This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Eurolyser Diagnostica GmbH, Bayernstrasse 11a, 5020 Salzburg, Austria
EVALUATION of SMART HBA1c
Author: Michael Gruber Eurolyser Diagnostica Bayernstrasse 11-A 5020 Salzburg AUSTRIA Co-Author: Dr. Xiaoru Chen Diazyme Laboratories A Division of General Atomics 12889 Gregg Ct. Poway, CA 92064 USA Date: 05/23/2008
1. Scope and goals of the evaluation 1.1 Method comparison Testing the correlation between the HBA1c measurement results in the SMART 700/340 analyzer from whole blood EDTA (Sarstedt pipettes) and the results of the clinical-chemical fully automated Cobas 6000 from Roche, with its HBA1c test kit “Tina Quant A1C-2”
1.2 Imprecision Characterization of the precision of the SMART HbA1c production model test at various HbA1c concentrations 1.3 Day/Day variation according to NCCLS EP10 protocol
1.4 Effect of hemoglobin variants Test of susceptibility to failure of hemoglobin variants by the Diazyme laboratory in the USA with the use of the enzymatic test that is also used in SMART. 2. Samples, reagent kits, and consumab les Samples: Samples taken from the daily routine of 3 days (04/08/2008 to 04/10/2008), and no longer needed, were used for the method comparison. The blood was anonymous and packed in Sarstedt EDTA pipettes. Material and methods: SMART 700/340 analyzers with the following serial numbers, which happen to have been taken from the current production lot: SMART1: SNO Aa0249 SMART2: SNO Ab0709 SMART3: SNO Ab0711 SMART4: SNO Ab0712 Reagent kits: 2 packages of article ST0110, with charge no. 100108-1 and expiration date 01/2009, were taken from the current production charge. To calibrate this production charge, the calibration kit provided by Diazyme Laboratories, with charge no. HBS100108 (Calibrator 1) and HBS200108 (Calibrator 2) was used. QC materials To check the correctness, the QC kit provided by Diazyme Laboratories, charge no. HBC00507, was used.
3 . Test processing with the SMART system: The reagent kit contains all the materials required to carry out the tests. 3.1 Sample Preparation Step-Lysation: 2 pump strokes of the lysation reagent were pumped into the brown-colored Eppendorf cup, using the pump dispenser included in the test kit. Pipetting of 20µl whole blood from the Sarstedt tubes, into the brown Eppendorf cup. Vigorous mixing in the Eppendorf cup, and 6 minutes wait time for lysation Using the Dual Bulb Pastette, 150µl of the lysate were transferred from the Eppendorf cup into the reagent tube=cuvette. Set the R2 cap in place and measure with the SMART photometer 4. Analysis: The raw data were documented in HbA1c values and mAbs values from the SMART photometer, and entered into an Excel spreadsheet.
Method comparison : The method comparison was carried out with 33 samples from the pool of the hospital “Krankenhaus der barmherzigen Brüder” in Salzburg, in the range of 5.4% to 10.9% HBA1c. The R² value of the linear regression was determined, as well as the k and d value, according to the formula y=kx+d (y=SMART HbA1c and x= Roche Tina Quant HbA1c) The chart and raw data are in the appendix. The samples were alternately measured with all 4 Smart units, so that an approximate device distribution was also taken into consideration for the test.
Chart 1: Correlation between SMART HBA1c (y) and Roche Tina Quant (x) R²=0.9415 Y=kx+d=1,3135x-2,3573 Bias 95% CI Constant -2.43 -3.37 to -1.41 Proportional 1.32 1.16 to 1.44
4.1.1 Deviation of the SMART HBA1c value, recalculated for a calibration to the Roche Tinaquant:
The table shows the SMART HBA1c values calculated as if the Diazyme calibrators had not been used to calibrate the SMART units, but rather the Roche TinaQuant values of the patient blood had been used.
