Glycated Hemoglobin: A test to measure long–term blood sugar control

Executive summary

Prolonged high blood sugar levels lead to diabetic complications. Glycated hemoglobin, or HbA_1c, is a good test that measures long–term blood glucose control.

Blood glucose attaches to hemoglobin in the red blood cells slowly over a period of time. This irreversible process, Glycation, leads to glycated hemoglobin, or HbA_1c. The more the blood sugar, the higher the formation of HbA_1c.

Since red blood cells live about 110 days, measuring the percentage of hemoglobin that is glycated, roughly tells us about average blood sugar levels over the past 2 to 3 months.

American Diabetes Association (ADA) considers HbA_1c below 5.7% as normal, 5.7% to 6.4% as pre–diabetic, and 6.5% and above as diabetic. The WHO considers HbA_1c below 6% as normal, instead of 5.7%.

Lowering HbA_1c reduces diabetes–related damage in kidneys, eyes, and nerves by 25% for every 1% reduction in HbA_1c. As a result, ADA advises bringing HbA_1c below 7.0% for a good sugar control. The International Diabetes Federation, on the other hand, advises 6.5% as the target level for HbA_1c. The article discusses about whom to follow.

There are various situations that give wrong HbA_1c readings. For example, different anemias, nutrient deficiencies, G6PD deficiency, stress, ethnicity, gender, surgery, and even the blood donation status affect HbA_1c levels. Finally, the classical interpretation of HbA_1c as a 2 to 3 month average is incorrect.

When glycated hemoglobin cannot be used as a test, tests for blood fructosamine or glycated albumin are used.

Continuous blood glucose monitoring remains the Gold standard. Plus, there are many new technologies coming up that will make long–term blood sugar monitoring easy. All of this is discussed in the full article.

HbA_1c, A_1c, glycated hemoglobin, or glycosylated hemoglobin (which is an incorrect term, as we will see later) test is a useful medical test for understanding long–term blood sugar control of the body.

Diabetics are often advised to do this test every 3 to 4 months to assess their blood sugar control.

Many medical associations also recommend this test for diagnosing diabetes in untested people (new research suggests caution against this, as you will see at the end).

We will look at the science behind this test, its advantages, and situations in which this test can give incorrect results, leading to wrong treatment.

Blood sugar levels

Blood sugar, blood glucose, plasma glucose, and serum glucose are terms used interchangeably. They refer to the amount of glucose in the blood plasma.

Glucose is needed for energy by cells in our body. Our blood transports this glucose across the entire body. Depending on when and what we eat as food, and our physical activities, the levels of blood glucose go up and down.

Tests for blood sugar levels

High blood sugar is known to damage body organs, leading to complications. So, knowing the level of blood sugar is important. This is more relevant in diabetic people, whose bodies cannot control blood glucose levels well.

Short–term, or intra–day, variations

Tests, such as fasting blood sugar, post–meal blood sugar, 2–hour post–glucose blood sugar, random blood sugar, and oral glucose tolerance test, help with this. They offer blood glucose levels at one time, or changes in it over a few hours.

Long–term variations

Continuous glucose monitoring and HbA_1c test fall into this category. Read here about some more upcoming technologies that measure blood sugar continuously. That is the future of blood sugar measurement.

My father was a diabetic, and he was extremely good at gaming the system. A couple of days before his blood sugar test was due, he would control his intake of carbohydrates. No wonder he would ‘pass’ his fasting blood glucose test with flying colours. Then, with a sly smile, he would ask us that he should be allowed to eat sweets.

A test that would track blood sugar levels over a longer period of time, such as a month, would have caught my father’s trick.

It is found that long–term presence of high blood glucose causes more damage compared to a short–term (less than a day) spurt in it. That was another reason a test was needed for assessing long–term blood sugar levels.

In 1970s, a test called glycated hemoglobin, or HbA_1c, was proposed that promised to quantify average blood sugar over 2 to 3 months.

The HbA_1c test

The test involves withdrawing blood, like any other blood test. The amount of a protein, glycated hemoglobin, in the blood is measured.

What percentage of total hemoglobin in the blood is glycated hemoglobin? That is the HbA_1c value — the test result — such as 6.3%.

Knowing this percentage, a empirically–derived conversion formula can tell us the average blood glucose over the previous 2 to 3 months. This is called estimated average glucose, or eAG.

