AUC values (advanced CGM use question)

Hello,

I’d like to tap into the knowledge base here for a helpful discussion of the concept of Area under the Curve (AUC), specifically, as displayed on my MiniMed pump when wearing the Enlite sensor, under the Utilities > Daily Totals menu.

I understand what that concept captures (don’t ask me to explain it back, I am not going to be able to, but I read a lot of definitions and explanations how to calculate it in principle, so I get the main point – it measures the length of time of hypo- and hyper- excursions, but also accounts for their severity/level off from my pre-set targets). I also understand intuitively that you want it to be as low as possible on both counts; less variability of BG is my current goal precisely. So far, so good.

But I want to know what number I should aim for: what is the target AUC that shows normal variability of BG over the course of a day? (On the logic that the target HbA1c is that of a normally functioning pancreas, with the understanding that it is not always possible to achieve for a PWD.)

I only started paying attention to them a couple of days ago, with the insertion of my last sensor, and currently see the following:
Today (still mid-day on Oct 24)
AUC High 0.6
AUC Low 0.0

AUC High 1.0 (fantastic day, when I felt what a difference CGM can really make)
AUC Low 0.1

Oct 22 (not one of my best days)
AUC High 6.7
AUC Low 0.5

Oct 21 (pretty average day)
AUC High 4.3
AUC Low 0.0

Oct 20 (pretty average day)
AUC High 4.6
AUC Low 1.7

I just have no good way of interpreting these mysterious units. Any knowledgeable discussion (Are there are any resident endos on this site?) welcome.

hi @Dessito,
I don’t see too many doctors on this site, and probably due to liability, they’d likely not say anything as doctors. Most of my doctors wouldn’t understand calculus anyway.

the definition you seek confounded Issac Newton for most of his life, don’t feel too bad that it isn’t super easy to get or understand.

forget “under the curve” for a moment. consider what “area” means. if you have an easy shape like a rectangle, Area = width X length and that’s all. Now take 2 rectangles,

  1. 1 inch high by 2 inches wide; area 2 square inches
  2. 2 inches high by 1 inch wide; area 2 square inches

these rectangles have the same area, one is wider than high, the other is higher than wide.

consider a third rectangle 200 inches high by .01 inches wide; this one is super tall but very narrow, guess what - area 2 square inches this 3rd rectangle can illustrate something bad happening for a very short amount of time, if you change the “high” by “wide” to “blood sugar level” X “duration in time”.

we are supposed to keep our blood sugar between 70 and 100 (sorry - I use mg/dl) all the time. suppose you get a high blood sugar say 200 but it is gone after an hour. is this better or worse than a 200 that lasts 6 hours? I say better. the AUC for the first excursion would be 1001 hour or 100, the excursion for the second is 100 * 6 hours or 600. (100 is the amount above your target) (my AUC "units are in mg/dlhours and may vary from your cgm calculation).

2 ways to make the AUC (high or low) larger: bad excursion or long time or both. the AUC is a cumulative number that describes the multiplication of both the excursion and the length of time.

why would anyone care? this is a good question. the reason anyone would care is there are few numerical ways we can describe control. if you go by blood sugar “average” or even a1c, you can get a decent average or even an a1c in the 6’s by following every high with a low. example

monday blood sugars every 4 hours 55 250 49 170 60 60 - average? 107! super awesome! NOT!
Tuesday blood sugars every 4 hours 75 99 67 103 130 98 - average? 107! actually awesome.

so average is a “bad” way to describe control, so is HbA1c, AUC adds a dimension that the average leaves out.

if my target was 90-130, the AUC high and low for Monday would both be large, while they AUC for Tuesday woudl only show a little number for AUC low on Tuesday and 0.0 for AUC high.

hey I talked too much. PM me I can do examples till the cows come home.

cheers!
Joe

Great explanation Joe! I’m transitioning to Enlite later this week and this is a huge help.

Levi

Yes, thanks @Joe – as I said, I had grasped the crux of this from more general discussions of AUC in pharmakineticswi, but seeing another description (and a very clear one) of the concept applying it directly to the diabetes context still helps.

I guess the units then are △mg/dl*hours. (I am in the US and measure in the mg/dL too.) And the target (however ideal) should be 0. That’s what I’ll take away from this until someone who knows better tells me otherwise. :slight_smile:

May all your AUCs be as small as possible,
D.