Incidence and clinical risk factors for bleeding and thrombotic complications in myeloproliferative disorders. A retrospective analysis of 260 patients

Myeloproliferative syndromes and the associated risk of coronary artery disease

INTRODUCTION

Of the major myeloproliferative syndromes (MPS) [polycythemia vera (PV), essential thrombocythemia (ET), chronic myeloid leukemia (CML) and myelofibrosis (MF)], PV and ET are reported to be associated with increased thrombotic complications. However, the relationship between these myeloproliferative syndromes and coronary artery disease (CAD) is unclear.

METHODS

We performed a retrospective chart review to evaluate the prevalence of CAD in patients with diagnosed with MPS between 1991 and 2001.

RESULTS

One hundred and eighty-one patients (100 males, 81 females) with a mean age of 72.5 years were included. Twenty-nine patients, 19 males and 10 females (16%, 95% CI: 12.0-24.0) had CAD. These included 6/53 (11.3%, 95% CI: 1.5-20.2) patients with CML, 1/26 (3.8%, 95% CI: -4.4 to 12.8) patients with PV, 5/30 (16.7%, 95% CI: 2.5-30.8) patients with ET, 3/7 (42.9%, 95% CI: 12.3-87.7) patients with MF and 14/65 (21.5%, 95% CI: 13.1-37.8) patients with co-existent MPS. Comparing the risk of CAD with CML as a baseline, MF had an OR of 8.2 (p < 0.01, 95% CI: 1.7-39), PV-0.4 (p = 0.4, 95% CI: 0.04-3.2), ET-1.6 (p = 0.7, 95% CI: 0.43-6.2) and patients with co-existent MPS-2.8 (p=0.07, 95% CI: 0.91-8.6). However, after adjusting for age, sex, dyslipidemia, diabetes, hypertension and tobacco use, the difference in the prevalence of CAD between the various categories of MPS was not significant.

CONCLUSION

Contrary to conventional belief, we did not find an increased prevalence of CAD in patients with either PV or ET. In fact, patients with MF had a significantly higher prevalence of CAD. However, this difference appears to be due to the increased age at diagnosis of MF. The conventional risk factors for CAD appear to be the major determinants of CAD among patients with MPS.

Avascular necrosis of the femoral head in chronic myeloid leukemia patients treated with interferon-alpha: a synergistic correlation?

BACKGROUND

The objectives of this study were to describe cases of avascular necrosis of the femoral head (ANFH) observed in chronic myeloid leukemia (CML) patients who were treated with interferon-alpha and to review the literature.

METHODS

The authors undertook a case review of the M. D. Anderson experience with ANFH occurring in CML patients who were managed with interferon-alpha-based therapy. MEDLINE (from 1966 to November 1999) and CancerLit (from 1983 to November 1999) searches were conducted to identify cases of avascular necrosis (AVN) associated with either CML or interferon-alpha.

RESULTS

Three patients with ANFH were identified from the authors' experience. No common features related to the disease or therapy were seen among them, except for the presence of thrombocytosis and loss of response. A literature review revealed seven cases of ANFH associated with CML with or without interferon-alpha-based therapy. ANFH was not reported in association with interferon-alpha use for indications other than the treatment of patients with CML.

CONCLUSIONS

ANFH may be the result of an interaction between CML and interferon-alpha therapy. ANFH that occurs in patients with CML who are treated with interferon-alpha should be recognized for treatment implications. Thrombocytosis with consequent microvascular thrombi and avascular necrosis manifesting in susceptible vascular or weight-bearing areas (e.g., the femoral head) may be an associated finding along with loss of response to interferon-alpha therapy.