Estimated average glucose, or eAG = (28.7 × HbA_1c) − 46.7

You don’t need to know this formula. The testing laboratories calculate and mention this value in their HbA_1c reports.

For example, HbA_1c of 6.3% corresponds to 134 mg/dL. It means that over the last 2 to 3 months, your blood sugar may have fluctuated a lot, but its average value was 134 mg/dL.

The science behind the test

When blood sugar comes in contact with proteins and fats, it attaches to them. This process is called Glycation. Glycated proteins and fats are irreversibly changed chemically.

This process continues further and produces very dangerous Advanced Glycation End–products (AGE). Many diabetes–related complications as well as health conditions are directly linked to AGEs. However, we will discuss that in some other article.

Red blood cells and hemoglobin

Our blood has red blood cells, which contain a protein called hemoglobin. Hemoglobin carries oxygen in the blood.

Actually, there are many hemoglobin molecules. Hemoglobin A is the prominent one, called adult hemoglobin, or HbA. It is about 92% to 97% of hemoglobin in our blood.

About 6% of this HbA is a fraction, or a derivative molecule, HbA₁. And, nearly most of this fraction HbA₁ is a sub–fraction, HbA_1c. So, HbA_1c is about 4 to 6% of total hemoglobin in a healthy individual.

Glycation

When blood sugar attaches to HbA, we get glycated hemoglobin, or HbA_1c. This reaction is concentration–dependent. That is, the higher the concentration of blood sugar, proportionately higher is the newly created glycated hemoglobin.

Glycated hemoglobin

This process continues throughout the life of red blood cells. As a red blood cell lives longer, more and more of its hemoglobin is converted into glycated hemoglobin.

Even at normal levels of blood sugar, some glycation is always taking place. If you measure HbA_1c in people with very good blood sugar control, it will be between 4% and 5.7%.

When blood sugar rises, the glycation increases in a linear proportion. More blood sugar, proportionately more glycation and HbA_1c.

Also, the longer is the presence of higher blood sugar, higher is the glycation. So, HbA_1c reflects how high was the blood sugar and for how long it was.

Adding it all up

If you plot a graph of blood sugar versus time, say over 3 months, the total glycated hemoglobin formed will be proportional to levels of blood sugars all along.

In high school mathematics, you may have learned something called area under the curve (AUC). HbA1c essentially corresponds to the area under the curve of blood sugar against time. You can call it an integral of blood sugar levels over 3 months.

Thus, glycated hemoglobin will represent average blood sugar levels over the 3 months.

Those of you who are generally inquisitive like me, below is a YouTube video that explains how area under the blood sugar curve will give you an average value of blood sugar over a certain time period.

While it is not at all needed to learn about HbA_1c test, people like me understand the HbA_1c test better, if they can visualise how the glycation process, kind of, integrates blood sugar levels over a period of 2 to 3 months.

I also wish such videos had existed when I was learning integral calculus in college. I, for sure, did not understand this concept in those days (1980s).

A bit of important digression

Another reason is the purpose of this website is not to offer any blanket advice. It is to make you think, question your biases, and derive your own conclusions, in light of modern science.

I have found that people who look for ready answers, struggle with their health all their life. When the scientific dogma changes in light of new evidence, they are flustered. So, they resist change, often staying with their comfortable, and no–longer–correct, views.

That is because their ‘ready answers’ have been proven wrong now. And, no one knows the right answers. So, they don’t know what to follow and whom to trust anymore.

Recollect the recent debates about fats versus sugars, statins and cholesterol, coconut oils: healthy or bad?, etc. Remember how frustrated some of you were, since the research community did not seem to agree on these?

Who is at fault?

Well, in my humble opinion, the fault is with you, not with the scientific community.

At the leading edge of science, our knowledge always keeps changing. In our ancient past, people believed in flat earth. And, scientists like Galileo had to face a death penalty to say things that primary school kids today know with absolute conviction: that earth revolves around the sun.

The ever–changing scientific scenario

The conclusions of science have changed and will keep changing. Some people are scared of this change. But, to me, this is a fascinating benefit. Every bit of new and genuine research is the nugget of wisdom that I did not have to invent or discover. I just have to know how to apply it correctly in my life.

