Portal de Periódicos da CAPES

Diagnosing Polycythemia Vera: A Paradigm Shift

1999; Elsevier BV; Volume: 74; Issue: 2 Linguagem: Inglês

10.4065/74.2.159

1942-5546

Ayalew Tefferi,

Erythrocyte Function and Pathophysiology

The primary objective during the evaluation of erythrocytosis is to ascertain the presence or absence of polycythemia vera (PV). Because of prognostic and treatment differences, PV must be distinguished from relative polycythemia and secondary erythrocytosis. This distinction is currently accomplished through the laboratory measurement of red blood cell mass, plasma volume, and arterial oxygen saturation and determination of oxygen pressure at 50% hemoglobin saturation (P50). Furthermore, according to the Polycythemia Vera Study Group guidelines, the demonstration of an increased red blood cell mass is an absolute criterion for the diagnosis of PV. This article discusses the use of the serum erythropoietin level and endogenous erythroid colony assay as a potential alternative in the diagnosis of PV. The primary objective during the evaluation of erythrocytosis is to ascertain the presence or absence of polycythemia vera (PV). Because of prognostic and treatment differences, PV must be distinguished from relative polycythemia and secondary erythrocytosis. This distinction is currently accomplished through the laboratory measurement of red blood cell mass, plasma volume, and arterial oxygen saturation and determination of oxygen pressure at 50% hemoglobin saturation (P50). Furthermore, according to the Polycythemia Vera Study Group guidelines, the demonstration of an increased red blood cell mass is an absolute criterion for the diagnosis of PV. This article discusses the use of the serum erythropoietin level and endogenous erythroid colony assay as a potential alternative in the diagnosis of PV. An increased hematocrit or hemoglobin level suggests polycythemia. These measurements, however, are not informative of the total red blood cell mass (ReM) and simply indicate the ratio of red blood cells to plasma volume. The total RCM can be indirectly estimated by measuring total blood volume and multiplying the result by the hematocrit level.1Fairbanks VF Klee GG Wiseman GA Hoyer JD Tefferi A Petitt RM et al.Measurement of blood volume and red cell mass: re-examination of 5 lCr and 1251 methods.Blood Cells Mol Dis. 1996; 22: 169-186Crossref PubMed Scopus (70) Google Scholar An isolated plasma volume depletion does not affect the RCM but may significantly increase the hematocrit or hemoglobin level (relative polycythemia). Conversely, a true increase in ReM, if accompanied by a proportional increase in plasma volume, may be overlooked because of a normal hematocrit or hemoglobin level (inapparent polycythemia).2Lamy T Devlllers A Bernard M Molsan A Grulois I Drenou B et al.Inapparent polycythemia vera; an unrecognized diagnosis.Am J Med. 1997; 102: 14-20Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar A true increase in RCM is a clinically relevant finding that necessitates further investigation. Moreover, an increase in RCM due to polycythemia vera (PV) is more serious (thrombogenic) than an equivalent erythropoietin (EPO)-mediated increase in normal red blood cells (secondaryerythrocytosis). Therefore, the primary objective in evaluating erythrocytosis is to ascertain the presence or absence of PV. Conventionally, the distinction among PV secondary erythrocytosis, relative polycythemia, and inapparent polycythemia has involved the laboratory measurement of RCM and plasma volume. These measurements have also been incorporated in the PV diagnostic criteria established by the Polycythemia Vera Study Group almost 30 years ago.3Berlin Nl Diagnosis and classification of the polycythemias.Semin Hematol. 1975; 12: 339-351PubMed Google Scholar During the past several years, it has become evident that the RCM measurement may not always be convenient, reliable, or necessary in the assessment of a patient with suspected polycythemia. First, the procedure is expensive and necessitates strictly controlled laboratory conditions. Second, patients with true PV who have either early disease or concurrent iron deficiency may not fulfill the absolute Polycythemia Vera Study Group criterion of an increased RCM. Third, measuring the RCM is a costly, redundant procedure when the hemoglobin level is increased higher than 18.5 g/dL in men and 16.5 g/dL in women. This degree of hemoglobin increase is almost always associated with an increased RCM.1Fairbanks VF Klee GG Wiseman GA Hoyer JD Tefferi A Petitt RM et al.Measurement of blood volume and red cell mass: re-examination of 5 lCr and 1251 methods.Blood Cells Mol Dis. 1996; 22: 169-186Crossref PubMed Scopus (70) Google Scholar, 4Pearson TC Apparent polycythaemia.Blood Rev. 1991; 5: 