Clinicopathological diagnosis and differential criteria of thrombocythemias in various myeloproliferative disorders by histopathology, histochemistry and immunostaining from bone marrow biopsies

Differentiation of essential thrombocythemia (ET) from thrombocythemias occurring in various subtypes of chronic myeloproliferative disorders (MPDs) is controversial, because of the lack of uniform clinical and morphological criteria. A retrospective clinicopathologic study was performed on 375 patients presenting with a MPD and a platelet count exceeding 500 x 10(9/)l. For comparison 35 patients with reactive thrombocytosis (RT) and five patients with a myelodysplastic syndrome (MDS-5q(-) syndrome) were enrolled into this study. In addition to a complete clinicopathological work-up, procedures included histochemical and immunological staining techniques and morphometry of bone marrow biopsies for proper evaluation of megakaryocytes (CD61) and erythroid precursors (Ret40f). Because of the high patient's age on admission, relative survival rates with corresponding disease-specific loss of life expectancy were calculated. Analysis of clinical and morphological characteristics, in particular megakaryopoiesis revealed features which enabled a clear-cut distinction between thrombocythemias in MPDs and thrombocythemic states in MDS. This rationale proved to be most important for the diagnostic discrimination of the 33 patients with initial (prefibrotic) stages of idiopathic myelofibrosis (IMF) from ET (40 patients). A new set of relevant criteria for the diagnosis of IMF with special regard to early stages and its distinction from ET has been proposed. Hemorrhagic episodes were more frequently observed in ET than in thrombocythemias associated with polycythemia vera (PV). Computation of specific loss of life expectancy revealed two extremes: thrombocythemia in CML (81%) and ET (3%), whereas thrombocythemias in PV and IMF did not show a significantly different life loss (19-22%). The revised criteria for ET, PV and IMF are reliable by taking histopathological features from bone marrow biopsies into consideration, particularly for the diagnosis of ET and its differentiation from thrombocythemias as a presenting symptom accompanying the various subtypes of MPDs.

Role of red blood cells in thrombosis

Most biomedical textbooks teach that coagulation and thrombosis are primarily a function of endothelial cells, platelets, and soluble coagulation factors. Red blood cells, in contrast, are generally regarded as innocent bystanders, passively entrapped in a developing thrombus as they flow through the vasculature. This review summarizes evidence that demonstrates an active role for red cells in normal and pathologic hemostasis. We then evaluate the possible molecular mechanisms whereby a usually inert erythrocyte can actively contribute to the processes of clot formation.

Treatment of polycythaemia vera and essential thrombocythaemia

The clinical course in both polycythaemia vera (PV) and essential thrombocythaemia (ET) is characterized by significant thrombohaemorrhagic complications and variable risk of disease transformation into myeloid metaplasia with myelofibrosis or acute myeloid leukaemia. Randomized studies have shown that the risk of thrombosis was significantly reduced in ET with the use of hydroxyurea (HU) and in PV with the use of chlorambucil or 32P. However, the use of chlorambucil or 32P has been associated with an increased risk of leukaemic transformation. Subsequently, other studies have suggested that both HU and pipobroman may be less leukaemogenic and as effective as chlorambucil and 32P for preventing thrombosis in PV. However, the results from these prospective studies have raised concern that even HU and pipobroman may be associated with excess leukaemic events in both ET and PV. The recent introduction of anagrelide as a specific platelet-lowering agent, the demonstration of treatment efficacy with interferon-alpha, and the revived interest in using low-dose acetylsalicylic acid provide the opportunity to initiate prospective randomized studies incorporating these treatments.

Reduced in vitro clot lysis and release of more active platelet PAI-1 in polycythemia vera and essential thrombocythemia

