Variation in levels on same meter

[bizzilizzi]

Hi,

So I'm a parent of a wonderfully resilient 8-year-old girl diagnosed six-months ago and several times we've had significant differences in blood glucose results taken within just a minute or so of each other. (Definitely nothing to do with sugary fingers!)

I've been reading up about the variation that you can get with meters and it's a bit terrifying to be honest. And the small print that says "system accuracy results for glucose concentrations ≥5.55mmol/L within +/- 5%: 64.9% accurate" etc. So, correct me if I'm wrong, but that basically means that for levels above 5.5 the chance that it is out by more than 5% is 35.1%? From what I understand, the higher the level the more potential variation there is in the test result also?

So is it normal/acceptable for a meter to read 16.7 followed by 11.4 a minute later? (The meter is the Freestyle Optium Neo).

Bearing in mind that she has a Libre, which is what we usually use, except when we want to double-check (because we're told "the meter is more accurate"). This time the Libre read 14.8 dropping sharply, and I recall the good doctor told us that the arrow pointing straight down means the BG is dropping 1mmol/L per 5 minutes. All this was checked within the space of 2-3 minutes.

Another question: we were also told not to use the Libre for finger-prick tests but not why, and from my reading of the leaflet provided within the strips box the two meters (Optium Neo and Libre) have exactly the same level of accuracy for this.

So it seems to me a bit pointless to be trusting the meter to give us a more accurate reading than the Libre (and yes, I understand the time-lag between blood and interstitial fluid). Is there anything we can trust or is it just always going to be feeling our way in the dark like this?!

Thanks for any advice!

Lizzi

PS. the previous two times we'e had significantly different readings were 1.8 (obviously inaccurate) followed by 4.something and 10 versus 6.

 

[urbanracer]

Hi @bizzilizzi

The Freestyle Libre is great for trending but less so for absolute blood glucose. Most of us seem to find that it's reasonably accurate between say 4 to 9 mmol but outside of this range the magnitude of it's errors increase the further away from 'normal' one gets.

I don't know what meter you are using but blood test meters (sold in the UK) must comply with the requirement to be accurate within +/- 15% for 95% of readings.

https://www.diabetes.co.uk/blood-glucose-meters/blood-glucose-meter-accuracy.html

As I understand the technology, the biggest problem isn't the meter itself, but the enzymes in the test strips which can be influenced by temperature and air time among other things.

You then have to consider other factors like levels rising or falling sharply, the fact that your glucose levels are constantly changing and that your blood is not completely homogenous. The glucose in blood near your digestive organs starts to rise before the blood in your finger tips, and then has to be pumped round everywhere else.

You can get special solutions from the manufacture (usually FOC) to test your meter with. Also worth trying a different pot of test strips.

For what it's worth I had a similar experience at lunchtime today with 6 readings in the range 4 to 7.9 mmol in a couple of minutes, from 1 finger prick.

Normally you should get better results than what you're reporting above.

Edited to add, if you type libre in the search bar top right, you will find many threads on the subject.

 

[bizzilizzi]

Thanks for the reply, and for the advice to try a different box of strips / control solution, both of which we'll do.

 

[searley]

They are only home testing equipment

The standard for reqular blood glucose meters I'd that 95% of reading for more than 5.5 should be within 15% accuracy of what the lab result would be.. for less than 5.5 the results should be within 0.83mmol/l

This is more than accurate enough to have a fair idea of what the blood glucose is doing

Face it it doesn't matter if it reads 17 or 22 you are too high you need to bring it down

Between 4.5 and 6.2 you a pretty much good and no action need taking

3.9 or lower then you probably need to raise it a bit

But it's not worth worrying to much about the exact reading as this is not possible on home equipment to get lab results

What we have today is far better than what we had 10 years ago

If you get a result that is massively different than expected test again before treating

 

[searley]

Ps.. you should generally make treatment decisions from a proper finger test as this is more precise as the glucose level change faster at the fingers in the blood than anywhere else

The libre doesn't test the blood it testing the fluid under the skin

Libre are not considered accurate enough for HGV drivers to use