| Objective American Diabetes Association officially regarded Hb A1c≥6.5% as diagnostic criteria for diabetes in 2010. However, the Hb A1 c detection technology standards ware different across countries, racial differences ware existed in the world, Hb A1c6. 5% was not the diagnostic criteria for diabetes in China. This study aimed to investigate the optimal glycosylated hemoglobin Alc cut-off points on screening for diabetes in this region.Method Selected January 2012 to December 2014 in Kailuan General Hospital, Located in Tangshan City, Hebei Province, 18 to 75 years old in all outpatient, who selected 75 g oral glucose tolerance test(OGTT)and Hb A1 c determination, a total of 613 cases, with an average age of 55.07±10.89 years, 368 cases of male, and 245 cases of female. Grouping principles refered to the gold standard in 1999 WHO diagnostic of diabetes, PG2h≥11.1mmol/L for the diabetic group, a total of 494 cases, aged 54.59±11.14 years old, male, female were 305 cases and 189 cases. PG2 h <11.1mmol/L for non-diabetic group, a total of 119 cases, aged 57.04±9.54 years old, male, female were 63 cases and 56 cases. Hexokinase method was used to tast blood glucose detection method of OGTT and high pressure liquid chromatography method was used to tast glycated hemoglobin. The optimal Hb Alc cut-off point for screening diabetes mellitus was identified with the ROC curve.Results 1 Frequency analysis showed that about 20% of patients in diabetic group with Hb A1 c <6.5%, and about 80.6% of the patients in non-diabetic group with Hb A1 c <6.25%. 2 Glycosylated hemoglobin A1 c levels rised in diabetic group, the difference was statistically significant(P<0.05). 3 Glycosylated hemoglobin A1 c and 2-hour postprandial blood glucose was significantly correlated(r=0.732,P<0.05). 4 Glycated hemoglobin A1 c as the dependent variable, 2-hour postprandial blood glucose as the independent variable, the regression equation was Hb A1c=3.031+0.290×PG2h(%), when the PG2h=11.1mmol/L, Hb A1c=6.25%. 5 2 hours postprandial glcose≥11.1mmol/L as the gold standard for the diagnosis of diabetes, draw receiver operating characteristic curve(ROC curve) from the use of all subjects glycosylated hemoglobin A1 c, the result showed that the area under the curve was 0.922(95 % confidence interval is 0.898~0.946,P<0.05), the optimal cut-off point of glycated hemoglobin A1 c was 6.25%(sensitivity is 87.85% and specificity is 80.67%). 6 When Hb A1c=6.25%, the sensitivity was 87.85%, specificity was 80.67%, misdiagnosis rate was 12.15%, misdiagnosis rate was 19.33%, the positive predictive value was 94.97%, the negative predictive value was 61.54%, the positive likelihood ratio was 4.54, negative likelihood ratio was 0.15, Youden index was 0.69, the Kappa value was 0.6. 7 The sensitivity and specificity of the test results showed that Hb A1c6.25% had a maximum youden index, glycosylated hemoglobin A1 c increased, decreased sensitivity and specificity. Description the higher glycated hemoglobin value, the higher rate of misdiagnosis, but the misdiagnosis rate would be reduced accordingly.Conclusions 1 The screening efficiency of Hb A1 c ≥ 6.25% for the screen of diabetes is 94.97%, it is suitable to be the cut-off point for early screen of diabetes mellitus. 2 Hb A1c≥6.25% for the screen of diabetes, the sensitivity is better than Hb A1c≥6.50%, and reduces the rate of misdiagnosis. |
PDF Full Text Request