Today we had out first A1c where all the time was spent in full puberty mode (for our last A1c, only half the time was spent in puberty mode). I was expecting something in the 6s but I was taken aback by the results: 5.4% (our last two were 5.4 and 5.3 in chronological order).
I am not 100% sure if the number is fully true, since my son has gone through a good number of lows in the mid-50s: I understand that deep lows can de-glycate hemoglobin (@Chris, you had some references on that recently?). I am not sure if these lows would have been enough to do so.
However, even if the “real” A1c is a bit higher, this really cheers me, because it means that it is possible to maintain a decent A1c during puberty. So, as a data point, I am sharing how we have changed my son’s diabetes management since he entered puberty. We have only been at it for 6 months, so this is a work in progress 
-
We have lowered our aim point from 120 to 100
-
We have adjusted our acceptable range to 70-120 from 80-150
-
We have decided to live in what we call “the lower half of the range:” if we are significantly in the upper half of the range (100-120), say, above 105 or 110, we inject to go to the lower half (70-100)
-
We changed the way we make decisions on insulin injections. We used to often wait for a few more points to make sure of a decision. Now, the moment the BG crosses 120, we “decide and inject,” assuming that we will get to a certain level (typically assuming 180), then keep on injecting more if we hit the level we were aiming for.
-
We also routinely stack, with caution. My son’s insulin activates after 40 minutes. If, after 1 hour 20, we don’t see significant corrective action, we stack – again, carefully. For instance, we had a particularly resilient peak 2 weeks ago – we stacked 5 injections on a row every 60-80 minutes to get the peak down (rare case though). FYI, my son’s DIA is about 5h 20.
What is the price we are paying for this? We had a stretch of several months where we were hitting too many lows in the mid-50s. We have worked very hard on it, and are still working on it – we still need to improve there. As a reference, over the past 2 weeks, we have spent less than 1% time below 60, and about 6% time below 70, per the Dexcom (although some of that is an artifact – the real numbers are less). Still, we have more lows than we used to before puberty.
At the same time, we have improved our % time low, and know how to do it better – we just have to practice it. So, imho, in the end the % time low will be quite low, and the real price to pay will be sleep time – my wife and I spend a lot of time up at night regulating the boy, so that he does not have to trash his nights and be tired in the morning. I’d say we average 3 alarms per night, low or high – for many of which you have to stay up 1.5 hours or more, to make sure, for instance, that the high is properly treated.
So, in the end, I think the primary cost is sleep, and it is a significiant cost – bad enough that we have to take turns: a week of it puts me on my knees :-). This limited cost is true only as long as the boy is willing to do the routine. When that changes, all bets are off.