Can hba1c be too low?

pavlosn

Well-Known Member
Messages
2,705
Type of diabetes
Type 2
Treatment type
Tablets (oral)
As a t2 non insulin dependent diabetic, I have been controlling my levels with a target of keeping my hba1c as low as possible, as I understood that this would improve my risk of avoiding the most common diabetic complications.

A non diabetic Hba1c was my initial target for reasons that appeared self evident to me. If I can keep my levels the same as a non diabetic then hopefully I am also keeping the risks I am facing down to a similar level as well.

But the non diabetic hba1c range stretches from 4% to 6% so it is quite a wide range. Where within this should i/we actually be aiming for?

I am currently waiting for my latest score but my last hba1c was a non diabetic 5,2% or 33mmol/mol, which led me to question where I go from here.

Do I continue to aim lower hoping perhaps to break into the 4% club with a sub 5% hba1c?

Is there any evidence that such a score is better than my current score or even a slightly higher score?

Conversely, is there evidence that such a low score my actually be an indicator of increased rather than decreased risk of health complications.

My concerns were compounded by reading stories of other non insulin dependent t2s that on recent visits to their doctors were advised to raise their hba1c scores from under 6% (42.1 mmol/mol) to the 6-6,5% (42,1-47,5) pre diabetic range. Advise that did not make sense to me but which I did not immediately dismiss because I am not a doctor and they are, so they must be basing it on some kind of evidence and research.

Which led me to start looking online for relevant literature.

The first question I tried to find the answer to was where within the normal non diabetic range did most non diabetics lie.

I could not find a direct answer to my question but from the sample used in this relevant study

http://www.nejm.org/doi/full/10.1056/NEJMoa0908359#discussion

I got this diagram suggesting that the vast majority of non diabetics have an hba1c higher than 5% (31,1 mmol/mol) with most being in the 5-5,5% band. ( 31.1 - 36.6 mmol/mol)


ImageUploadedByDCUK Forum1418642958.574287.jpg


But this study was based on a sample of about 11000 participants who were mainly in their fifties.


By looking at the sample characteristics of this study of 14000 non diabetics

http://m.circoutcomes.ahajournals.org/content/3/6/661.full

I got a similar looking graph

ImageUploadedByDCUK Forum1418642908.075886.jpg


The split seemed to be 75:25 in favor of above 5%( 31,1 mmol/mol) Hba1c amongst the participating diabetics, with the 5,0-5,5 ( 31.1-36.6 mmol/mol) band again being the most populous.

So it seems that if the above is correct the average non diabetic has an hba1c of 5-5,5. ( 31.1-36.6 mmol/mol).

But does most common equate to best. What if a smaller number of diabetics had a sub 5% (31.1 mmol) hba1c but these were the non diabetics that faced the least risk.

Two diagrams extracted from these two studies gave similar answers to the above question.

ImageUploadedByDCUK Forum1418643418.546648.jpg


ImageUploadedByDCUK Forum1418643463.303482.jpg


Both suggest a j shape relationship between all cause mortality and Hba1c I.e the risk of death from any cause decreases with decreasing hba1c as one reduces from the diabetic to the prediabetic to the non diabetic range. But at some point around the 5% 31.1 mmol/mol) mark reducing further actually seems to increase all purpose mortality risk.

Why this should be so I can not answer but this is what I understand from the above studies/graphs.

And both of these are dealing with non diabetics who can achieve these scores without resorting to any intervention whether dietary or medicinal.

I also found this online abstract

http://m.care.diabetesjournals.org/content/early/2014/11/13/dc14-1787.abstract

From a recent German study which concluded that "Excess mortality risk is associated with both known diabetes and undiagnosed diabetes but not for prediabetes, with lowest risk at HbA1c levels 5.4–5.6% (35.5-37.7 mmol/mol) (and a significantly increased risk at ≤5.0% (31.1 mmol) and ≥6.4% ( 46.4 mmol)"

All food for thought and all challenging my basic assumption that lower is necessarily better when it comes to Hba1c

Pavlos
 
Last edited by a moderator:

Bluetit1802

Legend
Messages
25,216
Type of diabetes
Type 2 (in remission!)
Treatment type
Diet only
A very well thought out post, Pavlos, putting it all together in one post and very interesting.

