Dr. Johnson) and Pulmonary Hypertension Centre (Dr. Granton), University Health Network, University of Toronto, Toronto; and Centre for Critical Illness Research (Dr. Mehta), Lawson Health Research Institute, Division of Respirology, London Health Sciences Centre, London. ON, Canada.
Correspondence to: Sanjay Mehta, MD, FCCP, Southwest Ontario Pulmonary Hypertension Clinic, Division of Respirology, Department of Medicine, London Health Sciences Center, Victoria Hospital, Room E2.624, Professional Block, 800 Commissioners Rd East, London, ON, Canada, N6A 4G5; e-mail: sanjay.mehta{at}lhsc.on.ca
Abstract
The role of thrombotic arteriopathy in the pathophysiology of idiopathic pulmonary arterial hypertension (IPAH) and the use of anticoagulants in the treatment of IPAH are currently controversial issues. This article reviews the evidence for a role of vascular thrombosis in the pathophysiology of IPAH. There is sufficient biological rationale to support the notion that thrombotic arteriopathy is an important pathophysiologic feature of pulmonary arterial hypertension (PAH) and that its progression materially contributes to disease progression. To date, the data from observational studies suggest that anticoagulation with warfarin is an effective intervention in patients with IPAH. Its efficacy in other causes of PAH remains speculative.
Key Words: idiopathic pulmonary arterial hypertension pulmonary hypertension thrombosis thrombotic arteriopathy
Pulmonary arterial hypertension (PAH) is a disease state characterized by an increase in pulmonary vascular resistance to blood flow across the pulmonary circulation, and results in an elevation of pulmonary artery pressure (PAP). The ensuing progressive right ventricular failure leads to progressive disability and frequent death of patients with PAH. PAH is most commonly associated with underlying systemic disease (APAH), eg, connective tissue disease (CTD), as well as congenital heart disease (CHD). PAH may also arise in the absence of underlying disease (idiopathic PAH [IPAH]), formerly called primary pulmonary hypertension.1 Although IPAH is a rare condition, PAH is a more common and increasingly recognized clinical problem.
PAH is a complex pulmonary vascular disease that demonstrates a broad range of abnormal vascular responses and pathology. The pathophysiology of PAH was initially thought to consist principally of abnormal pulmonary vasoconstriction. However, the lack of a vasodilator response in many patients with PAH suggested the presence of pulmonary vascular abnormalities other than a simple increase in vasomotor tone. Indeed, a pathologic study2 has demonstrated chronic alterations in the structure and composition of the walls of the pulmonary arteries, commonly referred to as remodeling. These alterations consist of complex disturbances in vascular smooth-muscle cell, endothelial cell, and fibroblast phenotype and function, vascular wall inflammation, and thrombosis, as well as ultrastructural and functional intercellular matrix changes.2 These biological vascular wall changes characterize many of the physiologic features seen in patients with PAH, including a decrease in pulmonary vascular capacitance and an increase in pulmonary vascular reactivity.
Of the many pathophysiologic features of PAH being investigated, the exact role of chronic thrombosis in the pulmonary arteries and microvasculature is one of the most controversial. Two views exist. First, it has been suggested that thrombotic arteriopathy is an epiphenomenon of the underlying hypertensive pulmonary vascular state and endothelial dysfunction of PAH. The alternate view is that the chronic organized thrombotic pulmonary vascular lesions are integral aspects of pulmonary vascular remodeling, luminal narrowing, and increased pulmonary vascular resistance, and contribute to the progression of PAH. This article reviews the pathologic and hematologic evidence evaluating the role of vascular thrombosis in the pathophysiology of IPAH and APAH due to CTD and CHD. We also review the clinical effectiveness of anticoagulation with warfarin in patients with PAH.
