American Association for Clinical Chemistry
Better health through laboratory medicine
February 2011 Clinical Laboratory News: HbA1c
CLN Banner Logo

February 2011: Volume 37, Number 2

An Overview of Current Analytical Testing Issues 

By Randie R. Little, PhD and Curt L. Rohlfing

The global prevalence of diabetes mellitus continues to increase rapidly, with more than 280 million people diagnosed worldwide. By 2030, experts predict that the number of people with diabetes will soar to 439 million. Measuring glycated hemoglobin (GHB) has long been fundamental to managing patients with diabetes. Reported by clinical laboratories as HbA1c, this marker is not only used to monitor long-term glycemic control, but also to adjust therapy, assess quality of care, and predict risk for the development of complications.

The results of the landmark Diabetes Control and Complications Trial (DCCT) and its continuation as the Epidemiology of Diabetes Interventions and Complications (EDIC) Trial, as well as the United Kingdom Prospective Diabetes Study (UKPDS) (1–3), collectively underscored the importance of HbA1c as a disease management tool. These two prospective long-term randomized trials conclusively demonstrated that intensive glycemic control significantly reduces the risk of long-term diabetes complications and allowed diabetologists to establish specific treatment goals based on HbA1c. For example, the American Diabetes Association (ADA) recommends that HbA1c levels be routinely obtained in all patients with diabetes at least two to four times per year and that non-pregnant adults should aim for HbA1c levels <7% (4). Other clinical organizations, worldwide, have also set treatment goals for HbA1c. More recently, ADA recommended using HbA1c levels to diagnose diabetes. These new recommendations set HbA1c levels >6.5% for diagnosis and levels 5.7–6.4% as an indication of increased risk for diabetes (4).

With this heightened attention on HbA1c for optimal diabetes treatment and diagnosis, it is imperative that clinical labs have accurate and reliable methods to measure the marker. This article will describe the current state of and efforts to improve HbA1c testing and discuss analytical issues laboratories need to consider for optimal assay performance.

Glycation of Hemoglobin

Glycated hemoglobin is derived from the non-enzymatic addition of glucose to amino groups of hemoglobin (Figure 1). HbA1c is a specific species of glycated hemoglobin resulting from attachment of glucose to the N-terminal valine of the hemoglobin β-chain (5). Its concentration is proportional to that of other glycated hemoglobins (glycation on the N-terminus of the α-chain or the ε-amino groups of lysine residues); therefore, methods that measure total GHB easily can be standardized to report % HbA1c.

Figure 1 
Nonenzymatic Glycation of Hemoglobin

Click for figure

Glucose reacts with an amino group on hemoglobin to form an aldimine linkage and subsequently undergoes an Amadori rearrangement to form a stable ketoamine (glycated hemoglobin).

The concentration of HbA1c, as well as other glycated hemoglobins, depends on both the concentration of glucose in blood and the lifespan of erythrocytes. Because these cells circulate in blood for approximately 120 days, HbA1c levels actually represent the integrated glucose concentration over the preceding 8–12 weeks (6). Long-term assessment of glycemia is advantageous not only because it eliminates the large fluctuations that occur daily in blood glucose concentrations, but in contrast to glucose measurements, HbA1c also provides an accurate result from blood drawn at any time of day without reference to prandial state. This is especially useful for both diabetes screening and diagnosis.

Measurement of HbA1c

Clinical laboratories use a wide variety of analytical techniques to separate and quantify HbA1c. These can be divided into two categories: separation by molecular charge and separation by molecular structure.

The first category includes cation-exchange chromatography, mostly high-performance liquid chromatography (HPLC), that specifically separates and quantitates levels of HbA1c. HbA1c is less positively charged at neutral pH than HbA0 and therefore does not bind as well to the negatively charged resin. Spectrophotometric analysis of the various eluants provides the percentage of each hemoglobin species in the patient sample. This type of analysis typically yields excellent assay precision and rapid hemoglobin separation.

Separation of total glycated hemoglobin or HbA1c based on molecular structure can be achieved by either immunoassay or boronate affinity chromatography. The immunoassays use monoclonal or polyclonal antibodies that recognize some number of amino acids at the amino terminus of the β-chain plus the glucose residue to quantitate HbA1c. The final concentration is calculated as a ratio of HbA1c to total hemoglobin.

