Hba1c: when the diagnosis of diabetes can be wrong

Words below taken from The Diabetes Diet Blog http://diabetesdietblog.com/

see here http://diabetesdietblog.com/2015/10/17/hba1c-when-the-diagnosis-of-diabetes-can-be-wrong/

Many doctors are now using the hba1c on its own  to diagnose type two diabetes. It means that there is no need to fast overnight, and that you don’t have to spend hours in the GP surgery. But the test relies on having an average turnover time of your red blood cells and this can lead to an incorrect diagnosis if certain conditions are not taken into account.

For some people they DO have diabetes but the blood sugar changes have been so rapid that the red cells have not had time to accumulate enough sugar on them. Therefore children and young people,  women who are pregnant, or who has been pregnant in the last two months,  and anyone in whom type one diabetes is suspected, who tend to have a short duration of symptoms, should still get glucose measurements to determine the diagnosis. Testing for blood or urinary ketones would also be good practice in this group of people.

In people who are anaemic or whose haemoglobin is fragile and gets destroyed earlier than the usual 120 days, blood sugars will be higher than they look for any given hba1c value. Therefore if the hba1c is relied on a diagnosis of diabetes could be missed. Those with haemoglobinopathies, renal failure or HIV infection will be in this group.

If someone is acutely ill, their blood cortisol rises, and effectively blocks the action of a person’s own insulin. Thus the blood sugar rises. This can make it look like someone has developed diabetes but blood sugars will settle back to normal once the person is over the illness.

Certain drugs such as corticosteroids and anti-psychotics also rapidly raise the blood sugar. If a person is acutely ill it is best to rely on blood glucose measurements but after two months of continued drug use the hba1c can be used as long as the person is not acutely unwell.

People with pancreatic damage or who have just had pancreatic surgery also may have a deficient insulin response. This may or may not recover sufficiently over time. Blood glucose measurements are again more reliable in this group.

One of the major reasons that hba1c testing was introduced was to facilitate the diagnosis of type two diabetes. Currently about one in 4 type two diabetics in the UK is thought to be undiagnosed. It is certainly easier to do a hba1c than a glucose tolerance test, but the oral glucose tolerance test still has its place.

Although I’m a doctor I wasn’t fully aware of all the types of people and situations in which the hba1c could be misleading.

Based on an article by Professor Andrew Farmer of Oxford BMJ 10th November 2012.

Good post Jan,something all diabetics should be aware of and factor in if they suffer with other health conditions as well...My A1c always comes out lower than my meter averages suggest so a higher than normal A1c result is always a massive cause of concern to me.

I agree. The criteria for diagnosing should change so that we catch it before complications set in. I had an A1c of 6.5 and a fasting glucose of 128 mg/dl when diagnosed with multiple complcations. That's too late!

I now think that insulin resistance should be one of the major criteria since it is abnormal levels of insulin that appear to start the whole process towards type 2. Higher levels of insulin itself also cause damage to our vascular system (ex: hypertension and heart disease) before diabetes appears.

I think A1c was adopted as the ONLY diagnostic criterion precisely because it catches FEWER diabetics, later in the progression, and thus improves the statistics. Many US doctors were instructed to use ONLY fasting BG for the same reason - the majority of Type 2 shows in postprandial numbers long before fasting is affected, and likewise you can have substantial "diabetic" complications before A1c reaches 6.5%.

People like me with significant postprandial spikes followed by post-postprandial lows can have effectively normal A1c because the one cancels out t'other. Over time I've seen a significant number of high or low glycators where A1c does not bear out spot readings. Fructosamine (as far as I know obly used by vets in the UK) often bears out the meter readings suggesting the A1c is at fault.

There's also some evidence that nondiabetic low carbers may have a higher than average A1c, maybe because the blood cells live significantly longer than "normal" and pick up more glycation from exposure to essentially normal BG over a longer time.

I doubt this will persuade Farmer to renege on his belief that testing is ineffective though.

chris c wrote:I think A1c was adopted as the ONLY diagnostic criterion precisely because it catches FEWER diabetics, later in the progression, and thus improves the statistics. Many US doctors were instructed to use ONLY fasting BG for the same reason - the majority of Type 2 shows in postprandial numbers long before fasting is affected, and likewise you can have substantial "diabetic" complications before A1c reaches 6.5%.

People like me with significant postprandial spikes followed by post-postprandial lows can have effectively normal A1c because the one cancels out t'other. Over time I've seen a significant number of high or low glycators where A1c does not bear out spot readings. Fructosamine (as far as I know obly used by vets in the UK) often bears out the meter readings suggesting the A1c is at fault.

There's also some evidence that nondiabetic low carbers may have a higher than average A1c, maybe because the blood cells live significantly longer than "normal" and pick up more glycation from exposure to essentially normal BG over a longer time.

I doubt this will persuade Farmer to renege on his belief that testing is ineffective though.

