• Guest - w'd love to know what you think about the forum! Take the 2025 Survey »

hba1c correlation with RBC lifespan

finsit

Well-Known Member
Messages
331
Location
UK
Type of diabetes
Type 2
Treatment type
Diet only
I am in a fix, been reading on how much RBC lifespan can affect your a1c results. I finally found how to calculate your mean RBC lifespan using your CBC results, its explained here https://forum.fudiabetes.org/t/a1c-test-results-calculating-your-rbc-lifespan/3965

Now i did the calculations and i found my average RBC lifespan is 78 days as opposed to 120 days which means my hba1c will always come lower than actual figure. My intensive finger testing estimated a1c over the past several months is about 32 mmol, however on blood results it comes 26. Does anyone has an idea how to correct a1c levels if you know the RBC lifespan ? Can you guys calculate your RBC lifespan and see if it correlates with your a1c results?

You will need your RBC value, haematocrit value and your retic count.

All help will be appreciated.
 
Personally if my HbA1c was anywhere between 26 and 32 I would be very happy. I don't see the point in paying for tests where whatever the result it wouldn't make any difference.
 
Mr_Pot said:
Personally if my HbA1c was anywhere between 26 and 32 I would be very happy. I don't see the point in paying for tests where whatever the result it wouldn't make any difference.
Click to expand...
That is if that is the correlation. I am afraid i may be running in 40's somehow, so thinking of a fructosamine test. My BGs are 90% of the time between 90-120 (5-6.7 mmol), what a1c does this sound like? I seldom cross 8 mmol
 
My finger prick test averages have never matched or even near matched my A1c tests.

I use the A1c as an indication of the direction of travel, not an absolute truth. When you think about it, if your lab A1c is 30, say, in most instances it is reported in full numbers, so it could actually be from 29.5 to 30.4.

Clearly, somewhere along the line you need to find peace with it all. You could try a fructosamine test if you want to and/or your GP will support it, but I doubt very much indeed my GP would support it on your numbers or mine.
 

Hi,

I would imagine it's a "merry-go- round" regarding the production & lifespan of red blood cells ? Some my not even last as long as thought whilst others may go longer?? ("Circle of life" stuff.)

Now oddly, my estimated A1c from finger pricking & logging is 15% higher than what comes back from the lab..
I feel you maybe forgetting the accuracy tolerance on your meter?
My meter strangely & by coincidence reads 15% higher than a calibrated standard nurse's NHS issue.
 
anyone want to tell me what RBC means? I hate it when acronyms are used with no explanation. We are not all scientist.
 
lucylocket61 said:
anyone want to tell me what RBC means? I hate it when acronyms are used with no explanation. We are not all scientist.
Click to expand...

Red blood cell, within the context of the question posed in the header..
 

Interesting find, @finsit.

So, even if we can't get a reticulocyte count (and I would guess this includes most of us), the relationship generally seems to be be, the higher your hematocrit value (holding reticulocyte count constant -- as we most likely have no value to plug into the equation), the more likely your lab-based HbA1c would come in higher than expected. Alternatively, with a low hematocrit count, your HbA1c would be lower than expected.

This seems to be also what I observe. Similar to @AndBreathe, my HbA1c always turns out higher than expected. One time it was off by about 0.6% or 7 mmol/mol (which, personally, I find quite a lot). However, the probability that the average daily blood glucose readings were in fact as high as indicated by the lab-based HbA1c, was a little less than 2%. In other words, the likelihood that the true average blood glucose levels were, in reality, lower than indicated by the HbA1c was 98%. (Btw, I also made sure, my finger pricks results were quite close to real average blood glucose levels).

This kind of seems to jive with what would be expected based on @finsit's find -- as I also always have a hematocrit value in the higher range between 42.5% and 47.5%.

As to @finsit's question what the corresponding correction factor would have to be, there is probably no answer to this. However, possibly there is the option of writing to the author/s of the article on which the formula is based.
 
My HbA1c has always been higher (significantly) than any other testing method indicates. I learnt this very early on and decided I must have long lived RBC. I simply look at my HbA1c results as a trend line rather than a true result, and it keeps my nurse happy. The only thing that worries me about it is how many newly diagnosed T2s are mis-diagnosed because they happen to have long lived or short lived RBC.
 

That's interesting. I have only had Hba1c tested in the last few years (I am mid-60's) and my first was 41. It was then 39, 40, 42, 40, and 40. Fasting glucose blood tests - which I have had for a longer time - have never been higher than 4.7, which is a much better level. I eat a low carb diet and am very active with a normal BMI. I have always wondered why my Hba1c was relatively high and where all this sugar is coming from! Maybe it is because my RBC are long lived.
 
Fasting levels are never going to correlate as they are a single point in time and could be 15% error of margin.

