Test Catalog


Hemoglobin A1c:

Diabetes Diagnosis and Management

September 2009

Glucose is a vital source of energy for the body. We obtain glucose either from the diet or from production by the liver. The liver stores or releases glucose to maintain the necessary level of glucose in the blood. Insulin, a hormone synthesized by the pancreas, allows cells to utilize glucose, which reduces blood glucose levels. The more glucose in your blood, the more insulin the pancreas releases. Diabetes is caused by an inability to properly process glucose, resulting in chronic hyperglycemia. Chronic hyperglycemia results in long-term damage to various organs such as the kidneys, heart, and eyes.1

There are 3 primary types of diabetes: type 1, type 2, and gestational (not discussed in this article). Type 1 diabetes, previously called insulin-dependent diabetes mellitus or juvenile-onset diabetes, accounts for ~5% to 10% of all diagnosed cases of diabetes and results from an inability to produce adequate amounts of insulin. Autoimmune, genetic, and environmental factors are involved in the destruction of the insulin-producing cells in the pancreas, resulting in development of type 1 diabetes. Patients require treatment with insulin. Type 2 diabetes, previously called adult-onset diabetes, accounts for the majority (90%–95%) of new diagnoses of diabetes. Development of type 2 diabetes is a result of age, genetic factors, and obesity. Affected individuals develop insulin resistance, requiring increased production of insulin, and may eventually require insulin treatment. Type 2 diabetes, the most common type of the disease, will be the focus this article.

Complications and Costs of Diabetes

Obesity is the major factor in the growing number of Americans developing type 2 diabetes. A majority of Americans are overweight and more than 30% of these individuals are classified as obese or extremely obese (body mass index >30% or >40%, respectively).2 The incidence and prevalence of obesity and type 2 diabetes in children and adolescents also has increased dramatically, reaching epidemic proportions.3

The Centers for Disease Control and Prevention (CDC) estimated that in 2007 nearly 24 million Americans had diabetes, an increase of nearly 3 million in roughly 2 years, largely due to increases in obesity.4 Approximately one-fourth (6 million individuals) of those individuals remain undiagnosed and at risk for premature development of micro- and macrovascular complications. (Figure 1) Complications of diabetes include5:

  • High blood pressure
          – 75% of diabetics are hypertensive or on medication for hypertension
          – Ischemic heart disease and stroke
  • Kidney disease
          – Leading cause of kidney failure
  • Blindness and diabetic retinopathy
          – Leading cause of blindness among adults aged 20 to 74
  • Nervous system disease
          – 60% to 70% develop nervous system damage
  • Nontraumatic amputations
          – Severe diabetes is a major contributing cause of lower-extremity amputations due to peripheral neuropathy, necrosis, and gangrene1
          – >60% of lower-limb amputations are in diabetics
  • Heart disease and stroke
          – 2 to 4 times higher risk of death or stroke
  • Dental disease
          – Approximately one-third of diabetics have severe periodontal disease
  • Weakened immune system making patients susceptible to infections (influenza, pneumonia, skin infections)
  • Complications of pregnancies
Figure 1

Figure 1. Complications of diabetes.

Previous research trial groups, including the Diabetes Control and Complications Trial (DCCT) and the United Kingdom Prospective Diabetes Study (UKPDS) both demonstrated a strong relationship between the level of plasma glucose control for both type 1 and type 2 diabetes and the risk of retinal, renal, and neurological complications. An estimated one-fourth of diabetics manifest complications at the time of diagnosis and there is an approximate 7-year gap between the onset of diabetes and its clinical diagnosis. Early identification of patients at risk for diabetes, identified either from obesity, demonstrated insulin resistance, or family history, allows for interventional lifestyle changes to decrease excess weight, increase exercise, and modify the diet. While individuals’ success at lifestyle changes varies, careful management of prediabetes and diabetes improves the long-term outcomes, reducing serious and life-threatening complications.

The critical importance of timely diagnosis of diabetes relates to the high prevalence of long-term medical complications in diabetics that, in turn, can have devastating effects on patients and place a burden on the health care system. Based on 2007 data, direct medical costs for diabetes in the United States were $174 billion.5 This is more than double the estimated expenditures if diabetes was not a factor.

