A neighbor was telling me he had blood work done and he was told not to fast, the doctor didn’t need it for whatever he was interested in the test for.
Anyway, his glucose was 102 and the normal range on the report he said went up to 99. 102 was considered high.
We were talking about it, and he wondered if he repeated the blood test, but fasting this time, would he expect the glucose level to be higher or lower than 102?
Fasting, it should be lower. BUT!
That range is for fasting glucose. Since he hadn’t fasted, the range does not apply. Depending on when was the last time he ate and what, it would have been much higher than that and still perfectly fine: a chunk of cake, a ball of ice cream and a glass of sweet white wine will raise your blood sugar higher than a dish of wholemeat spaghetti bolognese and a glass of water.
Doctors’ offices usually test A1c now, not the current blood sugar level. The A1c reading is an average of your blood sugar over the last three months. It’s more accurate for diagnostic purposes than a one-time blood sugar reading because your blood sugar can vary widely (and wildly) throughout the day. You don’t have to fast for the A1c. More detail here.
I think that’s right. After decades of having to fast before a blood test, the last two years I was told not to fast. My doctor always ordered an A1C as a check against my cheating (by eating very modestly for several days before) for the glucose level. Now I assume they don’t even bother with the latter. It is a measure of the percentage of the hemoglobin in my blood that in a hemoglobin glucose complex and changes only slowly. However, if you are on insulin (I am not), then blood glucose level is still important. It is dangerous for it to be too low. Too high is a long-term danger, but too low is an immediate danger.
Eating modestly for several days won’t do it (I’ve tried that, too). The A1c is an average of your blood sugar for the past THREE MONTHS. Busted!
When I’m looking at my glucose levels, 99 and 102 are the same, as far as I’m concerned. A click of humalog will lower my blood glucose at least 15 points, most days, so three points is no difference; I don’t even bother taking any unless it’s over 150.
Then again, if he did have an A1c done and it came back 102, that’s a little high, yeah.
But as I understand it, it’s glucose tying itself to hemoglobin, and the “3 month average” is because hemoglobin replaces itself over 3 months? Some of the hemo will have attachments for the last 3 months, some will only have attachments from the last 1 or 2 depending on its age. So presumably it’s more reflective of your most recent behaviour, since some of the older hemoglobin has been flushed from your system; but still, in a 90-day window changing behaviour for the last 4 days wouldn’t significantly change overall results.
My bold.
Yes, that’s my point.
If your blood sugar has been over 150* or 200* for 10 weeks most of the time, and you low-carb for the last two weeks, your A1c will bust you.
To get the equivalent European (or Canadian) levels divide the US reading by 18. One is in mmoles/l and the other is in mg/dl. Since the molecule wt of glucose is 180 and the ratio of l:dl is 10, that is the result. Clear? Now I don’t know the exact correlation of A1C percentage to glucose is, but I have recently had A1C around 6% and fasting glucose levels of 6, so that sounds like they are numerically close even if one is a percentage and the other is measured in certain units.
Non-fasting glucose is only really useful for curiosity or insulin dosing. I have a feeling the doctor ordered it as part of a panel looking for something else and didn’t particularly care about an accurate fasting glucose reading. Even if it was a fasting reading I personally would not be concerned about a reading of “102” where “99” is considered “normal” and “120” is considered bad. My doctor wasn’t even concerned about a “112”. Diagnosing if there’s a problem requires fasting readings and other testing.
For daily blood glucose levels, the normal range is 80 to 120. Anything in there is considered OK. Note how wide that range is – 40 points (150%) from bottom to top. His 3% difference is nothing. A couple times I’ve taken the test twice in a row, and had nearly a 10% variation.
Often the machine simply has “glucose” as part of the list of results it gives. You get your iron, and your glucose, and your three cholesterols… by pushing the single big button, and for all 100 samples in the tray. It’s a lot simpler to engineer and use that than to engineer and use a machine where individual results are requested for each of the individual samples.
That may be a guideline for diabetics on meds but the blood glucose of healthy people varies anywhere from 50 to 160 mg/dl.
But my point is the opposite. Because of the nature of what A1C measures, the result is “front end loaded”, that is - weighted somewhat more reflective of recent rather than older trends.
But - if your problem is that you have a high A1C and are trying to game the system, how low can you really get by having a low-carb diet if your basic problem is that even with a low-carb intake you have high blood sugar? If the goal is to get the reading for 3 months under 7, getting your average down to 6 for a month won’t help much if your reading was well above that for 2 months.
If you’re eating rock-bottom *ultra *low carb (<15-ish carbs per day) and you still have high blood sugar, then you probably don’t have type 2 diabetes. You may have type 1 or something else.
Maybe I’m stupid- but why would I even want to game the system, either with an A1C or glucose ? So the doctor doesn’t put me on medication/change the dose? Why am I even getting the test if I’m going to try to game it?
Depends. The best excuse I can think of is if you have to apply for life insurance or travel health insurance or if there’s some sort of license restrictions on something (although that last one is dangerous if all you are doing is hiding the problem). Basically, some reason why you don’t want it on the medical record that you are diabetic if it’s marginal and you think you can deal with it using diet.
Also complicated by the fact that a blood glucose under 50 can kill you, so if you’re already shooting for a range of 90-120 and you run on fumes for a couple weeks, the test might be thrown off by a process known as “death”.
It could cost you your livelihood.
For example, a truck driver’s license (CDL) is normally cancelled when a person tests as diabetic. Even if the diabetes can be readily controlled by medication, their license is still cancelled. They can only get it reinstated if they can show that it has been controlled for a year. (Good luck in surviving that year without a job!)
So truck drivers with pre-diabetes symptoms work hard to prevent failing an A1C test. Often avoiding doctor visits, or self-medicating to try to deal with it. That doesn’t usually work for long. Getting appropriate treatment would be better for everyone, but the bureaucratic DoT regulations don’t recognize that.
There’s also the possibility that you’ve been chowing down on surplus Halloween candy or your kid’s stash for a month, and now realize that could have license implications so go low-carb for a while.
The Atkins diet was all the rage for a while (when I tried it, I went from 235 to 208lb - but it was a pain to stay on it and of course I gained all the weight back and extra over time). That must do something for your running blood glucose levels.
But my point still stands - you replace hemoglobin over time, so the average reading is going to be more heavily weighted to more recent levels. The reading is an assumption it’s been pretty much the same (i.e. fluctuate within the same range) all along.