Got my first ever Libre, think it's faulty

[pawprint91]

So yesterday after a face to face group appointment at the hospital, I was presented with my first ever Libre sensor (Libre 2) - so exciting! Got home and got it on yesterday evening. Cleaned area with alcohol and actually got my dad (retired nurse) to put it on for me to make sure it was done right (didn't trust myself with the first one!). First reading seemed good - 6.2 (didn't check this against finger prick). However, within an hour my low alarm was going off, telling me I was 3.4 and dropping. Finger pricked and was actually 4.8. Ended up turning the alarms off as it was trying to tell me I was 3.0 when I was actually 6.1 following a biscuit before a delayed dinner. Tried to resist obsessively scannning ... before bed I was a 7.1 on a fingerprick, 6.2 on libre. This is more like it, I thought. Woke up this morning to several notifications coming in overnight saying that I had lost signal on the sensor. Rescanned this morning, 6.3. Fingerprick - 8.1?! after breakfast it told me I was 10.1. This is good again, I thought, as it's clearly responding to something and having a rise. Fingerpricked again and I was 12 - so whilst it rose by the correct ratio (+4 mmol), it is so out of whack I am wondering if it's worth it. Does anybody have any advice/tips? Could the heat be affecting it and should I just let it settle for the full 24 hours before complaining that it's not working properly? I understand there will be some inaccuracy and a lag but this just seems to far out to actually be of any use - considering contacting Abbott today.

 

[Robin]

Are you keeping well hydrated in the heat? Both times it’s been really hot this summer, my Libre has struggled if I’m a bit short on fluids.

Also, I have found that sometimes (but not always) the sensors take a day to fully 'bed in', I now allow at least 12 hours and put mine on first thing in the morning, if I'm going to be starting it in the evening.

Thirdly, because of the time lag in the sensor readings, behind a fingerprick, and the fact that the Libre has an algorithm built in to try and catch it up a bit, I sometimes find if I'm falling or rising quickly, that the Libre has overshot the true figure, and only sorts itself out when I look again in 15-20 minutes to see what the graph is actually telling me.

 

[Bruce Stephens]

Fingerpricked again and I was 12 - so whilst it rose by the correct ratio (+4 mmol), it is so out of whack I am wondering if it's worth it.

Remember that when you're outside the usual range (about 4-7) all our tests are less accurate. Also remember that the two are measuring different things: what Libre is measuring is about 15 minutes delayed from blood glucose (so measure with a test strip, wait 15 minutes, and then scan the Libre).

So you can really only compare the two (and get meaningful numbers) when your BG is pretty stable for a while. And Libre is using a prediction algorithm which tries to reduce the 15 minute delay, making it all more uncertain. But when you're fairly stable the algorithm shouldn't be doing much (another reason for waiting until the curve is roughly flat).

 

[helli]

@pawprint91 Libre can be frustrating but it is much less frustrating if you understand the limitations which, unfortunately, neither Abbott nor most DSNs bother telling us. I may forget some but here are the ones I remember
- Many of us find that our bodies take some time familiarising theirselves with an alien object inserted into our arm. This may affect the readings you get from Libre during the first 48 hours. For this reason, many of us insert the next Libre a couple of days before the current one runs out and only activate the new one once it has "bedded in"
- Libre is calibrated to be most accurate between about 4 and 9 mmol/l/ Outside of this range, it may less reliable. Therefore, unless you have a very very obvious hypo, it is not recommended to treat an alarm until you have double checked it with a finger prick reading.
- Libre can drift a little during the life of a sensor. Therefore, it is recommended to do a finger prick check once a day when your levels are flat(ish) and between 4 and 9.
- Libre is factory calibrated. Some of us seem to be different to "Factory Man" and need to be able to calibrate the sensor ourselves which is not possible with the LibreLink app. There are unofficial apps such as xDrip which use a slightly different algorithm to convert the Interstitial Fluid Reading to Blood Glucose. These apps allow calibration. Unfortunately, they are written and documented (where there is any documentation) by techies, If you are technically minded these are useful. If you are not technically minded, they can add to your frustration.
- Libre reads Interstitial Fluid not blood. These readings are about 15 minutes behind. The algorithm Libre uses attempts to overcome the delay by extrapolating the current trend. This works pretty well unless the trend changes direction in the last 15 minutes. This is especially obvious when you treat a hypo so don't only check that you are hypo with a finger prick when the alert goes off, also check your recovery with a finger prick and don't panic if Libre says you are going lower into hypo.
- Some sensors are faulty so you will need to contact Abbott in those scenarios.

 

[pawprint91]

Are you keeping well hydrated in the heat? Both times it’s been really hot this summer, my Libre has struggled if I’m a bit short on fluids.

