Defending "No Essential Performance" for a Surgical Video Camera

[infraredmed]

Hi all,

Looking for suggestions on the best way to approach defining that our medical device has NO Essential Performance.

Our device which we're seeking 60601 Ed. 3 certification for is essentially a surgical camera for streaming and capturing digital and infrared images. It is non-diagnostic, non-radiating, non-contacting, non-monitoring and is simply meant to record and capture images.

I believe it doesn't have EP because nothing that it does or provides translates to patient care. Absence or degradation of the device, whether it is completely non-functional or functioning but providing poor quality images, doesn't present unacceptable risk. However, absence or degradation of the device would prevent it from achieving the intended use of streaming and capturing images.

So, what it boils down to is whether or not the NRTL looks at EP as 1) freedom from unacceptable risk or 2) achieving intended use.

Anyone have suggestions on how to approach this? Is my rationale correct?

Thanks!!

 

[Mark Meer]

Just a thought: because the device is electrical, wouldn't this entail some elements of essential performance? Just about anything electrical will carry with it some potential electrical hazards, which presumably your device has some means of controlling?

 

[infraredmed]

Hi Mark,

Thanks for reply. Correct me if I'm wrong (I very well could be), but my understanding is that anything relating to potential electrical hazards would fall into basic safety and not essential performance.

 

[MIREGMGR]

What use is made of the captured image-stream, if that image-stream is not used for diagnosis or patient care in any way...?

How does the camera meet the definition of a medical device if it has no patient related function?

Per the MDD (paraphrased), a medical device must be intended to be used for:

• Diagnosis, prevention, monitoring, treatment, or alleviation of disease
• Diagnosis, monitoring, treatment, alleviation of, or compensation for an injury or handicap
• Investigation, replacement, or modification of the anatomy or of a physiological process

If the camera is intended to be used to do one of those things, it's a medical device, and it may have an Essential Performance. If it isn't intended to do one of those things, it seems to me that it's not a medical device.

 

[yodon]

We've bumped into this with at least one other device. Conceptually, it's difficult to accept the idea that a product has no essential performance (not exactly a good advertising angle!). The case we had was a bone stimulator. If it failed to perform, the bone would still heal (naturally), just maybe not as fast. So we were able to successfully argue no EP.

The electrical stuff, as I have had it explained to me, falls under "basic safety" and is not, as surmised, under essential performance. Same goes for sharp edges, etc.

Based on your description, though, you may need to review to see, as MIREMGR points out, whether it is a medical device.

So if it is a device, fall back to the risk analysis and determine if there is any risk with using the device and if lack of performance could lead to hazards. If there are none then you should be able to successfully defend the assertion of no essential performance.

 

[wally4u]

Personally, because it is an electrical device, I would expect an EP.

But since the manufacturer is allowed to define what the EP is, you could simply state:
"Since the device is not life supporting it is allowed to fall into a safe state (no output/power/etc) requiring operator intervention to restart)"

If you do that, you make a statement that there is an EP (you also need this of EMC testing), but it does not cause any harm when it fails.

 

[Peter Selvey]

Ed: after writing all this I realized that the video camera in question is non-contact, so the example for disinfection makes no sense .

This is a bit of a mess in the standard, as the IEC series uses "essential performance" for at least 3 distinct meanings and no one definition fits all.

In practice there will always be essential performance even for a low risk device. For imaging this will be things like resolution, contrast, dynamic range, color reproduction and so on. Failure provide a minimum level of performance can affect the clinical outcome so it is important.

The question then falls to how this fits into IEC 60601-1. In 2005 edition there is no general clause which requires essential performance to be confirmed (it seems like an oversight, but it could be deliberate). The requirement does pop up in special cases like home use equipment (IEC 60601-1-11), and some particular specify minimum performance (e.g. IEC 60601-2-25). And there is Clause 14, which is discussed below.

In the 2012 amendment (IEC 60601-1/A1:2012), essential performance is called up more often. Essential performance should not be affected by stress tests which represent normal condition (e.g. drop tests, IP tests, cleaning, sterilization etc).

