ABSTRACT

Glycated hemoglobin (HbA1c) measured using high-performance liquid chromatography (HPLC) assays with venous blood and dried blood spots (DBS) are compared for 143 paired samples collected in Aceh, Indonesia. Relative to gold-standard venous-blood values, DBS-based values reported by the HPLC are systematically upward biased for HbA1c<8% and the fraction diabetic (HbA1c ≥ 6.5%) is overstated almost five-fold. Inspection of chromatograms from DBS assays indicates the % glycosylated calculated by the HPLC excludes part of the hemoglobin A which is misidentified as a hemoglobin variant. Taking this into account, unbiased DBS-based values are computed using data from the machine-generated chromatograms. When the DBS are collected in a clinic-like setting, under controlled humidity/temperature conditions, the recalculated values are almost identical to venous-based values. When DBS are collected under field conditions, the recalculated values are unbiased, but only about half the HbA1c values are measured reliably, calling into question the validity of the other half. The results suggest that collection conditions, particularly humidity, affect the quality of the DBS-based measures. Cross-validating DBS-based HbA1c values with venous samples collected under exactly the same environmental conditions is a prudent investment in population-based studies.

摘要

比较了使用高效液相色谱（HPLC）测定法与静脉血和干血斑（DBS）测得的糖化血红蛋白（HbA1c），对在印度尼西亚亚齐（Aceh）收集的143个配对样品进行了比较。相对于金标准静脉血值，HPLC报告的基于DBS的值被系统性地偏向于HbA1c <8％，而糖尿病分数（HbA1c≥6.5％）被夸大了将近五倍。DBS分析色谱图的检查表明，HPLC计算得出的糖基化百分比不包括被误认为是血红蛋白变异体的部分血红蛋白A。考虑到这一点，将使用机器生成的色谱图中的数据来计算基于DBS的无偏值。当在类似诊所的环境中在受控的湿度/温度条件下收集DBS时，重新计算的值几乎与基于静脉的值相同。当在野外条件下收集DBS时，重新计算的值是无偏的，但是只有大约一半的HbA1c值可以可靠地测量，这质疑了另一半的有效性。结果表明，收集条件，尤其是湿度，会影响基于DBS的措施的质量。将基于DBS的HbA1c值与在完全相同的环境条件下收集的静脉样本进行交叉验证是对基于人群的研究的一种谨慎投资。影响基于DBS的度量的质量。将基于DBS的HbA1c值与在完全相同的环境条件下收集的静脉样本进行交叉验证是对基于人群的研究的一种谨慎投资。影响基于DBS的度量的质量。将基于DBS的HbA1c值与在完全相同的环境条件下收集的静脉样本进行交叉验证是对基于人群的研究的一种谨慎投资。