205-213Abstract Full Text PDF PubMed Scopus (43) Google Scholar One might argue that the RCM measurement with lower hemoglobin values (less than 18.5 g/dL in men and 16.5 g/dL in women) may distinguish “true erythrocytosis” from “relative polycythemia” and also unmask inapparent PV. In regard to the former issue, some studies have questioned the existence of “relative polycythemia” as a separate nosologic entity.5Brown SM Gilbert HS Krauss S Wasserman LR Spurious (relative) polycythemia: a nonexistent disease.Am J Med. 1971; 50: 200-207Abstract Full Text PDF PubMed Scopus (32) Google Scholar In a recent study of 109 consecutive RCM and plasma volume measurements obtained at Mayo Clinic Rochester, not a single case of “relative polycythemia” was encountered. Similarly, other studies have suggested that abnormally low body-venous hematocrit ratios, not reduced plasma volume, accounted for the “relative polycythemia” observed in patients with hypertension (so-called Gaisböck's syndrome).6Watts El Lewis SM Spurious polycythaemia—a study of 35 patients.Scand J Haematol. 1983; 31: 241-247Crossref PubMed Scopus (15) Google Scholar In regard to patients with “inapparent polycythemia,” the situation does not differ from that in patients with true PV demonstrating “normal” RCM values because of associated bleeding, iron deficiency, or occult or early disease. Therefore, the presence of a PV-related feature (Table 1) dictates the subsequent measurement of more specific biologic markers of PV regardless of the calculated RCM. In contrast, overreliance on RCM measurement may result in incorrectly labeling part of the normal population as having PV because of the statistically “narrow” limits of the reference interval.Table 1Polycythemia Vera-Related Clinical and Laboratory Features Persistent leukocytosisPersistent thrombocytosisMicrocytosisSplenomegalyGeneralized pruritus (after bathing)Unusual thrombosisErythromelalgia (acral dysesthesia and erythema) Open table in a new tab As previously mentioned, the primary reason for evaluating “erythrocytosis” is to ascertain the presence or absence of PV. PV differs from all other causes of erthrocytosis in its pathogenesis and therefore influences, in a unique fashion, the production of the erythroid growth factor (EPO). PV is a neoplastic (clonal) blood disorder with autonomous (EPO-independent) erythroid proliferation.7Hlnshelwood S Bench AJ Green AR Pathogenesis of polycythaemia vera.Blood Rev. 1997; 11: 224-232Abstract Full Text PDF PubMed Scopus (23) Google Scholar Because of the existence of a negative feedback mechanism, the erthrocytosis in PV down-regulates EPO production and results in below normal serum EPO levels. In contrast, an EPO-driven erthrocytosis characterizes secondary erythrocytosis, and this condition may be associated with either high or normal serum EPO levels. Of note, the physiologic effect of an increased hematocrit in hypoxia-related secondary erthrocytosis is to lower the initially increased serum EPO level to within the normal range. Similarly, because of the clonal, growth factor-independent nature of PV, endogenous (without the addition of EPO) erythroid colony (EEC) growth is a relatively specific manifestation associated with PV. Many investigators have repeatedly demonstrated the specificity of EEC growth to myeloproliferative disorders, especially to PV.8Weinberg RS In vitro erythropoiesis in polycythemia vera and other myeloproliferative disorders.Semin Hematol. 1997; 34: 64-69PubMed Google Scholar EEC assays, however, are not available in many hematology laboratories, and a high level of expertise is needed to perform this test. Regardless, determination of serum EPO levels and EEC growth pattern provides an alternative approach to the diagnosis of PV. The first step in the diagnostic work-up of the patient with erthrocytosis is to determine whether the observed hemoglobin level is truly abnormal. The particular value should be evaluated in reference to the appropriate sex- and race-adjusted “normal” range. Of importance, the stated “normal range” usually refers to the lower and upper limits observed in 95% of the normal population and does not necessarily imply that outlying values are always abnormal. Therefore, the initial evaluation should begin with review of previous laboratory test results that may or may not demonstrate an interim change. Obtaining an EPO level is reasonable if (1) the hemoglobin level is greater than 18.5 g/dL in men and 16.5 g/dL in women, (2) there is a documented interim change in the hemoglobin, and (3) a borderline increase in the hemoglobin is associated with a PV-related feature (Table 1). As previously mentioned, low serum EPO levels are relatively specific to PV, and above normal serum EPO levels are extremely unusual in PV.9Birgegard G Wide L Serum erythropoietin in the diagnosis of polycythaemia and after phlebotomy treatment.Br J Haematol. 