Hello! I am in need of some guidance... My company is a startup in the US and we produce extruded cereal. This is my first experience working in the extrusion industry (started in January) and we have only had a total of 3 months run time on our equipment this year. A small amount of water is introduced to our dry mix as it enters the extruder barrel. The dough is then extruded, shaped, dried through a forced air oven and ambient cooled before being packed for further processing by our customers into things like cereal mixes, protein bars, granola, etc. We are low risk RTE and are currently going through kill step validation as required by several big companies we are pursuing as customers. Our BRC consultant also advised the validation as well. Our first attempt at a validation study was just focusing on the oven and hoping we would achieve a 5 log reduction in Salmonella, but we only saw an average of 2.5. Our next attempt was incorporating the extruder and oven which we believe was successful but unfortunately were unable to confirm. Our third party lab inoculated a dry ingredient used in our dry mix but it died during transportation to our facility. We will be performing our third test in a couple weeks with another dry ingredient they will be inoculating, but this time they are accounting for the transportation loss. I have never setup a kill step validation before so I am heavily relying on our third party lab for guidance. I come from the meat industry where time and/or temperature are usually a CCP and this is what our lab has suggested controlling. However, with our extruded product I am afraid we will be boxing ourselves in if we require hitting x temperature, x time, x pressure, etc. because most of those variables fluctuate due to processing the product. My mind keeps gravitating towards controlling with our low aW (typically .2 - .3) but that only applies to listeria, correct? Can someone help point me in the direction of a resource to answer this or provide similar industry experience? Usually when I ask questions like this the responses are "risk analysis" which I intend on doing but I first need to find the resources and information to be able to properly educate myself and then risk asses. I just keep feeling like I am hitting a brick wall! Thank you in advance for your help!
I deduce you have contracted a lab to achieve your requirements. One would hope that the latter have expertise in the relevant area and would be able to advise you on an appropriate approach to carry out the validation.
Have you already been provided with information such as offered in Post 2 ?. (Readily available via Googling such as <<< validation of food extrusion to eliminate salmonella >>>> ?)
(Maybe/hopefully I am mistaken but I get the impression that the situation may be more like "the blind leading the blind".? If so you may be heading for an expensive and protracted undertaking).
Validation of this process is clearly not simple due the relative complexity of the dynamic situation. I would suggest you need to do some serious thinking about the precise scope of your validatory study although, hopefully, this already exists.
The work described in the useful 1st reference linked by Scampi seems to have been published (2018) with following abstract -
An increase in the number of foodborne outbreaks and recalls due to Salmonella in low-moisture foods has resulted in the need for the development and validation of process controls to ensure their microbiological safety. Furthermore, the FSMA's Preventive Controls for Human Food final rule requires food processors to validate their process controls to ensure food safety. The objective of this study was to develop a response surface model to predict Salmonella inactivation in oat flour, as affected by moisture, fat content, screw speed, and temperature. Oat flour was adjusted to different moisture (14 to 26% wet basis) and fat (5 to 15% [w/w]) contents and was then inoculated with a five-strain cocktail of Salmonella. Inoculated material was extruded through a single-screw extruder running at different screw speeds (75 to 225 rpm) and temperatures (65 to 85°C), without a die. Once steady-state conditions were attained, extruded samples were collected, cooled, and stored under refrigeration, and Salmonella survivors were enumerated. A split-plot central composite second-order response surface design was used, with the central point replicated six times. Temperature showed a significant (P < 0.0005) positive effect on microbial reduction. Moisture content showed significant linear (P = 0.0014) and quadratic (P = 0.0005) effects, whereas higher fat content showed a significant (P < 0.0001) protective effect on Salmonella destruction. The screw speed did not play a major role in inactivating Salmonella, but it had a significant (P = 0.0004) interactive effect with temperature. Results indicated that a >5.5-log reduction was achieved in oat flour extruded at a temperature above 85°C at all moisture and fat contents evaluated at a screw speed of 150 rpm. The developed response surface model can be used to identify the extrusion process conditions to achieve a desired reduction of Salmonella based on the moisture and fat contents of the product.
(also see "Conclusion", Pg 43 of Scampi's 1st link)
Hopefully, scope of yr system may have some similarity to that in above study.
JFI, this document illustrates Industry's degree of interest in seeking solutions to petfood extrusion systems
Kill step for low moisture extrusion, 2013.pdf 1.05MB
PS - Hi Scampi, sooo nice to see you have some time to be active again !!!