Hi,
@kitedoc , Dana Lewis, one of the pioneers of homemade artificial pancreas, wrote an interesting article on calculating carb absorption rates:
https://diyps.org/2014/05/29/determining-your-carbohydrate-absorption-rate-diyps-lessons-learned/
With me, peaks can be very variable. I'll usually see a rise start after about 30 to 45 mins, peaking about 60 to 90 mins later, but with some low gi foods, I can have a straight line for like a whole 3 hrs then it starts to rise from then. But other times the same meal just doesn't do that.
With cgm, I'm able to watch it and usually pin it with a few units correction before it gets too high, but sometimes it just does a flier and needs way more.
I'm suspecting one cause might be that if it's a particularly low gi meal, the digestion of the food is still kicking in well after the insulin has peaked.
Another possibility I've wondered about is that, while, after 30 yrs of T1, I have no obvious complications apart from mild background retinopathy, I wonder sometimes whether there is some mild vagus nerve damage, which can slow the empyting of the stomach into the intestine where glucose is absorbed. I've more or less ruled this out as I don't seem to have any of the other symptoms of gastroparesis, and it only happens once in a while.
I've been re-reading Stephen Ponder's Sugar Surfing recently. He says that during it's travels round the body after injection, some insulin is destroyed before it gets anywhere near doing its job, and the amount destroyed can, astonishingly, be as high as 90% depending on how one's biology is working that day.
He doesn't cite any references for that but if that is correct, it would go a long way to explaining T1 randomness - how 5u will be very effective one day, and seemingly do nothing the next.
I think what cgm.brings to the game is a heightened awareness of the user's general insulin sensitivity over the last few days; a keener appreciation of how 2 or 5 or 7 units might act that particular day; information on whether we're trending up or down or stable pre-meal so we can decide on adding or subtracting from the bolus; and accumulated info from long use of it to be clearer about appropriate pre-bolus timing for different types of meal.
Stitch all those factors together, and we can end up pushing lines which don't really have any post-prandial peak at all:
View attachment 32333
But, of course, T1 being T1, you then get rubbish like this, flat line for 2hrs post-meal, then an inexplicable rise to 9 which took 7u to drag down. I reckon there's something in what Ponder says - some days more insulin is being destroyed than on other days. I'm fairly relaxed about this sort of stuff now - it's T1, it happens, heck, it would be boring if it was too easy!
View attachment 32334