Essential thrombocythemia

What every physician needs to know:

Essential thrombocythemia (ET) is a disease of the bone marrow hematopoietic stem cell. It is classified as one of the “myeloproliferative neoplasms” (formerly termed “myeloproliferative disorders”).

Related myeloproliferative neoplasms (MPN), from which ET is sometimes difficult to differentiate, include the following:

  • Polycythemia vera (PV)

    Continue Reading

  • Myeloid metaplasia and myelofibrosis

  • Chronic myelogenous leukemia (CML)

The characteristic abnormality of ET on routine blood counts (CBC) is an elevated platelet count (thrombocytosis). In contrast, the characteristic predominant CBC finding in PV is an elevated hemoglobin and hematocrit; in myeloid metaplasia and myelofibrosis it is pancytopenia, and in CML, it is an elevated white blood count. ET is still mainly a diagnosis of exclusion, and requires that one rule out other causes of thrombocytosis.

In the past, the diagnosis of ET required a platelet count greater than 600,000 per cubic millimeter. However, it is now recognized that individuals with only borderline thrombocytosis (platelet counts of 450,000 to 600,000 per cubic mililiter) may have ET.

Are you sure your patient has essential thrombocythemia? What should you expect to find?

Most patients with ET are asymptomatic at presentation and are discovered to have the incidental finding of thrombocytosis on routine CBC. Features of the history that should raise suspicion that the elevated platelet count is due to ET, rather than to reactive or secondary form of thrombocytosis include an unexplained bleeding tendency and/or, paradoxically, venous or arterial thrombosis and vascular ischemic manifestations.

Bleeding manifestations of ET tend to occur at superficial sites, typical for platelet disorders (as opposed to coagulation factor deficiencies). Easy bruising, including unusually prominent bruising with the use of aspirin and non-steroidal antiinflammatory drugs, is characteristic. Other superficial bleeding manifestations of ET include mucosal hemorrhage, such as epistaxis, hemoptysis, gingival oozing or bleeding from the gastrointestinal or genitourinary tracts. Although it may seem counterintuitive, ET patients with extreme thrombocytosis (greater than 1,500,000 per cubic millimeter) are more prone to bleeding complications.

Arterial and microvascular thrombosis is more common than venous thromboembolism in ET. History of thrombosis may take the form of venous thromboembolism, most commonly deep vein thrombosis and/or pulmonary embolism, and less commonly portal, hepatic (Budd-Chiari syndrome), splenic vein and mesenteric thrombosis. Arterial thrombotic and vascular manifestations that may precede diagnosis of ET are most commonly cerebrovascular.

Neurologic complications occur in about 25% of ET patients; in addition to stroke, various forms of transient ischemic attacks, migraine, atypical migraine, other types of headache, episodes of dizziness, and visual symptoms may be involved. Coronary artery disease may occur even without established risk factors. Peripheral vascular disease characteristically manifests as microvascular ischemia, including the syndrome of “erythromelalgia”. Erythromelalgia is characterized by intense burning or throbbing pain in a patchy distribution over the plantar aspects of the feet or, less commonly, the palmar aspects of the hands. The pain of erythromelalgia is sometimes associated with the finding of corresponding areas of warmth and eythema, but without the loss of palpable peripheral pulses.

Younger women with ET may have a history of recurrent, spontaneous miscarriages. Systemic symptoms such as weight loss, fevers and night sweats are unusual in ET; their occurrence should raise the suspicion of an alternative diagnosis.

The physical exam in most patients with ET is normal. Splenomegaly is found in about 40% of ET patients, although it is rarely massive. Slight hepatomegaly is seen in about 20% of cases. Lymphadenopathy is not characteristic. Examination of the peripheral blood smear typically shows both an increased number of platelets and giant sized platelets. Other laboratory findings in ET are described below.

Beware of other conditions that can mimic essential thrombocythemia:

Thrombocytosis is much more commonly due to secondary (or reactive) causes than ET. This is true even where the platelet count is extremely high (greater than 1 million per cubic millimeter). Not infrequently, an isolated finding of thrombocytosis is not confirmed on repeat CBC; these cases are presumably due either to transient, subclinical illness or to laboratory artifact. The following are the more common causes of secondary (reactive) thrombocytosis:

  • Transient processes

– Acute blood loss

– Acute infection or inflammatory process

– Response to vigorous exertion

– Recovery (“rebound”) from thrombocytopenia

  • Sustained processes

– Cancer

– Chronic infectious or inflammatory disease (for example, tuberculosis, chronic pneumonia, inflammatory bowel disease, connective tissue disorder, temporal arteritis)

– Drug reactions (for example; vincristine, all-trans-retinoic acid, cytokines, growth factors)

– Iron deficiency

– Hemolytic anemia

– Asplenia (either surgical, following splenectomy, or functional, such as sickle cell disease)

While many of these disorders that cause secondary (reactive) thrombocytosis are clinically overt, others may be occult. Perhaps most worrisome in the differential diagnosis of thrombocytosis is the possible presence of an occult malignancy.

