1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. References

The diagnosis of an underlying chronic myeloproliferative disorder (CMPD) is often problematic in patients with primary extrahepatic portal vein obstruction (EHPVO) or Budd-Chiari syndrome (BCS); indeed, conventional clinical and hematological parameters usually yield insufficient information. To assess the diagnostic contribution of the gain-of-function mutation V617F of the JAK2 gene, 93 patients with EHPVO or BCS were investigated. JAK2 V617F was identified in 35.6% of 73 patients with EHPVO and in 40% of 20 patients with BCS. Taking the JAK2 mutation as a test with the highest positive predictive value for the diagnosis of CMPD, conventional clinical-hematological parameters had a sensitivity for CMPD lower than 48%. Bone marrow (BM) histology provided a diagnosis of CMPD in 41/74 (55.4%) patients, with a sensitivity of 93.5%. Clonality of hematopoiesis as assessed by granulocyte X-chromosome inactivation was present in 65.1% of 43 informative female patients, with a sensitivity of 86.6%. By resolving the sensitivity bias of the JAK2 mutation with the results of BM histology and clonality assay, CMPD was diagnosed in 53% of patients with EHPVO or BCS. In conclusion, CMPD is the major cause of primary EHPVO or BCS. JAK2 V617F is a very reliable and noninvasive molecular marker for CMPD and should be used as a first test for diagnosis. (HEPATOLOGY 2006;44:1528–1534.)

Extrahepatic portal vein obstruction (EHPVO) and Budd-Chiari syndrome (BCS) are relatively rare splanchnic vein thromboses (SVTs) often associated with an activation of the hemostatic system due to thrombophilic abnormalities (mainly carriership of the gain-of-function mutations factor V Leiden or A20210G prothrombin)1–3 and/or such clonal disorders of hemopoiesis as chromosome Philadelphia-negative chronic myeloproliferative disorders (CMPDs).4–6 The distinction between the two mechanisms may have important clinical implications, because anticoagulants are the most rational treatment in cases associated with thrombophilia, whereas cytoreductive therapy may be a necessary addition in CMPD. Whereas the diagnosis of thrombophilia is relatively simple and accurate, that of CMPD is often problematic in patients with EHPVO and BCS, because at the time of acute thrombosis—as well as in the postthrombotic period—hemodilution, occult bleeding, and hypersplenism due to portal hypertension may mask the changes of blood cell counts used to diagnose CMPD. Moreover, CMPD associated with SVT often presents with an atypical phenotype, making the conventional diagnostic criteria for typical CMPD elusive.6, 7 A meta-analysis of 120 patients with SVT indicated that, by using conventional criteria, the estimated prevalence of underlying CMPD was 49% for patients with BCS and 23% for patients with EHPVO.4 On the other hand, using endogenous erythroid colony formation as the only diagnostic criterion, the prevalence of CMPD was much higher (78% in BCS and 48% in EHPVO),4 but the assay lacks specificity.8, 9 Recently, Chait et al.6 used bone marrow biopsy as a diagnostic hallmark of CMPD and found that 53% of patients with BCS had the criteria for CMPD and that this proportion was 28% in those with EHPVO.6

Several molecular aberrations have been proposed as noninvasive markers for the diagnosis of Philadelphia-negative CMPD, such as clonality of hematopoiesis,10 reduced c-MPL protein expression,11 and PRV1 messenger RNA overexpression.12 Among them, only assays for clonality of hemopoiesis appear to contribute to the clinical diagnosis in CMPD13 and improve diagnostic accuracy in patients with atypical myeloproliferative disorders and thrombosis in unusual sites.14 More recently, an acquired mutation in the autoregulatory JH2 pseudokinase domain of the JAK2 gene (V617F) was detected in approximately 80% of patients with polycythemia vera, 40% of patients with essential thrombocytemia, 60% of patients with myelofibrosis with myeloid metaplasia, and in rare cases of atypical myeloid disorders.15–22 This gain-of-function mutation, which confers erythropoietin hypersensitivity and growth factor independence,15 is not detected in healthy individuals. The characteristic of high specificity for CMPD offers the opportunity to use the JAK2 V617F mutation to evaluate its contribution to the difficult diagnosis of CMPD in SVT. With this goal, we retrospectively analyzed a cohort of 93 patients with primary EHPVO or BCS well characterized for the presence of acquired or inherited thrombophilia and other risk factors for SVT. In most patients, bone marrow (BM) biopsy and an assay for clonality of hematopoiesis were used as additional diagnostic tools.

