Three-way comparison of methods for the measurement of the erythrocyte sedimentation rate

Abstract

An erythrocyte sedimentation rate (ESR) is a nonspecific sickness index, which is not diagnostic of any particular disease, but which when elevated may indicate the presence of inflammation, infection, rheumatologic disease, or neoplasm. Technological advances continue to evolve to make this old test to conform to requirements of a modern analytical laboratory. In this evaluation, two new semi-automated ESR measuring systems, the HumaSed and ESR-Auto Plus® (EAP), were compared with the Westergren method with regard to the accuracy and precision of ESR measurement, sample stability, and interference of various substances with the ESR assay. Samples from 125 patients were analyzed with the three methods and the results compared using linear regression and Bland-Altman plots. The mean ESR values of the HumaSed (32.10 ± 4.86) and EAP (38.09 ± 5.33) were comparable to that of the Westergren (31.54 ± 4.94). The high correlation coefficients of 0.910-0.96 and the Bland-Altman scatter plots revealed good association and agreement between the three methods. Bias between the three methods was small and the imprecision was within acceptable limits. ESR analysis beyond 4 hr was found to be unacceptable owing to sample instability. There was bilirubin and lipid but not heparin interference in the two automated systems. Overall, two automated analyzers were found to be fast, reliable, standardized, simplified, and safe instruments with accuracy and precision for ESR measurement comparable to the Westergren. © 2008 Wiley-Liss, Inc.