Several types of cancer are associated with elevated platelet counts. In patients with thrombocytosis on repeated blood counts, for whom ET is not felt to be a likely diagnosis and in whom secondary causes of thrombocytosis cannot be readily identified, occult malignancy should be seriously considered as the cause of the high platelet count. These individuals should have a thorough physical examination searching for evidence of cancer, stools examined for occult blood, chest radiogram, and further diagnostic studies as dictated by suspicious findings. Further, individuals with sustained thrombocytosis of unclear etiology in whom the diagnosis of ET has been ruled out clinically should have close follow-up.

In addition to the other secondary non-clonal causes of thrombocytosis listed above, the clonal myeloproliferative neoplasms related to ET (especially PV and sometimes myelofibrosis or CML) may be accompanied by thrombocytosis. Molecular markers of PV and CML, as noted below, should be diagnostic for these other related myeloproliferative neoplasms.

Finally, some myelodysplastic syndromes are associated with thrombocytosis. An elevated platelet count is particularly seen with the myelodysplastic syndrome associated with deletion of chromosome 5 (the 5q- syndrome).

Which individuals are most at risk for developing essential thrombocythemia:

ET is most frequently diagnosed in older individuals (median age of about 60 years at diagnosis), and the gender distribution is about equal between men and women. There is a less common subset of younger patients with apparently typical ET, the majority of whom are women. ET is rarely seen in children. No environmental risk factors for the disease have been identified.

An inherited predisposition to ET and other myeloproliferative neoplasms is recognized in occasional families with multiple affected members. These individuals are clinically indistinguishable from those with typical sporadic disease. Some of these unusual familial ET patients have low penetrance clonal hematopoiesis, as do typical patients with sporadic ET. In contrast, a completely separate form of familial thrombocytosis is occasionally seen as an inherited disorder that affects only the megakaryocyte lineage with Mendelian inheritance, high penetrance and polyclonal hematopoiesis. This latter form of familial thrombocytosis is associated with specific mutations in the genes for thrombopoietin or its receptor (see below).

The original cases were described as autosomal dominant disorders with gain-of-function mutations in the thrombopoietin gene, leading to overproduction of thrombopoietin and marked elevations of its plasma levels. However, these types of polyclonal familial thrombocytosis are now known to be genetically heterogeneous disorders.

What laboratory studies should you order to help make the diagnosis and how should you interpret the results?

Thrombocytosis must be first confirmed by repeat blood counts. Many individuals with an isolated high platelet count on initial CBC have normal platelet counts on subsequent study and do not require further evaluation.

For those patients with confirmed thrombocytosis, accompanying elevations in the hemoglobin/hematocrit and/or white blood count should raise the suspicion of a related but different myeloproliferative neoplasm. In those with isolated thrombocytosis, secondary causes should be sought by history and physical examination (see differential diagnosis above). For example, if the patient is found to be iron deficient, it should be confirmed that the thrombocytosis resolves with iron replacement. If the patient has an infectious or inflammatory disorder, the platelet count should promptly return to normal with treatment or resolution of the underlying process. Symptoms or signs of an occult malignancy should be sought.

While splenomegaly is found in about 40% of ET patients (see above), its presence is not diagnostic, as some of the inflammatory, infectious and malignant causes of secondary thrombocytosis may also be accompanied by splenomegaly.

Once the secondary causes of thrombocytosis have been eliminated, tests should be obtained that would implicate a myeloproliferative neoplasm such as ET.

