FREESTYLE LIBRE ON SALE!!!!

[PaulinaB]

I aim for non diabetic levels, have good hypo awareness, use a cgm on a constant basis (and have accurate readings). I already suffer from complications caused by late diagnosis and running my bg in a zone that makes my hospital team "comfortable" causes further issues with my health. My a1c is in the 5% zone, I feel great, my neuropathy and myopathy have improved,I did not have a serious hypo since I.took this tight control and haven't been close. I aim for bg level of 5mmol/l. I probably wouldn't be comfortable with that aim without a cgm, that's why I'm very happy to have one.
I'm looking forward to someone finding some kind of a cure, and I want my body to be in the best possible shape when this happens.

I can understand that many diabetic teams are not comfortable with people having tight control because a)many people don't actually take good care of themselves (where they achieve low levels by constantly running hypo) B)my nurse actually knows nothing about maintaining good every day control.
In the meantime I won't put up with myopathic pain to make a HCP, who I see twice a year, happy.

 

[smidge]

The irony of all this is that I listened to a consultant who has never had diabetes and a DSN who lacks the education/knowledge/intelligence to interpret the BG readings I try to show her - and by their own admission they never see any other diabetic with HbA1c as low as mine, so they clearly haven't an ounce of experience to share with me, but their constant nagging, whining and haranguing was too much, so I thought 'if I just do what they want for a few months, we'll all see it isn't working and I can try to focus them on the real issue - my basal insulin'. Well, it clearly hasn't worked because my levels are now all over the place and it's proving really hard to re-establish the control I had prior to this. In fact, without the timely intervention of the Libre, I think I'd be sunk! Heaven knows why I listened to these people!

Incidentally, an HbA1c isn't really an average. Hypos don't give us minus points. High BGs mean that extra glucose sticks to the red blood cells, but low BGs doesn't actually take any off - it just doesn't put any more on - if you see what I mean - so there is very little logic in equating a good HbA1c with a constant stream of hypos. It doesn't work like that! Just thought I'd clear that up as it always bugs me when people make the assumption that if you have a good HbA1c but you are having horrible spikes, you must be having hypos to compensate and bring down your average. It just doesn't work like that!!! OK, I feel better now LOL!

Smidge

 

[Hill28]

Have to add to my initial review, that the Test results of the Freestyle libre turn out to be very inaccurate, after I started double checking the results with traditional tests.

For example, this morning the Libre indicated 5,3 -> while te real value was 8.0. After an hour checked again to see if there was a delay. Libre indicated 5,8 -> while real value had risen til 9.0. So both the trend and the slightly to high value weren't indicated.

The Libre also indicated that sugar had been almost a flat line during the night, so it can not be a delay causing difference.

The sensor seems to be fine with 7 days to go and uncomfort getting less after some days. With differences up to 3mmol I will wait to buy a new sensor, a pitty..

 

[tim2000s]

@smidge, I think the issue is that there is a "perfect diabetic" level that consultants subscribe to, and this view has it that the characteristic of a diabetic is to have an Hba1c that is above a non-diabetic person. As you say, they view it that if you are close to normal as a diabetic then you must be having hypos as (and remember this is set in stone) diabetics are completely unable to control their blood sugar to that level of detail.

Now I'm not arguing that this approach is correct but it does appear that in recent years, many people have striven for better control as a result of the view on complications and the availability of better technology.
We all know how long it takes to change the medical profession's collective mind.

 

[LucySW]

The Christiansen study was just for 4 days, only 2 in normal 'daily life' It had fairly young subjects. Average age 27., the research from it's date and the description and pictures of the CGM used probably also used one of the earlier models.