1992; 81: 603-606Crossref PubMed Scopus (66) Google Scholar, 10Najean Y Schlageter MH Toubert ME Podgomlak MP Radioimmunoassay of immunoreactive erythropoietin as a clinical tool for the classification of polycythaemias.Nouv Rev Fr Hematol. 1990; 32: 237-240PubMed Google Scholar The serum EPO level, however, may be in the low to normal range in some patients with PV. Therefore, PV is a diagnostic possibility only when the serum EPO level is low or normal. A diagnosis of PV is extremely likely after confirmation of a low serum EPO level (Fig. 1). Even in patients undergoing active venesection, the serum EPO level is usually low in those with PV.9Birgegard G Wide L Serum erythropoietin in the diagnosis of polycythaemia and after phlebotomy treatment.Br J Haematol. 1992; 81: 603-606Crossref PubMed Scopus (66) Google Scholar A repeated serum EPO determination may be unnecessary if other clinical and laboratory findings are supportive of the diagnosis. In the absence of these supportive factors, however, the observed low value should be validated by a repeated test. A subsequent bone marrow examination with cytogenetic studies is indicated in all patients with a working diagnosis of PV. The bone marrow examination helps in confirming the diagnosis and also provides prognostically relevant information including the presence or absence of associated bone marrow fibrosis or a clonal cytogenetic abnormality (or both). If the diagnosis of PV is still uncertain, an EEC assay can be performed for further confirmation. Rarely, an EPO-receptor mutant may result in congenital pure erthrocytosis with a low serum EPO level. This uncommon disorder is associated with normal spleen size, normal leukocyte and platelet counts, and often a family history of the disorder. These features should provide adequate information to distinguish this familial disorder from PV. A normal serum EPO level does not absolutely exclude the possibility of PV. Investigative decisions regarding “suspected PV” with a normal serum EPO level depend primarily on the presence or absence of PV-related features (Fig. 1). In the presence of any PV-related feature (Table 1), proceeding with a bone marrow examination with cytogenetic studies is a reasonable approach. If the results are not consistent with PV, an EEC study may provide further insight. In the absence of a PV-related feature, a hemoglobin level of 18.5 g/dL or higher in men and 16.5 g/dL or higher in women and a normal serum EPO level could be consistent with either PV or secondary erythrocytosis.10Najean Y Schlageter MH Toubert ME Podgomlak MP Radioimmunoassay of immunoreactive erythropoietin as a clinical tool for the classification of polycythaemias.Nouv Rev Fr Hematol. 1990; 32: 237-240PubMed Google Scholar An initial work-up for secondary erythrocytosis is recommended, and if findings are normal, an EEC study can be done. In the absence of any PV-related feature, a normal serum EPO level associated with a hemoglobin level of less than 18.5 g/dL in men and 16.5 g/dL in women can be followed by a repeated hemoglobin determination periodically. The algorithm (Fig. 1) and the aforementioned discussion are provided as suggestions for an alternative approach to the evaluation of erythrocytosis. The discussion is based on the assumption that an accurate assay for measurement of EPO is available to the clinician. Often, additional supporting clinical characteristics are present to aid in the diagnostic process. In certain instances, the findings are equivocal, and additional studies including measurement of RCM are appropriate. Of note, however, measurement of RCM does not facilitate the distinction between PV and secondary erythrocytosis. My colleagues and I advise against overreliance on the serum EPO level, and we emphasize the primary importance of the clinical circumstances, including family history, age at onset of disease, clinical manifestations, presence or absence of other hematologic abnormalities, splenomegaly, thrombohemorrhagic manifestations, vasomotor symptoms (including erythromelalgia), or pruritus after bathing. The diagnosis of PV is not automatically confirmed or discarded on the basis of the serum EPO level alone. Of importance, laboratory inaccuracies in the measurement of serum EPO still persist, and clinicians are encouraged to use research laboratories for validation of discrepant values. Similarly, EEC assays are currently not widely available, and interpretation of the results should be done carefully. Regardless of these limitations, strict adherence to the original criteria of the Polycythemia Vera Study Group should no longer be considered a prerequisite to the diagnosis of PV.