How do you approach an isolated increased protein levels in the CSF in patients with cognitive decline?   

Specifically, a) what protein levels are considered significant and b) are there anything that can cause a false increase or a baseline increase in these levels that are unrelated to neurologic disease?