Pulmonary Hypertension Associated with Myeloproliferative Disorders: A Retrospective Study of Ten CasesGuilpain P.a · Montani D.c · Damaj G.a · Achouh L.c · Lefrère F.a · Le Pavec J.c · Marfaing-Koka A.d · Dartevelle P.e · Simonneau G.c · Humbert M.c · Hermine O.a, b
aDepartment of Clinical Hematology and bCentre National de Recherche Scientifique UMR 8147, Hôpital Necker-Enfants-Malades, Assistance Publique – Hôpitaux de Paris, Université Paris-V, Paris, and cVascular Pulmonary Diseases Center, Department of Pulmonary Diseases, and dDepartment of Hematology, Hôpital Antoine-Béclère, Assistance Publique – Hôpitaux de Paris, Université Paris-Sud, Clamart, and eUPRES EA2705, Department of Thoracic Surgery, Centre Chirurgical Marie-Lannelongue, Université Paris-Sud, Le Plessis-Robinson, France Corresponding Author
Background: Pulmonary hypertension (PH) is a severe hemodynamic disorder in which the pulmonary artery pressure is persistently elevated, leading to right-sided heart failure. Some studies have suggested an association between PH and myeloproliferative diseases (MPD). Objectives: This study describes clinical, hematological and hemodynamic characteristics of PH associated with MPD. Methods: We retrospectively reviewed 10 cases of PH associated with MPD: polycythemia vera (8 patients) and essential thrombocythemia (2 patients), followed between 1993 and 2002. The baseline evaluation was established by right-sided heart catheterization, ventilation/perfusion lung scan and pulmonary angiography if required. Results: Six patients had confirmed chronic thromboembolic pulmonary hypertension (CTEPH) and 4 had pulmonary arterial hypertension (PAH) associated with MPD without other risk factors for PAH. The hemodynamic characteristics of CTEPH and PAH associated with MPD were similar. The diagnosis of CTEPH was concomitant to that of MPD in all cases (5 polycythemia vera and 1 essential thrombocythemia). The PAH associated with MPD occurred later in the evolution of the MPD (3 polycythemia vera and 1 essential thrombocythemia) with a median of 162 months after the diagnosis of MPD, and it was associated with myeloid metaplasia (p < 0.01). Conclusion: We describe 2 distinct forms of PH in the context of MPD: CTEPH, which is diagnosed at an early stage of the MPD, and PAH, which occurs later in the course of the MPD and is associated with myeloid metaplasia. Progressively increasing dyspnea in a patient with an MPD warrants further investigation to rule out PAH and CTEPH, while a diagnosis of CTEPH warrants ruling out MPD.
Copyright © 2007 S. Karger AG, Basel
Pulmonary hypertension (PH) is a rare and severe condition, characterized by elevated pulmonary artery pressure leading to right-sided heart failure and death . PH is defined as a mean pulmonary artery pressure (mPAP) ≥25 mm Hg at rest or ≥30 mm Hg during exercise, measured during right-sided heart catheterization [1, 2]. The current clinical classification of PH, proposed in 2003, comprises apparently heterogeneous conditions which share comparable clinical presentations, pathophysiology and management . Pulmonary arterial hypertension (PAH) includes idiopathic PAH, familial PAH  and PAH associated with various conditions, including several hematological diseases , which have been associated with PH, such as hemoglobinopathies (sickle cell disease, thalassemia), red blood cell membrane abnormalities (spherocytosis), lymphomas and myeloproliferative diseases (MPD). The association between PH and MPD has been suggested by a few case reports and some small series [5,6,7,8,9,10,11]. Dingli et al. [9 ] described a possible association of PH and MPD, principally with myelofibrosis and myeloid metaplasia. MPD defines a group of acquired clonal disorders of hematopoietic stem cells, including polycythemia vera, essential thrombocythemia and myelofibrosis with myeloid metaplasia. Pulmonary complications in MPD are principally represented by infections and thromboembolic or hemorrhagic events. The clinical, hematological and hemodynamic characteristics and mechanisms leading to the development of PH in patients with MPD remain, however, largely unknown. Evolution and prognosis of polycythemia vera and essential thrombocythemia are relatively favorable but development of PH has been suggested to be a factor of worse prognosis , stressing the importance of characterizing PH associated with MPD. In our study, we analyzed the characteristics of 10 patients with PH associated with MPD, excluding PH associated with left-sided heart diseases. To our knowledge, this is the largest series of patients with a complete evaluation of PH including right-sided heart catheterization, and systematic exploration to confirm chronic thromboembolic pulmonary hypertension (CTEPH). We identified and described 2 distinct forms of PH associated with MPD: CTEPH and PAH.