I do know that I had complications at diagnosis with an A1c of 6.5 and fasting at 128 mg/dl. Also, Peter at Hyperlipid reports A1c's in the low 4s yet has higher fasting, so there may be a considerable variation in response to low carb especially at keto levels.

OldTech wrote:
I do know that I had complications at diagnosis with an A1c of 6.5 and fasting at 128 mg/dl. Also, Peter at Hyperlipid reports A1c's in the low 4s yet has higher fasting, so there may be a considerable variation in response to low carb especially at keto levels.

Yes I was getting postprandial neuropathy to the extent of tingling fingers and feet "going to sleep", soon followed by the rest of me, with an A1c of "only" 5.3%. I "only" scored 10.7 on the GTT after 1 hour and was back to 7.8 at 2 hours. This was obviously offset by the lows at 3 - 4 hours after eating. Also I had major "diabetic" dyslipidemia.

I soon had my postprandials under 7 which deleted the symptoms, but my A1c actually went UP to 5.8%. Further control has me generally around 5.6%. I also discovered that if my postprandial BG went over 8 I'd need to get up to piss in the night: over 10 and I'd be up and down every couple of hours, under 6 and I could sleep through the night again.

Likewise my mother had kidney failure and what I'm convinced was the same postprandial neuropathy while her A1c was "only" 6.4%

It's easy to blame the GPs but really it's the people who feed them duff information that are in serious need of stringing up.

EPIC-Norfolk and a huge New Zealand study relate A1c to CVD risk. Other work relates A1c to microvascular complications and postprandial spikes to macrovascular complications. Most GPs and especially nurses have no clue that any of this research exists.

chris c wrote:

OldTech wrote:
I do know that I had complications at diagnosis with an A1c of 6.5 and fasting at 128 mg/dl. Also, Peter at Hyperlipid reports A1c's in the low 4s yet has higher fasting, so there may be a considerable variation in response to low carb especially at keto levels.

Yes I was getting postprandial neuropathy to the extent of tingling fingers and feet "going to sleep", soon followed by the rest of me, with an A1c of "only" 5.3%. I "only" scored 10.7 on the GTT after 1 hour and was back to 7.8 at 2 hours. This was obviously offset by the lows at 3 - 4 hours after eating. Also I had major "diabetic" dyslipidemia.

I soon had my postprandials under 7 which deleted the symptoms, but my A1c actually went UP to 5.8%. Further control has me generally around 5.6%. I also discovered that if my postprandial BG went over 8 I'd need to get up to piss in the night: over 10 and I'd be up and down every couple of hours, under 6 and I could sleep through the night again.

Likewise my mother had kidney failure and what I'm convinced was the same postprandial neuropathy while her A1c was "only" 6.4%

It's easy to blame the GPs but really it's the people who feed them duff information that are in serious need of stringing up.

EPIC-Norfolk and a huge New Zealand study relate A1c to CVD risk. Another work relates A1c to microvascular complications and postprandial spikes to macrovascular complications. Most GPs and especially nurses have no clue that any of this research exists.

Thanks. What I have just realized recently is that it is not just glucose and A1c that does the damage. It is also insulin. Insulin is a cause of vascular disease, and its effects are felt long (even decades) before elevated glucose. See: http://www.hindawi.com/journals/ijhy/2013/230868/. High levels of injected insulin also explain the ACCORD study.  

Taubes' carbohydrate-insulin theory for metabolic disease along with understanding the damage that high levels of insulin causes go a long way in explaining why we do not have a cure for diabetes. The only 'cure' is to cut carbs! It is not a disease - it is just the body's response to abusing carbs! It's the same story for alcoholics.

OldTech wrote:
Thanks. What I have just realized recently is that it is not just glucose and A1c that does the damage. It is also insulin. Insulin is a cause of vascular disease, and its effects are felt long (even decades) before elevated glucose. See: http://www.hindawi.com/journals/ijhy/2013/230868/. High levels of injected insulin also explain the ACCORD study.

Yes and that goes back to Joseph Kraft, yet another name current doctors will never have heard of.

An old Type 1 friend used to quote a cardiologist as calling Type 2 diabetes "a cardiovascular disease sometimes associated with high blood glucose", which is pretty much spot on what modern research is finding and pretty much what Kraft had already discovered.

Nice paper! I hadn't seen that one before and will read it later.

And indeed credit should be given to Dr. Kraft and Dr. Bernstien. I have even read Kraft's book but apparently not everything sunk in at the time. Sadly it seems that I have to have a working hypothesis to fit the puzzle pieces together. At the time, I knew that high insulin levels occur when with eating excess carbs, but I did not know insulin itself causes vascular disease. Once I saw that I then knew where the insulin piece fit into the puzzle.

Yes it's yet another thing that doctors should know but aren't told. The other side of the coin of course is what they ARE told but shouldn't know, like fat causes diabetes and CVD.

I see the endocrine system as an analog computer, which is basically the body's operating system, and is governed by an intricate mechanism of levels of different hormones tied together by feedback and feedforward loops. Insulin, leptin and T3 are three of the "master hormones" and insulin is the easiest to manipulate through diet. Get that right and the leptin also readjusts and a whole bunch of other stuff comes into line. Get it wrong and you face the body's equivalent of a blue screen.

