Relation of platelet abnormalities to thrombosis and hemorrhage in chronic myeloproliferative disorders

Severe acquired platelet dysfunction because of primary myelofibrosis with full functional and morphological recovery after allogeneic hematopoietic cell transplantation

: Primary myelofibrosis (PMF) is a clonal hematopoietic stem cell disorder characterized by fibrosis of the marrow cavity, marked megakaryocyte atypia and progressive cytopenias. Although thrombosis predominates, bleeding is the primary manifestation in up to 20% of patients and may be life-threatening. In this report, we document restoration of megakaryocyte and platelet structure and function in PMF after allogeneic hematopoietic cell transplantation (HCT). A 59-year-old man presented with recurrent episodes of postoperative bleeding preceding a diagnosis of primary myelofibrosis (PMF). Platelet aggregation and secretion studies showed abnormal responses to all agonists tested (epinephrine, ADP, arachidonic acid, U46619, collagen, ristocetin) despite the presence of thrombocytosis. After an allogeneic HCT, platelet morphology and function studies were all normal. The pathophysiology of platelet dysfunction in myeloid neoplasia is not well understood but, as highlighted in our report, restoration of platelet function by HCT supports a clonal process involving an early hematopoietic progenitor cell.

Major hemorrhage in chronic lymphocytic leukemia patients in the US Veterans Health Administration system in the pre-ibrutinib era: Incidence and risk factors

Chronic lymphocytic leukemia (CLL) patients are at increased risk for major hemorrhage (MH). We examined incidence of and risk factors for MH in CLL patients before introduction of newer CLL therapies such as ibrutinib, which includes bleeding risk. This study included 24 198 CLL patients treated in the VA system before FDA approval of ibrutinib as CLL therapy. Data came from VA databases from 1999 to 2013. MH incidence was 1.9/100 person-years (95% CI: 1.8-1.9), with cumulative incidences of 2.3%, 5.2%, and 7.3% by year 1, 3, and 5, respectively. Median time from CLL diagnosis to MH was 2.8 years (range: 0-15.7 years). In multivariate analyses, concurrent anticoagulant and antiplatelet use (HR: 4.2; 95% CI: 3.2-5.6), anticoagulant use only (HR: 2.6; 95% CI: 2.3-3.1), and antiplatelet use only (HR: 1.5; 95% CI: 1.3-1.7) increased MH risk vs not receiving those medications; being nonwhite, male, having MH history, renal impairment, anemia, thrombocytopenia, and alcohol abuse were associated with increased MH risk. These pre-ibrutinib data are important for providing context for interpreting MH risk in ibrutinib-treated patients. As ibrutinib clinical use is increasing, updated analyses of MH risk among ibrutinib-treated VA patients with CLL may provide additional useful insight.

High platelet count at diagnosis is a protective factor for thrombosis in patients with essential thrombocythemia

To assess the role of platelet (PLT) count for thrombotic complications in Essential Thrombocythemia (ET), 1201 patients followed in 11 Hematological centers in the Latium region were retrospectively evaluated. At multivariate analysis, the following factors at diagnosis were predictive for a worse Thrombosis-free Survival (TFS): the occurrence of previous thrombotic events (p=0.0004), age>60years (p=0.0044), spleen enlargement (p=0.042) and a lower PLT count (p=0.03). Receiver Operating Characteristic (ROC) analyses based on thrombotic events during follow-up identified a baseline platelet count of 944×10⁹/l as the best predictive threshold: thrombotic events were 40/384 (10.4%) in patients with PLT count >944×10⁹/l and 109/817 (13.3%) in patients with PLT count <944×10⁹/l, respectively (p=0.04). Patients with PLT count <944×10⁹/l were older (median age 60.4years. vs 57.1years., p=0.016), had a lower median WBC count (8.8×10⁹/l vs 10.6×10⁹/l, p<0.0001), a higher median Hb level (14.1g/dl vs 13.6g/dl, p<0.0001) and a higher rate of JAK-2-V617F positivity (67.2% vs 41.6%, p<0.0001); no difference was observed as to thrombotic events before diagnosis, spleen enlargement and concomitant Cardiovascular Risk Factors. In conclusion, our results confirm the protective role for thrombosis of an high PLT count at diagnosis. The older age and the higher rate of JAK-2 V617F positivity in the group of patients with a baseline lower PLT count could in part be responsible of this counterintuitive finding.

How I treat polycythemia vera

Polycythemia vera (PV) is a chronic myeloproliferative neoplasm associated with JAK2 mutations (V617F or exon 12) in almost all cases. The World Health Organization has defined the criteria for diagnosis, but it is still unclear which parameter (hemoglobin or hematocrit) is the most reliable for demonstrating increased red cell volume and for monitoring response to therapy; also, the role of bone marrow biopsy is being revisited. PV is associated with reduced survival because of cardiovascular complications and progression to post-PV myelofibrosis or leukemia. Criteria for risk-adapted treatment rely on the likelihood of thrombosis. Controlled trials have demonstrated that incidence of cardiovascular events is reduced by sustained control of hematocrit with phlebotomies (low-risk patients) and/or cytotoxic agents (high-risk patients) and antiplatelet therapy with aspirin. Hydroxyurea and interferon may be used as first-line treatments, whereas busulfan is reserved for patients that are refractory or resistant to first-line agents. However, there is no evidence that therapy improves survival, and the significance of reduction of JAK2 mutated allele burden produced by interferon is unknown. PV is also associated with a plethora of symptoms that are poorly controlled by conventional therapy. This article summarizes my approach to the management of PV in daily clinical practice.

Changes in response to antiaggregatory treatment in patients with myeloproliferative neoplasms: a sequential study using multiple electrode aggregometry

In the present study, we used multiple electrode aggregometry (MEA) to investigate the response to aspirin and clopidogrel treatment, and its potential changes over a long-time disease course in patients with myeloproliferative neoplasms (MPNs). arachidonic acid (ASPI), ADP, and thrombin receptor activating peptide (TRAP) tests were performed at two timepoints between 32-50 months in 21 patients with MPN and 1-46 months in 29 controls. We further checked the medical records of the participants to identify a potential correlation of changes in the treatment response with clinical events. In MPN, four out of 13 patients treated with 100 mg of aspirin, no patients receiving 50 mg of aspirin, and one out of five clopidogrel-treated patients showed a therapeutic antiplatelet effect. In the subsequent examinations, five patients changed from response to nonresponse or vice versa. Initial nonresponse and changes from an initial response to nonresponse were observed in six patients with thrombotic events. In the controls, 25 out of 26 aspirin-treated patients and two out of three clopidogrel-treated patients showed an initially adequate in-vitro response. Except from one patient changing from initial aspirin nonresponse to response, all controls showed a stable response state. One control with two ischemic strokes showed a nonresponse to clopidogrel. In conclusion, MEA detects the response to antiaggregatory treatment, as well as its changes during the disease course in patients with MPN. An initial or subsequent nonresponse was observed in patients with thrombotic events.

