Freestyle libre reliability

[Djdiabetic]

Hi,

second post so if its irrelevant then please delete. I have just started a trial on the freestyle libre, I am just enquiring to see what the reliability is like and to see if I should still finger prick? I know its interstitial blood that it analyses. but has anyone found a big varience from their previous meter?

thanks for all advice

 

[Brunneria]

Yes. Please still finger prick, especially before driving and when calculating your insulin.
The Libre instructions are very clear that the Libre is not a substitute for a glucometer for these tests.

Have a good read of the paperwork that comes with the Libre and you will see it all explained there.

Personally, I find that each sensor is consistent, but that the consistency tends to run either higher or lower than my glucometer. So one sensor might be 0.2-0.1 mmol/l higher than my prick tests, while another might run 0.3 lower... I 'calibrate' each one against my glucometer and once I have the measure of it, I use prick tests very rarely for the life of the sensor - but then, I am not on insulin.

Hope that helps.

 

[Djdiabetic]

brilliant cheers I have done a finger prick test to try it out and see the difference it seems to be running 0.3mmol out which isn't too bad I feel that it may just be a case of getting used to the whole new concept of interstitial blood readings haha

thaks for your help

 

[Bluetit1802]

brilliant cheers I have done a finger prick test to try it out and see the difference it seems to be running 0.3mmol out which isn't too bad I feel that it may just be a case of getting used to the whole new concept of interstitial blood readings haha

thaks for your help

Don't forget the sensor runs 10 to 15 minutes behind the finger pricks, so it is finger prick, wait 10 to 15, scan.

 

[Scott-C]

I've been using it for about 8 months. They can be quirky at times, some have been closer than others and it can sometimes take a day or two before I decide whether it's a "good" sensor or not. Despite the quirks, I think it's still well worth it once you learn how to adjust for differences.

Sometimes the differences are useful: for example, if it normally runs lower than blood, but I've taken a bolus and I'm wondering whether it's at work yet, if a blood test then shows lower than the libre, that's a clue that the insulin is now working - quite useful to avoid possible insulin stacking.

Inserting it causes a small amount of damage which can throw readings out so many insert about a day or so before activating to allow it to "settle".

Also, I try to activate a new one just before I get up in the morning on the basis that that's the time of day that blood glucose will probably be closest to interstitial fluid glucose. Don't know if there's any science to back that up but it seems to work for me: last few I've done this way have been closer than others.

Hasn't made diabetes "fun" but definitely easier and less guesswork!

 

[GrantGam]

There is no such thing as an irrelevant post on the forum @Djdiabetic

As @Brunneria has mentioned, you should always continue with finger prick tests because of the following:

-to satisfy the requirements of the DVLA for driving prerequisites, if you live in the UK

-to ensure your Libre isn't giving you ropey readings.

You'll more than likely find your Libre is most inaccurate at times of rapid changing BG. Such times would include after meals and during exercise. This further reiterates the point that @Bluetit1802 has made regarding the 10-15 minute lag that the Libre can have being conventional blood glucose testing.

It's a great piece of kit and I was always happy to bolus from it (although that is considered taboo by many).

 

[Deleted Account]

My experience is that the accuracy of the Libre depends on your reading as the accuracy is quoted as a %. So the higher the reading the greater the actual variation.
When I trialled the Libre, I was very concerned by this variation so started to investigate the acceptable accuracy (the rep leading the Libre trials said they conform to these standard). According to diabetese.co.uk (http://www.diabetes.co.uk/blood-glucose-meters/blood-glucose-meter-accuracy.html):

Since the end of May last year, meters need to meet these accuracy guidelines 95% of the time:

• Within ± 0.83 mmol/l of laboratory results at concentrations of under 5.6 mmol/l
• Within ± 15% of laboratory results at concentrations of 5.6 mmol/l or more

This means, meters can be wrong 5% of the time. It also means, the higher the reading, the less accurate it has to be. So if you compare two different meters on blood tested to be 15mmol/l (above the level ketones are likely to be present) in the laboratory, one meter could read as high as 17.2 and the other could be as low as 12.7.

