# HbA1c Theory

Discussion in 'Blood Glucose Monitoring' started by Mr_Pot, Nov 6, 2017.

1. ### Mr_Pot Type 2 · Well-Known Member

Messages:
725
Likes Received:
336
Trophy Points:
103
HbA1c has been discussed in many threads. However there are aspects of it that I am still puzzled about and I would be interested in any light that can be thrown on the subject. I don’t think it makes much difference to treatment, at least in my case, so my interest is really “academic”.

It is said that HbA1c indicates the average blood glucose level over a period of about 12 weeks because that is an average time for the red blood cells to be replaced. Not everyone’s RBCs keep to the 12 weeks, so comparing one person’s HbA1c with another may not be appropriate and HbA1c may not agree with an average blood glucose calculated from regular blood testing.

1. Does anyone know, how the red blood cells becoming glycated but continually renewing, can be represented mathematically, to make it possible to calculate what effect different renewal rates would have on HbA1c?

2. It is also sometimes said that the period before the HbA1c test, maybe 4 weeks before, has more significance than the earlier period. Is there any evidence for this or is it perhaps a consequence of the above maths? If it is true and as long as your RBCs renew on average in more than 4 weeks, doesn’t that largely negate the argument that different renewal rates affect HbA1c?

#1
2. ### sally and james Family member · Well-Known Member

Messages:
751
Likes Received:
1,489
Trophy Points:
158
I don't know the answers to @Mr_Pot 's questions, but I'll have a go.
1. If you lose 10% of your red blood cells, in a steady way, every week and they are replaced in the same steady way AND you take in and produce glucose in the same even way, I'm sure that is the beginning of a mathematical equation. However, you have a late night, a slice of birthday cake, go on holidays, get flu and the whole thing is all over the place and your guess is as good as mine. The maths could give a vague idea, but won't be accurate in the real world.
2. I suppose the last four weeks is more relevant because you will almost certainly still have all the red blood cells created in this period, older ones are more likely to have died off, so the effects of sticky toffee pudding from eight weeks ago, may have largely gone, but last week's indulgence will definitely still be there. Regard recent weeks as being weighted, but not negated.
Hope this makes sense to somebody!
Sally

• Like x 2
• Winner x 1
#2
3. ### fletchweb Type 1 · Well-Known Member

Messages:
395
Likes Received:
444
Trophy Points:
103
Another interesting observation - I read somewhere that the margin of error fluctuates with ethnicity as well - when I do get my HbA1c I don't see it as the holy grail of success or failure but a loose measure of my BG levels over time - often more important to my health professionals than anyone else. But it's easy for me to say that as I've been living with Type 1 for 51 years, never had a good HbA1c and have no complications. I also have gone without the worries and apprehension too and there's much to say about avoiding those mental health issues/emotions.

Even HbA1c levels and complications are somewhat in doubt when many of these studies leave out other variables that may have a stronger link to complications than HbA1c levels. I'm not saying I'm right, (and the last time I said something similar to that I think I upset a lot of diabetics because I challenged their belief systems) it's important we question these things. As people we tend to jump on the band wagon on many medical discoveries and tools where faith seems to play a greater role than science.

Anyway, I can't answer your question but it is a good one. Here's an interesting link - https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2769981/

• Like x 1
#3
4. ### ringi Type 2 · Well-Known Member

Messages:
1,286
Likes Received:
456
Trophy Points:
83
It used to be assumed that "red blood cells becoming glycated" could only work in one direction, some people now think that if your BG reduces a lot, the "red blood cells can become UNglycated". Hence a much lower BG in the last few weeks can change the result of an AC1 by more than is expected.

#5
5. ### Mr_Pot Type 2 · Well-Known Member

Messages:
725
Likes Received:
336
Trophy Points:
103
I think I will have to wait until Xmas when I see my nephew who has actuarial training to answer question 1.

I have enlarged on your comments on question 2. below.

Assume cells live for 10 weeks and are renewed on a weekly basis rather than continuously, for simplicity…..

In a sample of blood there will be 10% of cells 1 week old, 10% of cells 2 weeks old, and so on to 10% at 10 weeks old.

Assume that the glucose level is constant for 9 of the weeks but high for 1 of the weeks; say due to high carbs on a holiday.

If the holiday was week 2 of the 10 week period, only 20% of the sample existed at that time and would have been affected. If the holiday was in week 8 then 80% of the cells in the sample existed at that time and would have been affected.

So it seems to be true that the latter weeks have more effect on HbA1c than the earlier weeks. In the example a high glucose level in week 8 would have 4 times the effect on HbA1c that the same level had in week 2.

• Informative x 1
#6
6. ### Bluetit1802 Type 2 · Oracle

Messages:
16,107
Likes Received:
25,451
Trophy Points:
278
Every minute/second of the day red blood cells die and are replaced by brand new ones in a continuous process. These brand new ones will have masses less glycated haemoglobin than the others. It is also true that if you donate blood or have a transfusion of blood, your HbA1c will be low if taken not long after. This is a well known fact.