Patients and Methods

  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. References


Samples were obtained from 93 consecutive patients with EHPVO or BCS referred from 1992 to 2005 to the Gastroenterology Unit and for hematological screening to the Hemophilia and Thrombosis Center of the Maggiore Hospital Foundation of Milan. None of the patients had a prior history of CMPD. Venous blood sampling was performed in patients 1 month to 1 year after the thrombotic event (median 6 mo). DNA isolated from peripheral blood granulocytes was analyzed for the JAK2 V617F mutation and clonality of hematopoiesis. DNA samples were also obtained from 20 healthy individuals as controls. Collection and use of blood samples were approved by the institutional review board of the Foundation, and all patients gave informed consent. EHPVO was diagnosed in 73 patients and BCS was diagnosed in 20 patients using such objective methods as Doppler ultrasonography, computed tomography, magnetic resonance imaging, or venography (in most patients, more than one diagnostic method was employed). Both EHPVO and BCS had occurred in the absence of neoplastic disease and liver cirrhosis. All patients had at least a 1-year clinical follow-up. None of them developed overt cancer or other disorders occult or not diagnosed at the time of SVT occurrence. Information on the risk factors for SVT was collected, and screening for thrombophilia was performed as previously described.3 Functional and/or antigenic assays for such anticoagulant proteins as protein C and protein S were performed only in patients not on oral anticoagulant therapy (n = 68). Because anticoagulant proteins are synthesized by hepatocytes and may be low due to the impairment of liver function, the hereditary nature of the deficiencies was established only in the presence of at least one first-degree deficient relative. Overall, 36.5% of patients (38.3% with EHPVO and 30% with BCS) had at least one thrombophilia abnormality (Table 1). Sixteen patients (17.2%) had at least one condition or disease known to predispose to thrombosis (paroxysmal nocturnal hemoglobinuria, Behçet disease, inflammatory bowel disease) or recent abdominal surgery, including splenectomy (Table 2). Twelve patients (12.9%) were taking oral contraceptives, and one was on hormone replacement therapy.

Table 1. Results of Thrombophilia in Patients With SVT (All Patients), EHPVO, and BCS
ResultAll Patients (N = 93)EHPVO (n = 73)BCS (n = 20)
  • NOTE. Data are presented as the number of patients (percentage).

  • *

    Performed only in 68 patients not on oral anticoagulant therapy (EHPVO, n = 51; BCS, n = 17).

Factor V Leiden mutation (G1691A)5 (5.4)2 (2.7)3 (15)
G20210A mutation of prothrombin gene15 (16.1)15 (20.5)0
Inherited protein C deficiency*2 (2.9)2 (3.9)0
Inherited protein S deficiency*1 (1.5)1 (2)0
Antithrombin deficiency3 (3.2)3 (4.1)0
Antiphospholipid antibody syndrome10 (10.7)8 (10.9)2 (10)
Hyperhomocysteinemia14 (15)8 (10.9)6 (30)
Patients with at least one abnormality34 (36.5)28 (38.3)6 (30)
Table 2. Conditions or Diseases Predisposing to SVT (All Patients), EHPVO, and BCS
Condition/DiseaseAll Patients (N = 93)EHPVO (n = 73)BCS (n = 20)
  • NOTE. Data are presented as the number of patients (percentage).

  • *

    In the 3 months preceding the occurrence of EHPVO and BCS.

Abdominal surgery*8 (8.6)8 (10.9)0
Splenectomy5 (5.4)5 (6.8)0
Abdominal trauma*4 (4.3)2 (2.7)2 (10)
Paroxysmal nocturnal hemoglobinuria2 (2.1)02 (10)
Behçet disease1 (1.1)01 (5)
Inflammatory/infectious abdominal diseases1 (1.1)01 (5)
Patients with at least one abnormality16 (17.2)11 (15.1)5 (25)

Study Protocol.

Venous blood was obtained for a complete blood count with differential, and at the time of blood collection spleen size was measured via ultrasonography by recording spleen diameter (i.e., the measure of the longitudinal spleen axis defined as the maximal width of the organ). Of 93 patients, 75 (78.5%) had a bone marrow biopsy at the time of SVT diagnosis or no more than 1 month after. Staining techniques were Giemsa, periodic acid Shiff reagent, and the silver impregnation method following Gomori's technique.

