Study On HbA_1C As Criteria For Diabetes Diagnosis

Objective:Blood glucose is the main criteria for diabetic diagnosis up to now,but it is unstable and affected by factors.HbA1C not only can reflect the level of blood glucose but also has the characteristic of stablity.Hence, the aim of this study was to explore whether HbA1C could be served as criteria for diabetic diagnosis and discusse the cut point for diabetic diagnosis when using HbA1C as diagnostic criteria.Methods:The subjects who performed both OGTT and HbA1C measurement and met certain standard during March 2008 to January 2010 were collected at the Second Affiliated Hospital of Dalian Medical University. Receiver operating characteristic curve(ROC curve)was set up by using FBG, PG2h and HbA1C of some receivers and the cut point of HbA1C corresponding to the level of blood glucose used as criteria for diabetic diagnosis could be got by analyzing the ROC curve.The other receivers diagnosed as impaired glucose regulation were followed up once every 1-3 months for measuring blood glucose to identify whether the cut point of HbA1C got from the above ROC curve was suitable to be as critical point for diabetic diagnosis.During the period of follow-up,once the subject was diagnosed as diabetes according to the criteria of blood glucose, the follow-up was terminated,otherwise the follow-up were continued until to March 2010.Results:1.377 subjects were collected firstly, the cut point of HbA1C got from the ROC curve corresponding to blood glucose criteria for diabetic diagnosis was 6.3%,its sensitivity was 67.8% and specificity was 72.7% for diagnosing diabetes.2.279 followed-up subjects were collected secondedly, two groups were divided as HbA1C≥6.3% group and HbA1C<6.3% group, the prevalence of diabetes was 74.42% and 31.09%(X2=45.237,P=0.000) respectively according to the level of blood glucose during the follow-up period.The length of followe-up was 4.69±0.39 months and 4.79±0.23 months (t=-0.249,P=0.803)respectively.3.If HbA1C 6.3% was selected as the cut point for diabetic diagnosis, the capacity of diagnosing diabetes was stronger than that using FBG, but was weaker than that using PG2h,the ability of diagnosing diabetes when HbA1C combined with FBG was even still weaker than that when PG2h jointed FBG. When HbA1C 6.3% and FBG were selected as criterias for diagnosing diabetes respectively, the prevalence of diabetes was 30.82% and 15.77%(X2=22.849,P=0.000)in the 279 subjects.There were 148 cases who performed OGTT among the 279 subjects,when HbA1C 6.3% and PG2h were selected as criterias for diagnosing diabetes respectively, the prevalence of diabetes was 31.08% and 40.54%(X2=16.861,P=0.000) Combined HbA1C with FBG as criteria for diagnosing diabetes, the prevalence of diabetes was 38.51%,combined PG2h with FBG as criteria, the prevalence of diabetes was 46.62%,there was statistically significant difference between them(X2=43.266,P=0.000)4.The 377 subjects were divided into two groups according to their age and ROC curves were set up respectively.The cut point of HbA1C got from the ROC curves were 6.4% in elder than or equal to 70 years old people and 6.3% in younger than 70 years old people.The sensitivity was 66.7% and the specificity was 71.0% to the former, the sensitivity was 67.8% and the specificity was 72.7% to the latter.5.In the elder than or equal to 70 years old people among the 279 subjects,when HbA1C 6.3% and 6.4% were respectively selected as the criterias for diabetic diagnosis,the diagnostic probability of diabetes was 49.25% and 41.79%(X2=49.560,P=0.000).In the younger than 70 years old people,when HbA1C 6.3% and 6.2% were respectively selected as the criterias for diabetic diagnosis,the diagnostic probability of diabetes was 25.00% and 32.08%(X2=149.674,P=0.000).6.If HbA1C 6.4% was selected as the criteria for diabetic diagnosis in the people whose age was elder than or equal to 70 years among the 279 subjects,its capacity of diagnosing diabetes was stronger than that by using FBG and same with that by using PG2h and also similar with that by combining PG2h with FBG. The capability of diagnosing diabetes when HbA1C 6.4% combined with FBG was still same with that when PG2h jointed FBG. when HbA1C 6.4% and FBG were respectively selected as criterias for diagnosing diabetes,the prevalence of diabetes was 41.79% and 25.37%(X2=20.000,P=0.000).There were 38 cases whose age were elder than or equal to 70 years old among the 148 subjects who had performed OGTT. When HbA1C 6.4% and PG2h were respectively selected as criterias for diagnosing diabetes,the prevalence of diabetes was 42.11% and 50.00% (X2=0.432,P=0.743).When jointing PG2h with FBG was selected as criteria for diagnosing diabetes,the prevalence of diabetes was 57.89% (57.89% vs 42.11% X2=3.317,P=0.100);When jointing HbA,c with FBG was selected as criteria for diagnosing diabetes,the prevalence of diabetes was 47.37%(47.37% vs 57.89% X2=5.546,P=0.388)If HbA1C 6.2% was selected as the criteria for diabetic diagnosis in the people whose age was younger than 70 years among the 279 subjects,its capacity of diagnosing diabetes was stronger than that by using FBG and same with that by using PG2h and weaker than that by combining PG2h with FBG. The capability of diagnosing diabetes when HbA1C 6.2% combined with FBG was same with that when PG2h jointed FBG. When HbA1C 6.2% and FBG were respectively selected as criterias for diagnosing diabetes,the prevalence of diabetes was 32.08% and 12.74%(X2=7.829,P=0.000).There were 110 cases whose age was younger than 70 years old among the 148 subjects who had performed OGTT. When HbA1C 6.2% and PG2h were respectively selected as criterias for diagnosing diabetes,the prevalence of diabetes was 31.82% and 37.27%(X2=14.327,P=0.377).When jointing PG2h with FBG was selected as criteria for diagnosing diabetes,the prevalence of diabetes was 42.73%(42.73% vs 31.82% X2=17.280, P=0.000).When jointing HbA1C with FBG was selected as criteria for diagnosing diabetes,the prevalence of diabetes was 37.27% (37.27% vs 42.73% X2=28.880,P=0.327)Conclusion:1.HbA1C could be used to diagnose diabetes,the cut point for diabetic diagnosis was 6.4% in the people whose age was elder than or equal to 70 years and was 6.2% in the people whose age was younger than 70 years.2.If HbA1C was selected as the indicator for diabetic diagnosis,its capacity of diagnosing diabetes was stronger than that using FBG. HbA1C 6.4% could be used to diagnose diabetes insteading of OGTT in the elder than or equal to 70 years old people.HbA1C 6.2% could be used to diagnose diabetes insteading of PG2h in the younger than 70 years old people.3.It would be more convenient to diagnose diabetes if HbA1C was stratified as criteria for diagnosing diabetes.