以常規檢測 HbA1c之異常圖形篩檢出糖尿病患者變異型血色素的比率

The Frequency of Detection of Hemoglobin Variants in Diabetic Patients Identified by Abnormal Hemoglobin Chromatograms Through Routine Use of an A1c Analyzer Tosoh HLC-723 G5

目的：糖化血色素(HbA1c)是評估長期血糖控制的重要指標，此外每減少1%HbA1c，可使因大血管病變引起的糖尿病死亡率降低25%，並減少18%致命或非致命性突發心臟病發生率。全世界已發表的變異型血色素，超過700種，台灣常見的變異型有Hb-G、Hb-J、Hb-E、Hb-M、Hb-H…等。變異型血色素常會影響”HbA1c”的測定值，臨床上經常是醫生發現血糖與HbA1c不相符，才發現變異型血色素之存在。如要在一般常規HbA1c檢測中發現其他變異型血色素，可使用Ion-exchange HPLC 方法，並注意HbA1c的圖形變化，有異樣則要進一步探討是否有變異型血色素的存在。為了探討變異型血色素在糖尿病患者精確之比率，以提供臨床醫生及醫檢同仁更精確的數據。方法： 我們收集了自民國94年3月至7月本科4239位糖尿病患者之血液，以本科所使用的機型Tosoh HLC723 G5（為陽離子高壓液相色層分析法），檢測其糖化血色素。當發現有異常圖形之HbA1c時，則依據平均色血球容積（MCV）或平均紅血球血色素量（MCH）的值及使用毛細管血色素電泳（Prince CE550）做變異型血色素檢定。結果： 使用Tosoh HLC723 G5 其intraassay及interassay之CV皆小於1﹪。4239位糖尿病患者男女比為1：1.3，年齡介於1-75歲（mean=62.6），HbA1c值介於0.6﹪-19.7﹪（mean=8.2﹪）。發現變異型血色素共15例（其中Hb-J 6例、Hb-E 1例、Hb-F 7例、Hb-H 1例）。所以本科糖尿病患者之變異型血色素比率為百分之0.35。結語： 台灣變異型血色素的盛行率小於千分之一。本科糖尿病患者之變異型血色素比率為百分之0.35，高於全體民眾之比率，而變異型血色素影響”HbA1c ”的測定值是難免的，因此強烈建議操作者要注意HbA1c圖形變化，希望能由常規之HbA1c檢驗篩選出變異型血色素，以提供臨床醫生參考。

Aims: Hemoglobin A1c (HbA1c) is an invaluable tool for monitoring long-term glycemic control. Every 1% reduction in HbA1c will decrease by 25% macrovascular mortality, and by 18% fatal or nonfatal cardiovascular events. However, more than 700 characterized Hb variants have been reported in the world. The variant types that have been found in Taiwan include Hb-G, Hb-J, Hb-E, Hb-M, Hb-H, etc. The presence of an Hb variant may produce a false HbA1c value. Clinically, an Hb variant is suspected when the HbA1c result varies substantially from the serum glucose level. An ion exchange HPLC method indicating an abnormal HbA1c level is able to identify the possible existence of an Hb variant. The aim of the present study is to evaluate the occurrence of Hb variants among diabetes patients. Methods: Blood samples from 4239 diabetes patients from the Division of Endocrinology and Metabolism of Kaohsiung Veterans General Hospital were collected from March to July 2005. A Tosoh HLC723 G5 (using an ion-exchange HPLC methodology) was used to measurement the patients’ HbA1c. A Prince CE 550 (using a capillary electrophoresis methodology) was used to detect Hb variants when an abnormal HbA1c level was found. Results: The intra-assay and inter-assay coefficients of variation for the Tosoh HLC723 G5 were below 1%. The mean age of the 4239 diabetes patients (male:female ratio of 1:1.3) was 62.6 years old (range 1-75 years old). The mean HbA1c was 8.2% (range 0.6% to 19.7%). Fifteen patients with Hb variants were detected (Hb-J in 6 patients, Hb-E in 1 patient, Hb-F in 7 patients, Hb-H in 1 patient). The prevalence of Hb variants among these diabetes patients was 0.35%. Conclusion: The prevalence of Hb variants in Taiwan is normally below 0.1%. The prevalence among diabetic patients visiting the Division of Endocrinology and Metabolism of Kaohsiung Veterans General Hospital was 0.35%, which is higher than that reported for Taiwan in general. Therefore, Hb variants among diabetes patients might influence the measurement of HbA1c and, as a result, any abnormal HbA1c levels during routine HbA1c measurement should be evaluated further in order to detect the presence of Hb variants.