常用抗-HIV、抗-TPELISA血液筛查试剂盒阳性判断值的验证

Confirmation and analysis of the cut-off value in anti-HIV and anti-TP ELISA kits for blood donation screening

目的应用约登指数(Youden index)验证常用抗-HIV、抗-TP ELISA血液筛查试剂盒的阳性判断值(S/CO)及其"灰区"设置的必要性。方法利用SPSS17.0软件,计算国家卫计委临床检验中心(NCCL)牵头组织的国内常用血液筛查试剂多中心评估项目中的抗-HCV、抗-TP ELISA试剂不同S/CO值时的灵敏度、特异性及约登指数[约登指数=(灵敏度+特异性)-1],以S/CO值为横坐标,约登指数为纵坐标,绘制曲线,观察每种试剂各自在最大约登指数下的S/CO值。结果在所评估的7家抗-HIV ELISA试剂盒中,有1种试剂的S/CO值为1、其余6种试剂的S/CO值均为5时,其约登指数达最大。在所评估的5家抗-TP ELISA试剂盒中,有2种试剂的S/CO值为1、2种试剂的S/CO值为3和1种试剂的S/CO值为5时,其约登指数达最大。结论国内常用血液筛查试剂多中心评估项目中的抗-HIV和抗-TP ELISA试剂盒灵敏度和特异性最佳时均为S/CO≥1(亦即试剂盒自身设定的判定值),因此均不必设置"灰区"。

Objective To evaluate the necessity of the application of Youden index in the cut-off value （S/CO） determi- nation and the gray zone range establishment in anti-HIV and anti-TP ELISA kits for blood donation screening. Methods Data was collected from the project organized by the National Center for Clinical Laboratories to evaluate anti-HIV and anti- TP ELISA kits by multiple centers and was analyzed by the SPSS [ 17.0] system. The Youden index under different S/CO value was calculated with the equation where Youden index = sensitivity＋ specificity-1. Each curve was plotted with the S/CO value on the horizontal axis and Youden index on the vertical. Thus the optimal S/CO values were defined as the Youden in- dex values reaehe maximum. Results Among 7 different anti-HIV ELISA kits provided, one presented a maximum Youden index value at a S/CO value of 1 while the rest possessed maximum Youden index values at S/CO values of 5. As far as the 5 anti-TP ELISA kits being concerned, the Youden indexes reaches maximum values at S/CO values of 1 （2 kits）, 3 （2 kits） and 5 （ 1 kit）. Conclusion Since the sensitivity and specific of 7 anti-HIV ELISA kits and 5 anti-TP ELISA kits all reaches ootimal when S/CO ≥ 1. there is no need to set gray zone range for these two kinds kits evaluated.