It is absolutely way too high. Here's a paper that puts it at about 1 per 13,000: Lagasse (2002): http://student.purduecal.edu/~jbojczuk/Research/Anesthesia%2...

It's worth noting that their data contains zero observed deaths in ASA Physical Status I Patients, which are basically healthy adults.

I would assume things have either stayed the same or improved: Sevoflurane was just starting to be used around the time that the data were taken for that paper.