 

[richyb]

I found sometimes the sensors are ok and sometimes not so good. I now sometimes use a dexcom which is very accurate. But unfortunately mine only seems to last 8 or 9 days. Yet I have read of others getting them to average much longer

 

[Andy_Totnes_]

Having been using the libre for nearly a year now I can give you the benefits of my experience.
Accuracy, according to the Freestyle helpline is +-14%. I find this is about right in the 6 to 12 range after allowing for the 10 minute delay. Below this they can read very low and above they tend to read high. For example I checked on waking one morning and it said 2.4 I then checked with a test strip in the Freestyle meter itself which said 3.9, then my old faithful contour next which said 4.5. Top end it tends to read up to 3 or so high but that depends on the individual sensor.
So the message is to calibrate each sensor by doing finger pricking.
As to convenience it is wonderful!
When they fail (5 so far) just call the helpline and they will ask for the error codes and the serial number and send you a new one. When your send the old one back make sure you get a proof of posting, they hassled my for months over one that didn't arrive.
Brilliant device Love it. Just waiting for Teresa May to tell the NHS to pay for it.

Andy in sunny Totnes

 

[Silas]

Hi,

second post so if its irrelevant then please delete. I have just started a trial on the freestyle libre, I am just enquiring to see what the reliability is like and to see if I should still finger prick? I know its interstitial blood that it analyses. but has anyone found a big varience from their previous meter?

thanks for all advice

Hi.
I have been using the Libre FreeStyle on and off for about 9 months since I went on the I HART diabetic research trial last year.

The Libre does not replace your blood glucose meter; it is supplementary to it. I have found that there can be significant deviations between the readings.

These are my observations:
1. Interstitial readings are not the same as blood glucose readings so are not comparable
2. The latter are more up to date by about 15-20 minutes than the Libre
3. The Libre must be using some kind of averaging algorithm to calculate a number that approximates to your blood glucose reading and to give an indication of the direction in which your reading is heading. This is not foolproof.
4. At times, no direction indication is given at all even though subsequent readings show falls or rises
5. I have had up to 4 consecutive readings during an hour on the Libre that were lower each time, but which we indicated as "Stable" i.e. not rising and not falling. This resulted in a hypo.
6. Because of the delay it can be hard to anticipate or avoid hypos if you have few warnings
7. When your blood glucose meter reading is higher than normal, the Libre system seems to indicate significantly higher results than my meter. I have experienced up to 45% difference!
8. I have checked my BG meter with control solutions and it is bang in the middle of the expected tolerances
9. In terms of reliability, I accidentally knocked a sensor during the I HART trial when I was working in a confined space. This rendered it "dead".
10. Since then, I had one sensor that stopped working just over 24 hours after it was applied. Abbott agreed to supply me with a replacement.
11. I have seldom had readings that agree with my BG meter. Those that are close are always at the low end of the spectrum.
12. You might think I am being critical of the Libre system but it has several benefits:
a. You can take as many readings as you like (I have even done them while driving)
b. Overnight it stores up to 8 hours data which it downloads to your reader so all the gaps are filled in
c. You do not need a reader if you have an Android smart phone (as I do) and the LibreLink App
d. Another person can also monitor your levels on their own phone (Useful for partners and parents)
e. The dedicated reader costs £160 so if you already have a smart phone you can avoid buying the reader
f. There are lots of useful statistics that are provided with the LibreLink App
g. These can help you identify regular patterns that can be addressed by dose changes either in your basal or bolus
h. The direction indicator is a real boon in warning about undetected hypos approaching
i. The sensor requires no calibration
j. The sensor is easy to apply
k. It is really useful during energetic sports activities for ease of checking
l. It would be good as a short term monitor for hospital monitoring and dose adjustment purposes
m. The Libre is waterproof for up to 20 minutes so showering is not a problem. I have not tested it for a long swimming session. I might do this when a sensor is nearing expiration.

