论文部分内容阅读
目的应用生物学变异和分析变异探讨糖化血红蛋白(glycosylated hemoglobin,HbA1c)测定结果对糖尿病诊断治疗的影响。方法以三级质量规范为标准,应用HbA1c生物学变异和参考变化值,分析实验室测量固有分析变异(analytical variance,CV_A)、个体内生物学变异(within-individual variation,CVI)以及偏倚对HbA1c测定结果的影响。结果 HbA1c的测定结果随CV_A的增加而导致的额外偏倚不呈线性增加。不同的生物学变异质量规范要求对HbA1c测定结果的影响有差异。不同的CV_A对非糖尿病个体和糖尿病患者的HbA1c测定结果离散范围影响不同。HbA1c检测要求分析方法的CV_A<3%。结论糖尿病诊断和治疗并不能单独依赖HbA1c测定结果的实际数值为判断标准。应结合CV_A及CVI对于HbA1c测定值偏倚的影响给出测定值,并附上置信区间。
Objective To investigate the effect of glycosylated hemoglobin (HbA1c) determination on the diagnosis and treatment of diabetes mellitus by using biological variation and analysis of variance. Methods Based on the three-level quality standard, the laboratory variance (HbA1c) and reference changes were used to analyze the analytical variance (CV_A), within-individual variation (CVI) and bias of HbA1c Determination of the impact of the results. Results The additional bias of HbA1c results with the increase of CV_A did not increase linearly. The impact of different biological variation quality specifications on the HbA1c assay results is different. Different CV_A had different effects on the discrete range of HbA1c in nondiabetic and diabetic subjects. The HbA1c assay requires CV_A <3% of the assay. Conclusion The diagnosis and treatment of diabetes can not depend solely on the actual value of HbA1c measurement as the criterion. The effect of CV_A and CVI on the bias of HbA1c measurements should be combined with confidence intervals.