CGM vs finger pricking: too much data?

[Timothy.M.Read]

I typically use a finger prick based approach to measuring my blood glucose level, before and two hours after each meal. These levels are within the range given to me by my doctor, so no worries.

Since this approach is a nuisance, I am trying a Freestyle Libre 3 CGM (continuous glucose monitor). It is very convenient, although very expensive, since my medical insurance does not cover it.

However, with all the data it provides, I am now more worried about my glucose levels than before. I see, quite often, that my glucose level 30 minutes after eating sometimes goes up to 200 mg/dL, and after an hour is sometimes between 140 mg/dL and 180 mg/dL (OK, according to my doctor but still higher than for non T2D people), although below 140 mg/dL after two hours.

So, if I am told to monitor my glucose levels TWO HOURS after eating (assuming a finger prick test), should I be worried about these "intermediate" results? Furthermore, what is a good glucose leave for T2D 30 minutes after eating.

I don't have a recent A1C result, which I know is the gold test, but the CGM does tell me that over the last 14 days my GMI (glucose management indicator) is 6.2%, 45 mmol/mol.

Thanks in advance.

TMR

 

[jjraak]

Hi @Timothy.M.Read

Yes, you do get a lot more info it seems,
But you are literally finger testing every 5 minutes (theoretically)

As for worrying...mmh.

If it helps, try to consider if you took your temperature using a similar technique to a CGM.

Would you worry if your temperature rose or fell ?

I doubt you would, if it wasn't sustained or unexpected

It would rise in relation to what you did & how your body reacted to it .

So.. sit still in the cold
V
Go jogging

And I'm sure we've all felt the flush of embarrassment, which must raises our temperature, no ?

My point being, yes our BG rises by what we put in our bodies, how we balance that intake with use of the carbs/glucose.

(And do bear in mind as a T2D, our response is slower/impaired/broken, so not matching non T2D, would be expected )

And a host of other day to day responses our bodies can have.

For example we respond to exercise, which can cause our BG to rise

And technically, any movement is exercise, we just grade it by how much effort we feel we are putting in ..aka gym v housework for instance

And when we get scared or stressed....

We get a higher BG level as we get a boost so we can fight or flee the danger.

We grade that 'danger' or level of stress, but simply mulling over being able to pay a bill or thinking over a witty response to what someone said to you earlier, MIGHT raise our levels.

All perfectly normal, if we return to a more normal level afterwards.

Which, I think will show in your trend of BG rises on your CGM.

Once you begin looking, I'm pretty sure you'll start being able to place any highs with a certain moment

So you'll begin to normalise how your body & it's glucose response to certain situations works for you, as an individual.

Something worth learning, I'd say

What you should be, I think concerning yourself with, is any extended spike that doesn't go back down or resolve.

The link to Jenny ruhls 'Diabetes 101' under my signature below goes into greater detail of her interpretation of why sustained high levels cause damage.

Does Type 2 Diabetes Always Get Worse?

Most dctors think all people with Type 2 develop complications but this research shows it isn't true.

www.bloodsugar101.com

Well worth a read, imho .

 

[jjraak]

 

[HSSS]

. I see, quite often, that my glucose level 30 minutes after eating sometimes goes up to 200 mg/dL, and after an hour is sometimes between 140 mg/dL and 180 mg/dL (OK, according to my doctor but still higher than for non T2D people), although below 140 mg/dL after two hours.

So, if I am told to monitor my glucose levels TWO HOURS after eating (assuming a finger prick test), should I be worried about these "intermediate" results? Furthermore, what is a good glucose leave for T2D 30 minutes after eating.

I don't have a recent A1C result, which I know is the gold test, but the CGM does tell me that over the last 14 days my GMI (glucose management indicator) is 6.2%, 45 mmol/mol.

Thanks in advance.

TMR

Everyone rises at 30mins and an hour - even no diabetics. The difference between normal levels and diabetics is how high and how long.

https://www.diabetes.co.uk/diabetes_care/blood-sugar-level-ranges.html This tells you after a minimum of 90 mins (we all seem to use 120 though) a non diabetic would usually be back to 7.8mmol (140mg). There are no figures for 30mins and one hour. So bearing in mind the timing differences are you actually higher? Perhaps a few meals still are a bit carb heavy for you if so. And remember even a non diabetic can be seen on a libre to have higher earlier spikes than 90mins. There’s some examples out on the net that have been linked to in here before.

