Is A1c Becoming Irrelevant?

Community Share Your Stories
1

from my blog Don’t Fear Diabetes

If there’s one number, one test that strikes fear and loathing into the hearts of diabetics everywhere, it’s the A1c. Not only is it the main number used to diagnose diabetes, it is the measuring stick wielded

by doctors, CDEs, insurers, and diabetics themselves for evaluating our“control”. But are its days numbered?



anti A1c graphhic

A1c, according to the American Diabetes Association, is “a test that measures a person’s average blood glucose level over the past 2 to 3 months. Hemoglobin (HEE-mo-glo-bin) is the part of a

red blood cell that carries oxygen to the cells and sometimes joins

with the glucose in the bloodstream. Also called hemoglobin A1C or

glycosylated (gly-KOH-sih-lay-ted) hemoglobin, the test shows the

amount of glucose that sticks to the red blood cell, which is

proportional to the amount of glucose in the blood.”

So, essentially, it is a measurement of your average blood sugar over a period of 2-3 months.

The biggest problem with this, as any diabetic will tell you, is that blood sugars move around. A lot. A whole big giant ton. So an average blood sugar over a period of as little as an hour could mask a

huge range of values (it would be very easy to have an average blood

sugar over an hour of 100 mg/dl, which sounds like a very good number,

and have a range of 40mg/dl-220mg/dl, which is a lot less desirable).

So if one hour’s average can hide so much, how about 2160 hours?

Probably a lot more. So why are we so hung up on A1cs?


Well, humans seem to have a need for standardized comparisons. The world is simply too complex to try to understand it all. In order to make decisions, it’s essential to simplify information. The A1c does

that for diabetes. The problem is, while it simplifies things, it does

it in only a marginally useful way. Sure, we can say that someone with

an A1c of 11 is worse off than someone with an A1c of 7.5. But about

the difference between 6.5 and 7.5? Conventional wisdom has it that the

6.5 is preferable (using the generally accepted lower=better paradigm).

But that’s not necessarily the case. Karen at Bittersweet wrote a very important post recently on celebrating her higher A1c.


What we all know (but until recently didn’t have the tools to measure) is that Standard Deviation is as (or, within a certain range, more) important than A1c. When you only test 6-10 times a day, getting

an accurate Standard Deviation is impossible (especially following the

convention of testing 2 hours after meals, well after most post-prandial

spikes have receded). But with modern CGMS, it’s as easy as pressing a

button. While I only have experience with Dexcom software, I’m sure all CGMS (Abbott Navigator, Medtronic Guardian)

have similar functionality. I can easily get my average (mean and

median) blood sugar, standard deviation, highs, lows, and many other

readings. This gives me a far more complete picture than an A1c ever

could. And if I still want an A1c, it’s easy to convert my average from my Dexcom.


Now that CGMS are on the rise, and the importance of blood sugar fluctuations are starting to register as more important than blood sugar averages, maybe we will start to see A1c ushered out as the yardstick

for diabetic care. Instead of hearing “that A1c isn’t where it should

be” maybe people will start to hear “I noticed that your post-prandials

for weekend dinners are much higher than any other meal. Let’s look at

your boluses for these meals and see if we can’t tighten those numbers

up a bit.” Wouldn’t that be more helpful?


2

I’m not a fan of A1C as the Joslin center wants you to be at 8% before going on the pump which can be impossible to achieve sometimes. Plus how is A1C really figured out??? What if you had a 60 carbs for breakfast and you go to the dr and they do an A1C test. Doesn’t what you eat effect that A1C?? I look forward on going on the CGM. I know it doesn’t replace finger sticks but at least it will give an idea of where my sugar levels are heading and going. Perhaps it can further research in curing Diabetes by looking at CGMs. Meanwhile 7% or lower is optimal for all Diabetes to reach, but those of us who just can not lower our BGs needs to try different methods of lowering it and having the correct A1C range.

3

Amy, I agree that setting an A1c threshold in order to approve DME’s is a potentially dangerous practice, that could lead people to pursue worse control, rather than better, out of desperation to have their equipment approved. That said, since A1c is a measurement of a 2-3 month average, what you eat the day you get tested shouldn’t affect the result (and fasting before the test won’t give you a lower number, but of course you need to fast if you are having lipids done at the same time).

4

sisiay, this is why I dont like A1C. Setting an A1C threshold in my opinion is agreeable to yours. I try my very best for a year now and still hover between 10% and 13%. Right now I’m at 11% and that is what it was in Mayish. But yet they wont approve me based on what the A1C was in May. So what gives doing an A1C?? It does change depending on whats going on. Even if they tell us get your A1C down and then we’ll approve but what if when you go back for that A1C and it still isn’t down?? Then what?? You have to wait another 3 months and try again. And the longer you wait to be approved for the DME, can take years when it can help at this moment. I just think there is a need for better help in control even if we are in control…

5

A1C will become irrelevant as soon as the good doctors can bill more for looking at the CGM stats than ordering an A1C test. I don’t hold my breath.

6

I wouldn’t say that A1c is totally irrelevant. I would only say it is not the entire story. It is this one number that can indicate if more investigation into individual BS is necessary?? I mean if i am a doctor and my patient is <6.0 i would look at his daily BS to see if he has a history of lows. I mean 80 might be OK for some people but for some other i would recommend staying above 90.
regarding the highs how bad are the after meal spikes. It is a known fact that even non-D (being a diabetic for so long i don’t want to call non-D as normal people…LoL!!!) experience spikes depending on the food they and hence athletes are advised not to have any high sugary content in their diet. One of the reason i can think of is such spikes can cause weight gain in athletes. So i still feel it is ok to have those spikes as long as BS return to normal at meal+2hours test.

