同一医院内两台凝血分析仪血浆凝血酶原时间的可比性验证

Comparable Verification of Patient Results of PT by Two Different Type Instruments Within One Health Care System

目的对同一医院内血浆凝血酶原时间在两台凝血分析仪上检测可比性进行验证。方法板差检验可比性方法包括：选择比对的分析物，选择比对的仪器，确定可比性检验接近的分析物浓度，计算用于比对样奉选择的期望的浓度，选择可比性的样本，设定可比性检验的临界差值，患者结果可比性检验可接受标准，确定检测重复的次数。结果两台凝血分析仪在两个浓度水平的不精密度都为3％。样本范围计算为13±13×20％（s），或10．4S15．6s和25±25×20％（s）或20．0s～30．0s。对于比对试验选择主要的分析仪样本最初的值为14．2s（样本1）和27．1s（样本2）。临界差值可接受标准定为15％。根据极差检验的临界差值（％）以及上述的信息可选择两个不同水平的患者样本，且每一样本只作单次检测来检验两台仪器之间是否存在差异。两台检测系统对于两个浓度水平的患者样本的极差值分别为6．14％和17．51％，调整极差分另1为3．86％和2．51％，因为调整极差的绝对值小于临界差值，对于两样本方法的可比性对于给定的已知的差值是可接受的。结论极差检验方法为评估在一个医疗机构内检测同一患者样本的不同系统的可比性提供了指南，比传统的方法比对试验简单、易实施。

Objective To verify the comparability of patients results of PT by two different types of instrument in one hospital. Methods The procedures of range comparison method include：selecting comparison analysis materials,selecting in- struments,calculating analysis material concentrations which were approaching the tests of comparability,calculating the concentrations or activities of expectations which could used to select comparable samples,selecting comparable samples, setting critical differentials of comparability tests ,acceptable standards of patients results comparability tests ,ensuring the number of test replication. Results Laboratories should evaluate the comparability of PT between major analysis instruments and reserved instruments. The imprecision levels of two types of analysis instruments were both 3%. Sample ranges were 13±13×20%（10.4s-15.6s） and 25±25×20% （20. 0s-30. 0s）. Selected 14.2s （sample 1） and 27. ls （sample 2） as the initial values of major analysis instruments. The acceptable standard of critical differentials was intended to be 15 %. According to the critical differentials （%） of studentized range tests and the information forward we could choose patient samples of two different levels while each sample only test once for each analysis instruments for testing whether there was difference between them. The mean percentage ranges for two analyzers results of each kind of samples were 6.14% and 17.51% respectively while the adjusted percentage ranges were 3.86% and 2.51%. Because the absolute values of the adjusted ranges were smaller than critical differentials,the comparability of the two methods for the given standard was acceptable. Conclusion Range test method supplies a guideline for evaluating the comparability for different systems in one health care ＇system of the same patient sample.