Chris, are you a system engineer or programmer? I too think that of our body as an analog computer, but from our perspective it is just a white box that we can not see inside. We only get imperfect hints at how it works. That's the reason that Dr. Bernstein's small steps mean small errors is a pragmatic solution.

So, what happens next? If the HbA1c is "normal" low carbing is adhered to but there is still inconsistent wildly fluctuating blood pressure.
Every time I think I am beginning to understand what is going on, another spectre looms. It is like trying to extinguish spot fires with a teacup full of water.

OldTech wrote:Chris, are you a system engineer or programmer? I too think that of our body as an analog computer, but from our perspective it is just a white box that we can not see inside. We only get imperfect hints at how it works. That's the reason that Dr. Bernstein's small steps mean small errors is a pragmatic solution.

I used to work in mechanical engineering but I've always had a mind that tends to find patterns in data.

merrylizard1314 wrote:So, what happens next? If the HbA1c is "normal" low carbing is adhered to but there is still inconsistent wildly fluctuating  blood pressure.
Every time I think I am beginning to understand what is going on, another spectre looms. It is like trying to extinguish spot fires with a teacup full of water.

Any chance you're on a calcium channel blocker?

My mother was on several BP meds towards the end. At one stage the hospital removed everything but a CCB and her BP started jumping up and down all over the place. The GP took her off it and replaced it with a low dose ARB which stabilised things. Anecdotally I've read similar several times and I'm pretty sure I dug up some research at the time that agreed. Especially for diabetics, ACE or ARB are superior - but CCBs are cheaper.

Ace inhibitor and beta blocker, at present, being titrated until BP comes under control. Calcium channel blockers give me bad side effects. I feel very well, and a cardiologist cannot find anything to worry about, endothelially speaking, but I do feel a bit worried at 207/100!
Had an appointment with a dietitian today, who wanted to know what my diet was, before low carb. I just pointed to the Healthy Eating for Adults chart on the wall in her office.

Merrylizard,
207/100 is indeed high, but are you measuring this at home, or just being measured at stressful visits to the doctor?  My husband's BP has been known to go up to these levels at moments of stress (which includes some medical surroundings), but is generally much, much lower at home.  He bought a home monitor a couple of years ago, of the type which goes round the upper arm, not the notoriously inaccurate wrist type.  This has been very useful, not only in bringing BP down in the first place, but in dropping some drugs as his health improved.
Low carb has not produced "perfect" blood pressure, as may be suggested in some places, but has almost certainly contributed to gradual improvements.

One thing I have noticed from his daily monitoring is that BP varies all the time, from minute to minute and day to day.  I see it as the body accommodating varying situations.  A higher than usual BP is often the first sign of any illness.  Raised BP at 7.00am and the first signs of a cold early afternoon.

I've also noticed that doctors/consultants, who are not being paid according to the number of patients with BP below a certain level, tend to take a far more relaxed attitude towards what is reasonable.  The old idea of 100 plus your age as your systolic measurement is not rejected by all, though I'm guessing that your 107th birthday is still a year or two off?
Sally

Sally
Using a home monitor, and recording in a blood monitor app, which has graphs and charts for results.
I do not have high results every time, but swooping from 110/72 - 200/95 in the course of a day, does concern me. Also, my pulse rate is slow, even with the very high numbers, often in the low 40s. Not sure what this means, but on the 3occasions in the past 2years when I have had AF, the diastolic reading has been high. It is annoying, particularly when I have managed to get the HbA1c in good order.
More research required.

Sometimes I think that "medicine by numbers" has got a bit above itself and that the important questions are, "do you feel well?", "do you look well?", "are you able to do the things you want to do and can reasonably expect to do at your age?" Too many doctors have forgotten this.
As you say, more research required.
Sally

merrylizard1314 wrote:Sally
Using a home monitor, and recording in a blood monitor app, which has graphs and charts for results.
I do not have high results every time, but swooping from 110/72 - 200/95 in the course of a day, does concern me. Also, my pulse rate is slow, even with the very high numbers, often in the low 40s. Not sure what this means, but on the 3occasions in the past 2years when I have had AF, the diastolic reading has been high. It is annoying, particularly when I have managed to get the HbA1c in good order.
More research required.

Possibly autonomic neuropathy. At one stage my pulse was locked to around 60 bpm whatever I did, my heart would beat harder but not faster. "Are you on a beta blocker? - oh no!" said my BG helpfully.

This actually resolved after BG was controlled for long enough, but it took at lot longer than the peripheral stuff to go away.

If I was your GP I'd also be looking at the rest of your endocrine system, my BP went through the roof and my HR went up with hypERthyroid. Something in your system appears to be cycling, pituitary problems can do this and be very hard to diagnose as you actually have to catch the thing overproducing which it may only be doing occasionally. NOT a diagnosis, just some places you may care to look based on what has happened to me and people I know.