Revision of Histological Diagnosis.

The histological diagnosis performed on BM biopsy at the time of SVT diagnosis was revised after the introduction of World Health Organization criteria.23 An ad hoc local panel of two pathologists reviewed all the available slides in an independent fashion, without prior knowledge of specific clinical data. They were requested to classify patients as having or not having CMPD. Five BM features were evaluated: BM cellularity (graded from 0 to 3),24 neutrophil granulopoiesis (normal, reduced, increased), erythropoiesis (normal, reduced, increased), fibrosis (grade 0 to 3),24 and megakaryopoiesis (quantity: normal, reduced, increased; clustering: loose clusters and dense clusters; size and nucleus morphology).

JAK2 V617F Genotyping.

DNA was extracted with standard procedures after isolation of total peripheral blood granulocytes by density gradient centrifugation over Histopaque 1077 (Sigma-Aldrich, Ayrshire, UK) and lysis of red cells using a hypotonic solution. Screening for the JAK2 mutation was initially performed by an allele-specific polymerase chain reaction in which the mutated allele was specifically amplified together with a fragment common to the mutated and wild type genes, with a detection limit for the mutated allele of 5 to 10%. Samples positive for the mutation were subsequently analyzed via polymerase chain reaction amplification and digestion with the restriction endonuclease BsaXI, which allows for an estimation of the ratio between mutated and wild-type alleles. Samples were scored as homozygous if the proportion of the mutant allele was >50%. A proportion of the mutant allele <50% may represent a heterozygous mutation, or an admixture of cells homozygous for the mutation and of cells with a wild-type JAK2 pattern, a condition defined as heterozygous/mixed clonality by Steensma et al.21 For simplicity, such samples were scored as heterozygous.

Clonality Analysis.

Clonality of hemopoiesis was analyzed by determination of the X-chromosome inactivation pattern in granulocytes from female patients. The assay is based on the evaluation of DNA methylation at the X-linked human androgen receptor and phosphoglycerate kinase loci as previously described,25 except that for human androgen receptor analysis a different set of polymerase chain reaction primers was used (sense primer 5′-AGAGGCCGCGAGCGCAGCACCT-3′, antisense primer 5′-GCTTGGGGAGAACCATCCTCACC-3′) and electrophoresis was performed on a 15% polyacrylamide gel or a 4.5% Metaphor agarose gel. Clonality was defined by an allelic cleavage ratio >3.0.25

Statistical Analysis.

The association between mutational status (heterozygous, homozygous, or wild-type) and patient characteristics was evaluated via chi-square or Fisher exact test for nominal variables. The Mann-Whitney U test was used for continuous variables. The sensitivity of BM biopsy for the diagnosis of CMPD was judged relatively in patients who were JAK2 V617F heterozygous or homozygous, taking the mutation state as a test with perfect specificity for the diagnosis of CMPD. The apparent false positive results (i.e., patients nonmutated for JAK2 but with a BM biopsy diagnostic for CMPD) were resolved by using the results of an ancillary test according to the method of discrepant analysis.26 With this method, the apparent false positive results undergo additional testing by means of clonality analysis. If the additional test yielded a positive result, then the original positive BM biopsy was considered to be a true positive result. Because of the very nature of this test, only female patients (n = 55) were analyzed for clonality. Results were considered statistically significant when P values were <.05. All computations were performed with STATISTICA software (Statsoft, Tulsa, OK).


  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. References

Patient Characteristics.

The hematological and clinical characteristics of the study cohort at the time of diagnosis of EHPVO or BCS are summarized in Table 3. A hematocrit level higher than 45% was present in 9.5% of patients, but no patient had a hematocrit level higher than 50%. The platelet count was higher than 400 × 109/L and the white blood cell (WBC) count was higher than 10 × 109/L in 26.6% and 36.1% of patients. A palpable spleen and a spleen diameter ≥20 cm was measured in 68.3% and 15.5% of patients. Only 3 patients had marked splenomegaly (diameter ≥20 cm) and a platelet count greater than 200 × 109/L, a combination suggested by Chait et al.6 to be an index with high specificity for CMPD. Five patients had been splenectomized before the occurrence of EHPVO, for reasons given in Table 3.

