Interesting. When I started reading this I wondered if you had encountered the "Differently aligned group" bug that can trigger recover to report a major problem with a file when none exists, but that would result in recover creating a new file.
Have you tried using FMDiff on this file? It might tell you something that Recover doesn't.
I've been there with this issue. "Do I spend many hours rebuilding the file or do I burn copious incense to Demon Murphy while hoping nothing catastrophic happens to my file"? The magnitude of the consequences for such a failure is usually the determining detail. If it's a file that won't bring business to a screeching halt and I have lots' of backups being made each day, I'm more likely to risk it.
One thing to watch out for if you choose to continue to use the file as is: A new release, new update, change in platform may cause this file to go from "tolerating" the issue recover is tripping on to a major failure. I've seen such differences before and you just don't know...
For those interested in reading about the bug I mentioned: FMPA 10: group graphic + align = corrupt file?
This is one of many acknowledged bugs that can be found in the Known Bug List here in the Report an Issue section of the forum.
It can also be downloaded as a database file from: http://www.4shared.com/file/8orL8apk/FMP_Bugs.html