Clinical manifestations and diagnosis of polycythemia vera

Author
Ayalew Tefferi, MD
Section Editor
Stanley L Schrier, MD
Deputy Editor
Rebecca F Connor, MD
Disclosures: Ayalew Tefferi, MD Nothing to disclose. Stanley L Schrier, MD Nothing to disclose.
Rebecca F Connor, MD Employee of UpToDate, Inc. Equity Ownership/Stock Options (Spouse
previously owned): Pharmacyclics [B cell lymphomas (Ibrutinib)].
Contributor disclosures are reviewed for conflicts of interest by the editorial group. When found,
these are addressed by vetting through a multi-level review process, and through requirements for
references to be provided to support the content. Appropriately referenced content is required of
all authors and must conform to UpToDate standards of evidence.
Conflict of interest policy
All topics are updated as new evidence becomes available and our peer review process is
complete.
Literature review current through: Aug 2014. | This topic last updated: Jun 09, 2014.
INTRODUCTION Polycythemia vera (PV, polycythemia rubra vera, maladie de Vaquez) is one
of the chronic myeloproliferative neoplasms, which are collectively characterized by clonal
proliferation of myeloid cells with variable morphologic maturity and hematopoietic efficiency
(figure 1). PV is distinguished clinically from the other MPNs by the presence of an elevated red
blood cell mass (RCM). However, an increased RCM alone is insufficient to establish the
diagnosis, since this is also observed in conditions associated with chronic hypoxia and with
erythropoietin-secreting tumors (table 1). (See "Overview of the myeloproliferative neoplasms".)
The clinical manifestations and diagnosis of PV will be discussed here. The prognosis and
treatment of PV and the overall approach to the patient with polycythemia are discussed
separately. (See "Prognosis and treatment of polycythemia vera" and "Diagnostic approach to the
patient with polycythemia".)
DISEASE INCIDENCE PV occurs in all populations, and all ages, including early adulthood and
occasionally in children and adolescents [1-3]. The median age at diagnosis for the first 325
patients entered into the Polycythemia Vera Study Group (PVSG) main protocol was 60 years
(range 20 to 85) [4], while it was 61 years (range 18 to 95) for the 1545 patients with WHO-defined
PV collected by the International Working Group for Myeloproliferative Neoplasms Research and
Treatment (IWG-MRT) [5].
The incidence of PV in Olmsted County, Minnesota during the period from 1935 through 1989 was
estimated to be 1.9/100,000 per year, approximately one-half that for multiple myeloma in the
same population [6]. The incidence of PV is slightly higher in men than women (2.8 versus 1.3
cases/100,000 per year), and is highest for men aged 70 to 79 years (24 cases/100,000 persons
per year) [6].
Survival Median survival of patients with untreated PV has been estimated at 6 to 18 months
from the time of diagnosis, whereas survival of treated patients exceeds 10 years, although it
appears to be shorter than that of an age- and sex-matched United States population [5,7]. (See
"Prognosis and treatment of polycythemia vera".)
The median survival of 1545 PV subjects (23 percent followed to death) collected by the IWG-
MRT was 18.9 years.
The median survival of 337 Mayo Clinic PV patients (44 percent followed to death) was 14.1
years.
In the PVSG-01 major protocol, which compared treatment with phlebotomy, chlorambucil, or
radioactive P-32, the most common causes of death included [8]:
Thrombosis (29 percent)
Hematologic malignancies (ie, acute myeloid leukemia or myelodysplastic syndrome, 23 percent)
Non-hematologic malignancies (16 percent)
Hemorrhage (7 percent)
Post-PV myelofibrosis (3 percent)
CLINICAL PRESENTATION Some patients with PV are discovered incidentally when an
elevated hemoglobin or hematocrit is noted on a complete blood count obtained for some other
reason. The following frequency of symptoms, physical, and laboratory findings were noted by the
PVSG (table 2) and the IWG-MRT [4,5]:
Nonspecific complaints Nonspecific complaints such as headache, weakness, dizziness, and
excessive sweating were present in 48, 47, 43, and 33 percent, respectively. Acute gouty arthritis
has been described in 5 to 20 percent.
Pruritus Pruritus following a warm bath or shower (aquagenic pruritus) is characterized by
strong sensations in the skin following contact with water without visible changes in the skin [9-11].
In a questionnaire completed in 441 patients with PV, 301 (68 percent) of which had aquagenic
pruritus, the following information was obtained [11]:
Symptoms were described as itching, tickling, stinging, or burning in 72, 21, 31, and 18 percent,
respectively.