Table 3. Clinical Characteristics of the 93 Patients With SVT (All Patients), EHPVO, and BCS at the Time of Sample Collection for JAK2 V617F Testing
CharacteristicsAll patients (N = 93)EHPVO (n = 73)BCS (n = 20)
  • *

    Splenectomy was performed before EHPVO in 5 patients because of splenic infarction, Evans syndrome, idiopathic thrombocytopenic purpura, thalassemia intermedia, and hereditary spherocytosis.

Median age, yr (range)38 (13–72)42 (13–66)33 (19–72)
Male, no. (%)37 (39.8)29 (39.7)8 (40)
Median hemoglobin level, g/L (range)13 (4.3–16)13.2 (4.3–16)12.1 (8–14.7)
Median WBC count, ×109/L (range)8.15 (2.7–21.9)8.1 (3.4–21.9)8.9 (2.7–17.2)
Median platelet count, ×109/L (range)255 (37–1400)263 (40–1400)253 (37–570)
Median spleen diameter, cm (range)16 (11–25)16 (11–24)15.5 (11–25)
Splenectomy, no.*550

JAK2 V617F Mutation.

No control DNA sample was positive for the mutation. The mutation was detected in 34 of 93 patients with SVT (36.5%), 26 of whom (76.5%) were heterozygous and 8 of whom (23.5%) were homozygous for the mutant allele. In all samples scored as homozygous, the mutated allele represented more than 70% of the polymerase chain reaction product. The mutation was identified in 26 of 73 (35.6%) patients with EHPVO and in 8 of 20 (40%) patients with BCS (P value not significant). There was no significant difference between patients with and without the mutation for age and sex (data not shown).

Patients carrying one or two mutated alleles differed from nonmutated patients for higher mean WBC count (10.3 × 109/L vs. 8.1 × 109/L; P = .02), higher platelet count (386 × 109/L vs. 278 × 109/L; P < .05), and larger spleen size (18.2 cm vs. 14.6 cm; P < .01). Patients with the heterozygous mutation were not different for age, WBC count, platelet count, and spleen diameter from nonmutated patients. Patients with the homozygous mutation had significantly higher WBC count (13.3 × 109/L vs. 8.1 × 109/L; P = .002) and spleen diameter (21.7 cm vs. 15.6 cm; P = .01) than nonmutated patients. Homozygotes had larger spleen diameter than heterozygotes (21.7 cm vs. 16.8 cm; P = .01).

The proportion of patients with at least one abnormality on thrombophilia screening was similar in JAK2-mutated and nonmutated patients (11/34 [32.3%] vs. 23/59 [38.9%]; P value not significant). The proportion of those with a disease or condition predisposing to SVT was lower in JAK2 mutated than in nonmutated patients (3/34 [8.8%] vs. 13/59 [22.0%]), but the difference was not statistically significant. The proportion of patients positive on thrombophilia screening or with a disease or condition predisposing to SVT was 13 of 34 (38.2%) in JAK2-mutated patients and 31 of 59 (52.5%) in nonmutated patients (P value not significant).

Clonality Pattern.

Fifty-five women with EHPVO or BCS were tested for heterozygosity at the human androgen receptor and phosphoglycerate kinase locus, 43 of whom (78.2%) were informative. Peripheral blood neutrophils expressed a monoclonal X-chromosome inactivation pattern in 28 of these women (65.1%). The frequency of clonal hematopoiesis was 24 of 36 (66.6%) in women with EHPVO and 4 of 7 (57.1%) in those with BCS (P value not significant). Analysis of the association between the occurrence of clonal hematopoiesis and clinical and laboratory characteristics showed no significant difference between patients with and without clonal hematopoiesis for age, hematocrit level, WBC count, platelet count, or spleen diameter. The proportion of patients who had at least one abnormality on thrombophilia screening was similar in the clonal and nonclonal groups (10/28 [35.7%] vs. 6/15 [40%]; P value not significant). The proportion of patients with a disease or condition predisposing to SVT was also similar (3/28 [12%] vs. 4/15 [26.6%]).

The high positive predictive power for CMPD of the heterozygous or homozygous JAK2 V617F mutation allowed us to calculate the sensitivity (true positive frequency) of the clonality assay in informative women. Two of 15 patients with the JAK2 mutation tested negative on the clonality assay, with a sensitivity of 86.6% (Table 4). The diagnostic sensitivity with respect to JAK2 mutation for a hematocrit level greater than 45%, WBC greater than 10 × 109/L, platelet count greater than 400 × 109/L, and spleen diameter greater than 20 cm were 10%, 47.8%, 46.1%, and 25%, respectively (Table 4).

