摘要： 目的 评价CardioChek PA便携式全血检测仪（便携式全血检测仪）检测血脂的可靠性，探讨其用于社区脑卒中高危人群血脂筛查的可行性及准确性。方法 根据美国国家临床实验室标准化委员会制定的患者样本比对实验及偏差评价指南 EP9-A 文件标准，分别以ADVIA2400全自动生化分析仪和便携式全血检测仪为参考仪器（目标检测系统）和试验系统，检测受试者血清总胆固醇、甘油三酯、高密度脂蛋白胆固醇和低密度脂蛋白胆固醇水平，计算两种检测系统相关系数和直线回归 方程，并对其预期偏差进行评估；在比对基础上应用便携式全血检测仪检测1263例社区年龄> 40岁的高危人群血脂水平。结果 便携式全血检测仪检测血脂与ADVIA2400全自动生化分析仪具有良好的相关性（r > 0.975，P < 0.05 ），不同医学决定水平可允许误差均位于预期偏差的95%CI 内，提示两种仪器检测总胆固醇、甘油三酯、高密度脂蛋白胆固醇、低密度脂蛋白胆固醇水平相近，数据可接受。社区筛查共检测1263例受试者，异常检出率为总胆固醇347 例（27.47%）、甘油三酯348 例（27.55%）、高密度脂蛋白胆固醇62例（4.91% ）、低密度脂蛋白胆固醇441 例（34.92%）。 结论 CardioChek PA便携式全血检测仪测定胆固醇、甘油三酯、高密度脂蛋白胆固醇和低密度脂蛋白胆固醇的结果与实验室常规ADVIA2400全自动生化分析仪测定结果相近，结果准确可信，可用于社区流行病学调查。

关键词: 高脂血症, 卒中, 普查, 社区卫生服务

Abstract: Objective  To evaluate the reliability of CardioChek PA portable blood analyzer in the detection of blood lipid and to explore the feasibility and accuracy of its application in lipid screening in high-risk stroke group. Methods  According to the profile of the National Committee for Clinical Laboratory Standards of Method Comparison and Bias Estimation Using Patient Samples (NCCLS-EP9-A), two different biochemical analysis systems, CardioChek PA blood analyzer (testing system) and Siemens ADVIA2400 automatic analyzer (reference system), were used respectively to test lipids including total cholesterol (TC), triglyceride (TG), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C) to obtain the correlation coefficient and the linear regression equation as well as predicted bias. Based on the comparison, CardioChek PA blood analyzer was used to detect blood lipid of 1263 persons (more than 40 years old) at high-risk of stroke. Results The outcomes of TC, TG, HDL-C and LDLC between CardioChek PA blood analyzer and Siemens ADVIA2400 automatic analyzer had significant correlations ( r > 0.975, P < 0.05) and error allowance at different medical decision levels were all between 95% confidence interval (CI) of predicted bias, indicating that there were comparability of TC, TG, HDL-C and LDL-C between two detecting systems and the outcomes were acceptable. Moreover, there were 347 cases (27.47%) with abnormal TC, 348 cases (27.55%) with abnormal TG, 62 cases (4.91%) with abnormal HDL-C and 441 cases (34.92%) with abnormal LDL-C among 1263 persons. Conclusion  There is comparability of TC, TG, HDL-C and LDL-C between CardioChek PA blood analyzer and ADVIA2400 biochemical automatic analyzer and the outcomes of CardioChek PA are accurate and confident, thus CardioChek PA can be used in epidemiological study.

Key words: Hyperlipidemias, Stroke, Mass screening, Community health services