Definition of Critical Equipment in Medical Device Production

[Georgethebest]

Hi everyone,


I'm currently trying to get a clearer understanding of how to define critical equipment in the context of medical device production. Specifically, I'm looking for guidance on the following points:

1. What criteria determine whether a piece of equipment is considered critical?
2. Are only monitoring and measurement equipment considered critical, or can process equipment (e.g., sealing machines, molding equipment) also be categorized as critical?
3. Which types of equipment require calibration certificates, and which do not?

I’ve gone through some regulatory documents (ISO 13485, MDR, etc.), but I’m having trouble finding a clear, practical definition or rule of thumb to apply during risk assessments or audits.


Any examples or insights would be much appreciated!


Thanks in advance,

 

[EmiliaBedelia]

The context of when/why you are defining "criticality" matters.

ISO 13485 tells you to assess the process risk with respect to product specifications and risk level, and control the measurement/qualification of the equipment and process commensurate to that risk. There's no need to explicitly define something as critical or not....but if you did, you'd assess that before applying risk controls so that you can determine how much risk control is necessary.
For the purposes of defining critical subcontractors/crucial suppliers (there's a definition in here: http://www.doks.nbog.eu/Doks/NBOG_BPG_2010_1.pdf), "critical" would be defined after risk controls because those steps can't be verified/controlled further and therefore have high residual risk.

To answer your specific questions:
1) Your process risk assessment. What is the risk to the patient if that equipment fails to perform?
2) Absolutely process equipment can be critical. Especially if the process cannot be measured/monitored (i.e., validated processes) - e.g., sterilization equipment, coatings, material treatment, etc.
3) If the output of the machine has to be accurate (see bullet 1), it has to be calibrated or otherwise qualified to ensure that it is performing appropriately. IQ/OQ/PQ might be more appropriate for some types of machines. Again, this depends on risk and what is necessary to show adequate controls/risk reduction for that step.

 

[Georgethebest]

Thanks, Emilia, that really helps clarify things—especially the distinction between when "critical" is defined in relation to risk controls and when it's about residual risk.

Would it be justified to say that if the equipment used (e.g., a measuring device) has a resolution or accuracy at least 10x better than what's required for the process, then the need for frequent calibration may be reduced or even unnecessary—provided there's documented evidence of long-term stability and low risk of drift?

Also worth mentioning: the ISO 13485 Practical Guide does refer to critical equipment, which suggests the term can have a practical use in risk-based thinking, even if it’s not strictly required for classification purposes.


Appreciate all the insights!

 

[Billy Milly]

Would it be justified to say that if the equipment used (e.g., a measuring device) has a resolution or accuracy at least 10x better than what's required for the process, then the need for frequent calibration may be reduced or even unnecessary

This kinda goes in a different direction than your original question. In the context of the first post, it does not matter how accurate the device is, it's what it does that matters. If I take your example, if you measure a sterile product's primary packaging width (kinda cosmetic), this is less risky. If you measure the seal width (critical to packaging integrity), this is high risk. As Emilia suggested, use your production process risk assesment (e.g. production process FMEA). This detremines critical production operations/steps/processes. Criticallity for different products on the same equipment can be different, because your basic determinator is the product and its influence to patient, not the equipment itself.
This should help with #1 and #2. For #3 - in general, calibrate all the production and measuring equiment you can calibrate. Extent and/or period for calibration can be risk-based (influence to patient/user). Stuff that cannot be calibrated, should be either validated (again proportionate to risk) or its output has to be 100% inspected (again with calibrated or validated equipment).

 

[ChrisM]

Process equipment: If one of your pieces of equipment fails, how will you cope? If you can go out and buy an equivalent or identical item off-the-shelf, it is not critical. In general terms, if it is built to your specifications so you would need to put some effort into finding someone who could build you a replacement, then it is critical as your production could stop for a while whilst you sourced this replacement

 

[Billy Milly]

Chris, you addresed the business risk prespective (no product - no revenue). Relevant and sensible, but not strictly mandatory from regulatory perspective. Me and EB addessed the "mandatory" product/user/patient risks. If no useful RMS is in place, the latter should be covered first.
Just clarifying for members not familiar with risk management.

 

[ChrisM]

Billy Milly, I was responding to the question of "critical equipment in the context of medical device production".
When conducting risk analysis, wherever I have worked one major factor has always been to look at continuity of production, and this takes into account machine breakdown. As you say, no product if you have such an event and no replacement or contingency plan.

Lack of production may not be critical in terms of the regulatory perspective, but then if you're not able to produce anything you don't have to worry too much about the regulations.....