Almost one-half of the respondents noted that warm water caused stronger symptoms than cold
water, while 38 percent noted no difference between the two.
In 78 percent, symptom onset was within less than 10 minutes after water contact.
The most commonly involved symptomatic areas were the chest, back, medial side of the arms,
and ventral side of the legs.
Aquagenic pruritus is often the chief complaint of a patient with PV, has been described as
"unbearable" in 15 percent of those with this symptom, and may be present for many years before
the diagnosis of PV is made. In the study noted above, this symptom was present, on average,
approximately three years before the diagnosis of PV was made, and was the basis for suspecting
the diagnosis of PV in 15 percent [11]. This complaint was present in 43 percent of the patients
enrolled in the PVSG, 36 percent of the patients with WHO-defined PV, and in 31 percent of
patients with PV seen at the Mayo Clinic, where, for unexplained reasons, it was associated with a
significantly lower incidence of arterial thrombosis [12].
The cause of pruritus in PV is unclear. It has been suggested that mast cell degranulation, with
release of histamine, fibrinolytic factors, prostaglandins, or interleukin-31 may play a role [13-16].
An alternative hypothesis is that the release of adenosine diphosphate from red cells or
catecholamines from adrenergic vasoconstrictor nerves when the skin is cooled down might cause
platelet aggregation in skin vessels, with local production of pruritogenic factors, such as
prostaglandins [15]. The observation that aspirin can relieve pruritus in at least some patients is
compatible with an important role for prostaglandins [15].
Erythromelalgia Erythromelalgia, or burning pain in the feet or hands accompanied by
erythema, pallor, or cyanosis, in the presence of palpable pulses (picture 1) was seen in 28
percent of the WHO-defined PV patients. Erythromelalgia and the associated symptom of acral
paresthesias, both of which can be considered to be forms of dysesthesia, are considered to be
pathognomonic microvascular thrombotic complications in PV and essential thrombocythemia, and
are associated with platelet counts usually >400,000/microL [17,18]. These symptoms respond
dramatically to aspirin in low doses or to reduction of the platelet count to normal with low-dose
myelosuppressive agents [17-19]; they may be related to abnormal arachidonic acid metabolism in
PV platelets [19]. (See "Diagnosis and clinical manifestations of essential thrombocythemia",
section on 'Vasomotor symptoms'.)
Thrombosis A history of venous or arterial thromboses is common in PV. These complications
were noted in 7 and 16 percent, respectively, in the WHO-defined PV patients. Although the
mechanisms involved in this hypercoagulable state are unclear, abnormalities in blood viscosity,
platelets, and leukocytes have been implicated [20].
A prior major thrombotic complication (eg, cerebrovascular event, myocardial infarction,
superficial thrombophlebitis, deep vein thrombosis, pulmonary embolus) was present in 15 percent
of patients entering into the PVSG studies. (See "Prognosis and treatment of polycythemia vera".)
An arterial thrombotic complication, venous thrombosis, or major hemorrhage was noted prior to
or at the time of diagnosis in 16, 7.4, and 4.2 percent of the WHO-defined PV patients.
Major thrombotic events can occur in patients who otherwise have few clinical and laboratory
features of PV. Examples include the Budd-Chiari syndrome, and portal, splenic, or mesenteric
vein thrombosis [21], in whom the ensuing portal hypertension and hypersplenism may mask the
increase in blood cell counts [22-27]. PV should be suspected in patients with these diagnoses,
particularly women under the age of 45. (See "Etiology of the Budd-Chiari syndrome", section on
'Myeloproliferative disorders'.)
Transient visual disturbance Transient visual disturbances (eg, transient ocular blindness
[amaurosis fugax], scintillating scotomata, ophthalmic migraine) can occur in PV, similar to those
seen in patients with essential thrombocythemia [28].
In one study of 374 patients with PV, 53 presented with visual disturbance as an initial symptom.
Fluorescein angiography in the 21 patients who presented with transient ocular blindness
indicated significantly delayed choroidal and retinal blood flow when compared with results
obtained in 30 non-PV control patients [29]. On multivariate analysis, an increased hematocrit was
the only parameter significantly correlated with reduced choroidal and retinal blood flow.
Treatment of these 21 patients with phlebotomy and hydroxyurea to reduce their hematocrit to less
than 50 percent resulted in significant improvement in choroidal and retinal blood flow and a
reduction in the subsequent incidence of transient blindness. (See "Diagnosis and clinical
manifestations of essential thrombocythemia", section on 'Vasomotor symptoms'.)
