Abbot Libre system

[UK T1]

Yes, I've read a few papers saying the libre isn't as accurate if you're high or low, which is exactly why the training includes instructing users to do a capillary blood test if the libre says you're high or low.
It is also worth remembering that things like hydration and libre location play a role. There are a number of documented cases where the sensor records a low reading when it is under pressure, which is worth remembering if you sleep on your side for example!

Seems I'm also a lucky one in that whenever I've done a capillary test, waited 10 mins to allow for the diffusion delay, then scanned, they have matched (within 0.1 mmol of each other)! Annoyingly as with many diabetes related remedies some people get on fine and have almost textbook responses and others try their hardest but aren't as lucky! So very annoying, but I'm happy to be amongst the lucky ones in this case!

Also - my GP surgery always does HbA1c and current glucose as standard every 6 months, then usual annual 'extras' as necessary when bloods get sent to the lab. Perhaps case of different CCGs having different standard tests?

 

[Twisticles]

...
As for the claim that there is
a 15-20% tolerance on any kind of home measurement
I am afraid the situation is much more complicated than that.
This link explains things in more detail
https://www.diabetes.co.uk/blood-glucose-meters/iso-accuracy-standards.html
and it is important to bear in mind that the ISO accuracy standards apply to blood glucose meters.
They do not apply to interstitial fluid meters.
If anyone has the equivalent standards published for interstitial fluid meters/sensors I would love to read them, so please post a link. Thank you in advance.
...

In all fairness, I'm not going to deep dive into the finer details of the ISO standard when referencing for brevity on a forum. 15-20% accuracy is a fair summary of the standard I think.
From your link:

In 2013 new, tighter accuracy standards (ISO: 15197:2013) were drawn up, requiring that 95% of blood glucose results should reach the following standard.

• Within ± 0.83 mmol/L of laboratory results at concentrations of under 5.6 mmol/L
  (Within ± 15 mg/dl of laboratory results at concentrations of under 100 mg/dL)
• Within ± 20% of laboratory results at concentrations of 5.6 mmol/L (100 mg/dL) or more

So the standard says within 15% under 5.6mmol/L and within 20% above 5.6mmol/L. That is still 15-20% accuracy, isn't it?

Noted that the ISO applies to BG meters not ISF measurements, but when we look at the clinical evidence, the MARD for Libre is also within the standards for BG meters (11% MARD against the YSI).
https://www.nice.org.uk/advice/mib110/chapter/Clinical-and-technical-evidence

My Libres almost always read 1-1.2mmol/l lower than my glucometer readings, and my HbA1cs.
It is very consistent.
So I agree that there is a clear disparity between blood testing results by fingerprick and interstitial fluid results by Libre.
This may be down to the individual person's body. My results are far too consistent to blame manufacturing variations for the sensors.

See, this is where I get concerned. You're a mod here with a ton of posts and very knowledgeable. But your opinion is only anecdotal. When enough people post anecdotes, they become "lore", and the Libre is de facto "always reads low". This does not align with the clinical and technical evidence as posted by NICE. Could it be that your BG meter consistently reads high? We KNOW that ISF meters lag behind SMBG meters. Of course the results will be different!

On the other hand, my fasting scan this morning showed 5.6mmol/L and my fingerprick showed 5.8mmol/L. If I post this often enough will people start to believe that Libres read high?

At the end of the day, the evidence points to better anticipation of hypos and shorter time in hypo when Libre is used Vs finger-pricking. And anything that helps us is good for us. Also noted that every manufacturer of CGM/FGM recommends SMBG if readings are abnormal and before corrective doses.

 

[Twisticles]

You seem strangely in love with the Libre sensor.

I am! It saves me around 250 finger prick holes a month! My Endo gets my scans immediately and calls if I need to adjust doses.

We, as a community, need this kind of technology to work to make our lives easier. As a relative newbie I am researching like mad and trying to filter the facts from the opinions. Given that this site is such a valuable resource for diabetics including newcomers, it behoves us to be accurate with information as far as possible. A few stories from people feeling like their FGM reads low does not match the clinical and technical evidence. I feel it is very important to redress the imbalance.

