I had posted this on another thread... Non-ferrous metal is not a problem in an MRI, from the standpoint of patient injury (i.e., the magnets pulling the metal right out of the patient). Non-ferrous metal can, however, cause some interference to the resulting images.
I have a titanium plate in my neck. I did have an MRI following the surgery, but I ended up having to have a different radiographic procedure done because the titanium did interfere with the quality of the images.
An aside to Osage Orange-- you wrote "Most orthopedic implants, even though they may be ferromagnetic, are fine because they are firmly embedded in bone. "
My sister and I just had a conversation about this the other night. Most orthopedic implants are non-ferrous. The impression I got from my sister (works in medical imaging) is that while ferrous metals attached to bone wouldn't get pulled out of the body, it might pull the part of the body that's attached to the metal to the MRI magnet-- at a rather high rate of speed.
But, it was rather late when we had that conversation, so maybe I am not re-calling it correctly...
Even if a ferrous-metal object is not pulled from the body at a high rate of speed, I imagine it might be given to high-speed oscillations/vibrations and/or some type of heating; either of which would play hell with the surrounding tissue.
I invite correction, but it is my belief that any implant of ferrous metal must be removed prior to an MRI.