The boronate affinity binding methods rely upon the strong interaction of the coplanar cis-diol groups on glycated hemoglobin with boronic acid. Agarose gels with immobilized boronic acid are used for the chromatography methods—mainly HPLC. The non-glycated hemoglobins elute directly from the agarose gel while glycated hemoglobins bind and require a counterligand for elution. This method measures total glycated hemoglobin, including HbA1c and Hb glycated at ε-amino groups on lysine residues; however, it also can be calibrated using HbA1c-specific assigned values to produce HbA1c-equivalent values. More recently, enzymatic assays for HbA1c have become commercially available, although they are not yet in widespread use.

Standardization of HbA1c Measurements

When the DCCT ended in 1993, HbA1c assay methods were not standardized among methods or laboratories, making it difficult for physicians and patients to relate test results to DCCT/UKPDS-based treatment goals. Given the positive impact that standardization of HbA1c determinations would have on the care of diabetic patients, the AACC Standards Committee established an HbA1c Standardization Subcommittee to create a standardization protocol. Subsequently, the National Glycohemoglobin Standardization Program (NGSP) was created in 1996 to implement the protocol developed by the AACC group (7).

NGSP has taken a pragmatic approach to standardizing HbA1c, using the DCCT reference method as the designated comparison method and establishing a network of primary and secondary reference laboratories. In 2001, a higher order reference method, developed by a working group of the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC), was approved and an IFCC HbA1c Laboratory Network was established (8). Continuous monitoring within and between the NGSP and IFCC laboratory networks provides security for the stability of reported HbA1c results worldwide. Even though some countries will soon report HbA1c in IFCC numbers and units (mmol/mol) and others, including the U.S., will continue to report in NGSP/DCCT numbers and units (%), there is an established linear relationship between them allowing conversion from one to the other using a published master equation, NGSP (%) = (0.0915 x IFCC (mmol/mol)) + 2.15 (9).

Today, most methods for measuring HbA1c are NGSP-certified. According to the NGSP manufacturer certification criteria, this means that for a panel 40 samples analyzed in duplicate under ideal conditions, the 95% confidence interval of the differences between the method being tested and a NGSP SRL is within +/–0.75% HbA1c.

The NGSP website contains a list of certified methods along with detailed information about the NGSP. Over the years, the certification criteria have become tighter to encourage improvement in the methods. But it is also important to examine proficiency testing results as an indication of how a method performs in routine clinical laboratories because method certification is performed by the manufacturer. The biannual College of American Pathologists’ (CAP) GH-2 survey for HbA1c includes more than 2,000 laboratories and uses fresh, whole blood specimens. Each specimen has a value assigned by the NGSP laboratory network that is used for accuracy-based grading.

Recently, CAP tightened the grading criteria for this survey. The criterion for 2010 was ±8% of the NGSP-assigned value; the criterion for 2011–12 will be ±7%. Comparing the specific method mean with the NGSP-assigned value is an indication of the average bias of a method during the time of the survey. The between-laboratory imprecision (CV) for each method is also listed on the survey reports. The combined bias and CV (total error) for each method determines the overall pass rate for each method on the survey. The pass rate on the most recent survey, for all methods combined, exceeded 95%; the pass rates by method ranged from 81–100%.

The pass rates for 2011, using ±7% as the cutoff, are expected to be a bit lower. The tighter criterion is intended to encourage manufacturers to improve methods even further and to encourage individual laboratories to take steps to improve their performance. For example, laboratories could take steps to improve their QC methods, ensure optimal instrument performance, and communicate problems to manufacturers.

Improvement in HbA1c Measurements

The combined effect of tightening the NGSP certification criteria and the CAP survey criteria has been to improve the overall performance of HbA1c methods. Figure 2 shows the improvement in laboratory performance from 1993 (when the DCCT ended) through 2010. In 1993, half of laboratories reported results as HbA1c while the other 50% reported results as either HbA1 or total GHB. As more laboratories started reporting % HbA1c, methods showed lower CVs and bias. In 2010, labs could only report HbA1c on the survey and most results were close to assigned values.

Figure 2
Laboratory Performance Data from the CAP GH-2 Survey

Click for figure

Each point and bar represents the mean result ±2SD, respectively, of laboratories using a particular method. The horizontal dotted line represents the NGSP/DCCT target value.

Symbols: ? results reported as HbA1c, ♦ results reported as HbA1, ? results reported as total GHB.

Clearly, there is less bias today between methods and also less variability between laboratories within each method group. The all-method CV for the normal range CAP sample in 2000 was >7%; by 2010 the CV fell to <4%. Remarkably, a few individual methods can provide within-laboratory CVs <1% and between-laboratory CVs <2%.