Pathologic Evidence
Pulmonary vascular thrombosis and thrombotic arteriopathy are common pathologic findings in PAH.34 Thrombotic lesions are most commonly recognized as nonlaminar, eccentric intimal fibrotic lesions, suggesting chronic organization of a previous thrombotic event. These are believed to be either the result of unrecognized pulmonary vascular emboli from distant sources or active in situ pulmonary vascular thrombosis. Occasionally, complete vascular obliteration by pulmonary vascular thrombosis may be seen, with evidence of organization as well as recanalization of the organized thrombus. When such thrombotic lesions are the predominant pathologic finding in the pulmonary vasculature, the term thrombotic arteriopathy may be applied.
In two retrospective cohort studies56 of IPAH patients evaluating histology, the prevalence rates of isolated thrombotic arteriopathy were 56% and 57%, respectively. Similarly, in a detailed study7 of pulmonary vascular histopathology from the National Institute of Health Primary Pulmonary Hypertension Registry of 48 patients with pulmonary arteriopathy (rather than veno-occlusive disease or capillary hemangiomatosis), classic thrombotic arteriopathy was present in 19 of 48 patients (40%) with IPAH, but thrombotic lesions were also present in another 9 patients. In this study, the presence of thrombotic arteriopathy did not characterize patients with regard to age, presentation, specific symptoms, functional class, or family history of IPAH. However, these authors7 found a significant relationship between the histologic classification and the severity of PAH. Patients with a predominant thrombotic arteriopathy had a significantly better median survival than other histologic forms of PAH.
Chronic pulmonary vascular thrombotic lesions have also been described in pathologic samples from patients with PAH associated with exogenous toxins (eg, aminorex) and in the setting of PAH associated with portal hypertension.8 As such, organized thrombotic lesions may not be specific to IPAH but rather a complication related to the severity and duration of PAH. For example, such chronic thrombotic lesions are unusual in children but are quite common in adults with IPAH.9 In contrast to the aforementioned study by Pietra et al,7 Wagenvoort and Mulder9 identified a relationship between the presence of thrombotic lesions and both age (p = 0.002) and duration of clinical illness in patients with PAH (p = 0.007). Taken together, these observational studies indicate a relatively high prevalence of thrombotic lesions in PAH. Furthermore, these studies suggest the presence of thrombotic lesions are related to age and disease duration, but are not specific for the etiology of PAH.
Abnormalities of Blood Coagulation and Fibrinolysis
Blood coagulation is a complex process characterized by the interaction of endothelial cells with both soluble elements (eg, plasma coagulation proteins) and cellular elements of blood (eg, platelets). In the healthy state, a balance exists between ongoing thrombosis and prevention of significant clot formation by both antithrombotic and fibrinolytic mechanisms. Among its vascular homeostatic roles, the endothelium is central in the regulation of this thrombotic-antithrombotic balance. The endothelium participates actively in the process of coagulation, activating factor X, facilitating formation of the thrombin-activating prothrombinase complex, and activating the extrinsic pathway of coagulation via release of tissue factor. In addition, endothelial cells produce and release von Willebrand factor (vWF), which functions as an adhesive protein in the interaction of platelets with the vessel wall, as well as a carrier for factor VIII.
The endothelium not only facilitates the thrombotic process but also actively inhibits thrombosis and promotes fibrinolysis. Endothelial cell production and release of nitric oxide and prostacyclin, two potent inhibitors of platelet aggregation, are important mechanisms in the prevention of thrombosis.10 As well, the expression of thrombomodulin, a high affinity receptor for thrombin, on the surface of endothelial cells prevents the cleavage of fibrinogen to fibrin. Endothelial cells are also a source of tissue plasminogen activator (t-PA), a key activator of plasminogen (factor XIII) in the fibrinolytic cascade. Of note, endothelial cells also synthesize and release plasminogen activator inhibitor-1 (PAI-1), an inhibitor of t-PA, highlighting the role of the endothelium in regulating the fine balance of prothrombotic and antithrombotic mediators and cascades.