Gastrointestinal symptoms Gastrointestinal complaints are common in PV, with a high
incidence of epigastric distress, history of peptic ulcer disease, and gastroduodenal erosions on
upper endoscopy [30]. These have been attributed to alterations in gastric mucosal blood flow due
to altered blood viscosity, and/or increased histamine release from tissue basophils, although one
study has indicated a high incidence of positivity for infection with Helicobacter pylori [30].
PHYSICAL EXAMINATION The major abnormal findings on physical examination in PV include
splenomegaly, facial plethora (ruddy cyanosis), and hepatomegaly which occurred in 70, 67, and
40 percent of patients entered into the original PVSG protocol, respectively. For the WHO-defined
PV patients, palpable splenomegaly was noted in only 36 percent.
Other physical findings which might be helpful in pointing to a diagnosis of PV include the
following:
Injection of the conjunctival small vessels and/or engorgement of the veins of the optic fundus
Excoriation of the skin, which might be extensive, suggesting the presence of severe pruritus, a
common complaint in PV (see 'Pruritus' above)
Stigmata of a prior arterial or venous thrombotic event (eg, stroke, deep vein thrombosis,
superficial thrombophlebitis)
Gouty arthritis and tophi
LABORATORY FINDINGS Laboratory findings in the original PVSG protocol patients included
an elevated hemoglobin/hematocrit and red blood cell mass in virtually all patients, a platelet count
>400,000/microL in 60 percent, and a white blood cell count >12,000/microL in 40 percent. Bone
marrow cellularity was increased in 90 percent of patients, and storage iron was absent from the
marrow in 94 percent.
For the WHO-defined PV patients, the following laboratory findings were noted [5]:
Hemoglobin >18.5 g/dL 73 percent
Total white blood cell count >10,500/microL 49 percent
Platelet count >450,000/microL 53 percent; platelet count >1 million/microL 4 percent
Leukoerythroblastic blood smear 6 percent
Elevated serum lactate dehydrogenase 50 percent
JAK2 mutation positivity 98 percent
Abnormal karyotype 12 percent
Subnormal serum erythropoietin level 81 percent
Endogenous erythroid colony formation 73 percent
Bone marrow fibrosis In a study of 526 subjects who met the WHO criteria for the diagnosis of
PV, 74 (14 percent) displayed minor (grade 1) reticulin fibrosis on their initial bone marrow
examination, and only two showed higher grade fibrosis [31]. Presenting clinical and laboratory
characteristics, including JAK2V617F allele burdens, between patients with and without fibrosis
were generally similar, although palpable splenomegaly was significantly more common in those
with fibrosis. While there was no difference between the two patient groups in terms of overall or
leukemia-free survival, those with fibrosis were significantly more prone to develop post-PV
myelofibrosis (2.2 versus 0.8 per 100 patient-years) and significantly less prone to experience
thrombosis during their clinical course (1.1 versus 2.7 per 100 patient-years).
Elevated cytokine levels Plasma levels of a number of cytokines are elevated in PV, with a
pattern differing from that seen in patients with primary myelofibrosis [32]. Direct phenotypic
correlations in 65 PV patients included levels of IL-2 with hematocrit; IL-1b, IL-2, IL-7, FGF-b, and
HGF with leukocytosis; and IFN-alpha and IFN-gamma with thrombocytosis. On multivariate
analysis, increased levels of MIP-1beta, older age, and leukocytosis were significantly associated
with shortened survival. (See "Prognosis and treatment of primary myelofibrosis", section on
'Elevated cytokine levels'.)
DIAGNOSTIC CRITERIA AND ALGORITHMS
Polycythemia Vera Study Group Criteria Diagnostic criteria for PV were proposed in the late
1960s by the Polycythemia Vera Study Group (PVSG) [4]. The criteria were based upon a
combination of major and minor findings (table 3) (calculator 1). They were developed prior to the
general availability of assays for erythropoietin, endogenous erythroid colony formation, specific
karyotypic, clonal, and JAK2 mutational analyses, and during the time when blood volume studies
were routinely available.
These PVSG criteria were designed originally to select patients for clinical trials during the period
from 1968 to 1974. The major objectives were to make diagnosis possible during an office visit
and to collect patients whose disease activity was relatively homogeneous. Because no specific
test for PV could serve as a gold standard, the criteria themselves became the gold standard over
time. However, more than 30 years of experience in applying these criteria to a broader range of
patients as well as the availability of genetic and molecular testing have led to a better
appreciation of both their accuracy and limitations.
