EO Sterilization Validation - Sterility Testing and Load Configuration

[htcoztrk]

Hello to all!

I searched the forum to see if this thread already been opened but couldn't find any. If it was already opened, I apologize in advance.

I just started to work in a medical device manufacturer of products to be sterilized by EO gas and they -we- are validating the EO sterilization process which I have very limited experience with. So I already prepared the protocol based on EN ISO 11135 and some guidances however there are two points that I am not very sure:

1) Product sterility testing: So the SAL is determined as 10^-6 and we will use biological indicator containing Bacillus atrophaeus spores. We will perform 1 fractional (to see the recovery), 3 half cycles and 2 full cycles. BIs will be incubated and tested for sterility after each cycle. Sterility of actual products on loads to be performed after half cycle. However after which half cycle should we tested the product sterility and on what basis? Or sterility of products need to be tested after every half cycle? But in this case it is a huge money burden for the company. I checked the old records but it was too complicated for me to understand. So any help to clarify at which point should be perform the product sterility will be very appreciated.

2) Worst case configuration: We will use the mixed load during validation studies and already determined the worst case based on mass, density, packaging, shape and etc. of products. However I am not sure how does the loading configuration of mixed load into the sterilizer affect the validation. In case we demonstrate that all points inside the sterilizer are within the specified parameters (by performing physical validation simultaneously), does it matter where we put the loads during validation study and routine sterilization?

-Hatice

 

[planB]

htcoztrk,

I presume you are following the overkill half-cycle approach in validation your cycle (State your validation approach somewhere explicitly in your protocol):

ad (1): Product sterility testing is most critical and important to be performed on sample that have been subjected to sub-lethal (fractional) cycles. These cycles are intended to demonstrate that your product bioburden is easier to kill than your challenge organism. Therefore, you aim at the following result in sub-lethal runs: sterile product, some survivors on your PCD (process challenge device).

Once you have demonstrated this, you preferably perform sterility testing on products also during half cycles, confirming the results of your sub-lethal runs, but you do not necessarily have to in all half-cycle runs. More important is the demonstration of complete inactivation of your PCD in orer to show that you achieve the required SAL of 10^(-6).

Cost: microbiological tests are typically not expensive - residuals testing will cost you much more per sample

ad (2) Load variations might affect both lethality (i.e. cycle effectiveness) and sterilisation residuals (i.e. cycle safety), since the load has an impact on the critical parameters EO concentration, temperature (distribution), and humidity (distribution). You might consider performing a bracketed validation, having a minimum and a maximum (worst-case) load in half cycles and thus demonstrating that the load configuration has no impact on the SAL. You should map these half cycles in order to generate evidence of the uniformity of physical data that you are talking about as part of your PPQ activities. In this way you can partially combine MPQ and PPQ runs.

In terms of residuals, you will have to generate a full-cycle worst-case exposing your products to the maximum amount of EO and potentially even introduce a challenge to aeration. Note that a maximum load may absorb EO and thus, may not represent the worst case in terms of residuals.

HTH,

Gerhard

 

[vyaas97]

Thanks Gerhard for taking the time to explain the answers. However, I had a follow-up question: What if during the fractional cycle, product sterility was achieved i.e NO growth on any of the products whereas ALL the BI samples showed growth (PCD).

Is that still valid (since it clearly shows that PCD is more resistant than the product bioburden) or should the fractional cycle be repeated to ensure partial kill of PCD (some of the BI samples show growth whereas others remain sterile)? If the fractional cycle has to be repeated, then in that case, for what purpose(s) is the partial kill of BI (and product sterility) necessary?

Note that the overkill half-cycle approach is used during this procedure. Since adequacy of recovery can be shown through the presence of growth in ALL BI samples, is it necessary to have partial kill of PCD in this step (some are survivors whereas others show growth)?

 

[planB]

What if during the fractional cycle, product sterility was achieved i.e NO growth on any of the products whereas ALL the BI samples showed growth (PCD). Is that still valid [...]?

Yes.

[...] is it necessary to have partial kill of PCD in this step [...]?

No.

Just take care that you do not employ an overly resistant/challenging PCD that may require you to design an overly lethal cycle.

HTH,