Libra sensor dangerously inaccurate.

[Tye7]

Hi, I'm a new member, and I feel I need to make people aware of the dangers of relying on a Libra sensor.
I've been using the sensor on and off now for some time, and I am astonished at how the NHS allowed this to hit the market
In my experience, the readings that the sensor shows are at least 2 mmol higher than a blood prick, which in itself is dangerous, especially if you are elderly, or you like to keep the numbers tight, like I do.
I have lost count of how many times I have changed the sensor and reported this to Libra.
But their response is always the same: we'll send a new sensor out.
This does not work!
I understand that there is going to be a difference compared to finger pricking, but in my opinion, 2 mmol, and more isn't acceptable.
An extreme example is when I attend the gym, at least 3 times per week.
The readings that I get are always in the very high teens, sensor 18, Blood 6.5.
That isn't an isolated case, it's every single time, (high teens)
Has anyone else experience this sort of nonsense.

 

[Robin]

The Libre , and its rival the Dexcom, both seem to have some people for whom it just doesn’t work, or work accurately enough, and I’m sorry if you turn out to be one of them.
I’ve been using it since 2015, and I’ve always found it spot on with my meter in the middle range, and not too far out at the extremes (apart from two sensor failures that needed to be replaced in the whole of that time). I do recognise that it has its limitations even for me, though, on that it will overshoot at the top end, until the algorithm sorts itself out, and will read low at the bottom end, and is prone to having the odd dip into the red at night if I lie on it and compress it.
I’m sure it was carefully evaluated by the NHS before they started funding, so I assume it works well enough for the majority of people who use it.

 

[helli]

In my experience, the readings that the sensor shows are at least 2 mmol higher than a blood prick

How do you know your finger prick is accurate?

I appreciate there can be some issues with CGMs and there are limitations. This is why I wrote this thread.

Despite how inaccurate you may find it (for you), I do not believe it is dangerous if you follow the instructions to always check unexpected readings with a finger prick.

 

[Tye7]

Thank you for your response, Robin.
I guess the NHS feels that the pros outweigh the cons regarding the sensor.
As I stated in my previous statement I like to keep my numbers tight, and my last hba1c was 48, and my previous hba1c readings for the last 10 years haven't been over 50.
So, I believe that my control is excellent.
I would hate for someone to react to those sensor readings, especially, if it is showing high, and that person/ persons compensate by giving a high corrective dose.

 

[Tye7]

How do you know your finger prick is accurate?

I appreciate there can be some issues with CGMs and there are limitations. This is why I wrote this thread.

Despite how inaccurate you may find it (for you), I do not believe it is dangerous if you follow the instructions to always check unexpected readings with a finger prick.

I don't understand your question about, how do you know that your finger pricking is accurate.
It is the hold standard.!
checked

 

[PattiEvans]

No, readings on blood meters have to be accurate to within 15%. Some are better than others.

I am one of the people who finds the Libre extremely accurate, always within 0.5mmol of my meter (I check periodically) and I find I can rely on it. I have been using it since it first came out and I self funded for years. I do think though that there are some people (probably due to body chemistry) who do not find it accurate. FWIW I also keep very tight control as a rule.

Oh and as an elderly person I do not find it dangerous.

 

[Tye7]

No, readings on blood meters have to be accurate to within 15%. Some are better than others.

I am one of the people who finds the Libre extremely accurate, always within 0.5mmol of my meter (I check periodically) and I find I can rely on it. I have been using it since it first came out, and I self-funded it for years. I do think, though, that there are some people (probably due to body chemistry) who do not find it accurate. FWIW I also keep very tight control as a rule.

Oh, and as an elderly person, I do not find it dangerous.

Hi, Patti, thank you for your response.
I guess you could write about certain people's body chemistry.
However, the countless number of people on all forums across the Internet who are describing the same issue is concerning.
And that is why I posted.
I know of one example in which it has caused a very serious issue.
To the point, that person could have lost their life.
You are very fortunate that you can rely on the sensor, but I could not.

 

[Inka]

Could you ask to try a Dexcom instead @Tye7 ?

 

[Tye7]

Could you ask to try a Dexcom instead @Tye7 ?

I'm considering it, thank you.

 

[Tdm]

I don't understand your question about, how do you know that your finger pricking is accurate.
It is the hold standard.!
checked

It does sound like libre isn't working out for you
and it is important to feel trust in your sensor, especially if going for tight control. Ask to try out a dexcom. I found dexcom eorked much better for me.
I find, normally, my sensor and finger prick agree, which gives me confidence.
And you can/should be able to trust your cgm, especially as we move over to HCL and the pump and the sensor will make autonomous decisions....

 

[Lucyr]

I would hate for someone to react to those sensor readings, especially, if it is showing high, and that person/ persons compensate by giving a high corrective dose.

If following the instructions you’d fingerprick before correcting a high anyway so no risk there though?

 

[SimonP]

I find it somewhat hit and miss, it's generally in agreement but sometimes can be wildly wrong.

XDrip+ calibration deals with most of those cases, and for anything that's too large to be fixed by calibration I get a replacement.

There was a thread somewhere with a link to a journal paper showing that the interstitial sensors are less accurate in lean individuals. Perhaps the performance will improve for me over the Xmas period 🙂

 

[helli]

There was a thread somewhere with a link to a journal paper showing that the interstitial sensors are less accurate in lean individuals.