 

[KK123]

The Libre combined with the Miaomiao is really good. The MM allows you to calibrate the Libre readings with a finger prick test. I happily puit these readings into my pump and bolus with them

That's good to know, I'm not massively technically minded but I am hoping to purchase the next step up to complement the libre when I've got my head around it all. x

 

[Brunneria]

In all fairness, I'm not going to deep dive into the finer details of the ISO standard when referencing for brevity on a forum. 15-20% accuracy is a fair summary of the standard I think.
From your link:

In 2013 new, tighter accuracy standards (ISO: 15197:2013) were drawn up, requiring that 95% of blood glucose results should reach the following standard.

• Within ± 0.83 mmol/L of laboratory results at concentrations of under 5.6 mmol/L
  (Within ± 15 mg/dl of laboratory results at concentrations of under 100 mg/dL)
• Within ± 20% of laboratory results at concentrations of 5.6 mmol/L (100 mg/dL) or more

So the standard says within 15% under 5.6mmol/L and within 20% above 5.6mmol/L. That is still 15-20% accuracy, isn't it?

Noted that the ISO applies to BG meters not ISF measurements, but when we look at the clinical evidence, the MARD for Libre is also within the standards for BG meters (11% MARD against the YSI).
https://www.nice.org.uk/advice/mib110/chapter/Clinical-and-technical-evidence



See, this is where I get concerned. You're a mod here with a ton of posts and very knowledgeable. But your opinion is only anecdotal. When enough people post anecdotes, they become "lore", and the Libre is de facto "always reads low". This does not align with the clinical and technical evidence as posted by NICE. Could it be that your BG meter consistently reads high? We KNOW that ISF meters lag behind SMBG meters. Of course the results will be different!

On the other hand, my fasting scan this morning showed 5.6mmol/L and my fingerprick showed 5.8mmol/L. If I post this often enough will people start to believe that Libres read high?

At the end of the day, the evidence points to better anticipation of hypos and shorter time in hypo when Libre is used Vs finger-pricking. And anything that helps us is good for us. Also noted that every manufacturer of CGM/FGM recommends SMBG if readings are abnormal and before corrective doses.

Yup. Part of my comment was anecdotal. It was also clearly explained as being anecdotal. That is a good thing because it shows (like other people who have posted on this thread) we all have individual experiences that may, or may not, match up with the manufacturers declared info, or the results of a study on people who may, or may not, have different circumstances from ourselves. Anecdotes are great, so long as they are not stated as universal facts, and I was very clear that my experience was mine, and no one else's.

One of the reasons this forum is so helpful to people is because all our varying opinions and experiences allow members to learn learn from personal experiences, share references, exchange views, and make up their own minds, then change their minds as they learn more. Opinions, like science, constantly evolve.

I am both a mod and member, so I get to learn and share my experiences and knowledge alongside everyone else. I also get to say (pretty frequently) 'please don't take anything anyone says on this forum as gospel, because they are just strangers on the internet' and that since this is a forum, please use it as a springboard for your own research, don't take anyone's opinion as proof, and seek a medical professional if necessary, because members here are not allowed to give medical advice. In that respect we are all exactly the same. I also take care (as I did in my post above) to distinguish between my own personal experiences and other people's personal experiences as well as links to other sites and information sources.

We have had many, many different threads on this forum about the Libre, its accuracy, Abbott's customer service, skin reactions to the glue, sensor failure, additional gadgets and Apps, delivery delays and different versions of the Libre around the world. All very useful to members. All peppered with anecdotes, references and informative discussion. I for one am looking forward to many more such discussions.

 

[KK123]

Just want to add I have followed this post and found it very informative across the board, I do hope that such posts (about 'new technology') don't turn into a 'this is rubbish V this is great' scenario. That would be a shame because anything that improves the life of a person with diabetes (even in a very small way) is something to be grateful for. Of course people will have had good and not so good experiences and it is all relevant but in the end as long as we are informed, we can make our own minds up, take on board any 'issues' and report back for others. In a nice way!

 

[LittleGreyCat]

Just to add to @Brunneria , if someone reports a personal experience then that is anecdotal and should be weighted as such.

If several people report similar experiences then that is also anecdotal but (assuming that they are all telling the truth which seems fair) this adds weight to the anecdotal evidence.