Increased red blood cell mass The original PVSG criterion required a directly measured
elevated red blood cell mass (RCM, >36 mL/kg in men and >32 mL/kg in women), using isotope
dilution methodology. (See "Diagnostic approach to the patient with polycythemia", section on
'Blood volume measurement'.)
These values were set at approximately 10 to 20 percent above the upper limits of normal and
were deliberately chosen to reject patients with relatively early or inactive disease. However, this
specific PVSG research protocol requirement does not apply to the general clinical setting, in
which it may be preferable to include a broader range of patients. As a result, two revised
approaches have been proposed:
The International Council for Standardization in Hematology (ICSH) has recommended that
blood volumes be calculated in terms of body surface area, since normalization of blood volume to
total body weight is less accurate in obese patients [33]. RCM, normalized to body surface area,
should be considered elevated if it is more than 25 percent above the mean expected value [34].
It has been our experience and that of others that the majority of female patients with a
hemoglobin concentration >16.5 g/dL (or hematocrit [Hct] above 50 percent) and all male patients
with a hemoglobin concentration >18.5 g/dL (or Hct above 56 percent) have an increased RCM
[35,36]. Slightly different cutoff values were found in reports from Belgium and Sweden (Hct
greater than 55 [37] or 60 percent [38] in women and greater than 60 percent in men [37,38]).
However, determination of the RCM by isotopic dilution is no longer available in many locations.
Importantly, direct determination of RCM may be unnecessary in patients with hemoglobin or
hematocrit concentrations exceeding the above-noted values, especially since it is felt that direct
determination of the RCM is prone to error, has suboptimal sensitivity, and that the diagnosis of
PV can be made via alternative clinical and laboratory tests [35,39]. This position is not universally
held [36,38,40-43].
Palpable splenomegaly Palpable splenomegaly was present in 70 percent of 325 patients
entered into the first PVSG protocol [4]. Since splenomegaly may not always be detectable by
physical examination, other diagnostic criteria accept the finding of a nonpalpable spleen that is
enlarged on an imaging test as a minor or secondary criterion [34,44].
Thrombocytosis and leukocytosis A platelet count >400,000/microL or a total white blood cell
(WBC) count >12,000/microL in the absence of fever or infection are PVSG minor criteria for the
diagnosis of PV (table 3). However, the total WBC may not accurately reflect disease activity,
since neutrophils and not lymphocytes or monocytes are increased in PV. Thus, newer criteria
have replaced an elevated total WBC count with an elevated absolute neutrophil count
>10,000/microL [34,44].
Leukocyte alkaline phosphatase and serum B12 studies An elevated leukocyte alkaline
phosphatase (LAP) score, and elevated serum B12 or unbound serum B12 binding capacity are
minor PVSG criteria (table 3).
Although an elevated LAP score is reasonably sensitive for PV (approximately 70 percent), it is
not specific.
B12 studies are neither sensitive nor specific for the diagnosis of PV. (See "Physiology of vitamin
B12 and folate deficiency", section on 'Elevated levels of vitamin B12'.)
Accordingly, the LAP score and B12 studies are not included as major or minor criteria in any of
the newer published algorithms (see 'Other diagnostic algorithms' below).
False positive results Because secondary causes of an elevated RCM are much more common
than PV, the criteria should only be applied to populations in which other causes have been
excluded. It has been estimated that the practicing hematologist sees 10 cases of relative or
secondary polycythemia for every new patient with PV [4].
Application of the PVSG criteria has led to false positive results in patients with cirrhosis who
smoke heavily. Cirrhosis may be associated with splenomegaly (one of the major criteria), while
concomitant smoking may lead to an elevated RCM (another major criterion) due to high levels of
carbon monoxide.
Similarly, since splenomegaly, leukocytosis, and thrombocytosis are common to all of the chronic
myeloproliferative neoplasms (figure 1), a patient with chronic myeloid leukemia, essential
thrombocythemia, or primary myelofibrosis who has an elevated hematocrit for another reason
(eg, hypoxia, smoking) may be falsely diagnosed as having PV. This is an important distinction
clinically because each of these disorders is treated in a different manner.
Despite these concerns, it has been estimated that the false positive rate using the PVSG criteria
is only about 0.5 percent [8].
False negative results In contrast to the relatively low rate of false positive results using the
PVSG criteria, false negative results occur in approximately 10 percent of patients [4]. A number of
factors can contribute to this problem:
Calculation of the red blood cell mass may be falsely low in patients who are obese.
