My (rubbish) T2 Diagnosis

[rebrascora]

That will be a rogue reading. Some meters, particularly the SinoCare and I think the Kinetic give false low readings if you don't get enough blood on the test strip. Plus 5% of the test strips can be rogue. The important thing is to retest straight away if you get a result which doesn't match with how you feel and you would feel a 2.5 that is for sure. Most likely you would be feeling like you were about to pass out.

 

[Eddy Edson]

Contrary to what the diabetic nurse said, now the practice is refusing to give me any more strips and lancets, on the basis that I am not on enough medications. I understand the reasoning; the NHS view is that blood glucose monitoring is really for people at risk of hypoglycemia.

They're not getting the reader back. I'll say I splashed blood on it if they do. Using a calibration solution, its good enough in terms of reproducibility. The supplier offers VAT free refills, but an outrageous delivery charge. The same consumables, with free delivery, are available, with VAT, on Amazon, for less tha the delivered VAT-free cost.

While HbAC1 can be regarded as a 3 month rolling average of blood glucose, I know its is not a simple and acuurate relationship between mmol/mol and mmol/L. Nevertheless, I think some sort of regular monitoring would be helpful to me.

Severe lack of info what I'm supposed to do with my BP log. I've mentioned that I've never been able to reproduce the single measure that the GP made that has put me on medication

Single BP measurements at a GP visit are "performance not nedicine" (in the words of my cardio). Your log is the meaningful thing.

(But I'd also advocate investigating European and US guidance on appropriate BP targets - generally lower than in the UK, for good reason IMO. )

 

[Bruce Stephens]

Single BP measurements at a GP visit are "performance not nedicine" (in the words of my cardio). Your log is the meaningful thing.

There was that recent guidance on how BP measurements should be made to be accurate (from memory): with the arm relaxed on a table at the same height as the heart, take measurements on both arms, rest for 5 minutes with no talking before taking them, etc. None of those seem likely to happen in a GP surgery.

 

[everydayupsanddowns]

My blood glucose levels seem to have plateaued, at about 5.7-6.1mmol/l (either early morning, or a few hours after lunch). My blood pressure is basically all within normal range throughout the day. I don't know what is the expectation of the Metformin.

Sounds like it’s working really well for you @saz9961

Metformin seems to be one of the meds that gives a general helping hand ‘in the background’, alongside diet and lifestyle changes.

It’s a common first-line med, it helps increase insulin sensitivity, and reduces glucose output from the liver - in a sense it doesn’t actively reduce glucose on a meal-by-meal basis (eg by increasing insulin output), so much as helping the body use existing insulin production more effectively.

 

[Eddy Edson]

I've been tracking my blood glucose. Since 29-10, I've been earnestly cutting down on carbohydrates, and calorific intake, about 1000-1100 kcal per day. I don't feel tired. Today I went for a 4km walk.

Since the 29-11, my blood glucose, at least measured by the NHS issue meter, have been consistantly falling. I've usually done a first thing in the morning measure, and then mid afternoon, a few hours after a light lunch. Evening dinner is usually 400-500kcal.

Tonight, I measured and it was 4.6mmol/L. I repeated, and it came back at 5.3. But if the first number is correct, is that too low. I've been on Metformin for about 3 weeks now, and about a week on 1000mg per day (1 in morning, 1 in evening).


View attachment 32350

4.6 is just fine. It's a "normal", non-diabetic read.

Really good looking numbers!

 

[rebrascora]

4.6 is a really great number. Nothing to be worried about. That is where your levels should be. Congratulations and wishing you many more like that.

 

[Leadinglights]

The finger prick test blood glucose ranges people with a Type 2 diabetic diagnosis will be aiming at are 4-7mmol/l fasting/morning and before meals and no more than 8-8.5mmol/l 2 hours after meals. They will do a strategic before and 2 hours after meals to determine what meals are suitable and to be able to make adjustments to their meals. This gives more useful information than random testing.

 

[Docb]

@saz9961 - re your consecutive readings of 4.6 and 5.3. Some time ago I did 10 consecutive readings to get some idea of the reproducibility of blood glucose measurements. You can see the results with some discussion here:-

Thread 'What's in a number?'

Mar 8, 2021

We have a lot of members, particularly those new to testing, asking about blood test results and what they mean or what is a "good" number. As many of you are aware I like numbers, mostly because they are what they are and do not have any opinion.

This morning, I woke up, got out of bed, did a finger prick and got a 4.3. Checked another finger and got a 4.2. Tried a third finger and got a 5.1. This led to me doing something I have wanted to do for a while and this was to take 10 consecutive tests on different fingers and both thumbs to see what the variation was. The next seven...

• Docb

• 

The upshot is that the reproducibility of finger prick testing is such that your two readings are not statistically different. My own view is that readings should only be considered different if when rounded to the nearest whole number (the number after the decimal point is illusory) they are two or more units in difference. I would report both your results as 5 and not consider that they are different and the sort of reading a non-diabetic would get.

If you are interested in this stuff, have you tried testing before and about 1 hour after eating your most carby meal? I would expect you to then to see a change with before readings of 5 and 1 hr readings of anything up to 9 or 10. That would be a statistically significant difference.

I would also add that readings around 5 are not consistent with an HbA1c of 88. This suggests that whatever you have been doing has been effective in getting your blood glucose levels down. Another HbA1c test would confirm that.

Finger prick testers are amazing bits of kit but an appreciation of the boundaries is needed to avoid reading more into the results than is valid.

 

[Bruce Stephens]

I realise this is a trivial point unrelated to the bulk of the comment.

disregard first drop of blood

Do any test strip manufacturers actually advise that? I've occasionally seen people write that they do that as an alternative to washing (which sounds plausible to me) but I've not seen any other good reason why it would be worthwhile. I can imagine why it might be, I guess. Maybe the first drop gets interstitial fluid in it and that contaminates it? Whenever I've tried to look up research the results have always been it's harmless but also pointless and they've not been able to find why anyone's recommended it (though people obviously have).

 

[Eddy Edson]

What is being measured is the oxidoreductase activity of the immobilised glucose oxidase on the test strip. Likely the glucose oxidase on any test strip sold here has come from British Biocell (BBI), who also make nearly all of the gold nanoparticles used in the COVID-19 rapid tests many have used in recent years. Its manufactured by fermentation of Aspergillus niger (usually). The batches made are huge, and possibly a single batch might be enough for the product lifespan of the associated reader. So batch variations are unlikely. Some strip degradation might be expected; the strips are sensitive to moisture, though coatings are applied to minimise this. Hence don't drop all the strips onto the kitchen floor, and pick them up, keep the top fitted, store in a cool, dry area.

On a tangent, I imagine you have some good insight into the economics of test strips? Mature technology, high volume, visible easily addressed market, low marginal production costs => you'd think that retail prices for strips would be very low if the market were competitive? Or no? Is the R&D & process & production investment big & ongoing enough that you still need high prices even decades after launch? Or is it more a case of unfortunate consequences of a regulated market?

 

[Docb]

Thanks for those insights @saz9961. I have often wondered about the science behind finger prick testing and your description goes a long way to satisfying that.

In terms of random errors I have also tended to think about sampling errors. A tiny drop of blood is taken from a large dynamic system and it is assumed that the value obtained from the drop represents the whole. Do you think that is reasonable and do you think that error is smaller or larger than the system measurement errors you have described.

 

[PerSpinasAdAstra]

95% of results have to be within 0.83mmol/l for measurements below 5.5mmol/l and 15% for measurements above 5.5mmol/l.

So, not surprisingly, less accurate at low concentrations. For all intents and purposes, the readings might as well be +/- 1mmol/l. Round up/down

So @ a recorded 5.0mmol/l, the actual blood glucose might be 4.2-5.8mmol/l. At 7mmol/l, it might be 5.95-8.05mmol/l, but in practice, that inaccuracy will reduce with higher concentrations. 115% of 5.5 is 6.3mmol/mol.

It's tempting to think of meter accuracy this way but it's not how it works in practise. In order to manufacture a cheap test strip that meets the basic standard the manufacturer will effectively choose a blood glucose concentration 'sweet spot' where the test strip will be closest to optimum objective accuracy and the meter will be less accurate both above and below that mark. They may also invest a little more or a little less in the quality and consistency in the manufacturing process to help make outlier readings less likely. The standards and the reality of creating and selling tests strips don't quite match up. Some meters exceed the basic standards considerably and the manufacturers charge a premium for the test strips, while many other meters have failed to meet the basic standard in independent testing.

The readings from a good meter will not be plus or minus 1 mmol/L in practical terms, they'll be plus or they'll be minus at a given concentration at least 95% of the time, not round up or round down - one or the other. The good meters will be within 15% of objective accuracy 100% of the time, within 10% close to 100% of the time, and often much closer to the mark than that almost all the time depending on the blood glucose concentration of the sample. There will be rogue readings, outliers, but they're the exception not the rule. You can't trust a meter to be objectively accurate but many of them are very consistent. Consistently inaccurate, with a bias. Testing before and two hours after meals, with a good meter, the difference between readings will not jump around by plus or minus 1 mmol/L but almost always be either plus or minus and the variation in the gap between average bias and objective accuracy will be more like 0 to 0.3 mmol/L the great majority of the time.

This is what it looks like when you test the same drop of blood with four different meters. I repeated that four times and the graph on the left is the from a CGM. It's temping to look at the standard and imagine great variation from test to test, but that's not how it actually works. The green bar is where I believe objective accuracy lies on the last test, though I'm still trying to figure that out.

Edit - I just read your previous posts and I see that you likely know much more about these devices than I do, so I assume I've misunderstood your meaning in the post above where you're rounding both readings of 4.5 and 5.4 to 5? It's not impossible for both of those readings to be so far off the mark of course but it would be two big consecutive outliers in 'opposite directions' with both of those readings being so close to 5. Not likely. Even the cheapest meter, the Navii on the far left in the bottom image is generally a lot better than that.

 

[Eddy Edson]

Testing before and two hours after meals, with a good meter, the difference between readings will not jump around by plus or minus 1 mmol/L but almost always be either plus or minus and the variation in the gap between average bias and objective accuracy will be more like 0 to 0.3 mmol/L the great majority of the time.

That was my experience when I was testing *a lot* back in the early days & trying to get a handle on things.

 

[saz9961]

It's tempting to think of meter accuracy this way but it's not how it works in practise. In order to manufacture a cheap test strip that meets the basic standard the manufacturer will effectively choose a blood glucose concentration 'sweet spot' where the test strip will be closest to optimum objective accuracy and the meter will be less accurate both above and below that mark. They may also invest a little more or a little less in the quality and consistency in the manufacturing process to help make outlier readings less likely. The standards and the reality of creating and selling tests strips don't quite match up. Some meters exceed the basic standards considerably and the manufacturers charge a premium for the test strips, while many other meters have failed to meet the basic standard in independent testing.

The readings from a good meter will not be plus or minus 1 mmol/L in practical terms, they'll be plus or they'll be minus at a given concentration at least 95% of the time, not round up or round down - one or the other. The good meters will be within 15% of objective accuracy 100% of the time, within 10% close to 100% of the time, and often much closer to the mark than that almost all the time depending on the blood glucose concentration of the sample. There will be rogue readings, outliers, but they're the exception not the rule. You can't trust a meter to be objectively accurate but many of them are very consistent. Consistently inaccurate, with a bias. Testing before and two hours after meals, with a good meter, the difference between readings will not jump around by plus or minus 1 mmol/L but almost always be either plus or minus and the variation in the gap between average bias and objective accuracy will be more like 0 to 0.3 mmol/L the great majority of the time.

This is what it looks like when you test the same drop of blood with four different meters. I repeated that four times and the graph on the left is the from a CGM. It's temping to look at the standard and imagine great variation from test to test, but that's not how it actually works. The green bar is where I believe objective accuracy lies on the last test, though I'm still trying to figure that out.
View attachment 32420
Edit - I just read your previous posts and I see that you likely know much more about these devices than I do, so I assume I've misunderstood your meaning in the post above where you're rounding both readings of 4.5 and 5.4 to 5? It's not impossible for both of those readings to be so far off the mark of course but it would be two big consecutive outliers in 'opposite directions' with both of those readings being so close to 5. Not likely. Even the cheapest meter, the Navii on the far left in the bottom image is generally a lot better than that.

True

 

[saz9961]

So, back from a Medical conference (!) in Germany (so talking all day about medtech. My take away is there isn't much genuine innovation going on outside of AI; AI is increasingly being applied to existing products to differentiate them), but a little puzzled.

Morning before travel; 5.4mmol/l indicated, empty stomach (08:30)
Followed by one of those 200kcal shake breakfasts. At noon, lunch of 2 sardines in oil on toast. 15 minute walk to pick the car up from the garage.
3 hour later; 4.5mmol/l indicated. (15:00)
3 hour after evening meal (think it was a stew), and I snuck a pepperami; 5.7mmol/l indicated (23:00)
Morning of travel; didn't bother with a baseline, its been 5.1-5.5 for a while. Had one of those shakes. Didn't eat until evening so went hungry. Bit of walking in the morning (train, tube, wander a little around Heathrow), not much in Dusseldorf (taxi, hotel, tv)
2 hours after a rump steak, mound of French fries, salad with dressing; 5mmol/l.
Day 2: Morning baseline; 5.8mmol/l.
Breakfast of bacon & eggs, coffee with sugar because I couldn't find the sweetener
Walking around all day, lunch at 3pm, thai curry and rice (!). Probably didn't drink enough
3h later; 5.8mmol/l
Dinner in local pub; Hunter Pork schnitzel and chips. 2h reading confounded me; 4.2mmol/l (11pm)

Skipped any readings yesterday, morning was more conference walking, lunch at 2 with Currywurst, french fries, curry ketchup and chilli cheese (!), no evening dinner, snacked at airport on a packet of crisps. This morning, a predictable 5.6mmol/l before breakfast, and 2 hours after lunch, 5.5mmol/l
Weight is pouring off me.

Currently on 2x500mg Metformin. The doc has given me no direction when to take doses, so have elected for 1 in the morning, 1 in the evening, to flatten out the serum levels.

By now the doc wants me to increase to 3x500 (again, with no further direction), unless "side effects too much". I must have a high tolerance, because right now, the side effects are mild and tolerable.

I am hesitant to further increase the dosage, because the current serum levels are indicative of it "doing its job", ie getting blood sugar levels to within physiological acceptable limits. I'm surprised at the relatively small deltas between high and low (I was expecting to see between 4.5 and 7-8). 2 or 3 hours after eating should have caught somewhere near maxima.

I see that the mechanistic action of metformin is not even barely understood, only hypothesized

Metformin: update on mechanisms of action and repurposing potential - Nature Reviews Endocrinology

This Review highlights the latest advances in our understanding of the mechanisms of action of metformin. Potential repurposing of metformin for other indications is also discussed.

www.nature.com

But then my proverbial eyes rollback when I see this article:

Metformin: historical overview - Diabetologia

Metformin (dimethylbiguanide) has become the preferred first-line oral blood glucose-lowering agent to manage type 2 diabetes. Its history is linked to Galega officinalis (also known as goat’s rue), a traditional herbal medicine in Europe, found to be rich in guanidine, which, in 1918, was shown...

link.springer.com

100 year history of pharmacological use, and in crude form, perhaps 1000 years of use, and no one really knows how it works.

Kidney function tests;

Urine Microalbumin; 10.5mg/l (well within normal range)
Urine albumin/creatinine ratio; 0.86mg/mmol (bang in the middle of normal)
Urine creatinine level; 12.2 mmol/l (for my age, should be more than 1.3, but less than 26, so again, very normal)
GFR calculated abbreviated MDRD; 76 mL/m/1.73m
The "normal" value varies depending on other measurements. For Kidney Research UK, Over 90, indicates no kidney issue. Between 60 and 90, if no other issues, also normal. However, for the National Institutes of Health, anything over 60 is normal.

So I've no reason to suspect kidney disease. If anything, those kidney function tests might even have gotten "better", now that my morning BP has averaged 119/75 over 19 days, and afternoon 120/74 over the same period (ie. if there was a kidney problem, I'd be overdosing on metformin, with concomitant risk increasing of lactic acidosis). But maybe the lack of high glucose peaks might be associated with not drinking enough (I typically have maybe a 1000mls of liquid a day in addition to meals, so nowhere near the benchmark 2000mls). I've no symptoms associated with lactic acidosis.

 

[Leadinglights]

So, back from a Medical conference (!) in Germany (so talking all day about medtech. My take away is there isn't much genuine innovation going on outside of AI; AI is increasingly being applied to existing products to differentiate them), but a little puzzled.

Morning before travel; 5.4mmol/l indicated, empty stomach (08:30)
Followed by one of those 200kcal shake breakfasts. At noon, lunch of 2 sardines in oil on toast. 15 minute walk to pick the car up from the garage.
3 hour later; 4.5mmol/l indicated. (15:00)
3 hour after evening meal (think it was a stew), and I snuck a pepperami; 5.7mmol/l indicated (23:00)
Morning of travel; didn't bother with a baseline, its been 5.1-5.5 for a while. Had one of those shakes. Didn't eat until evening so went hungry. Bit of walking in the morning (train, tube, wander a little around Heathrow), not much in Dusseldorf (taxi, hotel, tv)
2 hours after a rump steak, mound of French fries, salad with dressing; 5mmol/l.
Day 2: Morning baseline; 5.8mmol/l.
Breakfast of bacon & eggs, coffee with sugar because I couldn't find the sweetener
Walking around all day, lunch at 3pm, thai curry and rice (!). Probably didn't drink enough
3h later; 5.8mmol/l
Dinner in local pub; Hunter Pork schnitzel and chips. 2h reading confounded me; 4.2mmol/l (11pm)

Skipped any readings yesterday, morning was more conference walking, lunch at 2 with Currywurst, french fries, curry ketchup and chilli cheese (!), no evening dinner, snacked at airport on a packet of crisps. This morning, a predictable 5.6mmol/l before breakfast, and 2 hours after lunch, 5.5mmol/l
Weight is pouring off me.

Currently on 2x500mg Metformin. The doc has given me no direction when to take doses, so have elected for 1 in the morning, 1 in the evening, to flatten out the serum levels.

By now the doc wants me to increase to 3x500 (again, with no further direction), unless "side effects too much". I must have a high tolerance, because right now, the side effects are mild and tolerable.

I am hesitant to further increase the dosage, because the current serum levels are indicative of it "doing its job", ie getting blood sugar levels to within physiological acceptable limits. I'm surprised at the relatively small deltas between high and low (I was expecting to see between 4.5 and 7-8). 2 or 3 hours after eating should have caught somewhere near maxima.

I see that the mechanistic action of metformin is not even barely understood, only hypothesized

Metformin: update on mechanisms of action and repurposing potential - Nature Reviews Endocrinology

This Review highlights the latest advances in our understanding of the mechanisms of action of metformin. Potential repurposing of metformin for other indications is also discussed.

www.nature.com

But then my proverbial eyes rollback when I see this article:

Metformin: historical overview - Diabetologia

Metformin (dimethylbiguanide) has become the preferred first-line oral blood glucose-lowering agent to manage type 2 diabetes. Its history is linked to Galega officinalis (also known as goat’s rue), a traditional herbal medicine in Europe, found to be rich in guanidine, which, in 1918, was shown...

link.springer.com

100 year history of pharmacological use, and in crude form, perhaps 1000 years of use, and no one really knows how it works.

Kidney function tests;

Urine Microalbumin; 10.5mg/l (well within normal range)
Urine albumin/creatinine ratio; 0.86mg/mmol (bang in the middle of normal)
Urine creatinine level; 12.2 mmol/l (for my age, should be more than 1.3, but less than 26, so again, very normal)
GFR calculated abbreviated MDRD; 76 mL/m/1.73m
The "normal" value varies depending on other measurements. For Kidney Research UK, Over 90, indicates no kidney issue. Between 60 and 90, if no other issues, also normal. However, for the National Institutes of Health, anything over 60 is normal.

So I've no reason to suspect kidney disease. If anything, those kidney function tests might even have gotten "better", now that my morning BP has averaged 119/75 over 19 days, and afternoon 120/74 over the same period (ie. if there was a kidney problem, I'd be overdosing on metformin, with concomitant risk increasing of lactic acidosis). But maybe the lack of high glucose peaks might be associated with not drinking enough (I typically have maybe a 1000mls of liquid a day in addition to meals, so nowhere near the benchmark 2000mls). I've no symptoms associated with lactic acidosis.

If your metformin is slow release than you ca take both together usually with the largest meal as they are better taken with food but if not then as long as you take with food to minimise the side effects of a possible dodgy stomach then morning and evening is the usual unless you don't eat breakfast then lunch time works for most people

 

[saz9961]

If your metformin is slow release than you ca take both together usually with the largest meal as they are better taken with food but if not then as long as you take with food to minimise the side effects of a possible dodgy stomach then morning and evening is the usual unless you don't eat breakfast then lunch time works for most people

Its the "normal" version.

 

[Eddy Edson]

I'm surprised at the relatively small deltas between high and low (I was expecting to see between 4.5 and 7-8). 2 or 3 hours after eating should have caught somewhere near maxima.

Maximum would typically be 30-60min after eating.

The general guidance to test ~2 hours after eating with a target of <= 7.8 mmol/l or whatever stems from considerations of what single post-prandial measure correlates best with an AUC consistent with normoglycemia. Obviously this is very rough! More basically, if you're consistently still elevated at 2 hours then it's a reasonable sign that glucose metabolism is out of whack.

(The AUC is far more important than peak. Everybody can go high after eating; a peak reading gives very little info by itself.)

FWIW, it sounds like you've basically zapped it apart from the slightly high fasting read (less than 5.6 mmol/l for "normal). The metformin is likely doing very little for you.

 

[PerSpinasAdAstra]

Regarding the calibration 'sweet spot' of test strip manufacturing - I think I accidentally found it. Couldn't sleep so thought I'd slowly get out of bed and do a test to see if I could figure out how far off the mark my Libre 2 readings are at night. I half sleep-walked to the kitchen trying not to trigger foot to floor phenomenon with too much excitement, and I assume that lack of sleep meant no dawn phenomenon either. Tested the same drop of blood with four meters and this happened:

If that's a fluke it's a rare one indeed. The thermometer is there because I'm trying to figure out if temperature has a bearing on the relative bias of the meters. It appears it might, maybe. More 'science' required

 

[saz9961]

My continuing low (well to me) blood glucose level continue to perturb me. Monday, 2 hours after evening dinner, 3.9 mmol/l, in the morning, before breakfast, 3.2mmol/l, before recording at 4.7mmol/mol (before dinner, yes, I am a bit random). Yesterday morning, fairly shocked to see 2.0mmol/l on waking (I felt fine, not tired, but I have been recovering from a Cold since the weekend), rising to a reasonable 5.8mmol/l by dinI aner. 5.6 this morning, down to 3.9 this evening, before dinner.

I am having a 200kcal shake for breakfast. A 200kcal noodle sort of dish, with some crackers for lunch, and 400kcal wet meal in evening (usually a meat stew, chicken curry etc), again with a couple of crackers. During the day I am snacking on meat products, pepperami, beef jerky sort of things. I suppose my diet is low carb, high protein, though I had not planned it that way.

I left a message with the clinic for my GP to contact me about whether I need to increase the Metformin beyond the 2x500mg currently. All I got from the Doctor was a SMS I can't respond to saying I need to keep increasing Metformin to the maximum strength because UK research "says so" and, the doctor fairly apocalyptically reminds me, it might "extend your life". My body, the numbers don't seem to lie. This GP only seems to bother to come in twice a week. The clinic telephone typically takes 30-45 minutes on hold to talk to anyone except you can't request a specific GP to call you back. Her SMS just cheerily states I should contact the surgery if I have any concerns. Except that's easier said than done. I don't understand if the doctor found the time to use a surgery text message service (it's one you can't reply to), why she couldn't just pick up the phone and speak to me. This continues my "rubbish diagnosis". If this is a typical experience, I feel sorry for you.