Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Do not turn HbA1c into something it is not
- Eric S Kilpatrick (19 November 2004)
-
Why make HgbA1c more confusing to our patients
- Charles H Pointer IV RD LD (23 November 2004)
-
A little "backwardness" sometimes pays off
- Peter Boucek (25 November 2004)
-
Re: A little "backwardness" sometimes pays off
- Nils Olof Carlin (26 November 2004)
-
Relevant Issue not Addressed
- George Phillipov (6 December 2004)
|
|
|||
|
Eric S Kilpatrick, Consultant in Chemical Pathology Hull Royal Infirmary
Send response to journal:
|
Home and colleagues support the pragmatist view that the current means of reporting HbA1c values (harmonised to the method used in the DCCT/UKPDS studies) is fit for purpose (1). They believe that any move to report the test using the more recent 'true' standardised values could lead to confusion amongst health care staff because patient values and target numbers would be 1.5-2% lower than currently. The purist scientific view is equally simple: medical science moves forward because of the constant development of new techniques and improvement to existing ones. To halt the reporting of the 'correct' HbA1c result could have unforseen consequences for the interpretation of future research involving the test in patients with diabetes. Thus, although the move to new HbA1c numbers could cause short term confusion, remaining on existing values could potentially cause even longer term harm to patients. As I like to believe I am both a scientist and a pragmatist, I can see both sides of this argument. However, I am very concerned that Home's article suggests that the diabetes community may opt for neither of these approaches, but to move towards expressing HbA1c as an 'average blood glucose equivalent'. Having spent much time trying to explain to patients that the HbA1c value is not in fact the same as mean blood glucose (MBG), there is certainly a superficial appeal to this approach. Nevertheless, the scientist in me finds it inconceivable that one analyte could be expressed in terms of a second. Moreover, the pragmatist in me is aware that different patients with the same HbA1c can have markedly different MBG values: in the DCCT study individual patients with a HbA1c of 7% could have MBG values anywhere between 5 and 15 mmol/L (2). How would I then easily explain to these patients the discrepancy between the laboratory assessment of MBG and their own using a glucose meter? In addition, setting the same MBG 'equivalent' target for all patients would likely be unrealistically low for some (with a high risk of hypoglycaemia if it is attempted to be reached) while for others it would be achieved with apparent ease, thereby leaving the patient and clinician falsely reassured. Whatever the decision taken to move this issue forward, it is hoped that both the pragmatists and the scientists will reject 'average blood glucose equivalent' as an option. 1. Home P, Mbanya JC, Horton E. Standardisation of glycated haemoglobin. BMJ 2004; 329: 1196-7 2. Rohlfing CL, Wiedmeyer HM, Little RR, et al. Defining the relationship between plasma glucose and HbA1c: analysis of glucose profiles and HbA1c in the Diabetes Control and Complications Trial. Diabetes Care 2002 ;25:275-8 Competing interests: None declared |
|||
|
|
|||
|
Charles H Pointer IV RD LD, Registered Dietitian Providence Hospital Washington, DC 20017 USA
Send response to journal:
|
As a Registered, Licensed Clinical Dietitian in that evil North American morass, I cannot understand the hurry to make confusion of a system that currently works quite well. To use the A1c levels as the sole representation of a patient's control of blood glucose will almost certainly reduce control. Medicine is an art as well as a science and must be practiced that way. To attempt a one size fits all leads me to imagine two major problems: 1) Patients with good control will be frustrated with the lack of precision and; 2) Type I patients will be much harder to teach and bring into good control. Also our less educated patients will have quite a hard time learning A1c values. I currently use both systems with very good results. Competing interests: None declared |
|||
|
|
|||
|
Peter Boucek, Diabetes specialist Diabetes Centre IKEM, Prague
Send response to journal:
|
The current discussion on the new HbA1c IFCE standards reminds me very much of the ambitions of some to introduce all SI units into clinical practice in the 70-ies and 80-ies. In my country this included such progressive endeavours as to try to convince all physicians to measure and report blood pressure in kilopascals (or in the extreme even heart rate in cycles per seconds!). As expected by the more "backward" of us all of this has been by now luckily consigned to the dust-bin of history. However, once again we are one of the few who have immediately adapted the new standards. Though having resigned to the new reality I must agree with the authors that substantial doubts on the purpose and effects of the introduction of the new standards into daily clinical practice linger on. The situation remains even more confusing in publications of results of patient follow-up in long-term clinical studies where some sort of conversion of old and new standards becomes a necessity. Competing interests: None declared |
|||
|
|
|||
|
Nils Olof Carlin, radiologist lasarettet Ystad, 271 82 Sweden
Send response to journal:
|
Peter Boucek considers it in the extreme to give heart rate in cycles per seconds - surely the SI unit is Hertz [Hz], and what could be a more appropriate unit to measure heart rate? Competing interests: None declared |
|||
|
|
|||
|
George Phillipov, Research Director Endocrine, Queen Elizabeth Hospital, Woodville, S Australia 5011
Send response to journal:
|
The continuing focus on both the standardisation of HbA1c and the development of a reference assay by the International Federation of Clinical Chemists (IFCC) has diverted attention from a more important issue. The DCCT, and in particular the UKPDS, as they relate to determination of HbA1c (the UKPDS originally determined HbA1), are based on analytical assays known to have systematic error and insufficient selectivity. Accordingly, the HbA1c target and intervention levels, established by these latter trials are biased by their own analytical limitations. Numerical standardisation cannot correct for the inherent lack of selectivity within the DCCT and UKPDS HbA1c values determined for any individual. Thus the HbA1c targets and thresholds established cannot be assumed to be generalisable. Moreover the different glycated hemoglobin assays used throughout the UKPDS, further question the reliability of longitudinal HbA1c measurements. HbA1c method selectivity for the DCCT and UKPDS is critical, as common long-term complications, like progressive renal impairment, would concurrently increase HbA1c levels (due to carbamylated haemoglobin) independently of glycaemic control. I do not believe that the HbA1c values from either the DCCT or UKPDS were ever adjusted to compensate for this type of confounding and its attendant implications. As the National Glycohemoglobin Standardization Program is based on the DCCT method, the notion of assay selectivity does not qualify as a criterion for certification. Unfortunately the IFCC has now developed an analytical method which is not specific for HbA1c, nor is it uniquely selective for the N-terminal valine beta globulin chain (1). The facts support Abraira and Duckworth’s recent call for new glycaemic trials (2), but with the added caveat that the measurement of any glycaemic marker must be specific and readily adapted to routine use. Given the problems with HbA1c, which extend even to controversy surrounding the exact nature of the analyte (3), a plasma glycoprotein would seem most appropriate and practical. 1. Jeppsson JO, Kobold U, Barr J, et al. Approved IFCC reference method for the measurement of HbA1c in human blood. Clin Chem Lab Med. 2002;40:78-89 2. Abraira C, Duckworth W. The need for glycemic trials in type 2 diabetes. Clinical Diabetes 2003;21:107-111 3. Kobold U, Jeppsson JO, Dulffer T, Finke A, Hoelzel W, Miedema K. Standardization of hemoglobin A1c. Clin Chem. 1998;44:1068-1069 Competing interests: None declared |
|||