The measurement of venous haematocrit in patients with polycythaemia vera

Objective. In clinical practice, patients with polycythaemia vera (PV) are monitored by measurement of venous packed cell volume (PCV). However, where as treatment recommendations are still based upon studies in which the resu lts were obtained with the centrifuged microhaematocrit, currently in most instances automated blood cell counters are used to calculate PCV. In a gro up of patients with polycythaemia we therefore compared the results obtaine d by the microhaematocrit method with PCV calculated by haematology analyse rs. Design. The study was carried out on a prospective basis. Duplicate Venous blood samples were collected. The centrifuged microhaemotocrit was obtained by using an IEC Micro-MB Centrifuge. Depending on different routine method s used in the participating hospitals, the blood cell counter PCV was calcu lated using Coulter STKS, Bayer Technicon H2 or H3. Setting. Patients were included from four Swedish university hospitals: Aka demiska (Uppsala), Huddinge and Karolinska (Stockholm) and Sahlgrenska (Got eborg). Subjects. Seventy-four patients with PV and 10 patients with secondary poly cythaemia were included and a total of 150 duplicate brood samples were ana lysed from these subjects. Results. In the 150 measurements the mean blood cell counter calculated PCV was 0.448 +/- 0.037; the mean for centrifuged microhaematocrit was 0.467 /- 0.037 and the difference between means was highly significant (P = 6.8 x 10(-25)). The means for centrifuged haematocrit and calculated PCV differe d significantly in the groups of PV patients treated with phlebotomy only, hydroxyurea or radiophosphorous (P < 0.0001, respectively). In PV patients treated with alpha-interferon and in patients with secondary polycythaemia the difference in means did not reach statistical significance (P = 0.07 an d P = 0.13, respectively). The groups of patients with MCV <808 fL and grea ter than or equal to 80 fL, both presented significant differences between means for calculated PCV and centrifuged haematocrit. Conclusions, If PV patients are monitored with blood cell counter calculate d PCV it appears that the therapeutic goal should be to maintain the calcul ated PCV below 0.43, provided the local differences in calculated PCV and c entrifuged haematocrit are of the same magnitude as in this study.