Patients with PV who have had significant recent gastrointestinal blood loss ("bled down"
polycythemia) may present with a normal hematocrit. Such patients may be incorrectly diagnosed
as having iron deficiency or essential thrombocythemia since their high platelet counts, low red
blood cell indices, and low serum ferritin concentrations may be the most striking features.
Patients who have PV and are also hypoxic for unrelated reasons (eg, coexisting chronic lung
disease) will not fulfill the PVSG criteria. These patients were generally excluded from the original
PVSG studies through the requirement of a directly determined arterial oxygen saturation at rest
>92 percent.
An increased red cell mass or markedly increased hemoglobin or hematocrit may not be
apparent in some patients with PV who may otherwise meet other major and minor criteria for the
diagnosis [45-47]. Such patients have been termed "masked" PV. (See 'Increased red blood cell
mass' above and 'Establishing the diagnosis' below.)
Patients with the Budd-Chiari syndrome often present without classical features of PV and may
not fulfill the PVSG or revised WHO criteria. The diagnosis can be suspected because many are
positive for the JAK2 mutation. (See "Etiology of the Budd-Chiari syndrome", section on
'Myeloproliferative disorders'.)
Other available diagnostic tests Criteria other than those originally proposed by the PVSG have
been used to make the diagnosis of PV. These are enumerated below.
Bone marrow aspiration and biopsy An increased number of megakaryocytes in a moderately
to markedly hypercellular marrow has been considered one of the diagnostic hallmarks of PV
[44,48]. Although specific bone marrow findings were not included as either major or minor PVSG
criteria, 281 pretreatment bone marrow biopsies were obtained in the first PVSG study and the
following observations were noted [49]:
The most common abnormality was the absence of stainable iron in 94 percent
Cellularity varied from 36 to 100 percent (mean 82 percent, normal: 35 to 50 percent)
The numbers of megakaryocytes and amount of reticulin were variable, although both were
generally increased
At the Mayo Clinic, this constellation of bone marrow findings is considered a PV-related feature
[41]. While the diagnostic value of this feature, as with the determination of red blood cell mass,
has been emphasized by others [43], this position is not universally held.
Clonal markers Bone marrow examination can also be used to identify clonal markers that have
been associated with PV, such as karyotypic changes and, in women, X-linked inactivation
patterns or restriction length polymorphisms. As an example, deletion of the long arm of
chromosome 20, trisomy for chromosomes 8 or 9, or loss of heterozygosity on the short arm of
chromosome 9 are found in up to 30 percent of previously untreated patients with PV [50-55]. Of
interest, chromosome 9p24 houses the JAK2 gene, which carries a somatic point mutation in
virtually all patients with PV (see 'JAK2 mutations' below).
These clonal features initially helped to establish the malignant character of PV, but were not
incorporated into the PVSG criteria [56]. Increased appreciation of the malignant nature of PV has
led to their inclusion into some, but not all, diagnostic criteria, because of the low incidence of such
findings [41].
Serum erythropoietin Patients with PV usually have low serum erythropoietin (EPO)
concentrations [57-61]. Although this was known when the PVSG criteria were developed, assays
were difficult to perform and not widely available. Standardization of methods for measurement of
serum EPO levels has now permitted a better understanding of their role in the diagnosis of PV.
In two studies, for example, serum EPO levels in 42 patients with PV were compared with control
subjects with other causes of polycythemia (eg, hypoxia) [57,62]. The sensitivity and specificity of
serum EPO levels below the reference range of normal for the diagnosis of PV (using the PVSG
criteria as the gold standard) were 64 and 92 to 99 percent, respectively. The sensitivity increased
to 72 percent in patients who were tested on two occasions, as low serum EPO values were noted
on a repeat study in some patients with normal levels on the initial examination. The serum EPO
concentration remained low even when the red cell mass was normalized following phlebotomy.
Accordingly, most of the newer diagnostic criteria for PV have included a low serum EPO level
(see 'Other diagnostic algorithms' below).
While low EPO levels are highly specific for PV, levels above normal are unusual and suggest
secondary erythrocytosis, with a specificity of 98 percent [62]. (See "Diagnostic approach to the
patient with polycythemia".)
