I'd give the repserver changes a try before discounting them. While it may sound like a pain in the keister it is relatively easy to implement & test, especially compared to ...
A 'sp_rename' solution will require a thorough scrub of all existing T-SQL (including application/batch coding) and documentation, along with (ideally) a regression test of all modified code. At this point the repserver based change looks a bit more enticing, eh. ;-)
Message was edited by: Mark A. Parsons