Thanks for starting this topic. I was always under the naive and erroneous assumption that "dawn phenomenon" (DP) was a gluconeogenesis in the liver phenomenon, purely. I didn't feel I really suffered from it.
Then more recently, I have switched to a carnivore diet (on top of intermittent fasting where I don't eat before 13:00 every day), so have 0 carbs, essentially (certainly <10g in a day), and no food in the morning. I previously have done keto diets (not carnivore), and not observed DP then, but realise I was on Lantus then. I am now on Levemir.
Interestingly I never used to suffer from it before this year. I've not changed insulin, nor activity level, nor anything else I can think of.
The DP I see, where I wake up and my levels might be at 8-9mmol/l, and as soon as I get up they shoot up to 11-12mmol/l. This is happening without carbs in the evening (although I wonder if too much protein is contributing here), and instantly on getting out of bed it starts to go up. Work has been quite intense personally and I've put this down to increased cortisol on waking - cortisol is known to increase BS levels. However, I would hope (expect?) that longer term carnivore/very low carb diets would deplete glycogen store availability to reduce the effect of cortisol (I struggle with the idea of gluconeogenesis occurring so rapidly on waking, but perhaps there is the sleeping pattern regularity that my body is operating under and so as I wake up the same time each day pretty much, this is happening earlier and snowballs?).
It is interesting to read the comments above about reducing/eliminating DP by changing eating habits. My understanding is/was also that DP is related to glycogen -> glucose conversion from the liver (i.e. glucogenesis) due to the effects of hormones (as mentioned above by
@Marie 2 and others.
I've not tried different eating habits, but if I do lots of exercise the preceding day I don't get or get very diminished DP the following day (and actually the one after that too.) It appears that it can take a quite significant time to replenish both liver and muscle glycogen stores (much longer than I'd thought) - on the order of 12h to 24h depending on what you eat - I need to do some deeper diving into the numbers that are quoted to turn these into hours consuming something normal, vs lab tests with very high concentrations of glucose/sucrose.
This paper is quite interesting:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6019055/#:~:text=Liver glycogen is rapidly restored,conversion of lactate to glucose. especially the comment and links to papers (in the section entitled "TIME COURSE OF GLYCOGEN REPLENISHMENT") re replenishment (repletion is apparently the preferred term) especially the comment that those on low carb diets might "replete" 13% of their stores over 24h vs 93% for those on a high carb diet.
One assumes that the liver attempts to replenish (I can't deal with replete any more, sorry) itself to 100% if possible, so I'm guessing that the amount and timing of exercise those on low carb/intermittent diets are also doing also feeds into this.
My final assumption is that the liver is less inclined to release glucose due to cortisol/adrenaline (requires higher concentration for example) when it's in a depleted state, though this is purely an assumption, there are doubtless papers out there, if anyone finds anything relevant please post a linky.
There are a couple more papers that are linked from the first one that look interesting, I've not read them yet, only skimmed:
So I'm guessing the best solution for this in the near term sounds from the advice here to just instantly take some novorapid on waking (this can be a challenge, as I don't eat, and then often cycle to work which means novorapid + fasted state + exercise = large effect of reducing BS levels risking hypos following the exercise), but perhaps pursuing the pump may be a good way to manage this?
Do you find that the cycling first thing has any effect on BG at all re DP? If I leave first thing for a long ride, I (now) need to take bolus for my DP, and it's a fine line trying to judge how much to take to account for DP but not so much that I then go low. I do eat breakfast though, so I have safety in terms of that supplying glucose while I ride. I currently take about the same bolus to cover DP + breakfast as I would just to cover DP on day when I'm not riding.
Before the advent of DP for me, I'd leave in the morning (for long rides) without any bolus cover for my breakfast (though I would probably run a little high to start with, but it's hard to tease that effect apart from stress as I can ride without bolus cover if I leave at lunchtime). I therefore assume that my dosing requirement to cover DP is constant whether I ride or not, because I also eat breakfast which covers the insulin tail. I'm pretty sure (never tested mind you) that the DP effect is much shorter than the carbs coming from the digestion of my breakfast, it would be interesting to know if anyone has tested/noticed though.
That was very round-about and not directly relevant to you (sorry!) Are you willing/able to eat carbs in the morning in an emergency, so that you can try taking some very small bolus doses to see what effect they have but be able to avoid hypos? If so, I'd be tempted to try that starting at 0.5U and working up. I'd be very interested to hear what you find.
P.S. Yes, sort of, a pump would partly solve this problem. Some people suffer from a slowly increasing BG in the morning due to I'm guessing growth hormone (et al), for something like this a pump is ideal as you can increase the basal rate and remove the rise. For "foot on the floor" DP caused by adrenaline (afaiu) I can't see that a pump would help - you could perhaps try to taper your BG downward before you plan to get up so that you're heading low to offset the high, but you'll still end up needing to take insulin which has a long tail which will affect your fasted riding. I'm not on a pump though, so am more than willing to accept I'm wrong about this.