A rare disorder that can mimic the basic findings of PV (ie, increased RCM with low serum EPO) is
an activating mutation in the EPO receptor [63-65]. Patients with this disorder may also have
endogenous erythroid colony formation, as described in the next section [66]. However, colony
growth in these patients can be distinguished from true PV because the colonies demonstrate in
vitro EPO hypersensitivity rather than EPO independence [63-65]. A positive family history, early
age at disease onset, and the lack of PV-associated clinical findings should provide adequate
information to distinguish this familial disorder from PV.
Endogenous erythroid colony formation A common feature of several myeloproliferative
neoplasms is the demonstration, using in vitro culture techniques, of the formation of erythroid
colonies in the absence of added (exogenous) erythropoietin (ie, endogenous erythroid colonies
[EEC]). However, the measurement of EEC is not a standard FDA-approved test and is available
only in a limited number of research laboratories that are especially interested in PV and
hematopoiesis.
The presence of EEC strongly supports the diagnosis of PV if other criteria are present. One
study, for example, evaluated EEC formation in 89 patients with various forms of polycythemia
[66]. EEC formation was found in 100 percent of PV patients with no prior cytotoxic chemotherapy,
50 percent of PV patients with prior cytotoxic therapy, and in no patient with secondary
polycythemia.
Thus, in polycythemic patients who have not had prior cytotoxic chemotherapy, the sensitivity and
specificity of EEC formation for the diagnosis of PV is approximately 100 percent. This observation
has been confirmed in other series [67,68].
JAK2 mutations When sensitive quantitative assays are employed, 95 to 100 percent of
patients with PV have a JAK2 mutation involving either exon 14 or 12 [5,69-74].
Exon 14 mutation In multiple studies, 95 to 97 percent of patients with PV have the V617F
mutation in exon 14 of the JAK2 gene, which is absent in normal subjects as well as those with
secondary polycythemia. Thus, this mutation, when present, enables one to distinguish patients
with PV from those with secondary polycythemia [41,75,76]. The finding of the JAK2 V617F
mutation is not specific for PV, since it is also present in a substantial proportion of patients with
ET as well as primary myelofibrosis. (See "Overview of the myeloproliferative neoplasms", section
on 'JAK2 mutations'.)
In a study of 63 patients with PV evaluated at the Mayo Clinic, the JAK2 mutation was present in
58 (92 percent); 45 and 13 were heterozygous or homozygous for this mutation in peripheral blood
granulocytes, respectively [77]. A comparison between the heterozygotes and homozygotes did
not reveal significant differences with regard to duration of disease or incidence of thrombosis or
bleeding. However, homozygotes had higher hemoglobin levels, a greater incidence of pruritus, a
higher rate of fibrotic transformation, and higher PRV-1 transcript levels in peripheral blood
granulocytes than the heterozygotes.
In other studies, high levels of this mutation were found to correlate with higher granulocyte counts
[78-80], potentially linking this mutation to the higher incidence of thrombosis seen in PV patients
with leukocytosis [81]. (See "Prognosis and treatment of polycythemia vera", section on
'Thrombotic events'.)
A higher rate of homozygous V617F JAK2 mutations may be present when testing is performed in
erythroid burst-forming units (BFU-Es) rather than peripheral blood granulocytes. In one report
using this technology, homozygous V617F JAK2 mutations were found in all 17 patients with PV,
both patients with ET after polycythemic transformation, and none of the 15 patients with ET alone
[82].
Exon 12 mutations In one study, four different mutations in exon 12 of the JAK2 gene were
found in 10 of 11 patients clinically diagnosed as having PV who had tested negative for the exon
14 V617F mutation [69]. The mutations were frequently present at low levels in granulocyte DNA
but were readily identifiable in clonally-derived erythropoietin-independent erythroid colonies.
When transduced into BaF3/EpoR cells, all four JAK2 exon 12 mutations caused growth factor
hypersensitivity and activated biochemical pathways associated with erythropoietin signaling.
In a second report of 114 patients fulfilling the PVSG diagnostic criteria for PV, 111 ultimately
tested positive for the exon 14 V617F mutation (97 percent); a JAK2 exon 12 mutation was
present in the remaining three subjects (3 percent), suggesting that either an exon 14 or 12 JAK2
mutation is present in virtually all patients with PV [74,83,84].
A third study reported on 106 patients with PV and exon 12 mutations, in which 17 different
mutations were identified [85]. Irrespective of the mutation, two-thirds had isolated erythrocytosis,
with the remainder having erythrocytosis plus leukocytosis and/or thrombocytosis. The
panmyelosis observed in the bone marrow of most patients with PV is typically absent in those
with exon 12 mutations. Compared with JAK2 V617F-positive PV patients, those with exon 12
mutations had similar incidences of thrombosis, myelofibrosis, leukemia, and death [73,85].
