George Kelita, Joy Sarah, Kaladi Palliyalil Haseena, Shaji Shibin M, Ajayakumar Sunanda Vishakan, Sreedevi Aswathy
Community Medicine, Amrita Institute of Medical Sciences, Amrita Vishwa Vidyapeetham, Kochi, IND.
Community Medicine, Amrita Institute of Medical sciences, Amrita Vishwa Vidyapeetham, Kochi, IND.
Cureus. 2024 Oct 7;16(10):e70981. doi: 10.7759/cureus.70981. eCollection 2024 Oct.
Introduction The International Diabetes Federation states that India accounts for one in seven of all adults with diabetes. Adherence to self-monitoring of blood glucose is essential for effective management of diabetes. Despite the wide variety of glucometers and their clinical applicability, there is a lot of ambiguity regarding their accuracy. Therefore, this study aimed to evaluate if three point-of-care glucometers using three different methods for the estimation of blood glucose fulfil the minimum accuracy needed by ISO15197:2013 when compared with the gold standard hexokinase test. Methods A cross-sectional study was conducted at two primary health care centers in Ernakulam district, which included 73 participants with diabetes and 73 without diabetes. We evaluated three different enzymes in the test glucometers: glucose dehydrogenase with pyrroloquinoline quinone (GDH-PQQ), glucose dehydrogenase with flavin adenine dinucleotide (GDH-FAD), and glucose oxidase (GOD). The glucometer readings were compared with the gold standard hexokinase test and ISO15197:2013 standards. To compare the accuracy of each glucometer with the reference method, the Bland-Altman plot was used and to evaluate the clinical significance of the differences between the glucometer and reference value the Clarke error grid analysis was performed. Results When compared with the gold standard hexokinase result, only 78.08% (GOD), 92.4% (GDH-PQQ), and 95.8 (GDH-FAD) of results were within the limits stipulated by ISO15197:2013. All three glucometers showed blood glucose results within zones A and B of the consensus error grid, implying that it is clinically acceptable, and the Bland-Altman plot showed that the GDH-FAD method had the narrowest range between the upper and lower limit of agreement and a minimum bias of -0.89. Conclusion Only one test glucometer with the GDH-FAD method met the ISO15197:2013 criteria. Glucometers must meet the accuracy standards for the safety of the patients.
引言 国际糖尿病联盟指出，印度成年糖尿病患者占全球成年糖尿病患者总数的七分之一。坚持自我血糖监测对于有效管理糖尿病至关重要。尽管血糖仪种类繁多且具有临床适用性，但其准确性仍存在诸多模糊之处。因此，本研究旨在评估三种采用不同方法估算血糖的即时检测血糖仪与金标准己糖激酶检测相比，是否符合ISO15197:2013所要求的最低准确性。
方法 在埃纳库拉姆区的两个初级卫生保健中心开展了一项横断面研究，研究对象包括73名糖尿病患者和73名非糖尿病患者。我们评估了检测血糖仪中的三种不同酶：吡咯喹啉醌葡萄糖脱氢酶（GDH-PQQ）、黄素腺嘌呤二核苷酸葡萄糖脱氢酶（GDH-FAD）和葡萄糖氧化酶（GOD）。将血糖仪读数与金标准己糖激酶检测及ISO15197:2013标准进行比较。为了将每种血糖仪的准确性与参考方法进行比较，使用了布兰德-奥特曼图，并通过克拉克误差网格分析来评估血糖仪与参考值之间差异的临床意义。
结果 与金标准己糖激酶检测结果相比，只有78.08%（GOD）、92.4%（GDH-PQQ）和95.8%（GDH-FAD）的结果在ISO15197:2013规定的范围内。所有三种血糖仪的血糖结果均在共识误差网格的A区和B区内，这意味着在临床上是可接受的，并且布兰德-奥特曼图显示，GDH-FAD方法在一致性上限和下限之间的范围最窄，最小偏差为-0.89。
结论 只有采用GDH-FAD方法的一种检测血糖仪符合ISO15197:2013标准。血糖仪必须符合准确性标准以保障患者安全。
