This post attempts to draw-together two recent threads from here on the MRI Metal Detector blog. First, there was a long-running question about the FDA and their online-accessible database of medical device accidents which, for months, appeared to be malfunctioning, and recently was repaired. Second, there was my post in which I identified 5 MRI ‘Never Events’ which, if industry standard procedures are followed, should never occur.

As I mentioned in the article on the restoration of the full MAUDE narratives, I filed a Freedom of Information Act (FOI) request for the data, motivated by a concern that the problem with the online database would not be resolved in a timely fashion. Below are a handful of PDF files from my FOI request, enumerating just MRI projectile accidents (one of the five types of MRI ‘never events’) from part of the 2009 data…

• Bed tray-table (ambiguous injuries, including facial lacerations)

• Rolling cart seriously injures Siemens Apps Specialist (facial fractures & brain trauma)

• Flat-screen monitor hits research subject (facial fractures & surgery)

• Patient on gurney gets more of a ride than planned (foot, ankle, leg fractures)

• IV cart nearly strikes patient (near-miss)

• Knife slices patient (laceration requiring stitches)

• ‘Sand’ bag injures patient (brain hemorrhage, tongue laceration and facial injuries)

• Scissors seriously injures tech (embedded in forehead, surgical removal required)

The length and scariness of this list says two things to me… 1. Even without correcting for the presumed 1% reporting rate, this list is already too long suggesting that we have a long way to go, and 2. Why aren’t we taking a more proactive role in preventing these sorts of accidents when there are tools and techniques readily available?

Is it that crushing facial injuries, brain trauma and scissors embedded in someone’s forehead are collectively ‘minor exceptions’ even when these events (and many others) occur within weeks of one another?

To answer the rhetorical question posed by the title of this post, absolutely we continue to call them ‘never events’ because they should never happen. The fact that we have a long way to go to get close to that frequency is not an indictment of the validity of the goal, but it is a reason to call for professional / regulatory change if the industry can’t close the gap on its own.