Paul/Gail, thanks very much for both replies - very helpful.

I should've mentioned that we do already know the cause of the corruption, which was a failure of a SAN controller. Unfortunately, by the time the failure was identified and a DBCC CHECKDB was run, nearly a week had passed, by which time the opportunity to restore from a clean backup had passed. We do have a Monthly backup from before the corruption occurred, but restoring this backup will result in a lot of data not being restored.

Running DBCC PAGE as suggested allows me to identify some of the leaves that will be affected, which will help us to identify and restore any SharePoint docs that get deleted by the repair process. However, I think we'll have to accept that other data may also be deleted, which we'll not be able to identify.

Either way, it looks like we really don't have any option other than to run the repair and live with any consequences.