I think most people have missed the point as to why the TR protocol chart was updated.
In the past, conventional wisdom seemed to be that the difference between a human and pet meter was "30 points". Therefore, those following the TR protocol that were using a human meter would reduce the dose when they got a nadir "under 50".
If you were using a pet meter, the protocol said to reduce when nadir was "under 80".
Apparently, they determined that there wasn't a 30 point variance on BGs at the lower ranges of normal.
So they updated the chart so that the reduction was given when a cat read "under 68" on a pet meter instead of 80. I dunno, maybe they saw reductions fail at 80 but stick on 68.
What is important to keep in mind is that other than "at the lower range" (target range for TR), the updated protocol is "silent" on the difference between the two types of meters. They don't say the "new variance" is 18 points at any other range of BG. And they don't say that the "30 - 40% difference" applies across the range of BGs.
We don't KNOW that. We're making assumptions as to what the degree of difference between the two types of meters might be at BG ranges above normal.
In reality, what the difference is at higher ranges doesn't much matter if you are following TR, or SLGS with Lantus or Lev. If you get crappy nadirs, you're going to up the dose anyway. It might matter more to non-depot insulin users who use an Alphatrack, especially if they shoot using a sliding scale. If you use Prozinc, and your scale says to shoot more at a PS of 350, there's a big difference between the old standard of "30 points" (380), and 350 plus 35% (473). That's why it would nice to know what the difference between human and pet meters is at higher BG ranges...