Material and Methods
We retrospectively reviewed the charts of PH patients followed at the Center for Pulmonary Vascular Diseases, South Paris University, between 1993 and 2002 and selected those associated with MPD, such as polycythemia vera, essential thrombocythemia and myelofibrosis with myeloid metaplasia. Patients with chronic myelogenous leukemia or myelodysplastic syndromes were excluded from the study. Also excluded were those who had any of the following conditions: chronic hypoxic lung disease, left-sided heart failure, systemic-to-pulmonary congenital heart shunt, connective tissue disorders, HIV infection and appetite suppressant exposure. All patients with PAH seen in our center are included in a registry with all relevant clinical and biological information. This registry was set up in agreement with the Commission Nationale de l’Informatique et des Libertés, the organization dedicated to information technology and civil rights in France. All patients gave informed consent to be entered in the registry.
The diagnosis and classification of MPD were based on clinical and biological findings, the criteria of the Polycythemia Vera Study Group and/or the presence of endogenous erythroid colony formation . The decisive criteria for the diagnosis of polycythemia vera were: a hematocrit greater than 54% in males and 47% in females, increased red blood cell mass (≥36 ml/kg for males and 32 ml/kg for females) in the absence of hypoxia (oxygen saturation ≥92%), tumor or hepatic mass, and the presence of splenomegaly. Diagnosis of essential thrombocythemia was made based on evidence of an MPD with a platelet count of more than 500 × 109/l, normal leukocyte differential and red blood cell morphology, a normal ferritin level and the absence of an inflammatory syndrome. In some cases, karyotype examinations were performed to exclude chronic myelogenous leukemia and myelodysplastic syndromes. We used the test of erythropoietin-independent erythroid colony (EEC) formation as a criterion for the diagnosis in the absence of increased spleen volume , or to rule out a secondary polycythemia as a consequence of hypoxia. Signs of myelophthisis of the blood, including tear-drop-shaped red blood cells, were considered as indirect signs of myelofibrosis. Cases were classified as histologically confirmed myeloid metaplasia or probable myeloid metaplasia, in the presence of myelophthisis without a proven histologic diagnosis.
Hemodynamic evaluation was performed at baseline in all subjects according to our previously described protocol . Cardiac output was measured by the standard thermodilution technique, and the cardiac index (CI) was calculated as the cardiac output divided by the body surface area. The mean pulmonary arterial pressure, pulmonary capillary wedge pressure (PCWP), right-sided atrial pressure and mixed venous oxygen saturation were recorded during right-sided heart catheterization. The indexed total pulmonary resistance and indexed vascular pulmonary arterial resistance were calculated respectively as (mPAP × 80)/CI and (mPAP – PCWP) × 80/CI. PAH was defined by an mPAP greater than 25 mm Hg with a normal PCWP (<15 mm Hg) upon right-sided heart catheterization. Patients with elevated PCWP >15 mm Hg were excluded from the study. All patients had a ventilation/perfusion lung scan to screen for CTEPH. If segmental perfusion defects were demonstrated, a pulmonary angiography was performed, and the possibility of thromboendarterectomy was discussed. CTEPH was diagnosed by typical abnormalities on the ventilation/perfusion lung scan, confirmed by pulmonary angiography. PAH associated with MPD was defined as PAH occurring during MPD, without other risk factors or conditions associated with PAH.
Statistical analysis was performed using Statview version 5.0 (Abacus Concepts Inc., Berkley, Calif., USA). Data are presented as median (min.–max.), except if stated otherwise. Comparisons between subjects with CTEPH and PAH associated with MPD were assessed by the Mann-Whitney U test and a χ2 test. A p value of ≤0.05 was considered statistically significant.