The ADAG study had far more participants (80) with an average age of 41 and a wider variety of ethnic groups. It took place at least a couple of years later so presumably used later CGM models (might be relevant they have gradually got better)

It used CGMS for 48 hours at a time, repeating the monitoring on at least three occasions over a period of time with a median usage of 9.5 days
. During continuous monitoring all the subjects went above 6.1mmol/l. The median time above 6.1mmol/l was 395 min; 99% of the people went above 7mmol/l for a median time of 109 min.
There was though a big variation in time above each level . The range for the middle 50% of subjects (IQR) was 273-688 minutes of the day spent above 6.1mmol/l. ( ie non diabetics, in this study spent anything from 4.5 hours to 11.6 hours above this level) Most of the subjects (99%) spent longer above 7mmol/l than Christiansen's subjects spent between 6.6 and 7.7

As I'm 20 years older than the average in the ADAG study, that seems to have more relevance to me than a group of much younger subjects

Re driving , I've just noticed that the French equivalent of DUK now states you shouldn't take the wheel at less than 1g/l which is 5.5mmol/l.

Very interesting Phoenix, I'll read. Don't think age of CGMs is serious tho because they did just what we still need to do with the Libre: they made a spline curve around the actual BG measurements, just fitting the shape of the trend lines to those. The anchors for the curve were BG not CGM.

 

[Idan6a]

Have to add to my initial review, that the Test results of the Freestyle libre turn out to be very inaccurate, after I started double checking the results with traditional tests.

For example, this morning the Libre indicated 5,3 -> while te real value was 8.0. After an hour checked again to see if there was a delay. Libre indicated 5,8 -> while real value had risen til 9.0. So both the trend and the slightly to high value weren't indicated.

The Libre also indicated that sugar had been almost a flat line during the night, so it can not be a delay causing difference.

The sensor seems to be fine with 7 days to go and uncomfort getting less after some days. With differences up to 3mmol I will wait to buy a new sensor, a pitty..

sorry to hear,
although my third sensor shows a 1-2 mmol differences it is very consistent, that means the trend is very much real.

 

[tim2000s]

Following up on a comment in another topic, are we defining "Accuracy" of the Libre incorrectly? There have been numerous complaints that the values that the sensor shows are not the same as those of a fingerprick test. We know that the Libre tests something different from a blood glucose test.

Are we trying to define the accuracy of the sensor on the Libre in a frame of reference that is incorrect? Should we really be reviewing it as an Interstitial fluid glucose (IFG) tester and not a blood glucose (BG) tester? Granted there may be differences with specific sensors, but that is true of any system reliant on an implanted sensor. In addition, if we are measuring IFG, does this also mean that body make up will have an effect on the IFG levels observed, e.g. someone with a higher body fat level may see a difference in IFG due to the way that fat levels dilute or affect IFG?

With this in mind, is it that the IFG results are in fact accurate, and it is our frame of reference that isn't as we simply don't have experience of testing anything other than bloods?

The reason I ask this is in relation to hypos and coming out of them. I'm sure that other T1s have experienced the BG level showing that you are no longer in hypo but the feelings you have are still as though you are. Given the separation in values that we see in the Libre, is it not possible that this is really the case of how our bodies work?

Maybe the introduction of systems like this may require that we re-appraise precisely when a hypo is "over". Maybe it isn't when the BG has been raised at all. Maybe it's actually when that has propagated to other aspects of our anatomy?

This is just a hypothesis, but I'd be interested to know what you think...

 

[LucySW]

Should we really be reviewing it as an Interstitial fluid glucose (IFG) tester and not a blood glucose (BG) tester?

I think this is the only way that sanity lies, Tim. I've given up re-translating Libre readings back and forth. If we don't choose to work with the contour the Libre shows, it's not useful, methinks.

The problem with the current release of the Libre, though, is that there is a lot of error. Like the first or second overnight of each sensor, when something clearly isn't working because you get rows and rows of 2.2s.bAnd that something that isn't working presumably is something technical - which Abbott ought to work on some more.

 

[DunePlodder]

Hi Tim
I've had similar thoughts.
As we went from urine tests to blood on coloured BM sticks, to electronic blood testing we older ones changed how we used the readings.
If we weren't used to testing with a finger prick but had only ever tested IF using a CGM, then our user definition (as opposed to clinical definition) of hypos and what is low would perhaps be different.

For instance I treat 4.5 with a descending graph as a low or even as a hypo - with the lag I could be below 4. (Usually subject to a confirming blood test, but not always.)