13. There are some disadvantages:
a. Cost (It is not currently funded on the NHS, but this may change)
b. It cannot actively warn you that a hypo is imminent (You must scan to get a result) so night time hypos may be missed
c. Direction indication may be missing or incorrect at times
d. It is not a substitute for your BG meter
e. I have noticed some initial discomfort (which usually disappears and is replaced by itching) after applying the sensor
f. As yet, I have not found a way to download the data from my smart phone to Diasend
g. Because of the delay factor, it is very easy to over-react to an anticipated hypo resulting in larger than necessary swings in BG levels
h. When dealing with hypos, your BG meter should be used because the delay factor is too great to ascertain if your corrective action is working or not
i. I would not want to rely on it for dose calculation

Sorry if this has been overly detailed but I feel it is best to be honest. Overall I would give Libre 4 starts out of 5. Perhaps the calibration and direction algorithms will improve with time and controlled trials. It will never be able to actively warn you that you are going low. Only the Dexcom system can do that, but it is not available to buy in the UK yet.

 

[Books1]

If you want your Libre data on diasend you have to 'share' your info - bottom option on the menu in the Libre app.

 

[Melody125]

I find the readings are inconsistent, sometimes spot on, other times way out. However I find it really useful when I am doing my exercise classes and during the night, so much better to be able to scan in bed than have to get up to finger prick. I feel I have been able to see the trends of highs and lows so have adjusted my pump settings accordingly, resulted in fewer hypo's. Still finger prick 3ish times a day to check, especially if it us showing very high, or low readings. Wish I could afford to use it all the time. Also I knock them off really easily so tape them on now. I also swim for 1 hr every week with no problems. Hope this helps.

 

[donnellysdogs]

I had libre. For 6 months really good and accurate.
I can only use it in one arm.
After 6 months I started to have consistent 2ml lower readings... sensor, after sensor after sensor.
Spoke to my wonderful consuktant about this and he said, just use it as a guieline to patterns. Some peoples fluid does read lower than the bloods consistently.

I think my arm was worn out.

Anyhow, on Weds this week I was lucky enough to get my Guardian Connect CGM. I can use this in my fatty part of back and bum.
The rep advised not to use it in stomach as despite putting on almost 2stone and now up to a size 10 my stomach still has too much muscle and not enough fat.

So far I am back to really accurate readings with the fluid compared to blood again.

I just think my arm was worn out.

My consultant said to me that people do blame the libre rather than thinking its their bodies. I thought all the way along it was my body...

 

[Scott-C]

Hi.
I have been using the Libre FreeStyle on and off for about 9 months since I went on the I HART diabetic research trial last year.

The Libre does not replace your blood glucose meter; it is supplementary to it. I have found that there can be significant deviations between the readings.