The recognition of the critical role of the endothelium in this balance between prothrombotic and antithrombotic mechanisms suggested that endothelial dysfunction might contribute to the pathophysiology of PAH through abnormalities of the blood coagulation and fibrinolytic systems. Thus, active intravascular thrombosis may be present in PAH. Indeed, plasma levels of fibrinopeptide A, a byproduct and a marker of fibrin generation, were found to be elevated in all 31 IPAH patients in one study,11 and markedly so in 19 of 31 patients (61%). Moreover, an actual gradient of fibrinopeptide A was found across the lung in a patient with IPAH, suggestive of pulmonary vascular-specific fibrin formation rather than a generalized vascular prothrombotic state.12 Finally, elevated fibrinopeptide A levels and lung histologic evidence of thrombotic pulmonary arteriopathy were found in several members of a family with familial PAH.13
Abnormalities in the plasma levels of fibrinogen and fibrinogen metabolism have also been described in patients with IPAH. In one study,14 a decreased half-life of plasma fibrinogen was found in patients with IPAH. In contrast, another study15 reported that fibrinogen levels were higher (p < 0.01) in patients with IPAH (n = 25) and chronic thromboembolic pulmonary hypertension (CTEPH; n = 11) than in either control patients (n = 28) or patients with APAH due to CHD (n = 12). Furthermore, upper-extremity venous occlusion-stimulated fibrinolysis, assessed by the increase in t-PA activity, was blunted in patients with IPAH or CTEPH. Other abnormalities of the fibrinolytic system (eg, increased PAI-1 levels) have been reported in patients with IPAH compared to control subjects.11161718 In one of these studies,18 the increased PAI-1 levels in 12 patients with IPAH were associated with decreased plasma soluble thrombomodulin and a prolonged euglobulin lysis time, a global in vitro measure of fibrinolytic activity. Lower fibrinolytic activity correlated with higher mean PAP (r = 0.41, p = 0.003). Finally, 10% of patients with IPAH in one study16 had antibodies to fibrin-bound t-PA, suggesting another possible mechanism for an impaired fibrinolytic state.
vWF is a protein synthesized and stored in endothelial cells, megakaryocytes, and platelets that is essential in the interaction of platelets with endothelial cells. Abnormalities have been described in vWF levels and activity in patients with PAH.171920 IPAH patients have elevated in vitro vWF activity relative to immunologically measured vWF antigen levels, with milder increases seen in CHD patients with or without APAH.1921 Enhanced endothelial secretion of vWF can be stimulated by thrombin, fibrin, various cytokines, complement, and increased shear stress in the setting of PAH. vWF abnormalities in IPAH are likely a marker of endothelial injury or dysfunction rather than platelet defects, since in vitro assessment of plasma vWF activity is done with normal platelets from healthy blood donors.
Besides changes in activity, abnormalities in the composition of vWF have also been noted. vWF normally exists as a population of multimers of a basic subunit, with an apparent molecular weight of 1 to 20 x 106 d. Increased proteolytic degradation of the basic vWF subunit has been described in PAH, characterized by an increased proportion of faster moving bands, indicative of smaller vWF multimers (Table 1 ).192022 Indeed, multivariate regression analysis has identified both vWF activity levels and the proportion of low-molecular-weight vWF multimers as significant predictors of 1-year mortality in both IPAH and APAH patients.20 Furthermore, Veyradier et al23 not only corroborated the findings of increased proportion of low-molecular-weight multimers and vWF proteolytic fragments but also demonstrated partial correction of vWF proteolysis concomitant with hemodynamic improvements after 30 days of continuous IV epoprostenol therapy in 10 patients with severe PAH.