Ten patients were included in this study, 8 patients with polycythemia vera and 2 with essential thrombocythemia. No patient had primitive myelofibrosis with myeloid metaplasia at the time of diagnosis. The median age at diagnosis of the MPD group was 54 years (range: 31–72). Individual characteristics and evolution are summarized in table 1.
|Table 1. General characteristics of patients|
Polycythemia vera was diagnosed in 8 patients, including 3 with peripheral signs of myeloid metaplasia. The median age at the diagnosis of polycythemia vera was 55.5 years (range: 31–72). At the diagnosis of PH, 3 patients with polycythemia vera had already been treated with cytoreductive agents and had a low hematocrit. The median hematocrit value was 51.2% at the diagnosis of PH. Four patients (50%) with polycythemia vera had associated thrombocytosis. The median white blood cell count was 9,750/mm3 (range: 4,300–17,000), and the median platelet count was 502,000/mm3 (range: 183,000–980,000). One patient with early-stage polycythemia vera (patient 3) did not receive cytoreductive therapy because further blood cell counts and hematocrits were normal. All other patients were treated with hydroxyurea, and 1 patient also received pipobroman (patient 1).
Essential thrombocythemia was diagnosed in 2 patients including 1 with peripheral signs of myeloid metaplasia. At the diagnosis of PH, the median platelet count was 644,500/mm3 (range: 624,000–665,000), and the median hematocrit was 35.5% (range: 26.8–44.1). The 2 patients received hydroxyurea during the course of the disease. Because of uncontrolled thrombocytosis, 1 patient with essential thrombocythemia (patient 8) also received CCNU, pipobroman, melphalan, anagrelide and subsequently underwent cytapheresis.
All patients underwent complete hemodynamic measurements during right-sided heart catheterization, with moderate to severe PH in all cases. Two distinct forms of PH associated with MPD were observed, CTEPH and PAH. Clinical, biological and hemodynamic characteristics of patients according to the type of PH are presented in table 2. Interestingly, 1 patient had a high cardiac output in the setting of massive splenomegaly, as well as confirmed elevated pulmonary vascular resistance. The remaining 9 patients had normal or reduced cardiac output. All patients received anticoagulation, diuretics and supplemental oxygen if needed. Six patients (4 PAH associated with MPD and 2 with CTEPH) died during the first year, and 4 of these patients died from end-stage right-sided heart failure. All patients with PAH associated with MPD died within the first year after the diagnosis of PH. Three patients experienced severe hemorrhagic complications, 1 with internal bleeding and 2 with hemoptysis.
|Table 2. Characteristics of patients with PH associated with MPD|
Chronic Thromboembolic PH
Ventilation/perfusion lung scans demonstrated segmental perfusion defects in 6 out of 10 patients (60%). Subsequent pulmonary angiography confirmed CTEPH in all these cases. CTEPH was proximal in 4 patients and distal in 2 patients. Amongst these 6 patients with CTEPH, 5 presented with polycythemia vera (83.3%) and 1 with essential thrombocythemia (16.7%). The diagnosis of CTEPH was concomitant to that of MPD in all cases, and PH revealed the underlying MPD. Thus, patients with CTEPH had mainly untreated polycythemia vera and a significantly higher hematocrit than patients with PAH associated with MPD (median: 50.7 and 30.5%, respectively; p < 0.001), but they had similar white blood cell and platelet counts. Erythroid progenitor cells were cultured from 5 patients with CTEPH, and demonstrated in all of these cases spontaneous EEC formation. Four patients with polycythemia vera had previous thrombotic events including pulmonary embolism in 3 patients and acute arterial ischemia of the left lower limb in 1 patient. The patient with essential thrombocythemia had evidence of distal thrombi that could be consistent with silent emboli. CTEPH was proximal in 4 cases, resulting in successful thromboendarterectomy in 3 patients (patient 4 died a few days after surgery). In addition, 1 patient with CTEPH received iloprost, and 2 patients received bosentan. The 2 patients with distal CTEPH did not undergo thromboendarterectomy and were treated with anticoagulation, diuretics, oxygen supplementation and bosentan.