And whilst that’s only an estimate hba1c it’s looking at non diabetic levels if you keep going the way you are. Well done.

 

[Timothy.M.Read]

Many thanks for the interesting and useful replies. I have some homework to do here

 

[Timothy.M.Read]

I'd like to ask a follow on question, with such expertise available.

I notice that the glucose level from a finger prick test is consistently lower than that of the CGM. My hands are always cold and I find it hard to get a blood drop out for the test and often squeeze the area. My blood is quite thick and "sticky", maybe I need to drink more water... Could these factors be affecting my results?

Thanks in advance

 

[jjraak]

I believe many struggle on occasions to get blood easily.

Remedies seem to be shake hands for a while before testing.
(Think downward facing jazz hands, I suspect )

Let a warm water tap run over them for a bit before testing

And yes, I think too little water can have an impact .

I'd suspect cold hands also has an impact on getting blood out.


On CGM difference, I assume you mean pinpricks lower even when lag factored in (Upto 15 mins I read ).

No idea, sorry.

 

[Timothy.M.Read]

Thanks jjraak. Yes, the finger prick values are always lower by around 10 than the CGM.

Your suggestions are very welcome. Drinking water is a real problem. I don't mind coffee or tea but unless I am really thirsty, I hate drinking water...

 

[KennyA]

Thanks jjraak. Yes, the finger prick values are always lower by around 10 than the CGM.

Your suggestions are very welcome. Drinking water is a real problem. I don't mind coffee or tea but unless I am really thirsty, I hate drinking water...

I tried a libre but it rarely matched the fingerprick, even allowing for the fact that it tests interstitial fluid rather than blood, and therefore has a "lag" time. The best I got was a libre reading that tracked the fingerprick well but gave a consistent reading around 1.5/2.0 mmol/l below.

On the other hand, even with the allowance for inaccuracy, it did give me useful info about what happened while I was sleeping. It was worth doing as a one off (in my opinion) but not something I'd pay for regularly.

Just a point: as I understand it, the two hour reading is not primarily intended to find out how high you've gone after eating. Everyone's BG rises after taking in carbs. A raeding at two hours is intended to assess whether one's levels have returned to an acceptable range by the two hour mark, and by extension how well your system is responding to your insulin.

 

[Timothy.M.Read]

Thanks KennyA. I am also using a libre3. These babies are expensive, so I don't want to use them on a permanent basis. My interest is getting feedback as I tweak my diet, does A affect my BS level? what about B?, etc. Once I have a stable diet setup, then I won't need to check my BS so often. Having to do a finger prick test during the day is fine, a little inconvenient, but no big deal. However, as you say, the night measures are a problem. I have dinner late, around 10pm, so want to measure BS at 00:00, at which point I'm usually asleep in front the TV. It takes superhuman willpower to do the finger prick test then...

Just a point: as I understand it, the two hour reading is not primarily intended to find out how high you've gone after eating. Everyone's BG rises after taking in carbs. A raeding at two hours is intended to assess whether one's levels have returned to an acceptable range by the two hour mark, and by extension how well your system is responding to your insulin.

Understood. I would have thought it would be important to be able to note that right after eating, say 30 minutes later, whether your BG only goes up to 180 or rather goes up to 250. Or is it the case that these really brief rises in BS aren't really important if after 2 hours the level is under 140? should I not worry then about these short-term values? Is that why we use A1C or GMI as a measure now?

Just a final aclaratory question: after a laboratory test, are finger prick test the gold standard for measuring BS levels? and a CGM a nice to have? Or are both quality ways of measuring BS? If, hypothetically speaking, I won the lottery tomorrow and could fill my kitchen cupboard up with libre 3s, could I then toss my finger prick test kit in the bin?

Thanks in advance...

 

[In Response]

I continue to see the reports that Libre is 15 minutes behind finger pricks.
This is not the case. Interstitial fluid readings are 15 minutes behind finger pricks but the algorithm Libre 2 uses to convert interstitial fluid readings into blood glucose takes into consideration the 15 minute lag.
So, the reading you get form Libre is not 15 minutes behind finger pricks.

