Myers G L, Ross J W, Smith S J, Morris C H, Triplett R B, Groff M
Centers for Disease Control and Prevention, Atlanta, GA 30333, USA.
Arch Pathol Lab Med. 1995 Aug;119(8):686-94.
To evaluate the suitability of various commercial preparations for use by the College of American Pathologists as survey materials in assessing high-density lipoprotein cholesterol measurement performance.
Lyophilized human serum preparations from six vendors (vendors A through F) were evaluated to determine which material(s) best mimicked the commutability of fresh human serum. Two freshly collected unfrozen pools prepared from donor specimens were analyzed concurrently with the vendor materials to identify sources of variation and possible matrix bias. Each material was evaluated using 5 common precipitation reagents (phosphotungstate-magnesium, phosphotungstic acid, dextran sulfate [50K and 500K], and heparin-manganese). To evaluate how each reagent separates lipoproteins in each material, the lipoprotein separation patterns were profiled using high-pressure liquid chromatography and compared with separation patterns observed for the fresh human serum pools.
Similarities in performance characteristics of vendor material(s) were compared with fresh human serum.
Two of the six materials gave separation profiles for the lipoproteins similar to the typical patterns observed for human serum. Material from vendor B showed the best commutability across all of the precipitation reagents and had the best combination of low overall variability (10% for level 1 and 9.4% for level 2) and minimal concentration differences among reagents.
Vendor B was selected by the College of American Pathologists to provide materials for use in assessing performance of lipid and lipoprotein testing in the 1994 Comprehensive Chemistry Surveys. This study demonstrates the great variability that different vendor preparations introduce into the measurement of high-density lipoprotein cholesterol. It also emphasizes the effort required to evaluate the suitability of processed materials for use in proficiency testing.
评估各种商业制剂是否适合美国病理学家学会用作评估高密度脂蛋白胆固醇测量性能的调查材料。
对来自六个供应商（供应商A至F）的冻干人血清制剂进行评估，以确定哪种材料最能模拟新鲜人血清的互换性。同时分析从供体标本制备的两个新鲜采集且未冷冻的混合样本与供应商提供的材料，以识别变异来源和可能的基质偏差。使用5种常见的沉淀试剂（磷钨酸镁、磷钨酸、硫酸葡聚糖[50K和500K]以及肝素锰）对每种材料进行评估。为了评估每种试剂如何分离每种材料中的脂蛋白，使用高压液相色谱对脂蛋白分离模式进行分析，并与新鲜人血清混合样本中观察到的分离模式进行比较。
将供应商材料的性能特征与新鲜人血清进行比较。
六种材料中的两种给出的脂蛋白分离图谱与人血清中观察到的典型模式相似。供应商B的材料在所有沉淀试剂中表现出最佳的互换性，并且具有总体变异性低（1级为10%，2级为9.4%）和试剂间浓度差异最小的最佳组合。
美国病理学家学会选择供应商B提供用于1994年综合化学调查中评估脂质和脂蛋白检测性能的材料。本研究表明不同供应商的制剂在高密度脂蛋白胆固醇测量中引入了很大的变异性。它还强调了评估加工材料用于能力验证的适用性所需的努力。