Also, I have found that sometimes (but not always) the sensors take a day to fully 'bed in', I now allow at least 12 hours and put mine on first thing in the morning, if I'm going to be starting it in the evening.

Thirdly, because of the time lag in the sensor readings, behind a fingerprick, and the fact that the Libre has an algorithm built in to try and catch it up a bit, I sometimes find if I'm falling or rising quickly, that the Libre has overshot the true figure, and only sorts itself out when I look again in 15-20 minutes to see what the graph is actually telling me.

I'm putting this morning's fingerprick higher reading down to dehydration overnight. Thank you for the tip about 'bedding in', will remember this for next time and maybe give this one a bit of extra time to settle with it being so new.

Remember that when you're outside the usual range (about 4-7) all our tests are less accurate. Also remember that the two are measuring different things: what Libre is measuring is about 15 minutes delayed from blood glucose (so measure with a test strip, wait 15 minutes, and then scan the Libre).

So you can really only compare the two (and get meaningful numbers) when your BG is pretty stable for a while. And Libre is using a prediction algorithm which tries to reduce the 15 minute delay, making it all more uncertain. But when you're fairly stable the algorithm shouldn't be doing much (another reason for waiting until the curve is roughly flat).

Thank you - I met somebody yesterday on my course and have swapped numbers with her. Just been in touch with her and she says she only checks with a finger prick if she is low or high and if it's in range she just accepts it, so may go with this approach for now.

@pawprint91 Libre can be frustrating but it is much less frustrating if you understand the limitations which, unfortunately, neither Abbott nor most DSNs bother telling us. I may forget some but here are the ones I remember
- Many of us find that our bodies take some time familiarising theirselves with an alien object inserted into our arm. This may affect the readings you get from Libre during the first 48 hours. For this reason, many of us insert the next Libre a couple of days before the current one runs out and only activate the new one once it has "bedded in"
- Libre is calibrated to be most accurate between about 4 and 9 mmol/l/ Outside of this range, it may less reliable. Therefore, unless you have a very very obvious hypo, it is not recommended to treat an alarm until you have double checked it with a finger prick reading.
- Libre can drift a little during the life of a sensor. Therefore, it is recommended to do a finger prick check once a day when your levels are flat(ish) and between 4 and 9.
- Libre is factory calibrated. Some of us seem to be different to "Factory Man" and need to be able to calibrate the sensor ourselves which is not possible with the LibreLink app. There are unofficial apps such as xDrip which use a slightly different algorithm to convert the Interstitial Fluid Reading to Blood Glucose. These apps allow calibration. Unfortunately, they are written and documented (where there is any documentation) by techies, If you are technically minded these are useful. If you are not technically minded, they can add to your frustration.
- Libre reads Interstitial Fluid not blood. These readings are about 15 minutes behind. The algorithm Libre uses attempts to overcome the delay by extrapolating the current trend. This works pretty well unless the trend changes direction in the last 15 minutes. This is especially obvious when you treat a hypo so don't only check that you are hypo with a finger prick when the alert goes off, also check your recovery with a finger prick and don't panic if Libre says you are going lower into hypo.
- Some sensors are faulty so you will need to contact Abbott in those scenarios.

Thank you so much for this. Really good point about the alien object! Might have a look at xdrip as my partner is very tech minded, although I am not I definitely got confused yesterday when it said I was further into the hypo after I'd eaten something, so thank you for letting me know of the actual delay! As I said above, my friend advised me to only check if low or high so might go with that for the next couple of days and then see what's occuring.

Should I be concerned about the sensor losing signal alert? I still had readings overnight, so guessing it must have righted itself?

 

[Saoirse]

@pawprint91 Libre can be frustrating but it is much less frustrating if you understand the limitations which, unfortunately, neither Abbott nor most DSNs bother telling us. I may forget some but here are the ones I remember
- Many of us find that our bodies take some time familiarising theirselves with an alien object inserted into our arm. This may affect the readings you get from Libre during the first 48 hours. For this reason, many of us insert the next Libre a couple of days before the current one runs out and only activate the new one once it has "bedded in"
- Libre is calibrated to be most accurate between about 4 and 9 mmol/l/ Outside of this range, it may less reliable. Therefore, unless you have a very very obvious hypo, it is not recommended to treat an alarm until you have double checked it with a finger prick reading.
- Libre can drift a little during the life of a sensor. Therefore, it is recommended to do a finger prick check once a day when your levels are flat(ish) and between 4 and 9.
- Libre is factory calibrated. Some of us seem to be different to "Factory Man" and need to be able to calibrate the sensor ourselves which is not possible with the LibreLink app. There are unofficial apps such as xDrip which use a slightly different algorithm to convert the Interstitial Fluid Reading to Blood Glucose. These apps allow calibration. Unfortunately, they are written and documented (where there is any documentation) by techies, If you are technically minded these are useful. If you are not technically minded, they can add to your frustration.
- Libre reads Interstitial Fluid not blood. These readings are about 15 minutes behind. The algorithm Libre uses attempts to overcome the delay by extrapolating the current trend. This works pretty well unless the trend changes direction in the last 15 minutes. This is especially obvious when you treat a hypo so don't only check that you are hypo with a finger prick when the alert goes off, also check your recovery with a finger prick and don't panic if Libre says you are going lower into hypo.
- Some sensors are faulty so you will need to contact Abbott in those scenarios.

