NGSP News Archive: Summer 2009 Meetings

Clinical Advisory Committee | NGSP Steering Committee | Manufacturer Forum

2009 NGSP Clinical Advisory Committee Meeting

American Diabetes Association 69th Annual Scientific Sessions, June 2009

The CAC meeting was attended by representatives of various clinical diabetes organizations (e.g. IDF, ISPAD, ADA, EASD). The major topics of discussion were hemoglobin variants, variability among laboratories and methods, point-of-care methods, estimated average glucose (eAG) and the use of HbA1c for diagnosing diabetes.

1.  HbA1c/NGSP Update: Hemoglobin Variants--Randie Little, NGSP Network Coordinator

  • CAP survey data show that there has been considerable progress in reducing the variability of HbA1c reseults since 1993.
  • The most recent CAP survey (2009A): Most methods showed bias <0.3% HbA1c from NGSP target valus and CVs <5.0% across all three levels.  Not many users of  point-of-care (POC) instruments participate in the CAP survey so we do not have a good picture of how many labs use them or how they are performing in the field
  • Critical difference or reference change value: The critical difference (CD) is the change in a patient’s serial test results that can be considered significantly different (e.g. at a probability of 95%).  Many physicians consider a change of 0.5% HbA1c to be significant.  This requires an assay CV of <3%; many but not all current methods meet this criterion.  If the tests are performed by different labs the current all-method CV of ~4.5% must be considered.
  • Ways the NGSP is trying to decrease variability
    1. Tightening the NGSP certification criteria
      • The criteria for the 95% CI of the differences between the assay method and SRL will be tightened from ±0.85% HbA1c to ±0.75% (from ±0.75% to ±0.70% for Level 1 Laboratories) in January 2010.
    2. Use of accuracy-based criteria by CAP, based on values assigned by the NGSP, began in 2007.
      • In 2007 ±15% was used to determine pass/fail rates.  In 2008, it was reduced to ±12% and ±10% in 2009.  It will be further reduced to ±8% and ±6% in 2010 and 2011, respectively.
      • The manufacturers have received their pass rates with current criteria and what they would be when the criteria are set at +6%.
  • HbA1c interferences from the most common hemoglobin variants (HbAS, HbAC, HbAD and HbAE).
    1. There is a table on the NGSP website that shows the interferences from the most common hemoglobin variants.
    2. Based on the CAP data, 5% of labs are using methods that show clinically significant interference (defined as >10% difference at 6 and/or 9% HbA1c) from HbAS and HbAC.
    3. At least 20% of labs are using methods with clinically significant HbAE interference and at least 6% of labs are using methods with clinically significant HbAD interference.
    4. Manufacturers are improving their methods to get rid of the interferences where possible.


POC Methods
There was much concern regarding the lack of data on POC methods, the fact that these methods are CLIA-waived means that users of the methods are not required to participate in the CAP survey.  Nonetheless these methods are widely used, especially in the developing world, and therefore it is important to know how well they are performing in the field.  If manufacturers would have to apply to the FDA for their method to be used for diagnosis, these POC methods could be “un-waived” for diagnosis which would force their participation in the survey.

HbA1c for Diabetes Diagnosis
Methods that show a positive bias in the normal range could cause over-diagnosis.  If HbA1c is to be used for diagnosis there should be assay performance specifications in terms of CV and bias, and physicians need to be made aware of how well the lab doing their testing is performing.  The biological variability of HbA1c is lower than for glucose, which is an advantage, but interference from hemoglobin variants and factors that affect red cell lifespan are a concern.

2.  IFCC – WG update: D. Sacks:  The IFCC working group met in April 2009.

  •   In 2008 values for the samples to be used as future calibrators for the IFCC reference method did not match between the assigned values and those measured by the IFCC laboratories.  The cause is being investigated, this calibrator will not be used by the IFCC network.
  • The master equations developed between the IFCC and the NGSP networks and the IFCC and Swedish methods have remained stable, but the equation between the IFCC and Japan has drifted.  The cause is being investigated.
  • Global standardization is going in different directions.  Germany will be reporting only IFCC numbers; the United Kingdom is reporting both IFCC and NGSP numbers and will move only to reporting IFCC numbers in two years.  Several other countries are following the UK, eAG is not being reported outside the U.S. where it will be reported along with NGSP %HbA1c.

Discussion:  It is likely that dual-reporting of HbA1c numbers will continue in journals.  The UK decision was made after a meeting of officials from all healthcare disciplines as well as patients.  In the end it was decided that only one number should be reported and it should not be eAG.  It is of concern that different countries are going different directions, this could cause confusion.  The numbers coming out of the instruments are still the same; they are just being converted by equation prior to reporting of the final result.

3.  eAG Update: M. Peterson

  • The American Diabetes Association in the United States is moving forward with reporting eAG as an option.
  • We are not saying do not report HbA1c, but want to make eAG available as an option when discussing average glucose values with patients.
  • The ADA is providing education materials for conversion from HbA1c to eAG, and there is an on-line calculator that allows educators to determine their patients’ eAG value.
  • A professional website has been developed.
  • NDEP has shown interest in using eAG in their materials.

Discussion:  16% of labs that participated in the latest CAP survey are reporting eAG.  It is important that healthcare providers understand eAG so they can educate their patients and that labs use the correct equation from the ADAG study rather than the older equations.

4.  HbA1c for Diabetes Diagnosis: D. Nathan

  • The expert committee’s report will be in Diabetes Care, Volume 32, Number 7, July 2009.
  • The committee focused on several points:
    1. Potential advantages and disadvantages of HbA1c compared to glucose measurements as used for diagnosis of diabetes.
    2. The premise that we should diagnose diabetes based not on some absolute glucose level but on the clinical implications of glucose levels long-term and their relationship to organ disease.
      • There is a relationship between the level of glycemia and complications and that should be the premise on which we base the diagnosis of diabetes.
      • Fasting and 2-hour glucose, and HbA1c each represent a different feature of metabolism, yet all three are inter-correlated and are related to complications.
      • What is the advantage of switching from diagnosis based on glucose to HbA1c?
        1. The standardization of HbA1c has tightened the precision and accuracy of HbA1c methods, and glucose methods are far from perfect.
        2. The pre-analytical sample stability is better and the biological variability is lower with HbA1c compared to glucose measurements.
  • The committee decided to recommend that HbA1c be used to diagnose diabetes.
    1. A cutoff of 6.5% HbA1c was chosen because there is clearly an absence of diabetes related retinopathy below this level.
    2. In the case of pre-diabetes, the patients at the greatest risk are those closest to the cut-point of 6.5% HbA1c, therefore 6.0-6.4% HbA1c was selected as the category for patients with pre-diabetes.
    3. In those patients and places where you cannot do an HbA1c, glucose measurement should be used for diagnosis.

Discussion:  The ADA executive committee has approved the recommendations of the expert committee.  Other clinical organizations are discussing this issue.

5.  How HbA1c levels are (mis)interpreted in practice—D. Arons

  • Some physicians believe that a HbA1c of 7.1% is bad while a 6.9% is good without taking analytical variability into account.
  • There are various ways that variability can influence results: everything occurs in context.
  • Evidence shows that quality care increases the likelihood of good outcomes in individuals and populations.
  • Various factors affect where HbA1c targets should be, as the HbA1c level is lowered there is risk of hypoglycemia.
  • There is a balance for each individual patient.

Discussion:  Guidelines are important but individual patient preferences are equally important.