|
Deficiencies of the classical inhibitors of coagulation, (eg, antithrombin-3) and abnormal procoagulant factors (eg, factor V Leiden) are well-recognized risk factors for pulmonary thromboembolic disease. Overall, there is no evidence to suggest an increased tendency to PAH in these thrombophilic states, or an increased prevalence of these inherited disorders in patients with IPAH.24 Wolf et al25 found no difference in the frequency of antithrombin-3, protein C, protein S, factor V, and factor II mutations between PAH patients and control subjects. This finding is consistent with other published studies.2627
The presence of anti-phospholipid antibodies, directed against either membrane phospholipids or plasma proteins bound to anionic phospholipids (prothrombin, annexin V), is an important cause of thrombophilia. Anti-phospholipid antibody syndrome may be either primary (idiopathic) or more commonly seen in the setting of CTD such as systemic lupus erythematosus, and is associated with an increased tendency to both arterial and venous thrombosis. In particular, anti-prothrombin antibodies have been described in association with increased clotting and pulmonary hemorrhage.28 Although CTEPH has been well described in anti-phospholipid antibody syndrome, the association between anti-phospholipid antibodies and IPAH and APAH associated with CTD is uncommon.293031
Abnormalities of Platelet Function
Platelets not only participate in clot formation but are capable of releasing many vasoactive substances that promote vasoconstriction (eg, thromboxane-A2 [TxA2], serotonin) and thrombosis (eg, TxA2), as well as growth factors that stimulate proliferation of smooth-muscle cells, endothelial cells, and fibroblasts (eg, serotonin, platelet-derived growth factor).32 In addition, increased platelet aggregation is enhanced by the altered balance of vasoactive mediators in IPAH, such as increased TxA2 (proaggregatory) and decreased nitric oxide and prostacyclin (antiaggregatory).3334
Experimental models have implicated platelet abnormalities in the thrombotic tendency of PAH. For example, in the commonly used rat monocrotaline model of PAH, vascular thrombi are often found in the pulmonary vasculature,35 and the development of monocrotaline-induced PAH is attenuated by experimentally induced thrombocytopenia.36 There are only a few clinical studies of platelet function and activation in patients with IPAH. A case report12 described thrombocytosis in association with increased pulmonary vascular-specific fibrin generation and platelet activation in a patient with IPAH. Moreover, since TxA2 production is predominantly from platelets, the observed increase in the urinary metabolites of TxA2 in IPAH vs secondary PAH and control subjects is consistent with significant platelet activation in IPAH.33 A study of patients with moderate-to-severe APAH (mean PAP, 39 to 84 mm Hg) due to various etiologies demonstrated circulating platelet aggregates by scanning electron microscopy in 7 of 12 patients vs 1 of 6 control subjects.22 In addition, increased plasma ß-thromboglobulin levels in APAH patients vs control subjects (p < 0.025) indicated platelet activation.
One of the key mechanisms for platelet involvement in the pathophysiology of PAH may be the production and release of serotonin, a vasoactive substance with important effects on cell growth and proliferation.37 A family with a documented platelet serotonin storage disorder, resulting in high plasma serotonin levels, has been described in which one family member acquired IPAH > 20 years after the presence of the platelet defect.38 A similar inherited platelet defect associated with increased plasma serotonin levels in the Fawn-hooded rat is associated with a genetically determined, idiopathic form of PAH.3940 In IPAH patients, plasma serotonin was markedly elevated while the serotonin content of platelets, the major source of blood serotonin, was significantly reduced.41 Moreover, an in vitro platelet stimulation study41 demonstrated significantly greater serotonin release in response to epinephrine, adenosine diphosphate, and collagen by platelets from IPAH patients than from control patients. Most strikingly, in six IPAH patients studied before and 350 ± 30 days (mean ± SD) after heart-lung transplantation, the abnormal platelet and plasma serotonin concentrations persisted despite normalized pulmonary vascular hemodynamics after transplantation. This suggests a primary platelet serotonin-release defect rather than a secondary abnormality related to the abnormal pulmonary vascular hemodynamics in IPAH.
Supportive evidence for coagulation factor and platelet function abnormalities in IPAH also comes from studies looking at long-term response to chronic epoprostenol therapy. Although this agent is a potent pulmonary vasodilator, long-term benefit in severe IPAH has been shown even in patients without an acute vasodilator response to epoprostenol.4243 In one study44 cohort, plasma factor VIII levels and vWF antigen levels were abnormally high in 24 of 26 adult patients (92%) and 18 of 25 adult patients (72%) with IPAH, respectively. Moreover, vWF activity was abnormally high (> 120% normal) in 13 of 25 patients (52%), and ex vivo platelet aggregation findings demonstrated depressed responses in 87% of patients. One year of continuous IV epoprostenol therapy was associated with significant decreases in factor VIII levels, vWF antigen levels, and vWF activity. Platelet aggregation abnormalities had fully normalized in 83% of patients following long-term treatment with epoprostenol. Furthermore, hemodynamic improvement was significantly correlated with the improvement in platelet aggregation (p < 0.005) and the vWF activity/antigen ratio (p < 0.01).
