利用空腹血糖值及估算平均血糖值設立糖化血紅素自動驗證規則

Utilizing Fasting Glucose and Estimated Average Glucose to Establish Automatic Verification Rules for Glycosylated Hemoglobin

目的:監測糖化血紅素(HbA1c)可反應一段期間血中平均葡萄糖的濃度，用來診斷及監控糖尿病病人的病情，精確監控糖化血紅素(HbA1c)是重要的。隨著人口老化加劇，人們對於醫療的需求與日俱增，導致臨床實驗室必須處理大量的檢體及檢驗報告，如何快速地提供精確的檢驗報告一直是檢驗醫學中重要的課題。因此我們可以透過自動驗證來提高檢驗的效率和報告的準確性，但自動驗證條件的設立是需要考慮到實驗室的環境及檢體的族群分布，是相當複雜的。我們在此篇文章中分享糖化血紅素(HbA1c)自動驗證規則設定的方法。
材料與方法:回溯病例方式收集糖化血紅素(HbA1c)及空腹血糖(Glucose)檢驗數值，並利用數值分佈圖設定極端值檢查(limit ch eck)、檢驗差值比較檢查( delta check)及分析物交叉比較(Cross-Analyte Comparison)。
結果:透過檢驗數值分布設定極端值檢查(limit check)下限值為4.5%，根據血中葡萄糖臨床決策值推算出極端值檢查(limit check )上限值為15%。分析同一人近兩次糖化血紅素(HbA1c)差值分布圖，設定檢驗差值比較檢查(delta check )為40%。除此之外，糖化血紅素(HbA1c)數值可經由公式換算出估算平均血糖值Estimated Average G lucose (eAG)，可與空腹血糖數值進行分析物交叉比較(Cross-Analyte Comparison)。透過估算平均血糖值(eAG)與空腹血糖的比值，畫出eAG/Gluose比值分布圖，設立當比值超出0.74~2.60範圍則判斷為不合理數值，表示糖化血紅素(HbA1c)檢測流程可能有異常。
討論:我們設立糖化血紅素(HbA1c)之自動驗證參數，透過自動驗證可以實現檢驗過程的自動化和標準化，使有正常報告快速核發，異常之檢體可被自動驗證攔截，減少發出錯誤報告風險，提高了工作效率及檢驗品質。

Purpose : Monitoring of HbA1c can reflect the average glucose concentration in the blood for a period of time, which is used to diagnose and monitor the condition of diabetes patients. It is important to accurately monitor HbA1c. With the intensification of population aging, people’s demand for medical care is increasing day by day, which leads to the need for clinical laboratories to deal with a large number of samples and test reports. How to quickly provide accurate test reports has always been an important issue in laboratory medicine. Therefore, we can improve the efficiency of testing and the accuracy of reporting through auto-verification, but the establishment of auto-verification criteria needs to take into account the laboratory environment and the population distribution of the samples, which are quite complex. In this article, we share the method of automatic verification setting rule for glycosylated hemoglobin (HbA1c).
Materials and Methods : In a retrospective manner, glycated hemoglobin (HbA1c) and fasting blood glucose (Glucose) test values were collected. Numerical distribution charts were used to set up limit checks for extreme values, perform delta checks for comparing differences between tests, and conduct cross-analyte comparisons for analyzing different parameters.
Results : By testing the distribution of numerical values, the lower limit of limit check is set to 4.5%, and the upper limit of limit check is calculated to be 15% based on the clinical decision value of glucose in the blood. The distribution of the difference between two consecutive HbA1c values for the same individual was analyzed, and a delta check of 40% was set to compare test value differences. In addition, HbA1c values can be converted using a formula to estimate the Estimated Average Glucose (eAG), which can be compared with fasting blood glucose values through Cross-Analyte Comparison. By estimating the ratio of average blood glucose (eAG) to fasting blood glucose, the distribution chart of eAG/Gluose ratio is drawn. When the ratio exceeds the range of 0.74~2.60, it is judged as an unreasonable value, indicating that there may be some abnormalities exist in the detection process of glycosylated hemoglobin (HbA1c).
Conclusion : We have set up auto-verification parameters for glycosylated hemoglobin (HbA1c). Through auto-verification, the inspection process can be automated and standardized, so that normal reports can be issued quickly, and abnormal samples can be intercepted by auto-verification, which reduces the risk of sending out false reports as well as improves working efficiency and inspection quality.