Thrombin generation and procoagulant phospholipids in patients with essential thrombocythemia and reactive thrombocytosis

Thrombocytosis is a commonly encountered clinical scenario and can be either a secondary process (reactive thrombocytosis), or due to clonal disorder (i.e., essential thrombocythemia). This distinction is important as it carries implications for evaluation, prognosis and treatment. In this study we compared procoagulant potential in essential thrombocythemia and reactive thrombocytosis by measuring the thrombin generation and the level of circulating procoagulant phospholipids with functional tests. Twenty nine patients with essential thrombocythemia and 24 with reactive thrombocytosis were studied. Thrombin generation was determined by calibrated automated thrombography. Procoagulant phospholipids were detected by a chronometric standardised method (STA-Procoag-PPL). Patients with reactive thrombocytosis had a longer lag time, higher endogenous thrombin potential, peak of thrombin generation and velocity index than patients with essential thrombocythemia. The level of circulating procoagulant phospholipids was increased in patients with essential thrombocythemia as observed with the procoagulant phospholipids assay. Each parameter was analysed using ROC curves. Highest areas under the curve (AUC) were found for lag time and procoagulant phospholipids ratio (0.817 and 0.853, respectively), associated with high negative predictive value for ET (92.3% and 80%, respectively). In conclusion, patients with essential thrombocythemia and reactive thrombocytosis displayed significant differences in terms of thrombin generation and levels of procoagulant phospholipids. Among these parameters, lag time and procoagulant phospholipids ratio could help to differentiate between reactive thrombocytosis and essential thrombocythemia patients.

JAK2V617F leads to intrinsic changes in platelet formation and reactivity in a knock-in mouse model of essential thrombocythemia

The principal morbidity and mortality in patients with essential thrombocythemia (ET) and polycythemia rubra vera (PV) stems from thrombotic events. Most patients with ET/PV harbor a JAK2V617F mutation, but its role in the thrombotic diathesis remains obscure. Platelet function studies in patients are difficult to interpret because of interindividual heterogeneity, reflecting variations in the proportion of platelets derived from the malignant clone, differences in the presence of additional mutations, and the effects of medical treatments. To circumvent these issues, we have studied a JAK2V617F knock-in mouse model of ET in which all megakaryocytes and platelets express JAK2V617F at a physiological level, equivalent to that present in human ET patients. We show that, in addition to increased differentiation, JAK2V617F-positive megakaryocytes display greater migratory ability and proplatelet formation. We demonstrate in a range of assays that platelet reactivity to agonists is enhanced, with a concomitant increase in platelet aggregation in vitro and a reduced duration of bleeding in vivo. These data suggest that JAK2V617F leads to intrinsic changes in both megakaryocyte and platelet biology beyond an increase in cell number. In support of this hypothesis, we identify multiple differentially expressed genes in JAK2V617F megakaryocytes that may underlie the observed biological differences.

Dysfunction of the PI3 kinase/Rap1/integrin α(IIb)β(3) pathway underlies ex vivo platelet hypoactivity in essential thrombocythemia

Patients with myeloproliferative disorders (MPDs), such as essential thrombocythemia (ET) have increased risk of thrombosis and bleeding, which are major sources of morbidity and mortality. Most MPD patients have a gain of function mutation in Janus kinase 2 (JAK2V617F), but little is known how JAK2V617F affects platelet function. Here, we demonstrate that platelets from ET patients have impaired SFLLRN-mediated fibrinogen binding and have lost the potentiating effect of thrombopoietin (which couples to JAK2) on this pathway. In contrast, SFLLRN-mediated P-selectin expression, ATP secretion, phosphorylation of the PKC substrate pleckstrin, and Ca(2+) mobilization were unaffected in JAK2V617F positive platelets. In addition, thrombopoietin-mediated JAK2 phosphorylation was unchanged, suggesting that signaling pathways activated downstream of JAK2 are impaired. Indeed, we found that platelets from JAK2V617F positive ET patients have significantly reduced phosphorylation of the PI3 kinase substrate Akt, and have reduced activation of Rap1 in response to thrombopoietin, IGF-1,ADP, SFLLRN, and thrombin. This effect was independent of Giα P2Y12 purinergic receptor function as ADP-mediated inhibition of VASP phosphorylation was unchanged. These results demonstrate that the PI3 kinase/Rap1 pathway is intrinsically impaired in platelets from JAK2V617F-positive ET patients, resulting in diminished thrombin and thrombopoietin-mediated integrin α(IIb)β(3) activation.

Prevalence of venous thromboembolism in patients with secondary polycythemia

To investigate an association between secondary polycythemia and venous thromboembolism (VTE) risk, we performed a case-control study to compare the prevalence of VTE in participants with secondary polycythemia due to chronic obstructive pulmonary disease (COPD; N = 86) to that in age- and sex-matched controls with COPD without secondary polycythemia (N = 86). Although there was a significant difference in mean hematocrit between cases and controls (53.5% vs 43.6%, respectively; P < .005), we identified no difference in the number of total or idiopathic VTE events in the 2 groups. Patients with VTE, however, had a significantly higher body mass index than patients without VTE. Our findings suggest that secondary polycythemia alone may not be a significant risk factor for VTE but that VTE risk in this population may be related to known risk factors such as obesity. The role of phlebotomy for VTE risk reduction secondary polycythemia is therefore questionable.

