Journal List > Ann Lab Med > v.35(3) > 1091280

TOOLS
Umemoto, Hoshino, Miyashita, Tani, and Kuwa: Report on HbA1c Proficiency Testing in Asia in 2012
Brief Communication
Clinical Chemistry

Annals of Laboratory Medicine 2015; 35(3): 352-355.

Published online: 1 April 2015

DOI: https://doi.org/10.3343/alm.2015.35.3.352

Report on HbA1c Proficiency Testing in Asia in 2012

Masao Umemoto, Ph.D.1, Tadao Hoshino, M.D.2, Tetsuo Miyashita, Ph.D.3, Wataru Tani, Ph.D.4, Katsuhiko Kuwa, Ph.D.5

1Reference Material Institute for Clinical Chemistry Standards, Kanagawa, Japan.

2Reference Laboratory Institute of Biopathological Medicine, Kanagawa, Japan.

3Nihon University Itabashi Hospital, Tokyo, Japan.

4Reference Material Institute for Clinical Chemistry Standards, Kanagawa, Japan.

5National Metrology Institute of Japan, Ibaraki, Japan.

Corresponding author: Katsuhiko Kuwa. National Metrology Institute of Japan, 503 Surpass Tsukuba Kenkyugakuen, 5-11-1, Kenkyugakuen, Tsukuba Ibaraki 305-0817, Japan. Tel: +81-29-852-1654, Fax: +81-29-852-1654, kkuwa@gol.com

Received 19 June 2014       Revised 22 October 2014       Accepted 7 February 2015

© The Korean Society for Laboratory Medicine.

(open-access, http://creativecommons.org/licenses/by-nc/3.0/):

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

In 2010, the Japan Diabetes Society decided to introduce the National Glycohemoglobin Standardization Program (NGSP) values into clinical practice. Accordingly, NGSP Certification of Japanese manufacturers of HbA1c-related diagnostic reagents and instruments was initiated in February, 2012, through an NGSP network laboratory, the Asian Secondary Reference Laboratory (ASRL) #1. Traceability to the NGSP reference system can be endorsed by manufacturer certification, as well as by the College of American Pathologists (CAP) survey. Nevertheless, only a few manufacturers participate in the CAP survey in Japan. Thus, proficiency testing (PT) was proposed and executed by ASRL #1. Single-donor whole-blood samples were used for the PT. The participated measurement systems were NGSP certified. Twenty-two laboratories obtained certification through ASRL #1; 2 through the Secondary Reference Laboratory (SRL) #8; and 9 through the SRL #9. The combination plots of the bias data in this PT and in the NGSP certification performed in March and May in 2012 were consistent with each other: mean NGSP values at each level agreed well with the target value. In conclusion, PT using whole blood is useful in endorsing NGSP certification.

Keywords: HbA1c, National Glycohemoglobin Standardization Program, Proficiency testing, Asian Secondary Reference Laboratory

The Japan Diabetes Society (JDS) decided to use National Glycohemoglobin Standardization Program (NGSP) values, as well as JDS HbA1c values, in clinical practice for international clinical harmonization of HbA1c [1]. Accordingly, NGSP certification of Japanese manufacturers of diagnostic reagents and instruments related to HbA1c was initiated in February, 2012, through NGSP Network Laboratory, the Asian Secondary Reference Laboratory (ASRL) #1 [2]. As a result, 21 routine methods were certified in 2012 and 42 in 2013. Moreover, traceability to the NGSP reference system can now be endorsed by both manufacturer certification and the College of American Pathologists (CAP) survey. The CAP survey is useful for monitoring between-certification proficiency of certified methods. However, only a few manufacturers participate in the CAP survey in Japan. Thus, proficiency testing (PT) similar to the CAP survey targeting NGSP-certified manufacturers and laboratories in Asia has been proposed and executed by ASRL #1 in collaboration with the Japan Reference Measurement Institute (JRMI). The committee for the PT was organized by the Reference Material Institute for Clinical Chemistry Standards (ReCCS) in collaboration with JRMI.
Five PT samples were prepared: PT samples 1, 2, 4, and 5 were single-donor whole-blood aliquots, and PT sample 3 was a mixture of PT samples 1 and 4, with plasma replaced by physiological saline solution. PT samples were drawn from two Japanese individuals in Japan and two diabetes patients in USA with informed consent and with the approval of the Ethics Committee of ReCCS. PT sample 3 for A1c GEAR in Table 1 shows abnormally high HbA1c values because only whole blood can be applied to A1c GEAR; therefore, in the mean value obtained by immunoassay, the A1c GEAR result of this sample was excluded in Table 2.
ReCCS prepared and shipped the test samples. PT samples were stored in a refrigerator (2-8℃) until shipment, and at each laboratory, they were removed from the container immediately on receipt and then stored in a refrigerator (2-8℃) until use. Analyses were completed within two days of receipt when less than 14 days passed after each sample collection.
The participated measurement systems were NGSP certified (according to manufacturer's methods and laboratory certification) by ASRL #1, SRL #8, and SRL #9. Each PT sample was measured in duplicate. Measured HbA1c values were reported to 2 decimal places. Mean values (calculated to 3 decimal places) of the duplicate values for each PT sample were reported as final values, which were compared with target values determined by ASRL #1. The obtained values were consistent with those of SRL #8 and SRL #9. An overall test of coincidence by 2 least squares linear regression lines was performed using Microsoft Office Excel 2007.
A total of 33 laboratories (29 Japanese and four overseas laboratories) participated in the PT. Among these facilities, 22 were certified by NGSP through ASRL #1, two through SRL #8, and nine through SRL #9. Furthermore, 17 participating laboratories used immunoassay, four used enzymatic assay, and 12 used HPLC. PT samples 4 and 5 yielded HbA1c values over 12 NGSP%, which exceeded measurement ranges for some measurement instruments; therefore, these values were excluded for those systems. Target values of PT samples 1, 2, 3, 4, and 5 were determined by ASRL #1 to be 4.86 NGSP%, 5.64 NGSP%, 8.00 NGSP%, 12.68 NGSP%, and 12.78 NGSP%, respectively. The comparison of values as measured by instruments and the NGSP/ASRL target values are shown in Table1.
The regression line comparing the relationship between total mean values and target values was Y (the total mean values) = 1.013×(the target value)+0.141. That is, the total mean values were higher than the target values. Furthermore, the values of CV of PT samples 1 through 5 were 3.72%, 2.54%, 2.23%, 2.41%, and 2.71%, respectively. PT sample 3 (containing red cells and saline) did not show any difference compared with other hemolyzed blood samples. CV values were 2.51-4.51% by immunoassay, 1.05-2.53% by enzymatic assay, and 1.73-2.63% by HPLC (Table 2).
Compatibility with JDS criteria (less than or equal to a relative bias of ±5.0% by mean of duplicate measurement) was checked by comparing the measured values of test samples 1 through 3 for the certification of 5% to10.0% HbA1c. The results indicate that 26 out of 28 methods in test sample 1, 26 out of 28 in test sample 2, and 26 out of 27 in test sample 3 were compatible with JDS criteria [1]. Fig. 1 shows combination plots of bias data in this study and bias data from NGSP certification through ASRL #1 conducted during March and May, 2012. Bias plots of mean values versus target values of PT samples in this report are also shown for comparison, and they are in good agreement with each other.
The determined target values using the Central Primary Reference Laboratory reference panels (100 samples) correspond with measured values obtained by SRL #3 and SRL #9. In conclusion, the PT is useful to monitor NGSP-certified methods. Mean values at each level correspond well with the target value, including point-of-care testing.

Acknowledgments

We are grateful to Dr. R. Little, University of Missouri School of Medicine, for giving us the data of NGSP network Laboratory Monitoring for the measurement of HbA1c PT samples by the NGSP Laboratory Network.

Notes

Authors' Disclosures of Potential Conflicts of Interest: No potential conflicts of interests relevant to this article were reported.

References

1. Committee of the Japan Diabetes Society on the Diagnostic Criteria of Diabetes Mellitus. Seino Y, Nanjo K, Tajima N, Kadowaki T, Kashiwagi A, et al. Report of the Committee on the classification and diagnostic criteria of diabetes mellitus. J Diabetes Investig. 2010; 1:212–228.
crossref
2. NGSP Network Laboratory Members. Updated in 2010. http://www.ngsp.org/network.asp.
Fig. 1

Observed HbA1c values of each participating laboratory compared to the NGSP/ASRL target (dashed line) in 2012 based on ASRL PT data. Closed circles show within ±5% (dotted line) of relative bias. Open circles show within ±7% (dashed line) of relative bias. Open squares show biases of all manufacturers versus ASRL target PT samples.

Abbreviations: NGSP, National Glycohemoglobin Standardization Program; ASRL, Asian Secondary Reference Laboratory; PT, proficiency testing.
alm-35-352-g001
Table 1

Participating laboratories and supplemental data HbA1c unit : NGSP%

alm-35-352-i001
Method Participating laboratory Measurement system PT sample
1 2 3 4 5
Immunoassay Kyowa Medex Co., Ltd. Determiner HbA1c/DM-JACK 4.755 5.485 7.895 12.530 12.560
Immunoassay Kyowa Medex Co., Ltd. Determiner HbA1c/JCA-BM9130 4.805 5.500 7.940 12.565 12.625
Immunoassay Kyowa Medex Co., Ltd. Determiner L HbA1c/DM-JACK 4.795 5.520 8.005 12.540 12.625
Immunoassay Kyowa Medex Co., Ltd. Determiner L HbA1/JCA-BM9130 4.795 5.480 7.985 12.485 12.570
Immunoassay TFB, INC. Rapidia Auto HbA1c-L/Hitachi 7170s 4.570 5.400 7.660 12.200 12.420
Immunoassay ROHM Co., Ltd. Banalyst Ace HbA1c/Banalyst Ace 4.700 5.500 7.900 High* High*
Immunoassay ROHM Co., Ltd. Spotchem Banalyst HbA1c/Spotchem Banalyst SI-3610 4.800 5.600 8.000 High* High*
Immunoassay Wako Pure Chemical Industries, Ltd. Autokit HbA1c/Hitachi 7170S 4.790 5.440 8.075 12.585 12.535
Immunoassay SAKAE Corporation A1c GEAR 4.650 5.300 8.650† 12.050 12.500
Immunoassay Kotobiken Medical Laboratories, Inc. Biken Central Laboratory Tsukuba Determiner L HbA1c/JCA-BM9130 4.830 5.360 7.660 12.205 11.965
Immunoassay Siemens Healthcare Diagnostics DCA2000+ / DCA Vantage 4.950 5.600 8.150 13.050 13.300
Immunoassay Siemens Healthcare Diagnostics Dimension RxL MAX 5.400 6.065 8.440 13.190 13.180
Immunoassay Ortho Clinical Diagnostics Vitros d%A1c/Vitros 5,1FS 4.970 5.475 8.125 12.735 12.780
Immunoassay Ortho Clinical Diagnostics Vitros d%A1c/Vitros 5600 4.980 5.475 8.110 12.605 12.585
Immunoassay Roche Diagnostics TQ HbA1c Gen.3/cobas c501 4.650 5.475 8.275 12.495 12.590
Immunoassay Roche Diagnostics TQ HbA1c Gen.2/cobas c501 5.095 5.645 8.135 12.280 12.335
Immunoassay Boditech Med. Inc i-ChromaTM 4.450 5.600 8.050 12.150 12.450
Enzymatic assay Kyowa Medex Co., Ltd. MetaboLead HbA1c/JCA-BM9130 4.850 5.440 7.910 12.880 13.065
Enzymatic assay SEKISUI MEDICAL CO., LTD. Norudia N HbA1c/Hitachi 7170S 4.855 5.715 7.690 12.730 13.420
Enzymatic assay Hitachi Chemical Co., Ltd. Seratestum A1C 4.780 5.465 8.085 12.555 12.855
Enzymatic assay NIHON KOHDEN CORPORATION BM Test HbA1c/JCA-BM6010 4.870 5.445 8.140 13.205 13.315
Ion-exchange HPLC Kotobiken Medical Laboratories, Inc. Niigata Laboratory HLC-723 G8 4.740 5.460 7.920 12.690 12.840
Ion-exchange HPLC SRL, Inc. ADAMS A1c HA-8160 4.900 5.650 8.200 12.800 13.100
Ion-exchange HPLC Bio-Rad Laboratories Variant II Turbo 4.635 5.535 7.875 12.530 12.675
Ion-exchange HPLC Bio-Rad Laboratories D-10 4.800 5.600 7.950 12.500 12.600
Ion-exchange HPLC Tosoh Corporation HLC-723 G7 4.710 5.510 7.975 12.735 12.885
Ion-exchange HPLC Tosoh Corporation HLC-723 G8 4.680 5.500 7.965 12.805 12.865
Ion-exchange HPLC Tosoh Corporation HLC-723 G9 4.700 5.525 7.995 12.855 12.910
Ion-exchange HPLC Tosoh Corporation HLC-723 GX 4.695 5.495 7.980 12.835 12.890
Ion-exchange HPLC Korea Association of Health Promotion HLC-723 G8 4.740 5.585 8.100 13.035 13.120
Ion-exchange HPLC Seoul National University Bundang Hospital Variant II Turbo 4.600 5.600 8.050 13.100 13.400
Ion-exchange HPLC NEODIN MEDICAL INSTITUTE HLC-723 G8 4.900 5.750 8.250 13.150 13.250
Ion-exchange HPLC Chung-Ang University Hospital Variant II Turbo 4.450 5.300 7.650 12.500 12.600

*out of measurement range; †abnormally high values (whole blood only can be applied) for sample 3 (mixture of samples 1 and 4, replacing the plasma with physiological saline solution).

Abbreviations: NGSP, National Glycohemoglobin Standardization Program; PT, proficiency testing; ASRL, Asian Secondary Reference Laboratory; ReCCS, Reference Material Institure for Clinical Chemistry Standards.

Table 2

Comparison of three methods: immunoassay, enzymatic assay, and HPLC HbA1c unit: NGSP% for target value, Mean and SD

alm-35-352-i002
Sample 1 2 3 4 5
Target value 4.863 5.643 7.997 12.678 12.779
Method Immunoassay (17 measurement systems)
Mean 4.823 5.525 8.025 12.511 12.601
SD 0.217 0.165 0.199 0.317 0.317
CV (%) 4.51 2.99 2.47 2.53 2.51
Method Enzymatic assay (4 measurement systems)
Mean 4.855 5.552 7.997 12.843 13.164
SD 0.051 0.140 0.198 0.239 0.220
CV (%) 1.05 2.53 2.48 1.86 1.67
Method HPLC (12 measurement systems)
Mean 4.713 5.543 7.993 12.795 12.928
SD 0.124 0.110 0.155 0.222 0.250
CV (%) 2.63 1.99 1.94 1.73 1.93

Abbreviations: NGSP, National Glycohemoglobin Standardization Program.

TOOLS
Similar articles