Libre predicted HbA1c versus actual HbA1c

[Satsooma]

Good morning,

I started using the FreeStyle Libre a few weeks ago, and am finding it very interesting compared to finger prick testing. I began with the two week trial and have since continued to purchase as I’m enjoying learning from the all the data.

The Libre seems to run slightly higher than finger prick testing and has done on all sensors used so far. But my predicted HbA1c so far is lower than it would usually be.

I have my HbA1c done quarterly with the next one due next month. I’m curious to hear how the Libre predicted HbA1c has compared to the actual HbA1c for others.

 

[In Response]

Firstly, the Libre reports an Estimated HBA1C because it has no access to blood so cannot measure the glucose attached to the haemoglobin within it.
For me, with 3 or more months of data, Libre underestimated my HBA1C by over 20%.
I find it a useful metric to track to show my progress but not an indication of my blood test HBA1c.

 

[Satsooma]

Ah yes.. estimated, not predicted! Wrong choice of word there.

That is interesting, thank you.

 

[jaywak]

After my first review of using the Libre for six months it estimated my HbA1c would be 51 but the blood test showed 48 so not too far out !

 

[DeejayR]

I use a Libre every six months or so and it's always optimistic compared with the HbA1C figure. But it's useful as a 24-hour-a-day guide.

 

[Satsooma]

Thank you both.

I’m a bit suspicious of the lower estimated HbA1c but finding it very useful.

 

[Melgar]

I have found Libre under estimates my HbA1c.
My Contour Next One glucometer is not that accurate either, it always reads low.

 

[Jasmin2000]

HbA1c is a physical measure of BG control over 3 months. Estimated (e)A1c is an algorithm based on sensor glucose (not BG) and is not related to HbA1c methods. There were moves about 5 years ago to rename it to Glucose Management Indicator (GMI) to avoid confusion.

From the above reference: "Many patients and clinicians find the eA1C to be a helpful educational tool, but others are often confused or even frustrated if the eA1C and laboratory-measured A1C do not agree. In the US, the Food and Drug Administration determined that the nomenclature of eA1C needed to change. This led the authors to work toward a multipart solution to facilitate the retention of such a metric, which includes renaming the eA1C the glucose management indicator (GMI) and generating a new formula for converting CGM-derived mean glucose to GMI based on recent clinical trials using the most accurate CGM systems available."

So the GMI can give you a handle on how well your control is over different time periods, based on sensor readings. The algorithm includes a factor to compare sensor glucose to real BG so that the units come out the same, but it will never predict HbA1c as we all have different variables.

 

[Rachox]

I always find my Libre estimate is lower than the lab blood test HbA1c , my last two were
Libre 33 then 31
Lab tests 39 then 36

 

[Satsooma]

HbA1c is a physical measure of BG control over 3 months. Estimated (e)A1c is an algorithm based on sensor glucose (not BG) and is not related to HbA1c methods. There were moves about 5 years ago to rename it to Glucose Management Indicator (GMI) to avoid confusion.

From the above reference: "Many patients and clinicians find the eA1C to be a helpful educational tool, but others are often confused or even frustrated if the eA1C and laboratory-measured A1C do not agree. In the US, the Food and Drug Administration determined that the nomenclature of eA1C needed to change. This led the authors to work toward a multipart solution to facilitate the retention of such a metric, which includes renaming the eA1C the glucose management indicator (GMI) and generating a new formula for converting CGM-derived mean glucose to GMI based on recent clinical trials using the most accurate CGM systems available."

So the GMI can give you a handle on how well your control is over different time periods, based on sensor readings. The algorithm includes a factor to compare sensor glucose to real BG so that the units come out the same, but it will never predict HbA1c as we all have different variables.

Thank you for this.

 

[HairySmurf]

The Libre seems to run slightly higher than finger prick testing and has done on all sensors used so far.

I saw something similar, only the Libre 2 showed me numbers that were consistently lower than finger prick testing.

I was curious as to which was closer to the mark as I wanted a clearer idea of just how low I had managed to get my fasting BG. I tried to identify the most accurate finger prick tester I could find and came across this study: Link

Based on that study and review articles I found online I ordered a Contour Next meter and some test strips. For a time I tested each drop of blood twice when testing first thing in the morning - once with the Contour Next and once with my existing meter, a Glucomen Areo GK. The Next gave me results which were consistently lower than the Areo, averaging around 0.5 mmol/L difference. The Next also agreed with the fasting numbers I got from the Libre, though I've only used a single sensor so far so I can't say if that similarity applies to all Libre 2s.

I wouldn't suggest shelling out for a new meter by any means. The strips for the Contour Next are expensive and I certainly won't be self-funding lots of strips for it. No meter is even close to perfectly accurate and what really matters is consistency - I would imagine that the observed rise in BG 2 hours after eating for example would be similar using any meter, even if the specific readings vary between meters. As regards differences between Libre and a specific finger prick meter though - it could be either one that is off the mark, or both.

 

[In Response]

As regards differences between Libre and a specific finger prick meter though - it could be either one that is off the mark, or both.

Or it could be a compression low or it could be insertion trauma or it could just be the 15% accuracy tolerance results in a high difference when your BG is high.
However, the difference in the BG readings on a Libre is only a small part of the difference in HBA1C measure from blood and tested in a lab and the estimated HBA1C which uses a completely different approach on a Libre.

 

[TheSecretCarbAddict]

I've been using Freestyle Libre 2 reader and their Optium Blood Glucose strips (you can also test your ketones if you get the right type of test strips) for checking my lows. My thinking has been: the same manufacturer and the same device for both flash and fingerprick monitoring, so hopefully aligned + can capture all the data in one place. Libre 2 app was screaming at me yesterday evening, showing my level as 3.6, while my finger prick gave me 2.3. Out of curiosity, I did another comparison just now: phone app at 5.4, and finger prick at 4.4. Could be the different nature of the two tests, delay, etc, but if this difference holds consistently, it would indicate that my Libre 2 sensor would slightly overestimate my BG levels. Might play with other meters when I save up.

 

[Satsooma]

Thank you all.

 

[Taighnamona]

Libre estimated 38 and my actual A1c was 41 just a few days ago.

 

[eventhorizon]

Dexcom / xdrip estimated 52 actual was 48. A nice surprise!

 

[oldminer88]

A1C was 39 Libre actual HA1c 46

 

[Rachox]

My latest bloods last week showed HbA1c at 33, Libre was saying 31, so Libre still lower but less of a difference this time.