Platelet-induced thrombin generation by the calibrated automated thrombogram assay is increased in patients with essential thrombocythemia and polycythemia vera

The platelet contribution to the thrombophilic state of patients with myeloproliferative neoplasms (MPNs), i.e., essential thrombocythemia (ET) and polycythemia vera (PV), remains uncertain. In this study we aimed to characterize the thrombin generation (TG) potential expressed by platelets from these subjects, compare it to normal platelets, and identify what factors might be responsible for platelet TG. In a group of 140 MPN patients (80 ET and 60 PV) and 72 healthy subjects, we measured the global procoagulant potential of platelet-rich plasma (PRP) utilizing the TG assay by the calibrated automated thrombogram (CAT). To characterize the procoagulant contribution of platelets in PRP, the TG of both isolated platelets and platelet-poor plasma was measured, and the platelet surface expression of TF was determined. Finally, the activation status of platelets was assessed by the levels of P-selectin expressed on platelet surface. MPN patients had significantly increased PRP and isolated platelet TG potential compared to controls. This was associated to the occurrence of platelet activation. Patients carriers of the JAK2V617F mutation showed the highest values of TG and platelet surface TF and P-selectin. Platelet TG potential was significantly lower in hydroxyurea(HU) compared to non-HU-treated patients and was lowest in HU-treated JAK2V617F carriers. In subjects not receiving HU, platelet TG significantly increased by JAK2V617F allele burden increment (P < 0.05).This study demonstrates a platelet-dependent form of hypercoagulability in MPN patients, particularly in those carriers of the JAK2V617F mutation. The cytoreductive therapy with HU significantly affects this prothrombotic phenotype.

Evaluation of platelet function and pharmacological platelet inhibition in patients with myeloproliferative disorders using multiple electrode aggregometry

BACKGROUND

The aim of this study was to describe platelet aggregation characteristics by multiple electrode aggregometry (MEA) and to evaluate MEA for its potential to detect platelet dysfunction and response to anti-aggregatory drugs in patients with myeloproliferative disorders (MPD).

METHODS

We compared the platelet response to arachidonic acid (ASPI test), adenosine diphosphate (ADP test) and thrombin receptor activating peptide (TRAP test) in hirudin-anticoagulated blood of 55 patients with polycythaemia vera and essential thrombocythaemia and 75 controls.

RESULTS

Comparing MPD patients and controls no statistically significant difference indicative of platelet dysfunction was found in MPD patients. Analysis of covariance revealed platelet- and leukocyte count as a significant influencing factor on MEA function. Furthermore we could demonstrate that ASA and clopidogrel treatment results in a statistically significant lower ASPI (Controls: p<0.0001, MPD: p<0.0001) and ADPtest value (MPD: p=0.00125) compared to untreated patients thereby validating the method for monitoring of anti-aggregatory therapy.

CONCLUSION

In this study MEA was confirmed as a valid method for monitoring of ASA and clopidogrel treatment in patients with MPD and normal control subjects. The platelet and leukocyte count were identified as major influencing factors on MEA aggregation tests both in MPD patients and controls. No functional platelet abnormalities were detected in MPD patients.

Increased platelet glycoprotein Ib receptor number, enhanced platelet adhesion and severe cerebral ischaemia in a patient with polycythaemia vera

The present study describes severe multiplex cerebral ischaemic laesions in a male patient being diagnosed with polycythaemia vera (PV). In contrast to previous publications, unique platelet receptor pattern with normal platelet count was identified. Glycoprotein Ib receptor number on the surface of resting platelets was increased two-fold and almost three-fold in case of activated platelets compared to the controls. More over, in an in vitro study when whole blood was circulated both at venous and arterial shear conditions and shear rate was adjusted according to the blood viscosity, platelet aggregate/thrombus formation was characteristic on surfaces covered with purified von Willebrand factor while in case of controls the surface was covered with single platelets or platelet monolayer. Similar results with pathological findings have not been published in PV until now. Our result undersigns the necessity of antiplatelet therapy of PV patients, even at normal platelet count.

The role of the Platelet Function Analyzer (PFA)-100 and platelet aggregometry in the differentiation of essential thrombocythemia from reactive thrombocytosis

INTRODUCTION

The most crucial component of all diagnostic criteria for essential thrombocythemia (ET) has been the exclusion of reactive thrombocytosis (RT). Our aim was to evaluate the diagnostic performance of the PFA-100 collagen-epinephrine (CEPI) cartridge test and epinephrine-induced aggregometry individually, but mainly combined, in the differentiation of ET from RT.

MATERIALS AND METHODS

26 patients with ET and 25 with RT were studied. Platelet function was analyzed by the PFA-100 and by light transmission aggregometry with epinephrine and ADP. The JAK2 mutational status was identified and hematological parameters, plasma von Willebrand factor antigen and activity levels were also assessed.

RESULTS

The sensitivity (Se), specificity (Sp), positive predictive value (PPV), and the negative predictive value (NPV) of PFA-100 CEPI vs epinephrine-induced aggregometry in the differentiation of ET from RT were estimated as follows: Se (%): 78.9 vs 84.6, Sp (%): 92.0 vs 96.0, PPV (%): 88.2 vs 95.7, NPV (%): 85.2 vs 85.7, respectively. When both of these methods were combined, a lower sensitivity of 68.4%, but a specificity of 100% was attained. The PPV observed with this double abnormal combination was 100% and the NPV 80.6%. Lastly, when we assessed the abnormality for either CEPI CT or epinephrine-induced aggregometry, the sensitivity was 100%, the specificity 88.0%, PPV 86.4% and NPV 100%. Thus, an abnormal combination was strongly suggestive of ET, while normal results with both methods excluded ET.

CONCLUSIONS

If our results are replicated by further studies, these two methods could be used very effectively as adjunct markers in the differentiation between ET and RT.

Phospholipase A2 activity in platelets of patients with primary thrombocythemia

Patients with primary thrombocythemia (PT) have both, bleeding and thrombotic events. Although platelet aggregation tests are usually abnormal, synthesis of thromboxane B2 (TxB2) by platelets is increased. This feature could be the consequence of an increased phospholipase activity or a facilitated metabolism of arachidonate by prostaglandin synthetase pathway. We studied the activity of phospholipase A2 as well the arachidonate metabolism in platelets of patients suffering from PT. Eleven patients and 11 controls were included. Platelets were labelled with [14C]arachidonic acid ([14C]AA). Lost of radioactivity from phospholipids and new radioactive prostanoids were evaluated in calcium ionophore A23187 activated platelets, to explore phospholipase A2 activity. This assay was also carried out in aspirin-incubated platelets. We also studied the formation of prostanoids in platelets activated by radioactive free arachidonic acid. Platelet aggregation studies of patients were abnormal. [14C]AA incorporation in platelet phospholipids was normal. Ionophore activated platelets from patients and controls lost 26.1 +/- 8.3% and 24.1 +/- 10.5% of radioactivity, respectively, mainly from phosphatidylcholine. The main arachidonate metabolite was 12-L-hydroxy-5,8,10,14-eicosatetraenoic acid (HETE), which comprised 14.1 +/- 5.1% of the radioactivity released from phospholipids in patients, and a similar amount in the controls (14.4 +/- 7.5%). Formation of TxB2 was also similar in patients (5.5 +/- 1.2%) and controls (4.9 +/- 2.9%). Formation of 12-L-hydroxy-5,8,10-heptadecatrienoic acid (HHT) was also normal. Ionophore A23187 activation of aspirinized platelets of patients released 19.5 +/- 7.4% of radioactivity from phospholipids, which was completely metabolized to HETE. Formation of prostanoids HETE, HHT and TxB2 by arachidonic acid activated platelets of patients was normal. Phospholipase A2 activity as well both cyclooxygenase and lipoxygenase activities in platelets of patients with PT were found to be normal.

