Question about the A1c

I will be having an A1c done next week and want to know how much post-prandial spikes affects the test. I would assume that it balances out if you have other numbers in the 100’s. My spikes are between 200-300 mg every time I eat. I am not yet diagnosed so I have no idea about the test and how much the spikes will affect it. Any info in this would be appreciated!

DN has had roughly the same A1c’s in the low to mid sixes, and high sixes if numbers are abominable, and also used to spike to the high 200s, now usually low 200 or under. It must balance out because A1c still decent. Spikes, try our best, but seems like a hopeless cause. Never go a day without a spikey period. Using new insulin and postprandials better with Apidra but crazy and very high spikes if any interruption at all so we shall see.

You’re right, A1c’s are an average I think over several weeks, so if all you have are 300’s and 30’s your A1c will average pretty good! But we all know that it’s not. All it is is an estimate to give an idea of how you’re doing.

A1c are time-weighted averages. In other words, If you are 200 for 30 min/day and 100 for 23.5 hr day your average will be 102

and your A1c should be about 5

That’s simplistic, but it demonstrates the principle.

Despite other comments: the A1c is not an average. Lows will NOT compensate highs. The A1c measures the binding between haemoglobin and glucose. This binding is permanent and will stay stable until the blood cell ultimately dies after 3 months. It is fact that to establish the stable connection the blood glucose must be elevated for a longer period of time. Thus shorter higher spikes may not be visible in the A1c. It is still the most important long term indicator for the mean glucose level. A healthy A1c in combination with a moderate deviation of blood glucose within one day are the goals to achieve because it simulates the glucose regulation of a healthy person.

You state other numbers in the 100’s. Not 100… so with the spikes - assuming they are not really ‘spikes’ but rises. Your A1c should not be in the 5’s but in the 6’s. OK…that’s my guess for the day! I’m not even going to say high or low 6’s - not enough info for that guess (and really I can be wrong BIG TIME)
If you are not diagnosed and your fasting is still below diagnositic and your A1c is also below suspect if I were you and you have the insurance you want I would tell your doctor about your posts. Lots of ‘AND’s’ there…
The A1c is also not a true 3 month number since your red blood cells die off and replace so your A1c is more weighted on the last few weeks.
go to Jenny has info on the A1c there that will be helpful to you.

That’s not quite right, either. Lows don’t “compensate” for highs, but the binding rate doesn’t change, so less is bound when less is available in the blood to bind. You will never have a BG of 0, so you will always have binding going on.

A low is 50 and a healthy fasting number is 80. That is only a difference of 30 mg/dl. So you can not compensate 4 hours at 300 mg/dl with 4 hours at 50 mg/dl. You need significantly more hours being low to have a compensation. Yes, it is possible but only with medication like insulin and intention. It is not possible with untreated glucose intolerance as the starter of this discussion has. For her evey spike will be part of the A1c.

I heard it put once like this:

Low A1c
When your A1c is low (<7), Post meal spikes contribute alot (70 %)

High A1c
When your A1c is higher (>7), Post meal spikes contribute less and less (at A1c of about 10, only 30 % now)

I can’t remember where I read it, but a quick search confirmed my memory, as shown in this article here:


Could you elaborate further on an A1c not being an average? Are you saying that only highs that bind are evidenced in A1c results? Everything I’ve read says it is an average over approximately a three month period, with more recent readings have more of an effect.

It is an average but the binding behaviour is not balanced evenly. In a healthy person the fluctuation of blood glucose is between 80 and 120 and this leads to an A1c of 4.5 to 5.5 (maybe lower). If we experience a high the binding rate excels over proportion to the glucose increase. So higher number leave a significant trace that is found in the A1c. By having lower numbers you can not make these bindings disappear. Of course you will not add more bindings while being low. But a healthy person stays at 80 for the whole night. This is a very long time. A low starts with 50 and this is only 30 mg/dl less. The binding rate difference between these two states is not as significant as on the high side. As a result you need to have more hours at low numbers to compensate for hours of being high. Because of this imbalance I would not use the term average. To me average implies an unbiased binding behaviour. Maybe I am a little strict or scientific here.

What makes it difficult is the fact that different eating styles could bias the binding behaviour too. A high fructose eater will have a slightly higher A1c than a normal eating person (if they had the same glucose levels for 3 months). This is because the fructose has a slightly higher binding rate than glucose. Furthermore individual genetic differences may cause slightly different binding rates from individual to individual.

Thanks, Holger.

The A1c measures the cumulative glycation (damage) in your red blood cells. Glycation occurs at a rate that is linearly proportional to blood sugar levels, this linear relationship is backed up by numerous studies. Because red blood cells have a specific life, measuring your HbA1c if everything is steady state will result in the HbA1c being linearly proportional to your average blood sugar. There are a number of equations with this mapping, the most recent being the ADAG ( That being said, there are a number of things which will mess up this nice clean relationship. Holger notes one, which is fructose. Another is that the HbA1c actually heavily weights more recent values. And the third is that there appears to be significant individual differences in glycation rate, perhaps as much as +/- 20%. This is the so-called “high-glycator low glycator” theory ( Often the HbA1c will not correlate well with meter readings and this variation can be an ongoing source of confusion.