Pulmonary Arterial Hypertension Associated with MPD
Four patients (40%) presented PAH associated with MPD, 3 patients had polycythemia vera (75%) and 1 had essential thrombocythemia (25%). In contrast to CTEPH, PAH occurred later in the course of the disease (p < 0.01) with a median of 162 months (range: 120–240) after the diagnosis of MPD. Interestingly, PAH associated with MPD occurred only in females, in contrast to CTEPH (3 females and 3 males). No difference was observed in hemodynamic characteristics between CTEPH and PAH. One PAH patient received a continuous epoprostenol infusion, the other 3 PAH patients received no specific PAH therapy. PAH was more often associated with myeloid metaplasia than CTEPH. All patients with PAH had confirmed or probable myeloid metaplasia, whereas none of the patients with CTEPH had myeloid metaplasia (p < 0.01). In 3 patients, PAH was diagnosed concomitantly to the appearance of myeloid metaplasia signs. Two patients had previous thrombotic events (a deep vein thrombosis and a thrombosis of a renal artery) diagnosed by a normal lung perfusion scan. Necropsy was available for 1 patient with PAH associated with MPD and showed lung infiltration by hematopoietic cells consistent with pulmonary myeloid metaplasia.
Previous reports have suggested a possible association between PH and MPD [6,7,8,9,10]. However, the exact prevalence of PH in this population remains unknown. It is probably underestimated because the clinical signs of PH appear at an advanced stage of the disease, and, in some cases, the diagnosis of MPD is particularly difficult in the context of hypoxia. Dingli et al. [9 ] reported 26 cases of PH occurring in the evolution of chronic myeloid malignancies, including MPD, myelodysplastic syndrome and chronic myeloid leukemia. The authors showed that PH was a factor of worse prognosis with a median survival time of 18 months. However, in this study, only 5 patients underwent right-sided heart catheterization, and no systematic exploration of CTEPH was performed. Furthermore, in the 5 patients who underwent right-sided heart catheterization, 4 (80%) had elevated PCWP suggesting that left-sided heart disease could be the major cause of PH in patients with MPD. In a recent study of 26 patients with MPD and chronic myeloid leukemia, Garypidou et al. [14 ] found more than 40% of PH was diagnosed by echocardiography. These results are higher than previously reported , and it might be hypothesized that the rate of patients with PH was overestimated, as elevated pulmonary arterial pressure could be secondary to left-sided heart disease and no hemodynamic confirmation of PH was obtained. To describe and confirm a possible association of PH with MPD, patients with elevated PCWP were excluded from our study. In our series of 10 patients with PH associated with MPD, we described 2 distinct forms of PH with similar hemodynamic characteristics. CTEPH was diagnosed concomitantly with MPD at an early stage of the disease, while PAH was diagnosed later in the evolution of MPD and appeared to be mainly associated with myeloid metaplasia, secondary to either polycythemia vera or essential thrombocythemia.