Congenital and acquired polycythemias

INTRODUCTION

Polycythemias are characterized by an increased concentration of red blood cells. Because blood cell counts are a routine investigation, these disorders present to non-hematologic physicians. Polycythemia vera (PV), an acquired stem cell disease, is the most important variant.

METHODS

Selective literature review and the authors' own clinical experiences.

RESULTS AND DISCUSSION

Erythropoietin, which is produced in the kidneys, and its receptor system in the bone marrow, are of critical importance in polycythemia. Congenital polycythemias are caused by mutations of the Erythropoietin-receptor gene, hemoglobin variants, 2,3-bisphosphoglycerate mutase deficiency or by disturbances of renal oxygen sensing. Acquired polycythemias can occur secondary to hypoxia at high altitudes, or primarily through acquired mutations in the EPO-receptor signaling system (JAK2 mutations). Alternatively they may be caused by pulmonary or renal disease. An artificial erythrocytosis is induced by athletes through doping. Differential diagnosis comprises erythropoietin determination, JAK2 mutation analysis and if necessary hemoglobin electrophoresis. Only PV requires immediate treatment, because of a high thromboembolic risk. Epidemiological studies on polycythemias in German speaking countries are urgently needed.

Use of whole blood platelet lumi-aggregometry to optimize anti-platelet therapy in patients with chronic myeloproliferative disorders

Twenty-seven patients with chronic myeloproliferative disorders and in vitro evidence of platelet hyperactivity on whole blood platelet lumi-aggregometry were commenced on anti-platelet therapy comprising aspirin, clopidogrel, and/or odorless garlic and the studies were repeated to assess the efficacy of the therapeutic agent(s). Only 8 patients showed clear evidence of anti-platelet effect while receiving the standard low-dose (100 mg/day) aspirin therapy. Thirteen patients required a higher dosage of aspirin and/or an additional anti-platelet agent to achieve therapeutic adequacy. Lumi-aggregometry also proved useful to optimize therapy in the 6 patients who received clopidogrel or odorless garlic because of aspirin intolerance.

Thrombotic complications in essential thrombocythemia (ET): clinical facts and biochemical riddles

Hemostatic complications which can occur in the arterial or venous vasculature or in the microcirculation are the major causes of morbidity and mortality in patients with ET. In order to prevent these complications, often platelet reductive drugs are used. These agents are by themselves potentially toxic, i.e. may cause leukemia or cardiac side effects. In order to avoid these adverse effects, a better understanding of the mechanism of thrombus formation which is causative in ET is mandatory. Unfortunately, until now, no biomarkers have been identified which allow the estimation of the risk of thrombotic complications. Platelet number is not a good predictor per se since thrombotic complications can occur in some patients at low platelet numbers whereas others do not encounter a thrombosis even at very high platelet levels. On the other hand, lowering of the platelet count usually results in symptomatic improvement. In ET, morphological alterations of the megakaryocyte in the bone marrow and the circulating platelets are observed: megakaryocyte nuclei show a staghorn appearance, circulating platelets are characterized by anisocytosis and giant size. Functional studies indicate that these anatomically altered platelets function abnormally. When platelets are analyzed with a platelet function analyzer (PFA-100, which uses cartridges that measure how well a patient's platelets adhere and aggregate to form a platelet plug in the first phase of thrombus formation), in many patients with ET, closure time using collagen/ADP and collagen/epinephrine cartridges is prolonged. This seems paradoxical since these patients do not show an increased bleeding time. These results indicate that either receptors and/or consecutive signaling events are abnormal in ET platelets. Proteomic analysis of platelets of ET patients has revealed individual differences but not yet led to the identification of disease-specific proteins. Moreover, the search for alternative risk factors (factor V Leiden, prothrombin gene polymorphism, etc.) has not provided evidence for the contribution of these factors to the generation of the thrombotic risk in ET patients. In summary, despite intensive research over several decades, relatively little is known about the pathogenesis and risk factors for thrombosis in ET. I expect that this conference will contribute to the development of new strategies to identify patients at risk for hemostatic complications.

Lactate production by thrombin-activated platelets of patients with primary thrombocythemia

INTRODUCTION

Platelet activation needs a high energy demand which is supplied by the degradation of glucose into lactate. Platelet response to agonists in patients with primary thrombocythemia is defective. We studied the production of lactate by the platelets of patients with this disease and defective platelet aggregation.

MATERIAL AND METHODS

Ten patients suffering from primary thrombocythemia and ten controls were included in this study. The lactate generation was measured in resting and thrombin activated platelets in absence or presence of glucose.

RESULTS

Resting platelets incubated for 30 min in phosphate-buffered saline (PBS) generated the same amount of lactate in patients (44.6+/-21.6 micromol/10(11) cells) and controls (41.0+/-17.3 micromol/10(11) cells). Addition of glucose led to similar increases in lactate formation by platelets in patients (82.2+/-26.4 micromol/10(11) cells) and controls (88.1+/-34.5 micromol/10(11) cells). The addition of thrombin in absence of glucose did not modify the lactate formation respective to PBS. Finally, the incubation of platelets with both glucose and thrombin caused further increases in the generation of lactate in both groups, patients (236.9+/-83.9 micromol/10(11) cells) and controls (228.6+/-63.5 micromol/10(11) cells) without differences between them. The production of lactate in both groups was also similar when platelets were incubated for 10 min or 20 min with both thrombin and glucose. However at 5 min, platelets of patients generated more lactate (97.8+/-23.7 micromol/10(11) cells) than controls (66.5+/-38.7 micromol/10(11) cells, p<0.05).

CONCLUSIONS

These results suggest that thrombin is able to induce an initial hyperactivity of those pathways involved in the platelet energy production of patients with primary thrombocythemia.