In general, the ideal HbA1c will be insensitive to variability in blood sugar, and hence it won’t really show wild blood sugar swings. It will however capture time spent with higher postprandial blood sugar readings as having elevated blood sugars for a period of time will raise your average. I have found the best way to get a consistent long term measure of your postprandial is to compute your standard deviation (SD). You can take your readings from your meter or CGMS and compute this number. Keep your SD below 20 and you are doing well, keep them below 10 and you are a super star.

Every doc or endo I have ever seen always refers to the A1c as an overall pic of your diabetic control,and in fact is refered to as an average. A specific A1c level equates out to a bg average. Now, we all know that averages are effected by “outliers.” It may also depend on whether you are seeing a general doc or actual endo as to how they interpret your A1c. I have heard of docs who have told people not to worry about their bg levels when their A1c comes back in the normal, or just slightly above normal range. This happened to my father for years before a doc finally said, “I don’t care what your A1c says, you are having bg levels above normal and that means Type 2 diabetes.” Everything I have ever been told is that an A1c can only reflect up to 3 months of prior levels/control.
You must also realize that the A1c test is not perfect. There are too many people, and I assume there are some on this site, whose daily levels do NOT correlate to their A1c level. As previously mentioned, my father is one of them. Plus, the A1c takes into account what your body is doing all the time, not just when you are testing. There could be times when you are not testing that you are having high levels you do not know about (hence the onset of the continuous bg monitoring systems). There could also be times you are not testing that your levels are perfectly normal. It has always been told to me that is why the A1c is seen as the best (but not the only) indicator of diabetic control.od
Good luck on your test. And although you haven’t officially been diagnosed, I am sorry for you to be joining our club.

This may/may not be related to bsc’s post but:
your post meal bg should generally not be 72 mg/dl or 4 mmol greater than your premeal (fasting) levels

The 7.0 rule is outdated. To prevent long term complications you should be at least below 6.5. My personal recommendation is to aim below 6.0 for a type 1. Even this level has a likelyhood of complications above 0 but it is significantly smaller than the likeyhood at 7.

Hi Holger,
Momto3 above wants to know how her post meal spikes contribute to her A1c.
I think i have been the only one in this discussion to answer her original question.

A1 c - best ? mmm

Alot of studies show higher mortality associated with strict/tight glucose control (usually bc of the meds).

So what is best A1c?
Blood Sugar 101 site ( says if your A1c is over 5.5-6 you are probably in slight trouble. If you believe this (and I do), tight control should be aimed for. But at what expensse ? Mortality?

Where I’m going with this is many believe that adequate control (A1c between 6-7 ) is critical, and probably ok. However, to live a healthy life in the years to come, a controlled blood pressure (less than 130/80) is ALSO critical to avoid the microvascular complications (of small blood vessel), WHICH CAN LEAD TO HEART ATTACK, STROKES, RENAL NEPHROPATHY, RETINOPATHY etc etc etc


What studies are you referring to?

The ACCORD study, which cites lower A1cs & better control contributing to mortality, was a hugely flawed study. Unfortunately, most healthcare professionals only read the summary & headlines, not the methodology. The ACCORD study used test subjects who had existing heart disease. Subjects were given a highly risky combo of meds. They were also forced to eat an extremely high carb diet. A prescription for disaster–yes!

Better research studies refuted ACCORD’s findings.

Lower A1c’s (6 & below) are the route to avoiding microvascular damage.

Hi Gerri,
Yes, there were many flaws in the study.
So was the original question answered here - no - you could say the study was inconclusive -
So it remains… is tight control better ? Is there reduced risk of complications (without an effect on mortality?) ?

Point I want to make in this discussion is that there is a strong blood pressure control connection in all this:
…to live a healthy life in the years to come, a controlled blood pressure (less than 130/80) is ALSO critical to avoid the microvascular complications
… have a listen to the person who suggests this here

Dr. Claude K. Lardinois, an out-of-the-box-thinking endocrinologist, with strong opinions on this subject. Dr. Lardinois has many years of vast experience in treating diabetes,

PS Holger is referring to Type 1 s. So he is probably correct (thanks Holger; but you should provide the source for this as well to convince me).
…The debate continues in the case of T2s

Sorry, I don’t understand why you’re presenting the ACCORD study since you know it was flawed.

I’d say that the ACCORD study was more than inconclusive. Having read the study & critiques of it, the conclusions were incorrect. It’s harmful that doctors now advocate higher A1cs as a result of this terribly faulty research. ACCORD grabbed the headlines, while other better research hasn’t.

Have you read the ADVANCE study? It shows the results of tighter control as highly beneficial to preventing complications. ADVANCE was the largest study undertaken, involved over 11,000 subjects who were followed for five years.

ADVANCE test subjects were Type 2 diabetics, so that should answer your question. Aside from their results, it’s logical that tight control is best for any diabetic. Why would type make a difference?

Good BP is critical, as well.