I believe one should look at scientific research based on first principles (back to basics) and common sense. Never take any shortcut advice from anyone; learn the basics and master the methodology of science: how to think correctly about data, how to connect the dots, and how to proceed in light of uncertainty.

When faced with ambiguity, artificial intelligence proceeds by attaching varying weights to different views or findings. Why not do the same with your natural intelligence? Proceed by ever–decreasing surety of your older opinions, and ever–increasing certainty of the newer viewpoints.

Welcome to the jungle

For example, the scientists are now finding that there are more variants of diabetes. Here is a Mar 2018 article published in the renowned journal Lancet that says that there are 5 types of diabetes, and not two (type I and type II), as we always believed.

In plain English, here is an article describing these findings: Are there actually 5 types of diabetes?

It means that the treatment for a person who has type II diabetes (old classification), will differ based on whether he is in group 3 (SIRD), group 4 (MOD), or group 5 (MARD), of the new classification.

Haven’t you noticed that normal diabetes treatment does not work properly in some people, while it does in others? This may be the reason.

Now, what is your attitude towards these new findings? Frustrated? Upset? Angry that you have been getting treated wrongly all these years? If I were you, my attitude would be of gratefulness.

I am glad that, at least now, we know why some of us do not respond to standard diabetes treatment. My grandmother and my aunt were not so lucky. Both died of diabetes, with the aunt having to undergo a leg amputation due to diabetes–related gangrene.

There are no right answers

In our school days, we were taught that there is an answer to everything. There is one right answer, and everything else is a wrong answer. So, we are trained to find that right answer.

Later in our lives, we found that in some situations, there are no clear answers. No one knows what exactly is going to happen with global warming or world economy. You don’t even know if your father’s cancer is going to be cured, or your software programming job is going to survive the onslaught of machine intelligence.

But, you are taught that there is an answer to everything. So you seek that answer all the time. And when you don’t find it, you get frustrated.

The way forward

In my doctoral work at Princeton University, USA, I learned that at the leading edge of science, there are no answers. No one knows what the right thing is. Yet, you need to proceed, one step at a time: juggling multiple views simultaneously, adjusting their weights all the time as new evidence comes about. This thinking was completely different from what I had learned in my school days.

But, that is the methodology of science and my articles on this website will always aim to stick to it. I will highlight scientific evidence, and mention the right way to understand and assimilate it in your life.

There are no ready–made solutions in life, and I, for one, will not offer any on this website. And, to expect them at the cutting edge of scientific knowledge is to be like a school kid — always believing that someone else knows the final, right answer.

My advice: Grow up! And, welcome to the real world.

Hemoglobin recycling

Coming back to our HbA_1c test, what happens to this glycated hemoglobin as time progresses?

Red blood cells have a life of about 110 days. When they die, they are taken to an organ called spleen. Their hemoglobin is broken down and recycled.

The heme, or iron, component is carried to the bone marrow by proteins called transferrins. New red blood cells are born in the bone marrow, where this iron is used. Such new cells do not have any glycated hemoglobin since their hemoglobin is freshly formed using the recycled iron.

The globin, or protein, component is converted to bilirubin, a yellow substance that is secreted as bile. Bilirubin ends up in our stools and is excreted. Out goes the glycation component with this.

Every day, some highly glycated, old red blood cells die and non–glycated, new red blood cells take their place. Since about 1% of the red blood cells are recycled daily, your HbA_1c reflects nearly 100 days of glycation of hemoglobin.

The percentage of hemoglobin that is glycated is your HbA_1c number. If 6% of all the hemoglobin in the red blood cells is glycated, HbA_1c is 6%.

Standardising the test results

This section offers some perspective on why HbA_1c numbers are given in different units in different countries.

When the test was introduced in the 1980s, HbA_1c percentages used to be all over the place, for the same amount of average blood glucose in the blood. This is because various methods were used for measuring it.

Therefore, in the United States, they initiated something called the National Glycohemoglobin Standardization Program. This was to standardise the HbA_1c test results.

They used a major clinical trial called DCCT (Diabetes Control and Complications Trial) for benchmarking the HbA_1c values. Incidentally, that trial was started in 1982 and its first set of results came out in 1993. The trial also gave us the first glimpse into various diabetes–related complications.