Assessment of platelet activation in myeloproliferative disorders with complementary techniques

Bleeding and thrombosis in myeloproliferative disorders (MPD) are common events, sometimes both are present in the same patient during the course of the disease. Platelet activation in patients with MPD is often suggested. The present study analyses the presence of circulating activated platelets, using simultaneously flow cytometry and aggregometric studies in MPD. We studied 28 patients: 13 with polycythaemia vera, seven with essential thrombocythaemia, and eight chronic myeloid leukaemia. We performed functional tests, aggregation and adenosine triphosphate (ATP) release and flow cytometric assays (mepacrine staining and platelet activation markers CD62, CD63 and fibrinogen binding (B-FG)). Twenty-one MPD samples (75%) had reduced aggregation and ATP release. Acquired delta-SPD was detected in 11 of 28 MPD patients (39%), and we found no association between reduced mepacrine labelling and abnormal ATP release. High levels of activation markers were obtained: CD62 in 19 of 28 patients (68%), CD63 in 13 of 28 patients (46%) and B-FG in 19 of 28 patients (68%). The most prevalent abnormality was a reduced aggregation and ATP release. The lack of association between ATP release and mepacrine labelling suggests that other mechanisms, besides the deficit of intraplatelet ATP/adenosine diphosphate, might occur. High levels of activation markers were also observed. We conclude that both tests are complementary and necessary to understand the functional status of platelets in MPD.

[Unclear liver fibrosis in a 42-year-old patient with polycythemia vera]

A 42-year-old patient was admitted to hospital because of ascites and polyglobulia. Laboratory tests revealed reduced liver function and a significant elevation of all three hematopoietic cell lines. Liver fibrosis and polycythemia vera were diagnosed by histologic examination. The most frequent causes for liver fibrosis were serologically excluded. Ultrasound combined with Doppler imaging revealed an obstruction of the right hepatic vein, which was indicative of Budd-Chiari syndrome. BCS can occur under fulminant and nonfulminant conditions, which can result in progressive damage of the liver. Phlebotomy and combined therapy with low-dose aspirin and anagrelide achieved permanent reduction of the elevated blood parameters. In the follow-up the patient's clinical course was stable without hepatic decompensation.

Constitutively activated phosphatidylinositol 3-kinase primes platelets from patients with chronic myelogenous leukemia for thrombopoietin-induced aggregation

In this study, we examined the effect of thrombopoietin (TPO) on the aggregation of platelets from 40 patients with myeloproliferative disorders (MPDs), including 17 patients with chronic myelogenous leukemia in the chronic phase (CML-CP), 10 with polycythemia vera, 10 with essential thrombocythemia, and three with myelofibrosis. TPO by itself dose-dependently induced the aggregation of platelets from patients with CML-CP but not from those with other MPDs or with CML-CP in cytogenetical complete remission. The expression of CD63 in CML-CP platelets was induced by TPO treatment. Phosphatidylinositol 3-kinase (PI3-kinase) was constitutively activated in CML-CP platelets. Pretreatment with PI3-kinase inhibitors (wortmannin and LY294002) dose-dependently inhibited TPO-induced aggregation of CML-CP platelets. The Abl kinase inhibitor imatinib mesylate and the Jak inhibitor AG490 suppressed TPO-induced aggregation of CML-CP platelets. Pretreatment with imatinib mesylate, but not with AG490, inhibited the activity of PI3-kinase in CML-CP platelets. In addition, tyrosine phosphorylation of Jak2 was undetected in CML-CP platelets before TPO treatment. These findings indicate that the constitutive activation of PI3-kinase primes CML-CP platelets for the aggregation induced by TPO, and that Bcr-Abl, but not Jak family protein tyrosine kinases, are involved in the constitutive activation of PI3-kinase in CML-CP platelets.

Propensity for hemorrhage and thrombosis in chronic myeloproliferative disorders

Polycythemia vera (PV) and essential thrombocythemia (ET) are two myeloproliferative disorders (MPDs) with frequent thrombotic and hemorrhagic complications. Thrombosis is often the cause of mortality in PV and ET; hemorrhage occurs more commonly in idiopathic myelofibrosis patients, but is rarely fatal. Thromboses may occur in arteries or veins. Splanchnic, portal, hepatic, and splenic vein thromboses are not uncommon and thrombosis is also thought to cause placental vascular insufficiency and fetal wastage during pregnancies in MPD patients. These complications may result because of altered interactions between platelets, white blood cells, or endothelial cells, due to either altered receptor expression, receptor-ligand interactions, or signaling events. Age, leukocytosis, increased hematocrit, and a history of thrombotic events are risk factors for thrombosis. In determining a link between clonality and thrombosis using X-chromosome inactivation patterns in patients with ET, those who were polyclonal were less likely to experience thromboses. The search for hypercoagulability in these patients led to identification of changes in the expression patterns of coagulation proteins from the coagulation cascade. Mutations in factor V Leiden were examined and the incidence of mutations did not vary between normal and MPD patients. However, mutations in factor V Leiden were found to be risk factors for venous thrombotic events. Similarly, presence of a prothrombin gene mutation showed a higher risk for venous thromboembolic events. Proteolyzed thrombospondin appeared to contribute to hypercoagulability, and acquired von Willebrand factor disorder gave rise to hemorrhagic complications. These findings provide several potential reasons for thrombotic and hemorrhagic complications in MPD patients. Therefore, the best therapy for these patients is reduction of their platelet counts to less than 450,000/microL and close regulation of their hematocrits. The role of leukocytosis in bleeding or hemorrhage in this population remains to be elucidated.

The predictive value of megakaryocytic and erythroid colony formation and platelet function tests on the risk of thromboembolic and bleeding complications in essential thrombocythaemia

The predictive value of spontaneous in vitro colony formation of megakaryocytic and erythroid progenitors (154 patients), and defective platelet aggregation responses (55 patients) on the risk of thrombohaemorrhagic complications in patients with essential thrombocythaemia (ET) was evaluated retrospectively. In the in vitro cultures of haematopoietic progenitors, 114/154 patients (74%) showed either spontaneous megakaryocytic or erythroid colony formation or both. Forty-three per cent of patients with any spontaneous colony growth and only 20% of those without this phenomenon had an arterial thrombosis at diagnosis or during the follow-up (P = 0.02). In the whole patient group neither spontaneous megakaryocytic nor spontaneous erythroid colony formation alone predicted the risk of arterial thrombosis. In patients younger than 45 yr of age, the prognostic value of spontaneous megakaryocytic growth was statistically significant: 44% of the patients with spontaneous megakaryocytic colony formation, but only 14% of those without it, experienced arterial thrombosis (P = 0.04). The presence of spontaneous colony formation had no effect on the risk of bleeding complications. Forty-one of the 55 patients (75%) showed abnormalities in the platelet aggregation responses. There was no statistically significant correlation between the platelet function response and the risk of bleeding or thrombotic complications. No correlation was found between the platelet aggregation responses and the presence of spontaneous colony growth. In conclusion, spontaneous colony formation indicated an increased risk of thrombohaemorragic events but the platelet function test had no predictive value for these complications.