Diagnosis of Diabetes with Glucose

The laboratory plays an important role in the diagnosis and chronic management of diabetes. Traditionally, the diagnosis of diabetes is made by any 1 of the following tests1:

  • Fasting plasma glucose 126 mg/dL
  • Random plasma glucose 200 mg/dL with symptoms of hyperglycemia (increased thirst, increased urination, unexplained weight loss)
  • 2-Hour plasma glucose 200 mg/dL during an oral glucose tolerance test (OGTT)

In the absence of unequivocal hyperglycemia, laboratory results must be confirmed by repeat testing on a different day. Note: Point-of-care glucose analyzers are not sufficiently precise to be used for diagnosis of diabetes and should not be used for this purpose.

Measurement of the fasting plasma glucose has several advantages. Glucose is technically easy to measure and widely available on automated instruments. Only 1 specimen is needed, as opposed to timed specimens with the OGTT. A single cutoff (126 mg/dL) is used for diagnosis (prediabetes, also known as impaired glucose tolerance, is defined by a fasting glucose level of 100–125 mg/dL).

Disadvantages of using the fasting plasma glucose are that the patient must fast at least 8 hours (preferably 12 hours) and that there is a diurnal variation in glucose, such that specimens should be drawn in the morning when values are at their peak and individuals will not be underdiagnosed. These 2 issues present an inconvenience for the patient. In addition, there is a large biological variability with a fasting glucose, with intraindividual coefficient of variation (CV) of 5% to 8% and interindividual CV of 7% to 13%. Fasting plasma glucose also is considered to be less sensitive than the OGTT. There are preanalytical problems as well, and it is an often overlooked fact that use of a sodium fluoride collection tube does not prevent glucose degradation for the first 30 to 90 minutes after the specimen is drawn.6

Hemoglobin A1c

Assessment of chronic glycemic control typically focuses on measurement of glycated proteins such as HbA1c and fructosamine. HbA1c is a result of the nonenzymatic attachment of a glucose molecule to the N-terminal amino acid of the hemoglobin molecule. (Figure 2) The attachment occurs continually over the entire lifespan of the erythrocyte and is dependent on blood glucose concentration and the duration of exposure of the erythrocyte to blood glucose. It is also thought that the permeability of the red blood cell to glucose influences the amount of glycation and could explain the discordance noted in some hematologically normal people with diabetes in whom HbA1c results appear different from their other measures of glycemic control.7,8

Figure 2

Figure 2. Hemoglobin glycation.

In general, the HbA1c level reflects the mean glucose concentration over the previous 8 to 12 weeks and provides a much better indication of long-term glycemic control than blood and urinary glucose determinations. However, this is not true in patients who have altered red blood cell lifespans, including anemias, hemolysis, B12 or folate deficiencies, or hemoglobinopathies. Since the HbA1c assay reflects long-term fluctuations in blood glucose concentration, a diabetic patient who has in recent weeks come under good control may still have a high concentration of HbA1c. The converse is true for a diabetic previously under good control who is now poorly controlled.

The American Diabetes Association (ADA) recommends measurement of HbA1c (typically 3-4 times per year for type 1 and poorly controlled type 2 diabetic patients, and 2 times per year for well-controlled type 2 diabetic patients) to determine whether a patient’s metabolic control has remained continuously within the target range. In the 2009 ADA Summary of Revisions for the 2009 Clinical Practice Recommendations, the new HbA1c goal for nonpregnant adults is now <7%.9 This is a general guideline to reduce the risk of microvascular and neuropathic complications. Physicians must evaluate individual patients and determine whether the patient would benefit from more stringent HbA1c goals, or should have less stringent goals, based on their current state of health.

Until now, the use of HbA1c was not recommended for diagnosis of diabetes. In a recent announcement, the International Expert Committee, which includes representatives from the ADA, International Diabetes Federation, and European Association for the study of Diabetes (EASD), has recommended HbA1c for diagnosis of diabetes with cutpoint at 6.5%.10

Hemoglobin A1c Methods

Methods for analysis of HbA1c can essentially be divided into 2 categories: those that measure HbA1c based upon charge and those that quantitate HbA1c by structure. A common charge-based method utilizes cation-exchange high pressure liquid chromatography (HPLC), where hemoglobin species (eg, HbA, HbA2, HbF) elute from the cation-exchange column at different times, depending on their charge, with the application of buffers of increasing ionic strength. Changes or mutations in the hemoglobin molecule (hemoglobinopathies, also termed hemoglobin variants) that result in charge differences may or may not alter the normal elution time when using cation-exchange chromatography. In general, charge-based methods are considered more susceptible to interference from hemoglobin variants.