As for comparing readings, then try doing the same thing with 2 blood test meters & you would see what on the face of it seem like large differences that are actually within the quoted precision of the meter. As Lucy says that way madness lies.

The obviously bad 2.2s Lucy & several other people have mentioned are a different matter.

 

[tim2000s]

I've had a few overnight 2.2 flat recordings, and if I'm honest, the way I've felt when I woke up I would guarantee they were real hypos, if not necessarily at 2.2 when compared to bloods (there's a certain hungover feeling that's quite unpleasant when I've spent a few hours overnight below 3.5).

I haven't yet seen something I would disagree with on the overnights. That's not to say it doesn't happen! I think this is one of the questions that we need to try and get a better answer for. Is it actually possible that when you are sleeping, your IFG level does fall simply because there is nothing for your musculature to be doing? I don't know the answer to that.

 

[LucySW]

In the circs you describe Tim that might well be worth seriously thinking about.

 

[Charlesa8]

I tried to order one of these today from Abbott but was told I had to wait until early 2015! Does anyone know another way to order?

 

[smidge]

WRT the accuracy or otherwise of the Libre, I think there is a large element of it simply measuring something different as Tim and DunePlodder say, but when you get a bad sensor you will see that the readings are completely different and completely random. My third sensor has been spot on, the first two were random. My third sensor expires tomorrow and I'm really upset about it. I can rely on this sensor. If this had behaved like my first two, I would probably be seeing the next sensor as my last for some time and bringing my Libre experiment to a close, but having seen the difference a really accurate sensor is making to me, I'd be lost without it!

I was wearing the Medtronic sensor for a week last week and they have now sent me the data through, but it's on paper rather than electronic and very hard to match against the exact Libre readings for the same time period because of the different parameters they used. For example, they have set the whole period by meal time rather than time and as my mealtimes are pretty moveable, that isn't helpful in terms of comparison. However, there is a note on the report to say that the calibration readings with the Accuchek Mobile are accurate/within tolerance. When I see them next week, I'll try to get an electronic dataset so that I can manipulate it to do a proper comparison.

If anyone is feeling wealthy, it would be a really interesting experiment to have two Libre sensors (one on each arm!) and two readers (one for each sensor). If Tim's hypothesis is correct and the differences some of us are seeing are really just IFG v BG plus time lag, these should give identical results both on the scan and the continuous data shouldn't they? I'm not volunteering though LOL!

Smidge

 

[Idan6a]

WRT the accuracy or otherwise of the Libre, I think there is a large element of it simply measuring something different as Tim and DunePlodder say, but when you get a bad sensor you will see that the readings are completely different and completely random. My third sensor has been spot on, the first two were random. My third sensor expires tomorrow and I'm really upset about it. I can rely on this sensor. If this had behaved like my first two, I would probably be seeing the next sensor as my last for some time and bringing my Libre experiment to a close, but having seen the difference a really accurate sensor is making to me, I'd be lost without it!

I was wearing the Medtronic sensor for a week last week and they have now sent me the data through, but it's on paper rather than electronic and very hard to match against the exact Libre readings for the same time period because of the different parameters they used. For example, they have set the whole period by meal time rather than time and as my mealtimes are pretty moveable, that isn't helpful in terms of comparison. However, there is a note on the report to say that the calibration readings with the Accuchek Mobile are accurate/within tolerance. When I see them next week, I'll try to get an electronic dataset so that I can manipulate it to do a proper comparison.

If anyone is feeling wealthy, it would be a really interesting experiment to have two Libre sensors (one on each arm!) and two readers (one for each sensor). If Tim's hypothesis is correct and the differences some of us are seeing are really just IFG v BG plus time lag, these should give identical results both on the scan and the continuous data shouldn't they? I'm not volunteering though LOL!

Smidge

Smidge I'm with you,
no one should expect identical measurement from Libre or any CGM but it seems there are bad sensors and much better ones.
which is not acceptable in terms of manufacturing,

about you experiment- I'm in !! we just need to find an investor

 

[Idan6a]

I tried to order one of these today from Abbott but was told I had to wait until early 2015! Does anyone know another way to order?

can you buy it from anywhere else in Europe and make sure it's sent to you?

i logged in to Libre Spanish site (freestylelibre.es) and ordered it from there (using Holla IP blocking service its free).