After this standardisation, HbA_1c results became far less variable and hence, more reliable. These HbA_1c results are called DCCT HbA_1c and are given in percentage (%). These are the units used in the USA and some countries such as India.

When someone in India says that his HbA_1c is 6.4%, he is actually saying he has “DCCT HbA_1c of 6.4%”.

Europeans, meanwhile, were using a different measurement unit for HbA_1c. They measured HbA_1c in something called IFCC (International Federation of Clinical Chemistry and Laboratory Medicine) units. These HbA_1c results are called IFCC HbA_1c and are given in mmol/mol.

The conversion equation is:

IFCC HbA_1c (mmol/mol) = [DCCT HbA_1c (%) — 2.14] x 10.929

Once again, there is no need to remember any of this. It is meant to tell you that both HbA_1c are exactly the same values. They are not measured differently; they are just expressed in different units.

Eventually, the American Diabetes Association, European Association for the Study of Diabetes, and International Diabetes Federation agreed that HbA_1c will be measured in IFCC units. In future, all of us will start hearing about HbA_1c in IFCC units.

Interpreting test results

The test gives an HbA_1c number as a result. Table 1 below shows diagnosis of diabetes, versus DCCT HbA_1c, IFCC HbA_1c, and estimated average glucose.

* For confirmation of diagnosis (first–time conclusion) of diabetes, doctors require a second measurement. One test result is never relied upon, unless there are clear symptoms of diabetes.

Debate on HbA_1c cutoffs

The World Health Organisation (WHO) does not agree with the above cutoff for normal levels. They have a more lax definition. They consider DCCT HbA_1c of lower than 6% as normal, which is IFCC HbA_1c of 42 mmol/mol. This corresponds to average blood sugar (eAG) of 126 mg/dL or 7.0 mmol/L.

This article published in 2012 in the British Medical Journal also gives the WHO numbers as cutoffs for the diagnosis of diabetes.

Contrast this with the American Diabetes Association’s cutoff, in the table 1 above, of 5.7%.

Here is a conversion calculator of various HbA_1c and eAG values in different units.

Don’t get too lost in all these complicated numbers and units. Just select the units that are followed in your country and understand the diagnosis.

So, what should the cutoff for HbA_1c be for a normal person? Whatever your doctor chooses. Remember that whether your doctor calls you a pre–diabetic or normal, your blood sugar is the same and should need the same management. So, let your doctor take a call on this.

Complications of persistent high blood sugar

High levels of blood sugar for prolonged periods increase the risk of vascular complications of diabetes.

Macrovascular complications

Diabetes can damage large blood vessels that supply blood to large organs in the body. It can lead to:

• Heart disease;
• Heart attack;
• Heart failure;
• Stroke;
• Slowed emptying of the stomach (gastroparesis);
• Gangrene and resultant amputation.

Microvascular complications

Diabetes can damage small blood vessels that supply blood to delicate organs in the body. It can lead to:

• Kidney failure (nephropathy);
• Blindness (retinopathy);
• Loss of sensation, especially in the feet (neuropathy); and
• Erectile dysfunction.

So, it makes sense to lower the average blood sugar, and HbA_1c.

Why lower HbA_1c?

Two very large studies — the Diabetes Control and Complications Trial (DCCT) and the UK Prospective Diabetes Study (UKPDS) — showed that lowering HbA_1c reduced complications of diabetes.

After 30 years of follow–up in patients involved in the DCCT trial, it was found that patients who had maintained HbA_1c readings below 7% (or 53 mmol/mol) had significantly lower risks of kidney damage, eyesight damage, and heart problems.

The patients who underwent intensive blood sugar lowering therapy showed compared with patients not on intensive therapy:

• 57% reduction in heart disease, stroke and cardiovascular death;
• 39% lower incidence of early signs of kidney damage (microalbuminuria);
• 61% lower incidence of higher levels of protein in the urine (macroalbuminuria); and
• 46% lower risk of retinopathy.

The UKPDS and the DCCT showed that lowering HbA_1c by 1% (or 11 mmol/mol) reduced the risk of diabetic nephropathy (kidney), retinopathy (eyes) and neuropathy (nerves) by 25%.

An article published in 2000 in the British Medical Journal showed that people with type 2 diabetes who reduce their HbA_1c by 1% are:

• 19% less likely to suffer cataracts
• 16% less likely to suffer heart failure
• 43% less likely to suffer amputation or death due to peripheral vascular disease.

