POCT血糖仪的性能分析

目的比对5个品牌9款床边检测（POCT）血糖仪的主要分析性能。方法用高、中、低3个浓度的质控品在9款POCT血糖仪上进行检测，连续测试30d，计算变异系数（CV）。取高、中、低3种不同葡萄糖浓度的患者样本，同一样本取2份。一份为新鲜全血乙二胺四乙酸（EDTA）抗凝]，在9款POCT血糖仪上进行检测，计算同一样本在9款POCT血糖仪上检测所得结果的均值（x）、标准差（s）、CV；另一份为新鲜血清样本，用日立7060全自动生化分析仪进行检测，评价日立7060全自动生化仪与POCT血糖仪测定结果的相关性。结果质控品在9款POCT血糖仪上所测结果的CV在2．62％～5．86％之间，符合国家对血糖检测系统的测量重复性要求（CV不超过7．5％）。高浓度血糖用POCT血糖仪所测结果的CV在10％左右，中、低浓度血糖用POCT血糖仪所测结果的CV均〉10％，与美国临床实验室标准化委员会（NCCLS）对葡萄糖POCT的管理要求（总误差≤10％）不符。Johnson Ultra和ArkRay GT-1810 2款POCT血糖仪与日立7060全自动生化仪所测结果的相关系数（r）〈0．95，相关性略差；其他几款POCT血糖仪与日立7060全自动生化仪所测结果的r均t〉0．95，相关性较好。KBCT-X10、Roche Integra、Advantage和Johnson Ultra POCT血糖仪的斜率（b）超出1．0±0．05。结论POCT血糖仪仅可用于患者空腹血糖的监测或血糖过筛试验，不能代替实验室血糖的检测结果作为确诊试验。同时在选择POCT血糖仪时，不能仅一味追求其检测的快速，应该对血糖仪的分析性能进行综合评价。

Analytic evaluation of point of care testing blood glucometers

Objective To compare the main analytical ability of 9 point of care testing (POCT) glucometers of 5 brands. Methods The control materials with 3 gradual levels (high, medium and low) were detected continually for 30 days and then the coefficient of variation (CV) was calculated. The glucose concentrations were determined from patients′ samples with high, medium and low levels. Each sample was divided into two parts. One part was the fresh whole blood anti-coagulation with ethylenediaminetetraacetaic acid (EDTA)] which was detected on 9 POCT glucometers and calculated the mean value (), standard deviation (s) and CV (%). The other part was the fresh serum which was detected with Hitachi 7060 automatic biochemical analyzer and the correlation with glucometers was evaluated. Results The control materials CV values of 9 glucometers were from 2.62% to 5.86%, in accordance with the rule that required CV below 7.5%. High blood glucose concentrations had the CV value around 10%, and medium and low ones had the CV value more than 10%, which did not follow the National Committee for Clinical Laboratory Standards (NCCLS) requirements that CV should be below 10%. The results were not consistent with the requirement total error (bias and imprecision)≤10%] of the NCCLS. POCT glucometers had good correlations (r≥0.95) with automatic biochemical analyzer except Johnson Ultra and ArkRay GT-1810. The slopes (b) of KB CT-X10, Roche Integra, Advantage and Johnson Ultra were above 1.0±0.05. Conclusions POCT glucometer can be used for monitoring the fasting blood-glucose and for the blood-glucose screening test. POCT can not replace the traditional blood-glucose tests and can not act as the diagnostic test. Meanwhile, for selecting the POCT glucometers, we should focus not only on the testing speed but also on their analytical ability.