Determination of serum immunoreactive erythropoietin in the investigation of erythrocytosis.

We assessed the diagnostic value of determinations of serum levels of immunoreactive erythropoietin in 90 patients referred for the investigation of a raised packed red-cell volume (hematocrit) and possible erythrocytosis (a red-cell volume greater than that predicted by weight, body-surface area, or both). The mean values for erythropoietin were 16 mIU per milliliter (range, 8 to 22) in patients with polycythemia rubra vera (n = 24), 30 mIU per milliliter (range, 14 to 123) in patients with secondary erythrocytosis (n = 12), 27 mIU per milliliter (range, 13 to greater than 400) in patients with erythrocytosis of unknown origin (n = 19), and 25 mIU per milliliter (range, 18 to 35) in normal controls (n = 25). The values in the patients with polycythemia rubra vera were lower than those in the other three groups (P less than 0.01, P less than 0.05, and P less than 0.0001, respectively). In subjects with an increased packed red-cell volume but without erythrocytosis (n = 35), the levels of serum erythropoietin were the same as those in normal controls. Among patients with erythrocytosis with an unknown cause, abnormally high serum erythropoietin levels were found in 3 of 19 subjects, and in 1 of these 3 the abnormality was intermittent. Thus, measurement of serum erythropoietin in a single sample may be misleading and may not have high discriminatory value in distinguishing between polycythemia rubra vera and secondary erythrocytosis. This assay is useful in identifying patients with secondary erythrocytosis who have inappropriate erythropoietin secretion.