Klapuh-Bukvić Nermina, Kurtanović Zehra, Šeper Damir
Clinical Center University of Sarajevo, Sarajevo, B&H.
University of Sarajevo, Faculty of Pharmacy, Sarajevo, B&H.
J Med Biochem. 2025 Aug 21;44(5):1009-1019. doi: 10.5937/jomb0-55966.
Differentiation of leukocytes is one of the key diagnostic procedures in clinical medicine, and correct identification of them in a blood smear is of essential importance. Light microscopy is the reference method for leukocyte differentiation; however, it is time-consuming and must be performed by a highly qualified specialist. For this reason, automatic analysers capable of precise and accurate differentiation of blood cells in the examined sample are increasingly present in haematology laboratories. This paper aims to evaluate the performance of the Sysmex XN-3100 analyser, manufactured by SYSMEX CORPORATION, Kobe, Japan., with a focus on the advantages and disadvantages of its digital microscopy in the differentiation of leukocytes and give brief guidelines on the possibilities and limitations of everyday work on the basis of the obtained results.
Digital optical microscopy on 253 samples was performed with primary data (preclassification) collected after the completion of the autoanalysis. Before validating the obtained results, the data were reviewed by a medical biochemistry specialist who confirmed or corrected them. This generated secondary data (reclassification). The two groups of data were statistically analysed using Passing-Bablok regression analysis, Bland-Altman analysis and Spearman correlation.
The obtained results showed strong correlations between the primary and secondary analysis in all cells (highest in lymphocyte group (r=0.986), lowest in eosinophil group (r=0.870)) except immature granulocytes and blasts (significant deviation from linearity, p<0.01).
The haematology analyser Sysmex XN-3100 shows high performance in leukocyte analysis and differentiation using digital microscopy, but samples containing blasts and immature granulocytes must additionally be analysed by light microscopy.
白细胞分类是临床医学中的关键诊断程序之一，在血涂片上正确识别白细胞至关重要。光学显微镜检查是白细胞分类的参考方法；然而，该方法耗时且必须由高素质的专家进行操作。因此，血液学实验室越来越多地配备了能够对检测样本中的血细胞进行精确分类的自动分析仪。本文旨在评估日本神户SYSMEX公司生产的Sysmex XN - 3100分析仪的性能，重点关注其数字显微镜在白细胞分类方面的优缺点，并根据所得结果对日常工作的可能性和局限性给出简要指导。
对253份样本进行数字光学显微镜检查，自动分析完成后收集原始数据（预分类）。在验证所得结果之前，由医学生物化学专家对数据进行审核，确认或纠正数据。这产生了二次数据（重新分类）。使用Passing - Bablok回归分析、Bland - Altman分析和Spearman相关性对两组数据进行统计分析。
所得结果显示，除未成熟粒细胞和原始细胞外，所有细胞的一次分析和二次分析之间均具有强相关性（淋巴细胞组最高（r = 0.986），嗜酸性粒细胞组最低（r = 0.870））（显著偏离线性，p < 0.01）。
血液分析仪Sysmex XN - 3100在使用数字显微镜进行白细胞分析和分类方面表现出高性能，但含有原始细胞和未成熟粒细胞的样本必须另外通过光学显微镜进行分析。