How much to lower HbA_1c

Based on the above results, the American Diabetes Association (ADA) recommends HbA_1c to be below 53 mmol/mol (7.0 DCCT %) for most patients.

What complicates matter is that the International Diabetes Federation and the American College of Endocrinology prefer more aggressive control. They recommend HbA_1c values below 48 mmol/mol (6.5 DCCT %).

Which recommendation is right?

Both.

Lower HbA_1c values are better for preventing long–term diabetic complications. However, in some elderly people, such aggressive blood sugar control increases the risk of hypoglycemia, or dangerously low blood sugar.

Since your doctor knows your situation — your general health, allied medical problems, work schedule, compliance and attitude —, your doctor should take a call on this.

How should I lower my HbA_1c?

The answer to this question is beyond the scope of this article. Also, that is equivalent of asking a medical advice for sugar control. A website is not a good source for such discussion.

This problem is aggravated by the fact that scientists are now finding that sugar control works very differently in different people. Some people get insulin surges for foods that others don’t, while those others get them for a completely different set of foods.

So diabetes and sugar management, per force, has become an individual–based approach. Your doctor is the best person to help you with that.

An article can suggest general guiding principles of care, but specifics must be worked out by you with your doctor.

And mind you, diabetes is one of the most complicated diseases to manage. That is because there is just too much variation in how it behaves from person to person.

So, don’t expect your doctor to resolve your problem at a snap of a finger. He may have to do many trials and errors to arrive at the best solution. Allow him that freedom and don’t look for the holy grail of diabetes treatment.

Glycated or glycosylated?

HbA_1c is interchangeably called glycated or glycosylated, hemoglobin. However, the latter use is wrong. These two are different processes.

For medical professionals

In glycation, glucose attaches to proteins or fats. But, it is an irreversible reaction that does not need enzymes as catalysts. It also depends on the blood concentration of glucose.

Glycosylation is again a process in which glucose attaches to proteins or fats. But, it needs enzymes as catalysts. Also, it involves something called folding of proteins.

When proteins are first formed in the body, they are just chains of amino acids. Actually, first, the chains of amino acids form polypeptides. And, the chains of polypeptides form proteins.

These newly formed molecules are not stable. Through the process called folding, the amino acids and the polypeptides in the molecule rearrange themselves into a proper 3–D shape, which is stable. Such protein molecule can now function properly.

The first process of protein synthesis in the body (bio–synthesis) is called translation. The folding of protein is called post–translational modification. It is a chemical change in the protein, and is catalysed by enzymes. Glycosylation is also a post–translational modification process and so needs enzymes.

In 1980s, it was not known whether the glucose attachment to proteins was enzymatic or non–enzymatic. However, in the recent years, it has been confirmed that it is irreversible and non–enzymatic. So, the process must be called glycation.

Medical literature still uses the term ‘glycosylated’. But, the correct use is glycated hemoglobin.

Read more: Glycated or glycosylated?

Limitations of HbA_1c test

Personally, I am a ‘first principles’ person. I like to evaluate things based on ‘back to basics’. So when one wants to talk about the limitations of the HbA_1c test, I prefer to look at the basic science behind the test. That can tell us a lot about the limitations.

Fluctuations in glucose levels

The average of five numbers — 4, 6, 6, 5, and 4 — is 5. So is the average of another five numbers — 9, 2, 1, 7, and 6. But, as you can notice, the second set has a wider range or fluctuations in values.

Similar situation can happen in HbA_1c. We may get HbA_1c of 6.3%, indicating an average glucose value of 134 mg/dL. But, it does not tell us whether the blood sugar was quite stable or it had wild swings in the values, over the previous few months. If blood sugar dips too low, there is a risk of hypoglycemia, or dangerously low levels of blood sugar. That can lead to a person going into coma.

Thus, the spot values (values at a particular time, such as fasting or post–meal) of blood glucose are also important to judge the sugar control.

Scatter

A paper published in Aug 2008 in the journal Diabetes Care showed that 90% of the times, measured HbA_1c values were accurate within 15% of the actual value. Table 2 below shows measured HbA_1c values against measured average blood glucose values. The 95% confidence interval in the 3–month average blood glucose values is shown in the brackets.