Thromboembolic events are associated with a higher morbidity and mortality in patients with MPD. These include cerebrovascular stroke, myocardial infarction, peripheral vascular occlusion, pulmonary infarction and venous thrombosis. In essential thrombocythemia, the incidence of thrombosis is approximately 15% . In polycythemia vera, about 20% of patients present with thrombotic events [15,] and the risk of thrombosis is substantially greater in patients who have previously developed thrombosis and in older patients . In the study of the Gruppo Italiano Studio Policitemia , 41% of patients with polycythemia vera had at least 1 thrombotic event, which occurred either at presentation or before diagnosis in 64% of cases. Thrombotic episodes were seen more frequently in the 2 years preceding the diagnosis. In polycythemia vera, the degree of thrombocytosis and the presence of platelet dysfunction have not been correlated with the risk of thrombosis, and in essential thrombocythemia, the correlation with the platelet count also appears to be unclear [17,18,19,20]. However, local activated platelets may play a crucial role in the development of CTEPH. Rostagno et al. [21 ] reported a patient with CTEPH and long-standing thrombocytosis secondary to a splenectomy performed in infancy for minor thalassemia. In this case, the demonstration of the existence of a transpulmonary gradient for thromboxane A2, β-thromboglobulin and fibrinopeptide A are very suggestive of the role of local pulmonary platelet activation and thrombin generation in the pathogenesis of CTEPH . In the present study, the platelet count was not different between patients with CTEPH and PAH associated with MPD, but an elevated hematocrit was significantly associated with CTEPH compared to PAH associated with MPD. This result could be explained by the absence of treatment for MPD at the diagnosis of CTEPH. So, PAH associated with MPD was diagnosed in a late stage of MPD, which could explain a lower hematocrit secondary to specific hematologic therapy. However, the elevated hematocrit could be a causal factor of in situ arterial pulmonary thrombosis leading to the development of CTEPH. Furthermore, it is well established that spontaneous EEC formation may be associated with thrombosis, particularly of hepatic veins, in the absence of other peripheral blood abnormalities . Thus, the presence of EEC leads to the diagnosis of primary MPD in 78% of cases with apparent idiopathic Budd-Chiari syndrome, and in about half of patients with portal, splenic and/or mesenteric vein thrombosis . Spontaneous EEC formation has never been demonstrated in patients with CTEPH.In the present study, the diagnosis of CTEPH and MPD was made simultaneously in all patients, suggesting that CTEPH could be the first manifestation of MPD. Therefore, EEC formation tests might be included in the evaluation of CTEPH patients. Although the majority of patients with CTEPH had a previous history of thrombotic events, CTEPH in MPD may also occur as a consequence of silent emboli . A previous thrombotic event does not exclude the diagnosis of PAH associated with MPD since 2 of our patients had a history of thrombosis, without evidence of CTEPH on lung perfusion scan or angiography. It has been established that splenectomy is a risk factor for PH. Histopathologic examinations of lung specimens from patients with surgical asplenia show intimal fibrosis, plexiform lesions and thrombotic lesions . Moreover a splenectomy is a situation that increases the platelet count and may promote hematopoietic pulmonary infiltration in patients with myeloid metaplasia. However, none of our patients had undergone splenectomy.
Management of CTEPH should include pulmonary thromboendarterectomy in patients with proximal pulmonary artery thrombi . In the present study, such surgery was feasible and seemed safe in 3 patients. In other cases, medical therapy including diuretics, anticoagulants, antiplatelet agents, cytoreductive therapy and specific PAH therapy, such as bosentan [26, 27], should be discussed. In essential thrombocythemia and polycythemia vera, some studies have shown that hydroxyurea reduces the risk of thrombosis in high-risk patients [28, 29]. However, vitamin K antagonists and antiplatelet agents should be used with caution because of the potential risk of hemorrhage in patients with polycythemia vera and essential thrombocythemia [15, 17]. Recent studies have demonstrated that patients with CTEPH who cannot undergo surgery may benefit from specific PAH therapies, such as epoprostenol or the dual endothelin receptor antagonist, bosentan [4, 26,30,31,32]. No data on specific PAH therapy are available to date for patients with CTEPH associated with MPD.
