Can you use Mean Kinetic Temperature to Account for Spikes in Accelerated Aging Chambers?

Paul Daniel, Vaisala
Senior GxP Regulatory Compliance Expert
Industrial Measurements
Life Science

In this week's blog, Paul Daniel has a little back and forth on a question he received from a webinar attendee on using mean kinetic temperature in elevated-temperature storage applications.

Dear Paul, We just watched the webinar on MKT. We enjoyed it, but have a question about elevated-temperature storage. Is it ok to use if mapping can be proven to provide a MKT? In our case, I'm referring specifically to products stored for accelerated aging studies?  We are interested if you have any advice… R

Dear R, 

Thank you for joining the webinar! Current guidance is clear that MKT shouldn't be used above 30°C (from the MHRA) or above 40°C (from the USP). And, in your case, since you are doing elevated-temperature studies, it should be in a well-controlled stability chamber, so you shouldn't be dealing with excursions anyway.  

With that in mind, what do you want to use MKT analysis to achieve within your accelerated aging study? 

I look forward to your answer! 


Trend report Vaisala

Dear Paul, 

I am happy to provide more detail. We provide accelerated aging and real-time storage capabilities for multiple facilities.  Typically class II products only.  We monitor the real-time and accelerated validated chambers with data loggers. 

We provide temperature summary (histogram) reports to the facilities we service.  We see excursions when chambers have their doors opened to place or remove product.  Since multiple studies are in the same chambers, when it's time to provide a customer report of their study, the histogram shows spikes due to opening and closing of doors.

This is why we've been using MKT to explain the excursion; it assures the customer the product did not experience the temperature change during this excursion. Only the air temperature of the chamber was changed.  Additionally, we have mapped the real-time storage area and although there is no real call out for storing class II products, we offer 23 ± 3°C monitoring and storage capability, all tracked through a LIMS software. 

We use the MKT temperature value derived from our mapping.  We can use the default but wanted to explain to the customers that the spikes seen on a histogram report are explainable and had no adverse effect on the product.  Is this a proper use of the MKT values? 

Thanks again!

Dear R, 

Excellent outline! I wouldn't say your use of MKT is improper, but even though I follow your logic, in my experience it's unconventional. I fear I may be misunderstanding, but to give you as much information as possible within this letter, I'm going to make a few assumptions about your process and rationale. 

It sounds like the temperature spikes on the histogram are the transient result of door opening/closing, affecting only chamber air temperature, while the product temperature is stable. You presume this stability because the product is buffered by its thermal mass and/or product packaging.  This is a completely reasonable presumption.  Since you serve customers who aren't familiar with your processes, it's nice to have a way to SHOW the customers that this statement is true. 

In my experience, the typical way to show that this statement is true (product did not experience a significant change in temperature, only the air in the chamber) would be to load the chamber with placebos that are representative of the thermal properties of the product, and do some door opening tests while monitoring both the air temperatures and the product temperatures.  Then you have relevant data from actual excursions. You can prove that the product does not experience the same temperature changes measured in the chamber air. 

It sounds like your facility is using MKT to achieve this end.  The math of the MKT weighted average is based on the idea that the product IS ACTUALLY EXPERIENCING the temperatures that you a using in your calculation.  So the conclusions that can be drawn from it are different.

If your claim is that the product never really changed temperature, that it was only the air that changed temperature, the MKT value cannot represent this. Currently, your MKT analysis is saying this:  

"Even though your product experienced a temperature spike, the amount of chemical breakdown that happened during this time of elevated temperature is minimal, as can be seen by the fact that the MKT value, over the storage period, remained below 25°C (or whatever is your target value)". 

However, I would argue that MKT assumes the product actually experienced the temperature excursion.This doesn't mean that you can't use MKT analysis for your purpose, as long as you are clear about what the MKT value is meant to represent.  

You can, however, say this to your customer with MKT in mind…  

"There were several air temperature spikes due to door openings.  We believe based on our mapping data that the product does not change temperature significantly during these air temperature excursions.  However, if we assume the worst case scenario that the product temperature matched the air temperature, MKT analysis over the product storage period shows that because MKT remains below 25°C.  Therefore no product damage could have resulted from these excursions, because product temperature was always be lower than the air temperature for high side excursions."

If you are dealing with low-temperature excursions (which might be the case with accelerated studies at higher than ambient temperatures), MKT analysis theoretically isn't helpful. The value doesn't account for low-temperature affects.  However, in this case, you aren't trying to prove the product wasn't damaged, but you might be trying to ensure that the study wasn't compromised by time below the lower limit of 21°C. 

Please let me know your thoughts… Was this helpful? 

Best Regards, 

Paul Daniel


Dear Paul, 

Absolutely! Thank you for your response and I really appreciate and respect your opinion.  We did perform door opening tests with loaded and unloaded chambers with separate monitoring of internal product temperatures and external air temperatures to make the correlation. 

We also performed chamber-off worst case scenarios to see how much time it takes for the product to change temperature when the chamber lost power to provide us an indication of when we need to prompt an action.   

Your information is the first time I have been able to get a clear analysis for using MKT or not.  We will use caution when trying to explain these excursions and limit the references to MKT.   

Thanks again!  R

Add new comment