Table 4. Sensitivity of Conventional Clinical-Hematological Diagnostic Criteria for SVT-Associated CMPD and Sensitivity and Specificity of BM Biopsy
ParameterJAK2 V617F Mutation-Based Estimate of SensitivityDiscrepant Analysis (JAK2 Mutation Corrected by Clonality Analysis)-Based Estimate of Specificity
Hematocrit level ≥45%10%
Leukocyte count ≥10 × 109/L47.8%
Platelet count ≥400 × 109/L46.1%
Spleen diameter ≥20 cm25%
Clonality of hematopoiesis86.6%
BM biopsy93.5%91.6%

BM Biopsy.

Due to the retrospective nature of the study, some BM biopsies were no longer available for re-evaluation according to World Health Organization criteria. Hence, BM biopsy was available only in 74 patients with EHPVO or BCS. On the basis of these criteria, CMPD was diagnosed in 41 patients (55.4%), 56.3% with EHPVO and 52.6% with BCS. The presence of megakaryocyte hyperplasia was the feature that characterized CMPD in these disorders (Table 5). Clustering of megakaryocytes (including loose and dense clusters) was a diagnostic hallmark present in 42.1% of cases, associated with a histological diagnosis of CMPD in 83%. BM hypercellularity and fibrosis were additional features frequently present in CMPD.

Table 5. Histological Characteristics of BM Biopsy in 74 Patients With SVT Revised According to World Health Organization Criteria
DiagnosisIncreased CellularityIncreased GranulopoiesisIncreased ErythropoiesisIncreased MegakaryocytopoiesisClusters of Megakaryocytes*Fibrosis
  • *

    Including loose and dense clusters.

CMPD (n = 41)94.4%88.8%83.3%100%83.3%33.3%
No CMPD (n = 33)14.3%14.3%35%15.%5%0%

Patients with a BM biopsy diagnostic for CMPD had higher WBC counts (10.1 × 109/L vs. 7.96 × 109/L; P = .03) and platelet counts (403 × 109/L vs. 254 × 109/L; P = .008) and a higher frequency of splenomegaly (91% vs 52%; P = .001). The proportion of patients with at least one abnormality on thrombophilia screening was lower in those with a BM biopsy diagnostic of CMPD (13/41 [31.7%]) than in those without this diagnosis (15/33 [45.4%]), but the difference was not statistically significant. Thus, 60.8% of patients without thrombophilia had a diagnosis of CMPD on BM biopsy, while this proportion was 46.6% in patients with at least one abnormality (P value not significant). By contrast, the proportion of patients with a disease or condition predisposing to SVT was lower (P = .009) in patients with a histological diagnosis of CMPD (2/41 [4.9%]) than in those without (9/33 [27.3%]). Thus, 61.9% of patients without a disease or condition predisposing to SVT were positive for CMPD on BM biopsy, while only 18.1% of patients with a predisposing condition were positive (P = .008). Patients with a BM biopsy diagnostic for CMPD had a prevalence of JAK2 V617F mutation of 70.7% (29/41), significantly higher (P < .001) than in those with a BM biopsy not diagnostic for CMPD (2/33 [6%]).

Diagnostic Performance of BM Biopsy for the Diagnosis of CMPD.

Of 31 patients carrying the JAK2 mutation and in whom a revision of the BM biopsy was available, only 2 were not diagnosed with CMPD, yielding a sensitivity for BM biopsy of 93.5%. By using the JAK2 mutation as the reference diagnostic test, the specificity of BM biopsy for CMPD was 72%. However, the JAK2 V617F mutation does not carry a high negative predictive value for CMPD. To calculate a more reliable specificity value, apparent false positive cases (i.e., patients who had a negative JAK2 mutation and a positive BM biopsy) underwent discrepant analysis using clonality of hematopoiesis as an ancillary test. Three of 4 apparent false positive cases received a positive result on clonality analysis; consequently, the original JAK2-negative result was considered a false negative result. With this analysis, the adjusted estimate of specificity of BM biopsy was 91.6% (11/12) (Table 4).