Because platelets interact with fibrinolysis in a complex manner, it can be expected that with abnormal platelet numbers and quality this interference can be even more profound. The aim of this work was to study the lysis-resistance of platelet-rich clots in diseases with high platelet counts: polycythemia vera (PV), essential thrombocythemia (ET) and to make comparison with polyglobulia (PG). Platelet-rich plasma (PRP) and platelet-poor plasma (PPP) were analyzed by an in vitro clot lysis test. Plasminogen activator inhibitor-1 (PAI-1) activity was measured in plasma and in the supernatants of the washed and gel-filtered platelets after activation by thrombin. The lysis showed decreased speed of PPP-clots in PV and ET. This phenomenon was even more marked in PRP-clots from PV and ET, but further increased lysis resistance after retraction was not observed in PV and ET, most likely due to abnormal platelet functions. Our results suggest that the fibrinolytic activity is reduced in PV and ET, and may play a role both in the increased aptitude for venous thrombosis and in the arterial complications. These are partly caused by higher plasmatic PAI-1 activity as well as by more active platelet PAI-1. The PAI-1 activity was significantly higher in the supernatants of the washed and gel-filtered platelets of PV after activation by thrombin compared with controls. Other factors might have influenced the reduced fibrinolysis.

Platelet activation and thrombosis: studies in a patient with essential thrombocythemia

Recent advances permit the detection of activated platelets using specific monoclonal antibodies and flow cytometry. Nevertheless, there are few reports in which activated platelets have been studied over a period of time in patients at risk for thrombosis. Our patient S.D. has essential thrombocythemia and a prothrombotic state manifested in two major thrombotic episodes involving the portal vein and a mesenteric artery. Investigation revealed both spontaneous aggregation and hyperaggregability in response to ADP and the presence of activated platelets in platelet-rich plasma as revealed by flow cytometry. Interestingly, the activated platelets were recognized by an anti-RIBS ("receptor-induced binding site") monoclonal antibody that recognized bound fibrinogen but not by antibodies reactive with antigens whose presence on the platelet surface was secretion dependent. Treatment with aspirin inhibited spontaneous platelet aggregation but had little effect on the activated platelet profile. A change of therapy to ticlopidine suppressed expression of platelet activation markers. Treatment with ticlopidine has continued for 1 year so far without further thrombotic complications.

Essential thrombocytosis: underemphasized cause of large-vessel thrombosis

PURPOSE

The purpose of this study was to describe the clinical course of patients seen with large-vessel thrombosis in association with essential thrombocytosis (ET).

METHODS

This study was a retrospective review of all patients treated for large-vessel thrombosis caused by ET during a 2-year period at University of Washington teaching hospitals.

RESULTS

Five patients presented with arterial (femoral-popliteal-tibial: aortic), portal (two cases), or systemic venous (inferior vena cava) thrombosis and required operation. Two were known to have ET; in three others ET was diagnosed after operation when platelet counts persistently in excess of 500,000/mm3 were noted. The diagnosis of ET was established in each case by ruling out causes of reactive thrombocytosis and (in the three new cases) by evidence for megakaryocyte hyperplasia on bone marrow biopsy. Platelet counts in all five patients were reduced to normal levels by cytoreductive therapy, and no further thrombotic episodes have occurred during 18 months (mean) of follow-up. During this 2-year period ET accounted for more large-vessel thrombotic complications in our institutions than all other more frequently described hypercoagulable states combined.

CONCLUSIONS

ET is an underemphasized cause of large-vessel thrombosis.

Alpha-2a interferon therapy and antibody formation in patients with essential thrombocythemia and polycythemia vera with thrombocytosis

In ten patients with essential thrombocythemia and polycythemia vera with thrombocytosis we have investigated the therapeutic effect of recombinant alpha-2a interferon (Roceron-A) given subcutaneously in a maintenance dosage of 3 million units three times weekly. The aim was to normalize the platelet count (< or = 400 x 10(9)/L). One of the secondary aims was to study platelet activity measured as beta-thromboglobulin (beta-TG) in urine. All but one patient could administer the injections and in all patients a significant reduction in platelet values was seen. The treatment was discontinued in three patients due to side effects of interferon, two because of hair loss (one with irreversible alopecia), and one because of depression. Three patients developed antibodies to alpha-2a interferon and a concomitant rise in the platelet level; in one patient therapy was switched to leukocyte alpha-interferon with an excellent response. The initial levels of beta-TG were elevated in 9/10 patients and were significantly reduced at 6 months in 4/5 patients not developing antibodies. Six patients are still on alpha-interferon therapy with a long-term follow-up of 3-3.5 years. We conclude that alpha-interferon therapy may be an alternative in patients with thrombocytosis and/or complications necessitating treatment.

