Libre 2 inaccuracy compared to a fingerprick test

[DavidDK]

Hello

Is anyone else finding that the Freestyle Libre 2 seems to read consistently 3-4 mmols lower than a fingerprick test (even when one takes into account the reported reduced "timelag" between the Libre 2 interstitial fluid test and a fingerprick blood test of just over two minutes)?

I've been using the Libre 2 for a good few months now. Had some interesting experiences with the first Libre 2 sensors (I couldn't get the alarms to work with my mobile; sensors ending prematurely, sensor applicator not fitting into the sensor pack), but on each occasion Abbott were very good and replaced the sensor.

I try and place the sensors on my arm at least 24 hours before activating them and place the sensors in the area of the lower arm designated in the pictures that Abbott have in their instruction manual. I also try to stay well hydrated.

I had a meeting with my DSN recently who was congratulating me on how much better my time in target has been since wearing the Libre 2 sensor. Even I was surprised just how few variations there have been in pre meal and post meal readings, but I assumed that as the Libre 2 sensor is more accurate than the Libre 1, and has alarms, I have been getting much better as basaling and bolusing and taking carbs to minimise hypos.

But I noticed today (the Libre 2 sensor will stop in a few hours time) that there seems to be a difference of 3-4 mmols between the Libre 2 sensor reading and a fingerprick blood test.

I need to do some more checking and I hope the discrepancy is simply because the current sensor I'm wearing is coming to an end. If it isn't, it is concerning.

 

[Deleted member 527103]

I have been using Libre 2 since the start of the year and not noticed any significant difference with Libre 1 apart from when my levels are changing direction (more about that later). I continue to find
- The first couple of days are random
- Libre is less accurate above 8 or below 4
- Sensor accuracies drift during their 2 weeks so I never assume they can be "mentally calibrated" on day one and never tested after that.

With regard to values when changing direction: as has been discussed many times, interstital fluid, which is used by CGMs, is 15 minutes behind finger pricks. With Libre 1, if you numbers were rising or falling, Libre numbers could be out by a fair amount With Libre 2, they changed the algorithm to predict what your reading will be based on the trend of the graph. This is great with regard to falling or rising fast but, if you "fix" the rise or fall in those 15 minutes, Libre will not know so the prediction can be significantly out. This is obvious, for example, when correcting a hypo.

Over the years, with Libre, I decided that I did not like the factory calibration - it did not suit me. Therefore, I chose to use an alternative app to read my sensors which allow calibration. I calibrate it with finger pricks twice a day and found that is it pretty close.

 

[LionChild]

Hello

Is anyone else finding that the Freestyle Libre 2 seems to read consistently 3-4 mmols lower than a fingerprick test (even when one takes into account the reported reduced "timelag" between the Libre 2 interstitial fluid test and a fingerprick blood test of just over two minutes)?

I've been using the Libre 2 for a good few months now. Had some interesting experiences with the first Libre 2 sensors (I couldn't get the alarms to work with my mobile; sensors ending prematurely, sensor applicator not fitting into the sensor pack), but on each occasion Abbott were very good and replaced the sensor.

I try and place the sensors on my arm at least 24 hours before activating them and place the sensors in the area of the lower arm designated in the pictures that Abbott have in their instruction manual. I also try to stay well hydrated.

I had a meeting with my DSN recently who was congratulating me on how much better my time in target has been since wearing the Libre 2 sensor. Even I was surprised just how few variations there have been in pre meal and post meal readings, but I assumed that as the Libre 2 sensor is more accurate than the Libre 1, and has alarms, I have been getting much better as basaling and bolusing and taking carbs to minimise hypos.

But I noticed today (the Libre 2 sensor will stop in a few hours time) that there seems to be a difference of 3-4 mmols between the Libre 2 sensor reading and a fingerprick blood test.

I need to do some more checking and I hope the discrepancy is simply because the current sensor I'm wearing is coming to an end. If it isn't, it is concerning.

Well, so far, in my experience, Libre 2 is reading 0.8 or more, higher, than my finger pricks. So, there you go!

 

[DavidDK]

Thanks In Response and LionChild.

My results just seem almost too good to be true. According to Libreview my time in target is almost perfect; my average glucose is 6.5mmol and my GMI is 6.1% or 43mmol.

However, I am going to give the sensor I have on my arm at present the benefit of the doubt and assume that because it will stop working shortly, it isn't as accurate as it was a week ago.

And do some more checks once I have started the new sensor.

I spoke to Abbott to try and get some reassurance about the accuracy, and they very kindly offered to replace my existing sensor.

 

[KK123]

Libre is less accurate above 8 or below 4

I do like the libre (I'm still on Libre 1) but it's a shame if the accuracy wavers outside of the 4 to 8 range, if we were all within that range we wouldn't need a libre at all. I suspect for most of us we are more interested in our readings below 4 and above 8 as they are the ones we really need to be accurate in order to take action. I suppose I'm asking for too much as it's still a darn sight better than pricking fingers and getting snapshots 4 times a day and nothing else. x

 

[Deleted member 527103]

I do like the libre (I'm still on Libre 1) but it's a shame if the accuracy wavers outside of the 4 to 8 range, if we were all within that range we wouldn't need a libre at all. I suspect for most of us we are more interested in our readings below 4 and above 8 as they are the ones we really need to be accurate in order to take action. I suppose I'm asking for too much as it's still a darn sight better than pricking fingers and getting snapshots 4 times a day and nothing else. x

In my experience Libre is pretty good at reporting that we are under 4 or over 8 but it is not accurate enough to be able to make a correction without a finger prick.
So the Libre 2 alerts at about the right time but I will then finger prick to determine what kind of correction is needed.

 

[KK123]

In my experience Libre is pretty good at reporting that we are under 4 or over 8 but it is not accurate enough to be able to make a correction without a finger prick.
So the Libre 2 alerts at about the right time but I will then finger prick to determine what kind of correction is needed.

May I ask, do you use the libre for insulin dosing? I was told that I was to do so when prescribed the libre and they reduced my finger prick test strips to one box a month (around 1.5 per day essentially).

 

[Deleted member 527103]

May I ask, do you use the libre for insulin dosing? I was told that I was to do so when prescribed the libre and they reduced my finger prick test strips to one box a month (around 1.5 per day essentially).

Yes I do if the current value is been 4 and 8.
I test with finger pricks at least once a day when my levels are flattish (usually, first thing in the morning) to check the sensor is close enough so I am confident to dose from it.