• Like x 1
#7
7. ### Mr_Pot Type 2 · Well-Known Member

Messages:
725
Likes Received:
336
Trophy Points:
103
I didn't understand your point. Are you agreeing or disagreeing?

#8
8. ### Squire Fulwood Type 2 · Expert

Messages:
5,710
Likes Received:
4,590
Trophy Points:
178
#9
9. ### Bluetit1802 Type 2 · Oracle

Messages:
16,107
Likes Received:
25,451
Trophy Points:
278
#10
10. ### Bluetit1802 Type 2 · Oracle

Messages:
16,107
Likes Received:
25,451
Trophy Points:
278
Agreeing. I think.

#11
11. ### LittleGreyCat Type 2 · Well-Known Member

Messages:
1,607
Likes Received:
1,127
Trophy Points:
178
Haven't read the references, but isn't one of the issues that the calculation is based on the average life span of the red blood cells?
Ignoring for the moment another point raised, that the glycation of red blood cells changes up and down (I would assume slowly) as BG levels go up and down (which makes it even more complicated) let us assume that new red blood cells record the BG level at the moment that they are created, then preserve this value (effectively a meter reading) until they are destroyed.
Let us further assume that red blood cells have an average life of 100 to 120 days (giving a mean age of 50 to 60 days). This suggests that after 100 days (just over 3 month usually) there could be a significant die off and by 4 months nearly all the red blood cells are gone. However we are talking averages and some red blood cells could well have died off before the 3 month mark.
So HbA1c is likely to most accurately reflect the period where you have 100% of the blood cells surviving which presumably is at least 4-6 weeks, probably 8.
After that, assuming a 10 week/70 day range before the old red cells really start to decline I assume you would mathematically deprecate the portion of the red blood cells consisting of old red cells. That is, after 10 weeks you can no longer assume that the red blood cells are as accurate as the newly minted.

O.K. now I have argued myself into a pit of confusion.

Too many unknowns.

Thrashing about did raise one question. I assume that women with functioning ovaries have a higher turnover in red blood cells than men due to monthly periods, but that just reduces the total red cell count not the balance? Or do two monthly cycles reduce the percentage of old red cells and increase the percentage of brand new?

I think I will go somewhere quiet and stop thinking for a bit.

#12
12. ### Mr_Pot Type 2 · Well-Known Member

Messages:
725
Likes Received:
336
Trophy Points:
103
We seem to agree that HbA1c has its shortcomings that can make it an inaccurate measure of average blood glucose, in particular the variation between people as to the lifetime of their red blood cells and the greater significance of the most recent weeks before the test. Also a short term high glucose spike could be a source of damage in the body but not be registered by HbA1c so additional means of measuring variation in BG are advisable.

As I said in my original post my interest is just “academic” I find it fascinating how the human body works and the more I read, the more I realise how little I know and how complex everything is. I do have some observations about @Squire Fulwood ‘s blog and the post by @LittleGreyCat .

Glucose is carried around the body in the blood plasma, the liquid in which the blood cells are suspended not by the cells themselves. As far as I know the glucose “accidentally “gets attached to haemoglobin molecules in the red blood cells. As this attachment has been shown to be proportional to the concentration of glucose in the plasma it can be used as a measure of average blood glucose levels that have existed over the lifetime of a cell. I have not seen any evidence that this attachment only occurs when cells have just been created, I think it just happens randomly during their lifetime.

I hadn’t realised until recently the enormous numbers involved…..

There are about 5,000,000 red blood cells per cubic mm of blood.
If the human body contains 5 Litres of blood then there are about 25,000,000,000,000 RBC in total.
If they are replaced every 120 days = 120x24x60x60 seconds
Then red blood cells are created at a rate of about 2,400,000 per second.
Each red blood cell contains about 270,000,000 haemoglobin molecules.
So 648,000,000,000,000 haemoglobin molecules are created every second.

#13
13. ### Bluetit1802 Type 2 · Oracle

Messages:
16,107
Likes Received:
25,451
Trophy Points:
278
It's no wonder our meters are inaccurate, and consecutive readings wildly different!

I did read somewhere, in one of the papers about the unreliability of the HbA1c that the attachment of glucose to haemoglobin is a slow process - but how slow I have no idea.

#14
14. ### ringi Type 2 · Well-Known Member

Messages:
1,286
Likes Received:
456
Trophy Points:
83
At least hba1c is a lot more meaningful then BMI...... And is a very good way to track progress. No one would claim that a Type2 by diet along going from an AC1 of 80 to 50 has not improved!

• Agree x 1
#15
15. ### Lincka01 · Member

Messages:
8
Likes Received:
4
Trophy Points:
3
Made me lol

#16
• ### Meet the Community

Find support, connect with others, ask questions and share your experiences with people with diabetes, their carers and family.

Did you know: 7 out of 10 people improve their understanding of diabetes within 6 months of being a Diabetes Forum member. Get the Diabetes Forum App and stay connected on iOS and Android

Grab the app!