An EU-MDR question from a customer, Keith Morel shares his extended answer.

As EU-MDR experts, we are always willing to answer questions and share knowledge about the EU-MDR. This time, Keith Morel, VP Regulatory Affairs, gives an answer to a question about the use of Harmonized Standards in Clinical Evaluations, which might be interesting for all medical device manufacturers.

Keith: “I am happy to share with you that we received very positive feedbacks on the EU-MDR workshops that we organized this year in the Netherlands. Additionally, I recently received a question from an EU-MDR workshop follower and wanted to share this with you.”

The question from industry: 

In MEDDEV 2.7.1 rev 4, section 10.3, there are certain devices that may not require clinical data for demonstration of conformity. You mentioned the standards for electrosurgical devices (EN-60601-1) as an example.

The question is: If a device like an electro-surgery device that fulfills all relevant Harmonized Standards, the Risk Management report does not reveal any residual clinical risks/hazards, performance and usability can be shown with bench and pre-clinical data - would it be possible to demonstrate conformity, in your opinion, without the use of clinical data? Or is this clause (10.3) only intended for low-risk devices like a tongue depressor etc.?

In other words, is the intention with rev 4 that the (very) many medical devices on the EU market where there already are Harmonized Standards (electrosurgery, endoscopes, cardiac defib. etc = 60601-2-1 to 60601-2-68) would not need clinical data for the clinical evaluation using 10.3?

Keith: “The question described here, I think, is how the use of standards can be used to demonstrate conformity to safety, ER1.”

So, IF (and that can be quite a big IF) there are harmonized standards which address (i) all identified hazards, fully (ii) reduce the risks associated with those hazards/hazardous situations to a minimum and (ii) there are no “state of the art” considerations which supersede those standards, then I suppose that your suggestion could be possible. Of course, any gaps here would require clinical data and evaluation. 

BUT, remember, we also have to consider the other aspects of the essential requirements (ERs).

With regard to acceptable risk-benefit, we learn from the NBRG document on the interpretation of EN ISO 14971:2012 (Oct 2014), as well as from ISO/TF 24971:2013, that IF a standard has (i) a specific test method AND (ii) specific acceptance criteria (AC) which address a particular hazardous situation and its related risk, and a manufacturer (MFR) meets the requirements of that test method and those AC, THEN the residual risk can be accepted. 

In many cases, we either don’t have (i) or (ii) both! Besides, there’s also the aspect of performance (ER3). 
IF the standard(s) covered all the performances claimed by the manufacturer for a device (e.g. in the IFU and intended use statement, as well as any claims in promotional material) then, if the MFR uses full compliance to all those standards, then we could be covered there as well.

Lastly, we have the aspect of undesirable side effects (ER6). It’s much harder for me to see how compliance to a standard(s) can cover us here. However, I won’t say it’s impossible – e.g. for mature, well-established devices, with mature harmonized standards, that might be possible – those in the 60601-2-x arena come to mind, perhaps (but bear in mind, many harmonized standards include a requirement for clinical data within them, e.g. 11979-7 for IOLs; ISO 5840 for heart valves; ISO 25539-2 for coronary stents). 

Then, there are the specific ERs 7 till 13 for biocompatibility, sterilization, packaging, labeling etc. Again, we have harmonized standards in most of these general categories; but we have to be sure that all hazards/risks are appropriately considered, for the specific device in question. 


So, to conclude, your hypothesis is possible. I think one CAN justify, under MEDDEV 2.7.1 rev 4, section 10.3,  that a particular device (albeit rare) does NOT need clinical data to demonstrate conformity with the essential requirements - however, a clinical evaluation still needs to be done (last bullet of section 10.3). This clinical evaluation (based on bench tests, pre-clinical data, and data from harmonized standards tests) will have to justify NOT using clinical data, perhaps based on the maturity of the technology, the full understanding of the clinical hazards and related risks, as well as the clinical benefits, and the full use of applicable (harmonized) European standards to address these issues.  
It should be noted that there is still a lot of room for interpretation. As you point out, the concept of clinical evaluation is excellent but it requires that clinicians (i.e., the clinical evaluator) fully understand the guidelines. Furthermore, in light of the new MDR, also clinicians at the Notified Bodies and clinicians at Competent Authority level need to have the same understanding of the requirements.


I hope that my answer was clear and useful, should you wish to discuss your MDR program and regulatory details with your peers, then please join one of our (monthly) EU-MDR training in Europe or the US (November 2017, Irvine CA). 

Should you have any other question, then feel free to contact Qserve, we are always happy to discuss some details.


Keith Morel
Post date: August 16, 2017
Subscribe to the Qserve Newsletter and follow us on LinkedIn
Want more information? Please contact us.
Information request