Note how wide the variations are in the average blood sugar. It means for a measured HbA_1c value, the real average blood sugar spanned a wide range. You can turn the argument around and say that for an average blood sugar value, we may get a wide range of measured HbA_1c values (with 15% or more variation).

Age of the red blood cells

A paper published in Oct 2016 in the journal Science Translational Medicine showed that error in estimated average blood glucose using HbA_1c was 15 mg/dL in one-third (33%) of the patients.

They found that this error cropped up because of different lifespans of the red blood cells in different people. When they adjusted for this difference, the error of 15 mg/dL was noticed in only 10% of the patients, down from 33%.

To me, though, 15 mg/dL, appears to be a large number. It is best to keep this possibility in mind when using HbA_1c numbers.

Gender effect

The average age of red blood cells is 117 days in men, and 106 days in women. The longer the cells survive, the higher is the glycated hemoglobin in them, boosting up the HbA_1c value.

While I did not find any research paper on this, I would guess that the men would have higher HbA_1c values than women, for the same blood sugar levels.

Average versus weighted–average effect

We said that the HbA_1c test measures the average levels over 2 to 3 months. However, consider a simple analogy.

Assume that the average red blood cells life is exactly 100 days. That is, all of them die on the completion of 100 days. So every day, 1% of the cells die, and another 1% new cells are formed. Thus:

1% of the cells are exposed to blood sugar for 100 days.
1% of the cells are exposed to blood sugar for 99 days.
1% of the cells are exposed to blood sugar for 98 days.
…
1% of the cells are exposed to blood sugar for 3 days.
1% of the cells are exposed to blood sugar for 2 days.
1% of the cells are exposed to blood sugar for 1 day.

In other words:

1% of the cells are exposed to blood sugar level present 100 days ago.
2% of the cells are exposed to blood sugar level present 99 days ago.
3% of the cells are exposed to blood sugar level present 98 days ago.
…
98% of the cells are exposed to blood sugar level present 3 days ago.
99% of the cells are exposed to blood sugar level present 2 days ago.
100% of the cells are exposed to blood sugar level present 1 day ago.

That means:

The blood sugar levels 100 days ago, contribute roughly 1/5050 = 0.02% to the final HbA_1c value.
The blood sugar levels 99 days ago, contribute roughly 2/5050 = 0.04% to the final HbA_1c value.
The blood sugar levels 98 days ago, contribute roughly 3/5050 = 0.06% to the final HbA_1c value.
…
The blood sugar levels 3 days ago, contribute roughly 98/5050 = 1.94% to the final HbA_1c value.
The blood sugar levels 2 days ago, contribute roughly 99/5050 = 1.96% to the final HbA_1c value.
The blood sugar levels 1 day ago, contribute roughly 100/5050 = 1.98% to the final HbA_1c value.

As you can see, in the measured HbA_1c value, yesterday’s blood sugar contributes 100 times more than the blood sugar 100 days ago. To visualise this further, let us group these into five periods of 20 days each and add up the contribution. We will find:

The blood sugar levels from recent 1 to 20 days, contribute roughly 1810/5050 = 36% to the final HbA_1c value.
The blood sugar levels from recent 21 to 40 days, contribute roughly 1410/5050 = 28% to the final HbA_1c value.
The blood sugar levels from recent 41 to 60 days, contribute roughly 1010/5050 = 20% to the final HbA_1c value.
The blood sugar levels from recent 61 to 80 days, contribute roughly 610/5050 = 12% to the final HbA_1c value.
The blood sugar levels from the final 81 to 100 days, contribute roughly 210/5050 = 4% to the final HbA_1c value.

This clearly shows that the most recent 20 days contribute nearly 9 times more than the last 20 days.

Since we are having fun with mathematics, if we extend our calculations further, we will find that:

The blood sugar levels from the most recent 78 days, contribute roughly 4797/5050 = 95% of the final HbA_1c value. Perhaps, that is why we say that HbA_1c reflects 2 to 3 months average of blood sugar, though the red blood cells live for 100 days (in this example).

The blood sugar levels from the most recent 30 days, contribute roughly 2565/5050 = 50.79%, or nearly half the final HbA_1c value.

If you consider the average life span of red blood cells to be 117 days in men and 106 days in women, 50% of the contribution to HbA_1c value should come from the latest 35 days in men, and 32 days in women.

