RESUMO
BACKGROUND: Current haematology analysers have variable sensitivity and accuracy for counting nucleated red blood cells in samples with low values and in all those conditions characterised by altered sensitivity of red blood cells to the lysing process, such as in beta-thalassaemia or sickle-cell diseases and in neonates. The aim of our study was to evaluate the performance of the automated analyser XE-2100 at counting nucleated red blood cells in the above-mentioned three categories of subjects with potentially altered red blood cell lysis sensitivity and yet a need for accurate nucleated red blood cell counts. MATERIALS AND METHODS: We measured nucleated red blood cell count by XE-2100 in peripheral blood samples of 187 subjects comprising 55 patients with beta-thalassaemia (40 major and 15 traits), 26 sickle-cell patients, 56 neonates and 50 normal subject. Results were compared with those obtained by optical microscopy. Agreement between average values of the two methods was estimated by means of Pearson's correlation and bias analysis, whereas diagnostic accuracy was estimated by analysis of receiver operating characteristic curves. RESULTS: The comparison between the two methods showed a Pearson's correlation of 0.99 (95% CI; 0.98-0.99; p<0.001) and bias of -0.61 (95% CI, -1.5-0.3). The area under the curve of the nucleated red blood cell count in all samples was 0.98 (95% CI, 0.96-1.00; p<0.001). Sub-analysis revealed an area under curve of 0.99 (95% CI, 0.98-1.00; p<0.001) for patients with thalassaemia, 0.94 (95% CI, 0.85-1.00; p<0.001) for patients with sickle cell anaemia, and 1.00 (95% CI, 1.0-1.0) for neonates. DISCUSSION: XE-2100 has excellent performance for nucleated red blood cell counting, especially in critical populations such as patients with haemoglobinopathies and neonates.