Sticky platelet syndrome and thrombocythemia

Platelets are intimately involved in the pathogenesis of thromboembolic disorders, especially arterial forms of thrombosis. Although most arterial thromboses develop on the basis of endothelial injuries, some do not. In these instances "hyperactive" platelets could be the cause. Hyperaggregable platelets have been described in association with a number of acquired disease entities whereby the cause-and-effect relationship is unclear. In contrast, the sticky platelet syndrome is a congenital, autosomal dominant disorder, characterized by hyperaggregable platelets in response to ADP, epinephrine, or both. Patients usually seek treatment for transient or permanent arterial vascular occlusions. These are often precipitated by stressful events. Treatment with low-dose aspirin (81 mg/day) reverses clinical symptoms and hyperaggregability in the laboratory.

Oral anticoagulants as secondary prophylaxis of thrombosis in patients with polycythemia vera: a retrospective analysis of 15 patients

Thrombotic complications are the leading cause of mortality in patients with polycythemia vera (PV). Phlebotomy and myelosuppressive agents are able to reduce the frequency of thrombosis; however, despite adequate control of myeloproliferation, the thrombotic risk remains increased in PV. This has led to a search for additional antithrombotic agents suited to reduce the rate of thrombotic events in PV. We have retrospectively analyzed 15 patients with PV who received oral anticoagulants (OA) as secondary prophylaxis of thrombosis at our institution with the aim of investigating this treatment approach in patients with PV. Despite OA treatment, 2 (13%) of 15 patients displayed recurrent venous thromboembolism, 3 patients (20%) displayed recurrent arterial thromboembolism, whereas 1 patient (7%) developed recurrent restenosis/occlusion of a transjugular intrahepatic porto-systemic shunt (TIPSS) implanted because of an underlying Budd-Chiari syndrome. Three (20%) of 15 patients had a total of 5 bleeding events during OA treatment, two of which were major bleedings. In none of the major bleedings over-anticoagulation was the reason for bleeding. Within the limitations of a retrospective analysis in a small number of patients, our data suggest that meticulous control of anticoagulant therapy is of critical importance in patients with PV, since in several of our patients recurrent thromboembolic events were associated with a low anticoagulation intensity of INR<2.0. In our opinion, a prospective clinical trial similar to the recently performed ECLAP study would be needed to more thoroughly elucidate the efficacy of OA in the secondary prophylaxis of thrombosis in PV.

Markers of platelet activation and platelet-leukocyte interaction in patients with myeloproliferative syndromes

INTRODUCTION

Changes in platelet count and function contribute to thrombo-hemorrhagic episodes in chronic myeloproliferative syndromes (MPS). We used flow cytometry to study platelet-leukocyte conjugates and markers of platelet activation in patients with MPS.

METHODS

Whole blood from patients with chronic myelogenous leukemia (CML), polycythemia vera (PV), chronic myelofibrosis (MF), and essential thrombocythemia (ET) and from healthy volunteers was prepared for flow cytometry. Platelet microparticles and platelet microaggregates were identified with anti-CD42b and forward scatter, activated platelets with anti-CD62p. Anti-CD42b, anti-CD14, and anti-CD45 were used to study platelet-leukocyte conjugates.

RESULTS

The percentage of CD62p-positive platelets was elevated in all myeloproliferate syndrome subtypes. The median percentage of platelet microparticles was 5.2% in controls and significantly higher in PV (12.0%), MF (11.0%), and ET (11.0%, all p<0.05). There was an increased percentage of platelet-neutrophil conjugates in patients with PV (8.3%) and ET (10.4%) compared to normal controls (6.8%, all p<0.05). Platelet-monocyte conjugates were 8.0% in controls and elevated in PV (15.4%) and ET (15.0%, all p<0.05). Patients with a history of venous or arterial thrombotic events had slightly less platelet-leukocyte conjugates and slightly more microparticles than patients without thrombosis; however, this difference was not statistically significant.

CONCLUSIONS

These findings suggest that platelet-leukocyte conjugate formation occurs in myeloproliferative syndromes and indicates platelet activation. Also, platelet microparticles are elevated and might provide a catalytic surface for thrombin generation. This could explain the clinical observation that patients with myeloproliferative syndromes have an increased risk to experience arterial or venous thrombotic events.

Frequent occurrence of anticardiolipin antibodies, Factor V Leiden mutation, and perturbed endothelial function in chronic myeloproliferative disorders

Chronic myeloproliferative disorders (MPD) are characterized by a high incidence of thrombohaemorrhagic complications, possibly related to platelet abnormalities and disturbances of the coagulation system. In an attempt to define abnormalities in coagulation and fibrinolysis, we investigated risk markers for venous thromboembolism, fibrinolytic, and haemostatic system activation markers and antiphospholipid antibodies in blood samples from 50 MPD patients and 30 controls. Compared with controls median levels of protein S free and protein C were significantly decreased in the patients (0.27 vs. 0.38 arbitrary units; P < 0.001 and 0.86 vs. 0.99 arbitrary units; P < 0.001, respectively), and activated partial thromboplastin time was significantly prolonged in patients (33 vs. 27 sec; P < 0.001). No differences were observed in levels of antithrombin, thrombin-antithrombin complex, and fibrin D-dimer. In patients the median value of thrombomodulin was significantly increased indicating perturbed endothelial function. Anticardiolipin antibodies of IgM subtype (median 38 U/mL, 33-99) were detected in 11 patients (22%) and only in one control (3%) (P < 0.021; patients vs. controls). Seven patients were heterozygous for the Factor V Leiden, one patient was heterozygous for the prothrombin G20210A mutation, and one patient was homozygous for the Factor V Leiden mutation. Among controls, two were heterozygous for the Factor V Leiden mutation. Comparing the allele frequency of the Factor V Leiden mutation in patients with MPD and the background population disclosed a significantly increased allele prevalence of the Factor V Leiden mutation in the patients (9% vs. 3.4%; P = 0.003). Alterations in the level of anticoagulant proteins, disturbances of endothelial cell function, and the presence of cardiolipin antibodies and Factor V Leiden mutation may increase the cumulative thrombotic risk in MPD besides the risk imposed by platelet activation.

