Department of Endocrinology, Renji Hospital, Shanghai Jiaotong University, China.
Acta Diabetol. 2010 Sep;47(3):231-6. doi: 10.1007/s00592-009-0143-2. Epub 2009 Sep 17.
The aim of this study is to assess the validity of combined use of fasting plasma glucose (FPG) and glycated hemoglobin A1c (HbA1c) as screening tests for diabetes and impaired glucose tolerance (IGT) in high-risk subjects. A total of 2,298 subjects were included. All subjects underwent a 75-g oral glucose tolerance test (OGTT) and HbA1c measurement. Receiver operating characteristic curve (ROC curve) analysis was used to examine the sensitivity and specificity of FPG and HbA1c for detecting diabetes and IGT, which was defined according to the 1999 World Health Organization (WHO) criteria. (1) Based on the ROC curve, the optimal cut point of FPG related to diabetes diagnosed by OGTT was 6.1 mmol/l that was associated with a sensitivity and specificity of 81.5 and 81.0%, respectively; The optimal cut point of HbA1c related to diabetes diagnosed by OGTT was 6.1%, which was associated with a sensitivity and specificity of 81.0 and 81.0%, respectively; The screening model using FPG > or = 6.1 mmol/l or HbA1c > or = 6.1% had sensitivity of 96.5% for detecting undiagnosed diabetes; the screening model using FPG > or = 6.1 mmol/l and HbA1c > or = 6.1% had specificity of 96.3% for detecting undiagnosed diabetes. (2) Based on the ROC curve, the optimal cut point of FPG related to IGT diagnosed by OGTT was 5.6 mmol/l that was associated with a sensitivity and specificity of 64.1 and 65.4%, respectively; The optimal cut point of HbA1c related to IGT diagnosed by OGTT was 5.6%, which was associated with a sensitivity and specificity of 66.2 and 51.0%, respectively; The screening model using FPG > or = 5.6 mmol/l or HbA1c > or = 5.6% had sensitivity of 87.9% for detecting undiagnosed IGT; The screening model using FPG > or = 5.6 mmol/l and HbA1c > or = 5.6% had specificity of 82.4% for detecting undiagnosed IGT. Compared with FPG or HbA1c alone, the simultaneous measurement of FPG and HbA1c (FPG and/or HbA1C) might be a more sensitive and specific screening tool for identifying high-risk individuals with diabetes and IGT at an early stage.
本研究旨在评估空腹血糖(FPG)和糖化血红蛋白 A1c(HbA1c)联合用于筛查高危人群糖尿病和糖耐量受损(IGT)的有效性。共有 2298 名受试者入组。所有受试者均行 75g 口服葡萄糖耐量试验(OGTT)和 HbA1c 检测。采用受试者工作特征曲线(ROC 曲线)分析 FPG 和 HbA1c 对糖尿病和 IGT 的诊断价值,IGT 按照 1999 年世界卫生组织(WHO)标准定义。(1)基于 ROC 曲线,FPG 与 OGTT 诊断糖尿病相关的最佳切点值为 6.1mmol/L,此时的灵敏度和特异度分别为 81.5%和 81.0%;HbA1c 与 OGTT 诊断糖尿病相关的最佳切点值为 6.1%,此时的灵敏度和特异度分别为 81.0%和 81.0%;以 FPG≥6.1mmol/L 或 HbA1c≥6.1%作为筛查模型,对未诊断糖尿病的检出率为 96.5%;以 FPG≥6.1mmol/L 且 HbA1c≥6.1%作为筛查模型,对未诊断糖尿病的检出率为 96.3%。(2)基于 ROC 曲线,FPG 与 OGTT 诊断 IGT 相关的最佳切点值为 5.6mmol/L,此时的灵敏度和特异度分别为 64.1%和 65.4%;HbA1c 与 OGTT 诊断 IGT 相关的最佳切点值为 5.6%,此时的灵敏度和特异度分别为 66.2%和 51.0%;以 FPG≥5.6mmol/L 或 HbA1c≥5.6%作为筛查模型,对未诊断 IGT 的检出率为 87.9%;以 FPG≥5.6mmol/L 且 HbA1c≥5.6%作为筛查模型,对未诊断 IGT 的检出率为 82.4%。与 FPG 或 HbA1c 单独检测相比,同时检测 FPG 和 HbA1c(FPG 和/或 HbA1C)可能是一种更敏感和特异的筛查工具,可早期识别出糖尿病和 IGT 的高危人群。