A1c and red blood cell longevity

Diabetes Technology Glucose Monitoring
1

Continuing the discussion from Tips to quickly lower blood sugar readings:

As often happens in our discussion threads the discussion focus veered off topic. So when @Eric30 suggested that we start a separate thread, I decided to do just that.

I’ve always checked my fingerstick blood sugar more than the average person. For many years I averaged between 10-15 checks per day. Eight years ago I started using a CGM and my quantity of blood glucose data skyrocketed.

With the volume of data I collected I had confidence that it revealed representative blood glucose averages for 14, 30, 60 and 90 day periods. I was also aware that a math formula could be used to convert an average blood glucose number to its corresponding A1c number.

What I found when I did this over the years is that my lab derived A1c corresponded to a significantly higher average blood glucose than reported by my fingerstick meter and CGM. Stated another way, my meter and CGM averages predicted an A1c 0.5-1% lower than the lab measured.

I’ve read that the A1c test is based on an assumption that red blood cells live for a certain duration, something within 90-120 days. I’ve also read comments from the medical literature that for people whose red blood cells live longer or shorter than the assumed longevity in the formula, then their A1c would read falsely higher or lower. In my situation, with a lab A1c 0.5-1% higher than the averages my meter and CGM predicted, it made sense to me that my red blood cells may live longer than the number built into the A1c protocol.

I’ve raised this issue in a few other threads and was surprised that most responses say their meter and CGM averages do a very good job predicting their A1c’s.

I know there are other explanations for my experience. The obvious one is the fact that I’m basing my conclusion on averages from my blood glucose meter. Our meters are not the most accurate tools but it’s what we have.

At my last A1c blood draw, I did three fingersticks and recorded these numbers: 80, 85, 86 mg/dL (4.4, 4.7, 4.8 mmol/L). This calculates to an average of 83.7 (4.6). The lab came back with a glucose value of 82 (4.6), a difference of only 2%. So I conclude that my meter (Accu-Chek Aviva Connect) is reasonably accurate and using it to calibrate my CGM keeps my CGM numbers relatively accurate. I like to do this exercise at every A1c lab draw.

When my blood was drawn for the glucose and A1c, here are the averages from my CGM along with the predicted A1c:

90-day average = 99 = 5.1%
60-day average = 94 = 4.9%
30-day average = 90 = 4.8%
14-day average = 87 = 4.7%

My doctors have told me over the years and I have read that the most recent period is weighted more than the earlier periods. So, the last two weeks are weighted more heavily than the first two weeks of the three-month period. I think my CGM averages predicted an A1c of 4.9%. My A1c came in at 5.5%.

Now, I’m not complaining. I’m thrilled that my A1c is in the “non-diabetic” range. So I just used a lot of words to express a simple thought: I believe the explanation for my higher than expected A1c is due to my red blood cells living longer than the average.

@Eric30 was interested in the explanation of why I think my red blood cells live longer than the A1c formula presumes. I’d be happy to hear your thoughts, if any. Sorry, @Eric30 if the explanation provides too much detail.

1 Like
2

My CGM-predicted A1C (when the Dexcom program did that) was always .5 to1.0% lower than the lab value.

The red blood cell longevity could be a factor, but, seriously, who’s going to spend the :moneybag: on a study?

My A1C (the last) dropped about 0.4% and I didn’t do anything differently–except I started taking an iron supplement.

But it could just be a coincidence.

1 Like
3

You’re right. It may be an issue of interest to few people but the A1c is a flawed measure anyway. I wish the clinicians and researchers would use things like time in range. TIR is a better measure of glycemia, safety, and quality of life.

2 Likes
4

Terry, unfortunately, that fits the scenario of the hemoglobin living shorter than the “offiical” 120 life span (believe the 120 includes the “teen years” so to speak as reticulocyte) in my understanding. If it were to live longer, it would become more glycated, thus raise the A1C. The 90 day average takes that into account when it starts functioning as hemoglobin and functioning in it’s traditional role of carrying oxygen, etc, thus becoming glycated, hence the traditional 90-day average. But as pointed out in a lab and your discovery shows, for us diabeteics, maybe due to the glycation, the average “adult” life span is a bit shorter than the traditional 90 days.

Edit: Using NightWatch/xDrip, I’ve noticed the same. Hence my curiosity and deep dive into the topic (general internet and NIH archives).

Edit II: (Blessing/curse with reading and remember nearly everything) – They still don’t know C-Peptides purpose yet, perhaps that plays a role in hemoglobin lifespan. But as I’ve seen with everything else, even the “useless” appendix has a recently discovered role/function. :slight_smile:

Edit III: BTW, (this is the curse I live with, mind never stops): WIth the iron correlation, that is probably attributed to allowing the body produce more/new baby hemoglobin (reticulocyte count) versus the lifespan being longer, hence lowering your A1C. Iron from my understanding has more to do with birth versus lifespan.

1 Like
5

I had the opposite but explainable situation. I got an A1C of 4.6 when it should have been in the mid 5 range. It was during chemo when my blood counts were getting killed, wait a week for recovery, then whack them again.

1 Like
6

MM1 - Thank you for mentioning the opposite effect. Went back and re-read Terry’s post and realized I didn’t get all of Terry’s post. In this case of his A1C coming back higher than it should, Terry, you are correct in that math demonstrates the longer lifespan. That would correlate with longer lifespan if it came back higher than it should.

2 Likes
7

My experience has been similar, lab A1C consistently higher than indicated by CGM average BG. My average BG indicated by the A1C result seems to be consistently 16-18 mg/dl higher than the 90 day average provided by Dexcom Clarity. (I run in range of 5.9 to 6.4) My own theory has two parts: (1) The CGM does a poor job of catching peaks when BG rises sharply and then falls sharply. I have caught some of these peaks with tests and even when calibration is otherwise good the CGM seems to lop off the peaks. (2) Lows (below 70) do a lot to reduce our CGM average BG but do they have as much significance for the A1C lab test? I am guessing no.

1 Like
8

Sometimes my writing is not as clear as it could be. I feel like I wrote a mountain of words for an issue that affects few people. It’s more academic and less pertinent to our community. Thanks for your clarifying comment. I wasn’t sure how to respond to your first comment.

9

So, you’re saying the A1c is probably a true number since the CGM sometimes cuts off the hyperglycemia excursions. These CGM unrecorded excursion then skew the CGM data and show a lower average glucose value than it really is.

Did I get that right? If true, that could explain the discrepancy.

10

Terry, to add to your comment in being academic, this might actually be more applicable to help others understand the fight/struggle when they go to the doctor’s office and get results that don’t add up in their mind to what they expected. Like you and I, we are ademate about trying to find accurate glucometers (another topic on this site), not all glucose meters are “accurate” as advertised/understoood to be and as John pointed out, they’re only a snapshot in time. And CGMs, they lag in time so depending on the cross over of glucose rising and when insulin meets it’s peak, the peak can be missed.

Hence, the doctors typical/general advice to do the best you can and not dwell on it. If we spend that much time worrying/working on it, better to apply that effort towards enjoying life.

1 Like
11

Terry, Yes that is my theory.

12

Add to that the POC A1C that many doctors offices do, and there’s another margin of error. When possible I try to get A1C included on any other labs that are ordered, but timing doesn’t always work out, so I go with the POC.

Point-of-contact.