By recalculating the frequency of positive BM biopsies after adjustment for the rate of false negative results (as obtained by the sensitivity value [6.5%]) and false positive results (as obtained by the specificity value corrected by discrepant analysis [8.4%]), the estimated prevalence of CMPD in SVT via BM biopsy was 53.3%.


  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. References

The finding of the gain-of-function JAK2 V617F mutation in patients with CMPDs has raised questions regarding its potential influence on the diagnostic algorithm of these disorders.27 These issues are particularly relevant and critical in patients with SVT, because in them a mechanistic role for a masked or atypical CMPD in producing activation of the hemostastic system has been surmised for a long time. In this study, the question of the diagnostic contribution of the JAK2 mutation was tackled in a cohort of 93 consecutive patients with EHPVO or BCS who were fully characterized at the time of thrombosis for the presence of thrombophilic abnormalities and other surgical, infectious, or inflammatory risk factors. The available original BM biopsies were reviewed according to the recently established World Health Organization criteria,23 and the diagnosis was revised where appropriate.

JAK2 V617F was identified in 35.6% of patients with EHPVO and in 40% of those with BCS, confirming recent reports of BCS.28 In the entire population of patients with SVT, the frequency of the JAK2 V617F mutation was 36.5%. The mutation is a highly specific diagnostic marker of CMPD, because it is never detected in healthy individuals, and very few cases of acute leukemia or atypical myeloproliferative disorders have tested positive.21, 22

Starting from this assumption, we determined the sensitivity (true positive rate) of the current reference diagnostic method for SVT-associated CMPD (i.e., the BM biopsy).6 CMPD was diagnosed via BM histology in more than 93% of patients with a heterozygous or homozygous mutation of the JAK2 gene. However, the JAK2 mutation is not a diagnostic gold standard for CMPD, because the mutation may not be present in as many as 20% to 60% of patients with CMPD.15–20 To reconcile the different prevalences of CMPD diagnosis found in this study, we performed discrepant analysis to overcome the sensitivity bias of JAK2. The apparent false positive results of BM biopsies (i.e., those with a diagnosis of CMPD in patients who tested JAK2 negative) were corrected for their positivity for clonality of hemopoiesis. After this reclassification, the specificity of BM biopsies for CMPD was 91.6%. By recalculating the prevalence of positive BM biopsies after adjustment for JAK2-derived sensitivity and discrepant analysis–derived specificity, the estimated prevalence of CMPD in SVT was 53.3%. Even though discrepant analysis has been criticized for conceptual and logic problems,29 it has become a standard method for estimating the sensitivity and specificity of many diagnostic tests.26

In this series of cases, neither a diagnosis of CMPD through BM biopsy nor the result of a positive JAK2 mutation or clonality of hematopoiesis correlated with the presence of a thrombotic risk due to at least one abnormality on thrombophilia screening. In patients who tested negative for thrombophilia markers, the incidence of CMPD at BM biopsy was only slightly higher than in those who tested positive (60.8% vs. 46.4%; P value not significant). This observation strengthens the views that inherited or acquired thrombophilia plays a triggering role in SVT.1 A more mechanistic role was documented for abdominal surgery, inflammatory diseases, paroxysmal nocturnal hemoglobinuria, or Behçet disease, all of which predispose to thrombotic events. The prevalence of CMPD was 18.1% in patients with these conditions, whereas it was 61.9% in those without (P = .008). These results led us to speculate that CMPD is the major cause of EHPVO or BCS in patients in whom thrombosis occurs idiopathically.