I was told they need at least 5mm of fat for the filament. When I placed sensors on leaner parts of my arm, I found them very inaccurate and needed replacing. When I found some fat, they were much better

I don't understand your question about, how do you know that your finger pricking is accurate.
It is the hold standard.!

If your meter complies with that standard! But your finger prick meter maybe faulty.

 

[rebrascora]

I am another one who has been using Libre for nearly 5 years and it has really revolutionised my diabetes management and I absolutely love it. You do have to be aware of it's limitations. I find Libre usually reads up to 1mmol lower than my "official" BG meter which is a Caresens Duo Pro. In that time I think I have only had 1 or 2 sensors which were slightly more than 2mmols out and they were reading lower not higher, so erring on the safe side I suppose, but they were replaced without quibble. I have had the odd one or 2 that didn't last the full distance and in the early days a few that got dislodged. Abbott have always been very good about replacing even when it was my fault that I caught it on clothing when I was new to them.

I think some of the people who have problems with them, have not fully appraised themselves of the way they work and when they should double check results with a finger prick, but there does seem to be a small percentage of people like yourself whose "body chemistry" is perhaps not compatible with them and we have a few members of the forum who found the Dexcom system suited them better and thankfully people have the option to swap. You could apply to Dexcom for a free trial of their Dexcom One+ before you ask to switch prescription, so that you can run it alongside your Libre and make a direct comparison.

I too put a lot of effort into keeping my diabetes management as good as I can (last HbA1c was 46) and Libre helps me enormously not only to achieve that but also to achieve an average TIR of about 90% for the past 4+ years and most importantly, it takes a lot of the mental strain out of my diabetes management, so I am incredibly grateful; to Abbott for developing it and the NHS for prescribing it.

 

[Docb]

Hang on everybody... if I can a get a bit technical.

Neither the libre or the finger pricker give you an ACCURATE blood glucose reading. They both give you an ESTIMATE of your blood glucose. To meet the required standard, the finger prick tester has to give a reading within 15% of the known concentration of glucose in a test solution. That is something they all will do comfortably. I do not know if there is an equivalent requirement for the CGM's.

What this means is that it is unwise to consider one more "accurate" than the other. They each have their own characteristics.

What this means is that different people might find the characteristics of one system easier to work with than the other. I am not T1 and have never used CGM but my assessment of looking at members' experiences is that the vast majority of T1's use both systems in combination to manage their diabetes and do far better than they ever could without them. That said, some prefer finger pricking and don't use CGM, some seem to rely on CGM. It seems to me the trick is to sort out a regime that you are comfortable with and allows you to manage your diabetes. I would council anybody not be tempted to tell all those that do it differently, but with equal success, that they are doing it wrong.

In any case it really is unwise to condemn one system or the other based on random anecdotes.

I make my usual comment that I think it absolutely amazing that it is possible to have a device that gives a reading anywhere near enough to your blood glucose level to allow very good diabetes management for most people with diabetes. That should be rejoiced.

 

[Lucyr]

What this means is that it is unwise to consider one more "accurate" than the other. They each have their own characteristics.

Fingerprick is known to be more accurate than CGM though so I disagree with this. Fingerprick involves blood in indicating blood glucose levels. CGM doesn’t involve blood and is purely estimating it.

 

[Satan’s little helper]

Fingerprick is known to be more accurate than CGM though so I disagree with this. Fingerprick involves blood in indicating blood glucose levels. CGM doesn’t involve blood and is purely estimating it.

I don't think the doctor is suggesting this. He seems to be talking about the comparison with Libre & Dexcom?

 

[Lucyr]

I don't think the doctor is suggesting this. He seems to be talking about the comparison with Libre & Dexcom?

No I do think he’s saying fingerprick isn’t more accurate than CGM because of this claim?

Neither the libre or the finger pricker give you an ACCURATE blood glucose reading. They both give you an ESTIMATE of your blood glucose. To meet the required standard, the finger prick tester has to give a reading within 15% of the known concentration of glucose in a test solution. That is something they all will do comfortably. I do not know if there is an equivalent requirement for the CGM's.

What this means is that it is unwise to consider one more "accurate" than the other. They each have their own characteristics.

 

[mashedupmatt]

What I have found is if you are running high or low, the CGM is not as accurate as a finger prick (based on how I feel, not actual numbers) - it is almost identical when in mid-range - where it is very useful is seeing where you are trending (up or down, and to what extent) - all technology has its limitations but personally I'd be lost without it

 

[Docb]

@Lucyr. When I wrote that post, an early draft made that point but it all got a bit complicated and I left it out to keep things simple.

Yes, in principle you would expect the CGM to have a larger potential error because it is measuring interstitial fluid and not blood. I am sure that there is a lot of clever stuff going on in the software built into the device to try and compensate for this but there is no way we will know what effect that has.

I fall back on the overwhelming support for CGM from forum members suggesting that although the BG estimate from CGM might have a bigger potential error than that from a finger prick, the numbers it gives are good enough for diabetes management once the user has got to grips with its characteristics and how best to work with it.

@Satan’s little helper - If you are referring to my comment I am not "the" doctor but a proper Dr like many others on the with a PhD in a non medical subject. I make the point to make sure that new comers to the forum are not misled.

@lucy is right. My comment was about CGM in general, making no distinction between brands.

@mashedupmatt - you make my point very well except my pedantic technical mind would say that the finger prick is not more accurate but that in principal it is a better estimate of blood glucose than that from the CGM. The fact that it takes loads more measurements allowing monitoring of short term movements in blood glucose can make it far more useful than finger pricking even if the potential error is bigger.