CGMs are potentially better at guessing your hba1c but not sure if they do, maybe others who have them can say
 
So there is another issue in few of other studies that i have been reading that they are taking average red blood cell (RBC) lifespan somewhere between 40-60 days. I am not sure what is that??

Also this is really helpful study https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5094338/ but still i don't understand what are those non-age related erythrocytes...


 
This is the formula i used to calculate my RBC lifespan

RBC survival in days = 100 / (reticulocyte count (in percent) / reticulocyte life span)

 
Andydragon said:
Fasting levels are never going to correlate as they are a single point in time and could be 15% error of margin.

CGMs are potentially better at guessing your hba1c but not sure if they do, maybe others who have them can say
Click to expand...

In my experience. The Libre CGM set up estimate isn't far off my meter logging estimate.? Just gives a more detailed picture between tests. Fills in the blanks, so to speak.
Think of it a little like coming home & finding your house burgled? Then looking at your CCTV recorded footage & seeing evidence of the "culprit."
I can look at my graph & work out where it went wrong overnight, or whilst preoccupied?

It's also like "big brother" for any HCP with access to the shared data too?
 
Andydragon said:
Fasting levels are never going to correlate as they are a single point in time and could be 15% error of margin.

CGMs are potentially better at guessing your hba1c but not sure if they do, maybe others who have them can say
Click to expand...

Thanks. I have been toying with the idea of self-funding a CGM or just some finger prick testing strips as I would like to see what it is in my diet that causes any spikes. I am still surprised that if Hba1c is a 3-month average, why my fasting levels have never been high at all, and that therefore I must be experiencing some pretty high readings at other times of day.
 
Quite possibly, spikes will come after eating which is normal but in non diabetics they drop reasonably quickly after, hence why there is talk about the testing 2 hours after food

Some foods, especially carb heavy fatty foods will also cause I have found a spike that lasts longer than the 2 hours. Or is a delayed spike you don’t see without CGM

so 15% accuracy coupled with all that can easily lead to a hba1c that the fasting just cannot show you
 
I am very interested in this because my HbA1c results are far worse than my Libre 1/2 results predict.
The Libre tends to be near to a finger prick although not always as close as I would like.
Noting that for the last several months I have been running a Libre almost constantly so the aggregate results should be consistent with my interstitial fluid throughout the period.

The article is interesting, but numbers being quoted don't match my full blood count as regards units.

"Full blood count
Haemoglobin concentration 125 g/L [135.0 - 175.0]
Below low reference limit
Total white blood count 6.9 10*9/L [4.0 - 11.0]
Platelet count - observation 295 10*9/L [135.0 - 450.0]
Red blood cell count 3.83 10*12/L [4.5 - 5.5]
Below low reference limit
Mean cell volume 98 fL [80.0 - 100.0]
Haematocrit 0.38 L/L [0.38 - 0.5]
Mean cell haemoglobin level 32.6 pg [27.0 - 34.0]
Red blood cell distribution width 13.7 [10.0 - 15.0]
Neutrophil count 3.0 10*9/L [2.0 - 7.5]
Lymphocyte count 2.8 10*9/L [1.0 - 4.0]
Monocyte count - observation 0.5 10*9/L [0.1 - 1.0]
Eosinophil count - observation 0.4 10*9/L [0.0 - 0.5]
Basophil count 0.1 10*9/L [0.0 - 0.2]
Nucleated red blood cell count 0.0 10*9/L [0.0 - 0.2]"

The results show mild anaemia; the cause hasn't yet been tracked down.

Haematocrit is given in L/L but the article and comments are in percentages.


From the above posted results my Haematocrit is at the very low end of the normal range, yet my HbA1c is much higher than other indicators suggest.
Which doesn't tie in with the above quote.
Which leaves me wondering if I have an abnormal reticulocyte count or if something else is going on.

If I do have very long lived red blood cells then my past history of HbA1c doesn't really tie in. I don't recall and abnormally high readings in the past.

I would assume that with a low red blood cell count the amount of {whatever} that is tested for would also be lower than normal giving a lower than normal result, but this doesn't seem to be happening.

TL;DR - well confused!
 
Multiply haematocrit given with 100 and it will give you the %age.
To get your retic %age divide UK number by 44.8
 
finsit said:
Multiply haematocrit given with 100 and it will give you the %age.
To get your retic %age divide UK number by 44.8
Click to expand...

So for Haematocrit just move the decimal point (0.39 becomes 39%).
In that case the example using 25% is way below the standard range quoted of 38%-50%.

As you can no doubt see from the test results there is no value for reticulocyte count so I can't apply the calculation.
 
Retic count is not usually the part of CBC, you will need to ask your GP next time and this should be added easily I guess.
 
You must log in or register to reply here.
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn More.…