In patients with ET and other myeloproliferative neoplasms:

  • Examination of the peripheral blood smear often shows giant platelets

  • Bone marrow aspirate and biopsy should be performed in most cases

– It may show giant, dysplastic megakaryocytes with increased ploidy, associated with large masses of platelet debris in ET (although just an increased number of normal-appearing megakaryocytes in the marrow is also seen in secondary forms of thrombocytosis). There may be increased fibrosis in ET and other myeloproliferative neoplasms, as well as abnormal cytogenetics. Iron deficiency can be ruled out by visible iron stores with special stains of the marrow (although this by itself is not an indication to do the bone marrow exam)

  • Molecular analysis of blood for the Janus kinase 2 (JAK2)-V617F mutation

– A qualitative test for this mutation is routinely performed by reverse transcription polymerase chain reaction (PCR) on a peripheral blood sample. Quantitative tests are available to monitor mutant allele burden, but these are not routinely ordered at this time. Up to 50% of patients with clinical ET have the JAK2-V617F mutation (while 95% of PV have the mutation)

  • JAK2-V617F negative

– If the patient is JAK2-V617F negative, other molecular markers may be considered in selected cases to diagnose a myeloproliferative neoplasm that has taken an atypical clinical form of ET. Sometimes CML phenotypically and clinically masquerades as ET; this can be ruled out by a negative test for the BCR-ABL rearrangement. Other unusual mutations associated with JAK2-V617F-negative ET include JAK2 exon 12 mutations and mutations in the myeloproliferative leukemia gene (MPL) which encodes the thrombopoietin receptor, including MPL-W515 and MPL-S505 mutations (Table I)

Table I.
JAK2-V617F mutation (%) JAK2Exon12 mutation (%) MPL-W515 mutation (%)
Essential thrombocythemia (ET) 50 0 1
Polycythemia vera (PV) 95 2 to 4 0
Myelofibrosis (MF) 50 0 5 to 7

Caveats to interpreting these molecular tests include:

  • They detect mutations only in the specific exons tested, while mutations or polymorphisms in other locations will not be detected

  • False-negative results occur when the mutant allele burden in the peripheral blood is below detection

– In conjunction with clinical and other laboratory correlations, however, the detection of one of the described mutations is diagnostic of a myeloproliferative neoplasm and rules out secondary or reactive thrombocytosis.

What imaging studies (if any) will be helpful in making or excluding the diagnosis of essential thrombocythemia?

There are no definitive imaging studies available for the diagnosis of ET. Documenting the presence of splenomegaly by imaging is usually not necessary because it is not a specific finding for ET, and may occur in some secondary causes of thrombocytosis. However, if there is clinical suspicion of a more complex cause of splenomegaly, such as Budd-Chiari syndrome or portal vein thrombosis, abdominal imaging may be important to direct management.

If you decide the patient has essential thrombocythemia, what therapies should you initiate immediately?

In patients with secondary (reactive) thrombocytosis, all efforts should be directed at diagnosing and treating the underlying cause of thrombocytosis. In these patients, the high platelet count per se, however elevated, does not pose risk and therefore does not require treatment with platelet-lowering or antiplatelet drugs.

In patients with ET (or other related forms of clonal thrombocytosis), the first decision is whether or not any kind of treatment (as opposed to just observation) is needed. In those who have independent cardiovascular risk factors that would require antiplatelet prevention even in the absence of coexisting ET, low-dose aspirin should be used.

Therapy in ET requires urgent initiation only in patients with active cerebrovascular, coronary, or digital ischemia. In these cases, rapid platelet-lowering is most readily achieved by initiating treatment with oral hydroxyurea (see below). Additionally, antiplatelet therapy with aspirin can be started immediately unless there is an absolute contraindication. Platelet pheresis is rarely performed, but is the most rapid (albeit transient) way to reduce the platelet count when critically needed.

When patients present for the first time with thrombocytosis and acute or unstable vascular manifestations, there may not be sufficient time to pursue a logical series of diagnostic tests to determine the cause of the elevated platelet count. In these cases, the emergency measures described above can be initiated (unless absolutely contraindicated) under the assumption that ET is a possible diagnosis, until the patient is stabilized and further evaluation can be resumed.

More definitive therapies?

The long-term management of patients with ET should be dictated by stratification of risk for thrombotic complications. Current consensus is that advanced age (greater than60 years), previous history of thrombosis or vascular events defines ET patients at “high risk”. Some experts would add to the list of “high risk” criteria the presence of cardiovascular risk factors and coexisting leukocytosis. Interestingly, the degree of thrombocytosis is not a reliable indicator of thrombotic risk. “Low risk” patients with ET have none of these characteristics.

“Low risk” ET patients can be followed without therapy, even without aspirin (unless they have independent, coexisting cardiovascular risk factors). Although some clinicians still recommend platelet-reducing therapy in “low risk”, asymptomatic ET patients with only extreme thrombocytosis (platelet count greater than 1,500,000 per cubic millimeter), there is little evidence that the benefits of this approach outweigh potential risks attributable to long-term treatment, especially in younger patients with the disease.