Antibodies to Fibrin-Bound t-PA
In a study45 assessing the immunologic response to fibrin-bound t-PA, 9% of adults (4 of 45 patients) and 10% of children (4 of 41 patients) with IPAH had antibodies to fibrin-bound t-PA, compared to only 2.5% (1 of 40 children) with APAH due to CHD. In this small minority of patients with antibodies to fibrin-bound t-PA, there was a very high frequency (6 of 7 patients, 86%; odds ratio, 14.4; p = 0.05) of human leukocyte antigen-DQ7, compared to 29% in healthy control subjects. Furthermore, IPAH patients with antibodies to fibrin-bound t-PA commonly had a human leukocyte antigen amino acid epitope profile associated with the anti-phospholipid antibody syndrome.
Clinical Studies of Anticoagulation in IPAH
Several clinical studies support the hypothesis that ongoing pulmonary vascular thrombosis contributes to the pathogenesis and the progression of IPAH. The first study5 is a retrospective review of 120 patients followed up for an average of 14 years between 1955 and 1977 at the Mayo Clinic. In these patients, many with severe PAH (mean PAP, 64 mm Hg), 57% had evidence for chronic organized pulmonary vascular thromboses at autopsy. Overall survival was quite poor, with only 21% surviving 5 years; in a multiple linear regression analysis, one of the strongest, positive prognostic factors was the use of systemic anticoagulation therapy (p = 0.01).
In a recent retrospective study, Kawut et al46 evaluated predictors of outcome in a cohort of 84 consecutive adults with newly diagnosed, severe PAH (mean PAP, 55 mm Hg; range, 48 to 61 mm Hg), of which 79 patients (86%) were treated with warfarin. This cohort included 66 patients (78%) with IPAH, 14 patients (17%) with familial PAH, and 4 patients (5%) with anorexigen-associated PAH.46 Multivariate analyses of transplant-free survival indicated that warfarin use was a significant protective factor (hazard ratio, 0.35; 95% confidence interval, 0.12 to 0.99; p = 0.05).
The long-term effect of anticoagulation has also been looked at in a prospective, although nonrandomized, study47 in which systemic anticoagulation was selectively prescribed for 35 of 64 patients with IPAH in whom the perfusion lung scan revealed nonuniform pulmonary blood flow. Survival was better in those treated with anticoagulation than in those not receiving anticoagulants (p = 0.025). The improvement in survival was especially apparent in patients not receiving chronic calcium-channel blocker therapy over the 5-year follow-up period because of a lack of an acute calcium-channel blocker vasodilator response: 91%, 62%, and 47% survival at 1, 3, and 5 years, respectively, with anticoagulation, vs 52%, 31%, and 31% without anticoagulation.
Although these clinical studies of systemic anticoagulation in IPAH are methodologically limited, the apparent survival benefit has led to widespread recommendation and clinical use of anticoagulants in IPAH. In fact, the American College of Chest Physicians clinical practice guidelines,48 based on "fair" level of evidence and an "intermediate" level of benefit, recommend that patients with IPAH should receive anticoagulation with warfarin.
Discussion
Several lines of evidence from many studies1112131415161718192033343536384445 suggest that there are at least four important components in the conceptual model relating thrombotic arteriopathy and PAH: (1) abnormal hematologic parameters; (2) presence of a prothrombotic state; (3) thrombotic arteriopathy; and (4) PAH (Fig 1 ). In order for one to state that thrombotic arteriopathy is a cause of PAH, a number of scientific criteria must be fulfilled. The evidence must support biological plausibility, temporality, strength of association, and analogy.49 The conceptual model in Figure 1 is a biologically plausible causal pathway relating the observed abnormal hematologic parameters, development of thrombotic arteriopathy, and culminating in PAH. However, the current evidence does not fully support temporality, strength of association, and analogy. Some evidence suggests that abnormalities of blood coagulation factors, antithrombotic factors, and the fibrinolytic system contribute to a prothrombotic state in patients with IPAH. However, other lines of evidence do not support this relationship. For example, given its allelic distribution is similar to reference control populations, factor V Leiden does not appear to contribute to the pathophysiology of IPAH in most patients. In addition, PAH is quite uncommon in the setting of inherited thrombophilias, like anti-phospholipid antibody syndrome. Furthermore, it remains uncertain whether the hematologic abnormalities observed in IPAH analogously contribute to a prothrombotic state in APAH, and requires further study.