Low CVs and bias facilitate optimal clinical use of HbA1c in both diabetes monitoring and diagnosis. When evaluating HbA1c assay methods, laboratorians should consider not only each method’s NGSP-certification status but also its performance on the CAP survey to ensure that they deliver accurate lab results.


HbA1c provides an excellent measure of glycemic control for the vast majority of patients with diabetes and is also an excellent tool for diabetes diagnosis; however, measuring this marker may be unreliable in some situations. For example, disorders that affect the lifespan of the erythrocyte, such as hemolytic anemia, Sickle cell disease, or polycythemia, will falsely raise or lower HbA1c results. Severe iron-deficiency anemia has also been shown to interfere with HbA1c results. Some hemoglobin variants like HbS trait or hemoglobin adducts such as carbamylated hemoglobin can adversely affect HbA1c results depending on the analysis method. In addition, researchers have reported that other factors such as race or age influence HbA1c results, albeit to a small extent.

The most common interference in HbA1c analysis is from Hb variants. The four most common variants worldwide are HbS, HbE, HbC, and HbD. HbF, normally present at low levels in adults, may be increased in cases of hereditary persistence of fetal hemoglobin or diseases such as thalassemias or some types of leukemia. Interferences from heterozygous hemoglobin variants or elevated HbF are generally method-specific. Studies have evaluated most assay methods with respect to potential interference from the most common Hb variants, and a list summarizing these findings can be found on the NGSP website.

Currently, there are still a few immunoassay methods that show interference from HbAS and HbAC, although most do not. Because there is no way to detect the presence of Hb variants by immunoassay, the lab is likely to report an inaccurate result.

There are also a few ion-exchange HPLC methods that show interference from HbAE and HbAD. With these methods, the presence of an interfering variant can usually be seen on the chromatogram, but in some cases the indication is very subtle and can be overlooked.

Laboratorians should know the limitations of their method with respect to interference from the most prevalent Hb variants and should consider the prevalence of potentially interfering hemoglobin variants in their patient population when selecting an HbA1c assay method. As with any test, HbA1c results that do not fit the clinical impression must be evaluated further.

Working Towards Better Diabetes Care

Laboratory analysis of HbA1c is integral to the management of diabetes and has now been recommended for diagnosis of the disease. The variability among methods and laboratories continues to decline due to the combined efforts of the NGSP, CAP, and manufacturers of HbA1c methods. Surveillance by NGSP and IFCC will ensure that lab results continue to relate both to clinical outcome measures like the DCCT/UKPDS results and clinical guidelines and to reference methods of higher metrological order. Furthermore, recognizing potential interferences will enable labs to report accurate results by appropriate methods and better select methods that are suitable for their specific patient populations.


  1. DCCT Research Group. The effect of intensive treatment of diabetes on the development and progression of long term complications in insulin dependent diabetes mellitus. New Engl J Med 1993;329:977–86.
  2. The DCCT/EDIC Study Research Group. Intensive Diabetes Treatment and Cardiovascular Disease in Patients with Type 1 Diabetes. New Engl J Med 2005;353:2643–53.
  3. UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352:837–51.
  4. American Diabetes Association. Standards of Medical Care in Diabetes 2010. Diabetes Care 2010;33:S11–S61.
  5. Sacks DB. Carbohydrates. In: Burtis CA, Ashwood ER, Bruns DE, eds. Tietz Textbook of Clinical Chemistry and Molecular Diagnostics. 4th Ed. St. Louis: Elsevier Saunders 2006:837–902.
  6. Goldstein DE, Little RR, Lorenz RA, Malone JI, et al. Tests of glycemia in diabetes. Diabetes Care 2004;27:1761–73.
  7. Little RR, Rohlfing CL. HbA1c Standardization: Background, Progress and Current Issues. Lab Med 2009;40:368–73.
  8. Jeppsson JO, Kobold U, Barr J, Finke A, et al. Approved IFCC reference method for the measurement of HbA1c in human blood. Clin Chem Lab Med 2002;40:78–89.
  9. Hoelzel W, Weykamp C, Jeppsson JO, Miedema K, et al. IFCC reference system for measurement of hemoglobin A1c in human blood and the national standardization schemes in the United States, Japan, and Sweden: a method-comparison study. Clin Chem 2004;50:166–74.

Randie R. Little, PhD, is a research associate professor at the University of Missouri School of Medicine in Columbia and is the network coordinator for the NGSP.

Curt L. Rohlfing is a technical writer and research analyst at the University of Missouri School of Medicine in Columbia, and is the data manager for the NGSP.