“High risk” ET patients should have treatment to reduce their platelet count and maintain it within normal limits (preferably less than 400,000 per cubic millimeter). The drug of first choice is hydroxyurea, starting with an oral dose of 0.5 to 1g daily. The dose of hydroxyurea is subsequently adjusted upward by 0.5g per day every 1 to 2 weeks, with frequent monitoring of blood counts to attain and maintain the desired platelet level, without significantly suppressing the hemoglobin and white blood count. In addition to reversible marrow suppression, some studies have raised concern about the potential of hydroxyurea to cause acute leukemia after prolonged use, but this is disputed.

Anagrelide is a second line platelet-reducing therapy for “high risk” patients with ET. In a prospective, randomized study, hydroxyurea plus low-dose aspirin was found to be superior to anagrelide plus low-dose aspirin, for patients with ET at high risk for vascular events. Initial oral dose of anagrelide is 0.5mg four times daily or 1mg twice daily, with 0.5mg upward dose adjustments at weekly intervals (up to a maximum of 10mg per day) to maintain a normal platelet count.

Up to 30% of patients cannot tolerate this drug because of its vasodilatory and positive inotropic properties that may cause fluid retention, palpitations, arrhythmias, heart failure, and headaches. Use of this drug requires caution in elderly patients and in those with pre-existing cardiac disease. Anagrelide is also associated with reversible bone marrow fibrosis. An alternative second-line drug that is effective in reducing the platelet count is interferon alfa; its use is limited by severe side effects that make it intolerable for about 20% of patients. Pegylated interferon alfa has a superior pharmacokinetic and toxicity profile compared with standard interferon alfa; it has been recently reported to be effective in ET (using a starting dose of approximately 90mcg by subcutaneous injection weekly). In women with “high risk” ET who are considering pregnancy, interferon alfa is the treatment of choice, because hydroxyurea is teratogenic and anagrelide crosses the placenta (with as yet unknown safety implications).

Although not yet available for routine clinical use, small molecule tyrosine kinase inhibitors are in late-phase development. Early experience with some of these drugs shows that they are particularly useful in controlling splenomegaly and constitutional symptoms (which are much more common in myeloid metaplasia and myelofibrosis, than in ET or PV).

What other therapies are helpful for reducing complications?

Low-dose aspirin (75 to 325mg daily) can be given optionally to “low risk” ET patients, unless there is a history of significant bleeding complications. Antiplatelet therapy alone has been shown to reduce the incidence of venous thrombosis, specifically in JAK2-V617F mutation-positive “low risk” ET patients; these individuals would therefore particularly benefit from low-dose aspirin. For “high risk” ET patients, low-dose aspirin should be prescribed, along with one of the platelet-lowering drugs unless contraindicated. In aspirin-resistant or intolerant individuals, an alternative antiplatelet drug (for example, clopidogrel) could be used.

Screening for and treating or modifying traditional cardiovascular risk factors (hypertension, diabetes, smoking, obesity, hyperlipidemia) are at least as important in the management of patients with ET, as in the general population. Control of cardiovascular risk factors as adjunctive management with ET-specific treatment (as described above) is critical to minimizing risk of thrombotic and vascular complications in this disease.

What should you tell the patient and the family about prognosis?

Survival data for patients with ET is variable. One analysis of actuarial survival indicated no significant decrease in life expectancy. However, another population-based study did show that survival of patients with ET was worse than that for age and gender matched healthy control subjects. Thrombotic and vascular complications are the major causes of death in ET. In a small minority of patients, transformation to acute leukemia, myelodysplasia or myelofibrosis may occur, even without the use of cytoreductive therapy.

"What if" scenarios.

In general, the complications of ET are managed as they would be in other individuals who do not have ET. Primary or secondary prophylaxis against cardiovascular disease with aspirin is indicated for patients at risk. Venous thromboembolic events are treated with anticoagulation as in those without ET.


ET is a clonal disorder of the pluripotent hematopoietic stem cell in the bone marrow. It is related to other clonal chronic myeloproliferative neoplasms, including PV, myeloid metaplasia and myelofibrosis, and CML (as well as some less common diseases like chronic eosinophilic leukemia and systemic mast cell disease).