|
Irrespective of whether thrombotic arteriopathy is a cause or consequence of PAH, thrombotic arteriopathy may alter the progression and prognosis of IPAH and APAH patients. Interruption of ongoing thrombosis with effective, systemic anticoagulation therapy appears to predict a better prognosis, especially for patients with disease not responsive to vasodilators. The American College of Chest Physicians clinical guidelines48 support the use of anticoagulation with a grade "B" recommendation based on observational studies that indicate a survival benefit in anticoagulated IPAH patients. However, due to the methodologic limitations affecting these studies, the magnitude of this treatment effect is uncertain. Furthermore, the generalizability of these recommendations to patients with APAH (eg, scleroderma associated PAH) remains controversial given the potentially increased risk of GI hemorrhage from luminal telangectasia.50 To definitely evaluate the efficacy of anticoagulation with warfarin on survival and ascertain if the clinical benefit outweighs the potential risks, a randomized, controlled trial is needed. Both in the United States (D.B. Badesch, MD; personal communication; September 2005) and Canada (J.T. Granton, MD; personal communication; February 2004), proposals for randomized trials to evaluate the efficacy of warfarin in IPAH and scleroderma-associated PAH patients have been developed. Together, the results of these trials will clarify this issue.
Conclusions
The available evidence suggests that thrombotic arteriopathy is an important pathophysiologic feature of PAH. Further research on the abnormalities of endothelial cell and platelet function, abnormalities of coagulation factors, antithrombotic and fibrinolytic factors, and the genetics of IPAH and APAH will undoubtedly shed more light on the role of thrombotic arteriopathy in the pathophysiology of IPAH and APAH. The possibility of early identification of patients with preclinical PAH holds promise for early intervention to correct the described abnormalities of both cells and soluble mediators in order to attenuate the severity of PAH or prevent the clinical expression of disease.
Footnotes
Abbreviations: APAH = pulmonary arterial hypertension associated with an underlying condition; CHD = congenital heart disease; CTD = connective tissue disease; CTEPH = chronic thromboembolic pulmonary hypertension; IPAH = idiopathic pulmonary arterial hypertension; PAH = pulmonary arterial hypertension; PAI-1 = plasminogen activation inhibitor-1; PAP = pulmonary artery pressure; t-PA = tissue plasminogen activator; TxA2 = thromboxane-A2; vWF = von Willebrand factor
Dr. Johnson is the recipient of a research fellowship award from the Canadian Institutes for Health Research, Institute for Musculoskeletal Health and Arthritis, and The Arthritis Society.
Received for publication May 5, 2005. Accepted for publication January 24, 2006.
References
This article has been cited by other articles:
![]() |
E. J. van Beers, B. L. F. van Eck-Smit, M. R. Mac Gillavry, C. F. J. van Tuijn, J. W. J. van Esser, D. P. M. Brandjes, M. C. Kappers-Klunne, A. J. Duits, B. J. Biemond, J.-J. B. Schnog, et al. Large and Medium-Sized Pulmonary Artery Obstruction Does Not Play a Role of Primary Importance in the Etiology of Sickle-Cell Disease-Associated Pulmonary Hypertension Chest, March 1, 2008; 133(3): 646 - 652. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. R. Johnson, S. Mehta, and J. T. Granton Anticoagulation in pulmonary arterial hypertension: a qualitative systematic review. Eur. Respir. J., November 1, 2006; 28(5): 999 - 1004. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |