The National Glycohemoglobin Standardization Program: Over 20 Years of Improving Hemoglobin A1c Measurement

Abstract

Background: Measurement of hemoglobin A1c (HbA_1c) in the blood is integral to and essential for the treatment of patients with diabetes mellitus. HbA_1c reflects the mean blood glucose concentration over the preceding 8 to 12 weeks. Although the clinical value of HbA_1c was initially limited by large differences in results among various methods, the investment of considerable effort to implement standardization has brought about a marked improvement in analysis.

Content: The focus of this review is on the substantial progress that has been achieved in enhancing the accuracy and, therefore, the clinical value of HbA_1c assays.

Summary: The interactions between the National Glycohemoglobin Standardization Program and manufacturers of HbA_1c methods have been instrumental in standardizing HbA_1c. Proficiency testing using whole blood has allowed accuracy-based assessment of methods in individual clinical laboratories that has made an important contribution to improving the HbA_1c measurement in patient samples. These initiatives, supported by the efforts of the IFCC network, have led to a continuing enhancement of HbA_1c methods.Many of the factors that previously influenced HbA_1c results independently of blood glucose have been eliminated from most modern methods. These include carbamylation, labile intermediates, and common hemoglobin variants. Nevertheless, some factors (e.g., race and aging) may alter HbA_1c interpretation, but whether these differences have clinical implications remains contentious. HbA_1c has a fundamental role in the diagnosis and management of diabetes. Ongoing improvements in HbA_1c measurement and quality will further enhance the clinical value of this analyte.

Figure 1.

NGSP Network and process: CPRL, Central Primary Reference Laboratory; PRL, Primary Reference Laboratory; SRL, Secondary Reference Laboratory; IFCC, International Federation of Clinical Chemistry

Figure 2.

Mean HbA1c measured by the CPRL for nine quality control specimens. Each point represents the mean during each year of use. For each point the CV was <3% (4.4% IFCC).

Figure 3.

The 2019 NGSP certification criterion (36/40 results within ±5%) compared to the proposed 2020 CAP criterion of ±5%. The lines represent the bias and CV combinations required to pass the NGSP (solid lines) and CAP (dashed lines) criteria with 0.95 (top, light gray), 0.99 (middle, dark gray) and 0.999 (bottom, black) probabilities.

Figure 4.

Improvement in HbA1c over time. CAP data from a single sample in 1993, 1999, 2004, 2010, 2014 and 2018. The dashed line is the NGSP/DCCT Reference value for that sample. Each point is the mean of a single method; bars are 2SD ranges. Results were reported as HbA1c ▇, HbA1 ◆, and Total GHB ●.

Figure 5.

All method CVs over time in CAP samples from 2000 to 2018. A. Samples with target values; A. 4–6% (20–42 mmol/mol) HbA1c, B. 6–8% (42–64 mmol/mol) HbA1c and C. 8–10% (64–86 mmol/mol) HbA1c. Note that the NGSP/IFCC %CV relationships on the y-axes are different for each graph since the conversion is based on specific HbA1c levels. The HbA1c levels used for the conversion are 5% (31 mmol/mol), 7% (53 mmol/mol) and 9% (75 mmol/mol) for A, B, and C, respectively (10).