These are my observations:
1. Interstitial readings are not the same as blood glucose readings so are not comparable
2. The latter are more up to date by about 15-20 minutes than the Libre
3. The Libre must be using some kind of averaging algorithm to calculate a number that approximates to your blood glucose reading and to give an indication of the direction in which your reading is heading. This is not foolproof.
4. At times, no direction indication is given at all even though subsequent readings show falls or rises
5. I have had up to 4 consecutive readings during an hour on the Libre that were lower each time, but which we indicated as "Stable" i.e. not rising and not falling. This resulted in a hypo.
6. Because of the delay it can be hard to anticipate or avoid hypos if you have few warnings
7. When your blood glucose meter reading is higher than normal, the Libre system seems to indicate significantly higher results than my meter. I have experienced up to 45% difference!
8. I have checked my BG meter with control solutions and it is bang in the middle of the expected tolerances
9. In terms of reliability, I accidentally knocked a sensor during the I HART trial when I was working in a confined space. This rendered it "dead".
10. Since then, I had one sensor that stopped working just over 24 hours after it was applied. Abbott agreed to supply me with a replacement.
11. I have seldom had readings that agree with my BG meter. Those that are close are always at the low end of the spectrum.
12. You might think I am being critical of the Libre system but it has several benefits:
a. You can take as many readings as you like (I have even done them while driving)
b. Overnight it stores up to 8 hours data which it downloads to your reader so all the gaps are filled in
c. You do not need a reader if you have an Android smart phone (as I do) and the LibreLink App
d. Another person can also monitor your levels on their own phone (Useful for partners and parents)
e. The dedicated reader costs £160 so if you already have a smart phone you can avoid buying the reader
f. There are lots of useful statistics that are provided with the LibreLink App
g. These can help you identify regular patterns that can be addressed by dose changes either in your basal or bolus
h. The direction indicator is a real boon in warning about undetected hypos approaching
i. The sensor requires no calibration
j. The sensor is easy to apply
k. It is really useful during energetic sports activities for ease of checking
l. It would be good as a short term monitor for hospital monitoring and dose adjustment purposes
m. The Libre is waterproof for up to 20 minutes so showering is not a problem. I have not tested it for a long swimming session. I might do this when a sensor is nearing expiration.

13. There are some disadvantages:
a. Cost (It is not currently funded on the NHS, but this may change)
b. It cannot actively warn you that a hypo is imminent (You must scan to get a result) so night time hypos may be missed
c. Direction indication may be missing or incorrect at times
d. It is not a substitute for your BG meter
e. I have noticed some initial discomfort (which usually disappears and is replaced by itching) after applying the sensor
f. As yet, I have not found a way to download the data from my smart phone to Diasend
g. Because of the delay factor, it is very easy to over-react to an anticipated hypo resulting in larger than necessary swings in BG levels
h. When dealing with hypos, your BG meter should be used because the delay factor is too great to ascertain if your corrective action is working or not
i. I would not want to rely on it for dose calculation

Sorry if this has been overly detailed but I feel it is best to be honest. Overall I would give Libre 4 starts out of 5. Perhaps the calibration and direction algorithms will improve with time and controlled trials. It will never be able to actively warn you that you are going low. Only the Dexcom system can do that, but it is not available to buy in the UK yet.

Those are all fair points which you make, @Silas, (apart from cost of reader - it's not £160 on it's own, it's just over £50, it's about £150 for a starter pack of a reader and two sensors. I've used both reader and phone, reader is much more discreet, and I'm not sure you can get the fantastically useful AGP graphs out of librelink but could be wrong).

Re the inconsistencies, I've never viewed it as providing hard and fast answers. Instead, I think of it as providing more clues which I wouldn't have if using strips alone.

Strips, libre and past experience/intuition working together are a pretty powerful team despite all of them having limitations on their own.

I agree absolutely with your view that it is supplementary - Abbott's advertising pitch of no more pricking is misplaced and has maybe led to too high expectations. Still, that line is an easy sell for ad people, compared to explaining being able to see variability and the advantages of modal day/AGP over a simple hba1c result.

I recently picked up a cheap copy from Amazon of Wiley in vivo glucose sensing. All of it is way above my limited secondary school chemistry knowledge from 32 years ago, but I can get the gist of it enough to take my hat off to the people who've figured out the horrendously complex electrochemistry/engineering/computing/biology involved. It amazes me that cgm/fgm produces any result at all let alone ones which are a bit out but still ball park enough for me.

 

[grva05]

I had cause recently to be wearing two sensors at the same time - mine and another as part of a research project. The difference between the sensors logged 1.0mmol at low readings and could head towards 1.5mmol during blood sugar spikes (which was the target of the research and were deliberately induced by high carb meals). I'd never thought about the accuracy of the sensors before but this was an eye opener. Since then I've looked carefully at the results being logged by new sensors. The case of two sensors looks to have been a fairly extreme one though it is just within the tolerances declared by Abbott if you accept extreme high versus extreme low as the situation. Definitely baseline each new sensor against a reliable real blood test. That negative report aside I feel rather marooned when I take time out from wearing a sensor for financial reasons.