The algorithm extrapolates the current trend and "predicts" the current reading. This can lead to a few oddities when our trend changes in the 15 minutes (e.g someone treating a hypo with fast acting carbs)
- Libre will take longer to notice the change of direction of the trend (so a hypo may appear to last longer on a Libre)
- Libre will over "predict" before the change of direction (so a low may appear lower than it actually goes)
- Libre will overwrite a prediction when it gets the later data (you may see a low value after a scan but it could disappear later)

The inaccuracies some people see are mostly due to the factory calibration. Our bodies do not all behave the same and many behave differently to the fictitious "factory man". This is not a "consistent 10 lower". We are different, sensors are different and accuracy is expressed as a percentage not as an absolute. This is why it is recommended to occasionally check Libre with finger pricks to understand how your current sensor varies for you.

 

[sw600]

^^^ I have just discovered this over the last couple of days, after noticing that sometimes (often during sharp transitions e.g. hard exercise) it gives you one number (and the little arrow tells you it's heading even further that way) then if you re-test 10 mins later it turns out that never happened If you click a logbook entry for that day you'll see the guesstimate that it made, sometimes the little green circle is quite a way off the 'actual' graph. Check out my pic of today attached for an example; I've been fasting today as I had an hba1c test. As far as absolute accuracy goes, this one I have right now seems to over-read compared to the last couple. For the test they gave me the actual glucose as 5.5 where the libre seems to think it's 6.5+.

 

[In Response]

I've been fasting today as I had an hba1c test.

Why are you fasting for HBA1C?
As it measures an approximation of the last 3 months, a few hours of fasting makes no difference.

If you had cholesterol tests at the same time, I believe it is sometimes suggested you fast for these.
But if it was just HBA1C, my understanding is that this is not a fasting blood test.

 

[plantae]

In hospital the nurse took my BSL using a finger prick test because she said it would be different to my libre. So I did the libre test and showed her immediately after her test. They were both 6.9. In reality most of the time my tests are a few decimal points different but nothing I'd consider major or concerning. If I'm high (rarely) or low (fairly often at the moment) I still follow up with a finger prick... not always when I'm high, but always when I'm low. The arrows (rate of change) on the libre I find less reliable. Clearly, based on reports, this varies from person to person but all of my Libre sensors except one (which Abbot replaced) have pretty much exactly matched my Countour Next results...

 

[sw600]

Why are you fasting for HBA1C?
As it measures an approximation of the last 3 months, a few hours of fasting makes no difference.

If you had cholesterol tests at the same time, I believe it is sometimes suggested you fast for these.
But if it was just HBA1C, my understanding is that this is not a fasting blood test.

I was fasting because triglyceride levels fluctuate after meals. I want to make a valid comparison with my regular tests, which have always been done fasted. Non-fasting triglycerides are also a valid measure, just a different one.

 

[In Response]

I was fasting because triglyceride levels fluctuate after meals. I want to make a valid comparison with my regular tests, which have always been done fasted. Non-fasting triglycerides are also a valid measure, just a different one.

So you were not fasting for the HBA1C but for the other tests done at the same time.
That makes sense. Thank you for the clarification.

 

[sw600]

Yes cholesterol also tends to be a bit high unfortunately. And last time I went to the docs they reckoned my blood pressure was also slightly high, which it never has been before, just to complete the holy trinity.

 

[sw600]

But back to the OP's question, I don't think you can have too much data. If I only had my 3-6 monthly hba1c to look at then I have been between 40-44 for about 6 years depending on how well I apply myself to eating properly (and it's steadily getting more difficult to keep it low I think). But looking at the CGM shows quite large spikes from exercise and eating, with the rest of it relatively low. There seems to be a growing body of evidence to suggest that this variability (big spikes) creates more oxidative stress. So inside your single hba1c number there is a wealth of information about how you could better deal with your condition. I could just say hba1c 41 (which I just got yesterday) is 'normal', and ignore any carb restriction. But I can see that I get readings of 13-14 if I am not careful (moreso from exercise weirdly) which is definitely not 'normal'. If you only do a finger prick test 2 hours after a meal and never see the max values of the actual spike you are still losing information that you could use to your advantage.

edited to add: hba1c 41 from the blood test (let's call this the 'actual' value) was kind of halfway between the libre's optimistic "Estimated A1C" value of 38 and the pessimistic GMI value (in their online reporting tool) of 43. So, not sure what to make of either of those estimated numbers. Let's wait 3 months and I'll test again, if I can stomach the £100 per month for the libre sensors.