If a significant number of people report the same effect then that becomes strong anecdotal evidence, and should be considered as an indication that "the reported something" can happen.

Over the years we have seen a number of instances where personal experiences reported on this Forum have conflicted with the official line and are now being accepted, including statins, the Eatwell plate, LCHF, cholesterol (both eating and blood levels).

 

[Sailorsf]

I have been using Libre for 8 weeks now (on my 4th Libre now) and I have Miaomiao 2 for 3 weeks now. I didn't suspect the accuracy of the first 2 Libre (I didn't do any comparison). The 3rd one I pricked (using Contour next) and found the 3rd Libre to be low 15%-20%.

Now I'm on my 4th Libre. Here are the readings just 10 minutes ago (3pm PST):
Libre app (133 mg/dl or 7.4 mmol/l);
Tomato with mm2 (100 mg/dl or 5.6 mmol/l);
Contour (200/199 mg/dl or 11.1 mmol/l, pricked twice).
This is all within 1-2 minutes of sitting here.

What do you guys think of this? Is the mm2 out of wreck? Is the Libre 35%-40% off?

I love Libre for the trend but these numbers are all over the place. Very confused.

 

[Walking Girl]

Count me as one who has a more textbook experience with the Libre. I’ve had a few sensors that were whacky, usually they started very low and then fell off the charts, but they were so far off it was pretty easy to spot.

no technology is perfect. Heck, I’ve even read articles that say carb counts in packages can be 20-30% off, so it’s all a bit of a high-wire act. I have two meters, a Freestyle Lite, and a Bayer Contour Next. When tested against each other, the Contour next runs 10%-20% higher the vast majority of the time.

all in all, I’ve found the Libre to be very helpful. And my A1c has been with .2 to .3 of the estimated A1c based on the standard average BG to A1c conversion.

 

[Peter03]

I am finding the libre very accurate since the last update, yes you do get faulty or strangs reading with the odd senser but that is true of all sensers,

 

[Sailorsf]

I have been using Libre for 8 weeks now (on my 4th Libre now) and I have Miaomiao 2 for 3 weeks now. I didn't suspect the accuracy of the first 2 Libre (I didn't do any comparison). The 3rd one I pricked (using Contour next) and found the 3rd Libre to be low 15%-20%.

Now I'm on my 4th Libre. Here are the readings just 10 minutes ago (3pm PST):
Libre app (133 mg/dl or 7.4 mmol/l);
Tomato with mm2 (100 mg/dl or 5.6 mmol/l);
Contour (200/199 mg/dl or 11.1 mmol/l, pricked twice).
This is all within 1-2 minutes of sitting here.

What do you guys think of this? Is the mm2 out of wreck? Is the Libre 35%-40% off?

I love Libre for the trend but these numbers are all over the place. Very confused.

As an update, the above readings were taken 4 hours after the Libre was first applied. After a day or so, the readings among ALL three devices became about the same within 1-10% of each other (5 tests over 3days). This is amazing! This has gone from a spread of 40-50% off to 1-10% off among the 3 devices.

My take away here is: (1) the first few hours after applying the Libre can be wild; (2) I still don’t understand why is mm2 off by 30% from the Libre app in the first readings.

All in all, Libre has been a game changer for me.

 

[Sailorsf]

As an update, the above readings were taken 4 hours after the Libre was first applied. After a day or so, the readings among ALL three devices became about the same within 1-10% of each other (5 tests over 3days). This is amazing! This has gone from a spread of 40-50% off to 1-10% off among the 3 devices.

My take away here is: (1) the first few hours after applying the Libre can be wild; (2) I still don’t understand why is mm2 off by 30% from the Libre app in the first readings.

All in all, Libre has been a game changer for me.

As a further update on the Libre, i started to check the Libre again in the last 5-6 days of the same Libre (against the same Contour Next). The spread was 20%-30% between the Librelink app and Contour Next pricks (Libre was always lower). At first Tomato was closer to Librelink, then I figure out how to input values to Tomato and so Tomato adjusted its readings closer to the Contour.

Take away: regular finger pricks are still needed to calibrate the Libre - maybe once every other day or when the sugar level doesn't look right.