中国人群平均血糖与糖化血红蛋白转换关系的初步分析

目的 采用动态血糖监测（CGM）技术探讨不同糖代谢人群24 h平均血糖水平（MBG）与糖化血红蛋白（HbA1c）的关系,并初步建立不同HbA1c水平其相对应的MBG值.方法 选取2007年8月至2010年1月于上海交通大学附属第六人民医院门诊及病房进行口服葡萄糖耐量试验（OGTT）的受试者318人,其中正常糖调节组115名、糖调节受损组57例、新诊断2型糖尿病患者组146例.糖尿病、糖调节受损的诊断采用2003年美国糖尿病学会（ADA）糖尿病诊断标准.测量各组生化指标,并行OGTT,测定糖负荷后2 h血糖（2 h PG）.采用CGM系统估算MBG水平,高压液相法测定HbA1c水平,分析MBG与HbA1c间的关系.统计学采用单因素ANOVA方差分析、Kruskal-Wallis检验、χ^2检验、Spearman相关分析及线性模型进行数据统计.结果 318例研究对象中总体HbA1c水平为（6.6±1.5）%,MBG水平为（7.3±2.3）mmol/L.（2）MBG与HbA1c呈显著正相关（r=0.848,P〈0.01）,其回归方程为MBG=1.252×HbA1c-0.992（R^2= 0.718,P〈0.01）,即HbA1c每升高1%,MBG相应升高1.252 mmol/L.HbA1c=6.5%时,对应的MBG值为7.1 mmol/L.HbA1c=7.0%时,对应的MBG值为7.8 mmol/L.（3）仅包括146例新诊断2型糖尿病患者的MBG和HbA1c之间的关系与总人群的结果一致（r=0.788,P〈0.01）,线性回归方程为MBG=1.255×HbA1c-0.886（R2=0.621,P〈0.01）.当HbA1c= 6.5%时,MBG对应值为7.3 mmol/L.HbA1c=7.0%时,对应的MBG值为7.9 mmol/L.结论 初步建立了各HbA1c水平其相应的MBG值,为深入开展中国人群平均血糖与HbA1c转换关系研究提供了线索.

Investigation on translating glycated hemoglobin A1c into mean blood glucose in Chinese subjects

Objective To establish the corresponding value of mean blood glucose (MBG) for a specific target of glycated hemoglobin A1C (HbA1c), and evaluate the relationship between 24 hour MBG determined by continuous glucose monitoring and HbA1c in subjects with different states of glucose tolerance. Methods From August 2007 to January 2010, 318 subjects who accepted oral glucose tolerance test in out-patient department and ward of Shanghai Jiao Tong University Affiliated Sixth People′s Hospital were enrolled, including 115 with normal glucose regulation, 57 with impaired glucose regulation and 146 with newly diagnosed type 2 diabetes. Based on the diagnostic criteria specified in American Diabetes Association (ADA, 2003) guidelines, categories of glucose tolerance were defined as diabetes mellitus and impaired glucose regulation.Biochemical indicators were investigated in each group. Oral glucose tolerance test was performed and 2-h postchallenge glycemia (2 h PG) was also measured. MBG was measured with continuous glucose monitoring system. HbA1c was measured with high performance liquid chromatography method. The correlations among MBG, HbA1c and the other parameters monitored were analyzed. Statistical analysis was performed using one way ANOVA, Kruskal-Wallis test, Chi-square test, Spearman correlation analysis and Linear regression analysis. Results (1)The levels of HbA1c and MBG in 318 subjects were (6.6±1.5)% and (7.3±2.3) mmol/L, respectively.(2)MBG was significantly correlated with HbA1c (r=0.848, P<0.01). Linear regression analysis, using MBG and HbA1c summarized by patient (n=318), produced a relationship of MBG=1.252×HbA1c-0.992 (R2=0.718, P<0.01). That is, MBG will increase 1.252 mmol/L with 1% increase of HbA1c. When HbA1c level was 6.5%, the expected value of MBG was 7.1 mmol/L. When HbA1c level was 7.0%, the expected value of MBG was 7.8 mmol/L.(3)In 146 patients newly diagnosed type 2 diabetes, the relationship between MBG and HbA1c was similar to that in total subjects (r=0.788, P<0.01), and the linear regression analysis showed a relationship of MBG=1.255×HbA1c-0.886 (R2=0.621, P<0.01). When HbA1c level was 6.5%, the expected value of MBG was 7.3 mmol/L. When HbA1c level was 7.0%, the expected value of MBG was 7.9 mmol/L. Conclusion We initially establish the corresponding value of MBG for a specific HbA1c target and provide the basis for translating HbA1c into mean blood glucose in Chinese subjects.