Relationships between thrombohemorrhagic complications and platelet function in patients with essential thrombocythaemia

Three subgroups have been distinguished in essential thrombocythaemia (ET) patients, on the basis of clinical and laboratory findings. ET patients with bleeding incidents had smaller platelet volume, lower concentrations of beta-thromboglobulin and platelet factor 4 in their plasma, 10%, 26%, and 26% lower compared to patients without complications, respectively. ATP secretion from platelets of bleeders, clotters, and "no-complications" ET patients was found to be 75%, 36%, and 45%, respectively, lower than in healthy people. Spontaneous platelet aggregation appeared to be normal in about 90% of ET patients with no complications and in all bleeders but only in 35% patients with clotting incidents. All bleeders had abnormal agonist-evoked aggregation assays. Among remaining ET patients 30%-60% displayed normal values of different evoked aggregation tests. Thus, clinically distinguished group of bleeding ET patients may be differentiated from other subgroups on the basis of laboratory findings.

Increased circulating platelet-leukocyte aggregates in myeloproliferative disorders is correlated to previous thrombosis, platelet activation and platelet count

Platelet-leukocyte adhesion may occur as a consequence of platelet activation and possibly plays a key role in the deposition of activated platelets and fibrin in the thrombotic plug. The aim of the present study was to assess by whole blood flow cytometry the presence of circulating platelet-leukocyte aggregates (PLA) and the platelet-leukocyte response to platelet agonist stimulation (ADP and TRAP) in 50 patients with chronic myeloproliferative disorders (MPD) and 30 controls. PLA were identified as platelet-granulocyte/monocyte aggregates (PGMA), platelet-monocyte aggregates (PMA) and defined as the percentage of leukocytes coexpressing the platelet-specific marker glycoprotein Ib. Compared to controls the mean percentage of PGMA and PMA was increased in unstimulated whole blood from patients with MPD (7.98 vs. 1.76%; p<0.001 and 12.34 vs. 3.2%; p<0.001, respectively). The percentage of PGMA was correlated to the platelet count (r=0.46; p<0.001), percentage of P-selectin (r=0.69; p<0.001) and thrombospondin (r=0.58; p<0.001) positive platelets and platelet expression of GPIV (r=0.33; p=0.02). The mean percentage of PGMA and PMA was significantly increased in ADP-stimulated whole blood of patients (57.14 vs. 47.92%; p=0.009 and 54.91 vs. 45.89%; p<0.001, respectively). Compared to patients without a history of thrombosis, patients having experienced microvascular disturbances or a thrombotic event had a higher mean percentage of PGMA and PMA in non-stimulated whole blood (10.07 vs. 6.34%; p=0.025 and 14.81 vs. 10.48%; p=0.021, respectively) and a higher percentage of PGMA in ADP stimulated whole blood (64.32 vs. 51.50%; p<0.01). These data document an increased frequency of PLA in non-stimulated whole blood in MPD associated with a previous history of thrombosis or microvascular disturbances.

Abnormalities of platelet aggregation in chronic myeloproliferative disorders

A large variety of platelet dysfunctions has been described in chronic myeloproliferative disorders. These abnormalities may be due to deficiency of platelet granules, arachidonic acid metabolism defects or platelet membrane glycoproteins abnormalities. In this study we intend to detect the incidence of platelet function defects in 76 patients with various types of chronic myeloproliferative disorders. The platelet activity was studied in vitro by measuring platelet aggregation in response to ADP, epinephrine, collagen, arachidonic acid and ristocetin. These results were subsequently correlated with bleeding time and clinical aspects (bleeding or thrombosis). We found complex changes in platelet response with all agonists, in varied proportions. These abnormalities include absent, decreased or abnormal platelet aggregation response. In a few cases we found a markedly decreased, almost absent platelet response to all agonists while in some patients a normal platelet aggregation was noted. The correlation between these results and template bleeding time, thrombotic or hemorrhagic events and the type of diseases was difficult to establish and sometimes conflictual. Despite this fact, we consider that investigating platelet aggregation may be useful not only for the assessment of the hemostatic balance in chronic myeloproliferative disorders but also for a better insight into cell abnormalities occurring in these pathologic conditions.

High prevalence of hyperhomocysteinemia due to marginal deficiency of cobalamin or folate in chronic myeloproliferative disorders

Hyperhomocysteinemia is an established risk factor for thrombosis. In patients with myeloproliferative disorders, thrombotic events are common. Our aim was to investigate whether the increased burden of proliferating cells present in these patients implies a risk of homocysteine (HCY) accumulation secondary to depletion of folate and/or cobalamin. Fifty patients (PV, 25; ET, 10; IMF, 15) and 163 healthy volunteers (HV) participated in the study. The prevalence of hyperhomocysteinemia was 56.0% in PV, 70.0% in ET, 60.0% in IMF, and 34.9% in HV. The mean P-homocysteine (P-HCY) was 13.88 +/- 4.24 micromol/L in PV, 12.78 +/- 3.70 in ET, 11.34 +/- 4.22 in IMF, and 9. 71 +/- 2.76 in HV. In PV and ET, but not in IMF, the mean P-HCY was significantly higher than in the HV group (P < 0.001, P = 0.028, and P = 0.163, respectively). Thirty-three percent of the patients with hyperhomocysteinemia displayed metabolic changes compatible with cobalamin deficiency (P-HCY and P-methylmalonic acid both elevated), while 67% were folate deficient (P-HCY elevated, P-methylmalonic acid normal). Supplementation therapy with the relevant vitamin was implemented in 11 vitamin-deficient patients and led to normalization of metabolite levels in all cases. No correlation between hyperhomocysteinemia and thrombosis was found. Our data indicate that patients with PV, ET, and IMF frequently develop hyperhomocysteinemia due to discrete depletion of cobalamin or folate. Vitamin therapy leads to normalization of P-HCY and should be considered, even though hyperhomocysteinemia does not seem to be of crucial importance for the thrombotic tendency in the myeloproliferative disorders.