An article published in July 2004 in the journal Diabetes Care claimed exactly the same: 50% of the effect on HbA_1c is by blood sugar values in the first 30 to 35 days.

Thus, while we call HbA_1c to be average of blood sugar levels over 2 to 3 months, it is actually a weighted–average with the most recent one month contributing more weightage, about half the value.

So my view is, for a fair evaluation, treat your HbA_1c as an indicator of sugar control over the previous month, and not more.

Stress effect

Stress causes reduction in the life of red blood cells. You can expect lower HbA_1c values in stressed individuals, for identical blood sugar levels.

Note that, in general, stressed individuals have poorer sugar control. So they may have higher blood sugars than their equivalent counterparts. However, I am referring to two individuals with identical blood sugars: one with stress, another without. The stressed individual will show lower HbA_1c value.

Iron deficiency effects

As per the World Health Organization (WHO), anemia is diagnosed when blood hemoglobin is less than 13 gm/dL in men, and less than 12 gm/dL in women. Most women in developing countries are anemic. They are likely to show higher HbA_1c value than the actual one.

In iron deficiency anemia, the blood level of a compound called malondialdehyde (MDA) increases. This plasma MDA is known to increase hemoglobin glycation. Thus, in iron deficiency anemia, you will see higher levels for HbA_1c for the same blood sugar levels.

In people with deficiencies of vitamin B–9 (folate) or vitamin B–12 (cobalamin), red blood cells cannot be formed properly. As a result, the person develops vitamin–deficiency anemia. In such people, many red blood cells are immature and die early. Only the good cells survive long. Hence, the average age of surviving red blood cells is higher, showing higher levels of HbA_1c.

In late pregnancy, large amount of iron is consumed, leading to some iron deficiency. The HbA_1c will be elevated in such women, even if they are non–diabetic.

In such patients, a test called glycated albumin is advised, as albumin levels are not affected in iron deficiency conditions.

Hemoglobin–variant effect

People of African, Mediterranean, or Southeast Asian origin, or having family members with sickle cell anemia or a thalassemia, may have a different type of hemoglobin, such as HbS, known as a hemoglobin variant. This may give wrong HbA_1c results, because their glycated hemoglobin will be HbS_1c, and not HbA_1c, for example.

Conditions that reduce red blood cells life or number

Medical conditions such as G6PD deficiency or sickle cell anemia cause early death of red blood cells, reducing HbA_1c numbers.

Blood donation effect

In a blood donation, 500 ml blood is collected. We have about 5 to 6 litres of blood. So, about 10% of the red blood cells are donated.

After blood donation, blood plasma is replaced within 24 hours. However, the red blood cells are replaced over 4 to 6 weeks. That is why you are advised not to donate blood within 8 weeks of a previous donation.

If you have donated blood recently, we are proud of you. 🙂 Besides, there will be more new red blood cells in your blood than there are normally. So, your HbA_1c will be lower than the correct representative value.

On the other hand, if you had donated blood 3 months ago, many new red blood cells would have come in your blood 2 to 3 months ago. So, a whole lot of cells today would be 2 to 3 months old. Your HbA_1c, therefore, would be higher than actual.

Unpredictable results

In some conditions such as blood loss, blood transfusion, surgery (blood loss, again), chronic kidney disease, dialysis, or chronic liver disease, the HbA_1c results can be unreliable.

Some papers claim high dose vitamin C (1 gm/day) lowers HbA_1c readings by reducing the glycation of hemoglobin, while some other papers say they don’t.

Even if you measure HbA_1c in the same laboratory, temperature and sample handling make a difference in HbA_1c values measured.

Read here: Could your HbA_1c test be wrong?

For those of you who are medically inclined, here is the WHO’s annexure to its report on HbA_1c: Some of the factors that influence HbA_1c and its measurement.

Alternative glycation tests

When glycated hemoglobin test cannot be used reliably, fructosamine in the blood is measured. Fructosamine is the glycated part of all plasma proteins. That test is less accurate than glycated hemoglobin. It generally reflects average blood sugar over the previous two weeks.

Another test used is glycated albumin, because albumin does not get affected by iron deficiency nor by red blood cell life.

Read a wonderful and detailed guidelines article on the website of the National Institutes of Health, USA: The HbA_1c test and diabetes.