In contrast with CTEPH, PAH associated with MPD was diagnosed a median of 13 years after the recognition of MPD. In PAH, several causal factors may be implicated. First, portal hypertension is a well-described condition associated with PAH [2, 4] and is a well-known complication of myeloid metaplasia [33, 34]. However, in this study, only 1 patient had portal hypertension with esophageal varices. Second, the use of chemotherapy in these conditions may result in pulmonary damage or veno-occlusive disease. However, none of our patients had clinical or radiological evidence of pulmonary veno-occlusive disease . Third, pulmonary myeloid infiltration during the chronic phase of agnogenic myeloid metaplasia and leukemic infiltration during the acute transformation of the disease have been described [10, 36]. Marvin and Spellberg [6 ] reported a patient with myeloid metaplasia, thrombocytosis and circulating micromegakaryocytes. Microcirculatory obstruction was hypothesized by the authors as the mechanism of PH. Alveolar capillary plugging by malignant megakaryocytes has also been reported . Moreover, activated platelets, by releasing cytokines (like platelet-derived growth factor and transforming growth factor β) or other mediators, including serotonin, might be implicated in the development of PH . Necropsy was available for 1 patient showing evidence of pulmonary metaplasia. Three patients with a history of MPD were diagnosed as having PAH concomitant to the appearance of signs of myeloid metaplasia. This may not be a coincidental event, and we hypothesize that PAH associated with MPD may result from either cytokine release responsible for the development of myeloid metaplasia, and/or plugging of the alveolar capillary by hematopoietic cells. Dingli et al. [9 ] emphasized the major role of blood cell proliferation in the pathogenesis of PAH associated with MPD. They reported a correlation between right-sided ventricular systolic pressure and platelet count, in patients with myeloid metaplasia and essential thrombocythemia, and with hemoglobin levels in patients with polycythemia vera. Furthermore, Marvin and Spellberg  observed reversibility of PH by decreasing the platelet count. This indicates that patients may benefit from the administration of cytoreductive treatment (or from therapeutic phlebotomies). However, the effectiveness of antiplatelet agents for PH has not been established and one third of patients may develop PH while receiving aspirin . This finding might be explained by the inefficacy of aspirin on platelet cytokine release. It is noteworthy that none of our patients received drugs with potential pulmonary toxicity except for 1 patient who consecutively received anagrelide and CCNU for uncontrolled thrombocytosis. On a whole, it is difficult to establish the benefits of hematological therapy in PH associated with MPD. The prognosis of this clinical condition remains poor since 6 out of 10 patients died. In the study of Dingli et al. , the median survival time was 18 months and death was mainly related to cardiopulmonary failure. Although their effect is not demonstrated in this indication, PAH specific therapies like bosentan, epoprostenol and nebulized Ilomedine might be useful in those patients. However, even if PAH and CTEPH are considered as 2 distinct forms of PH , these 2 diseases share some similarities. Many patients with CTEPH develop a severe progressive small-vessel pulmonary arteriopathy. Furthermore, recent studies have shown that specific PAH therapies provide an alternative medical therapy to improve function and delay the progression of CTEPH in patients in whom surgery is not suitable [26, 27].
In conclusion, PH associated with MPD is a rare event; however, its true frequency may be underestimated. In the present study we describe 2 distinct forms of PH in the context of MPD, CTEPH and PAH associated with MPD. CTEPH is diagnosed at an early stage of MPD (especially polycythemia vera, which makes an elevated hematocrit suggestive of the diagnosis). In contrast, PAH associated with MPD occurs later in the course of the disease and seems to be associated with myeloid metaplasia, suggesting that pulmonary myeloid infiltration could participate in the development of PAH in these patients. CTEPH may unmask a PH associated with MPD. The treatment of PH associated with MPD is not yet established and depends on the diagnosis of PH. Pulmonary thromboendarterectomy seems to be safe and might be proposed in patients with proximal CTEPH, although anticoagulant drugs should be administered carefully because of the potential risk of hemorrhage. Cytoreductive treatment should be used in association with symptomatic treatment of PH. In PAH associated with MPD, the benefit of specific PAH therapies need to be confirmed.
Centre des Maladies Vasculaires Pulmonaires, Service de Pneumologie
Hôpital Antoine-Béclère, Assistance Publique – Hôpitaux de Paris
Université Paris-Sud, 157, rue de la Porte de Trivaux, FR–92140 Clamart (France)
Tel. +33 1 45 37 47 79, Fax +33 1 46 30 38 24, E-Mail email@example.com
P.G. and D.M. contributed equally to the paper.
Received: July 12, 2007
Accepted after revision: October 16, 2007
Published online: December 21, 2007
Number of Print Pages : 8
Number of Figures : 0, Number of Tables : 2, Number of References : 37
Respiration (International Journal of Thoracic Medicine)
Vol. 76, No. 3, Year 2008 (Cover Date: September 2008)
Journal Editor: Bolliger C.T. (Cape Town)
ISSN: 0025–7931 (Print), eISSN: 1423–0356 (Online)
For additional information: http://www.karger.com/RES