 

[Scott-C]

I had cause recently to be wearing two sensors at the same time - mine and another as part of a research project. The difference between the sensors logged 1.0mmol at low readings and could head towards 1.5mmol during blood sugar spikes (which was the target of the research and were deliberately induced by high carb meals). I'd never thought about the accuracy of the sensors before but this was an eye opener. Since then I've looked carefully at the results being logged by new sensors. The case of two sensors looks to have been a fairly extreme one though it is just within the tolerances declared by Abbott if you accept extreme high versus extreme low as the situation. Definitely baseline each new sensor against a reliable real blood test. That negative report aside I feel rather marooned when I take time out from wearing a sensor for financial reasons.

Can empathise with the marooned feeling! One of mine broke at work, the rest of the day till I got a new one on, it felt like going back to the Stone Age.

Re the differences between the two sensors, I'm not convinced that if you've got such a complex piece of plumbing as the bloodstream, which then deliberately leaks stuff into interstitial fluid, that the levels of glucose in the two sensor areas is going to be the same. Testing accuracy by comparing results from two different areas is worthless unless you're convinced that glucose in IF is always evenly distributed through every bit of your body and that seems unlikely to me. You wouldn't be comparing like with like, because the conditions in the two areas will be different.

Maybe capillaries in one area are "leakier", maybe one insertion cut a few more capillaries than the other - would that make a difference, can you tell whether that's even happened? Some of my detached sensors have blood on the underside, others don't - does that make a difference?

Meters are known to be inaccurate but near enough. That must be true for sensors too.

Sensors can get gummed up by all the junk floating around in IF, it's called "biofouling" and can hinder it reading properly like a dirty windscreen hinders seeing. Maybe one of your sensors has been junked up more than the other.

 

[Djdiabetic]

Hi how did you get the guardian connect? is there somewhere I can try it or buy it?


OTE="donnellysdogs, post: 1415530, member: 17713"]I had libre. For 6 months really good and accurate.
I can only use it in one arm.
After 6 months I started to have consistent 2ml lower readings... sensor, after sensor after sensor.
Spoke to my wonderful consuktant about this and he said, just use it as a guieline to patterns. Some peoples fluid does read lower than the bloods consistently.

I think my arm was worn out.

Anyhow, on Weds this week I was lucky enough to get my Guardian Connect CGM. I can use this in my fatty part of back and bum.
The rep advised not to use it in stomach as despite putting on almost 2stone and now up to a size 10 my stomach still has too much muscle and not enough fat.

So far I am back to really accurate readings with the fluid compared to blood again.

I just think my arm was worn out.

My consultant said to me that people do blame the libre rather than thinking its their bodies. I thought all the way along it was my body...[/QUOTE]

 

[donnellysdogs]

At the moment the Connect is only available to patients with funding in place. However it should become available for patients self funding some time in April.

It is really accurate. However, my levels generally do not fluctuate too much and I do tend to pick good times to calibrate it... it is good...

 

[gbswales]

I use the libre sensors for 2 weeks then prick for two weeks (on cost grounds). I am type2 on insulin and the sensor has been accurate enough to prevent hypos. Due to poor circulation finger picking is very hit and miss and I waste a lot of strips because of not getting enough blood out. The pain from multiple prick makes matters worse so that I rarely finger test more than once a day. Risky I know but my body gets plenty of warning of hypo which is enough to even wake me. However an accurate sensor red on a smart phone and paid by nhs is my dream

 

[TheBigNewt]

How much is the Libre and the sensors and whatever other stuff that goes with it? I kind of wonder why it's not available in the States. Can one purchase it online and just use it here?