My very personal opinion is that we all have our own normal natural level that we would see were we not diabetics. Some will be 4's, some 5's. I know from looking back at my fasting glucose results from before I was diabetic that my own personal level is in the 5's as far as fasting goes. They are all 5's. No 4's. (my husband's are all 4's) This is born out now as if I go to bed on 4's my liver always dumps me back to the 5's. If I go to bed on 5's my liver keeps me in the 5's.

I believe we should aim for a level we know we can sustain for life without any real deprivation or stress. Once that level is reached, hey ho. We should be happy, whether it be 4's or 5's.
 
  • Like
Reactions: 11 people

linda321

Well-Known Member
Messages
118
Type of diabetes
Type 1
Treatment type
Insulin
Interesting thread. I will be keeping an eye on what others add to the debate!
 
  • Like
Reactions: 4 people

sanguine

Well-Known Member
Messages
3,340
Type of diabetes
Type 2
Treatment type
Diet only
Dislikes
Intolerance, career politicians, reality TV and so-called celebrity culture, mobile phones in the quiet carriage.
Thanks Pavlos (but could you put the equivalent mmol/mol HbA1c values as well please? :))

5.0% is 31.1 mmol/mol, and apparently equivalent to 5.4 mmol/l although I take that latter conversion with a large pinch of salt.

Similarly, 6.4% is 46.4, 7.6 mmol/l equivalent, same pinch of salt.

If I can get to 5.3% (34.4) and keep it there I will be a happy bunny, although I don't have the knowledge to explain why I don't currently have an interest in taking it lower. Maybe when I get there I'll know more or have to research it further!
 
  • Like
Reactions: 6 people

Lamont D

Oracle
Messages
15,793
Type of diabetes
Reactive hypoglycemia
Treatment type
I do not have diabetes
@pavlosn, what a great and thought out post, I had to read it twice to take it in, I'm going for a third one after replying!
As a non diabetic and I do fast time to time, my body doesn't allow me to go below 4.1. I don't know the science or even bothered to find out!
My 'normal' level is between 4.3 & 5.5.
I have to maintain the range under 6 to stop me spiking then going hypo.
My consultant even on my 72 hour fasting test was pushing me to hypo whilst fasting but my body said no!
I believe that there is no definitive advice than to keep your bloods around the 4-6 range to. Feel better and be healthier. You shouldn't really be obsessive about getting down too low. I'm sure you by now know where you want to be!
If you want to experiment that's fine!
I do most of the time, seeing what certain combinations of food does to me, how much of a certain food can I tolerate. We are all different, unique!
There is a certain limit to generalise any tests with diabetes. It all depends on the ratio of insulin/ glucagon to diet!
There's no way I would tackle that!
Every study I've seen cannot be comprehensive in its findings.
Thanks again for the post, it is really good reading!
 
  • Like
Reactions: 2 people

Lamont D

Oracle
Messages
15,793
Type of diabetes
Reactive hypoglycemia
Treatment type
I do not have diabetes
Having read it again the repeating question is will I get complications if I stay in the diabetic range and if I get below that I will get less complications if I get as low as possible?
Now as you follow the footie thread my predictions are terrible! However I'm sure that the chance of complications lowers the lower you can keep them. There is no magic wand. You just have to trust yourself to do what's right for you!
A low carb diet has been shown to help diabetics of all types to control their BSLs. I have to stay in that range for my health, I can't fall off the wagon!
What would you do in my place?
 
  • Like
Reactions: 2 people
Messages
6,107
Type of diabetes
Type 2
Treatment type
Diet only
I watched a recorded lecture few weeks ago, it was one of those from the Libra presentation. The lecturer made a valid point about T1's and to paraphrase what he said .....Hba1c is an average and the lower it is the more time the patient has spent being hypo. He suggested a slightly raised Hba1c as a target would improve matters. I presume this applies to insulin dependent T2's as well..

As a T2 most of my time is spent finding ways to keep my Hba1c down and I am unlikely to go hypo much or at all. Chances are that my liver will play a part if I went too far but I have never put this to the test.
 