Chart 2: Correlation between SMART HBA1c recalibrated (y) and Roche Tina Quant (x) R²=0.9415 Y=kx+d = 0.9826x+0.5689 Bias 95% CI Constant -0.03 -0.74 to -0.75 Proportional 1.00 0.88 to -1.09
5. Imprec ision : The imprecision of the SMART HAB1c device was determined externally, at the laboratory of Diazyme Laboratories, by Dr. Xiaoro Chen. The values were measured with 3 different SMART units that were selected at random from the device pool at Diazyme. SMART1: SNO Aa0315 SMART2: SNO Aa0316 SMART3: SNO Ab0596 The measurement series were carried out on 03/27/2008, the raw data were listed in 10 x mAbs values (activity) and the HbA1c values.
Level1 Level2 Assigned values at Diazyme Labs USA mean of variance ( R ) 0,027 0,135 Level 1 5,50%
variance MV (S) 0,046 0,026 Level 2 9,50% T = (S)-R/3 0,038 -0,019
Total imprecision Variance U = (R+T) 0,064 0,116
Total Imprecision Standard Deviation V (Sqrt U) 0,253 0,341
Total MV 5,24 % 9,58 % Total Imprecision cv (%) 4,832 3,557
6. Evaluating the in fluence of hemoglob in variants There are significant differences in precision and specificity of the HbA1c measurement methods currently on the market The enzymatic method, as used in SMART, has demonstrably the best specificity.
This results from the fact that the “glycated valine” of the HbA1c is split in the enzymatic method. Table 4: Effects of chemically modified hemoglobin variants on known measurement methods:
Table 5: Effects of genetically determined hemoglobin variants:
Modification Methods Affected Meth ods Unaffected Hb C Immunoassay,HPLC assay, up to 40% errors Smart HbA1c Hb S Immunoassay,HPLC assay, up to 40% errors Smart Hba1c assay <10%
errors Hb Graz Immunoassay,HPLC assay, up to 30% errors Smart HbA1c Hb Sherwood Forest
Immunoassay,HPLC assay, up to 30% errors Smart HbA1c
Hb D Immunoassay,HPLC assay, up to 30% errors Smart HbA1c Hb O padova
Immunoassay,HPLC assay, up to 30% errors Smart HbA1c
It is to be demonstrated that the “direct enzymatic” measurement method used in the SMART HAB1c from DIAZYME is not affected by hemoglobin variants. This test was carried out on 01/03/2007 by Dr. Abhijit Datta of Diazyme Laboratories in the USA. Samples from the University of Missouri that were known to have HB variants were measured on 3 different analysis systems. The PRIMUS HPLC system is considered to be the standard method.
Table 4: Measurement of blood samples with known hemoglobin variants HbC HbE and HbS
7. Summary:
The direct enzymatic method used in the SMART analyzer has clear advantages regarding the sensitivity to the HBA1c parameter, relative to hemoglobin variants. It is therefore difficult to compare this very new test with HbA1c test systems that have been on the market for a long time using any kind of method comparison, since most of the typical methods are affected by sensitivity to errors, and there could therefore be blatant differences for samples with hemoglobin variants. If one assumes a population of patient samples that are free of HGB variants, then the SMART HbA1c is most comparable to the “Roche Tina Quant Gen2” test. The imprecision in the series, with cv values of 1.87 to 3.02%, is excellent for a point-of-care system.
8. Discussion of any problems: Users must take part in round-robin tests in a specific country. “Blind samples” are sent to labs as lyophilisates. Since the enzymatic HbA1c test is very new, there will not yet be any association in the correct data pool, which is why the system can probably not participate in typical round-robin tests without further effort. QC material, such as is used in typical systems, is not comparable with the enzymatic test. The round-robin test locations would actually have to provide a separate QC material for enzymatic HbA1c tests. This is available from Diazyme Laboratories in the USA, and NGSP certified. After the manufacturer of the SMART Hba1c reagent was able to convince well-known diagnostics companies of the enzymatic test, it can be assumed that, in the near to medium term, this test will also have a significant market volume in large laboratories, and that the round-robin test sites will then adjust to this method. 05/23/2008 Salzburg