This is an excellent post which would be helpful to many people especially new to libre. Maybe the mods could pin it somewhere?

 

[PhoebeC]

Yeah everything that @helli said!

I don't put mine on early, I just accept that it may be a little out for a few days, but for me 2 is crazy out.

And always test lows and highs. Like you may get an alarm for a high or low and can test again in a few minutes with the libre and it will have been a random spike and be back into range. So finger pricking is still here as a back up but a few a week for me rather than 10s a day 🙂

Also you will know pretty quickly if not on right, I've only had 2 since March 2020 that haven't gone on right. Which isn't too bad.

 

[pawprint91]

Yeah everything that @helli said!

I don't put mine on early, I just accept that it may be a little out for a few days, but for me 2 is crazy out.

And always test lows and highs. Like you may get an alarm for a high or low and can test again in a few minutes with the libre and it will have been a random spike and be back into range. So finger pricking is still here as a back up but a few a week for me rather than 10s a day 🙂

Also you will know pretty quickly if not on right, I've only had 2 since March 2020 that haven't gone on right. Which isn't too bad.

I think mine is placed on correctly - I can't even feel it and it seems pretty secure! Thank you 🙂

 

[Finn]

My first sensor was activated right after applying and it was pretty frustrating getting wonky readings, but my diabetes team had warned me that the first 24 hours (if activated right after applying) needed to done in tandem with finger pricks.

Now, I just apply 24 hours before activating. I plan it so one sensor ends and the other is all ready to start straight away (applied 24 hours earlier). Inside the lid of the box, I write the time and date the sensor was applied, activated, and when it ended if it was faulty. I also store the pieces with the sensor number inside the box until I know it's lasted its lifespan (this is also on the sensor and in the app/reader, but easier to read from the foil covering the applicator). This helps if you have a faulty sensor and need to contact Abbot about it. If a sensor ends early, keep the faulty sensor so you can send it back to them if they ask for it.

The 15 minute delay took a bit for me to wrap my head around as it's not super straight forward:
Take a reading with a finger prick at 10am
Scan at 10:15 (to allow for the 15 minute delay)
Look at what the graph shows you were at 10am, not 10:15am - the graph will 'catch up' and correct the numbers it extrapolated

If you scan and look at the reading right now, that's a projection based on your trend from 15 minutes earlier, not a reading that's 15 minutes delayed. The points plotted in the graph earlier than 15 minutes ago should be accurate (as far as FSL is accurate).

As others have said, readings that are high or low are less accurate on FSL, so go with your gut or with finger pricks if you doubt your sensor at all.

I have my low alarm set to just above hypo overnight and a bit higher during the day (so I can treat with things that are slower, but tastier).
I don't often use the high alarm unless I want to know I'm in the 'safe zone' after a hypo - then I set it to the lowest high alarm setting.
I completely turn off the 'lost signal' alarm.

 

[pawprint91]

My first sensor was activated right after applying and it was pretty frustrating getting wonky readings, but my diabetes team had warned me that the first 24 hours (if activated right after applying) needed to done in tandem with finger pricks.

Now, I just apply 24 hours before activating. I plan it so one sensor ends and the other is all ready to start straight away (applied 24 hours earlier). Inside the lid of the box, I write the time and date the sensor was applied, activated, and when it ended if it was faulty. I also store the pieces with the sensor number inside the box until I know it's lasted its lifespan (this is also on the sensor and in the app/reader, but easier to read from the foil covering the applicator). This helps if you have a faulty sensor and need to contact Abbot about it. If a sensor ends early, keep the faulty sensor so you can send it back to them if they ask for it.

The 15 minute delay took a bit for me to wrap my head around as it's not super straight forward:
Take a reading with a finger prick at 10am
Scan at 10:15 (to allow for the 15 minute delay)
Look at what the graph shows you were at 10am, not 10:15am - the graph will 'catch up' and correct the numbers it extrapolated

If you scan and look at the reading right now, that's a projection based on your trend from 15 minutes earlier, not a reading that's 15 minutes delayed. The points plotted in the graph earlier than 15 minutes ago should be accurate (as far as FSL is accurate).