HbA_1c test as a diagnostic test for diabetes

It is quite possible that in some people, a fasting blood glucose test may show diabetes when an HbA_1c test does not. This could be because the person consumed high levels of carbohydrates the day before the test, for example.

On the other hand, in some patients with early stages of diabetes, the blood sugar levels may not have risen high enough to show up every time a fasting glucose test is performed. In such patients, an HbA_1c test may diagnose diabetes, while the fasting blood sugar would not.

Doctors generally perform two different tests on two separate days to confirm the diagnosis.

Recent doubts about diagnostic capabilities of HbA_1c

Intra–day fluctuations in blood glucose don’t affect HbA_1c values. Hence, there is not need for the person to undertake a fast before giving a blood sample. Hence, HbA_1c is often preferred as a diagnostic test.

However, a study presented in March at ENDO 2019, the annual meeting of the Endocrine Society, USA, found that the HbA_1c test does not catch many diabetics among the untested population.

The researchers studied 9,000 adults who had not been diagnosed diabetics in the past. Both, HbA_1c test and oral glucose tolerance test, were given to them. Oral glucose tolerance test is almost a Gold standard for diagnosing diabetes, but it is not as simple nor convenient as the HbA_1c test.

The study found that 73% of the people who were diagnosed diabetics by oral glucose tolerance test, were labelled normal as per the HbA_1c test results. This false–negative number is too high. That means, the sensitivity of this test — its ability to catch diabetes — is quite low, only about 27%.

If the research is correct, HbA_1c cannot be used as a diagnostic test for diabetes. The study authors recommend that HbA_1c alone should not be used as a test for diagnosing diabetes.

My view

HbA_1c test has many problems in giving you accurate results. But, that is a part of most medical tests. There is almost no good test that will tell you about long–term sugar control, except continuous glucose monitoring, which is cumbersome and costly.

Look at the simplicity of the HbA_1c test. It does not even need fasting. Short–term dietary variations don’t affect it.

I would look at the trend in HbA_1c values for a person. For the reasons mentioned above, HbA_1c results may not be worthwhile to compare from person to person. It may also not be right to pay too much attention to variations between two consecutive HbA_1c test results for the same person.

However, for the same person, many parameters such as average red blood cell life, ethnicity, stress levels, and nutrition will be similar over multiple HbA_1c tests. So if there are errors, they will be consistent in all tests done at various times. So, the test should be more useful to know the trend in assessment of sugar control, than the absolute value of HbA_1c measured.

In conclusion

Prolonged high blood sugar levels lead to diabetic complications. Glycated hemoglobin, or HbA_1c, is a good test that measures long–term blood glucose control.

Blood glucose attaches to hemoglobin in the red blood cells slowly over a period of time. This irreversible process, Glycation, leads to glycated hemoglobin, or HbA_1c. The more the blood sugar, the higher the formation of HbA_1c.

Since red blood cells live about 110 days, measuring the percentage of hemoglobin that is glycated, roughly tells us about average blood sugar levels over the past 2 to 3 months.

American Diabetes Association (ADA) considers HbA_1c below 5.7% as normal, 5.7% to 6.4% as pre–diabetic, and 6.5% and above as diabetic. The WHO considers HbA_1c below 6% as normal, instead of 5.7%.

Lowering HbA_1c reduces diabetes–related damage in kidneys, eyes, and nerves by 25% for every 1% reduction in HbA_1c. As a result, ADA advises bringing HbA_1c below 7.0% for a good sugar control. The International Diabetes Federation, on the other hand, advises 6.5% as the target level for HbA_1c. The article discusses about whom to follow.

There are various situations that give wrong HbA_1c readings. For example, different anemias, nutrient deficiencies, G6PD deficiency, stress, ethnicity, gender, surgery, and even the blood donation status affect HbA_1c levels. Finally, the classical interpretation of HbA_1c as a 2 to 3 month average is incorrect.

When glycated hemoglobin cannot be used as a test, tests for blood fructosamine or glycated albumin are used.

Continuous blood glucose monitoring remains the Gold standard. Plus, there are many new technologies coming up that will make long–term blood sugar monitoring easy. All of this is discussed in the full article.

First published on 19th July, 2019

Image credit: Steve Buissinne from Pixabay