It is reasonable, for example, to expect that the minimum performance of the surgical video should not be affected by multiple disinfections using the harshest chemical shown in the operation manual.

This also fits with medical device regulation (performance shall not be adversely affected by the conditions and stresses of normal use).

However the standard remains rather spotty and inconsistent on this approach. Some tests like relative humidity test (48hr/93%) make no mention of essential performance, even though this is logical.

And of course, testing things like video performance is way out the test labs competence. They would rather like to stick in their comfort zone of electrical safety.

So expect this area to remain confused.

Finally there is the issue of Clause 14, design controls for programmable systems. This is a significant point as it is a lot of work to prepare the required documentation, and have it reviewed by a test lab. If all devices have some form of essential performance, it would seem like Clause 14 is always applicable if there is software involved. At the same time, the work seems excessive for low risk devices, and conflicts with regulations which usually do not require any design controls for low risk devices.

However, there is an out: Clause 14.1 says that if the risk management process considered the failure of the programmable system to have no significant risk, the remainder of Clause 14 can be excluded. Some test labs may argue against but the standard has a clear path linking exclusion to risk management, so it is really not up to the test lab to decide.

 

[Mark Meer]

For imaging this will be things like resolution, contrast, dynamic range, color reproduction and so on. Failure provide a minimum level of performance can affect the clinical outcome so it is important.

I agree that there is a lot of room for case-by-case interpretation, but nevertheless, the definition of Essential Performance in the the standard does explicitly say "...freedom from unacceptable risk.", and as the OP points out degradation of image quality is not such a risk.

...I think, as stated by the OP, essential performance is often misinterpreted as related to intended use, and not to risk.

 

[Peter Selvey]

I think this is an interesting philosophical area, but of course in practice for low risk devices I would never argue. If a manufacturer wants to claim no essential performance, it is OK as long as the documentation is consistent (for example, manufacturers often claim no essential performance but it does not fit with EMC test plans and other areas).

From philosophical side (semantics ...):

One of the problems in discussions is on essential performance is the hidden context. In that context it might be perfectly reasonable to ignore aspects of performance.

For example, earlier on many experts used a home use thermometer as an example of a device with no essential performance. Thier context was direct harm. However if we include indirect harm the performance is essential, and this has been confirmed by the standard ISO 80601-2-56.

Next, many people consider that "degradation of performance has no unacceptable risk". The hidden context with such a statement is that degradation is an abnormal or at least an unusual event. If the degradation occurred frequently, the risk should be deemed unacceptable. Otherwise, it calls into question what the medical benefit of the device is. A device without any reliable benefit (reliable performance) is not a medical device.

Of course, the hidden assumption may be entirely reasonable. Reputable manufacturers are not going to put a product on the market that fails most of the time. Quality manufacturers can be expected to throw everything at their product to make sure stands up well to expected normal use.

Nevertheless it is a hidden assumption, and not always true. As an auditor I can come across many copycat manufacturers that don't understand the technology, use fake parts and create stuff that really does fail a lot of the time (if not all the time). And sometimes even the best manufacturers will mess up and fail to realize that some of the stresses in normal use such as humidity, cleaning, mechanical, extreme ambient temperature, EMI cause degradation of performance with high probability.

For example, a ultrasound probe manufacturer may not realize that multiple cleaning with a common and recommended cleaning agent causes the plastic in the acoustic window to harden, degrading the image quality.

A general, broad, context insensitive, hidden assumption free definition of essential performance must have some performance. Otherwise, it is not a medical device.

 

[michm9]

This whole essential performance discussion is full of ambiguity. I've been trying to conduct a risk analysis on a cryogenic freezer that is considered class IIa in Europe. What is the clinical application in this case? Complete failure of the tank would cause no harm to anyone. Even if I assume they are storing life saving specimens in the freezer, the loss of which could delay patient treatment (or worse), how is that for the freezer manufacture to mitigate? Shouldn't the user of the freezer be following appropriate level of GMP control? Very frustrating. Also, great to see the Cove back up and running.