The problem is I don't think most brewers here have enough control over their fermentations to reliably predict where the beer will end up. However, with a good finishing hydrometer, of course, it's theoretically possible.
I think a similar and more predictable technique would be to bottle with krausen from a simultaneously fermenting batch. For example, if you have two pale ales, one done with secondary and one at high krausen, you could siphon off a measured amount of the wort in primary, measure the SG, calculate SG * volume to determine how much sugar you have, and or subtract a little wort to compensate, and then bottle.
I know an HBTer or two has tried that technique and I know some commercial breweries do it as well. It strikes me as being substantially more predictable because the yeast is being mixed with a yeast with a known terminal gravity... and the wort being added is a smaller proportion of the whole amount in secondary, so there is less opportunity to miscalculate your expected FG.
I can see the value in priming with high-krausen wort. I cannot, however, see the value of doing this unless you have a REALLY nice set up that is well protected from risks of infections. With couple of siphon tubes and an open bottling bucket thrown into the mix, I don't think the risks are worth the rewards.