I am of the view that a 6.7 with lower SD is better than a 5.8 with a larger SD!!! I know that this would open a huge debate and most may not agree with my view!! I strongly feel that when in doubt it is better to be a little high for a little while rather than experience a low. especially when u r traveling or forgot ur meter or don’t know how much to blous (especially u r trying a food for which u can’t really judge carbs.)

7

Only if you consistenly wear cgms can you dispense with the A1c. If testing ten times a day is not enough to get proper SD, fifteen times would only be a little better. We test various times per day and do a 2 hour postprandial check if she’s home. We do get A1cs every few months but I agree, not sure of how much weight to put on the A1c. Basically, I want to make sure it is not a high one. Hoping she will agree to wear cgms full time once Dexcom integrates with a pump.

8

If all diabetics have A1c numbers below 6.5 then the A1c itself would be less relevant. But unfortunatly the reality is not very pleasant. So the A1c is still a very good indicator how well someone is managing his diabetes. Not the fine tuning of course but the general management of the mechanics of D and its acceptance.

9

Everyone focues on A1C but little on your BG graphs. At least I havent met a Dr that was interested. Maybe thats why Dr B makes his patients make meticulus records of their blood sugars and carefully examines them.

10

I am going to have to respectfully disagree.

Based of course on what I have read and learned in the last few months, which I admit up front is limited. And I am not a professional, just a scared victim of this disease trying to learn everything I can.

I agree with your objections about the details of up and down BG being lost in the average. But the A1C is so much more than the average of your blood glucose measurements.

My understanding is that the A1C is THE important value. That is because it measures something real. It measures the percentage of red blood cells that are permanently glycated. The more glycated blood cells the fewer blood cells that can deliver oxygen to their tissue destination. A1C directly correlates with the chances of complications down the road, because it measures the mechanism that causes those complications (to the best of anyone’s knowledge).

We can only measure our blood glucose at the time of measurement. And so we manage blood glucose day by day, hour by hour, meal by meal. But except for extreme highs and extreme lows, the BG number itself is not important. We manage our blood glucose IN ORDER TO manage our A1C. Not the other way around.

My understanding is that if your A1C is good, then the occasional high or low BG measurement is irrelevant, and if your A1C is bad, then all your good BG measurements are kidding you, and you are missing the measurements at the times they would have been higher.

Please correct me if I am wrong, but I really think I have this mostly correct, frustrating as it is for you and for me.

11

The graphs of all my BG measurements are an aid in my learning what makes it go up and go down, and to correct what might be wrong in my attempts to control it. But the goal is getting the A1C down, and once I know my A1C, the graph of the BG meaurements that led to it becomes irrelevant. If my A1C is good I am doing a good job, and if its not I am not, regardless of what the graph says.

12

JeffD- Your point is good, but I think we disagree only in semantics. Yes, the amount of glycated blood cells IS important. But if you were to have a monthly average blood sugar of 90 mg/dl, and a standard deviation of 45, then you should have a very healthy level of glycated blood cells, and you will also know a lot more about HOW you were able to achieve such a good level. I would amend your last sentence to say "If my A1c is good and I am rarely low I am doing a good job…"
And yes, it’s true that this only applies w/ CGM. Finger sticks will never give enough data to give meaningful averages/standard deviations.

13

@JeffD thanks a lot for your post!! It is very informative and well said.Once again Thanks!!

14

JeffD,

I came across a lot of literature that claims that mean BG and SD are independent risk factors. Your best bet is to rein in both.

Should minimal blood glucose variability become the gold standard of glycemic control?

15

Thanks for that. Its pretty interesting.

I notice they say A1C below 7% “deemed appropriate for reducing risk…”. That was the thinking in 2005. I wonder if they did an update to todays thinking, where they looked at A1Cs below 6%, and the affect of an A1C of say 5.5% with wildly varying BGs and another 5.5% A1C with BGs controled to a small range.

Certainly reining in both makes a whole lot of sense, regardless, because widely varying BG, even if it is not critical, is an indication that things are not in as much control as one would like.

This so much sucks. I hate it. There are soooo many things I would rather pay attention to than all of this stuff. I play fiddle and there are tons of fiddle tunes I could be learning but no, I have to “study” my diabetes. I have to jab myself with sharp things, and do trigonometry just to eat a snack. I am so tired of this. I really am. But I discovered that bad moods and frustration raise my blood sugar, so I can’t even have a petulant frenzy.

16

It could be very helpful information, but most of us are not on pumps and won’t be anytime soon.

I don’t see this replacing a lab test anytime soon either - lab tests are more accurate than pumps.

I’ll stick with the A1c and my meter readings (on a calibrated meter). I’m sure my endo would back me up on that too.

17

@jag1 of course, you’re right, the A1c isn’t going to be replaced anytime soon. My point is more that as diabetics, using that as the standard for measuring our control is becoming less necessary with the far greater data set afforded by CGMS. Sure, if you don’t get good info from your CGM, it’s useless, but many people (myself included) find good correlation between meter and CGM. In a way, it’s a self-fulfilling pursuit: the less your BG fluctuates, the more likely that the CGM readings are correct. Sure, under 60 or over 180 readings tend to lose accuracy, but you don’t want to be there anyway.

What I meant by “What we all know … is that Standard Deviation is as (or, within a certain range, more) important than A1c.” was that the difference between a 6 and 6.5 A1c is probably less of a predictor of future complications than the amount of time spent above the renal threshold (140 mg/dl), which, for good A1cs, would correlate with higher Standard Deviations.