Ouch! I hope recovery was possible without excessive downtime...
In case people haven't met it already, the Refactoring Databases book by Pramod Sadalage has a number of strategies for this and other similar situations. The website has pretty much all you need to know, unless you like paper copies, and/or would like to support Pramod :)
I personally like the use of migration scripts, tested & deployed with the code through the standard pipeline (perhaps as part of a management application). Decoupling when they are applied and leaving control of that to the appropriate ops team keeps the humans happy too.
We're a place where coders share, stay up-to-date and grow their careers.
We strive for transparency and don't collect excess data.