Imperial Collage Healthcare
29 November

Diabetes – markers in a melting pot

Report: Diabetes – markers in a melting pot.

Maggie Hancock, Consultant Clinical Scientist

The UK endorsement of HbA1c in the diagnosis of type 2 diabetes(1) in the middle of 2012 knocked the oral glucose tolerance test (OGTT) off the perch it had occupied for decades. Since then the number of OGTT requests received in our laboratories has fallen dramatically with a corresponding rise in the number of requests for HbA1c.

The interpretation of HbA1c depends on the circulating haemoglobin being predominantly HbA. The lifespan of a red blood cell is 110-120 days; if this lifespan is significantly shorter then, at a set average blood glucose concentration, the exposure of an individual HbA molecule to glycation is reduced with a consequent lowering of the HbA1c result. We know that red blood cell life span is foreshortened in persons who are homozygous for HbS and HbC but, even for these common variants there are conflicting reports about red cell survival in heterozygotes. When you consider that there are now over 1,000 variant forms of haemoglobin on record, there is a lot of work still to do.

The wonderful melting pot that is London must surely have a higher incidence of healthy carriers of a haemoglobin trait than many other parts of the UK. Conscious of this, we at ICHNT, have always used methods of HbA1c analysis that can identify the presence of a non-A haemoglobin.

However, there are faster, and possibly cheaper, assay methods that measure total glycated haemoglobins; these will not detect the presence of variant haemoglobins. When the time came to go out to tender for reagents/equipment for HbA1c analysis we wanted to assess how much value our users attach to knowing whether there is a variant haemoglobin present in a sample submitted for HbA1c analysis.

A short survey was circulated to 115 GP practices; 27 individual GP responses were received from 17 practices. These responses are represented as Pi charts below:

Q1 Where the patient is heterozygous for HbA and a variant haemoglobin we add a comment to highlight the variant’s presence. Does this comment alter your application of the WHO cut-off point of 48 mmol/mol HbA1c/glycated haemoglobin for diagnosis of type 2 diabetes?









Q2 Where the patient is heterozygous for HbA and a variant haemoglobin we add a comment to highlight the variant’s presence. Does this comment alter your application of the NICE treatment target window of 48-59 mmol/mol HbA1c/glycated haemoglobin for type 2 diabetes?









Q3 Do you routinely send a sample for haemoglobin electrophoresis in response to a report highlighting the presence of a variant Hb?










Q4 Does the current turnaround time for HbA1c of 3 working days satisfy your clinical requirements?











Our interpretation of the responses received is, yes, our GPs are aware that HbA1c interpretation may differ in healthy carriers of a haemoglobin trait and that our users are, overall, satisfied with the service we currently provide.

These factors will be taken into account when reviewing our technical/clinical ratings for different providers in our imminent OJEU tender.


In the interim, we have reviewed the comments we add to our HbA1c reports in the light of the expert position statement(1). One instruction from this statement is ‘that laboratories should print concise warning information for clinicians on the laboratory report’ – this is then followed by 44 lines of text on the factors to consider when using HbA1c for diagnosis.


After much discussion, we have condensed the advice and will add the following comment to GP HbA1c reports:

HbA1c is accepted for the diagnosis of type 2 diabetes in the UK but should not be used to diagnose type 1 diabetes or in the following contexts; childhood, pregnancy, renal failure, haemoglobinopathy trait, anaemia, HIV, abnormal red cell turnover, or any recent drug treatment likely to affect glycaemia or red-cell turnover.


The comment added to HbA1c results in patients with a previously unreported variant haemoglobin will change to:

This sample exhibits a haemoglobin variant; caution is indicated when interpreting the HbA1c result.


1 Expert Position Statement

Use of HbA1c in the diagnosis of diabetes mellitus in the UK. The implementation of World Health Organization guidance 2011.

W.G.John on behalf of the UK Department of Health Advisory Committee on Diabetes.

Diabetic Med. 29, 1350-1357 (2012)

Imperial Laboratories Test Directory

Click the A-Z links and you will be able to view the names of services filtered by initial letter.