  • Like
Reactions: 8 people

pavlosn

Well-Known Member
Messages
2,705
Type of diabetes
Type 2
Treatment type
Tablets (oral)
Thanks Pavlos (but could you put the equivalent mmol/mol HbA1c values as well please? :))

5.0% is 31.1 mmol/mol, and apparently equivalent to 5.4 mmol/l although I take that latter conversion with a large pinch of salt.

Similarly, 6.4% is 46.4, 7.6 mmol/l equivalent, same pinch of salt.

If I can get to 5.3% (34.4) and keep it there I will be a happy bunny, although I don't have the knowledge to explain why I don't currently have an interest in taking it lower. Maybe when I get there I'll know more or have to research it further!
Original post now edited to include hba1c in mmol/mol as well
 

cold ethyl

Well-Known Member
Messages
3,210
Type of diabetes
Type 2
Treatment type
Diet only
Great post. I have my next HbA1c in January and hope to have shifted downwards significantly on the last one I had at 7 weeks post diagnosis. I think I'd like it to be about 5.5 then I'd be plumb in middle of that 5.4-5.6 range. I don't think even pre decline into diabetes I was happy running at low BS levels as they precipitate my panic attacks - I can feel a sense of unease and adrenalin around 4.5 BG so it's a balancing act between mental health and diabetes.
 
  • Like
Reactions: 9 people

pavlosn

Well-Known Member
Messages
2,705
Type of diabetes
Type 2
Treatment type
Tablets (oral)
I watched a recorded lecture few weeks ago, it was one of those from the Libra presentation. The lecturer made a valid point about T1's and to paraphrase what he said .....Hba1c is an average and the lower it is the more time the patient has spent being hypo. He suggested a slightly raised Hba1c as a target would improve matters. I presume this applies to insulin dependent T2's as well..

As a T2 most of my time is spent finding ways to keep my Hba1c down and I am unlikely to go hypo much or at all. Chances are that my liver will play a part if I went too far but I have never put this to the test.

I tried to take the risk of hypos out of the equation. Like you I am a t2, who is not on medication that are likely to lead to hypos, so they are not a major concern to me.

The above studies are actually dealing with non diabetics so risk of hypos was never an issue, yet they still indicated increased all purpose mortality below a certain hba1c level.

I appreciate that a lot of us are unlikely to reach so low that this becomes a practical issue but it is still something to bear in mind.

Pavlos
 
  • Like
Reactions: 2 people

pavlosn

Well-Known Member
Messages
2,705
Type of diabetes
Type 2
Treatment type
Tablets (oral)
Update.

Just had my latest hba1c result and it is up from 33 to 38; probably as a consequence of reducing my daily Metformin dosage from the prescribed 1650mg daily to 1000 daily.

My latest hba1c is still in the non diabetic range but it is a bit disappointing to see it go up by so much. At least I no longer have to worry about going to low.

At least it will keep me motivated over the Christmas period.

I would like to bring this down a bit by the time I go to see my doctor in mid January.

None of which should detract from the discussion of my OP, which deal with issues beyond my own personal hba1c score.
 

Bluetit1802

Legend
Messages
25,216
Type of diabetes
Type 2 (in remission!)
Treatment type
Diet only
@pavlosn do I recall correctly that you had a period of being "off the waggon" for a while when you disappeared from the forum? Could this account for the increase? In any event, I would be dancing naked in the snow if I got a 38. :) Please, if ever I do get to 38, don't hold me to it!
 
  • Like
Reactions: 9 people

Lamont D

Oracle
Messages
15,793
Type of diabetes
Reactive hypoglycemia
Treatment type
I do not have diabetes
Update.

Just had my latest hba1c result and it is up from 33 to 38; probably as a consequence of reducing my daily Metformin dosage from the prescribed 1650mg daily to 1000 daily.

My latest hba1c is still in the non diabetic range but it is a bit disappointing to see it go up by so much. At least I no longer have to worry about going to low.

At least it will keep me motivated over the Christmas period.

I would like to bring this down a bit by the time I go to see my doctor in mid January.