Exon 12 mutations were not found in granulocyte DNA in any of 55 patients with V617F-positive
PV, any of 25 patients with V617F-negative ET, or any of 12 patients with V617F-negative primary
myelofibrosis [69]. However, exon 12 mutations have been found in patients with
myeloproliferative disorders presenting with isolated erythrocytosis, familial PV, and, in two cases,
splanchnic vein thrombosis [86,87].
Investigational diagnostic methods Several markers are being investigated for their utility in
diagnosing PV and/or differentiating PV from secondary polycythemia (SP):
The observation that EPO-independent erythroid cell growth in PV is associated with
overexpression of Bcl-xL, an apoptosis-inhibiting oncoprotein, suggests a role for Bcl-xL
determination in diagnosis [88], and a possible role for a small-molecule Bcl-xL inhibitor as
treatment [89].
Patients with PV have reduced levels of the thrombopoietin (TPO) receptor c-mpl in platelets and
megakaryocytes, compared with those with secondary erythrocytosis as well as normal controls
[90,91]. Thus, proliferation of platelets in PV may be independent of TPO, in the same manner that
red blood cell proliferation is independent of EPO. (See "Biology and physiology of
thrombopoietin".)
Overexpression of the PRV-1 gene in peripheral blood granulocytes has been noted in PV,
essential thrombocythemia, and primary myelofibrosis, but not in patients with secondary
thrombocytosis or secondary erythrocytosis [54,92-95]. These differences were not seen when the
assay was performed on cells from the bone marrow [96].
In a prospective study, the median and range of platelet-rich plasma (PRP) serotonin
concentrations in 27 patients with PV were significantly lower than those in 22 patients with
secondary polycythemia (SP) [97]. PRP serotonin measurement performed as well as the PRV-1
assay in distinguishing PV from SP.
In one study, gene expression profiling correctly discriminated 40 patients with PV from 12
patients with secondary erythrocytosis [98].
Other diagnostic algorithms Because of the limitations of the original PVSG criteria (see
'Diagnostic criteria and algorithms' above), newer strategies for establishing the diagnosis have
been proposed [33,44,99,100]. Many of these are similar; all require the following basic criteria to
be met:
Demonstration of an increased red blood cell mass as determined either by blood volume studies
or the presence of a very much increased hemoglobin concentration or elevated hematocrit
Disorders causing secondary erythrocytosis are absent, including hypoxia, familial polycythemic
disorders, high-affinity hemoglobins, truncated EPO receptor, and inappropriate EPO production
by tumor
WHO criteria The original WHO criteria established the diagnosis of PV when, in addition to the
two basic criteria listed above, one or more of the following is also present: splenomegaly, a clonal
genetic abnormality other than the Philadelphia chromosome (or BCR-ABL fusion gene), or
spontaneous EEC [99]. If these are not present, the diagnosis can be made in the presence the
two basic criteria plus two (or more) of the following:
Platelet count >400,000/microL
White blood cell count >12,000/microL
Low serum EPO levels
Bone marrow showing panmyelosis with prominent erythroid and megakaryocytic proliferation
Revised WHO criteria A revised set of WHO criteria for the diagnosis of PV includes the
following (table 4) [100-104].
The two major criteria include an increased hemoglobin level (>18.5 g/dL in men or >16.5 g/dL in
women) or other evidence of increased red cell volume and presence of a JAK2 mutation
The three minor criteria include typical findings on the bone marrow aspiration/biopsy, a serum
EPO level below the reference range for normal, and endogenous erythroid colony formation in
vitro
The diagnosis of PV requires the presence of both major criteria and one minor criterion, or the
presence of the first major criterion together with two minor criteria.
British Committee for Standards in Hematology Criteria Criteria for the diagnosis of PV
according to the British Committee for Standards in Hematology are similar to those of the
Revised WHO criteria, with the exception that they are based upon an elevated hematocrit (>52
percent in men, >48 percent in women), rather than an elevated hemoglobin [105,106].
ESTABLISHING THE DIAGNOSIS Consensus has not yet been achieved for the optimal
diagnostic criteria for PV. Although several diagnostic approaches have been proposed since the
original PVSG studies, most have similar components. These diagnostic criteria should be applied
only to patients who have undergone the appropriate diagnostic evaluation to exclude secondary
causes of polycythemia [107]. However, the presence of the signs and symptoms described above
should raise suspicion for the presence of PV in any patient with polycythemia (table 2). (See
"Diagnostic approach to the patient with polycythemia".)