As others have said, readings that are high or low are less accurate on FSL, so go with your gut or with finger pricks if you doubt your sensor at all.

I have my low alarm set to just above hypo overnight and a bit higher during the day (so I can treat with things that are slower, but tastier).
I don't often use the high alarm unless I want to know I'm in the 'safe zone' after a hypo - then I set it to the lowest high alarm setting.
I completely turn off the 'lost signal' alarm.

Very helpful response, thank you. Does the lost signal alarm not mean alot?

 

[helli]

Very helpful response, thank you. Does the lost signal alarm not mean alot?

If you have lost signal, you will not get an alarm if your blood sugars get low (or high).
That sounds like a pretty useful thing to me.

However, the Bluetooth on some phones can be a bit temperamental which can result in dropped signal at times and too many alerts.

 

[Finn]

Very helpful response, thank you. Does the lost signal alarm not mean alot?

I personally don't find the lost signal alarm useful. I scan often enough and I can usually feel if I'm low, so I find the lost signal alarm doesn't give me any useful info. If I really need to know my level and the reader isn't scanning, I just do a finger prick.

 

[Inka]

I find some sensors read low the first day or two. My current one did that and, like yours, would alarm and tell me I was in the 3s when I was actually in the 6s. It took a couple of days to settle but is now much closer to my finger-pricks.

 

[pawprint91]

If you have lost signal, you will not get an alarm if your blood sugars get low (or high).
That sounds like a pretty useful thing to me.

However, the Bluetooth on some phones can be a bit temperamental which can result in dropped signal at times and too many alerts.

I have the alarms turned off at the moment anyway due to the figures being so off (hoping this is temporary). I think it was the bluetooth as I had about 5 notifications that it had lost signal in the space of 10 minutes, but then still had readings for overnight anyway.

 

[Bruce Stephens]

I have the alarms turned off at the moment anyway due to the figures being so off (hoping this is temporary).

Then you may as well turn off the alarm for losing connection (since alarms are the only thing the connection is used for).

but then still had readings for overnight anyway.

The sensor remembers 8 hours of readings, so as long as we scan at least every 8 hours we get a continuous record. (And that's nothing to do with the Bluetooth connection.)

 

[PhoebeC]

Yes I never use the loss of signal alarm. It was going off all the time and I just rescan if I know I have moved too far away.

 

[pawprint91]

I think my initial concern about being faulty was just a settling in period, it's been semi accurate (to within 1 m/mol) since about Wednesday lunchtime, felt so confident with it that I was out and about in London on Saturday and only checked with a fingerprick once on the train when it said I was at 4.3 (was 4.9).

However, I turned the alarms off on the day I got the sensor as it was being wildly inaccurate, and now it won't let me turn them back on? It either says 'alarms are only available when you scan a Freestyle Libre Sensor 2 with this device' or asks me to scan sensor, which I do, but then just get the same message. I can live without alarms for now, but should we be able to turn them on and off, or is it a once they're gone they're gone situation?

To those who apply the new sentence the day before the old one is done, which side do you then sleep on? Old sensor or new? Can we sleep on our sensors at all?

 

[Bruce Stephens]

I can live without alarms for now, but should we be able to turn them on and off, or is it a once they're gone they're gone situation?

No, you should be able to turn them on and off (and change the levels at which they alarm).

Can we sleep on our sensors at all?

I'm sure I do, without apparent harm. Probably depends on lots of factors (like how hard your mattress is). Worst case is that you get woken up by a compression low.

 

[Lynne888]

Just came here to post a similar thread and saw yours! I’ve started using Libre today. First reading was 5.7 but blood monitor was 6.8 and since then two further readings have been very different too. It will give me false hope that I am reversing diabetes with lower readings. I only got it because the finger pricking has become very painful, most likely because I’ve only been doing it about 10 weeks but I dread doing it atm they are so sore.

 

[Bruce Stephens]

First reading was 5.7 but blood monitor was 6.8

I wouldn't regard those as "very different", but maybe I'm just looking at the results from a different perspective.

From my perspective I'm just trying to stay within 3.9 and 10.0 most of the time, so the difference between 5.7 and 6.8 just doesn't make any difference in what I'm going to do. Now, if (as I suspect) the sensors tend to read a bit low then how much time I've spent really in range won't match what Libre shows, but I don't think I care that much. I'm pretty sure I saw a paper saying that such small differences don't matter that much and that just believing what the CGM says is useful.