[Evaluation of the validity of two hemoglobinometry methods for measurement of hemoglobin levels in whole blood donors in Yazd, 2010-2011]

ABSTRACT

Hemoglobin screening methods before blood donation need to be not only simple, rapid and inexpensive but also sensitive for detection of both low and high hemoglobin levels to prevent false rejection of potential donors. The aim of this study was to evaluate the validity of two methods of hemoglobinometry in blood donors. In this cross sectional study, 198 blood donors were randomly selected. The hemoglobin levels of two blood samples taken by means of a single finger stick were determined by Hemocue 201+ and Hemocontrol and hemoglobin level of a venous blood sample was measured by Cobas hematology analyzer as standard method. Venous hemoglobin range of 12.5-17.9 g/dl was regarded as inclusion criteria for blood donation. The sensitivity, specificity and correlation between reference method and each of the hemoglobinometers were assessed. Data analysis was performed by means of correlation tests, regression analysis and paired t-test. Mean hemoglobin levels measured by Cobas, Hemocue and Hemocontrol were 15.9 +/-1.68, 16.28 +/-1.91 and 16.30 +/-196 g/dl, respectively. Hemoglobin levels measured by hemoglobinometers were higher than that measured by the standard reference. This difference was not significant in low hemoglobin levels, contrary to high hemoglobin levels. Each of the two methods showed significant correlation with standard method when hemoglobin levels were lower than normal limit. Increased hemoglobin levels led to decreased correlation coefficient which was not significant with hemoglobin levels higher than 18g/dl. Our study results showed that these heomoglobinometers were not valid for screening high hemoglobin levels in blood donors and a large number of potential donors would be falsely rejected. Further studies to find valid methods are recommended