An important achievement of this study is that it allowed delineation of the most convenient algorithm for SVT-associated CMPD. The study confirms that conventional hematological criteria are very insensitive to diagnose CMPD. Neither polycythemia (hematocrit level higher than 45%) nor thrombocytosis (platelet count greater than 600 × 109/L) allowed to suspect myeloproliferation in the vast majority of cases. Only 6 patients, for instance, had more than 600 × 109/L platelet count, confirming the atypical nature of SVT-associated CMPD. BM biopsy was the single most useful technique for the diagnosis of SVT-associated CMPD; however, it is an invasive procedure that is difficult to standardize, and it requires an experienced clinical hematopathologist to distinguish a myeloproliferative disorder from a reactive BM picture. The finding that in SVT patients with a BM biopsy diagnostic for CMPD the prevalence of the JAK2 V617F mutation is more than 70% allows us to conclude that JAK2 V617F is a very reliable noninvasive molecular marker for the disease and should be viewed as the first test for the diagnosis of SVT-associated CMPD. This rule is also valid for SVT associated with genetic thrombophilia or thrombophilia due to systemic diseases, because such conditions do not exclude the coexistence of a CMPD. Because of the low negative predictive value of JAK2 mutation, in the absence of V617F a further search of a CMPD is mandatory. The results of this study and previous reports6, 13, 14 identify BM biopsy and clonality of hematopoiesis (only in females) as the tests to be sequentially performed. With all these tests negative, a SVT-associated CMPD is very unlikely. On the other hand, although the JAK2 mutation is diagnostic for the presence of a CMPD, BM biopsy must also be performed in JAK2-positive patients to assist in the diagnosis of the underlying CMPDs that may have different prognoses.

In conclusion, these improvements in the accuracy of the diagnosis of CMPD, as well as the results stemming from thrombophilia testing, should help to design randomized controlled trials based on cytoreductive or antithrombotic regimens in SVT, because currently there is no evidence-based therapeutic approach for the secondary prophylaxis of these conditions.


