When I was in RT School back in 1995 the Bennet 7200 was the most popular ventilator and some of the newer ones provided the ability to do weaning parameters on the machine without even disconnecting the patient from the ventilator. Our teacher said doing parameters on the ventilator was a bad idea.
He said it was a bad idea because the parameters would be not as good as if the patient were taken off the ventilator. For example, if you would get a 500 tidal volume by doing the parameters off the vent, you'd get a 450 on the vent. For NIF it would be the same: the value while on the vent will be less than if you did it by hand with the patient off the ventilator.
Fifteen years later we are fifteen years wiser. And I have to say I agree with my teacher that weaning parameters on the vent are a litte less. However, I don't agree that it's less effective. Now that we have the Servo i's we are able to do all parameters (VT, NIF and FVC) without even disconnecting the patient from the vent.
As a rule, as long as everyone does it the same way, that's all that matters. If my coworker does parameters one morning and gets a 400 tidal volume off the vent, and the next day I use the vent and get a 350, my 350 doesn't look so good. It looks like the patient is going in the wrong direction.
However, if we both would have done it the same way, then we would have known the patient was consistently getting the same tidal volumes. So consistency is the key.
At our hospital we have a protocol that if the ETT is 7.5 or greater we put the patient on a PEEP of 5 and a PS of zero, and if the ETT is smaller than 7.5 we use a PEEP of 5 and a PS of 5 to make up for the resistance of the ETT. We all use the Servo i (if the patient is on that ventilator). All parameters this way are consistent because we're all doing them the same way.