 

[Charlesa8]

can you buy it from anywhere else in Europe and make sure it's sent to you?

i logged in to Libre Spanish site (freestylelibre.es) and ordered it from there (using Holla IP blocking service its free).

OK thanks for the tip. I just tried the Spanish site and Dutch site but get the same message saying there won't have any more Libre's until 2015.

 

[Emmotha]

Lol @smidge as you and I both have trouble remembering which arm to scan with one sensor on I don't think we'd be good candidates for that experiment!

 

[Hill28]

I think the comments of tim2000s are philosophical. If the Libre is measuring something which doesn't actually reflect the thing that is having effect on your body, it is useles.

I think we can all agree that the bloodsugar value is the best indication of what your body needs or feels like. So alternative systems like the Libre should try to approach those bloodsugar values.

Furthermore if a 5mmol on the Libre sometimes means a 6.0 flat blood sugar and on other times a 8+ blood Sugar, it's useles to act on. The use for me is at the moment to small and I think I relied on the sensor for some days unjustified, because when I started checking the differences are just too big. I really hope they come up with accuracy improvements and hope they are already working on that..

That might be the reason no new customers are accepted.

 

[jddukes]

I am coming to the end of my 14 days of the first sensor. Here are some thoughts:

Insertion
Painless, did not feel it stuck on very well. Put at bottom of upper arm (when arm outstretched palm facing upwards) half way between armpit and elbow. Used physiotape over top to hold in place (with hole cut in middle so most of sensor exposed) which I replaced after about d10 as this was coming off in shower. Suspect the sensor would have come off had I not used this tape. No pain at all after first few hours of insertion.

Reading (NFC)
Bit disappointed here. First time trying nothing happened so was worried. Now I know that very often it will not transmit the first time (when you press the home button and it says scan). Most of time have to press the button again, then select scan glucose and then it works. Sometimes it does not work 3 times in a row and it times out. Bit tempermental and no obvious reason why (so sometimes I look like a lemon trying to scan it). Not sure if this is the sensor or the unit.

Settling period
Inserted it around 12pm one day, did not settle for a good 12-18h and read low. From <2.2mmol/l up to around 3.8 when bloods were saying 4.6-7mmol/l. However the following morning it was very accurate. Example readings:
18hr - Sensor = 6.3mmol/l, blood = 6.4mmol/l
24hr - S = 5.3, B = 5.5
30h S=4.4, B = 4.8
Etc.
Usual variability is 0.1-0.8mmol/l out from blood strip test, accounting for 15-20min lag. Only once or twice outside of 0.8mmol/l out, and seemed to be around hot shower/laying heavily on it, or when BG rapidly changing (apparently). On average about 0.3mmol/l out from blood.

Data generated
Biggest selling point. This is huge - and the reports are excellent, very detailed, very satisfying to a scientist! If I wasn't so busy right now I would be going mad with it all.

What it has helped me with
For this trial period I have stuck quite rigidly with my LCHF diet. This has really shown the benefits of this diet in keeping stable sugar levels. It has allowed me to observe the effect of exercise quite precisely and the timings around that (as I lift weights and then run every weekday morning) and shows that when my BS are out of range, it is due to exercise (on this diet) initially, with weights liberating the largest increase in BS. It has also helped me to answer the questions around certain foods and how I react - in particular I really wanted to know the true effect of Xylitol on my BS and this has been as I had hoped it would be, which was nice to see. It has confirmed that this diet is by far, the best for my control as I have not gone about around 9.2mmol/l in 2 weeks, and been in target (3.9-8.2mmol/l) for 91% of the time, with most of the 4% below coming from inaccurate readings early on due to settling (so probably in target 94%). I did not go above 10.0 on this diet during the period. It is pertinent to note that on a high carb diet with more insulin I regularly would find myself around 16.0 and sometimes in the 20s, at least 1-2x a week.

Here are my daily patterns:

J

 

[JANROU]

Regardless of all the comments I wish FC could put one of these in my stocking, so jealous of all those users.