I asked a similar question to this a few months ago, Pappers basically said he has never tried it but it couldnt hurt so I went ahead and did it. My observations of the results are as follows.
1) the primary issue with this process is that bottle carbing is a warm(ish) process but CO2 doesn't dissolve well in warm liquids so you end up loosing allot of carbonation when opening a bottle up. Ideally, you would need to refrigerate the test bottle a couple of days for the CO2 to properly desolve in the liquid but then it would not be a good representation of the rest of the batch. So.....
2) Observing the fact that it is not anywhere near an ideal process, I do it this way anyway (using a different test bottle each time). Once I have achieved a desirable level of carbonation (4-6 days), I pasteurize the un-tested bottles and leave the tested bottles sit another day or two and then just throw those in the fridge. I have done it this way twice now and the results have been good enough that I dont see changing methods any time soon.