Structural methods include immunoassays and boronate-affinity chromatography. Immunoassays use antibodies that target the β-N-terminal glycated amino acid in the first 4 to 10 amino acids of hemoglobin, depending upon the manufacturer. Interferences with immunoassays are uncommon. Boronate-affinity chromatography separates glycated from nonglycated compounds and is least affected by the presence of hemoglobin variants. However, these methods should be used with caution in populations where the prevalence of hemoglobin variants is high. The majority of laboratories in the United States utilize either an immunoassay or a cation-exchange HPLC method.

Although more than 950 different hemoglobin variants have been identified, most have a benign phenotype and patients are not affected clinically. In the United States, hemoglobin S is the most common variant, followed by hemoglobin C, hemoglobin E, and hemoglobin D (Punjab/Los Angeles), in that order. Worldwide, hemoglobin variants follow this similar trend, except hemoglobin E is more common than hemoglobin C. In the presence of a given hemoglobinopathy, the accuracy of HbA1c results depends on the method; some methods provide reliable results, others do not. For patients who have no hemoglobin A, such as homozygous S or C patients, some HbA1c assays may be inaccurate (providing falsely increased or falsely decreased results) or results may be lower due to increased red cell turnover rates. In general, it is recommended that an alternate test of glycemia, such as fructosamine, be used for these patients. Fructosamine is a general term that applies to any glycated protein formed by the nonenzymatic reaction of glucose with the amino groups of proteins. Albumin, the most abundant serum protein, accounts for 80% of the fructosamines. The nonenzymatic glycation of proteins in vivo is proportional to the prevailing glucose concentration during the lifetime of the proteins. In general, fructosamine levels reflect glycemic control over the previous 2 to 3 weeks. If fructosamine is unavailable, the glycemic targets using HbA1c should be adjusted accordingly.

The Estimated Average Glucose (eAG)

HbA1c is also being touted as a means to calculate estimated average glucose (eAG). Traditionally, many physicians correlate the HbA1c with the patient’s glucose meter history and fasting plasma glucose concentrations. After the DCCT and UKPDS trials, it was clear there was a direct relationship between HbA1c and mean blood glucose concentration. The major advantage to reporting HbA1c as an eAG is that both physicians and patients understand glucose results and patients are familiar with glucose values from their home glucose meters. Until recently, no reliable regression equations were available to calculate an eAG. However, in 2008, 2 major studies validated the mathematical relationship between HbA1c and eAG.11,12

In the A1c Derived Average Glucose (ADAG) study, 507 diabetic and nondiabetic subjects were evaluated.12 A linear regression was calculated that directly used HbA1c to calculate the eAG. The correlation was 0.84 and was deemed significant. The following table shows the relationship of the measured HbA1c with the calculated estimated average glucose.

HbA1c (%)
eAG (mg/dL)

Table. Derived Relationship of HbA1c to Estimated Average Glucose11

There are some limitations noted with the ADAG study. The majority of the patients were type 1 diabetics. Only a small number of ethnic groups were included, and most subjects were Caucasian. There was no data in children, pregnant women, or patients with renal impairment. In addition, there was enough scatter around the HbA1c values that the concept of the glycation gap was brought into the picture. It is hypothesized that some patients are high glycators and some are low glycators. This means that 2 patients may have the same average blood glucose, yet the high glycator will have a much higher HbA1c than the low glycator. This is notable if you examine the confidence intervals around a HbA1c of 7%, which translates to an eAG of 154 mg/dL. The confidence intervals around the HbA1c are 6.7% to 9.2%, and the eAG is anywhere from 123 mg/dL to 185 mg/dL. This wide margin of error has significant clinical and analytical implications.

Despite some of these shortcomings, reporting of the eAG has been endorsed by several clinical groups such as the ADA and International Diabetes Foundation, and laboratory medicine groups including the American Association of Clinical Chemistry and the International Federation of Clinical Chemistry.

A New Role for HbA1c

In addition to reporting the eAG, HbA1c has also been targeted for a potential role in the diagnosis of diabetes. The advantages to using HbA1c are:

  • It gives a measurement of chronic hyperglycemia, based on the individual’s red blood cell lifespan.
  • The standardization efforts from the National Glycohemoglobin Standardization Program (NGSP) have been largely successful and the accuracy of HbA1c testing is closely monitored by manufacturers and laboratories.
  • No fasting is necessary to measure HbA1c and there is very low intraindividual variability with a CV of <2%.
  • A single test could be used for both diagnosis and monitoring of diabetes.