Strictly defined PV The diagnosis of polycythemia vera (PV) is considered established in
patients who fulfill the original PVSG criteria (table 3), the revised WHO criteria (table 4), or the
revised British Committee for Standards in Hematology (BCSH) criteria. (See 'British Committee
for Standards in Hematology Criteria' above.)
If the only indication of PV is an elevated hemoglobin concentration or hematocrit, this test
should be repeated. Further testing is unnecessary if the hemoglobin/hematocrit returns to normal.
Serum EPO levels should be repeated in patients with a single normal test; the second value will
be low in more than one-quarter of these patients [57].
Patients with nondiagnostic bone marrow studies and/or absence of a JAK2 mutation should
have repeat measurement of the hemoglobin or hematocrit and serum EPO concentrations in
three months. A diagnosis other than PV should be entertained in such patients.
Bled-down PV Some patients with PV may not fulfill the major PVSG, BCSH, or WHO criterion
of an increased hemoglobin/hematocrit or increased red cell mass because of a prior bleeding
episode. This has been called "bled-down" polycythemia vera if they otherwise meet the major and
minor criteria for this diagnosis.
Masked PV When blood volume measurements are either unavailable or not performed, some
patients with a simultaneous increase in red cell mass and plasma volume due to an increase in
splenic size may not have an elevated hemoglobin/hematocrit [43]. Such patients have been
termed "masked" PV if they otherwise meet the major and minor criteria for the diagnosis of PV
[46,47,108]. (See 'False negative results' above.)
DIFFERENTIAL DIAGNOSIS Diagnostic possibilities in a patient who has fulfilled either the
PVSG, revised WHO, or revised BCSH criteria for PV are vanishingly small, and are likely
confined to patients with one of the other JAK2 mutation-positive myeloproliferative neoplasms (ie,
essential thrombocythemia, primary myelofibrosis) who have an unrelated secondary cause for
polycythemia (table 1). (See 'False positive results' above.)
Distinguishing patients with masked PV from those with essential thrombocythemia may be
difficult. A comparison of hematocrit or hemoglobin values in 257 patients with WHO-defined PV,
140 patients with masked PV, and 397 patients with ET has suggested that the best cut-offs to
distinguish masked PV from ET are the following [108]:
For hemoglobin >16.5 g/dL in men and >16.0 g/dL in women
For hematocrit >49 percent in men and >48 percent in women
The differential diagnosis for a patient with polycythemia but without the other major or minor
PVSG, WHO, or BCSH criteria is discussed separately. (See "Diagnostic approach to the patient
with polycythemia".)
INFORMATION FOR PATIENTS UpToDate offers two types of patient education materials,
"The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain
language, at the 5th to 6th grade reading level, and they answer the four or five key questions a
patient might have about a given condition. These articles are best for patients who want a general
overview and who prefer short, easy-to-read materials. Beyond the Basics patient education
pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to
12th grade reading level and are best for patients who want in-depth information and are
comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or
e-mail these topics to your patients. (You can also locate patient education articles on a variety of
subjects by searching on "patient info" and the keyword(s) of interest.)
Basics topic (see "Patient information: Polycythemia vera (PV) (The Basics)")
SUMMARY AND RECOMMENDATIONS
Suspecting the diagnosis Polycythemia vera (PV) should be suspected in any patient with an
increased red blood cell mass or increased hemoglobin/hematocrit and an arterial oxygen
saturation >92 percent. Additional evidence suggesting the diagnosis of PV includes the following.
(See 'Clinical presentation' above.)
Splenomegaly
Thrombocytosis and/or leukocytosis
Thrombotic complications
Erythromelalgia or pruritus
Making the diagnosis For establishing the diagnosis of PV, the patient must satisfy a
combination of the following criteria (table 4). (See 'Establishing the diagnosis' above.)
Disorders causing secondary erythrocytosis are absent (eg, hypoxia, familial polycythemia, high-
affinity hemoglobins, truncated EPO receptor, erythropoietin-secreting tumor)
An increased hemoglobin or hematocrit or other evidence of an increased red cell mass
Presence of a JAK2 mutation (see 'JAK2 mutations' above)
Bone marrow biopsy showing hypercellularity with prominent erythroid, granulocytic, and
megakaryocytic proliferation
A low serum erythropoietin concentration
Endogenous erythroid colony formation in vitro