Evaluation Of Five Different Methods For Measurement Of HbA1c

Objective:Glycosylated hemoglobin(GHb),an effective detection index designed not only to diagnosis diabetes mellitus(DM), but also to evaluate blood glucose detection and curative effect. In clinic, Hb A1 c always replaces GHb. There are many ways to measure Hb A1 c, as high performance liquid chromatography( HPLC), electrophoresis, boronic acid affinity chromatography, enzymatic and latex agglutination method. HPLC for the testing of Hb A1 c is recognized as the gold standard for monitoring blood glucose in diabetes patients because of its accuracy and repeatability improved greatly. The four other methods are compared and evaluated to HPLC in this paper, in order to offer a reference for clinicians and laboratory to select different methods for measurement of Hb A1 c. Methods:The experimental samples are mainly from the subjects of physical examination center, out-patient and in-patient ward in department of endocrinology and department of internal medicine. Specimens from 129 patients were divided into two groups according to with or without diabetes: normal control group of 65 cases without diabetes, male 33 cases, female 32 cases, aged 44-57 years old, with body mass index 18-24kg/m2; 64 cases of DM patients, male 33 cases, female 31 cases, aged 45-66 years old, with body mass index 18-24kg/m2. Each specimen was tested with five kinds of methods to detect Hb A1 c in the same conditions, and then the results were recorded for comparative analysis. 5 samples were used respectively from normal and abnormal samples on the basis of medical decision level. The within-run precision test was repeated 20 times in one day and the between-run precision detection of 10 tests were conducted in 7 d. Average( X),standard deviation(s) and coefficient of variation(CV) of within-run precision and between-run precision were calculated. The sample of Hb A1 c 10.0 was diluted with normal saline into 5 samples of concentration gradient for linear regression analysis from low to high density to observe the linear range. The sample of Hb A1 c 10.0 was diluted with normal saline into 1/5、2/5、3/5、4/5、5/5 of concentration gradient and the samples were tesed with five kinds of methods to detect Hb A1 c. Add glucose 56mmol/L, triacylglycerol 25mmol/L, bilirubin 3.4mmol/L to blood samples preparation respectively or do the physical hemolysis by low temperature freezing thawing to mild hemolysis(Hb<1g/L), moderate hemolysis(Hb 1~4 g/L), severe hemolysis(Hb>5g/L). The value of Hb A1 c under the influence of various interferences in different concentration, if the value of Hb A1 c vary 0.3%, will be regarded as interference. Compare the amount and the time of detecting Hb A1 c by five methods, and the regression equation and correlation coefficient were also compared. Results:In five methods, except the boronic acid affinity method, the other four methods are consistent with the requirements. Comparatively the precision of D-10 automatic glycosylated hemoglobin analyzer in HPLC was better. The data of five different methods from 4.0% to 14.2% was a good linear and the correlation coefficient(r) of measured value and theoretical value was in the range of 0.971 to 0.996. In the five methods HPLC was the best. When the interfering substance was add into the samples, Hb A1 c values measured by HPLC, enzymatic and latex agglutination method fluctuated <0.2%. So these three methods were stable performance and not easy to be affected by a variety of chemicals. The electrophoresis was particularly sensitive to impact of hemolytic. In boronic acid affinity chromatography method, when the hemolysis concentration fell too low or the lipidemia concentration rised too high, the value was barely detectable on the Nyco Card Reader II multifunctional total quantitative special protein analyzer. Through comparison between the numbers of samples, test time, CV and R index, the numbers of samples and test time in the electrophoresis were higher than those in other methods.The R in enzymatic is closest to 1. Conclusion:Within-run precision and between-run precision of HPLC, which was used as the gold standard, are the best in the five methods. The second is enzymatic method and the latex agglutination method. Yet repeatability of boronic acid affinity method is relatively poor. Electrophoresis results are related to scanning and electrophoretic peaks judgment by researchers, which contains more subjective arbitrariness and need laboratory staffs skilled and experienced, otherwise the results have a great difference. HPLC has the best correlation with the theoretical value, which has the best wide linear range. The linear graph of boronic acid affinity method presents S shape slightly. Where the concentration is higher, there is greater to deviate. The reason may be that the test result is total Hb A1c(means all the glycosylated hemoglobin). Interfering substance can not interfere with the HPLC method, enzyme method and latex agglutination method, which three methods show the stable performance and strong anti-interference ability. These three methods had less influenced by the interferential substance and have higher stability. Electrophoresis is affected by hemolysis a lot. Boronic acid affinity method was affected by interferential substance most obviously. When the same sample is used by five methods at the same time, it was enzymatic method which is closest to HPLC.