新生儿滤纸干血片自动打孔检测系统检测葡萄糖-6-磷酸脱氢酶的性能评价

Performance Evaluation of Automatic Perforation Detection System for Detecting Glucose-6-phosphate Dehydrogenase in Neonatal Filter Paper Dried Blood Slices

目的评价新生儿滤纸干血片自动打孔检测系统检测葡萄糖-6-磷酸脱氢酶(G6PD)的性能,并探讨在新生儿G6PD缺乏症筛查中的应用可行性。方法选取2020年12月至2021年7月来源于临床科研合作单位的104份新生儿滤纸干血片剩余样本,采用新生儿滤纸干血片自动打孔检测系统(简称自动打孔法)进行新生儿滤纸干血片G6PD检测,并对其检测低限、线性、准确度、精密度和一致性等性能指标进行评价,并与直接打孔法试验结果进行对比。结果自动打孔法检测G6PD的检测低限为0.29 U/gHb;在0.96~10.20 U/gHb范围内的样本,实测浓度与理论浓度的线性相关系数达0.9996,检测浓度与理论浓度的相对偏倚均在1/3允许总误差(±10%)内;检测质控品C1和C2的批内变异系数(CV)均在1/4允许总误差(小于7.5%)内,批间CV均在1/3允许总误差(小于10%)内;与直接打孔法平行检测临床样本的结果间差异无统计学意义(P>0.05),两种方法检测结果间具有高度相关性(r=0.9991,P<0.05),阳性判断值(2.50 U/gHb)预期偏倚的95%置信区间在可以接受范围内,采用Bland-Altman分析,两种方法有97.12%(101/104)的样本结果偏差在一致界限范围内。结论自动打孔法检测G6PD的检测低限、线性、准确度和精密度等性能指标均符合可接受标准,且自动打孔法与直接打孔法检测结果一致性良好,可用于新生儿滤纸干血片样本G6PD检测。

Objective The performance of the automatic punching detection system for detecting glucose-6-phosphate dehydrogenase(G6PD)on neonatal filter paper dry blood slice was evaluated,and the feasibility of its application in the screening of neonatal G6PD deficiency was investigated.Methods With the selection of remaining 104 samples of neonatal filter paper dry blood slices from clinical research cooperation units from December 2020 to July 2021,G6PD of neonatal filter paper dry blood slices were detected by using neonatal filter paper dry blood slice automatic punching detection system(Automatic punching method).Meanwhile,the detection lower limit,linearity,accuracy,precision,consistency and other performance indicators of the method were evaluated and compared with the detecting results of the direct punching method.Results The lower limit of G6PD detection by automatic punching method was 0.29 U/gHb;For the samples in the range from 0.96 U/gHb to 10.20 U/gHb,the linear correlation coefficient between the measured concentration and the theoretical concentration reached 0.9996,and the relative bias between the measured concentration and the theoretical concentration was within 1/3 of the allowable total error(±10%);The intra batch CVs of the tested quality control C1 and C2 were within 1/4 of the allowable total error(less than 7.5%),and the inter batch CVs were within 1/3 of the allowable total error(less than 10%);There was no statistically significant difference between the results of parallel detection of clinical samples by the automatic punching method and the direct drilling method(P>0.05).The results of the two methods were highly correlated(r=0.9991,P<0.05).The 95%confidence interval of the expected bias of the positive judgment value(2.50 U/gHb)was within the acceptable range.With Bland Altman analysis,97.12%(101/104)of the sample results of the two methods were within the consistent limit.Conclusions With the performance indicators such as lower limit of detection,linearity,accuracy and precision in G6PD detecting by automatic punching method meeting acceptable standards,the results of automatic punching method and direct punching method were consistent,which can be used to detect G6PD in neonatal filter paper dry blood sample.