中文摘要：

目的：探讨电化学发光法（ECLIA）不同批号癌胚抗原（CEA）试剂室内质控发生明显偏倚时对患者检测结果的影响。方法留取两批次 CEA 试剂检测的标本数：批号为171617检测的临床标本66例；批号为172356检测的临床标本58例。用批号为172356 CEA 试剂重新检测，计算两次检测结果的偏倚。结果不同批号试剂检测结果差异有统计学意义（P =0．012）；不同批号试剂检测结果偏倚与相同批号试剂检测结果偏倚两者比较差异有统计学意义（ P 〈0．001）；不同批号试剂检测结果正向偏倚和负向偏倚的比例与相同批号试剂检测结果正向偏倚和负向偏倚的比例，两者比较差异有统计学意义（P 〈0．001）。用室内质控的偏倚确定校正因子纠正患者结果，不同批号试剂纠正后结果差异没有统计学意义（P =0．828）；不同批号试剂纠正后偏倚与相同批号试剂偏倚比较差异无统计学意义（P =0．092）；不同批号试剂纠正后正向偏倚和负向偏倚的比例与相同批号试剂正向偏倚和负向偏倚的比例比较差异无统计学意义（P =0．774）。结论不同批号 CEA 试剂室内质控靶值存在明显偏倚时，患者检测结果会产生相似的偏倚，该偏倚可以通过校正因子进行纠正。

英文摘要：

Objective To investigate the effect of different batches reagent of carcino-embryonic antigen by the electrochemiluminescence on the patients when internal quality control has obvious bias. Methods We measured samples from two batches of CEA clinical specimen, 66 from batch 171617 and 58 from batch 172356. All of these samples were re-measured by batch 172356, and then the bias of the two measures was analyzed. Results The de-tection results in different batches of reagent were of statistical significance （P = 0. 012）. In result bias, there was a significant difference between the two groups （P 〈 0. 001）. In the ratio of the positive and negative bias, there was also a significant difference between the two groups （P 〈 0. 001）. There was no statistical significance （P =0. 828） in different batches of reagent following the adjustment of patients′ outcomes according to the bias of inter-nal quality control. There was no statistical significance （P = 0. 092） either after the adjustment in different batches of reagent compared with that of the same batch of the reagent. The proportions of the positive and the negative bias were of no statistical significance （P = 0. 774） compared with those of the same batch of reagent. Conclusion The patients′ outcomes would produce similar bias when the target value of CEA internal quality control detected by dif-ferent batches of reagents has significant bias and this bias can be adjusted by an adjustment factor.