None of which should detract from the discussion of my OP, which deal with issues beyond my own personal hba1c score.
You are still in the non diabetic range so now you have a target to achieve. You are doing well. Keep it up!
 
  • Like
Reactions: 4 people

tim2000s

Expert
Retired Moderator
Messages
8,934
Type of diabetes
Type 1
Treatment type
Other
I watched a recorded lecture few weeks ago, it was one of those from the Libra presentation. The lecturer made a valid point about T1's and to paraphrase what he said .....Hba1c is an average and the lower it is the more time the patient has spent being hypo. He suggested a slightly raised Hba1c as a target would improve matters.

This is, to be frank, a pile of absolute BS. If you are able to use a CGM/libre like device and manage glucose levels accordingly, there is no reason at all why you are definitely spending more time in a Hypo. It simply isn't a valid point for all T1s. What the likes of CGMs (and pumps) enable is that you can reduce the standard deviation of your blood testing and glucose levels (which works both up and down), and doing this reduces your Hba1C.

If you take that away then god-only knows what other consultants are believing. You can be a well controlled, limited hypo T1 with an Hba1C in the 5.7%-6.2% range without too much effort. Damnable HCPs who don't move with technology. :mad::mad::mad:
 
  • Like
Reactions: 2 people

pavlosn

Well-Known Member
Messages
2,705
Type of diabetes
Type 2
Treatment type
Tablets (oral)
@pavlosn do I recall correctly that you had a period of being "off the waggon" for a while when you disappeared from the forum? Could this account for the increase? In any event, I would be dancing naked in the snow if I got a 38. :) Please, if ever I do get to 38, don't hold me to it!
You remember correctly but that does not account for the increase.

I returned to the forum on 16/9/2014 and since then I have had three monthly (self funded) hba1c tests:
- October: 6% (42 mmol)
- November: 5,2% (33 mmol)
- December : 5,7% (38 mmol)

I am happy with all and any of the above counts by the way.

This thread is not motivated by any self serving interest in trying to justify that my latest increase in hba1c is a good thing.

I am still aiming for somewhere around 35 mmol/mol next time round.

As you know, my interest on this subject of whether there is such a thing as an optimal non diabetic hba1c score we should all be aiming for pre dates this post.

My aim is to initiate an honest exchange of views with other members of the forum, without risking that such a discussion would divert someone else's thread, as was the case whenever something similar came up in the past.

I honestly am interested in other people's views and I am happy to be persuaded by any contrasting evidence that may be presented.

I am not trying to convert anybody to my way of thinking as to be honest I do not think I have arrived at a final conclusion on the subject and, at any rate, I am not really into converting others. I am more interest at arriving to answers for my own questions.

All I attempted to do in my OP was to present my questions and the answers I was able to find to the extent I could understand the rather technical sources with my very limited admittedly expertise and knowhow.

Pavlos
 
Last edited by a moderator:
  • Like
Reactions: 13 people

NoCrbs4Me

Well-Known Member
Messages
3,700
Type of diabetes
I reversed my Type 2
Treatment type
Other
Dislikes
Vegetables
Having just had an HbA1c test myself that was not reflective of my daily self monitoring numbers, I did some investigating.
My last result was 42 mmol/mol. According to the converter tool (
http://www.diabetes.co.uk/hba1c-to-blood-sugar-level-converter.html) that's an average of 7 mmol/L. However, I've only had 2 postmeal levels above 7 in the last three months. Based on my meter readings my average should be around 5.5 mmol/L, which should result in an HbA1c of 5.1 mmol/mol.

One thing we all like to do on this forum is post our numbers without consideration of the accuracy of them, with a few acceptions.

So, with respect to how well HbA1c reflects blood glucose levels I found this (http://labtestsonline.org/understanding/analytes/a1c/tab/faq/):


"4. Why are my A1c and blood glucose different?
Beyond the difference in units used to report them, the A1c represents an average over time while your blood glucosereflects what is happening in your body now. Your blood glucose will capture the changes in your blood sugar that occur on a daily basis, the highs and the lows. Each blood glucose is a snapshot and each is different. The A1c is an indication that "in general" your glucose has been elevated over the last few months or "in general" it has been normal. It is inherently not a sensitive as a blood glucose. However, if your day-to-day glucose control is stable (good or bad), then both the A1c and blood glucose should reflect this. It is important to remember the time lag associated with the A1c. Good glucose control for the past 2-3 weeks will not significantly affect the A1c result for several more weeks.