The primary somatic “clonogenic” event in ET is still unknown. However, whatever the ET-initiating mutation may be in the bone marrow stem cell, it causes a growth advantage of that abnormal clone over nontransformed progenitors, leading to myeloproliferation with preferential (but not exclusive) dominance of the megakaryocyte/platelet blood cell line. When the JAK2-V617F mutation was discovered in many patients with chronic myeloproliferative neoplasms, including almost all patients with PV, it was hoped that it represented the initiating somatic mutation. However, JAK2-V617F could not explain all the clinicopathologic correlations in the various myeloproliferative neoplasms, and subsequently other contributing mutations have been identified.

Thrombopoietin (THPO) is the key hormone that regulates megakaryocyte differentiation and proliferation in the marrow. THPO stimulates platelet production in a manner that is comparable to erythropoietin in its primary role in regulating erythropoiesis. States associated with secondary or reactive thrombocytosis are driven either directly or indirectly by increased THPO production. In contrast, the thrombocytosis in ET is caused mainly by increased sensitivity of the megakaryocytic cell line to THPO. In fact, megakaryopoiesis may be almost THPO-independent in ET.

The “JAK/STAT” (Janus kinase/signal transducers and activators of transcription) pathway plays a central role in initiating signal transduction from the cell surface receptors of hematopoietic growth factors to stimulate cell proliferation. In the case of ET, this is mainly the THPO receptor, also known as MPL, on the surface membranes of megakaryocytes.

Under normal circumstances, initiation of intracellular JAK/STAT signaling is triggered by occupancy of the receptor (THPO-receptor or MPL) by its ligand (THPO). However, a single guanine-to-thymine somatic mutation encoding a valine-to-phenylalanine substitution at position 617 (V617F) in the catalytically inactive pseudokinase domain of JAK2, releases the normally autoinhibitory action of JAK2 in quiescent progenitor cells that have unoccupied growth factor receptors. This leads to constitutive activation of the JAK2 tyrosine kinase, and consequent activation of downstream intermediates that stimulate marrow stem cell proliferation, even in the absence (or with only normally subthreshold levels) of occupancy of the hematopoietic growth factor receptor to which it is coupled. Thus, JAK2-V617F is a gain-of-function mutation.

While JAK2-V617F may be a disease-initiating event in PV (where it is present in over 95% of cases), and indeed may be the diagnostic genotypic marker for true PV, it is likely that there is a “pre-JAK2” phase in the other myeloproliferative neoplasms, including ET. Since there is considerable phenotypic overlap between ET and PV (including the occurrence of thrombocytosis in both diseases), ET patients positive for the JAK2-V617F mutation (about 50% using current clinical diagnostic criteria) may actually represent a forme fruste of PV.

What other clinical manifestations may help me to diagnose essential thrombocythemia?


What other additional laboratory studies may be ordered?


What’s the evidence?

Beer, PA, Erber, WN, Campbell, PJ, Green, AR. “How I treat essential thrombocythemia”. Blood.. vol. 117. 2011. pp. 1472-1482. [Review of management of ET, including risk stratification, consideration of special situations, and areas of controversy.]

Harrison, CN, Campbell, PJ, Buck, G. ” Hydroxyurea compared with anagrelide in high-risk essential thrombocythemia”. N Engl J Med.. vol. 353. 2005. pp. 33-45. [Randomized Study.]

Schafer, AI. “Thrombocytosis”. N Engl J Med. vol. 350. 2004. pp. 1211-9. [Concise review with special emphasis on differentiating various forms of thrombocytosis.]

Skoda, RC. “Thrombocytosis”. Hematology Am Soc Hematol Educ Program.. 2009. pp. 156-67. [Review with emphasis on molecular mechanisms of clonal thrombocytosis.]

Spivak, JL. “Narrative review: thrombocytosis, polycythemia vera, and JAK2 mutations: the phenotypic mimicry of chronic myeloproliferation”. Ann Intern Med.. vol. 152. 2010. pp. 300-306. [Discussion of how JAK2 genotype determines phenotype of myeloproliferative neoplasms.]

Tefferi, A, Vainchenker, W. “Myeloproliferative neoplasms: molecular pathophysiology, essential clinical understanding, and treatment strategies”. J Clin Oncol.. vol. 29. 2011. pp. 573-582. [Comprehensive review of pathophysiology with clinical correlations.]

Teofili, L, Larocca, LM. “Advances in understanding the pathogenesis of familial thrombocythaemia”. Br J Haematol.. vol. 152. 2011. pp. 701-712. [Molecular pathogenesis of hereditary thrombocytosis, with an approach to the diagnosis and therapy of these syndromes.]