Some arguments against using HbA1c include the limited number of studies that have been performed thus far to derive an appropriate diagnostic cutoff. Other conditions can alter HbA1c values, including the presence of hemoglobin variants, uremia, transfusions, and anemias. There are also arguments that some analytical issues related to the imprecision of the assays remain unresolved, and HbA1c is a more expensive analyte to measure than glucose.

Very recently, HbA1c was endorsed by an International Expert Committee to be used in the diagnosis of diabetes.10 A cutpoint of 6.5% was recommended, based on the sensitivity and specificity demonstrated by several studies. An elevated HbA1c should be confirmed with a repeat measurement, except in those individuals who are symptomatic and also have a plasma glucose over 200 mg/dL. The committee recommended that the terms prediabetes, impaired fasting glucose, and impaired glucose tolerance be phased out to eliminate confusion. Patients who have an HbA1c 6.0% but 6.5% are considered at risk for developing diabetes in the future. HbA1c for diabetes screening is not endorsed at this time.

HbA1c Testing at Mayo

On April 28, 2009, Mayo Medical Laboratories changed HbA1c testing from an immunoassay to the BioRad Variant II Turbo, a cation ion-exchange HPLC method. Reflex testing is performed using boronate-affinity methodology if a result is unable to be obtained by HPLC. The HbA1c reference range also changed and is now 4% to 6%, which is a decision-based reference range from the ADA, as opposed to the previous reference range that was determined by normal population sampling.

#82080 Hemoglobin A1c, Blood

New reference values:
18 years: 4.0%–6.0 %
ADA-defined normal range

Previous reference values:
≥16 years: 4.7%–5.8%
Population-derived normal range

Reporting of the eAG at Mayo Clinic is still being debated at this time; however, the conversion equation is provided:
            eAG (mg/dL) = (28.7 × %HbA1c) – 46.7


The role of HbA1c is clearly changing and is a controversial topic in many laboratories. It is clear that the optimal diagnostic criteria for diabetes are still being debated. HbA1c has recently been endorsed by clinical groups for diagnosis of diabetes using a cutpoint of 6.5%. Reporting of the eAG at this time is still controversial and will vary from laboratory to laboratory depending on the preference of the physicians, pathologists, and endocrinologists.


  1. American Diabetes Association: Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 2009 Jan;32:Sup 1
  2. CDC FastStats: Overweight Prevalence available at http://www.cdc.gov/nchs/fastats/overwt.htm. Retrieved 8/6/09
  3. Robard HW: Diabetes Screening, Diagnosis, and Therapy in Pediatric Patients With Type 2 Diabetes. Medscape J Med 2008;10(8):184
  4. Department of Health and Human Services: Number of People with Diabetes Increases to 24 Million. Press Release, June 24, 2008. Retrieved 2/28/09. Available: www.cdc.gov/diabetes
  5. National Institute of Diabetes and Digestive and Kidney Diseases: National Diabetes Information Clearinghouse: National Diabetes Statistics, 2007. Retrieved 3/24/09. Available at diabetes.niddk.nih.gov/dm/pubs/statistics/index.htm#complications
  6. Chan AYW, Swaminathan R, Cockram CS: Effectiveness of sodium fluoride as a preservative of glucose in blood. Clin Chem 1989;35:315-317
  7. Cohen RM, Franco RS, Khera PK, et al: Red cell life span heterogeneity in hematologically normal people is sufficient to alter HbA1c. Blood 2008;112:4284-4291
  8. Khera PK, Joiner CH, Carruthers A, et al: Evidence for interindividual heterogeneity in the glucose gradient across the human red blood cell membrane and its relationship to hemoglobin glycation. Diabetes 2008 Sep;57(9):2445-2452
  9. American Diabetes Association: Summary of Revisions for the 2009 Clinical Practice Recommendations. Diabetes Care 2009 Jan;32(S1):S3-S5
  10. International Expert Committee Report on the Role of the A1C Assay in the Diagnosis of Diabetes. Diabetes Care 2009 July;32(7):1327-1334
  11. Nathan DM, Kuenen J, Borg R, et al: Translating the A1C Assay Into Estimated Average Glucose Values. Diabetes Care 2008 Aug;31(8):1473-1478
  12. Lenters-Westra E, Slingerland RJ: Hemoglobin A1c determination in the A1C-Derived Average Glucose (ADAG) study. Clin Chem Lab Med 2008;46(11):1617-1623

Back to top

Tell Us What You Think