Division of Biochemistry, The Ottawa Hospital, Canada; Department of Pathology and Laboratory Medicine, University of Ottawa, Canada.
Social Determinants and Science Integration Directorate, Public Health Agency of Canada, Canada.
Diabetes Res Clin Pract. 2018 Feb;136:143-149. doi: 10.1016/j.diabres.2017.11.035. Epub 2018 Jan 4.
Previous studies have shown varying sensitivity and specificity of hemoglobin A1c (HbA1c) to identify diabetes and prediabetes, compared to 2-h oral glucose tolerance testing (OGTT) and fasting plasma glucose (FPG), in different ethnic groups. Within the Canadian population, the ability of HbA1c to identify prediabetes and diabetes in First Nations, Métis and Inuit, East and South Asian ethnic groups has yet to be determined.
We collected demographic, lifestyle information, biochemical results of glycemic status (FPG, OGTT, and HbA1c) from an ethnically diverse Canadian population sample, which included a purposeful sampling of First Nations, Métis, Inuit, South Asian and East Asian participants.
Sensitivity and specificity using Canadian Diabetes Association (CDA) recommended cut-points varied between ethnic groups, with greater variability for identification of prediabetes than diabetes. Dysglycemia (prediabetes and diabetes) was identified with a sensitivity and specificity ranging from 47.1% to 87.5%, respectively in Caucasians to 24.1% and 88.8% in Inuit. Optimal HbA1c ethnic-specific cut-points for dysglycemia and diabetes were determined by receiver operating characteristic (ROC) curve analysis.
Our sample showed broad differences in the ability of HbA1c to identify dysglycemia or diabetes in different ethnic groups. Optimal cut-points for dysglycemia or diabetes in all ethnic groups were substantially lower than CDA recommendations. Utilization of HbA1c as the sole biochemical diagnostic marker may produce varying degrees of false negative results depending on the ethnicity of screened individuals. Further research is necessary to identify and validate optimal ethnic specific cut-points used for diabetic screening in the Canadian population.
既往研究显示，糖化血红蛋白（HbA1c）在不同种族人群中与 2 小时口服葡萄糖耐量试验（OGTT）和空腹血糖（FPG）相比，诊断糖尿病和糖尿病前期的敏感性和特异性存在差异。在加拿大人群中，HbA1c 用于识别第一民族、梅蒂斯人和因纽特人、东亚和南亚种族群体中糖尿病前期和糖尿病的能力尚未确定。
我们从一个种族多样化的加拿大人群样本中收集了人口统计学、生活方式信息以及血糖状态的生化结果（FPG、OGTT 和 HbA1c），其中包括有针对性地选择第一民族、梅蒂斯人、因纽特人、南亚和东亚参与者。
使用加拿大糖尿病协会（CDA）推荐的切点，不同种族人群的敏感性和特异性存在差异，对于糖尿病前期的识别差异更大。在白种人中，血糖异常（糖尿病前期和糖尿病）的识别敏感性和特异性范围分别为 47.1%至 87.5%，而在因纽特人中分别为 24.1%和 88.8%。通过接受者操作特征（ROC）曲线分析确定了 HbA1c 用于诊断不同种族人群中血糖异常和糖尿病的最佳种族特异性切点。
我们的样本显示，HbA1c 识别不同种族人群中血糖异常或糖尿病的能力存在广泛差异。所有种族人群中血糖异常或糖尿病的最佳切点均明显低于 CDA 的建议。单独使用 HbA1c 作为生化诊断标志物可能会因筛查个体的种族而异，产生不同程度的假阴性结果。需要进一步研究以确定和验证用于加拿大人群糖尿病筛查的最佳种族特异性切点。
