510(k) Approval Timeline

My little company got 510(k) approval a while back and is steaming right along with planned first man sometime in first quarter 2009. There is still some shelf life testing to finish and another software release to validate before we're ready.

The basic 510(k) and design and development timeline went something like this:

• June 2006: Idea
• September 2006: Concept Drawings
• September - December 2006: Solidworks modeling
  
• October 2006: Foam mockups
• December 2006: SLA Prototypes
• December 2006 - June 2007: Many design iterations including preliminary packaging testing
  
• June 2007: Clinical Review
  
• July - November 2007: Validation Build (Disposables and System)
• October - December 2007: Sterilization Validation
  
• December 2007: Verification Testing
  
• December 2007 - February 2008: Biocompatibility Testing
  
• March 2008: 510(k) Submitted
• June 2008: Initial FDA 510(k) Response, Request for more information
• July - August 2008: FDA Response and further correspondence and testing at FDA request
  
• November - December 2008: FDA 510(k) Approval

Actual engineering time of about three days, the rest was paperwork. Ha! I kid. Anyway, the whole 510(k) process went fairly well I think, there could have been some better focus and maybe two months would have been taken off the total time. There were also a couple FDA comments that could have been avoided with better company proofreading and checking, although neither of these caused delays.

Some things that I learned during the 510(k) process:

• Packaging sucks and can screw up your entire timeline, master cartons help. The good people at Mangar were a huge help.
  
• Leaving software features off the initial 510(k) submission then adding them in later saved a huge amount of time.
• Same with shelf life, just start it and tell the FDA its in process. Final shelf life testing has to include redoing any functionality testing unless justified.
  
• The FDA is very concerned about software and will go over your software verification very thoroughly, we didn't skimp on this documentation and I'm glad we didn't.
  
• Leave room on your product labeling for a CE Mark (doh.), but don't spend too much time on labeling, it is going to change.
  
• A small company does not have to spend a lot of money on a regulatory consultant to get the 510(k) approval. Just follow the instructions and have a guy/girl you can call.
• Risk mitigation, hazard analysis, or whatever you want to call it needs to be robust and to ISO 14971.
• Unless you are lucky, outside testing takes longer than you will be told, sterilization testing took at least a month longer, biocompatibility testing would have been on time if not for a false positive control failure, and Safety/EMC/IEC 60601-1 testing took a month longer than described as well.
  
• To move things along more quickly, hire more engineers and less management, even a new engineer is a huge help, when was the last time you saw management write a decent software design specification? What do you need to get a 510(k)- someone doing or someone saying it should be done (and that is if they actually know what is going on)? Exactly. Don't skimp on software licenses either.
• If you are management and you are confused about what to do, leaving engineers alone is a good start, look forward and think about what you'll need to actually sell the device, if you must "help"- follow up with vendors that the project is waiting on.
  
• Testing at the limits is good, but in our case the FDA required more "real world" verification testing. Cover your bases and throw in a couple mid-range points as well.
• Wiring detail can wait until before production. Just have a diagram and someone qualified to build to it.
• Buy a decent camera up front, getting good pictures saves a lot of time going back and forth with vendors or consultants if any problems come up.

That is all for now, have a good new year!