Increased platelet activation and abnormal membrane glycoprotein content and redistribution in myeloproliferative disorders

Chronic myeloproliferative disorders (MPDs) are characterized by a high incidence of thrombohaemorrhagic complications, possibly caused by platelet dysfunction. In an attempt to define platelet functional abnormalities, we assessed the expression of activation-dependent membrane proteins in unstimulated and agonist [ADP and thrombin receptor-activating peptide (TRAP)]-stimulated platelets using quantitative whole blood flow cytometry in samples from 50 MPD patients and 30 controls. The receptor densities of activation markers and glycoproteins (GPs) were quantified using standardized fluorescent beads. Compared with controls, the mean percentage of P-selectin-positive (15.3% vs. 7.2%; P < 0.001) and thrombospondin (TSP)-positive (6.6% vs. 3.7%; P = 0.003) platelets was increased in unstimulated platelets from patients. Patients having experienced a thrombotic event had a higher mean percentage of TSP-positive non-stimulated platelets than patients without a history of thrombosis (9.0% vs. 4.6%; P = 0.02) and a higher GPIV molecules of equivalent fluorochrome (MEF) value (33113 vs. 24471 MEF; P = 0.02). Mean MEF values of monoclonal antibodies (mAbs) against GPIb (34055 vs. 38945 MEF; P < 0.001) and GPIIb/IIIa (1416 vs. 1648 MEF; P < 0. 001) were significantly reduced among patients, whereas surface expression of GPIV was increased in patients (28273 vs. 16258 MEF; P < 0.001). In TRAP (10 micromol/l) stimulated whole blood, the MEF of P-selectin (9611 vs. 13293 MEF; P = 0.004) and CD63 (2385 vs. 5177 MEF; P < 0.001) and the ratio of PAC-1/GPIIb/IIIa MEF (0.98 vs. 2. 00; P < 0.001) was reduced in patients, indicating either a reduced granule GP content or an intrinsic cellular defect in receptor-mediated granule secretion and activation of the GPIIb/IIIa complex. Expressed as the relative change of MEF compared with unstimulated platelets, TRAP induced decrease of GPIb (7.8% vs. 45%; P < 0.001) and increase of GPIIb/IIIa (49.1% vs. 95.7%; P < 0.001) and GPIV expression (17.8% vs. 55.2%; P < 0.001) was attenuated in patients.

Transforming growth factor beta1 enhances platelet aggregation through a non-transcriptional effect on the fibrinogen receptor

Upon activation, platelets store and release large amounts of the peptide transforming growth factor beta1 (TGFbeta1). The released TGFbeta1 can then act on nearby vascular cells to mediate subsequent vessel repair. In addition, TGFbeta1 may circulate to bone marrow and regulate megakaryocyte activity. It is not known what effect, if any, TGFbeta1 has on platelets. Adult TGFbeta1-deficient mice exhibit thrombocythemia and a mild bleeding disorder that is shown to result from faulty platelet aggregation. TGFbeta1-deficient platelets are shown to contain functional receptors, and preincubation with recombinant TGFbeta1 improves aggregation, demonstrating that TGFbeta1 plays an active role in platelet aggregation. TGFbeta1-deficient platelets fail to retain bound fibrinogen in response to aggregation agonists, but they possess normal levels of the alpha(IIb)/beta(3) fibrinogen receptor. Signaling from agonist receptors is normal because the platelets change shape, produce thromboxane A(2), and present P-selectin in response to stimulation. Consequently, activation and maintenance of alpha(IIb)/beta(3) into a fibrinogen-binding conformation is impaired in the absence of TGFbeta1. 4-Phorbol 12-myristate 13-acetate treatment and protein kinase C activity measurements suggest a defect downstream of protein kinase C in its activation cascade. Because platelets lack nuclei, these data demonstrate for the first time a non-transcriptionally mediated TGFbeta1 signaling pathway that enhances the activation and maintenance of integrin function.

Platelet dense granules: structure, function and implications for haemostasis

The advances that have been made over the last decade in microscopic, biochemical, molecular, and genetic techniques have led to substantial improvement in our understanding of platelet dense granule structure and function, and the implications of dense granule deficiencies for haemostasis. However, much has still to be learned. For example, what is the specific mechanism of docking and fusion that occurs during dense granule exocytosis? What are the roles of dense granule membrane proteins during exocytosis or after expression on the surface of activated platelets? Finally, how do the genetic defects identified in HPS and CHS result in the clinical phenotype of these diseases, and what does this tell us about the origin and function of the affected subcellular organelles?

Acute coronary occlusion and in-stent thrombosis in a patient with essential thrombocythemia

We describe a case of essential thrombocythemia in a 34-year-old male who presented with acute anterior wall myocardial infarction and a platelet count of 2,100,000/mm3. Primary percutaneous coronary angioplasty and stenting were performed. Postangioplasty course was complicated by stent thrombosis requiring repeat coronary angioplasty and persistent femoral arterial bleeding that was treated with surgical exploration and repair. The patient was subsequently treated with platelet pheresis, acetylsalicylic acid, ticlopidine, hydroxyurea, and anagrelide without further complications.

Anagrelide for treatment of patients with chronic myelogenous leukemia and a high platelet count

Chronic myelogenous leukemia (CML) is usually treated with hydroxyurea or interferon-alpha. In some patients high platelet counts develop although leukocyte counts are well controlled with these drugs. If in such a situation cytoreductive therapy has to be intensified by a increase of the dosage, anemia and leukocytopenia as well as adverse effects of the drugs are likely to occur. In twelve CML patients we have therefore combined the basic CML treatment with anagrelide. This drug which selectively reduces platelet counts has been shown to be efficacious in the control of thrombocytosis in essential thrombocythemia. The diagnosis had been confirmed in all CML patients by cytogenetic and/or molecular biological analysis. The median age of our group was 58 years. Five were women and seven men. All patients were on treatment with hydroxyurea, some of them had previously received treatment with interferon-alpha (alone or in combination with hydroxyurea), busulfan or melphalan. Prior to the initiation of anagrelide treatment the platelet count was between 970,000 and 3,600, 000/microl (median about 2,000,000/microl). Seven patients had thrombohemorrhagic complications. All twelve patients, experienced hematologic responses, since their platelet counts decreased to less than 600,000/microl. The median platelet count after reduction was 343,000/microl. The median dosage required to achieve these responses and to maintain them for a period of at least four weeks was 1.9 mg/day. Thrombohemorrhagic complications disappeared or did not recur in all symptomatic patients. Adverse effects were seen in 3/12 patients: headache (1), tachycardia (1), palpitation (1) and fluid retention (1). Whereas these symptoms were mild and transitory they caused one patient to request discontinuation of treatment. Currently five patients are still on treatment with anagrelide (median duration of treatment 11 months) while therapy had to be discontinued in the seven others because of bone marrow transplantation, development of osteomyelofibrosis, blast crisis or on patient request. In our experience anagrelide is a useful therapeutic adjunct when thrombocytosis in patients with CML cannot properly controlled alone with traditional drugs.