High plasma levels of plasminogen activator inhibitor 1 (PAI-1) in polycythemia vera and essential thrombocythemia are associated with thrombosis

We attempted to determine if a hypercoagulability state exists in patients with polycythemia vera (PV) and essential thrombocythemia (ET). We studied the hematocrit level, platelet count, use of any antiaggregant drugs, thrombotic or bleeding accidents and plasma levels of antithrombin III, protein C, total protein S, free protein S, vWF:Ag (Von Willebrand's factor related antigen), thrombin-antithrombin complexes, D-dimer, fibrinolytic activity, tissue plasminogen activator, plasminogen and PAI-1 in 33 patients (19 with ET and 14 with PV). PAI-1 plasma concentration was significantly higher in, both ET and PV patients than in the control group, and were higher in those patients with previous thrombotic episodes than in asymptomatic patients or with previous bleeding episodes. Increasing age was associated to more thrombotic episodes while younger patients presented with more hemorrhagic complications. A linear correlation between platelet count and PAI-1 levels in PV patients (r = 0.44, p < 0.05) and ET patients (r = 0.30, p < 0.05) was found. Fibrinolytic activity in patients with ET was significantly decreased when compared to the control group. A hypofibrinolytic state could be an additional factor which could be used as a predictive index of the thrombotic or bleeding tendency in each patient.

Issues in the diagnosis and management of essential thrombocythemia

OBJECTIVE

To summarize the current trends in the diagnosis and management of essential thrombocythemia (ET) and to discuss the treatment of young and pregnant patients with ET.

DESIGN

We review our experiences in the diagnosis and management of ET.

MATERIAL AND METHODS

A definitive diagnosis of ET at initial examination is seldom possible because of the low incidences of associated clonal cytogenetic abnormalities and palpable splenomegaly. The criteria of the Polycythemia Vera Study Group for the diagnosis of ET are provided, as are the clinical and laboratory features that help to distinguish ET from reactive thrombocytosis.

RESULTS

Patients with ET have an almost normal life expectancy if thrombohemorrhagic complications are controlled. Although a previous history of thrombosis is an established risk factor for further episodes, the adverse effects of extreme thrombocytosis and age are less well defined. Determining which patients require therapy is controversial. The management of young or pregnant patients necessitates special considerations.

CONCLUSION

Until additional data become available, no strong recommendations can be given in support of or against specific therapy for asymptomatic young patients. We favor treatment in asymptomatic patients with cardiovascular risk factors but not in asymptomatic women who are pregnant or are of childbearing age.

Familial essential thrombocythemia: clinical characteristics of 11 cases in one family

Reports on familial occurrence of essential thrombocythemia (ET) are scanty. Many clinical and hematological aspects of familial ET have not been clarified yet. We studied 16 family members in four successive generations. By laboratory tests and bone marrow examination they were divided into a non-thrombocythemia group (n = 5) and into ET patients (n = 11). Five ET patients were asymptomatic, three patients had both vaso-occlusive and hemorrhagic symptoms, and three patients only vaso-occlusive symptoms. The platelet count ranged from 500 to 1700 x 10(9)/l. Symptoms correlated with age but not with platelet count. ADP-induced platelet aggregation distinguished best between patients and non-ET subjects. Four patients and four non-ET subjects had factor VIII:C or von Willebrand factor antigen abnormalities; all but one had blood group O. These abnormalities were not due to inherited von Willebrand's disease according to haplotype analysis. Two patients and three non-ET subjects had a bleeding diathesis. One of these two patients and all three non-ET subjects had a decreased factor VIII:C or vWF:Ag. No chromosome abnormalities were found. In conclusion, familial ET has a relatively benign course with clinical manifestations similar to nonfamilial cases, and it is probably transmitted by an autosomal dominant mode of inheritance.