Feb 9, 2008

Sterilization wrap up

I promised a while ago to summarize the sterilization situation, hopefully someone will find our experiences useful. Our disposable product was designed for EtO sterilization, an important potential add on to the product made by another company has been approved using EtO and changing their sterilization method at some point would be a huge hassle. So we built the validation samples, shipped them off to the sterilizer, they wrote the batch release protocol, based upon the add on's protocol and we ended up failing sterility testing for both the half cycle and full cycle.

A QA and I forced a meeting and flew immediately to the sterility test lab, I think the lab is still unhappy with us for this- they avoid talking to the QA at all now, but thats what we were told to do. Anyway, we sat around a table with the lab personnel and they told us in not so many words, what we already knew because they said the same things on the phone, its just one of those things. The lab actually handled it very well and had some useful minor suggestions, but there is not much they could do. You don't want to do sterility testing if you can avoid it because these things happen, obviously you have to do it a few times to validate. We made a few very minor changes to the testing procedure, doubled the sterilization time and flew back to help finish build more samples, working through many a weekend. When we tried again with the new protocol it worked out well for us, and it was done quickly enough.

Since at the time of the failure the company had more money than time, we decided to hedge our bets and also approached Anderson to run a batch release in almost parallel. The Anderson folks wanted to do some development work first, due to our previous failed sterilization attempt as generally their sterilization method is gentler than standard EtO sterilization so there was some question about if it would work at all. We got the samples to them, they ran the development work and everything came out roses- it was on. They sterilized the samples and sent them to Apptec for sterility testing and everything tested sterile. FWIW, the Anderson half cycle process sterilized what the 2 hour standard EtO cycle did not, but I don't put too much stock into that bit of info, as I know several devices which use a sterilization time less than ours and are triple the size and more complicated. Overall, the Anderson sterilization cost about half as much and was performed more quickly from start to report, if you take into into account that they did some development work first.

Going forward we're going to file the 510(k) initially with the standard EtO sterilization, but in the future add Anderson, especially since we have plans for a series of low volume, high cost disposables that would be perfect for Anderson. The FDA considers the Anderson sterilization a "non-traditional" method, but I don't think it would be much of a problem to get approval for as the method is becoming more common and is well documented.

Theoretically you can use 3 successful batch releases to validate sterilization, if they are exactly the same composition, that is not going to happen. I've been pushing the company to go through a full sterilization validation, but I don't think that is in the cards for this year. I have theoretically officially turned all sterilization duties over to QA although I think that will be forgotten next time we need to sterilize something.


Anonymous said...

Hi there,
congratulations on the completion of the big project. I am a microbiologist working for a medical device company, currently in the same situation that you were facing a couple months ago. All BI samples passed sterility test but not the NSP. We, however, suspected that it was the testing itself that could have created the contamination as the product was rather complex. On reviewing the ISO 1135 closer and consultation with an expert, we realized that ISO 11135 never really asked for NSP in half cycle. So we decided to go for sublethal cycle instead and pray for the best. Wish me luck!


JCK said...


Thanks for your encouragement, my company had the same suspicions and it actually set some tempers off. I have kept fairly quiet about them here because it is impossible to say for certain one way or another (at least right now with limited data) and you just have to work through it. At one point my boss and I discussed the wiggle room in ISO 11135, but we're not experts and our sterilization resources weren't as solid about it as we would have liked. Additionally, we are going to have to redo validation probably this year, so hopefully we can improve on what we have so far.

I hope you guys are successful!

Stewart said...

Sterilization is really important for all aspects. All must be sterilized well. Sterilization must be based on the standard.

omindustries said...

congratulations on the completion of the big project.Nice Big achievement .