  1. Top of page
  2. Abstract
  3. Patients and Methods
  4. Results
  5. Discussion
  6. References
  • 1
    Denninger MH, Chait Y, Casedevall N, Hillaire S, Guillin MC, Bezeaud A, et al. Cause of portal or hepatic venous thrombosis in adults: the role of multiple concurrent factors. HEPATOLOGY 2000; 31: 587591.
  • 2
    Amitrano L, Guardascione MA, Ames PR, Margaglione M, Antinolfi I, Iannaccone L, et al. Thrombophilic genotypes, natural anticoagulants, and plasma homocysteine in myeloproliferative disorders: relationship with splanchnic vein thrombosis and arterial disease. Am J Hematol 2003; 72: 7581.
  • 3
    Primignani M, Martinelli I, Bucciarelli P, Battaglioli T, Reati R, Fabris F, et al. Risk factors for thrombophilia in extrahepatic portal vein obstruction. HEPATOLOGY 2005; 41: 603608.
  • 4
    De Stefano V, Teofili L, Leone G, Michiels JJ. Spontaneous erythroid colony formation as the clue to an underlying myeloproliferative disorder in patients with Budd-Chiari syndrome or portal vein thrombosis. Semin Thromb Hemost 1997; 23: 411418.
  • 5
    Gruppo Italiano Studio Policitemia. Polycythemia vera: the natural history of 1213 patients followed for 20 years. Ann Intern Med 1995; 123: 656664.
  • 6
    Chait Y, Condat B, Cazals-Hatem D, Rufat P, Atmani S, Chaoui D, et al. Relevance of the criteria commonly used to diagnose myeloproliferative disorder in patients with splanchnic vein thrombosis. Br J Haematol 2005; 129: 553560.
  • 7
    Barosi G, Buratti A, Costa A, Liberato LN, Balduini C, Cazzola M, et al. An atypical myeloproliferative disorder with high thrombotic risk and slow disease progression. Cancer 1991; 68: 23102318.
  • 8
    Dudley JM, Westwood N, Leonard S, Eridani S, Pearson TC. Primary polycythemia: positive diagnosis using the differential response of primitive and mature erithroid progenitors to erythropoietin, interleukin3 and alfa-interferon. Br J Hematol 1990; 75: 188194.
  • 9
    Masters GS, Baines P, Jacobs A. Erythroid colony growth from peripheral blood and bone marrow in polycythaemia. J Clin Pathol 1990; 43: 937941.
  • 10
    Liu E, Jelinek J, Pastore YD, Guan Y, Prchal JF, Prchal JT. Discrimination of polycythemias and thrombocytoses by novel, simple, accurate clonality assays and comparison with PRV-1 expression and BFU-E response to erythropoetin. Blood 2003; 101; 32943301.
  • 11
    Moliterno AR, Hankins WD, Spivak JL. Impaired expression of the thrombopoietin receptor by platelets from patients with polycythemia vera. N Engl J Med 1998; 338: 572580.
  • 12
    Klippel S, Strunck E, Temerinac S, Bench AJ, Meinhardt G, Mohr U, et al. Quantification of PRV-1 mRNA distinguishes polycythemia vera from secondary erythrocytosis. Blood 2003; 102: 35693574.
  • 13
    Kralovics R, Buser AS, Teo SS, Coers J, Tichelli A, van der Maas AP, et al. Comparison of molecular markers in a cohort of patients with chronic myeloproliferative disorders. Blood 2003; 102: 18691971.
  • 14
    Chiusolo P, La Barbera EO, Laurenti L, Piccirillo N, Sora F, Giordano G, et al. Clonal hemopoiesis and risk of thrombosis in young female patients with essential thrombocythemia. Exp Hematol 2001; 29: 670676.
  • 15
    James C, Ugo V, Le Couedic JP, Staerk J, Delhommeau F, Lacout C, et al. A unique clonal JAK2 mutation leading to constitutive signalling causes polycythaemia vera. Nature 2005; 434: 11441148.
  • 16
    Kralovics R, Passamonti F, Buser AS, Teo SS, Tiedt R, Passweg JR, et al. A gain-of-function mutation of JAK2 in myeloproliferative disorders. N Engl J Med 2005; 352: 17791790.
  • 17
    Baxter EJ, Scott LM, Campbell PJ, East C, Fourouclas N, Swanton S, et al. Cancer Genome Project. Acquired mutation of the tyrosine kinase JAK2 in human myeloproliferative disorders. Lancet 2005; 365: 10541061.
  • 18
    Levine RL, Wadleigh M, Cools J, Ebert BL, Wernig G, Huntly BJ, et al. Activating mutation in the tyrosine kinase JAK2 in polycythemia vera, essential thrombocythemia, and myeloid metaplasia with myelofibrosis. Cancer Cell 2005; 7: 387397.
  • 19
    Zhao R, Xing S, Li Z, Fu X, Li Q, Krantz SB, et al. Identification of an acquired JAK2 mutation in polycythemia vera. J Biol Chem 2005; 280: 2278822792.
  • 20
    Jones AV, Kreil S, Zoi K, Waghorn K, Curtis C, Zhang L, et al. Widespread occurrence of the JAK2 V617F mutation in chronic myeloproliferative disorders. Blood 2005; 106: 21622168.
  • 21
    Steensma DP, Dewald GW, Lasho TL, Powell HL, McClure RF, Levine RL, et al. The JAK2 V617F activating tyrosine kinase mutation is an infrequent event in both “atypical” myeloproliferative disorders and myelodysplastic syndromes. Blood 2005; 106: 12071209.
  • 22
    Jelinek J, Oki Y, Gharibyan V, Bueso-Ramos C, Prchal JT, Verstovsek S, et al. JAK2 mutation 1849G>T is rare in acute leukemias but can be found in CMML, Philadelphia chromosome-negative CML, and megakaryocytic leukemia. Blood 2005; 106: 33703373.
  • 23
    Vardiman JW, Harris NL, Brunning RD. The World Health Organization (WHO) classification of the myeloid neoplasms. Blood 2002; 100: 22922302.
  • 24
    Thiele J, Kvasnicka HM, Facchetti F, Franco V, van der Walt J, Orazi A. European consensus on grading bone marrow fibrosis and assessment of cellularity. Haematologica 2005; 90: 11281132.
  • 25
    Tonon L, Bergamaschi G, Dellavecchia C, Rosti V, Lucotti C, Malabarba L, et al. Unbalanced X-chromosome inactivation in haemopoietic cells from normal women. Br J Haematol 1998; 102: 9961003.
  • 26
    Hadgu A, Dendukuri N, Hilden J. Evaluation of nucleic acid amplification tests in the absence of a perfect gold-standard test: a review of the statistical and epidemiologic issues. Epidemiology 2005: 16: 604612.
  • 27
    Tefferi A, Gilliland DG. The JAK2V617F tyrosine kinase mutation in myeloproliferative disorders: status report and immediate implications for disease classification and diagnosis. Mayo Clin Proc 2005; 80: 947958.
  • 28
    Patel RK, Lea NC, Heneghan MA, Westwood NB, Milojkovic D, Thanigaikumar M, et al. Prevalence of the activating JAK2 tyrosine kinase mutation V617F in the Budd-Chiari syndrome. Gastroenterology 2006; 130: 20312038.
  • 29
    Miller WC. Bias in discrepant analysis: when two wrongs don't make it a right. J Clin Epidemiol 1998; 51: 219231.