In addition to this, it is also important to remember that glycated hemoglobin and blood glucose are two different but related things. For unknown reasons, some peoples' A1c may not accurately reflect their average blood glucose."

I looked for graphs of HbA1c versus average blood glucose and all I found was this (http://care.diabetesjournals.org/content/25/2/275.long):

HbA1c_EMBG.jpg


So the conversion from HbA1c to average blood glucose levels should be taken with a grain of salt...or sugar.

I also found this discussion: "Why hemoglobin A1c is not a reliable marker": http://chriskresser.com/why-hemoglobin-a1c-is-not-a-reliable-marker

And of course the HbA1c only reflects an average number and does not tell you how much time you've spent at elevated levels.

In addition, the accuracy of the HbA1c test is +/- 6%.

One thing I've learned on this forum and researching type 2 diabetes is that we are all different. Almost all of these studies are based on large numbers of people, which means that it's not always applicable to look at our own numbers and predict what it means for us. So just because your HbA1c is a certain number, that doesn't necessarily translate into the average glucose level as someone else with the same HbA1c number.

So, to address your original question, I think the lower your blood glucose (fasting, postmeal, and HbA1c) is the better, with a few caveats: 1) you're not going hypo, 2) it's through diet and exercise, but not a crazy diet or a crazy exercise regime, 3) it's not with meds.

 
  • Like
Reactions: 8 people

jack412

Expert
Messages
5,618
Type of diabetes
Type 2
Treatment type
Tablets (oral)
@pavlosn I though we have already done this?
from your study when you take out the people who have hepatitis C.
people 4 - 4.4% have the best chance of living


model 4.... 0.92
upload_2014-12-3_18-57-0-png.8763
 
Last edited by a moderator:

douglas99

Well-Known Member
Messages
4,572
Type of diabetes
I reversed my Type 2
Treatment type
Other
It's an interesting question.
I'm firmly wanting to stay exactly where I am, in HbA1c, fasting levels, and peaks and troughs.

I think the subject, of the in depth study of BG, and diet, is too new, and any data is from my generation, so by definition there can't be any long term study, as we're it..
I'm diabetic, so even the study of the 'normal' population may not apply to me in the end.

But, I know uncontrolled diabetes is bad.
I know that direct complications come from having too much BG in the blood.
So some effects are a straight causality, so those I can hope to avoid.
None diabetics don't get them, so if I keep my BG in the normal range, hopefully I won't. But normal will do me, I don't need to go lower.
Indeed, as some data shows, low in itself may link to some other effects.
But it's all unproven so far, and could simply be two side effects of an unknown third cause for all I know.

So I've decided to go for the same numbers most other people have, who live an average life, with average illnesses, and average BG.
I seem to be able to hit the figures I hit, without devoting my life to it, I'm happy if it's 5's rather than 4's, as it's 'normal' for me by the look of it.
Maybe I'm wrong, I know I could force my BG down, but I'd be concentrating too hard on that, and I'd still only be 'normal' at the end of it.
Especially as you seem to have found, it's unproven good or bad.

I view any gain on my lifespan by good control as a bonus, so I'm not going to look back at the end and regret I could have got a year or two extra maybe, I'm going to look back at the extra decades (hopefully), and accept I did the best I could, based on my decisions on what I knew at the time.

It's the same on any gamble, my target on the stockmarket is always to sell in the top ten percent of gains, if I sell stock near the top, I'm happy at the gain I crystallise at the time, if it goes up again the next day, I'm not upset.
 
  • Like
Reactions: 5 people
Messages
6,107
Type of diabetes
Type 2
Treatment type
Diet only
This is, to be frank, a pile of absolute BS. If you are able to use a CGM/libre like device and manage glucose levels accordingly, there is no reason at all why you are definitely spending more time in a Hypo.

I think the man was making the point that if you didn't have a CGM then you had the problem so rush out and buy one.