The Great Debate
Should Labs Report eAG and HbA1c?
By Genna Rollins
In the world of diabetes management there perhaps is no greater controversy at the moment than whether labs should report estimated average glucose (eAG) in addition to glycated hemoglobin (HbA1c). This topic will be the subject of an afternoon symposium today in Room S100 A&B of the McCormick Place Convention Center. One clinician and one laboratorian each will present views and evidence on opposing sides of the issue. A poll will be taken of audience members before and after the debate to gauge how their thinking on the subject changes as a result of the discussion.
“There are groups vehemently opposed to reporting both numbers and there are those who are supporting it. So the idea was to get them in the same room and talk about the evidence that they believe supports their viewpoint,” explained Mitchell Scott, PhD, professor of pathology and immunology and co-medical director of the clinical chemistry laboratory at Washington University School of Medicine in St. Louis. Scott, who will moderate the debate, also is a past president of AACC. Evidence from the studies that were used to develop the relationship between eAG and HbA1c also will be presented, along with official positions of major professional organizations, including AACC.
The Path Towards Standardization
HbA1c has been used since the early 1990s as a clinical measure of patients’ diabetes control, but the original standardization, spearheaded by the National Glycohemoglobin Standardization Program (NGSP) and based on the Diabetes Control and Complications Trial (DCCT) had some shortcomings. This standardization did not use pure HbA1c for calibration—now defined as glycation of the N-terminal valine of the hemoglobin A beta chain—and therefore was not a true reference method. To address that concern, the International Federation of Clinical Chemistry (IFCC) developed a reference method using mass spectrometry, but at the same time suggested that HbA1c should be reported in mmol per mol rather than percent and that a switch be made to the compound’s chemical name, deoxy-fructosyl Hb. “Many laboratorians in this country thought that would be extremely confusing to both clinicians and patients, since all the evidence and education since the early 1990s had been based on HbA1c with a target value of 7 percent,” observed Scott.
The transition to using IFCC-aligned assays was complicated in that results produced by instruments calibrated to the IFCC reference standard are 1% to 2% lower than those calibrated to the NGSP/DCCT standard. For example, a 7% HbA1c result by NGSP/DCCT would be reported as 5.4% or 54 mmol per mol by IFCC. Leaders in laboratory medicine were concerned that reporting these different HbA1c results would be another source of confusion for both clinicians and patients. In addition, the IFCC method has been described as complicated and expensive (Diabetes Care 2008;31:1704–7).
Subsequently, the A1c-Derived Average Glucose (ADAG) study found a linear relationship between eAG and HbA1c (Diabetes Care 2008;31:1473–8). The study, which was funded by the American Diabetes Association (ADA) and the European Association for the Study of Diabetes, used continuous glucose monitoring and self-glucose monitoring in 507 patients to determine that HbA1c and eAG are reliably correlated and that a mathematical relationship can be used to report eAG from HbA1c. Using the method will enable clinicians to report HbA1c results to patients as an eAG using the same units (mg/dL or mmol/L) that diabetics see routinely in their blood glucose measurements. In June 2008, ADA called for labs to start reporting eAG in addition to HbA1c, in an effort to make it easier for patients to understand control of their disease. The organization recommends that labs report eAG in either mg/dL or mmol/L and maintains an HbA1c-eAG calculator on its website.
AACC issued a position statement in January 2009 advising U.S.-based labs to report both eAG and HbA1c. The statement also observed that at the present time there is inadequate data to justify reporting eAG only without HbA1c and called for additional research to determine the appropriateness of using eAG in less-studied populations, including pregnant women, adolescents, and various ethnic groups. In addition, AACC stressed the importance of achieving global harmonization between eAG and HbA1c and of developing a coordinated plan for educating clinicians and patients about eAG.
A Call for More Evidence
Despite these recommendations, the reporting of eAG has its detractors. “There are many groups of patients where we don’t have evidence that the conversion is valid, including patients with renal impairment, those who are pregnant, and children. There also are a number of ethnic groups where the number included in the studies was very small,” commented Ian Young, MB, BCH, professor of medicine and director of the Centre for Public Health at Queen’s University Belfast. Young, who will be the laboratorian speaking against reporting both eAG and HbA1c at this afternoon’s symposium, said his position reflects that of U.K. clinicians generally. The ADAG study group compared HbA1c with mean glucose in patients with either type 1 or type 2 diabetes at 10 centers in the U.S., Europe, and Africa (Cameroon). Of the 507 participants included in the final analysis, 8% were of African descent and 8% were Hispanic.
The British diabetic community also has concerns about the ADAG conversion formula itself. “We believe that when you apply the equation to convert HbA1c to eAG there is a considerable uncertainty in that conversion. So someone with an HbA1c of 8% may be given a particular eAG value as if it’s an accurate figure, but in fact, the true figure might be significantly above or below that,” contended Young. “Our feeling is that you introduce a substantial error when you convert HbA1c to eAG and that is potentially misleading.”
An a priori criterion of the ADAG study was that 90% of the estimates had to be within 15% of the regression line in order to establish a linear relationship between eAG and HbA1c. The study found that 89.95% did so. Some observers have argued that the study’s criterion allowed too great a degree of variation in HbA1c.
Although he supports reporting both eAG and HbA1c and will be speaking in favor of doing so at this afternoon’s symposium, the conversion formula also is a concern to David Sacks, MB, ChB, FRCPath, associate professor of pathology at Harvard Medical School. “An important thing to communicate to the lab community is that since the way eAGs would be calculated is by lab IT systems, it’s critical that labs use the correct equation to generate the correct number. If they use the old equation, then they’re going to produce erroneous information,” he observed. Prior to publication of the ADAG study, estimates of average glucose had been based on a retrospective analysis of the DCCT which was not designed to determine average glucose, he said.
What’s Best for Patients?
One of the reasons Sacks supports reporting both eAG and HbA1c is based on feedback from clinicians that using the two measures will aid in patient education. “A lot of patients don’t really understand why, when they have a disease they understand is related to increased sugar in the blood, they’re measuring something with the word hemoglobin,” he said. “In my discussions with the team of people who look after patients with diabetes, they believe reporting both eAG and HbA1c will help them communicate the concept of glucose monitoring with their patients,” he said.
In contrast, diabetologists, diabetes educators, and nutritionists in the U.K. generally do not believe the eAG value to be supportive of their educational initiatives, according to Young. “We had some concern that the term ‘eAG’ might be confusing for patients because their glucometers give an average glucose reading that is going to be completely different and separate from an eAG,” he observed. “We think that’s potentially confusing and difficult to explain.”
The airing of these and other issues in support of and against reporting both eAG and HbA1c should make for a lively and informative session, Sacks predicted. “When there’s a contentious issue like this, a debate format is a very good way to present the two sides of the arguments, and then ultimately it’s up to the audience to decide what they’re going to do in their own practices.”